EfSl^ilSsSJi
                                                 —^/l^Sfiil; *. i'^-CL|r%?ปi
                                               S^*r:^*:?Q
                                        m^^:?mm

                                                      '&Sff*
                                                          aas
                                              S^ฃSSSTpfS.V,iy
                                              m^m^mftm'K

t A *  *
                       WfSv3^S™*'w.iH

-------

-------
                         PWB Engineering Support Team


        The PWB Engineering Support Team consisted of University of Tennessee faculty and
 graduate students who developed analytical models for the project and/or authored sections of
 this document. Members of the Team and the sections to which they contributed are listed below:

 Exposure Assessment and Risk Characterization

 Dr. Chris D. Cox, Associate Professor of Civil and Environmental Engineering,
 Dr. R. Bruce Robinson, Professor of Civil and Environmental Engineering,
 and graduate research assistants in Civil and Environmental Engineering:
 Aaron Damrill, Jennie Ducker, Purshotam Juriasingani, and Jeng-hon Su.

 Cost Analysis

 Dr. Rupy Sawhney, Assistant Professor of Industrial Engineering and Director,
 Lean Production Laboratory, and graduate research assistants in Industrial Engineering-
 Aamer Ammer.                               .
                                     Disclaimer
       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 informatipn 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 Aye., N.W. (7407)
                                 Washington, DC 20460
                                 Phone: (202) 260-1023
                                   Fax: (202) 260-4659
                                  E-mail: ppic@epa.gov
                     Web site: www.epa.gov/opptintr/library/ppicdist.htm

       To learn more about the University of Tennessee Center for Clean Products and Clean
Technologies, visit the Center's web site at:

                                  eerc.ra.utk.edu/clean/
                                           11

-------
                                 Acknowledgments


        This Cleaner Technologies Substitutes Assessment (CTSA) was prepared by the
 University of Tennessee (UT) Knoxville Center for Clean Products and Clean Technologies and
 the PWB Engineering Support Team, with assistance from numerous UT students and staff. The
 authors would like to acknowledge the outstanding contributions of Lori Kincaid, UT Center for
 Clean Projects and Clean Technologies, who provided guidance throughout the project;
 Catherine Wilt, UT Center for Clean Projects and Clean Technologies, who researched and
 wrote the Regulatory status section of this document; James Dee, UT Center for Clean Projects
 and Clean Technologies, who assisted with the development of the Human Health and Ecological
 Hazards Summary; and Margaret Goergen, who was the document production manager.

        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 Pah/clad 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 Lnan 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 Hilhnan for their help in planning and
 conducting the performance demonstration. Recognition is also given to ADI/IsoIa, who
 supplied the materials to build the performance demonstration test boards, to Network Circuits
for volunteering to build the boards, and to the test facilities that ran the boards through their
 surface finish lines and provided critical data for the study. Performance demonstration
 contractor support was provided by Abt Associates, Inc., of Cambridge, MA, under the direction
of Cheryl Keenan.

-------
                      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
                   FranklynHall
                   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  ,

Electrochemical, 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.                                                         •
ADC 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

                                t

Parlex Corp.

Quality Circuits Inc.

Sanmina Corp. (formerly Altron 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  EPADfEProgram	-	1-2
       1.1.2  DfEPWB Program	 1-2
1.2    Overview of PWB Industry	1-5
       1.2.1  Types of Printed Wiring Boards	1-5
       1.2.2  Industry Profile	-	I'5
       1.2.3  Overview of Rigid Multi-Layer PWB Manufacturing  	1-7
1.3    CTSA Methodology	-..		:	.'	1-3
       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	-	-	•  1-1ฐ
       1.3.4  Primary Data Sources	• •  1-H
       1.3.5  Project Limitations	- -  1-1-3
1.4    Organization of this Report	•	  1-15
References	 - - -	!-l6

Chapter!
Profile of the Surface Finishing Use Cluster	2-1
2.1    Chemistry and Process Description of Surface Finishing Technologies	 2-1
       2.1.1  . Process Sequence of Surface Finishing Technologies  	2-1
       2.1.2  Overview of the Surface Finishing Manufacturing Process . . .	2-3
       2.1.3  Chemistry and Process Descriptions of Surface Finishing Technologies	2-4
       2.1.4   Chemical Characterization of Surface Finishing Technologies . ..	2-17
2.2    Additional Surface Finishing Technologies	2-23
       2.2.1  Immersion Palladium	2-23
References	2-25

Chapters
Risk Screening and Comparison	3-1
3.1    Source Release Assessment	3-1
       3.1.1  Data Sources and Assumptions	 3-2
       3.1.2   Overall Material Balance for Surface Finishing Technologies  '. .	3-3
       3.1.3   Source and Release Information for Specific Surface Finishing
              Technologies	3-17
       3.1.4   Uncertainties hi the Source Release Assessment	3-34
3.2    Exposure Assessment	- - - 3-35
       3.2.1   Exposure  Setting	 3-35
       3.2.2   Selection of Exposure Pathways	.....' 3-40
       3.2.3   Exposure-Point Concentrations	 . 3-43

                                           vi

-------
        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     HumanHealth and Ecological Hazards Summary	         ;   3.30
        3.3.1  Carcinogenicity	                   3_gQ
        3.3.2  Chronic Effects (Otherthan Carcinogenicity)	                  3.53
        3.3.3  Ecological Hazard Summary	        3.95
        3.3.4  Summary	                 3-102
 3.4     Risk Characterization	;                               3-104
        3.4.1  Summary of Exposure Assessment .	;    3-104
        3.4.2  Summary of Human Health Hazards Assessment  . .	              3-109
        3.4.3  Summary of Ecological Hazards Assessment  	 . . 3-109
        3.4.4  Methods Used to Calculate Human Health Risks	'..'.'.'.'. 3-110
        3.4.5  Results of Calculating Human Health Risk Indicators	         '.;.'. 3-113
        3.4.6  Evaluation of Lead Risks from Tin-Lead Solder Used in the HASL
              Process	                  3-125
        3.4.7  Results of Calculating Ecological (Aquatic) Risk Indicators	3-128
        3.4.8  Uncertainties	                     3-130
       3.4.9  Conclusions	    .                  3-132
3.5    Process Safety Assessment	                      3-137
       3.5.1   Chemical Safety Concerns	               3-137
       3.5.2  Hot Air Solder Leveling (HASL) Process Safety Concerns	'.[ 3-146
       3.5.3   Process Safety Concerns	            3-146
References	;	              	" ' .,  ,5,

Chapter 4
Competitiveness	 .                                               4 1
4.1    Performance Demonstration Results	       	4-1
       4.1.1  Background	         	4_j
       4.1.2  Performance Demonstration Methodology	                          4_2
       4.1.3  Test Vehicle Design		'.'.'.'.'.'.'.'.'.'.'.	"	4.4
       4.1.4  Environmental Testing Methodology 	   	4_8
       4.1.5  Analysis ofthe Test Results	'.'.'.'.'.'.'.'.'.'.'.'.'.           4-8
       4.1.6  Overview of Test Results	'.'.'.';'.'.	4-13
       4.1.7  HCLV Circuitry Performance Results	       . ' ^         4.13
       4.1.8  HVLC Circuitry Performance Results	      4-20
       4.1.9  High Speed Digital Circuitry Performance Results  	'.'.'.'.'.'.'.'.'.'" 4-21
       4.1.10 High Frequency Low Pass Filter Circuitry Performance Results  ......]... 4-22
       4.1.11 High Frequency Transmission Line Coupler Circuitry Performance
             Results .	                              v 2j
      4.1.12 Leakage Measurements Performance Results                              4 29
      4.1.13 Stranded Wires	.	'.'.'.'.'.'.'.'.	4.32
      4.1.14 Failure Analysis  	                 	      ^    4 33
      4.1.15 Summary and Conclusions ;	      	    4_40
      4.1.16 Boxplot Displays	       	4_42
                                         vu

-------
4.2    Cost Analysis	• •	4-55
       4.2.1  Overview of the Cost Methodology 	• 4-5,6
       4.2.2  Cost Categories and Discussion of Unquahtifiable Costs	4-57
       4.2.3  Simulation Modeling of Surface Finishing Processes	 4-62
       4.2.4  Activity-Based Costing	._..:	• •  • • 4-68
       4.2.5  Cost Formulation Details and Sample Calculations	4-71
       4.2.6  Results	'i	4-84
       4.2.7  Conclusions	- -  • - 4-86
4.3    Regulatory Assessment	4-88
       4.3.1  Clean Water Act	,	4-88
       4.3.2  Clean Air Act	4-93
       4.3.3  Resource Conservation and Recovery Act	 . . . '.	4-95
       4.3.4  Comprehensive Environmental Response, Compensation and
             Liability Act	4-97
       4.3.5  Superfund Amendments and Reauthorization Act and Emergency
             Planning and Community Right-To-Know Act	 4-98
       4.3.6  Toxic Substances Control Act	 4-100
       4.3.7  Summary of Regulations for Surface Finishing Technologies  	4-101
References	•'	•	4-108

Chapters
Conservation	•  -	5-1
5.1    Resource Conservation	5-1
       5.1.1  Consumption of Natural Resources	5-2
       5.1.2  Consumption of Other Resources	5-8
       5.1.3  Summary and Conclusions	5-10
5.2    Energy Impacts	5-12
       5.2.1  Energy Consumption During Surface Finishing Process Operation	5-12
       5.2.2  Energy Consumption Environmental Impacts	,	5-18
       5.2.3  Energy Consumption in Other Life-Cycle Stages	5-21
       5.2.4  Summary and Conclusions	5-22
References	•  • •	•	5-23

Chapter 6
Additional Environmental Improvements Opportunities	6-1
6.1    Pollution Prevention	:	6-2
       6.1.1  Management and Personnel Practices	6-3
       6.1.2  Materials Management and Inventory Control	6-6
       6.1.3  Material Selection	6-7
       6.1.4  Process Improvements	6-8
6.2    -Recycle, Recovery, and Control Technologies Assessment .	,. . 6-19
       6.2.1  Recycle and Resource Recovery Opportunities	6-19
       6.2.2  Control Technologies	 6-28
References	- - - - 6-38
                                          vui

-------
 7.1
 7.2
7:3
       .7.2.3
       7.2.4
       7.3.1
       7.3.2
       7.3.3
       7.3.4
       7.3.5
       7.3.6
References .
                                                                                   7-1
Chapter 7
Choosing Among Surface Finishing Technologies .;....
      Risk, Competitiveness, and Conservation Data Summary	
      7.1.1  Risk Summary 	         	           72
      7.1.2  Competitiveness Summary		7 8
      7.1.3  Resource Conservation Summary .............                      7-14
      Social Benefits/Costs Assessment	          	          716
      7.2.1  Introduction to Social Benefits/Costs Assessment.	        ..."'"     7-16
      7.2.2  Benefits/Costs Methodology and Data Availability	            7-18
             Private and External Benefits and Costs Associated with Choice of	
             Surface Finishing Alternative  	•..;..                         7_19
             Summary of Benefits and Costs	. "     	7"28
      Technology Summary Profiles	           " [       	7"30
             HASL Technology		7"30
             Nickel/Gold Technology		...........     	      '"' 7"35
             Nickel/Palladium/Gold Technology		" ' 7"39
             OSP Technology	'..'.'.'.'.'.'.'.'.'.'.'.'.	  7^44
             Immersion Silver Technology	              	       7 47
             Immersion Tin Technology	      	-----.  -
                                                                                 7-56
                                        IX

-------
                                   List of Tables
Table 1-1.    Surface Finishing Technologies Submitted by Chemical Suppliers	..  1-11
Table 1-2.    Responses to the PWB Workplace Practices Questionnaire	1-13
Table 2-1.    Use Cluster Chemicals and Associated Surface Finishing Processes	2-18
Table 3-1.    Water Usage of Surface Finishing Technologies From Questionnaire  	3-7
Table 3-2.    Reported Use of Chemical Flushing as a Tank Cleaning Method		3-8
Table 3-3.    Average Bath Dimensions and Temperatures for All Processes	3-10
Table 3-4.    Spent Bath Treatment and Disposal Methods  .....	3-13
Table 3-5.    RCRA Wastes and Container Types for Surface Finishing Technologies	3-16
Table 3-6.    Workplace Activities and Associated Potential Exposure Pathways	3-40
Table 3-7.    Potential Population-Exposure Pathways	3-42
Table 3-8.    Results of Workplace Air Modeling		3-47
Table 3-9.    Results of Ambient Air Modeling	3-51
Table 3-10.   Estimated Releases to Surface Water Following Treatment			3-54
Table 3-11.   Parameter Values for Workplace Inhalation Exposures	3-57
Table 3-12.   General Parameter Values for Workplace Dermal Exposures 	3-59
Table 3-13.   Parameter Values for Workplace Dermal Exposures for Line Operators
             onNon-Conveyorized Lines			•	3-60
Table 3-14.   Parameter Values for Workplace Dermal Exposure for Line Operators
             on Conveyorized Lines	3-61
Table 3-15.   Parameter Values for Workplace Dermal Exposure for a Laboratory
             Technician on Either Conveyorized or Non-Conveyorized Lines . . .	3-62
Table 3-16.   Estimated Average Daily Dose for Workplace Exposure From Inhalation
             and Dermal Contact	3-63
Table 3-17.   Estimated Concentration of Lead in Adult and Fetal Blood from Incidental
             Ingestion of Lead in Tin/Lead Solder	, 3-72
Table 3-18.   Parameter Values for Estimating Nearby Residential Inhalation Exposure .. . 3-74
Table 3-19.   Estimated Average Daily Dose for General Population Inhalation
             Exposure  .	3-74
Table 3-20.   Children's Blood-Lead Results from the IEUBK Model at Various Lead
             Air Concentrations	3-75
Table 3-21.   Available Carcinogenicity Information	3-81
Table 3-22.    Summary of RfC and RfD Information used in Risk Characterization for
             Non-Proprietary Ingredients	 3-84
Table 3-23.   NOAEL/LOAEL Values Used in Risk Characterization for
             Non-Proprietary Ingredients	3-87
Table 3-24.   Developmental Toxicity Values Used in Risk Characterization for
             Non-Proprietary Ingredients		3-88
Table 3-25.    Summary of Health Effects Information	3-89
Table 3-26.    Overview of Available Toxicity Data	 3-93
Table 3-27.   Estimated (Lowest) Aquatic Toxicity Values  and Concern Concentrations
             for PWB Surface Finishing Chemicals, Based on Measured Test Data or
              SAR Analysis	3-96
Table 3-28.   Environmental Hazard Ranking of PWB Finishing Chemicals	3-100

-------
 Table 3-29.   Gastrointestinal (GI) Absorption Factors			  3-112
 Table 3-30.   Summary of Human Health Risks From Occupational Inhalation
              Exposure for Selected Chemicals	  3-116
 Table 3-31.   Summary of Human Health Risks Results From Occupational Dermal
              Exposure for Selected Chemicals	3-118
 Table 3-32.   Summary of Potential Human Health Effects for Chemicals of Concern  ...  3-121
 Table 3-33.   Data Gaps for Chronic Non-Cancer Health Effects for Workers	3-122
 Table 3-34.   Risk Evaluation Summary for Lead	  3-126
 Table 3-35.   Summary of Aquatic Risk Indicators for Non-Metal Chemicals
              of Concern	 :	3-129
 Table 3-36.   Summary of Aquatic Risk Indicators for Metals Assuming No On-Site
              Treatment	               3-130
 Table 3-37.   Overall Comparison of Potential Human Health and Ecological Risks
              for the Non-Conveyorized HASL and Alternative Processes	3-136
 Table 3-38.    Flammable, Combustible, Explosive, and Fire Hazard Possibilities for
              Surface Finishing Processes	                    3-139
 Table 3-39.    Corrosive, Oxidizer, Reactive, Unstable, and Sudden Release of Pressure
              Possibilities for Surface Finishing Processes	3-141
 Table 3-40.    Sensitizer, Acute and Chronic Health Hazards, and Irreversible
              Eye Damage Possibilities for Surface Finishing Processes	3-143
 Table 4-1.     Electrical Responses for the Test PWA and Acceptance Criteria .......	4-6
 Table 4-2.     Distribution  of the Number of LRSTFPWAs by Surface Finish,
              Site, and Flux	              .... 4-7
 Table 4-3.     Listing of 23 Site/Flux Combinations Used in the Multiple Comparisons
              Analyses		                    4_U
 Table 4-4.     Number of Anomalies Observed at Each Test Time  .	    4-13
 Table 4-5.     Percentage of Circuits Meeting Acceptance Criteria at Each Test Time	4-14
 Table 4-6.     Comparison  of CCAMTF Pre-Test Ranges with DfEPre-Test
              Measurements 	                   4.^5
 Table 4-7.     Frequency Distribution of Post 85/85 Anomalies per PWA by
              Surface Finish	            . . .           4.15
 Table 4-8.     Frequency Distribution of Post-Thermal Shock Anomalies per PWA by
              Surface Finish .;	                    4_17
 Table 4-9.     Frequency Distribution of Post-Mechanical Shock Anomalies per PWA by
              Surface Finish	.	..'....                   4-18
 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_20
 Table 4-12.   P-Values forHVLC Test Results	       	    4-21
 Table4-13.   P-Values for HSD TestResults	'.'.I'.'.'.'.'.'.'.'.'.'.'.'.'.''"  4-22
 Table 4-14.   P-Values for HF LPF Test Results  	'.'.'.'.'.'.'.'.'.'.  4-23
 Table 4-15.   Frequency Distribution of HF LPF Anomalies at Post-Mechanical Shock
             per PWA	             4,25
Table 4-16.   Comparison of the Observed and Expected Number of Anomalies for
             the HF LPF PTH 50MHz Circuit by Surface Finish	4-26
                                         XI

-------
Table 4-17.   Comparison of the Observed and Expected Number of Anomalies Under the
             Hypothesis of Independence of Surface Finishes	4-26
Table 4-18.   P-Values for HF TLC Test Results	4-28
Table 4-19.   P-Values for Leakage Test Results	:'	4-30
Table 4-20.   P-Values for Stranded Wire Test Results	4-32
Table 4-21.   Identification of Assemblies Selected for Ion Chromatography Analysis .... 4-35
Table 4-22.   Ion Chromatography AnionwData (HLASL)	'.	4-36
Table 4-23.   Ion Chromatography AnionwData (Immersion Tin)	4-36
Table 4-24.   Ion Chromatography AnionWData (Immersion Silver)	4-37
Table 4-25.   Ion Chromatography AnionwData (Nickel/Gold)	4-37
Table 4-26.   Ion Chromatography Anionฐ Data (OSP)		.. 4-37
Table 4-27.   Ion Chromatography Anion w Data (Nickel/Palladium/Gold)  	4-38
Table 4-28.   Acceptance Levels for Weak Organic Acids	 4-39
Table 4-29.   Frequency of Anomalies by Individual Circuit Over Test Times	4-41
Table 4-30.   Surface Finishing Processes Evaluated in the Cost Analysis	 4-55
Table 4-31.   Cost Component Categories	4-58
Table 4-32.   Number of Filter Replacements by Surface Finishing Process 	4-62
Table 4-33.   Bath Volumes Used for Conveyorized Processes  ....,	4-64
Table 4-34.   Time-Related Input Values for Non-Conveyorized Processes	4-65
Table 4-35.   Time-Related Input Values for Conveyorized Processes	 4-65
Table 4-3 6.   Bath Replacement Criteria for Nickel/Gold Processes  .... 1	4-66
Table 4-37.   Frequency and Duration of Bath Replacements for Non-Conveyorized
             Nickel/Gold Process	 .	4-67
Table 4-3 8.   Production Time and Down Time for the Surface Finishing Processes to
             Produce260,000 ssf ofPWB 	\ . .	4-68
Table 4-39.   BOA for Transportation of Chemicals to the Surface Finishing
             Process Line		4-70
Table 4-40.   Costs of Critical Tasks	4-71
Table 4-41.   Materials Cost for the Non-Conveyorized Nickel/Gold Process  . . .	4-76
Table 4-42.   Chemical Cost per Bath Replacement for One Product Line of the
             Non-Conveyorized Nickel/Gold Process	4-76
Table 4-43.   Tiered Cost Scale for Monthly Wastewater Discharges to a POTW  ...	4-79
Table 4-44.   Summary of Costs for the Non-Conveyorized Nickel/Gold Process  	4-83
Table 4-45.   Total Cost  of Surface Finishing Technologies	4-84
Table 4-46.   Surface Finishing Alternative Unit Costs for Producing 260,000 ssf
             ofPWB	 .	. .  .. .	 . 4-86
Table 4-47.   CWA Regulations That May Apply to Chemicals in the Surface
             Finishing Process  ....••	4-89
Table 4-48.   Printed Circuit Board Facilities Discharging Less than  38,000 Liters per
             Day PSES Limitations (mg/L)  		.......	4-91
Table 4-49.   Printed Circuit Board Facilities Discharging 3 8,000 Liters per Day or
             More PSES Limitations (mg/L)  	4-91
Table 4-50.   PSNS for Metal Finishing Facilities	4-92
Table 4-51.   Amenable Cyanide Limitation Upon Agreement	4-92
Table 4-52.   PSES for All Plants Except Job Shops and Independent PWB
             Manufacturers ,.. :	4-92
                                         Xll

-------
 Table 4-53.   CAA Regulations That May Apply to Chemicals in the Surface
              Finishing Process	    4-93
 Table 4-54.   CERCLA RQs That May Apply to Chemicals in the Surface Finishing
              Process .	             4_97
 Table 4-55.   SARA and EPCRA Regulations That May Apply to Chemicals in the
              Surface Finishing Process	                 4,99
 Table 4-56.   TSCA Regulations and Lists That May Apply to Chemicals Used
              in Surface Finishing Processes ,.	............;... 4-100
 Table 4-57,   Summary of Regulations that May Apply to Chemicals Used in
              Hot Air Solder Leveling (HASL) Technology	4-102
 Table 4-58.   Summary of Regulations that May Apply to Chemicals Used in
              Nickel/Gold Technology	         4-103
 Table 4-59.   Summary of Regulations that May Apply to Chemicals Used in
              Nicfcel/Palladium/Gold Technology	.........:... 4-104
 Table 4-60.   Summary of Regulations that May Apply to Chemicals Used in
              OSP Technology  ...:	         4-105
 Table 4-61.   Summary of Regulations that May Apply to Chemicals Used in
              Immersion Silver Technology	                       4-106
 Table 4-62.    Summary of Regulations that May Apply to Chemicals Used in
              Immersion Tin Technology	 ..	      ..                 4-107
 Table 5-1.     Effects of Surface Finishing Technology on Resource Consumption	5-2
 Table 5-2.     Normalized Water Flow Rates of Various Water Rinse Types	      5-4
 Table 5-3.     Rinse Water Consumption Rates and Total Water Consumed by
              Surface Finishing technologies 	                  5,5
 Table 5-4.     Metal Deposition Rates and Total Metal Consumed by Surface
              Finishing Technologies	           5.7
 Table 5-5.     Energy-Consuming Equipment Used in Surface Finishing Process Lines	5-13
 Table 5-6.     Number of Surface Finishing Process Stages that Consume Energy by
              Function of Equipment	       5_14
 Table 5-7.     Energy Consumption Rates for Surface Finishing Equipment 	     5-15
 Table 5-8.     Hourly Energy Consumption Rates for Surface Finishing Technologies .	5-16
 Table 5-9.     Energy Consumption Rate per ssf of PWB Produced for Surface
              Finishing Technologies	            5_17
 Table 5-10.    Effects of Automation on Energy Consumption for Surface Finishing
              Technologies 	                     5_jg
 Table 5-11.   Pollution Resulting From the Generation of Energy Consumed by
              Surface Finishing Technologies	                   5_2o
 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	                       g_g
 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
                                        xtu

-------
Table 6-8.    Treatment Chemicals Used to Remove Metals From Chelated Wastewater .  . 6-34
Table 6-9.    Treatment Profile of PWB Surface Finishing Process Baths  ....		6-36
Table 7-1.    Surface Finishing Processes Evaluated in the CTSA . . .	 7-1
Table 7-2.    Surface Finishing Chemicals of Concern for Potential Occupational
             Inhalation Risk		7-4
Table 7-3.    Chemicals of Concern for Potential Dermal Risks ..	7-5
Table 7-4.    Aquatic Risk of Non-Metal Chemicals of Concern	7-7
Table 7-5.    Chemical Hazards	- -	7-8
Table 7-6.    Cost of Surface Finishing Technologies  		 7-11
Table 7-7.    Regulatory Status of Surface Finishing Technologies 	7-13
Table 7-8.    Energy and Water Consumption Rates of Surface Finishing Alternatives  .... 7-14
Table 7-9.    Glossary of Benefits/Costs Analysis Terms	 7-17
Table 7-10.   Overview of Potential Private and External Benefits or Costs	7-20
Table 7-11.   Overall Cost Comparison, Based on Manufacturing 260,000 ssf	7-21
Table 7-12.   Summary of Occupational Hazards, Exposures, and Risks of
             Potential Concern	• • • •	• •  • ^"^
Table 7-13.   Potential Health Effects Associated with Surface Finishing Chemicals of
             Concern	•	7r23
Table 7-14.   Number of Chemicals with Estimated Surface Water Concentration
             Above Concern Concentration	 7-26
Table 7-15.   Energy and Water Consumption of Surface Finishing Technologies  	7-27
Table 7-16.   Examples of Private Costs and Benefits Not Quantified  	7-28
Table 7-17.   Summary of Human Health and Environmental Risk Concerns for the
             HASL Technology	;	7-31
Table 7-18.   Number of HASL Chemicals Subject to Applicable Federal Regulations  .... 7-34
Table 7-19.   Summary of Human Health and Environmental Risk Concerns for
             the Nickel/Gold Technology	7-35
Table 7-20.   Number of Nickel/Gold Chemicals Subject to Applicable Federal
             Regulations	7-39
Table 7-21.   Summary of Human Health and Environmental Risk Concerns for the
             Nickel/Palladium/Gold Technology	 7-40
Table 7-22.   Number of Nickel/Palladium/Gold Chemicals Subject to  Applicable Federal
             Regulations	7-43
Table 7-23.   Summary of Human Health and Environmental Risk Concerns for the OSP
             Technology  	'....,	7-45
Table 7-24.   Number of OSP Chemicals Subject to Applicable Federal Regulations 	7-47
Table 7-25.   Summary of Human Health and Environmental Risk Concerns for the
             Immersion Silver Technology	7-48
Table 7-26.   Number of Immersion Silver Chemicals  Subject to Applicable Federal
             Regulations	7-50
Table 7-27.   Summary of Human Health and Environmental Risk Concerns for the
             Immersion Tin Technology	7-51
Table 7-28.   Number of Immersion Tin Chemicals Subject to Applicable Federal
             Regulations	; ...'..•	7-54
                                          XIV

-------
                                     List of Figures
  Figure 1-1.   PWBs Produced for World Market in 1998 (IPC) ..... /                   1_6
  Figure 1-2.   Number of PWBs Produced by U.S. Manufacturers in 1998 (IPC) '.'.'.'•'•""'  1-7
  Figure 2-1.   Typical Process Steps for Surface Finishing Technologies         	22
  Figure 2-2.   HASL Process Flow Diagram		26
  Figure 2-3.   Nickel/Gold Process Flow Diagram	        	2-8
  Figure 2-4.   Nickel/PaUadium/Gold Process Flow Diagram         	     211
  Figure 2-5.   OSP Process Flow Diagram	2-13
  Figure 2-6.   Immersion Silver Process Flow Diagram	   2-15
  Figure 2-7.   Immersion TinProcess Flow Diagram  		2-16
  Figure 3-1.   Schematic of Overall Material Balance for Surface Finishing Technologes     3-4
  Figure 3-2    Wastewater Treatment Process Flow Diagram	                        3-5
  Figure 3-3.   Generic HASL Process Steps and Typical Bath Sequence '."''	"  " 3_18
  Figure 3-4.   Generic Nickel/Gold Process Steps and Typical Bath Sequence    '•••••••• 3_2J
  Figure 3-5.   Generic Nickel/Palladium/Gold Process Steps and Typical Bath Sequence    3-24
  Figure 3-6.    Generic OSP Process Steps and Typical Bath Sequence ....            " 3-27
  Figure 3-7.    Generic Immersion Silver Process Steps and Typical Bath Sequence '•"•"" 3.29
  Figure 3-8.    Generic Immersion Tin Process Steps and Typical Bath Sequence          ' 3-32
  Figure 3-9.    Relationship Between Intake Rate and Blood-Lead Level for Both Adult
               and Fetus	                            3 73
  Figure 4-1.    Boxplot Displays for HCLV PTH Measurements (volts) at Pre-Test	
               by Surface Finish ,	-                        4 43
.Figure 4-2.    Boxplot Displays for HCLV PTH Post 85/85 - Pre-Test Measurements	 "
               (volts) by Surface Finish       	                         4 43
 Figure 4-3.   Boxplot Displays for HCLV PTH Post TS - Pre-Test Measurements
               (volts) by Surface Finish		                               4 44
 Figure 4-4.   Boxplot Displays for HCLV PTH Post MS - Pre-Test Measurements """
              (volts)  by Surface Finish	                            4 44
 Figure 4-5.   Boxplot Displays for HCLV SMT Measurements (volts) by Surface " " " "
              Fimsh	               i     4.45
 Figure4-6.   Boxplot Displays for HCLV PTH Post 85/85 -Pre-Test Measurements"
              (volts) by Surface Finish ..	                             4.45
 Figure4-7.    Boxplot Displays for HCLV PTH Post Ts'-Pre-Test Measurements"
              (volts) by Surface Finish	                          •  4 46
 Figure 4-8.    Boxplot Displays for HCLV PTH Post MS - Pre-Test Measurements	
              (volts) by Surface Finish	                           4 46
 Figure 4-9.    Boxplot Displays for HF PTH 50MHz Measurements (volts) bv   	
              Surface Finish	                  J            4 4J
 Figure4-10.   Boxplot Displays for HFPTH50MHz Post MS "-Pre-Test Measurements' "
              (volts) by Surface Finish ..	                                  4 47
 Figure 4-11.   Boxplot Displays for HF PTH f(-3dB) Post MS -Pretest Measurements
              (MHz) by Surface Finish       	                                  4 48
 Figure 4-12.   Boxplot Displays for HF PTH f(-40dB) P6st MS -Pretest Measurements
              (MHz)  by Surface Finish	            4_48
                                         XV

-------
Figure 4-13.   Boxplot Displays for HF SMT 50MHz Post Ms - Pre-Test Measurements
             (dB) by Surface Finish		4'49
Figure 4-14.   Boxplot Displays for HF SMT f(-3dB) Post MS - Pre-Test Measurements
             (MHz) by Surface Finish	,	•	4'49
Figure 4-15.   Boxplot Displays for HF PTH f(-40dB) Post MS - Pre-Test Measurements
             (MHz) by Surface Finish	•	4-50
Figure 4-16.   Boxplot Displays for HF TLC 50MHz Post MS - Pre-Test Measurements
             (dB) by Surface Finish		•	4'50
Figure 4-17.  Boxplot Displays for HF TLC 500MHz Post MS - Pre-Test Measurements
             (dB) by Surface Finish  	•	•  4'51
Figure 4-18.  Boxplot Displays for HF TLC RNR Post MS -Pre-Test Measurements
             (dB) by Surface Finish  .........	4'51
Figure 4-19.  Boxplot Displays for 10-Mil Pad Measurements (Iog10 ohms) at Pre-Test
             by Surface Finish	-	•	4'52
Figure 4-20.  Boxplot Displays for 10-Mil Pad 85/85 Pre-Test Measurements (Iog10 ohms)
             by SurfaceFinish  .	• • • • •	- -	4'52
Figure 4-21.  Boxplot Displays for PGA-A Measurements (Iog10 ohms) at Pre-Test by
             SurfaceFinish .....	4'53
Figure 4-22.  Boxplot Displays for PGA-B Measurements (Iog10 ohms) at Pre-Test by
             SurfaceFinish	4~53
Figure 4-23.  Boxplot Displays for the Gull Wing Measurements (Iog10 ohms) at Pre-Test
             by SurfaceFinish	 -	- - -	•	4'54
Figure 4-24.  Hybrid Cost Analysis Framework	- - - 4-56
Figure 5-1.   Water Consumption Rates of Surface Finishing Technologies	5-5
Figure 6-1.    Solder Reclaim System Diagram	6-21
Figure 6-2.   Flow Diagram of Combination Ion Exchange and Electrowinning
             Recovery System for Metal Recovery	6.-24
Figure 6-3.   Reverse Osmosis Water Reuse System	• • • 6-25
Figure 6-4.    TypicalPWB Waste Treatment System	6-32
 Figure 7-1.    Production Costs and Resource Consumption of Conveyorized
              HASL Technology		:	7'33
 Figure 7-2.    Production Costs and Resource Consumption of the Nickel/Gold
              Technology	• • •-	•	7"38
 Figure 7-3.    Production Costs and Resource Consumption of Nickel/Palladium/Gold
              Technology  	•  • •	7~43
 Figure 7-4.    Production Costs and Resource Consumption of OSP Technology	7-46
 Figure 7-5.    Production Costs and Resource Consumption of Immersion Silver
              Technology  		•	•	 7'49
 Figure 7-6.    Production Costs and Resource Consumption of Immersion Tin
              Technology	"	7-54
                                          xvi

-------
                                      Acronyms
 ABC
 ADD
 ADI
 ACGm
 AIM
 ANOVA
 ASF
 AT
 ATSDR
 BAT
 BCP
 EGA
 BKSF
 BOA
 BOD
 BPT
 Btu
 BW
 CAA
 CC
 CCAMTF
 CDC
 CEB
 CERCLA
 CFR
 CO
 CO2
 COB
 CSL
 CTSA
 CuSO4
 CWA
 DCT  .
 DfE
 ED
 EDTA
 EF
 EMS
 EPA
 EPCRA
HCLV
HVLC
HSD
 activity-based costing
 average daily dose
 acceptable daily intake
 American Conference of Governmental Industrial Hygenists, Inc.
 Adult Lead Methodology)
 an analysis of variance
 alternative surface finishing
 averaging time
 Agency for Toxic Substances and Disease Registry
 best available control technology economically achievable
 best conventional pollution control technology
 ball grid array
 biokhietic slope factor
 bill of activities
 biological oxygen demand
 best practicable control technology currently available
 British Thermal Units
 body weight
 Clean Air Act          .
 concern concentration
 Circuit Card Assembly and Materials Task Force
 Centers for Disease Control and Prevention
 Chemical Engineering Branch
 Comprehensive Environmental Response, Compensation and Liability Act
 Code of Federal Regulations
 carbon monoxide
 carbon dioxide
 chip on board
 Contamination Studies Laboratories, Inc.
 Cleaner Technologies Substitutes Assessment
 copper sulfate
 Clean Water Act
 dimethyl-dithiocarbamate
 Design for the Environment
 exposure duration
 ethylenediaminetetraacedic acid
 exposure frequency
 Environmental Management System
 Environmental Protection Agency
 Emergency Planning and Community Right-to-Know Act
 high current low voltage
high voltage low current
high speed digital
                                        xvii

-------
HFLPF      high frequency low pass filter
EDF 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       sulfuricacid
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
IETJBK      integrated exposure uptake biokinetic model for lead in Children
IPC          Institute for Interconnecting and Packaging Electronics Circuits
IQR         interquartile range                                 .
IR          intake rate
IRIS         Integrated Risk Information System
ISO          International Standards Organization
KUB         Knoxville. Utilities Board
kW          kilowatt
LADD       lifetime average daily dose
LOAEL      lowest-observed-adverse-effect level
LR          low residue
LRSTF      Low-Residue Soldering Task Force
LSD         least significant difference
LT          lab technical
MACT       Maximum Achievable  Control Technology
MHC        making holes conductive
MnO2        manganese dioxide
MOE        margin of exposure
MRL         minimum risk level
MSDS       material safety data sheet
MTL         Master Testing List
NaOH        caustic soda
NCP         National Contingency Plan
NIOSH      National Institute for Occupational Safety and Health
NOAEL      no-observed-adverse-effect level
                                        XVlll

-------
  NOx
  NPDES
  NPDWR
  NSDWR
  NSPS
  NTP
  ON
  OSP
  OSHA
  PEL
  PDR
  POTW
  PPE
  PSES
  PSNS
  psi
  PTH
  PWAs
  PWB
  RCRA
  REL
  RfC
  RฃD
 RO
 RTECS
 RQ
 SAR
 SARA
 SAT
 SDWA
 SERC
 SF
 SIC
 SMT
 SOX
 SPC
 ssf
 STEL
 SW
 TMT15
 TLC
 TLV
 TPY
TRI
TS
TSCA
  oxides of nitrogen
  National Pollutant Discharge Elimination System
  National Primary Drinking Water Regulations
  National Secondary Drinking Water Regulations
  New Source Performance Standards
  National Toxicology Program
  other networks
  organic solderability preservative
  Occupational Safety and Health Administration
  permissible exposure limit
  potential dose rate
  publicly owned treatment work
  personal protective equipment
  pretreatment standards for existing xources
  pretreatment standards for new sources
  per square inch
  plated-through holes
  printed wiring assemblies
  printed wiring board
  Resource Conservation and Recovery Act
  recommended exposure level
  reference concentration
  reference dose
  reverse osmosis
  Registry of Toxic Effects of Chemical Substances
  reportable quantity
  structure-activity relationship
  Superfund Amendments and Reauthorization Act
  Structure-Activity Team
 Safe Drinking Water Act
 State Emergency Response Commission
 slope factor
 standard industrial code
 surface mount technology
 sulfur oxides
 statistical process control
 surface  square feet
 short-term exposure limit
 stranded wire
tri-mercaptotriazine
transmission line coupler
threshold limit value
tons per year
Toxic Release Inventory
thermal shock
Toxic Substances Control Act
                                         XIX

-------
TSDS        treatment, storage, or disposal facility
TWA        time-weighed average
UC          unit cost
UF          uncertainty factor
UT          University of Tennessee
UR          utilization ratio
VOC        volatile organic compounds
WHO        World Health Organization
WO A        weak organic acid
WOE        weight-of-evidence
WS         ' water soluble
                                          XX

-------
                               Executive Summary
        The Printed Wiring Board Surface Finishes Cleaner Technologies Substitutes
 Assessment:  Volume lisa 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
 mckel/immersion gold (nickel/gold), electroless nickel/immersion palladium/immersion gold
 (mckel/paUadmm/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 H
 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
       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 redaction and pollution prevention opportunities.
DfE works on a voluntary basis with industry sectors to
evaluate the risks, performance, costs, and resource
requirements of alternative chemicals, processes, and
technologies.

  Additional goals of the program include:

•   Changing general business practices to incorporate
    environmental concerns.
•   Helping individual businesses undertake
    environmental design efforts through the application of
    specific tools and methods.

DfE Partners include:

    industry;
    professional institutions;
    academia;
    public-interest groups; and
    other government agencies.
                                            ES-2

-------
                                                                    EXECUTIVE SUMMARY
  tt     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 P WB 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
Nbn-Conveyorized
•
•
•
•

•
' ' Conveyorized
•


•
•
•
HI.   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

-------






















1
~

11
Jg

oo
o .

ill
glฐ

3s
*

g.
s



i
a

i
OS
8

I
C

|
e

i
6

1
3






>,


>>,


>,





>,




tฃ
.-
M

1
o
1

P


1
*ซJ

|
*

1




>
^




^









'



1
1

f
1

I
h
*,

•8
t
is
U

•s
^






^
^


'




>,


>,







a
9>

a
I
1

fe
O

a
P


.&
^
^c

s




•^
-^

s



>.
•





•






f



•

i
*

9B
1
s
•ft

1
3

1
ui '






f?
P


'

i
g

i
g


•









t
p
 O
•4*3 fiฃrf
aa ^*
tn 2
co CH
O 4**
S ฐ5
CL, a;,
a 2
ปg S
OS ^^
co .Si
•^ S?
^j pH
f j
"S ซ
g =
•a -2?
i ^
11
1 1
| --a
e^
1 .9 .
S-( T3
OS
s|

ES-5

-------
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 85 ฐC/85 percent humidity), followed by
thermal shock testing, and mechanical shock testing to distinguish variability in the performance
of the surface finish.  Comparative costs of the surface finishing technologies were estimated with
a hybrid cost model that combines traditional costs with simulation modeling and activity-based
costs.  Costs are presented in terms of dollars per surface square feet (ssf) of PWB produced.

       Federal environmental regulatory information is presented for the chemicals in the surface
finishing technologies.  This information is intended to provide an indication of the regulatory
requirements potentially associated with a technology, but not to serve as regulatory guidance.

       Quantitative resource consumption data are presented for the comparative rates of energy
and water use of the surface finishing technologies. The consumption of other non-renewable
resources such as process chemicals and metals also are analyzed.
                                          ES-6

-------
                                                                    EXECUTIVE SUMMARY
        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 H of the CTSA 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 One dimensions, and checked survey
        data for accuracy,
 •       The Supplier Data Sheet, which included  information on chemical cost, equipment cost,
       water consumption rates, product constraints, and the locations of test sites for the
       Performance Demonstration.

 Chemical Information

       Appendix B presents chemical properties and selected environmental  fate properties for
the non-proprietary chemicals in surface finishing chemical products.  Proprietary chemical
ingredients are not included to protect proprietary chemical identities.  Properties that were
measured or estimated (using a variety of standard EPA methods) included melting point,
solubility, vapor pressure, octanol-water partition coefficient, boiling point, and flash point.
                                          ES-7

-------
EXECUTIVE SUMMARY
These properties can be used to determine the environmental fate of the surface finishing
chemicals when they are released to the environment.

Health Hazard Assessments

       Inherent in determining the risk associated with the surface finishing chemicals is a
determination of the hazard or toxicity of the chemicals. Human health hazard information for
non-proprietary chemicals is presented in Section 3.3. Detailed toxicity data for proprietary
chemicals are not included to maintain the secrecy of the proprietary chemical formulations.
Many of the chemicals in the surface finishing chemical products have been studied to determine
their health effects, and data from those studies are available in published scientific literature.  In
order to collect available testing data for the surface finishing chemicals, literature searches were
conducted using standard chemical references and online databases, including EPA's Integrated
Risk Information System (IRIS) and the National Library of Medicine's Hazardous Substances
Data Bank (HSDB).

       For many of the chemicals, EPA has identified chemical exposure levels that are known to
be hazardous if exceeded or met (e.g., no- or lowest-observed-adverse-effect level [NOAEL or
LOAEL]), or levels that are protective of human health (reference concentration [RfC] or
reference dose [RfD]).  These values were taken from online databases and published literature.
For many of the chemicals lacking toxicity data, EPA's Structure-Activity Team (SAT) estimated
human health concerns based on analogous  chemicals.  Hazard information is combined with
estimated exposure levels to develop an estimate of the risk associated with each chemical.

Ecological Hazard Assessments

       Similar information was gathered on the ecological effects that may be  expected if surface
finishing chemicals are released to water. Acute and chronic toxicity values were taken from
online database searches (TOXNET and ACQUIRE), published literature, or were estimated
using structure-activity relationships if measured data were not available. Based on the toxicity
values, surface finishing chemicals were assigned concern concentrations (CCs).  A CC is the
concentration of a chemical in the aquatic environment which, if exceeded, may result in
significant risk to aquatic organisms.  CCs were determined by dividing acute or chronic toxicity
values by an assessment factor (ranging from one to 1,000) that incorporates the uncertainty
associated with toxicity data.

Limitations

       There are a number of limitations to the project, both because of the limit of the project's
resources, the predefined scope of the project, and uncertainties inherent to risk characterization
techniques.  Some of the limitations related to the risk, competitiveness, and conservation
components of the CTSA are summarized below. More detailed information on limitations and
uncertainties for a particular portion of the assessment is given in the applicable sections of this
document.  A limitation common to all components of the assessment is that the surface finishing
                                          ES-8

-------
                                                                    EXECUTIVE SUMMARY
 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 260,000  ssf of PWB. As with the risk characterization, this approach
 results in a comparative evaluation of cost, not an absolute evaluation or determination. The cost
 analysis focuses on the private costs that would be incurred by facilities implementing a
 technology.  However, the analysis is limited to costs that are solely attributable to the surface
 finishing process and does not evaluate costs associated with product quality or wastewater
 treatment. Community benefits or costs, such as reduced health effects to workers or the effects
 on jobs from implementing a more efficient surface, finishing technology, also are not quantified.
 The Social Benefits/Costs Assessment (see Section 7.2), however, qualitatively evaluates some of
these external benefits and costs.

       The regulatory information contained in the CTSA may be useful in evaluating the benefits
of implementing processes which no longer contain chemicals that trigger compliance issues;
however, this document is not intended to provide compliance assistance.  If the reader has '
                                          ES-9

-------
EXECUTIVE SUMMARY
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 cost analysis, consumption rates were
estimated based on using a surface finishing technology in a model facility to produce 260,000 ssf
ofPWB.
IV.    CLEANER TECHNOLOGIES SUBSTITUTES ASSESSMENT RESULTS

Occupational Exposures and Health Risks

       Health risks to workers are estimated for inhalation exposure to vapors and aerosols from
surface finishing baths and for dermal exposure to surface finishing bath chemicals. Inhalation
exposure estimates are based on the assumptions that emissions to indoor air from conveyorized
lines are negligible, that the air in the process room is completely mixed and chemical
concentrations are constant over time, and that no vapor control devices (e.g., bath covers) are
used in non-conveyorized lines. Dermal exposure estimates are based on the conservative
assumptions that workers do not wear gloves and that all non-conveyorized lines are operated by
manual hoist. Dermal exposure to line operators on non-conveyorized lines is estimated for
routine line operation and maintenance (e.g., bath replacement, filter replacement), and on
conveyorized lines for bath maintenance activities alone.

       Based on the number of chemicals with risk results above concern levels, some
alternatives to the non-conveyorized HASL process appear to pose lower occupational risks
(immersion silver, conveyorized and non-conveyorized immersion tin, and conveyorized HASL),
some may pose similar levels of risk (conveyorized and non-conveyorized OSP), and some may
pose higher risk (nickel/gold and nickel/palladium/gold). Surface finishing chemicals of concern
for potential occupational risk from inhalation are shown in Table ES-2.

       There also are occupational risk concerns for dermal contact with chemicals in the non-
conveyorized HASL, nickel/gold, nickel/palladium/gold, OSP, and immersion tin processes, and
the conveyorized HASL and OSP processes. Table ES-3 presents chemicals of concern for
potential occupational risk from dermal contact.
                                         ES-10

-------
                                                                                  EXECUTIVE SUMMARY
                Table ES-2. Surface Finishing Chemicals of Concern for Potential
Chemical
^ "• s
Alkyldiol
Ethylene glycol
Hydrochloric acid
Hydrogen peroxide
Nickel sulfate
Phosphoric acid
Propionic acid
Process (Non-Conveyorizedf 260,000 ssf)a
HASL

•





Mckel/Gold
•

•
•
•
•

Nickel/Pailadiiim/Gold
•

•
•
•
•
•
OSP

•





   Non-conveyorized immersion silver process not evaluated. Occupational exposure and risk from all conveybrized
 process configurations are below concern levels.
 • Line operator risk results above concern levels (non-cancer health effects).
Table ES-3. Chemicals of Concern for Potential Dermal Risks
Chemical
;
A I
Ammonia compound A
Ammonium chloride
Ammonium hydroxide
Copper ion
Copper salt C
Copper sulfate pentahydrate
Ethylene glycol monobulyl ether
Hydrogen peroxide
Inorganic metallic salt B
Lead
Nickel suliate
Urea compound C
HASL
0*9





••



X


HASL
(0





••



X


Nickel/Gold
(NO

•
•


••

•
••

••

" Nickel/
Palladium/Gold
~(NQ
•

•


••

•
••

••

OSB
(NQ
•


••
••
••






OSP
(C)



••
•
••






Immersion
Tm
(NQ






•




•
                                                        ซ<  	-——	— — _..V|U *,ซ. , •*,*. VM. WV4XV wjvy^.j^jw\4 i*^Alljlvl.ijJHJJLi U
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.                                    .
C: Conveyorized (horizontal) process configuration.
NC:  Non-conveyorized (vertical) process configuration.
                                                  ES-11

-------
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 Ijfe time cancer risk of 2 x 10'7 (one in five million) based on high end exposure.
Cancer risks less than 1 x 1CT6 (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 hi 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 hi tin-lead
solder in the HASL process. Thiourea.is used hi the immersion tin process. Urea compound B, a
confidential ingredient hi 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).  Sulfimc
acid is used hi diluted form hi every surface finishing process hi 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 onNOAEL lower than 100, or MQE based
on a LOAEL lower than 1,000 - were estimated, for occupational exposures to chemicals hi 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
 classified as a human carcinogen.2 All hazard quotients are less than one for ambient exposure to
 the general population, and all MOEs for ambient exposure are greater than 1,000 for all
 processes, indicating low concern from the estimated air concentrations for chronic non-cancer
 effects.

        Estimated ambient air concentrations of lead from a HASL process are well below EPA
 air regulatory limits for lead, and risks to the nearby population from airborne lead are expected to
 be below concern levels.

 Ecological Hazards

        Ecological risk indicators (RIECO) were calculated for non-metal surface finishing
 chemicals that may be released to surface water. Risk indicators for metals are not used for
 comparing alternatives because it is assumed that on-site treatment is targeted to remove metal so
 that permitted concentrations are not exceeded.  Estimated surface water concentrations for non-
 metals exceeded the CC for the processes as shown in Table ES-4. CCs are discussed in more
 detail in Section 3.3.3
Table ES-4. Aquatic Risk of Non-Metal Chemicals of Concern
Chemical v
1,4-Butenediol
AJkylaryl imidazole
Alkylaryl sulfonate
Hydrogen peroxide
Potassium peroxymonosulfate
HASL
(NQ
•

•
•
•
HASL
(Q


•
•
•
OSP
(Ncr

•



OSP
CQ

•



Immersion Silver
(Q



•

Immersion Tin
(NQ




•
Estimated surface water concentration > CC after publicly owned treatment works (POT W) treatment.
   f)                                                   „          -                 •
     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,ofMSDSV
-iv^
•fi.
33
19
18
9
4
14
V - - ' "-i Hazardous "Property' \ *>
.. E '
1


1


C






JE
1



1
1
JEH
3


2
1

CO
4
8
12
4
2
7
o
1
1
1
1
1

SRP
1
1
1
1


II
1



1

* For alternative processes with more than one product line, the hazard data reported represent the most hazardous bath
of each type for the two product lines (e.g., of the microetch baths from the two product lines, the one with the most
hazardous chemicals is reported).
b Formulations for HASL process baths were unavailable because cleaner and microetch bath chemistries are not made
specifically for the HASL process. Hazards reported for HASL bath types were reported as the worst case of the results
of similar baths from other processes.
F = Flammable; C = Combustible; E = Explosive; FH = Fire Hazard; CO = Corrosive; O = Oxidizer; SRP = Sudden
Release of Pressure; U = Unstable.
       Several unique process safety concerns arise from the operation of the HASL process.
Solder eruptions during start-up can lead to solder splattering onto workers causing serious burns.
The HASL process also poses a fire hazard due to the build-up of residual carbon from the use of
oil-based flux or other flammable materials. Other safety concerns include worker exposure to
acids in the flux, accidental contact with the molten solder, or exposure to the other chemical
hazards on the process line.

       Work-related injuries from equipment, improper use of equipment, bypassing equipment
safety features, failure to use personal protective equipment, and physical stresses that may appear
gradually as a result of repetitive motion are all potential process safety hazards to workers.
Reducing the potential for work-related injuries is critical in an effective and ongoing safety
training program.  Appropriate training can help reduce the number of work-related accidents and
injuries regardless of the technology used.
                                           ES-14

-------
                                                                    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 rion-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 hi 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 hi
each surface finishing technology that are subject to federal environmental regulations. Different
chemical suppliers of a technology do not always use the same chemicals in their particular
product lines.  Thus, all of these chemicals may not be present in any one product line.

Resource Conservation

       Energy and water consumption rates were evaluated for each of the surface finishing
process alternatives. Other resource consumption by the surface finishing technologies was
evaluated qualitatively due to the variability of factors that  affect the consumption of these
resources. Table ES-8 presents the energy and water consumption rates of the surface finishing
technologies.
    3  In some cases, state or local requirements may be more restrictive than federal requirements. Due to resource
 limitations, however, only federal regulations were reviewed.

                                           ES-16

-------
                        EXECUTIVE SUMMARY
-6. Cost Analyses Results B
 eo
;ฃ t— O O O O
j? o5 O O O O
•
3 r oฐ oo <-< en
^^ JC O O ^5 ,_<
i- S "•* ^
to * T-H
II
^g - oo -sr ON en
y ฐg 0 0 0 0
t i 09 69 69 09
ซ ! SS o? o?
2: en r~- oo
a! ' *s
1*
^( g O CS O T-H
- ซ9
O9 69 69 69
iS::;;:;;f;^u;p4!e&;FMwhซng1T^n^
HASL, Non-conveyorized (BASELINE)
HASL, Conveyorized
Nickel/Gold, Non-conveyorized
Nickel/Palladium/Gold, Non-conveyorized
sp sp sO sp sp
O^s ฃ- tfs ฃ~- ฃs
en ON i— i vo n-
10 VO _ fsj 00
00000
o o o o o
o o o o o
69 69 69 69 69
OO ^* ^^ VO T~H
en v^ >o ^f *
o o o o o
o o o o o
o o o o o
69 09 09 69 69
<0  ON T-I ^H
r^ c^ CN vo vo
* i i i i
ป— i CQ en CN <— i
t*~ t^ O *—< trH
0 0 CM TTH r-i
O C5 O O O
69 09 69 09 69
sO sp sp sP sP
0s- 0s- 0^ 0^ S--
O OO t^ <— i r> •^>
o o o o o
O O O O 0
09 09 09 69 09
sO sP sO sO sp
ON tS ts) 0 T-I
vo t^ CN wp en
>— 1 O OO OO -< CS 1-1 fs)
O O O O CD
&9 09 09 69 09 '
v^ar, iNon-conveyonzea
OSP, Conveyorized
Immersion Silver, Conveyorized
Immersion Tin; Non-conveyorized
Immersion Tin, Conveyorized

S
,9
e
1
5
1
•ct
1
1
U
I
1
i
2
I
3
i
3
n
3
1
K
2

•
3
(i
o
I
<<-,
J
•<
t
en
ซ
1
e
1
35



c
o

c

cS

<*>
i
i
;=|
<&
13
a>
j
3
ฃป
3
-f
yi
t
ง
-M

275%
CN
C
o
ฃe
ae
ro.
VO
c~
o
VO
OO
X
09
0
i
X
w
Nickel/Gold, Non-conveyorized

e^
O
o
oo
o
o
09
e<-

c
o
sj
OS
r<
•^i
w
I
1
X
w
Nickel/Palladmm/Gold, Non-conveyorized

0s
en
e<-
c
o
O9

m
c
^
o
en
oe-
?f
i
X
ft=h
UbP, Non-conveyonzed,

o^
en
en
oo
o
o
o
09
v>
vo
VO
c
SN
oo
v%
I

c
o"
09
eN
O
t-_
t~-
cs
c;
o
y
^
VO
09
^P
0^
X
t— i

immersion Tin, Non-conveyorized |
Tmmersinn Tin CnmrmnriveA 1
^
t~
c
o
O9
^
so
^

ri
^
s
U
"a3
-i US
3 rO
3 "S
I
3)
3
5 5
! I
5 S.
5 -g
if
4 O
!
H
ES-17

-------
EXECUTIVE SUMMARY






!?3
tS
w
o
0
•s
r
C**
01
4g
2
ซฃ2
n
E
1
CO
*S
CO

ฃ?

CO
s?
o
13
i
CO

3
cff
o

Q.
O
'
•s
*5P
1
tn
•a
0
"S
oj
O

o
i.
a>
S
ฃ.







s
1


u
r/j
H






,

8
p..
m




^f
•^
u






^
g
^


3
ป:"
*•'.
OS I-J
oo •<
PL,
^
^

Qjj
•o-fl
OO ซ/2
- ป
01 '
?•?
to -

o
f-H

S

r*

A
.ซ* .
,_,


T-t ,
'^

H
^ '
efj ,
CS
O
JD
O-

en "3
en o
cj ฐ3
S ซ

'
•
CO

^



CO


CO


^


VO

T 	 1


CO



CO


T-H
^


t-H

T-H


^
EH
•
•
CO

-*



T-H


CO


^


t— <

—


vo



1 — 1


VO
vo


vo

vo

•o
ickel/Gol
*Z
i
•-
^.

^.



\ — i


CO


VO


o
* — I

—


1/-J



<0


ซ
cs


^
-
ซ0
"o
53
Rt
1
"55
.'
'
T-H

cs



,-H


OJ*


cs


-

T-H


cs



CO


CS



cs

cs


PH
  •
                                       wJ    F~^ irH
                                       I
                                       ^3
                                       ^
              c> —<
              en fi
ฃฃฃ
&&&
u a o
     < ^ <
ซ+ << K K P5
S S o o o
^ ^ D-i Pu Cu
o o S m S
                                           ES-18

-------
                                                                   EXECUTIVE SUMMARY
       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-
 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.
Table ES-8. Energy and Water Consumption Rates of Surface Finishing Technologies
Process Type
^ ~~">-'?^=-ni^~~*i> --^
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) l
1.24
0.99
2.06
3.61
0.77
0.53
0.53
1.81
0.88
Energy Consumption
- * (Btu/ssf)
218
133
447
768
125
73
287
289
522
       Energy consumption by the surface finishing technologies was driven primarily by the
overall throughput efficiency of the technology.  Although HASL had the highest BTU per hour
rate of all the technologies, after normalizing the rate using the overall throughput (260,000 ssf),
only the OSP (conveyorized and non-conveyorized), and the conveyorized HASL were more
energy efficient than the HASL process. It also was found that for alternatives with both types of
automation, the conveyorized version of the process is typically the more energy efficient (HASL
and OSP), with the exception of the immersion tin process.

       The rate of deposition of metal was calculated for each technology along with the total
amount of metal consumed for 260,000 ssf of PWB produced. It was shown that the
consumption of close to 300 pounds of lead (per 260,000 ssf) could be eliminated by replacing the
baseline HASL process with an alternative technology (see Section 5.1, Resource Conservation).
In cases where waste solder is not routinely recycled or reclaimed, the consumption of as much as
2,500 pounds of lead (per 260,000 ssf) could be  eliminated by replacement of the HASL  process.
Although several of the alternative technologies rely on the use of small quantities of other metals
(especially nickel, palladium, gold, silver, and tin) the OSP technology eliminates metal
consumption entirely.
                                         ES-19

-------
EXECUTIVE SUMMARY
Social Benefits/Costs Assessment

       The social benefits and costs of the surface finishing technologies were qualitatively
assessed to compare the benefits and costs of switching from the baseline technology to an
alternative, while considering both the private and external costs and benefits.  Private costs
typically include any direct costs incurred by the decision-maker and are generally reflected in the
manufacturer's balance sheet.  By contrast, external costs are not borne by the manufacturer, but
by society. Therefore, the analysis considered both the impact of the alternative surface finishing
processes on the manufacturer itself (private costs and benefits) and the impact the choice of an
alternative had on external costs and benefits.

       Table ES-9 presents an overview of potential private and external benefits and costs
associated with the operation of the surface finishing line. Changes in the surface finishing
technology employed could potentially result in a net benefit (a change in a beneficial direction) or
cost ( a change in a detrimental direction) in each of the categories listed below. The type of
change and the magnitude will vary by facility.

       Table ES-9. Overview of Potential Private and External Benefits or Costs
Evaluation
Category
Manufacturing Costs
Occupational health/
Worker risk
Public Health/
Population risk
Wastewater and
Ecological risk
Energy use
Water use
Private Benefit or Cost? \
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 *
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).
                                           ES-20

-------
                                                                   EXECUTIVE SUMMARY
        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.


 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/paUadium/gold, OSP, immersion silver  and immersion
 tin.

       The results of the CTSA analyses of the surface finishing technologies were mixed.
 Analyses  of process costs, energy, and natural resource consumption each showed that some
 alternatives performed better than HASL, while others did not.  An evaluation of potential
 occupational risks from both inhalation and dermal exposures indicated that several alternatives
 posed lower occupational risks than HASL, based on the number of chemicals with risk results
 above concern concentrations. Ecological risks posed by the alternatives were all lower than the
 HASL process, also based on the number of chemicals exceeding concern concentrations  None
 of the surface finishing technologies, including HASL, posed a risk to populations living nearby
 Finally, alternatives to the traditional non-conveyorized HASL technology (the baseline process)
 were demonstrated to perform as well as HASL during performance testing; however, several of
 the alternatives improve upon the technical limitations of the HASL finish (e g  wire-bondabilitv
 surface planarity).

