Printed Wiring Board
Surface Finishes
Cleaner
Technologies
Substitutes
Assessment
VOLUME 1
Jack R. Geibig, Senior Research Associate
Mary B. Swanson, Research Scientist
and the
PWB Engineering Support Team
This document was produced by the University of Tennessee
Center for Clean Products and Clean Technologies under grant
#X825373 from EPA's Design for the Environment Branch,
Economics, Exposure, & Technology Division, Office of
Pollution Prevention and Toxics.
or
U.S.EPA
-------
PWB Engineering Support Team
The PWB Engineering Support Team consisted of University of Tennessee faculty and
graduate students who developed analytical models for the project and/or authored sections of
this document. Members of the Team and the sections to which they contributed are listed
below:
Exposure Assessment and Risk Characterization
Dr. Chris D. Cox, Associate Professor of Civil and Environmental Engineering,
Dr. R. Bruce Robinson, Professor of Civil and Environmental Engineering,
and graduate research assistants in Civil and Environmental Engineering:
Aaron Damrill, Jennie Ducker, Purshotam Juriasingani, and Jeng-hon Su.
Cost Analysis
Dr. Rupy Sawhney, Assistant Professor of Industrial Engineering and Director, Lean Production
Laboratory, and graduate research assistants in Industrial Engineering: Aamer Ammer.
Disclaimer
This document was written by the grantee. It has not been through a formal external peer review
process. Some information in this document was provided by individual technology vendors and
has not been independently corroborated by EPA or the University of Tennessee. The use of
specific trade names or the identification of specific products or processes in this document are
not intended to represent an endorsement by EPA or the U.S. Government. Discussion of federal
environmental statutes is intended for information purposes only; this is not an official guidance
document, and should not be relied on by companies in the printed wiring board industry to
determine applicable regulatory requirements.
-------
For More Information
To learn more about the Design for the Environment (DfE) Printed Wiring Board Project
or the DfE Program, please visit the DfE Program web site at:
www. epa.gov/dfe
The DfE web site also contains the document, Cleaner Technologies Substitutes Assessment: A
Methodology and Resources Guide, which describes the basic methodology used in this
assessment.
To obtain copies of DfE Printed Wiring Board Project technical reports, pollution
prevention case studies, and project summaries, please contact:
Pollution Prevention Information Clearinghouse (PPIC)
U.S. Environmental Protection Agency
1200 Pennsylvania Ave., N.W. (7407)
Washington, DC 20460
Phone:(202)260-1023
Fax: (202) 260-4659
E-mail: ppic@epa.gov
Web site: www.epa.gov/opptintr/library/ppicdist.htm
To learn more about the University of Tennessee Center for Clean Products and Clean
Technologies, visit the Center's web site at:
eerc.ra.utk.edu/clean/
11
-------
Acknowledgments
This Cleaner Technologies Substitutes Assessment (CTSA) was prepared under a grant
from the U.S. Environmental Protection Agency's Design for the Environment (DfE) Program,
Office of Pollution Prevention and Toxics, by the University of Tennessee (UT) Knoxville
Center for Clean Products and Clean Technologies and the PWB Engineering Support Team,
with assistance from numerous UT students and staff. The authors would like to acknowledge
the outstanding contributions of Lori Kincaid, UT Center for Clean Projects and Clean
Technologies, who provided guidance throughout the project; Catherine Wilt, UT Center for
Clean Projects and Clean Technologies, who researched and wrote the Regulatory status section
of this document; James Dee, UT Center for Clean Projects and Clean Technologies, who
assisted with the development of the Human Health and Ecological Hazards Summary; and
Margaret Goergen, who was the document production manager.
This document was produced as part of the DfE Printed Wiring Board Project, under the
direction of the project's Core Group members, including: Kathy Hart, Project Lead and Core
Group Co-Chair, U.S. EPA, Office of Pollution Prevention and Toxics (OPPT), Economics,
Exposure and Technology Division (EETD), DfE Branch; Holly Evans, formerly of
IPC-Association Connecting Electronics Industries (IPC), and Fern Abrams, IPC, Core Group
Co-Chairs; Dipti Singh, Technical Lead and Technical Workgroup Co-Chair, U.S. EPA, OPPT,
EETD, DfE Branch; John Sharp, Teradyne Inc., Technical Workgroup Co-Chair; Michael
Kerr, BHE Environmental, Inc., John Lott, DuPont Electronics; Greg Pitts, Ecolibrium; Gary
Roper, Substrate Technologies, Inc.; and Ted Smith, Silicon Valley Toxics Coalition. Many
thanks also to the industry representatives and other interested parties who participated in the
project's Technical Workgroup, especially David Ormerod of Dexter Electronic Materials (now
Polyclad Technologies - Enthone), Michael Schectman of Technic, Inc., David Hillman of
Rockwell International Corp., and Eric Brooman of Concurrent Technologies Corporation.
We would like to acknowledge Ron Iman of Southwest Technology Consultants for his
work in analyzing and presenting the results of the performance demonstration, and Terry
Munson of Contamination Studies Laboratory (CSL), Inc., for conducting a failure analysis and
helping to present the performance demonstration results. We also appreciate the efforts of Jeff
Koon of Raytheon Systems Company and David Hillman for their help in planning and
conducting the performance demonstration. Recognition is also given to ADI/Isola, who
supplied the materials to build the performance demonstration test boards, to Network Circuits
for volunteering to build the boards, and to the test facilities that ran the boards through their
surface finish lines and provided critical data for the study. Performance demonstration
contractor support was provided by Abt Associates, Inc., of Cambridge, MA, under the direction
of Cheryl Keenan.
in
-------
EPA Design for the Environment Workgroup
We would like to express appreciation to the U.S. EPA Office of Pollution Prevention and
Toxics Design for the Environment Workgroup, who provided valuable expertise in the
development of this CTSA and reviewed all draft documents.
Andrea Blaschka
Susan Dillman
Conrad Flessner
Franklyn Hall
Susan Krueger
Tim Lehman
Fred Metz
Dave Monroe
Jerry Smrchek
Participating Suppliers
We would like to thank the suppliers for their participation in the DfE Printed Wiring
Board Surface Finishes Project. In addition to supplying critical information regarding the
various technologies, these companies also made significant contributions in planning and
conducting the performance demonstration.
Alpha Metals
(now Polyclad Technologies-Enthone)
Steve Beigle
50 Burr Ridge Parkway
Burr Ridge, IL 60521
(630) 794-9329
Dexter Electronic Materials
(now Polyclad Technologies-Enthone)
David Ormerod
144 Harvey Road
Londonderry, NH 03053
(603) 645-0021, ext. 203
Electrochemicals, Inc.
Paul Galatis
5630 Pioneer Creek
Maple Plain, MN 55359
(612)479-2008
Florida CirTech
Mike Scimeca
13 09 North 17th Avenue
Greeley, CO 80631
(970) 346-8002
MacDermid, Inc.
Don Cullen
245 Freight Street
P.O. Box 671
Waterbury, CT 06702
(203)575-5658
Technic, Inc.
Michael Schectman
1 Spectacle Street
Cranston, RI 02910
(401)781-6100
IV
-------
Performance Demonstration Sites
We would like to recognize the companies and their facilities that participated in our
performance demonstration, for donating the time and materials necessary to carry out the
testing. We also appreciate the assistance they provided in gathering data necessary to conduct
this assessment.
AIK Electronics (UK) Ltd. (formerly
Philips Printed Circuits)
Artetch Circuits Ltd.
Central Circuits Inc.
Circuit Connect Inc.
Circuit Wise Inc.
Diversified Systems
GTC Circuits Corp. (formerly General
Technology Corp.)
H-R Industries
Parlex Corp.
Quality Circuits Inc.
Sanmina Corp. (formerly Altron Inc.)
Sanmina Corp. (formerly Hadco Corp.)
Solder Station One, Inc.
v
-------
Table of Contents
Executive Summary ES-1
Chapter 1
Introduction 1-1
1.1 Project Background 1-2
1.1.1 EPA Dffi Program 1-2
1.1.2 DfE PWB Program 1-2
1.2 Overview of PWB Industry 1-5
1.2.1 Types of Printed Wiring Boards 1-5
1.2.2 Industry Profile 1-5
1.2.3 Overview of Rigid Multi-Layer PWB Manufacturing 1-7
1.3 CTSA Methodology 1-8
1.3.1 Identification of Alternatives and Selection of Project Baseline 1-8
1.3.2 Boundaries of the Evaluation 1-9
1.3.3 Issues Evaluated 1-10
1.3.4 Primary Data Sources 1-11
1.3.5 Project Limitations 1-13
1.4 Organization of this Report 1-15
References 1-16
Chapter 2
Profile of the Surface Finishing Use Cluster 2-1
2.1 Chemistry and Process Description of Surface Finishing Technologies 2-1
2.1.1 Process Sequence of Surface Finishing Technologies 2-1
2.1.2 Overview of the Surface Finishing Manufacturing Process 2-3
2.1.3 Chemistry and Process Descriptions of Surface Finishing Technologies 2-4
2.1.4 Chemical Characterization of Surface Finishing Technologies 2-17
2.2 Additional Surface Finishing Technologies 2-23
2.2.1 Immersion Palladium 2-23
References 2-25
Chapter 3
Risk Screening and Comparison 3-1
3.1 Source Release Assessment 3-1
3.1.1 Data Sources and Assumptions 3-2
3.1.2 Overall Material Balance for Surface Finishing Technologies 3-3
3.1.3 Source and Release Information for Specific Surface Finishing
Technologies 3-17
3.1.4 Uncertainties in the Source Release Assessment 3-34
3.2 Exposure Assessment 3-35
3.2.1 Exposure Setting 3-35
3.2.2 Selection of Exposure Pathways 3-40
vi
-------
3.2.3 Exposure-Point Concentrations 3-43
3.2.4 Estimating Potential Dose Rates 3-56
3.2.5 Uncertainty and Variability 3-75
3.2.6 Summary 3-77
3.3 Human Health and Ecological Hazards Summary 3-80
3.3.1 Carcinogenicity 3-80
3.3.2 Chronic Effects (Other than Carcinogenicity) 3-83
3.3.3 Ecological Hazard Summary 3-96
3.3.4 Summary 3-102
3.4 Risk Characterization 3-104
3.4.1 Summary of Exposure Assessment 3-104
3.4.2 Summary of Human Health Hazards Assessment 3-109
3.4.3 Summary of Ecological Hazards Assessment 3-109
3.4.4 Methods Used to Calculate Human Health Risks 3-110
3.4.5 Results of Calculating Human Health Risk Indicators 3-113
3.4.6 Evaluation of Lead Risks from Tin-Lead Solder Used in the HASL
Process 3-125
3.4.7 Results of Calculating Ecological (Aquatic) Risk Indicators 3-128
3.4.8 Uncertainties 3-130
3.4.9 Conclusions 3-132
3.5 Process Safety Assessment 3-137
3.5.1 Chemical Safety Concerns 3-137
3.5.2 Hot Air Solder Leveling (HASL) Process Safety Concerns 3-146
3.5.3 Process Safety Concerns 3-146
References 3-151
Chapter 4
Competitiveness 4-1
4.1 Performance Demonstration Results 4-1
4.1.1 Background 4-1
4.1.2 Performance Demonstration Methodology 4-2
4.1.3 Test Vehicle Design 4-4
4.1.4 Environmental Testing Methodology 4-8
4.1.5 Analysis of the Test Results 4-8
4.1.6 Overview of Test Results 4-13
4.1.7 HCLV Circuitry Performance Results 4-18
4.1.8 HVLC Circuitry Performance Results 4-20
4.1.9 High Speed Digital Circuitry Performance Results 4-21
4.1.10 High Frequency Low Pass Filter Circuitry Performance Results 4-22
4.1.11 High Frequency Transmission Line Coupler Circuitry Performance
Results 4-27
4.1.12 Leakage Measurements Performance Results 4-29
4.1.13 Stranded Wires 4-32
4.1.14 Failure Analysis 4-33
4.1.15 Summary and Conclusions 4-40
4.1.16 Boxplot Displays 4-42
vn
-------
4.2 Cost Analysis 4-55
4.2.1 Overview of the Cost Methodology 4-56
4.2.2 Cost Categories and Discussion of Unquantifiable Costs 4-57
4.2.3 Simulation Modeling of Surface Finishing Processes 4-62
4.2.4 Activity-Based Costing 4-68
4.2.5 Cost Formulation Details and Sample Calculations 4-71
4.2.6 Results 4-84
4.2.7 Conclusions 4-86
4.3 Regulatory Assessment 4-88
4.3.1 Clean Water Act 4-88
4.3.2 Clean Air Act 4-93
4.3.3 Resource Conservation and Recovery Act 4-95
4.3.4 Comprehensive Environmental Response, Compensation and
Liability Act 4-97
4.3.5 Superfund Amendments and Reauthorization Act and Emergency
Planning and Community Right-To-Know Act 4-98
4.3.6 Toxic Substances Control Act 4-100
4.3.7 Summary of Regulations for Surface Finishing Technologies 4-101
References 4-108
Chapter 5
Conservation 5-1
5.1 Resource Conservation 5-1
5.1.1 Consumption of Natural Resources 5-2
5.1.2 Consumption of Other Resources 5-8
5.1.3 Summary and Conclusions 5-10
5.2 Energy Impacts 5-12
5.2.1 Energy Consumption During Surface Finishing Process Operation 5-12
5.2.2 Energy Consumption Environmental Impacts 5-18
5.2.3 Energy Consumption in Other Life-Cycle Stages 5-21
5.2.4 Summary and Conclusions 5-22
References 5-23
Chapter 6
Additional Environmental Improvements Opportunities 6-1
6.1 Pollution Prevention 6-2
6.1.1 Management and Personnel Practices 6-3
6.1.2 Materials Management and Inventory Control 6-6
6.1.3 Material Selection 6-7
6.1.4 Process Improvements 6-8
6.2 Recycle, Recovery, and Control Technologies Assessment 6-19
6.2.1 Recycle and Resource Recovery Opportunities 6-19
6.2.2 Control Technologies 6-28
References 6-38
Vlll
-------
Chapter 7
Choosing Among Surface Finishing Technologies 7-1
7.1 Risk, Competitiveness, and Conservation Data Summary 7-2
7.1.1 Risk Summary 7-2
7.1.2 Competitiveness Summary 7-8
7.1.3 Resource Conservation Summary 7-14
7.2 Social Benefits/Costs Assessment 7-16
7.2.1 Introduction to Social Benefits/Costs Assessment 7-16
7.2.2 Benefits/Costs Methodology and Data Availability 7-18
7.2.3 Private and External Benefits and Costs Associated with Choice of
Surface Finishing Alternative 7-19
7.2.4 Summary of Benefits and Costs 7-28
7.3 Technology Summary Profiles 7-30
7.3.1 HASL Technology 7-30
7.3.2 Nickel/Gold Technology 7-35
7.3.3 Nickel/Palladium/Gold Technology 7-39
7.3.4 OSP Technology 7-44
7.3.5 Immersion Silver Technology 7-47
7.3.6 Immersion Tin Technology 7-51
References 7-56
IX
-------
List of Tables
Table 1-1. Surface Finishing Technologies Submitted by Chemical Suppliers 1-11
Table 1-2. Responses to the PWB Workplace Practices Questionnaire 1-13
Table 2-1. Use Cluster Chemicals and Associated Surface Finishing Processes 2-18
Table 3-1. Water Usage of Surface Finishing Technologies From Questionnaire 3-7
Table 3-2. Reported Use of Chemical Flushing as a Tank Cleaning Method 3-8
Table 3-3. Average Bath Dimensions and Temperatures for All Processes 3-10
Table 3-4. Spent Bath Treatment and Disposal Methods 3-13
Table 3-5. RCRA Wastes and Container Types for Surface Finishing Technologies .... 3-16
Table 3-6. Workplace Activities and Associated Potential Exposure Pathways 3-40
Table 3-7. Potential Population Exposure Pathways 3-42
Table 3-8. Results of Workplace Air Modeling 3-47
Table 3-9. Results of Ambient Air Modeling 3-51
Table 3-10. Estimated Releases to Surface Water Following Treatment 3-54
Table 3-11. Parameter Values for Workplace Inhalation Exposures 3-57
Table 3-12. General Parameter Values for Workplace Dermal Exposures 3-59
Table 3-13. Parameter Values for Workplace Dermal Exposures for Line Operators
on Non-Conveyorized Lines 3-60
Table 3-14. Parameter Values for Workplace Dermal Exposure for Line Operators
on Conveyorized Lines 3-61
Table 3-15. Parameter Values for Workplace Dermal Exposure for a Laboratory
Technician on Either Conveyorized or Non-Conveyorized Lines 3-62
Table 3-16. Estimated Average Daily Dose for Workplace Exposure From Inhalation
and Dermal Contact 3-63
Table 3-17. Estimated Concentration of Lead in Adult and Fetal Blood from Incidental
Ingestion of Lead in Tin/Lead Solder 3-72
Table 3-18. Parameter Values for Estimating Nearby Residential Inhalation Exposure . . . 3-74
Table 3-19. Estimated Average Daily Dose for General Population Inhalation
Exposure 3-74
Table 3-20. Children's Blood-Lead Results from the IEUBK Model at Various Lead
Air Concentrations 3-75
Table 3-21. Available Carcinogenicity Information 3-81
Table 3-22. Summary of RfC and RfD Information used in Risk Characterization for
Non-Proprietary Ingredients 3-84
Table 3-23. NOAEL/LOAEL Values Used in Risk Characterization for
Non-Proprietary Ingredients 3-87
Table 3-24. Developmental Toxicity Values Used in Risk Characterization for
Non-Proprietary Ingredients 3-88
Table 3-25. Summary of Health Effects Information 3-89
Table 3-26. Overview of Available Toxicity Data 3-93
Table 3-27. Estimated (Lowest) Aquatic Toxicity Values and Concern Concentrations
for PWB Surface Finishing Chemicals, Based on Measured Test Data or
SAR Analysis 3-96
Table 3-28. Environmental Hazard Ranking of PWB Finishing Chemicals 3-100
-------
Table 3-29. Gastrointestinal (GI) Absorption Factors 3-112
Table 3-30. Summary of Human Health Risks From Occupational Inhalation
Exposure for Selected Chemicals 3-116
Table 3-31. Summary of Human Health Risks Results From Occupational Dermal
Exposure for Selected Chemicals 3-118
Table 3-32. Summary of Potential Human Health Effects for Chemicals of Concern ... 3-121
Table 3-33. Data Gaps for Chronic Non-Cancer Health Effects for Workers 3-122
Table 3-34. Risk Evaluation Summary for Lead 3-126
Table 3-35. Summary of Aquatic Risk Indicators for Non-Metal Chemicals
of Concern 3-129
Table 3-36. Summary of Aquatic Risk Indicators for Metals Assuming No On-Site
Treatment 3-130
Table 3-37. Overall Comparison of Potential Human Health and Ecological Risks
for the Non-Conveyorized HASL and Alternative Processes 3-136
Table 3-38. Flammable, Combustible, Explosive, and Fire Hazard Possibilities for
Surface Finishing Processes 3-139
Table 3-39. Corrosive, Oxidizer, Reactive, Unstable, and Sudden Release of Pressure
Possibilities for Surface Finishing Processes 3-141
Table 3-40. Sensitizer, Acute and Chronic Health Hazards, and Irreversible
Eye Damage Possibilities for Surface Finishing Processes 3-143
Table 4-1. Electrical Responses for the Test PWA and Acceptance Criteria 4-6
Table 4-2. Distribution of the Number of LRSTF PWAs by Surface Finish,
Site, and Flux 4-7
Table 4-3. Listing of 23 Site/Flux Combinations Used in the Multiple Comparisons
Analyses 4-11
Table 4-4. Number of Anomalies Observed at Each Test Time 4-13
Table 4-5. Percentage of Circuits Meeting Acceptance Criteria at Each Test Time 4-14
Table 4-6. Comparison of CCAMTF Pre-Test Ranges with DfE Pre-Test
Measurements 4-15
Table 4-7. Frequency Distribution of Post 85/85 Anomalies per PWA by
Surface Finish 4-16
Table 4-8. Frequency Distribution of Post-Thermal Shock Anomalies per PWA by
Surface Finish 4-17
Table 4-9. Frequency Distribution of Post-Mechanical Shock Anomalies per PWA by
Surface Finish 4-18
Table 4-10. P-Values for HCLV Test Results 4-19
Table 4-11. Number of HCLV PTH Anomalies at Post-Mechanical Shock by
Surface Finish 4-20
Table 4-12. P-Values for HVLC Test Results 4-21
Table 4-13. P-Values for HSD Test Results 4-22
Table 4-14. P-Values for HF LPF Test Results 4-23
Table 4-15. Frequency Distribution of HF LPF Anomalies at Post-Mechanical Shock
per PWA 4-25
Table 4-16. Comparison of the Observed and Expected Number of Anomalies for
the HF LPF PTH 50MHz Circuit by Surface Finish 4-26
XI
-------
Table 4-17. Comparison of the Observed and Expected Number of Anomalies Under the
Hypothesis of Independence of Surface Finishes 4-26
Table 4-18. P-Values for HF TLC Test Results 4-28
Table 4-19. P-Values for Leakage Test Results 4-30
Table 4-20. P-Values for Stranded Wire Test Results 4-32
Table 4-21. Identification of Assemblies Selected for Ion Chromatography Analysis .... 4-35
Table 4-22. Ion Chromatography AnionHData (HASL) 4-36
Table 4-23. Ion Chromatography Anion^Data (Immersion Tin) 4-36
Table 4-24. Ion Chromatography Anion^Data (Immersion Silver) 4-37
Table 4-25. Ion Chromatography AnionHData (Nickel/Gold) 4-37
Table 4-26. Ion Chromatography Anion^Data (OSP) 4-37
Table 4-27. Ion Chromatography Anion H Data (Nickel/Palladium/Gold) 4-38
Table 4-28. Acceptance Levels for Weak Organic Acids 4-39
Table 4-29. Frequency of Anomalies by Individual Circuit Over Test Times 4-41
Table 4-30. Surface Finishing Processes Evaluated in the Cost Analysis 4-55
Table 4-31. Cost Component Categories 4-58
Table 4-32. Number of Filter Replacements by Surface Finishing Process 4-62
Table 4-33. Bath Volumes Used for Conveyorized Processes 4-64
Table 4-34. Time-Related Input Values for Non-Conveyorized Processes 4-65
Table 4-35. Time-Related Input Values for Convey orized Processes 4-65
Table 4-36. Bath Replacement Criteria for Nickel/Gold Processes 4-66
Table 4-37. Frequency and Duration of Bath Replacements for Non-Conveyorized
Nickel/Gold Process 4-67
Table 4-38. Production Time and Down Time for the Surface Finishing Processes to
Produce 260,000 ssf of PWB 4-68
Table 4-39. BOA for Transportation of Chemicals to the Surface Finishing
Process Line 4-70
Table 4-40. Costs of Critical Tasks 4-71
Table 4-41. Materials Cost for the Non-Conveyorized Nickel/Gold Process 4-76
Table 4-42. Chemical Cost per Bath Replacement for One Product Line of the
Non-Conveyorized Nickel/Gold Process 4-76
Table 4-43. Tiered Cost Scale for Monthly Wastewater Discharges to a POTW 4-79
Table 4-44. Summary of Costs for the Non-Conveyorized Nickel/Gold Process 4-83
Table 4-45. Total Cost of Surface Finishing Technologies 4-84
Table 4-46. Surface Finishing Alternative Unit Costs for Producing 260,000 ssf
of PWB 4-86
Table 4-47. CWA Regulations That May Apply to Chemicals in the Surface
Finishing Process 4-89
Table 4-48. Printed Circuit Board Facilities Discharging Less than 38,000 Liters per
Day PSES Limitations (mg/L) 4-91
Table 4-49. Printed Circuit Board Facilities Discharging 38,000 Liters per Day or
More PSES Limitations (mg/L) 4-91
Table 4-50. PSNS for Metal Finishing Facilities 4-92
Table 4-51. Amenable Cyanide Limitation Upon Agreement 4-92
Table 4-52. PSES for All Plants Except Job Shops and Independent PWB
Manufacturers 4-92
Xll
-------
Table 4-53. CAA Regulations That May Apply to Chemicals in the Surface
Finishing Process 4-93
Table 4-54. CERCLA RQs That May Apply to Chemicals in the Surface Finishing
Process 4-97
Table 4-55. SARA and EPCRA Regulations That May Apply to Chemicals in the
Surface Finishing Process 4-99
Table 4-56. TSCA Regulations and Lists That May Apply to Chemicals Used
in Surface Finishing Processes 4-100
Table 4-57. Summary of Regulations that May Apply to Chemicals Used in
Hot Air Solder Leveling (HASL) Technology 4-102
Table 4-58. Summary of Regulations that May Apply to Chemicals Used in
Nickel/Gold Technology 4-103
Table 4-59. Summary of Regulations that May Apply to Chemicals Used in
Nickel/Palladium/Gold Technology 4-104
Table 4-60. Summary of Regulations that May Apply to Chemicals Used in
OSP Technology 4-105
Table 4-61. Summary of Regulations that May Apply to Chemicals Used in
Immersion Silver Technology 4-106
Table 4-62. Summary of Regulations that May Apply to Chemicals Used in
Immersion Tin Technology 4-107
Table 5-1. Effects of Surface Finishing Technology on Resource Consumption 5-2
Table 5-2. Normalized Water Flow Rates of Various Water Rinse Types 5-4
Table 5-3. Rinse Water Consumption Rates and Total Water Consumed by
Surface Finishing technologies 5-5
Table 5-4. Metal Deposition Rates and Total Metal Consumed by Surface
Finishing Technologies 5-7
Table 5-5. Energy-Consuming Equipment Used in Surface Finishing Process Lines ... 5-13
Table 5-6. Number of Surface Finishing Process Stages that Consume Energy by
Function of Equipment 5-14
Table 5-7. Energy Consumption Rates for Surface Finishing Equipment 5-15
Table 5-8. Hourly Energy Consumption Rates for Surface Finishing Technologies .... 5-16
Table 5-9. Energy Consumption Rate per ssf of PWB Produced for Surface
Finishing Technologies 5-17
Table 5-10. Effects of Automation on Energy Consumption for Surface Finishing
Technologies 5-18
Table 5-11. Pollution Resulting From the Generation of Energy Consumed by
Surface Finishing Technologies 5-20
Table 5-12. Pollutant Environmental and Human Health Concerns 5-21
Table 6-1. Management and Personnel Practices Promoting Pollution Prevention 6-4
Table 6-2. Materials Management and Inventory Control Pollution Prevention
Practices 6-6
Table 6-3. Pollution Prevention Practices to Reduce Bath Contaminants 6-9
Table 6-4. Methods for Reducing Chemical Bath Drag-Out 6-11
Table 6-5. Bath Maintenance Improvement Methods to Extend Bath Life 6-12
Table 6-6. Typical Value of Reclaimed Metals (1999) and Recovery Methods 6-26
Table 6-7. Applicability of Recovery/Reclamation Technologies by Bath Type 6-28
Xlll
-------
Table 6-8. Treatment Chemicals Used to Remove Metals From Chelated Wastewater . . 6-34
Table 6-9. Treatment Profile of PWB Surface Finishing Process Baths 6-36
Table 7-1. Surface Finishing Processes Evaluated in the CTSA 7-1
Table 7-2. Surface Finishing Chemicals of Concern for Potential Occupational
Inhalation Risk 7-4
Table 7-3. Chemicals of Concern for Potential Dermal Risks 7-5
Table 7-4. Aquatic Risk of Non-Metal Chemicals of Concern 7-7
Table 7-5. Chemical Hazards 7-8
Table 7-6. Cost of Surface Finishing Technologies 7-11
Table 7-7. Regulatory Status of Surface Finishing Technologies 7-13
Table 7-8. Energy and Water Consumption Rates of Surface Finishing Alternatives . . . 7-14
Table 7-9. Glossary of Benefits/Costs Analysis Terms 7-17
Table 7-10. Overview of Potential Private and External Benefits or Costs 7-20
Table 7-11. Overall Cost Comparison, Based on Manufacturing 260,000 ssf 7-21
Table 7-12. Summary of Occupational Hazards, Exposures, and Risks of
Potential Concern 7-22
Table 7-13. Potential Health Effects Associated with Surface Finishing Chemicals of
Concern 7-23
Table 7-14. Number of Chemicals with Estimated Surface Water Concentration
Above Concern Concentration 7-26
Table 7-15. Energy and Water Consumption of Surface Finishing Technologies 7-27
Table 7-16. Examples of Private Costs and Benefits Not Quantified 7-28
Table 7-17. Summary of Human Health and Environmental Risk Concerns for the
HASL Technology 7-31
Table 7-18. Number of HASL Chemicals Subject to Applicable Federal Regulations .... 7-34
Table 7-19. Summary of Human Health and Environmental Risk Concerns for
the Nickel/Gold Technology 7-35
Table 7-20. Number of Nickel/Gold Chemicals Subject to Applicable Federal
Regulations 7-39
Table 7-21. Summary of Human Health and Environmental Risk Concerns for the
Nickel/Palladium/Gold Technology 7-40
Table 7-22. Number of Nickel/Palladium/Gold Chemicals Subject to Applicable Federal
Regulations 7-43
Table 7-23. Summary of Human Health and Environmental Risk Concerns for the OSP
Technology 7-45
Table 7-24. Number of OSP Chemicals Subject to Applicable Federal Regulations 7-47
Table 7-25. Summary of Human Health and Environmental Risk Concerns for the
Immersion Silver Technology 7-48
Table 7-26. Number of Immersion Silver Chemicals Subject to Applicable Federal
Regulations 7-50
Table 7-27. Summary of Human Health and Environmental Risk Concerns for the
Immersion Tin Technology 7-51
Table 7-28. Number of Immersion Tin Chemicals Subject to Applicable Federal
Regulations 7-54
XIV
-------
List of Figures
Figure 1-1. PWBs Produced for World Market in 1998 (IPC) 1-6
Figure 1-2. Number of PWBs Produced by U.S. Manufacturers in 1998 (IPC) 1-7
Figure 2-1. Typical Process Steps for Surface Finishing Technologies 2-2
Figure 2-2. HASL Process Flow Diagram 2-6
Figure 2-3. Nickel/Gold Process Flow Diagram 2-8
Figure 2-4. Nickel/Palladium/Gold Process Flow Diagram 2-11
Figure 2-5. OSP Process Flow Diagram 2-13
Figure 2-6. Immersion Silver Process Flow Diagram 2-15
Figure 2-7. Immersion Tin Process Flow Diagram 2-16
Figure 3-1. Schematic of Overall Material Balance for Surface Finishing Technologes .... 3-4
Figure 3-2 Wastewater Treatment Process Flow Diagram 3-5
Figure 3-3. Generic HASL Process Steps and Typical Bath Sequence 3-18
Figure 3-4. Generic Nickel/Gold Process Steps and Typical Bath Sequence 3-21
Figure 3-5. Generic Nickel/Palladium/Gold Process Steps and Typical Bath Sequence . . 3-24
Figure 3-6. Generic OSP Process Steps and Typical Bath Sequence 3-27
Figure 3-7. Generic Immersion Silver Process Steps and Typical Bath Sequence 3-29
Figure 3-8. Generic Immersion Tin Process Steps and Typical Bath Sequence 3-32
Figure 3-9. Relationship Between Intake Rate and Blood-Lead Level for Both Adult
and Fetus 3-73
Figure 4-1. Boxplot Displays for HCLV PTH Measurements (volts) at Pre-Test
by Surface Finish 4-43
Figure 4-2. Boxplot Displays for HCLV PTH Post 85/85 - Pre-Test Measurements
(volts) by Surface Finish 4-43
Figure 4-3. Boxplot Displays for HCLV PTH Post TS - Pre-Test Measurements
(volts) by Surface Finish 4-44
Figure 4-4. Boxplot Displays for HCLV PTH Post MS - Pre-Test Measurements
(volts) by Surface Finish 4-44
Figure 4-5. Boxplot Displays for HCLV SMT Measurements (volts) by Surface
Finish 4-45
Figure 4-6. Boxplot Displays for HCLV PTH Post 85/85 - Pre-Test Measurements
(volts) by Surface Finish 4-45
Figure 4-7. Boxplot Displays for HCLV PTH Post TS - Pre-Test Measurements
(volts) by Surface Finish 4-46
Figure 4-8. Boxplot Displays for HCLV PTH Post MS - Pre-Test Measurements
(volts) by Surface Finish 4-46
Figure 4-9. Boxplot Displays for HF PTH 50MHz Measurements (volts) by
Surface Finish 4-47
Figure 4-10. Boxplot Displays for HF PTH 50MHz Post MS - Pre-Test Measurements
(volts) by Surface Finish 4-47
Figure 4-11. Boxplot Displays for HF PTH f(-3dB) Post MS - Pre-Test Measurements
(MHz) by Surface Finish 4-48
Figure 4-12. Boxplot Displays for HF PTH f(-40dB) Post MS - Pre-Test Measurements
(MHz) by Surface Finish 4-48
XV
-------
Figure 4-13. Boxplot Displays for HF SMT 50MHz Post Ms - Pre-Test Measurements
(dB) by Surface Finish 4-49
Figure 4-14. Boxplot Displays for HF SMT f(-3dB) Post MS - Pre-Test Measurements
(MHz) by Surface Finish 4-49
Figure 4-15. Boxplot Displays for HF PTH f(-40dB) Post MS - Pre-Test Measurements
(MHz) by Surface Finish 4-50
Figure 4-16. Boxplot Displays for HF TLC 50MHz Post MS - Pre-Test Measurements
(dB) by Surface Finish 4-50
Figure 4-17. Boxplot Displays for HF TLC 500MHz Post MS - Pre-Test Measurements
(dB) by Surface Finish 4-51
Figure 4-18. Boxplot Displays for HF TLC RNR Post MS - Pre-Test Measurements
(dB) by Surface Finish 4-51
Figure 4-19. Boxplot Displays for 10-Mil Pad Measurements (logic ohms) at Pre-Test
by Surface Finish 4-52
Figure 4-20. Boxplot Displays for 10-Mil Pad 85/85 Pre-Test Measurements (logio ohms)
by Surface Finish 4-52
Figure 4-21. Boxplot Displays for PGA-A Measurements (logio ohms) at Pre-Test by
Surface Finish 4-53
Figure 4-22. Boxplot Displays for PGA-B Measurements (logio ohms) at Pre-Test by
Surface Finish 4-53
Figure 4-23. Boxplot Displays for the Gull Wing Measurements (logio ohms) at Pre-Test
by Surface Finish 4-54
Figure 4-24. Hybrid Cost Analysis Framework 4-56
Figure 5-1. Water Consumption Rates of Surface Finishing Technologies 5-5
Figure 6-1. Solder Reclaim System Diagram 6-21
Figure 6-2. Flow Diagram of Combination Ion Exchange and Electrowinning
Recovery System for Metal Recovery 6-24
Figure 6-3. Reverse Osmosis Water Reuse System 6-25
Figure 6-4. Typical PWB Waste Treatment System 6-32
Figure 7-1. Production Costs and Resource Consumption of Conveyorized
HASL Technology 7-33
Figure 7-2. Production Costs and Resource Consumption of the Nickel/Gold
Technology 7-38
Figure 7-3. Production Costs and Resource Consumption of Nickel/Palladium/Gold
Technology 7-43
Figure 7-4. Production Costs and Resource Consumption of OSP Technology 7-46
Figure 7-5. Production Costs and Resource Consumption of Immersion Silver
Technology 7-49
Figure 7-6. Production Costs and Resource Consumption of Immersion Tin
Technology 7-54
XVI
-------
Executive Summary
The Printed Wiring Board Surface Finishes Cleaner Technologies Substitutes
Assessment: Volume 7 is a technical document that presents comparative risk, competitiveness,
and resource requirements information on six technologies for performing the surface finishing
function during printed wiring board (PWB) manufacturing. Surface finishing technologies are
used by PWB manufacturers to deposit a coating on the outside surfaces of the PWB that
provides a solderable surface for future assembly, while also protecting the surface from
contamination. The technologies evaluated include hot air solder leveling (HASL), electroless
nickel/immersion gold (nickel/gold), electroless nickel/immersion palladium/immersion gold
(nickel/palladium/gold), organic solderability preservative (OSP), immersion silver, and
immersion tin. Volume I describes the surface finishing technologies, methods used to assess the
technologies, and Cleaner Technologies Substitutes Assessment (CTSA) results. Volume II
contains appendices, including detailed chemical properties and methodology information.
Information presented in the CTSA was developed by the U.S. Environmental Protection
Agency (EPA) Design for the Environment (DfE) Printed Wiring Board (PWB) Project and the
University of Tennessee (UT) Center for Clean Products and Clean Technologies. The DfE PWB
Project is a voluntary, cooperative partnership among EPA, industry, public-interest groups, and
other stakeholders to promote implementation of environmentally beneficial and economically
feasible manufacturing technologies by PWB manufacturers. Project partners participated in the
planning and execution of this CTSA by helping define the scope and direction of the CTSA,
developing project workplans, reviewing technical information contained in this CTSA and
donating time, materials, and their manufacturing facilities for project research. Much of the
process-specific information presented here was provided by chemical suppliers for the PWB
industry, PWB manufacturers who completed project information requests, and PWB
manufacturers who volunteered their facilities for a performance demonstration of the baseline
and alternative technologies.
The CTSA is intended to provide PWB manufacturers with information that can assist
them in making decisions that incorporate environmental concerns, along with performance and
cost information, when choosing a surface finishing technology. The DfE PWB Project is
especially designed to assist PWB manufacturers who may not have the resources or expertise to
compare surface finishing technologies. The primary audience for the CTSA is environmental
health and safety personnel, chemical and equipment manufacturers and suppliers in the PWB
manufacturing industry, PWB assembly shops, community groups concerned about community
health risks, and other technically informed decision-makers.
ES-1
-------
I.
DESIGN FOR THE ENVIRONMENT PRINTED WIRING BOARD PROJECT
EPA's Design for the Environment Program
The EPA DfE Program was established by the Office
of Pollution Prevention and Toxics to use EPA's
expertise and leadership to facilitate information exchange
and research on risk reduction and pollution prevention
opportunities. DfE works on a voluntary basis with
industry sectors to evaluate the risks, performance, costs,
and resource requirements of alternative chemicals,
processes, and technologies.
Additional goals of the program include:
• Changing general business practices to incorporate
environmental concerns.
• Helping individual businesses undertake
environmental design efforts through the application
of specific tools and methods.
DfE Partners include:
• industry;
• professional institutions;
• academia;
• public-interest groups; and
• other government agencies.
The DfE PWB Project is a joint
effort of the EPA DfE Program and the
UT Center for Clean Products and Clean
Technologies in voluntary and
cooperative partnerships with the PWB
industry national trade association, the
IPC-Association Connecting Electronics
Industries (IPC); individual PWB
manufacturers and suppliers; and public-
interest organizations, including the
Silicon Valley Toxics Coalition.
In part, the project is an outgrowth
of industry studies to identify key cleaner
technology needs in electronic systems
manufacturing. These studies include
Environmental Consciousness: A
Strategic Competitiveness Issue for the
Electronics Industry (MCC, 1993), the
Electronics Industry Environmental
Roadmap (MCC, 1994), and the National
Technology Roadmap for Electronic
Interconnections (IPC, 1996). The first
two studies identified environmental issues as priority targets for improvement by industry, while
concluding that improvement would be accomplished most effectively through collaboration
with government, academia, and the public. The final study cited the development of non-
tin/lead metallic or organic coatings to retain solderability characteristics as an industry need over
the near term. The potential for improvement in these areas led EPA's DfE Program to forge the
working partnerships that resulted in the DfE PWB Project.
Since its inception in 1994, the PWB Project has fostered open and active participation in
addressing environmental challenges faced by the PWB industry. The Project also has identified,
evaluated, and disseminated information on viable pollution prevention opportunities in the
industry; conducted a study of industry pollution prevention and control practices; and
completed a study of making holes conductive alternatives, among other project efforts. Over the
long-term, the Project seeks to encourage companies to consider implementing cleaner
technologies that will improve the environmental performance and competitiveness of the PWB
industry. Toward this goal, the CTSA presents the complete set of information developed by the
Project on the risk, competitiveness (e.g., cost and performance), and resource requirements of
cleaner technologies for applying a surface finish to a PWB.
ES-2
-------
II. OVERVIEW OF SURFACE FINISHING TECHNOLOGIES
Until the late 1980s, virtually all PWB manufacturers employed a HASL process to apply
the final surface finish to PWBs. The HASL process applies a thin layer of solder to the panel
surface by submerging the panel in molten solder, then removing the excess solder with an air
knife as the panel is removed. Although the traditional HASL process is a mature technology
that produces reliable surface connections, the finish has become limiting with respect to state-of-
the-art component technology that requires special assembly. It is also a significant source of
lead consumption in the PWB manufacturing process. In recent years, the advancements in
component technology, along with public and private concerns over the use of lead, have led the
PWB industry to seek viable alternative surface finishes.
Process Description
Surface finishing processes typically consist of a series of sequential chemical processing
stages separated by water rinse tanks. The process can either be operated in a vertical, non-
conveyorized immersion-type mode, or in a horizontal, conveyorized mode. In either mode,
selected baths may be operated at an elevated temperature to facilitate required chemical
reactions, or agitated to improve contact between the panels and the bath chemistry. Agitation
methods employed by PWB manufacturers include panel agitation, air sparging, and fluid
circulation pumps.
Most process baths are followed by a water rinse tank to remove drag-out (i.e., the
clinging film of process solution covering the rack and boards when they are removed from a
tank). Rinsing is necessary to clean the surface of the rack and boards to avoid contaminating
subsequent process baths. Many PWB manufacturers employ a variety of rinse water reduction
methods to reduce rinse water usage and consequent wastewater generation rates. The nature
and quantity of wastewater generated from surface finishing process lines are discussed in
Section 3.1, Source Release Assessment, while rinse water reduction techniques are discussed in
Section 6.1, Pollution Prevention.
In the non-conveyorized mode, etched panels, covered with solder mask, are loaded onto
a rack and processed through the surface finishing process line. Racks may be manually moved
from tank to tank, or moved by a manually-controlled hoist or other means. Process tanks
usually are open to the atmosphere. To reduce volatilization of chemicals from the bath or
worker exposure to volatilized chemicals, process baths may be equipped with a local ventilation
system, such as a push-pull system, bath covers for periods of inactivity, or floating plastic balls.
Conveyorized systems typically are fully enclosed, with air emissions vented to a control
technology or to the air outside the plant.
The HASL process combines wet chemistry steps, similar to those described above, with
mechanical HASL equipment. First, panels are passed through a series of wet chemistry cleaning
and etching steps to prepare the surface of the panel for the solder. Then, the solder is applied to
the panel by dipping it into molten solder and removing the excess with high pressure air. After
leaving the HASL machine, panels are cleaned by a water-based, high pressure rinse system.
ES-3
-------
Generic Process Steps and Bath Sequences of Surface Finishing Technologies
Figure ES-1 presents the generic process steps and typical bath sequences evaluated in the
CIS A. The process baths depicted in the figure are an integration of the various products
submitted for evaluation by chemical suppliers within a technology category. For example, two
different OSP processes were submitted by chemical suppliers for evaluation in the CTSA, and
these and other suppliers offer additional OSP processes that may have slightly different bath
chemistries or bath sequences. In addition, the bath sequences (bath order and rinse tank
configuration) were aggregated from data collected from various PWB facilities using the
different surface finishing technologies. Thus, Figure ES-1 lists the types and sequences of baths
in generic process lines; however, the types and sequences of baths in actual lines may vary.
Table ES-1 presents the processes evaluated in the CTSA. These are distinguished both
by process technology and equipment configuration (non-conveyorized or conveyorized). The
non-conveyorized HASL process is the industry standard for performing the surface finishing
function and is the baseline process against which alternative technologies and equipment
configurations are compared.
Table ES-1. Surface Finishes Evaluated in the CTSA
Surface Finishing Technology
HASL (Baseline)
Nickel/Gold
Nickel/Palladium/Gold
OSP
Immersion Silver
Immersion Tin
Equipment Configuration
Non-Conveyorized
X
X
X
X
X
Conveyorized
X
X
X
X
III. CLEANER TECHNOLOGIES SUBSTITUTES ASSESSMENT METHODOLOGY
The CTSA methodology is a means of systematically evaluating and comparing human
health and environmental risk, competitiveness (e.g., performance and cost), and resource
requirements of traditional and alternative chemicals, manufacturing methods, and technologies
that can be used to perform the same function. The publication, Cleaner Technologies
Substitutes Assessment: A Methodology & Resource Guide (Kincaid et al., 1996), presents the
basic CTSA methodology in detail. Particular methods used in this assessment are described in
chapters 2 through 6 of this document, and in the appendices (Printed Wiring Board Surface
Finishes Cleaner Technologies Substitutes Assessment: Volume 2).
ES-4
-------
HASL
U. ivi;fi-«f>
E
Nickel/
Gold
Electroless
Nickel/
Palladium/
Gold
OSP
Cleaner
p MicroptrhI [' Microetch
,-„*„,„„*
|J' Air Knife
OSP
s. Electroless
AirKnite
T Dry
7 Electroless
Immersion
Silver
Immersion
Tin
L
4 Immersion | I4 Immersion Tin
Note: One or more intermediate rinse steps typically separate the process steps listed above. For simplicity, these intermediate rinse steps have
not been included in the diagram.
Figure ES-1. Typical Process Steps for Surface Finishing Technologies
ES-5
-------
Key to the successful completion of any CTSA is the active participation of
manufacturers and their suppliers. This assessment was open to any surface finishing chemical
supplier who wanted to submit a technology, provided the technology met the following criteria:
• it is an existing or emerging technology; and
• the equipment and facilities are available to demonstrate its performance.
In addition, suppliers were required to provide information about their technologies, including
complete chemical product formulation data, process schematics, process characteristics and
constraints (e.g., cycle time, bath immersion time, thickness of deposit), bath replacement criteria,
and cost information.
Issues Evaluated
The CTSA evaluated a number of issues related to the risk, competitiveness, and resource
requirements (conservation) of surface finishing technologies. These include the following:
• Risk: occupational health risks, public health risks, ecological hazards, and process safety
concerns.
• Competitiveness: technology performance, cost, and regulatory status.
• Conservation: energy and natural resource use.
Occupational and public health risk information is for chronic exposure to long-term, day-
to-day releases from a PWB facility, rather than short-term, acute exposures to high levels of
hazardous chemicals as could occur with a fire, spill, or periodic release. Risk information is
based on exposures estimated for a typical, model facility, rather than exposures estimated for a
specific facility. Ecological risks are evaluated for aquatic organisms that could be exposed to
surface finishing chemicals in wastewater discharges. Process safety concerns are summarized
from material safety data sheets (MSDSs) for the technologies and process operating conditions.
Technology performance is based on a snapshot of the performance of the surface
finishing technologies at volunteer test sites in the United States and abroad. Panels were tested
under accelerated aging conditions (three weeks of 85 °C/85 percent humidity), followed by
thermal shock testing, and mechanical shock testing to distinguish variability in the performance
of the surface finish. Comparative costs of the surface finishing technologies were estimated with
a hybrid cost model that combines traditional costs with simulation modeling and activity-based
costs. Costs are presented in terms of dollars per surface square feet (ssf) of PWB produced.
Federal environmental regulatory information is presented for the chemicals in the surface
finishing technologies. This information is intended to provide an indication of the regulatory
requirements potentially associated with a technology, but not to serve as regulatory guidance.
Quantitative resource consumption data are presented for the comparative rates of energy
and water use of the surface finishing technologies. The consumption of other non-renewable
resources such as process chemicals and metals also are analyzed.
ES-6
-------
Finally, a socio-economic costs and benefits analysis of the operation of the surface
finishing process line is presented for each of the process alternatives. The private costs and
benefits to the manufacturer resulting from the use of a technology, as well as the external costs
and benefits to workers and the community are evaluated quantitatively or qualitatively.
Data Collection
Determining the risks of the baseline and alternative surface finishing technologies
required information on the chemical products for each process. Chemical information provided
by chemical suppliers included the following publicly-available sources of information: MSDSs
for the chemical products in their surface finishing technology lines and Product Data Sheets,
which are technical specifications prepared by suppliers for PWB manufacturers that describe
how to mix and maintain the chemical baths. Suppliers also were asked to provide the identities
and concentrations of proprietary chemical ingredients to the project.
Data Collection Forms
Appendix A in Volume II of the CIS A presents data collection forms used by the project,
including the following:
• The PWB Workplace Practices Questionnaire, which requested detailed information on
facility size, process characteristics, chemical consumption, and worker activities related
to chemical exposure, water consumption, and wastewater discharges.
• The Facility Background Information Sheet (developed from the PWB Workplace
Practices Questionnaire) which was sent to PWB facilities participating in the
Performance Demonstration prior to their surface technology test date. This sheet
requested detailed information on facility and process characteristics, chemical
consumption, and worker activities related to chemical exposure, water consumption, and
wastewater discharges.
• The Observer Data Sheet, which was used by an on-site observer to collect data during
the Performance Demonstration. In addition to ensuring that the performance test was
performed according to the agreed-upon test protocol, the on-site observer collected
measured data, such as bath temperature and process line dimensions, and checked
survey data for accuracy.
• The Supplier Data Sheet, which included information on chemical cost, equipment cost,
water consumption rates, product constraints, and the locations of test sites for the
Performance Demonstration.
Chemical Information
Appendix B presents chemical properties and selected environmental fate properties for
the non-proprietary chemicals in surface finishing chemical products. Proprietary chemical
ingredients are not included to protect proprietary chemical identities. Properties that were
measured or estimated (using a variety of standard EPA methods) included melting point,
solubility, vapor pressure, octanol-water partition coefficient, boiling point, and flash point.
ES-7
-------
These properties can be used to determine the environmental fate of the surface finishing
chemicals when they are released to the environment.
Health Hazard Assessments
Inherent in determining the risk associated with the surface finishing chemicals is a
determination of the hazard or toxicity of the chemicals. Human health hazard information for
non-proprietary chemicals is presented in Section 3.3. Detailed toxicity data for proprietary
chemicals are not included to maintain the secrecy of the proprietary chemical formulations.
Many of the chemicals in the surface finishing chemical products have been studied to determine
their health effects, and data from those studies are available in published scientific literature. In
order to collect available testing data for the surface finishing chemicals, literature searches were
conducted using standard chemical references and online databases, including EPA's Integrated
Risk Information System (IRIS) and the National Library of Medicine's Hazardous Substances
Data Bank (HSDB).
For many of the chemicals, EPA has identified chemical exposure levels that are known
to be hazardous if exceeded or met (e.g., no- or lowest-observed-adverse-effect level [NOAEL or
LOAEL]), or levels that are protective of human health (reference concentration [RfC] or
reference dose [RfD]). These values were taken from online databases and published literature.
For many of the chemicals lacking toxicity data, EPA's Structure-Activity Team (SAT) estimated
human health concerns based on analogous chemicals. Hazard information is combined with
estimated exposure levels to develop an estimate of the risk associated with each chemical.
Ecological Hazard Assessments
Similar information was gathered on the ecological effects that may be expected if surface
finishing chemicals are released to water. Acute and chronic toxicity values were taken from
online database searches (TOXNET and ACQUIRE), published literature, or were estimated
using structure-activity relationships if measured data were not available. Based on the toxicity
values, surface finishing chemicals were assigned concern concentrations (CCs). A CC is the
concentration of a chemical in the aquatic environment which, if exceeded, may result in
significant risk to aquatic organisms. CCs were determined by dividing acute or chronic toxicity
values by an assessment factor (ranging from one to 1,000) that incorporates the uncertainty
associated with toxicity data.
Limitations
There are a number of limitations to the project, both because of the limit of the project's
resources, the predefined scope of the project, and uncertainties inherent to risk characterization
techniques. Some of the limitations related to the risk, competitiveness, and conservation
components of the CTSA are summarized below. More detailed information on limitations and
uncertainties for a particular portion of the assessment is given in the applicable sections of this
document. A limitation common to all components of the assessment is that the surface
finishing chemical products assessed in this report were voluntarily submitted by participating
suppliers and may not represent the entire surface finishing technology market.
ES-8
-------
Risk Screening and Comparison
The risk screening and comparison is a screening level assessment of multiple chemicals
used in surface finishing technologies. The focus of the risk characterization is chronic (long-
term) exposure to chemicals that may cause cancer or other toxic effects rather than acute
toxicity from brief exposures to chemicals. The exposure assessment and risk characterization
use a "model facility" approach, with the goal of comparing the exposures and health risks of the
surface finishing process alternatives to the baseline non-conveyorized HASL technology.
Characteristics of the model facility were aggregated from questionnaire data, site visits, and
other sources. This approach does not result in an absolute estimate or measurement of risk.
The estimates of exposure and risk reflect only a portion of the potential exposures within
a PWB manufacturing facility. Many of the chemicals found in surface finishing technologies
also may be present in other process steps of PWB manufacturing, and other risk concerns for
human health and the environment may occur from these other process steps. Incremental
reduction of exposures to chemicals of concern from a surface finishing process, however, will
reduce cumulative exposures from all sources in a PWB facility. Uncertainties and key
assumptions are described further in Chapter 3, Risk Screening and Comparison.
Competitiveness
The Performance Demonstration was designed to provide a snapshot of the performance
of different surface finishing technologies. The test methods used to evaluate performance were
intended to indicate characteristics of a technology's performance, not to define parameters of
performance or to substitute for thorough on-site testing. Because the test sites were not chosen
randomly, the sample may not be representative of all PWB manufacturing facilities in the United
States (although there is no specific reason to believe they are not representative).
The cost analysis presents comparative costs of using a surface finishing technology in a
model facility to produce 260,000 ssf of PWB. As with the risk characterization, this approach
results in a comparative evaluation of cost, not an absolute evaluation or determination. The cost
analysis focuses on the private costs that would be incurred by facilities implementing a
technology. However, the analysis is limited to costs that are solely attributable to the surface
finishing process and does not evaluate costs associated with product quality or wastewater
treatment. Community benefits or costs, such as reduced health effects to workers or the effects
on jobs from implementing a more efficient surface finishing technology, also are not quantified.
The Social Benefits/Costs Assessment (see Section 7.2), however, qualitatively evaluates some of
these external benefits and costs.
The regulatory information contained in the CTSA may be useful in evaluating the
benefits of implementing processes which no longer contain chemicals that trigger compliance
issues; however, this document is not intended to provide compliance assistance. If the reader
has questions regarding compliance concerns, they should contact their federal, state, or local
authorities.
ES-9
-------
Conservation
The analysis of energy and water consumption is also a comparative analysis, rather than
an absolute evaluation or measurement. Similar to the cost analysis, consumption rates were
estimated based on using a surface finishing technology in a model facility to produce 260,000 ssf
ofPWB.
IV. CLEANER TECHNOLOGIES SUBSTITUTES ASSESSMENT RESULTS
Occupational Exposures and Health Risks
Health risks to workers are estimated for inhalation exposure to vapors and aerosols from
surface finishing baths and for dermal exposure to surface finishing bath chemicals. Inhalation
exposure estimates are based on the assumptions that emissions to indoor air from conveyorized
lines are negligible, that the air in the process room is completely mixed and chemical
concentrations are constant over time, and that no vapor control devices (e.g., bath covers) are
used in non-conveyorized lines. Dermal exposure estimates are based on the conservative
assumptions that workers do not wear gloves and that all non-conveyorized lines are operated by
manual hoist. Dermal exposure to line operators on non-conveyorized lines is estimated for
routine line operation and maintenance (e.g., bath replacement, filter replacement), and on
conveyorized lines for bath maintenance activities alone.
Based on the number of chemicals with risk results above concern levels, some
alternatives to the non-conveyorized HASL process appear to pose lower occupational risks
(immersion silver, conveyorized and non-conveyorized immersion tin, and conveyorized HASL),
some may pose similar levels of risk (conveyorized and non-conveyorized OSP), and some may
pose higher risk (nickel/gold and nickel/palladium/gold). Surface finishing chemicals of concern
for potential occupational risk from inhalation are shown in Table ES-2.
There also are occupational risk concerns for dermal contact with chemicals in the non-
conveyorized HASL, nickel/gold, nickel/palladium/gold, OSP, and immersion tin processes, and
the conveyorized HASL and OSP processes. Table ES-3 presents chemicals of concern for
potential occupational risk from dermal contact.
ES-10
-------
Table ES-2. Surface Finishing Chemicals of Concern for Potential
Occupational Inhalation Risk
Chemical
Alkyldiol
Ethylene glycol
Hydrochloric acid
Hydrogen peroxide
Nickel sulfate
Phosphoric acid
Propionic acid
Process (Non-Conveyorized, 260,000 ssf) a
HASL
X
Nickel/Gold
X
X
X
X
X
Nickel/Palladium/Gold
X
X
X
X
X
X
OSP
X
a Non-conveyorized immersion silver process not evaluated. Occupational exposure and risk from all conveyorized
process configurations are below concern levels.
X Line operator risk results above concern levels (non-cancer health effects).
Table ES-3. Chemicals of Concern for Potential Dermal Risks
Chemical
Ammonia compound A
Ammonium chloride
Ammonium hydroxide
Copper ion
Copper salt C
Copper sulfate pentahydrate
Ethylene glycol monobutyl ether
Hydrogen peroxide
Inorganic metallic salt B
Lead
Nickel sulfate
Urea compound C
HASL
(NC)
XX
t
HASL
(C)
XX
t
Nickel/Gold
(NC)
X
X
XX
X
XX
XX
Nickel/
Palladium/Gold
(NC)
X
X
XX
X
XX
XX
OSP
(NC)
XX
XX
XX
OSP
(C)
XX
X
XX
Immersion
Tin
(NC)
X
X
Note: No risk results were above concern levels for the conveyorized immersion silver or conveyorized immersion
tin processes.
X Line operator risk results above concern levels (non-cancer health effects).
XX Line operator and laboratory technician risk results above concern levels (non-cancer health effects).
f: Risk indicators were not calculated for lead as with the other chemicals (see Section 3.4.6). Other information,
however, indicates that incidental ingestion of lead from contact with hands could result in lead exposure at levels of
concern.
C: Conveyorized (horizontal) process configuration.
NC: Non-conveyorized (vertical) process configuration.
ES-11
-------
The non-conveyorized nickel/gold process contains the only chemical for which an occupational
cancer risk has been estimated (inorganic metallic salt A). The line operator inhalation exposure
estimate for inorganic metallic salt A results in an estimated upper bound excess individual life
time cancer risk of 2 x 10"7 (one in five million) based on high end exposure. Cancer risks less
than 1 x 10"6 (one in one million) are generally considered to be of low concern. Risks to other
types of workers1 were assumed to be proportional to the average amount of time spent in the
process area, which ranged from 12 to 69 percent of the risk for a line operator.
Other identified chemicals in the surface finishing processes are suspected or known
carcinogens. Lead and thiourea have been determined by the International Agency for Research
on Cancer (IARC) to be possible human carcinogens (IARC Group 2B); lead has also been
classified by EPA as a probable human carcinogen (EPA Class B2). Lead is used in tin-lead
solder in the HASL process. Thiourea is used in the immersion tin process. Urea compound B, a
confidential ingredient in the nickel/gold and nickel/palladium/gold processes, is possibly
carcinogenic to humans. Exposure for workers from these chemicals has been estimated, but
cancer potency and cancer risks are unknown. Additionally, strong inorganic and acid mists of
sulfuric acid have been determined by IARC to be a human carcinogen (IARC Group 1).
Sulfuric acid is used in diluted form in every surface finishing process in this evaluation. It is not
expected, however, to be released to the air as a strong acid mist. There are potential cancer risks
to workers from these chemicals, but because there are no slope factors, the risks cannot be
quantified.
For non-cancer risk, risk indicators exceeding concern levels - a hazard quotient (HQ)
greater than one, a margin of exposure (MOE) based on NOAEL lower than 100, or MOE based
on a LOAEL lower than 1,000 - were estimated for occupational exposures to chemicals in the
non-conveyorized and conveyorized HASL processes, non-conveyorized nickel/gold process,
non-conveyorized nickel/palladium/gold process, non-conveyorized and conveyorized OSP
processes, and the non-conveyorized immersion tin process.
Based on calculated occupational exposure levels, there may be adverse health effects to
workers exposed to chemicals with a HQ exceeding 1.0 or an MOE less than 100 or 1,000. It
should be emphasized, however, that these conclusions are based on screening level estimates.
These numbers are used here for relative risk comparisons between processes and should not be
used as absolute indicators for actual health risks to surface finishing line workers.
Worker blood-lead levels measured at one PWB manufacturing facility were below any
federal regulation or guideline for workplace exposure. Modeling data, however, indicate that
blood-lead levels could exceed recommended levels for an adult and fetus, given high incidental
ingestion rates of lead from handling solder. Although these results are highly uncertain, this
indicates the need for good personal hygiene for HASL line operators, especially wearing gloves
and hand washing to prevent accidental hand-to-mouth ingestion of lead.
1 These include laboratory technicians, maintenance workers, and wastewater treatment operators. Other types of
workers may be present for shorter or longer times.
ES-12
-------
Public Exposures and Health Risks
Potential public health risks was estimated for inhalation exposure for the general public
living near a PWB facility. Public exposure estimates are based on the assumption that emissions
from both conveyorized and non-conveyorized process configurations are vented to the outside.
The risk indicators for ambient exposures to humans, although limited to airborne releases,
indicate low concern for nearby residents. The upper bound excess individual cancer risk for
nearby residents from inorganic metallic salt A in the non-conveyorized nickel/gold process was
estimated to be from approaching zero to 2 x 10"11 (one in 50 billion). This chemical has been
classified as a human carcinogen.2 All hazard quotients are less than one for ambient exposure to
the general population, and all MOEs for ambient exposure are greater than 1,000 for all
processes, indicating low concern from the estimated air concentrations for chronic non-cancer
effects.
Estimated ambient air concentrations of lead from a HASL process are well below EPA
air regulatory limits for lead, and risks to the nearby population from airborne lead are expected
to be below concern levels.
Ecological Hazards
Ecological risk indicators (RIECO) were calculated for non-metal surface finishing
chemicals that may be released to surface water. Risk indicators for metals are not used for
comparing alternatives because it is assumed that on-site treatment is targeted to remove metal so
that permitted concentrations are not exceeded. Estimated surface water concentrations for non-
metals exceeded the CC for the processes as shown in Table ES-4. CCs are discussed in more
detail in Section 3.3.3.
Table ES-4. Aquatic Risk of Non-Metal Chemicals of Concern
Chemical
1,4-Butenediol
Alkylaryl imidazole
Alkylaryl sulfonate
Hydrogen peroxide
Potassium peroxymonosulfate
HASL
(NC)
X
X
X
X
HASL
(C)
X
X
X
OSP
(NC)
X
OSP
(C)
X
Immersion Silver
(C)
X
Immersion Tin
(NC)
X
Estimated surface water concentration > CC after publicly owned treatment works (POTW) treatment.
A cancer classification of known human carcinogen has been assigned by either the EPA, IARC, and/or NTP.
Further details about the carcinogen classification are not provided in order to protect the confidential chemical
identity.
ES-13
-------
Process Safety
In order to evaluate the chemical safety hazards of the various surface finishing
technologies, MSDSs for chemical products used with each surface finish were reviewed. Table
ES-5 summarizes the hazardous properties listed on MSDSs for surface finishing chemical
products. Other potential chemical hazards posed by surface finishing chemicals include either
the hazardous decomposition of chemical products, or chemical product incompatibilities with
other chemicals or materials.
Table ES-5. Hazardous Properties of Surface Finishing Chemical Products
Process
HASLb
Nickel/Gold
Nickel/Palladium/Gold
OSP
Immersion Silver
Immersion Tin
No. ofMSDS3
33
19
18
9
4
14
Hazardous Property
F
1
1
C
E
1
1
1
FH
3
2
1
CO
4
8
12
4
2
7
O
1
1
1
1
1
SRP
1
1
1
1
U
1
1
For alternative processes with more than one product line, the hazard data reported represent the most hazardous
bath of each type for the two product lines (e.g., of the microetch baths from the two product lines, the one with the
most hazardous chemicals is reported).
b Formulations for HASL process baths were unavailable because cleaner and microetch bath chemistries are not
made specifically for the HASL process. Hazards reported for HASL bath types were reported as the worst case of the
results of similar baths from other processes.
F = Flammable; C = Combustible; E = Explosive; FH = Fire Hazard; CO = Corrosive; O = Oxidizer; SRP = Sudden
Release of Pressure; U = Unstable.
Several unique process safety concerns arise from the operation of the HASL process.
Solder eruptions during start-up can lead to solder splattering onto workers causing serious
burns. The HASL process also poses a fire hazard due to the build-up of residual carbon from
the use of oil-based flux or other flammable materials. Other safety concerns include worker
exposure to acids in the flux, accidental contact with the molten solder, or exposure to the other
chemical hazards on the process line.
Work-related injuries from equipment, improper use of equipment, bypassing equipment
safety features, failure to use personal protective equipment, and physical stresses that may
appear gradually as a result of repetitive motion are all potential process safety hazards to
workers. Reducing the potential for work-related injuries is critical in an effective and ongoing
safety training program. Appropriate training can help reduce the number of work-related
accidents and injuries regardless of the technology used.
ES-14
-------
Performance
The performance of the surface finishing technologies was tested using production run
tests following a strict testing protocol. Functional test boards were fabricated using a complex
test board design (a modified version of the IPC-B-24 board) developed by the Circuit Card
Assembly and Materials Task Force (CCAMTF). A surface finish was then applied to test boards
at each of thirteen volunteer PWB manufacturing facilities. Test boards were then collected
together and assembled at an assembly facility, using either a halide-free low-residue flux or a
halide-containing water-soluble flux, before being tested under thermal and mechanical stress,
and accelerated aging conditions. Additional residue testing was conducted to determine the
mechanism of failure.
The test vehicle measured roughly 6" x 5.8" x 0.062" and was designed to contain at least
80 percent of the circuitry used in military and commercial electronics. The test vehicle also
contained a variety of circuits, including high current low voltage (HCLV), high voltage low
current (HVLC), high speed digital (HSD), high frequency (HF), stranded wire (SW), and other
networks, which were used to measure current leakage. Overall, the vehicle provided 23 separate
electrical responses for testing the performance of the surface finish. Types of electrical
components in the HCLV, HVLC, HSD, and HF circuits included both plated through hole
(PTH) and surface mounted components.
Test sites were submitted by suppliers of the technologies, and included production
facilities and supplier testing facilities. Because the test sites were not chosen randomly, the
sample may not be representative of all PWB manufacturing facilities (although there is no
specific reason to believe that they are not representative). In addition, the number of test sites
for each technology ranged from one to four. Due to the smaller number of test sites for some
technologies, statistical relevance could not be determined.
The results of the performance testing showed that all of the surface finishes under study
were very robust to the environmental exposures, with two exceptions. Failures during the
mechanical shock testing, resulting in the separation of the surface mount components, were
attributable to the severity of the testing, and were spread evenly across all finishing technologies,
including the baseline HASL process. Failures in the high frequency, low pass filter circuits,
resulting from open PTH, were found to be attributable to a combination of board fabrication
materials and board design. From an overall contamination standpoint, the five non-HASL
surface finishes performed as well, if not better than the HASL finish. The few solder joint
cracking failures were greater with the HASL finish than with the alternative finishes.
Cost Analysis
Comparative costs were estimated using a hybrid cost model that combined traditional
costs with simulation modeling and activity-based costs. The cost model was designed to
determine the total cost of producing 260,000 ssf of PWB for each of the surface finishing
technologies using a model facility concept. Total costs were normalized to a cost per ssf of
PWB produced.
ES-15
-------
The cost components evaluated include capital costs (primary equipment, installation, and
facility costs), materials costs (limited to chemical costs), utility costs (water, electricity, and
natural gas costs), wastewater costs (limited to wastewater discharge cost), production costs
(production labor and chemical transport costs), and maintenance costs (tank cleanup, bath
setup, sampling and analysis, and filter replacement costs). Start-up costs for implementing a
surface finishing technology, as well as the costs of other process changes that may be required
to implement an alternative technology, were not considered in the cost evaluation. Other cost
components that contribute to overall costs, but which also could not be quantified include
quality costs, wastewater treatment cost, sludge recycling and disposal cost, and other solid waste
disposal costs.
Cost analysis results are presented in Table ES-6. With the exception of the two
technologies containing gold, an expensive precious metal, the results indicate that all of the other
surface finishing alternatives were more economical than the baseline non-conveyorized HASL
process. Three processes had a substantial cost savings of at least 50 percent of the cost per ssf
over that of the baseline HASL process (conveyorized OSP at 72 percent, non-conveyorized OSP
at 69 percent, and non-conveyorized immersion tin at 50 percent). Three other process
alternatives realized a somewhat smaller cost savings over the baseline HASL process
(conveyorized immersion tin at 31 percent, conveyorized immersion silver at 22 percent, and the
convey orized HASL process at 3 percent).
In general, conveyorized processes cost less than non-conveyorized processes. Chemical
cost was the single largest component cost for all nine of the processes, with the cost of labor a
distant second.
Regulatory Status
Discharges of surface finishing chemicals may be restricted by federal, state, or local air,
water, or solid waste regulations, and releases may be reportable under the federal Toxic Release
Inventory program. Federal environmental regulations were reviewed to determine the federal
regulatory status of surface finishing chemicals.3 Table ES-7 lists the number of chemicals used
in each surface finishing technology that are subject to federal environmental regulations.
Different chemical suppliers of a technology do not always use the same chemicals in their
particular product lines. Thus, all of these chemicals may not be present in any one product line.
Resource Conservation
Energy and water consumption rates were evaluated for each of the surface finishing
process alternatives. Other resource consumption by the surface finishing technologies was
evaluated qualitatively due to the variability of factors that affect the consumption of these
resources. Table ES-8 presents the energy and water consumption rates of the surface finishing
technologies.
3 In some cases, state or local requirements may be more restrictive than federal requirements. Due to resource
limitations, however, only federal regulations were reviewed.
ES-16
-------
Table ES-6. Cost Analyses Results a
Surface Finishing Technology
HASL, Non-conveyorized (BASELINE)
HASL, Conveyorized
Nickel/Gold, Non-conveyorized
Nickel/Palladium/Gold, Non-conveyorized
OSP, Non-conveyorized
OSP, Conveyorized
Immersion Silver, Conveyorized
Immersion Tin, Non-conveyorized
Immersion Tin, Conveyorized
Average Cost
$/ssf
$ 0.36
$ 0.35
$ 0.60
$ 1.54
$ 0.11
$ 0.10
$ 0.28
$ 0.18
$ 0.25
% change
~
-3%
67%
328%
-69%
-72%
-22%
-50%
-3 1%
Capital Cost
$/ssf
$ 0.038
$ 0.044
$ 0.039
$ 0.083
$ 0.008
$ 0.012
$ 0.044
$ 0.015
$ 0.074
% change
~
16%
4%
119%
-80%
-68%
17%
-61%
95%
Chemical Cost
$/ssf
$ 0.288
$ 0.289
$ 0.419
$ 1.235
$ 0.071
$ 0.072
$ 0.203
$ 0.112
$ 0.111
% change
-
0%
46%
329%
-75%
-75%
-29%
-61%
-61%
Water Cost
$/ssf
$ 0.003
$ 0.002
$ 0.005
$ 0.008
$ 0.002
$ 0.001
$ 0.001
$ 0.004
$ 0.003
% change
~
-20%
67%
191%
-38%
-57%
-57%
46%
-1%
Electricity Cost
$/ssf
$ 0.003
$ 0.002
$ 0.009
$ 0.016
$ 0.001
$ 0.001
$ 0.003
$ 0.002
$ 0.005
% change
~
-32%
253%
507%
-53%
-69%
11%
-26%
84%
Surface Finishing Technology
HASL, Non-conveyorized (BASELINE)
HASL, Conveyorized
Nickel/Gold, Non-conveyorized
Nickel/Palladium/Gold, Non-conveyorized
OSP, Non-conveyorized
OSP, Conveyorized
Immersion Silver, Conveyorized
Immersion Tin, Non-conveyorized
Immersion Tin, Conveyorized
Natural Gas Cost
$/ssf
$ 0.000
$ 0.000
$ 0.000
$ 0.000
$ 0.000
$ 0.000
$ 0.001
$ 0.001
$ 0.001
%
change
—
-50%
-100%
-100%
-24%
-65%
59%
82%
171%
Wastewater Cost
$/ssf
$ 0.004
$ 0.003
$ 0.008
$ 0.014
$ 0.003
$ 0.002
$ 0.002
$ 0.006
$ 0.005
%
change
~
-23%
86%
222%
-36%
-58%
-52%
47%
10%
Production Cost
$/ssf
$ 0.016
$ 0.007
$ 0.076
$ 0.101
$ 0.013
$ 0.006
$ 0.021
$ 0.027
$ 0.034
%
change
—
-53%
381%
539%
-19%
-65%
32%
70%
118%
Maintenance
Cost
$/ssf
$ 0.011
$ 0.007
$ 0.042
$ 0.080
$ 0.013
$ 0.008
$ 0.010
$ 0.015
$ 0.017
%
change
—
-36%
275%
610%
13%
-33%
-15%
28%
46%
Table lists costs and percent change in cost from the baseline.
ES-17
-------
Table ES-7. Regulatory Status of Surface Finishing Technologies
Process
Chemical
HASL
Nickel/Gold
Nickel/Palladium/Gold
OSP
Immersion Silver
Immersion Tin
Number of Chemicals Subject to Applicable Regulation
CWA
304b
1
6
5
2
1
1
307a
1
6
5
2
1
1
311
4
16
12
5
5
6
Priority
Pollutant
1
6
5
2
1
1
CAA
111
3
11
5
3
1
3
112b
3
6
5
2
1
2
112r
1
1
1
1
-
1
EPCRA
313
6
12
10
5
3
7
110
1
7
6
2
1
1
302a
3
3
3
2
3
2
TSCA
8d
HSDR
3
1
1
1
-
2
MTL
4
4
4
2
1
4
8a
PAIR
3
3
4
1
1
3
RCRA Waste
P
-
-
-
-
-
-
U
-
-
-
-
-
2
Abbreviations and definitions:
CAA - Clean Air Act
CAA 111 - Standards of Performance for New Stationary Sources of
Air Pollutants -Equipment Leaks Chemical List
CAA 112b - Hazardous Air Pollutant
CAA 112r - Risk Management Program
CWA-Clean Water Act
CWA 304b - Effluent Limitations Guidelines
CWA 307a - Toxic Pollutants
CWA 311 - Hazardous Substances
CWA - Priority Pollutants
EPCRA - Emergency Planning and Community Right-to-Know Act
EPCRA 302a - Extremely Hazardous Substances
EPCRA 313 - Toxic Chemical Release Inventory
RCRA - Resource Conservation and Recovery Act
RCRA P Waste - Listed acutely hazardous waste
RCRA U Waste - Listed hazardous waste
SARA - Superfund Amendments and Reauthorization Act
SARA 110 - Superfund Site Priority Contaminant
SDWA - Safe Drinking Water Act
SD WA NPD WR - National Primary Drinking Water Rules
SDWA NSDWR - National Secondary Drinking Water Rules
TSCA - Toxic Substances Control Act
TSCA 8d HSDR - Health & Safety Data Reporting Rules
TSCA MTL - Master Testing List
TSCA 8a PAIR - Preliminary Assessment Information Rule
ES-18
-------
The rate of water consumption is directly related to the rate of wastewater generation.
Several processes were found to consume less water than the HASL baseline, including
conveyorized versions of the immersion silver and immersion tin technologies and both versions
of the OSP process. Convey orized processes were found to consume less water than non-
conveyorized versions of the same process. Primary factors influencing the water consumption
rate included the number of rinse tanks and the overall efficiency of the convey orized processes.
Table ES-8. Energy and Water Consumption Rates of Surface Finishing Technologies
Process Type
HASL, Non-conveyorized (BASELINE)
HASL, Conveyorized
Nickel/Gold, Non-conveyorized
Nickel/Palladium/Gold, Non-conveyorized
OSP, Non-conveyorized
OSP, Conveyorized
Immersion Silver, Conveyorized
Immersion Tin, Non-conveyorized
Immersion Tin, Conveyorized
Water Consumption
(gal/ssf)
1.24
0.99
2.06
3.61
0.77
0.53
0.53
1.81
0.88
Energy Consumption
(Btu/ssf)
218
133
447
768
125
73
287
289
522
Energy consumption by the surface finishing technologies was driven primarily by the
overall throughput efficiency of the technology. Although HASL had the highest BTU per hour
rate of all the technologies, after normalizing the rate using the overall throughput (260,000 ssf),
only the OSP (conveyorized and non-conveyorized), and the Conveyorized HASL were more
energy efficient than the HASL process. It also was found that for alternatives with both types of
automation, the conveyorized version of the process is typically the more energy efficient (HASL
and OSP), with the exception of the immersion tin process.
The rate of deposition of metal was calculated for each technology along with the total
amount of metal consumed for 260,000 ssf of PWB produced. It was shown that the
consumption of close to 300 pounds of lead (per 260,000 ssf) could be eliminated by replacing
the baseline HASL process with an alternative technology (see Section 5.1, Resource
Conservation). In cases where waste solder is not routinely recycled or reclaimed, the
consumption of as much as 2,500 pounds of lead (per 260,000 ssf) could be eliminated by
replacement of the HASL process. Although several of the alternative technologies rely on the
use of small quantities of other metals (especially nickel, palladium, gold, silver, and tin) the OSP
technology eliminates metal consumption entirely.
ES-19
-------
Social Benefits/Costs Assessment
The social benefits and costs of the surface finishing technologies were qualitatively
assessed to compare the benefits and costs of switching from the baseline technology to an
alternative, while considering both the private and external costs and benefits. Private costs
typically include any direct costs incurred by the decision-maker and are generally reflected in the
manufacturer's balance sheet. By contrast, external costs are not borne by the manufacturer, but
by society. Therefore, the analysis considered both the impact of the alternative surface finishing
processes on the manufacturer itself (private costs and benefits) and the impact the choice of an
alternative had on external costs and benefits.
Table ES-9 presents an overview of potential private and external benefits and costs
associated with the operation of the surface finishing line. Changes in the surface finishing
technology employed could potentially result in a net benefit (a change in a beneficial direction)
or cost (a change in a detrimental direction) in each of the categories listed below. The type of
change and the magnitude will vary by facility.
Table ES-9. Overview of Potential Private and External Benefits or Costs
Evaluation Category
Manufacturing costs
Occupational health/
Worker risk
Public health/
Population risk
Wastewater and
Ecological risk
Energy use
Water use
Private Benefit or Cost a
Capital costs,
Materials (chemical) costs,
Utility costs,
Wastewater discharge costs,
Production costs, and
Maintenance costs.
Worker sick days, and
Health insurance costs to the PWB
manufacturer.
Potential liability costs.
Treatment costs to meet wastewater
permit requirements,
Possible fines if permits are violated,
and Increased liability costs.
Direct costs from the use of energy in
the manufacturing process.
Direct costs from the use of water in
the manufacturing process.
External Benefit or Cost a
NA
Medical costs to workers, and
Pain and suffering associated with
work-related illness.
Medical costs, and
Pain and suffering associated with
illness.
Loss of ecosystem diversity; and
Reduction in the recreational value of
streams and rivers.
Increased air emissions, and
Depletion of natural resources.
Water costs for the surrounding area,
Costs paid to treatment facilities to
clean the water,
Changes to water quality available to
society; and
Reduced water supply.
a A benefit would be a change in a beneficial direction (e.g., decreased capital costs), while a cost would be a
detrimental change (e.g., increased worker sick days).
ES-20
-------
Each alternative presents a mixture of private and external benefits and costs. In terms of
worker health risks, conveyorized processes have the greatest benefits for reduced worker
inhalation exposure to bath chemicals; they are enclosed and vented to the atmosphere.
However, dermal contact from bath maintenance activities can be of concern regardless of the
equipment configuration for HASL and OSP processes, as well as non-conveyorized nickel/gold,
nickel/palladium/gold, and immersion tin processes. Little or no improvement is seen in public
health risks because results were below concern levels for all technologies. Differences in
estimated wastewater contaminant levels and aquatic risk concerns suggest that alternatives to
non-convey orized HASL pose lower potential private and external costs (or higher benefits).
Conveyorized processes consumed less water than that consumed by non-conveyorized
processes, resulting in net private and external benefits. Only the OSP technology, along with the
conveyorized HASL technology, are expected to reduce potential private and external costs of
energy consumption, resulting in increased social benefits.
V. CONCLUSIONS
The CTSA evaluated the risk, competitiveness, and resource requirements of six
technologies for performing the surface finishing function during PWB manufacturing. These
technologies are HASL, nickel/gold, nickel/palladium/gold, OSP, immersion silver, and
immersion tin.
The results of the CTSA analyses of the surface finishing technologies were mixed.
Analyses of process costs, energy, and natural resource consumption each showed that some
alternatives performed better than HASL, while others did not. An evaluation of potential
occupational risks from both inhalation and dermal exposures indicated that several alternatives
posed lower occupational risks than HASL, based on the number of chemicals with risk results
above concern concentrations. Ecological risks posed by the alternatives were all lower than the
HASL process, also based on the number of chemicals exceeding concern concentrations. None
of the surface finishing technologies, including HASL, posed a risk to populations living nearby.
Finally, alternatives to the traditional non-conveyorized HASL technology (the baseline process)
were demonstrated to perform as well as HASL during performance testing; however, several of
the alternatives improve upon the technical limitations of the HASL finish (e.g., wire-bondability,
surface planarity).
Table ES-10 summarizes the CTSA analyses results for the surface finishing technologies,
relative to the non-conveyorized HASL baseline. It is important to note that there are additional
factors beyond those assessed in this CTSA that individual businesses may consider when
choosing among alternatives. The actual decision of whether or not to implement an alternative is
made outside of the CTSA process.
ES-21
-------
Table ES-10. Relative Benefits and Costs of Surface Finishing Alternatives Versus Baseline
Surface Finishing Technology
HASL, Non-Conveyorized (BASELINE)
HASL, Conveyorized
Nickel/Gold, Non-conveyorized
Nickel/Palladium/Gold, Non-
Conveyorized
OSP, Non-conveyorized
OSP, Conveyorized
Immersion Silver, Conveyorized
Immersion Tin, Non-conveyorized
Immersion Tin, Conveyorized
Production
Costs
($/ssf)
$0.36
=
-
~
++
++
+
+
+
Number of Chemicals of Concern
Worker Health
Risks a'b'c
Inhalatio
n
1
+
~
~
=
+
+
+
+
Dermal
2
=
~
~
-
=
+
=
+
Public
Health Risks
Inhalation
0
=
=
=
=
=
=
=
=
High Aquatic
Toxicity
Concern a
4
+
++
++
++
++
++
++
++
Water
Consumption
(gal/ssf)
1.24
+
-
~
+
++
++
-
+
Energy
Consumption
(Btu/ssf)
218
+
~
~
+
++
-
-
—
a For technologies with more than one chemical supplier (e.g., nickel/gold, OSP, immersion tin) all chemicals may not be present in any one product line.
b For the most exposed individual (e.g., a surface finishing line operator).
0 Because the risk characterization did not estimate the number of incidences of adverse health outcomes, the amount of reduced risk benefit cannot be quantified.
The comparison shown in this table is based on the number of chemicals of concern for the baseline.
Key:
Neutral, less than 10% increase or decrease from baseline.
10 to 100 percent worse.
100 percent worse.
Some benefit, 10 to 50 percent decrease from baseline.
Greater benefit, +50 percent or greater decrease from baseline.
ES-22
-------
To assist PWB manufacturers who are considering the implementation of an alternative
surface finish, the DfE PWB Project has prepared an implementation guide that describes lessons
learned by other PWB manufacturers who have begun using an alternative surface finishing
process.4 In addition, the University of Tennessee Department of Industrial Engineering can
provide technical support to facilities that would like to use the cost model developed for the
CTSA to estimate their own manufacturing costs should they switch to a surface finishing
alternative.
4 Implementing Cleaner Printed Wiring Board Technologies: Surface Finishes (EPA 744-R-00-002, March
2000). This and other DfE PWB Project documents can be obtained by contacting EPA's Pollution Prevention
Information Clearinghouse at (202) 260-1023 or from www.epa.gov/dfe/pwb.
ES-23
-------
REFERENCES
IPC (IPC-Association Connecting Electronics Industries). 1996. National Technology Roadmap
for Electronic Interconnections.
Kincaid, Lori E., Jed Meline and Gary Davis. 1996. Cleaner Technologies Substitutes
Assessment: A Methodology & Resource Guide. EPA Office of Pollution Prevention and
Toxics. Washington, D.C. EPA 744-R-95-002. December.
MCC (Microelectronics and Computer Technology Corporation). 1993. Environmental
Consciousness: A Strategic Competitiveness Issue for the Electronics and Computer Industry.
March.
MCC (Microelectronics and Computer Technology Corporation). 1994. Electronics Industry
Environmental Roadmap. December.
ES-24
-------
Acronyms
ABC activity-based costing
ADD average daily dose
ADI acceptable daily intake
ACGIH American Conference of Governmental Industrial Hygenists, Inc.
ALM Adult Lead Methodology)
ANOVA an analysis of variance
ASF alternative surface finishing
AT averaging time
ATSDR Agency for Toxic Substances and Disease Registry
BAT best available control technology economically achievable
BCP best conventional pollution control technology
EGA ball grid array
BKSF biokinetic slope factor
BOA bill of activities
BOD biological oxygen demand
BPT best practicable control technology currently available
Btu British Thermal Units
BW body weight
CAA Clean Air Act
CC concern concentration
CCAMTF Circuit Card Assembly and Materials Task Force
CDC Centers for Disease Control and Prevention
CEB Chemical Engineering Branch
CERCLA Comprehensive Environmental Response, Compensation and Liability Act
CFR Code of Federal Regulations
CO carbon monoxide
CO2 carbon dioxide
COB chip on board
CSL Contamination Studies Laboratories, Inc.
CTSA Cleaner Technologies Substitutes Assessment
CuSO4 copper sulfate
CWA Clean Water Act
DCT dimethyl-dithiocarbamate
DfE Design for the Environment
ED exposure duration
EDTA ethylenediaminetetraacedic acid
EF exposure frequency
EMS Environmental Management System
EPA Environmental Protection Agency
EPCRA Emergency Planning and Community Right-to-Know Act
HCLV high current low voltage
HVLC high voltage low current
HSD high speed digital
XVll
-------
HF LPF high frequency low pass filter
HF TLC high frequency transmission line coupler
g gram
gal gallon
GI gastrointestinal
GLM general linear models
gpm gallons per minute
GRAS generally recognized as safe
GSD geometric standard deviation
gpm gallons per minute
H2SO4 sulfuric acid
HASL hot air solder leveling
Hc Henry' s Law Constant
HCLV high current low voltage
HEAST Health Effects Assessment Summary Tables
HF high frequency
HQ hazard quotient
HSD high speed digital
HSDB Hazardous Substances Data Bank
HVLC high voltage low current
IARC International Agency for Research on Cancer
IEUBK integrated exposure uptake biokinetic model for lead in Children
IPC Institute for Interconnecting and Packaging Electronics Circuits
IQR interquartile range
IR intake rate
IRIS Integrated Risk Information System
ISO International Standards Organization
KUB Knoxville Utilities Board
kW kilowatt
LADD lifetime average daily dose
LOAEL lowest-observed-adverse-effect level
LR low residue
LRSTF Low-Residue Soldering Task Force
LSD least significant difference
LT lab technical
MACT Maximum Achievable Control Technology
MHC making holes conductive
MnO2 manganese dioxide
MOE margin of exposure
MRL minimum risk level
MSDS material safety data sheet
MTL Master Testing List
NaOH caustic soda
NCP National Contingency Plan
NIOSH National Institute for Occupational Safety and Health
NOAEL no-observed-adverse-effect level
XVlll
-------
NOx oxides of nitrogen
NPDES National Pollutant Discharge Elimination System
NPDWR National Primary Drinking Water Regulations
NSDWR National Secondary Drinking Water Regulations
NSPS New Source Performance Standards
NTP National Toxicology Program
ON other networks
OSP organic solderability preservative
OSHA Occupational Safety and Health Administration
PEL permissible exposure limit
PDR potential dose rate
POTW publicly owned treatment work
PPE personal protective equipment
PSES pretreatment standards for existing xources
PSNS pretreatment standards for new sources
psi per square inch
PTH plated-through holes
PWAs printed wiring assemblies
PWB printed wiring board
RCRA Resource Conservation and Recovery Act
REL recommended exposure level
RfC reference concentration
RfD reference dose
RO reverse osmosis
RTECS Registry of Toxic Effects of Chemical Substances
RQ reportable quantity
SAR structure-activity relationship
SARA Superfund Amendments and Reauthorization Act
SAT Structure-Activity Team
SDWA Safe Drinking Water Act
SERC State Emergency Response Commission
SF slope factor
SIC standard industrial code
SMT surface mount technology
SOX sulfur oxides
SPC statistical process control
ssf surface square feet
STEL short-term exposure limit
SW stranded wire
TMT15 tri-mercaptotriazine
TLC transmission line coupler
TLV threshold limit value
TPY tons per year
TRI Toxic Release Inventory
TS thermal shock
TSCA Toxic Substances Control Act
xix
-------
TSDS treatment, storage, or disposal facility
TWA time-weighed average
UC unit cost
UF uncertainty factor
UT University of Tennessee
UR utilization ratio
VOC volatile organic compounds
WHO World Health Organization
WOA weak organic acid
WOE weight-of-evidence
WS water soluble
XX
-------
Chapter 1
Introduction
This document presents the results of a cleaner technologies substitutes assessment
(CTSA) of six technologies for performing the surface finishing function during the manufacture
of printed wiring boards (PWBs). Surface finishing technologies deposit a coating on the outside
surfaces of the PWB that provides a solderable surface for future assembly, while protecting the
surface from exposure to the local environment. The technologies evaluated in the study are hot
air solder leveling (HASL), electroless nickel/immersion gold (nickel/gold), electroless
nickel/electroless palladium/immersion gold (nickel/palladium/gold), organic solderability
preservative (OSP), immersion silver, and immersion tin.
For the purposes of this evaluation, the non-conveyorized HASL process is considered
the baseline process against which alternative technologies and equipment configurations (i.e.,
non-conveyorized or conveyorized) are compared. This CTSA is the culmination of over two
years of research by the U.S. Environmental Protection Agency (EPA) Design for the
Environment (DfE) PWB Project and the University of Tennessee (UT) Center for Clean
Products and Clean Technologies on the comparative risk, performance, cost, and natural
resource requirements of the alternative technologies as compared to the baseline.
The DfE PWB Project is a voluntary, cooperative partnership among EPA, industry,
public-interest groups, and other stakeholders to promote implementation of environmentally
beneficial and economically feasible manufacturing technologies by PWB manufacturers.
Project partners participated in the planning and execution of this CTSA by helping define the
scope and direction of the CTSA, developing project workplans, donating time, materials, and
their manufacturing facilities for project research, and reviewing technical information contained
in this CTSA. Much of the process-specific information presented here was provided by
chemical suppliers to the PWB industry, PWB manufacturers who responded to project
information requests, and PWB manufacturers who volunteered their facilities for a performance
demonstration of the baseline and alternative technologies.
Section 1.1 presents project background information, including summary descriptions of
the EPA DfE Program and the DfE PWB Project. Section 1.2 is an overview of the PWB
industry, including the types of PWBs produced, the market for PWBs, and the overall PWB
manufacturing process. Section 1.3 summarizes the CTSA methodology, including a discussion
of how technologies were selected for evaluation in the CTSA, the boundaries of the evaluation,
issues evaluated, data sources, and project limitations. Section 1.4 describes the organization of
the remainder of the CTSA document.
1-1
-------
1.1 PROJECT BACKGROUND
The PWB is the connector between the semiconductors, computer chips, and other
electronic components that form an electronic circuit. Therefore, PWBs are an irreplaceable part
of many "high-tech" products in the electronics, defense, communications, and automotive
industries. PWB manufacturing, however, typically generates a significant amount of hazardous
waste, requires a substantial amount of water and energy, and uses chemicals that may pose
environmental and health risks.
To address these issues, the PWB industry has been actively seeking to identify and
evaluate cleaner technologies and pollution prevention opportunities. However, many PWB
manufacturers do not have the resources or experience to independently develop the data needed
to evaluate new technologies and redesign their processes. The DfE PWB Project was initiated to
develop that data, by forming partnerships between the EPA DfE Program, the PWB industry,
and other interested parties to facilitate the evaluation and implementation of alternative
technologies that reduce health and environmental risks and production costs. The EPA DfE
Program and the DfE PWB Project are discussed in more detail below.
1.1.1 EPA DfE Program
EPA's Office of Pollution Prevention and Toxics created the DfE Program in 1991. The
Program uses EPA's expertise and leadership to facilitate information exchange and research on
risk reduction and pollution prevention opportunities. DfE works on a voluntary basis with
industry sectors to evaluate the risks, performance, costs, and resource requirements of
alternative chemicals, processes, and technologies. Additional goals of the program include:
• changing general business practices to incorporate environmental concerns, and
• helping individual businesses undertake environmental design efforts through the
application of specific tools and methods.
The DfE Program encourages voluntary environmental improvement through stakeholder
partnerships. DfE partners include industry, trade associations, research institutions,
environmental and public-interest groups, academia, and other government agencies. By
involving representatives from each of these stakeholder groups, DfE projects gain the necessary
expertise to perform the project's technical work and improve the quality, credibility, and utility
of the project's results.
1.1.2 DfE PWB Project
The DfE PWB Project is a voluntary, cooperative partnership among EPA, industry,
public-interest groups, and other stakeholders to promote implementation of environmentally
beneficial and economically feasible manufacturing technologies by PWB manufacturers. In
part, the project is an outgrowth of industry efforts to identify key cleaner technology needs in
1-2
-------
electronics manufacturing. The results of these industry studies are presented in two reports
prepared by Microelectronics and Computer Technology Corporation (MCC), an industry
research consortium: Environmental Consciousness: A Strategic Competitiveness Issue for
the Electronics and Computers Industry (MCC, 1993) and Electronics Industry Environmental
Roadmap (MCC, 1994).
The first study identified wet chemistry processes used in PWB fabrication as water- and
energy-intensive processes that generate significant amounts of hazardous waste. The study
concluded that effective collaboration among government, industry, academia, and the public is
imperative to proactively address the needs of environmental technologies, policies, and practices
(MCC, 1993). To follow-up, the industry embarked on a collaborative effort to develop an
environmental roadmap for the electronics industry. The roadmap project involved more than
100 organizations, including EPA, the Department of Energy, the Advanced Research Projects
Agency, and several trade associations. The PWB industry national trade association, the IPC-
Association Connecting Electronics Industries (IPC), was instrumental in developing the
information on PWBs through its Environmental, Health, and Safety Committee.
The highest priority need identified for PWB manufacturers was for more efficient use,
regeneration, and recycling of hazardous wet chemistries. One proposed approach to meet this
need was to eliminate formaldehyde from materials and chemical formulations by researching
alternative chemical formulations. Another priority need was for industry to reduce water
consumption and discharge, which can be accomplished by using wet chemistries that have
reduced numbers of rinse steps. The electroless copper technologies for making holes conductive
(MHC) use formaldehyde and consume large amounts of water.
The potential for improvement in these areas led EPA's DfE Program to forge working
partnerships with IPC, individual PWB manufacturers and suppliers, research institutions such as
MCC and UT's Center for Clean Products and Clean Technologies, and public-interest
organizations, including the Silicon Valley Toxics Coalition. These partnerships resulted in the
DfE PWB Project.
Since its inception in 1994, the goal of the DfE PWB Project has been the identification
and evaluation of environmentally preferable alternative technologies for the PWB manufacturing
industry. The project initially focused on the evaluation of alternative technologies for the MHC
process. Seven MHC processes were evaluated for performance, cost, and their impact on
human health and the environment. The project results are published in the Printed Wiring
Board Cleaner Technologies Substitutes Assessment: Making Holes Conductive (U.S. EPA,
1998a).
The success of the MHC study led project partners to explore the possibility of a second
project with the PWB manufacturing industry. Results of the environmental roadmap from 1994
identified a top priority need for PWB manufacturers as the need to minimize the impact of
hazardous materials use in PWB fabrication. One proposed approach to meet this need was to
eliminate or reduce lead solder use when possible by validating the quality of lead plating
alternatives. Another priority need for the industry was to establish better supplier relationships
to enhance the development and acceptance of environmentally preferable materials.
1-3
-------
As a follow up to the environmental roadmap, the electronics industry embarked on a
study of industry technology trends, the results of which were published as The National
Technology Roadmap for Electronic Interconnections (IPC, 1996). The roadmap detailed
trends in PWB manufacturing and assembly technologies, and forecasted the technology needs
for the industry over the immediate future. The study concluded that major efforts are needed to
overcome the reluctance to trying new and innovative ideas, citing the environmental pressure to
reduce hazardous waste and the use of lead. The results also cited the development of non-
tin/lead metallic or organic coatings to retain solderability characteristics as an industry need over
the near term.
Recognizing the importance of reducing lead consumption in the PWB industry, and
building on the strong partnerships established during the previous work, the PWB surface
finishing project was begun in 1997 to evaluate alternative surface finishing technologies to
HASL. This CTSA is a culmination of this effort. During this time, the project has also:
• Prepared several additional case studies of pollution prevention opportunities (U.S. EPA,
1997a; U.S. EPA, 1997b; U.S. EPA, 1997c; U.S. EPA, 1999).
• Prepared an implementation guide for PWB manufacturers interested in switching from
HASL to an alternative surface finishing technology (U.S. EPA, 2000).
• Identified, evaluated, and disseminated information on viable pollution prevention
opportunities for the PWB industry through an updated review of a pollution prevention
and control practices industry study (U.S. EPA, 1998b).
Further information about the project, along with web-based versions of all the documents listed
above and other previous project work, can be obtained by visiting the Design for the
Environment Program website, located at www.epa.gov/dfe/pwb.
1-4
-------
1.2 OVERVIEW OF PWB INDUSTRY
1.2.1 Types of Printed Wiring Boards
PWBs may be categorized in several ways, either by the number of layers or by the type
of substrate. The number of circuit layers present on a single PWB give an indication of the
overall complexity of the PWB. The most common categories are multi-layer, double-sided, and
single-sided PWBs. Multi-layer PWBs contain more than two layers of circuitry, with at least one
layer imbedded in the substrate beneath the surface of the board. Multi-layer boards may consist
of 20 or more interconnected layers, but four, six, and eight layer boards are more common.
Double-sided boards have circuitry on both sides of a board, resulting in two interconnected
layers, while single-sided PWBs have only one layer of circuitry. Double-sided and single-sided
PWBs are generally easier to produce than multi-layer boards (U.S. EPA, 1995).
PWB substrates, or base material types, fall into three basic categories: rigid PWBs,
flexible circuits, and rigid-flex combinations. Rigid multi-layer PWBs dominate the domestic
production of all PWBs (see Section 1.2.2, below) and are the focus of this CTSA.
Rigid PWBs typically are constructed of glass-reinforced epoxy-resin systems that
produce a board less than 0.1" thick. The most common rigid PWB thickness is 0.062", but there
is a trend toward thinner PWBs. Flexible circuits (also called flex circuits) are manufactured on
polyamide and polyester substrates that remain flexible at finished thicknesses. Ribbon cables
are common flexible circuits. Rigid-flex PWBs are essentially combinations or assemblies of
rigid and flexible PWBs. They may consist of one or more rigid PWBs that have one or more
flexible circuits laminated to them during the manufacturing process. Three-dimensional circuit
assemblies can be created with rigid-flex combinations (U.S. EPA, 1995).
1.2.2 Industry Profile
The total world market for PWBs is about $31.4 billion, with U. S. production accounting
for about one quarter of the total (Wehrspann, 1999a). Although the United States and Japan are
the leading suppliers of PWBs, Hong Kong, Singapore, Taiwan, and Korea also have captured a
significant share of the world market. The U.S.-dominated world market for PWBs eroded from
1980 to 1990, but has come back slightly in recent years. The market share of the countries with
the largest PWB production is shown in Figure 1-1.
IPC estimates that the U.S. market for PWBs in 1998 totaled approximately $8.6 billion
for both rigid and flex PWBs. U.S. imports of PWBs were estimated to be approximately $500 to
$600 million annually, the majority of which come from Taiwan, Japan, Hong Kong, Korea, and
Thailand (Wehrspann, 1999b). The value of U.S. PWB exports reported for 1998 were
approximately $100 million, which represents two to three percent of total U.S. PWB production
(Wehrspann, 1999b).
1-5
-------
U.S.
Others
Japan
Taiwan
Figure 1-1. PWBs Produced for World Market in 1998 (IPC)
The United States had 652 independent PWB manufacturing plants in 1999 (Abrams,
2000). California, Minnesota, Texas, Illinois, Massachusetts, and Arizona have the highest
number of PWB manufacturing plants, but there are PWB manufacturing facilities in virtually all
50 states and territories. More than 75 percent of U.S.-made PWBs are produced by independent
shops (U.S. EPA, 1995).
About 80 percent of independent PWB manufacturers are small- to medium-sized
businesses with annual sales under $10 million, but these shops only account for 20 to 25 percent
of total U.S. sales. Conversely, about five percent of PWB manufacturers are larger independent
shops with annual sales over $20 million, but these shops account for about 70 percent of total
U.S. sales (Wehrspann, 1999b). Recent industry trends have seen the purchase of many smaller
companies by larger corporations with much larger annual sales.
Overall U.S. production accounted for 1.4 billion PWBs produced in 1998. While
demand for multi-layer PWBs continues to grow, both single- and double-sided PWBs are still
produced in greater numbers. The market for multi-layer boards was about $7.9 billion in 1998
(Wehrspann, 1999b), up from approximately $700 million in 1980 (U.S. EPA, 1995). A
breakdown of U.S. production by the type of PWB is shown in Figure 1-2.
1-6
-------
Single-Sided
Multi-layer
Double-Sided
Figure 1-2. Number of PWBs Produced by U.S. Manufacturers in 1998 (IPC)
The PWB industry directly employs about 75,000 people, with about 68 percent of
employment in production jobs. This is the highest ratio of production jobs for U.S. electronics
manufacturing (U.S. EPA, 1995). Additional jobs related to the industry are generated by PWB
material and equipment suppliers and the OEMs that produce PWBs for internal use. Further
information about the industry may be found in Printed Wiring Board Industry and Use Cluster
Profile (U.S. EPA, 1995) or from contacting the industry trade association, IPC.
1.2.3 Overview of Rigid Multi-Layer PWB Manufacturing
Multi-layer boards consist of alternating layers of conductor and insulating material
bonded together. Individual circuitry inner-layers are created and then assembled under high
temperature into a solid board. Holes are drilled through the boards, and then plated to provide
layer-to-layer connection on multi-layered circuits. The outside layers are imaged, plated, and
then etched to create the circuitry traces on the outside surfaces of the PWB. A solder mask is
then applied to the board prior to applying the final surface finish.
Application of the surface finish is the last major step in the PWB manufacturing process.
The function of the surface finish is to provide a clean, solderable surface for subsequent
assembly, while also protecting the surface from degradation or contamination from
environmental factors, such as water, temperature, and oil from handling. The surface finishing
technologies evaluated in this report all deposit this solderable layer, or coating. Traditionally, the
surface finish has been tin-lead solder, applied using the HASL technology.
1-7
-------
1.3 CTSA METHODOLOGY
The CTSA methodology is a means of systematically evaluating and comparing human
health and environmental risk, competitiveness (i.e., performance, cost, etc.), and resource
requirements of traditional and alternative chemicals, manufacturing methods, and technologies
in a particular use cluster. A use cluster is a set of chemical products, technologies, or processes
that can substitute for one another to perform a particular function. A CTSA document is the
repository for the technical information developed by a DfE project on a use cluster. Thus,
surface finishing technologies comprise the use cluster that is the focus of this CTSA.
The overall CTSA methodology used in this assessment was developed by the EPA DfE
Program, the UT Center for Clean Products and Clean Technologies, and other partners in
voluntary, industry-specific pilot projects. The publication, Cleaner Technologies Substitutes
Assessment: A Methodology & Resource Guide (Kincaid et al., 1996) presents the CTSA
methodology in detail. This section summarizes how the various technologies were selected for
evaluation in the CTSA, identifies issues evaluated and data sources, and describes the project
limitations. Chapters 2 through 6, and appendices, describe in detail the methods used to
evaluate the technologies.
1.3.1 Identification of Alternatives and Selection of Project Baseline
Once the use cluster for the CTSA was chosen, industry representatives identified
technologies that may be used to accomplish the surface finishing function. Initially, eight
technology categories were identified, including six inorganic metal-based technologies, and two
organic-based coatings. These include:
• Inorganic: HASL, nickel/gold, nickel/palladium/gold, immersion silver, immersion
palladium, and immersion tin.
• Organic: OSP (benzotriazole-based), and OSP (substituted immidizole-based).
Suppliers were contacted by EPA and asked to submit their product lines in these
technology categories for evaluation in the CTSA. Criteria for including a technology in the
CTSA were the following:
• it is an existing or emerging technology; and
• there are equipment and facilities available to demonstrate its performance.
In addition, suppliers were required to provide information about their technologies, including
chemical product formulation data, process schematics, process characteristics and constraints
(e.g., cycle time, limitations for the acid copper plating process, substrate and drilling
compatibilities, aspect ratio capacity, range of hole sizes), bath replacement criteria, and cost
information.
1-8
-------
Product lines were submitted, along with confidential process formulation data, for all of
the technologies except the benzotriazole-based OSP technology. After further review, it was
determined that the immersion palladium technology could not be demonstrated sufficiently
under production conditions, preventing the evaluation of the technology's performance and cost
of operation. As a result, only a process description of the immersion palladium technology is
presented in this CTSA. Thus, seven categories of technologies were carried forward for further
evaluation in the CTSA.
The HASL technology was selected as the project baseline for the following reasons:
• It is generally regarded to be the industry standard and holds the vast majority of the
market for surface finishing technologies.
• Possible risk concerns associated with lead exposure, the large amount of solid waste
generated by the HASL process, and the fact that the solder finish has become
technologically limiting with regard to current design and assembly practices have
prompted many PWB manufacturers to independently seek alternatives to HASL.
As with other surface finishing technologies, the HASL process can be operated using
vertical, immersion-type, non-conveyorized equipment or horizontal, conveyorized equipment.
Conveyorized surface finishing equipment is usually more efficient than non-conveyorized
equipment, but requires a substantial capital investment. Most facilities in the United States still
use a non-conveyorized HASL process to perform the surface finishing function. Therefore, the
baseline technology was further defined to only include non-conveyorized HASL processes.
Conveyorized HASL processes, and both non-conveyorized and conveyorized equipment
configurations of the other technology categories, are all considered to be alternatives to non-
conveyorized HASL.
1.3.2 Boundaries of the Evaluation
For the purposes of the environmental evaluation (i.e., human health and ecological
hazards, exposure, risk, and resource consumption), the boundaries of this evaluation can be
defined in terms of the overall life cycle of the surface finishing products and in terms of the
PWB manufacturing process. The life cycle of a product or process encompasses extraction and
processing of raw materials, manufacturing, transportation and distribution, use/re-use/
maintenance, recycling, and final disposal. As discussed in Section 1.2.3, rigid, multi-layer PWB
manufacturing encompasses a number of process steps, of which the surface finishing process is
the last one.
The activities evaluated in this study are primarily the use of surface finishing chemicals at
PWB facilities and the release or disposal of surface finishing chemicals from PWB facilities.
However, in addition to evaluating the energy consumed during surface finishing line operation,
the analysis of energy impacts (Section 5.2) also discusses the pollutants generated from
producing the energy to operate the surface finishing line, as well as energy consumed in other
life-cycle stages, such as the manufacture of chemical ingredients. In addition, information is
presented on the type and quantity of wastewater generated by the surface finishing process line,
1-9
-------
and the risk to the environment resulting from the discharge of the wastewater to nearby surface
water (Section 3.4). Finally, while information is presented on the generation and disposal of
solid waste from surface finishing technologies, there was insufficient information to characterize
the risk from these environmental releases. This is discussed in more detail in Section 3.1, Source
Release Assessment.
In terms of the PWB manufacturing process, this analysis focused entirely on the surface
finishing process, defined as beginning with a panel that has had solder mask applied, and ending
after a surface finish has been applied to the connecting surfaces of the PWB and the board has
been cleaned of any residual process chemistry. In cases where no solder mask is applied, the
use cluster would begin after the stripping of the etch resist from the outside board surfaces.
The narrow focus on surface finishing technologies yields some benefits to the evaluation,
but it also has some drawbacks. Benefits include the ability to collect extremely detailed
information on the relative risk, performance, cost, and resources requirements of the baseline
technology and alternatives. This information provides a more complete assessment of the
technologies than has previously been available and would not be possible if every step in the
PWB manufacturing process was evaluated. Drawbacks from such focused evaluations include
the inability to identify all of the plant-wide benefits, costs, or pollution prevention opportunities
that could occur when implementing an alternative to the baseline HASL technology. However,
given the variability in workplace practices and operating procedures at PWB facilities, these
other benefits and opportunities are expected to vary substantially among facilities and would be
difficult to assess in a comparative evaluation such as a CTSA. Individual PWB manufacturers
are urged to assess their overall operations for pollution prevention opportunities when
implementing an alternative technology.
1.3.3 Issues Evaluated
The CTSA evaluated a number of issues related to the risk, competitiveness, and resource
requirements of surface finishing technologies. These include the following:
• Risk: occupational health risks, public health risks, ecological hazards, and process safety
concerns.
• Competitiveness: technology performance, cost, and regulatory status.
• Conservation: energy and natural resource use.
Occupational and public health risk information is for chronic exposure to long-term,
day-to-day exposure and releases from a PWB facility rather than short-term, acute exposures to
high levels of hazardous chemicals as could occur with a fire, spill, or other periodic release. Risk
information is based on exposures estimated for a model facility, rather than exposures estimated
for a specific facility. Ecological risks are also evaluated for aquatic organisms that could be
exposed to surface finishing chemicals through wastewater discharges. Process safety concerns
are summarized from material safety data sheets (MSDSs) for the technologies and process
operating conditions.
1-10
-------
Technology performance is based on a snapshot of the performance of the surface
finishing technologies at volunteer test sites in the United States. Panels were electrically
prescreened, followed by electrical stress testing, accelerated aging, and mechanical testing, in
order to distinguish robustness of the applied surface finishes. Comparative costs of the surface
finishing technologies were estimated with a hybrid cost model that combines traditional costs
with simulation modeling and activity-based costs. Costs are presented in terms of dollars per
surface square feet (ssf) of PWB produced.
Federal environmental regulatory information is presented for the chemicals in the surface
finishing technologies. This information is intended to provide an indication of the regulatory
requirements associated with a technology, but not to serve as regulatory guidance.
Quantitative resource consumption data are presented for the comparative rates of metal,
energy, and water use by the surface finishing technologies. The consumption of other
resources, such as process and treatment chemicals, are qualitatively assessed.
1.3.4 Primary Data Sources
Much of the process-specific information presented in this CTSA was provided by
chemical suppliers to the PWB industry, PWB manufacturers who responded to project
information requests, and PWB manufacturers who volunteered their facilities for a performance
demonstration of the baseline and alternative technologies. The types of information provided by
chemical suppliers and PWB manufacturers are summarized below.
Chemical Suppliers
The project was open to all interested chemical suppliers, provided that they agreed to
disclose confidential chemical formulation data for use in this evaluation, and that
their technologies met the criteria described in Section 1.3.1. Table 1-1 lists the suppliers who
participated in the CTSA and the categories of surface finishing technologies they submitted for
evaluation. It should be noted that this is not a comprehensive list of surface finishing
technology suppliers. EPA made every effort to publicize the project through trade associations,
PWB manufacturers, industry conferences and other means, but some suppliers did not learn of
the project until it was too late to submit technologies for evaluation, or chose not to participate.
Table 1-1. Surface Finishing Technologies Submitted by Chemical Suppliers
Chemical Supplier
Polyclad Technologies-
Enthone
Electrochemicals, Inc.
Florida CirTech, Inc.
MacDermid, Inc.
Technic, Inc.
Surface Finishing Technology
Nickel/Gold
X
X
Nickel/Palladium/
Gold
X
OSP
X
X
Immersion
Silver
X
Immersion
Tin
X
X
1-11
-------
A supplier for HASL is not shown in Table 1-1 because the HASL technology is not sold
as a product line by a supplier. Instead, it consists a series of chemical cleaning and flux steps,
followed by HASL equipment, which mechanically applies the solder to PWB surface. The
board is then cleaned using a water rinse cleaning system. The chemical baths preceding the
HASL equipment are not designed specifically for use with the HASL process, and are similar to
those used by other surface finishing technologies. Chemical data from cleaning baths in other
processes were substituted for this analysis. HASL equipment is commercially available from a
number of suppliers.
Each of the chemical suppliers provided the following: MSDSs for the chemical products
in their surface finishing technology lines; Product Data Sheets, which are technical specifications
prepared by suppliers for PWB manufacturers that describe how to mix and maintain the
chemicals baths; and complete product formulation data. Suppliers were also asked to complete
a Supplier Data Sheet, designed for the project, which included information on chemical cost,
equipment cost, water consumption rates, product constraints, and the locations of test sites for
the Performance Demonstration. Appendix A contains a copy of the Supplier Data Sheet.
PWB Manufacturers
PWB manufacturers were asked to participate in a study of workplace practices. The
PWB Workplace Practices Questionnaire requested detailed information on facility size, process
characteristics, chemical consumption, worker activities related to chemical exposure, water
consumption, and wastewater discharges. The questionnaire was distributed by IPC to PWB
manufacturers. PWB manufacturers returned the completed questionnaires to IPC, which
removed all facility identification and assigned a code to the questionnaires prior to forwarding
them to UT's Center for Clean Products and Clean Technologies. In this manner, PWB
manufacturers were guaranteed confidentiality of data.
For the Performance Demonstration project the PWB Workplace Practices Questionnaire
was modified and divided into two parts: a Facility Background Information Sheet and an
Observer Data Sheet. The Facility Background Information Sheet was sent to PWB facilities
participating in the Performance Demonstration prior to their surface finishing technology test
date. It requested detailed information on facility and process characteristics, chemical
consumption, worker activities related to chemical exposure, and water consumption. The
Observer Data Sheet was used by an on-site observer to collect data during the Performance
Demonstration. In addition to ensuring that the performance test was conducted according to the
agreed-upon test protocol, the on-site observer collected measured data, such as bath temperature
and process line dimensions, and difficult to collect data, such as equipment loading rates and
energy usage. The observer also checked survey data collected on the Facility Background
Information Sheet for accuracy. Appendix A contains copies of the PWB Workplace Practices
Questionnaire, the Facility Background Information Sheet, and the Observer Data Sheet forms.
1-12
-------
Table 1-2 lists the number of PWB manufacturing facilities that completed the PWB
Workplace Practices Questionnaire by type of surface finishing process, excluding responses
with poor or incomplete data. Of the 54 responses to the questionnaire, 16 were Performance
Demonstration test sites.
Table 1-2. Responses to the PWB Workplace Practices Questionnaire
Surface Finishing
Technology
HASL
Nickel/Gold
Nickel/Palladium/Gold
No. of
Responses
29
8
1
Surface Finishing Technology
OSP
Immersion Silver
Immersion Tin
No. of
Responses
9
2
5
Information from the pollution prevention and control technologies survey conducted by
the DfE PWB Project was also used in the CTSA. These data are described in detail in the EPA
publication, Printed Wiring Board Pollution Prevention and Control Technology: Analysis of
Updated Survey Results (U.S. EPA, 1998b).
1.3.5 Project Limitations
There are a number of limitations to the project, both because of the predefined scope of
the project and data limitations inherent to the characterization techniques. Some of the
limitations related to the risk, competitiveness, and conservation components of the CTSA are
summarized below. More detailed information on limitations and uncertainties for a particular
portion of the assessment is given in the applicable sections of this document. A limitation
common to all components of the assessment is that the surface finishing chemical products
assessed in this report were voluntarily submitted by participating suppliers and may not
represent the entire surface finishing technology market. For example, the immersion palladium
and benzotriazole-based OSP technologies were not evaluated in the CTSA. Alternatives that are
evaluated were submitted by at least one supplier, but not necessarily by every supplier who
offers that surface finishing technology.
Risk
The risk characterization is a screening level assessment of multiple chemicals used in
surface finishing technologies. The focus of the risk characterization is on chronic (long-term)
exposure to chemicals that may cause cancer or other toxic effects, rather than on acute toxicity
from brief exposures to chemicals. The exposure assessment and risk characterization use a
"model facility" approach, with the goal of comparing the exposures and health risks of the
surface finishing process alternatives to the baseline HASL technology. Characteristics of the
model facility were aggregated from questionnaire data, site visits, and other sources, and are
based on the assumption of manufacturing 260,000 ssf per year. This approach does not result in
an absolute estimate or measurement of risk.
1-13
-------
In addition, the exposure and risk estimates reflect only a portion of the potential
exposures within a PWB manufacturing facility. Many of the chemicals found in surface
finishing technologies may also be present in other process steps of PWB manufacturing, and
other risk concerns for human health and the environment may occur from other process steps.
Incremental reduction of exposures to chemicals of concern from a surface finishing process,
however, will reduce cumulative exposures from all sources in a PWB facility, provided that
increased production does not increase plant-wide pollution.
Finally, information presented in this CTSA is based on publicly-available chemistry data
submitted by each of the participating suppliers, as well as proprietary data submitted by the
suppliers. Risk information for proprietary ingredients is included in this CTSA, but chemical
identities and chemical properties are not listed.
Competitiveness
The Performance Demonstration was designed to provide a snapshot of the performance
of different surface finishing technologies. The test methods used to evaluate performance were
intended to indicate characteristics of a technology's performance, not to define parameters of
performance or to substitute for thorough on-site testing. Because the test sites were not chosen
randomly, the sample may not be representative of all PWB manufacturing facilities in the United
States (although there is no specific reason to believe they are not representative).
The cost analysis presents comparative costs of using a surface finishing technology in a
model facility to produce 260,000 ssf of PWBs. As with the risk characterization, this approach
results in a comparative evaluation of cost, not an absolute evaluation or determination. The cost
analysis focuses on private costs that would be incurred by facilities implementing a technology.
It does not evaluate community benefits or costs, such as the effects on jobs from implementing
a more efficient surface finishing technology. However, the Social Benefits/Costs Assessment
(see Section 7.2) qualitatively evaluates some of these external (i.e., external to the decision-
maker at a PWB facility) benefits and costs.
The regulatory information contained in the CTSA may be useful in evaluating the
benefits of moving away from processes containing chemicals that trigger compliance issues.
However, this document is not intended to provide compliance assistance. If the reader has
questions regarding compliance concerns, they should contact their federal, state, or local
authorities.
Conservation
The analysis of energy and water consumption is also a comparative analysis, rather than
an absolute evaluation or measurement. Similar to the risk and cost analyses, consumption rates
were estimated based on using a surface finishing technology in a model facility to produce
260,000 ssf of PWB.
1-14
-------
1.4 ORGANIZATION OF THIS REPORT
This CTSA is organized into two volumes: Volume I summarizes the methods and
results of the CTSA; Volume II consists of appendices, including detailed chemical properties
and methodology information.
Volume I is organized as follows:
• Chapter 2 gives a detailed profile of the surface finishing use cluster, including process
descriptions of the surface finishing technologies evaluated in the CTSA and the
estimated concentrations of chemicals present in surface finishing chemical baths.
• Chapter 3 presents risk information, beginning with an assessment of the sources, nature,
and quantity of selected environmental releases from surface finishing processes (Section
3.1); followed by an assessment of potential exposure to surface finishing chemicals
(Section 3.2) and the potential human health and ecological hazards of surface finishing
chemicals (Section 3.3). Section 3.4 presents quantitative risk characterization results,
while Section 3.5 discusses process safety concerns.
• Chapter 4 presents competitiveness information, including performance demonstration
results (Section 4.1), cost analysis results (Section 4.2), and regulatory information
(Section 4.3).
• Chapter 5 presents conservation information, including an analysis of water and other
resource consumption rates (Section 5.1) and energy impacts (Section 5.2).
• Chapter 6 describes additional pollution prevention and control technology opportunities
(Sections 6.1 and 6.2, respectively).
• Chapter 7 organizes data collected or developed throughout the CTSA in a manner to
facilitate decision-making. Section 7.1 presents a summary of risk, competitiveness, and
conservation data. Section 7.2 assesses the social benefits and costs of implementing an
alternative as compared to the baseline. Section 7.3 provides summary profiles for the
baseline and each of the surface finishing alternatives.
1-15
-------
REFERENCES
Abrams, Fern. 2000. IPC-Association Connecting Electronics Industries). Personal
communication with Jack Geibig, UT Center for Clean Products and Clean Technologies.
December.
IPC (IPC-Association Connecting Electronics Industries). 1996. The National Technology
Roadmapfor Electronic Interconnections.
Kincaid, Lori E., Jed Meline and Gary Davis. 1996. Cleaner Technologies Substitutes
Assessment: A Methodology & Resource Guide. EPA Office of Pollution Prevention and
Toxics. Washington, D.C. EPA 744-R-95-002. December.
MCC (Microelectronics and Computer Technology Corporation). 1993. Environmental
Consciousness: A Strategic Competitiveness Issue for the Electronics and Computer Industry.
March.
MCC (Microelectronics and Computer Technology Corporation). 1994. Electronics Industry
Environmental Roadmap. December.
U.S. EPA (Environmental Protection Agency). 1995. Printed Wiring board Industry and Use
Cluster Profile. EPA Office of Pollution Prevention and Toxics. Washington, D.C.
EPA/744-R-95-005. September.
U.S. EPA (Environmental Protection Agency). 1997a. "Printed Wiring Board Case Study 6:
Pollution Prevention Beyond Regulated Materials." EPA Office of Pollution Prevention and
Toxics. Washington, D.C. EPA 744-F-97-006. May.
U.S. EPA (Environmental Protection Agency). 1997b. "Printed Wiring Board Case Study 7:
Identifying Objectives for your Environmental Management System." EPA Office of Pollution
Prevention and Toxics. Washington, D.C. EPA 744-F-97-009. July.
U.S. EPA (Environmental Protection Agency). 1997c. "Printed Wiring Board Case Study 8:
Building an EMS, H-R Industries Experience." EPA Office of Pollution Prevention and Toxics.
Washington, D.C. EPA 744-F-97-010. December.
U.S. EPA (Environmental Protection Agency). 1998a. Printed Wiring Board Cleaner
Technologies Substitutes Assessment: Making Holes Conductive. Design for the Environment
Printed Wiring Board Project. EPA Office of Pollution Prevention and Toxics. Washington,
D.C. EPA/744-R-98-004a and EPA/744-R-98-004b. August.
U.S. EPA (Environmental Protection Agency). 1998b. Printed Wiring Board Pollution
Prevention and Control Technology: Analysis of Updated Survey Results. Design for the
Environment Printed Wiring Board Project. EPA Office of Pollution Prevention and Toxics.
Washington, D.C. EPA 744-R-98-003. August.
1-16
-------
U.S. EPA (Environmental Protection Agency). 1999. "Printed Wiring Board Case Study 9:
Flexible Simulation Modeling of PWB Costs." EPA Office of Pollution Prevention and Toxics.
Washington, D.C. EPA 744-F-99-004. May.
U.S. EPA (Environmental Protection Agency). 2000. Implementing Cleaner Printed Wiring
Board Technologies: Surface Finishes. EPA Office of Pollution Prevention and Toxics.
Washington, D.C. EPA 744-R-00-002. March.
Wehrspann, Carla. 1999a. IPC-Association Connecting Electronics Industries. Personal
communication with Jeng Hon Su, University of Tennessee Center for Clean Products and Clean
Technologies. July.
Wehrspann, Carla. 1999b. IPC-Association Connecting Electronics Industries. Personal
communication with Jack Geibig, University of Tennessee Center for Clean Products and Clean
Technologies. June.
1-17
-------
Chapter 2
Profile of the Surface Finishing Use Cluster
This section of the Cleaner Technologies Substitutes Assessment (CTSA) describes the
technologies that comprise the surface finishes use cluster. A use cluster is a set of chemical
products, technologies, or processes that can substitute for one another to perform a particular
function. In this case, the function is the application of a final surface finish to the printed wiring
board (PWB). The set of technologies includes hot air solder leveling (HASL), which was
selected as the baseline, and the alternative surface finishes, including electroless
nickel/immersion gold (nickel/gold), electroless nickel/electroless palladium/immersion gold
(nickel/palladium/gold), organic solderability preservative (OSP), immersion silver, and
immersion tin.
Section 2.1 presents process descriptions for each of the surface finishing technologies
and describes the chemical composition of products that were evaluated in the CTSA. Section
2.2 briefly describes additional technologies that may be used to perform the surface finishing
function, but were not evaluated.
2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING
TECHNOLOGIES
This section introduces the surface finishing technologies evaluated in the CTSA and
details the process sequences. Typical operating conditions and operating and maintenance
procedures are described in an overview of the surface finishing manufacturing process. Then
the chemical processes occurring in each bath are detailed, along with additional process
information specific to each technology.
2.1.1 Process Sequences of Surface Finishing Technologies
Figure 2-1 depicts the six surface finishing technologies evaluated in the CTSA. Because
the function of applying a final surface finish can be performed using any of these technologies,
these technologies may be substituted for each other in PWB manufacturing. The surface
finishing technologies are all wet chemistry processes consisting of a series of chemical process
baths, often followed by rinse steps, through which the PWB panels are passed to apply the final
surface finish. The exception is the HASL process, which combines the typical cleaning and
etching chemical processes with a mechanical process of dipping a board into molten solder
followed by rinsing (described in Section 2.1.3).
For each of the surface finishes evaluated, the process steps depicted in the figure
represent an integration of the various commercial products within the technology category. For
example, chemical suppliers to the PWB industry submitted product data for two different OSP
processes. The chemical suppliers offer additional variations to the OSP process that may have
slightly different bath chemistries or process sequences, than the processes submitted. Figure 2-1
lists the process steps in a typical, or generic, OSP surface finishing line. The process steps in an
actual line may vary.
2-1
-------
HASL
p- Micrnptch
^
F «™
F^F
__ ...... -™_..
ff TVT • _
' ' : ' : •••""•'r^
Nickel/
Gold
L
Electroless
E
Nickel/
Palladium/
Gold
^ JUy-rnefoh J [' ii^~J'I±tki'i-
1 L
s- Electroless
F ^.•::V_V ..'....
7 Electroless
OSP
Cleaner
f Microetch
r Air Knife
Air Knife
Dry
Immersion
Silver
r
4 Immersion
K
Immersion
Tin
I F D ,./,,i;.> I
X ;
Note: One or more intermediate rinse steps typically separate the process steps listed above. For simplicity, these intermediate rinse steps have not been
included in the diagram.
Figure 2-1. Typical Process Steps for Surface Finishing Technologies
2-2
-------
2.1.2 Overview of the Surface Finishing Manufacturing Process
Surface finishing technologies typically consist of a series of sequential chemical
processing tanks (baths) separated by water rinse stages. The process can either be operated in a
vertical, non-conveyorized submersive-type mode, or in a horizontal, conveyorized mode. In
either mode, selected baths may be operated at elevated temperatures to facilitate required
chemical reactions or baths may be agitated to improve contact between the panels and the bath
chemistry. Agitation methods employed by PWB manufacturers include panel agitation,
ultrasonic vibration, air sparging, and fluid circulation pumps.
Most process baths are followed by a water rinse tank to remove drag-out, the clinging
film of process solution covering the rack and boards when they are removed from a tank.
Rinsing is necessary to provide a clean panel surface for further chemical activity and to prevent
chemical drag-out, which may contaminate subsequent process baths. PWB manufacturers
employ a variety of rinse water minimization methods to reduce rinse water usage and
consequent wastewater generation rates. The quantities of wastewater generated from surface
finishing lines are discussed in Section 5.1, Resource Conservation, while the composition of the
wastewater is modeled and presented in Section 3.2, Exposure Assessment. Rinse water
reduction techniques are discussed in Section 6.1, Pollution Prevention.
After the application, imaging, and development of the solder mask, panels are loaded
into racks (vertical, non-conveyorized mode) or onto a conveyor (horizontal, conveyorized
mode) for processing by the surface finishing line. Racks may be manually moved from tank to
tank, moved by a manually or automatically controlled hoist, or moved by other means. Process
tanks are usually open to the atmosphere. To reduce volatilization of chemicals from the bath or
worker exposure to volatilized chemicals, process baths may be equipped with a local ventilation
system, such as a push-pull system, or covered during extended periods of latency. Horizontal,
conveyorized systems are typically fully enclosed, with air emissions vented to a control
technology or to the atmosphere outside the plant.
The HASL process differs from the other alternatives in that it does not rely on a chemical
process to apply the final surface finish. Instead, the process combines the chemical processes of
board preparation and cleaning with a mechanical step to apply the finish.
Regardless of the mode of operation or type of alternative, process chemical baths are
periodically replenished to either replace solution lost through drag-out or volatilization, or to
return the concentration of constituents in the bath to within acceptable limits. During the course
of normal operations, bath chemistry can be altered by chemical reactions occurring within the
bath or by contamination from drag-in. Bath solution may be discarded and replaced with new
solution as required, with the frequency of replacement depending on analytical sampling results,
the number of panel surface square feet (ssf) processed, or the amount of time elapsed since the
last change-out. Process line operators also may clean the tank or conveyorized equipment
during bath change-out operations.
2-3
-------
Some process baths are equipped with filters to remove particulate matter that may be
introduced to the bath or formed as a precipitate through a chemical reaction. Process line
operators or other personnel periodically replace the bath filters based on criteria, such as
analytical sampling results from the process baths, elapsed time, or volume of product produced.
2.1.3 Chemistry and Process Descriptions of Surface Finishing Technologies
This section describes in detail the processes for applying a solderable and protective
coating, or surface finish, to the outside surfaces of a PWB. A brief description of the chemical
mechanisms or processes occurring in each of the process steps, along with other pertinent
process data such as flux compatibilities, storage limitations, assembly methods required, and
modes of operation (e.g., non-conveyorized or conveyorized), are presented for each technology.
For technologies with more than one chemical supplier (e.g., nickel/gold, OSP, immersion tin), a
process description for each chemical product line was developed in consultation with the
chemical supplier and then combined to form a generic process description for that technology.
Notable differences in the chemical mechanisms or processes employed in a single product line
from that of the generic process are detailed.
Each alternative surface finishing process evaluated in the CTSA uses one of the
following mechanisms to apply the final finish.
• Electroless process: This chemical process promotes continuous deposition of a metal
onto the PWB surface through an oxidation-reduction chemical reaction, without the use
of an external electrical potential. A reducing agent, such as sodium hypophosphite,
donates electrons to the positively charged metal ions in solution, thereby reducing the
metal and promoting its deposition onto the catalyzed metal surfaces of the PWB. This
reaction is considered auto-catalytic because it will continue to plate in the presence of
source metal ions and a reducing agent until the board is removed from the plating bath.
The thickness of plated deposits vary according to the amount of time spent in the plating
bath, but are typically in the 3 to 5 micron range.
• Immersion process: This chemical process uses a chemical displacement reaction to
deposit a metal layer onto the exposed metal surface of the PWB. In this reaction, the
base metal donates the electrons that reduce the positively charged metal ions in the
solution. Driven by the electrochemical potential difference, the metal ions in solution
(e.g., gold ions in the immersion gold portion of the nickel/palladium/gold process) are
deposited onto the surface of the board, simultaneously displacing ions of the surface
metal (e.g., nickel ions for the example above) back into solution. This reaction is
considered self-limiting, because once the surface metal is plated, there is no longer a
source of electrons and the reaction stops. Surface finish deposits of up to 0.2 microns
are considered typical for immersion processes.
• Coating: A protective coating is applied by submerging the PWB into a chemical bath.
Although a coating does not require an exchange of electrons to facilitate deposition of
the protective layer, some coatings may be formulated to adhere selectively to exposed
metal surfaces. Typical coating thicknesses range from 0.1 to 0.5 microns.
2-4
-------
Hot Air Solder Leveling (HASP
HASL has long been the standard surface finishing method used in the manufacture of
double-sided and multi-layered boards, because its excellent solderability during assembly.
However, due to its technological limitations, environmental concerns, and process safety issues,
assemblers and manufacturers have begun to seriously consider other surface finishes as viable
alternatives to HASL. During the HASL process, soldermask-coated boards are first cleaned and
etched to prepare the contact surfaces for the solder. Following the application of flux to a board,
a layer of solder is applied to the copper surfaces by submersing the panel in molten solder. The
excess solder is then blown from the board by an air knife, leaving a thin, protective layer of
solder on the exposed circuitry.
Any of these three process segments - board preparation, solder application, or cleaning
- may be automated or manual, or any combination thereof. These segments may also be
integrated into one entire conveyorized process, combining the chemical pretreatment and
cleaning steps with the solder application. Flux formulations are altered depending on the mode
of operation and the desired flux characteristics. HASL finishes are compatible with surface
mount technology (SMT) and typical through hole components; however, the lack of planarity,
or flatness, of the finish makes assembly with fine pitch, small components difficult to control.
In addition, the HASL finish cannot be wirebonded. Extended shelf life on a typical SMT pad or
plated through hole (PTH) annular ring is not a concern with HASL finished boards, because of
the durability of the finish. However, large flat surfaces can exhibit solderability problems after
storage due to removal of all but a very thin coating of solder by the HASL process. This thin
coating allows exposure of intermetallic surfaces that can create solderability problems (Carroll,
1999). Typically HASL finished PWBs have a shelf life of up to a year (Kerr, 1999).
A flow diagram of the process steps in a typical HASL process is presented in Figure 2-2.
A brief description of each of the process steps is also given.
Step 1: Cleaner: An acid-based cleaner removes surface oils, oxides, and any organic
residues left after the solder mask application. The cleaner provides a clean,
consistent copper surface to ensure uniform etching.
Step 2: Microetch: The microetch solution lightly etches the exposed copper surfaces of
the panel, including the barrels of the PTH, to remove any chemical contamination
and metal oxides present.
Step 3: Dry: The etched panels are then air-dried using a non-heated blower to minimize
the formation of oxides on the cleaned and etched copper surfaces.
2-5
-------
1 Cleaner I
y
Microetch 1
y
Water Rinse x 2 1
y
3 OiJ |
y
4 Flux |
y
5. Solder 1
y
Pressure Rinse 1
y
7 DI Rinse I
Figure 2-2. HASL Process Flow Diagram
Step 4: Flux: A chemical flux is applied to the panel to reduce the surface tension of the
copper pads, thereby maximizing the wetting of the copper surfaces. The flux is
composed of a heat transfer fluid, stabilizers, inhibitors, and activating agents.
Flux formulations may vary considerably depending on the characteristics desired.
Horizontal HASL system fluxes tend to be lower in viscosity and more highly
activated than fluxes for vertical, non-conveyorized systems.
Step 5: Solder: Solder is selectively applied to the copper surfaces of the panel by
submerging the preheated, fluxed panels in a bath of molten solder. The excess
solder is then removed from the board by an air knife when the panel is withdrawn
from the solder bath.
2-6
-------
Step 6-7: Pressure Rinse: A high-pressure water rinse is used to dislodge any solder balls or
excess solder flash that may be present on the PWB. The water rinse also
removes any remaining flux residue that was not vaporized in the solder bath.
This rinse stage may consist of several rinse tanks and include heated rinses or
rinses combined with mechanical scrubbing. A post-solder chemical cleaner may
also be used as a rinse aid if desired, or if water rinsing is insufficient. The final
step in the post-clean process is rinsing in de-ionized water to reduce ionic
contaminants on the surface finish.
Flux selection is critical to the sound operation of the HASL process. The flux is
responsible for creating the copper surface conditions required to achieve a high quality solder
deposit on the PWB. Fluxes are available in a variety of formulations with differing
characteristics such as viscosity, foam level, acidity, volatile content, and type of activator. The
type of HASL flux ultimately selected will depend on the type of chemicals and processes used
in previous manufacturing stages, type of solder mask, and the solder deposit characteristics
required.
The cleaning steps after the application of the solder can vary quite a bit, depending on
several factors including the type of flux, type of solder mask, and the cleanliness standards to be
met. The most commonly reported post-clean sequence by survey respondents utilized a series
of water rinse baths combined with either high pressure rinsing, scrubbing, or a mild detergent.
The post-clean system described above was selected to represent the HASL baseline.
Nickel/Gold
The nickel/gold process promotes the deposition of an initial, thick layer of nickel
followed by a thin, protective layer of gold onto the exposed copper surfaces of the PWB. Nickel
characteristics such as hardness, wear resistance, solderability, and uniformity of the deposit
make this process a durable alternative PWB surface finish. The thin layer of immersion gold
preserves the solderability of the finish by preventing oxidation of the highly active nickel
surface. Nickel/gold finishes can typically withstand as many as six or more thermal excursions
(heating cycles) during assembly without losing solderability.
This process can be operated in either a horizontal, conveyorized or vertical, non-
conveyorized mode. A nickel/gold finish is compatible with SMT, flip chip, and ball grid array
(EGA) technologies, as well as with typical through hole components. The thin layer of gold
makes the surface aluminum wire-bondable, with thicker gold deposits also allowing gold wire-
bonding. The high plating temperatures and low pH of the nickel/gold plating process can be
incompatible with solder masks with high acrylic content, although solder masks high in epoxy
content are unaffected by the plating solution. Nickel/gold plated boards have a shelf life of up to
two years or more.
A flow diagram of the process baths in a typical nickel/gold process is presented in Figure
2-3, followed by a brief description of each of the process steps.
2-7
-------
1. Cleaner 1
V
Water Rinse x 1 1
t
2. Microetch 1
1
Water Rinse x 1 1
1
3. Catalyst 1
V
Water Rinse x 1 1
1
4- Acid Dip 1
1
Water Rinse x 1 1
t
5. Electro less Nickel 1
t
Water Rinse x 2 1
t
6. Immersion Gold 1
1
Water Rinse x 2 1
Figure 2-3. Nickel/Gold Process Flow Diagram
Step 1: Cleaner: Grease, contaminants, and any organic solder mask residues are
removed from the PWB surface in an acidic cleaner solution. The cleaner
provides a clean, consistent copper surface to ensure uniform etching and prepares
the board for application of the palladium catalyst.
2-8
-------
Step 2: Microetch: The microetch solution lightly etches the exposed copper surfaces of
the panel, including the barrels of the PTHs, to remove any chemical
contamination and metal oxides present.
Step 3: Catalyst: The catalyst consists of a palladium salt in an acidic solution. Palladium
ions are deposited onto the surface of a PWB in a displacement reaction,
effectively exchanging the surface copper layer for palladium, thus forming a
catalytic layer for subsequent nickel plating.
Step 4: Acid Dip: The acid dip, usually a weak sulfuric or hydrochloric acid, removes any
residual catalyst from the non-copper surfaces of the PWB, to prohibit plating on
the solder mask or other unwanted areas of the board.
Step 5: Electroless Nickel: An electroless nickel solution is used to plate a layer of nickel
onto the surface of the palladium-covered areas in a high temperature, acidic bath.
The electroless nickel solution contains a source of nickel ions, phosphorous, and
a reducing agent, which is typically either sodium hypophosphite or
dimethylamine borane. In the presence of the palladium, the reducing agent
provides electrons to the positively charged nickel ions, causing reduction of the
nickel and the deposition of elemental nickel onto the exposed palladium catalyst
(Parquet and Sedacca, 1996). Phosphorous is co-deposited with the nickel, and
the resulting nickel-phosphorous alloy forms a corrosion-resistant layer protecting
the underlying copper. Because the bath is autocatalytic, it will continue plating
until the panel is removed from the nickel bath. Nickel layer thicknesses for
PWBs are typically 3 to 5 microns (120 to 200 microinches).
Step 6: Immersion Gold: A very thin, protective layer of pure gold is deposited onto the
surface of the nickel in the immersion gold plating bath. A chemical displacement
reaction occurs, depositing the thin layer of gold onto the metal surface while
displacing nickel ions into the solution. Because the reaction is driven by the
electrochemical potential difference between the two metals, the reaction ceases
when all of the surface nickel has been replaced with gold. Gold layer thicknesses
are typically 0.2 microns (8 microinches), but can be increased to allow gold wire-
bonding of the final surface.
Although electroless nickel plating processes all require the presence of a catalyst to plate
nickel onto a copper surface, the catalyst can at times be too aggressive and catalyze areas where
plating is undesirable, such as areas of fine pitch circuitry, causing unintended short-circuiting.
This problem is handled successfully (with typically less than a 5 percent failure rate) by
introducing the panel to an acid dip after the catalyst bath, as described above (Kerr, 1999). The
acid dip removes the unintended palladium salt deposits, without harming the elemental
palladium deposited onto the copper surfaces.
2-9
-------
A second method employed by some manufacturers is to use a less active catalyst, which
tends not to bridge fine pitch circuitry or adhere onto solder mask-covered PWB surfaces. A
ruthenium-based catalyst is used to deposit a ruthenium seed layer, in place of the more typical
palladium-based catalysts. A nickel surface is then plated to the ruthenium seed layer using a
sodium-hypophosphite-reduced nickel plating chemistry, until the desired nickel thickness is
obtained. The gold is then applied as described above.
Nickel/Palladium/Gold
The nickel/palladium/gold process is similar to the nickel/gold process described above,
except it uses a palladium metal layer that is deposited after the nickel layer, but prior to
depositing the final gold layer. The palladium layer is much harder than gold, providing added
strength to the surface finish for wirebonding and connector attachment, while protecting the
underlying nickel from oxidation.
The process can be operated in either a horizontal, conveyorized, or a vertical, non-
conveyorized mode. A nickel/palladium/gold finish is compatible with SMT, flip chip, and EGA
technologies, as well as with typical through hole components. The finish is also both gold and
aluminum wire-bondable. The high plating temperatures and low pH of the
nickel/palladium/gold plating process can be incompatible with solder masks with high acrylic
content, although solder masks high in epoxy content are unaffected by the plating solution.
Nickel/palladium/gold-plated boards can withstand as many as six thermal excursions during
assembly, and have a shelf life of up to two years or more.
A flow diagram of the process steps in a typical HASL process is presented in Figure 2-4.
A brief description of each of the process steps is also given.
Steps 1- 4: Cleaner/Microetch/Catalyst/Acid Dip: PWBs are cleaned, microetched, and a
palladium catalyst is applied to the exposed copper surfaces in a chemical process
similar to the one described previously for nickel/gold. An acid dip is then used to
remove the catalyst from areas of the board where plating is undesirable.
Step 5: Electroless Nickel: An electroless nickel solution plates a layer of nickel onto the
surface of the thin, initial nickel deposit. The electroless nickel bath is a slightly
alkaline solution containing a source of nickel ions, and a sodium hypophosphite
reducing agent. The reducing agent provides electrons to the positively charged
nickel ions, causing the reduction of the nickel and the plating of elemental nickel
onto the exposed nickel-boron layer. Phosphorous is co-deposited with the
nickel, causing the formation of a corrosion resistant layer of nickel-phosphorous
alloy that protects the underlying copper. Because the bath is autocatalytic, it will
continue plating until the panel is removed from the nickel bath. Nickel layer
thicknesses are typically 3 to 5 microns (120 to 200 microinches).
2-10
-------
1 Cleaner I
1
Water Rinse x 2 |
1
2- Microetch 1
1
Water Rinse x 2 1
*
3 Catalyst 1
1
Water Rinse x 2 1
1
4- Acid Dip 1
1
Water Rinse x 2 1
1
Electroless Nickel 1
1
Water Rinse x 2 1
1
Preinitiator 1
V
Electroless Palladium 1
y
Water Rinse x 2 1
t
Immersion Gold 1
y
Water Rinse x 2 1
Figure 2-4. Nickel/Palladium/Gold Process Flow Diagram
2-11
-------
Step 6: Preinitiator: The preinitiator reactivates the nickel surfaces by using a mineral acid
to remove oxide from the surface of the nickel. In addition, the preinitiator
deposits trace quantities of a catalytic metal that promotes homogeneous
palladium deposition, ensuring that all nickel surfaces begin plating quickly and at
the same time.
Step 7: Electroless Palladium: The electroless palladium bath deposits a thin layer of
palladium onto the nickel-covered circuitry through an oxidation-reduction
reaction. Hypophosphite or formate is used as the reducing agent, providing
electrons to the positively charged palladium ions, resulting in the plating of
palladium onto the nickel surfaces of the PWB. Palladium layer thicknesses are
typically 0.3 to 0.8 microns (12 to 32 microinches).
Step 8: Immersion Gold: A very thin, protective layer of pure gold is deposited onto the
surface of the palladium in the immersion gold plating bath. A chemical
displacement reaction occurs, depositing the thin layer of gold onto the metal
surface while displacing palladium ions into the solution. Because the reaction is
driven by the potential difference of the two metals, the reaction ceases when all
of the surface palladium has been replaced with gold. Gold layer thickness is
typically 0.2 microns (8 microinches).
Organic Solderability Preservative (OSP)
The OSP process selectively applies a flat, anti-oxidation film onto the exposed copper
surfaces of the PWB to preserve the solderability of the copper. This coating reacts with the
copper in an acid and water mixture to form the nearly invisible protective organic coating. OSP
processes can be based on benzimidazole chemistries that deposit thicker coatings, or on
benzotriazoles and imidazoles chemistries which deposit thinner coatings. The thicker OSP
coatings, which are evaluated in this CTSA, can withstand a minimum of three and up to as many
as five thermal excursions while still maintaining coating integrity. Coating thicknesses of 0.1 to
0.5 microns (4 to 20 microinches) are typical for the thicker coatings, as opposed to the
monomolecular layer formed by the thinner OSPs.
The process is typically operated in a horizontal, conveyorized mode but can be modified
to run in a vertical, non-conveyorized mode. OSP processes are compatible with SMT, flip chip,
and EGA technologies, as well as with typical through hole components. The OSP surface finish
cannot be wirebonded. OSP surfaces are compatible with all solder masks, can withstand 3 to 4
thermal excursions during assembly, and have a shelf life of up to one year; extended shelf life
times may result in a degradation of the coating.
A flow diagram of the process baths in a typical OSP process is presented in Figure 2-5,
followed by a brief description of each of the process steps.
2-12
-------
Cleaner
Water Rinse x 1
Microetch
Water Rinse x 1
Air Knife
OSP
Air Knife
Water Rinse x 1
Dry
Figure 2-5. OSP Process Flow Diagram
Step 1: Cleaner: Surface oils and solder mask residues are removed from the exposed
copper surfaces in a cleaner solution. The acidic solution prepares the surface to
ensure the controlled, uniform etching in subsequent steps.
Step 2: Microetch: The microetch solution, typically consisting of dilute hydrochloric or
sulfuric acid, etches the existing copper surfaces to further remove any remaining
contaminants and to chemically roughen the surface of the copper to promote
coating adhesion.
Step 3: Air Knife: An air knife removes excess water from the panel to limit oxidation
formation on the copper surfaces prior to coating application. This step also
minimizes drag-in of sulfates, which are harmful to the OSP bath.
2-13
-------
Step 4: OSP: A protective layer is formed selectively on the exposed copper surfaces by
the OSP in an acidic aqueous bath. The deposited protective layer chemically
bonds to the copper, forming an organometallic layer that preserves the
solderability of the copper surface for future assembly (Mouton, 1997).
Step 5: Air Knife: An air knife removes excess OSP from the panel and promotes even
coating across the entire PWB surface. The air knife also minimizes the chemical
losses through drag-out from the OSP bath.
Step 6: Dry: A warm-air drying stage cures the OSP coating and helps to remove any
residual moisture from the board.
Immersion Silver
The immersion silver process promotes the selective deposition of silver onto the exposed
copper surfaces of the PWB through a chemical displacement reaction. Silver surfaces are
protected from tarnishing by a co-deposited organic inhibitor that forms a hydrophobic layer on
top of the silver, thus preserving the coating's solderability. The final silver finish thickness is
typically 0.1 microns (3 to 4 microinches). The silver process submitted for evaluation is
operated exclusively as a horizontal, conveyorized process, however the process may be operated
in either vertical or horizontal mode. Immersion silver finishes are compatible with SMT, flip
chip, and EGA technologies, as well as with typical through hole components. They are also
both gold and aluminum wire-bondable. Silver finishes are compatible with all types solder
masks, can withstand up to five thermal excursions during assembly, and have a shelf life of at
least six months.
A flow diagram of the process steps in a typical HASL process is presented in Figure 2-6.
A brief description of each of the process steps is also given.
Step 1: Cleaner: An acid-based cleaner removes surface oils, oxides, and any organic
residues left after the solder mask application. The cleaner provides a clean,
consistent copper surface to ensure uniform etching.
Step 2: Microetch: The microetch solution lightly etches the exposed copper surfaces of
the panel, including the barrels of the PTHs, to remove any chemical
contamination and metal oxides present.
Step 3: Predip: Etched panels are processed through a predip solution prior to silver
deposition to remove any surface oxidation that may have occurred in the
previous rinse stage. The predip, which is chemically similar to that of the silver
deposition bath, is also used to protect the bath from any harmful drag-in
chemicals that may be detrimental to the deposition bath.
2-14
-------
Cleaner
Water Rinse x 1
Microetch
Water Rinse x 1
Predip
Immersion Silver
Water Rinse x 1
5
Dry
Figure 2-6. Immersion Silver Process Flow Diagram
Step 4: Immersion Silver: The immersion silver bath is a pH-neutral solution that
selectively deposits a 0.1 micron (3 to 4 microinch) layer of silver onto all of the
exposed copper surfaces of the PWB. Coating proceeds by a simple displacement
reaction, with silver ions displacing copper ions from the surface. The liberated
copper ions are benign to the bath chemistry and thus do not inhibit the bath
effectiveness as copper concentrations increase. Because the bath is an immersion
process, plating is self-limiting and will cease when the entire copper surface has
been coated.
Step 5: Dry: A drying stage removes any residual moisture from the board to prevent
staining and to ensure metal quality in the through holes.
2-15
-------
Immersion Tin
The immersion tin process utilizes a thiorea-based reducing agent to create an
electrochemical potential between the surface and stannous ions in solution, causing the
reduction of a layer of tin onto the copper surfaces of the PWB. An organo-metallic compound,
which is co-deposited along with the tin, acts to retard the formation of a tin-copper intermetallic
layer, preserving the solderability of the finish. The organo-metallic compound also inhibits the
formation of tin whiskers (i.e., dendritic growth). The process is typically operated in a
conveyorized fashion, but can be modified to run in a vertical, non-conveyorized mode.
Immersion tin surfaces are compatible with SMT, flip chip, EGA technologies, and typical
through hole components. The immersion tin surface cannot be wirebonded. Tin surfaces are
compatible with all solder masks, have a reported shelf life of one year and can typically
withstand a minimum of five thermal excursions during assembly.
A flow diagram of the process steps in a typical immersion tin process is presented in
Figure 2-7. A brief description of each of the process steps is given.
1 • Cleaner 1
Y
Water Rinse x 2 1
Y
2- Microetch 1
Y
Water Rinse x 2 1
Y
3- Predip 1
Y
Water Rinse x 1 1
Y
4. Immersion Tin 1
Y
Water Rinse x 2 1
Y
5 Dry I
Figure 2-7. Immersion Tin Process Flow Diagram
2-16
-------
Step 1: Cleaner: Surface oils and solder mask residues are removed from the exposed
copper surfaces in a cleaner solution. The acidic solution prepares the surface to
ensure controlled, uniform etching.
Step 2: Microetch: A microetch solution, typically consisting of dilute hydrochloric or
sulfuric acid, removes any remaining contaminants from the copper surface. The
etching also chemically roughens the copper surface to promote good tin-to-
copper adhesion.
Step 3: Predip: Etched panels are processed through a predip solution that is chemically
similar to that of the tin bath, thus protecting the plating bath from harmful drag-in
chemicals.
Step 4: Immersion Tin: A tin plating bath deposits a thin layer of tin onto the exposed
copper circuitry through a chemical displacement reaction that deposits stannous
ions while displacing copper ions into the plating solution. The bath is considered
self-limiting, because plating continues only until all the copper surfaces have
been coated with a tin deposit. The presence of a complexing agent, thiourea,
prevents the copper from interfering with the plating process. The complexed
copper is removed as a precipitate from solution by decantation.
Step 5: Dry: A drying stage removes any residual moisture from the board to prevent
staining and to ensure high metal quality in the through holes.
2.1.4 Chemical Characterization of Surface Finishing Technologies
This section describes the sources of bath chemistry information, methods used for
summarizing that information, and the use of bath chemistry data. Publicly-available
information, along with proprietary chemical information obtained from the chemical suppliers,
was used to assess exposure, risk, and cost for the processes. This section does not identify any
proprietary ingredients. Generic names have been submitted for the names of proprietary,
confidential chemicals to mask their identity.
Use of Chemical Product and Formulation Data
Assessment of releases, potential exposure, and characterizing risk for the surface
finishing technologies requires chemical-specific data, including concentrations for each chemical
in the various process baths. Although some bath chemistry data were collected in the PWB
Workplace Practices Questionnaire, the decision was made not to use this data because of
inconsistencies in the responses to questions pertaining to bath chemistry. Instead, the suppliers
participating in the Performance Demonstration each submitted complete chemical formulations
along with other publicly-available information on their respective product lines. This
information includes:
2-17
-------
material safety data sheets (MSDSs);
Product Data Sheets;
proprietary chemical product formulations; and
patent data, in isolated cases.
The chemical formulations identify the chemicals and concentrations present in the
chemical products while the MSDS provides physical property and worker hazard information on
the entire formulation. The Product Data Sheets describe how those products are mixed together
to make up the individual process baths. Patent information, when available, provided insight
into the mechanisms for chemical activity.
Table 2-1 presents all of the chemicals identified in surface finishing process lines and the
technologies in which they were used. Generic names have been substituted for the names of
proprietary, confidential chemicals to mask their identity. Although the confidential formulations
included all of the chemicals listed below, a chemical was considered publicly-available if it was
listed on a MSDS or patent.
Table 2-1. Use Cluster Chemicals and Associated Surface Finishing Processes
Chemical
Acetic acid
Aliphatic acid A
Aliphatic acid B
Aliphatic acid D
Aliphatic acid E
Aliphatic dicarboxylic acid A
Aliphatic dicarboxylic acid C
Alkylalkyne diol
Alkylamino acid A
Alkylamino acid B
Alkylaryl imidazole
Alkylaryl sulfonate
Alkyldiol
Alkylimine dialkanol
Alkylphenol ethoxylate
Alkylphenol polyethoxyethanol
Alkylpolyol
Amino acid salt
Amino carboxylic acid
Ammonium chloride
Ammonia compound A
Ammonia compound B
Ammonium hydroxide
HASL
X
X
X
X
X
Nickel/
Gold
X
X
X
X
X
X
X
X
X
X
X
Nickel/
Palladium/
Gold
X
X
X
X
X
X
X
X
X
X
X
X
OSP
X
X
Immersion
Silver
X
Immersion
Tin
X
X
X
X
X
X
2-18
-------
Chemical
Aromatic imidizole product a
Arylphenol
Bismuth compound
Citric acid
Copper ion
Copper salt C
Copper sulfate pentahydrate
Cyclic amide
Ethoxylated alkylphenol
Ethylenediamine
Ethylene glycol
Ethylene glycol monobutyl ether
Fatty amine
Fluoboric acid
Gum
Hydrochloric acid
Hydrogen peroxide
Hydroxy carboxylic acid
Hydroxyaryl acid
Hydroxyaryl sulfonate
Inorganic metallic salt A
Inorganic metallic salt B
Inorganic metallic salt C
Lead
Maleic acid
Malic acid
Methane sulfonic acid
Nickel sulfate
Nitrogen acid
Nonionic surfactant a
Palladium chloride
Palladium salt
Phosphoric acid
Potassium compound
Potassium gold cyanide
Potassium peroxymonosulfate
Propionic acid
Quantenary alkylammonium chlorides
Silver nitrate
HASL
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Nickel/
Gold
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
Nickel/
Palladium/
Gold
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
OSP
X
X
X
X
X
X
X
X
X
X
X
X
X
Immersion
Silver
X
X
X
X
X
X
Immersion
Tin
X
X
X
X
X
X
X
X
X
X
X
X
2-19
-------
Chemical
Silver salt
Sodium benzene sulfonate
Sodium hydroxide
Sodium hypophosphite
Sodium hypophosphite mono hydrate
Sodium phosphorus salt
Sodium salt
Stannous methane sulfonic acid
Substituted amine hydrochloride
Sulfuric acid
Surfactant a
Thiourea
Tin
Tin chloride
Transition metal salt a
Unspecified tartrate
Urea
Urea compound B
Urea Compound C
Vinyl polymer
HASL
X
X
X
X
Nickel/
Gold
X
X
X
X
X
X
X
X
Nickel/
Palladium/
Gold
X
X
X
X
X
X
X
X
OSP
X
X
Immersion
Silver
X
X
Immersion
Tin
X
X
X
X
X
X
X
X
X
X
X
a Dropped due to insufficient identification.
Determining Chemical Formulations
Determining the chemical formulations for each process step is critical for evaluating each
surface finishing technology. Each surface finishing product line submitted for evaluation was
divided into basic bath steps common to all the processes within that surface finishing category
(e.g., both OSP product lines submitted were divided into cleaner, microetch, and OSP baths).
The basic bath steps were combined to form a process flow diagram specific to each surface
finishing technology, as shown in Figure 2-1. The recommended formula for creating a new
bath, along with the individual formulations for each chemical product, were combined to
determine the individual chemical concentrations in the final bath. The individual chemical
concentrations in the baths were calculated by:
2-20
-------
Cb = (CcHEM)(CFORM)(D)(1000cm3/L)
where,
Cb = concentration of constituent in bath (g/L)
CCHEM = chemical concentration, by weight, in the product, from chemical product
formulations obtained from chemical suppliers (%)
CFORM = proportion of the product formulation volume to the total bath volume,
from Product Data Sheets (%)
D = density of the product (g/cm3)
An example calculation for the ethylene glycol concentration in the cleaner bath is shown
below for one supplier's OSP process. Each product's formulation lists the chemicals that are
contained in that product on a weight percentage basis. For ethylene glycol, this is 40 percent, or
40 grams ethylene glycol per 100 grams of product (CCHEM)- The supplier's Product Data Sheet
lists how much of that chemical product is used in the total bath make-up on a volume
percentage basis: in this case, ten percent, or ten liters of product per 100 liters of the total bath
(CFORM)- The remaining volume in the bath is made up of deionized water. The MSDS for the
product lists the specific gravity or density (D) of the product, which was multiplied by the
weight and volume percentages above to obtain the bath concentration (Cb) for that constituent.
(In some cases, the Product Data Sheets list chemicals or product packages on a mass per volume
basis. This was multiplied by the weight percentage from the MSDS for that product package to
obtain a concentration in the bath.) The example calculation is shown here:
After the product formulation and Product Data Sheet data were combined in the above
manner for each supplier's product line, a list of chemicals in each surface finishing technology
category (HASL, OSP, etc.) was compiled. This list shows all the chemicals that might be in each
bath, by technology, as well as the concentration range for each chemical. However, some of the
alternatives (e.g., OSP, nickel/gold, and immersion tin) have more than one chemical supplier
using different bath chemistries. It was decided to include all of the identified chemicals in the
formulations rather than selecting a typical or "generic" subset of chemicals.
Estimated concentration ranges (low, high, and average) were determined based on the
formulation data and are presented in Appendix B. Concentrations are for each bath in each
surface finishing process alternative.
Data Limitations
Limitations and uncertainties in the chemical characterization data arise primarily from
side reactions in the baths. Side reactions in the baths may result in changing concentrations over
time and/or formation of additional chemicals in the baths. This information is not reflected in
product formulation data, MSDSs or Product Data Sheets, but would affect bath concentrations
over time. As a result, bath concentrations are estimated; actual chemical constituents and
concentrations will vary by supplier and facility.
2-21
-------
In cases where the formulation data was reported as a " < " or " > " value, the reported
values were assumed in calculating bath concentrations. For example, if "< 5 percent" was
reported for a constituent by a product formulation, it is assumed that product contained 5
percent by weight (or volume, where appropriate) of that constituent. Also, some data were
reported as ranges. In these cases, mid-points for the ranges were used to estimate bath
concentrations (e.g., if 20 to 30 percent by weight was reported, 25 percent by weight was
assumed).
Chemical Properties
Appendix C contains chemical properties data for each of the non-proprietary chemicals
identified in surface finishing baths. For example, properties listed include molecular weight,
vapor pressure, solubility, Henry's Law Constant, and octanol-water partition coefficient. Basic
chemical properties information for each chemical is followed by a summary description of fate
and transport mechanisms for that chemical. In order to protect the identity of confidential
chemicals, chemical properties data was not included for proprietary chemicals.
2-22
-------
2.2 ADDITIONAL SURFACE FINISHING TECHNOLOGIES
The surface finishing technologies described in Section 2.1 represent the technologies that
were evaluated in the CTSA. However, additional surface finishing technologies exist which
were not evaluated in the CTSA for one or more of the following reasons:
• a product line was not submitted for the technology by any chemical supplier;
• the technology was not available to be tested in the Performance Demonstration; or
• the technology has only recently been commercialized since the evaluation began or was
submitted too late to be included in the evaluation.
Despite not being evaluated, these technologies are important because they are alternative
methods for surface finishing that accomplish the removal of lead from PWB manufacturing,
which is a goal of the PWB manufacturing industry. A brief description of one surface finishing
technology not evaluated in this CTSA is presented below. Other technologies may exist, but
they have not been identified by the project.
2.2.1 Immersion Palladium
The immersion palladium process uses a three step process to deposit a thin surface finish
of palladium on the exposed copper traces of the PWB. The process is similar to other wet
processes presented earlier in this chapter. It consists of a series of chemical baths separated by a
series of water rinse steps. The recommended bath sequence for the immersion palladium
process is as follows:
• cleaner;
• water rinse;
• microetch;
• water rinse;
• immersion palladium;
• water rinse; and
• dry.
A mild alkaline cleaner is first used to clean the surface of copper, removing oil and debris
from the boards' surface. The copper is then lightly etched to remove any copper oxide by the
microetch, providing a pristine surface for palladium deposition. Finally, a three microinch layer
of palladium is deposited onto the board by the immersion palladium bath via a chemical
displacement reaction. During the reaction, palladium ions are deposited onto only the exposed
copper surfaces of the board, displacing copper ions into the plating solution. Like other
immersion processes, the palladium deposition is self-limiting, halting once all of the exposed
copper has been covered by a layer of palladium. The displaced copper remains in solution,
continuing to build in concentration, until an electrolyte in the bath causes the copper to
precipitate out of solution, usually at a concentration of greater than 150 parts per million. The
precipitate is then filtered out of the bath. The bath can be operated without replacement as long
as the electrolyte and palladium content are maintained (Sedlak, 2000).
2-23
-------
The immersion palladium process is typically operated in a vertical, non-conveyorized
mode but can be modified to run in a horizontal, conveyorized mode. Immersion palladium
finishes are compatible with SMT, flip chip, and EGA technologies, as well as with typical
through hole components. The finish is also gold wire-bondable. Immersion palladium finishes
are sensitive to some of the more aggressive fluxs, so milder fluxes (e.g., no-clean fluxes) are
recommended. They can withstand four thermal excursions during assembly, and have a shelf
life of at least 12 months. The immersion palladium process has been run successfully at two
prototype facilities. However, the process could not be evaluated by the project because it could
not be tested under full production at the time of the Performance Demonstration.
2-24
-------
REFERENCES
Carroll, Thomas. 1999. Senior Staff Engineer, Hughes Space Communications Co., Materials
and Processes Operations. Electronic Communication to Abbey Hoates at U.S. EPA. June.
Kerr, Michael. 1999. Circuit Center, Inc. Written communication to Kathy Hart at U.S. EPA.
July.
Mouton, Roger. 1997. OSP Technology Improvements. Circuitree. March.
Parquet, Dan T. and Spence A. Sedacca. 1996. Electroless Ni/Au as an HASL alternative.
Electronic Packaging and Production. March.
Sedlak, Rudy. 2000. RD Chemical Company. Electronic communication with Jack Geibig,
University of Tennessee Center for Clean Products and Clean Technologies. July.
2-25
-------
Chapter 3
Risk Screening and Comparison
This chapter of the Cleaner Technologies Substitutes Assessment (CTSA) addresses the
health and environmental hazards, exposures, and risks that may result from using a surface
finishing technology. The information presented here focuses entirely on the surface finishing
technologies. It does not, nor is it intended to, represent the full range of hazards or risks that
could be associated with printed wiring board (PWB) manufacturing. This risk evaluation is a
screening-level assessment of multiple chemicals belonging to the surface finishing use cluster,
and is presented as a screening level rather than a comprehensive risk characterization, both
because of the predefined scope of the assessment and because of exposure and hazard data
limitations. The intended audience of this risk screening and comparison is the PWB industry
and others with a stake in the practices of this industry.
Section 3.1 identifies possible sources of environmental releases from surface finishing
and, in some cases, discusses the nature and quantity of those releases. Section 3.2 assesses
occupational and general population (i.e., the public living near a PWB facility; fishing streams
that receive wastewater from PWB facilities) exposures to surface finishing chemicals. This
section quantitatively estimates inhalation and dermal exposure to workers and inhalation
exposure to the public living near a PWB facility. Section 3.3 presents human health hazard and
aquatic toxicity data for surface finishing chemicals. Section 3.4 characterizes the risks and
concerns associated with the exposures estimated in Section 3.2. In all of these sections, the
methodologies or models used to estimate releases, exposures, or risks are described along with
the associated assumptions and uncertainties. Finally, Section 3.5 summarizes chemical safety
hazards from material safety data sheets (MSDSs) for surface finishing chemical products and
discusses process safety issues.
3.1 SOURCE RELEASE ASSESSMENT
The Source Release Assessment uses data from the PWB Workplace Practices
Questionnaire, together with other data sources, to identify sources and amounts of
environmental releases. Both on-site releases (e.g., evaporative or fugitive emissions from the
process) and off-site transfers (e.g., off-site recycling) are identified and, for those where
sufficient data exist from the questionnaire, numerical results are presented. The objectives of the
Source Release Assessment are to:
• identify potential sources of releases;
• characterize the source conditions surrounding the releases, such as a heated bath or the
presence of local ventilation; and
• characterize, where possible, the nature and quantity of releases under the source
conditions.
3-1
-------
Many of the releases may be mitigated and even be prevented through pollution prevention
techniques and good operating procedures such at those described in Chapter 6, Additional
Environmental Improvement Opportunities. However, they are included in this assessment to
illustrate the range of releases that may occur from surface finishing processes.
A material balance approach was used to identify and characterize environmental releases
associated with day-to-day operation of surface finishing processes. Air releases and releases of
organics to surface waters, which could not be quantified from the questionnaire data, are
modeled in Section 3.2, Exposure Assessment.
Section 3.1.1 describes the data sources and assumptions used in the Source Release
Assessment. Section 3.1.2 discusses the material balance approach used, release information,
and data pertaining to all surface finishing process alternatives. Section 3.1.3 presents source and
release information and data for specific surface finishing process alternatives. Section 3.1.4
discusses uncertainties in the Source Release Assessment.
3.1.1 Data Sources and Assumptions
This section presents a general discussion of data sources and assumptions for the Source
Release Assessment. Sections 3.1.2 and 3.1.3 present more detailed information about specific
inputs and releases for individual surface finishing alternatives.
Sources of data used in the Source Release Assessment include:
industry data collection forms, such as the PWB Workplace Practices Questionnaire and
Performance Demonstration Observer Data Sheets (Appendix A, Data Collection Sheets);
supplier-provided data, including bath chemistry data and supplier Product Data Sheets
describing how to mix and maintain baths (Appendix B, Publicly-Available Bath
Chemistry Data);
engineering estimates; and
• DfE PWB Proj ect publication, Printed Wiring Board Pollution Prevention and Control
Technologies: Analysis of Updated Survey Results (U.S. EPA, 1998a).
Bath chemistry data were collected in the PWB Workplace Practices Questionnaire, but
these data were not used due to inconsistencies in the responses to questions pertaining to bath
chemistry. Instead, surface finishing chemical suppliers participating in the Performance
Demonstration submitted confidential chemical formulation data along with publicly-available
Product Data Sheets on their respective product lines. Bath concentration ranges were
determined based on this information using the method discussed in Section 2.1.4, Chemical
Characterization of Surface Finishing Technologies. A general description of the PWB
Workplace Practices Questionnaire, including its distribution and overall general results, is
presented in Section 1.3.4, Primary Data Sources.
3-2
-------
Several assumptions or adjustments were made to put the PWB Workplace Practices
Questionnaire data into a consistent form for all surface finishing technologies. These include the
following:
Data reported on a daily basis were converted to an annual basis using the number of
days per year of process operation (Appendix A, questions 2.2 and 3.2). For data on a
weekly or monthly basis, 12 months per year and 52 weeks per year were assumed.
Data reported on a per shift basis was converted to a per day basis using the number of
hours per day the process was in operation, when available. Eight hours of operation was
assumed to be equivalent to one shift.
Bath names provided by questionnaire respondents were revised to be consistent with the
generic surface finishing process descriptions provided in Section 2.1.3, Chemistry and
Process Descriptions of Surface Finishing Technologies.
There were wide variations in submitted data due to the differences in size of PWB
facilities. To adjust for this, data are presented here both as reported in the questionnaire (usually
as an annual quantity consumed or produced), as well as normalized by annual surface square
feet (ssf) of PWB produced by the individual surface finishing technology. Normalizing the data,
however, may not fully account for possible differences in processing methods that could result
from different production levels.
3.1.2 Overall Material Balance for Surface Finishing Technologies
A general material balance is presented here to identify and characterize inputs and
potential releases from the surface finishing process alternatives. Due to limitations and gaps in
the available data, no attempt was made to perform a quantitative mass balance of inputs and
outputs. This approach is still useful, however, as an organizing tool for discussing the various
inputs to, and outputs from, surface finishing processes, and presenting the available data. Figure
3-1 depicts inputs to a generalized surface finishing process line, along with possible outputs,
including PWB product, solid waste, air emissions, and wastewater discharges.
Many PWB manufacturers have an on-site wastewater treatment system for pretreating
wastewater prior to direct discharge to a stream or lake, or indirect discharge to a publicly owned
treatment works (POTW). Figure 3-2 describes a simplified PWB wastewater treatment system,
including the inputs and outputs of interest in the Source Release Assessment.
3-3
-------
Bath chemicals
- bath replacement
- bath additions
Etched and solder
mask-coated PWBs
Water
- rinse water
- equipment cleaning
- bath make-up
Cleaning chemicals
- chemical flush
- equipment cleaning
A,
Evaporation and
aerosols from baths
- from surface
- from air sparging
Evaporation
from drying/
ovens
Surface
Process
Finishing
Line
Chemical
reaction
Wastewater
(primarily from
rinse tanks)
Spent bath solutions
(include waste
equipment cleaning
chemicals)
System Boundary
Physical Bath Boundary
Chemicals
incorporated
onto PWBs
Hazardous solid waste
- filters
- precipitates
- container residues
Non-hazardous solid waste
- filters
- precipitates
- container residues
Bath chemicals
-sampling,
-bail-out
Drummed solid or
liquid waste
I
r
On- or off-site
recycle or
disposal
On- or off-site wastewater treatment
or disposal (see Figure 2)
Figure 3-1. Schematic of Overall Material Balance for Surface Finishing Technologies
3-4
-------
W,
Spent bath solutions (include wasted)
equipment cleaning chemicals)
W,
Wastewater
(primarily from rinse tank)
I Bath chemicals I /
- sampling I /
- bail-out | ^
System Boundary
Storage Tank
Wastewater
Treatment
System
Treatment
Sludge to recycle
or disposal
Discharge to POTW or
Receiving Stream
Figure 3-2. Wastewater Treatment Process Flow Diagram
Inputs
Possible inputs to a surface finishing process line include process chemicals and materials,
etched and solder mask-coated PWBs that have been processed through previous PWB
manufacturing process steps, water, and cleaning chemicals.
The total inputs for the process are described by the equation:
where,
Ii
I2
I3
I4
I,o,al= Il
bath chemicals
etched and solder mask-coated PWBs
water
cleaning chemicals
These terms are discussed below.
\ Bath chemicals. This includes chemical formulations used for initial bath make-up, bath
bailout and additions, and bath replacement. Bath formulations and the chemical
constituents of those formulations were characterized based on Product Data Sheets and
bath formulation data provided by the chemical suppliers. A detailed description of the
3-5
-------
calculation of bath chemical concentrations is presented in Section 2.1.4, Chemical
Characterization of Surface Finishing Technologies. Calculated chemical bath
concentrations are reported in Appendix B. PWB manufacturers were asked to report the
quantity of surface finishing chemicals they use annually in the PWB Workplace Practices
Questionnaire. However, the resulting data were variable and poor in quality, preventing
the quantification of total chemical usage for process chemicals.
Etched and solder mask-coated PWBs. PWBs with solder mask-coated copper circuitry
that enter the surface finishing line could lose a small amount of copper to the process line
due to etching and dissolution. Trace amounts of other additives such as arsenic,
chromium, and phosphate may also be lost to the process. This applies to all surface
finishing alternatives where copper is etched off the boards in the microetch bath at the
beginning of the process.
PWB panels are the only source of copper for the surface finishing process. The rate at
which the copper is lost can vary depending on process conditions (e.g., bath
temperature, chemical concentration of bath, etc.) and the type of bath (whether a
microetch bath or a plating bath). The amount of copper lost through etching and
through displacement plating mechanisms is expected to be small, relative to other
chemical additions. This input is not quantified.
Water. Water, usually deionized, is used in the surface finishing process for rinse water,
bath make-up, and equipment cleaning. The water consumption of surface finishing
technologies varies according to the number and size of rinse tanks used by the process.
However, the number of rinse tanks can also vary from facility to facility within a
technology category due to differences in facility operating procedures, rinse
configuration, and water conservation measures.
Water usage data collected by the PWB Workplace Practices Questionnaire include the
daily volume of water used for rinse water and bath make-up. Daily water usage in
gallons was converted to annual water usage by multiplying by the number of days per
year the process was in operation. The value was then normalized by dividing the annual
water usage in gallons by the annual production in ssf of PWB produced for the same
line. Both annual and normalized water consumption data from the questionnaire for
each surface finishing technology are summarized in Table 3-1.
From the normalized data it can be seen that the nickel/gold and nickel/palladium/gold
processes consume more water per ssf than the other technologies. The increased water
consumption is due to the bath sequences of these technologies which are typically longer
and thus use more rinse tanks. Drawing other conclusions from this data is difficult,
given the variation in PWB throughput between reporting facilities and the relatively few
number of responses within some technology categories.
3-6
-------
Table 3-1. Water Usage of Surface Finishing Technologies From Questionnaire
Process Type
No. of Responses
Water Usage (I3)
(thousand gal/year)a
Water Usage (I3)
(gal/ssf)
HASL
Non-conveyorized
Conveyorized
6
17
0.3 - 750 (254)
910 - 3,740 (1,250)
0.970
4.89
Nickel/Gold
Non-conveyorized
8
17 - 1,620 (538)
101
Nickel/Palladium/Gold
Non-conveyorized
2
216- 1,710(961)
164
OSP
Non-conveyorized
Conveyorized
5
5
42-150(89.1)
8-1,580 (440)
1.93
14.3
Immersion Silver
Conveyorized
2
698 - 1,120 (907)
36.8
Immersion Tin
Non-conveyorized
Conveyorized
4
2
3.3-385 (209)
11.5-199(105)
11.0
0.333
a Average values from the PWB Workplace Practices Questionnaire data are shown in parentheses. Refer to Section
1.3.4 for a detailed discussion of questionnaire responses.
I4 Cleaning chemicals. This includes chemicals used for conveyor equipment cleaning,
tank cleaning, chemical flushing, rack cleaning, and other cleaning pertaining to the
surface finishing process line. Data were collected by the PWB Workplace Practices
Questionnaire regarding the use of chemicals to clean conveyors and tanks (questions 2.8,
3.8, 2.13, and3.13). Three respondents with OSP, one with immersion tin, and one with
the hot air solder leveling (HASL) technology use chemicals to clean their conveyor
systems.
Table 3-2 shows the number of times that chemical flushing was reported by respondents
as the method for tank cleaning for each process bath. The electroless nickel bath in the
nickel/gold process, and both the activator and electroless nickel baths in the
nickel/palladium/gold process are the only process baths that were consistently reported
to require chemical cleaning. The use of chemicals to clean other process baths was
reported infrequently and appeared to be based upon the operating practices of the
particular facility, rather than on any cleaning requirement specific to the technology.
3-7
-------
Table 3-2. Reported Use of Chemical Flushing as a Tank Cleaning Method
Process Type
HASL
Nickel/Gold
Nickel/Palladium/Gold
OSP
Immersion Silver
Immersion Tin
Bath Type
Microetch
Flux
Solder
Pressure Rinse
Acid Dip
Electroless Nickel
Immersion Gold
Microetch
Other Bath
Microetch
Acid Dip
Activator
Electroless Nickel
Electroless Palladium
Immersion Gold
OSP
Predip
Immersion Silver
Immersion Tin
Number of Respondents Using
Chemical Flushing"
1(27)
2(27)
5(28)
1(22)
1(8)
8(8)
1(8)
1(8)
5(9)
1(1)
1(1)
2(2)
2(2)
1(2)
1(1)
4(9)
1(2)
2(2)
1(4)
Total number of questionnaire responses for process bath are shown in parentheses.
Outputs
Possible outputs from a surface finishing process line include finished PWBs, air
emissions, wastewater discharges, and solid wastes.
Product Outputs. Product outputs include the following:
P! Chemicals incorporated onto PWBs during the surface finishing process. This includes
the PWBs along with lead, tin, silver, palladium, nickel, gold, and/or organic compounds
that are coated onto the PWB surface. This output is not quantified.
Air Releases. Chemical emission rates and air concentrations are estimated by air
modeling performed in Section 3.2, Exposure Assessment. The sources of air releases and
factors affecting emission rates are summarized below.
The total outputs to air are given by the equation:
Atotai = Aj + A2
3-8
-------
where,
Aj = evaporation and aerosol generation from baths
A2 = evaporation from drying/ovens
These terms are discussed below.
A! Evaporation and aerosol generation from baths. Potential air releases from the process
include volatilization from open surfaces of the baths as well as volatilization and aerosols
generated from air sparging, which is used in some baths for mixing. These releases to
both the occupational and outside environments are quantified in Section 3.2, Exposure
Assessment. Gases formed by chemical reactions, side reactions, and by chemical plating
in baths also contribute to air releases. However, they are expected to be small compared
to volatilization and aerosol losses, and are not quantified.
Air releases may be affected by open bath surface area, bath temperature, bath mixing
methods, and vapor control methods employed. Questionnaire data for bath agitation
and vapor control methods are summarized below:1
• Most facilities using conveyorized processes use fluid circulation pumps to mix the baths.
Panel agitation is also used as a mixing method by several facilities, while air sparging was
seldom reported (more than one method can be used simultaneously).
• The majority of vapor control methods reported are fully-enclosed and vented to the
outside. Only a few of the conveyorized processes use a push-pull2 system for vapor
control.
• For facilities using non-conveyorized processes, most use either panel agitation or
circulation pumps to mix the tanks. Only about ten percent of the facilities use air
sparging as a tank mixing method, which could generate aerosols and enhance
volatilization from the baths.
• Frequently-used vapor control methods for non-conveyorized process baths include vent-
to-outside (approximately 60 percent) and bath covers (20 percent), while seldom-
reported methods include push-pull systems or fully enclosed baths.
Table 3-3 lists average bath surface area, volume, and bath temperature data from the
PWB Workplace Practices Questionnaire. Some of this information (both surface area
and temperature) is used to model air releases in the Exposure Assessment. Surface areas
are calculated from reported bath length and width data. Larger bath surface areas
enhance evaporation. Most of the baths are maintained at elevated temperatures, which
also enhance evaporation.
1 From Questionnaire, questions 2.10 and 3.10.
2 Push-pull ventilation combines a lateral slot hood at one end of the tank with a jet of push air from the opposite
end. It is used primarily for large surface area tanks where capture velocities are insufficient to properly exhaust
fumes from the tank.
3-9
-------
Table 3-3. Average Bath Dimensions and Temj
Bath
No. of
Responses
Length
(in.)
Width
(in.)
jeratures for All Processes a
Surface Area b
(sq. in.)
Volume
(gal.)
Temp.
(°F)
HASL, Non-conveyorized
Cleaner
Microetch
Dry
Flux
Preheat
Solder
Air Knife
Pressure Rinse
3
5
1
7
1
6
1
6
28
28
-
33
-
34
-
63
20
27
-
22
-
23
-
32
540
720
-
760
-
870
-
1900
33
57
-
5
-
10
-
41
74
105
135
76
244
515
123
91
HASL, Conveyorized
Cleaner
Microetch
Dry
Flux
Preheat
HASL
Air Knife
Pressure Rinse
6
16
1
15
1
15
2
15
24
50
37
29
38
35
38
67
24
32
9
25
37
25
37
34
580
1700
330
810
1400
990
1400
2255
40
92
-
15
-
18
-
104
70
90
140
80
180
523
231
97
Nickel/Gold, Non-conveyorized
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Immersion Gold
6
7
6
7
7
7
25
26
23
26
27
26
17
17
17
17
19
17
310
370
300
360
430
370
44
43
33
42
52
43
118
93
165
75
185
181
Nickel/Palladium/Gold, Non-conveyorized
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Electroless Palladium
Immersion Gold
2
2
2
2
2
1
2
29
25
33
21
24
35
21
20
21
10
14
14
10
14
540
440
330
250
270
350
250
26
55
50
34
36
43
32
119
97
134
-
181
125
183
OSP, Non-conveyorized
Cleaner
Microetch
4
5
27
25
24
25
580
570
83
82
121
83
3-10
-------
Bath
OSP
No. of
Responses
4
Length
(in.)
27
Width
(in.)
24
Surface Area b
(sq. in.)
580
Volume
(gal.)
86
Temp.
(°F)
124
OSP, Conveyorized
Cleaner
Microetch
OSP
3
5
5
36
35
72
30
34
34
1100
1300
2600
56
63
125
113
99
108
Immersion Silver, Conveyorized
Cleaner
Microetch
Predip
Immersion Silver
Dry
2
2
2
2
1
34
42
47
143
-
31
31
31
31
-
1000
1300
1600
4400
-
65
80
60
142
-
81
73
86
113
149
Immersion Tin, Non-conveyorized
Cleaner
Microetch
Predip
Immersion Tin
2
2
1
2
27
27
30
27
18
18
24
18
500
500
720
500
49
49
60
47
104
103
-
150
Immersion Tin, Conveyorized
Cleaner
Microetch
Predip
Immersion Tin
Dry
2
2
2
3
2
39
39
31
47
-
31
31
14
31
-
1500
1500
450
1400
-
100
100
33
140
-
105
95
101
133
165
Based on PWB Workplace Practices Questionnaire data.
b All of the surface areas present in the table are average values of individual bath areas; they are not obtained by
multiplying the average length by the average width.
- No responses were given to this question in the questionnaire.
A2 Evaporation from drying/ovens. Air losses due to evaporation from drying steps apply
to HASL, OSP, immersion tin, and immersion silver processes with air knife, oven, or air
cool steps. Releases for each process type are discussed qualitatively in Section 3.1.3.
Water Releases. Potential outputs to water include chemical-contaminated wastewater
from rinse tanks, equipment cleaning, spent bath solutions, and liquid discharges from bath
sampling and bail-out. Wastewater streams from the surface finishing process line are typically
pre-treated by an on-site treatment system prior to being discharged from the facility. Spent bath
chemicals that are considered hazardous, or are too difficult to treat on-site, are drummed and
sent off-site for treatment. Waste streams with similar treatment requirements (e.g., chelated
waste streams) may be segregated from the other wastes and batch treated together. All
remaining liquid wastes are combined with similar wastes from other PWB manufacturing
3-11
-------
processes prior to treatment. The co-mingled wastewater streams are then treated to meet the
discharge limits for the facility. Once treated, the wastewater is discharged to a POTW or directly
to a receiving stream. Facilities that directly discharge to a stream require a National Pollution
Discharge Elimination System (NPDES) permit. Out of the 47 total survey respondents, 36
facilities indirectly discharge to POTWs while 10 facilities directly discharge to receiving streams.
A detailed description of on-site treatment systems is presented in Section 6.2, Recycle,
Recovery, and Control Technologies Assessment.
The total outputs to water are given by the equation:
Wtotal = Wi + W2+W3
where,
Wi = wastewater
W2 = spent bath solution
W3 = bath sampling and bail-out
These terms are discussed below.
Wi Wastewater. Chemical-contaminated rinse water is the largest source of wastewater from
the surface finishing process line, resulting primarily from drag-out. The term drag-out
refers to the process chemicals that are 'dragged' from chemical baths into the following
water rinse stages, where they are washed from the board, resulting in contamination of
the rinse water. Drag-out losses account for approximately 95 percent of uncontrolled
bath losses [i.e., losses other than from bath replacement, bail-out, and sampling (Bayes,
1996)]. Because the volume of water consumed by the rinse steps greatly exceeds the
water consumed by all other water uses, the quantity of wastewater generated by the
process is assumed to be equal to the overall water usage (I3). Daily water usage data
were collected in the PWB Workplace Practices Questionnaire (questions 2.6 and 3.6),
with the resulting data of variable to poor quality. The previous discussion of water usage
data also applies to wastewater amounts.
In the absence of quality data from industry, a model was developed to estimate the mass
loading of constituents within the wastewater, resulting from drag-out, during the
production of 260,000 ssf of PWB by the surface finishing process. The mass of chemical
transferred per day to the wastewater, as well as other model results, are presented in
Appendix E. A detailed description of the model along with the methods of model
development, validation and testing, and model limitations are presented in Prediction of
Water Quality from Printed Wiring Board Processes (Robinson et al., 1999), part of
which has been included in Appendix E. Operational practices, such as increased
drainage time, that can be used to reduce chemical losses, are described in Section 6.1,
Pollution Prevention.
W2 Spent bath solution. The concentration of chemicals within the process baths will vary,
both as PWBs are processed through them, and as the baths age (e.g., volatilization,
3-12
-------
evaporation, side reactions, etc.). These chemical baths are considered 'spent' once they
have become too contaminated or depleted to properly perform, and are replaced with a
new bath. During replacement, the spent bath chemistry is removed and the tank is
cleaned, sometimes with cleaning chemicals, before a new bath is created. Depending on
the chemicals involved, the spent bath chemistry will either undergo treatment on-site, or
may be drummed and shipped off-site for treatment when hazardous. Waste equipment
cleaning chemicals are also included in this waste stream.
Though requested, the data provided by industry respondents to the survey regarding the
annual volume of bath chemistry disposed for each bath type (questions 2.13, 2.15, 3.13,
and 3.15) was found to be of variable to poor quality. Instead, the annual volume of
chemical solution disposed per bath type was calculated by determining the number of
times a bath would require changing to produce a specific surface area of PWB, as
described in Section 4.2, Cost Analysis. For the purposes of this assessment, chemical
concentrations within the spent baths were assumed to be the same as concentrations at
the time of bath make-up.
The methods of on-site treatment or disposal for individual spent baths were identified by
questionnaire respondents. A summary of the spent bath treatment and disposal
responses by technology type is presented in Table 3-4.
Bath sampling and bail-out. This includes bath samples disposed of after analysis and
bath solution discarded through bail-out (sometimes done prior to bath additions). In
some cases sampling may be performed at the same time as bail-out if the process bath is
controlled by an automated monitoring system.
Routine bail-out activities, the practice of removing bath solution to make room for more
concentrated chemical additions, could result in large volumes of bath disposal. Bail-out
and bath addition data (e.g., frequency, duration and quantity) were collected in the PWB
Workplace Practices Questionnaire, with the resulting data being of poor quality.
Chemical loss due to bath sampling was assumed to be negligible.
Table 3-4. Spent Bath Treatment and Disposal Methods
Process
Alternative
HASL
Nickel/Gold
Nickel/Palladium/
Gold
OSP
Immersion Silver
Immersion Tin
Total No.
of Baths
113
55
14
28
8
17
Precipitation
Pretreatment a
29
35
8
14
o
6
o
6
pH
Neutralization a
24
25
o
J
15
o
J
6
Disposed
to Sewer a
1
0
0
0
1
0
Drummed a
11
2
7
4
2
5
Recycled
On-Site a
6
2
1
1
0
o
J
Recycle
Off-Site a
29
4
1
0
0
0
Others
8
5
0
0
0
0
a Number of affirmative responses for any bath from the PWB Workplace Practices Questionnaire, for all facilities
using a technology category.
3-13
-------
Wastewater Treatment. Figure 3-2 depicts the overall water and wastewater treatment
flows, including wastewater, bath chemicals, and spent bath solution inputs to treatment,
treatment performed on-site or off-site, sludge generated from either on-site or off-site treatment,
and final effluent discharge to a POTW or receiving streams. PWB manufacturers typically
combine wastewater effluent from other PWB manufacturing processes prior to on-site
wastewater treatment. Sludge from on-site wastewater treatment is typically sent off-site for
recycling or disposal. Detailed treatment system diagrams for each surface finishing technology
are presented and discussed in Section 6.2, Recycle, Recovery, and Control Technologies
Assessment.
EI Wastewater effluent from treatment. The mass-loading of chemical constituents within
the wastewater effluent is dependent on several factors including the type and mass-
loading of chemical inputs to the treatment process, the treatment technology employed,
the duration of treatment of the wastewater, and the discharge limit, if applicable.
Facilities that discharge to a POTW must treat their wastewater to meet the permit levels
set by the receiving POTW for targeted contaminants such as metals and biochemical
oxygen demand (BOD). Facilities that discharge wastewater directly to a receiving stream
must obtain aNPDES permit, which establishes limits for similar chemical contaminants.
No data were collected for this waste stream due to dependance on factors outside of the
surface finishing technology. However, organic chemical constituents resulting from the mass-
loading into the treatment process are calculated and organic releases to the receiving stream are
modeled in Section 3.2, Exposure Assessment.
Solid Waste. Solid wastes are generated by day-to-day surface finishing line operation
and by wastewater treatment of process effluent. Some of these solid wastes are recycled, while
others are sent to incineration or land disposal. The total solid waste outputs are given by the
equation:
O — O_|_O_|_O_|_O
Ototal 01 ^ 02 ^ 03 ^ 04
where,
Si = hazardous solid waste
S2 = non-hazardous solid waste
S3 = drummed solid or liquid waste
S4 = sludge from on-site wastewater treatment
These terms are discussed below.
Si Hazardous solid waste. Hazardous solid waste could include spent bath filters, solder
dross, packaging or chemical container residues, and other solid waste from the process
line which is contaminated with any hazardous material, as defined by the Resource
Conservation and Recovery Act (RCRA). For example, lead, which is a component of
the solder used in the HASL technology, is considered a hazardous solid waste (the
3-14
-------
RCRA waste code D008 is for lead).3 Container residue is estimated by EPA to be up to
four percent of the chemicals use volume (Froiman, 1996). An industry reviewer
indicated this estimate would only occur with very poor housekeeping practices and is not
representative of the PWB industry. RCRA waste codes which are applicable to the
surface finishing technologies are discussed in Section 4.3, Regulatory Status. Hazardous
solid waste is typically sent off-site to a hazardous waste landfill for disposal or is
incinerated.
S2 Non-hazardous solid waste. Non-hazardous solid wastes could include any spent bath
filters, packaging or chemical container residues, and other solid waste from the process
line that does not contain any RCRA-defined hazardous materials listed in CFR Section
261. These wastes may be recycled or sent to off-site disposal in a landfill.
S3 Drummed solid or liquid waste. This includes other liquid or solid wastes that are
drummed for off-site recycling or disposal. This includes spent bath chemicals which
cannot be treated on-site because they are considered hazardous or require treatment
beyond what can be provided by the facility. Hazardous chemical wastes are sent to a
hazardous waste treatment facility. Table 3-5 is a summary of responses indicating the
presence of a RCRA listed waste and the type of container in which it was stored.
Other chemical wastes are drummed and sent out for recycling to reclaim the metal
content from the solution (e.g., gold, silver, nickel, etc.). The number of responses which
indicated that a bath was drummed for disposal was shown in Table 3-4.
S4 Sludge from on-site wastewater treatment. Facilities were asked to report the amount of
sludge generated during on-site wastewater treatment that could be attributed to surface
finishing line effluents (question 1.3). Many PWB manufacturers have indicated that the
amount of sludge resulting from the surface finishing process cannot be reliably estimated
since effluents from various PWB manufacturing process steps are combined prior to
wastewater treatment. Other factors that also influence the amount of sludge generated
during wastewater treatment include the size of the facilities, the surface finishing
technology used, the treatment method used, facility operating procedures, the efficiency
with which bath chemicals and rinse water are used, and so on. Thus, the actual and
comparative amount of sludge generated due to the choice of surface finishing technology
could not be determined, nor were data available to characterize the concentrations of
metals contributed by the surface finishing line.
However, many respondents did report the annual amount of sludge generated from their
on-site waste treatment facility. The average sludge generated annually by the
respondents to the PWB Workplace Practices Questionnaire is 214,900 pounds. The
average water content of the sludge, which is typically pressed prior to disposal, ranges
from 60 to 70 percent (Sharp, 1999).
3 It is important to note that solder dross and solder pot dumps are excluded from the RCRA definition of solid
waste when they are recycled. Therefore, when they are recycled they are not considered a hazardous solid waste.
3-15
-------
Table 3-5. RCRA Wastes and Container Types for Surface Finishing Technologies
Process
Alternatives
HASL
Nickel/Gold
OSP
Immersion Tin
Bath Name
Cleaner
Microetch
Flux
Solder
Pressure Rinse
Cleaner
Microetch
Catalyst
Acid dip
Electroless. Nickel
Immersion Gold
Cleaner
Microetch
OSP
Cleaner
Immersion Tin
No. of
Baths
12
25
26
26
21
7
8
5
18
8
8
7
8
7
5
4
No. of
RCRA Wastes
1
8
7
7
2
1
2
1
3
0
3
2
1
0
0
0
Open Head
Drum
0
0
0
8
1
0
0
0
0
0
0
0
0
0
0
0
Close Head
Drum
2
9
12
6
1
2
3
2
6
3
4
1
1
1
1
1
Others
0
4
0
5
3
0
0
0
0
0
0
2
2
1
0
0
Transformations
Transformations within the surface finishing system boundary could include:
R! Chemical reaction gains or losses. This includes any chemical species consumed,
transformed, or produced in chemical reactions and side reactions occurring in the
process baths. Reactions and side reactions within the baths could result in either
chemical losses or production of new chemicals as degradation products. Although there
are almost certainly side reactions which occur, little research has been conducted to
identify them when they do not obstruct the desired reactions. This is not quantified.
Material Balance
A material balance approach is often used to describe and analyze a process. The
approach is based on the principle that the mass of the material inputs must equal the mass of the
material outputs if the process is at steady-state (i.e., there is no accumulation of material within
the process). Although the PWB Workplace Practices Questionnaire did not collect enough data
to quantify every stream, the approach is a useful way to identify and organize input and output
streams that cross the boundary of the system (the process in this case).
The general mass balance equation for a specific chemical is:
Input - Output + Production - Consumption = Accumulation
3-16
-------
Since there were no chemical transformations identified, the production and consumption terms
are dropped from the equation. When the system is considered to be running at steady-state, the
accumulation term is equal to zero and the mass balance equation becomes:
Inputs = Outputs
The material balance for Figure 3-1 (surface finishing process line prior to wastewater treatment)
includes the inputs Ib I2,13, and I4, and the outputs Pb Ab A2, Wb W2, W3, Sb S2, and S3.
Since the inputs must equal the outputs, the material balance for Figure 3-1 is:
T! + I2 + I3 + I4 = P! + A! + A2 + Wi + W2 + W3 + Si + S2 +S3
or:
I,o,al = Pi + Atotal + Wtotal + Si + S2 + S3
The material balance for Figure 3-2 (wastewater treatment) includes the inputs Wb W2, and W3i
and the outputs Ej and S4.
Thus, the material balance equation for Figure 3-2, wastewater treatment, is:
Wi + W2 + W3 = E! + S4
or:
Wtotai = E! + S4
These equations are presented to indicate that all the material flows have been accounted
for.
3.1.3 Source and Release Information for Specific Surface Finishing Technologies
This section applies the material balance approach described previously to the individual
surface finishing technologies. Each input and output is discussed as it applies to that surface
finishing technology, and quantified when possible. The numbers reported in this section
represent the actual responses to the PWB Workplace Practices Questionnaire, and thus, may
reflect wide variations in the data corresponding to the different operating profiles of the
respondents. To facilitate comparison among process alternatives and to adjust for wide
variations in the data due to differences in facility size and production levels, data are presented
both as reported in the PWB Workplace Practices Questionnaire, and normalized by production
amounts (annual ssf of PWB produced). Values reported in this summary are average values
calculated from questionnaire responses.
3-17
-------
The limited number of responses to the questionnaire for some technologies along with
differences in production levels and operating practices between facilities make it difficult to
make a comparison of technologies. To facilitate a comparative evaluation, the individual
technologies were modeled using a consistent production throughput in ssf of PWB produced.
The modeling of the surface finishing technologies is presented in Section 4.2, Cost Analysis.
Hot Air Solder Leveling
Figure 3-3 illustrates the generic HASL process steps and typical bath sequence evaluated
in the CTSA. The number and location of rinse steps shown in the figure are based on the PWB
Workplace Practices Questionnaire data. Thus, Figure 3-3 describes the types and sequence of
baths in a generic HASL line, but the types and sequence of baths in an actual line could vary. A
detailed description of HASL process stages is presented in Section 2.1.3, Chemistry and Process
Descriptions of Surface Finishing Technologies.
Cleaner
Hr
Microetch
y
3 • Water Rinse x 2
y
4 Dry
y
5 Flux
i
6 Preheat
y
7 HASL
y
8- Air Knife
i
Pressure Rinse
V
10 Water Rinse xl
I
|
|
I
I
I
I
|
I
|
Figure 3-3. Generic HASL Process Steps and Typical Bath Sequence
3-18
-------
Water Usage (I3) and Wastewater (Wt). Water usage data from the PWB Workplace
Practices Questionnaire is presented in Table 3-1; the volume of wastewater generated was
assumed to be equal to the amount of water used (I3). Of respondents using a HASL process, 21
facilities use the conveyorized process, while 9 facilities use the non-conveyorized process. In
summary:
• Reported water usage for the facilities using the conveyorized HASL process average 1.2
million gallons per year, or about 4.9 gallons per ssf of PWB produced.
• Reported water usage for the facilities using the non-conveyorized HASL process average
250 thousand gallons per year, or 0.97 gallons per ssf of PWB produced.
Chemical constituents and concentrations in wastewater could not be adequately
characterized from questionnaire data. In the absence of quality data from industry, a model was
developed to estimate the mass loading of constituents within the wastewater, resulting from
drag-out, during the production of 260,000 ssf of PWB by the surface finishing process. The
term drag-out refers to the process chemicals that are 'dragged' from chemical baths into the
following water rinse stages, where they are washed from the board, resulting in contamination of
the rinse water. The mass of chemical transferred per day to the wastewater, as well as other
model results, are presented in Appendix E.
Bath Chemicals Used (I^. Bath concentrations of individual chemical constituents are
presented in Appendix B. The volume of chemicals consumed per year was determined by
modeling the time it would take the generic HASL process described in Figure 3-3 to produce a
specific PWB throughput. A detailed description of the process modeling is presented in Section
4.2, Cost Analysis. The number of bath replacements (calculated from the modeled time) was
then multiplied by the volume of the bath to determine the volume of a bath chemical consumed
per year. The mass of solder consumed per year was calculated by using an estimate of the
amount of solder applied per ssf of PWB produced, then adjusted to account for solder waste.
When waste solder is not routinely recycled, as much as 2,500 Ibs of solder is consumed when
producing 260,000 ssf of PWB. Solder consumption is discussed further in Section 5.1, Resource
Conservation. Bath chemical consumption is presented Appendix G.
Cleaning Chemicals (I4). Nine out of 129 HASL baths were reported to be cleaned using
chemicals, however, data concerning the type of cleaning chemical(s) were not collected by the
questionnaire. The majority of chemical flushing reported for the HASL processes was used for
solder tank cleaning during bath replacement. Water is most frequently used to clean tanks prior
to new bath make-up.
Spent Bath Solutions (W2). The quantity of spent bath solution could not be determined
directly from the questionnaire data. However, the volume of spent bath chemistry was
calculated by determining the number of bath changes required per year and multiplying by the
average volume of the process tank (see Section 4.2, Cost Analysis). The concentrations of
chemical constituents within the spent bath solutions were assumed to be the same as make-up
bath concentrations.
3-19
-------
Spent bath treatment and disposal methods were presented in Table 3-4. Off-site
recycling, precipitation pretreatment, and pH neutralization are reported as common treatment
methods for the conveyorized HASL processes. Respondents for both the non-conveyorized,
vertical process and the mixed HASL processes reported that precipitation pretreatment, pH
neutralization, and off-site recycling are common treatment methods.
Evaporation From Baths (At). Air releases are modeled in Section 3.2, Exposure
Assessment. A summary of data collected from the questionnaire is presented below:
• For the conveyorized HASL processes, circulation pumps are used to mix all process
baths except for the cleaner bath. Full enclosure and venting are the most common
methods of vapor control reported by respondents for all baths and process steps.
For non-conveyorized HASL facilities, both panel agitation and circulation pumps are the
most reported mixing methods for all baths. Venting to the outside is the prevalent form
of vapor control reported, though 25 percent of the baths were reported to use bath
covers.
Table 3-3 lists the bath surface area, volume, and bath temperature data reported by
respondents to the PWB Workplace Practices Questionnaire.
Evaporation From Drying/Ovens (A2). Air knife and oven drying occur after the
microetch and HASL baths. Any solution adhering to the PWBs would be either blown off the
boards and returned to the sump, or volatilized in the oven. Air emissions from air knife or oven
drying were not quantified.
Chemicals Incorporated Onto PWBs (Pi). A coating of tin/lead solder is applied to the
surface of PWB panels in the HASL process. The amount of solder added to the panels depends
on the exposed surface area of the PWB panels being processed. The amount of solder
incorporated onto a PWB was calculated at 0.0369 oz/ssf. Solder consumption is discussed
further in Section 5.1, Resource Conservation.
Drummed Solid or Liquid Waste (S3). Questionnaire respondents indicated that
approximately 25 percent of HASL baths contain hazardous waste constituents as defined by
RCRA. These wastes were associated by respondents with the microetch, flux, and solder baths.
RCRA wastes are discussed in further detail in Section 4.3, Regulatory Status. In response to a
separate question regarding spent bath treatment (see Table 3-4), 11 out of 113 HASL baths were
reported by respondents to be drummed and sent off-site for recycling or disposal.
3-20
-------
Nickel/Gold Process
Figure 3-4 depicts the generic nickel/gold process steps and typical bath sequence
evaluated in the CIS A. The process baths shown in the figure represent an amalgamation of the
various products offered within the nickel/gold technology category. The number and location of
rinse steps displayed in the figure are based on PWB Workplace Practices Questionnaire
responses. Thus, Figure 3-4 describes the types and sequence of baths in a generic nickel/gold
line, but the types and sequence of process baths used by any particular facility could vary. A
detailed description of the nickel/gold process is presented in Section 2.1.3, Chemistry and
Process Descriptions of Surface Finishing Technologies.
1. Cleaner
V
2. Water Rinse x 1
V
3. Microetch
y
4. Water Rinse x 1
V
5. Catalyst
V
6. Water Rinse x 1
V
7- Acid Dip
V
"• Water Rinse x 1
V
9. Electroless Nickel
y
10. Water Rinse x 2
V
1 1 . Immersion Gold
*
12- Water Rinse x 2
1
1
1
1
1
1
1
1
1
1
1
1
Figure 3-4. Generic Nickel/Gold Process Steps and Typical Bath Sequence
3-21
-------
Water Usage (I3) and Wastewater (Wt). Water usage data from the PWB Workplace
Practices Questionnaire is presented in Table 3-1; the volume of wastewater generated was
assumed to be equal to the amount of water used (I3). All eight respondents report using the non-
conveyorized nickel/gold process. In summary:
• Reported water usage for the facilities using the non-conveyorized nickel/gold process
average 540 thousand gallons per year, or 100 gallons per ssf of PWB produced.
Chemical constituents and concentrations in wastewater could not be adequately
characterized from questionnaire data. In the absence of quality data from industry, a model was
developed to estimate the mass loading of constituents within the wastewater, resulting from
drag-out, during the production of 260,000 ssf of PWB by the surface finishing process. The
mass of chemical transferred per day to the wastewater, as well as other model results, are
presented in Appendix E.
Bath Chemicals Used (I^. Bath concentrations of individual chemical constituents are
presented in Appendix B. The volume of chemicals consumed per year was determined by
modeling the time it would take the generic nickel/gold process described in Figure 3-4 to
produce a specific PWB throughput. A detailed description of the process modeling is presented
in Section 4.2, Cost Analysis. The number of bath replacements (calculated from the modeled
time) was then multiplied by the volume of the bath to determine the volume of a bath chemical
consumed per year. Nickel/gold process chemical consumption is presented in Appendix G.
Cleaning Chemicals (I4). Twelve out of 47 reported nickel/gold baths require chemicals
to clean the tanks, however, data concerning the type of cleaning chemical(s) were not collected
by the questionnaire. Seven of the tanks that were reported to require chemical flushing belong
to electroless nickel baths. The remaining tanks requiring chemical flushing belong to baths
which are not part of the generic process sequence described in Figure 3-4. Water is most
frequently used to clean tanks prior to new bath make-up.
Spent Bath Solutions (W2). The quantity of spent bath solution could not be determined
directly from the questionnaire data. However, the volume of spent bath chemistry was
calculated by determining the number of bath changes required per year and multiplying by the
average volume of the process tank (see Section 4.2, Cost Analysis). The concentrations of
chemical constituents within the spent bath solutions were assumed to be the same as make-up
bath concentrations.
