=->. <•--ป._ *.".>,*.-.
          Assess mBnl

-------

-------
 Printed Wiring Board
Surface  Finishes
                     Cleaner
                     Technologies
                     Substitutes
                     Assessment
   VOLUME1
  Jack R. Geibig, Senior Research Associate
  Mary B. Swanson, Research Scientist
  and the
  PWB Engineering Support Team
U5.EPA*
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.
                                UT

-------

-------
                        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 assistant in Industrial Engineering:
Aamer Ammer.
                                    Disclaimer
       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/
                                           u

-------
                                Acknowledgments
       This Cleaner Technologies Substitutes .Assessment (CTS A) was prepared 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 and Claire VanRiper-Geibig, who were the
document production managers.

       Valuable contributions to the CTSA were produced by the project's Core Group
members, including: Kathy Hart, Project Lead and Core Group Co-Chair; Holly Evans,
formerly of IPC, and Fern Abrams, IPC, Core Group Co-Chairs; Dipti Singh, Technical Lead
and Technical Workgroup Co-Chair; 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 then-
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.
                                          111

-------
                      EPA Risk Management Workgroup
      We would like to express appreciation to the EPA Risk Management 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
1309 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, RI02910
(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 Alteon Inc.)

Sanmina Corp. (formerly Hadco Corp.)
                  }
Solder Station One Inc.

-------
                               Table of Contents
Executive Summary	ES-1
Chapter 1
Introduction	.	1-1
1.1    Project Background	1-2
       1.1.1  EPA DfE Program	1-2
       1.1.2  DfEPWB Project	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    GTS A 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 Sequences 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
                                          VI

-------
Chapters
Risk Screening and Comparison	               3_j
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	t       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	3.49
       3.2.3   Exposure-Point Concentrations	     3.43
       3.2.4   Estimating Potential Dose Rates	  3.55
       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	3.95
       3.3.4   Summary	3-103
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
              Hot Air Solder Leveling (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
                                         Vll

-------
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-g
       4.1.5  Analysis of the Test Results	4-8
       4.1.6  Overview of Test Results	 4-13
       4.1.7  High Current Low Voltage (HCLV) Circuitry Performance Results	 4-18
       4.1.8  High Voltage Low Current (HVLC) Circuitry Performance Results 	4-20
       4.1.9  High Speed Digital (HSD) Circuitry Performance Results	4-21
       4.1.10 High Frequency Low Pass Filter (HP LPF) Circuitry
             Performance Results	4-22
       4.1.11 High Frequency Transmission Line Coupler (HF TLC) 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 	4-41
       4.1.16 Boxplot Displays	4-43
4.2    Cost Analysis	4-56
       4.2.1  Overview of the Cost Methodology	4-57
       4.2.2  Cost Categories and Discussion of Unquantifiable Costs	4-58
       4.2.3  Simulation Modeling of Surface Finishing Processes	4-63
       4.2.4  Activity-Based Costing	4-70
       4.2.5  Cost Formulation Details and Sample Calculations 	4-73
       4.2.6  Results	4-86
       4.2.7  Conclusions	4-88
4.3    Regulatory Assessment	4-90
       4.3.1  Clean Water Act		4-90
       4.3.2  Clean Air Act	4-95
       4.3.3  Resource Conservation and Recovery Act  	4-97
       4.3.4  Comprehensive Environmental Response, Compensation and
             Liability Act	4-99
       4.3.5  Superfund Amendments and Reauthorization Act and
             Emergency Planning and Community Right-To-Know Act	4-100
       4.3.6  Toxic Substances Control Act	4-102
       4.3.7  Summary of Regulations for Surface Finishing Technologies  	4-103
References	4-110
                                         Vlll

-------
 Chapters           •    l_     '        '
 Conservation  ..'.	_                 5_j
 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	5-19
       5.2.3   Energy Consumption in Other Life-Cycle Stages	 5-2l
       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


 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_g
       7.1.3  Resource Conservation Summary	7_14
 7.2    Social Benefits/Costs Assessment	7-16
       7.211  Introduction to Social Benefits/Costs Assessment  	7-16
       7.2.2  Benefits/Costs Methodology and Data Availability  	7-18
       7.2.3  Social Benefits/Costs Associated with Choice of Surface
             Finishing Alternative	7_19
       7.2.4  Summary of Benefits and Costs	7_29
 7.3    Technology Summary Profiles  	7_31
       7.3.1  HASL Technology		7.31
       7.3.2  Nickel/Gold Technology	.	7.36
       7.3.3  Nickel/Palladium/Gold Technology	 7.40
       7.3.4  OSP Technology	' 7.45
       7.3.5  Immersion Silver Technology	7.49
       7.3.6  Immersion Tin Technology	7.53
References	7.53
                                          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 Ambient Population Exposure Pathways	3-42
Table 3-8.    Bath Surface Areas for Conveyorized Process Baths	3-44
Table 3-9.    Results of Workplace Air Modeling	3-47
Table 3-10.   Results of Ambient Air Modeling	3-51
Table 3-11.   Estimated Releases to Surface Water Following Treatment	3-54
Table 3-12.   Parameter Values for Workplace Inhalation Exposures  	3-57
Table 3-13.   General Parameter Values for Workplace Dermal Exposures	3-59
Table 3-14.   Parameter Values for Workplace Dermal Exposures for Line Operators
             on Non-Conveyorized Lines	3-60
Table 3-15.   Parameter Values for Workplace Dermal Exposure for Line Operators
             on Conveyorized Lines 	3-61
Table 3-16.   Parameter Values for Workplace Dermal Exposure for a Laboratory
             Technician on Either Conveyorized or Non-Conveyorized Lines	3-62
Table 3-17.   Estimated Average Daily Dose for Workplace Exposure from
             Inhalation and Dermal Contact	3-63
Table 3-18.   Estimated Concentration of Lead in Adult and Fetal Blood from
             Incidental Ingestion of Lead in Tin/Lead Solder	3-72
Table 3-19.   Parameter Values for Estimating Nearby Residential Inhalation Exposure ..  3-74
Table 3-20.   Estimated Average Daily Dose for General Population Inhalation
             Exposure	3-74
Table 3-21.   Children's Blood-Lead Results from the ffiUBK Model at Various Lead
             Air Concentrations	  3-75
Table 3-22.   Available Carcinogenicity Information	3-81
Table 3-23.   Summary of RfC and RfD Information Used in Risk Characterization for
             Non-Proprietary Ingredients	3-84
Table 3-24.   NOAEL/LOAEL Values Used in Risk Characterization for
             Non-Proprietary Ingredients	3-87
Table 3-25.   Developmental Toxicity Values Used in Risk Characterization for
             Non-Proprietary Ingredients	  3-88
Table 3-26.   Summary of Health Effects Information	3-89
Table 3-27.   Overview of Available Toxicity Data	3-93
Table 3-28.   Estimated (Lowest) Aquatic Toxicity Values and Concern Concentrations
             for PWB Surface Finishing Chemicals, Based on Measured Test Data or
             SAR Analysis	3-97
Table 3-29.   Environmental Hazard Ranking of PWB Finishing Chemicals	3-100

-------
 Table 3-30.   Gastrointestinal (GI) Absorption Factors  	3-112
 Table 3-31.   Summary of Human Health Risks from Occupational Inhalation
             Exposure for Selected Chemicals		3-116
 Table 3-32.   Summary of Human Health Risks Results from Occupational Dermal
             Exposure for Selected Chemicals 	3-118
 Table 3-33.   Summary of Potential Human Health Effects for Chemicals of Concern ... 3-121
 Table 3-34.   Data Gaps for Chronic Non-Cancer Health Effects for Workers  	3-122
 Table 3-35.   Risk Evaluation Summary for Lead	3-126
 Table 3-36.   Summary of Aquatic Risk Indicators for Non-Metal Chemicals
             of Concern	      3-129
 Table 3-37.   Summary of Aquatic Risk Indicators for Metals Assuming No On-Site
             Treatment	3-130
 Table 3-38.   Overall Comparison of Potential Human Health and Ecological Risks
             for the Non-Conveyorized HASL and Alternative Processes	3-136
 Table 3-39.   Flammable, Combustible, Explosive,  and Fire Hazard Possibilities for
             Surface Finishing Processes	 3-139
 Table 3-40.   Corrosive, Oxidizer, Reactive, Unstable, and Sudden Release of
             Pressure Possibilities for Surface Finishing Processes	3-141
 Table 3-41.   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 Tune			4-13
 Table 4-5.    Percentage of Circuits Meeting Acceptance Criteria at Each Test Tune  	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_jg
 Table4-10.   P-Values  for HCLV Test Results	....'.'.'.'.'.'.'.'.'.'.'.'.'.'.'. 4-19
 Table 4-11.   Number of HCLV PTH Anomalies at Post-Mechanical Shock
             by Surface Finish	4_2Q
 Table4-12.   P-Values  for HVLC Test Results	' 4.21
Table 4-13.   P-Values  for HSD Test Results	4_22
Table4^14.   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 50 MHz Circuit by Surface Finish  	4-26
Table 4T17.   Comparison of the Observed and Expected Number of Anomalies
             Under the Hypothesis of Independence of Surface Finishes	4-26
                                         XI

-------
Table4-18.   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 AnionฐData (HASL) 	4-36
Table 4-23.   Ion Chromatography Anion^Data (Immersion Tin)  	4-36
Table 4-24.   Ion Chromatography Anion0Data (Immersion Silver)  	4-37
Table 4-25.   Ion Chromatography AnionฐData (Nickel/Gold)  	4-37
Table 4-26.   Ion Chromatography AnionwData (OSP)	-.	4-38
Table 4-27.   Ion Chromatography Anion ฐ Data (Nickel/Palladium/Gold)	4-38
Table 4-28.   Acceptance Levels for Weak Organic Acids	 4-40
Table4-29.   Frequency of Anomalies by Individual Circuit Over Test Times  ...	 4-41
Table 4-30.   Surface Finishing Processes Evaluated in the Cost Analysis 	4-56
Table 4-31.   Cost Component Categories 	4-59
Table 4-32.   Number of Filter Replacements by Surface Finishing Process	4-63
Table 4-33.   Bath Volumes Used for Conveyorized Processes	4-65
Table 4-34.   Time-Related Input Values for Non-Conveyorized Processes 	* 4-66
Table 4-35.   Time-Related Input Values for Conveyorized Processes 	4-67
Table 4-36.   Bath Replacement Criteria for Nickel/Gold Processes 	4-68
Table 4-37.   Frequency and Duration of Bath Replacements for Non-Conveyorized
             Nickel/Gold Process	 4-69
Table 4-38.   Production Time and Down Time for the Surface Finishing Processes
             to Produce 260,000 ssf of PWB	 4-69
Table 4-39.   BOA for Transportation of Chemicals to the Surface Finishing
             Process Line	4-72
Table 4-40.   Costs of Critical Tasks	4-73
Table 4-41.   Materials Cost for the Non-Conveyorized Nickel/Gold Process	4-78
Table 4-42.   Chemical Cost per Bath Replacement for One Product Line of the
             Non-Conveyorized Nickel/Gold Process	4-78
Table 4-43.   Tiered Cost Scale for Monthly Wastewater Discharges to a POTW	4-81
Table 4-44.   Summary of Costs for the Non-Conveyorized Nickel/Gold Process	4-85
Table 4-45.   Total Cost of Surface Finishing Technologies	 4-86
Table 4-46.   Surface Finishing Alternative Unit Costs for Producing
             260,000 ssf of PWB	4-88
Table 4-47.   CWA Regulations that May Apply to Chemicals in the Surface
             Finishing Process	4-91
Table 4-48.   Printed Circuit Board Facilities Discharging Less than 38,000 Liters
             per Day PSES Limitations (mg/L)..:	 4-93
Table 4-49.   Printed Circuit Board Facilities Discharging 38,000 Liters per Day or
             More PSES Limitations (mg/L)	 4-93
Table 4-50.   PSNS for Metal Finishing Facilities	4-94
Table 4-51.   Amenable Cyanide Limitation Upon Agreement	4-94
Table 4-52.   PSES for All Plants Except Job Shops and Independent
             PWB Manufacturers	4-94
Table 4-53.   CAA Regulations that May Apply to Chemicals in the Surface
             Finishing Process	 4-95
                                          Xll

-------
Table 4-54.   CERCLA RQs that May Apply to Chemicals in the Surface
             Finishing Process	4.99
Table 4-55.   SARA and EPCRA Regulations that May Apply to Chemicals in the
             Surface Finishing Process 	4-101
Table 4-56.   TSCA Regulations and Lists that May Apply to Chemicals Used in
             Surface Finishing Processes	   4-102
Table 4-57.   Summary of Regulations that May Apply to Chemicals Used in
             Hot Air Solder Leveling (HASL) Technology 	4-104
Table 4-58.   Summary of Regulations that May Apply to Chemicals Used in
             Nickel/Gold Technology	4-105
Table 4-59.   Summary of Regulations that May Apply to Chemicals Used in
             Nickel/Palladium/Gold Technology	4-106
Table 4-60.   Summary of Regulations that May Apply to Chemicals Used in
             OSP Technology	4-107
Table 4-61.   Summary of Regulations that May Apply to Chemicals Used in
             Immersion Silver Technology	4-108
Table 4-62.   Summary of Regulations that May Apply to Chemicals Used in
             Immersion Tin Technology				4-109
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
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
                                         Xlll

-------
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_g
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 Technologies  .. 7-14
Table 7-9.    Glossary of Benefits/Costs Analysis Terms  	7-17
Table 7-10.   Potential Private Benefits or Costs Associated with the Selection of a
             Surface Finishing Technology	7-20
Table 7-11.   Overall Manufacturing Cost Comparison	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	 7.32
Table 7-18.   Number of HASL Chemicals Subject to Applicable Federal Regulations ... 7-35
Table 7-19.   Summary of Human Health and Environmental Risk Concerns for the
             Nickel/Gold  Technology	7-36
Table 7-20.   Number of Nickel/Gold Chemicals Subject to Applicable Federal
             Regulations	 7.40
Table 7-21.   Summary of Human Health and Environmental Risk Concerns for the
             Nickel/Palladium/Gold Technology  	7-41
Table 7-22.   Number of Nickel/Palladium/Gold Chemicals Subject to Applicable
             Federal Regulations	7-44
Table 7-23.   Summary of Human Health and Environmental Risk Concerns for the
             OSP Technology	7-46
Table 7-24.   Number of OSP Chemicals Subject to Applicable Federal Regulations	7-48
Table 7-25.   Summary of Human Health and Environmental Risk Concerns for the
             Immersion Silver Technology	7.49
Table 7-26.   Number of Immersion Silver Chemicals Subject to Applicable Federal
             Regulations	7-52
Table 7-27.   Summary of Human Health and Environmental Risk Concerns for the
             Immersion Tin Technology	7.53
Table 7-28.   Number of Immersion Tin Chemicals Subject to Applicable Federal
             Regulations	 7-56
                                         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 How Diagram	2-11
Figure 2-5.   OSP Process Row 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 Technologies  .. 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	4-44
Figure 4-2.   Boxplot Displays for HCLV PTH Post 85/85 - Pre-Test Measurements
             (volts) by Surface Finish	4-44
Figure 4-3.   Boxplot Displays for HCLV PTH Post TS - Pre-Test Measurements
             (volts) by Surface Finish	 4-45
Figure 4-4.   Boxplot Displays for HCLV PTH Post MS - Pre-Test Measurements
             (volts) by Surface Finish	4-45
Figure 4-5.   Boxplot Displays for HCLV SMT Measurements (volts)  at Pre-Test
             by Surface Finish	4-46
Figure 4-6.   Boxplot Displays for HCLV PTH Post 85/85 - Pre-Test Measurements
             (volts) by Surface Finish	4-46
Figure 4-7.   Boxplot Displays for HCLV PTH Post TS - Pre-Test Measurements
             (volts) by Surface Finish	4-47
Figure 4-8.   Boxplot Displays for HCLV PTH Post MS - Pre-Test Measurements
             (volts) by Surface Finish	4-47
Figure 4-9.   Boxplot Displays for HF PTH 50MHz Measurements (dB) at Pre-Test
             by Surface Finish	 4-48
Figure 4-10.  Boxplot Displays for HF PTH 50MHz Post MS - Pre-Test
             Measurements (dB) by Surface Finish	4-48
Figure 4-11.  Boxplot Displays for HF PTH f(-3dB) Post MS - Pre-Test
             Measurements (MHz) by Surface Finish	4-49
Figure 4-12.  Boxplot Displays for HF PTH f(-40dB) Post MS - Pre-Test
             Measurements (MHz) by Surface Finish	4-49
                                         xv

-------
Figure 4-13.  Boxplot Displays for HF SMT 50MHz Post MS - Pre-Test
             Measurements (dB) by Surface Finish	4-50
Figure 4-14.  Boxplot Displays for HF SMT f(-3dB) Post MS - Pre-Test
             Measurements (MHz) by Surface Finish	4-50
Figure 4-15.  Boxplot Displays for HF SMT f(-40dB) Post MS - Pre-Test
             Measurements (MHz) by Surface Finish	4-51
Figure 4-16.  Boxplot Displays for HF TLC 50MHz Post MS - Pre-Test
             Measurements (dB) by Surface Finish	4-51
Figure 4-17.  Boxplot Displays for HF TLC 500MHz Post MS - Pre-Test
             Measurements (dB) by Surface Finish	4-52
Figure 4-18.  Boxplot Displays for HF TLC RNR Post MS - Pre-Test
             Measurements (dB) by Surface Finish	4-52
Figure 4-19.  Boxplot Displays for 10-Mil Pad Measurements (Iog10 ohms)
             atPre-Testby Surface Finish	4-53
Figure 4-20.  Boxplot Displays for 10-Mil Pad Post 85/85 Pre-Test
             Measurements (Iog10 ohms) by Surface Finish  	4-53
Figure 4-21.  Boxplot Displays for PGA-A Measurements (Iog10 ohms) at Pre-Test by
             Surface Finish	4-54
Figure 4-22.  Boxplot Displays for PGA-B Measurements (Iog10 ohms) at Pre-Test by
             Surface Finish  		4-54
Figure 4-23.  Boxplot Displays for the Gull Wing Measurements (Iog10 ohms)
             at Pre-Test by Surface Finish	 4-55
Figure 4-24.  Hybrid Cost Analysis Framework	4-57
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	7-34
Figure 7-2.   Production Costs and Resource Consumption of the
             Nickel/Gold Technology	7-39
Figure 7-3.   Production Costs and Resource Consumption of Nickel/Palladium/Gold
             Technology	 7-44
Figure 7-4.   Production Costs and Resource Consumption of OSP Technology 	7-47
Figure 7-5.   Production Costs and Resource Consumption of Immersion Silver
             Technology  	;			7-51
Figure 7-6.   Production Costs and Resource Consumption of Immersion Tin
             Technology	7-56
                                         xvi

-------
                                  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
BGA         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
NOx         oxides of nitrogen
                                        xvni

-------
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        pretrearment standards for new sources
psi          pounds 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
S AR         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
TSDS        treatment, storage, or disposal facility
TWA        time-weighed average
                                        xix

-------
uc
UF
UT
UR
VOC
WHO
WOA
WOE
WS
unit cost
uncertainty factor
University of Tennessee
utilization ratio
volatile organic compounds
World Health Organization
weak organic acid
weight-of-evidence
water soluble
                                         xx

-------
                             Executive Summary
       The Printed Wiring Board Surface Finishes Cleaner Technologies Substitutes
Assessment: Volume 1 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

-------
EXECUTIVE SUMMARY
I.     DESIGN FOR THE ENVIRONMENT PRINTED WIRING BOARD PROJECT
       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.
    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 progranrinclude:

•  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.
                                           ES-2

-------
                                                                 EXECUTIVE SUMMARY
H.     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

-------
EXECUTIVE SUMMARY
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 CTSA. 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-ConveyorizedL .
•
•
•
•

•
Conveyorized
•


•
•
•
       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

-------
                       EXECUTIVE SUMMARY


a
.S
en
IM g
S fr"1
s

a
<8 >


&
0

ซ S "O
I'-sl


ซ 2
121

03
S



I
U

1*
O
ซ

i
o

i
6

i


I*
ฐ






>,
i





•









s
r*

e
b
-;

e
1

1
o
1

1
o
jj

"S
k o
1
ri









^
^
<












ซ
n

•3
ฃ,

I
*

I
3
u

-w
g
•5
e
<-

1C
(=
m



'


1>


•^
T*




'>


*
^



|
fe
•T

ersion
I5

CL|
O

f
1

i
. "O
3
-T

•M
u;




•

ซ
i
•• •ซ
i 3















c
Q
Hi




i
in

z
*t
1
&

o
J
1
•0

1
o
in

-




r?
Q
ri




1


1


i ^
1 *










Q





>,
1

a
c
•p
pi
i
. c









~


s
ซS rS oo

a
a>

>
>
1 ^
i.]



'

















1
CO
4ป>
CO
<1>
1
"1
.g

-------
EXECUTIVE SUMMARY
       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 85ฐC/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 hi 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.
                                          ES-6

-------
                                                                 EXECUTIVE SUMMARY
       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.

       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 CTS 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.
                                          ES-7

-------
EXECUTIVE SUMMARY
 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.

 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 confidentiality 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.
                                          ES-8

-------
                                                                  EXECUTIVE SUMMARY
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 CTS A 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.