       Table ES-10 summarizes the CTSA analyses results for the surface finishing technologies
 relative to the non-conveyorized HASL baseline. It is important to note that there are additional'
factors beyond those assessed in this CTSA that individual businesses may consider when
choosing among alternatives. The actual decision of whether or not to implement an alternative is
made outside of the CTSA process.
                                         ES-21

-------
EXECUTIVE SUMMARY






<
.ฃ
"<
c
G
w
c
a
SM
JW
hx
*^
en
C
*
I
2

c
j:
o
"c
fฃ
a
1
a
'S
V.
V.
"O
ec
42
1
a

>
**•ซ
1
0
CO
1

e
o
It!
ซ งs
•a
1

CJป gc OT
|ll
5^






e
at.
1
a
O'
en
•a
u
S""*

o>
a
o
s-
1
12;



^

"*•*"
ง -fj*
^*- •**"(•
-< 'R
•ง>H
S
O9
•••* cS
"1 •ฃ
ft, pg
S
S3


ฃ
&i
ซ ซ
** S
{ฃ '

8
0
-fi or &^
-|S|
8
Pi


o
"o
e
"i
_ i
1
u
a
Q








•^



a
e
bt
s
1,
g-
•J3
OS
>a
,e



•a
to
Q
O
•43
cs
•a
1
-












oo




^J.
o






^



0





01



ป-H


n
o
ee



|
9
S
*rt
n-Conveyorizei
0
1


+





+






+



II





II



+


11







1
>•
^
3
•\
ซ


-ป





— ^






+



II





— *



— ป


__>






g
1
3
NickeVGol


—





^






t



n





z:



^


-



"o
1
ฃ
6
ง
ladium/Gold, N
Nickel/Pall


+





+






J



n





-ป



n


+







conveyorized
1
3
O


j





+






t



u





n



+


t







1
3
*\
O








+






t



II





+



+


+






1
1
|
B"
Immersion








.-ป






+



H





n



+


-h





"O
1
i
[Immersion


-V





+






t



II





+



+


+






1
1
3
3
[Immersion

T3
s
J s*
3 "o

OH O
a €
o 
• 5 *^
"S 5
ง 13
ซ 8
2 s

a> a
•ฐ CM
>> g
cd O
S S
•3 2
.3 ^
s ฐ^
1 1 d

's -g .s
o> o "S
a & *^
O Q>  Q, =W S
S 0 ฐ J3
*> <1> ^ f!T
M.S J3 O
t> '"u ^ c^
^j ^u o> a
i; 'S ^ . ง
^ (jj cs ปy*j
"S ซ 6 a
I11-S
1 OT j_* - O
i j||

2 -a ?? ti
logies with more
st exposed indivi
e risk characteriz
son shown in this
" for techno
b For the mo
0 Because th
The compari
                                                              
1
S













d>
I
^
<1>
B
S
a,
o
0
^
o
I-H














Q>
2
i
l
ง
n.
0
o


S
^>
1

^

l-H
Some benefll
oป
3
x>
e
<ง

C/J
ca
ง
u
T3
S-<
S
ts
&
oo
&
-+— ป
a
• 8
f-4
O
&•
O
•0
+
ซ"
Greater bene
                                                M II
                                    ES-22

-------
                                                                       EXECUTIVE SUMMARY
        To assist PWB manufacturers who are considering the implementation of an alternative
 surface finish, the DfE PWB Project has prepared an implementation guide that describes lessons
 learned by other PWB manufacturers who have begun using an alternative surface finishing
 process.4 In addition, the University of Tennessee Department of Industrial Engineering can
 provide technical support to facilities that would like to use the cost model developed for the
 CTSA to estimate then- own manufacturing costs should they switch to a surface finishing
 alternative.
     Implementing Cleaner Printed Wiring Board Technologies: Surface Finishes (EPA 744-R-00-002, March
2000). This and other DfE PWB Project documents can be obtained by contacting EPA's Pollution Prevention
Information Clearinghouse at (202) 260-1023 or from www.epa.gov/dfe/pwb.

                                           ES-23

-------
EXECUTIVE SUMMARY
                                   REFERENCES
IPC (IPC-Association Connecting Electronics Industries).  1996. National Technology Roadmap
for Electronic Interconnections.

Kincaid, Lori E., Jed Meline and Gary Davis.  1996.  Cleaner Technologies Substitutes
Assessment: A Methodology & Resource Guide. EPA Office of Pollution Prevention and Toxics.
Washington, D.C. EPA 744-R-95-002. December.

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

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

-------
                                     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 CTSA, developing project workplans, donating time, materials, and their
 manufacturing facilities for project  research, and reviewing technical information contained in this
 CTSA.  Much of the process-specific information presented here was provided by chemical
 suppliers to the PWB industry, PWB manufacturers who responded to project information
 requests, and PWB manufacturers who volunteered their facilities for a performance
 demonstration of the baseline and alternative technologies.

       Section 1.1 presents project background information,  including summary descriptions of
the EPA DfE Program and the DfE PWB  Project.  Section 1.2 is an overview of the PWB
industry, including the types of PWBs produced, the market for PWBs, and the overall PWB
manufacturing process. Section 1.3 summarizes the CTSA methodology, including a discussion
of how technologies were selected for evaluation in the CTSA, the boundaries of the evaluation,
issues evaluated, data sources, and project limitations. Section 1.4 describes the organization of
the remainder of the CTSA document.
                                          1-1

-------
1.1 PROJECT BACKGROUND
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 hi more detail below.

1.1.1  EPA DfE Program

       EPA's Office of Pollution Prevention and Toxics created the DfE Program hi 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,
pubh'crinterest 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 hi
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 processes that generate significant amounts of hazardous waste. The study
 concluded that effective collaboration among government, industry, academia, and the public is
 imperative to proactively address the needs of environmental technologies, policies, and practices
 (MCC, 1993).  To. follow-up, the industry embarked on a collaborative effort to develop an
 environmental roadmap for the electronics industry.  The roadmap project involved more than 100
 organizations, including EPA, the Department of Energy, the Advanced Research Projects
 Agency, and several trade associations.  The PWB industry national trade association, the IPC-
, Association Connecting Electronics Industries (TPC), 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 pubUc-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 Roadmapfor Electronic Interconnections (TPC, 1996).  The roadmap detailed trends
in PWB manufacturing and assembly technologies, and forecasted the technology needs for the
industry over the immediate future. The study concluded that major efforts are needed to
overcome the reluctance to trying new and innovative ideas, citing the environmental pressure to
reduce hazardous waste and the use of lead. The results also cited the development of non-
tin/lead metallic or organic coatings to retain solderability characteristics as an industry need over
the near term.               ..                                         •

       Recognizing the importance of reducing lead consumption in the PWB industry, and
building on the strong partnerships established during the previous work, the PWB surface
finishing project was begun in 1997 to evaluate alternative surface finishing technologies to
HASL. This CTSA is a culmination of this effort. During this time, the project has also:

•      Prepared several additional case studies of pollution prevention opportunities (U.S. EPA,
       1997a; U.S. EPA, 1997b; U.S. EPA, 1997c; U.S. EPA, 1999).
•      Prepared an implementation guide for PWB manufacturers interested in switching from
       HASL to an alternative surface finishing technology (U.S. EPA, 2000).
•      Identified, evaluated, and disseminated information on viable pollution prevention
       opportunities for the PWB industry through an updated review of a pollution prevention
       and control practices industry study (U.S. EPA, 1998b).

Further information about the project, along with web-based versions of all the documents listed
above and other previous project work, can be obtained by visiting the Design for the
Environment Program websit, 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 surfece 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 hi this CTSA.  Thus, seven categories of technologies were carried forward for further
 evaluation hi 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 hi terms  of the overall life cycle of the surface finishing products and hi 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 hi Section  1.2.3, rigid, multi-layer PWB
 manufacturing encompasses a number of process steps, of which the surface finishing process is
 the last one.

       The activities evaluated in this study are primarily the use of surface finishing chemicals at
PWB facilities and the release or disposal of surface finishing chemicals from PWB facilities.
However, in addition to evaluating the energy consumed during surface finishing line operation,
the analysis of energy impacts (Section 5.2) also discusses the pollutants generated from
producing the energy to operate the surface finishing line, as well as energy consumed in other
life-cycle stages,  such as the manufacture of chemical ingredients. In addition, information is
presented on the  type and quantity of wastewater generated by the surface finishing process line,
                                           1-9

-------
1.3 CTSA METHODOLOGY
and the risk to the environment resulting from the discharge of the wastewater to nearby surface
water (Section 3.4). Finally, while information is presented on the generation and disposal of
solid waste from surface finishing technologies, there was insufficient information to characterize
the risk from these environmental releases.  This is discussed in more detail in Section 3.1, Source
Release Assessment.

       In terms of the PWB manufacturing process, this analysis focused entirely on the surface
finishing process, defined as beginning with a panel that has had solder mask applied, and ending
after a surface finish has been applied to the connecting surfaces of the PWB and the board has
been cleaned of any residual process chemistry. In cases where no solder mask is applied, the use
cluster would begin after the stripping of the etch resist from the outside board surfaces.

       The narrow focus on surface finishing technologies yields some benefits to the evaluation,
but it also has  some drawbacks. Benefits include the ability to collect extremely detailed
information on the relative risk, performance, cost,  and resources requirements of the baseline
technology and alternatives. This information provides a more complete assessment of the
technologies than has previously been available and would not be possible if every step in the
PWB manufacturing process was evaluated. Drawbacks from such focused evaluations include
the inability to identify all of the plant-wide benefits, costs, or pollution prevention opportunities
that could occur when implementing an alternative to the baseline HASL technology. However,
given the variability in workplace practices and operating procedures at PWB facilities, these
other benefits  and opportunities are expected to vary substantially among facilities and would be
difficult to assess in a comparative evaluation such as a CTSA. Individual PWB manufacturers
are urged to assess their overall operations for pollution prevention opportunities when
implementing an alternative technology.

1.3.3  Issues Evaluated

       The CTSA evaluated a number of issues related to the risk, competitiveness, and resource
requirements of surface finishing technologies. These include the following:

•      Risk: occupational health risks, public health risks, ecological hazards, and process safety
       concerns.                                      .
•      Competitiveness: technology performance, cost, and regulatory status.
•      Conservation: energy and natural resource use.
       Occupational and public health risk information is for chronic exposure to long-term, day-
to-day exposure and releases from a PWB facility rather than short-term, acute exposures to high
levels of hazardous chemicals as could occur with a fire, spill, or other periodic release.  Risk
information is based on exposures estimated for a model facility, rather than exposures estimated
for a specific facility. Ecological risks are also evaluated for aquatic organisms that could be
exposed to surface finishing chemicals through wastewater discharges. Process safety concerns
are summarized from material safety data sheets (MSDSs) for the technologies and process
operating conditions.
                                           1-10

-------
                                                                  1.3 CTSA METHODOLOGY
        Technology performance is based on a snapshot of the performance of the surface
 fishing 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.
TaWc 1-1. Surface Finishing Technologies Submitted bv Chemical Simnlierc
Chemical Supplier
^ jj^K. --3 -) *C""-t_^ i Tti
' - " 'ซ;',
Polyclad Technologies-Enthone
Electrochemicals, Inc.
Florida CirTech, Inc.
MacDemud, Inc.
Technic, Inc.
Surface Finishing Technology
Nickel/Gold



S
/
Mckel/Palladium/
Gold



S

OSP

S

S

Immersion
Silver
/




Immersion
Tin
^

/


                                          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 hi 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

-------
                                                                  13 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
SurfaceT5nishing 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
 tiie 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 oft 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 hi 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.

Conservations

       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 TfflS REPORT

       This CTSA is organized into two volumes: Volume I summarizes the methods and results
of the CTSA; Volume n 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. 1PC-Association Connecting Electronics Industries). Personal
communication with Jack Geibig, UT Center for Clean Products and Clean Technologies.
December.

JPC (EPC-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.

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

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

U.S. EPA (Environmental Protection Agency). 1995. Printed Wiring board Industry and Use
Cluster Profile. EPA Office of Pollution Prevention and Toxics. Washington, D.C.
EPA/744-R-95-005. September.

U.S. EPA (Environmental Protection Agency). 1997a. 'Trinted Wiring Board Case Study 6:
Pollution Prevention Beyond Regulated Materials." EPA Office of Pollution Prevention and
Toxics. Washington, D.C. EPA 744-F-97-006. May.

U.S. EPA (Environmental Protection Agency). 1997b. "Printed Wiring Board Case Study 7:
Identifying Objectives for your Environmental Management System." EPA Office of Pollution
Prevention and Toxics.  Washington, D.C. EPA 744-F-97-009. July.

U.S. EPA (Environmental Protection Agency). 1997c. "Printed Wiring Board Case Study 8:
Building an EMS, H-R Industries Experience." EPA Office of Pollution Prevention and Toxics.
Washington, D;C.  EPA744-F-97-010. December.

U.S. EPA (Environmental Protection Agency). 1998a. Printed Wiring Board Cleaner
Technologies Substitutes Assessment: Making Holes Conductive. Design for the Environment
Printed Wiring Board Project.  EPA Office of Pollution Prevention and Toxics. Washington,
D.C. EPA/744-R-98-004aandEPA/744-R-98-004b.  August.

U.S. EPA (Environmental Protection Agency). 1998b. Printed Wiring Board Pollution
Prevention and Control Technology: Analysis of Updated Survey Results.  Design for the
Environment Printed Wiring Board Project. EPA Office of Pollution Prevention and Toxics.
Washington, D.C.  EPA744-R-98-003. August.
                                        1-16

-------
                                                                          REFERENCES
U.S. EPA (Environmental Protection Agency). 1999. "Printed Wiring Board Case Study 9'
Flexible Simulation Modeling of PWB Costs." EPA Office of Pollution Prevention and Toxics
Washington, B.C. EPA 744-F-99-004.  May.

U.S. EPA (Environmental Protection Agency). 2000. Implementing Cleaner Printed Wiring
Board Technologies: Surface Finishes.  EPA Office of Pollution Prevention and toxics
Washington, D.C. EPA 744-R-00-002.  March.

Wehrspann, Carla.  1999a.  IPC-Association Connecting Electronics Industries. Personal
communication with Jeng Hon Su, University of Tennessee Center for Clean Products and Clean
Technologies.  July.                           ,

Wehrspann, Carla.  1999b.  IPC-Association Connecting Electronics Industries. Personal
communication with Jack Geibig, University of Tennessee Center for Clean Products and Clean
Technologies.  June.
                                        1-17

-------

-------
                                      Chapter 2

              Profile of the Surface Finishing Use Cluster

         This section of the Cleaner Technologies Substitutes Assessment (CTSA) describes the
  technologies that comprise the surface finishes use cluster.  A use cluster is a set of chemical
  products, technologies, or processes that can substitute for one another to perform a particular
  function.  In this case, the function is the application of a final surface finish to the printed wiring
  board (PWB). The set of technologies includes hot air solder leveling (HASL), which was
  selected as the baseline, and the alternative surface finishes, including electroless nickel/immersion
  gold (nickel/gold), electroless nickel/electroless 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
 details the process sequences.  Typical operating conditions and operating and maintenance
 procedures are described in an overview of the surface finishing manufacturing process.  Then the
. chemical processes occurring in each bath are detailed, along with additional process information
 specific to each technology.

 2.1.1  Process Sequences of Surface Finishing Technologies

        Figure 2-1 depicts the six surface finishing technologies evaluated in the CTSA. Because
 the function of applying a final surface finish can be performed using any of these technologies,
 these technologies may be substituted for each other in PWB manufacturing. The surface
 finishing technologies are all wet chemistry processes consisting of a series of chemical process
 baths, often followed by rinse steps, through which the PWB panels are passed to apply the final
 surface finish. The exception is the HASL process, which combines the typical cleaning and
 etching chemical processes with a mechanical process of dipping a board into molten solder
 followed by rinsing (described in Section 2.1.3).                                           •

       For each of the surface finishes evaluated, the process steps depicted in the figure
 represent an-integration of the various commercial products within the technology category. For
 example, chemical suppliers to the PWB industry submitted product data for two different OSP
 processes. The chemical suppliers offer additional variations to the OSP process that may have
 slightly different bath chemistries or process sequences, than the processes submitted. Figure 2-1
 lists the process steps in a typical, or generic, OSP surface finishing line. The process steps iri an
 actual line may vary.
                                           2-1

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

g
ฅ
ซ

H
o
•gg
a> >
a co

&
0

3is
•g T3 o
.2 ซC5
K'S0
PH

3s
^^
gU

J
1


1
a

i
0
U
•H

I
8

b
0

h
I
a

1
QJ
0








^^
^











1
F>
BW

1
JU
•J

1
_u
**ป

1
0
s
ft*

1
g
_ซ

1
fc,
r*








•^
f







^
^


P.
's
ฃ

_a
(^
ซ

ฃ
ฃ
3
ซJ

I
b
n

•s
"ซ
s

fc
Q
m








^
^




ป



^


C
H
a
0
OB
•T

O
"ES
u

&
O
^

o.
3
i

I
. ts
1

1
5
**



•

^
>



•


>fc
•








gซ
Q
V)


^

Q>
3
ซ

OG
OE
M
>u
a
S
*n

V
V
^a
-ป-
&
s

1
a
M
ซi





ฃป
Q
i


1

1
g

1
g

•w
^








t1
Q
VD-







a
a
_ซ
s
2
j
w











b
•fj
1
PH
>=

sion Gold
S
ui

>
!
>
• ^
'•^









^

2
S a S
-2-3 fl
2-3 -ฐ
^ "S^ ^ S
pS"S ง
WPn ง
t> tฃ




(U
V3
i
s-
,.ซs..


                                                               ง
                                                              1
                                                              _o



                                                              .!"
                                                              *ซ

                                                              8
                                                               i
                                                              •s
                                                              •a
J
"So
ฃ

o
fl
ซe
*p*
a
IS

02

1-
                                                                        2
                                                              I
                                                               o.
                                                               
-------
 	2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES

 2.1.2   Overview of the Surface Finishing Manufacturing Process

        Surface finishing technologies typically consist of a series of sequential chemical
 processing tanks (baths) separated by water rinse stages. The process can either be operated in a
 vertical, non-conveyorized submersive-type mode, or in a horizontal, conveyorized mode. In
 either mode, selected baths may be operated at elevated temperatures to facilitate required
 chemical reactions or baths may be agitated to improve contact between the panels and the bath
 chemistry. Agitation methods employed by PWB manufacturers include panel agitation,
 ultrasonic vibration, air sparging, and fluid circulation pumps.

        Most process baths are followed by a water rinse tank to remove drag-out, the clinging
 film of process solution covering the rack and boards when they are removed from a tank.
 Rinsing is necessary to provide a clean panel surface for further chemical activity and to prevent
 chemical drag-out, which may contaminate subsequent process baths.  PWB manufacturers
 employ a variety of rinse water minimization methods to reduce rinse water usage and consequent
 wastewater generation rates.  The quantities of wastewater generated from surface finishing lines
 are discussed in  Section 5.1, Resource Conservation, while the composition of the wastewater is
 modeled and presented in Section 3.2, Exposure Assessment. Rinse water reduction techniques
 are discussed in  Section 6.1, Pollution Prevention.

        After the application, imaging, and development of the solder mask, panels are loaded into
 racks (vertical, non-conveyorized mode) or onto a conveyor (horizontal, conveyorized mode) for
 processing by the surface finishing line.  Racks may be manually moved from tank to tank, moved
 by a manually or automatically controlled hoist, or moved by other means. Process tanks are
 usually open to the atmosphere. To reduce volatilization of chemicals from the bath or worker
 exposure to volatilized chemicals, process baths may be equipped with a local ventilation system,
 such as a push-pull system, or covered during extended periods of latency. Horizontal,
 conveyorized systems are typically fully enclosed, with air emissions vented to a control
 technology or to the atmosphere outside the plant.

       The HASL process differs from the other alternatives in that it does not rely on a chemical
 process to apply the final surface finish. Instead, the process combines the chemical processes of
 board preparation and cleaning with a mechanical step to apply the finish.

       Regardless of the mode of operation or type of alternative, process chemical baths are
 periodically replenished to either replace solution lost through drag-out or volatilization, or to
 return the concentration of constituents in the bath to within acceptable limits. During the course
 of normal operations, bath chemistry can be altered by chemical reactions occurring within the
 bath or by contamination from drag-in. Bath solution may be discarded and replaced with new
 solution as required, with the frequency of replacement depending on analytical sampling results,
the number of panel surface square feet (ssf) processed, or the amount of time elapsed since the'
last change-out.  Process line operators also may clean the tank or conveyorized equipment during
bath change-out operations.
                                          2-3

-------
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.

2.1.3  Chemistry and Process Descriptions of Surface Finishing Technologies

       This section describes hi 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 hi 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 tune spent in the plating
       bath, but are typically hi 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-frniting, 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 OTAST,)

        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 solderabffity 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 DESCKIPTION OF SURFACE FINISHING TECHNOLOGIES



*• Cleaner
V
2- Microetch

1

1
y
Water Rinse x 2
y
3- Dry
1

1
y
4- Flux
y
5- Solder
y
6- Pressure Rinse
y
7- DI Rinse

1

1

1

1

                        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, solderabflity, 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 sk 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. NiekeVgold 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
i
Water Rinse xl
I
2- Microetch
1
Water Rinse xl
i
3. Catalyst
V
Water Rinse xl
i
4- Acid Dip
y
Water Rinse xl
. y
1

I

I

I

I

I

I

I

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


I

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 AKP PROCESS DESCRIPTION OF SURFACE FESISHIIVG 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 sulfiiric 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.

 SteP5:        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. Mckel 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.
                                                                                 L
       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/palkdium/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/lvficroetch/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 AMP PROCESS DESCRIPTION OF SURFACE FEVTSHEVG TECHNOLOGIES


Cleaner
I
1
Water Rinse x 2 1
1
2- ' Microetch 1
-|
Water Rinse x 2 |
y
3 Catalyst I
^ ' . -
Water Rinse x 2

1
4- Acid Dip I
V
Water Rinse x 2

~]
1
5- Electroless Nickel 1
1
Water Rinse x 2 1
1 '
6 Preinitiator |
y
7 Electroless Palladium

1
V
Water Rinse x 2
I
* '
8* Immersion Gold 1
y
Water Rinse x 2 1


       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?:       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 Ihe nickel surfaces of the PWB. Palladium layer thicknesses are
             typically 0.3 to 0.8 microns (12 to 32 microinches).

Step 8:       Immersion Gold: A very thin, protective layer of pure gold is deposited onto the
             surface of the palladium in the immersion gold plating bath. A chemical
             displacement reaction occurs, depositing the thin layer of gold onto the metal
             surface while displacing palladium ions into the solution. Because the reaction is
             driven by the potential difference of the two metals, the reaction ceases when all of
             the surface palladium has been replaced with gold. Gold layer thickness is typically
             0.2 microns (8 microinches).

Organic Solderability Preservative (OSP)

       The OSP process selectively applies a flat, anti-oxidation film onto the exposed copper
surfaces of the PWB to preserve the solderability of the copper.  This coating reacts with the
copper in an acid and water mixture to form the nearly invisible protective organic coating.  OSP
processes can be based on benzunidazole 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 hi a horizontal, conveyorized mode but can be modified
to run hi 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 hi Figure 2-5,
followed by a brief description of each of the process steps.
                                           2-12

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



1- Cleaner
D
. • ' ' V
Water Rinse xl
V
2- Microetch
V
Water Rinse xl


^

D
t
3 Air Knife
J
y
4 OSP I
V
5 Air Knife
1
y
Water Rinse xl |
V
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 coating 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 organo'metallic layer that preserves the
              solderability of the copper surface for future assembly (Mouton, 1997).

Step 5:        AirKiiife: An air knife removes excess OSP from the panel and promotes even
              coating across the entire PWB surface. The ah- 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 hi 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 H
V
Water Rinse xl

D
t
2- Microetch
D
V
Water Rinse xl 1
t
3- Predip

~t
t
Immersion Silver
1
y
Water Rinse xl
D
V
-
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 TECHNOLOGIES

Immersion Tin

       The immersion tin process utilizes a thiorea-based reducing agent to create an
electrochemical potential between the surface and stannous ions in solution, causing the reduction
of a layer of tin onto the copper surfaces of the PWB. An organo-metallic compound, which is
co-deposited along with the tin, acts to retard the formation of a tin-copper intermetallic layer,
preserving the solderability of the finish.  The organo-metallic compound also inhibits the
formation of tin whiskers (i.e., dendritic growth). The process is typically operated in a
conveyorized fashion, but can be modified to run in a vertical, non-conveyorized mode.
Immersion tin surfaces are compatible with SMT, flip chip, EGA technologies, and typical
through hole components. The immersion tin surface cannot be wirebonded.  Tin surfaces are
compatible with all solder masks, have a reported shelf life of one year and can typically withstand
a minimum of five thermal excursions during assembly.

       A flow diagram of the process steps in a typical immersion tin process is presented in
Figure 2-7. A brief description of each of the process steps is given.


1- Cleaner 1
'V
Water Rinse x2 1
Y
2. Microctch 1
Y
Water Rinse x 2 1
Y
3 Predip |
V
Water Rinse x 1 1
Y
4. Immersion Tin 1
Y
Water Rinse x 2 1
Y
5. Dry |

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

-------
 	2.1 CHEMISTRY AMP PROCESS DESCRIPTION OF SURFACE FENISHIIV6 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:        Mcroetch:  A micrdetch solution, typically consisting of dilute hydrochloric or
               sulfuric acid, removes any remaining contaminants from the copper surface. The
               etching also chemically roughens the copper surface to promote good tin-to-
               copper adhesion.

 Step 3:        Predip: Etched panels are processed through a predip solution that is chemically
               similar to that of the tin bath, thus protecting the plating bath from harmful drag-in
               chemicals.

 Step 4:        Immersion Tin: A tin plating bath deposits a thin layer of tin onto the exposed
               copper circuitry through a chemical displacement reaction that deposits stannous
               ions while displacing copper ions into the plating solution. The bath is considered
               self-limiting, because plating continues only until all the copper surfaces have been
               coated with a tin deposit. The presence of a complexing agent, thiourea, prevents
               the copper from interfering with the plating process.  The complexed copper is
               removed as a precipitate from solution by decantation.

 Step 5:        Dry:  A drying stage removes any residual moisture from the board to prevent
               staining and to ensure high metal quality in the through holes.

 2.1.4  Chemical Characterization of Surface Finishing Technologies

       This section describes the sources of bath chemistry information, methods used for
 summarizing that information, and the use of bath chemistry data. Publicly-available information,
 along with proprietary chemical information obtained  from the chemical suppliers, was used to
 assess exposure, risk, and cost for the processes. This section does not identify any proprietary
 ingredients. Generic names have been submitted for the names of proprietary, confidential
 chemicals to mask their identity.

 Use of Chemical Product and Formulation Data

       Assessment of releases, potential exposure, and characterizing risk for the surface finishing
technologies requires chemical-specific data, including concentrations for each chemical in the
various process baths. Although some bath chemistry data were collected in the PWB Workplace
Practices Questionnaire, the decision was made not to use  this data because of inconsistencies in
the responses to questions pertaining to bath chemistry.  Instead, the suppliers participating in the
Performance Demonstration each submitted complete  chemical formulations along with other
publicly-available information on their respective product lines.  This information includes:
                                           2-17

-------
2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES
•      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 Ouster Chemicals and Associated Surface Finishing Processes
Chemical
^
Acetic acid
Aliphatic acid A
Aliphatic acid B
Aliphatic acid D
Aliphatic acid E
Aliphatic dicarboxylic acid A
Aliphatic dicarboxylic acid C
Alkylalkyne diol
Alkylamino acid A
Alkylamino acid B
Alkylaryl imidazole
Alkylaryl sulfonate
Alkyldiol
Alkylimine dialkanol
Alkylphenol ethoxylate
Alkylphenol polyethoxyethanol
Alkylpolyol
Amino acid salt
Amino carboxylic acid
Ammonium chloride
Ammonia compound A
HASL







•



•
•

•
•





Mckel/
.Gold

•
•

•
•
•


•


•


•



•

Nickel/,
Palladium/
, Gold


•

•
•
•


•


•



•
•
•

•
OSP^
•









•










Immersion
Silver








•












Immersion
Tin



•



•

•

•

•
•






                                          2-18

-------
2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES
Chemical -
' ,x ฐ *
~ >• x" ~ ' ^
ป "^ *~ "S*
Ammonia compound B
Ammonium hydroxide
Aromatic imidizole product a
Arylphenol
Bismuth compound
Citric acid
Copper ion
Copper salt C
Copper sulfate pentahydrate
Cyclic amide
Ethoxylated alkylpheriol
Ethylenediamine
Ethylene glycol
Ethylene glycol monobutyl ether
Fatty amine
Fluqboric acid
Gum
Hydrochloric acid
Hydrogen peroxide
Hydroxy carboxylic acid
Hydroxyaryl acid
Hydroxyaryl sulfonate
Inorganic metallic salt A
Inorganic metallic salt B
Inorganic metallic salt C
Lead
Maleic acid
Malic acid
Methane sulfonic acid
Nickel sulfate
Nitrogen acid
Nonionic surfactant a
Palladium chloride
Palladium salt
Phosphoric acid
Potassium compound
Potassium gold cyanide
HASL
*V -i



•

•


•

•

•
•

• .
•
•
•

•
•



•








•


Nickel/
Gold
•
•



•


•

•






•
•

•

•
•
•


•

•


•

•
•
•
Nickel/
Palladium/
Gold
•
•



•


•

•
•





•
•

•


•


•
•

•



•
•
•
•
OSP


•
•


•
•
•

•

•



•
•
•

•
•












•


Immersion
Silver














•



•











•
•


•


Immersion
Tin




•
•



•
•


•

•

•

•








•





•


                               2-19

-------
2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FINISHING TECHNOLOGIES
Chemical
•i ,, >4 > '!•*
>•. v"* X k jl> ""-"
- ,. \*
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
Surfactant0
Thiourea
Tin
Tin chloride
Transition metal salt a
Unspecified tartrate
Urea
Urea compound B
Urea Compound C
Vinyl polymer
HASL
•




•
•






•


•







Nickel/
Gold
ฃ
•





•
•
•

•

•
•




•


•


Nickel/
Palladium/
Gold

•




•

•

•

•
•
•



•


•


OSP






•






•










Immersion
Silver "



•


•






•










Immersion
Tin
3
•

•

•
•



•

•

•

•

•

•
•

•
•
* Dropped due to insufficient identification.
 Determining Chemical Formulations

       Determining the chemical formulations for each process step is critical for evaluating each
 surface finishing technology. Each surface finishing product line submitted for evaluation was
 divided into basic bath steps common to all the processes within that surface finishing category
 (e.g., both OSP product lines submitted were divided into cleaner, microetch, and OSP baths).
 The basic bath steps were combined to form a process flow diagram specific to each surface
 finishing technology, as shown in Figure 2-1. The recommended formula for creating a new bath,
 along with the individual formulations for each chemical product, were combined to determine the
 individual chemical concentrations in the final bath.  The individual chemical concentrations in the
 baths were calculated by:
                                          2-20

-------
         2.1 CHEMISTRY AND PROCESS DESCRIPTION OF SURFACE FBNISH1NC TECHNOLOGIES

                           Q, =  (CCHEM) (CpORM) (D) (1000 cm3/L)
 where,
 Q,    •-
 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 bath concentration (Q,) 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 (  IQL\( 1.27g\t 1000cm3]    ..g-
                     lOQg(lOQL)(  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 in each
 bath, by technology, as well as the concentration range for each chemical.  However, some of the
 alternatives (e.g., OSP, nickel/gold, and immersion tin) have more than one chemical supplier
 using different, bath chemistries. It was decided to include all of the identified chemicals in the
 formulations rather than selecting a typical or "generic" subset of chemicals.

      . Estimated concentration ranges (low, high, and average) were determined based on the .
 formulation data and are presented in Appendix B.  Concentrations are for each bath in each
 surface finishing process alternative.

 Data Limitations

       Limitations and uncertainties in the chemical characterization data arise primarily from
 side reactions in the baths.  Side reactions in the baths may result in changing concentrations over
time and/or formation of additional chemicals hi the baths. This information is not reflected hi
product formulation data, MSDSs  or Product Data Sheets, but would affect bath concentrations
                                          2-21

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

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 with Jack Geibig,
University of Tennessee Center for Clean Products and Clean Technologies. July.
                                        2-25

-------

-------
                                      Chapter 3
                      Risk Screening and Comparison
        This chapter of the Cleaner Technologies Substitutes Assessment (CTSA) addresses the
 health and environmental hazards, exposures, and risks that may result from using a surface
 finishing technology. The information presented here focuses entirely on the surface finishing
 technologies.  It does not, nor is it intended to, represent the full range of hazards or risks that
 could be associated with printed wiring board (PWB) manufacturing.  This risk evaluation is a
 screening-level assessment of multiple chemicals belonging to the surface finishing use cluster, and
 is presented as a screening level rather than a comprehensive risk characterization, both because
 of the predefined scope of the assessment and because of exposure and hazard data limitations.
 The intended audience of this risk screening and comparison is the PWB industry and others with
 a stake in the practices of this industry.

        Section S.ridentifies possible sources of environmental releases from surface finishing
 and, in  some cases, discusses the nature and quantity of those releases.  Section 3.2 assesses
 occupational and general population (i.e., the public living near a PWB facility; fishing streams
 that receive wastewater from PWB facilities) exposures to surface finishing chemicals. This
 section quantitatively estimates inhalation and dermal exposure to workers and inhalation
 exposure to the public living near a PWB facility.  Section 3.3 presents human health hazard and
 aquatic toxicity data for surface finishing chemicals.  Section 3.4 characterizes the risks and
 concerns associated with the exposures estimated in Section 3.2. In all of these sections, the
 methodologies or models used to estimate releases, exposures, or risks are described along with
 the associated assumptions and uncertainties. Finally, Section 3.5 summarizes chemical safety
 hazards from material safety data sheets (MSDSs) for surface finishing chemical products and
 discusses process safety issues.


 3.1    SOURCE RELEASE ASSESSMENT

       The Source Release Assessment uses data from the PWB Workplace Practices
 Questionnaire, together with other data sources, to identify sources and amounts of environmental
releases. Both on-site releases (e.g., evaporative or fugitive emissions from the process) and off-
site transfers (e.g., off-site recycling) are identified and, for those where sufficient data exist from
the questionnaire, numerical results are presented.  The objectives of the Source Release
Assessment are to:

•     identify potential sources of releases;
•     characterize the source conditions surrounding the releases, such as a heated bath or the
      presence of local ventilation; and
•     characterize, where possible, the nature and quantity of releases under the source
      conditions.
                                          3-1

-------
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 in the Source. Release Assessment include:

•      industry data collection forms, such as the PWB Workplace Practices Questionnaire and
       Performance Demonstration Observer Data Sheets (Appendix A, Data Collection Sheets);
•      supplier-provided data, including bath chemistry data and supplier Product Data Sheets
       describing how to mix and maintain baths (Appendix B, Publicly-Available Bath Chemistry
       Data);
•      engineering estimates; and
•      DfE PWB Proj ect publication, Printed Wiring Board Pollution Prevention and Control
       Technologies:  Analysis of Updated Survey Results (U.S. EPA, 1998a),

       Bath chemistry data were collected in the PWB Workplace Practices Questionnaire, but
these data were not used due to inconsistencies in the responses to questions pertaining to bath
chemistry. Instead, surface finishing chemical suppliers participating in the Performance
Demonstration submitted confidential chemical formulation data along with publicly-available
Product Data Sheets on their respective product lines. Bath concentration ranges were
determined based on this information using the method discussed in Section 2.1.4, Chemical
Characterization of Surface Finishing Technologies.  A general description of the PWB
Workplace Practices Questionnaire, including its distribution and overall general results, is
presented in Section 1.3.4, Primary Data Sources.
                                           3-2

-------
                                                         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
 faculties.  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
	 System Boundary
TtlinrrnSrinl T>ซ+1-> Tlnimrlrmr
>ป
|
3
3
q
-4
3
9 or
••4
^
A
-r
C
•zsl1*
oซ=ง

.*—
': , ' '•
^^
W_ ^""

CDJ
OT3 to

A,
Evaporation and
aerosols from baths


D)
•l> > r
S CO •<
•ง ซ
w'ro
EE
^^
• i — - —

CD
jco E co
ill
OQ , ,
•ป.
CO*
CO •••

Hazardous solid wast<
- filters
- precipitates
- container residues
^
•H
Non-hazardous solid waste
- filters
- precipitates
- container residues
A ^
D)
|.l
Surface
Process

J--
CO C
0 0
Et5 oE
co co

U A >L J
— -— .

'-'-,.'
i


LJ
04 CO J'
co
*ฐs
•ง1"
73^1
CO O
•o V
CD .^
x: co
UJ E
Water
- rinse water
- equipment cleaning
- bath make-up
^•M
M
. V'
ฐt
oi
81
•**
p

>.

•••• .
el
ia
^
b
CO (1)
•a to
(D CO
" S 3
..•••••
CO
CO
_o
& OJ
(D c -i f
r- P '
(TT CQ CO
ca •?•*?

o .E •
Spent bath s
(include was
equipment c
chemicals)
eo
CO
1
m
>.
Q)
E
"co
H-Q>
> if
s If
a= SS
o a
>^


Cleaning chemicals
- chemical flush
- equipment cleaning


Vastewater
primarily from
inse tanks)
•^ >**' l™
O CO
CS
>. o b
* '-'<-'

5

Figure 3-1. Schematic of Overall Material Balance for Surface Finishing Technologies
                                      3-4

-------
                                                           3.1 SOURCE RELEASE ASSESSMENT
                                                 w.
                    Spent bath solutions (include wasted!
                    equipment cleaning chemicals)    I
          Wastewater
      (primarily from rinse tank)
           Bath chemicals
           - sampling
           - bail-out
                                                                           r System Boundary
Storage Tank
                       Wastewater
                       Treatment
                        System
                                          Treatment
                                       Sludge to recycle
                                         or disposal
                                     Discharge to POTW or
                                       Receiving Stream
                 Figure 3-2. Wastewater Treatment Process Flow Diagram
Inputs
       Possible inputs to a surface finishing process line include process chemicals and materials,
etched and solder mask-coated PWBs that have been processed through previous PWB
manufacturing process steps, water,  and cleaning chemicals.


       The total inputs for the process are described by the equation:
where,

Ii     =     bath chemicals

I2    • =     etched and solder mask-coated PWBs
I3     =     water

I4     -     cleaning chemicals


These terms are discussed below.


li     Bath chemicals. This includes chemical formulations used for initial bath make-up, bath
       bailout and additions, and bath replacement. Bath formulations and the chemical

       constituents of those formulations were characterized based on Product Data Sheets and

       bath formulation data provided by the chemical suppliers.  A detailed description of the
                                            3-5

-------
3.1 SOURCE RELEASE ASSESSMENT
       calculation of bath chemical concentrations is presented in Section 2.1.4, Chemical
       Characterization of Surface Finishing Technologies.  Calculated chemical bath
       concentrations are reported in Appendix B. PWB manufacturers were asked to report the
       quantity of surface finishing chemicals they use annually hi 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 hi 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 hi
       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 hi gallons by the annual production hi 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 hi 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 hi PWB throughput between reporting facilities and the relatively few
       number of responses within some technology categories.
                                          3-6

-------
                                                          3.1 SOURCE RELEASE ASSESSMENT
       Table 3-1. Water Usage of Surface Finishing Technologies From Questionnaire
Process Type
No. of Responses
Water Usage (I3)
(thousand gal/year)a
~ Water Usage (I3)
(gal/ssf)
HASL
Non-conveyorized
Conveyorized
6
17
0.3 - 750 (254)
910-3,740(1,250)
0.970
4.89
Nickel/Gold
Non-conVeyorized
8
17-1,620(538)
101
Nickel/Palladium/Gold
Non-conveyorized
2
216-1,710(961)
164
OSP
Non-conveyorized
Conveyorized
5
5
42-150(89.1)
8-1,580(440)
1.93
14.3
Immersion Silver
Conveyorized
-2
698 - 1,120 (907)
36.8
Immersion Tin
Non-conveyorized
Conveyorized
4
2
3.3 - 385 (209)
11.5-199(105)
11.0
0.333
3 Average values from the PWB Workplace Practices Questionnaire data are shown hi parentheses. Refer to Section
1.3.4 for a detailed discussion of questionnaire responses.
I4     Cleaning chemicals. This includes chemicals used for conveyor equipment cleaning, tank
       cleaning, chemical flushing, rack cleaning, and other cleaning pertaining to the surface
       finishing process line. Data were collected by the PWB Workplace Practices
       Questionnaire regarding the use of chemicals to clean conveyors and tanks (questions 2.8,
       3.8, 2.13, 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
Process Type ,
HASL
Nickel/Gold
Nickel/PaUadium/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 JiTsing
Chemical Flushing3
1(27)
2(27)
5(28)
1 (22)
1 (8)
8(8)
1(8)
1(8)
5(9)
1(1)
1 (1)
2(2)
2(2)
1 (2)
1(1)
4(9)
1(2)
2(2)
1(4)
  Total number of questionnake 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 surf ace 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:

                                      Atotal  = Aj + A2
                                            3-8

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

AS     =

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 Ihe 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 fum<
from the tank.
                                                                                      tes
                                           3-9

-------
3.1 SOURCE RELEASE ASSESSMENT
        Table 3-3. Average Bath Dimensions and Temperatures for All Processes!
Bath
No- ฐf ,.
Responses
Length

-------
                                                           3.1 SOURCE RELEASE ASSESSMENT
Bath ". ,
* i ,. - -,
OSP
No. of
Responses;
4
Length
Cm.)
27
Width
0ซ.)
24
Surface Area b
(sq. in.)
580
-, Volume
(gal->
86
Temp."
(*F)
124
OSP, Conveyorized
Cleaner
Microetch
OSP
3
5
5
36
35
72
30
34
34
1100
1300
2600
56
63
125
113
99
108
Immersion Silver, Conveyorized
Cleaner
Microetch
Predip
Immersion Silver
Dry
2
2
2
2
1
34
42
47
143
- ' •
31
31
31
31
-
1000
1300
1600
4400
-
65
80
60
142
-
81
73
86
113
149
Immersion Tin, Non-conveyorized
Cleaner
Microetch
Predip
Immersion Tin
2
2
1
2
. 27
27
30
27
18
18
24
18
500
500
720
500
49
49
60
47
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
-
104
103
-
150

105
95
101
133
165
-a Based on PWB Workplace Practices Questionnaire data.
b All of the surface areas present in the table are average values of individual
multiplying the average length by the average width.
- No responses were given to this question in the questionnaire.
bath areas; they are not obtained by
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:
                                               W2+W3
where,
Wx
W2
W3
wastewater
spent bath solution
bath sampling and bail-out
These terms are discussed below.

Wj    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.
ofBaths
113
55
14
28
8
17
Precipitation
Prctreatment a
29
35
8
14
3
3
PH
Neutralization a
24
25
3
15
3
6
Disposed [Drummed a
to Sewer a j - *
11
0
0
0
1
0
2
7
4
2
5
Recycled
On-Sftea
6
2
1
1
0
3
Recycle
Off-Site *
• 29
4
1
0
0
,0
Others
f -
8
5
0
0
0
0
technology category.
                                                                                    i using a
                                           3-13

-------
3.1 SOURCE REUEASE 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.

E!     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. Faculties 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 dependence 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:
                                                     S4
where,
Si
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 RCRA
                                          3-14.

-------
                                                         3.1 SOURCE RELEASE ASSESSMENT
        waste code D008 is for lead).3  Container residue is estimated by EPA to be up to four
        percent of the chemicals use volume (Froiman, 1996).  An industry reviewer indicated this
        estimate would only occur with very poor housekeeping practices and is not representative
        of the PWB industry. RCRA waste codes which are applicable to the surface finishing
        technologies are discussed in Section 4.3, Regulatory Status. Hazardous solid waste is
        typically sent off-site to a hazardous waste landfill for disposal or is incinerated.

 S2     Non-hazardous solid-waste. Non-hazardous solid wastes could include any spent bath
        filters, packaging or chemical container residues, and other solid waste from the process
        line that does not contain any RCRA-defined hazardous materials listed in CFR Section
        261. These wastes may be recycled or sent to off-site disposal in a landfill.

 S3     Drummed solid or liquid waste. This includes other liquid or solid wastes that are
        drummed for off-site recycling or disposal. This includes spent bath chemicals which
        cannot be treated on-site because they are considered hazardous or require treatment
        beyond what can be provided by the facility.  Hazardous chemical wastes are sent to a
        hazardous waste treatment facility. Table 3-5 is a summary of responses indicating the
        presence of a RCRA listed waste and the type of container in which it was stored.

        Other chemical wastes are drummed and sent out for recycling to reclaim the metal
        content from the solution (e.g.,  gold, silver, nickel, etc.). The number of responses which
        indicated that a bath was drummed for disposal was shown in Table 3-4.

 S4     Sludge from on~site wastewater treatment.  Facilities were asked to report the amount of
        sludge generated during on-site wastewater treatment that could be attributed to surface
        finishing line effluents (question 1.3). Many PWB manufacturers have indicated that the
        amount of sludge resulting from the surface finishing process cannot be reliably estimated
        since effluents from various PWB manufacturing process steps are combined prior to
        wastewater treatment. Other factors that also influence the amount of sludge generated
        during wastewater treatment include the size of the facilities, the surface finishing
        technology used, the treatment method used, facility operating procedures, the efficiency
        with which bath chemicals and rinse water are used, and so on.  Thus, the actual and
        comparative amount of sludge generated due to the choice of surface finishing technology
        could not be determined, nor were data available to characterize the concentrations of
        metals contributed by the surface finishing line.

        However, many respondents did report the annual amount of sludge generated from their
        on-site waste treatment facility.  The average sludge generated annually by the
        respondents to the PWB Workplace Practices Questionnaire is 214,900 pounds.  The
        average water content of the sludge, which is typically pressed prior to disposal, ranges
       from 60 to  70 percent (Sharp, 1999).
     It is important to note that solder dross and solder pot dumps are excluded from the RCRA definition of solid
waste when they are recycled. Therefore, when they are recycled they are not considered a hazardous solid waste.

                                          3-15

-------
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.oC .
RCRA, Wastes
1
8
7
7
2
1
2
1
3
0
3
2
1
0
0
0
Open Head*
' Drum
0
0
0
8
1
0
0
0
0
0
0
0
0
0
0
0
Close Head
,Drum
2
9
12
6
1
2
3
2
6
3
4
1
1
1
1
1
Others
0
4
0
5
3
0
0
0
0
0
0
2
2
1
0
0
Transformations

       Transformations within the surface finishing system boundary could include:

RI     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 Il5 1^ I3, and I4, and the outputs P^A, A^ Wto W^ W3, S1; 83, and S3.

 Since the inputs must equal the outputs, the material balance for Figure 3-1 is:
or:
The material balance for Figure 3-2 (wastewater treatment) includes the inputs Wt, W2, and W3
and the outputs Ej and S4

Thus, the material balance equation for Figure 3-2, wastewater treatment, is:

                                 W!+.W2 + W3 =  EX + S4

or:                                                   •     •
for.
       These equations are presented to indicate that all the material flows have been accounted
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 hi 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
v-
•^ Microetch
t
3. Water Rinses 2
I

I

I
i
4 Dry
t
5- Plus
i
6- Preheat
i
7- HASL
t
8 Air Knife
I

I

I

I

~J
i
9- Pressure Rinse
V
10- Water Rinse xl

I

1

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

-------
                                                       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 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 (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 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 conveyorized HASL processes. Respondents for both the non-conveyorized,
vertical process and the mixed HASL processes reported that precipitation pretreatment, pH
neutralization, and off-site recycling are common treatment methods.

       Evaporation From Baths (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 (A2). Air knife and oven drying occur after the
microetch and HASL baths.  Any solution adhering to the PWBs would be either blown off the
boards and returned to the sump, or volatilized in the oven.  Air emissions from air knife or oven
drying were not quantified.

       Chemicals Incorporated Onto PWBs (PJ. A coating of tin/lead solder is applied to the
surface of PWB panels in the HASL process.  The amount of solder added to the panels depends
on the exposed surface area of the PWB panels being processed. The amount of solder
incorporated'onto a PWB was calculated at 0.0369 oz/ssf. Solder consumption is discussed
further in Section 5.1, Resource Conservation.

       Drummed Solid or Liquid Waste (S3). Questionnaire respondents indicated that
approximately 25 percent of HASL baths contain hazardous waste constituents as defined by
RCRA. These wastes were associated by respondents with the microetch, flux, and solder baths.
RCRA wastes are discussed in further detail in Section 4.3, Regulatory Status. In response to a
separate question regarding spent bath treatment (see Table 3-4), 11 out of 113 HASL baths were
reported by respondents to be drummed and sent off-site for recycling or disposal.
                                          3-20

-------
                                                       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 CTSA. The process baths shown in the figure represent an amalgamation of the
various products offered within the nickel/gold technology category. The number and location of
rinse steps displayed in the figure are based on PWB Workplace Practices Questionnaire
responses.  Thus, Figure 3-4 describes the types and sequence of baths in a generic nickel/gold
line, but the types and sequence of process baths used by any particular facility could vary.  A
detailed description of the nickel/gold process is presented in Section 2.1.3, Chemistry and
Process Descriptions of Surface Finishing Technologies.


1- Cleaner
y
2- Water Rinse xl
~)

~\
y
3- Microetch |
y
4. Water Rinses 1

n
y
5- Catalyst
u
y
6- Water Rinse si |
y
7- Acid Dip
1
y •' •
8- Water Rinse si
J
^
9. Electroless Nickel
1
y
10- , Water Rinses 2
J
. y

11. Immersion Gold I
y

12- Water Rinse x 2
1

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

-------
3.1 SOURCE RELEASE ASSESSMENT
       Water Usage (Ej) 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). All eight respondents report using the non-
conveyorized nickel/gold process.  In summary:

•      Reported water usage for the facilities using the non-conveyorized nickel/gold process
       average 540 thousand gallons per year, or 100 gallons per ssf of PWB produced.

       Chemical constituents and concentrations in wastewater could not be adequately
characterized from questionnaire data. In the absence of quality data from industry, a model was
developed to estimate the mass loading of constituents within the wastewater, resulting from
drag-out, during the production of 260,000 ssf of PWB by the surface finishing process. The
mass of chemical transferred per day to the wastewater, as well as other model results, are
presented in Appendix E.

       Bath Chemicals Used (I,). Bath concentrations of individual chemical constituents are
presented in Appendix B. The volume of chemicals consumed per year was determined by
modeling the time it would take the generic nickel/gold process described in Figure 3-4 to
produce a specific PWB throughput. A detailed description of the process modeling is presented
in Section 4.2, Cost Analysis.  The number of bath replacements (calculated from the modeled
time) was then multiplied by the volume of the bath to determine the volume of a bath chemical
consumed per year. Nickel/gold process chemical consumption is presented in Appendix G.

       Cleaning Chemicals (T4).  Twelve out of 47 reported nickel/gold baths require chemicals
to clean the tanks, however, data concerning the type of cleaning chemical(s) were not collected  •
by the questionnaire. Seven of the tanks that were reported to require chemical flushing belong to
electroless nickel baths.  The remaining tanks requiring chemical flushing belong to baths  which
are not part of the generic process sequence described in Figure 3-4. Water is most frequently
used to clean tanks prior to new bath make-up.

       Spent Bath Solutions (W2). The quantity of spent bath solution could not be determined
directly from the questionnaire data. However, the volume of spent bath chemistry was calculated.
by determining the number of bath changes required per year and multiplying by the average
volume of the process tank (see Section 4.2, Cost Analysis).  The concentrations of chemical
constituents within the spent bath solutions were assumed to be the same as make-up bath
concentrations.

       Spent bath treatment and disposal methods were presented in Table 3-4.  Respondents for
the non-conveyorized, vertical process reported that pH neutralization and precipitation
pretreatment are common treatment methods. OfF-site recycling was also reported as a treatment
option.

       Evaporation From Baths (AJ. Air releases are modeled in Section 3.2, Exposure
Assessment.  A summary of data collected from the questionnaire is presented below:
                                          3-22

-------
                                                        3.1 SOURCE RELEASE ASSESSMENT
 •      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
 Status. In response to a separate question regarding spent bath treatment (see Table 3-4), two
 out of 55 nickel/gold baths (3.6 percent) were reported by respondents to be drummed and sent
 off-site for recycling.  Section 5.1, Resource Conservation, presents methods commonly used to
 recover gold on-site.

 Nickel/Palladium/Gold Process

       Figure 3-5 depicts the generic nickel/palladium/gold process steps and typical bath
 sequence evaluated  in the CTSA. The number and location of rinse steps displayed in the figure
 are based on PWB Workplace Practices Questionnaire responses.  Thus, Figure 3-5 describes the
types and sequence  of baths in a generic nickel/palladium/gold line, but the types and sequence of
process baths used by any particular facility could vary.  A detailed description of the
nickel/palladium/gold process is presented in Section 2.1.3, Chemistry and Process Descriptions
of Surface Finishing Technologies.                                                ;
                                          3-23

-------
3.1 SOURCE RELEASE ASSESSMENT

*• Cleaner
\r

z- Water Rinse x 2
)
r
3- Microetch
\
r
4" Water Rinse x 2
>
r
5- Catalyst
>
f •
6- Water Rinse x 2
>
'
7- Acid Dip
•
\'
8- Water Rinses 2

I

I

I

I

I

i

i

i
V
9- Electroless Nickel

V
I0- Water Rinse x 2
1
f
31- Preinitiator
>
f
12- Electroless Palladium
' >
r
13- Water Rinse x 2
>
r
14- Immersion Gold
' ' >
f
1S- Water Rinse x 2

i

i

i

i

i

i

i

   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,OQO 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 (A,).  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).
require the use of a drying oven or air knife.
The nickel/palladium/gold process does not
       Chemicals Incorporated Onto PWBs (Pj). 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 hi Section 5.1, Resource Conservation.

       Drummed Solid or Liquid Waste (S3). Questionnaire respondents indicated that none
of the nickel/palladium/gold baths contain hazardous waste constituents as defined by RCRA. A
detailed discussion of RCRA wastes can be found in Section 4.3, Regulatory Status. In response
to a separate question regarding spent bath treatment (see Table 3-4), seven out of 14
nickel/palladium/gold baths (50 percent) were reported by respondents to be drummed and sent -
off-site for recycling or disposal.  Section 5.1, Resource Conservation, presents methods
commonly used to recover gold on-site.

Organic Soiderability Preservative

       Figure 3-6 depicts the generic OSP process steps and typical bath sequence evaluated hi
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 hi the figure are based on PWB Workplace Practices Questionnaire responses. Thus,
Figure 3-6 describes the types and sequence of baths hi 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 hi Section 2.1.3., Chemistry and Process Descriptions of Surface
Finishing Technologies.
                                          3-26

-------
                                                       3.1 SOURCE RELEASE ASSESSMENT


!- Cleaner
y
2. Water Rinses!
1

1
y
3- Microetch
y
1

4. Water Rinses! |
if
5- Air Knife |
y
6. . OSP

1
y
7- Air Knife 1
y
8. Water Rinse si |
. y
9- Dry |


            Figure 3-6.  Generic OSP 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 OSP process,
five facilities use the conveyorized OSP process while five other facilities use the non-
conveyprized OSP process. In summary:

•      Reported water usage for the facilities using the conveyorized OSP process average 440
       thousand gallons per year, or about 14 gallons per ssf of PWB produced.
•      Reported water usage for the facilities using the non-conveyorized OSP process average
       89 thousand gallons per year, or 1.9 gallons per ssf of PWB produced.       '

       Chemical constituents and concentrations .in wastewater could not be adequately
characterized from questionnaire data. In the absence of quality data from industry, a model was
developed to estimate the mass loading of constituents within the wastewater, resulting from
drag-out, during the production of 260,000 ssf of PWB by the surface finishing process.  The
mass of chemical transferred per day to the wastewater, as well as other model results, are
presented in Appendix E.           .
                                         3-27

-------
3.1 SOURCE RELEASE ASSESSMENT
       Bath Chemicals Used (I,). Bath concentrations of individual chemical constituents are
presented in Appendix B.  The volume of chemicals consumed per year was determined by
modeling the time it would take the generic OSP process described in Figure 3-6 to produce a
specific PWB throughput. A detailed description of the process modeling is presented in Section
4.2, Cost Analysis.  The number of bath replacements (calculated from the modeled time) was
then multiplied by the volume of the bath to determine the volume of a bath chemical consumed
per year. OSP process chemical consumption is presented in Appendix G.

       Cleaning Chemicals (I4). Three out of 31 OSP baths were reported to be cleaned using
chemicals, however, data concerning the type of cleaning chemical(s) were not collected by the
questionnaire. All of the chemical flushing reported for OSP processes was used for cleaning the
OSP tank during bath replacement. Water is most frequently used to clean tanks prior to new
bath make-up.

       Spent Bath Solutions (W2).  The quantity of spent bath solution could not be determined
directly from the questionnaire data. However, the volume of spent bath chemistry was calculated
by determining the number of bath changes required per year and multiplying by the average
volume of the process tank (see Section 4.2, Cost Analysis).  The concentrations of chemical
constituents within the spent bath solutions are assumed to be the same as make-up bath
concentrations.

       Spent bath treatment and disposal methods were presented in Table 3-4. Precipitation
pretreatment, pH neutralization, and drummed for off-site treatment are reported as common
treatment methods for the conveyorized OSP processes. Respondents for the non-conveyorized,
vertical process reported that pH neutralization and precipitation pretreatment are common
treatment methods.

       Evaporation From  Baths (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 Status. In response to a separate
question regarding spent bath treatment (see Table 3-4), four out of 28 OSP baths were reported
to be drummed and sent off-site for recycling or disposal.

Immersion Silver Process

       Figure 3-7 depicts the generic immersion silver process steps and typical bath sequence
evaluated in the CTS A 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 1
^
2. Water Rinse xl 1
V
3 . Microetch

D
• . ^
4. Water Rinse xl
3
y
5- Predip 1
>
ฐ- Immersion Silver 1
. y

7. Water Rinse xl
*

8- Dry
J

~J

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

-------
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 the two respondents using the
immersion silver process, both reported using the conveyorized process configuration. In
summary:

•      Reported water usage for the facilities using the conveyorized immersion silver process
       average 910 thousand gallons per year, or about 37 gallons per ssf of PWB produced.

       Chemical constituents and concentrations in wastewater could not be adequately
characterized from questionnaire data. In the absence of quality data from industry, a model was
developed to estimate the mass loading of constituents within the wastewater, resulting from
drag-out, during the production of 260,000 ssf of PWB by the surface finishing process. The
mass of chemical transferred per day to the wastewater, as well as other model results, are
presented in Appendix E.

       Bath Chemicals Used (lj).  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 (T4). 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 drammed for off-site treatment are reported as common
 treatment methods for the conveyorized immersion silver processes.

        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 conveyorized immersion silver processes, circulation pumps are used to mix all
        process wet chemistry baths.  The spraying of chemicals onto the surface of the PWB in
        the cleaner and microetch baths is also reported. All of the process baths were reported as
        fully enclosed. Only one out often process baths was reported to be vented to the
        outside.
 •       Table 3-3 lists bath the surface area, volume, and bath temperature data reported by
        respondents to the PWB Workplace Practices Questionnaire.

        Evaporation From Drying/Ovens (A2).  Oven drying occurs directly after the immersion
 silver bath.  Any solution adhering to the PWBs is volatilized during the drying of the PWBs by
 the oven. Air emissions resulting from oven drying were not modeled. No air knife is required by
 this process.                                          •

        Chemicals Incorporated Onto PWBs (PJ). Silver is added to the boards in the
 immersion silver processes. A hydrophobic  layer, formed with a co-deposited organic inhibitor, is
 also coated on top of the silver layer.  The amount of silver incorporated onto a PWB was
 calculated at 0.0013 oz/ssf. Silver consumption is discussed further in Section 5.1, Resource
 Conservation.

       Drummed Solid or Liquid Waste (S3). Questionnaire respondents indicated that none
 of the immersion silver baths contain hazardous waste constituents as defined by RCRA. A
 detailed discussion of RCRA wastes can be found in Section 4.3, Regulatory Status.  In response
 to a separate question regarding spent bath treatment (see Table 3-4), two out of eight immersion
 silver baths were reported to be drummed and sent off-site for recycling.

 Immersion Tin Process

       Figure 3-8 depicts the generic immersion tin process steps and typical bath sequence
 evaluated in the CTSA.  The process baths shown in the figure represent an amalgamation of the
various products offered within the immersion tin technology category. The number and location
of rinse steps displayed in the. figure are based on PWB Workplace Practices Questionnaire
responses. Thus, Figure 3-8 describes the types and sequence of baths in a generic immersion tin
line, but the types and sequence of immersion tin process baths used by any particular facility
could vary. A detailed description of the immersion tin process is presented in Section 2.1.3,
Chemistry and Process Descriptions of Surface Finishing Technologies.
                                          3-31

-------
3.1 SOURCE RELEASE ASSESSMENT


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

1

I

1

5- Prcdip 1
y
6. Water Rinse z 1

D
y
7. Immersion Tin
y
8. Water Rinse x 2
y
1

1

9. Dry )


       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 (I^. Bath concentrations of individual chemical constituents are
 presented in Appendix B. The volume of chemicals consumed per year was determined by
 modeling the time it would take the generic immersion tin process described hi Figure 3-8 to
 produce a specific PWB throughput. A detailed description of the process modeling is presented
 hi 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 (T4). 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 hi 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 npn-conveyorized, vertical process
 reported that pH neutralization, precipitation pretreatment, ion exchange with  on-site metal
 reclaim and drummed for off-site treatment are all treatment options reported by respondents.