Spent bath treatment and disposal methods were presented in Table 3-4. Respondents for
the non-conveyorized, vertical process reported that pH neutralization and precipitation
pretreatment are common treatment methods. Off-site recycling was also reported as a treatment
option.
Evaporation From Baths (At). Air releases are modeled in Section 3.2, Exposure
Assessment. A summary of data collected from the questionnaire is presented below:
3-22
-------
• For non-conveyorized nickel/gold processes, panel agitation and circulation pumps are
the most reported mixing methods for all baths. Venting to the outside is the most
prevalent form of vapor control reported (33 percent), though the use of bath covers and
push-pull systems are also reported.
Table 3-3 lists the bath surface area, volume, and bath temperature data reported by
respondents to the PWB Workplace Practices Questionnaire.
Evaporation From Drying/Ovens (A2). The nickel/gold process does not require the use
of a drying oven or air knife.
Chemicals Incorporated Onto PWBs (Pi). The nickel/gold process promotes the
deposition of an initial, thick layer of nickel followed by a thin, protective layer of gold onto the
exposed metal surfaces of the PWB. The amount of nickel incorporated onto a PWB was
calculated at 0.0337 oz/ssf, while gold was deposited at the rate of 0.0028 oz/ssf. Both nickel and
gold deposition rates are discussed further in Section 5.1, Resource Conservation.
Drummed Solid or Liquid Waste (S3). Questionnaire respondents indicated that
approximately 20 percent of nickel/gold baths contain hazardous waste constituents as defined
by RCRA. These wastes were associated by respondents with the microetch, acid dip, catalyst,
and immersion gold baths. RCRA wastes are discussed in further detail in Section 4.3,
Regulatory Status. In response to a separate question regarding spent bath treatment (see Table
3-4), two out of 55 nickel/gold baths (3.6 percent) were reported by respondents to be drummed
and sent off-site for recycling. Section 5.1, Resource Conservation, presents methods commonly
used to recover gold on-site.
Nickel/Palladium/Gold Process
Figure 3-5 depicts the generic nickel/palladium/gold process steps and typical bath
sequence evaluated in the CTSA. The number and location of rinse steps displayed in the figure
are based on PWB Workplace Practices Questionnaire responses. Thus, Figure 3-5 describes the
types and sequence of baths in a generic nickel/palladium/gold line, but the types and sequence
of process baths used by any particular facility could vary. A detailed description of the
nickel/palladium/gold process is presented in Section 2.1.3, Chemistry and Process Descriptions
of Surface Finishing Technologies.
3-23
-------
3.
7.
S.
10.
11.
12.
Cleaner
Water Rinse x 2
Microetch
Water Rinse x 2
Catalyst
Water Rinse x 2
I
Acid Dip
Water Rinse x 2
Electroless Nickel
Water Rinse x 2
Preinitiator
Electroless Palladium
13.
Water Rinse x 2
14.
Immersion Gold
IS.
Water Rinse x 2
Figure 3-5. Generic Nickel/Palladium/Gold Process Steps and Typical Bath Sequence
3-24
-------
Water Usage (I3) and Wastewater (Wt). Water usage data from the PWB Workplace
Practices Questionnaire is presented in Table 3-1; the volume of wastewater generated was
assumed to be equal to the amount of water used (I3). Of the two facilities using the
nickel/palladium/gold process included in this study, both report using the non-conveyorized
process configuration. In summary:
• Reported water usage for the facilities using the non-conveyorized nickel/palladium/gold
process average 960 thousand gallons per year, or 160 gallons per ssf of PWB produced.
Chemical constituents and concentrations in wastewater could not be adequately
characterized from questionnaire data. In the absence of quality data from industry, a model was
developed to estimate the mass loading of constituents within the wastewater, resulting from
drag-out, during the production of 260,000 ssf of PWB by the surface finishing process. The
mass of chemical transferred per day to the wastewater, as well as other model results, are
presented in Appendix E.
Bath Chemicals Used (I^. Bath concentrations of individual chemical constituents are
presented in Appendix B. The volume of chemicals consumed per year was determined by
modeling the time it would take the generic nickel/palladium/gold process described in Figure 3-5
to produce a specific PWB throughput. A detailed description of the process modeling is
presented in Section 4.2, Cost Analysis. The number of bath replacements (calculated from the
modeled time) was then multiplied by the volume of the bath to determine the volume of a bath
chemical consumed per year. Nickel/palladium/gold process chemical consumption is presented
in Appendix G.
Cleaning Chemicals (I4). Eight out of 14 reported nickel/palladium/gold baths require
chemicals to clean the tanks, however, data concerning the type of cleaning chemical(s) were not
collected by the questionnaire. Chemical flushing was reported at least once for the microetch,
acid dip, electroless nickel, electroless palladium, and immersion gold tanks. The remaining tanks
requiring chemical flushing belong to baths which are not part of the generic process sequence
described in Figure 3-5. Water is most frequently used to clean tanks prior to new bath make-up.
Hand scrubbing was also required for tank cleaning by several of the respondents.
Spent Bath Solutions (W2). The quantity of spent bath solution could not be determined
directly from the questionnaire data. However, the volume of spent bath chemistry was
calculated by determining the number of bath changes required per year and multiplying by the
average volume of the process tank (see Section 4.2, Cost Analysis). The concentrations of
chemical constituents within the spent bath solutions were assumed to be the same as make-up
bath concentrations.
Spent bath treatment and disposal methods were presented in Table 3-4. Respondents for
the non-conveyorized, vertical process reported that precipitation pretreatment was the prevalent
treatment method for spent bath solutions. Drummed for off-site treatment and pH
neutralization were also reported.
3-25
-------
Evaporation From Baths (At). Air releases are modeled in Section 3.2, Exposure
Assessment. A summary of data collected from the questionnaire is presented below:
For non-conveyorized nickel/palladium/gold processes, panel agitation and circulation
pumps are the most reported mixing methods for all baths, while the use of air sparging
for the electroless nickel bath was also reported. Vapor control methods were only
identified for two process baths by survey respondents. Both baths were reported to use
bath covers.
Table 3-3 lists the bath surface area, volume, and bath temperature data reported by
respondents to the PWB Workplace Practices Questionnaire.
Evaporation From Drying/Ovens (A2). The nickel/palladium/gold process does not
require the use of a drying oven or air knife.
Chemicals Incorporated Onto PWBs (Pi). Layers of nickel, palladium, and gold are
deposited onto the exposed metal surfaces of the PWBs through a series of chemical plating
reactions. The amount of nickel incorporated onto a PWB was calculated at 0.0337 oz/ssf,
palladium at 0.0015 oz/ssf, and gold at a rate of 0.0028 oz/ssf. The deposition rates of all three
metals are discussed further in Section 5.1, Resource Conservation.
Drummed Solid or Liquid Waste (S3). Questionnaire respondents indicated that none
of the nickel/palladium/gold baths contain hazardous waste constituents as defined by RCRA. A
detailed discussion of RCRA wastes can be found in Section 4.3, Regulatory Status. In response
to a separate question regarding spent bath treatment (see Table 3-4), seven out of 14
nickel/palladium/gold baths (50 percent) were reported by respondents to be drummed and sent
off-site for recycling or disposal. Section 5.1, Resource Conservation, presents methods
commonly used to recover gold on-site.
Organic Solderability Preservative
Figure 3-6 depicts the generic OSP process steps and typical bath sequence evaluated in
the CTSA. The process baths shown in Figure 3-6 represent an amalgamation of the various
products offered within the OSP technology category. The number and location of rinse steps
displayed in the figure are based on PWB Workplace Practices Questionnaire responses. Thus,
Figure 3-6 describes the types and sequence of baths in a generic OSP line, but the types and
sequence of OSP process baths used by any particular facility could vary. A detailed description
of the OSP process is presented in Section 2.1.3, Chemistry and Process Descriptions of Surface
Finishing Technologies.
3-26
-------
1 • Cleaner
V
2 Water Rinse x 1
V
3 • Microetch
V
4 . Water Rinse x 1
V
5 Air Knife
y
6. OSP
V
I
I
I
I
__j
I
7 Air Knife I
y
8 . Water Rinse x 1
V
9. Dry
I
__j
Figure 3-6. Generic OSP Process Steps and Typical Bath Sequence
Water Usage (I3) and Wastewater (Wt). Water usage data from the PWB Workplace
Practices Questionnaire is presented in Table 3-1; the volume of wastewater generated was
assumed to be equal to the amount of water used (I3). Of respondents using the OSP process,
five facilities use the conveyorized OSP process while five other facilities use the non-
conveyorized OSP process. In summary:
• Reported water usage for the facilities using the conveyorized OSP process average 440
thousand gallons per year, or about 14 gallons per ssf of PWB produced.
• Reported water usage for the facilities using the non-conveyorized OSP process average
89 thousand gallons per year, or 1.9 gallons per ssf of PWB produced.
Chemical constituents and concentrations in wastewater could not be adequately
characterized from questionnaire data. In the absence of quality data from industry, a model was
developed to estimate the mass loading of constituents within the wastewater, resulting from
drag-out, during the production of 260,000 ssf of PWB by the surface finishing process. The
mass of chemical transferred per day to the wastewater, as well as other model results, are
presented in Appendix E.
3-27
-------
Bath Chemicals Used (I^. Bath concentrations of individual chemical constituents are
presented in Appendix B. The volume of chemicals consumed per year was determined by
modeling the time it would take the generic OSP process described in Figure 3-6 to produce a
specific PWB throughput. A detailed description of the process modeling is presented in Section
4.2, Cost Analysis. The number of bath replacements (calculated from the modeled time) was
then multiplied by the volume of the bath to determine the volume of a bath chemical consumed
per year. OSP process chemical consumption is presented in Appendix G.
Cleaning Chemicals (I4). Three out of 31 OSP baths were reported to be cleaned using
chemicals, however, data concerning the type of cleaning chemical(s) were not collected by the
questionnaire. All of the chemical flushing reported for OSP processes was used for cleaning the
OSP tank during bath replacement. Water is most frequently used to clean tanks prior to new
bath make-up.
Spent Bath Solutions (W2). The quantity of spent bath solution could not be determined
directly from the questionnaire data. However, the volume of spent bath chemistry was
calculated by determining the number of bath changes required per year and multiplying by the
average volume of the process tank (see Section 4.2, Cost Analysis). The concentrations of
chemical constituents within the spent bath solutions are assumed to be the same as make-up
bath concentrations.
Spent bath treatment and disposal methods were presented in Table 3-4. Precipitation
pretreatment, pH neutralization, and drummed for off-site treatment are reported as common
treatment methods for the conveyorized OSP processes. Respondents for the non-conveyorized,
vertical process reported that pH neutralization and precipitation pretreatment are common
treatment methods.
Evaporation From Baths (At). Air releases are modeled in Section 3.2, Exposure
Assessment. A summary of data collected from the questionnaire is presented below:
For the conveyorized OSP processes, circulation pumps are used to mix all process wet
chemistry baths. Full enclosure and venting are the most common methods of vapor
control reported by respondents for all baths and process steps.
For non-conveyorized OSP processes, both panel agitation and circulation pumps are the
most reported mixing methods for all baths. Venting to the outside is the most prevalent
form of vapor control reported (66 percent), though a push-pull vapor control system is
also reported (33 percent).
Table 3-3 lists the bath surface area, volume, and bath temperature data reported by
respondents to the PWB Workplace Practices Questionnaire.
Evaporation From Drying/Ovens (A2). Air knife and oven drying occur after the
microetch and OSP baths. Any solution adhering to the PWBs would be either blown off the
boards and returned to the sump, or volatilized in the oven. Air emissions from air knife or oven
drying were not modeled.
3-28
-------
Chemicals Incorporated onto PWBs (Pi). A thin coating of a protective organic
compound is applied to the surfaces of the PWB to protect the solderability of the copper
surfaces.
Drummed Solid or Liquid Waste (S3). Questionnaire respondents indicated that
approximately 15 percent of OSP baths contain hazardous waste constituents as defined by
RCRA. These wastes were primarily associated by respondents with the cleaner bath. RCRA
wastes are discussed in further detail in Section 4.3, Regulatory Status. In response to a separate
question regarding spent bath treatment (see Table 3-4), four out of 28 OSP baths were reported
to be drummed and sent off-site for recycling or disposal.
Immersion Silver Process
Figure 3-7 depicts the generic immersion silver process steps and typical bath sequence
evaluated in the CTSA. The number and location of rinse steps displayed in the figure are based
on PWB Workplace Practices Questionnaire responses. Thus, Figure 3-7 describes the types and
sequence of baths in a generic immersion silver line, but the types and sequence of immersion
silver process baths used by any particular facility could vary. A detailed description of the
immersion silver process is presented in Section 2.1.3, Chemistry and Process Descriptions of
Surface Finishing Technologies.
1- Cleaner
V
2. Water Rinse x 1
V
3. Microetch
y
4. Water Rinse x 1
y
5- Predip
V
"• Immersion Silver
Y
7. Water Rinse x 1
1
8. Dry
I
I
I
I
I
I
I
I
Figure 3-7. Generic Immersion Silver Process Steps and Typical Bath Sequence
3-29
-------
Water Usage (I3) and Wastewater (Wt). Water usage data from the PWB Workplace
Practices Questionnaire is presented in Table 3-1; the volume of wastewater generated was
assumed to be equal to the amount of water used (I3). Of the two respondents using the
immersion silver process, both reported using the conveyorized process configuration. In
summary:
• Reported water usage for the facilities using the conveyorized immersion silver process
average 910 thousand gallons per year, or about 37 gallons per ssf of PWB produced.
Chemical constituents and concentrations in wastewater could not be adequately
characterized from questionnaire data. In the absence of quality data from industry, a model was
developed to estimate the mass loading of constituents within the wastewater, resulting from
drag-out, during the production of 260,000 ssf of PWB by the surface finishing process. The
mass of chemical transferred per day to the wastewater, as well as other model results, are
presented in Appendix E.
Bath Chemicals Used (I^. Bath concentrations of individual chemical constituents are
presented in Appendix B. The volume of chemicals consumed per year was determined by
modeling the time it would take the generic immersion silver process described in Figure 3-7 to
produce a specific PWB throughput. A detailed description of the process modeling is presented
in Section 4.2, Cost Analysis. The number of bath replacements (calculated from the modeled
time) was then multiplied by the volume of the bath to determine the volume of a bath chemical
consumed per year. Immersion silver process chemical consumption is presented in Appendix
G.
Cleaning Chemicals (I4). Three out of nine immersion silver baths were reported to be
cleaned using chemicals, however, the type of cleaning chemical(s) were not collected by the
questionnaire. The immersion silver process tanks reported to require chemical flushing prior to
bath replacement included two immersion silver process tanks and one pre-dip tank. Water is
most frequently used to clean tanks prior to new bath make-up.
Spent Bath Solutions (W2). The quantity of spent bath solution could not be determined
directly from the questionnaire data. However, the volume of spent bath chemistry was
calculated by determining the number of bath changes required per year and multiplying by the
average volume of the process tank (see Section 4.2, Cost Analysis).
The concentrations of chemical constituents within the spent bath solutions are expected
to vary significantly as PWBs are processed through the bath. Some new constituents, such as
copper displaced by an immersion-type plating reaction, will be present in solution, although they
are not part of the original bath chemistry. While the concentrations of these chemical
constituents can be significant, they are difficult to accurately estimate and will vary widely. For
the purposes of this analysis, the concentrations of chemical constituents within the spent bath
solutions were assumed to be the same as make-up bath concentrations.
3-30
-------
Spent bath treatment and disposal methods were presented in Table 3-4. Precipitation
pretreatment, pH neutralization, and drummed for off-site treatment are reported as common
treatment methods for the conveyorized immersion silver processes.
Evaporation From Baths (At). Air releases are modeled in Section 3.2, Exposure
Assessment. A summary of data collected from the questionnaire is presented below:
For conveyorized immersion silver processes, circulation pumps are used to mix all
process wet chemistry baths. The spraying of chemicals onto the surface of the PWB in
the cleaner and microetch baths is also reported. All of the process baths were reported as
fully enclosed. Only one out often process baths was reported to be vented to the
outside.
Table 3-3 lists bath the surface area, volume, and bath temperature data reported by
respondents to the PWB Workplace Practices Questionnaire.
Evaporation From Drying/Ovens (A2). Oven drying occurs directly after the immersion
silver bath. Any solution adhering to the PWBs is volatilized during the drying of the PWBs by
the oven. Air emissions resulting from oven drying were not modeled. No air knife is required
by this process.
Chemicals Incorporated Onto PWBs (Pi). Silver is added to the boards in the
immersion silver processes. A hydrophobic layer, formed with a co-deposited organic inhibitor,
is also coated on top of the silver layer. The amount of silver incorporated onto a PWB was
calculated at 0.0013 oz/ssf. Silver consumption is discussed further in Section 5.1, Resource
Conservation.
Drummed Solid or Liquid Waste (S3). Questionnaire respondents indicated that none
of the immersion silver baths contain hazardous waste constituents as defined by RCRA. A
detailed discussion of RCRA wastes can be found in Section 4.3, Regulatory Status. In response
to a separate question regarding spent bath treatment (see Table 3-4), two out of eight immersion
silver baths were reported to be drummed and sent off-site for recycling.
Immersion Tin Process
Figure 3-8 depicts the generic immersion tin process steps and typical bath sequence
evaluated in the CTSA. The process baths shown in the figure represent an amalgamation of the
various products offered within the immersion tin technology category. The number and location
of rinse steps displayed in the figure are based on PWB Workplace Practices Questionnaire
responses. Thus, Figure 3-8 describes the types and sequence of baths in a generic immersion tin
line, but the types and sequence of immersion tin process baths used by any particular facility
could vary. A detailed description of the immersion tin process is presented in Section 2.1.3,
Chemistry and Process Descriptions of Surface Finishing Technologies.
3-31
-------
1- Cleaner
y
2. Water Rinse x 2
y
3- Microetch
y
4. Water Rinse x 2
V
5- Predip
Y
6. Water Rinse x 1
y
7. Immersion Tin
y
8. Water Rinse x 2
y
9. Dry
I
I
I
I
I
I
I
I
I
Figure 3-8. Generic Immersion Tin Process Steps and Typical Bath Sequence
Water Usage (I3) and Wastewater (Wt). Water usage data from the PWB Workplace
Practices Questionnaire is presented in Table 3-1; the volume of wastewater generated was
assumed to be equal to the amount of water used (I3). Of respondents using the immersion tin
process, two facilities use the conveyorized immersion tin process while four other facilities use
the non-conveyorized process. In summary:
• Reported water usage for the facilities using the conveyorized immersion tin process
average 110 thousand gallons per year, or about 0.33 gallons per ssf of PWB produced.
• Reported water usage for the facilities using the non-conveyorized immersion tin process
average 210 thousand gallons per year, or 11 gallons per ssf of PWB produced.
Chemical constituents and concentrations in wastewater could not be adequately
characterized from questionnaire data. In the absence of quality data from industry, a model was
developed to estimate the mass loading of constituents within the wastewater, resulting from
drag-out, during the production of 260,000 ssf of PWB by the surface finishing process. The
mass of chemical transferred per day to the wastewater, as well as other model results, are
presented in Appendix E.
3-32
-------
Bath Chemicals Used (I^. Bath concentrations of individual chemical constituents are
presented in Appendix B. The volume of chemicals consumed per year was determined by
modeling the time it would take the generic immersion tin process described in Figure 3-8 to
produce a specific PWB throughput. A detailed description of the process modeling is presented
in Section 4.2, Cost Analysis. The number of bath replacements (calculated from the modeled
time) was then multiplied by the volume of the bath to determine the volume of a bath chemical
consumed per year. Immersion tin process chemical consumption is presented in Appendix G.
Cleaning Chemicals (I4). One out of 15 immersion tin baths were reported to be cleaned
using chemicals, however, data concerning the type of cleaning chemical(s) were not collected by
the questionnaire. The bath reported to require chemical flushing to clean the tank during bath
replacement was the immersion tin bath. Water is most frequently used to clean tanks prior to
new bath make-up.
Spent Bath Solutions (W2). The quantity of spent bath solution could not be determined
directly from the questionnaire data. However, the volume of spent bath chemistry was
calculated by determining the number of bath changes required per year and multiplying by the
average volume of the process tank (see Section 4.2, Cost Analysis). The concentrations of
chemical constituents within the spent bath solutions were assumed to be the same as make-up
bath concentrations.
Spent bath treatment and disposal methods were presented in Table 3-4. Drummed for
off-site treatment and pH neutralization are reported as common treatment methods for the
conveyorized immersion tin processes. Respondents for the non-conveyorized, vertical process
reported that pH neutralization, precipitation pretreatment, ion exchange with on-site metal
reclaim and drummed for off-site treatment are all treatment options reported by respondents.
Evaporation From Baths (At). Air releases are modeled in Section 3.2, Exposure
Assessment. A summary of data collected from the questionnaire is presented below:
• For the conveyorized immersion tin processes, circulation pumps are the most reported
mixing methods for all baths. Full enclosure and venting are the most common methods
of vapor control reported by respondents for baths other than the pre-dip bath.
For non-conveyorized immersion tin processes, panel agitation and circulation pumps are
the most reported mixing methods for all baths. Venting to the outside is the most
prevalent form of vapor control reported (33 percent), though the use of bath covers are
also reported.
Table 3-3 lists the bath surface area, volume, and bath temperature data reported by
respondents to the PWB Workplace Practices Questionnaire.
Evaporation From Drying/Ovens (A2). Oven drying occurs directly after the immersion
tin bath. Any solution adhering to the PWBs is volatilized during the drying of the PWB by the
oven. Air emissions resulting from oven drying were not modeled. No air knife is required by
this process.
3-33
-------
Chemicals Incorporated Onto PWBs (Pi). A layer of metallic tin is deposited onto the
PWB by the immersion tin processes. The amount of tin incorporated onto a PWB was
calculated at 0.0038 oz/ssf. Tin consumption is discussed further in Section 5.1, Resource
Conservation.
Drummed Solid or Liquid Waste (S3). Questionnaire respondents indicated that none
of the immersion tin baths contain hazardous waste constituents as defined by RCRA. A
detailed discussion of RCRA wastes can be found in Section 4.3, Regulatory Status. In response
to a separate question regarding spent bath treatment (see Table 3-4), five out of 17 immersion tin
baths were reported by respondents to be drummed and sent off-site for recycling or disposal.
3.1.4 Uncertainties in the Source Release Assessment
Uncertainties and variations in the data include both gaps in knowledge (uncertainty) and
variability among facilities and process alternatives. These are discussed below.
For the PWB Workplace Practices Questionnaire data:
• There may be uncertainties due to misinterpretation of a question, not answering a
question that applies to that facility, reporting inaccurate information or numbers in
different units (e.g., using a mass unit to report a volumetric measurement). Also,
because of a limited number of responses for the alternative processes, information more
typical for that process may not be reported.
• Variation can occur within or among process alternatives, or from difference due to
varying amounts of PWB produced. According to the questionnaire database query
results, data from facilities with small amounts of PWB produced often produce
unrealistic results. Again, for surface finishing process alternatives with a limited number
of responses, statistical summaries of the data may be precluded, and data may not be
representative of most PWB facilities.
For the supplier-provided data:
• Knowledge gaps include a lack of information on proprietary chemicals, incomplete bath
composition data, and the reporting of wide ranges of chemical concentrations on a
MSDS rather then specific amounts in the formulations.
• Variation in bath chemistries and process specifications among suppliers can occur for a
given process alternative. The publicly-available bath chemistry data, chemical
concentrations, and supplier recommendations may not apply to a specific facility due to
variation in process set-up and operation procedures.
Other uncertainties pertain to the applicability and accuracy of estimates and assumptions
used in this assessment.
3-34
-------
3.2 EXPOSURE ASSESSMENT
Evaluating exposure for the PWB CTSA involves a series of sequential steps. The first
step is characterizing the exposure setting, which includes describing the physical setting and
characterizing the populations of interest and their activities that may result in exposure. These
are described in Section 3.2.1 for both workplace and surrounding population (ambient)
exposure.
The next step is selecting a set of workplace and population exposure pathways for
quantitative evaluation from the set of possible exposure pathways. This is discussed in Section
3.2.2.
Next, chemical concentrations are collected or estimated in all media where exposure
could occur. For the surface finishing processes, this consists of estimating the chemical
concentrations in the surface finishing baths, and performing fate and transport modeling to
estimate workplace and ambient air concentrations and surface water concentrations (Section
3.2.3).
The exposure-point concentrations and other exposure parameters are combined in
exposure models to estimate potential dose rates (PDRs) for all quantified pathways. These
exposure models, parameter values, and resulting exposure estimates are presented in Section
3.2.4. The final step, characterizing uncertainties, is in Section 3.2.5. The exposure assessment is
summarized in Section 3.2.6.
Because this CTSA is a comparative evaluation, and standardization is necessary to
compare results for the surface finishing processes, this assessment focuses on a "model"
(generic) PWB facility and uses aggregated data. In addition, this assessment focuses on
exposure from chronic, long-term, day-to-day releases from a PWB facility, rather than short-
term exposures to high levels of hazardous chemicals as there could be with a fire, spill, or
periodic releases. Due to the fixed amount of resources available to the project and the lack of
information to characterize such releases, high level, acute exposures could not be assessed.
3.2.1 Exposure Setting
Characterizing the exposure setting includes the following analyses:
• characterizing the physical environment (in this case, a model PWB facility, its surface
finishing process area, and the surrounding environment);
• identifying potentially exposed workers and their activities, and any potentially exposed
populations, human or ecological, that may be exposed through releases to the ambient
environment from PWB facilities;
• defining the workplace exposure scenarios to evaluate (where a scenario describes a
specified physical setting, exposed population, and activities that may result in exposure);
and
• defining ambient exposure scenarios to evaluate.
3-35
-------
Physical Environment
The surface finishing technologies are all wet chemistry processes consisting of a series of
chemical process baths, often followed by rinse steps, through which the PWB panels are passed
to apply the final surface finish. The exception is the HASL process, which combines the typical
cleaning and etching chemical processes with a mechanical process of dipping a board into
molten solder. (Details of each process are presented in Section 2.1, Chemistry and Process
Description of Surface Finishing Technologies.)
PWB Workplace Practices Questionnaire and Performance Demonstration data, collected
for 54 PWB facilities and their surface finishing process areas, were used to characterize a model
PWB facility. The PWB Workplace Practices Questionnaire database includes information from
29 facilities using the HASL process, eight using nickel/gold, one using nickel/palladium/gold,
nine using OSP, two using immersion silver, and five using immersion tin. Data from the
questionnaire database used in the exposure models are discussed further in Section 3.2.4.
Potentially Exposed Populations
Potentially exposed populations include both workers in the PWB facilities and ecological
and human populations in the vicinity of the facilities. Each of these are discussed below.
PWB Facility Employees. The questionnaire included questions about the types of
workers who might be present in the surface finishing process area. These include:
• line operators;
• laboratory technicians;
• maintenance workers;
• supervisory personnel;
• wastewater treatment operators;
• quality inspectors; and
• other employees.
General Population Outside the Facility. PWB facilities that are included in the PWB
Workplace Practices Questionnaire and Performance Demonstration database are located
throughout the U.S. This assessment estimates potential exposure to a hypothetical community
living near a model PWB facility, based on a residential scenario. The primary exposure route is
inhaling airborne chemicals originating from a PWB facility.
Surface Water. Exposure to ecological populations could also occur outside a PWB
facility. In this assessment we evaluated exposure to aquatic organisms in a stream that receives
treated wastewater from a facility.
3-36
-------
Workplace Exposure Scenarios
A scenario describes the exposure setting, potentially exposed populations or individuals,
and activities that could lead to exposure. For workplace exposures, the setting involves the
surface finishing process in a PWB facility. PWB Workplace Practices Questionnaire data are
used here to determine the types of workers that may be exposed and to characterize their
activities. Worker activities include working in the process area, surface finishing line operation,
chemical bath sampling, chemical bath additions, chemical bath replacement, rack cleaning,
conveyor equipment cleaning, and filter replacement.
Working in the Process Area. Workers may inhale airborne chemicals in the surface
finishing process area. Line operators are expected to have the highest inhalation exposure,
because they are typically in the process area for the longest time each day. For other types of
workers, their inhalation exposure would be proportional to their time spent in the process area.
Surface Finishing Line Operation. Potential for exposure during surface finishing line
operation is expected to vary significantly among process methods. Non-conveyorized process
configurations can be operated manually, automatically, or with a semi-automated system. In
manual methods, a line operator stands at the bath and manually lowers and raises the panel
racks into and out of each bath. A vertical/automated method is completely automated, where
panel racks are lowered and raised into vertical tanks by a robotic arm; line operators load and
unload panels from the racks. A manually-controlled vertical hoist is a semi-automated system
where racks are lowered into and raised out of a series of vertical chemical baths by a line
operator-controlled hoist. The hoist is controlled by a hand-held control panel attached to the
hoist by a cable. The conveyorized process configuration uses an automated method where
panels are transported horizontally into and out of process baths by means of a conveyor; line
operators load and unload panels from the conveyor system. Based on the workplace practices
data:
• For HASL, eight out of 29 facilities reported using non-conveyorized lines, and 21
reported conveyorized lines.
• The eight nickel/gold and one nickel/palladium/gold facilities all reported using non-
conveyorized lines.
• For facilities using OSP, four reported non-conveyorized lines and five reported
conveyorized lines.
• Both facilities using immersion silver use conveyorized lines.
• For immersion tin, three facilities reported using non-conveyorized and two facilities use
conveyorized lines.
Of the non-conveyorized systems described in the questionnaire, ten are
vertical/automated systems, ten are completely manual, one uses a manually-controlled hoist,
one HASL line is partly conveyorized, and two other systems were undefined. As a conservative
but consistent assumption, we assumed that workers manually lower and raise panel racks for all
non-conveyorized process alternatives.
3-37
-------
Chemical Bath Sampling. Based on the questionnaire database, chemical baths are
normally sampled manually by dipping/ladling. Other methods include manual sampling with a
pipette or other device, and automated sampling. We assumed there could be dermal contact
with bath chemicals from this activity, and quantified dermal exposure for laboratory technicians
and for line operators on conveyorized lines.
Chemical Bath Additions. Methods of chemical additions, from the database, are as
follows:
• Most facilities pour chemicals directly into the bath or tank.
• Other reported methods include manual pumping, or some combination of pumping,
pouring, and/or scooping chemical formulations into a bath.
Data were collected for the length of time required to make chemical additions, and on the criteria
used to determine when to add chemicals to the baths. Some facilities base chemical addition
requirements on time elapsed, some on the surface area of boards processed, and some on the
concentration of key constituents. For these reasons, complicated by the fact that most facilities
running alternatives to HASL do not run those lines at full capacity, a typical addition frequency
could not be determined. Therefore, exposure was not quantified separately for this activity.
Chemical Bath Replacement. This process includes removing the spent bath, cleaning
the empty tank, and making up fresh bath solutions. In this process, a worker could be exposed
to chemicals in the spent bath, on the inside walls of the emptied bath, or to chemicals in the new
bath solution.
Rack Cleaning. The racks that hold PWB panels can be cleaned in a variety of ways.
These include cleaning in a chemical bath on the surface finishing line or using non-chemical
cleaning methods. Of the six facilities that provided information on rack cleaning, four reported
using non-chemical methods, one reported using a chemical bath on the surface finishing line,
and one reported shipping racks offsite for cleaning. Dermal exposure for rack cleaning is not
quantified separately for this activity.
Conveyor Equipment Cleaning. Conveyor equipment cleaning involves regular
equipment maintenance for conveyorized surface finishing lines. Methods include chemical
baths on the surface finishing line, chemical rinse, manual scrubbing with chemicals, non-
chemical cleaning, and continuous cleaning as part of the process line. It was assumed that
workers could be exposed to bath chemicals during cleaning.
Filter Replacement. Filter replacement could result in exposure to the material on the
filter or in the bath. Whether the pathway is significant to worker risk will depend, in part, on the
chemical constituents in the bath.
Use of Personal Protective Equipment (PPE). It is assumed that the only PPE used is
eye protection, and that the line operator's hands and arms may contact bath solutions. This is a
3-38
-------
conservative but consistent assumption for all process alternatives and worker activities,
particularly for dermal exposure. While many PWB facilities reported that line operators do wear
gloves for various activities, the assumption that the line operator's hands and arms may contact
bath solutions is intended to account for the fraction of workers who do not. For workers who
do wear gloves, dermal contact exposure is expected to be negligible.
Summary of Occupational Scenarios
Surface Finishing Line Operators. In general, line operators perform several activities,
as described above, including working in the surface finishing process area, surface finishing line
operation, chemical bath replacement, conveyor equipment cleaning, filter replacement, chemical
bath sampling, and making chemical bath additions. Some kind of local ventilation is typically
used for the process line.
There are two different scenarios for line operators depending on process configuration.
For non-conveyorized processes, dermal exposure could occur through routine line operation as
well as bath maintenance activities. Inhalation exposure could occur throughout the time period
a line operator is in the surface finishing process area. Conveyorized processes are enclosed and
the line operator does not contact the bath solutions in routine line operation; he or she only
loads panels at the beginning of the process and unloads them at the end of the process. For
conveyorized processes, dermal exposure is primarily expected through bath maintenance
activities such as bath replacement, filter replacement, bath sampling, and conveyor equipment
cleaning. Because the conveyorized lines are enclosed and typically vented to the outside,
inhalation exposure to line operators and other workers is much lower than for the conveyorized
processes and are not presented separately.4
Laboratory Technicians. In general, laboratory technicians perform one activity
pertaining to the surface finishing line, chemical bath sampling, in addition to working in the
surface finishing process area. Bath sampling exposure is quantified separately for laboratory
technicians.
Other Workers in the Surface Finishing Process Area. Other workers in the surface
finishing process area may include maintenance workers, supervisory personnel, wastewater
treatment operators, contract workers, and other employees. They perform activities not directly
related to the surface finishing line, but typically spend some time in the surface finishing process
area. Because the line operators spend the most amount of time per shift, exposure via inhalation
is quantified for them (for non-conveyorized processes), and is characterized for the other
employees in terms of the time spent in the process area relative to line operators.
4 Inhalation exposures for conveyorized process configurations were initially assumed to be negligible, and are
not presented separately here. Some inhalation exposure is possible, however, during sampling and bath replacement,
when the baths are opened for a short period of time. After characterizing risks from inhalation for non-conveyorized
lines, inhalation exposures and risks were estimated for the subset of inhalation chemicals of concern for
conveyorized lines. No chemical exposures from inhalation resulted in risks above concern levels for conveyorized
lines.
3-39
-------
Ambient Exposure Scenarios
Ambient refers to the nearby area outside of a PWB facility. As discussed in Section 3.1,
Source Release Assessment, chemicals may be released to air, surface water, and/or possibly
land. Receptors include members of general population living near a PWB facility and aquatic
organisms, such as fish, in surface water receiving treated wastewater from a PWB facility.
Exposure is also possible to animals on land or birds. The ecological assessment focused on
aquatic life because much more data are available.
3.2.2 Selection of Exposure Pathways
The definition of exposure scenarios leads to selection of the exposure pathways to be
evaluated. An exposure scenario may comprise one or several pathways. A complete exposure
pathway consists of the following elements:
a source of chemical and mechanism for release;
an exposure point;
a transport medium (if the exposure point differs from the source); and
an exposure route.
Tables 3-6 and 3-7 present an overview of the pathway selection for workplace and
surrounding population exposures, respectively. For the workplace, a potential pathway not
quantified is oral exposure to vapors or aerosols. For example, oral exposure could occur if
inhaled chemicals are coughed up and then swallowed.
Table 3-6. Workplace Activities and Associated Potential Exposure Pathways
Activities
Potential Pathways
Evaluation Approach and Rationale
Line Operators B
Surface Finishing Line
Operation
Working in Process Area
Dermal contact with
chemicals in surface
finishing baths.
Inhalation of vapors or
aerosols from surface
finishing baths.
Inhalation of vapors or
aerosols from surface
finishing baths.
Exposure quantified for non-conveyorized lines;
the highest potential dermal exposure is expected
from this activity. Exposure for conveyorized lines
assumed to be negligible for this activity.
Exposure quantified initially only for non-
conveyorized lines. Exposure for conveyorized
lines assumed to be much lower. b
Exposure quantified for non-conveyorized lines.
3-40
-------
Activities
Chemical Bath
Replacement;
Conveyor Equipment
Cleaning; Filter
Replacement;
Chemical Bath Sampling
Rack Cleaning
Chemical Bath Additions
Potential Pathways
Dermal contact with
chemicals in bath or on
filters.
Inhalation of vapors or
aerosols from surface
finishing baths.
Dermal contact with
chemicals on racks.
Inhalation of vapors or
aerosols from surface
finishing baths.
Dermal contact with
chemicals added.
Inhalation of vapors or
aerosols from surface
finishing baths or while
making bath additions.
Evaluation Approach and Rationale
Exposure quantified for conveyorized lines for all
activities together (bath sampling quantified
separately for laboratory technicians). Exposure
not quantified separately for these activities on
non -conveyorized lines.
Not quantified separately. Included in "working in
process area" for non-conveyorized lines; not
quantified due to modeling limitations for
conveyorized lines. b
Not quantified; limited data indicate this is not
performed by many facilities.
Not quantified separately. Included in "working in
process area" for non-conveyorized lines; not
quantified due to modeling limitations for
conveyorized lines.
Not quantified separately from chemicals already in
the baths.
Not quantified separately. Included in "working in
process area" for non-conveyorized lines; not
quantified due to modeling limitations for
conveyorized lines.
Laboratory Technicians
Chemical Bath Sampling
Working in Process Area
Dermal contact with
chemicals in surface
finishing baths.
Inhalation of vapors or
aerosols from surface
finishing baths.
Inhalation of vapors or
aerosols from surface
finishing baths.
Exposure quantified for conveyorized and
non-conveyorized lines.
Not quantified separately (included in "working in
process area").
Exposure quantified for line operators for non-
conveyorized lines; exposure for other workers is
proportional to their exposure durations.
Maintenance Workers, Supervisory Personnel, Wastewater Treatment Operators, Contract
Workers, and Other Workers
Working in Process Area
Inhalation of vapors or
aerosols from surface
finishing baths.
Dermal contact with
chemicals in surface
finishing baths.
Exposure quantified for line operators for non-
conveyorized lines; exposure for other workers is
proportional to their exposure durations.
Not quantified. a
a This assumes surface finishing line operators are the most exposed individuals and perform all direct maintenance
on the surface finishing line, including filter replacement and equipment cleaning.
b Inhalation exposures for conveyorized process configurations were initially assumed to be negligible. Some
inhalation exposure is possible, however, during sampling and bath replacement, when the baths are opened for a
short period of time. After characterizing risks from inhalation for non-conveyorized lines, inhalation exposures and
risks were estimated for the subset of inhalation chemicals of concern for conveyorized lines. No chemical exposures
from inhalation resulted in risks above concern levels for conveyorized lines.
3-41
-------
Table 3-7. Potential Population Exposure Pathways
Population
Residents Living
Near a PWB Facility
Ecological
Potential Pathways
Inhalation of chemicals released
to air.
Contact with chemicals released
to surface water directly or
through the food chain.
Exposure to chemicals released to
land or groundwater.
Exposure to chemicals released to
surface water.
Exposure to chemicals released to
air or land.
Evaluation Approach and Rationale
Exposure quantified for all potential
carcinogens and any other chemical released
at a rate of at least 23 kg/year.
Not evaluated. Direct exposure to surface
water is not expected to be a significant
pathway; modeling exposure through the
food chain (e.g., someone catching and
eating fish) would be highly uncertain.
Not evaluated. Not expected to be a
significant pathway; modeling releases to
groundwater from a landfill would be highly
uncertain.
Screening-level evaluation performed.
Not evaluated. The human (residential)
evaluation air exposure could be used as a
screening-level assessment for animals living
nearby. Releases directly to land are not
expected, and animals are not directly
exposed to groundwater.
Population exposures may occur through releases to environmental media (i.e., releases to
air, water, and land). The pathways for which exposure is estimated are inhalation of chemicals
released from a facility to a nearby residential area and releases of chemicals in wastewater to a
receiving stream, where aquatic organisms, such as fish, may be exposed through direct contact
with chemicals in surface water.
Air releases from the surface finishing process are modeled for the workplace. These
modeled emission rates are used in combination with an air dispersion model to estimate air
concentrations to a nearby population.
Exposures and risks from surface water are evaluated by identifying chemicals potentially
released to surface water from process rinse steps following wastewater treatment. This exposure
pathway is described in Section 3.2.3.
Possible sources of releases to land from surface finishing processes include bath filters
and other solid wastes from the process line, chemical precipitates from baths, and sludge from
wastewater treatment. These sources are discussed in Section 3.1, Source Release Assessment.
Reliable characterization data for potential releases to land are not available; therefore, the
exposure assessment does not estimate the nature and quantity of leachate from landfills or
effects on groundwater.
3-42
-------
3.2.3 Exposure-Point Concentrations
An exposure-point concentration is a chemical concentration in its transport or carrier
medium, at the point of contact (or potential point of contact) with a human or environmental
receptor. Sources of data for estimating exposure-point concentrations include monitoring data,
publicly-available bath chemistry data, some proprietary bath chemistry data, and fate and
transport models used to estimate air releases and air concentrations. Bath concentrations for
dermal exposure were estimated from bath chemistry data. Monitoring data were used for lead
from the HASL process. Fate and transport modeling were performed to estimate air
concentrations for workplace and surrounding population exposures. Use of monitoring data and
modeling used to estimate air concentrations are described in this section.
Monitoring Data
Air monitoring data for lead have been provided by one PWB manufacturing facility. A
combination of personal monitoring for HASL line operators and air samples from the HASL
process area result in an average air concentration of 0.003 mg/m3. It should be noted that these
monitoring data are limited to only one PWB manufacturer, and may vary from facility to facility.
In addition, air sampling results from hand soldering operations were reported in one study
(Monsalve, 1984), ranging from < 0.001 mg/m3 to 2 mg/m3.
Modeling Workplace Air Concentrations
Air emission models, combined with an indoor air dilution model, were used to estimate
chemical air concentrations for worker inhalation exposure from PWB surface finishing lines
(Robinson et al., 1997). Three air emission models were used to estimate worker exposure:
1. Volatilization of chemicals from the open surface of surface finishing tanks.
2. Volatilization of chemicals induced by air sparging.
3. Aerosol generation induced by air sparging.
The first model was applied to determine volatilization of chemicals from un-sparged
baths. For the air-sparged baths, the total air emission rate for chemicals was determined by
summing the releases from all three models. Modeled emission rates were then put into an
indoor air dilution model to estimate workplace air concentrations. For models 1 and 2,
volatilization was modeled only for those chemicals with a vapor pressure above 10"3 torr (a vapor
pressure less than 10"3 torr was assumed for inorganic salts even if vapor pressure data were not
available). A review of the relevant literature, descriptions of the models, and examples
demonstrating the use of the models are available in the December 22, 1995, Technical
Memorandum, Modeling Worker Inhalation Exposure (Appendix D).
Volatilization of Chemicals from the Open Surface of Surface Finishing Tanks.
Most plating tanks have a free liquid surface from which chemicals can volatilize into the
workplace air. Air currents across the tank will accelerate the rate of volatilization. The
3-43
-------
following model for evaporation of chemicals from open surfaces was used, based on EPA's
Chemical Engineering Branch (CEB) Manual (U.S. EPA, 1991a):
-|0.5
where,
y>°
A
vz
Az
volatilization rate of chemical y from open tanks (mg/min)
concentration of chemical y in bulk liquid (mg/L)
dimensionless Henry's Law Constant (Hc) for chemical y
bath surface area (m2)
molecular diffusion coefficient of chemical y in air (cm2/sec)
air velocity (m/sec)
pool length along direction of air flow (m)
Concentration of chemical in bulk liquid (cLj) is the bath concentration. The coefficient
of 1,200 includes a factor of 600 for units conversion.
Henry's constants were corrected for bath temperature. Bath temperature varies by
process type and bath type; bath temperature data from the questionnaire database were
determined by specific process type and bath type.
Bath surface areas used in the air modeling were determined from the questionnaire
database. For non-conveyorized lines, an overall average for all baths and all processes of 422 sq
in (0.280 m2) was used. For conveyorized lines, an average was used for each type of process
bath, as follows:
Conveyorized Bath Type
Cleaner baths
Immersion silver
Immersion tin
Microetch baths
OSP
Predip baths
Average Surface Area
(sq in)
1,078
4,364
1,436
1,629
2,573
1,004
Some limitations of the model should be pointed out. The model was developed to
predict the rate of volatilization of pure chemicals, not aqueous solutions. The model was also
derived using pure chemicals. As a result, the model implicitly assumes that mass transfer
resistance on the gas side is the limiting factor. The model may overestimate volatilization of
chemicals from solutions when liquid-side mass transfer is the controlling factor.
3-44
-------
Volatilization of Chemicals from Air-Sparged Surface Finishing Tanks.
Volatilization and aerosol generation from air-sparged baths were modeled only for those baths
that are mixed by air sparging, as indicated in the PWB Workplace Practices Questionnaire and
Performance Demonstration data; this includes the electroless nickel baths in nickel/gold and
nickel/palladium/gold processes. Mixing in the baths is commonly accomplished by sparging the
tank with air. The equation used for predicting the mass transfer rate from an aerated system is
based on volatilization models used in research of aeration in wastewater treatment plants:
1-expl JWl^
where,
FyiS = mass transfer rate of chemical y out of the system by sparging (mg/min)
QG = air sparging gas flow rate (L/min)
Hy = dimensionless Henry' s Law Constant (Hc) for chemical y
cLj = concentration of chemical y in bulk liquid (mg/L)
KOLJ = overall mass transfer coefficient for chemical y (cm/min)
a = interfacial area of bubble per unit volume of liquid (cm2/cm3)
VL = volume of liquid (cm3)
Data for the sparging air flow rate (QG) come from information supplied by a PWB manufacturer.
Aerosol Generation from Air-Sparged Surface Finishing Tanks. Aerosols or mists
are also potentially emitted from process baths. This was estimated for electroless nickel baths in
nickel/gold and nickel/palladium/gold processes. The rate of aerosol generation has been found
to depend on the air sparging rate, bath temperature, air flow rate above the bath, and the distance
between bath surface and the tank rim. The following equation is used to estimate the rate of
aerosol generation (Berglund andLindh, 1987):
[.. . i
S Sr/fl ^ID / A\+0 01 \ F F F
-J.-J-VJl/ \\sr^ ' **-1 Vy.\J 1 I 1
-------
M,
F = L f F
y'a Mh JlE y's
where,
Fyj3 = rate of mass transfer from the tank to the atmosphere by aerosols (mg/min)
fffi = fraction of bubble interface ejected as aerosols (dimensionless)
M! = mass of contaminant at the interface (mg)
Mb = mass of contaminant in gas bubble (mg)
The literature on aerosol generation indicates that the typical size of aerosols is one to ten
microns; this is important to note because particles in this range are more inhalable. Larger sized
particles tend to fall back into baths rather than remaining airborne and dispersing throughout the
room.
Calculation of Chemical Concentration in Workplace Air from Emission Rates. For
unsparged baths, the total emission rate is equal to Fyj0 calculated by the first equation. For
sparged baths, the total emission rate is equal to Fyj0 + FyiS + Fyja, as calculated by the three
equations described above. The indoor air concentration is estimated from the total emission rate
using the following equation (U.S. EPA, 1991a):
Cy = FyTl(Qk)
where,
Cy = workplace contaminant concentration (mg/m3)
Fy;r = total emission rate of chemical from all sources (mg/min)
Q = ventilation air flow rate (m3/min)
k = dimensionless mixing factor
The CEB Manual commonly uses values of the ventilation rate (Q) from 500 cubic feet
per minute (cfm) to 3,500 cfm; a ventilation rate for surface finishing lines of 13.6 m3/min (480
cfm) was determined by taking the 10th percentile air flow rates from the facility questionnaire
database for general ventilation. An average room volume of 505 m3 (18,200 ft3) was determined
by assuming a ten foot room height and using the average room size from the questionnaire
database. The combination of room volume and ventilation rate is equivalent to an air turnover
rate of 0.026 per minute (1.56 per hour). The mixing factor (k) could be used to account for slow
and incomplete mixing of ventilation air with room air; however, a value of 1.0 was used for this
factor consistent with the assumption of complete mixing.
Other assumptions pertaining to these air models include the following:
Deposition on equipment, condensation of vapors, and photodegradation are negligible.
Incoming air is contaminant-free.
The concentration of contaminant at the beginning of the day is zero.
3-46
-------
As much air enters the room as exits through ventilation (mass balance).
Room air and ventilation air mix ideally.
Sensitivity Analysis. Model sensitivity and uncertainty was examined for the making
holes conductive (MHC) project (U.S. EPA, 1998b) using Monte Carlo analysis, with the air
transport equations outlined above, and probability distributions for each parameter based on
data from the PWB Workplace Practices Questionnaire. The analysis was conducted using a
Monte Carlo software package (Crystal Ball™, Decisioneering, Inc., 1993) in conjunction with a
spreadsheet program. Because the same models are used for this surface finishing evaluation,
and the model facility is similar to that developed for MHC, the results of this sensitivity analysis
are relevant to surface finishing air modeling as well.
The sensitivity analysis suggested that a few parameters are key to modeling chemical
emissions from PWB tanks. These key parameters are air turnover rate, bath temperature,
chemical concentration in the bath, and Hc. The air turnover rate assumption contributes most to
overall model variance. The chemical bath concentration and bath temperature also contribute
variance to the model, but are less important than air turnover rate. This statement is supported
by the fact that relatively accurate information is available on their distributions. Hc appears to be
least important of the four, but may have more variability associated with it. The models appear
to be largely indifferent to small changes in most other parameters.
Modeled emission rates and workplace air concentrations are presented in Table 3-8.
Table 3-8. Results of Workplace Air Modeling
Chemical a
Total Emission
Rate (Fy,T)
(mg/min)
Workplace
Air Cone.
(Cy)
(mg/m3)
Federal OSHA and/or NIOSH
Permissible Inhalation Exposure
Limits (mg/m3) b
HASL, Non-conveyorized
1,4-Butenediol
Alkylaryl sulfonate
Arylphenol
Ethylene glycol
Ethylene glycol monobutyl ether
Hydrochloric acid
Hydrogen peroxide
Phosphoric acid
9.8
0.018
0.0060
12
120
0.89
5.2
1.5
0.75
0.0014
4.6E-04
0.94
8.9
0.068
0.40
0.12
none
NR
NR
no OSHA PEL or NIOSH REL
NIOSH REL: 24 (5 ppm)
OSHA PEL: 240 (50 ppm)
NIOSH REL, C: 7 (5 ppm)
OSHA PEL, C: 7 (5 ppm)
NIOSH REL: 1.4 (1 ppm)
OSHA PEL 1.4(1 ppm)
NIOSH REL: 1, STEL: 3
OSHA PEL: 1
Nickel/Gold, Non-conveyorized
Aliphatic acid A
Aliphatic acid B
Aliphatic acid E
77
5.4E-04
100
5.9
4.1E-05
7.8
NR
NR
NR
3-47
-------
Chemical a
Aliphatic dicarboxylic acid A
Aliphatic dicarboxylic acid C
Alkyldiol
Ammonia compound B
Ammonium hydroxide
Hydrochloric acid
Hydrogen peroxide
Inorganic metallic salt A
Inorganic metallic salt B
Inorganic metallic salt C
Malic acid
Nickel sulfate
Phosphoric acid
Potassium compound
Sodium hypophosphite
Urea compound B
Total Emission
Rate (Fy,T)
(mg/min)
0.10
0.049
22
0.025
1.2
26
3.8
3.1E-05
2.1E-03
2.2E-05
0.22
0.55
1.2
1.0
0.64
7.6E-04
Workplace
Air Cone.
(Cy)
(mg/m3)
0.0080
0.0038
1.6
0.0019
0.094
2.0
0.29
2.4E-06
1.6E-04
1.7E-06
0.017
0.042
0.092
0.079
0.048
5.8E-05
Federal OSHA and/or NIOSH
Permissible Inhalation Exposure
Limits (mg/m3) b
NR
NR
NR
NR
none
NIOSH REL, C: 7 (5 ppm)
OSHA PEL, C: 7 (5 ppm)
NIOSH REL: 1.4 (1 ppm)
OSHA PEL: 1.4 (1 ppm)
NR
NR
NR
none
NIOSH REL, Ca: 0.0 15
OSHA PEL: 1
NIOSH REL: 1, STEL: 3
OSHA PEL: 1
NR
none
NR
Nickel/Palladium/Gold, Non-conveyorized
Aliphatic acid B
Aliphatic acid E
Aliphatic dicarboxylic acid A
Aliphatic dicarboxylic acid C
Alkyldiol
Ammonia compound B
Ammonium hydroxide
Ethylenediamine
Hydrochloric acid
Hydrogen peroxide
Inorganic metallic salt B
Malic acid
Nickel sulfate
Phosphoric acid
Potassium compound
5.6E-04
140
0.11
0.051
22
0.026
2.0
0.064
28
3.7
0.0021
0.23
0.90
1.2
1.1
4.2E-05
11
0.0082
0.0039
1.7
0.0020
0.16
0.0048
2.1
0.28
1.6E-04
0.018
0.068
0.092
0.082
NR
NR
NR
NR
NR
NR
none
NIOSH REL: 25 (10 ppm)
OSHA PEL: 25 (10 ppm)
NIOSH REL, C: 7 (5 ppm)
OSHA PEL, C: 7 (5 ppm)
NIOSH REL: 1.4 (1 ppm)
OSHA PEL: 1.4 (1 ppm)
NR
none
NIOSH REL, Ca: 0.0 15
OSHA PEL: 1
NIOSH REL: 1, STEL: 3
OSHA PEL: 1
NR
3-48
-------
Chemical a
Propionic acid
Sodium hypophosphite
Urea compound B
Total Emission
Rate (Fy,T)
(mg/min)
26
0.85
0.0015
Workplace
Air Cone.
(Cy)
(mg/m3)
2.0
0.065
1.2E-04
Federal OSHA and/or NIOSH
Permissible Inhalation Exposure
Limits (mg/m3) b
NIOSHREL:30(10ppm)
STEL: 45 (15 ppm)
none
NR
OSP, Non-conveyorized
Acetic acid
Arylphenol
Ethylene glycol
Hydrochloric acid
Hydrogen peroxide
Phosphoric acid
74
0.0059
23
2.0
1.8
1.2
5.6
4.5E-04
1.7
0.15
0.14
0.092
NIOSH REL: 25 (10 ppm),
STEL: 37 (15 ppm)
OSHA PEL: 25 (10 ppm)
NR
no OSHA PEL or NIOSH REL
NIOSH REL, C: 7 (5 ppm)
OSHA PEL, C: 7 (5 ppm)
OSHA PEL, NIOSH REL: 1.4
(1 ppm)
NIOSH REL: 1, STEL 3
OSHA PEL: 1
Immersion Tin, Non-conveyorized
Aliphatic acid D
Alkylaryl sulfonate
Cyclic amide
Hydrochloric acid
Hydroxy carboxylic acid
Phosphoric acid
Urea compound C
27
0.026
22
0.090
37
0.74
250
2.1
0.0020
1.7
0.069
2.8
0.056
19
NR
NR
NR
NIOSH REL, C: 7 (5 ppm)
OSHA PEL, C: 7 (5 ppm)
NR
NIOSH REL: 1, STEL: 3
OSHA PEL: 1
NR
Only chemicals with calculated values are presented. A number was not calculated for a chemical if its vapor
pressure is below the 1 x 10~3 torr cutoff and it is not used in any air-sparged bath. For these chemicals, air
concentrations are expected to be negligible.
b Source: NIOSH, 1999. RELs and/or PELs for proprietary chemicals are not presented in order to protect
confidential chemical identities. Notes about these values follow:
NIOSH REL: Recommended exposure limit, a time-weighted average (TWA) concentrations for up to a
10-hour workday during a 40-hour workweek.
OSHA PEL: The OSHA permissible exposure limits, as found in Tables Z-l, Z-2, and Z-3 of the OSHA
General Industry Air Contaminants Standard (29 CFR 1910.1000). Unless noted otherwise, PELs are TWA
concentrations that must not be exceeded during any 8-hour workshift of a 40-hour workweek.
STEL: A short-term exposure limit; unless noted otherwise, this is a 15-minute TWA exposure that should
not be exceeded at any time during a workday.
C: A ceiling REL or PEL is designated by "C"; unless noted otherwise, the ceiling value should not be
exceeded at any time.
Ca: Any substance that NIOSH considers to be a potential occupational carcinogen is designated by the
notation "Ca."
Note: The numeric format used in these tables is a form of scientific notation, where the "E" replaces the " x 10s'.
Scientific notation is typically used to present very large or very small numbers. For example, 1.2E-04 is the same as
1.2 x 10"4, which is the same as 0.00012 in common decimal notation.
3-49
-------
Modeling Air Concentrations for Population Exposure
The following approach was used for dispersion modeling of air emissions from a single
facility:
• The Industrial Source Complex Long Term ISC(2)LT model was used from the
Risk*Assistant™ software program.
• A building (release) height of 3 meters was assumed.
• An area source with dimensions of 10 x 10m was assumed.
• Meteorological data for Oakland, California, Denver, Colorado, and Phoenix, Arizona
were used. (PWB facilities are located throughout the U.S., and many are in Southern
California. These three areas give the highest modeled concentrations around a facility for
any available city data in the model.)
• Regulatory default values were used for other model parameters. (These are model
defaults pertaining to plume rise, stack-tip downwash, buoyancy-induced dispersion,
wind profile exponents, vertical temperature gradient, and buildings adjacent to the
emission source.)
• An urban mode setting was used. (The setting can be either rural or urban. The urban
setting is appropriate for urban areas or for large facilities.)
• Because of the short time expected for chemical transport to nearby residents, chemical
degradation was not taken into account.
• A standard polar grid5 with 36 vector directions and one distance ring (at 100m) was used;
the highest modeled air concentration in any direction at 100 meters was used to estimate
population exposure.
An average emission rate-to-air concentration factor of 2.18 x 10"6 min/m3 was determined
using model results for the three locations. This factor was multiplied by the total emissions rate
for each chemical (in mg/min) to yield air concentrations at the receptor point, in units of mg/m3.
The emission rates calculated for workplace inhalation exposures (Table 3-8) are used for the
source emission rates to ambient air. Except for the carcinogen, inorganic metallic salt A,
ambient air concentrations are not reported for those chemicals with facility emission rates less
than 23 kg/year (44 mg/min), which is a screening threshold typically used by EPA.6 In addition,
ambient air concentrations for lead were estimated, based on this air dispersion model and HASL
workplace air monitoring data. Results of ambient air modeling are presented in Table 3-9.
5 A polar grid is a coordinate system that describes the location of a point by means of direction and distance in
relation to a central point (e.g., two miles northeast of the center). In the model, a series of regularly-spaced
concentric distance rings are defined at chosen intervals along with a defined number of direction vectors (e.g., north,
south, east, west, northeast, northwest, southeast, and southwest would be eight directions).
6 Under conservative assumptions, inhalation exposure to fugitive releases less than 23 kg/yr result in exposures
of less than 1 mg/yr for an individual.
3-50
-------
Table 3-9. Results of Ambient Air Modelim
Chemical
Emission Rate a Air Concentration b
(mg/min) (mg/m3)
HASL, Non-conveyorizd
Ethylene glycol monobutyl ether
Lead
120
0.039 c
2.55E-04
9.0E-08
HASL, Conveyorized
Ethylene glycol monobutyl ether
Lead
230
0.039 c
5.05E-04
9.0E-08
Nickel/Gold, Non-conveyorized
Aliphatic acid A
Aliphatic acid E
Inorganic metallic salt A
Nickel/Palladium/Gold, Non-conveyorized
Aliphatic acid E
OSP, Non-conveyorized
Acetic acid
OSP, Conveyorized
Acetic acid
Ethylene glycol
Immersion Tin, Non-conveyorized
Urea compound C
Immersion Tin, Conveyorized
Aliphatic acid D
Cyclic amide
Hydroxy carboxylic acid
Urea compound C
77
100
3.12E-05
140
74
280
46
250
67
53
90
610
1.68E-04
2.22E-04
6.81E-11
3.06E-04
1.62E-04
6.15E-04
9.94E-05
5.40E-04
1.46E-04
1.16E-04
1.96E-04
1.32E-03
a Only those chemicals with an emission rate of at least 23 kg/year (44 mg/min) are listed. Immersion silver had no
modeled emission rates above this cut-off.
b The numeric format used in this column is a form of scientific notation, where "E" replaces the " x 10X. Scientific
notation is typically used to present very large or very small numbers. For example, 1.2E-04 is the same as 1.2 x 10~4,
which is the same as 0.00012 in common decimal notation.
0 Based on air monitoring data from one facility, with an average workplace air concentration of 0.003 mg/m3.
3-51
-------
Surface Water
PWB manufacturers typically combine wastewater effluent from the surface finishing
process line with effluent from other PWB manufacturing processes prior to on-site wastewater
pretreatment. The pretreated wastewater is then discharged to a POTW. Consequently,
characterizing the process wastewater has been problematic. Because many of the chemical
constituents expected in the wastewater of the surface finishing process are also found in other
PWB manufacturing processes, testing data obtained from industry was not sufficient to
characterize what portion of the overall wastewater contamination was actually attributable to the
operation of the surface finishing process. Therefore, a model was developed to estimate
environmental releases to surface water for chemical constituents and concentrations in the
wastewater as a result of the operation of the surface finishing process alone.
In the absence of quality data from industry, this model was developed using laboratory
testing to determine the amount of drag-out from a wet chemistry process involving PWBs and
the amount of chemical disposed through bath replacement. The MHC process, which is similar
in operation to the surface finishing process, was used as the basis for the model because of the
availability of chemical formulation data at time of development. The term drag-out refers to the
process chemicals that are 'dragged' (lost) from chemical baths into the following water rinse
stages, during the processing of PWB panels through the surface finishing line. Residual process
chemicals are washed from the surface of the PWB by the rinse water stages resulting in
contamination of the rinse water. Industry has estimated that up to 95 percent of the chemical
contamination in the wastewater is attributable to drag-out (the remaining contamination results
from spent bath treatment and bath maintenance practices). The drag-out model is given by the
following linear regression equation:
DO = 18 + 201 (SIZE) - 60.1 (ELCTRLS) + 73 (WR/DT) - 20.9 (ALK) + 26 (HOLES) + 26.1 (WR) - 0.355 (DT)
where,
DO = drag-out from bath, ml/m2
SIZE = board area, m2
WR = withdraw rate, m/sec
DT = drain time, sec
ALK = 1 if the bath is an alkaline cleaner bath and = 0 otherwise
HOLES = 1 if the board is drilled and = 0 for undrilled boards (we assumed that all
boards were drilled)
ELCTRLS = 1 if the bath is an electroless copper bath and = 0 otherwise
The model was used to estimate the mass loading of constituents to the wastewater
resulting from drag-out during the production of 260,000 ssf of PWB by the surface finishing
process, by the following equation:
MDij = P * Cij * DOij / 1,000,000
where,
3-52
-------
MDij = drag-out mass of constituent I from bath j, g/d
P = PWB production rate, m2/day
Cij = concentration of constituent I in bath j, mg/L
The amount of chemical going to wastewater from bath replacement was calculated by:
Mbij = Fj/T * Vj * Cij / 1,000
where,
MBij = mass of constituent I from dumping bath j, g/d
Fj = replacement frequency for bath j, times/yr
T = operating time (from cost model, total production time minus down time), days/yr
V = volume of bath j,L
For non-conveyorized lines, the total mass per day going to wastewater is the sum of
drag-out mass and bath dumping mass for the constituent in all baths:
Mi = £ (MDij + MBij)
where,
Mi = total mass of constituent i going to wastewater, g/d, from all baths j containing
constituent i
Because conveyorized lines are designed to operate with minimal drag-out, and the drag-
out model was developed only for vertical configurations, bath replacement alone was considered
in estimating chemical amounts to wastewater. For conveyorized lines,
Mi = EMBij
j=i J
A detailed description of the model, along with the methods of model development,
validation and testing, and model limitations, are presented in Prediction of Water Quality from
Printed Wiring Board Processes (Robinson et al., 1999) and Appendix E.
Preliminary in-stream concentrations were then calculated from the mass loading by
considering dilution in the receiving stream and assuming no treatment, by:
Ci,sw=1000Mi/(Qsw + Qww)
where,
Ci,sw = preliminary surface water concentration of constituent i, assuming no treatment,
mg/L
Qsw = surface water flow rate, L/day
Qww = wastewater flow rate, L/day
3-53
-------
For surface water flow, we used a rate of 13.3 million liters/day. This is the 10th percentile
low flow rate (7Q10) for the distribution of streams associated with facilities with the Electronic
Components Manufacture SIC code. This type of flow rate is typically used by EPA for
comparisons of screening-level estimates of in-stream chemical concentrations versus concern
concentrations (CCs) for aquatic species.
These concentrations were then screened against CCs for toxicity to aquatic life (CCs are
discussed in Section 3.3.3 and Appendix H). For any chemicals with preliminary in-stream
concentrations exceeding CCs, a typical treatment efficiency was determined. For this purpose,
it was assumed that wastewater treatment consisted of primary treatment by gravitational settling
followed by complete-mix activated sludge secondary treatment and secondary settling
(clarification), as typically employed at POTWs. Treatment efficiencies were estimated on a
chemical-by-chemical basis using a combination of readily available data on the chemical or close
structural analogs and best professional judgment. Information sources included EPA's
Treatability Database, the Environmental Fate Data Base (Syracuse Research Corp., updated
periodically), the Handbook of Environmental Degradation Rates (Howard et al., 1991),
wastewater engineering handbooks such as Metcalf and Eddy, and various journal articles from
the published literature.
Treatment efficiencies were then applied to the chemical concentrations, and revised in-
stream concentrations were calculated. Select inorganic constituents that are targeted by industry
for treatment, such as metals, were assumed to be treated effectively by on-site treatment to
required effluent levels. These metals are not included in the surface water evaluation.
(Pretreatment is discussed further in Section 6.2, Control Technologies.) Results for chemicals,
excluding metals, where the initial stream concentration (without treatment) exceeded the CC for
that chemical are presented in Table 3-10. Full results are presented in Appendix E.
Table 3-10. Estimated Releases to Surface Water Following Treatment
Chemical a
Cone, in
Wastewater
Released to
Stream
(mg/L)
Stream Cone.
w/o POTW
Treatment
(mg/L)
Treatment
Efficiency
(%)
Stream
Cone, after
POTW
Treatment
(mg/L)
HASL, Non-conveyorized
1,4-Butenediol
Alkylaryl sulfonate
Citric acid
Ethylene glycol monobutyl ether
Hydrogen peroxide
Potassium peroxymonosulfate
49
2.3
94
71
195
390
0.10
0.0049
0.20
0.15
0.41
0.82
90
0
93
90
90
90
0.010
0.0049
0.014
0.015
0.041
0.082
HASL, Conveyorized
1,4-Butenediol
Alkylaryl sulfonate
Citric acid
23
1.0
42
0.076
0.0035
0.14
90
0
93
0.0076
0.0035
0.0099
3-54
-------
Chemical a
Ethylene glycol monobutyl ether
Hydrogen peroxide
Potassium peroxymonosulfate
Cone, in
Wastewater
Released to
Stream
(mg/L)
32
90
180
Stream Cone.
w/o POTW
Treatment
(mg/L)
0.11
0.30
0.61
Treatment
Efficiency
(%)
90
90
90
Stream
Cone, after
POTW
Treatment
(mg/L)
0.011
0.030
0.061
Nickel/Gold, Non-conveyorized
Hydrogen peroxide
Substituted amine hydrochloride
62
97
0.045
0.070
90
80
0.0045
0.014
Nickel/Palladium/Gold, Non-conveyorized
Hydrogen peroxide
Substituted amine hydrochloride
36
55
0.034
0.053
90
80
0.0034
0.011
OSP, Non-conveyorized
Alkylaryl imidazole
Hydrogen peroxide
200
110
0.33
0.18
90
90
0.033
0.018
OSP, Conveyorized
Alkylaryl imidazole
Hydrogen peroxide
75
61
0.18
0.15
90
90
0.018
0.015
Immersion Silver, Conveyorized
1,4-Butenediol
Fatty amine
Hydrogen peroxide
48
7.7
430
0.029
0.0047
0.26
90
95
90
0.0029
0.00023
0.026
Immersion Tin, Non-conveyorized
Alkylaryl sulfonate
Citric acid
Ethylene glycol monobutyl ether
Potassium peroxymonosulfate
Quantenary alkylammonium
chlorides
Thiourea
Urea compound C
1.2
660
36
200
42
170
35
0.0021
1.2
0.064
0.36
0.074
0.30
0.062
0
93
90
90
90
90
90
0.0021
0.082
0.0064
0.036
0.0074
0.030
0.0062
Immersion Tin, Conveyorized
Potassium peroxymonosulfate
68
0.041
90
0.0041
a This includes any chemicals, except metals, where the initial stream concentration (without treatment) exceeded the
CC for that chemical. Metals are not included; it was assumed that metals are targeted for effective on-site treatment.
3-55
-------
3.2.4 Estimating Potential Dose Rates
This section contains information on exposure models, parameter values, and resulting
exposure estimates for potential workplace and population exposures. Data on frequency and
duration of most activities were derived from the PWB Workplace Practices Questionnaire
database, Product Data Sheets from chemical suppliers (e.g., bath change out rates), and the
process simulation model (e.g., days of process operation). The general models for calculating
inhalation and dermal potential dose rates are discussed below.
Inhalation Exposures
The general model for inhalation exposure to workers is from CEB (U.S. EPA, 1991a):
I = (Ca)(IR)(ET)
where,
I = daily inhalation potential dose rate (mg/day)
Ca = airborne concentration of substance (mg/m3)
(Note: this term is denoted "Cy" in air modeling equation in Section 3.2.3.)
IR = inhalation rate (m3/hr)
ET = exposure time (hr/day)
Daily exposures are averaged over a lifetime (70 years) for carcinogens, and over the
exposure duration (e.g., 25 years working in a facility) for non-carcinogens.7 The following
equations are used to estimate average daily doses for inhalation:
LADD = (I)(EF)(ED)/[(BW)(ATCAR)]
ADD = (I)(EF)(ED)/[(BW)(ATNC)]
where,
LADD = lifetime average daily dose (mg/kg-day) (for carcinogens)
ADD = average daily dose (mg/kg-day) (for non-carcinogens)
I = daily inhalation potential dose rate (mg/day)
EF = exposure frequency (days/year)
ED = exposure duration (years)
BW = body weight (kg)
ATCAR = averaging time for carcinogenic effects (days)
ATNC = averaging time for non-carcinogenic effects (days)
7 Different averaging times are used for characterizing risk for carcinogenic and non-carcinogenic effects. For
carcinogenic agents, because even a single incidence of exposure is assumed to have the potential to cause cancer
throughout an individual's lifetime, the length of exposure to that agent is averaged over a lifetime. An additional
factor is that the cancer latency period may extend beyond the period of working years before it is discernible. For
chemicals exhibiting non-cancer health effects from chronic (longer-term) exposure, where there is an exposure
threshold (a level below which effects are not expected to occur); only the time period when exposure is occurring is
assumed to be relevant and is used as the averaging time.
3-56
-------
Parameter values for estimating workers' potential dose rates from inhalation are
presented in Table 3-11.
Table 3-11. Parameter Values for Workplace Inhalation Exposures
Parameter
Air Concentration
(Ca)
Inhalation Rate (IR)
Units
mg/m3
m3/hr
Value
Source of Data, Comments
Modeled from bath concentrations (see Table 3-9).
1.25
U.S. EPA, 1991a (data from
NIOSH, 1976).
Exposure Time (ET)
Line Operation
Working in Process
Area
hrs/day
hrs/day
8
laboratory technician 2.8
maintenance worker 1.6
supervisors 5 5
wastewater treatment operator . . 1
other employee 9
Default value for occupational
exposure.
90th percentile of hours/week
reported from PWB Workplace
Practices Questionnaire, assuming
a 5 -day work week.
Exposure Frequency (EF)
Line Operation &
Working in Process
Area
Exposure Duration
(ED)
Body Weight (BW)
Averaging Time (AT)
ATCAR
ATNC
days/yr
years
kg
days
HASL (NC) 44
HASL (C) 22
Nickel/Gold (NC) 212
Nickel/Palladium/Gold (NC) . 280
OSP (NC) 35
OSP (C) 16
Immersion Silver (C) 64
Immersion Tin (NC) 75
Immersion Tin (C) 107
25
70
25,550
9,125
From process cost model, based on
the number of days per year
required to produce 260,000 ssf of
finished PWB. Assumed this is the
time worked per year.
95th percentile for job tenure
(Bureau of Labor Statistics, 1990).
(Median tenure for U.S. males is 4
years; Bureau of Labor Statistics,
1997.)
Average for adults (U.S. EPA,
1991b).
70 yrs (average lifetime) x 365 d/yr
25 yrs (ED) x 365 d/yr
Workplace Dermal Exposures
Two approaches were considered for evaluating dermal exposure. The general model for
potential dose rate via dermal exposure to workers from the CEB Manual (U.S. EPA, 199la) is as
follows:
D = SQC
3-57
-------
where,
D = dermal potential dose rate (mg/day)
S = surface area of contact (cm2)
Q = quantity typically remaining on skin (mg/cm2)
C = concentration of chemical (percent)
Because a line operator is expected to have dermal contact with the chemicals in a given
bath several times a day in the course of normal operations, the total time of contact combined
with a flux rate (rate of chemical absorption through the skin) is believed to give a more realistic
estimate of dermal exposure. An equation based on flux of material through the skin (from on
U.S. EPA, 1992a), is as follows:
D = (S)(C)(f)(ET)(0.001)
where,
D = dermal potential dose rate (mg/day)
S = skin surface area of contact (cm2)
C = chemical concentration (mg/L)
f = flux through skin (cm/hour)
ET = exposure time (hours/day)
with a conversion factor of 0.001 L/cm3
This second equation was used for all workplace dermal exposure estimates.8
As indicated earlier, daily exposures are averaged over a lifetime (70 years) for
carcinogens, and over the exposure duration (e.g., 25 years working in a facility) for non-
carcinogens. The following equations are used to estimate average daily doses from dermal
contact:
LADD = (D)(EF)(ED)/[(BW)(ATCAR)]
ADD = (D)(EF)(ED)/[(BW)(ATNC)]
where,
D = dermal potential dose rate (mg/day)
General parameter values for estimating workers' potential dose rates from dermal
exposure are presented in Table 3-12.
8 This permeability coefficient-based approach is recommended over the absorption fraction approach for
compounds in an aqueous media or in air (U.S. EPA 1992a).
3-58
-------
Table 3-12. General Parameter Values for Workplace Dermal Exposures
Parameter
Chemical
Concentration (C)
Skin Surface Area
(S)
Flux Through Skin
(0
Exposure Duration
(ED)
Body Weight (BW)
Averaging Time (AT)
ATCAR
ATNC
Units
%
cm2
cm/hr
years
kg
days
Value
Source of Data, Comments
Range of reported values and average determined from bath chemistry
data.
800
Default for inorganics: 0.001 estimate
for organics by: log f = -2.72+0.71
log K™ - 0.0061(MW)
(KoW = octanol/water partition
coefficient, MW = molecular weight)
25
70
25,500
9,125
CEB, routine immersion, 2 hands,
assuming gloves not worn.
U.S. EPA, 1992a.
95th percentile for job tenure
(Bureau of Labor Statistics, 1990).
(Median tenure for U.S. males is 4
years; Bureau of Labor Statistics,
1997.)
U.S. EPA, 1991b.
70 yrs (average lifetime) x 365 d/yr
25 yrs (ED) x 365 d/yr
Dermal exposure was quantified for line operators performing routine line operation
activities on non-conveyorized lines. Parameter values used in the dermal exposure equations are
provided in Table 3-13.
3-59
-------
Table 3-13. Parameter Values for Workplace Dermal Exposures for Line Operators
on Non-Conveyorized Lines
Parameter/
Activity a
Units
Value
Source of Data, Comments
Exposure Time (ET)
Line Operation z
irs/day
Process / no. baths or steps
HASL (NC) / 10
Nickel/Gold (NC) / 14
Nickel/Palladium/Gold (NC) /
22
OSP (NC) / 9
Immersion Tin (NC) 712
Value
0.80
0.57
0.36
0.89
0.67
Based on a default value of 8
irs/day; corrected for typical
number of baths in a process,
including rinse baths, by dividing 8
irs/day by the number of baths
and/or steps in a typical process
ine.
Exposure Frequency (EF)
Line Operation "
days/yr
HASL (NC) 44
HASL (C) 22
Nickel/Gold (NC) 212
Nickel/Palladium/Gold (NC) 280
OSP (NC) 35
OSP (C) 16
Immersion Silver (C) 64
Immersion Tin (NC) 75
Immersion Tin (C} 107
From cost process simulation
model, based on a throughput of
260,000 ssf
Dermal exposure on non-conveyorized lines was quantified for line operation activities only, because this would
result in higher line operator exposure than any other activities that may be performed (e.g., bath sampling, filter
replacement).
Dermal exposure was quantified for line operators on conveyorized lines for chemical
bath replacement, conveyor equipment cleaning, filter replacement, and bath sampling activities.
Because convey orized lines are enclosed and the boards are moved through the line
automatically, it was assumed that dermal exposure from line operation would be negligible.
Parameter values used in the exposure equations for convey orized lines are provided in Table 3-
14.
3-60
-------
Table 3-14. Parameter Values for Workplace Dermal Exposure for Line Operators on
Conveyorized Lines
Parameter/
Activity a
Units b
Value
Source of Data, Comments
Exposure Time (ET)
Chemical Bath
Replacement
Filter
Replacement
Chemical Bath
Sampling
min/occur
min/occur
min/occur
HASL 264
OSP 190
Immersion Silver 198
Immersion Tin 120
15
HASL 15
OSP 22
Immersion Silver 1 0
Immersion Tin 5 0
90th percentile from survey.
Questionnaire data for
replacement duration were
combined regardless of
process configuration
90th percentile from PWB
Workplace Practices
Questionnaire, combined for
all process types.
90th percentile from PWB
Workplace Practices
Questionnaire. Questionnaire
data for sampling duration
were combined regardless of
process configuration.
Exposure Frequency (EF)
Chemical Bath
Replacement
Filter
Replacement
Chemical Bath
Sampling
occur/year
occur/year
occur/year
HASL cleaner 6
HASL microetch 6
OSP cleaner 6
OSP microetch 6
OSP OSP bath 1
Immersion Silver, cleaner & microetch . 6
Immersion Silver predip 5
Immersion Silver, imm. silver bath .... 1
Immersion Tin, cleaner & microetch ... 6
Immersion Tin predip 5
Immersion Tin, imm. tin bath 1
HASL 28
OSP 9
Immersion Silver 4
Immersion Tin 57
HASL 67
OSP 200
Immersion Silver 253
Immersion Tin 485
From cost process simulation
model, based on a throughput
of 260,000 ssf
From cost process simulation
model, based on a throughput
of 260,000 ssf.
From cost process simulation
model, based on a throughput
of 260,000 ssf.
3-61
-------
Parameter/
Activity a
Units b
Value
Source of Data, Comments
Exposure Frequency and Time combined (EF x ET)
Conveyor
Equipment
Cleaning
min/year
10,488
90th percentile of total
duration per year from PWB
Workplace Practices
Questionnaire for
conveyorized lines. Because a
correlation between EF and ET
was apparent, we did not take
the 90th percentile of each term
separately.
Dermal exposure on conveyorized lines is quantified for specific routine activities other than line operation because
on an enclosed, conveyorized line it is assumed that dermal contact from line operation would be negligible.
b min/occur = minutes per occurance; occur/year = number of occurances per year.
Dermal exposure was also quantified for a laboratory technician on all conveyorized and
non-conveyorized lines for chemical bath sampling activities. Parameter values used in the
exposure equations for a laboratory technician are provided in Table 3-15.
Table 3-15. Parameter Values for Workplace Dermal Exposure for a Laboratory
Technician on Either Conveyorized or Non-Conveyorized Lines
Parameter/
Activity
Units a
Value
Source of Data, Comments
Exposure Time (ET)
Chemical Bath
Sampling
min/occur
HASL 15
Nickel/Gold 10
Nickel/Palladium/Gold 1.5
OSP 22
Immersion Silver 1 0
Immersion Tin 5 0
Questionnaire data for
sampling duration were
combined regardless of
process configuration.
Exposure Frequency (EF)
Chemical Bath
Sampling
occur/year
HASL (NC) 135
HASL (C) 67
Nickel/Gold (NC) 954
Nickel/Palladium/Gold (NC) . . . 2,406
OSP (NC) 436
OSP (C) 200
Immersion Silver (C) 253
Immersion Tin (NC) 341
Immersion Tin (C} 485
From cost process simulation
model, based on a throughput
of 260,000 ssf
min/occur = minutes per occurance; occur/year = number of occurances per year.
3-62
-------
Results
Table 3-16 presents results for estimating ADDs for inhalation and dermal workplace
exposure for line operators and laboratory technicians.
Table 3-16. Estimated Average Daily Dose for Workplace Exposure From Inhalation
and Dermal Contact
Chemical
ADD3
(mg/kg-day)
Inhalation
Line
Operator
Dermal
Line
Operator
Laboratory
Technician
HASL, Non-conveyorized
1,4-Butenediol
Alkylalkyne diol
Alkylaryl sulfonate
Alkylphenol ethoxylate
Alkylphenolpolyethoxyethanol
Aryl phenol
Citric acid
Copper Sulfate Pentahydrate
Ethoxylated alkylphenol A
Ethoxylated alkylphenol B
Ethylene glycol
Ethylene glycol monobutyl ether
Fluoboric acid
Gum
Hydrochloric acid
Hydrogen peroxide
Hydroxyaryl acid
Hydroxyaryl sulfonate
Phosphoric acid
Potassium peroxymonosulfate
Sodium benzene sulfonate
Sodium hydroxide
Sulfuric acid
1.28E-02
NA
2.43E-05
NA
NA
7.86E-06
NA
NA
NA
NA
1.60E-02
1.53E-01
NA
NA
1.16E-03
6.81E-03
NA
NA
2.01E-03
NA
NA
NA
NA
2.05E-03
1.31E-05
5.50E-07
1.59E-27
1.50E-26
1.98E-03
4.25E-03
4.93E-02
1.26E-27
8.97E-28
5.17E-03
3.53E-02
1.35E-02
NAb
2.28E-02
5.55E-02
9.52E-04
3.35E-05
6.69E-02
1.11E-01
1.85E-07
1.86E-04
2.34E-01
2.82E-05
1.81E-07
7.58E-09
2.18E-29
2.06E-28
2.73E-05
5.85E-05
6.79E-04
1.73E-29
1.24E-29
7.13E-05
4.86E-04
1.86E-04
NAb
3.15E-04
7.66E-04
1.31E-05
4.62E-07
9.22E-04
1.53E-03
2.55E-09
2.57E-06
3.23E-03
HASL, Conveyorized
1,4-Butenediol
Alkylalkyne diol
Alkylaryl sulfonate
Alkylphenol ethoxylate
Alkylphenolpolyethoxyethanol
NA
NA
NA
NA
NA
8.53E-05
5.47E-07
2.29E-08
6.61E-29
6.23E-28
6.35E-06
4.07E-08
1.71E-09
4.92E-30
4.64E-29
3-63
-------
Chemical
Aryl phenol
Citric acid
Copper sulfate pentahydrate
Ethoxylated alkylphenol A
Ethoxylated alkylphenol B
Ethylene glycol
Ethylene glycol monobutyl ether
Fluoboric acid
Gum
Hydrochloric acid
Hydrogen peroxide
Hydroxyaryl acid
Hydroxyaryl sulfonate
Phosphoric acid
Potassium peroxymonosulfate
Sodium benzene sulfonate
Sodium hydroxide
Sulfuric acid
ADD3
(mg/kg-day)
Inhalation
Line
Operator
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Dermal
Line
Operator
8.26E-05
1.77E-04
2.05E-03
5.24E-29
3.74E-29
2.15E-04
1.47E-03
5.62E-04
NAb
9.51E-04
2.31E-03
3.97E-05
1.40E-06
2.79E-03
4.64E-03
7.72E-09
7.75E-06
9.76E-03
Laboratory
Technician
6.15E-06
1.32E-05
1.53E-04
3.90E-30
2.78E-30
1.60E-05
1.09E-04
4.19E-05
NAb
7.08E-05
1.72E-04
2.95E-06
1.04E-07
2.08E-04
3.45E-04
5.75E-10
5.77E-07
7.27E-04
Nickel/Gold, Non-conveyorized
Aliphatic acid A
Aliphatic acid B
Aliphatic acid E
Aliphatic dicarboxylic acid A
Aliphatic dicarboxylic acid C
Alkylamino acid B
Alkyldiol
Alkylphenolpolyethoxyethanol
Ammonia compound B
Ammonium chloride
Ammonium hydroxide
Citric acid
Copper sulfate pentahydrate
Ethoxylated akylphenol B
Hydrochloric acid
Hydrogen peroxide
Hydroxyaryl acid
4.86E-01
3.38E-06
6.43E-01
6.59E-04
3.12E-04
NA
1.37E-01
NA
1.61E-04
NA
7.76E-03
NA
NA
NA
1.63E-01
2.40E-02
NA
2.35E-02
1.56E-03
1.41E-02
4.94E-03
1.75E-03
5.38E-06
1.66E-02
5.18E-26
2.65E-04
2.08E-01
1.34E-01
4.79E-03
1.71E-01
3.11E-27
2.08E+00
1.36E-01
3.30E-03
1.53E-03
1.02E-04
9.16E-04
3.21E-04
1.13E-04
3.49E-07
1.08E-03
3.36E-27
1.72E-05
1.35E-02
8.71E-03
3.11E-04
1.11E-02
2.02E-28
1.35E-01
8.84E-03
2.14E-04
3-64
-------
Chemical
Inorganic metallic salt A
Inorganic metallic salt A (LADD) °
Inorganic metallic salt B
Inorganic metallic salt C
Malic acid
Nickel sulfate
Palladium chloride
Phosphoric acid
Potassium compound
Potassium gold cyanide
Potassium peroxymonosulfate
Sodium salt
Sodium hydroxide
Sodium hypophosphite
Substituted amine hydrochloride
Sulfuric acid
Transition metal salt
Urea compound B
ADD3
(mg/kg-day)
Inhalation
Line
Operator
1.97E-07
7.04E-08
1.31E-05
1.37E-07
1.41E-03
3.49E-03
NA
7.67E-03
6.59E-03
NA
NA
NA
NA
4.02E-03
NA
NA
NA
4.80E-06
Dermal
Line
Operator
8.00E-06
2.85E-06
5.31E-04
5.55E-06
2.10E-03
1.41E-01
5.01E-03
1.93E-01
2.66E-01
1.14E-02
NAd
3.41E-01
6.45E-04
1.62E-01
2.27E-01
8.55E-01
2.27E-03
2.40E-05
Laboratory
Technician
5.19E-07
1.85E-07
3.45E-05
3.61E-07
1.37E-04
9.17E-03
3.25E-04
1.25E-02
1.73E-02
7.39E-04
NAd
2.22E-02
4.19E-05
1.05E-02
1.48E-02
5.55E-02
1.48E-04
1.56E-06
Nickel/Palladium/Gold, Non-conveyorized
Aliphatic acid B
Aliphatic acid E
Aliphatic dicarboxylic acid A
Aliphatic dicarboxylic acid C
Alkylamino acid B
Alkyldiol
Alkylpolyol
Amino acid salt
Amino carboxylic acid
Ammonia compound A
Ammonia compound B
Ammonium hydroxide
Citric acid
Copper sulfate pentahydrate
Ethoxylated alkylphenol
4.63E-06
1.17E+00
8.98E-04
4.26E-04
NA
1.85E-01
NA
NA
NA
NA
2.20E-04
1.71E-02
NA
NA
NA
1.32E-03
1.58E-02
4.16E-03
1.47E-03
8.01E-06
1.40E-02
3.56-03
6.39E-04
1.11E-05
1.60E-01
2.23E-04
1.91E-01
4.91E-03
1.43E-01
2.61E-27
3.23E-05
3.88E-04
1.02E-04
3.61E-05
1.97E-07
3.43E-04
8.76E-05
1.57E-05
2.73E-07
3.92E-03
5.48E-06
4.70E-03
1.21E-04
3.53E-03
6.42E-29
3-65
-------
Chemical
Ethylenediamine
Hydrochloric acid
Hydrogen peroxide
Hydroxyaryl acid
Inorganic metallic salt B
Maleic acid
Malic acid
Nickel sulfate
Palladium salt
Phosphoric acid
Potassium compound
Potassium gold cyanide
Propionic acid
Sodium hydroxide
Sodium hypophosphite
Sodium salt
Substituted amine hydrochloride
Sulfuric acid
Surfactant
Transition metal salt
Urea compound B
ADD3
(mg/kg-day)
Inhalation
Line
Operator
5.32E-04
2.35E-01
3.11E-02
NA
1.79E-05
NA
1.92E-03
7.50E-03
NA
1.01E-02
8.98E-03
NA
2.13E-01
NA
7.11E-03
NA
NA
NA
NA
NA
1.28E-05
Dermal
Line
Operator
4.14E-04
3.92E-01
1.14E-01
2.77E-03
2.07E-03
1.36E-03
1.77E-03
1.87E-01
1.02E-02
1.62E-01
2.24E-01
9.56E-03
2.69E-02
5.42E-04
1.93E-01
4.78E-01
1.91E-01
4.99E-01
3.19E-04
1.91E-03
3.94E-05
Laboratory
Technician
1.02E-05
9.63E-03
2.81E-03
6.81E-05
5.08E-05
3.35E-05
4.34E-05
4.59E-03
2.51E-04
3.98E-03
5.50E-03
2.35E-04
6.60E-04
1.33E-05
4.75E-03
1.18E-02
4.70E-03
1.23E-02
7.83E-06
4.70E-05
9.67E-07
OSP, Non-conveyorized
Acetic acid
Alkylaryl imidazole
Aromatic imidizole product
Arylphenol
Copper ion
Copper salt C
Copper sulfate pentahydrate
Ethoxylated alkylphenol
Ethylene glycol
Gum
Hydrochloric acid
Hydrogen peroxide
7.79E-02
NA
NA
6.18E-06
NA
NA
NA
NA
2.38E-02
NA
2.04E-03
1.92E-03
3.75E-02
5.50E+00
6.33E-03
1.77E-03
4.95E-02
1.36E-03
4.41E-02
8.03E-28
4.63E-03
NAb
2.33E-02
1.78E-02
2.45E-03
3.59E-01
4.13E-04
1.16E-04
3.23E-03
8.89E-05
2.88E-03
5.24E-29
3.02E-04
NAb
1.52E-03
1.16E-03
3-66
-------
Chemical
Hydroxyaryl acid
Hydroxyaryl sulfonate
Phosphoric acid
Sodium hydroxide
Sulfuric acid
ADD3
(mg/kg-day)
Inhalation
Line
Operator
NA
NA
1.27E-03
NA
NA
Dermal
Line
Operator
8.52E-04
3.00E-05
4.98E-02
1.67E-04
2.55E-01
Laboratory
Technician
5.57E-05
1.96E-06
3.25E-03
1.09E-05
1.66E-02
OSP, Conveyorized
Acetic acid
Alkylaryl imidazole
Aromatic imidizole product
Aryl phenol
Copper ion
Copper salt C
Copper sulfate pentahydrate
Ethoxylated alkylphenol
Ethylene glycol
Gum
Hydrochloric acid
Hydrogen peroxide
Hydroxyaryl acid
Hydroxyaryl sulfonate
Phosphoric acid
Sodium hydroxide
Sulfuric acid
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
1.78E-03
2.61E-01
3.00E-04
8.93E-05
2.34E-03
6.45E-05
2.22E-03
4.04E-29
2.33E-04
NAb
1.17E-03
8.96E-04
4.29E-05
1.51E-06
2.50E-03
8.38E-06
1.28E-02
5.30E-04
7.78E-02
8.94E-05
2.51E-05
6.99E-04
1.92E-05
6.23E-04
1.13E-29
6.54E-05
NAb
3.30E-04
2.51E-04
1.20E-05
4.24E-07
7.03E-04
2.35E-06
3.60E-03
Immersion Silver, Conveyorized
1,4-Butenediol
Alkylamino acid A
Fatty amine
Hydrogen peroxide
Nitrogen acid
Nonionic surfactant
Phosphoric acid
Silver nitrate
Sodium hydroxide
Sulfuric acid
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
3.07E-04
1.71E-04
5.75E-01
1.85E-02
3.95E-03
9.23E-03
2.02E-02
1.51E-04
8.72E-03
7.55E-04
6.48E-06
3.79E-06
1.28E-02
3.91E-04
8.75E-05
2.04E-04
4.26E-04
3.48E-06
1.93E-04
1.59E-05
3-67
-------
Chemical
ADD3
(mg/kg-day)
Inhalation
Line
Operator
Dermal
Line
Operator
Laboratory
Technician
Immersion Tin, Non-conveyorized
Aliphatic acid D
Alkylalkyne diol
Alkylamino acid B
Alkylaryl sulfonate
Alkylimine dialkanol
Alkylphenol ethoxylate
Bismuth compound
Citric acid
Cyclic amide
Ethoxylated alkylphenol
Ethylene glycol monobutyl ether
Fluoboric acid
Hydrochloric acid
Hydroxy carboxylic acid
Methane sulfonic acid
Phosphoric acid
Potassium peroxymonosulfate
Quantenary alkylammonium chlorides
Silver salt
Sodium benzene sulfonate
Sodium phosphorus salt
Stannous methane sulfonic acid
Sulfuric acid
Thiourea
Tin chloride
Unspecified tartrate
Urea
Urea compound C
Vinyl polymer
6.14E-02
NA
NA
5.74E-05
NA
NA
NA
NA
4.90E-02
NA
3.75E-01
NA
2.03E-03
8.26E-02
NA
1.66E-03
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
5.55E-01
NA
8.22E-03
1.88E-05
1.79E-06
7.88E-07
1.84E-05
2.27E-27
4.02E-05
7.65E-02
1.15E-02
1.80E-27
5.06E-02
1.94E-02
1.13E-02
7.03E-03
1.62E+00
4.75E-02
1.60E-01
7.60E-04
6.03E-06
2.66E-07
1.41E-01
2.18E-02
4.62E-01
1.89E-02
2.19E-02
1.77E-03
3.68E-03
2.37E-02
1.81E-32
9.54E-05
2.19E-07
2.08E-08
9.15E-09
2.13E-07
2.64E-29
4.66E-07
8.88E-04
1.34E-04
2.09E-29
5.87E-04
2.25E-04
1.31E-04
8.16E-05
1.88E-02
5.51E-04
1.85E-03
8.83E-06
7.00E-08
3.08E-09
1.64E-03
2.53E-04
5.37E-03
2.20E-04
2.55E-04
2.06E-05
4.27E-05
2.75E-04
2.10E-34
Immersion Tin, Conveyorized
Aliphatic acid D
Alkylalkyne diol
Alkylamino acid B
Alkylaryl sulfonate
NA
NA
NA
NA
1.33E-03
3.17E-06
2.89E-07
1.33E-07
2.32E-04
5.31E-07
5.05E-08
2.22E-08
3-68
-------
Chemical
Alkylimine dialkanol
Alkylphenol ethoxylate
Bismuth compound
Citric acid
Cyclic amide
Ethoxylated alkylphenol
Ethylene glycol monobutyl ether
Fluoboric acid
Hydrochloric acid
Hydroxy carboxylic acid
Methane sulfonic acid
Phosphoric acid
Potassium peroxymonosulfate
Quantenary alkylammonium chlorides
Silver salt
Sodium benzene sulfonate
Sodium phosphorus salt
Stannous methane sulfonic acid
Sulfuric acid
Thiourea
Tin chloride
Unspecified tartrate
Urea
Urea compound C
Vinyl polymer
ADD3
(mg/kg-day)
Inhalation
Line
Operator
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
Dermal
Line
Operator
2.98E-06
3.83E-28
6.50E-06
1.24E-02
1.87E-03
3.04E-28
8.52E-03
3.26E-03
1.82E-03
1.14E-03
2.69E-01
8.00E-03
2.69E-02
1.23E-04
9.75E-07
4.48E-08
2.33E-02
3.52E-03
7.69E-02
3.05E-03
3.54E-03
2.86E-04
5.94E-04
3.82E-03
2.92E-33
Laboratory
Technician
5.17E-07
6.41E-29
1.13E-06
2.16E-03
3.25E-04
5.08E-29
1.43E-03
5.46E-04
3.18E-04
1.98E-04
4.56E-02
1.34E-03
4.50E-03
2.14E-05
1.70E-07
7.49E-09
3.98E-03
6.14E-04
1.30E-02
5.33E-04
6.19E-04
4.99E-05
1.04E-04
6.88E-04
5.09E-34
a Average Daily Dose (ADD) unless otherwise noted.
b Dermal absorption not expected due to large molecular size.
0 LADD is used for calculating cancer risk, and is calculated using a carcinogen averaging time (ATCAR) of 70 years.
Note: The numeric format used in these tables is a form of scientific notation, where "E" replaces the
" x 10*'. Scientific notation is typically used to present very large or very small numbers. For example, 1.2E-04 is
the same as 1.2 x 10~4, which is the same as 0.00012 in common decimal notation.
d Bath concentration not available.
NA: Not Applicable. Unless otherwise noted, a number was not calculated because the chemical's vapor pressure is
below the 1 x 10~3 torr cutoff and it is not used in any sparged bath. Inhalation exposures are therefore expected to be
negligible.
ND: Not determined because a required value was not available.
3-69
-------
Occupational Exposure to Elemental Lead
Modeling Occupational Lead Exposure. We estimated occupational exposure to lead
based on EPA guidelines for lead ingestion in soil (U.S. EPA, 1996a). This includes modeling
worker blood-lead levels using the following equation:
(Pbs)(BKSF)(IR)(AF)(EFs)^AT
where,
PbBadulti central = central estimate of adult blood-lead concentrations (//g/dl)
PbBadult, o = typical background adult blood-lead concentration (//g/dl)
Pbs = lead concentration (//g/g)
BKSF = biokinetic slope factor (//g/dl)
IRS = intake rate (g/day)
AFS = gastrointestinal absorption factor (unitless fraction)
EFS = exposure frequency (days/year)
AT = averaging time (days/year)
Lead can be easily passed along to an unborn fetus via the placenta. Using the EPA guidelines
(U.S. EPA, 1996a), we also estimated fetal blood-lead levels, assuming a pregnant worker, by:
PbBfetal o 95 = PbBadult central X GSD; adult X Rfetal/matemal
where,
PbBfetal 095 = 95 percent estimate of fetal blood-lead levels (//g/dl)
PbBadultiCentral = central estimate of adult blood-lead concentrations (//g/dl)
GSD; addt = estimated value of the individual geometric standard deviation
(dimensionless)
Rfetai/matemai = fetal/maternal lead concentration at birth (dimensionless)
These equations were developed for exposure to lead in soil and dust, and were modified
for the surface finishing situation by considering lead in solder, rather than soil. Our treatment of
each term in the model is discussed below.
Estimated Adult Blood-Lead Concentration (PbBadU]t centrai). This represents the central
estimate of blood-lead in adults exposed to the HASL process, measured in//g/dl.
Background Blood-Lead Concentration (PbBadun „)• This value represents the typical
blood-lead concentration of adults who are not exposed to lead at the site that is being assessed,
and is measured in //g/dl. A value of 1.95 is used, based on a typical range of 1.7 - 2.2 (//g/dl)
(U.S. EPA, 1996a).
3-70
-------
Lead Concentration in Source (Pbs). This is an average estimate of the amount of lead
that is present in solder, and is measured in //g/g. For PWB facilities, the lead concentration of
solder was used instead of soil lead concentration. A value of 37,000 //g/g (37 percent) was used,
based on typical proportion of lead in tin/lead solder.
Biokinetic Slope Factor. The biokinetic slope factor (BKSF) relates the increase of
typical adult blood-lead concentrations to the average daily lead uptake. The recommended
default value is 0.4 //g Pb/dl blood per //g Pb absorbed/day. This value is derived from Pocock et
al. (1983) and Sherlock et al. (1984) as cited by the U.S. EPA (1996a). (Both studies involved the
amount of lead in tap water, and both predict higher blood-lead concentrations than expected in
today's average U.S. population.)
Intake Rate. The use of this model is based on the assumption that solder could adhere
to a workers' hands from routine handling, and be subsequently ingested. Although no studies
were found that address the amount of lead that might be ingested by a worker handling solder
specifically for a HASL process, Monsalve (1984) investigated hand soldering and pot tinning
operations. Based on surface wipe samples and samples from worker's hands, a "conservative
overestimate" of 30 //g Pb per day ingested was calculated.9 In addition to this intake rate (IRs),
two values based on soil ingestion studies were used in the model: an average soil ingestion rate
for adults, based on tracer studies, of 10 mg (Stanek et al., 1997) and the adult central estimate for
soil ingestion of 50 mg from the EPA's Exposure Factors Handbook (U.S. EPA, 1997a).
Gastrointestinal Absorption Factor. The gastrointestinal absorption factor (AFS)
represents the absolute gastrointestinal absorption fraction for ingested lead in soil. This value
was determined by multiplying the absorption factor for soluble lead by the bioavailiability of
lead in soil. Three factors that were considered in determining this value are the variability of
food intake, lead intake, and lead form/particle size (U.S. EPA, 1996a). The soil value of 0.12 is
used due to the lack of information on the absorption of ingested metallic lead from tin-lead
solder.
Exposure Frequency. This represents the exposure frequency (EFS) to lead solder for a
worker in a PWB manufacturing facility. This is the number of days that a worker is exposed to
lead and is determined in days/year. The exposure frequency was increased from EPA's value of
219 (U.S. EPA, 1996a) to 250 days/year as a standard default value for occupational exposure.
Averaging Time. The averaging time (AT) is the total period of time that lead contact
may occur. We used one year, or 365 days, as the AT.
Estimated Fetal/Maternal Blood Lead Concentration (PbBfetai 0.95). This represents the
95th percentile estimate of fetal/maternal blood-lead, and is measured in //g/dl. These results are
also based on the intake rate, as discussed above.
9 Wipe samples from surfaces in the area ranged from 13 to 92 ,wg Pb per 100 cm2, and samples from solderer's
hands ranged from 3 to 32 ^g Pb per 100 cm2.
3-71
-------
Individual Blood Lead Geometric Standard Deviation (GSDj). The GSD; is used to
measure the inter-individual variability of blood-lead concentrations in a population whose
members are exposed to the same non-residential environmental lead levels. A value of 1.8 is
recommend for homogeneous populations and 2.1 for heterogeneous populations. The values for
GSD; are estimated in the population of concern. If this is not possible, the GSD; is estimated
using a surrogate population. Factors used to estimate the GSD; are variability in exposure,
biokinetics, socioeconomic/ethnic characteristics, degree of urbanization, and geographical
location. Using these factors can cause a high degree of uncertainty (U.S. EPA, 1996a).
Fetal/Maternal Blood Lead Concentration Ratio (Rfetai/maternai)- The Rfetai/matemai describes
the relationship between the umbilical cord and the maternal blood-lead concentration. The U.S.
EPA Technical Working Group for Lead recommends a default value of 0.9 (dimensionless).
This is based on two separate studies: one conducted by Goyer (1990) and the other by Graziano
et al. (1990). This value was derived by comparing the fetal/maternal blood-lead concentrations
at delivery. The 0.9 fetal/maternal blood-lead concentration can change due to physiological
changes that include the mobilization of bone/lead stores during pregnancy, and iron and calcium
deficiency due to poor nutrition (U.S. EPA, 1990; Franklin et al., 1995). The blood-lead
concentration also can decrease in the later stages of pregnancy due to an increase in plasma
volume, which dilutes the concentration, and an increased rate of transfer of lead to the placenta
or to fetal tissue (Alexander and Delves, 1981).
Modeling Results. According to the results of the blood-lead solder model, incidental
ingestion could result in blood-lead concentration for workers of 2 to 63 //g/dL, and of 3.2 to 102
for a fetus (Table 3-17). Estimated blood-lead levels will be compared to federal health-based
standards and guidlines in Section 3.4.
Table 3-17. Estimated Concentration of Lead in Adult and Fetal Blood from Incidental
Ingestion of Lead in Tin/Lead Solder
Intake Rate
(mg/day)
0.03
10
50
Ingestion Rate source, notes
"Conservative overestimate" based on surface wipe
samples in hand-soldering operations (Monsalve, 1984).
Average soil ingestion rate for adults, based on tracer
studies (Stanek et al., 1997).
Adult central estimate for soil ingestion (U.S. EPA,
1997a).
"bB adult, central
(^g/dl)
2.0
14
63
PkBfetal, 0 95
(Mg/dl)
3.2
23
102
= 1.95/ug/dl;PbS= 37,000 ,wg/g; BKSF = 0.4,wg/dL; AFS = 0.12;EFS=250 days/yr; AT= 365 days/yr;
= 1 8- anH R = 0 Q
1. o, clllU -tvfetal/matemal ^ •'
The intake rate is a major source of uncertainty in estimating exposure to workers from
handling solder. A range of intake rates were used to provide a possible range of modeled blood-
lead concentrations. These values provide bounding estimates only. It is expected that a smaller,
but unknown, amount of solder would be ingested from a workers hands than the estimates that
have been used here. Figure 3-9 shows the relationship between intake rate and blood-lead level
for both an adult and fetus.
3-72
-------
2T
1
d
o
o
•o
re
•a
o
o
m
Blood lead concentration vs intake rate
120 -i
100 -
on
60 -
A n
B
/
.s
^r j^
.S ..---'
41) .x _*•*
M^" A* •
LxjB*"
o r
•
1 1
— • — fetal
0 20 40 60
Intake rate (mg/day)
Figure 3-9. Relationship Between Intake Rate and Blood-Lead Level for Both
Adult and Fetus
Monitoring Data. Lead monitoring data for HASL line operators were made available by one
PWB manufacturer. For seven line operators monitored from 1986 to 1998, blood-lead levels
ranged from 5 to 12 |ig/dL.
Population Exposure
The equation for estimating ADDs from inhalation for a person residing near a facility is:
LADD = (Ca)(IR)(EF)(ED)/[(BW)(ATCAR)]
ADD = (Ca) (TR) (EF) (ED)/[(BW) (ATNC)]
where,
LADD
ADD
Ca
IR
EF
ED
BW
A IMP
lifetime average daily dose (mg/kg-day) (for carcinogens)
average daily dose (mg/kg-day) (for non-carcinogens)
chemical concentration in air (mg/m3) (from air dispersion modeling, described in
Section 3.2.3)
inhalation rate (m3/day)
exposure frequency (day/yr)
exposure duration (years)
average body weight (kg)
averaging time for carcinogenic effects (days)
averaging time for non-carcinogenic chronic effects (days)
Table 3-18 presents values used for these parameters. Results for general population inhalation
exposure are presented in Table 3-19.
3-73
-------
Table 3-18. Parameter Values for Estimating Nearby Residential Inhalation Exposure
Parameter
Air Concentration (Ca)
Inhalation Rate (IR)
Exposure Frequency
(EF)
Exposure Duration (ED)
Body Weight (BW)
Averaging Time (AT)
ATCAR
ATNC
Units
mg/m3
m3/day
days/yr
years
kg
days
Value
Source of Data, Comments
Modeled, varies by chemical and process type.
15
350
30
70
25,550
10,950
Total home exposures for adults based on activity patterns
and inhalation rates (U.S. EPA, 1997a).
Assumes 2 wks per year spent away from home (U.S. EPA,
1991b).
National upper 90th percentile at one residence (U.S. EPA,
1990).
Average value for adults (U.S. EPA, 1991b).
70 yrs x 365 days/year
ED x 365 days/year
Table 3-19. Estimated Average Daily Dose for General Population Inhalation Exposure
Chemical a
ADD (mg/kg-day) b
HASL, Non-conveyorized
Ethylene glycol monobutyl ether
HASL, Conveyorized
Ethylene glycol monobutyl ether
5.25E-05
1.04E-04
Nickel/Gold, Non-conveyorized
Aliphatic acid A
Aliphatic acid E
Inorganic metallic salt A (LADD)
3.45E-05
4.56E-05
5.99E-12
Nickel/Palladium/Gold, Non-conveyorized
Aliphatic acid E
OSP, Non-conveyorized
Acetic acid
OSP, Conveyorized
Acetic acid
Ethylene glycol
Immersion Tin, Non-conveyorized
Urea compound C
Immersion Tin, Conveyorized
Aliphatic acid D
Cyclic amide
Hydroxy carboxylic acid
Urea compound C
6.29E-05
3.33E-05
1.26E-04
2.04E-05
1.11E-04
2.99E-05
2.39E-05
4.03E-05
2.72E-04
a Only inorganic metallic salt A plus those chemicals with an emission rate of at least 23 kg/year (44 mg/min) are
listed (see Table 3-9). Immersion silver had no modeled emission rates above this cut-off.
b Unless otherwise noted.
Note: The numeric format used in this table is a form of scientific notation, where "E" replaces the " x KF'.
Scientific notation is typically used to present very large or very small numbers. For example, 1.2E-04 is the same as
1.2 x 10~4, which is the same as 0.00012 in common decimal notation.
3-74
-------
For lead, we did not calculate an ADD. The recommended approach for evaluating lead
exposure to nearby residents is to apply an EPA model, the Integrated Exposure Uptake
Biokinetic (TEUBK) Model for Lead in Children (U.S. EPA, 1994), to estimate blood-lead
concentrations in children based on local environmental concentrations (air, soil/dust, drinking
water, food, etc). The model includes defaults based on typical concentration levels in an urban
setting (U.S. EPA, 1994). The default air concentration used in the IEUBK model is 0.1 |ig/m3,
which is approximately the average 1990 U.S. urban air lead concentration (U.S. EPA, 1991b).
This default/background concentration is 1,000 times higher than the ambient air concentration of
0.00009 |ig/m3 estimated from a HASL process (Section 3.2.3). The model was run at various air
concentrations down to 0.001 jig/m3 (the model does not accept air concentration values less than
0.001 |ig/m3). At those levels, such small changes to the air concentration result in no real
difference in estimated blood-lead concentrations compared to results obtained from using the
default values (i.e., typical urban levels of lead to which a child may be exposed). These results
are shown in Table 3-20. Since the estimated air concentration of lead from HASL is so far
below the default/background level in air, and the model could not discern any change in
children's blood-lead levels from those at average urban air concentrations, it can be concluded
that general population exposure to airborne lead from the HASL process is negligible.
Table 3-20. Children's Blood-Lead Results from the IEUBK Model at Various Lead
Air Concentrations
Age
(year)
0.5- 1
1-2
2-3
3-4
4-5
5-6
6-7
Blood-Lead Results (jig/dL) at Various Airborne Lead Concentrations
1 (ug/m3 in air)
4.2
4.7
4.4
4.2
3.6
3.2
2.9
0.1 (jig/m3 in air)
4.1
4.5
4.2
4.0
3.4
3.0
2.7
0.01 (u£/m3 in air)
4.1
4.5
4.2
4.0
3.4
2.9
2.7
0.001 (jig/m3 in air)
4.1
4.5
4.2
4.0
3.4
2.9
2.7
Note: Model default values were used for concentrations in soil/dust, drinking water, and diet.
3.2.5 Uncertainty and Variability
Because of both the uncertainty inherent in the parameters and assumptions used in
estimating exposure, and the variability that is possible within a population, there is no one
number that can be used to describe exposure. In addition to data and modeling limitations,
discussed in Sections 3.2.3, sources of uncertainty in assessing exposure include the following:
• Accuracy of the description of exposure setting: how well the model facility used in the
assessment characterizes an actual facility; the likelihood of exposure pathways actually
occurring (scenario uncertainty).
• Missing data and limitations of workplace practices data: this includes possible effects of
any chemicals that may not have been included (e.g., minor ingredients in the
3-75
-------
formulations; possible effects of side reactions in the baths which were not considered;
and questionnaire data with limited facility responses).
• Estimating exposure levels from averaged data and modeling in the absence of measured,
site-specific data.
• Data limitations in the Source Release Assessment: releases to land could not be
characterized quantitatively, as discussed in Section 3.1.
• Chemical fate and transport model applicability and assumptions: how well the models
and assumptions represent the situation being assessed, and the extent to which the
models have been validated or verified (model uncertainty).
• Parameter value uncertainty, including measurement error, sampling error, parameter
variability, and professional judgement.
• Uncertainty in combining pathways for an exposed individual.
A method typically used to provide information about the position an exposure estimate
has in the distribution of possible outcomes is the use of exposure (or risk) descriptors. EPA's
Guidelines for Exposure Assessment (U.S. EPA, 1992b) provides guidance on the use of risk
descriptors, which include the following:
• High-end: approximately the 90th percentile of the actual (measured or estimated)
distribution. This is a plausible estimate of individual risk for those persons at the upper
end of the exposure distribution, and is not higher than the individual in the population
who has the highest exposure.
• Central tendency: either an average estimate (based on average values for the exposure
parameters) or a median estimate (based on 50th percentile or geometric mean values).
• What-if: represents an exposure estimate based on postulated questions (e.g., what if the
air ventilation rates were ...), in this case, making assumptions based on limited data so
that the distribution is unknown. If any part of the exposure assessment qualifies as a
"what-if' descriptor, then the entire exposure assessment is considered "what-if"
This exposure assessment uses whenever possible a combination of central tendency
(either an average or median estimate) and high-end (90th percentile)10 assumptions, as would be
used for an overall high-end exposure estimate. The 90th percentile is used for:
• hours per day of workplace exposure;
• exposure frequency (days per year);
• exposure duration in years (90th percentile for occupational and 95th percentile for
residential exposures);
• time required for chemical bath replacement; and
• the time and frequency of filter replacements, conveyor equipment cleaning and chemical
bath sampling (minutes per occurrence and number of occurrences per year).
Average values are used for:
10 For exposure data from the PWB Workplace Practices Questionnaire, this means that 90 percent of the
facilities reported a lower value, and ten percent reported a higher value.
3-76
-------
• body weight;
• concentration of chemical in bath;
• frequency of chemical bath replacements;
• the number of baths in a given process; and
• bath size.
However, because some data, especially pertaining to bath concentrations and inhalation
exposure are limited, and this exposure assessment does not apply to a specific facility, the entire
exposure assessment should be considered "what-if."
3.2.6 Summary
This exposure assessment uses a "model facility" approach, with the goal of comparing
the exposures and health risks of one surface finishing technology to the exposures and risks
associated with switching to another technology. As much as possible, reasonable and consistent
assumptions are used across alternatives. Data to characterize the model facility and exposure
patterns for each surface finishing technology were aggregated from a number of sources,
including PWB shops in the U.S., supplier data, and input from PWB manufacturers at project
meetings. Thus, the model facility is not entirely representative of any one facility, and actual
exposure (and risk) could vary substantially, depending on site-specific operating conditions and
other factors.
Chemical exposures to PWB workers and the general population from day-to-day surface
finishing line operations were estimated by combining information gathered from industry (PWB
Workplace Practices Questionnaire, MSDSs, and other available information) with standard EPA
exposure assumptions for inhalation rate, surface area of dermal contact, and other parameters.
The pathways identified for potential exposure from surface finishing process baths were
inhalation and dermal contact for workers, and inhalation contact only for the general populace
living near a PWB facility.
The possible impacts of short-term exposures to high levels of hazardous chemicals
addressed have not been addressed, such as those that could occur from chemical fires, spills, or
other episodic releases.
Inhalation exposure could occur by breathing air containing vapor or aerosol-phase
chemicals from the surface finishing process line. Inhalation exposures to workers are estimated
only for non-conveyorized lines; inhalation exposure to workers from conveyorized surface
finishing lines was assumed to be much lower because the lines are typically enclosed and vented
to the outside.11
11 Inhalation exposures for conveyorized process configurations were initially assumed to be negligible, and are
not presented separately here. Some inhalation exposure is possible, however, during sampling and bath replacement,
when the baths are opened for a short period of time. After characterizing risks from inhalation for non-conveyorized
lines, inhalation exposures and risks were estimated for conveyorized lines. No chemical exposures from inhalation
resulted in risks above concern levels for conveyorized lines.
3-77
-------
The daily intake for inhalation exposure to workers was calculated by first modeling
chemical emissions from surface finishing baths with three air-transport mechanisms: liquid
surface diffusion (desorption), bubble desorption, and aerosol generation and ejection. These
chemical emission rates were combined with information from the PWB Workplace Practices
Questionnaire regarding process room size and air turnover rate to estimate an average indoor air
concentration for each chemical for the process area. General room ventilation was assumed,
although the majority of shops have local ventilation on chemical tanks. An uncertainty and
sensitivity analysis of the air transport models (U.S. EPA, 1998b) suggests that the air turnover
(ventilation) rate assumption greatly influences the estimated air concentration in the process area
because of its large variability.
Inhalation exposure to the human population surrounding PWB plants was estimated
using the Industrial Source Complex - Long Term (ISCLT) air dispersion model. The modeled
air concentrations of each contaminant were determined at 100 meters radially from a PWB
facility, and the highest estimated air concentration was used. This model estimates air
concentration from the process bath emission rates. These emissions were assumed to be vented
to the ambient environment at the rate emitted from the baths, for all process alternatives.
Inhalation exposures estimated for the public living 100 meters away from a PWB facility were
very low (approximately 10,000 times lower than occupational exposures).
Dermal exposure could occur when a worker's skin comes in contact with the bath
solution while dipping boards, adding replacement chemicals, etc. Although the data suggest that
surface finishing line operators often do wear gloves, it was assumed in this evaluation that
workers do not wear gloves to account for the fraction that do not. Otherwise, dermal exposure
is expected to be negligible. For dermal exposure, the duration of contact for workers was
obtained from the PWB Workplace Practices Questionnaire information. A permeability
coefficient (rate of penetration through skin) was estimated for organics, and a default rate
assumption was used for inorganics. Another source of uncertainty in dermal modeling lies with
the assumed duration of contact. For non-conveyorized processes, the worker is assumed to
have potential dermal contact for the entire time spent in the surface finishing area, divided
equally among the baths. [This does not mean that a worker has both hands immersed in a bath
for that entire time; but that the skin is in contact with bath solution (i.e., the hands may remain
wet from contact).] This assumption may result in an overestimate of dermal exposure.
Assumptions and parameter values used in these equations are presented throughout this
section. Exposure estimates are based on a combination of high end (90th percentile)12 and
average values, as would be used for a high-end exposure estimate. The 90th percentile was used
for hours per day of workplace exposure, exposure frequency (days per year), exposure duration
in years (90th percentile for occupational and 95th percentile for residential exposures), and the
time and frequency of chemical bath and filter replacements, conveyor equipment cleaning and
chemical bath sampling (minutes per occurrence and number of occurrences per year) and
estimated workplace air concentrations. The average value was used for body weight,
12 For exposure data from the PWB Workplace Practices Questionnaire, this means that 90 percent of the
facilities reported a lower value, and ten percent reported a higher value.
3-78
-------
concentration of chemical in bath, and the number of baths in a given process. However,
because some data, especially pertaining to bath concentrations and inhalation exposure, are
limited and this exposure assessment does not apply to a specific facility, the entire exposure
assessment should be considered "what-if."
As a "what if exposure assessment, this evaluation is useful for comparing alternative
surface finishing processes to the baseline (non-conveyorized HASL) on a consistent basis. It is
also useful for risk screening, especially if actual facility conditions meet those that were assumed
(i.e., given similar production rates, what chemicals may be of concern if workers do not wear
gloves; what chemicals may be of concern if ventilation rates are similar to those assumed?).
Finally, this assessment points to the importance of preventing dermal contact by using gloves,
and of proper ventilation.
Surface water concentrations were estimated for bath constituents, with a focus on those
constituents that are not typically targeted for pre-treatment by PWB facilities. This was done for
conveyorized lines by estimating the amount of chemical going to wastewater from routine bath
replacement, and for non-conveyorized lines by estimating the amount of chemical going to
wastewater from bath replacement plus an estimated amount due to drag-out from the baths to
rinse water. These amounts were then included in a stream dilution model, and if estimated
surface water concentrations exceeded CCs for aquatic life, the model was refined using
estimated POTW treatment efficiencies.
These exposure results, taken by themselves, are not very meaningful for evaluating
surface finishing alternatives; it is the combination of hazard (Section 3.3) and exposure that
defines risk. Quantitative exposure estimates are combined with available hazard data in the risk
characterization (Section 3.4) for risk screening and comparison of the surface finishing process
configurations.
3-79
-------
3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
This section presents a summary of the human health and ecological hazards data that are
used in the risk characterization. This information is summarized from toxicity profiles prepared
for chemicals identified as constituents in the baths for the surface finishing technologies
evaluated. Table 2-1 lists these chemicals and identifies the surface finishing process or processes
in which these chemicals are used. HASL is the predominant method now used for surface
finishing. Section 2.1.4 includes more detailed information on bath constituents and
concentrations. Throughout this section, proprietary chemicals are identified only by generic
name, with limited information presented, in order to protect proprietary chemical identities.
3.3.1 Carcinogenicity
The potential for a chemical to cause cancer is evaluated by weight-of-evidence (WOE)
classifications and by cancer potency factors, typically determined from laboratory or
epidemiological studies. There are a large number of chemicals in commerce, however,
(approximately 15,000 non-polymeric chemicals produced in amounts greater than 10,000
Ib/year), and many of these chemicals have not yet been tested or assigned carcinogenicity
classifications. The WOE classifications referenced in this risk assessment are defined below.
In assessing the carcinogenic potential of a chemical, EPA classifies the chemical into one
of the following groups, according to the WOE from epidemiologic, animal and other supporting
data, such as genotoxicity test results:
• Group A: Human Carcinogen (sufficient evidence of carcinogenicity in humans).
• Group B: Probable Human Carcinogen (Bl - limited evidence of carcinogenicity in
humans; B2 - sufficient evidence of carcinogenicity in animals with inadequate or lack of
evidence in humans).
• Group C: Possible Human Carcinogen (limited evidence of carcinogenicity in animals
and inadequate or lack of human data).
• Group D: Not Classifiable as to Human Carcinogenicity (inadequate or no evidence).*
Group E: Evidence of Non-Carcinogenicity for Humans (no evidence of carcinogenicity
in adequate studies).
EPA has proposed a revision of its guidelines that would eliminate the above discrete
categories while providing a more descriptive classification.13
The International Agency for Research on Cancer (IARC) uses a similar WOE method for
evaluating potential human carcinogenicity based on human data, animal data, and other
supporting data. A summary of the IARC carcinogenicity classification system includes:
13 The "Proposed Guidelines for Carcinogen Risk Assessment" (U.S. EPA, 1996b) proposes the use of WOE
descriptors, such as "Likely" or "Known," "Cannot be determined," and "Not likely," in combination with a hazard
narrative, to characterize a chemical's human carcinogenic potential, rather than the classification system described
above.
3-80
-------
Group 1: Carcinogenic to humans.
Group 2A: Probably carcinogenic to humans.
Group 2B: Possibly carcinogenic to humans.
Group 3: Not classifiable as to human carcinogenicity.
Group 4: Probably not carcinogenic to humans.
Both of these classification schemes represent judgements regarding the likelihood of
human carcinogenicity. Table 3-21 lists all surface finishing chemicals that have been classified
by EPA or IARC. The National Toxicology Program (NTP) is an additional source used to
classify chemicals, but its classifications are based only on animal data from NTP studies.
Table 3-21. Available Carcinogenicity Information
Chemical Name a
Cancer Slope
Factor
(Inhalation Unit
Risk)
(ug/m3)1
Cancer Slope
Factor
(Oral)
(mg/kg-day)1
Comments/Classification
Known, probable, or possible human carcinogens
Inorganic metallic
salt A
Sulfuric acid d
Lead
Thiourea
Urea compound B
Not reported b
ND
ND
ND
ND
ND
ND
ND
ND
ND
Human carcinogen or probable human
carcinogen. °
IARC Group 1 e (IARC 1992).
EPA Class B2 f (IRIS, 1999); IARC
Group 2B 8 (IARC, 1987).
IARC Group 2B e (IARC 1974).
Possible human carcinogen. °
Other weight-of-evidence (WOE) or other information available
Nickel sulfate
Copper ion,
Copper salt, and
Copper sulfate
pentahydrate
Hydrochloric acid
Hydrogen peroxide
Vinyl polymer
Silver nitrate
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
Nickel refinery dust is IARC Group 1 e
(IARC, 1990). No assessment
available for soluble salts of nickel.
Copper is EPA Class D h (IRIS, 1998).
IARC Group 3 ' (HSDB, 1998), excess
lung and laryngeal cancer occurred in
workers exposed to HCL mist;
however, many of these cases
involved exposure to acid mixtures
(Perry etal., 1994).
IARC Group 3 ' (IARC, 1987),
stomach tumors occurred in mice (Ito
etal., 1981).
Not classifiable according to EPA
and/or IARC. c
Silver is EPA Class D h (IRIS, 1998).
3-81
-------
Chemical Name a
Silver salt
Stannous methane
sulfonic acid
Tin chloride
Palladium chloride
Propionic acid
Cancer Slope
Factor
(Inhalation Unit
Risk)
(ug/m3)1
ND
ND
ND
ND
ND
Cancer Slope
Factor
(Oral)
(mg/kg-day)1
ND
ND
ND
ND
ND
Comments/Classification
Not classifiable according to EPA
and/or IARC. c
EPA Class D h (U.S. EPA, 1987a).
EPA Class D h or IARC Group 31 (U.S.
EPA, 1987a).
No classification; mice administered
palladium in drinking water had a
significantly higher incidence of
malignant tumors (Schroeder and
Mitchener, 1971).
No classification; tumors in
forestomach of rats (Clayson et al.,
1991).
a Only those chemicals with available data or classifications are listed.
b The unit risk value is not reported here to protect confidential ingredient identity.
0 Specific EPA and/or IARC groups not reported in order to protect proprietary chemical identities.
d Classification pertains to the strong inorganic acid mist.
e IARC Group 1: Human Carcinogen.
f EPA Class B2: Probable Human Carcinogen (sufficient evidence of carcinogenicity in animals with inadequate or
lack of evidence in humans).
g IARC Group 2B: Possibly carcinogenic to humans.
h EPA Class D: Not classifiable as to human carcinogenicity.
1 IARC Group 3: Not classifiable as to its carcinogenicity to humans.
ND: No Data, a cancer slope factor has not been determined for this chemical.
For carcinogenic effects, there is presumably no level of exposure that does not pose a
small, but finite, probability of causing a response. This type of mechanism is referred to as
"non-threshold." When the available data are sufficient for quantification, EPA develops an
estimate of the chemical's carcinogenic potency expressed as a "slope factor." The slope factor
(qi*) is a measure of an individual's excess risk or increased likelihood of developing cancer if
exposed to a chemical (expressed in units of [mg/kg-day]"1). More specifically, qi* is an
approximation of the upper bound of the slope of the dose-response curve using the linearized,
multistage procedure at low doses. "Unit risk" is an equivalent measure of potency for air or
drinking water concentrations and is expressed as the upper bound excess lifetime cancer risk per
//g/m3 in air, or as risk per //g/L in water, for continuous lifetime exposures. (Unit risk is simply a
transformation of slope factor into the appropriate scale.) Slope factors and unit risks can be
viewed as quantitatively derived judgements of the magnitude of carcinogenic effect. These
estimates will continue to be used whether the current EPA WOE guidelines are retained or the
new proposals are adopted. Their derivation, however, may change for future evaluations.
3-82
-------
EPA risk characterization methods require a slope factor or unit risk to quantify the upper
bound, excess cancer risk from exposure to a known or suspected carcinogen. There is only one
chemical, inorganic metallic salt A, with a slope factor. Therefore, this is the only chemical for
which cancer risk can be characterized (see Section 3.4, Risk Characterization).
3.3.2 Chronic Effects (Other than Carcinogenicity)
Adverse effects, other than cancer and gene mutations, are generally assumed to have a
dose or exposure threshold. Therefore, a different approach is used to evaluate toxic potency and
risk for these "systemic effects." Systemic toxicity means an adverse effect on any organ system
following absorption and distribution of a toxicant to a site in the body distant from the
toxicant's entry point. A reference dose (RfD) is an estimate (with uncertainty spanning perhaps
an order of magnitude) of the daily exposure through ingestion to the human population
(including sensitive subgroups) that is likely to be without an appreciable risk of deleterious non-
cancer effects during a lifetime (in mg/kg-day). Similarly, a reference concentration (RfC) is an
estimate (with uncertainty spanning perhaps an order of magnitude) of the daily inhalation
exposure to the human population (including sensitive subgroups) that is likely to be without an
appreciable risk of deleterious non-cancer effects during a lifetime (in mg/m3) (Barnes and
Dourson, 1988). RfDs and RfCs also can be derived from developmental toxicity studies.
However, this was not the case for any of the surface finishing chemicals evaluated. RfDs and
RfCs are derived from EPA peer-reviewed study results (for values appearing in EPA's
Integrated Risk Information System [IRIS]), together with uncertainty factors regarding their
applicability to human populations. Table 3-22 presents a summary of the available RfC and RfD
information obtained from IRIS and EPA's Health Effects Assessment Summary Tables
(HEAST) for non-proprietary chemicals. An additional proprietary chemical has an RfC and an
RfD; these data are not reported in order to protect the identity of the confidential ingredient.
3-83
-------
Table 3-22. Summary of RfC and RfD Information Used in Risk Characterization for
Non-Proprietary Ingredients
Chemical
Name3
Ammonium
chloride,
Ammonium
lydroxide
Ethylenediamine
Ethylene glycol
Ethylene glycol
monobutyl ether
Hydrochloric
acid
Leadf
Nickel sulfate
Phosphoric acid
Potassium gold
cyanide
Silver nitrate
Inhalation
RfCb
(mg/m3)
O.ld (IRIS)
ND
ND
13 (IRIS)
0.02 (IRIS)
Comments c
(Inhalation)
Ammonia: decreased
lung function (IRIS,
1999).
Changes in red blood
cell count (IRIS,
1999).
Rats, hyperplasia of
nasal mucosa, larynx,
and trachea (IRIS,
1998).
Oral/Dermal
RfDb
(mg/kg/day)
0.2 e (IRIS)
0.02 (HEAST)
2 (IRIS)
0.5 (IRIS)
ND
Comments c
(Oral/Dermal)
Ammonium sulfamate:
rats, drinking water, 90
days, decreased body
weight (Gupta et al, 1979;
IRIS, 2000).
Rats, 3 months, increased
leart weight and
lematologic changes
(U.S. EPA, 1997b).
Rats, kidney toxicity
(IRIS, 1999).
Changes in mean
corpuscular volume
(IRIS, 1999).
ND: Some health effects of lead, particularly changes in the levels of certain blood
enzymes and in aspects of children's neurobehavioral development, may occur at
blood- lead levels so low as to be essentially without a threshold. EPA considers it
inappropriate to develop an RfD for inorganic lead (IRIS, 2000).
0.00053 s (MRL)
0.01 (IRIS)
ND
ND
Rats, lung
inflammation
(ATSDR, 1997a).
Rats, histologic
lesions in
tracheobronchiolar
region (IRIS, 1998).
0.02 (IRIS)
(soluble salts
of nickel)
221 (ADI)
0.02 h (IRIS)
0.005 ' (IRIS)
Rats, decreased body and
organ weight (IRIS,
1998).
(U.S. EPA, 1997c; WHO,
1974).
Cyanide: rats, 2 year,
weight loss, thyroid
effects and myelin
degeneration, (IRIS,
1998).
Silver-argyria (benign but
permanent bluish-gray
discoloration of skin)
(Gaul and Staud, 1935).
3-84
-------
Chemical
Name3
Stannous methane
sulfonic acid,
Tin, and
Tin chloride
Sulfuric acid
Inhalation
RfCb
(mg/m3)
ND
0.07 (HEAST)
Comments c
(Inhalation)
Acceptable air
concentration for
lumans based on
respiratory effects
(U.S. EPA, 1997b).
Oral/Dermal
RfDb
(mg/kg/day)
0.6 J (HEAST)
NDk
Comments c
(Oral/Dermal)
Tin and inorganic
compounds: rats, 2 year,
listopathologic study
(U.S. EPA, 1997b).
a Only non-proprietary chemicals with available data are listed.
b The type of value is noted in parentheses:
IRIS: EPA-derived and peer-reviewed values listed in the Integrated Risk Information System. IRIS values
are preferred and used whenever available.
HEAST: EPA-derived RfD or RfC listed in the Health Effects Assessment Summary Tables. These values
have not undergone the same level of review as IRIS values.
ADI: Acceptable daily intake, developed by the World Health Organization (WHO).
MRL: Minimal risk level, developed by the Agency for Toxic Substances and Disease Registry (ATSDR) in
a manner similar to EPA-derived values.
0 Comments may include exposure route, test animal, duration of test, effects, and source of data.
d In the risk calculations, conversion factors are used based on the molecular weights of ammonia, ammonium
chloride, and ammonium hydroxide.
e In the risk calculations, conversion factors are used based on the molecular weights of ammonium sulfamate,
ammonium chloride, and ammonium hydroxide.
f More information on lead is presented in Section 3.4.6 of the Risk Characterization.
g Value given represents a chronic inhalation minimum risk level (MRL). Although the test substance was nickel
sulfate hexahydrate, the reported value is 0.0002 mg/m3 as nickel. This was converted in the risk calculations based
on the molecular weights of nickel and nickel sulfate.
h A conversion factor is used in the risk calculations based on molecular weights of cyanide and potassium gold
cyanide. This RfD is only relevant to the oral route; potassium gold cyanide is expected to be chemically stable except
under highly acidic conditions such as those found in the stomach (pH 1-2).
1 A conversion factor is used in the risk calculations based on molecular weights of silver and silver nitrate.
] Conversion factors are used in the risk calculations based on molecular weights of tin, tin chloride, and stannous
methane sulfonic acid.
k Although chronic toxicity values have not been established, repeated skin contact with low concentrations of
sulfuric acid causes skin dessication, ulceration of the hands, and chronic inflammation around the nails.
ND: No data, RfC or RfD not available.
When an RfD or RfC was not available for a chemical, other toxicity values were used,
preferably in the form of a "no-observed-adverse-effect level" (NOAEL) or "lowest-observed-
adverse-effect level" (LOAEL). These toxicity values were obtained from the published scientific
literature, as well as unpublished data submitted to EPA on chemical toxicity in chronic or
subchronic studies. Typically, the lowest NOAEL or LOAEL value from a well-conducted study
was used. (If study details were not presented or the study did not appear to be valid, the
reported NOAEL/LOAELs were not used.) But, unlike the majority of RfD/RfCs,
NOAEL/LOAELs have not received EPA peer-review of the studies on which the values are
based, and uncertainty factors have not been considered.
3-85
-------
The LOAEL is the lowest dose level in a toxicity test at which there are statistically or
biologically significant increases in frequency or severity of adverse effects in the exposed
population over its appropriate control group (in mg/kg-day, or mg/m3 for inhalation). The
NOAEL is the highest dose level in a toxicity test at which there is no statistically or biologically
significant increase in the frequency or severity of adverse effects in the exposed population over
its appropriate control (in mg/kg-day, or mg/m3 for inhalation). LOAEL values are presented
only where NOAELs were not available. Table 3-23 presents a summary of the available NOAEL
and LOAEL values for non-proprietary chemicals. Chemicals having potential developmental
toxicity were identified based on the data provided in the toxicity profiles. These data are
summarized in Table 3-24. An additional 5 proprietary chemicals have inhalation NOAELs or
LOAELs, and 13 have oral NOAELs or LOAELs; these data are not reported in order to protect
the identity of confidential ingredients.
Neither RfDs/RfCs nor LOAELs/NOAELs were available for some chemicals in each
surface finishing process alternative. For these chemicals, no quantitative estimate of risk could
be calculated. EPA's Structure-Activity Team (SAT)14 has reviewed the chemicals without
relevant toxicity data to determine if these chemicals are expected to present a toxicity hazard.
This review was based on available toxicity data on structural analogues of the chemicals, expert
judgement, and known toxicity of certain chemical classes and/or moieties. Chemicals received a
concern level rank of high, moderate-high, moderate, moderate-low, or low. Results of the SAT
evaluation are presented in Table 3-25. A summary of toxicity data available for the chemicals is
presented in Table 3-26.
14 The SAT is a group of expert scientists at EPA who evaluate the potential health and environmental hazards of
new and existing chemicals.
3-86
-------
Table 3-23. NOAEL/LOAEL Values Used in Risk Characterization for Non-Proprietary
Ingredients
Chemical
Name a
Acetic acid
Copper ion,
Copper sulfate
pentahydrate
Ethylenediamine
Ethylene glycol
Hydrogen peroxide
Lead1
Propionic acid
Inhalation
NOAEL/
LOAEL b
(mg/m3)
NDd
0.6 (L) e
145 (N) g
31(L)
79 (L) h
lOug/dL
in blood
23(TClo)]
Comments c
(Inhalation)
Cupric chloride: rabbits, 6
hrs/day, 5 days/wk for 4-6 wks,
increase in lung tumors (U.S. Air
Force, 1990).
Rats, 7 hrs/day, 5 days/wk for 30
days, depilation (Pozzani and
Carpenter, 1954).
Humans, 20-22 hrs/day for 30
days, respiratory irritation,
headache, and backache
(ATSDR, 1997b).
Mouse, 6 weeks, 7/9 died (U.S.
EPA, 1988a).
Children, level concern in blood
(CDC, 1991).
Rats, subchronic exposure
(RTECS, 1998).
Oral/Dermal
NOAEL/
LOAEL b
(mg/kg-day)
195 (N)
0.056 (L) f
NA
NA
290 (L)
lOug/dL
in blood
150 (N)
Comments c
(Oral/Dermal)
Rats, drinking water, 2-4
months, no deaths (Sollmann,
1921).
Copper: humans, 1.5 years,
abdominal pain and vomiting
(ATSDR, 1990a).
RfD is available (Table 3-22).
RfD is available (Table 3-22).
Mice, 35 weeks, liver, kidney,
and GI effects (IARC, 1985).
Children, level concern in blood
(CDC, 1991).
Rats, diet, lesions in GI tract
(BASF, 1987; Mori, 1953;
Harrison et al., 1991; Rodrigues
etal., 1986).
a Only non-proprietary chemicals with available data are listed.
b (N) = NOAEL; (L) = LOAEL. When more than one NOAEL and/or LOAEL was available, only the lowest
available NOAEL or LOAEL was used and is listed here. If both NOAEL and LOAEL data are available, the NOAEL
is used and is listed here. If a chronic NOAEL or LOAEL was not available, other values (e.g., from shorter-term
studies) were used as noted.
0 Comments may include exposure route, test animal, duration of test, effects, and source of data.
d Although health effects have been noted in workers and laboratory tests from inhalation exposure to acetic acid, no
appropriate chronic inhalation toxicity value is available.
e Conversion factors are used in the risk calculations based on molecular weights of cupric chloride, copper ion, and
copper sulfate pentahydrate.
f A conversion factor is used in the risk calculations based on molecular weights of copper and copper sulfate
pentahydrate.
g Not considered a "chronic" value because the study duration was less than 90 days. The value was used, however,
as the best available value, rather than leaving a data gap for a chemical where adverse health effects have been noted.
h In the absence of other data, this value will be used as a LOAEL.
1 More information on lead is presented in Section 3.4.5 of the Risk Characterization.
] TClo = The lowest dose of a chemical that is expected to cause a defined toxic effect. In the absence of other data,
this is used as a LOAEL.
ND: No Data. A NOAEL or LOAEL was not available for this chemical.
NA: Not applicable. A NOAEL or LOAEL is not required because an RfC or RfD is available for this chemical.
3-87
-------
Table 3-24. Developmental Toxicity Values Used in Risk Characterization for Non-
Proprietary Ingredients
Chemical a
Ammonium
chloride
Copper ion,
Copper sulfate
pentahydrate
Ethylenediamine
Ethylene glycol
Ethylene glycol
monobutyl ether
Developmental
Inhalation
NOAEL /
LOAEL
(mg/m3) "
ND
ND
ND
150 (N)
ND
Comments c
(Inhalation)
Rats and mice, 6 hr/day, gd 6-
15, fetal malformations in
mice (exencephaly, cleft
palate, and abnormal rib and
facial bones) (Shell Oil, 1992;
Union Carbide, 1991).
Developmental
Oral/Dermal
NOAEL /
LOAEL b
(mg/kg-day)
1,691 (N)
3(L)e
470 (L)
500 (N)
100 (N)
Comments c
(Oral/Dermal)
Mice, drinking water, after gd d
7, no congenital effects
(Shepard, 1986).
Copper: mink, diet, increased
mortality (Aulerich et al., 1982;
ATSDR, 1990a).
Rats, gd 6-15 diet, resorption,
impaired growth, missing or
shortened innominate arteries,
and delayed ossification of
cervical vertebrae or phalanges
(DePassetal., 1987).
Rats, gd 6-15, gavage,
teratogenic effects at higher
dose levels. NOAEL based on
developmental effects (Bushy
Run, 1995).
Rats, gd 9-1 1, oral gavage,
developmental toxicity (Sleet
etal., 1989).
a Only those chemicals with available data are listed.
b (N) = NOAEL; (L) = LOAEL. When more than one NOAEL and/or LOAEL was available, only the lowest
available NOAEL or LOAEL was used and is listed here. If both NOAEL and LOAEL data are available, the NOAEL
is used and is listed here.
0 Comments may include test effects, test animal, duration during time of gestation, exposure route, and source of
data.
d gd = gestation day.
e Conversion factors are in the risk calculations based on molecular weights of copper ion and copper sulfate
pentahydrate.
ND: No data available.
3-S
-------
Table 3-25. Summary of Health Effects Information
(from Structure-Activity Team Reports)
Chemical
SAT Health Effects Pertaining to
Dermal or Inhalation Exposure
Overall
Concern Level
1,4-Butenediol
Expect good absorption via all routes of exposure. The
primary alcohols will oxidize to the corresponding acids
(fumaric or maleic) via aldehydes. There is concern for
mutagenicity as an unsaturated aldehyde. This compound is
expected to be irritating to the lungs and other mucous
membranes. Effects on the liver and kidney and neurotoxicity
(sedation) are also expected.
Low moderate
Aliphatic acid B
Expect no absorption by skin, but expect absorption by lungs
and GI tract. Related compound is reported to be positive in a
dominant lethal assay. Uncertain concerns for developmental
toxicity and kidney toxicity. Some concern for irritation.
Moderate
Aliphatic
dicarboxylic acid A
Absorption is expected to be poor through the skin and good
through the lungs and GI tract. As a free acid, this compound
is expected to be irritating to all exposed tissues. A mixture of
acids containing this compound was tested in rats. The
mixture was negative for mutagenicity but caused signs of
neurotoxicity. A mixture containing the dimethyl ester of this
compound was tested in acute inhalation and dermal studies
because blurring of vision had been reported in humans. An
increase in the anterior chamber depth in the eye was seen
following inhalation and dermal exposure. This could be an
indication of changes in circulation in the eye which could
lead to glaucoma. A mixture of the same compounds was
tested in a 1-generation reproduction study in rats via
inhalation, showing a decrease in postnatal pup weight and
irritation of the respiratory tract in parental animals.
Low moderate
Alkylalkyne diol
Expect poor absorption via all routes of exposure. This
compound may be irritating to the eyes, lungs, and mucous
membranes and cause defatting of the skin which can lead to
skin irritation. There is uncertain concern for neurotoxicity
and liver and kidney effects.
Low
Alkylamino acid A
Absorption is expected to be poor through the skin and good
through the lungs and GI tract. This compound is expected to
chelate metals such as calcium, magnesium, and zinc. Based
on its potential to chelate calcium, there is concern for
developmental toxicity, inhibition of blood clotting, and
effects on the nervous system and muscles including effects on
the heart. Chelation of zinc may cause immunotoxicity
(retardation of wound healing). This compound is expected to
be irritating to all exposed tissues and may be a dermal
sensitizer. A salt of this compound caused developmental
effects in rats. There is concern for oncogenicity and kidney
toxicity. There is also a potential for male reproductive
effects. This compound may be mutagenic.
Low moderate
3-89
-------
Chemical
SAT Health Effects Pertaining to
Dermal or Inhalation Exposure
Overall
Concern Level
Alkylaryl imidazole
Expect good absorption via the lungs and GI tract. Absorption
of the neat material is expected to be nil through the skin;
however, absorption is expected to be moderate through the
skin when in solution. There is concern for developmental
toxicity and neurotoxicity.
Low moderate
Alkylaryl sulfonate
Absorption is expected to be nil through the skin and poor
through the lungs and GI tract. There is uncertain concern for
irritation to mucous membranes.
Low
Alkylimine
dialkanol
Absorption is expected to be poor through the skin, moderate
through the GI tract, and good through the lungs. This
compound is a moderate to severe skin irritation and a severe
eye irritant. It has low acute toxicity. Another analog was
tested in a subchronic gavage study in rats and dogs. Cataracts
were noted in rats, stomach and lung lesions consistent with
irritation were seen, and liver effects were seen in female dogs.
There is concern for developmental toxicity. There is little
concern for mutagenicity by analogy to a similar compound.
Moderate
Amino acid salt
Absorption is expected to be nil through the skin and good
through the lungs and GI tract. There is uncertain concern for
developmental toxicity. This compound is an amino acid
analog and may be an antimetabolite. This chemical is also
expected to be an irritant to moist tissues such as the lungs and
respiratory tract.
Low moderate
Ammonia
compound B
Absorption is expected to be nil through the skin and good
through the lungs and GI tract. This material will be irritating
and/or corrosive to all exposed tissues. The degree of
irritation is a function of the concentration. Fluoride causes
dental fluorosis (pitting and discoloration of the teeth) and
crippling skeletal malformations. Additional concerns for this
compound are neurotoxicity, mutagenicity, and possibly
developmental toxicity. The uncertain concern for
developmental toxicity is by analogy to ammonium chloride.
Moderate high
Aryl phenol
Expect moderate absorption by all routes. Moderate concerns
for oncogenicity due to positive data; low moderate concerns
for mutagenicity due to positive Ames and mouse lymphoma
assays; low moderate concerns for renal effects and
developmental and reproductive toxicity due to presence of
phenolic moiety.
Moderate
Bismuth
compound
Absorption is expected to be nil through the skin and good
through the lungs and GI tract. In water, this compound will
cause irritation of all moist tissues. There is also concern for
neurotoxicity and possibly developmental toxicity. There is no
concern for mutagenicity based on negative results for DNA
damage. This compound has a relatively high oral LD50.a
Moderate, based
on irritation
Citric acid
Expect poor absorption by skin, but expect absorption by
lungs and GI tract. No health concerns identified.
Low
3-90
-------
Chemical
SAT Health Effects Pertaining to
Dermal or Inhalation Exposure
Overall
Concern Level
Ethoxylated
alkylphenol
Absorption is expected to be poor through the skin, moderate
through the GI tract, and good through the lungs. As a
surfactant, this compound may cause lung effects if inhaled.
This compound is expected to be a severe and persistent eye
irritant. Eye irritation is of particular concern because this type
of compound can anesthetize the eye so an individual will not
feel pain and rinse the material out of the eye. It is also
expected to be irritating to the lungs. Possible signs of lung
irritation (lung discoloration) were noted with a similar
chemical tested in an acute inhalation study in rats. There is
uncertain concern for reproductive effects and
immunotoxicity. By analogy to a related compound, this
chemical may be an endocrine disrupter. Liver and kidney
effects were noted in rats with a structural analog. Myocardial
degeneration has also been noted in several species with
related compounds. Developmental toxicity as demonstrated
by skeletal changes has been noted with dermal and oral
exposure.
Low moderate
Fatty amine
Absorption is expected to be poor through the skin, moderate
through the GI tract, and good through the lungs. This
compound is expected to be a strong irritant and/or corrosive
to exposed tissues. A similar compound was reported to be a
moderate skin irritant and a severe eye irritant. Oleyl amine is
a severe irritant. There is also concern for lung effects if
inhaled. Another analog was tested in a subchronic gavage
study in rats and dogs. Cataracts were noted in rats, stomach
and lung lesions consistent with irritation were seen, and liver
effects were seen in female dogs. There is concern for
developmental toxicity. There is little concern for
mutagenicity by analogy to a similar compound.
Moderate
Hydroxyaryl acid
Absorption is expected to be poor through the skin and good
through the lungs and GI tract. There is concern for
developmental toxicity and uncertain concern for effects on
blood clotting (slower time for clotting). This compound is
expected to have estrogenic activity. It has low acute toxicity.
It may also cause neurotoxicity and hypersensitivity. There is
some concern for mutagenicity.
Moderate
Hydroxyaryl
sulfonate
Absorption is expected to be nil through the skin and good
through the lungs and GI tract. There is concern for
developmental toxicity. This compound is also expected to be
an irritant (the free acid is corrosive to the eyes) and may
cause neurotoxicity.
Low moderate
Maleic acid
Expect no absorption by skin, but expect absorption by lungs
and GI tract. Maleic acid is reported to be negative in a NTP
Ames assay. According to Merck this chemical is strongly
irritating to corrosive.
Moderate
3-91
-------
Chemical
SAT Health Effects Pertaining to
Dermal or Inhalation Exposure
Overall
Concern Level
Malic acid
Expect no absorption by skin, but expect absorption by lungs
and GI tract. Concerns for mild irritation to skin and eyes.
Low moderate
Potassium
compound
Absorption/corrosion by all routes. Concentrated form is
corrosive to all tissues. Dilute form may be irritating. No
other health concerns identified.
High for
concentrated
form only,
otherwise low
Potassium
peroxymonosulfate
Absorption is expected to be nil through the skin and good
through the lungs and GI tract. The peroxymonosulfate
moiety is reactive with moisture (oxidizing agent). This
material will be an irritant as a concentrated solution.
Moderate
Quaternary
alkylammonium
chlorides
Absorption is expected to be poor through the skin, moderate
through the GI tract, and good through the lungs. This
chemical is expected to be a strong irritant and/or corrosive to
all exposed tissues. It is also expected to be neurotoxic. There
is also concern for lung effects if inhaled. There is concern
for developmental toxicity as an ethanolamine derivative. This
compound is expected to be in the moderately toxic range for
acute toxicity.
Moderate
Sodium benzene
sulfonate
Absorption is expected to be nil through the skin and good
through the lungs and GI tract. There is concern for
methemoglobinemia, neurotoxicity, and developmental
toxicity. Serious brain damage was noted in a 2-week
inhalation study with a related compound. There is uncertain
concern for oncogenicity. This compound is reported to be
negative in the Ames assay. It is expected to be irritating to
mucous membranes and the upper respiratory tract.
Moderate
concern
Sodium
hypophosphite;
Sodium
hypophosphite
monohydrate
Absorption is expected to be nil through the skin and good
through the lungs and GI tract. This compound has low acute
toxicity. It is irritating to mucous membranes and may cause
dermal sensitization. There is uncertain concern for
mutagenicity. It is reported to be effective in inhibiting the
growth of selected Gram-positive pathogenic bacteria.
Low moderate
concern
Substituted amine
hydrochloride
Absorption is expected to be nil through the skin and good
through the lungs and GI tract. This chemical has fairly high
acute toxicity. It is a severe skin irritant in guinea pigs and a
weak to moderate dermal sensitizer. In a repeated dose dietary
study in rats, the primary effects were on the red blood cells
(through methemoglobin production) and the spleen. This
compound is reported to be positive in a variety of
mutagenicity assays, although there are also some negative
responses. There is concern for oncogenicity based on the
mutagenicity results. There is uncertain concern for
developmental toxicity.
Moderate
concern
3-92
-------
Chemical
SAT Health Effects Pertaining to
Dermal or Inhalation Exposure
Overall
Concern Level
Transition metal
salt
Absorption is expected to nil through the skin and good
through the lungs and GI tract. This compound is expected to
be an irritant because it is hydroscopic. There is concern for
mutagenicity. There is also concern for neurotoxicity and
uncertain concern for allergic reactions.
Moderate
concern
LD50: Lethal dose to 50 percent of the test population.
Table 3-26. Overview of Available Toxicity Data
Chemical
1,4-Butenediol
Acetic acid
Aliphatic acid A
Aliphatic acid B
Aliphatic acid D
Aliphatic acid E
Aliphatic dicarboxylic acid A
Aliphatic dicarboxylic acid C
Alkylalkyne diol
Alkylamino acid A
Alkylamino acid B
Alkylaryl imidazole
Alkylaryl sulfonate
Alkyldiol
Alkylimine dialkanol
Alkylphenol ethoxylate
Alkylphenol
polyethoxyethanol
Alkylpolyol
Amino acid salt
Amino carboxylic acid
Ammonium chloride
Ammonia compound A
Ammonia compound B
Ammonium hydroxide
Cancer:
Slope Factor (SF),
Weight-of- Evidence
(WOE)
Classification
Inhalation:
RfC, NOAEL,
or LOAEL a
Yes
Yes
RfC (for
ammonia)
RfC (for
ammonia)
RfC (for
ammonia)
RfC (for
ammonia)
Oral/Dermal:
RfD, NOAEL,
or LOAEL a
NOAEL
Yes
Yes
Yes
Yes
Yes
Yes
D-NOAEL
RfD (for ammonium
sulfamate)
Yes
Yes
RfD (for ammonium
sulfamate)
SAT
Rank
X
X
X
X
X
X
X
X
X
X
X
X
3-93
-------
Chemical
Aromatic imidizole product
Arylphenol
Bismuth compound
Citric acid b
Copper ion
Copper salt C
Copper sulfate pentahydrate
Cyclic amide
Ethoxylated alkylphenol
Ethylenediamine
Ethylene glycol
Ethylene glycol monobutyl
ether
Fatty amine
Fluoboric acid
Gum
Hydrochloric acid
Hydrogen peroxide
Hydroxy carboxylic acid
Hydroxyaryl acid
Hydroxyaryl sulfonate
Inorganic metallic salt A
Inorganic metallic salt B
Inorganic metallic salt C
Lead
Maleic acid
Malic acid °
Methane sulfonic acid
Nickel sulfate
Nitrogen acid
Nonionic surfactant
Palladium chloride
Palladium salt
Phosphoric acid
Potassium compound
Potassium gold cyanide
Potassium peroxymonosulfate
Cancer:
Slope Factor (SF),
Weight-of- Evidence
(WOE)
Classification
Inhalation:
RfC, NOAEL,
or LOAEL a
Oral/Dermal:
RfD, NOAEL,
or LOAEL a
SAT
Rank
Not enough information to identify a specific chemical.
WOE (for copper)
WOE (for copper)
WOE (for copper)
WOE
WOE
SF, WOE
WOE
WOE (for nickel
dust)
LOAEL
Yes
LOAEL
Yes
NOAEL
LOAEL; D-
NOAEL
RfC
RfC
Other b
Yes
Yes
Yes
Yes
Other b
MRLd
Yes
LOAEL; D-LOAEL
Yes; D-LOAEL
LOAEL; D-LOAEL
Yes
RfD; D-LOAEL
RfD; D-NOAEL
RfD; D-NOAEL
Yes
LOAEL
Yes
Yes
Yes
Yes
Other b
RfD
X
X
X
X
X
X
X
X
X
X
X
X
Not enough information to identify specific chemical.
Some data (for Pd)
Some data (for Pd)
RfC
ADIe
RfDf
X
X
3-94
-------
Chemical
Propionic acid
Quantenary alkylammonium
chlorides
Silver salt
Silver nitrate
Sodium benzene sulfonate
Sodium hydroxide
Sodium hypophosphite
Sodium hypophosphite mono
hydrate
Sodium phosphorus salt
Sodium salt 8
Stannous methane sulfonic
acid
Substituted amine
hydrochloride
Sulfuric acid
Surfactant
Thiourea
Tin
Tin chloride
Transition metal salt
Unspecified tartrate
Urea
Urea compound B
Urea compound C
Vinyl polymer
Cancer:
Slope Factor (SF),
Weight-of- Evidence
(WOE)
Classification
Some data
WOE (for silver)
WOE (for silver)
WOE
WOE
Inhalation:
RfC, NOAEL,
or LOAEL a
Other c
Other c
Oral/Dermal:
RfD, NOAEL,
or LOAEL a
NOAEL
Yes
RfD (for silver)
RfD (for tin)
SAT
Rank
X
X
X
X
X
X
Not enough information to identify specific chemical.
WOE
WOE
WOE
WOE
RfD
RfD
Yes
Yes
Yes
X
"Yes" indicates a value is available (RfC or RfD, NOAEL or LOAEL) but the type of toxicity measure is not
specified in order to protect confidential ingredient identity. D-NOAEL/or D-LOAEL: Developmental NOAEL or
LOAEL available.
b Toxicity data other than RfD, NOAEL or LOAEL were used; see Tables 3-22 and 3-23 for details.
c Generally recognized as safe (GRAS) by the U.S. Food & Drug Administration (HSDB, 1995).
d MRL = minimal risk level.
e ADI = allowable daily intake.
f These values are only relevant to the oral route; potassium gold cyanide is expected to be chemically stable except
under highly acidic conditions such as those found in the stomach (pH 1-2).
g Not generally considered poisonous to humans or animals.
3-95
-------
3.3.3 Ecological Hazard Summary
Ecological hazards data are presented in two ways: through a CC and an aquatic hazard
concern level, each derived separately from aquatic toxicity data (fish, invertebrates, and algae).
Hazards to terrestrial species were not assessed because sufficient toxicity data were not
available. CCs are based on the most sensitive endpoint, modified by an assessment factor,
which reflects the amount and quality of toxicity data available for that chemical. CCs are
compared to estimated surface water concentrations as part of the Risk Characterization (Section
3.4). Aquatic hazard concern levels are based on where the lowest available toxicity value (i.e.,
the most sensitive endpoint) fits into pre-defined ranges of values, indicating relative toxicity
when compared to other chemicals.
Concern Concentration
Table 3-27 presents a summary of the available ecological hazards information. CCs were
determined for aquatic species (e.g., Daphnia, algae, and/or fish) using standard EPA
methodology. The method for determining CCs is summarized below and presented in more
detail in Appendix H.
Table 3-27. Estimated (Lowest) Aquatic Toxicity Values and Concern Concentrations for
PWB Surface Finishing Chemicals, Based on Measured Test Data or SAR Analysis
Chemical
1,4-Butenediol
Acetic acid
Aliphatic acid A
Aliphatic acid B
Aliphatic acid D
Aliphatic acid E
Aliphatic dicarboxylic acid A
Aliphatic dicarboxylic acid C
Alkylalkyne diol
Alkylamino acid A
Alkylamino acid B
Alkylaryl imidazole
Alkylaryl sulfonate
Alkyldiol
Alkylimine dialkanol
Alkylphenol ethoxylate
Alkylphenol
polyethoxyethanol
Alkylpolyol
Acute (a) Toxicity
(mg/L)
Fish Invert Algae
0.5
79 65
Chronic (c) Toxicity
(mg/L)
Fish Invert Algae
0.08
data omitted a
data omitted a
data omitted a
data omitted a
data omitted a
data omitted a
data omitted a
data omitted a
data omitted a
data omitted a
data omitted a
data omitted a
data omitted a
data omitted a
16 16 20
225
data omitted a
Concern
Concentration
(mg/L)
0.008 (c)
0.65 (a)
0.5 - 1 (a)
1- 5 (c)
5 - 10 (c)
>l(c)
>l(c)
>10
0.1 -0.5 (c)
500 - 1,000 (c)
0.1 -5 (c)
0.001 - 0.005 (c)
0.001 - 0.005 (c)
10 - 50 (c)
0.001 - 0.005 (c)
0.1 -0.5 (c)
0.2 (c)
5 - 10 (c)
3-96
-------
Chemical
Amino acid salt
Amino carboxylic acid
Ammonia compound A
Ammonia compound B
Ammonium chloride
Ammonium hydroxide
Arylphenol
Bismuth compound
Citric acid In soft water
In hard water
Copper ion
Copper salt C
Copper sulfate pentahydrate
Cyclic amide
Ethoxylated alkylphenol
Ethylenediamine
Ethylene glycol
Ethylene glycol monobutyl
ether b
Fatty amine
Fluoboric acid
Gum
Hydrochloric acid
Hydrogen peroxide
Hydroxyaryl acid
Hydroxyaryl sulfonate
Hydroxy carboxylic acid
Inorganic metallic salt A
Inorganic metallic salt B
Inorganic metallic salt C
Lead
Maleic acid
Malic acid
Methane sulfonic acid
Nickel sulfate
Nitrogen acid
Palladium chloride
Acute (a) Toxicity
(mg/L)
Fish
Invert
Algae
Chronic (c) Toxicity
(mg/L)
Fish
Invert
Algae
data omitted a
data omitted a
data omitted a
data omitted a
725
12
161
32
>30
1
3
>3
data omitted a
data omitted a
>100
0.14
>100
12.8
5
100
>10
>10
1
30
data omitted a
0.34
0.3
0.00002
0.022
0.0014
0.062
data omitted a
data omitted a
220
10,000
116
26.5
6,900
89
>100
31,000
620
5,400
10
0.16
710
3.9
8.3
440
32
data omitted a
>1,000
560
160
20
70
1.4
data omitted a
70
5.9
100
4.3
345
1.7
63
16
15
data omitted a
data omitted a
data omitted a
data omitted a
data omitted a
data omitted a
315
5,227
2,860
g/L
>1,000
1.28
143
1,199
2,380
g/L
>1,000
2.58
500
30,654
1,200
g/L
>1,000
1.9
4.1
204,000
>100
30
24,378
>100
993
14,339
>100
data omitted a
1,584
1,567
917
170
49
47
Concern
Concentration
(mg/L)
0.5 - 1 (c)
5 - 10 (c)
1 - 5 (a)
0.01 - 0.05 (c)
1.6 (a)
0.1 (c)
0.01 - 0.05 (c)
0.1 -0.5 (c)
0.1 (c)
3.0 (c)
0.001 (a)
0.005 - 0.01(c)
0.01 (c)
10 - 50 (c)
0.1 -0.5 (c)
0.02 (c)
44 (c)
0.04 (c)
0.001 - 0.005 (c)
0.14(c)
0.5 - 1 (c)
1.5 (c)
0.02 (a)
0.1 -0.5 (c)
1 - 5 (c)
1 - 5 (c)
0.0001-0.0005 (c)
0.001 - 0.005 (c)
0.001 - 0.005 (c)
0.41 (c)
99.3 (c)
1,434 (c)
10 (c)
0.01 (a)
1 - 5 (c)
4.7 (c)
3-97
-------
Chemical
Palladium salt
Phosphoric acid
Potassium compound
Potassium gold cyanide
Potassium peroxymonosulfate
Propionic acid
Quantenary alkylammonium
chlorides
Silver nitrate
Silver salt
Sodium benzene sulfonate
Sodium hydroxide
Sodium hypophosphite and
Sodium hypophosphite
monohydrate
Sodium phosphorus salt
Sodium salt
Stannous methane sulfonic
acid
Substituted amine
hydrochloride
Sulfuric acid
Thiourea
Tin
Tin chloride
Transition metal salt
Unspecified tartrate
Urea
Urea compound B
Urea compound C
Vinyl polymer
Acute (a) Toxicity
(mg/L)
Fish
Invert
Algae
Chronic (c) Toxicity
(mg/L)
Fish
Invert
Algae
data omitted a
1,751
25,817
13,761
2,405
394
278
data omitted a
>0.6
<1
1,369
>2
<3
587
>0.4
<3
6,644
>0.06
<0.1
1,216
>0.03
<0.3
318
>0.1
<1
292
data omitted a
0.007
0.0007
0.13
0.001
0.005
data omitted a
data omitted a
133,000
199,000
g/L
191,000
g/L
1,330
g/L
3,180
g/L
55,700
g/L
498,000
8,430
g/L
22,658
331,000
10,616
103,000
data omitted a
data omitted a
7
140
<8
0.2
0.9
<0.8
data omitted a
42
>100
2.7
1.89
5,200
g/L
9
55
19.5
250,000
4.8
<3
0.2
600,000
>60
0.07
0.4
4,222
0.9
0.35
42
2,241
0.3
<0.3
data omitted a
data omitted a
>1,000
>1,000
>1,000
>100
>100
>100
data omitted a
data omitted a
data omitted a
Concern
Concentration
(mg/L)
1 - 5 (c)
27.8 (c)
1,000 - 1,500 (c)
0.003 (c)
0.01 (c)
29.2 (c)
0.01 - 0.05 (c)
0.0001 (c)
0.0001 - 0.0005 (c)
>l(c)
1,062 (c)
10,300 (c)
10,000 - 50,000 (c)
50 - 100 (c)
0.02 (c)
0.01 - 0.05 (c)
224 (c)
0.03(c)
0.007 (c)
0.04 (c)
<1 - 5 (c)
1 - 5 (c)
>10 (c)
0.01 - 0.05 (c)
0.01 - 0.05 (c)
>1 -5
a Data omitted from table and a range reported for CC in order to protect identity of confidential ingredients.
b Diethylene glycol monobutyl ether reviewed instead; both chemicals are very similar.
The CC for each chemical in water was calculated using the general equation:
CC = acute or chronic toxicity value + UF
3-98
-------
where,
CC = aquatic toxicity concern concentration, the concentration of a chemical in the
aquatic environment below which no significant risk to aquatic organisms is
expected
UF = uncertainty factor, the adjustment value used in the calculation of a CC that
incorporates the uncertainties associated with: 1) toxicity data (e.g., laboratory
test versus field test, measured versus estimated data); 2) acute exposures versus
chronic exposures; and 3) species sensitivity. This factor is expressed as an order
of magnitude or as a power often (U.S. EPA, 1984).
If several acute or chronic toxicity values are available, the lowest one is used (most
sensitive tested species), unless poor or uncertain data quality disqualify one or more of the
values. UFs are dependent on the amount and type of toxicity data contained in a toxicity profile
and reflect the amount of uncertainty about the potential effects associated with a toxicity value.
In general, the more complete the toxicity profile and the greater the quality of the toxicity data,
the smaller the UF used.
The following approach was used, depending on availability and type of data:
• If the toxicity profile only contained one or two acute toxicity values (no chronic values),
UF = 1,000 and the CC was calculated by using the lower acute value.
• If the toxicity profile contained three or more acute values (no chronic values), UF = 100
and the CC was calculated by using the lowest acute value.
• If the toxicity profile contained at least one chronic value, and the value was for the most
sensitive species, UF = 10 and the CC was calculated by using the lowest chronic value;
otherwise, UF = 100 and the CC was calculated with the acute value for the most sensitive
species.
Hazard Concern Levels
Table 3-28 presents aquatic hazard concern levels; chemicals were assigned to aquatic
toxicity concern levels according to the following EPA criteria:
For chronic values:
< 0.1 mg/L High concern
> 0.1 to < 10 mg/L Moderate concern
> 10 mg/L Low concern
For acute values:
< 1 mg/L High concern
> 1 to < 100 mg/L Moderate concern
> 100 mg/L Low concern
Chronic toxicity ranking takes precedence over the acute ranking.
3-99
-------
Most surface finishing chemicals can theoretically be subject to spills and releases. Also,
PWB facilities routinely release wastewater to POTWs. Different geographic regions and
different POTWs have different levels of acceptability for such wastes, and the acceptable levels
can change over time. Discontinuing use of chemicals in Table 3-28 with Medium to High hazard
concern levels can help avoid potential problems.
Table 3-28. Environmental Hazard Ranking of PWB Finishing Chemicals
Chemical
1,4-Butenediol
Acetic acid
Aliphatic acid A
Aliphatic acid B
Aliphatic acid D
Aliphatic acid E
Aliphatic dicarboxylic acid A
Aliphatic dicarboxylic acid C
Alkylalkyne diol
Alkylamino acid A
Alkylamino acid B
Alkylaryl imidazole
Alkylaryl sulfonate
Alkyldiol
Alkylimine dialkanol
Alkylphenol ethoxylate
Alkylphenol polyethoxyethanol
Alkylpolyol
Amino acid salt
Amino carboxylic acid
Ammonia compound A
Ammonia compound B
Ammonium chloride
Ammonium hydroxide
Arylphenol
Bismuth compound
Citric acid
Copper ion
Copper salt C
Copper sulfate pentahydrate
Cyclic amide
Ethoxylated alkylphenol
Lowest Acute (a) or
Chronic (c) Value
(mg/L)
0.08 (c)
65 (a)
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
0.008 (c) to 2 (c)
NR
NR
NR
NR
NR
161(a)
l(c)
NR
NR
l(c)
0.14 (a)
NR
O.OOl(c)
NR
NR
Hazard
Rank3
H
L
L
L
L
L
L
L
M
L
M
H
H
L
H
MtoHb
MtoHb
L
L
L
L
H
L
M
M
M
M
H
H
H
L
MtoHb
3-100
-------
Chemical
Ethylenediamine
Ethylene glycol
Ethylene glycol monobutyl ether °
Fatty amine
Fluoboric acid
Gum
Hydrochloric acid
Hydrogen peroxide
Hydroxyaryl acid
Hydroxy aryl sulfonate
Hydroxy carboxylic acid
Inorganic metallic salt A
Inorganic metallic salt B
Inorganic metallic salt C
Lead
Maleic acid
Malic acid
Methane sulfonic acid
Nickel sulfate
Nitrogen acid
Palladium chloride
Palladium salt
Phosphoric acid
Potassium compound
Potassium gold cyanide
Potassium peroxymonosulfate
Propionic acid
Quantenary alkylammonium chlorides
Silver nitrate
Silver salt
Sodium benzene sulfonate
Sodium hydroxide
Sodium hypophosphite and Sodium hypophosphite
monohydrate
Sodium phosphorus salt
Sodium salt
Stannous methane sulfonic acid
Substituted amine hydrochloride
Sulfuric acid
Lowest Acute (a) or
Chronic (c) Value
(mg/L)
0.16(c)
440 (c)
3.9(c)
NR
1.4(c)
NR
15 (c)
1.7 (a)
NR
NR
NR
NR
NR
NR
4.1 (c)
993 (c)
14,339 (c)
>100 (c)
1.3 (a)
NR
47 (c)
NR
278 (c)
NR
>0.03 (c)
<0.1 (c)
292 (c)
NR
0.001 (c)
NR
NR
10,616 (c)
103,000 (c)
NR
NR
0.2 (c)
NR
2,241 (c)
Hazard
Rank3
M
L
M
H
M
L
M
M
M
L
L
H
H
H
M
L
L
L
M
L
L
L
L
L
H
H
L
M
H
H
L
L
L
L
L
M
M
L
3-101
-------
Chemical
Thiourea
Tin
Tin chloride
Transition metal salt
Unspecified tartrate
Urea
Urea compound B
Urea compound C
Vinyl polymer
Lowest Acute (a) or
Chronic (c) Value
(mg/L)
0.3 (c)
0.07 (c)
0.4 (c)
NR
NR
>100 (c)
NR
NR
NR
Hazard
Rank3
M
H
M
M
L
L
M
M
L
Ranking based on the lowest estimated acute or chronic value; H = high, M = medium, L = low.
b Toxicity of breakdown product results in high hazard rank.
0 Diethylene glycol monobutyl ether reviewed instead; both chemicals are very similar.
NR: Not reported in order to protect confidential ingredient identity.
3.3.4 Summary
For human health hazards, toxicity data in the form of RfDs, RfCs, NOAELs, LOAELs,
and cancer slope (cancer potency) factors were compiled for inhalation and dermal pathways.
Inorganic metallic salt A (a confidential ingredient used in the nickel/gold process) was the only
chemical with an established cancer slope (cancer potency) factor. Other chemicals in the surface
finishing processes are carcinogens or suspected carcinogens, but do not have established slope
factors. Strong inorganic acid mist of sulfuric acid has been determined by IARC to be a human
carcinogen (IARC Group 1). Sulfuric acid is used in every surface finishing process in this
evaluation. It is not expected, however, to be present as a strong acid mist because it is greatly
diluted in the aqueous baths. Lead and thiourea have been determined by IARC to be possible
human carcinogens (IARC Group 2B) and lead has also been classified by EPA as a probable
human carcinogen (EPA Class B2). Lead is used in tin-lead solder in the HASL process.
Thiourea is used in the immersion tin process. Urea compound B, a confidential ingredient in the
nickel/gold and nickel/palladium/gold processes, is possibly carcinogenic to humans.
A total of 83 chemicals are considered as part of the surface finishing use cluster. For
non-cancer health effects, eight surface finishing chemicals have inhalation RfCs available from
which to calculate hazard quotient (HQ) in the risk characterization. For the remaining chemicals,
12 have an inhalation NOAEL or LOAEL from which to calculate margin of exposure (MOE).
Pertaining to dermal exposure, 12 surface finishing chemicals have RfDs from which to calculate
HQs; of the remaining chemicals, 19 have an oral NOAEL or LOAEL from which to calculate
MOE. For a number of chemicals, no quantitative risk indicator could be calculated for direct
comparison of risk among alternatives. A qualitative assessment was done for 33 chemicals,
based on chemical structure, for which no quantitative non-cancer health effects measures were
available.
3-102
-------
An ecological hazards assessment was performed based on chemical toxicity to aquatic
organisms. CCs were estimated for surface finishing chemicals using an established EPA
method. A CC is an acute or chronic toxicity value divided by a UF. UFs are dependent on the
amount and type of toxicity data contained in a toxicity profile and reflect the amount of
uncertainty about the potential effects associated with a toxicity value. CCs were determined for
aquatic species (e.g., Daphnia, algae, and/or fish). CCs are compared to estimated surface water
concentrations modeled from PWB wastewater releases in Section 3.4.
Chemicals were also ranked for aquatic toxicity concern levels using established EPA
criteria (high, moderate, and low concern) based on the available toxicity data. The number of
chemicals with a high aquatic hazard concern level include eight in the HASL process, nine in
nickel/gold, five in nickel/palladium/gold, five in OSP, three in immersion silver, and six in the
immersion tin process.
3-103
-------
3.4 RISK CHARACTERIZATION
Risk characterization integrates the hazard and exposure components of a risk evaluation
and presents overall conclusions. Risk characterization typically includes a description of the
assumptions, scientific judgments, and uncertainties that are part of this process. The focus of
this risk characterization is on chronic (long-term) exposure to chemicals that may cause cancer
or other toxic effects, rather than on acute toxicity from brief exposures to chemicals. The focus
is also on those health effects from chronic exposures that could be used to measure risk. From
an ecological risk standpoint, the focus is on chronic exposure to chemicals that cause sublethal
effects (e.g., effects on growth and reproduction). The Process Safety Assessment (Section 3.5)
includes further information on chemical safety concerns for workers.
The goals of the PWB project risk characterization are to:
• present conclusions and uncertainties associated with a screening-level health risk
assessment of chemicals used in the surface finishing process of PWB manufacture;
• integrate chemical hazard and exposure information to assess potential risks from ambient
environment and occupational exposures from the surface finishing process;
• use reasonable and consistent assumptions across alternatives, so potential health risks
associated with one alternative can be compared to the potential health risks associated
with other alternatives; and
• identify the areas of concern that differ among the substitutes in a manner that facilitates
decision-making.
This section contains a summary of the exposure assessment (Section 3.4.1), a summary
of the human health hazards assessment (Section 3.4.2), and the ecological hazards assessment
(Section 3.4.3), a description of methods used to calculate risk indicators (Section 3.4.4), potential
human health risk results (Section 3.4.5), an evaluation of lead risks from tin-lead solder used in
the HASL process (Section 3.4.6), ecological (aquatic) risk results (Section 3.4.7), a discussion of
uncertainties (Section 3.4.8), and conclusions (Section 3.4.9). Detailed exposure and hazard data
are presented separately in the Exposure Assessment (Section 3.2) and Human Health and
Ecological Hazards Summary (Section 3.3), respectively.
3.4.1 Summary of Exposure Assessment
The exposure assessment uses a "model facility" approach where, as much as possible,
reasonable and consistent assumptions are used across alternatives. Data to characterize the
model facility and exposure patterns for each process alternative were aggregated from a number
of sources, including PWB shops in the U.S. and abroad, supplier data, and input from PWB
manufacturers at project meetings. Thus, the model facility is not entirely representative of any
one facility, and actual exposure (and risk) could vary substantially, depending on site-specific
operating conditions and other factors.
3-104
-------
Chemical exposures to PWB workers and the general population were estimated by
combining information gathered from industry (PWB Workplace Practices Questionnaire and
Performance Demonstration data, MSDSs, other information provided by product suppliers, and
other available information) with standard EPA exposure assumptions (e.g., for inhalation rate,
surface area of dermal contact, and other parameters). The pathways for which potential
exposure from surface finishing process baths was quantified include inhalation and dermal
contact for workers, inhalation for the general population living near a PWB facility, and contact
with aquatic organisms living in a stream that receives treated wastewater originating from a PWB
facility. Acute impacts, such as impacts from chemical spills, are not addressed due to the pre-
defined scope of this assessment.
Inhalation exposure could occur by breathing air containing vapor or aerosol-phase
chemicals from the surface finishing process line. Inhalation exposures to workers from non-
conveyorized lines are estimated in the exposure assessment. Inhalation exposure to workers
from conveyorized surface finishing lines is much lower than for non-conveyorized lines because
the lines are typically enclosed and vented to the outside.15 The model used to estimate daily
inhalation exposure is from the EPA Chemical Engineering Branch Manual for the Preparation
of'Engineering Assessments (U.S. EPA, 199la):
I = (Cm)(b)(h)
where,
I = daily inhalation potential dose rate (mg/day)
Cm = airborne concentration of substance (mg/m3)
b = inhalation rate (m3/hr)
h = duration (hr/day)
Daily exposures are then averaged over a lifetime (70 years) for carcinogens, and over the
exposure duration (e.g., 25 years working in a facility) for non-carcinogens,16 using the following
equations:
For carcinogens:
15 Inhalation exposures for conveyorized process configurations were initially assumed to be negligible, and are
not presented separately here. Some inhalation exposure is possible, however, during sampling and bath replacement,
when the baths are opened for a short period of time. After characterizing risks from inhalation for non-conveyorized
lines, inhalation exposures and risks were estimated for the subset of inhalation chemicals of concern for
conveyorized lines. No chemical exposures from inhalation resulted in risks above concern levels for conveyorized
lines.
16 Different averaging times are used for characterizing risk for carcinogenic and non-carcinogenic effects. For
carcinogenic agents, because even a single incidence of exposure is assumed to have the potential to cause cancer
throughout an individual's lifetime, the length of exposure to that agent is averaged over a lifetime. An additional
factor is that the cancer latency period may extend beyond the period of working years before it is discernible. For
chemicals exhibiting non-cancer health effects from chronic (longer-term) exposure, where there is an exposure
threshold (a level below which effects are not expected to occur), only the time period when exposure is occurring is
assumed to be relevant and is used as the averaging time.
3-105
-------
LADD = (I)(EF)(ED)/[(BW)(ATCAR)]
For non-carcinogens:
ADD = (I)(EF)(ED)/[(BW)(ATNC)]
where,
LADD = lifetime average daily dose (mg/kg-day)
ADD = average daily dose (mg/kg-day)
EF = exposure frequency (days/year)
ED = exposure duration (years)
BW = body weight (kg)
ATCAR = averaging time for carcinogenic effects (days)
ATNC = averaging time for non-carcinogenic chronic effects (days)
The daily intake for inhalation exposure to workers was calculated by first modeling
chemical emissions from surface finishing baths with three air-transport mechanisms: liquid
surface diffusion (desorption), bubble desorption, and aerosol generation and ejection. This
modeled chemical emission rate was combined with data from the PWB Workplace Practices
Questionnaire and Performance Demonstration Data Sheets regarding process room size and air
turnover rate to estimate an average indoor air concentration for the process area.
Modeled air concentrations were used to evaluate inhalation exposure to a nearby
population. This outdoor air modeling used the air emission rates that were estimated for the
process baths, assuming they are vented outside at the same rate they are emitted from the baths.
The Industrial Source Complex - Long Term (ISCLT) air dispersion model17 was used to estimate
air concentrations resulting from dispersion in the outdoor air. The modeled air concentrations of
each contaminant were determined at 100 meters radially from a PWB facility. The highest
estimated air concentration was used to estimate inhalation exposure to a hypothetical population
located near a model PWB facility. Inhalation exposures estimated for the public living 100
meters away from a PWB facility were very low (approximately 10,000 times lower than
occupational exposures).
Dermal exposure could occur when skin comes in contact with the bath solution while
dipping boards, adding bath replacement chemicals, etc. Although the data suggest that most
surface finishing line operators wear gloves for many activities, it was assumed in this evaluation
that workers do not wear gloves, to account for the fraction that do not. Otherwise, dermal
exposure is expected to be negligible. For dermal exposures, the flux of a material through the
skin was estimated based on U.S. EPA, 1992a:
D = (S)(C)(f)(h)(0.001)
17 This version of the ISCLT model is provided as part of the Risk* Assistant™ 2.0 software package (Hampshire
Research Institute, 1995).
3-106
-------
where,
D = dermal potential dose rate (mg/day)
S = surface area of contact (cm2)
C = concentration of chemical in the bath (mg/L)
f = flux through skin (cm/hour)
h = duration (hours/day) with a conversion factor of 0.001 (L/cm3)
It should be noted that the above equation was developed for exposures with an infinite
volume of liquid or boundary layer contacting the skin, such as swimming or bathing.
Occupational conditions of dermal contact are likely to be more finite in comparison, resulting in
possible overestimates of flux through the skin when using the above equation.
Similar to inhalation, daily dermal exposures were then averaged over the exposure
duration for non-carcinogens (cancer risk was not quantified because none of the surface
finishing chemicals have an oral or dermal cancer slope factor) using the following equation:
ADD = (D)(EF)(ED)/[(BW)(ATNC)]
For dermal exposure, the concentration of chemical in the bath and duration of contact
for workers was obtained from publicly-available bath chemistry data, disclosed proprietary
chemical information, supplier data sheets, and PWB Workplace Practices Questionnaire
information. A permeability coefficient (rate of penetration through skin) was estimated for
organic compounds and a default rate assumption was used for inorganic chemicals. Reliance on
such estimates in the absence of data is a source of uncertainty in the exposure assessment.
Key assumptions in the exposure assessment include the following:
• The exposure frequency (i.e., days/year of line operation) was based on the time required
to manufacture 260,000 ssf of PWB.
• For dermal exposure, it was assumed that line operators do not wear gloves. Although
the data suggest that many surface finishing line operators do wear gloves for various
activities, it was assumed for this evaluation that workers do not wear gloves, to account
for the subset of workers who do not wear proper personal protective equipment.
• For dermal exposure, it was assumed that all non-conveyorized lines are manual hoist.
• The worker on a non-conveyorized line is assumed to potentially have dermal contact for
the entire time spent in the surface finishing process area, and the contact time is assumed
to be divided equally among the baths over an 8-hour workday. This does not mean that
a worker has both hands immersed in a bath for that entire time but that the skin is in
contact with bath solution (i.e., the hands may remain wet from contact). This
assumption may result in an overestimate of dermal exposure.
• For estimating ambient (outdoor) air concentrations, it was assumed that no air pollution
control technologies are used to remove airborne chemicals from facility air prior to
venting it to the outside.
3-107
-------
• For inhalation exposure to workers, it was assumed that chemical emissions to air in the
process room from conveyorized lines are negligible, and that no vapor control devices
(e.g., bath covers) are used on baths in non-conveyorized lines.
• For air concentrations, the model assumes complete mixing in the process room and that
concentrations do not change with time (i.e., steady state).
• For all exposures, it was assumed that there is one surface finishing process line and one
line operator per shift in a process area.
• For characterizing the chemical constituents in the surface finishing process baths, it was
assumed that the form (speciation) and concentration of all chemicals in the baths are
constant over time.
Chemical concentrations in baths are based on publicly-available chemistry data,
including MSDSs, proprietary chemical information, and supplier Product Data Sheets that
describe how to mix and maintain chemical baths. Many MSDSs provided concentration ranges
for chemical constituents instead of absolute concentrations, in which case it was assumed that a
chemical is present at the mid-point of the reported concentration range. This assumption may
either overestimate or underestimate risk for chemicals, depending on their actual concentrations.
Assumptions and parameter values used in these equations, and results of the exposure
calculations, are presented in the Exposure Assessment (Section 3.2). In order to provide
information about the position an exposure estimate has in the distribution of possible outcomes,
exposure (or risk) descriptors are used following EPA's Guidelines for Exposure Assessment
(U.S. EPA, 1992b). For this risk characterization, whenever possible the exposure assessment
uses a combination of central tendency (either an average or median estimate) and high-end (90th
percentile)18 assumptions, as would be used for an overall high-end exposure estimate. The 90th
percentile is used for:
• hours per day of workplace exposure;
• exposure frequency;
• exposure duration in years (90th percentile for occupational and 95th percentile for
residential exposures);
• time required for chemical bath replacement;
• time and frequency of filter replacements, conveyor equipment cleaning, and chemical
bath sampling (minutes per occurrence and number of occurrences per year); and
• estimated workplace air concentrations.
Average values are used for:
• body weight;
• concentration of chemical in bath;
18 For exposure data from the PWB Workplace Practices Questionnaire, this means that 90 percent of the
facilities reported a lower value, and ten percent reported a higher value.
3-108
-------
• frequency of chemical bath replacements;
• number of baths in a given process; and
• bath size.
Some values used in the exposure calculations, however, are better characterized as
"what-if," especially pertaining to use of gloves, process area ventilation rates, and production
times (days/year) required to manufacture 260,000 ssf of PWB for the model facility. ("What-if
represents an exposure estimate based on postulated questions, making assumptions based on
limited data where the distribution is unknown.) Because some part of the exposure assessment
for both inhalation and dermal exposures qualifies as a "what-if descriptor, the entire
assessment should be considered "what-if."
3.4.2 Summary of Human Health Hazards Assessment
For human health hazards, toxicity data in the form of RfDs, RfCs, NOAELs, LOAELs,
and cancer slope (cancer potency) factors were compiled for inhalation and dermal pathways.
Inorganic metallic salt A (a confidential ingredient used in the nickel/gold process) was the only
chemical with an established cancer slope (cancer potency) factor. Other chemicals in the surface
finishing processes are known or suspected carcinogens, but do not have established slope
factors. Strong inorganic acid mist of sulfuric acid has been determined by IARC to be a human
carcinogen (IARC Group 1). Sulfuric acid is used in every surface finishing process in this
evaluation. It is not expected, however, to be present as a strong acid mist because it is greatly
diluted in the aqueous baths. Lead and thiourea have been determined by IARC to be possible
human carcinogens (IARC Group 2B) and lead has also been classified by EPA as a probable
human carcinogen (EPA Class B2). Lead is used in tin-lead solder in the HASL process.
Thiourea is used in the immersion tin process. Urea compound B, a confidential ingredient in the
nickel/gold and nickel/palladium/gold processes, is possibly carcinogenic to humans.
3.4.3 Summary of Ecological Hazards Assessment
An ecological hazard assessment was performed based on chemical toxicity to aquatic
organisms. CCs were estimated for surface finishing chemicals using an established EPA method
(see Table 3-27 and Appendix H). A CC is an acute or chronic toxicity value divided by a UF.
UFs are dependent on the amount and type of toxicity data contained in a toxicity profile, and
reflect the amount of uncertainty about the potential effects associated with a toxicity value.
Concern concentrations were determined for aquatic species (e.g., Daphnia, algae, and/or fish)
for each chemical. The lowest CCs are for inorganic metallic salt A, silver nitrate, and silver salt.
Chemicals also were ranked for aquatic toxicity concern levels using established EPA criteria
(high, moderate, and low concern) based on the available toxicity data (see Table 3-28). The
number of chemicals with a high aquatic hazard concern level include eight in the HASL process,
nine in nickel/gold, five in nickel/palladium/gold, five in OSP, three in immersion silver, and six in
the immersion tin process.
3-109
-------
3.4.4 Methods Used to Calculate Human Health Risks
Estimates of potential human health risk from chemical exposure are characterized here in
terms of excess lifetime cancer risk, HQ, and MOE. This section defines these risk indicators and
discusses the methods for calculating each of them.
Cancer Risk
Cancer risks are expressed as the excess probability of an individual developing cancer
over a lifetime from chemical exposure. For chemicals classified as carcinogens, an upper bound
excess lifetime cancer risk, expressed as a unitless probability, was estimated by the following
equation:
Cancer Risk = LADD x slope factor
where,
Cancer Risk = the excess probability of developing cancer over a lifetime as a result of
exposure to a potential carcinogen. The estimated risks are the upper
bound excess lifetime cancer risks for an individual. {Upper bound refers
to the method of determining a slope factor, where the upper bound value
for the slope of the dose-response curve is used. Excess means the
estimated cancer risk is in addition to the already-existing background risk
of an individual contracting cancer from all other causes.)
LADD = the lifetime average daily dose, the estimated potential daily dose rate
received during the exposure duration, averaged over a 70-year lifetime (in
mg/kg-day). LADDs were calculated in the Exposure Assessment
(Section 3.2).
Slope factor (qj *) is defined in Section 3.3.1.
Non-Cancer Risk Indicators
Non-cancer risk estimates are expressed either as an HQ or as an MOE, depending on
whether or not RfDs and RfCs are available. There is a higher level of confidence in the HQ than
the MOE, especially when the HQ is based on an RfD or RfC that has been peer-reviewed by
EPA (as with data from the EPA IRIS database). If an RfD or RfC is available, the HQ is
calculated to estimate risk from chemicals that exhibit chronic, non-cancer toxicity. (RfDs and
RfCs are defined in Section 3.3.2.) The HQ is the unitless ratio of the RfD (or RfC) to the
potential dose rate. For surface finishing chemicals that exhibit non-cancer toxicity, the HQ was
calculated by:
HQ = ADD/RfD
where,
3-110
-------
ADD = average daily dose rate, the amount of a chemical ingested, inhaled, or applied to
the skin per unit time, averaged over the exposure duration (in mg/kg-day)
ADDs were calculated in the Exposure Assessment (Section 3.2).
The HQ is based on the assumption that there is a level of exposure (i.e., the RfD or RfC)
below which it is unlikely, even for sensitive subgroups, to experience adverse health effects.
Unlike cancer risk, the HQ does not express probability and is not necessarily linear; that is, an
HQ often does not mean that adverse health effects are ten times more likely to occur than for an
HQ of one. However, the ratio of estimated dose to RfD/RfC reflects the level of concern.
For chemicals where an RfD or RfC was not available, an MOE was calculated by:
MOE = NOAEL/ADD or LOAEL/ADD
As with the HQ, the MOE is not a probabilistic statement of risk. The ratio for calculating MOE
is the inverse of the HQ, so that a high HQ (exceeding one) indicates a potential concern, whereas
a high MOE (exceeding 100 for a NOAEL-based MOE or 1,000 for a LOAEL-based MOE)
indicates a low concern level. (NOAELS and LOAELs are defined in Section 3.3.2.) As the
MOE increases, the level of concern decreases. (As the HQ increases, the level of concern also
increases.) In general, there is a higher level of confidence for HQs than for MOEs because the
toxicity data on which RfDs and RfCs are based have passed a more thorough level of review,
and test-specific uncertainty factors have been included.
Both the exposure estimates and toxicity data are specific to the route of exposure (i.e.,
inhalation, oral, or dermal). Very few RfDs, NOAELs, or LOAELs are available for dermal
exposure. If oral data were available, the following adjustments were made to calculate dermal
values based on EPA (1989) guidance:
iJ (GI absorption)
NOAEL/LOAELDER = (NOAEL or LOAEL0RAL) (GI absorption)
SFDER = (SF0RAL)/(GI absorption)
where,
RfDDER = reference dose adjusted for dermal exposure (mg/kg-day)
NOAEL/LOAELDER = NOAEL or LOAEL adjusted for dermal exposure (mg/kg-day)
SFDER = cancer slope factor adjusted for dermal exposure (mg/kg-day)"1
GI absorption = gastrointestinal absorption efficiency
This adjustment is made to account for the fact that the oral RfDs, NOAELs, and
LOAELs are based on an applied dose, while dermal exposure represents an estimated absorbed
dose. The oral RfDs, NOAELs, and LOAELs used to assess dermal risks therefore were adjusted
using GI absorption to reflect an absorbed dose. Table 3-29 lists the GI absorption data for
chemicals used in calculating risk from dermal exposure. (Data for some proprietary ingredients
are not presented in order to protect confidential chemical identities.)
3-111
-------
Table 3-29. Gastrointestinal (GI) Absorption Factors
Chemicals a
Acetic acid
Aliphatic acid A
Aliphatic acid D
Aliphatic dicarboxylic acid C
Alkyldiol
Alkylpolyol
Amino carboxylic acid
Ammonia compound A
Ammonia compound B
Ammonium chloride
Ammonium hydroxide
Aryl phenol
Copper ion, Copper salt C, and
Copper sulfate pentahydrate
Cyclic amide
Ethylene glycol
Ethylene glycol monobutyl ether
Ethylenediamine
Hydroxy carboxylic acid
Hydrogen peroxide
Inorganic metallic salt A
Inorganic metallic salt B
Inorganic metallic salt C
Nickel sulfate
Phosphoric acid
Potassium gold cyanide
Propionic acid
Silver nitrate
Silver salt
Stannous methane sulfonic acid
Tin chloride
Unspecified tartrate
GI Absorption Factor
0.9
0.9
0.5
0.2
NR
0.2
0.2
0.9
0.9
0.9
0.9
0.5
0.6
0.5
0.5
0.5
0.78
0.2
0.2
NR
0.15
0.15
0.05
0.2
0.2
0.2
0.08
NR
0.2
0.5
0.5
Source
chemical profile b
chemical profile b
NR
assumption °
NR
assumption °
assumption °
chemical profile b
chemical profile b
chemical profile b
chemical profile b
chemical profile b
midpoint of range, 0.15 - 0.97;
U.S. EPA, 1984
chemical profile b
midpoint of range;
HSDB, 1998
ATSDR, 1998
midpoint of range, 0.6 - 0.95
U.S. EPA, 1988b
assumption °
assumption °
NR
NR
NR
midpoint of range, 0.01 - 0.1,
chemical profile
U.S. EPA, 1995
assumption °
assumption °
midpoint of range, 0.05 - 0.1
(U.S. EPA, 1991c; ATSDR, 1990b)
NR
assumption °
Johnson and Greger, 1982
chemical profile b
3-112
-------
Chemicals a
Urea compound C
Vinyl polymer
GI Absorption Factor
0.2
0.1
Source
assumption °
chemical profile b
a Includes only chemicals for which dermal HQs or MOEs could be calculated.
b Good, moderate, and low GI absorption, as reported in EPA chemical profiles, were translated to assumed GI
absorption fractions of 0.9, 0.5, and 0.1, respectively.
0 An assumption of 20 percent GI absorption was made for chemicals with no available GI absorption data.
NR: Not reported; data for some proprietary ingredients are not presented in order to protect confidential chemical
identities.
Lead
Methods used to evaluate potential lead risks from tin-lead solder used in the HASL
process are described in Section 3.4.6.
3.4.5 Results of Calculating Human Health Risk Indicators
This section presents the results of calculating risk indicators for both the occupational
setting and the ambient (outdoor) environment. When considering these risk characterization
results, it should be remembered that the results are intended for use in comparing relative
potential risk between processes, based on a model PWB facility, and should not be used as
absolute indicators of actual health risks to surface finishing line workers or to the public.
Occupational Setting
Estimated cancer risks and non-cancer risk indicators from occupational exposure to
surface finishing chemicals are presented below. It should be noted that no epidemiological
studies of health effects among PWB workers were located.
Inhalation Cancer Risk. Nickel/gold is the only process containing a chemical for which
a cancer slope (cancer potency) factor is available. Inorganic metallic salt A, in the nickel/gold
process, is the only chemical for which an inhalation cancer risk has been estimated. This metal
compound is considered a human carcinogen.19
Inhalation exposure estimates are based on the assumptions that emissions to indoor air
from conveyorized lines are negligible, that the air in the process room is completely mixed and
chemical concentrations are constant over time, and that no vapor control devices (e.g., bath
covers) are used in non-conveyorized lines. The exposure estimates use 90th percentile modeled
air concentrations, which means that, based on the PWB Workplace Practices Questionnaire data
and available information on bath concentrations, approximately 90 percent of the facilities are
19 A cancer classification of known human carcinogen has been assigned by either the EPA, IARC, and/or the
National Toxicology Program (NTP). Further details about the carcinogen classification are not provided in order to
protect the confidential chemical's identity.
3-113
-------
expected to have lower air concentrations and, therefore, lower risks. Using 90th percentile data
is consistent with EPA policy for estimating upper-bound exposures.
The upper bound maximum individual cancer risk over a lifetime is 2 x 10"7 based on a
workplace concentration of 2.4 x 10"6 milligrams inorganic metallic salt A per cubic meter of air,
over an 8 hour-day, for line operators using the non-conveyorized nickel/gold process. Cancer
risks less than 1 x 10"6 (one in one million) are generally considered to be of low concern. The
use of modeled, steady state, workplace air concentrations instead of actual monitoring data of
average and peak concentrations thus emerges as a significant source of uncertainty in estimating
cancer risk to workers exposed to inorganic metallic salt A in this industry. The available
toxicological data do not indicate that dermal exposure to inorganic metallic salt A increases
cancer risk, but no dermal cancer studies were located.
Risks to other workers would be proportional to the amount of time spent in the process
area. The exposure from inhalation for a typical line operator is based on spending 8 hr/day in
the surface finishing process area. Exposure times (i.e., time spent in the process area) for
various worker types from the workplace practices database are listed below. The number in
parentheses is the ratio of average time for that worker type to the 8 hr/day exposure time for a
line operator.
• laboratory technician: 2.8 hr/day (0.35);
• maintenance worker: 1.6 hr/day (0.2);
• supervisor: 5.5 hr/day (0.69); and
• wastewater treatment operator: 1 hr/day (0.12).
(Other types of workers may be in the process area for shorter or longer times.)
Other Potential Cancer Risk. Slope factors (cancer potency values) are needed to
calculate estimates of cancer risk. In addition to the chemical discussed above, lead and thiourea
have been determined by IARC to be possible human carcinogens (IARC Group 2B); lead has
also been classified by EPA as a probable human carcinogen (EPA Class B2). Lead is used in
tin-lead solder in the HASL process. Thiourea is used in the immersion tin process. Urea
compound B, a confidential ingredient in the nickel/gold and nickel/palladium/gold processes, is
possibly carcinogenic to humans. There are potential cancer risks to workers from these
chemicals, and workplace exposures have been estimated, but cancer potency and cancer risks
are unknown. Additionally, strong inorganic acid mists of sulfuric acid have been determined by
IARC to be a human carcinogen (IARC Group 1). Sulfuric acid is used in every surface finishing
process in this evaluation. It is not expected, however, to be present as a strong acid mist
because it used in diluted form in the aqueous baths.
Non-Cancer Risk. HQs and MOEs were calculated for line operators and laboratory
technicians from workplace exposures. An HQ exceeding one indicates a potential concern.
Unlike cancer risk, the HQ does not express probability, only the ratio of the estimated dose to
3-114
-------
the RfD or RfC, and it is not necessarily linear (an HQ often does not mean that adverse health
effects are ten times more likely than an HQ of one).
EPA considers high MOE values, such as values greater than 100 for a NOAEL-based
MOE or 1,000 for a LOAEL-based MOE, to pose a low level of concern (Barnes and Dourson,
1988). As the MOE decreases, the level of concern increases. Chemicals are noted here to be of
potential concern if a NOAEL-based MOE is lower than 100, a LOAEL-based MOE is lower than
1,000, or an MOE based on an effect level that was not specified as a LOAEL (used in the
absence of other data) is less than 1,000. As with the HQ, it is important to remember that the
MOE is not a probabilistic statement of risk.
Inhalation risk indicators of concern are presented in Table 3-30. This includes chemicals
of potential concern based on MOE and/or HQ results, as well as cancer risk results for the one
chemical with a cancer slope factor. Inhalation exposure estimates are based on the assumptions
that emissions to air from conveyorized lines are negligible, that the air in the process room is
completely mixed and chemical concentrations are constant over time, and that no vapor control
devices (e.g., bath covers) are used in non-conveyorized lines.
Dermal risk indicators of concern are presented in Table 3-31. This includes chemicals of
potential concern based on MOE and/or HQ. Dermal exposure estimates are based on the
assumption that both hands are routinely immersed in the bath, the worker does not wear gloves,
and all non-conveyorized lines are operated by manual hoist.
Table 3-32 provides a summary of the potential health effects for the chemicals of
concern listed in Tables 3-30 and 3-31. It should be noted that Tables 3-30 and 3-31 do not
include chemicals for which toxicity data were unavailable. Table 3-33 lists chemicals where
inhalation or dermal exposure is expected to occur, but appropriate toxicity values are not
available. (Table 3-25 provides qualitative structure-activity information pertaining to chemical
toxicity for those chemicals without available measured toxicity data.)
3-115
-------
Table 3-30. Summary of Human Health Risks From Occupational Inhalation Exposure for Selected Chemicals
Chemical of Concern
Alkyldiol
Ethylene glycol
Hydrochloric acid
Hydrogen peroxide
Inorganic metallic salt A
Nickel sulfate
Phosphoric acid
Propionic acid
Human Health Risk Indicator a
HASL
(NC)
NA
MQE (3, 9)
550, line operator
LOAEL
NA
NA
NA
NA
NA
NA
Nickel/Gold
(NC)
line operator
NA
HQ (1, 2, 3)
29, line operator
MOE (9)
940, line operator
LOAEL
cancer risk
< 1 x 10"6, line operator
HQ (4)
23, line operator
HQ (3)
2.7, line operator
NA
Nickel/Palladium/Gold
(NC)
line operator
NA
HQ (2, 12)
41, line operator
MOE (9)
730, line operator
LOAEL
NA
HQ (4)
50, line operator
HQ (3)
3.5, line operator
MOE (5)
31, line operator
LOAEL
OSP
(NC)
NA
MQE (3, 9)
370, line operator
LOAEL
NA
NA
NA
NA
NA
NA
a This table includes results for chemicals and pathways with an MOE less than 1,000 if based on a LOAEL (or less than 100 if based on a NOAEL), an HQ greater
than one, or cancer risk. It does not include chemicals for which toxicity data were unavailable. Specific results are not presented for confidential ingredients in
order to protect proprietary ingredient identity.
3-116
-------
How to read this table:
A: Type of risk indicator for which results are reported (HQ, MOE, or cancer risk).
B: Process bath(s) in which the chemical is used. These are only shown for non-proprietary chemicals. Numbers in parentheses indicate the process bath(s) in
which the chemical is used:
(2) catalyst
(4) electroless nickel
(6) immersion gold
(8) immersion tin
(10)OSP
(12) preinitiator
C: Value calculated for risk indicator (cancer risk, HQ, or MOE).
D: Type of worker for which risk results are presented (line operator or laboratory technician).
E: Type of toxicity data used for MOE: NOAEL, LOAEL, or data from human exposures, which do not provide a range of exposures but identify levels that have
adverse effects on humans.
NA: Not applicable.
(1) acid dip
(3) cleaner
(5) electroless palladium
(7) immersion silver
(9) microetch
(ll)predip
3-117
-------
Table 3-31. Summary of Human Health Risks Results From Occupational Dermal Exposure for Selected Chemicals
Chemical of Concern a
Ammonia compound A
Ammonium chloride
Ammonium hydroxide
Copper ion
Copper salt C
Copper sulfate
pentahydrate
Hydrogen peroxide
Inorganic metallic salt B
Nickel sulfate
Urea compound C
Human Health Risk Indicator a'b
HASL
(NC)
NA
NA
NA
NA
NA
MOE (9)
2.7, line
operator
190, lab tech
LOAEL
NA
NA
NA
NA
HASL
(C)
NA
NA
NA
NA
NA
MOE (9)
64, line
operator
860, lab tech
LOAEL
NA
NA
NA
NA
Nickel/Gold
(NC)
NA
HQ (6)
2.3, line operator
HQ (6)
2.5, line operator
NA
NA
MOE (9)
0.77, line
operator
12, lab tech
LOAEL
MOE (9)
430, line
operator
LOAEL
ine operator,
ab tech
HQ (4)
140, line
operator
9.2, lab tech
NA
Nickel/
Palladium/Gold
(NC)
ine operator
NA
HQ (6)
3.5, line
operator
NA
NA
MOE (9)
0.92, line
operator
37, lab tech
LOAEL
MOE (9)
510, line
operator
LOAEL
ine operator,
ab tech
HQ (4)
190, line
operator
4.6, lab tech
NA
OSP
(NC)
NA
NA
NA
MOE (10)
0.68, line
operator
10, lab tech
LOAEL
ine operator
MOE (9)
3.0, line
operator
46, lab tech
LOAEL
NA
NA
NA
NA
OSP
(C)
NA
NA
NA
MOE (10)
14, line
operator
48, lab tech
LOAEL
NA
MOE (9)
59, line
operator
210, lab tech
LOAEL
NA
NA
NA
NA
Immersion
Tin
(NC)
NA
NA
NA
NA
NA
NA
NA
NA
NA
ine operator
a This table includes results for chemicals and pathways with an MOE less than 1,000 if based on LOAELs (or less than 100 based on NOAELs), an HQ greater
than one, or cancer risk above IxlO"6. It does not include chemicals for which toxicity data were unavailable. Specific results are not presented for confidential
ingredients in order to protect proprietary ingredient identity.
3-118
-------
How to read this table:
A: Type of risk indicator for which results are reported (HQ, MOE, or cancer risk).
B: Process bath(s) in which the chemical is used. Numbers in parentheses indicate the process bath(s) in which the chemical is used:
(1) acid dip
(3) cleaner
(5) electroless palladium
(7) immersion silver
(9) microetch
(ll)predip
(2) catalyst
(4) electroless nickel
(6) immersion gold
(8) immersion tin
(10)OSP
(12) preinitiator
C: Value calculated for risk indicator (cancer risk, HQ, or MOE).
D: Type of worker for which risk results are presented (line operator or laboratory technician).
E: Type of toxicity data used for MOE: NOAEL, LOAEL, or data from human exposures, which do not provide a range of exposures but identify levels that have
adverse effects on humans.
NA: Not applicable.