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 260^00 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
                                         ES-9

-------
EXECUTIVE SUMMARY
       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

-------
                                                                   EXECUTIVE SUMMARY
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 (RI^o) 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)
•

•
•
•
HASL
(C)


•
•
•
OSP
(NC)

•



OSP
(C)

•



Immersion
' Silver (C)



•

Immersion Tin
(NC) ,




•
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

-------
EXECUTIVE SUMMARY
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.ofMSDS"
33
19
18
9
4
14
Hazardous Property t
:'.F
i


i


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

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
safely training program. Appropriate training can help reduce the number of work-related
accidents and injuries regardless of the technology used.
                                           ES-14

-------
                                                                 EXECUTIVE SUMMARY
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

-------
EXECUTIVE SUMMARY
       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  conveyorized 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.
     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

-------
                                                                                       EXECUTIVE SUMMARY
    a
    as
    •8
    cs
    CO
       CN
                 VO
                 vo
                 Cvป
I
8
p
o
•69
                        o
                        p
              O

              69
       VO
              OO
              CO
                 vo
                 -*
                    o
                    o
                                  U




                                   I


                                  I
                                              Hi
                                             I
                    69
                                          O
                                          O

                                          69
                                             VO
                                             o
                                                                            69
          CN
          CO
                    CN
                                                    CO,
                                                           ON
                                                           CO
                                                                  VO
                                                                     CN
OO
OO
CN
0\
^
                 VI
                 CO
                 CM
       CN
       B
                                                        VD
                           69
                                             5?
                                                               CO
                                                              69
                                                                  VD
                                                                     CN
                                                                            s
                                                                  69
U


I
                 O\
                                                    a
                                          VO
                                          00
       oo
       q
       C5
       8
o
o
                 CN
                 i—i
                                                    CN
                                                    f^
                                                    CN
                                             VO
                                             CO
                                                                  00
                                                                  v>
                                CO
                                o
                                                    s
                                                    o
                                                                     CN
                                                    S
                                                    p
    •8
              vo
OO
CN
CO
                        cs
                           CN
                           cs
                                      I
       vo
       "3
                    o

                    •69
                              69
                                             o
                                             o
                                                                         CN
                                                                         OO
                                                                     o
                                                                     o
                                                       •69
                                                              •69
                                                                  69
                                                                               I
                                                                         I
       I
       ง
       u
       I
              u
              g
                 !
                 I
          i
                            ง
                 U

                 OH"
                 co
                 O
                                   CO
                                      CO
                                       a
                                       o

                                      I
                                                       !
                                                               r
                                                              I
                                                              eC
                                                              CO
                                                                  1
                                                    CO
                                                    O
                                                            !
                                                            o
                                                            o
                                                           I
                                                                  ง
                                                    ES-17

-------
EXECUTIVE SUMMARY







.ฃ
S
'o
gJJ
*?)
ฃ
3
1
1
o>
w
t2
tJ
a
co
*S
o
CO
B
,ff
CO
&
R
CO
I
V
#
t>
lt
W
ot
42
m
H
















e
.2

CS
1
5>
ซ
o
i
t
I
5
4*ป
3
Iff
9
CO


.^
g
o>
6
(M
O
b
tu
I
^










s
^
i
^

i






EPCRA







^
ViJ






^
r">





&

IX,

ซs
oo •— <
)_(







CO

ffi
1

,

CO
-

1— (



co

r-~
cs




1 (


VO


t—4
vo



vo



vo
vo




T3
0
ง
— 1



m
ri



i—*
f-H




fe
CO
o
'^

-------
                                                                 EXECUTIVE SUMMARY
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.

       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. Conveyorized processes were found to consume less water than non-
conveyqrized 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.
Table ES-8. Energy and Water Consumption Rates of Surface Finishing Technologies
JProeessType
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.
                                         ES-19

-------
EXECUTIVE SUMMARY
       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.
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.
                                          ES-20

-------
                                                                  EXECUTIVE SUMMARY
       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 *
%
Capital costs,
Materials (chemical) costs,
Utility costs,
Wastewater discharge costs,
Production costs, and
Maintenance costs.
Worker sick days;
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
NA
Medical costs to workers;
Pain and suffering associated with
work-related illness.
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;
Reduced water supply.
 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).

       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-conveyorized 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.
                                          ES-21

-------
EXECUTIVE SUMMARY
V.     CONCLUSIONS

       The CTS A 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 CTS A 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 CTS A 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 CTS A 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 CTS A process.

       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
CTS A 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.
                                         __                   -                 -

-------
                                                          EXECUTIVE SUMMARY
I
I
u
I
s
   ffS*
•a
             oo
e
C
Ni
C
onveyorized
S
nveyori
ri
n

gi
se
ed.
                                                                1
                                   ES-23

-------
EXECUTIVE SUMMARY
                                   REFERENCES

IPC-Association Connecting Electronics Industries.  1996. 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.

Microelectronics and Computer Technology Corporation (MCC). 1993. Environmental
       Consciousness: A Strategic Competitiveness Issue for the Electronics and Computer
       Industry. March.

Microelectronics and Computer Technology Corporation (MCC). 1994. Electronics Industry
       Environmental Roadmap. December.
                                        ES-24

-------
                                    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 CTS A 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 CTS A by helping define the
scope and direction of the CTS A, developing project workplans, donating tune, materials, and
their manufacturing facilities for project research, and reviewing technical information contained
in this CTS A. 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
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
electronics manufacturing.  The results of these industry studies are presented in two reports
prepared by Microelectronics and Computer Technology Corporation (MCC), an industry
                                           1-2

-------
                                                            1.1 PROJECT BACKGROUND
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 processe's 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

-------
1.1 PROJECT BACKGROUND
       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 CTS A 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    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

-------
1.2 OVERVIEW OF PWB INDUSTRY
                                                     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

-------
                                                        1.2 OVERVIEW OF PWB INDUSTRY
              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
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

-------
                                                               1.3 CTSA METHODOLOGY
       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,
and the risk to the environment resulting from the discharge of the wastewater to nearby surface
                                           1-9

-------
1.3  CTSA METHODOLOGY
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

-------
                                                                1.3 CTSA METHODOLOGY
       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
Electro-chemicals, Inc.
Florida CirTech, Inc.
MacDermid, Inc.
Technic, Inc.
Surface Finishing Technology
Nickel/Gold



/
/
Nickel/Palladium/
Gold



/

OSP

S

/

Immersion
Silver
V




Immersion
Tin
^

S


                                           1-11

-------
1.3 CTSA METHODOLOGY
       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

-------
                                                               1.3 CTSA METHODOLOGY
       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

-------
1.3 CTSA METHODOLOGY
       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
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
                                   REFERENCES

Abrams, Fern. 2000. IPC-Association Connecting Electronics Industries. Personal
       communication to Jack Geibig, UT Center for Clean Products and Clean Technologies.
       December.

IPC-Association Connecting Electronics Industries.  1996. The 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.

Microelectronics and Computer Technology Corporation (MCC). 1993. Environmental
       Consciousness: A Strategic Competitiveness Issue for the Electronics and Computer
       Industry. March.

Microelectronics and Computer Technology Corporation (MCC). 1994. Electronics Industry
       Environmental Roadmap. December.

U.S. Environmental Protection Agency  (EPA).  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. Environmental Protection Agency  (EPA).  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. Envkonmental Protection Agency  (EPA).  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. Envkonmental Protection Agency  (EPA).  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. Envkonmental Protection Agency  (EPA).  1998a. Printed Wiring Board Cleaner
       Technologies Substitutes Assessment: Making Holes Conductive. Design for the
       Envkonment 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. Envkonmental Protection Agency (EPA).  1998b. Printed Wiring Board Pollution
       Prevention and Control Technology: Analysis of Updated Survey Results.  Design for the
       Envkonment Printed Wiring Board Project. EPA Office of Pollution Prevention and
       Toxics. Washington, D.C. EPA 744-R-98-003. August.
                                         1-16

-------
                                                                   REFERENCES
                      •   A   ™rppA^  1999 "Printed Wiring Board Case Study 9:
U'S-                  SEX^•ฃ%. EPA Office ofWon Prevent and
      Toxics. WasMngton,D.C. EPA 744-F-99-004. May.



       Washington, D.C. EPA 744-R-00-002. March.



       July.



       June.

-------

-------
                                     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 paUadium/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
defines the process sequences.  An overview of each surface finish process describes the typical
operating conditions and maintenance procedures, the chemical processes occurring in each bath of
the process, and provides additional process information relevant to each surface finish 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 finish
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

-------
2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES

1l


n
•2 ..
is
JB

04
O2
o

111
2lw
Pk

w ปB
•83
go

&



>ซ
0)
a
n
o>
U
-•

1
OJ
U

he
U
ง
8
M

hi
U

1
ea
u

fc
H
u





>,


>,


>,


•w,






J3
ซ
P
_u
H

1
U
o
&
r!

1
ri

1
b
i^

1
1
H

1
OJ
O
&
"



1

>,


•w
^
<


•








ja
•3
ฃ
m

i
.E
<<

s
2
*
n

•*j
I
cs
U
•!

•w
i
3
n

&
a
m



>.
•

>,





>,
*


i



•^


a
e>
V
I
-f

a
Is
•ป

ง5
o
•*

*
^
•f

i
•n
5
-T

M
E
TT





s
o




•^
•J




>
J















P*





f
Q
VI


P^

0>
4
wi

V
s
1
.
•




a
Q
5
6
5
I
PL
xe











I
Pk
uS

sion Gold
u
I
1
Mi

1
>
ซ ^_
!>•











•s
^ i
^H ฎ
^^ H ^ ^^ *fS
•>1fl> 6
ป-3 C5 -0
w* i
h- 1
•J tปS


-

4ป
99
a
•JS
K
s
r-^

                                    2-2

-------
	2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES

 2.1.2   Overview of the Surface Finishing Manufacturing Process

        Surface finish 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 minrmization 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, conveyoriz-ed 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

-------
2.1  CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES

       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.

2.13  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

-------
       2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES

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

-------
2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES


1- Cleaner
y
Microetch
y
Water Rinse x 2
y
3- Dry
I

I

I

1
y
4- Flux
y
5- Solder
y


I

6- 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

-------
	2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES

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
(BGA) 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

-------
2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES


1. Cleaner 1
i
Water Rinse x 1

I
1
2. Microetcfa
1
1

Water Rinse xl I
i
3- Catalyst |
*
Water Rinse xl

1
4- Acid Dip
Y
I

Water Rinse x 1 1
1

5. Electroless Nickel 1
V
Water Rinse x 2 1
i
6. Immersion Gold
' 1
Water Rinse x 2

I

I

                      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

-------
	2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES

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

-------
2.1  CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES

       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

-------
2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES


Cleaner

1
Water Rinse x 2 1
1
2- Microetch 1
1
Water Rinse x 2
1
3- Catalyst
y
Water Rinse x 2
I

I

1
1
4- Acid Dip
1
Water Rinse x 2
1
5- Electroless Nickel
I

I

I
1
Water Rinse x 2
1
6- Preinitiator
t
I

I

7- Electroless Palladium I
' t
Water Rinse x 2

1
8- Immersion Gold
V
Water Rinse x 2

1

I

          Figure 2-4. Nickel/Palladium/Gold Process Flow Diagram
                                  2-11

-------
2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES

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 Solder-ability 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 benzirnidazole 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
BGA 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

-------



-------
       ;P5;
     -P6:

  •feป(
             ^4 Waon-
**ซ5L**^ฃ--
a/scv
^oca/o-Y  n-aojjฃOflf   ซcs/  j^    01uerahT   ltorth    fl- Si/v    wrฐt

                                                         e*Posed
                                 'ฐ'^>ฃ?b'*ปSto%*ซ'4,*
                                                          ^caz/y

                                                   'Us
                  2-14

-------
       2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES


1- Cleaner |
i
Water Rinse x 1

D
1
2- . Microetch 1
t
Water Rinse x 1 I
t
3- Air Knife

t
4. OSP
Y
1

5. Air Knife |
t
Water Rinses 1 |
-t
6. 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 coaling application. This step also
              minimizes drag-in of sulfates, which are harmful to the OSP bath.
                                           2-13

-------
2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES

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
BGA 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

-------
       2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES


1- Cleaner I
^^^
Water Rinse x 1
J
t
2. Microetch
t
Water Rinse xl
t
I

I

3- Predip |
y
4- Immersion Silver I
y
Water Rinse xl
. t
I

5. Dry |


                   Figure 2-6. Immersion Silver Process Flow Diagram
Step 4:
Step 5:
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.

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

-------
2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGffiS
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-metalh'c 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 hi 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
Y
Water Rinse x 2
Y
2. Microetch
Y
Water Rinse x 2
Y
J

I

I

I

3- Predip |
Y
Water Rinse xl I
Y

4. Immersion Tin 1
Y
Water Rinse x 2
Y
5. Dry


I

I

                     Figure 2-7.  Immersion Tin Process Flow Diagram
                                           2-16

-------
	2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES

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.
       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
                                           2-17

-------
2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES
Performance Demonstration each submitted complete chemical formulations along with other
publicly-available information on their respective product lines.  This information includes:

•      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
HASL







•



•
•

•
•



Nickel/
Gold.

•
•

•
•
•


•


•


•



Nickel/
Palladium/
Gold


•

•
•
•


•


•



•
•
•
OSP
•









•








Immersion
Sfflver








•










Immersion
Tin



•



•

•

•

•
•




                                          2-18

-------
2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES
Chemical
f
Ammonium chloride
Ammonia compound A
Ammonia compound B
Ammonium hydroxide
Aromatic imidizole product3
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
HASL





•

•


•

•

•
•

•
•
•
•

•
•



•








•

Nickel/
Gold
•

•
•



•


•

•






•
•

•

•
•
•


•

•


•

•
•
NidceV
Palladium/
Gold-

•
•
•



•


•

•
•





•
•

•


•


•
•

•



•
•
•
OSP




•
•


•
•
•

•

•



•
•
•

•
•












•

Immersion
Silver
















•



•











•
•


•

Immersion
TJn






•
•



•
•


•

•

•

•








•





•

                                 2-19

-------
2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES
Chemical
Potassium gold cyanide
Potassium peroxymonosulfate
Propionic acid
Quantenary alkylammonium chlorides
Silver nitrate
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

•




•
•






•


-•







Nickel/
Gold
•
•





•
•
•

•

•
•




•


•


Nickd/
Palladium/
Gold
•

•




•

•

•

•
•
•



•


•


OSP







•






•










Immersion
Silver




•


•






•










Immersion
Tin

•

•

•
•



•

•

•

•

•

•
•

•
•
• 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 bam 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

-------
       2.1 CHEMISTRY AMD PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES

                           Q = (CCHEM) (CFORM) (D) (1000 cm3/L)
where,
D
concentration of constituent in bath (g/L)
chemical concentration, by weight, in the product, from chemical product
formulations obtained from chemical suppliers (%)
proportion of the product formulation volume to the total bath volume,
from Product Data Sheets (%)
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 bam 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:
                      40g [  10LVl.27gWlOOOcm3>|_50gg
                      lOOgVOOLA cm3 )(     L    }     '  L
       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 hi 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 hi each surface
finishing process alternative.
                                           2-21

-------
2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES

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.

       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
2.2    ADDITIONAL SURFACE FINISHING TECHNOLOGIES

       The surface finishing technologies described in Section 2.1 represent the technologies
that were evaluated in the CTS A.  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.

-------
2.2 ADDITIONAL SURFACE FINISHING TECHNOLOGIES
       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
                                  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 to Jack Geibig, UT
       Center for Clean Products and Clean Technologies. July.
                                        2-25

-------

-------
                                     Chapters
                     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 sqme 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

-------
3.1  SOURCE RELEASE ASSESSMENT
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 hi 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 Project 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 then: 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

-------
                                                       3.1 SOURCE RELEASE ASSESSMENT
       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

-------
3.1 SOURCE RELEASE ASSESSMENT
       CO
       TJ



       O
       CQ



       CD


       I

       CO
O

PQ
OQ
O
                                      CO
                     .2 o>

                     E&ฐ-
                     CD O O
                     j= o-g

                     O.E o
                 fi

                 i
                 T3

                 O
                 CO
                 CO

                 O
                      I
                      •a
                                CO
                    CO 0)


                lit!
                CD y= a. o
                X i i  i
                           CD
                                p

                                '53
                                           S
                            .c c: '3 *s

                            1 -2 ฃ o
                              € Q.O
                                       *= o_
                                       ฐ m CO
                                       te-2 co
                                       o o o
                                        I ^^ Q.


                                       6 ง=ง
           s'e-
           g."0 m
           CO E  (Dt
          LU CO i
               D)
               _C


               (0  Q.

                                              J2   .E
                                              CO   C
                                     -



                                  111
                                       3-4

-------
                                                         3.1 SOURCE RELEASE ASSESSMENT
         Wastewater>
     (primarily from rinse tank)
                   Spent bath solutions (include wasted
                   equipment cleaning chemicals)
                                                                          System Boundary
Wastewater
Treatment
 System
                                         Treatment
                Sludge to recycle
                  or disposal
                                    Discharge to PO7W or
                                      Receiving Stream
                 Figure 3-2. Wastewater Treatment Process Flow Diagram
       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:
                                   Itotal =  Il
where,
Ii      =      bath chemicals
I2      = .     etched and solder mask-coated PWBs
I3      =      water
I4      =      cleaning chemicals

These terms are discussed below.

Ij      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
       calculation of bath chemical concentrations is presented in Section 2.1.4, Chemical
                                            3-5

-------
3.1 SOURCE RELEASE ASSESSMENT
       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

-------
                                                        3.1 SOURCE RELEASE ASSESSMENT
Process Type
No. of Responses
Water Usage (I3)
(thousand gal/year)?
Water Usage as)
(gaVssf)
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)
1.1.0
0.333
 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.
       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, and 3.13). 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

-------
3.1 SOURCE RELEASE ASSESSMENT
        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 Flushing8
1(27)
2 (27)
5(28)
1(22)
1 (8)
8(8)
1 (8)
1 (8)
5(9)
1(1)
KD
2(2)
2(2)
1(2)
KD
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:

                                     •"•total  = A! + A2
                                           3-8

-------
                                                        3.1 SOURCE RELEASE ASSESSMENT
where,

A;     =

These terms are discussed below.
              evaporation and aerosol generation from baths
              evaporation from drying/ovens
       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.
     From Questionnaire, questions 2.10 and 3.10.

     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.

                                           --    -        ._

-------
3.1 SOURCE RELEASE ASSESSMENT
       Table 3-3. Average Bath Dimensions and Temperatures for All Processes'
Bath
No. of
Responses
Length
(in.)
Width
On.),
Surface Area
(sq. in.)
Volume
(gal.) ;
Temp.
' *CF)
HASL, Non-conveyorized
Cleaner
VKcroetch
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

-------
                                                         3.1 SOURCE RELEASE ASSESSMENT
Bath
OSP
No. of
Responses
4
Length
(in.)
27
Width
(in.)
24
Surface Area
(sq. in.) '
580
Volume
(gal-)
86
Temp.
f 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

-------
3.1 SOURCE RELEASE ASSESSMENT
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:
where,
W,    =
W2    =
w,    =
                                 wtotal  =
wastewater
spent bath solution
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.
                                          3-12

-------
                                                      3.1 SOURCE RELEASE ASSESSMENT
W,
w,
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,
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
29
35
8
14
3
3
pH
Neutralization *
24
25
3
15
3
6
Disposed
to Sewer*
1
0
0
0
1
0
Drummed
11
2
7
4
2
5
Recycled
Oil-Site*
6
2
1
1
0
3
.Recycle
OflMSite
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

-------
3.1 SOURCE RELEASE ASSESSMENT
       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.

Et     '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 a NPDES 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:
                                   total
                                      ~  Sj + O2 + 03 + 04
where,
Si
S2
S3
hazardous solid waste
non-hazardous solid waste
drummed solid or liquid waste
sludge from on-site wastewater treatment
 These terms are discussed below.

 Sj    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

-------
                                                       3.1  SOURCE RELEASE ASSESSMENT
       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 Assessment.
       Hazardous solid waste is typically sent off-site to a hazardous waste landfill for disposal
       or is incinerated.

       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.

       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.

       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).
     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.

                               '           2ML5

-------
3.1 SOURCE RELEASE ASSESSMENT
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
Dram
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

-------
                                                       3.1 SOURCE RELEASE ASSESSMENT
 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 I15 I2, 13, and I4, and the outputs P15 A15 A2, Wlf W2, W3, S19 S2, and S3.

 Since the inputs must equal the outputs, the material balance for Figure 3-1 is:
or:
                           Itotai = PI + Atotal + Wtota] + Si + S2 + S3

The material balance for Figure 3-2 (wastewater treatment) includes the inputs
and the outputs El and S4

.Thus, the material balance equation for Figure 3-2, wastewater treatment, is:
                                                                          j, W2, and W3
or:
                                     Wtotal = Ej + S4
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

-------
3.1 SOURCE RELEASE ASSESSMENT
       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.

•

L Cleaner
*
2- Microetch
1

1
t
3. Water Rinse x 2

. t
4- Dry
y
5- Flux
*
6- Preheat
t
7. HASL
y
8- Air Knife
• v.
9- Pressure Rinse
1

1

1

I

I

I
V
10. Water Rinse xl

1

            Figure 3-3. Generic HASL Process Steps and Typical Bath Sequence
                                           3-18

-------
                                                     3.1 SOURCE RELEASE ASSESSMENT
       Water Usage (I3) and Wastewater (Wj). 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 (Ix).  Bath concentrations of individual chemical constituents are
presented in Appendix B.  The volume of chemicals consumed per year was determined by
modeling the tune 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

-------
3.1 SOURCE RELEASE ASSESSMENT
       , 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 convey orized 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 (Aj).  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 (Aj). 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 (P^. 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 hi further detail in Section 4.3,  Regulatory Status. In response to a
separate question regarding spent bath treatment (see Table 3-4), 1 1 out of 1 13 HASL baths were
reported by respondents to be drummed and sent off-site for recycling or disposal.
                                          3-20

-------
                                                      3.1 SOURCE RELEASE ASSESSMENT
Nickel/Gold Process

       Figure 3-4 depicts the generic nickel/gold process steps and typical bath sequence
evaluated in the CTS 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.


I- Cleaner
1
y
2. Water Rinse zl

V
3- Microetch 1
y ~
4. Water Rinse zl |
t '
5- Catalyst

y
6. Water Rinse zl
y
1

7- Acid Dip |
y
8- Water Rinse zl
y '
9. Electroless Nickel
1

1
t
10. Water Rinse z 2 j
^
1 1 . Immersion Gold
1
\f
12. Water Rinse z 2 |


        Figure 3-4. Generic Nickel/Gold Process Steps and Typical Bath Sequence
                                         3-21

-------
3.1 SOURCE RELEASE ASSESSMENT
       Water Usage (I3) and Wastewater (Wx). 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 (Ij). 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.
                                          3-22

-------
                                                       3.1 SOURCE RELEASE ASSESSMENT
       Evaporation From Baths (Aj). 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/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 (Pj). 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 Assessment. 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

-------
3.1 SOURCE RELEASE ASSESSMENT

1- Cleaner
V

2- Water Rinses 2
V

3- Microetch
V

4- Water Rinses 2
)

5- Catalyst
)
i
6- Water Rinses 2
)
i
I

I

I

]|

I

I

7- Acid Dip |

f
8- Water Rinses 2

V
9- Electroless Nickel

V

I

I

1
f
1L Preinitiator 1
>
<
12- Electroless Palladium
>
1
13. Water Rinse x 2
>
1
14- Immersion Gold
>
t

I

I

I

1S- Water Rinses 2 1


   Figure 3-5. Generic Nickel/Palladium/Gold Process Steps and Typical Bath Sequence
                                       3-24

-------
                                                      3.1 SOURCE RELEASE ASSESSMENT
       Water Usage (I3) and Wastewater (W^.  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 (Ij).  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

-------
3.1 SOURCE RELEASE ASSESSMENT
       Evaporation From Baths (Ax). 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 (P^. 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 Assessment. 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 Solderabilitv 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

-------
                                                      3.1 SOURCE RELEASE ASSESSMENT


!• Cleaner
J
V
2. Water Rinse xl
J
i
3- Microetcb
n
i
4. Water Rinse xl
n
i
5- Air Knife
j
t
6- OSP
j
i
7- Air Knife
I
i
8. Water Rinse xl
V
9- Dry

j

1

           Figure 3-6. Generic OSP Process Steps and Typical Bath Sequence
       Water Usage (I3) and Wastewater (W^. 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, of 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

-------
3.1 SOURCE RELEASE ASSESSMENT
       Bath Chemicals Used (Ij). 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 detennining 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 (Aj). 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

-------
                                                      3.1 SOURCE RELEASE ASSESSMENT
       Chemicals Incorporated onto PWBs (P^. 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 Assessment. 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
y
2. Water Rinse xl
>L
3. Microetch
t
J

I

I

4. Water Rinse xl |
y
5. Predip
i
6. Immersion Silver

I

n
t
7. Water Rinse xl
1
J

8. Dry |


      Figure 3-7.  Generic Immersion Silver Process Steps and Typical Bath Sequence
                                           3-29

-------
3.1 SOURCE RELEASE ASSESSMENT
       Water Usage (I3) and Wastewater (W^. 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 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 (Ix). 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

-------
                                                      3.1 SOURCE RELEASE ASSESSMENT
       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 (A^. Ah- 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 of ten process baths was reported to be vented to the
       outside.
•      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 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 (P^. 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 Assessment. 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 CTS A. 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

-------
3.1 SOURCE RELEASE ASSESSMENT



1- Cleaner 1
y
2. Water Rinse x 2
y
3- Microetch
y
4. Water Rinse x 2

I

^j

D
y
5- Predip
y
6. Water Rinse xl
y
7. Immersion Tin
y
8. Water Rinse x 2
y
9. Dry

1

1

1

1

1

       Figure 3-8. Generic Immersion Tin Process Steps and Typical Bath Sequence
       Water Usage (I3) and Wastewater (Wj). 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

-------
                                                      3.1 SOURCE RELEASE ASSESSMENT
       Bath Chemicals Used (Ix). 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 viable treatment options.