       Evaporation From Baths (At). Air releases are modeled in Section 3.2, Exposure
 Assessment. A summary of data collected from the questionnaire is presented below:

 •      For the conveyorized immersion tin processes, circulation pumps are the most reported
       mixing methods for all baths. Full enclosure and venting are the most common methods of
       vapor control reported by respondents for baths other than the pre-dip bath.
       For non-conveyorized immersion tin processes, panel agitation and circulation pumps are
       the most reported mixing methods for all baths. Venting to the outside is the most
       prevalent form of vapor control reported (33 percent), though the use of bath covers are
       also reported.
       Table 3-3 lists the bath surface area, volume, and bath temperature data reported by
       respondents to the PWB Workplace Practices Questionnaire.

       Evaporation From Drying/Ovens (A2).  Oven drying occurs directly after the immersion
tin bath.  Any solution adhering to the PWBs is volatilized during the drying of the PWB by the
oven.  Air emissions resulting from oven drying were not modeled. No air knife is required by
this process.
                                         3-33

-------
3.1 SOURCE RELEASE ASSESSMENT
       Chemicals Incorporated Onto PWBs (Pj). A layer of metallic tin is deposited onto the
PWB by the immersion tin processes. The amount of tin incorporated onto a PWB was
calculated at 0.0038 oz/ssf. Tin consumption is discussed further in Section 5.1, Resource
Conservation.

       Drummed Solid or Liquid Waste (S3).  Questionnaire respondents indicated that none
of the immersion tin baths contain hazardous waste constituents as defined by RCRA. A detailed
discussion of RCRA wastes can be found in  Section 4.3, Regulatory Status. In response to a
separate question regarding spent bath treatment (see Table 3-4), five out of 17 immersion tin
baths were reported by respondents to be drummed and sent off-site for recycling or disposal.

3.1.4  Uncertainties in the Source Release Assessment

       Uncertainties and variations in the data include both gaps in knowledge (uncertainty) and
variability among facilities and process alternatives. These are discussed below.

       For the PWB Workplace Practices Questionnaire data:

•      There may be uncertainties due to misinterpretation of a question, not answering a
       question that applies to that facility, reporting inaccurate information or numbers in
       different units (e.g., using a mass unit to report a volumetric measurement).  Also, because
       of a limited number of responses for  the alternative processes, information more typical
       for that process may not be reported.
•      Variation can occur within or among process alternatives, or from difference due to
       varying amounts of PWB produced.  According to the questionnaire database query
       results, data from facilities with small amounts  of PWB produced often produce unrealistic
       results. Again, for surface finishing process alternatives with a limited number of
       responses, statistical summaries of the data may be precluded, and data may not be
       representative of most PWB facilities.

       For the supplier-provided data:

•      Knowledge gaps include a lack of information on proprietary chemicals, incomplete bath
       composition data, and the reporting of wide ranges of chemical concentrations on a
       MSDS rather then specific amounts in the formulations.
•      Variation in bath chemistries and process specifications among suppliers can occur for a
       given process alternative. The publicly-available bath chemistry data, chemical
       concentrations, and supplier recommendations  may not apply to a specific facility due to
       variation in process set-up and operation procedures.

       Other uncertainties pertain to the applicability and accuracy of estimates and assumptions
used in this assessment.
                                          3-34

-------
                                                               3.2 EXPOSURE ASSESSME1ST
 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 hi 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 conveyorized lines.

       Chemical Bath Additions.  Methods of chemical additions, from the database, are as
follows:

•      Most facilities pour chemicals directly into the bath or tank.
•      Other reported methods include manual pumping, or some combination of pumping,
       pouring, and/or scooping chemical formulations into a bath.

Data were collected for the length of time required to make chemical additions, and on the criteria
used to determine when to add chemicals to the baths.  Some facilities base chemical addition
requirements on time elapsed, some on the surface area of boards processed, and some on the
concentration of key constituents. For these reasons, complicated by the fact that most facilities
running alternatives to HASL do not run those lines at full capacity, a typical addition frequency
could not be determined.  Therefore, exposure was not quantified separately for this activity.

       Chemical Bath Replacement. This process includes removing the spent bath, cleaning
the empty tank, and making up fresh bath solutions. In this process, a worker could be exposed
to chemicals hi the spent bath, on the inside walls of the emptied bath, or to chemicals hi the new
bath solution.

       Rack Cleaning. The racks that hold PWB panels can be cleaned in a variety of ways.
These include cleaning hi a chemical bath on the surface finishing line or using non-chemical
cleaning methods. Of the six facilities that provided information on rack cleaning, four reported
using non-chemical methods, one reported using a chemical bath on the surface finishing line, and
one reported shipping racks offsite for cleaning.  Dermal exposure for rack cleaning is not
quantified separately for this activity.

       Conveyor Equipment Cleaning. Conveyor equipment cleaning involves regular
equipment maintenance for conveyorized surface finishing lines. Methods include chemical baths
on the surface finishing line, chemical rinse, manual scrubbing with chemicals, non-chemical
cleaning, and continuous cleaning as part of the process line. It was assumed that workers could
be exposed to bath chemicals during cleaning.

       Filter Replacement. Filter replacement could result in exposure to the material on the
filter or in the bath. Whether the pathway is significant to worker risk will depend, in part, on the
chemical constituents in the bath.

       Use of Personal Protective Equipment (PPE). It is assumed that the only PPE used is
eye protection, and that the line operator's hands and arms may contact bath solutions. This is a
conservative but consistent assumption for all process alternatives and worker activities,
                                          3-38

-------
                                                                 3.2 EXPOSURE ASSESSMENT
 particularly for dermal exposure. While many PWB facilities reported that line operators do wear
 gloves for various activities, the assumption that the line operator's hands and arms may contact
 bath solutions is intended to account for the fraction of workers who do not.  For workers who
 do wear gloves, dermal contact exposure is expected to be negligible.

 Summary of Occupational Scenarios

        Surface Finishing Line Operators.  In general, line operators perform several activities,
 as described above, including working in the surface finishing process area, surface finishing line
 operation, chemical bath replacement, conveyor equipment cleaning, filter replacement, chemical
 bath sampling, and making chemical bath additions. Some kind of local ventilation is typically
 used for the process line.

        There are two different scenarios for line operators depending on process configuration.
 For non-conveyorized processes, dermal exposure could occur through routine line operation as
 well as bath maintenance activities. Inhalation exposure could occur throughout the time period a
 line operator is in the surface finishing process area.  Conveyorized processes are enclosed and the
 line operator does not contact the bath solutions in routine line operation; he or she only loads
 panels at the beginning of the process and unloads them at the end of the process.  For
 conveyorized processes, dermal exposure is primarily expected through bath maintenance
 activities such as bath replacement, filter replacement, bath sampling, and conveyor equipment
 cleaning. Because the conveyorized lines are enclosed and typically vented to the outside,
 inhalation exposure to line operators and other workers is much lower than for the conveyorized
 processes and are not presented separately.4

       Laboratory Technicians. In general, laboratory technicians perform one activity
 pertaining to the surface finishing line, chemical bath sampling, in addition to working in the
 surface finishing process area. Bath sampling exposure is quantified separately for laboratory
 technicians.

       Other Workers in the Surface Finishing Process Area. Other workers in the surface
 finishing process area may include maintenance workers,  supervisory personnel, wastewater
 treatment operators, contract workers, and other employees.  They perform activities not directly
 related to the surface finishing line, but typically spend some time in the surface finishing process
 area. Because the line operators spend the most amount of time per shift, exposure via inhalation
 is quantified for them (for non-conveyorized processes), and is characterized for the other
 employees in terms of the time spent in the process area relative to line operators.
     Inhalation exposures for conveyorized process configurations were initially assumed to be negligible, and are not
presented separately here. Some inhalation exposure is possible, however, during sampling and bath replacement, when
the baths are opened for a short period of time. After characterizing risks from inhalation for non-conveyorized lines,
inhalation exposures and risks were estimated for the subset of inhalation chemicals of concern for conveyorized lines.
No chemical exposures from inhalation resulted in risks above concern levels for conveyorized lines.

                                           3-39

-------
   EXPOSURE ASSESSMENT
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-conveyorized lines.
                                          3-40

-------
                                                                           3.2 EXPOSURE ASSESSMENT
'"' '^-Activities >
Chemical Bath
Replacement;
Conveyor Equipment
Cleaning; Filter
Replacement;
Chemical Bath Sampling
Rack Cleaning
Chemical Bath Additions
Potential Pathways >,
Dermal contact with
chemicals in bath or on
filters.
Inhalation of vapors or
aerosols from surface
finishing baths.
Dermal contact with
chemicals on racks.
Inhalation of vapors or
aerosols from surface
finishing baths.
Dermal contact with
chemicals added.
Inhalation of vapors or
aerosols from surface
finishing baths or while
making bath additions.
Evaluation Approach and Rationale
Exposure quantified for conveyorized lines for all
activities together (bath sampling quantified
separately for laboratory technicians). Exposure not
quantified separately for these activities on non-
conveyorized lines.
Not quantified separately. Included in "working in
process area" for non-conveyorized lines; not
quantified due to modeling limitations for
conveyorized lines. b
Not quantified; limited data indicate this is not
performed by many facilities.
Not quantified separately. Included in "working in
process area" for non-conveyorized lines; not
quantified due to modeling limitations for
conveyorized lines.
Not quantified separately from chemicals already in
the baths.
Not quantified separately. Included in "working in
process area" for non-conveyorized lines; not
quantified due to modeling limitations for
conveyorized lines.
Laboratory Technicians
Chemical Bath Sampling
Working in Process Area
Dermal contact with
chemicals in surface
finishing baths.
Inhalation of vapors or
aerosols from surface
finishing baths.
Inhalation of vapors or
aerosols from surface
finishing baths.
Exposure quantified for conveyorized and
non-conveyorized lines.
Not quantified separately (included in "working in
process area").
Exposure quantified for line operators for non-
conveyorized lines; exposure for other workers is
proportional to their exposure durations.
Maintenance Workers, Supervisory Personnel, Wastewater Treatment Operators, Contract
Workers, and Other Workers
Working in Process Area
Inhalation of vapors or
aerosols from surface
finishing baths.
Dermal contact with
chemicals in surface
finishing baths.
Exposure quantified for line operators for non-
conveyorized lines; exposure for other workers is
proportional to their exposure durations.
Not quantified. a .
  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 Population Exposure Pathways
.Population
Residents Living
Near a PWB Facility
Ecological
Potential Pathways
Inhalation of chemicals released to
air.
Contact with chemicals released to
surface water directly or through
the food chain.
Exposure to chemicals released to
land or groundwater.
Exposure to chemicals released to
surface water.
Exposure to chemicals released to
air or land.
Evaluation Approach and Rationale
Exposure quantified for all potential
carcinogens and any other chemical released at
a rate of at least 23 kg/year.
Not evaluated. Direct exposure to surface
water is not expected to be a significant
pathway; modeling exposure through the food
chain (e.g., someone catching and eating fish)
would be highly uncertain.
Not evaluated. Not expected to be a significant
pathway; modeling releases to groundwater
from a landfill would be highly uncertain.
Screening-level evaluation performed.
Not evaluated. The human (residential)
evaluation air exposure could be used as a
screening-level assessment for animals living
nearby. Releases directly to land are not
expected, and animals are not directly exposed
to groundwater.
       Population exposures may occur through releases to environmental media (i.e., releases to
air, water, and land). The pathways for which exposure, is estimated are inhalation of chemicals
released from a facility to a nearby residential area and releases of chemicals in wastewater to a
receiving stream, where aquatic organisms, such as fish, may be exposed through direct contact
with chemicals in surface water.

       Air releases from the surface finishing process are modeled for the workplace. These
modeled emission rates are used in combination with an air dispersion model to estimate air
concentrations to a nearby population.

       Exposures and risks from surface water are evaluated by identifying chemicals potentially
released to surface water from process rinse steps following wastewater treatment. This exposure
pathway is described in Section 3.2.3.         .                                     .

       Possible sources of releases to land from surface finishing processes include bath filters
and other solid wastes from the process line, chemical precipitates from baths, and sludge from
wastewater treatment. These sources are discussed in Section 3.1, Source Release Assessment.
Reliable characterization data for potential releases to land are not available; therefore, the
exposure assessment does not estimate the nature and quantity of leachate from landfills or effects
on groundwater.
                                          3-42

-------
                                                               3.2 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.  Bath concentrations for
 dermal exposure were estimated from bath chemistry data.  Monitoring data were used for lead
 from the HASL process. Fate and transport modeling were performed to estimate air
 concentrations for workplace and surrounding population exposures.  Use of monitoring data and
 modeling used to estimate air concentrations are described in this section.

 Monitoring Data

        Air monitoring data for lead have been provided by one PWB manufacturing facility.  A
 combination of personal monitoring for HASL line operators and air samples from the HASL
 process area result in an average air concentration of 0.003 mg/m3.  It should be noted that these
 monitoring data are limited to only one PWB manufacturer, and may vary from facility; to facility.
 In addition, air sampling results from hand soldering operations were reported in one study
 (Monsalve, 1984), ranging from < 0.001 mg/m3 to 2 mg/ni3.

 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 fines
 (Robinson et al., 1997). Three air emission models were used to estimate worker exposure:
 1.
 2.
 3.
Volatilization of chemicals from the open surface of surface finishing tanks.
Volatilization of chemicals induced by air sparging.
Aerosol generation induced by air sparging.
       The first model was applied to determine volatilization of chemicals from un-sparged
baths. For the air-sparged baths, the total air emission rate for chemicals was determined by
summing the releases from all three models.  Modeled emission rates were then put into an indoor
air dilution model to estimate workplace air concentrations. For models 1 and 2, volatilization
was modeled only for those chemicals with a vapor pressure above 10'3 torr (a vapor pressure less
than 10'3 torr was assumed for inorganic salts even if vapor pressure data were not available).  A
review of the relevant literature, descriptions of the models, and examples demonstrating the use
of the models are available in the December 22, 1995, Technical Memorandum, Modeling Worker
Inhalation Exposure (Appendix D).

       Volatilization of Chemicals from the Open Surface of Surface Finishing Tanks.
Most plating tanks have a free liquid surface  from which chemicals can volatilize into the
workplace air. Air currents across the tank will accelerate the rate of volatilization. The
                                          3-43

-------
3.2 EXPOSURE ASSESSMENT
following model for evaporation of chemicals from open surfaces was used, based on EPA's
Chemical Engineering Branch (CEB) Manual (U.S. EPA, 1991a):
                                                           -10.5
where,
 y.o
Az
volatilization rate of chemical;/ from open tanks (mg/min)
concentration of chemical^ in bulk liquid (mg/L)
dimensionless Henry's Law Constant (H,) for chemical^
bath surface area (m2)
molecular diffusion coefficient of chemical y in air (cm2/sec)
air velocity (m/sec)
pool length along direction of air flow (m)
       Concentration of chemical in bulk liquid (CL^) is the bath concentration. The coefficient of
 1,200 includes a factor of 600 for units conversion.

       Henry's constants were corrected for bath temperature. Bath temperature varies by
 process type and bath type; bath temperature data from the questionnaire database were
 determined by specific process type and bath type.

       Bath surface areas used in the air modeling were determined from the questionnaire
 database. For non-conveyorized lines, an overall average for all baths and all processes of 422 sq
 in (0.280 m2) was used.  For conveyorized lines, an average was used for each type of process
 bath, as follows:
Conveyorized Bath Type
Cleaner baths
Immersion silver
Immersion tin
Micrqetch baths
OSP
Predip baths
Average Surface Area (sq in)
1,078
4,364
1,436
1,629
2,573
1,004
       Some limitations of the model should be pointed out. The model was developed to
 predict the rate of volatilization of pure chemicals, not aqueous solutions. The model was also
 derived using pure chemicals. As a result, the model implicitly assumes that mass transfer
 resistance on the gas side is the limiting factor. The model may overestimate volatilization of
 chemicals from solutions when liquid-side mass transfer is the controlling factor.
                                           3-44

-------
                                                               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-exp -
  Fy,* =
 where,
 Fys    =     mass transfer rate of chemical 7 out of the system by sparging (mg/min)
 QG    =     air sparging gas flow rate (L/min)
 Hy     =     dimensionless Henry's Law Constant (He) for chemical;;
 c^y    =     concentration of chemical _y in bulk liquid (mg/L)
 KOLJ   =     overall mass transfer coefficient for chemical 3; (cm/min)
 a      =     interfacial area of bubble-per unit volume of liquid (cnrVcm3)
 VL     =     volume of liquid (cm3)

 Data for the sparging air flow rate (QG) come from information supplied by a PWB manufacturer.

        Aerosol Generation from Air-Sparged Surface Finishing Tanks. Aerosols or mists
 are also potentially emitted from process baths.  This was estimated for electroless nickel baths in
 nickel/gold and nickeypalladium/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 (Bergluhd and Lindh, 1987):
RA  = [5.5x10
                                                      FT FA FD
where,
RA    =      aerosol generation rate (ml/min/m2)
QG    =      air sparging gas flow rate (cnrYmin)
A     =      bath surface area (m2)
FT     ~      temperature correction factor
FA     =      air velocity correction factor
FD     =      distance between the bath surface and tank rim correction factor

       The emission of contaminants resulting from aerosols depends on both the rate of aerosol
generation arid the concentration of contaminants in the aerosol. The following equation is used
to estimate contaminant emission (flux) from aerosol generation:
                                          3-45

-------
3.2 EXPOSURE ASSESSMENT
where,
H    =
Mb    -
                                               flE Fy,s
              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, 1991a):
where,
Q
k
              workplace contaminant concentration (mg/m3)
              total emission rate of chemical from all sources (mg/min)
              ventilation air flow rate (m3/min)
              dimensionless mixing factor
       The CEB Manual commonly uses values of the ventilation rate (Q) from 500 cubic feet
per minute (cfrn) to 3,500 cfrn; a ventilation rate for surface finishing lines of 13.6 m3/min (480
cfin) was determined by taking the 10th percentile air flow rates from the facility questionnaire
database for general ventilation.  An average room volume of 505 m3 (18,200 ft3) was determined
by assuming a ten foot room height and using the average room size from the questionnaire
database.  The combination of room volume and ventilation rate is equivalent to an air turnover
rate of 0.026 per minute (1.56 per hour).  The mixing factor (k) could be used to account for slow
and incomplete mixing of ventilation air with room air; however, a value of 1.0 was used for this
factor consistent with the assumption of complete mixing.

       Other assumptions pertaining to these air models include the following:

•      Deposition on equipment, condensation of vapors, and photodegradation are negligible.
•      Incoming air is contaminant-free.
•      The concentration of contaminant at the beginning of the day is zero.
                                          3-46

-------
                                                              3.2 EXPOSURE ASSESSMENT
•      As much air enters the room as exits through ventilation (mass balance).
•      Room air and ventilation air mix ideally.

       Sensitivity Analysis. Model sensitivity and uncertainty was examined for the making
holes conductive (MHC) project (U.S. EPA, 1998b) using Monte Carlo analysis, with the air
transport equations outlined above, and probability distributions for each parameter based on data
from the PWB Workplace Practices Questionnaire. The analysis was conducted using a Monte
Carlo software package (Crystal Ball™, Decisioneering, Inc., 1993) in conjunction with a
spreadsheet program.  Because the same models are used for this surface finishing evaluation, and
the model facility is similar to that developed for MHC, the results of this sensitivity analysis are
relevant to surface finishing air modeling as well.

       The sensitivity analysis suggested that a few parameters are key to modeling chemical
emissions from PWB tanks. These key parameters are air turnover rate, bath temperature,
chemical concentration in the bath, and HC- The air turnover rate assumption contributes most to
overall model variance. The chemical bath concentration and bath temperature also contribute
variance to the model, but are less important than air turnover rate. This statement is supported
by the fact that relatively accurate information is available on their distributions. HC appears to be
least important of the four, but may have more variability associated with it. The models appear
to be largely indifferent to small changes in most other parameters.

       Modeled emission rates and workplace  air concentrations are presented in Table 3-8.

                     Table:
Chemical a
•>
Total Emission
JRateflE^-)
(mg/min)
Workplace
Air Cone.
(Cy}(mgftn3>
Federal OSHA and/or NTOSH
Permissible Inhalation Exposure "
Limits (mWm*) b
HASL, Noh-conveyorized
1,4-Butenediol
Alkylaryl sulfonate
Arylphenol
Ethylene glycol
Ethylene glycol monobutyl ether
Hydrochloric acid
Hydrogen peroxide
Phosphoric acid
9.8
0.018
0.0060
12
120
0.89
5.2
1.5
0.75
0.0014
4.6E-04
0.94
8.9
0.068
0.40
0.12
none
NR
NR
no OSHA PEL or NIOSH REL
NIOSH REL: 24 (5 ppm)
OSHA PEL: 240 (50 ppm)
NIOSH REL, C: 7 (5 ppm)
OSHA PEL, C: 7>,(5 ppm)
NIOSH REL: 1.4(1 ppm)
OSHA PEL 1.4(1 ppm)
NIOSHREL:1,STEL:3
OSHA PEL: 1
Nickel/Gold, Non-conveyorized
Aliphatic acid A
Aliphatic acid B
Aliphatic acid E
Aliphatic dicarboxylic acid A
77
5.4E-04
100
0.10
5.9
4.1E-05
7.8
0.0080
NR
NR
NR
NR
                                         3-47

-------
3.2 EXPOSURE ASSESSMENT
Chemical a
ซ i*~v*
,. ~^ ~ s -.
Aliphatic dicarboxylic acid C
Alkyldiol
Ammonia compound B
Ammonium hydroxide
Hydrochloric acid
Hydrogen peroxide
Inorganic metallic salt A
Inorganic metallic salt B
Inorganic metallic salt C
Malic acid
Nickel sulfate
Phosphoric acid
Potassium compound
Sodium hypophosphite
Urea compound B
Total Emission
RateO?^
(mg/min)
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
i.o
0.64
7.6E-04
Workplace
Air Cone*
(Cy)
-------
                                                                            3.2  EXPOSURE ASSESSMENT
" Chemical3
•> * *

Propionic acid
Sodium hypophosphite
Urea compound B
Total Emission
JlateCP^
- (mg/min)
26
0.85
0.0015
Workplace
, Air Cone.
(CyKwg/m3)
2.0
0.065
1.2E-04
Federal OSBA and/or MOSH
Permissible Inhalation Exposure
Limits (mgftn3) fr
NIOSH REL: 30 (10 ppm)
STEL:45(15ppm)
none
Ml
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(15ppm)
OSHAPEL:25(10ppm)
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)
NIOSHREL:1,STEL3
OSHA PEL: 1
Immersion Tin, Non-conveyorized
Aliphatic acid D
Alkylaiyl 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)
OSHAPEL,C:7(5ppra)
NR
NIOSHREL:1,STEL:3
OSHA PEL: 1
NR
      -         — 	—— -—'.••*•*ป*•*ปปป, jfiwwiiw**., ฃ i. U.UU.X.UVL woo .uui waivuuucu iui 
-------
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 hi 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 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, hi 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
ah* concentrations for lead were estimated, based on this air dispersion model and HASL
workplace air monitoring data. Results of ambient air modeling are presented in Table 3-9.
     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 mgitive releases less than 23 kg/yr result in exposures of
less than 1 mg/yr for an individual.

                                            _:_—..

-------
                                                                         3.2 EXPOSURE ASSESSMENT
fable 3-9. Results of Ambient Air Modelin
Chemical , >
*•• -~ "•* > *^ฃ*-*
„ -f „ _ _ V - .*- ,- •* -
^ .— -! t_
Emission Rate a
(mg/min)
g
Air Concentration^
(mg/m5)
HASL, Non-conveyorizd
Ethylene glycol monobutyl ether
Lead
120
0.039ฐ
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
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
pretreatment. The pretreated wastewater is then discharged to a POTW.  Consequently,
characterizing the process wastewater has been problematic.  Because many of the chemical
constituents expected in the wastewater of the surface finishing process are also found in other
PWB manufacturing processes, testing data obtained from industry was not sufficient to
characterize what portion of the overall wastewater contamination was actually attributable to the
operation of the surface finishing process. Therefore, a model was developed to estimate
environmental releases to surface water for chemical constituents and concentrations in the
wastewater as a result of the operation of the surface finishing process alone.

       Li 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 arid = 0 otherwise
       The model was used to estimate the mass loading of constituents to the wastewater
resulting from drag-out during the production of 260,000 ssf of PWB by the surface finishing
process, by the following equation:
                            MDij  = P * Cij * DOij / 1,000,000
where,
                                          3-52

-------
                                                              3.2 EXPOSURE ASSESSMENT
 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   =
 Fj
 T
 V
mass of constituent I from dumping bath j, g/d
replacement frequency for bath j, times/yr
operating time (from cost model., total production time minus down time), days/yr
volume of bath j, L
        For non-conveyorized lines, the total mass per day going to wastewater is the sum of
 drag-out mass and bath dumping mass for the constituent in all baths:


                                  Mi = E (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-
 put 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.

       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 + Q^)

where,
CW  =      preliminary surface water concentration of constituent i, assuming no treatment,
              mg/L
Qsw   =      surface water flow rate, L/day
Qww   =      wastewater flow rate, L/day
                                          3-53

-------
3.2 EXPOSURE ASSESSMENT
       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 AppendixH). For any chemicals with preliminary in-stream
concentrations exceeding CCs, a typical treatment efficiency was determined. For this purpose, it
was assumed that wastewater treatment consisted of primary treatment by gravitational settling
followed by complete-mix activated sludge secondary treatment and secondary settling
(clarification), as typically employed at POTWs. Treatment efficiencies were estimated on a
.chemical-by-chemical basis using a combination of readily available data on the chemical or close
structural analogs and best professional judgment. Information sources included EPA's
Treatability Database, the Environmental Fate Data Base (Syracuse Research Corp., updated
periodically), the Handbook of Environmental Degradation Rates (Howard et al., 1991),
wastewater engineering handbooks such as Metcalf and Eddy, and various journal articles from
the published literature.

       Treatment efficiencies were then applied to the chemical concentrations, and revised in-
stream concentrations were calculated. Select inorganic constituents that are targeted by industry
for treatment, such as metals,  were assumed to be treated effectively by on-site treatment to
required effluent levels. These metals are not included in the surface water evaluation.
(Pretreatment is discussed further in Section 6.2, Control Technologies.) Results for chemicals,
excluding metals, where the initial stream concentration (without-treatment) exceeded the CC for
that chemical are presented in Table 3-10. Full results are presented in Appendix E.

           Table 3-10. Estimated Releases to Surface Water Following Treatment
Chemical51 *
? "~ % -•
"Cone, in
Wastewater
Released to
Stream (mg/L)
Stream Cone.
w/oPOTW
Treatment
(mg/L) '
Treatment
Efficiency
(%)
Stream Cone.
after POTW
Treatment
(mg/L)
IIASL, Non-conveyorized
1,4-Butenediol
Alkylaryl sulfonate
Citric acid
Ethylene glycol monobutyl ether
Hydrogen peroxide
Potassium peroxymonosulfate
49
2.3 '
94
71
195
390
0.10
0.0049
0.20
0.15
0.41
0.82
90
0
93
90
90
90
0.010
0.0049
0.014
0.015
0.041
0.082
IIASL, Conveyorized
1,4-Butenediol
Alkylaryl sulfonate ' .
Citric acid
23
1.0
42
0.076
0.0035
0.14
90
0
93
0.0076
0.0035
0.0099
                                           3-54

-------
                                                                           3.2 EXPOSURE ASSESSMENT
Chemical"-
. ' ^ - '> %.
^A "• *•• ^. /*• •"• •"
' V -K
v* ^ - ^
Ethylene glycol monobutyl ether
Hydrogen peroxide
Potassium peroxymonosulfate
Cone, fia
Wastewafer
'-'Released to
Stream (mg/L)
32
90
180
Stream Cone.
w/oPOT\^
Treatment
'_ Cmg^L>
0.11
0.30
0.61
Treatment
Efficiency
(%)
90
90
90
Stream Cone.
, after POTW
Treatment'
~(rag/L) l
0.011
0.030
0.061
Nickel/Gold, Non-conveyorized
Hydrogen peroxide
Substituted amine hydrochloride
62
97 .
0.045
0.070
90
80
0.0045
0.014
Nickel/Palladium/Gold, No n-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
Q.0029
0.00023
0.026
[mmersion Tin, Non-conveyorized
Alkylaryl sulfonate
Citric acid
Ethylene glycol monobutyl ether
Potassium peroxymonosulfate
Quantenary alkylammonium chlorides
Thiourea
Urea compound C
1.2
660
36
200
42
170
35
0.0021
1.2
0.064
0.36
0.074
0.30
0.062
0 '
93
90
90
90
90
90
0.0021
0.082
0.0064
0.036
0.0074
0.030
0.0062
Immersion Tin, Conveyorized '
Potassium peroxymonosulfate
68
0.041
90
0.0041
a This includes any chemicals, except metals, where the initial stream concentration (without treatment) exceeded the
CC for that chemical. Metals are not included; it was assumed that metals are targeted for effective on-site treatment.
                                                  3-55

-------
3.2 EXPOSURE ASSESSMENT
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

IR
ET
                                     I =  (Ca)(IR)(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 (nrVhr)
exposure time (hr/day)
       Daily exposures are averaged over a lifetime (70 years) for carcinogens, and over the
exposure duration (e.g., 25 years working in a facility) for non-carcinogens.7 The following
equations are used to estimate average daily doses for inhalation:

                           LADD = (I)(EF)(ED)/[(BW)(ATCAR)]
                            ADD  = (T)(EF)(ED)/[(BW)(ATNC)]
where,
LADD
ADD
I
EF
ED
BW
ATCAR
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)
     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
        Parameter values for estimating workers' potential dose rates from inhalation are
 presented in Table 3-11.
Table 3-11. Parameter Values for Workplace Inhalation Exposures
Parameter
Air Concentration (Ca)
Inhalation Rate (IR)
Units
mg/m3
m3/hr.
Value
Source of Data, Comments
Modeled from bath concentrations (see Table 3-9).
1.25
U.S. EPA, 1991a (data from
NIOSH, 1976).
Exposure Time (ET) .
Line Operation
Working in Process Area
hrs/day
hrs/day
Exposure Frequency (EF)
Line Operation &
Working in Process Area
Exposure Duration
(ED)
Body Weight (BW)
Averaging Time (AT)
•"••••CAR
ATNC .
days/yr
years
kg
days
8
laboratory technician 	 	 28
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.

HASL(NC) 	 44
HASL (C) 	 22
Nickel/Gold (NC) 	 	 212
Nickel/Palladium/Gold (NC) . . 280
OSP(NC) 	 	 . 35
OSP(C) 	 	 	 16
Immersion Silver (C) 	 	 64
Immersion Tin (NC) 	 75
Immersion Tin (C) 	 	 	 107
25
70
25,550
9,125
From process cost model, based on
the number of days per year
required to produce 260,000 ssf of
finished PWB. Assumed this is the
time worked per year.
95th percentile for job tenure
(Bureau of Labor Statistics, 1990).
(Median tenure for U.S. males is 4
years; Bureau of Labor Statistics,
1997.)
Average for adults (U.S. EPA,
1991b).
70 yrs (average lifetime) x 365 d/yr
25 yrs (ED) x 365 d/yr
Workplace Dermal Exposures

       Two approaches were considered for evaluating dermal exposure.  The general model for
potential dose rate via dermal exposure to workers from the CEB Manual (IIS. EPA, 199la) is
as follows:

                                      D •= SQC
                                         3-57

-------
3.2 EXPOSURE ASSESSMENT
where,
D
S
Q
c
dermal potential dose rate (mg/day)
surface area of contact (cm2)
quantity typically remaining on skin (mg/cm2)
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

       As indicated earlier, daily exposures are averaged over a lifetime (70 years) for
carcinogens, and over the exposure duration (e.g., 25 years working in a facility) for non-
carcinogens. The following equations are used to estimate average daily doses from dermal
contact:

                          LADD = (D)(EF)(ED)/[(BW)(ATCAR)]
                           ADD  = (D)(EF)(ED)/[(BW)(ATNC)]
where,
D
dermal potential dose rate (mg/day)
       General parameter values for estimating workers' potential dose rates from dermal
exposure are presented in Table 3-12.
   8 This permeability coefficient-based approach is recommended over the absorption fraction approach for
compounds in an aqueous media or in air (U.S. EPA 1992a).

-------
                                                              3.2 EXPOSURE ASSESSMENT
Table 3-12. General Parameter Values for Workplace Dermal Exposures
Parameter
Chemical
Concentration (C)
Skin Surface Area (S)
Flux Through Skin (f)
Exposure Duration
(ED)
Body Weight (BW)
Averaging Time (AT)
ATCAR
ATNC
Units
%
cm2
cm/hr
years
kg
days
Value
Source of Data, Comments
Range of reported values and average determined from bath chemistry
data.
800
Default for inorganics: 0.001 estimate
for organics by: log f = -2.72+0.71
log Kow- 0.0061 (MW)
(K^ = octanol/water partition
coefficient, MW = molecular weight)
25 •
70
25,500
9,125
CEB, routine immersion, 2 hands,
assuming gloves not worn.
U.S. EPA, 1992a.
95th percentile for job tenure
(Bureau of Labor Statistics, 1990).
(Median tenure for U.S. males is 4
years; Bureau of Labor Statistics,
1997.)
U.S. EPA, 1991b.
70 yrs (average lifetime) x 365 d/yr
25 yrs (ED) x 365 d/yr
       Dermal exposure was quantified for line operators performing routine line operation
activities on non-conveyorized lines.  Parameter values used in the dermal exposure equations are
provided in Table 3-13.
                                         3-59

-------
   EXPOSURE ASSESSMENT
    Table 3-13. Parameter Values for Workplace Dermal Exposures for Line Operators
                               on Non-Conveyorized Lines
Parameter/
Activity*
Units
Value
*jv
Source of Data, Comments
v> -^ *"•ฃ
Exposure Time (ET)
Line Operation a
hrs/day
Process / no. baths or steps
HASL(NC)/10
Nickel/Gold (NC) 714
Nickel/Palladium/Gold (NC) / 22
OSP(NC)/9
Immersion Tin (NC) / 12
Value
0.80
0.57
0.36
0.89
0.67
Exposure Frequency (EF)
Line Operation "
days/yr
HASL (NC) 	 44
HASL (C) 	 	 22
Nickel/Gold (NC) 	 	 212
Nickel/Palladium/Gold (NC) 	 .... 280
OSP (NC) 	 	 	 35
OSP (C) 	 16
Immersion Silver (C) 	 64
Immersion Tin (NC) 	 	 	 75
Immersion Tin (C) 	 .107
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.

From cost process simulation model,
based on a throughput of 260,000
ssf.
 Dermal exposure on non-conveyorized lines was quantified for line operation activities only, because this would result
in higher line operator exposure than any other activities that may be performed (e.g., bath sampling, filter
replacement).
       Dermal exposure was quantified for line operators on conveyorized lines for chemical bath
replacement, conveyor equipment cleaning, filter replacement, and bath sampling activities.
Because 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 3-14.
                                           3-60

-------
                                                     3.2 EXPOSURE ASSESSMENT
Table 3-14. Parameter Values for Workplace Dermal Exposure for Line Operators on
                             Conveyorized Lines
Parameter/
Activity*
Units b
\
Value " ~ *
s " ?
-J f1 "• f-
Source of Data, Comments
Exposure Time (ET)
Chemical Bath
Replacement
Filter
Replacement
Chemical Bath
Sampling
min/occur
min/occur
min/occur
HASL 	 	 . 264
OSP 	 ion
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 mat dermal contact from line operation would be negligible.
b min/occur = minutes per occurance; occur/year = number of occurances per year.
       Dermal exposure was also quantified for a laboratory technician on all conveyorized and
non-conveyorized lines for chemical bath, sampling activities.  Parameter values used in the
exposure equations for a laboratory technician are provided in Table 3-15.

      Table 3-15. Parameter Values for Workplace Dermal Exposure for a Laboratory
               Technician on Either Conveyorized or Nou-Gonveyorized Lines
Parameter/
Activity
Units a
Value „
- " ' ' t „ '"" ซ.*<''
Source of Data, Comments
*\
, , -• * i
Exposure Time (ET)
Chemical Bath
Sampling
min/occur
HASL 	 - . 15
Nickel/Gold 	 10
Nickel/Palladium/Gold 	 1.5
OSP 	 22
Immersion Silver 	 1 .0
Immersion Tin 	 5.0
Questionnaire data for sampling
duration were combined
regardless of process
configuration.
Exposure Frequency (EF)
Chemical Bath
Sampling
occur/year
HASL (NC) 	 135
HASL (C) 	 67
Nickel/Gold (NC) 	 	 954
Nickel/Palladium/Gold (NC) 	 2,406
OSP(NC) 	 	 	 436
OSP (C) 	 200
Immersion Silver (C) ,.,.,....... 253
Immersion Tin (NC) 	 341
Immersion Tin (C) 	 485
From cost process simulation
model, based on a throughput
of 260,000 ssf.
  min/occur = minutes per occurance; occur/year = number of occurances per year.
                                             3-62

-------
                                                            3.2 EXPOSURE ASSESSMENT
Results

       Table 3-16 presents results for estimating ADDs for inhalation and dermal workplace
exposure for line operators and laboratory technicians.

   Table 3-16. Estimated Average Daily Dose for Workplace Exposure From Inhalation
                                 and Dermal Contact
Chemical
•- k ^ ^ ~ฐ ~~~
7 ADD*
(mg/kg-day) -
Inhalation
line
Operator
JDennal
Line
Operator
Laboratory
Technician
HASL, Non-conveyorized
1,4-Butenediol
Alkylalkyne diol
Alkylaryl sulfonate
Alkylphenol ethoxylate
Alkylphenolpolyethoxyethanol
Aryl phenol
Citric acid
Copper Sulfate Pentahydrate
Ethoxylated alkylphenol A
Ethoxylated alkylphenol B
Ethylene glycol
Ethylene glycol monobutyl ether
Fluoboric acid
Gum
Hydrochloric acid
Hydrogen peroxide
Hydroxyaryl acid
Hydroxyaryl sulfonate
Phosphoric acid
Potassium peroxymonosulfate
Sodium benzene sulfonate
Sodium hydroxide
Sulfunc 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
NA
NA
NA
NA
8.53E-05
5.47E-07
2.29E-08
6.61E-29
6.35E-06
4.07E-08
1.71E-09
4.92E-30
                                        3-63

-------
3.2 EXPOSURE ASSESSMENT
Chemical
1 >-v* •*
* , ,_*- „.*ป
•* * .* \
jr ,* ^
t
Alkylphenolpolyethoxyethanol
Aiyl 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
_, ** ,-^viv * * ^- <-•';„ **
r ~ ' (mg/kg-day) - „ _
Inhalation
Line
Operator
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
,. Dermal
Line
Operator
6.23E-28
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
4.64E-29
6.15E-06
1.32E-05
1.53E-04
3.90E-30
2.78E-30
1.60E-05.
1.09E-04
4.19E-05
NAb
7.08E-05
1.72E-04
2.95E-06
1.04E-07
2.08E-04
3.45E-04
5.75E-10
5.77E-07
7.27E-04
Nickel/Gold, Non-conveyorized
Aliphatic acid A
Aliphatic acid B
Aliphatic acid E
Aliphatic dicarboxylic acid A
Aliphatic dicarboxylic acid C
Alkylamino acid B
Alkyldiol
Alkylphenolpolyethoxyethanol
Ammonia compound B
Ammonium chloride
Ammonium hydroxide
Citric acid
Copper sulfate pentahydrate
Ethoxylated akylphenol B
Hydrochloric acid
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
2.35E-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
1.53E-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
                                     3-64

-------
                   3.2 EXPOSURE ASSESSMENT
Chemical f r <
t <(,,-• t
w -• ?
, r 4" " '
- > <•
-v ซV \
Hydrogen peroxide
Hydroxyaryl acid
Inorganic metallic salt A
Inorganic metallic salt A (LADD) ฐ
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
r , ADDa^ ,* „ , '
N * %* (mg/kg-day) * „ -
, Inhalation \
Line
Operator
2.40E-02
NA
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 *>- w " * *
' Line
Operator
1.36E-01
3.30E-03
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
Teehnician
8.84E-03
2.14E-04
5.19E-07
1.85E-07
3.45E-05
3.61E-07
1.37E-04
9.17E-03
3.25E-04
1.25E-02
1.73E-02
7.39E-04
NAd
2.22E-02
4.19E-05
1.05E-02
. 1.48E-02
5.55E-02
1.48E-04
1.56E-06
Nickel/Palladium/Gold, Non-conveyorized
Aliphatic acid B
Aliphatic acid E
Aliphatic dicarboxylic acid A
Aliphatic dicarboxylic acid C
Alkylamino acid B
Alkyldiol
Alkylpolyol
Amino acid salt
Ammo carboxylic acid
Ammonia compound A
Ammonia compound B
Ammonium hydroxide
Citric acid
Copper sulfate pentahydrate
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
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
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
3-65

-------
3.2 EXPOSURE ASSESSMENT
Chemical
- > v " *
j. ^ „ V-
* / >
^ *="
ซ, "^ * t
J. •ป- is 0-,T
",,.., <•'
Ethoxylated alkylphenol
Ethylenediamine
Hydrochloric acid
Hydrogen peroxide
Hydroxyaryl acid
Inorganic metallic salt B
Maleic acid
Malic acid
Nickel sulfate
Palladium salt
Phosphoric acid
Potassium compound
Potassium gold cyanide
Propionic acid
Sodium hydroxide
Sodium hypophosphite
Sodium salt
Substituted amine hydrochloride
Sulfuric acid
Surfactant
Transition metal salt
Urea compound B
/ j
^ -I •*&
Inhalation
Line
Operator
NA
5.32E-04
2.35E-01
3.11E-02
NA
1.79E-05
NA
1.92E-03
7.50E-03
NA
1.01E-02
8.98E-03
NA
2.13E-01
NA
7.11E-03
NA
NA
NA
NA
NA
1.28E-05
„: ^DDa, fc v
;(mงT^-dayj
~*
•" Permal
Line
Operator
2.61E-27
4.14E-04
3.92E-01
1.14E-01
2.77E-03
2.07E-03
1.36E-03
1.77E-03
1.87E-01
1.02E-02
1.62E-01
2.24E-01
9.56E-03
2.69E-02
5.42E-04
1.93E-01
4.78E-01
1.91E-01
4.99E-01
3.19E-04
1.91E-03
3.94E-05
Laboratory^
Technician
6.42E-29
1.02E-05
9.63E-03
2.81E-03
6.81E-05
5.08E-05
3.35E-05
4.34E-05
4.59E-03
2.51E-04
3.98E-03
5.50E-03
2.35E-04
6.60E-04
1.33E-05
4.75E-03
1.18E-02
4.70E-03
1.23E-02
7.83E-06
4.70E-05
9.67E-07
OSP, Non-conveyorized
Acetic acid
Alkylaryl imidazole
Aromatic imidizole product
Arylphenol
Copper ion
Copper salt C
Copper sulfate pentahydrate
Ethoxylated alkylphenol
Ethylene glycol
Gum
Hydrochloric acid
Hydrogen peroxide
7.79E-02
NA
NA
6.18E-06
NA
NA
NA
NA
2.38E-02
NA
2.04E-03
1.92E-03
3.75E-02
5.50E+00
6.33E-03
1.77E-03
4.95E-02
1.36E-03
4.41E-02
8.03E-28
4.63E-03
NAb
2.33E-02
1.78E-02
2.45E-03
3.59E-01
4.13E-04
L16E-04
3.23E-03
8.89E-05
2.88E-03
5.24E-29
3.02E-04
NAb •
1.52E-03
1.16E-03
                                     3-66

-------
                   3.2 EXPOSURE ASSESSMENT
Chemical
ป ^ ' f ~^'
- ~-*~ -ซ,,': .
* V ซ• *
•*- * *
*" 'ซ A • ~-" j v '-"
Hydroxyaryl acid
Hydroxyaryl sulfonate '
Phosphoric acid
Sodium hydroxide
Sulfunc acid
^ /., - - / 4^?^ , '-- '" * ^ "• >
\* - ~^ (mg/kg-day) " ^/"
Inhalation
Line
w
'• Operator
NA
NA
1.27E-03
NA
NA
Dermal
Line-
t Operator
8.52E-04
3.00E-05
4.98E-02
1.67E-04
2.55E-01
Laboratory
Technician'
5.57E-05
1.96E-06
3.25E-03
1.09E-05
1.66E-02
OSP, Conveyorized
Acetic acid
Alkylaryl imidazole
Aromatic imidizole product
Aryl phenol
Copper ion
Copper salt C
Copper sulfate pentahydrate
Ethoxylated alkylphenol
Ethylene glycol
Gum
Hydrochloric acid
Hydrogen peroxide
Hydroxyaryl acid
Hydroxyaryl sulfonate
Phosphoric acid
Sodium hydroxide
Sulfuric acid
NA
NA
NA
NA
NA
NA
NA
NA
NA .
NA
NA
NA
NA
NA
NA
NA
NA
1.78E-03
2.61E-01
3.00E-04
8.93E-05
2.34E-03
6.45E-05
2.22E-03
4.04E-29
2.33E-04
NAb
1.17E-03
8.96E-04
4.29E-05
1.51E-06
2.50E-03
8.38E-06
1.28E-02
5.30E-04
7.78E-02
8.94E-05
2.51E-05
6.99E-04
1.92E-05
6.23E-04
1.13E-29
6.54E-05
NAb
3.30E-04
2.51E-04
1.20E-05
4.24E-07
7.03E-04
2.35E-06
3.60E-03
Immersion Silver, Conveyorized
1,4-Butenediol
Alkylamino acid A
Fatty amine
Hydrogen peroxide
Nitrogen acid
Nonionic surfactant
Phosphoric acid
Silver nitrate
Sodium hydroxide
Sulfuric acid
NA
NA
NA.
NA
NA
NA
NA
NA
NA
NA
3.07E-04
1.71E-04
5.75E-01
1.85E-02
3.95E-03
9.23E-03
2.02E-02
1.51E-04
8.72E-03
7.55E-04
6.48E-06
3.79E-06
1.28E-02
3.91E-04
8.75E-05
2.04E-04
4.26E-04
3.48E-06
1.93E-04
1.59E-05
3-67

-------
3.2 EXPOSURE ASSESSMENT
Chemical
! w
•* "" „/" ?
. J*" * •=.
ADD" '_, -
-> ^"f A ^
, P , (mg/kg-day) " > ป- "
• Inhalation
Line
r '
Operator
Dermal
Line
Operator
Laboratory
Technician
Immersion Tin, Non-conveyorized
Aliphatic acid D
Alkylalkyne diol
Alkylamino acid B
Alkylaryl sulfonate
Alkylimine dialkanol
Alkylphenol ethoxylate
Bismuth compound
Citric acid
Cyclic amide
Ethoxylated alkylphenol
Ethylene glycol monobutyl ether
Fluoboric acid
Hydrochloric acid
Hydroxy carboxylic acid
Methane sulfonic acid
Phosphoric acid
Potassium peroxymonosulfate
Quantenaty alkylammonium chlorides
Silver salt •
Sodium benzene sulfonate
Sodium phosphorus salt
Stannous methane sulfonic acid
Sulfuricacid
Thiourea
Tin chloride
Unspecified tartrate
Urea
Urea compound C
Vinyl polymer
6.14E-02
NA
NA •
5.74E-05
NA
' NA
NA
NA
4.90E-02
NA
3.75E-01
NA
2.03E-03
8.26E-02
NA
1.66E-03
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
5.55E-01
NA
8.22E-03
1.88E-05
1.79E-06
7.88E-07
1.84E-05
2.27E-27
4.02E-05
7.65E-02
1.15E-02
1.80E-27
5.06E-02
1.94E-02
1.13E-02
7.03E-03
1.62E+00
4.75E-02
1.60E-01
7.60E-04
6.03E-06
2.66E-07
1.41E-01
2.18E-02
4.62E-01
1.89E-02
2.19E-02
1.77E-03
3.68E-03
2.37E-02
1.81E-32
9.54E-05
2.19E-07
2.08E-08
9.15E-09
2.13E-07
2.64E-29
4.66E-07
8.88E-04
1.34E-04
2.09E-29
5.87E-04
2.25E-04
1.31E-04
8.16E-05
1.88E-02
5.51E-04
1.85E-03
8.83E-06
7.00E-08
3.08E-09
1.64E-03
2.53E-04
5.37E-03
2.20E-04
2.55E-04
2.06E-05
4.27E-05
2.75E-04
. 2.10E-34
Immersion Tin, Conveyorized
Aliphatic acid D
Alkylalkyne diol
Alkylamino acid B
Alkylaryl sulfonate
NA
NA
NA
NA
1.33E-03
3.17E-06
2.89E-07
1.33E-07
2.32E-04
5.31E-07
5.05E-08
2.22E-08
                                    3-68

-------
                                                                             3.2 EXPOSURE ASSESSMENT
Chemical
* ~ '• ? J- f~*t- J; ^ /•
*v <• ~
t - •! >ซ- -J s, - x
-. ) 1 *
~ - x <• -^ - t T ,
I __ *" ~
rv, •ป * ^ •*
Alkyhmine dialkanol
Alkylphenol ethoxylate
Bismuth compound
Citric acid
Cyclic amide
Ethoxylated alkylphenol
Ethylene glycol monobutyl ether
Fluoboric acid
Hydrochloric acid
Hydroxy carboxylic acid
Methane sulfonic acid
Phosphoric acid
Potassium peroxymonosulfate
Quantenary alkylammonium chlorides
Silver salt
Sodium benzene sulfonate
Sodium phosphorus salt.
Stannous methane sulfonic acid
Sulfuric acid
Thiourea
Tin chloride
Unspecified tartrate
Urea
Urea compound C
Vinyl polymer
„ . ^AD0a ; . '
- " ~ , (mg/kg-day) '" *_-
Inhalation
Line
Operator
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
-Dermal "
Line
Operator
2.98E-06
3.83E-28
6.50E-06
1.24E-02
1.87E-03
3.04E-28
8.52E-03
3.26E-03
1.82E-03
1.14E-03
2.69E-01
8.00E-03
2.69E-02
1.23E-04
9.75E-07
4.48E-08
2.33E-02
3.52E-03
7.69E-02
3.05E-03
3.54E-03
2.86E-04
5.94E-04
3.82E-03
2.92E-33
Laboratory
Technician
5.17E-07
6.41E-29
1.13E-06
2.16E-03
3.25E-04
5.08E-29
1.43E-03
5.46E-04
3.18E-04
1.98E-04
4.56E-02
1.34E-03
4.50E-03
2.14E-05
1.70E-07
7.49E-09
3.98E-03
6.14E-04
1.30E-02
5.33E-04
6.19E-04
4.99E-05
1.04E-04
-6.88E-04
5.09E-34
b Dermal absorption not expected due to large molecular size.
0 LADD is used for calculating cancer risk, and is calculated using a carcinogen averaging time (AT^ of 70 years.
Note: .The numeric format used in these tables is a form of scientific notation, where "E" replaces the
" x 10*". Scientific notation is typically used to present very large or very small numbers. For example, 1.2E-04 is the
same as 1.2 x 10"4, which is the same as 0.00012  in common decimal notation.
d Bath concentration not available.
NA: Not Applicable.  Unless otherwise noted, a number was not calculated because the chemical's vapor pressure is
below the 1 x 10'3 torr cutoff and it is not used in any sparged bath. Inhalation exposures are therefore expected to be
negligible.
ND: Not determined because a required value was not available.
                                                   3-69

-------
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:                 *
where,
PbBa(Mt)Centrai
Pbs
BKSF
AFS
EFS
AT
                 central estimate of adult blood-lead concentrations
                 typical background adult blood-lead concentration
                 lead concentration C"g/g)
                 biokinetic slope factor (//g/dl)
                 intake rate (g/day)
                 gastrointestinal absorption factor (unitless 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:
                          fetat „ 95
                                                     adult X Rfetal/maternal
where,
PbB
PbB
GSD,
'fetal, 0.95

aUult,ccntral

 i, adult
  "clal'nutcm.Tl
95 percent estimate of fetal blood-lead levels (/ig/dl)
central estimate of adult blood-lead concentrations (/ug/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^ ceniia^. This represents the
central estimate of blood-lead hi 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 hi ^g/dl. A value of 1.95 is used, based on a typical range of 1.7 - 2.2 (/ug/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 ^g/g. For PWB facilities, the lead concentration of
 solder was used instead of soil lead concentration. A value of 37,000 /^g/g (37 percent) was used,
 based on typical proportion of lead in tin/lead solder.

        Biokinetic Slope Factor. The biokinetic slope factor (BKSF) relates the increase of
 typical  adult blood-lead concentrations to the average daily lead uptake. The recommended
 default value is 0.4 //g Pb/dl blood per ^g Pb absorbed/day. This value is derived from Pocock et
 al. (1983) and Sherlock et al. (1984) as cited by the U.S. EPA (1996a). (Both studies involved
 the amount of lead in tap water, and both predict higher blood-lead concentrations than expected
 in today's average U.S. population.)                             .

        Intake Rate. The use of this model is based on the assumption that solder could adhere
 to a workers' hands from routine handling, and be subsequently ingested. Although no studies
 were found that address the amount of lead that might be ingested by a worker handling solder
 specifically for a HASL process, Monsalve (1984) investigated hand soldering and pot tinning
 operations.  Based on surface wipe samples and samples from worker's hands, a "conservative
 overestimate" of 30 fj.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 (PbBfetal 0 95). This represents
the 95th percentile estimate of fetal/maternal blood-lead, and is measured in //g/dl.  These results
 are also based on the intake rate, as discussed above.
     Wipe samples from surfaces in the area ranged from 13 to 92 /jg Pb per 100 cm2, and samples from solderer's
hands ranged from 3 to 32 ywg Pb per 100 cm2.                  •

                                          3-71

-------
32 EXPOSURE ASSESSMENT
       Individual Blood Lead Geometric Standard Deviation (GSDj).  The GSD; is used to
measure the inter-individual variability of blood-lead concentrations in a population whose
members are exposed to the same non-residential environmental lead levels. A value of 1.8 is
recommend for homogeneous populations and 2.1 for heterogeneous populations. The values for
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 (Rfeta!/niateniai)'  The
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) arid 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 3-17). Estimated blood-lead levels will be compared to  federal health-based
standards and guidlines in Section 3.4.

  Table 3-17. Estimated Concentration of Lead in Adult and Fetal Blood from Incidental
                           Ingestion of Lead in Tin/Lead Solder
Intake Rate
(ing/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., 1 997).
Adult central estimate for soil ingestion (U.S. EPA, 1997a).
" W>adnlk central
(M2/dl>
2:0
14
63
•PW*fetalv095
<ฃgAdl)
3.2
23
. 102
PbBaduti0= 1.95 ^g/dl; PbS = 37,000 ^g/g; BKSF =
                                               ; AFS = 0.12; EFS= 250 days/yr; AT =365 days/yr;
GSD
    i. adult"
                       = 0.9
       The intake rate is a major source of uncertainty in estimating exposure to workers from
handling solder.  A range of intake rates were used to provide a possible range of modeled blood-
lead concentrations. These values provide bounding estimates only. It is expected that a smaller,
but unknown, amount of solder would be ingested from a workers hands than the estimates that
have been used here. Figure 3-9 shows the relationship between intake rate and blood-lead level
for both an adult and fetus.
                                          3-72

-------
                                                             3.2 EXPOSURE ASSESSMENT
                        Blood lead concentration vs intake rate
                  CO
                                   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 ug/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   =
ATCAR =
ATNC  =
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 (mVday)
exposure frequency (day/yr)
exposure duration (years)
average body weight (kg)
averaging time for carcinogenic effects (days)
averaging time for non-carcinogenic chronic effects (days)
Table 3-18 presents values used for these parameters. Results for general population inhalation
exposure are presented in Table 3-19.
                                         3-73

-------
33 EXPOSURE ASSESSMENT
Table 3-18. Parameter Values for Estimating Nearby Residential Inhalation Exposure
Parameter -
Air Concentration (Ca)
Inhalation Rate (IR)
Exposure Frequency (EF)
Exposure Duration (ED)
Body Weight (BW)
Averaging Time (AT)
ATcAR
ATW
Units
mg/m3
m3/day
days/yr
years
kg
days
Value
Source of Data, Comments
Modeled, varies by chemical and process type.
15
350
30
70
25,550
10,950
Total home exposures for adults based on activity patterns and
inhalation rates (U.S. EPA, 1997a).
Assumes 2 wks per year spent away from home (U.S. EPA,
1991b).
National upper 90th percentile at one residence (U.S. EPA,
1990).
Average value for adults (U.S. EPA, 1991b).
70 yrs x 365 days/year
ED x 365 days/year
Table 3-19. Estimated Average Daily Dose for General Population Inhalation Exposure
Chemical a
ADDtmg/kg-day)*
HASL, Non-conveyorized
Ethylene glycol monobutyl ether
HASL, Conveyorized
Ethylene glycol monobutyl ether
5.25E-05

1.04E-04
Nickel/Gold, Non-conveyorized
Aliphatic acid A
Aliphatic acid E
Inorganic metallic salt A (LADD)
3.45E-05
4.56E-05
5.99E-12
Nickel/Palladium/Gold, Non-conveyorized
Aliphatic acid E
OSP, Non-conveyorized
Acetic acid
OSP, Conveyorized
Acetic acid
Ethylene glycol
Immersion Tin, Non-conveyorized
Urea compound C
Immersion Tin, Conveyorized
Aliphatic acid D
Cyclic amide
Hydroxy carboxylic acid
Urea compound C
6.29E-05

3.33E-05

1.26E-04
2.04E-05

1.11E-04

2.99E-05
2.39E-05
4.03E-05
2.72E-04
* 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 (IEUBK) Model for Lead in Children (U.S. EPA, 1994), to estimate blood-lead
 concentrations in children based on local environmental concentrations (air, soil/dust, drinking
 water, food, etc).  The model includes defaults based on typical concentration levels in an urban
 setting (U.S. EPA, 1994).  The default air concentration used in the IEUBK model is 0.1 ug/m3,
 which is approximately the average 1990 U.S. urban air lead concentration (U.S. EPA, 1991b).'
 This default/background concentration is 1,000 times higher than the ambient air concentration of
 0.00009 ug/m3 estimated from a HASL process (Section 3.2.3). The model was run at various air
 concentrations down to 0.001  ug/m3 (the model does not accept air concentration values less than
 0.001 ug/m3).  At those levels, such small changes to the air concentration result in no real
 difference in estimated blood-lead concentrations compared to results obtained from using the
 default values (i.e., typical urban levels of lead to which a child may be exposed). These results
 are shown in Table 3-20. Since the estimated air concentration of lead from HASL is so far
 below the default/background level in air, and the model could not discern any change in
 children's blood-lead levels from those at average urban air concentrations, it can be concluded
 that general population exposure to airborne lead from the HASL process is negligible.

    Table 3-20. Children's Blood-Lead Results from the IEUBK Model  at Various Lead
                                    Air Concentrations
Age >-
^ (year) t
0.5-1
1-2
2-3
3-4
4-5
5-6
6-7
; Blood-Lead Results (jig/dL) at Various Airborneteaa^Concentrations
- I(jig/m3inair)
4.2
4.7
4.4
.4.2
3.6
3.2
2.9
0.1 Oig/m3inair) -
-, 4.1
4.5
4.2
4.0
3.4
3.0
2.7
0.01Gig/m3Inair)
4.1
4,5
4.2
4.0
3.4
2.9
2.7
0.0tfl
4.1
. 4.5
4.2
4.0
3.4
2.9
2.7
Note: Model default values were used for concentrations in soil/dust, drinking water, and diet.
3.2.5  Uncertainty and Variability

       Because of both the uncertainty inherent in the parameters and assumptions used in
estimating exposure, and the variability that is possible within a population, there is no one
number that can be used to describe exposure.  In addition to data and modeling limitations,
discussed in Sections 3.2.3, sources of uncertainty in assessing exposure include the following:

•      Accuracy of the description of exposure setting:  how well the model facility used in the
       assessment characterizes an actual facility; the likelihood of exposure pathways actually
       occurring (scenario uncertainty).
•      Missing data and limitations of workplace practices data:  this includes possible effects of
       any chemicals that may not have been included (e.g., minor ingredients in the formulations;
                                          3-75

-------
3.2 EXPOSURE ASSESSMENT
       possible effects of side reactions in the baths which were not considered; and questionnaire
       data with limited facility responses).
•      Estimating exposure levels from averaged data and modeling in the absence of measured,
       site-specific data.
•      Data limitations in the Source Release Assessment: releases to land could not be
       characterized quantitatively, as discussed in Section 3.1.
•      Chemical fate and transport model applicability and assumptions:  how well the models
       and assumptions represent the situation being assessed, and the extent to which the models
       have been validated or verified (model uncertainty).
•      Parameter value uncertainty, including measurement error, sampling error, parameter
       variability, and professional judgement.
•      Uncertainty in combining pathways for an exposed individual.

