Solders in
Electronics:
A Life Cycle
Assessment
US. EPA
-------
EPA 744-R-05-001
August 2005
Solders in
Electronics:
A Life-Cycle
Assessment
Jack R. Geibig
Maria Leet Socolof
US. EPA
This document was produced by the University of Tennessee Center for
Clean Products and Clean Technologies under grant # X-82931801 from
EPAs Design for the Environment Branch, Economics, Exposure, &
Technology Division, Office of Pollution Prevention and Toxics.
-------
Disclaimer
This document was written by the grantee. It has not been through a formal external peer
review process and does not necessarily reflect all of the most recent policies of the U.S.
Environmental Protection Agency (EPA), in particular those now under development. 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
environmental statutes is intended for information purposes only; this is not an official guidance
document and should not be relied upon to determine applicable regulatory requirements.
For More Information
To learn more about the Design for the Environment (DfE) Lead-Free Solder Project or
the DfE Program, please visit the DfE Program web site at:
www.epa.gov/dfe
To obtain copies of DfE Program technical reports, pollution prevention case studies, and
project summaries, please contact:
National Service Center for Environmental Publications
U.S. Environmental Protection Agency
P.O. Box 42419
Cincinnati, OH 45242
Phone: (513)489-8190
(800)490-9198
Fax:(513)489-8695
E-mail: ncepimal@one.net
To learn more about the University of Tennessee Center for Clean Products and Clean
Technologies, visit the Center's web site at:
www. cl eanproducts. org
-------
Ackno wled gm ents
This life-cycle assessment (LCA) was prepared by the University of Tennessee (UT)
Center for Clean Products and Clean Technologies under funding through a grant from the U.S.
Environmental Protection Agency's Design for the Environment (DfE) Program in the
Economics, Exposure, and Technology Division (EETD) of the Office of Pollution Prevention
and Toxics (OPPT), and through financial contributions from the following organizations:
• Agilent Technologies
• Cookson-Fry
• Delphi
Hewlett Packard
Intel
• International Business Machines (IBM)
Sematech
• Pitney Bowes
• Rockwell Collins
• Thomson Consumer Media
The authors would like to acknowledge the outstanding contributions of the UT staff, faculty,
and students who assisted the authors, including: UT undergraduate research assistants, Sarah
Surak and Brooke Weeks, who helped perform the technical work; Ulrika Kindesjo, a
University of Lund, Sweden graduate student who researched disposal and recycling processes;
and Claire VanRiper-Geibig, who assisted with document production.
This document was produced as part of the DfE Lead-Free Solder Partnership, under the
direction of the project's Core Group members, including: Kathy Hart, Project Lead and Core
Group Co-Chair, U.S. EPA OPPT, DfE Branch; Holly Evans, formerly of Electronic Industries
Alliance, Core Group Co-Chair; Fern Abrams, IPC, Core Group Co-Chair; and, Todd Brady,
Intel Corp.; Maria Socolof and Jack Geibig, University of Tennessee Center for Clean Products
and Clean Technologies; Anne Brinkley, IBM; Lee Vroom, Thomson Consumer Electronics;
Mark Corbett, Pitney Bowes; Walter Worth, International Sematech; Pete Palmer, Cookson-
Fry; Jerry Gleason, Hewlett Packard; and Brenda Baney, Delphi, all Core Group members.
Many thanks also to the industry representatives and other interested parties who participated in
the project's Technical Workgroup and who helped to organize and facilitate the start of this
project. The authors thank Vince Nabholz and Terry O'Bryan of EPA's Risk Assessment
Division, OPPT, for their assistance in reviewing and providing health and environmental
toxicity information for the project.
-------
Participating Organizations
In addition to the organizations listed above who provided funding and continued support
for this project, the authors would like to thank the following participants who provided life-
cycle inventory data, materials, research, or support for the project in some other significant way.
The LCA could not have been completed without their participation.
AIM Solder
Boliden
Celestica
Electronic Industries Alliance (EIA)
Flextronics
Hobi
Intel
IPC-Association for Connecting Electronics Industries
Kestor
Metalvert
Micrometallics
Noranda
NxtCycle
Omega Solder
Phillips
Senju
Siemens
Teradyne
U.S. EPA—Incineration data
U.S. Navy—Crane
University of Florida
Vitronic-Soltec
-------
Table of Contents
EXECUTIVE SUMMARY ES-1
CHAPTER 1: GOAL DEFINITION AND SCOPE
1.1 Introduction 1-1
1.2 Project Background 1-1
1.3 Goals and Scope: Why Perform a Life-Cycle Assessment of Solders? 1-2
1.3.1 Lead-Free Solder Project purpose 1-2
1.3.2 Previous research 1-2
1.3.3 Need for the project 1-3
1.3.4 Market trends 1-4
1.3.5 Target audience and use of the study 1-4
1.4 Summary of Life-Cycle Assessment Methodology 1-5
1.5 Product Systems 1-7
1.5.1 Solder alternatives 1-7
1.5.2 Functional unit 1-8
1.6 Assessment Boundaries 1-9
1.6.1 Life-Cycle stages and unit processes 1-9
1.6.2 Spatial and temporal boundaries 1-11
1.6.3 General exclusions 1-11
References 1-12
CHAPTER 2: LIFE-CYCLE INVENTORY
2.1 General Methodology 2-5
2.1.1 Data categories 2-5
2.1.2 Decision rules 2-7
2.1.3 Data collection and data sources 2-8
2.1.4 Allocation procedures 2-9
2.1.5 Data management and analysis software 2-10
2.1.6 Data quality 2-10
2.1.7 Critical review 2-11
2.2 Materials Extraction and Materials Processing (Upstream Life-Cycle Stages) 2-12
2.2.1 Methodology 2-12
2.2.1.1 Materials (metals) 2-12
2.2.1.2 Fuels and power sources 2-16
2.2.2 Limitations and uncertainties 2-19
-------
2.3 Product Manufacturing 2-20
2.3.1 Methodology 2-20
2.3.1.1 Data collection and allocation 2-20
2.3.1.2 Solder manufacturing 2-21
2.3.1.3 Post-industrial recycling 2-23
2.3.2 Limitations and uncertainties 2-26
2.3.2.1 Product system boundary uncertainties 2-26
2.3.2.2 Data collection process uncertainties 2-26
2.3.2.3 Data uncertainties 2-26
2.4 Solder Use/Application 2-28
2.4.1 Methodology 2-28
2.4.1.1 Paste solder 2-28
2.4.1.2 Bar solder 2-32
2.4.2 Limitations and uncertainties 2-34
2.4.2.1 Paste solder 2-34
2.4.2.2 Bar solder 2-34
2.5 End-of-Life 2-35
2.5.1 Methodology 2-35
2.5.1.1 Landfilling 2-37
2.5.1.2 Incineration 2-39
2.5.1.3 Post-consumer recycling: demanufacturing and
copper smelting 2-40
2.5.1.4 Post-consumer recycling: unregulated recycling
and disposal 2-41
2.5.2 Limitations and uncertainties 2-43
2.6 Baseline Life-Cycle Inventory Results 2-44
References 2-47
CHAPTER 3: LIFE-CYCLE IMPACT ASSESSMENT
3.1 Methodology 3-1
3.1.1 Classification 3-2
3.1.2 Characterization 3-6
3.2 Characterization and Results 3-8
3.2.1 Resource use (non-renewable and renewable) 3-10
3.2.1.1 Characterization 3-10
3.2.1.2 Paste solder results 3-11
3.2.1.3 Bar solder results 3-18
3.2.1.4 Limitations and uncertainties 3-25
3.2.2 Energy use 3-26
3.2.2.1 Characterization 3-26
3.2.2.2 Paste solder results 3-26
3.2.2.3 Bar solder results 3-31
3.2.2.4 Limitations and uncertainties 3-35
11
-------
3.2.3 Landfill space use impacts 3-36
3.2.3.1 Characterization 3-36
3.2.3.2 Paste solder results 3-36
3.2.3.3 Bar solder results 3-40
3.2.3.4 Limitations and uncertainties 3-44
3.2.4 Global warming impacts 3-46
3.2.4.1 Characterization 3-46
3.2.4.2 Paste solder results 3-46
3.2.4.3 Bar solder results 3-50
3.2.4.4 Limitations and uncertainties 3-54
3.2.5 Stratospheric ozone depletion impacts 3-55
3.2.5.1 Characterization 3-55
3.2.5.2 Paste solder results 3-55
3.2.5.3 Bar solder results 3-60
3.2.5.4 Limitations and uncertainties 3-65
3.2.6 Photochemical smog impacts 3-68
3.2.6.1 Characterization 3-68
3.2.6.2 Paste solder results 3-68
3.2.6.3 Bar solder results 3-72
3.2.6.4 Limitations and uncertainties 3-76
3.2.7 Acidification impacts 3-77
3.2.7.1 Characterization 3-77
3.2.7.2 Paste solder results 3-77
3.2.7.3 Bar solder results 3-81
3.2.7.4 Limitations and uncertainties 3-84
3.2.8 Air particulate impacts 3-86
3.2.8.1 Characterization 3-86
3.2.8.2 Paste solder results 3-86
3.2.8.3 Bar solder results 3-90
3.2.8.4 Limitations and uncertainties 3-94
3.2.9 Water eutrophication impacts 3-95
3.2.9.1 Characterization 3-95
3.2.9.2 Paste solder results 3-95
3.2.9.3 Bar solder results 3-99
3.2.9.4 Limitations and uncertainties 3-102
3.2.10 Water quality impacts 3-104
3.2.10.1 Characterization 3-104
3.2.10.2 Paste solder results 3-104
3.2.10.3 Bar solder results 3-108
3.2.10.4 Limitations and uncertainties 3-112
3.2.11 Occupational human health impacts 3-114
3.2.11.1 Characterization 3-114
3.2.11.2 Paste solder results 3-118
3.2.11.3 Bar solder results 3-130
3.2.11.4 Limitations and uncertainties 3-140
in
-------
3.2.12 Public human health impacts 3-144
3.2.12.1 Characterization 3-144
3.2.12.2 Paste solder results 3-145
3.2.12.3 Bar solder results 3-155
3.2.12.4 Limitations and uncertainties 3-165
3.2.13 Aquatic ecotoxicity impacts 3-169
3.2.13.1 Characterization 3-169
3.2.13.2 Paste solder results 3-171
3.2.13.3 Bar solder results 3-177
3.2.13.4 Limitations and uncertainties 3-181
3.3 Alternate Analyses 3-183
3.3.1 Reflow application energy analysis 3-183
3.3.2 Alternate silver inventory analysis 3-186
3.3.3 Alternate leachate analysis 3-190
3.4 Summary of Life-Cycle Impact Assessment characterization and results . . 3-192
3.4.1 Impact score equations 3-192
3.4.2 Life-Cycle Impact Assessment data sources and data quality 3-195
3.4.3 Paste solder results summary 3-196
3.4.4 Bar solder results summary 3-204
3.4.5 Limitations and uncertainties 3-211
3.4.5.1 General Life-Cycle Impact Assessment methodology
limitations
and uncertainties 3-211
3.4.5.2 General limitations and uncertainties
of results 3-212
References 3-213
APPENDIX A: LIFE-CYCLE INVENTORY DATA COLLECTION FORMS
APPENDIX B: USE/APPLICATION ENERGY TESTING
APPENDIX C: SOLDER LEACHIBILITY TESTING
APPENDIX D: LIFE-CYCLE IMPACT ASSESSMENT SUPPORT DATA (NON-TOXIC)
APPENDIX E: LIFE-CYCLE IMPACT ASSESSMENT SUPPORT DATA (TOXIC)
APPENDIX F: SUMMARY OF INDUSTRY PERFORMANCE TESTING OF SOLDER
APPENDIX G: LIFE CYCLE INVENTORY FUEL DATA
APPENDIX H: EXAMPLE TOXICITY CALCULATIONS
IV
-------
Tables
Table 1-1: Solders selected for evaluation 1-8
Table 2-1: LCI data categories 2-6
Table 2-2: Data types by life-cycle stage 2-9
Table 2-3: Base metal inventories: summary of information from secondary data 2-14
Table 2-4: Data sources and data quality for metals inventories
in the ME&P life-cycle stage 2-15
Table 2-5: Fuel and power inventories: summary of information from secondary data .2-17
Table 2-6: Data sources and data quality for fuel and power inventories
used in various life-cycle stages 2-18
Table 2-7: Inventory data sets for paste and bar solder manufacturing 2-23
Table 2-8: Average virgin content of base metals used in solder manufacturing 2-25
Table 2-9: Reflow profile specifications 2-30
Table 2-10: Reflow test vehicle specifications 2-31
Table 2-11: Paste solder reflow test data 2-32
Table 2-12: Wave solder test data 2-34
Table 2-13: General distribution of EOL electronics by disposition 2-36
Table 2-14: Data collection approach for EOL dispositions 2-37
Table 2-15: TCLP-based leachate data used to predict outputs from landfilling 2-38
Table 2-16: Percent distribution of incinerator outputs 2-39
Table 2-17: Fraction distribution of copper smelting outputs 2-40
Table 2-18: Unregulated recycling and disposal assumptions 2-43
Table 3-1: Inventory types and properties for classifying inventory items
into impact categories 3-5
Table 3-2: LCIA characterization approaches for the LFSP 3-7
Table 3-3: Process groups 3-9
Table 3-4: NRRuse impacts by life-cycle stage (paste solder) 3-11
Table 3-5: RRuse impacts by life-cycle stage (paste solder) 3-13
Table 3-6: NRRuse impacts by life-cycle stage and process group (paste solder) 3-14
Table 3-7: RRuse impacts by life-cycle stage and process group (paste solder) 3-16
Table 3-8: Top contributors to NRRuse impacts (paste solder) 3-17
Table 3-9: Top contributors to RRuse impacts (paste solder) 3-18
Table 3-10: NRRuse impacts by life-cycle stage (bar solder) 3-18
Table 3-11: RRuse impacts by life-cycle stage (bar solder) 3-20
Table 3-12: NRRuse impacts by life-cycle stage and process group (bar solder) 3-21
Table 3-13: RR use impacts by life-cycle stage and process group (bar solder) 3-22
Table 3-14: Top contributors to NRRuse impacts (bar solder) 3-23
Table 3-15: Top contributors to RRuse impacts (bar solder) 3-24
Table 3-16: Energy use impacts by life-cycle stage (paste solder) 3-26
Table 3-17: Energy use impacts by life-cycle stage and process group (paste solder) .... 3-28
Table 3-18: Top contributors to energy use impacts (paste solder) 3-30
Table 3-19: Energy use impacts by life-cycle stage (bar solder) 3-31
Table 3-20: Energy use impacts by life-cycle stage and process group (bar solder) 3-32
Table 3-21: Top contributors to energy use impacts (bar solder) 3-34
Table 3-22: Landfill space use impacts by life-cycle stage (paste solder) 3-36
Table 3-23: Landfill space use impacts by life-cycle stage and
process group (paste solder) 3-38
-------
Table 3-24: Top contributors to landfill space use impacts (paste solder) 3-40
Table 3-25: Landfill space use impacts by life-cycle stage (bar solder) 3-40
Table 3-26: Landfill space use impacts by life-cycle stage and
process group (bar solder) 3-42
Table 3-27: Top contributors to landfill space use impacts (bar solder) 3-43
Table 3-28: Global warming impacts by life-cycle stage (paste solder) 3-47
Table 3-29: Global warming impacts by life-cycle stage and
process group (paste solder) 3-48
Table 3-30: Top contributors to global warming impacts (paste solder) 3-50
Table 3-31: Global warming impacts by life-cycle stage (bar solder) 3-50
Table 3-32: Global warming impacts by life-cycle stage and
process group (bar solder) 3-52
Table 3-33: Top contributors to global warming impacts (bar solder) 3-53
Table 3-34: Stratospheric ozone depletion impacts by life-cycle stage (paste solder) .... 3-56
Table 3-35: Stratospheric ozone depletion impacts by life-cycle stage and process
group (paste solder) 3-58
Table 3-36: Top contributors to stratospheric ozone depletion impacts (paste solder) . . . 3-59
Table 3-37: Ozone-depleting substances in the LFSP inventories 3-60
Table 3-38: Stratospheric ozone depletion impacts by life-cycle stage (bar solder) 3-61
Table 3-39: Stratospheric ozone depletion impacts by life-cycle stage and
process group (bar solder) 3-63
Table 3-40: Top contributors to stratospheric ozone depletion impacts (bar solder) 3-64
Table 3-41: Geographic and temporal boundaries of inventories contributing
to the ozone depletion results 3-66
Table 3-42: Photochemical smog impacts by life-cycle stage (paste solder) 3-69
Table 3-43: Photochemical smog impacts by life-cycle stage and
process group (paste solder) 3-70
Table 3-44: Top contributors to photochemical smog impacts (paste solder) 3-72
Table 3-45: Photochemical smog impacts by life-cycle stage (bar solder) 3-73
Table 3-46: Photochemical smog impacts by life-cycle stage and
process group (bar solder) 3-74
Table 3-47: Top contributors to photochemical smog impacts (bar solder) 3-75
Table 3-48: Acidification impacts by life-cycle stage (paste solder) 3-78
Table 3-49: Acidification impacts by life-cycle stage and process group (paste solder) . . 3-79
Table 3-50: Top contributors to acidification impacts (paste solder) 3-80
Table 3-51: Acidification impacts by life-cycle stage (bar solder) 3-81
Table 3-52: Acidification impacts by life-cycle stage and process group (bar solder) .... 3-83
Table 3-53: Top contributors to acidification impacts (bar solder) 3-84
Table 3-54: Air particulate impacts by life-cycle stage (paste solder) 3-87
Table 3-55: Air paniculate impacts by life-cycle stage and process group (paste solder) . 3-88
Table 3-56: Top contributors to air particulate impacts (paste solder) 3-90
Table 3-57: Air particulate impacts by life-cycle stage (bar solder) 3-90
Table 3-58: Air particulate impacts by life-cycle stage and process group (bar solder) . . . 3-92
Table 3-59: Top contributors to air particulate impacts (bar solder) 3-93
Table 3-60: Water eutrophication impacts by life-cycle stage (paste solder) 3-96
Table 3-61: Water eutrophi cation impacts by life-cycle stage and
process group (paste solder) 3-97
Table 3-62: Top contributors to water eutrophication impacts (paste solder) 3-98
VI
-------
Table 3-63: Water eutrophication impacts by life-cycle stage (bar solder)
Table 3-64: Water eutrophi cation impacts by life-cycle stage and
process group (bar solder)
Table 3-65: Top contributors to water eutrophication impacts (bar solder)
Table 3-66: Water quality impacts by life-cycle stage (paste solder)
Table 3-67: Water quality impacts by life-cycle stage and
process group (paste solder)
Table 3-68: Top contributors to water quality impacts (paste solder)
Table 3-69: Water quality impacts by life-cycle stage (bar solder)
Table 3-70: Water quality impacts by life-cycle stage and process group (bar solder)
Table 3-71: Top contributors to water quality impacts (bar solder)
Table 3-72: Hazard values for carcinogenicity WOE if no slope factor is available . .
Table 3-73: Occupational non-cancer impacts by life-cycle stage (paste solder)
Table 3-74: Occupational cancer impacts by life-cycle stage (paste solder)
Table 3-75: Occupational non-cancer impacts by life-cycle stage and
process group (paste solder)
Table 3-76: Occupational cancer impacts by life-cycle stage and
process group (paste solder)
Table 3-77: Top contributors to occupational non-cancer impacts (paste solder)
Table 3-78: Top contributors to occupational cancer impacts (paste solder)
Table 3-79: Occupational non-cancer impacts by life-cycle stage (bar solder)
Table 3-80: Occupational cancer impacts by life-cycle stage (bar solder)
Table 3-81: Occupational non-cancer impacts by life-cycle stage and
process group (bar solder)
Table 3-82: Occupational cancer impacts by life-cycle stage and
process group (bar solder)
Table 3-83: Top contributors to occupational non-cancer impacts (bar solder)
Table 3-84: Top contributors to occupational cancer impacts (bar solder)
Table 3-85: Public non-cancer impacts by life-cycle stage (paste solder)
Table 3-86: Public cancer impacts by life-cycle stage (paste solder)
Table 3-87: Public non-cancer impacts by life-cycle stage and
process group (paste solder)
Table 3-88: Public cancer impacts by life-cycle stage and
process group (paste solder)
Table 3-89: Top contributors to public non-cancer impacts (paste solder)
Table 3-90: Top contributors to public cancer impacts (paste solder)
Table 3-91: Public non-cancer impacts by life-cycle stage (bar solder)
Table 3-92: Public cancer impacts by life-cycle stage (bar solder)
Table 3-93: Public non-cancer impacts by life-cycle stage and
process group (bar solder)
Table 3-94: Public cancer impacts by life-cycle stage and process group (bar solder)
Table 3-95: Top contributors to public non-cancer impacts (bar solder)
Table 3-96: Top contributors to public cancer impacts (bar solder)
Table 3-97: Aquatic ecotoxicity impacts by life-cycle stage (paste solder)
Table 3-98: Aquatic ecotoxicity impacts by life-cycle stage and
process group (paste solder)
Table 3-99: Top contributors to aquatic ecotoxicity impacts (paste solder)
Table 3-100: Aquatic ecotoxicity impacts by life-cycle stage (bar solder)
3-99
101
102
105
107
108
109
110
112
117
118
120
3-123
3-
3-
O
3-
O
125
127
128
130
132
3-134
3-
3-
O
3-
O
135
137
138
145
147
3-149
3-
3-
3-
3-
O
3-
3-
3-
3-
O
3-
O
3-
151
152
153
156
157
160
161
162
163
171
174
175
177
vn
-------
Table 3-101: Aquatic ecotoxicity impacts by life-cycle stage and
process group (bar solder) 3-179
Table 3-102: Top contributors to aquatic ecotoxicity impacts (bar solder) 3-180
Table 3-103: Energy estimates for the reflow application process 3-183
Table 3-104: Impact categories and alloys with majority of impacts from energy
used in reflow application of paste solders 3-184
Table 3-105: Use/application energy sensitivity analysis: percent contribution
of use/application stage to energy impacts 3-185
Table 3-106: Alternative silver production analysis (paste solder) 3-188
Table 3-107: Comparison of baseline and alternate LCA analyses (paste solder) 3-189
Table 3-108: Alternative silver production analysis (bar solder) 3-189
Table 3-109: Comparison of baseline and alternate LCA analyses (bar solder) 3-189
Table 3-110: Alternative lead leachate analysis for selected impact categories
in the paste solder results 3-190
Table 3-111: Alternative lead leachate analysis for selected impact categories
in the bar solder results 3-191
Table 3-112: Summary of natural resources impact scoring 3-193
Table 3-113: Summary of atmospheric resource impact scoring 3-193
Table 3-114: Summary of water resource impact scoring 3-194
Table 3-115: Summary of human health and ecotoxicity impact scoring 3-194
Table 3-116: Data sources for equivalency factors and hazard values 3-195
Table 3-117: Paste solder LCIA results 3-197
Table 3-118: Solder paste life-cycle stages contributing a majority of impacts 3-198
Table 3-119: Top contributing flows to SnPb solder paste impacts 3-200
Table 3-120: Top contributing flows to SAC solder paste impacts 3-201
Table 3-121: Top contributing flows to BSA solder paste impacts 3-202
Table 3-122: Top contributing flows to SABC solder paste impacts 3-203
Table 3-123: Bar solder LCIA results 3-204
Table 3-124: Bar solder life-cycle stages contributing a majority of impacts 3-205
Table 3-125: Top contributing flows to SnPb bar solder impacts 3-208
Table 3-126: Top contributing flows to SAC bar solder impacts 3-209
Table 3-127: Top contributing flows to SnCu bar solder impacts 3-210
Vlll
-------
Figures
Figure 1-1: Life-cycle stages of solder alternatives 1-7
Figure 1-2: Typical solder joints for both through-hole and surface mount connections . . 1-9
Figure 1-3: Solder life-cycle conceptual model 1-10
Figure 2-1: Unit process inventory conceptual diagram 2-1
Figure 2-2: SnPb paste solder life-cycle processes 2-2
Figure 2-3: SAC paste solder life-cycle processes 2-2
Figure 2-4: BSA paste solder life-cycle processes 2-3
Figure 2-5: SABC paste solder life-cycle processes 2-3
Figure 2-6: SnPb bar solder life-cycle processes 2-4
Figure 2-7: SAC bar solder life-cycle processes 2-4
Figure 2-8: SnCu bar solder life-cycle processes 2-5
Figure 2-9: Solder manufacturing process diagrams for bar and paste solders 2-22
Figure 2-10: Typical post-industrial recycling process flow diagram 2-24
Figure 2-11: Solder paste reflow process diagram 2-29
Figure 2-12: Reflow profiles for soldier pastes 2-30
Figure 2-13: Reflow test PWB assembly 2-31
Figure 2-14: Process flow diagram for wave solder 2-33
Figure 2-15: Unregulated recycling and disposal process flow diagram 2-42
Figure 2-16: Paste solder total mass inputs 2-44
Figure 2-17: Paste solder total mass outputs 2-45
Figure 2-18: Bar solder total mass inputs 2-45
Figure 2-19: Bar solder total mass outputs 2-46
Figure 3-1: Impact classification conceptual model 3-4
Figure 3-2: Solder paste total life-cycle impacts: NRR use 3-11
Figure 3-3: Solder paste total life-cycle impacts: RR use 3-13
Figure 3-4: Bar solder total life-cycle impacts: NRR use 3-19
Figure 3-5: Bar solder total life-cycle impacts: RRuse 3-20
Figure 3-6: Paste solder total life-cycle impacts: energy use 3-27
Figure 3-7: Bar solder total life-cycle impacts: energy use 3-31
Figure 3-8: Paste solder total life-cycle impacts: landfill space use 3-37
Figure 3-9: Bar solder total life-cycle impacts: landfill space use 3-41
Figure 3-10: Solder paste total life-cycle impacts: global warming 3-47
Figure 3-11: Bar solder total life-cycle impacts: global warming 3-51
Figure 3-12: Solder paste total life-cycle impacts: stratospheric ozone depletion 3-56
Figure 3-13: Bar solder total life-cycle impacts: stratospheric ozone depletion 3-61
Figure 3-14: Ozone depletion impacts with methyl bromide only (paste solder) 3-66
Figure 3-15: Solder paste total life-cycle impacts: photochemical smog 3-69
Figure 3-16: Bar solder total life-cycle impacts: photochemical smog 3-73
Figure 3-17: Solder paste total life-cycle impacts: acidification 3-78
Figure 3-18: Bar solder total life-cycle impacts: acidification 3-82
Figure 3-19: Solder paste total life-cycle impacts: air particulates 3-87
Figure 3-20: Bar solder total life-cycle impacts: air particulates 3-91
Figure 3-21: Solder paste total life-cycle impacts: water eutrophication 3-96
Figure 3-22: Bar solder total life-cycle impacts: water eutrophi cation 3-99
Figure 3-23: Solder paste total life-cycle impacts: water quality (BOD & solids) 3-105
Figure 3-24: Bar solder total life-cycle impacts: water quality (BOD & solids) 3-109
ix
-------
Figure 3-25: Solder paste total life-cycle impacts: occupational non-cancer 3-119
Figure 3-26: Solder paste total life-cycle impacts: occupational cancer 3-121
Figure 3-27: Bar solder total life-cycle impacts: occupational non-cancer 3-131
Figure 3-28: Bar solder total life-cycle impacts: occupational cancer 3-132
Figure 3-29: Solder paste total life-cycle impacts: public non-cancer 3-146
Figure 3-30: Solder paste total life-cycle impacts: public cancer 3-147
Figure 3-31: Bar solder total life-cycle impacts: public non-cancer 3-156
Figure 3-32: Bar solder total life-cycle impacts: public cancer 3-158
Figure 3-33: Comparative lead HV analysis (paste solder) 3-168
Figure 3-34: Comparative lead HV analysis (bar solder) 3-168
Figure 3-35: Solder paste total life-cycle impacts: aquatic ecotoxicity 3-172
Figure 3-36: Bar solder total life-cycle impacts: aquatic ecotoxicity 3-177
Figure 3-37: Sensitivity analysis of energy consumption during
reflow solder application 3-185
-------
Acronyms and Abbreviations
Ag silver
Amt amount
AP acidification potential
Bi Bismuth
BOD biological oxygen demand
BOM bill of materials
BSA bismuth-tin-silver
CAAA Clean Air Act Amendments
cc cubic centimeters
CFC chlorofluorocarbon
CFC-11 trichlorofluromethane
CFC-12 dichlorodifluorom ethane
CFC-13 chlorotriflurom ethane
CFC-114 dichlorotetrafluorethane
CHEMS Chemical Hazard Evaluation for Management Strategies
COD chemical oxygen demand
CO2 carbon dioxide
Cu Copper
D density
DEAM Database for Environmental Analysis and Management
DfE Design for the Environment
DHHS Department of Health and Human Services
DQI data quality indicator
EF equivalency factor
EFSOT environmentally friendly soldering technology
EIA Electronic Industries Alliance
EIA Energy Information Alliance
EOL end-of-life
EP eutrophication potential
EPA Environmental Protection Agency
EU European Union
GaBi life-cycle assessment software tool
GWP global warming potential
H heat value
HAP Hazardous Air Pollutant
HEAST Health Effects Assessment Summary Tables
HSDB Hazardous Substances Data Bank
HV hazard value
IARC International Agency for Research on Cancer
IPC Association Connecting Electronics Industries
IPCC Intergovernmental Panel on Climate Change
IRIS Integrated Risk Information System
IS impact score
ISO International Standards Organization
ITRI Interconnect Technology Research Institute
kg kilogram
XI
-------
Kow octanol water co-efficient
kW kilowatt
LC lethal concentration
LC50 lethal concentration to 50 percent of the exposed fish population
LCA life-cycle assessment
LCI life-cycle inventory
LCIA life-cycle impact assessment
LFSP Lead-Free Solder Project
LNG liquified natural gas
LOAEL lowest-observed-adverse-effect levels
LPG liquified petroleum gas
m3 cubic meter
MACT maximum achievable control technology
ME&P materials extraction and processing
mg milligrams
MITI Ministry of International Trade and Industry
MJ megajoule
MSW municipal solid waste
N nitrogen
NCMS National Center for Manufacturing Sciences
NEMI National Electronics Manufacturing Initiative
NMVOC non-methane volatile organic compounds
NOAEL no-observed-adverse-effect levels
NOEC no-observed-effect concentration
NOEL no-observed-effect level
NRR non-renewable resource
OD ozone depletion
ODP ozone depletion potential
OEM original equipment manufacturer
P phosphorus
Pb lead
PI post-industrial
PM particulate matter
PM10 particulate matter with an average aerodynamic diameter less than 10
micrometers
PO43" phosphate
POCP photochemical oxidant creation potential
POTW publicly-owned treatment works
PWB printed wiring board
QSAR quantitative structure-activity relationship
RC recycled content
RCRA Resource Conservation and Recovery Act
ROHS Restriction of Hazardous Substances
RR renewable resource
RSS ramp-soak-spike
RTECS Registry of Toxic Effects of Chemical Substances
SABC tin-silver-bismuth-copper
SAC tin-silver-copper
xn
-------
SAR structure-activity relationship
SET AC Society of Environmental Toxicology and Chemistry
SF slope factor
Sn tin
SnPb tin-lead
SnCu tin-copper
SO2 sulfur dioxide
SPLP synthetic precipitation leaching procedure
SWL solid waste landfill
TAL time above liquidous
TCLP toxic characteristic leachate procedure
TRI toxic release inventory
TSP total suspended particulates
TSS total suspended solids
UF University of Florida
UT University of Tennessee
VOC volatile organic carbons
WEEE Waste Electronics and Electronic Equipment
WOE weight of evidence
Zn Zinc
Xlll
-------
EXECUTIVE SUMMARY
This report presents the results of a voluntary, cooperative project among the Design for
the Environment (DfE) Program in the Economics, Exposure, and Technology Division of the
U.S. Environmental Protection Agency's (EPA) Office of Pollution Prevention and Toxics, the
University of Tennessee (UT) Center for Clean Products and Clean Technologies, the electronics
industry, and other interested parties to develop a life-cycle model and to assess the life-cycle
environmental impacts of lead-based and lead-free solders. Analyses are presented for both bar
and paste soldering applications used in electronics manufacturing.