       Evaporation From Baths (Aj).  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 is
       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 (Aj).  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

-------
3.1 SOURCE RELEASE ASSESSMENT
       Chemicals Incorporated Onto PWBs (Px). 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 Asessment. 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 differences due to the
       varying amount of PWBs produced.  According to the query results from the
       questionnaire.database, data from facilities with small amounts of PWB produced often
       produce unrealistic results.  For surface finish alternatives with a limited number of
       responses, enough data may not exist to have statistically meaningful results which are
       representative of most PWB facilities.

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 engineering estimates and
assumptions used in this assessment.
                                          3-34

-------
                                                            3.2 EXPOSURE ASSESSMENT
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

-------
3.2 EXPOSURE ASSESSMENT
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

-------
                                                            3.2 EXPOSURE ASSESSMENT
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

-------
3.2 EXPOSURE ASSESSMENT
       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 conveybrized 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.
                                          3-38

-------
                                                               3.2 EXPOSURE ASSESSMENT
       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
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-cpnveyorized 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.
     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.
—                                         _

-------
3.2 EXPOSURE ASSESSMENT
       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.

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 a
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-conyeyorized lines.
                                          3-40

-------
                                                                       3.2 EXPOSURE ASSESSMENT
Activities :;/-:\f
Chemical Bath
Replacement;
Conveyor Equipment
Cleaning; Filter
Replacement;
Chemical Bath Sampling
Rack Cleaning
Chemical Bath Additions
1 ^^n^^sOimsys'l
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.
-- " Evaliuai^wtt Alnprda^ a^ "•'
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

-------
3.2 EXPOSURE ASSESSMENT
               Table 3-7. Potential Ambient 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.
l^tiation Approa<* and Mtionale
3xposure 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 of
 effects on groundwater.
                                            3-42

-------
                                                            3.2 EXPOSURE ASSESSMENT
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. Monitoring data were used
for evaluating lead inhalation from the HASL process. Bath concentrations for dermal exposure
were estimated from bath chemistry data.  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.
Although lead is not volatile at he melting temperature, there may be some lead present that
could not be modeled.  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).
                                         3-43

-------
3.2 EXPOSURE ASSESSMENT
       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

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,
Az
volatilization rate of chemical y from an open tank (mg/min)
concentration of chemical y in bulk liquid (mg/L)
dimensionless Henry's Law Constant (HJ for chemical y
bath surface area (m2)
molecular diffusion coefficient of chemical y in air (cnrVsec)
air velocity (m/sec)
pool length along direction of air flow (m)
       Concentration of chemical in bulk liquid (cL_y) 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 shown hi Table 3-8:

              Table 3-8. Bath Surface Areas for Conveyorized Process Baths
Conveyorized Bath Type
Cleaner baths
Immersion silver
Immersion tin
Microetch baths
OSP
Predip baths
.,, Av^rageSurfece Area {swings
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

-------
                                                             3.2 EXPOSURE ASSESSMENT
       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-es
                                                   HQ
                                                    yo
where,
py,s
Qo
Hy
a
VL
mass transfer rate of chemical y out of the system by sparging (mg/min)
air sparging gas flow rate (L/min)
dimensionless Henry's Law Constant (Hc) for chemical y
concentration of chemical y in bulk liquid (mg/L)
overall mass transfer coefficient for chemical y (cm/min)
interfacial area of bubble per unit volume of liquid (cm2/cm3)
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 and Lindh, 1987): .
R
                = [5.5x10
                                             A)+O.Ol] FT FA FD
where,
RA
Qo
A
FT
FA
FD
aerosol generation rate (ml/min/m2)
air sparging gas flow rate (cm3/min)
bath surface area (m2)
temperature correction factor
air velocity correction factor
distance between the bath surface and tank rim correction factor
                                          3-45

-------
3.2 EXPOSURE ASSESSMENT
       The emission of contaminants resulting from aerosols depends on both the rate of aerosol
generation and the concentration of contaminants in the aerosol. The following equation is used
to estimate contaminant emission (flux) from aerosol generation:
                                           M,
                                             ,
                                       -  _ L f
                                               I
                                                 p
                                               IE   y's
where,
Fy>a

M,    =
             rate of mass transfer from the tank to the atmosphere by aerosols (mg/min)
             fraction of bubble interface ejected as aerosols (dimensionless)
             mass of contaminant at the interface (mg)
             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 Fy 0 calculated by the first equation. For
sparged baths, the total emission rate is equal to Fy 0 + Fy s + Fy a, 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, I991a):

                                    Cy=Fyj/(QK)
where,

F T
Qy>
k
              workplace contaminant concentration (mg/m3)
              total emission rate of chemical from all sources (mg/min)
              ventilation air flow rate (nrVmin)
              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 mVmin (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.
                                          3-46

-------
                                                              3.2 EXPOSURE ASSESSMENT
       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.
       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 ah- 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 3-9.

                     Table 3-9.  Results of Workplace Air Modeling
Chemical',,/
/
Total Emission
Rate(FyW)
(mg/min)
Workplace
Air Cone.

-------
3.2 EXPOSURE ASSESSMENT
Chemical a
Total Emission
RateCFy^)
(mg/min)
Workplace
Air Gone.
(^(ing/in3)
Federal OSHA and/or NIOSH
Permissible Inhalation Exposure
Limits (mg/m3)"
Mickel/Gold, Non-conveyorized
Aliphatic acid A
Aliphatic acid B
Aliphatic acid E
Aliphatic dicarboxylic acid A
Aliphatic dicarboxylic acid C
Alkyldiol
Ammonia compound B
Ammonium hydroxide
Hydrochloric acid
Hydrogen peroxide
fnorganic metallic salt A
[norganic metallic salt B
tnorganic metallic salt C
Malic acid
Nickel sulfate
Phosphoric acid
Potassium compound
Sodium hypophosphite
Urea compound B
77
5.4E-04
100
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
LO
0.64
7.6E-04
5.9
4.1E-05
7.8
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
NR
NR
NR
NR
NR
NR
NR
none
NIOSHREL,C:7(5ppm)
OSHA PEL, C: 7 (5 ppm)
NIOSHREL:1.4(lppm)
OSHA PEL: 1.4(1 ppm)
NR
NR
NR
none
NIOSH REL, Ca: 0.015
OSHA PEL: 1
NIOSHREL: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
5.6E-04
140
0.11
0.051
22
0.026
2.0
0.064
28
3.7
0.0021
0.23
0.90
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
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: 0.015
OSHA PEL: 1
                                     3-48

-------
                                                                              3.2  EXPOSURE ASSESSMENT
Chemical*
Phosphoric acid
Potassium compound
Propionic acid
Sodium hypophosphite
Urea compound B
Total Emission
RateO&W)
(mg/min)
1.2
1.1
26
0.85
0.0015
Workplace
Air Cone.
(Cv)(mg/m3)
0.092
0.082
2.0
0.065
1.2E-04
Federal OSHA and/or NIOSH
Permissible Inhalation Exposure
Limits (mg/m3)b
NIOSH REL: 1, STEL: 3
OSHA PEL: 1
MR
NIOSH REL: 30 (10 ppm)
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 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.
                                                      3-49

-------
3.2 EXPOSURE ASSESSMENT
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 10 m 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 hi 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 adjustment 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 3-10.
   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

-------
                                                                       3.2 EXPOSURE ASSESSMENT
Table 3-10. Results of Ambient Air Modeling
Chemical
Emission Rate a
(mg/min)
* Air Concentration11
(mgfm3)
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
  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 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.
c Based on air monitoring data from one facility, with an average workplace air concentration of 0.003 mg/m3.
                                                 3-51

-------
3.2 EXPOSURE ASSESSMENT
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
pretrearment. 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
SIZE
WR
DT
ALK
HOLES

ELCTRLS    =
drag-out from bath, ml/m2
board area, m2
withdraw rate, m/sec
drain time, sec
1 if the bath is an alkaline cleaner bath and = 0 otherwise
1 if the board is drilled and = 0 for undrilled boards (we assumed that all
boards were drilled)
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:
                                          3-52

-------
                                                            3.2 EXPOSURE ASSESSMENT
                           MDij =  P * Cij * DOij / 1,000,000

where,
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
Vj     =     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:

                                       n
                                 Mi = ฃ (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 =  gMBij


       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.
                                         3-53

-------
3.2 EXPOSURE ASSESSMENT
       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 = 1000 Mi / (Qsw +
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

       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, Recycle, Recovery, and Control
Technologies Assessment.)  Results for chemicals, excluding metals, where the initial stream
concentration (without treatment) exceeded the CC for that chemical are presented in Table 3-11.
Full results are presented in Appendix E.

          Table 3-11. Estimated Releases to Surface Water Following Treatment
Chemical "
Cone, in
Wastewater
Released to
Stream (mg/L)
Stream Cone.
w/oPOTW
Treatment
(mg/L)
Treatmen
, t
Efficiency
Stream Cone.
after POTW
Treatment
(mg/L)
HASL, Non-conveyorized
1,4-Butenediol
49
0.10 | 90
0.010
                                          3-54

-------
                  3.2 EXPOSURE ASSESSMENT
Chemical *
Alkylaryl sulfonate
Citric acid
Ethylene glycol monobutyl ether
Hydrogen peroxide
Potassium peroxymonosulfate
Cone, in
Wastewater
Released to
Stream (mg/L)
2.3
94
71
195
390
Stream Cone.
w/o POTW
Treatment
(mg/L)
0.0049
0.20
0.15
0.41
0.82
Treatmen
t
Efficiency
(%)
0
93
90
90
90
Stream Cone.
after POTW
' Treatment
: (mg/L)
0.0049
0.014
0.015
0.041
0.082
HASL, Conveyorized
1,4-Butenediol
Alkylaryl sulfonate
Citric acid
Ethylene glycol monobutyl ether
Hydrogen peroxide
Potassium peroxymonosulfate
23
1.0
42
32
90
180
Nickel/Gold, Non-conveyorized
Hydrogen peroxide
Substituted amine hydrochloride
62
97
0.076
0.0035
0.14
0.11
0.30
0.61
. 90
0
93
90
90
90
0.0076
0.0035
0.0099
0.011
0.030
• 0.061

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
1.2
660
36
200
42
0.0021
1.2
0.064
0.36
0.074
0
93
90
90
90
0.0021
0.082
0.0064
0.036
0.0074
3-55

-------
3.2 EXPOSURE ASSESSMENT
Chemical"



Thiourea
Urea compound C
Cone, in
Wastewater
Released to
Stream (mg/L)
170
35
Stream Cone.
w/oPOTW
Treatment
(mg/L)
0.30
0.062
Treatraen
t
Efficiency
(%)
90
90
Stream Cone.
after POTW
Treatment
(mg/L)
0.030
0.0062
Immersion Tin, Conveyorized
Potassium peroxymonosulfate
68 | 0.041
90
0.0041
* 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.
3.2.4  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):
where,
I
Ca

JR
ET
                       I = (Ca)OR)(ET)

daily inhalation potential dose rate (mg/day)
airborne concentration of substance (mg/m3)
(Note: this term is denoted "Cy" in air modeling equation in Section 3.2.3.)
inhalation rate (mVhr)
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:
   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

-------
                                                            3.2 EXPOSURE ASSESSMENT
where,
LADD
ADD
I
EF
ED
BW
AT,
AT,
CAR
   NC
                          LADD =  (I)(EF)(ED)/[(BW)(ATCAR)]
                           ADD =  (I)(EF)(ED)/[(BW)(ATNC)]
lifetime average daily dose (mg/kg-day) (for carcinogens)
average daily dose (mg/kg-day) (for non-carcinogens)
daily inhalation potential dose rate (mg/day)
exposure frequency (days/year)
exposure duration (years)
body weight (kg)
averaging time for carcinogenic effects (days)
averaging time for non-carcinogenic effects (days)
       Parameter values for estimating workers' potential dose rates from inhalation are
presented in Table 3-12.
Table 3-12. Parameter Values for Workplace Inhalation Exposures
Parameter ,
Air Concentration (Ca)
Inhalation Rate (IR)
Units
mg/m3
mVhr
. Value
Source of Data, Commeirtsl?
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 	 16
supervisors 	 55
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
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 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.
                                         3-57

-------
3.2 EXPOSURE ASSESSMENT
Parameter
Exposure Duration
(ED)
Body Weight (BW)
Averaging Time (AT)
ATcAR
ATnc
Units
years
kg
days
. Value ' r>: A ;--;;.;•
25
70
25,550
9,125
Source oiiDataV^ommenlS
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, 1991a) is
as follows:
                                       D =  SQC
where,
D
S
Q
c
dermal potential dose rate (mg/day)
surface area of contact (cm2)
quantity typically remaining on skin (mg/cm2-day)
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:
 where,
 D
 S
 C
 f
 ET
                                D = (S)(C)(f)(ET)(0.001)
dermal potential dose rate (mg/day)
skin surface area of contact (cm2)
chemical concentration (mg/L)
flux through skin (cm/hour)
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  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).
                                          _                    —   ,

-------
                                                             3.2 EXPOSURE ASSESSMENT
       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 3-13.
         Table 3-13.
Parameter
Chemical
Concentration (C)
Skin Surface Area (S)
Flux Through Skin (f)
Exposure Duration
(ED)
Body Weight (BW)
Averaging Tune (AT)
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 3-14.
                                          3-59

-------
3.2 EXPOSURE ASSESSMENT
   Table 3-14. Parameter Values for Workplace Dermal Exposures for Line Operators
                               on Non-Conveyorized Lines
Parameter/
Activity *
Units
Value
/ *
Source of Data, Comments
Exposure Time (ET)
Line Operation *
hrs/day
Process / no. baths or steps
HASL (NC) / 10
NickeVGold (NC) / 14
Nickel/Palladium/Gold (NC) / 22
OSP(NC)/9
Immersion Tin (NC) / 12
Value
0.80
0.57
0.36
0.89
0.67
Based on a default value of 8
hrs/day; corrected for typical
number of baths in a process,
including rinse baths, by dividing 8
hrs/day by the number of baths
and/or steps in a typical process
line.
Exposure Frequency (EF)
Line Operation "
days/yr
HASL (NC) 	 44
HASL (C) 	 22
Nickel/Gold (NC) 	 212
Nickel/PaUadium/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.
8 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 conveyorized 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 conveyorized lines are provided in
Table 3-15.
                                           3-60

-------
                                                     3.2 EXPOSURE ASSESSMENT
Table 3-15. Parameter Values for Workplace Dermal Exposure for Line Operators on
                             Conveyorized Lines
Parameter/
Activity**
Units"
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 	 10
Immersion Tin 	 50

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

-------
3.2 EXPOSURE ASSESSMENT
Parameter/
Activity *
Units"
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 rain/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 3-16.


     Table 3-16. Parameter Values for Workplace Dermal Exposure for a Laboratory
   	Technician on Either Conveyorized or Non-Conveyorized Lines
    Parameter/
     Activity
 Units'
              Value
     Source of Data, Comments
 Exposure Time (ET)
 Chemical Bath
 Sampling
min/occur
HASL	
Nickel/Gold	
Nickel/Palladium/Gold
OSP 	
Immersion Silver	
Immersion Tin	
 15
 10
 1.5
, 22
 1.0
 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

-------
                                                          3.2 EXPOSURE ASSESSMENT
Results

       Table 3-17 presents results for estimating ADDs for inhalation and dermal workplace
exposure for line operators and laboratory technicians.

          Table 3-17. Estimated Average Daily Dose for Workplace Exposure
                         from Inhalation and Dermal Contact
Chemical
ADD8
(mg&g-day)
Inhalation
Line
Operator
Dermal
. -t 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
Fhioboric 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

-------
3.2 EXPOSURE ASSESSMENT
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
ADD*
(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
NA"
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

-------
                 3.2 EXPOSURE ASSESSMENT
Chemical
Inorganic metallic salt A
Inorganic metallic salt A (LADD) c
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
ADD8
(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
Ammo acid salt
Ammo carboxylic acid
Ammonia compound A
Ammonia compound B
Ammonium hydroxide
Citric acid
Copper sulfate pentahydrate
Ethoxylated alkylphenol
Ethylenediamine
Hydrochloric acid
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
5.32E-04
2.35E-01
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
4.14E-04
3.92E-01
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
'1.02E-05
9.63E-03
3-65

-------
3.2 EXPOSURE ASSESSMENT
Chemical


Hydrogen peroxide
lydroxyaryl acid
inorganic metallic salt B
Vf aleic acid
Vlalic acid
Nickel sulfate
Palladium salt
Phosphoric acid
Potassium compound
Potassium gold cyanide
Propionic acid
Sodium hydroxide
Sodium hypophosphite
Sodium salt
Substituted amine hydrochloride
Sulfuricacid
Surfactant
Transition metal salt
Urea compound B
ADD8
(mg/kg-day)
Inhalation
Line
Operator
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
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
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
Hydroxyaryl acid
Hydroxyaryl sulfonate
Phosphoric acid
Sodium hydroxide
7.79E-02
NA
NA
6.18E-06
NA
NA .
NA
NA
2.38E-02
NA
2.04E-03
1.92E-03
NA
NA
1.27E-03
NA
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
8.52E-04
3.00E-05
4.98E-02
1.67E-04
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.Q2E-04
NAb
1.52E-03
1.16E-03
5.57E-05
1.96E-06
3.25E-03 '
1.09E-05
                                     3-66

-------
                  3.2 EXPOSURE ASSESSMENT
- Chemical
* r /
.. ,>•
Sulfunc acid
ADD*
(mg/kg-day)
Inhalation
Line
Operator
NA
Dermal -
Line
Operator
2.55E-01
Laboratory
Technician
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

-------
3.2 EXPOSURE ASSESSMENT
Chemical
ADD*
(mg/kg-day) -t
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 monpbutyl 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
Alkylimine dialkanol
NA
NA
NA
NA
NA.
1.33E-03
3.17E-06
2.89E-07
1.33E-07
2.98E-06
2.32E-04
5.31E-07
5.05E-08
2.22E-08
5.17E-07
                                    3-68

-------
                                                                       3.2 EXPOSURE ASSESSMENT
Chemical
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
ADD"
(mg&g-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
Dermal
Line ,
Operator
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
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.
c 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 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.
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

-------
3.2 EXPOSURE ASSESSMENT
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:
                  PbB
        = PbB
where,
PbB
    adulIfCCntral   =
Pbs
BKSF
AFS
EFS
AT
central estimate of adult blood-lead concentrations Cug/dl)
typical background adult blood-lead concentration (//g/dl)
lead concentration G^g/g)
biokinetic slope factor G/g/dl)
intake rate (g/day)
gastrointestinal absorption factor (unitiess fraction)
exposure frequency (days/year)
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:
where,
PbBaduftjCentaa
GSDi>aduIt
 R
  •fetal/maternal
                                - PbBadult>central X GSDjiaduh X
95th percentile estimate of fetal blood-lead levels
central estimate of adult blood-lead concentrations G"g/dl)
estimated value of the individual geometric standard deviation
(dimensionless)
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 (PbBadult)Central). This represents the
 central estimate of blood-lead in adults  exposed to the HASL process, measured in ^g/dl.

       Background Blood-Lead Concentration (PbBadult,0).  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 ywg/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

-------
                                                              3.2 EXPOSURE ASSESSMENT
       Lead Concentration in Source (Pbs). This is an average estimate of the amount of lead
that is present in solder, and is measured in ywg/g.  For PWB facilities, the lead concentration of
solder was used instead of soil lead concentration. A value of 37,000 jiig/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 ,ug 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 09S).  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 //g Pb per. 100 cm2, and samples from solderer's
hands ranged from 3 to 32 //g Pb per 100 cm2.
                                          —

-------
3.2 EXPOSURE ASSESSMENT
       Individual Blood Lead Geometric Standard Deviation (GSDj). The GSDj 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
GSDj 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 (Rfetavmaternai)'
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 ,ug/dL, and of 3.2 to 102
for a fetus (Table 3-18). Estimated blood-lead levels will be compared to federal health-based
standards and guidelines in Section  3.4.