       A method typically used to provide information about the position an exposure estimate
has in the distribution of possible outcomes is the use of exposure (or risk) descriptors.  EPA's
Guidelines for Exposure Assessment (U.S. EPA, 1992b) provides guidance on the use of risk
descriptors, which include the following:

•      High-end:  approximately the 90th percentile of the actual (measured or estimated)
       distribution. This is a plausible estimate of individual risk for those persons at the upper
       end of the exposure distribution, and is not higher than the individual in the population
       who has the highest exposure.
•      Central tendency: either an average estimate (based on average values for the exposure
       parameters) or a median estimate (based on 50th percentile or geometric mean values).
•      What-if:  represents an exposure estimate based on postulated questions (e.g., what if the
       air ventilation rates were ... ), in this case, making assumptions based on limited data so
       that the distribution is unknown.  If any part of the exposure assessment qualifies as a
       "what-if" descriptor, then the entire exposure assessment is considered "what-if."

       This exposure assessment uses whenever possible a combination of central tendency
(either an average or median estimate) and high-end (90th percentile)10 assumptions, as would be
used for an overall high-end exposure estimate.  The 90th percentile is used for:

•      hours per day of workplace exposure;
•      exposure frequency (days per year);
•      exposure duration in years (90th percentile for occupational and 95th percentile for
       residential exposures);
•      time required for chemical bath replacement; and
•      the time and frequency of filter replacements, conveyor equipment cleaning and chemical
       bath sampling (minutes per occurrence and number of occurrences per year).

Average values are used for:
   10
      For exposure data from the PWB Workplace Practices Questionnaire, this means that 90 percent of the facilities
reported a lower value, and ten percent reported a higher value.
                                          _

-------
                                                                 3.2 EXPOSURE ASSESSMENT
 •      body weight;
 •      concentration of chemical in bath;
 •      frequency of chemical bath replacements;
 •      the number of baths in a given process; and                                 ;
 •      bath size.

        However, because some data, especially pertaining to bath concentrations and inhalation
 exposure are limited, and this exposure assessment does not apply to a specific facility, the entire
 exposure assessment should be considered "what-if."                           •   '

 3.2.6  Summary

        This exposure assessment uses a "model facility" approach, with the goal of comparing the
 exposures and health risks of one surface finishing technology to the exposures and risks
 associated with switching to another technology. As much as possible, reasonable and consistent
 assumptions are used across alternatives.  Data to characterize the model facility and exposure
 patterns for each surface finishing technology were aggregated from a number of sources,
 including PWB shops in the U.S., supplier data, and input from PWB manufacturers at project
 meetings.  Thus, the model facility is not entirely representative of any one facility, and actual
 exposure (and risk) could vary substantially, depending on site-specific operating conditions and
 other factors.

        Chemical exposures to PWB workers and the general population from day-to-day surface
 finishing line operations were estimated by combining information gathered from industry (PWB
 Workplace Practices Questionnaire, MSDSs, and other available information) with standard EPA
 exposure assumptions for inhalation rate, surface area of dermal contact, and other parameters.
 The pathways identified for potential exposure from surface finishing process baths were
 inhalation and dermal contact for workers, and inhalation contact only for the general populace
 living near a PWB facility.

        The possible impacts of short-term exposures to high levels of hazardous chemicals
 addressed have not been addressed, such as those that could occur from chemical fires, spills, or
 other episodic releases.

        Inhalation exposure could occur by breathing air containing vapor or aerosol-phase
 chemicals from the surface finishing process line.  Inhalation exposures to workers are estimated
 only for non-conveyorized lines; inhalation exposure to workers from conveyorized surface
 finishing lines was assumed to be much lower because the lines are typically enclosed and vented
 to the outside.11
      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 bafts 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
       The daily intake for inhalation exposure to workers was calculated by first modeling
chemical emissions from surface finishing baths with three air-transport mechanisms: liquid
surface diffusion (desorption), bubble desorption, and aerosol generation and ejection.  These
chemical emission rates were combined with information from the PWB Workplace Practices
Questionnaire regarding process room size and air turnover rate to estimate an average indoor air
concentration for each chemical for the process area.  General room ventilation was assumed,
although the majority of shops have local ventilation on chemical tanks.  An uncertainty and
sensitivity analysis of the air transport models (U.S. EPA, 1998b) suggests that the air turnover
(ventilation) rate assumption greatly influences the estimated air concentration in the process area
because of its large variability.

       Inhalation exposure to the human population surrounding PWB plants was estimated
using the Industrial Source Complex - Long Term (ISCLT) air dispersion model. The modeled
air concentrations of each contaminant were determined at 100 meters radially from a PWB
facility, and the highest estimated air concentration was used. This model estimates air
concentration from the process bath emission rates. These emissions were assumed to  be vented
to the ambient environment at the rate emitted from the baths, for all process alternatives.
Inhalation exposures estimated for the public living 100 meters away from a PWB facility were
very low (approximately 10,000 times lower than occupational exposures).

       Dermal exposure could occur when a worker's skin comes in contact with the bath
solution while dipping boards, adding replacement chemicals, etc.  Although the data suggest that
surface finishing line operators often do wear gloves, it was assumed in this evaluation that
workers do not wear gloves to account for the fraction that do not. Otherwise, dermal exposure
is expected to be negligible. For dermal exposure, the duration of contact for workers  was
obtained from the PWB Workplace Practices Questionnaire information. A permeability
coefficient (rate of penetration through skin) was, estimated for organics, and a default rate
assumption was used for inorganics.  Another source of uncertainty in dermal modeling lies with
the assumed duration of contact.  For non-conveyorized processes, the worker is assumed to have
potential dermal contact for the entire time spent in the surface finishing area, divided equally
among the baths. [This does not mean that a worker has both hands immersed in a bath for that
entire time; but that  the skin is hi contact with bath solution (i.e., the hands may remain wet from
contact).] This assumption may result in an overestimate of dermal exposure.

       Assumptions and parameter values used in these equations are presented throughout this
section.  Exposure estimates are based on a combination of high end (90th percentile)12 and
average values, as would be used for a high-end exposure estimate. The 90th percentile was used
for hours per day of workplace exposure, exposure frequency (days per year), exposure duration
in years (90th percentile for occupational and 95th percentile for residential exposures), and the
time and frequency of chemical bath and filter replacements, conveyor equipment cleaning and
chemical bath sampling (minutes per occurrence and number of occurrences per year) and
estimated workplace air concentrations. The average value was used for body weight,
   12 For exposure data from Hie PWB Workplace Practices Questionnaire, this means that 90 percent of the facilities
reported a lower value, and ten percent reported a higher value.
                                          ______.        _.

-------
                                                              3.2 EXPOSURE ASSESSMENT
concentration of chemical in bath, and the number of baths in a given process. However, because
some data, especially pertaining to bath concentrations and inhalation exposure, are limited and
this exposure assessment does not apply to a specific facility, the entire exposure assessment
should be considered "what-if."

       As a "what if' exposure assessment, this evaluation is useful for comparing alternative
surface finishing processes to the baseline (non-conveyorized HASL) on a consistent basis. It is
also useful for risk screening, especially if actual facility conditions meet those that were assumed
(i.e.,  given similar production rates, what chemicals may be of concern if workers do not wear
gloves; what chemicals may be of concern if ventilation rates are similar to those assumed?).
Finally, this assessment points to the importance of preventing dermal contact by using gloves,
and of proper ventilation.

       Surface water concentrations were estimated for bath constituents, with a focus on those
constituents that are not typically targeted for pre-treatment by PWB facilities. This was done for
conveyorized lines by estimating the amount of chemical going to wastewater from routine bath
replacement, and for non-conveyorized lines by estimating the amount of chemical going to
wastewater from bath replacement plus an estimated amount due to drag-out from the baths to
rinse water.  These amounts were then included in a stream dilution model, and if estimated
surface water concentrations exceeded CCs for aquatic life, the model was refined using estimated
POTW treatment efficiencies.

       These exposure results, taken by themselves, are not very meaningful for evaluating
surface finishing alternatives; it is the combination of hazard (Section 3.3) and exposure that
defines risk.   Quantitative exposure estimates are combined with available hazard data in the risk
characterization (Section 3.4) for risk  screening and comparison of the surface finishing process
configurations.
                                          3-79

-------
3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
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
epidemiologjcal studies. There are a large number of chemicals in commerce, however,
(approximately 15,000 non-polymeric chemicals produced hi 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 hi
       humans; B2 - sufficient evidence of carcinogenicity in animals with inadequate or lack of
       evidence in humans).
•      Group C: Possible Human Carcinogen (limited evidence of carcinogenicity in animals and
       inadequate or lack of human data).
•      Group D: Not Classifiable as to Human Carcinogenicity (inadequate or no evidence).
•      Group E: Evidence of Non-Carcinogenicity for Humans (no  evidence of carcinogenicity
       in adequate studies).

       EPA has proposed a revision of its guidelines that would eliminate the above discrete
categories while providing a more descriptive classification.13

       The International Agency for Research on Cancer (IARC) uses a similar WOE method for
evaluating potential human carcinogenicity based on human data, animal data, and other
supporting data. A summary of the IARC carcinogenicity classification  system includes:
    13
      The "Proposed Guidelines for Carcinogen Risk Assessment" (U.S. EPA, 1996b) proposes the use of WOE
descriptors, such as "Likely" or "Known," "Cannot be determined," and "Not likely," in combination with a hazard
narrative, to characterize a chemical's human carcinogenic potential, rather than the classification system described
above.
                                           3-80

-------
                               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 3-21 lists all surface finishing chemicals that have been classified by
EPA or IARC. The National Toxicology Program (NTP) is an additional source used to classify
chemicals, but its classifications are based only on animal data from NTP studies.
Table 3-21. Available Carcinogenicity Information
Chemical Name "*
•* 4
t " J -
' Cancer Slope
, Factor
(Inhnlation Unit Risk)
' dig/of^
Cancer Slope
-Facto/ '
(Oral)
(m^kg-day)1
Comments/Classification
*. ^ i
".'"'* •"*'.'
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. ฐ
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-evidemce (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 ; (IARC, 1987),
stomach tumors occurred in mice (Ito et
al., 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 a
Silver salt
Stannous methane
sulfonic acid
Tin chloride
Palladium chloride
Propionic acid
Cancer Slope
Factor ~
(Inhalation Unit Risk)
(fig/m3)-1
ND
ND
ND
ND
ND
Cancer Slope
' Factor '
(Oral)
(mg/kg-day)"1
ND
ND
ND
ND
ND
Comments/Classification
' '
* ' < "• '„ป'
, t -i
Not classifiable according to EPA
and/or IARC.C
EPA Class D > (U.S. EPA, 1987a).
EPA Class D h or IARC Group 31 (U.S.
EPA, 1987a).
No classification; mice administered
palladium in drinking water had a
significantly higher incidence of
malignant tumors (Schroeder and
Mitchener, 1971).
No classification; tumors in
forestomach of rats (Clayson et al.,
1991).
• 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.
* IARC Group 1: Human Carcinogen.
f EPA Class B2: Probable Human Carcinogen (sufficient evidence of carcinogenicily in animals with inadequate or
lack of evidence in humans).
s IARC Group 2B: Possibly carcinogenic to humans.
h EPA Class D: Not classifiable as to human carcinogenicity.
' 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 (q!*) 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, q,* is an approximation of the
upper bound of the slope of the dose-response curve using the linearized, multistage procedure at
low doses. 'TJnit risk" is an equivalent measure of potency for air or drinking water
concentrations and is expressed as the upper bound excess lifetime cancer risk per //g/m3 in air, or
as risk per fj.g/L in water, for continuous lifetime exposures. (Unit risk is simply a transformation
of slope factor into the appropriate scale.) Slope factors and unit risks can be viewed as
quantitatively derived judgements of the magnitude of carcinogenic effect.  These estimates will
continue to be used whether the current EPA WOE guidelines are retained or the new proposals
are adopted. Their derivation, however, may change for future evaluations.
                                             3-82

-------
                                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
 PRISJ), together with uncertainty factors regarding their applicability to human populations.
 Table 3-22 presents a summary of the available RfC and RfD information obtained from IRIS and
EPA's Health Effects Assessment Summary Tables (HEAST) for non-proprietary chemicals  An
 additional proprietary chemical has an RfC and an RfD; these data are not reported in order to
protect the identity of the confidential ingredient.
                                         3-83

-------
33 HUMAN HEALTH AMP ECOLOGICAL HAZARDS SUMMARY
  Table 3-22. Summary of RfC and RfD Information Used in Risk Characterization for
Chemical
Name*
Ammonium
chloride,
Ammonium
iiydroxide-
Ethylenediamine
Ethylene glycol
Ethylene 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, r
(Inhalation)
* c
Ammonia: decreased
lung function (IRIS,
1999).


Changes in red blood
cell count (IRIS,
1999).
Rats, hyperplasia of
nasal mucosa, larynx,
and trachea (DRIS,
1998).
Oral/Dermal
RfD*
(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).
0.00053s (MRL)
0.01 (IRIS)
ND
ND
Rats, lung
inflammation
(ATSDR, 1997a).
Rats, histologic
lesions hi
trachebbronchiolar
region (IRIS, 1998).


0.02 (IRIS)
(soluble salts
of nickel)
221 (ADI)
0.02 h (IRIS)
0.005 l (IRIS)
Rats, decreased body and
organ weight (IRIS, 1998).
(US. 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

-------
                                     33 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
Chemical
Name a _-
* >
•p '•
Stannous methane
sulfonic acid,
Tin, and
Tin chloride
Sulfunc acid
Inhalation
RfCb
(mgft?)
ND
0.07 (HEAST)
'^Comments c
(Inhalation)
• *v- - -
J-

Acceptable air
concentration for
humans based on
respiratory effects
(U.S.EPA,1997b).
Oral/Dermal
RfDi
T- . ,'
(mg/fcg/day)
0.6 J (HEAST)
NDk
Comments0
- (Oral/Dermal)
^ ~5 v.
r ff ~"
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 ESC 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.
 e Comments may include exposure route, test animal, duration of test, effects, and source of data.
 d In the risk calculations, conversion factors are used based on the molecular weights of ammonia, ammonium chloride
 and ammonium hydroxide.                                                                           '
 e In the risk calculations, conversion factors are used based on the molecular weights of ammonium sulfamate,
 ammonium chloride, and ammonium hydroxide.
 f More information on lead is presented in Section 3.4.6 of the Risk Characterization.
 * Value given represents a chronic inhalation minimum risk level (MRL). Although the test substance was nickel
 sulfate hexahydrate, the reported value is 0.0002 mg/m3 as nickel. This was converted in the risk calculations based on
 the molecular-weights  of nickel and nickel sulfate.
 h A conversion factor is used in the risk calculations based  on molecular weights of cyanide and potassium gold
 cyanide.  This RfD is only relevant to the oral route; potassium gold cyanide is expected to be chemically stable except
 under highly acidic conditions such as those found in the stomach (pH 1-2).
 | A conversion factor is used in the risk calculations based on molecular weights of silver and silver nitrate
 > Conversion factors are used in the risk calculations based  on molecular weights of tin, tin chloride and stannous
 methane sulfonic acid.
 k Although chronic toxicity values have not been established, repeated skin contact with low concentrations of sulfuric
 acid causes skin dessication, ulceration of the hands, and chronic inflammation around the nails
 ND: No data, RfC or RfD not available.
        When an RfD or RfC was not available for a chemical, other toxicity values were used
preferably in the form of a "no-observed-adverse-effect level" (NOAEL) or "lowest-observed-
adverse-effect level" (LOAEL). These toxicity values were obtained from the published scientific
literature, as well as unpublished data submitted to EPA on chemical toxicity in chronic or
subchronic studies.  Typically, the lowest NOAEL or LOAEL value from a well-conducted study
was used. (If study details were not presented or the study did not appear to be valid, the
reported NOAEL/LOAELs were not used.) But, unlike the majority of RfD/RfCs,
NOAEL/LOAELs have not received EPA peer-review of the studies on which the'values are
based, and uncertainty factors have not been considered.
                                               3-85

-------
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 hi 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 hi the frequency or severity of adverse effects in the exposed population over
its appropriate control (hi mg/kg-day, or mg/m3 for inhalation). LOAEL values are presented
only where NOAELs were not available. Table 3-23 presents a summary of the available NOAEL
and LOAEL values for non-proprietary chemicals. Chemicals having potential developmental
toxicity were identified based on the data provided in the toxicity profiles. These data are
summarized in Table 3-24. An additional 5 proprietary chemicals have inhalation NOAELs or
LOAELs, and 13 have oral NOAELs or LOAELs; these data are not reported in order to protect
the identity of confidential ingredients.

       Neither RfDs/RfCs nor LOAELs/NOAELs were available for some chemicals in each
surface finishing process alternative. For these, chemicals, no quantitative estimate of risk could
be calculated. EPA's Structure-Activity Team (SAT)14 has reviewed the chemicals without
relevant toxicity data to determine if these chemicals are expected to present a toxicity hazard.
This review was based on available toxicity data on structural analogues of the chemicals, expert
judgement, and known toxicity of certain chemical classes and/or moieties. Chemicals received a
concern level rank of high, moderate-high, moderate, moderate-low, or low. Results of the SAT
evaluation are presented in Table 3-25. A summary of toxicity data available for the chemicals is
presented in Table 3-26.
    14 The SAT is a group of expert scientists at EPA who evaluate the potential health and environmental hazards of
new and existing chemicals.
                                          —

-------
                                      3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
   Table 3-23. NOAEL/LOAEL Values Used in Risk Characterization for Non-Proprietary
Chemical _
JVamea"
- x *'
Acetic acid
Copper ion,
Copper sulfate
pentahydrate
Ethylenediamine
Etiiylene glycol
Hydrogen peroxide
Lead '
Propionic acid
Inhalation
NOAEL/
LOAELb
(mg/m3)
NDd
0.6 (L) e
145 (N)s
31 (L)
79(L)h
10 ug/dL
in blood
23(TCIo)j
Comments c "
(Inhalation)
„ *v
/ *

Cupric chloride: rabbits, 6
hrs/day, 5 days/wk for 4-6 wks,
increase in lung tumors (U.S. Air
Force, 1990).
Rats, 7 hrs/day, 5 days/wk for 30
days, depilation (Pozzani and
Carpenter, 1954).
Humans, 20-22 hrs/day for 30
days, respiratory irritation,
headache, and backache
(ATSDR, 1997b).
Mouse, 6 weeks, 7/9 died (U.S.
EPA, 1988a).
Children, level concern in blood
(CDC, 1991).
Rats, subchronic exposure
(RTECS, 1998).
Oral/Dermal
NOAEL/
1LQAEL1*
(mg/kg-day)
195 (N)
0.056 (L) f
. NA
NA
290 (L)
10 ug/dL
in blood
150 (N)
„ t Comments'
(Oral/Dermal)
- v J * ซ "•
^ ft
Rats, drinking water, 2-4 months,
no deaths (Sollmann, 1921).
Copper: humans, 1.5 years,
abdominal pain and vomiting
(ATSDR, 1990a).
RfD is available (Table 3-22).
RfD is available (Table 3-22).
Mice, 35 weeks, liver, kidney,
and GI effects (IARC, 1985).
Children, level concern in blood
(CDC, 1991).
Rats, diet, lesions in GI tract
(BASF, 1987; Mori, 1953;
Harrison et al., 1991; Rodrigues
etal., 1986).
 b (N) - NOAEL; (L) = LOAEL.  When more than one NOAEL and/or LOAEL was available, only the lowest available
 NOAEL or LOAEL was used and is listed here.  If both NOAEL and LOAEL data are available, the NOAEL is used
 and is listed here. If a chronic NOAEL or LOAEL was not available, other values (e.g., from shorter-term studies) were
 used as noted.
 c Comments may include exposure route, test animal, duration of test, effects, and source of data.        '
 d Although health effects have been noted in workers and laboratory tests from inhalation exposure to acetic acid, no
 appropriate chronic inhalation toxicity value is available.
 e Conversion factors are used in the risk calculations based on molecular weights of cupric chloride, copper ion, and
 copper sulfate pentahydrate.
 f A conversion factor is used in the risk calculations based on molecular weights of copper and copper sulfate
 pentahydrate.
 s Not considered a "chronic" value because the study duration was less than 90 days. The value was used, however, as
 the best available value, rather than leaving a data gap for a chemical where adverse health effects have been noted
 h In the absence of other data, this value will be used as a LOAEL.
 ' More information on lead is presented in Section 3.4.5 of the Risk Characterization.
3 TClo = The lowest dose of a chemical that is expected to cause a defined toxic effect. In the absence of other data, this
 is used as a LOAEL.
ND: No Data.  A NOAEL or LOAEL was not available for this chemical.
NA: Not applicable. A NOAEL or LOAEL is not required because an RfC or RfD is available for this chemical.
                                                 3-87

-------
33 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
    Table 3-24. Developmental Toxicity Values Used in Risk Characterization for Non-
                                    Proprietary Ingredients
Chemical?1
Ammonium
chloride
Copper ion,
Copper sulfate
pentahydrate
Ethylenediamine
Ethylene glycol
Ethylene glycol
monobutyl ether
Developmental
Inhalation
NOAEL/
iOAEL
(mg/m3)" "
ND
ND
ND
150 (N)
ND
* Comments c
J (Inhalation)
i i
ซt ; J
tj^-
1 - J^- M* ,
^,*\. ^



Rats and mice, 6 hr/day, gd 6-
15, fetal malformation's in
mice (exencephaly, cleft
palate, and abnormal rib and
facial bones) (Shell Oil, 1992;
Union Carbide, 1991).

Developmental
Oral/Dermal
' NOAEL/
XOAEX*
(mg/kg-day) _
1,691 (N)
3(L)e
470 (L)
500 (N)
100 (N)
Comments c
(Oral/Dermal)
f ^ > < ;
, < : ',- ' •
Mice, drinking water, after gd d
7, no congenital effects
(Shepard, 1986).
Copper: mink, diet, increased
mortality (Aulerich et al.,
1982; ATSDR, 1990a).
Rats, gd 6-15 diet, resorption,
impaired growth, missing or
shortened innominate arteries,
and delayed ossification of
cervical vertebrae or phalanges
(DePassetal., 1987).
Rats, gd 6-15, gavage,
teratogenic effects at higher
dose levels. NO AEL based on
developmental effects (Bushy
Run, 1995).
Rats, gd 9-11, oral gavage,
developmental toxicity (Sleet
etal., 1989).
1 Only those chemicals with available data are listed.
b (N) - NO AEL; (L) = LOAEL. When more than one NO AEL and/or LO AEL was available, only the lowest available
NOAEL or LOAEL was used and is listed here.  If both NO AEL and LOAEL data are available, the NO AEL 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.
ฐ Conversion factors are in the risk calculations based on molecular weights of copper ion and copper sulfate
pentahydrate.
ND: No data available.
                                               3-88

-------
                                  33 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
                     Table 3-25.  Summary of Health Effects Information
                    	(from Structure-Activity Team Reports)
     Chemical
              SAT Health Effects Pertaining to  T
               <    '    *           *  "<-? ti  ฐ     <~~ ~
               Dermal or Inhalation Exposure
                                                                               Overall Concern,
                                                                                    XeveJ
 1,4-Butenediol
 Expect good absorption via all routes of exposure. The primary
 alcohols will oxidize to the corresponding acids (fumaric or
 maleic) via aldehydes. There is concern for mutagenicity as an
 unsaturated aldehyde. This compound is  expected to be irritating
 to the lungs and other mucous membranes. Effects on the liver
 and kidney and neurotoxicity (sedation) are also expected.
 Low moderate
Aliphatic acid B
 Expect no absorption by skin, but expect absorption by lungs
 and GI tract. Related compound is reported to be positive in a
 dominant lethal assay. Uncertain concerns for developmental
 toxicity and kidney toxicity. Some concern for irritation.
Moderate
Aliphatic
dicarboxylic acid A
 Absorption is expected to be poor through the skin and good
 through the lungs and GI tract. As a free acid, this compound is
 expected to be irritating to all exposed tissues. A mixture of
 acids containing this compound was tested in rats.  The mixture
 was negative for mutagenicity but caused signs of neurotoxicity.
 A mixture containing the dimethyl ester of this compound was
 tested in acute inhalation and dermal studies because blurring of
 vision had been reported in humans. An increase in the anterior
 chamber depth in the eye was seen following inhalation and
 dermal exposure. This could be an indication of changes in
 circulation in the eye which could lead to glaucoma. A mixture
 of the same compounds was tested in a 1-generation
 reproduction study in rats via inhalation, showing a decrease in
 postnatal pup weight and irritation of the respiratory tract in
 parental animals.
Low moderate
Alkylalkyne diol
Expect poor absorption via all routes of exposure. This
compound may be irritating to the eyes, lungs, and mucous
membranes and cause defatting of the skin which can lead to skin
irritation. There is uncertain concern for neurotoxicity and liver
and kidney effects.
Low
Alkylamirio acid A
Absorption is expected to be poor through the skin and good
through the lungs and GI tract This compound is expected to
chelate metals such as calcium, magnesium, and zinc. Based on
rts potential to chelate calcium, there is concern for
developmental toxicity, inhibition of blood clotting, and effects
on the nervous system and muscles including effects on the heart.
Chelation of zinc may cause immunotoxicity (retardation of
wound healing). This compound is expected to be irritating to all
exposed tissues and may be a dermal sensitizer. A salt of this ,
compound caused developmental effects in rats. There is
concern for oncogenicity and kidney toxicity.  There is also a
potential for male reproductive effects. This compound may be
mutagenic.
                                                                               Low moderate
                                             3-89

-------
33  HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
     Chemical
            -SAT Health Effects Pertaining to
             ' Dermal or Inhalation Exposure
Overall Concern
     Level
 Alkylaryl imidazole
Expect good absorption via the lungs and GI tract. Absorption
of the neat material is expected to be nil through the skin;
however, absorption is expected to be moderate through the skin
when in solution. There is concern for developmental toxicity
and neurotoxicity.
Low moderate
 Alkylaryl sulfonate
Absorption is expected to be nil through the skin and poor
through the lungs and GI tract. There is uncertain concern for
irritation to mucous membranes.
Low
 Alkylimine
 dialkanol
Absorption is expected to be poor through the skin, moderate
through the GI tract, and good through the lungs. This
compound is a moderate to severe skin irritation and a severe eye
irritant. It has low acute toxicity. Another analog was tested in a
subchronic gavage study in rats and dogs. Cataracts were noted
hi rats, stomach and lung lesions consistent with irritation were
seen, and liver effects were seen in female dogs. There is
concern for developmental toxicity. There is little concern for
mutagenicity by analogy to a similar compound.
Moderate
 Amino acid salt
Absorption is expected to be nil through the skin and good
through the lungs and GI tract. There is uncertain concern for
developmental toxicity.  This compound is an amino acid analog
and may be an antimetabolite. This chemical is also expected to
be an irritant to moist tissues such as the lungs and respiratory
tract.
Low moderate
 Ammonia
 compounds
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, hi water, this compound will
cause irritation of all moist tissues. There is also concern for
neurotoxicity and possibly developmental toxicity. There is no
concern for mutagenicity based on negative results for DNA
damage. This compound has a relatively high oral LD50.a
Moderate, based
on irritation
 Citric acid
Expect poor absorption by skin, but expect absorption by lungs
and GI tract. No health concerns identified.
Low
                                              3-90

-------
                                   3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
     Chemical
              SATHealth EffectsJPertaining to
               Dermal or Inhalation Exposure
 Overall Concern
      Level
 Ethoxylated
 alkylphenol
 Absorption is expected to be poor through the skin, moderate
 through the GI tract, and good through the lungs. As a
 surfactant, this compound may cause lung effects if inhaled.
 This compound is expected to be a severe and persistent eye
 irritant. Eye irritation is of particular concern because this type
 of compound can anesthetize the eye so an individual will not
 feel pain and rinse the material out of the eye. It is also expected
 to be irritating to the lungs. Possible signs of lung irritation
 (lung discoloration) were noted with a similar chemical tested in
 an acute inhalation study in rats. There is uncertain concern for
 reproductive effects and immunotoxicity. By analogy to a related
 compound, this chemical may be an endocrine disrupter. Liver
 and kidney effects were noted in rats with a structural analog.
 Myocardial degeneration has also been noted in several species.
 with related compounds.  Developmental toxicity as
 demonstrated by skeletal changes has been noted with dermal
 and oral exposure.
Low moderate
Fatty amine
 Absorption is expected to be poor through the skin, moderate
 through the GI tract, and good through the lungs. This
 compound is expected to be a strong irritant and/or corrosive to
 exposed tissues. A similar compound was reported to be a
 moderate skin irritant and a severe eye irritant. Oleyl amine is a
 severe irritant. There is also concern for lung effects if inhaled.
 Another analog was tested in a subchronic gavage study in rats
 and dogs. Cataracts were noted in rats, stomach and lung lesions
 consistent with irritation were seen, and liver effects were seen in
 female dogs.  There is concern for developmental toxicity.  There
 is little concern for mutagenicity by analogy to a similar
 compound.
Moderate
Hydroxyaryl acid
Absorption is expected to be poor through the skin and good
through the lungs and GI tract. There is concern for
developmental toxicity and uncertain concern for effects on
blood clotting (slower time for clotting). This compound is
expected to have estrogenic activity. It has low acute toxicity. It
may also cause neurotoxicity and hypersensitivity.  There is some
concern for mutagenicity.
Moderate
Hydroxyaryl
sulfonate
Absorption is expected to be nil through the skin and good
through the lungs and GI tract. There is concern for
developmental toxicity. This compound is also expected to be an
irritant (the free acid is corrosive to the eyes) and may cause
neurotoxicity.
Low moderate
Maleic acid
Expect no absorption by skin, but expect absorption by lungs
and GI tract. Maleic acid is reported to be negative in a NTP
Ames assay. According to Merck this chemical is strongly
irritating to corrosive.
                                                                                 Vtoderate
                                              3-91

-------
3.3  HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
     Chemical
             SAT Health Effects Pertaining to
           \  Dermal orlnhalation Exposure
Qverall Concern
 "r  Level
 Malic acid
Expect no absorption by skin, but expect absorption by lungs
and GI tract. Concerns for mild irritation to skin and eyes.
Low moderate
 Potassium
 compound
Absorption/corrosion by all routes. Concentrated form is
corrosive to all tissues. Dilute form may be irritating. No other
health concerns identified.
High for
concentrated form
only, otherwise
low
 Potassium
 peroxymonosulfate
Absorption is expected to be nil through the skin and good
through the lungs and GI tract. The peroxymonosulfate moiety
is reactive with moisture (oxidizing agent). This material will be
an irritant as a concentrated solution.
Moderate
 Quaternary
 alkylammonium
 chlorides
Absorption is expected to be poor through the skin, moderate
through the GI tract, and good through the lungs.  This chemical
is expected to be a strong irritant and/or corrosive to all exposed
tissues. It is also expected to be neurotoxic. There is also
concern for lung effects if inhaled.  There is concern for
developmental toxicity as an ethanolamine derivative. This
compound is expected to be in the moderately toxic range for
acute toxicity.
Moderate
 Sodium benzene
 sulfonate
Absorption is expected to be nil through the skin and good
through the lungs and GI tract.  There is concern for
methemoglobinemia, neurotoxicity, and developmental toxicity.
Serious brain damage was noted in a 2-week inhalation study
with a related compound.  There is uncertain concern for
oncogenicity. This compound is reported to be negative in the
Ames assay. It is expected to be irritating to mucous membranes
and the upper respiratory tract.
Moderate concern
 Sodium
 hypophosphite;
 Sodium
 hypophosphite
 monohydrate
Absorption is expected to be nil through the skin and good
through the lungs and GI tract. This compound has low acute
toxicity. It is irritating to mucous membranes and may cause
dermal sensitization.  There is uncertain concern for
mutagenicity. It is reported to be effective in inhibiting the
growth of selected Gram-positive pathogenic bacteria.
Low moderate
concern
 Substituted amine
 hydrochloride
Absorption is expected to be nil through the skin and good
through the lungs and GI tract. This chemical has fairly high
acute toxicity. It is a severe skin irritant in guinea pigs and a
weak to moderate dermal sensitizer. In a repeated dose dietary
study in rats, the primary effects were on the red blood cells
(through methemoglobin production) and the  spleen. This
compound is reported to be positive in a variety of mutagenicity
assays, although there are also some negative  responses. There
is concern for oncogenicity based on the mutagenicity results.
There is uncertain concern for developmental  toxicity.
Moderate concern
                                              3-92

-------
3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
' Chemical
Transition metal
salt
SAT HealtkEffects Pertaining to
4 " Dermal or Inhalation Exposure
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.
k LLou: i^etbal dose to 50 percent of the test population.
, QyeraB Concern.
* Level
Moderate concern

Table 3-26. Overview of Available Toxicitv Data
Chemical -
s
c*
t <~ * -.
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
Alkyhmine dialkanol
Alkylphenol ethoxylate
Alkylphenol polyethoxyethanol
Alkylpolyol
Amino acid salt
Amino carboxylic acid
Ammonium chloride
Ammonia compound A
Ammonia compound B
Ammonium hydroxide
Cancer:
SIopeFactor (SF),
Weight-of-Evidence
(WOE)
Classification
























Inhalation:
BfQNOAEL,
orLOAEL*




Yes








Yes






RfC (for ammonia)
IfC (for ammonia)
*fC (for ammonia)
RfC (for ammonia)
Oral/Dermab
RiD,NOAEL3
, orlsQAEL*
<• \

NOAEL
Yes

Yes


Yes





Yes



Yes

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


•


•

•
•

•
•

•
•
•

•



•

         3-93

-------
3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
Chemical
•>
_ i t
Aromatic imidizole product
Arylphenol
Bismuth compound
Citric acid b
Copper ion
Copper salt C
Copper sulfate pentahydrate
Cyclic amide
Ethoxylated alkylphenol
Ethylenediamine
Ethylene glycol
Sthylene 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 ฐ
Methane sulfonic acid
Nickel sulfate
Nitrogen acid
Nonionic surfactant
Palladium chloride
Palladium salt
Phosphoric acid
Potassium compound
Cancer:^ f
Slope Factor (SF),
Weight-of-Evidence
(WOE) '
Classification
Inhalation:
RfC,NOAEL,^
- orLOAEL* ^
Oral/Dermal:
RfD, NOAEL, ฐ
or LOAEL a
r
^ A
ซ
Not enough information to identify a specific chemical.



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








WOE
WOE



SF,WOE


WOE



WOE (for nickel
dust)




LOAEL
Yes
LOAEL
Yes

NOAEL
LOAEL; D-
NOAEL
RfC



RfC
Other b
Yes


Yes
Yes
Yes
Other"



MRLd

Yes


LOAEL; D-LOAEL
Yes; D-LOAEL
LOAEL; D-LOAEL
Yes

RfD; D-LOAEL
RfD; D-NOAEL
RfD; D-NOAEL


Yes

LOAEL
Yes


Yes
Yes
Yes
Other"



RfD


-------
                                     3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
Chemical
f1 ^ ""
* „ x i
y -•„ „ <
~ , ซ•"*
^_ **
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 8
Stannous methane sulfonic acid
Substituted amine
hydrochloride
Sulfuric acid
Surfactant
Thiourea
Tin
Tin chloride
Transition metal salt
Unspecified tartrate
Urea
Urea compound B
Urea compound C
Vinyl polymer
Cancer: s
Slope Factor (SF),
Weight-of-EvTdence
(WOE)
Classification


Some data

WOE (for silver)
WOE (for silver)






WOE

WOE
* Inhalation:
RfC, NOAEL,
or LOAEL3
•^


Other0











Other0
OraWDermal:
RfD, NOAEL,
of LOAEL a
RfDf

NOAEL

Yes
RfD (for silver)






RfD(fortin)


Not enough information to identify specific chemical.
WOE

WOE



WOE

WOE










RfD
RfD

Yes


Yes
Yes
SAT
Rank

•

•


•

•
•
•


•





•





  'Yes" indicates a value is available (RfC or RfD, NOAEL or LOAEL) but the type of toxieity measure is not
specified in order to protect confidential ingredient identity. D-NOAEL/or D-LOAEL:  Developmental NOAEL of
LOAEL available.
b Toxieity data other than RfD, NOAEL or LOAEL were used; see Tables 3-22 and 3-23 for details.
c Generally recognized as safe (GRAS) by the U.S. Food & Drug Administration (HSDB, 1995).
d MRL = minimal risk level.
e ADI = allowable daily intake.
f These values are only relevant to the oral route; potassium gold cyanide is expected to be chemically stable except
under highly acidic conditions such as those found in the stomach (pH 1-2).
8 Not generally considered poisonous to humans or animals.
                                                 3-95

-------
33 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
3.3.3  Ecological Hazard Summary

       Ecological hazards data are presented in two ways: through a CC and an aquatic hazard
concern level, each derived separately from aquatic toxicity data (fish, invertebrates, and algae).
Hazards to terrestrial species were not assessed because sufficient toxicity data were not available.
CCs are based on the most sensitive endpoint, modified by an assessment factor, which reflects
the amount and quality of toxicity data available for that chemical.  CCs are compared to
estimated surface water concentrations as part of the Risk Characterization (Section 3.4).
Aquatic hazard concern levels are based on where the lowest available toxicity value (i.e., the
most sensitive endpoint) fits into pre-defined ranges of values, indicating relative toxicity when
compared to other chemicals.

Concern Concentration

       Table 3-27 presents a summary of the available ecological hazards information. CCs were
determined for aquatic species (e.g., Daphnia, algae, and/or fish) using standard EPA
methodology.  The method for determining CCs is summarized below and presented in more
detail in Appendix H.

 Table 3-27. Estimated (Lowest) Aquatic Toxicity Values and Concern Concentrations for
     PWB Surface Finishing Chemicals, Based on Measured Test Data or SAR Analysis
Chemical
-*• •** i.
1,4-Butenediol
Acetic acid
Aliphatic acid A
Aliphatic acid B
Aliphatic acid D
Aliphatic acid E
Aliphatic dicarboxylic acid A
Aliphatic dicarboxylic acid C
Alkylalkyne diol
Alkylamino acid A
Alkylamino acid B
Alkylaryl imidazole
Alkylaryl sulfonate
Alkyldiol
Alkylimine dialkanol
Alkylphenol ethoxylate
Alkylphenol polyethoxyethanol
Alkylpolyol
Acute (a) Toxicity
""•",. (mg/L)
Fish Invert Algae
0.5
79 65
Chronic (c) Toxichy ,
t (mg/L) f :
Fish
0.08

data omitted 3
Invert



Algae



data omitted a
data omitted a
data omitted"
data omitted a
data omitted a
data omitted a
data omitted3
data omitted a
data omitted a
data omitted3
data omitted3
data omitted a
data omitted3
16 16 20
2
data omitted3
2

5

Concern
Concentration
> Cmg/I>)
0.008 (c)
0.65 (a)
0.5 - 1 (a)
.l-5(c)
5 - 10 (c)
>l(c)
>l(c)
>10
0.1-0.5(c)
500 - 1,000 (c)
0.1 - 5 (c)
0.001 - 0.005 (c)
0.001- 0.005 (c)
10 - 50 (c)
0.001 - 0.005 (c)
0.1 - 0.5 (c)
0.2 (c)
5 - 10 (c)
                                          3-96

-------
3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
Chemical
^ i f-
A. ^ j •ฃ- ^ ^
' -f - "
^ ,_% "* V -ฃ5-
V
Amino acid salt
Amino carboxylic acid
Ammonia compound A
Ammonia compound B
Ammonium chloride
Ammonium hydroxide
Arylphenol
Bismuth compound
Citric acid. In soft water
In hard water
Copper ion
Copper salt C
Copper sulfate pentahydrate
Cyclic amide
Ethoxylated alkylphenol
Ethylenediamine
Ethylene glycol
Ethylene glycol monobutyl
ether b
Fatty amine
Fluoboric acid
Gum
Hydrochloric acid
Hydrogen peroxide
Hydroxyaryl acid
Hydroxyaryl sulfonate
Hydroxy carboxylic acid
Inorganic metallic salt A
Inorganic metallic salt B
Inorganic metallic salt C
Lead
Maleic acid
Malic acid
Methane sulfenic acid
Nickel suhate
Nitrogen acid
Palladium chloride
Acute (a) Toxicity
(mg/L) .;
Fish
* '>
Invert
Algae
ChrooieXc) Toxicny
' (mgflL)
Fish
Invert
Algae
dataomitteda
data omitted a
data omitted"
data omitted a
725
12
161
32

>30

1

3

>3
data omitted1
data omitted a
>100
0.14
>100
12.8
5
100

>10

>10

1
30

data omitted a
0.34
0.3
0.00002
0.022
0.0014
0.062
data omitted a
data omitted a
220
10,000
116
26.5
6,900
89
>100
31,000
620

5,400
10
0.16
710
3.9
8.3
440
32
data omitted a
>1,000
560
160
20
70
1.4
data omitted a
70
5.9
100
4.3
345
1.7
63

16

15

data omitted a
data omitted a
data omitted a
data omitted a
data omitted a
data omitted a
315
5,227
2,860
g/L
>1,000
1.28
143
1,199
2,380
g/L
>1,000
2.58
500
30,654
1,200
g/L
>1,000
1.9
4.1

204,000
>100

30

24,378
>100


993
14,339
>100

data omitted a
1,584
1,567
917
170
49
47
Concern
Concentration
(mg/L) ,
S !,
0.5 - 1 (c)
5 - 10 (c)
1-5 (a)
0.0 1- 0.05 (c)
1.6 (a)
0.1 (c)
0.01 - 0.05 (c)
0.1 - 0.5 (c)
0.1 (c)
3.0 (c)
0.001 (a)
0.005 - 0.01(c)
0.01 (c)
10-50(c)
0.1-0.5(c)
0.02 (c)
44 (c)
0.04 (c)
0.001 - 0.005 (c)
0.14 (c)
0.5 - 1 (c)
1.5 (c)
0.02 (a)
. 0.1- 0.5 (c)
l-5(c)
- 1 - 5 (c)
0.0001-0.0005 (c)
0.001 - 0.005 (c)
0.001 - 0.005 (c)
0.41 (c)
99.3 (c)
1,434 (c)
10 (c)
0.01 (a)
1 - 5 (c)
4.7 (c)
         3-97

-------
33 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
Chemical" ~
•f
%
>
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
SuUEuric acid
Thiourea
Tin
Tin chloride
Transition metal salt
Unspecified tartrate
Urea
Urea compound B
Urea compound C
Vinyl polymer
Acute (a) Tosicity
,, 0.6
<1
1,369
>2
0.4
<3
6,644
>0.06
<0.1
1,216
X).03
<0.3
318
>0.1
<1
292
data omitted1
0.007
0.0007
0.13
0.001
0.005

data omitted a
data omitted a
133,000
199,000
g/L
191,000
g/L
1,330
g/L
3,180
g/L
55,700
g/L
498,000
8,430
g/L
22,658
331,000
10,616
103,000
data omitted a
data omitted a
7
140
<8
0.2
0.9
<0.8
data omitted a
42
>100
2.7
1.89
5,200
g/L
9
55
19.5
250,000
4.8
<3
0.2
600,000
>60
0.07
0.4
4,222
0.9
0.35
42
2,241
0.3
<0.3

data omitted a
data omitted a
>1,000
>1,000
>1,000
>100
>100
>100
data omitted a
data omitted a
data omitted a
- Concern
Concentration
(mg/L)
l-5(c)
27.8 (c)
1,000 - 1,500 (c)
0.003 (c)
0.01 (c)
29.2 (c)
0.0 1- 0.05 (c)
0.0001 (c)
0.0001- 0.0005 (c)
>1 (c)
1,062 (c)
10,300 (c)
10,000 - 50,000 (c)
50 - 100 (c)
0.02 (c)
0.0 1- 0.05 (c)
224 (c)
0.03(c)
0.007 (c)
0.04 (c)
,<-5(c)
l-5(c)
>10 (c)
0.01 - 0.05 (c)
0.0 1- 0.05 (c)
>l-5
a Data omitted from table and a range reported for CC in order to protect identity of confidential ingredients.
b Diethylene glycol monobutyl ether reviewed instead; both chemicals are very similar.
        The CC for each chemical in water was calculated using the general equation:

                           CC  = acute or chronic toxicity value •*- UF
                                              3-98

-------
                                3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
where,
CC    =
UF    =
              aquatic toxicity concern concentration, the concentration of a chemical in the
              aquatic environment below which no significant risk to aquatic organisms is
              expected
              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.

       The following approach was used, depending on availability and type of data:

 •      If the toxicity profile only contained one or two acute toxicity values (no chronic values),
       UF = 1,000 and the CC was calculated by using the lower acute value.
 •      If the toxicity profile contained three or more acute values (no chronic values), UF = 100
       and the CC was calculated by using the lowest acute value.
 •      If the toxicity profile contained at least one chronic value, and the value was for the most
       sensitive species, UF = 10 and the CC Was calculated by using the lowest chronic value;
       otherwise, UF = 100 and the CC was calculated with the acute value for the most sensitive.
       species.

 Hazard Concern Levels

       Table 3-28 presents aquatic hazard concern levels; chemicals were assigned to aquatic
toxicity concern levels according to the following EPA criteria:

For chronic values:
       < 0.1 mg/L	High concern
       > 0.1 to < 10 mg/L	Moderate concern
       >10mg/L	Low concern      .

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

Chronic toxicity ranking takes precedence over the acute ranking.
                                         3-99

-------
33 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
       Most surface finishing chemicals can theoretically be subject to spills and releases. Also,
PWB facilities routinely release wastewater to POTWs. Different geographic regions and
different POTWs have different levels of acceptability for such wastes, and the acceptable levels
can change over time.  Discontinuing use of chemicals in Table 3-28 with Medium to High hazard
concern levels can help avoid potential problems.
Table 3-28. Environmental Hazard Ranking of PWB Finishing Chemicals
,. <~ • -, " ; i "* r tปV
* - Chemical , ~ ~~ ~ „
^ , - * ?ฃฃ ^ - j"-
1,4-Butenediol
Acetic acid
Aliphatic acid A
Aliphatic acid B
Aliphatic acid D
Aliphatic acid E
Aliphatic dicarboxylic acid A
Aliphatic dicarboxylic acid C
Alkylalkyne diol
Alkylamino acid A
Alkylamino acid B
Alkylaryl imidazole
Alkylaryl sulfonate
Alkyldiol
Alkylimine dialkanol
Alkylphenol ethoxylate
Alkylphenol polyethoxyethanol
Alkylpolyol
Amino acid salt
Amino carboxylic acid
Ammonia compound A
Ammonia compound B
Ammonium chloride
Ammonium hydroxide
Arylphenol
Bismuth compound
Citric acid
Copper ion
Copper salt C
Copper sulfate pentahydrate
Cyclic amide
Ethoxylated alkylphenol
Lowest Acute (a) or
Chronic (c) Value(mg/L)
0.08 (c)
65 (a)
NR
MR
NR
NR
: NR
NR
NR
NR
NR
NR
NR
NR
NR
NR
0.008 (c) to 2 (c)
NR
NR
NR
NR
NR
161(a)
l(c)
NR
NR
l(c)
0.14 (a)
NR
0.001(c)
NR
NR
Hazard
Rank"
H
L
L
L
L
L
L
L
M
L
M
H
H
L
H
MtoHb
MtoHb
L
L
L
L
H
L
M
M
M
M
H
H
H
L
MtoHb
                                          3-100

-------
3.3 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
/ e , ^ ^Chenficai f "\ „'- /, * 7
^ F ^ V
Ethylenediamine
Ethylene glycol
Ethylene glycol monobutyl ether ฐ
Fatty amine
Fluoboric acid
Gum
Hydrochloric acid
Hydrogen peroxide
Hydroxyaryl acid
Hydroxy aryl sulfonate
Hydroxy carboxylic acid
Inorganic metallic salt A
Inorganic metallic salt B
Inorganic metallic salt C
Lead
Maleicacid •
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
LowesfcAcute^a) or
Chronic (c) Value (mgflL
0.16 (c)
440 (c)
3.9 (c)
NR
1.4(c)
NR
15 (c)
1.7 (a)
NR
NR
NR
NR
NR
NR
. 4.1 (c)
993 (c)
14,339 (c)
>100 (c)
1.3 (a)
NR
47 (c)
NR
278 (c)
NR
>0.03 (c)
<0.1 (c)
292 (c)
NR
0.001 (c)
NR
NR
10,616 (c)
103,000 (c)
NR •
NR
0.2 (c)
NR
2,241 (c)
Hazard
Rank*
M
L
M
H
- M
L
M
M
M
L
L
H
H
, H
; M
L
L
L
M
L
L
L
; L.
L
H
H
L
M
H
H
L
L
L
L
L
M
M
L
        3-101

-------
33 HUMAN HEALTH AND ECOLOGICAL HAZARDS SUMMARY
t f < sซ- _ * -%~ *
* Chemical
.. i ป +
Thiourea
Tin
Tin chloride
Transition metal salt
Unspecified tartrate
Urea
Urea compound B
Urea compound C
Vinyl polymer
Lowest Acute (a) or
Chronic (c) Value (mg/L)
0.3 (c)
0.07 (c) .
0.4 (c)
NR
NR
•>100(c)
NR
NR
NR
Hazard
Rank3
M
H
M
M
L
L
M
M
L
• Ranking based on the lowest estimated acute or chronic value; H = high, M = medium, L - law.
b Toxicity of breakdown product results m Mgh hazard rank.
0 Diethylene glycol monobutyl ether reviewed instead; both chemicals are very similar.
NR: Not reported in order to protect confidential ingredient identity.
3.3.4  Summary

       For human health hazards, toxicity data in the form of RfDs, RfCs, NOAELs, LOAELs,
and cancer slope (cancer potency) factors were compiled for inhalation and dermal pathways.
Inorganic metallic salt A (a confidential ingredient used in the nickel/gold process) was the only
chemical with an established cancer slope (cancer potency) factor.  Other chemicals in the surface
finishing processes are carcinogens or suspected carcinogens, but do not have established slope
factors.  Strong inorganic acid mist of sulfuric acid has been determined by IARC to be a human
carcinogen (LARC 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 tbiourea 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 NO AEL or LOAEL from which to calculate
MOE. For a number of chemicals, no quantitative risk indicator could be calculated for direct
comparison of risk among alternatives. A qualitative assessment was done for 33 chemicals,
based on chemical structure, for which no quantitative non-cancer health effects measures were
available.
                                           3-102

-------
                                3.3 HUMAN HEALTH AND ECOLOGICAL, HAZARDS SUMMARY
       An ecological hazards assessment was performed based on chemical toxicity to aquatic
organisms. CCs were estimated for surface finishing chemicals using an established EPA method.
A CC is an acute or chronic toxicity value divided by a UF. UFs are dependent on the amount
and type of toxicity data contained in a toxicity profile and reflect the amount of uncertainty about
the potential effects associated with a toxicity value. CCs were determined for aquatic species
(e.g., Daphnia, algae, and/or fish). CCs are compared to estimated  surface water concentrations
modeled from PWB wastewater releases in Section 3.4.

       Chemicals were also ranked for aquatic toxicity concern levels using established EPA
criteria (high, moderate, and low concern) based on the available toxicity data.  The number of
chemicals with a high aquatic hazard concern level include eight in the HASL process, nine in
nickel/gold, five in nickel/palladium/gold, five in OSP, three in immersion silver, and six in the
immersion tin process.
                                         3-103

-------
3.4 RISK CHARACTERIZATION
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, MSDSs, other information provided by product suppliers, and
  other available information) with standard EPA exposure assumptions (e.g., for inhalation rate,
  surface area of dermal contact, and other parameters).  The pathways for which potential
  exposure from surface finishing process baths was quantified include inhalation and dermal
  contact for workers, inhalation for the general population Irving 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 (\J.S. EPA, 1991a): ^
 where,
 j      =
 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:

 For carcinogens:
      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 balhs are opened for a short.period of time. After characterizing risks from inhalation for non-conveyorized lines
inhalation exposures and risks were estimated for the subset of inhalation chemicals of concern for conveyorized lines
No chemical exposures from inhalation resulted in risks above concern levels for conveyorized lines.

   16 Different averaging times are used for characterizing risk for carcinogenic and non-carcinogenic effects. For
carcinogenic agents, because even a single incidence of exposure is assumed to have the potential to cause cancer
throughout an individual's lifetime, the length of exposure to that agent is averaged over a lifetime. An additional factor
is that the cancer latency period may extend beyond the period of working years before it is discernible For chemicals
exhibiting non-cancer health effects from chronic (longer-term) exposure, where there is an exposure threshold (a level
below which effects are not expected to occur), only the time period when exposure is occurring is assumed to be
relevant and is used as the averaging time.


                                             3-105

-------
3.4 RISK CHARACTERIZATION
For non-carcinogens:
                          LADD  = a)(EF)(ED)/[(BW)(ATCAE)]
                           ADD = (I)(EF)(ED)/[(BW)(ATNC)]
where,
LADD =
ADD  =
EF    =
ED    =
BW   =
ATC  —
   NC
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:

                             .    D = (S)(C)(f)(h)(0.001)
    17 This version of the ISCLT model is provided as part of the Risk* Assistant™ 2.0 software package (Hampshire
 Research Institute, 1995).

 '                         '.            :    3-106

-------
                                                            3.4 RISK CHARACTERIZATION
 where,
 D
 S
 c
 f
 h
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 from publicly-available bath chemistry data, disclosed proprietary chemical
 information, supplier data sheets, and PWB Workplace Practices Questionnaire information. A
 permeability coefficient (rate of penetration through skin) was estimated for organic compounds
 and a default rate assumption was used for inorganic chemicals. Reliance on such estimates in the
 absence of data is a source  of uncertainty in the exposure assessment.

       Key assumptions in the exposure assessment include the following:
                                                         /
       The exposure frequency (i.e., days/year of line operation) was based on the time required
       to manufacture 260,000 ssf of PWB.
•      For dermal exposure, it was assumed that line operators do not wear gloves.  Although
       the data suggest that many surface finishing line operators do wear gloves for various
       activities, it was  assumed for this evaluation that workers do not wear gloves, to account
       for the subset  of workers who dp not wear proper personal protective equipment.
•      For dermal exposure, it was assumed that all  non-conveyorized lines are manual hoist.
•      The worker on a non-conveyorized line is assumed to potentially have dermal contact for
       the entire time spent in the surface finishing process area, and the contact time is assumed
       to be divided equally among the baths over an 8-hour .workday.  This does not mean that a
       worker has both hands immersed in a bath for that entire time but that the skin is in
       contact with bath solution (i.e., the hands may remain wet from contact).  This assumption
       may result in an overestimate of dermal exposure.
       For estimating ambient (outdoor) air concentrations, it was assumed that no air pollution
       control technologies are used to remove airborne chemicals from facility air prior to
       venting it to the outside.
                                        3-107

-------
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 hi 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 ah" 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 tower value, and ten percent reported a higher value.

                '                  "~~      3-108                                         ~~

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

         Some values used in the exposure calculations, however, are better characterized as
  "what-if," especially pertaining to use of gloves, process area ventilation rates, and production
  times (days/year) required to manufacture 260,000 ssf of PWB for the model facility.  ('What-if'
  represents an exposure estimate based on postulated questions, making assumptions based on
  limited data where the distribution is unknown.) Because some part of the exposure assessment
  for both inhalation and dermal exposures qualifies as a "what-if' descriptor, the entire assessment
  should be considered "what-if."

  3.4.2   Summary of Human Health Hazards Assessment

        For human health hazards, toxicity data in the form of RfDs, RfCs, NOAELs LOAELs
  and cancer slope (cancer potency) factors were compiled for inhalation and dermal pathways
  Inorganic metallic salt A (a confidential ingredient used in the nickel/gold process) was the only
  chemical with an established cancer slope (cancer potency) factor. Other chemicals in the surface
 finishing processes are known or suspected carcinogens, but do not have established slope factors
  Strong inorganic acid mist of suffiiric acid has been determined by IARC to be a human
 carcinogen ([ARC Group 1).  Suffiiric acid is used in every surface finishing process in this
 evaluation. It is not expected, however, to be present as a strong acid mist because it is greatly
 diluted in the aqueous baths.  Lead and thiourea have been determined by IARC to be possible
 human carcinogens (IARC Group 2B) and lead has also been classified by EPA as a probable
 human carcinogen (EPA Class B2).  Lead is used in tin-lead solder in the HASL process
 Thiourea is used in the immersion tin process.  Urea compound B, a confidential ingredient in the
 nickel/gold and nickel/palladium/gold processes, is possibly carcinogenic to humans.

 3.4.3  Summary of Ecological Hazards Assessment

       An ecological hazard assessment was performed based on chemical toxicity to aquatic
 organisms. CCs were estimated for surface finishing chemicals using an established EPA method
 (see Table 3-27 and Appendix H). A CC is an acute or chronic toxicity value divided by a UF
 UFs are dependent on the amount and type of toxicity data contained in a toxicity profile and
 reflect the amount of uncertainty about the potential effects associated with a toxicity value
 Concern concentrations were determined for aquatic species (e.g., Daphnia, algae and/or fish)
 for each chemical.  The lowest CCs are for inorganic metallic salt A, silver nitrate, and silver salt
 Chemicals also were ranked for aquatic toxicity concern levels using established EPA criteria
 (high, moderate, and low concern) based on the available toxicity data (see Table 3-28)  The
number of chemicals with a high aquatic hazard concern level include eight in the HASL process
nine in nickel/gold, five in nickel/palladium/gold, five in OSP, three in immersion silver  and six in
the immersion tin process.                                                     '
                                         3-109

-------
3.4 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 in 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 (hi
                    mg/kg-day). LADDs were calculated in the Exposure Assessment (Section
                    3.2).

Slope factor (qx *) is defined in Section 3.3.1.

Non-Cancer Risk Indicators

       Non-cancer risk estimates are expressed either as an HQ or as an MOE, depending on
whether or not RfDs and RfCs are available. There is a higher level of confidence in the HQ than
the MOE, especially when the HQ is based on an RfD or RfC that has been peer-reviewed by
EPA (as with data from the EPA IRIS database). If an RfD or RfC is available, the HQ is
calculated to estimate risk from chemicals that exhibit chronic, non-cancer toxicity.  (RfDs and
RfCs are defined in Section 3.3.2.) The HQ is the unitless ratio of the RfD (or RfC) to  the
potential dose rate. For surface finishing chemicals that exhibit non-cancer toxicity, the HQ was
calculated by:
                                    HQ = ADD/RฃD
 where,
                                          3-110

-------
                                                            3.4 RISK CHARACTERIZATION
  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/Tcg-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 RflC)
  below which it is unlikely, even for sensitive subgroups, to experience adverse health effects.
  Unlike cancer risk, the HQ does not express probability and is not necessarily linear; that is, an
  HQ often does not mean that adverse health effects are ten times more likely to occur than for
  HQ of one.  However, the ratio of estimated dose to RfD/RfC reflects the level of concern.
                                                          an
        For chemicals where an RfD or RfC was not available, an MOE was calculated by:

                          MOE = NOAEL/ADD or LOAEL/ADD

 As with the HQ, the MOE is not a probabilistic statement of risk. The ratio for calculating MOE
 is the inverse of the HQ, so that a high HQ (exceeding one) indicates a potential concern, whereas
 a high MOE (exceeding 100 for a NOAEL-based MOE or 1,000 for a LOAEL-based MOE)
 indicates a low concern level. (NOAELS and LOAELs are defined in Section 3.3.2.) As the
 MOE increases, the level of concern decreases. (As the HQ increases, the level of concern also
 increases.) In general, there is a higher level of confidence for HQs than for MOEs because the
 toxicity data on which RfDs and RfCs are based have passed a more thorough level of review, and
 test-specific uncertainty factors have been included.

        Both the exposure estimates and toxicity data are specific to the route of exposure (i e
 inhalation, oral, or dermal).  Very few RfDs, NOAELs, or LOAELs are available for dermal  '
 exposure. If oral data were available, the following adjustments were made to calculate dermal
 values based on EPA (1989) guidance:
              RfD,
                 'DER
where,
SFDER
GI absorption
              CHORAL) (GI absorption)
              (NOAEL or LOAEL?RAL) (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
       This adjustment is made to account for the fact that the oral RfDs, NOAELs, and
LOAELs are based on an applied dose, while dermal exposure represents'an estimated absorbed
dose. The oral RfDs, NOAELs, and LOAELs used to assess dermal risks therefore were adjusted
using GI absorption to reflect an absorbed dose. Table 3-29 lists the GI absorption data for
chemicals used in calculating risk from dermal exposure. (Data for some proprietary ingredients
are not presented in order to protect confidential chemical identities.)
                                        3-111

-------
3.4 RISK CHARACTERIZATION
Table 3-29. Gastrointestinal (GI) Absorption Factors
; Chemicals a
Acetic acid
Aliphatic acid A
Aliphatic acid D
Aliphatic dicarboxylic acid C
Alkyldiol
Alkylpolyol
Amino carboxylic acid
Ammonia compound A
Ammonia compound B
Ammonium chloride
Ammonium hydroxide
Aryl phenol
Copper ion, Copper salt C, and
Copper sulfate pentahydrate
Cyclic amide
Ethylene glycol
Ethylene glycol monobutyl ether
Ethylenediamine
Hydroxy carboxylic acid
Hydrogen peroxide
Inorganic metallic salt A
Inorganic metallic salt B
Inorganic metallic salt C
Nickel sulfate
Phosphoric acid
Potassium gold cyanide .
Propionic acid
Silver nitrate
Silver salt
Stannous methane sulfonic acid
Tin chloride
Unspecified tartrate
GI AbsorptionFactor
0.9
0.9
0.5
0.2
Ml
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
Ml
0.15
0.15
0.05
0.2
0.2
0.2
0.08
Ml
0.2
0.5
0.5
Source
chemical profile b
chemical profile b
Ml
assumption c
Ml
assumption- c
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 c
assumption ฐ
Ml
Ml
Ml
midpoint of range, 0.01 - 0.1,
chemical profile
U.S. EPA, 1995
assumption ฐ
assumption c
midpoint of range, 0.05 - 0.1
(U.S. EPA, 1991c; ATSDR, 1990b)
Ml
assumption0
Johnson and Greger, 1982
chemical profile b
                                       3-112

-------
                                                                3.4 RISK CHARACTERIZATION
', - ,€hemicalsa I %
Urea compound C
Vinyl polymer
GI Absorption Factor
0.2
0.1
v Source
assumption0 .
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 fecflity, and should not be used as
 absolute indicators of actual health risks to surface finishing line workers or to the public.