The DfE Lead-Free Solder Project (LFSP) used life-cycle assessment (LCA) as an
environmental evaluation tool that looked at the full life cycle of the product from materials
acquisition to manufacturing, use, and final disposition. As defined by the Society of
Environmental Toxicology and Chemistry (SETAC), there are four major components of an
LCA study: goal definition and scoping, in which the goals of the study and boundaries of the
assessment are determined; life-cycle inventory (LCI), in which data on material and energy
inputs and outputs for each process in each life-cycle stage are gathered; life-cycle impact
assessment (LCIA), in which the LCI data are entered into a tool-kit, and impact scores are
generated for each impact category in each life-cycle stage; and improvement assessment. The
more recent International Standards Organizations (ISO) definition of LCA includes the same
first three components, but replaces the improvement assessment component of LCA with a life-
cycle interpretation component. During the interpretation component, the user weighs the impact
scores from the different categories and determines how to improve a product, or decides which
product poses an environmentally preferable profile. As is the case with this study, this last step
of the LCA process is often left to the user of the results, because it involves weighting the
results toward the impact categories that are of most concern to the user. However, there are
many accepted methods for performing this step including the eco-indicator '99 method, or the
analytical hierarchy process, which is a technique for multi-attribute decision making.
Commercially available software packages are available for conducting such analyses.
LCAs are generally global and non-site specific in scope. The LFSP uses the LCA
methodology developed and refined in a previous DfE LCA of desktop computer displays (EPA
2003a) and published in Socolof et a/., 2003. LCAs evaluate the potential environmental impacts
from each of the following major life-cycle stages: raw materials extraction and processing,
product manufacturing, product use/application, and final disposition at end-of-life (EOL). The
inputs (e.g., resources and energy) and outputs (e.g., products, emissions, and waste) within each
life-cycle stage are evaluated to determine the environmental impacts.
In this study and project report, the goal and scope of the LFSP are the subject of
Chapter 1. The life-cycle inventory (LCI), which describes the method of quantification of raw
material and fuel inputs, along with solid, liquid, and gaseous emissions and effluents, is the
subject of Chapter 2. The life-cycle impact assessment (LCIA) involves the translation of the
environmental burdens identified in the LCI into environmental impacts and is described in
detail in Chapter 3. The improvement assessment or life-cycle interpretation is left to the
electronics industry or any other interested party given the results of this study.
ES-1
-------
I. GOAL DEFINITION AND SCOPE
Purpose and Need
The purpose of this study is three-fold: (1) to establish a scientific baseline that evaluates
the potential life-cycle environmental impacts of selected lead-based and lead-free solder
alternatives using LCA methodologies; (2) to evaluate the effects of lead-free solders on
teachability, recycling, and reclamation at end-of-life; and (3) to identify data gaps or other
potential areas of analysis for future investigation by EPA or industry. This study is designed to
provide the electronics industry with the information needed to improve the environmental
attributes of electronics and electronic equipment containing solder. The evaluation considers
impacts related to material consumption, energy, air resources, water resources, landfills, human
toxicity, and ecological toxicity, as well as teachability and recycling. It is intended to provide
valuable data not previously published, and an opportunity to use the model developed for this
project in future improvement evaluations that consider life-cycle impacts. It also will provide
the industry and regulating authorities with valuable information to make environmentally
informed decisions regarding solders and electronics, and enable them to consider the relative
environmental merits of an alternative solder along with its performance and cost.
Solder is the chief method for attaching components to a printed wiring board (PWB)
during the manufacturing of electronic assemblies. Eutectic tin-lead (SnPb) solder has long been
the primary choice for assembling electronics due to its reflow properties, low melting point, and
the relative ductility of the solder joints formed. Lead, however, has come under increasing
regulatory scrutiny due to its relatively high toxicity to human health and the environment. In
2001, the European Union (EU) proposed the Waste Electronics and Electronic Equipment
(WEEE), and the associated Restriction of Hazardous Substances (ROHS) directives, that bans
the use of lead in electronics devices sold in the EU beginning in July 2006. The directives have
since been finalized. In Japan, subsequent to takeback (recycling) legislation that took effect in
that country in 2001, the Japanese EPA and Ministry of International Trade and Industry (MITI)
suggested a voluntary phase-out of lead, with lead levels reduced to half by 2000, and by two-
thirds by 2005, along with increased EOL product recycling. In response, electronics industry
members have undertaken the development and evaluation of alternative lead-free alloys as
potential replacements for the SnPb solder. Thus far, the focus of industry research has been on
performance-based issues. While there have been some screening-level assessments of the life-
cycle environmental impacts of paste solder, there has not to-date been a comprehensive
quantitative study of the leading lead-free paste solder alternatives, nor has there been any study
of bar solders. Given the importance of solder during the manufacture of electronics, the
likelihood of the impending EU ban, and the unknown environmental profiles of the leading
solder alternatives, there is a need for an independently conducted, science-based evaluation of
the potential life-cycle environmental impacts of the SnPb solder and the leading alternative
solder alloys.
ES-2
-------
Targeted Audience and Use of the Study
The electronics industry is expected to be one of the primary users of the LFSP study
results. The project aims to provide the industry with an objective analysis of the life-cycle
environmental impacts of selected lead-free solders. Scientific verification of these relative
impacts will allow industry to consider environmental concerns equitably along with
traditionally evaluated parameters of cost and performance, and to potentially redirect efforts
towards products and processes that reduce solder's environmental footprint, including energy
consumption, releases of toxic chemicals, and risks to health and the environment. Based on the
study results, the industry can perform an improvement assessment of solder alternatives.
This study was designed to provide the electronics industry with information needed to
identify impacts throughout the life-cycle of various solder alternatives that can lead to
improving the environmental attributes of solders. The LFSP study also allows the electronics
industry to make environmentally informed choices about solder alternatives when assessing and
implementing improvements such as changes in product, process, and activity design; raw
material use; industrial processing; consumer use; and waste management.
Identification of impacts from the life-cycle of lead-free solders also can encourage
industry to implement pollution prevention options such as the development and demonstration
projects, and to foster technical assistance and training. The electronics industry can use the
tools and data provided by this study to evaluate the health, environmental, and energy
implications of the solder alternatives. Using this evaluation, the U.S. electronics industry may
be better prepared to meet the growing demand for extended product responsibility; to help guide
public policy towards informed, scientifically based solutions that are environmentally
preferable; and to be better able to meet the competitive challenges of the world market.
Potentially, the LCA model and results presented by this study provide a baseline upon which
solder alternatives not included in the study can be evaluated. This will allow for further,
expedited LCA studies, whose growing popularity within the industry puts them in demand by
original equipment manufacturers (OEMs) and international organizations.
The information generated in this study also can be used by the electronics industry to
select the lead-free solders that work well for a given application and that pose the fewest risks
to public health and the environment over their entire life cycles. The study results should
inform the activities of community action groups and help governmental organizations to better
manage their electronics purchasing and EOL disposition activities.
Product System
The product system was divided into two groups—bar solders and paste solders—based
on the manner that they are applied to the circuit assembly. Bar solders are melted in a solder
pot and then pumped through a nozzle that forms a defined wave over which the assembly is
passed. Wave soldering is used to attach large surface devices and through-hole components.
Paste solders are screened onto the boards to facilitate placement of components, then reflowed
by passing the assembly though a high-temperature oven. Reflow soldering is used to attach
surface mount components and other micro-componentry to a circuit board during assembly.
ES-2
-------
The solders evaluated in the study are listed in Table ES-1. Solders were selected for
evaluation by project participants based on the results of initial industry research on solder
performance, the likelihood of industry-wide adoption of the solder, and the prioritized interests
of project stakeholders. Eutectic SnPb solder (bar and paste) was selected as the baseline for
both wave and reflow applications. Tin/silver/copper (SAC) was selected because of its ability
to function in both the wave and reflow solder environment, and because it has emerged as a
leading candidate for adoption as an alternative solder during industry testing (NEMI, 2002).
Other solder pastes included two bismuth containing solders, selected for their low melting
temperatures and to evaluate their impacts at end-of-life. For bar solders, in addition to SnPb
and SAC, tin-copper (SnCu) was included as a potential low-cost alternative that is currently in
limited use.
Product systems in an LCA are evaluated on a functionally equivalent basis to provide a
reference for relating process inputs and outputs to the inventory and impact assessment across
alternatives. For this project, the functional unit is a unit volume of solder required to form a
viable surface mount or through-hole connection between the PWB and the component, or
multiples thereof. The selection of the functional unit was based on the knowledge that a similar
volume of solder is required to fill the space in a solder joint regardless of the type of solder
used. A volume of one thousand cubic centimeters (cc) of solder was selected for use as the
functional unit in the LCA. The selection of this functional unit is independent of PWB design
or configuration because the number and types of connections formed by the solder would be the
same for each alternative.
Table ES-1. Solders selected for evaluation
Solder alloys
Tin-Lead (SnPb) (baseline)
Tin-Copper (SnCu)
Tin-Silver-Copper (SAC)
Bismuth-Tin-Silver (BSA)
Tin-Silver-Bismuth-Copper
(SABC)
Composition
63 Sn 737 Pb
99.2Sn/0.8Cu
95.5 Sn/3.9Ag/0.6Cu
57Bi/42Sn/1.0Ag/
96 Sn 72.5 Ag/l.OBi/0.5 Cu
Density
(g/cc)
8.4
7.3
7.35
8.56
7.38
Melting
Point (°c)
183
227
218
138
215
Application type
Paste and Bar
Bar
Paste and Bar
Paste
Paste
Assessment Boundaries
In a comprehensive cradle-to-grave analysis, the solder system includes five life-cycle
stages: (1) raw materials extraction/acquisition; (2) materials processing; (3) product
manufacture; (4) product use/application; and (5) final disposition/EOL.
The geographic boundaries of this assessment depend on the life-cycle stage. For
example, the raw materials acquisition and processing of the metals comprising the solder alloys
is done throughout the world and is represented by worldwide data sets. Product manufacturing
also occurs worldwide; however, all of the solders selected for evaluation in this project are
ES-4
-------
manufactured in the U.S. Although a worldwide geographic boundary was considered for the
manufacturing stage, ultimately the data were obtained primarily from the U.S. Similarly, solder
application in the use stage is done worldwide; but, given the geographic location of the project
researchers, data were only collected from manufacturers in the U.S. The EOL evaluation
focuses on solders and electronic products containing solder that reach the end of their lives in
the U.S. Due to limited availability of U.S. EOL data (e.g., on recycling), however, EOL data
from other countries also were used. For purposes of this study, the geographic boundaries for
all life-cycle stages are worldwide; however, several stages are primarily represented by data
collected in the U.S.
Temporal boundaries of the LFSP are defined from 2001 to 2003, the period representing
the majority of data collected. Data for manufacturing and use/application life-cycle stages
reflect the period stated. Unlike most products, solder does not have a use life-cycle that extends
over a large time frame, instead it occurs over the relatively short period of time required to
assemble a printed wiring board. While EOL disposition for electronics can be temporarily
displaced for many years, data used to assess EOL impacts were based during the time period
mentioned.
Impacts from the transportation and distribution of materials, products, and wastes
throughout the life-cycle of a solder are included in most of the upstream processes where
secondary data are used that already include transportation. For the primary data collected from
solder manufacturers, PWB assemblers, and recyclers, transportation was not included in the
scope, mostly due to limited project resources. The differences in transportation among the
different solder alloys in the associated life-cycle stages (i.e., manufacturing, use, and EOL) are
not expected to be significant. Therefore, excluding transportation from primary data collection
is not expected to adversely affect the study results.
ES-5
-------
II. LIFE-CYCLE INVENTORY (LCI)
General Methodology
A LCI is the identification and quantification of the material and resource inputs and
emission and product outputs from the unit processes in the life cycle of a product system. For
the LFSP, LCI inputs include materials used in the solder products; ancillary materials used in
processing and manufacturing the solders; and energy and other resources consumed in the
manufacturing, use, or final disposition of the solders. Outputs include products, air emissions,
water effluents, and releases to land. Figure ES-1 shows the unit processes that are included in
the scope of this project for the SnPb solder paste life cycle. While process diagrams for solder
alternatives may vary somewhat from solder to solder, and from paste to bar, a scope for each
alternative is similar to that shown for the SnPb paste solder alloy. The differences include the
following: (1) the upstream production of lead will be replaced with the appropriate alternate
metals found in each alloy; (2) liquified petroleum gas (LPG) also is used as a fuel input in bar
manufacturing, in addition to the fuels used inpaste manufacturing (i.e., natural gas, heavy fuel
oil); and (3) for the BSA alloy, due to the high bismuth content and the potentially prohibitive
cost of copper smelting due to the bismuth content, flows from demanufacturing are assumed to
be sent to landfilling or incineration instead of copper smelting.
Fuel Oil
-Hvy
Fuel Oil
-Lt
i Electronic
i Product Use-
(Not modeled)
SnPb Paste
Manufacturing
Unregulated
Recycling and
Disposal
Natural Gas
Fuel Oil-Hvy
Bold= Primary Data
Dash= Not modeled
Figure ES-1. SnPb Paste Solder Life-Cycle Processes
ES-6
-------
Data also were collected on the final disposition of emissions outputs, such as whether
outputs are released directly to the environment, recycled, treated, or disposed. This information
was used to determine which impacts will be calculated for a particular inventory item. Methods
for calculating impacts are discussed in Chapter 3, Life-Cycle Impact Assessment.
Given the enormous amount of data involved in inventorying all of the inputs and outputs
for a product system, decision rules were used to determine which materials or unit processes to
include in the LCI. Decision rules are designed to make data collection manageable while still
representative of the product system and its impacts; they were based on mass, environmental,
energy, and functional significance. Data were collected from both primary and secondary
sources. Table ES-2 lists the types of data (primary or secondary) used for each life-cycle stage.
In general, greater emphasis was placed on collecting data and developing models for the
product manufacturing, use, and EOL life-cycle stages.
Table ES-2. Data types by life-cycle stage
Life-cycle stage
Upstream
(materials extraction and processing)
Solder manufacturing
Use (Solder Application)
Final disposition
(Leachability, recycling and/or disposal)
Data types
Secondary data
Primary data
Primary data
Primary and secondary data
In the LFSP, LCI data were allocated to the functional unit (i.e., 1,000 cubic centimeters
of solder) as appropriate. The data that were collected for this study were either obtained using
questionnaires developed for this project, site visits, and performance testing (i.e., primary data),
or from existing databases (i.e., secondary data). LCI data were imported into GaBi, a publicly
available life-cycle assessment tool in which customized life-cycle process profiles were
developed for each of the solder alloys.
LCI data quality was evaluated based on the following data quality indicators (DQIs):
(1) the source type (i.e., primary or secondary data sources); (2) the method in which the data
were obtained (i.e., measured, calculated, estimated); and (3) the time period for which the data
are representative. Any proprietary information required for the assessment was aggregated to
protect confidentiality.
A critical review process was maintained in the LFSP LCA to help ensure that
appropriate methods were employed and study goals were met. A project Core Group and
Technical Work Group, both consisting of representatives from industry, academia, government,
and other interested parties provided critical reviews of the assessment. The Core Group served
as the project steering committee and was responsible for approving all major scoping
assumptions and decisions, as well as for providing guidance on technical issues. The Technical
Work Group also provided technical guidance and were given the opportunity to review all
major project deliverables, including the final LCA report.
ES-7
-------
Upstream Life-Cycle Stage Methodology
The materials extraction and processing inventories for lead, tin, copper, and silver were
available as secondary data. The lead, copper, and silver inventories were contained within the
GaBi software and databases (GaBi, 2000). The tin inventory was obtained from Ecobilan in
their Database for Environmental Analysis and Management (Ecobilan, 1999). No secondary
data sets were publicly available for bismuth, so a bismuth data set was constructed from the
lead and copper inventories weighted to represent the percentage of bismuth co-mined with each
metal.
In the upstream processes for metals production, fuel and energy data are included within
the secondary inventory data sets. For the primary data collected in the other life-cycle stages of
this analysis, fuel and energy production inventory data are included as separate processes.
Although these processes are described in the "Upstream Life-Cycle Stage Methodology"
section of this report, the inventory and impact results associated with fuel/energy production are
presented with the appropriate life-cycle stage in which the fuel or energy is used. For example,
SnPb solder manufacturing requires natural gas as an input, therefore, the impacts associated
with the production of natural gas (needed during solder manufacturing) are presented within the
manufacturing life-cycle stage results. Fuel inventories were obtained from secondary data
sources. The natural gas, fuel oils, and diesel fuel inventories were contained within the GaBi
databases, while the LPG inventory was obtained from DEAM. Electricity generation inventory
data was obtained from a GaBi data set based on the U.S. electric grid.
Manufacturing Stage Methodology
The inventories for the product manufacturing life-cycle stage were developed from
primary data collected from manufacturers in North America and Japan. Five companies
provided primary data for the analyses. For the paste alloys, data were obtained from three
manufacturers, and for the bar alloys, data were collected from all five manufacturers. All told,
these five solder manufacturers account for approximately eighty percent of the U.S. market
demand. Data were collected through site-visits to three of the manufacturing facilities
throughout North America and through questionnaires forwarded to the remaining participating
companies. Manufacturers provided inventory data for the manufacture of both lead-based and
lead-free solders, as well as for the processes used to reclaim or recycle post-industrial solder
waste returned by customers. Allocation of data to the functional unit was conducted as
necessary. Processes for which more than one company's data were collected were averaged
together.
The quality of the manufacturing stage data is dependent on how the data were obtained,
measured, calculated, or estimated. Because solder manufacturers have been producing SnPb in
volume for many years, the majority of SnPb solder manufacturing data were measured or
calculated based on known process parameters and experience. Demand for the lead-free
solders, though increasing, had not yet been enough to require them to be made in anything other
than batch mode. Data for the lead-free solders, therefore, was often estimated based on batch
production data and on required process parameters for lead-free solder manufacture.
ES-8
-------
Product Use/Application Life-Cycle Stage Methodology
The use stage for solder was defined as the process of applying solder to the PWB during
the assembly process. LCI data were collected for the use/application stage through performance
testing conducted at two manufacturing facilities. Data measured during testing were then
compared to published data for verification and validation of testing protocols.
Protocols for testing were developed in conjunction with industry experts. Testing was
conducted for both reflow (paste solders) and wave (bar solder) assembly processes for each of
the solder alloys. For reflow application, inventory data were measured directly during testing
conducted at two manufacturing facilities using an identical protocol. Testing sites were selected
to vary the type and age of the reflow equipment so that the inventory would represent a range of
industry conditions. Energy consumption data collected were converted to a functional unit
basis and then averaged. Additional inventory data (e.g., flux consumption) were estimated from
established usage rates and experience. Wave application data were measured during
performance testing at a single facility and then compared to published data to validate the
testing. Inventory data for each of the bar solders was collected using a single protocol
developed by industry experts.
Inventory data collected for this life-cycle stage are considered to be of high quality.
Alternate analyses were conducted using the high and low energy consumption values to address
the potential effects of uncertainties in the data.
End-of-Life (EOL) Methodology
The EOL stage assumes that the solder on a PWB is in a product that has reached its end
of life. The EOL analysis does not address the disposition of the entire PWB. To be consistent
with the functional unit, the focus is on the solder and where the associated metals in the solder
are distributed at the EOL. The EOL dispositions that are considered in this analysis, followed
by the assumptions for the percent distribution of electronics to those dispositions are as follows:
•• landfilling (solid and hazardous)—72 percent;
•• incineration (waste to energy)—19 percent; and
• • recycling—9 percent;
- demanufacturing (i.e., disassembly/shredding and copper smelting)—4.5 percent;
- unregulated recycling and disposal—4.5 percent.
The unregulated recycling and disposal disposition was included based on an acknowledgment
that electronics sent for recycling are sometimes diverted to locations where unregulated
recycling and disposal may be occurring.
Primary data were collected for demanufacturing and copper smelting, while secondary
data were used for the landfilling and incineration processes. Assumptions based on the physical
properties of the solder were used to estimate releases in the unregulated recycling and disposal
disposition. The demanufacturing data were collected from three companies, and the copper
smelting data were obtained from two smelters. The data from these companies represent
facility operations ranging from 2001 to 2003.
ES-9
-------
LCI Limitations and Uncertainties
Several factors contribute to the overall quality of data for each life-cycle stage. For
example, the manufacturing stage includes data that were collected from several different
companies. The quality of one data set from one company may be different from that of another
company. Relative data quality estimates have been made for each life-cycle stage (Table ES-3).
The table also lists the major limitations associated with each life-cycle stage.
Table ES-3. Relative data quality and major limitations
Life-cycle stage
Upstream
Manufacturing
Use
EOL
Relative data quality
Moderate
Moderate to high
High
Moderate
Major limitations
Used only secondary data, not originally collected for
the purpose of the LFSP.
SnPb data expected to have few limitations; more
uncertainty with alternatives, which were not yet in
full production when data were collected.
Data are based on testing protocols developed for the
LFSP, thus few limitations expected; however, data
that were averaged had a relatively large range.
Used secondary data or assumptions for incineration,
landfilling, and unregulated recycling/disposal
processes.
Baseline LCI Results
Figures ES-2 and ES-3 present the total mass quantity of inputs and outputs, respectively,
for each paste alloy. Figures ES-4 and ES-5 present the inputs and outputs, respectively, for
each of the bar alloys. These LCI results are only intended to be used as an interim step to
conducting the LCIA; therefore, only a brief discussion is provided here. The paste solders show
similar total mass input quantities for SnPb, SAC and SABC, with SAC having the greatest mass
inventory inputs (Figure ES-2). BSA has the fewest mass inputs. The greatest contributor to
these mass inputs is water as a resource. The outputs from the paste solder life-cycles (Figure
ES-3) show SnPb, SAC, and SABC to be about equivalent to one another and BSA to have a
lower mass output. The outputs also are dominated by water emissions.
ES-10
-------
Reflow Solder Total Mass Inputs
en nnn
4n nnn
M
E 30 000
Oon nnn
12
10 000
D Waste for recycling
• Deposited goods
pj Valuable substances
D Resources
i
SnPb SAC BSA SABC
solder
Figure ES-2. Paste Solder Total Mass Inputs
Reflow Solder Total Mass Outputs
40,000
35,000
30,000
E 25,000
I) 20,000
= 15,000
^ 10,000
5,000
0
D Waste for recycling
• Deposited goods
D Emissions to soil
• Emissions to water
D Emissions to air
D Valuable substances
SnPb SAC BSA
solder
SABC
Figure ES-3. Paste Solder Total Mass Outputs
For the bar solder inventories, SAC has the greatest mass quantity of inputs, and SnPb
and SnCu mass inputs are nearly equivalent. The outputs follow the same pattern. Similar to the
paste solder, most of the inputs are from water resources. The outputs also are dominated by
emissions to water.
ES-11
-------
14 nnn
1? 000
m nnn
£ 8 nnn
o) c nnn
5 4 nnn _
9 nnn
n
Bar Solder Total Mass Inputs
•
3 E
D Waste for recycling
nValuable substances
• Resources
SnPb SAC SnCu
solder
Figure ES-4. Bar Solder Total Mass Inputs
(A
E
TO
O)
O
Bar Solder Total Mass Outputs
rjWaste for recycling
• Deposited goods
QEmissions to soil
• Emissions to water
DEmissions to air
nValuable substances
Figure ES-5. Bar Solder Total Mass Outputs
ES-12
-------
III. LIFE-CYCLE IMPACT ASSESSMENT (LCIA)
LCIA Methodology
LCIA involves the translation of the environmental burdens identified in the LCI into
environmental impacts. LCIA does not seek to determine actual impacts, but rather to link the
data gathered from the LCI to impact categories and to quantify the relative magnitude of
contribution to the impact category (Fava et al, 1993; Barnthouse et al, 1997). Further, impacts
in different impact categories are generally calculated based on differing scales and, therefore,
cannot be directly compared.
Within LCA, the LCI is a well-established methodology; however, LCIA methods are
less defined and continue to evolve (Barnthouse et al., 1997; Fava et al., 1993). Fortoxicity
impacts in particular, there are some methods being applied in practice (Guinee et a/., 1996;
ILSI, 1996; Curran, 1996), for example, toxicity potentials, critical volume, and direct valuation,
while others are in development. There is currently no general consensus among the LCA
community as to one method over another.
The UT LCIA methodology employed in this study calculates life-cycle impact category
indicators for a number of traditional impact categories, such as global warming, stratospheric
ozone depletion, photochemical smog, and energy consumption. Furthermore, the method
calculates relative category indicators for potential chronic human health, aquatic ecotoxicity,
and terrestrial ecotoxicity impacts in order to address the interest of project partners in human
and ecological toxicity, and to fill a common gap in LCIAs.