  Table 3-18. Estimated Concentration of Lead in Adult and Fetal Blood from Incidental
                          Ingestion of Lead in Tin/Lead Solder
Intake Rate
(rag/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).
PbBudui^jeatjai
(Atg/dl) *
2.0
14
63
1^*6^0,95

-------
                                                           3.2 EXPOSURE ASSESSMENT
                       Blood lead concentration vs intake rate
                 CQ
                        0         20        40
                               Intake rate (mg/day)
                                                       60
      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 jttg/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) (IR) (EF) (ED)/[(BW) (ATNC)]
where,
LADD =
ADD  =
Ca

IR
EF
ED
BW   =
             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 3-19 presents values used for these parameters. Results for general population inhalation
exposure are presented in Table 3-20.
                                         3-73

-------
3.2 EXPOSURE ASSESSMENT
   Table 3-19. 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)
A.TCAR
ATNf>
Units
mg/m3
m3/day
days/yr
years
kg
days
Value
: •.."•'•'• •'.'•' Source MDatai;Commenis,^^
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 3-20. Estimated Average Daily Dose for General Population Inhalation Exposure
Chemical *
ADD (mg/kg-dayXKii
IIASL, 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
* 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 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.
                                              3-74

-------
                                                             3.2 EXPOSURE ASSESSMENT
       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 (ffiUBK) 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 EEUBK model is 0.1 jig/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 flie ambient air concentration
of 0.00009 fig/m3 estimated from a HASL process (Section 3.2.3). The model was run at various
air concentrations down to 0.001 /tg/m3 (the model does not accept air concentration values less
than 0.001 jtig/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 3-21. 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 3-21. 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 0*g/dL) at Various Airborne Lead Concentrations
1 (jtg/m3 in air)
4.2
4.7
4.4
4.2
3.6
3.2
2.9
0,10ig/m3Jna!r)
4.1
4.5
4.2
4.0
3.4
3.0
2.7
0.010tg/m3inair)
4.1
4.5
4.2
4.0
3.4
2.9
2.7
0^001 Otg/m3 in air)
4.1
4.5
4.2
4.0
3.4
2.9
2.7
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
       formulations; possible effects of side reactions in the baths which were not considered;
       and questionnaire data with limited facility responses).
                                         3-75

-------
3.2 EXPOSURE ASSESSMENT
•      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:

 •     body weight;
 •      concentration of chemical in bath;
    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

-------
                                                               3.2 EXPOSURE ASSESSMENT
 •      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

       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
      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

-------
3.2 EXPOSURE ASSESSMENT
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,
 concentration of chemical in bath,  and the number of baths in a given process.  However,
    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

-------
                                                             3.2 EXPOSURE ASSESSMENT
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
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
       hi 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.

-------
                              3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
•      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 3-22 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 3-22. Available Carcinogenicity Information
Chemical Name8
Y1 •> / tl
"
Cancer Slope
Factor
(Inhalation Unit
Risk) •
Otg/m3)-1'-
Cancer Slope
Factor
(Oral)
(nag/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. c
IARC Group 1 e (IARC 1992).
EPA Class B2 f (IRIS, 1999); IARC
Group 2B ซ (IARC, 1987).
IARC Group 2B 8 (IARC 1974).
Possible human carcinogen. c
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 I (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 et al., 1994).
IARC Group 3 J (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

-------
33 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
Chemical Name *
Silver salt
Stannous methane
sulfonic acid
Tin chloride
Palladium chloride
Propionic acid
Cancer Slope
Factor
(Inhalation Unit
Risk)
Otg/m3)-1
ND
ND
ND
ND
ND
Cancer Slope
Factor
(Oral)
(mg/kg-day)4
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 pr IARC Group 3s
(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).
  Only those chemicals with available data or classifications are listed.
b The unit risk value is not reported here to protect confidential ingredient identity.
c 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).
* 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
(qt*) 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, qx* 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
/zg/m3 in air, or as risk per /ug/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

-------
                               3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
       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 3-23 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

-------
33 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
  Table 3-23. Summary of RfC and RfD Information Used in Risk Characterization for
                          Non-Proprietary Ingredients
Chemical
Name*
Ammonium
chloride,
Ammonium
lydroxide
ithylenediamine
Bthylene glycol
Bthylene glycol
monobutyl ether
Hydrochloric
acid
Leadf
Nickel sulfate
Phosphoric acid
Potassium gold
cyanide
Silver nitrate
Inhalation
RfCb
(mg/m3)
0.1 d (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
RฃDb
(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
heart weight and
hematologic 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 g (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

-------
                                  3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
Chemical
Name*
Stannous methane
sulfonic acid,
Tin, and
Tin chloride
Sulfunc acid -
Inhalation
RfCb/
(rag/m3)
ND
0.07 (HEAST)
Comments c
(Inhalation)

Acceptable air
concentration for
humans based on
respiratory effects
(U.S. EPA, 1997b).
Oral/Denna!
MfDb
(mg/kgMay) ,
0.6 J (HEAST)
NDk
Comments c
(Oral/Dermal)
Tin and inorganic
compounds: rats, 2 year,
histopathologic study
(U.S. EPA, 1997b).

 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.
 c 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 tabulations, 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.
 6 Value given represents a chronic inhalation minimum risk level (MRL). Although the test substance was nickel
 sulfate hexahydrate, the reported value is 0.0002 rag/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).
 ' A conversion factor is used in the risk calculations based on molecular weights of silver and silver nitrate.
3 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

-------
3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
       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 3-24 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 3-25. 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, LOAELs/NOAELs, or other data (e.g. TCLO) 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 3-26. A
summary of toxicity data available for the chemicals is presented in Table 3-27.
    14 The SAT is a group of expert scientists at EPA who evaluate the potential health and environmental hazards
 of new and existing chemicals.
                                           _              .                         _

-------
                                   3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
             Table 3-24.  NOAEL/LOAEL Values Used in Risk Characterization
Chemical
Name*
Acetic acid
Copper ion,
Copper sulfate
pentahydrate
Ethylenediamine
Ethylene glycol
Hydrogen
peroxide
Lead1
Propionic acid
Inhalation
NOAEL/
LOAEL b

-------
3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
        Table 3-25. Developmental Toxicity Values Used in Risk Characterization
                              for Non-Proprietary Ingredients
Chemical*
Ammonium
chloride
Copper ion,
Copper sulfate
pentahydrate
Ethyl-
enediamine
Ethylene glycol
Ethylene glycol
monobutyl ether
Developmental
Inhalation
NOAEL/
LOAEL
(mg/m3)b
ND
ND
ND
150 (N)
ND
Comments'
(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 "
(mg/kg-day)
1,691 (N)
3(L)ฐ
470 (L)
500 (N)
100 (N)
Comments*,
(Oral/Dermal)
? *
&•
vlice, drinking water, after gd
d 7, no congenital effects
(Shepard, 1986).
Copper: mink, diet, increased
mortality (Aulerich et al.,
1982; ATSDR, 1990a).
Hats, gd 6-15 diet, resorption,
unpaired growth, missing or
shortened innominate arteries,
and delayed ossification of
cervical vertebrae or
phalanges (DePass et al.,
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).
 " 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.
 c Comments may include test effects, test animal, duration during time of gestation, exposure route, and source of
 data.
 d gd = gestation day.
 c Conversion factors are hi the risk calculations based on molecular weights of copper ion and copper sulfate
 pentahydrate.
 ND: No data available.
                                             3-88

-------
                                 3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
                    Table 3-26.  Summary of Health Effects Information
                            (from Structure-Activity Team Reports)	
     Chemical
             SAT Health Effects Pertaining to
              Dermal or Inhalation Exposure
                                                                                 Overall
                                                                              Concern JLevel
 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 neurotoxieity
 (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
neurotoxieity.  A mixture containing the dimethyl ester of this
compound was tested hi 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 neurotoxieity 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
 :helate 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
 leart. Chelation of zinc may cause irnmunotoxicity
 retardation of wound healing). This compound is expected to
 )e irritating to all exposed tissues and may be a dermal
sensitizer. A salt of this compound caused developmental
 sffects in rats.  There is concern for oncogenicity and kidney
:oxicity. There is also a potential for male reproductive effects.
 [his compound may be mutagenic.
Low moderate
                                            3-89

-------
3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
     Chemical
             SAT HealtK 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;
 lowever, absorption is expected to be moderate through the
 skin when in solution. There is concern for developmental
 oxicity 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
 :ye irritant. It has low acute toxicity. Another analog was
tested hi 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
 sxpected 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 LD5Q."
 Moderate, based
 on irritation
                                              3-90

-------
                                3.3 HUMAN HEALTH AMD ECOLOGICAL HAZARDS SUMMARY
     Chemical
             SAT Health Effects Pertaining to
              Dermal or Inhalation Exposure
                                                                                  Overall
                                                                               Concern Level
Citric acid
Expect poor absorption by skin, but expect absorption by lungs
and GI tract.  No health concerns identified.
                                                                              Low
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 hi 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
                                             3-91

-------
3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
     Chemical
             SAT Health Effects Pertaining to
             Dermal or Inhalation Exposure
    Overall
 Concern Level
 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
 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
                                             3-92

-------
                                 3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARF
    Chemical
             SAT Health Effects Pertaining to
              Dermal or Inhalation Exposure
                                                                                  Overall
                                                                               Concern Level
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
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 3-27. Overview of Available Toxicitv 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
Cancer:
Slope Factor (SF),
Weigfrt-qf-
Evidence - -
(WOE)
Classification
















Inhalation:
BfC,NOAEL,
j>rฃOAELa




Yes








Yes


Oral/Dermal:
Rfl>,NOAEL,
orLOAEfc8
-5ฃ,v, -^
/

NOAEL
Yes

Yes


Yes





Yes


SAT
Rank
•


•


•

•
•

•
•

•
•
                                             3-93

-------
3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
Chemical
Alkylphenol
polyethoxyethanol
Alkylpolyol
Amino acid salt
Amino carboxylic acid
Anunonium chloride
Ammonia compound A
Ammonia compound B
Ammonium hydroxide
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
Cancer:
Slope Factor (SF),
Weight-of-
Evidence
(WOE)
Classification








Inhalation:
RfC,NQAEL,
or LOAEL "




RfC (for ammonia)
RfC (for ammonia)
RfC (for ammonia)
RfC (for ammonia)
Oral/Dermal:
RfD, NOAEL,
orLOAEL8

Yes .

Yes
D-NOAEL
RED (for ammonium
sulfamate)
Yes
Yes
RfD (for ammonium
sulfamate)
,SAT-
Rank
•

•



•

Not enough information to identify a specific chemical.



WOE (for copper)
WOE (for copper)
WOE (for copper)








WOE
WOE



SF, WOE





LOAEL
Yes
LOAEL
Yes

NOAEL
LOAEL; D-
NOAEL
RfC



RfC
Other"
Yes


Yes
Yes
Yes
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
•
•
•



•
•



•
•



•
•
•



                                  3-94

-------
3.3 HUMAN HEALTH AND ECOLOGICAJL HAZARDS SUMMARY
Chemical
Lead
Maieic acid
Malic acid c
Methane sulfonic acid
Nickel sulfate
Nitrogen acid
Nonionic surfactant
Palladium chloride
Palladium salt
Phosphoric acid
Potassium compound
Potassium gold cyanide
Potassium peroxymonosulfate
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 g
Stannous methane sulfonic
acid
Substituted amine
hydrochloride
Sulfuric acid
Surfactant
Thiourea
Tin
Tin chloride
Transition metal salt
Unspecified tartrate
Cancer:
Slope Factor (SF),
Weight-of-
Evidence
(WOE)
Classification
WOE



WOE (for nickel
dust)

Inhalation:
RfC,NOAEL,
orLOAEL*
Other b



MRLd

Oral/Dermal:
RfD,NOAEL,
orLOAEL"
Other b



RfD

Not enough information to identify specific chemical.
Some data (for Pd)
Some data (for Pd)




Some data

WOE (for silver)
WOE (for silver)



-"


WOE

WOE


RfC



Other c











Other c


ADIe

RfDf

NOAEL

Yes "
RfD (for silver)






RfD (for tin)


Not enough information to identify specific chemical.
WOE

WOE








RfD
RfD

Yes
SAT
Rank

•
•







•

•

•


•

•
•
•


•





•

            3-95

-------
33 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
Chemical





Urea
Urea compound B
Urea compound C
Vinyl polymer
Cancer:
Slope Factor (SF),
Weigbt-of-
Evidence
(WOE)
Classification

WOE

WOE
.Inhalation:
RfC, NOAEL,
or LOAEL a

-





Oral/Dermal:
RfD, NOAEL,
or LOAEL'


'


Yes
Yes
SAT
Rank



-




  "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 3-23 and 3-24 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.
' 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).
8 Not generally considered poisonous to humans or animals.
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 3-28 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.
                                            3-96

-------
                           3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
Table 3-28. Estimated (Lowest) Aquatic Toxicity Values and Concern Concentrations for
   PWB Surf ace 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
Alkylammo 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 In soft water
In hard water
Copper ion
Copper salt C
Copper sulfate pentahydrate
Cyclic amide
Ethoxylated alkylphenol
Ethylenediamine
Ethylene glycol
Acute (a) Toxicity
(mg/L)
Fish
0.5
79
Invert

65
Algae


Chronic (c) Toxicity
(mg/L)
Fish
0.08

Invert


Algae


data omitted a
data omitted8
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
2
2
5
data omitted a
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
26.5
6,900
>100
31,000

5,400
0.16
710
8.3
440
Concern
Concentration
(mg/L)
0.008 (c)
0.65 (a)
0.5 - 1 (a)
1- 5 (c)
5 - 10 (c)
>1 (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)
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)
                                      3-97

-------
3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
Chemical

Ethylene glycol monobutyl
ether"
Fatty araine
Fluoboric acid
Gum
Hydrochloric acid
Hydrogen peroxide
Hydroxyaryl acid
Hydroxyaryl sulfonate
Sydroxy carboxylic acid
[norganic metallic salt A
[norganic 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
Acute (a) Toxicity
(mg/L)
Fish
116
Invert
89
Algae
620
Chronic (c) ToxicUy
(mg/L) ,
Fish
10
Invert,
3.9
Algae
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 omitted3
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
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
Concern
Concentration
(mg/L)
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)
l-5(c)
l-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)
l-5(c)
4.7 (c)
, l-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)
                                   3-98

-------
                                3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
Chemical
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
AcB*e(a)Toxicityx
(mg/L)
Fish
7
.Invert
140
Algae
<8
Chronic (c) Toxicity
(mg/L)
Fish
0.2
Invert
0.9
Algae
<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 omitted3
data omitted a
data omitted *
Concern'
Concentration
.. (mg/L)
0.02 (c)
0.01 - 0.05 (c)
224 (c)
0.03(c)
0.007 (c)
0.04 (c)
<1 - 5 (c)
l-5(c)
>10 (c)
0.01 - 0.05 (c)
0.01 - 0.05 (c)
>1 - 5
2 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 -f UP
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 of ten (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.
                                            3-99

-------
33 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
       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 3-29 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 ฃ  10 mg/L	Moderate concern
       > 10 mg/L	Low concern

For acute values:
       < 1 mg/L	High concern
       > 1 to ฃ 100 mg/L	Moderate concern
       > 100 mg/L	Low concern

Chronic toxicity ranking takes precedence over the acute ranking.

       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 3-29 with Medium to High
hazard concern levels can help avoid potential problems.

        Table 3-29. 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
Lowest Acute (a) or
Chronic (c) Value
(mg/L)
0.08 (c)
65 (a)
MR
NR
NR
NR
NR
Hazard
Bank8
H
L
L
L
L
L
L
                                          3-100

-------
3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
Chemical
Aliphatic dicarboxylic acid C
Alkylalkyne diol .
Alkylammo acid A
Alkylaimnb acid B
Alkylaryl iniidazole
Alkylaryl sulfonate
Alkyldiol
Alkylimine dialkanol
Alkylphenol ethoxylate
Alkylphehol polyethoxyethanol
Alkylpolyol
Amino acid salt
Ammo 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
Ethylenediamine
Ethylene glycol
Ethylene glycol monobutyl ether c
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
Lowest Acute (a) or
Chronic (c) Value
(mg/L)
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
1 (c)
0.14 (a)
NR
0.001(c)
NR
NR
0.16 (c)
440 (c)
3.9 (c)
NR
1.4 (c)
NR
15 (c)
1.7 (a)
NR
NR
NR
NR
NR
Hazard
Rank"
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
M
L
M
H
M
L
M
M
M
L
L
H
H
          3-101

-------
33 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
Chemical
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
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)
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)
0.3 (c)
0.07 (c)
0.4 (c)
NR
NR
>100 (c)
NR
NR
NR
Hazard
Rank8
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
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.
 c Diethylene glycol monobutyl ether reviewed instead; both chemicals are very similar.
 NR: Not reported in order to protect confidential ingredient identity.
                                              3-102

-------
                               3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
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.

       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 CCis 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
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

-------
                                                              3.4 RISK CHARACTERIZATION
       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, MSDSl, 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, 1991a):
where,
I
Cm   =
b
h
                                       I = (Cm)(b)(h)
daily inhalation potential dose rate (mg/day)
airborne concentration of substance (mg/m3)
inhalation rate (m3/hr)
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:
      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.

   1  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 mere 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:~

-------
3.4 RISK CHARACTERIZATION
For carcinogens:
For non-carcinogens:
                          LADD = (I)(EF)(ED)/[(BW)(ATCAR)]
                           ADD =  (I)(EF)(ED)/[(BW)(ATNC)]
where,
LADD =
ADD  =
EF    =
ED    =
BW   =
             lifetime average daily dose (mg/kg-day)
             average daily dose (mg/kg-day)
             exposure frequency (days/year)
             exposure duration (years)
             body weight (kg)
             averaging time for carcinogenic effects (days)
             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:
   17 This version of the ISCLT model is provided as part of the Risk* Assistant™ 2.0 software package
(Hampshire Research Institute, 1995).
                                ,         -_                    ___       _

-------
                                                           3.4 RISK CHARACTERIZATION
where,
D
S
c
f
h
                                 D  = (S)(C)(f)(h)(0.001)
dermal potential dose rate (mg/day)
surface area of contact (cm2)
concentration of chemical in the bath (mg/L)
flux through skin (cm/hour)
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 froni 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 j 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

-------
3.4 RISK CHARACTERIZATION
•      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;
•      frequency of chemical bath replacements;
   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

-------
                                                           3.4 RISK CHARACTERIZATION
•      number of baths in a given process; and
•      bath size.                                p-

       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 hi 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 3-28 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 CGs 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 3-29):
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 RISK CHARACTERIZATION
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
LADD
                    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 hi addition to the already-existing background risk
                    of an individual contracting cancer from all other causes.)
                    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 (ql *) 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:
                                         3-110

-------
                                                          3.4 RISK CHARACTERIZATION
                                   HQ =  ADD/RfD
where,
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 of ten 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 levelof concern.

       For chemicals where an RfD or RfC was not available, an MOE was calculated by:

                         MOE = NOAEL/ADDorLOAEL/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:
             RfD
                 DER
where,
SFDER
GI absorption
             (RfDORAL) (GI absorption)
             (NOAEL or LOAELORAL) (GI absorption)
             (SFORAL)/(GI absorption)
reference dose adjusted for dermal exposure (mg/kg-day)
NOAEL or LOAEL adjusted for dermal exposure (mg/kg-day)
cancer slope factor adjusted for dermal exposure (mg/kg-day)"1
gastrointestinal absorption efficiency
                                         3-111

-------
3.4 RISK CHARACTERIZATION
       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 3-30 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.)
Table 3-30. Gastrointestinal (GI) Absorption Factors
Chemicals *
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 8
Inorganic metallic salt C
Nickel sulfate
Phosphoric acid
Potassium gold cyanide
Propionic acid
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
:.•-„.,. ''v'VxvM-Sourre
chemical profile b
chemical profile b
NR
assumption c
NR
assumption ฐ
assumption c
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 ฐ
assumption c
NR
NR
NR
midpoint of range, 0.01-0.1,
chemical profile
U.S. EPA, 1995
assumption c
assumption c
                                         3-112

-------
                                                              3.4 RISK CHARACTERIZATION
-:vV^.^ฃh^^
Silver nitrate
Silver salt
Stannous methane sulfonic acid
Tin chloride
Unspecified tartrate
Urea compound C
Vinyl polymer
'^^OI^MTP^f.'f
0.08
MR
0.2
0.5
0.5
0.2
0.1
KKma-&:D/^r;:Souirce '
midpoint of range, 0.05 - 0.1
(U.S. EPA, 1991c; ATSDR, 1990b)
NR
assumption ฐ
Johnson and Greger, 1982
chemical profile b
assumption c
chemical profile b
" 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.
c 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
    19 A cancer classification of known human carcinogen has been assigned by either the EPA, I ARC, 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

-------
3.4 RISK CHARACTERIZATION
       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
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
lexicological 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 hi 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.
                                         3-114

-------
                                                           3.4 RISK CHARACTERIZATION
       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 exprlss probability, only the ratio of the estimated dose to
the RfD or RfC, and it is not necessarily linear (an HQ of ten 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 3-31.  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 3-32. 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 3-33 provides a summary of the potential health effects for the chemicals of
concern listed in Tables 3-31 and 3-32. It should be noted that Tables 3-31 and 3-32 do not
include chemicals for which toxicity data were unavailable. Table 3-34 lists chemicals where
inhalation or dermal exposure is expected to occur, but appropriate toxicity values are not
available.  (Table 3-26 provides qualitative structure-activity information pertaining to chemical
toxicity for those chemicals without available measured toxicity data.)
                                         3-115

-------
3.4 RISK CHARACTERIZATION
i
*ง
0
e
o>
•*•*
o
M
"33
02
1

Inhalation Exposi
a
1

Risks From Occu
r— -i
1
i
s
ffi
*ฉ
^"
cs
ง
OS

?o
to
0>
1









1
3
1
d
03










0* Q
5^




2
o
Nickel/Palladium
(NC)


tj
If
i


5_G>
2 P^
S3




e
8.
i
Chemical of C






g


line operator



2
1
1







<;
Iz;



1
S
1

1
.S pq

llcfl







o7
C-
3
ง

c H
W ^^

ง ซO t_J



Ethylene glycol






g
i-H
r.
tH
1
1
gj^r
co1
of
r-T
V~'
|

-------
                                                            3.4 RISK CHARACTE1OZAT/O2V
g
B^
^1
< 'a
o ง
2 ,2
ง1
T-t ฃซ
u S
1|
hs •ง
ง |
I
Si
i
cl
•S
!
''S
CO
8-
to
. 1
3
•5
-22
1 ' o f
'1
a
12
s
JO
en
2
- • |
'S
a
                                             0)
                                             00
j'l
o ,ซ
OJ 42
g'l
•81
ซ
5 [,
o B

:sults are reported (HQ, MOE, o
ical is used. These are only sho'
a a
indicator for which
s) in which the chei
nical is used:
III
*o M -S
3 >C 2 -a
J W H ฃ -g
<* pj .a
2
a
1
•ง
Hi
't'^
.S s
B *•"*
•6 ป
O O
ฃd
S B
2 S
t—< G
O ง
cd ซfi
•~ ,s
o S
ง3
ill
•2 to .
>-< .S *~*
1? a ill
c so -a fe . S |-J "
ซ s c i ^ 8 u
ง3 .2 .2 •ซ -c B w
'S'o 2 S2- 'eป- a'>H to toCu.5 S too
X_^^.VX_,^-S0--3W
ri^ssS-^ci o s? O
s" 3-2
1 |-|c2 g
is ง ^ !ง "a 1
Ul |1 fa
ei-s ^,2lง . .
tgifistSB^ss
s ง g | .b a -3 "^..a se S
es'o'ei3.Se/-^^&2(D'S<
^ co  •— i ฐ'oto ^ ea
• J3 to Q} ^^ Q
^ >ป >^ 03 ^i

U Q W J3 ^
                                         3-117

-------
3.4 RISK CHARACTERIZATION


J9
.2

G
0>
^
1
•3

o
ซu
u
en
O
i
*•*
a
3
o>
Q
•ซ
1
a
I
a
o
ฃ
S
ซ
I
1
w
i
ฃ
g?
ss
g
3
CC
ro
s
















^
<•
I
"3
1
a
Health Risk
g
gj
a








s
o
w
JO ^ tf>

J** ^*J



ftl^-s
O
O




II



.•e
S
Nickel/
Pa!Iad!unt/(
(NC)
-
*o
if
1
z
J_
S—


Sฎ
gs

ซ , • •
Chemical of Concent






g




<;
%,




<;
I2H


line operator



g


g



g

(Ammonia compound A






g




<;
is




<;
J25


^
s s
"ert
&
o
1
SIS


g



g

Ammonium chloride






g




<;
%Z




<;
J2
@ 0
V~/ Is
1
as
@ s
*— >
i
u
w
as


g



^

Ammonium hydroxide






g
ง|

„, g .ง H
o "i- ~- cj
SI >-<•ซ* >j
ง
^
4=
O 4)
wi ""1 ta ca 52
O vง |Lo"C!
21 O O r- < I-J


^



g


g



g

Copper ion






g




<;
Is


I
r*>.
0
J



g



g


•g



g '.