 Occupational Setting                                                    •       .

        Estimated cancer risks and non-cancer risk indicators from occupational exposure to
 surface finishing chemicals are presented below. It should be noted that no epidemiological
 studies of health effects among PWB workers were located.

        Inhalation Cancer Risk. Nickel/gold is the only process containing a chemical for which
 a cancer slope (cancer potency)  factor is available. Inorganic metallic salt A, in the nickel/gold
 process, is the only chemical for which an inhalation cancer risk has been estimated. This metal
 compound is considered a human carcinogen.19

        Inhalation exposure estimates are based on the assumptions that emissions to indoor air
 from conveyorized lines are negligible, that the air in the process room is completely mixed and
 chemical concentrations are constant over time, and that no vapor control devices (e.g.-, bath
 covers) are used in non-conveyorized lines.  The exposure estimates use  90th percentile modeled
 air concentrations, which means that, based on the PWB Workplace Practices Questionnaire data
 and available information on bath concentrations, approximately 90 percent of the facilities are
    19                                                                               '
      A cancer classification of known human carcinogen has been assigned by either the EPA, IARC, and/or the
National Toxicology Program (NTP). Further details about the carcinogen classification are not provided hi order to
protect the confidential chemical's identity.
                                            3-113

-------
3.4 RISK CHARACTERIZATION
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 W6 milligrams inorganic metallic salt A per cubic meter of air,
over an 8 hour-day, for line operators using the non-conveyorized nickel/gold process.  Cancer
risks less than 1 x 10"6 (one in one million) are generally considered to be of low concern.  The
use of modeled, steady state, workplace air concentrations instead of actual monitoring data of
average and peak concentrations thus emerges as a significant source of uncertainty in estimating
cancer risk to workers exposed to inorganic metallic salt A in this industry. The available
toxicological data do not indicate that dermal exposure to inorganic metallic salt A increases
cancer risk, but no dermal cancer studies were located.

       Risks to other workers would be proportional to the amount of time spent in the process
area.  The exposure from inhalation for a typical line operator is based on spending 8 hr/day in the
surface finishing process area. Exposure times (i.e., time spent in the process area) for various
worker types from the workplace practices database are listed below. The number in parentheses
is the ratio of average time for that worker type to the 8 hr/day exposure time-for a line operator.

•      laboratory technician: 2.8 hr/day (0.35);
•      maintenance worker:  1.6 hr/day (0.2);
•      supervisor: 5.5 hr/day (0.69); and
•     wastewater treatment operator:   1 hr/day (0.12).

(Other types of workers may be in the process area for shorter or longer tunes.)

       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 (IAR.C Group 2B); lead has
also been classified by EPA as a probable human carcinogen (EPA Class B2). Lead is used in tin-
lead solder in the HASL process. Thiourea is used in the immersion tin process. Urea compound
B, a confidential ingredient in the nickel/gold and nickel/palladium/gold processes, is possibly
carcinogenic to humans. There are potential cancer risks to workers from these chemicals, and
workplace exposures have been estimated, but cancer potency and cancer risks are unknown.
Additionally, strong inorganic acid mists of sulfuric acid have  been determined by IARC to be a
human carcinogen (IARC Group 1).  Sulfuric acid is used in every surface finishing process in this
evaluation. It is not expected, however, to be present as a strong acid mist because it used in
diluted form in the aqueous baths.

      Non-Cancer Risk. HQs and MOEs were calculated for line operators and laboratory
technicians from workplace exposures.  An HQ exceeding one indicates a potential concern.
Unlike cancer risk, the HQ does not express probability, only the  ratio of the estimated dose to
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).
                                         3-114

-------
                                                           3.4 RISK CHARACTERIZATION
       EPA considers high MOE values, such as values greater than 100 for a NOAEL-based
MOE or 1,000 for a LOAEL-based MOE, to pose a low level of concern (Barnes and Dourson,
1988). As the MOE decreases, the level of concern increases.  Chemicals are noted here to be of
potential concern if a NOAEL-based MOE is lower than 100, a LOAEL-based MOE is lower
than 1,000, or an MOE based on an effect level that was not specified as a LOAEL (used in the
absence of other data) is less than 1,000. As with the HQ, it is important to remember that the
MOE is not a probabilistic statement of risk.

       Inhalation risk indicators of concern are presented in Table 3-30. This includes chemicals
of potential concern based on MOE and/or HQ results, as well as cancer risk results for the one
chemical with a cancer slope factor. Inhalation exposure estimates are based on the assumptions
that emissions to air from conveyorized lines are negligible, that the air in the process room is
completely mixed and chemical concentrations are constant over time, and that no vapor control
devices (e.g., bath covers) are used in non-conveyorized lines.

       Dermal risk indicators of concern are presented in Table 3-31. This includes chemicals of
potential concern based on MOE and/or HQ. Dermal exposure estimates are based on the
assumption that both hands are routinely immersed in the bath, the worker does not wear gloves,
and all non-conveyorized lines are operated by manual hoist.

        Table 3-32 provides a summary of the potential health effects for the chemicals of
concern listed in Tables 3-30 and 3-31. It should be noted that Tables 3-30 and 3-31 do not
include chemicals for which toxicity data were unavailable. Table 3-33 lists chemicals where
inhalation or dermal exposure is expected to occur, but appropriate toxicity values are not
available. (Table 3-25 provides qualitative structure-activity information pertaining to chemical
toxicity for those chemicals without available measured toxicity data.)
                                        3-115

-------
3.4 RISK CHARACTERIZATION


1
•*ซ
S
it
A
u
•a
"S
M
b
^o

-
• P5
-•c
I







. •




. '.•••'-'-.•-•'
%(&
^^-. J2T-
W Ci'
'"/V "-1;,

.. _ -^ ป - - _


'?o'."'.'' /

-Q _.'.
C^ ••"•'
.-^ ; • " "
Q . ,""=* -
||



'••s)v'"7;:
Nickel/Gol
• ^,:^-':m-:


' .'"-"


j ^| ^_^
|s





"tS '--:'•' '•••• •"'
flj ""*" ' "
;B-:. ••'•':'-.;-
O ':' -
O :
^o- -:
"S , ,,", •









S







line operator




line operator









g








J3
en"
^~^

o
S
Q>
0
J M
f^j] *l ^?
il'e?3













/— ^
v\
en"
^^

S
1 •
o
•S g
^1  1— 1






^_
o
Ethylene gly









n
c?
T— 4


"S S
a S •

"T 'S
O 53
2 2
' • 5-j
s
S -*2
n a LOAEL (or less ti
lilable. Specific resul
•O KJ
1 *
J 0>
**H Q>
0 ?
nMOE less than 1,00
for which toxicity data
-H en
~S "3 rj.
IE .3 ซ
ฃ ! ง
pt o ^_,
•5 -ง o
ro s "-f
Q. ••* 'O
*rt ^ ^
S -^ ฃP
J 1
O ซ> ^
"S o ."ง
1^ 1
^w ^co -^
OT *-i a>

OT O *"•
ฃ i o
^ o •*-*
0> i, SH
•o O a>
"o S" o
a This table in
greater than 01
ingredients in
                                       3-116

-------
                                                        3.4 RISK CHARACTERIZATION
•s
2
                                      3-117

-------
3.4 RISK CHARACTERIZATION
I

JS

U

T3
!
                                                                            •o.
                                                                         <= "8

                                                                         "S 2
                                                a a
                                               ^ ซ
                                               ^S ซB
                                                03 "O

                                               J I
                                                gCO

                                                — * *r^
                                        3-118

-------
                                                                3.4 RISK CHARACTERIZATION
 o
I
 03
'ฃ
T3
 C3
 S
3
 ฃ
                **
                                   o
                                           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 in non-conveyorized HASL;
•      alkyldiol, hydrochloric acid, hydrogen peroxide, nickel sulfate, and phosphoric acid in non-
       conveyorized nickel/gold;
•      alkyldiol, hydrochloric acid, hydrogen peroxide, nickel sulfate, phosphoric acid, and
       propionic acid in non-conveyorized nickel/palladium/gold; and
•      ethylene glycol in non-conveyorized OSP.

       Chemicals with HQs from dermal exposure greater than one, NOAEL-based MOEs lower
than 100, or LOAEL-based MOEs lower than 1,000, include:
       copper sulfate pentahydrate in non-conveyorized and conveyorized HASL;
       ammonium chloride, ammonium hydroxide, copper sulfate pentahydrate, hydrogen
       peroxide, inorganic metallic salt B, and nickel sulfate in non-conveyorized nickel/gold;
       ammonia compound A, ammonium hydroxide,- copper sulfate pentahydrate, hydrogen
       peroxide, inorganic metallic 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 3-32. Summary of Potential Human Health Effects for Chemicals of Concern
Chemical of Concern
Ammonia compound A,
Ammonium chloride, and
Ammonium hydroxide
Alkyldiol
Copper ion,
Copper sulfate pentahydrate,
and Copper salt C
Ethylene glycol
Hydrochloric acid
Hydrogen peroxide
Inorganic metallic salt A
Inorganic metallic salt B
Nickel sulfate
Phosphoric acid
Propionic acid
Urea compound C
Potential Health Effects
Contact with ammonium chloride solution or fumes irritate the eyes.
Large doses of ammonium chloride may cause nausea, vomiting, thirst,
headache, hyperventilation, drowsiness, and altered blood chemistry.
Ammonia fumes are extremely irritating to skin, eyes, and respiratory
passages. The severity of effects depends on the amount of dose and
duration of exposure.
Can affect the respiratory system if inhaled, and kidneys if absorbed into
the body.
Long-term exposure to high levels of copper may cause liver damage.
Copper is not known to cause cancer. The seriousness of the effects of
copper can be expected to increase with both level and length of
exposure.
In humans, low levels of vapors produce throat and upper respiratory
irritation. When ethylene glycol breaks down in the body, it forms
chemicals that crystallize and can collect in the body, which prevent
kidneys from working. The seriousness of the effects can be expected to
increase with both level and length of exposure.
Hydrochloric acid in the air can be corrosive to the skin, eyes, nose,
mucous membranes, respiratory tract, and gastrointestinal tract.
hydrogen peroxide in the air can irritate the skin, nose, and eyes.
Ingestion can damage the liver, kidneys, and gastrointestinal tract
exposure can cause flu-like symptoms, weakness and coughing, and has
been linked to lung cancer and kidney disease.
Exposure to this material can damage the nervous system, kidneys, and
immune system.
Skin effects are the most common effects in people who are sensitive to
nickel. Workers who breath very large amounts of nickel compounds
lave developed lung and nasal sinus cancers. '
Inhaling phosphoric acid can damage the respiratory tract
Nk> 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 3-33. Data Gaps for Chronic Non-Cancer Health Effects for Workers
Chemical
-.^ *x-*.t ** ^
Inhalation a or Dermal b
Exposure Potential
SATRank , ,
(if available)
HASL
1,4-Butenediol
Alkylaryl sulfonate
Arylphenol
Fluoboricacid
Hydrochloric acid
Sodium hydroxide
Sulfuric acid
Tin
Inhalation and Dermal
Inhalation
Inhalation
Dermal
Dermal
Dermal
Dermal
Dermal
Low-moderate
Low
Moderate





Nickel/Gold
Aliphatic acid A
Aliphatic acid B
Aliphatic acid E
Aliphatic dicarboxylic acid A
Aliphatic dicarboxylic acid C
Alkylamino acid B
Ammonia compound B
Hydrochloric acid
Malic acid
Palladium chloride
Potassium compound
Sodium hydroxide
Sodium hypophosphite
Sulfuric acid
Urea compound B
Inhalation
Inhalation
Inhalation and Dermal
Inhalation
Inhalation
Dermal
Inhalation
Dermal
Inhalation
Dermal
Inhalation and Dermal
Dermal
Inhalation
Dermal
Inhalation and Dermal

Moderate

Low-moderate


Moderate-high
•
Low-moderate

Low

Low-moderate


Nickel/Palladium/Gold
Aliphatic acid B
Aliphatic acid E
Aliphatic dicarboxylic acid A
Aliphatic dicarboxylic acid C
Ammonia compound B
Hydrochloric acid
Malic acid
Palladium salt
Potassium compound
Sodium hydroxide
Inhalation
Inhalation and Dermal
Inhalation
Inhalation
Inhalation
Dermal
Inhalation
Dermal
Inhalation and Dermal
Dermal
Moderate

Low-moderate

Moderate-high

Low-moderate

Low

                                     3-122

-------
                                                             3.4 RISK CHARACTERIZATION
'' -• ' Chemical, v
-* -,"1 /"' " - t Ir "
Sodium hypophosphite monohydrate
Sulfunc acid
Urea compound B
Inhalation * or Dermal b
Exposure Potential
Inhalation
Dermal
Inhalation and Dermal
SATRank
(if available) - *
Low-moderate


OSP
Acetic acid
Alkylaryl imidazole
Aromatic imidazole product
Arylphenol
Hydrochloric acid
Sodium hydroxide
Sulfunc acid
Inhalation
Dermal
Dermal
Inhalation
Dermal
Dermal
Dermal

Low-moderate

Moderate



Immersion Silver
1,4-Butenediol
Nitrogen acid
Sodium hydroxide
Sulfunc acid
Dermal
Dermal
Dermal
Dermal
Low-moderate



Immersion Tin
Alkylaryl sulfonate
Fluoboric acid
Hydrochloric acid
Methane sulfonic acid
Sulfunc acid
Thiourea
Urea compound C
Inhalation
Dermal
Dermal
Dermal
Dermal
Dermal
Inhalation
Low






a Applies only to the non-cpnveyorized 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 nearbv
model PWB facility.
a
                                          3-123

-------
3.4 RISK CHARACTERIZATION
       Cancer Risk. As with the occupational setting, the nickel/gold process is theonly
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. An MOE was
calculated for chemicals if an inhalation LOAEL or NOAEL was available and an RfC was not.
All MOEs for ambient exposure are greater than 1,000 for all processes, indicating low concern.

        These results suggest there is low risk to nearby residents, based on incomplete but best
available data.  Data limitations include the use of modeled air concentrations using data compiled
for a model facility rather than site-specific, measured concentrations. For estimating ambient
(outdoor) air concentrations, one key assumption is that no air pollution control technologies are
used to remove airborne chemicals from facility air prior to venting it to the outside. Other data
limitations are the lack of solid waste data to characterize exposure routes in addition to
inhalation, and lack of toxicity data for many chemicals.
    20 Upper bound refers to the method of determining a slope factor, where the upper bound value (generated from a
 certain probability statement) for the slope of the dose-response curve is used. Excess means the estimated cancer risk
 is in addition to the already-existing background risk of an individual contracting cancer from all other causes.

    21 A cancer classification of known human carcinogen has been assigned by either the EPA, IARC, and/or NTP. ..
 Further details about the carcinogen classification are not provided in order to protect the confidential chemical's
 identity.

-------
                                                            3.4 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 hi Children
 (U.S. EPA, 1994). Both of these models relate estimated exposure levels to a lead concentration
 in blood, which can then be compared to blood-lead levels at which health effects are known to
 occur.  These models are described further in Section 3.2.4 of the Exposure Assessment.

        Table 3-34 presents federal (and other) regulations and guidelines for lead.  This table also
 presents comparable lead exposure values for workers and the ambient environment potentially
 resulting from the lead in tin-lead solder used in the HASL process. For workers, the lowest
 federal target or action levels are from OSHA and ACGffl, at 30 ng/dL hi 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
 ug/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 ug/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 hi contact with soil (Stanek et al., 1997 and
 U.S. EPA 1997a), respectively.22 (Ingestion data are not available specifically for a HASL worker
handling solder.) However, these results do indicate that there may be risk from lead exposure via
the ingestion route from poor hygiene practices.                              .        •
     10 mg/day is an average estimate; 50 mg/day is a central tendency estimate.

                                         3-125

-------
3.4 RISK CHARACTERIZATION






i.
.0
<ซ->
S?
es
S
S
(/)
c
a
1
3
ป
^yj
W
*
Tt
CO
i
g






">
 *™l
•fS
O Ut
. j_^ a>

\fe





"S
(U ฐ
H-(
L^
3
<*^ - ;
1 Regulations and Guidelines
-!••.'
(*•! '

































Workplace
CN
i
>o


Occupational blood-lead
monitoring data.
-5 GO •*••
fib g oi
en S ^
vo <>i as
1 ox P


Modeled (ALM) blood-lead
data for an adult worker.


j_i
T>
a.
CD
cn

to
o
o
^
=sl
0
en
%
S
*o
2
x!
ป.,
S
OSHA, adults "who wish to bf

|
o





cj
fl>
O
OSHA, blood-lead level of cor

i
0
U1








OSHA, medical removal

i
0
en








ACGIH (ACGIH, 1998)
DO
O
0
o

"3
ซ-t
2P
g
M
ฃ
•w*
NIOSH, level to be maintainec
concentrations

*T3 W
rt ^T!
tt> >
i  t*
cnSg


Modeled (ALM) fetal blood-
lead level.


j
•2
=L
O


60
O
O
!3>
^
o
cn







O

i
IS.
o








O
8
•a J
*S ra
:- ^s
g p
^^ OjJ
a*rt OT
O 'flS
W o ^
1
en



Workplace air monitoring
data (average of HASL
process area monitoring data
provided by one P WB
manufacturer).


"s
"5s
a.
o



1
o








OSHA PEL (8 hr TWA)

1
0
o








NIOSH REL (NIOSH, 1994)

1
o




oo"
ON
ON
r— (
ACGIH TLV TWA (ACGIH,
CO
O
5-c
(D
t-<
'S
1
11
























Sซ
a
P
_sซ
^
Ambient En
1
o
0
o
0
o


Ambient air concentrations
near a P WB facility based on
HASL workplace air
monitoring data and air
dispersion model.


"S .
"So

IO
T— C

^^
III
2 Jj, 1
o

'O
c3
c*1
ง

M
National Ambient Air Quality
(U.S.. EPA, 1987b)




,
il
11- .
8 ^ S Jง
-g r- O *3
| Jj> ฎ •- ^
"S o -a "2 ^ so '•;
T1 ^"^ M M ^^ J-1 t-J
o — < m
'S ^
C8 0>
q> Ji

ซ t
"S "S
o  . *K
ฃ S.
O .
o so

&s CD
o> 01
^5 A
tl— 1 E/1
3 have an elevated level of lead i;
for all children with blood level
II
2 a
!S *ฐ
o "^ /-^x
S2 S ON
a3 2 ON
a CDC consid
remediation si
^g/dL(RTI,l
                                       3-126

-------
                               3.4 RISK CHARACTERIZATION
 I
 0>
f
 CS
.53
•8
bo
o
o
NOTES
ACGIH:
CD
EP
NI
O
           3-127

-------
3.4 RISK CHARACTERIZATION
       In addition to an adult worker, we used the AIM 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 ng/dL can be compared to the
guidance level from CDC and EPA of 10 ng/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 ug/m3 can be compared to the lowest federal regulatory
level of 50 ug/m3 (an OSHA, 8-hour, time-weighted average permissible exposure limit). For
ambient air near a facility, an estimated air concentration of 0.0001 ug/m3 is well below the EPA
air regulation of 1 .5 ug/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 (RlEco) fฐr aquatic organisms as a unitless 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
Medical evaluation and environmental remediation should be done for all children with blood-lead levels 2:20 jig/dL.
Medical treatment may be necessary in children if the blood-lead concentration is > 45 ng/dL (RTI, 1999).

   24 Results from both personal monitoring for HASL line operators and air samples from the HASL process area
were averaged.

                                          3-128:!:

-------
                                                             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 3-35.
 Estimated surface water concentrations of several non-metals exceed the CC, as follows:

 •      alkylaryl sulfonate, 1,4-butenediol, hydrogen peroxide, and potassium peroxymonosulfate
       in the non-conveyorized HASL process;
 •      alkylaryl sulfonate, hydrogen peroxide,and potassium peroxymonosulfate in the
       conveyorized HASL process;
 •      alkylaryl imidazole in non-conveyorized and 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 3-35. Summary of Aquatic Risk Indicators for Non-Metal Chemicals of Concern
Chemical
1,4-ButenedioI
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 ffisk Indicator {RIjjco) 1
HASL

-------
3.4 RISK CHARACTERIZATION
          Table 3-36. Summary of Aquatic Risk Indicators for Metals Assuming
                                  No On-Site Treatment
Chemical
Copper ion
Copper-sulfate pentahydrate
Nickel sulfate
Potassium gold cyanide
CC
(mg/L)
*
0.001
0.01
0.01
0.003
*• <-. ? Aquatic Risk Indicator (RIECO) < "<•
HASL
(NQ
NA
5.1
NA
NA
HASL
(Q
NA
3.8
NA
NA
NTckeJ/
Cold
NA
NA
5.1
1.5
Nickel/ Palladium/
Gold
NA
NA
5.5
NA
OSP
(NQ
46
6.3
NA
NA
OSP
CO
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 fete 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 (effects 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 comes from use of structure-activity relationships (SARs)
 for estimating human health hazards in the absence of experimental toxicity data. Specifically this
 was done for: aliphatic acid B, aliphatic dicarboxylic acid A, alkylalkyne diol, alkylamino acid A,
 alkylaryl imidazole, alkylaryl sulfonate, alkylimine dialkanol, amino acid salt, ammonia compound
 B, aryl phenol, bismuth compound, 1,4-butenediol, citric acid, ethoxylated alkylphenol fatty
 amine, hydroxyaryl acid, hydroxyaryl sulfonate, maleic acid, malic acid, potassium compound
 potassium peroxymonosulfate, quaternary alkylammonium chlorides, sodium benzene sulfonate
 sodium hypophosphite, sodium hypophosphite monohydrate, substituted amine hydrochloride '
 and transition metal salt.            .                                                 '
       Uncertainties in assessing risk from dermal exposure come from the use of toxicological
potency factors from studies with a different route of exposure than the one under evaluation (i e
using oral toxicity measures to estimate dermal risk).  This was done for chemicals with oral RfDs
and chemicals with oral NOAELs or LOAELs (as noted in Tables 3-25 and 3-26) Uncertainties
                                          3-131

-------
3.4 RISK CHARACTERIZATION
in dermal risk estimates also stem from the use of default values for missing gastrointestinal
absorption data. Specifically, this was done for: aliphatic acid E, aliphatic dicarboxylic acid C,
alkylamino acid B, alkylpolyol, amino carboxylic acid, ftuoboric 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 uncertainly in using oral data for dermal exposure and in estimating dermal absorption rates,
which could result in either over- or under-estimates of exposure and risk.

       A third significant source of uncertainty is from the use of SARs to estimate toxicity in the
absence of measured toxicity data, and the lack of peer-reviewed toxicity data for many  surface
finishing chemicals.  Other uncertainties associated with the toxicity data include the possible
effects of chemical interactions on health risks, and extrapolation of animal data to estimate
human health risks from exposure to inorganic metallic salt A and other PWB chemicals.

       Another major source of uncertainty in estimating exposure is the reliance on modeled
data (i.e., modeled ah- 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 assumptions, as would be used for an overall high-
 end exposure estimate. Some values used in the exposure calculations, however, are better
 characterized as "what-if," especially pertaining to exposure frequency, bath concentrations, use
 of gloves, and process area ventilation rates for a model facility. Because some part of the
 exposure assessment for both inhalation and dermal exposures qualifies as a "what-if' descriptor,
 the entire assessment should be considered "what-if."

 Occupational Exposures and Risks

        Health risks to workers were estimated for inhalation exposure to vapors and aerosols
 from surface finishing baths and for dermal exposure to surface finishing bath chemicals.
 Inhalation exposure estimates are based on the assumptions that emissions to indoor air from
 qonveyorized 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 occup'ational 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 (one in one million) are generally considered to be of
 low concern; Risks to other types of workers25 were assumed to be proportional to the average
 amount of time spent in the process area, which ranged from 12 to 69 percent of the risk for a line
 operator.
     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 hi 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 has been
                                          3-134

-------
                                                              3.4 RISK CHARACTERIZATION
 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 hi 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 3-37 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 fhe EPA, IARC, and/or NTP.
Further details about the carcinogen classification are not provided in order to protect toe confidential chemical identity.

                                          3-135

-------
3.4 RISK CHARACTERIZATION
 Table 3-37.  Overall Comparison of Potential Human Health and Ecological Risks for the
                     Nou-Conveyorized HASL and Alternative Processes
Process
V
•f 1
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)
"• \ ~ ~ ^ •* Numbejr^of Qiemicals ^ - r ^ ' f
Potential ,
Carcmbgena
'•, * !
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 ^
".Gaps*-.
3
0
10
9
'2
0
0
'2
0
Dermal
Data
Gaps'
6
6
8
7
5
5
4
5
5
Aquatic
Concern1
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-
21).
b The number of chemicals for which the HQ for worker inhalation exceeds 1, the NOAEL-based MOE is less than
100, or the LOAEL-based MOE is less than 1,000. See Table 3-30 for detailed results.
c The number of chemicals for which the HQ for dermal contact by workers exceeds I, the NOAEL-based MOE is less
than 100, or the LOAEL-based MOE is less than 1,000. See Table 3-30 for detailed results.
d The number of chemicals for which worker inhalation exposure is possible, but appropriate toxicity data are not
available for calculating a risk indicator (see Table 3-33).
* The number of chemicals for which worker dermal contact is possible but appropriate toxicity data are not available
for calculating a risk indicator (see Table 3-33).
f The number of chemicals for which the ecological risk indicators exceeds the concern level (i.e., RI^ >  1.0). See
Table 3-35 for detailed results.
                                              3-136

-------
                                                        3.5 PROCESS SAFETY ASSESSMEJNT
 3.5    PROCESS SAFETY ASSESSMENT

        Process safety is a concern and responsibility of employers and employees alike. Each
 company has the obligation to provide its employees with a safe and healthy work environment,
 while each employee is responsible for his/her own safe personal work habits. In the surface
 finishing process of PWB manufacturing, hazards may be either chemical or process hazards.
 Chemicals used in the surface finishing process can be hazardous to worker health and, therefore,
 must be handled and stored properly, using appropriate personal protective equipment and safe
 operating practices. Automated equipment can be hazardous to employees if safe procedures for
 cleaning, maintaining, and operating the equipment are not established and regularly performed.
 These hazards can result in serious injury and health problems to employees, and potential damage
 to equipment.

        The U.S. Department of Labor and the Occupational Safety and Health Administration
 (OSHA) have established safety standards and regulations to assist employers in creating a safe
 working environment and protecting workers from potential workplace hazards.  In addition,
 individual states may also have safety standards regulating chemical and physical workplace
 hazards for many industries.  Federal safety standards and regulations affecting the PWB industry
 can be found in the Code of Federal Regulations (CFR) Title 29, Part 1910, and are available by
 contacting your local OSHA field office.  State and local regulations are available from the
 appropriate state office.

        An effective process safety program identifies potential workplace hazards and, if possible,
 seeks to eliminate or at least reduce their potential for harm.  Some companies have successfully
 integrated the process safety program into their ISO 14000 certification plan, often  establishing
 process safety practices that go beyond OSHA regulations.  This section of the CTSA presents
 chemical and process safety concerns associated with the surface finishing baseline technology and
 substitutes, as well as OSHA requirements to mitigate these concerns.

 3.5.1   Chemical Safety Concerns

        As part of its mission, OSHA's Hazard Communication Standard (29 CFR 1910.1200)
 requires that chemical containers be labeled properly with chemical name and warning information
 [.1200(f)], that employees be trained in chemical handling and safety procedures [,1200(h)], and
that a MSDS be created and made available to employees for every chemical or chemical
formulation used in the workplace [. 1200(g)]. Each MSDS must be in English and include
information regarding the specific chemical identity and common name of the hazardous chemical
ingredients. In addition, information must be provided on the physical and chemical.
characteristics of the hazardous chemical(s), known acute and chronic health effects and related
health information, exposure limits, whether the chemical is a carcinogen,  emergency and first-aid
procedures, and the identification of the organization preparing the data sheet. Copies of MSDSs
for all of the chemicals/chemical formulations used must be kept and made available to workers
who may come into contact with the process chemicals during their regular work shift.
                                         3-137

-------
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 3-38 presents a breakdown of surface finishing chemical products that, when in
concentrated form, are flammable, combustible, explosive, or pose a fire hazard.  The following
lists OSHA definitions for chemicals in these categories, and discusses the data presented in the
table.                                   .
                                          3-138

-------
                                                             3.5 PROCESS SAFETY ASSESSMENT
        Table 3-38. Flammable, Combustible, Explosive, and Fire Hazard Possibilities
Surface Finishing ProcessJ C
^•>^ i -/
HASLC
OSP d (2 product lines)
Immersion Silver
Immersion Tin d (2 product lines)
Bath
Type
Cleaner
Microetch
Microetch
Cleaner
Immersion Tin
Hazardous Property "• b
Flammable
1(3)
1(3)


Combustible




Explosive
1(1)

1(1)
1(4)
Fire Hazard
1(1)
: 2(3)
2(3)
1(1)

        —,	_. ^. „„._. ^.vuv "^"& j.vj.*ij.M.t. - Tr VFJL piv/uuvio lutcimg woxjLtt. uciilUUOIl lOr ulc given flaZaTOOUS
 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., sulfiuic 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 wim 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.

        Two chemical products are reported as flammable according to MSDS data. Although the
chemicals are flammable in their concentrated form, none of the  chemical baths in the surface
finishing line contain flammable aqueous solutions.
                                            3-139

-------
3.5 PROCESS SAFETY ASSESSMENT
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 their MSDSs.

Fire Hazard - A chemical product that is a potential fire hazard is required by OSHA to be
reported on the product's MSDS.  According to MSDS data, six chemical products are reported
as potential fire hazards.

Corrosive. Oxidizer. and Reactive Surface Finishing Chemical Products

       A breakdown of surface finishing chemical baths containing chemical products that are
corrosive, oxidizers, or reactive in their concentrated form is presented in Table 3-39. The table
also lists process baths that contain chemical products that may cause a sudden release of pressure
when opened. The following lists OSHA definitions for chemicals in these categories and
discusses the data presented in the table.

Corrosive - As defined by OSHA (29 CFR 1910.1200 [Appendix A]), a chemical is considered
corrosive if it causes visible destruction of, or irreversible alterations in, living tissue by chemical
action at the site of contact following ah 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 3-39.  Corrosive, Oxidizer, Reactive, Unstable, and Sudden Release of Pressure
                          Possibilities for Surface Finishing Processes
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
Immersion Tin
~_ Hazardous Property a>b ' -" -- /
Corrosive
1(1)
3(4)
KD
3(4)
3(3)
KD
1(1)
3(4)
3(3)
1(4)
3(3)
1(3)
1(1)
3(4)
1(1)
1(3)
1(2)
2(2)
1(1)
3(4)
Oxidizer
1(3)
1(4)
1(4)
1(3)
1(3)

Reactive






Unstable
1(3)



1(3)

Sudden Release
ofPressure
1(4)
1(4)
1(4)
1(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, 3(4) means that four of the five products in the bath were classified as corrosive
per OSHA criteria, as reported by the products' MSDSs.
b Data for pure chemicals (e.g., sulfuric acid) not sold as products were obtained from the Merck Index (Budavari,
1989) and included in category totals.                                                    •
0 Formulations for HASL process baths were unavailable because cleaner and microetch bath chemistries are not made
specifically for the HASL process. Hazards reported for HASL bath types were reported as the worst case of the results
of similar baths from other processes.                                             .
d For alternative processes with more than one product line, the hazard data reported represents the most hazardous bath
of each type for the two product lines (e.g., of the microetch baths from the two product lines, the one with the most
hazardous chemicals is reported).
Reactive - A chemical is considered reactive if it is readily susceptible to change and the possible
release of energy.  EPA gives a more precise definition of reactivity for solid wastes.  As defined
by EPA (40 CFR 261.23), a solid waste is considered reactive if a representative sample of the
waste exhibits any of the following properties: 1) is normally unstable and readily undergoes
violent change without detonating; 2) reacts violently or forms potentially explosive mixtures with
water; 3) when mixed with water, generates toxic gases, vapors, or fumes in a quantity sufficient
to present a danger to human health or the environment (for a cyanide or sulfide bearing waste,
                                              3-141

-------
3.5 PROCESS SAFETY ASSESSMENT
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 3-40. Also discussed in this section are surface finishing
chemical products that are potential eye or dermal irritants and suspected carcinogens.  The
following presents OSHA definitions for chemicals in these categories and discusses the data in
Table 3-40, where appropriate.

Sensitizer - A sensitizer is defined by OSHA [29 CFR 1910.1200 Appendix A (mandatory)] as a
chemical that causes a substantial proportion of exposed people or animals to develop an allergic
reaction in normal tissue after repeated exposure to the chemical.  Sixteen chemical products are
reported as sensitizers by MSDS data.

Acute and Chronic Health Hazards - As defined by OSHA (29 CFR 1910.1200 Appendix A), a
chemical is considered a health hazard if there is statistically significant evidence based on at least
one study conducted in accordance with established scientific principles that acute or chronic
health effects may occur in exposed employees. Health hazards are classified using the criteria
below:

•      acute health hazards are those whose effects occur rapidly as a result of short-term
       exposures, and are usually of short duration; and
•      chronic health hazards are those whose effects occur as a result of long-term exposure,
       and are of long duration.

Chemicals that are considered a health hazard include carcinogens., toxic or highly toxic agents,
reproductive toxins, irritants, corrosives, sensitizers, hepatotoxins, nephrotoxins, neurotoxins,
agents that act on the hematopoietic system, and agents which damage the  lungs, skin, eyes, or
mucous membranes.
                                          3-142

-------
                                                                   3.5 PROCESS SAFETY ASSESSMENT
   Table 3-40.  Sensitizer, Acute and Chronic Health Hazards, and Irreversible Eye Damage
                            Possibilities for Surface Finishing Processes
Surface Finishing
Process
'•f- a f '"
/
/!
HASLC

Nickel/Gold d
(2 product lines)




Nickel/Palladium/Gold






OSPd
2 product lines)
Immersion Silver

Immersion Tin d
2 product lines)


Bath Type
-
" V '* "" -~
•-1 * -
Cleaner
Microetch
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Cmmersion Gold
Cleaner
Microetch
Catalyst
Activator
Electroless Nickel
Electroless Palladium
Immersion Gold
Cleaner
Vficroetch
Cleaner
Microetch
Cleaner
Vficrbetch
Predip
Immersion Tin
Hazardous Property a>b - >:;
Sensitizer
j^

1(2)
2(3)
1(2)
1(2)
1(1)



1(4)
1(3)

1(3)



2(3)

1(3)
1(2)
1(2)

2(4)
Acute
Health
Hazard
1(1)
3(4)
1(1)
3(4)
2(3)
1(1)
2(2)
2(2)
1(1)
3(4)
2(3)
4(4)
3(3)
2(3)
1(2)
1(1)
3(3)
1(1)
2(3)
1(2)
1(2)

1(4)
Chronic
Health
Hazard
1(1)
3(3)
1(1)
2(2)
1(2)
1(1)
2(2)
2(2)
1(1)
1(4)
1(3)
2(4)
2(3)
1(3)
1(2)
1(1)
3(3) .
1(1)
2(3)
1(2)
1(2)

1(4)
Carcinogen


1(1)

1(1)


1(2)




2(4)

1(3)








1(1)
Irreversible
Eye Damage
o
1(1)
3(4)
1(1)
3(4)
1(2)
1(1)

1(2)

3(4)
1(3)
1(4)
2(3)
3(3)

1(1)
3(4)
1(1)
2(3)
1(2)
2(2)
1(1)
2(4)
              	~~~ "***& *ซ-i-"*ซ*ซ.  a vj. jsi v/uut/iL) xuwvtuig v^t^Ajtn. uoiujuiitwi iur uie 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 hi the bath were classified as
sensitizers per OSHA criteria, as reported by the products' MSDSs.
b Data'for pure chemicals (e.g., sulfuric acid) not sold as products were obtained from the Merck Index (Budavari
1989) and included in category totals.
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).
                                                 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 LARC, 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 (KIT); or 3) it is regulated by
OSHA as a carcinogen. A review of MSDS data indicates that seven chemical products are
reported as potential carcinogens, by either NTP, IARC, or EPA WOE Classifications.  Suspected
carcinogens include nickel sulfate, thiourea, and various lead compounds that are commonly used
in several processes. Suspected carcinogens are discussed in more detail in the human health and
ecological hazards summary, Section 3.3.

Dermal or Eye Irritant - An irritant is defined by OSHA [29 CFR 1910.1200 Appendix A
(mandatory)] as a chemical, that is not corrosive, but which causes a reversible inflammatory
effect on living tissue by chemical action at the site of contact.  A chemical is considered a dermal
or eye irritant if it is so determined under the testing procedures detailed in 16 CFR 1500.41- 42
for four hours exposure.  Table 3-40 does not include this term, because all of the surface
finishing chemical products are reported as either dermal or eye irritants.

Irreversible Eye Damage - Chemical products that, upon coming in contact with eye tissue, can
cause irreversible damage to the eye are required by OSHA to be identified as  such on the
product's MSDS. A review of MSDS data shows that 34 chemical products are reported as
having the potential to cause irreversible eye damage.

Other Chemical Hazards

       Surface finishing chemical products that have the potential to form hazardous
decomposition products are presented below.  In addition, chemical product incompatibilities with
other chemicals or materials are described, and other chemical hazard categories are presented.
The following lists OSHA definitions for chemicals in these categories and summarizes the MSDS
data, where appropriate.

Hazardous Decomposition - A chemical product, under specific conditions, may decompose to
form chemicals that are considered hazardous.  The MSDS data for the chemical products in the
surface finishing process indicate that over half of the products have the possibility of
decomposing to form potentially hazardous chemicals. Each chemical product should be
examined to determine its decomposition products so that potentially dangerous reactions and
exposures can be avoided.  The following are examples of hazardous decomposition of chemical
products that are employed in the surface finishing alternatives:
                                         3-144

-------
                                                         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 130 ฐF);
•      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 500 ฐF, poses several safety concerns, such as accidental
contact with the molten metal by workers, exposure to acids in the flux, and the potential for fire.

       Solder eruptions often occur during process startup as the solid solder is heated. Solder
melts from bottom to top,  and pressure may build up from thermal expansion causing the solder
to erupt. Splattering of the melted solder onto workers could cause serious burns. Caution
should be exercised during process startup to avoid worker injury. Heat resistant clothing, face
shields, protective aprons, long sleeve gloves, and shoes should be required when working around
the solder bath.

       Fire is possible at the solder bath and the exhaust/ventilation system, although it does not •
occur frequently. When fire occurs, small amounts of hazardous gases, such as hydrogen
chloride and carbon dioxide, can be released.  Causes of fire include the build-up of carbon
residual from the use of oil-based flux and other flammable materials kept too close to the
process.  Isolating flammable materials from the process area and regular cleaning of the HASL
machine will prevent a fire from occurring.

       Other safety concerns include workers exposed to small amounts of acid in the flux, lead
in the solder bath, and to process chemicals in the cleaner and microetch baths.  Risk from
exposure to process chemicals is addressed in detail in Section 3.4, Risk Characterization.  Like
other surface finishing processes, federal safety standards and regulations concerning the HASL
process can be found in CFR Title 29, Part 1910, and are available from the appropriate state
office.

3.5.3  Process Safety Concerns

       Exposure to chemicals is just one of the safety issues that PWB manufacturers may have
to address during their daily activities.  Preventing worker injuries should be a primary concern
for employers and employees alike. Work-related injuries may result from faulty equipment,
improper use of equipment, bypassing equipment safety features, failure to use personal protective
equipment, and physical stresses that may appear gradually as a result of repetitive motions (i.e.,
                                          3-146

-------
                                                        3.5 PROCESS SAFETY ASSESSMENT
 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 hi 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 staffer
 outside parties who  are familiar with the PWB manufacturing process and the pertinent safety
 concerns. The training should be held for each new employee, as well as periodic retraining
 sessions when necessary (e.g., when a new surface finishing process is instituted), or on a regular
 schedule.  The training program should inform the workers about the types of chemicals with
 which they work and the precautions to be used when handling or storing them, when and how
 personal protection equipment should be worn, and how to operate and maintain equipment
 properly.

 Storing and Using Chemicals Properly

       Because the  surface finishing process requires handling a variety of chemicals, it is
 important that workers know and follow the correct procedures for the use and storage of the
 chemicals.  Much of the use, disposal, and storage information about surface finishing process
 chemicals may be obtained from the MSDSs provided by the manufacturer or supplier of each
 chemical or formulation. Safe chemical storage and handling involves keeping chemicals in their
 proper place, protected from adverse environmental conditions, as well as from other chemicals
with which they may react. Examples of supplier recommended storage procedures found on the
MSDSs for surface finishing chemicals are listed below.

 •      store chemical containers in a cool, dry place away from direct sunlight and other sources
       of heat;
•      chemical products should only be stored in their properly sealed original containers and
       labeled with the common name of the chemical contents;
                                         3-147

-------
 3.5 PROCESS SAFETY ASSESSMENT
 •      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
ง to the PWB manufacturing industry. These standards address general safety and certification
 requirements (29 CFRPart 1910.132), the use of eye and face protection (Part 1910.133), head
 protection (Part 1910.135), foot protection (Part 1910.136), and hand protection (Part
 1910.138). The standards for eye, face, and hand protection are particularly important for the
 workers operating the surface finishing process where there is close contact with a variety of
 chemicals, of which nearly all irritate or otherwise harm the skin and eyes. In order to prevent or
 minimize exposure to such chemicals, workers should be trained in the proper use of personal
 safety equipment.

        The recommended personal protective equipment for a worker handling chemicals is also
 indicated on the MSDS. For the majority of surface finishing chemicals, the appropriate
 protective equipment indicated by the MSDS includes:

 •       goggles to prevent the splashing of chemical into the eyes;
 •       chemical aprons or other impervious clothing to prevent splashing of chemicals on
        clothing;
 •       gloves to prevent dermal exposure while operating the process; and
 •       boots to protect against chemical spills.
                                          3-148

-------
                                                        3.5 PROCESS SAFETY ASSESSMENT
        Additional personal protective equipment recommended for workers operating the HASL
 process includes:

 •       heat resistant gloves to prevent burns by accidental contact with molten solder; and
 •       face shield to protect face and eyes from solder splatter.

        Other items less frequently suggested include chemically resistant coveralls and hats. In
 addition to the personal protective equipment listed above, some MSDSs recommend that other
 safety equipment be readily available. This equipment includes first aid kits, oxygen supplies
 (SCBA), fire extinguishers, ventilation equipment, and respirators.

        Other personal safety considerations  are the responsibility of the worker. Workers should
 be prohibited from eating or keeping food near the surface finishing process. Because automated
 processes contain moving parts, workers should also be prohibited from wearing jewelry or loose
 clothing, such as ties, that may become caught in the machinery and cause injury to the worker or
 the machinery itself. In particular, the wearing of rings or necklaces may lead to injury.  Workers
 with long hair that may also be caught in the machinery should be required to securely pull their
 hair back or wear a hair net.

 Use of Equipment Safeguards

        In addition to the use of proper personal protection equipment for all workers, OSHA has
 developed safety standards (29 CFR Part 1910.212) that apply to the equipment used in a PWB
 surface finishing process. Among the safeguards recommended by OSHA that may be used for
•conveyorized equipment are barrier guards, two-hand trip devices, and electrical safety devices.
 Safeguards for the normal operation of conveyor equipment are included in  the standards for
 mechanical power-transmission apparatus (29 CFR Part 1910.219) and include belts, gears,
 chains, sprockets, and shafts. PWB manufacturers should be familiar with the safety requirements
 included in these standards and should contact their local OSHA office or state technical
 assistance program for assistance in determining how to comply with them.

        In addition to normal equipment operation standards, OSHA also has a lockout/tagout
 standard (29 CFR Part 1910.147). This standard is designed to prevent the  accidental start-up of
 electric machinery during cleaning or maintenance operations, and apply to the cleaning of
 conveyorized equipment as well as other operations. OSHA has granted an exemption for minor
 servicing of machinery, provided the equipment has other appropriate safeguards, such as a
 stop/safe/ready button that overrides all other controls and is under the exclusive control of the
 worker performing the servicing. Such minor servicing of conveyorized equipment can include
 clearing fluid heads, removing jammed panels, lubricating, removing rollers,  minor cleaning,
 adjustment operations, and adding chemicals. Rigid finger guards should also extend across the
 rolls, above and below the area to be cleaned. Proper training of workers is required under the
 standard whether lockout/tagout is employed or not. For further information on the applicability
 of the OSHA lockout/tagout standard to surface finishing process operations contact the local
 OSHA field office.
                                          3-149

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

       OSHAhas also developed standards (29 CFRPart 1910.95) that apply to occupational
noise exposure. These standards require protection against the effects of noise exposure when the
sound levels exceed certain levels specified in the standard. No data were collected on actual
noise levels from surface finishing process lines.
                                         3-150

-------
                                                                           REFERENCES
                                     REFERENCES

 ACGDB (American Conference of Governmental Industrial Hygienists).  1998.  1998 TLVs and
 BEIs. Threshold limit values for chemical substances and physical agents.  Biological exposure
 indices.  Cincinnati, OH.

 ATSDR (Agency for Toxic Substances and Disease Registry). 1990a. Toxicological Profile for
 Copper. U.S. Department of Health and Human Services, Public Health Service.  Atlanta, GA.

 ATSDR (Agency for Toxic Substances and Disease Registry). 1990b. Toxicological Profile for
 Silver. U.S. Department of Health and Human Services, Public Health Service, Atlanta, GA.

 ATSDR (Agency for Toxic Substances and Disease Registry). 1997a. Toxicological Profile for
 Nickel (Update). ATSDR, U.S. Department of Health and Human Services, Public Health
 Service, Atlanta.

 ATSDR (Agency for Toxic Substances and Disease Registry). 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.

 ATSDR (Agency for Toxic Substances and Disease Registry). 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 48:35-39.

 Aulerich, R.J., RK. Ringer, M.R. Bleavins, et al. 1982. Effects of supplemental dietary copper
 on growth, reproductive performance and kit survival of standard dark mink and the acute
 toxicity of copper to mink. J Animal Sci. 55:337-343 (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, p. 471-486.

 BASF Corporation.  1987. Results of several investigations with propionic acid with cover letter
 dated 100687.  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.

Budavari, S. (Ed.). 1989.  The Merck Index, llth ed.  Rahway: Merck & Co., Inc.
                                        3-151

-------
REFERENCES
Bureau of Labor Statistics. 1990. Statistical Summary: Tenure with Current Employer as of
January 1987.  As cited in EPA, 199 Ib.

Bureau of Labor Statistics. 1997. Job Tenure Summary. Labor Force Statistics from the current
Population Survey. Web site . Information
downloaded on 3/24/97.

Bushy Run (Bushy Run Research Center).  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 12/5/95. FYI-OTS-0790-
0692; FYI-OTS-0990-0323.

CDC (Centers for Disease Control and Prevention). 1991. Preventing lead poisoning in young
children. U.S. Department of Health and Human Services, Public Health Service, Atlanta, GA.

Clayson, D.B., F. Iverson, E.A. Nera and E. Lok.  1991. Title not available. MutatRes. 248:321
(cited in Katz and Guest, 1994, HSDB, 1998).
                                                             it
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. 9:687-697 (cited in U.S. EPA, 1988).

Franklin, C.A., M.J. Inskip, C.L. Baccanale, E.J. O'Flaherty, W.I. Manton, D.L. Schanzer, J.
Blenkinsop and C.M. Edwards. 1995. Transplacental transfer of lead in non-human primates
(Macacafascicularis): use of serially administered stable isotope tracers hi 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 15:194.

Froiman, Gail. 1996. U.S. Environmental Protection Agency (EPA). Personal communication to
Debbie Boger. April.

Gaul, L.E. and AJH. Staud. 1935. Clinical spectroscopy.  Seventy cases of generalized argyrosis
following organic and colloidal silver medication. J. Am. Med. Assoc. 104:1387-1390.  Cited in
IRIS, 1998 and U.S. EPA, 1991.

Goyer, R.A. 1990. Transplacental transport of lead.  Environmental Health Perspectives
89:101-105.

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. Stem.  1990.
Determinants of elevated blood lead levels during pregnancy in a population surrounding a lead
smelter Kosovo, Yugoslavia. Environmental Health Perspectives 89:95-100.
                                         3-152

-------
                                                                         REFERENCES
 Gupta, B.N., R.N. Khanna and K.K. Datta. 1979. Toxicological studies of ammonium sulfamate
 in rat after repeated oral administration. Toxicology 13:45-49.

 Hampshire Research Institute. 1995. Risk*Assistant™ Version 2.0 software.

 Harrison, P.T., P. Grasso and V. Badescu.  1991. Title not available.  FoodChem. Toxicio.
 29:367 (cited in Katz & Guest, 1994).

 Howard, P.H., R. S. Boethling, W.F. Jarvis, W.M. Meylan and E.M. Mchalenko.  1991.
 Handbook of Environmental Degradation Rates. Lewis Publishers, Chelsea, MI.

 HSDB (Hazardous Substance Data Bank).  1995. MEDLARS Online Information Retrieval
 System, National Library of Medicine, National Toxicology Program (via TOXNET), Bethesda,
 MD.

 HSDB (Hazardous Substance Data Bank).  1998. MEDLARS Online Information Retrieval
 System, National Library of Medicine, National Toxicology Program (via TOXNET), Bethesda,
 MD. June.

 IARC (International Agency for Research on Cancer). 1974. IARCMonographs on the
 Evaluation of the Carcinogenic Risk of Chemicals to Humans 7:95-109. IARC, Lyon, France.

 IARC (International Agency for Research on Cancer). 1985. Hydrogen Peroxide. In: IARC
 Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to Humans. Ally!
 Compounds, Aldehydes, Epoxides and Peroxides, Vol. 36. IARC, Lyon, France, pp. 285-314.

 IARC (International Agency for Research on Cancer). 1987. IARC Monographs on the
 Evaluation of the Carcinogenic Risk of Chemicals to Man.  Overall evaluations of
 carcinogenicity. An updating of Vols. 1 to 42. IARC, Lyon, France, p. 64.

 IARC (International Agency for Research on Cancer).  1990. Nickel and Nickel Compounds.
 Monographs on the Evaluation of the Carcinogenic Risks to Humans 49:257-445. IARC, Lyon,
 France.

 IARC (International Agency for Research on Cancer).  1992.  Occupational exposures to mists
 and vapours from suMuric 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 IARC Lvon, France on
 41-119.                                                            '            'PF"

 IARC (International Agency for Research on Cancer).  1993.  Cadmium and Cadmium
 Compounds.  IARC Monographs on the Evaluation of the Carcinogenic Risk of Chemicals to
Humans 58:119-237.  IARC, Lyon, France.
                                        3-153

-------
REFERENCES
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. Gannll: 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. 35:655-660.

Monsalve, E.R.  1984.  Lead ingestion hazard in hand soldering environments. Presentation for
the Eighth Annual Seminar, Solder Technology and Product Assurance, 22-23 February, 1984.
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
44:421-427 (cited in FDA, 1979).

NIOSH (National Institute for Occupational Safety and Health).  1976. A Guide to Industrial
Respiratory Protection. Cincinnati, OH: NIOSH, U.S. Department of Health and Human
Services. NEW Pub. 76-189.

NIOSH (National Institute for Occupational Safety and Health).  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.

NIOSH (National Institute for Occupational Safety and Health).  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. n, Part F. New York:  John
Wiley and Sons, pp. 4487-4505.

Pocock, S.J., A.G. Shaper, M. Walker, C.J. 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. /. Epi. Comm. Health 37:1-7.
                                        3-154

-------
                                                                           REFERENCES
 Pozzani, U.C. and C.P. Carpenter. 1954. Response of rats to repeated inhalation of
 ethylenediamine vapors. Arch. Ind. Hyg. Occup. Med. 9:223-226 (cited in U.S. EPA, 1998).

 RTECS (Registry of Toxic Effects of Chemical Substances).  1998. MEDLARS Online
 Information Retrieval System, National Library of Medicine.

 Robinson, R.B., C.D. Cox, N.D. Jackson and M.B. Swanson.  1997. "Estimating Worker
 Inhalation Exposure to Chemicals From Plating Baths at Printed Wiring Board Facilities."
 Presented at the 1997 Canadian Society of Civil Engineers - American Society of Civil Engineers
 Environmental Engineering conference, Edmonton, Alberto, Canada. July.

 Robinson, R.B., C.D. Cox and J. Drucker.  1999.  Prediction of Water Quality From Printed
 Wiring Board Processes.  Final report to the University of Tennessee Center for Clean Products
 and Clean Technologies and to the U.S. Environmental Protection Agency.  Part of the
 Verification of Finishing Technologies Project, EPA Grant X825373-01-2 (Amendment No. 2).
 August.

 Rodrigues, C., E.  Lok, E. Nera, F. Iverson, D. Page, K. Karpinski and D.B. Clayson. 1986. Title
 not available.  Toxicology 38:103 (cited in Katz & Guest, 1994).

 RTI (Research Triangle Institute).  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. 101:1431-1438.

 Sharp, John.  1999. Teradyne, Inc. Personal communication to Dipti Singh at U.S.
 Environmental Protection Agency (EPA). September.

 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, 1992. HSE-84-0072 (002334). 8EHQ-0492-3402 Mt.

 Shepard, T.H. 1986.  Catalog of Teratogenic Agents, 5th ed. The Johns Hopkins University
Press, Baltimore, p. 38 (cited hi HSDB, 1998).

 Sherlock, J.C., D.  Ashby, H.T. Delves, G.I. Forbes, M.R. Moore, WJ. Patterson, SJ. Pocock,
M.J. 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. 3:383-192.

Sleet, R.B. et al. 1989. National Toxicology Program, Report NTP 89-058.  NTIS PB89-
 165849 (cited in Gingell et al., 1994).
                                        3-155

-------
REFERENCES
Sollmann, T. 1921. Title not available. J. Pharmacol. Exp. Ther. 16:463-474 (cited in Katz,
Guest and Astffl, 1982).                                          ,

Stanek, Edward J., Edward J. Calabrese, Ramon Barnes and Penelope Pekow. 1997. Soil
Ingestion in Adults. Ecotoxicology and Environmental Safety 36:249-257.

SRC (Syracuse Research Corporation). Updated periodically. Environmental Fate Database
(EEDBฉ). 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 7,1991. 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. EPA (Environmental Protection Agency).  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. EPA (Environmental Protection Agency).  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. EPA (Environmental Protection Agency).  1987b.  Code of Federal Regulations.  40 CFR
50.12.

U.S. EPA (Environmental Protection Agency).  1988a.  Reportable Quantity Document for
Hydrogen Peroxide. EPA Office of Solid Waste and Emergency Response, Environmental
Criteria and Assessment Office, Cincinnati, OH, 9 p.

U.S. EPA (Environmental Protection Agency).  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. EPA (Environmental Protection Agency).  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. EPA (Environmental Protection Agency).  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.
                                        3-156

-------
                                                                        REFERENCES
 U.S. EPA (Environmental Protection Agency).  1991a. Chemical Engineering Branch Manual
for the Preparation of Engineering Assessments. EPA Office of Toxic Substances, Washington,
 D.C.

 U.S. EPA (Environmental Protection Agency).  1991b. Human Health Evaluation Manual,
 Supplemental Guidance:  "Standard Default Exposure Factors." EPA Office of Solid Waste and
 Emergency Response, Washington D.C.

 U.S. EPA (Environmental Protection Agency).  1991c. Drinking Water Health Advisory for
 Silver. Office of Water, Washington, D.C. NTISNo. PB92-135516.

 U.S. EPA (Environmental Protection Agency).  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. EPA (Environmental Protection Agency.  1992b.  Guidelines for Exposure Assessment.
 Washington, D.C. EPA 600-2-92-001.

 U.S. EPA (Environmental Protection Agency).  1994. Guidance Manual for the Integrated
 Exposure Uptake Biokinetic Model for Lead in Children.  EPA/540/R-93/081, and IEUBK
 Model, version 0.99D.

 U.S. EPA (Environmental Protection Agency).  1995. Supplemental Guidance to RAGS: Region
 4 Bulletins, Human Health Risk Assessment (Interim Guidance). Waste Management Division,
 Office of Health Assessment.       .                            .

 U.S. EPA (Environmental Protection Agency).  1996a. Recommendations of the Technical
 Workgroup for Lead for an Interim Approach to Assessing Risks Associated with Adult
 Exposures to Lead in Soil. US EAA Technical Workgroup for Lead, December.

 U.S. EPA (Environmental Protection Agency).  1996b. 'Proposed Guidelines for Carcinogen
 Risk Assessment." EPA Office of Research and Development, Washington, D C  EPA/600/P-
 92/1003C.                      .

U.S. EPA (Environmental Protection Agency).  1997a. Exposure Factors Handbook. EPA
 Office of Research and Development, Washington, D.C. EPA/600/p-95/1002Fa.

U.S. EPA (Environmental Protection Agency).  1997b. Health Effects Assessment Summary
Tables. FY_1997. Office of Research and Development, Office of Energy and Remedial
Response.                                                                • ',       '

U.S. EPA (Environmental Protection Agency).  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.
                                       3-157

-------
REFERENCES	                    .

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. EPA744-R-98-003. August.

U.S. EPA (Environmental Protection Agency). 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. EPA744-R-97/002a. June.

WHO (World Health Organization). 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]).

WHO (World Health Organization). 1986. Regional Office for Europe: Air Quality Guidelines.
Vol. 11. Geneva, Switzerland: World Health Organization, 1-34.
                                        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 COMPETITIVENESS
       85/85:  85ฐC and 85% relative humidity for three weeks),
       thermal shock (TS): 200 cycles between -50ฐC and 125ฐC, and
•      mechanical shock (MS):  dropped 25 tunes 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 firorn fabrication and assembly process residues and the electrical anomalies. In
addition, the boards were inspected visually to identify any obvious anomalies or defects.  The
results indicated that the failures were not a result of residue, and that solder cracking was the
most common visual defect. HASL had more solder cracks than the other finishes.

4.1.2  Performance Demonstration Methodology

       The general plan for the performance demonstration was to collect data on alternative
surface finishing processes, during actual production runs, at sites where the processes were
already in use. These demonstration sites were production facilities, customer testing facilities
(beta sites), or supplier testing facilities. Whenever possible, production facilities were used.
Each demonstration site received standardized test boards, which were run through the surface
                                           4-2 .

-------
                                                                    4.1 COMPEHTlVEJNESS
 finishing operation during their normal production. The information collected through the
 demonstrations was intended to provide a "snapshot" of the way the technology was performing
 at that particular site at that particular time.  This methodology was developed by consensus with
 the technical workgroup, which included suppliers, trade association representatives, EPA, and
 PWB manufacturers.  A detailed performance demonstration methodology is included in
 Appendix F.                                                          ,

       Each supplier was asked to submit the names of up to two facilities at which the
 demonstrations of their technology were to be conducted.  This selection process encouraged the
 suppliers to nominate the facilities where the technology was performing at its best. This, in turn,
 provided for more consistent comparisons across technologies. The demonstration sites included
 13 facilities, at which 16 different demonstrations were run.

       To minimize differences in performance due to processes other than surface finishing, the
 panels used for testing were all manufactured at one facility, Network Circuits, in Irving, Texas.
 After fabrication, the panels were numerically coded for tracking purposes, and six panels
 (containing  four boards per panel), were shipped to each demonstration site, where the appropriate
 surface finish was applied.

       An observer from the DfE project team was present at each demonstration site to monitor
 the processing of the test panels. Observers were present to confirm that the processing was
 completed according to the methodology and to record data. Surface finish application at each
 demonstration site was completed within one day, while performance processing at all sites was
 completed over a three-month period.

       When the processing was completed, the panels were put into sealed bags and shipped to
 a single facility, which acted as a collection point for all performance demonstration panels.
 Completed panels were then shipped back to Network Circuits, where the panels were cut into
 boards.  All  coded boards were then shipped to a single facility (EMPF/American Competitiveness
 Institute) for assembly. One subgroup was assembled using low-residue (LR) flux and the other
 with water soluble (WS) flux.