LCIAs generally classify the consumption and loading data from the inventory stage into
various impact categories (know as "classification"). "Characterization" methods are then used
to quantify the magnitude of the contribution that loading or consumption could have in
producing the associated impact. The impact categories included in the LFSP LCIA are as
follows: renewable resource use, nonrenewable materials use, energy use, landfill space use,
global warming, stratospheric ozone depletion, photochemical smog, air acidification, air
particulates, water eutrophication (nutrient enrichment), water quality (biological oxygen
demand [BOD] and total suspended solids [TSS]), occupational human health effects (cancer and
non-cancer), public human health effects (cancer and non-cancer), and aquatic ecotoxicity.
Classification of an inventory item into impact categories depends on whether the
inventory item is an input or output, what the disposition of the output is, and, in some cases, the
material properties of the inventory item. Outputs with direct release dispositions are classified
into impact categories for which impacts will be calculated in the characterization phase of the
LCIA. Outputs sent to treatment or recycle/reuse are considered inputs to treatment or
recycle/reuse processes, and impacts are not calculated until direct releases from these processes
occur. Once impact categories for each inventory item are classified, life-cycle impact category
indicators are quantitatively estimated through the characterization step.
The characterization step of LCIA includes the conversion and aggregation of LCI results
to common units within an impact category. Different assessment tools are used to quantify the
magnitude of potential impacts, depending on the impact category. Three types of approaches
are used in the characterization method for the LFSP:
ES-13
-------
•• Loading—An impact score is based on the inventory amount (e.g., resource use).
•• Equivalency—An impact score is based on the inventory amount weighed by a certain
effect, equivalent to a reference chemical (e.g., global warming impacts relative to carbon
dioxide [CO2]).
Full equivalency—All substances are addressed in a unified, technical model.
Partial equivalency—A subset of substances can be converted into equivalency
factors.
•• Scoring of inherent properties—An impact score is based on the inventory amount
weighed by a score representing a certain effect for a specific material (e.g., toxicity
impacts are weighed using a toxicity scoring method).
The scoring of inherent properties method is employed for the human and ecological
toxicity impact categories, based on the CHEMS-1 method described by Swanson etal. (1997).
The scoring method provides a hazard value (HV) for each potentially toxic material, which is
then multiplied by the inventory amount to calculate the toxicity impact score.
Using the various approaches, the UT LCIA method calculates impact scores for each
inventory item within applicable impact category. Impact scores are based on either a direct
measure of the inventory amount or some modification (e.g., equivalency or scoring) of that
amount based on the potential effect the inventory item may have on a particular impact
category. The specific calculation methods for each impact category are detailed in Chapter 3.
Impact scores are then aggregated within each impact category to calculate the various life-cycle
impact category indicators.
General LCIA Methodology Limitations and Uncertainties
The purpose of an LCIA is to evaluate the relative potential impacts of a product system
for various impact categories. There is no intent to measure the actual impacts or provide spatial
or temporal relationships linking the inventory to specific impacts. The LCIA is intended to
provide a screening-level evaluation of impacts. In addition to lacking temporal or spatial
relationships and providing only relative impacts, LCA also is limited by the availability and
quality of the inventory data. Data collection can be time consuming and expensive.
Confidentiality issues may also inhibit the availability of primary data.
Uncertainties are inherent in each parameter used to calculate impacts. For example,
toxicity data require extrapolations from animals to humans and from high to low doses (for
chronic effects) and can have a high degree of uncertainty.
Uncertainties also are inherent in such chemical ranking and scoring systems as the
scoring of inherent properties approach used for human health and ecotoxicity effects. In
particular, systems that do not consider the fate and transport of chemicals in the environment
can contribute to misclassifications of chemicals with respect to risk. Also, uncertainty is
introduced where it was assumed that all chronic endpoints are equivalent, which is likely not the
case. The human health and ecotoxicity impact characterization methods presented here are
screening tools that cannot substitute for more detailed risk characterization methods. It should
be noted, however, that in LCA, chemical toxicity is often not considered at all. This
ES-14
-------
methodology is an attempt to consider chemical toxicity where it is often ignored.
Uncertainty in the inventory data depends on the responses to the data collection
questionnaires and other limitations identified during inventory data collection. These
uncertainties are carried into the impact assessment. In this LCA, there was uncertainty in the
inventory data, which included, but was not limited to the following:
•• missing individual inventory items,
•• missing processes or sets of data,
• • measurement uncertainty,
•• estimation uncertainty,
•• allocation uncertainty/working with aggregated data, and
•• unspeciated chemical data.
The goal definition and scoping process helped reduce the uncertainty from missing data,
although it is certain that some missing data still exist. As far as possible, the remaining
uncertainties were reduced primarily through quality assurance/quality control measures (e.g.,
performing systematic double-checks of all calculations on manipulated data).
Baseline LCIA Results
Tables ES-4 and ES-5 display the baseline LCIA indicator results for paste and bar
solders, respectively. Bolded numbers in the tables indicate a score that is the greatest score for
that category among all of the solders displayed in a table. Likewise, results that are shaded
indicate the lowest impact score among the solders for that category. The indicator results
presented in the tables are the result of the characterization step of LCIA methodology, where
LCI results are converted to common units and aggregated within an impact category. It should
be noted that the impact category indicator results are in a number of different units and,
therefore, cannot be summed or compared across impact categories.
For paste solders, as shown in Table ES-4, SnPb solder has the highest score among the
solders in six impact categories, while SAC has the highest impact score in the remaining ten
impact categories. Conversely, BSA has the lowest impact scores in eleven of the 16 categories,
with SnPb having the lowest scores in the remaining five categories. When considering only the
lead-free solder paste alternatives, SAC has the highest impact scores of the remaining solders
in fourteen of the sixteen categories, with SABC having the highest impact score in the
remaining two categories (occupational cancer and aquatic ecotoxicity). BSA has the lowest
impact scores among the lead-free alternatives in every category except non-renewable resource
consumption.
As shown in Table ES-5 for bar solders, it is SAC with the highest impact score among
the bar solders in twelve of sixteen impact categories, while SnPb has the higher score in the
remaining four categories. On the other hand, SnCu has the lowest impact score of any of the
three bar solder alloys in eleven of the sixteen categories. When only the lead-free solders are
considered, SAC has the highest impact score in every impact category, while SnCu has the
lowest scores. Details of each impact category and major contributors to the impacts in those
ES-15
-------
categories are presented in Chapter 3.
Table ES-4. Paste solder LCIA results
Impact category
Non-renewable resource use
Renewable resource use
Energy use
Landfill space
Global warming
Ozone depletion
Photochemical Smog
Acidification
Paniculate matter
Eutrophication
Water quality
Occupational non-cancer
Occupational cancer
Public non-cancer
Public cancer
Aquatic ecotoxicity
Units per
functional unit*
kg
kg
MJ
m3
kg CO2-equiv.
kgCFC-11-equiv.
kg ethene-equiv.
kg SO2-equiv.
kg
kg phosphate-equiv.
kg
kg noncancertox-equiv.
kg cancertox-equiv.
kg noncancertox-equiv.
kg cancertox-equiv.
kg aquatictox-equiv.
Quality
rating**
M-H
M-H
H
M-H
H
L-M
M-H
M-H
M-H
H
H
M-H
L-M
M-H
L-M
M-H
SnPb
1.61E+03
3.48E+04
1.25E+04
2.75E-03
8.17E+02
9.95E-05
3.13E-01
6.50E+00
4.52E-01
1.22E-01
1.79E-01
5.60E+05
7.62E+01
8.80E+04
6.96E+00
1.27E+03
SAC
1.82E-KJ3
3.47E+04
1.36E+04
1.62E-02
8.73E+02
1.10E-04
6.18E-01
1.25E+01
1.30E-H)0
1.18E-01
2.26E-01
8.12E+03
7.20E+01
1.05E+04
7.05E+00
3.64E+01
BSA
1.76E+03
2.64E+04
9.76E+03
6.57E-03
6.31E+02
7.98E-05
3.61E-01
7.32E+00
5.85E-01
9.06E-02
1.64E-01
2.34E+03
6.34E+01
5.01E+03
5.15E+00
2.34E+01
SABC
1.72E+03
3.41E+04
1.31E+04
1.13E-02
8.49E+02
1.04E-04
5.05E-01
1.03E+01
1.01E+00
1.17E-01
2.06E-01
5.25E+03
7.23E+01
7.84E+03
6.51E+00
3.85E+01
* The functional unit is 1,000 cc of solder applied to a printed wiring board.
** Quality rating summarizes the overall relative data quality associated with each impact category: high (H),
medium (M), or low (L). Further explanation is provided in Section 3.2.1.3.
Notes: Bold impact scores indicate the alloy with the highest score for an impact category.
Shaded impact scores indicate the alloy with the lowest score for an impact category.
Table ES-5. Bar solder LCIA results
Impact category
Non-renewable resource use
Renewable resource use
Energy use
Landfill space
Global warming
Ozone depletion
Photochemical smog
Acidification
Paniculate matter
Eutrophication
Water quality
Occupational non-cancer
Occupational cancer
Public non-cancer
Public cancer
Aquatic ecotoxicity
Units per
functional unit*
kg
kg
MJ
mj
kg CO2-equiv.
kgCFC-11-equiv.
kg ethene-equiv.
kg SO,-equiv.
kg
kg phosphate-equiv.
kg
kg noncancertox-equiv.
kg cancertox-equiv.
kg noncancertox-equiv.
kg cancertox-equiv.
kg aauatictox-equiv.
Quality
rating**
M-H
M-H
H
M-H
H
L-M
M-H
M-H
M-H
H
H
M-H
L-M
M-H
L-M
M-H
SnPb
3.15E+02
6.03E+03
2.91E+03
1.34E-03
1.87E+02
1.87E-05
6.98E-02
1.43E+00
1.49E-01
2.14E-02
3.98E-02
7.15E+05
5.94E+01
1.33E+05
4.13E+00
1.55E+03
SAC
7.68E+02
8.76E+03
5.77E+03
2.14E-02
3.57E+02
4.13E-05
5.51E-01
1.10E+01
1.47E+00
2.57E-02
1.20E-01
1.09E+04
5.75E+01
1.22E+04
5.04E+00
1.98E+02
SnCu
3.12E+02
5.83E+03
3.40E+03
1.33E-03
2.16E+02
1.78E-05
7.06E-02
1.53E+00
1.99E-01
2.06E-02
3.64E-02
6.53E+01
5.49E+01
7.26E+02
2.58E+00
8.70E+00
* The functional unit is 1,000 cc of solder applied to a printed wiring board.
** Quality summarizes the overall relative data quality associated with each impact category: high (H), medium
(M), or low (L). Further explanation is provided in section 3.2.1.3.
Notes: Bold impact scores indicate the alloy with the highest score for an impact category.
Shaded impact scores indicate the alloy with the lowest score for an impact category.
ES-16
-------
Top Contributors by Impact Category for Paste Solders
For paste solders, Table ES-6 through ES-9 list the top contributing flows and their
associated processes and life-cycle stages for each impact category for each of the solders. The
tables show that the majority of impact categories are driven by resource flows from processes
associated with either the use/application or upstream life-cycle stages. Resource flows from
use/application life-cycle stage processes are the primary contributor to fourteen of sixteen
impact categories for SnPb, and to at least ten or more categories for each of the lead-free
alternatives, with the electricity generation process being the single largest driver. While the
upstream life-cycle stage does not drive any of the impacts for SnPb, resource flows from
upstream processes are the primary contributors to six impact categories for SAC, two categories
for SABC, and one for BSA. When considering the impacts from all of the resource flows from
each life-cycle stage, however, not just the top contributors are shown in the tables; upstream
processes are the major contributors to at least three, and as many as six categories for each of
the lead-free alternatives.
Many top contributing flows comprise a large majority of the total contribution to the
alloy's life-cycle impacts within a category. In the SnPb results, eleven of the sixteen impact
categories had top flows representing a majority of total impacts. By contrast, for lead-free
solders, only seven of the sixteen categories had flows contributing fifty percent or more. The
major contributing flow for a particular impact category varied depending on the solder.
ES-17
-------
Table ES-6. Top contributing flows to SnPb solder paste impacts
Impact category
Non-renewable resource use
Renewable resource use
Energy
Landfill space use
Global warming
Ozone depletion
Photochemical smog
Air acidification
Air particulates
Water eutrophication
Water quality
Occupational
health — non-cancer
Occupational health — cancer
Public human
health — non-cancer
Public human health — cancer
Aquatic ecotoxicity
Life-cycle stage
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
End-of-life
Use/application
End-of-life
Process
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Sn-Pb reflow application
Electricity generation
Solder landfilling (SnPb)
Electricity generation
Solder landfilling (SnPb)
Flow
Inert rock
Water
Hard coal
(resource)
Sludge (hazardous
waste)
Carbon dioxide
CFC-114
Sulphur dioxide
Sulphur dioxide
Dust (unspecified)
Chemical oxygen
demand
Solids (suspended)
SnPb solder paste
Natural gas
Lead emissions to
water
Nitrogen oxides
Lead emissions to
water
%
Contrib.
76.8
88.8
46.8
64.8
87.7
39.3
65.1
65.4
79.1
97.1
86.9
31.2
43.2
72.6
32.8
78.3
ES-18
-------
Table ES-7. Top contributing flows to SAC solder paste impacts
Impact category
Non-renewable resource
use
Renewable resource use
Energy
Landfill space use
Global warming
Ozone depletion
Photochemical smog
Air acidification
Air particulates
Water eutrophication
Water quality
Occupational
health — non-cancer
Occupational
health — cancer
Public human
health — non-cancer
Public human
health — cancer
Aquatic ecotoxicity
Life-cycle stage
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
Upstream
Upstream
Upstream
Use/application
Use/application
Use/application
Use/application
Upstream
Use/application
Upstream
Process
Electricity generation
Electricity generation
Electricity generation
Silver production
Electricity generation
Electricity generation
Silver production
Silver production
Silver production
Electricity generation
Electricity generation
SAC reflow
application
Electricity generation
Silver production
Electricity generation
Silver production
Flow
Inert rock
Water
Hard coal (resource)
Slag (hazardous
waste)
Carbon dioxide
CFC-114
Sulphur dioxide
Sulphur dioxide
Dust (unspecified)
Chemical oxygen
demand
Solids (suspended)
SAC solder paste
Natural gas
(resource)
Sulphur dioxide
Nitrogen oxides
Cadmium emissions
to water
%
Contrib.
64.1
83.7
40.5
77.8
77.1
33.4
47.9
49.5
63.9
94.1
64.7
31.5
43.0
38.7
30.4
45.7
ES-19
-------
Table ES-8. Top contributing flows to BSA solder paste impacts
Impact category
Non-renewable resource
use
Renewable resource use
Energy
Landfill space use
Global warming
Ozone depletion
Photochemical smog
Air acidification
Air particulates
Water eutrophication
Water quality
Occupational
health — non-cancer
Occupational
health — cancer
Public human
health — non-cancer
Public human
health — cancer
Aquatic ecotoxicity
Life-cycle stage
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
End-of-life
Process
Electricity generation
Electricity generation
Electricity generation
Silver production
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
BSAreflow
application
Electricity generation
Electricity generation
Electricity generation
Unregulated recycling
and disposal (BSA)
Flow
Inert rock
Water
Hard coal
Slag (hazardous
waste)
Carbon dioxide
CFC-114
Sulphur dioxide
Sulphur dioxide
Dust (unspecified)
Chemical oxygen
demand
Solids (suspended)
BSA solder paste
Natural gas
(resource)
Sulphur dioxide
Nitrogen oxides
Silver emissions to
water
%
Contrib.
51.7
85.9
44.0
57.1
83.4
36.0
41.5
42.7
45.0
95.7
69.8
32.5
37.9
41.2
32.4
63.3
ES-20
-------
Table ES-9. Top contributing flows to SABC solder paste impacts
Impact category
Non-renewable resource use
Renewable resource use
Energy
Landfill space use
Global warming
Ozone depletion
Photochemical smog
Air acidification
Air particulates
Water eutrophication
Water quality
Occupational
health — non-cancer
Occupational health — cancer
Public human
health — non-cancer
Public human health — cancer
Aquatic ecotoxicity
Life-cycle stage
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
End-of-life
Process
Electricity generation
Electricity generation
Electricity generation
Silver production
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Silver production
Electricity generation
Electricity generation
SABC reflow
application
Electricity generation
Electricity generation
Electricity generation
Unregulated recycling
and disposal (SABC)
Flow
Inert rock
water
Hard coal
Slag (hazardous
waste)
Carbon dioxide
CFC-114
Sulphur dioxide
Sulphur dioxide
Dust
(unspecified)
Chemical oxygen
demand
Solids
(suspended)
SABC solder
paste
Natural gas
(resource)
Sulphur dioxide
Nitrogen oxides
Silver emissions
to water
%
Contrib.
67.9
85.5
42.0
71.3
79.6
34.5
38.1
39.0
53.2
95.1
71.2
31.5
42.9
33.7
33.1
32.8
ES-21
-------
Top Contributors by Impact Category for Bar Solders
Tables ES-10 through ES-12 list the top contributing flows and their associated processes
and life-cycle stages for each impact category for the bar solders. Like the paste solders, the
majority of impact categories are driven by resource flows from processes associated with the
use/application or upstream life-cycle stages. Resource flows from use/application life-cycle
stage processes are the primary contributor to twelve of sixteen impact categories for SnPb, and
to at least six or more categories for each of the lead-free alternatives. Flows associated with
electricity generation are the largest contributors to the impacts in these categories.
While the use/application stage is the primary driver for the SnPb and SnCu, resource
flows associated with upstream processes are top contributors to nine impact categories for the
SAC alloy, and for two categories for the SnCu alloy. Flows from EOL processes also are
significant, being the top contributor to three impact categories for the SnPb alloy and two
categories for SnCu.
Many top contributing flows comprise a large majority of the total contribution to the
alloy's life-cycle impacts within a category. For each of the solder alloys, a minimum of eight of
the sixteen impact categories had top flows contributing fifty percent or more, with the major
contributing flow for a particular impact category dependent on the solder type.
ES-22
-------
Table ES-10. Top contributing flows to SnPb bar solder impacts
Impact category
Non-renewable resource use
Renewable resource use
Energy
Landfill space use
Global warming
Ozone depletion
Photochemical smog
Air acidification
Air particulates
Water eutrophication
Water quality
Occupational
health — non-cancer
Occupational health — cancer
Public human
health — non-cancer
Public human health — cancer
Aquatic ecotoxicity
Life-cycle stage
Use/application
Use/application
Use/application
End-of-life
Use/application
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
Use/application
Use/application
End-of-life
Use/application
End-of-life
Process
Electricity generation
Electricity generation
Electricity generation
Landfilling
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Tin production
Electricity generation
Electricity generation
SnPb wave application
SnPb wave application
Solder landfilling (SnPb)
Sn-Pb wave application
Solder landfilling (SnPb)
Flow
Inert rock
Water
Hard coal
(resource)
SnPb solder to
landfill
Carbon dioxide
CFC-114
Sulphur dioxide
Sulphur dioxide
Dust (unspecified)
Chemical oxygen
demand
Solids (suspended)
SnPb bar solder
SnPb bar solder
Lead emissions to
water
Flux material F
Lead emissions to
water
%
Contrib.
62.3
81.1
31.8
53.7
60.5
33.1
46.3
47.2
56.3
87.4
62.0
29.8
15.5
53.3
25.5
71.4
ES-23
-------
Table ES-11. Top contributing flows to SAC bar solder impacts
Impact category
Non-renewable resource
use
Renewable resource use
Energy
Landfill space use
Global warming
Ozone depletion
Photochemical smog
Air acidification
Air particulates
Water eutrophication
Water quality
Occupational
health — non-cancer
Occupational
health — cancer
Public human
health — non-cancer
Public human
health — cancer
Aquatic ecotoxicity
Life-cycle stage
Upstream
Use/application
Use/application
Upstream
Use/application
Upstream
Upstream
Upstream
Upstream
Use/application
Upstream
Use/application
Upstream
Upstream
Use/application
End-of-life
Process
Silver production
Electricity generation
Electricity generation
Silver production
Electricity generation
Silver production
Silver production
Silver production
Silver production
Electricity generation
Silver production
SAC wave application
Tin production
Silver production
SAC wave application
Unregulated recycling
and disposal (SAC)
Flow
Zinc-Pb-Cu Ore
Water
Hard coal (resource)
Slag (hazardous
waste)
Carbon dioxide
Halon(1301)
Sulphur dioxide
Sulphur dioxide
Dust (unspecified)
Chemical oxygen
demand
Solids (suspended)
SAC bar solder
Natural gas
(resource)
Sulphur dioxide
Flux material C
Silver emissions to
water
%
Contrib.
26.7
56.5
16.2
87.2
32.1
20.3
79.9
83.5
83.8
73.5
69.8
29.1
20.7
49.6
16.9
81.8
ES-24
-------
Table ES-12. Top contributing flows to SnCu bar solder impacts
Impact category
Non-renewable resource
use
Renewable resource use
Energy
Landfill space use
Global warming
Ozone depletion
Photochemical smog
Air acidification
Air particulates
Water eutrophication
Water quality
Occupational health — non-
cancer
Occupational
health — cancer
Public human
health — non-cancer
Public human
health — cancer
Aquatic ecotoxicity
Life-cycle stage
Use/application
Use/application
Use/application
End-of-life
Use/application
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
End-of-life
Process
Electricity generation
Electricity generation
Electricity generation
Landfilling
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Tin production
Electricity generation
Electricity generation
SnCu wave
application
Tin production
Electricity generation
SnCu wave
application
Unregulated recycling
and disposal (SnCu)
Flow
Inert rock
Water
Hard coal (resource)
SnCu solder to
landfill
Carbon dioxide
CFC-114
Sulphur dioxide
Sulphur dioxide
Dust (unspecified)
Chemical oxygen
demand
Solids (suspended)
SnCu bar solder
Natural gas
(resource)
Sulphur dioxide
Flux material C
Copper emissions to
water
%
Contrib.
63.5
84.8
28.0
53.8
53.3
35.2
46.3
44.5
68.9
91.6
68.5
14.8
16.7
61.9
21.3
90.4
ES-25
-------
Alternate Reflow Energy Analysis
Several alternate analyses were performed to evaluate the impact of key assumptions and
uncertainties on the overall results of the LCA. These analyses either were performed because
they evaluated data with the largest uncertainty or were major contributors to the inventory
results. One such analysis focused on the potential effect that the large range in energy
consumption data measured during reflow testing might have on the LCA results.
Energy consumed during the use/application life-cycle stage constituted a majority of the
impacts for many of the impact categories evaluated. For paste solder, nearly all of the
use/application energy consumption occurs during the reflow soldering process. The power
consumed during the reflow application process was based on primary data collected from two
facilities where test runs were conducted. The two ovens in which these tests were performed
represent different technologies with different thermal efficiencies resulting in a large range in
energy consumption rates. For the baseline analysis, an average energy consumption value from
these two test runs was used in the determination of the life-cycle impacts. The alternate
analyses re-evaluate the impacts using both the high and low energy consumption values
measured during the performance testing tojietermine the sensitivity of the baseline impact
results to these variations. Only the impacts for the energy use impact category were re-
evaluated. Other impacts categories would also be affected by the differences in power
consumption, but are unlikely to be as sensitive given the dominance of the reflow process on
energy use.
As shown in Figure ES-6, for all three scenarios (low energy, baseline, and high energy),
SAC has the highest impacts, followed by SABC, SnPb, and finally BSA. When the low and
high energy data points are used to generate life-cycle impact results for each type of solder
paste, the magnitude of the impact scores change; however, the relative comparison among
alloys remains the same. The analyses indicated that the contribution of the reflow energy to the
energy use impact category remains substantial, even when the low energy value is used (from
seventy-three to ninety percent, depending on the alloy).
Although only the energy use impact category was re-evaluated using the alternate data,
it is not necessary to re-evaluate the other impact categories. None of the other categories had a
higher percentage of their impacts attributable to the reflow energy consumption and are unlikely
to be as affected by a change in the reflow data. Overall, this analyses suggests that the relative
results between solders and the overall conclusions of the study are not too sensitive to the
variations in the reflow energy data (assuming the range used in this sensitivity analyses
represents a true or realistic range of the energy estimates for reflow applications process).
ES-26
-------
Low energy
High energy
Figure ES-6. Sensitivity Analysis of Energy Consumption
during Reflow Solder Application
Alternate Silver Inventory Analysis
Upstream silver production was the greatest contributing process group for many of the
impact categories of the lead-free solder pastes in the baseline LCA. For example, silver
production during the SAC life-cycle dominated six of the sixteen impact categories evaluated,
and was a major contributor in several others. The production of silver also contributed
significantly to the other silver-based lead-free alternatives, though to a lesser extent. Due to the
large influence that silver production had on many of the impact categories, an alternate analysis
was performed by substituting a BEAM silver data set for the GaBi silver mix data set used to
calculate the baseline results.
The results of the alternate analysis are dramatic and can be readily observed in Tables
ES-13 and ES-14, which compare the results of the alternate analysis to the baseline results for
both paste and bar solders, respectively. For the paste solders, the DEAM silver data set resulted
in a significant shift in the relative scores of the solders, increasing the number of categories in
which SnPb has the highest impact score from six to fourteen impact categories. SAC on the
other hand, while having many scores very close to SnPb, has the highest score in only one
category. BSA remains the solder with the lowest relative impacts compared to the other
solders. The overall shift in results is due to various flows in the DEAM silver inventory that
have lower values than the associated flows in GaBi. Due to a lack of available documentation
for the DEAM data, it is unclear what is causing the differences in the data sets. Some potential
reasons could be different scoping boundaries of the inventories, different processes included, or
different mines or processing plants represented.
ES-27
-------
Table ES-13. Comparison of paste solder baseline and alternate LCA analysis
Solder
Alloy
SnPb
SAC
BSA
SABC
Baseline
Highest
Score* Lowest Score*
6 5
10 0
0
0
11
0
Alternate
Highest
Score* Lowest Score*
14 0
1 1
1
0
15
0
* Numbers indicate the number of impact categories where solder has the highest or lowest score.
Table ES-14. Comparison of bar solder baseline and alternate LCA analysis
Solder
Alloy
SnPb
SAC
SnCu
Baseline
Highest
Score*
Lowest Score*
4
12
0
6
0
10
Alternate
Highest
9
7
0
Score*
Lowest Score*
6
5
5
* Numbers indicate the number of impact categories where solder has the highest or lowest score.
Likewise, the alternate analysis for bar solders results in an overall decrease in
importance of the silver mining process. As shown in the table, the number of categories for
which SnPb has the highest relative impact score rises from four to nine, while SAC decreases
from twelve to only seven. This is not as dramatic a change as was seen with the paste results;
however, several impact-specific conclusions were altered. Unlike the paste solders results, the
solder with the lowest relative impact score for any category is split among the solders.
These results indicate the high sensitivity of the overall life-cycle results for paste solders
to the silver data set. The baseline GaBi data set is believed to be of good quality and attempts
to verify the BEAM data set were inconclusive. Thus, the GaBi data set was chosen for this
analysis. These results show the possible variability and sensitivity of the results to the silver
inventory data, and suggest that additional effort to further resolve the silver mining and
extraction data would be well spent.
Alternate Leachate Analysis
The teachability study conducted for this project was used to estimate the outputs of
metals from landfilling PWB waste or residual metals in ash. Lead was found to leach to a much
greater extent than the other metals in the solders being analyzed in this study. These
teachability results contributed to the large public non-cancer and aquatic ecotoxicity impacts for
the SnPb as compared to the other alloys for both the paste and the bar solder results (see
Sections 3.2.12 and 3.2.13). The toxic characteristic leachate procedure (TCLP) teachability
study is based on a standard EPA TCLP test protocol using acetic acid, a substance known to
readily leach lead. It is unknown to what extent these test conditions represents actual landfill
ES-28
-------
conditions, which can vary dramatically over the lifetime of a landfill. As a result, the alternate
analysis was conducted using the detection limit of lead during the testing as a lower bound to
determine the sensitivity of the results to the lead teachability.
Results of the analysis indicated that even with the assumption that the lead essentially
does not leach (i.e., assuming the study detection limit for the teachability of lead), the SnPb
alloy impact scores are still at least 2.5 times higher than the score of the next closest alloy for
public non-cancer impacts, and a full order of magnitude higher for aquatic ecotoxicity. The
relative differences between SnPb and the lead-free alloys are far less than in the baseline
analysis. This analysis suggests that any elevation of the teachability data for SnPb due to the
aggressive nature of acetic acid towards the lead-based solder was unlikely to have changed the
overall impacts for SnPb relative to the other solders. The SnPb alloy would still have the higher
potential impacts for both public non-cancer and aquatic ecotoxicity than the other solder alloys,
based primarily on its relative toxicity.