Copper salt C

ง



o
2
1
1,1
o *-•
•s-^
oc2"o
 ^q*
OH O
O O
15


g



g
CQ
CO
.y
1
!






g




<
&




<
&
S o
^1
ป [ง
^ ^
S
h
•ja ^
SIงS:


g



g

Nickel sulfate


03
S3
o.
0
1




<
•z




<;
'Z


<



g


g



g

Urea compound C



































                                     3-118

-------
                                                    3.4 RISK CHARACTERIZATION
*" ' •
d-S-
                                       g
.2 .a
  -ง
  31
ซ
•3 g.9
                                     3-119

-------
3.4 RISK CHARACTERIZATION
       For inhalation exposure to workers, the following chemicals result in an HQ greater than
one or an MOE below the concern levels:

•      ethylene glycol hi 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 salt B, 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

-------
                                                  3.4 RISK CHARACTERIZATION
Table 3-33. Summary of Potential Human Health Effects for Chemicals 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
> PotentialKealtfi 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 ah- 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

-------
3.4 RISK CHARACTERIZATION
Table 3-34. Data Gaps for Chronic Non-Cancer Health Effects for Workers
Chemical
Inhalation " or Dermal b
Exposure Potential
SAT Rank
(if available)
HASL
1,4-Butenediol
Alkylaryl sulfonate
Arylphenol
Huoboric acid
Hydrochloric acid
Sodium hydroxide
Sulfuric acid
Tin
Inhalation and Dermal
Inhalation
Inhalation
Dermal
Dermal
Dermal
Dermal
Dermal
Low-moderate
Low
Moderate





NickeVGoId
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
Sodium hypophosphite monohydrate
Inhalation
Inhalation and Dermal
Inhalation
Inhalation
Inhalation
Dermal
Inhalation
Dermal
Inhalation and Dermal
Dermal
Inhalation
Moderate

Low-moderate

Moderate-high

Low-moderate

Low

Low-moderate
                                   3-122

-------
                                                             3.4 RISK CHARACTERIZATION
Chemical
Sulfuric acid
Urea compound B
Inhalation ฐ or Dermal b
Exposure Potential
Dermal
Inhalation and Dermal
SAT Rank
(if available)


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 Isulfonic acid
Sulfuric acid
Thiourea
Urea compound C
Inhalation
Dermal
Dermal
Dermal
Dermal
Dermal
Inhalation
Low






3 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

-------
3.4 RISK CHARACTERIZATION
       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"11.  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.  AnMOEwas
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"""

-------
                                                           3.4 RISK CHARACTERIZATION
       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 3-35 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 hi tin-lead solder used in the HASL process. For workers, the
lowest federal target or action levels are from OSHA and ACGIH, at 30 jttg/dL in blood. By
comparison, the 5 to 12 jig/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 ftg/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.  .
   22 10 mg/day is an average estimate; 50 mg/day is a central tendency estimate.
                                          -_

-------
3.4 RISK CHARACTERIZATION









h4
eg
*ป•*
frrf
W
S
s
8
JO
1
>
#
f\^
•
in
CO
Table 3-








S
1
o
g
s
&
$?
Comparable Lead ฃ
"3
** S
fe'rs
1 2

*t3
fa



ซ
fu
>^
O




Occupational blood-lead
monitoring data.
1-1 * /-N
a ง*ฅ
=s-^3 S
S S^
• S4 S
cs -a **


T^
Modeled (ALM) blood-leai
data for an adult worker.

J
"Sb
3,
o
CO

o
T— 1
"Bb
SL
O
CO



c3
o
10 wish to b
f
iv

O
O
'ง>
a.
o
VO
•a
3
I
*3
•o
j maintaine
NIOSH, level to b<
air concentrations


•a M
ซ ^3
i O
Q ฃ•
i|
11
If
J ง1
1) g>Jj ^
ง|| I
|  fiQ
o •ง
& 13
C -O CG
eo o 13

"s
•ง>
=5.
CO



Rl
2
Workplace air monitoring
data (average of HASL
process area monitoring da
provided by one PWB
manufacturer).

"S
I)
a.
o
VI

a.
0






s-^
OSHAPEL(8hr1
1
o
o






f
c\
I— 1
ffi
CO
NIOSH REL (NIC
I
O


^_^
So
o\
o\
*•"•*
1
**
ACGIHTLV TW^
 ง .s ซ
•o P Ti -^  CO 51
Sj Cj 1 Hjrf ป—J |^\
Not determined. TheffiUI
blood- lead levels for child
years. However, estimated
concentration from a HASI
times lower than the defaul
model. BEUBK model resu
values range from 2.7 to 4.'.

"s
"Sb

o
T-H

"Sb
o







H







c^
O
Tib
a.
o
CO






E







t'
o
CN

'S
^_4
>
^^
TlJ
13
0
(*) ^ฎ
International: WH(
concern (WHO, 19
B
O
•a
o
S^ *^3
i^
J o
ll
3 rt '































                                    3-126

-------
               3.4 RISK CHARACTERIZATION
c S
ง ซ
> ง

o -a
"a J
c "S
o o
'is
S 
S sg
o g
*~* Q>
to
03 O
JS *ฐ
.S2 S
T3 "S
o S
3 fi
(D (U
^j -D
ฃ• ^3
•<"* Q
1|
*s .
jl
(U CS
-3 A
t*-( U)
il
J^ T3
s o
Is CDC considers children to have an elevated level i
remediation should be done for all children with bio
































>45|tg/dL(RTI,1999).
NOTES:



























1
4
g
•a
^
1
s
-o
\ ACGIH: American Conference of Governmental In
































1 CDC: Centers for Disease Control and Prevention.





























































1
& .
•ง_ง
CS ^S
EPA: U.S. Environmental Protection Agency.
NIOSH: National Institute for Occupational Safety
OSHA: Occupational Safety and Health Administra
WHO: World Health Organization.
































I PEL: Permissible Exposure Limit.
































1 REL: Recommended Exposure Level.



























-




i
i
"8
I
"53
ฃ
































J TLV: Threshold limit value.
































1 ALM: Adult Lead Methodology.




J
•o .
S

i
8
t— t
s
1
g1 • . .
"S
.ง
2
0,
a
.53
1 ' . '
rO
eo
8
*—< '
"O
o
o
3
It
6| '
"^ i>
3iง
-~ <3
y ^^
o 1— *
11
||
^ o
II
tS H
•g .J '
2 "ง>
So =t
e ซ
gg -*^
3-127

-------
3.4 RISK CHARACTERIZATION
       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 #g/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 /ig/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 ptg/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 unitiess ratio:
where,
Csw   =
CC    =
                                                '
              estimated surface water concentration following treatment in a POTW (mg/1)
              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
fig/dL, Medical evaluation and environmental remediation should be done for all children with blood-lead levels
^20 /ig/dL. Medical treatment may be necessary in children if the blood-lead concentration is > 45 jig/dL (RTJ,
1999).
   24
     Results from both personal monitoring for HASL line operators and air samples from the HASL process area
were averaged.


-------
                                                          3.4 RISK CHARACTERIZATION
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 3-36.
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 conveyorized configurations of the OSP
       process;
•      hydrogen peroxide in the conveyorized 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 3-36. 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
(€)
NA
NA
0.7 - 3.5
1.5
6.1
NA
OSP
(NC)
NA -
6.6 - 33
NA
NA
NA
NA
OSP
.CC)
NA
3.6-18
NA
NA
NA
NA
Imm.
Sliver (C)
NA
NA
NA
1.3
NA
NA
faun. Tin
(NC)
NA
NA
NA
NA
3.6
1.0 a
 Estimated surface water concentration is equal to the CC;
NA: Not applicable; estimated surface water concentration
that process configuration.
NC: Non-conveyorized.
C:  Conveyorized.
this is not counted as an exceedance.
is less than CC or the chemical is not an ingredient of
       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 3-37. 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

-------
3.4 RISK CHARACTERIZATION
          Table 3-37. 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 (RIj-co)
HASL
(NO;,
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

-------
                                                            3.4 RISK CHARACTERIZATION
       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 (effect's may be independent,
       additive, synergistic,  or antagonistic); and
 •      possible effects of substances not evaluated because of a lack of chronic/subchronic
       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 conies 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 lexicological
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 3-26 and 3-27).
Uncertainties in dermal risk estimates also stem from the use of default values for missing
                                          3-131

-------
3.4 RISK CHARACTERIZATION
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 S ARs 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

-------
                                                           3.4 RISK CHARACTERIZATION
       The Exposure Assessment for this risk characterization, whenever possible, used a
 combination of central tendency and high-end assuriiptidns, 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
 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.,
 conveyorized immersion silver, conveyorized and non-conveyorized immersion tin, and
 conveyorized 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 (oiie 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.
     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

-------
3.4 RISK CHARACTERIZATION
       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 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 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
                                         3-134

-------
                                                            3.4 RISK CHARACTERIZATION
has been 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 conveyorized OSP process, one in the conveyorized immersion silver process, and one in
the non-conveyorized immersion tin process.

Overall Risk Screening and Comparison Summary

       Table 3-38 presents an overall comparison of potential human health and ecological risks
for the baseline (non-conveyorized HASL) and the alternative process configurations.
     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          :           ',               "~

-------
3.4 RISK CHARACTERIZATION
 Table 3-38.  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*
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
Gapsd
3
0
10
9
2
0
0
2
0
Dermal
Data
Gaps8
6
6
8
7
5
5
4
5
5
Aquatic
Concern f
4
3
0
0
1
1
1
1
0
  The number of chemicals with an EPA cancer WOE of A, Bl, or B2, or an IARC WOE of 1,2A, or 2B (see Table
3-22).
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 3-31 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 3-31 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 3-34).
c 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 3-34).
f The number of chemicals for which the ecological risk indicators exceeds the concern level (i.e., RIreo > 1.0). See
Table 3-36 for detailed results.
                                               3-136

-------
                                                      3.5 PROCESS SAFETY ASSESSMENT
 3.5    PROCESS SAFETY ASSESSMENT
                                                                           i
       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)3, 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

-------
3.5 PROCESS SAFETY ASSESSMENT
       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 3-39 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

-------
                                                           3.5 PROCESS SAFETY ASSESSMENT
       Table 3-39.  Flammable, Combustible, Explosive, and Fire Hazard Possibilities
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
-Hazardoiis^opeily *ป b
Flammable
1(3)
1(3)


Combustible




Explosive
1(1)

1(1)
1(4)
r^Fire
. Hazard
KD
2(3)
2(3)
1(1)

       	—	______ _ป ซ..v ซ.v>uu ...ป_ฃ, ซ. V.M..U.M.UI,  ,, VM. ^r*. uuuwu ,II..LW/UJJI^ vyvj-LifT. Uw.LuU.u.tjtll i\~fi Uiw tlM. Vdl XICLฃiCUvlUUo
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, 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.
c 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 ฐF (37.8 ฐC), except any mixture having
       components with flashpoints of 100 ฐF (37.8 ฐC) 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.
                                            3-139

-------
3.5 PROCESS SAFETY ASSESSMENT
       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.

Combustible Liquid - As defined by OSHA [29 CFR 1910.1200(c)], a liquid that is considered
combustible has a flashpoint at or above 100 ฐF (37.8 ฐC), but below 200 ฐF (93.3 ฐC), except
any mixture having components with flashpoints of 200 ฐF (93.3 ฐC), 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 then-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 3-40. 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

-------
                                                             3.5 PROCESS SAFETY ASSESSMENT
    Table 3-40. Corrosive, Oxidizer, Reactive, Unstable, and Sudden Release of Pressure
Surface Finishing
Process
HASLC
Nickel/Gold"
(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 c
Microetch
Predip
[mmersion Tin
Hazardous Property %b
Corrosiv
e
1(1)
3(4)
KD
3(4)
3(3)
KD
1(1)
3(4)
3(3)
1(4)
3(3)
1(3)
KD
3(4)
KD
1(3)
1(2)
2(2)
KD
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)


                	__ _	..	^ _________  .. __. ฃ,._. vv__vvu .M.WUU^ *~rป~r*.J-l X \_-WJ. 14.* JLblA_*._.i J.VSJ. U1W glV^/ll HClXiCUUULlo
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.
c 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
                                             3-141

-------
3.5 PROCESS SAFETY ASSESSMENT
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 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 then-
concentrated form is presented in Table 3-41.  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 3-41, 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 mat 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

-------
                                                               3.5 PROCESS SAFETY ASSESSMENT
  Table 3-41. Sensitizer, Acute and Chronic Health Hazards, and Irreversible Eye Damage
                          Possibilities for Surface Finishing Processes
Surface Finishing
Process
HASLC
NickeVGold 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 % 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)
KD
2(2)
2(2)
KD
3(4)
2(3)
4(4)
3(3)
2(3)
1(2)
KD
3(3)
10)
2(3)
1(2)
1(2)
1(4)
Chronic
Health
Hazard
1(1)
3(3)
1(1)
2(2)
1(2)
KD
2(2)
2(2)
1(1)
1(4)
1(3)
2(4)
2(3)
1(3)
1(2)
KD
3(3)
1(1)
2(3)
1(2)
1(2)
1(4)
Carcinogen
1(1)
1(1)
1(2)
2(4)
1(3)


KD
Irreversible
Eye Damage
KD
3(4)
1(1)
3(4)
1(2)
1(1)
1(2)
3(4)
1(3)
1(4)
2(3)
3(3)
KD
3(4)
KD
2(3)
1(2)
2(2)
KD
2(4)
  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, 2(4) means that three of the five products in the bath were classified as
sensiti/ers 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 die 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

-------
3.5 PROCESS SAFETY ASSESSMENT
       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 3-41 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

-------
                                                       3.5 PROCESS SAFETY ASSESSMENT
•      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 130ฐ F);
•      stored as a compressed gas;
                                         3-145

-------
3.5 PROCESS SAFETY ASSESSMENT
•      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 500ฐ 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 hi 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

-------
                                                      3.5 PROCESS SAFETY ASSESSMENT
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:
                                         3-147

-------
3.5 PROCESS SAFETY ASSESSMENT
•      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;
•      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 CFR Part 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.
                                         3-148

-------
                                                      3.5 PROCESS SAFETY ASSESSMENT
       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 on
       clothing;
•      gloves to prevent dermal exposure while operating the process; and
•      boots to protect against chemical spills.

       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 applies 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

-------
3.5 PROCESS SAFETY ASSESSMENT
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.
Occupational Noise Exposure

       OSHA has also developed standards (29 CFR Part 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
                                    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." /. 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

-------
REFERENCES	'               	

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., R Iverson, E.A. Nera and E. Lok. 1991. Tide not available. Mutat Res. Vol.
       248, p.321 (cited in Katz and Guest, 1994, HSDB, 1998).

Decisioneering, Inc. 1993. Crystal Ballฎ 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., MJ. Inskip, C.L. Baccanale, EJ. O'Flaherty, W.I. Manton, D.L. Schanzer, J.
       Blenkinsop and C.M. Edwards. 1995. "Transplacental Transfer of Lead in Non-Human
       Primates (Macaco, 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." /. 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

-------
                                                                        REFERENCES
Graziano, J.H., D. Popovac, P. Factor-Litvak, 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.  "Toxicological 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

-------
REFERENCES	_________^	

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

-------
                                                                        REFERENCES
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, PartF,
       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. Dracker. 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." JNutr.
       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

-------
REFERENCES	

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).

SoUmann, T. 1921. Title not available. /. Pharmacol Exp. Ther.. 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
       (EFDBฉ).  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.

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.
                                        3-156

-------
	___	REFERENCES

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
      lEUBK 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.

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.
                                        3-157

-------
REFERENCES
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 (HF 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

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
       85/85:  85ฐC and 85% relative humidity for three weeks),
       thermal shock (TS):  200 cycles between-50 ฐC and 125ฐC, 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 hi 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

-------
                                           4.1 PERFORMANCE DEMONSTRATION RESULTS
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  85ฐC 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.
                                          4-3

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
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
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 hi 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.
                                          4-4

-------
                                           4.1 PERFORMANCE DEMONSTRATION RESULTS
       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.
•      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

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
        Table 4-1. Electrical Responses for the Test PWA and Acceptance Criteria
Electrical
Response
Circuitry
Acceptance Criteria 5
# " ** *
High Current Low Voltage
1
2
HCLVPTH
HCLV SMT
Change in voltage from pre-test < 0.50V
Change in voltage from pre-test < 0.50V
High Voltage Low Current
3
4
HVLCPTH
HVLCSMT
4/i A < 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)
HFSMTSOMHz
HFSMTf(-3dB)
HFSMTf(-40dB)
Within ฑ 5dB of pre-test
Within ฑ 50MHz of pre-test
Within ฑ 50MHz of pre-test
Within ฑ 5dB of pre-test
Within ฑ 50MHz of pre-test
Within ฑ 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 + [ฑ?] 5dB of pre-test
Within ฑ 5dB of pre-test
Within ฑ 5dB of pre-test
Within ฑ 50MHz of pre-test
< lOdB increase over pre-test
Other Networks: Leakage
18
19
20
21
10-mil Pads
PGA-A
PGA-B
Gull Wing
Resistance > 7.7 Iog10 ohms
Resistance > 7.7 Iog10 ohms
Resistance > 7.7.1og10 ohms
Resistance > 7.7 Iog10 ohms
Stranded Wire
22
23
Stranded Wire 1
Stranded Wke 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

-------
                                           4.1 PERFORMANCE DEMONSTRATION RESULTS
       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.
                                                                       i
 Table 4-2. Distribution of the Number of LRSTF PWAs by Surface Finish, Site, and Flux
Surface Finish
HASL


Nickel/Gold


Nickcl/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 PERFORMANCE DEMONSTRATION RESULTS
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 ฐ C and 85% relative humidity.
•      200 cycles of thermal shock with the PWAs rotated between chambers at -50 ฐ C and
       125 ฐC 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
Thermal Shock
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 hi 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

-------
                                           4.1 PERFORMANCE DEMONSTRATION RESULTS
       An ANOVA performed on the 164 pre-test measurements for HCLV PTH produced the
following:
   Source
   Site/Flux
   Error
   Total
DF      Sum of Squares  Mean Square
22           0.2908         0.0132
141          2.6796         0.0190
163          2.9704
F-Statistic
   0.70
P-Value
 0.838
       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 -I) = 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

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
       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(F22441>Fsite/Flux) =  Prob(F22(141 > 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 hot make a significant
contribution to the overall variation hi 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:
N
                                                    J-+J-
                                                    nt  HJ
                                          4-10

-------
                                           4.1 PERFORMANCE DEMONSTRATION RESULTS
where,
a
t

MSB
n and n
level of significance
the o/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
NickeyGold
Nickel/Gold
NickeyGold
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

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
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 (X^) 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 X25 and X25 -
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

 u
Upper
Quartile
                         A Boxplot Used to Display Test Results
                                          4-12

-------
                                           4.1 PERFORMANCE DEMONSTRATION RESULTS
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 i
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 tune 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

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
    Table 4-5. Percentage of Circuits Meeting Acceptance Criteria at Each Test Time2
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%
^•'•^iix^-^^^'/l
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 wke
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
   2 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

-------
                                            4.1 PERFORMANCE DEMONSTRATION RESULTS
have affected the dielectric constant and hence the HF LPF responses. Another possibility is that
the board 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

Circuit [units]
1 HCLV PTH [V]
2 HCLV SMT [V]
3 HVLCPTH^A]
4 HVLCSMTfytA]
5 HSD PTH Propagation Delay [p sec]
6 HSD SMT Propagation Delay [fi sec]
7 HF PTH 50MHz [dB]

8 HFPTHf(-3dB)[MHZ]
9 HF PTH f (-40dB) [MHZ]
10 HF PTH 50MHz [dB]
11 HFSMTf(-3dB)[MHZJ
12 HFSMTf(-40dB)[MHZ]
13 HF TLC 50MHz Forward Response [dB]
14 HF TLC 500MHz Forward Response [dB]
15 HF TLC IGHz Forward Response [dB]
16 HF TLC Reverse Null Frequency [MHZ]
17 HF TLC Reverse Null Response [dB]
18 10-mil Pads [k>gj0 ohms]
19 PGA-A [Iogj0 ohms]
20 PGA-B [Iog10 ohms]
21 Gull Wing [Iog10 ohms]
22 SW 1 [mV]
23 SW2[mV]
CCAMTF Pre-Test
Min
6.60
6.96
5.00
4.92
12.66
4.28
-0.320

239.4
425.3
-0.296
275.0
642.6
-49.74
-21.47
-16.91
624.2
-74,53
10.01
8.94
8.72
9.71
5
19
Max
7.20
7.44
5.25
4.97
13.50
DfE Pre-Test
Min
6.80
7.00
5.00
4.81

5-45
0.094

262.6
454.9
0.081
283.3
674.0
-36.48
-17.54
-12.08
659.8
-38.22
15.00
15.00
15.00
14.00
19
28


^^^Jfe^^ar"', i^''^

^^sSS^^^*
, Max
7.52
7.44
5.25
5.39









-50.87
-19.91
-15.01

-43.67
10.10
10.38
10.07
9.01
7
19
-42.66
-15.28
-12.89

-32.08
15.00
14.00
13.70
13.70
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

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
       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 HF TLC Reverse Null
Response (in dB).  The HF TLC Reverse Null Frequency ranged from approximately 624MHz to
660MHz in the CCAMTF program, while the HF TLC Reverse Null Response ranged from
approximately -75dB to -38dB. However, the null point of the reverse response function for the
Dffi 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 -32.1dB. The
nearly constant value of HF Reverse Null Frequency relegates any subsequent analysis of the
uncertainly 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/Gol
d
(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

-------
                                           4.1 PERFORMANCE DEMONSTRATION RESULTS
       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
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/Palladiam/
- •< 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 ppst-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

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
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 > 0.01 have been shaded).
                                         4-18

-------
                                            4.1 PERFORMANCE DEMONSTRATION RESULTS
                      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
, t ^_^-Q<838 j f. :.-?r^
* ^_ 0.933,- />-„_ .
 f
& • ~v '6.109 ป">'•""* -
**"i^ * tfbป*' ~-"~ซ
\"\ " 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 ckcuits. 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

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
       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
DeltaS
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
NickeyGold
Nickel/Palladium/Gold
OSP
Immersion Silver
Immersion Tin
Anomalies
1
0
0
3
5
3
Notl^el^sS'
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.
                                        4-20

-------
                                           4.1 PERFORMANCE DEMONSTRATION RESULTS
       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).