3-119
-------
For inhalation exposure to workers, the following chemicals result in an HQ greater than
one or an MOE below the concern levels:
• ethylene glycol in non-conveyorized HASL;
• alkyldiol, hydrochloric acid, hydrogen peroxide, nickel sulfate, and phosphoric acid in
non-conveyorized nickel/gold;
• alkyldiol, hydrochloric acid, hydrogen peroxide, nickel sulfate, phosphoric acid, and
propionic acid in non-conveyorized nickel/palladium/gold; and
• ethylene glycol in non-conveyorized OSP.
Chemicals with HQs from dermal exposure greater than one, NOAEL-based MOEs lower
than 100, or LOAEL-based MOEs lower than 1,000, include:
copper sulfate pentahydrate in non-conveyorized and conveyorized HASL;
ammonium chloride, ammonium hydroxide, copper sulfate pentahydrate, hydrogen
peroxide, inorganic metallic salt B, and nickel sulfate in non-conveyorized nickel/gold;
ammonia compound A, ammonium hydroxide, copper sulfate pentahydrate, hydrogen
peroxide, inorganic metallic saltB, and nickel sulfate in non-conveyorized
nickel/palladium/gold;
copper ion, copper salt C, and copper sulfate pentahydrate in non-conveyorized OSP;
copper ion and copper sulfate pentahydrate in conveyorized OSP; and
urea compound C in non-conveyorized immersion tin.
3-120
-------
Table 3-32. Summary of Potential Human Health Effects for Chemicals of Concern
Chemical of Concern
Ammonia compound A,
Ammonium chloride, and
Ammonium hydroxide
Alkyldiol
Copper ion,
Copper sulfate pentahydrate,
and Copper salt C
Ethylene glycol
Hydrochloric acid
Hydrogen peroxide
Inorganic metallic salt A
Inorganic metallic salt B
Nickel sulfate
Phosphoric acid
Propionic acid
Urea compound C
Potential Health Effects
Contact with ammonium chloride solution or fumes irritate the eyes.
Large doses of ammonium chloride may cause nausea, vomiting,
thirst, headache, hyperventilation, drowsiness, and altered blood
chemistry. Ammonia fumes are extremely irritating to skin, eyes, and
respiratory passages. The severity of effects depends on the amount
of dose and duration of exposure.
Can affect the respiratory system if inhaled, and kidneys if absorbed
into the body.
Long-term exposure to high levels of copper may cause liver damage.
Copper is not known to cause cancer. The seriousness of the effects
of copper can be expected to increase with both level and length of
exposure.
In humans, low levels of vapors produce throat and upper respiratory
irritation. When ethylene glycol breaks down in the body, it forms
chemicals that crystallize and can collect in the body, which prevent
kidneys from working. The seriousness of the effects can be expected
to increase with both level and length of exposure.
Hydrochloric acid in the air can be corrosive to the skin, eyes, nose,
mucous membranes, respiratory tract, and gastrointestinal tract.
Hydrogen peroxide in the air can irritate the skin, nose, and eyes.
Ingestion can damage the liver, kidneys, and gastrointestinal tract.
Exposure can cause flu-like symptoms, weakness and coughing, and
has been linked to lung cancer and kidney disease.
Exposure to this material can damage the nervous system, kidneys,
and immune system.
Skin effects are the most common effects in people who are sensitive
to nickel. Workers who breath very large amounts of nickel
compounds have developed lung and nasal sinus cancers.
Inhaling phosphoric acid can damage the respiratory tract.
No data were located for health effects of propionic acid exposure in
humans, although some respiratory effects were seen in laboratory
mice.
Dermal exposure to urea compound C has resulted in allergic contact
dermatitis in workers, and exposure has caused weight loss in mice.
3-121
-------
Table 3-33. Data Gaps for Chronic Non-Cancer Health Effects for Workers
Chemical
Inhalation a or Dermal b
Exposure Potential
SAT Rank
(if available)
HASL
1,4-Butenediol
Alkylaryl sulfonate
Arylphenol
Fluoboric acid
Hydrochloric acid
Sodium hydroxide
Sulfuric acid
Tin
Inhalation and Dermal
Inhalation
Inhalation
Dermal
Dermal
Dermal
Dermal
Dermal
Low-moderate
Low
Moderate
Nickel/Gold
Aliphatic acid A
Aliphatic acid B
Aliphatic acid E
Aliphatic dicarboxylic acid A
Aliphatic dicarboxylic acid C
Alkylamino acid B
Ammonia compound B
Hydrochloric acid
Malic acid
Palladium chloride
Potassium compound
Sodium hydroxide
Sodium hypophosphite
Sulfuric acid
Urea compound B
Inhalation
Inhalation
Inhalation and Dermal
Inhalation
Inhalation
Dermal
Inhalation
Dermal
Inhalation
Dermal
Inhalation and Dermal
Dermal
Inhalation
Dermal
Inhalation and Dermal
Moderate
Low-moderate
Moderate-high
Low-moderate
Low
Low-moderate
Nickel/Palladium/Gold
Aliphatic acid B
Aliphatic acid E
Aliphatic dicarboxylic acid A
Aliphatic dicarboxylic acid C
Ammonia compound B
Hydrochloric acid
Malic acid
Palladium salt
Potassium compound
Sodium hydroxide
Inhalation
Inhalation and Dermal
Inhalation
Inhalation
Inhalation
Dermal
Inhalation
Dermal
Inhalation and Dermal
Dermal
Moderate
Low-moderate
Moderate-high
Low-moderate
Low
3-122
-------
Chemical
Sodium hypophosphite monohydrate
Sulfuric acid
Urea compound B
Inhalation a or Dermal b
Exposure Potential
Inhalation
Dermal
Inhalation and Dermal
SAT Rank
(if available)
Low-moderate
OSP
Acetic acid
Alkylaryl imidazole
Aromatic imidazole product
Arylphenol
Hydrochloric acid
Sodium hydroxide
Sulfuric acid
Inhalation
Dermal
Dermal
Inhalation
Dermal
Dermal
Dermal
Low-moderate
Moderate
Immersion Silver
1,4-Butenediol
Nitrogen acid
Sodium hydroxide
Sulfuric acid
Dermal
Dermal
Dermal
Dermal
Low-moderate
Immersion Tin
Alkylaryl sulfonate
Fluoboric acid
Hydrochloric acid
Methane sulfonic acid
Sulfuric acid
Thiourea
Urea compound C
Inhalation
Dermal
Dermal
Dermal
Dermal
Dermal
Inhalation
Low
a Applies only to the non-conveyorized process configuration.
b Applies to either process configuration.
Lead
Risk results for workers from lead in the HASL process are presented in Section 3.4.6.
Ambient (Outdoor) Environment
Potential risks are evaluated from exposure to chemicals released to outdoor air from a
PWB facility. Inhalation is the only exposure route to be quantified for people living nearby a
model PWB facility.
3-123
-------
Cancer Risk. As with the occupational setting, the nickel/gold process is the only
process for which cancer risk to humans in the ambient (outdoor) environment has been
estimated. These results for the non-conveyorized nickel/gold process, assuming that emissions
are vented to the outside, are an upper bound excess20 individual lifetime cancer risk for nearby
residents of 2 x 10~n. Inorganic metallic salt A is a human carcinogen.21 These estimates indicate
low concern and are interpreted to mean that, over a lifetime, an individual resident is expected to
have no more than one chance in 50 billion of developing cancer from exposure to inorganic
metallic salt A from a nearby facility using the non-conveyorized process.
None of the other process alternatives use chemicals for which cancer slope factors were
available, so no other cancer risks were estimated. Other identified chemicals in the surface
finishing processes are suspected carcinogens, but do not have established slope factors. Lead
and thiourea have been determined by IARC to be possible human carcinogens (IARC Group
2B); lead has also been classified by EPA as a probable human carcinogen (EPA Class B2). Lead
is used in tin-lead solder in the HASL process. Thiourea is used in the immersion tin process.
Urea compound B, a confidential ingredient in the nickel/gold and nickel/palladium/gold
processes, is possibly carcinogenic to humans. Exposure for nearby residents from these
chemicals has been estimated, but cancer potency and cancer risks are unknown. Additionally,
strong inorganic and acid mists of sulfuric acid have been determined by IARC to be a human
carcinogen (IARC Group 1). Sulfuric acid is used in diluted form in every surface finishing
process in this evaluation. It is not expected, however, to be released to the environment as a
strong acid mist.
Non-Cancer Risk. All HQs are less than one for ambient exposure to the general
population, indicating low concern from the estimated air concentrations. An MOE was
calculated for chemicals if an inhalation LOAEL or NOAEL was available and an RfC was not.
All MOEs for ambient exposure are greater than 1,000 for all processes, indicating low concern.
These results suggest there is low risk to nearby residents, based on incomplete but best
available data. Data limitations include the use of modeled air concentrations using data
compiled for a model facility rather than site-specific, measured concentrations. For estimating
ambient (outdoor) air concentrations, one key assumption is that no air pollution control
technologies are used to remove airborne chemicals from facility air prior to venting it to the
outside. Other data limitations are the lack of solid waste data to characterize exposure routes in
addition to inhalation, and lack of toxicity data for many chemicals.
20 Upper bound refers to the method of determining a slope factor, where the upper bound value (generated from
a certain probability statement) for the slope of the dose-response curve is used. Excess means the estimated cancer
risk is in addition to the already-existing background risk of an individual contracting cancer from all other causes.
21 A cancer classification of known human carcinogen has been assigned by either the EPA, IARC, and/or NTP.
Further details about the carcinogen classification are not provided in order to protect the confidential chemical's
identity.
3-124
-------
Lead. Risk results for people living near a PWB facility from lead in the HASL process
are presented below in Section 3.4.6.
3.4.6 Evaluation of Lead Risks from Tin-Lead Solder Used in the HASL Process
Although classified as a probable carcinogen by EPA, and known to cause other serious
health effects from chronic exposure, EPA has not derived a cancer slope factor, an RfD, or an
RfC for lead. Therefore, it is not possible to calculate a cancer risk, and the standard approach of
calculating an HQ to assess non-cancer health risks is not used for lead. Instead, lead exposure is
estimated using one of two exposure-biokinetic models, the Interim Adult Lead Methodology
(U.S. EPA, 1996a) and the Integrated Exposure Uptake Biokinetic Model for Lead in Children
(U.S. EPA, 1994). Both of these models relate estimated exposure levels to a lead concentration
in blood, which can then be compared to blood-lead levels at which health effects are known to
occur. These models are described further in Section 3.2.4 of the Exposure Assessment.
Table 3-34 presents federal (and other) regulations and guidelines for lead. This table also
presents comparable lead exposure values for workers and the ambient environment potentially
resulting from the lead in tin-lead solder used in the HASL process. For workers, the lowest
federal target or action levels are from OSHA and ACGIH, at 30 |ig/dL in blood. By comparison,
the 5 to 12 |ig/dL blood-lead levels from actual facility monitoring data for HASL line operators
are below this level. These monitoring data are limited to one facility, however.
We also modeled worker blood-lead levels using EPA's Adult Lead Methodology.
Estimated adult worker blood-lead levels (central estimate) from the model range from 2 to 63
|ig/dL, depending on the worker's lead intake rate. This estimate is higher than the limited
available monitoring data, with workers' measured blood-lead levels from 5 to 12 |ig/dL.
Estimated lead exposure using this model are very uncertain and could vary greatly depending on
worker activities. The ALM model was run based on the assumption that a worker gets lead on
his/her hands from handling solder, and then accidentally ingests some of that lead (e.g., by
eating or smoking without thoroughly washing their hands). The amount of lead ingested this
way is highly uncertain. Results from the model are based on a "conservative overestimate"
from surface wipe samples in hand soldering operations of 0.03 mg/day (Monsalve, 1984) and on
a range of soil ingestion rates of 10 to 50 mg/day for an adult in contact with soil (Stanek et al.,
1997 and U.S. EPA 1997a), respectively.22 (Ingestion data are not available specifically for a
HASL worker handling solder.) However, these results do indicate that there may be risk from
lead exposure via the ingestion route from poor hygiene practices.
10 mg/day is an average estimate; 50 mg/day is a central tendency estimate.
3-125
-------
Table 3-34. Risk Evaluation Summary for Lead
Federal Regulations and Guidelines for Lead
Lowest
Federal Level
Comparable Lead Exposure Values
Workplace
Worker blood-lead
target/action levels
Pregnant worker: fetal
blood-lead
target/action levels
Workplace air
exposure limit
OSHA, adults "who wish to bear
children"
OSHA, blood-lead level of concern
OSHA, medical removal
ACGIH (ACGIH, 1998)
NIOSH, level to be maintained
through air concentrations
OSHA
CDC
OSHA PEL (8 hr TWA)
NIOSH REL (NIOSH, 1994)
ACGIH TLV TWA (ACGIH, 1998)
30 ug/100g
40 ug/dL
50 ug/dL
30 ug/dL
60 ug/100 g
30 ug/100g
10 ug/dL a
50 ug/m3
100 ug/m3
50 ug/m3
30 ug/dL
10 ug/dL
50 ug/m3
Occupational blood-lead
monitoring data.
Modeled (ALM) blood-
lead data for an adult
worker.
Modeled (ALM) fetal
blood-lead level.
Workplace air
monitoring data
(average of HASL
process area monitoring
data provided by one
PWB manufacturer).
5-12 ug/dL
2 - 63 ug/dL
(depending on
intake rate)
3-102 ug/dL
(depending on
maternal intake rate)
3 ug/m3
Ambient Environment
Ambient air
concentration
Blood-lead
target/action levels for
child
National Ambient Air Quality
Standard, (U.S. EPA, 1987b)
CDC
OSHA
International: WHO blood lead level of
concern (WHO, 1986)
1.5 ug/m3
(averaged
over 3 mo.)
10 ug/m3
30 ug/100g
20 ug/dL
1.5 ug/m3
10 ug/m3
Ambient air
concentrations near a
PWB facility based on
HASL workplace air
monitoring data and air
dispersion model.
0.00009 ug/m3
Not determined. The IEUBK model estimates
blood- lead levels for children age 0 through 6
years. However, estimated ambient air
concentration from a HASL process are 1,000
times lower than the default value for air in the
model. IEUBK model results using default
values range from 2.7 to 4.5 ug/dL.
a CDC considers children to have an elevated level of lead if the amount of lead in the blood
remediation should be done for all children with blood levels > 20 ug/dL. Medical treatment
ug/dL (RTI, 1999).
is at least 10 ug/dL. Medical evaluation and environmental
may be necessary in children if the blood lead concentration is > 45
3-126
-------
NOTES:
ACGIH: American Conference of Governmental Industrial Hygienists, Inc.
CDC: Centers for Disease Control and Prevention.
EPA: U.S. Environmental Protection Agency.
NIOSH: National Institute for Occupational Safety and Health.
OSHA: Occupational Safety and Health Administration.
WHO: World Health Organization.
PEL: Permissible Exposure Limit.
REL: Recommended Exposure Level.
TWA: Time-weighted average.
TLV: Threshold limit value.
ALM: Adult Lead Methodology.
IEUBK: Integrated Exposure Uptake Biokinetic Model for Lead in Children.
About units: ug/dL = micrograms of elemental lead per deciliter (100 mL) of blood; 100 g blood is approximately equal to 100 mL or 1 dL.
3-127
-------
In addition to an adult worker, we used the ALM to model potential fetal blood-lead
levels, assuming a pregnant HASL line operator is exposed to lead through incidental ingestion.
Estimated 95th percentile fetal blood-lead levels of from 3.2 to 100 |ig/dL can be compared to the
guidance level from CDC and EPA of 10 |ig/dL for children.23 Again, these estimates are based
on uncertain ingestion rates.
Estimated workplace and ambient air concentration of lead also can be compared directly
to air regulations and guidelines for airborne lead from federal agencies (e.g., U.S. EPA, OSHA)
and the World Health Organization (WHO). For the workplace, an average of air monitoring data
from one PWB manufacturer24 of 3 jig/m3 can be compared to the lowest federal regulatory level
of 50 |ig/m3 (an OSHA, 8-hour, time-weighted average permissible exposure limit). For ambient
air near a facility, an estimated air concentration of 0.0001 |ig/m3 is well below the EPA air
regulation of 1 .5 |ig/m. (Ambient air modeling from a PWB facility is described further in
Section 3.2.3 of the Exposure Assessment.) It should be noted that these air monitoring data are
also limited to only one PWB manufacturer, and may vary from facility to facility.
The recommended approach for evaluating lead exposure to nearby residents is to apply
the IEUBK model to estimate blood-lead levels in children who may be exposed. (This is
discussed further in Section 3.2.4.) The default air concentration set in the model, based on
average 1990 U.S. urban air levels, is 1,000 times higher than the ambient air concentration
estimated from a HASL process. The IEUBK model could not discern any difference in
children's blood-lead levels based on such a small incremental increase in background air
concentrations. Based on these results, risks from lead exposure to nearby residents is expected
to be below concern levels.
3.4.7 Results of Calculating Ecological (Aquatic) Risk Indicators
We calculated ecological risk indicators (RIEco) for aquatic organisms as a unitless ratio:
= Csw / CC
where,
Csw = estimated surface water concentration following treatment in a POTW (mg/1)
CC = concern concentration (mg/1)
The method for estimating surface water concentrations is described in Section 3.2.3 of
the Exposure Assessment. Exposure concentrations below the CC are assumed to present low
risk to aquatic species. An ecological risk indicator greater than one indicates that the estimated
23 CDC considers children to have an elevated level of lead if the amount of lead in the blood is at least 10 ug/dL.
Medical evaluation and environmental remediation should be done for all children with blood-lead levels > 20 ug/dL.
Medical treatment may be necessary in children if the blood-lead concentration is > 45 ug/dL (RTI, 1999).
24 Results from both personal monitoring for HASL line operators and air samples from the HASL process area
were averaged.
3-128
-------
chemical concentration exceeds the concentration of concern for the aquatic environment based
on chemical toxicity to aquatic organisms. The level of concern increases as the ratio of exposure
concentration to CC increases, the derivation of CCs is described in Section 3.3.3 of the Human
Health and Ecological Hazards Summary and in Appendix H.
The results for non-metal surface finishing chemicals are summarized in Table 3-35.
Estimated surface water concentrations of several non-metals exceed the CC, as follows:
alkylaryl sulfonate, 1,4-butenediol, hydrogen peroxide, and potassium
peroxymonosulfate in the non-conveyorized HASL process;
alkylaryl sulfonate, hydrogen peroxide, and potassium peroxymonosulfate in the
conveyorized HASL process;
alkylaryl imidazole in non-conveyorized and convey orized configurations of the OSP
process;
hydrogen peroxide in the convey orized immersion silver process; and
potassium peroxymonosulfate in the non-conveyorized the immersion tin process (the
estimated surface water concentration per thiourea is equal to the CC).
Table 3-35. Summary of Aquatic Risk Indicators for Non-Metal Chemicals of Concern
Chemical
1,4-Butenediol
Alkylaryl imidazole
Alkylaryl sulfonate
Hydrogen peroxide
Potassium
peroxymonosulfate
Thiourea
CC
(mg/L)
0.008
0.001 -0.005
0.001 -0.005
0.02
0.01
0.03
Aquatic Risk Indicator (RIECo)
HASL
(NC)
1.3
NA
1 -5
2.0
8.2
NA
HASL
(C)
NA
NA
0.7-3.5
1.5
6.1
NA
OSP
(NC)
NA
6.6 - 33
NA
NA
NA
NA
OSP
(C)
NA
3.6- 18
NA
NA
NA
NA
I mm.
Silver (C)
NA
NA
NA
1.3
NA
NA
mm. Tin
(NC)
NA
NA
NA
NA
3.6
1.0 a
a Estimated surface water concentration is equal to the CC; this is not counted as an exceedance.
NA: Not applicable; estimated surface water concentration is less than CC or the chemical is not an ingredient of that
process configuration.
NC: Non-conveyorized.
C: Conveyorized.
It is assumed that on-site treatment is targeted to remove metals so that permitted
concentrations are not exceeded. If on-site treatment is not used to remove metals, high aquatic
risk indicators are possible. The ratio of estimated surface water concentration to CC for metals
is presented in Table 3-36. These data are presented to highlight the importance of on-site
treatment for toxic metals; because on-site treatment is expected to be performed to meet water
discharge permit requirements, these results are not used in comparing potential aquatic risks
among surface finishing alternatives.
3-129
-------
Table 3-36. Summary of Aquatic Risk Indicators for Metals Assuming
No On-Site Treatment
Chemical
Copper ion
Copper sulfate
pentahydrate
Nickel sulfate
Potassium gold cyanide
CC
(mg/L)
0.001
0.01
0.01
0.003
Aquatic Risk Indicator (RIECo)
HASL
(NC)
NA
5.1
NA
NA
HASL
(C)
NA
3.8
NA
NA
Nickel/
Gold
NA
NA
5.1
1.5
Nickel/ Palladium/
Gold
NA
NA
5.5
NA
OSP
(NC)
46
6.3
NA
NA
OSP
(C)
25
5.1
NA
NA
NA: Not applicable; estimated surface water concentration is less than CC or the chemical is not an ingredient of that
process configuration.
NC: Non-conveyorized.
C: Conveyorized.
3.4.8 Uncertainties
An important component of any risk characterization is the identification and discussion
of uncertainties. There are uncertainties involved in the measurement and selection of hazard
data, and in the data, models, and scenarios used in the Exposure Assessment. Any use of the
risk characterization should include consideration of these uncertainties.
Uncertainties in the Exposure Assessment include the following:
• accuracy of the description of exposure setting: how well the model facility used in the
assessment characterizes an actual facility; the likelihood of exposure pathways actually
occurring (scenario uncertainty);
• missing data and limitations of workplace practices data: this includes possible effects of
any chemicals that may not have been included (e.g., minor ingredients in the
formulations); possible effects of side reactions in the baths which were not considered;
and questionnaire data with limited facility responses;
• estimating exposure levels from averaged data and modeling in the absence of measured,
site-specific data;
• data limitations in the Source Release Assessment: releases to land could not be
characterized quantitatively;
• chemical fate and transport model applicability and assumptions: how well the models
and assumptions represent the situation being assessed and the extent to which the
models have been validated or verified (model uncertainty);
• parameter value uncertainty, including measurement error, sampling (or survey) error,
parameter variability, and professional judgement; and
• uncertainties in estimating exposure to lead, especially with assumptions made about
hand-to-mouth lead intake rates for workers.
Key assumptions made in the Exposure Assessment are discussed in Section 3.4.1.
3-130
-------
Uncertainties in the human health hazard data (as typically encountered in a hazard
assessment) include the following:
• using dose-response data from high dose studies to predict effects that may occur at low
levels;
• using data from short-term studies to predict the effects of long-term exposures;
• using dose-response data from laboratory animals to predict effects in humans;
• using data from homogeneous populations of laboratory animals or healthy human
populations to predict the effects on the general human population, with a wide range of
sensitivities (uncertainty due to natural variations in human populations);
• using LOAELs and NOAELs in the absence of peer-reviewed RfDs and RfCs;
• possible increased or decreased toxicity resulting from chemical interactions;
• assuming a linear dose-response relationship for cancer risk (in this case for inorganic
metallic salt A);
• effects of chemical mixtures not included in toxicity testing (effects may be independent,
additive, synergistic, or antagonistic); and
• possible effects of substances not evaluated because of a lack of chronic/sub chronic
toxicity data.
Uncertainties in the ecological hazards data and ecological risk characterization, which
attempt to evaluate potential ecotoxicity impacts to aquatic organisms from long-term (chronic)
exposure in a receiving stream, include the following:
• use of laboratory toxicity data to evaluate the effects of exposure in a stream;
• use of estimated toxicity data in the absence of measured data;
• use of data from acute exposure to evaluate the effect of chronic exposures;
• variation in species sensitivity; and
• uncertainties in estimating surface water concentrations from the drag-out model and
predicted POTW treatment efficiencies; also, surface water concentrations are based on
estimated releases to a modeled stream flow for the electronics industrial sector.
Another source of uncertainty comes from use of structure-activity relationships (SARs)
for estimating human health hazards in the absence of experimental toxicity data. Specifically,
this was done for: aliphatic acid B, aliphatic dicarboxylic acid A, alkylalkyne diol, alkylamino
acid A, alkylaryl imidazole, alkylaryl sulfonate, alkylimine dialkanol, amino acid salt, ammonia
compound B, aryl phenol, bismuth compound, 1,4-butenediol, citric acid, ethoxylated
alkylphenol, fatty amine, hydroxyaryl acid, hydroxyaryl sulfonate, maleic acid, malic acid,
potassium compound, potassium peroxymonosulfate, quaternary alkylammonium chlorides,
sodium benzene sulfonate, sodium hypophosphite, sodium hypophosphite monohydrate,
substituted amine hydrochloride, and transition metal salt.
Uncertainties in assessing risk from dermal exposure come from the use of toxicological
potency factors from studies with a different route of exposure than the one under evaluation
(i.e., using oral toxicity measures to estimate dermal risk). This was done for chemicals with oral
RfDs and chemicals with oral NOAELs or LOAELs (as noted in Tables 3-25 and 3-26).
3-131
-------
Uncertainties in dermal risk estimates also stem from the use of default values for missing
gastrointestinal absorption data. Specifically, this was done for: aliphatic acid E, aliphatic
dicarboxylic acid C, alkylamino acid B, alkylpolyol, amino carboxylic acid, fluoboric acid, gum,
hydrogen peroxide, hydroxy carboxylic acid, nitrogen acid, potassium gold cyanide, propionic
acid, stannous methane sulfonic acid, and sulfuric acid, and urea compound C.
Finally, the risk characterization does not address the potential adverse health effects
associated with acute exposure to peak levels of chemicals. This type of exposure is especially
important when evaluating developmental risks associated with exposure.
3.4.9 Conclusions
This risk characterization uses a health-hazard based framework and a model facility
approach to compare the potential health risks of one surface finishing process technology to the
potential risks associated with switching to an alternative technology. As much as possible,
reasonable and consistent assumptions are used across alternatives. Data to characterize the
model facility and exposure patterns for each process alternative were aggregated from a number
of sources, including PWB shops in the U.S., supplier data, and input from PWB manufacturers
at project meetings. Thus, the model facility is not entirely representative of any one facility, and
actual risk could vary substantially, depending on site-specific operating conditions and other
factors.
When using the results of this risk characterization to compare potential health effects
among alternatives, it is important to remember that this is a screening level rather than a
comprehensive risk characterization, both because of the predefined scope of the assessment and
because of exposure and hazard data limitations. It should also be noted that this approach does
not result in any absolute estimates or measurements of risk, and even for comparative purposes,
there are several important uncertainties associated with this assessment.
Another significant source of uncertainty is the limited data available for dermal toxicity
and the use of oral to dermal extrapolation when dermal toxicity data were unavailable. There is
high uncertainty in using oral data for dermal exposure and in estimating dermal absorption rates,
which could result in either over- or under-estimates of exposure and risk.
A third significant source of uncertainty is from the use of SARs to estimate toxicity in
the absence of measured toxicity data, and the lack of peer-reviewed toxicity data for many
surface finishing chemicals. Other uncertainties associated with the toxicity data include the
possible effects of chemical interactions on health risks, and extrapolation of animal data to
estimate human health risks from exposure to inorganic metallic salt A and other PWB
chemicals.
Another major source of uncertainty in estimating exposure is the reliance on modeled
data (i.e., modeled air concentrations) to estimate worker and ambient exposure. It should also
be noted that there is no comparative evaluation of the severity of effects for which HQs and
MOEs are reported.
3-132
-------
The Exposure Assessment for this risk characterization, whenever possible, used a
combination of central tendency and high-end assumptions, as would be used for an overall high-
end exposure estimate. Some values used in the exposure calculations, however, are better
characterized as "what-if," especially pertaining to exposure frequency, bath concentrations, use
of gloves, and process area ventilation rates for a model facility. Because some part of the
exposure assessment for both inhalation and dermal exposures qualifies as a "what-if descriptor,
the entire assessment should be considered "what-if."
Occupational Exposures and Risks
Health risks to workers were estimated for inhalation exposure to vapors and aerosols
from surface finishing baths and for dermal exposure to surface finishing bath chemicals.
Inhalation exposure estimates are based on the assumptions that emissions to indoor air from
conveyorized lines are negligible, that the air in the process room is completely mixed and
chemical concentrations are constant over time, and that no vapor control devices (e.g., bath
covers) are used in non-conveyorized lines. Dermal exposure estimates are based on the
assumption that workers do not wear gloves and that all non-conveyorized lines are operated by
manual hoist. Dermal exposure to line operators on non-conveyorized lines is estimated for
routine line operation and maintenance (e.g., bath replacement, filter replacement), and on
convey orized lines for bath maintenance activities alone.
Based on the number of chemicals with risk results above concern levels, some
alternatives to the non-conveyorized HASL process appear to pose lower occupational risks (i.e.,
conveyorized immersion silver, conveyorized and non-conveyorized immersion tin, and
convey orized HASL), some may pose similar levels of risk (i.e., conveyorized and non-
conveyorized OSP), and some may pose higher risk (i.e., non-conveyorized nickel/gold and
nickel/palladium/gold). There are occupational inhalation risk concerns for chemicals in the non-
conveyorized HASL, nickel/gold, nickel/palladium/gold, and OSP processes. There are also
occupational risk concerns for dermal contact with chemicals in the non-conveyorized HASL,
nickel/gold, nickel/palladium/gold, OSP, and immersion tin processes, and the conveyorized
HASL and OSP processes.
Cancer Risk. The non-conveyorized nickel/gold process contains the only chemical for
which an occupational cancer risk has been estimated (inorganic metallic salt A). The line
operator inhalation exposure estimate for inorganic metallic salt A results in an estimated upper
bound excess individual life time cancer risk of 2 x 10"7 (one in five million) based on high end
exposure. Cancer risks less than 1 x 10"6 (one in one million) are generally considered to be of
low concern. Risks to other types of workers25 were assumed to be proportional to the average
amount of time spent in the process area, which ranged from 12 to 69 percent of the risk for a line
operator.
25 These include laboratory technicians, maintenance workers, supervisors, and wastewater treatment operators.
Other types of workers may be present for shorter or longer times.
3-133
-------
Other identified chemicals in the surface finishing processes are suspected or known
carcinogens. Lead and thiourea have been determined by IARC to be possible human
carcinogens (IARC Group 2B); lead has also been classified by EPA as a probable human
carcinogen (EPA Class B2). Lead is used in tin-lead solder in the HASL process. Thiourea is
used in the immersion tin process. Urea compound B, a confidential ingredient in the nickel/gold
and nickel/palladium/gold processes, is possibly carcinogenic to humans. Exposure for workers
from these chemicals has been estimated, but cancer potency and cancer risks are unknown.
Additionally, strong inorganic and acid mists of sulfuric acid have been determined by IARC to
be a human carcinogen (IARC Group 1). Sulfuric acid is used in diluted form in every surface
finishing process in this evaluation. It is not expected, however, to be released to the air as a
strong acid mist. There are potential cancer risks to workers from these chemicals, but because
there are no slope factors, the risks cannot be quantified.
Non-Cancer Risk. For non-cancer risk, HQs greater than one, NOAEL-based MOEs
lower than 100, or LOAEL-based MOEs lower than 1,000 were estimated for occupational
exposures to chemicals in the non-conveyorized and conveyorized HASL processes, non-
conveyorized nickel/gold process, non-conveyorized nickel/palladium/gold process, non-
conveyorized and convey orized OSP processes, and the non-conveyorized immersion tin
process.
Based on calculated occupational exposure levels, there may be adverse health effects to
workers exposed to chemicals with an HQ exceeding 1.0 or an MOE less than 100 or 1,000.
However, it should be emphasized that these conclusions are based on screening level estimates.
These numbers are used here for relative risk comparisons between processes, and should not be
used as absolute indicators for actual health risks to surface finishing line workers.
Lead. Worker blood-lead levels measured at one PWB manufacturing facility were below
any federal regulation or guideline for workplace exposure. Modeling data, however, indicate
that blood-lead levels could exceed recommended levels for an adult and fetus, given high
incidental ingestion rates of lead from handling solder. These results are highly uncertain;
ingestion rates are based on surface wipe samples from hand soldering operations and on
incidental soil ingestion rates for adults in contact with soil. However, this indicates the need for
good personal hygiene for HASL line operators, especially wearing gloves and washing hands to
prevent accidental hand-to-mouth ingestion of lead.
Public Health Risks
Potential public health risk was estimated for inhalation exposure for the general public
living near a PWB facility. Public exposure estimates are based on the assumption that emissions
from both conveyorized and non-conveyorized process configurations are vented to the outside.
The risk indicators for ambient exposures to humans, although limited to airborne releases,
indicate low concern for nearby residents. The upper bound excess individual cancer risk for
nearby residents from inorganic metallic salt A in the non-conveyorized nickel/gold process was
estimated to be from approaching zero to 2 x 10"11 (one in 50 billion). This chemical has been
3-134
-------
classified as a human carcinogen.26 All hazard quotients are less than one for ambient exposure
to the general population, and all MOEs for ambient exposure are greater than 1,000 for all
processes, indicating low concern from the estimated air concentrations for chronic non-cancer
effects.
Estimated ambient air concentrations of lead from a HASL process are well below EPA
air regulatory limits for lead, and risks to the nearby population from airborne lead are expected
to be below concern levels.
Ecological Risks
We calculated ecological risk indicators (RIEco) for non-metal surface finishing chemicals
that may be released to surface water. Risk indicators for metals are not used for comparing
alternatives because it is assumed that on-site treatment is targeted to remove metals so that
permitted concentrations are not exceeded. Estimated surface water concentrations for non-
metals exceeded the CC in the following processes: four in the non-conveyorized HASL
process, three in the conveyorized HASL process, one in the non-conveyorized OSP process,
one in the convey orized OSP process, one in the convey orized immersion silver process, and one
in the non-conveyorized immersion tin process.
Overall Risk Screening and Comparison Summary
Table 3-37 presents an overall comparison of potential human health and ecological risks
for the baseline (non-conveyorized HASL) and the alternative process configurations.
26 A cancer classification of known human carcinogen has been assigned by either the EPA, IARC, and/or NTP.
Further details about the carcinogen classification are not provided in order to protect the confidential chemical
identity.
3-135
-------
Table 3-37. Overall Comparison of Potential Human Health and Ecological Risks for the
Non-Conveyorized HASL and Alternative Processes
Process
HASL (NC) (Baseline)
HASL (C)
Nickel/Gold (NC)
Nickel/Palladium/Gold (NC)
OSP (NC)
OSP (C)
Immersion Silver (C)
Immersion Tin (NC)
Immersion Tin (C)
Number of Chemicals
Potential
Carcinogen a
2
2
3
1
1
1
1
1
1
Inhalation
Concern b
1
0
5
6
1
0
0
0
0
Dermal
Concern c
1+ lead
1+ lead
6
6
3
2
0
1
0
Inhalation
Data
Gaps"
3
0
10
9
2
0
0
2
0
Dermal
Data
Gaps6
6
6
8
7
5
5
4
5
5
Aquatic
Concern f
4
3
0
0
1
1
1
1
0
a The number of chemicals with an EPA cancer WOE of A, B1, or B2, or an IARC WOE of 1, 2A, or 2B (see Table
3-21).
b The number of chemicals for which the HQ for worker inhalation exceeds 1, the NOAEL-based MOE is less than
100, or the LOAEL-based MOE is less than 1,000. See Table 3-30 for detailed results.
0 The number of chemicals for which the HQ for dermal contact by workers exceeds 1, the NOAEL-based MOE is
less than 100, or the LOAEL-based MOE is less than 1,000. See Table 3-30 for detailed results.
d The number of chemicals for which worker inhalation exposure is possible, but appropriate toxicity data are not
available for calculating a risk indicator (see Table 3-33).
e The number of chemicals for which worker dermal contact is possible but appropriate toxicity data are not available
for calculating a risk indicator (see Table 3-33).
f The number of chemicals for which the ecological risk indicators exceeds the concern level (i.e., RL^ > 1-0). See
Table 3-35 for detailed results.
3-136
-------
3.5 PROCESS SAFETY ASSESSMENT
Process safety is a concern and responsibility of employers and employees alike. Each
company has the obligation to provide its employees with a safe and healthy work environment,
while each employee is responsible for his/her own safe personal work habits. In the surface
finishing process of PWB manufacturing, hazards may be either chemical or process hazards.
Chemicals used in the surface finishing process can be hazardous to worker health and, therefore,
must be handled and stored properly, using appropriate personal protective equipment and safe
operating practices. Automated equipment can be hazardous to employees if safe procedures for
cleaning, maintaining, and operating the equipment are not established and regularly performed.
These hazards can result in serious injury and health problems to employees, and potential
damage to equipment.
The U.S. Department of Labor and the Occupational Safety and Health Administration
(OSHA) have established safety standards and regulations to assist employers in creating a safe
working environment and protecting workers from potential workplace hazards. In addition,
individual states may also have safety standards regulating chemical and physical workplace
hazards for many industries. Federal safety standards and regulations affecting the PWB
industry can be found in the Code of Federal Regulations (CFR) Title 29, Part 1910, and are
available by contacting your local OSHA field office. State and local regulations are available
from the appropriate state office.
An effective process safety program identifies potential workplace hazards and, if
possible, seeks to eliminate or at least reduce their potential for harm. Some companies have
successfully integrated the process safety program into their ISO 14000 certification plan, often
establishing process safety practices that go beyond OSHA regulations. This section of the
CTSA presents chemical and process safety concerns associated with the surface finishing
baseline technology and substitutes, as well as OSHA requirements to mitigate these concerns.
3.5.1 Chemical Safety Concerns
As part of its mission, OSHA's Hazard Communication Standard (29 CFR 1910.1200)
requires that chemical containers be labeled properly with chemical name and warning
information [. 1200(f)], that employees be trained in chemical handling and safety procedures
[.1200(h)], and that a MSDS be created and made available to employees for every chemical or
chemical formulation used in the workplace [. 1200(g)]. Each MSDS must be in English and
include information regarding the specific chemical identity and common name of the hazardous
chemical ingredients. In addition, information must be provided on the physical and chemical
characteristics of the hazardous chemical(s), known acute and chronic health effects and related
health information, exposure limits, whether the chemical is a carcinogen, emergency and first-
aid procedures, and the identification of the organization preparing the data sheet. Copies of
MSDSs for all of the chemicals/chemical formulations used must be kept and made available to
workers who may come into contact with the process chemicals during their regular work shift.
3-137
-------
In order to evaluate the chemical safety concerns of the various surface finishing
processes, MSDSs for 37 chemical products comprising six surface finishing technology
categories were collected and reviewed for potential hazards to worker safety. MSDSs were not
received for five confidential chemical products. Chemical safety data for pure chemical
compounds not sold as products were obtained from the Merck Index (Budavari, 1989).
Evaluating the chemical safety concerns specific to the HASL process baths was not
possible because there are no suppliers of microetch or cleaner baths made specifically for the
HASL process. Manufacturers will typically use the same microetch and cleaner formulation that
is used as part of another process line (e.g., the microetch and cleaner used in the making holes
conductive line). The chemical safety hazards for the HASL baths are similar to those reported
by the other processes for the same bath type. Therefore, the worse case bath from another
process was selected and reported for the HASL process to indicate the maximum safety hazard
which could be associated with the HASL process bath. Actual safety hazards for the bath will
depend greatly on the bath chemistry selected, and so may be less than the stated values.
Alternative processes with more than one product line submitted for evaluation were
treated in a similar manner to the HASL process. For each bath category, the actual bath which
posed the greatest hazard for each chemical hazard category was listed. For example, the
microetch bath which posed the greatest hazard, out of the two microetch baths submitted for
OSP, was listed for the OSP process
The results of that review are summarized and discussed in the sections below. General
information on OSHA storage and handling requirements for chemicals is located in Section
3.5.3. For a more detailed description of OSHA storage and handling requirements for surface
finishing chemical products, contact your area OSHA field office or state technical assistance
program.
Flammable. Combustible, and Explosive Surface Finishing Chemical Products
Table 3-38 presents a breakdown of surface finishing chemical products that, when in
concentrated form, are flammable, combustible, explosive, or pose a fire hazard. The following
lists OSHA definitions for chemicals in these categories, and discusses the data presented in the
table.
3-138
-------
Table 3-38. Flammable, Combustible, Explosive, and Fire Hazard Possibilities
for Surface Finishing Processes
Surface Finishing Process
HASLC
OSP d (2 product lines)
Immersion Silver
Immersion Tin d (2 product
lines)
Bath
Type
Cleaner
Microetch
Microetch
Cleaner
Immersion Tin
Hazardous Property a' b
Flammable
1(3)
1(3)
Combustible
Explosive
1(1)
1(1)
1(4)
Fire Hazard
1(1)
2(3)
2(3)
1(1)
a Table entries are made in the following format - # of products meeting OSHA definition for the given hazardous
property, as reported in the products' MSDSs (Total # of products in the process bath). A blank entry means that
none of the products for the specific process bath meet the OSHA reporting criteria for the given properly.
Example: For the immersion tin bath, 1 (4) means that one of the four products in the bath were classified as
explosive per OSHA criteria, as reported on the products' MSDSs.
b Data for pure chemicals (e.g., sulfuric acid) not sold as products were obtained from the Merck Index (Budavari,
1989) and included in category totals.
0 Formulations for HASL process baths were unavailable because cleaner and microetch bath chemistries are not
made specifically for the HASL process. Hazards reported for HASL bath types were reported as the worst case of the
results of similar baths from other processes.
d For alternative processes with more than one product line, the hazard data reported represents the most hazardous
bath of each type for the two product lines (e.g., of the microetch baths from the two product lines, the one with the
most hazardous chemicals is reported).
Flammable - A flammable chemical is defined by OSHA [29 CFR 1910.1200(c)] as one of the
following:
• An aerosol that, when tested by the method described in 16 CFR 1500.45, yields a flame
projection exceeding 18 inches at full valve opening, or a flashback at any degree of valve
opening.
• A gas that: 1) at ambient temperature and pressure, forms a flammable mixture with air at
a concentration of 13 percent by volume or less; or 2) when it, at ambient temperature and
pressure, forms a range of flammable mixtures with air wider than 12 percent by volume,
regardless of the lower limit.
• A liquid that has a flashpoint below 100 °F (37.8 °C), except any mixture having
components with flashpoints of 100 °F (37.8 °C) or higher, the total of which make up 99
percent or more of the total volume of the mixture.
• A solid, other than a blasting agent or explosive as defined in 29 CFR 1910.109(a), that is
liable to cause fire through friction, absorption of moisture, spontaneous chemical change,
or retained heat from manufacturing or processing, or which can be ignited readily and
when ignited burns so vigorously and persistently as to create a serious hazard.
Two chemical products are reported as flammable according to MSDS data. Although
the chemicals are flammable in their concentrated form, none of the chemical baths in the surface
finishing line contain flammable aqueous solutions.
3-139
-------
Combustible Liquid - As defined by OSHA [29 CFR 1910.1200(c)], a liquid that is considered
combustible has a flashpoint at or above 100 °F (37.8 °C), but below 200 °F (93.3 °C), except any
mixture having components with flashpoints of 200 °F (93.3 °C), or higher, the total volume of
which make up 99 percent or more of the total volume of the mixture. None of the chemical
products have been reported as combustible by their MSDSs.
Explosive - As defined by OSHA [29 CFR 1910.1200(c)], a chemical is considered explosive if it
causes a sudden, almost instantaneous release of pressure, gas, and heat when subjected to
sudden shock, pressure, or high temperature. Three chemical products are reported as explosive
by their MSDSs.
Fire Hazard - A chemical product that is a potential fire hazard is required by OSHA to be
reported on the product's MSDS. According to MSDS data, six chemical products are reported
as potential fire hazards.
Corrosive. Oxidizer. and Reactive Surface Finishing Chemical Products
A breakdown of surface finishing chemical baths containing chemical products that are
corrosive, oxidizers, or reactive in their concentrated form is presented in Table 3-39. The table
also lists process baths that contain chemical products that may cause a sudden release of
pressure when opened. The following lists OSHA definitions for chemicals in these categories
and discusses the data presented in the table.
Corrosive - As defined by OSHA (29 CFR 1910.1200 [Appendix A]), a chemical is considered
corrosive if it causes visible destruction of, or irreversible alterations in, living tissue by chemical
action at the site of contact following an exposure period of four hours, as determined by the test
method described by the U.S. Department of Transportation, 49 CFR Part 173, Appendix A.
This term does not apply to chemical action on inanimate surfaces. A review of MSDS data
found that 37 surface finishing chemical products are reported as corrosive in their concentrated
form. Some surface finishing baths may also be corrosive, but MSDSs do not provide data for
the process chemical baths once they are prepared.
Oxidizer - As defined by OSHA (29 CFR 1910.1200[c]), an oxidizer is a chemical other than a
blasting agent or explosive as defined by OSHA [29 CFR 1910.109(a)], that initiates or promotes
combustion in other materials, thereby causing fire either of itself or through the release of
oxygen or other gases. Five chemical products are reported as oxidizers, according to MSDS
data.
3-140
-------
Table 3-39. Corrosive, Oxidizer, Reactive, Unstable, and Sudden Release of Pressure
Possibilities for Surface Finishing Processes
Surface Finishing
Process
HASLC
Nickel/Gold d
(2 product lines)
Nickel/Palladium/Gold
OSPd
(2 product lines)
Immersion Silver
Immersion Tin d
(2 product lines)
Bath Type
Cleaner
Microetch
Cleaner
Microetch
Catalyst
Acid Dip
Cleaner
Microetch
Catalyst
Activator
Electroless Nickel
Electroless Palladium
Cleaner
Microetch
Cleaner
Microetch
Cleaner °
Microetch
Predip
Immersion Tin
Hazardous Property a' b
Corrosive
1(1)
3(4)
1(1)
3(4)
3(3)
1(1)
1(1)
3(4)
3(3)
1(4)
3(3)
1(3)
1(1)
3(4)
1(1)
1(3)
1(2)
2(2)
1(1)
3(4)
Oxidizer
1(3)
1(4)
1(4)
1(3)
1(3)
Reactive
Unstable
1(3)
1(3)
Sudden Release
of Pressure
1(4)
1(4)
1(4)
1(4)
a Table entries are made in the following format - # of products meeting OSHA definition for the given hazardous
property, as reported in the products' MSDSs (Total # of products in the process bath). A blank entry means that
none of the products for the specific process bath meet the OSHA reporting criteria for the given property.
Example: For the immersion tin bath, 3 (4) means that four of the five products in the bath were classified as
corrosive per OSHA criteria, as reported by the products' MSDSs.
b Data for pure chemicals (e.g., sulfuric acid) not sold as products were obtained from the Merck Index (Budavari,
1989) and included in category totals.
0 Formulations for HASL process baths were unavailable because cleaner and microetch bath chemistries are not
made specifically for the HASL process. Hazards reported for HASL bath types were reported as the worst case of the
results of similar baths from other processes.
d For alternative processes with more than one product line, the hazard data reported represents the most hazardous
bath of each type for the two product lines (e.g., of the microetch baths from the two product lines, the one with the
most hazardous chemicals is reported).
Reactive - A chemical is considered reactive if it is readily susceptible to change and the possible
release of energy. EPA gives a more precise definition of reactivity for solid wastes. As defined
by EPA (40 CFR 261.23), a solid waste is considered reactive if a representative sample of the
waste exhibits any of the following properties: 1) is normally unstable and readily undergoes
violent change without detonating; 2) reacts violently or forms potentially explosive mixtures
with water; 3) when mixed with water, generates toxic gases, vapors, or fumes in a quantity
sufficient to present a danger to human health or the environment (for a cyanide or sulfide
bearing waste, this includes exposure to a pH between 2 and 12.5); 4) is capable of detonation or
explosive reaction if subjected to a strong initiated source or if heated under confinement; or 5) is
3-141
-------
explosive reaction if subjected to a strong initiated source or if heated under confinement; or 5) is
readily capable of detonation or explosive decomposition or reaction at standard temperature and
pressure. A review of MSDS data shows that none of the chemical products used in the surface
finishing processes are considered reactive.
Unstable - As defined by OSHA (29 CFR 1910.1200[c]), a chemical is unstable if in the pure
state, or as produced or transported, it will vigorously polymerize, decompose, condense, or will
become self-reactive under conditions of shock, pressure, or temperature. Only two of the
chemical products are reported as unstable, according to MSDS data.
Sudden Release of Pressure - OSHA requires the reporting of chemical products that, while
stored in a container subjected to sudden shock or high temperature, causes a pressure increase
within the container that is released upon opening. MSDS data indicates four chemical products
that are potential sudden release of pressure hazards.
Surface Finishing Chemical Product Health Hazards
A breakdown of surface finishing process baths that contain chemical products that are
sensitizers, acute or chronic health hazards, or irreversible eye damage hazards in their
concentrated form is presented in Table 3-40. Also discussed in this section are surface finishing
chemical products that are potential eye or dermal irritants and suspected carcinogens. The
following presents OSHA definitions for chemicals in these categories and discusses the data in
Table 3-40, where appropriate.
Sensitizer - A sensitizer is defined by OSHA [29 CFR 1910.1200 Appendix A (mandatory)] as a
chemical that causes a substantial proportion of exposed people or animals to develop an allergic
reaction in normal tissue after repeated exposure to the chemical. Sixteen chemical products are
reported as sensitizers by MSDS data.
Acute and Chronic Health Hazards - As defined by OSHA (29 CFR 1910.1200 Appendix A), a
chemical is considered a health hazard if there is statistically significant evidence based on at least
one study conducted in accordance with established scientific principles that acute or chronic
health effects may occur in exposed employees. Health hazards are classified using the criteria
below:
• acute health hazards are those whose effects occur rapidly as a result of short-term
exposures, and are usually of short duration; and
• chronic health hazards are those whose effects occur as a result of long-term exposure,
and are of long duration.
Chemicals that are considered a health hazard include carcinogens, toxic or highly toxic agents,
reproductive toxins, irritants, corrosives, sensitizers, hepatotoxins, nephrotoxins, neurotoxins,
agents that act on the hematopoietic system, and agents which damage the lungs, skin, eyes, or
mucous membranes.
3-142
-------
Table 3-40. Sensitizer, Acute and Chronic Health Hazards, and Irreversible Eye Damage
Possibilities for Surface Finishing Processes
Surface Finishing
Process
HASLC
Nickel/Gold d
(2 product lines)
Nickel/Palladium/Gold
OSPd
(2 product lines)
Immersion Silver
Immersion Tin d
(2 product lines)
Bath Type
Cleaner
Microetch
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Immersion Gold
Cleaner
Microetch
Catalyst
Activator
Electroless Nickel
Electroless Palladium
Immersion Gold
Cleaner
Microetch
Cleaner
Microetch
Cleaner
Microetch
Predip
Immersion Tin
Hazardous Property a' b
Sensitizer
1(2)
2(3)
1(2)
1(2)
1(1)
1(4)
1(3)
1(3)
2(3)
1(3)
1(2)
1(2)
2(4)
Acute
Health
Hazard
1(1)
3(4)
1(1)
3(4)
2(3)
1(1)
2(2)
2(2)
1(1)
3(4)
2(3)
4(4)
3(3)
2(3)
1(2)
1(1)
3(3)
1(1)
2(3)
1(2)
1(2)
1(4)
Chronic
Health
Hazard
1(1)
3(3)
1(1)
2(2)
1(2)
1(1)
2(2)
2(2)
1(1)
1(4)
1(3)
2(4)
2(3)
1(3)
1(2)
1(1)
3(3)
1(1)
2(3)
1(2)
1(2)
1(4)
Carcinogen
1(1)
1(1)
1(2)
2(4)
1(3)
1(1)
Irreversible
Eye
Damage
1(1)
3(4)
1(1)
3(4)
1(2)
1(1)
1(2)
3(4)
1(3)
1(4)
2(3)
3(3)
1(1)
3(4)
1(1)
2(3)
1(2)
2(2)
1(1)
2(4)
a Table entries are made in the following format - # of products meeting OSHA definition for the given hazardous
property, as reported in the products' MSDSs (Total # of products in the process bath). A blank entry means that
none of the products for the specific process bath meet the OSHA reporting criteria for the given properly.
Example: For the immersion tin bath, 2(4) means that three of the five products in the bath were classified as
sensitizers per OSHA criteria, as reported by the products' MSDSs.
b Data for pure chemicals (e.g., sulfuric acid) not sold as products were obtained from the Merck Index (Budavari,
1989) and included in category totals.
0 Formulations for HASL process baths were unavailable because cleaner and microetch bath chemistries are not
made specifically for the HASL process. Hazards reported for HASL bath types were reported as the worst case of the
results of similar baths from other processes.
d For alternative processes with more than one product line, the hazard data reported represents the most hazardous
bath of each type for the two product lines (e.g., of the microetch baths from the two product lines, the one with the
most hazardous chemicals is reported).
3-143
-------
A review of MSDS data shows that 41 chemical products are reported as potentially
posing acute health hazards, and 32 chemical products potentially pose chronic health hazards.
OSHA does not require reporting of environmental hazards such as aquatic toxicity data, nor are
toxicity data on MSDSs as comprehensive as the toxicity data collected for the CTSA. OSHA
health hazard data are presented here for reference purposes only, and are not used in the risk
characterization component of the CTSA.
Carcinogen - As defined by OSHA (29 CFR 1910.1200 Appendix A), a chemical is considered to
be a carcinogen if: 1) it has been evaluated by the IARC, and found to be a carcinogen or
potential carcinogen; 2) it is listed as a carcinogen or potential carcinogen in the Annual Report
on Carcinogens published by the National Toxicology Program (NTP); or 3) it is regulated by
OSHA as a carcinogen. A review of MSDS data indicates that seven chemical products are
reported as potential carcinogens, by either NTP, IARC, or EPA WOE Classifications. Suspected
carcinogens include nickel sulfate, thiourea, and various lead compounds that are commonly
used in several processes. Suspected carcinogens are discussed in more detail in the human
health and ecological hazards summary, Section 3.3.
Dermal or Eye Irritant - An irritant is defined by OSHA [29 CFR 1910.1200 Appendix A
(mandatory)] as a chemical, that is not corrosive, but which causes a reversible inflammatory
effect on living tissue by chemical action at the site of contact. A chemical is considered a dermal
or eye irritant if it is so determined under the testing procedures detailed in 16 CFR 1500.41- 42
for four hours exposure. Table 3-40 does not include this term, because all of the surface
finishing chemical products are reported as either dermal or eye irritants.
Irreversible Eye Damage - Chemical products that, upon coming in contact with eye tissue, can
cause irreversible damage to the eye are required by OSHA to be identified as such on the
product's MSDS. A review of MSDS data shows that 34 chemical products are reported as
having the potential to cause irreversible eye damage.
Other Chemical Hazards
Surface finishing chemical products that have the potential to form hazardous
decomposition products are presented below. In addition, chemical product incompatibilities
with other chemicals or materials are described, and other chemical hazard categories are
presented. The following lists OSHA definitions for chemicals in these categories and
summarizes the MSDS data, where appropriate.
Hazardous Decomposition - A chemical product, under specific conditions, may decompose to
form chemicals that are considered hazardous. The MSDS data for the chemical products in the
surface finishing process indicate that over half of the products have the possibility of
decomposing to form potentially hazardous chemicals. Each chemical product should be
examined to determine its decomposition products so that potentially dangerous reactions and
exposures can be avoided. The following are examples of hazardous decomposition of chemical
products that are employed in the surface finishing alternatives:
3-144
-------
• products used in the predip and immersion tin baths of the immersion tin process, or in
the microetch and OSP baths of the OSP process, may decompose to release carbon
monoxide and carbon dioxide gas;
• oxygen gas may be generated by some of the microetch baths from the nickel/gold
process, posing a potential combustion hazard;
• thermal decomposition under fire conditions of certain chemical bath constituents in the
nickel/gold or the nickel/palladium/gold process can result in releases of toxic oxide gases
such as nitrogen, sulfur, or chlorine;
• some chemical products used in the nickel/gold and nickel/palladium/gold processes will
release toxic chlorine fumes if they are allowed to react with persulfate compounds; and
• one product present in the cleaner bath of the immersion silver process will react with
reactive metals to release flammable hydrogen gas.
Incompatibilities - Chemical products are often incompatible with other chemicals or materials
with which they may come into contact. A review of MSDS data shows that over 80 percent of
the surface finishing chemical products have incompatibilities that can pose a threat to worker
safety if the proper care is not taken to prevent such occurrences. Reported incompatibilities
range from specific chemicals or chemical products, such as acids or cyanides, to other
environmental conditions, such as direct heat or sunlight. Chemical incompatibilities that are
common to products from all the surface finishing processes include acids, bases, alkalies,
oxidizing and reducing agents, metals, and combustible materials. Incompatibilities were also
found to exist between chemical products used on the same process line. Individual chemical
products for each process bath should be closely examined to determine specific
incompatibilities, and care should be taken to avoid contact between incompatible chemicals and
chemical products, textiles, and storage containers.
The following are examples of chemical incompatibilities that exist for chemical products
used in the surface finishing alternatives:
• some products in the catalyst baths of both the nickel/gold and nickel/palladium/gold
processes are incompatible with strong bases, alkalies, and oxidizing agents;
• organic materials, combustible materials, and oxidizing and reducing agents should be
kept away from the microetch bath of the OSP process, and strong alkaline materials
should be avoided in the microetch baths for all of the processes; and
• persulfate should be avoided in the electroless palladium bath of the nickel/palladium/gold
process, because it will react with the chemicals in the bath to release chlorine gas.
Other Chemical Hazard Categories - OSHA requires the reporting of several other hazard
categories on the MSDSs for chemicals or chemical products that have not already been
discussed above. These additional categories include chemical products that are:
• water-reactive (react with water to release a gas that presents a health hazard);
• pyrophoric (will ignite spontaneously in air at temperatures below 130 °F);
• stored as a compressed gas;
3-145
-------
• classified as an organic peroxide; or
• chemicals that have the potential for hazardous polymerization.
A review of MSDS data indicates that none of the chemical products are reported as being
water-reactive, pyrophoric, a compressed gas, an organic peroxide, or as having the potential for
hazardous polymerization.
3.5.2 Hot Air Solder Leveling (HASL) Process Safety Concerns
Several unique process safety concerns arise from the operation of the HASL process,
due to differences in the way the final surface finish is applied. Although the cleaning and
microetch baths are similar to those used by the other alternatives, the solder finish is applied by
the physical process of manually contacting the PWB with molten solder, rather than applying
the surface finish through a chemical plating or coating process. The molten solder bath, which is
typically operated at a temperature of up to 500 °F, poses several safety concerns, such as
accidental contact with the molten metal by workers, exposure to acids in the flux, and the
potential for fire.
Solder eruptions often occur during process startup as the solid solder is heated. Solder
melts from bottom to top, and pressure may build up from thermal expansion causing the solder
to erupt. Splattering of the melted solder onto workers could cause serious burns. Caution
should be exercised during process startup to avoid worker injury. Heat resistant clothing, face
shields, protective aprons, long sleeve gloves, and shoes should be required when working
around the solder bath.
Fire is possible at the solder bath and the exhaust/ventilation system, although it does not
occur frequently. When fire occurs, small amounts of hazardous gases, such as hydrogen
chloride and carbon dioxide, can be released. Causes of fire include the build-up of carbon
residual from the use of oil-based flux and other flammable materials kept too close to the
process. Isolating flammable materials from the process area and regular cleaning of the HASL
machine will prevent a fire from occurring.
Other safety concerns include workers exposed to small amounts of acid in the flux, lead
in the solder bath, and to process chemicals in the cleaner and microetch baths. Risk from
exposure to process chemicals is addressed in detail in Section 3.4, Risk Characterization. Like
other surface finishing processes, federal safety standards and regulations concerning the HASL
process can be found in CFR Title 29, Part 1910, and are available from the appropriate state
office.
3.5.3 Process Safety Concerns
Exposure to chemicals is just one of the safety issues that PWB manufacturers may have
to address during their daily activities. Preventing worker injuries should be a primary concern
for employers and employees alike. Work-related injuries may result from faulty equipment,
improper use of equipment, bypassing equipment safety features, failure to use personal
protective equipment, and physical stresses that may appear gradually as a result of repetitive
3-146
-------
motions (i.e., ergonomic stresses). Any or all of these types of injuries may occur if proper
safeguards or practices are not in place and adhered to. An effective worker safety program
includes:
• an employee training program;
• employee use of personal protective equipment;
• proper chemical storage and handling; and
• safe equipment operating procedures;
The implementation of an effective worker safety program can have a substantial impact
on business, not only in terms of direct worker safety, but also in reduced operating costs as a
result of fewer days of absenteeism, reduced accidents and injuries, and lower insurance costs.
Maintaining a safe and efficient workplace requires that both employers and employees recognize
and understand the importance of worker safety and dedicate themselves to making it happen.
Employee Training
A critical element of workplace safety is a well-educated workforce. To help achieve this
goal, the OSHA Hazard Communication Standard requires that all employees at PWB
manufacturing facilities (regardless of the size of the facility) be trained in the use of hazardous
chemicals to which they are exposed. A training program should be instituted for workers,
especially those operating the surface finishing process, who may come into contact with, or be
exposed to, potentially hazardous chemicals. Training may be conducted by either facility staff
or outside parties who are familiar with the PWB manufacturing process and the pertinent safety
concerns. The training should be held for each new employee, as well as periodic retraining
sessions when necessary (e.g., when a new surface finishing process is instituted), or on a regular
schedule. The training program should inform the workers about the types of chemicals with
which they work and the precautions to be used when handling or storing them, when and how
personal protection equipment should be worn, and how to operate and maintain equipment
properly.
Storing and Using Chemicals Properly
Because the surface finishing process requires handling a variety of chemicals, it is
important that workers know and follow the correct procedures for the use and storage of the
chemicals. Much of the use, disposal, and storage information about surface finishing process
chemicals may be obtained from the MSDSs provided by the manufacturer or supplier of each
chemical or formulation. Safe chemical storage and handling involves keeping chemicals in their
proper place, protected from adverse environmental conditions, as well as from other chemicals
with which they may react. Examples of supplier recommended storage procedures found on the
MSDSs for surface finishing chemicals are listed below.
• store chemical containers in a cool, dry place away from direct sunlight and other sources
of heat;
• chemical products should only be stored in their properly sealed original containers and
labeled with the common name of the chemical contents;
3-147
-------
• incompatible chemical products should never be stored together; and
• store flammable liquids separately in a segregated area away from potential ignition
sources or in a flammable liquid storage cabinet.
Some products have special storage requirements and precautions listed on their MSDSs
(e.g., relieving the internal pressure of the container periodically). Each chemical product should
be stored in a manner consistent with the recommendation on the MSDS. In addition, chemical
storage facilities must be designed to meet any local, state, and federal requirements that may
apply.
Not only must chemicals be stored correctly, but they must also be handled and
transported in a manner that protects worker safety. Examples of chemical handling
recommendations from suppliers include:
• wear appropriate protective equipment when handling chemicals;
• open containers should not be used to transport chemicals;
• use only spark-proof tools when handling flammable chemicals; and
• transfer chemicals using only approved manual or electrical pumps to prevent spills
created from lifting and pouring.
Proper chemical handling procedures should be a part of the training program given to
every worker. Workers should also be trained in chemical spill containment procedures and
emergency medical treatment procedures in case of chemical exposure to a worker.
Use of Personal Protective Equipment
OSHA has developed several personal protective equipment standards that are applicable
to the PWB manufacturing industry. These standards address general safety and certification
requirements (29 CFRPart 1910.132), the use of eye and face protection (Part 1910.133), head
protection (Part 1910.135), foot protection (Part 1910.136), and hand protection (Part 1910.138).
The standards for eye, face, and hand protection are particularly important for the workers
operating the surface finishing process where there is close contact with a variety of chemicals, of
which nearly all irritate or otherwise harm the skin and eyes. In order to prevent or minimize
exposure to such chemicals, workers should be trained in the proper use of personal safety
equipment.
The recommended personal protective equipment for a worker handling chemicals is also
indicated on the MSDS. For the majority of surface finishing chemicals, the appropriate
protective equipment indicated by the MSDS includes:
goggles to prevent the splashing of chemical into the eyes;
chemical aprons or other impervious clothing to prevent splashing of chemicals <
clothing;
gloves to prevent dermal exposure while operating the process; and
boots to protect against chemical spills.
3-148
-------
Additional personal protective equipment recommended for workers operating the HASL
process includes:
• heat resistant gloves to prevent burns by accidental contact with molten solder; and
• face shield to protect face and eyes from solder splatter.
Other items less frequently suggested include chemically resistant coveralls and hats. In
addition to the personal protective equipment listed above, some MSDSs recommend that other
safety equipment be readily available. This equipment includes first aid kits, oxygen supplies
(SCBA), fire extinguishers, ventilation equipment, and respirators.
Other personal safety considerations are the responsibility of the worker. Workers should
be prohibited from eating or keeping food near the surface finishing process. Because automated
processes contain moving parts, workers should also be prohibited from wearing jewelry or loose
clothing, such as ties, that may become caught in the machinery and cause injury to the worker
or the machinery itself. In particular, the wearing of rings or necklaces may lead to injury.
Workers with long hair that may also be caught in the machinery should be required to securely
pull their hair back or wear a hair net.
Use of Equipment Safeguards
In addition to the use of proper personal protection equipment for all workers, OSHA has
developed safety standards (29 CFR Part 1910.212) that apply to the equipment used in a PWB
surface finishing process. Among the safeguards recommended by OSHA that may be used for
conveyorized equipment are barrier guards, two-hand trip devices, and electrical safety devices.
Safeguards for the normal operation of conveyor equipment are included in the standards for
mechanical power-transmission apparatus (29 CFR Part 1910.219) and include belts, gears,
chains, sprockets, and shafts. PWB manufacturers should be familiar with the safety
requirements included in these standards and should contact their local OSHA office or state
technical assistance program for assistance in determining how to comply with them.
In addition to normal equipment operation standards, OSHA also has a lockout/tagout
standard (29 CFR Part 1910.147). This standard is designed to prevent the accidental start-up of
electric machinery during cleaning or maintenance operations, and apply to the cleaning of
conveyorized equipment as well as other operations. OSHA has granted an exemption for minor
servicing of machinery, provided the equipment has other appropriate safeguards, such as a
stop/safe/ready button that overrides all other controls and is under the exclusive control of the
worker performing the servicing. Such minor servicing of conveyorized equipment can include
clearing fluid heads, removing jammed panels, lubricating, removing rollers, minor cleaning,
adjustment operations, and adding chemicals. Rigid finger guards should also extend across the
rolls, above and below the area to be cleaned. Proper training of workers is required under the
standard whether lockout/tagout is employed or not. For further information on the applicability
of the OSHA lockout/tagout standard to surface finishing process operations, contact the local
OSHA field office.
3-149
-------
Occupational Noise Exposure
OSHA has also developed standards (29 CFRPart 1910.95) that apply to occupational
noise exposure. These standards require protection against the effects of noise exposure when
the sound levels exceed certain levels specified in the standard. No data were collected on actual
noise levels from surface finishing process lines.
3-150
-------
REFERENCES
American Conference of Governmental Industrial Hygienists (ACGIH). 1998. 1998
TLVs and BEIs. Threshold Limit Values for Chemical Substances and Physical Agents.
Biological Exposure Indices. Cincinnati, OH.
Agency for Toxic Substances and Disease Registry (ATSDR). 1990a. "Toxicological
Profile for Copper." ATSDR, U.S. Department of Health and Human Services, Public
Health Service. Atlanta, GA.
Agency for Toxic Substances and Disease Registry (ATSDR). 1990b. "Toxicological
Profile for Silver." ATDSR, U.S. Department of Health and Human Services, Public
Health Service, Atlanta, GA.
Agency for Toxic Substances and Disease Registry (ATDSR). 1997a. "Toxicological
Profile for Nickel (Update)." ATSDR, U.S. Department of Health and Human Services,
Public Health Service, Atlanta.
Agency for Toxic Substances and Disease Registry (ATDSR). 1997b. "Technical
Report for Ethylene Glycol/Propylene Glycol." Draft for Public Comment. ATSDR,
U.S. Department of Health and Human Services, Public Health Service, Atlanta.
Agency for Toxic Substances and Disease Registry (ATDSR). 1998. "Toxicological
Profile for 2-Butoxyethanol and 2-Butoxyethanol Acetate." ATSDR, U.S. Department of
Health and Human Services, Public Health Service, Atlanta.
Alexander, F.W. and H.T. Delves. 1981. "Blood Lead Levels during Pregnancy." Int.
Arch. Occup. Environ. Health. Vol. 48, pp.35-39.
Aulerich, R.J., R.K. Ringer, M.R. Bleavins, et al. 1982. "Effects of Supplemental
Dietary Copper on Growth, Reproductive Performance and Kit Survival of Standard Dark
Mink and the Acute Toxicity of Copper to Mink." J. Animal Sci. Vol. 55 pp.337-343 (as
cited in ATSDR, 1990).
Barnes, D.G. and M. Dourson. 1988. "Reference Dose (RfD): Descriptions and Uses in
Health Risk Assessments." Regulatory Toxicology and Pharmacology. Vol. 8, pp. 471-
486.
BASF Corporation. 1987. Results of several investigations with propionic acid with
cover letter. October 6. Submission to EPA under TSCA, section FYI.
Bayes, Martin. 1996. Shipley Company. Personal communication to Jack Geibig, UT
Center for Clean Products and Clean Technologies. January.
Berglund, R. and E. Lindh. 1987. "Prediction of the Mist Emission Rate from Plating
Baths." Proc. Am. Electroplaters and Surface Finishers Soc. Ammu. Tech. Conf.
3-151
-------
Budavari, S. (Ed). 1989. The Merck Index, 11th ed. Rahway: Merck & Co., Inc.
Bureau of Labor Statistics. 1990. "Statistical Summary: Tenure with Current Employer
as of January 1987." As cited in EPA, 1991b.
Bureau of Labor Statistics. 1997. Job Tenure Summary. Labor Force Statistics from the
current Population Survey. Website: http://stats.bls.gov/news.release/tenure.nws.htm.
Information downloaded on March 24.
Bushy Run Research Center (Bushy Run). 1995. Developmental Toxicity Evaluation of
Ethylene Glycol Administered by Gavage to CD Rats: Determination of a No
Observable Effect Level with Cover Letter. TSCATS data submissions retrieved from
TOXLINE, December 5. FYI-OTS-0790-0692; FYI-OTS-0990-0323.
Centers for Disease Control and Prevention (CDC). 1991. "Preventing Lead Poisoning in
Young Children." U.S. Department of Health and Human Services, Public Health
Service, Atlanta, GA.
Clayson, D.B., F. Iverson, E.A. Nera and E. Lok. 1991. Title not available. MutatRes.
Vol. 248, p.321 (cited in Katz and Guest, 1994, HSDB, 1998).
Decisioneering, Inc. 1993. Crystal Ball® software.
DePass, L.R., R.S.H. Yang and M.D. Woodside. 1987. "Evaluation of the
Teratogenicity of Ethylenediamine Dihydrochloride in Fischer 344 Rats by Conventional
and Pair-Feeding Studies." Fund. Appl. Toxicol. Vol.9, pp.687-697 (cited in U.S. EPA,
1988).
Franklin, C.A., M.J. Inskip, C.L. Baccanale, E.J. O'Flaherty, W.I. Manton, D.L.
Schanzer, J. Blenkinsop and C.M. Edwards. 1995. "Transplacental Transfer of Lead in
Non-Human Primates (Macaca fascicularis): Use of Serially Administered Stable Isotope
Tracers in Lead to Elicit Contribution of Maternal Bone Lead to Blood Lead and the
Fetus." Poster presented at the 1995 meeting of the Society of Toxicology, Baltimore,
MD. The Toxicologist Vol.15, p.194.
Froiman, Gail. 1996. U.S. Environmental Protection Agency (EPA). Personal
communication to Debbie Boger. April.
Gaul, L.E. and A.H. Staud. 1935. "Clinical spectroscopy. Seventy Cases of Generalized
Argyrosis Following Organic and Colloidal Silver Medication." J. Am. Med. Assoc.
Vol.104, pp. 1387-1390. Cited in IRIS, 1998 and U.S. EPA, 1991.
Goyer, R.A. 1990. "Transplacental Transport of Lead." Environmental Health
Perspectives Vol. 89, pp. 101-105.
3-152
-------
Graziano, J.H., D. Popovac, P. F actor-Li tvak, P. Shrout, J. Kline, MJ. Murphy, Y. Zhao,
A. Mehmeti, X. Ahmedi, B. Rajovic, Z. Zvicer, D. Nenezic, N. Lolacono and Z. Stein.
1990. "Determinants of Elevated Blood Lead Levels during Pregnancy in a Population
Surrounding a Lead Smelter Kosovo, Yugoslavia." Environmental Health Perspectives
Vol. 89, pp.95-100.
Gupta, B.N., R.N. Khanna and K.K. Datta. 1979. "lexicological Studies of Ammonium
Sulfamate in Rat after Repeated Oral Administration." Toxicology Vol.13, pp.45-49.
Hampshire Research Institute. 1995. Risk*Assistant™ Version 2.0 software.
Harrison, P.T., P. Grasso and V. Badescu. 1991. Title not available. Food Chem.
Toxicio.
Vol. 29, p.367 (cited in Katz & Guest, 1994).
Howard, P.H., R.S. Boethling, W.F. Jarvis, W.M. Meylan and E.M. Michalenko. 1991.
Handbook of Environmental Degradation Rates. Lewis Publishers, Chelsea, MI.
Hazardous Substance Data Bank (HSDB). 1995. MEDLARS Online Information
Retrieval System, National Library of Medicine, National Toxicology Program (via
TOXNET), Bethesda, MD.
Hazardous Substance Data Bank (HSDB). 1998. MEDLARS Online Information
Retrieval System, National Library of Medicine, National Toxicology Program (via
TOXNET), Bethesda, MD. June.
International Agency for Research on Cancer (IARC). 1974. IARC Monographs on the
Evaluation of the Carcinogenic Risk of Chemicals to Humans Vol.7, pp.95-109. IARC,
Lyon, France.
International Agency for Research on Cancer (IARC). 1985. Hydrogen Peroxide. In:
IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans.
Allyl Compounds, Aldehydes, Epoxides and Peroxides, Vol. 36, pp. 285-314. IARC,
Lyon, France.
International Agency for Research on Cancer (IACR). 1987. IARC Monographs on the
Evaluation of the Carcinogenic Risk of Chemicals to Man. Overall evaluations of
carcinogenicity. An updating of Vols. 1- 42, p.64. IARC, Lyon, France.
International Agency for Research on Cancer (IARC). 1990. "Nickel and Nickel
Compounds." Monographs on the Evaluation of the Carcinogenic Risks to Humans.
Vol.49, pp.257-445. IARC, Lyon, France.
3-153
-------
International Agency for Research on Cancer (IARC). 1992. "Occupational Exposures
to Mists and Vapours from Sulfuric Acid and Other Strong Inorganic Acids." IARC
Monograph on the Evaluation of the Carcinogenic Risks to Humans. Occupational
Exposures to Mists and Vapours from Sulfuric Acid; and Other Industrial Chemicals.
Vol. 54, pp. 41-119. IARC, Lyon, France.
International Agency for Research on Cancer (IARC). 1993. "Cadmium and Cadmium
Compounds." IARC Monographs on the Evaluation of the Carcinogenic Risk of
Chemicals to Humans. Vol. 58, pp.119-237. IARC, Lyon, France.
IRIS. Integrated Risk Information System. 1998. U.S. EPA, Office of Health and
Environmental Assessment, Cincinnati, OH.
IRIS. Integrated Risk Information System. 1999. U.S. EPA, Office of Health and
Environmental Assessment, Cincinnati, OH.
IRIS. Integrated Risk Information System. 2000. U.S. EPA, Office of Health and
Environmental Assessment, Cincinnati, OH.
Ito, A., H. Watanabe, M. Naito and Y. Naito. 1981. "Induction of Duodenal Tumors in
Mice by Oral Administration of Hydrogen Peroxide. Gann. Vol. 72, pp. 174-175 (cited
in IARC, 1985).
Johnson, M.A. and J.L. Greger. 1982. "Effects of Dietary Tin on Tin and Calcium
Metabolism of Adult Males. Am. J. Clin. Nutr. Vol. 35, pp.655-660.
Monsalve, E.R. 1984. "Lead Ingestion Hazard in Hand Soldering Environments."
Presentation for the Eighth Annual Seminar, Solder Technology and Product Assurance.
February 22-24. Soldering Technology Branch, Product Assurance Division,
Engineering Department, Naval Weapons Center, China Lake, CA.
Mori, K. 1953. "Production of Gastric Lesions in the Rat by the Diet Containing Fatty
Acids." Gann. Vol. 44, pp.421-427 (cited in FDA, 1979).
National Institute for Occupational Safety and Health (NIOSH). 1976. A Guide to
Industrial Respiratory Protection. Cincinnati, OH: NIOSH, U.S. Department of Health
and Human Services. NEW Pub. 76-189.
National Institute for Occupational Safety and Health (NIOSH). 1994. NIOSH Manual
of Analytical Methods, 4th edition. Methods 7082 (Lead by Flame AAS), 7105 (Lead by
HGAAS), 7505 (Lead Sulfide), 8003 (Lead in Blood and Urine), 9100 (Lead in Surface
Wipe Samples). U.S. Department of Health and Human Services, Centers for Disease
Control, Cincinnati, OH.
3-154
-------
National Institute for Occupational Safety and Health (NIOSH). 1999. NIOSH Pocket
Guide to Chemical Hazards and Other Databases. (CD-ROM). NIOSH, Cincinnati, OH.
DHHS (NIOSH) Publication No. 99-115.
Perry, W.G., F.A. Smith and M.B. Kent. 1994. "The Halogens." In: Clayton, G.D. and
F.E. Clayton. Patty's Industrial Hygiene and Toxicology, 4th ed., Vol. II, Part F,
pp. 4487-4505. John Wiley and Sons, New York.
Pocock, S.J., A.G. Shaper, M. Walker, CJ. Wale, B. Clayton, T. Delves, R.F. Lacey,
R.F. Packham and P. Powell. 1983. "Effects of Tap Water Lead, Water Hardness,
Alcohol, and Cigarettes on Blood Lead Concentrations." J. Epi. Comm. Health. Vol. 37,
pp.1-7.
Pozzani, U.C. and C.P. Carpenter. 1954. "Response of Rats to Repeated Inhalation of
Ethylenediamine Vapors." Arch. Ind. Hyg. Occup. Med. Vol. 9, pp.223-226. (cited in
U.S. EPA, 1998).
RTECS. Registry of Toxic Effects of Chemical Substances. 1998. MEDLARS Online
Information Retrieval System, National Library of Medicine.
Robinson, R.B., C.D. Cox, N.D. Jackson and M.B. Swanson. 1997. "Estimating Worker
Inhalation Exposure to Chemicals From Plating Baths at Printed Wiring Board
Facilities." Presented at the 1997 Canadian Society of Civil Engineers - American
Society of Civil Engineers Environmental Engineering conference, Edmonton, Alberto,
Canada. July.
Robinson, R.B., C.D. Cox and J. Drucker. 1999. Prediction of Water Quality From
Printed Wiring Board Processes. Final report to the University of Tennessee Center for
Clean Products and Clean Technologies and to the U.S. Environmental Protection
Agency. Part of the Verification of Finishing Technologies Project, EPA Grant
X825373-01-2 (Amendment No. 2). August.
Rodrigues, C., E. Lok, E. Nera, F. Iverson, D. Page, K. Karpinski and D.B. Clayson.
1986. Title not available. Toxicology 38:103 (cited in Katz & Guest, 1994).
Research Triangle Institute (RTI). 1999. Toxicological Profile for Lead (Update).
Prepared for the U.S. Department of Health and Human Services, Public Health Service,
Agency for Toxic Substances and Disease Registry, Atlanta, GA.
Schroeder, H.A. and M. Mitchener. 1971. "Scandium, Chromium (VI), Gallium,
Yttrium, Rhodium, Palladium, Indium in Mice: Effects on Growth and Life Span." J
Nutr. Vol. 101, pp.1431-1438.
Sharp, John. 1999. Teradyne, Inc. Personal communication to Dipti Singh at U.S.
Environmental Protection Agency (EPA). September.
3-155
-------
Shell Oil (Shell Oil Company). 1992. Final Report Teratologic Evaluation of Ethylene
Glycol (CAS No. 107-21-1) Administered to CD Rats on Gestational Days 6 through 15.
Cover letter dated March 5. HSE-84-0072 (002334). 8EHQ-0492-3402 Init.
Shepard, T.H. 1986. Catalog of Teratogenic Agents, 5th ed. The Johns Hopkins
University Press, Baltimore, p. 38 (cited in HSDB, 1998).
Sherlock, J.C., D. Ashby, H.T. Delves, G.I. Forbes, M.R. Moore, WJ. Patterson, SJ.
Pocock, MJ. Quinn, W.N. Richards and T.S. Wilson. 1984. "Reduction in Exposure to
Lead from Drinking Water and its Effect on Blood Lead Concentrations." Human
Toxicol. Vol. 3, pp.383-192.
Sleet, R.B. et al. 1989. National Toxicology Program, Report NTP 89-058.
NTIS PB89-165849 (cited in Gingell et al., 1994).
Sollmann, T. 1921. Title not available. J. Pharmacol. Exp. Then. Vol. 16, pp.463-474
(cited in Katz, Guest and Astill, 1982).
Stanek, Edward J., Edward J. Calabrese, Ramon Barnes and Penelope Pekow. 1997.
"Soil Ingestion in Adults." Ecotoxicology and Environmental Safety. Vol. 36, pp.249-
257.
Syracuse Research Corporation (SRC). Updated periodically. Environmental Fate
Database (EFDB©). Syracuse, NY.
Union Carbide (Union Carbide Chemicals and Plastics Company, Inc.). 1991.
Evaluation of the Teratogenic Potential of Ethylene Glycol Aerosol in the CD Rat and the
CD-I Mouse. Cover letter dated November 7. 8EHQ-1191-1508 Init.
U.S. Air Force. 1990. "Copper - Elemental Copper." In: The Installation Restoration
Toxicology Guide. Wright-Patterson Air Force Base, OH. Vol. 5, pp. 77-1 - 77-44.
U.S. Environmental Protection Agency (EPA). 1984. Health Effects Assessment for
Copper. Office of Research and Development, Office of Emergency and Remedial
Response, Washington, D.C., Office of Health and Environmental Assessment,
Environmental Criteria and Assessment Office, Cincinnati, OH.
U.S. Environmental Protection Agency (EPA). 1987a. Health Effects Assessment for
Tin. Prepared for Office of Solid Waste and Emergency Response by Environmental
Criteria and Assessment Office, Cincinnati, OH.
U.S. Environmental Protection Agency (EPA). 1987b. Code of Federal Regulations.
40CFR50.12.
3-156
-------
U.S. Environmental Protection Agency (EPA). 1988a. Reportable Quantity Document
for Hydrogen Peroxide. EPA Office of Solid Waste and Emergency Response,
Environmental Criteria and Assessment Office, Cincinnati, OH, 9 p.
U.S. Environmental Protection Agency (EPA). 1988b. Health and Environmental
Effects Document for Ethylenediamine. Prepared by Environmental Criteria and
Assessment Office, Office of Health and Assessment, Cincinnati, OH, for the Office of
Solid Waste and Emergency Response.
U.S. Environmental Protection Agency (EPA). 1989. Risk Assessment Guidance for
Superfund, Volume 1: Human Health Evaluation Manual (Part A). Interim Final. Office
of Emergency and Remedial Response. Washington, D.C. EPA/540/1-89/002.
(Appendix A).
U.S. Environmental Protection Agency (EPA). 1990. Supplement to the 1986 EPA Air
Quality Criteria Document for Lead - Volume 1 Addendum. Office of Research and
Development, Office of Health and Environmental Assessment, Washington, D.C.
EPA-600/8-89/049A.
U.S. Environmental Protection Agency (EPA). 1991a. Chemical Engineering Branch
Manual for the Preparation of Engineering Assessments. EPA Office of Toxic
Substances, Washington, D.C.
U.S. Environmental Protection Agency (EPA). 1991b. Human Health Evaluation
Manual, Supplemental Guidance: Standard Default Exposure Factors. EPA Office of
Solid Waste and Emergency Response, Washington D.C.
U.S. Environmental Protection Agency (EPA). 1991c. Drinking Water Health Advisory
for Silver. Office of Water, Washington, D.C. NTIS No. PB92-135516.
U.S. Environmental Protection Agency (EPA). 1992a. Dermal Exposure Assessment:
Principles and Applications, Interim Repot. EPA Office of Research and Development,
Washington, D.C. EPA/600/18-91/01 IB.
U.S. Environmental Protection Agency (EPA). 1992b. Guidelines for Exposure
Assessment. Washington, D.C. EPA 600-2-92-001.
U.S. Environmental Protection Agency (EPA). 1994. Guidance Manual for the
Integrated Exposure Uptake Biokinetic Model for Lead in Children. EPA/540/R-93/081,
and IEUBK Model, version 0.99D.
U.S. Environmental Protection Agency (EPA). 1995. Supplemental Guidance to RAGS:
Region 4 Bulletins, Human Health Risk Assessment (Interim Guidance). Waste
Management Division, Office of Health Assessment.
3-157
-------
U.S. Environmental Protection Agency (EPA). 1996a. "Recommendations of the
Technical Workgroup for Lead for an Interim Approach to Assessing Risks Associated
with Adult Exposures to Lead in Soil." US EPA Technical Workgroup for Lead,
December.
U.S. Environmental Protection Agency (EPA). 1996b. "Proposed Guidelines for
Carcinogen Risk Assessment." EPA Office of Research and Development, Washington,
D.C. EPA/600/P-92/1003C.
U.S. Environmental Protection Agency (EPA). 1997a. Exposure Factors Handbook.
EPA Office of Research and Development, Washington, D.C. EPA/600/p-95/1002Fa.
U.S. Environmental Protection Agency (EPA). 1997b. "Health Effects Assessment
Summary Tables. FY_1997." Office of Research and Development, Office of Energy
and Remedial Response.
U.S. Environmental Protection Agency (EPA). 1997c. Toxic Release Inventory Relative
Risk-Based Environmental Indicators: Interim Toxicity Weighting Summary Document.
Office of Pollution, Prevention and Toxics (MS 7406), Washington, D.C. June.
U.S. Environmental Protection Agency (EPA). 1998a. Printed Wiring Board Pollution
Prevention and Control Technology: Analysis of Updated Survey Results. EPA Office of
Pollution Prevention and Toxics, Washington, D.C. EPA 744-R-98-003. August.
U.S. Environmental Protection Agency (EPA). 1998b. Printed Wiring Board Cleaner
Technologies Substitutes Assessment: Making Holes Conductive. Design for the
Environment Printed Wiring Board Project. EPA Environmental Technology Initiative
Program, Washington, D.C. EPA 744-R-97/002a. June.
World Health Organization (WHO). 1974. Toxicological Evaluation of Certain Food
Additives with a Review of General Principles and of Specifications. Seventeenth Report
of the Joint FAO/WHO Expert Committee on Food Additives. WHO Technical Report
Series No. 539 (cited in U.S. EPA, 1997 [HEAST]).
World Health Organization (WHO). 1986. Regional Office for Europe: Air Quality
Guidelines. Vol. 11, 1-34. Geneva, Switzerland: World Health Organization.
3-158
-------
Chapter 4
Competitiveness
4.1 PERFORMANCE DEMONSTRATION RESULTS
4.1.1 Background
This section of the Cleaner Technologies Substitutes Assessment (CTSA) summarizes the
performance testing of the surface finishing technologies. To conduct the performance
evaluation, a test board was designed and fabricated, then 16 different surface finishes were
applied at 13 volunteer printed wiring board (PWB) manufacturing sites, in the U.S. and England,
during "performance demonstrations" between May and July, 1998. The performance of the
alternative surface finishes, taken in conjunction with risk, cost, and other information in this
document, provides a more comprehensive assessment of alternative technologies.
In a joint and collaborative effort, Design for the Environment (DfE) project partners
organized and conducted the performance demonstrations. The demonstrations were open to any
supplier of surface finishing technologies who chose to submit product line information and
nominate appropriate demonstration sites. Prior to the start of the demonstrations, DfE project
partners advertised the project and requested participation from all interested suppliers through
trade shows, conferences, and direct telephone calls.
A summary of the methodologies used and key results are presented in this chapter.
Additional results, details on testing and analysis methodologies, and more information on the test
board design, can be found in Appendix F.
The assembled PWBs used in the performance demonstration provided electrical
responses for 23 individual circuits that fall into seven major circuit groups. The first four circuit
groups had both plated through hole (PTH) and surface mount technology (SMT) components.
• high current low voltage (HCLV),
• high voltage low current (HVLC),
high speed digital (HSD),
• high frequency low pass filter (HF LPF),
• high frequency transmission line coupler (FIF TLC),
• leakage networks, and
• stranded wire (SW).
The design of the assembled PWB made it an excellent discriminating test vehicle to
discover problem areas associated with new technologies, materials, and processes. Test boards
were exposed to the following test conditions ("environmental testing") to accelerate the
discovery process:
4-1
-------
85/85: 85°C and 85% relative humidity for three weeks),
thermal shock (TS): 200 cycles between -50°C and 125°C, and
• mechanical shock (MS): dropped 25 times from a height of three meters onto a concrete
surface)
In general, each of the surface finishes applied during the performance demonstrations
were very robust under the conditions of the three tests. Some problem areas did develop,
however, during the testing. In particular, a problem area with HF LPF circuits related to open
PTHs was identified. The number of HF LPF anomalies was compared to the amount that would
be expected under a hypothesis that anomalies are independent of surface finish. This analysis led
to the following summary statements about the HF LPF circuits with respect to each surface
finish:
• Hot air solder leveling (HASL) anomalies were close to the expected values throughout
the three tests.
• Nickel/gold had far fewer anomalies than expected for all circuits.
• Nickel/palladium/gold had far fewer anomalies than expected for all circuits.
• Organic solderability preservative (OSP) anomalies were close to expected values, except
for one HF LPF SMT circuit where there were more anomalies than expected.
• Immersion silver had many more anomalies than expected for all circuits.
• Immersion tin anomalies were close to expected for PTH circuits, but were higher than
expected for SMT circuits.
The number of open PTH anomalies in the HF LPF circuit may have been related to the
inherent strength of the metals, as well as to board design (i.e., the small diameter vias in this
circuit). Product designers should be aware of these phenomena when considering a change of
surface finishes.
Other notable anomalies were in the HCLV SMT and HVLC SMT circuits in the
mechanical test, during which SMT components across all surface finishes fell off the board.
A failure analysis was conducted on the test boards that failed the 85/85 test and on a
control group not subjected to the test, in order to determine if any links existed between board
contamination from fabrication and assembly process residues and the electrical anomalies. In
addition, the boards were inspected visually to identify any obvious anomalies or defects. The
results indicated that the failures were not a result of residue, and that solder cracking was the
most common visual defect. HASL had more solder cracks than the other finishes.
4.1.2 Performance Demonstration Methodology
The general plan for the performance demonstration was to collect data on alternative
surface finishing processes, during actual production runs, at sites where the processes were
already in use. These demonstration sites were production facilities, customer testing facilities
(beta sites), or supplier testing facilities. Whenever possible, production facilities were used.
Each demonstration site received standardized test boards, which were run through the surface
4-2
-------
finishing operation during their normal production. The information collected through the
demonstrations was intended to provide a "snapshot" of the way the technology was performing
at that particular site at that particular time. This methodology was developed by consensus with
the technical workgroup, which included suppliers, trade association representatives, EPA, and
PWB manufacturers. A detailed performance demonstration methodology is included in
Appendix F.
Each supplier was asked to submit the names of up to two facilities at which the
demonstrations of their technology were to be conducted. This selection process encouraged the
suppliers to nominate the facilities where the technology was performing at its best. This, in turn,
provided for more consistent comparisons across technologies. The demonstration sites included
13 facilities, at which 16 different demonstrations were run.
To minimize differences in performance due to processes other than surface finishing, the
panels used for testing were all manufactured at one facility, Network Circuits, in Irving, Texas.
After fabrication, the panels were numerically coded for tracking purposes, and six panels
(containing four boards per panel) were shipped to each demonstration site, where the appropriate
surface finish was applied.
An observer from the DfE project team was present at each demonstration site to monitor
the processing of the test panels. Observers were present to confirm that the processing was
completed according to the methodology and to record data. Surface finish application at each
demonstration site was completed within one day, while performance processing at all sites was
completed over a three-month period.
When the processing was completed, the panels were put into sealed bags and shipped to
a single facility, which acted as a collection point for all performance demonstration panels.
Completed panels were then shipped back to Network Circuits, where the panels were cut into
boards. All coded boards were then shipped to a single facility (EMPF/American Competitiveness
Institute) for assembly. One subgroup was assembled using low-residue (LR) flux and the other
with water soluble (WS) flux.
Following assembly, the boards were sent to Raytheon Systems Inc. in McKinney, Texas,
where the performance characteristics of the assembled boards were tested. Testing included
Circuit Electrical Performance and Circuit Reliability testing. The Electrical Performance testing
assessed the circuit performance of the printed wiring assemblies (PWAs), or assembled PWBs,
before and after exposure to 85°C temperatures at 85% relative humidity for 3 weeks. For the
Circuit Reliability testing, the same PWAs were tested after being subject to thermal shock and
mechanical shock conditions.
Limitations of the Performance Demonstration Methodology
The performance demonstration was designed to provide a snapshot of the performance of
different surface finishing technologies. Because the demonstration sites were not chosen
4-3
-------
randomly, the sample may not be representative of all PWB manufacturing facilities in the U.S.
(However, there is no specific reason to believe that they are not representative.)
4.1.3 Test Vehicle Design
The test vehicle design was based on a test board designed by the Sandia National
Laboratory Low-Residue Soldering Task Force (LRSTF). This test vehicle was used by the
Circuit Card Assembly and Materials Task Force (CCAMTF), a joint industry and military
program evaluating several alternative surface finishing technologies. The design is a functional
PWA designed to test process effects resulting from changing materials and processes. The PWA
and the test/data analysis methodology are considered excellent discriminators in comparing
processes fluxes, surface finishes, or other process technologies. It should be noted that circuit
technology continues to change rapidly; the PWA design is based on 1994 technology and does
not incorporate some more recent, state-of-the-art, circuitry. It is unknown whether the results
might have differed with newer circuit technology. However, the test PWA was designed to
contain approximately 80 percent of the circuitry used in military and commercial electronics. In
addition, use of this test vehicle by the DfE PWB Project provided great savings in cost and time
that would be required to develop a new test vehicle, and allows for some comparison with
CCAMTF results.
The test vehicle was designed to be representative of a variety of extreme circuits: high
voltage, high current, HSD, low-leakage current, and high frequency (HF) circuits. A designer
can use the resulting measurements to make some analytical judgments about the process being
tested. The test PWA was not intended to be a "production" board, which would typically be too
narrow in breadth to represent a wide variety of these circuit extremes. Even though some
technology complexities/advancements are not duplicated, the basic types are represented, and
comparison of baseline technologies can be extrapolated, in some cases, to more current
technology by analysis. The performance results are assessed based on the acceptance criteria
developed by the CCAMTF project, which are described in Table 4-1.
The test PWA measures 6.05" x 5.8" x 0.062". See Appendix F for more details on the
design of the test PWA. The PWA is divided into six sections, each containing one of the
following types of electronic circuits:
HCLV;
HVLC;
HSD,
HF;
SW; and
• other networks.
The components in the HCLV, HVLC, HSD, and HF circuits represent two principal
types of soldering technology:
• PTH: Leaded components are soldered through vias in the circuit board by means of a
wave soldering operation.
4-4
-------
• SMT: Leadless components are soldered to pads on the circuit board by passing the
circuit board through a reflow oven
The other networks used for current leakage measurements are 10-mil pads, a socket for a
pin grid array (PGA), and a gull wing. The two stranded wires (SW) are hand soldered.
The test vehicle provides 23 separate electrical responses as shown in Table 4-1. The
criteria for 17 of the 23 circuits require a comparison to the pre-test measurements (i.e., before
exposure to test environments), while the criteria for the remaining six circuits are based on
absolute responses. The CCAMTF project conducted baseline testing for 480 test PWAs, which
were used as the basis of the acceptance criteria that were published in their Joint Test Protocol.
These criteria are also shown in Table 4-1.
It should be noted that these acceptance criteria are not absolutes, but rather guidelines
based on engineering judgement and experience with the particular circuit. Therefore, in some
cases when values that are just outside the acceptance criterion, they may be considered "not of
practical significance." This would be the case when a single observation is close to the
acceptance criterion. For example, if the criterion specifies an acceptable increase of lOdB and
the increase for one board was measured at 10.2dB, it would be difficult to make any conclusion
from a single observation so close to the acceptance criterion. However, if all HASL boards, for
example, measured 10.2dB while all other surface finishes were below lOdB, it would be
reasonably clear that there is an effect due to the surface finish.
The test PWAs were manufactured with the following six surface finishes for the DfE
PWB Project:
HASL;
• nickel/gold;
• nickel/palladium/gold;
OSP;
• immersion silver; and
• immersion tin.
Additional information about each technology, including a process flow diagram and a description
of each process step, is presented in Section 2.1, Chemistry of Use and Process Description.
4-5
-------
Table 4-1. Electrical Responses for the Test PWA and Acceptance Criteria
Electrical
Response
Circuitry
Acceptance Criteria
High Current Low Voltage
1
2
HCLV PTH
HCLV SMT
Change in voltage from pre-test < 0.50V
Change in voltage from pre-test < 0.50V
High Voltage Low Current
3
4
HVLC PTH
HVLC SMT
4|iA < x < 6|iA
4|iA < x < 6|iA
High Speed Digital
5
6
HSD PTH Propagation Delay
HSD SMT Propagation Delay
< 20% increase from pre-test
< 20% increase from pre-test
High Frequency Low Pass Filter
7
8
9
10
11
12
HF PTH 50MHz
HFPTHf(-3dB)
HFPTHf(-40dB)
HF SMT 50MHz
HFSMTf(-3dB)
HFSMTf(-40dB)
Within ± 5dB of pre-test
Within ± 50MHz of pre-test
Within ± 50MHz of pre-test
Within ± 5dB of pre-test
Within ± 50MHz of pre-test
Within ± 50MHz of pre-test
High Frequency Transmission Line Coupler
13
14
15
16
17
HF TLC 500MHz Forward Response
HF TLC 500MHz Forward Response
HF TLC IGHz Forward Response
HF TLC Reverse Null Frequency
HF TLC Reverse Null Response
Within + [±?] 5dB of pre-test
Within ± 5dB of pre-test
Within ± 5dB of pre-test
Within ± 50MHz of pre-test
< lOdB increase over pre-test
Other Networks: Leakage
18
19
20
21
10-milPads
PGA-A
PGA-B
Gull Wing
Resistance > 7.7 Iog10 ohms
Resistance > 7.7 Iog10 ohms
Resistance > 7.7 Iog10 ohms
Resistance > 7.7 Iog10 ohms
Stranded Wire
22
23
Stranded Wire 1
Stranded Wire 2
Change in voltage from pre-test< 0.356V
Change in voltage from pre-test < 0.356V
Abbreviations and Definitions:
HCLV - high current low voltage
HF - high frequency
HSD - high speed digital
HVLC - high voltage low current
PGA - pin grid array
PTH - plated through hole
SMT - surface mount technology
TLC - transmission line coupler
4-6
-------
These surface finishes were applied at one or more of the different demonstration sites.
Table 4-2 provides a summary of the 164 PWAs that were subjected to environmental testing by
surface finish, manufacturing site, and flux type. Table 4-2 also shows that both fluxes were not
used with all demonstration sites, and that 84 PWAs were processed with low residue flux, while
80 PWAs were processed with water soluble flux.
Table 4-2. Distribution of the Number of LRSTF PWAs by Surface Finish, Site, and Flux
Surface Finish
HASL
Nickel/Gold
Nickel/Palladium/Gold
OSP
Immersion Silver
Immersion Tin
Site
1
2
3
13
14
15
16
4
5
6
11
12
7
8
9
10
Total No. of Boards:
Low Residue Flux
8
8
—
4
8
—
8
4
8
8
8
—
4
8
8
—
84
Water Soluble Flux
8
—
8
8
—
8
4
—
8
8
4
8
8
—
—
8
80
Due to the uneven distribution of fluxes during assembly, the number of PWAs are
different for each surface finish, as follows:
Surface Finish
HASL
Nickel/Gold
Nickel/Palladium/Gold
OSP
Immersion Silver
Immersion Tin
Total
No. of PWAs (Percent of total)
32 (19.5%)
28(17.1%)
12 ( 7.7%)
36 (22.0%)
20 (12.2%)
36 (22.0%)
164
4-7
-------
4.1.4 Environmental Testing Methodology
Each of the 164 PWAs summarized in Table 4-2 was exposed to the following
environmental test sequence:
• Exposure to three weeks of 85°C and 85% relative humidity.
• 200 cycles of thermal shock with the PWAs rotated between chambers at -50°C and
125°C with 30 minute dwells at each temperature.
• Mechanical shock where the PWA is mounted in a rectangular fixture and dropped 25
times on a concrete surface from a height of 1 meter.
3 weeks of 85/85 I —> I Thermal Shock I —> I Mechanical Shock
The PWAs were functionally tested prior to exposure to these environments and after each
environment. Although the sequential nature of the tests may affect the results (i.e., the PWAs
may be weakened in the mechanical test because of the previous two tests), the testing sequence
was planned to minimize any carryover effect. The 85/85 environment was the first test because it
is relatively benign with respect to impacts on the functionality of the PWA. In contrast, the
mechanical shock test was performed last because it can cause separation of SMT components
and therefore permanent damage to the PWA.
4.1.5 Analysis of the Test Results
General Linear Models
General linear models (GLMs) were used to analyze the test data for each of the 23
electrical circuits in Table 4-1 at each test time. The GLM analysis determines which
experimental factors or combinations of factors (interactions) explain a statistically significant
portion of the observed variation in the test results, and in quantifying their contribution.
Analysis of Variance and Multiple Comparisons of Means
Another statistical approach can be used to determine which groups of site/flux means are
significantly different from one another for a given electrical response from the test PWA. This
procedure begins with an analysis of variance (ANOVA) of the test results (Iman, 1994) for a
given circuit. An ANOVA is perhaps best explained via an example.
4-8
-------
An ANOVA performed on the 164 pre-test measurements for HCLV PTH produced the
following:
Source DF Sum of Squares Mean Square F-Statistic P-Value
Site/Flux 22 0.2908 0.0132 0.70 0.838
Error 141 2.6796 0.0190
Total 163 2.9704
The meaning of the terms in each of the columns of the ANOVA table is now given.
Source. The entries in this column represent the following:
• Site/Flux refers to the 23 site/flux combinations listed in Table 4-3.
• Error refers to the random/unexplained variation in the HCLV PTH voltage
measurements.
• Total refers to the total variation in the data.
Degrees of Freedom. The numbers in this column represent a statistical term known as
the degrees of freedom (DF). The degrees of freedom associated with each source are calculated
as follows:
Site/Flux 23-1 (the number of site/flux combinations - 1) = 22
Error Total DF - Site/Flux DF = (164 - 1) - (23 -1) = 163 - 22 = 141
Total 164-1 (the number of test measurements -1) = 163
Sum of Squares. The entries in this column are the sums of squares associated with each
source of variation. The Total Sum of Squares is calculated by summing the squares of the
deviations of the 164 data points from the sample mean. If this number were divided by 164 - 1,
the result would be the usual sample variance (i.e., s2 = 2.9704/163 = 0.0182). The other sums
of squares in this column represent a partitioning of the total sum of squares. Note that they sum
to the total sum of squares:
0.2908 + 2.6796 = 2.9704
The calculations for these other sums of squares are somewhat more involved than the
total sum of squares and will not be discussed here. The interested reader can find details of these
calculations in Iman, 1994.
Mean Square. The values in this column are obtained by dividing the sum of squares in
each row by their respective degrees of freedom:
Mean Square for Site/Flux = 0.2908/22 = 0.0132
Mean Square for Error = 2.6796/141 = 0.0190
4-9
-------
The Mean Square Error calculation is an estimate of the standard error for the experiment.
Note that this estimate (0.0190) differs from the sample variance (0.0182), as the standard error is
computed after the other source of variation in the data (Site/Flux) has been taken into account.
Note that these two variance estimates are close in this particular example, but they can differ
greatly.
F-Statistic. This column contains the f-statistic that is used to determine if Site/Flux
makes a statistically significant contribution to the total variation. The f-statistic is the ratio of the
Mean Square for Site/Flux and Mean Square Error:
F-statistic = (Mean Square Site/Flux)/(Mean Square Error) = 0.0132/0.0190 = 0.70
In the surface finishes analysis, when the f-statistic was significant, the Least Significant
Difference (LSD) procedure (described below) was used to compare the means of the 23 site/flux
combinations given in Table 4-3, and results are displayed in boxplots (also described below).
P-Value. The statistical significance of the f-statistic (the p-value) is given in the last
column of the ANOVA table. This value is determined by comparing the f-statistic to its
probability distributions (the larger the f-statistic, the more significant the contribution).
Probability distributions for f-statistics are indexed by two parameters known as degrees of
freedom. The degrees of freedom for the f-statistic are Site/Flux DF = 22 and Error DF = 141.
The p-value is computed as the tail probability for the f-statistic as:
P-Value = Prob(F22jl41>FsiteMJ = Prob(F22jl41 > 0.70) = 0.838
Whenever a p-value in this analysis is less than 0.01, the corresponding source of variation
can be regarded as making a significant contribution to the overall variation. In this example the
p-value is quite large, which signifies that Site/Flux does not make a significant contribution to the
overall variation in the data. Thus, there is no need to check for differences in the means of the
site/flux combinations. In statistical analyses, the level of significance frequently is a p-value less
than 0.05. The more stringent 0.01 is used in this report because of the relatively small cell sizes
(4 or 8 samples per circuit per site/flux combination) and the potentially important decisions that
may be made based on these test results.
Least Significant Difference. In the event that the p-values associated with the F-
statistics are less than 0.01, the sample means can be incorporated into a test statistic for
determining which population means are significantly different from one another. The "measuring
stick" used to compare the sample means is known as Fisher's LSD. In particular, two population
means are declared significantly different from one another if the absolute difference of the
corresponding sample means exceeds LSDa, which is defined as:
L^Da= WfcV^XE
\
1+1
H, H,
4-10
-------
where,
a,
t
MSB
ri and
level of significance
the a/2 quantile from a Student's t distribution with n-k degrees of freedom
mean square error for the model
sample sizes for the means being compared
When the F-statistic is significant, the LSD procedure will be used to compare the means
of the 23 site/flux combinations given in Table 4-3. Results of the multiple comparisons will be
displayed in boxplots (described below).
Table 4-3. Listing of 23 Site/Flux Combinations Used in the Multiple
Comparisons Analyses
Site/Flux
Combination
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Surface Finish
HASL
HASL
HASL
HASL
OSP
OSP
OSP
OSP
OSP
Immersion Tin
Immersion Tin
Immersion Tin
Immersion Tin
Immersion Tin
Immersion Silver
Immersion Silver
Immersion Silver
Nickel/Gold
Nickel/Gold
Nickel/Gold
Nickel/Gold
Nickel/Palladium/Gold
Nickel/Palladium/Gold
Flux Type
LR
WS
LR
WS
LR
WS
LR
WS
LR
LR
WS
LR
LR
WS
LR
WS
WS
LR
WS
LR
WS
LR
WS
Site No.
1
1
2
3
4
4
5
5
6
7
7
8
9
10
11
11
12
13
13
14
15
16
16
No. of Observations
8
8
8
8
4
8
8
8
8
4
8
8
8
8
8
4
8
4
8
8
8
8
4
Abbreviations and Definitions:
LR - low residue
WS - water soluble
4-11
-------
Boxplot Displays
A boxplot is simply a rectangular box with lines extending from the left-hand and right-
hand sides of the box as shown below. The left-hand side of the box represents the lower quartile
(X25) or lower 25 percent of the sample data. The right-hand side of the box represents the upper
quartile (X75), or upper 25 percent of the sample data (or lower 75 percent). Thus, the box
covers the middle 50 percent of the sample data. A vertical line inside the box connecting the top
and bottom sides represents the sample median (X 50).
The interquartile range (IQR) is the difference between the upper quartile and the lower
quartile. A horizontal line at the right-hand side of the box extends to the maximum observation
in the interval from X75 to X75 +1.5 IQR. This line never extends beyond X75 +1.5 IQR. A
horizontal line on the left-hand side extends to the smallest observation between X7, and X7, - 1.5
IQR. This line never extends below X25 - 1.5 IQR. Any observations outside of these limits are
regarded as outliers and are marked with an asterisk or other symbols. A heavy dot is frequently
added to a boxplot to identify the sample mean.
Boxplots can be constructed in either a horizontal or vertical position. When the boxplot
is constructed vertically, the top side of the box represents the upper quartile and the bottom side
represents the lower quartile.
Boxplot displays have advantages over traditional plots of means, such as: 1) the median
is not heavily influenced by outlying or unusual observations that can be misleading; and 2) the
information about the variability in the data, captured in a boxplot, is lost in a plot of the means.
Boxplots will be used as the basis for graphical displays of the multiple comparisons results for
each electrical response for each test.
Lower
Quartile
Median
Upper
Quartile
v
A
.2S
X.
75
A Boxplot Used to Display Test Results
4-12
-------
4.1.6 Overview of Test Results
The 164 PWAs as summarized in Table 4-2 were functionally tested at the following four
times:
• Pre-test;
Post-85/85;
Post-TS; and
Post-MS.
At each of these test times, 3,772 electrical test measurements were recorded (164 PWAs
x 23 individual circuits). An overall summary of success rates based on 3,608 measurements1 at
each test time is shown in Table 4-4.
Table 4-4. Number of Anomalies Observed at Each Test Time
Test Time
Pre-test
Post-85/85
Post-TS
Post-MS
Anomalies
2
17
113
527
Success Rate
99.9%
99.6%
96.9%
85.4%
Abbreviations and Definitions:
MS - mechanical shock
TS - thermal shock
An overview of the test results at each test time is discussed in this section. A discussion
of the results for each test time for each major circuit group is presented in Sections 4.1.7 through
4.1.13. An overview of the circuits meeting the acceptance criteria after each testing sequence is
summarized in Table 4-5 for each major circuit group.
1 Since HF TLC RNF gave a constant response of 50MHz throughout, there is no variability to analyze.
4-13
-------
Table 4-5. Percentage of Circuits Meeting Acceptance Criteria at Each Test Time
Circuitry
HCLV
HVLC
HSD
HFLPF
HFTLC
Other Networks
SW
Totals
85/85
100%
99.7%
99.7%
98.7%
99.8%
99.8%
100%
99.5%
Thermal Shock
100%
99.7%
98.8%
89.4%
99.5%
100%
99.7%
96.9%
Mechanical Shock
48.2%(7.1%SMT)
50.0% (0.0% SMT)
99.1% (99.3% SMT)
82.6% (74.8% SMT)
97.9%
100%
98.5%
85.4%
Abbreviations and Definitions:
HCLV - high current low voltage
HF LPF - high frequency low pass filter
HF TLC - high frequency transmission line coupler
HSD - high speed digital
HVLC - high voltage low current
SMT - surface mount technology
SW - stranded wire
Overview of Pre-Test Results
The electrical measurements were compared to the acceptance criteria given in Table 4-1
at each test time. Note that the acceptance criteria require a comparison to pre-test results for all
but six of the 23 electrical circuits (#'s 3, 4, 18-21 in Table 4-1). Hence, pre-test comparisons to
the acceptable criteria can only be made for those six circuits. There were no pre-test anomalies
observed for those six circuits. Pre-test measurements for the remaining 17 circuits were
compared to CCAMTF pre-test results. Table 4-6 presents this comparison of the ranges of the
measurements for each of the 23 circuits with pre-test measurements for the PWAs used in the
CCAMTF 85/85 testing.
Table 4-6 shows that the two sets of ranges for circuits 5 through 12 and 16 do not even
overlap. The lack of overlap in the ranges for the HSD PTH and HSD SMT circuits (#'s 5 and 6)
is due to different components being used on the DfE PWAs than were used in processing the
PWAs in the CCAMTF program. The differences in the HF LPF circuits 7 through 12 are more
difficult to pinpoint. The most likely explanation lies in the fact that the actual boards used in the
DfE program and those in the CCAMTF PWAs were produced by two different manufacturers.
FR-4 epoxy was used for the board laminate material. HF LPF responses are sensitive to the
dielectric constant of the board laminate material. Differences in FR-4 epoxy at the two
manufacturing locations used by the DfE program and the CCAMTF program could have affected
the dielectric constant and hence the HF LPF responses. Another possibility is that the board
The total number of measurements, rather than the number of measurements meeting the acceptance criteria
after the previous test, is used to calculate these percentages. While it is possible to adjust for anomalies resulting
from the previous test, doing so: 1) would make the calculation conditional on the previous test and therefore
would require a very careful interpretation ; and 2) would not reflect the "curing" that can occur with a circuit that
is an anomaly in one test but meets the acceptance criteria in the subsequent test.
4-14
-------
layers manufactured at the two locations might not be the same thickness. A microsection was
required to make this determination, which was beyond the scope of this analysis. A final
possibility is that the two sets of boards might have used a different lot of ceramic capacitors, but
this is not likely, as all the parts for the DfE and CCAMTF boards were ordered at the same time.
Table 4-6. Comparison of CCAMTF Pre-Test Ranges with DfE Pre-Test Measurements
CCAMTF Pre
Circuit [units] jyjjn j
-Test DfE Pre-Test
Max Min Max
1 HCLVPTH[V] 6.60 7.20 6.80 7.52
2 HCLVSMT[V] 6.96 7.44 7.00 7.44
3 HVLC PTH [|iA] 5.00 5.25 5.00 5.25
4 HVLC SMT [|iA] 4.92 4.97 4.81 5.39
5 HSD PTH Propagation Delay [|i sec] 12.66 13.50 16.76 18.20
6 HSD SMT Propagation Delay [|i sec] 4.28 5.45 8.89 9.52
7 HF PTH 50MHz [dB] -0.320 0.094 -1.176 -0.365
8 HFPTHf(-3dB) [MHZ] 239.4 262.6 274.4 287.5
9 HFPTHf(-40dB)[MHZ] 425.3 454.9 456.7 485.2
10 HF PTH 50MHz [dB] -0.296 0.081 -0.901 -0.617
11 HFSMTf(-3dB) [MHZ] 275.0 283.3 313.0 338.0
12 HFSMTf(-40dB) [MHZ] 642.6 674.0 811.2 951.9
13 HFTLC 5 0MHz Forward Response [dB] -49.74 -36.48 -50.87 -42.66
14 HF TLC 500MHz Forward Response [dB] -21.47 -17.54 -19.91 -15.28
15 HFTLC 1 GHz Forward Response [dB] -16.91 -12.08 -15.01 -12.89
16 HFTLC Reverse Null Frequency [MHZ] 624.2 659.8 50.0 79.7
17 HFTLC Reverse Null Response [dB] -74.53 -38.22 -43.67 -32.08
18 10-mil Pads [Iog10 ohms] 10.01 15.00 10.10 15.00
19 PGA-A [Iog10 ohms] 8.94 15.00 10.38 14.00
20 PGA-B [Iog10 ohms] 8.72 15.00 10.07 13.70
21 Gull Wing [Iog10 ohms] 9.71 14.00 9.01 13.70
22 SW 1 [mV] 5
23 SW2[mV] 19
19 7 19
28 19 28
Abbreviations and Definitions:
HCLV - high current low voltage
HF - high frequency
HSD - high speed digital
HVLC - high voltage low current
PGA - pin grid array
PTH - plated through hole
SMT - surface mount technology
SW - stranded wire
TLC - transmission line coupler
4-15
-------
The reverse response function (#16 in Table 4-6) provides a range of responses from
50MHz to IGHz. The low point of this curve is referred to as the null point. The coordinates of
the null point are the HF TLC Reverse Null Frequency (in MHZ) and FTP TLC Reverse Null
Response (in dB). The FIF TLC Reverse Null Frequency ranged from approximately 624MHz to
660MHz in the CCAMTF program, while the FTP TLC Reverse Null Response ranged from
approximately -75dB to -38dB. However, the null point of the reverse response function for the
DfE PWAs occurred at the beginning of the curve, which is approximately 50MHz. In fact, all
but two of the HF Reverse Null Frequency measurements were 50MHz with the others being
77.3MHz and 79.7MHz. The Reverse Null Response ranged from -43.7dB to -32. IdB. The
nearly constant value of HF Reverse Null Frequency relegates any subsequent analysis of the
uncertainty to a moot point. As discussed further in subsequent sections, none of the
discrepancies could be attributed to the performance of the surface finishes.
Overview of 85/85 Results
At the conclusion of the 85/85 test, 99.5% of the electrical measurements met the
acceptance criteria given in Table 4-1. There were 17 anomalies distributed across 10 PWAs, as
shown in Table 4-7. Among the PWAs with anomalies, five were assembled with the low-residue
flux and five were assembled with the water-soluble flux. The anomalies are summarized in
Appendix F, Table F-l. Table F-l also contains observations made by the testing technician that
are useful in identifying the source of the anomaly for those cases where a problem was obvious,
such as an open PTH, a burnt etch, or a failed device.
Table 4-7. Frequency Distribution of Post-85/85 Anomalies per PWA by Surface Finish
(Sample sizes are given in parentheses)
Number of
Anomalies per
PWA
None
1
2
3
Total Anomalies
HASL
(32)
31
1
3
Nickel/Gold
(28)
26
2
2
Nickel/Palladium/Gold
(12)
12
0
OSP
(36)
35
1
3
Immersion
Silver
(20)
18
1
1
4
Immersion
Tin
(36)
32
3
1
5
Overview of Thermal Shock Results
The number of anomalies increased from 17 at the post-85/85 test to 113 at the post-TS
test, so that 96.9% of the electrical measurements met the acceptance criteria given in Table 4-1.
Of the 17 anomalies at post-85/85, 16 carried over to post-TS, so that the thermal shock test
introduced 97 new anomalies. 91% of the post-TS anomalies occurred for HF LPF circuits. As
shown in Table 4-8, the 113 anomalies affected 42 PWAs, with 19 PWAs accounting for 88 of the
anomalies. Of the PWAs with anomalies, 16 were assembled with low residue flux and 26 were
assembled with water soluble flux. The anomalies are summarized by surface finish in Appendix
F, Table F-2. This summary includes several observations made by the testing technician that are
useful in identifying the source of the anomaly.
4-16
-------
A chi-square test of independence (Iman, 1994) indicates that the anomalies are not
uniformly distributed over the surface finishes, with immersion silver and immersion tin having
more than expected. The p-value for this test is 0.025. The chi-square test does not indicate a
difference in anomalies with respect to flux type.
Table 4-8. Frequency Distribution of Post-Thermal Shock Anomalies per PWA
by Surface Finish
(Sample sizes are given in parentheses)
Number of
Anomalies per
PWA
None
1
2
3
4
5
6
Total Anomalies
HASL
(32)
25
4
2
1
16
Nickel/Gold
(28)
25
1
2
7
Nickel/Palladium/
Gold
(12)
11
1
1
OSP
(36)
28
2
3
2
1
20
Immersion
Silver
(20)
11
2
3
1
3
33
Immersion
Tin
(36)
22
3
3
5
3
36
Overview of Mechanical Shock Results
The number of anomalies increased greatly from 113 at post-TS to 527 at post-MS. 85%
of the electrical measurements met the acceptance criteria given in Table 4-1. Of the 113
anomalies at post-TS, 97 carried over to post-MS, hence the mechanical shock test introduced
430 new anomalies. These new anomalies included 157 from HCLV SMT — in contrast, there
was only one HVLC SMT anomaly at post-TS. In addition, there were 163 new anomalies from
the HVLC SMT circuits. Thus, these two circuits accounted for 320 of the 430 new anomalies.
The anomalies for these two SMT circuits were attributable to SMT components coming off the
board during the execution of the mechanical shock test. This affected every board and has no
relation to site, surface finish, or flux.
All anomalies, except for those associated with HCLV SMT and HVLC SMT (since these
affected every PWA), are summarized in Appendix F, Table F-3. In addition, this table includes
comments made by the test technician. There were five minor stranded wire anomalies that are
not listed in Table F-3.
At post-MS, every PWA had at least one anomaly. Table 4-9 provides a breakdown of
the number of anomalies per PWA for each surface finish. The last row in this table gives the
median number of anomalies per PWA for each surface finish. The hypothesis that the mean
number of anomalies is the same for all surface finishes is easily rejected with a p-value of 0.000
based on the Kruskal-Wallis test (Iman, 1994). Immersion silver has the most anomalies per
PWA with nickel/gold and nickel/palladium/gold having the least. HASL and OSP had
approximately the same number of anomalies, with immersion tin slightly higher than these two.
4-17
-------
The following section provides insight on the source of the anomaly disparities relative to surface
finish.
Table 4-9. Frequency Distribution of Post-Mechanical Shock Anomalies
per PWA by Surface Finish
(Sample sizes are given in parentheses)
Number of
Anomalies per
PWA
none
1
2
3
4
5
6
7
8
9
10
Total Anomalies
Median
HASL
(32)
1
16
8
5
2
98
2
Nickel/Gold
(28)
1
20
5
1
1
65
2
Nickel/Palladium/Gold
(12)
11
1
25
2
OSP
(36)
14
14
3
3
2
113
3
Immersion
Silver
(20)
5
5
3
2
2
1
2
95
4
Immersion
Tin
(36)
12
11
2
3
6
1
1
131
3
4.1.7 High Current Low Voltage (HCLV) Circuitry Performance Results
Pre-test measurements and deltas were analyzed with GLM for the main effects of site and
flux and their interactions, where the base case was defined as HASL at Site 1 and processed with
low residue flux. These data were also subjected to a second GLM analysis for the main effects
of surface finish and flux, where the base case was defined as HASL processed with low residue
flux. The specific equations used for these two analyses are given in Appendix F as Equation F-l
and F-2, respectively.
The results of the GLM analyses indicate that the experimental parameters surface finish,
site, and flux do not significantly affect the HCLV voltage measurements at pre-test, nor do they
affect the changes in the voltage after exposure to each of the three test environments. That is,
the HCLV measurements are robust with respect to surface finish, site, and flux. The results for
the two GLMs used in the analysis are examined in more detail in Appendix F.
Multiple comparison procedures for comparing the means of the 23 site/flux combinations
given in Table 4-3 were explained previously. The overall ANOVA that precedes the use of
multiple comparisons produced a significant f-statistic only at post-MS for HCLV PTH. The
HCLV SMT was close to significance at post-TS with a p-value of 0.018, as shown in Table 4-10
(non-significant values > 0.0 1 have been shaded).
4-18
-------
Table 4-10. P-Values for HCLV Test Results
Test Time
Pre-test
Post-85/85
Post-TS
Post-MS
P- Value for HCLV PTH
0.838
0.953
0.496
0.001
P- Value for HCLV SMT
0.442
0.109
0.018
0.861
Abbreviations and Definitions:
HCLV - high current low voltage
MS - mechanical shock
PTH - plated through hole
SMT - surface mount technology
TS - thermal shock
Boxplot Displays of Multiple Comparison Results. Boxplot displays provide a
convenient way to display multiple comparison results. Multiple comparison procedures are only
justified for HCLV PTH at post-MS since the other f-statistics were not significant. However,
boxplot displays are given in Figure 4-1 to 4-8 for all HCLV circuits for purposes of comparison.
Figures 4-1 to 4-4 display the test results at each test time for HCLV PTH circuits and Figures 4-
5 to 4-8 do the same for HCLV SMT circuits. For improved readability, all boxplots referenced
in this chapter can be found in Section 4.1.16 at the end of the performance results discussion.
Additional boxplots, where findings were not significant, can be found in Appendix F.
Some explanation of the contents of each graph of boxplots should facilitate
understanding. The test time and circuit type are labeled in the upper left-hand corner of each
boxplot display. The numbers (1 to 23) on the horizontal axis in each figure correspond
respectively to the 23 site/flux combinations listed in Table 4-3. The label WS on the horizontal
axis signifies those demonstration sites for which water soluble flux was used; otherwise, the flux
type was low residue (LR). The boxplots are grouped by surface finish, which are identified with
labels across the top of each graph. At pre-test, the vertical axis corresponds to the absolute test
measurement. After pre-test, the vertical axis either corresponds to the absolute test
measurement or the difference from the pre-test measurement as specified in the acceptance
criteria. The sample mean is identified in each boxplot with a solid circle
Note that there is a lot of overlap in all boxplots in Figure 4-1, which is consistent with the
lack of significance in the f-statistics for equality of means and in the results for the GLMs. Also
note that the total variation in the boxplots is approximately 0.3V, which most likely is not of
concern. Figures 4-2 to 4-4 display the differences between the current HCLV PTH
measurements and those obtained at pre-test. Note that all differences in Figures 4-2 and 4-3 are
well below the acceptance criteria of AV < 0.5V. However, several of the differences are well
above the acceptance criteria following mechanical shock, as illustrated in Figure 4-4. The
significant difference in means in Figure 4-4 at post-MS is attributable mostly to immersion silver
at Site 17 processed with a water soluble flux. It should be noted, however, that the other two
immersion silver sites showed no anomalies. This may indicate a site-specific problem and not a
surface finish problem. Additional failure analysis would be needed to draw further conclusions.
4-19
-------
Figures 4-5 to 4-8 are similar to those for HCLV PTH. Figure 4-8 for the HCLV SMT circuit is
especially worthy of note as it reflects the increase in voltage due to the loss of one or two
resistors (as illustrated in detail in Equations 2.1 to 2.3 in Appendix F). The loss of resistors has
caused an increase in voltage of 2V to 3V, which exceeds the acceptance criteria.
Comparison to Acceptance Criteria. The acceptance criteria for HCLV PTH and
HCLV SMT (responses 1 and 2 in Table 4-1) are based on the following differences between test
measurements:
Delta 1
Delta 2
Delta 3
85/85 - pre-test
thermal shock - pre-test
mechanical shock - pre-test
Specifically, these differences are not to exceed 0.50V.
None of the HCLV PTH or HCLV SMT voltage measurements exceeded the acceptance
criterion of AV < 0.50V after exposure to 85/85 or thermal shock. However, following
mechanical shock there were 12 HCLV PTH anomalies and 158 HCLV SMT anomalies. Whereas
the HCLV SMT anomalies affected almost every PWA, the HCLV PTH anomalies were
distributed unevenly among surface finishes, as shown in Table 4-11.
Table 4-11. Number of HCLV PTH Anomalies at Post-Mechanical Shock
by Surface Finish
Surface Finish
HASL
Nickel/Gold
Nickel/Palladium/Gold
OSP
Immersion Silver
Immersion Tin
Anomalies
1
0
0
3
5
3
No. ofPWAs
32
28
12
36
20
36
4.1.8 High Voltage Low Current (HVLC) Circuitry Performance Results
Results of the GLM analyses for HVLC PTH and HVLC SMT circuits are given in Tables
F-6 and F-7, respectively. The GLM analyses show no practical significance relative to the
acceptance criteria, which indicates that site, flux, and surface finish parameters do not influence
the HVLC measurements.
Unlike the resistors in the HCLV circuit that were in a parallel design, the HVLC resistors
were in a series circuit design. Thus, when one resistor is missing the circuit is open.
Boxplot Displays of Multiple Comparison Results. The overall ANOVA that precedes
the use of multiple comparisons did not produce significant f-statistics for HVLC PTH (level of
significance = 0.01). However, it did produce significant f-statistics for the first three test times
for the HVLC SMT circuit, as shown in Table 4-12 (shaded entries are > 0.01).
4-20
-------
Table 4-12. P-Values for HVLC Test Results
Test Time
Pre-test
Post-85/85
Post-TS
Post-MS
P- Value for HVLC PTH
0.046
0.028
0.625
0.274
P- Value for HVLC SMT
0.000
0.000
0.000
0.742
Abbreviations and Definitions:
HCLV - high current low voltage
MS - mechanical shock
PTH - plated through hole
SMT - surface mount technology
TS - thermal shock
Figures F-l to F-8 give boxplots for the HVLC PTH and SMT circuits. It is important to
keep the vertical scale in mind relative to the acceptance criteria when viewing these boxplots.
That is, the acceptance criteria indicates that the current should be between 4|iA and 6|iA. These
boxplots are centered close to 5|iA, and the total spread is on the order of 0.02|iA for the PTH
circuits and approximately O.SjiA for SMT circuits. Hence, even though there are some
statistically significantly differences, they are not likely to be of practical concern. Note the
boxplots in Figure F-8 for HCLV SMT at post-MS. These values are all either OjiA, or very
close to it, reflecting the fact that the resistors came off the PWA during the mechanical shock
test. This loss of components occurred on every PWA and was not related to the site, surface
finish, or flux.
Comparison to Acceptance Criterion. The acceptance criteria for HVLC PTH and
HVLC SMT are listed in Table 4-1 (responses 3 and 4). All HVLC PTH circuits met the
acceptance criteria of 4|iA and 6|iA for the entire sequence of tests. Only one HVLC SMT
current measurement failed to meet the acceptance criterion after exposure to 85/85 (see Table F-
1). In turn, this same PWA also was the only one that did not meet the acceptance criteria after
thermal shock. The test technician noted that this PWA exhibited a burnt edge after 85/85.
However, after the mechanical shock test all HVLC SMT circuits failed to meet the acceptance
criteria due to resistors coming off the PWA resulting in an open circuit.
4.1.9 High Speed Digital (HSD) Circuitry Performance Results
The pre-test measurements for HSD PTH and HSD SMT circuits were subjected to GLM
analyses, as were the deltas after 85/85, thermal shock, and mechanical shock. The complete
results of the GLM analyses are given in Tables F-8 and F-9, respectively. The GLM analyses
indicate that the experimental parameters under evaluation do not influence the HSD total
propagation delay measurements.
Boxplot Displays of Multiple Comparison Results. The overall ANOVA that precedes
the use of multiple comparisons did not produce significant f-statistics for either HSD PTH or
HSD SMT circuitry. The p-values for the respective f-statistics are given in Table 4-13 (shaded
entries are > 0.01).
4-21
-------
Table 4-13. P-Values for HSD Test Results
Test Time
Pre-test
Post-85/85
Post-TS
Post-MS
P- Value for HSD PTH
0.442
0.443
0.491
0.487
P- Value for HSD SMT
0.585
0.359
0.954
0.760
Abbreviations and Definitions:
HSD - high speed digital
MS - mechanical shock
PTH - plated through hole
SMT - surface mount technology
TS - thermal shock
Figures F-9 and F-10 give boxplots of pre-test measurements of total propagation delay
for the HSD PTH and HSD SMT circuits, respectively. Note that most total propagation delays
in Figure F-9 for HSD PTH are a little over 17 nanoseconds (ns) with a range of about Ins.
Figure F-10 shows that the total propagation delays for HSD SMT have a range of about 0.4ns
and are centered about 9.2ns. The percentage changes in the total propagation delay
measurements were small and well within the acceptance criteria so boxplot displays of these
measurements are not presented.
Comparison to Acceptance Criterion. The acceptance criteria for HSD PTH and HSD
SMT are listed in Table 4-1 (responses 5 and 6). One HSD SMT did not give a response after
exposure to 85/85 (see Table F-l). This same circuit also failed to give a response after thermal
shock, as did one additional HSD SMT circuit and two HSD PTH circuits. At post-MS, two
HSD PTH circuits and one HSD SMT circuit did not give a response. The testing technician
indicated that the HSD device had failed. Previous testing with the test PWA in other programs
has indicated a failure of the HSD components, which is independent of the experimental
parameters under evaluation in the DfE program. All other HSD circuits were well within the
acceptance criterion.
4.1.10 High Frequency Low Pass Filter (HF LPF) Circuitry Performance Results
Pre-test measurements for all HF LPF circuits were subjected to GLM analyses, as were
the deltas after 85/85, thermal shock, and mechanical shock. The results of the GLM analyses are
given in Tables F-10 to F-15. The GLM analyses indicate that the parameters under evaluation
(site, surface finish, or flux) do not influence the HF LPF measurements. The same is true at
post-85/85, post-TS and post-MS. However, the test measurements contained many extreme
outlying observations at both of these later two test times, which greatly increases the sample
variance and in turn hinders the interpretation of the GLM results. As indicated in Tables F-l, F-
2, and F-3 there were many anomalous HF LPF test measurements (171 at post-
MS). The principal source of these outliers was open PTHs, is discussed in more detail under
Comparison to Acceptance Criteria.
4-22
-------
Boxplot Displays of Multiple Comparison Results
The ANOVA that precedes the use of multiple comparisons produced significant
f-statistics for HF LPF PTH 50MHz at all test times and for three other HF LPF circuits at post-
TS. The p-values for the respective f-statistics are given in Table 4-14 (all p-values > 0.01 are
shaded).
Table 4-14. P-Values for HF LPF Test Results
Test Time
Pre-test
Post-85/85
Post-TS
Post-MS
P- Value for
HFPTH
50MHz
0.002
0.000
0.004
0.002
P- Value for
HFPTH
f(-3dB)
0.052
0.484
0.578
0.001
P- Value for
HFPTH
f(-40dB)
0.024
0.487
0.594
0.028
P- Value for
HFSMT
50MHz
0.241
0.227
0.016
0.000
P- Value for
HFSMT
f(-3dB)
0.092
0.258
0.074
0.112
P- Value for
HFSMT
f (-40dB)
0.057
0.970
0.023
0.000
Abbreviations and Definitions:
HF - high frequency
LPF - low pass filter
MS - mechanical shock
PTH - plated through hole
SMT - surface mount technology
TS - thermal shock
These results are discussed separately for each of the six HF LPF circuits. Boxplot
displays of all test results for HF LPF circuits have been created to aid in the interpretation. Only
the boxplots showing statistical and practical significance are shown here (Figures 4-9 to 4-15);
the rest are in Appendix F.
HF LPF PTH 50MHz. While the p-values for the associated f-statistic were highly
significant at all test times, Figure 4-9 identifies the source of this significance at pre-test, where
the responses for nickel/gold applied at Site 18 and subsequently processed with low residue flux
are much lower than the others. Post-85/85 and post-TS results indicate just the opposite for this
demonstration site (see Figures F-l 1 and F-12). What occurred is that the problem circuit
returned to normal at post-85/85 and post-TS, but those measurements were then compared to
their low pre-test measurements, which caused the differences to be large in the positive direction.
Hence, the large values at post-85/85 and post-TS are an artifact of the pre-test measurements
and should most likely be ignored as the circuit performance was in line with all others. More
importantly, the significant differences at pre-test are too small to be of practical concern. The
range depicted in Figure 4-9 is approximately 0.7dB and the acceptance criterion allows a change
of+5dB. On the other hand, Figure 4-10 is of concern as several of the surface finishes have
measurements well below the lower bound acceptance criterion of -5dB. In particular, one of the
five OSP PWAs, two of the three immersion silver PWAs, and one of the five immersion tin
PWAs. This circuit had 15 anomalies at post-MS.
HF LPF PTH f(-3dB). Figure 4-11 shows the boxplot for the HF LPF PTH f(-3dB)
circuit at the post-MS test time. Boxplots for the other three test times can be found in Appendix
4-23
-------
F, since the p-values for the associated F-statistic were not significant, except for post-MS.
Figure 4-11 shows notable variation in the magnitude of the differences — note the vertical scale.
Several cases are well outside the acceptance criterion. In particular, one of four nickel/gold
PWAs, two of the three immersion silver PWAs, and one of five immersion tin PWAs are quite
low. This circuit had 18 anomalies at post-MS.
HF LPF PTH f(-40dB). Figure 4-12 shows the boxplot for the HF LPF PTH f(-40dB)
circuit at the post-MS test time. Boxplots for the other three test times can be found in Appendix
F. While the p-values for the associated f-statistic were not significant at any of the test times,
Figure 4-12 shows notable variation in the magnitude of the differences (note the vertical scale).
Several cases are well outside the acceptance criterion of +50MHz. In particular, two of the three
immersion silver PWAs and one of five immersion tin PWAs are quite low. This circuit had 14
anomalies at post-MS.
HF LPF SMT 50MHz. Figure 4-13 shows the boxplot for the HF LPF SMT 50MHz
circuit at the post-MS test time. Boxplots for the other three test times can be found in Appendix
F, since the p-values for the associated f-statistic were not significant, except for post-MS. The
magnitude of the changes at post-85/85 and post-TS are too small to be of practical concern
relative to the acceptance criteria of +5dB. On the other hand, the post-MS results are of serious
concern, as nine of 23 cases are well below the lower acceptance bound of-5dB. It is noteworthy
that neither nickel/gold or nickel/palladium/gold had any anomalies. This circuit had 30 anomalies
at post-MS.
HF LPF SMT f(-3dB). Figure 4-14 shows the boxplot for the HF LPF SMT f(-3dB)
circuit at the post-MS test time. Boxplots for the other three test times can be found in Appendix
F. While the p-values for the associated F-statistic were not significant at any of the test times,
Figure 4-14 shows notable variation in the magnitude of the differences (note the vertical scale).
Several cases are well outside the acceptance criterion of+50MHz. It is noteworthy that neither
nickel/gold or nickel/palladium/gold had any anomalies. This circuit had 29 anomalies at post-
MS.
HF LPF SMT f(-40dB). Figure 4-15 shows the boxplot for the HF LPF SMT f(-40dB)
circuit at the post-MS test time. Boxplots for the other three test times can be found in Appendix
F, since the p-values for the associated f-statistic were not significant, except at post-MS. This
circuit had the most anomalies (65) at post-MS. Some of the anomalies may be due to the high
variability in the frequency when measured at -40dB. Figure 4-15 shows notable variation in the
magnitude of the differences (note the vertical scale). Most cases are well outside the acceptance
criterion of+50MHz. Nickel/gold and nickel/palladium gold are again noteworthy as they have
very few anomalies.
Comparison to Acceptance Criteria
The acceptance criteria for the six HF LPF circuits are shown in Table 4-1 (responses 7
through 12). Thirteen of 984 HF LPF test measurements did not meet the acceptance criterion
after exposure to 85/85 (see Table F-l). These 13 responses occurred on six PWAs, with 12 of
the 13 occurring with PTH components. After exposure to thermal shock, the number of HF LPF
anomalies increased to 104 (see Table F-2). Thirteen of these 103 HF LPF anomalies carried over
4-24
-------
from the 85/85 test. At post-MS, the number of anomalies increased to 171 with 97 carrying over
from thermal shock.
PWAs with HF LPF anomalies generally have multiple anomalies. This can be seen in
Table 4-15, which shows the frequency distribution of the number of HF LPF anomalies per PWA
at post-MS (see Tables F-l to F-3).
Table 4-15. Frequency Distribution of HF LPF Anomalies
at Post-Mechanical Shock per PWA
No. of HF LPF Anomalies per PWA
at Post-Mechanical Shock
None
1
2
3
4
5
6
Frequency
90
36
5
20
4
5
4
The test technician comments indicate that most of the HF LPF anomalies were due to an
open PTH, which affects both PTH and SMT. To explain further, a circuit board consists of
alternating layers of epoxy and copper through which a hole is drilled during fabrication. This via
is plated with a very thin layer of electroless copper to provide a "seed bed" for the primary
coatings. Copper is then electroplated over the electroless copper strike. The final surface finish
(HASL, OSP, etc.) is then applied. Failure to make an electrical connection between the copper
etches on the opposite sides of the board is known as an open PTH. The opens occurred in very
small vias in the HF LPF circuit. Small vias can be very difficult to plate. Opens were present
during in-circuit testing and at pre-test. In some cases, a z-wire was inserted through the via to
make an electrical connection between the etches on the opposite side of the board. It appears
that test conditions may accelerate the problem.
Although an open PTH is a fabrication issue, there does appear to be a relationship with
surface finish. The HF LPF anomalies are summarized by surface finish in Table 4-17 for each of
the six HF LPF circuits. Under the assumption that the anomalies occur independent of surface
finish, the expected number of anomalies can be calculated for each cell. For example, consider
Table 4-16, which summarizes the observed and expected anomalies for the HF LPF PTH 50MHz
circuit.
4-25
-------
Table 4-16. Comparison of the Observed and Expected Number of Anomalies for
the HF LPF PTH 50MHz Circuit by Surface Finish
Surface Finish
HASL
Nickel/Gold
Nickel/Palladium/Gold
OSP
Immersion Silver
Immersion Tin
Column Total
Observed Anomalies
(Expected)
1 (2.9)
2 (2.6)
0(1.1)
2(3.3)
6(1.8)
4(3.3)
15
Observed Non-Anomalies
(Expected)
31 (29.1)
26 (25.4)
12(10.9)
34 (32.7)
14(18.2)
32(32.7)
149
Row Total
32
28
12
36
20
36
164
Abbreviations and Definitions:
HF - high frequency
LPF - low pass filter
PTH - plated through hole
Under the hypothesis of independence of row and column classifications, the expected
number of observations in each cell is the product of the cell's row and column totals divided by
the grand total. For example, the expected number of anomalies for HASL is computed as
(32)(15)/164 = 2.9. The expected values for all cells are shown in parentheses in the example.
A chi-square statistic is calculated on the differences of the observed and expected number in each
cell (Iman, 1994). The chi-square distribution is used to approximate the p-value for the chi-
square statistic. For the above example, the p-value is 0.016, which is not significant at the 0.01
level. With this level of significance, the hypothesis of independence is not rejected for the FTP
LPF PTH 50MHz circuit. That is, there are no significant differences in the number of anomalies
among the surface finishes for the HF LPF PTH 50MHz circuit.
Table 4-17. Comparison of the Observed and Expected Number of Anomalies
Under the Hypothesis of Independence of Surface Finishes
No. of
PWAs
HASL
Nickel/Gold
Nickel/Palladium/Gold
OSP
Immersion Silver
Immersion Tin
Totals
32
28
12
36
20
36
164
p-value
50MHz
1 (2.9)
2 (2.6)
0(1.1)
2(3.3)
6(1.8)
4(3.3)
15
0.016
HFLPF
PTH
f(-3dB)
2(4.1)
3 (3.6)
0(1.5)
2 (4.6)
6 (2.6)
5 (4.6)
18
0.051
f(-40dB)
1 (2.9)
2 (2.6)
0(1.1)
1 (3.3)
7(1.8)
3 (3.3)
14
0.001
50MHz
6(5.9)
0(5.1)
0 (2.2)
6 (6.6)
7(3.7)
11 (6.6)
30
0.006
HFLPF
SMT
f(-3dB)
7(5.9)
0(5.1)
0 (2.2)
5 (6.6)
6(3.7)
11 (6.6)
29
0.008
f(-40dB)
15(13.1)
1(11.4)
1 (4.9)
20(14.7)
11 (8.2
17(14.7)
65
0.000
Abbreviations and Definitions:
HF - high frequency
LPF - low pass filter
PTH - plated through hole
SMT - surface mount technology
4-26
-------
Such is not the case for the last four HF LPF circuits listed in Table 4-17, where the
p-values at the bottom of the table indicate that the anomalies are not independent of surface
finish. The expected values for anomalies appear in parenthesis in each cell in that table. These
comparisons show:
• HASL anomalies are close to the expected values throughout.
• Nickel/gold has far fewer anomalies than expected.
• Nickel/palladium/gold has far fewer anomalies than expected.
• OSP anomalies are close to expected, except for the last column, where they have more
anomalies than expected.
• Immersion silver has many more anomalies that expected for all circuits.
• Immersion tin anomalies are close to expected for PTH circuits, but are higher than
expected for SMT circuits.
The number of open PTH anomalies may be related to the inherent strength of the metals.
Tin and silver are relatively weak; OSP has no metal, while nickel makes the PTH stronger. To
determine the relevancy of metal strength to the open PTH anomalies, the HF LPF circuits would
need to be subjected to failure analysis to check for copper plating thickness and PTH voids in the
vias, as both of these may be problems in small vias. In addition, the chemical removal of copper
from the via may be much greater in immersion tin and immersion silver, depending on how they
were processed.
4.1.11 High Frequency Transmission Line Coupler (HF TLC) Circuitry Performance
Results
Pre-test measurements for all HF TLC circuits except Reverse Null Frequency were
subjected to GLM analyses, as were the deltas after 85/85, thermal shock, and mechanical shock.
The results of the GLM analyses are given in Tables F-16 to F-20. The GLM analyses indicate
that the experimental parameters do not influence the pre-test HF TLC measurements, except for
those at 50 MHZ. The results for the 50MHz case are examined in further detail.
The predicted response at pre-test for HF TLC 50MHz for the base case (HASL at Site 1
processed with low residue flux) based on the Site & Flux GLM was -47.43dB. The predicted
differences from the base case are given in Appendix F in Table F-21. The results show that the
demonstration sites that produced nickel/gold and nickel/palladium/gold (# 13 - 16) have
predicted increases of less than 3dB. While statistically significant, this change is rather small
compared to the base case value and is probably not of practical utility. Overall, some of the
demonstration sites differ from the base case by approximately -1.5dB to 2.9dB. These changes
again may not have any practical significance, since the important concept is not so much the
magnitude of the response, but rather its stability when subject to environmental stress conditions,
which is the basis for the acceptance criteria.
The predicted response at pre-test for HF TLC 50MHz for the base case (HASL
processed with low residue flux) based on the Surface Finish & Flux GLM was -46.73dB, which
is almost identical to that for the Site & Flux GLM. The predicted differences from the base case
are given in Appendix F in Table F-22. These predictions are consistent with those in
4-27
-------
Table F-21, and show that immersion tin and immersion silver are approximately l.OdB lower
than the base case, and nickel/gold and nickel/palladium/gold are approximately 1 to 2dB higher
than the base case. Again, these differences are most likely not of practical utility.
Boxplot Displays of Multiple Comparison Results. The ANOVA that precedes the use
of multiple comparisons produced a significant f-statistic for only the HF TLC 50MHz circuit at
pre-test. The p-values for the respective f-statistics are given in Table 4-18 (all p-values > 0.01
are shaded).
Table 4-18. P-Values for HF TLC Test Results
Test Time
Pre-test
Post-85/85
Post-TS
Post-MS
P- Value for
HFTLC
50MHz
0.000
0.285
0.344
0.313
P- Value for
HFTLC
500MHz
0.070
0.111
0.560
0.390
P- Value for
HFTLC
IGHz
0.250
0.299
0.650
0.568
P- Value for
HFTLC
RNR
0.418
0.201
0.770
0.359
Abbreviations and Definitions:
HF TLC - high frequency transmission line coupler
MS - mechanical shock
TS - thermal shock
Boxplot displays of the test results for HF TLC 50MHz are given in Appendix F. While
the F-statistic is significant at pre-test, the post-85/85 results show that the changes from the base
case are centered about OdB and well within the acceptance criteria of+5dB. Thus, while the
magnitude of the individual responses at pre-test may or may not be of practical concern in a
particular application, the acceptance criteria is focused on the stability of the response when the
circuit is subsequently subjected to environmental stress. The post-85/85 and post-TS results
confirm that changes in the responses are all acceptable. However, post-MS shows several
anomalies (seven by count), as shown in Figure 4-16. Five of these seven anomalies were for
immersion silver, while HASL and immersion tin each had one anomaly.
Figure 4-17 displays the boxplot of the test results for HF TLC 500MHz post-MS. The
HF TLC 500MHz results for the other test times are quite similar to those for HF TLC 50MHz,
and boxplots of these results can also be found in Appendix F. Post-MS results for HF TLC
500MHz had only one slight anomaly compared to seven for HF TLC 50MHz. This anomaly was
only -5.22dB, compared to the lower bound of -5dB, so it is of no concern. Boxplots displays for
HF TLC IGHz are not given to conserve space. The total variation at pre-test for HF TLC IGHz
was only 2dB, and there was only one slight anomaly of-5dB at post-MS, which is not of
concern.
Figure 4-18 displays the boxplot of the test results for HF TLC RNR post-MS. None of
the F-statistics were significant for testing equality of means; boxplots of results from the other
three test times can be found in Appendix F. The reader should keep in mind that the decreases in
the HF TLC RNR response in Figure 4-18 are favorable outcomes. The acceptance criterion only
specifies an upper bound of either 5dBb or lOdB for the increase, depending on the magnitude of
4-28
-------
the pre-test values. There were five slight anomalies at post-MS, with immersion tin having three,
while HASL and immersion silver each had one.
Comparison to Acceptance Criteria. The acceptance criteria for HF TLC circuitry are
listed in Table 4-1 (responses 13 to 17). Only one HF TLC RNR measurement failed to meet the
acceptance criterion after exposure to 85/85 (see Table F-l). This measurement showed an
increase of 10.2dB, which is only slightly above the acceptance criteria of lOdB and not of
practical interest. At post-TS, this value was 10.02. One other FTP TLC RNR measurement had
an increase of 7.93dB at post-TS. All other changes were less than 5dB. One FTP TLC IGHz
measurement was just below the lower limit of -5dB at -5.65dB. There were five anomalies at
post-MS, none of which were of practical interest.
4.1.12 Leakage Measurements Performance Results
Four features were included in the design of the test PWA to check for current leakage:
10-mil pads, PGA socket (PGA-A, PGA-B), and a gull wing component (responses 18 to 21 in
Table 4-1). The PGA hole pattern has four concentric squares that are electrically connected by
traces on the top layer of the board. Two leakage current measurements were made: 1) between
the two inner squares (PGA-A; and 2) between the two outer squares (PGA-B). Solder mask
covers the pattern of the PGA-B, allowing a direct comparison of similar patterns with and
without solder mask. Rather than an actual PGA device, a socket was used, because it provides
the same soldering connections as a PGA device.
The leakage measurements were subjected to GLM analyses at pre-test and after 85/85,
thermal shock, and mechanical shock. The results of the GLM analyses are given in Appendix F
(Tables F-23 to F-26).
10-Mil Pads
Tables F-27 and F-28 give the predicted changes from their respective base cases for all
leakage measurements at pre-test for the GLMs. Examination of the GLM results for 10-mil pad
shows evidence of site-to-site variation and some interaction between site and flux that affects
resistance either positively or negatively by up to an order of magnitude. Demonstration sites
applying the OSP surface finish (Sites 6, 7, 8, and 9), as well as Sites 10 and 11 with immersion
tin, do not differ from the base case when low residue flux is used. When sites are dropped from
the GLM and replaced by surface finishes, the results show slight increases in resistance over the
base case for OSP, immersion tin, and immersion silver.
The differences from the base case for both GLMs essentially disappear after exposure to
the 85/85 test environment. This result is not unusual and may be due to a cleansing effect from
the 85/85 test environment that removes residues resulting from board fabrication, assembly, and
handling. This same phenomenon was observed for the other three leakage circuits.
Boxplot Displays of Multiple Comparison Results. As with the other circuits, an
ANOVA was performed to determine if there was a significant difference in the mean leakage
measurements for each of the four leakage circuits. The p-values for the respective f-statistics for
4-29
-------
all four leakage measurements are given in the following summary (all p-values > 0.01 are
shaded). Table 4-19 shows significant differences in the means at pre-test and post-85/85 for the
10-mil pads.
Table 4-19. P-Values for Leakage Test Results
Test Time
Pre-test
Post-85/85
Post-TS
Post-MS
P- Value for
10-mil Pads
0.000
0.000
0.047
0.213
P- Value for
PGA-A
0.000
0.510
0.048
0.125
P- Value for
PGA-B
0.000
0.198
0.026
0.093
P- Value for
Gull Wing
0.000
0.551
0.432
0.243
Abbreviations and Definitions:
MS - mechanical shock
PGA - pin grid array
TS - thermal shock
Boxplot displays of the leakage measurements for 10-mil pads are given in Figures 4-19
and 4-20 for pre-test and post-85/85, respectively. Boxplots for post-TS and post-MS are in
Appendix F. Figure 4-19 illustrates the impact of flux that was identified as significant in the
GLM analyses. Every case with water soluble flux is higher (better) than the corresponding low
residue analog. In Figure 4-20, the differences due to flux have disappeared. As mentioned
above, this change is likely due to a cleansing effect from the 85/85 test environment, which
removed residues resulting from board fabrication, assembly, and handling. The statistical
significance of the F-statistic at post-85/85 is attributable to immersion tin produced at Site 13,
which had lower resistance (i.e. higher current leakage). However, the resistance is still well
above the acceptance criteria of 7.7.
Comparison to Acceptance Criterion. The acceptance criterion for the leakage
measurements requires the resistance to be greater than 7.7 when expressed as log 10 ohms.
There were no anomalies for the 10-mil pads at pre-test, post-85/85, post-TS, or post-MS.
Pin Grid Array-A
Examination of the GLM results in Table F-27 for PGA-A shows evidence of site-to-site
variation and some interaction between site and flux that affects resistance either positively or
negatively by up to an order of magnitude. Nine of the demonstration sites do not differ from the
base case when low residue flux is used.
Table F-28 in Appendix F shows a flux effect of approximately 2.05 orders of magnitude
as determined using GLM analyses surface finish, indicating there are no meaningful differences
due to surface finishes. As was the case with the 10-mil pads, the differences from the base case
for both GLMs essentially disappear after exposure to the 85/85 test environment.
Boxplot Displays of Multiple Comparison Results. The p-values for the ANOVA
given above show the only test indicating a significant difference in mean leakage for the PGA-A
circuit was the pre-test (shown in Figure 4-21). Boxplot displays of the other leakage
4-30
-------
measurements for PGA-A are given in Appendix F. Figure 4-21 illustrates the impact of flux that
was identified as significant in the GLM analyses. As was true with 10-mil pads, every case with
water soluble flux is higher (better) than the corresponding low residue analog, although all test
responses were above the acceptance criteria. In subsequent results, the differences due to flux
disappear. As mentioned above, this change is likely due to a cleansing effect from the 85/85 test
environment.
Comparison to Acceptance Criterion. There were no anomalies for PGA-A at pre-test,
post-85/85, post-TS, or post-MS.
Pin Grid Array-B
Examination of the GLM results in Table F-27 for PGA-B shows a strong effect due to
flux of approximately 2.77 orders of magnitude. Thirteen of the demonstration sites do not differ
from the base case when low residue flux is used, and the other two only differ slightly. Table
F-28 also shows a strong flux effect of approximately 2.71 orders of magnitude as determined in
the GLM analyses surface finish, indicating there are no meaningful differences due to surface
finishes. As was the case with the 10-mil pads and PGA-A, the differences from the base case for
both GLMs essentially disappear after exposure to the 85/85 test environment.
Boxplot Displays of Multiple Comparison Results. The p-values for the ANOVA
given above show the only test indicates a significant difference in mean leakage for the PGA-B
circuit at pre-test. Figure 4-22 illustrates the impact of flux that was identified as significant in the
GLM analyses. As was true with 10-mil pads and PGA-A, every case with water soluble flux is
higher (better) than the corresponding low residue analog, though all test responses were above
the acceptance criteria. In boxplots for the other test times, the differences due to flux disappear.
As mentioned above, this change is likely due to a cleansing effect from the 85/85 test
environment. Boxplot displays of the other leakage measurements for PGA-B are given in
Appendix F.
Comparison to Acceptance Criterion. There were no anomalies for PGA-B at pre-test,
post-85/85, post-TS, or at post-MS.
Gull Wing
Examination of the GLM results in Table F-27 for the Gull Wing shows a moderate effect
due to flux of approximately 0.81 orders of magnitude. There is evidence of modest site-to-site
variation and some interaction between site and flux. Eleven of the demonstration sites do not
differ from the base case when low residue flux is used, and the other two only differ slightly.
Table F-28 shows a flux effect of approximately 1.09 orders of magnitude as determined in the
GLM analyses by surface finish, indicating there are no meaningful differences due to surface
finishes. As was the case with the 10-mil pads, PGA-A, and PGA-B, the differences from the
base case for both GLMs essentially disappear after exposure to the 85/85 test environment.
4-31
-------
Boxplot Displays of Multiple Comparison Results. The p-values for the ANOVA
given above show only the test time, with a significant difference in mean leakage for the gull
wing circuit at pre-test, as illustrated in Figure 4-23. Boxplot displays of the leakage
measurements for the gull wing at the other test times are given in Appendix F. Figure 4-23
illustrates the impact of flux that was identified as significant in the GLM analyses. As was true
with 10-mil pads, PGA-A, and PGA-B, every case with water soluble flux is higher that the
corresponding low residue analog, though all test responses were abofe the acceptance criteria.
At subsequent test times, the differences due to flux disappear. As mentioned above, this change
is likely due to a cleansing effect from the 85/85 test environment.
Comparison to Acceptance Criterion. There was one slight anomaly for the Gull Wing
following 85/85. This value was 7.27 compared to the acceptance criteria of 7'.7, so it is not of
concern. There were no anomalies at post-TS or post-MS.
4.1.13 Stranded Wires
Two stranded wires were hand-soldered on the PWA (responses 22 and 23 in Table 4-1).
One wire was soldered into PTHs, and the other was soldered to two terminals. Pre-test
measurements for the stranded wire circuits were subjected to GLM analyses, as were the deltas
after 85/85, thermal shock, and mechanical shock. The results of the GLM analyses are given in
Tables F-29 and F-30. The GLM analyses indicate that the experimental parameters do not
influence the stranded wire voltage measurements.
Boxplot Displays of Multiple Comparison Results. As with other circuits, an ANOVA
was used to determine if there was a significant difference in the mean leakage measurements for
the two stranded wire measurements. The p-values for the respective f-statistics for these two
sets of voltage measurements are given in the following summary (all p-values > 0.01 are shaded).
Table 4-20 shows no significant differences in the means at any test time. Boxplot displays of the
pre-test voltage measurements (mV) can be found in Appendix F.
Table 4-20. P-Values for Stranded Wire Test Results
Test Time
Pre-test
Post-85/85
Post-TS
Post-MS
P- Value for 10-mil Pads
0.951
0.410
0.537
0.396
P-Value for PGA-A
0.203
0.407
0.440
0.408
Abbreviations and Definitions:
MS - mechanical shock
PGA - pin grid array
TS - thermal shock
Comparison to Acceptance Criterion. The acceptance criteria requires changes in
voltage to be within 0.356V of their pre-test measurements. There were no anomalies for either
Stranded Wire 1 or 2 at pre-test or following 85/85. There was one minor anomaly at post-TS
for SW2 where the measured increase in voltage was 0.371V, compared to the upper acceptable
limit of 0.356V. At post-MS, there was one minor anomaly for SW1 (0.375) and four minor
4-32
-------
anomalies for SW2 (0.359, 0.37O, 0.365, and 0.357). All of these anomalies were right at the
upper limit and are not of concern.
4.1.14 Failure Analysis
Following the analysis of the test boards, ion chromatography was used as a tool to
analyze boards that failed 85°C/85% relative humidity exposure. Contamination Studies
Laboratories, Inc. (CSL) in Kokomo, Indiana, conducted this failure analysis. The purpose of the
analysis was to determine if any links exist between board contamination from fabrication and
assembly process residues and the electrical anomalies.
Test Sample Identification
Twenty boards were selected for the ion chromatography analysis including: 1) a test
group of boards that failed after exposure to 85°C/85% relative humidity; and 2) a control group
of boards that were not subjected to the 85°C/85% relative humidity environment. The test group
consisted of 10 boards (identified in Table F-l) that exhibited various anomalies following
85°C/85% relative humidity testing. For the control group, the 10 boards selected represented
each of the six surface finishes and a variety of assembly processes and sites. Table 4-21
summarizes the 20 boards selected for ion chromatography analysis.
Visual Observations
The test group of boards was visually inspected to identify any obvious anomalies or
defects. All 10 boards exhibited visual anomalies in varying degrees. The most common
anomalies were solder cracking and discoloration of the surface metalization. Less common were
pinholes and foreign material (e.g., solder balls). The following photographs show examples of
the more prominent visual defects.
4-33
-------
Table 4-21. Identification of Assemblies Selected for Ion Chromatography Analysis
Finish
Board #
Assembly Process
Site
Untested Board (Control Group)
HASL
HASL
Nickel/Gold
Nickel/Gold
Nickel/Palladium/Gold
OSP
Immersion Silver
Immersion Silver
Immersion Tin
Immersion Tin
077-4
096-2
068-4
017-4
001-4
061-2
085-4
074-3
103-4
034-4
LR
WS
WS
LR
LR
WS
WS
LR
WS
LR
1
2
7
12
15
3
8
9
4
10
Post-85/85 Exposure (Anomaly Group)
HASL
Nickel/Gold
Nickel/Gold
OSP
Immersion Silver
Immersion Silver
Immersion Tin
Immersion Tin
Immersion Tin
Immersion Tin
083-2
013-1
015-4
056-4
082-2
094-4
030-4
032-4
086-2
102-4
WS
LR
LR
LR
LR
WS
WS
LR
WS
WS
1
13
14
5
11
12
9
8
7
10
Abbreviations and Definitions:
LR - low residue flux
WS - water soluble flux
Test Method
The fundamental steps for conducting ion chromatography analysis per IPC-TM-650,
method 2.3.28 are as follows:
1. The lab technician (LT) placed the test board(s) into clean KAPAK™ (heat-sealable
polyester film) bag(s).
2. The LT introduced a mixture of isopropanol (75 percent volume) and deionized water (25
percent volume) into the bag(s), immersing the test board(s). NOTE: The heat-sealed
bag(s) included an opening for ventilation.
3. The LT inserted the bag(s) into an 80 °C water bath for one hour.
4. The LT removed the bag(s) from the water bath.
5. The LT separated the test board(s) from the bags.
6. The LT placed the test board(s) on a clean holding rack for air drying at room
temperature.
4-34
-------
7. The LT performed controls and blanks on the Dionex ion chromatography system before
the test began. NOTE: NIST-traceable standards for system calibration were used.
8. The LT injected a 1.5 ml sample of each test sample's extract solution using a 5 mM
sodium bicarbonate eluent.
Failure Analysis Results
The following tables show the ion chromatography data for each surface finish analyzed,
reported as micrograms of the residue species per square inch of extracted surface (jig/in2).
NOTE: This measure should not be confused with micrograms of sodium chloride equivalent per
square inch, which is the common measure for most ionic cleanliness test instruments.
Table 4-22. Ion Chromatography Anion°Data (HASL)
Sample
Description
Assembly
Process
Site
Ion Chromatography Data
Cl
Br
WOA
Untested Boards (Control Group)
Board #077-4
Board #096-2
LR
WS
1
2
5.87
14.53
3.82
10.01
154.33
3.01
Tested Boards (Anomaly Group)
Board #083-2
WS
1
5.36
2.73
7.15
Test results reported as micrograms of the residue species per square inch of extracted surface (ug/in2).
Br" = bromide ion; Cl" = chloride ion; LR = low residue flux; WOA = weak organic acids; WS = water soluble
flux.
Table 4-23. Ion Chromatography Anion ° Data (Immersion Tin) a
Sample
Description
Assembly
Process
Site
Ion Chromatography Data
Cl
Br
WOA
Untested Boards (Control Group)
Board #034-4
Board #103 -4
LR
WS
10
4
0.87
5.10
5.26
2.98
140.45
3.30
Tested Boards (Anomaly Group)
Board #032-4
Board # 030-4
Board #086-2
Board #102-4
LR
WS
WS
WS
8
9
7
10
1.75
1.70
2.99
2.33
4.12
5.68
3.30
3.16
15.78
15.46
9.23
4.63
Test results reported as micrograms of the residue species per square inch of extracted surface (ug/in2).
Abbreviations and Definitions:
Br" - bromide ion
Cl" - chloride ion
LR - low residue flux
WOA - weak organic acids
WS - water soluble flux
4-35
-------
Table 4-24. Ion Chromatography Anion ° Data (Immersion Silver) a
Sample
Description
Assembly
Process
Site
Ion Chromatography Data
Cl
Br
WOA
Untested Boards (Control Group)
Board #074-3
Board #085-4
LR
WS
9
8
0.60
4.77
6.53
2.64
159.48
5.22
Tested Boards (Anomaly Group)
Board #082-2
Board #094-4
LR
WS
11
12
2.59
2.53
3.25
4.65
4.28
5.78
Test results reported as micrograms of the residue species per square inch of extracted surface (ug/in2).
Abbreviations and Definitions:
Br" - bromide ion
Cl" - chloride ion
LR - low residue flux
WOA - weak organic acids
WS - water soluble flux
Table 4-25. Ion Chromatography Anion ° Data (Nickel/Gold)
Sample
Description
Assembly
Process
Site
Ion Chromatography Data
Cl
Br-
WOA
Untested Boards (Control Group)
Board #01 7-4
Board #068-4
LR
WS
12
7
1.01
4.57
5.34
1.78
150.81
3.08
Tested Boards (Anomaly Group)
Board #01 3-1
Board #01 5 -4
LR
LR
13
14
2.44
1.63
3.56
2.80
15.13
14.04
a Test results reported as micrograms of the residue species per square inch of extracted surface (ug/in2).
Abbreviations and Definitions:
Br" -bromide ion
Cl" - chloride ion
LR - low residue flux
WOA - weak organic acids
WS - water soluble flux
4-36
-------
Table 4-26. Ion Chromatography Anion°Data (OSP)
Sample
Description
Assembly
Process
Site
Ion Chromatography Data
Cl
Br | WOA
Untested Boards (Control Group)
Board #06 1-2
WS
3
3.57
3.45 | 2.57
Tested Boards (Anomaly Group)
Board #056-4
LR
5
2.40
4.28 | 26.41
Test results reported as micrograms of the residue species per square inch of extracted surface (ug/in2).
Abbreviations and Definitions:
Br" -bromide ion
Cl" - chloride ion
LR - low residue flux
WOA - weak organic acids
WS - water soluble flux
Table 4-27. Ion Chromatography Anion ° Data (Nickel/Palladium/Gold) a
Sample
Description
Assembly
Process
Site
Ion Chromatography Data
Cl
Br | WOA
Untested Boards (Control Group)
Board #00 1-4
LR
15
0.84
5.15 | 151.18
Test results reported as micrograms of the residue species per square inch of extracted surface (ug/in2).
Abbreviations and Definitions:
Br" - bromide ion
Cl" - chloride ion
LR - low residue flux
WOA - weak organic acids
Chloride. Chloride ion (Cl") is one of the more detrimental materials found on printed
circuit assemblies. Chloride, which can come from a variety of sources, is most often attributable
to flux residues. Chloride will generally initiate and propagate electrochemical failure
mechanisms, such as metal migration and electrolytic corrosion, when combined with water vapor
and an electrical potential. The tolerance for chloride on an assembly depends on the flux
chemistry that an assembler uses. An assembly processed with high-solids rosin fluxes (RA or
RMA) can tolerate higher levels of chloride due to the encapsulating nature of the rosin, water
soluble fluxes and no-clean fluxes, which flux manufacturers typically formulate using resins or
very low levels of rosin, do not have this encapsulating protection. Therefore, they require lower
levels of flux on final assemblies.
CSL recommends a maximum chloride level of no more than 4.5 to 5.0 |ig/in2 for finished
assemblies processed with water-soluble fluxes, and no more than 2.5 |ig/in2 for finished
assemblies processed with low solids (no-clean) fluxes. Although these recommended maximums
do not presently appear in any nationally-accepted specifications or standards, years of failure
analysis experience dealing with CSL's numerous customers serves as a basis or starting point.
4-37
-------
With the exception of the HASL boards, all untested and tested assemblies exhibit levels at
or below CSL's recommended guidelines. Therefore, the observed chloride levels are not
considered to be detrimental from an electrochemical standpoint.
The two untested (control) boards with the HASL finish exhibit levels significantly above
CSL's recommended limits and are therefore at risk for electrochemical failures. For the board
processed with low residue (no clean) flux, CSL suspects that the high chloride is due mainly to
the board fabricator's use of a chloride-activated HASL flux coupled with an ineffective
post-HASL cleaning process. For the board processed with water-soluble flux, high chloride may
be the result of both HASL residues and water soluble flux residues. In both cases, ineffective
cleaning is the likely culprit.
The one tested HASL board with the reported anomaly exhibits a level only slightly above
CSL's recommended limit. Although the chloride in the observed amount places the assembly at
slight risk for electrochemical failures, CSL does not believe in this case that chloride
contamination is the root cause for reported open PTH failures on Board #083-2.
Based on the fact the tested boards with known anomalies exhibit levels near or below
CSL's recommended guidelines, there is reasonable confidence that the anomalies identified in the
performance testing are not the result of chloride residues. The majority of the anomalies are
either mechanical in nature (e.g., poor solder joint integrity) or component non-conformities (e.g.,
wrong value and device failures).
Bromide. Bromide ion (Br") is generally attributable to the bromide fire retardant added
to epoxy-glass laminates to give fire resistance, and which is subsequently extracted in the ion
chromatography analytical procedure. Bromide can also sometimes come from solder masks,
marking inks, or fluxes that have a bromide activator material. Bromide, when from the fire
retardant, is not a material that typically degrades the long-term reliability of electronic
assemblies. If bromide comes from a flux residue, it can be corrosive, as other halides can be.
The level of bromide varies depending on the porosity of the laminate and/or mask, the degree of
over/under cure of the laminate or mask, or the number of exposures to reflow temperatures.
For epoxy-glass laminate, bromide levels typically fall within the range of 0 to 7 |ig/in2,
depending upon the amount of fire retardant the laminate manufacturer has added. Exposure to
reflow conditions tends to increase the porosity of the laminate and mask. With several exposures
to reflow conditions, bromide can reach levels as high as 10 to 12 |ig/in2. The testing laboratory,
CSL, does not presently consider bromide levels under 12 |ig/in2 to be detrimental on organic
PWBs. However, CSL considers levels between 12 |ig/in2 to 20 |ig/in2to be a borderline risk for
failures if attributable to corrosive flux residues. Furthermore, levels above 20 |ig/in2 are
considered to be a significant threat for failures if attributable to corrosive flux residues.
Based on CSL's guidelines, the bromide levels on the assemblies are acceptably low and as
such do not pose a threat for electrochemical failures. CSL attributes these bromide levels to the
fire retardant material in the FR-4 laminate.
Weak Organic Acids. Weak organic acids (WOAs), such as adipic or succinic acid,
serve as activator compounds in many fluxes, especially no-clean fluxes. WOAs are typically
4-38
-------
benign materials and are therefore not a threat to long term reliability. In order to avoid
formulation disclosure difficulties with flux manufacturers, all detected WOA species were
grouped together and referred to collectively as WO As.
WOA levels vary greatly, depending on the delivery method (e.g., foam vs. spray) and the
preheat dynamics. In general, water-soluble fluxes have a much lower WOA content than do
low-solids (no clean) fluxes, and the amount of residual WOA is proportional to the amount of
residual flux. Bare boards typically do not contain WOA residues.
Table 4-28. Acceptance Levels for Weak Organic Acids
Process
Spray-applied, low solids solder paste deposition
Foam-applied flux process w/air knife
Spray-applied, low solids flux
Level
0 - 20 (ig/in2
20-120 (ig/in2
250 - 400 (ig/in2
When WOA levels are under 400 |ig/in2, the residues are generally not detrimental.
Excessive WOA amounts (appreciably greater than 400 |ig/in2) present a significant reliability
threat for finished assemblies. Low levels of WOA can also create electrical performance
problems in certain applications.
• An excessive amount of flux can produce the situation in which the thermal energy of
preheat is spent driving off the solvent, therefore not allowing the flux to reach its full
activation temperature. Unreacted flux residues readily absorb moisture that promotes the
formation of corrosion and the potential for current leakage failures.
• Fully reacted and therefore benign WO As act as insulators that, even at levels as low as
10 |ig/in2, can potentially create a high resistance contact-to-contact resistance problem on
devices such as switches.
The observed levels of WO As on all 20 boards are typical and therefore are not
detrimental from an electrochemical standpoint. As expected, more WOA is evident on the
boards processed with low residue fluxes than on those processed with water soluble fluxes.
4.1.15 Summary and Conclusions
The test PWA provides electrical responses for 23 individual circuits that fall into the
following seven major circuit groups:
• high current low voltage (HCLV);
• high voltage low current (HVLC);
high speed digital (HSD);
• high frequency low pass filter (HF LPF);
• high frequency transmission line coupler (HF TLC);
• leakage networks; and
• stranded wire (SW).
4-39
-------
The first four circuit groups have both PTH and SMT components.
These characteristics make the test PWA an excellent discriminating test vehicle to
discover problem areas associated with new circuit card technologies, materials, and processes.
Exposure to environmental conditions such as the 85/85, thermal shock, and mechanical shock
used in this test program can accelerate the discovery process. Table 4-29 illustrates how
problem areas developed during the three tests.
Table 4-29 clearly identifies the HF LPF circuits as a problem area. The main problem
was related to open PTHs, which were discussed previously in Section 4.1.10. The HF LPF
anomalies resulted from a combination of board fabrication materials and processes and board
design (i.e., the small diameter vias in the FTP LPF circuit). Product designers should be aware of
these phenomena when considering a change to the new surface finishes.
Table 4-29. Frequency of Anomalies by Individual Circuit Over Test Times
Circuitry
Post-
85/85
Post-
Thermal
shock
Post-
Mechanical
Shock
Comments
HCLV
1
2
HCLV PTH
HCLV SMT
0
0
0
0
12
158
Some should be subjected to failure analysis.
SMT components came off board during
mechanical shock.
HVLC
3
4
HVLC PTH
HVLC SMT
0
1
0
1
0
164
Excellent performance throughout.
SMT components came off board during
mechanical shock.
HSD
5
6
HSD PTH
HSD SMT
0
1
2
2
2
1
Component problem.
Component problem.
HFLPF
7
8
9
10
11
12
HF PTH 50MHz
HF PTH f(-3dB)
HF PTH f(-40dB)
HF SMT 50MHz
HF SMT f(-3dB)
HFSMT
f(-40dB)
4
4
4
0
0
1
15
15
13
18
16
27
15
18
14
30
29
65
Perform failure analysis related to open PTH
(see Section 4. 1.10).
Perform failure analysis related to open PTH
(see Section 4.1.10).
Perform failure analysis related to open PTH
(see Section 4. 1.10).
Perform failure analysis related to open PTH
(see Section 4.1.10).
Perform failure analysis related to open PTH
(see Section 4. 1.10).
Perform failure analysis related to open PTH
(see Section 4.1.10).
4-40
-------
Circuitry
Post-
85/85
Post-
Thermal
shock
Post-
Mechanical
Shock
Comments
HFTLC
13
14
15
16
17
HF TLC 50MHz
HF TLC 500MHz
HF TLC IGHz
HF TLC RNF
HF TLC RNR
0
0
0
1
0
0
1
2
7
1
1
5
Minor anomalies.
Minor anomalies.
Minor anomalies.
Minor anomalies.
Leakage
18
19
20
21
10-milPads
PGA-A
PGA-B
Gull Wing
0
0
0
1
0
0
0
0
0
0
0
0
Excellent performance throughout.
Excellent performance throughout.
Excellent performance throughout.
Excellent performance throughout.
Stranded Wire
22
23
SW 1
SW2
0
0
0
1
1
4
Excellent performance throughout.
Minor anomalies.
Abbreviations and Definitions:
HCLV - high current low voltage
HF - high frequency
HSD - high speed digital
HVLC - high voltage low current
PGA - pin grid array
PTH - plated through hole
SMT - surface mount technology
SW - stranded wire
TLC - transmission line coupler
With the exception of the HCLV SMT and HVLC SMT circuits in the mechanical shock
test, the surface finishes under study were very robust to the environmental exposures. When
assessing the HCLV SMT and HVLC SMT results, product and process designers should
consider the severity of the mechanical shock test (25 drops, five times on each edge excluding
the connector edge and five times on each face, to a concrete surface from a height of one meter).
Also, HCLV SMT and HVLC SMT anomalies due to SMT components coming off the board
during the execution of the mechanical shock test were equally distributed across all surface
finishes including the HASL baseline.
Based on the results of the Failure Analysis:
• Observed levels of bromide and WOA on all 20 assemblies are typical and therefore not
detrimental from an electrochemical standpoint.
• Based on the fact the tested boards with known anomalies exhibit levels near or below
CSL's recommended guidelines, there is reasonable confidence that the anomalies are not
the result of chloride, bromide, or WOA contamination.
4-41
-------
From an overall contamination standpoint, the five non-HASL surface finishes tested in
this analysis performed as well if not better against the HASL finish.
Solder joint cracking failures were greater with the HASL finish than with the alternative
finishes. The opens occurred along the interface of the component leads on these older
PTH technology boards.
4.1.16 Boxplot Displays
Boxplot displays are presented here for selected results as discussed in this Chapter.
Boxplots of the remaining test results are presented in Appendix F.
4-42
-------
Pre-Test
HCLV PTH
Boxplots of HCLV PTH by SiteFlux
(means are indicated bysolid circles)
HASL
7.5 —
7.4 —
7.3 —
1 72~
§ 7.1-
I
7.0 —
6.9 —
6.8 —
SiteFlux
JL
.
1
„
i
0
X
OSP ImmSn Imm
*
JL
"
^ ,
T I I
- CN co -^ in
WS WS
h
I
"1 In
n A
pJL.
1
•
1 •
1 R u y •
- T T
1 U I
T 1
1
i
1 1
• h
• H .
•
U
I I I I I ill
cor^cocnot-cNco-^-in
WS WS WS
WS
JL
.
-
1
CO
Ag Ni/Au Ni/Au/Pd
JL
•
-
rl A
T
i
•
T
• •
J U T * a
I
T
*
ii ill
h- co o o t- CN co
wsws ws ws ws
Figure 4-1. Boxplot Displays for HCLV PTH Measurements (volts) at Pre-test by Surface Finish
Post 85/85
HCLV PTH
HASL
Boxplots of DPHCLV P by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg Ni/Au Ni/Au/Pd
0.5 —
0.4 —
0.3 —
0.2 —
0.1 —
>
J
) 0.0 —
i -0.1 —
-0.2 —
-0.3 —
-0.4 —
SiteFlux
T-
1
-
1
I
CN
J
C
*
\ J
o
X
1 1
> -*r in tc
\
> r
|
1
-
T
i
•^
c
l
T
1
o
c
i
T
i
n
c
-
T
i
D
|
1
1
1
C
1
~
T
i
N
C
1
1
T
i
o
1
•q
1
ri
r
X
1
r in c
1
•
-
1
o
t^
|
1
_L
,
T
i
»
1
-
T
1
l
3)
1
1
C
O
r
3
J
1
m
I
1
;—
>J
1
•
1
1
>i
-vj
1
1
CO
CN
ws ws
ws ws
ws
ws
wsws
ws ws
ws
Figure 4-2. Boxplot Displays for HCLV PTH Post 85/85 - Pre-test Measurements (volts) by Surface
Finish
(Acceptance Criterion = A<0.5V)
4-43
-------
Post Thermal Shock
HCLV PTH
HASL
Boxplots of DTHCLV P by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
0.5 —
] 0.0 —
)
1
-0.5 —
1
1
E
1
!
T
1
i
SiteFlux --
O
I
2 —
0 —
Boxplots of DM HCLV P by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
i
SiteFlux
i i r
CN CO I1
WS WS
\ r
in CD
i r
co en
nr
o
\ r
CN CO
\ r
in CD
\ r
CO O)
n i i r
ws ws
ws
ws
wsws
ws ws
ws
Figure 4-4. Boxplot Displays for HCLV PTH Post MS - Pre-Test Measurements (volts) by Surface
Finish
(Acceptance Criterion = A<0.5V)
4-44
-------
Pre-Test
HCLV SMT
Boxplots of HCLV SMT by SiteFlux
HASL
(means are indicated bysolid circles)
OSP ImmSn ImmAg Ni/Au Ni/Au/Pd
o
7.45 —
7.40 —
7.35 —
7.30 —
7.25 —
7.20 —
7.15 —
7.10 —
7.05 —
7.00 —
SiteFlux
1
i
c
-
•
*
1
N
C
-
1
1
O
1
1
1
1
T
L
JL
i
J
\
n
c
1
-
•
1
1
D
t
\
1
1
•^
C
-
T
i
o
J
a
L
1
c
JL
T
i
D
-
I
T
T
i
i
i
^
i
r
J C
1
1
*
1
o
JL
•
T
i
;r
L
1
T
1
n
c
1
•
1
I
D
r
i
1
hr
T
i
i
1
•
i
»
i
a
)
-
T
i
D
-vj
-
-
1
*
1
r—
-vj
0
O
1-1
1 •
1
J CO
J CM
WS WS WS WS
ws
ws wsws ws ws ws
Figure 4-5. Boxplot Displays for HCLV SMT Measurements (volts) at Pre-Test by Surface Finish
u~ Ifo^T Boxplots of DPHCLV S by SiteFlux
HULV olvl 1
(means are indicated bysolid circles)
0.3 —
0.2 —
0.1 —
^
O "°'1 ~~
-0.3 —
-0.4 —
-0.5 —
SiteFlux
HASL OSP I mm
'
h
r
1
/i L (i i
• o i1! n t\
, = • . i H y
"n B
T
*
Sn
i
T
i i i i i i iii
•t-CMCO-^-inCOI-^CQCnOt-CMCO'
1
1
•
1
ij-
ImmAg
J
a
m i
L
i
i
1
* D
i i
in CD r~
Ni/Au Ni/Au/Pd
i
T
1
»
1
i
•
T
i
3)
1
* n
. a
*
1-1
•
•
T
i i i
O T- CN CO
CM CM CM CM
WS WS WS WS
ws
ws wsws ws ws ws
Figure 4-6. Boxplot Displays for HCLV PTH Post 85/85 - Pre-Test Measurements (volts) by Surface
Finish
(Acceptance Criterion = A<0.5V)
4-45
-------
Post Thermal Shock
HCLV SMT
HASL
Boxplots of DTHCLV S by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
0.5
0.4
0.3
0.2
0.1
0.0
H -0.1
-0.2
-0.3
-0.4
SiteFlux
o
\
CO
nr
i
r
en
r
CD
nr
CO
n i i r
ws ws ws ws ws ws wsws ws ws ws
Figure 4-7. Boxplot Displays for HCLV PTH Post TS - Pre-Test Measurements (volts) by Surface
Finish
(Acceptance Criterion = A<0.5V)
Post Mechanical Shock
HCLV SMT
HASL
O
3 —
1 -
0 —
Boxplots of DM HCLV S by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
SiteFlux
J
T-
ll
'
1
JL
.
I
1
o
J
•^
lal^
58
*
i i
1
t
x-
1
1
]
n
c
rq
"
T
i
i
D
J
1
T-
1
<
J^
T
X-
1
N
1
r
1
r
X
1
JL
•
T
1
l
n
n
1
•
1
i
T
1
»
,
•
T
j>
J
i
1
•)
c
o
L
r
f
3
J
1
T
;—
>j
j
i
n
*
0
o
L
1
r
f
^
u
n
•
T
1
•o
-sj
WS WS WS WS
ws
WS WSWS WS WS WS
Figure 4-8. Boxplot Displays for HCLV PTH Post MS - Pre-Test Measurements (volts) by Surface
Finish
(Acceptance Criterion = A<0.5V)
4-46
-------
Pre-Test
HF PTH 50MHz
HASL
o
LO
I
CL
LL
Boxplots of HF PTH50 by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
-0.3 —
-0.4 —
-0.5 —
-0.6 —
-0.7 —
-0.8 —
-0.9 —
-1.0 —
-1.1 —
-1.2 —
UX
t $'0
i i i i
i-
8i?S?
*
1 1 1 1 1
O t- O) O •<-
fl 8
i i
-------
Post Mechanical Shock
HF PTH f(-3dB)
HASL
Boxplots of DMHF PTH by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
CO
i
200 —
150 —
100 —
50 —
-50 —
-100 —
-150 —
-200 —
-250 —
lux
X
X
X
. . • * *
X
I I I I I I I I I
WS WS WS WS
X-
•
I
O i-
: * *
i
n "
u
T
*
•
U
•
X
T
i i i i i i i i
WS WS WSWS W!
I I I I
O t-
-------
Post Mechanical Shock
HF SMT 50MHz
HASL
o
LO
Boxplots of DMHF SMT by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
0 —
-10 —
-20 —
-30 —
-40 —
-50 —
-60 —
-70 —
-80 —
-90 —
- * n
If
- , .
'
* I
* 1
j
•
•
-
.
-
T
i
x
T "
WS WS WS WS
P
T
D
-
-
•
1
*
-
1
= »
T
-
B
T
J
1 L
1 x It
L »*••*•
1
1 1 1 1 1 1 1 1 1 1 1 1
WS WS WSWS WS WS WS
Figure 4-13. Boxplot Displays for HF SMT 50MHz Post MS - Pre-Test Measurements (dB) by Surface
Finish
(Acceptance Criterion = +5dB of Pre-test)
Post Mechanical
HF SMT f(-3dB)
HASL
Boxplots of DMHF SMT by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
600 —
500 —
400 —
**•
i i i i
co en o i-
WS WSWS WS WS
* *
CN CO
CN CN
WS
Figure 4-14. Boxplot Displays for HF SMTf(-SdB) Post MS - Pre-Test Measurements (MHz) by Surf.
Finish
(Acceptance Criterion = +50MHz of Pre-test)
4-49
-------
Post Mechanical Shock
HF SMT f(-40dB)
HASL
Boxplots of DMHFSMT- by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
0 —
§ "20° ~
U? -400 -
^
Q
-600 —
-800 —
SiteFlux
•
i
X
I
t- ^
J C
1
^
o
1
1
1
T
1
1
=r
L
•
o
1
1
•i
C£
1 1
«
) t--
I
C
1
T
1
o
n
i
* '
i
r
) T
1
1
T
1
* e
1*
*
1
i p i
i i
r
X
•i
f
1
L
«
) C
JL
,
T
1
D
t^
i •
1 "
i
X
1 1 1 1 1 1
WS WS WS WS
WS
WS WSWS WS WS WS
Figure 4-15. Boxplot Displays for HF SMTf(-40dB) Post MS - Pre-Test Measurements (MHz) by
Surface Finish
(Acceptance Criterion = +50MHz of Pre-test)
Post Mechanical Shock
HF TLC 50MHz
HASL
o
LO
LL
40 —
30 —
20-
10-
0 —
SiteFlux
Boxplots of DMHF TL by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
X
x •
x-
8 • • *•
*
X
B
X
1
'
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
WS WS WS WS
*
1
-
1
* * * *
*
1 1 1 1
CO O) O t-
WS WS WSWS WS WS
.. d
i i
CN CO
CN CN
WS
Figure 4-16. Boxplot Displays for HF TLC 50MHz Post MS - Pre-Test Measurements (dB) by Surface
Finish
(Acceptance Criterion = +5dB of Pre-test)
4-50
-------
Boxplots of DMHF TL5 by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
SiteFlux
\ ii i r
CN co i- in
Ni/Au Ni/Au/Pd
4 —
3 —
2 —
1 —
0 —
-1 —
-2 —
-3 —
-4 —
-5 —
-6 —
I
1 i
5 i + 0
X
*
h
•
I
*
s * h *
H
v
1
*
JL r
5 i a
* 1
X
1
- «
r
X
X fl
e D
X
1
.
•
T
X
X
• 8 a i
*
X
X
X
$ i
\ ii i i i i i i i r
ws ws ws ws
ws
ws wsws ws ws
ws
Figure 4-17. Boxplot Displays for HF TLC 500MHz Post MS - Pre-Test Measurements (dB) by Surface
Finish
(Acceptance Criterion = +5dB of Pre-test)
Post Mechanical Shock
HF TLC RNR
HASL
Boxplots of DMHFTLRN by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
10 —
0 —
3
z
OL
t -10 -
^
^
Q
-20 —
-30 —
SiteFlux
X
* X
-s.l
*
•
X
*
*
I I I I
i- CN CO I1
g.
* * - ? -
1 1 1 1 1
in CD r~ co en
X
*
X
h - * * *
•1 X-
LJ x-
T *
i i i i i
O T- (N CO ^1
r
L
L
ir
i
JL
in
M
T
i i
) CD r^
* X
X
1 1 1 1
co en o i-
•• 1^
1 1
CN CO
CN CN
WS WS WS WS
ws
ws wsws ws ws
ws
Figure 4-18. Boxplot Displays for HF TLC RNR Post MS - Pre-Test Measurements (dB) by Surface
Finish
(Acceptance Criterion = <10dB increase over Pre-test)
4-51
-------
Prp-TpQt
^ * lir* Boxplots of Pads by SiteFlux
10-Mil Pads
(means are indicated bysolid circles)
HASL OSP ImmSn ImmAg Ni/Au Ni/Au/Pd
15 —
14 —
13 —
en
T3
CO
CL
12 —
11 —
10 —
w
* r\ *
.
rj
¥
*
J :
' i
j
*
^ n
i i
a n
Y
•f
h
1
* 9
i
S '
LI
?
f
I I I I I I I I I I I I I I I I I I I I I I
SiteFlux i-cNco-^-incQi-^cQcnoi-cNco-^-incQi-^cQcnoi-cNco
WS WS WS WS WS WS WSWS WS WS WS
Figure 4-19. Boxplot Displays for 10-Mil Pad Measurements (Iog10 ohms) at Pre-Test by Surface Finish
(Acceptance Criterion = Resistance > 7.7 Iog10 ohms)
Post 85/85
10-Mil Pads
Boxplots of DPPads by SiteFlux
(means are indicated bysolid circles)
HASL
14 —
13 —
en
T3
CO 19
CL 12 -
CL
Q
11 —
10 —
SiteFlux
1
•
T
t
H
M
T
i
i
CN C
WS
-
T
OSP
1
1
1
9
Qn
*
1 1 1 1
j^
•
ImmSn ImmAg Ni/Au Ni/Au/Pd
i iii
j
T
o i- in CD r~ co
L
_
1
J
9 B Ij
p
~
•
T
A
y
Bfif
5 1
e ? Q 6
X
X
n
~ V
i i i i i i i i i i i i i i
OOi-CNCO-^-inCQI-^CQCnOi-CNCO
WS WS WS WS WS WSWS WS WS WS
Figure 4-20. Boxplot Displays for 10-Mil Pad Post 85/85 - Pre-Test Measurements (Iog10 ohms) by
Surface Finish
(Acceptance Criterion = Resistance > 7.7 Iog10 ohms)
4-52
-------
Pre-Test
PGA-A
HASL
Boxplots of PGA A by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
14 —
13 —
<
$ I*'
CL
11 —
10 —
SiteFlux
X
A 6
n
M
X
1
T
t
n *
H n
M
M
T
1 1 1 1 1 1 1
i
•
I
i
fl
f
1
8
i i
1
.
-
A
T h
1
-cNco^rLncor^cocno^cN
WS WS WS WS
WS
1
f
u
7
1 1
1
co -^ in co
T
a ,
rl
B
V 15
¥
8
1 1 1 1 1 1 1
I"- 00 O) O T- (N CO
WS WSWS WS WS WS
Figure 4-21. Boxplot Displays for PGA-A Measurements (Iog10 ohms) at Pre-Test by Surface Finish
(Acceptance Criterion = Resistance > 7.7 Iog10 ohms)
Pre-Test
PGA-B
HASL
Boxplots of PGA B by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
14 —
13 —
CO
CL
11 —
10 —
SiteFlux
X
1
1
T
i
r^i
fl *
X
fl
1 1 1
CN CO I1
WS WS
.
X
a
1 1
in co t
WS
1
1
•^
X
1
CO C
WS
1
T
1
n
1
8
I I
0 T- C
WS
1
.
-
T
1
N
B
^
X
1 1
co ^r
WS
B H
i i i
in co i~-
WSWS
8 5
i
X
.
f
1 1 1 1
CO 0) 0 T-
WS WS
T
1
CN
V
V
I
1
1
•^
K
Figure 4-22. Boxplot Displays for PGA-B Measurements (Iog10 ohms) at Pre-Test by Surface Finish
(Acceptance Criterion = Resistance > 7.7 Iog10 ohms)
4-53
-------
Pre-Test
Gull Wing
Boxplots of GullWing by SiteFlux
(means are indicated bysolid circles)
HASL
14 —
13 —
12 —
D)
^
"5 11
10 —
9 —
Site Flux
X PI
1 p
1
1
J.
i
T
1 1
t- CN CO
X
(\
u
JL
•
T
1 1
•^- in CD
OSP ImmSn ImmAg Ni/Au
u
* !
V
1
ft i
i *
^ i i *
T
X
Q
1 *
A
y
*
J
T
JL
^
Ni/Au/Pd
1
•
T
1
J
u
I
1 1 1
t- CN CO
CN CN CN
WS WS WS WS
WS
WS WSWS WS WS WS
Figure 4-23. Boxplot Displays for the Gull Wing Measurements (Iog10 ohms) at Pre-Test by Surface
Finish
(Acceptance Criterion = Resistance > 7.7 Iog10 ohms)
4-54
-------
4.2 COST ANALYSIS
Operating an efficient and cost-effective manufacturing process with strict control of
material and production costs is the goal of every successful company. Consumer demand for
smaller and lighter electronics is fueling rapid and continuous advancements in circuit technology,
such as higher aspect-ratio holes and tighter circuit patterns. This in turn forces manufacturers to
evaluate and replace aging manufacturing processes in order to keep up with the ever-increasing
technology threshold. These new processes represent a major capital investment to a company,
and emphasize the importance of selecting an efficient, cost-effective process that will allow the
company to remain competitive. As a result, manufacturers are seeking comprehensive and more
detailed cost data before investing in alternative processes.
This section presents a comparative cost analysis of the surface finishing technologies.
Costs were developed for each technology and equipment configuration (vertical, immersion-type
equipment; or horizontal, conveyorized equipment) for which data were available from the PWB
Workplace Practices Questionnaire and Performance Demonstration. Table 4-30 presents the
processes (alternatives and equipment configurations) evaluated.
Table 4-30. Surface Finishing Processes Evaluated in the Cost Analysis
Surface Finishing Alternative
HASL
Nickel/Gold
Nickel/Palladium/Gold
OSP
Immersion Silver
Immersion Tin
Non-Conveyorized
Conveyorized
Costs were analyzed using a cost model developed by the University of Tennessee
Department of Industrial Engineering. The model employs generic process steps and functional
groups (see Section 2.1, Chemistry and Process Description of Surface Finishing Technologies) to
form a typical bath sequence (see Section 3.1, Source Release Assessment) for each process
alternative. To develop comparative costs on a $/surface square foot (ssf) basis, the cost model
was formulated to calculate the cost of performing the surface finishing function on a job
consisting of 260,000 ssf (value corresponds to the average annual throughput for facilities using
HASL in the PWB Workplace Practices Questionnaire database).
Processes were also modeled at a throughput of 60,000 ssf, a number which corresponds
to the average annual throughput for facilities using a non-HASL alternative. This additional
modeling run was performed to examine the effects, if any, that operating throughput will have on
the normalized cost for each process. Although the calculations presented in this section are
based on the higher production operating conditions, similar calculations were performed using
lower production level data and the results of the two runs are compared at the end of the cost
analysis.
4-55
-------
The overall objective of this analysis was to determine the comparative costs of the
surface finishing technologies using a cost model that adheres to fundamental principles of cost
analysis. Other objectives were to make the analysis flexible and to consider environmental costs.
The cost model was designed to estimate the comparative costs of fully operational surface
finishing process lines. It does not estimate start-up costs for a facility switching to an alternative
surface finishing technology or the cost of other process changes that may be required to
implement a new surface finishing line. Section 4.2.1 gives an overview of the cost methodology
used in this analysis. Section 4.2.2 presents the cost categories defined for the analysis and
discusses the categories that could not be quantified. Section 4.2.3 presents an overview of the
simulation model purpose, approach and results, while Section 4.2.4 describes the activity-based
costing techniques and results. Section 4.2.5 details of the individual cost formulations and
presents sample cost calculations. Section 4.2.6 contains analysis results and conclusions.
4.2.1 Overview of the Cost Methodology
The costs of the surface finishing technologies were analyzed by identifying the steps in
each process, breaking each step down into its cost components, and determining the cost of each
component. Component costs were determined using a combination of traditional costing
mechanisms, computer simulation, and activity-based costing (ABC). Figure 4-24 presents the
hybrid cost formulation framework:
Surface Finish
Technologies
Development of
Cost Categories
Development of
Simulation Model
_L
Traditional Costs
Components
Activity-Based Cost
Components
Cost
Analysis
Figure 4-24. Hybrid Cost Analysis Framework
4-56
-------
The generic process descriptions, chemical baths, typical bath sequences, and equipment
configurations for each surface finishing process form the basis of the cost analysis and are
presented in Table 4-30 and Figure 2-1 from Chapter 2, Profiling of the Surface Finishing Use
Cluster. The process information was used to identify critical variables and to define the cost
categories to be calculated by the cost analysis. The cost categories were analyzed to identify the
data required to calculate the costs (e.g., unit costs; utilization or consumption rates; criteria for
performing an activity, such as chemical bath replacement; the number of times an activity is
performed). For each process, a computer simulation was developed using ARENA® computer
simulation software. The simulation model then was used to model each process under similar
operating conditions to determine operating data, such as overall production time, required by the
cost analysis. Individual cost formulas were developed using traditional cost techniques, while
costs typically allocated to overhead were quantified using ABC techniques. The costs were then
calculated and compared to the cost of the baseline, non-conveyorized, HASL process. A more
detailed description of each step is presented later in this chapter.
4.2.2 Cost Categories and Discussion of Unquantifiable Costs
Cost Categories
Table 4-31 summarizes the cost components considered in this analysis, gives a brief
description of each cost component and key assumptions, and lists the primary sources of data for
determining the costs. Section 4.2.5 gives a more detailed accounting of the cost components,
including sample cost calculations for each component. In addition to traditional costs, such as
capital, production, and maintenance costs, the cost formulation identifies and captures some
environmental costs associated with the technologies. In this regard, both simulation and ABC
assist in analyzing the impact of the surface finishing technologies on the environment.
Specifically, the amounts of energy and water consumed, as well as the amount of wastewater
generated, are determined for each surface finishing process.
Unquantifiable Cost Categories
The goal of this cost analysis was to perform a comparative cost analysis on the surface
finishing alternatives in the evaluation. Although every effort was made to characterize each cost
component listed in Table 4-31, data and/or process limitations prevented the quantification of
every component. A qualitative discussion of each of these costs is presented below.
4-57
-------
Table 4-31. Cost Component Categories
Cost Category
Capital
Cost
Material
Cost
Utility
Cost
Wastewater
Cost
Component
Primary
Equipment &
Installation
Facility
Process
Chemicals5858
Water
Electricity
Natural Gas
Publicly Owned
Treatment Works
(POTW) Permit
Wastewater
Pretreatment Cost
Wastewater
Discharge Costs
Description of Cost Component
Annualized cost of equipment with throughput
capacity of 60 panels/hr x URa; includes the cost of
delivery and installation of equipment; assumes 10
year equipment life and straight-line depreciation.
Annualized cost of floor space required to operate
surface finish process equipment x URa; assumes 25
year facility life and straight-line depreciation.
Costs of chemicals used in initial bath setup, bath
maintenance additions, and replacement of spent
process baths.
Water consumption costs based on number of rinse
stages per process line and normalized water flow
rates per stage.
Electricity costs based on daily electrical
consumption of surface finish process equipment
and days to complete job.
Natural gas consumption based on daily natural gas
consumption from drying ovens and days to
complete job.
Cost for permit to discharge wastewater to POTW.
Cost to pretreat wastewater prior to discharge to
POTW.
Fees for wastewater discharge assessed by local
utility.
Sources of Cost Data
Vendor quote for equipment and installation costs; time to
complete job from simulation.
Floor space requirements from IPC Workplace Practices
Questionnaire; unit cost for industrial floor space from
published sources.
Vendor quotes for chemical product cost; bath sizes from
IPC Workplace Practices Questionnaire; bath replacement
criteria from supplier data; number of bath replacements
required for job from simulation.
Number of rinse stages and normalized water flow rates
per stage from Section 5.1, Resource Conservation; cost of
water based on results reported by manufacturers from the
Pollution Prevention and Control Survey.
Daily electricity consumption from Section 5.2, Energy
Impacts; days to complete job from simulation; cost of
electricity based on national power grid from the Internal
Energy Agency.
Daily natural gas consumption from Section 5.2, Energy
Impacts; days to complete job from simulation; cost of
natural gas from the Knoxville Utilities Board (KUB).
Not quantified; assumed to be the same for all alternatives.
Not quantified; pretreatment costs are expected to differ
significantly among the alternatives, but inability to
separate pretreatment of surface finish wastes from other
process wastes made it impossible to reliably estimate
these costs.
Quantity of wastewater discharged assumed equal to water
usage; discharge fees based on fees charged by KUB.
4-58
-------
Cost Category
Production
Cost
Maintenance
Cost
Waste Disposal
Cost
Quality
Cost
Component
Labor
Transportation of
Materials
Bath Cleanup
Bath Setup
Sampling and
Analysis
Filter Replacement
Sludge Disposal
Filter Disposal
Defective Boards
Description of Cost Component
Labor costs for line operator, excluding labor costs
for maintenance activities (included under
maintenance costs). Assumes one line operator per
day per conveyorized process, 1 . 1 line operators per
day per non-conveyorized process, to reflect the
greater level of labor required.
Cost to transport chemicals required for bath
replacement from storage to process line.
Labor and material (excluding chemicals) costs to
clean up a chemical tank during bath replacement.
Labor and equipment costs to set up a chemical tank
after bath replacement.
Labor and materials costs for sampling and analysis
of chemical baths.
Labor costs for replacing bath filters.
Disposal cost to recycle or disposal of sludge from
wastewater treatment.
Disposal cost to recycle or dispose of bath filters.
Costs of defective boards due to failure of the
surface finish process line to apply an adequate
finish to the surface of the PWB.
Sources of Cost Data
Number of line operators based on IPC Workplace
Practices Questionnaire data and site visits; days to
produce job from simulation; labor rate based on published
data.
Cost of transporting materials from a bill of activity
(BOA); number of bath replacements required from
simulation.
Cost to clean up tank from BOA; number of bath cleanups
(replacements) required from simulation.
Cost to set up bath from BOA; number of bath setups
required from simulation.
Assumes analytical work done in-house. Cost for one
activity from BOA; annual number of samples from IPC
Workplace Practices Questionnaire adjusted using URa.
Labor cost for one activity from BOA; annual number of
filters replaced from IPC Workplace Practices
Questionnaire adjusted using URa.
Not quantified; sludge disposal costs are expected to differ
significantly among the alternatives, but insufficient data
were available to reliably estimate these costs. Factors
affecting sludge disposal cost include the characteristics of
the sludge (e.g., type of metal content, percent solids, waste
classification, etc.), the amount of sludge generated, and
the type of disposal (e.g., reclaim, disposal to landfill, etc).
Not quantified; filter disposal costs are not expected to
differ significantly among the alternatives, but insufficient
data on the type and size of waste filters made it difficult to
reliably estimate these costs. Factors affecting filter
disposal cost include the waste classification of the filter,
the size (weight and volume) of the filter, and the number
of waste filters generated.
Not quantified; assumed equal among the alternatives.
Performance Demonstration showed that all alternatives
can work at least as well as the baseline process as long as
they are operated according to supplier specifications.
1UR = utilization ratio = the time in days required to process 260,000 ssf ^ one operating year (280 days).
4-59
-------
Wastewater Treatment and Sludge Disposal Costs. PWB manufacturing consists of a
number of process steps (see Section 1.2.3 for an overview of rigid multi-layer PWB
manufacturing). In addition to the surface finishing process line, these steps include electroplating
operations and other steps which consume large quantities of rinse water and, consequently,
generate large quantities of wastewater. Most PWB manufacturers combine the effluents from
various process lines into one wastewater stream which is treated on-site in a continuous process
prior to discharge.
As part of the Pollution Prevention and Control Survey (U.S. EPA, 1998), PWB
manufacturers were asked to provide the following about their on-site wastewater treatment
facility:
a process flow diagram for wastewater treatment;
the quantity of sludge generated from wastewater treatment;
the percent solids of the sludge;
the costs of on-site wastewater treatment; and
the method and costs of sludge recycle and disposal.
Capital costs for wastewater treatment ranged from $1.2 million for a system purchased in 1980
with a capacity of 135 gallons per minute (gpm) to $4,000 for a system purchased in 1987 with a
capacity of nine gpm. Costs for operating an on-site wastewater treatment system were as high as
3.1 percent of total annual sales. The median cost for wastewater treatment operation was 0.83
percent, and the average was 1.02 percent of annual sales.
Wastewater treatment sludges from PWB electroplating operations are classified as an
F006 hazardous waste under the Resources Conservation and Recovery Act (RCRA); most
facilities combine effluents from the electroplating line with other process wastewaters. Eighty-
eight percent of respondents to the Pollution Prevention and Control Survey reported that
wastewater treatment sludges are sent to an off-site recycling facility to recover the metals. The
average and median costs for off-site recovery of sludge were $0.48/lb and $0.21/lb, respectively.
In general, the lower costs experienced by some respondents compared to others were due to
larger-size shipments and shorter distances to the recycling sites. In some cases, respondents
whose sludge had a higher solids content also reported lower costs; dewatered sludge has a higher
recovery value.
The PWB Workplace Practices Questionnaire attempted to characterize costs by
collecting information about the percentage contribution of the surface finishing line to the overall
wastewater and sludge generation rates. However, most manufacturers were unable to provide
this information and the data that were reported were of variable to poor quality.
A drag-out model was developed to determine the extent of chemical contamination of the
wastewater resulting from drag-out. The model was used to estimate quantities of the chemical
constituents in the wastewater. Model results are presented in Section 3.2, Exposure Assessment
and Appendix E. However, since the streams are co-mingled prior to treatment, industry sources
explained that it would be difficult to reliably quantify the effect of the surface finishing
wastewater stream on the treatment of the entire stream (e.g., a treatment chemical used to treat
4-60
-------
the surface finishing wastewater may have a stronger affinity for another compound that
may be present in the wastewater from another source, thus negatively affecting the treatment of
the surface finishing wastewater).
Because the surface finishing line is only one of several process lines that discharge
effluent to wastewater treatment, and because little or no information is available on the
contribution of the surface finishing line to overall wastewater effluents, on-site wastewater
treatment and sludge disposal costs could not be reliably estimated. However, costs of
wastewater treatment and sludge disposal are expected to differ significantly among the
alternatives, based on the compounds involved. For example, the presence of thiourea in the
immersion tin process may require an additional treatment step to break down the compound
prior to release. Silver is tightly regulated, thus the addition of an immersion silver process to a
facility may require additional treatment to prevent exceeding the relatively low effluent limit. A
detailed discussion of treatment concerns, systems, and options for each surface finishing process
is presented in Section 6.2, Recycle, Recovery, and Control Technologies Assessment.