ES-29
-------
REFERENCES
Barnthouse, L., J. Fava, K. Humphreys, R. Hunt, L. Laibson, S. Noesen, J. Owens, J. Todd, B.
Vigon, K. Weitz, J. Young (Eds.). 1997. Life-Cycle Impact Assessment: The State-of-
the-Art. Society of Environmental Toxicology and Chemistry, Pensacola, Florida.
Curran, M.A. 1996. Environmental Life-Cycle Assessment. McGraw-Hill,
New York, New York.
Ecobilan, 1999. Database for Environmental Analysis and Managment (DEAM) life cycle
inventory database developed by Ecobilan Group.
EPA (Environmental Protection Agency). 2001. Socolof M.L., J.G. Overly, L.E. Kincaid, J.R.
Geibig. Desktop Computer Displays: A Life-Cycle Assessment, Volumes 1 and2. U.S.
Environmental Protection Agency, EPA 744-R-01-004a,b, 2001. Available at:
http ://www. epa.gov/dfe/pub s/comp-lic/lca/toc 1. pdf.
Fava, J., R. Denison, R. Jones, B. Curran, M. Vigon, B. Selke, S. & J.A. Barnum. 1991.
Technical Framework for Life-Cycle Assessment. SET AC & SET AC Foundation for
Environmental Education, Inc. Washington, DC.
GaBi. 2000. GaBi3: The Software System for Life-Cycle Engineering. Produced by PE & IKP
(PE Product Engineering GmbH & IKP University of Stuttgart), Stuttgart, Germany.
Guinee, J., R. Heijungs, L. van Oers, D. van de Meent, T. Vermeire, M. Rikken. 1996. LCA
Impact Assessment of Toxic Releases. The Hague, The Netherlands.
ILSI (International Life Sciences Institute). 1996. Human Health Impact Assessment in Life
Cycle Assessment: Analysis by an Expert Panel. Washington, DC.
ISO (International Standards Organization). 1996. ISO 14040, Environmental Management -
Life-cycle Assessment Principles and Framework. TC 2071 SC 5N 77. International
Standards Organization, Paris.
NEMI (National Electronics Manufacturing Initiative). 2002. Press Release: "NEMI's Lead-
Free Assembly Project Reports Latest Results at APEX 2002," January 21. Available at:
http://www.nemi.org/Newsroom/PR/PR012102b.html, downloaded March 22, 2002.
SET AC (Society of Environmental Toxicology and Chemistry). 1994. Life-Cycle Assessment
Data Quality: A Conceptual Framework. SET AC and SET AC Foundation for
Environmental Education, Inc. Washington, DC.
Socolof, M.L., J.R. Geibig, M.B. Swanson. 2003. Cradle to Gate Toxic Impacts of Solders: A
ES-30
-------
Comparison of Impact Assessment Methods. IEEE ISEA Proceedings, Boston,
Massachusetts. May, 2003.
Swanson, M.B., G.A. Davis, L.E. Kincaid, T.W. Schultz, I.E. Bartmess, S.L. Jones, E.L. George.
1997. "A Screening Method for Ranking and Scoring Chemicals by Potential Human
Health and Environmental Impacts." Environmental Toxicology and Chemistry.
16(2): 372-383.
ES-31
-------
Chapter 1
GOAL DEFINITION AND SCOPE
1.1 INTRODUCTION
This project report presents the results of a life-cycle assessment (LCA) of selected lead-
based and lead-free solder alternatives. The structure of the report follows the formal components
of a LCA: goal definition and scope are the subject of Chapter 1, inventory analysis is the subject
of Chapter 2, and impact assessment is the subj ect of Chapter 3. The LCA's fourth component,
improvement assessment (or interpretation of results), is not directly addressed in this report, but
remains for the project partners who review the report.
This chapter provides an overview of the project that is the basis for this report. It
includes background information, the project's goals and scope, a summary of the methodology
employed in this LCA of lead-based and lead-free solder alternatives, descriptions of the product
systems analyzed, and an explanation of parameters that determine the project boundaries.
1.2 PROJECT BACKGROUND
The Lead-Free Solder Project (LFSP) is a voluntary, cooperative project among partners
that include the Design for the Environment (DfE) Program of the U.S. Environmental Protection
Agency's (EPA) Office of Pollution Prevention and Toxics, the Electronic Industries Alliance
(EIA), the IPC-Association Connecting Electronics Industries (IPC), individual electronics
industry companies, a high-technology research group (International Sematech), and the
University of Tennessee (UT) Center for Clean Products and Clean Technologies. The purpose
of the LFSP is to objectively assess the environmental life-cycle impacts of selected lead-free
solders as substitutes for lead-based solder. Aside from offering a baseline life-cycle assessment
of lead-based and lead-free solders, the DfE LFSP analysis also provides an assessment of the
recyclability and teachability of both types of solder.
EPA's Office of Pollution Prevention and Toxics established the DfE Program in 1992 to
encourage businesses to incorporate environmental concerns into their business decisions. The
EPA DfE Program promotes risk reduction, pollution prevention, energy efficiency, and other
resource conservation measures through process choices at a facility level. DfE industry projects
are cooperative, joint efforts among trade associations, businesses, public interest groups, and
academia to assist specific industries in identifying and evaluating environmentally sound
products, processes, and technologies. The DfE LFSP partnership consists of solder
manufacturers, manufacturers that use solder in their products, original equipment manufacturers
(OEMs) that incorporate components containing solder into their products, industry trade
association members, academic institutions, public interest groups, and EPA. The direction and
focus of the LFSP was determined by the project partners.
The DfE LFSP used LCA as a tool to evaluate the environmental effects of lead-based and
lead-free solders. A LCA is a comprehensive method for evaluating the full life cycle of the
product system, from materials acquisition through manufacturing to use and final disposition.
There are four major components of a LCA study: (1) goal definition and scope, (2) life-cycle
inventory (LCI), (3) life-cycle impact assessment (LCIA), and (4) interpretation of results (also
1-1
-------
called improvement assessment). LCAs are generally global in scope and non-site-specific. The
LFSP study incorporates goal definition and scoping as it is recommended in the
LCA process (e.g., Curran, 1996; Fava etal, 1991; ISO, 1996).
1.3 GOALS AND SCOPE: WHY PERFORM A LIFE-CYCLE ASSESSMENT OF
SOLDERS?
Defining goals and scope, the first phase of any LCA, is crucial to the project's success
because it determines why the LCA is being conducted and its general intent, as well as
specifying the product systems and data categories to be studied. These are addressed in the
sections below, which describe the project's purpose, prior research, the need for the LFSP,
market trends, and its target audience. A description of the LCA methodology specific to this
project follows in Section 1.4, and descriptions of the product systems assessed and assessment
boundaries used in the LCA can be found in Sections 1.5 and 1.6.
1.3.1 Lead-Free Solder Project Purpose
The purpose of the LFSP study is three-fold:
(1) to establish an objective, scientific baseline that evaluates the potential life-cycle
environmental impacts of selected lead-based and lead-free solder alternatives using LCA
methodologies;
(2) to evaluate the effects of lead-free solders on teachability, recycling, and reclamation
at the end-of-life; and
(3) to identify data gaps or other potential areas of analysis for future investigation by
EPA or industry.
This study evaluates both lead-based and lead-free solder alternatives, and considers impacts
related to material consumption, energy use, air resources, water resources, landfills, human
toxicity, and ecological toxicity, as well as teachability and recycling.
1.3.2 Previous Research
Substantial research has been conducted on lead-free solders that focuses on the
performance aspects of potential substitute alloys, including research by the National Center for
Manufacturing Sciences (NCMS), National Electronics Manufacturing Initiative (NEMI), and
Interconnect Technology Research Institute (ITRI). A number of these research efforts, along
with their findings, have been summarized in a separate DfE project report entitled Summary of
Lead-Free Solder Performance Based on Existing Data Provided by the Electronics Industry
(EPA, 2002). A summary of which is included in Appendix F.
In addition, some work on the health and environmental impacts of lead-free solder
alternatives has been conducted or is on-going in other countries and by individual companies.
The European Union has focused its research on the risks associated with lead solder, while a
multi-national, Japanese-based "Next Generation of Environmentally Friendly Soldering
Technology" (EFSOT) project is addressing, among other things, the life-cycle impacts of
products using lead-free solders. The multi-national research effort should be completed in 2005.
1-2
-------
There also have been a couple of screening-level LCAs that evaluated select lead-free solders
over a limited scope, primarily using pre-existing data or focusing on select life-cycle stages
(Warburg, 2003; Van der Wei, 2002). A quantitative LCA addressing both lead-based and the
lead-free solder alternatives like those selected for this study has not been completed, however,
nor has there been adequate evaluation of the teachability and recyclability of lead-free solder
alternatives.
1.3.3 Need for the Project
Lead is a key ingredient in electronic products. Releases of lead into the ambient or
workplace environment may occur from the mining or processing of lead or from the
recycling or disposing of products containing lead. Lead is a heavy metal that has been linked to
developmental abnormalities in fetuses and in children who ingest or absorb lead. Small amounts
of lead may cause hypertension and permanent mental dysfunction in adults. The Department of
Health and Human Services (DHHS) has determined that, based on animal studies, lead acetate
and lead phosphate may reasonably be anticipated to be carcinogens. Further, lead is a toxic
chemical that persists and bioaccumulates in the environment (DHHS, 1999). The toxic nature of
lead has resulted in global efforts to reduce its use.
Concern over lead's toxic effects and ensuing market and regulatory pressures have led
the U.S. electronics industry to commit to adopting lead-free solders. Such a commitment
requires that industry know as soon as possible which solder alternatives present the fewest
potential risks to both the environment and public health. Many other organizations and
individuals in the United States and abroad have expressed interest in obtaining objective,
detailed information about the life-cycle impacts of lead-free solders.
Various compositions of alloys containing tin, silver, copper, bismuth, and antimony have
been identified as leading candidates for solder substitutes. The performance of the metals and
fluxes of many of the alternatives has been studied, but their toxicity and environmental impacts
have not yet been evaluated. It is crucial to identify the potential impacts of the most promising
solder alternatives in order to determine whether any of the lead-free solders may present
significant health or environmental impacts or if previously unrecognized consequences may arise
from their use. In addition to the question of impacts, issues such as the availability of certain
metals and potential differences in workplace exposures need to be addressed. The use of lead-
free solder alternatives is a significant technological change. The electronics industry would like
to be confident that the choices made over the next few years will not be found later to pose
significant, unexpected risks.
Switching to lead-free solder will require substantial capital expenditures and could have a
broad impact on public health and the environment. Managing the environmental impacts posed
by this change is crucial to the long-term environmental sustainability of both the U.S. and global
economies. As a result, the electronics industry, public-interest groups, and governmental
organizations are all concerned about assessing the environmental and human-health impacts of
the lead-free alternatives to lead-based solder.
Given the current trends toward lead-free solders, the environmental concerns about lead-
based solder, and the fact that the relative environmental impacts of solder alternatives have not
yet been completed, this study fills a need for a quantitative environmental life-cycle analysis of
lead-free solders. The LFSP offers the opportunity to mitigate current and future risks by
assisting the electronics industry to identify lead-free solders that are less toxic and that pose the
fewest risks over their life cycle. In addition, when this study began, only limited information on
1-3
-------
leachability and recycling was available for some of the alternatives; this report addresses both of
these issues.
1.3.4 Market Trends
In the year 2000, approximately 48,000 tons (97 million pounds) of lead-based solder
were used worldwide (Bernier, 2002). Initiatives in Europe and Japan mandate or require
voluntary elimination of lead from electronic products. In Europe, the Restriction of Hazardous
Substances (ROHS) in Electrical and Electronic Equipment (2002/95/EC) stipulates restrictions
on the use of hazardous substances and will require lead and other selected toxic chemicals in
electrical and electronic equipment to be replaced by July, 2006. In Japan, subsequent to
takeback (recycling) legislation that took effect in that country in 2001, the Japanese EPA and
Ministry of International Trade and Industry (MITI) suggested a voluntary phase-out of lead, with
lead levels reduced to half by 2000, and by two-thirds by 2005, along with increased end-of-life
(EOL) product recycling.
Electronics in the United States is a $400 billion-per-year industry facing significant
legislative and market pressures to phase out the use of lead-based solder and switch to lead-free
alternatives (CEA, 2003). Consumer demand for lead-free products also may increase as the
general public becomes more aware of lead issues, for example, as a result of EPA's successful
efforts to eliminate lead in gasoline, paint, and dust/soil. All these forces combine to drive the
U.S. electronics market inexorably toward lead-free solders.
1.3.5 Target Audience and Use of the Study
The electronics industry is expected to be one of the primary users of the LFSP study
results. The project aims to provide the industry with an objective analysis of the life-cycle
environmental impacts of selected lead-free solders. Scientific verification of these relative
impacts will allow industry to consider environmental concerns along with traditionally evaluated
parameters of cost and performance, and to potentially redirect efforts towards products and
processes that reduce solder's environmental footprint, including energy consumption, releases of
toxic chemicals, and risks to health and the environment. Based on the study results, the industry
can perform an improvement assessment of solder alternatives.
This study was designed to provide the electronics industry with information needed to
identify impacts throughout the life-cycle of various solder alternatives. This can lead to
improving the environmental attributes of solders. The LFSP study also allows the electronics
industry to make environmentally informed choices about solder alternatives when assessing and
implementing improvements such as changes in product, process, and activity design; raw
material use; industrial processing; consumer use; and waste management.
Identification of impacts from the life-cycle of lead-free solders also can encourage
industry to implement pollution prevention options such as development and demonstration
projects, and to foster technical assistance and training. The electronics industry can use the tools
and data provided by this study to evaluate the health, environmental, and energy implications of
the solder alternatives. Using this evaluation, the U.S. electronics industry may be better
prepared to meet the growing demand for extended product responsibility; to help guide public
policy towards informed, scientifically-based solutions that are environmentally preferable; and to
be better able to meet the competitive challenges of the world market. In addition, the LCA
model and results presented by this study provide a baseline upon which solder alternatives not
1-4
-------
included in the study can be evaluated. This will allow for further, expedited LCA studies, whose
growing popularity within the industry puts them in demand by original equipment manufacturers
(OEMs) and international organizations.
The information generated in this study also can be used by the electronics industry to
select the lead-free solders that work well for a given application and that pose the fewest impacts
to public health and the environment over their entire life cycles. The study results also should
help governmental organizations to better manage their electronics purchasing and EOL
disposition activities, and to inform the activities of community action groups.
1.4 SUMMARY OF LIFE-CYCLE ASSESSMENT METHODOLOGY
As defined by the Society of Environmental Toxicology and Chemistry (SET AC), the four
major components of an LCA are:
(1) goal definition and scoping;
(2) inventory analysis;
(3) impact assessment; and
(4) improvement assessment.
More recently, ISO 14040: Environmental Management—Lifecycle
Assessment—Principles and Framework, has defined the four major components of an LCA as:
(1) goal and scope;
(2) inventory analysis;
(3) impact assessment; and
(4) interpretation of results.
The SETAC and International Standards Organization (ISO) LCA frameworks are essentially
synonymous with respect to the first three components, but differ somewhat with respect to the
fourth component, "improvement assessment" vs. "interpretation of results." "Improvement
assessment" is the systematic evaluation of opportunities for reducing the environmental impacts
of a product, process, or activity. "Interpretation of results" is the phase of an LCA in which the
findings from the inventory analysis and the impact assessment are combined together, consistent
with the defined goal and scope, in order to reach conclusions and recommendations. Under
either definition, this fourth component of the LFSP LCA remains for the project partners and is
not addressed in this report. The first three components of the LCA (which are essentially the
same for both the SETAC and ISO standards) for lead-based and lead-free solders are detailed in
separate chapters of this report.
The goals and scope of the lead-based and lead-free solder LCA, introduced in
Section 1.3, are the overall subject of Chapter 1 of the report. The second component, inventory
analysis, involves the quantification of raw material and fuel inputs, and solid, liquid, and gaseous
emissions and effluents. The approach to the LCI in this study involved defining product
materials (e.g., solders), developing a bill of materials (BOM) of the products, and collecting
inventory data for each process within each life-cycle stage. Details of the LCI data-gathering
activities are provided in Chapter 2.
The third component of the LCA, LCIA, involves the translation of the environmental
1-5
-------
burdens identified in the LCI into environmental impacts. LCIA is typically a quantitative
process involving characterization of burdens and assessment of their effects on human and
ecological health, as well as other effects such as smog formation and global warming. This
project uses an LCIA methodology that incorporates more detailed health effects compared to
many other typical LCIA methods, to more fully reflect the concerns of policy makers, public
interest groups, and the electronics industry. Details of the LCIA methodology are presented in
Chapter 3.
From a general perspective, LCA evaluates the life-cycle environmental impacts from
each of the following major life-cycle stages:
raw materials extraction/acquisition;
materials processing;
product manufacture;
product use; and
• final disposition/end-of-life.
Figure 1-1 briefly describes each of these stages for a solder product system. The resource flows
(e.g., materials and energy inputs) and the emissions, waste, and product flows (e.g., outputs)
within each life-cycle stage, as well as the interaction between each stage (e.g., transportation) are
evaluated to determine the environmental impacts.
1-6
-------
INPUTS
LIFE-CYCLE STAGES
OUTPUTS
Materials
Energy
Resources
RAW MATERIALS EXTRACTION/ACQUISITION (UPSTREAM)
Activities related to the acquisition of natural resources, including
mining non-renewable material, harvesting biomass, and
transporting raw materials to processing facilities.
MATERIALS PROCESSING (UPSTREAM)
Processing natural resources by reaction, separation, purification, and
alteration steps in preparation for the manufacturing stage; and
transporting processed materials to product manufacturing facilities.
PRODUCT MANUFACTURE
Processing materials into solder and solder alternatives.
USE/APPLICATION
Application of the solders as the solders are used in manufacturing
various products (e.g., printed wiring board and component
manufacturing processes).
END-OF-LIFE
At the end of its useful life, the solders, which are part of another product,
as produced in the use stage, are retired. If reuse and recycle of
the solder is feasible, the product can be transported to an appropriate
facility and disassembled or demanufactured for materials recovery.
Materials that are not recoverable are then transported to appropriate
facilities and treated (if required or necessary) and/or disposed of.
Emissions
Wastes
•Products
Product System Boundary
Figure 1-1. Life-Cycle Stages of Solder Alternatives
1.5 PRODUCT SYSTEMS
The following sections describe the product systems that are the subject of the LFSP LCA
and how the solder alternatives are compared for the purposes of the study.
1.5.1 Solder Alternatives
The solders investigated in this study are listed in Table 1-1. Solders were selected for
evaluation by the project participants based on such factors as current trends and performance
studies (see Section 1.3.2). Tin-lead (SnPb) solder was selected as the baseline solder for the
evaluation. Tin-copper (SnCu) was selected because it is currently being used by segments of the
industry as a low-cost substitute for SnPb in wave solder applications. Tin-silver-copper (SAC)
was selected because of its ability to function in both a bar solder and paste environment, and
because it appears—through testing— to be emerging as a top choice for a possible substitute for
SnPb (NEMI, 2002). Finally, the evaluation group includes two bismuth-containing solders to
assess their environmental impacts, particularly at the EOL, because they are currently being
considered by several project partners as viable replacements for lead-based solder.
The solders selected represent both paste and bar solders. Paste solders are used for
attaching surface mount components to the surface of printed wiring boards (PWBs). In general,
where circuitry is sufficiently complex or the size of the assembly is an important design criteria,
1-7
-------
most high-end applications and much of the consumer market electronics require assembly with
paste solders. Conversely, low-complexity electronics applications (e.g., many toys) often use
single-sided or double-sided PWBs and lower cost through-hole components. These lower-end,
low-cost applications are often assembled using bar solders, which are simpler to apply and less
costly to produce.
Table 1-1. Solders selected for evaluation
Solder alloys
Tin-Lead (SnPb) (baseline)
Tin-Copper (SnCu)
Tin-Silver-Copper (SAC)
Bismuth-Tin-Silver (BSA)
Tin-Silver-Bismuth-Copper
(SABC)
Composition
63Sn/37Pb
99.2 Sn/ 0.8 Cu
95. 5 Sn 73.9 Ag/0.6Cu
57Bi/42Sn/1.0Ag/
96 Sn 72.5 Ag/
1.0Bi/0.5Cu
Density (g/cc)
8.4
7.3
7.35
8.56
7.38
Melting
Point (°c)
183
227
218
138
215
Solder type
Paste and Bar
Bar
Paste and Bar
Paste
Paste
Ag=Silver; Bi=Bismuth; Cu=Copper; Pb=Lead; Sn=Tin
1.5.2 Functional Unit
The product systems being evaluated in this project are either lead-based or lead-free
solders currently in use within the electronics industry. In an LCA, product systems are evaluated
on a functionally equivalent basis. The functional unit normalizes data based on equivalent use to
provide a reference for relating process inputs and outputs to the inventory and impact assessment
across alternatives.
For this project, the functional unit is a unit volume of solder required to form a viable
surface mount or through-hole connection between the PWB and the component (Figure 1-2).
The functional unit is based on the understanding that a similar volume of solder is required to fill
the space in a solder joint regardless of the type of solder used. The selection of this functional
unit is independent of PWB design or configuration, since the number and types of connections
formed by the solder would be the same for each alternative. As a result, a volume of one
thousand cubic centimeters (cc) of solder was selected for use as the functional unit in the LCA.
Figure 1-2. Typical Solder Joints for Both Through-hole and
Surface Mount Connections
-------
1.6 ASSESSMENT BOUNDARIES
The following sections explain more about the data categories, the physical and
geographic limitations, and general exclusions to the LFSP LCA, all of which combine to
determine the project boundaries.
1.6.1 Life-Cycle Stages and Unit Processes
As noted above, in a comprehensive cradle-to-grave analysis such as this LCA, the
product system includes five life-cycle stages:
(1) raw materials acquisition;
(2) materials processing;
(3) product manufacture;
(4) product use/application; and
(5) final disposition/EOL.
Also included are the activities that are required to affect movement between the stages (e.g.,
transportation). The major processes within the life cycles of the solders that were modeled in
this study are depicted in Figure 1-3. Each process box represents a unit process that has its own
inventory of inputs and outputs.
Because of process differences during the product use/application stage between paste and
bar solders, the two groups of solders could not be evaluated together within a single LCA. As
such, separate LCAs were conducted comparing each group of solders identified in Table 1-1.
LCA results for both paste and bar solder are presented separately within each section of this
report.
1.6.2 Spatial and Temporal Boundaries
Geographic boundaries are used in a LCA to show where impacts are likely to occur for
each life-cycle stage. This is important for assessing the impact of such activities as
transportation of materials between life-cycle stages. For example, acquisition and processing of
materials used in the manufacture of the metals comprising the solder alloys is done throughout
the world and is represented by a worldwide database. Product manufacturing also occurs
worldwide with data being collected from U.S. and Japanese sources. Similarly, solder
application in the use stage is done worldwide; however, given the geographic location of the
project researchers, data were only collected from manufacturers in the U.S. The EOL evaluation
focuses on solders and electronic products containing solder that reach the end of their lives in the
U.S. Due to limited availability of U.S. EOL data (e.g., on recycling), however, EOL data from
other countries also were used. For purposes of this study, the geographic boundaries for all life-
cycle stages are worldwide; however, several stages are primarily represented by data collected in
the U.S.
1-9
-------
INPUTS:
Materials
Energy
Resources
ft
ft
ft
Materials extraction
Extraction of each
metal in alloy: Pb, Sn,
Cu, Ag, Bi
Extraction of other
materials used in
product or in
manufacturing the
product*
Materials processing
Processing of each
metal in alloy: Pb, Sn,
Cu, Ag, Bi
Processing of other
materials used in
product or in
manufacturing the
product*
Product manufacturel
Manufacturing of each
solder
Use/Application
Wave application of
each bar solder (as
applicable) to a generic
board
Reflow application of
each paste solder (as
applicable) to a generic
board
End-of-Life
Recycling
Landfilling
Incineration
Unregulated disposal
ft
ft
ft
ft
OUTPUTS:
Products
Emissions
Effluents
Wastes
* Additional materials will be included if they meet project decision rules.
Figure 1-3. Solder Life-Cycle Conceptual Model
-------
1.6.3 General Exclusions
A number of items have been excluded from the LCA. General exclusions of processes or data
are as follows:
• impacts associated with the infrastructure needed to support manufacturing facilities (e.g.,
general plant maintenance);
• use of the final product in which a PWB is installed where no flows of or exposure to
solder metals are likely to occur (e.g., use of a personal computer; however, energy and
solder flows from EOL recycling or disposal of that final product is included in the LCA's
scope);
• lead or other solder metals used in non-solder parts of a PWB (e.g., on the surface finish);
and
• transportation between life-cycle stages (due to the large diversity of materials sources
and intended markets).
1-11
-------
REFERENCES
Bernier, D. 2002. Personal communication between Dennis Bernier of Kester and Jerry Gleason
of Hewlett-Packard, March 25.
CEA (Consumer Electronics Association). 2003. Annual Electronic Market Data Book 2003.
Consumer Electronics Association.
Curran, M.A. 1996. Environmental Life-Cycle Assessment. McGraw-Hill, New York, NY.
DHHS (U.S. Department of Health and Human Services). 1999. Toxicological Profile for Lead.
Public Health Service, Agency for Toxic Substances and Disease Registry, prepared by Research
Triangle Institute, July.
EPA (U.S. Environmental Protection Agency). 2002. Summary of Lead-Free Solder
Performance Based on Existing Data Provided by the Electronics Industry. Design for the
Environment Program, Office of Pollution Prevention and Toxics. Washington, DC.
Available at: http://eerc.ra.utk.edu/ccpct/lfsp-docs.html
Fava, J., R. Denison, R. Jones, B. Curran, M. Vigon, B. Selke, S. & J.A. Barnum. 1991.
Technical Framework for Life-Cycle Assessment. Society of Environmental Toxicology and
Chemistry and SETAC Foundation for Environmental Education, Inc. Washington, DC.
ISO (International Standards Organization). 1996. ISO 14040, Environmental Management -
Life-Cycle Assessment Principles and Framework. TC 2071 SC 5N 77. International Standards
Organization, Paris.
NEMI (National Electronics Manufacturing Initiative). 2002. Press Release: "NEMFs Lead-Free
Assembly Project Reports Latest Results at APEX 2002." January 21. Available at:
http://www.nemi.org/Newsroom/PR/PR012102b.html, downloaded March 22, 2002.
SETAC (Society of Environmental Toxicology and Chemistry). 1994. Life-Cycle Assessment
Data Quality: A Conceptual Framework. SETAC and SETAC Foundation for Environmental
Education, Inc. Washington, DC.
Van der Wei, H. 2002. E-mail communication between Jack Geibig, of UT, and Has Van der
Wei, of Phillips CFT, May 21.
Warburg, N. 2003. "Life Cycle Analysis of Lead-free Solders." Presentation at IPC Printed
Circuits Expo 2003. Anaheim, California. April 1.
1-12
-------
Chapter 2
LIFE-CYCLE INVENTORY
A LCI is the identification and quantification of the material, resource, emission, waste,
and product flows from the unit processes in the life-cycle of a product system (Figure 2-1). For
the DfE LFSP, LCI inputs (a.k.a. resource flows) include materials used in the solders
themselves, ancillary materials used in processing and manufacturing of the solders, and energy
and other resources consumed in the manufacturing, use (application), or final disposition of the
solders. LCI process output flows include primary and co-products, as well as releases to air,
water, and land. A conceptual model of the specific unit processes for solders was represented
previously by the boxes in Figure 1-3. Each unit process has flows particular to that process.
Figures 2-2 through 2-5 show each unit process for the life-cycles of the paste solders, and
Figures 2-6 through 2-8 show those for the bar solders. The figures graphically display how
processes in the product life-cycle are linked to one another and what processes are evaluated
within the scope of this LCA.
Solder materials
Ancillary material!
Energy/resources
Unit Process
Emissions (air and water)
Wastes (solid, hazardous,
and radioactive)
Primary products
Co-products
Figure 2-1. Unit process inventory conceptual diagram
Chapter 2 describes the approach taken for collecting and evaluating LCI data in the
LFSP and summarizes the LCI results. Section 2.1 describes the general methodology for LCI
data collection. Sections 2.2 through 2.5 present the specific methodologies, data sources, data
quality, limitations and uncertainties for each life-cycle stage. Section 2.6 summarizes the
baseline LCI data results for the paste and bar solder categories.