                      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
'•>'' ' P O.M0 T* , j^r-vy
'-?1" . ^~"f *$028 -^ xl_^~"i
-~C '-'"/I, ;-;&&ฃ: "*?,_ - ;.
r'- ^"*tft274: . *-rt r
P-Vaiue fcrlHVtC -SMZ^
0.000
0.000
0.000
' ~~r WnAn '—•.- - ป-*
-i - ,0.74,2, " " " -,
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^A and 6/^A.
These boxplots are centered close to 5//A, and the total spread is on the order of 0.02//A for the
PTH circuits and approximately 0.5/^A 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//A, 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^A and 6fj.A 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.
                                         4-21

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
       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).

                       Table 4-13. P-Values for HSD Test Results
Test Time
Pre-test
Post-85/85
Post-TS
Post-MS
P-ValueforHSDPTH
0.442- _,> ^ ,
0,443 ,; x ,
_ 0.491 r. x^>^
0.487 ' -
P-VahieforHSDSMT
,< > ^^ 0.585* ''; ""^^ .;
* , V 0.35^'^, '. ^
, , <*. ' ass*1 , ' v,"*^'
"o:t6o^ ., , ;:;
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
                                          4-22

-------
                                           4.1 PERFORMANCE DEMONSTRATION RESULTS
(171 at post-MS). The principal source of these outliers was open PTHs, is discussed in more
detail under Comparison to Acceptance Criteria.

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
P- Value tor
 HFPTH
  50MHz
P-Vahiefor
 HFPTH
  f(~3
-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
       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
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 tunes 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 ฑ50MHz.  Nickel/gold and nickel/palladium gold are again noteworthy as they have
very few anomalies.
                                         4-24

-------
                                          4.1 PERFORMANCE DEMONSTRATION RESULTS
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 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-Medtianieal 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

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
    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
HF 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
NickeyGold
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
0.031 'j
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

-------
                                           4.1 PERFORMANCE PEMONSTRATICW RESULTS
       Such is not the case for the last four HP 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.

4.1.1J. 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

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
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
L 22:0.344 _
I ,;. .;.:,. 0313 .
P- Value for
HFTLC
500MHz
; 0.070- ,.
AHI '*\
o.56o;cr';
o.m „.,,
P-Valuefor
HFTLC
IGHz
* v% 0,250, ;?*
*"' -429$ , ' ;
~''; ' 0,65Q;^r'^
" ,0.568" ?,":"
P- Value for
HFTLC
RNR
/' 0418 *r'
T^%204 tฃ"'
^$.770^ "
S; >/Q359' ,*''
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 hi 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 hi 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
                                         4-28

-------
                                          4.1 PERFORMANCE DEMONSTRAITOJV RESULTS
only specifies an tipper bound of either 5dBb or lOdB for the increase, depending on the
magnitude of 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 HF TLC RNR measurement had
an increase of 7.93dB at post-TS. All other changes were less than 5dB.  One HF 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 then- 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
                                         4-29

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
measurements for each of the four leakage circuits. The p-values for the respective f-statistics
for 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 , „
0.04$, '
0.125 -
P-Value for
PGA-B
0.000
- ," 0.198/: ' "
_ f >~ S,^ /
s
„ ป . 0,093 .
P-Value for
Gull Wing
0.000
• j, ^*QJ5i f ",~
- "-"0.432 ',
^ , 0.243f- /
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.
                                           4-30

-------
                                           4.1 PERFORMANCE DEMONSTRATION RESULTS
       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). Boxplbt displays of the other leakage
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
                                         4-31

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
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.

       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 ฃ•' T
0.410
".a3ฃr~~.?r '"
0.396' """"" -
P- Value for PGA-A
'*." ~'ikv*0,203 *
";; <- 4407 >-; -_ t
' -' - r /y044oj^-' ^-ป3
: j ^ror^8/ -„ T^*f
Abbreviations and Definitions:
MS - mechanical shock
PGA - pin grid array
TS - thermal shock
                                          4-32

-------
                                           4.1 PERFORMANCE DEMONSTRATION RESULTS
       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
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 85ฐC/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 85ฐC/85% relative humidity; and 2) a control group
of boards that were not subjected to the 85ฐC/85% relative humidity environment. The test
group consisted of 10 boards (identified in Table F-l) that exhibited various anomalies following
85ฐC/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 pinhples and foreign material (e.g., solder balls).  The following photographs show
examples of the more prominent visual defects.
                                         4-33

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
                      Visual Observations of Anomalies on Select Test Boards
       A.
       C.
       E.
       G.
B.
D.
H.
      A.  Burnt etches near U102 on Board #082-2
      B.  Excessive WS paste residues beneath SMT component on Board #030-4
      C.  Pin hole in feed through on Board #056-4
      D.  Solder crack around perimeter of J5 connector feed through on Board #056-4
      E.  Solder ball between pins of U3 on Board #013-1
      F.  Discoloration on trace connected to J6 on Board #015-4
      G.  Solder crack around perimeter of filled via near C16 on Board #102-4
      H.  Solder crack around perimeter of J9 connector feed through Board #086-2
                                                4-34

-------
                                           4.1 PERFORMANCE DEMONSTRATION RESULTS
Table 4-21. Identification of Assemblies Selected for Ion Chromatography Analysis
Finish
Board #
Assembly JProtess
:".'•*:'. ::• "'Sfie:;:i-:---':;
Untested; Board (Control Group)
HASL
HASL
Nickel/Gold
NickeUGold
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
i
2
7
12
15
3
8
9
4
10
Post-85/85 Exposure (Anomaly Group)
HASL
NickeyGold
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 ฐC 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.
                                          4-35

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
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 (//g/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) a
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 Gug/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
cr
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
1 Test results reported as micrograms of the residue species per square inch of extracted surface
Abbreviations and Definitions:
Br" - bromide ion
Cl" - chloride ion
LR - low residue flux
WOA - weak organic acids
WS - water soluble flux
                                           4-36

-------
                                                  4.1 PERFORMANCE DEMONSTRATION RESULTS
Table 4-24. Ion Chromatography Anion ฐ Data (Immersion Silver) a
Sample
Description.
Assembly
Process
Site
f Ion Chromatography Data
cr
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 ฐData (Nickel/Gold) a
Sample ,
Description
Assembly
Process <
Site
i
Ion Chromatography Data
cr
Br- , >'-
*WOA.< !
Untested Boards (Control Group)
Board #017-4
Board #068-4
LR
WS
12
7
1.01
4.57
5.34
1.78
150.81
3.08
Tested Boards (Anomaly Group)
Board #013-1
Board #015-4
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 Cug/in2).
Abbreviations and Definitions:
Br" - bromide ion
Cl"-chloride ion
LR - low residue flux
WOA - weak organic acids
WS - water soluble flux
                                                 4-37

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
                 Table 4-26. Ion Chromatography Anion^Data (OSP)a
Sample
Description
Assembly
Process
Site
ton Chromatography Data
cr
Br
WOA
Untested Boards (Control Group)
Board #061-2
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 G/g/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
Cr | Br*
WOA
Untested Boards (Control Group)
Board #001-4
LR 1 15
0.84 | 5.15
151.18
' Test results reported as micrograms of the residue species per square inch of extracted surface Gug/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 //g/in2 for
finished assemblies processed with water-soluble fluxes, and no more than 2.5 jug/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.
                                          4-38

-------
                                            4.1 PERFORMANCE DEMONSTRATION RESULTS
       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 conies 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
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 jug/in2.  The testing
laboratory, CSL, does not presently consider bromide levels under 12 Aig/in2 to be detrimental on
organic PWBs. However, CSL considers levels between 12 /^g/in2 to 20 /zg/in2to be a borderline
risk for failures if attributable to corrosive flux residues. Furthermore, levels above 20 yug/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.
                                          4-39

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
       Weak Organic Acids. Weak organic acids (WOAs), such as adipic or succinic acid,
serve as activator compounds in many fluxes, especially no-clean fluxes.  WO As are typically
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 WOAs.

       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 ^g/in2
20 - 120 //g/in2
250 - 400 /^g/in2
       When WOA levels are under 400 Aig/in2, the residues are generally not detrimental.
Excessive WOA amounts (appreciably greater than 400 Atg/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 WOAs act as insulators that, even at levels as low as
       10 Aig/in2, can potentially create a high resistance contact-to-contact resistance problem
       on devices such as switches.

       The observed levels of WOAs 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.
                                          4-40

-------
                                           4.1 PERFORMANCE DEMONSTRATION RESULTS
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).

       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 HF 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
Oroiitry
Post-
85/85
Past-
Thermal
shock
Post-
Mechanical
Shock
~ Comments
/
i
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.
                                         4-41

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
Circuitry
Post-
85/85
Post-
Thermal
shock
Post-
Mechanical
Shock
Comments
HFLPF
7
8
9
10
11
12
HFPTHSOMHz
HFPTHf(-3dB)
HFPTHf(-40dB)
HFSMT 50MHz
HFSMTf(-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).
HFTLC
13
14
15
16
17
HFTLC 50MHz
HFTLC 500MHz
HFTLC IGHz
HFTLC RNF
HFTLC RNR
0
0
0

1
0
0
1

2
7
1
1
^
5
Minor anomalies.
Minor anomalies.
Minor anomalies.
>3< * * ^ • < J-^. ***" '^^ ' ' '
Minor anomalies.
Leakage
18
19
20
21
10-mil Pads
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
SW1
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
                                            4-42

-------
                                          4.1 PERFORMANCE DEMONSTRATION RESULTS
       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.
•      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.
                                         4-43

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
 Pre-Test
 HCLV PTH
               HASL
Boxplots of HCLV PTH by SiteFlux
    (means are indicated bysolid circles)
  OSP         ImmSn     ImmAg     Ni/Au.  Ni/Au/Pd
75-
7.4 —
73 —
fc 72-
>
c! 7.1-
^
7.0 —
63-
6.8 —

SiteFlux









V


u
r
L
i r

r


c\


T
J





J P

.
*
i

0 ^

*

1
> Tf If



^


1


> It

•1

I
' R 0
'
f u

i i
> rซ- co <
*

0
1
y
T .


1
33 <

1


•

T

1
D

1




1


T-

1
1

1




cv

I




1


i tr

,
n
u


.

i



i


•
I

I
n









<


i

Ul



1
D









r

JL

I
•

T

|
s.












1
?\

i
T

|
X>

J







a

1
, I
r



^

9 ง




J


t

3

i
•

I

1

1
\









c




E



1
j e\i
                ws   ws   ws   ws      ws      ws   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          Imm Sn      Imm Ag     Ni/Au    Ni/Au/Pd
05 —
0.4 —
03 —
I ฐ2~
ฐ- 0.1-
>]
งi ฐ-ฐ~

Q -01 —
•02 —
-03 —
-0.4 —








I .
i

1


1


\
rh rl
i y
• kvj
^

^

1
1


J











1

1


T












I


T







1


















J




^ i










i
| 1

,
_
P
*


j
I



1


















•
j
U

u
I


L
i
1 \
„ '
\
ifj

f


                                                                     i   i  r
                                                                   CM Si  CM  CM
                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)
                                         4-44

-------
                                        4.1 PERFORMANCE DEMONSTRATION RESULTS
Post Thermal Shock
HCLV PTH-
 I
              HASL
      0.5—1
      0.0-
      -0.5-

  SiteFlux
Boxplots of DTHCLV P by SiteFlux
    (means are indicated by solid circles)
  OSP         ImmSn     ImmAg     Ni/Au   Ni/Au/Pd








1
i
•



1
i
1











<
1



j


— r








,
i
T









I
i
1



1
i I
i
IT

• MI-


1"
1
*



i
i
T




1
.
1





^




i
r

1 T~

1 ,
y1
' I
V


J

1 ill '
r




1




rt
1
T




A




L
ป 1
f!



11
•


              ws   ws   ws   ws
             ws
                  i—T-T

ws    wsws   ws   ws   ws
  Figure 4-3,  Boxplot Displays for HCLV PTH Post TS - Pre-Test Measurements (volts) by Surface
                                       Finish
                             (Acceptance Criterion = A<0.5V)
Post Mechanical Shock  D    ,.  • xr>..L1^iwou  o-* t-i
HPI  V PTH              Boxplots of DMHCLV P by SiteFlux
              HASL
    (means are indicated by solid circles)
  OSP         Imm Sn     Imm Ag
              Ni/Au    Ni/Au/Pd
2 —

Q.
Q
0 —


iteFlux


*
• ill
Y p 1

i- CM CO t


1
1 . ; .
T T
T 1
I I I I t
in CD t-~ co en

*

iS*
'

1
o i- CM  •*
	

•
H


i i i
in 
-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
Pre-Test
HCLV SMT
Boxplots of HCLV SMT by SiteFlux
(means are indicated bysolid circles)
HASL
7.45 —
•7.40 -
7.35-
7.30-
fr_
S 755 —
5j 750-
Q
X 7.15 -

7.10-
7.05-
7.00 —













SiteFlux '



i


i

























•





*
r i
- CJ


















1







CO
OSP












rS



i




T
J






1






T





l|




r
1 1




1

^









,


T






1 1 1 1 1
M- 1O CD t-- CO
'












L


1








-L
1


T






Imm Sn Imm Ag













\
O> O T-



1

i 1





T






1


J,
1

J





JL



•













* T



i











i



*








i


1

•g
:,,







CM co TJ- ir> CD r*ซ
1














a.
^
i








CO
Ni/Au











] fi
i Fj

1 I






Ni/Au/Pd





™
•





*
i i i
O> O •>-
1 '










i

































Si 13
               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
 Post 85/85
 HCLV SMT
               HASL
Boxplots of DPHCLV S by SiteFlux
    (means are indicated bysolid circles)
  OSP         ImmSn     ImmAg     Ni/Au   Ni/Au/Pd
0.3 —
02 —
0.1 —
ป
5 -05 —
-0.3 —
-0.4 —
-0.5 —

SiteFlux
I

T




|
T- O


r
r




i
J CO ^
I
J
1
1
n





t ir
|
i
1





) tฃ








) r-

• J
r
f
i



a








> <
JL
'
T




i
ji

j
i
i




c

rn


X


1
> T-


'




1
M








C
iLi i
J • * • JL, n
3 : : | i -
1 '• "
J ;• JL T 1 -
'f" i;(
T



1 1 1 1 1 1 1


LJ
•
*




i i
CM CM O

;
T




I
a co
J 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)
                                        4-46

-------
                                         4.1 PERFORMANCE DEMONSTRATION RESULTS
Post Thermal Shock
HCLV SMT
              HASL
   Boxplots of DTHCLV S by SiteFlux
       (means are indicated by solid circles)
     OSP          ImmSn      ImmAg     Ni/Au   Ni/Au/Pd
0.5-
0.4 —
0.3 —
0.2 —
5 - 0.1 -
>
j
) 0.0-

i -0.1 —
-0.2 —
-0.3 —
•0.4 —

SiteFlux



.I
PI

LJ





1
T- C\

J

J,
U

T
1
1



1
co *a
1
1


• .JL

ฃ

D

I

1
c m
1
* * rl
-L a
S 1' J_ *
; IJR !) PN IH

• U. T * U ^ 1J ปT
'•' • ' ' D T
y 1 i 1 T . Tl
j ฅ



 O T-
1

1
i _
11 ™
rl *

! T ;
1 Ti
1 1



1 1
CM CO
J CM CM
              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)
              HASL
       3 —
   Q   1 -
       0 —
  SiteFlux
Boxplots of DMHCLV S by SiteFlux
    (means are indicated by solid circles)
  OSP         ImmSn     ImmAg
                                        Ni/Au   Ni/Au/Pd
               i   i
               CM   CO
 I  I   I   I   I
LO  CD   I-  CO  O>
               I   I
              i-  CM
 I
CO
 I
•*
 I
I-.
              WS   WS   WS   WS
                 ws
                                                                  CM  CM
                      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)
                                        4-47

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
 Pre-Test
 HF PTH 50MHz
               HASL
Boxplots of HF PTH50 by SiteFlux
    (means are indicated by solid 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 —
-12 —
UX


.
6 fl ^ 8
i



i- CM TO T


^1 *? T IM W
M f V



i i i i i
in 


S * fi "$ i




O -r- OJ CO -t


a-i
1_J t



1 1 1
LO CD r-


r*

i

1
s
ป


f 9



i i i
) O> O i-



T



1 I
CM CO
C\l CM
               WS   WS   WS   WS
              WS
WS   WSWS    WS   WS   WS
  Figure 4-9. Boxplot Displays for HF PTH 50MHz Measurements at Pre-Test (dB) by Surface Finish
 Post Mechanical Shock
 HF PTH 50MHz
               HASL
Boxplots of DMHF PTH by SiteFlux
    (means are indicated by solid circles)
  OSP         Imm Sn      Imm Ag
              Ni/Au    Ni/Au/Pd

0^^
"™~
-10 —
-20 —
-30 —
-40 —
-50-
-60 —
-70 —
-80-

UX









till
ป- CM CO *T









I I
m to t-


i


1

'

1 1
>• CO O>









1
0 -r
1
•



*



1 1 1 1
- OJ CO ^


•
> •:•
\ i


1

i i i
in 
-------
                                         4.1 PERFORMANCE DEMONSTRATION RESULTS
 Post Mechanical Shock
 HFPTHf(-SdB)
               HASL
Boxplots of DMHF PTH by SiteFlux
    (means are indicated by solid circles)
  OSP         Imm Sn     Imm Ag
                                    Ni/Au   Ni/Au/Pd
200 —

150 —
100 —
50 —
CO
ฃ
!••• \j
a.
ฃ -50-
2
Q -100 —
-150 —

-200 —
-250 —

SiteFlux

	 : 	 1

X
•



•- — -m- —




'
•

•
X
I I 1 I
*- CM CO tf
WS WS
1 	 1
X




ป•"•*•*


X





I I I I I
to co r*ป oo o>
WS WS
1 	














1

w
1
X




*
""









• CM m •*
S WS
i 	






S* i
,
'
.;?


i
X ^f
i 1 i

WSWS















1

5 W








*
.
1
|


T

ป
ซ
	 1







*






| |

CM CM
5 WS
t 	














I t

CM CM
WS
Figure 4-11. Boxplot Displays for HF PTH f(-3dB) Post MS - Pre-Test Measurements (MHz) by Surface
                                        Finish
                         (Acceptance Criterion = ฑ50MHz of Pre-test)
 Post Mechanical Shock
 HFPTHf(-40dB)
               HASL
Boxplots of DMHFPTH- by SiteFlux
    (means are indicated by solid circles)
  OSP         Imm Sn     Imm Ag
                                    Ni/Au   Ni/Au/Pd
200 —
100 —
0 —
-100 —
-200 —
• -300 —
-400 —
SiteFlux



• •• -*• f ป
• * X
"
.
'

till
f CM CO 1C
WS WS


*****

X


1 1 1 1 1
WS WS
1
j
•




o

X

@ ป 5 •




i i i i
WS WS







ir



• r*
1
I
f
T
1
WSWJ


•-; e *

X


1 1 1 1
oo en o i-
3 WS WS
.

* *




I i
CM m
CM CM
WS
 Figure 4-12. Boxplot Displays for HF PTH f(-40dB) Post MS - Pre-Test Measurements (MHz) by Surf.
                                        Finish
                         (Acceptance Criterion = ฑ50MHz of Pre-test)
                                        4-49

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
 Post Mechanical Shock
 HFSMT 50MHz

               HASL
     Boxplots of DMHF SMT by SiteFlux
         (means are indicated bysolid circles)
       OSP         Imm Sn     Imm Ag
                                       Ni/Au   Ni/Au/Pd
0 —
-10 —
-20 —
O -30 —
LO
S -40 —
X -50-
Q -60 —
-70-
-80-
-90 —
~. * n
?