Other Solid Waste Disposal Costs. Two other types of solid wastes were identified
among the technologies that could have significantly different waste disposal costs: filter disposal
cost and defective boards disposal costs. Table 4-32 presents the number of filters that would be
replaced in each process during a job of 260,000 ssf This is based on data from the PWB
Workplace Practices Questionnaire and a utilization ratio (UR) calculated for each process from
simulation results (Simulation results are discussed further in Section 4.2.3). The UR is the
percentage of time during the year required for the process to manufacture the required
throughput. While these results illustrate that the number of waste filters generated by the
processes differ significantly, no information is available on the characteristics of the filters used
by the processes. For example, the volume or mass of the filters and waste classification of the
filters (hazardous or non-hazardous) would significantly affect the unit cost for disposal.
Therefore, filter disposal costs were not estimated.
The number of defective boards produced by a process has significance not only from the
standpoint of quality costs, but also from the standpoint of waste disposal costs. Clearly, a higher
defect rate leads to higher scrap and, therefore, waste of resources. However, the Performance
Demonstration showed that each of the technologies can perform as well as the HASL process if
operated according to specifications. Thus, for the purposes of this analysis, no differences would
be expected in the defect rate or associated costs of the technologies.
4-61
-------
Table 4-32. Number of Filter Replacements by Surface Finishing Process
Surface Finishing Process
HASL, Non-conveyorized
HASL, Conveyorized
Nickel/Gold, Non-conveyorized
Nickel/Palladium/Gold, Non-conveyorized
OSP, Non-conveyorized
OSP, Conveyorized
Immersion Silver, Conveyorized
Immersion Tin, Non-conveyorized
Immersion Tin, Conveyorized
Filter Replacements
per Year a
354
354
119
162
150
150
19.5
150
150
Filter Replacements
Required to Produce
260,000 ssf b
55
28
90
162
19
9
4
40
57
a 90th percentile data based on PWB Workplace Practices Questionnaire data. Data not adjusted for throughput or
to account for differing maintenance policies at individual PWB manufacturing facilities.
b Values calculated by multiplying the filter replacements per year for a process by the utilization ratio for that
process.
4.2.3 Simulation Modeling of Surface Finishing Processes
A computer simulation was developed using ARENA® computer simulation software for
each surface finishing process. The purpose of the modeling is to simulate the operation of each
process on the computer under identical conditions to predict a set of key metrics (e.g., overall
production time, process down time, number of bath replacements) required to perform a
comparative cost analysis. The model is necessary because the data collected from actual
facilities, if available, would reflect the individual operating practices of each facility (e.g., bath
maintenance frequencies, rise water flow rates, PWB feed rates) preventing a valid comparison of
any process costs. Appendix G presents a graphic representation of the simulation models
developed for each of the surface finishing technologies.
Simulation modeling provides a number of benefits to the cost analysis, including the
following:
Simulation modeling replicates a production run on the computer screen, allowing the
analyst to observe a process when the actual process does not exist: in this case, the
generic surface finishing technologies, as defined in Figure 2-1, may not exist within any
one facility.
Simulation allows for process-based modifications and variations, resulting in inherent
flexibility within the system: models can be designed to vary the sequence of operations,
add or delete operations, or change process times associated with operations, materials
flows, and other variables.
Simulation modeling facilitates the comparison of technologies by modeling each
technology operating under a single, consistently applied performance profile developed
from data collected from industry.
4-62
-------
Simulation enables a study of the sensitivity of critical performance measures to changes in
underlying input variables (constant input variables may be modified in the sensitivity
analysis to determine the uncertainty associated with these input variables).
Direct results of the simulation model and results derived from simulation outputs include
the following:
the overall time the surface finishing line operates to produce the job;
the number of repetitions of an activity (e.g., bath replacements) over the course of
the job;
consumption rates (e.g., water, energy, and chemical consumption); and
production rates (e.g., wastewater generation).
Simulation results were combined with traditional cost components to adjust these costs
for the specified job. An example of this is the determination of equipment cost. Simulation
results were used to calculate a UR, defined as the amount of time in days required to produce
260,000 ssf divided by one operating year. A 280-day operating year was selected to match the
longest modeled operating time for any process (nickel/palladium/gold). Annualized equipment
costs were determined using industry sources for equipment price and depreciation guidelines
from the Internal Revenue Service. These costs were multiplied by the UR to determine the
equipment costs for the job being evaluated.
Simulation Model Assumptions
Several assumptions used in the simulation model are based on the characteristics of a
model facility presented in the Source Release Assessment and Exposure Assessment (Sections
3.1 and 3.2, respectively). Assumptions include the following:
The facility operates a surface finishing process line 280 days/year, one shift/day; [Note:
many facilities operate two shifts, but the Exposure Assessment and this analysis use first
shift data as representative. This assumption could tend to underestimate labor costs for
companies that pay higher rates to second shift workers. Alternatively, it could tend to
overestimate equipment costs for a company running two shifts and using equipment more
efficiently. However, because this assumption is used consistently across technologies, the
effects on the comparative cost results are expected to be minor.]
the surface finishing process line operates an average of 6.8 hr/shift;
the surface finishing line is down at least 1.2 hr/day for start-up time and for maintenance,
including lubricating of equipment, sampling of baths, and filter replacement;
additional down time occurs when the surface finishing line is shut down to replace a spent
or contaminated bath;
PWB panels that have been processed up to the surface finishing step are available
whenever the surface finishing process line requires them;
if a chemical bath is replaced at the end of the day, and the amount of time required to
replace the bath exceeds the time remaining in the shift hours, employees will stay after
hours and have the bath ready by the beginning of the next shift;
the entire surface finishing process line is shut down whenever a bath requires replacing,
4-63
-------
but partially processed racks or panels are finished before the line is shut down;
the surface finishing process only shuts down at the end of a shift and for bath
replacement, when required; and
the process is empty of all panels or racks at the end of each shift, and starts the process
empty at the beginning of a shift.
Further simulation assumptions have to be defined separately for conveyorized and non-
conveyorized systems. Conveyorized surface finishing process assumptions are as follows:
the size of a panel is 17.5 x 23.1" (from PWB Workplace Practices Questionnaire data for
conveyorized processes);
panels are placed on the conveyor whenever space on the conveyor is available, and each
panel requires 18 inches (including space between panels);
conveyor speed is constant, thus, the volume (gallons) of chemicals in a bath varies by
bath type (i.e., microetch, conditioner, etc.) and with the length of the process step (e.g.,
bath or rinse tank) to provide the necessary contact time (see Table 4-33 for bath
volumes); and
the conveyor speed, cycle time, and process down time are critical factors that determine
the time to complete a job.
Table 4-33. Bath Volumes Used for Conveyorized Processes
Chemical Bath
Cleaner
Microetch
Flux
Solder
OSP
Predip
Immersion Silver
Immersion Tin
Bath Volume by Surface Finishing Alternative
(gallons)
HASL
66.5
86.6
13.2
17.4
NA
NA
NA
NA
OSP
66.5
86.6
NA
NA
125
NA
NA
NA
Immersion Silver
66.5
86.6
NA
NA
NA
46.2
142
NA
Immersion Tin
66.5
86.6
NA
NA
NA
46.2
NA
140
NA: Not applicable.
Non-conveyorized surface finishing process assumptions are as follows:
the average volume of a chemical bath is 51.1 gallons (from PWB Workplace Practices
Questionnaire data for non-conveyorized processes);
only one rack of panels can be placed in a bath at any one time;
a rack contains 20 panels (based on PWB Workplace Practices Questionnaire data,
including the dimensions of a bath, the size of a panel, and the average distance between
panels in a rack);
the size of a panel is 4.22 ssf to give 84.4 ssf/rack;
the frequency at which racks are entered into the process is dependent upon the bottleneck
4-64
-------
or rate limiting step; and
the duration of the rate limiting step, cycle time, and process down time are critical factors
that determine the time to complete a job.
Simulation Model Input Values
Input values for the critical factors identified above (cycle time, down time, and conveyor
speed for conveyorized processes, and cycle time, down time, and duration of rate limiting step
for non-conveyorized processes) were developed from the PWB Workplace Practices
Questionnaire data and Product Data Sheets (Product Data Sheets, which are prepared by
suppliers, describe how to mix and maintain chemical baths). Tables 4-34 and 4-35 present time-
related inputs to the simulation models for non-conveyorized and convey orized processes,
respectively.
Table 4-34. Time-Related Input Values for Non-Conveyorized Processes
Non-Conveyorized Surface
Finishing Technology
HASL
Nickel/Gold
Nickel/Palladium/Gold
OSP
Immersion Tin
Time Required to
Replace a Bath b
(minutes)
136
116
113
149
85
Rate Limiting
Bath
Cleaner
Electroless Nickel
Electroless Nickel
Cleaner
Immersion Tin
Time in Rate
Limiting Bath b
(minutes)
3.47
18.3
18.3
3.47
8.55
Process
Cycle Time b
(minutes)
7.94
86.8
109
22.6
27.0
a Values may represent chemical products from more than one supplier. For example, two suppliers of nickel/gold
chemical products participated in the project. Input values may underestimate or overestimate those of any one
facility, depending on factors such as individual operating procedures, the chemical or equipment supplier, and the
chemical product used.
b Average values from the PWB Workplace Practices Questionnaire and Performance Demonstration observer
sheets.
Table 4-35. Time-Related Input Values for Conveyorized Processes
Conveyorized Surface
Finishing Technology
HASL
OSP
Immersion Silver
Immersion Tin
Time Required
to Replace a
Bath"
(minutes)
136
149
114
85
Length of
Conveyor b
(feet)
41.3
54.1
34.0
20.0
Process Cycle
Timeb
(minutes)
4.86
5.22
11.2
12.3
Conveyor
Speed c
(ft/min)
8.50
10.4
3.04
1.63
a Values may represent chemical products from more than one supplier. For example, two suppliers of OSP
chemical products participated in the project. Input values may underestimate or overestimate those of any one
facility, depending on factors such as individual operating procedures, the chemical or equipment supplier, and the
chemical product used.
b Average values from PWB Workplace Practices Questionnaire and Performance Demonstration observer sheets.
0 Conveyor speed was calculated by dividing the length of conveyor by the process cycle time.
4-65
-------
The input values for the time required to replace a bath (in Tables 4-34 and 4-35) are used
together with bath replacement criteria in the calculation of down time. Suppliers provide
instructions with their products (called Product Data Sheets for the purposes of this project) that
describe when a bath should be replaced because it is expected to be spent or too contaminated to
be used. These replacement criteria are usually given in one of four forms:
as a bath production capacity in units of ssf per gallon of bath;
as a concentration-based criterion that specifies an upper concentration limit for
contaminants in the bath, such as grams of copper per liter in the microetch bath;
as elapsed time since bath make-up; or
as a number of chemistry (or metal) turnovers before replacement.
Bath replacement criteria submitted by suppliers were supplemented with PWB
Workplace Practices Questionnaire data and reviewed to determine average criteria for use in the
simulation models. Criteria in units of ssf/gallon were preferred because these could be correlated
directly to the volume of a bath. For baths with replacement criteria expressed in number of
chemical turnovers, the ssf/gallon for that bath was adjusted by a factor equal to the number of
metal turnovers (e.g., the replacement criteria for a 750 ssf/gal bath with two metal turnovers was
considered to be 1500 ssf/gal of bath). Table 4-36 presents bath replacement criteria used to
calculate input values for the nickel/gold processes, as an example. Appendix G presents the bath
replacement criteria recommended by chemical suppliers, and the input values used in this analysis
for the remaining surface finishing technologies.
Table 4-36. Bath Replacement Criteria for Nickel/Gold Processes
Chemical Bath
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Immersion Gold
Bath Replacement Criteria a
(ssf/gal)
750
570
830
1500
130
890
a Values were selected from data provided by two nickel/gold suppliers. To convert to
units of racks per bath replacement for non-conveyorized processes, multiply by 51.1
gallons (the average bath size) and divide by 84.4 ssf (ssf per rack). To convert to units
of panels per bath replacement for conveyorized processes, multiply by the bath size in
gallons and divide by 5.66 ssf/panel.
Simulation Model Results
Simulation models were run for each of the surface finishing processes. Simulation
outputs used in the cost analysis include:
frequency and duration of bath replacements;
overall production time required for each process; and
down time incurred while producing 260,000 ssf.
4-66
-------
For example, the frequency and duration of bath replacements for nickel/gold that were obtained
from the simulation modeling are shown in Table 4-37. The frequency of bath replacements for
each bath type was calculated by the simulation model using the bath replacement criteria
presented for each bath in Table 4-36. Using the average time of bath replacement determined
from the PWB Workplace Practices Questionnaire data, the total down time associated with the
replacement of each bath type was determined. Summing over all baths, bath replacement
consumed 36.7 hours (2,200 minutes) to produce 260,000 ssf when using the non-conveyorized
nickel/gold process. Bath replacement simulation outputs for the other surface finishing processes
are presented in Appendix G.
Table 4-37. Frequency and Duration of Bath Replacements for Non-Conveyorized
Nickel/Gold Process
Chemical Bath
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Immersion Gold
Total
Frequency of
Replacement
7
9
6
4
40
6
72
Avg. Time of Replacement
(minutes)
116
116
116
116
116
116
116
Total Time of Replacement
(minutes)
812
1,044
696
464
4,640
696
8,352
Table 4-38 presents the other simulation outputs: the total production time required and
the down time incurred by the surface finishing processes while producing 260,000 ssf of PWB.
Total production time is the sum of actual operating time and down time. The operating time is
based on the process producing 260,000 ssf of PWB and operating 6.8 hr/day. The down time
includes the remaining 1.2 hr/day that the line is assumed inactive, plus the time the process is
down for bath replacements. The amount of process down time due to a bath replacement, shown
in Table 4-37, may be adjusted by the model if the bath changeout occurs at the end of the day,
when the replacement duration exceeds the time remaining in the day. (7,670 minutes of
downline are reported in Table 4-38, indicating that 680 minutes of the 8,352 reported in Table
14-37 occurred at the end of the day.) In this instance, the worker is considered to complete the
bath replacement during the remaining 1.2 hours of the day set aside for process maintenance.
The simulation model output reports for each process are presented in Appendix G.
4-67
-------
Table 4-38. Production Time and Down Time for the Surface Finishing
Processes to Produce 260,000 ssf of PWB
Surface Finishing Process
HASL, Non-conveyorized
HASL, Conveyorized
Nickel/Gold, Non-conveyorized
Nickel/Palladium/Gold, Non-conveyorized
OSP, Non-conveyorized
OSP, Conveyorized
Immersion Silver, Conveyorized
Immersion Tin, Non-conveyorized
Immersion Tin, Conveyorized
Total Production Time a
minutes
17,830
8,890
86,500
114,240
14,360
6,570
26,190
30,680
43,660
days
43.7
21.8
212
280
35.2
16.1
64.2
75.2
107
Total Down Time a
minutes
2,330
938
7,670
11,380
2,530
1,020
1,390
1,880
1,020
days
5.7
2.3
18.8
27.9
6.2
2.5
3.4
4.6
2.5
To convert from minutes to days, divide by 6.8 hr/day (408 minutes).
4.2.4 Activity-Based Costing
ABC is a method of allocating indirect or overhead costs to the products or processes that
actually incur those costs. ABC complements the traditional costing /modeling efforts of this
assessment by allowing the cost of tasks that are difficult to quantify, or are just unknown by
industry, to be determined. Activity-based costs are determined by developing a BOA for critical
tasks, which are defined as tasks required to that support the operation of the surface finish
process line. A BOA is a listing of the component activities involved in the performance of a
certain task, together with the number of times each component activity is performed. The BOA
determines the cost of a task by considering the sequence of actions and the resources utilized
while performing that task. In this analysis, the costs of critical tasks determined by a BOA are
combined with the number of times a critical task is performed, derived from simulation results to
determine the total costs of that activity.
BOAs were developed for the following critical tasks performed during the operation of
the surface finishing process:
chemical transport from storage to the surface finishing process;
tank cleaning;
bath setup;
bath sampling and analysis; and
filter replacement.
These BOAs were developed based on information developed for earlier projects
involving similar tasks and on information gathered through site visits and general process
knowledge. The following discussion uses the BOA for chemical transport, presented in Table
4-39, as an example of how BOAs were developed and used. Appendix G presents the BOAs for
the remaining activities.
4-68
-------
Key assumptions were developed to set the limits and to designate the critical activity's
characteristics. For chemical transport, the assumptions were as follows:
• chemical costs are not included in the BOA, but are considered within material costs;
• labor costs considered are independent of those included within production costs;
• employee labor rate is $10.24 per hour, consistent with the rate for an operator-level job;
• multiple chemicals are required for each bath replacement;
• all chemicals for a bath replacement are transported on one forklift trip;
• chemicals are purchased in containers larger than the line containers used to move
chemicals to the surface finishing process;
• all chemicals are stored in a central storage location;
• chemicals are maintained in central storage via inventory tracking and physical monitoring;
• forklift operation costs are $580/month or $0.06/minute, which includes leasing,
maintenance, and fuel;
• forklifts are used to move all chemicals; and
• forklifts are parked in an assigned area when not in use.
Each critical task was broken down into primary and secondary activities. For example,
chemical transport has six primary activities: paperwork associated with chemical transfer,
moving forklift to chemical storage area, locating chemicals in storage area, preparation of
chemicals for transfer, transporting chemicals to the surface finishing process, and transporting
chemicals from the surface finishing process to the bath. The secondary activities for the primary
activity of "transport chemicals to the surface finishing process" are: move forklift with
chemicals, unload line containers, and park forklift in assigned parking area. For each secondary
activity the labor, material, and forklift costs are calculated. The forklift costs are a function of
the time that labor and the forklift are used. On a BOA, the sum of the costs of a set of secondary
activities equals the cost of the primary activity.
Continuing the example, for a chemical transport activity that requires two minutes, the
labor cost is $0.34 (based on a labor rate of $10.24/hour) and the forklift cost is $0.12 (based on
$0.06/minute). Materials costs are determined for materials other than chemicals and tools
required for an activity. The total of $9.28 shown in Table 4-39 represents the cost of a single act
of transporting chemicals to the surface finishing line. The same BOAs are used for all surface
finishing technologies because either the activities are similar over all technologies or information
is unavailable to distinguish between them. However, individual facilities could modify a BOA to
best represent their unique situations. Table 4-40 presents costs to perform each of the critical
tasks one time.
4-69
-------
Table 4-39. BOA for Transportation of Chemicals to the Surface Finishing Process Line
Activities
A. Paperwork and Maintenance
1 . Request for chemicals
2. Updating inventory logs
3. Safety and environmental record keeping
B. Move Forklift to Chemical Storage Area
1 . Move to forklift parking area
2. Prepare forklift to move chemicals
3. Move to line container storage area
4. Prepare forklift to move line container
5. Move forklift to chemical storage area
C. Locate Chemicals in Storage Area
1 . Move forklift to appropriate areas
2. Move chemical containers from storage to
staging
3. Move containers from staging to storage
D. Preparation of Chemicals for Transfer
1 . Open chemical container(s)
2. Utilize correct tools to obtain chemicals
3. Place obtained chemicals in line
container(s)
4. Close chemical container(s)
5 . Place line container(s) on forklift
E. Transport Chemicals to Line
1 . Move forklift to line
2. Unload line container(s) at line
3 . Move forklift to parking area
F. Transport Chemicals from Line to Bath
1 . Move line container(s) to bath
2. Clean line container(s)
3 . Store line container(s) in appropriate area
Time
(min)
2
1
2
2
5
2
3
2
1
2
2
1
3
3
1.5
1
2
1
2
1
2
1
Resources
Labor a
$0.34
$0.17
$0.34
$0.34
$0.85
$0.34
$0.51
$0.34
$0.17
$0.34
$0.34
$0.17
$0.51
$0.51
$0.26
$0.17
$0.34
$0.17
$0.34
$0.17
$0.34
$0.17
Materials b
$0.10
$0.05
$0.10
$0.00
$0.00
$0.00
$0.00
$0.00
$0.00
$0.00
$0.00
$0.05
$0.05
$0.00
$0.00
$0.00
$0.00
$0.00
$0.00
$0.00
$0.20
$0.00
Forklift c
$0.00
$0.00
$0.00
$0.12
$0.30
$0.12
$0.18
$0.12
$0.06
$0.12
$0.12
$0.00
$0.00
$0.00
$0.00
$0.06
$0.12
$0.06
$0.12
$0.00
$0.00
$0.00
Total Cost per Transport
Cost
($/transport)
$0.44
$0.22
$0.44
$0.46
$1.15
$0.46
$0.69
$0.46
$0.23
$0.46
$0.46
$0.22
$0.56
$0.51
$0.26
$0.23
$0.46
$0.23
$0.46
$0.17
$0.54
$0.17
$9.28
a Labor rate = $10.24 per hour.
b Materials do not include chemicals or tools.
0 Forklift operating cost = $0.06 per minute.
4-70
-------
Table 4-40. Costs of Critical Tasks
Task
Transportation of Chemicals
Tank Cleaning
Bath Setup
Sampling and Analysis
Filter Replacement
Cost
$9.28
$67.00
$15.10
$3.70
$17.50
4.2.5 Cost Formulation Details and Sample Calculations
This section develops and describes in detail the cost formulation used for evaluating the
surface finishing process alternatives. The overall cost was calculated from individual cost
categories that are common to, but expected to vary with, the individual process alternatives. The
cost model was validated by cross-referencing the cost categories with Tellus Institute (White et
al., 1992), and Pacific Northwest Pollution Prevention Research Center (Badgett et al., 1995).
The cost model for an alternative is as follows:
TC = C + M + U + WW + P + MA
where,
TC
C
M
U
WW =
P
MA =
total cost to produce 260,000 ssf
capital cost
material cost
utility cost
wastewater cost
production cost
maintenance cost
The unit cost of producing 260,000 ssf is then represented as follows:
Unit Cost ($/ssf) = TC ($) / 260,000 ssf
The following sections present a detailed description of cost calculation methods together
with sample calculations, using the non-conveyorized nickel/gold process as an example. Cost
summary tables for all of the process alternatives are presented at the end of this section.
Capital Costs
This section presents methods and sample calculations for calculating capital costs.
Capital costs are one-time or periodic costs incurred in the purchase of equipment or facilities. In
this analysis, capital costs include the costs of primary equipment including equipment installation,
and facility space utilized by the surface finishing process line. Primary equipment is the
equipment vital to the operation of the surface finishing process without which the process would
not be able to operate (i.e., bath tanks, heaters, rinse water system, etc.). Installation costs
4-71
-------
include costs to install the process equipment and prepare it for production. Facility space is the
floor space taken up by the actual process equipment plus an additional buffer area necessary for
operation of the equipment by workers and access for maintenance and repair.
Total capital costs for the surface finishing processes were calculated as follows:
C = (E + I + F)xUR
where,
E = annualized capital cost of equipment ($/yr)
I = annualized capital cost of installation ($/yr), which is sometimes included in the
cost of the equipment
F = annualized capital cost of facility ($/yr)
UR = utilization ratio, defined as the time in days required to manufacture 260,000 ssf
divided by one operating year (280 days)
The UR adjusts annualized costs for the amount of time required to process 260,000 ssf,
determined from the simulation model for each process alternative. The components of capital
costs are discussed further below followed by sample calculations of capital costs.
Equipment and Installation Costs. Primary equipment and installation cost estimates
were provided by equipment suppliers and include delivery of equipment, installation, and sales
tax. Equipment estimates were based on basic, no frills equipment capable of processing the
modeled throughput rates determined by the simulation model, presented in Table 4-38.
Equipment estimates did not include auxiliary equipment, such as statistical process control or
automated sampling equipment sometimes found on surface finishing process lines.
Annual costs of the equipment (which includes installation) were determined assuming
5-year, straight-line depreciation and no salvage value. These annual costs were calculated using
the following equation:
E = equipment cost ($) + 5 years
Facility Costs. Facility costs are capital costs for the floor space required to operate the
surface finishing line. Facility costs were calculated assuming industrial floor space costs $76/ft2
and the facility is depreciated over 25 years using straight-line depreciation. The cost per square
foot of floor space applies to Class A light manufacturing buildings with basements. This value
was obtained from the Marshall Valuation Service (Vishanoff, 1995) and mean square foot costs
(Ferguson, 1996). Facility costs were calculated using the following equation:
F = [unit cost of facility utilized ($/ft2) x footprint area/process step (ft2/step) x number of steps]
+- 25 years
The "footprint area" is the area of floor space required by surface finishing equipment,
plus a buffer zone to maneuver equipment or have room to work on the surface finishing process
4-72
-------
line, and to maintain and repair it.2 The footprint area per process step was calculated by
determining the equipment dimensions of each process alternative, adjusting the dimensions for
working space, and then determining the area per process step. Because the footprint area
depends on the type of process automation, the average dimensions of both conveyorized (8' x
40') and non-conveyorized (5' x 23') equipment, irrespective of surface finish technology, were
determined from the PWB Workplace Practices Questionnaire data. Because these dimensions
account for the equipment only, an additional 8 ft was added to every dimension to allow space
for line operation, maintenance, and chemical handling. The footprint area required by either
equipment type, including the buffer zone, was calculated as 1,344 ft2 for convey orized processes
and 819 ft2 for non-conveyorized processes. The area required per process step was determined
by first identifying the process alternative with the fewest process steps for each automation type,
and then dividing the required floor space by that number of steps. This method conservatively
estimated the amount of floor space required per process step for conveyorized processes as 168
ft2/step, and for non-conveyorized processes as 91 ft2/step. The overall area required for each
process alternative was then calculated using the following equations:
Conveyorized:
Fc = [$76/ft2 x 168 ft2/step x number of steps per process] + 25 years
Non-conveyorized:
FN = [$76/ft2 x 91 ft2/step x number of steps per process] + 25 years
Example Capital Cost Calculations. This section presents example capital costs
calculations for the non-conveyorized nickel/gold process. From Figure 2-1, the non-
conveyorized nickel/gold process consists of 14 chemical bath and rinse steps. Simulation outputs
in Table 4-38 indicate this process takes 212 days to manufacture 260,000 ssf of PWB.
Equipment vendors estimated equipment and installation costs at a combined $48,000 (Harbor,
2000). The components of capital costs are calculated as follows:
E = $48,000 - 5 yrs = $9,600/yr
FN = ($76/ft2 x 91 ft2/step x 14 steps) - 25 yr = $3,870/yr
UR = 212 days - 280 days/yr = 0.757 yr
Thus, the capital costs for the non-conveyorized nickel/gold process to produce 260,000
ssf of PWB are as follows:
C = ($9,600/yr+$3,870/yr)x 0.757 yr = $10,200
Material Costs
Materials costs were calculated for the chemical products consumed during the operation
of the surface finishing process lines, through the initial setup and subsequent replacement of
2 PWB manufacturers and their suppliers use the term "footprint" to refer to the dimensions of process
equipment, such as the dimensions of the surface finishing process line.
4-73
-------
process chemical baths. The following presents equations for calculating materials costs, along
with some sample materials cost calculations.
Materials Cost Calculation Methods. Chemical suppliers were asked to provide
estimates of chemical costs ($/ssf), along with the other process data required by the project.
While some suppliers furnished estimates for one or more of their process alternatives, several
suppliers did not provide chemical cost estimates for any of their surface finishing process lines
being evaluated. Still others provided incomplete cost estimates, or did not provide any
supporting documentation of assumptions used to estimate chemical costs. Therefore, these data
could not be used in the comparative cost estimates. Instead, chemical costs were estimated using
the methods detailed below.
Chemical baths are typically made up of one or more separate chemical products mixed
together at specific concentrations to form a chemical solution. As PWBs are processed by the
surface finishing line, the chemical baths become contaminated or depleted and require chemical
additions or replacement. Baths are typically replaced according to analytical results or by
supplier recommended replacement criteria specific to each bath. When the criteria are met or
exceeded, the spent bath is removed and a new bath is created. The chemical cost to replace a
specific bath one time is the sum of the costs of each chemical product in the bath, and is given by
the following equation:
n
Cost per bath replacement = [chemical product I cost/bath ($/gal) x
1=1
% chemical product I in bath x total volume of bath (gal)]
where,
n = number of chemical products in a bath
Price quotes were obtained from chemical suppliers in $/gallon or $/lb for process
chemical products. Chemical compositions of the individual process baths were determined from
the corresponding Product Data Sheets submitted by the chemical suppliers of each process
alternative. The average volume of a chemical bath for non-conveyorized processes was
calculated to be 51.1 gallons from PWB Workplace Practices Questionnaire data. For
conveyorized processes, however, conveyor speed is constant; thus, the volume of chemicals in a
bath varies by bath type to provide the necessary contact time (see Table 4-33 for conveyorized
process bath volumes). These data were used in the above equation to calculate the chemical cost
per bath replacement for each product line. The bath replacement costs were then averaged
across like product lines (e.g., chemical costs from various suppliers of the OSP were averaged by
bath type) to determine an average chemical cost per replacement for each process bath.
To obtain the total materials cost, the chemical cost per bath replacement for each bath
was multiplied by the number of bath replacements required (determined by simulation) and then
summed over all the baths of an alternative process. The cost of chemical additions was not
included, because no data were available to determine the amount and frequency of chemical
additions. However, for process baths that are typically maintained rather than replaced (e.g.,
4-74
-------
baths with expensive metal ions such as tin, gold, silver and palladium), the replacement criteria
were adjusted to reflect the number of bath chemical turnovers that occur between bath
replacements, thereby accounting for the additional chemical usage. A complete change of bath
chemistry through bath maintenance, such as chemical additions, was considered one chemical
turnover. The number of chemical turnovers for each bath is represented on Table 4-41 as the
multiplying factor. Materials costs (M) are given by the following equation:
m
M = [chemical cost j /bath replacement ($) x number of replacements/bath]
where,
m = number of baths in a process
The frequency of replacement for individual process baths was determined using supplier
recommended criteria provided on Product Data Sheets and from PWB Workplace Practices
Questionnaire data. Simulation models were used to determine the number of times a bath would
be replaced during the production of 260,000 ssf of PWB by the surface finishing process.
Appendix G presents bath replacement criteria used in this analysis and summaries of chemical
product cost by supplier and by surface finishing technology.
Example Materials Cost Calculations. Table 4-42 presents an example of chemical
costs per bath replacement for one of the two nickel/gold product lines that were submitted by
chemical suppliers for evaluation. From the data in the table, the total cost of chemicals per bath
was calculated by multiplying the average chemical cost for a bath (calculated by computing the
chemical cost per bath of the second product line not shown in Table 4-42, then averaging the
costs for a bath from both product lines) by the number of bath replacements required to process
260,000 ssf, as determined by the process simulation. The costs for each bath were then summed
to give the total materials cost for the overall non-conveyorized nickel/gold product line. Data for
each of the product lines submitted, including the other electroless nickel/immersion gold product
line, are presented in Appendix G.
Table 4-41 presents the chemical cost per bath replacement, the number of bath
replacements required, as determined by simulation, the total chemical cost per bath, and the total
material cost for the non-conveyorized nickel/gold process. The chemical costs per process bath
for both product lines were averaged to determine the average chemical cost per bath for the non-
conveyorized nickel/gold process. Similar material cost calculations are presented in Appendix G
for each of the surface finishing process alternatives.
4-75
-------
Table 4-41. Materials Cost for the Non-Conveyorized Nickel/Gold Process
Bath
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Immersion Gold
Chemical Cost/Bath
Replacement a
$92.80
$386
$1,640
$315
$890
NAC
Number of Bath
Replacements b
7
9
6
4
40
6
Total Chemical
Cost
$649
$3,470
$9,830
$1,260
$35,500
$57,900
Total Materials Cost $108,600
a Reported data represents the chemical cost per bath replacement averaged from two nickel/gold product lines.
b Number of bath replacements required to process 260,000 ssf, as determined by process simulation.
0 The immersion gold replacement cost was calculated differently than the other baths because of the wide
disparity in costs and throughput between product lines. The overall cost for the gold bath was calculated for each
product line and then averaged together to give the gold chemical cost for the process.
Table 4-42. Chemical Cost per Bath Replacement for One Product Line of the
Non-Conveyorized Nickel/Gold Process
Bath
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Immersion Gold
Chemical
Product
A
B
C
D
E
F
G
H
I
J
K
L
M
N
Product
Cost a ($)
$25.0/gal
$5.66/gal
$9.39/gal
$27.3/kg
$1.20/gal
$127/gal
$54.0/gal
$51.0/gal
$29. I/kg
$24.1/gal
$30.9/gal
$28.4/gal
$21.4/gal
$40.0/g
Percentage of
Chemical Product b
10
3
3
45g/l
8.5
30
20
12
2g/l
6.6
15
6.6
50
3g/l
Multiplying
Factor c
1
1
1
1
1
1
1
1
1
6
6
5
1
3
Chemical Cost/Bath
Replacement d ($)
$128
$266
$2,810
$11.3
$2,390
$70,200
a Product cost from supplier of the chemical product.
b The percentage of a chemical product by volume in each process bath was determined from Product Data Sheets
provided by the supplier of the chemical product.
0 Multiplying factors reflect the number of chemical turnovers expected before the bath is replaced. A chemical
turnover is considered to be a complete change of bath chemistry through bath maintenance such as chemical
additions. Multiplying factors are used for baths that are typically maintained, rather than replaced.
d Cost per bath calculated assumes a bath volume of 51.1 gallons, as determined by PWB Workplace Practices
Questionnaire data for non-conveyorized processes.
4-76
-------
Utility Costs
Utility costs for the surface finishing process include water consumed by rinse tanks,3
electricity used to power the panel transportation system, heaters and other process equipment,
and natural gas consumed by drying ovens employed by some process alternatives. The following
example presents utility costs calculation methods and utility costs for the nickel/gold process.
Utility Cost Calculation Methods. The rate of water consumption depends on both the
number of distinct water rinse steps and the flow rate of the water used in those steps. The
typical number of water rinse steps for each surface finishing alternative was determined using
supplier- provided data together with data from the PWB Workplace Practices Questionnaire.
Based on Questionnaire data, the average normalized water flow rate per rinse stage for individual
rinse types was 0.176 gal/ssf for conveyorized processes, 0.258 gal/ssf for non-conveyorized
processes, and 0.465 gal/ssf for high pressure rinse tanks, regardless of automation type.
However, it was assumed that the rinse steps are shut off during periods of process down time.
Therefore, daily water consumption rates were adjusted for the percentage of time the process
was in operation. The total volume of water consumed was calculated by multiplying the number
of each type of rinse tank occurring in the process by the appropriate water flow rates for each
rinse. Water consumption rates for surface finishing technologies, along with a detailed
description of the methodology used to calculate them, are presented in Section 5.1, Resource
Conservation.
The cost of water was calculated by multiplying the water consumption rate of the entire
surface finishing process by the unit cost factor for water. A unit cost of $2.19/1,000 gallons of
water was obtained from the Pollution Prevention and Control Survey (U.S. EPA, 1998). The
equation for calculating water cost (W) is:
W = quantity of rinse water consumed (gal) x $2.19/1,000 gal
The rate of electricity consumption for each surface finishing alternative is dependent upon
the equipment required to operate each alternative. Differences in required process equipment,
such as the number of heaters, pumps, and type and extent of panel agitation, directly affect
electricity consumption. The cost of electricity is calculated by multiplying the electricity
consumption rate of the process alternative by the production time required to produce 260,000
ssf of PWB, and then applying a unit cost factor to the total. Electricity consumption rates for
surface finishing alternatives are presented in Section 5.2, Energy Impacts, while the required
production time was determined by the simulation model. A unit cost of $0.069/kW-hr was
obtained from the International Energy Agency. The energy cost (E) was calculated using the
following equation:
E = hourly consumption rate (kW) x required production time (hr) x $0.069/kW-hr
Natural gas is used to fire the drying ovens required by many of the surface finishing
processes. All processes with the exception of the nickel/gold and the nickel/palladium/gold
3 Water is also used in surface finishing chemical baths to dilute chemical products to the appropriate
concentration, but this use of water was assumed negligible compared to the water consumed in rinse tanks.
4-77
-------
processes required gas-fired ovens for panel drying. The amount of gas consumed was determined
by multiplying the natural gas consumption rate for the process alternative by the amount of
operating time required by the process to produce 260,000 ssf of PWB, and then applying a unit
cost to the result. Knoxville Utilities Board (KUB) charges $0.4028 per therm of natural gas
consumed (KUB, 2000a), while the production time required to produce 260,000 ssf of PWB
came from simulation results. Thus, the cost of natural gas consumption (G) was calculated by
the following equation:
G = natural gas consumption rate (therm/hr) x required production time (hrs) x $0.4028/therm
The total utility cost (U) for a surface finishing process was determined as follows:
U = W + E + G
where,
W = cost of water consumed ($/ssf) to produce 260,000 ssf
E = cost of electricity consumed ($/ssf) to produce 260,000 ssf
G = cost of natural gas consumed ($/ssf) to produce 260,000 ssf
Example Utility Cost Calculations. The above methodology was used to calculate the
utility costs for each of the surface finishing alternatives. This section presents example utility
cost calculations for the non-conveyorized nickel/gold process.
Simulation results indicate the non-conveyorized nickel/gold process is down 18.8 days
and takes 212 days overall (at 6.8 hrs/day) to produce 260,000 ssf. It is comprised of eight rinse
steps which consume approximately 537,000 gallons of water during the course of the job (see
Section 5.1, Resource Conservation). Electricity is consumed at a rate of 26.0 kW-hr of
operation (see Section 5.2, Energy Impacts). The non-conveyorized nickel/gold process has no
drying ovens and, therefore, does not consume natural gas. Based on this information, water,
electricity, and gas costs were calculated as follows:
W = 537,000 gallons x $2.19/1,000 gal = $1,180
E = 26.0 kW x (212 days - 18.8 days) x 6.8 hr/day x $.069/kW-hr = $2,360
G = $0
Thus, the utility cost for the process was determined by the calculation:
U = $1,180 +$2,360+ $0 = $3,540
4-78
-------
Wastewater Costs
Wastewater Cost Calculation Method. Wastewater costs for the surface finishing
processes were only determined for the cost of discharging wastewater to a POTW. The analysis
assumes that discharges are made in compliance with local allowable limits for chemical
concentrations and other parameters so that no fines are incurred.
Wastewater quantities were assumed equal to the quantity of rinse water used. Rinse
water usage was calculated in Section 5.1, Resource Conservation, and used to calculate water
costs in the Utility Costs section. The unit costs for fees charged by a POTW for both city and
non-city discharges of wastewater were obtained from KUB, and were averaged for use in
calculating wastewater cost (KUB, 2000b). These average unit costs are not flat rates applied to
the total wastewater discharge, but rather combine to form a tiered cost scale that applies an
incremental unit cost to each level of discharge. The tiered cost scale for wastewater discharges
to a POTW is presented in Table 4-43.
Wastewater Discharge
Quantity
(ccf/month)
0-2
3- 10
11-100
101 -400
401 -5,000
City Discharge
Cost
($/ccf/month)
$6.30
$2.92
$2.59
$2.22
$1.85
Non-City
Discharge Cost
($/ccf/month)
$7.40
$3.21
$2.85
$2.44
$2.05
Average Discharge
Cost
($/ccf/month)
$6.85
$3.06
$2.72
$2.33
$1.95
Source: KUB, 2000b.
ccf: 100 cubic ft.
The unit costs displayed for each level of discharge are applied incrementally to the
quantity of monthly discharge. For example, the first two cubic feet of wastewater discharged in
a month are assessed a charge of $6.85, while the next eight cubic feet cost $3.06, and so on. The
production time required to produce 260,000 ssf of PWB comes from the simulation models.
Thus, wastewater costs (WW) were calculated as follows:
WW = [quantity of discharge in tier k (ccf/mo) x tier cost factor ($/ccf)] x
k=l
required production time (months)
where,
P
ccf =
number of cost tiers
100 cubic ft
4-79
-------
Example Wastewater Cost Calculations. This section presents example wastewater
calculations for the non-conveyorized nickel/gold process. Based on rinse water usage, the total
wastewater release was approximately 537,000 gallons. The required production time in months
was calculated using the required production time from Table 4-38 and an operating year of 280
days (212 days + 280 days/year x 12 months/yr = 9.1 months). Thus, the monthly wastewater
release was 78.9 ccf (537,000 gallons ^9.1 months + 748 gal/ccf). To calculate the wastewater
cost for the non-conveyorized nickel/gold process, the tiered cost scale was applied to the
quantity of discharge, and the resulting costs per tier were summed, as follows:
$6.85 x 2 ccf/month = $13.70 ccf/month
$3.06 x 8 ccf/month = $24.48 ccf/month
$2.72 x 68.9 ccf/month = $187.40 ccf/month
Monthly discharge cost = $13.70 + $24.48 + $187.40 = $226/month
The monthly cost was then multiplied by the number of months required to produce
260,000 ssf of PWB to calculate the overall wastewater treatment cost:
WW = $226/monthx 9.1 month = $2,050
Production Costs
Production Cost Calculation Methods. Production costs for the surface finishing
process include both the cost of labor required to operate the process and the cost of transporting
chemicals to the production line from storage. Production costs (P) were calculated by the
following equation:
P = LA + TR
where,
LA = production labor cost ($/ssf) to produce 260,000 ssf
TR = chemical transportation cost ($/ssf) to produce 260,000 ssf
Production labor cost is a function of the number and type of employees and the length of
time required to complete a job. The calculation of production labor cost assumes that line
operators perform all of the daily activities, excluding bath maintenance, vital to the operation of
the surface finishing process. Labor costs associated with bath maintenance activities, such as
sampling and analysis, are presented in the discussion of maintenance costs, below. An average
number of line operators was determined for conveyorized (one line operator) from PWB
Workplace Practices Questionnaire data and site visit observations. Although no significant
difference in the number of line operators by automation type was reported in the data, the
number of line operators for non-conveyorized processes was adjusted upward to 1.1 to reflect
the greater level of labor content for these processes, as compared to convey orized processes.
The labor time required to complete the specified job was calculated assuming an average
shift time of eight hours per day, and using the number of days required to produce 260,000 ssf of
PWB from simulation results. A labor wage of $10.24/hr was obtained from the American Wages
4-80
-------
and Salary Survey (Fisher, 1999) and utilized for surface finishing line operators. Therefore, labor
costs for process alternatives were calculated as follows:
LA = number of operators x $10.24/hr x 8 hr/day x required production time (days)
The production cost category of chemical transportation cost includes the cost of
transporting chemicals from storage to the process line. A BOA, presented in Appendix G, was
developed and used to calculate the unit cost per chemical transport. Because chemicals are
consumed whenever a bath is replaced, the number of trips required to supply the process line
with chemicals equals the number of bath replacements required to produce 260,000 ssf of PWB.
Chemical transportation cost was calculated as follows:
TR = number of bath replacements x unit cost per chemical transport ($)
Example Production Cost Calculations. For the example of the non-conveyorized
nickel/gold, production labor cost was calculated assuming 1.1 operators working for 212 days
(see Table 4-38). Chemical transportation cost was calculated based on a cost per chemical
transport of $9.28 (see Table 4-40 and Appendix G) and 72 bath replacements (see Table 4-37).
Thus, the production cost was calculated as follows:
LA = 1.1 x $10.24 x 8 hr/day x 212 days = $19,100
TR = 72 x $9.28 = $668
thus,
P= $19,100+ $668 = $19,768
Maintenance Costs
Maintenance Cost Calculation Method. The maintenance costs for the surface finishing
process include the costs associated with tank cleaning, bath setup, sampling and analysis of bath
chemistries, and bath filter replacement. Maintenance costs were calculated as follows:
MA = TC + BS + FR+ST
where,
TC = tank cleanup cost ($/ssf) to produce 260,000 ssf
BS = bath setup cost ($/ssf) to produce 260,000 ssf
FR = filter replacement cost ($/ssf) to produce 260,000 ssf
ST = sampling cost ($/ssf) to produce 260,000 ssf
The maintenance costs listed above depend on the unit cost per repetition of the activity
and the number of times the activity was performed. For each maintenance cost category, a BOA
was developed to characterize the cost of labor, materials, and tools associated with a single
repetition of that activity. The BOA and unit cost per repetition for each cost category are
presented in Appendix G. It was assumed that the activities and costs characterized on the BOAs
are the same, regardless of the surface finishing process or process baths. Unit costs per
4-81
-------
repetition for both tank cleanup and bath setup were determined to be $67.00 and $15.10,
respectively.
The number of tank cleanups and bath setups equals the number of bath replacements
obtained from process simulation results (see Appendix G). Each time a bath is replaced, the tank
is cleaned before a replacement bath is created. The costs of tank cleanup and bath setup are thus
given by the following:
TC = number of tank cleanups x $67.00
BS = number of bath setups x $15.10
The PWB Workplace Practices Questionnaire data for both filter replacement and bath
sampling and analysis were reported in occurrences per year, instead of as a function of
throughput, and are represented in Section 3.2, the Exposure Assessment. These frequencies
were adjusted for this analysis using the URs for the production time required to manufacture
260,000 ssf of PWB. Using the unit costs determined by the BO As developed for filter
replacement ($17.50 per replacement), and bath sampling and testing ($3.70 per test), the costs
for these maintenance activities were calculated as follows:
ST = annual number of sampling & testing x UR x $3.70
FR = annual number of filter replacement x UR x $17.50
The total maintenance cost for each process alternative was determined by first calculating
the individual bath maintenance costs using the above equations and then summing the results for
all baths in that process.
Maintenance Costs Example Calculations. This section presents example maintenance
costs calculations for the non-conveyorized nickel/gold process. From Table 4-38, this process
has a production time of 212 days, which gives a UR of 0.76 (UR = 212 + 280). The number of
tank cleanups and bath setups equals the number of bath replacements reported in Table 4-37 (72
bath replacements). As reported in Section 3.2, Exposure Assessment, chemical baths are
sampled and tested 1,260 times per year, and filters are replaced 119 times per year. Thus, the
maintenance costs for the non-conveyorized nickel/gold process are:
TC = 72 x $67.00 = $4,820
BS = 72 x $15.10 = $1,090
ST = 1,260 x 0.76 x $3.70 = $3,530
FR = 119 x 0.76 x $17.50 = $1,580
Therefore, the overall maintenance cost for the process is:
MA = $4,820+ $1,090+ $3,530+ $1,580 = $11,000
4-82
-------
Determination of Total Cost and Unit Cost
The total cost for surface finishing process alternatives was calculated by summing the
totals of the individual costs categories. The cost per ssf of PWB produced, or unit cost (UC),
can then be calculated by dividing the total cost by the amount of PWBs produced. Table 4-44
summarizes the total cost of manufacturing 260,000 ssf of PWB using the non-conveyorized
nickel/gold process.
The UC for the non-conveyorized nickel/gold process was calculated as follows:
UC = total cost (TC) - 260,000 ssf
= $156,000-260,000 ssf
= $0.60/ssf
Table 4-44. Summary of Costs for the Non-Conveyorized Nickel/Gold Process
Cost Category
Capital Cost
Material Cost
Utility Cost
Wastewater Cost
Production Cost
Maintenance Cost
Component
Primary Equipment & Installation
Facility
Chemical(s)
Water
Electricity
Natural Gas
Wastewater Discharge
Transportation of Material
Labor for Line Operation
Tank Cleanup
Bath Setup
Sampling and Analysis
Filter Replacement
Component Cost a
$7,260
$2,930
$109,000
$1,180
$2,360
$0
$2,050
$668
$19,100
$4,820
$1,090
$3,530
$1,580
Total Cost
Totals a
$10,200
$109,000
$3,540
$2,050
$19,800
$11,000
$156,000
a Costs of producing 260,000 ssf of PWB by the process.
4.2.6 Results
Table 4-45 presents the costs for each of the surface finishing technologies. Table 4-46
presents unit costs ($/ssf). The total cost of producing 260,000 ssf ranged from a low of $26,300
for the conveyorized OSP process to a high of $399,000 for the non-conveyorized
nickel/palladium/gold process, with the corresponding unit costs ranging from $0.10/ssf to
$1.54/ssf for the same two processes. With the exception of the two technologies containing
gold, all of the other surface finishing alternatives were less expensive than the baseline, non-
conveyorized HASL process.
4-83
-------
Total cost data in Table 4-45 illustrate that chemical cost is the largest cost for all of the surface
finishing processes. Labor costs were the second largest cost component, though far smaller than
the cost of process chemicals.
Table 4-45. Total Cost of Surface Finishing Technologies
Cost Category
Capital Cost
Material Cost
Utility Cost
Wastewater Cost
Production Cost
Maintenance Cost
Cost Components
Primary Equipment & Installation
Facility
Chemical(s)
Water
Electricity
Natural Gas
Wastewater Discharge
Transportation of Material
Labor for Normal Production
Tank Cleanup
Bath Setup
Sampling and Testing
Filter Replacement
Total Cost
HASL,
NC
$9,360
$432
$74,800
$706
$669
$88
$1,100
$167
$3,940
$1,210
$272
$499
$967
$94,200
HASL,
C
$11,000
$398
$75,200
$565
$452
$45
$851
$130
$1,790
$938
$211
$249
$482
$92,400
Nickel/Gold,
NC
$7,260
$2,930
$109,000
$1,180
$2,360
$0
$2,050
$668
$19,100
$4,820
$1,090
$3,530
$1,580
$156,000
4-84
-------
Table 4-45. Total Cost of Surface Finishing Technologies
Cost Category
Capital Cost
Material Cost
Utility Cost
Wastewater Cost
Production Cost
Maintenance Cost
Cost Components
Primary Equipment & Installation
Facility
Chemical(s)
Water
Electricity
Natural Gas
Wastewater Discharge
Transportation of Material
Labor for Normal Production
Tank Cleanup
Bath Setup
Sampling and Testing
Filter Replacement
Total Cost
Nickel/Palladium/Gold,
NC
$15,400
$6,090
$321,000
$2,060
$4,050
$0
$3,530
$1,030
$25,200
$7,430
$1,680
$8,900
$2,840
$399,000
cont.)
OSP,
NC
$1,640
$320
$18,500
$441
$313
$66
$702
$159
$3,170
$1,140
$257
$1,610
$330
$28,700
OSP,
c
$2,880
$264
$18,800
$301
$208
$31
$463
$121
$1,320
$871
$196
$738
$151
$26,300
Table 4-45. Total Cost of Surface Finishing Technologies (cont.)
Cost Category
Capital Cost
Material Cost
Utility Cost
Wastewater Cost
Production Cost
Maintenance Cost
Cost Components
Primary Equipment & Installation
Facility
Chemical(s)
Water
Electricity
Natural Gas
Wastewater Discharge
Transportation of Material
Labor for Normal Production
Tank Cleanup
Bath Setup
Sampling and Testing
Filter Replacement
Total Cost
Immersion
Silver, C
$10,540
$937
$52,700
$301
$739
$140
$529
$167
$5,260
$1,210
$272
$937
$80
$73,800
Immersion
Tin, NC
$2,950
$892
$29,000
$1,030
$494
$162
$1,620
$204
$6,780
$1,470
$332
$1,260
$705
$46,900
Immersion
Tin, C
$16,800
$2,340
$28,900
$702
$1,230
$240
$1,220
$167
$8,770
$1,210
$272
$1,800
$1,000
$64,700
4-85
-------
Table 4-46. Surface Finishing Alternative Unit Costs for Producing 260,000 ssf of PWB
Surface Finishing Alternative
HASL, Non-conveyorized
HASL, Conveyorized
Nickel/Gold, Non-conveyorized
Nickel/Palladium/Gold, Non-conveyorized
OSP, Non-conveyorized
OSP, Conveyorized
Immersion Silver, Conveyorized
Immersion Tin, Non-conveyorized
Immersion Tin, Conveyorized
Total Cost
($)
94,200
92,400
156,000
399,000
28,700
26,300
73,800
46,900
64,700
Unit Cost
($/ssf)
0.36
0.35
0.60
1.54
0.11
0.10
0.28
0.18
0.25
Cost Savings a
(%)
—
3
-67
-327
69
72
22
50
31
a Cost savings measured by comparing cost of the surface finish to the cost of the baseline non-conveyorized HASL
process. Positive results represent percent savings from the costs incurred had the baseline process been used,
while negative results represent percent lost.
4.2.7 Conclusions
This analysis generated comparative costs for six surface finishing technologies, including
HASL, nickel/gold, nickel/palladium/gold, OSP, immersion silver, and immersion tin processes.
Costs were developed for each technology and equipment configuration for which data were
available from the PWB Workplace Practices Questionnaire and Performance Demonstration, for
a total of nine processes (five non-conveyorized processes and four Conveyorized processes).
Costs were estimated using a hybrid cost model that combines traditional costs with simulation
modeling and activity-based costs. The cost model was designed to determine the total cost of
processing a specific amount of PWBs through a fully operational surface finishing line, in this
case 260,000 ssf. The cost model does not estimate start-up costs for a facility switching to a
surface finishing alternative, which could factor significantly in the decision to implement a
technology. Total costs were divided by the throughput (260,000 ssf) to determine a unit cost in
$/ssf
The cost components considered include capital costs (primary equipment and installation,
and facility costs), materials costs (limited to chemical costs), utility costs (water, electricity, and
natural gas costs), wastewater costs (limited to wastewater discharge cost), production costs
(production labor and chemical transport costs), and maintenance costs (tank cleanup, bath setup,
sampling and analysis, and filter replacement costs). Other cost components may contribute
significantly to overall costs, but were not quantified because they could not be reliably estimated.
These include wastewater treatment cost, sludge recycling and disposal cost, other solid waste
disposal costs, and quality costs.
Overall, the costs ranged from $0.10/ssf for the Conveyorized OSP process to $1.54/ssf
for the non-conveyorized nickel/palladium/gold process. The cost of the baseline non-
conveyorized HASL process was calculated to be $0.36/ssf.
4-86
-------
Based on the results of this analysis, six of the eight alternative surface finishing processes
are more economical than the baseline non-conveyorized HASL process. Three processes had a
substantial cost savings of at least 50 percent of the cost per ssf over that of the baseline HASL
process (conveyorized OSP at 72 percent cost savings, non-conveyorized OSP at 69 percent, and
non-conveyorized immersion tin at 50 percent). Three other process alternatives realized a
somewhat smaller cost savings over the baseline HASL process (conveyorized immersion tin at 31
percent, conveyorized immersion silver at 22 percent, and the conveyorized HASL process at 3
percent.)
Two processes were more expensive than the baseline. The exceptions were the
electroless nickel/immersion gold process and the electroless nickel/palladium/immersion gold
process, both of which had chemical costs exceeding the entire cost of the non-conveyorized
HASL process, due to the precious metal content of the surface finish.
In general, conveyorized processes cost less than non-conveyorized processes of the same
technology due to the cost savings associated with their higher throughput rates. The exception
to this was immersion tin, which was more costly because the combination of process cycle time
and conveyor length resulted in a lower throughput rate than its non-conveyorized version.
Chemical cost was the single largest component cost for all of the nine processes. Labor
costs were the second largest cost component, though far smaller than the cost of process
chemicals.
4-87
-------
4.3 REGULATORY ASSESSMENT
This section describes the federal environmental regulations that may affect the use of
chemicals in the surface finishing processes during PWB manufacturing. Discharges of these
chemicals may be restricted by air, water, or solid waste regulations, and releases may be
reportable under the federal Toxic Release Inventory (TRI) program. This section discusses and
illustrates pertinent portions of federal environmental regulations that may be pertinent to surface
finishing operations, including the Clean Water Act (Table 4-47), the Clean Air Act (Table 4-53),
the Resource Conservation and Recovery Act, the Comprehensive Environmental Response,
Compensation, and Liability Act (Table 4-54), the Superfund Amendments and Reauthorization
Act and Emergency Planning and Community Right-To-Know Act (Table 4-55), and the Toxic
Substances Control Act (Table 4-56). This section is intended to provide an overview of
environmental regulations triggered by the use of the identified chemicals; it is not intended to be
used as regulatory guidance.
The primary sources of information for this section were the EPA Register of Lists (U.S.
EPA, 1998) and the EPA document, "Federal Environmental Regulations Affecting the
Electronics Industry" (U.S. EPA, 1995). The former is a database of federal regulations
applicable to specific chemicals that can be searched by chemical. The latter was prepared by the
DfE PWB Project. Of the 83 chemicals reportedly used in one or more of the evaluated surface
finishing technologies, no regulatory listings were found for 40 chemicals.
4.3.1 Clean Water Act
The Clean Water Act (CWA) is the basic federal law governing water pollution control in
the U.S. today. The various surface finishing processes used by the PWB industry produce a
number of pollutants that are regulated under the CWA. Applicable provisions, as related to
specific chemicals, are presented in Table 4-47; these particular provisions and process-based
regulations are discussed in greater detail below.
CWA Hazardous Substances and Reportable Quantities
Under Section 31 l(b)(2)(A) of the CWA, the Administrator designates hazardous
substances which, when discharged to navigable waters or adjoining shorelines, present an
imminent and substantial danger to the public health or welfare, including fish, shellfish, wildlife,
shorelines, and beaches. 40 Code of Federal Regulations (CFR) Part 117 establishes the
Reportable Quantity (RQ) for each substance listed in 40 CFR Part 116. When an amount equal
to or in excess of the RQ is discharged, the facility must provide notice to the federal government
of the discharge, following Department of Transportation requirements set forth in 33 CFR
Section 153.203. Liability for cleanup can result from such discharges. This requirement does
not apply to facilities that discharge the substance under a National Pollutant Discharge
Elimination System (NPDES) Permit or a CWA Section 404 dredge and fill permit, or to a
publicly owned treatment works (POTW), as long as any applicable effluent limitations or
pretreatment standards have been met. Table 4-47 lists RQs of hazardous substances under the
CWA that may apply to chemicals used in the surface finishing process.
4-S
-------
Table 4-47. CWA Regulations that May Apply to Chemicals in the
Surface Finishing Process
Chemical a
Acetic acid
Ammonium hydroxide
Copper ion
Copper sulfate pentahydrate
Ethylenediamine
Hydrochloric acid
Nickel sulfate
Nitric acid
Phosphoric acid
Propionic acid
Silver nitrate
Sodium hydroxide
Sulfuric acid
Urea
CWA 311 RQ
(Ibs)
5,000
1,000
5,000
5,000
100
1,000
5,000
5,000
1
1,000
1,000
CWA Priority
Pollutant
CWA 307a
CWA 304(b)
a In addition to the chemicals listed, there are 29 confidential business information (CBI) chemicals that would fall
under CWA regulations.
Abbreviations and Definitions:
CWA - Clean Water Act
CWA 304(b) - Effluent Limitations Guidelines
CWA 3 07 (a) - Toxic Pollutants Pretreatment Standards
CWA 311 - Hazardous Substances
RQ - Reportable Quantity
The NPDES permit program (40 CFR Part 122) contains regulations governing the
discharge of pollutants to waters of the U.S. Forty-three states and one territory are authorized
to administer NPDES programs that are at least as stringent as the federal program; EPA
administers the program in states or territories that are not authorized to do so, and on Native
American lands. The following discussion covers federal NPDES requirements. Facilities may be
required to comply with additional state requirements not covered in this document.
The NPDES program requires permits for the discharge of "pollutants" from any "point
source" into "navigable waters" (except those covered by Section 404 dredge and fill permits).
The CWA defines all of these terms broadly, and a source is required to obtain an NPDES permit
if it discharges almost anything other than dredge and fill material directly to surface water. A
source that sends its wastewater to a POTW is not required to obtain an NPDES permit, but may
be required to obtain an industrial user permit from the POTW to cover its discharge.
4-89
-------
CWA Priority Pollutant
In addition to other NPDES permit application requirements, facilities need to be aware of
priority pollutants listed in 40 CFR Part 122, Appendix D; this list of 126 compounds was
developed by EPA to define a specific list of chemicals to be given priority consideration in the
development of effluent limitation guidelines. Each PWB applicant for an NPDES permit must
provide quantitative data for those priority pollutants that the applicant knows or has reason to
believe, will be discharged in greater than trace amounts. Each applicant also must indicate if it
knows, or has reason to believe, it discharges any of the other hazardous substances or
non-conventional pollutants listed at 40 CFR Part 122, Appendix D. In some cases, quantitative
testing is required for these pollutants; in other cases, the applicant must describe why it expects
the pollutant to be discharged and provide the results of any quantitative data about its discharge
for that pollutant.
CWA Effluent Limitation Guidelines TCWA 30Ub). 304(b)1
A principal means for attaining water quality objectives under the CWA is the
establishment and enforcement of technology-based effluent limitations, which are based on the
pollutant control capabilities of available technologies, taking into consideration the economic
achievability of these limitations and a number of other factors. Because of differences in
production processes, quantities, and composition of discharges, separate national standards are
established for discharges associated with different industry categories. These standards are
referred to as technology-based effluent limitation guidelines.
The effluent limitation to be applied to a particular pollutant in a particular case depends
on the following:
whether the pollutant is conventional, non-conventional, or toxic;
whether the point source is a new or existing source; and
whether the point source discharges directly to the waters of the U.S. or to a POTW.
(Facilities that discharge to POTWs must comply with the pretreatment standards.)
Existing sources must comply with either best practicable control technology currently
available (BPT), best conventional pollution control technology (BCT), or best available control
technology economically achievable (BAT) standards. New facilities must comply with New
Source Performance Standards. NPDES permits also must contain any more stringent permit
limitations based on state water quality standards.
In the absence of effluent limitation guidelines for a facility category, permit writers
establish technology-based controls using their best professional judgement. In essence, the
permit writer undertakes an effluent guideline-type analysis for a single facility. The permit writer
will use information such as permit limits from similar facilities using similar treatment technology,
performance data from actual operating facilities, and scientific literature. Best Professional
Judgement may not be used in lieu of existing effluent guidelines. These guidelines apply only to
direct dischargers of wastewater.
4-90
-------
Pretreatment Standards
Only those facilities that discharge pollutants into waters of the U.S. need to obtain an
NPDES permit. Facilities that discharge to POTWs, however, must comply with pretreatment
requirements, as set out in Section 307(a) of CWA. These requirements were developed because
of concern that dischargers' waste containing toxic, hazardous, or concentrated conventional
industrial wastes might "pass through" POTWs, or that pollutants might interfere with the
successful operation of the POTW. EPA has established national, technology-based "categorical
pre-treatment standards" by facility category. In addition, or for industry categories without
national standards, POTWs may establish "local limits" or individual industrial facilities.
Wastewater emission standards for the PWB industry can be found at 40 CFR Part 413
and 433, which include the Pretreatment Standards for Existing Sources (PSES) and Pretreatment
Standards for New Sources (PSNS) that regulate PWB industry and wastewater, respectively.
The major constituents of PWB wastewater are heavy metals and other cations.
The PSES and the PSNS establish maximum concentration levels of several metals that
cannot be exceeded. They also regulate cyanide, which is used in some surface finishing
alternatives. Generally speaking, PSNS puts more stringent regulations on pollutants than PSES.
A summary of PSES for metals is included in Tables 4-48 and 4-49.
Table 4-48. Printed Circuit Board Facilities Discharging Less than 38,000 Liters per
Day PSES Limitations (mg/L)
Pollutant or Pollutant
Property
Cyanide (CN)
Lead (Pb)
Cadmium (Cd)
Max. Value for Any 1
Day (ppm)
5.0
0.6
1.2
Average Daily Values for 4 Consecutive Monitoring
Days that Shall Not be Exceeded mg/L (ppm)
2.7
0.4
0.7
Table 4-49. Printed Circuit Board Facilities Discharging 38,000 Liters per Day
or More PSES Limitations (mg/L)
Pollutant or Pollutant
Property
Copper (Cu)
Nickel (Ni)
Lead (Pb)
Cadmium (Cd)
Silver (Ag)
Total Metals
Cyanide (CN)
PH
Max. Value for Any 1
Day (ppm)
4.5
4.1
0.6
1.2
1.2
10.5
1.9
7.5
-------
Both 40 CFR Part 433.17, PSNS, and Part 433.16, New Source Performance Standards
(NSPS), have the same and more stringent regulated metal levels. Tables 4-50 and 4-51
summarize these sections.
Table 4-50. PSNS for Metal Finishing Facilities
Pollutant or Pollutant
Property
Copper (Cu)
Nickel (Ni)
Lead (Pb)
Cadmium (Cd)
Silver (Ag)
Cyanide (CN)
PH
Max. Value for Any 1
Day (ppm)
3.38
3.98
0.69
0.11
0.43
1.20
6.0
-------
4.3.2 Clean Air Act
The Clean Air Act (CAA), with its 1990 amendments, sets the framework for air pollution
control in the U.S. The various surface finishing processes produce a number of pollutants that
are regulated under the CAA. Applicable provisions, as related to specific chemicals, are
presented in Table 4-53; these particular provisions and process-based regulations are discussed
below.
Table 4-53. CAA Regulations that May Apply to Chemicals in the
Surface Finishing Process
Chemical a
Acetic acid
Ethylene glycol
Ethylenediamine
Hydrochloric acid
Malic acid
Nickel sulfate
Propionic acid
Sulfuric acid
CAA 111
CAA 112b
CAA 112r
a In addition to the chemicals listed, there are 16 CBI chemicals that have been identified as falling under the
CAA regulations discussed.
Abbreviations and Definitions:
CAA - Clean Air Act
CAA 111 - Standards of Performance for New Stationary Sources of Air Pollutants-Equipment Leaks Chemical
List
CAA 112b - Hazardous Air Pollutant
CAA 112r - Risk Management Program
Hazardous Air Pollutants
Section 112 of the CAA established a regulatory program for 188 hazardous air pollutants
and directed EPA to add other pollutants to the list, as needed. EPA is required to establish
Maximum Achievable Control Technology (MACT) standards for source categories that emit at
least one of the pollutants on the list in major quantities. Chemicals listed in Section 112 (b) of
the CAA that are used in surface finishing are shown in Table 4-53. EPA has identified categories
of industrial facilities that emit substantial quantities of any of these 188 pollutants and plan to
develop emissions limits for those industry categories between 1992 and 2000.
Section 112(r) of the CAA deals with sudden releases of, or accidents involving acutely
toxic, explosive, or flammable chemicals. This provision, added by the CAA Amendments of
1990, establishes a list of substances which, if present in a process in a quantity exceeding a
threshold, would require the facility to establish a Risk Management Program to prevent chemical
accidents. This program must include preparation of a risk management plan, which is submitted
to the state and local emergency planning organizations.
4-93
-------
Minimum Standards for State Operating Permit Programs
The CAA and its implementing regulations (at 40 CFR Part 70) define the minimum
standards and procedures required for state operating permit programs. The permit system is a
new approach established by the 1990 Amendments that is designed to define each source's
requirements and to facilitate enforcement. In addition, permit fees generate revenue to fund the
program's implementation.
Any facility defined as a "major source" is required to secure a permit. Section 70.2 of the
regulations defines a major source, in part, based upon if the source emits or has the potential to
emit:
10 tons per year (TPY) or more of any hazardous air pollutant;
25 TPY or more of any combination of hazardous air pollutants; or
100 TPY of any air pollutant.
For ozone non-attainment areas, major sources are defined as sources with the potential to
emit:
100 TPY or more of volatile organic compounds (VOCs) or oxides of nitrogen (NOx) in
areas classified as marginal or moderate;
50 TPY or more of VOCs or NOx in areas classified as serious;
25 TPY or more of VOCs or NOx in areas classified as severe; and
10 TPY or more of VOCs or NOx in areas classified as extreme.
Section 70.2 also defines certain other major sources in ozone transport regions and
serious non-attainment areas for carbon monoxide and particulate matter. In addition to major
sources, all sources that are required to undergo New Source Review, sources that are subject to
New Source Performance Standards or section 112 air toxics standards, and any affected source,
must obtain a permit.
By November 15, 1993, each state was required to submit an operating permit program to
EPA for approval. EPA was required to either approve or disapprove the state's program within
one year after submission. Once approved, the state program went into effect.
Major sources, as well as other sources identified above, were to submit their permit
applications to the state within one year of approval of the state program. Once a source submits
a timely and complete application, it may continue to operate until the permit is issued. Permit
issuance may take years because permit processing allows time for terms and conditions to be
reviewed by the public and neighboring states as well as by EPA.
When issued, the permit includes all federal air requirements applicable to the facility, such
as compliance schedules, emissions monitoring, emergency provisions, self-reporting
responsibilities, and emissions limitations. States may also choose to include state air
requirements in the permit. Five years is the maximum permit term.
4-94
-------
As established in 40 CFR Part 70, the states are required to develop fee schedules to
ensure the collection and retention of revenues sufficient to cover permit program costs. The
CAA has set a presumptive minimum annual fee of $25 per ton for all regulated pollutants
(except carbon monoxide), indexed for inflation, but states may set higher or lower fees as long as
they collect sufficient revenues to cover program costs.
4.3.3 Resource Conservation and Recovery Act
One purpose of the Resource Conservation and Recovery Act (RCRA) of 1976 (as
amended in 1984) is to set up a cradle-to-grave system for tracking and regulating hazardous
waste. EPA has issued regulations, found in 40 CFR Parts 260-299, which implement the federal
statute. These regulations are Federal requirements. Currently, 47 states have been authorized to
implement the basic RCRA program and may include more stringent requirements in their
authorized RCRA programs. In addition, non-RCRA-authorized states (Alaska, Hawaii, and
Iowa) may have state laws establishing hazardous waste management requirements. A facility
always should check with its state when analyzing which requirements apply to its activities.
To be an RCRA "hazardous waste", a material must first be a solid waste, which is
defined broadly under RCRA and RCRA regulations. Assuming the material is a solid waste, the
first evaluation to be made is whether or not it is also considered a hazardous waste. 40 CFR Part
261 addresses the identification and listing of hazardous waste. Waste generators are responsible
for determining whether a waste is hazardous, and what classification, if any, may apply to the
waste. Generators must undertake testing, or use their own knowledge and familiarity with the
waste, to determine if it is hazardous. Generators may be subject to enforcement penalties for
improperly determining that a waste is not hazardous.
RCRA Hazardous Waste Codes
Wastes can be classified as hazardous either because they are listed by EPA through
regulation in 40 CFR Part 261, or because they exhibit certain characteristics; namely toxicity,
corrosivity, reactivity, and ignitability. Listed hazardous wastes are specifically named (e.g.,
discarded commercial toluene, spent non-halogenated solvents). Characteristic hazardous wastes
are solid waste which "fail" a characteristic test, such as the RCRA test for ignitability.
There are four separate lists of hazardous wastes in 40 CFR Part 261. If any waste from a
PWB facility is on any of these lists, the facility is subject to regulation under RCRA (there are
two CBI chemicals used in a surface finishing process that have been identified as "U" listed
wastes). The listing is often defined by industrial processes, but all wastes are listed because they
were determined to be hazardous (these hazardous constituents are listed in Appendix VII to Part
261). Section 261.31 lists wastes from non-specific sources and includes wastes generated by
industrial processes that may occur in several different industries; the codes for such wastes
always begin with the letter "F." The second category of listed wastes (40 CFR Section 261.32)
includes hazardous wastes from specific sources; these wastes have codes that begin with the
letter "K." The remaining lists (40 CFR Section 261.33) cover commercial chemical products
that have been or are intended to be discarded; these have two letter designations, "P" and "U."
4-95
-------
Waste codes beginning with "P" are considered acutely hazardous, while those beginning with
"U" are simply considered hazardous.
Generator Status
A hazardous waste generator is defined as any person, by site, who creates a hazardous
waste or makes a waste subject to RCRA Subtitle C. Generators are divided into three
categories:
1. Large Quantity Generators—facilities that generate at least 1,000 kg (approximately 2,200
Ibs) of hazardous waste per month, or greater than 1 kg (2.2 Ibs) of acutely hazardous
waste per month.
2. Small Quantity Generators—facilities that generate greater than 100 kg (approximately
220 Ibs) but less than 1,000 kg of hazardous waste per month, and up to 1 kg (2.2 Ibs) per
month of acutely hazardous waste.
3. Conditionally Exempt Small Quantity Generators—facilities that generate no more than
100 kg (approximately 220 Ibs) per month of hazardous waste and up to 1 kg (2.2 Ibs) per
month of acutely hazardous waste.
Large and small quantity generators must meet many similar requirements. 40 CFR Part
262 provides that small quality generators may accumulate up to 6,000 kg of hazardous waste
on-site, at any one time, for up to 180 days without being regulated as a treatment, storage, or
disposal facility (TSDF), which requires a TSDF permit. The provisions of 40 CFR 262.34(f)
allow small quality generators to accumulate waste on-site for 270 days without having to apply
for TSDF status, provided the waste must be transported over 200 miles. Large quantity
generators have only a 90-day window to ship wastes off-site without needing a RCRA TSDF
permit. Keep in mind that most provisions of 40 CFR Parts 264 and 265 (for hazardous waste
treatment, storage and disposal facilities) do not apply to generators who send their wastes
off-site within the 90- or 180-day window, whichever is applicable.
Hazardous waste generators that do not meet the conditions for being conditionally
exempt, small quantity generators must (among other requirements such as record keeping and
reporting):
obtain a generator identification number;
accumulate and ship hazardous waste in suitable containers or tanks (for accumulation
only);
manifest the waste properly;
maintain copies of the manifest, a shipment log covering all hazardous waste shipments,
and test records;
comply with applicable land disposal restriction requirements; and
report releases or threats of releases of hazardous waste.
4-96
-------
TSDF Status
As mentioned above, Subtitle C of RCRA (40 CFR Parts 264 and 265) establishes
substantive permit requirements for facilities that treat, store, or dispose of hazardous wastes.
Generators (unless exempt, e.g., through the conditionally exempt, small quantity generators
exemption [see 40 CFR Part 261.5(g)]), no matter what monthly waste output, with waste on
site, for more than 90 days are classified as TSDFs. TSDFs must comply with 40 CFR Part
264-267 and Part 270, including permit requirements and stringent technical and financial
responsibility requirements. Generators who discharge hazardous waste into a POTW, or from a
point source regulated by an NPDES permit, are not required to comply with TSDF regulations.
4.3.4 Comprehensive Environmental Response, Compensation and Liability Act
The Comprehensive Environmental Response, Compensation and Liability Act (also
known as CERCLA, or more commonly as "Superfund") was enacted in 1980. CERCLA is the
Act that created the Superfund hazardous substance cleanup program and set up a variety of
mechanisms to address risks to public health, welfare, and the environment caused by hazardous
substance releases.
CERCLA ROs
Substances defined as hazardous under CERCLA are listed in 40 CFR Section 302.4.
Under CERCLA, EPA has assigned a RQ to most hazardous substances; regulatory RQs are
either 1, 10, 100, 1,000, or 5,000 pounds (except for radionuclides). If EPA has not assigned a
regulatory RQ to a hazardous substance, typically its RQ is one pound (Section 102). Any person
in charge of a facility (or vessel) must immediately notify the National Response Center as soon as
a person has knowledge of a hazardous substance release in an amount that is equal to or greater
than its RQ. There are some exceptions to this requirement, including the exceptions for federally
permitted releases. There is also streamlined reporting for certain continuous releases (see 40
CFR 302.8). Table 4-54 lists RQs of substances under CERCLA that may apply to chemicals
used in surface finishing processes.
Table 4-54. CERCLA RQs that May Apply to Chemicals in the Surface Finishing Process
Chemical a
Acetic acid
Ammonium hydroxide
Copper ion
Ethylene glycol
Ethylenediamine
Hydrochloric acid
Nickel sulfate
CERCLA RQ (Ibs)
5,000
1,000
1
5,000
5,000
5,000
100
Chemical a
Phosphoric acid
Propionic acid
Silver nitrate
Sodium hydroxide
Sulfuric acid
Thiourea
CERCLA RQ (Ibs)
5,000
5,000
1
1,000
1,000
10
a In addition to the chemicals listed, there are 17 CBI chemicals with reportable quantities under CERCLA.
Abbreviations and Definitions:
CERCLA - Comprehensive Environmental Response, Compensation and Liability Act
RQ - Reportable Quantity
4-97
-------
CERCLA Liability
CERCLA further makes a broad class of parties liable for the costs of removal or
remediation of the release, or threatened release, of any hazardous substance at a facility. Section
107 specifies the parties liable for response costs, including the following: 1) current owners and
operators of the facility; 2) owners and operators of a facility at the time hazardous substances
were disposed; 3) persons who arranged for disposal, treatment, or transportation for disposal or
treatment of such substances; and 4) persons who accepted such substances for transportation for
disposal or treatment. These parties are liable for: 1) all costs of removal or remedial action
incurred by the federal government, a state, or an Indian tribe not inconsistent with the National
Contingency Plan (NCP); 2) any other necessary costs of response incurred by any person
consistent with the NCP; 3) damages for injury to natural resources; and d) costs of health
assessments.
4.3.5 Superfund Amendments and Reauthorization Act and Emergency Planning
and Community Right-To-Know Act
CERCLA was amended in 1986 by the Superfund Amendments and Reauthorization Act
(SARA). Title III of SARA is also known as the Emergency Planning and Community
Right-To-Know Act (EPCRA). Certain sections of SARA and EPCRA may be applicable to
surface finishing chemicals and PWB manufacturers. Table 4-55 lists applicable provisions as
related to specific chemicals.
SARA Priority Contaminants
SARA Section 110 addresses Superfund site priority contaminants. This list contains the
275 highest-ranking substances of the approximately 700 prioritized substances. These chemical
substances, found at Superfund sites, are prioritized based on their frequency of occurrence,
toxicity rating, and potential human exposure. Once a substance has been listed, the Agency for
Toxic Substances and Disease Registry must develop a toxicological profile containing general
health/hazard assessments with effect levels, potential exposures, uses, regulatory actions, and
further research needs.
EPCRA Extremely Hazardous Substances
Section 302(a) of EPCRA regulates extremely hazardous substances and is intended to
facilitate emergency planning for response to sudden toxic chemical releases. Facilities must
notify the State Emergency Response Commission (SERC) if these chemicals are present in
quantities greater than their threshold planning quantities. These same substances also are subject
to regulation under EPCRA Section 304, which requires accidental releases in excess of
reportable quantities to be reported to the SERC and Local Emergency Planning Committee.
4-98
-------
EPCRA Toxic Release Inventory
Under EPCRA Section 313, a facility in a covered Standard Industrial Code (SIC), that
has 10 or more full-time employees, or the equivalent, and that manufactures, processes, or
otherwise uses a toxic chemical listed in 40 CFR Section 372.65 above the applicable reporting
threshold, must either file a toxic chemical release inventory reporting form (EPA Form R)
covering release and other waste management activities, or if applicable, an annual certification
statement (EPA Form A). The activity thresholds are 25,000 pounds per year for manufacturing
(including importing) and processing, and 10,000 pounds per year for the otherwise use of a listed
toxic chemical. Facilities that do not manufacture, process, or otherwise use more than one
million pounds of a toxic chemical, and have a total annual reportable amount of no greater than
500 pounds for the chemical, may utilize the briefer Form A certification statement. The Form R,
or form A if applicable, must be filed with the EPA and a state agency where the facility is
located. Beginning in the 1991 reporting year, facilities must also report pollution prevention and
recycling data for TRI chemicals on Form R pursuant to Section 6607 of the Pollution Prevention
Act, 42 U.S.C. 13106. Table 4-55 lists chemicals used in surface finishing processes that are
listed in the TRI.
Table 4-55. SARA and EPCRA Regulations that May Apply to Chemicals in the Surface
Finishing Process
Chemical a
Ammonium hydroxide
Copper ion
Copper sulfate pentahydrate
Ethylene glycol
Ethylenediamine
Nickel sulfate
Palladium chloride
Phosphoric acid
Sulfuric acid
SARA 110
EPCRA 302a
EPCRA 313
a In addition to the chemicals listed, there are 14 CBI chemicals identified as falling under the SARA and EPCRA
regulations discussed.
Abbreviations and Definitions:
SARA - Superfund Amendments and Reauthorization Act
SARA 110 - Superfund Site Priority Contaminant
EPCRA - Emergency Planning & Community Right-To-Know Act
EPCRA 302a - Extremely Hazardous Substances
EPCRA 313 - Toxic Chemical Release Inventory
4-99
-------
4.3.6 Toxic Substances Control Act
The Toxic Substances Control Act (TSCA), 15 U.S.C. Sections 2601-2692 (Regulations
found at 40 CFR part 700-799), originally passed in 1976 and subsequently amended, applies to
the manufacturers, importers, processors, distributors, users, and disposers of chemical substances
or mixtures. Table 4-56 lists TSCA regulations and testing lists that may be pertinent to surface
finishing processes.
Table 4-56. TSCA Regulations and Lists that May Apply to Chemicals Used in
Surface Finishing Processes
Chemical a
Ethylene glycol
Palladium chloride
TSCA 8d HSDR
TSCA MTL
TSCA 8a PAIR
a In addition to the chemicals listed, there are 10 CBI chemicals identified as falling under the TSCA regulations
discussed.
Abbreviations and Definitions:
TSCA - Toxic Substances Control Act
TSCA 8d HSDR - Health & Safety Data Reporting Rules
TSCA MTL - Master Testing List
TSCA 8a PAIR - Preliminary Assessment Information Rule
Testing Requirements
Section 4 authorizes EPA to require the testing of any chemical substance or mixture for
potential adverse health and environmental effects. On finding that such testing is necessary, due
to insufficient data from which the chemical's effects can be predicted, and that either: 1)
activities involving the chemical may present an unreasonable risk of injury to health or the
environment; or 2) the chemical is produced in substantial quantities and enters the environment in
substantial quantities, or there is significant or substantial human or environmental exposure to the
chemical.
The TSCA Master Testing List (MTL) is a list compiled by the EPA Office of Pollution
Prevention and Toxics to set the Agency's testing agenda. The major purposes are to: 1) identify
chemical testing needs; 2) focus limited EPA resources on those chemicals with the highest
priority testing needs; 3) publicize EPA's testing priorities for industrial chemicals; 4) obtain
broad public comments on EPA's testing program and priorities; and 5) encourage initiatives by
industry to help EPA meet those priority needs. The 1996 MTL now contains over 500 specific
chemicals in 10 categories.
4-100
-------
Unpublished Health and Safety Data Reporting Requirements
Under section 8(d) of TSCA, EPA has promulgated regulations that require that any
person who manufactures, imports, or, in some cases, processes (or proposes to manufacture,
import, or, in some cases, process) a chemical substance or mixture identified under 40 CFR part
716, must submit to EPA copies of unpublished health and safety studies with respect to that
substance or mixture.
Preliminary Assessment Information Rule
Under section 8(a) of TSCA, EPA has promulgated regulations at 40 CFR part 712,
Subpart B (the Preliminary Assessment Information Rule (PAIR), which establishes procedures
for chemical manufacturers and importers to report production, use, and exposure-related
information on listed chemical substances. Any person (except a small manufacturer or importer)
who imports or manufactures chemicals identified by EPA in this rule, must report information on
production volume, environmental releases, and certain other releases. Small manufacturers or
importers may be required to report such information on certain chemicals.
4.3.7 Summary of Regulations for Surface Finishing Technologies
Tables 4-57 through 4-62 provide a summary of regulations that may apply to chemicals in
each of the surface finishing technology categories.
4-101
-------
Table 4-57. Summary of Regulations that May Apply to Chemicals Used in Hot Air Solder Leveling (HASL) Technology
Chemicals Subject to Applicable Regulation
Process Chemical
Copper sulfate pentahydrate
Ethylene glycol
Hydrochloric acid
Hydrogen peroxide
Phosphoric acid
Sodium hydroxide
Sulfuric acid
CBI chemicals (13)
CWA
304b
/
307a
/
311
/
/
/
/
Priority
Poll.
/
CAA
111
/
/
1
112b
/
/
1
112r
/
EPCRA
313
/
/
/
/
2
302a
/
/
/
SARA
110
/
TSCA
8d
HSDR
3
MTL
/
/
/
1
8a
PAIR
3
RCRA Waste
P
U
Note: For technologies with more than one process submitted for evaluation (i.e., nickel/gold, OSP, immersion tin), the number of listed chemicals subject
to regulation reflects the total number of chemicals for both processed.
Abbreviations and definitions:
CAA - Clean Air Act
CAA 111 - Standards of Performance for New Stationary Sources of
Air Pollutants-Equipment Leaks Chemical List
CAA 112b - Hazardous Air Pollutant
CAA 112r - Risk Management Program
CWA - Clean Water Act
CWA 304b - Effluent Limitations Guidelines
CWA 307a - Toxic Pollutants
CWA 311 - Hazardous Substances
CWA Priority Pollutants
EPCRA - Emergency Planning and Community Right-to-Know Act
EPCRA 302a - Extremely Hazardous Substances
EPCRA 313 - Toxic Chemical Release Inventory
RCRA - Resource Conservation and Recovery Act
RCRA P Waste - Listed acutely hazardous waste
RCRA U Waste - Listed hazardous waste
SARA - Superfund Amendments and Reauthorization Act
SARA 110 - Superfund Site Priority Contaminant
TSCA - Toxic Substances Control Act
TSCA 8d HSDR - Health & safety data reporting rules
TSCA MTL - Master Testing List
TSCA 8a PAIR - Preliminary Assessment Information Rule
4-102
-------
Table 4-58. Summary of Regulations that May Apply to Chemicals Used in Nickel/Gold Technology
Chemicals Subject to Applicable Regulation
Process Chemical
Ammonium hydroxide
Copper sulfate pentahydrate
Hydrochloric acid
Hydrogen peroxide
Malic acid
Nickel sulfate
Palladium chloride
Phosphoric acid
Sodium hydroxide
Sulfuric acid
CBI Chemicals (19)
CWA
304b
/
/
4
307a
/
/
4
311
/
/
/
/
/
/
10
Priority
Poll.
/
/
4
CAA
111
/
/
9
112b
/
/
4
112r
/
SARA
110
/
/
/
4
EPCRA
313
/
/
/
/
/
7
302a
/
/
/
TSCA
8d
HSD
R
1
MTL
/
/
2
8a
PAIR
/
2
RCRA Waste
P
U
Note: For technologies with more than one process submitted for evaluation (i.e., nickel/gold, OSP, immersion tin), the number of listed chemicals subject
to regulation reflects the total number of chemicals for both processed.
Abbreviations and definitions:
CAA - Clean Air Act
CAA 111 - Standards of Performance for New Stationary Sources of
Air Pollutants-Equipment Leaks Chemical List
CAA 112b - Hazardous Air Pollutant
CAA 112r - Risk Management Program
CWA - Clean Water Act
CWA 304b - Effluent Limitations Guidelines
CWA 307a - Toxic Pollutants
CWA 311 - Hazardous Substances
CWA Priority Pollutants
EPCRA - Emergency Planning and Community Right-to-Know Act
EPCRA 302a - Extremely Hazardous Substances
EPCRA 313 - Toxic Chemical Release Inventory
RCRA - Resource Conservation and Recovery Act
RCRA P Waste - Listed acutely hazardous waste
RCRA U Waste - Listed hazardous waste
SARA - Superfund Amendments and Reauthorization Act
SARA 110 - Superfund Site Priority Contaminant
TSCA - Toxic Substances Control Act
TSCA 8d HSDR - Health & safety data reporting rules
TSCA MTL - Master Testing List
TSCA 8a PAIR - Preliminary Assessment Information Rule
4-103
-------
Table 4-59. Summary of Regulations that May Apply to Chemicals Used in Nickel/Palladium/Gold Technology
Chemicals Subject to Applicable Regulation
Process Chemical
Ammonium hydroxide
Copper sulfate pentahydrate
Ethylenediamine
Hydrochloric acid
Hydrogen peroxide
Nickel sulfate
Palladium chloride
Phosphoric acid
Propionic acid
Sodium chloride
Sodium hydroxide
Sulfuric acid
CBI Chemicals (20)
CWA
304b
/
/
3
307a
/
/
3
311
/
/
/
/
/
/
6
Priority
Poll.
/
/
3
CAA
111
/
4
112b
/
/
3
112r
/
SARA
110
/
/
/
3
EPCRA
313
/
/
/
/
/
5
302a
/
/
/
TSCA
8d
HSDR
1
MTL
/
/
2
8a
PAIR
/
3
RCRA Waste
P
U
Note: For technologies with more than one process submitted for evaluation (i.e.
to regulation reflects the total number of chemicals for both processed.
Abbreviations and definitions:
CAA-Clean Air Act
CAA 111 - Standards of Performance for New Stationary Sources of
Air Pollutants-Equipment Leaks Chemical List
CAA 112b - Hazardous Air Pollutant
CAA 112r - Risk Management Program
CWA - Clean Water Act
CWA 304b - Effluent Limitations Guidelines
CWA 307a - Toxic Pollutants
CWA 311 - Hazardous Substances
CWA Priority Pollutants
EPCRA - Emergency Planning and Community Right-to-Know Act
EPCRA 302a - Extremely Hazardous Substances
EPCRA 313 - Toxic Chemical Release Inventory
nickel/gold, OSP, immersion tin), the number of listed chemicals subject
RCRA - Resource Conservation and Recovery Act
RCRA P Waste - Listed acutely hazardous waste
RCRA U Waste - Listed hazardous waste
SARA - Superfund Amendments and Reauthorization Act
SARA 110 - Superfund Site Priority Contaminant
TSCA - Toxic Substances Control Act
TSCA 8d HSDR - Health & safety data reporting rules
TSCA MTL - Master Testing List
TSCA 8a PAIR - Preliminary Assessment Information Rule
4-104
-------
Table 4-60. Summary of Regulations that May Apply to Chemicals Used in OSP Technology
Chemicals Subject to Applicable Regulation
Process Chemical
Acetic acid
Copper ion
Copper sulfate
pentahydrate
Ethylene glycol
Hydrogen peroxide
Hydrochloric acid
Phosphoric acid
Sodium hydroxide
Sulfuric acid
CBI Chemicals (9)
CWA
304b
/
/
307a
/
/
311
/
/
/
/
/
Priority
Poll.
/
/
CAA
111
/
/
/
112b
/
/
112r
/
SARA
110
/
/
EPCRA
313
/
/
/
/
1
302a
/
/
TSCA
8d
HSD
R
1
MTL
/
/
8a
PAIR
1
RCRA Waste
P
U
Note: For technologies with more than one process submitted for evaluation (i.e., nickel/gold, OSP, immersion tin), the number of listed chemicals subject
to regulation reflects the total number of chemicals for both processed.
Abbreviations and definitions:
CAA-Clean Air Act
CAA 111 - Standards of Performance for New Stationary Sources of
Air Pollutants-Equipment Leaks Chemical List
CAA 112b - Hazardous Air Pollutant
CAA 112r - Risk Management Program
CWA - Clean Water Act
CWA 304b - Effluent Limitations Guidelines
CWA 307a - Toxic Pollutants
CWA 311 - Hazardous Substances
CWA Priority Pollutants
EPCRA - Emergency Planning and Community Right-to-Know Act
EPCRA 302a - Extremely Hazardous Substances
EPCRA 313 - Toxic Chemical Release Inventory
RCRA - Resource Conservation and Recovery Act
RCRA P Waste - Listed acutely hazardous waste
RCRA U Waste - Listed hazardous waste
SARA - Superfund Amendments and Reauthorization Act
SARA 110 - Superfund Site Priority Contaminant
TSCA - Toxic Substances Control Act
TSCA 8d HSDR - Health & safety data reporting rules
TSCA MTL - Master Testing List
TSCA 8a PAIR - Preliminary Assessment Information Rule
4-105
-------
Table 4-61. Summary of Regulations that May Apply to Chemicals Used in Immersion Silver Technology
Chemicals Subject to Applicable Regulation
Process Chemical
Hydrogen peroxide
Phosphoric acid
Silver nitrate
Sodium hydroxide
Sulfuric acid
CBI chemicals (5)
CWA
304b
X
307a
X
311
X
X
X
X
1
Priority
Poll.
X
CAA
111
X
112b
1
112r
SARA
110
X
EPCRA
313
X
X
1
302a
X
X
1
TSCA
8d
HSDR
MTL
X
8a
PAIR
X
RCRA Waste
P
U
Note: For technologies with more than one process submitted for evaluation (i.e., nickel/gold, OSP, immersion tin), the number of listed chemicals subject to
regulation reflects the total number of chemicals for both processed.
Abbreviations and definitions:
CAA - Clean Air Act
CAA 111 - Standards of Performance for New Stationary Sources of
Air Pollutants-Equipment Leaks Chemical List
CAA 112b - Hazardous Air Pollutant
CAA 112r - Risk Management Program
CWA-Clean Water Act
CWA 304b - Effluent Limitations Guidelines
CWA 307a - Toxic Pollutants
CWA 311 - Hazardous Substances
CWA Priority Pollutants
EPCRA - Emergency Planning and Community Right-to-Know Act
EPCRA 302a - Extremely Hazardous Substances
EPCRA 313 - Toxic Chemical Release Inventory
RCRA - Resource Conservation and Recovery Act
RCRA P Waste - Listed acutely hazardous waste
RCRA U Waste - Listed hazardous waste
SARA - Superfund Amendments and Reauthorization Act
SARA 110 - Superfund Site Priority Contaminant
TSCA - Toxic Substances Control Act
TSCA 8d HSDR - Health & safety data reporting rules
TSCA MTL - Master Testing List
TSCA 8a PAIR - Preliminary Assessment Information Rule
4-106
-------
Table 4-62. Summary of Regulations that May Apply to Chemicals Used in Immersion Tin Technology
Chemicals Subject to Applicable Regulation
Process Chemical
Hydrochloric acid
Phosphoric acid
Silver nitrate
Sulfuric acid
Thiourea
Urea
CBI Chemicals (16)
CWA
304b
/
307a
/
311
/
/
/
/
2
Priority
Poll.
/
CAA
111
/
2
112b
/
1
112r
/
SARA
110
/
EPCRA
313
/
/
/
/
3
302a
/
/
TSCA
8d
HSDR
2
MTL
/
3
8a
PAIR
/
2
RCRA Waste
P
U
U219
1
Note: For technologies with more than one process submitted for evaluation (i.e., nickel/gold, OSP, immersion tin), the number of listed chemicals subject
to regulation reflects the total number of chemicals for both processed.
Abbreviations and definitions:
CAA - Clean Air Act
CAA 111 - Standards of Performance for New Stationary Sources of
Air Pollutants-Equipment Leaks Chemical List
CAA 112b - Hazardous Air Pollutant
CAA 112r - Risk Management Program
CWA - Clean Water Act
CWA 304b - Effluent Limitations Guidelines
CWA 307a - Toxic Pollutants
CWA 311 - Hazardous Substances
CWA Priority Pollutants
EPCRA - Emergency Planning and Community Right-to-Know Act
EPCRA 302a - Extremely Hazardous Substances
EPCRA 313 - Toxic Chemical Release Inventory
RCRA - Resource Conservation and Recovery Act
RCRA P Waste - Listed acutely hazardous waste
RCRA U Waste - Listed hazardous waste
SARA - Superfund Amendments and Reauthorization Act
SARA 110 - Superfund Site Priority Contaminant
TSCA - Toxic Substances Control Act
TSCA 8d HSDR - Health & safety data reporting rules
TSCA MTL - Master Testing List
TSCA 8a PAIR - Preliminary Assessment Information Rule
4-107
-------
REFERENCES
Badgett, Lona, Beth Hawke and Karen Humphrey. 1995. Analysis of Pollution Prevention and
Waste Minimization Opportunities Using Total Cost Assessment: A Study in the Electronic
Industry. Seattle, WA. Pacific Northwest Pollution Prevention Research Center Publication.
Ferguson, John H. 1996. Mean Square Foot Costs: Means-Southern Construction
Information Network. Kingston, MA: R.S. Means, Co., Inc. Construction. Publishers and
Consultants.
Fisher, Helen S. 1999. American Wages and Salary Survey., 3rd Ed. Detroit, MI: Gale Research
Inc. (An International Thompson Publishing Co.)
Iman, R.L. et al. 1995. "Evaluation of Low-Residue Soldering for Military and Commercial
Applications: A Report from the Low-Residue Soldering Task Force." June.
Iman, R.L., J.F. Koon, et al. 1997. "Screening Test Results for Developing Guidelines for
Conformal Coating Usage and for Evaluating Alternative Surface Finishes." CCAMTF Report.
June.
Iman, R.L., J. Fry, R. Ragan, J.F. Koon and J. Bradford. 1998. "A Gauge Repeatability and
Reproducibility Study for the CCAMTF Automated Test Set." CCAMTF Report. March.
Joint Group on Acquisition Pollution Prevention (JG-APP) Joint Test Protocol CC-P-1-1 for
Validation of alternatives to Lead-Containing Surface Finishes, for Development of Guidelines
for Conformal Coating Usage, and for Qualification of Low-VOC Conformal Coatings. 1998.
Iman, R.L. 1994. A Data-Based Approach to Statistics. Duxbury Press.
The Institute for Interconnecting and Packaging Electronic Circuits. 1995. "Ionic Analysis of
Circuit Boards Ion Chromatography Method." IPC-TM-650 Test Methods Manual.
Lincolnwood, IL: IPC.
Vishanoff, Richard. 1995. Marshall Valuation Service: Marshall and Swift the Building Cost
People. Los Angeles, CA: Marshall and Swift Publications.
White, Allan L., Monica Becker and James Goldstein. 1992. Total Cost Assessment.
Accelerating Industrial Pollution Prevention Through Innovative Project Financial Analysis:
With Application to Pulp and Paper Industry. U.S. EPA's Office of Pollution Prevention and
Toxics, Washington, DC.
U. S. EPA (Environmental Protection Agency). 1998. Pollution Prevention and Control Survey.
4-108
-------
Chapter 5
Conservation
Businesses are finding that by conserving resources, both natural and man-made, and
conserving energy, they can cut costs, improve the environment, and improve their
competitiveness. Due to the substantial amount of rinse water consumed and wastewater
generated by the printed wiring board (PWB) manufacturing process, water conservation is an
issue of particular concern to board manufacturers and to the communities in which they are
located. This chapter of the Cleaner Technologies Substitutes Assessment (CTSA) evaluates the
comparative resource consumption and energy use of the surface finishing technologies. Section
5.1 presents a comparative analysis of the resource consumption rates of the surface finishing
technologies, including the relative amounts of rinse water and metals consumed, and a
discussion of factors affecting process and wastewater treatment chemicals consumption.
Section 5.2 presents a comparative analysis of the energy impacts of the surface finishing
technologies, including the relative amount of energy consumed by each process and the
environmental impacts of the energy consumption.
5.1 RESOURCE CONSERVATION
Resource conservation is an increasingly important goal for all industry sectors,
particularly as global industrialization increases demand for limited resources. A PWB
manufacturer can conserve resources through its selection of a surface finishing process and the
manner in which it is operated. By reducing the consumption of resources, a manufacturer will
not only minimize process costs and increase process efficiency, but also will conserve resources
throughout the entire life-cycle chain. Resources typically consumed by the operation of the
surface finishing process include water used for rinsing panels, metals that form the basis of
many of the surface finishing technologies, process chemicals used on the process line,
wastewater treatment chemicals, and energy used to heat process baths and power equipment. A
summary of the effects of the surface finishing technology on the consumption of resources is
presented in Table 5-1.
To determine the effects that surface finishing technologies have on the rate of resource
consumption during the operation of the surface finishing process, specific data were gathered
from chemical suppliers of the various technologies, Performance Demonstration participants,
and from PWB manufacturers through the Workplace Practices Questionnaire and Observer Data
Sheets. Data gathered through these means to determine resource consumption rates include:
• process specifications (e.g., type of process, facility size, process throughput, etc.);
• physical process parameters and equipment description (e.g., automation level, bath size,
rinse water system configuration, pollution prevention equipment, etc.);
• operating procedures and employee practices (e.g., process cycle-time, individual bath
dwell times, bath maintenance practices, chemical disposal procedures, etc.); and
• resource consumption data (e.g., rinse water flow rates, frequency of bath replacement,
criteria for replacement, bath formulations, frequency of chemical addition, etc.).
5-1
-------
Table 5-1. Effects of Surface Finishing Technology on Resource Consumption
Resource
Water
Metals
Process Chemicals
Treatment Chemicals
Energy
Effects of Surface Finishing Technology on Resource Consumption
Water consumption can vary significantly according to the surface finishing
process and level of automation. Other factors such as the cost of water,
sewage costs, and operating practices also affect water consumption rates.
Both the type and quantity of metal consumed is dependent on the surface
finishing technology used by a facility. Metal plating thicknesses are crucial
to surface finishing performance and are set forth in strict guidelines from
process suppliers to PWB manufacturers. Facility operating practices can
influence metal consumption if baths are not maintained properly causing
increased process chemical waste.
Reduction in the number of chemical baths comprising the surface finishing
process typically leads to reduced chemical consumption. The quantity of
process chemicals consumed also is dependent on other factors such as
expected bath lives [e.g., the number of surface square feet (ssf) processed
before a bath must be replaced or chemicals added], process throughput, and
individual facility operating practices.
Water consumption rates and the associated quantities of wastewater
generated, as well as the presence of metal ions and other chemical
constituents, can result in differences in the type and quantity of treatment
chemicals consumed.
Energy consumption rates can differ substantially among the baseline and
alternative processes. Energy consumption is discussed in Section 5.2.
The focus of this section is to perform a comparative analysis of the resource
consumption rates of the baseline [non-conveyorized hot air solder leveling (HASL)] and the
alternative surface finishing technologies. Section 5.1.1 discusses the types and quantities of
natural resources consumed during a surface finishing process operation, while section 5.1.2
focuses on other resources. Section 5.1.3 presents the conclusions drawn from this analysis.
5.1.1 Consumption of Natural Resources
Process resources that can be found naturally in the environment are considered to be
natural resources. Over the last several years there has been a movement towards making society
and the world more sustainable. By limiting the consumption of natural resources to a rate at
which they can replenished, the availability of these precious resources will be assured for future
generations. The concept of sustainability has been adopted by members of the manufacturing
community as part of a successful environmental management program, meant to improve
environmental performance and, by extension, profitability.
A surface finishing process primarily consumes two natural resources: water and metals.
A comparative analysis of the rate of natural resource consumption by each of the surface
finishing technologies is presented below.
5-2
-------
Water Consumption
The surface finishing process line consists of a series of chemical baths which are
typically separated by one or more water rinse steps. These water rinse steps account for
virtually all of the water consumed during the operation of the surface finishing process. The
water baths dissolve or displace residual chemicals from the panel surface, preventing
contamination of subsequent baths, while creating a clean panel surface for future chemical
activity. The number of rinse stages recommended by chemical suppliers for their surface
finishing processes range from three to nine, but can actually be much higher depending on
facility operating practices. The number of separate water rinse stages reported by respondents to
the PWB Workplace Practices Questionnaire ranged from three to seventeen.
The flow rate required by each process rinse tank depends on several factors, including
the time of panel submersion, the type and amount of chemical residue to be removed, the type
of agitation used in the rinse stage, and the purity of rinse water. Because proper water rinsing is
critical to the application of the surface finish, manufacturers often use more water than is
required to ensure that panels are cleaned sufficiently. Other methods, such as flow control
valves and sensors, are available to ensure that sufficient water is available to rinse PWB panels,
while minimizing the amount of water consumed by the process.
PWB manufacturers often use multiple rinse water stages between chemical process steps
to facilitate better rinsing. The first rinse stage removes the majority of residual chemicals and
contaminants, while subsequent rinse stages remove any remaining chemicals. Counter-current
or cascade rinse systems minimize water use by feeding the water effluent from the cleanest rinse
tank, usually at the end of the cascade, into the next cleanest rinse stage, and so on, until the
effluent from the most contaminated, initial rinse stage is sent for treatment or recycle. Other
water reuse or recycle techniques include ion exchange, reverse osmosis, as well as reusing rinse
water in other plant processes. A detailed description of methods to reduce water consumption,
including methods to reuse or recycle contaminated rinse water, is presented in Chapter 6 of this
CTSA.
Water consumption rates for each alternative were calculated using data collected from
both the PWB Workplace Practices Questionnaire and from the Observer Data Sheets completed
during the performance demonstration. Because of the wide variation in the overall, yearly
production of the respondents, it was necessary to normalize the water consumption data to
account for the variety in the overall throughput of the surface finishing process and the
associated water consumption. The daily water consumption for each water rinse reported by a
facility was divided by the overall daily production of the facility to develop a water consumption
rate in gallons per ssf of PWB produced (gal/ssf) for each rinse. An average water consumption
rate was then determined for each automation type and for any specialized rinse conditions (e.g.,
high pressure rinses). The resulting normalized flow rates for each water rinse type are shown in
Table 5-2.
5-3
-------
Table 5-2. Normalized Water Flow Rates of Various Water Rinse Types
Rinse Type
Water Rinse, Non-conveyorized
Water Rinse, Conveyorized
High Pressure Water Rinse, All automation types
Normalized Water Flow Rate a
(gal/ssf)
0.258
0.176
0.465
a Data were normalized to account for differences in facility production rates by dividing the yearly water
consumption by the total PWB produced for each facility. The individual normalized data points were then averaged.
The normalized flow rates were then combined with the standard configuration for each
surface finishing technology (see Section 3.1, Source Release Assessment) to develop an overall
water consumption rate for the entire surface finishing process line. The total water consumption
rate for each surface finishing process was calculated by multiplying the number of rinse stages
(Table 5-3) by the appropriate water flow rate (Table 5-2) for each water rinse category, then
summing the results. The calculations are described by the following equation:
WCRtotal = / j [NRS; x NWCRJ
where,
WCR,otal = total water consumption rate (gal/ssf)
NRS; = number of rinse water stages of type I
NWCR; = normalized water consumption rate for rinse type I (gal/ssf)
The resulting overall rate represents the total water consumption for the entire surface finishing
technology in gallons per ssf of PWB produced. Finally, the total volume of water consumed
while producing 260,000 ssf was calculated using the total water consumption rate for the
process. The number of rinse stages in a standard configuration of each technology, the water
consumption rate of the entire surface finishing process, and the total water consumed by the
application of the surface finish to 260,000 ssf of PWB for each technology is shown in Table 5-3.
The amount of rinse water consumed for each alternative is also displayed graphically in
Figure 5-1, from the lowest to the highest total consumption.
An analysis of the data shows that the type of surface finishing technology, as well as the
level of automation, have a profound affect on the amount of water that a facility will consume
during normal operation of the surface finishing process line. Five surface finishing processes
consume less water than the baseline HASL process, including the convey orized versions of the
HASL, immersion silver, and immersion tin technologies, along with both versions of the organic
solderability preservative (OSP) process. Three surface finishing processes consume more water
than the baseline HASL process: the non-conveyorized versions of the immersion tin,
nickel/gold, and the nickel/palladium/gold technologies.
5-4
-------
Table 5-3. Rinse Water Consumption Rates and Total Water
Consumed by Surface Finishing Technologies
Surface Finishing Technology
HASL, Non-conveyorized
HASL, Conveyorized
Nickel/Gold, Non-conveyorized
Nickel/Palladium/Gold, Non-conveyorized
OSP, Non-conveyorized
OSP, Conveyorized
Immersion Silver, Conveyorized
Immersion Tin, Non-conveyorized
Immersion Tin, Conveyorized
No. of Rinse
Stages a
Normal
Flow
3
3
8
14
3
3
3
7
5
High
Pressure
1
1
-
-
-
-
-
-
-
Total Water
Consumption
Rate"
(gal/ssf)
1.24
0.99
2.06
3.61
0.77
0.53
0.53
1.81
0.88
Rinse Water
Consumed
(gal/260,000 ssf)
3.22 xlO5
2.58 xlO5
5.37 xlO5
9.39 xlO5
2.01 x 105
1.37 xlO5
1.37 xlO5
4.69 x 105
2.29 x 105
a Data reflects the number of rinse stages required for the standard configuration of each surface finishing technology
as reported in Section 3.1, Source Release Assessment.
b Rinse water consumption rate was calculated by multiplying the number of rinse stages for each rinse type by the
corresponding consumption factor listed in Table 5-2. The individual rates were then totaled and divided by 1,000 to
determine the overall consumption rate for that technology.
Immersion Silver (c)
OSP (c)
OSP (nc)
Immersion Tin (c)
HASL(c)
HASL(nc)
Immersion Tin (nc)
Nickel/Gold (nc)
Nickel/Palladium/Gold (nc)
1 2
(gal/ssf)
c: Conveyorized
nc: non-conveyorized
Figure 5-1. Water Consumption Rates of Surface Finishing Technologies
5-5
-------
The rate of water usage is primarily attributable to the number of rinse stages required by
the processes. All of the processes with fewer rinse stages than the baseline HASL process show
reduced water consumption, while all the processes that consumed more water had significantly
more water rinse stages. Only the conveyorized immersion tin process had more water rinse
steps than HASL while consuming less water, due primarily to the high pressure rinse tanks used
by the HASL process.
The table also demonstrates that the conveyorized version of a process will consume less
water during operation than the non-conveyorized version of the same process, a result attributed
to the increased efficiency of the conveyorized processes over their non-conveyorized
counterparts. The increased efficiency is a result of the higher throughput and shorter cycle time
of the conveyorized systems, and is reflected in the normalized water flow rates for rinse stages
for each automation type (Table 5-2).
To minimize water usage, some companies have gone a step farther by developing
equipment systems that monitor water quality and usage in order to optimize water rinse
performance. This pollution prevention technique is recommended to reduce both water
consumption and wastewater generation. The actual water usage experienced by manufacturers
employing such a system may be less than that calculated in Table 5-3.
Metal Consumption
Many of the surface finishes are formed by the deposition of metal ions onto the surface
of the PWB, forming a reliable, solderable finish for further assembly. The metals range from
relatively inexpensive, widely available metals such as tin and lead, found in solder, to expensive
'precious' metals such as silver, gold, and palladium. While a portion of the metal consumed can
be found in the surface finish of the PWB, metal is also lost through drag-out of the plating bath
to subsequent stages, and through the replacement of spent or contaminated plating solutions. In
the case of HASL, solder is also lost through the continual removal of dross, a film of
contaminated solder.
The amount of metal consumed through the deposition, or plating, of the surface finish is
dependent on the thickness of the metal deposit, the amount of PWB surface area that must be
plated, and the density of the metal being applied. The recommended plating thickness for a
surface finishing technology can be obtained from the appropriate chemical supplier. In addition,
plating specifications for surface finishes have been established through testing by both chemical
suppliers and by industry. These specifications set forth strict guidelines on minimum plating
thicknesses required to insure a reliable, solderable surface finish. The metal deposition rates and
the total metal deposited by the surface finishing technologies are presented in Table 5-4.
5-6
-------
Table 5-4. Metal Deposition Rates and Total Metal Consumed by
Surface Finishing Technologies
Process
HASL
Nickel/Gold,
Nickel/Palladium/Gold
Immersion Silver
Immersion Tin
Metal
Tin
Lead
Nickel
Palladium
Gold
Silver
Tin
Density a
(g/cm3)
7.4
11.4
8.1
12.0
19.3
10.5
7.4
Thickness b
(HHI)
126 d
74 d
200
6
7
6
25
Metal Plated c
(oz. per ssf)
0.0194
0.0175
0.0337
0.0015
0.0028
0.0013
0.0038
Total Metal
Consumed
(lb/260K ssf)
315
285
547
24.3
45.6
21.3
62.5
a Source: Chemical Engineers 'Handbook, 1994.
b Thicknesses of deposits recommended by suppliers of individual product lines unless otherwise noted.
0 Calculations assume that 25 percent of the PWB surface area requires metal deposition.
d Plating thickness calculated using a 200 \a in deposit and 63/37 tin-lead solder.
In addition to the metal consumed by the process through deposition or plating, metal is
also lost through drag-out of bath chemicals into subsequent process baths and chemical
degradation through contamination. Metal lost through drag-out along with other process
chemicals were estimated with the use of a model developed specifically for estimating drag-out
in the PWB surface finishing process. A description of the model along with model results are
presented in Section 3.2.3 of the Exposure Assessment.
Calculating the metal lost to bath degradation and subsequent bath replacement is
problematic due to the variability of metal ion concentrations at the time of replacement. The
metal ion concentrations of plating baths are typically replenished regularly rather than replaced
to maintain optimal operating conditions and to prevent depletion of the bath. However, because
the metal baths are valuable, especially the ones containing precious metals, these baths are
typically monitored very closely to prevent a build-up of contaminants and to minimize bath
replacement. When replaced, the spent bath solutions are typically sent off for metal
reclamation. Section 6.2.1, Recycle and Resource Recovery Opportunities, describes reclamation
options and costs for various metals.
A significant amount of solder is also lost through the removal of dross during the
operation of the HASL process. Dross is a solid film of contaminated solder that covers the top
of the molten solder, requiring constant removal through either manual or mechanical means.
Dross is composed of both copper contamination of the solder and the oxidation products of the
tin-lead through contact with air. The amount of solder lost through dross removal can be
significant, estimated to be as much as 90 percent of the solder consumed (Sharp, 2000), though
much can be reclaimed through recycling. If not recycled, dross must be treated as a hazardous
waste. A detailed discussion of solder recycling, including methods of recycling and reclamation
costs, is presented in Section 6.2.1, Recycle and Resource Recovery Opportunities.
5-7
-------
Table 5-4 shows that the use of HASL results in 600 pounds of metal being consumed
through deposition onto the PWB, including 285 pounds of lead, a known environmental toxin.
Only the nickel/palladium/gold process consumes nearly as much metal. It should be noted also
that the values in Table 5-4 only reflect the metal deposited onto the PWBs and do not include
any metal consumed or lost through drag-out, bath contamination, or any other losses such as
dross removal. These losses can be significant as in the case of HASL, where the amount of lead
consumed can be as much as 2,500 pounds if waste solder is not routinely recycled or reclaimed.
Although Table 5-4 shows the relative quantities of metal deposited, any determination of
the relative importance of metal savings on the environment also must consider the availability of
the metal, the toxicity of the metal at disposal, the price of the metal consumed, and the
environmental impacts of mining the metal. While much of this impact analysis is beyond the
scope of this project, the risks to human health and the environment are presented and discussed
in Chapter 3, Risk Screening and Comparison. The cost of process chemicals containing the
metals for each technology are presented in Section 4.2, Cost Analysis.
5.1.2 Consumption of Other Resources
Several resources consumed by the surface finishing processes fall under the category of
man-made, rather than natural, resources. These include process chemicals, treatment chemicals,
bath filters, board laminate, packaging waste, cleaning materials, and any other consumable
materials. Both process chemicals and treatment chemicals are the only resources listed whose
consumption rates are expected to vary significantly between the different surface finishing
technologies. The remaining resources listed are of little concern to this comparative evaluation
because they are either consumed in small quantities, or their consumption rate is not dependent
on the type of surface finishing technology, and so will not vary greatly. A comparative analysis
of the rate of consumption of man-made resources for each of the surface finishing technologies
is presented below.
Process Chemicals Consumption
Bath chemicals that constitute the various chemical baths or process steps are consumed
in large quantities during the normal operation of the surface finishing process, either through co-
deposition with the metals onto the surface of the PWB or degradation through chemical
reaction. Process chemicals are also lost through volatilization, bath depletion, bath drag-out to
subsequent process stages, or contamination as PWBs are cycled through the surface finishing
process. Lost or consumed process chemicals are replaced through chemical additions, or if the
build-up of contaminants is too great, the bath is replaced. Methods for limiting unnecessary
chemical loss and thus minimizing the amount of chemicals consumed are presented in Chapter 6
in this CTSA.
Presenting a chemical-by-chemical analysis of process chemical consumption is not
possible without disclosing the composition and concentration of the proprietary chemical
formulations collected from the chemical suppliers (the actual chemical consumption is a
combination of the quantity and concentration of chemicals present, factors which vary greatly,
5-8
-------
even with processes within a similar technology category). Legal constraints prevent the
disclosure of this information. However, two of the primary conclusions drawn from the analysis
are the effects of the chemical consumption on the process cost and on human health. These
conclusions are presented in detail in the Risk Characterization (Section 3.4) and in the Cost
Analysis (Section 4.2) portions of this document. A qualitative discussion of the factors found to
contribute to the consumption of process chemicals is presented below.
Performing a comparative analysis of the process chemical consumption rates is
problematic due to both the site-specific nature of many of the factors that contribute to process
chemical consumption, and the differences in concentration and chemical composition of the
solutions involved (i.e., would the consumption of one pound of hydrochloric acid be equivalent
to one pound of ethylene glycol?). Factors affecting the rate at which process chemicals are
consumed through the operation of the surface finishing process include:
• characteristics of the process chemicals (i.e., composition, concentration, volatility, etc.);
• process operating parameters (i.e., number of chemical baths, process throughput,
automation, etc.); and
• bath maintenance procedures (i.e., frequency of bath replacement, replacement criteria,
frequency of chemical additions, etc.).
The chemical characteristics of the process chemicals determine the rate at which
chemicals are consumed in the surface finishing process. A chemical bath containing a highly
volatile chemical, or mixture of chemicals, can experience significant chemical losses to the air.
A more concentrated process bath will lose a greater amount of process chemicals in the same
volume of drag-out than a less concentrated bath. These chemical characteristics not only vary
among surface finishing alternatives, but can also vary considerably among surface finishing
processes offered by different chemical suppliers within the same technology category.
The physical operating parameters of the surface finishing process also have a significant
impact on the consumption rate of process chemicals. One such parameter is the number of
chemical baths contained within the surface finishing process (the surface finishing process is
comprised of several process stages, some of which are chemical process baths). The number of
chemical process baths through which a panel must be processed to perform the surface finishing
function varies widely among the technologies, with a corresponding affect on chemical
consumption. The number of chemical baths (excluding rinse stages) range from three for OSP
to eight in the nickel/palladium/gold technology. The process throughput, or quantity of PWBs
passed through the surface finishing process, also affects chemical usage since the higher the
throughput, the more process chemicals are consumed. However, conveyorized processes tend
to consume less chemicals per ssf than non-conveyorized versions of the same process due to the
smaller bath sizes and higher efficiencies of the automated processes.
The greatest impact on process chemical consumption can result from the bath
maintenance procedures of the facility operating the process. The frequency with which baths
are replaced and the bath replacement criteria used are key chemical consumption factors.
Chemical suppliers typically recommend that chemical baths be replaced using established
5-9
-------
testing criteria, such as concentration thresholds of bath constituents (e.g., 2 g/L of copper).
Other bath replacement criteria include ssf of PWB processed and elapsed time since the last bath
replacement. The practice of making regular adjustments to the bath chemistry through additions
of process chemicals consumes process chemicals, but will extend the operating life of the
process baths, reducing chemical use over time. Despite the supplier recommendations, project
data showed a wide range of bath replacement practices and criteria for manufacturing facilities
operating the same, as well as different, surface finishing technologies.
Wastewater Treatment Chemicals Consumption
The extent to which the consumption of treatment chemicals will be reduced, if any, is
dependent on several factors, some of which include the rate at which wastewater is generated
(e.g., the amount of rinse water consumed), the type of treatment chemicals used, composition of
waste streams from other plant processes, percentage of treatment plant throughput attributable
to the surface finishing process, the resulting reduction in surface finishing waste volume realized,
and the extent to which the former surface finishing process was optimized for waste reduction.
Because many of the above factors are site-specific and not dependent on the type of surface
finishing process, a quantitative evaluation would not be meaningful. However, there is a direct
correlation between the amount of treatment chemicals required and the amount of process
chemicals lost to drag-out that must be treated. A description of a typical wastewater treatment
process, along with the types of treatment chemicals used to treat contaminated wastewater, is
presented in Section 6.2.2, Control Technologies.
Alternative treatment processes to conventional precipitation treatment may be available
to reduce the amount of treatment chemical consumption depending on the type of surface
finishing process being operated. A discussion of treatment options for each technology,
including a treatment profile for each type of process bath, also is presented in Section 6.2.2,
Control Technologies.
5.1.3 Summary and Conclusions
A comparative analysis of the water consumption rates was performed for the surface
finishing technologies. A daily water flow rate was developed for each surface finishing
technology using survey data provided by industry. Calculated water consumption rates ranged
from a low of 0.53 gal/ssf for the immersion silver and OSP conveyorized processes, to a high of
3.6 gal/ssf for the non-conveyorized nickel/palladium/gold process. Several processes were
found to consume less water than the HASL baseline including convey orized versions of the
immersion silver and immersion tin technologies, along with both versions of the OSP process.
Convey orized processes were found to consume less water than non-conveyorized versions of
the same process. Primary factors influencing the water consumption rate included the number
of rinse tanks and the overall efficiency of the conveyorized processes.
Metals are another natural resource consumed by a surface finishing process. The rate of
deposition of metal was calculated for each technology along with the total amount of metal
consumed for 260,000 ssf of PWB produced. It was shown that the consumption of close to 300
5-10
-------
pounds of lead could be eliminated by replacing the baseline HASL process with an alternative
technology. In cases where waste solder is not routinely recycled or reclaimed, the consumption
of as much as 2,500 pounds of lead could be eliminated by replacement of the HASL process.
Although several of the alternative technologies rely on the use of small quantities of other
metals, the OSP technology eliminates metal consumption entirely. Other factors influencing
metal consumption were identified and discussed.
A quantitative analysis of both process chemicals and treatment chemicals consumption
could not be performed due to the variability of factors that affect the consumption of these
resources, and for reasons of confidentiality. The role the surface finishing process has in the
consumption of these resources was presented and the factors affecting the consumption rates
were identified and discussed.
5-11
-------
5.2 ENERGY IMPACTS
Energy conservation is an important goal for PWB manufacturers, as companies strive to
cut costs and seek to improve environmental performance and global competitiveness. Energy
use has become an important consideration in the manufacture of PWB s, as much of the
manufacturing process requires potentially energy-intensive operations, such as heating the
process baths. This is especially true during the operation of the surface finishing process, where
energy is consumed by process equipment such as immersion heaters, fluid and air pumps,
agitation devices such as vibrating motors, and by conveyorized transport systems. The focus of
this section is to perform a comparative analysis of the relative energy consumption rates of the
baseline HASL process and alternative surface finishing technologies.
Data collected for this analysis focus on the energy consumed during the application of
the surface finish. Traditional life-cycle analysis indicates that energy consumption during other
life-cycle stages also can be significant and should be considered when possible. Although a
quantitative life-cycle analysis is beyond the scope and resources of this project, the impacts to
the environment from the manufacture of the energy required by the surface finishing process is
briefly analyzed and presented at the end of this chapter.
Section 5.2.1 discusses energy consumption during the application of the surface finish,
while Section 5.2.2 discusses the environmental impacts of this energy consumption. Section
5.2.3 briefly discusses the energy consumption of other life-cycle stages. Section 5.2.4 presents
conclusions of the comparative energy analysis.
5.2.1 Energy Consumption During Surface Finishing Process Operation
To determine the relative rates of energy consumption during the operation of the surface
finishing technologies, specific data were collected regarding energy consumption through the
Performance Demonstration project and through dissemination of the PWB Workplace Practices
Questionnaire to industry members. Energy data collected include the following:
• process specifications (i.e., type of process, facility size, etc.);
• physical process parameters (i.e., number of process baths, bath size, bath conditions
such as temperature and mixing, etc.);
• process automation (i.e., conveyorized, computer-controlled hoist, manual, etc.);
• equipment description (i.e., heater, pump, motor, etc.); and
• equipment energy specifications (i.e., electric load, duty, nominal power rating,
horsepower, etc.).
Each of the surface finishing technologies contains a series of chemical baths that are
typically separated by one or more water rinse steps. In some processes, these chemical stages
are supplemented by other stages such as a drying oven or a HASL machine, which applies the
solder to the PWB using a mechanical type of process. In order for the process to perform
properly, each process stage should be operated within specific supplier recommended
parameters, such as parameters for bath temperature and mixing, oven temperatures, or air knife
5-12
-------
pressures. Maintaining these process stages within the desired parameters often requires energy-
consuming equipment such as immersion heaters, fluid circulation pumps, and air compressors.
In addition, the degree of process automation affects the relative rate of energy consumption.
Clearly, conveyorized equipment requires energy to operate, but also non-conveyorized systems
require additional equipment not found in conveyorized systems, such as panel agitation
equipment.
Table 5-5 lists the types of energy-consuming equipment typically used during the
operation of a surface finishing process and the function of the equipment. In some cases, one
piece of equipment may be used to perform a function for the entire process line. For example,
in a non-conveyorized system, panel vibration is typically performed by a single motor used to
rock an apparatus that extends over all of the process tanks. The apparatus provides agitation to
each individual panel rack that is connected to it, thus requiring only a single motor to provide
agitation to every bath on the process line that may require it. Other equipment types such as
immersion heaters affect only one process stage, so each process bath or stage may require a
separate piece of energy-consuming equipment.
Table 5-5. Energy-Consuming Equipment Used in Surface Finishing Process Lines
Type of Equipment
Conveyor Drive Motor
Immersion Heater
Fluid Pump
Air Pump
Panel Agitation Motor
Gas Heater
Solder Pot
Ventilation Equipment
Function
Powers the conveyor system required to transport PWB panels through the
surface finishing process. Not required for non-conveyorized, vertical
processes.
Raises and maintains temperature of a process bath to the optimal operating
temperature.
Circulates bath fluid to promote flow of bath chemicals through drilled
through holes and to assist filtering of impurities from bath chemistries.
Compresses and blows air into process baths to promote agitation of bath to
ensure chemical penetration into drilled through holes. Also provides
compressed air to processes using an air knife to remove residual chemicals
from PWB panels.
Moves the apparatus used to rock panel racks back and forth in process
baths. Not required for conveyorized processes.
Heats PWB panels to promote drying of residual moisture on the panel
surface. Can also be used to cure a chemical coating.
Melts solder and maintains the molten solder at optimal operating
temperature, usually between 480 to 550 °F.
Provides ventilation required for surface finishing baths and to exhaust
chemical fumes.
To assess the energy consumption rate for each surface finishing technology, an energy
use profile was developed that identified typical sources of energy consumption during the
application of the surface finish. The number of surface finishing process stages that result in the
consumption of energy during operation was determined from Performance Demonstration and
PWB Workplace Practices Questionnaire data. This information is listed in Table 5-6 according
to the function of the energy-consuming equipment. For example, a typical non-conveyorized
5-13
-------
OSP process consists of two heated chemical baths, three chemical baths requiring fluid
circulation, two process stages requiring compressed air (for air knives in this case), and a single
heated drying stage to cure the OSP coating. Panel agitation for the entire process is provided by
a single motor used to rock an apparatus that extends over all of the process tanks. Ventilation
equipment is not presented in Table 5-6 because the necessary data were not collected during the
Performance Demonstration or in the PWB Workplace Practices Questionnaire. However, the
amount of ventilation required varies according to the type of chemicals, bath operating
conditions, and the configuration of the process line. Because they are enclosed, the ventilation
equipment for conveyorized processes are typically more energy efficient than non-conveyorized
processes.
Table 5-6. Number of Surface Finishing Process Stages that Consume Energy
by Function of Equipment
Process Type
HASL, Non-conveyorized
HASL, Conveyorized
Nickel/Gold, Non-conveyorized
Nickel/Palladium/Gold,
Non-conveyorized
OSP, Non-conveyorized
OSP, Conveyorized
Immersion Silver, Conveyorized
Immersion Tin, Non-conveyorized
Immersion Tin, Conveyorized
Function of Equipment a
Conveyor
0
1
0
0
0
1
1
0
1
Panel
Agitation
1
0
1
1
1
0
0
1
0
Bath
Heat
1
1
4
6
2
2
2
3
3
Ur Knife/
Sparging
2
2
1
1
2
2
0
0
0
Fluid
Circulation
3
4
3
3
3
3
4
4
3
Panel
Drying
1
1
0
0
1
1
1
1
1
Solder
Heater
1
1
0
0
0
0
0
0
0
Table entries for each surface finishing alternative represent the number of process stages requiring each specific
function. All functions are supplied by electric equipment, except for drying, which is performed by gas-fired oven.
b Processes reporting panel agitation for one or more process stages are entered as one in the summary regardless of
the number since a single motor can provide agitation for the entire process line.
0 Air sparging is used selectively by some manufacturers to enhance bath performance. Sparging may not be required
for all product lines or facilities using a surface finishing technology.
The electrical energy consumption of surface finishing line equipment, as well as
equipment specifications (power rating, average duty, and operating load), were collected during
the Performance Demonstration. In cases where electricity consumption data were not available,
the electricity consumption rate was calculated using the following equation:
EC = NPR x OL x AD x (lkW/0.746 HP)
where,
5-14
-------
EC
NPR =
OL =
AD =
electricity consumption rate (kWh/day)
nominal power rating (HP)
operating load (percent), or the percentage of the maximum load or output of
the equipment that is being used
average duty (hr/day), or the amount of time per day that the equipment is
being operated at the operating load
Electricity consumption data for each equipment category were averaged to determine the
average amount of electricity consumed per hour of operation for each type of equipment per
process. The natural gas consumption rate for a drying oven was supplied by an equipment
vendor. Electricity and natural gas consumption rates for surface finishing equipment per
process stage are presented in Table 5-7.
Table 5-7. Energy Consumption Rates for Surface Finishing Equipment
Function of Equipment
Conveyorized Panel Automation
Panel Agitation
Bath Heater
Fluid Circulation
Air Knife/Sparging
Panel Drying
Solder Heater
Type of Equipment
Conveyor System
Panel Agitation Motor
Immersion Heater
Fluid Pump
Air Pump
Gas Drying Oven
Solder Pot
Energy Consumption Rates Per
Equipment Type
Electricity a
(kW)
14.1
3.1
4.1
0.9
3.8
-
20
Natural Gas b
(ft3/hr)
-
-
-
-
-
90
Electricity consumption rates for each type of equipment were calculated by averaging energy consumption data per
stage from the performance demonstrations. If required, consumption data were calculated from device specifications
and converted to total kW per bath using 1 HP = 0.746 kW.
b Natural gas consumption rate for the gas heater was estimated by an equipment vendor (Exair Corp.).
The total electricity consumption rate for each surface finishing alternative was calculated
by multiplying the number of process stages that consume electricity (Table 5-6) by the
appropriate electricity consumption rate (Table 5-7) for each equipment category, then summing
the results. The calculations are described by the following equation:
where,
ECRtotai
NFS;
ECR;
[NFS; x ECRJ
total electricity consumption rate (kW)
number of process stages requiring equipment i
energy consumption rate for equipment i (kW)
5-15
-------
Natural gas consumption rates were calculated using a similar method. The individual
energy consumption rates for both natural gas and electricity were then converted to British
Thermal Units (Btu) per hour and summed to give the total energy consumption rate for each
surface finishing technology. The individual consumption rates for both natural gas and
electricity, as well as the hourly energy consumption rate calculated for each of the surface
finishing technologies, are listed in Table 5-8.
These energy consumption rates include only the types of equipment listed in Table 5-5,
which are commonly recommended by chemical suppliers to successfully operate a surface
finishing process. However, equipment such as ultrasonics, automated chemical feed pumps,
vibration units, panel feed systems, or other types of electrically powered equipment may be part
of the surface process line. The use of this equipment may improve the performance of the
surface finishing process, but is not required in a typical process for any of the surface finishing
technologies.
Table 5-8. Hourly Energy Consumption Rates for Surface Finishing Technologies
Process Type
HASL, Non-conveyorized
HASL, Conveyorized
Nickel/Gold, Non-conveyorized
Nickel/Palladium/Gold, Non-conveyorized
OSP, Non-conveyorized
OSP, Conveyorized
Immersion Silver, Conveyorized
Immersion Tin, Non-conveyorized
Immersion Tin, Conveyorized
Energy Consumption Rates
Electricity
(kW)
37.5
49.4
26.0
34.2
21.6
32.6
25.9
19.0
29.1
Natural Gas
(ft3/hr)
90
90
-
-
90
90
90
90
90
Hourly
Consumption
Rate a (Btu/hr)
219,800
260,400
88,700
116,700
165,500
203,100
180,200
156,700
191,100
a Electrical energy was converted at the rate of 3,413 Btu per kilowatt hour. Natural gas consumption was converted
at the rate of 1,020 Btu per cubic feet of gas consumed.
To determine the overall amount of energy consumed by each technology, the hourly
energy consumption rate from Table 5-8 was multiplied by the amount of time needed for each
alternative to manufacture 260,000 ssf of PWB (the average HASL throughput of respondents to
the PWB Workplace Practices Questionnaire). Because insufficient survey data exist to
accurately estimate the amount of time required for each process to produce the 260,000 ssf of
board, the operating time was simulated using a computer model developed for each surface
finishing technology. The results of the simulation, along with a discussion of the data and
parameters used to define each technology, are presented in Section 4.2, Cost Analysis. The
hours of surface finishing operation required to produce 260,000 ssf of board from the
simulation, the total amount of energy consumed, and the energy consumption rate per ssf of
board produced for each technology are presented in Table 5-9.
5-16
-------
Table 5-9. Energy Consumption Rate per ssf of PWB Produced
for Surface Finishing Technologies
Process Type
HASL, Non-conveyorized
HASL, Conveyorized
Nickel/Gold, Non-conveyorized
Nickel/Palladium/Gold, Non-conveyorized
OSP, Non-conveyorized
OSP, Conveyorized
Immersion Silver, Conveyorized
Immersion Tin, Non-conveyorized
Immersion Tin, Conveyorized
Process
Operating Time a
(hours)
258
133
1,310
1,710
197
93
414
480
710
Total Energy
Consumed
(Btu/260,000
ssf)
5.67 xlO7
3.46 xlO7
1.16xl08
2.00 x 10s
3.26 xlO7
1.89 xlO7
7.46 x 107
7.52 xlO7
1.36 x 10s
Energy
Consumption Rate
(Btu/ssf)
218
133
447
768
125
73
287
289
522
a Times listed represent the operating time required to manufacture 260,000 ssf of PWB by each process as simulated
by computer model. Operating time was considered to be the overall process time minus the downtime of the
process.
Table 5-9 shows that three of the process alternatives consumed less energy than the
baseline, non-conveyorized, HASL process. Both the non-conveyorized and Conveyorized
versions of the OSP process, along with the Conveyorized HASL process, consumed significantly
less energy than the baseline process. The reductions were primarily attributable to the efficiency
of the three processes, which resulted in operating times significantly less than that of the
traditional non-conveyorized HASL process. Both the immersion silver process and the
Conveyorized immersion tin processes performed roughly equal to the baseline process, utilizing
a lower hourly consumption rate to offset a small disadvantage in operating time.
Three processes consumed significantly more energy than the baseline process. Despite
having the lowest hourly consumption rate of all the surface finishing technologies, the
nickel/gold process consumed more than twice the energy of the baseline due to its long process
operating time. Other processes with high energy consumption rates include
nickel/palladium/gold and Conveyorized immersion tin.
The performance of specific surface finishing technologies with respect to energy is
primarily dependent on the hourly energy consumption rate (Table 5-8) and the overall operating
time for the process (Table 5-9). Non-conveyorized processes typically have lower hourly
consumption rates than Conveyorized processes of the same type because the operation of
Conveyorized equipment is more energy-intensive. Although Conveyorized processes typically
have higher hourly consumption rates, these differences are usually more than offset by the
shorter operating times that are required to produce an equivalent quantity of PWB s.
When the non-conveyorized and Conveyorized versions of a surface finishing technology
are compared, the Conveyorized versions of the technology seem to be typically more energy
efficient. Table 5-10 compares the energy consumption data for those technologies that are
5-17
-------
operated in both conveyorized and non-conveyorized modes. This table shows that, although the
conveyorized version of all three processes requires more energy per hour to operate than the
non-conveyorized mode, the added efficiency of the conveyorized system (reflected in the
shorter operating time) results in less energy usage per ssf of board produced. The immersion tin
processes are the exceptions. The non-conveyorized configuration of this process not only has a
better hourly consumption rate than the conveyorized, but also benefits from a faster operating
time, a condition due to the long overall cycle-time required for the conveyorized process. These
factors combine to give the non-conveyorized immersion tin process a lower energy
consumption rate than the conveyorized version. Despite this exception, the overall efficiency of
conveyorized systems typically will result in less energy usage per ssf of board produced, as it
did for both the HASL and OSP processes.
Table 5-10. Effects of Automation on Energy Consumption
for Surface Finishing Technologies
Process Type
HASL, Non-conveyorized
HASL, Conveyorized
OSP, Non-conveyorized
OSP, Conveyorized
Immersion Tin, Non-
conveyorized
Immersion Tin, Conveyorized
Hourly
Consumption Rate
(1,000 Btu/ssf)
220
260
165
203
156
191
Process
Operating Time a
(hours)
258
133
197
93
480
710
Energy Consumption
Rate
(Btu/ssf)
218
133
125
73
289
522
Times listed represent the operating time required to manufacture 260,000 ssf of PWB by each process as simulated
by computer model. Operating time was considered to be the overall process time minus the downtime of the
process.
Finally, it should be noted that the overall energy use experienced by a facility will
depend greatly upon the operating practices and the energy conservation measures adopted by
that facility. To minimize energy use, several simple energy conservation opportunities are
available and should be implemented. These include insulating heated process baths, using
thermostats on heaters, and turning off equipment when not in use.
5.2.2 Energy Consumption Environmental Impacts
The production of energy results in the release of pollution into the environment,
including pollutants such as carbon dioxide (CO2), sulfur oxides (SOX), carbon monoxide (CO),
sulfuric acid (H2SO4), and particulate matter. The type and quantity of pollution depends on the
method of energy production. Typical energy production facilities in the U.S. include
hydroelectric, nuclear, and coal-fired generating plants.
The environmental impacts attributable to energy production resulting from the
differences in energy consumption among surface finishing technologies were evaluated using a
computer program developed by the EPA National Risk Management Research Laboratory, P2P-
5-18
-------
version 1.50214 (U.S. EPA, 1994). This program can, among other things, estimate the type and
quantity of pollutant releases resulting from the production of energy as long as the differences in
energy consumption and the source of the energy used (e.g., electrical energy from a coal-fired
generating plant, thermal energy from a oil-fired boiler, etc.) are known. The program uses data
reflecting the "national average" pollution releases per kilowatt-hour derived from particular
sources. Electrical power derived from the average national power grid was selected as the
source of electrical energy, while natural gas was used as the source of thermal energy for this
evaluation. Energy consumption rates from Table 5-8 were multiplied by the operating time
required to produce 260,000 ssf of board reported for each technology in Table 5-9. These totals
were then divided by 260,000 to get the electrical and thermal energy consumed per ssf of board,
which were then used as the basis for the analysis. Results of the environmental impact analysis
from energy production are summarized and presented in Table 5-11. Appendix H contains
printouts from the P2P program for each alternative.
Although the pollutant releases reported in Table 5-11 are combined for all media (i.e., air,
water, and land), they often occur in one or more media, where they may present different
hazards to human health or the environment. To allow a comparison of the relative effects of any
pollution that may occur, it is necessary to identify the media of releases. Table 5-12 displays the
pollutants released during the production of energy, the media into which they are released, and
the environmental and human health concerns associated with each pollutant.
The information presented in Tables 5-11 and 5-12 show that the generation of energy is
not without environmental consequences. Pollutants released to air, water, and soil resulting
from energy generation can pose direct threats to both human health and the environment. As
such, the consumption of energy by the surface finishing process contributes directly to the type
and magnitude of these pollutant releases. Primary pollutants released from the production of
electricity include CO2, solid wastes, SOX, and nitrogen oxides. These pollutants contribute to a
wide range of environmental and human health concerns. Natural gas consumption results
primarily in releases of CO2 and hydrocarbons, which typically contribute to environmental
problems such as global warming and smog. Minimizing the amount of energy usage by the
surface finishing process, either by selection of a more energy efficient process or by adopting
energy efficient operating practices, will decrease the quantity of pollutants released into the
environment resulting from the generation of the energy consumed.
5-19
-------
"i s o
£2 "2,
Sfl
a
4)
o
OH
a $
S
"3
u
1
o
42
U
s
•O
Sr w
jj; 15
to ~5
B
fig
e '-o
•I £
S g
1^
P* BO
^"^ 43
o
0
ON
CN
es
O
!i
T3 O
s s^
OS >•
I?
.a §
ON
00
zed
-co
ON
CN
I
§
o
oT
(N
oo
OO
S
o
m W
O
O
ON
oo
O
tal
5-20
-------
Table 5-12. Pollutant Environmental and Human Health Concerns
Pollutant
Carbon Dioxide (CO2)
Carbon Monoxide (CO)
Dissolved Solids
Hydrocarbons
Nitrogen Oxides (NOX)
Particulates
Solid Wastes
Sulfur Oxides (SOX)
Sulfuric Acid (H2SO4)
Medium
of Release
Air
Air
Water
Air
Air
Air
Soil
Air
Water
Environmental and Human Health Concerns
Global warming
Toxic organic,3 smog
Dissolved solids b
Odorant, smog
Toxic inorganic,3 acid rain, corrosive, global warming, smog
Particulates °
Land disposal capacity
Toxic inorganic,3 acid rain, corrosive
Corrosive, dissolved solids b
a Toxic organic and inorganic pollutants can result in adverse health effects in humans and wildlife.
b Dissolved solids are a measure of water purity and can negatively affect aquatic life as well as the future use of the
water (e.g., salinity can affect the water's effectiveness at crop irrigation).
0 Paniculate releases can promote respiratory illness in humans.
5.2.3 Energy Consumption in Other Life-Cycle Stages
When performing a comparative evaluation among surface finishing technologies, the
energy consumed throughout the entire life cycle of the chemical products in the technology
should be considered. The product use phase is only one aspect of the environmental
performance of a product. A life-cycle analysis considers all stages of the life of a product,
beginning with the extraction of raw materials from the environment, and continuing on through
the manufacture, transportation, use, recycle, and ultimate disposal of the product.
Each stage within this life cycle consumes energy. It is possible for a product to be
energy efficient during the use phase of the life cycle, yet require large amounts of energy to
manufacture or dispose of the product. There are energy consumption differences also in the
transportation of wastes generated by a surface finishing process. The transportation of large
quantities of sludge resulting from the treatment of processes with chelated waste streams (i.e.,
nickel/gold) will consume more energy than the transportation of smaller quantities of sludge
resulting from processes that do not use chelators. These examples show that energy use from
other life-cycle stages can be significant and should be considered when evaluating the energy
performance of a product. However, a comprehensive assessment of other life-cycle stages was
beyond the scope of this study.
5-21
-------
5.2.4 Summary and Conclusions
A comparative analysis of the relative energy consumption rates was performed for the
surface finishing technologies. An hourly energy consumption rate was developed for the
baseline non-conveyorized HASL process and each alternative using data collected from industry
through a survey. A computer simulation was used to determine the operating time required to
produce 260,000 ssf of PWB, and an energy consumption rate per ssf of PWB was calculated.
The energy consumption rates ranged from 73 Btu/ssf for the conveyorized OSP process to 768
Btu/ssf for the non-conveyorized nickel/palladium/gold process. The results indicate that three
surface finishing processes are more energy efficient than the traditional non-conveyorized
HASL process, while two others are roughly comparable. It was found also that for alternatives
with
both types of automation, the conveyorized version of the process is typically the more energy
efficient (HASL and OSP), with the notable exception of the immersion tin process.
An analysis of the impacts directly resulting from the production of energy consumed by
the surface finishing process showed that generation of the required energy is not without
environmental consequence. Pollutants released to air, water, and soil can result in damage to
both human health and the environment. The consumption of natural gas tends to result in
releases to the air which contribute to odor, smog and global warming, while the generation of
electricity can result in pollutant releases to all media, with a wide range of possible affects.
Minimizing the amount of energy usage by the surface finishing process, either by selection of a
more energy efficient process or by adopting energy efficient operating practices, will decrease
the quantity of pollutants released into the environment resulting from the generation of the
energy consumed.
5-22
-------
REFERENCES
Chemical Engineers 'HandbooR.994. McGraw-Hill Book Company.
Sharp, John. 2000. Teradyne, Inc. Personal communication to Jack Geibig, UT Center for Clean
Products and Clean Technologies.
U.S. EPA (Environmental Protection Agency). 1994. P2P-version 1.50214 computer software
program. Office of Research and Development, National Risk Management Lab, Cincinnati, OH.
5-23
-------
Chapter 6
Additional Environmental Improvement Opportunities
This chapter of the Cleaner Technologies Substitute Assessment (CTSA) identifies and
qualitatively discusses techniques that can be used by printed wiring board (PWB) manufacturing
facilities to prevent pollution, minimize waste, recycle and recover valuable resources, and control
releases. The Pollution Prevention Act of 1990 set forth the following hierarchy to waste
management in order of desirability:
• pollution prevention at the source;
• recycling in an environmentally safe manner;
• treatment in an environmentally safe manner; and
• disposal or other release into the environment only as a last resort and in an
environmentally safe manner.
This hierarchy has been adopted by EPA as the preferred method of waste management
to reduce or eliminate potential releases by industry. The hierarchy reflects the common sense
notion that preventing pollution in an environmentally safe manner is preferable to any
subsequent response, be it recycling, treatment, or disposal. Acceptable pollution prevention
methods include product and process redesign and the selection of safe substitutes for problem
processes/chemicals, along with other traditional pollution prevention techniques that reduce
pollution at the source (Kling, 1995).
The hierarchy also recognizes that pollution prevention is not always possible and that
other waste management methods are often required. When pollution prevention is not possible,
we should turn in order to recycling, treatment, and finally disposal if no other option remains. A
manufacturing facility often combines pollution prevention techniques with these other
approaches to effectively reduce emissions from a production process. While pollution
prevention is generally the most desirable of the above choices, the most important aspect of this
hierarchy is to reduce the environmental impacts of the overall process as much as is feasible
while maintaining the quality, performance, and safety criteria for the products being
manufactured.
This chapter focuses on the application of the waste management hierarchy to waste
streams generated by the surface finishing process of the PWB industry. Techniques are
identified, organized, and presented in an order corresponding to the hierarchy. Pollution
prevention techniques are presented in Section 6.1, while methods for minimizing waste,
recycling or recovering resources, and controlling releases are presented in Section 6.2. While the
focus of this chapter is on the surface finishing line, many of the techniques described here can
be applied to other processes used in PWB manufacturing. A series of pollution prevention case
studies developed by the EPA Design for the Environment (DfE) Program for the PWB industry
present examples of the successful implementation of techniques available to industry (U.S. EPA,
1995a; U.S. EPA, 1995b; U.S. EPA, 1996a; U.S. EPA, 1996b; U.S. EPA, 1996c; U.S. EPA, 1997a;
U.S. EPA, 1997b; U.S. EPA, 1997c; U.S. EPA, 1999).
6-1
-------
6.1 POLLUTION PREVENTION
Pollution prevention, defined in the Pollution Prevention Act of 1990, is the reduction in
the amounts or hazards of pollution at the source and is often referred to as source reduction.
Source reduction, also defined in the Pollution Prevention Act, is any practice which: 1) reduces
the amount of any hazardous substance, pollutant, or contaminant entering any waste stream or
otherwise released into the environment (including fugitive emissions) prior to recycling,
treatment, or disposal; and 2) reduces the hazards to public health and the environment
associated with the release of such substances, pollutants, or contaminants. Source
reduction/pollution prevention includes equipment or technology modifications, process or
procedure modifications, reformulation or redesign of products, substitution of raw materials,
and improvements in housekeeping, maintenance, training, or inventory control.
EPA's regulations are moving towards incorporating pollution prevention options. For
example, the EPA Office of Water is currently developing a set of proposed effluent guidelines
for the metal products and machinery industries, which are expected to be published in October,
2000. The proposed rule will discuss ten options that can be employed to meet effluent
guidelines and standards, five of which include specific pollution prevention technologies.
Current pollution prevention practices within the PWB industry were identified and data
were collected through contact with industry personnel, extensive review of published accounts,
and through the design and dissemination of two information requests to PWB manufacturers.
The PWB Workplace Practices Questionnaire, conducted as part of this CTSA, specifically
focused on the surface finishing process to identify important process parameters and operating
practices for the various surface finishing technologies. For a breakdown of respondents by
alternative, refer to Section 1.3 of the Introduction. Facility characteristics of respondents are
presented in Section 3.2, Exposure Assessment. The PWB Workplace Practices Questionnaire is
presented in Appendix A.
The Pollution Prevention and Control Technology Survey (hereafter referred to as the
Pollution Prevention Survey) was an update to a previous survey and was designed to collect
information about past and present pollution prevention procedures and control technologies for
the entire PWB manufacturing process. This Survey was performed by the DfE PWB Project
and is documented in the EPA publication, Printed Wiring Board Pollution Prevention and
Control Technology: Analysis of Updated Survey Results (U.S. EPA, 1998). The Survey results
presented periodically throughout this chapter are compiled from responses to the Pollution
Prevention Survey unless otherwise indicated. Results from the Pollution Prevention Survey
pertaining to recycle or control technologies are presented in Section 6.2 of this chapter.
Opportunities for pollution prevention in PWB manufacturing were identified in each of
the following areas:
6-2
-------
• management and personnel practices;
• materials management and inventory control;
• materials selection; and
• process improvements.
The successful implementation of pollution prevention practices can lead to reductions in
waste treatment, pollution control, environmental compliance, and liability costs. Cost savings
can result directly from pollution prevention techniques that minimize water usage, primary or
ancillary material consumption, and process waste generation.
6.1.1 Management and Personnel Practices
Pollution prevention is an ongoing activity that requires the efforts of both management
and employees to achieve the best results. While pollution prevention initiatives, such as an ISO
14000-type environmental management system, require a commitment and continued support
from management, any pollution prevention measures taken are ultimately implemented by the
process employees, making them an integral part of any pollution prevention effort.
Management and employees must work together to form an effective pollution prevention
program.
Just under two thirds (60.9 percent) of the PWB companies responding to the Pollution
Prevention Survey reported having a formal pollution prevention policy statement while just over
half (55.1 percent) of the survey respondents reported having a pollution prevention program.
Over two thirds (71.2 percent) of PWB companies surveyed reported conducting employee
education for pollution prevention. Each of these statistics in the current Pollution Prevention
Survey increased between three and eight percent over the same statistics in the prior survey,
showing improvement in company perspectives on pollution prevention since the previous
survey was conducted.
The scope and depth of pollution prevention planning and the associated activities will
vary with the size of the facility. While larger facilities may go through an entire pollution
prevention planning exercise (as described below), smaller facilities may require as little as a
commitment by the owner to pollution prevention along with cooperation and assistance from
employees to meet any stated goals. A list of management and personnel practices that promote
pollution prevention, along with the benefits, are listed in Table 6-1.
A company's commitment to pollution prevention begins with a pollution prevention and
waste reduction policy statement. This statement, which is the company's public proclamation
of its dedication to preventing pollution and reducing waste, should clearly state why a program
is being undertaken, include specific pollution prevention and waste reduction goals, and assign
responsibility for accomplishing those goals. The statement details to the public and to its
employees the depth of the company's commitment to pollution prevention.
6-3
-------
Table 6-1. Management and Personnel Practices Promoting Pollution Prevention
Method
Create a company pollution prevention and
waste reduction policy statement.
Develop a written pollution prevention and
waste reduction plan.
Provide periodic employee training on pollution
prevention.
Make employees accountable for their pollution
prevention performance and provide feedback
on their performance.
Promote internal communication between
management and employees.
Implement total cost accounting or activity-based
accounting system.
Benefits
Communicates to employees and states publicly the
company commitment to achieving pollution
prevention and waste reduction goals.
Communicates to employees how to accomplish the
goals identified in the company's policy statement.
Identifies in writing specific implementation steps
for pollution prevention.
Educates employees on pollution prevention
practices.
Provides incentives to employees to improve
pollution prevention performance.
Informs employees and facilitates input on pollution
prevention from all levels of the company.
Identifies true costs of waste generation and the
benefits of pollution prevention.
A pollution prevention plan is needed to detail how the pollution prevention and waste
reduction goals described in the company's policy statement will be achieved. The pollution
prevention plan builds on the company's policy statement by:
creating a list of waste streams and their point sources;
identifying opportunities for pollution prevention;
evaluating and prioritizing waste reduction options;
developing an implementation strategy for options that are feasible;
creating a timetable for pollution prevention implementation; and
detailing a plan for measuring and evaluating pollution prevention and waste reduction
progress.
The plan is best developed with input drawn from the experiences of a team of people
selected from levels throughout the company. The team approach provides a variety of
perspectives to pollution prevention and helps to identify pollution prevention opportunities and
methods for implementing them. Team members should include representatives from
management, supervisory personnel, and line workers who are familiar with the details of the
daily operation of the process. The direct participation of line workers in the development of the
pollution prevention plan is important since it is the employees who are responsible for
implementing the plan.
Data should be collected by performing an assessment of the process(es) being targeted.
It is not possible to develop a pollution prevention plan unless there exists good data on the rate
at which primary and ancillary materials are used and wastes are generated. Once the assessment
and data collection are complete, pollution prevention options should be evaluated and prioritized
based on their cost, feasibility of implementation, and their overall effectiveness of eliminating or
6-4
-------
reducing waste. After an implementation strategy and timetable is established, the plan, along
with expected benefits, should be presented to the remaining company employees to
communicate the company's commitment to pollution prevention.
Once the pollution prevention plan has been finalized and implementation is ready to
begin, employees must be given the skills to implement the plan. Training programs play an
important role in educating process employees about current pollution prevention practices and
opportunities. The goal of the training program is to educate each employee on how waste is
generated, its effects on worker safety and the environment, possible methods for waste
reduction, and on the overall benefits of pollution prevention.
Employee training should begin at the time of new employee orientation, introducing
them to the company's pollution prevention plan, thus highlighting the company's dedication to
reducing waste. More advanced training focusing on process operating procedures, potential
sources of release, and pollution prevention practices already in place should be provided after a
few weeks of work or when an employee starts a new position. Retraining employees
periodically will keep them focused on the company's goal of pollution prevention.
Effective communication between management and employees is an important part of a
successful pollution prevention program. Reports to employees on the progress of implementing
pollution prevention recommendations, as well as the results of actions already taken, reiterate
management's commitment to reducing waste, while keeping employees informed and intimately
involved in the process. Employee input should also be solicited both during and after the
creation of the pollution prevention plan to determine if any changes in the plan are warranted.
Assigning responsibility for each source of waste is an important step in closing the
pollution prevention loop. Making individual employees and management accountable for
chemical usage and waste generated within their process or department provides incentive for
employees to reduce waste. The quantity of waste generated should be tracked and the results
reported to employees who are accountable for the process generating the waste. Progress in
pollution prevention should be an objective upon which employees will be evaluated during
performance reviews, once again emphasizing the company's commitment to waste reduction.
Employee initiative and good performance in pollution prevention areas should be
recognized and rewarded. Employee suggestions that prove feasible and cost effective should be
implemented and the employee recognized either with a company commendation or with some
kind of material award. These actions will ensure continued employee participation in the
company's pollution prevention efforts.
Implementing an activity-based or total cost accounting system will identify the costs of
waste generation that are typically hidden in overhead costs by standard accounting systems.
These cost accounting methods identify cost drivers (activities) within the manufacturing process
and assign the costs incurred through the operation of the process to the cost drivers. By
identifying the cost drivers, manufacturers can correctly assess the true cost of waste generation
and the benefits of any pollution prevention efforts.
6-5
-------
The International Standards Organization has developed the ISO 14001 standard which
defines specific Environmental Management System (EMS) criteria for certification by the
organization. Although the standard has been recently established, many companies are already
seeking certification to demonstrate their commitment to environmental performance. More
information on the ISO environmental standards can be found at the ISO's website:
.
An alternative to the ISO 14001 model for EMS is the DfE EMS. It is based on the
structure outlined in the ISO 14001 standard and incorporates the five phases of Commitment,
Policy, Planning, Implementation, Evaluation and Review. While generally consistent with the
ISO 14001 standard, the DfE EMS places less emphasis on management infrastructure and
documentation and more emphasis on pollution prevention and risk reduction. The DfE EMS is
designed for small- and medium-sized businesses and provides technical guidance and detailed
methods for developing an EMS. The DfE EMS allows a company to create a simple yet
effective EMS aimed at improving environmental performance by focusing on substitutes
assessments, chemical risk reduction, pollution prevention opportunities, and resource and cost
savings. DfE has developed an EMS guidance manual and several assessment tools that are
available on the DfE EMS website: .
6.1.2 Materials Management and Inventory Control
Materials management and inventory control focuses on how chemicals and materials
flow through a facility in order to identify opportunities for pollution prevention. A proper
materials management and inventory control program is a simple, cost effective approach to
preventing pollution. Table 6-2 presents materials management and inventory control methods
that can be used to prevent pollution.
Table 6-2. Materials Management and Inventory Control Pollution Prevention Practices
Practice
Minimize the amount of chemicals kept on the
floor at one time.
Manage inventory on a first-in, first-out basis.
Return unused chemicals to inventory.
Centralize responsibility for storing and distributing
chemicals.
Store chemical products in closed, clearly marked
containers.
Use a pump to transfer chemical products from
stock to transportation container.
Benefits
Provides incentives to employees to use less
chemicals.
Reduces materials and disposal costs of expired
chemicals.
Reduces chemical and disposal costs.
Provides incentives to employees to use less
chemicals.
Reduces materials loss; increases worker safety
by reducing worker exposure.
Reduces potential for accidental spills; reduces
worker exposure.
6-6
-------
Controlling inventory levels and limiting access to inventory are widely used practices in
the PWB manufacturing industry (82.7 percent of Pollution Prevention Survey respondents).
Keeping track of chemical usage and limiting the amount of chemicals on the process floor
provides process operators an incentive to use the minimum quantity of chemical required to do
the job. Using chemicals on a first-in/first-out basis reduces the time chemicals spend in storage
and the amount of expired chemicals that are disposed. Some companies have contracted with a
specific chemical supplier to provide all of their process chemicals and manage their inventory.
In exchange for the exclusive contract, the chemical supplier assumes many of the inventory
management duties including managing the inventory, material safety data sheets (MSDSs),
ordering the chemicals, distributing the chemicals throughout the plant, and disposing of spent
chemicals and packaging (Brooman, 1996).
Chemical storage and handling practices also provide pollution prevention opportunities.
Ensuring that all chemical containers are kept closed when not in use minimizes the amount of
chemical lost through evaporation or volatilization. When transferring chemicals from container
to container, utilizing a hand pump can reduce the amount of chemical spillage. These simple
techniques not only result in less chemical usage representing a cost savings, but also result in
reduced worker exposure and an improved worker environment.
6.1.3 Material Selection
Often times, decreasing the amount of pollution a particular process generates can be as
simple as selecting different materials for use in the process. This could include primary materials
such as bath chemicals or ancillary materials such as racks or rack coverings, and is dependent
upon the availability of alternatives to the currently chosen material.
For example, the selection of the proper flux can greatly reduce the air emissions from the
hot air solder leveling (HASL) process. In the HASL process, the boards are immersed in a bath
of flux followed by submersion in a bath of solder mixed with oil. A hot air knife is then utilized
to remove excess solder and oil from the board. An air emission is created during these steps that
is the result of the bath chemicals being heated to fairly high temperatures (e.g., 450°F for the oil
and solder mixture) and both the oil and flux having vapor pressures that when heated encourage
a portion to evaporate and condense as fine droplets (Lee, 1999).
Most flux manufacturers fabricate multiple types of flux for use in the many different
environments that exist in PWB manufacturing, some producing as many as 30 to 40 different
fluxes. Each flux is manufactured to work most effectively in a particular environment (e.g., low
viscosity, high acidity). Carefully choosing the right flux for a particular PWB application can
reduce flux losses, the subsequent emissions generated, and the associated costs.
Another example would include choosing the most appropriate type of racking system
surface material. With several different types of racking system materials available (e.g.,
aluminum, iron, stainless steel, plastic, rubber-coated), the unnecessary build-up of bath
chemicals on the racks can be reduced. For instance, the use of plastic racks can prevent the
deposition of metal on the racks in plating baths, eliminating the need to strip them, thereby
reducing the amount of time, effort, and cost that goes into rack cleaning.
6-7
-------
6.1.4 Process Improvements
Improving the efficiency of a production process can significantly reduce waste
generation at the source. Process improvements include process or procedural changes in
operations carried out by employees, process equipment modification or automation, and
redesign of the process altogether. Process improvements that lead to pollution prevention in
surface finishing are categorized by the following goals:
• extend chemical bath life;
• reduce air emissions;
• reduce water consumption;
• improve process efficiency through automation; and
• segregate waste streams to reduce sludge generation.
Pollution prevention through process improvement does not always have to be expensive.
In fact, some of the most cost effective pollution prevention techniques are simple, inexpensive
changes in production procedures. Process improvements that help achieve the goals listed
above, along with their benefits, are discussed in detail in the sections below.
Extend Chemical Bath Life
The surface finishing process involves the extensive use of chemicals, many of which are
costly and pose a hazard to human health and the environment. Improvements in the efficient
usage of these chemicals can occur by accomplishing the following:
reducing chemical bath contamination;
reducing chemical bath drag-out; and
improving bath maintenance.
Inefficiencies in the use of chemicals can result in increased chemical usage, higher
operating costs, increased releases to the environment, and increased worker exposure.
Techniques to improve the efficient use of chemicals by the surface finishing and other PWB
process steps are discussed in detail below.
Reduce Bath Contaminants. The introduction of contaminants to a chemical bath will
affect its performance and significantly shorten the life of the chemical bath. Bath contaminants
include chemicals dragged in from previous chemical baths, chemical reaction by-products, and
particulate matter which may be introduced to the bath from the air. Process baths are replaced
when impurities reach a level where they degrade product quality to an unacceptable level. Any
measure that prevents the introduction of impurities will not only result in better bath
performance, but also will reduce chemical usage and generate less waste. Table 6-3 presents
pollution prevention methods for reducing bath contamination.
6-8
-------
Table 6-3. Pollution Prevention Practices to Reduce Bath Contaminants
Practices
Improve the efficiency of the water rinse system.
Use distilled or deionized water during chemical
bath make-up.
Maintain and rebuild panel racks.
Clean process tanks efficiently before new bath
make-up.
Utilize chemical bath covers when process baths
are not in operation.
Remove immediately foreign objects that have
fallen into chemical tank.
Filter contaminants continuously from process
baths.
Benefits
Rinses off any residual bath chemistries and
dislodges any particulate matter from panels and
racks.
Reduces chemical contamination resulting from
water impurities.
Prevents the build-up of deposits and corrosion
that can dislodge or dissolve into chemical baths.
Prevents contamination of the new bath from
residual spent bath chemistries.
Reduces the introduction of unwanted airborne
particulate matter; prevents evaporation or
volatilization of bath chemistries.
Prevents the contamination and premature
degradation of bath chemicals.
Prevents the build-up of any contaminants.
Thorough and efficient water rinsing of process panels and the racks that carry them is
crucial to preventing harmful chemical drag-in and to prolonging the life span of the chemical
baths. The results of the PWB Workplace Practices Questionnaire indicate that nearly every
chemical bath in the surface finishing process is preceded by at least one water rinse tank.
Improved rinsing can be achieved by using spray rinses, panel and/or water agitation, warm
water, or by several other methods that do not require the use of a greater volume of water. A
more detailed discussion of these methods is presented in the reduced water consumption
portion in this section.
A rack maintenance program is also an important part of reducing chemical bath
contamination and is practiced by 87 percent of the respondents to the Pollution Prevention
Survey. By cleaning panel racks regularly and replacing corroded metal parts, preferably with
parts of plastic or stainless steel, chemical deposition and build-up can be minimized.
Respondents to the PWB Workplace Practices Questionnaire typically perform rack cleaning
using either a chemical process that is either part of the process or a separate acid bath, or a
mechanical method. Mechanical methods, such as peeling or filing away the majority of any
metal deposits before applying a weak acid solution, can be used to prevent pollution by reducing
the quantity of acid required. An added benefit is that the reclaimed metal can be sold or reused
in the process.
According to the PWB Workplace Practices Questionnaire, 42 percent of the respondents
reported using bath covers on at least some of their baths during periods when the surface
finishing process was not operating. Respondents were not specifically questioned about the
other methods for reducing bath contamination described above; consequently, no information
was collected.
6-9
-------
Chemical Bath Drag-Out Reduction. The primary loss of bath chemicals during the
operation of the surface finishing process comes from chemical bath drag-out. This loss occurs
as the rack full of panels is being removed from the bath, dragging with it a film of chemical
solution still coating the panels. The drag-out is then either removed from the panels by a hot air
knifing process, which uses air to remove excess chemical solution retained on the boards, or is
simply carried into the next bath. In most cases, the panels are deposited directly into the next
process bath without first being air knifed.
As an extension of the making holes conductive and surface finishing DfE projects, a
mathematical tool was developed to help predict the volume of bath chemistry lost through panel
drag-out. The model identifies multiple process parameters (e.g., number of through holes, size
of panel, length of drip time, etc.) and bath characteristics (e.g., bath temperature, viscosity, etc.)
that directly affect the volume of drag-out. Process data for the model were obtained from the
PWB Workplace Practices Questionnaire and from data provided by individual chemical
suppliers. Because the primary daily loss of bath chemistry is through drag-out, using the model
to minimize drag-out will result in extended bath life, decreases in rinse water and bath chemistry
usage, and a reduction in treatment sludge. The drag-out model along with a complete
description of the method of development, individual factors in the model, and the model
limitations is presented in Appendix E. Drag-out model results for the surface finishing
alternatives are presented in Section 3.2, Exposure Assessment.
Techniques that minimize bath drag-out also prevent the premature reduction of bath
chemical concentration, extending the useful life of a bath. In addition to extended bath life,
minimizing or recovering drag-out losses also has the following effects:
• minimizes bath chemical usage;
• reduces the quantity of rinse water used;
• reduces chemical waste;
• requires less water treatment chemical usage; and
• reduces overall process cost.
Methods for reducing or recovering chemical bath drag-out are presented in Table 6-4 and
discussed below.
The two most common methods of drag-out control employed by respondents to the
Pollution Prevention Survey that require no capital investment are increased panel drainage time
(76.3 percent) and practicing slow rack withdrawal from process tanks (60.5 percent). Increasing
the time allowed for the panels to drain over the process bath allows a greater percentage of
potentially removable chemicals to remain in the bath. Practicing slow rack withdrawal during
rack removal is another step used relatively often to allow more time for the bath chemicals to
drip back into the bath. Neither of these techniques requires capital investment and both are
effective methods for reducing drag-out.
6-10
-------
Another viable option is to use drip shields, which are plastic panels that extend the wall
height of the process tank. Drip shields are inexpensive, effective drag-out control options, and
require no space between process steps, making them very practical where process space is an
issue.
Much of the chemical solution lost to drag-out can be recovered through the use of either
static drag-out tanks or drip tanks. A static drag-out tank is a batch water bath that immediately
follows the process bath from which the drag-out occurs. The panels are submerged and agitated
in the static rinse water, washing the residual chemicals from the panel's surface. When
sufficiently concentrated, the rinse water and chemical mixture can be used to replenish the
original bath. Drip tanks are similar to static drag-out tanks except that they contain no water.
The drip tank collects chemical drag-out which can then be returned to the process bath. Static
drag-out tanks are most suitably used in conjunction with heated process baths which lose water
by evaporation, requiring frequent replacement.
Table 6-4. Methods for Reducing Chemical Bath Drag-Out
Methods
Remove panels slowly from process baths.
Increase panel drainage time over process
bath.
Agitate panels briefly while draining.
Install drain boards.
Install drip shields between process baths.
Add static drag-out tanks/drip tanks to
process line where needed.
Utilize non-ionic wetting agents in the
process bath chemistries.
Utilize air knives directly after process bath
in conveyorized system. a
Employ fog rinses/spray rinses over heated
baths. a
Benefits
Reduces the quantity of residual chemical on panel
surfaces.
Allows a greater volume of residual bath chemistries to
drip from the panel back into the process bath.
Dislodges trapped bath chemistries from drilled through
holes.
Collects and returns drag-out to process baths.
Prevents bath chemical loss due to splashing.
Recovers chemical drag-out for use in bath replenishment.
Reduces surface tension of bath solutions, thereby
reducing residual chemicals on panel surfaces.
Blows residual process chemistries from process panels
which are recaptured and returned to process bath.
Rinses drag-out from the panels as they are removed from
the solution.
May not be a viable pollution prevention technique unless system is fully enclosed to prevent worker exposure to
bath chemicals introduced to the air.
Bath Maintenance Improvements. The surface finishing processes and other wet
chemistry processes in PWB manufacturing consist of a complex, carefully balanced series of
formulated chemical mixtures, each one designed to operate at specific conditions, working
together to perform an overall function. A bath testing and control program is essential in
preventing the chemical breakdown of process baths, thus extending their useful lives and
preventing their premature disposal. The premature disposal of process chemistries results in
increased chemical costs for both bath and treatment chemicals, prolonged process down-time,
and increased process waste.
6-11
-------
Bath maintenance, or control, refers to maintaining a process bath in peak operating
condition by identifying and controlling key operating parameters, such as bath temperature,
individual chemical concentrations, pH, and the concentration of contaminants. Proper control
of bath operating parameters will result in more consistent bath operation, less water usage, and
better, more consistent quality of work.
According to Pollution Prevention Survey respondents, the majority of PWB
manufacturing facilities (72.4 percent) have a preventative bath maintenance program already in
place. Typical bath maintenance methods and their benefits are presented in Table 6-5 below.
Table 6-5. Bath Maintenance Im
Methods
Monitor bath chemistries by testing
frequently.
Replace process baths according to chemical
testing.
Maintain operating chemical balance through
chemical additions according to testing.
Filter process baths continuously.
Employ steady state technologies.
Install automated/statistical process control
system.
Utilize temperature control devices.
Utilize bath covers.
provement Methods To Extend Bath Life
Benefits
Determines if process bath is operating within
recommended parameters.
Prevents premature chemical bath replacement of good
process baths.
Maintains recommended chemical concentrations
through periodic chemical replenishment as required.
Prevents the build-up of harmful impurities that may
shorten bath life.
Maintains steady state operating conditions by filtering
precipitates or regenerating bath solutions continuously.
Provides detailed analytical data of process operating
parameters, facilitating more efficient process operation.
Regulates bath temperatures to maintain optimum
operating conditions.
Reduces process bath losses to evaporation and
volatilization.
Frequent monitoring and adjustment of the various chemical concentrations within a
process bath are the foundations on which a good bath maintenance program is built.
Monitoring is done by regularly testing the bath concentrations of key chemicals to ensure that
the bath is chemically balanced. If chemical concentrations are outside of the operating levels
recommended by the supplier, a volume of chemical is added to the bath to bring it back into
balance. When the concentration of contaminants reaches an established critical level, or some
other criterion provided by the supplier, the bath is disposed of and replaced with a new bath.
Bath testing and adjustment can be performed manually or with an automated system
that can perform both functions. Either way, controlling the bath through regular testing and
bath additions is an inexpensive, effective method for extending bath life and reducing pollution.
Nearly all of the PWB facilities surveyed (93.1 percent) reported testing chemical bath
concentrations, adding chemicals as necessary and maintaining records of the analysis and
additions.
6-12
-------
Bath replacement should be based upon chemical testing, instead of some other
predetermined criteria. Predetermined criteria, such as times or production volumes, are often
given by suppliers as safe guidelines for bath replacement for facilities that do not regularly test
their process baths. These criteria are conservative estimates of the effective life of the process
bath, but possibly could be exceeded with a proper bath testing and maintenance program. By
replacing the process bath only when chemical testing indicates it is required, bath life can be
extended while chemical usage and waste are reduced. Most (95.0 percent) of the surveyed PWB
facilities reported replacing their process baths only when testing indicated.
The build-up of contaminants in a process bath will eventually require the bath to be
replaced. Bath contaminants can be solid matter, such as paniculate matter and precipitates, or
undesired chemical species in solution, such as reaction by-products or drag-in chemicals.
Installing standard cartridge or bag filters to continuously remove solid impurities from the bath
is an inexpensive, yet effective method to extend bath life.
Additionally, some baths may be maintained at steady state conditions using readily
obtainable systems capable of regenerating or filtering process bath chemistries. Although these
systems may require capital investment, maintaining steady state conditions keeps a bath within
the optimal operating conditions resulting in extended bath life and increased cost savings
(Edwards, 1996).
Statistical process control (SPC) is a method of analyzing the current and past
performance of a process bath, using chemical testing results and operating condition records to
optimize future bath performance. SPC will lead to more efficient bath operation and extended
bath life by indicating when a bath needs maintenance through the tracking and analysis of
individual operating parameters and their effect on past performance (Fehrer, 1996).
A method of limiting evaporative losses from process baths is to cover the surface of the
solution with floating plastic balls that will not react with the process solution. The plastic balls,
similar to ping pong balls which do not interfere with the work pieces being processed, prevent
the evaporation of the bath solution by limiting the surface area of solution exposed to the air.
Hexagonal shaped balls are now available that leave even less surface area exposed to the air
(Brooman, 1996). This method is especially effective for higher temperature process baths where
evaporative losses tend to be high. This method is inexpensive, easy to utilize, and will decrease
the air emissions from the bath, limiting the amount of operator exposure to the chemicals.
Reduce Air Emissions
During surface finishing, air emissions are generated from some chemical baths. When
the chemicals being used pose a hazard to human health, hoods are utilized to collect the
emissions and move them away from the workers. These emissions are ducted to air emission
control devices as necessary. These emissions increase the costs associated with PWB
manufacture, thus efforts that reduce these emissions not only produce cost savings but reduce
worker exposure and reduce the environmental impacts of the process.
6-13
-------
One particularly troublesome source of air emissions during the HASL process is the
application of a flux and a subsequent solder to the PWB, which generates air emissions that can
include oil mist, oxides of lead and tin, hydrogen chloride or hydrogen bromide, and copper
chloride or copper bromide (chlorine or bromine is typically used as the flux activator). This
process typically requires pollution control equipment like a wet scrubber followed by a
diffusion-type fiber bed filter, to control not only the pollutants but also the odors created by
their release.
The most prominent option available to reduce these HASL process air emissions comes
in the form of process redesign, or utilizing an alternative surface finishing (ASF) technology.
Although most of the ASF technologies being evaluated in this CTSA also have air emissions of
one type or another, it is the current understanding that one or more will offer a reduction in the
overall quantity and/or toxicity of the air emissions generated while maintaining product quality
and performance criteria. Depending on the characteristics of the particular boards needing
surface finishing (e.g., their aspect ratio), an ASF technology might provide performance either
similar to or better than the HASL process while reducing the surface finishing process'
environmental impacts.
Reduced Water Consumption
Contaminated rinse water is one of the primary sources of heavy metal ions discharged to
waste treatment processes from the surface finishing process and other wet chemistry process
lines (Bayes, 1996). These contaminants, which are introduced to the rinse water through
chemical drag-out, must be treated and removed from the water before it can be reused in the
process or discharged to the sewer. Because rinsing is often an uncontrolled portion of the
process, large quantities of water are consumed and treated unnecessarily. Reducing the amount
of water used by the surface finishing process has the following benefits:
• decreases water and sewage costs;
• reduces wastewater treatment requirements, resulting in less treatment chemical usage
and reduced operating costs;
• reduces the volume of sludge generated from wastewater treatment, which results in
reduced sludge treatment or disposal costs; and
• improves opportunities to recover process chemicals from more concentrated waste
streams.
The surface finishing process line consists of a series of chemical baths, which are
typically separated by at least one, and sometimes more, water rinse steps. These water rinse
steps account for virtually all of the water used during the operation of the surface finishing line.
The water baths act as a buffer, dissolving or displacing any residual drag-in chemicals from the
panel surface. The rinse baths prevent contamination of subsequent baths while creating a clean
surface for future chemical activity.
Improper rinsing not only leads to shortened bath life through increased drag-in, as
discussed previously, but can also lead to a host of problems affecting product quality, such as
peeling, blistering, and staining. Insufficient rinsing of panels can lead to increased chemical
6-14
-------
drag-in quantities and will fail to provide a clean panel surface for subsequent chemical activity.
Excessive water rinsing, done by exposing the panels too long to water rinsing, can lead to
oxidation of the copper surface and may result in peeling, blistering, and staining. To avoid
insufficient rinsing, manufacturers often use greater water flow rates than are necessary, instead
of using more efficient rinsing methods that reduce water consumption but may be more
expensive to implement. These practices were found to be true among survey respondents,
where facilities with low water and sewage costs typically used much larger amounts of water
than comparable facilities with high water and sewer costs.
Many techniques are available that can reduce the amount of water consumed while
rinsing. These techniques are categorized by the following:
• methods to control water flow;
• techniques to improve water rinse efficiency; and
• good housekeeping practices.
Flow control methods focus on controlling the flow of water, either by limiting the
maximum rate that water is allowed to flow into the rinse system, or by stopping and starting the
water flow as it is needed. These methods seek to limit the total water usage while ensuring that
sufficient water is made available to cleanse the PWB panels. Examples of these techniques
include the use of flow restrictors or smaller diameter piping to limit the maximum flow of water,
and control valves that provide water to the rinse baths only when it is needed. Control valves
can be either manually operated by an employee, or automated using some kind of sensing
device such as conductivity meters, pH meters, or parts sensors. All of the methods are effective
water reduction techniques that can be easily installed.
Pollution prevention techniques directed at improving water efficiency in the rinse system
seek to control or influence the physical interaction between the water and the panels. This can
be done by increasing bath turbulence, improving water quality, or by using a more efficient rinse
configuration. All of these methods, discussed below, seek to improve rinsing performance while
using less water.
Increasing bath turbulence can be accomplished through the use of ultrasonics, panel
agitation, eductors (nozzles below the surface that circulate solution), or air sparging. All of these
agitation methods create turbulence in the bath, increasing contact between the water and the
part, thereby accelerating the rate that residual chemicals are removed from the surface. Agitating
the bath also keeps the water volume well mixed, distributing contaminants throughout the bath
and preventing concentrations of contaminants from becoming trapped. However, agitating the
bath can also increase air emissions from the bath unless pollution prevention measures are used
to reduce air losses.
Water quality can be improved by using distilled or deionized water for rinsing instead of
tap water that may include impurities such as carbonate and phosphate precipitates, calcium,
fluoride, and iron. Finally, utilizing more efficient rinse configurations such as countercurrent
rinse stages, spray rinses, or fog rinses will increase the overall efficiency of the surface finishing
rinse system while reducing the volume of wastewater generated. PWB manufacturers often use
6-15
-------
multiple rinse water stages between chemical process steps to facilitate better rinsing. The first
rinse stage removes the majority of residual chemicals and contaminants, while subsequent rinse
stages remove any remaining chemicals. Counter-current or cascade rinse systems minimize
water use by feeding the water effluent from the cleanest rinse tank, usually at the end of the
cascade, into the next cleanest rinse stage, and so on, until the effluent from the most
contaminated, initial rinse stage is sent for treatment or recycle.
Good housekeeping practices focus on keeping the process equipment in good repair and
fixing or replacing leaky pipes, pumps, and hoses. These practices can also include installing
devices such as spring loaded hose nozzles that shut off when not in use, or water control timers
that shut off water flow in case of employee error. These practices often require little investment
and are effective in preventing unnecessary water usage. For a more detailed discussion on
methods of improving water rinse efficiency and reducing water consumption, refer to Section
5.1, Resource Conservation.
Improve Process Efficiency Through Automation
The operation of the surface finishing process presents several opportunities for important
and integral portions of the process to become automated. By automating important functions,
operator inconsistencies can be eliminated, allowing the process to be operated more efficiently.
Automation can lead to the prevention of pollution by:
• gaining a greater control of process operating parameters;
• performing the automated function more consistently and efficiently;
• eliminating operator errors; and
• making the process compatible with newer and cleaner processes designed to be operated
with an automated system.
Automating a part of the surface finishing process can be expensive. The purchase of
some automated equipment can require a significant initial investment, which may prevent small
companies from automating. Other costs that may be incurred include those associated with
installing the equipment, training employees, any lost production due to process down-time, and
redesigning other processes to be compatible with the new system. Although it may be
expensive, the benefits of automation on productivity and waste reduction will result in a more
efficient process that can save money over the long run.
Installation of automated equipment such as a rack or panel transportation system,
chemical sampling equipment, or an automated system to make chemical additions can have a
major impact on the quantity of pollution generated during the day-to-day operation of the
surface finishing process and can also reduce worker exposure. Surface finishing process steps
or functions that can be automated effectively include:
• rack transportation;
• bath maintenance; and
• water flow control.
6-16
-------
Rack transportation systems present an excellent opportunity for automation, due to the
repetitive nature of transporting panel racks. Various levels of automation are available ranging
from a manually operated vertical hoist to a computer controlled robotic arm. All of these
methods allow for greater process control over panel movement through the surface finishing
process line. By building in drag-out reduction methods such as slower panel withdrawl and
extended drainage times into the panel movement system, bath chemical loss and water
contamination can be greatly reduced.
Automating bath maintenance testing and chemical additions can result in longer bath life
and reduced waste. These systems monitor bath solutions by regularly testing bath chemistries
for key contaminants and concentrations. The system then adjusts the process bath by making
small chemical additions, as needed, to keep contaminant build-up to a minimum and the process
bath operating as directed. The resulting process bath operates more efficiently, resulting in
prolonged bath life, less chemical waste, reduced chemical cost, and reduced drag-out.
Controlling rinse water flow is an inexpensive process function to automate. Techniques
for controlling rinse water flow were discussed previously. The reduction in fresh water usage as
a result of automating these techniques will not only reduce water costs, but will also result in
reduced treatment chemical usage and less sludge.
A conveyorized system integrates many of the methods described above into a complete
automated surface finishing system. The system utilizes a series of process stages connected by
a horizontal conveyor to transport the PWB panels through the surface finishing process. Drag-
out is greatly reduced due, in part, to the separate process stages, and to the vertical alignment of
the drilled holes that trap less chemicals. Since drag-out is reduced, much less rinse water is
required to cleanse the panel surfaces, resulting in reduced water and treatment costs. A single
water tank is sufficient between process baths, whereas multiple stages may be required in a non-
conveyorized process. Thus, automation dramatically reduces the number of process stages
required, resulting in a much shorter cycle time and reduced floor space requirements. The
enclosed process stages limit evaporative losses, reducing chemical costs, while also reducing the
amount of chemical to which an employee is exposed. Several surface finishing alternative
chemistry processes have been designed to operate effectively using this type of convey orized
system.
A convey orized system should also take advantage of other pollution prevention
techniques, such as water flow controllers, bath maintenance techniques and other methods
discussed throughout this module, to further reduce waste. By integrating all of these methods
together into a single surface finishing system, the process operates more efficiently, reducing
water and chemical consumption, resulting in less process waste and employee exposure.
6-17
-------
Segregate Wastewater Streams to Reduce Sludge Generation
The segregation of wastewater streams is a simple and cost effective pollution prevention
technique for the surface finishing process. In a typical PWB facility, wastewater streams from
different process steps are often combined and then treated by an on-site wastewater treatment
process to comply with local discharge limits.
Some waste streams from the surface finishing process, however, may contain chelating
agents. These chelators, which permit metal ions to remain dissolved in solution at high pH
levels, must first be broken down chemically before the waste stream can be treated and the
heavy metal ions removed. Treatment of waste containing chelators requires extra treatment
steps or more active chemicals to break down the chelating agents and precipitate out the heavy
metal ions from the remaining water effluent. Because the chelator-bearing streams are
combined with other non-chelated streams before being treated, a larger volume of waste must be
treated for chelators than is necessary, which also results in a larger volume of sludge.
To minimize the amount of treatment chemical used and sludge produced, the chelated
waste streams should be segregated from the other non-chelated wastes and collected in a storage
tank. When enough waste has been collected, the chelated wastes should be batch treated to
breakdown the chelator and remove the heavy metals. The non-chelated waste streams can then
be treated by the on-site wastewater treatment facility without additional consideration. By
segregating and batch treating the chelated heavy metal wastes from other non-hazardous waste
streams, the volume of waste undergoing additional treatment is minimized and treatment
chemical usage and sludge generation reduced.
6-18
-------
6.2 RECYCLE, RECOVERY, AND CONTROL TECHNOLOGIES ASSESSMENT
While pollution prevention is the preferred method of waste management, the pollution
prevention hierarchy recognizes that pollution prevention is not always practical. Companies
often supplement their pollution prevention efforts with additional waste management techniques
to further reduce emissions. These techniques, presented in order of preference, include
recycling or reclamation, treatment, and disposal. Techniques for pollution prevention are
presented in Section 6.1. This section presents waste management techniques typically used by
the PWB industry to recycle or recover valuable process resources (Section 6.2.1), and to control
emissions to water and air (Section 6.2.2) from the surface finishing process. Typical treatment
configurations presented in this section were developed and reviewed by PWB manufacturers
participating in this project.
6.2.1 Recycle and Resource Recovery Opportunities
PWB manufacturers have begun to re-emphasize recycle and recovery technologies, due
to more stringent pretreatment effluent limits. Recycling or reclamation is the recovery of
process material effluent, either on-site or off-site, which would otherwise become a solid waste,
air emission, or would be discharged to a wastewater stream. Technologies that recycle water
from waste streams concentrate the final effluent, making subsequent treatment more efficient,
which reduces the volume of waste generated and lowers overall water and sewer costs. As a
result, these technologies are being used more frequently by industry to recycle or recover
valuable process resources, while also minimizing the volume of waste that is sent to disposal.
This trend was supported by the respondents of the Printed Wiring Board Pollution Prevention
and Control Technology: Analysis of Updated Survey Results (U.S. EPA, 1998), 81 percent of
whom reported using some type of recycle or resource recovery technology.
Recycle and resource recovery technologies include those that recover materials from
waste streams before disposal, or recycle waste streams for reuse in another process.
Opportunities for both types of technologies exist within a surface finishing process. Rinse water
can be recycled and reused in further rinsing operations, while valuable metals such as copper,
silver, palladium, and gold can be recovered from waste streams before disposal and sold to a
metals reclaimer. These recycle and recovery technologies may be either in-line (dedicated and
built into the process flow of a specific process line) or at-line (employed at the line as desired, as
well as at other places in the plant), depending on what is required (Brooman, 1996). Each waste
stream that cannot be prevented should be evaluated to determine its potential for effective
recycle or resource recovery as part of a pollution prevention and waste management plan.
The decision of whether to purchase a recycle or resource recovery process should be
based on several factors. Economic factors, such as process operating and effluent disposal costs
for the current system, must be compared with those estimated for the new technology. The
initial capital investment of the new technology, along with any potential cost savings, and the
length of the payback period must also be considered. Other factors such as the characteristics of
the waste stream(s) considered for treatment, the ability of the process to accept reused or
6-19
-------
recycled materials, and the effects of the recycle or recovery technology on the overall waste
treatment process also should be considered.
The entire PWB manufacturing process must be considered when assessing the economic
feasibility of a recycle or resource recovery process. An individual recovery process can recover
metal from a single stream originating from a surface finishing process, or it may recover the
metal from streams that originate from other processes, as well. Only by considering the new
technology's impact on the entire PWB manufacturing process can an accurate and informed
decision be made. While this section focuses on technologies that may be used to recycle or
recover resources from the waste streams that are generated by the surface finishing processes,
many of these technologies are also applicable to other PWB process lines. Workplace practices
that can lead to the recycle or reuse of resources (e.g., manually recovering copper from panel
racks, water recycle using cascade water rinse systems) are discussed in Section 6.1.
Solder Recycling
The application of solder to the surface of PWBs by HASL has been the industry
standard finish for many years. The process has long been considered to provide a reliable finish
which facilitates assembly and introduces few defects. However, as the concentration of
impurities in the solder increases to above 0.3 percent by weight, the quality and appearance of
the applied solder finish deteriorates. The solder begins to appear grainy and takes on a dull gray
color.
The primary impurity is copper, which is introduced to the molten solder as a by-product
of the process reaction. Tin from the molten solder is exchanged with the copper ions on the
surface of the exposed copper pads, forming a tin-copper intermetallic layer upon which the
solder can adhere. The displaced copper ions remain in the molten solder as a contaminant
where they build in concentration until the contamination begins to affect the quality of the
solder deposit.
To restore the HASL process to optimum operating conditions, the solder pots typically
are refreshed to reduce the contaminant concentration. This maintenance process is performed
with the solder in molten form by discarding a substantial quantity of the contaminated solder
and replacing it with fresh solder. The contaminated solder (a.k.a. solder dross) can be returned
to a recycler to be reclaimed for credit. The effectiveness of the dilution is dependent on the
amount of solder replaced, with a 40 percent by weight replacement of solder resulting in roughly
a 33 percent decrease in copper contamination, dropping the concentration from 0.3 to 0.2
percent copper. This process is repeated as required to maintain operation of the HASL process
(Fellman, 1997).
Solder skimming is another method of purifying the solder. The solder is cooled to a
temperature just above the melting point (360 °F), causing the copper impurities to become
insoluble. The copper-tin needles which form are then skimmed from the surface of the solder
and handled as waste. Because only the impurities are removed, along with a minimal amount of
solder, the skimming process results in much less solder usage over time. However, this method
6-20
-------
requires open access to the molten solder pot to perform the skimming, so it is typically only
associated with some vertical, non-conveyorized HASL machines.
A solder saver, or solder reclaim system, will purify the solder in HASL machines where
access to the solder is restricted by air knives, rollers, pumps, or some other equipment, such as
in some vertical HASL machines and nearly all horizontal, conveyorized HASL machines. A
solder reclaim system diagram is shown in Figure 6-1. The solder saver continuously siphons a
portion of the molten solder from the HASL machine to a separate solder pot, where the
temperature is lowered and the impurities are skimmed from the solder.
Process Work - PWBs
Purified Solder
Dross to
Reclaim
Figure 6-1. Solder Reclaim System Diagram
Impurities that have been skimmed from the solder are collected in a compartment of the
machine for removal and disposal, and an equal weight of fresh solder is manually added to
maintain operating solder levels. Transfer of the solder from the pot takes place in a heated pipe
to prevent the solder from solidifying during the transfer process. The purified portion of the
solder is then pumped back through a second
heated pipe to the HASL solder pot. The
solder reclaim system is an off-line system that
operates continuously, without disrupting the
operation of the HASL process.
One study found that approximately
96 percent of the solder was retained after
skimming with a solder reclaim system,
resulting in a remaining copper concentration
of 0.16 percent, or a purification efficiency of
better than 90 percent (Fellman, 1997). One
PWB manufacturer reported a yearly decrease
of 86 percent in solder consumption (see
inset), decreasing their overall lead usage to
below reportable levels (Sharp, 1999).
Solder Recovery Case Study of PWB
Manufacturer
Before Solder Reclaim:
• 202,000 Ibs solder usage/year
• 75,000 Ibs lead usage reported
After Solder Reclaim:
• 27,000 Ibs solder usage/year
• Lead usage below reportable level
(less than 10,000 Ibs)
Cost Comparison:
• Net cost of solder $0.50/lb ($2.10/lb solder -
$1.60/lb dross reclaim credit)
• Total solder usage reduction of 175,000 Ibs/yr
• Total cost savings of $82,000/yr
• Equipment cost is $70,000
• Payback period is approximately one year
6-21
-------
The average capital cost of a solder reclaim unit was reported to be $60,000 to $80,000. A
cost analysis performed by one large PWB manufacturer found the expected payback period for
this equipment to be one year, based upon an annual solder usage of 200,000 pounds per year,
prior to the installation of the equipment.
Electrolytic Recovery
Electrolytic recovery, also known as electrowinning, is a common metal recovery
technology employed by the PWB industry. Electrowinning uses an electrolytic cell to recover
dissolved metal ions from solution. Operated either in continuous or batch mode, electrowinning
can be applied to various process fluids including spent microetch, drag-out rinse water, and ion
exchange regenerant. An advantage is its ability to recover only the metal from solution, leaving
behind the other impurities that are present. The recovered metal can then be sold as scrap or
traded for credit towards future bath chemistry. Electrowinning is typically used by PWB
manufacturers to recover copper (effluent limit concerns) and gold (high price) from process
baths or rinse tanks. It can also be used to recover other metals such as tin or silver, but this is
not usually done because the metal does not exceed effluent treatment limits, or the recovery of
the metal is not economically viable. Nickel recovery using electrowinning requires close control
of pH; therefore, it is performed less frequently than for other metals, such as copper and gold
(U.S. EPA, 1998).
The electrolytic cell is comprised of a set of electrodes (cathodes and anodes) placed in
the metal-laden solution. An electric current, or voltage, is applied across the electrodes and
through the solution. The positively charged metal ions are drawn to the negatively charged
cathode where they deposit onto the surface. The solution is kept thoroughly mixed using air
agitation or other proprietary techniques, to permit the use of higher current densities (the
amount of current per surface area of cathode). These higher current densities shorten deposition
time and improve the recovery efficiency. As the metal recovery continues, the concentration of
metal ions in solution becomes depleted, requiring the current density to be reduced to maintain
efficiency at an acceptable level. When the concentration of metal becomes too low for its
removal to be economically feasible, the process is discontinued and the remaining solution is
sent to final treatment.
Electrowinning is most efficient with concentrated solutions. Dilute solutions (less than
100 mg/1 of metal) become uneconomical to treat due to the high power consumption relative to
the amount of metal recovered (Coombs, 1993). Waste streams that are to be treated by
electrowinning should be segregated, only combining streams containing the same metal, to
prevent dilution, and to create a pure metal deposit free of other metal impurities. Already diluted
solutions can be concentrated first, using ion exchange or evaporation techniques, to improve the
efficiency and cost effectiveness of metal recovery.
Process waste solutions containing chlorine ions in any form should not be processed
using electrolytic recovery methods, because the electrolysis of these solutions could generate
chlorine gas. Solutions containing copper chloride salts should first be converted to non-chloride
6-22
-------
copper salt (e.g., copper sulfate) solutions, using ion exchange methods, before undergoing
electrowinning to recover the copper content (Coombs, 1993).
The recovered metal(s) can be sold as scrap to a metals reclaimer. Typical metal removal
efficiencies of 90 to 95 percent have been achieved using electrolytic methods (U.S. EPA, 1990).
The remaining effluent will still contain small amounts of metal and will be acidic in nature (i.e.,
low pH). Adjusting the pH may not be sufficient for the effluent to meet the standards of some
POTW authorities; therefore, further treatment may be required.
Eighteen percent of the Pollution Prevention Survey respondents reported using
electrowinning as a resource recovery technology, with 89 percent of those being satisfied with its
performance. The median cost of an electrowinning unit reported by the respondents was
$15,000; however, electrowinning capital costs are dependent on the capacity of the unit.
Ion Exchange
Ion exchange is a process used by the PWB industry mainly to recover metal ions, such
as copper, tin, or palladium, from rinse waters and other solutions. This process uses an
exchange resin to remove the metal from solution and concentrate it on the surface of the resin.
It is particularly suited to treating dilute solutions, since at lower concentrations the resin can
process a greater volume of wastewater before becoming saturated. As a result, the relative
economics of the process improve as the concentration of the feed solution decreases. Aside
from recovering metals such as copper and silver, ion exchange also can be used for treating
wastewater, deionizing feed water, and recovering chemical solutions.
Ion exchange relies on special resins, either cationic or anionic, to remove the desired
chemical species from solution. Cation exchange resins are used to remove positively charged
ions such as copper, tin, or other metals. When a feed stream containing a metal is passed
through a bed of cation exchange resin, the resin removes the metal ions from the stream,
replacing them with hydrogen ions from the resin. For example, if a feed stream containing
copper sulfate (CuSO4) is passed through the ion exchange resin, the copper ions are removed
and replaced by hydrogen ions to form sulfuric acid (H2SO4). The remaining water effluent is
either further processed using an anion exchange resin and then recirculated into the rinse water
system, or pH neutralized and then directly sewered. Ion exchange continues until the exchange
resin becomes saturated with metal ions and must be regenerated.
Special chelating resins have been designed to capture specific metal ions that are in the
presence of chelating agents, such as metal ions in electroless plating baths. These resins are
effective in breaking down the chemical complexes formed by chelators that keep metal ions
dissolved in solution, allowing them to be captured by the resin. Hard water ions, such as
calcium and magnesium, are not captured, creating a purer concentrate. Chelating resins require
that the feed stream be pH-adjusted to reduce acidity, and filtered to remove suspended solids
that will foul the exchange bed (Coombs, 1993).
6-23
-------
Regeneration of the cation or chelating exchange resin is accomplished using a
moderately concentrated (e.g., ten percent) solution of a strong acid, such as sulfuric acid.
Regeneration reverses the ion exchange process by stripping the metal ions from the exchange
resin and replacing them with hydrogen ions from the acid. The concentration of metal ions in
the remaining regenerant depends on the concentration of the acid used, but typically ranges
from
10 to 40 g/1 or more (Coombs, 1993).
Ion exchange can be combined with electrowinning to recover metal from solutions that
would not be cost effective to recover using either technology alone. A typical flow diagram for
this type of system is shown in Figure 6-2. It can be used to concentrate a dilute solution of
metal ions for electrolytic recovery that would otherwise be uneconomical to recover. For
example, a dilute copper chloride solution can be treated by an ion exchange unit that is
regenerated using sulfuric acid, producing a concentrated copper sulfate solution. The
electrowinning unit can then be used to recover the copper from the solution while regenerating
the acid, which could then be used for the next regeneration cycle. The recovery of gold from the
drag-out and rinse tanks, following the immersion gold bath, is another example of where this
configuration is typically used. The high cost of gold makes this system cost effective over the
long term.
Spent Baths/
Rinses
Spent
- Regenerant-
Metal
Spent
Rfigenerant
Waste
Treatment
Figure 6-2. Flow Diagram of Combination Ion Exchange and
Electrowinning Recovery System for Metal Recovery
A benefit of ion exchange is the ability to control the type of metallic salt that will be
formed by selecting the type of acid used to regenerate the resin. In the previous example, the
copper chloride was converted to copper sulfate while being concentrated by the ion exchange
system. This is particularly useful when electrowinning is used, because it cannot process
solutions containing the chlorine ion without usually generating toxic chlorine gas.
Forty-four percent of the respondents to the Pollution Prevention Survey reported using
an ion exchange process as a water recycle/chemical recovery technology. Of these facilities, 90
percent indicated that they were satisfied with its overall performance. The average capital cost of
a unit, which is related to its capacity, was reported to be $65,000 (with a low of $10,000 and a
high of $120,000).
6-24
-------
Reverse Osmosis
Reverse osmosis (RO) is a recovery process used by the PWB industry to regenerate rinse
waters and to reclaim process bath drag-out for return to the process (U.S. EPA, 1990). It relies
on a semi-permeable membrane to separate the water from metal impurities, allowing bath
solutions to be reused. It can be used as a recycling or recovery technology to reclaim or
regenerate a specific solution, or it can be part of an overall waste treatment process to
concentrate metals and impurities before final treatment.
The semi-permeable membrane permits only certain components to pass through, and
pressure is used as a driving force to separate the components at a useful rate. The membrane is
usually made of a polymer compound (e.g., nylon) with hole sizes ranging from 0.0004 to 0.06
microns in diameter. Pumping of the waste stream, at pressures typically ranging from 300 to
1,500 pounds per square inch (psi) force the solution through the membrane (Capsule
Environmental Engineering, Inc., 1993). The membrane allows the water to pass while inhibiting
the metal ions, collecting them on the membrane surface. The concentrated metal ions are
allowed to flow out of the system, where they are reused as bath make-up solution, or they are
sent to treatment. The relatively pure water can be recycled as rinse water or directly sewered
(sent to a POTW).
A typical RO system for recycling rinse water is shown in Figure 6-3. The effluent from
rinse water tanks throughout the facility is collected in a conditioning tank. Any pretreatment
that may be required, such as pH adjustment, takes place in the conditioning tank. The
conditioning tank also acts to smooth out any chemical concentration spikes that may occur in
the rinse effluent. The water is then passed through the RO membrane, where the metals and
other dissolved solids are removed. The purified water is then passed on to a storage tank to be
used for further rinsing operations, where required. The removed solids and other materials are
sent to the wastewater treatment system to be processed. An RO system of this design will have
an efficiency of 70 to 85 percent, with the remainder being sent to waste treatment (Hosea, 1998).
Rinse
Water
Effluent
Co
>
<£-
HIS
nditioning
Tank
>.
RO
Unit
up ^
>,
Storage
Tank
^
Reusable
Rinse
Water
Waste
Treatment
Figure 6-3. Reverse Osmosis Water Reuse System
6-25
-------
The RO process has some limitations. The types of waste streams suitable for processing
are limited by the ability of the polymeric membranes to withstand the destructive nature of the
given waste stream. The membranes are sensitive to solutions with extreme pH values, either low
or high, which can degrade them. Pure organic streams likewise are not treatable. Waste streams
with suspended solids should be filtered prior to separation to keep the solids from fouling the
membrane, to avoid reducing the efficiency of the process. Process membranes also may have a
limited life due to the long-term pressure of the solution on the membrane (Coombs, 1993). Data
regarding the usage of RO technology by the PWB industry were not collected in the Pollution
Prevention Survey.
Off-Site Refining/Reclamation
Many of the surface finishing technologies are based on the deposition of precious
metals. Due to the high cost of replacement, these baths are typically recharged rather than
discarded, replacing the metal that has been plated to maintain proper operating concentrations.
Should the baths become too contaminated to operate properly, the baths are replaced with new
chemistry and the spent bath solution is sent to a chemical refinery to reclaim the value of the
remaining precious metal content. The most likely solutions to be refined to recover their value
are those containing gold and palladium. The value of the recovered metal is based on the current
spot market price of the metal. Table 6-6 lists the current value of the metal and the typical
methods of recovery.
Table 6-6. Typical Value of Reclaimed Metals (1999) and Recovery Methods
Metal
Gold
Palladium
Silver
Copper
Solder
Price a'b
$283/oz
$636/oz
$4.98/oz
$0.80/lb
$1.60/lb
Recovery Method c
Off-site refining or electrolytic
Off-site refining
Off-site refining or ion exchange
On-site electrolytic or ion exchange
Manual or solder recovery system
Metal prices received will be current market prices minus a 2 to 5 percent refining fee. Prices listed are spot prices
on 7/6/00 obtained from www.kitco.com.
b Solder cost obtained from Alpha Metals (03/00). Copper price reflects London metal exchange price on 7/6/00
obtained from www.nickelalloy.com.
0 Methods of recovery are typical methods and do not represent all recovery options.
Some chemical suppliers provide this service to their customers, accepting spent bath
solution in exchange for credit toward future chemical purchases. While the fee charged to
recover the metal from the bath is similar to that charged by a refinery service (2 to 5 percent),
PWB manufacturers may find it easier to deal with a single company to both supply bath
chemicals and to reclaim the spent bath solution, rather than contracting with a separate waste
recovery service (Schectman, 1999).
6-26
-------
The chemical supplier also benefits from providing this service, because the companies
that receive credit are more likely to continue purchasing their chemical products. Chemical
suppliers also may be able to reuse the spent solution, regenerating the stock into new bath
solution, rather than treating and discarding the remaining solution.
Both gold and palladium plating baths are routinely refined to recover the value of the
remaining metal. The value of the metals combined with the high concentration of metal ions
remaining, even in a spent bath, makes refining worthwhile. Silver plating baths do not typically
have sufficiently high concentrations of silver ions to warrant refining for economic reasons.
However, in some instances, silver baths may be combined with other silver-bearing waste
streams, such as photo developing solutions, before being refined, making it more cost effective
to recover the metal (Sharp, 1999).
Although the low recovery value of copper, tin, and nickel prevent refining from being
economically advantageous, these solutions are at times sent off-site to a reclaimer, at a cost to
the PWB manufacturer, because the facility lacks the capability to treat the solution or does not
want to deal with the extra treatment steps and risks involved. The value of the metal recovered
from the solution is credited to the PWB manufacturer, but is usually insufficient to cover the
entire expense of the refining and disposal (Schectman, 1999). These metals, particularly copper,
also can be recovered on-site using ion exchange or electrowinning, when the recovered metal
can be sold to a reclaimer to partially offset the cost of recovery.
Applicability of Recovery Technologies
Recovery and reclamation technologies typically are quite efficient, but are designed for a
specific application, which is usually chemical-specific in nature (e.g., electrowinning removes
positively charged metal ions), often limiting their applicability. Because surface finishing
processes are comprised of a series of chemical baths of different chemical characteristics, it is
appropriate to match the recovery technologies with individual chemical baths when identifying
opportunities for recycling or reclaiming materials. Table 6-7 displays the applicability of the
various recycling and recovery technologies to each of the surface finishing chemical baths. Bath
types that do not require additional recycling, are not economically feasible to recycle, or those
for which a recycling technology does not exist are not listed in the table. Recovery technologies
can sometimes be combined (e.g., ion exchange followed by electrowinning to recover metal)
into a more cost effective recovery system that achieves greater removal efficiency.
6-27
-------
Table 6-7. Applicability of Recovery/Reclamation Technologies by Bath Type
Bath Type
Drag -out Rinse
(following gold,
palladium)
Gold
Microetch
Nickel
Palladium
Immersion Silver
Solder
Immersion Tin
Water Rinse
Process(es)
Nickel/Gold and
Nickel/Palladium/Gold
Nickel/Palladium/Gold
All
Nickel/Gold and
Nickel/Palladium/Gold
Nickel/Palladium/Gold
Immersion Silver
HASL
Immersion Tin
All
Solder
Recovery
X
Ion
Exchange
X
X
X
X
X
X
Electrolytic
Recovery
X
X
X
X
X
X
Reverse
Osmosis
X
Off-Site
Refining
X
X
X
X
6.2.2 Control Technologies
If the release of a hazardous material cannot be prevented or recycled, it may be possible
to treat or reduce the impact of the release using a control technology. Control technologies are
engineering methods that minimize the toxicity and/or volume of released pollutants. Most of
these methods involve altering either the physical or chemical characteristics of a waste stream to
isolate, destroy, or alter the concentration of target chemicals. While this section focuses on
technologies that are used to control on-site releases from a surface finishing process, many of
these technologies are also applicable to other PWB process lines.
Control technologies are typically used to treat on-site releases to both water and air
resulting from the application of a surface finish to the PWB. Wastewater containing
concentrations of heavy metal ions, along with chelators and complexing agents, are of particular
concern. Water effluent standards require the removal of most heavy metals and toxic organics
from the plant effluent before it can be disposed to the sewer. On-site releases of concern to air
include acid vapors and solvent fumes. This section identifies the control technologies used by
PWB manufacturers to treat or control wastewater and air emissions released by the operation of
the surface finishing processes.