2-1
-------
Electronic
Product Use-
(Not modeled)
Unregulated
Recycling and
Disposal
Bold= Primary Data
Dash= Not modeled
Figure 2-2. SnPb Paste Solder Life-Cycle Processes
Unregulated
Recycling and
Disposal
r 1 ^ (Not Modeled)
r* ^
Bold= Primary Data
Dash= Not modeled
Figure 2-3. SAC Paste Solder Life-Cycle Processes
2-2
-------
Natural Gas
Fuel Oil-Hvy
Unregulated
Recycling and
Disposal
I
Electronic
Product Use - ]
(Not Modeled)
Solder Landfilling
Bold= Primary Data
Dash= Not modeled
Figure 2-4. BSA Paste Solder Life-Cycle Processes
Electronic
Product Use -
(Not Modeled)
Unregulated
Recycling and
Disposal
Bold= Primary Data
Dash= Not modeled
Figure 2-5. SABC Paste Solder Life-Cycle Processes
-------
Bold= Primary Data
Dash= Not modeled
Figure 2-6. SnPb Bar Solder Life-Cycle Processes
Bold= Primary Data
Dash= Not modeled
Figure 2-7. SAC Bar Solder Life-Cycle Processes
2-4
-------
Fuel Oil
-Hvy
Fuel Oil
-Lt
Electronic
Product Use -
(Not Modeled)
Unregulated
Recycling and
Disposal
Application
Fuel Oil- Hvy '
U.S. Elect
LPG Prod
fc.
\
, v
A.
**
SnCu
V
Post-consumer Cu
Smelting (SnCu)
Fuel Oil- Lt
Bold= Primary Data
Dash= Not modeled
Figure 2-8. SnCu Bar Solder Life-Cycle Processes
2.1 GENERAL METHODOLOGY
This section describes the data categories evaluated in the LFSP LCI, decision rules used
to determine which materials to evaluate in the study, and data collection methods. It also
describes procedures for allocating inputs and outputs from a process to the product of interest
(i.e., a solder) when the process is used in the manufacture, recycle, or disposal of more than one
product type at the same facility. Finally, it describes the data management and analysis
software used for the project, and methods for maintaining overall data quality and critical
review.
2.1.1 Data Categories
Table 2-1 describes the data categories for which inventory data were collected,
including material and resource flows (inputs), and emission, waste, and product flows (outputs).
In general, inventory data were normalized to either (1) the mass of an input or output per
functional unit, or (2) energy input (i.e., megajoules, [MJ]) per functional unit. As discussed in
Chapter 1 (see Section 1.5.2), the functional unit is a unit volume of a particular solder equal to
1,000 cubic centimeters (cc) of solder. Solder density was used to convert the normalized data
from mass to volume (see Table 1-1 for solder densities).
Data that reflect production for one year of continuous processes were scaled to distribute
over time the excessive material or energy consumption associated with startups, shutdowns, and
2-5
-------
changeovers. Consequently, any modeling associated with the impact assessment reflects
continuous emissions when equilibrium concentrations may be assumed. If data were reported
over a period of less than one year for any inventory item, the analysis was adjusted as
appropriate to the functional unit. Data were collected on the final disposition of emissions and
waste flows, such as whether these flows are recycled, treated, and disposed. This information
was used to determine which impacts will be calculated for each particular inventory item.
Methods for calculating impacts are discussed in Chapter 3, Life-Cycle Impact Assessment.
Table 2-1. LCI data categories
Data category
Description
Material and resource flows (inputs)
Material flows
(kilograms [kg]
per functional
unit)
Energy flows (MJ
per functional
unit)
Actual materials that make up the final product for a particular process (primary materials)
and materials that are used in the processing of a product for a particular process. Process
materials from solder application could include, for example, fluxes. Materials may be non-
renewable (i.e., materials extracted from the ground that are non-renewable or stock
resources such as coal), renewable, or flow resources such as water and limestone.
Process energy and pre-combustion energy (i.e., energy expended to extract, process, refine,
and deliver a usable fuel for combustion) consumed by any process in the life-cycle. The
energy flows modeled in this analysis are generally from non-renewable sources.
Emissions, wastes, and product flows (outputs)
Emissions to air
(kg per functional
unit)
Emissions to
water (kg per
functional unit)
Emissions to soil
(kg per functional
unit)
Deposited goods
(kg per functional
unit)
Primary products
(kg of material or
number of
components per
functional unit)
Co-products (kg
per functional
unit)
Mass of a product or material that is considered a pollutant within each life-cycle stage. Air
outputs represent actual or modeled gaseous or paniculate releases to the environment from
a point or diffuse source, after passing through emission control devices, if applicable.
Mass of a product or material that is considered a pollutant within each life-cycle stage.
Water outputs represent actual or modeled discharges to either surface or groundwater from
point or diffuse sources, after passing through any water treatment devices.
Mass of chemical constituents that are considered pollutants and emitted to soil within each
life-cycle stage. Soil emissions represent actual or modeled discharges to soil from point or
diffuse sources.
Mass of a product or material that is deposited as solid or hazardous waste in a landfill or
deep well. Represents actual disposal of either solids or liquids that are deposited either
before or after treatment (e.g., incineration, composting), recovery, or recycling processes.
Material or component outputs from a process that are received as input by a subsequent
unit process within the solder life-cycle.
Material outputs from a process that can be used, either with or without further processing,
that are not used as part of the final functional unit product.
2-6
-------
2.1.2 Decision Rules
Given the enormous amount of data involved in inventorying all of the flows for a
product system, decision rules are typically employed to make the data collection manageable
and representative of the product system and its impacts. Decision rules are a set of criteria
established by project participants used to determine if a given process or material flow is to be
evaluated in the LCA.
In this project, decision rules as to which processes within the materials extraction and
processing (i.e., "upstream") stage to include are based on the materials used to manufacture
solders. In considering upstream materials, a combination of several factors, including
availability of existing data, plus manufacturer's willingness to participate, were considered;
including all of the upstream processes in the scope of the project can unnecessarily lengthen the
project period and expend project resources on materials that are unlikely to be influential to the
impact results. For example, while it is beneficial to include in the LCA scope the manufacture
of the solder flux, it is not necessarily practical to include the manufacturing processes for each
of the chemicals that comprise the flux material.
To help determine which upstream processes to include in the LFSP LCI, first the bill of
materials of the primary solder materials (Table 1-1 in Section 1.5.1) was reviewed. Note that
Table 1-1 does not include non-metallic components of solders, such as flux, which are
considered to be ancillary materials. Because of the limited number of metals in the solders,
each metal was included in the upstream inventories. Secondary inventory data exist for lead,
tin, copper, and silver. As bismuth is a co-product of lead and copper mining, an inventory of
materials associated with the extraction and processing of bismuth was developed from the lead
and copper inventories. Material inventories for flux components were assessed once the flux
formulations were obtained. Inclusion of the fluxes and other ancillary materials, as well as
energy sources (e.g., fuels or electric power) associated with manufacturing the solders or
applying the solders were determined based on decision rules.
The decision rule process begins by assessing the additional materials used in the various
processes within the life-cycle of the solders for the following attributes:
1. The quantity contribution of each material or energy source. Materials or energy sources
used in large quantities have the potential for even more materials and resources to be
associated with their manufacture, and thus have a higher potential for having a
significant environmental impact.
2. Materials that are of known or suspected environmental significance (i.e., toxic). In an
environmental life-cycle assessment, consideration of materials or components that are
known to or are suspected to exhibit an environmental hazard are to be included to the
extent feasible.
3. Materials that are known or suspected to have a large energy contribution to the systems
energy requirements. Significant environmental impacts are associated with the
production of energy, therefore, priorities will be given to include materials or processes
that are known to or suspected to consume large amounts of energy.
2-7
-------
4. Materials that are physically unique to one solder over another. The physical
uniqueness of a material or component has the potential to accentuate the environmental
differences among solders and, thus, are included in the study, where possible.
5. Materials that are functionally significant to the solder. "Functionally significant" is
defined as important to the technically successful use of the solder as it functions to allow
the successful operation of a PWB. For example, each base metal is considered
functionally significant.
In general, materials or energy sources that are greater than one percent of the total mass
or energy required to manufacture the solder were included in the scope. Materials comprising
between one and five percent, however, also were evaluated for whether or not upstream
inventories were required. Inclusion of materials falling into the one to five percent range were
then based on the other decision rule criteria, as well as availability of data. Materials of known
or suspected environmental or energy significance were included, regardless of their mass
contribution. Additionally, materials that were physically unique or functionally significant to a
solder alternative were included if they would have been otherwise eliminated based on the mass
cutoff. For example, copper production for SAC, SABC, and SnCu was included as an upstream
process although it is less than one percent of the mass of the alloys, due to its technological
importance and physical uniqueness (i.e., it is not found in the SnPb baseline alloy).
2.1.3 Data Collection and Data Sources
Data were collected from both primary and secondary sources. Primary data are directly
accessible, plant-specific, measured, modeled, or estimated data generated for the particular
project at hand. Secondary data are from literature sources or other LCAs, but are specific to
either a product, material, or process used in the manufacture of the product of interest.
Table 2-2 lists the types of data (primary or secondary) employed for each life-cycle stage in the
LFSP LCI. Where both primary and secondary data were lacking, modeled data or assumptions
served as defaults.
2-8
-------
Table 2-2. Data types by life-cycle stage
Life-cycle stage
Upstream
(materials extraction and processing)
Product manufacturing
Use (solder application)
Final disposition
(teachability, recycling, and/or disposal)
Data types
Secondary data
Primary data
Primary data
Primary and secondary data.
Modeling for some processes
Scope
Less emphasis
Greater emphasis
Greater emphasis
Greater emphasis
2.1.4 Allocation Procedures
An allocation procedure is required when a process within a system shares a common
management structure, or where multiple products or co-products are produced. In the LFSP
LCI, allocation procedures were required when processes or services associated with the
functional unit were used in more than one product line at the same facility. Flows are allocated
among the product lines to avoid over-estimating the environmental burdens associated with the
product under evaluation. For example, energy consumption data collected during the
manufacture of solder must first be allocated by the quantity of each of the solder alloys
produced, in order to accurately determine the energy consumption attributable to the
manufacture of the solder in question.
The International Standards Organization (ISO, 1996) recommends that wherever
possible, allocation should be avoided or minimized. This may be achieved by sub-dividing the
unit process into two or more sub-processes, some of which can be excluded from the system
under study. In the example above, if a manufacturer uses only one type of solder, no allocation
would be necessary from that manufacturer. It is more likely that the manufacturer would
produce multiple solders, however. This requires allocation of flows from the manufacturing
using several solders to those associated only with the one solder alloy of interest. As suggested
by ISO, if sub-processes within the facility can be identified that distinguish between solders
used during application, the sub-processes using the solders that are not of interest can be
eliminated from the analysis, thus reducing allocation procedures.
In this study allocation procedures were used as follows:
• Inventory data for utilities and services common to several processes are allocated to
reflect the relative use of the service. For example, fuel inputs and emission outputs from
electric utility generation are allocated to a solder according to the actual or estimated
electricity consumed during the applicable process.
•• Where a unit process produces co-products, the burdens associated with the unit process
are allocated to the co-product on a mass or volume basis, as appropriate.
2-9
-------
2.1.5 Data Management and Analysis Software
Data collected for the study were either obtained from site visits, telephone interviews, or
electronic mail correspondences using a standardized data collection form developed for this
project; from existing databases; or other secondary data collected by the UT Center for Clean
Products and Clean Technologies. All these data were normalized to the study functional unit
and then imported into GaBi3, a commercially available life-cycle assessment software program
(GaBi, 2000). The GaBi3 software tool stores and organizes life-cycle inventory data and
calculates life-cycle impacts for a product profile. It is designed to allow flexibility in
conducting life-cycle design and life-cycle assessment functions, and to provide the means to
organize inventory data, investigate alternative scenarios, evaluate impacts, and assess data
quality.
2.1.6 Data Quality
LCI data quality can be evaluated based on the following data quality indicators (DQIs):
(1) the source type (i.e., primary or secondary data sources); (2) the method in which the data
were obtained (e.g., measured, calculated, estimated); and (3) the time period for which the data
are representative. LCI DQIs are discussed further in Life-Cycle Assessment Data Quality: A
Conceptual Framework (SETAC, 1994). LFSP data quality for each life-cycle stage is discussed
in detail in Sections 2.2 through 2.5, and summarized below.
For the primary data collected in this project, participating companies reported the
method in which their data were obtained and the time period for which the data are
representative. Data from 2002 and 2003 were sought. The time period of secondary data and
method in which the data were originally obtained were recorded, where available.
Anomalies and missing data are common hurdles in any data collection exercise.
Anomalies are extreme values within a given data set. Any anomaly identified during the course
of this project that was relevant to project results was highlighted for the project team and
investigated to determine its source (i.e., mis-reported values). If an anomaly could be traced to
an event inherently related to the process, it was left in the data set. If, however, an anomaly
could not be accounted for, it was removed from the data set.
Missing data were replaced hierarchically. That is, if specific primary data were missing,
secondary data were used. Where neither primary nor secondary data were available,
assumptions were made. When assumptions or choices in data drove results, modified scenarios
were applied to the analysis to help understand the sensitivity of the results to those assumptions
or data. In the case where no data were found, or reasonable assumptions could not be made,
these deficiencies are reported.
Any proprietary information required for the assessment was subject to confidentiality
agreements between the Center for Clean Products and Clean Technologies and the participating
company. Proprietary data are presented as aggregated data to avoid revealing the source of the
data, or not reported at all if data aggregation is insufficient to protect the confidentiality of the
data.
2-10
-------
2.1.7 Critical Review
Critical review is a technique to verify whether an LCA has met the requirements of the
study for methodology, data, and reporting, as defined in the goal definition and scoping phase.
A critical review process was maintained in the LFSP LCA to help ensure that the following
criteria were met:
•• the methods used to carry out assessments were consistent with the EPA, SET AC, and
ISO assessment guidelines;
•• the methods used to carry out assessments were scientifically and technically valid within
the LCA framework;
•• the data used were appropriate and reasonable in relation to the goals of the study;
•• any interpretations reflect the limitations identified, and the goals of the study; and
•• the study results were transparent and consistent.
A project Core Group and Technical Work Group were identified, consisting of
representatives from industry, academia, public interest groups, and EPA. Both groups provided
critical reviews of the project assessments. The Core Group served as the project steering
committee and was responsible for approving all major scoping assumptions and decisions. The
Technical Work Group (which also includes the members of the Core Group) provided technical
guidance and reviews of major project deliverables including the LCA report. In addition to the
critical review process, primary data were double-checked with the original source to ensure
accuracy.
2-11
-------
2.2 MATERIALS EXTRACTION AND MATERIALS PROCESSING (UPSTREAM
LIFE-CYCLE STAGES)
This section describes the extraction and processing inventories for the major primary
materials (i.e., base metals) used in each solder alternative, as well as processes associated with
fuel production and the generation of electricity. The fuel and electricity generation inventories
are linked to processes in other life-cycle stages where fuels or electricity are used as process
inputs. In the presentation of inventory results (Section 2.6) and impact results (Chapter 3), the
fuel and electricity processes are presented as part of the life-cycle stage to which they are
linked. Section 2.2.1 provides the methodologies, including discussions on data sources and data
quality, for the major materials (Section 2.2.1.1), and for the major fuels and power sources
(Section 2.2.1.2). Section 2.2.2 presents the limitations and uncertainties applicable to both
materials and fuel/power sources.
2.2.1 Methodology
2.2.1.1 Materials (metals)
The major primary materials being evaluated in the upstream life-cycle stage are the base
metals in each solder alternative. These metals include lead, tin, copper, silver, and bismuth.
Both the extraction and processing of these metals are included in the scope of this analysis. For
each metal, this LCA combines the extraction and processing (e.g., smelting), along with
associated transportation of each metal into one process inventory.
The inventories for lead, tin, copper and silver were available as secondary data. Where
multiple data sets were available, data were selected based on reported data quality, timeliness of
data, and consistency with other data sets used in the LFSP analyses. The lead, copper and silver
inventories used for the LFSP were contained within GaBi3 software and databases (GaBi,
2000). The tin inventory was obtained from Ecobilan in their Database for Environmental
Analysis and Management (Ecobilan, 1999).
No secondary data sets were available for bismuth. Worldwide, bismuth is primarily co-
mined with other metals, including lead (35 percent), copper (35 percent), tungsten (15-20
percent, from China), and tin and other miscellaneous metals (10 to 15 percent) (Palmieri, 2002).
As lead and copper co-mining consist of the majority (70 percent) of the worldwide bismuth
supply, and because inventories for tungsten and the other metals were not readily available, the
bismuth mining and processing inventory was developed from the inventories for lead and
copper mining and processing, assuming they represent 100 percent of bismuth production.
Thus, the resulting bismuth inventory assumes that 50 percent of bismuth is co-mined with lead
and 50 percent with copper. In addition, research showed that the ratio of lead to bismuth
production is approximately 14:1 (Miller, 2002). Lacking additional information for copper,
2-12
-------
both the lead and copper inventories were thus scaled by a 14:1 ratio to represent the bismuth
inventory.1 The uncertainties in this approach are discussed in Section 2.2.2.
Solder manufacturers reported data on the origin of virgin metals purchased for solder
manufacturing. The data indicate that the majority of bismuth purchased for the manufacture of
solder is derived from lead and copper mining processes. Judgments on the applicability and the
level of confidence in the secondary mining and extraction data sets were based on the data
collected. Table 2-3 lists the processes included and the basic assumptions used to develop the
materials extraction and processing (ME&P) metals inventories.
The secondary inventories listed in the table include the primary production of the metals
(e.g., production from virgin sources) and are provided as material and energy flows per
kilogram of metal produced. In this analysis, these upstream inventories are linked to the
associated solder manufacturing processes and scaled in two ways: (1) to the mass of each metal
required as input to each solder manufacturing process, and (2) to the virgin content of each
metal used in manufacturing. The percentages of base metals that are of virgin origin were
estimated from primary data collected from five solder manufacturers and presented in Table 2-8
(see Section 2.3, Product Manufacturing). The mass of each metal input to the manufacturing
process was estimated assuming a process in full production. These estimates are predictions,
however, because most alternative solders are currently made only in batch processes to meet
customer demand, rather than in full production.
The remainder of the solder not manufactured from virgin materials is made from
recycled metal content. In this study, it is assumed that all the recycled content is from post-
industrial recycling (as opposed to post-consumer recycling). Post-industrial recycling, in some
form, is performed by most solder manufacturing facilities, and is included in this study as a
separate unit process in the solder manufacturing life-cycle stage. Thus, if a metal has a high
virgin content, more of the inventory will be represented in the upstream life-cycle stage than in
the manufacturing stage; while, alternatively, if a metal has a high recycled content, more of the
inventory will be represented in the manufacturing stage than in the upstream stage. While the
LFSP does not model post-consumer waste recycled directly back into the product, the process
of recycling solder from PWBs (via demanufacturing and copper smelting) is accounted for in
the EOL life-cycle stage.
1 Estimation of a 14:1 lead to bismuth ration is based on data from one mine. Additional research produced
an anecdotal, yet unconfirmed estimate that bismuth production might require ten times the materials as does lead
mining and processing (CEFIC et al., 2002). The more conservative 14:1 ratio was used, however, potentially
causing the results of this study to overestimate the impacts from bismuth production if the 10:1 ratio is indeed more
accurate.
2-13
-------
Table 2-3. Base metal inventories: summary of information from secondary data
Base metal
Inventory
Lead (GaBi)
Tin (DEAM)
Copper (GaBi)
Silver (GaBi)
Bismuth
(derived from
GaBi)a
Processes included
German-based primary lead production (99.995 percent lead), includes ore mining; ore
beneficiation; production of concentrate; sintering (with sulfuric acid); processing via
traditional shaft furnace (70 percent), QueneauSchuhmann-Lurgi (QSL) plants (20 percent),and
imperial smelting (10 percent); and refinery. Breakdown of shaft, QSL and imperial smelting
processes based on processing activities in Germany. Includes transportation and worldwide
mix of electric power generation.
Open mining of Casserite (SnO2), which is 55 percent tin. Otherwise, processes not specified.
German-based pyro-metallurgical primary copper production (from sulphidic ore); includes:
mining (mixture of opencast and underground mining in Chile, Canada, Russia and the U.S.),
beneficiation by flotation, transport, oxidation, and final electrolysis. Germany electric power
grid inventory applied to electricity use.
Global mix of data (including Canadian- and Swedish-based data). Primarily a by-product of
lead and copper (assumes 62.5 percent as a by-product from lead and 37.5 percent as a by-
product from copper; this is based on scaling up percentages of 50 percent as a by-product from
lead and 30 percent as a by-product of copper [GaBi, 2000]). Swedish silver production data
are based on the Ronnskar production facility in Sweden where copper, lead, zinc, gold and
silver are produced. The ores are mined in Laisvall (Zn, Pb), Litik (Cu) and Garpenberg
(Zn/Pb/Cu/Ag/Au). The non-ferrous metals are produced from metal ores, while the precious
metals are produced through recycling of secondary raw materials (i.e., scrap). Includes the
mining and smelting. The inventory for silver from the Swedish data is based on the allocation
of the market value of the pure silver produced from the overall production (from both mining
and smelting). The silver process is linked to (1) ore mining, which includes both opencast
mining and underground mining; (2) ore beneficiation, which involves extracting of valuable
minerals, removal of unwanted impurities, and separation of several valuable minerals, and (3)
sintering, which is a high temperature agglomeration process. The global mix silver production
data also combines primary lead production data from Canada, which includes mining,
concentrate production, sintering, and further processing at an acid plant, blast furnace and
refinery. Country -specific energy and transportation are included.
Primarily a by-product of lead and copper (assumes 50 percent as a by-product of lead and 50
percent as a by-product of copper and a 14: 1 ratio of lead or copper production to bismuth
production).
a The bismuth inventory developed for use in this analysis was derived from the GaBi inventories for lead and
copper.
Table 2-4 summarizes sources of secondary data and data quality information (e.g.,
original source of data, year of data, and geographic boundaries) for the metals ME&P
inventories used in this study.
2-14
-------
Table 2-4. Data sources and data quality for metals inventories
in the ME&P life-cycle stage
Materials
Lead
Tin
Copper
Silver
Bismuth
Year
f
01
data
1995a
1983-
1989b
1994a
1994a
See
lead
and
copper
above.
Geographic boundaries
Extraction
Germany
information
not readily
available
Chile,
Canada,
Russia,
United
States
Sweden,
Canada
See lead and
copper
above.
Processing
Germany
information
not readily
available
Germany
Sweden,
Canada
See lead and
copper
above.
Sources
GaBi, 2000 (which
cites Wiley-VHS,
1997)
Ecobilan, 1999
(which cites
IDEMAT, 1995;
which cites the
following primary
sources: Chapman
and Roberts, 1983
andU.S.BOM, 1989)
GaBi, 2000 (which
cites Wiley-VHS,
1997)
Silver mix data was
developed by GaBi
based on co-mining
with lead and copper.
The global mix silver
data is a combination
of two inventories:
silver production in
Sweden and lead
production in Canada
(GaBi, 2000).
Bismuth mix
developed by UT,
based on lead and
copper inventories in
GaBi3 (see above).
Data quality description
Average industry data. GaBi
states, "the data describe the
modeled process in a sufficient
quality" (GaBi, 2000).
IDEMAT rates both data
reliability and completeness as
average.
Average industry data. GaBi3
states that this is "a good
estimation for the production of
copper under consideration of the
described conditions" (GaBi,
2000)
For the Swedish mine silver
production data, GaBi3 states that
the data quality "...is quite
reasonable," although these data
are only representative of
conditions in Sweden, which only
contributes a low percentage of
the total world production. For
the Canadian lead production
process data quality is "relatively
good" as reported by GaBi3;
however, it should also be noted
that the lead-based data does not
include secondary raw materials
(scrap) and, thus, is considered a
worst case scenario for the lead
available on the market (GaBi,
2000).
See lead and copper above.
2-15
-------
a Reference year of data
b Date of publication of primary data source; however, reference year of actual data expected to be slightly earlier,
but actual year not known.
Sources: IDEMAT, 1995; Chapman and Roberts, 1983; U.S.BOM, 1989; GaBi, 2000; Ecobilan, 1999; and Wiley,
1997.
As shown in the table, the geographic boundaries of the data encompass mining
operations worldwide spanning four continents. In addition, the temporal boundaries of the data
range from 1983 to 1995. All of these factors create some inconsistencies among the data sets
and reduce the data quality when used for the purposes of the LFSP; however, this difficulty is
common with LCA, which typically uses data from secondary sources for upstream processes to
limit the scope and budget of an LCA.
2.2.1.2 Fuels and power sources
Fuels and electricity are used in various processes in each life-cycle stage, as depicted in
Figures 2-2 through 2-8. The inventories associated with the production of the major fuels and
electricity (i.e., contributing greater than one percent of total energy sources per the decision
rules outlined in Section 2.1.2) are included in the LCI of each solder. Flows from the
production of the fuels and electricity in the ME&P life-cycle stages are already incorporated
into the associated metals inventories provided from secondary data sources. In the other life-
cycle stages (i.e., solder manufacturing, application, and EOL), the production processes for
fuels and electricity are not incorporated into individual processes. Thus, separate processes
(i.e., inventories of the flows from the fuel production or electricity generation) are included in
the appropriate life-cycle stages.
The following inventories are included in the solder LCIs:
• • natural gas
light or distillate fuel oil (fuel oil #2)
• • heavy fuel oil (fuel oil #6)
•• liquified petroleum gas (LPG)
•• diesel fuel
•• electricity generation
Although the fuel and power inventories are presented in this section of the report under
"materials extraction and processing," they are presented in the inventory and impact results with
the life-cycle stage of the process that uses that fuel or power source. For example, during the
manufacture of solder, natural gas is used as a fuel; therefore, flows from the processing of
natural gas, which is needed to fuel solder manufacturing activities, are included in the
manufacturing stage LCI and impact results.
The fuel and power inventories were obtained from secondary data sources. The
inventories of natural gas, fuel oils, diesel fuel, and electric power were contained within the
GaBi3 databases. The LPG inventory was obtained from DEAM. The electric grid inventory
used in this study was obtained from the GaBi database and is based on a 1995 reference year.
This data set matched closely with the U.S. electric grid inventory developed by the UT in 1997
2-16
-------
(see Socolof et a/., 2001, Appendix E). Despite the fact that the UT data set was slightly more
recent, the GaBi data set was used for the evaluation because it required fewer project resources
to include in the analysis and the two data sets closely match. Table 2-5 describes the processes
included in the fuel and power inventories.
Table 2-5. Fuel and power inventories: summary of information from secondary data
Fuel Inventory
Natural gas (GaBi)
Light fuel oil (#2)
(GaBi)
Heavy fuel oil (#6)
(GaBi)
LPG (DEAM)
Diesel fuel (GaBi)
Electricity generation
(GaBi)
Processes included
Exploration, extraction, processing, and distribution (via pipeline or liquified natural gas
[LNG] tanker) to the end customer.
Crude oil extraction, pipeline and tanker transport, crude oil desalinization, atmospheric
distillation, desulphurization (i.e., medium distillates to hydrofiner), medium distillates
mix plant that produces light fuel oil.
Crude oil extraction, pipeline and tanker transport, crude oil desalinization, atmospheric
distillation, residue to fuel mix plant that produces heavy fuel oil.
Domestic and foreign crude oil production (onshore conventional, advanced recovery
and offshore conventional recovery), transport (fluvial, pipeline, rail, sea, and road) to
the refineries in the U.S., crude oil refining into LPG, and transport (pipeline and road)
from refinery to end user.
Crude oil extraction, pipeline and tanker transport, crude oil desalinization, atmospheric
distillation, desulphurization (i.e., medium distillates to hydrofiner), medium distillates
mix plant that produces diesel fuel.
Assumes a grid of 52.3 percent hard coal, 22.7 percent nuclear power, 12.4 percent
natural gas, 4.2 percent crude oil, 3.5 percent lignite, 3.4 percent hydro, and 1.5 percent
other.2
Table 2-6 summarizes data sources and data quality information for the fuel and power
source inventories used in this study. Like the metals inventories discussed previously, all of
the fuel and power inventories are secondary data for the purposes of the LFSP.
The GaBi data are based on a 1995 reference year. In comparison, the U.S. Energy Information Alliance
(EIA) reported in 1999 that the U.S. grid consisted of 57 percent coal (includes hard coal and lignite), 20 percent
nuclear, 11 percent hydro, 9 percent natural gas, 3 percent petroleum (crude oil), and 1 percent other.