*



i * *

i








•



I












•



1



-




*








•
T













.
•


1


•ป








-
ป



1



— *




*



r
'
::


1



J








L •***••



1




   SiteFlux
             1   I
             T-  OJ  CO
 1
TJ-
         I   I
in  to  t^ co  cn
 I  I
i-  CM
 I  1
in  co
                                                          I  T
               ws   ws   ws   ws
                   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 = ฑ5dB of Pre-test)
Post Mecl
HFSMTf(
600 —
500-
400 —
ep 300 —
^ 200 —
X 100 —
Q o
-100-
-200 —
-300 —
SiteFlux
-Id'ef ' ShฐCk BoxP|ots of DMHF SMT by SiteFlux
' (means are indicated bysolid circles)
HASL OSP ImmSn ImmAg Ni/Au Ni/Au/Pd
1
* *

*
* * f
1
i i i i i i i
v- CM co *t in CD N.
WS WS WS

f •
CO O
ws
1
' n *
1
r v
T
1 1 1 1
O ซ- CM CO
ws


* * rt
i
1

i i i i i i i i
ws wsws ws ws


1 1
SI S3
ws
  Figure 4-14. Boxplot Displays for HF SMT f(-3dB) Post MS - Pre-Test Measurements (MHz) by Surf.
                                        Finish
                          (Acceptance Criterion = ฑ50MHz of Pre-test)
                                        4-50

-------
                                         4.1 PERFORMANCE DEMONSTRATION RESULTS
Post Mechanical Shock
HFSMTf(-40dB)
              HASL
        o-
      -200 —
ffi    -400 H
      -600 —



      -800 —

  SiteFlux
Boxplots of DMHFSMT- by SiteFlux
    (means are indicated by solid circles)
  OSP         ImmSn     ImmAg     Ni/Au   Ni/Au/Pd

;

;