Wastewater Treatment
The PWB industry typically uses a sophisticated treatment system to pretreat process
wastewater and spent bath chemistries prior to discharge. The treatment system is comprised of
several parts, including a versatile waste collection system, a flow-through precipitation process, a
series of batch treatment tanks, and a sludge thickening process. The treatment also may be
supplemented by other treatment technologies, depending on the treatment concerns for the
6-28
-------
facility and the effluent permit limits. Together these processes form a complete treatment
system capable of treating the waste streams generated by the PWB manufacturing process,
including those from the surface finishing line.
A diagram of a typical PWB facility treatment system is presented in Figure 6-4, while the
individual treatment processes are discussed below. References to key points of the diagram are
included in the descriptions, and are denoted with reference number in brackets.
Waste Collection and Segregation System. Waste streams are collected from processes
located throughout the facility by a sophisticated piping and collection system that conducts the
individual waste streams to the waste treatment process. The collection system must be versatile,
allowing the waste treatment operators complete control over the destination of an incoming
wastewater flow. In the case of a chemical spill or harmful accidental discharge, operators must
have the ability to divert the wastewater flow into a holding tank to prevent any violations that
might be caused by overloading the treatment system.
The treatment process typically has a waste collection tank and one or more holding
tanks. The collection system deposits the individual waste streams into one or more collection
tanks at the operator's discretion. Waste streams are typically co-mingled in the main collection
tank (1) for a period of time prior to entering the waste treatment system, to allow complete
mixing and to smooth out any concentration spikes that might occur during normal process
operation.
Difficult-to-treat streams, such as those containing chelators or requiring special
treatment, are segregated from the others at the source and fed into separate holding tanks.
Metal-bearing rinses should be segregated from streams which do not contain metals. Specific
segregation of cyanide, solvents, flux, and reflow oils is critically important (Iraclidis, 1998).
Waste streams containing oxidizing agents also typically are segregated from others because of
the difficulty oxidizing agents present during the flocculation and settling stages (oxidizing agents
evolve gas that can hinder floe settling) (Sharp, 1999).
Flow-Through Chemical Precipitation System. In the PWB industry, the majority of
facilities surveyed (61 percent) reported using a conventional chemical precipitation system to
accomplish the removal of heavy metal ions from wastewater. Chemical precipitation is a
process for treating wastewater that depends on the water solubility of the various compounds
formed during treatment. Heavy metal cations present in the wastewater are reacted with certain
treatment chemicals to form hydroxides, sulfides, or carbonates that have relatively low water
solubilities. The resulting heavy metal compounds are precipitated from the solution as an
insoluble sludge that is subsequently sent off-site to reclaim the metals content, or sent to
disposal. Chemical precipitation can be carried out in a batch process, but is typically operated in
a continuous flow-through process to treat wastewater.
In the chemical precipitation treatment of wastewater from PWB manufacturing, the
removal of heavy metals may be carried out by a unit sequence of rapid mix precipitation,
flocculation, and clarification. The process begins by adjusting the pH of the incoming
wastewater (2) to optimum operating conditions (pH 6 to 8). The optimum pH for treatment is
6-29
-------
dependant on both the treatment chemistry and the metals being removed from the wastewater.
Adjustments are made through the addition of acid or lime/caustic. Treatment chemicals are then
dispersed into the wastewater input stream under rapid mixing conditions. The initial mixing unit
(3) is designed to create a high intensity of turbulence in the reactor vessel, promoting multiple
encounters between the metal ions and the treatment chemical species, which then react to form
insoluble metal compounds. The type of chemical compounds formed depends on the treatment
chemical employed; this is discussed in detail later in this section. These insoluble compounds
form a fine precipitate at low pH levels and remain suspended in the wastewater.
The wastewater then enters the flocculation tank (4). The purpose of the flocculation step
is to transform smaller precipitates into large particles that are heavy enough to be removed from
the water by gravity settling in the clarification step. The flocculation tank uses slow mixing to
promote collisions of precipitate particles suspended in the wastewater. The degree of
flocculation is enhanced through the use of flocculating chemicals such as cationic or anionic
polymers. These chemicals promote interparticle adhesion by adding charged particles to the
wastewater, which attach themselves to the precipitate, thereby increasing the growth rate of the
precipitate particles.
Wastewater effluent from the flocculation stage is then fed into a clarification tank (5)
where the water is allowed to collect undisturbed. The rather large precipitate particles settle out
of the water by gravity, forming a blanket of sludge at the bottom of the clarification tank. A
portion of the sludge, typically 10 to 25 percent, is often recirculated to the head of the
flocculation step to reduce chemical requirements, as well as to enhance the rate of precipitation
(Frailey, 1996). The sludge particles provide additional precipitation nuclei that increase the
probability of particle collisions, resulting in a more dense sludge deposit. The remaining 75 to 90
percent of the sludge from the clarifier is fed into the sludge-thickening tank.
The remaining supernatant from the clarifier is decanted through a weir into the bottom of
a sand filter (6). As the water flows upward through the sand filter, the sand traps any remaining
suspended solids, polishing the treated wastewater stream. When the sand filter becomes
saturated with particles, and the effluent quality begins to deteriorate, the filter is taken off-line
and back flushed to remove the particulate matter, cleansing the filter for further use. The
collected particulate matter is sent to the sludge treatment system.
The treated wastewater then undergoes a final pH adjustment (7) to meet effluent
guidelines and is then pumped into a final collection tank prior to being discharged. The
collection tank allows for final testing of the water and also can act as a holding tank to capture
any water that fails inspection due to a system overload of contaminant or some other treatment
system failure. Water from this tank can be returned by the operator to the start of the process if
required.
Other process steps are sometimes employed in the case of unusually strict effluent limits.
Filtration, reverse osmosis, ion exchange, or additional precipitation steps are sometimes
employed to further reduce the concentration of chemical contaminants present in the wastewater
effluent.
6-30
-------
Batch Treatment of Process Baths. Most spent process baths can be mixed with other
wastewater and treated by the on-site wastewater treatment process using chemical precipitation.
Chemical suppliers, however, recommend that some process baths be treated separately from the
usual waste treatment process. The separate treatment of these baths is usually recommended
due to the presence of strong chelating agents, high metal concentrations or other chemicals, such
as additives or brighteners, which require additional treatment measures before they can be
disposed of properly. Spent bath solution requiring special treatment measures can be processed
immediately, but is typically collected and stored until enough has accumulated to warrant
treatment. Batch treatment (8) of the accumulated waste is then performed in a single tank or
drum, following the specific treatment procedures provided by the chemical supplier for that
bath.
Following batch treatment, the remaining solution may be transferred to the flow-through
precipitation system for further treatment, drummed for disposal, or discharged directly. Sludge
from the process is dewatered by a sludge press and then combined with other treatment sludge
to be dried.
Sludge Thickening Process. Sludge formed in the clarifier needs to be thickened and
dewatered prior to being shipped off-site. Clarifier sludge is typically light (4 to 5 percent solids)
and not very well settled prior to entering the thickening tank (9). Once in the tank, the
precipitate is compressed as it moves downward by the weight of the precipitate above and by
the constricting funnel at the bottom of the thickening tank. The supernatant separates from the
sludge as it thickens. It is pumped from the top of the thickener and returned to the wastewater
collection tank to be processed through the treatment system once again. The dense, thickened
sludge (8 to 10 percent solids) is then pumped from the bottom of the thickening tank to a sludge
press.
The sludge press (10) and sludge dryer (11) minimize the volume of sludge by increasing
the solids content through dewatering, thus reducing the cost of disposal. The sludge press is
usually a plate filter press, but belt filter presses also may be used. Dewatering occurs when the
sludge is passed under high pressure through a series of cloth covered plates. The cloth quickly
becomes coated with sludge, forming a layer that retains the solids, while the water is forced
through the cloth. The sludge cake (30 to 35 percent solids) is sufficiently dry for direct disposal
or recovery (Pontius, 1990). A sludge dryer (up to 70 percent solids) may be utilized to further
dewater the sludge, if desired.
Treatment of Non-Chelated Wastewater. The absence of complexing chemicals (e.g.,
ammonia) or chelating agents (e.g., EDTA) in the wastewater stream simplifies the removal of
metal ions by precipitation. Metal removal from such waste streams is accomplished through
simple pH adjustment using hydroxide precipitation. Caustic soda is typically used while other
treatment chemicals include calcium hydroxide and magnesium hydroxide. The heavy metal
ions react with the caustic soda to form insoluble metal hydroxide compounds that precipitate
out of solution at a high pH level. After the precipitate is removed by gravity settling, the effluent
is pH adjusted to a pH of seven to nine and then sewered. The treatment can be performed in a
chemical precipitation process similar to the one shown in Figure 6-4, resulting in a sludge
contaminated with metals that is then sent to recycling or disposal.
6-31
-------
a
V
S
H
>
H
•*
so
|
6-32
-------
Treatment of Wastewater Containing Chelated Metals. The presence of complexing
chemicals or chelators, such as EDTA, formaldehyde, thiourea, and quadrol require a more
vigorous effort to achieve a sufficient level of metal removal. Chelators are chemical compounds
that inhibit precipitation by forming chemical complexes with the metals, allowing them to
remain in solution beyond their normal solubility limits. These chemicals are found in spent
surface finishing plating baths, in cleaners, and in the water effluent from the rinse tanks
following these baths. Treatment chemicals enhance the removal of chelated metals from water
by breaking the chelate-to-metal bond, destroying the soluble complex. The freed metal ions
then react to form insoluble metal compounds, such as metal hydroxides, that precipitate out of
solution.
Several different chemicals are currently being used to effectively treat chelator
contaminated wastewater resulting from the manufacture of PWBs. Some common chemicals
used in the treatment of wastewater produced by the surface finishing process are briefly
described in Table 6-8. For more information regarding individual treatment chemicals and their
applicability to treating specific wastes, consult a supplier of waste treatment chemicals.
Chelated waste streams are typically segregated from non-chelated streams to minimize
the consumption of expensive treatment chemicals. Treatment of small volumes of these waste
streams is typically done in a batch treatment tank. A facility with large volumes of chelated
waste often will have a separate, dedicated flow-through chelated precipitation system to remove
the chelated metals from the wastewater.
Alternative Treatment Processes. Although chemical precipitation (61 percent of those
surveyed) is the most common process for treating wastewater used by PWB manufacturers,
other treatment processes exist. Survey respondents reported the use of ion exchange (30
percent) to successfully treat wastewater generated from the manufacture of PWBs. Thirty-six
percent of the ion exchange systems also combined with electrowinning to enhance treatment.
These processes operate separately or in combination to efficiently remove metal ions from
chelated or non-chelated waste streams, typically yielding a highly concentrated sludge for
disposal. These processes were discussed in Section 6.2.1.
Despite the supplier's recommendations, PWB facilities sometimes treat individual
process baths using their typical wastewater treatment process. Spent bath solutions can be
mixed slowly, in small quantities, with other wastewater before being treated, thus diluting the
concentration of the chemical species requiring treatment. However, the introduction of
concentrated wastes to the wastewater could result in increased treatment chemical consumption
and more sludge produced than if batch treated separately. Also, the introduction of a chemical
species not typically found in the wastewater may adversely affect the treatment process or
require more vigorous treatment chemicals or processes. Factors affecting the success of this
approach include the type of treatment chemicals used, the contaminant concentrations in the
wastewater, and the overall robustness of the existing, in-house treatment process.
6-33
-------
Table 6-8. Treatment Chemicals Used to Remove Metals From Chelated Wastewater
Chemical
Ferrous Sulfate
DTC (Dimethyl-
dithiocarbamate)
Sodium Sulfide
Polyelectrolyte
Sodium Borohydride
Ferrous Dithionite
TMT 15 (Tri-mercaptotriazine)
Description
Inexpensive treatment that requires iron concentrations in excess of 8: 1 of
copper and other metals to form an insoluble metal hydroxide precipitate
(Coombs, 1993). Ferrous sulfate is first used as a reducing agent to break
down the complexed copper structures under acidic conditions before
forming the metal hydroxide during subsequent pH neutralization.
Drawbacks include the large volumes of sludge generated and the
presence of iron which reduces the value of sludge to a reclaimer.
Moderately expensive chemical that acts as a complexing agent, exerting
a stronger reaction to the metal ion than the chelating agent, effectively
forming an insoluble heavy metal complex. The sludge produced is light
in density and difficult to separate by gravity (Guess, 1992; Frailey, 1996).
Forms metal sulfides with extremely low solubilities that precipitate even
in the presence of chelators. Produces large volume of sludge that is
slimy and difficult to dewater (Guess, 1992).
Polymers that remove metals effectively without contributing to the
volume of sludge. Primary drawback is the high chemical cost (Frailey,
1996).
Strong reducing agent reduces metal ions, then precipitate out of solution
forming a dense, low volume sludge. Drawbacks include its high
chemical cost and the evolution of potentially explosive hydrogen gas
(Guess, 1992; Frailey, 1996).
Reduces metal ions under acidic conditions to form metallic particles that
are recovered by gravity separation. Excess iron is regenerated instead of
being precipitated, producing a low volume sludge (Guess, 1992).
Designed specifically to precipitate silver ions, which are unaffected by
other treatment chemicals, from wastewater. Primary drawbacks are the
high chemical to silver removed weight ratio and the high chemical cost
(Sharp, 1999).
Individual Alternative Treatment Profiles
There are often many approaches from which a facility can choose to properly treat and
dispose of a process waste stream. Several of the approaches, which have been discussed in this
CTSA chapter, include reclamation, recycling, treatment, disposal, or a combination of these.
The treatment or recycling method used by a facility for each waste stream is dependent on a
number of factors including discharge permit effluent limits (is more vigorous treatment required
to meet effluent limits?), economics (is the treatment cost effective?), the capability of on-site
treatment system (e.g., the presence of reclamation technologies), the treatment requirements of
processes other than the surface finishing line (e.g., can the waste stream be combined with other
waste streams to make other treatment options more applicable), and a facility's preference,
based on experience. One, or a combination of several of these factors, will dictate the treatment
options available to a particular facility.
6-34
-------
Chemical suppliers offer guidance on the proper treatment and disposal of their process
chemicals and are available to consult with facilities investigating treatment options. Process
baths often contain proprietary ingredients that are known only to the chemical manufacturer.
These may impact the manner in which the bath can be successfully treated. Prior to deciding on
a treatment method for a particular bath, a PWB manufacturer should consult with the chemical
supplier to confirm the applicability of the method and to identify any problems or concerns that
may arise.
A profile for treating PWB surface finishing chemical baths is given in Table 6-9. The
profile was developed and reviewed by PWB manufacturers participating in this project as an
example of the treatment requirements of the individual chemical baths. Treatment of similar
baths by individual facilities may differ from that presented in Table 6-9, according to the
requirements/preferences of each facility.
Batch treatment is indicated for bath types containing chemicals or metals that require
special treatment considerations beyond that provided by the precipitation system. Batch
treatment could be required due to the presence of chelating agents, oxidizers, pH concerns,
chemical constituents not affected by precipitation (e.g., organic compounds, silver which is
unaffected by typical treatment chemicals, etc.), or to minimize the use of expensive treatment
chemicals. After batch treatment, the remaining supernatant may be fed through the precipitation
system for additional treatment, if required, drummed and sent out, or disposed directly to the
POTW, if it meets the effluent limits of the facility.
The batch treatment of microetches is typically done separately from other process wastes
due to the presence of chemical oxidizers in the microetch baths. Oxidizers commonly found in
PWB waste streams include nitric acid, peroxides, persulfates, and permanganates. These
compounds evolve gas during the treatment process, which hinders floe settling and, thereby,
reduces the overall efficiency of the treatment process. Waste streams containing oxidizers can
often be combined during treatment.
Metal reclamation is indicated for baths with metal concentrations that might typically
exceed effluent limits, or that are too valuable to simply discard. Metals reclamation can be
performed on-site using one, or a combination of metal recovery technologies, or can be sent off-
site to a metal refiner.
6-35
-------
Table 6-9. Treatment Profile of PWB Surface Finishing Process Baths
Bath Type
Acid Dip
Catalyst
Cleaner
Drag -out Rinse
(following gold,
palladium)
Electroless Gold
Electroless
Nickel
Electroless
Palladium
Flux
Immersion
Silver
Immersion Tin
Microetch
OSP
Predip
Solder/Dross
Water Rinse
Process(es)
Nickel/Gold and
Nickel /Palladium/Gold
Nickel/Gold and
Nickel /Palladium/Gold
All
Nickel/Gold and
Nickel /Palladium/Gold
Nickel/Gold and
Nickel /Palladium/Gold
Nickel/Gold and
Nickel /Palladium/Gold
Nickel/Gold and
Nickel /Palladium/Gold
HASL
Immersion Silver
Immersion Tin
All
OSP
Immersion Tin and
Immersion Silver
HASL
All
Chelated
N
N
Y
Y
Y
Y
Y
N
Y
Y
N
N
N
N
N
Typical Treatment Method a
Batch treatment - no oxidizers.
Metals reclamation on-site or off-site.
Batch treatment - no oxidizers.
Metals reclamation on-site or off-site.
Metals reclamation on-site or off-site.
Batch treatment - no oxidizers for chelated
waste streams.
Metals reclamation on-site or off-site.
Hazardous waste disposal.
Point of generation treatment equipment (e.g.,
ion exchange, iron exchange, etc.) to remove
silver, then to batch treatment - no oxidizers for
chelated streams.
Batch treatment for the destruction of thiourea
followed by precipitation treatment to remove
the remaining tin.
Batch treatment - oxidizers only.
Batch treatment - no oxidizers.
Batch treatment - no oxidizers.
Metals reclamation off-site.
Flow -through precipitation system.
Source: Sharp, 1999.
a Treatment methods represent the typical method by which the bath is treated. Indicated method is not the only way a
bath may be treated by an individual facility. Typical methods were developed and reviewed by PWB manufacturer
project participants.
Air Pollution Control Technologies
Air pollution control technologies are often used by the PWB industry to cleanse air
exhaust streams of harmful fumes and vapors. Exactly half (50 percent) of the PWB facilities
surveyed have installed air scrubbers to control air emissions from various manufacturing
processes, and almost a quarter of the facilities (23 percent) scrub air releases from surface
finishing processes. The first step of any air control process is the effective containment of
fugitive air emissions at their source of release. This is accomplished using fume hoods over the
6-36
-------
process areas from which the air release of concern occurs. These hoods may be designed to
continuously collect air emissions for treatment by one of the methods described below.
Gas Absorption. One method for removing pollutants from an exhaust stream is by gas
absorption in a technique sometimes referred to as air scrubbing. Gas absorption is defined as
the transfer of material from a gas to a contacting liquid or solvent. The pollutant is chemically
absorbed and dispersed into the solvent, leaving the air free of the pollutant. The selection of an
appropriate solvent should be based on the liquid's solubility for the solute, and the cost of the
liquid. Water is used for the absorption of water-soluble gases, while alkaline solutions are
typically used for the absorption of acid gases. Air scrubbers are used by the PWB industry to
treat wet process air emissions, such as formaldehyde and acid fumes, and emissions from other
processes other than the surface finishing process.
Gas absorption is typically carried out in a packed gas absorption tower, or scrubber. The
gas stream enters the bottom of the tower and passes upward through a wetted bed of packing
material before exiting the top. The absorbing liquid enters the top of the tower and flows
downward through the packing before exiting at the bottom. Absorption of the air pollutants
occurs during the period of contact between the gas and liquid. The gas is either physically or
chemically absorbed and dispersed into the liquid. The liquid waste stream then is sent to water
treatment before being discharged to the sewer. Although the most common method for gas
absorption is the packed tower, other methods exist such as plate towers, sparged towers, spray
chambers, or venturi scrubbers (Cooper and Alley, 1990).
Gas Adsorption. The removal of low concentration organic gases and vapors from an
exhaust stream can be achieved by the process of gas adsorption. Adsorption is the process in
which gas molecules are retained on the interface surfaces of a solid adsorbent by either physical
or chemical forces. Activated carbon is the most common adsorbent, but zeolites, such as
alumina and silica, are also used. Adsorption is used primarily to remove volatile, organic
compounds from air, but is also used in other applications such as odor control and drying
process gas streams (Cooper and Alley, 1990). In a surface finishing process, gas adsorption can
be used to recover volatile organic compounds, such as formaldehyde.
Gas adsorption occurs when the vapor-laden air is collected and then passed through a
bed of activated carbon or another adsorbent material. The gas molecules are adsorbed onto the
surface of the material, while the clean, vapor-free air is exhausted from the system. The
adsorbent material eventually becomes saturated with organic material and must be replaced or
regenerated. Adsorbent canisters, which are replaced on a regular basis, are typically used to treat
small gas flow streams. Larger flows of organic pollutants require packed beds of adsorbent
material, which must be regenerated when the adsorbent becomes saturated (Cooper and Alley,
1990).
Regeneration of the adsorbent is typically accomplished by a steam-stripping process.
The adsorbent is contacted with low-pressure steam which desorbs the adsorbed gas molecules
from the surface of the packed bed. Following condensation of the steam, the organic material is
recovered from the water by either decanting or distillation (Campbell and Glenn, 1990).
6-37
-------
REFERENCES
Bayes, Martin. 1996. Shipley Company. Personal communication to Jack Geibig, UT Center for
Clean Products and Clean Technologies. January.
Brooman, Eric. 1996. Concurrent Technologies Corporation. Personal communication to Lori
Kincaid, UT Center for Clean Products and Clean Technologies. August 5.
Campbell, M. and W. Glenn. 1990. "Profit from Pollution Prevention." Pollution Probe
Foundation.
Capsule Environmental Engineering, Inc. 1993. "Metal Finishing Pollution Prevention Guide."
Prepared for Minnesota Association of Metal Finishers in conjunction with The Minnesota
Technical Assistance Program. Prepared by Capsule Environmental Engineering, Inc., 1970
Oakcrest Avenue, St. Paul, MN 55113. July.
Coombs, Jr., Clyde. 1993. Printed Circuits Handbook. 4th ed. McGraw-Hill.
Cooper, David C. and F.C. Alley. 1990. Air Pollution Control: A Design Approach. Waveland
Press, Prospect Heights, IL.
Edwards, Ted. 1996. Honeywell. Personal communication to Lori Kincaid, UT Center for Clean
Products and Clean Technologies. July 10.
Fehrer, Fritz. 1996. Silicon Valley Toxics Coalition. Personal communication to Lori Kincaid,
UT Center for Clean Products and Clean Technologies. July 22.
Fellman, Jack D. 1997. "On-Site Solder Purification for HASL." In: Proceedings of the IPC
Printed Circuits Expo 97, San Jose, CA, March 9-13.
Frailey, Dean. 1996. Morton International. Personal communication to Jack Geibig, UT Center
for Clean Products and Clean Technologies. May 7.
Guess, Robert. 1992. Romar Technologies. United States Patent # 5,122,279. July 16.
Hosea, J. Michael. 1998. "Water Reuse for Printed Circuit Boards - When Does It Make Sense?"
In: Proceedings of the IPC Printed Circuits Expo 98, Long Beach, CA. April 26-30.
Iraclidis, Taso. 1998. "Wastewater Treatment Technologies of Choice for the Printed Circuit
Board Industry." In: Proceedings of the IPC Printed Circuits Expo 98, Long Beach, C A. April
26-30.
Kling, David J. 1995. Director, Pollution Prevention Division, Office of Pollution Prevention and
Toxics. Memo to Regional OPPT, Toxics Branch Chiefs. February 17.
Lee, Matthew A. 1999. "Controlling Emissions Stemming From The Hot Air Solder Leveling
6-38
-------
Process'' From the Proceedings of the Technical Conference, IPC Printed Circuits Exposition
1999, March 14-18, Long Beach, CA. Prepared by Ceco Filters, Conshohocken, PA.
Pontius, Frederick W. (ed). 1990. Water Quality and Treatment: A Handbook of Community
Water Supplies. 4th ed. American Water Works Association, McGraw-Hill, Inc.
Schectman, Michael. 2000. Technic. Personal communication to Jack Geibig, UT Center for
Clean Products and Clean Technologies. (Series of personal communications.)
Sharp, John. 1999. Teradyne. Personal communication to Jack Geibig, UT Center for Clean
Products and Clean Technologies. (Series of personal communications.)
U.S. EPA (Environmental Protection Agency). 1990. Guides to Pollution Prevention: The
Printed Circuit Board Manufacturing Industry. EPA Office of Resource and Development,
Cincinnati, OH. EPA/625/7-90/007. June.
U.S. EPA (Environmental Protection Agency). 1995a. "Printed Wiring Board Case Study 1:
Pollution Prevention Work Practices." Pollution Prevention Information Clearinghouse (PPIC).
Washington, D.C. EPA 744-F-95-004. July.
U.S. EPA (Environmental Protection Agency). 1995b. "Printed Wiring Board Case Study 2:
On-Site Etchant Regeneration." Pollution Prevention Information Clearinghouse (PPIC).
Washington, D.C. EPA744-F-95-005. July.
U.S. EPA (Environmental Protection Agency). 1996a. "Printed Wiring Board Project:
Opportunities for Acid Recovery and Management." Pollution Prevention Information
Clearinghouse (PPIC). Washington, D.C. EPA 744-F-95-009. September.
U.S. EPA (Environmental Protection Agency). 1996b. "Printed Wiring Board Project: Plasma
Desmear: A Case Study." Pollution Prevention Information Clearinghouse (PPIC).
Washington, D.C. EPA 744-F-96-003. September.
U.S. EPA (Environmental Protection Agency). 1996c. "Printed Wiring Board Project: A
Continuous-Flow System for Reusing Microetchant." Pollution Prevention Information
Clearinghouse (PPIC). Washington, D.C. EPA 744-F-96-024. December.
U.S. EPA (Environmental Protection Agency). 1997a. "Pollution Prevention beyond Regulated
Materials." Pollution Prevention Information Clearinghouse (PPIC). Washington, D.C.
EPA744-F-97-006. May.
U.S. EPA (Environmental Protection Agency). 1997b. "Identifying Objectives for Your
Environmental Management System." Pollution Prevention Information Clearinghouse (PPIC).
Washington, D.C. EPA744-F-97-009. December.
U.S. EPA (Environmental Protection Agency). 1997c. "Building an Environmental Management
System - HR Industry Experience." Pollution Prevention Information Clearinghouse (PPIC).
6-39
-------
Washington, D.C. EPA744-F-97-010. December.
U.S. EPA (Environmental Protection Agency). 1998. Printed Wiring Board Pollution
Prevention and Control Technology: Analysis of Updated Survey Results. Design for the
Environment Printed Wiring Board Project. EPA Office of Pollution Prevention and Toxics.
Washington, D.C. EPA 744-R-98-003. August.
U.S. EPA (Environmental Protection Agency). 1999. "Pollution Prevention beyond Regulated
Materials." Pollution Prevention Information Clearinghouse (PPIC). Washington, D.C.
EPA744-F-97-004. May.
6-40
-------
Chapter 7
Choosing Among Surface Finishing Technologies
This chapter of the Cleaner Technologies Substitutes Assessment (CTSA) organizes data
collected or developed throughout the assessment of the baseline non-conveyorized hot air
soldering level (HASL) process and alternatives, in a manner that facilitates decision-making.
First, risk, competitiveness, and conservation data are summarized in Section 7.1. This
information is used in Section 7.2 to assess the private and external benefits and costs (which
constitute the societal benefits and costs) of implementing an alternative as compared to the
baseline. Section 7.3 provides summary profiles for the baseline and alternatives.
Information is presented for six technologies for performing the surface finishing
function. These technologies are HASL, nickel/gold, nickel/palladium/gold, organic solderability
preservative (OSP), immersion silver, and immersion tin. All of these technologies are wet
chemistry processes, except the HASL technology, which combines a wet chemistry pre-cleaning
process with the mechanical process of applying the solder. The wet chemistry processes can be
operated using vertical, immersion-type, non-conveyorized equipment or horizontal,
conveyorized equipment. The HASL process can be applied in either equipment mode. Table 7-
1 presents the processes (alternatives and equipment configurations) evaluated in the CTSA.
Table 7-1. Surface Finishing Processes Evaluated in the CTSA
Surface Finishing Technology
HASL (Baseline)
Nickel/Gold
Nickel/Palladium/Gold
OSP
Immersion Silver
Immersion Tin
Equipment Configuration
Non-Conveyorized
X
X
X
X
X
Conveyorized
X
X
X
X
The results of the CTSA comparing alternative surface finishes are mixed, with some of
the alternatives offering environmental and/or economic benefits, or both, when compared to the
baseline non-conveyorized HASL process. The results of the risk screening and comparison of
the alternatives were also mixed, while results of the performance demonstration indicate that all
of the alternative finishes perform as well as the baseline. In addition, it is important to note that
there are additional factors beyond those assessed in this CTSA that individual businesses may
consider when choosing among alternatives. None of these sections make value judgements or
recommend specific alternatives. The intent of this document is to provide information for
decision-makers to consider, although the actual decision of whether or not to implement an
alternative is made outside of the CTSA process.
7-1
-------
7.1 RISK, COMPETITIVENESS, AND CONSERVATION DATA SUMMARY
Earlier sections of the CTSA evaluated the risk, performance, cost, and resource
requirements of the baseline surface finishing technology as well as the alternatives. This section
summarizes the findings associated with the analysis of surface finishing technologies. Relevant
data include the following:
• Risk information: occupational health risks, public health risks, ecological hazards, and
process safety concerns.
• Competitiveness information: technology performance, cost and regulatory status, and
international information.
• Conservation information: energy and natural resource use.
Sections 7.1.1 through 7.1.3 present risk, competitiveness, and conservation summaries,
respectively.
7.1.1 Risk Summary
The risk screening and comparison uses a health-hazard based framework and a model
facility approach to compare the potential health risks of one surface finishing process
technology to the potential risks associated with switching to an alternative technology. As much
as possible, reasonable and consistent assumptions are used across alternatives. Data to
characterize the model facility and exposure patterns for each process alternative were aggregated
from a number of sources, including printed wiring board (PWB) shops in the United States,
supplier data, and input from PWB manufacturers at project meetings. Thus, the model facility is
not entirely representative of any one facility, and actual risk could vary substantially, depending
on site-specific operating conditions and other factors.
When using the risk results to compare potential health effects among alternatives, it is
important to remember that this is a screening level rather than a comprehensive risk
characterization, both because of the predefined scope of the assessment and because of
exposure and hazard data limitations. It should also be noted that this approach does not result
in any absolute estimates or measurements of risk, and even for comparative purposes, there are
several important uncertainties associated with this assessment (see Section 3.4).
The Exposure Assessment, whenever possible, used a combination of central tendency
and high-end assumptions, as would be used for an overall high-end exposure estimate. Some
values used in the exposure calculations, however, are better characterized as "what-if,"
especially pertaining to exposure frequency, bath concentrations, use of gloves, and process area
ventilation rates for a model facility. Because some part of the exposure assessment for both
inhalation and dermal exposures qualifies as a "what-if descriptor, the entire assessment should
be considered "what-if."
7-2
-------
As with any evaluation of risk, there are a number of uncertainties involved in the
measurement and selection of hazard data, and in the data, models, and scenarios used in the
exposure assessment. Uncertainties arise both from factors common to all risk characterizations
(e.g., extrapolation of hazard data from animals to humans, extrapolation from the high doses
used in animal studies to lower doses to which humans may be exposed, and missing toxicity
data, including data on the cumulative or synergistic effects of chemical exposure), and other
factors that relate to the scope of the risk characterization (e.g., the surface finishing
characterization is a screening level characterization rather than a comprehensive risk
assessment). Key uncertainties in the risk characterization include the following:
• The risk estimates for occupational dermal exposure are based on limited dermal toxicity
data, using oral toxicity data with oral to dermal extrapolation when dermal toxicity data
were unavailable. Coupled with the high uncertainty in estimating dermal absorption
rates, this could result in either over- or under-estimates of exposure and risk.
• The exposure assessment is based on modeled estimates of average, steady-state chemical
concentrations in air, rather than actual monitoring data of average and peak air
concentrations.
• The exposure assessment does not account for any side reactions occurring in the baths,
which could either underestimate exposures to toxic reaction products or overestimate
exposures to toxic chemicals that react in the bath to form more benign chemicals.
• Due to resource constraints, the risk screening and comparison does not address all types
of exposures that could occur from surface finishing processes or the PWB industry,
including short-term or long-term exposures from sudden releases due to fires, spills, or
periodic releases.
• For aquatic risk, surface water concentrations are based on estimated releases to a
modeled, representative stream flow for the electronics industrial sector.
The Risk Characterization section of the CTSA (Section 3.4) discusses the uncertainties in this
characterization in more detail.
Occupational Health Risks
Health risks to workers are estimated for inhalation exposure to vapors and aerosols from
surface finishing baths and for dermal exposure to surface finishing bath chemicals. Inhalation
exposure estimates are based on the assumptions that emissions to indoor air from conveyorized
lines are negligible, that the air in the process room is completely mixed and chemical
concentrations are constant over time, and that no vapor control devices (e.g., bath covers) are
used in non-conveyorized lines. Dermal exposure estimates are based on the conservative
assumptions that workers do not wear gloves and that all non-conveyorized lines are operated by
manual hoist. Dermal exposure to line operators on non-conveyorized lines is estimated for
routine line operation and maintenance (e.g., bath replacement, filter replacement), and on
conveyorized lines for bath maintenance activities alone.
Based on the number of chemicals with risk results above concern levels, some
alternatives to the non-conveyorized HASL process appear to pose lower occupational risks (i.e.,
immersion silver, conveyorized and non-conveyorized immersion tin, and conveyorized HASL),
7-3
-------
some may pose similar levels of risk (i.e., conveyorized and non-conveyorized OSP), and some
may pose higher risk (i.e., nickel/gold and nickel/palladium/gold). There are occupational
inhalation risk concerns for chemicals in the non-conveyorized HASL, nickel/gold,
nickel/palladium/gold, and OSP processes. There are also occupational risk concerns for dermal
contact with chemicals in the non-conveyorized HASL, nickel/gold, nickel/palladium/gold, OSP,
and immersion tin processes, and the conveyorized HASL and OSP processes.
Table 7-2 presents chemicals of concern for potential occupational risk from inhalation.
Table 7-3 presents chemicals of concern for potential occupational risk from dermal contact.
Table 7-2. Surface Finishing Chemicals of Concern for Potential
Occupational Inhalation Risk
Chemical
Alkyldiol
Ethylene glycol
Hydrochloric acid
Hydrogen peroxide
Nickel sulfate
Phosphoric acid
Propionic acid
Process a
(Non-Conveyorized, 260,000 ssf)
HASL
X
Nickel/Gold
X
X
X
X
X
Nickel/Palladium/Gold
X
X
X
X
X
X
OSP
X
Non-conveyorized immersion silver process not evaluated. Occupational exposure and risk from all conveyorized
process configurations are below concern levels.
X Line operator risk results above concern levels (non-cancer health effects).
The non-conveyorized nickel/gold process contains the only chemical for which an
occupational cancer risk has been estimated (inorganic metallic salt A). The line operator
inhalation exposure estimate for inorganic metallic salt A results in an estimated upper bound
excess individual life time cancer risk of 2 x 10~7 (one in five million) based on high end exposure.
Cancer risks less than 1 x 10~6 (one in one million) are generally considered to be of low concern.
Risks to other types of workers1 were assumed to be proportional to the average amount of time
spent in the process area, which ranged from 12 to 69 percent of the risk for a line operator.
1 These include laboratory technicians, maintenance workers, and wastewater treatment operators. Other types of
workers may be present for shorter or longer times.
7-4
-------
Table 7-3. Chemicals of Concern for Potential Dermal Risks
Chemical
Ammonia compound A
Ammonium chloride
Ammonium hydroxide
Copper ion
Copper salt C
Copper sulfate pentahydrate
Ethylene glycol monobutyl
ether
Hydrogen peroxide
Inorganic metallic salt B
Lead
Nickel sulfate
Urea compound C
Process Configuration a
HASL
(NC)
XX
t
HASL
(C)
XX
t
Nickel/Gold
(NC)
X
X
XX
X
XX
XX
Nickel/
Palladium/Gold
(NC)
X
X
XX
X
XX
XX
OSP
(NC)
XX
XX
XX
OSP
(C)
XX
X
XX
Immersion
Tin
(NC)
X
X
a No risk results were above concern levels for the conveyorized immersion silver or conveyorized immersion tin
processes.
X Line operator risk results above concern levels (non-cancer health effects).
XX Line operator and laboratory technician risk results above concern levels (non-cancer health effects).
f: Risk indicators were not calculated for lead as with the other chemicals (see Section 3.4.6). Other information,
however, indicates that incidental ingestion of lead from contact with hands could result in lead exposure at levels of
concern.
C: Conveyorized (horizontal) process configuration
NC: Non-conveyorized (vertical) process configuration.
Other identified chemicals in the surface finishing processes are suspected or known
carcinogens. Lead and thiourea have been determined by IARC to be possible human
carcinogens (IARC Group 2B); lead has also been classified by EPA as a probable human
carcinogen (EPA Class B2). Lead is used in tin-lead solder in the HASL process. Thiourea is
used in the immersion tin process. Urea compound B, a confidential ingredient in the nickel/gold
and nickel/palladium/gold processes, is possibly carcinogenic to humans. Exposure for workers
from these chemicals has been estimated, but cancer potency and cancer risks are unknown.
Additionally, strong inorganic and acid mists of sulfuric acid have been determined by IARC to
be a human carcinogen (IARC Group 1). Sulfuric acid is used in diluted form in every surface
finishing process in this evaluation. It is not expected, however, to be released to the air as a
strong acid mist. There are potential cancer risks to workers from these chemicals, but because
there are no slope factors, the risks cannot be quantified.
For non-cancer risk, risk indicators exceeding concern levels - a hazard quotient (HQ)
greater than one, a margin of exposure (MOE) based on no-observed adverse effect level
(NOAEL) lower than 100, or MOE based on a lowest-observed adverse effect level (LOAEL)
lower than 1,000 - were estimated for occupational exposures to chemicals in the non-
conveyorized and conveyorized HASL processes, non-conveyorized nickel/gold process, non-
7-5
-------
conveyorized nickel/palladium/gold process, non-conveyorized and conveyorized OSP
processes, and the non-conveyorized immersion tin process.
Based on calculated occupational exposure levels, there may be adverse health effects to
workers exposed to chemicals with a HQ exceeding 1.0 or an MOE less than 100 or 1,000.
However, it should be emphasized that these conclusions are based on screening level estimates.
These numbers are used here for relative risk comparisons between processes, and should not be
used as absolute indicators for actual health risks to surface finishing line workers.
Worker blood-lead levels measured at one PWB manufacturing facility were below any
federal regulation or guideline for workplace exposure. Modeling data, however, show that it
may be possible for blood-lead levels to exceed recommended levels for an adult and fetus, given
high incidental ingestion rates of lead from handling solder. These results are highly uncertain;
ingestion rates are based on incidental soil ingestion rates for adults in contact with soil.
However, this indicates the need for good personal hygiene for HASL line operators, especially
wearing gloves and hand washing to prevent accidental hand-to-mouth ingestion of lead.
Public Health Risks
Potential public health risk was estimated for inhalation exposure for the general public
living near a PWB facility. Public exposure estimates are based on the assumption that emissions
from both conveyorized and non-conveyorized process configurations are vented to the outside.
The risk indicators for ambient exposures to humans, although limited to airborne releases,
indicate low concern for nearby residents. The upper bound excess individual cancer risk for
nearby residents from inorganic metallic salt A in the non-conveyorized nickel/gold process was
estimated to be from approaching zero to 2 x 10"11 (one in 50 billion). This chemical has been
classified as a human carcinogen.2 All hazard quotients are less than one for ambient exposure to
the general population, and all MOEs for ambient exposure are greater than 1,000 for all
processes, indicating low concern from the estimated air concentrations for chronic non-cancer
effects.
Estimated ambient air concentrations of lead from a HASL process are well below EPA
air regulatory limits for lead, and risks to the nearby population from airborne lead are expected
to be below concern levels.
Ecological Risks
We calculated ecological risk indicators (RIECo) for non-metal surface finishing chemicals
that may be released to surface water. Risk indicators for metals are not used for comparing
alternatives because it is assumed that on-site treatment is targeted to remove metal so that
permitted concentrations are not exceeded. Estimated surface water concentrations for non-
metals exceeded the concern concentration (CC) in the following processes: four in the non-
2 A cancer classification of known human carcinogen has been assigned by either the EPA, IARC, and/or NTP.
Further details about the carcinogen classification are not provided in order to protect the confidential chemical
identity.
7-6
-------
conveyorized HASL process, three in the conveyorized HASL process, one in the non-
conveyorized OSP process, one in the conveyorized OSP process, one in the conveyorized
immersion silver process, and one in the non-conveyorized immersion tin process. Table 7-4
presents chemicals of concern based on ecological risk indicator results.
Table 7-4. Aquatic Risk of Non-Metal Chemicals of Concern
Chemical
Alkylaryl imidazole
Alkylaryl sulfonate
1 ,4-Butenediol
Hydrogen peroxide
Potassium peroxymonosulfate
HASL
(NC)
X
X
X
X
HASL
(C)
X
X
X
OSP
(NC)
X
OSP
(C)
X
Immersion Silver
(C)
X
Immersion Tin
(NC)
X
Estimated surface water concentration > concern concentration (CC) after POTW treatment.
A CC is the concentration of a chemical in the aquatic environment which, if exceeded,
may result in significant risk to aquatic organisms. CCs were determined by dividing acute or
chronic toxicity values by an assessment factor (ranging from one to 1,000) that incorporates the
uncertainty associated with toxicity data. CCs are discussed in more detail in Section 3.3.3.
Process Safety
Workers can be exposed to two types of hazards affecting occupational safety and health:
chemical hazards and process hazards. Workers can be at risk through exposure to chemicals
and because of close proximity to automated equipment. In order to evaluate the chemical safety
hazards of the various surface finishing technologies, material safety data sheets (MSDSs) for
chemical products used with each of the surface finishing technologies were reviewed. Table 7-5
summarizes the hazardous properties of surface finishing chemical products.
Other potential chemical hazards can occur because of hazardous decomposition of
chemical products, or chemical product incompatibilities with other chemicals or materials. With
few exceptions, most chemical products used in surface finishing technologies can decompose
under specific conditions to form potentially hazardous chemicals. In addition, all of the surface
finishing processes have chemical products with incompatibilities that can pose a threat to worker
safety if the proper care is not taken to prevent such occurrences.
7-7
-------
Table 7-5. Chemical Hazards
Process
HASL
Nickel/Gold
Nickel/Palladium/Gold
OSP
Immersion Silver
Immersion Tin
No. of
MSDSa
33
19
18
9
4
14
Hazardous Property b
F
1
1
C
E
1
1
1
FH
3
2
1
CO
4
8
12
4
2
7
0
1
1
1
1
1
SRP
1
1
1
1
U
1
1
a For alternative processes with more than one product line, the hazard data reported represent the most hazardous
bath of each type for the two product lines (e.g., of the microetch baths from the two product lines, the one with the
most hazardous chemicals is reported).
b Formulations for HASL process baths were unavailable because cleaner and microetch bath chemistries are not
made specifically for the HASL process. Hazards reported for HASL bath types were reported as the worst case of the
results of similar baths from other processes.
F = Flammable; C = Combustible; E = Explosive; FH = Fire Hazard; CO = Corrosive; O = Oxidizer; SRP = Sudden
Release of Pressure; U = Unstable
Work-related injuries from equipment, improper use of equipment, bypassing equipment
safety features, failure to use personal protective equipment, and physical stresses that may
appear gradually as a result of repetitive motion are all potential process safety hazards to
workers. Regardless of the technology used, of critical importance is an effective and ongoing
safety training program. Characteristics of an effective worker health and safety program include:
• an employee training program;
• employee use of personal protective equipment;
• proper chemical storage and handling; and
• safe equipment operating procedures.
Without appropriate training, the number of worker accidents and injuries is likely to
increase, regardless of the technology used. A key management responsibility is to ensure that
training is not compromised by pressure to meet production demands or by cost-cutting efforts.
7.1.2 Competitiveness Summary
The competitiveness summary provides information on basic issues traditionally
important to the competitiveness of a business: the performance characteristics of its products
relative to industry standards; the direct and indirect costs of manufacturing its products; and its
need or ability to comply with environmental regulations. The final evaluation of a technology
involves considering these traditional competitiveness issues along with issues that business
leaders now know are equally important issues: the health and environmental impacts of
alternative products, processes, and technologies.
-------
Performance
The performance of the surface finishing technologies was tested using production run
tests following a strict testing protocol. Functional test boards were fabricated using a complex
test board design (a modified version of the IPC-B-24 board) developed by the Circuit Card
Assembly and Materials Task Force (CCAMTF). A surface finish was then applied to test boards
at each of thirteen volunteer PWB manufacturing facilities. Test boards were then collected
together and assembled at an assembly facility, using either a halide-free low-residue flux or a
halide-containing water-soluble flux, before being tested under thermal and mechanical stress,
and accelerated aging conditions. Additional residue testing was conducted to determine the
mechanism of failure. The test methods used to evaluate performance were intended to indicate
characteristics of a technology's performance, not to define parameters of performance or to
substitute for thorough on-site testing; the study was intended to be a "snapshot" of the
technologies. The Performance Demonstration was conducted with extensive input and
participation from PWB manufacturers, their suppliers, and PWB testing laboratories. The testing
protocol was designed to be consistent with the industry-led CCAMTF testing of surface finishes.
The technologies tested included HASL (baseline), nickel/gold, nickel/palladium/gold,
OSP, immersion silver, and immersion tin. The test vehicle measured roughly 6" x 5.8" x 0.062"
and was designed to contain at least 80 percent of the circuitry used in military and commercial
electronics. The test vehicle was also designed to be representative of a variety of circuits,
including high current low voltage (HCLV), high voltage low current (HVLC), high speed digital
(HSD), high frequency (HF), stranded wire (SW) and other networks, which were used to
measure current leakage. Overall, the vehicle provided 23 separate electrical responses for testing
the performance of the surface finish. Types of electrical components in the HCLV, HVLC,
HSD, and HF circuits included both plated through hole (PTH) and surface mounted
components.
Test sites were submitted by suppliers of the technologies, and included production
facilities and supplier testing facilities. Because the test sites were not chosen randomly, the
sample may not be representative of all PWB manufacturing facilities (although there is no
specific reason to believe that they are not representative). In addition, the number of test sites
for each technology ranged from one to four. Due to the smaller number of test sites for some
technologies, statistical relevance could not be determined.
The results of the performance testing showed that all of the surface finishes under study
were very robust to the environmental exposures, with two exceptions. Failures during the
mechanical shock testing, resulting in the separation of the surface mount components, were
attributable to the severity of the testing, and spread evenly across all finishing technologies,
including the baseline HASL process. Failures in the high frequency, low pass filter circuits,
resulting from open PTH, were found to be attributable to a combination of board fabrication
materials and board design. From an overall contamination standpoint, the five non-HASL
surface finishes performed as well, if not better than the HASL finish. The few solder joint
cracking failures were greater with the HASL finish, than with the alternative finishes.
7-9
-------
Cost
Comparative costs were estimated using a hybrid cost model that combined traditional
costs with simulation modeling and activity-based costs. The cost model was designed to
determine the total cost of processing a specific amount of PWB through a fully operational
surface finishing line, in this case, 260,000 surface square feet (ssf). Total costs were divided by
the throughput to determine a unit cost in $/ssf. Costs not related to the steady-state operation of
the surface finishing line, such as start-up costs or the costs of process changes required to other
process to implement a change in surface finishing technology, can vary widely by facility and
were not estimated by the model.
The cost components considered include capital costs (primary equipment & installation
costs, and facility costs), materials costs (limited to chemical costs), utility costs (water,
electricity, and natural gas costs), wastewater cost (limited to wastewater discharge cost),
production costs (production labor and chemical transport costs), and maintenance costs (tank
cleanup, bath setup, sampling and analysis, and filter replacement costs). Other cost components
may contribute significantly to overall costs, but were not quantified because they could not be
reliably estimated. These include wastewater treatment cost, sludge recycling and disposal cost,
other solid waste disposal costs, and quality costs (i.e., costs from decreased production
efficiency due to boards that do not meet quality specifications). However, Performance
Demonstration results indicate that each surface finishing technology has the capability to
achieve comparable levels of performance to HASL. Thus, quality costs are not expected to
differ among the alternatives.
Table 7-6 presents results of the cost analysis. The results indicate that all of the surface
finishing alternatives were more economical than the baseline non-conveyorized HASL process,
with the exception of the two technologies containing gold, an expensive precious metal. Unit
costs ranged from $0.10/ssf for the conveyorized OSP process to $1.54/ssf for the non-
conveyorized nickel/palladium/gold process. Three processes had a substantial cost savings of at
least 50 percent of the cost per ssf over that of the baseline HASL process (conveyorized OSP at
72 percent, non-conveyorized OSP at 69 percent, and non-conveyorized immersion tin at 50
percent). Three other process alternatives realized a somewhat smaller cost savings over the
baseline HASL process (conveyorized immersion tin at 31 percent, conveyorized immersion
silver at 22 percent, and the conveyorized HASL process at 3 percent).
In general, conveyorized processes cost less than non-conveyorized processes of the
same technology due to the cost savings associated with their higher throughput rates. The lone
exception, immersion tin, was more costly because the combination of process cycle time and
conveyor length resulted in a lower throughput rate than its non-conveyorized version.
Chemical cost was the single largest component cost for all of the nine processes. Labor
costs were the second largest cost component, though far less than the cost of process chemicals.
7-10
-------
Table 7-6. Cost of Surface Finishing Technologies
Cost Category
Capital Cost
Material Cost
Utility Cost
Wastewater Cost
Production Cost
Maintenance Cost
Cost Components
Primary Equipment & Installation
Facility
Chemicals
Water
Electricity
Natural Gas
Wastewater Discharge
Transportation of Material
Labor for Line Operation
Tank Cleanup
Bath Setup
Sampling and Testing
Filter Replacement
Total Cost
Unit Cost ($/ssf)
HASL
(NC)
$9,360
$432
$74,800
$706
$669
$88
$1,100
$167
$3,940
$1,210
$272
$499
$967
$94,200
$0.36
HASL
(C)
$11,100
$398
$75,200
$565
$452
$45
$851
$130
$1,790
$938
$211
$249
$482
$92,400
$0.35
Nickel/Gold
(NC)
$7,260
$2,930
$109,000
$1,180
$2,360
$0
$2,050
$668
$19,100
$4,820
$1,090
$3,530
$1,580
$156,000
$0.60
Cost Category
Capital Cost
Material Cost
Utility Cost
Wastewater Cost
Production Cost
Maintenance Cost
Cost Components
Primary Equipment & Installation
Facility
Chemicals
Water
Electricity
Natural Gas
Wastewater Discharge
Transportation of Material
Labor for Line Operation
Tank Cleanup
Bath Setup
Sampling and Testing
Filter Replacement
Total Cost
Unit Cost ($/ssf)
Nickel/Palladium/Gold
(NC)
$15,400
$6,090
$321,000
$2,060
$4,050
$0
$3,530
$1,030
$25,200
$7,440
$1,680
$8,900
$2,830
$399,000
$1.54
OSP
(NC)
$1,640
$313
$18,500
$441
$313
$67
$704
$158
$3,170
$1,140
$257
$1,610
$330
$28,700
$0.11
OSP
(C)
$2,880
$264
$18,800
$301
$208
$32
$462
$121
$1,320
$871
$196
$738
$151
$26,300
$0.10
7-11
-------
Table 7-6. Cost of Surface Finishing Technologies (cont.)
Cost Category
Capital Cost
Material Cost
Utility Cost
Wastewater Cost
Production Cost
Maintenance Cost
Cost Components
Primary Equipment & Installation
Facility
Chemicals
Water
Electricity
Natural Gas
Wastewater Discharge
Transportation of Material
Labor for Line Operation
Tank Cleanup
Bath Setup
Sampling and Testing
Filter Replacement
Total Cost
Unit Cost ($/ssf)
Immersion
Silver (C)
$10,500
$937
$52,700
$301
$739
$140
$529
$167
$5,260
$1,210
$272
$937
$80
$73,800
$0.28
Immersion
Tin (NC)
$2,950
$892
$29,000
$1,030
$494
$162
$1,620
$204
$6,780
$1,470
$332
$1,260
$705
$46,900
$0.18
Immersion
Tin (C)
$16,800
$2,340
$28,900
$702
$1,230
$240
$1,215
$167
$8,770
$1,210
$272
$1,800
$1,000
$64,700
$0.25
Regulatory Status
Discharges of surface finishing chemicals may be restricted by federal, state, or local air,
water, or solid waste regulations, and releases may be reportable under the federal Toxics Release
Inventory program. Federal environmental regulations were reviewed to determine the federal
regulatory status of surface finishing chemicals.3 Table 7-7 lists the number of chemicals used in
a surface finishing technology with federal environmental regulations restricting or requiring
reporting of their discharges. Different chemical suppliers of a technology do not always use the
same chemicals in their particular product lines. Thus, all of these chemicals may not be present
in any one product line.
3 In some cases, state or local requirements may be more restrictive than federal requirements. However, due to
resource limitations, only federal regulations were reviewed.
7-12
-------
Table 7-7. Regulatory Status of Surface Finishing Technologies
Process
Chemical
HASL
Nickel/Gold
Nickel/Palladium/Gol
d
OSP
Immersion Silver
Immersion Tin
Number of Chemicals Subject to Applicable Regulation
CWA
304b
1
6
5
2
1
1
307a
1
6
5
2
1
1
311
4
16
12
5
5
6
Priority
Pollutant
1
6
5
2
1
1
CAA
111
3
11
5
3
1
3
112b
3
6
5
2
1
2
112r
1
1
1
1
-
1
EPCRA
313
6
12
10
5
313
7
110
1
7
6
2
1
302a
3
3
3
2
2
TSCA
8d
HSDR
3
1
1
1
-
2
MTL
4
4
4
2
1
4
8a
PAIR
3
3
4
1
1
3
RCRA Waste
P
-
-
-
-
-
-
U
-
-
-
-
-
2
Abbreviations and definitions:
CAA - Clean Air Act
CAA 111 - Standards of Performance for New Stationary Sources of
Air Pollutants -Equipment Leaks Chemical List
CAA 112b - Hazardous Air Pollutant
CAA 112r - Risk Management Program
CWA - Clean Water Act
CWA 304b - Effluent Limitations Guidelines
CWA 307a - Toxic Pollutants
CWA 311 - Hazardous Substances
CWA - Priority Pollutants
EPCRA - Emergency Planning and Community Right-to-Know Act
EPCRA 302a - Extremely Hazardous Substances
EPCRA 313 - Toxic Chemical Release Inventory
RCRA - Resource Conservation and Recovery Act
RCRA P Waste - Listed acutely hazardous waste
RCRA U Waste - Listed hazardous waste
SARA - Superfund Amendments and Reauthorization Act
SARA 110 - Superfund Site Priority Contaminant
SDWA - Safe Drinking Water Act
SDWA NPDWR - National Primary Drinking Water Rules
SDWA NSDWR - National Secondary Drinking Water Rules
TSCA - Toxic Substances Control Act
TSCA 8d HSDR - Health & Safety Data Reporting Rules
TSCA MTL - Master Testing List
TSCA 8a PAIR - Preliminary Assessment Information Rule
-------
7.1.3 Resource Conservation Summary
Resources typically consumed by the operation of the surface finishing process include
water used for rinsing panels, process chemicals used in the process line, energy used to heat
process baths and power equipment, and wastewater treatment chemicals. A quantitative
analysis of the energy and water consumption rates of the surface finishing process alternatives
was performed to determine if implementing an alternative to the baseline process would reduce
consumption of these resources during the manufacturing process. A quantitative analysis of
both process chemical and treatment chemical consumption could not be performed due to the
variability of factors that affect the consumption of these resources. Section 5.1 discusses the
role that the surface finishing process has in the consumption of these resources and the factors
affecting the consumption rates.
The relative water and energy consumption rates of the surface finishing process
alternatives were determined as follows:
• the daily water consumption rate and hourly energy consumption rate of each alternative
were determined based on data collected from the PWB Workplace Practices
Questionnaire;
• the operating time required to produce 260,000 ssf of PWB was determined using
computer simulations models of each of the alternatives; and
• the water and energy consumption rates per ssf of PWB were calculated based on the
consumption rates and operating times.
Table 7-8 presents the results of these analyses.
Table 7-8. Energy and Water Consumption Rates of Surface Finishing Alternatives
Process Type
HASL, Non-conveyorized (BASELINE)
HASL, Conveyorized
Nickel/Gold, Non-conveyorized
Nickel/Palladium/Gold, Non- convey orized
OSP, Non-conveyorized
OSP, Conveyorized
Immersion Silver, Conveyorized
Immersion Tin, Non-conveyorized
Immersion Tin, Conveyorized
Water Consumption
(gal/ssf)
1.24
0.99
2.06
3.61
0.77
0.53
0.53
1.81
0.88
Energy Consumption
(Btu/ssf)
218
133
447
768
125
73
287
289
522
7-14
-------
The water consumption rates for the surface finishing alternatives ranged from a low of
0.53 gal/ssf for the immersion silver and OSP conveyorized processes to a high of 3.6 gal/ssf for
the non-conveyorized nickel/palladium/gold process. Several processes were found to consume
less water then the HASL baseline, including conveyorized versions of the immersion silver and
immersion tin technologies, along with both versions of the OSP process. Conveyorized
processes were found to consume less water than non-conveyorized versions of the same
process. Primary factors influencing the water consumption rate included the number of rinse
tanks and the overall efficiency of the conveyorized processes.
The energy consumption rates for the surface finishing alternatives ranged from 73
Btu/ssf for the conveyorized OSP process to 768 Btu/ssf for the non-conveyorized
nickel/palladium/gold process. The results indicate that three surface finishing processes are
more energy efficient than the traditional non-conveyorized HASL process (conveyorized HASL,
non-conveyorized OSP, and conveyorized OSP), while two others are roughly comparable
(conveyorized immersion silver and non-conveyorized immersion tin). It was also found that for
alternatives with both types of automation, the conveyorized version of the process is typically
the more energy efficient (HASL and OSP), with the notable exception of the immersion tin
process.
An analysis of the impacts directly resulting from the consumption of energy by the
surface finishing process showed that the generation of the required energy has environmental
impacts. Pollutants released to air, water, and soil can result in damage to both human health and
the environment. The consumption of natural gas tends to result in releases to the air which
contribute to odor, smog, and global warming, while the generation of electricity can result in
pollutant releases to all media with a wide range of possible effects. Minimizing the amount of
energy usage by the surface finishing process, either by selection of a more energy efficient
process or by adopting energy efficient operating practices, will decrease the quantity of
pollutants released into the environment resulting from the generation of the energy consumed.
Metals are another natural resource consumed by the surface finishing process. The rate
of deposition of metal was calculated for each technology along with the total amount of metal
consumed for 260,000 ssf of PWB produced, the average annual PWB production rate reported
by facilities using HASL. It was shown that the consumption of close to 300 pounds of lead
could be eliminated by replacing the baseline HASL process with an alternative technology (see
Section 5.1, Resource Conservation). In cases where waste solder is not routinely recycled or
reclaimed, the consumption of as much as 2,500 pounds of lead could be eliminated by
replacement of the HASL process. Although several of the alternative technologies rely on the
use of small quantities of other metals (especially nickel, palladium, gold, silver, and tin) the OSP
technology eliminates metal consumption entirely.
7-15
-------
7.2 SOCIAL BENEFITS/COSTS ASSESSMENT
7.2.1 Introduction to Social Benefits/Costs Assessment
Social benefits/costs analysis4 is a tool used by policy makers to systematically evaluate the
impacts to all of society resulting from individual decisions. The decision evaluated in this
analysis is the choice of a surface finishing technology. PWB manufacturers have a number of
criteria they may use to assess which surface finishing technology they will use. For example, a
PWB manufacturer might ask what impact their choice of a surface finishing alternative might
have on operating costs, compliance costs, liability costs, and insurance premiums. This business
planning process is unlike social benefit/cost analysis, however, because it approaches the
comparison from the standpoint of the individual manufacturer and not from the standpoint of
society as a whole.
A social benefits/costs analysis seeks to compare the benefits and costs of a given action,
while considering both the private and external costs and benefits.5 Therefore, the analysis will
consider both the impact of the alternative surface finishing processes on the manufacturer itself
(private costs and benefits) and the impact the choice of an alternative has on external costs and
benefits, such as environmental damage and the risk of illness for the general public. External
costs are not borne by the manufacturer, but by society. Table 7-9 defines a number of terms
used in benefit/cost analysis, including external costs and external benefits.
4 The term "analysis" is used here to refer to a more quantitative analysis of social benefits and costs, where a
monetary value is placed on the benefits and costs to society of individual decisions. Examples of quantitative
benefits/costs analyses are the regulatory impact analyses done by EPA when developing federal environmental
regulations. The term "assessment" is used here to refer to a more qualitative examination of social benefits and
costs. The evaluation performed in the CTSA process is more correctly termed an assessment because many of the
social benefits and costs of the surface finishing technologies are identified, but not monetized.
5 Private costs typically include any direct costs incurred by the decision-maker and are generally reflected in
the manufacturer's balance sheet. In contrast, external costs are incurred by parties other than the primary
participants to the transaction. Economists distinguish between private and external costs because each will affect
the decision-maker differently. Although external costs are real costs to some members of society, they are not
incurred by the decision-maker and firms do not normally take them into account when making decisions. A
common example of these "externalities" is the electric utility whose emissions are reducing crop yields for the
farmer operating downwind. The external costs experienced by the farmer in the form of reduced crop yields are
not considered by the utility when making decisions regarding electricity production. The farmer's losses do not
appear on the utility's balance sheet.
7-16
-------
Table 7-9. Glossary of Benefits/Costs Analysis Terms
Term
Definition
Exposed
Population
The estimated number of people from the general public or a specific population
group who are exposed to a chemical through wide dispersion of the chemical in the
environment (e.g., DDT). A specific population group could be exposed to a
chemical due to its physical proximity to a manufacturing facility (e.g., residents who
live near a facility using a chemical), use of the chemical or a product containing a
chemical, or through other means.
Exposed Worker
Population
The estimated number of employees in an industry exposed to the chemical, process,
and/or technology under consideration. This number may be based on market share
data as well as estimations of the number of facilities and the number of employees in
each facility associated with the chemical, process, and/or technology under
consideration.
Externality
A cost or benefit that involves a third party who is not a part of a market transaction;
"a direct effect on another's profit or welfare arising as an incidental by-product of
some other person's or firm's legitimate activity" (Mishan, 1976). The term
"externality" is a general term which can refer to either external benefits or external
costs.
External Benefits
A positive effect on a third party who is not a part of a market transaction. For
example, if an educational program results in behavioral changes which reduce the
exposure of a population group to a disease, then an external benefit is experienced
by those members of the group who did not participate in the educational program.
For the example of non-smokers exposed to second-hand smoke, an external benefit
can be said to result when smokers are removed from situations in which they expose
non-smokers to tobacco smoke.
External Costs
A negative effect on a third party who is not part of a market transaction. For
example, if a steel mill emits waste into a river which poisons the fish in a nearby
fishery, the fishery experiences an external cost as a consequence of the steel
production. Another example of an external cost is the effect of second-hand smoke
on non-smokers.
Human Health
Benefits
Economic benefit from reduced health risks to workers in an industry or business as
well as to the general public as a result of switching to less toxic or less hazardous
chemicals, processes, and/or technologies. An example would be switching to a less
volatile organic compound, lessening worker inhalation exposures as well as
decreasing the formation of photochemical smog in the ambient air.
Human Health
Costs
The cost of adverse human health effects associated with production, consumption,
and disposal of a firm's product. An example is respiratory effects from stack
emissions, which can be quantified by analyzing the resulting costs of health care and
the reduction in life expectancy, as well as the lost wages as a result of being unable
to work.
Illness
Costs
A financial term referring to the liability and health care insurance costs a company
must pay to protect itself against injury or disability to its workers or other affected
individuals. These costs are known as illness benefits to the affected individual.
Indirect Medical
Costs
Indirect medical costs associated with a disease or medical condition resulting from
exposure to a chemical or product. Examples would be the decreased productivity of
patients suffering a disability or death and the value of pain and suffering borne by
the afflicted individual and/or family and friends.
7-17
-------
Term
Definition
Private
(Internalized)
Costs
The direct costs incurred by industry or consumers in the marketplace. Examples
include a firm's cost of raw materials and labor, a firm's costs of complying with
environmental regulations, or the cost to a consumer of purchasing a product.
Social
Costs
The total cost of an activity that is imposed on society. Social costs are the sum of
the private costs and the external costs. Therefore, in the example of the steel mill,
social costs of steel production are the sum of all private costs (e.g., raw material and
labor costs) and the sum of all external costs (e.g., the costs associated with the
poisoned fish).
Social
Benefits
The total benefit of an activity that society receives (i.e., the sum of the private
benefits and the external benefits). For example, if a new product yields pollution
prevention opportunities (e.g., reduced waste in production or consumption of the
product), then the total benefit to society of the new product is the sum of the private
benefit (value of the product that is reflected in the marketplace) and the external
benefit (benefit society receives from reduced waste).
Willingness-to-Pay
Estimates used in benefits valuation are intended to encompass the full value of
avoiding a health or environmental effect. For human health effects, the components
of willingness-to-pay include the value of avoiding pain and suffering, impacts on the
quality of life, costs of medical treatment, loss of income, and, in the case of
mortality, the value of life.
7.2.2 Benefits/Costs Methodology and Data Availability
The methodology for conducting a social benefits/costs assessment can be broken down
into four general steps: 1) obtain information on the relative human and environmental risk,
performance, cost, process safety hazards, and energy and natural resource requirements of the
baseline and the alternatives; 2) construct matrices of the data collected; 3) when possible,
monetize the values presented within the matrices; and 4) compare the data generated for the
alternative and the baseline in order to produce an estimate of net social benefits. Section 7.1
presented the results of the first task by summarizing risk, competitiveness, and conservation
information for the baseline and alternative surface finishing technologies. Section 7.2.3 presents
information relevant to private and external benefits and costs, in matrix form and in monetary
terms where possible. Section 7.2.4 presents the private and external benefits and costs together
to produce an estimate of net social benefits.
Ideally, the analysis would quantify the social benefits and costs of using the alternative
and baseline surface finishing technologies, allowing identification of the technology whose use
results in the largest net social benefit. However, because of resource and data limitations and
because individual users of this CTSA will need to apply results to their own particular situations,
the assessment presents a qualitative description of the risks and other external effects associated
with each substitute technology compared to the baseline. Benefits derived from a reduction in
risk are described and discussed, but not quantified. Nonetheless, the information presented can
be very useful in the decision-making process. A few examples are provided to qualitatively
illustrate some of the benefit considerations. Personnel in each individual facility will need to
examine the information presented, weight each piece according to facility and community
characteristics, and develop an independent choice.
7-18
-------
7.2.3 Social Benefits/Costs Associated with Choice of Surface Finishing Alternative
The selection of a surface finish results in costs and benefits to society, in the form of
both private and external costs and benefits. For example, an alternative that releases less toxic
chemicals into the workplace air results in both private and external benefits. The manufacturer
pays less for health care costs and worker sick time, while workers benefit from working in a
healthier environment. Society as a whole benefits from a more competitive company in the
marketplace and from reduced long-term health care costs; in other words, from the cumulative
affect of the benefits or costs, both the private and external. This type of example is why
particular aspects of the surface finishing process are discussed in terms of both private benefits
and costs and external benefits and costs.
Private and/or external costs and benefits may occur in a number of areas, including:
• manufacturing
• occupational health/worker risk;
• public health/population risk;
• wastewater contaminants and ecological risk; and
• energy and natural resource consumption.
Table 7-10 presents an overview of potential private benefits or costs and external benefits
or costs associated with the evaluated areas. Each of these is discussed in turn below. While it is
difficult to obtain an overall number to express the private benefits and costs of alternative surface
finishing processes, some data were quantifiable, such as manufacturing costs. However, in order
to determine the overall private benefit/cost comparison, a qualitative discussion of the data is
also necessary. Following the discussion of manufacturing costs are discussions of costs
associated with occupational and population health risks and other costs or benefits that could not
be put in terms of monetary equivalents, but are important to the decision-making process.
Manufacturing
The cost of manufacturing is considered strictly a private cost, with little or no bearing on
social costs and benefits. The cost analysis estimated the average manufacturing costs of the
surface finishing technologies for several categories of costs. Results of the cost analysis are
shown in Table 7-11. Results show that implementation of several of the alternative processes are
likely to result in reduced private costs to the manufacturing facility, and that reductions were
primarily due to the lower cost of process chemicals between surface finish processes. Other cost
components may contribute significantly to overall private costs for a surface finish, but were not
quantified because they could not be reliably estimated. These include wastewater treatment
cost, sludge recycling and disposal cost, other solid waste disposal costs, and quality costs. Refer
to Chapter 4.2, Cost Assessment, for a more detailed discussion of the methodology and results
of the cost assessment for surface finish alternatives.
7-19
-------
Table 7-10. Potential Private Benefits or Costs Associated with the Selection
of a Surface Finish Technology
Evaluation
Category
Manufacturing Costs
Occupational Health/
Worker Risk
Public Health/
Population Risk
Wastewater and
Ecological Risk
Energy Use
Water Use
Private Benefit or Cost a
Capital costs.
Materials (chemical) costs.
Utility costs.
Wastewater discharge costs.
Production cost.
Maintenance costs.
Worker sick days.
Worker efficiency.
Health insurance costs to the PWB
manufacturer.
Potential liability costs.
Treatment costs to meet wastewater
permit requirements.
Possible fines if permits are
violated.
Increased liability costs.
Direct costs from the use of energy
in the manufacturing process.
Direct costs from the use of water
in the manufacturing process.
External Benefit or Cost a
Not Applicable
Long-term medical costs to workers.
Pain and suffering associated with work-
related illness.
Long-term medical costs.
Pain and suffering associated with illness.
Loss of ecosystem diversity.
Reduction in the recreational value of
streams and rivers.
Increased air emissions.
Depletion of natural resources.
Water costs for the surrounding area.
Costs paid to treatment facilities to clean
the water.
Changes to water quality available to
society.
a A benefit would be a change in a beneficial direction (e.g.,
detrimental change (e.g., 0«ue=>\eworker sick days).
Due=>\eazspitol costs), while a cost would be a
7-20
-------
Table 7-11. Overall Manufacturing Cost Comparison
Process
HASL, Non-conveyorized
HASL, Conveyorized
Nickel/Gold, Non-conveyorized
Nickel/Palladium/Gold, Non-conveyorized
OSP, Non-Conveyorized
OSP, Conveyorized
Immersion Silver, Conveyorized
Immersion Tin, Non-conveyorized
Immersion Tin, Conveyorized
Estimated Cost to Manufacture 260,000 ssf
($/ssf)
$0.36
$0.35
$0.60
$1.54
$0.11
$0.10
$0.28
$0.18
$0.25
Private Benefits/Costs. Reductions in the cost of manufacturing are reflected primarily in
reduced private costs for the PWB manufacturer. Implementation of an alternative surface finish
can potentially result in significant operating cost savings for a manufacturing facility, as shown
above. Decreased manufacturing costs allow companies more operating flexibility and are critical
to the long-term ability of the manufacturer to remain competitive in the global marketplace.
External Benefits/Costs. There are no significant external benefits derived directly from
the cost of manufacturing. However, several aspects that affect the manufacturing cost of the
process result in external benefits. For instance, the conservation of water or material results in a
more sustainable operating process with reduced environmental burdens that must be borne by
society. See the discussion of cost and benefits based on energy and natural resource
consumption presented later in this section for a more complete discussion of the external benefits
Costs and Benefits Based on Occupational Health
Operation of the surface finish process requires workers to work in close contact with
chemicals, some of which may pose a threat to occupational health. Unacceptably high risks to
workers from chemicals in the workplace may hurt company and worker alike. The reduction of
risks to workers through the implementation of an alternative surface finish can result in tangible
benefits, both private and external.
Health risks to workers were estimated for inhalation exposure to vapors and aerosols
from surface finishing baths, and for dermal exposure to surface finishing bath chemicals. Worker
risk to chemicals were compared to EPA guidelines for acceptable risk to identify chemicals of
concern within the workplace. Occupational cancer risks were estimated for inhalation exposure
to inorganic metallic salt A, a suspected or probable human carcinogen in the non-conveyorized
nickel/gold process. The cancer risks to worker health from inorganic metallic salt A are below
the EPA concern level of one in one million for inhalation exposure. Occupational cancer risks
associated with other suspected carcinogens could not be quantified because cancer slope factors
have not yet been established for these chemicals.
7-21
-------
Table 7-12 summarizes the number of chemicals of concern for the exposure pathways
evaluated and lists the number of suspected carcinogens in each technology. Table 7-13 lists
potential health effects associated with surface finishing chemicals of concern. Detailed
descriptions of the risk assessment methodology and results are presented in Chapter 3, while the
risk results are also summarized in Chapter 7.1.
Table 7-12. Summary of Occupational Hazards, Exposures, and Risks of
Potential Concern
Surface Finishing Technology
HASL, Non-conveyorized (BASELINE)
HASL, Conveyorized
Nickel/Gold, Non-conveyorized
Nickel/Palladium/Gold, Non-conveyorized
OSP, Non-conveyorized
OSP, Conveyorized
Immersion Silver, Conveyorized
Immersion Tin, Non-conveyorized
Immersion Tin, Conveyorized
No. of Chemicals of
Concern by Pathway a
Inhalation b
1
0
5
6
1
0
0
0
0
Dermal °
1
1
6
6
3
3
0
1
0
No. of
Suspected
Carcinogens d
2
2
3
1
1
1
1
1
1
a Number of chemicals of concern for a surface finishing line operator (the most exposed individual).
Occupational health risks could not be quantified for one or more chemicals in each surface finish due to lack of
toxicity or chemical property data. See Chapter 3.3 for a more detailed explanation.
b See Table 3-30 for further information on inhalation risks.
c See Table 3-31 for further information dermal risks.
d See Table 3-21 for further information on cancer classifications.
Health endpoints potentially associated with surface finishing chemicals of concern include:
• skin, eye, nose, throat, and respiratory irritation or damage;
• allergic contact dermatitis;
• gastrointestinal/digestive pain or damage;
• kidney damage;
• liver damage; and
• damage to the nervous system and immune system.
Based on the number of chemicals with risk results above concern levels (Table 7-12),
some alternatives to the non-conveyorized HASL process may have private and external benefits
resulting from reduced occupational risks. These alternatives include the Conveyorized HASL,
Conveyorized immersion silver, and Conveyorized and non-conveyorized immersion tin processes.
Some alternatives, however, may have increased costs due to higher risks; these include the non-
conveyorized nickel/gold and nickel/palladium/gold processes. Potential risks from Conveyorized
and non-conveyorized OSP are similar to those of non-conveyorized HASL.
7-22
-------
It is important to note that surface finishing chemicals are not the only factor contributing toward
the illnesses described in Table 7-13; other PWB manufacturing process steps may also contribute
toward adverse worker health effects. With the exception of determining the cancer risk from
inorganic metallic salt A, the risk characterization did not link exposures of concern with
particular adverse health outcomes or with the number of incidences of adverse health outcomes.6
Thus, the benefits or costs of illnesses avoided by switching to a surface finishing alternative could
not be quantified
Private Costs/Benefits. There are potential economic benefits associated with reduced
exposure to surface finishing chemicals. Private benefits for PWB manufacturers may include
increased worker productivity, increased worker morale, reduced worker absenteeism due to
illness, and a reduction in liability and health care insurance costs. While reductions in insurance
premiums as a result of pollution prevention are not currently widespread, the opportunity exists
for changes in the future.
External Costs/Benefits. External benefits are not as easily quantifiable, but no less
important than the private benefits listed above. Many of the health endpoints described in Table
7-13 lead to long-term illnesses in workers that result in hardship for the entire family. Many
states are struggling under the economic burden of providing adequate health care to an aging
population using an overburdened health care system experiencing rapidly increasing health care
costs. External benefits of a switch to an alternative surface finish system may include reductions
in illness to workers along with the associated decreases in both short-term and long-term medical
costs and insurance premiums. Other benefits include a higher quality of life for workers and their
families.
Table 7-13. Potential Health Effects Associated with Surface Finishing
Chemicals of Concern
Chemical of
Concern
Ammonium
chloride
Ammonia
compound A
Ammonium
hydroxide
Alkyldiol
Alternatives with
Exposure Levels of
Concern
Nickel/Gold
Nickel/Palladium/Gold
Nickel/Gold,
Nickel/Palladium/Gold
Nickel/Gold,
Nickel/Palladium/Gold
Pathway
of
Concern a
Dermal
Dermal
Dermal
Inhalation
Potential Health Effects
Contact with ammonium chloride solution or
fumes irritate the eyes. Large doses of
ammonium chloride may cause nausea, vomiting,
thirst, headache, hyperventilation, drowsiness,
and altered blood chemistry. Ammonia fumes are
extremely irritating to skin, eyes, and respiratory
passages. The severity of effects depends on the
amount of dose and duration of exposure.
Can affect the respiratory system if inhaled, and
kidneys if absorbed into the body.
6 Cancer risk from inorganic metallic salt A exposure was expressed as a probability, but the exposure
assessment did not determine the size of the potentially exposed population (e.g., number of surface finishing line
operators and others working in the process area). This information would be necessary to estimate the number of
illnesses avoided by switching to an alternative from the baseline.
7-23
-------
Chemical of
Concern
Copper ion and
copper salt C
Copper sulfate
pentahydrate
Ethylene glycol
Hydrochloric
acid
Hydrogen
peroxide
Inorganic
metallic salt B
Nickel sulfate
Phosphoric acid
Propionic acid
Urea compound
C
Alternatives with
Exposure Levels of
Concern
OSP
HASL, Nickel/Gold,
Nickel/Palladium/Gold,
OSP
HASL,
OSP
Nickel/Gold,
Nickel/Palladium/Gold
Nickel/Gold,
Nickel/Palladium/Gold
Nickel/Gold,
Nickel/Palladium/Gold
Nickel/Gold,
Nickel/Palladium/Gold
Nickel/Gold,
Nickel/Palladium/Gold
Nickel/Gold,
Nickel/Palladium/gold
Nickel/Gold,
Nickel/Palladium/Gold
Nickel/Palladium/Gold
Immersion Tin
Pathway
of
Concern a
Dermal
Dermal
Inhalation
Inhalation
Inhalation
Dermal
Dermal
Inhalation
Dermal
Inhalation
Inhalation
Dermal
Potential Health Effects
Long-term exposure to high levels of copper may
cause liver damage. Copper is not known to
cause cancer. The seriousness of the effects of
copper can be expected to increase with both level
and length of exposure.
In humans, low levels of vapors produce throat
and upper respiratory irritation. When ethylene
glycol breaks down in the body, it forms
chemicals that crystallize and can collect in the
body, which prevent kidneys from working. The
seriousness of the effects can be expected to
increase with both level and length of exposure.
Hydrochloric acid in air can be corrosive to the
skin, eyes, nose, mucous membranes, respiratory
tract, and gastrointestinal tract.
Hydrogen peroxide in air can irritate the skin,
nose, and eyes. Ingestion can damage the liver,
kidneys, and gastrointestinal tract.
Exposure to this material can damage the nervous
system, kidneys, and immune system.
Skin effects are the most common effects in
people who are sensitive to nickel. Workers who
breathed very large amounts of nickel compounds
have developed lung and nasal sinus cancers.
Inhaling phosphoric acid can damage the
respiratory tract.
No data were located for health effects of
propionic acid exposure in humans, although
some respiratory effects were seen in laboratory
mice.
Dermal exposure to urea compound C has
resulted in allergic contact dermatitis in workers,
and exposure has caused weight loss in mice.
Inhalation concerns only apply to non-conveyorized processes. Dermal concerns may apply to non-conveyorized
and/or conveyorized processes (see Table 7-3).
7-24
-------
Costs and Benefits Based on Public Health
In addition to worker exposure, members of the general public in close physical proximity
to a PWB plant may be exposed to surface finishing chemicals dispersed into the air. Both private
and external cost savings could be realized if an alternative surface finish reduced public health
risks.
Public health risk was estimated for inhalation exposure for the general populace living
near a facility. Risk was characterized for long-term ambient exposures to the population, rather
than short-term exposures to high levels of hazardous materials (e.g., fire, spill). The risk
indicators for ambient exposures to humans, although limited to airborne releases, indicated low
concern from the estimated air concentrations for chronic non-cancer effects. The excess cancer
risks were also found to be well below EPA concern levels (one in 50 billion). Refer to Chapter 3
for a description of the risk assessment methodology and results.
These results suggest little change in public health risks would result from a switch to an
alternative surface finish technology. While the study found little difference among the
alternatives for those public health risks that were assessed, it was not within the scope of this
comparison to assess all community health risks. Risk was not characterized for exposure via
other pathways (e.g., drinking water, fish ingestion) or short-term or long-term exposures to high
levels of hazardous chemicals when there is a spill, fire, or other periodic release.
Private Costs/Benefits. Private benefits could result from reductions in potential liability
costs resulting from adverse effects of emissions released from the facility into the environment.
Risk results for the nearby public from inhalation of air emissions from a PWB facility indicate
that no substantive difference in risk, and thus, liability cost would be realized. However, private
cost savings could result from reduced liability for other types of emissions (e.g. releases to
surface water) should they pose a threat to human health.
External Costs/Benefits. External benefits could result from reduced medical costs for
members of the public who become ill as a result of exposure to emissions from a nearby PWB
manufacturing facility. However, because the health risks from air emissions are all of low
concern, a change in alternatives would not be expected to result in significant changes to public
health. The effects of other emissions on the public, and the resulting differences in external
costs/benefits are unknown.
Costs and Benefits Based on Wastewater and Ecological Risks
Surface finishing chemicals in wastewater are potentially damaging to terrestrial and
aquatic ecosystems, resulting in private costs borne by manufacturers as well as external costs
borne by society. The CTSA evaluated the ecological risks of the baseline and alternatives for
aquatic life by calculating ecological risk indicators for non-metal surface finishing chemicals
(metals were assumed to be removed by treatment) that may be released to surface water.
Table 7-14 presents the number of chemicals in each technology with an estimated surface
water concentration above their CC. CCs are discussed in more detail in Section 3.3.3. These
7-25
-------
results suggest that all of the alternatives may pose lower private and external costs based on
wastewater contaminants and ecological risks than the baseline process.
Table 7-14. Number of Chemicals with Estimated Surface Water Concentration Above
Concern Concentration
Surface Finishing Technology
HASL, Non-conveyorized (BASELINE)
HASL, Conveyorized
Nickel/Gold, Non-conveyorized
Nickel/Palladium/Gold, Non-conveyorized
OSP, Non-conveyorized
OSP, Conveyorized
Immersion Silver, Conveyorized
Immersion Tin, Non-conveyorized
Immersion Tin, Conveyorized
No. of Chemicals
4
3
0
0
1
1
1
1
0
Private Costs/Benefits. The primary cost borne by the manufacturer is the cost of
pretreating the wastewater to meet wastewater permit requirements. Pretreatment could include
both in-line (e.g. electrowinning) or end-of-pipe treatment techniques (see Chapter 6.2). Other
potential private costs include possible fines if permits are violated and increased liability costs.
External Costs/Benefits. Pollution of streams and rivers can damage the aquatic
ecosystems, endangering species and reducing ecosystem diversity. Wastewater discharged to
streams and other surface waters, even if within permit levels, can have effects on the complex
ecosystems in ways that are difficult to predict. Reductions in chemicals of concern through the
adoption of alternative surface finish technologies preserves ecosystem diversity, while
maintaining the recreational value of surface waters for society.
Costs and Benefits Based on Energy and Natural Resources
Conservation of energy and natural resources has become a national priority with effects
on both society and the private sector. Energy shortages in western states have caused periodic
rolling blackouts responsible for large economic losses to companies, while at the same time
driving up energy costs for citizens and companies alike.
The natural resource and energy consumption of the surface finish technologies was
assessed in this CTSA. A detailed discussion of the methods used in evaluating individual
consumption rates is presented in Chapter 5, Conservation. Table 7-15 summarizes the water and
energy consumption rates and percent changes in consumption from the baseline to the surface
finishing alternatives. The results suggest that several of the alternatives use less water per ssf,
less energy per ssf, or both, than the baseline non-conveyorized HASL process. The
consumption rates of other natural resources, such as precious metals, were also evaluated in
Chapter 5.
7-26
-------
Table 7-15. Energy and Water Consumption of Surface Finishing Technologies
Surface Finishing Technology
HASL, Non-conveyorized (BASELINE)
HASL, Conveyorized
Nickel/Gold, Non-conveyorized
Nickel/Palladium/Gold, Non-conveyorized
OSP, Non-conveyorized
OSP, Conveyorized
Immersion Silver, Non-conveyorized
Immersion Tin, Non-conveyorized
Immersion Tin, Conveyorized
Water Consumption
gal/ssf
1.24
0.99
2.06
3.61
0.77
0.53
0.53
1.81
0.88
% change
-20
+66
+191
-38
-57
-57
+46
-29
Energy Consumption
Btu/ssf
218
133
447
768
125
73
287
263
522
% change
-39
+105
+252
-43
-66
+32
+21
+239
Private Costs/Benefits. Private benefits associated with the conservation of energy and
natural resources are reflected in reduced manufacturing costs for the process (see the discussion
of costs and benefits associated with manufacturing presented previously in this section).
Indirect private costs may occur in situations of extreme energy or water shortages,
affecting the availability and the cost of the resource required. Energy shortages in some western
states resulted in energy price increases and rolling blackouts that at times caused the complete
shut down of manufacturing facilities, and the loss of income associated with that shut down.
Conservation of energy protects the company and society from the affects of an energy crisis, and
acts to prevent another crisis from occurring.
External Costs/Benefits. While the private costs of natural resource and energy
consumption are reflected directly in the PWB manufacturers bottom line, the external costs and
benefits of conservation are no less tangible, becoming a key issue in the national and local debate
of public policy. Companies and governments worldwide are moving towards sustainable
production goals that will insure the continued availability of our natural resources, while
protecting the business and environmental climates.
Energy shortages have placed energy conservation on the front page of public discussion.
Reduced energy consumption through conservation results in the preservation of non-renewable
supplies of energy-producing raw materials such as coal, natural gas, or oil. Conservation also
acts to reduce air emissions resulting from the generation of energy, including compounds such as
carbon dioxide, nitrogen dioxide, carbon monoxide, sulfur oxide compounds (SOJ, and
particulate matter. Pollution resulting from the generation of energy consumed by surface finish
technologies was summarized in Table 5-11 in Section 5.2, Energy Impacts. Environmental and
human health concerns associated with these pollutants include global warming, smog, acid rain,
and health effects from toxic chemical exposure.
The use of water and consequent generation of wastewater also results in external costs to
society. While the private costs of this water usage are included in the cost estimates in Table 7-
7-27
-------
15, the external costs are not. Clean water is quickly becoming a scarce resource, and activities
that utilize water therefore impose external costs on society. Higher water costs, inadequate
water supplies, decreased water supply quality, and higher costs for public treatment facilities due
to increased sewage volumes are all potential external costs bourne by society as a result of
increased industrial water consumption.
Other Private Benefits and Costs
Table 7-16 gives additional examples of private costs and benefits that could not be
quantified in this CTSA. These include wastewater treatment, solid waste disposal, compliance,
and improvements in company image that accrue from implementing a substitute. Some of these
were mentioned above, but are included in the table due to their importance to overall benefits
and costs.
7.2.4 Summary of Benefits and Costs
The objective of a social benefits/costs assessment is to identify those technologies or
decisions that maximize net benefits. Ideally, the analysis would quantify the social benefits and
costs of using the alternative and baseline surface finishing technologies in terms of a single unit
(e.g., dollars) and calculate the net benefits of using an alternative instead of the baseline
technology. Due to data limitations, however, this assessment presents a qualitative description
of the benefits and costs associated with each technology compared to the baseline.
Each alternative presents a mixture of private and external benefits and costs. In terms of worker
health risks, conveyorized processes have the greatest benefits for reduced worker inhalation
exposure to bath chemicals; they are enclosed and vented to the atmosphere. However, dermal
contact from bath maintenance activities can be of concern regardless of the equipment
configuration for HASL and OSP processes, as well as non-conveyorized nickel/gold,
nickel/palladium/gold, and immersion tin processes. Little or no improvement is seen in public
health risks because concern levels were very low for all technologies. Differences in estimated
wastewater contaminant levels and aquatic risk concerns suggest that alternatives to non-
conveyorized HASL post lower potential private and external costs (or higher benefits).
Conveyorized processes consumed less water than that consumed by non-conveyorized processes,
resulting in net private and external benefits. Only the OSP technology, along with the
conveyorized HASL technology, are expected to reduce potential private and external costs of
energy consumption, resulting in increased social benefits.
Other benefits and costs discussed qualitatively include wastewater treatment, solid waste
disposal, compliance costs, and effects on the company image. The effects on jobs of wide-scale
adoption of an alternative was not evaluated in the CTSA.
7-28
-------
Table 7-16. Examples of Private Costs and Benefits Not Quantified
Category
Description of Potential Costs or Benefits
Wastewater
Treatment
Alternatives to the baseline HASL technology may provide cost savings by reducing
the quantity and improving the treatability of process wastewaters. In turn, these
cost savings can enable the implementation of other pollution prevention measures.
Several alternatives to the baseline process use less rinse water and, consequently,
produce less wastewater. However, some alternatives may also introduce additional
metals, such as silver or nickel, whicht are toxic to aquatic organisms. These
metals, which might not otherwise be present in the plant wastewater, may require
additional treatment steps. All of these factors contribute to both the private
benefits and costs of implementing a surface finishing alternative.
Solid Waste
Disposal
All of the alternatives result in the generation of sludge, off-specification PWBs,
and other solid wastes, such as spent bath filters or solder dross. These waste
streams must be recycled or disposed of, some of them as hazardous waste. For
example, many PWB manufacturers send the contaminated copper waste generated
by the HASL process to a recycler to reclaim the metal content. Solder wastes that
cannot be effectively reclaimed will likely be landfilled. It is likely that the
manufacturer will incur costs in order to recycle or landfill these solid wastes;
however, these costs were not quantified (reducing the volume and toxicity of solid
waste also provides social benefits to the community).
Compliance
Costs
The cost of complying with all environmental and safety regulations affecting the
surface finish process line was not quantified. However, chemicals and wastes from
several of the surface finish alternatives posed similar environmental compliance
problems as the HASL baseline. Two alternatives were subject to greater overall
federal environmental regulations than the baseline, suggesting that implementing
those alternatives could potentially increase compliance costs. It is easier to assess
the relative cost of complying with OSHA requirements, because several of the
alternatives pose reduced occupational safety hazards (non-automated, non-
conveyorized equipment may also pose less overall process hazards than working
with mechanized equipment).
Company
Image
Many businesses are finding that using cleaner technologies results in less tangible
benefits, such as an improved company image and improved community relations.
The elimination of lead from consumer products has been a key feature in many
company marketing plans. While it is difficult to put a monetary value on these
benefits, they should be considered in the decision-making process.
7-29
-------
7.3 TECHNOLOGY SUMMARY PROFILES
This section of the CTSA presents summary profiles of each of the surface finishing
technologies. The profiles summarize key information from various sections of the CTSA,
including the following:
• generic process steps, typical bath sequences, and equipment configurations evaluated in
the CTSA;
• human health and environmental hazards data and risk concerns for non-proprietary
chemicals;
• production costs and resource (water and energy) consumption data;
• Federal environmental regulations affecting chemicals in each of the technologies; and
• conclusions of the social benefits/costs assessment.
The summary profiles in this section present data for the HASL, nickel/gold,
nickel/palladium/gold, OSP, immersion silver, and the immersion tin technologies, respectively.
Data are presented for both the non-conveyorized and the conveyorized equipment
configurations, when applicable.
As discussed in Section 7.2, each of the alternatives appear to provide benefits in at least
one or more areas over the non-conveyorized HASL (the baseline process). However, the overall
benefits or costs associated with the alternatives could not be quantified without a more thorough
assessment of the factors involved. The actual decision of whether or not to implement an
alternative occurs outside of the CTSA process. Individual decision-makers may consider a
number of additional factors, such as their individual business circumstances and community
characteristics, together with the information presented in this CTSA.
7.3.1 HASL Technology
Generic Process Steps and Typical Bath Sequence
|-> Mfcr-tch 1^ W*irMic»2 L>. Biy 1^. FllS L
U. ™- I
r^l •"• I
Equipment Configurations Evaluated: Non-conveyorized (the baseline process) and
conveyorized.
7-30
-------
Risk Screening and Comparison
Table 7-17 summarizes human and environmental hazards and risk concerns for
chemicals in the HASL technology. The risk characterization identified occupational inhalation
risk concerns for one chemical in the non-conveyorized HASL process and dermal risk concerns
for two chemicals for either equipment configuration. No public health risk concerns were
identified for the pathways evaluated.
Table 7-17. Summary of Human Health and Environmental Risk Concerns for the
HASL Technology
Chemical
1 ,4-Butenediol
Alkylalkyne diol
Alkylaryl sulfonate
Alkylphenol ethoxylate
Alkylphenol
polyethoxyethanol
Arylphenol
Citric acid
Copper sulfate
pentahydrate
Ethoxylated alkylphenol
Ethylene glycol
Ethyleneglycol monobutyl
ether
Fluoboric acid
Gum
Hydrochloric acid
Hydrogen peroxide
Hydroxyaryl acid
Hydroxyaryl sulfonate
Lead
Phosphoric acid
Human Health Hazard and
Occupational Risk a
Inhalation
Risk
Concerns b
NE
NA
NE
NA
NA
NE
NA
NA
NA
Yes
No
NA
NA
No
No
NA
NA
No
No
Dermal
Risk
Concerns °
NE
Noe
Noe
Noe
Noe
No
Noe
Yes
Noe
No
No
NE
Noe
NE
No
Noe
Noe
Yesf
No
SAT
Rankd
LM
L
L
LM
LM
M
L
LM
M
LM
Carcinogenicity
Weight-of-
Evidence
Classification
None
None
None
None
None
None
None
Not classifiable
(EPA Class D)
None
None
None
None
None
Not classifiable
(IARC Group 3)
Not classifiable
(IARC Group 3)
None
None
Probable or
possible human
carcinogen
(EPA Class B2;
IARC Group 2 B)
None
Aquatic
Risk Concerns
NC: Yes
C:No
No
Yes
No
No
No
No
Not considered
No
No
No
No
No
No
Yes
No
No
No water releases
expected
No
7-31
-------
Chemical
Potassium
peroxymonosulfate
Sodium benzene sulfonate
Sodium hydroxide
Sulfuric acid
Tin
Summary
Human Health Hazard and
Occupational Risk a
Inhalation
Risk
Concerns b
NA
NA
NA
NA
NA
No or NA: 20
NE:3
Yes: 1
Dermal
Risk
Concerns °
Noe
Noe
NE
NEs
NE
No: 16
NE:6
YES: 2
SAT
Rankd
M
M
Carcinogenicity
Weight-of-
Evidence
Classification
None
None
None
Human carcinogen
(IARC Group 1)
None
2 suspected or
known
Aquatic
Risk Concerns
Yes
No
No
No
No water releases
expected
No: 19
Yes: 4
Not considered: 1
Risk concerns are for surface finishing line operators (the most exposed individual).
b Inhalation risk concerns for non-conveyorized process only. Inhalation risk from fully enclosed, conveyorized
process is assumed to be negligible.
c Dermal risk concerns apply to both conveyorized and non-conveyorized equipment.
d Structure-Activity Team rank for human health concerns:
L: Low concern; LM: Low-Moderate concern; M: Moderate concern.
e Chemical has very low skin absorption (based on EPA's Structure-Activity Team evaluation); risk from dermal
exposure not expected to be of concern.
f Lead evaluated by modeling potential blood-lead levels from incidental ingestion.
g Although chronic toxicity values have not been established, repeated skin contact with low concentrations of
sulfuric acid causes skin desiccation, ulceration of the hands, and chronic inflammation around the nails.
NE: Not Evaluated; due to lack of toxicity measure.
NA: Not Applicable. Inhalation exposure was not calculated because the chemical is not volatile (vapor pressure
below 1 x 10"3 torr) and is not used in any air-sparged bath.
Performance
The performance of the HASL technology was demonstrated at four test facilities, one of
which operated conveyorized HASL equipment. Performance test results were not differentiated
by the type of equipment configuration used. The Performance Demonstration determined that
each of the alternative technologies has the capability of achieving comparable levels of
performance to the HASL finish.
Production Costs and Resource Consumption
Computer simulation was used to model key operating parameters, including the time
required to process a job consisting of 260,000 ssf and the amount of resources (water and
energy) consumed. This information was analyzed with a hybrid cost model of traditional cost
(i.e., capital costs, etc.) and activity-based costs to determine average manufacturing costs per ssf
and water and energy consumption per ssf.
7-32
-------
Average manufacturing costs for the baseline process (the non-conveyorized HASL
process) were $0.36/ssf, while water and energy consumption were 1.24 gal/ssf and 218 Btu/ssf,
respectively. However, the conveyorized HASL process consumed less water and energy and
was more cost-effective than the baseline process (non-conveyorized HASL). Figure 7-1 lists the
results of the production cost and resource consumption analyses for the conveyorized HASL
process and illustrates the percent changes in costs and resource consumption from the baseline.
Manufacturing costs, water consumption, and energy consumption are less than the baseline by
three percent, 20 percent, and 39 percent, respectively.
-20%
-40%
-60%
($0.35/ssf)
(0.99 gal/ssf)
(133 Btu/ssf)
HASL-- Conveyorized
I Production Costs DWater Consumption D Energy Consumption
Figure 7-1. Production Costs and Resource Consumption of Conveyorized
HASL Technology
(Percent Change from Baseline with Actual Values in Parentheses )
Regulatory Concerns
Chemicals contained in the HASL technology are regulated by the Clean Water Act
(CWA), the Clean Air Act (CAA), the Emergency Planning and Community Right-to-Know Act
(EPCRA), the Superfund Amendments and Reauthorization Act (SARA), and the Toxic
Substances Control Act (TSCA). A summary of the number of HASL chemicals subject to
applicable federal regulations is presented in Table 7-18.
7-33
-------
Table 7-18. Number of HASL Chemicals Subject to Applicable Federal Regulations
Regulation
CWA
CAA
304b
307a
311
Priority Pollutant
111
112b
112r
No. of Chemicals
1
1
4
1
3
3
1
Regulation
EPCRA
SARA
TSCA
RCRA
313
302a
110
8d HSDR
MTL
8a PAIR
U
No. of Chemicals
6
3
1
3
4
3
--
Abbreviations and Definitions:
CWA 304b - Effluent Limitations Guidelines
CWA 307a - Toxic Pollutants
CWA 311 - Hazardous Substances
CAA 111 - Standards of Performance for New Stationary Sources of Air Pollutants-Equipment Leaks Chemical List
CAA 112b - Hazardous Air Pollutant
CAA 112r - Risk Management Program
EPCRA 313 - Toxic Chemical Release Inventory
EPCRA 302a - Extremely Hazardous Substances
SARA 110 - Superfund Site Priority Contaminant
TSCA 8d HSDR - Health & Safety Data Reporting Rules
TSCA MTL - Master Testing List
TSCA 8a PAIR - Preliminary Assessment Information Rule
RCRA U Waste - Characteristic hazardous waste
Social Benefits and Costs
Social cost is the total cost that an activity imposes on society (i.e., the sum of private and
external costs) while social benefit is the total benefit of an activity that society receives (i.e., the
sum of the private benefits and the external benefits). A qualitative assessment of the social
benefits and costs of the baseline and alternative technologies was performed to determine if
there would be net benefits or costs to society if PWB manufacturers switched to alternative
technologies from the baseline. (Net cost or benefit could not be completely assessed without a
more thorough assessment of effects on jobs and wages.)
In comparing the baseline (non-conveyorized HASL) to conveyorized HASL, there
appears to be a net benefit for switching to conveyorized HASL because — for the aspects
included in the evaluation — results are similar to or better than the baseline. Specifically,
changing from baseline to conveyorized HASL may result in:
• benefits from decreased worker and ecological risk (based on fewer chemicals of
concern), decreased water use, and decreased energy use; and
• no discernible cost or benefit for manufacturing cost and risk to the public.
7-34
-------
7.3.2 Nickel/Gold Technology
Generic Process Steps and Typical Bath Sequence
}-
-**
1
H —* H
H "—H-|
Equipment Configurations Evaluated: Conveyorized.
Risk Screening and Comparison
Table 7-19 summarizes human and environmental hazards and risk concerns for
chemicals in the nickel/gold technology. The risk characterization identified occupational
inhalation risk concerns for five chemicals and dermal risk concerns for six chemicals in the non-
conveyorized nickel/gold process. No public health risk concerns were identified for the
pathways evaluated, although cancer risks as high as one in 50 billion were estimated for the non-
conveyorized nickel/gold process.
Table 7-19. Summary of Human Health and Environmental Risk Concerns for the
Nickel/Gold Technology
Chemical
Aliphatic acid A
Aliphatic acid B
Aliphatic acid E
Aliphatic dicarboxylic acid
A
Aliphatic dicarboxylic acid
Alkylamino acid B
Alkyldiol
Alkylphenol
wlyethoxyethanol
Ammonia compound B
Ammonium chloride
Human Health Hazard and
Occupational Risks a
Inhalation
Risk
Concerns b
NE
NE
NE
NE
NE
NA
Yes
NA
NE
NA
Dermal
Risk
Concerns °
No
Noe
NE
Noe
No
NE
No
Noe
Noe
Yes
SAT
Rankd
M
LM
LM
MH
Carcinogenicity
Weight-of-Evidence
Classification
None
None
None
None
None
None
None
None
None
None
Aquatic Risk
Concerns
No
No
No
No
No
No
No
No
No
No
7-35
-------
Chemical
Ammonium hydroxide
Citric acid
Copper sulfate pentahydrate
Ethoxylated alkylphenol
Hydrochloric acid
Hydrogen peroxide
Hydroxyaryl acid
Inorganic metallic salt A
Inorganic metallic salt B
Inorganic metallic salt C
Malic acid
Nickel sulfate
Palladium chloride
Phosphoric acid
Potassium compound
Potassium gold cyanide
Potassium
peroxymonosulfate
Sodium hydroxide
Sodium hypophosphite
Sodium salt
Substituted amine
hydrochloride
Sulfuric acid
Transition metal salt
Urea compound B
Human Health Hazard and
Occupational Risks a
Inhalation
Risk
Concerns b
No
NA
NA
NA
Yes
Yes
NA
No
No
No
NE
Yes
NA
Yes
NE
NA
NA
NA
NE
NA
NA
NA
NA
NE
Dermal
Risk
Concerns °
Yes
Noe
Yes
Noe
NE
Yes
Noe
No
Yes
No
Noe
Yes
NE
No
NE
No
Noe
NE
Noe
No
Noe
NE8
Noe
NE
SAT
Rankd
L
LM
M
M
L
M
LM
M
M
Carcinogenicity
Weight-of-Evidence
Classification
None
None
Not classifiable
(EPA Class D)
None
Not classifiable
(IARC Group 3)
Not classifiable
(IARC Group 3)
None
Human carcinogen
or probable human
carcinogen f
Probable or possible
human carcinogen f
Probable or possible
human carcinogen f
None
None
None
None
None
None
None
None
None
None
None
Human carcinogen
(IARC Group 1)
None
Possible human
carcinogen f
Aquatic Risk
Concerns
No
No
Not considered
No
No
No
No
Not considered
Not considered
Not considered
No
Not considered
Not considered
No
No
Not considered
No
No
No
No
No
No
Not considered
No
7-36
-------
Chemical
Summary
Human Health Hazard and
Occupational Risks a
Inhalation
Risk
Concerns b
No or NA:
19
NE: 10
Yes: 5
Dermal
Risk
Concerns °
No: 20
NE:8
Yes: 6
SAT
Rankd
Carcinogenicity
Weight-of-Evidence
Classification
5 suspected or known
Aquatic Risk
Concerns
No: 26
Yes: 0
Not considered:
8
a Risk concerns are for surface finishing line operators (the most exposed individual).
b Inhalation risk concerns for non-conveyorized process only. Inhalation risk from fully enclosed, conveyorized
process is assumed to be negligible.
c Dermal risk concerns apply to both conveyorized and non-conveyorized equipment.
d Structure-Activity Team rank for human health concerns:
L: Low concern; LM: Low-Moderate concern; M: Moderate concern; MH: Moderate-High concern.
e Chemical has very low skin absorption (based on EPA's Structure-Activity Team evaluation); risk from dermal
exposure is not expected to be of concern.
f Specific EPA and/or IARC groups not reported in order to protect proprietary chemical identities.
g Although chronic toxicity values have not been established, repeated skin contact with low concentrations of
sulfuric acid causes skin desiccation, ulceration of the hands, and chronic inflammation around the nails.
NE: Not Evaluated; due to lack of toxicity measure.
Performance
The performance of the nickel/gold technology was demonstrated at three test facilities.
The Performance Demonstration determined that this technology has the capability of achieving
comparable levels of performance to the HASL finish. In addition, the nickel/gold process is
both gold and aluminum wire-bondable, though testing of wire-bondability was not included in
the performance testing protocol.
Production Costs and Resource Consumption
Computer simulation was used to model key operating parameters, including the time
required to process a job consisting of 260,000 ssf and the amount of resources (water and
energy) consumed. This information was analyzed with a hybrid cost model of traditional cost
(i.e., capital costs, etc.) and activity-based costs to determine average manufacturing costs per ssf
and water and energy consumption per ssf.
Analyses results determined that the non-conveyorized nickel/gold technology consumed
more water and energy and was less cost-effective than the baseline non-conveyorized HASL.
Average production costs for nickel/gold were $0.60/ssf, while water and energy consumption
rates were determined to be 2.06 gal/ssf and 447 Btu/ssf, respectively. Figure 7-2 lists the results
of these analyses and illustrates the percent changes in costs and resources consumption from the
baseline. Manufacturing costs, water consumption, and energy consumption are more than the
baseline by 67 percent, 66 percent, and 105 percent, respectively.
7-37
-------
Nickel/Gold—Non-Conveyorized
I Production Costs H Water Consumption D Energy Consumption
Figure 7-2. Production Costs and Resource Consumption of the Nickel/Gold Technology
(Percent Change from Baseline with Actual Values in Parentheses)
Regulatory Concerns
Chemicals contained in the nickel/gold technology are regulated by the CWA, CAA,
EPCRA, SARA, and TSCA. None of the nickel/gold process chemicals were regulated under
RCRA. A summary of the number of nickel/gold chemicals subject to applicable federal
regulations is presented in Table 7-20.
Social Benefits and Costs
A qualitative assessment of the private and external benefits and costs of the this
technology suggests a mixture of benefits and costs to society if PWB manufacturers switched to
the nickel/gold technology from the baseline. (Net social cost or benefit could not be
determined.) For the aspects included in the evaluation, changing from baseline to nickel/gold
may result in:
• costs from increased manufacturing cost, increased worker risk (based on fewer chemicals
of concern), increased water and energy use;
• benefits from decreased ecological risk (based on fewer chemicals of concern); and
• no discernible cost or benefit for risk to the public.
7-38
-------
Table 7-20. Number of Nickel/Gold Chemicals Subject to Applicable Federal Regulations
Regulation
CWA
CAA
304b
307a
311
Priority Pollutant
111
112b
112r
No. of
Chemicals
6
6
16
6
11
6
1
Regulation
EPCRA
SARA
TSCA
RCRA
313
302a
110
8d HSDR
MTL
8a PAIR
U
No. of
Chemicals
12
3
7
1
4
3
--
Abbreviations and Definitions:
CWA 304b - Effluent Limitations Guidelines
CWA 307a - Toxic Pollutants
CWA 311 - Hazardous Substances
CAA 111 - Standards of Performance for New Stationary Sources of Air Pollutants-Equipment Leaks Chemical List
CAA 112b - Hazardous Air Pollutant
CAA 112r - Risk Management Program
EPCRA 313 - Toxic Chemical Release Inventory
EPCRA 302a - Extremely Hazardous Substances
SARA 110 - Superfund Site Priority Contaminant
TSCA 8d HSDR - Health & Safety Data Reporting Rules
TSCA MTL - Master Testing List
TSCA 8a PAIR - Preliminary Assessment Information Rule
RCRA U Waste - Characteristic hazardous waste
7.3.3 Nickel/Palladium/Gold Technology
Generic Process Steps and Typical Bath Sequence
2|—>
\+\ w-—4
Crtrijct
•Vim
b
Equipment Configurations Evaluated: Non-conveyorized.
7-39
-------
Risk Screening and Comparison
Table 7-21 summarizes human and environmental hazards and risk concerns for
chemicals in the nickel/palladium/gold technology. The risk characterization identified
occupational inhalation risk concerns for six chemicals and dermal risk concerns for six
chemicals in the non-conveyorized nickel/palladium/gold process. No public health risk concerns
were identified for the pathways evaluated.
Table 7-21. Summary of Human Health and Environmental Risk Concerns for the
Nickel/Palladium/Gold Technology
Chemical
Aliphatic acid B
Aliphatic acid E
Aliphatic dicarboxylic acid A
Aliphatic dicarboxylic acid C
Alkylamino acid B
Alkyldiol
Alkyl polyol
Amino acid salt
Amino carboxylic acid
Ammonia compound A
Ammonia compound B
Ammonium hydroxide
Citric acid
Copper sulfate pentahydrate
Ethoxylated alkylphenol
Ethylenediamine
Hydrochloric acid
Hydrogen peroxide
Hydroxyaryl acid
Inorganic metallic salt B
Maleic acid
Malic acid
Nickel sulfate
Human Health Hazard and
Occupational Risks a
Inhalation
Risk
Concerns b
NE
NE
NE
NE
NA
Yes
NA
NA
NA
NA
NE
No
NA
NA
NA
No
Yes
Yes
NA
No
NA
NE
Yes
Dermal
Risk
Concerns °
NE
No
NE
No
No
No
No
NE
No
Yes
NE
Yes
Noe
Yes
Noe
No
NE
Yes
Noe
Yes
Noe
Noe
Yes
SAT
Rankd
M
LM
LM
MH
L
LM
M
M
LM
Carcinogenicity
Weight-of-
Evidence
Classification
None
None
None
None
None
None
None
None
None
None
None
None
None
Not classifiable
(EPA Class D)
None
None
Not classifiable
(IARC Group 3)
Not classifiable
(IARC Group 3)
None
Probable or
possible human
carcinogen f
None
None
None
Aquatic Risk
Concerns
No
No
No
No
No
No
No
No
No
No
No
No
No
Not considered
No
No
No
No
No
Not considered
No
No
Not considered
7-40
-------
Chemical
Palladium salt
Phosphoric acid
Potassium compound
Potassium gold cyanide
Propionic acid
Sodium hydroxide
Sodium hypophosphite
monohydrate
Sodium salt
Substituted amine
hydrochloride
Sulfuric acid
Surfactant
Transition metal salt
Urea compound B
Summary
Human Health Hazard and
Occupational Risks a
Inhalation
Risk
Concerns b
NA
Yes
NE
NA
Yes
NA
NE
NA
NA
NA
NA
NA
NE
No or NA: 21
NE:9
Yes: 6
Dermal
Risk
Concerns °
NE
No
NE
No
No
NE
Noe
No
Noe
NEs
NE
Noe
NE
No: 19
NE: 11
Yes: 6
SAT
Rankd
L
LM
M
M
Carcinogenicity
Weight-of-
Evidence
Classification
None
None
None
None
None
None
None
None
None
Human carcinogen
(IARC Group 1)
None
None
Possible human
carcinogen f
2 suspected or
known
Aquatic Risk
Concerns
Not considered
No
No
Not considered
No
No
No
No
No
No
NE
Not considered
No
No: 29
Yes: 0
Not considered: 6
Risk concerns are for surface finishing line operators (the most exposed individual).
b Inhalation risk concerns for non-conveyorized process only. Inhalation risk from fully enclosed, conveyorized
process is assumed to be negligible.
c Dermal risk concerns apply to both conveyorized and non-conveyorized equipment.
d Structure-Activity Team rank for human health concerns:
L: Low concern; LM: Low-Moderate concern; M: Moderate concern; MH: Moderate-High concern.
e Chemical has very low skin absorption (based on EPA's Structure-Activity Team evaluation); risk from dermal
exposure not expected to be of concern.
f Specific EPA and/or IARC groups not reported in order to protect proprietary chemical identities.
g Although chronic toxicity values have not been established, repeated skin contact with low concentrations of
sulfuric acid causes skin desiccation, ulceration of the hands, and chronic inflammation around the nails.
NE: Not Evaluated; due to lack of toxicity measure.
NA: Not Applicable. Inhalation exposure level was not calculated because the chemical is not volatile (vapor
pressure below 1 x 10"3 torr) and is not used in any air-sparged bath.
7-41
-------
Performance
The performance of the nickel/palladium/gold technology was demonstrated at one test
facility. The Performance Demonstration determined that this technology has the capability of
achieving comparable levels of performance to the HASL finish. In addition, the
nickel/palladium/gold process is both gold and aluminum wire-bondable, though testing of wire-
bondability was not included in the performance testing protocol.
Production Costs and Resource Consumption
Computer simulation was used to model key operating parameters, including the time
required to process a job consisting of 260,000 ssf and the amount of resources (water and
energy) consumed. This information was analyzed with a hybrid cost model of traditional cost
(i.e., capital costs, etc.) and activity-based costs to determine average manufacturing costs per ssf
and water and energy consumption per ssf.
The non-conveyorized nickel/palladium/gold technology consumed more water and
energy than the baseline process (non-conveyorized HASL). Average production costs for
nickel/palladium/gold were $1.54/ssf, while water and energy consumption rates were 3.61 gal/ssf
and 768 Btu/ssf, respectively. Figure 7-3 lists the results of these analyses and illustrates the
percent changes in resources consumption from the baseline. Manufacturing costs, water
consumption, and energy consumption are greater than the baseline by 327 percent, 191 percent,
and 252 percent, respectively.
Regulatory Concerns
Chemicals contained in the nickel/palladium/gold technology are regulated by the CWA,
CAA, EPCRA, SARA, and TSCA. None of the nickel/palladium/gold process chemicals were
regulated under RCRA. A summary of the number of nickel/palladium/gold chemicals subject to
applicable federal regulations is presented in Table 7-22.
Social Benefits and Costs
A qualitative assessment of the private and external benefits and costs of the this
technology suggests a mixture of benefits and costs to society if PWB manufacturers switched to
the nickel/palladium/gold technology from the baseline. (Net social cost or benefit could not be
determined.) For the aspects included in the evaluation, changing from baseline to
nickel/palladium/gold may result in:
• costs from increased manufacturing cost, increased worker risk (based on fewer chemicals
of concern), increased water and energy use;
• benefits from decreased ecological risk (based on fewer chemicals of concern); and
• no discernible cost or benefit for risk to the public.
7-42
-------
Figure 7-3. Production Costs and Resource Consumption of
Nickel/Palladium/Gold Technology
(Percent Change from Baseline with Actual Values in Parentheses)
Table 7-22. Number of Nickel/Palladium/Gold Chemicals Subject to Applicable
Federal Regulations
Regulation
CWA
CAA
304b
307a
311
Priority Pollutant
111
112b
112r
No. of
Chemicals
5
5
12
5
5
5
1
Regulation
EPCRA
SARA
TSCA
RCRA
313
302a
110
8d HSDR
MTL
8a PAIR
U
No. of
Chemicals
10
3
6
1
4
4
-
Abbreviations and Definitions:
CWA 304b - Effluent Limitations Guidelines
CWA 307a - Toxic Pollutants
CWA 311 - Hazardous Substances
CAA 111 - Standards of Performance for New Stationary Sources of Air Pollutants-Equipment Leaks Chemical List
CAA 112b - Hazardous Air Pollutant
CAA 112r - Risk Management Program
EPCRA 313 - Toxic Chemical Release Inventory
EPCRA 302a - Extremely Hazardous Substances
SARA 110 - Superfund Site Priority Contaminant
TSCA 8d HSDR - Health & Safety Data Reporting Rules
TSCA MTL - Master Testing List
TSCA 8a PAIR - Preliminary Assessment Information Rule
RCRA U Waste - Characteristic hazardous waste
7-43
-------
7.3.4 OSP Technology
Generic Process Steps and Typical Bath Sequence
w
n 1
•UT | >
1 ^
1 L
' \-> n
i
Equipment Configurations Evaluated: Non-conveyorized and conveyorized.
Risk Screening and Comparison
Table 7-23 summarizes human and environmental hazards and risk concerns for
chemicals in the OSP technology. The risk characterization identified occupational inhalation risk
concerns for one chemical in the non-conveyorized OSP process and dermal risk concerns for
three chemicals in the non-conveyorized OSP process and two chemicals in the conveyorized
OSP process. No public health risk concerns were identified for the pathways evaluated.
Performance
The performance of the OSP technology was demonstrated at three test facilities, one of
which operated conveyorized OSP equipment. Performance test results were not differentiated
by the type of equipment configuration used. The Performance Demonstration determined that
this technology has the capability of achieving comparable levels of performance to the HASL
finish.
Production Costs and Resource Consumption
Computer simulation was used to model key operating parameters, including the time
required to process a job consisting of 260,000 ssf and the amount of resources (water and
energy) consumed. This information was analyzed with a hybrid cost model of traditional cost
(i.e., capital costs, etc.) and activity-based costs to determine average manufacturing costs per ssf
and water and energy consumption per ssf.
Both the non-conveyorized and conveyorized OSP technologies consume less water and
energy and are more cost-effective than the baseline (non-conveyorized HASL process). Figure
7-4 lists the results of these analyses and illustrates the percent changes in costs and resource
consumption from the baseline. Manufacturing costs, water consumption, and energy
consumption for the non-conveyorized OSP process are less than the baseline by 69 percent, 38
percent, and 43 percent, respectively. The conveyorized OSP process is even more efficient than
its non-conveyorized counterpart, reducing manufacturing costs from that of the baseline by 72
percent, and reducing water and energy consumption by 57 percent and 67 percent, respectively.
7-44
-------
Table 7-23. Summary of Human Health and Environmental Risk Concerns for the
OSP Technology
Chemical
Acetic acid
Alkylaryl imidazole
Aromatic imidizole
product
Arylphenol
Copper ion
Copper salt C
Copper sulfate
pentahydrate
Ethoxylated alkylphenol
Ethylene glycol
Gum
Hydrochloric acid
Hydrogen peroxide
Hydroxyaryl acid
Hydroxy aryl sulfonate
Phosphoric acid
Sodium hydroxide
Sulfuric acid
Summary
Human Health Hazard and
Occupational Risks a
Inhalation Risk
Concerns b
NE
NA
NA
NE
NA
NA
NA
NA
Yes
NA
No
No
NA
NA
No
NA
NA
NoorNA: 14
NE:2
Yes: 1
Dermal
Risk
Concerns0
No
NE
NE
No
Yes
Yes e
Yes
Nof
No
Nof
NE
No
NE
Nof
No
NE
NEs
No: 8
NE:6
Yes: 3
SAT
Rankd
LM
M
LM
LM
Carcinogenicity
Weight-of-
Evidence
Classification
None
None
None
None
Not classifiable
(EPA Class D)
Not classifiable
(EPA Class D)
Not classifiable
(EPA Class D)
None
None
None
Not classifiable
(IARC Group 3)
Not classifiable
(IARC Group 3)
None
None
None
None
Human
carcinogen (IARC
Group 1)
1 suspected or
known
Aquatic Risk
Concerns
No
Yes
NE
No
Not considered
Not considered
Not considered
No
No
No
No
No
No
No
No
No
No
No: 12
Yes: 1
Not considered:
3
Risk concerns are for surface finishing line operators (the most exposed individual).
b Inhalation risk concerns for non-conveyorized process only. Inhalation risk from fully enclosed, conveyorized
process is assumed to be negligible.
c Dermal risk concerns apply to both conveyorized and non-conveyorized equipment unless otherwise noted.
d Structure-Activity Team rank for human health concerns:
LM: Low-Moderate concern; M: Moderate concern.
e Applied to non-conveyorized configuration only.
f Chemical has very low skin absorption (based on EPA's Structure-Activity Team evaluation); risk from dermal
exposure not expected to be of concern.
g Although chronic toxicity values have not been established, repeated skin contact with low concentrations of
sulfuric acid causes skin desiccation, ulceration of the hands, and chronic inflammation around the nails.
NA: Not Applicable. Inhalation exposure level was not calculated because the chemical is not volatile (vapor
pressure below 1 x 10"3 torr) and is not used in any air-sparged bath.
NE: Not Evaluated; due to lack of toxicity measure.
7-45
-------
.77
gal/ssf) (125
Btu/ssf)
($0.10/oof)
"ff
Figure 7-4. Production Costs and Resource Consumption of OSP Technology
(Percent Change from Baseline with Actual Values in Parentheses)
Regulatory Concerns
Chemicals contained in the OSP technology are regulated by the CWA, CAA, EPCRA,
SARA, and TSCA. None of the OSP process chemicals were regulated under RCRA. A
summary of the number of OSP chemicals subject to applicable federal regulations is presented
in Table 7-24.
Social Benefits and Costs
A qualitative assessment of the private and external benefits and costs of the this
technology suggests a mixture of benefits and costs to society if PWB manufacturers switched to
the OSP technology from the baseline. For the aspects included in the evaluation, changing from
baseline to OSP may result in:
• benefits from decreased manufacturing cost and ecological risk (based on fewer chemicals
of concern), decreased water and energy use;
• mixed results for worker risk (based on fewer carcinogens or suspected carcinogens used
in the process, but more chemicals of concern for non-cancer worker risk); and
• no discernible cost or benefit for risk to the public.
7-46
-------
Table 7-24. Number of OSP Chemicals Subject to Applicable Federal Regulations
Regulation
CWA
CAA
304b
307a
311
Priority Pollutant
111
112b
112r
No. of
Chemicals
2
2
5
2
3
2
1
Regulation
EPCRA
SARA
TSCA
RCRA
313
302a
110
8d HSDR
MTL
8a PAIR
U
No. of Chemicals
5
2
2
1
2
1
--
Abbreviations and Definitions:
CWA 304b - Effluent Limitations Guidelines
CWA 307a - Toxic Pollutants
CWA 311 - Hazardous Substances
CAA 111 - Standards of Performance for New Stationary Sources of Air Pollutants-Equipment Leaks Chemical List
CAA 112b - Hazardous Air Pollutant
CAA 112r - Risk Management Program
EPCRA 313 - Toxic Chemical Release Inventory
EPCRA 302a - Extremely Hazardous Substances
SARA 110 - Superfund Site Priority Contaminant
TSCA 8d HSDR - Health & Safety Data Reporting Rules
TSCA MTL - Master Testing List
TSCA 8a PAIR - Preliminary Assessment Information Rule
RCRA U Waste - Characteristic hazardous waste
7.3.5 Immersion Silver Technology
Generic Process Steps and Typical Bath Sequence
w
Clnir 1^
Y— .. — . — — - a^j. 1
WrtarMm 1 ^
HkrMteh 1— >. WrtarltaH 1 ^,
1
"""" n
Equipment Configurations Evaluated: Conveyorized.
Risk Screening and Comparison
Table 7-25 summarizes human and environmental hazards and risk concerns for
chemicals in the immersion silver technology. The risk characterization did not identify any
occupational or dermal risk concerns for chemicals in the conveyorized immersion silver process.
No public health risk concerns were identified for the pathways evaluated.
7-47
-------
Table 7-25. Summary of Human Health and Environmental Risk Concerns for the
Immersion Silver Technology
Chemical
1 ,4-Butenediol
Alkylamino acid A
Fatty amine
Hydrogen peroxide
Nitrogen acid
Phosphoric acid
Silver nitrate
Sodium hydroxide
Sulfuric acid
Summary
Human Health Hazard and
Occupational Risks a
Inhalation
Risk
Concerns b
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA: 9
Dermal
Risk
Concerns °
NE
Noe
Noe
No
NE
No
No
NE
NEf
No: 5
NE:4
SAT
Rankd
LM
LM
M
Carcinogenicity
Weight-of-
Evidence
Classification
None
None
None
Not classifiable
(IARC Group 3)
None
None
Not classifiable
(EPA Class D)
None
Human carcinogen
(IARC Group 1)
1 suspected or
known
Aquatic
Risk Concerns
No
No
No
Yes
No
No
Not considered
No
No
No: 7
Yes: 1
Not considered: 1
Risk evaluated for conveyorized process only. Inhalation risk from fully enclosed, conveyorized process is assumed
to be low. Risk concerns are for line operator (the most exposed individual).
b Inhalation risk concerns for non-conveyorized process only. Inhalation risk from fully enclosed, conveyorized
process is assumed to be negligible.
c Dermal risk concerns apply to both conveyorized and non-conveyorized equipment.
d Structure-Activity Team rank for human health concerns:
LM: Low-Moderate concern; M: Moderate concern.
e Chemical has very low skin absorption (based on EPA's Structure-Activity Team evaluation); risk from dermal
exposure is not expected to be of concern.
f Although chronic toxicity values have not been established, repeated skin contact with low concentrations of
sulfuric acid causes skin desiccation, ulceration of the hands, and chronic inflammation around the nails.
NE: Not Evaluated; due to lack of toxicity measure.
NA: Not Applicable. Inhalation exposure level was assumed to be negligible for conveyorized lines.
Performance
The performance of the immersion silver technology was demonstrated at two test
facilities. The Performance Demonstration determined that this technology has the capability of
achieving comparable levels of performance to the HASL finish. In addition, the immersion
silver process is both gold and aluminum wire-bondable, though testing of wire-bondability was
not included in the performance testing protocol.
7-48
-------
Production Costs and Resource Consumption
Computer simulation was used to model key operating parameters, including the time
required to process a job consisting of 260,000 ssf and the amount of resources (water and
energy) consumed. This information was analyzed with a hybrid cost model of traditional cost
(i.e., capital costs, etc.) and activity-based costs to determine average manufacturing costs per ssf
and water and energy consumption per ssf.
Analysis results showed that the conveyorized immersion silver process consumed less
water and was more cost-effective than the baseline non-conveyorized HASL process, while
consuming more energy. Average production costs for immersion silver were $0.28/ssf, while
water and energy consumption rates were determined to be 0.53 gal/ssf and 287 Btu/ssf,
respectively. Figure 7-5 lists the results of these analyses and illustrates the percent changes in
costs and resource consumption from the baseline. Manufacturing costs and water consumption
are less than the baseline by 22 percent and 57 percent, respectively, while energy consumption
increased by 32 percent.
Figure 7-5. Production Costs and Resource Consumption of Immersion Silver Technology
(Percent Change from Baseline with Actual Values in Parentheses)
7-49
-------
Regulatory Concerns
Chemicals contained in the immersion silver technology are regulated by the CWA, CAA,
EPCRA, SARA, and TSCA. None of the immersion silver process chemicals were regulated
under RCRA. A summary of the number of immersion silver chemicals subject to applicable
federal regulations is presented in Table 7-26.
Table 7-26. Number of Immersion Silver Chemicals Subject to Applicable
Federal Regulations
Regulation
CWA
CAA
304b
307a
311
Priority Pollutant
111
112b
112r
No. of
Chemicals
1
1
5
1
1
1
--
Regulation
EPCRA
SARA
TSCA
RCRA
313
302a
110
8d HSDR
MTL
8a PAIR
U
No. of
Chemicals
3
3
1
--
1
1
--
Abbreviations and Definitions:
CWA 304b - Effluent Limitations Guidelines
CWA 307a - Toxic Pollutants
CWA 311 - Hazardous Substances
CAA 111 - Standards of Performance for New Stationary Sources of Air Pollutants-Equipment Leaks Chemical List
CAA 112b - Hazardous Air Pollutant
CAA 112r - Risk Management Program
EPCRA 313 - Toxic Chemical Release Inventory
EPCRA 302a - Extremely Hazardous Substances
SARA 110 - Superfund Site Priority Contaminant
TSCA 8d HSDR - Health & Safety Data Reporting Rules
TSCA MTL - Master Testing List
TSCA 8a PAIR - Preliminary Assessment Information Rule
RCRA U Waste - Characteristic hazardous waste
Social Benefits and Costs
A qualitative assessment of the private and external benefits and costs of the this
technology suggests a mixture of benefits and costs to society if PWB manufacturers switched to
the immersion silver technology from the baseline. For the aspects included in the evaluation,
changing from baseline to immersion silver may result in:
• benefits from decreased manufacturing cost, worker and ecological risk (based on fewer
chemicals of concern), and decreased water use;
• costs from increased energy use; and
• no discernible cost or benefit for risk to the public.
7-50
-------
7.3.6 Immersion Tin Technology
Generic Process Steps and Typical Bath Sequence
[-> BMMMlmTte [_> Wd»»li.«a[->.
Equipment Configurations Evaluated: Non-conveyorized and conveyorized.
Risk Screening and Comparison
Table 7-27 summarizes human and environmental hazards and risk concerns for
chemicals in the immersion tin technology. The risk characterization identified occupational
dermal risk concerns for one chemical for either equipment configuration. No occupational
inhalation concerns or public health risk concerns were identified for the pathways evaluated.
Table 7-27. Summary of Human Health and Environmental Risk Concerns for the
Immersion Tin Technology
Chemical
Aliphatic acid D
Alkylalkyne diol
Alkylimine dialkanol
Alkylamino acid B
Alkylaryl sulfonate
Alkylphenol ethoxylate
Bismuth compound
Citric acid
Cyclic amide
Ethoxylated alkylphenol
Ethylene glycol monobutyl
ether
Fluoboric acid
Hydrochloric acid
Hydroxy carboxylic acid
Methane sulfonic acid
Phosphoric acid
Potassium peroxymonosulfate
Human Health Hazard and
Occupational Risks a
Inhalation
Risk
Concerns b
No
NA
NA
NA
NE
NA
NA
NA
No
NA
No
NA
No
No
NA
No
NA
Dermal
Risk
Concerns °
No
Noe
Noe
No
Noe
Noe
Nof
Noe
No
Noe
No
NE
NE
No
NE
No
Noe
SAT
Rankd
L
M
L
LM
M
L
LM
M
Carcinogenicity
Weight-of-
Evidence
Classification
None
None
None
None
None
None
None
None
None
None
None
None
Not classifiable
(IARC Group 3)
None
None
None
None
Aquatic
Risk Concerns
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
NC: Yes
C:No
7-51
-------
Chemical
Quantenary alkyl ammonium
chlorides
Silver salt
Sodium benzene sulfonate
Sodium phosphorus salt
Stannous methane sulfonic
acid
Sulfuric acid
Thiourea
Tin chloride
Unspecified tartrate
Urea
Vinyl polymer
Urea compound C
Summary
Human Health Hazard and
Occupational Risks a
Inhalation
Risk
Concerns b
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NE
No or NA: 27
NE:2
Yes: 0
Dermal
Risk
Concerns °
Noe
No
Noe
NE
No
No
NE
No
No
No
No
Yes
No: 23
NE:5
Yes: 1
SAT
Rankd
M
M
Carcinogenicity
Weight-of-
Evidence
Classification
None
Not classifiable g
None
None
Not classifiable
(EPA Class D)
Human
carcinogen
(IARC Group 1)
Possibly
carcinogenic
(IARC Group
2B)
Not classifiable
(EPA Class D;
IARC Group 3)
None
None
Not classifiable 8
None
2 suspected or
known
Aquatic
Risk Concerns
No
Not considered
No
No
Not considered
No
No
Not considered
No
No
No
No
No: 25
Yes: 1
Not considered:
3
Risk concerns are for surface finishing line operators (the most exposed individual).
b Inhalation risk concerns for non-conveyorized process only. Inhalation risk from a fully enclosed, conveyorized
process is assumed to be negligible.
c Dermal risk concerns apply to both conveyorized and non-conveyorized equipment.
d Structure-Activity Team rank for human health concerns:
L: Low concern; LM: Low-Moderate concern; M: Moderate concern.
e Chemical has very low skin absorption (based on EPA's Structure-Activity Team evaluation); risk from dermal
exposure is not expected to be of concern.
f No absorption expected through skin, however, in water this compound will cause irritation of all moist tissues
(SAT report).
g Specific EPA and/or IARC groups not reported in order to protect proprietary chemical identities.
NE: Not Evaluated; due to lack of toxicity measure.
NA: Not Applicable. Inhalation exposure level was not calculated because the chemical is not volatile (vapor
pressure below 1 x 10"3 torr) and is not used in any air-sparged bath.
7-52
-------
Performance
The performance of the immersion tin technology was demonstrated at four test facilities,
two of which operated conveyorized immersion tin equipment. Performance test results were not
differentiated by the type of equipment configuration used. The Performance Demonstration
determined that this technology has the capability of achieving comparable levels of performance
to the HASL finish.
Production Costs and Resource Consumption
Computer simulation was used to model key operating parameters, including the time
required to process a job consisting of 260,000 ssf and the amount of resources (water and
energy) consumed. This information was analyzed with a hybrid cost model of traditional cost
(i.e., capital costs, etc.) and activity-based costs to determine average manufacturing costs per ssf
and water and energy consumption per ssf.
Both the non-conveyorized and conveyorized methods of immersion tin were more
economical than the baseline process, with the non-conveyorized process proving less expensive
($0.18/ssf vs. $0.25/ssf) overall. Only the conveyorized immersion tin process showed a
reduction in water consumption, while both equipment configurations consumed more energy
than the baseline. Figure 7-6 lists the results of these analyses and illustrates the percent changes
in costs and resource consumption for either equipment configuration from the baseline. Non-
conveyorized immersion tin manufacturing costs are less than the baseline by 50 percent, while
the water and energy consumption rates increased by 46 percent and 33 percent, respectively.
Manufacturing costs and the water consumption for the conveyorized immersion tin process are
less than the baseline by 31 percent and 29 percent respectively, while energy consumption
increased 139 percent.
Regulatory Concerns
Chemicals contained in the immersion tin technology are regulated by the CWA, CAA,
EPCRA, SARA, and TSCA. In addition, two of the chemicals in the immersion tin process
chemicals is regulated under RCRA. A summary of the number of immersion tin chemicals
subject to applicable federal regulations is presented in Table 7-28.
7-53
-------
($0.25/ssf) (0.88gal/ssf)
Figure 7-6. Production Costs and Resource Consumption of Immersion Tin Technology
(Percent Change from Baseline with Actual Values in Parentheses)
Table 7-28. Number of Immersion Tin Chemicals Subject to Applicable
Federal Regulations
Regulation
CWA
CAA
304b
307a
311
Priority Pollutant
111
112b
112r
No. of
Chemicals
1
1
6
1
3
2
1
Regulation
EPCRA
SARA
TSCA
RCRA
313
302a
110
8d HSDR
MTL
8a PAIR
U
No. of
Chemicals
7
2
1
2
4
3
2
Abbreviations and Definitions:
CWA 304b - Effluent Limitations Guidelines
CWA 307a - Toxic Pollutants
CWA 311 - Hazardous Substances
CAA 111 - Standards of Performance for New Stationary Sources of Air Pollutants-Equipment Leaks Chemical List
CAA 112b - Hazardous Air Pollutant
CAA 112r - Risk Management Program
EPCRA 313 - Toxic Chemical Release Inventory
EPCRA 302a - Extremely Hazardous Substances
SARA 110 - Superfund Site Priority Contaminant
TSCA 8d HSDR - Health & Safety Data Reporting Rules
TSCA MTL - Master Testing List
TSCA 8a PAIR - Preliminary Assessment Information Rule
RCRA U Waste - Characteristic hazardous waste
7-54
-------
Social Benefits and Costs
A qualitative assessment of the private and external benefits and costs of the this
technology suggests a mixture of benefits and costs to society if PWB manufacturers switched to
the immersion tin technology from the baseline. For the aspects included in the evaluation,
changing from baseline to non-conveyorized immersion tin may result in:
• benefits from decreased manufacturing cost, worker and ecological risk (based on fewer
chemicals of concern);
• costs from increased water and energy use; and
• no discernible cost or benefit for risk to the public.
Changing from baseline to conveyorized immersion tin may result in:
• benefits from decreased manufacturing cost, worker and ecological risk (based on fewer
chemicals of concern) and decreased water use;
• costs from increased energy use; and
• no discernible cost or benefit for risk to the public.
7-55
-------
REFERENCES
Mishan, E.J. 1976. Cost-Benefit AnalysisPraeger Publishers: New York.
7-56
-------
Printed Wiring Board
Surface Finishes
Cleaner
Technologies
Substitutes
Assessment
VOLUME 2:
Appendices
Jack R. Geibig, Senior Research Associate
Mary B. Swanson, Research Scientist
and the
PWB Engineering Support Team
This document was produced by the University of Tennessee
Center for Clean Products and Clean Technologies under grant
#X825373 from EPA's Design for the Environment Branch,
Economics, Exposure, & Technology Division, Office of
Pollution Prevention and Toxics.
or
U.S.EPA
-------
Appendix A
Data Collection Sheets
-------
Contents
Workplace Practices Questionnaire A-l
Facility Background Information Sheet A-58
Observer Data Sheet A-67
Supplier Data Sheet A-74
-------
Workplace Practices Questionnaire
IPC
Design for the Environment
Printed Wiring Board Project
Workplace Practices Questionnaire
Please complete this questionnaire, make a copy for
your records, and send the original to:
Jack Geibig
UT Center for Clean Products
311 Conference Center Building
Knoxville TN 37996
Phone: (423) 974-6513
Fax: (423) 974-1838
FACILITY AND CONTACT INFORMATION
Facility Identification
Company Name:
Site Name:
Street Address:
City:
| State: |
Zip:
Contact Identification Enter the names of the persons who can be contacted regarding this survey.
Name:
Title:
Phone:
Fax:
E-Mail:
A-l
-------
—INSTRUCTION SHEET—
FOR THE DESIGN FOR THE ENVIRONMENT (DFE)
ALTERNATIVE SURFACE FINISHES (ASF) PROJECT
WORKPLACE PRACTICES QUESTIONNAIRE
INTRODUCTION
This questionnaire was prepared by the University of Tennessee Center for Clean Products and Clean Technologies
in partnership with The US EPA Design for the Environment (DfE) Printed Wiring Board (PWB) Program, IPC, and
other members of the DfE PWB Industry Project Work Groups.
The purpose of this questionnaire is to collect data that will be used in preparation of a Design for the Environment
(DfE) Alterative Surface Technologies report. This report will present an analysis and evaluation of the risk,
performance, and costs associated with operating each of the alternative surface finish processes. Much of this report
will be based on data submitted by PWB manufacturing facilities. You can obtain more information about this
project and other DfE PWB projects from the US EPA's website at http://www.epa.gov/opptintr/dfe/pwb/pwb.html).
CONFIDENTIALITY
All information and data that is entered into this questionnaire is confidential. The sources of responses are only
known to the IPC and have been coded by the IPC for industry research purposes. Any use or publication of the data
will not identify the names or locations of the respondent companies or the individuals completing the forms.
INSTRUCTIONS
Respondents must complete Sections 1 (Facility Characterization) and Section 2 (HASL Process) of
this questionnaire.
Section 3 is divided into five processes (3 A through 3E) as shown below:
3A. Organic Solder Preservative (OSP) Process
3B. Immersion Silver Process
3C. Immersion Tin Process
3D. Electroless Nickel/Immersion Gold Process
3E. Electroless Nickel/Electroless Palladium/Immersion Gold Process
Of these five subsections, 3A-3E, please fill out only the top two alternative processes, based on PWB
through-put, that are currently being implemented at your facility.
If your responses do not fit in the spaces provided, please photocopy the section to provide more space or use
ordinary paper and mark the response with the section number to which it applies.
Please make a copy of the completed sections and retain them for your records.
If you have questions regarding the survey, please contact Jack Geibig of the University of Tennessee Center for
Clean Products and Clean Technologies at (telephone 423/975-6513; fax 423/974-1838; emailjgeibig@utk.edu) or
Star Summerfield at IPC (telephone 847/790-5347; fax 847/509-9798; email summst@ipc.org).
Please return the completed questionnaire by January 8,1999 to:
Star Summerfield, IPC
2215 Sanders Road, Northbrook, IL 60062-6135
Phone: 847/790-5347, FAX 847/509-9798, email summst@ipc.org
A RETURN LABEL TO IPC IS ENCLOSED FOR YOUR CONVENIENCE.
A-2
-------
Section 1. Facility Characterization
This section focuses on general information specific to the facility. This information is not process-specific. Please
estimate manufacturing data for the previous 12 month period, or other convenient time period of 12 consecutive
months (e.g., FY97). Only consider the portion of the facility dedicated to PWB manufacturing when entering
employee and facility size data.
1.1
General Information
Size of portion of facility used for
manufacturing PWBs:
sq. ft.
Overall amount of PWB produced
in surface square feet (ssf):
ssf/yr
1.2 Process Type
Estimate the percentage of PWBs manufactured at your facility using the following methods for surface finishing
(SF). Specify "other" entry.
Surface Finish Process
HASL
OSP-Thick
OSP-Thm
(benzotriazole-based)
Immersion Tin
Electroless Palladium
Immersion Silver
Percent of Total
%
%
%
%
%
%
Surface Finish Process
Electroless Nickel/ Immersion Gold
Electroless Nickel/Electroless
Palladium/ Immersion Gold
Other:
Other:
Other:
Total
Percent of Total
%
%
%
%
%
100%
1.3 Wastewater Discharge and Sludge Data
Wastewater discharge method
(circle one):
Direct Indirect Zero
(to stream) (to POTW)
Throughput of facility wastewater treatment system:
Annual weight of sludge generated:
Is sludge dewatered prior to disposal (circle one)?
Water content prior to dewatering:
Water content after dewatering:
gals/day
Ibs
Yes No
%
%
A-3
-------
Section 2. HASL Process
2.1 Process Schematic: HASL
Fill in the figure below for your HASL surface finishing processses. Using the key at the bottom of the page, identify which letter corresponds w
the first step in your HASL process and write that letter in the first box (see example). Continue using the key to fill in boxes for each step intil y
entire HASL process is represented. If your particular process step is not represented by the key below, complete the figure by writing in the na
of the process step in your particular surface finishing line in the corresponding box(es). Finish by responding, to the questions at the bottom of
page.
Type of Process
Process
Step Letter
(see key below)
Chemical Supplier:
Process Line Installation Date:
i.
>,
2.
3.
>,
•
4.
>
5.
6.
>fc
•
7.
1
8.
>•
9.
>.
10.
11.
>,
12.
13.
>,
14.
>.
15.
Is the entire HASL process, as described in the chart above,
co-located in the same room: Yes No
* If no (process steps performed in more than one room), please
circle the steps above that are in a separate room.
Type of Process Automation for the HASL line: (circle one)
Conveyorized Automated non-conveyorized Manually-controlled hoist
Manual (no automation) Other (specify):
HASL Process Step Key
[A] - Cleaner
[B] - Microetch
[C] - Flux
[D] - Solder
[E] - Air Cool
[G] - Post-cleaner
[H] - Dryer
[I] - Water Rinse
[J] - Air Knife
[K] - Other
(Specify in the
appropriate box)
A-4
-------
2.2
General Data-HASL
Number of days HASL line is in
operation:
Estimated scrap rate (% of defective
product) for HASL process:
days/yr
%
Number of hours per day the HASL
line is in operation:
Total of P WB surface square feet
processed by HASL line per year:
hrs/day
ssf/yr
2.3 Process Area Employees—HASL
Complete the following table by indicating the number of employees of each type that perform work duties in the
same process room as the HASL line, and for what length of time. Consider only workers who have regularly
scheduled responsibilities that require them to be physically within the process room. Specify "other" entry. Enter
"N/A" in any category that is not applicable.
Type of HASL
Area Worker
Line Operators
Lab Technicians
Maintenance Workers
Wastewater Treatment Operators
Supervisory Personnel
Other (specify):
Number of Employees
in HASL Process Area
Average Hours per Week per
Employee in HASL Process Area
hrs
hrs
hrs
hrs
hrs
hrs
2.4 Physical Settings-HASL
Size of the room containing the HASL
process:
Are the overall process areas/rooms
ventilated (circle one)?
Do you have local vents (circle one)?
sq. ft.
Yes No
Yes No
Height of room:
Air flow rate:
Local vent air flow rate:
ft.
cu. ft./min.
cu. ft./min.
Overall surface finishing process line dimensions
Length (ft.): Width (ft.): Height (ft.):
2.5
Rack Dimensions—HASL
Average number of panels per rack:
Average size of panel in rack:
Length (in.):
Average space between panels in rack:
in.
Width (in.):
Do you purposely slow the withdraw rate of your panels from process baths
to reduce drag-out? (Circle one)
Yes No
A-5
-------
2.6 Rinse Bath Water Usage-HASL
Consult the process schematic in section 2.1 to obtain the process step numbers associated with each of the water rinse
baths present in your HASL process. Enter, in the table below, the process step number along with the flow control
method and flow rate data requested for each water rinse bath. If the water rinse bath is part of a cascade, you need
only report the daily water flow rate of one bath in the cascade.
| Total volume of water used by the HASL line when operating:
gal./day |
Process Step
Number a
Example: 8
Flow Control b
R
Daily Water
Flow Rate c
2,400 gal./day
gal. /day
gal./day
gal./day
gal./day
gal./day
gal./day
a Process step number - Consult the process schematic in question 2.1 and
enter the process step number of the specific water rinse tank.
b Flow control - Consult key at right and enter the letter for the flow control
method used for that specific rinse bath.
c Daily water flow rate - Enter the average daily flow rate for the specific
water rinse tank.
d Cascade water process steps - Use the step numbers for rinses that are
cascaded together.
Cascade Water
Process Steps d
8-»6
Flow Control Methods Key
[C] - Conductivity Meter
[P] - pH Meter
[V] - Operator Control Valve
[R] - Flow Restricter
[N] - None (continuous flow)
[O] - Other (explain)
2.7 Filter Replacement-HASL
Not Applicable
n
Bath(s) filtered
(enter process step # from flow diagram in 2. 1)
Frequency of replacement:
Duration of replacement process:
Personal protective equipment (see key):
Personal Protective Equipment Key:
[E] - Eye Protection [G] - Gloves [Z] - All except Respiratory Protection
[L] - Lab coat/Sleeved garment [A] - Apron [N] - None
[R] - Respiratory Protection [B] - Boots
A-6
-------
2.8 Rack or Conveyor Cleaning—HASL
Not Applicable Q
Rack Cleaning Method (see key): OR
Conveyor Cleaning Method (see key):
Frequency of rack or conveyor cleaning:
Number of personnel involved:
Personal protective equipment (see key):
Average time required to clean:
min.
Rack Cleaning Method:
[C]-Chemical bath on SF process line
[D]-Chemical bath on another line
[T]-Temporary chemical bath
[S]-Manual scrubbing with chemical
[M]-Non-chemical cleaning
[N]-None
[O]-Continuous cleaning
Conveyor Cleaning Method:
[C]-Chemical rinsing or soaking
[S]-Manual scrubbing with chemical
[M]-Non-chemical cleaning
[N]-None
[O]-Continuous cleaning
Personal Protective Equipment:
[E]-Eye Protection [G]-Gloves
[L]-Lab coat/Sleeved garment [A]-Apron
[R]-Respiratory Protection [B]-Boots
[O]-Continuous Cleaning [N]-None
2.9 Solder Unit Maintenance and Waste disposal Pl-^1 excePl Respiratory Protection
Complete the following maintenance and waste disposal questions for only the unit of the process that
performs the hot air solder leveling
Frequency of maintenance:
Duration of maintenance :
Personal protective equipment
(see key):
Number of personnel involved:
min.
d Personal Protective Equipment - Enter the letters of all
the protective equipment used by the workers who physically
replace the spent bath.
[E] - Eye protection [B] - Boots
[A] - Apron [G] - Gloves
[L] - Lab coat/Sleeved garment
[R] - Respiratory protection
[Z] - All except Respiratory Protection
[N] - None
Method of dross removal:
Frequency of dross removal:
Quantity of solder waste disposed
(per day):
Method of solder waste disposal
(see key):
Method Of Solder Waste Disposal - Indicate method of
solder waste disposal from key below:
[M] - Metals reclaimed off-site
[R] - Recycled on-site
[RO] - Recycled off-site
[D] - Drummed and treated as hazardous waste
[O] - Other (specify)
A-7
-------
2.10 Physical Data and Operating Conditions—HASL
Complete the tables below by entering the data requested for each specific type of chemical bath listed. If two tanks of the same type are used within the process,
list the data for each tank separately.
Average cycle time for a panel to complete entire HASL
process
(includes cleaning and post cleaning steps, if any):
mm.
Bath
Cleaner
Microetch
Flux
Solder
Post-Clean
Other (specify)
Physical Data
Length
(inches)
in.
in.
in.
in.
in.
Width
(inches)
in.
in.
in.
in.
in.
Nominal
Volume
(gal)
gal.
gal.
gal.
gal.
gal.
Process Data
Immersion
Time3
(seconds)
sec.
sec.
sec.
sec.
sec.
a Immersion Time - Enter the average elapsed time a rack of panels is immersed in
the specific process bath.
b Drip Time - Enter the average elapsed time that a rack of panels is allowed to hang
above the specific process bath to allow drainage from panels.
DripJTime
(seconds)
sec.
sec.
sec.
sec.
sec.
Operating Conditions
Temp
(°F)
°F
°F
°F
°F
°F
Agitation Methods Key:
[PA]- Panel agitation
[CP]- Circulation pump
[AS]- Air sparge
[O]- Other (explain)
Agitation
(see key)
Vapor Control
(see key)
Vapor Control Methods Key:
[BC]- Bath cover
[FE]- Fully enclosed
[VO]- Vent to outside
[VC]- Vent to control
[PP]- Push pull
[O]- Other (explain)
2.11 Initial Chemical Bath Make-Up Composition-HASL
A-8
-------
Complete the chart below for each chemical component of the bath type listed. Provide the manufacturer name if the chemical used is known only by trade name. If more
room is needed, please attach another sheet with the additional information. If two tanks of the same type are used within the process, list the data for a single tank only.
Bath
Cleaner
Microetch
Flux
Solder
Post-Clean
Other (specify)
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
Chemical Product Name
Manufacturer
(if applicable)
Annual Quantity Used"
(gallons)
a Annual Quantity Used - If the amount of a particular chemical used is measured by weight (i.e., crystalline chemicals) instead of volume, enter the weight in
pounds and clearly specify the units (Ibs).
A-9
-------
2.12 Chemical Bath Bailout and Additions-HASL
Complete the following chart detailing the typical bath bailout and chemical additions that are made to maintain the chemical balance of each specific process
bath. If more than three chemicals are added to a specific bath, attach another sheet with the additional information. If chemical additions to a bath are made
automatically, do not complete the last three columns for that bath. If two tanks of the same type are used within the process, list the data for a single tank only.
Bath Type
Cleaner
Microetch
Flux
Solder
Post-Clean
Other
(specify)
Bailout
Frequency
Bailout
Duration c
(minutes)
min.
min.
min.
min.
min.
min.
Bailout
Quantity
Personal
Protective
Equipment d
Chemical Products Added
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
Criteria for
Addition"
Method of
Chemical
Additioij to Tank
Duration of
Addition0
(minutes )
min.
min.
min.
min.
min.
min.
* Criteria for Additions - Enter the b Method of Chemical Addition to Tank - Enter the letter for d Personal Protective Equipment - Enter the letters of all the
letter for the criteria typically used to the method typically used to add chemicals to the tanks. protective equipment used by the workers who physically replace the
determine when bath additions are [PR] - Poured spent bath.
necessary. [P] - Pumped manually [O] - Other E] - Eye protection
[S] - Statistical process control „, , A - Apron
P] - Panel square feet processed c Duration of Bailowp or'A^fififron - Enter the elapsed time from L - Lab coat/Sleeved Sgroegoots
C - Chemical testing the retrieval of the chemical stock through the completion of the R - Respiratory protection
T - Time addition of all chemicals. For bailout, enter the time required to Z] - All except Respiratory Protection
[O - Other bailout the bath prior to making additions. [N - None
FG1 - Gloves
A-10
-------
2.13 Chemical Bath Replacement - HASL
Complete the chart below by providing information on the process of replacing, treating, and disposing of a spent chemical bath.
Bath Type
Cleaner
Microetch
Flux
Solder
Post-Clean
Other (specify)
Criteria
for
Replacement3
Replacement
Frequencyb
a Criteria for Replacement -
[S] - Statistical process control
P] - Panel square feet processed
[C] - Chemical testing
T] - Time
[O] - Other (specify)
b Frequency - Enter the average amount of time
elapsed, or number of square feet processed, between
bath replacements. Clearly specify units (e.g., hours,
sq.ft.).
Method of Spent
Bath Removal c
Tank
Cleaning Method11
c Methods of Spent Bath Removal-
[P] - Pump spent bath from tank
S] - Siphon spent bath from tank
[D] - Drain spent bath from tank
O] - Other (specify)
d Tank Cleaning Method -
[C] - Chemical flush
W] - Water rinse
[H] - Hand scrub
[O] - Other (specify)
6 Duration of Replacement - Enter the
elapsed time from the beginning of bath
removal until the replacement bath is
finished.
Duration of
Replacement
Procedure6
min.
min.
min.
min.
min.
min.
Personal
Protective
Equipment'
F Personal Protective Equip. - Enter the letters
of all the protective equipment used by the
workers who physically replace the spent bath.
E - Eye protection
[G - Gloves
L] - Lab coat/sleeved garment
A - Apron
[R] - Respiratory protection
[B] - Boots
[Z] - All except respiratory protection
[N] - None
A-ll
-------
2.14 Chemical Bath Sampling—HASL
Bath Type
Example:
Cleaner
Microetch
Other (specify):
Type of
Sampling a
A
Frequency b
3 per day
Duration of
Sampling c
5 min
Protective
Equipment d
E, G, A
Method of
Sampling "
P
' Type of Sampling c Duration of Sampling: Enter the e Method of Sampling:
[A] - Automated average time required to manually take a [D] - Drain or spigot
[M] - Manual sample from the tank. [P] - Pipette
[N] - None [L] - Ladle
11 Protective Equipment: Consult the [O] - Other (specify)
b Frequency: Enter the average time key for the above table and enter the
elapsed or number of panel sq. ft. letters for all protective equipment used
processed between samples. Clearly by the person performing the chemical
specify units (e.g., hours, sq.ft.) sampling.
2.15 Process Waste Disposal - HASL
Bath Type
Cleaner
Microetch
Flux
Solder
Post-Clean
Other
(specify):
Annual Volume
Treated or Disposed a
a Annual Volume Treated
or Disposed - Enter the
yearly amount of the specific
bath treated or disposed. Be
sure to consider the volume
treated from both bath
change outs and bailout
before entering the total
Method of
Treatment or
Disposal b
RCRA Waste
Code (if
applicable)
B Methods of Treatment or Disposal -
P] - Precipitation pretreatment on-site
N] - pH neutralization pretreatment on-site
S] - Disposed directly to sewer with no
treatment
[D] - Drummed for off-site treatment or
disposal
[RN] - Recycled on-site
[RF] - Recycled off-site
[O] - Other (specify)
Container
Type
Container Type -
Indicate the type of
container used for
disposal of bath wastes
OH]- Open-head drum
CH]- Closed-head drum
T]- Chemical tote
O]- Other (specify)
A-12
-------
Section 3. Electroless Nickel/Immersion Gold Process
3.1 Process Schematic: Nickel/Gold
Fill in the figure below for your electroless nickel/immersion gold surface finishing processses. Using the key at the bottom of the page, iedntify
which letter corresponds with the first step in your process and write that letter in the first box (see example). Continue using the key to fill in bo
for each step until your entire electroless nickel/immersion gold process is represented. If a particular process step is not represented by the key
below, complete the figure by writing in the name of the process step in your particular surface finishing line in the corresponding box(es). Finis
by responding to the questions at the bottom of the page.
Type of Process
Process ____^^
Step Letter
(see key below)
Ex.
-^fc.
-A
Chemical Supplier:
Process Line Installation Date:
i.
>.
2.
3.
>„
4.
>.
•
5.
6.
>_
*
7.
>„
'
8.
>-
9.
>.
10.
11.
>„
*
12.
^_
^
13.
>_
••
14.
>„
*
15.
Is the entire electroless nickel/immersion gold, as described in the
chart above, co-located in the same room:
Yes No
* If no (process steps performed in more than one room), please
circle the steps above that are in a separate room.
Type of Process Automation for the nickel/gold line: (circle one)
Conveyorized Automated non-conveyorized Manually-controlled hoist
Manual (no automation) Other (specify):
Electroless Nickel/Immersion Gold Process
Step Key
[A] - Conditioner/Cleaner
[B] - Microetch
[C] - Catalyst
[D] - Acid Dip
[E] - Activator
[F] - Predip
[G] - Electroless Nickel
[H] - Immersion Gold
[I] - Water Rinse
[J] - Other (Specify in the appropriate box)
A-13
-------
3.2
General Data-Nickel/Gold
Number of days the nickel/gold line
is in operation:
Estimated scrap rate (% of defective
product) for the nickel/gold process:
days/yr
%
Number of hours per day the nickel/gold
line is in operation:
Total of P WB surface square feet
processed by the nickel/gold line per year:
hrs/day
ssf/yr
3.3 Process Area Employees—Nickel/Gold
Complete the following table by indicating the number of employees of each type that perform work duties in the
same process room as the nickel/gold line, and for what length of time. Consider only workers who have regularly
scheduled responsibilities that require them to be physically within the process room. Specify "other" entry. Enter
"N/A" in any category that is not applicable.
Type of Surface Finish
Area Worker
Line Operators
Lab Technicians
Maintenance Workers
Wastewater Treatment Operators
Supervisory Personnel
Other (specify):
Number of Employees
in Surface Finish Process Area
Average Hours per Week per
Employee in Surface Finish
Process Area
hrs
hrs
hrs
hrs
hrs
hrs
3.4 Physical Settings-Nickel/Gold
Size of the room containing the
surface finish process:
Are the overall process areas/rooms
ventilated (circle one)?
Do you have local vents (circle one)?
sq. ft.
Yes No
Yes No
Height of room:
Air flow rate:
Local vent air flow rate:
ft.
cu. ft./min.
cu. ft./min.
Overall surface finishing process line dimensions
Length (ft.): Width (ft.): Height (ft.):
3.5
Rack Dimensions—Nickel/Gold
Average number of panels per rack:
Average size of panel in rack:
Length (in.):
Average space between panels in rack:
in.
Width (in.):
Do you purposely slow the withdraw rate of your panels from process baths
to reduce drag-out? (Circle one)
Yes No
A-14
-------
3.6 Rinse Bath Water Usage-Nickel/Gold
Consult the process schematic in section 2.1 to obtain the process step numbers associated with each of the water
rinse baths present in your nickel/gold process. Enter, in the table below, the process step number along with the
flow control method and flow rate data requested for each water rinse bath. If the water rinse bath is part of a
cascade, you need only report the daily water flow rate of one bath in the cascade.
Total volume of water used by the surface finish line when operating:
gal./day |
Process Step
Number a
Example: 8
Flow Control b
R
Daily Water
Flow Rate c
2,400 gal./day
gal./day
gal./day
gal./day
gal./day
gal./day
gal./day
a Process step number - Consult the process schematic in question 2.1 and
enter the process step number of the specific water rinse tank.
b Flow control - Consult key at right and enter the letter for the flow control
method used for that specific rinse bath.
c Daily water flow rate - Enter the average daily flow rate for the specific
water rinse tank.
d Cascade water process steps - Use the step numbers for rinses that are
cascaded together.
Cascade Water
Process Steps d
8->6
Flow Control Methods Key
[C - Conductivity Meter
\P] - pH Meter
V] - Operator Control Valve
R - Flow Restricter
N - None (continuous flow)
[O - Other (explain)
3.7 Filter Replacement-Nickel/Gold
Not Applicable
D
Bath(s) filtered
(enter process step # from flow diagram in 2.1)
Frequency of replacement:
Duration of replacement process:
Personal protective equipment (see key):
Personal Protective Equipment Key:
[E] - Eye Protection [G] - Gloves [Z] - All except Respiratory Protection
[L] - Lab coat/Sleeved garment [A] - Apron [N] - None
[R] - Respiratory Protection [B] - Boots
A-15
-------
3.8 Rack or Conveyor Cleaning—Nickel/Gold
Not Applicable |
Rack Cleaning Method (see key): OR
Conveyor Cleaning Method (see key):
Frequency of rack or conveyor cleaning:
Number of personnel involved:
Personal protective equipment (see key):
Average time required to clean:
min.
Rack Cleaning Method:
[C]-Chemical bath on SF process line
[D]-Chemical bath on another line
[T]-Temporary chemical bath
[S]-Manual scrubbing with chemical
[M]-Non-chemical cleaning
N]-None
O]-Continuous cleaning
Conveyor Cleaning Method:
[C]-Chemical rinsing or soaking
[S]-Manual scrubbing with chemical
[M]-Non-chemical cleaning
;N]-None
O]-Continuous cleaning
Personal Protective Equipment:
[E]-Eye Protection [G]-Gloves
[L]-Lab coat/Sleeved garment [A]-Apron
[R]-Respiratory Protection [B]-Boots
[O]-Continuous Cleaning [N]-None
[ZJ-A11 except Respiratory Protection
3.9 Chemical Bath Sampling —Nickel/Gold
Bath Type
Example:
Cleaner/
Conditioner
Microetch
Catalyst
Acid Dip
Acivator
Electroless
Nickel
Immersion Gold
Other (specify):
Type of
Sampling a
A
- Type of Sampling
[A] - Automated
[M] - Manual
[N] - None
b Frequency: Ent
time elapsed or nun
ft. processed betws
Clearly specify unit
ft).
sr the average
aber of panel sq.
;en samples.
s (e.g., hours, sq.
Frequency b
3 per day
Duration of
Sampling c
5 min
- Duration of Sampling: Enter the a
verage time required to manually take
a sample from the tank.
- Protective Equipment: Consult
the key for the above table and enter
the letters for all protective
equipment used by the person
performing the chemical sampling.
Protective
Equipment d
E,G,A
Method of
Sampling e
P
- Method of Sampling:
[D] - Drain or spigot
[P] - Pipette
{L] - Ladle
[O] - Other (specify)
A-16
-------
3.10 Physical Data and Operating Conditions—Nickel/Gold
Complete the tables below by entering me data requested for each specific type of chemical bath listed. If two tanks of the same type are used within the process,
list the data for each tank separately.
(Average cycle time for a panel to complete entire nickel/gold process
(includes cleaning and post cleaning steps, if any):
min.
Bath
Cleaner/
Conditioner
Microetch
Catalyst
Acid Dip
Activator
Electroless Nickel
Immersion Gold
Other (specify);
Physical Data
Length
(inches)
in.
in.
in.
in.
in.
in.
in.
in.
Width
(inches)
in.
in.
in.
in.
in.
in.
in.
in.
Nominal
Volume
(gal)
gal.
gal.
gal.
gal.
gal.
gal.
gal.
gal.
Process Data
Immersion
Time"
(seconds)
sec.
sec.
sec.
sec.
sec.
sec.
sec.
sec.
a Immersion Time - Enter the average elapsed time a rack of panels is immersed in
the specific process bath.
b Drip Time - Enter the average elapsed time that a rack of panels is allowed to hang
above the specific process bath to allow drainage from panels.
Drip bTime
(seconds)
sec.
sec.
sec.
sec.
sec.
sec.
sec.
sec.
Operating Conditions
Temp
(°F)
°F
°F
°F
°F
°F
°F
°F
°F
Agitation Methods Kev:
[PA] - Panel agitation
CP] - Circulation pump
AS] - Air sparge
[O] - Other (explain)
Agitation
(see key)
Vapor Control
(see key)
Vapor Control Methods Kev:
[BC] - Bath cover
[FE] - Fully enclosed
[VO] - Vent to outside
JVC] - Vent to control
PP] - Push pull
[O] - Other (explain)
A-17
-------
3. 1 1 Initial Chemical Bath
Complete the chart below for each
oom is needed nlease attach anot
Bath
Cleaner
Microetch
Catalyst
Acid Dip
Activator
Electroless Nickel
Immersion Gold
Other (specify)
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
i Make-Up Composition —Nickel/Gold
chemical component of the bath type listec
ler sheet with the additional information
Chemical Product Name
. Provide the manufacturer name if the chemical used is known only by trade name. If more
f two tanks of the same tvne are used within thejirocess list the data for a single tank onlv
Manufacturer (if applicable)
Annual Quantity Used a (gallons)
a Annual Quantity Used -
pounds and clearly specify
If the amount of a particular chemical used is measured by weight (i.e., crystalline chemicals) instead of volume, enter the weight in
the units (Ibs).
A-18
-------
3.12 Chemical Bath Bailout and Additions-Nickel/Gold
Complete the following chart detailing the typical bath bailout and chemical additions that are made to maintain the chemical balance of each specific process
bath. If more than three chemicals are added to a specific bath, attach another sheet with the additional information. If chemical additions to a bath are made
automatically, do not complete the last three columns for that bath. If two tanks of the same type are used within the process, list the data for a single tank only.
Bath Type
Cleaner/
Conditioner
Microetch
Catalyst
Acid Dip
Activator
Electroless
Nickel
Immersion
Gold
Other
(specify)
Bailout
Frequency
Bailout
Duration c
(minutes)
min.
min.
min.
min.
min.
min.
min.
min.
Bailout
Quantity
Personal
Protective
Eauiument d
Chemical Products Added
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
* Criteria for Additions - Enter the b Method of Chemical Addition to Tank - Enter the letter for
letter for the criteria typically used to the method typically used to add chemicals to the tanks.
determine when bath additions are [PR] - Poured
necessary. [P] - Pumped manually [O] - Other
[S] - Statistical process control
P] - Panel square feet processed c Duration of Bailowp or'A^fiflfron - Enter the elapsed time
C] - Chemical testing from the retrieval of the chemical stock through the completion
T] - Time of the addition of all chemicals. For bailout, enter the time
[O] - Other required to bailout the bath prior to making additions.
Criteria for
Addition"
Method of
Chemical
Addition to Tank"
Duration of
Addition0
(minutes )
min.
min.
min.
min.
min.
min.
min.
min.
d Personal Protective Equipment - Enter the letters of all the
protective equipment used by the workers who physically replace the
spent bath.
E] - Eye protection
A - Apron
L - Lab coat/Sleeved &JffQe]goots
R - Respiratory protection
Z] - All except Respiratory Protection
N - None
TGI - Gloves
A-19
-------
3.13 Chemical Bath Replacement — Nickel/Gold
Complete the chart below by providing information on the process of replacing, treating, and disposing of a spent chemical bath.
Bath Type
Cleaner/Conditioner
Microetch
Catalyst
Acid Dip
Activator
Electroles Nickel
Immersion Gold
Other (specify)
Criteria
for Replacement a
Replacement
Frequency b
a Criteria for Replacement -
[S] - Statistical process control
P] - Panel square feet processed
[C] - Chemical testing
T] - Time
[O] - Other (specify)
b Frequency - Enter the average amount of time
elapsed, or number of square feet processed, between
bath replacements. Clearly specify units (e.g., hours,
sq.ft.).
Method of Spent
Bath Removal c
Tank
Cleaning Method d
c Methods of Spent Bath Removal-
[P] - Pump spent bath from tank
S] - Siphon spent bath from tank
[D] - Drain spent bath from tank
O] - Other (specify)
d Tank Cleaning Method -
[C] - Chemical flush
W] - Water rinse
[H] - Hand scrub
[O] - Other (specify)
e Duration of Replacement - Enter the
elapsed time from the beginning of bath
removal until the replacement bath is
finished.
Duration of
Replacement
Procedure e
min.
min.
min.
min.
min.
min.
min.
min.
Personal
Protective
Equipment f
f Personal Protective Equip. - Enter the letters
of all the protective equipment used by the
workers who physically replace the spent bath.
E - Eye protection
[G -Gloves
L] - Lab coat/sleeved garment
A - Apron
[R] - Respiratory protection
[B] - Boots
[Z] - All except respiratory protection
[N] - None
A-20
-------
3.14 Process Waste Disposal — Nickel/Gold
Bath Type
Cleaner/Conditi
oner
Microetch
Catalyst
Acid Dip
Activator
Electroless
Nickel
Immersion Gold
Other (specify):
Annual Volume
Treated or Disposed a
a Annual Volume Treated or
Disposed - Enter the yearly
amount of the specific bath
treated or disposed. Be sure to
consider the volume treated
from both bath change outs
and bailout before entering the
total
Method of Treatment
or Disposal b
RCRA Waste
Code (if applicable)
Container
Type
b Methods of Treatment or Disposal - Container Type -
P] - Precipitation pretreatment on-site Indicate the type of
N] - pH neutralization pretreatment on-site container used for disposal
S] - Disposed directly to sewer with no treatment of bath wastes
D] - Drummed for off-site treatment or disposal [OH]- Open-head drum
RN] - Recycled on-site [CH]- Closed-head drum
RF] - Recycled off-site [T]- Chemical tote
O] - Other (specify) [O]- Other (specify)
A-21
-------
Section 4. Electroless Nickel/Electroless Palladium/Immersion Gold Process
4.1 Process Schematic: Nickel/Palladium/Gold
Fill in the figure below for your electroless nickel/ electroless palladium/immersion gold surface finishing processses. Using the key at the bottoi
of the page, identify which letter corresponds with the first step in your process and write that letter in the first box (see example). Continue usin
the key to fill in boxes for each step until your entire nickel/palladium/gold process is represented. If a particular process step is not represented
the key below, complete the figure by writing in the name of the process step in your particular surface finishing line in the corresponding boxeO
Finish by responding to the questions at the bottom of the page.
Type of Process
Process
Step Letter
(see key below)
Chemical Supplier:
Process Line Installation Date:
1.
>.
2.
•^
^
3.
>„
*
4.
*^
^
5.
6.
>„
'
7.
1
8.
>»
9.
•^
^
10.
11.
>.
12.
>,
•
13.
>
14.
>„
15.
Is the entire nickel/palladium/gold, as described in the chart above,
co-located in the same room:
Yes No
* If no (process steps performed in more than one room), please
circle the steps above that are in a separate room.
Type of Process Automation for the nickel/palladium/gold line: (circle one)
Conveyorized Automated non-conveyorized Manually-controlled hoist
Manual (no automation) Other (specify):
Nickel/Palladium/Gold Process Step Key
[A] - Conditioner/Cleaner
[B] - Microetch
[C] - Catalyst
[D] - Acid Dip
[E] - Activator
[F] - Electroless Nicke.
[G] - Electroless Palladium
[H] - Immersion Gold
[I] - Water Rinse
[J] - Other (Specify in the appropriate box)
A-22
-------
4.2
General Data-Nickel/Palladium/Gold
Number of days the nickel/palladium/gold
line is in operation:
Estimated scrap rate (% of defective
product) for the nickel/palladium/gold
process:
days/y
r
%
Number of hours per day the
nickel/palladium/gold line is in operation:
Total of PWB surface square feet
processed by the nickel/palladium/gold line
per year:
hrs/day
ssf/yr
4.3 Process Area Employees—Nickel/Palladium/Gold
Complete the following table by indicating the number of employees of each type that perform work duties in the
same process room as the nickel/palladium/gold line, and for what length of time. Consider only workers who have
regularly scheduled responsibilities that require them to be physically within the process room. Specify "other" entry.
Enter "N/A" in any category that is not applicable.
Type of Surface Finish
Area Worker
Line Operators
Lab Technicians
Maintenance Workers
Wastewater Treatment Operators
Supervisory Personnel
Other (specify):
Number of Employees
in Surface Finish Process Area
Average Hours per Week per
Employee in Surface Finish
Process Area
hrs
hrs
hrs
hrs
hrs
hrs
4.4 Physical Settings-Nickel/Palladium/Gold
Size of the room containing the
surface finish process:
Are the overall process areas/rooms
ventilated (circle one)?
Do you have local vents (circle one)?
sq. ft.
Yes No
Yes No
Height of room:
Air flow rate:
Local vent air flow rate:
ft.
cu. ft./min.
cu. ft./min.
Overall surface finishing process line dimensions
Length (ft.): Width (ft.): Height (ft.):
4.5 Rack Dimensions—Nickel/Palladium/Gold
Average number of panels per rack:
Average size of panel in rack:
Length (in.):
Average space between panels in rack:
in.
Width (in.):
Do you purposely slow the withdraw rate of your panels from process baths
to reduce drag-out? (Circle one)
Yes No
A-23
-------
4.6 Rinse Bath Water Usage-Nickel/Palladium/Gold
Consult the process schematic in section 2.1 to obtain the process step numbers associated with each of the water
rinse baths present in your nickel/palladium/gold process. Enter, in the table below, the process step number along
with the flow control method and flow rate data requested for each water rinse bath. If the water rinse bath is part
of a cascade, you need only report the daily water flow rate of one bath in the cascade.
Total volume of water used by the surface finish line when operating:
gal./day |
Process Step
Number a
Example: 8
Flow Control b
R
Daily Water
Flow Rate c
2,400 gal./day
gal./day
gal./day
gal./day
gal./day
gal./day
gal./day
a Process step number - Consult the process schematic in question 2.1 and
enter the process step number of the specific water rinse tank.
b Flow control - Consult key at right and enter the letter for the flow control
method used for that specific rinse bath.
c Daily water flow rate - Enter the average daily flow rate for the specific
water rinse tank.
d Cascade water process steps - Use the step numbers for rinses that are
cascaded together.
Cascade Water
Process Steps d
8-»6
Flow Control Methods Key
[C] - Conductivity Meter
[P] - pH Meter
V] - Operator Control Valve
R - Flow Restricter
TST - None (continuous flow)
[O] - Other (explain)
4.7 Filter Replacement-Nickel/Palladium/Gold
Not Applicable 1 1
Bath(s) filtered
(enter process step # from flow diagram in 2. 1)
Frequency of replacement:
Duration of replacement process:
Personal protective equipment (see key):
Personal Protective Equipment Key:
[E] - Eye Protection [G] - Gloves
[L] - Lab coat/Sleeved garment [A] - Apron
[R] - Respiratory Protection [B] - Boots
[Z] - All except Respiratory Protection
[N] - None
A-24
-------
4.8 Rack or Conveyor Cleaning—Nickel/Palladium/Gold
Not Applicable Q
Rack Cleaning Method (see key): OR
Conveyor Cleaning Method (see key):
Frequency of rack or conveyor cleaning:
Number of personnel involved:
Personal protective equipment (see key):
Average time required to clean:
min.
Rack Cleaning Method:
[C]-Chemical bath on SF process line
[D]-Chemical bath on another line
[T]-Temporary chemical bath
[S]-Manual scrubbing with chemical
[M]-Non-chemical cleaning
[N]-None
[O]-Continuous cleaning
Conveyor Cleaning Method:
[C]-Chemical rinsing or soaking
[S]-Manual scrubbing with chemical
[M]-Non-chemical cleaning
;N]-None
O]-Continuous cleaning
Personal Protective Equipment:
[E]-Eye Protection [G]-Gloves
[L]-Lab coat/Sleeved garment [A]-Apron
[R]-Respiratory Protection [B]-Boots
[O]-Continuous Cleaning [N]-None
[ZJ-A11 except Respiratory Protection
4.9 Chemical Bath Sampling -Nickel/Palladium/Gold
Bath Type
Example:
Cleaner/
Conditioner
Microetch
Catalyst
Acid Dip
Acivator
Electroless
Nickel
Electroless
Palladium
Immersion Gold
Other (specify):
Type of
Sampling a
A
- Type of Sampling
[A] - Automated
[M] - Manual
[N] - None
b Frequency: Ent
time elapsed or nun
ft. processed betwj
Clearly specify unit
ft).
sr the average
aber of panel sq.
;en samples.
s (e.g., hours, sq.
Frequency b
3 per day
Duration of
Sampling c
5 min
- Duration of Sampling: Enter the a
verage time required to manually take
a sample from the tank.
- Protective Equipment: Consult
the key for the above table and enter
the letters for all protective
equipment used by the person
performing the chemical sampling.
Protective
Equipment d
E,G,A
Method of
Sampling e
P
- Method of Sampling:
[D] - Dram or spigot
[P] - Pipette
{L] - Ladle
[O] - Other (specify)
A-25
-------
4.10 Physical Data and Operating Conditions—Nickel/Palladium/Gold
Complete the tables below by entering me data requested for each specific type of chemical bath listed. If two tanks of the same type are used within the process,
list the data for each tank separately.
(Average cycle time for a panel to complete entire surface finish process
(includes cleaning and post cleaning steps, if any):
min.
Bath
Cleaner/Conditioner
Microetch
Catalyst
Acid Dip
Activator
Electroless Nickel
Electroless Palladium
Immersion Gold
Other (specify);
Physical Data
Length
(inches)
in.
in.
in.
in.
in.
in.
in.
in.
in.
Width
(inches)
in.
in.
in.
in.
in.
in.
in.
in.
in.
Nominal
Volume
(gal)
gal.
gal.
gal.
gal.
gal.
gal.
gal.
gal.
gal.
Process Data
Immersion
Time"
(seconds)
sec.
sec.
sec.
sec.
sec.
sec.
sec.
sec.
sec.
a Immersion Time - Enter the average elapsed time a rack of panels is immersed in
the specific process bath.
b Drip Time - Enter the average elapsed time that a rack of panels is allowed to hang
above the specific process bath to allow drainage from panels.
DripbTime
(seconds)
sec.
sec.
sec.
sec.
sec.
sec.
sec.
sec.
sec.
Operating Conditions
Temp
(°F)
°F
°F
°F
°F
°F
°F
°F
°F
°F
Agitation Methods Kev:
[PA] - Panel agitation
CP] - Circulation pump
AS] - Air sparge
[O] - Other (explain)
Agitation
(see key)
Vapor Control
(see key)
Vapor Control Methods Kev:
[BC] - Bath cover
FE] - Fully enclosed
[VO] - Vent to outside
JVC] - Vent to control
PP] - Push pull
[O] - Other (explain)
A-26
-------
4. 1 1 Initial Chemical Bath
Complete the chart below for each
oom is needed nlease attach anot
Bath
Cleaner
Microetch
Catalyst
Acid Dip
Activator
Electroless Nickel
Elect roless
Palladium
Immersion Gold
Other (specify)
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
i Make-Up Composition —Nickel/Palla
chemical component of the bath type listec
ler sheet with the additional information
Chemical Product Name
dium/Gold
. Provide the manufacturer name if the chemical used is known only by trade name. If more
f two tanks of the same tvne are used within the, process list the data for a single tank onlv
Manufacturer (if annlicahle)
Annual Ouantitv Used a (gallons')
a Annual Quantity Used -
pounds and clearly specify
If the amount of a particular chemical used is measured by weight (i.e., crystalline chemicals) instead of volume, enter the weight in
the units (Ibs).
A-27
-------
4.12 Chemical Bath Bailout and Additions-Nickel/Palladium/Gold
Complete the following chart detailing the typical bath bailout and chemical additions that are made to maintain the chemical balance of each specific process
bath. If more than three chemicals are added to a specific bath, attach another sheet with the additional information. If chemical additions to a bath are made
automatically, do not complete the last three columns for that bath. If two tanks of the same type are used within the process, list the data for a single tank only.
Bath Type
Cleaner/
Conditioner
Microetch
Catalyst
Acid Dip
Activator
Electroless
Nickel
Electroless
Palladium
Immersion
Gold
Other
(specify)
Bailout
Frequency
Bailout
Duration c
(minutes)
min.
min.
min.
min.
min.
min.
min.
min.
min.
Bailout
Quantity
Personal
Protective
Eauiument d
Chemical Products Added
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
Criteria for
Addition"
Method of
Chemical
Addition to Tank"
Duration of
Addition0
(minutes )
min.
min.
min.
min.
min.
min.
min.
min.
* Criteria for Additions - Enter the b Method of Chemical Addition to Tank - Enter the letter for d Personal Protective Equipment - Enter the letters of all the
letter for the criteria typically used to the method typically used to add chemicals to the tanks. protective equipment used by the workers who physically replace the
determine when bath additions are [PR] - Poured spent bath.
necessary. [P] - Pumped manually [O] - Other E] - Eye protection
[S] - Statistical process control A - Apron
P] - Panel square feet processed c Duration of Bailowp or'A^fiflfron - Enter the elapsed time from L - Lab coat/Sleeved &JffQe]goots
C] - Chemical testing the retrieval of the chemical stock through the completion of the R - Respiratory protection
T] - Time addition of all chemicals. For bailout, enter the time required to Z] - All except Respiratory Protection
[O] - Other bailout the bath prior to making additions. [N - None
TGI - Gloves
A-28
-------
4.13 Chemical Bath Replacement — Nickel/Palladium/Gold
Complete the chart below by providing information on the process of replacing, treating, and disposing of a spent chemical bath.
Bath Type
Cleaner/Conditioner
Microetch
Catalyst
Acid Dip
Activator
Electroless Nickel
Electroless Palladium
Immersion Gold
Other (specify)
Criteria
for Replacement a
Replacement
Frequency b
a Criteria for Replacement -
[S] - Statistical process control
P] - Panel square feet processed
[C] - Chemical testing
T] - Time
[O] - Other (specify)
b Frequency - Enter the average amount of time
elapsed, or number of square feet processed, between
bath replacements. Clearly specify units (e.g., hours,
sq.ft.).
Method of Spent
Bath Removal c
Tank
Cleaning Method d
c Methods of Spent Bath Removal-
[P] - Pump spent bath from tank
S] - Siphon spent bath from tank
[D] - Drain spent bath from tank
O] - Other (specify)
d Tank Cleaning Method -
[C] - Chemical flush
W] - Water rinse
[H] - Hand scrub
[O] - Other (specify)
e Duration of Replacement - Enter the
elapsed time from the beginning of bath
removal until the replacement bath is
finished.
Duration of
Replacement
Procedure e
min.
min.
min.
min.
min.
min.
min.
min.
min.
Personal
Protective
Equipment f
f Personal Protective Equip. - Enter the letters
of all the protective equipment used by the
workers who physically replace the spent bath.
E - Eye protection
[G -Gloves
L] - Lab coat/sleeved garment
A - Apron
R - Respiratory protection
[B] - Boots
[Z] - All except respiratory protection
A-29
-------
4.14 Process Waste Disposal — Nickel/Palladium/Gold
Bath Type
Cleaner/
Conditioner
Microetch
Catalyst
Acid Dip
Activator
Electroless
Nickel
Electroless
Palladium
Immersion Gold
Other (specify):
Annual Volume
Treated or Disposed a
a Annual Volume Treated or
Disposed - Enter the yearly
amount of the specific bath
treated or disposed. Be sure to
consider the volume treated
from both bath change outs
and bailout before entering the
total
Method of Treatment
or Disposal b
RCRA Waste
Code (if applicable)
Container
Type
b Methods of Treatment or Disposal - Container Type -
P] - Precipitation pretreatment on-site Indicate the type of
N] - pH neutralization pretreatment on-site container used for disposal
S] - Disposed directly to sewer with no treatment of bath wastes
D] - Drummed for off-site treatment or disposal [OH]- Open-head drum
RN] - Recycled on-site [CH]- Closed-head drum
RF] - Recycled off-site [T]- Chemical tote
O] - Other (specify) [O]- Other (specify)
A-30
-------
Section 5. Organic Solder Preservative (OSP) Process
5.1 Process Schematic: OSP
Fill in the figure below for your OSP surface finishing process. Using the key at the bottom of the page, identify which letter corresponds with tl
first step in your process and write that letter in the first box (see example). Continue using the key to fill in boxes for each step until your entir
process is represented. If a particular step is not represented by the key below, complete the figure by writing in the name of the process step in
your particular surface finishing line in the corresponding box(es). Finish by responding to the questions at the bottom of the page.
Type of Process
Process -____^
Step Letter ~~ — —
(see key below)
Ex.
^^A
Chemical Supplier:
Process Line Installation Date:
1.
>.
2.
3.
•^
^
4.
>.
•
5.
6.
*^
^
7.
8.
>.
9.
>„
*
10.
11.
>»
12.
13.
>.
*
14.
>.
15.
Is the entire OSP process, as described in the chart above, co-locatec I
in the same room:
Yes No
* If no (process steps performed in more than one room), please
circle the steps above that are in a separate room.
Type of Process Automation for the OSP line: (circle one)
Conveyorized Automated non-conveyorized Manually-controlled hoist
Manual (no automation) Other (specify):
OSP Process Step Key
[A] - Cleaner
[B] - Microetch
[C] - Predip
[D] - OSP
[E] - Water Rinse
[F] - Air Knife
[G] - Other (Specify in the appropriate box
A-31
-------
5.2
General Data-OSP
Number of days the OSP line is in
operation:
Estimated scrap rate (% of defective
product) for the OSP process:
days/yr
%
Number of hours per day the OSP line is in
operation:
Total of P WB surface square feet
processed by the OSP line per year:
hrs/day
ssf/yr
5.3 Process Area Employees—OSP
Complete the following table by indicating the number of employees of each type that perform work duties in the
same process room as the OSP line, and for what length of time. Consider only workers who have regularly
scheduled responsibilities that require them to be physically within the process room. Specify "other" entry. Enter
"N/A" in any category that is not applicable.
Type of Surface Finish
Area Worker
Line Operators
Lab Technicians
Maintenance Workers
Wastewater Treatment Operators
Supervisory Personnel
Other (specify):
Number of Employees
in Surface Finish Process Area
Average Hours per Week per
Employee in Surface Finish
Process Area
hrs
hrs
hrs
hrs
hrs
hrs
5.4 Physical Settings-OSP
Size of the room containing the
surface finish process:
Are the overall process areas/rooms
ventilated (circle one)?
Do you have local vents (circle one)?
sq. ft.
Yes No
Yes No
Height of room:
Air flow rate:
Local vent air flow rate:
ft.
cu. ft./min.
cu. ft./min.
Overall surface finishing process line dimensions
Length (ft.): Width (ft.): Height (ft.):
5.5
Rack Dimensions~OSP
Average number of panels per rack:
Average size of panel in rack:
Length (in.):
Average space between panels in rack:
in.
Width (in.):
Do you purposely slow the withdraw rate of your panels from process baths
to reduce drag-out? (Circle one)
Yes No
A-32
-------
5.6 Rinse Bath Water Usage-OSP
Consult the process schematic in section 2.1 to obtain the process step numbers associated with each of the water
rinse baths present in your OSP process. Enter, in the table below, the process step number along with the flow
control method and flow rate data requested for each water rinse bath. If the water rinse bath is part of a cascade,
you need only report the daily water flow rate of one bath in the cascade.
Total volume of water used by the surface finish line when operating:
gal./day |
Process Step
Number a
Example: 8
Flow Control b
R
Daily Water
Flow Rate c
2,400 gal./day
gal./day
gal./day
gal./day
gal./day
gal./day
gal./day
a Process step number - Consult the process schematic in question 2.1 and
enter the process step number of the specific water rinse tank.
b Flow control - Consult key at right and enter the letter for the flow control
method used for that specific rinse bath.
c Daily water flow rate - Enter the average daily flow rate for the specific
water rinse tank.
d Cascade water process steps - Use the step numbers for rinses that are
cascaded together.
Cascade Water
Process Steps d
8-»6
Flow Control Methods Key
[C - Conductivity Meter
[P] - pH Meter
V] - Operator Control Valve
R - Flow Restricter
N - None (continuous flow)
[O - Other (explain)
5.7 Filter Replacement-OSP
Not Applicable 1 1
Bath(s) filtered
(enter process step # from flow diagram in 2. 1)
Frequency of replacement:
Duration of replacement process:
Personal protective equipment (see key):
Personal Protective Equipment Key:
[E] - Eye Protection [G] - Gloves
[L] - Lab coat/Sleeved garment [A] - Apron
[R] - Respiratory Protection [B] - Boots
[Z] - All except Respiratory Protection
[N] - None
A-33
-------
5.8
Rack or Conveyor Cleaning~OSP
Not Applicable Q
Rack Cleaning Method (see key): OR
Conveyor Cleaning Method (see key):
Frequency of rack or conveyor cleaning:
Number of personnel involved:
Personal protective equipment (see key):
Average time required to clean:
min.
Rack Cleaning Method:
[C]-Chemical bath on SF process line
[D]-Chemical bath on another line
[T]-Temporary chemical bath
[S]-Manual scrubbing with chemical
[M]-Non-chemical cleaning
N]-None
O]-Continuous cleaning
Conveyor Cleaning Method:
[C]-Chemical rinsing or soaking
[S]-Manual scrubbing with chemical
[M]-Non-chemical cleaning
;N]-None
O]-Continuous cleaning
Personal Protective Equipment:
[E]-Eye Protection [G]-Gloves
[L]-Lab coat/Sleeved garment [A]-Apron
[R]-Respiratory Protection [B]-Boots
[O]-Continuous Cleaning [N]-None
[ZJ-A11 except Respiratory Protection
5.9
Chemical Bath Sampling --OSP
Bath Type
Example:
Cleaner
Microetch
Other (specify):
Type of
Sampling a
A
- Type of Sampling
[A] - Automated
[M] - Manual
[N] - None
b Frequency: Ent
time elapsed or nun
ft. processed betws
Clearly specify unit
ft).
sr the average
aber of panel sq.
;en samples.
s (e.g., hours, sq.
Frequency b
3 per day
Duration of
Sampling c
5 min
- Duration of Sampling: Enter the a
verage time required to manually take
a sample from the tank.
- Protective Equipment: Consult
the key for the above table and enter
the letters for all protective
equipment used by the person
performing the chemical sampling.
Protective
Equipment d
E,G,A
Method of
Sampling e
P
- Method of Sampling:
[D] - Drain or spigot
[P] - Pipette
{L] - Ladle
[O] - Other (specify)
A-34
-------
5.10 Physical Data and Operating Conditions~OSP
Complete the tables below by entering me data requested for each specific type of chemical bath listed. If two tanks of the same type are used within the process,
list the data for each tank separately.
(Average cycle time for a panel to complete entire OSP process
(includes cleaning and post cleaning steps, if any):
min.
Bath
Cleaner
Microetch
Flux
Solder
Post-Clean
Other (specify);
Physical Data
Length
(inches)
in.
in.
in.
in.
in.
in.
Width
(inches)
in.
in.
in.
in.
in.
in.
Nominal
Volume
(gal)
gal.
gal.
gal.
gal.
gal.
gal.
Process Data
Immersion
Time3
(seconds)
sec.
sec.
sec.
sec.
sec.
sec.
a Immersion Time - Enter the average elapsed time a rack of panels is immersed in
the specific process bath.
b Drip Time - Enter the average elapsed time that a rack of panels is allowed to hang
above the specific process bath to allow drainage from panels.
Drip bTime
(seconds)
sec.
sec.
sec.
sec.
sec.
sec.
Operating Conditions
Temp
(°F)
°F
°F
°F
°F
°F
°F
Agitation Methods Kev:
[PA] - Panel agitation
CP] - Circulation pump
AS] - Air sparge
[O] - Other (explain)
Agitation
(see key)
Vapor Control
(see key)
Vapor Control Methods Kev:
[BC] - Bath cover
FE] - Fully enclosed
[VO] - Vent to outside
JVC] - Vent to control
PP] - Push pull
[O] - Other (explain)
A-35
-------
5. 1 1 Initial Chemical Bath
Complete the chart below for each
oom is needed nlease attach anot
Bath
Cleaner
Microetch
Flux
Solder
Post-Clean
Other (specify)
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
i Make-Up Composition -OSP
chemical component of the bath type listec
ier sheet with the additional information
Chemical Product Name
. Provide the manufacturer name if the chemical used is known only by trade name. If more
f two tanks of the same tvne are used within the, process list the data for a single tank onlv
Manufacturer (if applicable)
Annual Quantity Used a (gallons)
a Annual Quantity Used -
pounds and clearly specify
If the amount of a particular chemical used is measured by weight (i.e., crystalline chemicals) instead of volume, enter the weight in
the units (Ibs).
A-36
-------
5.12 Chemical Bath Bailout and Additions-OSP
Complete the following chart detailing the typical bath bailout and chemical additions that are made to maintain the chemical balance of each specific process
bath. If more than three chemicals are added to a specific bath, attach another sheet with the additional information. If chemical additions to a bath are made
automatically, do not complete the last three columns for that bath. If two tanks of the same type are used within the process, list the data for a single tank only.
Bath Type
Cleaner
Microetch
Flux
Solder
Post-Clean
Other
(specify)
Bailout
Frequency
Bailout
Duration c
(minutes)
min.
min.
min.
min.
min.
min.
Bailout
Quantity
Personal
Protective
Eauinment d
Chemical Products Added
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
Criteria for
Addition"
Method of
Chemical
Addition to Tank"
Duration of
Addition0
(minutes )
min.
min.
min.
min.
min.
min.
* Criteria for Additions - Enter the b Method of Chemical Addition to Tank - Enter the letter for d Personal Protective Equipment - Enter the letters of all the
letter for the criteria typically used to the method typically used to add chemicals to the tanks. protective equipment used by the workers who physically replace the
determine when bath additions are [PR] - Poured spent bath.
necessary. [P] - Pumped manually [O] - Other E] - Eye protection
[S] - Statistical process control „, , A - Apron
P] - Panel square feet processed c Duration of Bailowp or'A^fififron - Enter the elapsed time from L - Lab coat/Sleeved Sgroegoots
C] - Chemical testing the retrieval of the chemical stock through the completion of the R - Respiratory protection
T] - Time addition of all chemicals. For bailout, enter the time required to Z] - All except Respiratory Protection
[O] - Other bailout the bath prior to making additions. [N - None
FG1 - Gloves
A-37
-------
5.13 Chemical Bath Replacement - OSP
Complete the chart below by providing information on the process of replacing, treating, and disposing of a spent chemical bath.
Bath Type
Cleaner
Microetch
Flux
Solder
Post-Clean
Other (specify)
Criteria
for Replacement a
Replacement
Frequency b
a Criteria for Replacement -
[S] - Statistical process control
P] - Panel square feet processed
[C] - Chemical testing
T] - Time
[O] - Other (specify)
b Frequency - Enter the average amount of time
elapsed, or number of square feet processed, between
bath replacements. Clearly specify units (e.g., hours,
sq.ft.).
Method of Spent
Bath Removal c
Tank
Cleaning Method d
c Methods of Spent Bath Removal-
[P] - Pump spent bath from tank
S] - Siphon spent bath from tank
[D] - Drain spent bath from tank
O] - Other (specify)
d Tank Cleaning Method -
[C] - Chemical flush
W] - Water rinse
[H] - Hand scrub
[O] - Other (specify)
e Duration of Replacement - Enter the
elapsed time from the beginning of bath
removal until the replacement bath is
finished.
Duration of
Replacement
Procedure e
min.
min.
min.
min.
min.
min.
Personal
Protective
Equipment f
f Personal Protective Equip. - Enter the letters
of all the protective equipment used by the
workers who physically replace the spent bath.
E - Eye protection
[G - Gloves
L] - Lab coat/sleeved garment
A - Apron
[R] - Respiratory protection
[B] - Boots
[Z] - All except respiratory protection
[N] - None
A-38
-------
5.14 Process Waste Disposal — OSP
Bath Type
Cleaner
Microetch
Flux
Solder
Post-Clean
Other (specify):
Annual Volume
Treated or Disposed a
a Annual Volume Treated or
Disposed - Enter the yearly
amount of the specific bath
treated or disposed. Be sure to
consider the volume treated
from both bath change outs
and bailout before entering the
total
Method of Treatment
or Disposal b
RCRA Waste
Code (if applicable)
b Methods of Treatment or Disposal -
P] - Precipitation pretreatment on-site
N] - pH neutralization pretreatment on-site
S] - Disposed directly to sewer with no treatment
D] - Drummed for off-site treatment or disposal
RN] - Recycled on-site
RF] - Recycled off-site
O] - Other (specify)
Container
Type
Container Type -
Indicate the type of
container used for disposal
of bath wastes
OH]- Open-head drum
CH]- Closed-head drum
T]- Chemical tote
O]- Other (specify)
A-39
-------
Section 6. Immersion Silver Process
6.1 Process Schematic: Immersion Silver
Fill in the figure below for your immersion silver surface finishing process. Using the key at the bottom of the page, dentify which letter
corresponds with the first step in your process and write that letter in the first box (see example). Continue using the key to fill in boxes for each
step until your entire immersion silver process is represented. If a particular process step is not represented by the key below, complete the figu:
by writing in the name of the process step in your particular surface finishing line in the corresponding bbox(es) . Finish by responding to the
questions at the bottom of the page.
Type of Process
Process _____^
Step Letter — —
(see key below)
Ex.
^A
Chemical Supplier:
Process Line Installation Date:
1.
^.
2.
>,
•
3.
^
^
4.
>.
•
5.
6.
-
7.
•
8.
>.
9.
>„
*
10.
11.
>.
*
12.
13.
>^
-
14.
>_
*
15.
Is the entire immersion silver process, as described in the chart abov
co-located in the same room:
Yes No
* If no (process steps performed in more than one room), please
circle the steps above that are in a separate room.
Type of Process Automation for the immersion silver line: (circle one)
Conveyorized Automated non-conveyorized Manually-controlled hoist
Manual (no automation) Other (specify):
Immersion Silver Process Step Key
[A] - Pre-Cleaner
[B] - Microetch
[C] - Pre-Conditioner
[D] - Immersion Silver
[£] - Water Rinse
[F] - Other (Specify in the appropriate box)
A-40
-------
6.2
General Data—Immersion Silver
Number of days the immersion silver
line is in operation:
Estimated scrap rate (% of defective
product) for the immersion silver
process:
days/yr
%
Number of hours per day the immersion
silver line is in operation:
Total of PWB surface square feet
processed by the immersion silver line per
year:
hrs/day
ssf/yr
6.3 Process Area Employees—Immersion Silver
Complete the following table by indicating the number of employees of each type that perform work duties in the
same process room as the immersion silver line, and for what length of time. Consider only workers who have
regularly scheduled responsibilities that require them to be physically within the process room. Specify "other" entry.
Enter "N/A" in any category that is not applicable.
Type of Surface Finish
Area Worker
Line Operators
Lab Technicians
Maintenance Workers
Wastewater Treatment Operators
Supervisory Personnel
Other (specify):
Number of Employees
in Surface Finish Process Area
Average Hours per Week per
Employee in Surface Finish
Process Area
hrs
hrs
hrs
hrs
hrs
hrs
6.4 Physical Settings—Immersion Silver
Size of the room containing the
surface finish process:
Are the overall process areas/rooms
ventilated (circle one)?
Do you have local vents (circle one)?
sq. ft.
Yes No
Yes No
Height of room:
Air flow rate:
Local vent air flow rate:
ft.
cu. ft./min.
cu. ft./min.
Overall surface finishing process line dimensions
Length (ft.): Width (ft.): Height (ft.):
6.5
Rack Dimensions—Immersion Silver
Average number of panels per rack:
Average size of panel in rack:
Length (in.):
Average space between panels in rack:
in.
Width (in.):
Do you purposely slow the withdraw rate of your panels from process baths
to reduce drag-out? (Circle one)
Yes No
A-41
-------
6.6 Rinse Bath Water Usage—Immersion Silver
Consult the process schematic in section 2.1 to obtain the process step numbers associated with each of the water
rinse baths present in your nickel/gold process. Enter, in the table below, the process step number along with the
flow control method and flow rate data requested for each water rinse bath. If the water rinse bath is part of a
cascade, you need only report the daily water flow rate of one bath in the cascade.
Total volume of water used by the surface finish line when operating:
gal./day |
Process Step
Number a
Example: 8
Flow Control b
R
Daily Water
Flow Rate c
2,400 gal./day
gal./day
gal./day
gal./day
gal./day
gal./day
gal./day
a Process step number - Consult the process schematic in question 2.1 and
enter the process step number of the specific water rinse tank.
b Flow control - Consult key at right and enter the letter for the flow control
method used for that specific rinse bath.
c Daily water flow rate - Enter the average daily flow rate for the specific
water rinse tank.
d Cascade water process steps - Use the step numbers for rinses that are
cascaded together.
Cascade Water
Process Steps d
8-»6
Flow Control Methods Key
[C] - Conductivity Meter
[P] - pH Meter
V] - Operator Control Valve
R] - Flow Restricter
N] - None (continuous flow)
[O] - Other (explain)
6.7 Filter Replacement—Immersion Silver
Not Applicable
D
Bath(s) filtered
(enter process step # from flow diagram in 2.1)
Frequency of replacement:
Duration of replacement process:
Personal protective equipment (see key):
Personal Protective Equipment Key:
[E] - Eye Protection [G] - Gloves
[L] - Lab coat/Sleeved garment [A] - Apron
[R] - Respiratory Protection [B] - Boots
[Z] - All except Respiratory Protection
[N] - None
A-42
-------
6.8
Rack or Conveyor Cleaning—Immersion Silver
Not Applicable Q
Rack Cleaning Method (see key): OR
Conveyor Cleaning Method (see key):
Frequency of rack or conveyor cleaning:
Number of personnel involved:
Personal protective equipment (see key):
Average time required to clean:
min.
Rack Cleaning Method:
[C]-Chemical bath on SF process line
[D]-Chemical bath on another line
[T]-Temporary chemical bath
[S]-Manual scrubbing with chemical
[M]-Non-chemical cleaning
[N]-None
[O]-Continuous cleaning
Conveyor Cleaning Method:
[C]-Chemical rinsing or soaking
[S]-Manual scrubbing with chemical
[M]-Non-chemical cleaning
[N]-None
[O]-Continuous cleaning
Personal Protective Equipment:
[E]-Eye Protection [G]-Gloves
L]-Lab coat/Sleeved garment [A]-Apron
R]-Respiratory Protection [BJ-Boots
[O]-Continuous Cleaning [N]-None
[ZJ-A11 except Respiratory Protection
6.9
Chemical Bath Sampling —Immersion Silver
Bath Type
Example:
Pre-Cleaner
Microetch
Pre-Conditioner
Immersion
Silver
Other (specify):
Type of
Sampling a
A
- Tvpe of Sampling
[A] - Automated
[M] - Manual
[N] - None
b Frequency: Ent
time elapsed or nun
ft. processed betws
Clearly specify unit
ft).
sr the average
aber of panel sq.
;en samples.
s (e.g., hours, sq.
Frequency b
3 per day
Duration of
Sampling c
5 min
- Duration of Sampling: Enter the a
verage time required to manually take
a sample from the tank.
- Protective Equipment: Consult
the key for the above table and enter
the letters for all protective
equipment used by the person
performing the chemical sampling.
Protective
Equipment d
E,G,A
Method of
Sampling e
P
- Method of Sampling:
[D] - Dram or spigot
[P] - Pipette
{L] - Ladle
[O] - Other (specify)
A-43
-------
6.10 Physical Data and Operating Conditions—Immersion Silver
Complete the tables below by entering me data requested for each specific type of chemical bath listed. If two tanks of the same type are used within the process,
list the data for each tank separately.
(Average cycle time for a panel to complete entire immersion silver process
(includes cleaning and post cleaning steps, if any):
min.
Bath
Pre-Cleaner
Microetch
Pre-Conditioner
Immersion Silver
Other (specify):
Physical Data
Length
(inches)
in.
in.
in.
in.
in.
Width
(inches)
in.
in.
in.
in.
in.
Nominal
Volume
(gal)
gal.
gal.
gal.
gal.
gal.
Process Data
Immersion
Time3
(seconds)
sec.
sec.
sec.
sec.
sec.
a Immersion Time - Enter the average elapsed time a rack of panels is immersed in
the specific process bath.
b Drip Time - Enter the average elapsed time that a rack of panels is allowed to hang
above the specific process bath to allow drainage from panels.
Drip bTime
(seconds)
sec.
sec.
sec.
sec.
sec.
Operating Conditions
Temp
(°F)
°F
°F
°F
°F
°F
Agitation Methods Kev:
[PA] - Panel agitation
CP] - Circulation pump
AS] - Air sparge
[O] - Other (explain)
Agitation
(see key)
Vapor Control
(see key)
Vapor Control Methods Kev:
[BC] - Bath cover
FE] - Fully enclosed
[VO] - Vent to outside
JVC] - Vent to control
PP] - Push pull
[O] - Other (explain)
A-44
-------
6.11 Initial Chemical Bath Make-Up Composition —Immersion Silver
Complete the chart below for each chemical component of the bath type listed. Provide the manufacturer name if the chemical used is known only by trade name. If more
room is needed, please attach another sheet with the additional information. If two tanks of the same type are used within the process, list the data for a single tank only.
Bath
Pre-Cleaner
Microetch
Pre-Conditioner
Immersion Silver
Other (specify)
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
Chemical Product Name
Manufacturer (if applicable)
Annual Ouantitv Used a (gallons')
a Annual Quantity Used -
pounds and clearly specify
If the amount of a particular chemical used is measured by weight (i.e., crystalline chemicals) instead of volume, enter the weight in
the units (Ibs).
A-45
-------
6.12 Chemical Bath Bailout and Additions—Immersion Silver
Complete the following chart detailing the typical bath bailout and chemical additions that are made to maintain the chemical balance of each specific process
bath. If more than three chemicals are added to a specific bath, attach another sheet with the additional information. If chemical additions to a bath are made
automatically, do not complete the last three columns for that bath. If two tanks of the same type are used within the process, list the data for a single tank only.
Bath Type
Pre-Cleaner
Microetch
Pre-
Conditioner
Immersion
Silver
Other
(specify)
Bailout
Frequency
Bailout
Duration c
(minutes)
min.
min.
min.
min.
min.
min.
Bailout
Quantity
Personal
Protective
Eauinment d
Chemical Products Added
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
Criteria for
Addition"
Method of
Chemical
Addition to Tank"
Duration of
Addition0
(minutes )
min.
min.
min.
min.
min.
min.
* Criteria for Additions - Enter the b Method of Chemical Addition to Tank - Enter the letter for d Personal Protective Equipment - Enter the letters of all the
letter for the criteria typically used to the method typically used to add chemicals to the tanks. protective equipment used by the workers who physically replace the
determine when bath additions are [PR] - Poured spent bath.
necessary. [P] - Pumped manually [O] - Other E] - Eye protection
[S] - Statistical process control „, , A - Apron
P] - Panel square feet processed c Duration of Bailowp or'A^fififron - Enter the elapsed time from L - Lab coat/Sleeved Sgroegoots
C] - Chemical testing the retrieval of the chemical stock through the completion of the R - Respiratory protection
T] - Time addition of all chemicals. For bailout, enter the time required to Z] - All except Respiratory Protection
[O] - Other bailout the bath prior to making additions. [N - None
FG1 - Gloves
A-46
-------
6.13 Chemical Bath Replacement —Immersion Silver
Complete the chart below by providing information on the process of replacing, treating, and disposing of a spent chemical bath.