2-17
-------
Table 2-6 Data sources and data quality for fuel and power inventories
used in various life-cycle stages
Materials
Natural gas
Light fuel
oil (#2)
Heavy fuel
oil (#6)
LPG
Diesel fuel
Electricity
generation
Year of
data
1995
1994
1994
References
range from
1983 to
1994
1994
1995
Geographic boundaries
Extraction
Canada,
Mexico,
United States,
Algeria
Unclear
(various
country-based
data sources
cited)
Unclear
(various
country-based
data sources
cited)
"Domestic and
foreign crude
oil
production"
Unclear
(various
country-based
data sources
cited)
Multiple
countries, fuel
dependent
Processing
United
States
Germany
Germany
United
States
refinery
operations
Germany
United
States
Sources
GaBi, 2000 (a)
GaBi, 2000 (b)
GaBi, 2000 (b)
Ecobilan, 1999
(c)
GaBi, 2000 (b)
GaBi, 2000 (d)
Data quality description
GaBiS states the data quality is:
"... good. The important flows
are considered. Natural gas
supply is representative."
GaBiS describes the data quality
as "good." It is average
industrial data from 1994.
GaBiS describes the data quality
as "good." It is average
industrial data from 1994.
No data quality description
provided by DEAM; data appear
complete.
GaBiS describes the data quality
as "good." It is average
industrial data from 1994.
GaBi describes the data quality
as "good." They claim to use
consistent statistics and a
comparable information basis
for every state.
(a) GaBi, 2000: Natural Gas Production (sources are from secondary literature, see References at the end of this chapter).
(b) GaBi, 2000: Refinery data (light fuel oil, heavy fuel oil, diesel fuel production) (sources are from secondary literature, see
References at the end of this chapter).
(c) Ecobilan, 1999: LPG production (sources are from secondary literature, see References at the end of this chapter).
(d) GaBi, 2000: U.S. electric power grid electricity generation (sources are from secondary literature, see References at the end
of this chapter).
As discussed in Section 1.6.2, the geographic boundaries of this project are worldwide
for most life-cycle stages, but most downstream processes using electricity were U.S.-based
data; therefore, the inventory associated with electricity generation is based on the U.S. electric
grid. Some of the other processes are represented by countries that might not be completely
2-18
-------
representative of operations applicable to this study; however, because the ME&P stage was
given lower priority in terms of expending resources for primary data, already available and
easily accessible data were often chosen.
2.2.2 Limitations and Uncertainties
The limitations and uncertainties associated with the ME&P stage inventories are
primarily due to the fact that these inventories were derived from secondary sources and are not
tailored to the specific goals and boundaries of the LFSP. Because the data are based on a
limited number of facilities and have different geographic and temporal boundaries they are not
necessarily representative of current industry practices in the geographic and temporal
boundaries, defined for the LFSP (see Section 1.6.2). These limitations and uncertainties are
common to LCA, which strives to evaluate the life-cycle environmental impacts of entire
product systems and is, therefore, limited by resource constraints which do not allow the
collection of original, measured data for every unit process within a product life-cycle.
Recognizing the limited resources available for this LCA, project partners elected to rely on
secondary data for the ME&P life-cycle stage to permit collection of primary data for other
solder life-cycle stages for which data had not been previously compiled.
The potential inconsistent inclusion of transportation data in ME&P inventory data for
some processes is another limitation. These data become particularly important when, for
example, raw materials are uncommon and must be transported long distances for processing or
when the particular transport mode used for a particular materials tends to have high
environmental impacts. The lack of transportation data for ME&P processes is not unique to the
secondary databases employed in this project or to the LFSP LCI, but a common limitation of
other LCIs as well.
Specific to the metals ME&P inventories, uncertainties are associated with the
methodology used for deriving the bismuth inventory from the lead and copper inventories as
well as limitations in the resulting data set which may not account for flows from ME&P of
bismuth when it is a co-product of other metals (e.g., tungsten, tin, and other miscellaneous
metals). The uncertainty in the ratio of flows from bismuth production to those of lead and
copper production could lead to an overestimate of bismuth impacts if a lower ratio (e.g., 10:1)
of bismuth to lead is more accurate than the 14:1 ratio. Similarly, the results may be either over-
or under-estimated should the bismuth to copper ratio be different from the 10:1 ratio assumed
for the study.
The percentages of base metals that are of virgin origin were estimated from primary data
collected from five solder manufacturers. For the alternative alloys, the estimates attempted to
predict operations in full production; however, these are indeed predictions and may not
represent what will actually occur in full production. The effects on the ME&P stage are caused
by the virgin content, which dictates how much mining and extraction is done to process the
virgin metal.
Specific to the electric grid inventory, uncertainties exist in the weighting values applied
to the various fuel sources from which the power is generated for the U.S. electric grid. The
factors were based on a reference year of 1995 and, thus, may vary given the volatility of the oil
supply and the current U.S. energy policy.
2-19
-------
2.3 PRODUCT MANUFACTURING
The solder product manufacturing life-cycle stage is made up of two distinct processes:
solder manufacturing and post-industrial solder recycling. This section describes the details of
the processes from which inventory data were collected for use in the LCA analyses of the
solders. It also details the methods used to collect and validate the data.
As noted in Section 1.5.1, the solders investigated in this study were selected by the
project participants based on a number of factors, including performance, likelihood of industry-
wide adoption, and prioritized interest of project stakeholders. Solder manufacturers and other
industry experts were consulted to accurately define the major manufacturing processes, in terms
of resources used and potential importance to environmental impacts. These processes were then
targeted and the collection of process data prioritized in our primary data collection effort.
Through consultation with our industry partners, and in collaboration with the Solder
Products Value Council of IPC, solder manufacturers were identified and approached about
supplying data on their individual solder manufacturing processes for both paste and bar solder.
2.3.1 Methodology
2.3.1.1 Data collection and allocation
Data were collected through site visits or through the distribution of data collection
forms. Site visits were performed at several solder manufacturing facilities to capture data that
reflect the varying methods of bar and paste solder manufacturing for each of the solder alloys
being evaluated. Altogether, four solder manufacturing facilities were visited representing three
solder manufacturers, one each in the countries of Mexico and Canada, and two in the U.S.
Data collection forms were developed by the UT research team and approved by the
Technical Work Group to most efficiently collect and organize inventory data needed for the
LCA. Appendix F provides a copy of the data collection form. Data forms were completed
during site visits by project researchers or directly by companies when site visits were not
possible. The data that were collected included brief process descriptions; primary and ancillary
material inputs; utility inputs (e.g., electricity, fuels, water); air, water and waste outputs; product
outputs; and associated transportation. Quantities of inputs and outputs provided by companies
were converted to mass per unit of product. Transport of materials to and products or wastes
from the manufacturing facility also were reported.
Site visits were conducted to observe and to collect inventory data for the post-industrial
recycling process. Post-industrial recycling is the common practice among solder manufacturers
of reclaiming base metal content from process wastes resulting from solder manufacture or from
solder wastes generated during the solder use/application process. Reclaimed alloys are
preferred to post-consumer recycled content as they only need to be refined to common alloy
mixtures (e.g., 60 Sn/40 Pb) rather than refined to a pure alloy. The refined alloy is then
modified to the desired alloy through further refinement and mixing. Altogether, data were
collected during site visits to three post-industrial solder recycling facilities located in Mexico,
2-20
-------
Canada, and the U.S. Together, the data collected represent a variety of processes operated under
a variety of conditions and environmental requirements.
During each site visit, UT staff completed a data collection form similar to those
completed by facilities that were not visited. Each site visit took approximately a full day, and
included an extensive tour of the processes, interviews with process personnel, and a period of
time spent completing and reviewing the data on the collection form for accuracy. Data were
either measured on the spot, obtained from previously measured or collected data by the facility,
or estimated with the assistance of process personnel with appropriate experience and process
knowledge. Data were collected, when possible, on a per mass of solder produced basis.
Calculations to convert the data for the LCA based on data collected during the site visits were
then verified through direct follow-up with the facility at a later date prior to use in the LCA.
Data collected from processes often had to be allocated to solder alloys based on the
functional unit defined for this project: a volume equal to 1,000 cubic centimeters of solder.
Since much of the process data collected was based on mass (i.e., per kg solder), these data were
converted to the functional unit using the solder density. In cases where data collected covered
the processing of two or more solder alloys (i.e., monthly energy consumption for a process
producing multiple solder alloys), data were allocated to the various solder alloys based on the
mass of solder produced, then converted to the functional unit using density. Other data were
allocated to the solders using appropriate conversions, where applicable.
Multiple data sets collected for a single process (i.e., energy consumed during SnPb
solder manufacture from five facilities) were aggregated before being used in the study. Data
were aggregated to generate a single value for each inventory item, and to protect the
confidentiality of individual data points.
2.3.1.2 Solder manufacturing
Solder manufacturing data were collected through a series of site visits to solder
manufacturing facilities or through the distribution of data collection forms. While the process
of manufacturing solder varied by facility, the overall process of manufacturing followed a
similar series of process steps for both bar and paste solder manufacturing. Figure 2-9 displays a
flow diagram for both bar and paste solder manufacturing. The diagram depicts the primary
process steps for which life-cycle inventory data were collected.
Bar solder manufacturing begins with the formation of the alloy from the base metals,
which occurs in a large smelting pot. Metals are added in a metallurgically defined sequence to
a gas-fired pot, melting the base alloy, and then adding each of the other alloys until the required
composition is achieved. The time required to smelt the metals is dependent on a number of
factors including temperature, number and type of metals, and the order in which the metals are
added to the alloy.
The smelting is followed by a refining step during which undesired metals are removed
from the alloy through the use of additives. Undesirable metals are precipitated, and then
removed from the alloy typically through skimming or decanting the contaminant from the
desired alloy. Finally, once the metal alloy has reached the desired purity and composition, the
metal is poured into a series of molds in a casting step to form the solder bar product.
2-21
-------
Bar Manufacturing
Paste Manufacturing
Bar Solder
Product
Paste Solder
Product
Figure 2-9. Solder Manufacturing Process Diagrams for Bar and Paste Solders
The early steps of solder paste manufacturing are similar to that of bar solder. Basic
solder alloys are often prepared in advance and cast into bars for use at a later time, frequently as
a feedstock for solder paste manufacture. Solder bars are re-melted in a smaller smelting pot and
then fed into a process to generate solder powder. Powder is manufactured in one of three ways,
by spinning disk, by ultrasonic dispersion, or by dispersion via air venturi. In all of these
methods, the molten solder is introduced into the top of a column or tower, and ultimately
dispersed into tiny particles, using one of the methods mentioned. The particles cool as they fall
through the column forming small spheres. The spheres are sifted through a series of screens
that ensure the size and spherical geometry of the powder. Out-of-specification solder spheres
(fifty percent or more by volume) are reintroduced into the small smelting pot and the process
repeated until the desired amount of solder powder is created.
Flux is blended in a separate process, combining chemicals in a formulation specific to
each type of solder alloy and application. Flux chemistry is tailored to provide a variety of
characteristics (e.g., no clean) to meet customer needs, and is considered quite confidential by
solder manufacturers. As such, data were obtained for only one no-clean flux formulation during
the project from a single manufacturer. This chemical formulation was used for all of the paste
solder types.
The flux carrier is finally blended with the solder powder to create the solder paste. SnPb
solder paste was considered to be a blend of ninety percent powder and ten percent flux when
allocating inventory data. The lead-free alloys of SAC, SABC, and BSA were considered
blended at eighty-nine percent solder and eleven percent flux, due to the differences in metal
density. Solder paste is then packaged into various forms such as syringe tubes, squeezable
tubes, or jars.
2-22
-------
Table 2-7 displays the number of individual data sets collected for solder manufacturing
by solder type for both paste and bar solders.
Table 2-7 Inventory data sets for paste and bar solder manufacturing
Solder type
SnPb
SAC
BSA
SABC
SnCu
Paste data sets
3
3
2
3
N/A
Bar data sets
4
3
N/A
N/A
3
N/A=Not applicable
Being the predominant solder technology for a number of decades, SnPb solder
manufacturing is a mature technology performed using full-scale production processes.
Although the methods for manufacturing the solder alternatives are similar to those for SnPb,
involving smelting and refining processes, these solders are only produced in small-scale, batch
operations. Data are not yet available for full-scale production, therefore, product manufacturing
inventories for the solder alternatives were scaled from batch production data or from SnPb
production data combined with factors to account for the different melting points of the solder
constituents. This is a limitation and uncertainty of this study, discussed further in Section 2.3.2.
2.3.1.3 Post-industrial recycling
Process wastes from the use/application process (i.e., solder dross from wave soldering)
are often returned to the solder manufacturer for reclamation and reuse. These wastes are
considered to be of high value because they seldom contain other hard to separate metals or
compounds, and are already in a composition that requires minimal effort to recycle into new
solder. Other similar materials, such as solder manufacturing wastes (e.g., out of spec solder
paste) and even high purity non-solder related wastes (e.g., lead-based wiring), are often
accepted as material for recycling, depending on the manufacturer and the capabilities of the
reclamation process.
Figure 2-10 presents a typical flow diagram for a post-industrial recycling process
operated by a solder manufacturer. The process depicted is representative of the processes from
which inventory data were collected, though process steps differ between facilities.
2-23
-------
Scrap Metal
Smelting
Refining
Secondary
Metal Casting
Cathode
Casting
Electro-
Refining
Further
refining or
separation
Bar or Paste
Manufacture
Figure 2-10 Typical Post-Industrial Recycling Process Flow Diagram
Scrap metal is first smelted to melt the alloy and to combust any organic content
contained within the scrap. The molten metal is poured into ingots, and tested to identify any
contaminants. The ingots are then sent to refining, where the metal is reheated in pots and the
undesirable metal content is separated through the use of additives. Metals are not refined back
to pure elements, but rather into combinations of metals similar to those required for future
solder manufacturing (e.g., SnPb). Desirable metal combinations are sent to casting where they
are cast into large ingots of secondary metals that are later used as a feedstock for the paste or
bar solder manufacturing process. Metals or combinations of metals that cannot be separated
during the refining process are cooled and cast into cathodes in preparation for electro-refining.
Electrorefining uses an electrochemical cell to plate out the pure copper content while
simultaneously depositing the other metal content onto the cathode. The high purity metal
anodes are sent to bar and paste manufacturing as a feedstock, while the remaining metal content
is deposited as a sludge that is scraped off the cathode into a bin. The sludge is later sold to an
appropriate refiner or is sold as is to a customer using the remaining metal content. The sludge
sometimes will undergo further refining using methods suited for the particular metal content.
Inventory from three facilities performing post-industrial recycling were collected
through site-visits by project personnel. Data were collected on input and output materials,
natural resource consumption (e.g., natural gas), energy consumption, and basic process
parameters such as process throughput. Inventory data were allocated to solders based on the
composition of the alloys produced.
The solder alloys generated from the post-industrial recycling process are primarily used
as secondary metal feedstock to the solder manufacturing process. The percentages of base
metals that are primary, or virgin, materials were estimated from data collected from solder
manufacturers, and are displayed in Table 2-8 below. For each solder, the majority of each metal
2-24
-------
comes from virgin content, with the remaining secondary content coming primarily from the
post- industrial recycling content. Since all the secondary metal content was assumed to have
been generated through the post-industrial recycling process, the inventory data were weighted
to reflect the ratio of virgin content to secondary, recycled content. For example, sixty-eight
percent of the metals for SnPb came from virgin content, therefore, sixty-eight percent of the
inventory data representing metals production came from the upstream materials extraction and
processing data set. Similarly, thirty-two percent of both the Sn and Pb content of SnPb is
recycled, thus thirty-two percent of the metals inventory data came from the post-industrial
solder recycling data. As a result, if a metal has a high virgin content, more of the inventory will
be represented in the upstream life-cycle stage than in the manufacturing stage; while,
alternatively, if a metal has a high recycled content, more of the inventory will be represented in
the manufacturing stage than in the upstream stage.
Table 2-8. Average virgin content of base metals used in solder manufacturing
Solder Type
SnPb
SAC
BSA
SABC
SnCu
Sn
68 percent
74 percent
74 percent
74 percent
74 percent
Pb
68 percent
—
—
—
—
Ag
—
68 percent
68 percent
68 percent
—
Cu
—
93 percent
—
68 percent
81 percent
Bi
—
—
99 percent
99 percent
—
Note: No data were provided for SnCu, therefore, the content was assumed to be the average virgin content of Sn
from Sn-bearing alternatives (i.e., SAC, BSA, and SABC); and the average virgin content of Cu from Cu-bearing
alternatives (i.e., SAC and SACB)
For the alternative alloys, the estimates attempted to predict operations in full production;
however, these are indeed predictions and may not represent what will occur in full production.
The estimates are difficult to determine at this time because the limited production of these
alternative solders are currently made in batch processes as required, rather than in full
production. Post-industrial recycling is performed at some solder manufacturing facilities and,
in this study, is included as a separate unit process in the solder manufacturing life-cycle stage.
While the LFSP does not model post-consumer waste recycled directly back into the product, the
process of recycling solder from PWBs (via demanufacturing and copper smelting) is accounted
for in the EOL life-cycle stage.
2-25
-------
2.3.2 Limitations and Uncertainties
The limitations and uncertainties associated with the manufacturing stage are related to
the following categories:
•• the product system boundaries (scope),
•• the data collection process, and
•• the data.
Specific limitations/uncertainties for each of these categories are briefly described below.
2.3.2.1 Product system boundary uncertainties
In this LCA, all secondary metal content was assumed to have been generated through
post-industrial recycling, rather than through post-consumer recycling. This may lead to an over
estimate of impacts in post-industrial recycling. In practice, secondary material is obtained first
from post-industrial recycling, and then from outside, post-consumer sources when additional
material is required. Post-industrial content is more cost-effective as it requires less energy to
refine into a common alloy, rather than to create the alloy from material of a composition
significantly different from the desired alloy, or with unpredicted contaminants. This
assumption leads to uncertainty in the project results.
2.3.2.2 Data collection process uncertainties
Limitations and uncertainties related to the data collection process include the fact that
companies were self-selected, which could lead to selection bias (i.e., those companies that are
more advanced in terms of environmental protection might be more willing to supply data than
those that are less progressive). Companies providing data also may have a vested interest in the
project outcome, which could result in biased data being provided. Much of the data collected
for the solder manufacturing life-cycle stage was obtained through site-visits by project
personnel, however, limiting the opportunity for bias through reporting by the manufacturer.
Where possible, multiple sets of data were obtained for this project to develop life-cycle
processes. The peer review process and employment of the Core and Technical Work Groups as
reviewers in this project is intended to help identify and reduce any such bias.
2.3.2.3 Data uncertainties
Additional limitations to the manufacturing stage inventory are related to the data
themselves. Specific data with the greatest uncertainty include the scaling of full production
data for lead free alternatives from data collected for batch processes and from manufacturers'
professional experience. In some cases, solder manufacturing inventory for lead-free alloys was
developed from the batch process data adjusted to account for scaled-up production, and for
required process changes estimated through the experience of process engineers.
2-26
-------
Due to the confidentiality of flux chemistries and the variability in chemistries
manufactured by companies for use with the various solders, data for flux manufacturing was
based only on flux formulation. Variability in chemical constituents used for fluxes and any
associated process changes required to manufacture other fluxes results in uncertainty in the
study results.
2-27
-------
2.4 SOLDER USE/APPLICATION
Solder is primarily used to attach electronic components to PWBs during the assembly
process. In addition, the selection of the type of solder has no effect on the energy consumed
over the lifetime of the product the assembly becomes a part of; thus, for the purposes of the
LCA, the use stage is defined as the solder application process, and does not include the period
of time during which the electronics assembly is used for its intended application.
The process of solder application differs for paste and bar solders. Paste solders are
applied through a reflow soldering process that uses a heated oven to melt, or reflow the solder
paste. Paste solder is used to attach surface mount components to the surface of the PWB. Bar
solder is applied using a wave soldering process that requires passing the populated PWB over a
defined wave of molten solder. Wave soldering is used to attach through-hole components and
other hardware, such as connectors to the surface of a PWB. Some boards require assembly
using both methods to attach all of the boards components.
The electricity consumed during application is directly dependent on the melting point of
the individual solder alloys, which vary significantly. Because these energy differences were
suspected to be important within the solder life-cycle, collection of primary, measured data from
the solder application/use stage was given priority. Testing of electricity consumption was
performed at two facilities, and the data were linked to the electricity generation process in the
use stage LCI. This section presents the methodology and results of testing, and compares
results to other studies of electricity consumption. In the test results, it also discusses data
quality, and limitations and uncertainties.
2.4.1 Methodology
2.4.1.1 Paste solder
Solder paste is applied to a PWB using a reflow soldering process, which is shown in
Figure 2-11. A screen is first prepared with a stencil defining the pattern of solder application
for a specific PWB design. Solder paste is then introduced to the screen, and applied to the
PWBs using a squeegee to control the amount of solder paste applied. After the boards are
populated with components using a pick and place machine, applying surface mount components
to the pads covered with solder paste. Components are held in place by the paste and prevented
from moving throughout the remainder of the assembly process.
Populated boards are passed through an oven comprised of six to twelve temperature
controlled zones, configured to create a temperature-time reflow profile to control the manner in
which the solder paste is melted to form the solder joints. PWBs are then passed through a
chiller (optional) or allowed to cool in air. Depending on the type of flux, PWBs may need to
pass through a cleaning step to remove any flux residue prior to assembly.
2-28
-------
Energy
Energy
Paste
Screening
>,
Component
Placement
>,
Solder
Reflow
Chiller
(optional)
Figure 2-11. Solder Paste Reflow Process Diagram
Life-cycle inventory data for the solder paste reflow process were collected through the
execution of a detailed testing protocol developed in consultation with industry experts
knowledgeable about reflow assembly and the overall goals of the LCA project. The developed
protocol balanced the need to collect data in a timely and cost-efficient manner with the desire to
measure the primary factors of power consumption during assembly; namely, the shape of the
oven temperature profile, conveyor speed, oven loading, and the overall mass of the PWB
assembly. In order to evaluate the power consumption under typical operating conditions, it was
assumed that the ovens would be operating continuously throughout the day or that work would
be scheduled to minimize the cost of operation. Therefore, testing was confined to the
measurement of power consumption during periods of steady-state operation, neglecting the
preheat cycle.
As a result of prior testing performed by Intel, assembly profiles describing the rate and
duration of the incremental temperature changes the assembly must undergo to obtain a
functioning solder joint were already available for all but BSA. A suggested profile for BSA
was obtained from Hewlett Packard and used by Intel to develop an appropriate reflow profile.
The suggested profile was adjusted using a set of thermocouples attached to the surface of the
panel. The panel was then passed repeatedly through the temperature zones of the reflow oven
while the profile was adjusted until the surface temperature of the panel met the minimum peak
melting temperature of the solder. The resulting profile for each solder is depicted in Figure 2-
12.
The profiles presented in the figure represent ramp-soak-spike (RSS) assembly profiles,
so named for their quick ramp up to melting temperature, followed by a period of slow
temperature increase to promote the proper flow characteristics, before a final spike up to the
target peak temperature. Other assembly approaches (i.e., ramp to spike) may also be valid and
might result in slightly different energy consumption data. The time domain has been removed
from the profiles in Figure 2-12 to protect the confidentiality of the research conducted by Intel.
2-29
-------
Figure 2-12. Reflow Profiles for Solder Pastes
For comparison purposes, each profile was developed using a constant conveyor speed
across profiles to ensure a constant and comparable oven loading during periods of energy
measurement. Characteristics of the solder profiles are presented in Table 2-9.
Table 2-9. Reflow profile specifications
Solder
SnPb
SAC
BSA
SABC
Peak
temperature (°C) a
204.4-219.1
235.2-248.8
160.2-170.1
235.2-248.8
Average TAL
(seconds) b
51
65
65
65
Change in
temperature (°C)
14.7
13.6
9.9
13.6
a Peak temperature represents the peak temperatures taken at different points on the PWB surface, reported as a
range.
b Time above liquidous (TAL) is the period of time a board is heated above the liquidous temperature of the solder.
Note: The same reflow profile was used for both SABC and SAC
2-30
-------
Because solder reflow occurs once the joint reaches the minimus temperature required for
the particular solder, and because the scope of the testing was limited to energy consumption and
not joint testing, preassembled Intel micro ATX motherboards were used to limit the cost of the
testing. The Intel motherboard was selected because it is at the upper end of applications typical
for the consumer electronics market in terms of size, mass, and complexity. A photo of the test
board is shown in Figure 2-13. Specifications for the test assembly are presented in Table 2-10.
Figure 2-13. Reflow Test PWB Assembly
Table 2-10. Reflow test vehicle specifications
Category
PWB type
Length
Width
Assembly mass
Solder mass
Specification
Intel Micro ATX
Motherboard
9.6 inches
9.6 inches
225 grams
2.5 grams/board
Initial testing was conducted at Intel using a ten zone forced convection reflow oven with
an attached water-cooled chiller unit to cool the assemblies following reflow. A second phase of
testing was conducted at Vitronics-Soltec using an eight heating zone forced convection oven
with two cooling zones. Power consumption was measured at both facilities using a data logger
connected to the main power. Assemblies were fed into the oven at a controlled rate of 35.5
inches per minute until the oven achieved a fully loaded condition under the design profile.
Electricity consumption data were collected from the time the first assembly entered the oven
2-31
-------
until the final assembly exited. Assemblies exiting the oven were allowed to reach room
temperature before being reintroduced to the oven for the next test run.
Results of the reflow testing are presented in Table 2-11. Measured power consumption
data from the testing were converted to energy consumed using the time of the individual test
run, then normalized based on the amount of solder applied to the PWBs. Mass of solder applied
to the board was estimated by Intel and compared to measured data for a similar Intel ATX
mother board. Energy consumption data for each of the test runs were averaged and converted to
megajoules per kilogram of solder for entry into the LCA.
Table 2-11. Paste solder reflow test data
Solder Alloy
SnPb
SAC
BSA
SABC
Power Consumption (kW)
Vitronics- Soltec
8.3
9.1
6.8
9.1
Intel
23.3
25.2
15.7
25.2
Average energy
consumption (MJ/kg)
412
447
297
447
Note: power consumption data were converted to an average energy consumption using the following method:
[power (kilowatt [kW]) * 3.6 (MJ/kW-h)]/ time of test run (h)
2.4.1.2 Bar Solder
Bar solder is applied during PWB assembly in a soldering process known as wave
soldering. Basic process steps associated with wave soldering are displayed in Figure 2-14.
PWBs already populated with through-hole components and hardware (e.g., connectors) are first
coated with flux to facilitate the proper solder flow across the surface of the circuit pads. PWBs
are then loaded onto a conveyor and passed over a pot of molten solder that is pumped through a
nozzle with a defined flow profile, or wave. The solder, which is allowed only to contact the
bottom surface of the board, wicks up into the through-holes, forming a solder joint. Boards are
then allowed to cool in air and are inspected for defects before going on to further processing, if
required.
2-32
-------
Component
Placement
>,
Flux
Application
Energy
1
PWB
Preheat
Energy
1
Wave
Solder
Figure 2-14. Process Flow Diagram for Wave Solder
Wave solder data was collected through the development of a detailed test protocol in
conjunction with industry experts. The protocol balanced the need to collect data in a timely and
cost efficient manner with the desire to measure the key parameters of wave assembly; namely,
the pot temperature, conveyor speed, flux usage, and the overall mass of solder applied to the
PWB assembly.
Wave solder testing was conducted at Vitronics-Soltec using a PWB assembly measuring
7 inches wide x 10 inches long, designed specifically for wave solder application. In order to
evaluate the power consumption under typical operating conditions, it was assumed that the
solder pot would be operating continuously throughout the day; therefore, power consumption
measurements were confined to periods of steady-state operation, neglecting the solder pot
preheat cycle.
Testing was conducted using both water-based and alcohol-based flux. Energy data were
measured using a continuous data logger connected to the main power feed of the wave solder
machine. PWBs were fed into the wave solder machine with a board length spacing between
assemblies.^Energy data were collected from the time the first board was placed on the
conveyor until the time of the exit of the final board from the machine. Flux use was measured
by diverting the flow of flux into a collection jar over the span of the test period. Assemblies
were weighed both before and after soldering to determine the mass of solder applied to the
assemblies.
Table 2-12 presents the results of the wave solder testing described above. Energy use
data for both alcohol and water-based flux were averaged and then normalized for the amount of
solder applied to each PWB. The amount of solder applied was measured by comparing the
mass of the board after assembly with the initial mass of the board measured just before wave
soldering. Flux use also was normalized for the mass of solder applied.