       Following assembly, the boards were sent to Raytheon Systems Inc. in McKinney, Texas,
 where'the performance characteristics of the assembled boards were tested. Testing included   '
 Circuit Electrical Performance and Circuit Reliability testing.  The Electrical Performance testing
 assessed the circuit performance of the printed wiring assemblies (PWAs), or assembled PWBs,
before and after exposure to 85 ฐC 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 COMPETITIVENESS
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, and to
mitigate as much risk as possible in process change.  The PWA and the test/data analysis
methodology are considered excellent discriminators in comparing processes fluxes, surface
finishes, or other process technologies. It should be noted that circuit technology continues to
change rapidly; the PWA design is based on 1994 technology and does not incorporate some
more recent, state-of-the-art, circuitry. It is unknown whether the results might have differed
with newer circuit technology.  However, the test PWA was designed to contain approximately
80 percent of the circuitry used in military and commercial electronics. In addition, use of this
test vehicle by the DfE PWB Project provided great savings in cost and time that would be
required to develop a new test vehicle, and allows for some comparison with CCAMTF results;

       The test vehicle was designed to be representative of a variety of extreme circuits: high
voltage, high current, HSD, low-leakage current, and high frequency (HF) circuits. A designer
can use the resulting measurements to make some analytical judgments about the process being
tested.  The test PWA was not intended to be a "production" board, which would typically be too
narrow in breadth to represent a wide variety of these circuit extremes.  Even though some
technology complexities/advancements are not duplicated, the basic types are represented, and
comparison of baseline technologies can be extrapolated, in some cases, to  more current
technology by analysis.  The performance results are assessed based on the acceptance criteria
developed by the CCAMTF project, which are described in Table 4-1.

       The test PWA measures 6.05" x 5.8" x 0.062".  See Appendix F for more details on the
design of the test PWA.  The PWA is divided into six sections, each containing one of the
following types of electronic circuits:
       HCLV;
       HVLC;
       HSD,
       HF;
       SW;and
       other networks.
                                           4-4

-------
                                                                   4.1 COMPETITIVENESS
        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  COMPETITIVENESS
Table 4-1. Electrical Responses for the Test PWA and Acceptance Criteria
Electrical
Response
Circuitry
f „ <•
Acceptance Criteria
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
HVLC PTH
HVLC SMT
4uA 7.7 Iog10 ohms
Resistance^ 7.7 Iog10 ohms
Resistance > 7.7 Iog10 ohms
Resistance > 7.7 Iog10 ohms
Stranded Wire
22
23
Stranded Wire 1
Stranded Wire 2
Change in voltage from pre-test < 0.356V
Change in voltage from pre-test < 0.356V
HCLV = high current low voltage; HF = high frequency;
PGA = pin grid array; PTH = plated through hole; SMT
HSD = high speed digital; HVLC = high voltage low current;
= surface mount technology; TLC = transmission line coupler.
                                                4-6

-------
                                                                    4.1 COMPETITIVENESS
       These surface finishes were applied at one or more of the different demonstration sites.
 Table 4-2 provides a summary of the 164 PWAs that were subjected to environmental testing by
 surface finish, manufacturing site, and flux type. Table 4-2 also shows that both fluxes were not
 used with all demonstration sites, and that 84 PWAs were processed with low residue flux, while
 80 PWAs were processed with water soluble flux.
Table 4-2. Distribution of the Number of LRSTF PWAs by Surface Finish, Site, and Flux
Surface Finish
HASL


NickeyGoId


Nickel/PaUadium/GoId
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 effluxes during assembly, the number of PWAs are
different for each surface finish^ as follows:
                     Surfece 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 COMPETmVENESS
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 ฐC and
       125 ฐ C with 3 0 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

Genera! Linear Models

       General linear models (GLMs) were used to analyze the test data for each of the 23
electrical circuits in Table 4-1 at each test time. The GLM analysis determines which
experimental factors or combinations of factors (interactions) explain a statistically significant
portion of the observed variation in the test results, and in quantifying their contribution.

Analysis of Variance and Multiple Comparisons of Means

       Another statistical approach can be used to determine which groups of site/flux means are
significantly different from one another for a given electrical response from the test PWA This
procedure begins with an analysis of variance (ANOVA)  of the test results (Iman, 1994) for a
given circuit.  An ANOVA is perhaps best explained via an example.

       An ANOVA performed on the 164 pre-test measurements for HCLV PTH produced the
following:
                                          4-8

-------
                                                                  4.1 COMPETITIVENESS
Source
Site/Flux
Error
Total
DF
22
141
163
Sum of Squares
0.2908
2.6796
2.9704
Mean Square
0.0132
0.0190

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 - 1) '=22
 Error        Total DF - Site/Flux DF = (164 - 1) - (23 -1) = 163 - 22 = 141
 Total         164-1 (the number of test measurements -1) = 163

       Sum of Squares. The entries in this column are the sums of squares associated with each
 source of variation.  The Total Sum of Squares is calculated by summing the squares of the
 deviations of the 164 data points from the sample mean.  If this number were divided by 164 - 1,
 the result would be the usual sample variance (i.e.,s2 = 2.9704/163 = 0.0182).  The other sums
 of squares in this column represent a partitioning of the total sum of squares. Note that they sum
to the total sum of squares:

                               0.2908 + 2.6796  = 2.9704

       The calculations for these other sums of squares are somewhat more involved than the
total sum of squares  and will not be discussed here. The interested reader can find details of these
calculations in Iman, 1994.

       Mean Square. The values in this column are obtained by dividing the sum of squares in
each row by their respective degrees of freedom:

                     Mean Square for Site/Flux  = 0.2908/22 =  0.0132
                     Mean Square for Error  = 2.6796/141 = 0.0190
                                         4-9

-------
4.1 COMPETITIVENESS
       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(F2SUl41>FI5WHll?) =  Prob(FA141 > 0.70) = 0.838

       Whenever a p-value in this analysis is less than 0.01, the corresponding source of variation
can be regarded as making a significant contribution to the overall variation. In this example the
p-value is quite large, which signifies that Site/Flux does not make a significant contribution to the
overall variation 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 hi 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 LSD.,, which is defined as:
                                                   ^
-l+J-
n.  n.
                                           4-10

-------
                                                                  4.1 COMPETITIVENESS
where,
a            =•     level of significance
t            =     the a/2 quantile from a Student's t distribution with n-k degrees of freedom
MSB      ., =     mean square error for the model
ii; and rij      =     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
SiteMux ,
Combination
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Surface Finish
'
HASL
HASL
HASL
HASL
OSP
OSP
OSP
OSP
OSP
Immersion Tin
Immersion Tin
Immersion Tin
Immersion Tin
Immersion Tin
Immersion Silver
Immersion Silver
Immersion Silver
Nickel/Gold
Nickel/Gold
Nickel/Gold
Nickel/Gold
Nickel/Palladium/Gold
Nickel/Palladium/Gold
HuxType
t
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
LR = low residue; WS = water soluble.
                                        4-11

-------
4.1 COMPETITIVENESS
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
(Xt2s) or lower 25 percent of the sample data.  The right-hand side of the box represents the upper
quartile (X-75), 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 (X50).

       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 X7S to X7S+1.5 IQR.  This line never extends beyond X7S + 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 X2S -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.
                       Lover
                       Qaartile
                                      Median
Upper
Quartile
                                               X.
                                                  75
                         A Boxplot Used to Display Test Results
                                          4-12

-------
                                                                    4.1 COMPETITIVENESS
4.1.6  Overview of Test Results

       The 164 PWAs as summarized in Table 4-2 were functionally tested at the following four
times:
       Pre-test;
       Post-85/85;
       Post-TS; and
       Post-MS.
       At each of these test times, 3,772 electrical test measurements were recorded (164 PWAs
x 23 individual circuits). An overall summary of success rates based on 3,608 measurements1 at
each test time is shown in Table 4-4.

              Table 4-4.  Number of Anomalies Observed at Each Test Time
Test Time
Pre-test
Post-85/85
Post-TS
Post-MS
Anomalies
2
17
113
527
Success Rate
99.9%
99.6%
96.9%
85.4%
              MS = mechanical shock.; TS = thermal shock.
       An overview of the test results at each test time is discussed in this section. A discussion
of the results for each test time for each major circuit group is presented in Sections 4.1.7 through
4.1.13. An overview of the circuits meeting the acceptance criteria after each testing sequence is
summarized in Table 4-5 for each major circuit group.
    Since HF TLC RNF gave a constant response of 50MHz throughout, there is no variability to analyze. •
                                          4-13

-------
4.1 COMPETITIVENESS
    Table 4-5.  Percentage of Circuits Meeting Acceptance Criteria at Each Test Time 2
..•;•. Circuitry.1 •;%,;'- ,^i
HCLV
HVLC
HSD
HFLPF
HFTLC
Other Networks
SW
Totals
^-wfl^85/85^;;;l|5|!f:
100%
99.7%
99.7% .
98.7%
99.8%
99.8%
100%
99.5%
S^^EKan^Si^cCi?!!
100%
99.7%
98.8%
89.4%
99.5%
100%
99.7%
96.9%
'iM MecJlpScaii^liftc^j'J ^
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%
HCLV — high current low voltage; HF LPF = high frequency low pass filter; HF TLC = high frequency transmission
line coupler; HSD = high speed digital; HVLC = high voltage low current; SMT = surface mount technology;
SW = stranded wire.
Overview of Pre-Test Results

       The electrical measurements were compared to the acceptance criteria given in Table 4-1
at each test time. Note that the acceptance criteria require a comparison to pre-test results for all
but six of the 23 electrical circuits (#'s 3, 4, 18-21 in Table 4-1).  Hence, pre-test comparisons to
the acceptable criteria can only be made for those six circuits. There were no pre-test anomalies
observed for those six circuits.  Pre-test measurements for the remaining 17 circuits were
compared to CCAMTF pre-test results. Table 4-6 presents this comparison of the ranges of the
measurements for each of the 23 circuits with pre-test measurements for the PWAs used in the
CCAMTF 85/85 testing.

       Table 4-6 shows that the two sets of ranges for circuits 5 through 12 and 16 do not even
overlap.  The lack of overlap in the ranges for the HSD PTH and HSD SMT circuits (#'s 5 and 6)
is due to different components being used on the DfE PWAs than were used in processing the
PWAs in the CCAMTF program. The differences in the HF LPF circuits 7 through 12  are more
difficult to pinpoint. The most likely explanation lies in the fact that the actual boards used in the
DfE program and those in the CCAMTF PWAs were produced by two different manufacturers.
FR-4 epoxy was used for the board laminate material.  HF LPF responses are sensitive to the
dielectric constant of the board laminate material. Differences in FR-4 epoxy at the two
manufecturing locations used by the DfE program and the CCAMTF program could have affected
the dielectric constant and hence the HF LPF responses. Another possibility is that the board
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
     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 ah anomaly in one
test but meets the acceptance criteria in the subsequent test.
                                           4-14

-------
                                                                     4.1 COMPETITIVENESS
 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.
*• s f-
Circuit [units] '
'CAMTFPre-Test
Mia Max
1 HCLVPTH[V] 6.60 7.20
2 HCLVSMT[V] 6.96 7.44
. 3 HVLCPTH[uA] , 5.00 5.25
4 HVLC SMT [uA] 4.92 4.97
DfEPre-Test
Mm
6.80
7.00
5.00
4.81
5 HSDPTH Propagation Delay fo sec] 12.66 13.50
6 HSD SMT Propagation Delay [u. sec] 4.28 5.45
7 HF PTH 50MHz [dB] -0.320 0.094
8 HFPTHf(-3dB)[MHZJ 239.4 262.6 '•
9 HFPTHf(-40dB)[MHZ] 425.3 454.9 !
Max
7.52
7.44
5.25
5.39
SKf$iifiyj&33ป
s8Sias8mW*sgi




10 HF PTH 50MHz [dB] • -0.296 0.081
11 HFSMTf(-3dB)[MHZ] 275.0 283.3 {
12 HF SMT f (-40dB) [MHZ] . 642.6 674.0 ;
13 HFTLC 50MHz Forward Response [dB] -49.74 -36.48
14 HFTLC 500MHz Forward Response [dB] -21.47 -17.54
15 HFTLC IGHz Forward Response [dB] -16.91 -12.08


-50.87
-19.91
-15.01





-42.66
-15.28
-12.89
16 HFTLC Reverse Null Frequency [MHZ] 624.2 659.8
17 HFTLC Reverse Null Response [dB] -74.53 -38.22
. 18 10-mil Pads [Iog10 ohms] 10.01 15.00
19 PGA-A [logw ohms] 8.94 15.00
20 PGA-B[log10ohms] 8.72 15.00
21 Gull Wing [Iog10 ohms] , 9.71 14.00
22 SW 1 [mV]
23 SW2[mV]
5 19
19 28
-43.67
10.10
10.38
10.07
9.01
7
19
-32.08
15.00
-14.00
13.70
13.70
19
28
         ,_,            	ey~?	  	o~ ~~-J ป-•••ซ••'j 9 *-f.**j-,  *ij.^u U|^VVVA vu.giLCU9 J.J. V J_i\^ — lllgll VLUlOgC i
 PGA = pin grid array; PTH = plated through hole; SMT = surface mount technology; SW = stranded wire;
 TLC = transmission line coupler.
        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
•DfE PWAs occurred at the beginning of the curve, which is approximately 50MHz.  In fact, all
                                           4-15

-------
4.1 COMPETITIVENESS
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
uncertainty to a moot point.  As discussed further in subsequent sections, none of the
discrepancies could be attributed to the performance of the surface finishes.

Overview of 85/85 Results

       At the conclusion of the 85/85 test, 99.5% of the electrical measurements met the
acceptance criteria given in Table 4-1. There were 17 anomalies distributed across 10 PWAs, as
shown in Table 4-7. Among the PWAs with anomalies, five were assembled with the low-residue
flux and five were assembled with the water-soluble flux. The anomalies are summarized in
Appendix F, Table F-l. Table F-l also contains observations made by the testing technician that
are useful in identifying the source of the anomaly for those cases where a problem was obvious,
such as an open PTH, a burnt etch, or a failed device.

  Table 4-7. Frequency Distribution of Post-85/85 Anomalies per PWA by Surface Finish
                           (Sample sizes are given in parentheses)
Number of
Anomalies per
PWA
None
1
2
3
Total Anomalies
HASL
(32)
31


1
3
Nickel/Gold
(28)
26
2


2
Nickel/Palladiiim/Gold
(12) " ' ".
12 .



0
OSP
(36)
35


1
3
Immersion
Silver
<20)
18
1

1
4
Immersion
Tin
(36)
32
3
1

5
 Overview of Thermal Shock Results

       The number of anomalies increased from 17 at the post-85/85 test to 113 at the post-TS
 test, so that 96.9% of the electrical measurements met the acceptance criteria given in Table 4-1.
 Of the 17 anomalies at post-85/85,16 carried over to post-TS, so that the thermal shock test
 introduced 97 new anomalies. 91% of the post-TS anomalies occurred for HFLPF 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.

       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.
                                          4-16

-------
                                                                  4.1 COMPETITIVENESS
       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
Mckel/Palladium/Gold
(12)
11
1





1
OSP
(36)
28
2
3
2


1
20
Immersion
Silver
(20)
11
2

3
1

3
33
Immersion
Tin
(36)
22
3
3
5
3


36
 Overview of Mechanical Shock Results

       The number of anomalies increased greatly from 113 at post-TS to 527 at post-MS. 85%
 of the electrical measurements met the acceptance criteria given in Table 4-1. Of the 113
 anomalies at post-TS, 97 carried over to post-MS, hence the mechanical shock test introduced
 430 new anomalies. Theses new anomalies included 157 from HCLV SMT — in contrast, there
 was only one HVLC SMT anomaly at post-TS. In addition, there were 163 new anomalies from
 the HVLC SMT circuits. Thus, these two circuits accounted for 320 of the 430 new anomalies.
 The anomalies for these two SMT circuits were attributable to SMT components coming off the
 board during the execution of the mechanical shock test.  This affected every board and has no
 relation to site, surface finish, or flux.

       All anomalies, except for those associated with HCLV SMT and HVLC SMT (since these
 affected every PWA), are summarized in Appendix F, Table F-3. In addition, this table includes
 comments made by the test technician. There were five minor stranded wire anomalies that are
 not listed in Table F-3.

       At post-MS, every PWA had at least one anomaly.  Table 4-9. provides a breakdown of
 the number of anomalies per PWA for each surface finish. The last row in this table gives the
 median number of anomalies per PWA for each surface finish. The hypothesis that the mean
 number of anomalies is the same for all surface finishes is easily rejected with a p-value of 0.000
 based on the Kruskal-Wallis test (Iman, 1994). Immersion silver has the most anomalies per
 PWA with nickel/gold and nickel/palladium/gold having the least. HASL and OSP had
 approximately the same number of anomalies, with immersion tin slightly higher than these two.
The following section provides insight on the source of the anomaly disparities relative to surface
finish.      •->..'                                  •
                                         4-17

-------
4.1 COMPETITIVENESS
   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)
" ., t •*


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  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 COMPETITIVENESS
                        Table 4-10. P-Values for HCLV Test Results
 HCLV = high current low voltage; MS = mechanical shock; PTH = plated through hole; SMT = ^ace monat
 technology; TS = thermal shock.
        Boxplot Displays of Multiple Comparison Results. Boxplot displays provide a
 convenient way to display multiple comparison results. Multiple comparison procedures are only
 justified for HCLV PTH at post-MS since the other f-statistics were not significant. However,
 boxplot displays are given in Figure 4-1 to 4-8 for all HCLV circuits for purposes of comparison.
 Figures 4-1 to 4-4 display the test results at each test time for HCLV PTH circuits and Figures 4-
 5 to 4-8 do the same for HCLV SMT circuits. For improved readability, all boxplots referenced
 in this chapter can be found in Section 4.1.16 at the end of the performance results discussion.
 Additional boxplots, where findings were not significant, can be found in Appendix F.

        Some explanation of the contents of each graph of boxplots should facilitate
 understanding. The test time and circuit type are labeled in the upper left-hand corner of each
 boxplot display. The numbers (1 to 23) on the horizontal axis in each figure correspond
 respectively to the 23 site/flux combinations listed in Table 4-3.  The label WS on the horizontal
 axis signifies those demonstration sites for which water soluble flux was used; otherwise,  the flux
 type was low residue (LR). The boxplots are grouped by surface finish, which are identified with
 labels across the top of each graph.  At pre-test, the vertical axis corresponds to the absolute test
 measurement. After pre-test, the vertical axis either corresponds to the absolute test
 measurement or the difference from the pre-test measurement as specified in the acceptance
 criteria.  The sample mean is identified in each boxplot with a solid circle

       Note that there is a lot of overlap in all boxplots in Figure 4-1, which is consistent with the
 lack of significance in the f-statistics for equality of means and in the results for the GLMs. Also
 note that the total variation in the boxplots is approximately 0.3V, which most likely is not of
 concern. Figures 4-2 to 4-4 display the differences between the current HCLV PTH
 measurements and those obtained at pre-test. Note that all differences in Figures 4-2 and 4-3 are
 well below the acceptance criteria of AV < 0.5V. However, several of the differences are well
 above the acceptance criteria following mechanical shock, as illustrated in Figure 4-4. The
 significant difference in means in Figure 4-4 at post-MS is attributable mostly to immersion silver
 at Site 17 processed with a water soluble flux. It should be noted, however, that the other two
immersion silver sites showed no anomalies. This may indicate a site-specific problem and not a
surface finish problem.  Additional failure analysis would be needed to draw further conclusions.
                                          4-19

-------
4.1 COMPETITIVENESS
       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 AY" < 0.50V after exposure to 85/85 or .thermal shock. However,, following
mechanical shock there were 12 HCLV PTH anomalies and 158 HCLV SMT anomalies. Whereas
the HCLV SMT anomalies affected almost every PWA, the HCLV PTH anomalies were
distributed unevenly among surface finishes, as shown in Table 4-11..

         Table 4-11.  Number of HCLV PTH Anomalies at Post-Mechanical Shock
                                   by Surface Finish
Surface Finish
HASL
Nickel/Gold
Nickel/Palladium/Gold
OSP
Immersion Silver
Immersion Tin
Anomalies
1
0
0
3
5
3
NoJofPWAs
32
28
12
36
20
36
 4.1.8  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 paraEel design, the HVLC resistors
 were in a series circuit design.  Thus, when one resistor is missing the circuit is open.
                                         4-20

-------
                                                                    4.1 COMPETITIVENESS
                                                                               /
        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
                                                                sVMifc for HVEC SM3?
 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/iA.  These
 boxplots are centered close to 5fj.A, and the total spread is on the order of 0.02/M. 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 feted in Table 4-1 (responses 3 and 4).  All HVLC PTH circuits met the
 acceptance criteria of 4/uA and 6,aA for the entire sequence of tests. Only one HVLC SMT
 current measurement foiled 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 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 COMPETITIVENESS
       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
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 dekys
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 foiled. 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 Circuitry Performance Results

       Pre-test measurements for all HF LPF circuits were subjected to GLM analyses, as were
the deltas after 85/85, thermal shock, and mechanical shock. The results of the GLM analyses are
given in Tables F-10 to F-15.  The GLM analyses indicate that the parameters under evaluation
(site, surface finish., or flux) do not influence the HF LPF measurements.  The same is true at
post-85/85, post-TS and post-MS. However, the test measurements contained many extreme
outlying observations at both of these later two test times, which greatly increases the sample
variance and in turn hinders the interpretation of the GLM results. As indicated in Tables F-l,
F-2, and F-3 there were many anomalous HF LPF test measurements (171 at po^st-MS). The
principal source of these outliers was open PTHs, is discussed in more detail under Comparison
to Acceptance Criteria.
                                          -—-

-------
                                                                     4.1 COMPETITIVENESS
 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
 Pre-test
0.002
 Post-85/85
0.000
 Post-TS
0.004
 Post-MS
0.002
0.001
0.000
                                                                                0.000
 HF = high frequency; LPF = low pass filter; MS = mechanical shock; PTH = plated through hole; SMT = surface mount
 technology; TS = thermal shock.                                                        [
       These results are discussed separately for each of the six HF LPF circuits.  Boxplot
displays of all test results for HF LPF circuits have been created to aid in the interpretation.  Only
the boxplots showing statistical and practical significance are shown here (Figures 4-9 to 4-15);
the rest are in Appendix F.

       HF LPF PTH 50MHz. While the p-values for the associated f-statistic were highly
significant at all test times, Figure 4-9 identifies the source of this significance at pre-test, where
the responses for nickel/gold applied at Site 18 and subsequently processed with low residue flux
are much lower than the others. Post-85/85 and post-TS results indicate just the opposite for this
demonstration site (see Figures F-l 1 and F-12).  What occurred is that the problem circuit
returned to normal at post-85/85 and post-TS, but those measurements were then compared to
their low pre-test measurements, which caused the differences to be large in the positive direction.
Hence, the significance at post-85/85 and post-TS are an artifact of the pre-test measurements
and should most likely be ignored as the circuit performance was in line with all others. More
importantly, the significant differences at pre-test are too small to be of practical concern. The
range depicted in Figure 4-9 is approximately 0.7dB and the acceptance criterion allows a change
of ฑ5dB.  On the other hand, Figure 4-10 is of concern as several of the surface finishes have
measurements well below the lower bound acceptance criterion of-5dB. In particular, -one of the
five OSP PWAs, two of the three immersion silver PWAs, and one of the five immersion tin
PWAs. This circuit had 15 anomalies at post-MS.

       HF LPF PTH f(-3dB). Figure 4-11 shows the boxplot for the HF  LPF PTH f(-3dB)
circuit at the post-MS test time. Boxplots for the other three test times can be found in Appendix
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
                                           4-23

-------
4.1 COMPETITIVENESS
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 ฑ50MHz. 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 tunes 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-rTS are too small to be of practical concern
relative to the acceptance criteria of ฑ5dB. 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 ฑ50MHz. It is noteworthy that neither
nickel/gold or nickel/palladium/gold had any anomalies. This circuit had 29 anomalies at post-
MS.

       HF LPF SMT f(-40dB). Figure 4-15 shows the boxplot for the HF LPF SMT f(-40dB)
circuit at the post-MS test time. Boxplots for the other three test times can be found in Appendix
F, since the p-values for the associated f-statistic were not significant, except at post-MS. This
circuit had the most anomalies (65) at post-MS. Some of the anomalies may be due to the high
variability  in the frequency when measured at -40dB. Figure 4-15  shows notable variation in the
magnitude of the differences (note the vertical scale). Most cases are well outside the acceptance
criterion of ฑ50MHz. Nickel/gold and nickel/palladium gold are again noteworthy as they have
very few anomalies.

Comparison to Acceptance Criteria

       The acceptance criteria for the sk HF LPF circuits are shown I Table 4-1 (responses 7
through 12). Thirteen of 984 HF LPF test measurements did not meet the acceptance criterion
after exposure to 85/85 (see Table F-l). These 13 responses occurred on six PWAs, with 12 of
the 13 occurring with PTH components. After exposure to thermal shock, the number of HF LPF
anomalies  increased to 104 (see Table F-2). Thirteen of these 103 HF LPF anomalies carried over
                                         4-24

-------
                                                                   4.1 COMPETITIVENESS
 from the 85/85 test. At post-MS, the number of anomalies increased to 171 with 97 carrying over
 from thermal shock.

        PWAs with HF LPF anomalies generally have multiple anomalies.  This can be seen in
 Table 4-15, which shows the frequency distribution of the number of HF LPF anomalies per PWA
 at post-MS (see Tables F-l to F-3).

       Table 4-15. Frequency Distribution of HF LPF Anomalies at Post-Mechanical
                                    Shock per PWA
No. of HF LPF Anomalies per PWA
at Post-Mechanical Shock
None
1
2
3
4
5
6
Frequency
90
36
5
20
4
5
4
       The test technician comments indicate that most of the HF LPF anomalies were due to an
open PTH, which affects both PTH and SMT. To explain further, a circuit board consists of
alternating layers of epoxy and copper through which a hole is drilled during fabrication. This via
is plated with a very thin layer of electroless copper to provide a "seed bed" for the primary
coatings. Copper is then electroplated over the electroless copper strike. The final surface finish
(HASL, OSP, etc.) is then applied. Failure to make an electrical connection between the copper
etches on the opposite sides of the board is known as an open PTH. The opens occurred in very
small vias in the HF LPF circuit. Small vias can be very difficult to plate. Opens were present
during in-circuit testing and at pre-test.  In some cases, a z-wire was inserted through the via to
make an electrical connection between the etches on the opposite side of the board.  It appears
that test conditions may accelerate the problem.

      Although an open PTH is a fabrication issue, there does appear to be a relationship with
surface finish. The HF LPF anomalies are summarized by surface finish in Table 4-17 for each of
the sk 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 COMPETITIVENESS
     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 INpn-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
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.oฃ
PWAs
HASL
Nickel/Gold
Nickel/Palladium/Gold
OSP
Immersion Silver
Immersion Tin
Totals

32
28
12
36
20
36
164
p-value
50MHz
1 (2.9)
2 (2.6)
0(1.1)
2 (3.3)
6 (1.8)
4(3.3)
15

>HFLPF
PTH
f(-3dB)
2(4.1)
3 (3.6)
0(1.5)
2(4.6)
6 (2.6)
5(4.6)
18
XfcJSsirfiSiSfei
sr?M*?%feSS
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
HF = high frequency; LPF = low pass filter; PTH = plated through hole; SMT = surface mount technology.
                                          4-26

-------
                                                                   4.1 COMPETITIVENESS
        Such is not the case for the last four HF LPF circuits listed in Table 4-17, where the
 p-values at the bottom of the table indicate that the anomalies are not independent of surface
 finish.  The expected values for anomalies appear in parenthesis in each cell hi that table.  These
 comparisons show:

 •      HASL anomalies are close to the expected values throughout.
 •      Nickel/gold has far fewer anomalies than expected.
 •      Nickel/palladium/gold has far fewer anomalies than expected.
 •      OSP anomalies are close to expected, except for the last column, where they have more
        anomalies than expected.
 •      Immersion silver has many more anomalies that expected for all circuits.
 •      Immersion tin anomalies are close to expected for PTH circuits, but are higher than
        expected for SMT circuits.

        The number of open PTH anomalies may be related to the inherent strength of the metals.
 Tin and silver are relatively weak; OSP has no metal, while nickel makes the PTH stronger. To
 determine the relevancy of metal strength to the open PTH anomalies, the HF LPF circuits would
 need to be subjected to failure analysis to check for copper plating thickness and PTH voids in the
 vias,  as both of these may be problems in small vias.  In addition, the chemical removal of copper
 from the via may be much greater in immersion tin and immersion silver, depending on how they
 were processed.

 4.1.11  High Frequency Transmission Line Coupler 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 hi Appendix F hi Table F-22.  These predictions are consistent with those in Table F-21,
                                          4-27

-------
4.1 COMPETITIVENESS
and show that immersion tin and immersion silver are approximately 1 .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
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 ฑ5dB. Thus, while the
magnitude of the individual responses at pre-test may or may not be of practical concern in a
particular application, the acceptance criteria is focused on the stability of the response when the
circuit is subsequently subjected to environmental stress.  The post-85/85 and post-TS results
confirm that changes in the responses are all acceptable. However, post-MS shows several
anomalies (seven by count), as shown in Figure 4-16. Five of these seven anomalies were for
immersion silver, while HASL and immersion tin each had one anomaly.

       Figure 4-17 displays the boxplot of the test results for HF TLC 500MHz post-MS. The
HF TLC 500MHz results for the other test times are quite similar to those for HF TLC 50MHz,
and boxplots of these results can also be found in Appendix F. Post-MS results for HF TLC
500MHz had only one slight anomaly compared to seven for HF TLC 50MHz. This anomaly was
only -5.22dB, compared to the lower bound of-5dB, so it is of no concern. Boxplots displays for
HF TLC IGHz are not given to  conserve space. The total variation at pre-test for HF TLC IGHz
was only 2dB, and there was only one slight anomaly of -5dB at post-MS, which is not of
concern.

       Figure 4-18 displays the boxplot of the test results for HF TLC RNR post-MS. None of
the F-statistics were significant for testing equality of means; boxplots of results from the other
three test times can be found in Appendix F. The reader should keep in mind that the decreases in
the HF TLC RNR response in Figure 4-18 are favorable outcomes. The acceptance criterion only
specifies an upper bound of either 5dBb or lOdB for the increase, depending on the magnitude of
                                          4-28

-------
                                                                  4.1 COMPETITIVENESS
 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 KNR 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 riven in Appendix F
 (Tables F-23 to F-26).     .

 10-Mil  Pads

       Tables F-27 and F-28 give the predicted changes from their respective base cases for all
 leakage measurements at pre-test for the GLMs.  Examination of the GLM results for 10-mil pad
 shows evidence of site-to-site variation and some interaction between site and flux that affects
 resistance either positively or negatively by up  to an order of magnitude.  Demonstration sites
 applying the OSP surface finish (Sites 6, 7,  8, and 9), as well as Sites 10 and 11 with immersion
 tin, do not differ from the base case when low residue flux is used.  When sites are dropped from
 the GLM and replaced by surface finishes, the results show slight increases in resistance over the
 base case for OSP, immersion tin, and immersion silver.

      The differences from the base case for both GLMs essentially disappear after exposure to
 the 85/85 test environment. This result is not unusual and may be due to a cleansing effect from
the 85/85 test environment that removes residues resulting from board fabrication, assembly, and
handling. This same phenomenon was observed for the other three leakage circuits.
                                         4-29

-------
4.1 COMPETITIVENESS
       Boxplot Displays of Multiple Comparison Results. As with the other circuits, an
ANOVA was performed to determine if there was a significant difference in the mean leakage
measurements for each of the four leakage circuits. The p-values for the respective f-statistics for
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
0.000
0.000
0.000
0.000
 Post-85/85
0.000
 Post-TS
 Post-MS
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 Arrav-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 COMPETTITVEISESS
       Boxplot Displays of Multiple Comparison Results.  The p-values for the ANOVA
 given above show the only test indicating a significant difference in mean leakage for the PGA-A
 circuit was the pre-test (shown in Figure 4-21). Boxplot displays of the other leakage
 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 tejst
 environment. Boxplot displays of the other leakage measurements for PGA-B are given in
 Appendix F.

       Comparison to Acceptance Criterion. There were no anomalies for PGA-B at pre-test,
 post-85/85, post-TS,  or at post-MS.

 Gull Wing

     .  Examination of the GLM results in Table F-27 for the Gull Wing shows a moderate effect
 due to flux of approximately 0.81 orders of magnitude. There is evidence of modest site-to-site
variation and some interaction between  site and flux.  Eleven of the demonstration sites do  not
differ from the base case when low residue flux is used, and the other two only differ slightly.
Table F-28 shows a flux effect  of approximately 1.09 orders of magnitude as determined in the
GLM analyses by surface finish, indicating there are no meaningful differences due to surface
                                         4-31

-------
4.1 COMPETITIVENESS
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
MS = mechanical shock; PGA = phi grid array; TS = thermal shock.
                                          4-32

-------
                                                                  4.1 COMPETITIVENESS
       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 (0.359,0.370,0.365, 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
 pinholes and foreign material (e.g., solder balls).  The following photographs  show examples of
the more prominent visual defects.
                                         4-33

-------
4.1 COMPETITIVENESS

                                            ^


             Solder ball between pins of U3 ™ Board #013-1
          ;  Discoloration on trace connec ed\to^6 on B~nJ *01ซ     #1Q2.4
                                                      4-34

-------
                                                                   4.1 COMPETITIVENESS
Table 4-21. Identification of Assemblies Selected for Ion Chromatography Analysis
' Froish
Board #
Assembly Process
Site
Untested Board (Control Group)
HASL
HASL
Nickel/Gold
Nickel/Gold
Nickel/Palladium/Gold
OSP
Immersion Silver
Immersion Silver
Immersion Tin
Immersion Tin
077-4
096-2
068-4
017-4
001-4
061-2
085-4
074-3
103-4
034-4
LR
WS
WS
LR
LR
WS
WS
LR
WS
LR
1
2
7
12
15
3
8
9
4
10
Post-85/85 Exposure (Anomaly Group)
HASL
Nickel/Gold
Nickel/Gold
OSP
Immersion Silver
Immersion Silver
Immersion Tin
Immersion Tin
Immersion Tin
Immersion Tin
083-2
013-1
015-4
056-4
082-2
094-4
030-4
032-4
086-2
102-4
WS
LR
LR
LR
LR
WS
WS
LR
WS
WS
LR - low residue flux; WS = water soluble flux.
1
13
14
5 •
11
12
9
8
7
10

Test Method

       The fundamental steps for conducting ion chromatography analysis per IPC-TM-650,
method 2.3.28 are as follows:
1.
3.
4.
5.
6.
The lab technician (LT) placed the test board(s) into clean KAPAK™ (heat-sealable
polyester film) bag(s).
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.
The LT inserted the bag(s) into an 80 ฐC water bath for one hour.
The LT removed the bag(s) from the water bath.
The LT separated the test board(s) from the bags.
The LT placed the test board(s) on a clean holding rack for air drying at room
temperature.
                                          4-35

-------
4.1 COMPETITIVENESS
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 ปData (HASL)
Sample
Description
Assembly -
Process
Site
\
Ion Chromatography Data '
Q- f 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 (y^g/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)
Sample
Description
Assembly
Process
Site
Ion Chromatography Data
a
Br
WOA
Untested Boards (Control Group)
Board #034-4
Board #103-4
LR
WS
10
4
0.87
5.10
5.26
2.98
140.45
3.30 .
Tested Boards (Anomaly Group) .
Board #032-4
Board #030-4
Board #086-2
Board #102-4
LR
WS
WS
WS
8
9
7
10
1.75
1.70
2.99
2.33
4.12
5.68
3.30
3.16
15.78
15.46
9.23
4.63
  Test results reported as micrograms of the residue species per square inch of extracted surface (//g/in2).
Br" = bromide ion; Cl" = chloride ion; LR = low residue flux; WOA = weak organic acids; WS = water soluble flux.
                                            4-36

-------
                                                                            4.1 COMPETITIVENESS
Table 4-24. Ion Chromatography Anion ฐData (Immersion Silver) a
Sample
Description
Assembly
Process
Site
loa Chromatography Data
a
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 Og/in2).
Br" = bromide ion; Cl" = chloride ion; LR = low residue flux; WO A = weak organic acids; WS = water soluble flux.
Table 4-25. Ion Chromatography Anion0 Data (Nickel/Gold)3
Sample
Description '
Assembly
Process "
Site
Ion Chromatography Data
a
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 #01 5-4
LR
LR
13
14
2.44
1.63
3.56
2.80
15.13
. 14.04
Test results reported as micrograms of the residue species per square inch of extracted surface Og/in2).
Br- = bromide ion; Cl - = chloride ion; LR = low residue flux; WOA = weak organic acids; WS = water soluble flux.
Sample
Description
.Assembly
Process
Site
Ion Chromatography Data
CT
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 j 4.28
26.41
Br- - bromide ion; Cl" = chloride ion; LR = low residue flux; WOA = weak organic acids; WS = water soluble flux.
                                               4-37

-------
4.1 COMPETITIVENESS
        Table 4-27. Ion Chromatography Anion <-> Data (Nickel/PaUadium/Gold)
Sample
Description
Assembly *
Process*
Site
Ion Chromatography Data ,
- CT
Br~
WOA
Untested Boards (Control Group)
Board #001-4
LR
15
0.84
5.15
,151.18
* 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.
       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 ,ug/in2 for finished
assemblies processed with water-soluble fluxes, and no more than 2.5 //g/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.

       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.
                                           4-38

-------
                                                                    4.1 COMPETITIVENESS
        Based on the fact the tested boards with known anomalies exhibit levels near or below
 CSL's recommended guidelines, there is reasonable confidence that the anomalies identified in the
 performance testing are not the result of chloride residues.  The majority of the anomalies are
 either mechanical in nature (e.g., poor solder joint integrity) or component non-conformities (e.g.,
 wrong value and device failures).

        Bromide. Bromide ion (Br') is generally attributable to the bromide fire retardant added
 to epoxy-glass laminates to give fire resistance, and which is subsequently extracted in the ion
 chromatography analytical procedure. Bromide can also sometimes come from solder masks,
 marking inks, or fluxes that have a bromide activator material.  Bromide, when from the fire
 retardant, is not a material that typically degrades the long-term reliability of electronic
 assemblies. If bromide comes from.a flux residue, it can be corrosive, as other halides can be.
 The level of bromide varies depending on the porosity of the laminate and/or mask, the degree of
 over/under cure of the laminate or mask, or the number of exposures  to reflow temperatures.

       For epoxy-glass laminate, bromide  levels typically fall within the range of 0 to  7 /^g/in2, -
 depending upon the amount of fire retardant the laminate manufacturer has added. Exposure to
 reflow conditions tends to increase the porosity of the laminate and mask. With several exposures
 to reflow conditions, bromide can reach levels as high as 10 to  12 Mg/in2. The testing laboratory,
 CSL, does not presently consider bromide  levels under 12 //g/in2 to be detrimental on organic
 PWBs. However, CSL considers levels between 12 jug/in2 to 20 //g/in2to be a borderline risk for
 failures if attributable to corrosive flux residues.  Furthermore,  levels above 20 Mg/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 feilures.  CSL attributes these bromide levels to the
 fire retardant material in the FR-4 laminate.

       Weak Organic Acids. Weak organic acids (WOAs), such as  adipic or succinic acid,
 serve as activator compounds in many fluxes, especially no-clean fluxes. WOAs are typically
 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
 20 - 120
250 - 400 us/in2
                                          4-39

-------
4.1 COMPETITIVENESS
       When WOA levels are under 400 //g/in2, the residues are generally not detrimental.
Excessive WOA amounts (appreciably greater than 400 //g/in2) present a significant reliability
threat for finished assemblies. Low levels of WOA can also create electrical performance
problems in certain applications.

•      An excessive amount of flux can produce the situation in which the thermal energy of
       preheat is spent driving off the solvent, therefore not allowing the flux to reach its full
       activation temperature. Unreacted flux residues readily absorb moisture that promotes the
       formation of corrosion and the potential for current leakage failures.
•      Fully reacted and therefore benign WO As act as insulators that, even at levels as low as
      • 10 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.

4.1.15 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.
                                          4-40

-------
                         4.1 COMPETITIVENESS
Table 4-29. Frequency of Anomalies by Individual Circuit Over Test Times
Circuitry"
\ >> *,"- ^
" * *
Post-
85/85
Post-
Thermal
shock
Post-
Mechanical
Shock'
- _ Comments k t~
HCLV
1
2
HCLVPTH
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
HVLCPTH
HVLC SMT
0
1
0
1
0
164
Excellent performance throughout.
SMT components came off board during
mechanical shock.
HSD
5
6
HSDPTH
HSD SMT
0
1
2
2
2
1
Component problem.
Component problem.
HFLPF
7
8
9
10
11
12
HFPTH 50MHz
HF PTH f(-3dB)
HFPTH f(-40dB)
HF SMT 50MHz
HF SMT f(-3dB)
HF SMT 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
16
16
17
HF TLC 50MHz
HF TLC 500MHz
HF TLC IGHz
HFTLCKNF
HFTLCRNR
0
0
0
0!iฃ^Jtlpi?
1
0
0
1

2
7
1
1

5
Minor anomalies.
Minor anomalies.
Minor anomalies.

Minor anomalies.
Leakage •
18
19
20
21
10-milPads
PGA-A
PGA-B
Gull Wing
0
0
0
1
0
0
0
0
0
0
0
0
Excellent performance throughout.
Excellent performance throughout.
Excellent performance throughout.
Excellent performance throughout.
Stranded Wire
22
SW1
0
0
1 (Excellent performance throughout.
4-41

-------
4.1 COMPETITIVENESS
Grciritry
f v
-
23 |SW2
Post-
85/85

0
Post-
Thermal
shock
1
Post-
Mechanical
Shock
4
Comments ' ,
^ , j r, , (
, " f „". ',. -
' * •ป
Minor anomalies.
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; S W = stranded wire;
TLC = transmission line coupler.
       With, the exception of the HCLV SMT and HVLC SMT circuits in the mechanical shock
test, the surface finishes under study were very robust to the environmental exposures.  When
assessing the HCLV SMT and HVLC SMT results, product and process designers should
consider the severity of the mechanical shock test (25 drops, five times on each edge excluding
the connector edge and five times on each face, to a concrete surface from a height of one meter).
Also, HCLV SMT and HVLC SMT anomalies due to SMT components coming off the board
during the execution of the mechanical shock test were equally distributed across all surface
finishes including the HASL baseline.

       Based on the results of the Failure Analysis:

•      Observed levels of bromide and WOA on all 20 assemblies are typical and therefore not
       detrimental from an electrochemical standpoint.
•      Based on the feet 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.
                                           4-42

-------
                                                                 4.1 COMPETITIVENESS
 Pre-Test
 HCLV PTH
  Q_

  O
               HASL
 Boxplots of HCLV PTH by SiteFlux
    (means are indicated bysolid circles)
  OSP         Imm Sn      Imm Ag
                                                                Ni/Au    Ni/Au/Pd
   SiteFlux

7.5 —

7.4-
7.3 —
7^2-
7.1 -
7.0-
6.9 —
6.8 —

JX



1 1
T i



"~ ^



i
j
j



c*

i
i
I .:
i i
i 'i
* i

!
I
. 1
I
T. ""



L
i
:'
: r
'•i



cc



X
i


I
r- ce

X

1 1


. 1
1
en a



[
:
:



sr



1
I
j



Cv



j
|


I
C'
	 I—
|
i
i
!
1
!
i!
ง!
!
i
l *
i
*i
i
r~H
2 ฃ



I
'



C



4
1



0








r~
— i —
i
f
f
|!
f i
:J i
i •
P
i
}
~^
cc



\ I
!
;
•



) a



1
i
I
• :
1

. 91

i g



i
1 i
i i
3J



T-
fM
_j 	
I
i
i
i
i
i
i
i
1
r
1
!


o
n




i i
i
•
f :•

i
J CO
J r\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          ImmSn      ImmAg
                                                                Ni/Au   Ni/Au/Pd
0.5 —

0.4 —
0.3-
0.2-
L 0.1 -
>
J 0.0 -
5 -0.1 -
-0.2-
-0.3 —
-0.4 —
SiteFlux







T~



J



Cv







V

1
X. i
i
1
1
if
1
1
r
!
1 T
•^ in cc



1
j


IV



1
*
T

r
CO







cr

I
I
I
i
1 '
I '
i 1 *
I
rซ
11
i
c




i I
M

•r-



I
i r
i
T

r
CM



J



(T



(
:.
i
i
' i
; :
: :
1

" i
-3
— r 	
$
i
I
i i
1
5
i
1
i
i

10 c



1
j


r
D







r^-
	 ; —
i
, i
i
i i
i r
i i
i
i
J
i
i
i

K



'
•
•

) O







) C
	 T —
1

I
i
i
. , i
i
is
l!
T! [
i
i
i
i

' T- cs
1 CM tN



1,
1


1
i n
l r\i
               WS
                    ws
                          ws
     ws
                                        ws
                      ws   wsws   ws    ws   ws-
 Figure 4-2. Boxplot Displays for HCLV PTH Post 85/85 - Pre-test Measurements (volts) by Surface
                                         Finish
               '               (Acceptance Criterion = A<0.5V)
                                        4-43

-------
4.1 COMPETITIVENESS
Post Thermal Shock
HCLV PTH
               HASL
Boxplots of DTHCLV P by SiteFlux
    (means are indicated bysolid circles)
  OSP         ImmSn      ImmAg
                                                                 Ni/Au    Ni/Au/Pd
0.5 —




3 o.o-
\
J
3




-0.5-



J












|



1
jj
T
1





|
SiteFlux ••- CM




11





i


-

T

i
co •q
T
i
I
j
I
I
i
i
i
\
i
i
i
i
!








J
l
|
•:•.
T




|
r in
WS WS





L

:




1





i i



1
|
1
1
'




1
CD t^ CO
- B~~



•
I
(



*

|




i
1
f





1

















ฃ
1

1




1
O) O T-
WS WS
{ J 'J
! I I
* JL* 1




1 J
1
j
'
J


;
i
,
; i
: 1
.
i
{

1
1

1
I
: { tJ
1!"
1
i
I
I
I
!
!
!
r
$
i
J
i i
[I :
f T i
i
i
i
.11 L
] il 1
if
.1 i
I1 I
?!
' 1
•:•
i ' ฃ;
i
Mr


i
i
i i
i i
11 f


i i i i i i
III*
CM co "^a1 in co h- COOTOT-CMCO
ws 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)
        2 —
    cu

    o

    Q
        0-
                           Boxplots of DMHCLVP by SiteFlux
                               (means are indicated bysolid circles)
   SitoFIux
HASL OSP immSn ImmAg Ni/Au Ni/Au/Pd
t fl t
l l
! i
i !
I * • 1 .
i i
{ i
i
i Jr.
I 1
I 1 ' _,_
II .|, [I
1 i 1 j |:| JL S 1^ ra
H R i. M i ^ B ^ m ^ H n ^
flifjllTlil fj
T @ 1 1 T T i
I
i

.
i
i I
I ra
^ I !i ^
1 iff
*
! !
! {
i i

! !
^ 1 j
i ! i
: i !
si i
; i . [
l !
* 1 rti 1 j 1
' S El (a il j &
i 1 1 i In f ^
i ||| T ' il i
iT III i . 1 ' i
; i : ! i
i i i i i i i i i . i i i iii i i i i i i
ws ws ws ws ws ws wsws ws ws ws
  Figure 4-4. Boxplot Displays for HCLV PTH Post MS - Pre-Test Measurements (volts) by Surface
                                          Finish
                               (Acceptance Criterion = A<0.5V)
                                          4-44

-------
                                                                 4.1 COMPETITIVENESS
 Pre-Test
 HCLV SMT
 CO
 >
Boxplots of HCLV SMT by SiteFlux
    (means are indicated by so lid circles)
r-
7.45 —
7.40 —

7.35 —
7.30 -
725-
7.20-

7.15 -

7.10 —
7.05-
7.00-




i
1





1ASL OSP . ImmSn



1
I





X


Flux ••- CM
i
i
i


!
1





j
i
. i
<
i
i
L



11
r
i
c
i
1





1






*
CO ^J" , IO tD Is-- t



1
T






A
} 1
\ \



it










i
i
i

i
-




*.

i~l 1 1
0 O) O. T- CM C
WS WS WS WS WS
i
L$
: i
c
{
f
f
i
i
JI
I
i
fj -^r u
ws
Imm Ag Ni/Au Ni/Au/Pd
i
i
i
i
T



"

j

1
j

L!
} I
i I
• t
t
Ji
i
|
J

1
i
: ซ
! !
I
•




: I
T I
%
0 

T- C\

L 1 i
13
!••*




E

CO „
T 	 r
1S ft !
!!• 1 !
i l 1 1 i
! I a f JL s
'" 1'
l งi
i i
l f
i ;
f i
i !
i i
i i
! i
1O CD h-
	 3 —
1 [
JL R * F3 !
i * S s
IT* !
i !
r |

i
i
"r
i

n — i — f — r1— i
CO O) • O -i— O


'• ง
: S
1






J CO
J CM
              WS    WS   WS   WS
                                        WS
                                                ws    ws ws   ws
                                                                    ws   ws
Figure 4-6. Boxplot Displays for HCLV PTH Post 85/85 - Pre-Test Measurements (volts) by Surface
                                        Finish
                              (Acceptance Criterion = A<0.5V)
                                        4-45

-------
4.1 COMPETITIVENESS
 Post Thermal Shock
 HCLVSMT
               HASL
   Boxplots of DTHCLVS by SiteFlux

      (means are indicated bysolid circles)

    OSP      '   Imm Sn      Imm Ag
                                          Ni/Au   Ni/Au/Pd
       0.5 —
       0.4 —
       0.3 —
       0.2 —
       0.1 -
       0.0 —
  a    -0.1 -

       -0.2 —


       -0.3 —

       -0.4-

   SiteFIux
1!
                                                   I
   I   I   I
   in   
      Boxpiots of DMHCLVS by SiteFlux

         (means are indicated bysolid circles)

        OSP         I mm Sn      I mm Ag
                                        Ni/Au   Ni/Au/Pd
        3 —
        2-
    Q   1 -
        0 —
   S'rteFlux
                             Wl
                              I
                              I
                              I
                              i
                             * i
                              i
                              r
                              i

                              i
                            	L
                                    i
                                    i
                                    1 T
                                    i

                                    I

                                    i
                                    i
                                    l
                                   J_
 i
^j-
 i  l
in  CD
                              I
                              f-
               ws   ws    ws   ws
                   ws
                                                 WS    WSWS   WS   WS
                                                     WS.
  Figure 4-8. Boxplot Displays for HCLV PTH Post MS - Pre-Test Measurements (volts) by Surface
                                         Finish
                               (Acceptance Criterion = A<0.5V)
                                         4-46

-------
                                                                4.1 COMPETITIVENESS
 Pre-Test
 HF PTH 50MHz
               HASL
 o
 in
 Boxplots of HF PTH50 by SiteFlux
    (means are indicated bysolid circles)
  OSP         ImmSn     ImmAg
                                                               Ni/Au   Ni/Au/Pd
-0.3 —
-0.4 —

-0.5-

-0.6 —
-0.7-

-0.8 —

-0.9 —

-1.0 —

-1.1 —
-1.2-

i i\f
ux

J
i
I
I
I

i i S i 'I i & A f*l
H fi * H H is T ง P|
T •"P T {lip I] l|l
5
i
f
i
!
i
I
i i i i i r r^
^~ CM CO . ^J" IO (O t*ซ- OO O3
WS WS WS WS
J-
I
i
I
j
I
S & ! ^
@ @ $ @ M
T Tj t]
* J
!
f
i
i
i
1
T 1 1 I i~l
o T- CM n -q- ir
WS WS
	 r~
!
1
i
j

L L i
1 KJ J
,j r}
I
i
I
i
!
1
1
i — rH
to N. cc
wsws
. 	 ; — j —
* *
j
1
I
1
I
1
'i li fl p^i i @
T P i T
f
i

1
j
1
I
— 1 I 1 s 1
) O> 0 T- CM
WS WS.







ง








1
CO
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 bysolid circles)
  OSP         ImmSn     ImmAg
                                                              Ni/Au    Ni/Au/Pd
o 	

-10 —
-20 —
C3
10 -30 —
0- • -40-
u.
X
1 -50 ~

-60 —

-70 —
-80 —
SiteFlux
~ Ix *
By •*• 9t *& J flSJ 1Sr e
B }
! i
i i
S i
! !
! ' i
i
* 5

1

I

1
1 '. 1 1 1 \ 1 1
•<- CM to T ir> (o r-

•Sfr -8& ^
'










1 1 1
	 : 	 , 	
— -a ฎ- j
JC '
88 i
i
I
5
1
i
i
1
* 1
5
I
i
1
i
i
i i r^n
CM CO -^ If]

1 *
i
;
:









r — r
(o r--
— I— 	 5 	 : 	
ijj@ g •*• ป j * ซ•
: ! 88 ป !
J 1 I
: 8 ! -
J i
H 1
J J
i :
i „ j
3t
1 i

1 !
3 ! i
* . <
! 1
r . 1 •
. i i I ' I i;— '
co o> o T— CM ro
              ws   ws    ws   ws
             ws
                                                ws   ws ws
                                                             ws
                                         ws   ws
Figure 4-10. Boxplot Displays for HF PTH 50MHz Post MS - Pre-Test Measurements (dB) by Surface
                                        Finish
                          (Acceptance Criterion = ฑ5dB of Pre-test)
                                       4-47

-------
4.1 COMPETITIVENESS
Post Mechanical Shock
HFPTHf(-3dB)
              HASL
Boxplots of DMHF PTH by SiteFlux
    (means are indicated bysolid circles)
  OSP         I mm Sn     Imm Ag
                                                               Ni/Au   Ni/Au/Pd
200 —

150 —

100 —

50 —
5:
0-
5 -50-
O -100 —
-150-

-200-

-250 —

SiteFlux
i i
I *
* i
1
I
I
I
s \ a
8! j
j , . i
I
I
I
I
1 i

1 1 1 1, 1 1 1 1 1 1 .
T-CMCOM-mCOI^-COCDO-r
i
* I
I
I
J
I
I
a i
\ I™ i
; i
\ -i \
\ • i i
i i i
I j. t

i i i i i
- CM m •? in co r-
1
i
i
i
i
i
I
I
1 I
I i
i f
i ! i
* 1
1 I
:l I
Jl 1
*
1
- CO 0
(
I
I
• 1
1
1
i
1
* !
-i
i
"
1
I
\
1 I II

               ws   -WS   ws   ws
                                        ws
                      ws    wsws
WS    WS •  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
 HF PTH f(-40dB)
               HASL
 o
 I
       200 —
       100 —
        0 —
      -100 —
      -200-
      -300 —
      -400 —
                          Boxplots of DMHFPTH- by SiteFlux
                              (means are indicated bysolid circles)
                            OSP         ImmSn     immAg
                                    .Ni/Au   Ni/Au/Pd
   SiteFlux
	 ' 	 5 	 1 	 f—
I i • !
{ s !
i 1 . • i
i 1 a J


ซ * K | * i • * j i
I i f
I i i
i i i
i n
r K * !
* s s
i !
5 5
i 9
j j
1 — I • I I 1 1 i i i I i i i i i





&




i





1 1 1
_j 	 j 	
i I
i !
i !
$ i
$ i
-.{$$ JL {— <#•{•&• &•

i a * i
i i
. I I
' S i
!l * i
M i
ii * i
!i - i

r^ i i ii i
             T-  CM CO
               WS   WS   WS   WS
                                        WS
                                                WS
                                                      WSWS
                                    ws
                                                                   ws
           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)
                                        4-48

-------
                                                                 4.1 COMPETITIVEIYESS
Post Mechanical Shock
HF SMT 50MHz
               HASL
  O
  v>
  u.
        0 —
       -10 —
       -20 —
       -30-
       -40 —
       -50 —
  S
  Q    -60 —

       -70-

       -80 -

       -90 —

  SiteFlux
Boxplots of DMHF SMT by SiteFlux
     (means are indicated bysolid circles)
  OSP          ImmSn     InimAg
                                                                Ni/Au   Ni/Au/Pd

~ * R F
IS g
T



•
j
X
X





i 1 1
•<- CM CO -51
	 J 	
^1 '
i i
: ;
: i
: ,

:
i i
i
!
S
i
!
!
1 I i
10 co r-

'•
I \
I

\
:
|

f






CO
1

i ป
i
:
i
:

' 1
* i
1
I



i i
O) O















T-

3 **












1 i
CM r
—. j
1 ~ i *
i s ;
! !
: '
1 I
>
i i
J 5
f.i
i
i
i
S
i
S

•f IO CC

T
:
i
|

:
|
:
'
f






r^
— i 	 i 	 : 	 1
L|ป -^ ซป>* 88-
h i
: i |

r i
i i
s - s
! I
I I
Jj 1
I I
1 I
i I
f i
I i
; t
1 I.I 1 '-[-—I — '
CO O> O - . v- CM CO
               ws    ws
ws
ws
                                         ws
                       ws   wsws
Figure 4-13.  Boxplot Displays for HF SMT 50MHz Post MS - Pre-Test Measurements (dB) by Surface
                                         Finish
                           (Acceptance Criterion = ฑ5dB of Pre-test)
Boxplots of DMHF SMT by SiteFlux
    (means are indicated bysolid circles)
  OSP         Imm Sn      Imm Ag
                                                                Ni/Au    Ni/Au/Pd





I
1
5
S

M C2] til


X *

1 1
T- 0) CO
ws. \
f
i
1
f
i
i
I
I
I
i
i
i
m i* *
i !
1 i
f !
1
~l 1 1
Tf 1C CO t
NS WS

*







s
X
1
I

T
^











ป
T



i r^
CO O) C
ws
	 : 	 1 	 ! 	 ! 	 : 	 1
! ! !
1 I \
I i i
, i 5 f-
f 1 f
I I I
11 I
: : i
ii i
i i i
. „ .J. f j. ...[..
1 1 i II i :
? 1 | 1 l I
:•:•:•:.* g * I
1 1 ! M i
i T ** |l !
i i i
i i i i i i i i 1 1 1 1 "1 —
T-cMco^rmcDh-ooOTOT-cM co
ws ws wsws ws ws 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)
                                        4-49

-------
4.1 COMPETITIVENESS
 Post Mechanical Shock
 HFSMTf(-40dB)
              HASL
 o
 "T
 co
 Q
Boxplots of DMHFSMT- by SiteFlux
    (means are indicated bysolid circles)
  OSP         ImmSn     ImmAg
              Ni/Au   Ni/Au/Pd
0 —

-200-


-400-



-600-

-800 —


SiteFlux
(ฑ1

8







X


1
T- r














•l e














5 T
I
1
i
i
1
f
I
{
I

i
I
I
f!
|

f


1










i
10
1 1

1





X




1
ID t>
.













- a
, i

I *




i



;

1 1
3 O> "C














5 T
9 P?
X lฃ
f-:;:



:

! *




i
CM CO
i

jl
i
i
;
1
* .
*
I ^
i
I
\
1 1
T m S x
* } * 5 *
5 i
lllliittiiiiii.il
r-cMco-q-incDr~cocno-r-CMcoM;incD
kg Ni/Au Ni/Au/Pd
"I • i
I !
] i
i !
$ i
i :
I i
: i
! !
! !
1 i '1
| ! i
1 l l

•:• K!
^*ฎ ^? ija ซ&*l-ฎ si
! * j

J J j [ p* p^ J
h- CO O) O T- CM CO
               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)
                                        4-50

-------
                                                               4.1 COMPETITIVENESS
 Post Mechanical Shock

 HF TLC 500MHz

               HASL
   o
   o
   LO
   U.
 Boxplots of DMHF TL5 by SiteFlux

     (means are indicated bysolid circles)

   OSP         Imm Sn      Imm Ag
                                                               Ni/Au   Ni/Au/Pd
  SiteFlux
4 —
3 —
2 —

1 -
0 —

-1 -
-2-
-3-
-4-

-5 —

-6 —

JX

I
i
I
I ' '
I J, <
m -L & J ;

i
1
i
* !
* i
i
I
i
!