Bath Type
Pre-Cleaner
Microetch
Pre-Conditioner
Immersion Silver
Other (specify)
Criteria
for Replacement a
Replacement
Frequency b
a Criteria for Replacement -
[S - Statistical process control
[P - Panel square feet processed
C - Chemical testing
T - Time
[O - Other (specify)
b Frequency - Enter the average amount of time
elapsed, or number of square feet processed, between
bath replacements. Clearly specify units (e.g., hours,
sq.ft.).
Method of Spent
Bath Removal c
Tank
Cleaning Method d
c Methods of Spent Bath Removal-
[P] - Pump spent bath from tank
[S] - Siphon spent bath from tank
D] - Drain spent bath from tank
O] - Other (specify)
d Tank Cleaning Method -
[C] - Chemical flush
W] - Water rinse
[H] - Hand scrub
[O] - Other (specify)
e Duration of Replacement - Enter the
elapsed time from the beginning of bath
removal until the replacement bath is
finished.
Duration of
Replacement
Procedure e
min.
min.
min.
min.
min.
Personal
Protective
Equipment f
f Personal Protective Equip. - Enter the letters
of all the protective equipment used by the
workers who physically replace the spent bath.
E - Eye protection
[G - Gloves
L] - Lab coat/sleeved garment
A - Apron
[R] - Respiratory protection
[B] - Boots
[Z] - All except respiratory protection
[N] - None
A-47
-------
6.14 Process Waste Disposal — Immersion Silver
Bath Type
Pre-Cleaner
Microetch
Pre-Conditioner
Immersion
Silver
Other (specify):
Annual Volume
Treated or Disposed a
a Annual Volume Treated or
Disposed - Enter the yearly
amount of the specific bath
treated or disposed. Be sure to
consider the volume treated
from both bath change outs
and bailout before entering the
total
Method of Treatment
or Disposal b
RCRA Waste
Code (if applicable)
b Methods of Treatment or Disposal -
P] - Precipitation pretreatment on-site
N] - pH neutralization pretreatment on-site
S] - Disposed directly to sewer with no treatment
D] - Drummed for off-site treatment or disposal
RN] - Recycled on-site
RF] - Recycled off-site
O] - Other (specify)
Container
Type
Container Type -
Indicate the type of
container used for disposal
of bath wastes
OH]- Open-head drum
CH]- Closed-head drum
T]- Chemical tote
O]- Other (specify)
A-48
-------
Section 7. Immersion Tin Process
7.1 Process Schematic: Immersion Tin
Fill in the figure below for your immersion tin surface finishing processses. Using the key at the bottom of the page, dentify which letter
corresponds with the first step in your process and write that letter in the first box (see example). Continue using the key to fill in boxes for each
step intil your entire immersion tin process is represented. If a particular process step is not represented by the key below, complete the figure b
writing in the name of the process step in your particular surface finishing line in the corresponding boxe(s). Finish oy responding to the
questions at the bottom of the page.
Type of Process
Process _____^
Step Letter — —
(see key below)
Ex.
^A
Is the entire immersion tin process, as described in the chart above,
co-located in the same room:
Yes No
* If no (process steps performed in more than one room), please
circle the steps above that are in a separate room.
Chemical Supplier:
Process Line Installation Date:
1.
^.
2.
>,
•
3.
^
^
4.
>.
•
5.
6.
-
7.
•
8.
>.
9.
>„
*
10.
11.
>.
*
12.
13.
>^
-
14.
>_
*
15.
Type of Process Automation for the immersion tin line: (circle one)
Conveyorized Automated non-conveyorized Manually-controlled hoist
Manual (no automation) Other (specify):
Immersion Tin Process Step Key
[A] - Cleaner
[B] - Microetch
[C] - Predip
[D] - Immersion Tin
[£] - Water Rinse
[F] - Dry
[G] - Other (Specify in the appropriate box
A-49
-------
7.2
General Data—Immersion Tin
Number of days the immersion tin line
is in operation:
Estimated scrap rate (% of defective
product) for the immersion tin
process:
days/yr
%
Number of hours per day the immersion tin
line is in operation:
Total of P WB surface square feet
processed by the immersion tin line per
year:
hrs/day
ssf/yr
7.3 Process Area Employees—Immersion Tin
Complete the following table by indicating the number of employees of each type that perform work duties in the
same process room as the immersion tin line, and for what length of time. Consider only workers who have regularly
scheduled responsibilities that require them to be physically within the process room. Specify "other" entry. Enter
"N/A" in any category that is not applicable.
Type of Surface Finish
Area Worker
Line Operators
Lab Technicians
Maintenance Workers
Wastewater Treatment Operators
Supervisory Personnel
Other (specify):
Number of Employees
in Surface Finish Process Area
Average Hours per Week per
Employee in Surface Finish
Process Area
hrs
hrs
hrs
hrs
hrs
hrs
7.4 Physical Settings—Immersion Tin
Size of the room containing the
surface finish process:
Are the overall process areas/rooms
ventilated (circle one)?
Do you have local vents (circle one)?
sq. ft.
Yes No
Yes No
Height of room:
Air flow rate:
Local vent air flow rate:
ft.
cu. ft./min.
cu. ft./min.
Overall surface finishing process line dimensions
Length (ft.): Width (ft.): Height (ft.):
7.5
Rack Dimensions—Immersion Tin
Average number of panels per rack:
Average size of panel in rack:
Length (in.):
Average space between panels in rack:
in.
Width (in.):
Do you purposely slow the withdraw rate of your panels from process baths
to reduce drag-out? (Circle one)
Yes No
A-50
-------
7.6 Rinse Bath Water Usage—Immersion Tin
Consult the process schematic in section 2.1 to obtain the process step numbers associated with each of the water
rinse baths present in your immersion tin process. Enter, in the table below, the process step number along with the
flow control method and flow rate data requested for each water rinse bath. If the water rinse bath is part of a
cascade, you need only report the daily water flow rate of one bath in the cascade.
Total volume of water used by the surface finish line when operating:
gal./day |
Process Step
Number a
Example: 8
Flow Control b
R
Daily Water
Flow Rate c
2,400 gal./day
gal./day
gal./day
gal./day
gal./day
gal./day
gal./day
a Process step number - Consult the process schematic in question 2.1 and
enter the process step number of the specific water rinse tank.
b Flow control - Consult key at right and enter the letter for the flow control
method used for that specific rinse bath.
c Daily water flow rate - Enter the average daily flow rate for the specific
water rinse tank.
d Cascade water process steps - Use the step numbers for rinses that are
cascaded together.
Cascade Water
Process Steps d
8-»6
Flow Control Methods Key
[C] - Conductivity Meter
[P] - pH Meter
V] - Operator Control Valve
R] - Flow Restricter
N] - None (continuous flow)
[O] - Other (explain)
7.7 Filter Replacement—Immersion Tin
Not Applicable
Bath(s) filtered
(enter process step # from flow diagram in 2.1)
Frequency of replacement:
Duration of replacement process:
Personal protective equipment (see key):
Personal Protective Equipment Key:
[E] - Eye Protection
[L] - Lab coat/Sleeved garment
[R] - Respiratory Protection
G] - Gloves
A] - Apron
[B] - Boots
Zl - All except Respiratory Protection
=- ~ "M
- None
A-51
-------
7.8
Rack or Conveyor Cleaning—Immersion Tin
Not Applicable Q
Rack Cleaning Method (see key): OR
Conveyor Cleaning Method (see key):
Frequency of rack or conveyor cleaning:
Number of personnel involved:
Personal protective equipment (see key):
Average time required to clean:
min.
Rack Cleaning Method:
[C]-Chemical bath on SF process line
[D]-Chemical bath on another line
[T]-Temporary chemical bath
[S]-Manual scrubbing with chemical
[M]-Non-chemical cleaning
[N]-None
[O]-Continuous cleaning
Conveyor Cleaning Method:
[C]-Chemical rinsing or soaking
[S]-Manual scrubbing with chemical
[M]-Non-chemical cleaning
[N]-None
[O]-Continuous cleaning
Personal Protective Equipment:
[E]-Eye Protection [G]-Gloves
L]-Lab coat/Sleeved garment [A]-Apron
R]-Respiratory Protection [BJ-Boots
[O]-Continuous Cleaning [N]-None
[ZJ-A11 except Respiratory Protection
7.9
Chemical Bath Sampling -Immersion Tin
Bath Type
Example:
Cleaner
Microetch
Predip
Immersion Tin
Other (specify):
Type of
Sampling a
A
- Type of Sampling
[A] - Automated
[M] - Manual
[N] - None
b Frequency: Ent
time elapsed or nun
ft. processed betwj
Clearly specify unit
ft).
sr the average
aber of panel sq.
;en samples.
s (e.g., hours, sq.
Frequency b
3 per day
Duration of
Sampling c
5 min
- Duration of Sampling: Enter the a
verage time required to manually take
a sample from the tank.
- Protective Equipment: Consult
the key for the above table and enter
the letters for all protective
equipment used by the person
performing the chemical sampling.
Protective
Equipment d
E,G,A
Method of
Sampling e
P
- Method of Sampling:
[D] - Drain or spigot
[P] - Pipette
{L] - Ladle
[O] - Other (specify)
A-52
-------
7.10 Physical Data and Operating Conditions—Immersion Tin
Complete the tables below by entering me data requested for each specific type of chemical bath listed. If two tanks of the same type are used within the process,
list the data for each tank separately.
(Average cycle time for a panel to complete entire immersion tin process
(includes cleaning and post cleaning steps, if any):
min.
Bath
Cleaner
Microetch
Predip
Immersion Tin
Other (specify):
Physical Data
Length
(inches)
in.
in.
in.
in.
in.
Width
(inches)
in.
in.
in.
in.
in.
Nominal
Volume
(gal)
gal.
gal.
gal.
gal.
gal.
Process Data
Immersion
Time3
(seconds)
sec.
sec.
sec.
sec.
sec.
a Immersion Time - Enter the average elapsed time a rack of panels is immersed in
the specific process bath.
b Drip Time - Enter the average elapsed time that a rack of panels is allowed to hang
above the specific process bath to allow drainage from panels.
Drip bTime
(seconds)
sec.
sec.
sec.
sec.
sec.
Operating Conditions
Temp
(°F)
°F
°F
°F
°F
°F
Agitation Methods Key:
[PA] - Panel agitation
CP] - Circulation pump
AS] - Air sparge
[O] - Other (explain)
Agitation
(see key)
Vapor Control
(see key)
Vapor Control Methods Key:
[BC] - Bath cover
FE] - Fully enclosed
[VO] - Vent to outside
JVC] - Vent to control
PP] - Push pull
[O] - Other (explain)
A-53
-------
7.11 Initial Chemical Bath Make-Up Composition —Immersion Tin
Complete the chart below for each chemical component of the bath type listed. Provide the manufacturer name if the chemical used is known only by trade name. If more
room is needed, please attach another sheet with the additional information. If two tanks of the same type are used within the process, list the data for a single tank only.
Bath
Cleaner
Microetch
Predip
Immersion Tin
Other (specify)
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
Chemical Product Name
Manufacturer (if applicable)
Annual Quantity Used a (gallons')
a Annual Quantity Used -
pounds and clearly specify
If the amount of a particular chemical used is measured by weight (i.e., crystalline chemicals) instead of volume, enter the weight in
the units (Ibs).
A-54
-------
7.12 Chemical Bath Bailout and Additions—Immersion Tin
Complete the following chart detailing the typical bath bailout and chemical additions that are made to maintain the chemical balance of each specific process
bath. If more than three chemicals are added to a specific bath, attach another sheet with the additional information. If chemical additions to a bath are made
automatically, do not complete the last three columns for that bath. If two tanks of the same type are used within the process, list the data for a single tank only.
Bath Type
Cleaner
Microetch
Predip
Immersion
Tin
Other
(specify)
Bailout
Frequency
Bailout
Duration c
(minutes)
min.
min.
min.
min.
min.
min.
Bailout
Quantity
Personal
Protective
Eauinment d
Chemical Products Added
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
1
2
3
Criteria for
Addition"
Method of
Chemical
Addition to Tank"
Duration of
Addition0
(minutes )
min.
min.
min.
min.
min.
min.
* Criteria for Additions - Enter the bMethod of Chemical Addition to Tank - Enter the letter for the d Personal Protective Equipment - Enter the letters of all the
letter for the criteria typically used to method typically used to add chemicals to the tanks. protective equipment used by the workers who physically replace the
determine when bath additions are [PR] - Poured spent bath.
necessary. [P] - Pumped manually [O] - Other E] - Eye protection
[S] - Statistical process control „, , A - Apron
P] - Panel square feet processed c Duration of Bailowp or'A^fififron - Enter the elapsed time L - Lab coat/Sleeved Sgroegoots
C] - Chemical testing from the retrieval of the chemical stock through the completion R - Respiratory protection
T] - Time of the addition of all chemicals. For bailout, enter the time Z] - All except Respiratory Protection
[O] - Other required to bailout the bath prior to making additions. [N - None
FG1 - Gloves
A-55
-------
7.13 Chemical Bath Replacement — Immersion Tin
Complete the chart below by providing information on the process of replacing, treating, and disposing of a spent chemical bath.
Bath Type
Cleaner
Microetch
Predip
Immersion Tin
Other (specify)
Criteria
for Replacement a
Replacement
Frequency b
a Criteria for Replacement -
[S] - Statistical process control
P] - Panel square feet processed
[C] - Chemical testing
T] - Time
[O] - Other (specify)
b Frequency - Enter the average amount of time
elapsed, or number of square feet processed, between
bath replacements. Clearly specify units (e.g., hours,
sq.ft.).
Method of Spent
Bath Removal c
Tank
Cleaning Method d
c Methods of Spent Bath Removal-
[P] - Pump spent bath from tank
S] - Siphon spent bath from tank
[D] - Drain spent bath from tank
O] - Other (specify)
d Tank Cleaning Method -
[C] - Chemical flush
W] - Water rinse
[H] - Hand scrub
[O] - Other (specify)
e Duration of Replacement - Enter the
elapsed time from the beginning of bath
removal until the replacement bath is
finished.
Duration of
Replacement
Procedure e
min.
min.
min.
min.
min.
Personal
Protective
Equipment f
f Personal Protective Equip. - Enter the letters
of all the protective equipment used by the
workers who physically replace the spent bath.
E - Eye protection
[G - Gloves
L] - Lab coat/sleeved garment
A - Apron
[R] - Respiratory protection
[B] - Boots
[Z] - All except respiratory protection
[N] - None
A-56
-------
7.14 Process Waste Disposal — Immersion Tin
Bath Type
Cleaner
Microetch
Predip
Immersion Tin
Other (specify):
Annual Volume
Treated or Disposed a
a Annual Volume Treated or
Disposed - Enter the yearly
amount of the specific bath
treated or disposed. Be sure to
consider the volume treated
from both bath change outs
and bailout before entering the
total
Method of Treatment
or Disposal b
RCRA Waste
Code (if applicable)
b Methods of Treatment or Disposal -
P] - Precipitation pretreatment on-site
N] - pH neutralization pretreatment on-site
S] - Disposed directly to sewer with no treatment
D] - Drummed for off-site treatment or disposal
RN] - Recycled on-site
RF] - Recycled off-site
O] - Other (specify)
Container
Type
Container Type -
Indicate the type of
container used for disposal
of bath wastes
OH]- Open-head drum
CH]- Closed-head drum
T]- Chemical tote
O]- Other (specify)
A-57
-------
Facility Background Information
Design
for the
Environment
Printed Wiring Board Project
Performance Demonstration Questionnaire
Please complete this questionnaire, make a copy for
your records, and send the original to:
Ellen Moore
Abt Associates
55 Wheeler St.
Cambridge, MA 02138
Fax: (617) 349-2660
Note: The completed questionnaire must be returned PRIOR TO the
scheduled site visit.
FACILITY AND CONTACT INFORMATION
^acilitv Identifies
Company
Name:
Site Name:
Street Address:
City:
tion-
| State: |
Zip:
Contact Td
Name:
Title:
Phone:
Fax:
E-Mail:
pntification' Filter the names of the persons wV
o can he contacted repardinp this snrvev
A-58
-------
Section 1. Facility Characterization
Estimate manufacturing data for the previous 12 month period or other convenient time period of 12
consecutive months (e.g., FY96). Only consider the portion of the facility dedicated to PWB
manufacturing when entering employee and facility size data.
1.1
General Information
Size of portion of facility used for
manufacturing PWBs.
Size of portion of facility used for
surface finishing.
Sq. Ft.
Sq. Ft.
Number of days Surface Finish line is
in operation:
days/yr
1.2 Process Type
Estimate the percentage of PWBs manufactured at your facility using the following methods for surface
finishing (SF). Specify "other" entry.
Type of PWB process
HASL
OSP-Thick
OSP-Thm
Immersion Tin
Immersion Silver
Percent of total
%
%
%
%
%
Type of PWB process
Electroless Palladium
Electroless Nickel/Immersion Gold
Other:
Other:
TOTAL
Percent of Total
%
%
%
%
100%
1.3 General Process Line Data
Process Data
Number of hours the Surface Finishing line is in operation per day:
Hours
1.4 Process Area Employees
Complete the following table by indicating the number of employees of each type that perform work
duties in the same process room as the Surface Finishing line and for what length of time. Report the
number of hours per employee. Consider only workers who have regularly scheduled responsibilities
physically within the process room. Specify "other" entry. Enter "N/A" in any category not applicable.
Type of Process
Area Worker
Line Operators
Lab Technicians
Maintenance Workers
Wastewater Treatment Operators
Supervisory Personnel
Other:
Other:
Number of Employees
in Process Area
Average Hours per Week per
Employee in Process Area
Hrs.
Hrs.
Hrs.
Hrs.
Hrs.
Hrs.
Hrs.
A-59
-------
1.5 Wastewater Discharge and Sludge Data
Wastewater discharge type (check one)
Direct
Indirect
Annual weight (quantity in pounds) of sludge generated:
Is sludge dewatered prior to disposal?
% water content prior to dewatering:
% water content after dewatering:
Zero
Section 2. Process Description: Immersion Tin
2.1 Process Schematic
Fill in the following table by identifying what type of surface finishing process (e.g., HASL) your facility
uses. Then, using the proper key at the bottom of the page, identify which letter corresponds with the first
step in your process and write that letter in the first box (see example). Continue using the key to fill in
boxes for each step in your process until your entire surface finishing process is represented. If your
process is not represented by a key below, complete the chart by writing in the name of each process step
in your particular surface finishing line.
A-60
-------
Name of Process:
Process
Step Letter
(see key below)
1.
2.
3.
>„
*
6.
11.
7.
8.
-
9.
>,
12.
13.
>„
•
14.
10.
15.
Immersion Tin Process Step Key
[A] - Cleaner [D] - Immersion Tin
[B] - Microetch [E] - Water Rinse
[C] - Predip [F] - Other (specify step)
A-61
-------
2.2 Rinse Bath Water Usage
Consult the process schematic in section 2.1 to obtain the process step numbers associated with each of
the water rinse baths present. Enter, in the table below, the process step number along with the flow
control method and flow rate data requested for each water rinse bath. If the water rinse bath is part of
a cascade, you need only report the daily water flow rate of one bath in the cascade.
Amount of water used by the surface finishing line when operating:
Process Step Number a
Example: 8
Flow Control b
R
Daily Water Flow
Ratec
2,400 gal./day
gal. /day
gal./day
gal./day
gal./day
gal/, day
gal./day
gal./day
gal./day
a Process step number - Consult the process schematic in question 2. 1 and
enter the process step number of the specific water rinse tank.
b Flow control - Consult key at right and enter the letter for the flow control
method used for that specific rinse bath.
c Daily water flow rate - Enter the average daily flow rate for the specific
water rinse tank.
d Cascade water process steps - Use the step numbers for rinses that are
cascaded together.
gal/day
Cascade Water Process Steps d
86
Flow Control Methods Key
[C
P
[V
[R
[N
[O
- Conductivity Meter
- pH Meter
- Operator Control Valve
- Flow Restricter
- None (continuous flow)
- Other (explain)
2.3
Process Parameters
Size of the room containing the process
Are the overall process
areas (not tank vent) ventilated? (Circle one)
Air flow rate:
Do you have local vents?
Local vent air flow rate:
sq. ft.
Height of room
No
cu.ft.min.
No
cu. Ft./min.
Type of process automation for surface finishing line: (circle one)
Automated non-conveyorized Automated convey orized Manually controlled hoise
Manual (no automation) Other, specify:
A-62
-------
2.4 Physical, Process, and Operating Conditions
Complete the table below by entering the data requested for each specific type of chemical bath listed. If
two tanks of the same type are used within the process, list the data for a single tank only.
BATH
Acid cleaner
Microetch
Acid predip
Immersion tin
Other (specify)
LENGTH (inches)
in.
in.
in.
in.
in.
in.
in.
in.
WIDTH (inches)
in..
in..
in..
in..
in..
in..
in..
in..
NOMINAL VOLUME
gal.
gal.
gal.
gal.
gal.
gal.
gal.
gal.
A-63
-------
2.5 Initial Chemical Bath Make-Up Composition
Complete the chart below for each chemical component of the bath type listed. Provide the manufacturer name if the chemical used is
known only by trade name. If more room is needed, please attach another sheet with the additional information. If two tanks of the
same type are used within the process, list the data for a single tank only.
BATH
CLEANER
MICROETCH
ACID PREDIP
IMMERSION
TIN
OTHER
(specify)
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
CHEMICAL PRODUCT NAME
MANUFACTURER
(if applicable)
ANNUAL QUANTITY USED3
(gallons)
a Annual Quantity Used - If the amount of a particular chemical used is measured by weight (i.e., crystalline chemicals) instead of
volume, enter the weight in pounds and clearly specify the units (Ibs).
A-64
-------
2.6 Chemical Bath Replacement
Complete the chart below by providing information on the process of replacing, treating, and disposing of a spent chemical bath.
Bath Type
Criteria
for
Replacement
Frequency
Tank
Cleaning
Method'
Duration of
Replacement
Procedured
Personal
Protective
Equipment'
Method of
Treatment or
Disposalf
Annual
Volume
Treated or
Disposedg
ACID CLEANER
mm.
MICROETCH
mm.
ACID PREDIP
mm.
IMMERSION
TIN
mm.
Criteria for Replacement -
- Statistical process control
- Panel square feet processed
- Chemical testing
- Time
- Other (specify)
b Frequency - Enter the average amount
of time elapsed, or number of square feet
processed, between bath replacements.
Clearly specify units (e.g., hours, sq.ft.,
etc.)
cTank Cleaning Method
C] - Chemical Flush
- Water Rinse
C]
W]
H] - Hand Scrub
O] - Other (specify)
d Duration of Replacement-
Enter the elapsed time from the beginning
of bath removal until the replacement bath
is finished.
f Personal Protective Equip. -
Enter the letters of all the protective
equipment used by the workers who
mysically replace the spent bath.
E] - Eye protection
G - Gloves
L] - Lab coat/sleeved garment
A - Apron
R - Respiratory protection
B -Boots
Z] - All except respiratory protection
N -None
f Methods of Spent Bath Removal -
T] - Precipitation Pretreatment on-site
N] - PH Neutralization Pretreatment on-site
S] - Disposed directly to sewer with no treatment
D] - Drummed for off-site treatment or disposal
RN] - Recycled on-site
'RF] - Recycled off-site
O] - Other (specify)
g Annual Vol. Treat. Or Disp. -
Enter the yearly amount of the specific bath treated or
disposed. Needed only if water testing is not done.
A-65
-------
2.7 Chemical Bath Additions
Complete the following chart detailing the typical chemical additions that are made to maintain the chemical balance of each specific process bath.
If more than four chemicals are added to a specific bath, attach another sheet with the additional information. If chemical additions to a bath are
made automatically, do not complete the last three columns for that bath. If two tanks of the same type are used within the process, list the data for
Bath Type
CLEANER
MICROETCH
ACID PREDIP
IMMERSION
TIN
OTHER (specify):
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
Chemical Products
Added
"Criteria for Replacement -
Enter the letter for the criteria typically used to determine when bath
replacement is necessary.
[S] - Statistical Process Control
P] - Panel Square Feet Processed
C] - Chemical Testing
T] -Time
O] - Other
Criteria for
Replacement"
Method of Chemical
Addition to Tankb
"Method of Chemical Addition to Tank -
Enter the letters for the method typically used to add chemicals to
the tanks.
[P] - Pumped Manually
[PR] - Poured
[S] - Scooped
[O] - Other
'Duration of Addition - Enter the average elapsed time from the
retrieval of the chemical stock through the completion of the
addition of all chemicals
Duration of
Addition0 (minutes)
Personal Protective
Equipment11
"Personal Protective Equipment - Enter the letters of all the
protective equipment worn by the workers physically replacing
the spent bath.
[E] - Eye protection
G] - Gloves
L] - Labcoat/Sleeved garment
A - Apron
R - Respiratory protection
B -Boots
Z] - All except Respiratory Protection
N - None
A-66
-------
Observer Data Sheet
Observer Data Sheet
DfE PWB Performance Demonstrations
Facility name and location:
Surface finishing process type and name:
Date: Contact Name:
Installation date:
Test Panel Run
Overall Surface Finishing process line dimensions
Length (ft.): Width (ft.):
Height (ft.):
Average number of panels per rack:
Average space between panels in rack:
Average size of panel in rack: Length(in):
Width (in.):
At what % of capacity is the line currently
running?
At what % of capacity is the line typically
running?
What is the overall throughout?
How is it calculated:
surface sq.ft./year
Estimated yield for surface finishing line:
Number of thermal cycles the finished board can withstand:
Note any unusual storage conditions or oxidation.
Load system with layer 4 facing up or toward the operator.
While running the test panels, verify each process step and complete the table on the next page.
Test Panel Serial Numbers
Test Board
1.
2.
Serial #
Test Board
3.
4.
Serial #
Test Board
5.
6.
Serial #
A-67
-------
Test Panel Run
Bath Name
(from schematic)
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
Equipment"
Bath
Temp
Immersion
Time
Drip
Time
Overall System Time:
a List Number, type of
Agitation: Vapor Control: Filter Type: Heater Control: Water Rinses:
[PA] - Panel agitation [ BC] - Bath Cover [BF] - Bag [TH] - Thermostat [CN] - Continuous
[CP] - Circulation Pump []FE> -Fully Enclosed [CF] - Cartridge [TM] - Timer [DP] - Continuous During Process
[AS] - Air Sparge [ VO] - Vent to Outside [PR] - Programmed [PP] - Partial During Process
[VC]- Vent to Control
A-68
-------
Verification of Part A (mark any changes on working copy of Part A):
Ventilation:
Verify the type of ventilation as recorded in the Questionnaire:
Tank Volumes:
Verify the length, width, and volume of each tank, as recorded in the Questionnaire:
Water use:
Verify water use data, for each tank: .—.
Daily water flow rate verified I—I
Cascade process steps verified LJ
Pollution Prevention:
Have you used any other pollution prevention techniques on the surface finishing line (e.g., covered
tanks to reduce evaporation, measures to reduce dragout, changes to conserve water, etc.)?
If yes, describe and quantify results (note: if results have not been quantified, please provide an
estimate):
If your throughput changed during the time new pollution prevention techniques were
implemented, estimate how much (if any) of the pollution reductions are due to the throughput
changes:
A-69
-------
Filter Replacement
Bath(s) filtered (enter process step #)
Frequency of replacement:
Duration of replacement process:
Personal protective equipment (see key):
Personal Protective Equipment Key:
- Eye Protection
- Labcoat/Sleeved garment
- Respiratory Protection
[A] - Apron
[B] - Boots
[G] - Gloves
[Z] - All except Respiratory Protection
[N] - None
Equipment Maintenance
Estimate the maintenance requirements (excluding filter changes and bath changes) of the surface
finishing process equipment for both outside services calls (maintenance by vendor or service
company) and in-house maintenance (by facility personnel).
Describe the typical maintenance activities associated with the surface finishing process line (e.g.,
motor repair/replacement, conveyor repairs, valve leaks, etc.)
Average time spent per week:
Average downtime:
If there a recurring maintenance problem?
If yes, describe:
A-70
-------
Rack or Conveyor Cleaning Not Applicable Q
Frequency of rack or conveyor cleaning:
Rack Cleaning Method (see key): OR
Conveyor Cleaning Method (see key):
Number of personnel involved:
Personal protective equipment (see key):
Average time required to clean:
- Chemical bath on SF process line
- Chemical bath on another line
- Temporary chemical bath
- Manual scrubbing with chemical
- Non-chemical cleaning
-None
- Continuous cleaning
Conveyor Cleaning Method:
rCl - Chemical rinsing or soaking
- Manual scrubbing with chemical
- Non-chemical cleaning
-None
- Continuous cleaning
Personal Protective Equipment:
[E]
G
L
R
O
Z]
- Eye Protection
- Gloves
- Labcoat/Sleeved garment
- Respiratory Protection
- Continuous Cleaning
- All except Respiratory Protection
A] - Apron
B] - Boots
N] - None
Chemical Bath Sampling
Bath Type
Cleaner
Microetch
Flux
Solder
Post Clean
Other
(specify)
Other
(specify)
Type of
Sampling"
Frequency b
Duration of
Sampling c
Protective
Equipment d
Method of
Sampling e
" Type of Sampling c Duration of Sampling: ' Method of Obtaining Samples:
[A] - Automated Enter the average time for manually [D] - Drain or spigot
[M] - Manual taking a sample from the tank [p] - Pipette
[N] - None [L] - Ladle
[O] - Other (specify)
" Frequency: * Protective Equipment:
Enter the average time elapsed Consult the key for the above table
or number of panel sq. ft. processed and enter the letters for all protective
between samples. Clearly specify equipment worn by the person performing
units (e.g., hours, sq. ft., etc.) the chemical sampling.
A-71
-------
Process Description
Process Schematic
Fill in the table below by identifying what type of alternative surface finishing process (e.g., immersion tin) your company uses. Then, using the key at the bottom left of the
page, identify which letter corresponds with the first bath step in your process and write that letter in the first box (see example). Continue using the key to fill in boxes for each
step in your process until your entire alternative surface finishing process is represented. If your process step is not represented by the key below, complete the chart by writing in
the name of the process step in your particular surface finishing fine.
Process Automation
Letter (see key below right)
Type of Process
(write in process name)
Process Steps of
Your Facility
(begin here)
Standard Bath Types
A
B
C
D
E
F]
- Center
- Conditioner
- Micro Etch
- Pre-dip
- Catalyst
- Activator
G] - Accelerator
H] - Enhancer
J] - Electroless Nickel
K] - Electroless Gold
L] - Electroless Palladium
M] - Immersion
'alladium
[N] - Immersion Gold
P] - Immersion Tin
Q] - Immersion Silver
R] - OSP
S] - Anti-tarnish
W] - Water rinse
[O] - Other (specify step)
Process Automation
Please list all the process types with which the above process may be operated in:
Process Automation Key
orized
[A] - All of the above
[PJ - Automated on-conveyorized [S] - Manually controlled hoist
'Q] - Automated conveyorized [T] - Manual (no information)
'R] -
[V] - Other (specify)
- Partially automated
A-72
-------
Comparative Evaluation:
If the facility has switched from a previous system to the current system, complete this page.
Product Quality:
What, if any, changes were noticed in the quality of the boards produced? (Yield change?)
Installation:
How long was the debug period when this system was installed?
What were the types of problems encountered:
Manufacturing Process Changes: How did you change your upstream or downstream processes
when this system was installed (e.g., did you have to make changes in your solder mask)?
Waste Treatment:
Have any of your waste treatment methods or volumes changed due to the installation of this system
(not associated with volume changes due to throughput changes)?
If yes, describe the change(s) and attach quantitative information, if available:
Process Safety:
Have any additional OSHA-related procedures or issues arisen as a result of changing to the present
system (e.g., machinery lock-outs while cleaning, etc)? If so, describe:
Customer Acceptance:
Have customers accepted the new process? Why or why not:
Other:
Describe any other issues that have arisen as a result of the new process.
A-73
-------
Supplier Data Sheet
DfE Printed Wiring Board Project
Alternative Technologies for Surface Finishing
Manufacturer/Supplier Product Data Sheet
Manufacturer Name:
Address:
Contact:
Phone: .
Fax: —
How many alternative making holes conductive product lines will you submit for testing?
Please complete a Data Sheet for each product line you wish to submit for testing. In addition,
if you have not already done so, please submit the material safety data sheets (MSDS), product
literature, and the standard manufacturer instructions for each product line submitted.
Product Line Name: Category:*
* Categories of Product Lines:
A. HASL
B. Immersion Tin
C. Immersion Palladium
D. Electroless Nickel/Immersion Gold
E. Nickel/Palladium/Immersion Gold
F. OSP-(Thin)
G OSP-(Thick)
For the product line listed above, please identify one or two facilities that are currently using the
produ9t line at which you would like your product demonstrated. Also, identify the location of the
site (city, state) and whether the site is 1) a customer production site, 2) a customer test site, or 3)
your own supplier testing site.
Facility 1 Name and Location:
Type of Site:
Facility Contact:
May we contact the facility at this time (yes or no):
Facility 2 Name and Location:
Type of Site:
Facility Contact Phone:
May we contact the facility at this time (yes or no):
A-74
-------
Energy Usage
For each piece of equipment in the surface finishing line using energy, complete the table below:
Equipment Type
Tank or
Station # a
Power Rating
(from nameplate)
Load
(1% capacity in use)
Equipment
Cost
Period of Usage
continuous
_ continuous during process cycle
partial during process cycle. If partial, record
how often:
_ other:
continuous
_ continuous during process cycle
partial during process cycle. If partial, record
how often:
_ other:
continuous
_ continuous during process cycle
partial during process cycle. If partial, record
how often:
_ other:
continuous
_ continuous during process cycle
partial during process cycle. If partial, record
how often:
_ other:
continuous
_ continuous during process cycle
partial during process cycle. If partial, record
how often:
_ other:
Machine Control
_ timer
_ program
operator/manual
_ other:
_ timer
_ program
operator/manual
_ other:
_ timer
_ program
operator/manual
_ other:
_ timer
_ program
operator/manual
_ other:
_ timer
_ program
operator/manual
_ other:
Specify whether tank number of process flow diagram step numbers are used.
A-75
-------
Special Product Characteristics
1. Does the process operate as a vertical process, horizontal process, or either?
2. Average number of thermal excursions the finished board can withstand?
3. Most likely process step preceding the beginning of the surface finish application?
4. Should the application of solder mask occur after the application of the surface finish, or before?
5. Which of the following technologies is the surface finish compatible with?
(Circle all applicable choices.)
A. SMT D. Gold Wire Bonding
B. Flip Chip E. Aluminum Wire Bonding
C. EGA F. Other, Explain:
6. Please state cycle time of surface finish process line.
7. Please describe any special process equipment recommended (e.g., high pressure rinse, air knife, dryer,
aging equipment, etc.).
Product Line Constraints
1. Please list any substrate incompatibilities (e.g., BT, cyanate ester, Teflon, Kevlar, copper invar copper,
polyethylene, other [specify])
2. Please list compatibilities with solder masks.
3. Are there any special requirements needed for the soldering process (e.g., type of flux, etc.)?
4. Average shelf-life of finished boards?
5. Other general comments about the product line (include any known impacts on other process
steps).
A-76
-------
Bath Life
Please fill in the following table (for bath listings, please refer back to your process description on page
2).
Bath
1.
2.
3.
4.
5.
6.
7.
8.
Recommended
Treatment/Disposal
Method a
Criteria for Dumping
Bath
(e.g., time, surface sq ft of
panel processed,
concentration, etc.)
Recommended Bath
Life
(in terms of criteria listed
at left)
Attach and reference materials, if necessary.
A-77
-------
Costs:
Chemical Cost
Please provide the cost per gallon (or pound) of chemical for each chemical product required to operate
this alternative surface finishing product line. It is recognized that the cost of chemicals is, in part,
dependant on the amount of chemical purchased (i.e., volume discounts) and may vary accordingly. If
cost would decrease, please write decreased cost in margin along with volume of chemical required for
pricing discount.
Bath Name
1.
2.
3.
4.
5.
6.
7.
Product Name
A.
B.
C.
A.
B.
C.
A.
B.
C.
A.
B.
C.
A.
B.
C.
A.
B.
C.
A.
B.
C.
Chemical Cost
($/gal or $/lb)
A-78
-------
Equipment Cost
Do you recommend or suggest any specific equipment manufacturers to customers for obtaining process
equipment to operate this surface finish line? If so, why? Please provide the contact information for
equipment manufacturer below.
Equipment Company # 1
Company Name:
Contact Name:
Phone number:
Equipment Type:
Equipment Company # 2
Company Name:
Contact Name:
Phone number:
Equipment Type:
Do either of the companies listed above manufacture equipment specifically designed for your product
line? Which one?
If so, what is special or different about the equipment design?
A-79
-------
Appendix B
Bath Chemistry Data
-------
Contents
Table B-l. Bath Concentrations for the HASL Technology
Table B-2. Bath Concentrations for the Electroless Nickel/Immersion Gold Technology
Table B-3. Bath Concentrations for the Electroless Nickel/Electroless Palladium/Immersion
Gold Technology
Table B-4. Bath Concentrations for the OSP Technology
Table B-5. Bath Concentrations for the Immersion Silver Technology
Table B-6. Bath Concentrations for the Immersion Tin Technology
-------
Table B-l. Bath Concentrations for the HASL Technology
Bath
Chemicals
Concentration in Bath (g/1)
Cleaner
Alkylphenolpolyethoxyethanol
Ethylene glycol monobutyl ether
Fluoboric acid
Phosphoric acid
Sulfuric acid
*9 other confidential chemicals
18.00
22.90
12.33
61.11
110.40
Microetch
1,4-Butenediol
Copper sulfate pentahydrate
Hydrogen peroxide
Sodium hydroxide
Sulfuric acid
*7 other confidential chemicals
12.72
45.00
50.73
0.170
103.50
Table B-2. Bath Concentrations for the Electroless Nickel/Immersion Gold Technology
Bath
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Immersion Gold
Chemicals
Phosphoric acid
Sulfuric acid
Hydrochloric acid
Alkylphenolpolyethoxyethanol
*Two other confidential chemicals
Sodium hydroxide
Hydrogen peroxide
Copper sulfate pentahydrate
Sulfuric acid
*Two other confidential chemicals
Hydrochloric acid
*Four other confidential chemicals
*Two confidential chemicals
Nickel sulfate
*13 other confidential chemicals
Potassium gold cyanide
*Four other confidential chemicals
Concentration in Bath (g/1)
50.8
138
17.85
18.00
0.170
35.88
45.00
87.40
55.80
37.24
2.999
B-l
-------
Table B-3. Bath Concentrations for the Electroless Nickel/Electroless
Palladium/Immersion Gold Technology
Bath
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Preinitiator
Electroless Palladium
Immersion Gold
Chemical
Phosphoric acid
*2 other confidential chemicals
Sodium hydroxide
Hydrogen peroxide
Copper sulfate pentahydrate
Sulfuric acid
* 1 other confidential chemical
*4 confidential chemicals
* 1 confidential chemical
Nickel sulfate
*10 other confidential chemicals
*4 confidential chemicals
Ethylenediamine
Propionic acid
Maleic acid
*6 other confidential chemicals
Potassium gold cyanide
*4 other confidential chemicals
Concentration in Bath (g/1)
50.80
0.17
35.88
45.00
156.40
58.65
4.45
7.30
2.00
3.00
Table B-4. Bath Concentrations for the OSP Technology
Bath
Cleaner
Microetch
OSP
Chemical
Phosphoric acid
Sulfuric acid
*3 other confidential chemicals
Sodium hydroxide
Hydrogen peroxide
Copper sulfate pentahydrate
Sulfuric acid
*6 other confidential chemicals
Copper ion
*5 other confidential chemicals
Concentration in Bath (g/1)
50.80
9.20
0.170
18.165
45.00
250.70
50.50
B-2
-------
Table B-5. Bath Concentrations for the Immersion Silver Technology
Bath
Cleaner
Microetch
Predip
Immersion Silver
Chemical
Phosphoric acid
1,4-Butenediol
Sulfuric acid
Hydrogen peroxide
Sodium hydroxide
*4 other confidential chemicals
Sodium hydroxide
*5 other confidential chemicals
Concentration in Bath
(g/1)
122.90
12.72
4.60
113.00
29.36
26.43
Table B-6. Bath Concentrations for the Immersion Tin Technology
Bath
Chemical
Concentration in Bath (g/1)
Cleaner
Ethylene glycol monobutyl ether
Fluoboric acid
Sulfuric acid
Phosphoric acid
*6 other confidential chemicals
22.90
12.33
184.00
30.25
Microetch
Sulfuric acid
* 1 other confidential chemical
18.40
Predip
Methane sulfonic acid
Sulfuric acid
*10 other confidential chemicals
337.50
0.0092
Immersion Tin
Sulfuric acid
Urea
1,3 -Diethylthiourea
Tin chloride
Methane sulfonic acid
Stannous methane sulfonic acid
*14 other confidential chemicals
92.18
90.00
20.00
13.98
69.17
111.80
B-3
-------
Appendix C
Chemical Properties Data
-------
Contents
1,3-Diethylthiourea C-l
1,4-Butenediol C-3
Acetic Acid C-4
Branched Octylphenol, Ethoxylated C-8
Ammonium Chloride C-9
Ammonium Hydroxide C-12
Sodium Citrate (citric acid) C-14
Cupric Sulfate (copper ion) C-16
Cupric Acetate (copper sulfate pentahydrate) C-l8
Ethylenediamine C-20
Ethylene Glycol C-22
Ethylene Glycol Monobutyl Ether C-24
Fluoroboric Acid (fluoride) C-25
Hydrochloric Acid C-30
Hydrogen Peroxide C-32
Lead C-34
Maleic Acid C-35
Malic Acid C-37
Methanesulfonic Acid C-39
Nickel Sulfate C-41
Palladium Chloride C-43
Phosphoric Acid C-45
Potassium Aurocyanide C-47
Potassium Peroxymonosulfate C-49
Propionic Acid C-51
Silver Nitrate C-54
Sodium Hydroxide C-56
Sodium Hypophosphite and Sodium Hypophosphite Monohydrate C-58
Stannous Methanesulfonic Acid C-61
Sulfuric Acid C-63
Thiourea C-65
Tin C-67
Tin Chloride C-68
Urea C-70
References C-72
-------
CHEMICAL SUMMARY FOR 1,3-DIETHYLTHIOUREA
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of 1,3-diethylthiourea are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF 1,3-DIETHYLTHIOUREA
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
Data
105-55-5
N,N-diethylthiourea
C5H12N2S
C2H5NHCSNHC2H5
buff solid
132.32
78 °C
decomposes
4.56 g/L
1.11 mg/m3
no data
49 (estimated)
0.57
0.240 mm Hg at 25 °C (estimated)
no data
no data
no data
no data
no data
no data
6.9xlO"8 atm mVmole (estimated)
2 (estimated)
no data
Reference
Lide(1995)
Lide(1995)
Lide(1995)
Lewis (1993)
Lewis (1993)
Lide(1995)
Lide(1995)
Lide(1995)
PHYSPROP(1998)
Ohm (1997)
HSDB(1998)
PHYSPROP(1998)
PHYSPROP(1998)
PHYSPROP(1998)
HSDB(1998)
IL ENVIRONMENTAL FATE
A. Aquatic Fate
If released into aquatic waters, 1,3-diethylthiourea is not expected to adsorb to suspended solids and sediments in
water based upon an estimated Koc of 49 (HSDB, 1998; Swann et al., 1983), determined from a log Kow of 0.57
(Covers et al., 1986, as cited in PHYSPROP, 1998) and a regression-derived equation (Lyman et al., 1990).
Volatilization from the water column to the atmosphere is not expected to occur (Lyman et al., 1990) based on an
estimated Henry's Law constant of 6.9xlO'8 atm-mVmole (PHYSPROP, 1998; SRC, 1998). Since thiourea, a
structurally similar compound, was found to be stable to hydrolysis and photolysis (Schmidt-Bleek et al., 1982, as
cited in HSDB, 1998), 1,3-diethylthiourea is also expected to be stable to both hydrolysis and photolysis. According
C-l
-------
to a classification scheme (Franke et al., 1994), an estimated BCF of 2 (HSDB, 1998; Lyman et al., 1990) suggests
that the potential for bioconcentration in aquatic organisms is low.
B. Atmospheric Fate
According to a model of gas/particle partitioning of semivolatile organic compounds in the atmosphere (Bidleman,
1988), 1,3-diethylthiourea, which has an estimated vapor pressure of 0.24 mm Hg at 25 °C(PHYSPROP, 1998;
SRC, 1998), should exist solely as a vapor in the ambient atmosphere. The predominant removal process of 1,3-
diethylthiourea from the atmosphere is reaction with photochemically-produced hydroxyl radicals; the half-life for
this reaction in air is estimated to be 4 hours (Atkinson, 1988). 1,3-diethylthiourea, which has a high estimated water
solubility of 4.56 g/L (PHYSPROP, 1998; SRC, 1998), is expected to adsorb onto atmospheric paniculate material;
the small amount of 1,3-diethylthiourea deposited onto paniculate material may be physically removed by wet and dry
deposition (HSDB, 1998).
C. Terrestrial Fate
Based on a classification scheme (Swann et al., 1983), an estimated Koc of 49 (HSDB, 1998), determined from a log
Kow of 0.57 (Covers et al., 1986, as cited in PHYSPROP, 1998) and a regression-derived equation (Lyman et al.,
1990), indicates that 1,3-diethylthiourea is expected to have very high mobility in soil. Volatilization of 1,3-
diethylthiourea from moist soil surfaces is not expected to be important (Lyman et al., 1990) given an estimated
Henry's Law constant of 6.9xlO"8 atm-mVmole (PHYSPROP, 1998). In addition, 1,3-diethylthiourea is not expected
to volatilize from dry soil given its estimated vapor pressure of 0.24 mm Hg (PHYSPROP, 1998; SRC, 1998).
D. Summary
If released to air, an estimated vapor pressure of 0.24 mm Hg at 25 ° C indicates that 1,3-diethylthiourea should exist
solely as a vapor in the ambient atmosphere. Gas-phase 1,3-diethylthiourea will be degraded in the atmosphere by
reaction with photochemically-produced hydroxyl radicals; the half-life for this reaction in air is estimated to be 4
hours. 1,3-Diethylthiourea is not expected to adsorb to suspended solids and sediments in water. An estimated BCF
of 2 suggests the potential for bioconcentration in aquatic organisms is low. If released to soil, 1,3-diethylthiourea is
expected to have very high mobility based upon an estimated Koc of 49, and, therefore, it has the potential to leach to
groundwater. Volatilization from water and from moist soil surfaces is not expected to be an important fate process
based upon a Henry's Law constant of 6.9xlO"8 atm-mVmole. Volatilization from dry soil surfaces is not expected to
occur based upon the vapor pressure of this compound.
C-2
-------
SUMMARY FOR 1,4-BUTENEDIOL
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of 1,4-butenediol are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF 1,4-BUTENEDIOL
Characteristic/Property Data Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Difrusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
110-64-5
2-butene-l,4-diol (mixed isomers)
C4H802
HOCH2CH=CHCH2OH
pale, yellow liquid
88.1
4 °C(cis);25 °C (trans)
235 °C (cis); 135 °C @ 12 mm Hg (trans)
soluble; estimated to be >lx!03 g/1
specific gravity = 1.07 @ 25 °C (liquid)
no data
8.6 (estimated)
-0.81
4.7x1 Q-3 mm Hg @ 25 °C (extrapolated)
no data
not flammable: flash point>100 °F
263 ° F (Cleveland open cup)
no data
no data
no data
1.54xlO"10 atm mVmole (estimated)
0.14 (estimated)
no data
Graijeetal. (1985)
Graijeetal. (1985)
Graijeetal. (1985)
Graijeetal. (1985)
Graijeetal. (1985)
Graijeetal. (1985)
Howard and Meylan (1997)
Howard and Meylan (1997)
Graije et al. (1985); SRC (1998)
Weiss (1986)
Lymanetal. (1990)
Hanschetal. (1995)
Graijeetal. (1985)
Cote (1997)
Flick (1991)
Meylan and Howard (1991)
Boethlingetal. (1994)
IL ENVIRONMENTAL FATE
A. Aquatic Fate
An estimated Koc of 8.6, determined from a log Kow of -0.81 (Hansch et al., 1995) and a regression-derived
equation (Lyman et al., 1990), indicates that 1,4-butenediol is not expected to adsorb to suspended solids and
sediment in water. Also, an estimated Henry's Law constant of 1.54xlO"10atmm3/mole at 25 ° C (Meylan and
Howard, 1991) indicates that 1,4-butenediol is not expected to volatilize from water surfaces (Lyman et al., 1990).
Hydrolysis is not expected to be an important fate process for 1,4-butenediol due to the lack of hydrolyzable
functional groups (Lyman et al., 1990). No data were available in the scientific literature for the biodegradation of
1,4-butenediol in aquatic media under aerobic or anaerobic conditions. However, using a structure estimation
C-3
-------
method (Boethling et al., 1994), aerobic biodegradation is expected to be rapid (days to weeks). According to a
classification scheme (Franke et al., 1994), an estimated BCF of 0.14 (Lyman et al., 1990), obtained from the log
Kow, suggests the potential for bioconcentration of 1,4-butenediol in aquatic organisms is low.
B. Atmospheric Fate
According to a model of gas/particle partitioning of semivolatile organic compounds in the atmosphere (Bidleman,
1988), 1,4-butenediol, which has an extrapolated vapor pressure of 4.7xlO~3 mm Hg at 25 ° C (Grafje et al., 1985), is
expected to exist solely as a gas in the ambient atmosphere. Gas-phase 1,4-butenediol is degraded in the atmosphere
by reaction with photochemically -produced hydroxyl radicals; the half-life for this reaction in air is estimated to be 5-
6 hours, depending upon the isomer (Meylan and Howard, 1993). The half-life for the reaction of 1,4-butenediol
with ozone in the atmosphere is estimated to be 1-2 hours, depending upon the isomer (Meylan and Howard, 1993).
1,4-Butenediol is not expected to directly photolyze in the atmosphere due to the lack of absorption in the
environmental UV spectrum greater than 290 nm (Lyman et al., 1990). Because 1,4-butenediol is miscible with
water, physical removal from the atmosphere by wet deposition may occur.
C. Terrestrial Fate
An estimated Koc of 8.6 (Lyman, 1990), determined from a log Kow of -0.81 (Hansch et al., 1995), indicates that
1,4-butenediol is expected to have very high mobility in soil (Swann et al., 1983). Volatilization of 1,4-butenediol
from moist soil surfaces is not expected to be important (Lyman et al., 1990) given an estimated Henry's Law
constant of 1.54xlO"10 arm mVmole (Meylan and Howard, 1991). In addition, an extrapolated vapor pressure of
4.7xlO"3 mm Hg (Grafje et al., 1985) indicates that 1,4-butenediol is not expected to volatilize from dry soil surfaces.
No data were available in the scientific literature for the biodegradation of 1,4-butenediol in soil under aerobic or
anaerobic conditions. However, using a structure estimation method (Boethling et al., 1994), aerobic biodegradation
is expected to be rapid (days to weeks).
D. Summary
1,4-Butenediol exists as a mixture of the cis and trans isomers that are expected to behave similarly in the
environment. If released to air, an extrapolated vapor pressure of 4.7xlO"3 mm Hg at 25 °C indicates 1,4-butenediol
should exist solely as a gas in the ambient atmosphere. Gas-phase 1,4-butenediol will be degraded in the atmosphere
by reaction with photochemically-produced hydroxyl radicals; the half-life for this reaction in air is estimated to be 5-
6 hours, depending upon the isomer. The gas phase reactions of 1,4-butenediol with photochemically produced
ozone corresponds to a half-life of 1-2 hours. Physical removal of gas-phase 1,4-butenediol from the atmosphere
may also occur via wet deposition processes based on the miscibility of this compound with water. If released to soil,
1,4-butenediol is expected to have very high mobility and is not expected to adsorb to soil surfaces. Volatilization
from water and moist soil surfaces is not expected to be an important fate process based upon an estimated Henry's
Law constant of 1.54xlO"10 arm mVmole. In addition, volatilization from dry soil surfaces is not expected to occur
based upon the vapor pressure of 1,4-butenediol. Biodegradation data were not available from the scientific
literature; however, a computer model estimates that aerobic biodegradation in both soil and water may occur within
days to weeks. In water, 1,4-butenediol is not expected to bioconcentrate in fish and aquatic organisms based on its
estimated BCF of 0.14.
C-4
-------
CHEMICAL SUMMARY FOR ACETIC ACID
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of acetic acid are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF ACETIC ACID
Characteristic/Property Data Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconcentration Factor
Odor Threshold
64-19-7
ethanoic acid; vinegar acid
C2H402
CH3COOH
clear liquid
60.05
16.7 °C
118 °C
Ixl03g/l, 25 °C
d25/25,1.049
no data
6.5-228
-0.17
15.7mmHg@25 °C
corrosive, particularly when dilute
flammable
103 °F(39 °C), closed cup
pKa = 4.76
no data
no data
1.00x10-'atmm3/mole@ 25 °C
<1 (calculated)
no data
Howard and Neal (1992)
Howard and Neal (1992)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
U.S. EPA (1981)
Budavarietal. (1996)
Sansoneetal. (1987)
Hanschetal. (1995)
Daubert and Danner (1985)
Weiss (1986)
Budavarietal. (1996)
Budavarietal. (1996)
Serjeant and Dempsey (1979)
Gaffneyetal. (1987)
Lymanetal. (1990)
IL ENVIRONMENTAL FATE
A. Aquatic Fate
The dominant environmental fate process for acetic acid in water is expected to be biodegradation. A large number of
biological screening studies have determined that acetic acid biodegrades readily under both aerobic (Zahn and
Wellens, 1980; Dore et al., 1975; Price et al., 1974; Placak and Ruchhoft, 1947 as cited in HSDB, 1998) and
anaerobic (Kameya et al., 1995; Mawson et al., 1991; Swindoll et al., 1988 as cited in HSDB, 1998) conditions.
Two aqueous adsorption studies found that acetic acid exists primarily in the water column and not in sediment
(Hemphill and Swanson, 1964; Gordon and Millero, 1985 as cited in HSDB, 1998). In general, organic ions are not
expected to volatilize from water to adsorb to paniculate matter in water to the degree that would be predicted for
C-5
-------
their neutral counterparts. Volatilization from the water column to the atmosphere is not expected to occur (Lyman
et al., 1990 as cited in HSDB, 1998) based on a Henry's Law constant of IxlO'9 atm-nfVmole at pH 7 (Gaffney et al.,
1987 as cited in HSDB, 1998). According to a classification scheme (Franke et al., 1994 as cited in HSDB, 1998),
an estimated BCF of <1 (Lyman, 1990 as cited in HSDB, 1998), calculated from a log Kow of -0.17 (Hansch et al.,
1995 as cited in HSDB, 1998), suggests that the potential for bioconcentration in aquatic organisms is low.
B. Atmospheric Fate
According to a model of gas/particle partitioning of semivolatile organic compounds in the atmosphere (Bidleman,
1988 as cited in HSDB, 1998), acetic acid, which has a vapor pressure of 15.7 mm Hg at 25 ° C (Daubert and Danner,
1989 as cited in HSDB, 1998), should exist solely as a gas in the ambient atmosphere. This is consistent with a study
in which over 91% of the total measured acetic acid in an air sample was found to be in the gas phase (Khwaja, 1995
as cited in HSDB, 1998). Acetic acid has been identified as one of the major sources of free acidity in precipitation
from remote regions of the world (Keene and Galloway, 1984 as cited in HSDB, 1998), indicating that physical
removal by wet deposition is an important fate process (Hartmann et al., 1989 as cited in HSDB, 1998). Another
important removal process of acetic acid from the atmosphere is reaction with photochemically-produced hydroxyl
radicals; the half-life for this reaction in air is estimated to be 22 days (Atkinson, 1989 as cited in HSDB, 1998).
Acetic acid has also been detected adsorbed to atmospheric paniculate material as the acetate (Gregory et al., 1986;
Khwaja, 1995 as cited in HSDB, 1998); the small amount of acetic acid associated with paniculate material may be
physically removed by wet and dry deposition (Grosjean, 1992).
C. Terrestrial Fate
The major environmental fate process for acetic acid in soil is expected to be biodegradation. A large number of
biological screening studies have determined that acetic acid biodegrades readily under both aerobic ,Zahn and
Wellens, 1980; Dore et al., 1975; Price et al., 1974; Placak and Ruchhoft, 1947 as cited in HSDB, 1998) and
anaerobic (Kameya et al., 1995; Mawson et al., 1991; Swindoll et al., 1988 as cited in HSDB, 1998) conditions.
Based on a classification scheme (Swann et al., 1983 as cited in HSDB, 1998), Koc values of 6.5 to 228 (Sansone et
al., 1987 as cited in HSDB, 1998) indicate that acetic acid is expected to have moderate to very high mobility in soil.
This is consistent with a study in which no sorption was reported for three different soils/sediments (Von Oepen et al.,
1991 as cited in HSDB, 1998). Volatilization of acetic acid from moist soil surfaces is not expected to be important
(Lyman et al., 1990, as cited in HSDB, 1998) given a Henry's Law constant of IxlO"9 atm-mVmole (Gaffney et al.,
1987 as cited in HSDB, 1998) and because acetic acid will exist predominantly as the acetate at environmental pH's.
However, the potential for volatilization of acetic acid from dry soil surfaces may exist based on it's vapor pressure
of 15.7 mm Hg (Daubert and Danner, 1989 as cited in HSDB, 1998). Volatilization will be attenuated depending
upon pH and the amount of acetic acid dissociated.
D. Summary
Acetic acid occurs throughout nature as a normal metabolite of both plants and animals. Consequently, acetic acid's
fate in the environment will, in part, be dependent on its participation in natural cycles. With a pKa of 4.76, acetic
acid and its conjugate base will exist in environmental media in varying proportions that are pH dependent; under
typical environmental conditions (pHs of 5 to 9), acetic acid will exist almost entirely in the ionized (dissociated)
form. If released to air, a vapor pressure of 15.7 mm Hg at 25 ° C indicates that acetic acid should exist solely as a
gas in the ambient atmosphere. Gas-phase acetic acid will be degraded in the atmosphere by reaction with
photochemically-produced hydroxyl radicals; the half-life for this reaction in air is estimated to be 22 days. Physical
removal of vapor-phase acetic acid from the atmosphere may occur via wet deposition processes based on its
miscibility with water. An estimated BCF of <1 suggests the potential for bioconcentration on aquatic organisms is
low. Adsorption studies indicate that acetic acid is not expected to adsorb to suspended solids and sediments in water.
If released to soil, acetic acid is expected to have very high to moderate mobility based upon measured Koc values
ranging from 6.5 to 228 and, therefore, it has the potential to leach to groundwater. If released to soil in high
concentrations, such as those encountered in a spill, acetic acid may travel through soil and reach groundwater.
Volatilization from water and from moist soil surfaces is not expected to be an important fate process based upon a
Henry's Law constant of IxlO"9 atm-mVmole. Yet, volatilization from dry soil surfaces may occur based upon the
vapor pressure of this compound. However, volatilization of acetic acid will be pH dependent; if acetic acid is
C-6
-------
dissociated, very little (about 1%) will be available for volatilization. Biodegradation is expected to be rapid and may
be the dominant fate process in both soil and water under non-spill conditions; a large number of biological screening
studies have determined that acetic acid biodegrades readily under both aerobic and anaerobic conditions.
C-7
-------
CHEMICAL SUMMARY FOR BRANCHED OCTYLPHENOL, ETHOXYLATED1
(alkylphenol polyethoxyethanol)
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources. The
search identified sources of information for branched octylphenol, ethoxylated.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of branched octylphenol, ethoxylated1 are summarized in Table
1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF BRANCHED
OCTYLPHENOL, ETHOXYLATED1
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
Data
9036-19-5,9002-93-1
Triton X-1001, OPIOSP
C14H220.(C2H40)100
(C8H17)C6H40(C2H40)100
Clear viscous liquid
polymer, >4000
7.2°C
271°C
Dispersible, >100g/L
d25, 1.07
>1
No data
No data
<0.001 torr
No data
No data
288°C
No data
No data
No data
No data
No data
No data
Reference
Howard and Neal (1992)
Howard and Neal (1992)
Howard and Neal (1992)
Howard and Neal (1992)
MSDS
Howard and Neal (1992)
MSDS
MSDS
MSDS
MSDS
MSDS
MSDS
MSDS
1 The properties are given for TritonXlOO (manufacturer Rohm and Haas).
C-8
-------
CHEMICAL SUMMARY FOR AMMONIUM CHLORIDE
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of ammonium chloride are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF AMMONIUM CHLORIDE
Characteristic/Property Data Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
12125-02-9
Ammonium muriate
C1FL.N
NFLC1
colorless cubic crystals
53.492
sublimes at 350°C
no data
approximately 300 g/L'
1.519g/cm3
no data
no data; expected to be < 10
no data; expected to be < 1
1.84X1Q-12 mm Hg at 25 °C (extrapolated)
no data
not flammable
not flammable
dissociates to NH4+ and Cl"
no data
no data
no data; expected to be < IxlO"8
no data
odorless
CAS (1998)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Lide(1995)
Lide(1995)
Lewis (1993)
Estimated
Lide(1995)
Estimated
Estimated
Daubert and Danner (1992)
Weiss (1986)
Weiss (1986)
Bodeketal. (1988)
Estimated
Weiss (1986)
1 Estimated from a reported solubility of 37 parts in 100 parts water at 20°C (Dean 1985).
IL ENVIRONMENTAL FATE
A. Aquatic Fate
If ammonium chloride is released into water, it is expected to dissociate into ammonium (NH4+) and chloride (Cl")
ions (Bodek et al., 1988). The counter ion associated with the NH4+ will vary depending on the concentration and
type of ions available and the pH in the receiving water. In addition, NH4+ and NH3 (ammonia) are in equilibrium in
the environment and since the pKa of the ammonium ion, NH4+, is 9.26, most ammonia in water is present as the
protonated form rather than as NH3 (Manahan, 1991). Ammonia is, however, present in the equilibrium and will
volatilize to the atmosphere (based upon its Henry's Law constant of 1.6X10"5 atm mVmole [Betterton, 1992 as cited
in PHYSPROP, 1998]); the rate of volatilization will increase with increasing pH and, to a lesser degree, temperature
C-9
-------
(ATSDR, 1990). In the aquatic environment, ammonium can undergo sequential transformation by the nitrification
and denitrification processes of the nitrogen cycle; within this process, ionic nitrogen compounds are formed
(ATSDR, 1990). In addition, ammonium can be taken up by aquatic plants as a source of nutrition, and the uptake of
ammonium by fish has also been documented (ATSDR, 1990). Adsorption of ammonium to sediment should
increase with increasing organic content, increased metal content, and decreasing pH; however, ammonium can be
produced in, and subsequently released from, sediment (ATSDR, 1990). The dissociation of ammonium chloride into
its component ions indicates that ammonium chloride is not expected to bioconcentrate in aquatic organisms.
Ammonium ions may be adsorbed by negatively charged surfaces of sediment in the water column, however
ammonium ions are expected to be replaced by other cations present in natural waters (Evans, 1989). The chloride
ion may complex with heavy metals, thereby increasing their solubility (Bodek et al., 1988). Adsorption of the
chloride ion to suspended solids and sediment in the water column is not expected to be an important fate process.
B. Atmospheric Fate
If ammonium chloride is released to the atmosphere, this compound's low vapor pressure (Daubert and Danner,
1992) indicates it will exist as a paniculate in the ambient atmosphere. Ammonium chloride is expected to undergo
wet deposition (ATSDR, 1990) in rain, snow, or fog based upon its high water solubility (Dean, 1985). Dry
deposition of ammonium chloride is expected to be an important fate process in the atmosphere (ATSDR, 1990). The
rate of dry deposition will depend on the prevailing winds and particle size (Bodek et al., 1988). In addition, NH4+
and NH3 (ammonia) are in equilibrium. The gas-phase reactions of ammonia with photochemically produced
hydroxyl radicals has been reported to be 1.6xlO~13 cmVmolc-sec, with a calculated half-life of approximately 100
days; this process contributes approximately 10% to the removal of atmospheric ammonia (ATSDR, 1990).
C. Terrestrial Fate
If ammonium chloride is released to soil, it is expected to dissociate into its component ions in moist soils. As noted
above, NH4+ and NH3 (ammonia) are in equilibrium in the environment and since the pKa of the ammonium ion,
NH4+, is 9.26, most ammonia in water is present as the protonated form rather than as NH3 (Manahan, 1991). The
low vapor pressure and Henry's Law constant expected for an ionic salt indicates that ammonium chloride will not
volatilize from either dry or moist soil surfaces. Nonetheless, ammonia is present in the equilibrium and will
volatilize to the atmosphere (based upon its Henry's Law constant of 1.6X10"5 atm mVmole [Betterton, 1992 as cited
in PHYSPROP, 1998]); the rate of volatilization will increase with increasing pH and, to a lesser degree, temperature
(ATSDR, 1990). The mobility of ammonium ions through soil may be attenuated by attraction to negatively charged
surfaces of soil particles, however ammonium ions are expected to be replaced by other cations present in soil (Evans,
1989). In soil, ammonium will serve as a source of nutrient taken up by plants and other organisms and converted to
organic-nitrogen compounds. Ammonium can be converted to nitrate by microbial populations through nitrification;
the nitrate formed will either leach through soil or be taken up by plants and other organisms. It has been determined
that minerals and dry soils can rapidly and effectively adsorb ammonia from air. Chloride is extremely mobile in soils
(Bodek et al., 1988). The chloride ion may complex with heavy metals, thereby increasing their solubility (Bodek et
al., 1988) and potential for leaching into groundwater.
D. Summary
If released into water, ammonium chloride is expected to dissociate into ammonium and chloride ions. The
dissociation of ammonium chloride into its component ions indicates that ammonium chloride is not expected to
bioconcentration in aquatic organisms. Ammonium, however, will be used as a nutrient source by microorganisms
and plants, and rapid uptake is anticipated. Ammonium is in equilibrium with ammonia, but the majority will be in
the ammonium form under most environmental pHs. When present, ammonia's Henry's Law constant indicates that
volatilization from water surfaces may occur. If released to soil, ammonium chloride is expected to dissociate into
its component ions in moist soils and will be used as a nutrient by microorganisms and plants. The dissociation of
ammonium chloride into its component ions in moist soils indicates that volatilization of ammonium from moist soil
surfaces is not expected to occur. The mobility of ammonium ions in soil is expected to be attenuated by cation
exchange processes. The low vapor pressure expected for an ionic salt indicates that ammonium chloride is not
expected to volatilize from dry soil surfaces, however, when ammonia is present in equilibrium, volatilization may
C-10
-------
occur. If released to the atmosphere, ammonium chloride's low vapor pressure indicates this compound will exist as
a paniculate. Wet and dry deposition will be the dominant fate processes in the atmosphere. The rate of dry
deposition will depend on the prevailing wind patterns and particle size. Some atmospheric oxidation may occur.
C-ll
-------
CHEMICAL SUMMARY FOR AMMONIUM HYDROXIDE
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of ammonium hydroxide are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF AMMONIUM
HYDROXIDE
Characteristic/Property Data Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Difiusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
1336-21-6
ammonia solution; aqua ammonia; ammonium hydrate
H5NO
NH.OH
colorless liquid
35.05
no data
no data
soluble in water
no data
no data
no data; estimated to be < 10
no data; estimated to be < 1
no data
incompatible w/ HC1, FINO3, Ag compounds
not flammable
no data; estimated to be > 350 °C
9.26 (water solution)
no data
no data
no data1
no data
no data
Lide(1995)
Lewis (1993)
PHYSPROP(1998)
Lide(1995)
Lewis (1993)
Lide(1995)
Sax (1984)
Estimated
Estimated
Sax (1984)
Weiss (1986)
Estimated
Manahan (1991)
1 In the environment, ammonium ion is expected to predominate in the ammonia-ammonium ion equilibrium; however, this equilibrium is
highly dependent on both pH and temperature (ATSDR, 1990). Ammonia is expected to have a very high Henry's Law constant, while
ammonium is expected to have a negligible Henry's Law constant (SRC, 1998).
IL ENVIRONMENTAL FATE
A. Aquatic Fate
If released into the water column at low concentrations, ammonia or ammonium hydroxide will volatilize to the
atmosphere; the rate of volatilization will increase with increasing pH and, to a lesser degree, temperature (ATSDR,
1990). Since the pKa of the ammonia is 9.26, most ammonia in most environmental waters is present as the
protonated, NH4+, form rather than as NH3 (Manahan, 1991). In the aquatic environment, ammonia can undergo
C-12
-------
sequential transformation by the nitrification and denitrification processes of the nitrogen cycle; within this process,
ionic nitrogen compounds are formed (ATSDR, 1990). In addition, ammonia can be taken up by aquatic plants as a
source of nutrition, and the uptake of ammonia by fish has also been documented (ATSDR, 1990). Adsorption of
ammonia to sediment should increase with increasing organic content, increased metal content, and decreasing pH;
however, ammonia can be produced in, and subsequently released from sediment (ATSDR, 1990). Large releases of
the concentrated base into water, such as may result from a spill, will result in an increase of the pH (ATSDR, 1990).
B. Atmospheric Fate
If ammonia is released to the atmosphere, its vapor pressure indicates it will exist as a vapor in the ambient
atmosphere. If ammonium hydroxide is released to the atmosphere, it is anticipated that the dominant form will be as
a paniculate, but during equilibrium between ammonium and ammonia, the ammonia will rapidly leave the particle as
a vapor. The dominant fate process for the removal of ammonia from the atmosphere is the reaction with acid air
pollutants to form ammonium compounds (e.g., ammonium sulfate, ammonium nitrate); these ammonium
compounds can then be removed by wet or dry deposition (ATSDR, 1990). In addition, gas-phase reactions of
ammonia with photochemically produced hydroxyl radicals has been reported to be 1.6xlO"13 cmVmolc-sec, with a
calculated half-life of approximately 100 days; this process contributes approximately 10% to the removal of
atmospheric ammonia (ATSDR, 1990).
C. Terrestrial Fate
If ammonia or ammonium hydroxide is released to soil, it will serve as a source of nutrient taken up by plants and
other organisms and converted to organic-nitrogen compounds. Ammonia can be converted to nitrate by microbial
populations through nitrification; the nitrate formed will either leach through soil or be taken up by plants and other
organisms. It has been determined that minerals and dry soils can rapidly and effectively adsorb ammonia from air.
Specifically, ammonia may be either bound to soil or undergo volatilization to the atmosphere. (ATSDR, 1990)
D. Summary
Ammonia is a base, and as such, the environmental fate of ammonia is pH and temperature dependent. If released into
water, ammonia and ammonium hydroxide will volatilize to the atmosphere, depending on the pH. At high pHs,
where the equilibrium more favors ammonia, volatilization will become increasingly important. At low pHs,
volatilization will be less important. Adsorption of ammonia to sediment and suspended organic material can be
important under proper conditions (i.e., organic matter content, metal content, and pH). In addition, ammonia will be
taken up by aquatic organisms and plants as a source of nutrition. The dominant fate of ammonia in water will be its
participation in the nitrogen cycle. The predominant removal process of ammonia and ammonium hydroxide from the
atmosphere is expected to be wet and dry deposition. To a lesser extent, reactions with photochemically-produced
hydroxyl radicals will occur. If released to soil, ammonia is expected to be taken up by plants and other organisms
and converted to organic-nitrogen compounds. These compounds will either be taken up by plants or other organisms
or leach through the soil. Volatilization of ammonia from soil surfaces is expected to occur.
C-13
-------
CHEMICAL SUMMARY FOR SODIUM CITRATE (citric acid)
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of sodium citrate are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF SODIUM CITRATE
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
KOC
LogKoW
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Henry's Law Constant
Molecular Diffusivity Coefficient
Air Diffusivity Coefficient
Fish Bioconcentration Factor
Odor Threshold
Conversion Factors
Data
68-04-2
trisodium citrate; sodium citrate anhydrous; 2-hydroxy- 1,2,3 -
propanetricarboxylic acid, trisodium salt
C6H5Na307
CH2(COONa)C(OH)(COONa)CH2COONa
dihydrate, white crystals, granules, or powder; pentahydrate,
relatively large, colorless crystals or white granules
258.07
150°C(-2H2O)
decomposed at red heat
72 g/100 mL at25°C (dihydrate)
1.9
no data
no data
no data
no data
0 (nonreactive, NFPA classification);
aqueous solution slightly acid to litmus
1 (slightly combustible, NFPA classification);
no data
no data
no data
no data
no data
no data
no data; odorless
no data
Reference
Lockheed Martin 1991
Budavari et al. 1989
Osol 1980
Budavari et al. 1989
Budavari et al. 1989
Fisher Scientific 1985
Lewis 1993
Weast 1983-1984
Fisher Scientific 1985
Lockheed Martin 1991
Osol, 1980
Lockheed Martin 1991
Lewis 1993
IL ENVIRONMENTAL FATE
A. Environmental Release
Sodium citrate is a solid with a cool, saline taste that is soluble in water (Fisher Scientific 1985). It is used in soft
drinks, frozen desserts, meat products, cheeses, and as a nutrient for cultured buttermilk; in photography; in
detergents; as a sequestrant and buffer; as an anticoagulant for blood withdrawn from the body; and in the removal of
sulfur dioxide from smelter waste gases (Lewis 1993). Medicinally, sodium citrate is used as expectorant and
C-14
-------
systemic alkalizer. Sodium citrate is a chelating agent and has been used to facilitate elimination of lead from the
body (Osol 1980).
No data were found on the environmental releases of sodium citrate. The chemical is not listed on U.S. EPA's Toxics
Release Inventory, requiring certain U.S. industries to report on chemical releases to the environment (TRI93 1995).
The chemical could potentially enter the environment when used for the removal of sulfur dioxide from smelter waste
gases.
B. Transport
No data were found on the environmental transport of sodium citrate in the secondary sources searched. Its water
solubility suggests that the sodium citrate would remain in the water phase.
C. Transformation/Persistence
No data were found on the transformation/persistence of potassium bisulfate in the secondary sources searched.
C-15
-------
CHEMICAL SUMMARY FOR CUPRIC SULFATE (copper ion)
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of cupric sulfate are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF CUPRIC SULFATE
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Reactivity
Data
7758-99-8
cupric sulfate pentahydrate; blue Vitriol
CuO4S-5H2O
CuSO4-5H2O
large, blue, triclinic crystals; blue powder
249.68
decomposes @ 110°C
decomposes to CuO @ 650°C
316g/L@0°C
2.286 g/cm3
no data
no data
no data
no data
reacts with Mg to produce Cu2O, MgSO4, and H2
reacts with NH4C1 producing (NH4)2SO4 and CuCl2;
Reference
Lide(1995)
Budavarietal. (1996)
ATSDR(1990)
Lide(1995)
Budavarietal. (1996)
Lide(1995)
Lide(1995)
ATSDR(1990)
Weastetal. (1985)
Lide(1995)
U.S. Air Force (1990)
HSDB(1998)
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
reacts with alkali (R)OH to produce Cu(OH)2 and
RSO4; reacts with excess aq. NH3 producing
Cu(NH3)22+ + OH";decomposition products include
SO2.
non-flammable HSDB(1998)
non-flammable HSDB(1998)
no data
no data
no data
no data
10-100 for copper; 30,000 for copper in oysters ATSDR (1990)
no data
IL ENVIRONMENTAL FATE
A. Aquatic Fate
Copper (Cu) commonly exists in three valence states, Cu° (metal), Cu+ (cuprous), and Cu2+ (cupric). It can also be
oxidized to a Cu3+ state, but Cu3+ ions are rapidly reduced to Cu2+ in the environment (ATSDR, 1990). Copper in
solution is present almost exclusively as the Cu2+ valence state (U.S. EPA, 1987). Copper in the Cu2+ valence state
C-16
-------
forms compounds and complexes with a variety of organic and inorganic ligands binding to -NH2, -SH, and, to a
lesser extent, -OH groups (ATSDR, 1990). The predominant form of copper in aqueous solution is dependent on the
pH of the solution. Below pH 6, the cupric ion (Cu2+) predominates; copper complexes with carbonate usually
predominate above pH 6 (U.S. EPA, 1987; ATSDR, 1990). The association of copper with organic or inorganic
ligands also depends on the pH and on the CaCO3 alkalinity. Most of the copper entering surface water is in the form
of paniculate matter, which settles out, precipitates, or adsorbs to organic matter, hydrous iron and manganese oxides,
and clay; however, the predominating form can change with the amount of rain, pH, content of runoff, and the
availability of ligands (ATSDR, 1990). The processes of complexation, adsorption and precipitation limit the
concentration of copper (Cu2+) to very low values in most natural waters (ATSDR, 1990). Calculations of the
bioconcentration factor in fish for copper have ranged from 10 to 100; however, the majority of copper
measurements in fish tissues under environmental conditions have indicated little, if any, bioconcentration. Filter
feeding shellfish, especially oysters, however, were found to significantly concentrate copper with bioconcentration
factors as high as 30,000 (ATSDR, 1990).
B. Atmospheric Fate
Most of the copper in the air is in the form of paniculate matter (dust) or is adsorbed to paniculate matter. Larger
particles (>5 um) are removed by gravitational settling, smaller particles are removed by other forms of dry and wet
deposition (ATSDR, 1990). Atmospheric copper resulting from combustion is associated with sub-micron particles
that can remain in the troposphere for an estimated 7-30 days and may be carried long distances (ATSDR, 1990).
C. Terrestrial Fate
Most of the copper deposited in the soil is strongly adsorbed primarily to organic matter, carbonate minerals, clay
minerals, and hydrous iron and manganese oxides. Movement through the soil is dependent on the presence of these
substances, the pH, and other physical and chemical parameters. The greatest potential for leaching is seen in sandy
soils with low pH (ATSDR, 1990). Laboratory experiments using controlled models and field experiments utilizing
core samples have shown that very little copper moves through the soil. Core samples showed that some movement
occurred as far as the 22.5-25 cm layer of soil, but little, if any, moved below this zone. The evidence indicates that
hazardous amounts of copper should not leach into groundwater from sludge, even from sandy soils (ATSDR, 1990).
D. Summary
Copper (Cu) commonly exists in three valence states, Cu° (metal), Cu+ (cuprous), and Cu2+ (cupric). It can also be
oxidized to a Cu3+ state, but there are no important industrial Cu3+ chemicals, and Cu3+ ions are rapidly reduced to
Cu2+ in the environment. If released to water, copper in solution will be present almost exclusively as the Cu2+
valence state. The predominant form of copper in aqueous solution is dependent on the pH of the solution. Most of
the copper entering surface water is in the form of paniculate matter; however, the predominating form can change
with the amount of rain, pH, content of runoff, and the availability of ligands. Copper in the Cu2+ valence state will
form compounds and complexes with a variety of organic and inorganic ligands. Calculations of the bioconcentration
factor in fish for copper have ranged from 10 to 100; however, the majority of copper measurements in fish tissues
under environmental conditions have indicated little, if any, bioconcentration. If released to soil, the majority of
copper deposited in the soil is strongly adsorbed. Movement through the soil is dependent on the presence of organic
matter, carbonate minerals, clay minerals, hydrous iron and manganese oxides, the pH, and other physical and
chemical parameters. The greatest potential for leaching is seen in sandy soils with low pH. If released into the
atmosphere, copper is expected to exist as a dust paniculate or adsorb to paniculate matter. Studies have shown that
copper can remain in the atmosphere up to 30 days and be carried long distances.
C-17
-------
CHEMICAL SUMMARY FOR CUPRIC ACETATE (copper sulfate pentahydrate)
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of cupric acetate are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF COPPER ACETATE
Characteristic/Property Data Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Difrusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
6046-93-1
copper (II) acetate monohydrate
(CH3CO2)2Cu-H2O
Cu(C2H3O2 )2-H2O
dark, green monoclinic crystals
greenish-blue, fine powder
199.65
115 °C
decomposes at 240 °C
72 g/L cold water; 200 g/L hot water
1.88g/cm3
no data
no data
no data
no data
stable
not flammable
not flammable
no data
no data
no data
no data
10-100 for copper; 30,000 for copper in oysters
no data
Lide(1995)
Lide(1995)
Aldrich(1996)
Lide(1995)
Budavarietal. (1996)
Lewis (1993)
Lide(1995)
Lide(1995)
Lide(1995)
Weastetal. (1985)
Lide(1995)
Weiss (1986)
Weiss (1986)
Weiss (1986)
ATSDR(1990)
IL ENVIRONMENTAL FATE
A. Aquatic Fate
Copper (Cu) commonly exists in three valence states, Cu° (metal), Cu+ (cuprous), and Cu2+ (cupric). It can also be
oxidized to a Cu3+ state, but Cu3+ ions are rapidly reduced to Cu2+ in the environment (ATSDR, 1990). Copper in
solution is present almost exclusively as the Cu2+ valence state (U.S. EPA, 1987). Copper in the Cu2+ valence state
forms compounds and complexes with a variety of organic and inorganic ligands binding to -NH2, -SH, and, to a
lesser extent, -OH groups (ATSDR, 1990). The predominant form of copper in aqueous solution is dependent on the
pH of the solution. Below pH 6, the cupric ion (Cu2+) predominates; copper complexes with carbonate usually
C-18
-------
predominate above pH 6 (U.S. EPA, 1987; ATSDR, 1990). The association of copper with organic or inorganic
ligands also depends on the pH and on the CaCO3 alkalinity. Most of the copper entering surface water is in the form
of paniculate matter, which settles out, precipitates, or adsorbs to organic matter, hydrous iron and manganese oxides,
and clay; however, the predominating form can change with the amount of rain, pH, content of runoff, and the
availability of ligands (ATSDR, 1990). The processes of complexation, adsorption and precipitation limit the
concentration of copper (Cu2+) to very low values in most natural waters (ATSDR, 1990). Calculations of the
bioconcentration factor in fish for copper have ranged from 10 to 100; however, the majority of copper
measurements in fish tissues under environmental conditions have indicated little, if any, bioconcentration. Filter
feeding shellfish, especially oysters, however, were found to significantly concentrate copper with bioconcentration
factors as high as 30,000 (ATSDR, 1990).
B. Atmospheric Fate
Most of the copper in the air is in the form of paniculate matter (dust) or is adsorbed to paniculate matter. Larger
particles (>5 um) are removed by gravitational settling, smaller particles are removed by other forms of dry and wet
deposition (ATSDR, 1990). Atmospheric copper resulting from combustion is associated with sub-micron particles
that can remain in the troposphere for an estimated 7-30 days and may be carried long distances (ATSDR, 1990).
C. Terrestrial Fate
Most of the copper deposited in the soil is strongly adsorbed primarily to organic matter, carbonate minerals, clay
minerals, and hydrous iron and manganese oxides. Movement through the soil is dependent on the presence of these
substances, the pH, and other physical and chemical parameters. The greatest potential for leaching is seen in sandy
soils with low pH (ATSDR, 1990). Laboratory experiments using controlled models and field experiments utilizing
core samples have shown that very little copper moves through the soil. Core samples showed that some movement
occurred as far as the 22.5-25 cm layer of soil, but little, if any, moved below this zone. The evidence indicates that
hazardous amounts of copper should not leach into groundwater from sludge, even from sandy soils (ATSDR, 1990).
D. Summary
Copper (Cu) commonly exists in three valence states, Cu° (metal), Cu+ (cuprous), and Cu2+ (cupric). It can also be
oxidized to a Cu3+ state, but there are no important industrial Cu3+ chemicals, and Cu3+ ions are rapidly reduced to
Cu2+ in the environment. If released to water, copper in solution will be present almost exclusively as the Cu2+
valence state. The predominant form of copper in aqueous solution is dependent on the pH of the solution. Most of
the copper entering surface water is in the form of paniculate matter; however, the predominating form can change
with the amount of rain, pH, content of runoff, and the availability of ligands. Copper in the Cu2+ valence state will
form compounds and complexes with a variety of organic and inorganic ligands. Calculations of the bioconcentration
factor in fish for copper have ranged from 10 to 100; however, the majority of copper measurements in fish tissues
under environmental conditions have indicated little, if any, bioconcentration. If released to soil, the majority of
copper deposited in the soil is strongly adsorbed. Movement through the soil is dependent on the presence of organic
matter, carbonate minerals, clay minerals, hydrous iron and manganese oxides, the pH, and other physical and
chemical parameters. The greatest potential for leaching is seen in sandy soils with low pH. If released into the
atmosphere, copper is expected to exist as a dust paniculate or adsorb to paniculate matter. Studies have shown that
copper can remain in the atmosphere up to 30 days and be carried long distances.
C-19
-------
CHEMICAL SUMMARY FOR ETHYLENEDIAMINE
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of ethylene diamine are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF ETHYLENE DIAMTNE
Characteristic/Property Data Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconcentration Factor
Odor Threshold
107-15-3
1,2-diamineethane; 1,2-ethanediamine
C2H8N2
H2NCH2CH2NH2
colorless, clear, thick, liquid
60.10
8.5 °C
116-117 °C
Ixl03g/l@25 °C
d25'4, 0.898
no data
2 (calculated)
-2.04
12.0 mm Hg @ 25 °C
volatile w/ steam; absorbs CO2 from air
flammable
110 °F(43 °C), closed cup
pKaj = 9.92;pKa2 = 6.86
no data
no data
1.73x10-'atmm3/mole@ 25 °C
0.02 (calculated)
100% recognizable @ 11.2 ppm
Howard and Neal (1992)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Riddicketal. (1986)
Budavarietal. (1996)
Lymanetal. (1990)
Hanschetal. (1995)
Boubliketal. (1984)
Budavarietal. (1996)
Aldrich(1997)
Budavarietal. (1996)
Perrin(1972)
Hine and Mookerjee (1975)
Lymanetal. (1990)
Verschueren(1996)
IL ENVIRONMENTAL FATE
A Aquatic Fate
The dominant environmental fate process for ethylenediamine in surface water is expected to be biodegradation. A
number of biological screening studies have determined that ethylenediamine biodegrades readily under aerobic
conditions (Price et al., 1974; Takemoto et al., 1981; Fitter, 1976 ; Mills and Stack, 1955, as cited in HSDB, 1998).
No data were available for the biodegradation of ethylenediamine under anaerobic conditions. An estimated Koc
value of 2, determined from an experimental log Kow of -2.04 (Hansch et al., 1995) and a regression-derived
equation (Lyman et al., 1990), indicates that ethylenediamine is not expected to adsorb to suspended solids and
sediment in water. In general, organic ions are not expected to volatilize from water or adsorb to paniculate matter in
water to the degree that would be predicted for their neutral counterparts. Based on an estimated BCF of 0.02
C-20
-------
(Lyman et al., 1990) calculated from the log Kow, a classification scheme (Franke et al., 1994) suggests the potential
forbioconcentration in aquatic organisms is low. Ethylenediamine is not expected to volatilize from water surfaces
(Lyman et al., 1990) based upon an experimental Henry's Law constant of 1.73xlO"9 atm-mVmole (Hine and
Mookerjee, 1975). However, volatilization of ethylenediamine will be pH dependent and attenuated if it is
protonated; very little, about 1%, will be available for volatilization. Hydrolysis of ethylenediamine is not expected
to occur due to the lack of hydrolyzable functional groups (Lyman et al., 1990).
B. Atmospheric Fate
According to a model of gas/particle partitioning of semivolatile organic compounds in the atmosphere (Bidleman,
1988), ethylenediamine, which has a vapor pressure of 12 mm Hg at 25 ° C (Boublik et al, 1984), should exist solely
as a gas in the ambient atmosphere. Gas-phase ethylenediamine is degraded in the atmosphere by reaction with
photochemically-produced hydroxyl radicals; the half-life for this reaction in air is estimated to be 6 hours (Meylan
and Howard, 1993). Due to its miscibility with water, ethylenediamine may also be removed physically from the
atmosphere by wet deposition. Ethylenediamine is not expected to directly photolyze in the atmosphere due to the
lack of absorption in the environmental UV spectrum (>290 nm) (Lyman et al., 1990).
C. Terrestrial Fate
The major environmental fate process for ethylenediamine in aerobic soils is expected to be biodegradation. A
number of biological screening studies have determined that ethylenediamine biodegrades readily under aerobic
conditions (Price et al., 1974; Takemoto et al., 1981; Fitter, 1976 ; Mills and Stack, 1955, as cited in HSDB, 1998).
No data on the biodegradation of ethylenediamine under anaerobic conditions were located in the available literature.
An estimated Koc value of 2 (Lyman et al., 1990), determined from an experimental log Kow of -2.04 (Hansch et al.,
1995), indicates that ethylenediamine is expected to have very high mobility in soil (Swann et al., 1983).
Volatilization of ethylenediamine from moist soil surfaces is not expected to be important (Lyman et al., 1990) given
an experimental Henry's Law constant of 1.73xlO"9 atm-mVmole (Hine and Mookerjee, 1975), although it may
volatilize from dry soil surfaces based upon a vapor pressure of 12 mm Hg (Boublik et al., 1984). However, at
environmental pH's of 5-7, ethylenediamine will most likely be a salt and volatilization will be attenuated.
D. Summary
The dominant removal mechanisms of ethylenediamine from the environment are expected to be biodegradation on
the earth's surface and reaction with photochemically-produced hydroxyl radicals in the atmosphere. In both soil and
water, biodegradation is expected to be rapid; a large number of biological screening studies have determined that
ethylenediamine biodegrades readily under aerobic conditions. If released to air, a vapor pressure of 12 mm Hg
indicates ethylenediamine should exist solely as a gas in the ambient atmosphere. Gas-phase ethylenediamine will be
degraded in the atmosphere by reaction with photochemically-produced hydroxyl radicals; the half-life for this
reaction in air is estimated to be 6 hours. Physical removal of gas-phase ethylenediamine from the atmosphere may
also occur via wet deposition processes based on the miscibility of this compound with water. With a pKaj of 9.92,
ethylenediamine and its conjugate acid will exist in environmental media in varying proportions that are pH
dependent. If released to soil, ethylenediamine may display very high mobility based upon an estimated Koc of 2. If
released to soil in high concentrations, such as those encountered in a spill, ethylenediamine may travel through soil
and reach groundwater. Volatilization of ethylenediamine from water and moist soil surfaces is not expected to be an
important fate process based upon a Henry's Law constant of 1.73xlO"9 atm-mVmole, although its vapor pressure
indicates that volatilization from dry soil surfaces may occur. However, at environmental pH's of 5-7,
ethylenediamine will most likely be a salt and volatilization will be attenuated. In water, ethylenediamine is not
expected to bioconcentrate in fish and aquatic organisms based on an estimated BCF of 0.02.
C-21
-------
CHEMICAL SUMMARY FOR ETHYLENE GLYCOL
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of ethylene glycol are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF ETHYLENE GLYCOL
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Difrusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
Data
107-21-1
1,2-ethanediol
C2H602
HOCH2CH2OH
slightly viscous liquid
62.07
-13 °C
197.6 °C
miscible (1,000 g/1)
1.11 g/cm3
2.1
4 (estimated)
-1.36
0.092 mm Hg
no data
combustible
240 °F (115 °C)
15.1
no data
no data
6.0xlO"8atmm3/mol
10
25 ppm
Reference
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Riddicketal(1986)
Budavarietal. (1996)
Verschueren(1996)
SRC (1998)
Hansch et al. (1995), as cited in HSDB
(1998)
Daubert and Danner (1989)
no data
Lewis (1993)
Budavarietal. (1996)
Howard and Meylan (1997)
Howard and Meylan (1997)
HSDB (1998)
ECDIN (1998)
IL ENVIRONMENTAL FATE
A. Aquatic Fate
The dominant environmental fate process for ethylene glycol in water is expected to be biodegradation. A large
number of biological screening studies have determined that ethylene glycol biodegrades readily under both aerobic
and anaerobic conditions (Bridie et al. 1979; Fitter 1976; and Price et al. 1974, as cited in HSDB, 1998). Aerobic
degradation is essentially complete in
-------
HSDB, 1998) and a regression-derived equation (Lyman et al., 1990, as cited in HSDB, 1998). Volatilization from
the water column to the atmosphere is not expected to occur (Lyman et al., 1990, as cited in HSDB, 1998) based on a
Henry's Law constant of 6.0xlO"8 atm-mVmole (Butler and Ramchandani 1935, as cited in Howard and Meylan
1997). Ethylene glycol is expected to be stable to hydrolysis in the pH range of 5-9 typically encountered in the
environment (Lyman et al., 1990, as cited in HSDB, 1998). According to a classification scheme (Franke et al.,
1994), a BCF of 10 in golden ide fish (Freitag et al., 1985, as cited in HSDB, 1998) suggests that the potential for
bioconcentration in aquatic organisms is low.
B. Atmospheric Fate
According to a model of gas/particle partitioning of semivolatile organic compounds in the atmosphere (Bidleman
1989, as cited in HSDB, 1998), ethylene glycol, which has a vapor pressure of 0.092 mm Hg at 25 ° C (Daubert and
Danner 1989), should exist solely as a gas in the ambient atmosphere. Nonetheless, ethylene glycol has been detected
adsorbed onto atmospheric paniculate material (Abdelghani et al., 1990, as cited in HSDB, 1998); the small amount
of ethylene glycol deposited onto paniculate material may be physically removed by wet and dry deposition. The
predominant removal process of ethylene glycol from the atmosphere is reaction with photochemically-produced
hydroxyl radicals; the half-life for this reaction in air is estimated to be 50 hours (Atkinson 1989, as cited in HSDB,
1998). Ethylene glycol may undergo some degradation by direct photolysis; 12.1% of applied ethylene glycol was
degraded after 17 hours following irradiation by light > 290 nm (Freitag et al., 1985, as cited in HSDB, 1998).
C. Terrestrial Fate
The major environmental fate process for ethylene glycol in soil is expected to be biodegradation. A large number of
biological screening studies have determined that ethylene glycol biodegrades readily under both aerobic and
anaerobic conditions; complete biodegradation was shown in one soil within 2 days and 97% biodegradation in 12
days was reported for a second soil (McGahey and Bower 1992, as cited in HSDB, 1998). Based on a classification
scheme (Swannetal., 1983, as cited in HSDB, 1998), an estimated Koc of 4, determined from a log Kow of-1.36
(Hansch et al., 1995, as cited in HSDB, 1998) and a regression-derived equation (Lyman, 1990 et al., as cited in
HSDB, 1998), indicates that ethylene glycol is expected to have very high mobility in soil. Percent adsorption to 4
soils (2 clay and 2 sandy clay soils) ranged from 0-0.5% (Abdelghani et al 1990, as cited in HSDB, 1998).
Volatilization of ethylene glycol from moist soil surfaces is not expected to be important (Lyman et al., 1990, as cited
in HSDB, 1998) given a Henry's Law constant of 6.0xlO"8 atm-mVmole (Butler and Ramchandani 1935, as cited in
Howard and Meylan 1997). Ethylene glycol may volatilize from dry soil given its vapor pressure of 0.092 mmHg
(Daubert and Danner, 1989); this may be attenuated by hydrogen bonding to soil materials (SRC, 1998).
D. Summary
If released to air, a vapor pressure of 0.092 mm Hg at 25 ° C indicates that ethylene glycol should exist solely as a gas
in the ambient atmosphere; however, experimental results show that at least some ethylene glycol is associated with
atmospheric particulates. Gas-phase ethylene glycol will be degraded in the atmosphere by reaction with
photochemically-produced hydroxyl radicals; the half-life for this reaction in air is estimated to be 50 hours.
Adsorption studies indicate that ethylene glycol is not expected to adsorb to suspended solids and sediments in water.
A BCF of 10 in golden ide fish suggests the potential for bioconcentration in aquatic organisms is low. If released to
soil, ethylene glycol is expected to have very high mobility based upon an estimated Koc of 4, and, therefore, it has the
potential to leach to groundwater. Volatilization from water and from moist soil surfaces is not expected to be an
important fate process based upon a Henry's Law constant of 6.0xlO"8 atm-mVmole. Volatilization from dry soil
surfaces may occur based upon the vapor pressure of this compound, although this may be attenuated by hydrogen
bonding to soil materials. Biodegradation is expected to be rapid and may be the dominant fate process in both soil
and water under non-spill conditions; a large number of biological screening studies have determined that ethylene
glycol biodegrades readily under both aerobic and anaerobic conditions.
C-23
-------
CHEMICAL SUMMARY FOR ETHYLENE GLYCOL MONOBUTYL ETHER
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources. The
search identified sources of information for ethylene glycol monobutyl ether.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of ethylene glycol monobutyl ether are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF ETHYLENE GLYCOL
MONOBUTYL ETHER
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Difrusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
Data
111-76-2
BUCS, butoxyethanol, Dowanol EB
C6H14O2
CH3(CH2)3OCH2CH2OH
Clear, colorless liquid
118.18
-70°C
171°C, 743mmHg
>1000g/L, 25°C
d20'20, 0.9012
4.07
1
0.83
0.88 mm Hg @ 25°C
Inert
Combustible
60°C
No data
No data
No data
2.08x1 0-8atmm3/mol
No data
No data
Reference
Howard and Neal (1992)
Howard and Neal (1992)
Howard and Neal (1992)
Howard and Neal (1992)
HSDB(1998)
Howard and Neal (1992)
Budavarietal. (1996)
Budavarietal. (1996)
HSDB(1998)
HSDB(1998)
HSDB(1998)
EPI
Howard and Meylan (1997)
Howard and Meylan (1997)
Sax and Lewis (1987)
Sax and Lewis (1987)
HSDB(1998)
Howard and Meylan (1997)
C-24
-------
CHEMICAL SUMMARY FOR FLUOROBORIC ACID (fluoride)
This chemical was identified by one or more suppliers as a bath ingredient for the electroless copper and tin-palladium
processes. This summary is based on information retrieved from a systematic search limited to secondary sources.
The only exception is summaries of studies from unpublished TSCA submissions that may have been included. These
sources include online databases, unpublished EPA information, government publications, review documents, and
standard reference materials. No attempt has been made to verify information in these databases and secondary
sources. Very little information on the environmental fate and toxicity of fluoroboric acid or fluoroborates was
found in the available secondary sources. Supplemental information is provided for fluoride which may be a
degradation product and for sodium bifluoride.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of fluoroboric acid are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF FLUOROBORIC ACID
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
KOC
LogKoW
Vapor Pressure
Vapor Density
Reactivity
Flammability
Flash Point
Dissociation Constant (-pK)
Henry's Law Constant
Molecular Diffusivity Coefficient
Air Diffusivity Coefficient
Fish Bioconcentration Factor
Odor Threshold
Conversion Factors
Data
16872-11-0
hydrogen tetrafluoroborate
fluoboric acid
hydrofluoroboric acid
HBF4
B-F4-H
colorless liquid
87.82
-90 °C
130°C (decomposes)
miscible;
sol. in hot water
~1.84g/mL
NA
NA
5.1mmHgat20°C
3.0
strong acid; corrosive
NA
NA
-4.9
NA
NA
NA
NA
NA
Na
Reference
HSDB(1995)
HSDB(1995)
HSDB(1995)
Fisher Scientific
HSDB(1995)
HSDB(1995)
Fisher Scientific
HSDB(1995)
HSDB(1995)
HSDB(1995)
Fisher Scientific
Fisher Scientific
HSDB(1995)
HSDB(1995)
(1993)
(1993)
(1993)
(1993)
The chemical identity and physical/chemical properties of sodium tetrafluoroborate are summarized in Table 2.
C-25
-------
TABLE 2. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF SODIUM
TETRAFLUOROBORATE
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
KOC
LogKoW
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant (-pK)
Henry's Law Constant
Molecular Diffusivity Coefficient
Air Diffusivity Coefficient
Fish Bioconcentration Factor
Odor Threshold
Conversion Factors
Data
013755-29-8
sodium fluoroborate
STB
sodium borfluoride
sodium boron tetrafluoride
NaNF4
Na-F4-B
white crystalline powder
109.82
384°C
108g/100mLat26°C
210g/100mLatlOO °C
2.470
NA
NA
NA
reacts with strong oxidizing
agents; sensitive to moisture
noncombustible
NA
NA
NA
NA
NA
NA
Reference
Lockheed Martin (1994)
Lockheed Martin (1994)
Sigma- Aldrich (1992)
Budavarietal. (1989)
Budavarietal. (1989)
Budavarietal. (1989)
Sigma- Aldrich (1992)
Sigma- Aldrich (1992)
Lockheed Martin (1994)
The chemical identity and physical/chemical properties of sodium fluoride are summarized in Table 3.
C-26
-------
TABLE 3. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF SODIUM FLUORIDE
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
KOC
LogKoW
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant (-pK)
Henry's Law Constant
Molecular Diffusivity Coefficient
Air Diffusivity Coefficient
Fish Bioconcentration Factor
Odor Threshold
Conversion Factors
Data
7681-49-4
sodium hydrofluoride
sodium monfluoride
floridine
NaF
Na-F
crystals
42.00
993 °C
1704°C
4.0g/100mLat 15°C
4.3 g/100mLat25 °C
2.78
NA
NA
lmmHgatl077°C
stable under normal conditions
nonflammable
Reference
Budavarietal. (1989)
Budavarietal. (1989)
Budavarietal. (1989)
Budavarietal. (1989)
Budavarietal. (1989)
Budavarietal. (1989)
Keith and Walters (1985)
Keith and Walters (1985)
Keith and Walters (1985)
The chemical identity and physical/chemical properties of sodium bifluoride are summarized in Table 4.
C-27
-------
TABLE 4. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF SODIUM BIFLUORIDE
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
KOC
LogKoW
Vapor Pressure
Vapor Density
Reactivity
Flammability
Flash Point
Dissociation Constant (-pK)
Henry's Law Constant
Molecular Diffusivity Coefficient
Air Diffusivity Coefficient
Fish Bioconcentration Factor
Odor Threshold
Conversion Factors
Data
1333-83-1
sodium hydrogen difluoride
sodium hydrogen fluoride
sodium acid fluoride
NaHF2
F2-H-Na
white, crystalline powder
62.01
decomposes on heating
NA
soluble in cold and hot water
2.08
NA
NA
NA
NA
aqueous solution corrodes glass
slightly combustible
NA
NA
NA
NA
NA
NA
NA
NA
Reference
HSDB(1995)
HSDB(1995)
Lewis (1993)
HSDB(1995)
Budavarietal. (1989)
Budavarietal. (1989)
Lewis (1993)
Lide(1991)
Lewis (1993)
Budavarietal. (1989)
Lockheed Martin (1990)
IL ENVIRONMENTAL FATE
A.
Environmental Release
Fluoroboric acid may be released into the environment in emissions and effluents from facilities involved in its
manufacture or use. It is used primarily in industrial metal plating solutions (60%), in the synthesis of diazo salts
(20%), and in metal finishing (20%) (HSDB 1995). It is used in bright dipping solutions for Sn-Pb alloys in printed
circuits and other electrical components (HSDB 1995).
B.
Transport
No information was found in the available secondary sources on the environmental transport of fluoroboric acid. Its
miscibility with water indicates that transport in aqueous systems is very likely.
C-28
-------
C. Transformation/Persistence
FLUOROBORIC ACID:
1. Air — No information was found in the available secondary sources on the transformation and persistence
of fluoroboric acid or fluoroborates in the atmosphere.
2. Soil — No information was found in the available secondary sources on the transformation and persistence
of fluoroboric acid or fluoroborates in soil. Fluoroboric acid may undergo limited hydrolysis in moist soils
(Budavari et al. 1989).
3. Water — Fluoroboric acid undergoes limited hydrolysis in water to form hydroxyfluoroborate ions, the
major product is BF3OH- (Budavari et al. 1989).
4. Biota — No information was found in the available secondary sources on the biotransformation or
bioconcentration of fluoroboric acid or fluoroborates. Rapid urinary excretion of tetrafluoroborates
suggests that these salts would not bioaccumulate.
FLUORIDES:
1. Air — Gaseous inorganic fluorides undergo hydrolysis in the atmosphere; however, paniculate forms are
relatively stable and do not hydrolyze readily (ATSDR 1993).
2. Soil — Fluorides tend to persist in soils as fluorosilicate complexes under acidic conditions and as calcium
fluoride under alkaline conditions. Sandy acidic soils favor the formation of soluble forms (ATSDR 1993).
3. Water — In dilute solutions and at neutral pH, fluoride is generally present as dissolved fluoride ion. High
calcium carbonate levels may lead to precipitation as calcium fluoride (ATSDR 1993).
4. Biota — Fluorides have been shown to accumulate in some aquatic organisms (ATSDR 1993). Soluble
forms of fluoride are taken up by terrestrial plants and converted into fluoro-organic compounds (ATSDR
1993).
C-29
-------
CHEMICAL SUMMARY FOR HYDROCHLORIC ACID
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of hydrochloric acid are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF HYDROCHLORIC ACID
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
Data
7647-01-0
muriatic acid
HC1
HC1
filming liquid
36.46
-25.4 °C (39.17% soln)
108. 58 °C at 760 mm Hg
479.1 g/l(40%soln)
1.20 g/cm3 (39. 11% soln)
1.639g/l
expected to be < 50
expected to be < 1
no data
toxic, corrosive fumes w/H2O or steam
non-combustible
no data
~ 3
no data
no data
no data
no data
no data
Reference
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Lewis (1993)
Lide(1995)
Budavarietal. (1996)
Budavarietal. (1996)
Weastetal. (1985)
Budavarietal. (1996)
Austin and Glowacki (1989)
SRC (1998)
SRC (1998)
Sax (1984)
Lewis (1993)
Bodeketal. (1988)
IL ENVIRONMENTAL FATE
A. Aquatic Fate
If hydrochloric acid is released into the water column at low concentrations, a pKa of- -3.00 (Bodek et al., 1988)
indicates it will dissociate completely into chloride (Cl~) and hydrogen (H+) ions. The amount of gaseous
hydrochloric acid dissolved in water is affected by the pH of the solution. A higher pH allows more aqueous
hydrochloric acid to dissociate, thereby increasing the solubility of hydrochloric acid gas (Bodek et al., 1988). As a
result, dilute solutions of hydrochloric acid are not expected to volatilize from water surfaces or to bioconcentrate in
aquatic organisms. Chloride ions generally do not react with many species in water and are harmless at relatively low
concentrations (Manahan, 1991). Hydrochloric acid will protonate amines and other electron pair donators present in
C-30
-------
natural waters, forming salts; this will be dependent upon pH. Large releases of the concentrated acid into water, such
as may result from a spill, will result in a lowering of the pH (Bodek et al., 1988).
B. Atmospheric Fate
If hydrochloric acid is released to the atmosphere, its vapor pressure indicates it will exist as a vapor in the ambient
atmosphere. Wet deposition of hydrochloric acid in rain, snow, or fog is expected to be the dominant fate process in
the atmosphere based upon its high water solubility (Arimoto, 1989).
C. Terrestrial Fate
If hydrochloric acid is released to soil, it will dissociate into chloride and hydrogen ions in moist soils. Hydrochloric
acid will protonate amines and other electron pair donators present in soils, forming salts; this will be dependent upon
pH. The chloride ion is extremely mobile in soils and almost no soil retention occurs (Bodek et al., 1988). Chloride
is typically the predominant ion in saline soils and the second most abundant anion in sodic soils; thus, it is readily
available for the formation of metal complexes in soil (Bodek et al., 1988; SRC, 1998).
D. Summary
If released into water, hydrochloric acid will dissociate into chloride (Cl~) and hydrogen (H+) ions. Therefore,
hydrochloric acid is not expected to adsorb to suspended solids or sediment in the water column, bioconcentrate in
aquatic organisms, nor volatilize from water surfaces. Chloride ions generally do not react with many species in
water and are harmless at relatively low concentrations. Hydrochloric acid will protonate amines and other electron
pair donators present in natural waters and soils, forming salts; this will be dependent upon pH. If released to soil,
hydrochloric acid is expected to dissociate into its component ions in moist soils. Because the chloride ion is
extremely mobile in soils, almost no soil retention occurs. Chloride is typically the predominant ion in saline soils
and the second most abundant anion in sodic soils; thus, it is readily available for the formation of metal complexes in
soil. Volatilization of hydrochloric acid from soil surfaces is not expected to occur. If released to the atmosphere,
hydrochloric acid is expected to exist as a gas. Hydrochloric acid is expected to be physically removed from the
atmosphere by wet deposition based upon its high water solubility.
C-31
-------
CHEMICAL SUMMARY FOR HYDROGEN PEROXIDE
This chemical was identified by one or more suppliers as a bath ingredient for the electroless copper, non-
formaldehyde electroless copper, and tin-palladium processes. This summary is based on information retrieved from
a systematic search limited to secondary sources (see Attachment C-l). These sources include online databases,
unpublished EPA information, government publications, review documents, and standard reference materials. No
attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of hydrogen peroxide are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF HYDROGEN PEROXIDE
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
KOC
LogKoW
Vapor Pressure
Reactivity
Data
7722-84-1
hydrogen dioxide; hydroperoxide; albone; hioxyl
H202
H202
colorless, unstable liquid, bitter taste
34.02
-0.43°C
152°C
miscible
1.463 @0°C
no data
no data
no data
1 .97 mm Hg @ 25 ° C (measured)
strong oxidizer; may decompose violently if traces of
Reference
Budavari et al. 1989
Budavari et al. 1989
IARC 1985
Budavari et al. 1989
Budavari et al. 1989
Budavari et al. 1989
Budavari et al. 1989
Budavari et al. 1989
Budavari et al. 1989
CHEMFATE 1995
Budavari et al. 1989
Flammability
impurities are present
molecular additions, substitutions, oxidations, IARC 1985
reduction; can form free radicals
not flammable, but can cause spontaneous combustion HSDB 1995
of flammable materials
Flash Point
Dissociation Constant
Henry's Law Constant
Molecular Diffusivity Coefficient
Air Diffusivity Coefficient
Fish Bioconcentration Factor
Odor Threshold
Conversion Factors
no data
no data
no data
no data
no data
no data
odorless
1 ppm = 1.39 mg/m3
1 mg/m3 = 0.72 ppm
30%solnl.lkg/L
anhydrous 1 .46 kg/L
Budavari et al. 1989
IARC 1985
Budavari et al. 1989
C-32
-------
IL ENVIRONMENTAL FATE
A. Environmental Release
No information was found in the secondary sources searched regarding the environmental release of hydrogen
peroxide. Solutions of hydrogen peroxide gradually deteriorate (Budavarietal., 1989). Hydrogen peroxide is a
naturally occurring substance. Gaseous hydrogen peroxide is recognized to be a key component and product of the
earth's lower atmospheric photochemical reactions, in both clean and polluted atmospheres. Atmospheric hydrogen
peroxide is also believed to be generated by gas-phase photochemical reactions in the remote troposphere (IARC,
1985)
B. Transport
No information was found in the secondary sources searched regarding the transport of hydrogen peroxide.
C. Transformation/Persistence
1. Air — Hydrogen peroxide may be removed from the atmosphere by photolysis giving rise to hydroxyl
radicals, by reaction with hydroxyl radicals, or by heterogenous loss processes such as rain-out (IARC,
1985).
2. Soil — No information was found in the secondary sources searched regarding the transformation or
persistence of hydrogen peroxide in soil, however, solutions of hydrogen peroxide gradually deteriorate
(Budavarietal., 1989).
3. Water — Hydrogen peroxide is a naturally occurring substance. Surface water concentrations of hydrogen
peroxide have been found to vary between 51-231 mg/L, increasing both with exposure to sunlight and the
presence of dissolved organic matter (IARC, 1985).
4. Biota — Hydrogen peroxide is a naturally occurring substance. Endogenous hydrogen peroxide has been
found in plant tissues at the following levels (mg/kg frozen weight): potato tubers, 7.6; green tomatoes, 3.5;
red tomatoes, 3.5; and castor beans in water, 4.7 (IARC, 1985).
C-33
-------
CHEMICAL SUMMARY FOR LEAD
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources. The
search identified sources of information for Lead.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of Lead are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF LEAD
Characteristic/Property Data Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Difrusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
7439-92-1
Pb
N/A
Metal
207.2
327.4°C
1740°C
Insoluble
10.65
no data
no data
no data
1.77mmHg@ 1000°C
Flammable solid
no data
no data
no data
no data
no data
no data
no data
no data
Howard and Neal (1992)
Howard and Neal (1992)
Weast(1983)
Weast(1983)
Weast(1983)
Weast(1983)
Weast(1983)
Budavarietal. (1996)
Budavarietal. (1996)
C-34
-------
CHEMICAL SUMMARY FOR MALEIC ACID
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of maleic acid are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF MALEIC ACID
Characteristic/Property Data Reference
CAS No.
Common Synonyms
Common Synonyms
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Difrusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
110-16-7
(Z)-butenedioic acid; toxilic acid
cis-l,2-ethylenedicarboxylic acid
maleinic acid
C4HA
HOOCCH=CHCOOH
white crystals
116.07
130.5°C
no data
441 g/lat25 °C
1.59g/cm3at20°C
no data
16 (estimated)
-0.34
3.06x10J mm Hg at 25 °C
stable
combustible
not pertinent
pKj = 1.83; pK2 = 6.07
no data
no data
no data; estimated to be < IxlO"8 atm mVmol
10-11
no data
Lide(1995)
Budavarietal. (1996)
Budavarietal. (1996)
Lewis (1993)
Budavarietal. (1996)
Aldrich(1996)
Budavarietal. (1996)
Budavarietal. (1996)
Lide(1995)
PHYSPROP(1998)
Lide(1995)
Lymanetal. (1990)
Hanschetal. (1995)
Daubert and Danner (1991)
Weiss (1986)
Lewis (1993)
Weiss (1986)
Ho ward (1989)
Estimated
HSDB(1998)
IL ENVIRONMENTAL FATE
A. Aquatic Fate
If released into water, maleic acid is not expected to adsorb to suspended solids or sediments in water based upon an
estimated Koc of 16 (Swann et al., 1983), determined from a log Kow of-0.34 (Hansch et al., 1995) and a
regression-derived equation (Lyman et al., 1990). Volatilization from the water column to the atmosphere is not
expected to occur (Lyman et al., 1990) based on an estimated Henry's Law constant of <10"8 atm-mVmole . Maleic
acid is expected to be stable to hydrolysis in the pH range of 5-9 typically encountered in the environment (Lyman et
C-35
-------
al, 1990). According to a classification scheme (Franke et al., 1994), a BCF of 10 in golden ide fish (Freitag, 1985,
as cited in HSDB, 1998) suggests that the potential for bioconcentration in aquatic organisms is low. Maleic acid
was determined to be readily degraded in biodegradation screening tests; however, no biodegradation studies were
available in environmental waters (Howard, 1989).
B. Atmospheric Fate
According to a model of gas/particle partitioning of semivolatile organic compounds in the atmosphere (Bidleman,
1988), maleicacid, which has a vapor pressure of 3.06xlO"5 mm Hg at 25 °C (Daubert and Banner, 1991), is
expected to exist as both a paniculate and vapor in the ambient atmosphere. Because maleic acid has pKa's of 1.83
and 6.07 (Howard, 1989), it is expected to exist in the dissociated form in the environment and form salts with
cations (HSDB, 1998). Removal of maleic acid from the atmosphere by reaction with photochemically-produced
hydroxyl radicals results in an estimated half-life of 2 days (Meylan and Howard, 1993). The reaction of maleic acid
with ozone in the atmosphere results in a gas-phase half-life ranging from 7-13 days (Meylan and Howard, 1993).
Maleic acid may undergo some degradation by direct photolysis; 17% of applied maleic acid was degraded after 17
hours following irradiation by light > 290 nm (Freitag et al., 1985, cited in HSDB, 1998). Wet deposition of maleic
in rain, snow, or fog is expected to be an important transport process in the atmosphere based upon its high water
solubility (Arimoto, 1989).
C. Terrestrial Fate
Based on a classification scheme (Swann et al., 1983), an estimated Koc of 16, determined from a log Kow of -0.34
(Hansch et al., 1995) and a regression-derived equation (Lyman et al., 1990), indicates that maleic acid is expected to
have very high mobility in soil. Volatilization of maleic acid from moist soil surfaces is not expected to be important
(Lyman et al., 1990) given an estimated Henry's Law constant of <10"8 atm-mVmole. In addition, maleic acid is not
expected to volatilize from dry soil given its vapor pressure of 3.06x10"5 mm Hg (Daubert and Danner, 1991). While
maleic acid is readily biodegradable in screening studies, no degradation data were available for soil systems
(Howard, 1989).
D. Summary
If released to air, a vapor pressure of 3.06xlO~5 mm Hg at 25 ° C indicates that maleic acid should exist as both a gas
and paniculate in the ambient atmosphere. Gas-phase maleic acid will be degraded in the atmosphere by reaction with
photochemically-produced hydroxyl radicals; the half-life for this reaction in air is estimated to be 2 hours. The
reaction of maleic acid with ozone in the atmosphere results in a gas-phase half-life ranging from 7-13 days. Wet
deposition of maleic acid from the atmosphere is expected to be an important transport process. Screening studies
suggest that direct photolysis if maleic acid may occur. A BCF of 10 in golden ide fish suggests the potential for
bioconcentration in aquatic organisms is low. If released to soil, maleic acid is expected to have very high mobility
based upon an estimated Koc of 16, and, therefore, it has the potential to leach to groundwater. Volatilization from
water and from moist soil surfaces is not expected to be an important fate process based upon an estimated Henry's
Law constant of <10"8 atm-mVmole. Volatilization from dry soil surfaces is not expected to occur based upon the
vapor pressure of this compound. Maleic acid was determined to be readily biodegraded in screening studies, although
no data were available for biodegradation in water or soil.
C-36
-------
CHEMICAL SUMMARY FOR MALIC ACID
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of malic acid are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF MALIC ACID
Characteristic/Property Data Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Difrusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
6915-15-7
hydroxysuccinic acid; apple acid
C4H605
COOHCH2CH(OH)COOH
colorless crystals
134.09
100 °C
140 °C, decomposes
592g/lat25°C
1.6g/cm3
no data
5 (estimated)
-1.26
3.28xlO-8mmHgat25°C
no data
combustible
no data
3.40
no data
no data
no data; expected to be < 10"8 atm mVmol
no data; expected to be <1
no data
Lewis (1993)
Lewis (1993)
Budavarietal. (1996)
Lewis (1993)
Lewis (1993)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
PHYSPROP(1998)
Lewis (1993)
Lymanetal. (1990)
Hanschetal. (1995)
Yaws (1994)
Lewis (1993)
PHYSPROP(1998)
Estimated
Estimated
IL ENVIRONMENTAL FATE
A. Aquatic Fate
If released into aquatic waters, malic acid is not expected to adsorb to suspended solids and sediments in water based
upon an estimated Koc of 5 (Swann et al., 1983), determined from a log Kow of -1.26 (Hansch et al., 1995) and a
regression-derived equation (Lyman et al., 1990). Volatilization from the water column to the atmosphere is not
expected to occur (Lyman et al., 1990) based on an estimated Henry's Law constant of <10"8 atm-mVmole . Malic
acid is expected to be stable to hydrolysis in the pH range of 5-9 typically encountered in the environment (Lyman et
al., 1990). According to a classification scheme (Franke et al., 1994), an estimated BCF of <1 suggests that the
potential for bioconcentration in aquatic organisms is low and not an important fate process. Results from a number
C-37
-------
of biological screening tests have shown that malic acid biodegrades relatively fast (Fischer et al., 1974; Malaney and
Gerhold, 1969; Heukelekian and Rand, 1955; as cited in HSDB, 1998).
B. Atmospheric Fate
According to a model of gas/particle partitioning of semivolatile organic compounds in the atmosphere (Bidleman,
1988), malic acid, which has a vapor pressure of 3.28xlO~8mmHg at 25 °C(Yaws, 1994), should exist almost
entirely as a paniculate in the ambient atmosphere. Removal of malic acid from the atmosphere by reaction with
photochemically-produced hydroxyl radicals results in an estimated half-life of 2 days (Meylan and Howard, 1993).
Wet deposition of malic acid in rain, snow, or fog is expected to be the dominant transport process in the atmosphere
based upon its high water solubility (Arimoto, 1989). Because carboxylic acids are generally resistant to hydrolysis,
malic acid is not expected to hydrolyze in environmental media (Lyman et al., 1990).
C. Terrestrial Fate
Based on a classification scheme (Swannetal., 1983), an estimated Koc of 5, determined from a log Kow of-1.26
(Hansch et al., 1995) and a regression-derived equation (Lyman et al., 1990), indicates that malic acid is expected to
have very high mobility in soil and may leach to groundwater. Volatilization of malic acid from moist soil surfaces is
not expected to be important (Lyman et al., 1990) given an estimated Henry's Law constant of <10"8 atm-mVmole. In
addition, malic acid is not expected to volatilize from dry soil given its vapor pressure of 3.28xlO"8 mm Hg (Yaws,
1994). Biodegradation screening studies reveal that malic acid biodegrades relatively fast (Fischer et al., 1974;
Malaney and Gerhold, 1969; Heukelekian and Rand, 1955; as cited in HSDB, 1998).
D. Summary
If released to air, a vapor pressure of 3.28xlO~8 mm Hg at 25 ° C indicates that malic acid should exist as a paniculate
in the ambient atmosphere. Removal of malic acid from the atmosphere by reaction with photochemically-produced
hydroxyl radicals results in an estimated half-life of 2 days. Wet deposition is expected to be the dominant transport
process of malic acid from the atmosphere. An estimated BCF of <1 suggests the potential for bioconcentration in
aquatic organisms is low. If released to soil, malic acid is expected to have very high mobility based upon an
estimated Koc of 5, and, therefore, it has the potential to leach to groundwater. Volatilization from water and from
moist soil surfaces is not expected to be an important fate process based upon an estimated Henry's Law constant of
<10"8 atm-mVmole, also volatilization from dry soil surfaces is not expected to occur based upon the vapor pressure
of this compound. Hydrolysis of malic acid in environmental media is not expected to occur. Malic acid was
determined to be readily biodegraded in screening studies, although no data were available for biodegradation in water
or soil.
C-38
-------
CHEMICAL SUMMARY FOR METHANESULFONIC ACID
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of methanesulfonic acid are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF METHANESULFONIC
ACID
Characteristic/Property Data Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Difrusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
75-75-2
methylsulfonic acid
CH4O3S
CH3SO2OH
solid
liquid at room temperature
96.11
20 °C
200 °C; 167 °C at 10 mm Hg
1.0xl03g/Lat20 °C
1.48g/cm3
no data
1 (estimated)
no data; estimated to be < 1
4.28x1 Q4 mm Hg at 25 °C
thermally stable at mod. elevated temps
no data
112°C
-1.86
no data
no data
1.3xlO"8 atm mVmol (estimated)
3 (estimated)
no data
Lide(1995)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Lewis (1993)
Budavarietal. (1996)
Lide(1995)
Lewis (1993); Lide (1995)
PHYSPROP(1998)
Lide(1995)
HSDB(1998)
Estimated
Daubert and Danner (1991)
Budavarietal. (1996)
ECDIN (1998)
Serjeant and Dempsey (1979)
Meylan and Howard (1991)
Meylanetal. (1997)
IL ENVIRONMENTAL FATE
A. Aquatic Fate
If released into aquatic waters, methanesulfonic acid is not expected to adsorb to suspended solids and sediments in
water based upon an estimated Koc of 1 (Swann et al., 1983), determined from a structure fragment estimation
method (Meylan et al., 1992). Volatilization from the water column to the atmosphere is not expected to occur
(Lyman et al., 1990) based on an estimated Henry's Law constant of 1.3xlO"8 atm-mVmole (Meylan and Howard,
1991; SRC, 1998). Methanesulfonic acid is expected to be stable to hydrolysis in the pH range of 5-9 typically
C-39
-------
encountered in the environment (Lyman et al., 1990). According to a classification scheme (Franke et al., 1994), an
estimated BCF of 3 (Meylan et al., 1997) suggests that the potential for bioconcentration in aquatic organisms is low.
It was determined that many bacterial types can degrade methanesulfonic acid through diverse routes and at different
rates, although specifics were not given (Baker et al., 1991, as cited in HSDB, 1998). Because methanesulfonic acid
has pKa of-1.86 (Serjeant and Dempsey, 1979), it is expected to exist in the dissociated form in the environment.
B. Atmospheric Fate
According to a model of gas/particle partitioning of semivolatile organic compounds in the atmosphere (Bidleman,
1988), methanesulfonic acid, which has a vapor pressure of 4.28xlO~4mmHgat25 °C (Daubert and Banner, 1991),
has the potential to exist as both a vapor and paniculate in the ambient atmosphere. Because methanesulfonic acid
has pKa of -1.86 (Serjeant and Dempsey, 1979), it is expected to exist in the dissociated form in the environment.
Removal of methanesulfonic acid from the atmosphere by reaction with photochemically- produced hydroxyl radicals
results in an estimated half-life of 58 days (Meylan and Howard, 1993). In the atmosphere, methanesulfonic acid is
concentrated in the smaller size particles, 0.25-2 um in diameter (Kolaitis et al., 1989, as cited in HSDB, 1998).
Removal of paniculate methanesulfonic acid from the atmosphere can occur through wet and dry deposition (HSDB,
1998).
C. Terrestrial Fate
Based on a classification scheme (Swann et al., 1983), an estimated Koc of 1, determined from a structure fragment
estimation method (Meylan et al., 1992), indicates that methanesulfonic acid is expected to have very high mobility in
soil. Volatilization of methanesulfonic acid from moist soil surfaces is not expected to be important (Lyman et al.,
1990) given an estimated Henry's Law constant of 1.3xlO"8 atm-mVmole (Meylan and Howard, 1991; SRC, 1998).
In addition, methanesulfonic acid is not expected to volatilize from dry soil given its vapor pressure of 4.28xlO"4 mm
Hg (Daubert and Danner, 1991). It was determined that many bacterial types can degrade methanesulfonic acid
through diverse routes and at different rates, although specifics were not given (Baker et al., 1991, as cited in HSDB,
1998).
D. Summary
If released to air, a vapor pressure of 4.28 x 10~4 mm Hg at 25 ° C indicates that methanesulfonic acid has the
potential to exist as both a vapor and paniculate in the ambient atmosphere. Gas-phase methanesulfonic acid will be
degraded in the atmosphere by reaction with photochemically-produced hydroxyl radicals; the half-life for this
reaction in air is estimated to be 58 hours. Removal of paniculate methanesulfonic acid from the atmosphere can
occur through wet and dry deposition. An estimated BCF of 3 suggests the potential for bioconcentration in aquatic
organisms is low. If released to soil, methanesulfonic acid is expected to have very high mobility based upon an
estimated Koc of 1, and, therefore, it has the potential to leach to groundwater. Volatilization from water and from
moist soil surfaces is not expected to be an important fate process based upon an estimated Henry's Law constant of
1.3xlO"8atm-m3/mole. Hydrolysis of methanesulfonic acid is not expected to occur. Volatilization from dry soil
surfaces is not expected to occur based upon the vapor pressure of this compound. Methanesulfonic acid was
determined to be biodegraded by many bacterial types, although specifics were not given.
C-40
-------
CHEMICAL SUMMARY FOR NICKEL SULFATE
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources. The
search identified sources of information for nickel and soluble salts of nickel, including nickel sulfate and nickel
sulfate hexahydrate.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of nickel sulfate are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF NICKEL SULFATE
Characteristic/Property Data Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Difrusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
7786-81-4
sulfuric acid, nickel (2+) salt
NiO4S
NiSO4
green-yellow orthorhombic crystals
154.757
840 °C, decomposes
no data
293g/LatO °C
4.01 g/cm3
no data
no data; expected to be < 10
no data; expected to be < 1
no data; expected to be <10-6 mm Hg at 25 C
no data
not flammable
no data; expected to be > 350 °C
no data
no data
no data
no data; expected to be < IxlO"8
no data
no data
Lide(1995)
Howard and Neal (1992)
Budavarietal. (1996)
Lide(1995)
Lide(1995)
Lide(1995)
Lide(1995)
Dean (1985)
Lide(1995)
SRC (1998)
SRC (1998)
Estimated
Prager(1995)
SRC (1998)
SRC (1998)
IL ENVIRONMENTAL FATE
A. Aquatic Fate
If released into water, nickel sulfate is expected to dissociate into nickel (Ni2+) and sulfate [(SO4)2"] ions. The
dissociation of nickel sulfate into its component ions indicates that the compound nickel sulfate is not expected to
volatilize from water surfaces. In aqueous solutions, nickel exists as the hexaquonickel ion, [Ni(H2O)62+]; this ion is
poorly absorbed by most living organisms (Sunderman and Oskarsson, 1991). In natural waters, nickel exists both in
the ionic form and as stable organic complexes (Sunderman and Oskarsson, 1991). Nickel compounds are generally
C-41
-------
soluble at pH values less than 6.5, but at pH values greater than 6.7 nickel exists predominantly as insoluble nickel
hydroxides (Sunderman and Oskarsson, 1991). Shellfish and Crustacea generally contain higher concentrations of
nickel in their flesh than do other species offish (Sunderman and Oskarsson, 1991).
B. Atmospheric Fate
If released to the atmosphere, nickel sulfate's high melting point (Lide, 1995) and low vapor pressure (SRC, 1998)
indicate that it will exist as a paniculate (Bidleman, 1988). Wet and dry deposition of nickel sulfate is expected to be
the dominant fate process in the atmosphere (Arimoto, 1989). The rate of dry deposition will depend on the
prevailing winds and particle size (Bodek et al., 1988). Nickel sulfate's high water solubility (Dean, 1985) indicates
that it is expected to undergo wet deposition in rain, snow, or fog.
C. Terrestrial Fate
If nickel sulfate is released to soil, it is expected to dissociate into Ni2+ and (SO4)2" ions in the presence of moisture.
Iron and manganese oxides, clay minerals, and organic matter may be important sorbents of nickel (Bodek et al.,
1988) and will retard its migration through soil. Complexing ligands, such as organic acids, may reduce the sorption
of nickel (Bodek et al., 1988). Acid rain has a tendency to mobilize nickel from soil and increase leaching into
groundwater due to the high solubility of nickel compounds at pH values less than 6.5 (Sunderman and Oskarsson,
1991). The high melting point, low vapor pressure, and low Henry's Law constant expected for an ionic salt indicate
that nickel sulfate will not volatilize from either moist or dry soil surfaces (Bodek et al., 1988).
D. Summary
If released into water, nickel sulfate is expected to dissociate into nickel (Ni2+) and sulfate (SO4)2" ions. Therefore,
nickel sulfate is not expected to adsorb to suspended solids or sediment in the water column, bioconcentrate in
aquatic organisms, or volatilize from water surfaces. In natural waters, nickel exists in both the ionic form and as
stable organic complexes; at pH values greater than 6.7 it exists as insoluble nickel hydroxides. In moist soils, nickel
sulfate is expected to dissociate into its component ions. Ionic nickel may be sorbed by iron and manganese oxides,
clay minerals, and organic matter; acid rain and complexing ligands may reduce the sorption of nickel. Volatilization
of nickel sulfate from soil surfaces is not expected to occur. If released to the atmosphere, nickel sulfate is expected
to exist as a paniculate. Nickel sulfate is expected to be physically removed from the atmosphere by wet and dry
deposition. The rate of dry deposition will depend on particle size and prevailing wind patterns.
C-42
-------
CHEMICAL SUMMARY FOR PALLADIUM CHLORIDE
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of palladium chloride are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF PALLADIUM CHLORIDE
Characteristic/Property Data Reference
CAS No.
Common Synonyms
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
7647-10-1
Palladous chloride
Palladium (II) chloride
Cl2Pd
PdCl2
red rhombohedral crystals; hygroscopic
177.33
500°C (decomposes)
decomposed at high temperatures
soluble1
4.0 g/cm3
no data
no data; expected to be <10
no data; expected to be <1
no data; expected to be <10"6 mm Hg
no data
no data
no data
expected to dissociate into Pd2+ and Cl"
no data
no data
no data; expected to be <1X10"8
no data
no data
CAS (1998)
Budavarietal. (1996)
Lide(1995)
Budavarietal. (1996)
Budavarietal. (1996)
Lide(1995)
Lide(1995)
Lide(1995)
Budavarietal. (1996)
Dean (1985)
Lide(1995)
SRC (1998)
SRC (1998)
SRC (1998)
SRC (1998)
SRC (1998)
1 This form of expressing solubility cannot be converted into g/L units
IL ENVIRONMENTAL FATE
A. Aquatic Fate
If palladium chloride is released into the water column, it is expected to dissociate into palladium (Pd2+) and chloride
(Cl") ions. The dissociation of palladium chloride into its component ions indicates that palladium chloride is not
expected to bioconcentrate in aquatic organisms or volatilize from water surfaces. Palladium ions may adsorb to
charged surfaces of suspended sediments and humic materials in the water column (Evans, 1989). The chloride ion
may complex with heavy metals in natural waters, thereby increasing their solubility (Bodek et al., 1988). Adsorption
C-43
-------
of the chloride ion to suspended solids and sediment in the water column is not expected to be an important fate
process.
B. Atmospheric Fate
If palladium chloride is released to the atmosphere, the low vapor pressure expected for an ionic salt indicates that it
will exist as a paniculate. Dry deposition of palladium chloride is expected to be the dominant fate process in the
atmosphere (Arimoto, 1989). The rate of dry deposition will depend on the prevailing winds and particle size (Bodek
et al, 1988). Palladium chloride is expected to undergo wet deposition (Arimoto, 1989) in rain, snow, or fog, based
upon its water solubility (Dean, 1985).
C. Terrestrial Fate
If palladium chloride is released to soil, it is expected to dissociate into its component ions in moist soils. The
dissociation of palladium chloride in moist soils indicates that palladium chloride is not expected to volatilize from
moist soil surfaces. While no specific information concerning the sorption of ionic palladium in soils was available,
some metals adsorb to charged surfaces of soil particles or form inner sphere complexes with variable-charge soil
surfaces (Evans, 1989). If this occurs with palladium then its rate of migration through soil may be slow. Chloride is
extremely mobile in soils (Bodek et al., 1988). The chloride ion may complex with heavy metals, thereby increasing
their solubility (Bodek et al., 1988) and potential for leaching into groundwater. The low vapor pressure expected for
an ionic salt indicates that palladium chloride will not volatilize from dry soil surfaces.
D. Summary
If released into water, palladium chloride will dissociate into palladium and chloride ions. Therefore, palladium
chloride is not expected to adsorb to suspended solids or sediment in the water column, bioconcentrate in aquatic
organisms, nor volatilize from water surfaces. Palladium ions may adsorb to charged surfaces of suspended
sediments and humic matter in the water column. Adsorption of the chloride ion to suspended solids and sediment in
the water column is not expected to be an important fate process. If released to soil, palladium chloride is expected to
dissociate into its component ions in moist soils. The dissociation of palladium chloride into its component ions
indicates that palladium chloride is not expected to volatilize from moist soil surfaces. Ionic palladium may adsorb
to charged surfaces of soil particles or form inner sphere complexes with variable-charge soil surfaces. Chloride is
extremely mobile in soils. The low vapor pressure expected for an ionic salt indicates that volatilization of palladium
chloride from soil surfaces is not expected to be an important fate process. If released to the atmosphere, palladium
chloride is expected to exist as a paniculate. Palladium chloride is expected to be physically removed from the
atmosphere by wet and dry deposition. The rate of dry deposition will depend on particle size and prevailing wind
patterns.
C-44
-------
CHEMICAL SUMMARY FOR PHOSPHORIC ACID
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of phosphoric acid are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF PHOSPHORIC ACID
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Data
7664-38-2
orthophosphoric acid
H304-P
H3P04
unstable, orthorhombic crystals; clear, syrupy liquid
98.00
42.35 °C (crystals); -11.8 °C (30% soln)
261 °C (crystals); 101.8 °C (30% soln)
5,480 g/1 at 20 °C (crystals);
Reference
Lide(1995)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Gard(1996)
Gard(1996)
Weastetal. (1985)
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Difiusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
354.1 g/1 at 20 °C (30% soln)
1.86 g/cm3 at 25 °C (crystals);
1.18 g/cm3 at 25 °C (30% soln)
no data
expected to be < 10
expected to be < 1
0.03 mm Hg at 20 °C (crystals);
16.3 mm Hg at 20 °C (30% soln)
relatively unreactive at room temperature
no data
no data
pKj: 2.15; pK2: 7.09; pK3: 12.32
no data
no data
expected to be < 1x10"8 atm m3/mole
no data
no data
Gard(1996)
SRC (1998)
SRC (1998)
Gard(1996)
Gard(1996)
Budavarietal. (1996)
SRC (1998)
IL ENVIRONMENTAL FATE
A. Aquatic Fate
Phosphoric acid is a weak tribasic acid with a pKj of 2.15 (Budavari et al., 1996) and, if released into the water
column at low concentrations, it will dissociate into dihydrogen phosphate (H2PO4) and hydrogen (H+) ions.
Dihydrogen phosphate then dissociates into hydrogen phosphate ion (HPO4~2; pK2 of 7.09) and orthophosphate ion
(PO4~3; pK3 of 12.32). As a result, phosphoric acid is not expected to volatilize or bioconcentrate in aquatic
organisms. The phosphates become available in the water column and form salts, thus affecting biological
C-45
-------
productivity (Bodek et al, 1988). Phosphorous, in the form of phosphate, is an essential nutrient to plants in aquatic
environments (Bodek et al., 1988). In addition, the phosphates can complex with metal ions in sediment and water to
form insoluble species such as FePO4 and CaHPO4 (Bodek et al., 1988).
B. Atmospheric Fate
If phosphoric acid is released to the atmosphere, its vapor pressure indicates it will exist predominantly as a vapor in
the ambient atmosphere. Wet deposition of phosphoric acid in rain, snow, or fog is expected to be the dominant fate
process in the atmosphere based upon its solubility in water (Arimoto, 1989).
C. Terrestrial Fate
If phosphoric acid is released to soil, it will dissociate into dihydrogen phosphate and hydrogen ions, ultimately
dissociating to the orthophosphate ion at high pH's. Phosphate added to soil as fertilizer is quickly sorbed and later
"fixed" (probably precipitated) into less soluble forms (Bodek et al., 1988). A similar fate is anticipated for
phosphate species from phosphoric acid. While the exact mechanism of sorption is uncertain, phosphate fixation is
appreciable in all but very coarse-textured soils; only about one-fourth of the fertilizer phosphate is usable by plants,
the rest being lost to the occluded soil fraction (Bodek etal., 1988). Phosphorous, in the form of phosphate, is an
essential nutrient to plants (Bodek et al., 1988).
D. Summary
Phosphoric acid is a tribasic acid in which the first hydrogen is strongly ionizing, the second moderately weak, and the
third very weak. Both acidic and basic salts can be formed from phosphoric acid. If released into water, phosphoric
acid will dissociate into dihydrogen phosphate (H2PO4) and hydrogen (H+) ions, eventually dissociating into the
orthophosphate ion (PO4~3) under the proper conditions. Therefore, phosphoric acid is not expected to adsorb to
suspended solids or sediment in the water column, bioconcentrate in aquatic organisms, nor volatilize from water
surfaces. The phosphates become available in the water column and form salts, affecting biological productivity, and
complexing with metal ions form insoluble species such as FePO4 and CaHPO4. If released to soil, phosphoric acid is
expected to dissociate into its component ions in moist soils. Phosphate added to soil as fertilizer is quickly sorbed
and later "fixed" into less soluble forms; phosphate fixation is appreciable in all but very coarse-textured soils; only
about one-fourth of the fertilizer phosphate is usable by plants, the rest being lost to the occluded soil fraction.
Phosphorous, in the form of phosphate, is an essential nutrient for aquatic and terrestrial plants. Volatilization of
phosphoric acid from soil surfaces is not expected to occur. If released to the atmosphere, phosphoric acid is
expected to exist as a gas. Phosphoric acid is expected to be physically removed from the atmosphere by wet
deposition based upon its water solubility.
C-46
-------
CHEMICAL SUMMARY FOR POTASSIUM AUROCYANIDE
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of potassium aurocyanide are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF POTASSIUM
AUROCYANIDE1
Characteristic/Property Data Reference
CAS No. (deleted)
CAS No.
Common Synonyms
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Difiusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
554-07-4
13967-50-5
gold potassium cyanide
potassium dicyanoaurate(I)
C2AuKN2
KAu(CN)2
dihydrate, crystalline powder
288.13
no data; expected to be > 350 °C
no data; expected to be > 500 °C
Approximately 130 g/L2
1 g dissolves in 0.5 ml boiling H2O
3.45 g/cm3
no data
no data; expected to be < 10
no data; expected to be < 1
no data; expected to be <10-6 mm Hg at 25 C
stable in aqueous solution2
not flammable
no data; expected to be > 350 °C
readily dissociates to K+ and [Au(CN)J~
no data
no data
no data; expected to be < IxlO"8
no data
no data
CAS (1998)
CAS (1998)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
SRC (1998)
SRC (1998)
Budavarietal. (1996)
Budavarietal. (1996)
Weast(1986)
SRC (1998)
SRC (1998)
SRC (1998)
Cotton and Wilkinson (1966)
ECDIN (1998)
SRC (1998)
Cohn and Stern (1994)
SRC (1998)
1 Both electrochemical and electroless gold plating processes that use potassium aurocyanide under basic conditions may contain potassium
cyanide as a complexing agent (Gmelin, 1998; Cohn and Stern, 1994; McDermott, 1974). The concentration of KCN is typically
approximately 6 g/L (0.1 M), although values as high as 200 g/L (3 M) have been reported (Gmelin, 1988).
2 Estimated from a reported solubility of 1 g dissolves in 7 ml H2O (Budavari et al., 1996).
3 Potassium aurocyanide is stable in aqueous solution under both basic and neutral conditions (Cotton and Wilkinson, 1966; Cohn and Stern,
1994). It is also stable in aqueous solutions under acidic conditions (Cohn and Stern, 1994), although common acids such as HC1, H2SO4,
FINO3, and H2S are known to degrade potassium aurocyanide (Gmelin, 1998) and release HCN and gold monocyanide (Budavari et al., 1996;
Gmelin, 1998). Concentrated acids and elevated temperatures, or both, are required (Gmelin, 1998). Potassium aurocyanide is commonly
used in warm (35-55°C) acidic plating solutions at a pH of approximately 4 (Gmelin, 1998) and stabilized acidic plating baths containing
C-47
-------
potassium aurocyanide have been reported down to a pH of 1.5 (McDermott, 1974), yet it is generally considered stable in water above pH 3
(Renner and Johns, 1989). These data indicate that potassium aurocyanide is expected to be chemically stable in the pH range 5-9 typically
found in the environment (Lyman et al, 1990), but not under highly acidic conditions such as those found in the stomach (pH 1-2).
IL ENVIRONMENTAL FATE
A. Aquatic Fate
If released to water, potassium aurocyanide will rapidly and completely dissociate into potassium (K+) and
aurocyanide ([Au(CN)2]~) ions (Cohn and Stern, 1994). The aurocyanide ion is expected to be stable to hydrolysis in
the pH range of 5-9 typically encountered in the environment (Lyman et al, 1990; SRC, 1998). The dissociation of
potassium aurocyanide into its component ions also indicates that it is not expected to volatilize from water surfaces
to the atmosphere, adsorb to sediment and suspended organic matter, orbioconcentrate in fish and aquatic organisms
(Bodeketal., 1988).
B. Atmospheric Fate
If released to the atmosphere, potassium aurocyanide will exist as a paniculate. Its atmospheric fate will be
dominated by deposition to the Earth's surface via wet and dry processes, as potassium aurocyanide is not expected to
undergo degradation by the most common atmospheric oxidant, hydroxyl radicals (Lyman et al, 1990; SRC, 1998).
The rate of dry deposition will be dependent on the prevailing winds and particle size; fine particles of potassium
aurocyanide have the potential to be transported significant distances from their original point of release (Bodek et al,
1988). Potassium aurocyanide is expected to undergo efficient wet deposition in either rain or fog due to its water
solubility. Dissolution in clouds followed by wet deposition may also occur. Potassium aurocyanide is stable to
light (Cohn and Stern, 1994), and is not expected to undergo degradation by direct photolyis.
C. Terrestrial Fate
If potassium aurocyanide is released to soil, it is expected to display very high mobility based on its water solubility
of 143 g/L (Budavari, 1996). Therefore, it has the potential to leach into groundwater. Its rate of leaching through
soil may be attenuated by the formation of insoluble soil/aurocyanide complexes that can arise from reactions with
metals naturally present in soil (Bodek et al, 1988). The importance of complex formation for potassium aurocyanide
in soil is not known. The very high melting point and low vapor pressure expected for an ionic salt indicates that
potassium aurocyanide will not volatilize from either moist or dry soils to the atmosphere (Bodek et al, 1988).
D. Summary
If released to water, potassium aurocyanide will dissociate into K+ and [Au(CN)2]~ ions. Therefore, it is not expected
to adsorb to sediment and suspended organic matter, bioconcentrate in fish and aquatic organisms, or volatilize from
water surfaces to the atmosphere. The aurocyanide ion is expected to be chemically stable and it is not expected to
hydrolyze in the pH range 5-9 typically found in the environment. In soil, potassium aurocyanide is likely to display
very high mobility as a result of its relatively high water solubility and it has the potential to leach to groundwater. Its
rate of leaching through soil may be attenuated by the formation of insoluble soil/aurocyanide complexes although
the importance of this process is not known. Volatilization from soil surfaces to the atmosphere is not expected to
occur. If released to the atmosphere, potassium aurocyanide is expected to exist as a paniculate. Its atmospheric fate
is expected to be dominated by wet and dry deposition to the Earth's surface. Efficient removal from the atmosphere
during rain events is expected although the rate of dry deposition will be dependent on its particle size and the
prevailing wind patterns. Therefore, fine particles of potassium aurocyanide have the potential to travel significant
distances from their original point of release.
C-48
-------
CHEMICAL SUMMARY FOR POTASSIUM PEROXYMONOSULFATE
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of potassium peroxymonosulfate are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF POTASSIUM
PEROXYMONOSULFATE
Characteristic/Property Data Reference
CAS No.
Common Synonyms
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Difrusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
10058-23-8
Monopotassium peroxymonosulfurate
Peroxymonosulfuric acid, monopotassium salt
HO5S.K
HOOS(O)(O)OK
no data
153.18
no data
no data
no data
no data
no data
no data; expected to be <10
no data; expected to be <1
no data; expected to be <1X10"6 mm Hg
no data
no data
no data
expected to dissociate
no data
no data
no data; expected to be <1X10"8
no data
no data
CAS (1998)
Howard and Neal (1992)
Howard and Neal (1992)
Howard and Neal (1992)
CAS (1998)
Howard and Neal (1992)
Estimated
Estimated
Estimated
Bodeketal.(1988)
Estimated
IL ENVIRONMENTAL FATE
A. Aquatic Fate
Most potassium salts are highly dissociated in natural waters (Bodek et al., 1988). Therefore, if potassium
peroxymonosulfate is released into water, it is expected to dissociate into potassium (K+) and peroxymonosulfate
(SO5~) ions. The potassium ion is expected to exist predominately as the free ion in most natural waters (Bodek et al.,
1988). Ion exchange processes with suspended solids and sediment in the water column are expected to remove ionic
potassium from solution; however, ionic potassium may be displaced by other cations present in natural waters with a
C-49
-------
higher affinity for ion exchange sites (Bodek et al., 1988). Aqueous solutions of the impure potassium
peroxymonosulfate, i.e., those containing dipotassium sulfate and monopotassium sulfate, decompose yielding
mainly O2 and sulfate (SO42~), hydrogen peroxide and peroxydisulfate (S2O82~) occur in small amounts (Cotton and
Wilkinson, 1980). Sulfate ions may participate in oxidation-reduction reactions or react with cations present in the
water column to form soluble complexes or insoluble precipitates (Bodek et al., 1988). Sulfate-reducing
microorganisms are important mediators in redox reactions involving this ion (Bodek et al., 1988). Peroxy
compounds are short-lived because of the inherent instability of the O-O bond and are expected to degrade rapidly
(U.S. EPA, 1993).
B. Atmospheric Fate
If potassium peroxymonosulfate is released to the atmosphere, the low vapor pressure expected for an ionic salt
indicates that potassium peroxymonosulfate will exist as a paniculate. Wet and dry deposition of potassium
peroxymonosulate is expected to be an important fate process in the atmosphere (Arimoto, 1989). The rate of dry
deposition will depend on the prevailing winds and particle size (Bodek et al., 1988).
C. Terrestrial Fate
If potassium peroxymonosulfate is released to soil, it may decompose in moist soils; the importance of this process is
not known. The low vapor pressure expected for an ionic salt indicates that potassium peroxymonosulfate will not
volatilize from dry soil surfaces. The uncomplexed potassium ion is expected to be the predominant species in well-
drained soils from pH 4 to pH 10 (Bodek et al., 1988). Ion exchange reactions are expected to attenuate the mobility
of the potassium ion in the subsurface environment, however ionic potassium may be displaced by other cations with
a higher affinity for ion exchange sites (Bodek et al., 1988). Peroxy compounds are short-lived because of the
inherent instability of the O-O bond and are expected to degrade rapidly (U.S. EPA, 1993).
D. Summary
If released into water, potassium peroxymonosulfate is expected to dissociate into potassium and peroxymonosulfate
ions. The dissociation of potassium peroxymonosulfate into its component ions indicates that potassium
peroxymonosulfate is not expected to volatilize from water surfaces or bioconcentrate in aquatic organisms. In most
natural waters, the potassium ion is expected to exist predominately as the free ion. Ion exchange processes with
suspended solids and sediment in the water column are expected to remove ionic potassium from solution; however
ionic potassium may be displaced by other cations in natural waters with a higher affinity for ion exchange sites.
Sulfate ions may participate in oxidation-reduction reactions or react with cations present in the water column to
form soluble complexes or insoluble precipitates; sulfate-reducing microorganisms are important mediators in redox
reactions involving this ion. If released to soil, potassium peroxymonosulfate may decompose in moist soils or
dissociate into its component ions. As a result, potassium peroxymonosulfate is not expected to volatilize from
moist soil surfaces. The low vapor pressure expected for an ionic salt indicates that potassium peroxymonosulfate is
not expected to volatilize from dry soil surfaces. The mobility of potassium ions will be retarded by ion exchange
processes with charged surfaces of soil particles. However, since the potassium ion is held weakly by ion exchange
processes, it may leach into groundwater. Peroxy compounds are short-lived because of the inherent instability of the
O-O bond and are expected to degrade rapidly. If released to the atmosphere, potassium peroxymonosulfate is
expected to exist as a paniculate based upon the low vapor pressure expected for an ionic salt. Wet and dry
deposition is expected to be the dominant fate process in the atmosphere.
C-50
-------
CHEMICAL SUMMARY FOR PROPIONIC ACID
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of propionic acid are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF PROPIONIC ACID
Characteristic/Property Data Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconcentration Factor
Odor Threshold
79-09-4
methyl acetic acid; ethyl formic acid
C3H602
CH3CH2COOH
oily liquid
74.08
-21.5 °C
141.1 °C
lxlO+3g/l@25 °C
d25'4, 0.99336
no data
36 (calculated)
0.33
3.53mmHg@25 °C
corrodes steel, metal
combustible
136 °F(58 °C), open cup
pKa = 4.88
no data
no data
4.45x1 Q-7 atm mVmole @ 25 °C
0.02 (calculated)
no data
Howard and Neal (1992)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
U.S. EPA (1981)
Budavarietal. (1996)
Lymanetal. (1990)
Hanschetal. (1995)
Daubert and Danner (1985)
Weiss (1986)
Lewis (1993)
Budavarietal. (1996)
Serjeant and Dempsey (1979)
Butler and Ramchandani (1935)
Lymanetal. (1990)
IL ENVIRONMENTAL FATE
A. Aquatic Fate
Aerobic biodegradation is likely to be the most important removal mechanism of propionic acid from aquatic systems
(Dias and Alexander, 1971, as cited in HSDB, 1998). WithapKaof 4.88 (Serjeant and Dempsey, 1979), propionic
acid and its conjugate base will exist in environmental waters in varying proportions that are pH dependent. Under
neutral and alkaline conditions, propionic acid is expected to exist predominantly as its conjugate base, the propionate
ion (Lyman et al., 1990). In addition, at a pH of 4.88 propionic acid is 50% dissociated; even under mildly acidic
conditions, it will exist predominantly as the conjugate base. In general, organic ions are not expected to volatilize
from water or adsorb to paniculate matter in water to the degree that would be predicted for their neutral
C-51
-------
counterparts. An estimated Koc of 36, determined from a log Kow of 0.33 (Hansch et al., 1995), indicates propionic
acid should not partition from the water column to organic matter contained in sediments and suspended solids.
Similarly, the Kow indicates that bioconcentration in fish and aquatic organisms is not an important fate process.
Propionic acid's Henry's Law constant of 4.45xlO~7 atm mVmole (Butler and Ramchandani, 1935) indicates that
volatilization of propionic acid from environmental waters should be extremely slow (Lyman et al., 1990).
Volatilization will be attenuated depending upon pH and the amount of propionic acid that is dissociated. Since
carboxylic acids are generally resistant to aqueous hydrolysis (Lyman et al., 1990), it is not expected to be an
important fate process for propionic acid. The direct photolysis (Calvert and Pitts, 1966, as cited in HSDB, 1998)
and reaction of propionic acid with photochemically-generated hydroxyl radicals in water (Anbar and Neta, 1967, as
cited in HSDB, 1998) are also not expected to be important fate processes.
B. Atmospheric Fate
Based on a vapor pressure of 3.53 mm Hg at 25 ° C (Daubert and Banner, 1985, as cited in HSDB, 1998), propionic
acid is expected to exist almost entirely in the vapor phase in the ambient atmosphere (Bidleman, 1988). The rate
constant for the reaction of propionic acid with photochemically-produced hydroxyl radicals in air has been
experimentally determined to be 1.22 x 10~12 cmVmolecule-sec at 25 °C (Daugautetal., 1988, as cited in HSDB,
1998). This corresponds to an atmospheric half-life of approximately 13 days. Since low molecular weight organic
acids have absorption bands at wavelengths well below the environmentally important range of 290 nm, the direct
photolysis of propionic acid in air is not expected to be important (Calvert and Pitts, 1966, as cited in HSDB, 1998).
Extensive monitoring data (Chapman et al., 1986; Hoffman and Tanner, 1986; Winkeler et al., 1988; Mazurek and
Simoneitt, 1986, as cited in HSDB, 1998) has shown that physical removal of propionic acid from the air by wet
deposition (rainfall, dissolution in clouds, etc.) may be an important fate process under the appropriate atmospheric
conditions.
C. Terrestrial Fate
Biodegradation is likely to be the most important removal mechanism of propionic acid from aerobic soil (Dias and
Alexander, 1971, as cited in HSDB, 1998). With a pKa of 4.88 (Serjeant and Dempsey, 1979), propionic acid and its
conjugate base will exist in varying proportions that are dependent on the pH of the soil. A Henry' s Law Constant of
4.45xlO"7 atm mVmole (Butler and Ramchandani, 1935) indicates that volatilization of propionic acid from moist
soil should be extremely slow (Lyman et al., 1990). Yet, propionic acid should volatilize rapidly from dry surfaces
based upon a vapor pressure of 3.53 mm Hg at 25 ° C (Daubert and Danner, 1985, as cited in HSDB, 1998).
Volatilization will be attenuated depending upon pH and the amount of propionic acid dissociated. An estimated Koc
of 36, determined from a log Kow of 0.33 (Hansch et al., 1995), indicates that propionic acid may be highly mobile
in soil (Swann et al., 1983). In addition, monitoring data has shown that propionic acid can leach to groundwater
(Stuermer et al., 1982; Burrows and Rowe, 1975; Lema et al., 1988, as cited in HSDB, 1998). Organic ions
generally do not volatilize from moist soil surfaces and do not undergo adsorption to the extent of their neutral
counterparts, which is consistent with propionic acid's potential for displaying high mobility through soils under
conditions where rapid biodegradation does not occur.
D. Summary
With a pKa of 4.88, propionic acid and its conjugate base will exist in environmental media in varying proportions
that are pH dependent; under typical environmental conditions, propionic acid will exist predominantly as its
conjugate base. A Henry's Law constant of 4.45xlO"7 atm mVmole at 25 °C indicates that volatilization of propionic
acid from environmental waters and moist soil should be extremely slow. Yet, based on a vapor pressure of 3.53 mm
Hg, propionic acid should volatilize rapidly from dry surfaces. However, volatilization of propionic acid will be pH
dependent; if propionic acid is dissociated, very little (about 1%) will be available for volatilization. A relatively low
estimated Koc indicates that propionic acid should not partition from the water column to organic matter contained in
sediments and suspended solids; the Koc also indicates that it should be highly mobile in soil. However, monitoring
data has shown that propionic acid has the potential to leach to groundwater under the appropriate conditions.
Propionic acid is miscible with water and monitoring data has shown that physical removal from air by wet deposition
is an important removal mechanism. Biodegradation is likely to be the most important removal mechanism of
propionic acid from aerobic soil and water. In the atmosphere, propionic acid is expected to exist almost entirely in
the gas phase and oxidative removal by photochemically-produced hydroxyl radicals has a half-life of 13 days. The
C-52
-------
hydrolysis in water, photolysis in air, and bioconcentration in aquatic organisms are not expected to be important fate
processes for propionic acid.
C-53
-------
CHEMICAL SUMMARY FOR SILVER NITRATE
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources. The
search identified sources of information for silver nitrate, other nitrate salts and silver.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of silver nitrate are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF SILVER NITRATE
Characteristic/Property Data Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
7761-88-8
silver(I)nitrate
AgNO3
AgNO3
colorless, rhombohedral crystals
169.873
212 °C
440 °C decomposes
2,500 g/L water
4.35 g/cm3
no data
no data; expected to be < 10
no data; expected to be < 1
no data; expected to be <10-6 mm Hg at 25 °C
can explode on contact with soot, organics
not flammable
no data; expected to be > 350 °C
no data
no data
no data
no data; expected to be < IX10"8
no data
no data
Lide(1995)
Lide(1995)
Budavarietal. (1996)
Lide(1995)
Lide(1995)
Lide(1995)
Lide(1995)
Lide(1995)
Budavarietal. (1996)
Lide(1995)
SRC (1998)
SRC (1998)
Estimated
Renner(1993)
Prager(1995)
SRC (1998)
SRC (1998)
IL ENVIRONMENTAL FATE
A. Aquatic Fate
If silver nitrate is released into water, it is expected to dissociate into silver (Ag+) and nitrate (NO3)~ ions. The
dissociation of silver nitrate into its component ions indicates that silver nitrate is not expected to volatilize from
water surfaces or bioconcentrate in aquatic organisms (Bodek et al., 1988). Ionic silver may form complexes with
hydroxide, sulfide ligands, halide ligands, and chelating organics (Bodek et al., 1988). Silver-organic complexes may
be important (Bodek et al., 1988). In aquatic systems with high halide concentrations, precipitation of insoluble
silver halides may occur (Bodek et al., 1988). Silver ions may sorb to organic matter and sediment that has high
C-54
-------
manganese dioxide, iron oxide, and clay content (Bodek et al., 1988). Nitrate is a minor constituent in natural waters,
where its concentration is limited by biological reactions that consume it (Bodek et al., 1988). In aquatic systems
where nitrogen is a limiting nutrient, high loadings of nitrate into surface waters can cause algal blooms (Bodek et al.,
1988). In natural waters with a low nitrate concentration, complexation with transition metals is not expected to be
an important process (Bodek et al., 1988).
B. Atmospheric Fate
If released to the atmosphere, silver nitrate's low vapor pressure indicates that it will exist as a paniculate (Bidleman,
1988). Wet and dry deposition of silver nitrate is expected to be the dominant fate process in the atmosphere
(Arimoto, 1989). Silver nitrate's high water solubility (Budavari et al., 1996) indicates that it is expected to undergo
wet deposition in rain, snow, or fog. The rate of dry deposition will depend on the prevailing winds and particle size
(Bodek et al., 1988). Pure silver nitrate is not photosensitive (Cappel, 1997); however, trace amounts of organic
material promote its photodegradation (Budavari et al., 1996).
C. Terrestrial Fate
If released to soil, silver nitrate is expected to dissociate into its component ions in the presence of moisture. Silver
may adsorb to manganese dioxide, iron oxides, clays, and organic matter (Bodek et al., 1988); therefore, its rate of
migration through soil may be slow. The high boiling point, low vapor pressure, and low Henry's Law constant
expected for an ionic salt (SRC, 1998) indicates that silver nitrate will not volatilize from either moist or dry soil
surfaces. Inoic silver may form complexes with hydroxide, sulfide ligands, halide ligands, and chelating organics
(Bodek et al., 1988). Nitrate ions may be converted to gaseous N2 or nitrous oxide (N2O) by microorganisms under
anaerobic conditions or may be assimilated by plants (Bodek et al., 1988). Sorption of nitrate ions by soils is
generally insignificant and therefore nitrate ions are expected to leach into groundwater (Bodek et al., 1988).
D. Summary
If released into water, silver nitrate will dissociate into silver and nitrate ions. Therefore, silver nitrate is not expected
to adsorb to suspended solids or sediment in the water column, bioconcentrate in aquatic organisms, or volatilize
from water surfaces. In natural waters, the concentration of nitrate is limited by biological reactions that consume it.
High loadings of nitrate into surface waters can cause algal blooms if nitrogen is a limiting nutrient. Silver nitrate is
expected to dissociate into its component ions in moist soils, and ionic silver may adsorb to manganese dioxide, iron
oxides, and clays. Nitrate is highly mobile in soils and therefore may leach into groundwater. Under anaerobic
conditions nitrate may be converted to gaseous N2 or nitrous oxide by microorganisms. Volatilization of silver
nitrate from soil surfaces is not expected to occur. If released to the atmosphere, silver nitrate is expected to exist as a
paniculate. Silver nitrate is expected to be physically removed from the atmosphere by wet and dry deposition. Dry
deposition will depend on particle size and prevailing wind patterns. Pure silver nitrate is not photosensitive and will
not degrade in sunlight; trace amounts of organic material promote silver nitrate's photodegradation.
C-55
-------
CHEMICAL SUMMARY FOR SODIUM HYDROXIDE
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of sodium hydroxide are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF SODIUM HYDROXIDE
Characteristic/Property Data Reference
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
1310-73-2
Caustic soda
HNaO
NaOH
white orthohombic crystals; hygroscopic
39.997
323°C
1388°C
571.9g/L
2.13g/cm3
not pertinent
no data; expected to be <10
no data; expected to be <1
no data; expected to be <10"6 mm Hg
when wet, attacks metals such as aluminum, tin, lead,
and zinc to produce flammable hydrogen gas
not flammable
not flammable
readily dissociates into Na+ and OH"
no data
no data
no data; expected to be <1X10"8
no data
not pertinent
CAS (1998)
Bodeketal.(1988)
Budavarietal. (1996)
Budavarietal.(1996)
Lide(1995)
Lide(1995)
Lide(1995)
Lide(1995)
Weastetal. (1985)
Lide(1995)
Weiss (1986)
SRC (1998)
SRC (1998)
Weiss (1986)
Weiss (1986)
Weiss (1986)
Weiss (1986)
SRC (1998)
SRC (1998)
Weiss (1986)
IL ENVIRONMENTAL FATE
A. Aquatic Fate
If sodium hydroxide is released into water, it will dissociate into sodium (Na+) and hydroxide (OH") ions (Bodek et
al., 1988). The dissociation of sodium hydroxide into its component ions indicates that sodium hydroxide is not
expected to volatilize from water surfaces or bioconcentrate in aquatic organisms. Because it is strongly basic,
sodium hydroxide will react with any protic acids to form salts. Hydroxide is the conjugate base of water;
protonation of hydroxide produces water. The presence of hydroxide in natural waters is entirely dependent on the pH
of the water, but massive amounts of sodium hydroxide may raise the pH of the receiving water. Metals present in
natural waters may form complexes with the hydroxide ion; complexes with transition metals will result in
C-56
-------
precipitation of the sparingly soluble metal hydroxides (Bodek et al., 1988). The sodium ion is expected to exist
predominately as the free ion in most natural waters (Bodek et al., 1988). Ion exchange processes with suspended
solids and sediment in the water column are expected to remove ionic sodium from solution; however, sodium binds
weakly to ion exchange sites and is expected to be displaced by other cations present in natural waters (Bodek et al.,
1988).
B. Atmospheric Fate
If sodium hydroxide is released to the atmosphere, it is expected to exist as a paniculate based upon the low vapor
pressure expected for this compound. Wet deposition of sodium hydroxide (Arimoto, 1989) in rain, snow, or fog is
expected to be the dominant fate process in the atmosphere based upon its high water solubility (Budavari et al.,
1996); however, carbon dioxide dissolved in atmospheric water may react with sodium hydroxide to form sodium
carbonate.
C. Terrestrial Fate
If sodium hydroxide is released to soil, it is expected to dissociate into its component ions in moist soils and react
with any protic acids present in soil to form the sodium salt and water. The low vapor pressure and low Henry's Law
constant expected for an ionic salt indicates that sodium hydroxide will not volatilize from either moist or dry soil
surfaces. In soil, ion exchange processes are important in retarding the mobility of sodium ions, however they may be
replaced by other soil cations since the sodium ion is held weakly by soils (Evans, 1989).
D. Summary
If released into water, sodium hydroxide will dissociate into sodium and hydroxide ions. The dissociation of sodium
hydroxide into its component ions indicates that sodium hydroxide is not expected to volatilize from water surfaces
or bioconcentrate in aquatic organisms. The hydroxide ion will react with protic acids to form water. Massive
amounts of sodium hydroxide may raise the pH of the water. The sodium ion is expected to participate in ion
exchange reactions with charged surfaces of suspended sediments and sediment in the water column. If released to
soil, sodium hydroxide is expected to dissociate into its component ions in moist soils and react with protic acids to
form water. Sodium hydroxide is not expected to volatilize from moist or dry soil surfaces. The mobility of sodium
ions will be retarded by ion exchange processes with charged surfaces of soil particles. However, since the sodium
ion is held weakly by ion exchange processes, it may leach into groundwater. If released to the atmosphere, sodium
hydroxide is expected to exist as a paniculate based upon the low vapor pressure expected for an ionic compound.
Sodium hydroxide reacts with carbon dioxide to form sodium carbonate. Wet deposition in rain, snow, or fog is
expected to be the dominant fate process in the atmosphere based upon sodium hydroxide's high water solubility.
C-57
-------
CHEMICAL SUMMARY FOR SODIUM HYPOPHOSPHITE AND
SODIUM HYPOPHOSPHITE MONOHYDRATE
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of sodium hypophosphite and its monohydrate are summarized
in Tables 1 and 2, respectively.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF SODIUM
HYPOPHOSPHITE
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Data
7681-53-0
Phosphinic acid, sodium salt
H2NaO2P
NaH2PO2
colorless, pearly, crystalline plates or white granular
powder
87.98
no data
decomposes
approximately 500 g/L '
no data
no data
no data; expected to be <10
no data; expected to be <1
no data; expected to be < 10"6 mm Hg
Explosion risk when mixed with strong oxidizing
Reference
CAS (1998)
Budavarietal. (1996)
Budavarietal. (1996)
Budavarietal. (1996)
Lewis (1993)
Budavarietal. (1996)
Dean (1985)
Estimated
Estimated
Estimated
Estimated
Lewis (1993)
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
agents.
no data
no data
2.1 (phosphinic acid)
no data
no data
no data; expected to be <1X10"8
no data
no data
Fee etal. (1996)
Estimated
1 Estimated from a reported solubility of 100 parts in 100 parts at 25 °C for the monohydrate (Dean 1985).
C-58
-------
TABLE 2. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF SODIUM
HYPOPHOSPHITE MONOHYDRATE
Characteristic/Property
Data
Reference
CAS No.
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
10039-56-2
NaPH2CyH2O
NaPH2CyH2O
white, monoclinic
105.99
loses water at 200° C
decomposes
approximately 500 g/L'
no data
no data
no data; expected to be <10
no data; expected to be <1
no data; expected to be < 10"6 mm Hg
no data
no data
no data
2.1 (phosphinic acid)
no data
no data
no data; expected to be <1X10"8
no data
no data
CAS (1998)
Dean (1985)
Dean (1985)
Dean (1985)
Dean (1985)
Dean (1985)
Dean (1985)
Estimated
Estimated
Estimated
Estimated
Fee etal. (1996)
Estimated
1 Estimated from a reported solubility of 100 parts in 100 parts at 25 °C (Dean 1985).
IL ENVIRONMENTAL FATE
A. Aquatic Fate
Almost all sodium salts are highly dissociated in natural waters (Bodek et al, 1988). Therefore, if sodium
hypophosphite is released into water, it is expected to initially hydrate to form the monohydrate then dissociate into
hypophosphite (H2PC)2~) and sodium (Na+) ions. The pKa of phosphinic acid indicates that hypophosphite will exist
mainly in the dissociated state in the environment. The dissociation of sodium hypophosphite into its component ions
indicates sodium hypophosphite will not volatilize from water surfaces or bioconcentrate in aquatic organisms. The
sodium ion is expected to exist predominately as the free ion in most natural waters (Bodek et al., 1988). Ion
exchange processes with suspended solids and sediment in the water column are expected to remove ionic sodium
from solution; however, sodium binds weakly to ion exchange sites and is expected to be displaced by other cations
present in natural waters (Bodek et al., 1988). No information specifically regarding the environmental fate of the
phosphinic acid or hypophosphite ion in water was located in the available literature. Phosphinic acid and its salts are
a strong reducing agents; they are oxidized to phosphonic acid or phosphonate (H3PO3 or HPO32") (Fee et al., 1996).
It is unclear how rapidly this process will occur in the environment.
B.
Atmospheric Fate
If sodium hypophosphite or its monohydrate are released to the atmosphere, it is expected to exist as a paniculate
based upon the low vapor pressure expected for this compound. Particulates of the unhydrated salt may also hydrate
when exposed to moisture in the atmosphere to form the monohydrate. Wet deposition of sodium hypophosphite in
C-59
-------
rain, snow, or fog is expected to be the dominant fate process in the atmosphere (Arimoto, 1989) based upon its high
water solubility (Betterman et al, 1991).
C. Terrestrial Fate
If sodium hypophosphite is released to soil, it is expected to initially hydrate to form the monohydrate then dissociate
into its component ions in moist soils. The pKa of phosphinic acid indicates that it will exist mainly in the
dissociated state in the environment. The low vapor pressure and low Henry's Law constant expected for an ionic salt
indicates that neither sodium hypophosphite nor its hydrate will volatilize from either moist or dry soil surfaces. In
soil, ion exchange processes are important in retarding the mobility of sodium ions, however they may be replaced by
other soil cations since the sodium ion is held weakly by soils (Evans, 1989). No information specifically regarding
the environmental fate of the phosphinic acid or hypophosphite ion in soils was located in the available literature.
Phosphinic acid and its salts are a strong reducing agents; they are oxidized to phosphonic acid or phosphonate
(H3PO3 or HPO32") (Fee et al., 1996). It is unclear how rapidly this process will occur in the environment.
D. Summary
If released into water, sodium hypophosphite and its hydrate are expected to dissociate into sodium and hypophosphite
ions. The dissociation of sodium hypophosphite into its component ions indicates that it will not volatilize from
water surfaces or bioconcentrate in aquatic organisms. The sodium ion is expected to participate in ion exchange
reactions with charged surfaces of suspended sediments and sediment in the water column. If released into soil,
sodium hypophosphite and its hydrate are expected to dissociate into its component ions in moist soils. As a result,
sodium hypophosphite is not expected to volatilize from moist soil surfaces. The mobility of sodium ions will be
retarded by ion exchange processes with charged surfaces of soil particles. However, since the sodium ion is held
weakly by ion exchange processes, it may leach into groundwater. Phosphinic acid and its salts are a strong reducing
agents; they are oxidized to phosphonic acid or phosphonate (H3PO3 or HPO32"). It is unclear how rapidly this process
will occur in either soil or water environments. The low vapor pressure expected for an ionic salt indicates that
neither sodium hypophosphite nor its monohydrate are expected to volatilize from dry soil surfaces. If released to the
atmosphere, the low vapor pressure expected for an ionic salt indicates that sodium hypophosphite will exist as a
paniculate in the ambient atmosphere. Wet and dry deposition will be the dominant fate process in the atmosphere.
C-60
-------
CHEMICAL SUMMARY FOR STANNOUS METHANESULFONIC ACID
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of stannous methanesulfonic acid are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF STANNOUS
METHANESULFONIC ACID
Characteristic/Property Data Reference
CAS No.
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Difrusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
53408-94-9 CAS (1998)
C2H8O6S2Sn SRC (1998)
[H3CS(O)(O)O]Sn[OS(O)(O)CH3] SRC (1998)
no data
310.89 SRC (1998)
no data
no data
no data
no data
no data
no data; expected to be <10 Estimated
no data; expected to be <1 Estimated
no data; expected to be <10"6mm Hg at25°C Estimated
no data
no data
no data
no data
no data
no data
no data; expected to be <10"8 Estimated
no data
no data
IL ENVIRONMENTAL FATE
A. Aquatic Fate
If stannous methanesulfonic acid is released into water, it is expected to dissociate into tin (Sn2+) and
methanesulfonate (CH3SO3~) ions. The dissociation of stannous methanesulfonic acid into its component ions
indicates that stannous methanesulfonic acid is not expected to bioconcentrate in aquatic organisms or volatilize from
water surfaces. Ionic tin may adsorb to charged surfaces of suspended sediments and humic materials in the water
column (Evans, 1989). Methanesulfonic acid has a pKa of -1.86 (Serjeant and Dempsey, 1979 as cited in
PHYSPROP, 1998) indicating that it will exist in the ionized at pH values typically encountered in the environment.
Therefore, volatilization of methanesulfonate from water surfaces is not expected to be an important fate process.
Methanesulfonate ions may adsorb to charged surfaces of suspended solids and sediment in the water column,
C-61
-------
although the importance of this process in the environment is not known. Limited data indicate that biodegradation of
methanesulfonate may be an important fate process (HSDB, 1998). An estimated BCF of 3 for methane surfonic acid
(Meylan et al., 1997) suggests the potential for bioconcentration in aquatic organisms is low (Franke et al., 1994).
B. Atmospheric Fate
If stannous methanesulfonic acid is released to the atmosphere, the low vapor pressure expected for an ionic salt
indicates that it will exist as a paniculate. Dry deposition of stannous methanesulfonic acid is expected to be the
dominant fate process in the atmosphere (Arimoto, 1989). The rate of dry deposition will depend on the prevailing
winds and particle size (Bodek et al., 1988). Wet deposition of stannous methanesulfonic acid may occur (Arimoto,
1989) in rain, snow, or fog.
C. Terrestrial Fate
If stannous methanesulfonic acid is released to soil, it is expected to dissociate into its component ions in moist soils.
The dissociation of stannous methanesulfonic acid into its component ions in moist soils indicates that stannous
methanesulfonic acid is not expected to volatilize from moist soil surfaces. The low vapor pressure expected for an
ionic salt indicates that stannous methanesulfonic acid is not expected to volatilize from dry soil surfaces. Ionic tin
may adsorb to charged surfaces of soil particles or form inner sphere complexes with variable-charge soil surfaces
(Evans, 1989) and therefore its rate of migration through soil may be slow. Methanesulfonic acid has apKaof-1.86
(Serjeant and Dempsey, 1979 as cited in PHYSPROP, 1998) indicating it will exist in the ionized form in moist soils
in the environment. Therefore, volatilization of methanesulfonate from moist soil surfaces will not occur.
Methanesulfonate ions may adsorb to charged surfaces of soil particles, however the importance of this process in the
environment is unknown. Limited data indicate that biodegradation of methanesulfonate may be an important fate
process (HSDB, 1998).
D. Summary
If released into water, stannous methanesulfonic acid is expected to dissociate into tin and methanesulfonate ions.
The dissociation of stannous methane sulfonic acid into it component ions indicates that stannous methanesulfonic
acid is not expected to bioconcentrate in aquatic organisms nor volatilize from water surfaces. Ionic tin may adsorb
to charged surfaces of suspended sediments and humic materials in the water column. Methanesulfonate ions may
adsorb to charged surfaces of suspended sediments and humic materials in the water column, however the importance
of this process in the environment is unknown. If released to soil, stannous methanesulfonic acid is expected to
dissociate into its component ions in moist soils. The dissociation of stannous methanesulfonic acid into its
component ions in moist soils indicates that volatilization from soil surfaces is not expected to be an important fate
process. Ionic tin may adsorb to charged surfaces of soil particles or form inner sphere complexes with variable-
charge soil surfaces and therefore its rate of migration through soil may be slow. Methanesulfonate ions may adsorb
to charged surfaces of soil particles, however the importance of this process in the environment is unknown. The low
vapor pressure expected for an ionic salt indicates that stannous methanesulfonic acid is not expected to volatilize
from dry soil surfaces. Limited data indicate that biodegradation of methanesulfonate may be an important fate
process. If released to the atmosphere, stannous methanesulfonic acid is expected to exist as a paniculate in the
ambient atmosphere based upon the low vapor pressure expected for an ionic salt. Wet and dry deposition will be the
dominant fate process in the atmosphere. The rate of dry deposition will depend on the prevailing winds and particle
size.
C-62
-------
CHEMICAL SUMMARY FOR SULFURIC ACID
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of sulfuric acid are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF SULFURIC ACID
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Data
7664-93-9
Battery acid
H204S
H2S04
colorless oily liquid
98.080
10.31°C
337°C
1000g/Lat25° C
1.8g/cm3
not pertinent
no data; expected to be <10
no data; expected to be <1
5.98X10-5mmHgat25°C
very reactive, dissolves most metals; concentrated
Reference
CAS (1998)
Weiss (1986)
Budavarietal. (1996)
Budavarietal. (1996)
Lide(1995)
Lide(1995)
Lide(1995)
Lide(1995)
Gunther et al. (1968) as cited in
PHYSPROP(1998)
Lide(1995)
Weiss (1986)
Estimated
Estimated
Daubert and Danner (1987)
Lewis (1993)
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Difiusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
acid oxidizes, dehydrates, or sulfonates most organic
compounds, often causes charring.
notflammable Weiss (1986)
notflammable Weiss (1986)
pKal = -3.00, pKa2= 1.99 Bodeketal. (1988)
no data
no data
no data; expected to be <1X10"8 Estimated
no data
greater than 1 mg/m3 Weiss (1986)
IL ENVIRONMENTAL FATE
A. Aquatic Fate
If sulfuric acid is released into the water column at low concentrations, a pKal of -3.00 (Bodek et al., 1988) indicates
sulfuric acid will dissociate into bisulfate (HSO4~) and hydrogen (H+) ions. In virtually all natural waters, the
bisulfate ion will also dissociate into sulfate (SO42~) and hydrogen ions based upon a pKa of 1.99 (Bodek et al., 1988).
Sulfuric acid will form salts with basic components in water. The dissociation of sulfuric acid into its component
ions indicates that sulfuric acid is not expected to volatilize from water surfaces orbioconcentrate in aquatic
C-63
-------
organisms. Sulfate ions may participate in oxidation-reduction reactions or react with cations present in the water
column to form soluble complexes or insoluble precipitates (Bodek et al., 1988). Sulfate-reducing microorganisms
are important mediators in redox reactions involving this ion (Bodek et al., 1988). Large releases of the concentrated
acid into water, such as may result from a spill, will result in a lowering of the pH (Bodek et al., 1988).
B. Atmospheric Fate
If sulfuric acid is released to the atmosphere, its vapor pressure (Daubert and Banner, 1987) indicates it will exist as a
paniculate in the ambient atmosphere. Wet deposition of sulfuric acid in rain, snow, or fog is expected to be the
dominant fate process in the atmosphere (Arimoto, 1989) based upon its high water solubility (Gunther et al., 1968
as cited in PHYSPROP, 1998). In the atmosphere, SO2 is oxidized to sulfuric acid (Graedel et al., 1986).
C. Terrestrial Fate
If sulfuric acid is released to soil, it will dissociate into sulfate and hydrogen ions in moist soils and will form salts
with basic soil components. The dissociation of sulfuric acid into its component ions indicates that volatilization
from moist soil surfaces is not expected to occur. Sulfate is generally weakly retained by soils (Bodek et al., 1988)
and therefore it may leach into groundwater. Adsorption of the sulfate ion may be important in humic soils
containing Al and Fe oxides (Bodek et al., 1988). Sulfuric acid's vapor pressure (Daubert and Banner, 1987)
indicates that volatilization from dry soil surfaces is not expected to be an important fate process.
D. Summary
If released into water, sulfuric acid will dissociate into sulfate (SO42~) and hydrogen (H+) ions. Therefore, sulfuric
acid is not expected to adsorb to suspended solids or sediment in the water column, bioconcentrate in aquatic
organisms, nor volatilize from water surfaces. Sulfate ions may participate in redox reactions or react with cations
present in the water column. Sulfate-reducing microorganisms have been identified as important mediators in redox
reactions involving the sulfate ion. Sulfuric acid will form salts with basic components in water. If released to soil,
sulfuric acid is expected to dissociate into its component ions in moist soils and will form salts with basic soil
components. The dissociation of sulfuric acid into its component ions indicates that volatilization from moist soil
surfaces is not expected to be an important fate process. In general, sulfate is weakly retained by soils and therefore it
may leach into groundwater. Adsorption of the sulfate ion may be important in soils with high organic matter content
or soils containing Al and Fe oxides. Sulfuric acid's vapor pressure indicates that volatilization from dry soil surfaces
is not expected to occur. If released to the atmosphere, sulfuric acid is expected to exist as a paniculate. Sulfuric
acid is expected to be physically removed from the atmosphere by wet deposition based upon its high water solubility.
C-64
-------
CHEMICAL SUMMARY FOR THIOUREA
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of thiourea are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF THIOUREA
Characteristic/Property
CAS No.
Common Synonyms
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Difrusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
Data
62-56-6
Thiocarbamide
Urea, 2-thio
CH,N2S
H2NC(=S)NH2
crystals
76.12
182°C
no data
201 g/Lat20°C
1.405g/cm3at25°C
no data
no data; estimated to be 2.8
-1.02
S.llXlO^mm Hg at25°C (extrapolated)
no data
no data
no data
no data
no data
no data
no data; estimated to be 1 .6X10"'
<0.2to<2incarp
no data
Reference
CAS (1998)
Lide(1995)
Howard and Neal (1992)
Lide(1995)
Budavarietal. (1996)
Budavarietal. (1996)
Lide(1995)
Lide(1995)
Yalkowsky and Dannenfelser (1992)
Lide(1995)
Meylanetal. (1992)
Hanschetal. (1995)
Daubert and Danner (1992)
Meylan and Howard (1991)
Chemicals Inspection and Testing
Institute (1992)
IL ENVIRONMENTAL FATE
A. Aquatic Fate
If thiourea is released into water, an estimated Koc value of 2.8 (Meylanetal., 1992) indicates that thiourea is not
expected to adsorb to suspended solids and sediment in the water column (Swann et al., 1983). According to a
classification scheme (Franke et al., 1994), BCFs of <0.2 and <2 in carp (Chemicals Inspection and Testing Institute,
1992) indicate that bioconcentration in aquatic organisms is low. An estimated Henry's Law constant of 1.6X10"7
atm mVmole at 25 ° C (Meylan and Howard, 1991) indicates that thiourea is expected to be essentially nonvolatile
C-65
-------
from water surfaces (Lyman et al., 1990). Thiourea has been demonstrated to be resistant to biodegradation in a
variety of standard biodegradation tests (HSDB, 1998). Thiourea reached 2.6% of its theoretical biological oxygen
demand over 2 weeks in the Japanese MITI test using an activated sludge seed and an initial chemical concentration of
30 mg/L (Chemicals Inspection and Testing Institute, 1992). In the OECD-screening test, 3% degradation was
observed (Schmidt-Bleek et al., 1982 as cited in HSDB, 1998) and 17% CO2 evolution was measured in a 5-day
German GSF Biodegradation Test (Rott et al., 1982 as cited in HSDB, 1998). Thiourea is stable to hydrolysis at
environmental pHs (Schmidt-Bleek et al., 1982 as cited in HSDB, 1998).
B. Atmospheric Fate
If thiourea is released to the atmosphere, an extrapolated vapor pressure of 3.1IX 10~4 mm Hg at 25 ° C (Daubert and
Danner, 1992) indicates that thiourea will exist as a gas in the ambient atmosphere (Bidleman, 1988). The rate
constant for the gas-phase reaction of urea with photochemically-produced hydroxyl radicals has been estimated to be
4.2X10"11 cmVmolecule-sec at 25 ° C (Meylan and Howard, 1993); this corresponds to a half-life of 9.2 hours.
C. Terrestrial Fate
If thiourea is released to soil, an estimated Koc value of 2.8 (Meylan et al., 1992) indicates that thiourea is expected
to have very high mobility in soils (Swann et al., 1983). Thiourea is not expected to volatilize from moist soil
surfaces (Lyman et al., 1990) based upon its estimated Henry's Law constant (Meylan and Howard, 1991) or from dry
soils based on its vapor pressure. Biodegradation of thiourea by soil microorganisms may be an important fate
process, although microflora activity may be suppressed for extended periods of time by high concentrations of this
compound (HSDB, 1998). Degradation of thiourea was also observed in sterilized soils (Kolyada, 1969 as cited in
HSDB, 1998) indicating that abiotic degradation may be an important fate process.
D. Summary
If released into water, thiourea is not expected to be adsorb to suspended solids and sediment in the water column.
Bioconcentration in aquatic organisms and volatilization from water surfaces are not expected to be important fate
processes. Several biodegradation tests indicate that thiourea may be resistant to biodegradation. Thiourea is stable
to hydrolysis at environmental pHs. If released to the atmosphere, thiourea is expected to exist as a gas in the ambient
atmosphere based upon its extrapolated vapor pressure. Gas-phase thiourea is expected to be degraded in the
atmosphere by reaction with photochemically-produced hydroxyl radicals; the half-life for this reaction in air has been
estimated to be 9.2 hours. If released to soil, thiourea is expected to have very high mobility and therefore may leach
into groundwater. Volatilization from moist or dry soil surfaces is not expected to be an important fate process.
Biotic and abiotic degradation of thiourea may be important fate processes, however, no rates were available for these
processes. High concentrations of thiourea may suppress the activity of soil microorganisms for extended periods of
time.
C-66
-------
CHEMICAL SUMMARY FOR TIN
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources. The
search identified sources of information for Tin.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of Tin are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF TIN
Characteristic/Property
CAS No.
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Difrusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
Data
7440-31-5
Tin white
Sn
Metal
118.69
231.9°C
2260°C
Insoluble
7.31g/mL
no data
no data
no data
no data
Flammable solid
no data
no data
no data
no data
no data
no data
no data
no data
Reference
Howard and Neal (1992)
Weast(1983)
Howard and Neal (1992)
Budavarietal. (1996)
Budavarietal. (1996)
Weast(1983)
Weast(1983)
Weast(1983)
Weast(1983)
Budavarietal. (1996)
C-67
-------
CHEMICAL SUMMARY FOR TIN CHLORIDE
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of tin chloride are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF TIN CHLORIDE
Characteristic/Property
CAS No.
Common Synonyms
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Data
7772-99-8
Tin (II) chloride
Stannous chloride
Cl2Sn
SnCl2
white orthorhombic crystals
189.615
247 °C
623°C
approximately 600 g/L '
3.90 g/cm3
no data
no data; expected to be <10
no data; expected to be <1
no data; expected to be < 10"6 mm Hg
violent reactions with BrF3, CaC2, ethylene oxide,
Reference
CAS (1998)
Lide(1995)
Lewis (1993)
Sax (1984)
Lide(1995)
Lide(1995)
Lide(1995)
Lide(1995)
Lide(1995)
Estimated
Lide(1995)
SRC (1998)
SRC (1998)
SRC (1998)
Sax (1984)
Flammability
Flash Point
Dissociation Constant
Molecular Diffusivity Constant
Air Difiusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
hydrazine hydrate, nitrates, K, Na, H2O2
no data
no data
expected to dissociate into Sn2+ and Cl"
no data
no data
no data; expected to be <1X10"8
no data
no data
SRC (1998)
SRC (1998)
1 Estimated from a reported solubility of 84 parts in 100 parts water (Dean, 1985).
IL ENVIRONMENTAL FATE
A. Aquatic Fate
Water hydrolyzes tin halides (Cotton and Wilkinson, 1980). Therefore, if tin chloride is released into water, it is
expected to dissociate into tin (Sn2+) and chloride (Cl") ions. In waters containing excess chloride ion, tin chloride is
expected to dissolve, yielding SnCl3" (Cotton and Wilkinson, 1980). As a result, tin chloride is not expected to
volatilize from water surfaces or bioconcentrate in aquatic organisms. Ionic tin may adsorb to charged surfaces of
C-68
-------
suspended sediments and humic materials in the water column (Evans, 1989). The chloride ion may complex with
heavy metals, thereby increasing their solubility (Bodek et al., 1988). Adsorption of the chloride ion to suspended
solids and sediment in the water column is not expected to be an important fate process.
B. Atmospheric Fate
If tin chloride is released to the atmosphere, the low vapor pressure expected for an ionic salt indicates that it will
exist as a paniculate. Dry deposition of tin chloride is expected to be the dominant fate process in the atmosphere
(Arimoto, 1989). The rate of dry deposition will depend on the prevailing winds and particle size (Bodek et al.,
1988). Tin chloride is expected to undergo wet deposition (Arimoto, 1989) in rain, snow, or fog due to its high
water solubility (Dean, 1985).
C. Terrestrial Fate
Water hydrolyzes tin halides (Cotton and Wilkinson, 1980). Therefore, if tin chloride is released to soil, it is expected
to dissociate into its component ions in moist soils. Ionic tin may adsorb to charged surfaces of soil particles or form
inner sphere complexes with variable-charge soil surfaces (Evans, 1989) and therefore its rate of migration through
soil may be slow. The dissociation of tin chloride into its component ions in moist soils indicates that tin chloride is
not expected to volatilize from moist soil surfaces. The low vapor pressure expected for an ionic salt indicates that tin
chloride is not expected to volatilize from dry soil surfaces. Chloride is extremely mobile in soils (Bodek et al.,
1988). The chloride ion may complex with heavy metals, thereby increasing their solubility (Bodek et al., 1988) and
potential for leaching into groundwater.
D. Summary
If released into water, tin chloride is expected to dissociate into tin and chloride ions. The dissociation of tin chloride
into its component ions indicates that tin chloride is not expected to volatilize from water surfaces or
bioconcentration in aquatic organisms. Ionic tin may adsorb to charged surfaces of suspended sediments and humic
materials in the water column. The chloride ion may complex with heavy metals, thereby increasing their solubility.
Adsorption of the chloride ion to suspended solids and sediment in the water column is not expected to be an
important fate process. If released to soil, tin chloride is expected to dissociate into its component ions in moist soils.
The dissociation of tin chloride into its component ions in moist soils indicates that tin chloride is not expected to
volatilize from moist soil surfaces. The low vapor pressure expected for an ionic salt indicates that tin chloride is not
expected to volatilize from dry soil surfaces. Ionic tin may adsorb to charged surfaces of soil particles or form inner
sphere complexes with variable-charge soil surfaces and therefore its rate of migration through soil may be slow. The
chloride ion is extremely mobile in soils; it may complex heavy metals, thereby increasing their solubility and the
potential to leach into groundwater. If released to the atmosphere, tin chloride is expected to exist as a paniculate in
the ambient atmosphere based upon the low vapor pressure expected for an ionic salt. Wet and dry deposition will be
the dominant fate process in the atmosphere. The rate of dry deposition will depend on the prevailing winds and
particle size.
C-69
-------
CHEMICAL SUMMARY FOR UREA
This summary is based on information retrieved from a systematic search limited to secondary sources. The only
exception is summaries of studies from unpublished TSCA submissions that may have been included. These sources
include online databases, unpublished EPA information, government publications, review documents, and standard
reference materials. No attempt has been made to verify information in these databases and secondary sources.
I. CHEMICAL IDENTITY AND PHYSICAL/CHEMICAL PROPERTIES
The chemical identity and physical/chemical properties of urea are summarized in Table 1.
TABLE 1. CHEMICAL IDENTITY AND CHEMICAL/PHYSICAL PROPERTIES OF UREA
Characteristic/Property
CAS No.
Common Synonyms
Common Synonyms
Molecular Formula
Chemical Structure
Physical State
Molecular Weight
Melting Point
Boiling Point
Water Solubility
Density
Vapor Density (air =1)
Koc
Log Kow
Vapor Pressure
Reactivity
Flammability
Flash Point
Data
57-13-6
Carbamide
Carbonyldiamide
CH,N2O
H2NC(=O)NH2
Tetragonal prisms
60.06
132.7°C
decomposes
545g/Lat25°C
1.3230 g/cm3 at 20°C
not pertinent
8
-2.11
1 .2X 1 0'5 mm Hg at 25 ° C (extrapolated)
no reaction with water or common materials
not flammable
not flammable
Reference
CAS (1998)
Lide(1995)
Budavarietal. (1996)
Lide(1995)
Budavarietal. (1996)
Budavarietal. (1996)
Lide(1995)
Lide(1995)
Lide(1995)
Yalkowsky and Dannenfelser (1992)
Lide(1995)
Weiss (1986)
Hance (1965) as cited in HSDB (1998)
Hanschetal. (1995)
Jones (1960) as cited in PHYSPROP
(1998)
Weiss (1986)
Weiss (1986)
Weiss (1986)
Dissociation Constant
Molecular Diffusivity Constant
Air Diffusivity Constant
Henry's Law Constant
Fish Bioconcentration Constant
Odor Threshold
no data
no data
no data
no data; estimated to be less than IX 10"8
not pertinent
PHYSPROP (1998)
Freitag et al. (1985) as cited in HSDB
(1998)
Weiss (1986)
IL ENVIRONMENTAL FATE
A. Aquatic Fate
If urea is released into water, a Koc value of 8 (Hance, 1965 as cited in HSDB, 1998) indicates that urea is not
expected to adsorb to suspended solids and sediment in the water column (Swann et al., 1983). According to a
classification scheme (Franke et al., 1994), a BCF of <10 in golden ide (Freitag et al., 1985 as cited in HSDB, 1998)
indicates that bioconcentration in aquatic organisms is low. An estimated Henry's Law constant of <1X10~8 atm
mVmole at 25 °C (PHYSPROP, 1998) indicates that urea is expected to be essentially nonvolatile from water
C-70
-------
surfaces (Lyman et al, 1990). In natural waters, biodegradation of urea is expected to be an important fate process;
ammonia and carbon dioxide have been identified as degradation products (HSDB, 1998). The rate of biodegradation
is expected to decrease with decreasing temperatures; at 8 ° C, negligible degradation was observed after incubation in
river water for 14 days, while at 20 ° C complete degradation was observed after 4 to 6 days incubation (Evans and
Patterson, 1973 as cited in HSDB, 1998). The presence of naturally-occurring phytoplankton in water is expected to
increase the rate of biodegradation (HSDB, 1998). Urea is used as an agricultural fertilizer (Lewis, 1993) and will be
taken up by plants as a source of nitrogen. Abiotic hydrolysis of urea occurs slowly yielding ammonium carbamate
(HSDB, 1998). At 5°C, 0.35% of urea hydrolyzed during a 10-day test period in demineralized/distilled water
(Atkinson, 1971 as cited in HSDB, 1998).
B. Atmospheric Fate
If urea is released to the atmosphere, a vapor pressure of 1.2X10~5mmHgat25°C (Jones, 1960 as cited in
PHYSPROP, 1998) indicates that urea will exist as both a paniculate and a gas in the ambient atmosphere (Bidleman,
1988). The rate constant for the gas-phase reaction of urea with photochemically-produced hydroxyl radicals has
been estimated to be 2.0X10"12 cmVmolecule-sec at 25 ° C (Meylan and Howard, 1993); this corresponds to a half-life
of 8.0 days. Particulate-phase urea is expected to be physically removed from the atmosphere by wet and dry
deposition (Arimoto, 1989).
C. Terrestrial Fate
If urea is released to soil, it is expected to hydrolyze to ammonia through soil urease activity (HSDB, 1998). The rate
of hydrolysis can range from 24 hours to weeks depending upon soil type, moisture content, and urea formulation
(Malhi and Nyborg, 1979 as cited in HSDB, 1998). Urea is used as an agricultural fertilizer (Lewis, 1993) and will
be taken up by plants as a source of nitrogen. While no specific studies were identified in the literature, it is
anticipated that urea will biodegrade rapidly in soil as has been reported in water. A Koc value of 8 (Hance, 1965 as
cited in HSDB, 1998) indicates that urea is expected to have very high mobility in soils (Swann et al., 1983). Urea is
not expected to volatilize from soil surfaces based upon its vapor pressure and estimated Henry's Law constant.
D. Summary
If released into water, urea is expected to be biodegraded yielding ammonia and carbon dioxide. Biodegradation is
expected to be more rapid in waters containing phytoplankton and during summer months when warmer water
temperatures prevail. Urea will be taken up by plants and used as a source of nitrogen. Bioconcentration in aquatic
organisms, adsorption to suspended solids and sediment in the water column, and volatilization from water surfaces
are not expected to be important fate processes. If released to the atmosphere, urea is expected to exist as both a
paniculate and as a gas based upon its vapor pressure. Gas-phase urea is expected to be degraded in the atmosphere
by reaction with photochemically-produced hydroxyl radicals; the half-life for this reaction in air has been estimated
to be 8.0 days. Particulate-phase urea is expected to be physically removed from the atmosphere by wet and dry
deposition. The rate of dry deposition will depend upon particle size and prevailing wind patterns. If released to soil,
urea is expected to hydrolyze to ammonia through the activity of soil urease as well as biodegrade as is the case in
water. The rate of hydrolysis can range from 24 hours to weeks depending upon soil type, moisture content, and urea
formulation. Urea is used an agricultural fertilizer as a source of nitrogen. Urea is expected to have very high
mobility in soils and therefore may leach into groundwater. Volatilization from moist and dry soil surfaces is not
expected to be an important fate process.
C-71
-------
REFERENCES
Abdelghani, A. A., A.C. Anderson, G.A. Khoury and S.N. Chang. 1990. Fate of ethylene glycol in the environment.
Louisiana Department of Transportation and Development in Cooperation with the US Department of Transportation
Federal Highway Administration. FHWA/LA-90/228.PB90-264672. Cited in HSDB, 1998.
Aldrich. 1996. Handbook Catalog of Fine Chemicals. Aldrich Chemical Co., Milwaukee, WI.
Aldrich. 1997. Handbook Catalog of Fine Chemicals. Aldrich Chemical Co., Milwaukee, WI.
AnbarM. andP. Neta. 1967. Title not available. Int JAppl Radiation Isotopes 18:493-523. Cited in HSDB, 1998.
Arimoto, R. 1989. Atmospheric deposition of chemical contaminants to the Great Lakes. J Great Lakes Res
15:339-56.
Atkinson, R. 1971. Some aspects of the use of urea for deicing airfield runways. Water Treat Exam 20: 193-203.
Cited in HSDB, 1998.
Atkinson, R. 1988. Estimation of gas-phase hydroxyl radical rate constants for organic chemicals. Environ Toxicol
Chem 7:435-42.
Atkinson, R. 1989. Title not available. JPhys Chem Ref Data. Monograph No. 1. Cited in HSDB, 1998.
ATSDR (Agency for Toxic Substances and Disease Registry). 1990. Toxicological Profile for Ammonia. U.S.
Department of Health and Human Services, Public Health Service. Contract No. 205-88-0608. Chamblee, GA.
125pp.
ATSDR (Agency for Toxic Substances and Disease Registry). 1993. Toxicological Profile for Fluorides, Hydrogen
Fluoride, and Fluorine (F). ATSDR/TP-91-17. ATSDR, Atlanta, GA.
Austin, S. and A. Glowacki. 1989. Hydrochloric acid. In: Ullmann 's Encyclopedia of Industrial Chemistry, 5th
Edition, Vol A13. VCH Publishers, Deerfield Beach, FL. pp. 283-96.
Baker, S.C., D.P. Kelly and J.C. Murrell. 1991. Microbial degradation of methane sulphonic acid: a missing link in
the biogeochemical sulphur cycle. Nature 350: 627-28. Cited in HSDB, 1998.
Betterman, G., G. Reiss, and T. Hofmann. 1991. Phosphorus Compounds, Inorganic. In: Ullmann's Encyclopedia
of Industrial Chemistry. VCH Publishers, New York, NY. p. 531.
Betterton, E.A. 1992. Henry's Law constants of soluble and moderately soluble organic gases. In: Gaseous
Pollutants: Characterization and Cycling, J.O. Nriagu, Ed. John Wiley & Sons, Inc. pp. 1-50.
Bidleman, T.F. 1988. Atmospheric processes. Environ Sci Technol 22: 361-67. Cited in HSDB, 1998.
Bodek, I., W.J. Lyman, W.F. Reehl and D.H. Rosenblatt. 1988. Environmental Inorganic Chemistry. Pergamon
Press, Elmsford, NY.
Boethling, R.S., P.H. Howard, W. Meylan, W. Stiteler, J. Beauman and N. Tirado. 1994. Group contribution method
for predicting probability and rate of aerobic biodegradation. Environ Sci Technol 28: 459-465.
Boublik, T., V. Fried and E. Hala. 1984. The Vapor Pressures of Pure Substances: Selected Values of the
Temperature Dependence of the Vapor Pressures of Some Pure Substances in the Normal and Low Pressure Region,
Vol. 17. Elsevier Sci. Publ., Amsterdam, Netherlands.
Bridie, A.L., C. J.M.Wolff and M. Winter. 1979. BOD and COD of some petrochemicals. WaterRes 13: 627-30.
Cited in HSDB, 1998.
C-72
-------
Budavari, S.,M.J. O'Neil, A. SmithandP.E. Heckelman (Eds). 1989. The Merck Index, llth Edition. Merck&Co.,
Inc., Rahway, NJ. p. 1360.
Budavari, S., MJ. O'Neil, A. Smith, P.E. Heckelman and J.F. Kinneary (Eds). 1996. The Merck Index: An
Encyclopedia of Chemicals and Drugs, 12th Edition. Merck and Co., Inc., Whitehouse Station, NJ.
Burrows, W.D. and R.S.Rowe. 1975. Title not available. J Water Pollut Control Fed 47: 92-93. Cited in HSDB,
1998.
Butler, J.A.V. and C.N. Ramchandani. 1935. The solubility of non-electrolytes. Part II. The influence of the polar
group on the free energy of hydration of aliphate compounds. JChemSoc 952-55. Cited in Howard and MeyIan,
1997.
Calvert, J.G. and J.N. Pitts. 1966. Photochemistry. John Wiley & Sons, Inc., New York, NY. pp 427-30. Cited in
HSDB, 1998.
Cappel, C.R. 1997. Silver Compounds. In: Kirk-Othmer Encyclopedia of Chemical Technology, 4th Edition, Vol.
22: 179-95. VCH Publishers, Deerfield Beach, FL.
CAS Registry On-line. 1998.
Chapman, E.G., D.S. Sklarew and J.S. Flickinger. 1986. Organic acids in springtime Wisconsin precipitation
samples. AtmosEnviron 20: 1717-27. Cited inHSDB, 1998.
CHEMFATE. 1995-1996. Syracuse Research Corporation's Environmental Fate Data Bases. Syracuse Research
Corporation, Syracuse, NY.
Chemicals Inspection and Testing Institute. 1992. Biodegradation and Bioaccumulation Data of Existing Chemicals
Based on the CSCL Japan. Japan Chemical Industry Ecology - Toxicology and Information Center. ISBN 4-89074-
101-12-50. p. 2-104.
Cohn, J.G. and E.W. Stern. 1994. Gold and Gold Compounds. In: Kirk-Othmer Encyclopedia of Chemical
Technology, 4th Edition, Vol. 12. John Wiley & Sons, Inc., New York. pp. 738-767.
Cote, A.E. 1997. Fire Protection Handbook, 18th Edition. National Fire Protection Association, Quincy, MA. p.
4-60.
Cotton, F.A. and F.R.S. Wilkinson. 1966. Advanced Inorganic Chemistry. Interscience Publishers, New York, NY.
Cotton, F. A. and G. Wilkinson. 1980. Advanced Inorganic Chemistry. A Comprehensive Text, 4th Edition. John
Wiley & Sons, New York, NY. p. 535.
Daubert, T.E. and R.P. Danner. 1985. Physical and Thermodynamic Properties of Pure Chemicals: Data
Compilation. Design Inst Physical Property Data. American Institute of Chemical Engineering. Hemisphere Pub.
Corp., New York, NY.
Daubert, T.E. and R.P. Danner. 1987. Data Compilation Tables of Properties of Pure Compounds. American
Institute of Chemical Engineers, V5.
Daubert, T.E. and R.P. Danner. 1989. Physical and Thermodynamic Properties of Pure Chemicals: Data
Compilation. Design Inst Physical Property Data. American Institute of Chemical Engineering. Hemisphere Pub.
Corp., New York, NY. Cited inHSDB, 1998.
Daubert, T.E. and R.P. Danner. 1991. Physical and Thermodynamic Properties of Pure Chemicals: Data
Compilation. Design Inst Physical Property Data. American Institute of Chemical Engineering. Hemisphere Pub.
Corp, New York, NY. 5 Vol.
C-73
-------
Daubert, T.E. and R.P. Banner. 1992. Data Compilation Tables of Properties of Pure Compounds. American
Institute of Chemical Engineers, V5.
Daugaut, et al. 1988. Title not available. Int J Chem Kinet 20: 331-338. Cited in HSDB, 1998.
Dean, J.A. 1985. Lange's Handbook of Chemistry, 13th Edition. McGraw-Hill Company, New York, NY.
Dias, F.F. and M. Alexander. 1971. Title not available. ApplMicrob 22: 1114-118. Cited in HSDB, 1998.
Dore, M, N. Brunei and B. Legube. 1975. Title not available. Trib Cebedeau 28:3-11. Cited in HSDB, 1998.
ECDIN (Environmental Chemicals Data and Information Network). 1998. On-lineaddress:
http://ulisse.etoit.eudra.org/Ecdin/E_hinfo.html.
EPI (Estimation Program Interface). 1994. Syracuse Research Corporation, Syuacuse, NY.