2-33
-------
Table 2-12. Wave solder test data
Solder Type
SnPb
SAC
SnCu
Flux Type
Alcohol-based
Water-based
Alcohol-based
Water-based
Alcohol-based
Water-based
Energy Use
(MJ/kg solder)
56.4
60.9
65.4
70.3
65.9
70.8
Flux Use
(kg flux/kg solder)
0.733
1.133
0.838
1.294
0.843
1.073
2.4.2 Limitations and Uncertainties
2.4.2.1 Paste solder
Due to the limited number of PWBs available for assembly testing, unpopulated
motherboards without solder were used to measure the energy consumption during reflow
testing. This allowed for the reuse of the boards, after they were allowed to cool, for testing of
the remaining alternatives. The measurement of energy data will likely underestimate the overall
energy load measured in the testing resulting in uncertainty. In addition, the mass of solder used
to normalize the energy test data was developed based on data already measured for another
similar Intel test vehicle. Data were compared for validation to an estimate of solder applied per
surface area of PWB developed from a series of similar PWB designs.
2.4.2.2 Bar solder
Wave testing was performed at a single facility using the test protocol described. The
resulting data represent that process, but may not be reflective of all wave soldering. The single
set of data presents uncertainty in the overall results.
2-34
-------
2.5 END-OF-LIFE
2.5.1 Methodology
The functional unit in this analysis is a unit volume of solder used on an arbitrary PWB
design. At EOL, the solder is inextricably linked to a PWB which, in turn, becomes part of some
electronic product. To the extent possible, this project follows the solder itself to its final
disposition. Where EOL activities involve processing the entire product or the PWB on which
the solder lies, the flows from those activities are allocated by the mass of product or PWB being
processed. For example, if the mass of the solder accounted for a third of the total mass of
material processed, only one third of the process flows are attributed to the solder. The
allocation prevents the results from being unduly influenced from processing that is unrelated to
the amount of solder.
Allocation is not an issue in earlier life-cycle stages including the upstream,
manufacturing, and use/application stages. At this point the solder has not yet been incorporated
into another product. In order to remain as consistent as possible with the functional unit, the
EOL outputs are limited to the metals in the solders. For example, while incineration energy
inputs are allocated to the mass of the waste going in (which contains the solder), only the metal
outputs from the solder are characterized as outputs (and not outputs such as dioxins from
incinerating the boards). Further details are provided in subsections that follow.
For the EOL analysis, a PWB is assumed to have reached EOL status when:
• it has served its useful life;
• it is no longer functional; and
it is rendered unusable due to technological obsolescence.
The major EOL dispositions considered in this analysis are as follows:
• landfilling - includes hazardous and non-hazardous waste landfills;
• incineration - waste to energy incineration; and
• post-consumer recycling3
regulated demanufacturing followed by copper smelting, and
unregulated recycling and disposal.
The various EOL dispositions were allocated as the probability of a PWB going to a
certain EOL disposition. The U.S. EPA estimated that 9 percent of electronic waste is recycled
(EPA, 2002). No direct estimates on the amount of electronic waste landfilled or incinerated
were identified. The EPA reported, however, that of the municipal solid waste (MSW) generated
in the United States in 2000, 55.3 percent was landfilled and 14.5 percent was incinerated. The
remaining was either recovered for recycling or composted. Based upon the proportions of the
3Post-industrial recycling is included in the solder manufacturing stage (see Section 2.3).
2-35
-------
MSW being landfilled or incinerated, it was assumed the fate of the remaining 91 percent of
electronic waste is 72 percent to landfilling and 19 percent to incineration. An independent
verification on these estimates was conducted by UT researchers. Individual states were
contacted, and the percentages estimated above were consistent with what was found in many
states.
Among electronics waste that is recycled, two possible scenarios were included: (1)
regulated demanufacturing followed by copper smelting for materials recovery and, (2)
unregulated recycling and disposal. The latter was included in response to reports of recycling
overseas under uncontrolled or unregulated conditions (BAN & SVTC, 2002). Half of the
electronic waste being sent for recycling was assumed to be processed under controlled
conditions and the other half under uncontrolled conditions. Table 2-13 presents the
assumptions used for the EOL life-cycle stage dispositions for most of the alloys.
The distribution of BSA at EOL differs somewhat from what is represented in Table 2-
13. The flow charts showing how each alloy is modeled (refer to Figures 2-2 through 2-8) show
which processes are included in the life-cycles of each alloy. For BSA (see Figure 2-4), which
has a 57 percent bismuth content, the PWBs with BSA are assumed to be demanufactured under
the first recycling scenario, but are assumed not to be sent to a copper smelter. This is due to the
high bismuth content that makes cost-effective metals recovery difficult. Therefore, in the BSA
life-cycle model, we assume that once the electronic waste has been demanufactured (i.e.,
disassembled and/or shredded), it is then sent to a landfill or an incinerator, in the same
proportions assumed for the non-recycled waste (as described earlier). As a result, there is no
copper smelting process for BSA, and more landfilling and incineration than modeled for the
other alloys.
Table 2-13. General distribution of EOL electronics by disposition
Disposition
Landfilling
Incineration
Recycling: demanufacturing and copper smelting*
Recycling: unregulated recycling and disposal
Distribution
72 percent
19 percent
4.5 percent
4.5 percent
*Note: The BSA life-cycle does not include the copper smelting process. After demanufacturing, the
waste PWBs with BSA are sent to either a landfill or an incinerator.
The methodologies for each disposition are presented in subsections 2.5.1.1 through
2.5.1.4. Included in each methodology section is also a discussion of data sources and data
quality. Table 2-14 lists the general data collection approaches for each disposition. Limitations
and uncertainties for all dispositions are presented in Section 2.5.2.
2-36
-------
Table 2-14. Data collection approach for EOL dispositions
Disposition
Landfilling
Incineration
Recycling: demanufacturing and copper smelting
Recycling: unregulated recycling and disposal
Data source
Literature and
leachability analysis
Literature
Primary data
Primary data and assumptions
2.5.1.1 Landfilling
The inputs to the landfilling process modeled in this study include the fuels required to
run the landfill equipment (i.e., diesel fuel) and the PWBs or electronic waste assumed to be sent
to the landfill. Outputs include the solder metals, which were quantitatively measured based on
a leachability analysis commissioned for the LFSP (Appendix C). The transport and collection
of waste is not included since these activities would be similar for any of the solder types being
analyzed. While this will not affect solder to solder comparisons, it can affect comparisons
across life-cycle stages. The exclusion of transportation results in less total overall landfill
impacts.
A literature search was conducted to estimate the fuel requirements needed for operating
landfill equipment. Data were not available on landfilling of PWBs or electronics alone as there
are not dedicated electronics landfills. Energy requirements for landfilling MSW was used as a
surrogate for processing electronics waste, as it is expected that electronics waste will be
combined with all types of waste. Further, the operation of landfill equipment is not expected to
vary greatly depending on the type of waste. Denison (1996) reported that 230,800 BTU of
energy per ton of MSW (equivalent to 0.288 MJ/kg MSW) are used for landfill equipment.
Diesel fuel was assumed to be used to operate the heavy equipment. The diesel fuel production
process from the GaBi3 databases was included in the landfill inventory for each solder type and
allocated to the amount of fuel consumed.
The outputs from the landfilling process were based on a leachability study conducted by
the University of Florida (UF) in support of the LFSP. The study conducted the EPA-approved
toxicity characteristic leachate procedure (TCLP) test on each of the solder types included in the
LFSP. In addition to the TCLP test, a less aggressive test method called the synthetic
precipitation leaching procedure (SPLP) also was conducted. The TCLP test uses acetic acid
and sodium hydroxide in the leaching fluid, and is expected to represent conditions in a landfill.
The SPLP uses sulfuric acid and nitric acid, which is intended to be more representative of
rainwater. Appendix C presents the draft report describing the methodology and results. The
leachate output data are used to represent potential releases to water from landfilling. No further
fate and transport modeling is done in the context of this LCA, since the LCA does not address
specific locations for impacts and does not have the ability to incorporate site specific fate and
transport parameters. The output data used in the LFSP are derived from the TCLP study;
2-37
-------
however, the acetic acid contained in the TCLP leachate is known to more aggressively leach
lead than other metals. In response to concerns about whether the TCLP will over-estimate the
leaching from SnPb solder, an alternate analysis also was conducted using the detection limits as
a lower bound (Section 3.3.3).
From the teachability study results, which were provided in concentrations of metal per
liter of leachate, the data were converted to kilograms of metal outputs per kilogram of solder
(see Appendix C). Table 2-15 presents the data used as the landfilling process outputs based on
the teachability study. The table shows that lead in the SnPb alloy leached to the greatest extent,
followed by bismuth in BSA. In addition, other outputs from the landfilling process group
include outputs from the diesel fuel production process.
Table 2-15. TCLP-based leachate data used to predict outputs from landfilling
Solder Alloy
SnPb
SnPb
SAC
SAC
SAC
BSA
BSA
BSA
SABC
SABC
SABC
SABC
SnCu
SnCu
Solder type
Paste and bar
Paste and bar
Paste and bar
Paste and bar
Paste and bar
Paste
Paste
Paste
Paste
Paste
Paste
Paste
Bar
Bar
Metal
Lead
Tin
Silver
Tin
Copper
Bismuth
Tin
Silver
Bismuth
Copper
Silver
Tin
Copper
Tin
Fraction leached
(kg metal/kg solder)
1.88E-01
2.93E-05
1.86E-05
1.86E-05
1.34E-05
2.39E-02
5.18E-04
2.03E-05
9.09E-04
3.59E-05
2.39E-05
2.39E-05
2.72E-05
2.39E-05
The inputs to landfilling include only the fuel inputs from landfill equipment and PWB
waste entering the landfill. Other inputs such as fill materials were not included. Thus, the
inputs to this data set are considered incomplete; however, the fuel is expected to be a major
input and the production associated with the diesel fuel used in the landfilling process is included
in this process. The energy data used for landfilling was estimated from data on MSW and, thus,
does not exactly match the waste being considered in this study. It is expected that activities for
processing electronic waste at a landfill would be similar to processing MSW, however. With
differences in the density of the wastes, there would likely be differences in the fuel consumption
during processing. The quality of the output data is considered to be much higher. The
teachability tests were done directly to support this project, and measured the fraction of each
metal that leached from each solder type.
2-38
-------
2.5.1.2 Incineration
Direct data for the flows associated with incinerating electronic waste were unavailable;
therefore, literature reviews were conducted to estimate incineration flows. Energy inputs are
based on a waste to energy combustion facility that can process 500 metric tonnes per day of
MSW as presented by Harrison et al. (2000). The total energy recovered during MSW
combustion was reported as being equivalent to 6.36 MJ/kg of MSW. This value was
mathematically derived from a series of calculations according to Harrison et al. that determined
that the heat generated from combustion of the waste more than offset the energy consumed to
fire the incinerator. For the purposes of modeling the solder life-cycles, natural gas was assumed
to be used as the fuel for the combustion facility. Incineration of electronics would likely result
in an even higher net energy gain because the BTU content for a PWB exceeds that for a similar
mass of MSW. The energy gain was applied to the system as an offset, acting as a credit to
natural gas production (shown as a negative number in the LCI) and the associated process flows
for its' production.
The metal outputs were estimated by predicting the percent distribution of outputs to
three dispositions: bottom ash, fly ash, and fumes. Table 2-16 presents the percentages that are
applied to the mass of metal outputs. Metals in the bottom ash were assumed to be landfilled, and
the teachability results presented in Section 2.5.1.1. were used to predict the resulting landfill
outputs to water.
Table 2-16. Percent distribution of incinerator out
Species
Copper (a)
Lead (a)(b)
Silver (c)
Tin(b)(d)
Bismuth (b)
Bottom ash
94.8
64
82
65
81
Fly ash
4.75
34.5
17
34
18
Fumes
0.5
1.5
1
1
1
puts
Total
100
100
100
100
100
(a) Average of four data points from Chang-Hwan (no date) and Abanades (2002).
(b) At 800°C.
(c) At 1100°C. Disposition based upon EPA reference for MACT technology and metal volatility states. Note:
Listed as hazardous constituent under RCRA Appendix VIII of Section 261; however not a Hazardous Air
Pollutant (HAP) under Clean Air Act, therefore not categorized under maximum achievable control technology
(MACT) metals volatility groups directly. Listed disposition based on cement kiln burning Hazardous wastes.
(d) Chang-Hwan (no date).
The data for the incineration inputs include data obtained through secondary literature for
energy saved and from the GaBi3 database for the natural gas inventory. The data quality
description for the natural gas inventory is provided in Table 2-6. The outputs were estimated
from literature describing the fate of metals from incineration. Overall, the incineration data
quality for the purposes of the LFSP is moderate, as it is from secondary data and required
estimates from data on general thermal treatment.
2-39
-------
2.5.1.3 Post-consumer recycling: demanufacturing and copper smelting
Primary data were collected from three demanufacturing facilities and copper smelting
data were obtained from two copper smelters, Noranda and Boliden. Data for each process were
averaged from each data set collected using an EOL data collection form (see Appendix F). See
Section 2.3.1 for more information on primary data collection conducted for the LFSP.
PWBs sent to demanufacturing are dismantled and shredded and then sent to a copper
smelter for materials recovery. The demanufacturing process simply includes electric power
used to operate dismantling and shredding equipment and the waste PWBs as inputs The
generation of electricity from the U.S. electric grid, as described in Section 2.2, is linked to the
demanufacturing process in proportion to the amount of electricity required to process waste
PWBs. The mass of solder is assumed to remain constant throughout the demanufacturing
process, thus the mass of waste PWB (and associated solder) as an input is equal to the mass of
the shredded PWB (and associated solder) as an output. The shredded PWBs are the only direct
outputs from the demanufacturing process. Indirect outputs are emissions associated with
electricity generation.
The shredded PWBs containing each alloy (except BSA) are assumed to be sent to a
copper smelter. BSA is assumed to be sent to incineration or landfilling after demanufacturing
(discussed above). Based on averaged data, the copper smelting process is fueled by electricity,
LPG, light fuel oil, heavy fuel oil, and kerosene. Only kerosene did not meet the mass cut off
based on the decision rules as described in Section 2.1.2 and, thus, upstream inventories of all
the fuels, except kerosene, were linked to the copper smelting process (as depicted in Figures 2-2
through 2-8).
Estimates of outputs from copper smelting were obtained from interviews and site visits.
Process outputs for solder metals were allocated according to the smelting process distributions
presented in Table 2-17.
Data for regulated recycling (i.e., demanufacturing and copper smelting) were from
primary data sources and are considered of good quality. The demanufacturing process data are
expected to be of greater quality than the copper smelting data, as there were more data sets
which were used to average the primary data received.
Table 2-17. Fraction distribution of copper smelting outputs.
Species
Tin
Lead
Silver
Copper
Bismuth
Air
0.0023
0.0023
0
0
0.00092
Slag/tailings
impoundment
0.9977
0.05
0.05
0.05
0.79908
Product
Negligible
Negligible
0.95
0.95
0.2
Lead to
recovery
N/A
0.9477
N/A
N/A
N/A
Total
1
1
1
1
1
N/A=not applicable
2-40
-------
2.5.1.4 Post-consumer recycling: unregulated recycling and disposal
The unregulated PWB recycling and disposal process evaluated in the LFSP is modeled
after descriptions of processes in various Asian cities in a recent report (hereafter referred to as
the BAN report) by a coalition of environmental groups (BAN & SVTC, 2002). These processes
involve heat application to remove valuable components and recover solder from PWBs
followed by open burning or dumping of the stripped PWBs. Unregulated recycling and disposal
processes are expected to result in uncontrolled air emissions, water discharges, and soil releases
of solder metals. Although some air emissions may occur during the heating process to recover
valuable components and solder metals, the vast majority of environmental releases are expected
to occur from open dumping or burning of stripped PWBs. Figure 2-15 presents a process flow
diagram and describes the unregulated recycling and disposal processes in the BAN report in
more detail.
Descriptions of unregulated recycling and disposal processes for a few locations are
presented in the BAN report, but it should be noted that the processes may not be representative
of unregulated disposal processes at other locations.
The LFSP did not attempt to determine precise environmental releases at various steps in
the unregulated recycling and disposal process. Rather, our approach was to estimate: (1) the
amount of solder entering these facilities on PWBs; (2) the amount recovered for resale; and (3)
the distribution of the remainder among releases to air, soil, and water. The environmental
outputs and associated impacts from combustion of the plastics and flame retardants contained in
PWBs are not included in the analysis.
The amount of solder entering unregulated facilities was calculated assuming the amount
per functional unit (e.g., per 1000 cc of solder as applied to an arbitrary PWB design) is directly
proportional to the percent of waste electronics being exported for recycling and disposal.
Therefore, assuming 4.5 percent of EOL electronics is being exported to unregulated facilities,
4.5 percent of the functional unit (45 cc solder) also is being exported.
The amount of solder recovered from PWBs was estimated based on the amount
theoretically available for recovery adjusted to account for inefficiencies in the solder recovery
process. The amount theoretically available for recovery was defined as the mass of solder used
in connections, not including solder used in surface finishing. Based on data for SnPb solder
collected by the LFSP, approximately 65 percent of the solder on a PWB can be recovered;
however, since the solder recovery process employed by unregulated facilities is not likely to be
100 percent efficient, 50 percent recovery of the solder was assumed.
2-41
-------
PWBs
Componen
to resale
Solder A
fumes
t
Component
removal
t
Heat
Solder
s to
resale
Solder A
fumes
4
1
1
„, Solder
recovery
T *
Heat
Stripped
PWBs
Copper
recovery
4
1
I
r*
Uncontrolled
land disposal
1
Soil, surface
water, and
groundwater
releases
Solder fumes &
other air emissions
t
^
Open
burning
_(0_ptip_n_al)_
I
Soil, surface
water, and groundwater
releases
Figure 2-15. Unregulated Recycling and Disposal Process Flow Diagram
Estimating the distribution of the remainder among releases to air, soil, and water is more
problematic. The BAN report presents metals concentrations found in a limited number of soil,
sediment, and water samples from locations along a river in China where PWBs and wires are
treated and burned. These data cannot be related to the LFSP functional unit since there is no
record of the number of PWBs treated and disposed at these sites. Furthermore, the BAN data
do not include air emissions. EPA is currently conducting research to measure air emissions
from the open burning of electronics waste.
Pending release of EPA's data on air emissions from open burning of electronics waste,
the LFSP assumed 75 percent of the solder not recovered for resale is released to air and soil
with the remaining 25 percent released to water via surface water runoff and leaching to
groundwater. It should be noted that, in this instance, the relative distribution between soil and
water does not affect LCIA results for public toxicity impacts because releases to soil are
uncontained, unlike disposal in a controlled landfill. This means that there is potential for
exposure to all of the soil releases just as there is potential for exposure to all air releases. The
LCIA method for public toxicity impacts uses release amounts as a surrogate for exposure
together with a toxicity value. More information on the LCIA methodology for public toxicity
impacts can be found in
Section 3.2.12.
Table 2-18 summarizes the assumptions used to calculate the unregulated solder
inventory. The overall quality of the data for the unregulated recycling and disposal process is
2-42
-------
considered low. Assumptions were made based on limited available data; however, the project
Core Group agreed that it was important to recognize this scenario by including it even with
general assumptions about the fate of the solder metals.
Table 2-18. Unregulated recycling and disposal assumptions
Parameter
Assumption
(volume solder per
functional unit)
Basis
Volume solder entering
unregulated facilities
onPWBs
45 cc
4.5 percent of the solder functional unit. Assumes the volume
of solder entering unregulated facilities is directly proportional
to the percent of waste electronics being exported.
Volume recovered for
resale
22.5 cc
50 percent of the volume of solder entering unregulated
facilities on PWBs. Based on the percent of solder that can
theoretically be recovered from a typical PWB (e.g., used in
connections instead of as a surface finish) minus losses in the
recovery process
Volume released to air
and soil
16.9cc
37.5 percent of the volume of solder entering unregulated
facilities on PWBs. Assumes 75 percent of solder remaining
after solder recovery has a final disposition in air or soil. This
value is subject to change pending results of open burning
trials being conducted by EPA.
Volume released to
water
5.6 cc
12.5 percent of the volume of solder entering unregulated
facilities on PWBs. Assumes 25 percent of solder remaining
after solder recovery is released to water either through
leachate or surface water runoff from dumps and burn piles.
This value is subject to change pending results of open burning
trials being conducted by EPA.
2.5.2 Limitations and Uncertainties
Assumptions about the disposition percentages may not truly represent the actual
dispositions. Sensitivity analyses, which vary these assumptions, can be conducted if results
show enough impacts at EOL to warrant further analysis. For incineration and landfilling
inventories, predictions about process flows were often based on processing MSW rather than
specifically on processing solder or PWBs. For regulated post-consumer recycling, fewer
limitations exist as primary data were collected for demanufacturing and copper smelting.
2-43
-------
2.6 BASELINE LIFE-CYCLE INVENTORY RESULTS
Figures 2-16 and 2-17 present the total mass quantity of inputs and outputs, respectively,
for each paste alloy. Figures 2-18 and 2-19 present the inputs and outputs, respectively, for each
of the bar alloys. These LCI results are only intended to be used as an interim step to conducting
the LCIA; therefore, only a brief discussion is provided here. The reflow solders show similar
total mass input quantities for SnPb, SAC and SABC, with SAC having the greatest mass
inventory inputs (Figure 2-16). BSA has the fewest mass inputs. The greatest contributor to
these mass inputs is water as a resource. The outputs from the paste solder life-cycles
(Figure 2-17) show SnPb, SAC, and SABC to be almost equivalent to one another and BSA to
have a lower mass output. The outputs are also dominated by water emissions.
Reflow Solder Total Mass Inputs
en nnn
/in nnn
M
E -30 nnn
o on nnn
'2
10 000
n
D Waste for recycling
• Deposited goods
n Valuable substances
• Resources
SnPb SAC BSA SABC
solder
Figure 2-16. Paste Solder Total Mass Inputs
2-44
-------
Reflow Solder Total Mass Outputs
40,000
35,000
30,000
1 25,000
I) 20,000
= 15,000
^ 10,000
5,000
0
D Waste for recycling
• Deposited goods
D Emissions to soil
• Emissions to water
D Emissions to air
D Valuable substances
SnPb SAC BSA
solder
SABC
Figure 2-17. Paste Solder Total Mass Outputs
For the bar solder inventories, SAC has the greatest mass quantity of inputs, and SnPb
and SnCu mass inputs are nearly equivalent. The outputs follow the same pattern. Similar to the
paste solder, most of the inputs are from water resources. The outputs are also dominated by
emissions to water.
Bar Solder Total Mass Inputs
n Waste for recycling
rjValuable substances
• Resources
Figure 2-18. Bar Solder Total Mass Inputs
2-45
-------
Bar Solder Total Mass Outputs
12,000
(A
E
TO
O)
O
dWaste for recycling
• Deposited goods
nEmissions to soil
• Emissions to water
nEmissions to air
DValuable substances
Figure 2-19. Bar Solder Total Mass Outputs
2-46
-------
REFERENCES
Abanades, S, G. Flamant, B. Gagnepain, and D. Gauthier. 2002. "Fate of Heavy Metals During
Municipal Solid Waste Incineration." Waste Management and Research. 20:55-68.
BAN (Basal Action Network) & SVTC (Silicon Valley Toxics Coalition). 2002. Exporting
Harm: The High Tech Trashing of Asia, February
(www. svtc. org/cl eancc/pub s/technotrash. htm).
CEFIC etal. 2002. European Chemical Industry Council (CEFIC), European Electronic
Component Manufacturers Association (EECA), European Information, Communications
and Consumer Electronics Industry Technology Association (EICTA), and European
Association of Metals (EUROMETAUX). "Guidance Document on the Appliance of
Substances under Special Attention in Electric and Electronic-Products, Version 2.2."
November 25, 2002.
Chapman, P. F., F. Roberts. 1983. Metal andResources and Energy. Butterworth's
Monographs in Materials.
Denison, R. A. 1996. "Environmental Life-Cycle Comparisons of Recycling, Landfilling, and
Incineration: A Review of Recent Studies." Ann. Rev. Energy Env.
21:191-237.
Ecobilan, 1999. Database for Environmental Analysis and Managment (DEAM) life cycle
inventory database developed by Ecobilan Group.
EPA (U.S. Environmental Protection Agency). 2002. Municipal Solid Waste in the United
States: 2000 Facts and Figures.
GaBi. 2000. GaBi3: The Software System for Life-Cycle Engineering. Produced by PE & IKP
(PE Product Engineering GmbH & IKP University of Stuttgart), Stuttgart, Germany.
Harrison, K. W., R. D. Duan, M. A. Barlaz, S. R. Nishtala. 2000. "A Life-Cycle Inventory
Model of Municipal Solid Waste Combustion." Journal of Air and Waste Management
Association. 50:993-1003.
IDEMAT (Industrial Design Materials/ 1995. J.A.M Rammerswaal and J. Rombouts, Delft
University of Technology. Industrial Design Materials (IDEMAT). The Netherlands.
ISO (International Standards Organization). 1996. ISO 14040, Environmental Management -
Life-Cycle Assessment Principles and Framework. TC 2071 SC 5N 77. International
Standards Organization, Paris.
2-47
-------
Jung Chang-Hwan. No date. "Metal Flows in Thermal Treatment System of MSW in Japan."
Candidate's Degree of Master. Found at: http://ws3-
er.eng.hokudai.ac.jp/egpsee/alumni/abstracts/jung.pdf, Downloaded October 23, 2003.
Miller. 2002. Personal communication with H. Miller of SBC Global and S. Surak of
University of Tennessee, Dec 2, 2002.
Palmieri. 2002. Personal communication with Y. Palmieri, former President of the former
Bismuth Institute, and M. L. Socolof of University of Tennessee, April 16, 2002.
SET AC (Society of Environmental Toxicology and Chemistry). 1994. Life-Cycle Assessment
Data Quality: A Conceptual Framework. SET AC and SET AC Foundation for
Environmental Education, Inc. Washington, DC.
Socolof M.L., J.G. Overly, L.E. Kincaid, J.R. Geibig. 2001. Desktop Computer Displays: A
Life-Cycle Assessment, Volumes 1 and2. U.S. Environmental Protection Agency, EPA
744-R-01-004a,b, 2001. Available at: http://www.epa.gov/dfe/pubs/comp-
lic/lca/tocl.pdf.
U.S.BOM, (U.S. Bureau of Mines), 1989. Minerals Yearbook, "Minerals" Section. U.S.
Department of the Interior, Washington, DC.
Wiley-VHS, 1997. Ullmann's Encyclopedia of Industrial Chemistry, 5th Edition. Weinheim.
2-48
-------
Chapter 3
LIFE-CYCLE IMPACT ASSESSMENT
Within LCA, the LCI is a well-established methodology; however, LCIA methods are
less well-defined and continue to evolve (Barnthouse et a/., 1997; Fava et a/., 1993). For LCIA
toxicity impacts in particular, some of the methods commonly being applied include toxicity
potential, critical volume, and direct valuation (Guinee etal., 1996; ILSI, 1996; Curran, 1996).
There is currently no general consensus among the LCA community concerning which, if any, of
these methods are preferable, however. Efforts are under way to determine the appropriate level
of analytical sophistication in LCIA for various types of decision-making requirements and for
adequately addressing toxicity impacts (Bare, 1999).
Section 3.1 of this chapter presents the general LCIA methodology used in this LFSP
study, which takes a more detailed approach to chemical toxicity impacts than some of the
methods currently being used. This section also describes the data management and analysis
software used to calculate LCIA results. Section 3.2 presents the detailed characterization
methodologies for each impact category as well as the baseline LCIA results from the paste and
bar analyses. This section also discusses data sources, data quality, and the limitations and
uncertainties in this LCIA methodology as well as in the LCIA results. Section 3.3 presents
alternative analyses of the baseline results.
Our LCIA methodology calculates life-cycle impact category indicators using established
calculation methods for a number of traditional impact categories, such as global warming,
stratospheric ozone depletion, photochemical smog, and energy consumption. In addition, this
method calculates relative category indicators for potential impacts on human health and aquatic
ecotoxicity, impacts not always considered in traditional LCIA methodology. The toxicity
impact method is based on work for Saturn Corporation and the EPA Office of Research and
Development by the UT Center for Clean Products and Clean Technologies and used in the DfE
Computer Display Project (Socolof etal., 2001).
3.1 METHODOLOGY
In its simplest form, LCIA is the evaluation of potential impacts to any system as a result
of some action. LCIAs generally classify the consumption and loading data from the inventory
stage to various impact categories. Characterization methods are used to quantify the magnitude
of the contribution that loading or consumption could have in producing the associated impact.