1 1 I
•^- CM co ^r in



* •' *

lS(4ld]*Srli^

1 ' * 1
T
'



I
l
1 1 i I~l i i 	 1
•CD I^CO CT> O. T- CM f
	 ; 	 : 	
i
i
i
* f
1 * 1
I ca H
* . GS * Ijf? Ep
? ! i
i i
* i * ;
I
i
I
i
i
i
r^i — i — i
•^ u> CD r>
— ! 	 T 	
I I
i ^ j

i * ! *
(1 1
; , ^ _ is

I1 J
i! * . *!
]i. * i'
i !
i :
i i
! i
_j_ 	 |
i l 1 i i 	 T 	 '
CO O> O •*- f\l CO
               ws   ws
ws
                               ws
                                       ws
                      ws
                                                     wsws    ws   ws   ws
Figure 4-17. Boxplot Displays for HF TLC 500MHz Post MS - Pre-Test Measurements (dB) by Surface

                                        Finish

                          (Acceptance Criterion = ฑ5dB of Pre-test)
Post Mechanical Shock

HF TLC RNR  '

              HASL
Boxplots of DMHFTLRN by SiteFlux

    (means are indicated bysolid circles)

  OSP         ImmSn     ImmAg
                                                              Ni/Au   Ni/Au/Pd
10 —



0 —
~3
Z
t -10 —
X
^^
Q

-20-



-30 —
SiteFlux
* i I
I .* *l
i x I
* * 1 ^ f
— >ฎv,&^fyi}.3$ig,v, fs — S.&, ss!
ss X ' R * ' * i
* ' ' Ip ป i
ป 1
* -V 1
i J
I I
I I
\ I
1 f
i !
* i r


	 j- 	 5 — . 	
f i
J 5
1 i 1
I ! * ^ {
H ffl i a &. 5 il
1 P i ** x * f ^
i T i i
1 ' 1 . ^
:i i !
3 I i
I f
i |
i 1
1 1 .
i j
i I

1 1 1 1 1 1 ' 1 'T~'
              WS    WS   WS   'WS
                                      • ws
                                               ws .  wsws
                                    ws
                                                                   ws   ws
 Figure 4-18. Boxplot Displays for HF TLC RNR Post MS - Pre-Test Measurements (dB) by Surface

                                       Finish

                     (Acceptance Criterion = <10dB increase over Pre-test)
                                      4-51

-------
4.1 COMPETITIVENESS
^Mrf P rt Boxplots of Pads by SiteFlux
10-MllradS (means are indicated bysolid circles)
HASL OSP Imm Sn Imm Ag
15 —


14-


13 —
in
"8
Q.
.12-


11 -


10-
	 1 I >
.1 *l
I I
\ I
?.

fi i ft i
1 jlj g, 1 * I '
ฃ ! P PI i i ch
Bl i B A A rh I B
pi , 11 lj n R i' @
i i T @ f I i T
? 1 1
i !
1 i
i J
I j
*- f I
1 I I I I I I I I 1 I I I I I I I
SiteFlUX •e-encov'inaJi-.comO'.-oiCT'jriouJ^;
WS WS WS WS WS WS WSWS

Ni/Au
f
i
f
; •ป
i
I
I
1

;
|
i m ฎ
i B *
i
f
i
!
1 1 T
CO O) O
WS

Ni/Au/Pd
I
j
j
Eg j CB
ฅ j v
i
J
i

;
!
! fa
j '
i
i
i
. 1 . .

















1 i 1
T- CM CO
WS WS ..
Figure 4-19. Boxplot Displays for 10-MiI Pad Measurements (Iog10 ohms) at Pre-Test by Surface Finish
(Acceptance Criterion = Resistance > 7.7 Iog10
Post 85/85 Boxplots of DPPads by SiteFlux
"ID-Mil Paris
1 WIVIII 1 C3UO /^,^ซซ_. „... ;ซrt;,~*+Arl kworOM r*irfltm\
ohms)



               HASL
OSP         ImmSn      ImmAg
                                                                  Ni/Au    Ni/Au/Pd
        14—1
        13-
   tn
   •a

   S.
   Q.
   Q
        11 -
        10-


   SiteFlux

1
yi i
1 i
0 H
T T










i i
r- CM C1
WS
1

^ PI i PI rl rll
i ! j T 1
' 1 T
1 J
i
i
i
i
1
1
i
1
j
f
II 1 1 1
•> TT in  7.7 Iog10 ohms)
                                          4-52

-------
                                                                  4.1 COMPETITIVENESS
 P re-Test
 PGA-A
               HASL
       14 —
       13-
       12 —
       11 —
       10 —
  SiteFlux
 Boxplots of PGA A by SiteFlux
   (means are indicated bysolid circles)
 OSP      .    ImmSn      ImmAg
                                                                 Ni/Au    Ni/Au/Pd
- I 1 : ; i 	 1 	 , 	 .
• 1 • M 1 ' A i 1

















& ' it] Ii ' fa ml
4, * ' ' S 8 I
r"'l ! Iffi f^j a
Bซ f " , ' Eฃ] |
* ' , • i 1 1 i
H * |
; • *lf B] ' !
i T 1 i
f n * iS
iff
, 8 '
T EI S m ml K)
i 11 { T i
i ^T
' ซ 1
II i
.1 I
i 1 I i
i i "i •'ป f
I 1 . 1 I
1 1 JL 1 , 1
I i !i Si-'
• i ' T ! i ' * i ST
T i 1 i i i?
1 i i i
* i i ;
I 1 1 1 I 1 1 Fl I 1
> i i
i i — i — i — r~i — i — •
- cMCOTrmcDr--cocno-^ev 7.7 Iog10 ohms)
Pre-Test
PGA-B
               HASL
Boxplots of PGA B by SiteFlux
  (means are indicated bysolid circles)
OSP         ImmSn      ImmAg
                                                                Ni/Au   Ni/Au/Pd
14 —


13 —



DQ
CL
11 —


10 —
SiteFlux

|
*' ^ ! A * '
5 i p
5
i
i
! *
I
I
i
Ijl
1 E*| 5 1
111 1
Y i i Ip
i i | | ซ-] -| [ |
ws ws ws ws

i
T



•

u
:
i
:
\
: ^
r
1 T 1
WS
	 j 	 j 	 , 	
Ii i is i i! f
! S I Ii 1
i I 1 i
i i * 1 ;
! | - | i
* f t •
I I i "
L * 1 i ,
i i i 1
\ i i i
j ! j, 1

n ' ' H I S i
PI ' Y t ^ i di
^ I .1 T $ Ip
i i ii i — r 8 i ii — H— i — 7
ws wsws ws ws w.

':

i
•











Figure 4-22.  Boxplot Displays for PGA-B Measurements (Iog10 ohms) at Pre-Test by Surface Finish
                      (Acceptance Criterion = Resistance > 7.7 Iog10 ohms)
                                        4-53

-------
4.1 COMPETITIVENESS
prTw-St Boxplots of GullWing by SiteFlux
V3UII Wing (means are indicated bysolid circles)

HASL OSP ImmSn ImmAg Ni/Au
14 —


13-


12 —
D)
3 11-

10-


Q _
1 B > *
I i n !
1 JL i ! H ^ !
i! i. 1 ง '! i! i
1 l^l T J, I * ' 1 . @
Ig i pi III A fsl la Fl ^ ^ fl 1
•;' ^ j_ ' 1 H H lj:| ^ ^| i i \ ra
il $ 1 ii*i| * r i
II 1 * 5 \ • i I i
Tli 1 \ฎ \
1^1 i 1
5 11
I * i I
f i 1
I i K i

— I — | 	 | I I I I I I i i i i i i i i i





j.
1
1
f





T

Ni/Au/Pd
*
i
i
J
- 1
~
•


fS
! , 1
i JL y
J 1
l|
il
j
i
L
i
i
s















1 1
SiteFlUX '-cMCOTincoi-.coroo^cvjcoj.ncDh-cocDO^CNcg
WS    WS    WS  -WS
                                         WS
                                                  WS
                                        WSWS
WS
WS    WS
 Figure 4-23. Boxplot Displays for the Gull Wing Measurements (Iog10 ohms) at Pre-Test by Surface
                                          Finish
                       (Acceptance Criterion = Resistance > 7.7 log™ ohms)
                                         4-54

-------
                                                                       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
;>^Spn^^
HASL
Nickel/Gold
Nickel/Palladium/Gold
OSP
Immersion Silver
Immersion Tin
/ ^Qn-CjDttVejfbBiBeSS1-*
•
•
•
•

•
?:ซ^imybnraci><',
•


•
•
•
       Costs were analyzed using a cost model developed by the University of Tennessee
Department of Industrial Engineering.  The model employs generic process steps and functional
groups (see Section 2.1, Chemistry and Process Description of Surface Finishing Technologies) to
form a typical bath sequence (see Section 3.1, Source Release Assessment) for each process
alternative. To develop comparative costs on a $/surface square foot (ssf) basis, the cost model
was formulated to calculate the cost of performing the surface finishing function on a job
consisting of 260,000 ssf (value corresponds to the average annual throughput for facilities using
HASL in the PWB Workplace Practices Questionnaire database).

      Processes were also modeled at a throughput of 60,000 ssf, a number which corresponds
to the average annual throughput for facilities using a non-HASL alternative. This additional
modeling run was performed to examine the effects, if any, that operating throughput will have on
the normalized cost for each process. Although the calculations presented in this section are
based on the higher production operating conditions, similar calculations were performed using
lower production level data and the results of the two runs are compared at the end of the cost
analysis.
                                          4-55

-------
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
Actraty-Based Cost
   Components
                                          Cost
                                        Analysis
                      Figure 4-24.  Hybrid Cost Analysis Framework
                                           4-56

-------
                                                                      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 TJnquantifiable Costs

 Cost Categories '

       Table 4-31 summarizes the cost components considered in this analysis, gives a brief
 description of each cost component and key assumptions, and lists the primary sources of data for
 determining the costs. Section 4.2.5 gives a more detailed accounting of the cost components,
 including sample cost calculations for each component.  In addition to traditional costs, such as
 capital, production, and maintenance costs, the cost formulation identifies and captures some
 environmental costs associated with the technologies. In this regard, both simulation and ABC
 assist in analyzing the impact of the surface finishing technologies on the environment.
 Specifically, the amounts of energy and water consumed, as well as the amount of wastewater
 generated, are determined for each surface finishing process.

 Unquantifiable Cost Categories

       The goal of this cost analysis was to perform a comparative cost analysis on the surface
 finishing alternatives in the evaluation. Although every effort was made to characterize each cost
component listed in Table 4-31, data and/or process limitations prevented the quantification of
every component.  A qualitative discussion of each of these costs is presented below.
                                          4-57

-------
4.2 COST ANALYSIS
       d
       o

  113
                                    III
                                             irs
                                                   1
                                                   •8
                                                   S
                                                         &s
 s

5
-uป
 WJ

s
 


                                     8*

                                     •S
                                                                 PH
     -
             1
I
3


I
     o o
                            .^
                            •3 8
                                       -J2
                                       CO

                                       O





                                       S)
                                   4-58

-------
                                                                            4.2 COSTAIVAJLFS1S

ber
tion
latio
rials from a bill of activity (BO
nts ired from mulati
Cost of transporting
um of bath pla
er of bath cl
ts
Cost to c
qu
C
re
r
n
Ass alytical wo
activity from BOA;
Workplace Ptice
es
fr
la
ber of
ual n
ices
one activity from BOA;
from IPC Workplace Pr
UR8
or o
aced
f
c
naire a
Labor cos
filters
uest
N
si
w
ed; fi disposal
cantly among the
data on the type and size of
reliably estimate these
disposal cost includ
gen
ze (weight
ste filters
ce
         6

1
1
1
    1    il*
    O    g Oj  +j
           o 1
           to  "-1
•   -3 S -. s
8 .3 ง -- ง
   ฃฃ
   S 83
     .S
op
vi
Assu
proc
rized
req
Labor and material
clean up a chemical
up
replac
eq
p
Labor
after b

                                 <8
                                 ซ
                                 I  -
                                   !
Labo
of ch
                                                 <ง
                                                  u
                                                 f
e or d
                                                                                         ฃ•?
ts for repa
Labor
                   o
                   ft
                   CO
                   •3
                   s
                   i
                   I
                   s
                   I
                   T3
                   C/1
                   O
                   I
P
defe
                                                g
                                                o
                                                •1-
o
o

VO
oce
                                                                                                    p
                                                                                                    a
                                                                                                    3
                                                                                                   -t
O
a
   o
   XI
   C3
                   O

                   I
                n
                        O
                              CO
                                    rt
                                    & ,

 CS
 I
a
 
-------
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
                                          4-60

-------
                                                                       4.2 COST ANALYSIS
 wastewater stream on the treatment of the entire stream (e.g., a treatment chemical used 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 in Section 6.2, Recycle, Recovery, and Control Technologies Assessment.

       Other Solid Waste Disposal Costs. Two other types of solid wastes were identified
 among the technologies that could have  significantly different waste  disposal costs:  filter disposal
 cost and defective boards disposal costs. Table 4-32 presents the number of filters that would be
 replaced in each process during a job of 260,000 ssf. This is based on data from the PWB
 Workplace Practices Questionnaire and a utilization ratio (UR) calculated for each process from
 simulation results (Simulation results are discussed further in Section 4.2.3). The UR is the
 percentage of time during the year required for the process to manufacture the required
 throughput.  While these results illustrate that the number of waste filters generated-by the
 processes differ significantly,  no information is available on the characteristics of the filters used
 by the processes. For example, the volume or mass of the filters and waste classification of the
 filters (hazardous or non-hazardous) would significantly affect the unit cost for disposal.
 Therefore, filter disposal costs were not  estimated.

       The number of defective boards produced by a process has significance not only from the
 standpoint of quality costs, but also from the standpoint of waste disposal costs.  Clearly, a higher
defect rate leads to higher scrap and, therefore, waste of resources. However, the Performance
Demonstration showed that each of the technologies can perform as  well as the HASL process if
operated according to specifications.  Thus, for the purposes of this analysis, no differences  would
be expected in the defect rate or associated costs of the technologies.
                                          4-61

-------
4.2 COST ANALYSIS
         Table 4-32.  Number of Filter Replacements by Surface Finishing Process
Surface Finishing Process
HASL, Non-conveyorized
HASL, Conveyorized
Nickel/Gold, Non-conveyorized
Nickel/Palladium/Gold, Non-conveyorized
OSP, Non-conveyorized
OSP, Conveyorized
Immersion Silver, Conveyorized
Immersion Tinx Non-conveyorized
Immersion Tin, Conveyorized
Filter Replacements
per Year3
T
r-
354
354
119
162
150
150
19.5
150
150
Filter Replacements
Required to Produce
260^000 ssffr
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
facilities, if available, would reflect the individual operating practices of each facility (e.g., bath
maintenance frequencies, rise water flow rates, PWB feed rates) preventing a valid comparison of
any process costs. Appendix G presents a graphic representation of the simulation models
developed for each of the surface finishing technologies.

       Simulation modeling provides a number of benefits to the cost analysis, including the
following:

•      Simulation modeling replicates a production run on the computer screen, allowing the
       analyst to observe a process when the actual process does not exist: in this case, the
       generic surface finishing technologies, as defined in Figure 2-1, may not exist within any
       one facility.
•      Simulation allows for process-based modifications and variations, resulting in inherent
       flexibility within the system: models can be designed to vary the sequence of operations,
       add or delete operations, or change process times associated with operations, materials
       flows, and other variables.
•      Simulation modeling facilitates the comparison of technologies by modeling each
       technology operating under a single, consistently applied performance profile developed
       from data collected from industry.
                                            4-62

-------
                                                                       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;
                                          4-63

-------
43. 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 17.5x23.1" (from PWB Workplace Practices Questionnaire data for
       conveyorized processes);
•      panels are placed on the conveyor whenever space on the conveyor is available, and each
       panel requires 18 inches (including space between panels);
•      conveyor speed is constant, thus, the volume (gallons) of chemicals in a bath varies by
       bath type (i.e., microetch, conditioner, etc.) and with the length of the process step (e.g.,
       bath or rinse tank) to provide the necessary contact time (see Table 4-33 for bath
       volumes); and
•      the conveyor  speed, cycle time, and process down time are critical factors that determine
       the time to complete a job.

               Table 4-33. Bath Volumes Used for Conveyorized Processes
Chemical Bath
Cleaner
Microetch
Flux
Solder
OSP
Predip
Immersion Silver
Immersion Tin
Bath Volume by Surface Finishing Alternative
(gallons) ' , ..
HASL
66.5
86.6
13.2
17.4
NA
NA
NA
NA
OSP
66.5
86.6
NA
NA
125
NA
NA
NA
Immersion Silver
66.5
86.6
NA
NA
NA
46.2
142
NA
Immersion Tin
66.5
86.6
NA
NA
NA
46.2
NA
140
NA: Not applicable.
       Non-conveyorized surface finishing process assumptions are as follows:

       the average volume of a chemical bath is 51.1 gallons (from PWB Workplace Practices
       Questionnaire data for non-conveyorized processes);
       only one rack of panels can be placed in a bath at any one time;
       a rack contains 20 panels (based on PWB Workplace Practices Questionnaire data,
       including the dimensions of a bath, the size of a panel, and the average distance between
       panels in a rack);
                                          4-64

-------
                                                                             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 Inputs 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.
Non-Conveyojrized 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 Time6
- (minutes)
7.94
86.8
109
22.6
27.0
a Values may represent chemical products from more than one supplier. For example, two suppliers of nickel/gold
chemical products participated in the project. Input values may underestimate or overestimate those of any one facility,
depending on factors such as individual operating procedures, the chemical or equipment supplier, and the chemical
product used.
b Average values from the PWB Workplace Practices Questionnaire and Performance Demonstration observer sheets.
            Table 4-35. Time-Related Input Values for Couveyorized Processes
Conveyorized Surface
Finishing Technology ,
HASL
OSP
Immersion Silver
Immersion Tin
Time Required to
Replace a Bath b
(minutes)
136
149
114
85
Length of
Conveyor6
(feet)
41.3
54.1
34.0
20.0
Process Cycle
Time6
(minutes)
4.86
5.22
11.2
12.3
Conveyor
Speed c
(ft/min)
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.
c Conveyor speed was calculated by dividing the length of conveyor by the process cycle time.
                                              4-65

-------
   COST ANALYSIS
       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 ssfi'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 ssfgallon for that bath was adjusted by a factor equal to the number of
metal turnovers  (e.g., the replacement criteria for a 750 ssFgal bath with two metal turnovers was
considered to be 1500 ssi/gal of bath). Table 4-36 presents bath replacement criteria used  to
calculate input values for the nickel/gold processes, as an example.  Appendix G presents the bath
replacement criteria recommended by chemical suppliers, and the input values used in this analysis
for the remaining surface finishing technologies.

             Table 4-36. Bath Replacement Criteria for Nickel/Gold Processes
Chemical Bath
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Immersion Gold
Bath Replacement Criteria a
(ssffgal)
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 ssfpanel.
Simulation Model Results

       Simulation models were run for each of the surface finishing processes.  Simulation
outputs used in the cost analysis include:
                                           4-66

-------
                                                                     4.2 COST ANALYSIS
 •      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.

     Table 4-37. Frequency and Duration of Bath Replacements for Non-Conveyorized
                                  Nickel/Gold Process
Chemical Bath
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Immersion Gold
Total
~ Frequency of
- Replacement
7
9
6
4
40
6
72
Avg. Time of Replacement
{minutes)
116
116
116
116
116
116
116
Total Time of Replacement
(minutes)
812
1,044
696
464
4,640
696
8,352
       Table 4-38 presents the other simulation outputs: the total production time required and
the down time incurred by the surface finishing processes while producing 260,000 ssf of PWB.
Total production time is the sum of actual operating time and down time.  The operating time is
based on the process producing 260,000 ssf of PWB and operating 6.8 hr/day.  The down time
includes the remaining 1.2 hr/day that the line is assumed inactive, plus the time the process is
down for bath replacements.  The amount of process down time due to a bath replacement, shown
in Table 4-37, may be adjusted by the model if the bath changeout occurs at the end of the day,
when the replacement duration exceeds the time remaining in the day. (7,670 minutes of
downtine are reported in Table 4-38, indicating that 680 minutes of the 8,352 reported in Table
14-37 occurred at the end of the day.) In this instance, the worker is considered to complete the
bath replacement during the remaining  1.2 hours of the day set aside for process maintenance.
The simulation model output reports for each process are presented in Appendix G.
                                         4-67

-------
4.2 COST ANALYSIS
          Table 4-38.  Production Time and Down Time for the Surface Finishing
                         Processes to Produce 260,000 ssf of PWB
Surface Finishing Process
HASL, Non-conveyorized
HASL, Conveyorized
Nickel/Gold, Non-conveyorized
Nickel/Palladium/Gold, Non-conveyorized
OSP, Non-conveyorized
OSP, Conveyorized
Immersion Silver, Conveyorized
Immersion Tin, Non-conveyorized
Immersion Tin, Conveyorized
Total Production Time"
minutes
17,830
8,890
86,500
114,240
14,360
6,570
26,190
30,680
43,660
days5
43.7
21.8
212
280
35.2
16.1
64.2
75.2
107
Total Down Time*
minutes
2,330
938
7,670
11,380
2,530
1,020
1,390
1,880
1,020
days
5.7
2.3 '
18.8
27.9
6.2
2.5
3.4
4.6
2.5.
  To convert from miflutes to days, divide by 6.8 hr/day (408 minutes).
4.2.4  Activity-Based Costing

       ABC is a method of allocating indirect or overhead costs to the products or processes that
actually incur those costs.  ABC complements the traditional costing /modeling efforts of this
assessment by allowing the cost of tasks that are difficult to quantify, or are just unknown by
industry, to be determined. Activity-based costs are determined by developing a BOA for critical
tasks, which are defined as tasks required to that support the operation of the surface finish
process line. A BOA is a listing of the component activities involved in the performance of a
certain task, together with the number of times each component activity is performed. The BOA
determines the cost of a task by considering the sequence of actions and the resources utilized
while performing that task. In this analysis, the costs of critical tasks determined by a BOA are
combined with the number of times a critical task is performed, derived from simulation results to
determine the total costs of that activity.

       BOAs were developed for the following critical tasks performed during the operation of
the surface finishing process:

•      chemical transport from storage to the surface finishing process;
•      tank cleaning;
•      bath setup;
•      bath sampling and analysis; and
•      filter replacement.

       These BOAs were developed based on information developed for earlier projects
involving similar tasks and on information gathered through site visits and general process
knowledge. The following discussion uses the BOA for chemical transport, presented in Table
4-39, as an example of how BOAs were developed and used. Appendix G presents the BOAs for
the remaining activities.
                                          4-68

-------
                                                                        4.2 COST ANALYSIS
        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 hi central storage via inventory tracking and physical monitoring;
 •       fqrklift 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 hi an assigned area when not in use.

        Each critical task was broken down into primary and secondary activities.  For example,
 chemical transport has six primary activities: paperwork associated with chemical transfer,
 moving forklift to chemical storage area, locating chemicals in storage area, preparation of
 chemicals for transfer, transporting chemicals to the surface finishing process, and transporting
 chemicals from the  surface finishing process to the bath.  The secondary activities for the primary
 activity of "transport chemicals to the surface finishing process" are: move forklift with
 chemicals.; unload line containers, and park forklift in assigned parking area.  For each secondary
 activity the labor, material, and forklift costs are calculated. The forklift costs are a function of
 the time that labor and the forklift are used. On a BOA, the sum of the costs of a set of secondary
 activities equals the cost of the primary activity.

        Continuing the example, for a chemical transport activity that requires two minutes, the
 labor cost is $0.34 (based On a labor rate of $10.24/hour) and the forklift cost is $0.12 (based on
 $0.06/minute). Materials costs are determined for materials other than chemicals and tools
 required for an activity.  The total of $9.28 shown in Table 4-39 represents the cost of a single act
 of transporting chemicals to the surface finishing line.  The same BO As are used for all surface
 finishing technologies because either the activities are similar over all technologies or information
 is unavailable to distinguish between them. However, individual facilities could modify a BOA to
 best represent their unique situations. Table 4-40 presents costs to perform each of the  critical
tasks one time.
                                           4-69

-------
43 COST ANALYSIS
Table 4-39. BOA for Transportation of Chemicals to the Surface Finishing 'Process Line
Activities ?
; ™ T,
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 containers)
2. Utilize correct tools to obtain chemicals
3. Place obtained chemicals in line containers)
4. Close chemical containers)
*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 containers) to bath
2. Clean line container(s)
3. Store line containers) in appropriate area
Time
(min)

2
1
2

2
5
2
3
2

1
2
2

1
3
3
1.5
1.

2
1
2

1
2
1
Resources
Labor a

$0.34
$0.17
$0.34

$0.34
$0.85
$0.34
$0.51
$0.34

$0.17
$0.34
$0.34

$0.17
$0.51
$0.51
$0.26
$0.17

$0.34
$0.17
$0.34

$0.17
$0.34
$0.17
Materials b

$0.10
$0.05
$0.10

$0.00
$0.00
$0.00
$0.00
$0.00

$0.00
$0.00
$0.00

$0.05
$0.05 .
$0.00
$0.00
$0.00

$0.00
$0.00
$0.00

$0.00
$0.20
$0.00
Forklift c

$0.00
$0.00
$0.00

$0.12
$0.30
$0.12
$0.18
$0.12

$0.06
$0.12
$0.12

$0.00
$0.00
$0.00 '
$0.00
$0.06

$0.12
$0.06
$0.12

$0.00
$0.00
$0.00
Total Cost per Transport
Cost
($/transport)

$0.44
$0.22
$0.44

$0.46
$1.15
$0.46
$0.69
$0.46

$0.23
$0.46
$0.46

$0.22
$0.56
$0.51
$0.26
$0.23

$0.46
$0.23
$0.46

$0.17
$0.54
$0.17
$9.28
• Labor rate = $10.24 per hour.
b Materials do not include chemicals or tools.
c Forklift operating cost=$0.06 per minute.
                                               4-70

-------
                                                                        4.2 COST ANALYSIS
                            Table 4-40. Costs of Critical Tasks
Task
Transportation of Chemicals
Tank Cleaning
Bath Setup
Sampling and Analysis
Filter Replacement
Cost
$9.28
. $67.00
$15.10
$3.70
$17.50
 4.2.5  Cost Formulation Details and Sample Calculations

       This section develops and describes in detail the cost formulation used for evaluating the
 surface finishing process alternatives. The overall cost was calculated from individual cost
 categories that are common to, but expected to vary with, the individual process alternatives. The
 cost model was validated by cross-referencing the cost categories with Tellus Institute (White et
 aL, 1992), and Pacific Northwest Pollution Prevention Research Center (Badgett et al., 1995).

       The cost model for an alternative is as follows:
WW + P + MA
                            TC =
where,
TC    =      total cost to produce 260,000 ssf
C     =      capital cost
M     =      material cost
U     =      utility cost
WW  ==      wastewater cost
P     =      production cost
MA   =      maintenance cost

The unit cost of producing 260,000 ssf is then represented as follows:

                          Unit Cost ($/ssf) = TC($) 7260,000 ssf

       The following sections present a detailed description of cost calculation methods together
with sample calculations, using the non-conveyorized nickel/gold process as an example.  Cost
summary tables for all of the process alternatives' are presented at the end of this section.

Capital Costs

       This section presents methods and sample calculations for calculating capital costs.
Capital costs are one-time or periodic costs incurred in the purchase of equipment or facilities. In
this analysis, capital costs include the costs of primary equipment including equipment installation,
and facility space utilized by the surface finishing process line. Primary equipment is the
equipment vital to the operation of the surface finishing process without which the process would
                                           4-71

-------
4.2 COST ANALYSIS
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
I

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 tune in days required to manufacture 260,000 ssf
divided by one operating year (280 days)
       The UR. adjusts annualized costs for the amount of tune required to process 260,000 ssf,
determined from the simulation model for each process alternative. The components of capital
costs are discussed further below followed by sample calculations of capital costs.

       Equipment and Installation Costs. Primary equipment and installation cost estimates
were provided by equipment suppliers and include delivery of equipment, installation, and sales
tax.  Equipment estimates were based on basic, no frills equipment capable of processing the
modeled throughput rates determined by the simulation model, presented in Table 4-38.
Equipment estimates did not include auxiliary equipment, such as statistical process control or
automated sampling equipment sometimes found on surface finishing process lines.

       Annual costs of the equipment (which includes installation) were determined assuming
5-year, straight-line depreciation and no salvage value.  These annual costs were calculated using
the following equation:

                             E =  equipment cost ($) •*• 5 years

       Facility Costs. Facility costs are capital costs for the floor space required to operate the
surface finishing line. Facility costs were calculated assuming industrial floor space costs $76/ft2
and the facility is depreciated over 25 years using straight-line depreciation.  The cost per square
foot of floor space applies to Class A light manufacturing buildings with basements.  This value
was obtained from the Marshall Valuation Service (Vishanoff, 1995) and mean square foot costs
(Ferguson, 1996). Facility costs were calculated using the following equation:

F =  [unit cost of facility utilized ($/fi?) x footprint area/process step (ft2/step) x number of steps] •*- 25 years
                                           4-72

-------
                                                                        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 conveyorized (8' x
 40') and non-conveyorized (51 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
 fWstep, and for non-conveyorized processes as 91 flrVstep. The overall area required for each
 process alternative was then calculated using the following equations:

 Conveyorized:

             Fc  = [$76/ft2x 168 fWstepx number of steps per process] -^-25 years

 Non-conveyorized:
FN =
                           x 91 fP/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 4-3 8 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 fWstep x 14 steps) - 25 yr = $3,870/yr
       UR  = 212 days •*• 280 days/yr = 0.757 yr

       Thus, the capital costs for the non-conveyorized nickel/gold process to produce 260,000
 ssf of PWB are as follows:

       C  = ($9,600/yr + $3,870/yr) x 0.757 yr  =  $10,200
   2 PWB manufacturers and their suppliers use the term "footprint" to refer to the dimensions of process equipment,
such as the dimensions of the surface finishing process line.
                                          4-73

-------
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
           Cost per bath replacement = JLr [chemical product I  cost/bath ($/gal) x
                                      1=1
                  % chemical product I in bath x total volume of bath (gal)]
where,
n
number of chemical products in a bath
       Price quotes were obtained from chemical suppliers in $/gallon or $/lb for process
chemical products. Chemical compositions of the individual process baths were determined from
the corresponding Product Data Sheets submitted by the chemical suppliers of each process
alternative. The average volume of a chemical bath for non-conveyorized processes was
calculated to be 51.1 gallons from PWB Workplace Practices Questionnaire data.  For
conveyorized processes, however, conveyor speed is constant; thus, the volume of chemicals in a
bath varies by bath type to provide the necessary contact tune (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.
                                          4-74

-------
                                                                       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 - 2-, [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 tunes 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 hi this analysis and summaries of chemical
 product cost by supplier and by surface finishing technology.                     "

       Example Materials Cost Calculations.  Table 4-42 presents an example of chemical
 costs per bath replacement for one of the two nickel/gold product lines that were submitted by
 chemical suppliers for evaluation.  From the data in the table, the total cost of chemicals per bath -
 was calculated by multiplying the average chemical cost for a bath (calculated by computing the
 chemical cost per bath of the second product line not shown in Table 4-42, then averaging the
 costs for a bath from both product lines) by the number of bath replacements required to process
 260,000 ssf, as determined by the process simulation. The costs for each bath were then summed
 to give the total materials cost for the overall non-conveyorized nickel/gold product line.  Data for
 each of the product lines submitted, including the other electroless nickel/immersion gold product
 line, are presented in Appendix G.

       Table 4-41 presents the chemical cost per bath replacement, the number of bath
 replacements required, as determined by simulation, the total chemical cost per bath, and the total
 material cost for the non-conveyorized nickel/gold process.  The chemical costs per process bath
 for both product lines were averaged to determine the average chemical cost per bath for the non-
 conveyorized nickel/gold process.  Similar material cost calculations are presented in Appendix G
for each of the surface finishing process alternatives.
                                          4-75

-------
4.2 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
Replacement3
$92.80
$386
$1,640
$315
$890
NAฐ
Number of Bath
Replacements b
7
9
6
4
40.
6
TotaTOiemical
-' Cost *
$649
$3,470
$9,830
$1,260
$35,500
$57,900
Total Materials Cost $108,600
  Reported data represents the chemical cost per bath replacement averaged from two nickel/gold product lines.
b Number of bath replacements required to process 260,000 ssf, as determined by process simulation.
c The immersion gold replacement cost was calculated differently than the other baths because of the wide disparity in
costs and throughput between product lines. The overall cost for the gold bath was calculated for each product line and
then averaged together to give the gold chemical cost for the process.
       Table 4-42. Chemical Cost per Bath Replacement for One Product Line of the
                              Non-Conveyorized Nickel/Gold Process
Bath
Cleaner
Microetch
Catalyst
Acid Dip
Electroless Nickel
Immersion Gold
Chemical
Product
A
B
C
D
E
F
G
H
I
J
K
L
M
N
Product "
Costa($)
$25.0/gal
$5.66/gal
$9.39/gal
$27.3/kg
$1.20/gal
$127/gal
$54.0/gal
$51.0/gal
$29. I/kg
$24.1/gal
$30.9/gal
$28.4/gal
$21.4/gal
$40.0/g
' - Percentage of
Chemical Product b
10
3
3
45g/l
8.5
30
20
12
2g/l
6.6
15
6.6
50
3g/l
Multiplying
Factor c
1
1
1
1
1
1
1
1
1
6
6
5
1
3
Chemical Cost/Bath
Replacement d($)
$128
$266
$2,810
$11.3
$2,390
$70,200
* 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.
c 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 Costper bath calculated assumes abath volume ofSl.l gallons, as determined by PWB Workplace Practices
Questionnaire data for non-conveyorized processes.
                                                 4-76

-------
                                                                       4.2 COST ANALYSIS
 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.
                                          4-77

-------
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:

                                    U = W + E + G
where,
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
                                          4-78

-------
                                                                       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 ~
(S/ccf/month)
$6.30
$2.92
$2.59
$2.22
$1.85
Non-City
Discharge Cost
($/ccf/month)
$7.40
$3.21
$2.85 '
$2.44
'$2.05
Average Discharge
'Cost,
(S/ccfi'month)
$6.85
$3.06
$2.72
$2.33
$1.95
 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 $3.06, and so on. The
production time required to produce 260,000 ssf of PWB comes from the simulation models
Thus, wastewater costs (WW) were calculated as follows:
        WW - L*i [quantity of discharge in tier k (ccf/mo)'x tier cost factor ($/ccf)] x
               k=l
                             required production time (months)
where,
P
ccf    =
number of cost tiers
100 cubic ft
                                          4-79

-------
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 •*• 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 ccฃ7month = $13.70 ccffmonth
       $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/month x 9.1 month = $2,050

Production  Costs

       Production Cost Calculation Methods. Production costs for the surface finishing
process include both the cost of labor required to operate the process and the cost of transporting
chemicals to the production line from storage.  Production costs (P) were calculated by the
following equation:
                                     P = LA + TR
where,
LA
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.
                                          4-80

-------
                                                                      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, and using the number of days required to produce 260,000 ssf of
 PWB from simulation results. A labor wage of $10.24/hr was obtained from the American Wages
 and Salary Survey (Fisher, 1999) and utilized for surface finishing line operators. Therefore, labor
 costs for process alternatives were calculated as follows:

      LA = number of operators x $10.24/hr x 8 hr/day x required production time (days)

        The production cost category of chemical transportation cost includes the cost of
 transporting chemicals from storage to the process line. A BOA, presented in Appendix G, was
 developed and used to calculate the unit cost per chemical transport.  Because chemicals are
 consumed whenever a bath is replaced, the number of trips required to supply the process line
 with chemicals equals the number of bath replacements required to produce 260,000 ssf of PWB;
 Chemical transportation cost was calculated as follows:

            TR  = number of bath replacements x unit cost per chemical transport ($)

        Example Production Cost Calculations. For the example of the non-conveyorized
 nickel/gold, production labor cost was calculated assuming  1.1 operators working for 212 days
 (see Table 4-38). Chemical transportation cost was calculated based on a cost per chemical
 transport of $9.28 (see Table 4-40 and Appendix G) and 72 bath replacements (see Table 4-37).
 Thus, the production cost was calculated as follows:
       LA = 1.1 x $10.24 x 8 hr/day x 212 days = $19,100
       TR = 72 x $9.28 =  $668
thus,
       P= $19,100+ $668  = $19,768

Maintenance Costs

       Maintenance Cost Calculation Method.  The maintenance costs for the surface finishing
process include the costs associated with tank cleaning, bath setup, sampling and analysis of bath
chemistries, and bath filter replacement. Maintenance costs were calculated as follows:
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
                                         4-81

-------
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 hi 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 x UR x $ 17.50

       The total maintenance cost for each process alternative was determined by first calculating
the individual bath maintenance costs using the above equations and then summing the results for
all baths 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 •*• 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 hi 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 = 119x0.76 x $17.50 = $1,580

Therefore, the overall maintenance cost for the process is:
                                          _

-------
                                                                      4.2 COST ANALYSIS
       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) - 260,000 ssf
           =  $156,000-260,000 ssf
           =  $0.60/ssf

       Table 4-44.  Summary of Costs for the Non-Conveyorized Nickel/Goid Process
Cost Category
Capital Cost
Material Cost
Utility Cost
Wastewater Cost
Production Cost
Maintenance Cost
Component
Primary Equipment & Installation
Facility
Chemical(s)
Water
Electricity
Natural Gas
Wastewater Discharge
Transportation of Material
Labor for Line Operation
Tank Cleanup
Bath Setup
Sampling and Analysis
Filter Replacement
Component Cost a
$7,260
$2,930
$109.000
$1,180
$2,360
$0
$2,050
$668
$19,100
$4,820
$1,090
$3,530
$1,580
Total Cost
Totals3
$10,200
$109,000
$3,540
$2,050
$19,800
$11,000
$156,000
a Costs of producing 260,000 ssf of PWB by the process
                                          4-83

-------
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
CostCategory
Capital Cost
Material Cost
Utility Cost
Wastewater Cost
Production Cost
Maintenance Cost
Cost Components ."
•' '" - . ' Y'r- .'"- ':,'
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
v/H&SL, 0>
f -
$11,000
$398
$75,200
$565
$452
$45
$851
$130
$1,790
$938
$211
$249
$482
$92,400
Mckel/Gold,
' NC '
$7,260
$2,930
$109,000
$1,180
$2,360
$0
$2,050.
$668
$19,100
$4,820
. $1,090
$3,530
$1,580
$156,000
                                           4-84

-------
                            4.2 COST ANALYSIS
Table 4-45. Total Cost of Surface Finishing Technologies 1
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
Mckel/PaUadium/GoId,
NC - " y -
$15,400
$6,090
$321,000
$2,060
$4,050
$0
$3,530
$1,030
$25,200
$7,430
$1,680
$8,900
$2,840
$399,000
[cont.)
OSF,
NC
$1,640
$320
$18,500
$441
$313
$66
$702
$159
$3,170
$1,140
$257
$1,610
$330
$28,700
OSP,
c
$2,880
$264
$18,800
$301
$208
$31
$463
$121
$1,320
$871
$196
$738
$151
$26,300
Table 4-45. Total Cost of Surface Finishing Technologies (cont.
Cost Category
Capital Cost
Material Cost
Utility Cost
Wastewater Cost
Production Cost
Maintenance Cost
Cost Components
Primary Equipment & Installation
Facility
Chemical(s)
Water
Electricity
Natural Gas
Wastewater Discharge
Transportation of Material
Labor for Normal Production
TankCIeanup
Bath Setup
Sampling and Testing
Filter Replacement
Total Cost
Immersion
Silver, C
$10,540
$937
$52,700
$301
$739
$140
$529
$167
$5,260
$1,210
$272
$937
$80
$73,800
Immersion
Tin,NC
$2,950
$892
$29,000
$1,030
$494
$162
$1,620
$204
$6,780
$1,470
$332
$1,260
$705
$46,900

Immersion
"Tin,C
$16,800
$2,340
$28,900
$702
$1,230
$240
$1,220
$167
$8,770
: $1,210
$272
$1,800
$1,000
$64,700
4-85

-------
4.2 COST ANALYSIS
  Table 4-46.  Surface Finishing Alternative Unit Costs for Producing 260,000 ssf of PWB
Surface Finishing Alternative
, t \ X~
HASL, Non-conveyorized
HASL, Conveyorized
Nickel/Gold, Non-conveyorized
Nickel/Palladium/Gold, Non-conveyorized
OSP, Non-conveyorized
OSP, Conveyorized
Immersion Silver, Conveyorized
Immersion Tin, Non-conveyorized
Immersion Tin, Conveyorized
Total Cost
($>
94,200
92,400
156,000
399,000
28,700
26,300
73,800
46,900
64,700
Unit Cost
($/ssf|
0.36
0.35
0.60
1.54
0.11
0.10
0.28
0.18
0.25
Cost Savings a
(%)
—
3
-67
-327
69
72
22
50
31
* Cost savings measured by comparing cost of the surface finish to the cost of the baseline non-conveyorized HASL
process. Positive results represent percent savings from the costs incurred had the baseline process been used, while
negative results represent percent lost.
4.2.7  Conclusions

       This analysis generated comparative costs for six surface finishing technologies, including
HASL, nickel/gold, nickel/palladium/gold, OSP, immersion silver, and immersion tin processes.
Costs were developed for each technology and equipment configuration for which data were
available from the PWB Workplace Practices Questionnaire and Performance Demonstration, for
a total of nine processes (five non-conveyorized processes and four Conveyorized processes).
Costs were estimated using a hybrid cost model that combines traditional costs with simulation
modeling and activity-based costs. The cost model was designed to determine the total cost of
processing a specific amount of PWBs through a fully operational surface finishing line, in this
case 260,000 ssf. The cost model does not estimate start-up costs for a facility switching to a
surface finishing alternative, which could factor significantly in the decision to implement a
technology.  Total costs were divided by the throughput (260,000 ssf) to determine a unit cost in
$/ssf.                            .                                                 -

       The cost components  considered include capital costs (primary equipment and installation,
and facility costs), materials costs (limited to chemical costs), utility costs (water, electricity, and
natural gas costs), wastewater costs (limited to wastewater discharge cost), production costs
(production labor and chemical transport costs), and maintenance costs (tank cleanup, bath setup,
sampling and analysis, and filter replacement costs).  Other cost components may contribute
significantly to overall costs, but were not quantified because they could not be reliably estimated.
These include wastewater treatment cost, sludge recycling and disposal cost, other solid waste
disposal costs, and quality costs.

       Overall, the costs ranged from $0.10/ssf for the Conveyorized OSP process to $1.54/ssf
for the non-conveyorized nickel/palladium/gold process. The cost of the baseline non-
conveyorized HASL process was  calculated to be $0.36/ssf.
                                           4-86

-------
                                                                       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 nickel/palladium/immersion gold
process, both of which had chemical costs exceeding the entire cost of the non-conveyorized
HASL process, due to the precious metal content of the surface finish;

       In general, conveyorized processes cost less than non-conveyorized processes of the same
technology due to the cost savings associated with their higher throughput rates. The exception
to this was immersion tin, which was more costly because the combination of process cycle time
and conveyor length resulted in a lower throughput rate than its non-conveyorized version.

       Chemical cost was the single largest component cost for all of the nine processes. Labor
costs were the second largest cost component, though far smaller than the cost of process
chemicals.
                                          4-87

-------
4.3 REGULATORY ASSESSMENT
4.3    REGULATORY ASSESSMENT

       This section describes the federal environmental regulations that may affect the use of
chemicals in the surface finishing processes during PWB manufacturing.  Discharges of these
chemicals may be restricted by air, water, or solid waste regulations, and releases may be
reportable under the federal Toxic Release Inventory (TRT) 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, Tederal 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
DIE PWB Project.  Of the 83  chemicals reportedly used in one or more of the evaluated surface
finishing technologies, no regulatory listings were found for 40 chemicals.

4.3.1  Clean Water Act

       The Clean Water Act (CWA) is the basic federal law governing water pollution control in
the U.S. today.  The various surface finishing processes used by the PWB industry produce  a
number of pollutants that are regulated under the CWA. Applicable provisions, as related to
specific chemicals,  are presented in Table 4-47; these particular provisions and process-based
regulations are discussed in greater detail below.

CWA Hazardous  Substances and Reportable Quantities

       Under Section 31 l(b)(2)(A) of the CWA, the Administrator designates hazardous
substances which, when discharged to navigable waters or adjoining shorelines, present an
imminent and substantial danger to the public health or welfare, including fish, shellfish, wildlife,
shorelines, and beaches.  40 Code of Federal Regulations (CFR) Part 117 establishes the
Reportable Quantity (RQ)  for each substance listed in 40 CFR Part 116. When an amount equal
to or in excess of the RQ is discharged, the facility must provide notice to the federal government
of the discharge, following Department of Transportation requirements set forth in 33 CFR
Section 153.203. Liability for cleanup can result from such discharges. This requirement does
not apply to facilities that discharge the substance under a National Pollutant Discharge
Elimination System (NPDES) Permit or a CWA Section 404 dredge and fill permit, or to  a
publicly owned treatment works (POTW), as long as any applicable effluent limitations or
pretreatment standards have been met. Table 4-47 lists RQs of hazardous substances under the
CWA that may apply to chemicals used in the surface finishing process.
                                          4-88

-------
                                                            4.3 REGULATORY ASSESSMENT
             Table 4-47.  CWA Regulations That May Apply to Chemicals in the
                                 Surface Finishing Process
Chemical3 ,
i A * * -
Acetic acid
Ammonium hydroxide
Copper ion
Copper sulfate pentahydrate
Ethylenediamine
Hydrochloric acid
Nickel sulfate
Nitric acid
Phosphoric acid
Propionic acid
Silver nitrate
Sodium hydroxide
Sulfuric acid
Urea
CWA 311 RQ

-------
43 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 in greater than trace amounts. Each applicant also must indicate if it
knows, or has reason to believe, it discharges any of the other hazardous substances or
non-conventional pollutants listed at 40 CFR Part 122, Appendix D. In some cases, quantitative
testing is required for these pollutants; in other cases, the applicant must describe why it expects
the pollutant to be discharged and provide the results of any quantitative data about its discharge
for that pollutant.

CWA  Effluent Limitation Guidelines [CWA 301(b). 304fbl]

       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-basedeffluent limitation guidelines.

       The effluent limitation to be applied to a particular pollutant in a particular case depends
on the following:

•      whether the pollutant is conventional, non-conventional, or toxic;
•      whether the point source is a new or existing source; and
•      whether the point source discharges directly to the waters of the U.S. or to a POTW.
       (Facilities that discharge to POTWs must comply with the pretreatment standards.)
       Existing sources must comply with either best practicable control technology currently
available (BPT), best conventional pollution control technology (BCT), or best available control
technology economically achievable (BAT) standards.  New facilities must comply with New
Source Performance Standards. NPDES permits also must contain any more stringent permit
limitations based on state water quality standards.

       In the absence of effluent limitation guidelines for a facility category, permit writers
establish technology-based controls using their best professional judgement. In essence, the
permit writer undertakes an effluent guideline-type analysis for a single facility. The permit writer
will use information such as permit limits from similar facilities using similar treatment technology,
performance data from actual operating facilities, and scientific literature.  Best Professional
Judgement may not be used in lieu of existing effluent guidelines. These guidelines apply only to
direct dischargers of wastewater.
                                          4-90

-------
                                                          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
                             Pay PSES Limitations (mg/L)
Pollutant or Pollutant
Property
Cyanide (CN)
Lead(Pb)
Cadmium (Cd)
Max^ Value for Any 1
Day (ppm) _ „
5.0
0.6
1.2
Average Daily Values for 4 Consecutive Monitoring
Days that Shall Not be Exceeded mg/L (ppm)
2.7
0.4
0.7
     Table 4-49. Printed Circuit Board Facilities Discharging 38,000 Liters per Day
                           or More PSES Limitations (mgflL)
Pollutant or Pollutant
Property
Copper (Cu)
Nickel (Ni)
Lead(Pb)
Cadmium (Cd)
Silver (Ag)
Total Metals
Cyanide (CN)
PH
Max. Value for Any 1
Day (ppm)
4.5
4.1
0.6
1.2
1.2
10.5
1.9
7.5 
-------
43 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
, - Chemical3
Acetic acid
Ethylene glycol
Ethylenediamine
Hydrochloric acid
Malic acid
Nickel sulfate
Propionic acid
Sulfuric acid
CAA111
^
/
S

S

S
S
CAAll2b

S

S

S


CAA112r


S
^




regulations discussed.
Abbreviations and Definitions:
CAA - Clean Air Act
CAA III- Standards of Performance for New Stationary Sources of Air Pollutants-Equipment Leaks Chemical List
CAA112b-Hazardous Air Pollutant                                                    i
CAA 112r-RJsk Management Program
Hazardous Air Pollutants

       Section 112 of the CAA established a regulatory program for 188 hazardous air pollutants
and directed EPA to add other pollutants to the list, as needed. EPA is required to establish
Maximum Achievable Control Technology (MACT) standards for source categories that emit at
least one of the pollutants  on the list in major quantities. Chemicals listed in Section 112 (b) of
the CAA that are used in surface finishing are shown in Table 4-53. EPA has identified categories
of industrial facilities that emit substantial quantities of any of these 188 pollutants and plan to
develop emissions limits for those industry categories between 1992 and 2000.

       Section 112(r) of the CAA deals with sudden releases of, or accidents involving acutely
toxic, explosive, or flammable chemicals.  This provision,- added by the CAA Amendments of
1990, establishes a list of substances which, if present in a process in a quantity exceeding a
threshold, would require the facility to establish a Risk Management Program to prevent chemical
accidents. This program must include preparation of a risk management plan, which is submitted
to the state and local emergency planning organizations.
                                           4-93

-------
43 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 required for state operating permit programs.  The permit system is a
new approach established by the 1990 Amendments that is designed to define each source's
requirements and to facilitate enforcement.  In addition, permit fees generate revenue to fund the
program's implementation.

       Any facility defined as a "major source" is required to secure a permit.  Section 70.2 of the
regulations defines a major source, in part, based upon if the source emits or has the potential to
emit:

•      10 tons per year (TPY) or more of any hazardous air pollutant;
•      25 TPY or more of any combination of hazardous air pollutants; or
•      100 TPY of any air pollutant.
       For ozone non-attainment areas, major sources, are defined as sources with the potential to
emit:
•      100 TPY or more of volatile organic compounds (VOCs) or oxides of nitrogen (NOx) in
       areas classified as marginal or moderate;
•      50 TPY or more of VOCs or NOx in areas classified as serious;
•      25 TPY or more of VOCs or NOx in areas classified as severe; and
•      10 TPY or more of VOCs or NOx in areas classified as extreme.-

       Section 70.2 also defines certain other major sources in ozone transport regions and
serious non-attainment areas for carbon monoxide and particulate matter.  In addition to major
sources, all sources that are required to undergo New Source Review, sources that are subject to
New Source Performance Standards or section 112 air toxics standards, and any affected source,
must obtain a permit.

       By November 15,1993, each state was required to submit an operating permit program to
EPA for approval. EPA was required to either approve or disapprove the state's program within
one year after submission. Once approved, the state program went into effect.

       Major sources, as well as other sources identified above, were to submit their permit
applications to the state within one year of approval of the state program.  Once a source submits
a timely and complete application, it may continue to operate until the permit is issued. Permit
issuance may take years because permit processing allows time for terms and conditions to be
reviewed by the public and neighboring states as well as by EPA.

       When issued, the permit includes all federal air requirements applicable to the facility, such
as compliance schedules, emissions monitoring, emergency provisions, self-reporting
responsibilities, and emissions limitations.  States may also choose to include state air
requirements in the permit.  Five years is the maximum permit term.
                                         4-94

-------
                                                           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 CER 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,
 corrosivity, reactivity, and ignitability.  Listed hazardous wastes are specifically named (e.g.,
 discarded commercial toluene, spent non-halogenated solvents). Characteristic hazardous wastes
 are solid waste which "fail" a characteristic test, such as the RCRA test for ignitability.

       There are four separate lists of hazardous wastes in 40 CFR Part 261.  If any waste from a
PWB facility is on any of these lists, the facility is subject to regulation under RCRA (there are
two CBI chemicals used in a surface finishing process that have been identified as "U" listed
wastes).  The listing is often defined by industrial processes, but all wastes are listed because they
were determined to be hazardous (these hazardous constituents are listed in Appendix VII to Part
261).  Section 261.31 lists wastes from non-specific sources and includes wastes generated by
industrial processes that may occur in several different industries; the codes for such wastes
always begin with the letter "F." The second category of listed wastes (40 CFR Section 261.32)
includes hazardous wastes from specific sources; these wastes have codes that begin with the
letter "K." The remaining lists (40 CFR Section 261.33) cover commercial chemical products
that have been or are intended to be discarded; these have two letter designations, "P" and CCU."
                                          4-95

-------
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 of acutely hazardous waste.

       Large and small quantity generators must meet many similar requirements.  40 CFR Part
262 provides that small quality generators may accumulate up to 6,000 kg of hazardous waste
on-site, at any one time, for up to 180 days without being regulated as a treatment, storage, or
disposal facility (TSDF), which requires a TSDF permit. The provisions of 40 CFR 262.34(f)
allow small quality generators to accumulate waste on-site for 270 days without having to apply
for TSDF status, provided the waste must be transported over 200 miles. Large quantity
generators have only a 90-day window to ship wastes off-site without needing a RCRA TSDF
permit.  Keep in mind that most provisions of 40 CFR Parts 264 and 265 (for hazardous waste
treatment, storage and disposal facilities) do not apply to generators who send their wastes
off-site within the 90- or. 180-day window, whichever is applicable.   '

       Hazardous waste generators that do not meet the conditions for being conditionally
exempt, small quantity generators must (among other requirements such as record keeping and
reporting):

•      obtain a generator identification number;                                           ,
•      accumulate and ship hazardous waste in suitable containers  or tanks (for accumulation
       only);
•      manifest the waste properly;
•      maintain copies of the manifest, a shipment log covering all  hazardous waste shipments,
       and test records;
•      comply with applicable land disposal restriction requirements; and
•      report releases or threats of releases of hazardous waste.
                                          4-96

-------
                                                          43 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.

 CERCLA ROs

       Substances defined as hazardous under CERCLA are listed in 40 CFR Section 302.4.
 Under CERCLA, EPA has assigned a RQ to most hazardous substances; regulatory RQs are
 either 1,10,100,1,000, or 5,000 pounds (except for radionuclides). If EPA has not assigned a
 regulatory RQ to a hazardous substance, typically its RQ is one pound (Section 102). Any person
 in charge of a facility (or vessel) must immediately notify the National Response Center as soon as
 a person has.knowledge of a hazardous substance release in an amount that is equal to or greater
 than its RQ. There are some exceptions to this requirement, including the exceptions for federally
 permitted releases. There is also streamlined reporting for certain continuous releases (see 40
 CFR 302.8).  Table 4-54 lists RQs of substances under CERCLA that may apply to chemicals
 used in surface finishing processes.

 Table 4-54. CERCLA RQs That May Apply to Chemicals in the Surface Finishing Process
- Chemical a
Acetic acid
Ammonium hydroxide
Copper ion
Ethylene glycol
Ethylenediamine
Hydrochloric acid
Nickel sulfate
CERCLA RQ0bs)
5,000
1,000
1
5,000
5,000
5,000
100
Chemical a
Phosphoric acid
Propionic acid
Silver nitrate
Sodium hydroxide
Sulfuric acid
Thiourea

CERCLA RQflbs)
5,000
5,000
1
1,000
1,000
10

Abbreviations and Definitions:
CERCLA - Comprehensive Environmental Response, Compensation and Liability Act
RQ - Reportable Quantity
                                         4-97

-------
4.3 REGULATORY ASSESSMENT
CERCIA Liability

       CERCLA further makes a broad class of parties liable for the costs of removal or
remediation of the release, or threatened release, of any hazardous substance at a facility. Section
107 specifies the parties liable for response costs, including the following:  1) current owners and
operators of the facility; 2) owners and operators of a facility at the time hazardous substances
were disposed; 3) persons who arranged for disposal, treatment, or transportation for disposal or
treatment of such substances; and 4) persons who accepted such substances for transportation for
disposal or treatment. These parties are liable for: 1) all costs of removal or remedial action
incurred by the federal government, a state, or an Indian tribe not inconsistent with the National
Contingency Plan (NCP); 2) any other necessary costs of response incurred by any person
consistent with the NCP; 3) damages for injury to natural resources; and d) costs of health
assessments.

4.3.5   Superfund Amendments and Reauthorization Act and Emergency Planning
       and Community Right-To-Know Act

       CERCLA was amended in 1986 by the Superfund Amendments and Reauthorization Act
(SARA). Title IH of SARA is also known as the Emergency Planning and Community
Right-To-Know Act (EPCRA).  Certain sections of SARA and EPCRA may be applicable to
surface finishing chemicals and PWB manufacturers.  Table 4-55 lists applicable provisions as
related to specific chemicals.

SARA Priority Contaminants

       SARA Section 110 addresses Superfund site priority contaminants.  This list contains the
275 highest-ranking substances of the approximately 700 prioritized substances. These chemical
substances, found at Superfund sites, are prioritized based on their frequency of occurrence,
toxicity rating, and potential human exposure. Once a substance has been listed, the Agency for
Toxic Substances and Disease Registry must develop a toxicological profile containing general
health/hazard assessments with effect levels, potential exposures, uses, regulatory actions, and
further research needs.

EPCRA Extremely Hazardous Substances         .

       Section 302(a) of EPCRA regulates extremely hazardous substances and is intended to
facilitate emergency planning for response to  sudden toxic chemical releases.  Facilities must
notify the State Emergency Response Commission (SERC) if these chemicals are present in
quantities greater than then* threshold planning quantities.  These same substances also are subject
to regulation under EPCRA Section 304, which requires accidental releases in excess of
reportable quantities to be reported to the SERC and Local Emergency Planning Committee.
                                          4-98

-------
                                                            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
Chemical a
Ammonium hydroxide
Copper ion
Copper sulfate pentahydrate
Ethylene glycol
Ethylenediamine
Nickel sulfate
Palladium chloride
Phosphoric acid
Sulfuric acid
.SARAllO ,

S
/


/
/


EPCRA 302a




S



S
EPCRA 313
/
^
S
S

S

S
S
regulations discussed.
Abbreviations and definitions:
SARA - Superfund Amendments and Reauthorization Act
SARA 110 - Superfund 'Site Priority Contaminant
EPCRA - Emergency Planning & Community Right-To-Kriow Act
EPCRA 302a - Extremely Hazardous Substances
EPCRA 313 - Toxic Chemical Release Inventory
                                          4-99

-------
43 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 a
Ethyleneglycol
Palladium chloride
: TSCASdHSDR


TSCA MTL
S

TSCA 8a PAIR

^
* In addition to the chemicals listed, there are 10 CBI chemicals identified as falling under the TSCA regulations
discussed.
Definitions and abbreviations:  .
TSCA - Toxic Substances Control Act
TSCA 8d HSDR - Health & Safely Data Reporting Rules
TSCA MTL - Master Testing List
TSCA SaPAIR - 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 rinding that such testing is necessary, due
to insufficient data from which the chemical's effects can be predicted, and that either:  1)
activities involving the chemical may present an unreasonable risk of injury to health or the
environment; or 2) the chemical is produced in substantial quantities and enters the environment in
substantial quantities, or there is significant or substantial human or environmental exposure to the
chemical.

       The TSCA Master Testing List (MTL) is a list compiled by the EPA Office of Pollution
Prevention and Toxics to  set the Agency's testing agenda.  The major purposes are to:  1) identify
chemical testing needs; 2) focus limited EPA resources on those chemicals with the highest
priority testing needs; 3) publicize EPA's testing priorities for industrial chemicals; 4) obtain
broad public comments on EPA's testing program and priorities; and 5) encourage initiatives by
industry to help EPA meet those priority needs. The 1996 MTL now contains over 500 specific
chemicals in 10 categories.
                                          4-100

-------
                                                         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 in this rule, must report information on
production volume, environmental releases, and certain other releases. Small manufacturers or
importers may be required to report such information on certain chemicals.

4.3.7  Summary of Regulations for Surface Finishing Technologies

       Tables 4-57 through 4-62 provide a summary of regulations that may apply to chemicals in
each of the surface finishing technology categories.
                                        4-101

-------
43 REGULATORY ASSESSMENT

5
<
*c
a
j
H
/— *


^t
te
w
n
e
c
a>
h-J


GO

a
e
•a
•>
•w
CS
s
.S
'•C
3
i
CM
0
CS
E
s
CO
•
^^
tf)
•3
00 •ซ
PH
^
H ' •
•S:
.,
•S
OO HM
K
i.'--'

";'04 "
f


CO ' '

__

S

JJ
T-( ,;

T— 1
•ฃ? '.

AJ^.
TH
co
ctt
JJ'"
s
CO
• ' \
3 ':
6
tn
en
U
u
M : •• •















S




S










S



s
S

1
•&
•a
(Copper sulfate peni

.







%










S






S

S










1
3)
I
(3









s^







S


S



S


S






^*





[Hydrochloric acid









^







S






















Hydrogen peroxide




















S













^s





|Phosphoric acid


































S





Sodium hydroxide

















S











S




s





Sulfuricacid






en


,_,



en






CN






^

ป-i










42
1
3
3
q
O
e
o
ฅ
CO
"3
o
s
U
•S

S
tH
^>
e
SS
 O
^ t-i
3 ^
ie process s
micals for '
II

0> O
3 -g
ซ 'S
Note: For technologle
regulation reflects the
                                  4-102

-------
                  4.3 REGUIATORY ASSESSMENT




Si
o
c
&
A
o

5?
^
JJjj
.a
.5
Jp
05

ซ

1

o
*"*
^^
^
o

-w
ulations th
W

MH
f?
S
s
8
00

?
V
3

•









-

^~

^ f

I
A


2
•g
yซ
a,
a,
•
ts
OS

-S*
I
&+




3
ฃ




.j
-Jj
***
M


1
cw




^ซ
d


^

p" >
1-1

>

p

*

fft
es P
00 •<
PH
1
PS!
*o O
00 or.
a
1


T-l 1

•is
1
o
u
JO
s
2

•J3
g

g
g
•"
O
•d"
S)
•3
*ซ^
o
G
CD"
*c3
jmitted for ev
th processed.
•3 O.
5 J=

05 c2
0 ^
2 'ง
I!
s ^i

1 J.
ซ s
Note: For technologie
regulation reflects the
                               CO CO CO
                               H H H
4-103

-------
4.3 REGULATORY ASSESSMENT

;>..
fcjf
O
O
e
C.
ฃ
2
a
i
.5
s
1
^
*-*
.is
TS
U
"S
•tm
O!