Evans, W.H. and S. Patterson. 1973. Water Res 7:97'5-85. As cited in Environment Canada Tech. Info, for Problem
Spills: Urea. p. 48. Cited in HSDB, 1998.
Evans, LJ. 1989. Chemistry of metal retention by soils. Environ Sci Technol 23: 1046-56.
Fee, D.C., D.R. Gard and C-H Yang. 1996. Phosphorus Compounds. In: Kirk-Othmer Encyclopedia of Chemical
Technology, 4th Edition, Vol. 18. J.I. Kroschwitz and M. Howe-Grant (Eds). John Wiley & Sons, New York. p.
777.
Fischer, W.K., P. Gerike and R. Schmid. 1974. Combination method of successive tests and estimation methods for
the biological degradability of synthetic substances, i.e. organic complex compounds, based on common applicable
summary parameters (BOD, CO2, COD, TOC). (GER). Wasser-und-Abwasser-Forchung 7: 99-118. Cited in
HSDB, 1998.
Fisher Scientific. 1985. Sodium citrate. Materials Safety Data Sheet.
Flick, E.W. (Ed). 1991. Industrial Solvents Handbook, 4th Edition. Noyes Data Corporation, Park Ridge, NY.
Franke, C., G. Studinger and G. Berger, et al. 1994. The assessment of bioaccumulation. Chemosphere 29: 1501-
14.
Franke, C., G. Studinger, G. Berger, S. Boehling, U. Bruckmann, D. Cohors-Fresenborg and U.E. Joehncke. 1994.
The assessment of bioaccumulation. Chemosphere 29: 1501-1514. Cited in HSDB, 1998.
Freitag, D., L. Ballhorn, H. Geyer and F. Korte. 1985. Environmental hazard profile of organic chemicals.
Chemosphere 14: 1589-616. CitedinHSDB, 1998.
Gaffney, J.S., G.E. Streit, W.D. Spall and J.H. Hall. 1987. Beyond Acid Rain - Do Soluble Oxidants and Organic
Toxins Interact with SO2 and Nox to Increase Ecosystem Effects. Environ Sci Technol 21:519-23. Cited in HSDB,
1998.
Gard, D.R. 1996. Phosphoric acid and phosphates. In: Kirk-Othmer Encyclopedia of Chemical Technology, 4th
Edition, Vol 18. John Wiley & Sons, New York, NY. pp. 669-718.
Gmelin. 1998. Gmelin Inorganic Database, GMEL9704. Crossfire Searching Software. Beilstein Chemiedaten und
Software GmbH. Online May 28.
Gordon, A.S. and F.J. Millero. 1985. Micro Ecol 11: 289-98. Cited in HSDB, 1998.
Covers, H., C. Ruepert, T. Stevens and C.J. Vanleeuwen. 1986. Experimental determination and prediction partition
coefficients of thioureas and their toxicity to photobacterium phosphoreum. Chemosphere 15:383-93. Cited in
C-74
-------
PHYSPROP.
Graedel, T.E., D.T. Hawkins and L.D. Claxton. 1986. Atmospheric Chemical Compounds: Sources, Occurrence, and
Bioassay. Academic Press, Inc., New York. p. 70.
Grafje, H., W. KornigandH.M. Weitz, et al. 1985. Butanediols, Butenediol, andButylendiol. In: Ullmann's
Encyclopedia of Industrial Chemistry, 5th Edition, Vol. A4. VCH Publishers, Deerfield Beach, FL. pp. 455-62.
Gregory, G.L., R.C. Harriss, R.W. Talbot, R.A. Rasmussen, M. Garstang, M.O. Andreae, R.R. Hinton, E.V. Browell
and S.M. Beck. 1986. Title not available. J GeophysRes 91: 8603-612. Cited in HSDB, 1998.
Grosjean, D. 1992. Formic acid and acetic acid: Emissions, atomospheric formation and dry deposition at two
Southern California locations. Atmospheric Environment 26:3279.
Gunther, F.A., W.E. WestlakeandP.S. Jaglan. 1968. ResRev 20: 1-148. Cited in PHYSPROP, 1998.
Hance,RJ. 1965. The adsorption of urea and some of its derivatives by a variety of soils. Weed Res 5: 98. Citedin
HSDB, 1998.
Hansch C., A. Leo, and D. Hoekman. 1995. Exploring QSAR: Hydrophobic, Electronic, and Steric Constants. ACS
Professional Reference Book. American Chemical Society, Washington, DC. Cited in HSDB, 1998.
Hartmann, W.R., M.O. Andreae and G. Helas. 1989. Title not available. Atmos Environ 23: 1531-33. Cited in
HSDB, 1998.
Hemphill, L. and W.S. Swansea 1964. Proceed of the 18th Industrial Waste Conf, Eng Bull. Purdue Univ,
Lafayette, IN. 18:204-17. Cited in HSDB, 1998.
Heukelekian, H. and M.C. Rand. 1955. Biochemical oxygen demand of pure organic compounds. J Water Pollut
ControlAssoc27: 1040-53. CitedinHSDB, 1998.
Hine, J. and P.K. Mookerjee. 1975. The intrinsic hydrophilic character of organic compounds: correlations in terms
of structural contributions. J Org Chem 40: 292-28.
Hoffman, W.A. and R.L. Tanner. 1986. Detection of Organic Acids in Atmospheric Precipitation. BNL-51922.
NTIS DE86 005294. Brookhaven National Laboratory. Environ Chem Div, Dept Appl Sci. pp.21. Cited in HSDB,
1998.
Howard, P.H. 1989. Handbook of Environmental Fate and Exposure Data for Organic Chemicals, Vol. \. Lewis
Publishers, Chelsea, MI. p. 377-81.
Howard, P.H. and M. Neal. 1992. Dictionary of Chemical Names and Synonyms. Lewis Publishers, Chelsea, MI.
Howard, P.H. and W.M. Meylan. 1997. Handbook oj'Physical Properties oj'Organic Compounds. Lewis
Publishers, Chelsea, MI.
HSDB. (Hazardous Substances Data Bank). 1995-1996. MEDLARS Online Information Retrieval System, National
Library of Medicine, National Toxicology Program (via TOXNET), Bethesda, MD.
HSDB (Hazardous Substances Data Bank). 1998. MEDLARS Online Information Retrieval System, National
Library of Medicine, National Toxicology Program (via TOXNET), Bethesda, MD. June.
IARC. (International Agency for Research on Cancer). 1985. Hydrogen Peroxide. IN: IARCMonographs on the
Evaluation of Carcinogenic Risk of Chemicals to Humans: Allyl Compounds, Aldehydes, Eposides and Peroxides,
Vol. 36. IARC, Lyon, France.
Jones, A.H. 1960. Sublimation-pressure data for organic compounds. J Chem Eng Data 5: 196-200. Citedin
C-75
-------
HSDB, 1998.
Kameya, T., T. Murayama, K. Urano andM. Kitano. 1995. Title not available. Sci Tot Environ 170: 43-51. Cited
in HSDB, 1998.
Keene, W.C. and IN. Galloway. 1984. Title not available. Atmos Environ 18: 2491-97. Cited in HSDB, 1998.
Keith, L.H. and D.B. Walters (Eds.) 1985. Compendium of Safety Data Sheets for Research and Industrial
Chemicals. VCH Publishers, Deerfield Beach.
Khwaja, H.A. 1995. Title not available. Atmos Environ 29: 127-39. Cited in HSDB, 1998.
Kolaitis, N., S.F.J. Bruynseel, R.E. Vangrieken and M.O. Andreae. 1989. Determination of methanesulfonic acid and
non-sea-salt sulfate in single marine aerosol particles. Environ Sci Technol 23: 236-40. Cited in HSDB, 1998.
Kolyada, T.I. 1969. Decomposition of thiourea in the soil. Vestsi Akad Navuk Belarus 3: 36-40. CitedinHSDB,
1998.
Lema, J.M., R. Mendez and R. Blazquez. 1988. Title not available. Water Air Soil Pollut 40: 223-250. Cited in
HSDB, 1998.
Lewis, RJ. 1993. Haw ley's Condensed Chemical Dictionary, 12th Edition. Van Nostrand Reinhold Company,
New York, NY.
Lide, D.R. 1995. CRC Handbook of Chemistry and Physics, 76th Edition. CRC Press, Boca Raton, FL.
Lockheed Martin (Lockheed Martin Energy Systems, Inc.). 1994. Material Safety Data Sheet for Sodium
Tetrafluoroborate.
Lockheed Martin (Lockheed Martin Energy Systems, Inc.). 1991. Materials Safety Reference Sheet.
Lyman, W.J., W.F. Reehl and D.H. Rosenblatt. 1990. Handbook of Chemical Property Estimation Methods.
American Chemical Society, Washington, DC. Cited in HSDB, 1998.
Malaney, G.W. and R.M. Gerhold. 1969. Structural determinants in the oxidation of aliphatic compounds by
activated sludge. J Water Pollut Control Fed 41: R18-R33. CitedinHSDB, 1998.
Malhi, S. S. and M. Nyborg. 1979. Rate of hydrolysis of urea (in soils) as influenced by thiourea and pellet size.
Plant Soil 51: 177-86. CitedinHSDB, 1998.
Manahan, S.E. 1991. Environmental Chemistry, 5th Edition. Lewis Publishers, Chelsea, MI.
Mawson A.J., R.I. Earle and V.F. Larsen. 1991. Title not available. Wat Res 25: 1549-54. CitedinHSDB, 1998.
Mazurek, M. A. and B.R.T. Simoneitt. 1986. Organic components in bulk and wet-oily precipitation. CRC Critical
Review. Environ Control 16: 140. CitedinHSDB, 1998.
McDermott, J. 1974. Plating of Plastics With Metals. Noyes Data Corporation. Park Ridge, New Jersey.
McGahey, C. and EJ. Bouwer. 1992. Biodegradation of ethylene glycol in simulated subsurface environments. Wat
Sci Technol26: 41-49. CitedinHSDB, 1998.
Meylan, W.M. and P.H. Howard. 1991. Bond contribution method for estimating Henry's Law constants. Environ
ToxicolChem 10: 1283-1293.
Meylan, W.M., P.H. Howard and R.S. Boethling. 1992. Molecular topology/fragment contribution method for
predicting soil sorption coefficients. Environ Sci Technol 26: 1560-67.
C-76
-------
Meylan, W.M. and P.H. Howard. 1993. Computer estimation of the atmospheric gas-phase reaction rate of organic
compounds with hydroxyl radicals and ozone. Chemosphere 26: 2293-2299.
Meylan, W.M., P.H. Howard, D. Aronson, H. Printup and S. Gouchie. 1997. Improved Method for Estimating
Bioconcentration Factor (BCF) from Octanol-Water Partition Coefficient. Third Update Report. Prepared for
Boethling, R.S., U.S. EPA.
Mills, EJ. and V.T.J. Stack. 1955. Title not available. Sewage Indust Wastes 27: 1061-1064. Cited in HSDB,
1998.
MSDS. Material Safety Data Sheet Archive at University of Vermont (www.siri.org, site 3).
Ohm, R.F. 1997. Rubber Chemicals. In: Kirk-Othmer Encyclopedia of Chemical Technology, 4th Edition., Vol.
21. John Wiley & Sons, New York, NY. pp. 460-81.
Osol, A. (Ed). 1980. Sodium citrate. In: Remington's Pharmaceutical Sciences, 16th Edition. Mack Printing Co.,
Easton,PA. p. 769.
Perrin, D.D. 1972. Dissociation Constants of Organic Bases in Aqueous Solution. IUPAC Chem Data Series,
Supplement 1972. Buttersworth, London.
PHYSPROP. 1998. Syracuse Research Corporation's Physical Properties Database in ISIS/Base format.
Fitter, P. 1976. Determination of biological degradability of organic substances. Water Res 10: 231-35. Cited in
HSDB, 1998.
Placak, O.R. and C.C. Ruchhoft. 1947. Studies of Sewage purification, XVII. The utilization of organic substrates
by activated sludge. Sewage Works J19: 423-40. Cited in HSDB, 1998.
Prager, J.C. 1995. Environmental Contaminant Reference Databook, Vol. I. Van Nostrand Reinhold, New York,
NY.
Price, K.S., G.T. Waggy and R.A. Conway. 1974. Brine shrimp bioassay and seawater BOD of petrochemicals. J
Water Pollut Control Fed 46: 63-77. Cited in HSDB, 1998.
Rentier, H. and M. Johns. 1989. Gold, Gold Alloys, and Gold Compounds. In: Ullman 's Encyclopedia of
Industrial Chemistry, 5th Edition, Vol. 12. VCH Publishers, New York, NY. pp. 499-533.
Rentier, H. 1993. Silver, Silver Compounds, and Silver Alloys. In: Ullmann 's Encyclopedia of In dustrial
Chemistry, 5th Edition, Vol. A24. VCH Publishers, Deerfield Beach, FL. pp. 107-67.
Riddick, J. A., W.B. Bunger and T.K. Sakano. 1986. Organic Solvents: Physical Properties and Methods of
Purification. Techniques of Chemistry, 4th Edition. Wiley-Interscience. New York, NY
Rott, B., R. Viswanathan, D. Freitag and F. Korte. 1982. Comparison of the applicability of various tests for
screening the degradability of environmental chemicals. Chemosphere 11: 531-8. Cited in HSDB, 1998.
Sansone, J.F., C.C. Andrews and M.Y. Okamoto. 1987. Title not available. Geochimica et Cosmochimica Acta 51:
1889-1896. Cited in HSDB, 1998.
Sax, N.I. 1984. Dangerous Properties of Industrial Materials, 6th Edition. Van Nostrand Reinhold Company, New
York, NY.
Sax, N.I. and R. J. Lewis, Sr. 1987. Hawley 's Condensed Chemical Dictionary. Van Nostrand Reinhold Company,
New York, NY.
Schmidt-Bleek, F., W. Haberland, A.W. Klein and S. Caroli. 1982. Steps towards environmental hazard assessment
C-77
-------
of new chemicals (including a hazard ranking scheme, based upon directive 79/83 I/EEC). Chemosphere 11: 383-
415. Cited in HSDB, 1998.
Serjeant, E.P. and B. Dempsey. 1979. lonisation Constants of Organic Acids in Aqueous Solution. IUPAC Chem
DataSerNo. 23. Pergamon Press, New York, NY. p. 989. Cited in HSDB, 1998.
Sigma-Aldrich Corporation. 1992. Material Safety Data Sheet for Sodium Tetrafluoroborate. Sigma-Aldrich,
Milwaukee, WI.
SRC (Syracuse Research Corporation). 1998. Estimated value based on structure-activity relationships.
Stuermer, D.H., D.J. Ng and C.J. Morris. 1982. Title not available. Environ Toxicol Chem 16: 582-87. Cited in
HSDB, 1998.
Sunderman, Jr., F.W. 1989. Mechanisms of nickel carcinogenesis. Scand J Work Environ Health 15: 1-12. Cited
inATSDR, 1997.
Sunderman, Jr., F.W. and A. Oskarsson. 1991. Nickel. In: Metals and Their Compounds in the Environment, E.
Merian,Ed. VCH Verlagsgesellschaft, New York, NY. pp. 1101-1126. Cited inATSDR, 1997.
Swarm, R.L., D. A. Laskowski, P. J. Mccall, K. Vanderkuy and H. J. Dishburger. 1983. A rapid method for the
estimation of the environmental parameters octanol/water partition coefficient, soil sorption constant, water to air
ratio, and water solubility. ResRev$5: 23. Cited in HSDB, 1998.
Swindol, C.M., C.M. Aelion, D.C. Dobbins, O. Jiang and F.K. Pfaender. 1988. Title not available. Environ Toxicol
Chem 7:291-99. Cited in HSDB, 1998.
Takemoto, S., Y. Kuge andM. Nakamoto. 1981. Title not available. Suishitsu Okaku Kenkyu 4: 80-90. Cited in
HSDB, 1998.
TRI93. 1995. Toxics Release Inventory, Public Data Release. Office of Pollution Prevention and Toxics, U.S. EPA,
Washington, DC.
U.S. Air Force. 1990. Copper - Elemental Copper. In: The Installation Restoration Toxicology Guide, Vol. 5.
Wright-Patterson Air Force Base, OH. pp: 77-1 - 77-44.
U.S. EPA. 1981. Treatability Manual I. Treatability Data. EPA-600/2-82-001A. U.S. EPA, Washington, DC.
U.S. EPA. 1987. Drinking Water Criteria Document for Copper. Environmental Criteria and Assessment Office,
Office of Health and Environmental Assessment, Environmental Criteria and Assessment Office, Cincinnati, OH.
ECAO-CIN-477.
U.S. EPA. 1993. Reregistration Eligibility Document (RED): Peroxy Compounds. EPA738-R-93-030. Office of
Pesticide Programs, Special Review and Reregistration Division.
Verschueren, K. 1996. Handbook of Environmental Data on Organic Chemicals, 3rd Edition. VanNostrand
Reinhold, New York, NY.
VonOepenB., W. KoerdelandW. Klien. 1991. Title not available. Chemosphere 22: 285-304. Cited in HSDB,
1998.
Weast, R.C. 1983. CRC Handbook of Chemistry and Physics. CRC Press, Boca Raton, FL.
Weast, R.C. (Ed). 1983-1984. CRC Handbook of Chemistry an d Physics, 64th Edition. CRC Press, Inc., Boca
Raton, Florida, p. B-139.
Weast, R.C., M.J. Astle and W.H. Beyer. 1985. CRC Handbook of Chemistry and Physics, 66th Edition. CRC
C-78
-------
Press, Boca Raton, FL. p. D-245.
Weast,R.C. 1986. CRC Handbook on Organic Compounds. (Cited in HSDS 1959).
Weiss, G. 1986. Hazardous Chemicals Data Book, 2nd Edition. Noyes Data Corporation, Park Ridge, NJ.
Winkeler, H.D.,U. PuttinsandK. Levsen. 1988. Title not available. Vom Wasser 70: 107-17. Cited in HSDB,
1998.
Yalkowsky, S.H. andR.M. Dannenfelser. 1992. Aquasol Database of Aqueous Solubility, Versions. College of
Pharmacy, Univ. of Ariz, Tucson, AZ. PC Version.
Yaws, C.L. 1994. Handbook of Vapor Pressure, Vol. 1 - Cl to C4 Compounds. Gulf Publishing Co, Houston, TX.
Zahn,R. and H.Wellens. 1980. Title not available. Z Wasser Abwasser Forsch 13: 1-7. Cited in HSDB, 1998.
C-79
-------
Appendix D
Supplemental Exposure
Assessment Information
-------
Technical Memorandum RE: Modeling Worker Inhalation Exposure
D-l
-------
D.I Technical Memorandum RE: Modeling Worker Inhalation Exposure
TECHNICAL MEMORANDUM
TO: Debbie Boger
PWB Project File, EPA # X823941-01-0
cc: Lori Kincaid, Jack Geibig, Dean Menke, Diane Perhac
FROM: Bruce Robinson, Chris Cox, Nick Jackson, Mary Swanson
DATE: December 22, 1995 (Revised 8/96)
RE: MODELING WORKER INHALATION EXPOSURE
I. INTRODUCTION
This technical memorandum is submitted for review by the RM2 work group. Air transport
models to estimate worker inhalation exposure to chemicals from printed wiring board (PWB)
making holes conductive (MHC) lines are presented here for review and comment. The purpose
is to reach agreement on our technical approach before proceeding with further analysis.
Three air transport models will be required to estimate worker exposure:
! Volatilization of chemicals induced by air sparging.
! Aerosol generation induced by air sparging.
! Volatilization of chemicals from the open surface of MHC tanks.
The total transport of chemicals from the air-sparged baths will be determined by summing the
releases calculated using each of the three models described above. Air-sparged baths include the
electroless-copper baths and some cleaning tanks. Only the third model will be applied to
determine the atmospheric releases of chemicals from unsparged baths. This document includes
a review of the relevant literature, descriptions of the models, and examples demonstrating the
proposed use of the models. The results of the model calculations will be compared to available
occupational monitoring data.
D-2
-------
II. VOLATILIZATION OF CHEMICALS FROM AIR-SPARGED PWB
MANUFACTURING TANKS
Mixing in plating tanks, e.g., the electroless copper plating tank, is commonly accomplished by
sparging the tank with air. This is similar to aeration in wastewater treatment plants, and the
volatilization of chemicals from these plants has been the focus of recent research. The
volatilization models used in that research are based on well accepted gas transfer theory,
discussed below.
Background
Volatilization of chemicals from water to air has been investigated by many researchers (Liss and
Slater, 1974; Smith etal, 1980; Roberts, 1983; Peng etal., 1993). In PWB manufacturing,
volatilization due to air sparging of process tanks is expected to be one of the main pathways for
contaminant transfer to the air. In bubble aeration systems, the volatilization rate is dependent
upon the volumetric gas flow rate, partial pressure of the gas, and the mass transfer rate
coefficient (Matter-Muller, 1981). The volatilization characteristics for different diffuser types
and turbulent conditions were evaluated by Matter-Muller (1981), Peng (1995), and Hsieh (1994).
Volatilization from aerated systems has been mainly quantified using the two-film theory (Cohen
et a/., 1978; Mackay and Leinonen, 1975). This work is discussed below and is used to model
chemical transfer rates from air-sparged PWB process tanks. The main assumption of the theory
is that the velocity at a fluid interface is zero. Molecular diffusion across the interfacial liquid film
is the limiting factor for mass transfer to the air, and it is used to develop a simple equation
relating the overall mass transfer coefficient to the diffusion coefficient of the chemical in water.
The two-film model of gas transfer was expanded to include mass transfer in diffused aeration
systems (Matter-Muller et a/., 1981). Matter-Muller et al. assumed that the system was
isothermal, hydraulic conditions were steady, and that pressure and volume changes within the
bubbles were negligible. Further, an overall mass transfer coefficient was applied to represent
transfer of contaminants to the bubble as they rose through the homogeneous liquid volume.
Parker (1993) demonstrated that liquid-phase concentration can be assumed constant during the
rise time of the bubble. Under these assumptions, Matter-Muller et al. derived the following
relationship predicting the mass transfer rate from an aerated system:
1-expl -
where:
FyiS = mass transfer rate of chemical y out of the system by sparging (m/t)
QG = gas flow rate (P/t)
Hy = dimensionless Henry' s constant for chemical y
cLj = concentration of chemical y in bulk liquid (m/13)
= overall mass transfer coefficient for chemical y (1/t)
D-3
-------
a = interfacial area of bubble per unit volume of liquid (I2/!3)
VL = volume of liquid (I3)
The overall mass-transfer coefficient is defined as the inverse sum of the reciprocals of the liquid
and gas-phase mass transfer coefficients; but, because molecular diffusion of oxygen and
nonpolar organic substances is 103 times greater in air than in water (Matter-Muller et a/., 1981), it
is set equal to the liquid phase coefficient only. The mass transfer coefficient of a chemical can
then be related to oxygen using the following equation:
ID
y_\ v
7^ \^OL,02
where:
Dy = molecular diffusion coefficient for chemical y in water (!2/t)
D02 = molecular diffusion coefficient for oxygen in water (!2/t)
= 2.1xlO-5 cm2/cm @ 25° C (Cussler, 1984)
KOLJ = overall mass transfer coefficient for chemical y (1/t)
K0L,o2 = overall mass transfer coefficient for oxygen in water (1/t)
The value of KOL,o2 at 25°C in diffused aeration systems can be estimated using a correlation
developed by Bailey and Ollis (1977):
1/3
"b (3)
where:
db = bubble diameter (1)
Pmo = density of water (m/13)
Pair = density of air (m/13)
g = gravitational constant (1/t2)
Hmo = vi scosity of water (m/1 -t)
If a measured value of Dy is not available, then it can be calculated from the Hayduk and Laudie
correlation (Lyman etal., 1982):
~ , 2, N 13.26x70 ~5
D (cm /sec)=
1.14 /0.589
m (4)
where:
Vm = molar volume of solute (cm3/mol)
|^H2o = viscosity of water (centipoise)
D-4
-------
The mass transfer coefficient can be corrected for the bath temperature (°C) as follows
(Tschabanoglous, 1991):
KOL,y,T=KOL,y,250cl.024(T-25> (5)
Bailey and Ollis (1977) developed a relationship for the interfacial area per unit volume (a) as a
function of the bubble diameter, gas flow rate, and tank geometry:
6 Qr th
a=. G b
VL db (6)
where:
h = tank depth (1); and
Tb
IS h
db (PH20-Palr)g (7)
Values of Hy are often reported at 25°C. The Henry's constant can be corrected to the bath
temperature using the van't Hoff equation:
H ~H exp
A// -A
gas
r>
HJ 1
{ 298.15 2'
1
G.is+r
(8)
where:
Ar|gas = enthalpy of the chemical in the gas phase (cal/mol)
AHaq = enthalpy of the chemical in the aqueous phase (cal/mol)
R = gas constant (1.987 cal/mol-K)
Matter-Muller (1981) concluded that surfactants do not significantly alter the rate of volatilization
from the water. Some agents did lower the overall mass transfer coefficient, but most showed no
appreciable difference. This was attributed to an increase in the specific interfacial area, a, when
the interfacial energy, or mass transfer coefficient, was decreased. The transfer rate of volatile
organic compounds (VOCs) was found to depend heavily upon the type of aerators used, and the
degree of saturation of the bubbles rising through the liquid.
III. AEROSOL GENERATION FROM BATHS MIXED BY SPARGING WITH AIR
Aerosols or mists have been identified as a major source of contaminants released by
electroplating baths to the atmosphere (Burgess, 1981) and should be investigated as a potential
source of contaminants from electroless baths. At least two sources of aerosols exist in
electroplating baths: 1) aerosols generated due to liquid dripping from parts as they are removed
D-5
-------
from the bath (drag-out drips); and 2) aerosols generated due to bursting of the bubbles at the
surface. Drag-out drips are insignificant compared to other sources of aerosols (Berglund and
Lindh, 1987; Cooper et al., 1993).
Bubbles in electroplating baths can originate from the dissociation of water at the electrode, or
mixing of the bath via air sparging. Bubbles in other plating baths (e.g., electroless plating baths)
can originate from reactions in the bath or mixing of the bath via air sparging. The rate of aerosol
generation per unit bubble volume decreases with increasing bubble size. Bubbles generated by
water dissociation are typically smaller than those generated by air sparging; therefore, aerosol
generation in electroless plating processes may be less significant than in electroplating
operations. The focus of this memo is aerosols generated by air sparging. Except for the
conductive polymer and non-formaldehyde electroless alternatives, MHC processes in PWB
manufacturing do not use electroplating and therefore would not dissociate water to form gas
bubbles. Information collection is continuing to allow prediction of aerosol formation in MHC
processes that do have an electroplating step. Importantly, Berglund and Lindh (1987) report that
aerosol generation from electroplating tanks is greatly reduced by sparging; the relatively large air
bubbles formed during air sparging coalesce the smaller bubbles formed by hydrolysis and
electroless plating reactions.
To estimate the emission of contaminants resulting from aerosols, the rate of aerosol generation
and the concentration of contaminant in the aerosol are required. Limited information
concerning the rate of aerosol formation was found in the literature. The following sources were
consulted:
! U.S. EPA (1991). Chemical Engineering Branch Manual for the Preparation of
Engineering Assessments.
I
Chemical Abstracts, 1986 to date.
! Current and past text books in air pollution, chemical engineering, and water and
wastewater treatment.
i
Perry's Handbook (1984) related to entrainment in distillation trays.
The last five years of Water Environment Research and ASCE Journal of the
Environmental Engineering Division.
A title key-word search of holdings in the library of the University of Tennessee.
The ASPEN model commonly used for modeling chemical manufacturing processes. (It
was found that any aerosol formation routines within ASPEN would be relevant to
entrainment in devices such as distillation trays and not relevant to sparging of tanks.)
The manager of the US EPA Center for Environmental Assessment Modeling in Athens,
Georgia, as well as an expert in the Air and Energy Lab - Emission Modeling Branch in
North Carolina.
D-6
-------
In this work, the aerosol formation rates will be predicted based upon limited measurements of
aerosol generation in electroplating (Berglund and Lindh, 1987) and other air-sparged baths
(Wangwongwatana et al., 1988; Wangwongwatana et al., 1990) found in the literature.
Berglund and Lindh (1987) developed several graphs relating aerosol generation to air sparging
rate (Figure la), bath temperature (Figure Ib), air flow rate above the bath (Figure Ic), and
distance between bath surface and the tank rim (Figure Id). Using Figures la-Id, the following
relationship may be developed:
[.. . i
5.5x10 5(Qr I 4+0.01 FT FA Fn
\^G ) J T A D (9)
where:
RA = aerosol generation rate (ml/min/m2)
QG/A = air sparging rate per unit bath area (1/min/m2)
FT = temperature correction factor
FA = air velocity correction factor
FD = distance between the bath surface and tank rim correction factor
Wangwongwatana et al. (1988) presented figures relating the number of aerosol droplets
generated as a function of air flow rate, bubble rise distance, bubble size, and colloid
concentration (Figure 2). Droplet size distribution measurements by these researchers indicate
volume mean diameters of 5 to 10 jim. The aerosol generation rate can be calculated using the
following equation:
„ _
(10)
where:
Cd = droplet concentration (I"3)
Vd = droplet volume (1)
A = bath area (I2)
Contaminants may be present in aerosols at elevated concentration relative to the bath
concentration. Colloidal contaminants may be collected on the bubble surface as it rises through
the bath. As the bubble bursts, the contaminants on the bubble surface are incorporated into
aerosols. Wangwongwatana et al. (1990) report that in their experiments about one in two
aerosols contain polystyrene latex spheres, compared to about one in 250 expected based upon
the concentration of latex sphere in the bath. Organic contaminants may also partition at the air-
water interface. A correlation for the water-interface partitioning coefficient for nonpolar
compounds, kiw, defined as the ratio of the mass of contaminant per unit area of interface to the
mass of contaminant per unit volume of water is given by Hoff et al. (1993):
log V=-8-58 -0.769 log C^ (n)
D-7
-------
where:
Csw = saturated aqueous solubility of the contaminant.
For more polar compounds a more complicated relationship is required:
log V = -7.508+log Jw+as(owa-osa-l35oJ/2303RT (12)
where:
yw = activity coefficient of the contaminant in water (dimensionless)
as = molar area of the solute (cm2/mol)
R = gas constant (8.314x107 erg/mol K)
aWA = surface tension of the water-air interface (dyne/cm)
aSA = surface tension of the solute-air interface (dyne/cm)
asw = surface tension of the solute-water interface (dyne/cm)
Hoff et al. (1993) also present a relationship for the ratio of the mass of contaminant sorbed at the
air-water interface to the mass of contaminant in the gas volume of the bubble:
Mi kiw
Mh Hy(dh I 6)
where:
M! = mass of contaminant at the interface
Mb = mass of contaminant in gas bubble
Only a small fraction of the bubble interface will be ejected as aerosols. It may be calculated
from the following equation:
= RA Adb
6 QG lb (14)
where:
fffi = fraction of bubble interface ejected as aerosols (dimensionless)
lb = thickness of bubble film (1)
The rate of mass transfer from the tank to the atmosphere by aerosols, FyiS (m/t) is given by:
Mj
Fy- = W/1E^ (15)
D-8
-------
IV. VOLATILIZATION OF CHEMICALS FROM THE OPEN SURFACE OF MHC
TANKS
Most plating tanks have a free liquid surface from which chemicals can volatilize into the
workplace air. Air currents across the tank will accelerate the rate of volatilization. The model
presented in the Chemical Engineering Branch Manual for the Preparation of Engineering
Assessments (CEBMPEA) (US EPA, 1991) has potential application in this case. Some
limitations of the model should be pointed out. The model was developed to predict the rate of
volatilization of pure chemicals, not aqueous solutions. The model was also validated using pure
chemicals. As a result, the model implicitly assumes that mass transfer resistance on the gas side
is limiting. The model may fail in describing volatilization of chemicals from solutions when
liquid-side mass transfer controls.
CEBMPEA models the evaporation of chemicals from open surfaces using the following model:
Fy,0 = 2 cL,y Hy A [Dyjairvz/(7iz)]°-5 (16)
where:
Fyj0 = volatilization rate of chemical y from open tanks (m/t)
Dyjair = molecular diffusion coefficient of chemical y in air (!2/t)
vz = air velocity (1/t)
z = distance along the pool surface (1)
The value of vz recommended by CEBMPEA is 100 ft-min"1. The value of Dyair can be estimated
by the following formula (US EPA, 1991):
Dyjair = 4.09xlO-5 TL9 (1/29 + 1/Mf5 M-°33/P, (17)
where:
Dyjair = molecular diffusion coefficient of chemical y in air (cm2/s)
T = air temperature (K)
M = molecular weight (g/mol)
P, = total pressure (atm)
This equation is based on kinetic theory and generally gives values of Dyjair that agree closely with
experimental data.
V. CALCULATION OF CHEMICAL CONCENTRATION IN WORKPLACE AIR
FROM EMISSION RATES
The indoor air concentration will be estimated from the following equation (US EPA, 1991):
Cy = Fy,T/(VRRvk) (18)
D-9
-------
where:
Cy = workplace contaminant concentration (m/13)
FyjT = total emission rate of chemical from all sources (m/t)
VR = room volume (P/t)
Rv = room ventilation rate (t"1)
k = dimensionless mixing factor
The mixing factor accounts for slow and incomplete mixing of ventilation air with room air.
CEBMPEA sets this factor to 0.5 for the typical case and 0.1 for the worst case. CEBMPEA
commonly uses values of the ventilation rate Q from 500 ft3/min to 3,500 ft3/min. Appropriate
ventilation rates for MHC lines will be chosen from facility data and typical industrial
recommendations.
VI. EXAMPLE MODELING OF FORMALDEHYDE RELEASE TO ATMOSPHERE
FROM AIR-SPARGED ELECTROLESS COPPER BATH
In the examples below, the values of some parameters are based upon a site visit to SM
Corporation in Asheville, NC. Except where stated otherwise, final values of the various
parameters used in the models will be chosen based on the results of the Workplace Practices
Questionnaire, chemical suppliers information, site visits, and performance demonstrations. All
parameter values are based on preliminary information and are subject to change.
Values of site-specific parameters assumed in the example
Tank volume = 242 L
Tank depth = 71 cm
Tank width = 48 cm
Tank length = 71 cm
Air sparging rate = 53.80 L/min
Tank temperature = 51.67°C
H2CO Concentration in tank = 7,000 mg/L
Bubble diameter at tank surface = 2.00 mm
Room length = 20 m
Room width = 20 m
Room height = 5 m
Air turnovers/hour = 4 hr"1
Air velocity across tank surface = 0.508 m/s
Dimensionless mixing factor = 0.5
Site visit to SM Co., Asheville, NC
Assumed
Assumed
Assumed
Midpoint of values given in Perry's Handbook,
1985, pg 19.13
Site visit to SM Co., Asheville, NC
Product data sheets
Assumed
Assumed
Assumed
Assumed
Assumed
Default recommended by US EPA, 1991
Default recommended by US EPA, 1991
Volatilization induced by air sparging
Calculating overall mass transfer coefficient for oxygen in water:
D-10
-------
1/3
£02
db
= 0.0113 cm/sec
= 0.678 cm/min
where:
db = 0.2 cm
pH2o = 0.997 g/cm3 (Dean, 1985)
pgas = 0.00118 g/cm3 (Dean, 1985)
g = 980 cm/sec2
|iH2o = 0.0089 (g/cm-sec) (Dean, 1985)
D02 = 2.1x1 Q-5 cm2/sec (Cussler, 1984)
Calculating molecular diffusion coefficient of formaldehyde in water:
13.26x10-5
1.14 T/0.589
_
y 1.14 T/0.5
V-H2O ym
= 1.81xlO-5cm2/sec
where:
Vm =36.8 cm3/mol
Hmo = 0.89 centipoise
Calculating mass transfer coefficient of formaldehyde in water:
^i^k^te^ii.o.™
= 0.584 cm/min
Correcting mass transfer coefficient for temperature:
KOL,y, 51.67 = KOL,y,25°c 1.024(T-25) = 0.584* 1.024(5L67-25) =1.10 cm/min
Calculating tb:
18 h
db
= 0.291 sec
= 4.85xlO-3min
D-ll
-------
where:
h = 71 cm
Calculating inter facial area per unit volume:
QQ tb
= 0.0323 cm2/cm3
where:
QG = 53,800 cm3/min
VL = 242,000 cm3
Correcting Henry's constant for temperature:
R \ 298.15 273.15+7J
= 1.99xlO"5 (dimensionless)
where:
Hy,25°C = 1.7xlO'7 atm-m3/mol (Risk Assistant, 1995)
= 6.38xlO"6 (dimensionless)
DHgas =-27,700 cal/mol
DHaq = -35,900 cal/mol
R =1.987cal/mol-K
Calculating mass transfer rate of formaldehyde by air sparging:
aVT
= 7.49 mg/min
The argument of the exponential function is -8031. This indicates that the formaldehyde
concentration in the air bubbles is essentially in equilibrium with the bath concentration.
Transport in aerosols
The aerosol generation rate will be estimated using data presented by both Berglund and Lindh
(1987) and Wangwongwatana et al. (1988).
Calculating aerosol generation rate using Berglund and Lindh (1987) data:
D-12
-------
RA = 5.5xlO-5(QG/A)+O.Ol FT FA FD
= 0.0187mL/min/m2
where:
QG/A = (53.8*10,000)7(71*48) = 158 (L/min/m2)
FT = 0.95 @ 5 1 .67°C (Figure Ib)
FA = 1.2 @ 0.508 m/s (Figure Ic)
FD =1.0 assumed (Figure Id)
Calculating aerosol generation rate using Wangwongwatana et al. (1988) data:
The air sparging rate used in the example (53.8 L/min) must be converted to an equivalent rate in
the experimental apparatus using the ratio of the area of the example bath (0.341 m2) to the area
of the experimental apparatus (0.123 m2). The equivalent rate is 19.4 L/min. The bubble rise
distance would be approximately 0.6 m. From Figure 2, it can be inferred that the droplet
concentration is not much greater than 100 droplets/cm3. The aerosol generation rate can now be
calculated:
= 8.27x1 0-3ml/m2/min
where:
QG =53800cm3/min
Cd =100 droplets/cm3
Vd =(p/6)dd3 = 5.24xlO-10cm3
dd = 0.001 cm (upper end of range reported by Wangwongwatana et al., 1988)
A = 0.341m2
The aerosol generation rates calculated by the two methods agree quite well. The model of
Berglund and Lindh (1987) will be used because it gives a slightly greater generation rate and is
easier to use.
Emission rate from bath. If it is assumed that the formaldehyde concentration in the aerosols is
equal to the bath concentration (7 mg/mL) then the formaldehyde emission rate is:
Fy,a = (7 mg/mL) • (0.0187 mL/m2/min) • (0.341 m2) = 4.46xlQ-2 mg/min
To determine if accumulation of the contaminant at the air- water interface is significant, kiw must
be estimated using Equation 1 1 . Since formaldehyde is a gas at the temperatures of interest,
interfacial tension data are not available; however, average values of other aldehydes may be used
(Hoff et al., 1993). Calculation of kIW@25°C is summarized below; information was not available
for calculating kiw at other temperatures.
D-13
-------
log V = -7.508+log jw+as(awa-asa-l.35aj I 2303RT
= -6.848
where:
YW = 1.452 Method 1, page 11-10 in Lyman et al. (1982)
as =9.35xl08cm2/mol Calculated from: as = 8.45x107 Vm2'3
R =8.314x10 7erg/molK
OWA = 72 dyne/cm Hoff et al. (1993)
OSA =21.9 dyne/cm Value for acetaldehyde, Weast, 1980
osw = 14.6 dyne/cm Average value for n-heptaldehyde and benzaldehyde, Girfalco
and Good, 1957
kiw = 1.418xlO-7cm
Formaldehyde emissions due to aerosols can now be calculated:
Calculating the ratio of contaminant mass sorbedat the air-water interface to mass in gas
volume of bubble:
Mi kiw
Mh Hy(dhl6)
= 0.2138
Calculating fraction of bubble interface ejected as aerosols:
RA A d,
f _ A b
•/arT&T
= 4.35xlO-3
where:
lb = 5xlO'7 cm (Rosen, 1978)
Calculating formaldehyde mass transfer rate via aerosols from tank to the atmosphere:
_M1
Fy,a = —
= 0.00697 mg/min
Volatilization from open tanks
Calculating molecular diffusion coefficient of formaldehyde in air:
Dyjair = 4.09x10-5 T1-9 (1/29 + 1/Mf5 M'033 / Pt
D-14
-------
= 0.174cm2/sec
where:
T = 298.15 K
M = 30.03 g/mol
P, = 1 atm
Calculating volatilization rate of formaldehyde from open tanks:
= 13.8 mg/min
where:
Dyjair = molecular diffusion coefficient of chemical in air (!2/t)
Vz = 0.508 m/sec
z = 0.48 m (shortest tank dimension gives highest mass transfer rate)
The gas side mass transfer coefficient (kg) in the above model is:
kg = 2[Dy,airvz/(pz)]a5
= 0.484 cm/sec
Thibodeaux (1979) reports a value of the liquid side mass transfer coefficient (k;) in large water
bodies of about 6xlO"4 cm/sec for wind speeds of 0.5 m/sec. Although not directly applicable to
the current situation, it can be used as a first estimate to determine the potential for liquid film
resistance to control the mass transfer rate.
Liquid side resistance = Hy/ ^ = 3.3xlO"2 sec/cm
Gas side resistance = l/kg = 2.1 sec/cm
It can be concluded that formaldehyde volatilization from open tanks is controlled by gas-side
mass transfer resistance; therefore, the CEBMPEA equation appears to be valid. It should be
noted that it may be necessary to consider liquid-side mass transfer resistance for chemicals with
larger Henry's constants. In this case the CEBMPEA model would not be valid.
Surprisingly, volatilization due to air sparging is less significant than that from open tanks.
Although the concentration of formaldehyde in the bubbles is high (virtually at equilibrium with
the formaldehyde concentration in the bath), the volume of air sparged is small compared to the
volume of room air flowing over the top of the tanks.
D-15
-------
Concentration of formaldehyde in workplace air
Cy =Fy,T/(VRRvk)
= 0.326 mg/m3
= 0.265 ppmv
where: FyT = 7.49 mg/min + 0.421 mg/min + 13.8 mg/min = 21.71 mg/min
VR = 20 m • 20 m • 5 m = 2000 m3
Rv =4 hr-1 = 0.0667 min4
k =0.5
VII. COMPARISON OF PREDICTED FORMALDEHYDE CONCENTRATIONS IN
WORKPLACE AIR TO MONITORING DATA
In this section, the concentrations of formaldehyde in the workplace air predicted by the model
are compared to available monitoring data. The purpose of the comparison is not to validate the
model but to determine if the modeling approach gives reasonable values of formaldehyde
concentration. Model validation would require calculation of formaldehyde concentrations using
the conditions specific to the monitoring sites. Such data are not available.
The results of an OSHA database (OCIS) search of monitoring data for formaldehyde (provided
by OPPT) include 43 measured air concentrations for 10 facilities in Standard Industrial
Classification (SIC) 3672 (printed circuit boards). The concentrations range from not detected to
4.65 ppmv. Most of the concentrations (37/42) range from < 0.04 to 0.6 ppmv, with all but one
less than 1.55 ppmv. Cooper et al. reports formaldehyde concentrations from three electroless
plating operations measured over a two day period. The mean concentrations ranged from 0.088
to 0.199 ppmv. The predicted concentration of formaldehyde in the workplace air was 0.263
ppmv. Thus the predicted value is within the range of concentrations determined by monitoring,
and less than the OSHA time-weighted-average concentration of 0.75 ppmv. The authors
conclude that the results are reasonable.
D-16
-------
REFERENCES
Bailey and Ollis. Biochemical Engineering Fundamentals. New York: McGraw-Hill, Inc., 1977.
Berglund, R. and E. Lindh. "Prediction of the Mist Emission Rate from Plating Baths." Proc.
Am. Electroplaters and Surface Finishers Soc. Annu. Tech. Conf., 1987.
Burgess, W.H. Recognition of Health Hazards in Industry: A Review of Materials and
Processes. New York: John Wiley and Sons, 1981.
Cohen, Y. and W. Cocchio. Laboratory Study of Liquid-Phase Controlled Volatilization Rates in
Presence of Wind Waves. Environ. Sci. Technol., 12:553, 1978.
Cooper, C.D., R.L. Wayson, J.D. Dietz, D. Bauman, K. Cheze and PJ. Sutch. Atmospheric
Releases of Formaldehyde from Electroless Copper Plating Operations. Proceedings of the
80th AESF Annual Technical Conference, Anaheim, CA. 1993.
Cussler, E.L. Diffusion: Mass Transfer in Fluid Systems. Cambridge: Cambridge University
Press, 1984.
Dean, J.A. (Ed). Lange's Handbook of Chemistry, 13th ed. New York: McGraw Hill, 1985.
Girifalco, L.A. and R.J. Good. "A Theory for the Estimation of Surface and Interfacial Energies:
I. Derivation and Application to Interfacial Tension." J. Phys. Chem., 61(7):904-909, 1957.
Hoff, J.T., D. Mackay, R. Gillham and W.Y. Shiu. "Partitioning of Organic Chemicals at the Air-
Water Interface in Environmental Systems." Environ. Sci. Technol, 27(10):2174-2180, 1993.
Hsieh, C., R. Babcock and M. Strenstrom. Estimating Semivolatile Organic Compound Emission
Rates and Oxygen Transfer Coefficients in Diffused Aeration. Water Environ. Research., 66:206,
1994.
Liss, P.S. and P.G. Slater. Flux of Gases Across the Air-Sea Interface. Nature, 247:181, 1974.
Lyman, W. J., W.F. Reehl and D.H. Rosenblatt. Handbook of Chemical Property Estimation
Methods, Washington DC: American Chemical Society, 1982.
Mackay, D. and PJ. Leinonen. Rate of Evaporation of Low Solubility Contaminants from Water
Bodies to Atmosphere. Environ. Sci. Technol., 9:1178, 1975.
Matter-Miiller, C., W. Gujer and W. Giger. Transfer of Volatile Substances from the Water to
the Atmosphere. Institute for Water Resources and Water Pollution Control (EAWAG), Swiss
Federal Institute of Technol., CH-8600 Dubendorf, Switzerland, 15:1271, 1981.
Parker, W., D. Thompson and J. Bell. Fate of Volatile Organic Compounds in Municipal
Activated Sludge Plants. Water Environ. Research, 65:58, 1993.
D-17
-------
Peng, J., J.K. Bewtra and N. Biswas. Transport of High-Volatility Chemicals from Water into
Air. Proceeding of 1993 Joint CSCE-ASCE National Conf. on Environmental Eng., 120:662,
1993.
Peng, J., J. Bewtra and N. Biswas. Effect of Turbulence on Volatilization of Selected Organic
Compounds from Water, Water Environ. Research, 67:000, 1995.
Perry, R.H., D.W. Green and J.O. Maloney (Eds). Perry's Chemical Engineers' Handbook, New
York: McGraw-Hill Book Company, 1984.
Risk Assistant Software. Alexandria, VA: Thistle Publishing, 1995.
Roberts, P.V., P. Dandliker and C. Matter-Muller. Volatilization of Organic Pollutants in
Wastewater Treatment-Model Studies, EPA-R-806631. U.S. EPA, Munic. Environ. Res. Lab.,
Cincinnati, Ohio, 1983.
Rosen, MJ. Surfactants andInterfacial Phenomena. New York: John Wiley & Sons, 1978.
Smith, J. H., D.C. Bomberger and D.L. Haynes. Prediction of the Volatilization Rates of
High-Volatility Chemicals from Natural Water Bodies, Environ. Sci. Technol, 14:1332, 1980.
Thibodeaux, LJ. Chemodynamics: Environmental Movement of Chemicals in Air, Water and
Soil. New York: John Wiley & Sons, 1979.
Tschabanoglous, G. andF.L. Burton. Wastewater Engineering: Treatment, Disposal, and
Reuse. New York: McGraw-Hill, Inc., 1991.
U.S. Environmental Protection Agency. Chemical Engineering Branch Manual for the
Preparation of Engineering Assessments. Washington, DC: U.S. EPA Office of Toxic
Substances. February 28, 1991.
Wangwongwatana, S., P.V. Scarpino and K. Willeke. "Liquid-to-Air Transmission of Aerosols
from a Bubbling Liquid Surface." J. Aerosol Sci., 19(7):947-951, 1988.
Wangwongwatana, S., P.V. Scarpino, K. Willeke and P.A . Baron. "System for Characterizing
Aerosols from Bubbling Liquids." Aerosol Sci. Technol., 13(3):297-307, 1990.
Weast, R.C. (Ed.) CRC Handbook of Chemistry and Physics, 61st ed. Boca Raton, FL: CRC
Press, 1980.
D-18
-------
Iv *
t- o s
o
o
JJ
si
a. v
= 3-
2*
.a
OJ
o
TJ
o
li
-o
01
Hi)
3
to
e'S-
e t
(.UI/UjUI/llU)
SUOISSIUI3 |OSOJ»V
o
u
2
3
OJG
jj
OJ>
o
cs
o
o
O)
o
)
D-19
-------
-£ I02
O
6
f
d
C
o
l_
(U
/"»
r—
1 10
jir?
Q.
O
20
Q = 8 Lpm
- to*/
2.6 S
.40 60 20
Bubble rise distance (cm)
Q = 13 Lpm
- 10
40
60
Figure 2. Effect of bubble rise distance on droplets number concentration. (From
Wangwongwatana et al., 1990)
D-20
-------
Appendix E
Drag-Out Model
-------
Contents
Summary of Non-conveyorized HASL Chemicals in Process Wastewater E-l
Summary of Conveyorized HASL Chemicals in Process Wastewater E-2
Summary of Non-Conveyorized Nickel/Gold Chemicals in Process Wastewater E-3
Summary of Non-conveyorized Nickel/Palladium/Gold Chemicals in Process Wastewater . . E-4
Summary of Non-conveyorized OSP Chemicals in Process Wastewater E-5
Summary of Convey orized OSP Chemicals in Process Wastewater E-6
Summary of Convey orized Immersion Silver Chemicals in Process Wastewater E-7
Summary of Non-conveyorized Immersion Tin Chemicals in Process Wastewater E-8
Summary of Convey orized Immersion Tin Chemicals in Process Wastewater E-9
Prediction of Water Quality From Printed Wiring Board Processes E-10
-------
Estimates of Drag-out, Wastewater and Surface Water Concentrations
Process Name:
Production Rate, sq.m./d:
Number of Process Tanks:
Plant WW Flowrate, L/d:
Stream Flow rate, L/d:
Non-conveyorized HASL
553
2
27911
13,300,000
Summary of all Chemicals in Process Wastewater
Chemical Name
1,4-Butenediol
Alkylakyne diol
Alkylaryl sulfonate
Alkylphenol ethoxylate
Alkylphenolpolyethoxyethanol
Aryl phenol
Citric acid
Copper sulfate pentahydrate
Ethoxylated alkylphenol
Ethylene glycol
Ethylene glycol monobutyl ether
Fluoboric acid
Gum
Hydrochloric acid
Hydrogen peroxide
Hydroxyaryl acid
Hydroxyaryl sulfonate
Phosphoric acid
Potassium peroxymonosulfate
Sodium benzene sulfonate
Sodium hydroxide
Sulfuric acid
Drag-out,
g/d
861
8.4
42
106
999
2.9
1679
3046
144
3087
1271
684
12
1157
3434
16
28
3391
6883
8.3
12
13132
Bath
Replacement,
g/d
507
4.7
23
59
558
1.7
937
1792
80
1731
709
382
6.8
646
2021
10
17
1893
4051
4.6
6.8
7543
Total in
Wastewater,
g/d
1368
13
65
165
1557
4.6
2616
4838
224
4818
1980
1066
18
1802
5454
26
45
5285
10934
13
18
20675
Concentration
in Wastewater,
mg/L
49
0.47
2.3
5.9
56
0.16
94
173
*
173
71
38
0.66
65
195
0.92
1.6
189
392
0.46
0.65
741
Stream
Concentration
w/o Treatment,
mg/La
0.10
0.00098
0.0049
0.012
0.12
0.00034
0.20
0.36
0.02
0.36
0.15
0.080
0.0014
0.14
0.41
0.0019
0.0034
0.40
0.82
0.00097
0.0014
1.6
Treatment
Efficiency,
%
90
0
93
86
90
90
90
Stream
Concentration
Following POTW
Treatment, mg/L
0.010
0.0049
0.014
0.051
0.015
0.041
0.082
a Numbers in bold indicate the estimated stream concentration (without wastewater treatment) that exceeds the aquatic toxicity concern concentration.
-------
Estimates of Wastewater and Surface Water Concentrations
Process Name:
Production Rate, sq.m./d:
Number of Process Tanks:
Plant WW Flowrate, L/d
Stream Flow rate, L/d:
Conveyorized HASL
1108
2
44829
13,300,000
Summary of all Chemicals in Process Wastewater
Chemical Name
1 ,4-Butenediol
Alkylakyne diol
Alkylaryl sulfonate
Alkylphenol ethoxylate
Alkylphenolpolyethoxyethanol
Aryl phenol
Citric acid
Copper sulfate pentahydrate
Ethoxylated alkyphenol
Ethylene glycol
Ethylene glycol monobutyl ether
Fluoboric acid
Gum
Hydrochloric acid
Hydrogen peroxide
Hydroxyaryl acid
Hydroxyaryl sulfonate
Phosphoric acid
Potassium peroxymonosulfate
Sodium benzene sulfonate
Sodium hydroxide
Sulfuric acid
Bath
Replacement,
g/d
1016
9.4
47
119
1118
3.4
1879
3593
161
3470
1422
766
14
1294
4050
19
33
3795
8120
9.3
14
15120
Concentration in
Wastewater,
mg/L
23
0.21
1.0
2.6
25
0.076
42
80
3.6
77
32
17
0.30
29
90
0.43
0.75
85
181
0.21
0.30
337
Stream
Concentration w/o
Treatment, mg/L a
0.076
0.00070
0.0035
0.0089
0.084
0.00025
0.14
0.27
0.0121
0.26
0.11
0.057
0.0010
0.097
0.30
0.0014
0.0025
0.28
0.61
0.00070
0.0010
1.1
Treatment
Efficiency,
%
90
0
93
86
90
90
90
Stream Concentration
Following POTW
Treatment, mg/L
0.0076
0.0035
0.0099
0.038
0.011
0.030
0.061
Numbers in bold indicate the estimated stream concentration (without wastewater treatment) that exceeds the aquatic toxicity concern concentration.
-------
Estimates of Drag-out, Wastewater and Surface Water Concentrations
Process Name:
Production Rate, sq.m./d:
Number of Process Tanks:
Plant WWFIowrate, L/d
Stream Flow rate, L/d:
Non-conveyorized Nickel/Gold
113.9
6
9595
13,300,000
Summary of all Chemicals in Process Wastewater
Chemical Name
Aliphatic acid A
Aliphatic acid B
Aliphatic acid E
Aliphatic dicarboxylic acid A
Aliphatic dicarboxylic acid C
Alkylamino acid B
Alkyl diol
Alkylphenolpolyethoxyethanol
Ammonia compound B
Ammonium chloride
Ammonium hydroxide
Citric acid
Copper sulfate pentahydrate
Ethoxylated alkylphenol
Hydrochloric acid
Hydrogen peroxide
Hydroxyaryl acid
Inorganic metallic salt A
Inorganic metallic salt B
Inorganic metallic salt C
Malic acid
Nickel sulfate
Palladium chloride
Phosphoric acid
Potassium compound
Potassium gold cyanide
Sodium hydroxide
Sodium hypophosphite mono hydrate
Sodium salt
Substituted amine hydroxhloride
Sulfuric acid
Transition metal salt
Urea compound B
Drag-out, g/d
136
20
306
96
45
337
581
206
1.0
745
480
134
627
12
7601
500
3.3
0.029
1.9
0.020
205
508
18
581
959
41
2.4
585
1229
818
2796
8.2
0.7
Bath
Replacement,
g/d
82
12
184
58
27
45
93
33
0.57
100
65
16
123
2.0
569
98
0.66
0.017
1.1
0.012
123
306
2.4
93
577
5.5
0.47
352
164
109
491
1.1
0.4
Total in
Wastewater,
g/d
219
32
491
154
73
383
673
239
1.5
845
545
150
750
14
8170
598
4.0
0.046
3.1
0.032
328
814
20
673
1535
46
2.8
936
1393
928
3287
9.3
1.1
Concentration in
Wastewater, mg/L
23
3.4
51
16
7.6
40
70
25
0.16
88
57
16
78
1.5
851
62
0.42
0.0048
0.32
0.0033
34
85
2.1
70
160
4.8
0.30
98
145
97
343
1.0
0.1
Stream Concentration
w/o Treatment, mg/L a
0.016
0.0024
0.037
0.012
0.0055
0.029
0.051
0.018
0.00011
0.064
0.041
0.011
0.056
0.0011
0.61
0.045
0.00030
0.0000035
0.00023
0.0000024
0.025
0.061
0.0015
0.051
0.12
0.0035
0.00021
0.070
0.10
0.070
0.25
0.00070
0.00008
Treatment
Efficiency, %
86
90
24
66
80
Stream Concentration
Following POTW
Treatment, mg/L
0.0079
0.0045
0.051
0.0045
0.014
a Numbers in bold indicate the estimated stream concentration (without wastewater treatment) that exceeds the aquatic toxicity concern concentration.
-------
Estimates of Drag-out, Wastewater and Surface Water Concentrations
Process Name:
Production Rate, sq.m./d:
Number of Process Tanks:
Plant WW Flowrate, L/d
Stream Flow rate, L/d:
Non-conveyorized Nickel/Palladium/Gold
86
8
12703
13,300,000
Summary of all Chemicals in Process Wastewater
Chemical Name
Aliphatic acid B
Aliphatic acid E
Aliphatic dicarboxylic acid A
Aliphatic dicarboxylic acid C
Alkylamino acid B
Alkyldiol
Alkylpolyol
Amino acid salt
Amino carboxylic acid
Ammonia compound A
Ammonia compound B
Ammonium hydroxide
Citric acid
Copper sulfate pentahydrate
Ethoxylated alkylphenol
Ethylenediamine
Hydrochloric acid
Hydrogen peroxide
Hydroxyaryl acid
Inorganic metallic salt B
Maleic acid
Malic acid
Nickel sulfate
Palladium salt
Phosphoric acid
Potassium compound
Potassium gold cyanide
Propionic acid
Sodium hydroxide
Sodium hypophosphite mono hydrate
Sodium salt
Substituted amine hydrochloride
Sulfuric acid
Surfactant
Transition metal salt
Urea compound B
Drag-out, g/d
15
308
72
34
451
438
389
21
10
513
0.72
615
124
474
9.3
46
1268
378
2.5
6.6
20
155
604
33
438
724
31
75
1.8
625
1548
618
1646
1.0
6.2
1.0
Bath
Replacement, g/d
9.2
186
44
21
61
70
892
1.4
23
69
0.44
83
15
93
1.5
105
159
74
0.50
13
47
93
365
74
70
437
4.1
171
0.35
463
166
83
324
2.3
0.83
0.62
Total in
Wastewater, g/d
24
494
116
55
512
509
1282
22
34
582
1.2
698
139
567
11
150
1427
452
3.0
19
67
248
969
107
509
1160
35
246
2.1
1088
1714
701
1970
3.4
7.0
1.7
Concentration in
Wastewater,
mg/L
1.9
39
9.1
4.3
40
40
101
1.7
2.7
46
0.091
55
11
45
0.85
12
112
36
0.24
1.5
5.3
20
76
8.4
40
91
2.7
19
0.17
86
135
55
155
0.27
0.55
0.13
Stream
Concentration w/o
Treatment, mg/L a
0.0018
0.037
0.0087
0.0041
0.038
0.038
0.096
0.0017
0.0025
0.044
0.000087
0.052
0.010
0.043
0.00081
0.011
0.11
0.034
0.00023
0.0015
0.0051
0.019
0.073
0.0080
0.038
0.087
0.0026
0.018
0.00016
0.082
0.13
0.053
0.15
0.00025
0.00053
0.00120
Treatment
Efficiency, %
86
90
82
24
80
a Numbers in bold indicate the estimated stream concentration (without wastewater treatment) that exceeds the aquatic concern concentration.
Stream Concentration
Following POTW
Treatment, mg/L
0.0060
0.0034
0.00026
0.055
0.011
-------
Process Name:
Production Rate, sq.m./d:
Number of Process Tanks:
Plant VWV Flow/rate, L/d
Stream Flow rate, L/d:
Non-Conveyorized OSP
686
3
21631
13,300,000
Summary of all Chemicals in Process Wastewater
Chemical Name
Acetic acid
Alkylaryl imidazole
Aromatic imidizole product b
Aryl phenol
Copper ion
Copper salt C
Copper sulfate pentahydrate
Ethoxylated alkyphenol
Ethylene glycol
Gum
Hydrochloric acid
Hydrogen peroxide
Hydroxyaryl acid
Hydroxyaryl sulfonate
Phosphoric acid
Sodium hydroxide
Sulfuric acid
Drag-out, g/d
4951
4054
519
3.6
4054
112
3778
74
3829
14
1639
1525
20
35
3497
14
21683
Bath
Replacement,
g/d
339
277
35
2.1
277
8
2225
42
2149
8
916
898
12
21
1954
8
12751
Total in
Wastewater,
g/d
5289
4332
554
5.7
4332
119
6003
116
5978
23
2555
2423
32
56
5451
23
34433
Concentration in
Wastewater,
mg/L
245
200
26
0.26
200
5.5
278
5.4
276
1.1
118
112
1.50
2.6
252
1.10
1592
Stream
Concentration w/o
Treatment, mg/La
0.40
0.33
0.042
0.00430
0.33
0.0089
0.45
0.0087
0.45
0.0017
0.19
0.18
0.0024
0.0042
0.41
0.0017
2.6
Treatment
Efficiency,
%
90
86
86
86
90
Stream Concentration
Following POTW
Treatment, mg/L
0.033
0.046
0.00130
0.063
0.018
Numbers in bold indicate the estimated stream concentration (without Wastewater treatment) that exceeds the aquatic toxicity concern concentration.
This ingredient not evaluated further as there was not enough information provided to identify a specific chemical.
-------
Process Name:
Production Rate, sq.m./d:
Number of Process Tanks:
Plant WW Flowrate, L/d
Stream Flow rate, L/d:
Conveyorized OSP
1500
3
32232
13,300,000
Summary of all Chemicals in Process Wastewater
Chemical Name
Acetic acid
Alkylaryl imidazole
Aromatic imidizole product b
Arylphenol
Copper ion
Copper salt C
Copper sulfate pentahydrate
Ethoxylated alkyphenol
Ethylene glycol
Gum
Hydrochloric acid
Hydrogen peroxide
Hydroxyaryl acid
Hydroxyaryl sulfonate
Phosphoric acid
Sodium hydroxide
Sulfuric acid
Bath
Replacement,
g/d
2963
2427
310
4.6
2427
67
4865
91
4699
18
2002
1964
26
45
4272
18
27877
Concentration in
Wastewater,
mg/L
92
75
10
0.14
75
2.1
151
2.8
146
0.6
62
61
0.81
1.4
133
0.57
865
Stream
Concentration w/o
Treatment, mg/l_a
0.22
0.18
0.023
0.00034
0.18
0.0050
0.36
0.0068
0.35
0.0014
0.15
0.15
0.0019
0.0034
0.32
0.0014
2.1
Treatment
Efficiency,
%
90
86
86
86
90
Stream Concentration
Following POTW
Treatment, mg/L
0.018
0.025
0.00070
0.051
0.015
Numbers in bold indicate the estimated stream concentration (without wastewater treatment) that exceeds the aquatic toxicity concern concentration.
This ingredient not evaluated further as there was not enough information provided to identify a specific chemical.
-------
Estimates of Wastewater and Surface Water Concentrations
Process Name: Conveyorized Immersion Silver
Production Rate, sq.m./d: 376
Number of Process Tanks 4
Plant WW Flowrate, L/d 8083
Stream Flow rate, L/d: 13,300,000
Summary of all Chemicals in Process Wastewater
Chemical Name
1,4-Butenediol
Alkylamino acid A
Fatty amine
Hydrogen Peroxide
Nitrogen acid
Nonionic Surfactant b
Phosphoric acid
Silver Nitrate
Sodium hydroxide
Sulfuric acid
Bath
Replacement,
g/d
390
1603
62
3462
281
345
2891
8.4
621
141
Concentration in
Wastewater,
mg/L
48
198
7.7
428
35
43
358
1.0
77
17
Stream
Concentration w/o
Treatment, mg/L a
0.029
0.12
0.0047
0.26
0.021
0.026
0.22
0.00063
0.047
0.011
Treatment
Efficiency,
%
90
95
90
96
Stream Concentration
Following POTW
Treatment, mg/L
0.0029
0.00023
0.026
0.000025
Numbers in bold indicate the estimated stream concentration (without wastewater treatment) that exceeds the aquatic toxicity concern concentration.
This ingredient not evaluated further as there was not enough information provided to identify a specific chemical.
-------
Estimates of Drag-out, Wastewater and Surface Water Concentrations
Process Name:
Production Rate, sq.m./d:
Number of Process Tanks:
Plant WW Flowrate, L/d
Stream Flow rate, L/d:
Non-conveyorized Immersion Tin
321
4
23624
13,300,000
Summary of all Chemicals in Process Wastewater
Chemical Name
Aliphatic acid D
Alkylalkyne diol
Alkylamino acid B
Alkylaryl sulfonate
Alkylimine dialkanol
Alkylphenol ethoxylate
Bismuth compound
Citric acid
Cyclic amide
Ethoxylated alkylphenol
Ethylene glycol monobutyl ether
Fluoboric acid
Hydrochloric acid
Hydroxy carboxylic acid
Methane sulfonic acid
Phosphoric acid
Potassium peroxymonosulfate
Quantenary alkylammonium chlorides
Silver salt
Sodium benzene sulfonate
Sodium phosphorus salt
Stannous methane sulfonic acid
Sulfuric acid
Thiourea
Tin chloride
Unspecified tartrate
Urea
Urea compound C
Vinyl polymer
Drag-out,
g/d
493
4.9
779
24
26
61
1.0
14599
1983
49
738
397
279
1633
15636
974
3996
922
0.15
4.8
3475
4352
10239
3799
544
973
3503
779
493
Bath
Replacement,
g/d
33
0.78
51
3.9
1.7
9.8
0.066
1056
131
7.8
118
63
18
108
1046
156
785
61
0.010
0.77
231
288
1325
251
36
64
231
51
33
Total in
Wastewater,
g/d
526
5.7
830
28
28
71
1.1
15655
2115
57
856
461
298
1741
16682
1130
4780
983
0.16
5.6
3706
4640
11564
4050
580
1037
3735
830
526
Concentration
in Wastewater,
mg/L
22
0.24
35
1.2
1.2
3.0
0.045
663
90
2.4
36
19
13
74
706
48
202
42
0.0067
0.24
157
196
490
171
25
44
158
35
22
Concentration
in Stream,
mg/La
0.039
0.00042
0.062
0.0021
0.0021
0.0054
0.000080
1.2
0.16
0.0042
0.064
0.035
0.022
0.13
1.3
0.085
0.36
0.074
0.000012
0.00042
0.28
0.35
0.87
0.30
0.044
0.078
0.28
0.062
0.039
Treatment
Efficiency,
%
0
93
90
90
90
40
90
40
90
Concentration in
Stream following
POTW
Treatment, mg/L
0.0021
0.082
0.0064
0.036
0.0074
0.21
0.030
0.026
0.0062
Numbers in bold indicate the estimated stream concentration (without wastewater treatment) that exceeds the aquatic toxicity concern concentration.
-------
Estimates of Wastewater and Surface Water Concentrations
Process Name:
Production Rate, sq.m./d:
Number of Process Tanks:
Plant WW Flowrate, L/d
Stream Flow rate, L/d:
Conveyorized Immersion Tin
226
4
8106
13,300,000
Summary of all Chemicals in Process Wastewater
Chemical Name
Aliphatic acid D
Alkylalkyne diol
Alkylamino acid B
Alkylaryl sulfonate
Alkylimine dialkanol
Alkylphenol ethoxylate
Bismuth compound
Citric acid
Cyclic amide
Ethoxylated alkylphenol
Ethylene glycol monobutyl ether
Fluoboric acid
Hydrochloric acid
Hydroxy carboxylic acid
Methane sulfonic acid
Phosphoric acid
Potassium peroxymonosulfate
Quantenary alkylammonium chlorides
Silver salt
Sodium benzene sulfonate
Sodium phosphorus salt
Stannous methane sulfonic acid
Sulfuric acid
Thiourea
Tin chloride
Unspecified tartrate
Urea
Urea compound C
Vinyl polymer
Bath
Replacement,
g/d
23
0.55
36
2.7
1.2
6.9
0.046
742
92
5.5
83
45
13
76
735
109
551
43
0.0069
0.54
163
202
932
176
25
45
163
36
23
Concentration in
Wastewater,
mg/L
2.8
0.067
4.5
0.34
0.15
0.85
0.0057
92
11
0.67
10
5.5
1.6
9.4
91
13
68
5.3
0.00086
0.067
20
25
115
22
3.1
5.6
20
4.5
2.8
Stream
Concentration w/o
Treatment, mg/L a
0.0017
0.000041
0.0027
0.00021
0.000092
0.00052
0.0000035
0.056
0.0069
0.00041
0.0062
0.0033
0.0010
0.0057
0.055
0.0082
0.041
0.0032
0.00000052
0.000041
0.012
0.015
0.070
0.013
0.0019
0.0034
0.012
0.0027
0.0017
Treatment
Efficiency,
%
90
Stream Concentration
Following POTW
Treatment, mg/L
0.0041
Numbers in bold indicate the estimated stream concentration (without wastewater treatment) that exceeds the aquatic toxicity concern concentration.
-------
PREDICTION OF WATER QUALITY
FROM PRINTED WIRING BOARD PROCESSES
Final Report to the University of Tennessee Center for Clean Products and
Clean Technologies and to the U.S. Environmental Protection Agency
Part of the Verification of Finishing Technologies Project
EPA Grant X825373-01-2 (Amendment No. 2)
By
Dr. R. Bruce Robinson
Dept. of Civil and Environmental Engineering
73 Perkins Hall, University of Tennessee, Knoxville, TN 37996
Office: 865/974-7730, FAX: 865/974-2669, E-Mail: rbr@utk.edu
Dr. Chris Cox
Dept. of Civil and Environmental Engineering
73 Perkins Hall, University of Tennessee, Knoxville, TN 37996
Office: 865/974-7729, FAX: 865/974-2669, E-Mail: ccox9@utk.edu
Jennie Ducker
Dept. of Civil and Environmental Engineering
73 Perkins Hall, University of Tennessee, Knoxville, TN 37996
August 6,1999
E-10
-------
TABLE OF CONTENTS
INTRODUCTION
Objectives
LITERATURE REVIEW
Pollutant Generation Rate and Waste Generation Volume
Drag-out Tests at Micom, Inc.
Other Published Drag-out Estimates
Discussions with Experts in the Surface Finishing Industry
Summary of Drag-out Studies
Drag-out Prediction Equations
Rinsing Theory
Other Rinsing Theory Studies
Printed Wiring Board Pollution Prevention and Control Technology
Water Use Rates from Survey of MHC Facilities
RESEARCH APPROACH
LABORATORY DRAG-OUT EXPERIMENTS
Apparatus
Procedure
Quality Assurance and Quality Control (QA/QC)
Results and Discussion
DRAG-OUT MODEL DEVELOPMENT
PWB WASTEWATER MODEL
COLLECTION AND ANALYSIS OF FIELD SAMPLES
Process Characterization
Sample Collection
Temperature
pH
Conductivity
Viscosity
Specific Gravity
Surface Tension
Metals Analysis
Quality Assurance and Quality Control (QA/QC)
Results from Analysis of Field Samples
DYNAMIC MASS BALANCE MODEL FOR INTERPRETATION OF FIELD DATA
E-ll
-------
MODEL VALIDATION
CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK
Conclusions
Recommendations
REFERENCES
LIST OF SYMBOLS
E-12
-------
INTRODUCTION
The Design for the Environment (DFE) Project Printed Wiring Boards (PWB) Cleaner
Technologies Substitutes Assessment: Making Holes Conductive (MHC) was performed by the
Center for Clean Products and Clean Technologies (CCPCT) at the University of Tennessee. The
project and results were well received by industry and the U.S. Environmental Protection
Agency. However, all parties agreed that one weakness in the project was the evaluation of
impacts of chemicals in the wastewater discharges of bath solutions from the MHC plating lines.
Evaluation of these impacts was more difficult than anticipated partly because of insufficient
information from surveyed facilities on the water quality of their discharges. Attempts at a mass
balance to predict chemical discharges were also unsatisfactory due to insufficient data on
chemical use and ultimate fate.
An estimate of the pollutants in the raw wastewater from PWB plating processes is needed in
order to evaluate health risks, impacts on the environment, impacts on municipal wastewater
plants, and overall manufacturing costs which includes treatment/disposal costs. The main
source of pollutants in the raw wastewater is the drag-out from the baths. Hence, drag-out is the
key variable for determining pollutant mass.
PWB facilities analyze at most only a couple of chemicals in their wastewater, and the facilities
generally have insufficient data to calculate chemical mass balances. Therefore, a different
approach is required to estimate the pollutant loads and wastewater quality of the PWB
wastewater discharges. This report discusses the development, validation, and use of predictive
tools to satisfy this need.
Objectives:
The objectives of this research were:
• Develop tools and methodologies to predict, but more importantly to compare the mass
of pollutants in the raw wastewater discharges from PWB plating processes.
• Validate these tools and methodologies against data available in the literature and against
samples collected at PWB facilities.
E-13
-------
LITERATURE REVIEW
Literature was identified through a computerized search on several key words. Additional papers
were found from the references in papers and from a manual search of recent Chemical
Abstracts (1998).
Pollutant Generation Rate and Waste Generation Volume
The sources of the pollutants in the wastewater generated in the MHC and surface finishing
processes for PWB manufacturing are the chemicals that escape from the process baths and from
other processes such as stripping racks of plating deposits. Our assumption for estimating the
pollutant mass generation rate, e.g., kg Cu/day, is that the source of the pollutants is
predominantly the drag-out from the process baths. Whatever chemicals are drug out of the
process tanks by solution adhering to the surface of the boards and racks will be removed in the
rinse tanks and ultimately end up in the raw wastewater discharge before any treatment or metals
recovery. This is consistent with the literature (Mooney 1991) and is expressed in a simple mass
balance:
fmass of pollutants^ fmass of pollutants^
• , = • • 1- 1
V in drag-out ) \in rinse discharge^
As discussed later, the etchant process baths themselves are generally not dumped into the
wastewater at the end of their useful life, but are typically sent off-site for processing. Other
process baths are apparently not sent off-site and do need to be accounted for in the waste
generation. Although pollutants from the stripping of racks may be significant at times, the
average mass pollutants originating from this process should be less than that contributed by
drag-out. Therefore, an estimate of the expected drag-out from various process tanks under
differing conditions is critical for estimating the waste mass generation rate. The arrangement of
the rinse tanks and the rinse flow rates will not change the total mass of contaminants released,
only the concentration and the volume of wastes. The waste generation volume primarily
depends on the rinse flow rates since this is the main source of wastewater discharge. If certain
assumptions are made, then conventional rinsing theory may be used to estimate the volume of
waste based on the drag-out and needed final rinse water quality. Importantly, the primary goal
of this work is a methodology that can be used to compare the relative amounts of wastes
generated from alternative PWB surface finishing manufacturing processes.
There are many references giving advice on minimizing drag-out and rinse water. Factors that
will reduce the drag-out include slow withdrawal from the process tank, longer drainage times,
tilting the boards so that the liquid drains to a corner, using drip shields, using drag-out/drag-in
tanks, as well as others. Solution density, viscosity, which depends on the bath chemistry and
temperature, and surface tension also affect how well the liquid drains off the boards, and hence
affects drag-out. Because of the number of variables which have complex relationships with
drag-out, estimating drag-out for a series of baths is a difficult, unsolved problem. The following
sections review what is known about estimating drag-out, including several references that
include predictive equations and experimental measurements.
E-14
-------
Drag-Out Tests at Micom, Inc.
The MnTAP/EPA Write study (Pagel 1992) at Micom, Inc. evaluated the ability of two
modifications to reduce waste from PWB surface finishing processes. At the time of the study,
Micom produced 92 -111 m2/day of double-sided and multilayered PWBs with the average
board being 0.46 m by 0.53 m and having 8000 holes. Micom had already implemented several
waste reduction measures, including countercurrent rinses, flow restrictors, softened water in the
rinses (softened water improved the rinsing and increased the efficiency of the ion exchange
waste treatment system), and air and mechanical agitation. However, Micom evaluated whether
changes to the way PWBs were transferred from process baths to the rinse tanks could further
reduce the amount of waste by reducing the drag-out.
Two processes were tested at Micom, Inc. in their MHC line: 1) a micro-etch bath and the
countercurrent rinse tanks following it; and 2) an electroless copper bath and the countercurrent
rinse tanks following it. The PWBs were moved from tank to tank in racks. The racks were 0.86
m high by 0.50 m wide by 0.33 m deep and could hold 24 boards. Typically, the operator
controlled a hoist and allowed the rack to drain for 3-5 seconds before going into the next tank.
The residence time was about 75 seconds in the micro-etch tank, 30 minutes in the electroless
copper tank, which held two racks at a time, and 2-3 minutes in each rinse tank.
The modifications evaluated at Micom were: 1) slowing the withdrawal rate of the racks from the
process bath; and 2) using an intermediate rack withdrawal rate combined with a longer drain
time over the process bath before going into the rinse tanks. Slowing the withdrawal rate was
achieved by lowering the speed of the motor on the mechanical hoist used to move the racks.
Installation of new equipment prohibited matching the withdrawal rates used in the first
modification with tests on the second modification, hence the designation of "intermediate"
withdrawal rate. Withdrawal time was defined as the time it took to raise the boards from the
bath to a height needed to clear the tank walls, a total of 0.91 m. Increasing the drain time was
achieved by the operator simply waiting longer before placing the boards in the next bath. Drain
time was defined from the moment that the rack cleared the water surface until half of the rack
was over the adjacent rinse tank. Measurement of drag-out was accomplished by shutting off the
rinse water and then measuring the increase in copper concentration after a known quantity of
boards had been rinsed. Copper was measured by atomic absorption spectrophotometry. The
electroless copper samples were preserved with a hydrochloric/nitric acid mixture rather than just
nitric, because copper precipitated out of solution as the solution cooled when nitric acid alone
was used. There were some analytical difficulties of unknown origin in that the copper
measurements done by an outside laboratory showed about 1800-2200 mg/L of copper whereas
Micom's laboratory analyses showed about 2400 mg/L.
Baseline drag-out measurements were made over a twelve day period using 136 samples for 12
pairs of racks. The first modification experiments were also made using 136 samples for 12 pairs
of racks, and the second modification experiments used 109 samples for 9 pairs of racks.
The results of the experiments are summarized in Tables 1 and 2. It should be noted that the
values for drag-out, withdraw rate, and drain time are averages of a rather broad range of values
grouped by relative magnitude by Page 1.
E-15
-------
Table 1. Drag-Out Test Results on the Microetch Bath at Micom, Inc.
Parameter
Drag-out, mL/m2
Withdrawal time, sec
Withdrawal rate, m/sec
Drain time, sec
Total time, sec
Surface area/rack, m2
Water flow rate, 1pm
Baseline
129
1.7
0.51
3.4
5.1
8.2
9.8
Slow Withdrawal Rate
72.1
14.9
0.056
2.5
17.4
7.7
—
Intermediate Withdrawal Rate
& Longer Drain Time
76.4
4.3
0.20
12.1
16.4
8.6
—
Table 2. Drag-Out Test Results on the Electroless Bath at Micom, Inc.
Parameter
Drag -out, mL/m2
Withdrawal time, sec
Withdrawal rate, m/min
Drain time, sec
Total time, sec
Surface area/rack, m2
Water flow rate, 1pm
Baseline
64.6
1.8
0.48
5.2
7.0
15.7
12.5
Slow Withdrawal
Rate
32.3
13.9
0.061
3.2
17.1
15.0
—
Intermediate Withdrawal Rate
& Longer Drain Time
31.4
4.3
0.175
11.9
16.3
16.3
—
For the micro-etch bath, the first modification reduced the drag-out by 45% while the second
modification reduced drag-out by 41%. For the electroless copper bath, the reductions were 50%
and 52%, respectively. Because it was easier for Micom to control the drain time than the
withdrawal rate, they implemented a longer drain time.
It should be noted that reducing the drag-out from the micro-etch affects the bath. This bath
removes copper until the etchants are exhausted. Make-up chemicals may be added to replace
etchant solution is lost in drag-out. Reducing drag-out may mean that the entire bath must be
replaced more frequently, because of increased copper build-up in the bath. However, Micom
preferred to retain the copper in the bath and replace the bath, because there is greater
opportunity to recover metals in the etchant bath than in the rinses. For the electroless bath,
drag-out reduction helps retain the chemicals in the bath and increase its life, providing that build-
up of impurities does not offset this advantage. Reduction of drag-out from upstream baths
would help in this regard.
E-16
-------
Other Published Drag-Out Estimates
Slip (1990) evaluated several ways to minimize drag-out, including the effect of the inclination
angle during drainage, the withdrawal rate, and the drainage time. Several experiments focused
on the inclination angle in the design of electroplating product holders and its effect on drag-out.
The holders were not for PWBs but apparently for a variety of electroplated products. The
holders typically had horizontal cross-braces or struts. Sup noted that the drag-out from the
holder could be as much as 50% of the total drag-out in these cases. Sup experimented with
holder designs that had struts of different angles and showed that drag-out could be reduced
significantly. The effect of the inclination angle of the struts on drag-out is shown in Table 3.
Struts tilted at a 45° angle to horizontal had only 36% of the drag-out as a horizontal one.
Table 3. Effect of Inclination Angle of the Product Holder Strut on Drag-Out
Angle to Horizontal
0°
15°
30°
45°
90°
Drag-Out
niL/m2
44
35
25
16
22
% of Maximum
100
80
57
36
50
Slip (1990) also experimented with chromium plated sheets suspended from the holders to
determine the effect of drainage time and inclination angle of the sheet. The experiments used
either 19-20 g/L or 240-250 g/L CrO3 electrolytes. The effect of drainage time and inclination
angle is shown in Table 4. (Note: the data reported in Table 4 were read from two graphs in Sup
(1990) and include representative data, but not all the data.). As seen in the table, a 45° inclination
angle had about 33% less drag-out at short drainage times compared to a horizontal angle and
nearly 50% less drag-out at long drainage times. An increase in the drainage time greatly
reduced drag-out up to about 20-30 seconds, but had a relatively small effect for longer times.
Further experiments were conducted on the effect of withdrawal rate and inclination angle of the
sheet. The effect of withdrawal rate is shown in Table 5. Slower withdrawal rates reduced the
drag-out, but not as much as inclination angle. A plate withdrawn at 60 m/min had roughly 25-
30% more drag-out volume than a plate withdrawn at 6 m/min. The drag-out volumes reported
by Sup are approximately a factor of two less than the drag-out volumes reported in the Micom
study (Pagel 1992) discussed above. One explanation for the difference may be that the boards
in the Sup study did not contain holes but the boards used in the Micom study did. It should be
noted that Sup was not clear how the drag-out was calculated. It appears to be American practice
to report the drag-out in terms of the area of one side of the board. It is possible that Sup
calculated his drag-out based on the area of both sides of the board, leading to numbers which
are half as large. If this were the case, then to be comparable to American practice, his drag-out
volumes should be doubled. However, in a later paper, Sup (1992) used an equation which was
developed for drag-out on the basis of one side of the board. It is likely that he was aware of the
assumptions built into the equation, and considering that his values are comparable to the Micom
study, we will assume that Sup's drag-out volumes are directly comparable to other values. In
either case, the trends are the same.
E-17
-------
Table 4. Effect of Drainage Time and Inclination Angle on Drag-Out.
Drainage
Time, s
0
10
20
30
45
60
Drag-Out, niL/m2
280-320 g/L CrO3,
0° angle, 40°C
57
28
22
20
19
19
280-320 g/L CrO3,
45° angle, 40°C
~
21
13
11
~
10
20 g/L CrO3,
0° angle, 20°C
64
33
28
25
21
19
20 g/L CrO3,
45° angle, 20°C
~
24
19
15
13
11
Table 5. Effect of Withdrawal Rate on Drag-Out.
Withdrawal Rate,
m/min
3.6
6
9
18
36
60
Drag-Out
240-250 g/L CrO3
(40±1°C)
niL/m2
17
22
24.5
26.5
27
28
19-20 g/L CrO3
(20±1°C)
niL/m2
21
26
29
32
33
33
In a second paper, Slip (1992) evaluated two drag-out prediction equations by comparing
measured volumes of drag-out to predicted values. The first equation was from Kushner (1951):
/ = 0.02
Eqn2
or:
E-18
-------
Eqn3
where:
f = film thickness, cm
|i = dynamic viscosity of electrolyte, g/(cnrs)
h = height of metal sheet
p = density of electrolyte, gm/cm3
tw = withdrawal time, s
v = kinematic viscosity, cm2/s
VA = withdrawal rate of metal sheet, cm/s
The second equation was:
Eqn4
where:
g = gravity, 981 cm/s2
tdr = drainage time, s
Experiments were conducted on 21.0 x 21.4 cm metal sheets which had no holes. The sheets
were withdrawn from the bath at 20 cm/s and allowed to drain for 10 seconds.
Neither of the two equations predicted the measured values very well. Sixteen different
electrolytes were tested with concentrations ranging from 17 to 300 gm/L of material, densities
ranging from 1.015 to 1.562 g/cm3, dynamic viscosities ranging from 0.713 to 8.6 cP, and
temperatures ranging from 18 to 59.5°C. The average measured drag-out was 47.4 mL/m2 with a
standard deviation of 16.3 mL/m2. The average predicted drag-out and standard deviation
predicted by equation 3 were 96.8 and 17.8 mL/m2, respectively, while equation 4 had average
predicted drag-out and standard deviation of 15.6 and 2.06 mL/m2, respectively. A linear
regression of measured versus predicted drag-out volumes gave an r2 of 0.021 and 0.012 for
equations 3 and 4, respectively. Taking an average of the two equations yielded no better results.
A scatter plot of the measured drag-out and the predicted drag-out is shown in Figure 1.
E-19
-------
6X1 on
2
•a 60
1 40
_u
•3
B
a.
Measured drag-out vs predicted
20 40 60 80
Measured drag-out, mL/sq.m
100
Figure 1. Measured Versus Predicted Drag-Out for Results by Siip (1992).
Slip commented that the equations do not account for electrolyte that adheres to the surface and
bottom edge even after long drain times, i.e., there is a minimal film thickness left. This becomes
increasingly important for rougher surfaces. Slip recommended that drag-out estimations for use
in recycling procedures and wastewater treatment should be based on measurements rather than
calculations. Part of the reason that poor correlation was found between Slip's measured drag-
out and the predictive equations is that Slip's drag-out showed little variation with viscosity as
shown in Figure 2.
Viscosity vs Drag-out by SuP's data
1 DO
a 80 .
%
J
j"
5 AC\ -
M
*" 9O -
0 2U
0 -
0 0
• i '
»
»
1
1 * '
^
5 1 1
Kinematic viscosity
* ' •
,
522
, cSt
5
Figure 2. Measured Drag-out as a Function of Kinematic Viscosity for
Results of Sup (1992).
E-20
-------
McKesson and Wegener (1998) at RD Chemical Company experimentally measured the amount
of drainage from PWBs as a function of time. They pointed out that longer "hang" or drainage
times allows more liquid to drain from the PWB with consequently less drag-in into the rinse
tanks and thus more efficient rinsing. However, too long of a drainage time may result in lower
PWB quality due to drying and tarnishing. McKesson and Wegener tested two outer layer
boards with solder mask and solder plated and one inner layer board with no holes. A typical
result is shown in Figure 3. (This figure is reconstructed from a figure in McKesson and
Wegener.)
50
100
150
200
250
300
Hang time, seconds
Figure 3. Drainage vs Hang Time (McKesson and Wegener 1998).
The results for all three PWBs lay virtually on top of each other in Figure 3. The authors chose to
report just the percentage of liquid that remains on the board rather than mass or volume. This
allowed the authors to see the great similarities in drainage among varying conditions. The figure
shows two drainage phases. For short times, the liquid drains very quickly followed at longer
times by a much slower drainage rate. The authors concluded that 30 seconds appeared to be an
optimal drain time. The authors also studied the effect of surfactants and found very little
difference. They also tested canting the boards at about a 15-20° angle and saw only minor
differences.
It appears that the most influential reference for typical drag-out volumes is the Electroplating
Engineering Handbook (Pinkerton 1984). These values seem to go back to work by Soderberg
published in 1936. Typical drag-out volumes are given in Table 6 as reported by Pinkerton.
E-21
-------
Table 6. Drag-Out per Unit Area (Pinkerton 1984).
Condition
Vertical parts, well drained
Vertical parts, poorly drained
Vertical parts, very poorly drained
Horizontal parts, well drained
Horizontal parts, very poorly drained
Cup shaped parts, very poorly drained
Drag-Out niL/m2
16.21
82
160
32
410
320-980
1 Suggested by Pinkerton as being the absolute minimum for drag-out on a vertical sheet.
Hanson and Zabban (1959) discussed the design of a wastewater treatment plant at an IBM plant.
To design the plant, an estimate of the wastewater quality was needed. Because a primary source
of the contaminants was the plating lines, the drag-out was estimated based on information
published by Graham in the Electroplating Engineering Handbook. (Note: the data given are
the same as that in a more recent version of the Handbook given by Pinkerton [1984] and
experimental data from another IBM plant which showed drag-out volumes ranging from 100 to
160 mL/m2.) For design, a drag-out value of 200 mL/m2 was used.
Yost (1991) studied the effect of various rinsing arrangements on the costs of cadmium
electroplating wastewater costs. In doing the calculations, Yost arbitrarily assumed drag-out of
200 mL/m2 with no reference for the value.
Chang and McCoy (1990) used a drag-out value of 160 mL/ft2 to evaluate waste minimization for
PWB manufacture. No source was given for their drag-out value, but this value appears to be
commonly used by several researchers.
Discussions with Experts in the Surface Finishing Industry
Contacts were made with several experts in the surface finishing industry. One expert source
(Sharp 1998) had the following comments on drag-out:
• CH2M-Hill did a drag-out study for Merix Corporation sometime in the mid-80s (our
efforts to obtain the report from Merix were unsuccessful). CH2M-Hill used a bath tank
and one rinse tank and dipped the boards in the bath and rinsed them sequentially and
monitored the conductivity of the rinse tank. The boards were vertical and had no holes
(interlayer boards about 20 mils thick), but the hang time and other variables can only be
found in the original report. The amount of drag-out was T/2 gallons of process bath
liquid per 3,000 ft2 (102 mL/m2) of board area (one side only).
• Holes make a difference for drag-out since the holes are small enough that the liquid does
not drain out of them very well. "Hang time" also affects the drag-out.
E-22
-------
• Horizontal lines have drag-out of about 2-5 gallons per 3,000 ft2 (39-66 mL/m2) of board
area (one side only) for boards with no holes. The drag-out is lower for horizontal lines
compared to vertical lines because of the rollers used to squeegee the water off. Vertical
boards are the older process, and the trend is to go to horizontal boards. Currently, the
industry is about 1A vertical and 1A horizontal.
• One vendor has suggested that the drag-out is about 15 gallons per 3,000 ft2 (200 mL/m2)
of board area (one side only). However, this appears too high because the experts's mass
balances on his own plating line didn't work out using this number.
• Based on the mass balances on the expert's surface finishing line, i.e., accounting for the
amount of chemicals added, consumed, and those in the waste, etc., the drag-out ought to
be about 7 gallons per 3,000 ft2 (95 mL/m2) of board area (one side only) for circuit
boards with holes, and about 3 gallons per 3,000 ft2 (41 rnL/ft2) for interlayer boards.
• There are not any available computer models that could be used to predict wastewater
concentrations, flows, etc. for plating lines.
Most of the baths used at the expert's facility (Sharp 1998) have a specific gravity of about 1.08,
but the the viscosity and surface tension are unknown. The expert thought that chemical supply
companies know the viscosity or surface tension of the process baths, but it is nearly impossible
to get those data from the suppliers.
Summary of Drag-Out Studies
Table 7 summarizes the reported drag-out quantities from researchers and practitioners.
E-23
-------
Table 7. Summary of Reported Drag-Out Volumes in the Literature.
Board
Orientation
Vertical
ii
ii
ii
ii
ii
Vertical
Horizontal
Vertical
ii
ii
ii
ii
ii
Not specified
ii
Vertical
Vertical
Bath
Microetch
ii
ii
Electroless
ii
ii
Not
specified
ii
ii
ii
ii
ii
ii
ii
Not
specified
ii
19-20 g/L &
240-250 g/L
CrO3
Various
electrolytes
Conditions/Description
Baseline
Slow withdrawal rate
Intermediate withdrawal rate & longer
drain time
Baseline
Slow withdrawal rate
Intermediate withdrawal rate & longer
drain time
CH2M-Hill study
Based on experience
Boards with holes
Interlayer boards without holes
Vertical parts, well drained
Vertical parts, poorly drained
Vertical parts, very poorly drained
Rack plating (used to estimate metals in
wastewater for design of wastewater
treatment system)
Drag -out value assumed in order to
compare costs of rinsing alternatives
Drag-out value assumed to evaluate waste
minimization
Studies at varying drainage angles,
drainage times, and withdrawal rates
Experimental determinations to test
theoretical equations
Drag-Out,
mL/m2
130
72
76
65
32
31
103
27-67
95
41
161
82
160
203
162
160
12-65
18-94
Reference
Pagel 1992
II
II
II
II
II
Sharp 1998
II
II
II
Pinkerton 1984
ii
ii
Hansan &
Zabban 1959
Yost
Chang &
McCoy 1990
Slip 1990
Slip 1992
1 Suggested by Pinkerton as being the absolute minimum for drag-out on a vertical sheet.
Drag-Out Prediction Equations
Kushner (195 la) was one of the first researchers to study drag-out in detail. Kushner
distinguished two stages in the generation of drag-out. The first stage is the "withdrawal" stage in
which the work piece is moving out of the liquid but is still in contact with it. The second stage is
"drainage" in which the work piece is completely out of the liquid, but is still over the bath and
liquid is still running off the piece. Kushner considered the withdrawal stage the more important,
because the withdrawal determined the thickness of the adhering liquid film. The factors that
E-24
-------
control the film thickness are the velocity of withdrawal, viscosity of the liquid, density of the
liquid, and surface tension of the liquid although he believed surface tension was a minor factor.
Using dimensional analysis, Kushner derived the following equation:
where:
f = film thickness
K = unknown constant determined by experiments
V = velocity of withdrawal
|i = viscosity
p = density
g = acceleration of gravity
m = unknown exponent determined by experiments
Based on experimental work of others, Kushner concluded that the best fit equation was equation
3 presented earlier:
7 = 0.02.7^7 Eqn3
Note that although equation 3 was derived by dimensional analysis, it does not appear
dimensionally consistent, because the acceleration of gravity is dropped as a term. This is also
the equation referenced by Pinkerton and Graham in the Electroplating Engineering Handbook
(1984). Importantly, this equation is for work pieces with smooth surfaces, unlike PWBs which
have many small holes. This equation will tend to underestimate drag-out for PWBs. Notably,
this is one of two equations tested by Slip (1992) and discussed above. The equation performed
poorly in predicting drag-out for a variety of electrolytes.
Kushner (195 Ib) argued that equation 3 gives good drag-out predictions for short drainage times,
but increasingly overestimates the drag-out with longer drainage times, because it does not allow
for the liquid that drains off the work piece. Conceptually for a rectangular sheet, the volume of
liquid that drains off the sheet is:
t±V=A-fdr = A-Fdr(f,P,g^,0,tdr} Eqn6
where:
AV = volume of liquid that drains from the rectangular sheet
A = area of the sheet
fdr = thickness of the film that drains off the sheet
Fdr = function describing a relationship between the independent variables and
thickness of the film that drains from the sheet
odr = surface tension of the liquid
tdr = drainage time
E-25
-------
Hence, the net film thickness or the drag-out volume per unit area after any drainage time, tdr, is:
Eqn 7
The volume of liquid that drains from the board is a complex process and Kushner was not able
to develop a predictive equation. He did, however, make qualitative statements about the effect
of several variables. Kushner believed that viscosity was the most important property of the
plating solution. Higher viscosities tend to increase the liquid adhering to the sheet as it is
withdrawn from the bath and tend to decrease the liquid that drains. Some chemicals in
particular are surface active and have molecular structures that increase viscosity. These
chemicals may cause a "surface viscosity" that give higher drag-out. Higher densities tend to
decrease the liquid adhering to the sheet and increase the drainage. However, the increase in
density due to a higher concentration of chemicals in solution is usually outweighed by the
increase in viscosity. Kushner gave an example of increasing a sucrose solution from 20% to
60%. This increases the density by 18% while the viscosity increases by 2700%. Lower surface
tension will thin the film thickness as the sheet is withdrawn and also increase the drainage as
well as reducing the volume of the bead of liquid along the bottom edge of the sheet. Of course,
wetting agents are surface active and will concentrate in the drag-out, and hence will be removed
at a higher rate than other chemicals. Longer withdrawal times and drain times will reduce drag-
out, but Kushner believed that it is better to have a longer withdrawal time than a longer drain
time. His rationale was to start with the smallest volume on the work piece to begin with. He
also referenced work by Soderberg that drainage times beyond 60 seconds have little effect.
Finally, Kushner recommended that work pieces be oriented to minimize the drainage distance
and that the pieces be tilted.
Rinsing Theory
The primary source of the quantity of wastewater generated is rinse water. Most process baths
are followed by two rinses, but sometimes just one rinse and sometimes three rinses. The
development of rinsing theory can be traced at least as far back as Kushner (1949). Pinkerton and
Graham (1984) summarized some of the fundamental mathematical relationships for rinsing. For
a non-running rinse tank and assuming that ideal, instantaneous mixing occurs, the concentration
of a contaminant in the rinse tank is given by:
1-
D)
Eqn 8
where:
C,
C0
v,
D
n
concentration of contaminant in rinse tank after t min
concentration of contaminant solution being drug into rinse tank
volume of rinse tank
volume of drag-over or drag-out on rack and work rinsing operation
number of rinsing operations in t min
E-26
-------
Most rinse operations at larger facilities use multiple countercurrent cascade rinses. In this case,
the concentration in the effluent from the rth rinse tank is:
Eqn9
(Q-t/D)r+l -
where:
Cr = concentration of contaminant in the effluent of the rth rinse tank
Q = rate of fresh water flow
t = time interval between rinsing operations
r = number of rinse tanks in series
Talmadge (1968) presents equations similar to the above but with an extra term to account for
imperfect mixing, i.e., imperfect removal of the contaminant from the work piece.
An approximate equation for multiple, countercurrent rinses has apparently been used by some
(Hanson andZabban 1959; Mohler 1984):
=££J EqnlO
Mohler (1984) discussed how rinsing equations can be used in practice. In general the rinse must
not cause a loss in product quality. There is, then, a maximum allowable concentration in the
final rinse called the "contamination limit." The ratio of the concentration in the drag-in, C0, into
the first rinse tank (or drag-out from the process bath) to the concentration in the final rinse, Cr, is
the dilution factor or "rinsing ratio," C
-------
The approach above is consistent with Kushner (1949). Kushner observed that the purpose of
the rinse tanks are to "stand guard between baths to keep one solution from mixing with another
and contaminating it." The rinse water flow rate partially determines the concentration of
carryover into the next plating tank and thus the plating quality. Kushner believed that each rinse
system in a facility would have its own unique rinsing ratio, C
-------
Other Rinsing Theory Studies
Several other rinsing theory studies have been conducted by various researches. Some of these
have focused on how well the drag-out is dispersed into the rinsing tank. While interesting, these
studies are not applicable to this project, because sufficient rinsing is used in practice such that
most of the drag-out ends up in the rinse water and thence the wastewater. For example,
Talmadge and Sik (1969) developed equations to describe the dispersing of the bead of liquid at
the bottom of a plate into the rinse water. They extended previous work that used diffusion
theory to predict the residual contaminant on a plate in a rinse tank. Talmadge and Buffham
(1961) and Talmadge et al. (1962) made detailed investigations of rinsing effectiveness in the
absence of mixing or agitation other than the flow of rinse water in the tank, i.e., molecular
diffusion is the dominant mass transfer mechanism. They found in such cases that about 10% of
the contaminant is left in the film a flat sheet as compared to typically less than 0.1% when using
ideal mixing rinse equations. However, the situation is not typical of practice, and as mentioned
above, using the ideal complete mixing equations gives a conservative estimate of contaminant in
the wastewater, i.e., less contaminant is left on the board.
PWB Pollution Prevention and Control Technology: Analysis of Updated Survey Results
As part of an EPA funded project, a questionnaire survey form on pollution prevention was sent
to 400 PWB shops in 1995 and 40 shops responded. A shortened survey was sent in 1997 to 250
PWB shops in California and 45 responded for a total of 85 between the two surveys. A
summary of information relevant to this project follows (U.S. EPA 1998).
Wastewater generation. Most of the wastewater generated is from rinsing. The best estimate of
water usage is 10 gallons/(layer-ft2 of production) or 410 1/m2 which is the "wetted" surface area
and was "calculated based on the total surface area of all layers of boards manufactured." This
value is the mean of the 20 largest shops. Large shops had the most reliable data. Smaller shops
were encouraged to estimate their data if they did not know, and this made their data suspect.
Recycle, recovery, and bath maintenance. The survey revealed several practices for recycle,
recovery, and bath maintenance:
• Nearly all shops responding to the survey reported using off-site recycling for one or
more of their spent process baths although the percentage recycled for each bath type was
not reported. The most common bath sent for recycle was spent etching because the
baths have high copper concentrations of about 150 g/L. About 80-85% of the
responders used an ammoniacal etchant and most of the rest used cupric chloride. The
volume of spent ammoniacal etchant solutions generated was 1 gallon per 30 ft2 (1.41/m2)
of inner- and outer-layer panels. Other types of spent baths were far less likely to be sent
off-site for recycle. Tin and/or tin-lead stripping solutions were the next most common
spent bath sent off-site and was reported by 20% of the survey responders.
Approximately 50% of the responders used a tin outer-layer etch resist and 50% used a
tin-lead etch resist. Only 10% of responders indicated that spent rack stripping solutions
are sent off-site.
E-29
-------
This stripping solution results from removing plating deposits from racks used to hold the
PWBs. This solution can be a significant waste. Based on the survey report, we will
assume that only spent etchant baths are sent off-site for recycle.
• The use of various technologies to recycle and recover baths and waste streams on-site
varied. Ion exchange was used by 45% of the responders to treat and recover discharges,
but many times this was part of their waste treatment system.
• The volume of wastes generated from spent baths was estimated as shown in Table 9.
Wastewater treatment. Wastewater treatment systems removed the metals by conventional
precipitation systems, ion exchange, or a combination of the two. Wastewater treatment sludges
generated are typically (88% of responders) sent off-site for recycle rather than disposed of in a
landfill. Sludge generation data were few. The three largest facilities reporting data had sludge
generation rates of 0.02, 0.31, and 0.24 kg/m2. The smallest number, 0.02 kg/m2, came from a
facility making only single sided boards whereas the other two had a larger mix of products
which generated more waste.
Drag-out reduction practices. Table 10 shows the drag-out reduction or recovery practices used
by the responders.
Drag-out reduction can reduce pollution, but it can cause problems for the process baths due to
greater build-up of contaminants in the bath. One or more bath maintenance techniques may be
required.
Water Use Rates from Survey of MHC Facilities
As part of a U.S. EPA sponsored research project, the University of Tennessee CCPCT (1997)
surveyed MHC PWB plating facilities. Part of the survey addressed water use for various MHC
process alternatives. Table 11 shows the estimated water consumption for MHC alternatives
based on the survey data and normalizing assumptions.
These water consumption rates are of the same order of magnitude as those from the U.S. EPA
(1998) survey discussed earlier which estimated water usage to be 10 gallons/(layer-ft2 of
production) as the mean of the 20 largest shops.
E-30
-------
Table 9. Selected Waste Volume Estimates From Spent Baths.
Process
Etching, inner and outer layers
Dry film resist developer
Dry film resist stripper
Tin-lead stripper
Soldermask developer
Microetch; inner and outer layers
Sulfuric acid dips
Electroless copper
Board trim
Waste
Spent etchant
Spent developer
Spent stripping solution
Spent stripping solution
Spend developer
Spent micro-etchant
Spent sulfuric acid baths
Waste electroless Cu bath
Waste copper-clad material
Volume1
(per 1,000 ft2 of
4 layer boards)
140 gallons
200 gallons
6 gallons
17 gallons
60 gallons
16 gallons
12 gallons
26 gallons
187.5 ft2, 42.9 Ibs Cu
Volume1
(per m2 of
4 layer boards)
5. 7 liters
8.1 liters
0.24 liters
0.69 liters
2.4 liters
0.65 liters
0.48 liters
1 . 1 liters
0.1875m2, 19.6kg
1 Assumptions:
a) Ammoniacal etchant used for both inner- and outer-layers, 70% of copper foils etched, 1 oz. copper used on all
layers, and 20 oz/gal carrying capacity of etchant.
b) 50% of film developed (30% outer, 70% inner), developer carrying of 3 mil-ft2/gal, and 1 mil film is used
throughout.
c) 50% of film stripped (70% outer, 30% inner), stripper carrying capacity of 100 mil-ft2/gal, and 1 mil film is used
throughout.
d) 30% metal area, tin-lead resist is 0.3 mil thick and stripper capacity of 15 oz/gal of metal.
e) 30% of mask developed, 1 mil thickness, 10 mil-ft2/gal carrying capacity.
f) Oxide, electroless Cu, and pre-pattern plate microetches (50%, 100%, and 30% of surface area etched,
respectively) considered. Many facilities may employ additional baths.
g) Microetches average etch and 4 oz/gal carrying capacity.
h) Bath life of 1 gallon/500 ssf, 3 sulfuric dips (oxide, electroless copper, and pattern plate lines).
I) 18x24 panels with 0.75 inch thief area and 0.25 inch spacing of 6 step-and-repeats, outer layer 2 oz copper (80%
trim area), inner layer 1 oz copper (50% trim area).
E-31
-------
Table 10. Drag-out Reduction or Recovery Practices Used by the Responders.
Drag-Out Reduction or Recovery Practice
Allow for long drip times over process tanks
Have drip shields between process and rinse tanks
Practice slow rack withdrawal from process tanks
Use drag-in/drag-out rinse tank arrangements
Use drag-out tanks and return contents to process baths
Use wetting agents to lower viscosity
Use air knives to remove drag -out
Use drip tanks and return contents to process baths
Use fog or spray rinses over heated process baths
Operate at lowest permissible chemical concentrations
Operate at highest permissible temperatures
PWB Responders
Using, Vo1
76.3
60.5
52.6
34.2
34.2
31.6
26.3
10.5
10.5
7.9
5.2
Plating Shops
Using, %2
60. 43
56.9
38. 13
20. 83
61.03
32.4
2.23
27.03
18.93
34.6
17.9
Data from PWB survey.
2 Data from 1993-1994 survey of for the metal finishing industry.
3 Data are for manually operated methods, which are the predominant type for the plating operations surveyed during
the NCMS/NAMF project.
Table 11. Water Consumption Rates of PWB MHC Alternatives.
Process Type
Electroless copper, non-conveyorized
Electroless copper, conveyorized
Carbon, conveyorized
Conductive polymer, conveyorized
Graphite, conveyorized
Non-formaldehyde electroless copper, non-conveyorized
Organic-palladium, non-conveyorized
Organic-palladium, conveyorized
Tin-palladium, non-conveyorized
Tin-palladium, conveyorized
Water Consumption1
(gal/ft2)
11.7
1.15
1.29
0.73
0.45
3.74
1.35
1.13
1.80
0.57
(1/m2)
476
46.8
52.5
30
18
152
54.9
46.0
73.2
23
1 Based on wetted board surface area.
E-32
-------
RESEARCH APPROACH
The objective of this study was to develop and validate methods to predict the quality of waste
water generated from PWB manufacturing processes. The methods can then be used to compare
alternative manufacturing processes in the PWB industry. In the DFE studies, industrial and
environmental exposure and risk are evaluated on a chemical-specific basis for individual
manufacturing operations. Wastewater data collected during routine regulatory sampling are
inadequate for these purposes because data are collected for only a few specific pollutants and
the samples contain wastewater from the entire plant rather than an individual process line. For
these reasons, a mass-balance calculation is the most suitable approach to estimating the load of
each pollutant emanating from a given process line.
The literature review revealed that drag-out was the source of most of the contaminants in the
wastewater from a given process. Process-specific waste loads originating from drag-out can be
estimated by the product of the drag-out volume and the chemical concentration in the process
baths. The latter are determined as an existing component of the DFE process. However,
according to the literature review, drag-out volume from PWBs and other flat, vertical pieces can
vary between about 10 and 120 mL/m2. Drag-out was affected by variables such as bath
chemistry, board withdraw rate, drain time, and orientation of the boards during withdraw.
Board surface characteristics and the number and geometry of holes drilled in the board may also
be significant, but these variables have not been systematically investigated to date. Equations
presently available in the literature fail to accurately predict the volume of drag-out from vertical
plates (Sup 1992).
The MHC process was selected as the basis of the research because a significant data base
already existed for this process as a result of the previously concluded DFE project. Also, the
research team was most experienced and familiar with this process line. The results of this work
apply to other PWB processes that employ process baths in which the boards are vertically
oriented.
The specific steps in the research plan were:
• To conduct limited laboratory drag-out experiments for the purpose of supplementing
existing data in the literature.
• To identify or develop an accurate and comprehensive drag-out model for PWB using a
data-base that includes data developed in this study and by others.
• To develop a computer model to predict wastewater quality and quantity from a PWB
processes that incorporates the new drag-out model.
• To validate the model using data from process bath and rinse water samples collected
from three MHC process lines.
E-33
-------
LABORATORY DRAG-OUT EXPERIMENTS
Laboratory drag-out experiments were conducted to supplement existing drag-out data in the
literature. Existing drag-out equations do not accurately predict the effect of fluid properties on
drag-out from vertical flat pieces such as PWBs (Slip 1992). While some studies have
investigated the effect of viscosity, another parameter that may exert significant influence, surface
tension, has received virtually no attention. The scope of this study did not allow a
comprehensive evaluation of the effect of these parameters. Instead, an alkaline cleaner bath was
selected as a bath that was more difficult to drain and a microetch bath was selected as one that
would be relatively easy to drain. During the study, viscosity and surface tension would be
measured to gain an indication of the relative influence of these parameters on drag-out.
The procedures for the laboratory drag-out experiments were devised to simulate conditions
occurring in the PWB manufacturing process. The drag-out volume was measured
gravimetrically as the boards were withdrawn from the process tanks. Experiments were
conducted using two heated process baths to determine the range of expected drag-out volumes
under various conditions. Because the alkaline cleaner/condition and microeth baths have
significantly different chemical compositions and properties, these baths were chosen for the
experiments to provide a realistic range of drag-out volumes. The board size was 0.457 m by
0.610 m. Experimental conditions that were studied were the orientation of the board during the
drain time, the length of the drain time, the board withdraw rate from the bath, and shaking the
board at the beginning of the drain period. Withdraw rates of 0.076 m/sec and 0.305 m/sec were
tested, and the boards were drained with the long edge horizontally, vertically, or at a 45° angle.
Drain periods of 10 seconds, 20 seconds, and 30 seconds were studied. The basic operating
conditions (BOC) for the majority of the tests were: 0.076 m/sec withdraw rate, 10 second drain
time, no shaking after board withdraw, 45° drain angle, and the board oriented with the long edge
horizontal. Nine sets of experiments were conducted on each bath for a total of eighteen drag-
out experiments. Several additional experiments were conducted with the microetch bath for a
drilled board with a different hole density and design. The matrix of experimental conditions that
were tested for each of the two baths is presented in Table 12.
For the alkaline cleaner/conditioner experiments, generally five repetitions were made for each
condition, with the circuit board remaining submersed in the bath for one minute on each test.
Since the etching process changed both the properties of the circuit board and the chemical
composition of the bath, only three repetitions for each condition were performed and the boards
were only allowed to remain submersed for 30 seconds. These conditions were taken into
account by assuming that the copper etch rate would remain constant over the duration of the
experiments. This assumption was verified by weighing the boards before and after the tests to
determine the mass of copper etched from the board.
E-34
-------
Table 12. Experimental Matrix for Laboratory Study of Drag-out Volumes for
Each Bath Type.
Experimental
Conditions
0.076 m/sec withdraw
45° drain angle
10 sec drip time
no shaking
0.076 m/sec withdraw
long edge horizontal
10 sec drip time
no shaking
0.076 m/sec withdraw
long edge vertical
10 sec drip time
no shaking
0.076 m/sec withdraw
45° drain angle
20 sec drip time
no shaking
0.305 m/sec withdraw
45° drain angle
30 sec drip time
no shaking
1.0 fps withdraw
45° drain angle
10 sec drip time
no shaking
0.076 m/sec withdraw
45° drain angle
10 sec drip time
shake board
Drilled Board
!
!
!
!
!
!
!
Undrilled Board
!
Drilled, Etched Board
!
Apparatus
10 cm by 61 cm by 76 cm high density polyethylene (HDPE) tank, supported and
stabilized to prevent tipping.
Magna-Whirl Constant Temperature Water Bath, Model MW-1140A-1.
Pump, ITT Jabsco Self-Priming, Model 12290-0001, 115 volt, 3.3 amp, with thermal
overload protection.
6 m of 1.3 cm diameter stainless steel tubing, coiled to fit inside bottom of HDPE tank.
1.3 cm ID. Nalgene tubing, lab/food grade, with connection clamps.
48 liters bath solution (Alkaline Cleaner/Conditioner or Microetch).
Mettler Toledo Electronic Analytical Balance, Model PR5002, Maximum 5100 grams,
with cardboard air current shield.
E-35
-------
• 0.457 m by 0.610 m circuit boards (copper clad with holes; copper clad without holes;
etched, with holes).
• Plastic bags, 0.50 mil, 110 1 capacity.
• Whittner Taktell Super-Mini Metronom, Model 886051, set at 120 beats per minute.
• Laboratory clamps and clips.
Procedure
1. For the first set of experiments, the Alkaline Cleaner/Conditioner bath was prepared
according to the manufacturer's specifications by filling the HDPE tank with 24 L of
deionized water. Next, 2.88 L of Electro-Brite ML-371 were added, and the tank was
brought to a volume of 48 L with deionized water to produce a 6% (by volume)
concentration. The solution was gently mixed. For the second set of experiments, the
Microetch bath was prepared according to the manufacturer's specifications by filling the
HDPE process tank with 24 L of tap water and adding 720 g of copper sulfate
pentahydrate (CuSO45H2O) and 8.64 L of 66° Baume sulfuric acid (H2SO4). The acid was
added very slowly, taking care that the temperature of the mixture remained below 54° C.
A laboratory thermometer was inserted into the mixture to monitor temperature. Next,
3.34 L of Co-Bra Etch Inhibitor Makeup were added, and the mixture was brought to a
volume of 48 L with tap water.
2. The stainless steel heating coil was placed into the HDPE tank containing the simulated
bath. The coil inlet was connected to tubing from the water bath (with the in-line pump),
and the coil outlet connected to tubing discharging back to the water bath. The
experimental set up is presented as Figure 4.
3. The Magna-Whirl water bath was filled with approximately 95 liters of hot tap water. The
water bath heater and pump were turned on, allowing the bath to equilibrate to 57° C for
the alkaline cleaner/conditioner, and 52° C for the microetch bath. The water bath
thermostat was set, and a thermometer was placed in the bath to monitor the bath
temperature.
4. The bath temperature, pH, and density were measured in-situ in the tank. Conductivity,
viscosity, and surface tension were measured on a sample collected from the tank.
Analyses were performed as described later in the section entitled: COLLECTION AND
ANALYSIS OF FIELD SAMPLES.
5. The circuit board was cleaned with tap water and detergent, and thoroughly rinsed with
deionized water. The board was dried using compressed air to ensure no moisture
remained entrapped in the holes.
6. The board was centered on the analytical balance, and the weight was recorded to the
nearest 0.01 g.
7. A clean new plastic bag was weighed on the analytical balance, and the results recorded to
the nearest 0.01 g.
8. The plastic bag was opened, and carefully attached to the outside of the HDPE tank using
small laboratory clips.
E-36
-------
9. The metronome was turned on, and two laboratory clamps were attached to the circuit
board to serve as handles. The circuit board was slowly lowered into the tank so the
entire surface was completely submerged in the bath. The board was agitated slightly to
remove entrapped air bubbles, and then allowed to remain submerged for approximately
one minute in the alkaline cleaner/conditioner bath or 30 seconds in the microetch bath.
The process was timed by counting ticks on the metronome.
10. The board was removed vertically at the appropriate withdraw rate, stopping several
inches above the bath surface. Depending on the experiment, the board was then either
held steady or given one quick shake, and the board held so that its edge was either level
or at a 45° angle during the allotted drain time. The appropriate withdraw rates, drain
positions, and drain times were specified in the Table 12. Both the withdraw rate and drip
time were timed by ticks of the metronome.
11. The board was immediately placed into the plastic bag attached to the tank. Extra care
was taken to ensure that any drips after the specified drain period fell into the bag, and
that the sharp corners of the board did not puncture the bag.
12. The clamps were removed from the board, along with the clips holding the bag to the
tank. The bag was carefully sealed, removing as much air as possible.
13. The sealed bag containing the circuit board and drag-out was centered on the analytical
balance and weighed, the results were recorded to the nearest 0.01 g.
14. The circuit board was carefully removed from the bag, and the process was repeated,
beginning with weighing a clean new plastic bag.
15. After the specified number of runs were completed for each set of conditions, the bath
temperature, pH, and density were again measured in-situ in the tank. Conductivity,
viscosity, and surface tension were measured on a sample collected from the tank.
Analyses were performed immediately after collecting the sample, and the results were
recorded.
16. The drag-out volumes were calculated.
Before the actual drag-out experiments were conducted using PWB bath chemicals, a series of
four preliminary tests were conducted to validate the proposed methodology and to verify that
the drag-out could be measured accurately and precisely. The preliminary tests also served as
practice runs, and allowed for any necessary adjustments to the procedure and apparatus. The
coefficients of variation for the first two tests were 0.039 and 0.056, for eleven and nine trials,
respectively. The coefficients of variation in the third and fourth tests improved to 0.007 and
0.008, respectively, for series of seven trials each. Since preliminary tests were not designed to
cover the full range of operating variables, the following representative variables were selected: 1)
ambient temperature tap water was used to simulate bath chemicals; 2) a 0.265 m x 0.457 m
drilled etched board was used in the first two preliminary tests, and a 0.457 m by 0.610 m drilled
copper clad board was used for the third and fourth tests; and (3) the circuit board was
withdrawn at 0.15 m/sec, given one quick shake after removal, and allowed to drip for 10
seconds.
E-37
-------
Quality Assurance and Quality Control (QA/QC)
Prior to the experiments, all laboratory equipment was thoroughly cleaned with detergent
followed by a thorough deionized water rinse. The analytical balance used for weighing the
boards was allowed to warm up for at least 30 minutes before any measurements were made.
The balance was calibrated using calibration weights at the beginning and end of each laboratory
session, to ensure the instrument had not drifted. A large shield was placed around the balance to
decrease the effects of drafts while weighing the board.
Prior to mixing the actual baths, 500 ml batches of the solution were prepared per the
manufacturers' product information sheets. Measurements of viscosity, specific gravity, surface
tension, conductivity and pH were compared between the 500 ml batches and the full bath
volume. Temperature was monitored continuously during the drag-out experiments in the baths
by suspending a laboratory thermometer in the tank. Before the tests, the timing of the
metronome was checked with a clock to ensure proper timing. The tank was positioned in front
of a fume hood for adequate ventilation, and a large strip of tape was affixed to the fume hood
shield at a 45° angle from the horizontal to use as a guide during drain periods. Personal
protection equipment such as safety goggles, gloves, and aprons were used whenever feasible.
All waste material including plastic bags contaminated with the drag-out chemicals and the used
bath solutions were stored for proper disposal. All laboratory experimental information and data
were recorded in a laboratory notebook, with carbon copies given to the principal investigators
upon test completion.
Results and Discussion
Results of the laboratory drag-out volume experiments are presented in Tables 13 and 14 for the
alkaline cleaner/conditioner and microeth baths, respectively.
Table 13. Drag-Out Results for Alkaline Cleaner/Conditioner Bath.
Test
BOC
BOC, board edge horizontal
BOC, board edge vertical
BOC, 20 sec. drip time
BOC, 30 sec. drip time
BOC, 1 fps withdraw
BOC, with shake
BOC
BOC
Board Type
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
undrilled
drilled, etched
Drag-Out (ml/sq.m)
77.8
75.6
81.3
68.2
64.5
98.7
77.8
38.6
89.2
Coeff. of Variation
0.032
0.015
0.021
0.040
0.047
0.013
0.032
0.016
0.038
Note: Design 1, 5619 holes; Design 2, 7824 holes.
E-38
-------
Table 14. Drag-Out Results for Microetch Bath.
Test
BOC (2/2/99)
BOC (2/13/99)
BOC (2/13/99)
BOC, board edge horizontal
BOC, board edge vertical
BOC, 20 sec. drip time
BOC, 30 sec. drip time
BOC, 1 fps withdraw
BOC, with shake
BOC
BOC, etched board
BOC, etched board
Board Type
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
drilled, design 2
Drag-Out, ml/sq m
108.9
107.8
93.4
120.9
113.0
98.1
94.4
133.1
111.9
69.8
112.3
118.3
Coeff. of Variation
0.043
0.023
0.038
0.006
0.006
0.015
0.007
0.016
0.021
0.038
0.022
0.021
Note: Design 1, 5619 holes; Design 2, 7824 holes.
The drag-out volume for each experimental condition was calculated using the mean drag-out
weight from the group of tests for the specific condition. This was generally five runs for the
alkaline cleaner/conditioner, and three runs for the microetch. In addition to calculating the mean
drag-out weight (in grams), the standard deviation and the coefficient of variation of the
measurements were checked for each condition. The coefficient of variation was less than 0.05
for all experiments.
The mean drag-out volume for all experimental conditions for the alkaline cleaner/conditioner
was 74.7 ml/m2, which is approximately 30% less than the mean drag-out volume of 108 ml/m2
for the microetch bath. The mean drag-out for all experimental conditions for both baths
combined was 91.1 ml/m2, and was calculated using only data from the same board hole design
so as not to skew the results. It appears that drain time has an affect on drag-out volume, as
reflected in the decreasing drag-out volumes as drain time increased. It also appears that the
drag-out volume increases as the board withdraw rate decreases. Board tilt and orientation did
not appear to affect the drag-out volume; however, drilled boards had more drag-out than
undrilled boards, as expected.
Results from the microetch experiments compare favorably to those performed at Micom, Inc.
(Pagel 1992), although a direct comparison was difficult since operating conditions were different.
Board hole density for both tests were similar, with Micom boards having 33,000 holes/m2
compared to 28,000 holes/m2 for the boards used in the microetch experiments in this study.
Pagel's drag-out volumes appear to be less than those measured in this study. At a withdraw rate
of 0.20 m/sec and drain time of 12.1 sec, Pagel reported a drag-out volume of 76.4 mL/m2. Under
similar conditions, specifically a withdraw rate of 0.305 m/sec and a drain time of 10 seconds, this
study resulted in a drag-out of 130 mL/m2. Other differences in experimental
E-39
-------
procedures that could affect drag-out volumes include: 1) a 45° drain angle used in this study,
compared to a 0° angle used by Pagel; 2) Pagel's experiments included drag-out associated with
the racks; and 3) drag-out was measured by completely different approaches; specifically, Pagel
used a concentration approach whereas this study used a weight approach.
Analyses of parameters for the alkaline cleaner/conditioner and microetch simulated baths were
performed before the drag-out tests were run, and again after the tests were completed. Results
of the tests are presented in Tables 15 and 16.
Table 15. Alkaline Cleaner/Conditioner Bath Properties.
Parameter
pH
Conductivity mS/cm
Specific Gravity
Surface Tension, dynes/cm
Viscosity, cP
Before Experiments
8.65 @ 58°C
0.21 @35°C
8.65 @ 57°C
34.7
0.85
After Experiments
8.47 @ 57°C
0.23 @ 35°C
0.995 @ 57°C
34.7
0.87
Table 16. Microetch Bath Analyses.
Parameter
pH
Conductivity mS/cm
Specific Gravity
Surface Tension, dynes/cm
Viscosity, cP
Before Experiments
-0.42 @ 53°C
1374 @ 22°C
1.175 @53°C
71
1.44@49°C
After Experiments
-0.62 @ 55°C
1562 @ 22°C
1.205@57°C
60
0.87 @ 50°C
As expected, there was no significant variation in the bath parameters for the alkaline
cleaner/condition bath comparing values before and after the drag-out tests. There were,
however, significant variations in the microetch bath characteristics, as expected. Conductivity,
specific gravity, hydrogen ion concentration and viscosity all increased, possibly due to the
increase in copper in the bath as a result of etching from the PWBs during the drag-out tests.
E-40
-------
DRAG-OUT MODEL DEVELOPMENT
As stated previously the goal of this project was to develop and validate methods for predicting
the quality of wastewater generated during PWB manufacturing. Drag-out and bath dumps are
the two major sources of process wastewater. The literature reports drag-out rates for flat panels
and PWBs ranging from 10 to 160 ml/m2. Currently-available models utilize solution viscosity
and withdraw rate as the primary independent variables. Slip (1992) has demonstrated that drag-
out rates predicted using these models are poorly correlated with results from experiments.
Clearly there is a need for a more a more accurate means of predicting drag-out for PWB
manufacturing.
In addition to the drag-out data collected as part of this study, three data sets containing extensive
drag-out data for PWBs or flat panels were available in the literature (Slip 1990; Slip 1992; Pagel
1992; Ducker). An attempt was made to develop regression models to predict drag-out volumes
as a function of PWB manufacturing practices. Possible model variables that were either
recorded or varied in each study are summarized in Table 17.
Table 17. Potential Variables for PWB Drag-Out Prediction Model.
Board Size
Withdraw Rate
Drain Time
Board Orientation
Board Angle
Board Surface
Holes
Shaking or Vibration
Bath Type
Kinematic Viscosity
Surface Tension
Sup 1990
Slip 1992
Pagel
1992
This Study
Of the variables listed in the table above, not all were evaluated for inclusion in the model. Board
surface (etched or unetched) and shaking were not included in the parameters to be evaluated
because the little data that were available for these parameters indicated they have a minor effect
on drag-out volumes. Board orientation during draining was also not considered because
relatively few data were available and it is not one of the waste minimization practices commonly
practiced. We hypothesized that kinematic viscosity and surface tension were two fluid properties
that may be most significant in determining drag-out volumes. However, Slip (1992) showed
that drag-out volume was poorly correlated with kinematic viscosity. Furthermore, Pagel's data
set did not include data for either kinematic viscosity or surface tension of the baths and Slip's
data did not include any surface tension data. It was judged that the quantity of data and range of
values for these two variables were insufficient to justify their inclusion in the model.
E-41
-------
In the data base used to develop the model, board size (m2), withdraw rate (m/sec), and drain
time (sec) were treated quantitatively by using the numerical value of the variable. Three other
variables were treated qualitatively using indicator variables having values of 1 or 0. The indicator
variable for board angle was assigned a value of 1 if the board was angled and a value of 0 if the
board edge was kept horizontal. Similarly, the indicator variable for holes was assigned a value of
1 if it contained holes and a value of 0 if the board did not contain holes. The hole density for the
drilled boards in the data base ranged from 20,000 to 33,000 holes/m2; however, data needed to
further quantify the effect of drilled holes, such as hole diameter and aspect ratio, were not
available. Three different indicator variables were included to specify bath type: alkaline cleaner,
micro-etch and electroless copper. The obvious disadvantage of this approach is that the model
can make bath-specific predictions only for these three bath types, but insufficient viscosity and
surface tension data are available to make the model more general.
The data set was not ideal for development of the model. The work of Slip (1990, 1992) was not
specific to the PWB industry; therefore, he did not use standard PWB process baths, his boards
were smaller than those often used in the PWB industry, and his boards did not contain drilled
holes. As a result, variables describing board size and holes were strongly correlated (0.904),
making it difficult to distinguish between the effects of these two parameters. Also, Slip did not
use actual PWB process baths, thus bath type and board size were also correlated. During model
development, it was necessary to be aware of the effects that these peculiarities may have on the
developed model.
Both a linear regression model and a multiplicative regression model were tested. The linear
model was:
WR
DO = a(]+ a, SIZE + aJVR + a^DT + a, + aWR • DT +
U 1 2. 5 4 T~X rr-i J
a6HOLES + a, ANGLE + a.ALK + a9MICRO + awELCTRS
where:
DO = drag-out volume, mL/m2
SIZE = board area, m2
WR = withdraw rate, m/sec
DT = drain time, sec
HOLES = 1 if the board is drilled and = 0 for undrilled boards
ANGLE = 1 of the board is tilted during draining and = 0 if the board is kept
horizontal
ALK = 1 if the bath is an alkaline cleaner bath and = 0 otherwise
MICRO = 1 if the bath is a micro-etch bath and = 0 otherwise
ELCTRLS = 1 if the bath is an electroless copper bath and = 0 otherwise
The multiplicative model was:
E-42
-------
HOLES „ ANGLE
DO=a0- SIZEai • WR"2 • DTa^ • a6HUL^ • a7
ALK MICRO ELCTRLS
8 • <29 • <210 Eqn 12
which was rewritten in linear form for analysis by linear regression:
logDO = Ioga0 + av logSIZE+ a2 logWR+ a3 logDT+ HOLESloga6 +
ANGLE\oga7 + ALKloga, + MICRO\oga9 + ELCTRLS\ogaw
Both models were evaluated using stepwise regression (SSPS ver. 9). This procedure adds or
removes independent variables to the model based on criteria related to the reduction in the sum
of squares achieved by inclusion of the variable. The final model includes only the variables that
result in a statistically significant reduction in the sum of squares error. The stepwise regression
procedure yielded an r2 = 0.883 for the linear model and 0.814 for the multiplicative model. The
linear model was:
WR
DO = 3.63 + 694-SIZE - 180- ELCTRLS+ 89.6- E 14
- 155- ALK + 38.6-HOLES + 29.9-WR - 0.443-DT-127 -MICRO
The statistical package did not include the variables of ANGLE and WR-DT in the model because
they were not statistically significant. Inspection of this equation reveals that all three bath-type
coefficients are relatively large negative numbers, which would cause it to predict an erroneously
large drag-out for large boards (ca. 0.25 m2) with bath-types not explicitly accounted for in the
model. For small boards (ca. 0.05 m2) used with the bath-types accounted for in the model, it
could predict negative drag-out values. These anomalies were the result of correlation of the
independent variables, as described earlier. To correct this problem it was necessary to eliminate
one of the three bath types as a variable in the model. Each of the three bath types was evaluated
for elimination, the best fit was given by eliminating MICRO as a variable (R2 =0.852). The final
drag-out model was:
DO = 18 + 201-SIZE - 60.1-ELCTRLS +73- —
./->'-/ -•—i -t r
-20.9-ALK + 26.0-HOLES + 26.1-WR - 0.355-DT qn
A comparison of predicted and measured drag-out volumes is shown in Figure 5. The groups of
vertically-aligned data points occur when the model predicts a near-constant drag-our for
conditions in which the measured drag-out is variable. While some of the variability is random
error, some is also the result of variation of the independent variables, indicating that the model is
not able to accurately account for all the variables that affect drag-out. A more comprehensive
data base in which the independent variables are systematically varied is needed if more accurate
predictions of drag-out from PWB manufacturing processes are desired.
E-43
-------
(N
E
E,
-i—*
o
en
2
Q
T3
§
CD
0
^
IOU
160 -
140 -
120 -
100 -
80 -
60 -
40 -
20 -
n -
• Ducker, 1999 (Alkaline Cleaner)
O Ducker, 1999 (Microetch) T
V Pagel, 1992 (Microetch) T
V Pagel, 1992 (Electroless Coppeer) Q *T T
• Sup, 1992 jp \
D Sup, 1990 P V V
P T
§ ^O ^
i~ ® O
Jb T
w
s« W
v
-------
PWB WASTEWATER MODEL
Given the volume of drag-out from and chemical composition of each bath, it is possible to
calculate the mass of each contaminant that would enter the waste stream for a given PWB
process line. A computer model was developed to facilitate such calculations. The model was
based on the following assumptions:
1. Contaminants in wastewater are from drag-out from process baths and from dumping of
some baths at the end of their useful life. Contaminants from the stripping of racks from
deposits are ignored.
2. Essentially 100% of the drag-out ends up in the wastewater, i.e., very efficient rinsing.
3. Predictions are for vertical boards only.
4. Various predictive equations reported in literature are of limited value for estimating
absolute values of drag-out as evidenced by the results of Slip's work comparing
predicted versus measured drag-out. Equation 15 was used to estimate drag-out in the
model here.
5. Insufficient information exists to include surface tension as a variable although the
authors recognize that it may be an important variable.
6. The estimate of drag-out of contaminants in g/d is based on the PWB production rate,
chemical composition of each bath, and the estimated drag-out from each bath, according
to the following equation:
( PWB production) f Concentration of ^ ( drag - out from ^
contaminant i = , 2 r i • • u u • / T I ' i i i • T / 21 F™ i fi
I rate, m2/d ) \i mbathj, mg/L) (bath), mL/m2) Eqn 16
Vfrombathj j
The model is coded in an Excel Spreadsheet and utilizes a Visual Basic Macro. The user is
required to enter information in a separate spreadsheet defining the operating conditions of the
process line and the chemical composition of the baths. The effect of bath dumps on the overall
pollutant load can be included by specifying their frequency. The model calculates the mass of
contaminants coming from each process tank, together with the contaminant mass and
concentration for the entire process line. A user's manual is included in the Appendix.
E-45
-------
COLLECTION AND ANALYSIS OF FIELD SAMPLES
Samples of plating baths and rinse waters were collected from the MHC process line from three
different PWB facilities for the purpose of verifying the drag-out model. Three process baths at
each plant were selected for sampling: microetch, electroless copper, and Anti-Tarnish. Sodium
or potassium were selected as tracers for each bath because they are common ions in PWB baths,
and they tend to be relatively stable in solution. The relative amount of sodium and potassium in
the bath and downstream rinses can be used to estimate the drag-out from each tank and to verify
the overall mass balance approach to modeling wastewater quality from PWB facilities. In
addition to sodium and potassium, fluid properties (viscosity, surface tension and specific
gravity) that might effect the quantity of drag-out were measured. Routine measurements of
conductivity and pH were taken too. The project QA/QC plan (Robinson and Cox 1998),
submitted to and approved by EPA, was followed except where field conditions necessitated
minor changes.
Process Characterization
Operating practices affect the amount of drag-out and the concentration of contaminants in the
rinse-tank effluent. Extensive data characterizing the operating practices used at each site were
collected during the site visits. Operating practices potentially affecting the amount of drag-out or
the rinsing process are summarized in Tables 18-20. These data were later used to predict the
drag-out from each process bath using equation 15 and to independently calculate the drag-out
via a dynamic mass balance approach described later.
Table 18. Summary of MHC Operating Practices for the Field Sites.
Plant 1
Plant 2
Plants
Cycle Time, min
30
37
27
Withdraw Rate, m/sec
0.173
0.163
0.234
Board Tilt,
degrees
5
0
0
Hole Density, #/m2
100,000 to 570,000
NA
50,000
Table 19. Summary of Drip Times for Process Baths at Field Sites.
Bath
Plant 1 ME
Plant 1 EC
Plant 1 AT
Plant 2 ME
Plant 2 EC
Plant 2 AT
Plant 3 ME
Plant 3 EC
Plant 3 AT
Drip Time,
sec
5
25
5
10
15
10
5
10
5
E-46
-------
Table 20. Summary of Rinsing Practices Used at Field Sites.
Plant 1 ME Rinse 1
Plant 1 ME Rinse 2
Plant 1 EC Rinse 1
Plant 1 EC Rinse 2
Plant 1 AT Rinse 1
Plant 1 AT Rinse 2
Plant 2 ME Rinse 1
Plant 2 EC Rinse 1
Plant 2 AT Rinse 1
Plant 3 ME Rinse 1
Plant 3 EC Rinse 1
Plant 3 EC Rinse 2
Plant 3 AT Rinse 1
Rinse Time
(min:sec)
1:20
1:00
2:10
1:00
3:20
2:00
2:05
8:00
3:55
1:15
2:00
4:20
6:04
Rinse Tank
Vol (1)
832
832
832
832
832
832
415
415
415
892
892
892
892
Rinse Flow
Rate (1/min)
7.6
7.6
7.6
7.6
7.6
7.6
3.8
3.8
3.8
9.8
7.6
7.6
7.6
Rinse Water
Source
ME Rinse 2
city
EC Rinse 2
city
AT Rinse 2
city
city
AT Rinse 1
city
H2SO4 rinse
EC Rinse 2
AT Rinse 1
city
Mixing1
1,2
1,2
1,2
1,2
1,2
1,2
1,2
1,2
1,2
1,2
1,2
1,2
1
1 Mixing: 1 = Board Agitation; 2 = Aeration.
Sample Collection
Samples were collected for analyses from the laboratory drag-out study tanks in the UT
laboratory and from actual process baths and rinse tanks during the PWB industry site visits. For
the laboratory drag-out study in the UT laboratory, grab samples were collected for surface
tension and viscosity. The samples were collected directly from the experiment tank in a clean
beaker, and the analyses were immediately performed.
Samples were collected during the PWB site visits from the microetch (ME), electroless copper
(EC), and anti-tarnish (AT) process baths and their succeeding rinse tanks in the MHC process
line. Grab samples were collected using either a plastic measuring cup or a sampling beaker,
which consisted of a plastic beaker with a long handle attached. The sampling container was
thoroughly rinsed with the sampling fluid prior to sample collection. The grab sample was then
immediately transferred from the sampling cup or beaker into a clean 500 ml HPDE sample bottle
and capped. Before the sampling event, pre-printed labels were prepared in duplicate, with
one label pre-attached to the sample bottle. After the sample was collected, the remaining label
was attached to the Sub-Unit Data Collection Log, and the sample description, person taking the
sample, time of sample, sample volume, and method of preservation was recorded in ink.
Duplicate samples taken in identical manner were collected at plants 1 and 2. At plant 3, the two
samples were taken at different times in the board cycle. The first sample was taken just prior to
the boards entering the rinse tank while the second was taken just after the boards were removed.
Replicates were taken for approximately 20% of the samples. The sample bottles were sealed
with color-coded tamper-proof tape (to identify the sampler and establish chain-of-custody), and
placed in plastic lined containers for transport to the UT laboratory.
E-47
-------
Temperature
Temperature was measured in-situ in the laboratory drag-out tanks. In the field, temperature was
measured on grab samples collected from the process and rinse tanks. Measurements were made
immediately after collection.
pH
pH was measured in-situ in the laboratory drag-out tanks. In the field, pH was measured on grab
samples collected from the process and rinse tanks. Measurements were made immediately after
collection.
Apparatus
• Orion Digital Portable pH Meter, Model 250A.
Orion Triode™ pH Electrode, Model 91-57BN.
Procedure for pHMeasurements
1. After the meter was calibrated, the electrode was placed into the laboratory drag-out tank
or sample and agitated slightly.
2. When the pH display was stable, the pH was recorded on the Sub-Unit Data Collection
Log.
3. The electrode was rinsed with deionized water, and the process repeated.
The pH meter was calibrated prior to taking measurements for each sub unit. A two buffer
calibration was performed using the 4.01 and 7.00 buffers for the acid sub units, and 7.00 and
10.01 buffers for the alkaline sub units. The first measurement in a sub unit was made in the
samples from the last rinse tank, and the measurements progressed up-line, with the last
measurement made on the process bath sample.
Conductivity
Conductivity measurements were performed both in the UT laboratory and at the PWB site visits.
The instrument automatically compensates for temperature effects to a certain degree, except for
acids. Since many of the PWB baths and rinses were acids, and temperature could have a
significant effect on the conductance of these solutions, it was determined that all conductivity
measurements should be made at the reference temperature of 25° C. The conductivity
measurements originally made in the field at the PWB sites were re-analyzed on samples in the
UT laboratory at a controlled temperature of approximately 25° C. At the beginning of each lab
session, the conductivity meter was checked against a solution of known conductance to verify
accuracy.
The conductivity measurements of the rinse tanks were within the meter range of 0.0 to 199.9
mS/cm; however, as anticipated, the values of some of the process baths were higher. Since
conductivity is a nearly linear function of total dissolved solids (Snoeyink and Jenkins 1980), a
1:10 or 1:100 dilution with deionized water was performed on the sample if the initial reading was
above the highest range on the meter. The measurement was then taken on the diluted sample,
and the meter reading multiplied by the dilution factor.
E-48
-------
Two temperature and conductivity readings were taken on each sample, with the mean values
reported.
Apparatus
• Orion Conductivity/Temperature Meter, Model 122.
Viscosity
Viscosity was measured on site from grab samples collected from the rinse tanks, process baths,
and laboratory drag-out tanks.
Apparatus
• Gilmont Falling Ball Viscometer, size 1, with stainless steel ball, range 1 to 10 centipoise.
Procedure
1. The temperature of the rinse tank or process bath was taken using the laboratory
thermometer.
2. A grab sample was collected from the tank using a 2000 ml beaker. The viscometer,
stainless steel ball, and thermometer were immediately submerged into the sample for
approximately one minute to allow the laboratory equipment to equilibrate to the liquid
temperature.
3. The inside of the viscometer was rinsed with the sample, then slowly filled with rinse or
process bath liquid, making sure no air bubbles adhered to the sides of the viscometer.
4. The temperature of the liquid in the beaker was checked and compared with the tank
temperature. In general, if the temperature difference was more than approximately 5°C,
the beaker was emptied and a new sample collected.
5. The viscometer was held vertical in the center of the 2000 ml beaker. (The beaker still
contained the rinse or process liquid, which acted as a temperature bath for the
viscometer.) The stainless steel ball was carefully placed by hand into the filled
viscometer, making sure no air bubbles stuck to the ball.
6. A stopwatch was used to time the descent of the ball between the fiducial lines on the
viscometer. The time was recorded on the Sub-Unit Data Collection Log.
7. The viscometer and beaker were emptied, and the process repeated.
Using the mean descent time, the viscosity was calculated as follows:
Eqnl?
where:
m = viscosity, centipoise
K = viscometer constant (0.257 with stainless steel ball, based on laboratory calibration
tests using deionized water and sucrose solutions, described below)
rf = density of ball, mg/1 (8.02 for stainless steel ball)
r = density of liquid, mg/1
t = time of descent, minutes
E-49
-------
The viscosity was recorded on the Sub-Unit Data Collection Log.
The viscometer, stainless steel ball, and beaker were thoroughly rinsed with deionized water prior
to the next test.
Before viscosity measurements were made in the field and on the laboratory drag-out tanks, a
series of tests were performed to establish the viscometer constant, K, for the falling ball
viscometer. The constant was obtained by measuring the time of descent of the stainless steel
ball in standard solutions of known viscosity, and was calculated using the following relationship:
EqnlS
Three solutions were used in the investigation: 30 percent sucrose (by weight), 40 percent
sucrose (by weight), and deionized water. Before the sucrose solutions were prepared, the
sucrose was dried in a desiccator, and all glassware was cleaned and completely air dried. A 1000
ml volumetric flask was weighed on an electronic analytical balance, and the weight recorded to
the nearest 0.01 gram. The appropriate amount of sucrose was weighed on the analytical balance
(338. 10 g and 470.60 g for the 30 percent and 40 percent solutions, respectively), and added to the
clean, dry volumetric flask. Approximately 500 ml of deionized water was added to the flask,
and the mixture agitated by swirling. Additional deionized water was added slowly, while being
swirled, until the sucrose was completely dissolved and the bottom of the meniscus reached the
1000 ml reference line on the volumetric flask. The solution was allowed to rest to allow any
entrapped air bubbles to rise. The volumetric flask containing the solution was weighed on the
analytical balance, and the temperature was measured with a laboratory thermometer; both
measurements were recorded in a laboratory research notebook.
The density of the sucrose solutions and the deionized water was calculated using the following
relationship:
v Eqn 19
where:
D = density, g/ml
m = mass of solution = mass of flask and solution - mass of flask, g/L
v = volume of solution, ml
Prior to the experiments to determine the viscometer constant, the sucrose solutions were gently
stirred to ensure a homogeneous mixture. A laboratory thermometer was used to measure the
temperatures of the sucrose solutions and deionized water, and the results were recorded in a
laboratory research notebook. The same procedure as described above was used except the
constant temperature bath was not needed because the experiments were done at ambient
temperature. Instead, the filled viscometer was held vertical in a 50 ml glass cylinder. The
viscometer constant, K, was determined to be 0.257 by fitting equation 17 to the experimental
time and literature values of viscosity.
E-50
-------
Specific Gravity
Specific gravity was measured in-situ in the laboratory drag-out tanks. In the field, specific
gravity was measured on grab samples collected from the process and rinse tanks.
Measurements were made immediately after collection.
Apparatus
• Hydrometer, Fisherbrand, range 0.890 to 1.000.
• Hydrometer, Fisherbrand, range 1.000 to 1.600.
• 500 ml glass cylinder (optional).
Before the hydrometers were used for measurements for the rinse tanks, process baths and
laboratory drag-out tests, the accuracy of the instruments was verified. Hydrometer readings
were taken on deionized water and a 40 percent (by weight) sucrose solution. The temperature of
the water and sucrose solution was measured with a laboratory thermometer, and the specific
gravity measurements were compared with published values. Results of the verification for
deionized water resulted in a value 0.15% higher than the expected published value of 1.000 at
20° C, and 0.5% less than the published value of 1.176 for the 40 percent sucrose solution at
20° C.
Surface Tension
Surface tension was measured in the UT laboratory on grab samples collected from the rinse
tanks, process baths, and laboratory drag-out tanks.
Apparatus
• Fisher Surface Tensiomat, Model 21, with platinum-iridium ring.
• 5 cm inch diameter glass vessel, approximately 1.3 cm deep.
• Magna-Whirl water bath.
Procedure
1. A water bath was prepared to simulate the temperature of the rinse tank or process bath as
measured in the field and recorded on the Sub-Unit Data Collection Log.
2. The rinse tank or process bath sample bottles were placed in the water bath, and allowed
to equilibrate to the bath temperature. The water bath and sample temperatures were
intermittently monitored using the thermometer. The sample bottles remained in the
water bath until used for the surface tension measurement.
3. The clean platinum-iridium ring was placed on the hook on the lever arm of the tensiomat.
4. A clean 5 cm diameter glass vessel was filled with a portion of the sample (transferred
immediately from the water bath) and placed on the sample table inside the tensiomat.
5. The sample table was raised until the ring was immersed in the liquid to a depth of
approximately 3 mm.
6. The torsion arm on the tensiomat was released, and the instrument was adjusted to a zero
reading by turning the knob on the right side of the case until the index and its image were
in line with the mark on the mirror. Care was taken to ensure the ring remained in the
liquid by adjusting the height of the sample table. The knob on the front of the case
beneath the main dial was adjusted until the vernier read zero on the outer scale of the
dial.
E-51
-------
7. The sample table was lowered until the ring was at the surface of the liquid. At the same
time, the knob on the right side of the case was adjusted to keep the index in line with the
mark on the mirror. The two simultaneous adjustments were continued until the
distended film at the surface of the liquid broke.
8. The reading on the scale at the breaking point (surface tension in dynes per centimeter)
was recorded on the Sub-Unit Data Collection Log.
9. The liquid was emptied from the glass vessel, and the process was repeated.
10. Both the platinum-iridium ring and glass vessel were rinsed with deionized water prior to
the next test.
Prior to the surface tension tests, the calibration of the tensiomat was checked and the platinum-
iridium ring was thoroughly cleaned.
To verify the calibration according to the instrument's instruction manual, the ring was placed on
the lever arm and the instrument was adjusted to a zero reading. A 600 mg piece of aluminum
foil was placed on the ring, and the knob on the right side of the case was adjusted until the index
and its image were in line with the mark on the mirror. The dial reading was recorded, and
compared with the calculated surface tension:
Eqn 20
where:
S = dial reading = apparent surface tension in dynes/cm
M = weight (0.6 grams)
g = acceleration of gravity (980 cm/sec2)
L = mean circumference of ring (6.00 cm)
The platinum-iridium ring was cleaned per the manufacturer' s instructions: the ring was: 1)
soaked in concentrated nitric acid for approximately 2 minutes, then rinsed with deionized water;
2) rinsed with acetone, followed by deionized water; and 3) flamed with a Bunsen burner.
Before surface tension measurements were made, the surface tension of deionized water was
checked at 20°C to verify accuracy. Seven measurements were made, with a mean value of 74.96
dynes/cm, a standard deviation of 2.03 dynes/cm. This mean value is 4.2 percent higher than the
expected value of 72 dynes/cm for the deionized water.
Metals Analysis
Sodium and/or Potassium analyses were conducted in the UT laboratory on grab samples
collected from the process baths and rinse tanks.
Apparatus
• Allied Analytical Systems Atomic Absorption Spectrophotometer, IL Video 12, Serial
Number 1857.
• Sartorius Analytical Balance, Model AC 120S, UT ID Number 427286.
E-52
-------
Reagents
• Sodium calibration standard, Fisher Scientific, 1000 mg/L.
• Potassium calibration standard, Fisher Scientific, 1000 mg/L.
• Potassium chloride (KC1), Fisher Scientific, certified grade.
• Lanthanum chloride (LaCl 6H2O), Fisher Scientific, certified grade.
Procedure
1. Stock potassium chloride solution was prepared by dissolving 23.84 g. of potassium
chloride in 250 ml of deionized water in a volumetric flask. This produced a solution of
50,000 mg/L as K, which was used as an ionization suppressant for the sodium samples.
A stock solution of lanthanum chloride was prepared by dissolving 12.72 g. of lanthanum
chloride in 100 ml of deionized water in a volumetric flask. This produced a solution of
50,000 mg/L as La, which was used as an ionization suppressant for the potassium
samples.
2. Sodium and potassium standards were prepared by diluting the Fisher Scientific
calibration standards with deionized water to achieve the desired standards
concentrations.
3. The samples were prepared by performing dilutions with deionized water to get the
anticipated analyte concentrations within the linear range of the instrument. Volumetric
pipettes and volumetric flasks were used, and the samples were transferred to new, clean
125 ml HDPE sample bottles. Samples were acidified with ultrapure nitric acid, and
ionization suppressants were added to achieve a concentration of 2000 mg/L as K for the
sodium samples, and 1000 mg/L as La for the potassium samples.
4. The appropriate lamp was inserted in the atomic absorption spectrophotometer, and a
safety check of all settings was performed. The instrument electronics were turned on
and allowed to warm up for approximately 30 minutes.
5. The instrument printer, compressed air, and acetylene were turned on. The pilot was lit,
the flame adjusted, and the sampling tube was placed in a fresh beaker of deionized water.
6. The instrument was calibrated with the appropriate sodium or potassium standards. A
standards curve was printed, and a linear regression performed to check linearity of the
curve. If the value of r2 value was below 0.9950, the instrument was re-calibrated with
fresh standards.
7. The prepared samples were analyzed, beginning with the rinse samples and progressing
up-line to the process tank. Approximately ten analyses were run per sample, each lasting
approximately eight seconds. Results were printed and transferred to an Excel
spreadsheet.
8. The method of standard additions was performed on process bath samples to reduce
matrix effects. The samples were diluted 1:1 with known standards and analyzed in the
absorption mode. Generally, 0, 50, 100 and 200 mg/L standards were used for potassium
analyses, and 0, 20, 50 and 100 mg/L standards were used for sodium analyses; however
there was some variation since it was necessary to keep concentrations within the
instrument's linear range. A plot of absorption verses concentration of added standards
was then prepared, from which the actual concentration in the sample was derived. If
necessary, standard additions were performed on the succeeding rinse tanks, as described
later in this section.
E-53
-------
Before and during the atomic absorption analyses, all laboratory glassware and sample bottles
were acid washed in accordance with Standard Methods.
The analyte (sodium or potassium) was determined based on process bath composition, as
provided by either industry representatives, manufacturers' material safety data sheets, or
previous research conducted by the University of Tennessee's CCPCT.
Because of the extremely high anticipated concentration of analyte in some of the process baths,
along with the wide range of anticipated concentrations between the process baths and rinse
tanks, atomic absorption analyses were conducted using the least sensitive wavelengths (330.2
nm for sodium, and 404.4 nm for potassium) whenever possible. Dilutions were still necessary
on many of the samples. For sodium samples with very low sodium concentrations, it was
necessary to use the most sensitive wavelength of 589.0 nm.
The instrument was calibrated at the beginning of each lab session by using generally five
calibration standards within the linear range of the instrument, including a zero standard. The
standards used for the least sensitive wavelength for sodium (330.2 nm) were usually 0, 20, 50,
100, and 150 mg/L; however these occasionally varied depending on the anticipated
concentration of the sample. In all cases, the standards were chosen to best bracket the sample
concentration. Standards used for the most sensitive sodium analyses (589.0 nm wavelength)
were usually 0, 0.25 0.50. 0.75, 1.0 and 1.25 mg/L. Calibration standards for the least sensitive
wavelength for potassium (404.4 nm) were usually 0, 50, 100, 200 and 600. As with the sodium
analyses, standards were chosen to best bracket the sample potassium concentration. Standards
checks were performed during the measurements to ensure the instrument had not drifted. The
checks usually were performed after every four or five measurements, but always after ten
measurements were taken.
The samples were prepared for analysis by dilution with deionized water to achieve an anticipated
analyte concentration within the linear range of the instrument. The anticipated concentrations
were based on previous research conducted by the University of Tennessee's CCPCT. Alkali
salts were added to the samples and standards as an ionization suppressant. Potassium chloride
was added to sodium samples at 2000 mg/L, and lanthanum chloride at 1000 mg/L was added to
the potassium samples. Process and rinse tank samples and standard solutions were acidified to
pH < 2 in accordance with Standard Methods, using ultrapure concentrated nitric acid.
Electroless copper samples were not acidified due to the possibility of the baths containing
cyanide.
As an interference check, a standard additions analysis was performed on one sample for each
process bath, and compared with analysis results performed without standard additions.
Whenever there was a difference greater than 10 percent between the two measurements, a
standard addition analysis was performed on the duplicate bath sample, and the standard addition
results were used. If standard additions were necessary for the process bath samples, the
succeeding rinse tank samples were also checked, to determine if standard additions should be
used.
E-54
-------
Quality Assurance and Quality Control (QA/QC)
Prior to the site visit to collect the samples, the 500 ml new HDPE sample bottles were
thoroughly cleaned with detergent, triple rinsed with deionized water, and allowed to air dry.
Field blanks were used to monitor any contamination from the bottles. The field blanks were pre-
labeled and filled with deionized water in the UT laboratory prior to the site visits. During the
visit, the bottles were opened for approximately two minutes, then re-sealed.
All laboratory equipment transported to the site was thoroughly cleaned according to Standard
Methods prior to leaving the UT laboratory, and was again thoroughly cleaned between sites. All
laboratory equipment, including reagents and deionized water was transported from the UT
laboratory, including cleaning supplied. The samples remained in the custody of the sampling
team until arrival back to the UT laboratory, where they were placed in a limited access, locked
cold room until analyses.
Results from Analysis of Field Samples
Mean values of temperature, specific gravity, viscosity, conductivity, surface tension for each of
the field samples are summarized in Table 21.
Measurements of conductivity, specific gravity, surface tension, viscosity were all completed in
duplicate. The coefficients for all measurements were all excellent (conductivity 0.04, surface
tension 0.005, specific gravity 0.001% and viscosity 0.073).
Sodium and potassium concentrations are summarized in Table 22. Replicate samples at plants 1
and 2 were taken in identical manner, and the results were averaged and reported as a single
value. At plant 3, two samples were taken at different times in the board cycle time. Samples
labeled "A" were taken just prior to the boards entering the rinse tank and should normally
correspond to the lowest concentration present in the rinse tank. Samples "B" and "R" were
taken just after the boards were removed from the rinse tank and should be near the maximum
concentration in the rinse cycle. The individual samples from plant 3 were not averaged, but
reported individually. Details of the analytical procedure used for each sample are summarized in
the Appendix.
E-55
-------
Table 21. Temperature, Specific Gravity, Viscosity, Conductivity, Surface Tension for Field
Samples.
Sample Name
Plant 1 ME Process
Plant 1 ME Rinse 1
Plant 1 ME Rinse 2
Plant 1 EC Process
Plant 1 EC Rinse 1
Plant 1 EC Rinse 2
Plant 1 AT Process
Plant 1 AT Rinse 1
Plant AT Rinse 2
Plant 1 FB
Plant 2 ME Process
Plant 2 ME Rinse 1
Plant 2 EC Process
Plant 2 EC Rinse 1
Plant 2 AT Process
Plant 2 AT Rinse
Plant 2 FB
Plant 3 ME Process
Plant EC Process
Plant 3 EC Rinse 1
Plant 3 EC Rinse 2
Plant 3 AT Process
Plant 3 AT Rinse
Plant 3 FB
Temp.,
°C
30
20
20
45.5
21
20
19
20
20
NA
37
15
38
20
19
16.5
NA
29
54
27
30
25
30.5
NA
Specific
Gravity
1.110
1.005
1.004
1.170
1.003
1.005
1.004
1.002
1.002
NA
1.175
1.004
1.110
1.002
1.005
1.005
NA
1.145
1.115
1.002
1.003
1.005
0.994
NA
Viscosity,
cP
1.140
1.112
1.142
1.218
.977
1.097
1.172
1.097
1.022
NA
1.246
1.172
1.421
.932
1.202
1.037
NA
1.340
1.139
0.992
NA
1.127
0.798
NA
Conductivity,
mS/cm, 25 °C
304,000
1,935
213
224,000
1,043
224
341
229
223
1.8
477,000
2,170
119,600
676
353
256
1.9
168,400
261,000
736
155
543
156
1.8
Surface Tension,
dynes/cm
76.2
75.9
75.6
73.2
76.0
76.3
72.2
74.4
76.2
76.2
78.0
77.0
51.2
73.2
75.0
76.3
76.1
77.6
56.2
74.0
75.4
72.2
73.6
75.0
Table 22. Metals Concentrations Measured in Field Samples.
Sample Name
Plant 1 ME Process
Plant 1 ME Rinse 1
Plant 1 ME Rinse 2
Plant 1 EC Process
Plant 1 EC Rinse 1
Plant 1 EC Rinse 2
Sodium, mg/L
67,750
242
24.5
Potassium, mg/L
20,380
77.4
<7.5
Method
Standard Additions
Standard Curve
Standard Curve
Standard Additions
Standard Curve
Standard Curve
E-56
-------
Sample Name
Plant 1 AT Process
Plant 1 AT Rinse 1
Plant 1 AT Rinse 2
Plant 1 Makeup water
Plant 1 FB
Plant 2 ME Process
Plant 2 ME Rinse 1
Plant 2 EC Process
Plant 2 EC Rinse 1
Plant 2 AT Process
Plant 2 AT Rinse
Plant 2 Makeup water
Plant 2 FB
Plant 3 ME Process
Plant 3 ME Rinse 1-A
Plant 3 ME Rinse 1-B
Plant 3 ME Rinse 1-R
Plant 3 EC Process
Plant 3 EC Rinse 1-A
Plant 3 EC Rinse 1-B
Plant 3 EC Rinse 1-R
Sample Name
Plant 3 EC Rinse 2-A
Plant 3 EC Rinse 2-B
Plant 3 AT Process
Plant 3 AT Rinse 1-A
Plant 3 AT Rinse 1-B
Plant 3 AT Rinse 1-R
Plant 2 Makeup water
Plant 3 FB
Sodium, mg/L
2.8
20.15
63,450
128.6
30.8
34.5
31.36
<0.01
41,550
173.6
242
289
72,950
109.3
173.5
191.7
Sodium, mg/L
24.3
24.4
111
19.1
19.1
23.2
23.1
<0.1
Potassium, mg/L
94
<7.5
<7.5
<7.5
<7.5
62,300
98.8
<7.5
<7.5
<7.5
Potassium, mg/L
<7.5
Method
Standard Additions
Standard Curve
Standard Curve
Standard Curve
Standard Curve
Standard Additions
Standard Curve
Standard Additions
Standard Curve
Standard Additions
Standard Curve
Standard Curve
Standard Curve
Standard Additions
Standard Additions
Standard Additions
Standard Additions
Standard Additions
Standard Curve
Standard Additions
Standard Curve
Method
Standard Curve
Standard Curve
Standard Additions
Standard Curve
Standard Curve
Standard Curve
Standard Curve
Standard Curve
E-57
-------
The pooled instrumental relative standard deviation for potassium was determined to be 0.77%,
based on eighteen potassium samples with a mean sample concentration of 113.6 mg/L, and a
pooled instrumental standard deviation of 0.87 mg/L. The pooled instrumental relative standard
deviation for sodium was determined to be 1.6% based on seventy-three analyses with a mean
concentration of 60.6 mg/L. The pooled instrumental standard deviation was 0.97 mg/L. Data on
which these calculations are based are included in the Appendix.
The relative standard deviation for duplicate potassium samples ranged from 0.17 to 6.95% for
tests run with no standard additions, with a pooled standard deviation of 3.46 mg/L. There were
no duplicate or replicate analyses for potassium using the method of standard additions. The
relative standard deviation for duplicate sodium measurements without standard additions ranged
from 0.11 percent to 18.94 percent, with a pooled standard deviation of 8.05 mg/L. The relative
standard deviation for duplicate sodium analyses performed with standard additions ranged from
0.52 to 6.13%, with a pooled standard deviation of 2.76 mg/L. Data for duplicate samples from
which these results were determined are listed in the Appendix.
E-58
-------
DYNAMIC MASS BALANCE MODEL
FOR INTERPRETATION OF FIELD DATA
The field data collected at the PWB manufacturers was used to validate the drag-out component
of the wastewater generation model. The output from the model is the average mass rate of
contaminant in the rinse water from a particular process bath; the model can also calculate
average concentrations in the rinse tank effluent by dividing by the rinse flow rate.
However, the average concentration predicted by the model does not correspond directly to the
contaminant concentrations measured in the field samples. The MHC process is dynamic in that
the concentrations of contaminants in the rinse effluent change as a function of time. The
operation cycle of a given rinse tank consists of a short period of time in which a board is
immersed in the tank, followed by a longer period of time during which no boards are in the
tanks. Contaminants are continually flushed from the rinse tank during the entire operation time
of the bath. As a result of this operational practice, the rinse-tank concentration history will be a
periodic saw-tooth wave function. In the field, instantaneous grab samples were collected from
the rinse tanks, usually immediately after removal of the board. Clearly, the concentrations in the
instantaneous grab samples may not be directly comparable to the average concentration
calculated by the model; therefore, a means of verifying the model is needed. A dynamic
material balance model was used to compare the concentration of contaminant in the grab
samples with the average concentration of contaminant predicted by the models.
The following material balance equation describes the concentration of contaminant in a
completely-mixed rinse tank:
dC Eqn21
~dt
where:
Q = flow rate through the tank, L3/t
V = tank volume, L3
C = concentration of contaminant in the tank, M/L3
C0 = concentration of contaminant in the feed water to the tank, M/L3
t = time, t
The concentration of contaminant in the tank as a function of time can be determined by
separating the variables in equation 21 and integrating using appropriate boundary conditions.
Assume that when the line is first started (before the first board is dipped in the tank) that the
contaminant concentration in the tank is equal to the feed water concentration. Also assume that
at t=0 a rack of boards, containing mass of contaminant M, instantly releases all of its
contaminant to solution. Under these conditions, the concentration in the tank at t=0 is:
i^ o z-'
C = C0 +
0 V
M Eqn22
E-59
-------
The solution to equation 21 describing the concentration history after removal of the first board is
then given by:
ft Q r
JO \/ ~ Jl
dC Eqn 23
„ n M ( Q{\ Eqn 24
C = C+—exp -¥-
V \ V)
As time progresses additional boards will enter the rinse tank. Assume that additional boards
enter the tank at a constant period of 1. It is convenient to redefine t as:
t = n A + 9 Eqn25
where
n = number of cycles completed since t = 0
q = time elapsed in the current cycle, t
The effluent history during the rinsing cycle for the second board processed after start-up would
be given by:
\ —dO =
Jo l/
(M/l/)[1+exp(-CW,/>/)]+C0 Q _|
Eqn27
V V) V \ V j
This result can be extended to represent the effluent history for the rinsing period after the nth
board is rinsed:
r_r M QVKJ -<-
u-u, + —exp-—./.BAH -_ Eqn 28
E-60
-------
Steady-state is defined to occur when n = °°. Substituting
VexDr kW 1 Ecln 29
^° V v r,_exp(^
\ V )
yields an expression concentration history for a single rinse tank, operating at steady-state:
C-C0+-exp- —^^forq<1 Eqn3Q
v y1-exp ——,
Example:
A rinsing tank receives a rack containing 60 ft2 of boards every 30 minutes. The drag-out rate is
10 mL/ft2 and the contaminant concentration in the process tank is 3000 mg/L. The rinse rate is 2
gpm and the tank is 220 gallons in volume. The feed water contains 40 mg/L of the contaminant.
Calculate the effluent concentration history during the 30 minute cycle period under steady-state
conditions:
= 21 6mglL
= 40 + 2.16exp 7
I 2201 f-2»3(
1 - exp
\ 220
( 26
C=40 + 2.16exp
I TJD I I — > • -ill \
Eqn 32
Equation 32 is plotted over the course of one process cycle in Figure 6.
E-61
-------
40
10
15
time (min)
20
25
30
Figure 6. Example Concentration History of Rinse Tank Effluent During One
Plating Cycle.
E-62
-------
MODEL VALIDATION
The purpose of the field samples was to validate the drag-out prediction model and the overall
mass balance approach to predicting wastewater quality from PWB facilities. The dynamic
material balance model for the rinsing process was developed in the previous section to facilitate
this comparison. First, equation 30 was solved for the mass of contaminant in the drag-out:
1 - exp,
V ' Eqn33
The volume of the drag-out could then be calculated by dividing the mass of contaminant in the
drag-out by the bath concentration:
M
34
The drag-out volumes calculated from the field data and the dynamic mass balance (equations 33
and 34) are compared to those predicted using the drag-out regression model (equation 15) in
Table 23. Replicate samples at the plants 1 and 2 were taken in identical manner, and the results
were averaged and reported as a single value. At plant 3, the duplicate samples were taken at
different times in the board cycle time. Samples labeled "A" were taken just prior to the boards
entering the rinse tank and should normally correspond to the lowest concentration present in the
rinse tank. Samples "B" and "R" were taken just after the boards were removed from the rinse
tank and should be near the maximum concentration in the rinse cycle. Samples 3MER1-A and -
B were taken soon after the MHC line had been shut down for a short period of time and may
have been erroneously low. The individual samples from plant 3 were not averaged; separate
calculations were made for each one. Sodium and potassium concentrations in the anti-tarnish
rinse tanks were too low to accurately calculate either the mass of contaminant in the drag-our or
the drag-out volume.
The drag-out volumes calculated from the field data are consistently less than those predicted by
the drag-out model. They are also significantly less than those measured both in the laboratory
experiments performed as a part of this work and the data collected by Pagel (1992). For
example, the drag-out volumes from Microetch baths calculated from our field data ranged from
22.8 to 53.6 mL/m2, compared to a range of 76 to 122 mL/m2 in this study and a range of 57 to
145 mL/m2 in Pagel' s work. Similarly, the drag-out volumes from the Electroless baths
calculated from our field data ranged from 9.73 to 32.9 mL/m2, compared to a range of 20.4 to
81.8 mL/m2 in Pagel's work. A possible explanation is that the drag-out volumes calculated from
the field data were based on the assumption in the dynamic mass balance model that all the
contaminant was released instantaneously from the PWB and that the rinse tank was perfectly
mixed. The rinsing tanks used in PWB plants may not approximate this ideal behavior. Rinse
water typically enters the bottom of the rinse tank and flows over a weir at the water surface. As
the board enters the tank, it is likely that a significant fraction of the pollutant flows over the weir
prior to being mixed
E-63
-------
throughout the tank. Fluid shear may contribute to the loss of contaminant near the water
surface of the tank as the board enters the tank. The grab samples were generally collected
immediately following removal of the board from the rinse tank. We hypothesize that the short-
circuiting process described above may have caused a large fraction of the contaminant to be
removed from the rinse tank prior to the time that we collected the sample. Our laboratory drag-
out study, and the work of Pagel (in which the rinse water flow rate was set to zero during the
sampling) were not subject to this influence.
Table 23. Comparison of Drag-Out Volumes Calculated from Field Samples to Those
Predicted by Regression Model.
Sample Description
Plant 1, Microetch
Plant 1, Electroless Copper
Plan 2, Microetch
Plant 2, Electroless Copper
Plant 3, Microetch A
Plant 3, Microetch B
Plant 3, Microetch R
Plant 3, Electroless A
Plant 3, Electroless B
Plant 3, Electroless R
Drag-Out Volume Calculated
from Field Data, mL/m2
53.6
32.9
22.8
23.2
28.2
41.0
37.9
9.73
6.83
10.9
Drag-Out Volume Calculated
from Regression Model,
mL/m2
127
59.1
102
39.9
98.2
98.2
98.2
34.7
34.7
34.7
A regression equation was fitted to the data in Table 23, resulting in the following relationship(r2
= 0.71):
field
0.68
Eqn35
where:
^
predicted
= drag-out volume calculated from the field data
is the drag-out volume predicted by the regression model
The slope of the regression equation suggests that about 2/3 of the total mass of contaminant
flows over the weir prior to being mixed. The relatively good correlation coefficient indicates that
the field and predicted drag-out volumes were comparative on a relative basis. This suggests that
the drag-out regression model and overall mass balance approach may be valid for making
relative comparisons between process alternatives.
E-64
-------
CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE WORK
Conclusions
• Contaminant mass in PWB process wastewaters can be expressed as a mass balance in
which the mass of contaminant in the wastewater is equal to the mass of contaminant
released via drag-out from the process baths (which ultimately ends up in the rinse tanks),
periodic dumping of process tanks into the wastewater, and stripping deposits from racks.
Drag-out is generally considered to be the major contaminant source. Data quantifying
composition of the process baths, the volume of wastewater produced, and the frequency
of bath dumps are usually collected during the course of the DFE process. For example,
this information was collected for the MHC process during a previous study by the
University of Tennessee CCPCT (Kincaid et al. 1997).
• Very little data exists quantifying the rate of drag-out from PWB processes, i.e., the mass
or volume of drag-out per unit surface area of PWB, e.g., mL/m2. A study reported by
Pagel at Micom, Inc. is the only readily available study on PWB facilities. Limited drag-
out research has been conducted on flat pieces, most notably by Slip. However, the
numerous small holes present in PWBs renders application of drag-out volumes
measured from non-PWB pieces problematic. Practitioners tend to use rules-of-thumb or
historically accepted values for drag-out. This one-size-fits-all approach ignores process
specific information such as bath type, viscosity, surface tension, board withdrawal rate,
or drain time. Drag-out rates reported in the literature for vertically-oriented flat pieces.
range from 10 to 160 mL/m2.
• Commonly-cited equation found in the literature offer predictions of the drag-out rate as a
function of kinematic viscosity and board withdrawal rate. Slip showed that this equation
does not predict drag-out very well for the rectangular flat pieces that he studied. There
was no relationship between kinematic viscosity and drag-out, and two previously
proposed predictive equations performed poorly.
• Several variables have been shown to affect the drag-out rate. Studies at Micom, Inc.
reported by Pagel (1992) showed the importance of longer drainage time and slower
withdrawal rate in reducing drag-out. Slip (1990, 1992) also found that these variables are
important as well as the angle of the board during drainage. No research was found that
addressed the effect of surface tension. Based on the present study, surface tension may
be an important variable.
• Considerable literature exists on rinsing theory which appears highly developed and well
studied for ideal mixing situations. While rinsing theory is not as well developed for non-
ideal mixing, previous researchers have concluded the assumption of ideal mixing is valid
for estimating long-term-average wastewater concentrations because nearly all of the
drag-out ultimately reaches the wastewater effluent.
• Laboratory studies conducted as part of this research expanded the data base of drag-out
rates for two PWB process baths and several operating conditions. The experimental
procedures showed good reproducibility, and the data were consistent with previous
research.
E-65
-------
A regression model for predicting drag-out volume was developed using the available data
bases of Slip (1990, 1992), Pagel (1992), and the present study. The dependent variables
were a choice of two types of process baths (plus a default for any other type of bath),
board withdrawal rate, drain time, board size, and presence of drilled holes. The model
had an R2 of 0.852.
The regression model for predicting drag-out rate was incorporated in a computer model
for predicting contaminant mass loading and mean pollutant concentrations from PWB
manufacturing process lines. The model was written as a Visual Basic macro within an
EXCEL spreadsheet. Input variables included facility production rate, board size, number
and types of process baths, bath composition, frequency of bath dumps, and wastewater
production rate.
Samples were collected from three PWB facilities in order to validate the drag-out model.
Samples were collected from various process and rinse tanks and analyzed for
temperature, specific gravity, viscosity, surface tension, conductivity, and potassium or
sodium concentration. Since it was not convenient to collect composite samples from the
rinse tanks, grab samples were collected at various times after a board was inserted into a
rinse tank. An equation was developed to relate the time-dependent concentration of
potassium or sodium in the rinse tank to the drag-out volume. Unfortunately, it appears
that poor mixing in the rinse tanks led to unrepresentative sampling. Although the
apparent drag-out rates were about 1/3 of the predicted rates, a comparison of drag-out
rates between process tanks showed a correlation (r2 = 0.71) with the previously
developed regression model, and in that sense lend support to it.
Recommendations
The authors believe that this work has resulted in a more universally applicable method
for estimating the mass and concentration of contaminants in a PWB process wastewater
than currently exists, especially for relative evaluations. However, much can still be done
to improve the model since the existing data are so limited. Previous work has not
studied the effect of surface tension, but the laboratory studies in this work showed that
surface tension may be an important variable. Indeed, one of the drag-out reduction best
practices is to use a wetting agent which would reduce surface tension. The effect of
viscosity was previously thought to be important, but neither Slip nor this work found it
to be significant. There has also been only one previously reported study of an actual
PWB facility. The authors believe that a better quantitative understanding of the variables
affecting drag-out could lead to the development of a better prediction equation. The first
phase of such research should focus determining the effect of bath fluid properties and
operating variables under controlled laboratory conditions. Expansion and testing of the
model could be accomplished by samples collected at PWB facilities during a second
phase of the study.
Beyond determining the wastewater quality emanating from PWB manufacturing
processes, there is a need to assess the fate of the chemicals in the PWB wastewater both
in the onsite treatment processes at PWB manufacturing facilities and at Publicly-Owned
Treatment Works (POTW). Information of the effect of chemical speciation on the fate of
E-66
-------
pollutants is especially needed. For example, metals are one of the primary pollutants of
concern in PWB wastewater, and it is likely that many of the metals are chemically
complexed in PWB wastewater. On-site treatment processes are likely to preferentially
remove the least stable metal complexes, while the most stable complexes are discharged
to the POTW. Standard removal efficiencies for metals in activated sludge processes are
probably not applicable to these highly complexed metals and the potential for release of
the metals to the aquatic environment may be underestimated.
A third issue needful of better understanding is the volatilization of chemicals from tanks
and baths such as in PWB plating processes and other manufacturing processes. The
volatilization models used in the previous assessment of emissions for the MHC process
and the present assessment of surface finishing assume gas-side control of the mass
transfer, i.e., volatilization, of chemicals from the process baths. In the MHC, and
presumably in the surface finishing process, there were several chemicals whose mass
transfer appeared to be liquid-side controlled. The mass transfer model used does not
apply for this situation and could lead to an overestimate of the emission and consequent
risk for these chemicals. It would be productive to research the literature to find more
appropriate liquid-side control mass transfer models and applicable constants for various
types of manufacturing process tanks. For example, there is a body of literature available
on surface renewal theory models which would be more appropriate for liquid-side mass
transfer control. This literature search could be completed within a year and a decision
made at that time as to whether any lab based research is warranted.
E-67
-------
REFERENCES
American Chemical Society, Chemical Abstracts., American Chemical Society, Washington,
D.C., 1998.
Chang, L., and McCoy, B. J., "Waste Minimization for Printed Circuit Board Manufacture,"
Hazardous Waste & Hazardous Materials, 7, No. 3, 293-318 (1990).
Hanson, N. H., and Zabban, W., "Design and Operation Problems of a Continuous Automatic
Plating Waste Treatment Plant," Plating, 909-918 (August, 1959).
Hatschek, Emil, "The Viscosity of Liquids" D. Van Nostrand Company, New York, 1928.
Kincaid, L.E., Geibig, J.R., Swanson, M.B., and PWB Engineering Support Team, Printed
Wiring Board Cleaner Technologies Substitutes Assessment: Making Holes Conductive., Center
for Clean Products and Clean Technologies, University of Tennessee, Knoxville, Tennessee
(1997).
Kushner, J.B., "Rinsing: I. With Single-Compartment Tank," Plating, 36, August, p. 798-801,
866 (1949).
Kushner, J.B., "Where Do We Go from Here? Part III - Water Control," Metal Finishing, 47,
No. 12,52-58,67(1951).
Kushner, J.B., "Dragout Control - Part I," Metal Finishing, 49, November, 59-64 (1951).
Kushner, J.B., "Dragout Control - Part II," Metal Finishing, 49, December, 58-61,67 (1951).
Kushner, J.B., "Rinsing Techniques," in Metal Finishing: 47th Guidebook-Directory Issue 1979,
Vol. 77, No. 13, (January, 1979).
McKesson, Doug, and Wgener, M.J., "Rinsewater Quality .... Hard Data," Proceedings of the
Technical Conference IPC Printed Circuits Expo '98, p. S09-1-1 to S09-1-5, The Institute for
Interconnecting and Packaging Electronic Circuits, Long Beach, California, 1998.
Mohler, J.B., "Water Rinsing," in Metal Finishing: 52nd Guidebook-Directory Issue 1984, Vol.
82, No. 1A, (January, 1984).
Mooney, T., "Water Rinsing," in Metal Finishing: 59th Guidebook-Directory Issue 1991, Vol.
89, No. 1A, (January, 1991).
Pagel, Paul, Modifications to Reduce Drag out at a Printed Wiring Board Manufacturer,
EPA/600/R-92/114, Risk Reduction Laboratory, Office of Research and Development, U.S.
Environmental Protection Agency, Cincinnati, Ohio (1992).
E-68
-------
Pinkerton, H.L., and Graham, A.K., "Rinsing," in Electroplating Engineering Handbook.,
Lawrence J. Durney, Ed., van Nostrand Rheinhold, New York, 1984.
Robinson, R.B. and Cox, C.D., QA/QC Plan for Verification of Hot Solder Finishing:
Prediction of Water Quality from Printed Wiring Board Processes., University of Tennessee,
Knoxville, TN, 1998.
Sharp, J., Teradyne Corp., Nashua, NH., Personal communication (September, 1998).
Slip, Von M., "Technologische Mapnahmen zur Minimierung von Ausschleppverlusten
(Technological Measures for Minimizing Drag Out)," Galvanotechnik, 81, No. 11, 3873-3877,
(1990).
Slip, Von M., "Bestimmung Elektrolytspezifischer Ausschleppverluste (Detemination of
Electrolyte Specific Losses Due to Drag-out)," Galvanotechnik, 83, No. 2, 462-465, (1992).
Talmadge, John A., "Improved Rinse Design in Electoplating and Other Industries," Proceedings
of the Second Mid-Atlantic Industrial Waste Conference, p.217-234, (1968).
Talmadge, John A., Buffham, Bryan A., and Barbolini, Robert R., "A Diffusion Model for
Rinsing," AIChE Journal, 8, No. 5, 649-653, (1962).
Talmadge, John A., and Buffham, Bryan A., "Rinsing Effectiveness in Metal Finishing," J. Water
Pollution Control Federation, 33, No. 8, 817-828, (1961).
Talmadge, J.A., and Sik, U.L., "A Drop Dispersal Model for Rinsing," AlChE Journal, 15, No. 4,
521-526, (1969).
U.S. Environmental Protection Agency, Printed Wiring Board Pollution Prevention and Control
Technology: Analysis of Updated Survey Results, EPA/744-R-98-003, Design for the
Environment Branch, Economics, Exposure, and Technology Division, Office of Pollution
Prevention and Toxics, Washington, D.C. (1998).
Yost, K.J., "The Computer Analysis of Waste Treatment for a Model of Cadmium Electroplating
Facility," Proceedings Third International Cadmium Conference, p. 56-59 (1991).
Zaytsev, Ivan Demitrievich, and Aseyev, Georgiy Georgievich, Editors, Properties of Solutions
of Electrolytes, M. A. Lazarev and V. R. Sorochenko, Translators, CRC Press, Boca Raton, 1992.
E-69
-------
LIST OF SYMBOLS
A = area of the sheet
c; = mass content of the component is kg of component per kg of solution
C0 = concentration of contaminant solution being drug into rinse tank
Cr = concentration of contaminant in the effluent of the rth rinse tank
C, = concentration of contaminant in rinse tank after t min
D = volume of drag-over or drag-out on rack and work rinsing operation
D; = coefficient calculated as shown below for each component for use in the
method given by Zaytsev and Aseyev
da, du, d2 = empirical coefficients chosen for each electrolyte component from a table
for use in the method given by Zaytsev and Aseyev
f = film thickness, cm
fdr = thickness of the film that drains off the sheet
Fdr = function describing a relationship between the independent variables and
thickness of the film that drains from the sheet
g = gravity (981 cm/s2)
h = height of metal sheet
K = unknown constant determined by experiments
m = unknown exponent determined by experiments
n = number of rinsing operations in t min
Q = rate of fresh water flow
r = number of rinse tanks in series
t = time interval between rinsing operations
T = temperature of solution, °C
tdr = drainage time, s
tw = withdrawal time, s
V = velocity of withdrawal
VA = withdrawal rate of metal sheet, cm/s
V, = volume of rinse tank
AV = volume of liquid that drains from the rectangular sheet
v = kinematic viscosity, cm2/s
p = density of electrolyte, gm/cm3
|i = dynamic viscosity of electrolyte, g/(cnrs)
|i0 = viscosity of water, Pa-s
odr = surface tension of the liquid
E-70
-------
Appendix F
Supplemental Performance
Demonstration Information
-------
APPENDIX F
F.I Modeling the Test Results
General linear models (GLMs) were used to analyze the test data for each of the 23 electrical
circuits in Table 4.1 at each test time. The GLM analysis determines which experimental factors or,
when possible, combinations of factors (interactions) explain a statistically significant portion of the
observed variation in the test results.
A GLM used to analyze the test results with respect to sites, flux type, and their interactions
(where possible) is expressed as the following 22-term equation:
Y = po + (31D1 + p2D2 + p3D3 + p4D4 + p5D5 + p6D6 + p7D7 + p8D8 + p9D9 + p10D10 + puDn (F.1)
+ p12D12 + Pi30i3 + Pi4D14 + PisDis + PieD16 (Main effects)
+ pi7D3D16 + p18D4D16 + p19D6D16 + p20D10D16 (Two-factor interactions)
+ P21D12D16 + P22D15D16
The coefficients in the GLM ((3o, (3i, (32, ...) are estimated using ordinary least squares regression
techniques. The dummy variables, DI to Die, are set equal to 1 to identify type of surface
finish/manufacturing site and type of flux that are associated with individual test results. Otherwise,
the dummy variables are set to 0. The following dummy variables can be used to represent the
experimental variables for each test environment for each electrical response variable.
Dj = 0 if surface finish is not HASL - Site 2
= 1 if surface finish is HASL - Site 2
D2 = 0 if surface finish is not HASL - Site 3
= 1 if surface finish is HASL - Site 3
D3 = 0 if surface finish is not OSP - Site 4
= 1 if surface finish is OSP - Site 4
D4 = 0 if surface finish is not OSP - Site 5
= 1 if surface finish is OSP - Site 5
D5 = 0 if surface finish is not OSP - Site 6
= 1 if surface finish is OSP - Site 6
D6 = 0 if surface finish is not immersion Sn - Site 7
= 1 if surface finish is immersion Sn - Site 7
D7 = 0 if surface finish is not immersion Sn - Site 8
= 1 if surface finish is immersion Sn - Site 8
D8 = 0 if surface finish is not immersion Sn - Site 9
= 1 if surface finish is immersion Sn - Site 9
D9 = 0 if surface finish is not immersion Sn - Site 10
= 1 if surface finish is immersion Sn - Site 10
DIQ = 0 if surface finish is not immersion Ag - Site 11
= 1 if surface finish is immersion Ag - Site 11
DH = 0 if surface finish is not immersion Ag - Site 12
= 1 if surface finish is immersion Ag - Site 12
Di2 = 0 if surface finish is not Ni / Au - Site 13
= 1 if surface finish is Ni / Au - Site 13
Di3 = 0 if surface finish is not Ni / Au - Site 14
= 1 if surface finish is Ni / Au - Site 14
DH = 0 if surface finish is not Ni / Au - Site 15
= 1 if surface finish is Ni / Au - Site 15
D15 = 0 if surface finish is not Ni / Pd / Au - Site 16
= 1 if surface finish is Ni / Pd / Au - Site 16
Die = 0 if flux is not water soluble
= 1 if flux is water soluble
F-l
-------
APPENDIX F
The "base case" is obtained by setting all D; = 0. Note that the surface finish/manufacturing site is
HASL / Site 1 if DI = D2 = D3 = D4 = D5 = D6 = D7 = D8 = D9 = DID = DH = Di2 = D0 = DM = Di5 =
0. Likewise, if Die = 0, the flux is low-residue. Thus, the base case is HASL / Site 1 with LR flux.
Note the GLM in Equation F.I contains six interactions terms that represent the last six sites in
Table 4.2 (5, 6, 7, 11, 13, and 16) for which both LR and WS fluxes were used.
The GLM approach provides a tool for identifying the statistically significant experimental
variables and their interactions. That is, all terms in the model that are significantly different from the
base case are identified through tests of statistical hypotheses of the form:
H0: ^ = 0 versus Hj: P, ^ 0 for all i
If the null hypothesis is rejected, then the coefficient of the corresponding term in the GLM is
significantly different from 0, which means that the particular experimental conditions represented by
that term (surface finish or flux type) differ significantly from the base case. If the null hypothesis is
not rejected, then the coefficient of the corresponding term in the GLM is not significantly different
from 0 and, therefore, the experimental conditions represented by that term do not differ significantly
from the base case. Such terms are sequentially eliminated from the GLM (see Iman, 1994, for
complete details).
The GLM approach is quite flexible and easily adaptable to a variety of requirements. For
example, if the focus is on surface finishes and not sites; the GLM in Equation F. 1 would be replaced
by one of the following form:
Y = Po + Pi0! + p2D2 + p3D3 + p4D4 + p5D5 + p6D6 F.2
This model contains only main effects where the dummy variables are defined as follows.
DI = 0 if surface finish is not OSP
= 1 if surface finish is OSP
D2 = 0 if surface finish is not immersion Sn
= 1 if surface finish is immersion Sn
D3 = 0 if surface finish is not immersion Ag
= 1 if surface finish is immersion Ag
D4 = 0 if surface finish is not Ni / Au
= 1 if surface finish is Ni / Au
D5 = 0 if surface finish is not Ni / Pd / Au
= 1 if surface finish is Ni / Pd / Au
De = 0 if flux is not water soluble
= 1 if flux is water soluble
As before, the "base case" is obtained by setting all D; = 0, which is HASL with LR flux. Note
that the base case associated with the GLM in Equation F. 1 was also HASL with LR flux, but also
required Site 1. That requirement is not part of the latter model since sites are not included in the
model in Equation F.2.
As a final illustration of the flexibility of the GLM approach consider a subset of the data base that
only includes the results for Sites 1, 4, 5, 7, 11, 13, and 16 in Table 4.2. These sites were selected
because their surface finish was processed with both LR and WS fluxes, which allows an interaction
term to be added to the model in Equation F.2 for each surface finish and flux combination. However,
by excluding the other sites, the number of data points is reduced from 164 to 92.
F-2
-------
APPENDIX F
Example of GLM Analysis
The data base for the electrical responses incorporates the dummy variables used to define the
experimental parameters for each measurement. The data base contains 164 rows (one for each PWA).
Sample data base entries for the GLM in Equation F.2 for leakage measurement on the 10-mil pads
(response number 18 in Table 4.1) in logic ohms could appear as follows:
Row OSP Imm Sn Imm Ag Ni/Au Ni/Pd/Au Flux Leakage
1
2
3
4
0
1
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
1
0
1
0
1
12.8
11.9
12.1
11.8
The interpretation of these data base entries is as follows. The first row has zeros for OSP,
immersion Sn, immersion Ag, Ni/Au, and Ni/Pd/Au. This implies that the surface finish is HASL.
The surface finishes for rows 2, 3, and 4 are OSP, immersion Sn, and Ni/Pd/Au, respectively. Water
soluble flux is used on rows 2 and 4. The leakage measurements are given in the last column. The
above table would be expanded to include other experimental parameters or products (interactions) of
the experimental parameters depending on the requirements of the GLM such as given in Equation F.I.
The above table would also include columns containing the other 22 electrical measurements.
Computer software is used with the entries in the data base to find the least squares estimates of
coefficients in the GLM. For example, such an analysis for the GLM in Equation F.2 could produce
an estimated equation such as the following for leakage for the 10-mil pads.
Y = 12.5 - 0.200 OSP + 0.192 Immersion Sn - 0.164 Immersion Ag + 0.006 Ni/Au - 0.292 Ni/Pd/Au - 1.04 Flux
Note that the least squares process has simply solved a set of equations to determine an estimated
coefficient for each term appearing in the GLM in Equation F.2. However, it does not necessarily
follow that each of the terms in this estimated model makes a statistically significant contribution
toward explaining the variation in the leakage measurements. Rather, this determination is
accomplished by subjecting the coefficients in thefiill model to the following hypothesis test in a
sequential (stepwise) manner to determine if they are significantly different from 0:
H0: ft = 0 versus Hj: ft *0
If the coefficient is not significantly different from 0, it is eliminated from the model. Thus, the
only terms remaining in the model at the conclusion of this sequence of tests are those that are declared
to be significantly different from 0. This stepwise process eliminates some of the terms from the
model and the least squares calculations are repeated without those terms, which produces a reduced
model such as:
Y = 12.35 - 0.34 OSP - 0.38 Immersion Ag - 0.24 Ni/Pd/Au - 1.06 Flux
The intercept in this model, 12.35, is the estimated resistance for the base case—HASL processed with
LR flux. Mean predictions for other combinations of the experimental parameters can be made by
substituting the appropriate dummy variables into the model. For example, the mean prediction for a
OSP (Di=l, D2=0, D3=0, D4=0, D5=0) PWA processed with WS flux (D6=l) is found as:
F-3
-------
APPENDIX F
Y = 12.35 - 0.34 (1) - 1.06 (1) = 10.95
F.2 Overview of Test Results
Table F.I Anomaly Summary by Surface Finish after Exposure to 85/85
HASL
MSN
083-2
Site
1
Flux
WS
7
8
9
Circuit
HF PTH 50MHz
HF PTH f(-3dB)
HF PTH f(-40dB)
Test Technician Comments
Open PTH
Open PTH
Open PTH
OSP
056-4
Immersion Sn
030-4
032-4
086-2
102-4
5
9
8
7
10
LR
WS
LR
WS
WS
7
8
9
4
7
8
12
17
HF PTH 50MHz
HF PTH f(-3dB)
HF PTH f(-40dB)
HVLC SMT
HF PTH 50MHz
HF PTH f(-3dB)
HF SMT f(-40dB)
HF TLC RNR
Open PTH
Open PTH
Open PTH
Open PTH
Open PTH
Waveform
did not go to -40dB
Immersion Ag
082-2
094-4
11
12
LR
WS
21
7
Gull Wing
HF PTH 50MHz
Burnt etch
Open PTH
in multiple places
8 HFPTHf(-3dB)
9 HF PTH f(-40dB)
Open PTH
Open PTH
Ni/Au
013-1
015-4
13
14
LR
LR
6 HSDSMT
9 HF PTH f(-40dB)
Device failed, U3
Wrong value capacitor
Table F.
2 Anomaly Summary After Exposure to Thermal Shock
HASL
MSN
079-4
083-2
096-4
098-3
098-4
099-1
111-3
Site
1
1
3
3
3
3
3
Flux
WS
WS
WS
WS
WS
WS
WS
Circuit
12
7
8
9
10
11
12
10
11
12
10
11
12
11
12
23
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
SMT
PTH
PTH
PTH
SMT
SMT
SMT
SMT
SMT
SMT
SMT
SMT
SMT
SMT
SMT
f(-40dB)
50MHz
f(-3dB)
f(-40dB)
50MHz
f(-3dB)
f(-40dB)
50MHz
f(-3dB)
f(-40dB)
50MHz
f(-3dB)
f(-40dB)
f(-3dB)
f(-40dB)
Stranded Wire 2
Test Technician Comments
Open PTH
Open PTH
Open PTH
Open PTH
Open PTH
Open PTH
Open PTH
Open PTH
Open PTH
Open PTH
Open PTH
Open PTH
Waveform shifted
Distorted Waveform (does not quite go to -40dB, reads at-
3dB)
Minor
OSP
006-4
009-2
5
6
LR
LR
12
10
11
12
HF
HF
HF
HF
SMT
SMT
SMT
SMT
f(-40dB)
50MHz
f(-3dB)
f(-40dB)
Distorted waveform (goes to 40db but flattens and crosses
beyond 900mhz
Open PTH on coil
Open PTH on coil
Open PTH on coil
F-4
-------
APPENDIX F
014-3
056-2
056-4
058-1
060-1
060-2
Immersion
028-2
030-4
032-4
033-2
037-2
084-1
086-2
087-3
088-3
089-1
089-2
089-4
090-2
102-4
Immersion
071-1
072-1
5
5
5
5
5
5
Sn
9
9
8
8
9
7
7
7
7
7
7
7
7
10
Ag
11
11
LR
LR
LR
WS
WS
WS
LR
LR
LR
LR
LR
LR
WS
WS
WS
LR
WS
WS
WS
WS
WS
LR
LR
10
11
12
7
8
7
8
9
10
11
12
10
12
12
10
12
10
12
4
7
8
17
5
10
11
12
10
11
12
5
12
7
8
9
12
10
11
12
7
8
9
12
10
11
12
10
11
12
7
8
9
17
10
11
12
7
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT f(-40dB)
HF PTH 50MHz
HF PTH f(-3dB)
HF PTH 50MHz
HF PTH f(-3dB)
HF PTH f(-40dB)
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT f(-40dB)
HF SMT 50MHz
HF SMT f(-40dB)
HF SMT f(-40dB)
HF SMT 50MHz
HF SMT f(-40dB)
HF SMT 50MHz
HF SMT f(-40dB)
HVLC SMT
HF PTH 50MHz
HF PTH f(-3dB)
HF TLC RNR
HSD PTH
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT f(-40dB)
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT f(-40dB)
HSD PTH
HF SMT f(-40dB)
HF PTH 50MHz
HFPTHf(-3dB)
HFPTHf(-40dB)
HF SMT f(-40dB)
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT f(-40dB)
HF PTH 50MHz
HF PTH f(-3dB)
HF PTH f(-40dB)
HF SMT f(-40dB)
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT f(-40dB)
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT f(-40dB)
HF PTH 50MHz
HF PTH f(-3dB)
HF PTH f(-40dB)
HF TLC RNR
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT f(-40dB)
HF PTH 50MHz
Open PTH
Open PTH
Open PTH
Open PTH
Open PTH
2 open PTHs
2 open PTHs
2 open PTHs
2 open PTHs
2 open PTHs
2 open PTHs
Open PTH
Open PTH
Open PTH
Open PTH
Open PTH
Open PTH
Burnt etch (visual)
Open PTH
Open PTH
Likely component failure
Open PTH
Open PTH
Open PTH
Open PTH
Open PTH
Open PTH
Likely component failure
Distorted Waveform
High resistance on coil (acts
High resistance on coil (acts
High resistance on coil (acts
High resistance on coil (acts
Open PTH
Open PTH
Open PTH
Open PTH
Open PTH
Open PTH
Open PTH
High resistance on coil (acts
High resistance on coil (acts
High resistance on coil (acts
Open PTH
Open PTH
Open PTH
Open PTH on coil
Open PTH on coil
Open PTH on coil
Open PTH on coil
Open PTH on coil
Open PTH on coil
Open PTH
like open PTH)
like open PTH)
like open PTH)
like open PTH)
like open PTH)
like open PTH)
like open PTH)
F-5
-------
APPENDIX F
073-3 11 LR
082-2 1 1 WS
085-1 12 WS
085-2 12 WS
091-4 12 WS
094-1 12 WS
094-4 12 WS
8 HFPTHf(-3dB)
9 HF PTH f(-40dB)
7 HF PTH 50MHz
8 HFPTHf(-3dB)
9 HF PTH f(-40dB)
15 HRTLClGHz
12 HF SMT f(-40dB)
7 HF PTH 50MHz
8 HFPTHf(-3dB)
9 HF PTH f(-40dB)
10 HF SMT 50MHz
1 1 HF SMT f(-3dB)
12 HF SMT f(-40dB)
7 HF PTH 50MHz
8 HFPTHf(-3dB)
9 HF PTH f(-40dB)
10 HF SMT 50MHz
1 1 HF SMT f(-3dB)
12 HF SMT f(-40dB)
12 HF SMT f(-40dB)
7 HF PTH 50MHz
8 HFPTHf(-3dB)
9 HF PTH f(-40dB)
10 HF SMT 50MHz
1 1 HF SMT f(-3dB)
12 HF SMT f(-40dB)
7 HF PTH 50MHz
8 HFPTHf(-3dB)
9 HF PTH f(-40dB)
Open PTH
Open PTH
Open PTH
Open PTH
Open PTH
Burnt etch
Open PTH
Open PTH
Open PTH
Open PTH
Open PTH
Open PTH
Open PTH (2 places)
Open PTH (2 places)
Open PTH (2 places)
Open PTH (2 places)
Open PTH (2 places)
Open PTH (2 places)
Burnt Etch, High Resistance PTH,
Burnt Etch, High Resistance PTH,
Burnt Etch, High Resistance PTH,
Burnt Etch, High Resistance PTH,
Burnt Etch, High Resistance PTH,
Burnt Etch, High Resistance PTH,
Open PTH
Open PTH
Open PTH
and Open PTH
and Open PTH
and Open PTH
and Open PTH
and Open PTH
and Open PTH
Ni/Au
013-1 13 LR
015-2 14 LR
055-1 13 WS
6 HSDSMT
7 HF PTH 50MHz
8 HFPTHf(-3dB)
9 HF PTH f(-40dB)
7 HF PTH 50MHz
8 HFPTHf(-3dB)
9 HF PTH f(-40dB)
Device failed, U3
Open PTH on coil
Open PTH on coil
Open PTH on coil
Open PTH
Open PTH
Open PTH
Ni/Pd/Au
036-1 16 WS
6 HSDSMT
Likely component failure
Table F.3 Anomaly Summary After Mechanical Shock
(shaded entries signify carry over TS anomalies)
HASL
MSN Site Flux
039-2 2 LR
046-1 2 LR
046-2 2 LR
046-4 2 LR
076-1 1 LR
076-2 1 LR
079-4 1 WS
Circuit
12 HF SMT f(-40dB)
10 HF SMT 50MHz
1 1 HF SMT f(-3dB)
12 HF SMT f(-40dB)
10 HF SMT 50MHz
1 1 HF SMT f(-3dB)
12 HF SMT f(-40dB)
12 HF SMT f(-40dB)
10 HF SMT 50MHz
1 1 HF SMT f(-3dB)
12 HF SMT f(-40dB)
1 HCLVPTH
12 HF SMT f(-40dB)
Test Technician Comments
Waveform distorted
Open PTH
Open PTH
Distorted waveform
High resistance
Waveform does not go to -40dB
F-6
-------
APPENDIX F
080-4
083-2
096-4
098-2
098-3
098-4
099-1
099-4
100-3
1
1
3
3
o
J
3
o
J
o
J
o
J
WS
WS
WS
WS
WS
WS
WS
WS
WS
12
7
8
9
11
12
7
10
11
12
13
12
10
11
12
10
11
12
12
12
12
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
SMT
PTH
PTH
PTH
SMT
SMT
PTH
SMT
SMT
SMT
TLC
SMT
SMT
SMT
SMT
SMT
SMT
SMT
SMT
SMT
SMT
f(-40dB)
50MHz
f(-3dB)
f(-40dB)
f(-3dB)
f(-40dB)
f(-3dB)
50MHz
f(-3dB)
f(-40dB)
50MHz
f(-40dB)
50MHz
f(-3dB)
f(-40dB)
50MHz
f(-3dB)
f(-40dB)
f(-40dB)
f(-40dB)
f(-40dB)
Open PTH
Open PTH, distorted waveform
Open PTH
Open PTH
Waveform shifted
Distorted waveform
Distorted waveform
Distorted waveform
OSP
006-4
007-3
009-2
010-1
010-2
010-4
014-1
014-3
056-1
056-2
056-3
056-4
057-1
058-1
060-1
060-2
060-4
061-4
6
6
6
4
4
4
5
5
5
5
5
5
5
5
5
5
5
4
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
WS
WS
WS
WS
WS
WS
12
12
10
11
12
1
12
12
14
10
11
12
1
12
1
7
8
9
10
12
12
7
8
9
10
11
12
12
10
11
12
12
7
9
12
12
HF
HF
HF
HF
HF
SMT
SMT
SMT
SMT
SMT
f(-40dB)
f(-40dB)
50MHz
f(-3dB)
f(-40dB)
HCLVPTH
HF SMT f(-40dB)
HF
HF
HF
HF
HF
SMT
TLC
SMT
SMT
SMT
f(-40dB)
500MHz
50MHz
f(-3dB)
f(-40dB)
HCLVPTH
HF
SMT
f(-40dB)
HCLVPTH
HF PTH 50MHz
HFPTHf(-3dB)
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT f(-40dB)
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
HF
SMT
PTH
PTH
PTH
SMT
SMT
SMT
SMT
SMT
SMT
SMT
SMT
SMT
SMT
SMT
SMT
f(-40dB)
50MHz
f(-3dB)
f(-40dB)
50MHz
f(-3dB)
f(-40dB)
f(-40dB)
50MHz
f(-3dB)
f(-40dB)
f(-40dB)
50MHz
f(-40dB)
f(-40dB)
f(-40dB)
Distorted waveform
Open PTH
Distorted waveform
Open etch
Open PTH
Waveform does not go to -40 at the correct frequency
Open PTH
Waveform shifted
Open PTH - 2 places
Waveform does not go to -40dB
Open PTH
Distorted waveform
Open PTH
Distorted waveform
F-7
-------
APPENDIX F
062-1
062-4
065-1
065-4
Immersion
026-4
028-2
029-1
029-2
030-4
032-4
033-2
037-2
040-3
084-1
084-2
084-4
086-2
087-1
087-3
087-4
088-3
089-1
089-2
089-4
090-2
4
4
4
4
Sn
9
9
9
9
9
8
8
9
8
7
7
7
7
7
7
7
7
7
7
7
7
WS
ws
WS
ws
^m
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
LR
WS
WS
ws
WS
LR
WS
WS
WS
WS
12
12
12
12
•
5
10
11
12
1
17
9
7
9
17
5
10
11
12
9
10
11
12
9
10
11
12
10
11
12
15
1
12
12
8
10
11
12
12
10
11
12
7
8
9
12
10
11
12
7
8
10
11
12
7
8
10
11
HF SMT f(-40dB)
HF SMT f(-40dB)
HF SMT f(-40dB)
HF SMT f(-40dB)
^H
HSD PTH
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT f(-40dB)
HCLVPTH
HF TLC RNR
HF PTH f(-40dB)
HF PTH 50MHz
HF PTH f(-40dB)
HF TLC RNR
HSD PTH
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT f(-40dB)
HF PTH f(-40dB)
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT f(-40dB)
HF PTH f(-40dB)
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT f(-40dB)
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT f(-40dB)
HFTLClGHz
HCLVPTH
HF SMT f(-40dB)
HF SMT f(-40dB)
HFPTHf(-3dB)
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT f(-40dB)
HF SMT f(-40dB)
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT f(-40dB)
HF PTH 50MHz
HFPTHf(-3dB)
HF PTH f(-40dB)
HF SMT f(-40dB)
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT f(-40dB)
HF PTH 50MHz
HF PTH f(-3dB)
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT f(-40dB)
HF PTH 50MHz
HF PTH f(-3dB)
HF SMT 50MHz
HF SMT f(-3dB)
Distorted waveform
Waveform shifted
High resistance
Bad HSD PTH device
Open etch
Burnt etch (visual)
Open PTH
Open etch
Distorted waveform
Open PTH
Open PTH
Open PTH
Distorted waveform
Open PTH 2 places SMT & PTH
Distorted waveform
Open PTH
Open PTH
Waveform does not go to -40dB
Open PTH
Open PTH - 2 places
Open PTH 2 places SMT & PTH
F-8
-------
APPENDIX F
102-4
104-4
113-1
Immersion
072-1
072-2
072-4
073-3
075-2
075-3
082-2
082-3
085-1
085-2
091-4
094-1
094-3
094-4
095-4
10
10
10
Ag
11
11
11
11
11
11
11
11
12
12
12
12
12
12
12
WS
WS
WS
LR
LR
LR
LR
LR
LR
WS
WS
WS
WS
WS
WS
WS
WS
WS
12
17
12
10
11
12
7
8
9
12
12
7
8
9
12
13
10
12 |
13
12
7
8
9
10
11
1
7
8
9
10
11
12
1
10
11
12
7
8
9
10
11
12
13
9
12
13
17
1
7
8
9
10
11
12
13
1
HF SMT f(-40dB)
HF TLC RNR
HF SMT f(-40dB)
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT f(-40dB)
HF PTH 50MHz
HF PTH f(-3dB)
HF PTH f(-40dB)
HF SMT f(-40dB)
HF SMT f(-40dB)
HF PTH 50MHz
HF PTH f(-3dB)
HF PTH f(-40dB)
HF SMT f(-40dB)
HF TLC 50MHz
HF SMT 50MHz
HF SMT f(-40dB)
HF TLC 50MHz
HF SMT f(-40dB)
HF PTH 50MHz
HF PTH f(-3dB)
HF PTH f(-40dB)
HF SMT 50MHz
HF SMT f(-3dB)
HCLVPTH
HF PTH 50MHz
HF PTH f(-3dB)
HF PTH f(-40dB)
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT f(-40dB)
HCLV PTH
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT f(-40dB)
HF PTH 50MHz
HF PTH f(-3dB)
HF PTH f(-40dB)
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT f(-40dB)
HF TLC 50MHz
HF PTH f(-40dB)
HF SMT f(-40dB)
HF TLC 50MHz
HF TLC RNR
HCLVPTH
HF PTH 50MHz
HF PTH f(-3dB)
HF PTH f(-40dB)
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT f(-40dB)
HF TLC 50MHz
HCLV PTH
Open PTH
Open PTH
Waveform shifted
Waveform does not go to -40dB
Open PTH
Distorted waveform
Open PTH
Open PTH, distorted waveform
Open PTH - 2 places
Open PTH
Open etch
Open PTH - 2 places
Waveform distorted
Open PTH - 2 places
Open etch
F-9
-------
APPENDIX F
Ni/Au
013-1
015-2
051-2
054-4
055-1
055-4
13
14
13
13
13
13
LR
LR
WS
ws
WS
ws
6 HSDSMT
7 HFPTH 50MHz
9 HF PTH f(-40dB)
8 HF PTH f(-3dB)
8 HFPTHf(-3dB)
7 HFPTH 50MHz
8 HFPTHf(-3dB)
9 HF PTH f(-40dB)
12 HF SMT f(-40dB)
HSD device fail
Open etch
Open etch
Waveform distorted
Ni/Pd/Au
036-2 16 WS
12 HF SMT f(-40dB)
F.3 HCLV Circuitry
Pre-test measurements and deltas were analyzed with the GLM in Equation F. 1 for the main
effects site and flux and their interactions. These data were also subjected to a second GLM analysis
based on Equation F.2 for the main effects surface finish and flux. The base case for the GLM in
Equation F. 1 is defined as HASL at Site 1 and processed with LR flux. The base case for the GLM in
Equation F.2 is defined as HASL processed with LR flux.