*
1
i










*- cw c
ws
•









T
o










1
>i

i


S


1
I









~~~l 1
TJ- W <
I






#


L
•






i

1
!
V'



1
• r







j T
* '
1 "1 1 1
J





I '


•
'
r ซi
:






i'Sf
T =



*
*. ,
*


r
of*-cooso*-C4e
WS WS WS WS








"f
1

:J
]

I
X
™

* X
9








1
1 . 1 1 1 1 1 1 1 1 1 	 [ 	 '
0^fW t
*


i
i
i?

1





* * * *
*
till
CO O) O T-





. 1

1 1
CM CO
ft! (XI
              ws   ws   ws   ws
              ws
ws    wsws   ws   ws   ws
Figure 4-16. Boxplot Displays for HF TLC 50MHz Post MS - Pre-Test Measurements (dB) by Surface
                                        Finish
                          (Acceptance Criterion = ฑ5dB of Pre-test)
                                       4-51

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
 Post Mechanical Shock
 HF TLC 500MHz
               HASL
Boxplots of DMHF TL5 by SiteFlux
    (means are indicated bysolid circles)  •
  OSP         Imm Sn     Imm Ag
                                   ' Ni/Au   Ni/Au/Pd
4 —
3 —

2 —

1 —
o
ง
I- -1 —
ง -2 —
Q
-3 —
-4 —
-5-
-6 —

SiteFlux
'



I
1 I
1 i * i


*
*


iiii
i- C\J CO ^J-






1
i



I

i
in




^


U
T
1



IIII
CO !•*• CO O)




*
rt 1
fi*ป.
t,
X





1 1 1
O f- CM O



X


1 *
t

X



1
3 •*<•





x D 1
e y .
•:
1

*
I


1 1 1
in 
-------
                                           4.1 PERFORMANCE DEMONSTRATION RESULTS
A n T>ri r> ซi Boxplots of Pads by SiteFlux
10-MilPads
(means are indicated by solid circles)
HASL OSP . ImmSn ImmAg Ni/Au Ni/Au/Pd
15 —
14 —
13 —
a.
12 —
11 —
10 —

V
* rji *
r
i
X

T *
R
B rt I
I



X
'lp H
*
1 1 1
?


*•!
i
,
|





I .:


; 9

f






1 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
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 85/8
10-Mil Pa<
14 —
13 —

T3
ง. 12 —
Q.
Q
11 —
10 —
SiteFlux
5
is
HASL







1
i |
T J




fl
I
4
T




Boxplots of DPPads by SiteFlux
(means are indicated by solid circles)
OSP ImmSn ImmAg Ni/Au Ni/Au/Pd
I
fapj





B 1
1 "


*

-
1 ci m 4 ซ ป i! <
SL
;
5




8 i i
| ?






,

i

•
.
r
ง6
i W
n T
8 T


.


Mil



•


i 9
1 1
T





i i i i i i i i i i i i i i i i
                ws    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)
                                          4-53

-------
4.1 PERFORMANCE DEMONSTRATION RESULTS
 Pre-Test
 PGA-A
                HASL
        14 —
        13-
        12-
        11 —i
        10-
   SiteFlux
                   t
Boxplots of PGA A by SiteFlux
  (means are indicated bysolid circles)
OSP         ImmSn      ImmAg
                                  B
                                        8
                                                 Ni/Au   Ni/Au/Pd
                                                  I
 i   i   i   i  i   i   r  i   i   i   i   i   i   i   i  i   i   i   i   i   i   r
CMcoi-in 7.7 Iog10 ohms)
 Pre-Test
 PGA-B
                HASL
        14—)
        13 —
    CO
    <    12-1
    2
         11 -



         10 —

    SiteFlux
              Boxplots of PGA B by SiteFlux
                (means are indicated bysolid circles)
              OSP         Imm Sn      Imm Ag
                                    Ni/Au    Ni/Au/Pd
*





' 0 *
ซ


1
) fl
i
*

a J

i
i
*

.A.

1
T
I




a
0


,


fl ^



1

9 B



y *
f





1






          I   I   I   I   I   I  I   I   I   I   I   I   I   I   I   I   I
                WS   WS   WS    WS
                          ws
                     ws   wsws    ws   ws   ws
  Figure 4-22.  Boxplot Displays for PGA-B Measurements (Iog10 ohms) at Pre-Test by Surface Finish
                        (Acceptance Criterion = Resistance > 7.7 Iog10 ohms)
                                          4-54

-------
                                           4.1 PERFORMANCE DEMONSTRATION RESULTS
Pre-Test
Gull Wing
               HASL
       14 —
       13 —
       12-
       11-
       10-
        9 —
Boxplots of GullWing by SiteFlux
   (means are indicated bysolid circles)
 OSP          ImmSn      ImmAg     Ni/Au    Ni/Au/Pd










X
I,



„
J

	 1 —




JL

1
r
T

1




X

— i —

j
pH
^1



,


i y
JL r*i
IB i'-l

H
T

— i — i — r-1
1 •• 	 : 	 1
'
ii 0
y -
1 *
n J.
^J rl

8 S *
Y ป

— I 	 1 	 1 	 1 	 r—
1 	 i
ij



ix


X
1 	 : 	
1
rh 1
f'l 1

n
"r

•

i
, i


J '
I T


T 	
|
1 U
1*1 1

H
I


  SiteFlux
            i-  CM  CO
         10 ' (O  h-  000>0ป;C;jCT'q-tnOi-
ws   ws    ws    ws
             ws
                                                                            CMeO

                                                   ws    wsws    vvs    ws  w vvs
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)
                                         4-55

-------
4.2 COST ANALYSIS
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
Noti^
•
•
•
•

•
BCซwveypr|z^'/
•


•
•
•
       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.
                                          4-56

-------
                                                                     4.2 COST ANALYSIS
       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
                     Traditional Costs
                       Components
                           T-
Activity-Based Cost
   Components
                                         Cost
                                       Analysis
                     Figure 4-24. Hybrid Cost Analysis Framework
                                          4-57

-------
4.2 COST ANALYSIS
       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 ARENAฎ 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.
                                          4-58

-------
                                                        4.2 COST ANALYSIS
   0
   ซ*Mf
   '"Sft
   a
•c
SD
3
w
i
a
o
a
S
o
U
p
u
e
zes
la<
er
ti
tes
lace
fro
                            2 ?„'
d
C
an
5.1,
ed
of
th
S
si
Daily electricity
Impacts; days to
electricity based
Energy Agency.
5
ul
B
ti
to
m
w
li
on
tme
fi
                                                                   2 ฃ"

  Q
   o
  ra
  I
  "i
  a
eci
nitial bath setup, bath
replacement of spent
in
-o
es
cals
ditic
ze
of chem
ce add
ths.
ba
o
•S ฃ
tf
If
i t3
a S
a s
O c3
^3 C
M 2
CQ K
^ -O
CO S
1J
O co
'fง  •
i ||
Eo CU J2
G ^ w
  SID ..
  a, S
VH CO d<
-2J 2^ ซ
III
al
equipment
                               -1!
                               tl
                               *O "S
                               C 155
                               O „. '
ge to
iter prior to d
               u O o
               III
               •g g -ฐ
               Bill
                    CJ
                            I
              I
              s
                                                  11
                       i
s .^
cj O
CJ CJ
                      i
                    S 0
         s
      D 0
                         ฐ
                         0
                                  4-59

-------
4.2 COST ANALYSIS


Sources of Cost Data



1
a
o
a.
E
o
Description of Cost C

•*-*
g
8
e
a
o


o
1

O
g
O


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.
12 S3
W CX
8 g g -5
j> 1^1
ii>41 & ง<2

C ง S J3 co"
•3 ^ w ^ S
5 ป U Zw H •
o *O e . R T3
Labor costs for line operator, ex
for maintenance activities (inclu
maintenance costs). Assumes 01
day per conveyorized process, 1
per day per non-conveyorized pi
the greater level of labor require





o
,0
.3


Cost of transporting materials from a bill of activity
(BOA); number of bath replacements required from
simulation.


rrt
1 ซ
o i
ซ ^
•O en
S2 S
•t? ?5
Cost to transport chemicals requ
replacement from storage to pro<

o
o
•S
"C w
p<-c
c ฃ
Q cd
ฃ 5

S3
o
•a
o
3
•3 *-•
1 o
Pi o


Cost to clean up tank from BOA; number of bath
cleanups (replacements) required from simulation.
0 -g
" e
to S
8 8
/ — . ~
42 13,
T3 u
o ^
'ง ^S
o ^
f-1 ปQ
Labor and material (excluding c
clean up a chemical tank during


g,

O

PQ


[Cost to set up bath from BOA; number of bath setups
1 required from simulation.

"cd
'H
1
o
cS

3
_4_j
Labor and equipment costs to se
tank after bath replacement.



1

*rl
Is
W
MH
O
*c3
23
Disposal cost to recycle or dispc
wastewater treatment.

1

3

I

H ซ *ซ- &
O 'o F5 B ^ •ง
Not quantified; filter disposal costs are not expected t
differ significantly among the alternatives, but insuffi
data on the type and size of waste filters made it diffi<
to reliably estimate these costs. Factors affecting filte
disposal cost include the waste classification of the fil
the size (weight and volume) of the filter, and the nun
of waste filters generated.

1
53
^
r\
f, ,
O


Disposal cost to recycle or dispc


"c3
o
CO*
Q
1
s.
"S
1
Q

Is
ฃ o
                                      4-60

-------
                                                                     4.2 COST ANALYSIS
       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

-------
43 COST ANALYSIS
to treat 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 hi 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.
                                           4-62

-------
                                                                      4.2 COST ANALYSIS
         Table 4-32. Number of Filter Replacements by Surface Finishing Process
Surf ace 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 ssfb
55
28
90
162
19
9
4
40
57
 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 ARENAฎ 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
faculties, 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.
                                           4-63

-------
4.2 COST ANALYSIS
•      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;
                                          4-64

-------
                                                                     4.2 COST ANALYSIS
•      the entire surface finishing process line is shut down whenever a bath requires replacing,
       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 Surf ace 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);
                                          4-65

-------
4.2 COST ANALYSIS
•      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 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 conveyorized processes,
respectively.

         Table 4-34. Time-Related Input Values for Non-Conyeyorized 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
' 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.
                                           4-66

-------
                                                                       4.2 COST ANALYSIS
           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
- Time"
(minutes)

4.86
5.22
11.2
12.3
Conveyor
Speed6
(ft/inin)
'
8.50
10.4
3.04
1.63
 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.
       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.
                                           4-67

-------
4.2  COST ANALYSIS
             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
; '. ; -*v.%$a^*;^> ;-"^-
750
570
830
1500
130
890
              * 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.

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.
                                           4-68

-------
                                                                   4.2 COST ANALYSIS
    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. lime 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.
         Table 4-38.  Production Tune and Down Time for the Surface Finishing
                        Processes to Produce 260,000 ssf of PWB
Surface Finishing Process
f -- -T •
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 •
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 Time8
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
a To convert from minutes to days, divide by 6.8 hr/day (408 minutes).
                                        4-69

-------
4.2 COST ANALYSIS
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.

       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.
                                          4-70

-------
                                                                      4.2 COST ANALYSIS
       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.
                                           4-71

-------
4.2 COST ANALYSIS
  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 cpntainer(s)
5. Place line containers) on forklift
E. Transport Chemicals to Line
1. Move forklift to line
2. Unload line containers) at line
3. Move forklift to parking area
F. Transport Chemicals from Line to Bath
1. Move line container^) to bath
2. Clean line containers)
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
•':-r ;.'.*;-:< -^-" Resources ,; "'•' i1;"^;-.-''
Labor8

$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

$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^
• /• • ," -•- --.:•. '-;'

$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
$0.00
Total Cost per Transport
•r.^COSlK;
($/transport)
,: -. •-. -;•' ,;:,-_ ,f
~~ ' -' - ''-•

$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
" Labor rate = $10.24 per hour.
b Materials do not include chemicals or tools.
c Forklift operating cost = $0.06 per minute.
                                            4-72

-------
                                                                     4.2 COST ANALYSTS
                           Table 4-40. Costs of Critical Tasks
Task
Transportation of Chemicals
Tank Cleaning
Bath Setup
Sampling and Analysis
Filter Replacement
- , Cost -;':„:: ..:/••., v..
$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
                                          4-73

-------
4.2 COST ANALYSIS
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
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
T      _


F
UR    =
annualized capital cost of equipment ($/yr)
annualized capital cost of installation ($/yr), which is sometimes included in the
cost of the equipment
annualized capital cost of facility ($/yr)
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 ($) -f- 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 (fiVstep) x number of steps]
                                         •f- 25 years
                                           4-74

-------
                                                                     4.2 COST ANALYSIS
       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
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 cpnveyorized (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 conveyorized
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 ftVstep.  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] -f 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-f-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
   r\
     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.

                            ...             4^75

-------
4.2 COST ANALYSIS
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
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
where,
n
           Cost per bath replacement  = Li [chemical product I cost/bath ($/gal) x
                                      1=1
                  % chemical product I in bath x total volume of bath (gal)]
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.
                                          4-76

-------
                                                                    4.2 COST ANALYSIS
       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.,
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 = Z^i [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.
                                         4-77

-------
    COST ANALYSIS
         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
Replacement11
$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
  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.
c 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-Coiiveyorized 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*ฎ)
$25.0/gal
$5.66/gal
$9.39/gal
$27.3/kg
$1.20/gal
$127/gal
$54.0/gal
$51.0/gal
$29.1/kg
$24.1/gal
$30.9/gal
$28.4/gal
$21.4/gal
$40.0/g
Percentage of
Chemical Product b
10
3
3
45g/l
8.5
30
20
12
2g/l
6.6
15
6.6
50
3g/l
Multiplying
, Factor* ,
1
1
1
1
1
1
1
1
1
6
6
5
1
3
Chemical Cost/Bath
Replacement d ($)
$128
$266
$2,810
$11.3
$2,390
$70,200
* 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.
e 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.
                                               4-78

-------
                                                                     4.2 COST ANAL YSIS
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
     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.
                                -         ——

-------
4.2 COST ANALYSIS
       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
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:
where,
W
E
G
                                    U = W + E + G
cost of water consumed ($/ssf) to produce 260,000 ssf
cost of electricity consumed ($/ssf) to produce 260,000 ssf
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
                                          4-80

-------
                                                                    4.2 COST ANALYSIS
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.

      Table 4-43.  Tiered Cost Scale for Monthly Wastewater Discharges to a POTW
Wastewater Discharge
Quantity •"-
(ccf/month)
0-2
3-10
11-100
101 - 400
401 - 5,000
City Discharge
, .Cost
; ($/c
$7.40
$3.21
$2.85
$2.44
$2.05
AveragelMscharge
r - 'Cost
($/ccffmonth)
$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 $3.06, and so on.
The production time required to produce 260,000 ssf of PWB comes from the simulation models.
Thus, wastewater costs (WW) were calculated as follows:
        WW = 2^ [quantity of discharge in tier k (ccf/mo) x tier cost factor ($/ccf)] x
               k=l
                            required production time (months)
where,
P
ccf    =
number of cost tiers
100 cubic ft
                                         4-81

-------
4.2 COST ANALYSIS
       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 -r 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:
where,
LA    =
TR    =
                                     P = LA + TR
production labor cost ($/ssf) to produce 260,000 ssf
chemical transportation cost ($/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 conveyorized processes.
                                         4-82

-------
                                                                     4.2 COST ANALYSIS
       The labor time required to complete the specified job was calculated assuming an average
 shift time of eight hours per day, arid 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 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 categoiy 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:
 where,
 TC
 BS
 FR
 ST
                               MA = TC + BS + FR + ST
tank cleanup cost ($/ssf) to produce 260,000 ssf
bath setup cost ($/ssf) to produce 260,000 ssf
filter replacement cost ($/ssf) to produce 260,000 ssf
sampling cost ($/ssf) to produce 260,000 ssf
                                          4-83

-------
4.2 COST ANALYSIS
       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
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 BOAs 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 $3.70
                 FR =  annual number of filter replacement xURx$ 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 hi 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 -r 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
                                          4-84

-------
                                                                    4.2 COST ANALYSIS
Therefore, the overall maintenance cost for the process is:

       MA = $4,820+ $1,090+ $3,530+ $1,580 =  $11,000

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)-r 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 *
$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
Totals8
$10,200
$109,000
$3,540
$2,050
$19,800
$11,000
$156,000
 Costs of producing 260,000 ssf of PWB by the process.
                                         4-85

-------
4.2 COST ANALYSIS
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.

       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
                                           4-86

-------
                                                       4.2 COST ANALYSIS
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
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
OSP,
HC -
$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 "'
SHyer/C
$10,540
$937
$52,700
$301
$739
$140
$529
$167
$5,260
$1,210
$272
$937
$80
$73,800
Immersion
Ti^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
                             4-87

-------
4.2 COST ANALYSIS
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 *
\<%)
...
3
-67
-327
69
72
22
50
31
* 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.
                                           4-88

-------
                                                                    4.2 COST ANALYSIS
       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 nickeypalladium/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.
                                          4-89

-------
4.3 REGULATORY ASSESSMENT
43    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.

43.1  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 311(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.
                                         4-90

-------
                                                           4.3 REGULATORY ASSESSMENT
            Table 4-47. CWA Regulations that May Apply to Chemicals in the
                                Surface Finishing Process
Chemical"
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
Obs)
5,000
1,000


5,000
5,000
100
1,000
5,000
5,000
1
1,000
1,000

CWA Priority
- Pollutant


/
/


/



/



CWA307a


/
/


/



/



CWA304{b)


/
/


/



/



 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 307(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.
                                           4,91

-------
4.3 REGULATORY ASSESSMENT
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 hi 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 FCWA 301(b). 304(^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.
                                          4-92

-------
                                                         4.3 REGULATORY ASSESSMENT
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 Limitation s (mg/L)
Pollutant or
Pollutant Property
Cyanide (CN)
Lead (Pb)
Cadmium (Cd)
Max. Value for Any '
1 Bay (ppm)/ ,
5.0
0.6
1.2
Average'Daily Values for 4 Consecutive, Moititoring
, Bays 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 Properly
Copper (Cu)
Nickel (Ni)
Lead(Pb)
Cadmium (Cd)
Silver (Ag)
Total Metals
Cyanide (CN)
PH
Max. Value for Any
/ 1 Pay (ppm)
4.5
4.1
0.6
1.2
1.2
10.5
1.9
7.5 < pH < 10.0
Average Daily Values for 4 Consecutive Monitoring
Days that Shall Not be Exceeded mg/L (ppm)
2.7
2.6
0.4
0.7
0.7
6.8
1.0
7.5 < pH < 10.0
                                          4-93

-------
   REGULATORY ASSESSMENT
      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 REGULATORY ASSESSMENT
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*
Acetic acid
Ethylene glycol
Ethylenediamine
Hydrochloric acid
Malic acid
Nickel sulfate
Propionic acid
Sulfuric acid
CAA 111, ,
/
/
/

^

/
/
CAA112b-- •:,*!!

/

/

/


^:.x----'CAA.M2r ;ฃ • •


S
s




" 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.
                                           4-95

-------
4.3 REGULATORY ASSESSMENT
Minimum Standards for State Operating Permit Programs

       The CAA and its implementing regulations (at 40 CFR Part 70) define the minimum
standards and procedures requked 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 requked 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 ak pollutants; or
•      100 TPY of any ak 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 ak 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 requked 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 requkements 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
requkements in the permit.  Five years is the maximum permit term.
                                         4-96

-------
                                                         4.3 REGULATORY ASSESSMENT
       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 hi 40 CFR Part 261, or because they exhibit certain characteristics; namely toxicity,
cprrosivity, 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."
                                          4-97

-------
43 REGULATORY ASSESSMENT
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 pf 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.
                                          4-98

-------
                                                        4.3 REGULATORY ASSESSMENT
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.
CERCLARQs

       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*
Acetic acid
Ammonium hydroxide
Copper ion
Ethylene glycol
Ethylenediamine
Hydrochloric acid
Nickel sulf ate
CERCI^RQObs)
5,000
1,000
1
5,000
5,000
5,000
100
Chemical'
Phosphoric acid
Propionic acid
Silver nitrate
Sodium hydroxide
Sulfuric acid
Thiourea

cmx^/R^^^i
5,000
5,000
1
1,000
1,000
10

  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                                               -
                                         4-99

-------
4.3 REGULATORY ASSESSMENT
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.
43.5   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 DI 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 lexicological 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.
                                         4-100

-------
                                                          4.3 REGULATORY ASSESSMENT
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 *
Ammonium hydroxide
Copper ion
Copper sulfate pentahydrate
Ethylene glycol
Ethylenediamine
Nickel sulfate
Palladium chloride
Phosphoric acid
Sulfuric acid
SARA 110

S
/


/
/


EPCRA 302a




/



^
EPCRA 313
/
^
/
S

S

S
S
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
                                         4-101

-------
               4.3 REGULATORY ASSESSMENT
               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*
Ethylene glycol
Palladium chloride
TSCASdHSDR >
.

: _ :;^SGAOT^:/-[::>
/

:* ^'TSCASa-PAlRM^

/
               1 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 MIL - 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.
_
                                                         4-102

-------
                                                        4.3 REGULATORY ASSESSMENT
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 hi 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.
                                        4-103

-------
4.3 REGULATORY ASSESSMENT

1
"o
ฃ
S
1
2
u
.M
1
0
a>
1

3
1
>a
^^
4O
to
1
f.^
Table 4-57. Summary o







a
j>
es
•2
3
1
S
ซs
S
•ซ
^
Chemicals
w





1
U

1

.05.



^c


^ffi


Process Chemical


P
fc
ซi
i
si
ง
ซ
8,
i-c
u
S
IM
*C sa
to
^
to
I








S

S



S

S
S
Copper sulfate pentahydrate





S



S

S
S




Ethylene glycol





S


S
S
S
S


S


Hydrochloric acid





S


S








Hydrogen peroxide









S




S


1
1
•a
1
PH





-








S


Sodium hydroxide








S



S

S


!
I/I




CO
r— (
CO


CO

-
T-H




CBI chemicals (13)


2
_••< "ฐ ซ
1
*^ ซซSS^O60?^
ง• 5g^ซsuca.sa
•s iซ:fg358.=!ซ-i
ง H^|-ge|#|
e r9ซs2-<;52>5Hhl3
S ^•TH3^te'=oSsli
•-- 8^^11,-S^J^
I giiltil^g •
3 ll^fo^K^S
I 2?225^s"
^ ^^^^^<<<<
| B882g2ง8888
•g we^p^i/iooHHHH
o
ฅ
S -a *S t5
S 1 S <
ซ s -a &
•d o s S
w P o a
S o< OT t2
If ^ 2
3-S ' 8 • S3
w fa -a en
S <2 ง (2
u co tn T* g C"
S 1 s ts .-^ a S
Cu .3 S "S ซ S S a
ซ S iz*-1 2 งSง
8J feS 1 !•! 5
12 ซs-|s^ 1^7
^ ง1 1 1 • S ' 5 3-1
P S3 ง 61 f ซ S "ง 1 •ง
|i a- -8 111 ill •sI'B
|3 .2 ฃ-3-^1 ll-ง iS'i
11
1!
>-e -a ซ '2^ii-^S.5'sWo-<
ฃ| Iฐ3|3SVSงS2<55'
si |<^(ฃซ^^^<
-------
                 4.3 REGULATORY ASSESSMENT





01
o
I
3
"o
r T
^*^
1
d
•!•*
"S
5
en
Chemical
3
^*>
"7*^
<
*S
ซ
1
1
S
CM
O
&
n
s
g
GC
00
V)
•4
>2ซ




*







-
-
4
(9!
i
"3
a
A
S
s>
is
1
8-
'

-



f


&
jsj
^
eง
U




o
p


,
i
,3?
"S-i
m

i
" \


1

,\

X ป
1
w.

0-


'
p


A*
-

ปi

,
I .

-tfc
Sgซ
^ :,
K^'

O
'A
f-
J*4 > "


S -,
•t-i

Sf
|i *
ฃ'
<
1*5
ซs
1
ja
2
w ?
ป
^
5 / \















S











S







| Ammonium hydroxide















S

S






S




s

s

3
s
| Copper sulfate pentahyc










S



S
s


N


S





S







| Hydrochloric acid










S



S




















1
S
C?
i























s











3















S

S



s


s


s

s

s



1
I







s









s

















(Palladium chloride















S











S







'i
•s
o
I



























S







| Sodium hydroxide














S








S



s







3
CO







M


CM


-1

^

-



-

Os
^


O

^

•^f



1 CBI Chemicals (19)
o
8
3
CO
U
ti

U
1
ฃ
3
I
3
C

•S
•I
1
OH
S
2
o
c

ฅ
ta
!•*
l|
ll
i ซง
2 'a
a
c "S
Jj "s
0 U
1 %
-ฃ2 C
Note: For technologies wit
regulation reflects the total
4-105

-------
43 REGULATORY ASSESSMENT


^
bt
o

y
cu
H
S
o
0
•VM
iNickel/PaHad
.a
T3

A
.S
la.
2
ts
'T>
S
^
"s
-a
U








-ft

ฃ
ซซjj
n
s





s
B

i
5


01
ซi


•^E
•^E
.u






.
i





P


o^

A>
Cff
eง ง
ft*

1
P3
sl
1
i"


s
1H

j.
S

1-(

1-t
-
5?
gฃ



y :
03
_ฃ>
g
CO


Process Chemical














s
















s






[Ammonium hydroxide














S


S









S




s

s



Copper sulfate
pentahydrate






































| Ethylenediamine











S

S
S





s
s









s






[Hydrochloric acid











S

S
























[Hydrogen peroxide














S


S



S





s



s
s

s



i
eg
CO
I








s








s




















[Palladium chloride














S
















S






[Phosphoric acid






































| Propionic acid






































[ Sodium chloride































S






| Sodium hydroxide













S










S






s






•o
!
"3
CO








CO


cs
r— <

V)


CO



CO


•-ง i
ซ Ilill-a
| fli ||1
i ils^il


^ S-^^SS1?
^ c2 K ^ ^ ^ ^
•^ D^ fc U U U ^
.ง ftwxxw
e
•2
C3
5 <4-<
5 t3 ฐ
ซง 1 8
.ง &, M
•ง5 a
ซo •" o
1 ซs • i
i.3 83
g 6 ^ -a
Jo ซS -a S
"E 1 .s i S>
11 illS
^ ง M ซง 3^ s
Note: For technologies wit
regulation reflects the total
Abbreviations and definitio
CAA - Clean Air Act
CAA 1 1 1 - Standards of Pe
Air Pollutants-Equipment I
CAA 112b - Hazardous Air
CAA 1 12r - Risk Managem

                                                                        o
                                                                        i
                                   4-106

-------
               4.3 REGULATORY ASSESSMENT





12
*3
s
4
H
nr
0
.ฃ
TJ
,w
H^
c^
1
1

u
0

&
a
"^3

ซ
ง

a
.2

1
^•t
'S
n
|
st
V)
•
4

.0
CS
H








.
-,



.,x.


-?
0
fซ

ซ
s
• ••
%
a
4jl

?
S*
I
"si
1
1


' *•
}


-
. ,
'

"
IF
•S
ซ•*•
•^
u


^
&>
^1



^
^
At
^^
*
.

i
^ซ5
*/•


V-
•*5
r8*
>
>> /•
'
;
,'

S
•"""*


'
to
^/

!
--
^ *<^
ftซ
•ซ ,
ง "
A
'a*
^ S ^
.*? !
'ear* ^ ,
ซ
^ r ^

^n -
fV

a -*
^ ซ
^tซ '•*
$2
^Mj ^

j^ t
-ฃ<'*. ~
1*4
-3 • '*
ฃ>• •- "
''C s
g^ftป -.

3^-:
•68
i: -
&
wป
V
"
s ^ *"
*M ^ v
Js 1
0





























s




s








1
1


















s

s










s




s
s





(Copper ion


















S

S










s




s
s





Copper sulfate
pentahydrate








S









S








S

s













Ethylene glycol














S






















•





[Hydrogen peroxide








S





S



s




s



s






s








(Hydrochloric acid


































S








(Phosphoric acid


































S








(Sodium hydroxide





























S




S








•t-H
CO





,_<





T— 1






1— 1
























CBI Chemicals (9)
tt
O
3
CO
1
o
(/}
tg
&
s
s
•S

•o
o
a
o
1
ฃ
o
13
"3
-S?
M
•3



e
o

f
<ง ™
1 2
J-s
5a ^
X g
R S
a 03
&.a
11
CB ซ-i
•s ฐ

-------
4.3 REGULATORY ASSESSMENT






ง2
O
1
in Immersion Silver T<
13
o>
C/3
&
1
U
o
•*•ซ
>J
'"35
ง
<
>,
i
*•ป
03
ปO
53
1
S
3
DJC
$
CM
O
&
S
cc
1H
VO
•4
O
1









c
tt
Pi
JJ
1
g
51
<
5
•s
<5
in
>o
s
o
*3
1





<ป
1
^
"*ฃ
ฃjt
o
ซ


g
IS
1
w

.s


5
o







^
s


P



fw


ซ S
CO •ซ
sw
1
-a
"g
1

ง

T-l

h
A
*"<
^

^
^
*2 ?*•<
S 'o
lft<;

TS
53
v ป
1
I


Process Chemical









S

S






















Hydrogen peroxide













S














S





[Phosphoric acid








S






S









s


s

s
s


Silver nitrate













S














S





Sodium hydroxide











S









S






S





1
•s
1
00











T— H

i-H




^









T— 4





1 CBI chemicals (5)

2
*J
8
S> - .
=1
03
CO
•a
u
•g .0
S t> 3
ซ 1 - c
1 - J 1 1
fl it !| I 1
1 ii, 11 i I
ง s||*$ซi |
1 ijlll^l^l
S g^l II I'S 3-5 ' •
a -a S ซ S fi S ซ t>o fr
ง g3i-i.il0* -Si
ซ sซ*gป3s^'g-ง
W CO'rtT3S5--O-ilir<^P
22 GoiBS^c^^-S
C n •^ rri 3 ป•? S? fli t 1*^*
1 |aaliJaji
8 111^11^
•a (S^^rS^J^P0-
2 K ~ •— >C/3l-|HT3Sc3
•3 • &< •=> i -< v S ง -<ป
^ ^<^<<ซซ
1 • RR^3^^^^^
•fS UCJU-<
-------
                                                             4.3 REGULATORY ASSESSMENT
                                      o

                                      •w
                                      o
      5
          flu
               S
            w
               S
               S
         s
S
                                   co
0)
   4>

            s
                                      u
                             
-------
REFERENCES	

                                   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. Pacific Northwest Pollution Prevention Research Center
       Publication, Seattle, WA.

Ferguson, John H. 1996. Mean Square Foot Costs: Means-Southern Construction Information
       Nehvork. R.S. Means, Co., Inc. Construction, Publishers and Consultants, Kingston,
       MA.

Fisher, Helen S.  1999. American Wages and Salary Survey, 3rd Ed. Gale Research Inc.,
       Detroit, MI.

Iman, R.L.  1994. A Data-Based Approach to Statistics. Duxbury Press.

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
       Confbrmal Coaling 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.

The Institute for Interconnecting and Packaging Electronic Circuits.  1995.  "Ionic Analysis of
       Circuit Boards Ion Chromatography Method." IPC-TM-650 Test Methods Manual. IPC,
       Lincolnwood, EL.

Vishanoff, Richard.  1995. Marshall Valuation Service: Marshall and Swift the Building Cost
       People. Marshall and Swift Publications, Los Angeles, CA.

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, Office of Pollution
       Prevention and Toxics, Washington, DC.

U.S. Environmental Protection Agency (EPA).  1998. Pollution Prevention and Control Survey.
                                         4-110

-------
                                    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 (CTS A) 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 hi 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 mat 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

-------
5.1 RESOURCE CONSERVATION
      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

-------
                                                          5.1 RESOURCE CONSERVATION
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

-------
5.1 RESOURCE CONSERVATION
Table 5-2. Normalized Water Flow Rates of Various Water Rinse Tvnes
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
0 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:
where,
NRSj
NWCRj
                                            [NRSjXNWCRJ
total water consumption rate (gal/ssf)
number of rinse water stages of type I
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 effect 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 conveyorized 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

-------
                                                                 5.1 RESOURCE CONSERVATION
                Table 5-3. Rinse Water Consumption Rates and Total Water
- Surface Finishing Technology
-.s "
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
Rateb
(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 x 105
2.58 x 1.0s
5.37 x 10s
9.39 x 10s
2.01 x 10s
1.37 x 105
1.37 x 105
4.69 x 105
2.29 x 10s
  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.
" 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 total consumption rate listed in this column for each technology.
                Immersion Silver (c)

                         OSP(c)

                        OSP(nc)

                  Immersion Tin (c)

                        HASL (c)

                       HASL (nc)

                 Immersion Tin (nc)

                   Nickel/Gold (nc)

           Nickel/PalladiunVGold (nc)
                               c: Conveyorized
                               nc: non-conveyorized

          Figure 5-1. Water Consumption Rates of Surface Finishing Technologies
                                                5-5

-------
5.1 RESOURCE CONSERVATION
       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 rninimize 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

-------
                                                          5.1 RESOURCE CONSERVATION
             Table 5-4. Metal Deposition Rates and Total Metal Consumed by
Process
HASL
Nickel/Gold,
Nickel/Palladium/Gold
Immersion Silver
Immersion Tin
Metal
Tin
Lead
Nickel
Palladium
Gold
Silver
Tin
Density"
(Ib/ft3)
462
712
506
749
1200
655
462
Thickness b
(pin)
126 d
74 d
200
6
7
6
25
Metal Blated c
(oz.perssf> \
0.0194
0.0175
0.0337
0.0015
0.0028 '
0.0013
0.0038
Total Metal
Consumed

-------
5.1 RESOURCE CONSERVATION
       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 die 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 CTS A.

       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,
even with processes within a similar technology category). Legal constraints prevent the
                                           5-8

-------
                                                          5.1 RESOURCE CONSERVATION
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
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
                                          5-9

-------
5.1 RESOURCE CONSERVATION
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 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.

       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
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.
                                          5-10

-------
                                                          5.1 RESOURCE CONSERVATION
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
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 PWBs, 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 title 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
pressures. Maintaining these process stages within the desired parameters often requires
                                          5-12

-------
                                                                  5.2 ENERGY IMPACTS
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
l^pe^^nijpinent
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 ฐF.
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
                                          5-13

-------
 5.2 ENERGY IMPACTS
non-conveyorized 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
FuiactionofE^
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
Air Knife/
Sparging*
2
2
1
1
2
2
0
0
0
Fluid
Circulation
3
4
3
3
3
3
4
4
3
Panel
Drytag
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.
e 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:
                                            5-14

-------
                                                                   5.2 ENERGY IMPACTS
where,
EC    =
NPR   =
OL    =

AD    =
                        EC = NPR x OL x AD x (lkW/0.746 HP)
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
Electricity3
(kW) '
14.1
3.1
4.1
0.9
3.8
- -
20
Natural Gas b
W/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:
                                           5-15

-------
  5.2 ENERGY IMPACTS
 where,
 ECRtotal
 NPSj
 ECRj
                                      -.z
                              x ECRj]
total electricity consumption rate (kW)
number of process stages requiring equipment i
energy consumption rate for equipment i (kW)
       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.
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
(ftVhr)
90
90
-
-
90
90
90
90
90
Hourly
Consumption
Rate * 
-------
                                                                   5.2 ENERGY IMPACTS
       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.

             Table 5-9.  Energy Consumption Rate per ssf of PWB Produced
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 *
(hours)
258
133
1,310
1,710
197
93
414
480
710
Total Energy
''Consumed ,-,
(Btu/260,000 ssf)
5.67 x 107
3.46 x 107
1.16xl08
2.00 xlO8
3.26 x 107
1.89 x 107
7.46 x 107
7.52 x 107
1.36 x 108
Energy
Consumption Rate
/ - (Btu/ssf)
218
133
447
. 768
125
73
287
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.
       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.
                                          5-17

-------
 5.2 ENERGY IMPACTS
       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 PWBs.

       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
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 conveyorizeH 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*
(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 miriimize 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.
                                          5-18

-------
                                                                   5.2 ENERGY IMPACTS
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 paniculate 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 hi energy consumption among surface finishing technologies were evaluated using a
computer program developed by the EPA National Risk Management Research Laboratory,
P2P- 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

-------
 S3. ENERGY IMPACTS
 "5
 a

 |


 "c
 c
.8

|


i
a


I

ฃ
 B
ง

O
ot
PL,
tn
•8
            8
            CO
        .-o Jg
        S

        ti
        ซ5
        llฃ
          1

            *
        ง
        CJ Q>

        o tป
                 cs
                 0\
                 
-------
                                                                    5.2 ENERGY IMPACTS
            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 ฐ
Land disposal capacity
Toxic inorganic,3 acid rain, corrosive
Corrosive, dissolved solids b
 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).
c 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 ENERGY IMPACTS
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 hi 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
                                  REFERENCES

Chemical Engineers'Handbook. 1994. McGraw-Hill Book Company.

Sharp, John. 2000.  Teradyne, Inc. Personal communication to Jack Geibig, UT Center for
      Clean Products and Clean Technologies.

U.S. Environmental Protection Agency (EPA). 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
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 CTS A, 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.
                                          6-2

-------
                                                          6.1 POLLUTION PREVENTION
       Opportunities for pollution prevention in PWB manufacturing were identified in each of
the following areas:
       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

-------
6.1 POLLUTION PREVENTION
     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.
• -,,.- '••-' ..'-' : • C' ' "'• 'Benefits;, •^•-: •^:-:^"-:^-
-------
                                                            6.1 POLLUTION PREVENTION
and prioritized based on their cost, feasibility of implementation, and their overall effectiveness
of eliminating or 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 pollutibn 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.
                                           6-5

-------
6.1 POLLUTION PREVENTION
By identifying the cost drivers, manufacturers can correctly assess the true cost of waste
generation and the benefits of any pollution prevention efforts.

       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:
http://www.iso.ch/welcome.html.

       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 each of five phases: 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: http://www.epa.gov/opptintr/dfe/tools/ems/ems.html.

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.
•-•-., V.v;'-y:-^'y •- 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

-------
                                                            6.1 POLLUTION PREVENTION
       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/fkst-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.,  450ฐF
for the oil and solder mixture) and both the oil and flux having vapor pressures that allow a
portion to evaporate  and condense as fine  droplets at elevated temperatures (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 POLLUTION PREVENTION
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 hi 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

-------
                                                            6.1 POLLUTION PREVENTION
         Table 6-3. Pollution Prevention Practices to Reduce Bath Contaminants
'•'::,-^;''-''-"-^:'- - - i::->'''^:Vr-:':v'.
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

-------
6.1 POLLUTION PREVENTION
       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,
minmiizing 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

-------
                                                             6.1 POLLUTION PREVENTION
       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 statiq 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

-------
6.1 POLLUTION PREVENTION
       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 Improvement Methods To Extend Bath Life
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.
•' •':"• •'••--'• :•'•"•••- v -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

-------
                                                            6.1 POLLUTION PREVENTION
       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

-------
6.1 POLLUTION PREVENTION
       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

-------
                                                            6.1 POLLUTION PREVENTION
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
                                          6-15

-------
6.1 POLLUTION PREVENTION
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.

       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 hi 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

-------
                                                           6.1 POLLUTION PREVENTION
       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 conveyorized
system.

        A  conveyorized 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

-------
6.1 POLLUTION PREVENTION
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

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 aind 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

-------
6.2 RECYCLE, RECOVERY, AND CONTROL TECHNOLOGIES ASSESSMENT	

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 ฐF), 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

-------
	6.2 RECYCLE, RECOVERY, AND CONTROL TECHNOLOGIES ASSESSMENT

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
                           Contaminated
                               Solder
HASL

>
t
f
\.

Solder
Reclaim
  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

-------
6.2 RECYCLE, RECOVERY, AND CONTROL TECHNOLOGIES ASSESSMENT	______

       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 copper salt (e.g., copper sulfate) solutions, using ion exchange methods, before
undergoing electrowinning to recover the copper content (Coombs, 1993).
                                          6-22

-------
	6.2 RECYCLE, RECOVERY, AND CONTROL TECHNOLOGIES ASSESSMENT

       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

-------
63 RECYCLE, RECOVERY, AND CONTROL TECHNOLOGIES ASSESSMENT	

       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
                                                            claimed
                                                            Metal
                                                   Spent
                                                 Regenerant
                                                           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

-------
	6.2 RECYCLE, RECOVERY, AND CONTROL TECHNOLOGIES ASSESSMENT

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
>•
ke-
nns
nditioning
Tank



>.
*
RO
Unit


up "^


>.
•
Storage
Tank

w
                                                            Reusable
                                                             Rinse
                                                             Water
                                                       Waste
                                                      Treatment
                    Figure 6-3. Reverse Osmosis Water Reuse System
                                          6-25

-------
6.2 RECYCLE, RECOVERY, AND CONTROL TECHNOLOGIES ASSESSMENT
       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
Price8*
$283/oz
$636/oz
$4.98/oz
$0.80/lb
$1.60/lb
• •••'..•• .. .;;-•- V---Reซove%^tK^
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.
e 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

-------
	6.2 RECYCLE, RECOVERY, AND CONTROL TECHNOLOGIES ASSESSMENT

       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

-------
6.2 RECYCLE, RECOVERY, AND CONTROL TECHNOLOGIES ASSESSMENT
       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






•


Ion
Exchange
•
•
•
•

•

•

Electrolyti
c Recovery
•
•
•
•

•

•

Reverse ;
Osmosis








•
Off-Site
Refining
•
•


•
•



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
facility and the effluent  permit limits.  Together these processes form a complete treatment
                                          6-28

-------
	•    	6.2 RECYCLE, RECOVERY, AND CONTROL TECHNOLOGIES ASSESSMENT

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
dependant on both the treatment chemistry and the metals being removed from the wastewater.
                                          6-29

-------
6.2 RECYCLE, RECOVERY, AND CONTROL TECHNOLOGIES ASSESSMENT	

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

-------
                    6.2 RECYCLE, RECOVERY, AND CONTROL TECHNOLOGIES ASSESSMENT
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

-------
6.2 RECYCLE, RECOVERY, AND CONTROL TECHNOLOGIES ASSESSMENT
                                                                   I
                                                                   &
                                                                   I
                                                                   1
                                                                   I
                                  6-32

-------
   .         	6.2 RECYCLE, RECOVERY, AND CONTROL TECHNOLOGIES ASSESSMENT

       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

-------
6.2 RECYCLE, RECOVERY, AND CONTROL TECHNOLOGIES ASSESSMENT
    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
CIS A 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
                                         6-34

-------
	6.2 RECYCLE, RECOVERY, AND CONTROL TECHNOLOGIES ASSESSMENT

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.

       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

-------
63. RECYCLE. RECOVERY, AND CONTROL TECHNOLOGIES ASSESSMENT
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 /PaUadium/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*' "
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.
* 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 ah" control process is the effective containment of
fugitive air emissions at their source of release. This is accomplished using fume hoods over the
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.
                                           6-36

-------
	        6.2 RECYCLE, RECOVERY, AND CONTROL TECHNOLOGIES ASSESSMENT

       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
                                   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, JJL.

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,JackD. 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, CA.
      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.
                                        6-38

-------
Lee, Matthew A. 1999. "Controlling Emissions Stemming From The Hot Air Solder Leveling
       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. Environmental Protection Agency  (EPA).  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. Environmental Protection Agency  (EPA).  1995a. "Printed Wiring Board Case Study 1:
       Pollution Prevention Work Practices."  Pollution Prevention Information Clearinghouse
       (PPIC). Washington, B.C.  EPA 744-F-95-004.  July.

U.S. Environmental Protection Agency  (EPA).  1995b.  "Printed Wiring Board Case Study 2:
       On-Site Etchant Regeneration."  Pollution Prevention Information Clearinghouse (PPIC)
       Washington, D.C. EPA 744-F-95-005. July.

U.S. Environmental Protection Agency  (EPA).  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. Environmental Protection Agency  (EPA).  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. Environmental Protection Agency  (EPA).  1996c. "Printed Wiring Board Project: A
       Continuous-How System for Reusing Microetchant."  Pollution Prevention Information
       Clearinghouse (PPIC). Washington, D.C. EPA 744-F-96-024.  December.

U.S. Environmental Protection Agency  (EPA).  1997a. "Pollution Prevention beyond Regulated
       Materials." Pollution Prevention Information Clearinghouse (PPIC).  Washington, D.C.
       EPA744-F-97-006. May.

'U.S. Environmental Protection Agency  (EPA).  1997b. "Identifying Objectives for Your
       Environmental Management System." Pollution Prevention Information Clearinghouse
       (PPIC). Washington, D.C.  EPA744-F-97-009. December.
                                        6-39

-------
REFERENCES	   •	_____	

U.S. Environmental Protection Agency (EPA).  1997c. "Building an Environmental
       Management System - HR Industry Experience." Pollution Prevention Information
       Clearinghouse (PPIC). Washington, D.C. EPA744-F-97-010. December.

U.S. Environmental Protection Agency (EPA).  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. Environmental Protection Agency (EPA).  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
Nott-Conveyorized
•
•
•
•

•
Conveyorized
•


•
•
•
       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	

7.1    RISK, COMPETITIVENESS, AND CONSERVATION DATA SUMMARY

       Earlier sections of the CTS A 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

-------
	7.1 RISK, COMPETITIVENESS, AND CONSERVATION DATA SUMMARY

       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

-------
7.1 RISK, COMPETITIVENESS, AND CONSERVATION DATA SUMMARY
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*
(Non-Conveyorized, 260,000 ssf)
HASL

•





Nickel/Gold
•

•
•
•
•

Nickel/Palladium/Gold
•

•
•
•
•
•
OSP

•





ฐ Non-conveyorized immersion silver process not evaluated. Occupational exposure and risk from all conveyorized
process configurations are below concern levels.
• 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

-------
                        7.1 RISK, COMPETITIVENESS, AND CONSERVATION DATA SUMMARY

                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 "
HASL
(NO





••



X


HASL
(O





••



X


Nickel/Gold
. (NO

•
•


••

•
••

••

Nickel/,
Palladium/Gold
(NC)
•

•


••

•
••

••

OSP
(NC)



••
••
••






OSP
(O



••
•
••






Immersion
Tin
(NC)






•




•
 No risk results were above concern levels for the conveyorized immersion silver or conveyorized immersion tin
processes.
•  Line operator risk results above concern levels (non-cancer health effects).
•• Line operator and laboratory technician risk results above concern levels (non-cancer health effects).
X: 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.
Abbreviations and Definitions:
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.  Add