^
ฃ
B
•"5
et
jฃป3
W
CS
pis
en
O
•-o
•S
U

O
g?

E
E
-i


j^^
ซ
3















e
o
Vf
o
"S?
S
1,
a,
o
S
f
w
•a
u
'is
ป :
"T* I™N
O *O
•f* ftj
p^
_.
1-1
I
JS .'
o
co



•a- . -
.2
S -:-•
u
en
S
o
















S

















S








•8
| Ammonium hydrox
















S
S












S




S

S

.
^
Copper sulfate pent












































1 Ethylenediamine









S



S


S




S

S










S









1 Hydrochloric acid









S



S






























| Hydrogen peroxide
















S
S





S






S



S
S

S






1
z
1






S










S


























1 Palladium chloride
















S

















S









1 Phosphoric acid












































1 Propionicacid












































1 Sodium chloride


































S









1 Sodium hydroxide













S












S







S









1
-g
ซง
eg






co



jj

0>
ง
1
03
O *O
t2 CO
•0 S
(U O
p k
"ง ฐ
CO fc
8ซ8
8 "3
a. -2
|-S
J3 O

S -R
S g
J3 0

^ S
Note: For technologie
regulation reflects the 1
                                    4-104

-------
              4.3 REGULATORY ASSESSMENT
             o
             on

             ง
            i

                           -
              •.

              02
            ซ.   s
                lt
l
                     S


                 ,

                      f!



                  l

4-105

-------
        ATORY ASSESSMENT
.a
|


ฃ
w
es

S



CO



vo

4

-ง
s <
                                                 ง
If
.11
    _ ง1
    a "<3 o-

    &o N -a
    •S S S
""   ง   S


s IE i s

                                  <:oo
-------
                 4.3 REGULATORY ASSESSMENT
4-107

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

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

Fisher, Helen S.  1999. American Wages and Salary Survey, 3rd Ed.  Detroit, MI: Gale
Research Inc. (An International Thompson Publishing Co.)

Iman, R.L. et al.  1995. "Evaluation of Low-Residue Soldering for Military and Commercial
Applications: A Report from the Low-Residue Soldering Task Force." June.

Iman, R.L., J.F. Koon, et al.  1997.  "Screening Test Results for Developing Guidelines for
Confbnnal Coating Usage and for Evaluating Alternative Surface Finishes."  CCAMTF Report.
June.

Iman, R.L., J. Fry, R. Ragan, J.F. Koon and J. Bradford.  1998. "A Gauge Repeatability and
Reproducibility Study for the CCAMTF Automated Test Set." CCAMTF Report. March.

Joint Group on Acquisition Pollution Prevention (JG-APP) Joint Test Protocol CC-P-1-1 for
Validation of alternatives to Lead-Containing Surface Finishes, for Development of Guidelines for
Conformal Coating Usage, and for Qualification of Low-VOC Conformal Coatings.  1998.

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

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

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

White, Allan L., Monica Becker and James Goldstein.  1992.  Total Cost Assessment.
Accelerating Industrial Pollution Prevention Through Innovative Project Financial Analysis:
With Application to Pulp and Paper Industry. U.S. EPA's Office of Pollution Prevention and
Toxics, Washington, DC.

U.S. EPA (Environmental Protection Agency).  1998.  PollutionJPrevention and Control Survey.
                                        4-108

-------
                                       Chapters
                                   Conservation
        Businesses are finding that by conserving resources, both natural and man-made, and
 conserving energy, they can cut costs, improve the environment, and improve their
 competitiveness. Due to the substantial amount of rinse water consumed and wastewater
 generated by the printed wiring board (PWB) manufacturing process, water conservation is an
 issue of particular concern to board manufacturers and to the communities in which they are
 located. This chapter of the Cleaner Technologies Substitutes Assessment (CTSA) evaluates the
 comparative resource consumption and energy use of the surface finishing technologies. Section
 5.1 presents a comparative analysis of the resource consumption rates of the surface finishing
 technologies, including the relative amounts of rinse water and metals consumed, and a discussion
 of factors affecting process and wastewater treatment chemicals consumption.  Section 5,2
 presents a comparative analysis of the energy impacts of the. surface finishing technologies,
 including the relative amount of energy consumed by each process and the environmental impacts
 of the energy consumption.
 5.1    RESOURCE CONSERVATION

       Resource conservation is an increasingly important goal for all industry sectors,
 particularly as-global industrialization increases demand for limited resources. A PWB
 manufacturer can conserve resources through its selection of a surface finishing process and the
 manner in which it is operated. By reducing the consumption of resources, a manufacturer will
 not only minimize process costs and increase process efficiency, but also will conserve resources
 throughout the entire life-cycle chain. Resources typically consumed by the operation of the
 surface finishing process include water used for rinsing panels, metals that form the basis of many
 of the surface finishing technologies, process chemicals used on the process line, wastewater
 treatment chemicals,  and energy used to heat process baths and power equipment.  A summary of
 the effects of the surface finishing technology on the consumption of resources is presented in
 Table 5-1.

       To determine the effects that surface finishing technologies have on the rate of resource
 consumption during the operation of the surface finishing process, specific data were gathered
 from chemical suppliers of the various technologies, Performance Demonstration participants, and
from PWB  manufacturers through the Workplace Practices Questionnaire and Observer Data'
 Sheets. Data gathered through these means to determine resource consumption rates include:

•      process specifications (e.g., type of process, facility size, process throughput, etc.);
       physical process parameters and equipment description (e.g., automation level, bath size,
       rinse water system configuration, pollution prevention equipment, etc.);
       operating procedures and employee practices (e.g., process cycle-time,'individual bath
       dwell times, bath maintenance practices, chemical disposal procedures, etc.); and
       resource consumption data (e.g., rinse water flow rates, frequency of bath replacement,
       criteria for replacement, bath formulations, frequency of chemical addition, etc.).

                                          5-.1

-------
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
cornmunity 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 hi 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 Types
Rinse Type - - , ~ a\
*. * *" f „""•'" *•,„ ป n <
' . >- ' -s. >$ ^ * ^
Water Rinse, Non-conveyorized
Water Rinse, Conveyorized
High Pressure Water Rinse, All automation types
Normalized Water Row Rate a
(gal/ssf)
0.258
0.176
0.465
* 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:
                                            [NRS; xNWCRJ
where,
              =      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 affect on the amount of water that a facility will consume
during normal operation of the surface finishing process line.  Five surface finishing processes
consume less water than the baseline HASL process, including the 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/palkdium/gold technologies.
                                           5-4

-------
                                                                  5.1 RESOURCE CONSERVATION
                Table 5-3. Rinse Water Consumption Rates and Total Water
                         Consumed by Surface Finishing Technologies
Surface Finishing Technology
^ y -* *>*,? *• & ^ -~ - x. ~^
* ~ j ^
A^ •ฃ >-
^ " -~" •"- -i *"""
,s - ^ -ป . - p. v
*" ^ ^ ป "~ *ซ -a. ~ ^ ,, *-
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"
Normal
Mow
3
3
8
14
3
3
3
7
5
Hi^
Pressure
1
1

-
-
-
-
•• -
-
Total Water
Consumption
Rate*
(gal/ssf)
1.24
0.99
2.06
3.61
0.77
0.53
0.53
1.81
0.88
Rinse Water
Consumed
(gal/260,000 ssf)
3.22 xlO5
2.58x1 0s
5.37 xlO5
9.39 x 10s
2.01 x 10s
1.37 x 10s
1.37 x 10s
4.69 xlO5
2,29 xlO5
f Data reflects the number of rinse stages required for the standard configuration of each surface finishing technology as
reported in Section 3.1, Source Release Assessment.          '
b Rinse water consumption rate was calculated by multiplying the number of rinse stages for each rinse type by the
corresponding consumption factor listed in Table 5-2. The individual rates were then totaled and divided by 1,000 to
determine the overall consumption rate for mat technology.
                Immersion Silver (c)

                         OSP(c)

                     ,   OSP(nc)

                  Immersion Tin (c)

                        HASL(c)

                       HASL(nc)

                 Immersion Tin (nc)

                   JvSckeVGoId (nc)

           ^ketfRailadiunYGold (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 minimize water usage, some companies have gone a step farther by developing
equipment systems that monitor water quality and usage in order to optimize water rinse
performance. This pollution prevention technique is recommended to reduce both water
consumption and wastewater generation.  The actual water usage experienced by manufacturers
employing such a system may be less than that calculated in Table 5-3.

Metal Consumption

       Many of the surface finishes are formed by the deposition of metal ions onto the surface of
the PWB, forming a reliable, solderable finish for further assembly.  The metals range from
relatively inexpensive, widely available metals such as tin and lead, found in solder, to expensive
'precious' metals such as silver, gold, and palladium. While a portion of the metal consumed can
be found in the surface finish of the PWB, metal is also lost through drag-out of the plating bath
to subsequent stages, and through the replacement of spent or contaminated plating solutions.  In
the case of HASL, solder is also lost through the continual removal of dross, a film of
contaminated solder.

       The amount of metal consumed through the deposition, or plating, of the surface finish is
dependent on the thickness of the metal deposit, the amount of PWB surface area  that must be
plated, and the density of the metal being applied. The recommended plating thickness for a
surface finishing technology can be obtained from the appropriate chemical supplier. In addition,
plating specifications for surface finishes have been established through testing by both chemical
suppliers and by industry. These specifications set forth strict guidelines on minimum plating
thicknesses required to .insure a reliable, solderable surface finish. The metal deposition rates and
the total metal deposited by the surface finishing technologies are presented in Table 5-4.  .
                                          5-6

-------
                                                            5.1 RESOURCE CONSERVATION
             Table 5-4. Metal Deposition Rates and Total Metal Consumed by
                              Surface Finishing Technologies
Process -*"*"
J -
--^ - f *. ~ -S^
HASL

Nickel/Gold,
Nickel/Palladium/Gold

Immersion Silver
Immersion Tin
z- Metal
( -~
^
Tin
Lead
Nickel
Palladium
Gold
Silver
Tin
Density *
~ 
-------
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 the surface finishing technologies is
presented below.

Process Chemicals Consumption

       Bath chemicals that constitute the various chemical baths or process steps are consumed in
large quantities during the normal operation of the surface finishing process, either through co-
deposition with the metals  onto the surface of the PWB or degradation through chemical reaction.
Process chemicals are also  lost through volatilization, bath depletion, bath drag-out to subsequent
process stages, or contamination as PWBs are cycled through the surface finishing process. Lost
or consumed process chemicals are replaced through chemical additions, or if the build-up of
contaminants is too great, the bath is replaced. Methods for limiting unnecessary chemical loss
and thus minimizing the amount of chemicals consumed are presented in Chapter 6 in this CTSA.

       Presenting a chemical-by-chemical analysis of process chemical consumption is not
possible without disclosing the composition and concentration of the proprietary chemical
formulations collected from the chemical suppliers (the actual chemical consumption is a
combination of the quantity and concentration of chemicals present, factors which vary greatly,
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
                                           5-9

-------
5.1 RESOURCE CONSERVATION
replacement criteria include ssf of PWB processed and elapsed time since the last bath
replacement.  The practice of making regular adjustments to the bath chemistry through additions
of process chemicals consumes process chemicals, but will extend the operating life of the process
baths, reducing chemical use over time. Despite the supplier recommendations, project data
showed a wide range of bath replacement practices and criteria for manufacturing facilities
operating the same, as well as different, surface finishing technologies.

Wastewater Treatment Chemicals Consumption

       The extent to which the consumption of treatment chemicals will be reduced, if any, is
dependent on several factors, some of which include the rate at which wastewater is generated
(e.g., the amount of rinse water consumed), the type of treatment chemicals used, composition of
waste streams from other plant processes, percentage of treatment plant throughput attributable
to the surface finishing process, the resulting reduction hi 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
                                          5-10

-------
                                                           5.1 RESOURCE CONSERVATION
technology.  In cases where waste solder is not routinely recycled or reclaimed, the consumption
of as much as 2,500 pounds of lead could be eliminated by replacement of the HASL process.
Although several of the alternative technologies rely on the use of small quantities of other metals,
the OSP technology eliminates metal consumption entirely. Other factors influencing metal
consumption were identified and discussed.

       A quantitative analysis of both process chemicals and treatment chemicals consumption
could not be performed due to the variability of factors that affect the consumption of these
resources, and for reasons of confidentiality. The role the surface finishing process has in the
consumption of these resources was presented and the factors affecting the consumption rates
were identified and discussed.
                                          5-11

-------
5.2 ENERGY IMPACTS
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 the surface finishing process, where
energy is consumed by process equipment such as immersion heaters, fluid and air pumps,
agitation devices such as vibrating motors, and by conveyorized transport systems. The focus of
this section is to perform a comparative analysis of the relative energy consumption rates of the
baseline HASL process and alternative surface finishing technologies.

       Data collected for this analysis focus on the energy consumed during the application of the
surface finish. Traditional life-cycle analysis indicates that energy consumption during other life-
cycle stages also can be significant and should be considered when possible.  Although a
quantitative life-cycle analysis is beyond the scope and resources of this project, the impacts to the
environment firom the manufacture of the energy required by the surface finishing process is
briefly analyzed and presented at the end of this  chapter.

       Section 5.2.1 discusses energy consumption during the application of the  surface finish,
while Section 5.2.2 discusses the environmental  impacts of this energy consumption.  Section
5.2.3 briefly discusses the energy consumption of other life-cycle stages.  Section 5.2.4 presents,
conclusions  of the comparative energy analysis.

5.2.1  Energy Consumption During Surface  Finishing Process Operation

       To determine the relative rates of energy consumption during the operation of the surface
finishing technologies, specific data were collected regarding energy consumption through the
Performance Demonstration project and through dissemination of the PWB Workplace Practices
Questionnaire to industry members. Energy data collected include the following:

•      process specifications (i.e., type of process, facility size, etc.);
•      physical process parameters (i.e., number of process baths, bath size, bath conditions such
       as temperature and mixing, etc.);
•      process automation (i.e., conveyorized, computer-controlled hoist, manual, etc.);
•      equipment description (i.e., heater, pump, motor, etc.); and
•      equipment energy specifications (i.e., electric load, duty, nominal power rating,
       horsepower,  etc.).

       Each of the surface finishing technologies contains a-series of chemical baths that are
typically separated by one or more water rinse steps.  In some processes, these chemical stages
are supplemented by other stages such as a drying oven or a HASL machine, which applies the
solder to the PWB using a mechanical type of process. In order for the process to perform
properly, each process stage should be operated  within specific supplier recommended
parameters, such as parameters for bath temperature and mixing, oven temperatures, or air knife
                                          5-12

-------
                                                                    5.2 ENERGY IMPACTS
pressures.  Maintaining these process stages within the desired parameters often requires energy-
consuming equipment such as immersion heaters, fluid circulation pumps, and air compressors.  In
addition, the degree of process automation affects the relative rate of energy consumption.
Clearly, conveyorized equipment requires energy to operate, but also non-conveyorized systems
require additional equipment not found in conveyorized systems, such as panel agitation
equipment.

       Table 5-5 lists the types of energy-consuming equipment typically used during the
operation of a surface finishing process and the function of the equipment. In some cases, one
piece of equipment may be used to perform a function for the entire process line. For example,  in
a non-conveyorized system, panel vibration is typically performed by a single motor used to rock
an apparatus that extends over all of the process tanks.  The apparatus provides agitation to each
individual panel rack that is connected to it, thus requiring only a single motor to provide
agitation to every bath on the process line that may require it. Other equipment types such as
immersion heaters affect only one process stage, so each process bath or stage may require a
separate piece of energy-consuming equipment

     Table 5-5. Energy-Consuming Equipment Used in Surface Finishing Process Lines
Type of Equipment
Conveyor Drive Motor
Immersion Heater
Fluid Pump
Air Pump
Panel Agitation Motor
Gas Heater
Solder Pot
Ventilation Equipment
- ' - ' -' Function v"' " -r
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
                                          5-13

-------
5.2 ENERGY IMPACTS
to the function of the energy-consuming equipment.  For example, a typical 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
<-*'"„ /""FuncfionofEqiBptaent* v '„,* ^ '"'<-* %
Conveyor
0
1
0
0
0
1
1
0
1
Panel '
Agitation b
r
0
i
i
i
0
0
i
0
Bath
Heat
1
1
4
6
2
2
2
3
3
Air Knife/
Sparging c
2
2
1
1
2
2
0
0
0
Fluid
Circulation
3
4
3
3
3
3
4
4
3
Panel
Drying
1
1
0
0
1
1
1
1
J
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.
c Air sparging is used selectively by some manufacturers to enhance bath performance. Sparging may not be required
for all product lines or facilities using a surface finishing technology.
       The electrical energy consumption of surface finishing line equipment, as well as
equipment specifications (power rating, average duty, and operating load), were collected during
the Performance Demonstration.  In cases where electricity consumption data were not available,
the electricity consumption rate was calculated using the following equation:
                          EC = NPRxOLxADx(lkW/0.746HP)
where,
                                            5-14

-------
                                                                     5.2 ENERGY IMPACTS
 EC    =      electricity consumption rate (kWh/day)
 NPR   =      nominal power rating (HP)     -    '  .    '
 OL    =      operating load (percent), or the percentage of the maximum load or output of
               the equipment that is being used
 AD    =      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 " *"
"" ~* ., „' - v * _ ""
_, t
-V..V - • '
Conveyorized Panel Automation
Panel Agitation
Bath Heater
Fluid Circulation
Air Knife/Sparging
Panel Drying
Solder Heater
TypeflfEquipment "
V,
V
Conveyor System
Panel Agitation Motor
Immersion Heater
Fluid Pump
Air Pump
Gas Drying Oven
Solder Pot
- Energy Consumption Rates Per -
- Equipment Type
Electricity a
- 
-------
5.2 ENERGY IMPACTS
       Natural gas consumption rates were calculated using a similar method. The individual
energy consumption rates for both natural gas and electricity were then converted to British
Thermal Units (Btu) per hour and summed to give the total energy consumption rate for each
surface finishing technology. The individual consumption rates for both natural gas and
electricity, as well as the hourly energy consumption rate calculated for each of the surface
finishing technologies, are listed in Table 5-8.

       These energy consumption rates include only the types of equipment listed in Table 5-5,
which are commonly recommended by chemical suppliers to successfully operate a surface
finishing process.  However, equipment such as ultrasonics, automated chemical feed pumps,
vibration units, panel feed systems, or other types of electrically powered equipment may be part
of the surface process line.  The use of this equipment may improve the performance of the
surface finishing process, but is not required in a typical process for any of the surface finishing
technologies.

     Table 5-8. Hourly Energy Consumption Rates for Surface Finishing Technologies
Process Type *'" * *" >"""
l ^ - ", ~ -r> ,
„ ,v * i
- -~ ^ K
_ -. „. - w *- ' /
f \ -' /,/,> ^ ' '
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
}- I* *ฃ -r
-------
                                                                     5.2 ENERGY IMPACTS
              Table 5-9. Energy Consumption Rate per ssf of PWB Produced
                            for Surface Finishing Technologies
< Process Type
• . - -_ , ' r ^ ^ - ^
-, v ^ •* v-
HASL, Non^conveyorized
HASL, Conveyorized
Nickel/Gold, Non-conveyorized
Nickel/Palladium/Gold, Non-conveyorized
OSP, Non-conveyorized
OSP, Conveyorized
Immersion Silver, Conveyorized
Immersion Tin, Non-conveyorized
Immersion Tin, Conveyorized
.-Process
Operating Time a
> (hours)
258
133
1,310
1,710
197
93
414
480
710
" Total Energy
Consumed .
(Btu/260,000 ssf)
5.67 x 107
3.46 x 107
1.16x10*
2.00 x 108
3.26 xlO7
1.89xl07
7.46 x 1C7
7.52 xlO7
1.36 xlO8
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.

       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 ofiset 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
                                           5-17

-------
5.2 ENERGY IMPACTS
operated in both conveyorized and non-conveyorized modes.  This table shows that, although the
conveyorized version of all three processes requires more energy per hour to operate than the
non-conveyorized mode, the added efficiency of the conveyorized system (reflected in the shorter
operating time) results in less energy usage per ssf of board produced.  The immersion tin
processes are the exceptions. The non-conveyorized configuration of this process not only has a
better hourly consumption rate than the conveyorized, but also benefits from a faster operating
time, a condition due to the long overall cycle-time required for the conveyorized process. These
factors combine to give the non-conveyorized immersion tin process a lower energy consumption
rate than the conveyorized version.  Despite this exception, the overall efficiency of conveyorized
systems typically will result in less energy usage per ssf of board produced, as it did for both the
HASL and OSP processes.

               Table 5-10. Effects of Automation on Energy Consumption
                           for Surface Finishing Technologies
Process Type* - -
';"--•- _ v*^
HASL, Non-conveyorized
HASL, Conveyorized
OSP, Non-conveyorized
OSP, Conveyorized
Immersion Tin, Non-conveyorized
Immersion Tin, Conveyorized
-Hourly *
Consumption Rate
(l,OOOBtu/ssf)
220
260
165
203
156
191
Process _
Operating Time~a
(hours)"
258
133
197
93
480
710
Energy Consumption
Rate
(Bto/ssf)
218
133
125
73
289
522. •
* Times listed represent the operating time required to manufacture 260,000 ssf of PWB by each process as simulated
by computer model. Operating time was considered to be the overall process time minus the downtime of the process.
       Finally, it should be noted that the overall energy use experienced by a facility will depend
greatly upon the operating practices and the energy conservation measures adopted by that
facility.  To minimize energy use, several simple energy conservation opportunities are available
and should be implemented. These include insulating heated process baths, using thermostats on
heaters, and turning off equipment when not in use.

5.2.2  Energy Consumption Environmental Impacts

       The production of energy results in the release of pollution into the environment, including
pollutants such as .carbon dioxide (CO2), sulfur oxides (SOJ, carbon monoxide (CO), sulfuric acid
(Ky3O4), and participate matter. The type and quantity of pollution depends on the method of
energy production. Typical energy production facilities in the U.S. include hydroelectric, nuclear,
and coal-fired generating plants.

       The environmental impacts attributable to energy production resulting from the differences
in energy consumption among surface finishing technologies were evaluated using a computer
program developed by the EPA National Risk Management Research Laboratory, P2P- version
1.50214 (U.S. EPA, 1994). This program can, among other things, estimate the type and quantity
                                          5-18

-------
                                                                    5.2 ENERGY IMPACTS
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 hi 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 CO^ solid wastes, SO^ 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

-------
5.2 ENERGY IMPACTS
AK
   s
   a>
         'O
       co
si
o ซ
CO >•
S



1

ff
        -
       m

       O
       -Q
       O
       2
       •o

       &
       Si en
       A Tป
       ^ S3
       s-
       o'
       a
         cu
       sis
  I
  s
   BO

  I
  •a
  s
   s
                OJ
             o
                CO
               O

           01
                      01
                      O-i
                      VO
                      00
                      VO
                      0
                      o>
                      0)
O
VO
             -rf
             s
             II
             =1 ^
             II
             is
                  OO
                  o
                  o
                  o
                              01
                              VO
                              OJ
                         01
                         01
            ON
            -sf
                                  vo
                           oo

                           01
                         oo
                           en
                           o
                           en
                           01
                         o
                         o
                    s
                       o
                       o
          o
          o
                                  O
                         en
                         o
                       00
nveyo
                           CO
tal
ta
                             ฃ
                                5-20

-------
                                                                      5.2 ENERGY IMPACTS
             Table 5-12. Pollutant Environmental and Human Health Concerns
. 'Pollutant -
< '
Carbon Dioxide (CQJ
Carbon Monoxide (CO)
Dissolved Solids
Hydrocarbons
Nitrogen Oxides (NOJ
Particulates
Solid Wastes
Sulfur Oxides (SOJ
Sulfuric Acid (H2SO4)
ป Medium
ofRelease
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 wanning, 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 Particulate 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 in pollutant releases to all media, with a wide range of possible affects.
Minimizing the amount of energy usage by the surface finishing process, either by selection of a
more energy efficient process or by adopting energy efficient operating practices, will decrease the
quantity of pollutants released into the environment resulting from the generation of the energy
consumed.                                                                         ,
                                          5-22

-------
                                                                       REFERENCES

                                  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. EPA (Environmental Protection Agency).  1094. 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 CTSA, specifically
focused on the surface finishing process to identify important process parameters and operating
practices for the various surface finishing technologies. For a breakdown of respondents by
alternative, refer to Section 1.3 of the Introduction. Facility characteristics of respondents are
presented in Section 3.2, Exposure Assessment. The PWB Workplace Practices Questionnaire is
presented in Appendix A

       The Pollution Prevention and Control Technology Survey (hereafter referred to as the
Pollution Prevention Survey) was an update to a previous survey and was designed to collect
information about past and present pollution prevention procedures and control technologies for
the entire PWB manufacturing process.  This Survey was performed by the DfE PWB Project and
is documented in the EPA publication, Printed Wiring Board Pollution Prevention and Control
Technology: Analysis of Updated Survey Results (U.S. EPA,  1998). The Survey results
presented periodically throughout this chapter are compiled from responses to the Pollution
Prevention Survey unless otherwise indicated. Results from the Pollution Prevention Survey
pertaining to recycle or control technologies are presented in Section 6.2 of this chapter.

       Opportunities for pollution prevention in PWB manufacturing were identified in each of
the following areas:                                                                   :
                                          6-2

-------
                                                             6.1 POLLUTION PREVENTION
•      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 sayings
can result directly from pollution prevention techniques that minimise 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 bdth 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.
\ JBenefits^ , ^
Communicates to employees and states publicly the
company commitment to achieving pollution
prevention and waste reduction goals.
Communicates to employees how to accomplish the
goals identified in the company's policy statement.
Identifies in writing specific implementation steps
for pollution prevention.
Educates employees on pollution prevention practices.
Provides incentives to employees to improve pollution
prevention performance.
Informs employees and facilitates input on pollution
prevention from all levels of the company.
Identifies true costs of waste generation and the
benefits of pollution prevention.
       A pollution prevention plan is needed to detail how the pollution prevention and waste
reduction goals described, in the company's policy statement will be achieved. The pollution
prevention plan builds on the company's policy sta.tem.ent by:

•      creating a list of waste streams and their point sources;
•      identifying opportunities for pollution prevention;
•      evaluating and prioritizing waste reduction options;
•      developing an implementation strategy for options that are feasible;
•      creating a timetable for pollution prevention implementation; and
•      detailing a plan for measuring and evaluating pollution prevention and waste reduction '
       progress.

       The plan is best developed with input drawn from the experiences of a team of people
selected from levels throughout the company.  The team approach provides a variety of
perspectives to pollution prevention and helps to identify pollution prevention opportunities and
methods for implementing them.  Team members should include representatives from
management, supervisory personnel, and line workers who are familiar with the details of the daily
operation of the process.  The direct participation of line workers in the development of the
pollution prevention plan is important since it is the employees who are responsible for
implementing the plan.

       Data should be collected by performing an assessment of the process(es) being targeted.
It is not possible to develop a pollution prevention plan unless there exists good data on the rate
at which primary and ancillary materials are used and  wastes are generated.  Once the assessment
                                           6-4

-------
                                                            6.1 POLLUTION PREVENTION
 and data collection are complete, pollution prevention options should be evaluated and prioritized
 based on their cost, feasibility of implementation, and their overall effectiveness of eliminating or
 reducing waste.  After an implementation strategy and timetable is established, the plan, along
 with expected benefits, should be presented to the remaining company employees to communicate
 the company's commitment to pollution prevention.

       Once the pollution prevention plan has been finalized and implementation is ready to
 begin, employees must be given the skills to implement the plan. Training programs play an
 important role in educating process employees about current pollution prevention practices and
 opportunities.  The goal of the training program is to educate each employee on how waste is
 generated, its effects on worker safety and the environment, possible methods for waste
 reduction, and on the overall benefits of pollution prevention.

       Employee training should begin at the time of new employee orientation, introducing them
 to the company's pollution prevention plan, thus highlighting the company's dedication to
 reducing waste. More advanced training focusing on process operating procedures, potential
 sources of release, and pollution prevention practices already in place should be provided after a
 few weeks of work or when an employee starts a new position. Retraining employees periodically
 will keep them focused on the company's goal of pollution prevention.

       Effective communication between management and employees is an important  part of a
 successful pollution prevention program. Reports to employees on the progress of implementing
 pollution prevention recommendations, as well as the results of actions already taken, reiterate
 management's commitment to reducing waste, while keeping employees informed and intimately
 involved in the process. Employee input should also be solicited both during and after the
 creation of the pollution prevention plan to determine if any changes in the plan are warranted.

       Assigning responsibility for each source of waste is an important step in closing the
 pollution prevention loop. Making individual employees and management accountable for
 chemical usage and waste generated within their process or department provides incentive for
 employees to reduce waste.  The quantity of waste generated should be tracked and the results
 reported to employees who are accountable for the process generating the waste.  Progress in
 pollution prevention should be an objective upon which employees will be evaluated during
 performance reviews, once again emphasizing the company's commitment to waste reduction.

      Employee initiative and good performance in pollution prevention areas should be
 recognized and rewarded. Employee suggestions that prove feasible and cost effective should be
 implemented and the employee recognized either with a company commendation or with some
 kind of material award. These actions will ensure continued employee participation in the
 company's pollution prevention efforts.

      Implementing an activity-based or total cost accounting system will identify the costs of
waste generation that are typically hidden in overhead costs by standard accounting systems.
These cost accounting methods identify cost drivers (activities) within the manufacturing process
and assign the costs incurred through the operation of the process to the cost drivers. By
                                          6-5

-------
6.1 POLLUTION PREVENTION
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:
.
                                                            •
       An alternative to the ISO 14001 model for EMS is the DfE EMS. It is based on the
structure outlined in the ISO 14001 standard and incorporates the five phases of Commitment,
Policy, Planning, Implementation, Evaluation and Review. While generally consistent with the'
ISO 14001 standard, the DfE EMS places less emphasis on management infrastructure and
documentation and more emphasis on pollution prevention and risk reduction. The DfE EMS is
designed for small- and medium-sized businesses and provides technical guidance and detailed
methods for developing an EMS.  The DfE EMS allows a company to create a simple yet.
effective EMS aimed at improving environmental performance by focusing on substitutes
assessments, chemical risk reduction, pollution prevention opportunities, and resource and cost
savings. DfE has developed an EMS guidance manual and several assessment tools that are
available on the DfE EMS website: .

6.1.2  Materials Management and Inventory Control

       Materials management and inventory control focuses on how chemicals  and materials flow
through a facility in order to identify opportunities for pollution prevention.  A proper materials
management and inventory control program is a simple, cost effective approach to  preventing
pollution. Table 6-2 presents materials management and inventory control methods that can be
used to prevent pollution.

  Table 6-2.  Materials Management aud Inventory Control Pollution Prevention Practices
-' : .. •.- '- ."^'practice _ •.
Minimize the amount of chemicals kept on the floor
at one time.
Manage inventory on a first-in, first-out basis.
Return unused chemicals to inventory.
Centralize responsibility for storing and distributing
chemicals.
Store chemical products in closed, clearly marked
containers.
Use a pump to transfer chemical products from stock
to transportation container.
Benefits ' <
Provides incentives to employees to use less
chemicals.
Reduces materials and disposal costs of expired
chemicals.
Reduces chemical and disposal costs.
Provides incentives to employees to use less
chemicals.
Reduces materials loss; increases worker safety by
reducing worker exposure.
Reduces potential for accidental spills; reduces
worker exposure.
                                           6-6

-------
                                                              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/first-out basis reduces the time chemicals spend in storage
and the .amount of expired chemicals that are disposed. Some companies have contracted with a
specific chemical supplier to provide all of their process chemicals and manage their inventory. In
exchange for the exclusive contract, the chemical supplier assumes many of the inventory
management duties including managing the inventory, material safety data sheets (MSDSs),
ordering the chemicals, distributing the chemicals throughout the plant, and disposing of spent
chemicals and packaging (Brooman, 1996).

       Chemical storage and handling practices also provide pollution prevention opportunities.
Ensuring that all chemical containers are kept closed when not in use minimizes the amount of
chemical lost through evaporation or volatilization. When transferring chemicals from container
to container, utilizing a hand pump can reduce the amount of chemical spillage. These simple
techniques not only result in less chemical usage representing a cost savings, but also result in
reduced worker exposure and an improved worker environment.

6.1.3  Material Selection

       Often times, decreasing the amount of pollution a particular process generates can be as
simple as selecting different materials for use in the process. This could include primary materials
such as bath chemicals or ancillary materials, such as racks or rack coverings, and is dependent
upon the availability of alternatives to the currently chosen material.

       For example, the selection of the proper flux can greatly reduce the air emissions from the
hot air solder leveling (HASL) process.  In the HASL process, the boards are immersed in a bath
of flux followed by submersion in a bath of solder mixed with oil. A hot air knife is then utilized
to remove excess solder and oil from the board. An air emission is created during these steps that
is the result of the bath chemicals being heated to fairly high temperatures (e.g., 450 ฐF for the oil
and solder mixture) and both the oil and flux having vapor pressures that when heated encourage
a portion to evaporate and condense as fine droplets (Lee,  1999).

       Most flux manufacturers fabricate multiple types of flux for use in the many different
environments that exist in PWB manufacturing, some producing as many as 30 to 40 different
fluxes. Each flux is manufactured to work most effectively in a particular environment (e.g., low
viscosity, high acidity).  Carefully choosing the right flux for a particular PWB application can
reduce flux losses, the subsequent emissions generated, and the associated  costs.

       Another example would include choosing the most appropriate type of racking system
surface material.  With several different types of racking system materials available (e.g.,
aluminum, iron, stainless steel, plastic, rubber-coated), the unnecessary build-up of bath chemicals
on the racks can be reduced.  For instance, the use of plastic racks can prevent the deposition of
                                           6-7

-------
6.1 POLLUTION PREVENTION
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.

6.1.4  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.

ffxtend Chemical Bath Life

       The surface finishing process involves the extensive use of chemicals, many of which are
costly and pose a hazard to human health and the environment.  Improvements in the efficient
usage of these chemicals can occur by accomplishing the following:

•      reducing chemical bath contamination;
•      reducing chemical bath drag-out; and
•      improving bath maintenance.

       Inefficiencies in the use of chemicals can result in increased chemical usage, higher
operating costs, increased releases to the environment, and increased worker exposure.
Techniques to improve the efficient use of chemicals by the surface finishing and other PWB
process steps are discussed in detail below.

       Reduce Bath Contaminants. The introduction of contaminants to a chemical bath will
affect its performance and significantly shorten the life of the chemical bath. Bath contaminants
include chemicals dragged in from previous chemical baths, chemical reaction by-products, and
particulate matter which may be introduced to the bath from the air. Process baths are replaced
when impurities reach a level where they degrade product quality to an unacceptable level. Any
measure that prevents the introduction of impurities will not only result in better bath
performance, but also will reduce chemical usage and generate less waste. Table 6-3 presents
pollution prevention methods for reducing bath contamination.
                                          6-8

-------
                                                              6.1 POLLUTION PREVENTION
          Table 6-3. Pollution Prevention Practices to Reduce Bath Contaminants
~. ,A Practices^ * " <
Improve the efficiency of the water rinse system.
Use distilled or deionized water during chemical
bath make-up.
Maintain and rebuild panel racks.
Clean process tanks efficiently before new bath
make-up.
Utilize chemical bath covers when process baths are
not in operation.
Remove immediately foreign objects that have fallen
into chemical tank.
Filter contaminants continuously from process
baths.
Benefits ,~ * %
Rinses off any residual bath chemistries and
dislodges any particulate matter from panels and
racks.
Reduces chemical contamination resulting from
water impurities.
Prevents the build-up of deposits and corrosion that
can dislodge or dissolve into chemical baths.
Prevents contamination of the new bath from
residual spent bath chemistries.
Reduces the introduction of unwanted airborne
particulate matter; prevents evaporation or
volatilization of bath chemistries.
Prevents the contamination and premature
degradation of bath chemicals.
Prevents the build-up of any contaminants.
       Thorough and efficient water rinsing of process panels and the racks that carry them is
 crucial to preventing harmful chemical drag-in and to prolonging the life span of the chemical
 baths. The results of the PWB Workplace Practices Questionnaire indicate that nearly every
 chemical bath in the surface finishing process is preceded by at least one water rinse tank.
 Improved rinsing can be achieved by using spray rinses, panel and/or water agitation, warm water,
 or by several other methods that do not require the use of a greater volume of water.  A more
 detailed discussion of these methods is presented in the reduced water consumption portion in this
 section.

       A rack maintenance program is also an important part of reducing chemical bath
 contamination and is practiced by 87 percent of the respondents to the Pollution Prevention
 Survey. By cleaning panel racks regularly and replacing corroded metal parts, preferably with
 parts of plastic or stainless steel, chemical deposition and build-up can be minimised,
 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 coaling the panels. The drag-out is then either removed from the panels by a hot air knifing
process, which uses air to remove excess chemical solution retained on the boards, or is simply
carried into the next bath. In most cases, the panels are deposited directly into the next process
bath without first being air knifed.

       As an extension of the making holes conductive and surface finishing DfE projects, a
mathematical tool was developed to help predict the volume of bath chemistry lost through panel
drag-Out. The model identifies multiple process parameters (e.g., number of through holes, size
of panel, length of drip time, etc.) and bath characteristics (e.g., bath temperature, viscosity, etc.)
that directly affect the volume of drag-out. Process data for the model were obtained from the
PWB Workplace Practices Questionnaire and from data provided by individual chemical suppliers.
Because the primary daily loss of bath chemistry is through drag-out, using the model to minimize
drag-out will result in extended bath life, decreases in rinse water and bath chemistry usage, and a
reduction in treatment sludge.  The drag-out model along with a complete description of the
method of development, individual factors in the model, and the model limitations is presented in
Appendix E. Drag-out model results for the surface finishing alternatives are presented in Section
3.2, Exposure Assessment.

       Techniques that minimize bath drag-out also prevent the premature reduction of bath
chemical concentration, extending the useful life of a bath. In addition to extended bath life,
minimizing or recovering drag-out losses also has the following effects:  •  •

•      minimizes bath chemical usage;
•      reduces the quantity of rinse water used;
•      reduces chemical waste;
•      requires less water treatment chemical usage; and
•      reduces overall process cost.

       Methods for reducing or recovering chemical bath drag-out are presented in Table 6-4 and
discussed below.

       The two most common methods of drag-out control employed by respondents to the
Pollution Prevention Survey that require no capital investment are increased panel drainage time
(76.3  percent) and practicing slow rack withdrawal from process tanks (60.5 percent). Increasing
the time allowed for the panels to drain over the process bath allows a greater percentage of
potentially removable chemicals to remain in the bath. Practicing slow rack withdrawal during
rack removal is another step used relatively often to allow more time for the bath chemicals" to
drip back into the bath. Neither of these techniques requires capital investment and both are
effective methods for reducing drag-out.
                                          6-10

-------
                                                               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 static drag-out tanks/drip tanks to
process line where needed.
Utilize non-ionic wetting agents in the
process bath chemistries.
Utilize air knives directly after process bath
in conveyorized system. a
Employ fog rinses/spray rinses over heated
baths.3 •
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 arid 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.
a 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
:-.: -'../:"- •>;.;Meatods ~ '•> ] - - \- ^Benefits / - \
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.
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 particulate 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 diflusion-
  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.        v

        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 drag-
                                           6-14

-------
                                                             6.1 POLLUTION PREVENTION
in quantities and will fail to provide a clean panel surface for subsequent chemical activity.
Excessive water rinsing, done by exposing the panels top 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, utilising more efficient rinse configurations such as countercurrent
rinse stages, spray rinses, or fog rinses will increase the overall efficiency of the surface finishing
                                          6-15

-------
 6.1 POLLUTION PREVENTION
rinse system while reducing the volume of wastewater generated. 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.

       Good housekeeping practices focus on keeping the process equipment in good repair and
fixing or replacing leaky pipes, pumps, and hoses. These practices can also include installing
devices.such as spring loaded hose nozzles that shut off when not in use, or water control timers
that shut off water flow in case of employee error. These practices often require little investment
and are effective in preventing unnecessary water usage. For a more detailed discussion on
methods of improving water rinse efficiency and reducing water consumption, refer to Section
5.1, Resource Conservation.

Improve Process Efficiency Through Automation

       The operation of the surface finishing process presents several opportunities for important
and integral portions of the process to become automated. By automating important functions,
operator inconsistencies can be eliminated, allowing the process to be operated more efficiently.
Automation can lead to the prevention of pollution by:

•      gaining a greater control of process operating parameters;
•      performing the automated function more consistently and efficiently;
•      eliminating operator errors; and
•      making the process compatible with newer and cleaner processes designed to be operated
       with an automated system.

       Automating a part of the surface finishing process  can be expensive. The purchase of
some automated equipment can require a significant initial investment, which may prevent small
companies from automating.  Other costs that may be incurred include those associated with
installing the equipment, training employees, any lost production due to process down-time, and
redesigning other processes to be compatible with the new system. Although it may be  -
expensive, the benefits of automation on productivity and waste reduction will result in a more
efficient process that can save money over the long run.

      Installation of automated equipment such as a rack or panel transportation system,
chemical sampling equipment, or an automated system to make chemical additions can have a
major impact on the quantity of pollution generated during the day-to-day operation of the
surface finishing process and can also reduce worker exposure.  Surface finishing process steps or
functions that can be automated effectively include:
                                         6-16

-------
                                                              6.1 POLLUTION PREVENTION
•      rack transportation;
•      bath maintenance; and
•      water flow control.

       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 tunes 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 $3.$. EPA, 1998), 81 percent of whom
 reported using some type of recycle or resource recovery technology.

        Recycle and resource recovery technologies include those that recover materials from
 waste streams before disposal, or recycle waste streams for reuse in another process.
 Opportunities for both types of technologies exist within a surface finishing process. Rinse water
 can be recycled and reused in further rinsing operations, while valuable metals such as copper,
 silver, palladium, and gold can be recovered from waste streams before disposal and sold to a
 metals reclaimer. These recycle and recovery technologies may be either in-line (dedicated and
 built into the process flow of a specific process line) or at-line (employed at the line as desired, as
 well as at other places in the plant), depending on what is required (Brooman, 1996). Each waste
 stream that cannot be prevented should be evaluated to determine its potential for effective
 recycle  or resource recovery as part of a pollution prevention and waste management plan.

       The decision of whether to purchase a recycle or resource recovery process should be
based on several factors. Economic factors, such as process operating and effluent disposal costs
for the current system, must be compared with those estimated for the new technology.  The
initial capital investment of the new technology, along with any potential cost savings, and the
length of the payback period must also be considered.  Other factors such as the characteristics of
the waste  stream(s) considered for treatment, the ability of the process to accept reused or
                                          6-19

-------
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 ASSESSMEINT

 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
   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/Ib ($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. Mckel recovery using electrowinning requires close control
of pH; therefore, it is performed less frequently than for other metals, such as copper and gold
(U.S. EPA, 1998).

       The electrolytic cell is comprised of a set of electrodes (cathodes and anodes) placed in
the metal-laden solution.  An electric current, or voltage, is applied across the electrodes and
through the solution. The positively charged metal ions are drawn to the negatively charged
cathode where they deposit onto the surface. The solution is kept thoroughly mixed using air
agitation or other proprietary techniques, to permit the use of higher current densities (the  amount
of current per surface area of cathode).  These higher current densities shorten deposition time
and improve the  recovery efficiency. As the metal recovery continues, the concentration of metal
ions in solution becomes depleted, requiring the current density to be reduced to maintain
efficiency at an acceptable level. When the concentration of metal becomes too low for its
removal to be economically feasible, the process is discontinued and the remaining solution is sent
to final treatment.

       Electrowinning is most efficient with concentrated solutions.  Dilute solutions (less than
100 mg/1 of metal) become uneconomical to treat due to the high power consumption relative to
the amount of metal recovered (Coombs, 1993). Waste streams that are to be treated by
electrowinning should be segregated, only combining streams containing the same metal, to
prevent dilution,  and to create a pure metal deposit free of other metal impurities. Already diluted
solutions can be concentrated first, using ion exchange or evaporation techniques, to improve the
efficiency and cost effectiveness of metal recovery.

       Process waste solutions containing chlorine ions in any form should not be processed
using electrolytic recovery methods, because the electrolysis of these solutions could generate
chlorine gas. Solutions containing copper chloride salts should first be converted to non-chloride
                                          6-22

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

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

       The recovered metal(s) can be sold as scrap to a metals reclaimer.  Typical metal removal
efficiencies of 90 to 95 percent have been achieved using electrolytic methods (U.S. EPA, 1990).
The remaining effluent will still contain small amounts of metal and will be acidic in nature (i.e.,
low pH). Adjusting the pH may not be sufficient for the effluent to meet the standards of some
POTW authorities; therefore, further treatment may be required.

       Eighteen percent of the Pollution Prevention Survey respondents reported using
electrowinning as a resource recovery technology, with 89 percent of those being satisfied with its
performance. The median cost of an electrowinning unit reported by the respondents was
$15,000; however, electrowinning capital costs are dependent on the capacity of the unit.

Ion Exchange

       Ion exchange is a process used by the PWB industry mainly to recover metal ions, such as
copper, tin, or palladium, from rinse waters and other solutions. This process uses an exchange
resin to remove the metal from solution and concentrate it on the surface of the resin. It is
particularly suited to treating dilute solutions, since at lower concentrations the resin can process
a greater volume of wastewater before becoming saturated. As a result, the relative economics of
the process improve as the concentration of the feed solution decreases. Aside from recovering
metals such as copper and silver, ion exchange also can be used for treating wastewater,
deionizing feed water, and recovering chemical solutions.

       Ion exchange relies on special resins, either cationic or anionic, to remove the desired
chemical species from solution. Cation exchange resins are used to remove positively charged
ions such as copper, tin, or other metals. When a feed stream containing a metal is passed
through a bed of cation exchange resin, the resin removes the metal ions from the stream,
replacing them with hydrogen ions from the resin. For example, if a feed stream containing
copper sulfate (CuSO4) is passed through the ion exchange resin, the copper ions are removed
and replaced by hydrogen ions to form sulfuric acid (H2SO4). The remaining water effluent is
either further processed using an anion exchange resin and then recirculated into the rinse  water
system, or pH neutralized and then directly sewered.  Ion exchange continues until the exchange
resin becomes saturated with metal ions and must be regenerated.

       Special chelating resins have been designed to capture specific metal ions that are in the
presence of chelating agents, such as metal ions in electroless plating baths. These resins are
effective in breaking down the chemical complexes formed by chelators that keep metal ions
dissolved in solution, allowing them to be captured by the resin. Hard water ions, such as calcium
and magnesium, are not captured, creating a purer concentrate.  Chelating resins require that the
feed stream be pH-adjusted to reduce acidity, and filtered to remove suspended solids that will
foul the exchange bed (Coombs, 1993),
                                          6-23

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

       Regeneration of the cation or cbelating 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
>,

Ion
Exchange
>
r
>.

Electrolytic
Recovery
1
	 ^Reclaimed
Metal
Spent
Regenerant
r Waste
               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 RECYOLE, 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
(senttoaPOTW).

       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
>
ce-
ms
nditiorring
Tank


>.

RO
Unit


up ^
•
>.

Storage
Tank

^
                                                            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
ofrecovery.

        Table 6-6. Typical Value of Reclaimed Metals (1999) and Recovery Methods
Metal
Gold
Palladium
Silver
Copper
Solder
Price3*
$283/oz
$636/oz
$4.98/oz
$0.80/lb
$1.60/lb
Recovery Method c
Off-site refining or electrolytic
Off-site refining
Off-site refining or ion exchange
On-site electrolytic or ion exchange
Manual or solder recovery system
* Metal prices received will be current market prices minus a 2 to 5 percent refining fee. Prices listed are spot prices on
7/6/00 obtained from wwwJdtco.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 ofrecovery 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 eledrowinning to recover metal) into
a more cost effective recovery system that achieves greater removal efficiency.
                                          6-27

-------
63. RECYCLE, RECOVERY, AND CONTROL TECHNOLOGIES ASSESSMENT
Table 6-7. Applicability of Recovery/Reclamation Technologies bj
Bath Type
Drag-out Rinse
(following gold,
palladium)
Gold
Microetch
Nickel
Palladium
Immersion Silver
Solder
Immersion Tin
Water Rinse
, Processes)
Nickel/Gold and
Nickel/Palladium/Gold
Nickel/Palladium/Gold
• All
Nickel/Gold and
Nickel/Palladium/Gold
Mckel/Palladium/Gold
Immersion Silver
HASL
Immersion Tin
All
Solder
Recovery






•


-Ion
Exchange
•
•
•
•

•

•

Electrolytic
Recovery
•
•
•
•

•

•

r Bath Type
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
                                          6-28

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

facility and the effluent permit limits. Together these processes form a complete treatment system
capable of treating the waste streams generated by the PWB manufacturing process, including
those from the surface finishing line.    .

       A diagram of a typical PWB facility treatment system is presented in Figure 6-4, while the
individual treatment processes are discussed below.  References to key points of the diagram are
included in the descriptions, and are denoted with reference number in brackets.

       Waste Collection and Segregation System. Waste streams are collected from processes
located throughout the facility by a sophisticated piping and collection system that conducts the
individual waste streams to the waste treatment process. The collection system must be versatile,
allowing the waste treatment operators complete control over the destination of an incoming
wastewater flow.  In the case of a chemical spill or harmful accidental discharge, operators must
have the ability to divert the wastewater flow into  a holding tank to prevent any violations that
might be caused by overloading the treatment system.

       The treatment process typically has a waste collection tank and one or more holding tanks.
The collection system deposits the individual waste streams into one or more collection tanks at
the operator's discretion. Waste streams are typically co-mingled in the main collection tank (1)
for a period of time prior to entering the waste treatment system, to allow complete mixing and to
smooth out any concentration spikes that might occur during normal process operation.

       Difficult-to-treat streams, such as those containing chelators or requiring special
treatment, are segregated from the others at the source and fed into  separate holding tanks.
Metal-bearing rinses should be segregated  from streams which do not contain metals.  Specific
segregation of cyanide, solvents, flux, and  reflow oils is critically important (Iraclidis, 1998).
Waste streams containing oxidizing  agents also typically are segregated from others because of
the difficulty oxidizing agents present during the flocculation and settling stages (oxidizing agents
evolve gas that can hinder floe settling) (Sharp, 1999).

       Flow-Through Chemical Precipitation System. In the PWB industry, the majority of
facilities surveyed (61 percent) reported using a conventional chemical precipitation system to
accomplish the removal of heavy metal ions from wastewater.  Chemical precipitation is a process
for treating wastewater that depends on the water  solubility of the various compounds formed
during treatment.  Heavy metal cations present hi the wastewater are reacted with certain
treatment chemicals to form hydroxides, sulfides, or carbonates that have relatively low water
solubilities. The resulting heavy metal compounds are precipitated from the solution as an
insoluble sludge that is subsequently sent off-site to reclaim the metals content,  or sent to
disposal. Chemical precipitation can be carried out in a batch process, but is typically operated in
a continuous flow-through process to treat wastewater.

       In the chemical precipitation treatment of wastewater from PWB manufacturing, the
removal of heavy metals may be carried out by a unit sequence of rapid mix precipitation,
flocculation, and clarification. The process begins by adjusting the pH of the incoming
wastewater (2) to optimum operating conditions (pH 6 to 8),  The optimum pH for treatment is
                                          6-29

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

dependant on both the treatment chemistry and the metals being removed from the wastewater.
Adjustments are made through the addition of acid or lime/caustic. Treatment chemicals are then
dispersed into the wastewater input stream under rapid mixing conditions.  The initial mixing unit
(3) is designed to create a high intensity of turbulence in the reactor vessel, promoting multiple
encounters between the metal ions and the treatment chemical species, which then react to form
insoluble metal compounds. The type of chemical compounds formed depends on the treatment
chemical employed; this is discussed in detail later in this section.  These insoluble compounds
form a fine precipitate at low pB[ 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 hi the case of unusually strict effluent limits.
Filtration, reverse osmosis, ion exchange, of additional precipitation steps are sometimes
employed to further reduce the concentration of chemical contaminants present hi the wastewater
effluent.
                                          6-30

-------
             	6.2 RECYCUE, 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) rninimize 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

                                                                        s
                                                                        2
                                                                        H
                                                                        I
                                                                        "3
                                                                        u
                                                                        vo
                                                                        s

                                                                        .SP
                                                                        fa
                                   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
hi solution beyond their normal solubility limits. These chemicals are found in spent surface
finishing plating baths, hi 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
: sGhemical
Ferrous Sulfate
DTC (Dimethyl-dithiocarbamate)
Sodium Sulfide
Polyelectrolyte
Sodium Borohydride
Ferrous Dithionite
TMT 15 (Tri-mercaptotriazine)
> ^ Description t p _ * J
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, 1 992).
Polymers that remove metals effectively without contributing to the
volume of sludge. Primary drawback is the high chemical cost (Frailey,
1996).
Strong reducing agent reduces metal ions, then precipitate out of solution
forming a dense, low volume sludge. Drawbacks include its high
chemical cost and the evolution of potentially explosive hydrogen gas
(Guess, 1992; Frailey, 1996).
Reduces metal ions under acidic conditions to form metallic particles that
are recovered by gravity separation. Excess iron is regenerated instead of
being precipitated, producing a low volume sludge (Guess, 1992).
Designed specifically to precipitate silver ions, which are unaffected by
other treatment chemicals, from wastewater. Primary drawbacks are the
high chemical to silver removed weight ratio and the high chemical cost
(Sharp, 1999).
Individual Alternative Treatment Profiles

       There are often many approaches from which a facility can choose to properly treat and
dispose of a process waste stream. Several of the approaches, which have been discussed in this
CTSA chapter, include reclamation, recycling, treatment, disposal, or a combination of these.
The treatment or recycling method used by a faculty for each waste stream is dependent on a
number of factors including discharge permit effluent limits (is more vigorous treatment required
to meet effluent limits?), economics  (is the treatment cost effective?), the capability of on-site
treatment system (e.g., the presence of reclamation technologies), the treatment requirements of
processes other than the surface finishing line (e.g., can the waste stream be combined with other
waste streams to make other treatment options more applicable), and a faculty's preference, based
on experience. One, or a combination of several of these factors, will dictate the treatment
options available to a particular facility.
                                          6-34

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

       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

-------
6.2 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
, Processes)
Nickel/Gold and
Nickel /Palladium/Gold
Nickel/Gold and
Nickel /Palladium/Gold
All
Nickel/Gold and
Nickel /Palladium/Gold
Nickel/Gold and
Nickel /Palladium/Gold
Nickel/Gold and
Nickel /Palladium/Gold
Nickel/Gold and
Nickel /Palladium/Gold
HASL
Immersion Silver
Immersion Tin
All
OSP
Immersion Tin and
Immersion Silver
HASL
All
Chelated
N
N
Y
Y
Y
Y
Y
N
Y
Y
N-
N
N
N
N
' Typical Treatment Method*
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 bam is treated. Indicated method is not the only way a
bath may be treated by an individual facility. Typical methods were developed and reviewed by PWB manufacturer
project participants.
Air Pollution Control Technologies

       Air pollution control technologies are often used by the PWB industry to cleanse air
exhaust streams of harmful fumes and vapors. Exactly half (50 percent) of the PWB facilities
surveyed have installed air scrubbers to control air emissions from various manufacturing
processes, and almost a quarter of the facilities (23 percent) scrub air releases from surface
finishing processes. The first step of any air control process is the effective containment of
fugitive air emissions at their source of release. This is accomplished using fume hoods over the
                                           6-36

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

process areas from which the air release of concern occurs. These hoods may be designed to
continuously collect air emissions for treatment by one of the methods described below.

       Gas Absorption, One method for removing pollutants from an exhaust stream is by gas
absorption in a technique sometimes referred to as air scrubbing. Gas absorption is defined as the
transfer of material from a gas to a contacting liquid or solvent.  The pollutant is chemically
absorbed and dispersed into the solvent, leaving the air free of the pollutant. The selection of an
appropriate solvent should be based on the liquid's solubility for the solute, and the cost of the
liquid. Water is used for the absorption of water-soluble gases,  while alkaline solutions are
typically used for. the absorption of acid gases. Air scrubbers are used by the PWB industry to
treat wet process air emissions, such as formaldehyde and acid fumes, and emissions from other
processes other than the surface finishing process.

       Gas  absorption is typically carried out in a packed gas absorption tower, or scrubber. The
gas stream enters the bottom of the tower and passes upward through a wetted bed of packing
material before exiting the top. The absorbing liquid enters the top of the tower and flows
downward through the packing before exiting at the bottom. Absorption of the air pollutants
occurs during the period of contact between the gas and liquid.  The gas is either physically or
chemically absorbed and dispersed into the liquid. The liquid waste stream then is sent to water
treatment before being discharged to the sewer. Although the most common method for gas
absorption is the packed tower, other methods exist such as plate towers, sparged towers, spray
chambers, or venturi scrubbers (Cooper and Alley, 1990).

       Gas Adsorption. The removal of low concentration organic gases and vapors from an
exhaust stream can be achieved by the process of gas adsorption.  Adsorption is the process in
which gas molecules are retained on the interface  surfaces of a solid adsorbent by either physical
or chemical  forces. Activated carbon is the most common adsorbent, but zeolites, such as
alumina and silica, are also used.  Adsorption is used primarily to remove volatile, organic
compounds from air, but is also used in other applications such as odor control and drying
process gas  streams (Cooper and Alley, 1990). In a surface finishing process, gas adsorption can
be used to recover volatile organic compounds, such as formaldehyde.

       Gas adsorption occurs when the vapor-laden air is collected and then passed through a
bed of activated carbon or another adsorbent material.  The gas  molecules are adsorbed onto the
surface of the material, while the clean, vapor-free air is exhausted from the system. The
adsorbent material eventually becomes saturated with organic material and must be replaced or
regenerated. Adsorbent  canisters, which are replaced on a regular basis, are typically used to treat
small gas flow streams. Larger flows of organic pollutants require packed beds Of adsorbent
material, which must be regenerated when the adsorbent becomes saturated (Cooper and Alley
1990).

       Regeneration of the adsorbent is typically accomplished by a steam-stripping process. The
adsorbent is contacted with low-pressure steam which desorbs the adsorbed gas molecules from
the surface of the packed bed. Following condensation of the steam, the organic material is
recovered from the water by either decanting or distillation (Campbell and Glenn, 1990).
                                          6-37

-------
 REFERENCES
                                    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 JJPC
 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.

Jxaclidis, Taso.  1998.  "Wastewater Treatment Technologies of Choice for the Printed Circuit
Board Industry." In: Proceedings of the JJPC 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

-------
                                                                        REFERENCES

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. EPA (Environmental Protection Agency).  1990. Guides to Pollution Prevention: The
Printed Circuit Board Manufacturing Industry. EPA Office of Resource and Development,
Cincinnati, OH. EPA/625/7-90/007.  June.

U.S. EPA (Environmental Protection Agency),  1995a.  "Printed Wiring Board Case Study 1:
Pollution Prevention Work Practices." Pollution Prevention Information Clearinghouse (PPIC).
Washington, D.C. EPA 744-F-95-004. July.

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

U. S. EPA (Environmental Protection Agency).  1996a.  "Printed Wiring Board Project:
Opportunities for Acid Recovery and Management."  Pollution Prevention Information
Clearinghouse (PPIC).  Washington,  D.C. EPA 744-F-95-009. September.

U.S. EPA (Environmental Protection Agency).  1996b.  "Printed Wiring Board Project: Plasma
Desmear;  A Case Study." Pollution Prevention Information Clearinghouse (PPIC). Washington,
D.C. EPA744-F-96-003.  September.

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

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

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

-------
REFERENCES	_^__

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

U.S. EPA (Environmental Protection Agency). 1998. Printed Wiring Board Pollution
Prevention and Control Technology: Analysis of Updated Survey Results.  Design for the
Environment Printed Wiring Board Project. EPA Office of Pollution Prevention and Toxics.
Washington, D.C.  EPA744-R-98-003. August.

U.S. EPA (Environmental Protection Agency). 1999. "Pollution Prevention beyond Regulated
Materials."  Pollution Prevention Information Clearinghouse (PPIC). Washington, D.C.
EPA744-F-97-OQ4. 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, cohveyorized
equipment. The HASL process can be applied in either equipment mode.  Table 7-1 presents the
processes (alternatives and equipment configurations) evaluated in the CTSA.

              Table 7-1. Surface Finishing Processes Evaluated in the CTSA
Surface Finishing Technology
HASL (Baseline)
Nickel/Gold
Nickel/Palladium/Gold
OSP
Immersion Silver
Immersion Tin
Equipment Configuration
Non-Conveyorized
•
•
•
•

•
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 CTSA evaluated the risk, performance, cost, and resource
requirements of the baseline surface finishing technology as well as the alternatives. This section
summarizes the findings associated with the analysis of surface finishing technologies. Relevant
data include the following:

•      Risk information: .occupational health risks, public health risks, ecological hazards, and