Tables F.4 and F.5 summarize the results of these GLM analyses for HCLV PTH and HCLV
SMT. The upper portion of these tables contain the GLM results for Equation F.I while the lower
portion of these tables contain the GLM results for Equation F.2. The rows labeled "Constant" in
these tables contain the least squares estimates of P0 in Equations F.I and F.2 for each test time. The
numbers in the columns beneath the "Constants" are the estimated coefficients of the terms in
Equations F.I and F.2 that are significantly different from the base case. Shaded cells signify that the
corresponding term in the GLM is not significantly different from the base case.
The rows labeled Model R2 in Tables F.4 and F.5 show the percent of variation in the voltage
measurements explained by the respective estimated model. This value can range from 0% to 100%.
The model R2s for Equations F. 1 and F.2 for the HCLV circuitry are summarized as follows for each
test time.
GLM
Circuit
Pre-test
85/85 TS
MS
Site and Flux
Surface Finish and Flux
HCLV PTH
HCLV SMT
HCLV PTH
HCLV SMT
2.0%
4.2%
0.7%
1.5%
2.3%
7.7%
1.3%
0.3%
3.7%
10.9%
1.7%
9.8%
19.1%
2.1%
7.7%
0.7%
High R2 values would indicate a strong cause and effect relationship between the parameters of
surface finish, site, flux, and the voltage measurements at pretest. However, these R2s are all quite
small, which indicates that the experimental parameters: surface finish, site, and flux do not
significantly affect the HCLV voltage measurements at Pre-test nor do they affect the changes in the
voltage after exposure to each of the three test environments. That is, the HCLV measurements are
robust with respect to surface finish, site, and flux. The results for the two GLMs used in the analysis
are now examined in more detail.
GLM Results for Site and Flux
The uppermost portion of Table F.4 for HCLV PTH shows that only two experimental factors
(Site 2 and Site 8) are significantly different from the base case for the GLM in Equation F.I. The
F-10
-------
APPENDIX F
estimated GLM at Pre-test for Equation F. 1 is obtained from the estimated coefficients in the second
column of Table F.4 as:
Y = 7.14 + 0.06 Site2 + 0.07 Site 8
where Y represents the voltage response. The predicted voltage from this estimated model is 7.14V
for all site and flux combinations except Sites 2 and 8. The predictions for these two sites are 7.14V +
0.06V = 7.20V and 7.14V + 0.07V = 7.21V, respectively. Note that even though these two terms are
statistically significant, they represent very small changes from the base case voltage and, as such, are
not of practical interest. Moreover, the model R2 is only 2.0%, which has no practical value. Similar
comments hold for the GLM analyses at Pre-test for HCLV SMT.
Columns 3 to 5 in Tables F.4 and F.5 give the HCLV PTH and HCLV SMT GLM results for
Delta 1, 2, and 3, respectively. Note that these latter three analyses are based on changes in the voltage
measurements from Pre-test. The model R2 values after 85/85 and TS are also quite small, which
implies that the experimental parameters did not influence the HCLV measurements after exposure to
the 85/85, TS, and MS test environments.
In spite of the lack of significant experimental parameters in the HCLV GLMs, there is one very
interesting aspect of the model for HCLV SMT at Post MS. Note that the estimate of the constant term
in the last column of Table F.5 is 2.48, whereas, the estimated constants at Post 85/85 and Post TS
were 0.04 and 0.05, respectively. This is an increase of approximately 2.43V. The explanation of this
increase requires a review of the HCLV circuit, which is given in Section F. 10. In particular, Section
F. 10 explains that the HCLV circuit has seven 10Q resistors, RI, R2,..., R? in parallel. The overall
circuit resistance, Rtotai, is the parallel combination of these seven resistors, which is given as:
=J_+J_+J_+ +J_ = JL_ (F3)
Rtotal Rl R2 R2
D 10Q
^W=—
Since a current (I) of 5 A was applied to the circuit, Ohm's Law gives the resulting voltage (V) as
,.- „
(F.4)
4V (F.5)
During the MS test, it was noted that one to three of the resistors frequently fell off the board. In fact,
158 of the 164 PWAs were missing at least one of these resistors. If a single resistor is missing,
Equation F.5 would be revised as follows:
V = IR=5Ax^^= 8.33V (F.6)
6
Likewise, two missing resistors increase the voltage to 10V. Next consider the following dotplot of
voltage measurements at Post MS.
F-ll
-------
APPENDIX F
+ + + + + + Voltage
7.20 7.80 8.40 9.00 9.60 10.20
Note how the voltages are lumped around the points at 7.14V, 8.33V, and 10V, which
corresponds to the loss of no, one, or two resistors. Thus, the constant term in the GLM represents an
average increase in voltage of 2.48V over the nominal expected value of 7.14V, which is between one
and two missing resistors.
GLM Results for Surface Finish and Flux
The lower portion of Table F.4 for HCLV PTH shows that only one experimental factor
(Ni/Pd/Au) is significantly from the base case at Pre-test for the GLM in Equation F.2. The estimated
model is:
Y = 7.15-0.04Ni/Pd/Au
where Y represents the voltage response. The predicted voltage from this estimated model is 7.15V
for all surface finish and flux combinations except for Ni/Pd/Au processed with either flux, in which
case the prediction is decreased by 0.04V or 7.15V - 0.04V = 7.11V. As was just discussed with the
previous GLM, even though the coefficient for Ni/Pd/Au is statistically significant, it actually
represents a very small change from the base case and, as such, is not of practical interest. Moreover,
the model R2 is only 0.7%, which has no practical value. Similar comments hold for the GLM
analyses at Pre-test for HCLV SMT.
These low R2 values imply that the experimental parameters do not differ significantly from the
base case in terms of their impact on the voltage of the HCLV PTH and HCLV SMT circuits. That is,
there is no practical difference from the base case voltage measurements due to surface finish or flux
type. This result is to be expected since there were no difference among sites for these circuits in the
GLM analysis based on Equation F.I.
Columns 3 to 5 in Tables F.4 and F.5 give the HCLV PTH and HCLV SMT GLM results for
Delta 1, 2, and 3, respectively. The model R2 values at Post 85/85, Post TS, and Post MS are also quite
small, which implies that the experimental parameters did not influence the HCLV measurements after
exposure to the 85/85 and TS test environments. However, as just explained for the Site and Flux
model, the constant term in the last column of Table F.5 is affected by the missing resistors.
F-12
-------
APPENDIX F
Table F.4 Significant Coefficients for the Two GLM Analyses by Test Time for HCLV PTH
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
Site 3
Site 4
Site 5
Site 6
Site?
Site 8
Site 9
Site 10
Site 1 1
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 11 *Flux
Site 13 *Flux
Site 16 * Flux
Model R2
Standard Deviation
Pre-Test
7.14
0.06
2.0%
0.13
85/85
(Delta 1)
0.04
0.13
-0.16
2.3%
0.18
Thermal Shock
(Delta 2)
0.05
-0.17
3.7%
0.17
Mech Shock
(Delta 3)
0.14
0.80
19.1%
0.36
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
7.15
-0.04
0.7%
0.10
85/85
(Delta 1)
0.03
0.07
1.3%
0.10
Thermal Shock
(Delta 2)
0.04
0.07
1.7%
0.17
Mech Shock
(Delta 3)
0.13
0.34
7.7%
0.38
F-13
-------
APPENDIX F
Table F.5 Significant Coefficients for the Two GLM Analyses by Test Time for HCLV SMT
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
SiteS
Site 4
Site 5
Site 6
Site?
Site 8
Site 9
Site 10
Site 11
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 11 *Flux
Site 13* Flux
Site 16 * Flux
Model R2
Standard Deviation
Pre-Test
7.26
0.06
-0.07
4.2%
0.09
85/85
(Delta 1)
0.04
-0.09
-0.14
7.7%
0.12
Thermal Shock
(Delta 2)
0.05
-0.10
0.11
-0.11
10.9%
0.13
Mech Shock
(Delta 3)
2.48
-0.48
2.1%
0.70
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Pre-Test
85/85
(Delta 1)
Thermal Shock
(Delta 2)
Mech Shock
(Delta 3)
Constant
7.26
0.03
0.07
OSP
Immersion Sn
Immersion Ag
-0.08
-0.02
Ni/Au
Ni/Pd/Au
Flux
-0.10
-0.02
Model Rz
Standard Deviation
1.5%
0.09
0.3%
0.1
9.8%
0.13
0.7%
0.70
F-14
-------
APPENDIX F
F.4 HVLC Circuitry
Results of the GLM analyses for HVLC PTH and HVLC SMT circuits are given in Tables F.6 and
F.7, respectively. Columns 3 to 5 in these tables give the GLM results for 85/85, TS, and MS,
respectively. The model R2s for Equations F. 1 and F.2 for the HVLC circuitry are summarized as
follows for each test time.
GLM Circuit Pre-test 85/85 TS MS
Site and Flux
Surface Finish and Flux
HVLC PTH
HVLC SMT
HVLC PTH
HVLC SMT
13.3%
20.9%
7.6%
14.0%
5.2%
14.0%
2.5%
15.3%
0.0%
18.7%
2.6%
12.9%
3.2%
NA
3.2%
NA
These model R2 values are generally higher that those observed for the HCLV measurements.
However, the magnitudes of the coefficients were too small to be of practical significance relative to
the JTP acceptance criteria, which indicates that these parameters do not influence the HVLC
measurements. To further explain this point, consider the coefficients for site and flux in Table F.6 at
Pre-test where the constant term is 5.018|iA. The largest coefficient at Pre-test is -0.008|iA for the
interaction of Site 4 and Flux. Thus, this interaction can decrease the constant term to 5.018|iA -
0.008|iA = 5.010|iA, which is so far from the lower and upper limits of 4 |iA and 6|iA that it is not of
practical interest. Note that there are no R2 values listed for HVLC SMT at Post MS. This is due to
resistors coming off the PWA during the MS test, which caused the HVLC SMT circuit to give a
constant response for reasons that will now be explained.
Boxplot Displays of Multiple Comparison Results
Figures F.I to F.8 give boxplots for the HVLC PTH and SMT circuits. It is important to keep the
vertical scale in mind relative to the acceptance criteria when viewing these boxplots. That is, the
acceptance criteria indicates that the current should be between 4|iA and 6|iA. These boxplots are
centered close to 5|iA and the total spread is on the order of 0.02|iA for the PTH circuits and
approximately 0.5|iA for SMT circuits. Hence, even though there are some statistically significantly
differences, they are not likely to be of practical concern. Note the boxplots in Figure F.8 for HCLV
SMT at Post MS. These values are all either 0|iA for very close to it, reflecting the fact that the
resistors came off the PWA during the MS test.
F-15
-------
APPENDIX F
Table F.6 Significant Coefficients for the Two GLM Analyses by Test Time for HVLC PTH
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
SiteS
Site 4
Site 5
Site 6
Site?
SiteS
Site 9
Site 10
Site 1 1
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 1 1 * Flux
Site 13 *Flux
Site 16* Flux
Model R2
Standard Deviation
P re-Test
5.018
0.007
0.005
0.004
0.004
-0.008
13.3%
0.005
85/85
5.004
0.006
0.006
5.2%
0.006
Thermal Shock
4.999
0.0%
0.006
Mech Shock
4.998
-0.005
3.2%
0.006
GLM from Eq. F.2: Surface Finishes and Flux
OSP
Immersion Sn
Immersion Ag
Model R2
Standard Deviation
F-16
-------
APPENDIX F
Table F.7 Significant Coefficients for the Two GLM Analyses by Test Time for HVLC SMT
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
Site 3
Site 4
Site 5
Site 6
Site?
Site 8
Site 9
Site 10
Site 1 1
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 11 *Flux
Site 13 *Flux
Site 16 * Flux
Model R2
Standard Deviation
P re-Test
5.038
0.172
0.111
0.122
0.125
20.9%
0.100
85/85
5.034
0.173
0.111
0.125
0.126
21.5%
0.100
Thermal Shock
5.039
0.170
0.109
0.120
0.125
18.7%
0.112
Mech Shock
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
5.032
0.095
0.087
14.0%
0.100
85/85
5.027
0.100
0.090
15.3%
0.100
Thermal Shock
5.033
0.097
0.085
12.9%
0.110
Mech Shock
F-17
-------
APPENDIX F
F.5 HSD Circuitry
The complete results of the GLM analyses are given in Tables F.8 and F.9, respectively. Columns
3 to 5 in these tables give the GLM results for 85/85, TS, and MS, respectively. Note that these latter
three analyses are based on changes in total propagation delay from Pre-test. The model R2s for
Equations F. 1 and F.2 for the HSD circuitry are summarized as follows for each test time.
GLM Circuit Pre-test 85/85 TS MS
Site and Flux
Surface Finish and Flux
HSD PTH
HSD SMT
HSD PTH
HSD SMT
5.1%
6.1%
0.9%
1.0%
9.8%
6.4%
1.6%
0.3%
4.3%
0.0%
1.8%
0.8%
9.5%
2.3%
6.7%
0.2%
All these model R2 values are quite small at each test time, which indicates that the experimental
parameters under evaluation do not influence the HSD total propagation delay measurements.
Boxplot Displays of Multiple Comparison Results
Figures F.9 and F.10 give boxplots of Pre-test measurements of total propagation delay for the
HSD PTH and HSD SMT circuits, respectively. Note that most total propagation delays in Figure F.9
for HSD PTH are a little over 17 ns with a range of about Ins. Figure F. 10 shows that the total
propagation delays for HSD SMT have a range of about 0.4ns and are centered about 9.2ns. The
percentage changes in the total propagation delay measurements were small and well within the
acceptance criteria so boxplot displays of these measurements are not presented.
F-18
-------
APPENDIX F
Table F.8 Significant Coefficients for the Two GLM Analyses by Test Time for HSD PTH
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
Site 3
Site 4
Pre-Test
17.13
0.14
Site 5
Site 6
Site?
Site 8
Site 9
Site 10
Site 1 1
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 1 1 * Flux
Site 13* Flux
Site 16* Flux
Model R2
Standard Deviation
0.19
5.1%
0.19
85/85
(Delta 1)
0.55
0.61
1.89
9.8%
1.30
Thermal Shock
(Delta 2)
0.98
-0.46
-1.00
4.3%
1.33
Mech Shock
(Delta 3)
0.37
2.60
-2.30
-3.50
9.5%
3.52
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
17.13
0.05
0.9%
0.20
85/85
(Delta 1)
0.88
-0.35
1.6%
1.00
Thermal Shock
(Delta 2)
0.88
-0.36
1.8%
1.30
Mech Shock
(Delta 3)
0.52
-2.89
6.7%
3.5
F-19
-------
APPENDIX F
Table F.9 Significant Coefficients for the Two GLM Analyses by Test Time for HSD SMT
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
Site 3
Site 4
Site 5
Site 6
Site?
Site 8
Site 9
Site 10
Site 1 1
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 1 1 * Flux
Site 13 *Flux
Site 16* Flux
Model R2
Standard Deviation
Pre-Test
9.23
0.12
-0.10
6.1%
0.13
85/85
(Delta 1)
0.94
-1.59
-1.27
6.4%
1.65
Thermal Shock
(Delta 2)
1.16
0.0%
1.99
Mech Shock
(Delta 3)
-0.002
-1.60
2.3%
2.25
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Pre-Test
9.21
85/85
(Delta 1)
0.77
Ni/Pd/Au 0.35
Flux
Model R2
Standard Deviation
0.03
1.0%
0.10
0.3%
1.00
Thermal Shock
(Delta 2)
1.23
-0.56
0.8%
1.90
Mech Shock
(Delta 3)
-0.04
-0.25
0.2%
2.2
F-20
-------
APPENDIX F
F.6 HF LPF Circuitry
Pre-test measurements for all HF LPF circuits were subjected to GLM analyses, as were the deltas
after 85/85, TS, and MS. The results of the GLM analyses are given in Tables F. 10 to F. 15. Columns
3 to 5 in these tables give the GLM results for 85/85, TS, and MS, respectively.
Note that these latter three analyses are based on changes from Pre-test measurements. The model
R2s for Equations F.I and F.2 for the FTP LPF circuitry are summarized as follows for each test time.
GLM Circuit Pre-test 85/85 TS MS
Site and Flux
Surface Finish and Flux
PTH 50MHz
PTH f(-3dB)
PTH f(-40dB)
SMT 50MHz
SMT f(-3dB)
SMT f(-40dB)
PTH 50MHz
PTH f(-3dB)
PTH f(-40dB)
SMT 50MHz
SMT f(-3dB)
SMT f(-40dB)
20.6%
7.1%
14.3%
3.9%
8.8%
5.3%
4.3%
7.8%
4.5%
2.7%
0.7%
5.2%
29.5%
10.8%
9.6%
10.3%
10.5%
2.3%
2.3%
0.2%
1.8%
0.6%
1.5%
0.3%
24.1%
10.2%
7.6%
21.1%
19.1%
16.1%
0.3%
1.6%
1.6%
0.8%
5.0%
4.9%
20.5%
23.4%
13.5%
32.2%
14.3%
29.4%
8.1%
10.9%
10.9%
6.1%
3.0%
14.4%
The model R2 values are quite small at Pre-test, which indicates that the parameters under
evaluation do not influence the HF LPF measurements. The same is true at Post 85/85. The model R2
values are also quite small at Post TS and Post MS. However, the test measurements contained many
extreme outlying observations at both of these later two test times, which greatly increases the sample
variance and in turn hinders the interpretation of the GLM results. As indicated in Tables F. 1, F.2, and
F.3 there were many anomalous HF LPF test measurements (171 at Post MS).
Boxplot Displays of Multiple Comparison Results
Boxplot displays of all test results for HF LPF circuits have been created to aid in the
interpretation of the results. Figures 4.9 to 4.15 in Chapter 4 show the boxplots for the analyses with
significant differences or values not meeting acceptance criteria. Figures F. 11 to F.27 show all
remaining boxplots associated with the HF LPF results.
F-21
-------
APPENDIX F
Table F.10 Significant Coefficients for the Two GLM Analyses by Test Time for HF PTH 50 MHz
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
Site 3
Site 4
Site 5
Site 6
Site?
Site 8
Site 9
Site 10
Site 1 1
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 1 1 * Flux
Site 13 *Flux
Site 16* Flux
Model R2
Standard Deviation
Pre-Test
-0.721
-0.180
0.160
20.6%
0.055
85/85
(Delta 1)
-0.034
0.197
-0.206
29.5%
0.048
Thermal Shock
(Delta 2)
-0.002
0.192
-0.073
-0.180
24.1%
0.063
Mech Shock
(Delta 3)
-2.666
-28.1
-18.5
20.5%
14.1
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Pre-Test
85/85
(Delta 1)
Thermal Shock
(Delta 2)
Mech Shock
(Delta 3)
Constant
-0.720
-0.034
0.003
-3.28
OSP
Immersion Sn
Immersion Ag
-0.010
-13.6
Ni/Au
Ni/Pd/Au
Flux
-0.034
0.023
Model R2
Standard Deviation
4.3%
0.060
2.3%
0.050
0.3%
0.072
8.1%
15.00
F-22
-------
APPENDIX F
Table F.ll Significant Coefficients for the Two GLM Analyses by Test Time for HF PTH f(-3dB)
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
Site 3
Site 4
Site 5
Site 6
Site?
Site 8
Site 9
Site 10
Site 1 1
Site 12
Site 13
Pre-Test
283.0
-1.8
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 1 1 * Flux
Site 13 *Flux
Site 16* Flux
Model R2
Standard Deviation
-1.5
7.1%
2.0
85/85
(Delta 1)
-0.9
0.7
-1.2
II
10.8%
0.9
Thermal Shock
(Delta 2)
0.5
-2.2
10.2%
1.5
Mech Shock
(Delta 3)
-1.05
-116
-68
-79
23.4%
58.5
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Pre-Test
85/85
(Delta 1)
Thermal Shock
(Delta 2)
Mech Shock
(Delta 3)
Constant
283.0
-1.0
0.5
4.19
OSP
Immersion Sn
Immersion Ag
0.1
-0.5
Ni/Au
Ni/Pd/Au
Flux
-l.C
Model R2
Standard Deviation
7.8%
2.0
0.2%
0.9
1.6%
1.5
10.9%
62.0
F-23
-------
APPENDIX F
Table F.12 Significant Coefficients for the Two GLM Analyses by Test Time for HF PTH f(-40dB)
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
Site 3
Site 4
Site 5
Site 6
Site?
Site 8
Site 9
Site 10
Site 1 1
Site 12
Site 13
Pre-Test
472.9
-3.8
-5.7
-5.1
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 1 1 * Flux
Site 13 *Flux
Site 16* Flux
Model R2
Standard Deviation
-4.5
14.3%
5.1
85/85
(Delta 1)
-0.2
0.9
-1.5
9.6%
1.2
Thermal Shock
(Delta 2)
-0.2
-1.8
2.6
7.6%
1.5
Mech Shock
(Delta 3)
-11.7
-140
13.5%
77.1
GLM from Eq. F.2: Surface Finishes and Flux
Thermal Shock
(Delta 2)
OSP
Immersion Sn
Immersion Ag
Model Rz
Standard Deviation
F-24
-------
APPENDIX F
Table F.13 Significant Coefficients for the Two GLM Analyses by Test Time for HF SMT 50 MHz
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
Site 3
Site 4
Site 5
Site 6
Site?
Site 8
Site 9
Site 10
Site 1 1
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 1 1 * Flux
Site 13 *Flux
Site 16* Flux
Model R2
Standard Deviation
Pre-Test
-0.733
0.031
0.021
3.9%
0.039
85/85
(Delta 1)
-0.018
-0.049
-0.047
10.3%
0.037
Thermal Shock
(Delta 2)
0.005
-0.112
-0.126
21.1%
0.069
Mech Shock
(Delta 3)
-3.1
-19.2
-13.5
-49.7
-31.4
25.0
32.2%
17.2
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
-0.733
0.020
2.7%
0.030
85/85
(Delta 1)
-0.023
0.008
0.6%
0.030
Thermal Shock
(Delta 2)
-0.010
0.017
0.8%
0.077
Mech Shock
(Delta 3)
-5.62
-10.6
-10.7
6.1%
20.0
F-25
-------
APPENDIX F
Table F.14 Significant Coefficients for the Two GLM Analyses by Test Time for HF SMT f(-3dB)
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Pre-Test
85/85
(Delta 1)
Thermal Shock
(Delta 2)
Mech Shock
(Delta 3)
Constant
-1.3
0.7
Flux
Site 2
Site 3
-15.5
Site 4
Site 5
Site 6
Site?
Site 8
Site 9
Site 10
Site 11
Site 12
1.5
Site 13
Site 14
Site 15
3.7
-3.9
Site 16
Site 4 * Flux
Site 5 * Flux
-3.7
Site 7 * Flux
Site 11 * Flux
Site 13 *Flux
Site 16* Flux
11.9
-4.4
-102
Model R2
Standard Deviation
8.8%
1.9
10.5%
1.1
19.1%
4.7
14.3%
112
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Pre-Test
85/85
(Delta 1)
Thermal Shock
(Delta 2)
Mech Shock
(Delta 3)
Constant
319.7
-1.3
OSP
Immersion Sn
Immersion Ag
0.4
0.5
Ni/Au
Ni/Pd/Au
Flux
-41.0
Model R2
Standard Deviation
0.7%
2.0
1.5%
1.0
5.0%
5.0
3.0%
11.0
F-26
-------
APPENDIX F
Table F.15 Significant Coefficients for the Two GLM Analyses by Test Time for HF SMT f(-40dB)
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
Site 3
Site 4
Site 5
Site 6
Site?
Site 8
Site 9
Site 10
Site 1 1
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 1 1 * Flux
Site 13 *Flux
Site 16* Flux
Model R2
Standard Deviation
Pre-Test
865.5
-10.7
85/85
(Delta 1)
1.7
4.9
2.2
-19.7
5.3%
21.0
2.3%
7.6
Thermal Shock
(Delta 2)
-8.1
-23.7
•
16.1%
9.1
Mech Shock
(Delta 3)
-80.3
-244
-171
-430
-365
29.4%
221
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Pre-Test
85/85
(Delta 1)
Thermal Shock
(Delta 2)
Mech Shock
(Delta 3)
Constant
861.2
2.0
-6.8
-146.2
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
13.4
-4.4
192.0
171.0
-117.0
Model R2
Standard Deviation
5.2%
21.0
0.3%
7.0
4.9%
9.7
14.4%
24.0
F-27
-------
APPENDIX F
F.7 HF TLC Circuitry
Pre-test measurements for all HF TLC circuits except RNF were subjected to GLM analyses, as
were the deltas after 85/85, TS, and MS. The results of the GLM analyses are given in Tables F. 16 to
F.20. Columns 3 to 5 in those tables give the HF TLC PTH and HF TLC SMT GLM results for 85/85,
TS, and MS, respectively. Note that these latter three analyses are based on changes from Pre-test
measurements. The model R2s for Equations F. 1 and F.2 for the HF TLC circuitry are summarized as
follows for each test time, except for HF TLC RNF, which gave a constant response.
GLM
Circuit
Pre-test
85/85 TS
MS
Site and Flux
50MHz
500MHz
IGHz
RNF
RNR
62.3%
10.7%
13.2%
•
2.7%
6.7% 0.0%
8.1% 0.0%
10.9% 6.1%
3.2%
2.4%
14.7%
8.1%
7.9%
•
6.2%
Surface Finish and Flux
50MHz
500MHz
IGHz
RNF
RNR
48.1%
2.5%
0.9%
•
3.6%
6.6%
0.9%
2,8%
•
0.6%
5.0%
1.8%
4.1%
•
3.5%
9.1%
1.4%
0.7%
^
2.0%
The model R2 values for HF TLC are all quite small at Pre-test except for those at 50MHz, which
are of moderate size. The small R2 values indicate that the experimental parameters do not influence
the Pre-test HF TLC measurements. The moderate sized R2 values for the 50MHz case are examined
in further detail below (repeated from Chapter 4).
The predicted response at Pre-test for HF TLC 50MHz for the base case (HASL at Site 1
processed with LR flux) based on the Site & Flux GLM was -47.43dB. The predicted differences from
the base case are given in Appendix F in Table F.21. The results show that the sites that produced
Ni/Au and Ni/Au/Pd (#13-16) have predicted increases of less than 3dB. While statistically
significant, this change is rather small compared to the base case value and is probably not of practical
utility. Overall, some of the sites differ from the base case by approximately -1.5dB to 2.9dB. These
changes again may not have any practical significance since the important concept is not so much the
magnitude of the response, but rather its stability when subject to environmental stress conditions,
which is the basis for the acceptance criteria.
The predicted response at Pre-test for HF TLC 50MHz for the base case (HASL processed with
LR flux) based on the Surface Finish & Flux GLM was -46.73dB, which is almost identical to that for
the Site & Flux GLM. The predicted differences from the base case are given in Appendix F in Table
F.22. These predictions are consistent with those in Table F.21 and show that immersion Sn and
immersion Ag are approximately 1 .OdB lower than the base case and Ni/Au and Ni/Pd/Au are
approximately 1 to 2 dB higher than the base case. Again, these differences are most likely not of
practical utility.
Boxplot Displays of Multiple Comparison Results
HF TLC 50MHz. A boxplot display of the Post MS test results is given in Figure 4.16. Boxplots
for the other three test times are displayed in Figures F.28 to F.30.
HF TLC 500MHz. A boxplot display of the Post MS test results is given in Figure 4.17.
Boxplots for the other three test times are displayed in Figures F.31 to F.33.
F-28
-------
APPENDIX F
HF TLC IGHz. Boxplots displays for are not given for the HF TLC IGHz test results to
conserve space. The total variation at Pre-test for HF TLC IGHz was only 2dB and there was only
one slight anomaly of-5dB at Post MS, which is not of concern.
HF TLC RNR. A boxplot display of the Post MS test results is given in Figure 4.18. Boxplots
for the other three test times are displayed in Figures F.34 to F.36.
Table F.16 Significant Coefficients for the Two GLM Analyses by Test Time for HF TLC 50 MHz Forward
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Pre-Test
85/85
(Delta 1)
Thermal Shock
(Delta 2)
Mech Shock
(Delta 3)
Constant
0.22
-0.08
0.04
Flux
Site 2
Site 3
4.40
Site 4
Site 5
Site 6
Site?
Site 8
Site 9
Site 10
Site 11
Site 12
3.20
7.60
Site 13
Site 14
Site 15
-1.17
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 11 * Flux
Site 13 *Flux
Site 16 * Flux
Model R2
Standard Deviation
0.0%
1.01
14.7%
4.80
GLM from Eq. F.2: Surface Finishes and Flux
Model Rz
Standard Deviation
Experimental Factor
Thermal Shock
(Delta 2)
Mech Shock
(Delta 3)
Constant
0.29
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
F-29
-------
APPENDIX F
Table F.17 Significant Coefficients for the Two GLM Analyses by Test Time for HF TLC 500 MHz Forward
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
Site 3
Site 4
Site 5
Site 6
Site?
Site 8
Site 9
Pre-Test
-17.48
0.64
0.45
0.53
Site 10 0.56
Site 1 1
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 11 *Flux
Site 13 *Flux
Site 16 * Flux
Model R2
Standard Deviation
10.7%
0.66
85/85
(Delta 1)
0.06
-1.13
1.35
8.1%
0.62
Thermal Shock
(Delta 2)
-0.23
0.0%
0.60
Mech Shock
(Delta 3)
-0.14
-1.32
-0.85
1.50
8.1%
0.93
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Pre-Test
85/85
(Delta 1)
Thermal Shock
(Delta 2)
Mech Shock
(Delta 3)
Constant
-17.41
0.02
OSP
Immersion Sn
Immersion Ag
0.27
-0.28
0.20
-0.09
Ni/Au
Ni/Pd/Au
Flux
0.23
-0.22
Model R2
Standard Deviation
2.5%
0.60
0.9%
0.60
1.8%
0.59
1.4%
0.96
F-30
-------
APPENDIX F
Table F.18 Significant Coefficients for the Two GLM Analyses by Test Time for HF TLC 1 GHz Forward
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
Site 3
Site 4
Site 5
Site 6
Site?
Site 8
Site 9
Site 10
Site 1 1
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 1 1 * Flux
Site 13 *Flux
Site 16* Flux
Model R2
Standard Deviation
Pre-Test
-14.11
-0.16
-0.30
0.37
0.21
0.46
13.2%
0.37
85/85
(Delta 1)
0.11
-0.46
-0.35
0.59
10.9%
0.31
Thermal Shock
(Delta 2)
-0.39
-0.51
6.1%
0.52
Mech Shock
(Delta 3)
-0.22
-1.26
1.00
7.9%
0.69
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
-14.16
0.09
0.9%
0.30
85/85
(Delta 1)
0.11
-0.15
2.8%
0.30
Thermal Shock
(Delta 2)
-0.38
-0.33
4.1%
0.52
Mech Shock
(Delta 3)
-0.30
0.14
0.7%
0.71
F-31
-------
APPENDIX F
Table F.19 Significant Coefficients for the Two GLM Analyses by Test Time for HF TLC Rev Null Freq
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
Site 3
Site 4
SiteS
Site 6
Site?
Site 8
Site 9
Site 10
Site 1 1
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 1 1 * Flux
Site 13 *Flux
Site 16* Flux
Model R2
Standard Deviation
P re-Test
85/85
(Delta 1)
Thermal Shock
(Delta 2)
Mech Shock
(Delta 3)
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
85/85
(Delta 1)
Thermal Shock
(Delta 2)
Mech Shock
(Delta 3)
F-32
-------
APPENDIX F
Table F.20 Significant Coefficients for the Two GLM Analyses by Test Time for HF TLC Rev Null Resp
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
Site 3
Site 4
Site 5
Site 6
Site?
Site 8
Site 9
Site 10
Site 1 1
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 11 *Flux
Site 13 *Flux
Site 16 * Flux
Model R2
Standard Deviation
P re-Test
-33.90
1.13
-1.25
2.7%
1.40
85/85
(Delta 1)
0.20
-3.23
3.60
8.2%
1.70
Thermal Shock
(Delta 2)
-0.05
-1.60
2.4%
2.20
Mech Shock
(Delta 3)
0.02
-3.50
6.2%
3.56
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Thermal Shock
(Delta 2)
Mech Shock
(Delta 3)
Constant
0.03
OSP
Immersion Sn
Immersion Ag
-1.26
-0.74
Ni/Au
Ni/Pd/Au
Flux
1.03
Model R2
Standard Deviation
3.6%
1.00
0.6%
1.00
3.5%
2.1
2.0%
3.6
F-33
-------
APPENDIX F
Table F.21 Predicted Changes from the Base Case at Pre-test for HF TLC 50MHz for the GLM in
Equation F.I
LRFlux WSFlux
Site 2
Site 3
Site 4
SiteS
Site 6
Site?
SiteS
Site 9
Site 10
Site 11
Site 12
Site 13
Site 14
Site 15
Site 16
0.98
1.19
1.48
-1.51
0.90
-1.40
2.90
2.69
2.05
2.19
0.98
-0.18
1.48
-1.51
0.90
-1.40
2.90
2.69
2.05
0.69
Table F.22 Predicted Changes from the Base Case at Pre-test for HF TLC 50MHz
for the GLM in Equation F.2
LRFlux WSFlux
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
-0.71
-0.97
2.24
1.19
-0.59
-1.30
-1.56
1.65
0.60
F.8 Leakage Measurements
The results of the GLM analyses are given in Tables F.23 to F.26. Columns 3 to 5 in these tables
give the GLM results for 85/85, TS, and MS, respectively. The model R2s for Equations F. 1 and F.2
for the GLM analyses of the leakage measurements are summarized as follows.
GLM Circuit Pre-test 85/85 TS MS
Site and Flux
Surface Finish and Flux
10-MilPads
PGA-A
PGA-B
Gull Wing
10-MilPads
PGA-A
PGA-B
Gull Wing
85.6%
88.4%
89.4%
55.4%
74.8%
81.3%
88.7%
48.2%
22.7%
3.9%
5.6%
3.3%
1.9%
2.0%
5.6%
1.9%
10.8%
9.7%
15.5%
2.8%
3.4%
9.7%
16.0%
2.8%
8.6%
9.0%
12.5%
1.7%
1.7%
6.3%
6.7%
2.6%
It is of interest to note that the model R2 values at Pre-test for all but the Gull Wing are all quite
large. However, these values decrease to close to zero after exposure to the 85/85 environment. These
results are now examined in detail for each of the four leakage circuits.
Tables F.27 and F.28 give the predicted changes from their respective base cases for all leakage
measurements at Pre-test for the GLMs in Equations F. 1 and F.2, respectively.
F-34
-------
APPENDIX F
Table F.23 Significant Coefficients for the Two GLM Analyses by Test Time for 10-Mil Pads
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
SiteS
Site 4
Site 5
Site 6
Site?
Site 8
Site 9
Site 10
Pre-Test
12.20
0.74
-0.97
1.02
0.93
0.85
1.00
Site 1 1
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 1 1 * Flux
Site 13* Flux
Site 16* Flux
Model R2
Standard Deviation
0.91
-0.89
-0.75
0.98
-0.76
0.85
1.06
1.95
1.74
85.6%
0.42
85/85
13.29
-1.24
0.23
22.7%
0.51
Thermal Shock
14.45
-0.95
0.55
M
10.8%
0.70
Mech Shock
14.76
-0.84
8.6%
0.59
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Pre-Test
11.75
0.73
0.33
0.48
Ni/Pd/Au
Flux
Model R2
Standard Deviation
1.77
74.8%
0.50
85/85
13.21
0.21
1.9%
0.50
Thermal Shock
14.30
Mech Shock
14.69
0.31
0.27
3.4%
0.72
1.7%
0.61
F-35
-------
APPENDIX F
Table F.24 Significant Coefficients for the Two GLM Analyses by Test Time for PGA-A
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Pre-Test
85/85
Thermal Shock
Mech Shock
Constant
11.88
12.50
13.66
13.69
Flux
Site 2
Site 3
1.58
-1.19
0.348
0.22
Site 4
Site 5
Site 6
-0.54
Site?
Site8
Site 9
-0.81
Site 10
Site 11
Site 12
-0.34
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
-0.64
-0.94
0.63
Site 7 * Flux
Site 11 * Flux
Site 13 * Flux
Site 16 * Flux
Model R2
Standard Deviation
3.9%
0.71
9.7%
0.52
9.0%
0.49
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Pre-Test
85/85
Thermal Shock
Mech Shock
Constant
11.38
12.41
13.66
13.66
OSP
Immersion Sn
Immersion Ag
0.35
0.25
Ni/Au
Ni/Pd/Au
Flux
-0.35
2.05
0.34
0.256
Model R2
Standard Deviation
81.3%
0.5
2.0%
0.70
9.7%
0.51
6.3%
0.49
F-36
-------
APPENDIX F
Table F.25 Significant Coefficients for th
GLM from Eq
Experimental Factor
Constant
Flux
Site 2
SiteS
Site 4
SiteS
Site 6
Site?
SiteS
Site 9
Site 10
Site 11
Site 12
Site 13
Site 14
Site 15
Pre-Test
10.71
2.77
0.57
Site 16 -0.34
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 1 1 * Flux
Site 13 * Flux
Site 16 * Flux
Model R2
Standard Deviation
89.4%
0.47
e Two GLM Analyses by Test Time for PGA-B
. F.I: Sites and Interactions with Flux
85/85
12.52
-0.41
-0.61
0.72
8.0%
0.53
Thermal Shock
13.69
0.40
-0.44
15.5%
0.56
Mech Shock
13.83
-0.49
-0.63
-0.42
0.69
12.5%
0.50
GLM from Eq
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
Pre-Test
10.77
-0.38
2.71
88.7%
0.4
. F.2: Surface Finishes and Flux
85/85
12.55
-0.23
-0.40
5.6%
0.50
Thermal Shock
13.72
-0.33
0.39
16.0%
0.56
Mech Shock
13.70
-0.21
0.20
6.7%
0.51
F-37
-------
APPENDIX F
Table F.26 Significant Coefficients for the Ti
GLM from Eq
Experimental Factor
Constant
Flux
Site 2
SiteS
Site 4
SiteS
Site 6
Site?
SiteS
Site 9
Site 10
Site 11
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 1 1 * Flux
Site 13 * Flux
Site 16 * Flux
Model R2
Standard Deviation
Pre-Test
11.72
0.81
0.37
0.47
-0.65
0.54
0.47
1.61
55.4%
0.54
ivo GLM Analyses by Test Time for the Gull Wing
. F.I: Sites and Interactions with Flux
85/85
12.59
0.67
0.66
3.3%
1.1
Thermal Shock
13.76
-0.37
2.8%
1.10
Mech Shock
13.32
-0.64
1.7%
1.06
GLM from Eq. F.2: Surface Finishes and Flux
Model Rz
Standard Deviation
Experimental Factor
Constant
OSP
Immersion Sn
Immersion Ag
Ni/Au
Ni/Pd/Au
Flux
F-38
-------
APPENDIX F
Table F.27 Predicted Changes from the Base Case at Pre-test for the Leakage Measurements for the GLM in
Equation F.I
10-MilPads PGA-A PGA-B Gull Wing
LRFlux WSFlux LRFlux WS Flux LRFlux WS Flux LRFlux WS Flux
Site 2
SiteS
Site 4
Site 5
Site 6
Site?
Site 8
Site 9
Site 10
Site 1 1
Site 12
Site 13
Site 14
Site 15
Site 16
-0.97
1.02
0.93
0.85
0.91
-0.89
-0.75
0.98
-0.76
-0.23
1.76
1.67
1.59
0.74
1.59
0.74
0.74
1.74
1.80
1.65
1.80
-0.01
1.72
1.72
-1.19
-0.81
-0.34
-0.64
-0.94
-1.14
0.39
1.58
1.58
1.58
1.58
1.58
1.58
0.77
1.58
1.24
1.58
1.85
0.64
1.58
1.78
2.77
2.77
2.77
2.77
2.77
2.77
0.57 3.34
2.77
2.77
2.77
2.77
2.77
2.77
2.77
-0.34 2.43
0.37
0.47
-0.65
0.54
0.81
0.81
0.81
1.18
0.81
1.28
0.81
0.81
1.28
1.77
1.35
0.81
0.81
0.81
0.81
Table F.28 Predicted Changes from the Base Case at Pre-test for the Leakage Measurements for the
GLM in Equation F.2
10-MilPads PGA-A PGA-B Gull Wing
LRFlux WSFlux LRFlux WSFlux LRFlux WSFlux LRFlux WSFlux
OSP
Imm Sn
Imm Ag
Ni/Au
Ni/Pd/Au
0.73
0.33
0.48
2.50
2.10
2.25
1.77
1.77
0.35
-0.35
2.40
2.05
2.05
2.05
1.70
2.71
2.71
2.71
2.71
-0.38 2.33
0.30
0.27
1.39
1.36
1.09
1.09
1.09
10-Mil Pads
Examination of the GLM results in Table F.27 for 10-mil pads shows an effect due to flux of
approximately 0.74 orders of magnitude (see column 1 in uppermost portion of Table F.23). There is
also evidence of site-to-site variation and some interaction between site and flux that affects resistance
either positively or negatively by up to an order of magnitude. Sites applying the OSP surface finish
(Sites 6, 7, 8, and 9) as will as Sites 10 and 11 with immersion Sn do not differ from the base case
when LR flux is used.
Table F.28 shows a flux effect of approximately 1.77 orders of magnitude when sites are dropped
from the GLM and replaced by surface finishes. These results show slight increases in resistance over
the base case for OSP, immersion Sn, and immersion Ag.
The differences in the model R2s for both GLMS essentially disappear after exposure to the 85/85
test environment. This result is not unusual and may be due to a cleansing effect from the 85/85 test
environment that removes residues resulting from board fabrication, assembly, and handling. This
same phenomenon was observed for the other three leakage circuits.
Boxplot Displays of Multiple Comparison Results. Boxplot displays of the Pre-test and Post
85/85 test results are given in Figure 4.19 and 4.20. Boxplots for the other test times are displayed in
Figures F.37 and F.38. There are not great changes in the leakage measurements at Post TS and Post
MS as shown in the boxplots.
F-39
-------
APPENDIX F
PGA-A
Examination of the GLM results in Table F.27 for PGA-A shows an effect due to flux of
approximately 1.58 orders of magnitude. There is also evidence of site-to-site variation and some
interaction between site and flux that affects resistance either positively on negatively by up to an order
of magnitude. Nine of the sites do not differ from the base case when LR flux is used.
Table F.28 shows a flux effect of approximately 2.05 orders of magnitude when sites are dropped
from the GLM and replaced by surface finishes, but no meaningful differences due to surface finishes.
As was the case with the 10-mil pads, the differences in the model R2s for both GLMS essentially
disappear after exposure to the 85/85 test environment.
Boxplot Displays of Multiple Comparison Results. A boxplot display of the Pre-test results is
given in Figure 4.21. Boxplots for the other three test times are displayed in Figures F.39 to F.41.
PGA-B
Examination of the GLM results in Table F.27 for PGA-B shows a strong effect due to flux of
approximately 2.77 orders of magnitude. Thirteen of the sites do not differ from the base case when
LR flux is used and the other two only differ slightly. Table F.28 also shows a strong flux effect of
approximately 2.71 orders of magnitude when sites are dropped from the GLM and replaced by
surface finishes, but no meaningful differences due to surface finishes.
As was the case with the 10-mil pads and PGA-A, the differences in the model R2s for both
GLMS essentially disappear after exposure to the 85/85 test environment.
Boxplot Displays of Multiple Comparison Results. A boxplot display of the Pre-test results is
given in Figure 4.22. Boxplots for the other three test times are displayed in Figures F.42 to F.44.
Gull Wing
Examination of the GLM results in Table F.27 for the Gull Wing shows a moderate effect due to
flux of approximately 0.81 orders of magnitude. There is evidence of modest site-to-site variation and
some interaction between site and flux. Eleven of the sites do not differ from the base case when LR
flux is used and the other two only differ slightly. Table F.28 shows a flux effect of approximately
1.09 orders of magnitude when sites are dropped from the GLM and replaced by surface finishes, but
no meaningful differences due to surface finishes.
As was the case with the 10-mil pads, PGA-A, and PGA-B the differences in the model R2s for
both GLMS essentially disappear after exposure to the 85/85 test environment.
Boxplot Displays of Multiple Comparison Results. A boxplot display of the Pre-test results is
given in Figure 4.23. Boxplots for the other three test times are displayed in Figures F.45 to F.47.
F-40
-------
APPENDIX F
F.9 Stranded Wires
Pre-test measurements for the stranded wire circuits were subjected to GLM analyses, as were the
deltas after 85/85, thermal shock, and mechanical shock. The results of the GLM analyses are given in
Tables F.29 and F.30. Columns 3 to 5 in these tables give the results for 85/85, TS, and MS,
respectively. Note that these latter three analyses are based on changes from Pre-test measurements.
The model R2s for Equations F. 1 and F.2 for the stranded wire circuitry are summarized as follows for
each test time.
GLM Circuit Pre-test 85/85 TS MS
Site and Flux
Surface Finish and Flux
St. Wire 1
St. Wire 2
St. Wire 1
St. Wire 2
3.6%
8.6%
1.8%
0.8%
6.5%
8.2%
1.6%
0.9%
12.5%
8.2%
4.5%
7.4%
11.7%
4.1%
2.1%
2.2%
The model R2 values are all near zero at each test time, which indicates that the experimental
parameters do not influence the stranded wire voltage measurements.
Boxplot Displays of Multiple Comparison Results. Boxplots displays of the Pre-test voltage
measurements (mV) for both stranded wires are displayed in Figures F.48 and F.49.
F-41
-------
APPENDIX F
Table F.29 Significant Coefficients for the Two GLM Analyses
GLM from Eq. F.I: Sites and
by Test Time for Stranded Wire
Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
SiteS
Site 4
SiteS
Site 6
Site?
SiteS
Site 9
Site 10
Site 11
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 1 1 * Flux
Site 13 * Flux
Site 16 * Flux
Model R2
Standard Deviation
Pre-Test
12.90
0.55
,21
3.6%
2.57
85/85
(Delta 1)
0.000
-0.001
-0.001
0.002
6.5%
0.002
Thermal Shock
(Delta 2)
0.001
0.024
12.5%
0.014
Mech Shock
(Delta 3)
0.005
0.042
0.079
11.7%
0.041
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Pre-Test
85/85
(Delta 1)
Thermal Shock
(Delta 2)
Mech Shock
(Delta 3)
Constant
12.94
0.000
0.001
0.006
OSP
Immersion Sn
Immersion Ag
-0.001
1.06
0.010
0.019
Ni/Au
Ni/Pd/Au
Flux
Model R2
Standard Deviation
1.8%
2.00
1.6%
0.001
4.5%
0.014
2.1%
0.043
F-42
-------
APPENDIX F
Table F.30 Significant Coefficients for the Two GLM Analyses by Test Time for Stranded Wire 2
GLM from Eq. F.I: Sites and Interactions with Flux
Experimental Factor
Constant
Flux
Site 2
SiteS
Site 4
SiteS
Site 6
Site?
SiteS
Site 9
Site 10
Site 1 1
Site 12
Site 13
Site 14
Site 15
Site 16
Site 4 * Flux
Site 5 * Flux
Site 7 * Flux
Site 1 1 * Flux
Site 13 * Flux
Site 16 * Flux
Model R2
Standard Deviation
Pre-Test
23.44
-1.56
-2.31
8.6%
1.90
85/85
(Delta 1)
-.000
0.003
-0.002
8.2%
0.003
Thermal Shock
(Delta 2)
0.011
0.077
0.074
8.2%
0.067
Mech Shock
(Delta 3)
0.033
0.130
4.1%
0.098
GLM from Eq. F.2: Surface Finishes and Flux
Experimental Factor
Pre-Test
85/85
(Delta 1)
Thermal Shock
(Delta 2)
Mech Shock
(Delta 3)
Constant
23.34
0.000
-0.001
0.021
OSP
Immersion Sn
Immersion Ag
-0.43
-0.001
0.038
Ni/Au
Ni/Pd/Au
Flux
0.026
0.029
Model R2
Standard Deviation
0.8%
2.00
0.9%
0.002
7.4%
0.067
2.2%
0.099
F-43
-------
APPENDIX F
HVLC PTH
HASL
5.03 —
5.02 -
5.01 —
5.00-
Boxplots of HVLC PTH by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
\iiiiiiiiiiiiiiiiiiiiir^
SJt6 FlUX ^— CNCO^LOCOI^COOTOT— CNCO'^'LOCOh* CO O) O •*— CN CO
WS WS WS WS WS WS WSWS WS WS WS
Figure F.I Boxplot Displays for HVLC PTH Measurements (jiA) at Pre-test by Surface Finish
(Acceptance Criterion = 4|jA< X <6|jA)
Post 85/85
HVLC PTH
HASL
i
D.
^
I
CL
Q
Boxplots of DPHVLC P by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg Ni/Au Ni/Au/Pd
5.03-
5.02-
5.01 —
5.00-
4.99 —
SiteFlux
*
'
-
J
i
1
1
-
-
i
1
-
1
1
1
A
'-
\
1
A
'
\
m
-
1
1
i r
r
1
1
-
-
|
L *
A
-
T
| |
A
'
1
-
-------
APPENDIX F
Post Thermal Shock
HVLC PTH
HASL
Boxplots of DTHVLC P by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
5.01 —
-r- 5.00 —
1
Q_
O
5!
I
Q 4.99 —
4.98 —
—
i
i
, T T T
*
I
SiteFlux - CM co ^r
WS WS
1
1
1
X
i
T »
i i
.-,
T
i
f> CD h- co
*
1
1
O) O
PI
•
T
1
1
1
1
«- CN
WS WS WS
J
1
1 * 1
1 •
1
1 1
1
1
1
1
T
PI
1
1 1
1
PI
T
1
co ^ in CD Is* oo o
~\ ~ n n
•
..YD
1 1 1
O T- CN CO
CN CN CN CN
WS WSWS WS WS WS
Figure F.3 Boxplot Displays for HVLC PTH Post TS - Pre-test Measurements
(Acceptance Criterion = 4|jA< X <6|jA)
by Surface Finish
Post Mechanical Shock
HVLC PTH
HASL
Boxplots of DMHVLC P by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
5.01 —
_1_
Q_
O
>
Q 4.99 -
4.98 —
i
*
i
. 1
i
r
i
•
X
T
i
-
T
1
1
i
T
* I1
1
1 *
•
i
1
-
1
\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ r
gj-fg FlUX T— CNCO^lOCDI^OOOOT— CNCO^lOCDI^ CO O) O •*— CN CO
WS WS WS WS WS WS WSWS WS WS WS
Figure F.4 Boxplot Displays for HVLC PTH Post MS - Pre-test Measurements (\\A) by Surface Finish
(Acceptance Criterion = 4|jA< X <6|jA)
F-45
-------
APPENDIX F
Drp.Tpqt
u!fl /- cl/i-r Boxplots of HVLC SMT by SiteFlux
HVLU oM 1
(means are indicated bysolid circles)
5.4 —
5.3 —
5.2 —
| 5-1-
5.0 —
4.9 —
4.8 —
SiteFlux
HASL OSP
Bin.'iif
T | ~ ^ 'VI
I *
*
I I I I I I I I
WS WS WS WS
ImmSn ImmAg Ni/Au Ni/Au/Pd
1
.
l
D
_L
T
T
1
1
J
1
L
; i
i n
•
i
1
i
1
1
1 1
''•ills
T T
T
II 1 1 1 1 1 II
(NCO^LnCQI-^CQOTOT-CNCO
WS WS WSWS WS WS WS
Figure F.5 Boxplot Displays for HVLC SMT Measurements (jiA) at Pre-test by Surface Finish
(Acceptance Criterion = 4|jA< X <6|jA)
uw? r?o(f!rr Boxplots of DPHVLC S by SiteFlux
rIVLU oM 1
(means are indicated bysolid circles)
5.4 —
5.3 —
^ 5.2-
O>
0 5.1-
D.
Q 5.0 —
4.9-
4.8-
SiteFlux
HASL OSP
X X
J
Q i (, i i §
yppBuTp^
*
ImmSn ImmAg Ni/Au Ni/Au/Pd
1
-
i i i i i i i i i
WS WS WS WS
J_^
•
T
1
1
j
1
L
" 1
1
•
i
T
1
i
1
'l
1
1
" ,
6
?
II 1 1 1 1 1
CNCO^LOCOI^COa)
WS WS WSWS WS
1
I •
uB
u y r
1 1
O T- CN CO
CN CN CN CN
WS WS
Figure F.6 Boxplot Displays for HVLC PTH Post 85/85 - Pre-test Measurements
(Acceptance Criterion = 4|jA< X <6|jA)
F-46
by Surface Finish
-------
APPENDIX F
Post Thermal Shock
HVLC SMT
HASL
Boxplots of DTHVLC S by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
5.5-
5.4 —
5.3 —
W 5.2 —
O
i 5.1-
H
0
5.0 —
4.9 —
4.8-
SiteFlux
I
-
1
X
X
JU
n 0 ?
i i i
CN CO ^
WS WS
X
X
• rn
8 f ? D ?
1 1 1 1 1
in CD Is* oo o
ws ws
1
.
1
D
J
1
f±q
n
1
1
^- o
ws
L
1
J
1
_
"
T
1
•o
—
1
J
WS
1
1
a
1
6 1
T
i i
) (D f~
WSW!
^
1 1
00 O)
ws
1
'
1
o
•vj
V
\
J
1
1
N
i/S
h
^9
1 1
CN CO
CN CN
WS
Figure F.7 Boxplot Displays for HVLC PTH Post TS - Pre-test Measurements
(Acceptance Criterion = 4|jA< X <6|jA)
by Surface Finish
Post Mechanical Shock
HVLC SMT
HASL
Boxplots of DMHVLC S by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
0.05-
0.04-
I 0.02 -
Q
0.01 —
0.00 —
SiteFlux
I I I I
^- CN CO ^
WS WS
X X
I I I I I
in (D r~ co en
WS WS
X
I I I I I
O i-
-------
APPENDIX F
P re-Test
HSD PTH
Boxplots of HSD PTH by SiteFlux
(means are indicated bysolid circles)
HASL
18.0 —
I
Q- 17.5 —
Q
I
17.0 —
X
SB^
D
X
T
1
•
-
Bill1]
5 D
T ^
I I I I I I I I I
SiteFlux T-cNnTinoji^cooj
ImmSn ImmAg
I
X
(\ d R i n
8H!
h h A
8?
i
Ni/Au Ni/Au/Pd
JL
•
T
1
•
T
I I I I I I I I I I
O^-CNCO^LOCQI^COO)
C
IS
r
1
1
•
hr
i i i
D T- (N CO
T-T-T-T-T-T-T-T- T-T-CNCNCNCN
ws ws ws ws
ws
ws wsws ws ws
ws
Figure F.9 Boxplot Displays for HSD PTH Measurements (nsec) at Pre-test by Surface Finish
P re-Test
HSD SMT
HASL
Boxplots of HSD SMT by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
Q
OT
I
9.5 —
9.4 —
9.3 —
9.2 —
9.1 —
9.0 —
8.9 —
SiteFlux
J
[
"
1
1
i i
L
i
l
-
1
•
\
\
i
i
i
r
1
1
-
X
J
a.
i T
J
i
1
- J
i
i
,1
1
r
l
i
1
T
i
1
_
1
r
i
1
i
r
1
r
\ \ \ \ I I \ \ \ \ \
^-CNCO^LOCDI^COOTO^-CN
\ \ \ \ \ I \ \ \ \ \
^•incDt^ooajo^-cNco
WS WS WS WS
ws
ws ws ws
ws
ws ws
Figure F.10 Boxplot Displays for HSD SMT Measurements (nsec) at Pre-test by Surface Finish
F-48
-------
APPENDIX F
HFSPTH50MHZ
HASL
Boxplots of
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
0.5 —
0.4 —
0.3 —
O n9
LO u-^ ~~
I
D- 0.1 —
LL
0- 0.0 —
-0.1 —
-0.2 —
-0.3-
aega
X
X X
X
X
• * * i
M
T
*
X
X
x
ft • ¥
X
1
•
-
'
* 6 *
X
• fl
Site Flux
i i i i i i i i i i i i i i i i i i i i i i i
ws ws ws ws
T-T-T-T-T-T-T-T- T-T-CNCNCNCN
WS WS WSWS WS WS WS
Figure F.ll Boxplot Displays for HF PTH 50MHz Post 85/85 - Pre-test Measurements (dB) by Surf. Finish
(Acceptance Criterion = +5dB of Pre-test)
Post Thermal Shock
HF PTH 50MHz
HASL
Boxplots of DTHF PTH by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
0.5 —
3
2
: o.o-
3
-0.5 —
SiteFlux
i $ 4- *
1 1 1 1
^- CN CO ^
WS WS
g-M*
X
1 1 1 1 1
LO CD f~ CO O)
WS WS
.-.-.
1 1 1 1 1
O i- CN CO I1
WS WS
fl'8
1 1 1
LO CD f~
WSWS
1
'
1
CO
^ a *
X
X
1 1 1
O) O i-
ws ws
B «
1 1
CN CO
ws
Figure F.12 Boxplot Displays for HF PTH 50MHz Post TS - Pre-test Measurements (dB) by Surface Finish
(Acceptance Criterion = +5dB of Pre-test)
F-49
-------
APPENDIX F
P re-Test
HF PTH f(-3dB)
CO
±
285-
280-
275-
SiteFlux
HASL
Boxplots of HF PTH-3 by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg Ni/Au Ni/Au/Pd
1
1
T
1
•
T
•
i
*
j_.
•
T
•
i
8
i
•
1] i ij
("'lii'l '
1 T
Y g
n i i i i i i i i i i i i i i i i i i i i i r
^- CN CO
WS WS WS WS
WS
WS
WS WS
WS
WS WS
Figure F.13 Boxplot Displays for HF PTH f(-3dB) Measurements (MHz) at Pre-test by Surface Finish
(Acceptance Criterion = +50Mhz of Pre-test)
Post 85/85
HF PTH f(-3dB)
HASL
Boxplots of DPHF PTH by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
0 —
CO
±
1
Q_
LL
Q ~5 —
-10 —
SiteFlux
f 1 i ^
I I I I
•<- CN CO ^
WS WS
fi u g P p
' X
1 1 1 1 1
in CD r~ co en
WS WS
X
1 1 1 1 1
O i- CN CO I1
WS WS
B a ^
1 1 1
in CD f~
WSWS
.fit
1 1 1 1
co en o t-
WS WS
f 8
i i
CN CO
WS
Figure F.14 Boxplot Displays for HF PTH f(-3dB) Post 85/85 - Pre-test Measurements (MHz) by Surf. Finish
(Acceptance Criterion = +50Mhz of Pre-test)
F-50
-------
APPENDIX F
Post Thermal Shock
HF PTH f(-3dB)
HASL
Boxplots of DTHF PTH by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
5 —
± o-
CL
LL
1
° -5-
-10 —
SiteFlux
T a T- •
i i i i
^- CN CO ^
WS WS
X
8\\ l\ rn
.
I I I I
in CD Is* oo a
WS WS
I
1
5
ifr PI i***! 1
U
1
*
i i i i i
O •<-
1
-
ri
I
|
N
1
•
T
1
•o
i
y i
y
1
i i
^r in u
h " ,
r1! PI P
i D n i B
• n T
•
' 0 0 T -
i i i i i i
3 r^ co en o T- r-j
1
J
1
CO
WS WS WS WS WS WS WSWS WS WS WS
Figure F.16 Boxplot Displays for HF PTH f(-40dB) Measurements (MHz) at Pre-test by Surface Finish
(Acceptance Criterion = +50Mhz of Pre-test)
F-51
-------
APPENDIX F
Post 85/85
HF PTH f(-40dB)
HASL
Boxplots of DPHFPTH- by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
5 —
o °-
±
Q_
LL
a! -5-
Q
-10 —
m m 9 m
*
* g o g y
X |
X
a a g 1 -
*
* D
• 8 H
X
8 - f *
x
t8
SiteFlux
T
CO
T
in
T
CO
T
CO
T
en
WS WS WS WS
\ r
WS
\ i i r
\ i i r
WS
wsws
WS
WS WS
Figure F.17 Boxplot Displays for HF PTH f(-40dB) Post 85/85 - Pre-test Measurements (MHz) by Surf. Fin.
(Acceptance Criterion = +50Mhz of Pre-test)
Post Thermal Shock
HF PTH f(-40dB)
HASL
Boxplots of DTHFPTH- by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
o —
o
t -5-
LL
I
Q
-10 —
-15 —
SiteFlux
t * * t
I I I I
^- CN CO ^
WS WS
• 5 i 9 Q
n
X
I I I I I
in co r~ co en
WS WS
. ^ 6 g 8
*..
I I I I I
O t- CN CO I1
WS WS
1 1 1
in co r~
WSWS
1
\
a
j y i
H w 15
) x
1 1 1
D en o i-
WS WS
T LJ
1 1
CN CO
WS
Figure F.18 Boxplot Displays for HF PTH f(-40dB) Post TS - Pre-test Measurements (MHz) by Surf. Finish
(Acceptance Criterion = +50Mhz of Pre-test)
F-52
-------
APPENDIX F
P re-Test
HF SMT 50MHz
HASL
Boxplots of HF SMT50 by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg Ni/Au
Ni/Au/Pd
-0.6 —
-0.7 —
o
LO
tt>
LL
X -0.8-
-0.9 —
SiteFlux
1
rh i n n
i
I
•
I
U W
n 9
T
1 1
JL
i i i i
CN co ^ in
WS WS
JL
•
T
i
CD
0
M
M
i
1
"
1
h- CO
I
T
1
o>
WS WS
1
1
1^1 n H
LJ U
T I
X
1 1 1 1
O •<- CN CO
1
1
-
T
U 1
U T
i i i
^r in CD
1
T
1
*^
1
-
T
1
D
I
1
1
y
] D
1 1
1
P
U
i i i
00 O) O T- CN CO
^- T- CN CN CN CN
WS WS WSWS WS WS WS
Figure F.19 Boxplot Displays for HF SMT 50MHz Measurements (dB) at Pre-test by Surface Finish
Post 85/85
HF SMT 50MHz
HASL
Boxplots of DPHF SMT by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
0.1 —
0.0 —
° -0.1 —
^
OT
1 1
I -0.2 -
CL
Q
-0.3 —
-0.4 —
SiteFlux
!l*fi
X
1 1 1 1
•<- CN CO -^
X
X
•)*:'
I I I I I
in CD Is* oo o
. . e a ?
X
1 1 1 1 1
O T- CN CO ^~
X
i
x
1
in a
X
1 *
•
r
3 r~
* * e a
i i i i
00 O) O T-
T— T— rvi rvi
y i
X
1 1
CN CO
rvi rvi
WS WS WS WS
WS
WS
wsws
WS
WS WS
Figure F.20 Boxplot Displays for HF SMT 50MHz Post 85/85 - Pre-test Measurements (dB) by Surf. Finish
(Acceptance Criterion = +5 dB of Pre-test)
F-53
-------
APPENDIX F
Post Thermal Shock
HF SMT 50MHz
HASL
Boxplots of DTHF SMT by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
0.1 -
0.0 —
-0.1 -
o
| -0.2 -
U= -0.3 -
I—
Q -0.4 -
-0.5 —
-0.6 —
SiteFlux
f 8 *
i i i
•<- CN CO
JL
,
_
X
1
1
* - - . '
X
1 1 1 1 1
in CD Is* oo o
.
-
i
D
J
1
] 8 *&
i
X
1 1 1
CN CO ^
fe-
i
i i i
in co r~
i
1
0
3 t ? i
'
i i i
D O) O T-
— T— rvi rvi
\%
\ Y
i i
CN CO
rvi rvi
WS WS WS WS
WS
WS
WSWS
WS
WS WS
Figure F.21 Boxplot Displays for HF SMT 50MHz Post TS - Pre-test Measurements (dB) by Surface Finish
(Acceptance Criterion = +5 dB of Pre-test)
P re-Test
HF SMT f(-3dB)
HASL
Boxplots of HF SMT-3 by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
340-
330 —
CO
LL
I
320 —
310 —
SiteFlux
MB?
X
I I I I
^- CN CO ^1
s»t^
X
I I I I I
in co f~ co en
Bfi + ??
T
i i i i i
O t- CN CO I1
^ * i
g w
X
X
I I I
in co r~
1
1
TT
1
00
$ i *
X
1 1 1
en o i-
4 %
i i
CN CO
WS WS WS WS
WS
WS WSWS WS WS WS
Figure F.22 Boxplot Displays for HF SMT f(-3dB) Measurements (MHz) at Pre-test by Surface Finish
(Acceptance Criterion = +50Mhz of Pre-test)
F-54
-------
APPENDIX F
Post 85/85
HF SMT f(-3dB)
HASL
Boxplots of DPHF SMT by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
10 —
5 —
CO
2
LL
I
D- n
Q u
-5 —
SiteFlux
X
H D T 1
1 1 1 1
•<-
-------
APPENDIX F
P re-Test
HF SMT f(-40dB)
HASL
Boxplots of HFSMT-40 by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
950-
900 —
o
tt>
LL
I
850-
800-
SiteFlux
1
i
1
1
1
1
~
T
1
i
t- CN
I
i i
fl k
H
i r
• •
hr
i i i i
1
•
X
A
h
n P
r
T
1
co ^ in CD Is*
WS WS WS
X
0 D
T
1 1 1 1
*
1
•
T
1
00 O) O T- CN
WS WS
1 A
n n
P P
T
1
1
h
i i
fl
i
1
•
1
1
co ^ in co r~
WS
1
m
-
1
.
T
^
1 1
U
i i i
00 O) O T- CN CO
^- T- CN CN CN CN
wsws ws ws ws
Figure F.25 Boxplot Displays for HF SMT f(-40dB) Measurements (MHz) at Pre-test by Surface Finish
(Acceptance Criterion = +50Mhz of Pre-test)
Post 85/85
HF SMT f(-40dB)
HASL
Boxplots of DPHFSMT- by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
50 —
40 —
o 30-
jl
1 20-
1 1
I
D.
Q 10 —
0 —
-10 —
SiteFlux
I
A I
Wi
I I I I
^- CN CO ^
WS WS
*
X
i
jfl'98
T
I I I I I
in co f~ co O)
WS WS
*
-_'
I I
O i-
WS
X
1
T
|
N
46
i i
CO I1
WS
pg!
i i i
in co r~
WSWS
X
Sr
. lj
'
i i i
CO O) O T
WS W
1
1
S
IB
1
i i
CN CO
WS
Figure F.26 Boxplot Displays for HF SMT f(-40dB) Post 85/85 - Pre-test Measurements (MHz) by Surf. Fin.
(Acceptance Criterion = +50Mhz of Pre-test)
F-56
-------
APPENDIX F
Post Thermal Shock
HF SMT f(-40dB)
HASL
Boxplots of DTHFSMT- by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
40 —
30 —
20 —
10 —
o
T o —
1 -10-
LL
jE -20 —
Q
-30 —
-40 —
-50 —
-60 —
SiteFlux
J
I
T-
1
J
\
N
X
9
i
CO
T
1
T
• fl J j j
1 1 1 1 1
in CD Is* oo o
§
i
o
•
-
1
eu
1
i i i
CN CO -^
X
1
n u
J
I
r
i
3 f~
I
T
\
CO
n
i
O)
1
T
o
•V
j
1
rvi
X
D |
T
I I
CN CO
rvi rvi
ws ws ws ws
ws
ws
wsws
ws
ws ws
Figure F.27 Boxplot Displays for HF SMT f(-40dB) Post TS - Pre-test Measurements (MHz) by Surf. Finish
(Acceptance Criterion = +50Mhz of Pre-test)
P re-Test
HF TLC 50MHz
HASL
Boxplots of HF TL 50 by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
-42-
-43-
-44 —
-45-
° -46 —
-48 —
-49 —
-50 —
-51 -
SiteFlux
J
T-
I
r
i i
CN CO
1
1
1
T
nl
1
i i i
in CD r~ a
J
I
U
j
3 a
L
i
>
JL r
- '
)
T
i i i
O i- CN P
h
i
•>
i
T
1
CD r~
j
1
0
L -
^ fl
II
0
1
D cn c
U
X
1
D i-
1
1
T
i i
CN CO
WS WS WS WS
ws
ws
wsws
ws ws
ws
Figure F.28 Boxplot Displays for HF TLC 50MHz Measurements (dB) at Pre-test by Surface Finish
F-57
-------
APPENDIX F
o
LO
Q_
Q
HASL
Boxplots of
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
5 —
4 —
3 —
2 —
0-
-1 —
-2 —
-3 —
-4 —
X
j 0 £?
* X
X
X
H P £ JL
U * " P
X
X
X
n * fi
a 5 a g a
_j
_
i
i
1
L,
|
-
'
T
X
V
61
IX
X
1
§
X
\ i i i i i i i i i i i i i i i i i i i i i r
WS WS WS WS WS WS WSWS WS WS WS
Figure F.29 Boxplot Displays for HF TLC 50MHz Post 85/85 - Pre-test Measurements (dB) by Surf. Finish
(Acceptance Criterion = +5 dB of Pre-test)
Post Thermal Shock
HF TLC 50MHz
HASL
Boxplots of DTHF TL by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
5 —
4 —
3 —
2 —
o
LO 1 —
I- -1 —
Q
-2 —
-3 —
-4 —
-5 —
SiteFlux
X
X
I
JL
?^B
I I I I
^- CN CO ^1
*
JL
•f *8*
X
I I I I I
in eo f~ co O)
• ? 9 • f
x
i i i i i
O i-
-------
APPENDIX F
P re-Test
HF TLC 500MHz
HASL
Boxplots of HF TL500 by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
-15 —
-16 —
o -17-
o
LO
I —
I 1
-19 —
-20 —
SiteFlux
I
I
-
\
1
•
T
1
N
M
1
I
1 1
CO I1
S WS
Bl
1
.
1
1 1
LO CO
WS
1
I
T
1
*^
A
n
Y
1
CO
WS
1
_
T
1
1
3)
_
T
1
D
JL
0 i
?
1
•<- o
WS
1
J p
5
V
1
.
X
1
i
i/S
1
1
n i
T fi B
T5T
1 1 1
LO CO f~
WSWS
1
CO
J
i
c
W
L
i
*
•> c
c
S
1
f
3
g
V
JL
I
1
CN
i/S
1 n
pL,
0 H
0 U
*
i i
CN CO
CN CN
WS
Figure F.31 Boxplot Displays for HF TLC 500MHz Measurements (dB) at Pre-test by Surface Finish
Post 85/85
HF TLC 500MHz
HASL
Boxplots of DPHF TL5 by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
3 —
2 —
1 —
o
o
LO
_i
H 0-
LL
I
D.
Q -1-
-2 —
-3-
SiteFlux
X
8x
i * 9
n x
i
i i i i
^- CN CO ^1
T
I
LO
1
;
|
|
D
ft i g
X
X
1 1 1
1^ CO O)
X
X
1
PI
* 3 5 0 $
X
1 1 1 1 1
O T- CN CO ^~
X
* 8 *
X
1 1 1
LO CO t-^
-
|
T
1
CO
X
X
X
O
S B a
X
X
1 1 1
O) O i-
X
f
5 9
1 1
CN CO
r\i r\i
WS WS WS WS
ws
WS WSWS WS WS WS
Figure F.32 Boxplot Displays for HF TLC 500MHz Post 85/85 - Pre-test Measurements (dB) by Surf. Finish
(Acceptance Criterion = +5 dB of Pre-test)
F-59
-------
APPENDIX F
Post Thermal Shock
HF TLC 500MHz
HASL
o
o
LO
I
Q
1 —
0 —
-2 —
-3 —
Boxplots of DTHF TL5 by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
SiteFlux
1
1
M n S 0
X
1 1 1 1
•<- CN CO ^
\NS \NS
-JL
•
T
i
in
X
? $ 4 |
X
1 1 1 1
CD 1^ 00 O)
WS WS
*
a + 5
1 1 1
O T- CN
WS
1
T
\
•o
X
5
1
1-
WS
it
? B B
X
1 1 1
in CD h-
WSWS
*
*
. ' - '
"
X
1 1 1 1
co en o i-
•?-•?- CN CN
WS WS
X
B B
1 1
CN CO
CN CN
WS
Figure F.33 Boxplot Displays for HF TLC 500MHz Post TS - Pre-test Measurements (dB) by Surface Finish
(Acceptance Criterion = +5 dB of Pre-test)
P re-Test
HF TLC RNR
Boxplots of HFTLRNul by SiteFlux
(means are indicated bysolid circles)
-34 —
-5
CC
1 —
t£ -39 —
-44 —
SiteFlux
1
i
]
1
;—
HASL
fl**
1 1 1
CN CO ^
WS WS
OSP
X
1 1 1 1 1
in CD f~ co en
WS WS
ImmSn
* D •
X
X
X
1 1 1 1 1
O i- CN CO I1
ws ws
ImmAg
9 e ^
X
1 1 1
in CD r~
WSWS
Ni/Au
M R n n
X
X
1 1 1 1
CO O) O i-
WS WS
Mi/Au/Pd
'5
T
i i
CN CO
WS
Figure F.34 Boxplot Displays for HF TLC RNR Measurements (dB) at Pre-test by Surface Finish
F-60
-------
APPENDIX F
Post 85/85
HF TLC RNR
HASL
Boxplots of DPHFTLRN by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
10 —
5 —
z
_l
LL
I 0 —
CL
Q
-5 —
I
n *
a - w :
#
X
X
as.-,
X
X
X
X
• g - • ~
X
X
X
• • T
X *
S • 7
X
I
J
SiteFlux
n i i i i i i i i i i i i i i i i i i i i i r
^-cNco^Lncoi^coOTO^-cNco^LOcoh- OOOO^-CNCO
ws ws ws ws
ws
ws wsws ws ws ws
Figure F.35 Boxplot Displays for HF TLC RNR at Post 85/85 - Pre-test Measurements (dB) by Surf. Finish
(Acceptance Criterion = <10 dB increase over Pre-test)
Post Thermal Shock
HF TLC RNR
HASL
Boxplots of DTHFTLRN by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
10 —
^
z
5 o-
LL
I—
Q
-10 —
X X
X
X * X
X
• * • a ?
X
X
X
X
X
• •
_ •
• 1
1
* i
(
1
I
'
«
5s"
X
X
a B
*
X
SiteFlux
\ \ \ \ I I \ \ \ \ \ \ \ \ \ \ \ I \ \ \ \ \
^-cNco^mcDi^ooajo^-cNco^incDi^ OOOO^-CNCO
WS WS WS WS
ws
ws wsws ws ws ws
Figure F.36 Boxplot Displays for HF TLC RNR Post TS - Pre-test Measurements (dB) by Surface Finish
(Acceptance Criterion = <10 dB increase over Pre-test)
F-61
-------
APPENDIX F
Post Thermal Shock
10-MilPads
HASL
Boxplots of DTPads by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
15 —
14 —
13 —
T3
(0
D.
|—
Q 12 —
11 —
10 —
SiteFlux
Sr
I
^- o
V
\ 0 =
U ,
I |
J CO ^
/s ws
m
V
-
T
i
D
V
u •
Yt
i
Is* a
s v\
5
IS
-
.
I
J>
h
y
i
o
V
-
-
I
V
3
-
-
I
N
-
i
T
i
•o
V
-
-
«•
JS
Sn u
r
u u
i i i
LO CD Is*
WSWS
0
D C
w
•)
-
s
-
•
X
o
•vj
*
I
CN
ws
U V
I
CN CO
CN CN
WS
Figure F.37 Boxplot Displays for 10-Mil Pad Post TS - Pre-test Measurements (logic ohms) by Surf. Finish
(Acceptance Criterion = Resistance > 7.7 logic ohms)
Post Mechanical Shock
10-Mil Pads
HASL
Boxplots of DMPads by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
15 —
14 —
T3
(0
D.
1 13-
12 —
SiteFlux
*
i
1
1
- * 1
1 " "
r ,
-
-
1
1 III
T-CNCO^LOCQI-^CQ
l
1
r
*
O)
•
T
D
WS WS WS WS
-
-
1
•
* 1
1
1 • •
|Yf J,
II 1 1 1 1
WS WS WSWS WS WS WS
Figure F.38 Boxplot Displays for 10-Mil Pad Post MS - Pre-test Measurements (logic ohms) by Surf. Finish
(Acceptance Criterion = Resistance > 7.7 logic ohms)
F-62
-------
APPENDIX F
Post 85/85
PGA-A
Boxplots of DPPGA A by SiteFlux
(means are indicated bysolid circles)
HASL OSP
13 —
12 —
^j.
CL
CL n _
Q
10 —
u
T
n r^ n
0 ^
m
_l
X
*Q
-
.
i
X
X
ImmSn ImmAg Ni/Au Ni/Au/Pd
•
LP
I
i $ a - n i .i
H l
i i i
x
L
1
1
1
n
i
ee
•
X
L.
i
T
SiteFlUX
\ r
\ i i i i i i i i i i i i i r
WS WS WS WS WS WS WSWS WS WS WS
Figure F.39 Boxplot Displays for PGA-A Post 85/85 - Pre-test Measurements (logic ohms) by Surface Finish
(Acceptance Criterion = Resistance > 7.7 logic ohms)
Post Thermal Shock
PGA-A
HASL
Boxplots of DTPGA A by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
15 —
14 —
< 13 —
^"
D.
Q 12 —
11 —
10 —
SiteFlux
• ' o *
X
I I I
7.7 logic ohms)
F-63
-------
APPENDIX F
Post Mechanical Shock
PGA-A
HASL
Boxplots of DMPGA A by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg Ni/Au
Ni/Au/Pd
14— n
13 —
10 —
• *
p
X
n
,
• B : B
1 "
X
-
I
n
\ v
i
m
1
1 1 1 1 1 1 1 1 1 1 1 1 1 1
a • ^ *
n *
U
i i i i i
00 O) O T- CN CO
^- T- CN CN CN CN
WS WS WS WS WS WS WSWS WS WS WS
Figure F.41 Boxplot Displays for PGA-A Post MS - Pre-test Measurements (logic ohms) by Surface Finish
(Acceptance Criterion = Resistance > 7.7 logic ohms)
Post 85/85
PGA-B
Boxplots of DPPGA B by SiteFlux
(means are indicated bysolid circles)
HASL
13 —
12 —
CO
G5
CL
CL
o „_
10-
SiteFlux
8*
*
X
I I
WS
-
T
X
OSP
n
i
"l J
1
1
jj
X
1
ImmSn Imm
1
1 f] t B i!
' T
r ' x
|
i
T
o^^cor-cioo^^^j^co
WS WS WS
WS WS
Ag Ni/Au Ni/Au/Pd
98
1
•
T
X
e@
IB
1
II II 1
Is* 00 O) O T— CN CO
WSWS WS WS WS
Figure F.42 Boxplot Displays for PGA-A Post 85/85 - Pre-test Measurements (logic ohms) by Surface Finish
(Acceptance Criterion = Resistance > 7.7 logic ohms)
F-64
-------
APPENDIX F
HASL
Boxplots of DTPGA B by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
15 —
14 —
CQ
G5 13-
CL
I
Q
12 —
11 —
SiteFlux
X i
y
•
X
*
I I I I
T- 7.7 logic ohms)
Post Mechanical Shock
PGA-B
HASL
Boxplots of DMPGA B by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
15 —
14 —
CQ
r? 13~
S
Q
12 —
11 —
SiteFlux
X
x E
i
T- o
U'
X
I I
a co ^r
B "
X
I I
in CD h
'
X
I
CO
JL
T
X
I
J>
^ • n y
X
I I I I I
o T- CN co ^r
X
Q
I I I
in CD i^
,
1
CO
.
T
1
1
0)
if
1
1 1
O i-
rvi rvi
r
I
1
1 1
CN CO
rvi rvi
ws wsws ws ws ws
Figure F.44 Boxplot Displays for PGA-A Post MS - Pre-test Measurements (logic ohms) by Surface Finish
(Acceptance Criterion = Resistance > 7.7 logic ohms)
F-65
-------
APPENDIX F
Wing
14 —
13 —
12 —
CD
D.
Q
HASL
Boxplots of DPGulIWi by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
Site Flux
\ i i i i i i i i i i i i i i i i i i i i i r
ws ws ws ws
WS
ws wsws ws ws ws
Figure F.45 Boxplot Displays for the Gull Wing Post 85/85 - Pre-test Measuremts. (logic ohms) by Surf. Fin.
(Acceptance Criterion = Resistance > 7.7 logic ohms)
Post Thermal Shock
Gull Wing
HASL
15—1 X
14 —
13 —
12 —
11 —
10 —
7
Boxplots of DTGulIWi by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
D
Ni/Au Ni/Au/Pd
f
SiteFlux
\ I I I I I I I I I T
T-CNCO^LOCQI-^COOTOT-
WS WS WS WS
WS
\ I I I I I I I I I I
WS WSWS WS WS WS
Figure F.46 Boxplot Displays for the Gull Wing Post TS - Pre-test Measurements (logic ohms) by Surf. Fin.
(Acceptance Criterion = Resistance > 7.7 logic ohms)
F-66
-------
APPENDIX F
Post Mechanical Shock DMGu||Wi
Gull Wmq
(means are indicated bysolid circles)
HASL OSP
14 —
13 —
D)
C
§
=5 11-
(D
Q 10 —
9-
8-
0
r
v
I
Hh
B
1
D r
* 1
Bn n
u w
T
1
X
*
1
JL
i
ImmSn ImmAg
-
T
T 1
' a
i
X
Hh
1
|
n
V
i
T
i_
i
i
-
i
T
|
Ni/Au Ni/Au/Pd
Bn n yi
u
u
i
#
-
I
B
v
i
Site Flux
n i i i i i i i i i i i i i i i i i i i i i r
ws ws ws ws
ws
ws wsws ws ws ws
Figure F.47 Boxplot Displays for the Gull Wing Post MS - Pre-test Measurements (logic ohms) by Surf. Fin.
(Acceptance Criterion = Resistance > 7.7 logic ohms)
P re-Test
Stranded Wire 1
HASL
20 —
15 —
CO
10 —
Boxplots of StWire 1 by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
1
i
T
1
T
-
T
B
V
JL
I
-
X
1 1
^ 1 T
1
T
1
1
I
1
T
1
T
j
i
1
1
1
~
JL
T
1
I
i
1
i
r
m
T
SiteFlux
\ I I I T
\ i i i r
ws ws ws ws
ws
\ i i i i i i i i i r
ws wsws ws ws ws
Figure F.48 Boxplot Displays for the Stranded Wire 1 Measurements (volts) at Pre-test by Surface Finish
F-67
-------
APPENDIX F
P re-Test
Stranded Wire 2
HASL
Boxplots of StWire2 by SiteFlux
(means are indicated bysolid circles)
OSP ImmSn ImmAg
Ni/Au Ni/Au/Pd
28 —
27 —
26-
25-
OJ 24 —
SL
§ 23-
OT
22-
21 —
20-
19 —
SiteFlux
I
i
T
JL
B
I
1
•vj
C*
1 J
|
•) ^
L
1
r
JL
i
T
i
n
!
1
C£
J
1
r
> i-
i
i
i
a
• I
1
3 a
i
J
\
t
JL
i
T
i
D
1
1
-
1
_
T
1
N
1
f
1
r
i
1
•
T
1
1
L
1
T
1
o
•V
1
r
ID
n
T
i i
CN CO
«.i rvi rvi
ws ws ws ws
ws
ws
wsws
ws
ws ws
Figure F.49 Boxplot Displays for the Stranded Wire 2 Measurements (volts) at Pre-test by Surface Finish
F-68
-------
APPENDIX F
F.10 Design and CCAMTF Baseline Testing of the Test PWA
F.10.1 Test PWA
As mentioned in Chapter 4, the primary test vehicle used in both the DfE project and in the
CCAMTF evaluation of low-residue technology was an electrically functional PWA. This assembly
was designed at Sandia National Laboratories in Albuquerque based on input from LRSTF members
and from military and industry participants during open review meetings held by the task force. The
PWA measures 6.05" x 5.8" x 0.062" and is divided into six sections, each containing one of the
following types of electronic circuits:
• Fligh current low voltage (HCLV) • Fligh frequency (HF)
• High voltage low current (FTVLC) • Other networks (ON)
• High speed digital (HSD) • Stranded wire (SW)
The layout of the functional assembly is shown in Figure F.50. The components in the HCLV,
HVLC, HSD, and HF circuits represent two principal types of soldering technology:
• Plated through hole (PTH)—leaded components are soldered through vias in the circuit board
by means of a wave soldering operation
• Surface mount technology (SMT)—leadless components are soldered to pads on the circuit
board by passing the circuit board through a reflow oven.
The other networks (ON) are used for current leakage measurements: 10-mil pads, a socket for a
PGA, and a gull wing. The two stranded wires (SW) are hand soldered.
The subsections for PTH and SMT components form separate electrical circuits. The PWA
includes a large common ground plane, components with heat sinks, and mounted hardware.
Each subsection shown in Figure F.50 contains both functional and nonfunctional components
(added to increase component density). A 29-pin PTH edge connector is used for circuit testing. High
frequency connectors are used to ensure proper impedance matching and test signal fidelity as
required. Board fabrication drawings, schematics, and a complete listing of all components are
available by contacting the authors of this report. A discussion of each of the sections of the test PWA
is now given. This discussion is supplemented with baseline test results for each of the 23 electrical
responses listed in Table 4.1.
F.10.2 High Current Low Voltage
The HCLV section of the board is in the upper left-hand corner of PWA (see Figure F.50). The
upper left-hand portion of this quadrant contains PTH components with SMT components immediately
beneath.
Purpose of the HCLV Experiment
Performance of high-current circuits is affected by series resistance. Resistance of a conductor
(including solder joints) is determined by the following equation:
F-69
-------
APPENDIX F
pL
(F.7)
where p = resistivity, the proportionality constant
L = length of the conductor
AC = cross-sectional area of the conductor (solder joints)
Resistance is most likely to change due to cracking or corrosion of the solder joint that may be
related to the soldering process. These conditions decrease the cross-sectional area of the solder joints,
thus increasing resistance as shown in Equation F.7. Use of high current to test solder joint resistance
makes detection of a change in resistance easier. A 5 Amperes (A) current was selected as a value that
would cover most military applications. A change of resistance is most conveniently determined by
measuring the steady state performance of the circuit, which will now be discussed.
— 6.05"
—I
High Current Low Voltage (HCLV)
PTH
High Speed Digital (HSD)
PTH
High Current Low Voltage (HCLV)
SMT
High Voltage Low Current (HVLC)
PTH
High Voltage Low Current (HVLC)
SMT
High Speed
Digital (HSD)
SMT
TL2
TL4
High Frequency (HF) Transmission Line
T
Figure F.50 Layout of the PWA Illustrating the Four Major Sections and Subsections
F-70
-------
APPENDIX F
Steady State Circuit Performance
Overall circuit resistance, Rtotai, is the parallel combination of the seven resistors, RI, R2, ..., Ry,
(all resistors = 10Q) used in the HCLV circuit:
1 1 7
(R8)
10Q
****= —
Since a current (I) of 5 A will be applied to the circuit, the resulting voltage (V), according to
Ohm's Law, is
10Q
V = IR = 5Ax - = 7. 14V (F.1
Changes in resistance are thus detected by changes in voltage. However, a pulse width had to be
chosen that would not overstress the circuit components. With current equally divided among the
seven parallel resistors, the power (P) dissipated in each resistor, according to Joule's Law, is:
, (5AV
p = I2R = \ —
Since the power rating for the PTH wire-wound resistor is 3W, the rating is exceeded by a factor
of 1 .7 for steady state (5.1/3). Design curves from the resistor manufacturer indicate the PTH wire-
wound resistors could tolerate the excess power for about 100ms. The SMT resistors are rated at 1W,
so the steady state rating is exceeded by a factor of five. With the manufacturer unable to provide the
pulse current capability of the SMT resistors, a pulse derating factor could not be determined. A pulse
width of 100)05 was selected, which is three orders of magnitude less than the capability of the wire-
wound resistors. This width is also sufficiently long for the circuit to achieve steady state before the
measurement is taken.
Circuit Board Design
Traces carrying the 5 A current were placed on an inner layer of the circuit board because: (1) the
primary concern was the possible degradation of the solder connections as discussed above and (2) the
bulk electrical characteristics (resistivity) of the traces should not be affected by flux residues. High-
current trace widths were designed to be 250 mils whenever possible (following MIL-STD-275). This
width with a 5 A current should cause no more than a 30°C temperature rise under steady-state
conditions.
The resistor and capacitor values were selected to be readily available. If other values are used,
care should be taken to not over-stress the parts, as discussed above.
Baseline Testing Results for HCLV
A gauge repeatability and reproducibility (GR&R) study (Iman et al, 1998) was conducted for the
CCAMTF ATS as part of the CCAMTF program. The LRSTF PWA was utilized in this study. In
particular, 120 LRSTF PWAs were tested for each of the following four surface finishes: OSP,
F-71
-------
APPENDIX F
immersion Ag, immersion Au/Pd and HASL with solder mask. Half the PWAs in each surface finish
group were processed with low-residue (LR) flux and the other half with water soluble (WS) flux.
Data modeling showed that surface finish and flux type did not significantly affect the voltage
measurements for HCLV PTH and HCLV SMT. Figures F.51 and F.52 provide dotplot displays of 4
x 120 = 480 voltage measurements for HCLV PTH and 480 voltage measurements for HCLV SMT,
respectively. The summary statistics HCLV PTH and HCLV SMT voltages are given in Table F.31.
6.60
6.72
+ -
6.84
+ + + Volts
6.96 7.08 7.20
Figure F.51. Dotplot for 480 HCLV PTH Voltage Measurements
(each dot represents up to 10 points)
6.90
.
7.00
.
7.10
7.2C
) 7.30
7.40
Volts
Figure F.52. Dotplot for 480 HCLV SMT Voltage Measurements
(each dot represents up to 16 points)
Table F.31. Summary Statistics for HCLV Circuitry Test Measurements
Circuitry Mean Median St. Dev. Min Max
HCLV PTH
HCLV SMT
6.88V
7.20V
6.96
7.20
0.163
0.106
6.60
6.88
7.20
7.44
F.10.3 High Voltage Low Current
The HVLC circuitry is immediately below the HCLV circuitry and above the high frequency
transmission lines in Figure F.50. The PTH circuitry is in the upper part of this subsection and the
SMT circuitry is in the lower part.
Purpose of the HVLC Experiment
Flux residues could decrease the insulation resistance between conductors. The impact of this
decrease could be significant in circuits with a high voltage gradient across the insulating region.
Decreased resistance can be detected by an increase in current when a high voltage is applied to the
circuit. A voltage of 250V was selected as the high potential for this test. The change in leakage
current is determined by measuring the steady-state performance of the circuit, which will now be
discussed.
F-72
-------
APPENDIX F
Steady State Circuit Performance
Steady-state operation of the HVLC circuit can be determined by considering only the resistors. The
total resistance of the series combination is the sum of the resistances.
Rtotai = R,+R2+R3+R4=R5 = 50MQ (F. 1 2)
since all resistors are 10MQ each. From Ohm's law, the current flowing into the circuit with 250V
applied is
V 250V
Care was taken to not overstress the individual components in the circuits. The voltage stress across
each resistor-capacitor pair is one-fifth of the applied 250V, or 50V. The voltage ratings are 250V for
the PTH resistors, 200V for the SMT resistors, and 250V for all the capacitors. Power rating is not a
concern due to the low current.
Circuit Board Design
High voltage traces were placed next to ground potential traces by design. The spacings between
the high voltage and intermediate traces were selected using MIL-STD-275.
Voltage
0-100
101-300
301-500
Spacing Between Traces (mils)
5
15
30
These guidelines were followed except the 5-mil spacing, where 10 mils was used to facilitate board
fabrication. Table F.32 lists the voltage on various board circuit traces and the spacing to the adjacent
ground trace.
Resistors and capacitors were selected to have readily available values—different values could have
been used to achieve particular experimental goals. For instance, higher resistance values could be
used with lower value capacitors. Reverse biased, low-leakage diodes could also be used for higher
sensitivity to parasitic leakage resistance.
Baseline Testing Results for HVLC
Data modeling showed that surface finish and flux type had very little effect on the voltage
measurements for HVLC PTH and HVLC SMT. Figures F.53 and F.54 provide dotplot displays of
480 voltage measurements for HVLC PTH and HVLC SMT, respectively. The summary statistics for
HVLC PTH and HVLC SMT voltages are given in Table F.33. Note that two sight outliers for HVLC
PTH are identified in Table F.33, but are not included in Figure F.53.
F-73
-------
APPENDIX F
Table
F.32 HVLC
Technology Trace Connected to:
Resistor
PTH R15
R16
R17
R18
R19
SMT R20
R21
R22
R23
R24
Capacitor
C21
C22
C23
C24
C25
C26
C27
C28
C29
C30
Circuit Board Trace
Potential (V)
250
200
200
150
150
100
100
50
50
250
200
200
150
150
100
100
50
50
Potentials
Trace Length at
Potential (in)
0.8
0.4
0.4
NA
NA
0.4
0.4
NA
NA
5.0
1.0
1.0
NA
NA
0.9
0.9
NA
NA
Spacing
(mils)
30
15
15
10
10
30
15
15
10
10
NA = not applicable since no 50V or 150V traces were adjacent to ground potential
Circuitry
Table F.33 Summary Statistics for HVLC Circuitry Test Measurements (sans outliers)
Mean Median St. Dev. Min Max Outliers
HVLC PTH
HVLC SMT
5.04|jA
4.95|jA
5.04
4.95
0.024
0.011
4.972
4.914
5.148
4.976
5.203 5.232
- + -
4.970
+ -
5.005
+ -
5.040
+ -
5.075
+ -
5.110
+ -
5.145
-uA
Figure F.53 Dotplot of 478 Voltage Measurements for HVLC PTH
(each dot represents up to 2 points)
F-74
-------
APPENDIX F
+ -
4.920
+ -
4.932
+ -
4.944
+ -
4.956
+ -
4.968
uA
4.980
Figure F.54 Dotplot of 480 Voltage Measurements for HVLC SMT
(each dot represents up to 2 points)
F.10.4 High Speed Digital
The HSD circuitry is in the upper righthand corner of the LRSTF PWA shown in Figure F.50.
This subsection contains the PTH circuitry and consists of two 14-pin Dual In-line Package (DIP)
integrated circuits (ICs). The SMT subsection 1C is a single 20-pin leadless chip carrier (LCC)
package. Each of these ICs is a "Fast" bi-polar digital "QUAD-DUAL-INPUT-NAND-GATE." Both
subsections contain two ceramic capacitors that bypass spurious noise on the power input line (VCC)
to the ICs and an output high-frequency connector. Inputs to both subsections are applied through the
edge-connector on the right side of the board. Figure F.55 shows a simplified schematic of the ICs.
5V
2.5V
Pulse
1
VCC
Quad-Dual-lnput-NAND-Gate 1C
V
out
Ground
Figure F.55 Simplified Schematic of the ICs in the HSD Subsection
F-75
-------
APPENDIX F
Purpose of the HSD Experiment
The output signal of each gate in Figure F.55 is opposite in polarity to the input signal. If the
traces of these two signals are in close proximity on the printed circuit board (capacitively coupled),
the gate switching speed might be affected by the presence of flux residues. A 5VDC bias is applied to
the VCC inputs during environmental testing to accelerate aging. One PTH 1C (U02) is hand soldered
during assembly to introduce hand solder flux residue in the experiment.
Circuit Description
The schematic in Figure F.55 represents the ICs in the PTH and SMT subsections. The ICs are
random logic circuits that are NAND (Not AND) gates. An AND gate's output is high only when all
inputs are high. The logic of a NAND gate is opposite the logic of an AND gate. Therefore, the
output of a NAND gate is low only when all inputs are high, otherwise the output is high. With the
two connected inputs, the output of each gate is opposite the input. Since the four gates are connected
in series, the output of the last gate is the same logic level (high or low) as the input, with a slight lag.
The output pulse does not change logic levels instantaneously, but the switching times from low to
high (rise time) and from high to low (fall time) should be less than Ins. ICs should perform within
these criteria if the VCC input is 5±0.5V DC, the output load does not exceed specifications, and the
circuit has a proper ground plane as shown in Figure F.55. The HSD circuits also provide an
intermediate test for high frequencies, with switching time dictating a high frequency spectrum. The
frequency spectrum of switching circuits can be expressed in terms of bandwidth (BW). For a
switching circuit, the respective BWs (in Hertz) for rise (tr) and fall (tf) times are:
0.35 0.35
BWr= Hz and BWf = Hz (F.14)
tr tf
Bipolar technology was used rather than a complementary metal oxide semiconductor (CMOS)
since it is not as vulnerable to electrostatic discharge (BSD) damage. Available military bipolar
technologies have the following typical switching speeds and bandwidths:
Technology
5404 TTL
54LS04 Low
Power Schottky
54S04 Schottky
54F04 Advanced
Schottky (Fast)
Typical tr or f(ns)
12
9
3
2.5
Bandwidth (MHz)
29
39
117
140
The Fast technology was selected since it had the shortest switching time and largest bandwidth,
which provides the widest frequency spectrum for this test.
Circuit Board Design
Ground planes were provided for proper circuit operation of the ICs. The PTH subcircuit utilized
the large common ground plane on layer 3 since most of the input and output traces are on layer 4.
Since the SMT circuit traces are on the top layer, a smaller ground plane was added on layer 2. The
"QUAD-DUAL-INPUT-NAND-GATE" was selected since other solder studies of national attention
have used that particular type of 1C, which makes direct comparisons with these studies possible.
F-76
-------
APPENDIX F
Baseline Testing Results for HSD
Data modeling showed that surface finish and flux type had very little effect on the total
propagation delay measurements (msec) for HSD PTH and HSD SMT. Figures F.56 and F.57 provide
dotplot displays of 480 voltage measurements for HSD PTH and HSD SMT, respectively. The
summary statistics HSD PTH and HSD SMT total propagation delay are given in Table F.34 (Note one
slight outlier for HSD PTH).
+ -
12.64
+ -
12.80
+ -
12.96
+ -
13.12
+ -
13.28
+- u sec
13.44
Figure F.56 Dotplot of 480 Measurements of Total Propagation Delay for HSD PTH
(each dot represents up to 2 points)
• + -
4.80
+ -
4.92
+ -
5.04
+ -
5.16
+ -
5.28
+-u sec
5.40
Figure F.57 Dotplot of 480 Measurements of Total Propagation Delay for HSD SMT
(each dot represents up to 2 points)
Table F.34 Summary Statistics for HSD Circuitry Total Propagation Delay (jisec)
Test Measurements (sans outliers)
Circuitry
HSD PTH
HSD SMT
Mean
13.04|j,sec
5.02|J, sec
Median
13.04
5.02
St. Dev.
0.124
0.086
Min
12.56
4.75
Max
13.44
5.39
Outliers
14.40
4.20 4.29
F-77
-------
APPENDIX F
F.10.5 High Frequency
The HF section shown in the lower right-hand corner of Figure F.50 contains two major
subsections, the low-pass filters (LPF) and the transmission line coupler (TLC). The TLC traces on
layer 4 of the board are on the backside of the board. The LPF/PTH subsection is above the LPF/SMT
subsection. Each of these subsections has discrete ceramic capacitors and three inductor-capacitor
(LC) filters, with the inductor printed on the circuit board in a spiral pattern. The HF circuits allow
evaluation of circuit performance up to IGHz (1000MHz).
Purpose of the High Frequency Experiment
Flux residues may affect the performance of LPF printed circuit inductors and transmission lines
due to parasitic resistances and parasitic capacitances. Since the transmission lines are separated by
only 10 mils, flux residues between the lines may affect their performance.
LPF Circuit Description
An inductor-capacitor (LC) LPF consists of a series inductor followed by a shunt capacitor. A
low-frequency signal passes through the LPF without any loss since the inductor acts as a short circuit
and the capacitor acts as an open circuit for such signals. Conversely, a high-frequency signal is
blocked by the LPF since the inductor acts as an open circuit and the capacitor acts as a short circuit
for such signals.
When a sine wave test signal is passed through an LPF, its amplitude is attenuated as a function of
frequency. The relationship between the output and input voltage amplitudes can be expressed as a
transfer function. The transfer function, Vout / Vm, was measured to determine any effects of the 1 ow-
residue fluxes.
The transfer function is measured in decibels (dB) as a function of frequency. A decibel can be
expressed in terms of voltage as follows:
= 20 log
The PTH transfer function differs from the SMT transfer function due to the self inductance of the
capacitor through-hole leads.
LPF Circuit Board Design
The three LC LPFs for each of the SMT and PTH circuits were designed to have the following
cutoff frequencies: 800, 400, and 200 MHz. Cutoff frequency is that frequency for which the transfer
function is -3 dB. The respective component values chosen for the LC filters are 16 nH (nano-Henries)
and 6.4pF (pico-Farads), 32 nH and 13 pF, and 65 nH and 24pF. Most LPF circuitry was placed on
Layer 1, with Layer 2 used as a ground plane. Crossovers needed to connect the LPF circuits are on
Layer 4.
The LPF circuits were designed to operate with a 50Q test system, so all interconnect traces
longer than 0.10 in were designed as 50Q transmission lines to avoid signal distortion. The LPF
circuits were predicted to have less than 2 dB loss below 150 MHz, approximately 6 dB loss near 235
F-78
-------
APPENDIX F
MHz, and greater than 40 dB loss at 550 MHz and beyond. The measured response of the LPF/SMT
circuit is close to that predicted except that the transfer function decreases more rapidly than predicted
above 3 50 MHz. As stated previously, the PTH circuit transfer function did not perform similarly to
the SMT, particularly at frequencies above 150 MHz.
+ •
-0.325
+ -
-0.300
+ -
-0.275
+ -
-0.250
+ -
-0.225
+ -
-0.200
-dB
Figure F.58 Dotplot of 473 Measurements of the Response for HF PTH at 50 MHz
(each dot represents up to 2 points)
- + -
240.0
+ -
244.0
+ -
248.0
+ -
252.0
+ -
256.0
+ -
260.0
-MHz
Figure F.59 Dotplot of 472 Measurements of the Frequency for HF PTH at -3dB
(each dot represents up to 2 points)
- + -
424.0
+ -
432.0
+ -
440.0
+ -
448.0
+ -
456.0
+ -
464.0
-MHz
Figure F.60 Dotplot of 474 Measurements of the Frequency for HF PTH at^40dB
(each dot represents up to 2 points)
F-79
-------
APPENDIX F
Baseline Testing Results for HF LPF
Data modeling showed that surface finish and flux type had slights effects on the HF LPF
frequencies and responses for HF PTH 50 MHz, HF PTH f(-3dB), HF PTH f(-4Q6B), HF SMT 50
MHz, and HF SMT f(-3dB). The response, HF SMT f(-40dB), was 5 to 12 MHz lower for PWA with
OSP, immersion Ag, or immersion Au/Pd surface finishes. However, the range of frequencies for this
response was only from 630.7 MHz to 680.60 MHz, so the changes in frequency are relatively small.
Figures F.58 to F.59 provide dotplot displays of 480 measurements for the six HF LPF responses. The
summary statistics for these responses are given in Table F.35 (Note there are several outliers
identified in this table).
-dB
-0.315
-0.280
-0.245
-0.210
-0.175
-0.140
Figure F.61 Dotplot of 473 Measurements of the Response for HF SMT at 50 MHz
(each dot represents up to 2 points)
- + -
273.6
+ -
275.2
+ -
276.8
+ -
278.4
+ -
280.0
+ -
281.6
-MHz
Figure F.62 Dotplot of 469 Measurements of the Frequency for HF SMT at -3dB
(each dot represents up to 7 points)
- + -
630
- - - + -
640
- - - + -
650
- - - + -
660
- - - + -
670
- - - + -
680
-MHz
Figure F.63 Dotplot of 469 Measurements of the Frequency for HF SMT at -40dB
(each dot represents up to 2 points)
The distribution in Figure F.59 is different from the other 22 electrical responses in that it displays
a bimodal distribution for HF PTH f(-3dB) with one group of frequencies centered at approximately
245MHz and the other group at 256MHz. Data modeling showed that the differences between these
F-80
-------
APPENDIX F
two groups were not related to any of the experimental parameters (surface finish or flux) nor were
they related to fixture or time of test. A possible explanation for the bimodal distribution is differences
in date lots for the components. However, date lot information were not recorded prior to processing
and thus, the date lot hypothesis cannot be confirmed. Since the JTP acceptance criterion is based on
change after exposure to environmental conditions, the bimodal distribution could potentially be
important if the measurements were not repeatable. Twenty board serial numbers were randomly
selected for retestto see if the measurements were repeatable with 10 boards from the distribution
centered at 245MHz and 10 boards from the distribution centered at 256MHz. These two groups of 10
were equally split between fixtures A and B on the CCAMTF ATS. Table F.36 gives the differences
between the initial baseline measurements and those from the repeat test. The differences in this table
are all quite small. The correlation of the measurements on fixture A is 0.995 and on fixture B it is
0.982, which indicates excellent repeatability. Thus, other than being a curiosity, the bimodal
distribution for HF PTH f(-3dB) will have no practical effect on the test results.
Table F.35 Summary Statistics for 393 Test Measurements for Response (dB) or Frequency (MHz)
for HF LPF (sans outliers)
Circuitry Mean Median St. Dev. Min Max Outliers
HF PTH 50 MHz -0.254 dB -0.252 0.022
HFPTH-3dB 250.6MHz 250.7 5.65
HFPTH-40dB 440.7MHz 440.1 6.01
HFSMTSOMHz -0.242 dB -0.242 0.023
HFSMT-3dB 278.3MHz 278.6 1.20
HFSMT-40dB 660.2MHz 661.0 7.66
-0.319 -0.194 -0.351
-0.148
-0.130
-0.096
240.0 260.8 227.4
305.3
307.1
308.3
425.3 464.4 506.6
507.8
513.7
-0.329 -0.144 -0.447
-0.066
-0.061
273.8 282.2 225.2
299.4
302.9
355.2
383.1
389.6
630.7 680.6 694.8
708.5
721.5
862.8
877.7
-0.150
-0.138
-0.107
230.5
306.5
307.7
308.9
507.2
513.1
514.3
-0.074
-0.062
295.8
301.8
302.9
381.9
384.3
701.9
719.8
758.3
872.3
890.2
924.6
F-81
-------
APPENDIX F
Test
Table F.36 Results from Repeat Testing of the HF PTH f(-3dB) Circuit
Fixture A Fixture B
Baseline Repeat Difference Baseline Repeat Difference
1
2
3
4
5
6
7
8
9
10
244.2
245.3
246.5
247.1
253.1
255.4
256.0
257.2
259.0
259.6
243.0
244.8
246.5
247.1
254.3
255.4
256.0
257.8
259.0
259.0
1.23
0.55
-0.03
-0.03
-1.15
-0.04
-0.03
-0.61
0.00
0.60
242.4
244.2
245.3
246.5
248.9
253.7
254.8
256.0
257.8
259.0
243.0
245.3
245.9
244.2
250.1
255.4
255.4
258.4
258.4
259.0
-0.57
-1.14
-0.64
2.34
-1.19
-1.74
-0.64
-2.41
-0.61
0.00
TLC Circuit Description
Figure F.64 shows a diagram of the TLC subsection. The LPFs described above are lumped
element circuits since the capacitors are discrete components. The TLC lines are distributed element
circuits with the resistors, inductors, and capacitors distributed along the lines. A circuit model for the
lines is shown in Figure F.65.
J9
\
J7
J8
J10 / \
Figure F.64 Diagram of the HF/TLC Subsection
R
trace
© Vin
R
leakage
J ?
Rtrace LL
"* CL Rleakage '.
•
* ~r
R
trace
LL
CL
R
leakage '.
Vout
CL
Figure F.65 HF/TLC Distributed Element Model
The inductance and capacitance for a transmission line with a ground plane are, respectively:
LL = Q.
(F.16)
in
(F.17)
F-82
-------
APPENDIX F
where Ro = characteristic resistance and er = dielectric constant of the board material.
The TLC RO was designed to be 50Q for operation with a 50Q test system. For FR-4 epoxy
(board substrate material), LL is about 9.6 nH/in and CL is about 3.8pF/in.
The TLC was tested with a sine wave signal similar to the one used in testing the LPFs. The
source resistance was 50Q and the three output terminals were connected to 50Q loads.
TLC Circuit Board Design
The transmission line coupler (TLC) circuit has a pair of coupled 50 Q transmission lines with
required measurable performance frequencies less than 1000 MHz. Layer 4 of the printed wiring board
(PWB) was used to route the TLC circuit, with Layer 3 used as the ground plane. The TLC circuit is a
5 in long pair of 0.034 in wide 50Q transmission lines spaced 0.010 in apart. The circuit design
incorporated the board dielectric constant of about 3.8 and the .020 in spacing between copper layers.
A computer-aided circuit design tool (Libra) was used to model the TLC circuit. Performance
measured on a test PWB agreed very closely with the forward and reverse coupling predictions
between 45 MHz and 1000 MHz.
Baseline Testing Results for HF TLC
Data modeling showed that surface finish and flux type had very slight effect on the FTP TLC
frequencies and responses for HF TLC 50 MHz, HF TLC 500 MHz, HF TLC 1000 MHz, HF TLC
Reverse Null Frequency, and HF TLC Reverse Null Response. Figures F.66 to F.70 provide dotplot
displays of 480 measurements for the five HF TLC responses. Summary statistics for these responses
are given in Table F.37 (Note the outliers identified in this table).
-42.0 -40.0 -38.0 -36.0 -34.0 -32.0
Figure F.66 Dotplot of 479 Measurements of the Response for HF TLC at 50 MHz
(each dot represents up to 4 points)
-18.90 -18.20 -17.50 -16.80 -16.10 -15.40
Figure F.67 Dotplot of 479 Measurements of the Response for HF TLC at 500 MHz
(each dot represents up to 3 points)
F-83
-------
APPENDIX F
-13.20 -12.80 -12.40 -12.00 -11.60 -11.20
Figure F.68 Dotplot of 478 Measurements of the Response for HF TLC at 1000 MHz
(each dot represents up to 2 points)
--- + --------- + --------- + --------- + --------- + --------- + ---
636.0 642.0 648.0 654.0 660.0 666.0
Figure F.69 Dotplot of 479 Measurements of the HF TLC Reverse Null Frequency
(each dot represents up to 2 points)
_ + + + + + + (JB
-66.0 -60.0 -54.0 -48.0 -42.0 -36.0
Figure F.70 Dotplot of 479 Measurements of the HF TLC Reverse Null Response
(each dot represents up to 2 points)
Table F.37 Summary Statistics for 480 Test Measurements for Response (dB) or Frequency (MHz) for HF TLC
(sans outliers)
Circuitry Mean Median St. Dev. Min Max Outliers
HF TLC 50 MHz
HF TLC 500 MHz
HF TLC 1000 MHz
HF TLC RNF
HF TLC RNR
-37.57 dB
-18.34 dB
-12.56 dB
649.6 MHz
-44.82 dB
-37.34
-18.43
-12.60
649.1
-44.01
0.974
0.403
0.258
4.77
5.25
-42.74
-19.29
-13.15
636.6
-64.89
-33.05
-15.57
-11.07
665.1
-34.12
-6.13
-6.90
-7.05
935.3
-9.67
-8.94
F-84
-------
APPENDIX F
F.10.6 Other Networks (Leakage Currents)
The test PWA also contains three test patterns to provide tests for current leakage: (1) the pin grid
array (PGA), (2) the gull wing (GW), and (3) 10-mil spaced pads. A 100V source was used to
generate leakage currents.
Purpose of the Experiments
The PGA, GW, and 10-mil pads allow leakage currents to be measured on test patterns that are
typical in circuit board layouts. These patterns contain several possible leakage paths and the leakage
could increase with the presence of flux residues and environmental exposure. In addition, solder
mask was applied to portions of the PGA and GW patterns to evaluate its effect on leakage currents
and the formation of solder balls.
Pin Grid Array
The PGA hole pattern has four concentric squares that are electrically connected by traces on the
top layer of the board as shown in Figure F.71. The pattern also has four vias just inside the corners of
the innermost square that are connected to that square. Four vias were placed inside the innermost
square to trap flux residues. Two leakage current measurements were made: (1) between the two inner
squares (PGA-A) and (2) between the two outer squares (PGA-B), as shown in Figure F.71. Solder
mask covers the holes of the two outer squares on the bottom layer, allowing a direct comparison of
similar patterns with and without solder mask.
Rather than an actual PGA device, a socket was used since it provided the same soldering
connections as a PGA device. Also, obtaining leakage measurements on an actual PGA is nearly
impossible due to complexity of its internal semiconductor circuits.
Gull Wing
The upper half of the topmost GW lands and the lower half of the bottom most GW lands were
covered with solder mask to create a region that is susceptible to the formation of solder balls. The
lands were visually inspected to detect the presence of solder balls. A nonfunctional GW device is
installed with every other lead connected to a circuit board trace forming two parallel paths around the
device. Total leakage current measurements were made on adjacent lands of the GW device
10-mil Pads
The 10-mil pads were laid out in two rows of five pads each. The pads within each row were
connected on the bottom layer of the board and leakage between the rows was measured.
Baseline Testing Results for Leakage Currents
The leakage currents are converted to resistance (ohms) through the basic equation R = V/I. Since
the applied voltage is 100 V and the current is measured in nanoamps, this equation can be expressed
as logio R = 11 - logio I.
F-85
-------
APPENDIX F
PGA-B "
PGA-A •.
t Solder
Mask
Figure F.71 PGA Hole Pattern with Solder Mask
Table F.38 Significant Coefficients for the GLM Analyses of Leakage Currents
Experimental Variables 10-MilPad PGA A PGA B Gull Wing
Constant
11.43
10.63
9.8
11.57
OSP
Immersion Ag
Immersion Au/Pd
0.68
0.59
0.28
0.92
0.84
0.49
1.22
1.22
1.52
0.61
0.67
0.40
Flux
OSP*Flux
Ag*Flux
Au/Pd*Flux
1.61
-0.33
-0.37
1.77
-0.26
2.74
-0.60
-0.90
-0.90
0.89
-0.31
Model R2
Standard Deviation
60.99
0.606
74.52
0.542
88.12
0.432
35.04
.681
General linear modeling (GLM) results for logio R are given in Table F.38. The GLM results
show that surface finish and flux type strongly affect leakage currents. To illustrate these effects,
dotplot displays of 480 measurements for the four leakage responses are given by surface finish and
flux in Figures F.72 to F075 and by flux in Figure F.76. The summary statistics for these responses are
given in Tables F.39 and F.40.
F-86
-------
APPENDIX F
. + + + + OSP LR
. + + + + OSP WS
+ + + + Ag LR
+ + + + Ag WS
+ + + + Au/Pd LR
+ + + + Au/Pd WS
+ + + + HASL LR
+ + + + HASL WS
10.0 11.0 12.0 13.0 14.0
Figure F.72 Dotplots for 480 Measurements of Leakage on 10-Mil Pads by Surface Finish and Flux
F-87
-------
APPENDIX F
+ + + + OSP LR
+ + + + OSP WS
+ + + + Ag LR
+ + + + Ag WS
+ + + + Au/Pd LR
+ + + + Au/Pd WS
+ + + + HASL LR
+ + + + HASL WS
10.0 11.0 12.0 13.0 14.0
Figure F.73 Dotplots for 480 Measurements of Leakage on PGA A by Surface Finish and Flux
F-8
-------
APPENDIX F
LR
WS
. + + + + Ag LR
. + 4 4 4 Ag WS
- + + + + Au/Pd LR
+ + + + Au/Pd WS
+ + + + HASL LR
+ + + + HASL WS
10.0 11.0 12.0 13.0 14.0
Figure F.74 Dotplots for 480 Measurements of Leakage on PGA B by Surface Finish and Flux
F-8
-------
APPENDIX F
. + + + + OSP LR
. + + + + OSP WS
+ + + + Ag LR
+ + + + Ag WS
+ + + + Au/Pd LR
+ + + + Au/Pd WS
+ + + + HASL LR
+ + + + HASL WS
10.0 11.0 12.0 13.0 14.0
Figure F.75 Dotplots for 480 Measurements of Leakage on the Gull Wing by Surface Finish and Flux
F-90
-------
APPENDIX F
Table F.39 Summary Statistics for Leakage Currents Test Measurements
and Flux
Circuitry Surface Finish Flux Mean Median St. Dev.
10-Mil Pads OSP
Immersion Ag
Immersion Au/Pd
HASL
PGA A OSP
Immersion Ag
Immersion Au/Pd
HASL
PGA B OSP
Immersion Ag
Immersion Au/Pd
HASL
Gull Wing OSP
Immersion Ag
Immersion Au/Pd
HASL
LR
WS
LR
WS
LR
WS
LR
WS
LR
WS
LR
WS
LR
WS
LR
WS
LR
WS
LR
WS
LR
WS
LR
WS
LR
WS
LR
WS
LR
WS
LR
WS
12.11
13.39
12.02
13.26
11.81
13.22
11.29
13.15
11.59
13.28
11.47
12.98
11.23
12.78
10.45
12.56
11.10
13.23
11.10
12.94
11.47
13.16
9.74
12.70
12.15
13.10
12.23
13.14
11.99
12.53
11.57
12.44
11.94
13.52
11.90
13.30
11.73
13.22
11.29
13.40
11.62
13.30
11.39
12.94
11.20
12.80
10.46
12.66
11.11
13.30
11.12
13.00
11.44
13.10
9.75
12.70
12.40
13.22
12.32
13.46
12.02
12.66
11.52
12.70
0.77
0.55
0.76
0.38
0.54
0.60
0.33
0.67
0.67
0.26
0.66
0.33
0.56
0.62
0.28
0.58
0.43
0.25
0.47
0.27
0.50
0.39
0.29
0.35
0.90
0.65
0.60
0.70
0.57
0.64
0.39
0.86
by Surface Finish
Min Max
10.91
11.12
10.73
12.48
10.47
11.91
10.34
11.57
10.38
12.12
10.16
12.18
10.18
11.67
9.94
11.29
9.91
11.85
10.13
12.19
10.09
12.51
9.11
11.65
9.01
11.44
10.66
10.91
10.35
10.69
10.26
9.48
15.00
14.00
15.00
14.00
14.00
15.00
12.30
15.00
13.15
13.70
13.22
14.00
13.15
15.00
11.10
13.40
12.09
13.52
12.40
13.30
13.15
15.00
10.35
13.40
13.52
16.00
13.52
14.00
13.22
14.00
12.62
13.52
F-91
-------
APPENDIX F
. + + + + lOmilPad LR
. + + + + lOmilPad WS
. + + + + PGA A LR
Each dot represents up to 2 points
+ + + + PGA A WS
10.0 11.0 12.0 13.0 14.0
Figure F.76 Dotplots for 480 Leakage Measurements by Flux
F-92
-------
APPENDIX F
+ + + + PGA B LR
Each dot represents up to 3 points ::
+ + + + PGA B WS
. + + + + _ GullWing LR
+ + + + GullWing WS
10.0 11.0 12.0 13.0 14.0
Figure F.76 Continued
F-93
-------
APPENDIX F
F.10.7 Stranded Wires
Two 22-gauge stranded wires were hand soldered just to the left of the edge connector. One wire
was soldered directly into the board through holes and the other were soldered to two terminals, El7
and E18. Each wire is 1.5 in long, is silver coated, and has white PTFE insulation. All wires were
stripped, tinned, and cleaned in preparation for the soldering process.
Purpose of the Stranded Wire Experiment
Stranded wires were used to evaluate flux residues and subsequent corrosion.
Table F.40 Summary Statistics for Leakage Currents Test Measurements by Flux
Circuitry Flux Mean Median St. Dev. Min Max
10-Mil Pads
PGA A
PGAB
Gull Wing
LR
WS
LR
WS
LR
WS
LR
WS
11.80
13.25
11.18
12.90
10.85
13.01
11.99
12.80
11.68
13.30
11.10
13.00
11.00
13.07
12.02
12.94
0.70
0.56
0.72
0.54
0.79
0.38
0.68
0.78
10.34
11.12
9.94
11.29
9.11
11.65
9.01
9.48
15.00
15.00
13.22
15.00
13.15
15.00
13.52
16.00
Circuit Description
The 5 A lOOjis pulse used to test the HCLV circuit was injected into each of the stranded wires for
electrical test. A separate PWB trace was connected to each end of the stranded wire. Test wires were
connected to the separate traces allowing to provide the means to measure the voltage drop across the
stranded wires. In this manner, the voltage drop was measured independently from any voltage drop in
the test wires conducting the 5 A pulse to the stranded wires.
Baseline Testing Results for Stranded Wires
Surface finish and flux type had very little effect on the HF TLC frequencies and responses for HF
TLC 50 MHz, HF TLC 500 MHz, HF TLC 1000 MHz, HF TLC Reverse Null Frequency, and HF
TLC Reverse Null Response. Figures F.77 and F.78 provide dotplot displays of 480 measurements for
the two stranded wire voltages. The summary statistics for these responses are given in Table F.41.
8.0 10.0 12.0 14.0 16.0 18.0
Figure F.77 Dotplots for 480 Voltage Measurements for Stranded Wire 1
(each dot represents up to 11 points)
F-94
-------
APPENDIX F
Circuitry
+ + + + + + -mV
20.0 22.0 24.0 26.0 28.0 30.0
Figure F.80 Dotplots for 476 Voltage Measurements for Stranded Wire 2
(each dot represents 8 points)
Table F.41 Summary Statistics for Stranded Wires Voltage Test Measurements
Mean Median St. Dev. Min Max Outliers
Stranded Wire 1
Stranded Wire 2
11.75mV
24. 82m V
12.00
25.00
1.60
2.41
8.00
19.00
18.00
30.00
42,43, 45, 45
F.10.8 Summary Statistics for All Baseline Measurements
For ease of reference, Table F.42 gives the summary statistics for all 23 electrical responses from
the test PWA.
F.10.9 Listing of Components
All functional component types conformed to commercial specifications and were ordered pre -
tinned (to the extent possible). Components were not pre-cleaned before use. A listing of all
components is given in the Table F.43.
F-95
-------
APPENDIX F
Table F.42 Summary Statistics for All Baseline 480 Measurements (sans outliers)
Circuitry Mean Median St. Dev. Min Max Outliers
Hi
HCLV PTH
HCLV SMT
6.88V
7.20V
Hi
HVLC PTH
HVLC SMT
5.04MA
4.95MA
[h Current Low Voltage
6.92
7.20
0.16
0.10
6.60
6.88
7.20
7.44
^h Voltage Low Current
5.04
4.95
0.024
0.011
4.972
4.914
5.148
4.976
5.203 5.232
High Speed Digital
HSD PTH
HSD SMT
13.04|j,sec
5.02\i sec
0.12
0.08
13.04
5.02
12.56
4.75
13.44
5.39
14.40
High Frequency Low Pass Filter
HF PTH 50 MHz
HF PTH -3dB
HFPTH-40dB
HF SMT 50 MHz
HFSMT-3dB
HFSMT-40dB
-0.254 dB
250.5MHz
440.5MHz
-0.242 dB
278.4 MHz
660.7 MHz
-0.253
249.2
440.1
-0.241
278.6
661.6
0.024
5.74
5.96
0.022
1.21
7.46
-0.319
230.5
425.3
-0.329
273.8
639.0
-0.194
260.8
464.4
-0.173
282.2
680.6
-0.351 -0.150
-0.148 -0.138
-0.130 -0.107
-0.096
227.6 230.5
305.3 306.5
307.2 307.7
308.3 308.9
506.6 507.2
507.8 513.1
513.7 514.3
-0.447 -0.164
-0.144 -0.074
-0.066 -0.062
-0.061
225.2 295.8
299.4 301.8
302.9 302.9
355.2 381.9
383.1 384.3
389.6
694.8 701.9
708.5 719.8
721.5 758.3
862.8 872.3
877.7 890.2
924.6
High Frequency Transmission Line Coupler
HF TLC 50 MHz
HF TLC 500 MHz
HF TLC 1000 MHz
HF TLC RNF
HF TLC RNR
-37.61 dB
-18.31 dB
-12.55 dB
649.5MHz
-44.68 dB
-37.38
-18.40
-12.58
649.1
-43.96
0.957
0.389
0.254
4.87
5.208
-42.74
-19.29
-13.15
636.6
-64.89
-33.05
-15.57
-11.07
665.1
-34.12
-6.13
-6.90
-7.05 -8.94
935.3
-9.67
Leakage (resistance in log 10 ohms)
10-Mil Pads (LR)
10-Mil Pads (WS)
PGA A (LR)
PGA A (WS)
PGA B (LR)
PGA B (WS)
Gull Wing (LR)
Gull Wing (WS)
11.79
13.27
11.17
12.89
10.84
13.01
12.03
12.81
11.69
13.40
11.11
13.05
11.04
13.10
12.05
12.96
0.64
0.56
0.70
0.52
0.80
0.34
0.66
0.71
10.63
11.12
10.01
11.29
9.11
11.65
10.15
10.52
15.00
15.00
13.15
14.00
12.46
13.52
13.52
14.00
Stranded Wire
Stranded Wire 1
Stranded Wire 2
11.75mV
24.71mV
12.00
25.00
1.50
2.38
8.00
19.00
18.00
30.00
42, 43, 45, 45
F-S
-------
APPENDIX F
Table F.43 Listing of Components for the Test PWA
MFGP/N Description Quantity per Supplier
Assembly
ACC916228-2
350-60-2
402-632-38-0110
231-632-A-2
RWR89N10ROFR
M55342M09B10MOM
RLR07C1005FR
M55342M09B10POM
2309-2-00-44-00-07-0
KA29/127BPMCTH
C1825N474K5XSCxxxx
C0627104K1X5CS7506
C1825N104K1XRC
C062T105K5X5CSxxxx
C052G130J2G5CR
CDR31BP130BJWR
C052G240J2G5CRxxxx
C0805N240J1 GRC373 17537
C0805N629B1GSC37317535
C052G629D2G5CR7535
JM38510/33001B2A
JM38510/33001BCA
QFP80T25
CS1
CKR06
SC1210E7Axxxx
D034
RN65
RN55(sub for CS1, Qty 800)
SR1210E7A
T05
T0220M-3
5162-5013-09
131-3701-201
PGA Socket, 18X18 (223 PINS)
6 Split washer
6-32 UNC Mach Screw
6-32 UNC Mach Screw Nut
Resistor, 10 Ohm, Axial
Resistor, 10 Ohm, Surface Mnt
Resistor, 1 OMeg Axial
Resistor, lOMeg Surface Mount
Swage pin
29 Pin Connector,Pretin
CAP, .47 UF, Surf Mnt
CAP, 0.1 UF, Radial
CAP, 0.1 UF, Surf Mnt
CAP, 1 UF, Radial
CAP, 13 PF, Radial
CAP, 13PF, Surf Mnt
CAP, 24 PF, Radial
CAP, 24 PF, Surf Mnt
CAP, 6.2 PF +0.5%, Surf Mnt
CAP, 6.2 PF, +0.5%, Radial
20 Pin LCC
14PinDual-In-Line
80 Pin SQ Flat Pack
Cap
Cap
Cap
Diode
Resistor
Resistor
Resistor
Transistor
Transistor
Connector, RF, OMNI Spec
Sub for 5 162-50 13 -09
1
3
3
3
7
7
5
5
17
1
7
7
7
7
1
1
1
1
1
1
1
2
1
1
2
13
13
1
5
18
4
3
10
10
AMP
Barnhill Bolt
Barnhill Bolt
Barnhill Bolt
Dale
Dale
Dale
Dale
Harrison HEC
Hypertonics
Kemet
Kemet
Kemet
Kemet
Kemet
Kemet
Kemet
Kemet
Kemet
Kemet
TI (808810.1001)
TI (808810.1)
Top Line
Top Line
Top Line
Top Line
Top Line
Top Line
Top Line
Top Line
Top Line
Top Line
TTI
Penstock
F-97
-------
APPENDIX F
F.ll Design for the Environment Printed Wiring Board Project Performance
Demonstration Methodology for Alternative Surface Finishes
Note: This methodology is based on input from members of a Performance Demonstration Technical
Workgroup, which includes representatives of the printed wiring board (PWB) industry manufacturers,
assemblers, and designers; industry suppliers; public interest group; Environmental Protection Agency
(EPA); the University of Tennessee Center for Clean Products and Clean Technologies; and other
stakeholders. As the testing continues, there may be slight modifications to this methodology.
I. OVERVIEW
A. Goals
The U.S. Environmental Protection Agency's (EPA=s) Design for the Environment (DfE) Printed
Wiring Board (PWB) Project is a cooperative partnership among EPA, the PWB industry, public
interest groups, and other stakeholders. The project encourages businesses to incorporate
environmental concerns into their decision-making processes, along with the traditional parameters of
cost and performance, when choosing which technologies and processes to implement. To accomplish
this goal, the DfE PWB Project collects detailed data on the performance, cost, and risk aspects of one
Ause cluster® or manufacturing operation, and makes it available to all interested parties. This use
cluster focuses on surface finishes used in PWB manufacturing. Analyses on the performance, cost,
and risk of several alternative surface finishes will be conducted throughout this project, and the results
will be documented in the final project report, titled the Cleaner Technologies Substitutes Assessment
or CTSA. This methodology provides the general protocol for the performance demonstration portion
of the DfE PWB Project. The CTSA is intended to provide manufacturers and designers with detailed
information so that they can make informed decisions, taking environmental and health risks into
consideration, on what process is best suited for their own facility.
Surface finishes are applied to PWBs to prevent oxidation of exposed copper on the board, thus
ensuring a solderable surface when components are added at a later processing stage . Specifically, the
goals of the DfE PWB Surface Finishes Project are:
1) to standardize existing information about surface finish technologies;
2) to present information about surface finish technologies not in widespread use, so PWB
manufacturers and designers can evaluate the environmental and health risks, along with the cost and
performance characteristics, among different technologies; and
3) to encourage PWB manufacturers and designers to follow the example of this project and evaluate
systematically other technologies, practices, and procedures in their operations that affect the
environment.
B. General Performance Demonstration Plan
The most widely used process for applying surface finishes in commercial PWB shops is hot air solder
leveling (HASL). In this process, tin-lead is fused onto exposed copper surfaces. This process was
selected as the focus of the Design for the Environment Project because HASL is a source of lead
waste in the environment and because there are several alternative surface finishes available on the
market. A comprehensive evaluation of these technologies, including performance, cost, and risk,
however, has not been conducted. In addition, a major technical concern is that the HASL process
F-98
-------
APPENDIX F
does not provide a level soldering surface for components.
The general plan for the performance demonstration portion of the Project is to collect data on
alternative surface finish processes during actual production runs at sites where the processes are
already in use. Demonstration facilities will be nominated by suppliers. These sites may be customer
production facilities, customer testing facilities (beta sites), or supplier testing facilities, in that order of
preference. Each demonstration site will receive standardized test boards which they will run through
their surface finish operation during their normal production operation.
The test vehicle design will be tested on the test board designed by the Sandia National Laboratory
Low-Residue Soldering Task Force (LRSTF). The same test vehicle was used by the Circuit Card
Assembly and Materials Task Force (CCAMTF). CCAMTF is a joint industry and military program
evaluating several alternative technologies including Organic Solderability Preservative (OSP),
Immersion Silver, Electroplated Palladium/Immersion Gold, Electroless Nickel/Immersion Gold, and
Electroplated Palladium. CCAMTF conducted initial screening tests on coupons for each of these
surface finishes, however, they will conduct functionality tests only for the OSP (thick), Electroplated
Palladium/Immersion Gold, and Immersion Silver technologies.
II. PERFORMANCE DEMONSTRATION PROTOCOL
A. Technologies to be Tested
The technologies that the DfE Project plans to test include:
1. HASL (baseline)
2. OSP - Thick
3. Immersion Tin
4. Immersion Silver
5. Electroless Nickel/Immersion Gold
6. Nickel/Palladium/Gold
B. Step One: Identify Suppliers and Test Sites/Facilities
Performance Demonstration Technical Workgroup members identified suppliers of the above product
lines. Any supplier of these technologies who wanted to participate was eligible to submit its product
line, provided that it agreed to comply with the testing methodology and submit the requested
information, including chemical formulation data. All proprietary information submitted is bring
handled as Confidential Business Information. For each product line submitted, the supplier
completed a Supplier Data Sheet detailing information on the chemicals used, equipment requirements,
waste treatment recommendations, any limitations of the technology, and other information on the
product line.
Performance demonstration sites were nominated by suppliers. They identified sites that are currently
using their alternative surface finish product line in the following order of preference:
customer production facilities (first preference)
beta sites - customer testing facilities (second preference)
supplier testing facilities (third preference)
The final number of product lines evaluated for each type of alternative surface finish was determined
based on the number of suppliers interested in participating and on the resources available. Each
F-99
-------
APPENDIX F
accepted product line was tested at one or two sites. If a supplier has more than one substantially
different product line within a technology, the supplier was allowed to submit names of test facilities
for each of the products.
C. Step Two: Fabricate Test Vehicles
Test board were fabricated based on the Sandia National Laboratory Low-Residue Soldering Task
Force (LRSTF) test board design. This general design was also used in the CCAMTF testing. For the
DfE Project, uncoated test boards with comb pattern spacing of 8 mil, 12 mil, 16 mil, and 20 mil will
be used.
All test boards are of the same design, and were fabricated at a single shop to minimize the variables
associated with board production. All manufacturing steps, up to but not including the soldermask
application, were completed by the test board fabricator. For each supplier's product line, 24 boards
were shipped to the demonstration site where the alternative surface finish was applied, beginning with
the soldermask application step.
The design of the LRSTF PWB was based on input from a large segment of the manufacturing
community, and thus reflects the multiple requirements of the commercial sector. Each quadrant of the
LRSTF PWA contain one of the following types of circuity:
High-current low-voltage (HCLV)
High-voltage low current (HVLC)
High speed digital (HSD)
High frequency (HF)
The components in each quadrant represent two principal types of soldering technology:
Plated through hole (PTH) - leaded components are soldered through vias in the circuit board by
means of a wave soldering operation.
Surface mount technology (SMT) - components manufactured with solder tips on two of their
opposite ends are temporarily attached to the substrate with an adhesive and then they are soldered to
pads on the circuit board by passing the circuit board through a reflow oven to reflow the solder tips.
The LRSTF PWA also has two stranded wires (SW) that are secured to the circuit board with hand
soldering, such as used in repair operations. This assembly also contains other networks that are used
to monitor current leakage.
D. Step Three: Collect Background Information
After the suppliers identified appropriate test facilities and completed a supplier data sheet, an
independent observer contacted the designated facilities. The observer scheduled a date for the on-site
performance demonstration. A questionnaire was sent to each facility prior to the site visit to collect
information on the surface finish technology used and background information on the facility, such as
the size and type of product produced. On the day of the performance demonstration, the observer
reviewed the background questionnaire and discussed any ambiguities with the facility contect.
F-100
-------
APPENDIX F
E. Step Four: Conduct the Surface Finish Performance Demonstration
After test boards were distributed to the demonstration sites, the surface finish performance
demonstrations were conducted. The surface finish was applied to the test boards as part of the normal
production run at the facility. The test boards were placed in the middle of the run to reflect actual
production conditions. The facility applied the solder mask it normally uses in production. The usual
process operator operated the line to minimize error due to unfamiliarity with the technology. All test
boards were processed in the same production run.
On the day of the performance demonstration, the observer collected data on the surface finish process.
During the demonstration, the observer recorded information on surface finish technology
performance, including information on chemicals, equipment, and waste treatment methods used. In
addition, other information needed for the performance, cost, or risk analyses, as described below, was
collected.
1. Product Cost: A cost per square foot of panel processed will be calculated. This number will be
based on information provided by product suppliers, such as purchase price, recommended bath
life and treatment/disposal methods, and estimated chemical and equipment costs per square foot
panel per day. Any "real world" information from PWB manufacturers, such as actual dumping
frequencies, treatment/disposal methods, labor requirements, and chemical and equipment costs,
will be collected during performance demonstrations, as required for use in the cost analysis. The
product cost may differ for difference shop throughput categories.
2. Product Constraints: Information on any incompatibilities such as soldermask, flux, substrate
type, or assembly process will be included. This information will be submitted by the suppliers
and may also be identified as a result of the performance demonstrations.
3. Special storage, safety, and disposal requirements: Information on flammability or special
storage requirements of the chemicals used in the process will be requested from the suppliers.
Suppliers will provide recommendations on disposal or treatment of wastes associated with the use
of their product lines. Information on these issues was also collected from participating facilities
during the performance demonstrations. The storage and disposal costs will be a factor in
determining the adjusted cost of the product. This project does not entail a life cycle analysis for
disposal of the boards.
4. Ease of use: During the performance demonstration, the physical effort required to use the various
surface finishes effectively will be qualitatively assessed based on the judgement of the operator in
comparison to the baseline technology, HASL. Specific questions such as the following will be
asked: What process operating parameters are needed to ensure good performance? What are the
ranges of those parameters, and is there much flexibility in the process steps? How many hours of
training are required to use this type of surface finish?
5. Duration of Production Cycle: The measured time of the surface finish application process and
the number of operators required will be recorded during the performance demonstration. This
information will be used to measure the labor costs associated with the use of the product line.
Labor costs will be based on the operator time required to run the process using an industry
standard worker wage. The process cycle has been defined as the activities following soldermask
application up to, but not including, gold tab plating. The facilities participating in the
performance demonstration will use the same soldermask they typically use in production
conditions. The observer recorded the type of soldermask used, and information on the facilities'
experiences with other soldermasks to determine if any known incompatibilities exist.
F-101
-------
APPENDIX F
6. Effectiveness of Technology, Product Quality: The performance characteristics of the
assembled boards will be tested after all demonstrations are complete and the boards are assembled
with the functional components. Circuit electrical Performance will be tested to assess the circuit
performance of the functional test vehicle under applicable environmental stress. Circuit
Reliability Testing (functional tests) conditions will include Thermal Shock and Mechanical
Shock. These tests are described in greater detail in Step 5. Qualitative information on shelf life
considerations were collected through the performance demonstrations, where applicable.
7. Energy and Natural Resource Data: Information will be collected from the suppliers and during
the performance demonstrations to evaluate the variability of energy consumption for the use of
different surface finishes. The analysis will also address material use rates and how the rates vary
with the different surface finishes.
8. Exposure Data: Exposure data will be used to characterize chemical exposures associated with
the technologies. Exposure information collected during the performance demonstration may be
supplemented with data from other sources, where available.
F. Step Five: Assemble and Test the Boards
After the surface finish was applied to the test boards at each demonstration facility, the facility sent
the processed boards to one site for assembly. Two different assembly processes were used: a halide-
free, low-residue flux and a halide-containing, water-soluble flux. Table 1 shows the different
assembly methods, and number of test vehicles used for each method. The boards were not assembled
as originally planned, resulting in the uneven distribution of assembly methods.
Table 1: Test Vehicle Distribution by Site and Flux
Site#
1
2
6
o
J
13
16
4
5
10
11
8
9
7
12
14
Surface Finishes*
HASL
HASL
HASL
HASL Totals
OSP-Thick
OSP-Thick
OSP-Thick
OSP Totals
Immersion Tin
Immersion Tin
Immersion Tin
Immersion Tin
Immersion Tin Totals
Immersion Silver
Immersion Silver
Immersion Silver Totals
Electroless Ni/Immersion Au
Electroless Ni/Immersion Au
Electroless Ni/Immersion Au
NI/Au Totals
Subtotals
# of Boards
Assembled with Low
Residue Flux
8
0
8
16
4
8
8
20
0
4
8
8
20
0
8
8
0
8
4
12
84
# of Boards
Assembled with
Water Soluble Flux
8
8
0
16
8
8
0
16
8
8
0
0
16
8
4
12
8
0
8
16
80
Total test boards: 164
Total Boards by
Site and by Surface
Finish
16
8
8
32
12
16
8
36
8
12
8
8
36
8
12
20
8
8
12
28
* Corresponding board identification numbers are listed in Appendix A.
Following assembly, the performance characteristics of the assembled boards will be tested. Testing
will include Circuit Electrical Performance testing and Circuit Reliability Testing.
F-102
-------
APPENDIX F
Circuit Electrical Performance
This test assesses the circuit performance of a functional test vehicle under applicable environmental
stress. The assembled test vehicles will be exposed to 85 ° C at 85% relative humidity for 3 weeks.
The assemblies will be tested prior to exposure, and at the end of three weeks of exposure. Good
experimental design practices will be followed to control extraneous sources of variation. For
example, the assemblies will be placed randomly in the test chamber. If all assemblies cannot be
accommodated in the test chamber at the same time, they will be randomized to maintain balance
among the experimental factors at each test time. A staggered ramp will be used to prevent
condensation (during ramp-up, the temperature will be raised to test level before the humidity is raised
and the procedure will be reversed during ramp-down). The pre-tests and post-tests will be identical.
Circuit Reliability Testing
The same test vehicles used to test circuit electrical performance will be used for the circuit reliability
tests, which include:
Thermal Shock
Mechanical Shock
The electrical functionality of the LRSTF PWA will be evaluated through 23 electrical responses, as
follows:
HCLVPTH voltage
HCLV SMT voltage
Stranded wire 1 voltage
Stranded wire 2 voltage
HVLC PTH current
HVLC SMT current
10-mil spaced pads current leakage
PGA A current leakage
PGA B current leakage
Gull wing current leakage
HSD PTH total propagation delay
HSD SMT total propagation delay
HF LPF PTH 50 MHz response
HF LPF PTH frequency response at -3 dB
HF LPF PTH frequency response at -40 dB
HP LPF SMT 50 MHz response
HF LPF SMT frequency response at -3 dB
HF LPF SMT frequency response at -40 dB
HF TLC 50 MHz forward response
HF TLC 500 MHz forward response
HF TLC 1000 MHz forward response
HF TLC reverse null frequency
HF TLC reverse null response
Table 2 shows the total number of electrical responses that will be measured.
Table 2. Number of Tests to be Conducted
Test Environment
85/85
Thermal Shock
Mechanical Shock
Totals
Number of
PWBs
164
164
Number of Test
Times
2
1
1
4
Number of
Tests
164x2 = 328
164x1 = 164
164x1 = 164
656
Number of Electrical
Responses Measured
164x2x23 = 7,544
164x1x23 = 3,772
164x1x23 = 3,722
15,088
F-103
-------
APPENDIX F
G. Analyze Data and Present Results
The details of the data analysis and results are presented in the "Technical Proposal for this project, in
Appendix B.
III. PERFORMANCE DEMONSTRATION PARTICIPANT REQUIREMENTS
A. From the Facilities/Process Operators:
1. Participating facilities were contacted by the project observer to arrange a convenient data for the
performance demonstration. The observer sent a fact sheet describing the facility's role in the
project.
2. Each facility was asked to complete a background questionnaire prior to the scheduled date of the
performance demonstration and return it to the observer.
3. Each facility was asked to make its process line/process operators available to run the 24 test
boards on the agreed upon date.
4. The process operator met with the independent observer before running the test boards through the
line to explain the unique aspects of the line to the observer. The process operator was asked to be
available to assist the independent observer in collecting information about the line.
B. From the Suppliers of the Process Line Alternatives:
1. Suppliers were asked to submit product data sheets, on which they provided information on
product formulations, product constraints, recommended disposal/treatment etc. The information,
including chemical formulation information, was requested prior to testing. Any proprietary
information was submitted to the University of Tennessee as Confidential Business Information.
2. Suppliers were asked to identify and contact the demonstration sites.
3. Suppliers were asked to attend the on-site performance demonstration if they wishes to do so, but
they were not required to attend.
Attachment A to this Methodology lists "Identification Numbers for Assembled Boards." To
conserve space this information as not been reprinted as part of the CTSA.
Attachment B to this Methodology is the "Technical/Management Proposal for Validation of
Alternatives to Lead Containing Surface Finishes." This Attachment contains the testing and analysis
methodology submitted by Dr. Ronald L. Inman, President, Southwest Technology Consultants in
Albuquerque, MN. Dr. Inman's methodology and results are presented in Chapter 6 of the CTSA and
in Appendix F, and therefore, Attachment B of the Methodology is not repeated here.
F-104
-------
Appendix G
Supplemental Cost Analysis Information
-------
G-l Example Graphic Representation of Cost Simulation Model
G-2 Bath Replacement Criteria for Surface Finishing Processes
G-3 Bills of Activities for Surface Finishing Processes
G-4 Simulation Model Outputs for Surface Finishing Processes
G-5 Chemical Costs by Bath for Individual Surface Finishing Processes
G-6 Total Materials Cost for Surface Finishing Processes
G-l
-------
G-l. Example Graphic Representation of Cost Simulation Model
G-2
-------
G-2 Bath Replacement Criteria for Surface Finishing Processes
Process: HASL
Chemical Bath
Cleaner
Microetch
Flux
Solder
Bath Replacement Criteria'
(ssf/gal)
750
570
NAb
NAb
" Values were selected by averaging the replacement criteria for similar bath types from other alternatives.
b This bath is refilled or continuously maintained through chemical additions rather than replaced. The number of bath
replacements was set at one to reflect the initial bath make-up for the purposes of the computer simulation.
Process: Electroless Nickel/Immersion Gold
Chemical Bath
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Immersion Gold
Bath Replacement Criteria'
(ssf/gal)
750
570
830
1,500
130
890
" Values were determined from data provided by two electro less nickel/immersion gold suppliers. To convert to units of racks per
bath replacement for non-conveyorized processes, multiply by 51.1 gallons and divide by 84.4 ssf/rack.
Process: Electroless Nickel/Electroless Palladium/Immersion Gold
Chemical Bath
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Preinitiator
Electroless Palladium
Immersion Gold
Bath Replacement Criteria'
(ssf/gal)
750
570
830
1,500
130
1,200
150
890
" Values were determined from data provided by two electroless nickel/immersion gold suppliers and one electroless
nickel/palladium/immersion gold supplier. To convert to units of racks per bath replacement for non-conveyorized processes,
multiply by 51.1 gallons and divide by 84.4 ssf/rack.
G-3
-------
Process: OSP
Chemical Bath
Cleaner
Microetch
OSP
Bath Replacement Criteria'
(ssf/gal)
750
570
NAb
Values were determined from data provided by two OSP suppliers. To convert to units of racks per bath replacement for non-
conveyorized processes, multiply by 51.1 gallons and divide by 84.4 ssf/rack. To convert to units of panels per bath replacement
for conveyorized process, multiply by the size of the bath in gallons and divide by 5.66 ssf/panel.
b This bath is refilled or continuously maintained through chemical additions rather than replaced. The number of bath
replacements was set at one to reflect the initial bath make-up for the purposes of the computer simulation.
Process: Immersion Silver
Chemical Bath
Cleaner
Microetch
Predip
Immersion Silver
Bath Replacement Criteria'
(ssf/gal)
750
570
1,000
NAb
Values were determined from data provided by two OSP suppliers. To convert to units of panels per bath replacement for
conveyorized process, multiply by the size of the bath in gallons and divide by 5.66 ssf/panel.
b This bath is refilled or continuously maintained through chemical additions rather than replaced. The number of bath
replacements was set at one to reflect the initial bath make-up for the purposes of the computer simulation.
Process: Immersion Tin
Chemical Bath
Cleaner
Microetch
Predip
Immersion Tin
Bath Replacement Criteria'
(ssf/gal)
750
570
1,250
NAb
Values were determined from data provided by two OSP suppliers. To convert to units of racks per bath replacement for non-
conveyorized processes, multiply by 51.1 gallons and divide by 84.4 ssf/rack. To convert to units of panels per bath replacement
for conveyorized process, multiply by the size of the bath in gallons and divide by 5.66 ssf/panel.
' This bath is refilled or continuously maintained through chemical additions rather than replaced. The number of bath
replacements was set at one to reflect the initial bath make-up for the purposes of the computer simulation.
G-4
-------
G-3 Bills of Activities for Surface Finishing Processes
Activities Associated with the Bath Setup
Activity Description
Wear masks, goggles, rubber gloves, and suitable clothing
Go to storage area
Locate protective equipment
Put on protective equipment
Return to tank
Put in base liquid (usually water)
Open water valve
Wait for measured amount
Close water valve
Document water amount/level
Mix the bath solution
Open the chemical containers
Add the chemicals to the bath
Turn on the agitator
Wait for mixing
Turn off the agitator
Titrate sample
Document
Repeat as necessary
Flush containers
Turn on water valve
Spray containers
Turn off water valve
Place empty container in storage area
Take container to storage
Documentation
Return to tank
Total =
Cost Driver
S/bath setup
labor
labor
labor
protective equipment
labor
S/bath setup
labor
labor
labor
labor
S/bath setup
labor
labor
labor
labor
labor
labor
labor
labor
S/bath setup
labor
labor
labor
S/bath setup
labor
labor
labor
Sper testing
Cost/Activity
S2.50
S2.60
S5.00
S3. 00
S2.00
S15.10
G-5
-------
Activities Associated with the Tank Cleanup
Activity Description
Rinse with water
Obtain spray/rinse equipment
Turn water on
Spray equipment
Turn water off
Obtain scrubbing and cleaning tools
Go to storage area
Find necessary tools
Return to tank
Hand scrub tank
Put on gloves, choose tool
Scrub tank
Return cleaning tools
Go to the storage area
Place tools in correct place
Return to tank
Spray according to schedule
Wait for time to elapse before spraying
Obtain spray equipment
Turn spray on
Spray all cleaning solution from tank
Turn spray off
Operator opens control valve
Find correct control valve
Open valve
Water goes to treatment facility
Wait for water to drain
Operator closes control valve
Locate correct control valve
Close valve
Total =
Cost Driver
S/cleanup
labor
labor
labor
labor
S/cleanup
labor
labor
labor
S/cleanup
labor
labor
cleaning supplies
S/cleanup
labor
labor
labor
S/cleanup
labor
labor
labor
labor
labor
S/cleanup
labor
labor
S/cleanup
labor
S/cleanup
labor
labor
Sper testing
Cost/Activity
S25.00
SI. 00
S30.00
S1.25
S5.00
SI. 00
S2.75
SI. 00
S67.00
G-6
-------
Activities Associated with Sampling and Testing
Activity Description
Get sample
Go to the line
Titrate small sample into flask
Transfer to lab
Test sample
Request testing chemicals
Document request
Locate chemicals
Add chemicals to sample
Mix
Document the results
Return testing chemicals
Relay information to line operator
Return to line
Inform operator of results
Document
Total =
Cost Driver
S/testing
labor
labor
materials
labor
S/testing
labor
labor
labor
labor
materials
labor
labor
labor
S/testing
labor
labor
labor
Sper testing
Cost/Activity
S1.35
S1.35
SI. 00
S3.70
G-7
-------
Activities Associated with Filter Replacement
Activity Description
Check old filter
Pull canister from process
Inspect filter
Decide if replacement is necessary
Get new filer
Go to storage area
Locate new filters
Fill out paper work
Return to tank
Change filter
Pull old filter from canister
Replace with new filter
Replace canister
Fill out paper work
Dispose of old filter
Take old filter to disposal bin/area
Dispose of filter
Return to tank
Fill out paper work
Total =
Cost Driver
S/replacement
labor
labor
labor
S/replacement
labor
labor
labor
labor
S/replacement
labor
labor
filter
labor
labor
S/replacement
labor
labor
labor
labor
Sper replacement
Cost/Activity
S1.50
S1.75
S12.25
S2.00
S17.50
G-8
-------
Activities Associated with Transportation
Activity Description
Paperwork and maintenance
Request for chemicals
Updating inventory logs
Safety and environmental record keeping
Move forklift to chemical storage area
Move to forklift parking area
Prepare forklift to move chemicals
Move to line container storage area
Prepare forklift to move line container
Move forklift to chemical storage area
Locate chemicals in storage area
Move forklift to appropriate areas
Move chemical containers from storage to staging
Move containers from staging to storage
Preparation of chemicals for transfer
Open chemical container(s)
Utilize correct tools to obtain chemicals
Place obtained chemicals in line container(s)
Close chemical containers)
Place line container(s) on forklift
Transport chemicals to line
Move forklift to line
Unload line containers) at line
Move forklift to parking area
Transport chemicals from line to bath
Move line container(s) to bath
Clean line container(s)
Store line container(s) in appropriate area
Total =
Cost Driver
S/transportation
labor
labor
labor
S/transportation
labor
labor
labor
labor
labor
S/transportation
labor
labor
filter
S/transportation
labor
labor
labor
labor
labor
S/transportation
labor
labor
labor
S/transportation
labor
labor
labor
Sper testing
Cost/Activity
S1.10
S3.22
S1.15
S1.78
S1.15
S.88
S9.28
G-9
-------
G-4 Simulation Model Outputs for Surface Finishing Processes
NAME:
Throughput:
HASL, non-conveyorized
260,OOOK ssf
ARENA Simulation Results
Replication ended at time: 17831.4 min.
Tally Variables
Identifier Average Half Width Minimum
Maximum
Observations
Takt Time
Time in system
Counters
5.7866
19.957
(Corr)
4.8613
1.4700
7.9560
141.10
168.71
3080
3081
Identifier Count Limit
Parts Done 3081 Infinite
Frequencies
Identifier
STATE (Microetch3_R)
STATE (Cleaner3_R)
STATE (flux3_R)
STATE (solder3_R)
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
Number
3075
3075
9
2251
2250
7
3081
3082
1
3081
3082
1
AvgTime
1.4728
3.9279
136.00
4.7494
2.7503
136.00
.18000
5.5615
136.00
.12600
5.6155
136.00
Percent
25.40
67.74
6.86
59.96
34.70
5.34
3.11
96.13
0.76
2.18
97.06
0.76
Percent
25.40
67.74
6.86
59.96
34.70
5.34
3.11
96.13
0.76
2.18
97.06
0.76
G-10
-------
NAME:
Throughput:
HASL, non-conveyorized
60,OOOK ssf
ARENA Simulation Results
Replication ended at time:
Tally Variables
2876.64 min.
Identifier Average Half Width Minimum Maximum Observations
TaktTime 3.8531
Time in system 89.058
Counters
Identifier Count
Parts Done 711
Frequencies
Identifier
STATE (Microetch3_R)
STATE (Cleaner3_R)
STATE (flux3_R)
STATE (solder3_R)
.69813
(Corr)
Limit
Infinite
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
3.4700
7.9560
Number
577
575
3
3
1
2
711
712
1
711
712
1
139.47
279.95
AvgTime
1.8113
2.4756
136.00
822.39
137.47
136.00
.18000
3.6694
136.00
.12600
3.7233
136.00
710
711
Percent
36.33
49.48
14.18
85.77
4.78
9.46
4.45
90.82
4.73
3.11
92.16
4.73
Percent
36.33
49.48
14.18
85.77
4.78
9.46
4.45
90.82
4.73
3.11
92.16
4.73
G-ll
-------
NAME:
Throughput:
HASL, conveyorized
60,OOOK ssf
ARENA Simulation Results
Replication ended at time:
Tally Variables
2348.82 min.
Identifier
Takt time
Time in system
Counters
Identifier
Depart 33_C
Frequencies
Identifier
STATE (Cleaner_R)
STATE (solder_R)
STATE (flux_R)
STATE (Microetch_R)
Average Half Width Minimum Maximum Observations
.19281 .02704
19.009 (Corr)
Count Limit
10601 Infinite
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
.16654
4.9888
Number
9825
9823
2
10601
10601
1
10601
10601
1
10601
10601
1
136.00
140.82
AvgTime
.00539
.17549
136.00
.00500
.17544
136.00
.00500
.17544
136.00
.00500
.17544
136.00
10600
10601
Percent
2.59
84.14
13.28
2.59
90.77
6.64
2.59
90.77
6.64
2.59
90.77
6.64
Percent
2.59
84.14
13.28
2.59
90.77
6.64
2.59
90.77
6.64
2.59
90.77
6.64
G-12
-------
NAME:
Throughput:
HASL, conveyorized
260,OOOK ssf
ARENA Simulation Results
Replication ended at time: 8908.24 min.
Tally Variables
Identifier Average Half Width Minimum Maximum Observations
Time in system 21.188
Takttime .18000
Counters
Identifier Count
Depart 33_C 45936
Frequencies
Identifier
STATE (Cleaner_R)
STATE (solder_R)
STATE (Microetch_R)
STATE (flux_R)
10.277
(Corr)
Limit
Infinite
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
4.9888
.16654
Number
42056
42051
6
45936
45936
1
45936
45932
6
45936
45937
1
140.91
136.00
AvgTime
.00546
.17506
136.00
.00500
.17506
136.00
.00500
.16027
136.00
.00500
.17506
136.00
45936
45935
Percent
2.73
87.56
9.71
2.73
95.65
1.62
2.73
87.56
9.71
2.73
95.65
1.62
Percent
2.73
87.56
9.71
2.73
95.65
1.62
2.73
87.56
9.71
2.73
95.65
1.62
G-13
-------
NAME:
Throughput:
Nickel/Palladium/Gold, non-conveyorized
260,OOOK ssf
ARENA Simulation Results
Replication ended at time:
Tally Variables
114576.0 min.
Identifier Average Half Width Minimum Maximum Observations
Time in system 116.79
TaktTime 38.848
Counters
Identifier Count
Parts Done 3081
Frequencies
Identifier
STATE (Acid Dip_R)
STATE (Catalyst_R)
STATE (Cleaner_R)
STATE (Electroless Palla
STATE (Immersion Gold_R
STATE (Preinitiator_R)
STATE (Electroless Nicke
STATE (Microetch_R)
1.0484
(Corr)
Limit
Infinite
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
106.86
17.830
Number
3073
3070
4
3075
3070
6
3069
3062
7
3008
2975
34
2803
2798
6
3081
3082
5
2872
2833
40
3064
3056
9
278.21
131.33
AvgTime
1.6342
37.226
113.00
3.7372
35.045
113.00
3.4835
35.362
113.00
4.7321
34.179
113.00
19.598
22.926
113.00
2.3000
36.375
113.00
19.663
20.743
113.00
1.4781
37.373
113.00
308
3080
Percent
4.19
95.43
0.38
9.60
89.84
0.57
8.93
90.41
0.66
11.89
84.91
3.21
45.87
53.56
0.57
5.92
93.61
0.47
47.16
49.07
3.77
3.78
95.37
0.85
Percent
4.19
95.43
0.38
9.60
89.84
0.57
8.93
90.41
0.66
11.89
84.91
3.21
45.87
53.56
0.57
5.92
93.61
0.47
47.16
49.07
3.77
3.78
95.37
0.85
G-14
-------
NAME:
Throughput:
Nickel/Palladium/Gold, non-conveyorized
60,OOOK ssf
ARENA Simulation Results
Replication ended at time: 25807.8 min.
Tally Variables
Identifier Average Half Width Minimum Maximum Observations
Time in system 115.87
TaktTime 38.929
Counters
Identifier Count
Parts Done 711
Frequencies
Identifier
STATE (Acid Dip_R)
STATE (Cleaner_R)
STATE (Catalyst_R)
STATE (Electroless Palla
STATE (Immersion Gold_R
STATE (Preinitiator_R)
STATE (Electroless Nicke
STATE (Microetch_R)
1.7495
(Corr)
Limit
Infinite
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
106.86
17.830
Number
711
712
1
709
707
2
707
706
2
695
688
8
652
651
2
711
711
1
670
663
9
707
706
3
199.39
131.33
AvgTime
1.6300
37.269
113.00
3.4797
35.522
113.00
3.7511
35.311
113.00
4.7263
34.329
113.00
19.443
22.895
113.00
2.3000
36.651
113.00
19.451
20.751
113.00
1.4783
37.427
113.00
711
710
Percent
4.17
95.43
0.41
8.87
90.32
0.81
9.54
89.65
0.81
11.81
84.94
3.25
45.59
53.60
0.81
5.88
93.71
0.41
46.87
49.48
3.66
3.76
95.02
1.22
Percent
4.17
95.43
0.41
8.87
90.32
0.81
9.54
89.65
0.81
11.81
84.94
3.25
45.59
53.60
0.81
5.88
93.71
0.41
46.87
49.48
3.66
3.76
95.02
1.22
G-15
-------
NAME:
Throughput:
Nickel/Gold, non-conveyorized
260,OOOK ssf
ARENA Simulation Results
Replication ended at time: 86437.5 min.
Tally Variables
Identifier Average Half Width Minimum
Maximum
Observations
TaktTime 27.062 1.2220E-14 17.830 134.
Time in system 98.948 2.0602 86.100 286.
Counters
Identifier Count Limit
Parts Done 3081 Infinite
Frequencies
Identifier
STATE (Microetch2_R)
STATE (Acid Dip2_R)
STATE (Electroless Nickel)
STATE (Cleaner2_R)
STATE (Catalyst2_R)
STATE (Immersion Gold2
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
Number
3056
3048
9
3068
3065
4
2448
2409
40
3063
3056
7
3067
3062
6
2966
2961
6
AvgTime
1.4820
25.546
116.00
1.6369
25.432
116.00
23.069
9.2664
116.00
3.4903
23.538
116.00
3.7470
23.268
116.00
18.521
9.3911
116.00
33 3080
16 3081
Percent
5.43
93.32
1.25
6.02
93.42
0.56
67.69
26.75
5.56
12.81
86.21
0.97
13.77
85.39
0.83
65.84
33.33
0.83
Percent
5.43
93.32
1.25
6.02
93.42
0.56
67.69
26.75
5.56
12.81
86.21
0.97
13.77
85.39
0.83
65.84
33.33
0.83
G-16
-------
NAME:
Throughput:
Nickel/Palladium/Gold, non-conveyorized
60,OOOK ssf
ARENA Simulation Results
Replication ended at time:
Tally Variables
19427.7 min.
Identifier Average Half Width Minimum Maximum Observations
TaktTime 27.150
Time in system 95.321
Counters
Identifier Count
Parts Done 711
Frequencies
Identifier
STATE (Electroless Nicke
STATE (Acid Dip2_R)
STATE (Microetch2_R)
STATE (Cleaner2_R)
STATE (Catalyst2_R)
STATE (Immersion Gold2
(Corr)
4.1505
Limit
Infinite
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
17.830
86.100
Number
605
597
9
711
712
1
705
704
3
708
706
2
711
710
2
684
683
2
134.33
193.43
AvgTime
21.541
8.9632
116.00
1.6300
25.495
116.00
1.4825
25.617
116.00
3.4847
23.694
116.00
3.7300
23.300
116.00
18.533
9.5440
116.00
710
711
Percent
67.08
27.54
5.37
5.97
93.44
0.60
5.38
92.83
1.79
12.70
86.11
1.19
13.65
85.16
1.19
65.25
33.55
1.19
Percent
67.08
27.54
5.37
5.97
93.44
0.60
5.38
92.83
1.79
12.70
86.11
1.19
13.65
85.16
1.19
65.25
33.55
1.19
G-17
-------
NAME:
Throughput:
OSP, non-conveyorized
260,OOOK ssf
ARENA Simulation Results
Replication ended at time: 14371.9 min.
Tally Variables
Identifier
Average
Half Width Minimum
Maximum
Observations
Takt Time
Time in System
Counters
4.7599
399.53
.59985
(Corr)
4.6200
21.330
150.67
513.90
3080
3081
Identifier Count Limit
Depart 7_C 3081 Infinite
Frequencies
Identifier
STATE (Cleaner_R)
Category
BUSY
IDLE
FAILED
Number
2301
2294
7
AvgTime
4.6462
1.2850
149.00
Percent
72.82
20.08
7.10
Percent
72.82
20.08
7.10
STATE (Osp_R)
STATE (Microetch_R)
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
3081
3081
1
2711
2703
9
1.6700
3.0469
149.00
1.6706
3.2600
149.00
35.04
63.94
1.01
30.85
60.02
9.13
35.04
63.94
1.01
30.85
60.02
9.13
G-18
-------
NAME:
Throughput:
OSP, non-conveyorized
60,OOOK ssf
ARENA Simulation Results
Replication ended at time: 3731.92 min.
Tally Variables
Identifier
Average
Half Width Minimum
Maximum
Observations
Takt Time
Time in System
Counters
5.0236
172.58
.57885
(Corr)
4.6200
21.330
150.47
322.15
710
711
Identifier Count Limit
Depart 7_C 711 Infinite
Frequencies
Identifier
STATE (Cleaner_R)
Category
BUSY
IDLE
FAILED
Number
581
579
2
AvgTime
4.2464
1.6696
149.00
Percent
66.11
25.90
7.99
Percent
66.11
25.90
7.99
STATE (Osp_R)
STATE (Microetch_R)
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
711
711
1
619
618
3
1.6700
3.3692
149.00
1.6884
3.6241
149.00
31.82
64.19
3.99
28.01
60.02
11.98
31.82
64.19
3.99
28.01
60.02
11.98
G-19
-------
NAME:
Throughput:
OSP, conveyorized
260,OOOK ssf
ARENA Simulation Results
Replication ended at time: 6568.83 min.
Tally Variables
Identifier Average Half Width Minimum
Maximum
Observations
Takt time
Time in system
Counters
.14724
30.442
.01562
14.465
.13961
5.1777
149.00
154.12
45936
45937
Identifier Count Limit
Depart 22_C 45937 Infinite
Frequencies
Identifier
STATE (Microetch2_R)
STATE (Cleaner2_R)
STATE (osp_R)
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
Number
45937
45932
6
40587
40582
6
45937
45937
1
AvgTime
.00500
.12290
149.00
.00566
.13910
149.00
.00500
.13911
149.00
Percent
3.39
83.40
13.21
3.39
83.40
13.21
3.39
94.41
2.20
Percent
3.39
83.40
13.21
3.39
83.40
13.21
3.39
94.41
2.20
G-20
-------
NAME:
Throughput:
OSP, conveyorized
60,OOOK ssf
ARENA Simulation Results
Replication ended at time: 2002.0 min.
Tally Variables
Identifier Average Half Width Minimum
Maximum
Observations
Takt Time
Time in System
Counters
.15805
27.077
.03019
(Corr)
.1356
5.1777
149.00
154.07
1060
10600
Identifier Count Limit
Depart 22_C 10601 Infinite
Frequencies
Identifier
STATE (Microetch2_R)
STATE (Cleaner2_R)
STATE (OSP_R)
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
Number
10601
10601
1
9531
9530
2
10601
10601
1
AvgTime
.00500
.16979
149.00
.00556
.17324
149.00
.00500
.16979
149.00
Percent
2.65
89.91
7.44
2.65
82.47
14.89
2.65
89.91
7.44
Percent
2.65
89.91
7.44
2.65
82.47
14.89
2.65
89.91
7.44
G-21
-------
NAME:
Throughput:
Immersion Silver, conveyorized
60,OOOK ssf
ARENA Simulation Results
Replication ended at time: 5425.08 min.
Tally Variables
Identifier Average Half Width Minimum Maximum Observations
Time in System 14.998
Takttime .51074
Counters
Identifier Count
depart 44_C 10601
Frequencies
Identifier
STATE (Microetch3_R)
STATE (Cleaner3_R)
STATE (Immersion Silver)
STATE (prodip_R)
5.9815
(Corr)
Limit
Infinite
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
11.189
.48953
Number
10601
10601
1
10372
10370
2
10601
10601
1
10601
10600
2
125.07
113.99
AvgTime
.00500
.49600
114.00
.00511
.49605
114.00
.00500
.49600
114.00
.00500
.48529
114.00
10601
10600
Percent
0.98
96.92
2.10
0.98
94.82
4.20
0.98
96.92
2.10
0.98
94.82
4.20
Percent
0.98
96.92
2.10
0.98
94.82
4.20
0.98
96.92
2.10
0.98
94.82
4.20
G-22
-------
NAME:
Throughput:
Immersion Silver, conveyorized
260,OOOK ssf
ARENA Simulation Results
Replication ended at time:
Tally Variables
26206.7 min.
Identifier Average Half Width Minimum Maximum Observations
Time in System 18.921
Takt Time .50495
Counters
Identifier Count
depart 44_C 45937
Frequencies
Identifier
STATE (Microetch3_R)
STATE (Cleaner3_R)
STATE (Immersion Silver)
STATE (prodip_R)
4.1632
(Corr)
Limit
Infinite
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
11.189
.48995
Number
45937
45932
6
44792
44786
6
45937
45937
1
45021
45017
5
238.69
114.03
AvgTime
.00500
.48535
114.00
.00513
.49777
114.00
.00500
.49770
114.00
.00510
.49775
114.00
45937
45936
Percent
0.99
96.06
2.95
0.99
96.06
2.95
0.99
98.52
0.49
0.99
96.55
2.46
Percent
0.99
96.06
2.95
0.99
96.06
2.95
0.99
98.52
0.49
0.99
96.55
2.46
G-23
-------
NAME:
Throughput:
Immersion Tin, non-conveyorized
260,OOOK ssf
ARENA Simulation Results
Replication ended at time: 30669.2 min.
Tally Variables
Identifier Average Half Width Minimum Maximum Observations
TaktTime 9.8516
Time in System 40.215
Counters
Identifier Count
Depart 7_C 3081
Frequencies
Identifier
STATE (Cleaner_R)
STATE (predip_R)
STATE (Immersion Tin R)
STATE (Microetch_R)
(Corr)
4.5278
Limit
Infinite
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
8.5500
26.010
Number
3009
3002
7
3049
3045
5
2003
2003
1
3008
3000
9
93.550
185.18
AvgTime
3.5530
6.3568
85.000
1.1822
8.6500
85.000
13.151
1.9678
85.000
1.5056
8.3583
85.000
3080
3081
Percent
35.20
62.84
1.96
11.87
86.73
1.40
86.74
12.98
0.28
14.91
82.57
2.52
Percent
35.20
62.84
1.96
11.87
86.73
1.40
86.74
12.98
0.28
14.91
82.57
2.52
G-24
-------
NAME:
Throughput:
Immersion Tin, non-conveyorized
60,OOOK ssf
ARENA Simulation Results
Replication ended at time:
Tally Variables
7144.18min.
Identifier Average Half Width Minimum Maximum Observations
TaktTime 9.9108
Time in System 36.380
Counters
Identifier Count
Depart 7_C 711
Frequencies
Identifier
STATE (Cleaner_R)
STATE (Predip_R)
STATE (Immersion Tin R)
STATE (Microetch_R)
.36935
7.8297
Limit
Infinite
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
8.5500
26.010
Number
699
697
2
711
712
1
527
527
1
693
692
3
88.470
104.68
AvgTime
3.5295
6.4663
85.000
1.1700
8.7462
85.000
11.535
1.8598
85.000
1.5081
8.4451
85.000
710
711
Percent
34.53
63.09
2.38
11.64
87.17
1.19
85.09
13.72
1.19
14.63
81.80
3.57
Percent
34.53
63.09
2.38
11.64
87.17
1.19
85.09
13.72
1.19
14.63
81.80
3.57
G-25
-------
NAME:
Throughput:
Immersion Tin, conveyorized
260,OOOK ssf
ARENA Simulation Results
Replication ended at time:
Tally Variables
43501.6 min.
Identifier Average Half Width Minimum Maximum Observations
Takt Time .95367
Time in System 21.375
Counters
Identifier Count
Depart 22_C 45937
Frequencies
Identifier
STATE (Microetch2_R)
STATE (Cleaner2_R)
STATE (Predip_R)
STATE (Immersion Tin R)
(Corr)
(Corr)
Limit
Infinite
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
.93728
12.350
Number
45936
45931
6
45487
45481
6
45576
45572
5
45937
45937
1
85.005
160.23
AvgTime
.00500
.91794
85.000
.00505
.92702
85.000
.00504
.92704
85.000
.00500
.92707
85.000
45936
45937
Percent
0.54
98.28
1.19
0.54
98.28
1.19
0.54
98.47
0.99
0.54
99.27
0.20
Percent
0.54
98.28
1.19
0.54
98.28
1.19
0.54
98.47
0.99
0.54
99.27
0.20
G-26
-------
NAME:
Throughput:
Immersion Tin, conveyorized (Tin h 60)
60,OOOK ssf
ARENA Simulation Results
Replication ended at time:
Tally Variables
10029.78 min.
Identifier Average Half Width Minimum Maximum Observations
Takt Time .95796
Time in Systemn 23.910
Counters
Identifier Count
Depart 22_C 10601
Frequencies
Identifier
STATE (Microetch2_R)
STATE (Cleaner2_R)
STATE (Predip_R)
STATE (Immersion Tin R)
(Corr)
(Corr)
Limit
Infinite
Category
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
BUSY
IDLE
FAILED
.93728
12.364
Number
10601
10601
1
10476
10475
2
10601
10600
2
10601
10601
1
85.260
110.71
AvgTime
.26000
.67102
85.000
.26310
.67098
85.000
.26000
.66307
85.000
.26000
.67102
85.000
10600
10601
Percent
27.69
71.46
0.85
27.69
70.60
1.71
27.69
70.60
1.71
27.69
71.46
0.85
Percent
27.69
71.46
0.85
27.69
70.60
1.71
27.69
70.60
1.71
27.69
71.46
0.85
G-27
-------
G-5 Chemical Costs by Bath for Individual Surface Finish Processes
G-28
-------
Process: Hot Air Solder Leveling (HASL)a
Bath
Cleaner
Microetch
Flux
Volume in Bath
(in gallons)
Horizontal
66.5
86.6
NA
Volume in Bath
(in gallons)
Vertical
51.1
51.1
NA
Supplier
ID
#1
#2
#3
#4
#5
#6
#7
#8
#1
#2
#3
#4
#5
#6
#7
#8
Unit Vol.
Chemical
Cost
$14.4/gal
$5.42/gal
$1.38/gal
$1.13/gal
$2.50/gal
$1.00/gal
$1.02/gal
$2.50/gal
$1.43/gal
$2.14/gal
$0.757/gal
$9.88/gal
$5.20/gal
$5.20/gal
$1.05/gal
$5.20/gal
$12.50/gal
Avg.
Chemical
Cost
$3.67/gal
$3.86/gal
Total Cost of
the Bath
(Horizontal)
$244
$344
$12.50/galb
Total Cost of
the Bath
(Vertical)
$188
$197
$12.50/galb
a No suppliers of HASL were identified. Chemical costs for baths similar to other alternatives were calculated by averaging the individual bath costs from other
alternatives.
b Flux is refilled as it is consumed. The flux cost per gallon was obtained by industry estimate. (Personal communication with Mark Carey, February, 2000.)
G-29
-------
Process: Immersion Silver
Supplier #1
Bath
Cleaner
Microetch
Predip
Immersion Silver
Volume in Bath
(in gallons)
Horizontal
66.5
86.6
46.2
NA
Volume in Bath
(in gallons)
Vertical
No data
No data
No data
No data
Chemical
Name
A
B
C
D
E
F
G
Percentage of
Chemical in
Bath
100
5
0.25
10
100
90
10
Cost of
Chemicals
$14.4/gal
$26.6/gal
$1.20/gal
$1.00/gal
$26.0/gal
$26.0/gal
$75.0/gal
Multiplying
Factor
1
1
1
1
1
1
1
Total Cost
of the Bath
(Horizontal)
$958
$124
$1,200
$30.9/gal"
Total Cost
of the Bath
(Vertical)
No data
No data
No data
No data
a The silver bath is not replaced, but rather maintained as it becomes depleted. The total material cost of the silver bath required to produce 260,000 ssf of PWB
will be calculated directly from the price per gallon of bath solution and the total gallons of bath solution required.
G-30
-------
Process: Immersion Tin
Supplier #2
Bath
Cleaner
Microetch
Predip
Immersion Tin
Volume in Bath
(in gallons)
Horizontal
66.5
86.6
46.2
NA
Volume in Bath
(in gallons)
Vertical
51.1
51.1
51.1
NA
Chemical
Name
A
B
C
D
E
F
G
H
I
Percentage of
Chemical in
Bath
7
10
1.251b/gal
1
0.5
5
200 g/L
10
5
Cost of
Chemicals
$20.0/L
$1.20/gal
$1.70/lb
$1.20/gal
S40.0/L
$1.20/gal
$40.0/kg
S40.0/L
S40.0/L
Multiplying
Factor
1
1
1
1
1
1
2.24
3.48
5.94
Total Cost
of the Bath
(Horizontal)
$360
$185
$34.9
$166/gaP
Total Cost
of the Bath
(Vertical)
$277
$109
$38.7
$166/gaP
a The tin bath is not replaced, but rather maintained as it becomes depleted. The total material cost of the tin bath required to produce 260,000 ssf of PWB will be
calculated directly from the price per gallon of bath solution and the total gallons of bath solution required.
G-31
-------
Process: Immersion Tin
Supplier #3
Bath
Cleaner
Microetch
Predip
Immersion
Tin
Volume in Bath
(in gallons)
Horizontal
66.5
86.6
46.2
NA
Volume in Bath
(in gallons)
Vertical
51.1
51.1
51.1
NA
Chemical
Name
A
B
C
D
E
Percentage of
Chemical in
Bath
12.5
60g/L
1
25
100
Cost of
Chemicals
$11.0/gal
$1.49/lb
$1.20/gal
$100/gal
$100/gal
Multiplying
Factor
1
1
1
1
1
Total Cost
of the Bath
(Horizontal)
$91.4
$65.6
$1,160
SlOO/gal"
Total Cost
of the Bath
(Vertical)
$70.3
$38.7
$1,280
SlOO/gal"
a The tin bath is not replaced, but rather maintained as it becomes depleted. The total material cost of the tin bath required to produce 260,000 ssf of PWB will be
calculated directly from the price per gallon of bath solution and the total gallons of bath solution required.
G-32
-------
Process: Electroless Nickel/Immersion Gold
Supplier #4
Bath
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Immersion Gold
Volume in
Bath (in
gallons)
Horizontal
No data
No data
No data
No data
No data
No data
Volume in
Bath (in
gallons)
Vertical
51.1
51.1
51.1
51.1
51.1
51.1
Chemical
Name
A
B
C
D
E
F
G
H
J
K
L
Percentage of
Chemical in
Bath
15
1.881b/gal
1
10
17
40
5
15
5
0.250 unit/gal
(225 mL/gal)
8 oz/gal
Cost of
Chemicals
$7.50/gal
$5.25/lb
$1.20/gal
$40.0/gal
S8.00/L
S8.00/L
$14.5/gal
$20.0/gal
$23.0/gal
$344/unit
$3.25/lb
Multiplying
Factor
1
1
1
1
1
1
5
1
4
1
1
Total Cost
of the Bath
(Horizontal)
No data
No data
No data
No data
No data
No data
Total Cost
of the Bath
(Vertical)
$57.5
$505
$467
$619
$574
$58,500a
a Immersion gold replacement cost was calculated differently than other baths because of the wide disparity in costs and throughput between product lines. The
overall cost for the gold bath was calculated for each product line and then averaged together to give the gold cost for the process.
G-33
-------
Process: Electroless Nickel/Immersion Gold
Supplier #5
Bath
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Immersion Gold
Volume in
Bath (in
gallons)
Horizontal
No data
No data
No data
No data
No data
No data
Volume in
Bath (in
gallons)
Vertical
51.1
51.1
51.1
51.1
51.1
51.1
Chemical
Name
A
B
C
D
E
F
G
H
I
J
K
L
M
N
Percentage of
Chemical in
Bath
10
3
3
45g/L
8.5
30
20
12
2g/L
6.6
15
6.6
50
3g/L
Cost of
Chemicals
$25.0/gal
$5.66/gal
$9.39/gal
$27.3/kg
$1.20/gal
$127/gal
$54.0/gal
$51.0/gal
$29. I/kg
$24.1/gal
$30.9/gal
$28.4/gal
$21.4/gal
$40.0/g
Multiplying
Factor
1
1
1
1
1
1
1
1
1
6
6
5
1
3
Total Cost
of the Bath
(Horizontal)
No data
No data
No data
No data
No data
No data
Total Cost
of the Bath
(Vertical)
$128
$266
$2,810
$11.3
$2,390
$57,350a
a Immersion gold replacement cost was calculated differently than other baths because of the wide disparity in costs and throughput between product lines. The
overall cost for the gold bath was calculated for each product line and then averaged together to give the gold cost for the process.
G-34
-------
Process: OSP
Supplier #6
Bath
Cleaner
Microetch
OSP
Volume in Bath
(in gallons)
Horizontal
66.5
86.6
NA
Volume in Bath
(in gallons)
Vertical
51.1
51.1
NA
Chemical
Name
A
B
C
D
E
F
G
Percentage of
Chemical in
Bath
10
3
3
45.0g/L
8.5
6
23
Cost of
Chemicals
$10.0/gal
$5.66/gal
$9.39/gal
$27.3/kg
$1.20/gal
$324/gal
$321/gal
Multiplying
Factor
1
1
1
1
1
1
1
Total Cost
of the Bath
(Horizontal)
$66.5
$451
$93.6/gal"
Total Cost
of the Bath
(Vertical)
$51.1
$261
$93.6/gal"
a The OSP bath is not replaced, but rather maintained as it becomes depleted. The total material cost of the OSP bath required to produce 260,000 ssf of PWB
will
be calculated directly from the price per gallon of bath solution and the total gallons of bath solution required.
G-35
-------
Process: OSP
Supplier #7
Bath
Cleaner
Microetch
OSP
Volume in Bath
(in gallons)
Horizontal
66.5
86.6
NA
Volume in Bath
(in gallons)
Vertical
51.1
51.1
NA
Chemical
Name
A
B
C
D
E
Percentage of
Chemical in
Bath
10
2.5
7
18.5
100
Cost of
Chemicals
$10.2/gal
$7.62/gal
$9.12/gal
$1.20/gal
$117/gal
Multiplying
Factor
1
1
1
1
1
Total Cost
of the Bath
(Horizontal)
$67.8
$91.0
$117/gaP
Total Cost
of the Bath
(Vertical)
$52.1
$53.7
$117/gaP
a The OSP bath is not replaced, but rather maintained as it becomes depleted. The total material cost of the OSP bath required to produce 260,000 ssf of PWB
will
be calculated directly from the price per gallon of bath solution and the total gallons of bath solution required.
G-36
-------
Process: Electroless Nickel/Electroless Palladium/Immersion Gold
Supplier #8
Bath
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Preinitiator
Electroless Palladium
Immersion Gold
Volume in
Bath (in
gallons)
Horizontal
No data
No data
No data
No data
No data
No data
No data
No data
Volume in
Bath (in
gallons)
Vertical
51.1
51.1
51.1
51.1
51.1
51.1
51.1
NA
Chemical
Name
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Q
R
S
T
U
Percentage
of Chemical
in Bath
10
3
3
45g/L
8.5
30
20
12
2g/L
6.6
15
6.6
20
10
1.4
2.5
20
2.5
0.05
50
3g/L
Cost of
Chemicals
$25.0/gal
$5.66/gal
$9.39/gal
$27.3/kg
$1.20/gal
$127/gal
$54.0/gal
$51.0/gal
$29. I/kg
$24.1/gal
$30.9/gal
$28.4/gal
$160/gal
$152/gal
$8.00/L
$943/gal
$23.8/gal
$48.2/gal
$13.3/gal
$21.4/gal
$40.0/g
Multiplying
Factor
1
1
1
1
1
1
1
1
1
6
6
5
1
1
1
3
1
2
3
1
3
Total Cost
of the Bath
(Horizontal)
No data
No data
No data
No data
No data
No data
No data
No data
Total Cost
of the Bath
(Vertical)
$128
$266
$2,810
$11.3
$2,390
$2,430
$3,980
$57,900a
a Immersion gold replacement cost was calculated differently than other baths because of the wide disparity in costs and throughput between product lines. The
overall cost for the gold bath was calculated for each product line and then averaged together to give the gold cost for the process.
G-37
-------
G-6 Total Materials Cost for Surface Finishing Processes
Process: HASL, non-conveyorized
Throughput: 260K ssf of PWB
Bath
Cleaner
Microetch
Flux
Solder
Chemical Cost/Bath
Replacement '
$188
$197
$16,250 c
$55,460 d
Number of Bath
Replacements "
7
9
1
1
Total Materials Cost
Total Chemical Cost
$1,320
$1,770
$16,250
$55,460
$74,800
" Reported chemical cost per bath replacement reflects the average bath cost of all processes submitted for evaluation in this
surface finishing category.
b Number of bath replacements required to process 260,000 ssf as determined by process simulation.
1 Flux bath is not replaced, but rather refilled as flux is consumed. Cost of flux was calculated at $12.50/gal and is consumed at
200 ssf/gal.
d Solder is not replaced, but rather refilled as solder is consumed. Cost of solder was calculated using a solder cost of $2.57/lb
and an average solder consumption rate, including solder wastage, of 0.083 Ib/ssf which was obtained from three PWB facilities.
Process: HASL, conveyorized
Throughput: 260K ssf of PWB
Bath
Cleaner
Microetch
Flux
Solder
Chemical Cost/Bath
Replacement '
$244
$344
$16,250 c
$55,460 d
Number of Bath
Replacements "
6
6
1
1
Total Materials Cost
Total Chemical Cost
$1,460
$2,060
$16,250
$55,460
$75,200
" Reported chemical cost per bath replacement reflects the average bath cost of all processes submitted for evaluation in this
surface finishing category.
b Number of bath replacements required to process 260,000 ssf as determined by process simulation.
1 Flux bath is not replaced, but rather refilled as flux is consumed. Cost of flux was calculated at $12.50/gal and is consumed at
200 ssf/gal.
d Solder is not replaced, but rather refilled as solder is consumed. Cost of solder was calculated using a solder cost of $2.57/lb
and an average solder consumption rate, including solder wastage, of 0.083 Ib/ssf which was obtained from three PWB facilities.
G-38
-------
Process: Electroless Nickel/Immersion Gold, non-conveyorized
Throughput: 260K ssf of PWB
Bath
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Immersion Gold
Chemical Cost/Bath
Replacement '
$92.8
$386
$1,640
$315
$890
NAC
Number of Bath
Replacements "
7
9
6
4
40
6
Total Materials Cost
Total Chemical Cost
$649
$3,470
$9,830
$1,260
$35,500
$57,900
$108,600
" Reported chemical cost per bath replacement reflects the average bath cost of all processes submitted for evaluation in this
surface finishing category.
b Number of bath replacements required to process 260,000 ssf as determined by process simulation.
1 Immersion gold replacement cost was calculated differently than other baths because of the wide disparity in costs and
throughput between product lines. The overall cost for the gold bath was calculated for each product line and then averaged
together to give the gold cost for the process.
Process: Electroless Nickel/Electroless Palladium/Immersion Gold, non-conveyorized
Throughput: 260K ssf of PWB
Bath
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Preinitiator
Electroless Palladium
Immersion Gold
Chemical Cost/Bath
Replacement '
$128
$266
$2,810
$11.3
$2,390
$2,430
$3,980
NAC
Number of Bath
Replacements "
7
9
6
4
40
5
34
6
Total Materials Cost
Total Chemical Cost
$900
$2,390
$16,860
$45
$95,600
$12,150
$135,300
$57,900
$321,000
" Reported chemical cost per bath replacement reflects the average bath cost of all processes submitted for evaluation in this
surface finishing category.
b Number of bath replacements required to process 260,000 ssf as determined by process simulation.
1 Immersion gold replacement cost was calculated differently than other baths because of the wide disparity in costs and
throughput between product lines. The overall cost for the gold bath was calculated for each product line and then averaged
together to give the gold cost for the process.
G-39
-------
Process: OSP, non-conveyorized
Throughput: 260K ssf of PWB
Bath
Cleaner
Microetch
OSP
Chemical Cost/Bath
Replacement '
$51.6
$157
$16,750 c
Number of Bath
Replacements "
7
9
1
Total Materials Cost
Total Chemical Cost
$361
$1,420
$16,750
$18,500
" Reported chemical cost per bath replacement reflects the average bath cost of all processes submitted for evaluation in this
surface finishing category.
b Number of bath replacements required to process 260,000 ssf as determined by process simulation.
1 OSP bath is not replaced, but rather refilled as the OSP is consumed. Cost of OSP was calculated at $105/gal and is consumed
atl,630ssf/gal.
Process: OSP, conveyorized
Throughput: 260K ssf of PWB
Bath
Cleaner
Microetch
OSP
Chemical Cost/Bath
Replacement '
$67.2
$271
$16,750 c
Number of Bath
Replacements "
6
6
1
Total Materials Cost
Total Chemical Cost
$403
$1,630
$16,800
$18,800
" Reported chemical cost per bath replacement reflects the average bath cost of all processes submitted for evaluation in this
surface finishing category.
b Number of bath replacements required to process 260,000 ssf as determined by process simulation.
1 OSP bath is not replaced, but rather refilled as the OSP is consumed. Cost of OSP was calculated at $105/gal and is consumed
atl,630ssf/gal.
G-40
-------
Process: Immersion Silver, conveyorized
Throughput: 260K ssf of PWB
Bath
Cleaner
Microetch
Predip
Immersion Silver
Chemical Cost/Bath
Replacement '
$958
$124
$1,200
$40,170 c
Number of Bath
Replacements "
6
6
5
1
Total Materials Cost
Total Chemical Cost
$5,750
$744
$6,000
$40,200
$52,700
" Reported chemical cost per bath replacement reflects the average bath cost of all processes submitted for evaluation in this
surface finishing category.
' Number of bath replacements required to process 260,000 ssf as determined by process simulation.
1 Silver bath is not replaced, but rather maintained as the silver bath is depleted. The cost of the silver bath was calculated at
$30.9/gal and is consumed at 200 ssf/gal.
Process: Immersion Tin, non-conveyorized
Throughput: 260K ssf of PWB
Bath
Cleaner
Microetch
Predip
Immersion Tin
Chemical Cost/Bath
Replacement '
$174
$74
$659
$23,850 c
Number of Bath
Replacements "
7
9
5
1
Total Materials Cost
Total Chemical Cost
$1,220
$665
$3,300
$23,850
$29,000
" Reported chemical cost per bath replacement reflects the average bath cost of all processes submitted for evaluation in this
surface finishing category.
b Number of bath replacements required to process 260,000 ssf as determined by process simulation.
1 Tin bath is not replaced, but rather maintained as the tin bath is depleted. The cost of the tin bath was calculated at $133/gal and
is consumed at 1,450 ssf/gal.
Process: Immersion Tin, conveyorized
Throughput: 260K ssf of PWB
Bath
Cleaner
Microetch
Predip
Immersion Tin
Chemical Cost/Bath
Replacement '
$226
$125
$597
$23,850 c
Number of Bath
Replacements "
6
6
5
1
Total Materials Cost
Total Chemical Cost
$1,350
$752
$2.990
$23,850
$28,900
" Reported chemical cost per bath replacement reflects the average bath cost of all processes submitted for evaluation in this
surface finishing category.
b Number of bath replacements required to process 260,000 ssf as determined by process simulation.
1 Tin bath is not replaced, but rather maintained as the tin bath is depleted. The cost of the tin bath was calculated at $133/gal and
is consumed at 1,450 ssf/gal.
G-41
-------
Appendix H
Environmental Hazard Assessment and Ecological
Risk Assessment Methodology
-------
H-l. HAZARD PROFILE
The environmental hazard assessment of chemicals consists of the identification of the
effects that a chemical may have on organisms in the environment. An overview of this
assessment process has been reported by, for example, Smrchek and Zeeman (1998) and by
Zeeman and Gilford (1993). The effects are expressed in terms of the acute and chronic toxicity
of a chemical on the exposed organisms. These are generally given as either the lethal
concentration (LC) or as the effective concentration (EC) that describe the type and seriousness
of the effect for a known concentration of a chemical. When the effective concentrations for a
range of species for a chemical are tabulated, the tabulation is called a hazard profile or toxicity
profile. A more detailed discussion of a comprehensive hazard profile has been presented by
Nabholz (1991). The most frequently used hazard profile for the aquatic environment consists of
a set of six effective concentrations as reported by Nabholz et al. (1993a). These are:
• Fish acute value (usually a fish 96-hour LC50 value)
• Aquatic invertebrate acute value (usually a daphnid 48-hour LC50 value)
• Green algal toxicity value (usually an algal 96-hour EC50 value)
• Fish chronic value (usually a fish 28-day chronic value [ChV])
• Aquatic invertebrate chronic value (usually a daphnid 21-day ChV)
• Algal chronic value (usually an algal 96-hour NEC or GMATC value for biomass)
For the acute values, the LC50 (lethality or mortality) (EC50) (non-lethal/lethal effects)
refers to the concentration that results in 50 percent of the test organisms affected at the end of
the specified exposure period in a toxicity test. The chronic values represent the concentration of
the chemical that results in no statistically significant sublethal effects on the test organism
following an extended or chronic exposure.
The hazard profile can be constructed using effective concentrations based on toxicity test
data (with measured test chemical concentrations) or estimated toxicity values based on structure
activity relationships (SARs). The measured values are preferred because they are based on
actual test data, but in the absence of test data SAR estimates, if available for the chemical class,
can be used. Thus the hazard profile may consist of only measured data, only predicted values,
or a combination of both. Also, the amount of data in the hazard profile may range from a
minimum of one acute or chronic value to the full compliment of three acute values and three
chronic values.
In the absence of measured toxicity values, estimates of these values can be made using
SARs. SAR methods include quantitative structure activity relationships (QSARs), qualitative
SARs, or use of the chemical analogs. The use of SARs by OPPT has been described (Clements,
1988; Clements, 1994). The use and application of QSARs specifically for the hazard assessment
of TSCA new chemicals has been presented (Clements et al., 1993a). The development,
validation, and application of SARs in OPPT have been presented by OPPT staff (Zeeman et al.,
1993b; Boethling, 1993; Clements et al., 1993b; Nabholz et al., 1993b; Newsome et al., 1993 and
Lipnick, 1993).
H-l
-------
The predictive equations (QSARs) are used in lieu of actual test data to estimate a toxicity
value for aquatic organisms within a specific chemical class. A total of 140 have been listed
(Clement et al., 1995; Smrchek and Zeeman, 1998). Although the equations are derived from
correlation and linear regression analysis based on measured data, the confidence intervals
associated with the equation are not used to provide a range of toxicity values. Even with
measured test data, the use of the confidence limits to determine the range of values is not used.
H-2. DETERMINATION OF CONCERN CONCENTRATION
Upon completion of a hazard profile, a concern concentration (CC) is determined. A
concern concentration is that concentration of a chemical in the aquatic environment, which, if
exceeded, may cause a significant risk to aquatic organisms. Conversely, if the CC is not
exceeded, the assumption is made that probability of a significant risk occurring is low and no
regulatory action is required. The CC for each chemical is determined by applying assessment
factors (AsF) (U.S. EPA, 1984) or uncertainty factors (UF) (Smrchek et al., 1993) to the effect
concentrations in the hazard profile.
These factors incorporate the concept of the uncertainty associated with: 1) toxicity data,
laboratory tests versus field tests, and measured versus estimated data; and 2) species sensitivity.
For example, if only a single LC50 value for a single species is available, there are several
uncertainties to consider. First, how reliable is the value itself? If the test were to be done again
by the same laboratory or a different laboratory, would the value differ and, if so, by how much?
Second, there are differences in sensitivity (toxicity) among and between species that have to be
considered. If the species tested the most or the least sensitive? In general, if only a single
toxicity value is available, there is a large uncertainty about the applicability of this value to other
organisms in the environment and a large assessment factor, i.e., 1000, is applied to cover the
breadth of sensitivity known to exist among and between organisms in the environment.
Conversely, the more information that is available results in more certainty concerning the
toxicity values and requires the use of smaller factors. For example, if toxicity values are derived
from field tests, then an assessment factor of 1 is used because tests measure chemical effects on
field organisms.
Four factors are used by OPPT to set a CC for chronic risk: 1, 10, 100, and 1000. The
factor used is dependent on the amount and type of toxicity data contained in the hazard profile
and reflects the amount of uncertainty about the potential effects associated with a toxicity value.
In general, the more complete the hazard profile and the higher the quality of the generated
toxicity data, the smaller a factor that is used. The following discussion describes the use and
application of uncertainty or assessment factors.
1. If the hazard profile only contains one or two acute toxicity values, the concern
concentration is set at 1/1000 of the acute value.
2. If the hazard profile contains three acute values (called the base set), the concern
concentration is set at 1/100 of the lowest acute value.
H-2
-------
3. If the hazard profile contains one chronic value, the concern concentration is set at 1/10 of
the chronic value if the value is for the most sensitive species. Otherwise, it is 1/100 of the
acute value for the most sensitive species.
4. If the hazard profile contains three chronic values, the concern concentration is set at 1/10
of the lowest chronic value.
5. If the hazard profile contains a measured chronic value from a field study, then an
assessment factor of 1 is used.
H-3. HAZARD RANKING
Chemicals can be also ranked by their hazard concern levels for the aquatic environment.
This ranking can be based upon the acute toxicity values expressed in milligrams per liter (mg/L).
The generally accepted scoring used by OPPT is as follows (Smrchek et al., 1993; Wagner et al.,
1995):
High Concern (H) < 1
Moderate (or Medium) Concern (M) > 1 and < 100
Low Concern (L) > 100
This ranking can also be expressed in terms of chronic values as follows:
High Concern (H) <0.1
Moderate (or Medium) Concern (M) > 0.1 and < 10.0
Low Concern (L) > 10.0
Chronic toxicity ranking takes precedent over the acute ranking.
H-3
-------
REFERENCES
Boethling, R. S. 1993. Structure Activity Relationships for Evaluation of Biodegradability in the
EPA's Office of Pollution Prevention and Toxics. Environmental Toxicology and Risk
Assessment, 2nd Volume. ASTM STP 1216. J.W. Gorsuch, F. James Dwyer, Christopher G.
Ingersoll, and Thomas W. La Point (Eds.). American Society for Testing and Materials,
Philadelphia, pp. 540-554.
Clements, R.G. (Ed.) 1988. Estimating Toxicity of Industrial Chemicals to Aquatic Organisms
Using Structure-Activity Relationships. U.S. Environmental Protection Agency, Office of
Pollution Prevention and Toxics, Health and Environmental Review Division. Environmental
Effects Branch. Washington, DC. EPA-560/6-88/001, NTIS #PB89-117592.
Clements, R.G. (Ed.) 1994. "Estimating Toxicity of Industrial Chemicals to Aquatic Organisms
Using Structure Activity Relationships," 2nd Ed. EPA-748-R-93-001. Environmental Effects
Branch, Health and Environmental Review Division (7403). Office of Pollution Prevention and
Toxics, U.S. EPA, Washington, DC. PB94-108206, National Technical Information Services
(NTIS), U.S. Department of Commerce, Springfield, VA 22161.
Clements, R.G., J.V. Nabholz, D.W. Johnson, and M. Zeeman. 1993a. "The Use and
Application of QSAR's in the Office of Toxic Substances for Ecological Hazard Assessment of
New Chemicals." Environmental Toxicology and Risk Assessment. ASTM STP 1179. Wayne
G. Landis, Jane S. Hughes, and Michael A. Lewis (Eds.). American Society for Testing and
Materials, Philadelphia, pp. 56-64.
Clements, R.G., J.V. Nabholz, D.W. Johnson, and M. Zeeman. 1993b. "The Use of Quantitative
Structure-Activity Relationships (QSARs) as Screening Tools in Environmental Assessment."
Environmental Toxicology and Risk Assessment, 2nd Volume. ASTM STP 1216. J.W. Gorsuch,
F. James Dwyer, Christopher G. Ingersoll, and Thomas W. La Point (Eds.). American Society
for Testing and Materials, Philadelphia, pp. 555-570.
Clements, R.G., J.V. Nabholz, M.G. Zeeman, and C. Auer. 1995. "The Relationship of
Structure-Activity Relationships (SARs) in the Aquatic Toxicity Evaluation of Discrete Organic
Chemicals. SAR and QSAR in Environmental Research 3:203-215.
Lipnick, R.L. 1993. "Baseline Toxicity QSAR Models: A Means to Assess Mechanism of
Toxicity for Aquatic Organism and Mammals." Environmental Toxicology and Risk
Assessment, 2nd Volume. ASTM STP 1216. J.W. Gorsuch, F. James Dwyer, Christopher G.
Ingersoll, and Thomas W. La Point (Eds.). American Society for Testing and Materials,
Philadelphia, pp. 610-619.
Nabholz, J.V. 1991. "Environmental Hazard and Risk Assessment Under the United States
Toxic Substances Control Act." The Science of the Total Environment 109/110:649-665.
H-4
-------
Nabholz, J.V., P. Miller, and M. Zeeman. 1993a. "Environmental Risk Assessment of New
Chemicals Under the Toxic Substances Control Act (TSCA) Section Five." Environmental
Toxicology and Risk Assessment. ASTM STP 1179. Wayne G. Landis, Jane S. Hughes, and
Michael A. Lewis (Eds.). American Society for Testing and Materials, Philadelphia, pp. 40-55.
Nabholz, J.V., R.G. Clements, M.G. Zeeman, K.C. Osborn, and R. Wedge. 1993b. "Validation
of Structure Activity Relationships Used by the U.S. EPA's Office of Pollution Prevention and
Toxics for the Environmental Hazard Assessment of Industrial Chemicals." Environmental
Toxicology and Risk Assessment, 2nd Volume. ASTM STP 1216. J.S. Gorsuch, F. James
Dwyer, Christopher G. Ingersoll, and Thomas W. La Point (Eds.). American Society for Testing
and Materials, Philadelphia, pp. 571-590.
Newsome, L.D., D.E. Johnson, and J.V. Nabholz. 1993. "Quantitative Structure-Activity
Predictions for Amine Toxicity to Algae and Daphnids." Environmental Toxicology and Risk
Assessment, 2nd Volume. ASTM STP 1216. J.W. Gorsuch,. F. James Dwyer, Christopher G.
Ingersoll, and Thomas W. La Point (Eds.). American Society for Testing and Materials,
Philadelphia, pp. 591-609.
Smrchek, J.C. and M.G. Zeeman. 1998. "Assessing Risks to Ecological Systems from
Chemicals." Handbook of Environmental Risk Assessment and Management. P. Calow (Ed.),
Blackwell Science Ltd, Oxford, UK, pp. 24-90.
Smrchek, J.C., R. Clements, R. Morcock, and W. Robert. 1993. "Assessing Ecological Hazards
under TSCA: Methods and Evaluation of Data." Environmental Toxicology and Risk
Assessment. ASTM STP 1179. Wayne G. Landis, Jane S. Hughes, and Michael A. Landis (Eds.).
American Society for Testing and Materials, Philadelphia, pp 22-39.
U.S. EPA (Environmental Protection Agency). 1984. "Estimating Concern Levels for
Concentrations of Chemical Substances in the Environment." Environmental Effects Branch,
Health and Environmental Review Division (7403), Office of Pollution Prevention and Toxics,
U.S. EPA, Washington, DC.
Wagner, P.M., J.V. Nabholz, and R.J. Kent. 1995. "The New Chemicals Process at the
Environmental Protection Agency (EPA): Structure-Activity Relationships for Hazard
Identification and Risk Assessment." Toxicology Letters 79:67-73.
Zeeman, M.G. and James Gilford. 1993. "Ecological Hazard Evaluation and Risk Assessment
Under EPA's Toxic Substances Control Act (TSCA): An Introduction." Environmental
Toxicology and Risk Assessment. ASTM STP 1179. Wayne G. Landis, Jane S. Hughes, and
Michael A. Lewis (Eds.). American Society for Testing and Materials, Philadelphia, pp. 7-21.
Zeeman, M.G., J.V. Nabholz, and R.G. Clements. 1993. "The Development of SAR/QSAR for
Use Under EPA's Toxic Substances Control Act (TSCA): An Introduction." Environmental
Toxicology and Risk Assessment, 2nd Volume. ASTM STP 1216. J.W. Gorsuch,. F. James
Dwyer, Christopher G. Intersoll, and Thomas W. La Point (Eds.). American Society for Testing
and Materials, Philadelphia, pp. 523-539.
H-5
------- |