LCIA does not seek to determine actual impacts, but rather to link the data gathered from the
LCI to impact categories and to quantify the relative magnitude of contribution to the impact
category (Fava etal., 1993; Barnthouse etal., 1997). Further, impacts in different impact
categories are generally calculated based on differing scales and, therefore, cannot be directly
compared.
Conceptually, there are three major phases of LCIA, as defined by the SET AC (Fava et
al, 1993):
3-1
-------
• • Classification—The process of assignment and initial aggregation of data from inventory
studies to impact categories (i.e., greenhouse gases or ozone depletion compounds).
•• Characterization—The analyses and estimation of the magnitude of potential impacts
for each impact category, derived through the application of specific impact assessment
tools. (In the LFSP, "impact scores" are calculated for inventory items that have been
classified into various impact categories and then aggregated into life-cycle impact
category indicators.)
•• Valuation—The assignment of relative values or weights to different impacts, and their
integration across impact categories to allow decision makers to assimilate and consider
the full range of relevant impact scores across impact categories.
The international standard for life-cycle impact assessment, ISO 14042, considers
classification and characterization to be mandatory elements of LCIA; valuation ("weighting") is
an optional element to be included depending on the goals and scope of the study. Both the
classification and characterization steps are completed in the LFSP, while the valuation step is
left to industry or others interested stakeholders. The methodologies for life-cycle impact
classification and characterization are described in Sections 3.1.1 and 3.1.2, respectively.
3.1.1 Classification
In the first step of classification, impact categories of interest are identified in the scoping
phase of the LCA. The categories included in the LFSP LCIA are listed below:
• • Natural Resource Impacts
renewable resource use
non-renewable materials use/depletion
energy use
solid waste landfill use
hazardous waste landfill use
•• Abiotic Ecosystem Impacts
global warming
stratospheric ozone depletion
photochemical smog
acidification
air quality (particulate matter loading)
water eutrophication (nutrient enrichment)
water quality (biological oxygen demand [BOD] and total suspended solids [TSS]
•• Potential Human Health and Ecotoxicity Impacts
chronic cancer human health effects—occupational
chronic cancer human health effects—public
chronic non-cancer human health effects—occupational
chronic non-cancer human health effects—public
aquatic ecotoxicity
5-2
-------
Radioactivity and radioactive landfill waste are not included as impact categories because they
are simply proportional to the use of electricity across all alternatives. Terrestrial ecotoxicity is
not included as a separate impact category because the method for calculating chronic non-
cancer public health impacts would be the same as for terrestrial ecotoxicity.
The second step of classification is assigning inventory flows to applicable impact
categories. Classification includes whether the inventory item is an input or output, the
disposition of the output, and, in some cases, the material properties for a particular inventory
item. Figure 3-1 shows a conceptual model of classification for the LFSP. Table 3-1 presents
the inventory types and material properties used to define which impact category will be
applicable to an inventory item. One inventory item may have multiple properties and, therefore,
would have multiple impacts. For example, methane is a global warming gas and has the
potential to create photochemical oxidants (to form smog).
Output inventory items from a process may have such varying dispositions as direct
release (to air, water, or land), treatment, or recycle/reuse. Outputs with direct release
dispositions are classified into impact categories for which impacts will be calculated in the
characterization phase of the LCIA. Outputs sent to treatment are considered inputs to a
treatment process and impacts are not calculated until direct releases from that process occur.
Similarly, outputs to recycle/reuse are considered inputs to previous processes and impacts are
not directly calculated for outputs that go to recycle/reuse. Figure 3-1 graphically depicts the
relationships between inventory type, dispositions, and impact categories. Note that a product is
also an output of a process; however, product outputs are not used to calculate any impacts.
Once impact categories for each inventory item are classified, life-cycle impact category
indicators are quantitatively estimated through the characterization step.
-------
£* <1>
2 £
| TO lr"PUtS
1 1
c
*- o
3 ^3
Q- w
=> y
n °-
U c/)
T3
ir
-H- Resource/
o
g_ w energy use
E -c
— o
2 §?
c ro
£ °
0
« s r
.y t Certain ch
0 o. calculated
^ 0
0 fc.
Non-renewable
resource use
W i 1
.^ Renewable
§) resource use
0)
CD
^ Energy use
o
E
o
i^
1
Air
|
,r
Human J_
* health ^
1
Air
resources
Water
i
r
Landfil
1
Water
resources
Ecoto
^^^^M
emical properties indicate wh ch specific im
(e.g , greenhouse gases wil
i r ^
1 ^Occupational
non-cancer
Occupational
' " cancer
Public
non-cancer
Public cancer
o
CD
Q.
CO
\\ave a globbal
r
Global
warming
Stratospheric
ozone depl.
Photochem.
smog
Air
acidification
Air
particulates
Eutrc
Wat
Outputs
>act categories
varming impact
'Equations for calculating impact scores for each category are provided in Section 3.2
Figure 3-1. Impact Classification Conceptual Model
5-4
-------
Table 3-1. Inventory types and properties for classifying inventory items
into impact categories
Inventory type
Input
Output
Chemical/Material properties
Impact category
Natural Resource Impacts
Material, fuel
Material, water
Electricity, fuel
—
—
—
—
waste to landfill
Non-renewable
Renewable
Energy
Solid, hazardous, and radioactive waste
Non-renewable resource
use/depletion
Renewable resource use
Energy use
Landfill space use (volume)
Abiotic Ecosystem Impacts
—
—
—
—
—
—
—
Air
Air
Air
Air
Air
Water
Water
Global warming gases
Ozone depleting substances
Substances that can be photochemically
oxidized
Substances that react to form hydrogen
ions (H+)
Airparticulates (PM10, TSP) a
Substances that contain available nitrogen
or phosphorus
BODaandTSSa
Global warming
Stratospheric ozone depletion
Photochemical smog
Acidification
Air particulates
Water eutrophication (nutrient
enrichment)
Water quality
Human Health and Ecotoxicity
Material
—
Material
—
—
Air, soil, water
Air, soil, water
Water
Toxic material (carcinogenic)
Toxic material (carcinogenic)
Toxic material (non-carcinogenic)
Toxic material (non-carcinogenic)
Toxic material
Carcinogenic human health
effects — occupational
Carcinogenic human health
effects — public
Chronic, non-carcinogenic
human health effects —
occupational
Chronic, non-carcinogenic
human health effects — public
(and terrestrial ecotoxicity)
Aquatic ecotoxicity
a Acronyms: particulate matter with average aerodynamic diameter less than 10 micrometers (PM10); total suspended
particulates (TSP); biological oxygen demand (BOD); total suspended solids (TSS).
3-5
-------
3.1.2 Characterization
The characterization step of LCIA includes the conversion and aggregation of LCI results to
common units within an impact category. Different assessment tools are used to quantify the
magnitude of potential impacts, depending on the impact category. Three types of approaches
are used in the characterization method for the LFSP:
•• Loading—An impact score is based on the inventory amount.
•• Equivalency—An impact score is based on the inventory amount weighed by a certain
effect, equivalent to a reference chemical.
Full equivalency—all substances are addressed in a unified, technical model.
Partial equivalency—a subset of substances can be converted into equivalency
factors.
•• Scoring of inherent properties—An impact score is based on the inventory amount
weighed by a score representing a certain effect for a specific material (e.g., toxicity
impacts are weighed using a toxicity scoring method).
Table 3-2 lists the characterization approach used with each impact category. The
loading approach either uses the direct inventory amount to represent the impact or slightly
modifies the inventory amount to change the units into a meaningful loading estimate, such as
characterizing the impact of either non-renewable resource depletion or landfill use. Use of
nonrenewable resources is directly estimated as the mass loading (input amount) of that material
consumed; use of landfill space applies the mass loading (output amount) of hazardous, non-
hazardous, or radioactive waste, and converts that loading into a volume to estimate the landfill
space consumed.
The equivalency method uses equivalency factors in certain impact categories to convert
inventory amounts to common units relative to a reference chemical. Equivalency factors are
values that provide a measure (weighting) to relate the impact of an inventory amount of a given
chemical to the effect of the same amount of the reference chemical. For example, for the
impact category "global warming potential (GWP)," the equivalency factor is an estimate of a
chemical's atmospheric lifetime and radiative forcing that may contribute to global climate
change compared to the reference chemical carbon dioxide (CO2); therefore, GWPs are given in
units of CO2 equivalents.
Scoring of inherent properties is applied to impact categories that may have different
effects for the same amount of various chemicals, but for which equivalency factors do not exist
or are not widely accepted. The scores are meant to normalize the inventory data to provide
measures of potential impacts. Scoring methods are employed for the human and ecological
toxicity impact categories, based on the Chemical Hazard Evaluation Management Strategies
(CHEMS-1) method described by Swanson et al. (1997) and presented below. The scoring
method provides a relative score, or hazard value, for each potentially toxic material that is then
multiplied by the inventory amount to calculate the toxicity impact score.
Using the various approaches, the LFSP LCIA method calculates impact scores for each
inventory item for each applicable impact category. These impact scores are based on either a
direct measure of the inventory amount or some modification (e.g., equivalency or scoring) of
3-6
-------
that amount based on the potential effect the inventory item may have on a particular impact
category. Impact scores are then aggregated within each impact category to calculate the
various life-cycle impact category indicators.
Inventory amounts are identified on a functional unit basis and used to calculate impact
scores. For each inventory item, an individual score is calculated for each applicable impact
category. The detailed characterization equations for each impact category are presented in
Sections 3.2.1 through 3.2.13 and summarized in Section 3.4. The equations presented in those
subsections calculate impacts for individual inventory items that could later be aggregated as
defined by the user. Impact scores represent relative and incremental changes rather than
absolute effects or threshold levels.
Table 3-2. LCIA characterization approaches for the LFSP
Impact category
Characterization approach
Natural Resource Impacts
Non-renewable materials use/depletion
Renewable resource use
Energy use
Landfill space use
Loading
Loading
Loading
Loading
Abiotic Ecosystem Impacts
Global wanning
Stratospheric ozone depletion
Photochemical smog
Acidification
Air particulates
Water eutrophication (nutrient enrichment)
Water quality (BOD, TSS)
Equivalency (full)
Equivalency (full)
Equivalency (partial)
Equivalency (full)
Loading
Equivalency (partial)
Loading
Human Health and Ecotoxicity
Cancer human health effects — occupational
Cancer human health effects — public
Chronic non-cancer human health effects — occupational
Chronic non-cancer human health effects — public
Aquatic ecotoxicity
Scoring of inherent properties
Scoring of inherent properties
Scoring of inherent properties
Scoring of inherent properties
Scoring of inherent properties
5-7
-------
3.2 CHARACTERIZATION AND RESULTS
This section presents the impact assessment characterization methods and the impact
results by impact category. Within each impact category subsection (3.2.1 through 3.2.13), the
characterization equations are presented, followed by both the paste and bar solder results. A
discussion of the limitations and uncertainties associated with that impact category concludes
each section. The LCIA results are based on the boundaries outlined in Chapter 1 and the
inventory described in Chapter 2. Within the results subsections of Sections 3.2.1 through
3.2.13, the impacts are presented by life-cycle stage as well as by process. Individual flows that
are the greatest contributors to the life-cycle impacts also are presented. Section 3.4 briefly
summarizes the characterization methods and the overall life-cycle impact category indicators
for the sixteen impact categories for both the paste and bar alloys. A summary of the limitations
and uncertainties also is provided in Section 3.4.
For results presented at the process level, processes that consume energy (e.g., electricity
during solder application) are presented together as a process group with the associated
processes of electricity generation or fuel production. Table 3-3 lists the processes that are
grouped together as presented in Sections 3.2.1 through 3.2.13. Note that the metals extraction
and processing (ME&P) processes are not included in this list because they are from secondary
data that incorporate electricity generation and fuel production into the individual processes
themselves. Thus, the ME&P processes inherently include upstream energy sources.
The associated fuels for each process, as described above, also are depicted in the process
flow charts of the solder life-cycles in the figures in Chapter 2. For the upstream metals
production processes, fuel or energy production data are embedded in the inventories for those
processes. Fuel and energy production are included in the upstream results, but are not shown as
separate processes in the life-cycle process models shown in the figures in Chapter 2.
It should be reiterated that the LCIA results presented throughout this section are
indicators of the relative potential impacts of SnPb and the lead-free solders in various impact
categories and are not a measure of actual or specific impacts. The LCIA is intended to provide
a screening level evaluation of impacts and in no way provides absolute values or measures
actual effects. Results herein are referred to as impact category indicators (representing the total
impact score of an alloy in an impact category), impact results, impact scores, or simply impacts.
Each of these terms refers to relative potential impacts and should not be confused with an
assessment of actual impacts.
-------
Table 3-3. Process groups
Process group
Solder manufacturing
Post-industrial
recycling
Solder application
Landfilling
Incineration
Demanufacturing
Copper smelting
Associated processes
Paste solder
Paste solder manufacturing
Electric power production
Natural gas production
Heavy fuel oil (#6) production
Post-industrial recycling
Electric power production
Heavy fuel oil (#6) production
Light fuel oil (#2) production
LPG production
Reflow solder application on a PWB
Electric power production
Landfilling
Diesel fuel production
Incineration
Natural gas production
Demanufacturing
Electric power production
Copper smelting
Electric power production
Heavy fuel oil (#6) production
Light fuel oil (#2) production
LPG production
Bar solder
Bar solder manufacturing
Electric power production
Natural gas production
Heavy fuel oil (#6) production
Liquified petroleum gas (LPG) production
Post-industrial recycling
Electric power production
Heavy fuel oil (#6) production
Light fuel oil (#2) production
LPG production
Wave solder application on a PWB
Electric power production
Landfilling
Diesel fuel production
Incineration
Natural gas production
Demanufacturing
Electric power production
Copper smelting
Electric power production
Heavy fuel oil (#6) production
Light fuel oil (#2) production
LPG production
3-9
-------
3.2.1 Resource Use (Non-renewable and Renewable)
3.2.1.1 Characterization
Natural resources are materials that are found in nature in their basic form rather than
being manufactured. Non-renewable ("stock") natural resources are typically abiotic, such as
mineral ore or fossil fuels. Impacts to both of these natural resource types are calculated using
the loading approach (described in Section 3.1.2). Renewable ("flow") natural resources are
those that can be regenerated, typically biotic resources, such as forest products or other plants,
animal products, and water. Consumption impacts from non-renewable resources (NRRs) and
renewable resources (RRs) are calculated using direct consumption values (e.g., material mass)
from the inventory.
For the non-renewable materials use/depletion category, depletion of materials results
from the extraction of non-renewable resources. Non-renewable resource impact scores are
based on the amount of material inputs (which can be product or process materials), water, and
fuel inputs of non-renewable materials. To calculate the loading-based impact scores, the
following equation is used:
(ISNRP)t = [AmtmRx(l-RC)]t
where:
ISjwx equals the impact score for use of non-renewable resource / (kg) per functional
unit;
AmtmR equals the inventory input amount of non-renewable resource /' (kg) per functional
unit; and
RC equals the fraction recycled content (post-industrial and post-consumer) of
resource /'.
Renewable resource impact scores are based on the following process inputs in the LCI:
material inputs (which can be product or process materials), water, and fuel inputs of renewable
materials. To calculate the loading-based impact scores, the following equation is used:
(ISRR),=[AmtRRX(l-RC)],
where:
ISm equals the impact score for use of renewable resource /' (kg) per functional unit;
Amtm equals the inventory input amount of renewable resource / (kg) per functional
unit; and
RC equals the fraction recycled content (post-industrial and post-consumer) of
resource /'.
3-10
-------
Depletion of materials, which results from the extraction of renewable resources faster
than they are renewed, may occur, but is not specifically modeled or identified in the renewable
resource impact score.
3.2.1.2 Paste solder results
Total Resource Use Impacts by Life-Cycle Stage (Paste Solder)
Table 3-4 and Figure 3.2 present the solder paste results for NRR use impacts by
life-cycle stage. Table 3-5 and Figure 3.3 present the solder paste results for RR use impacts by
life-cycle stage. The tables list the impact scores per functional unit for the life-cycle stages of
each alloy, as well as the percent contribution of each life-cycle stage to the total impacts for
each alloy.
Table 3-4. NRR use impacts by life-cycle stage (paste solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
4.79E+01 2.97
1.89E+01 1.17
1.55E+03 95.8
1.23E+00 0.0761
1.61E+03 100
SAC
Score* %
3.43E+02 18.9
2.04E+01 1.12
1.45E+03 79.9
1.06E+00 0.0586
1.82E+03 100
BSA
Score* %
6.15E+02 34.9
1.15E+01 0.65
1.14E+03 64.5
-3.35E-02 -0.0019
1.76E+03 100
SABC
Score* %
2.42E+02 14.1
2.04E+01 1.19
1.46E+03 84.7
1.07E+00 0.0620
1.72E+03 100
*The impact scores are in units of kilograms of resources/1,000 cc of solder applied to a printed wiring board.
2 500
2,000
*j
'c
D
15
c 1 500
£ I.3UU
4-1
o
c
jo 1 000
-Z.
D)
"* 500
0 -
—
mnnnnnnnnnnnnnnnn
SnPb SAC BSA SABC
• EOL
D Use/application
• Mfg
D Upstream
Figure 3-2. Solder Paste Total Life-Cycle Impacts: NRR Use
3-11
-------
SAC solder paste has the greatest impact category indicator for NRR use at 1,820 kg of
NRR per functional unit, closely followed by BSA and SABC at 1,760 and 1,720 kg of NRR per
functional unit, respectively. The indicators for all three lead-free alloys exceed the NRR impact
category indicator for SnPb (1,610 kg/functional unit), but only by about 8 to 14 percent1. As
shown in the table and figure, the use/application stage dominates NRR use impacts for all of the
solders, accounting for 65 to 96 percent of NRR use depending on the alloy. The impact scores
from the use/application stage include resources consumed to generate electricity for solder
application. The upstream life-cycle stage (ME&P) is the second greatest contributor to NRR
use for all alloys, accounting for approximately 3 to 35 percent of the total score, depending on
the alloy. The manufacturing stage, which includes solder paste manufacturing and post-
industrial recycling, contributes minor amounts (approximately 1 percent). The EOL stage is a
negligible contributor (less than 0.1 percent) to the overall life-cycle impacts for each alloy.
An interesting note is that although SnPb has the lowest overall NRR impacts compared
to all the alternatives, it has the greatest impact from the use/application stage (1,550
kg/functional unit), which is the dominant stage for all of the alloys. This is due to the fact that
more electricity is required to reflow 1,000 cc of SnPb solder than the lead-free alloys. Although
the melting point of SnPb is lower than SAC and SABC, which taken alone would result in
lower energy needs for reflow, the energy requirements on a functional unit basis are greater
since SnPb is more dense (e.g., more mass per unit volume of solder is applied to a board).
Despite the fact that SnPb has the highest NRR impacts from application, the contribution from
upstream processes are greater for the lead-free alternatives than for SnPb, resulting in total NRR
impacts for all three alternatives that exceed that of SnPb.
Table 3-5 and Figure 3-3, which present RR use impacts, show a different trend than the
NRR impacts. The greatest RR impact category indicator is for SnPb at 34,800 kg/functional
unit. The SAC indicator is slightly less at 34,700 kg/functional unit and the SABC indicator
follows at 34,100 kg/functional unit. BSA has the lowest total impact score at
26,400 kg/functional unit. The use/application stage dominates each alloy's life-cycle RR use
impacts, accounting for 93 to 99 percent of the total scores. The upstream stage contributes
between 0.3 and 6 percent, and the solder manufacturing stage contributes approximately 1
percent to the overall life-cycle impacts of each alloy. The EOL stage is negligible compared to
the impact scores from the other stages (e.g., less than 0.1 percent for all).
lrThe actual difference in the scores from SnPb range from 110 kg to 210 kg of NRR per 1,000 cc
of solder applied. To help put this in perspective, say those 110 to 210 kg were made entirely of
automobile gasoline, then the amount can be equated to 39 to 75 gallons of automobile gasoline
(assuming a density of 2.79 kg/gal). If a driver consumes 20 gallons per week, this would be equivalent
to approximately 2 to 4 weeks of driving a car.
3-12
-------
Table 3-5. RR use impacts by life-cycle stage (paste solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
9.60E+01 0.276
3.70E+02 1.062
3.43E+04 98.6
2.75E+01 0.0791
3.48E+04 100
SAC
Score* %
2.04E+03 5.87
3.98E+02 1.15
3.22E+04 92.9
2.38E+01 0.0687
3.47E+04 100
BSA
Score* %
l.OOE+03 3.79
2.25E+02 0.852
2.52E+04 95.3
3.52E+00 0.0133
2.64E+04 100
SABC
Score* %
1.32E+03 3.86
3.98E+02 1.17
3.23E+04 94.9
2.39E+01 0.0702
3.41E+04 100
*The impact scores are in units of kilograms of resources/1,000 cc of solder applied to a printed wiring board
45 000
40 000
-3c nnn
'E
= 30 000
ro
Ooc nnn
'S
u
c 90 000
£ -ic nnn
D)
^ 10000-
5 000
0 -
SnPb
SAC
BSA
• EOL
n Use/application
• Mfg
H Upstream
SABC
Figure 3-3 Solder Paste Total Life-Cycle Impacts: RRUse
Similar to the NRR use impacts, SnPb has the highest RR impacts from the
use/application stage alone; however, the upstream impacts from SAC and SABC cause their
total impact scores to slightly exceed that of SnPb. Although BSA's upstream impact score
exceeds that of SnPb, BSA still has a smaller total score.
Resource Use Impacts by Process Group (Paste Solder)
Table 3-6 lists the NRR use impacts for the process groups in the life-cycle of a solder.
In addition to production processes typically associated with solder manufacturing, process
groups include fuel or energy production associated with a particular process (see Table 3-3).
Impacts from the use/application stage, which is the dominant stage contributing to the life-cycle
impacts, are due entirely to the production of electricity for the solder reflow process.
Upstream impacts arise from the materials consumed in the extraction and processing of
the various metals present in the alloys. Of note is that bismuth production for the BSA alloy is
3-13
-------
the single greatest contributor to upstream NRR use for all of the alloys (507 kg/functional unit),
causing BSA to exceed the impact scores of the other three alloys in the upstream stage. As a
result, bismuth production (which contributes 27 percent to the overall life-cycle impacts of
BSA), and to a much lesser degree, silver production (which contributes 5 percent) cause BSA's
overall NRR impacts to exceed SnPb.
Table 3-6. NRR use impacts by life-cycle stage and process group (paste solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
BSA
Score* %
SABC
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Bi production
Total
2.34E+01 1.31
2.45E+01 1.37
N/A N/A
N/A N/A
N/A N/A
4.79E+01 2.68
3.43E+01 1.72
N/A N/A
3.03E+02 15.2
6.00E+00 0.300
N/A N/A
3.43E+02 17.2
1.76E+01 0.929
N/A N/A
9.04E+01 4.79
N/A N/A
5.07E+02 26.8
6.15E+02 32.6
3.46E+01 1.82
N/A N/A
1.95E+02 10.2
5.02E+00 0.264
7.67E+00 0.403
2.42E+02 12.7
MANUFACTURING
Solder
manufacturing
Post-industrial
recycling
Total
6.81E+00 0.381
1.21E+01 0.679
1.89E+01 1.06
1.04E+01 0.519
1.01E+01 0.504
2.04E+01 1.02
6.54E+00 0.346
4.96E+00 0.263
1.15E+01 0.609
1.04E+01 0.547
l.OOE+01 0.527
2.04E+01 1.07
USE/APPLICATION
Solder application
Total
1.72E+03 96.2
1.72E+03 96.2
1.63E+03 81.7
1.63E+03 81.7
1.26E+03 66.8
1.26E+03 66.8
1.64E+03 86.1
1.64E+03 86.1
END-OF-LIFE
Landfill
Incineration
Demanufacturing
Cu smelting
Unregulated
Total
GRAND
TOTAL
4.96E-02 0.00278
-2.26E-01 -0.0126
1.52E-01 0.00851
1.25E+00 0.0702
O.OOE+00 0.00
1.23E+00 0.0688
1.79E+03 100
4.29E-02 0.00215
-1.95E-01 -0.0098
1.32E-01 0.00659
1.08E+00 0.0543
O.OOE+00 0.00
1.06E+00 0.0533
2.00E+03 100
5.31E-02 0.00281
-2.42E-01 -0.0128
1.55E-01 0.00820
N/A N/A
O.OOE+00 0.00
-3.35E-02 -0.0018
1.89E+03 100
4.31E-02 0.00227
-1.96E-01 -0.0103
1.32E-01 0.00695
1.09E+00 0.0573
O.OOE+00 0.00
1.07E+00 0.0562
1.90E+03 100
*The impact scores are in units of kilograms of resources/1,000 cc of solder applied to a printed wiring board.
N/A=not applicable
Silver production contributes significantly to the upstream impacts for SAC and SABC,
causing these alloys to have greater total impacts than SnPb. Silver processing in SAC and
SABC dominates the upstream impacts, even though silver comprises a much smaller percentage
of the overall alloy content than tin. For example, SAC is 95.5 percent tin (Sn) and only 3.9
percent silver (Ag), yet its impacts from silver production are far greater than those from tin
production (15 percent of total NRR impacts for silver versus 2 percent for tin). This illustrates
the relatively high resource consumption of silver extraction and processing compared to the
other solder metals. For BSA, the NRR impacts from silver processing account for about 5
3-14
-------
percent of total impacts compared to about 27 percent for bismuth processing. In this case,
BSA's impacts from silver processing are disproportionately higher than its silver content, but
less so than with SAC and SABC. BSA contains 57 percent Bismuth (Bi) and 1 percent Ag.
Manufacturing impacts are small compared to the upstream and use/application life-cycle
stages, and are nearly evenly distributed between solder manufacturing and post-industrial
recycling for the lead-free alternatives. SnPb, on the other hand, consumes almost 80 percent
more NRR in post-industrial recycling than in solder manufacturing. The differences in the
distribution of impacts between solder manufacturing and post-industrial recycling among the
alloys are due to two factors: (1) there are varying amounts of secondary alloy used in
manufacturing each of the alloys, and (2) the alloys have different melting temperatures that
affect their relative resource use. SnPb has the greatest amount of secondary alloy used in
manufacturing and requires more post-industrial recycling than the lead-free alloys; however,
SAC and SABC have higher melting points and, therefore, require more resources per unit of
secondary alloy produced. Although BSA has a lower melting point than SnPb, data were not
obtained on the resulting differences in resource inputs for post-industrial recycling of BSA; the
inputs were assumed to be the same as for SnPb (this is considered a conservative estimate since
the melting point of SnPb is higher than that of BSA). A more detailed discussion of this
assumption is presented in Section 2.3.
EOL processes contribute less than 0.08 percent of life-cycle NRR impacts for all of the
solders, with the majority of the SnPb, SAC, and SABC EOL impact scores coming from
smelting processes to recover copper and other valuable metals from waste electronics. No
impacts are shown for copper smelting of BSA-containing PWBs because the LFSP LCA
assumes these boards are not sent to copper smelting facilities at EOL. Copper smelting is not
included in the BSA inventory since its bismuth content exceeds allowable bismuth levels at
these facilities (see Chapter 2). Negative impacts from incineration are due to an energy credit
for incineration, which creates negative impacts from natural gas production. No resource
impacts are shown for unregulated disposal, as the inventory for this process did not include any
resource inputs; however, some energy is consumed when waste PWBs are heated to recover
solder and components. The amount of energy and associated resources consumed in this
process are not known, but they are expected to be small.
Table 3-7 lists the RR use impacts for the process groups in the life-cycle of a solder. As
with the NRR use category, impacts from the use/application stage dominate the life-cycle
impacts and are due entirely to production of electricity consumed during the solder reflow
process.
Upstream impacts arise from the materials consumed in the extraction and processing of
the various metals present in the alloys. Silver production dominates the upstream impacts of the
silver-containing alloys, despite their relatively low silver content. In addition, the impact scores
related to silver processing range from 607 kg/functional unit to 2,030 kg/functional unit,
depending on the silver-bearing alloy, while the impact scores from lead in the SnPb alloy are
only 96 kg/functional unit.
Manufacturing impacts are small compared to the upstream and use/application life-cycle
stages, and are nearly evenly distributed between solder manufacturing and post-industrial
recycling for SAC and SABC. SnPb has twice as many RR impacts from post-industrial
3-15
-------
recycling than from solder manufacturing. BSA, on the other hand, consumes about 23 percent
more RR in manufacturing than in post-industrial recycling. As explained above, the
discrepancy in the distribution of impacts between SnPb and the lead-free alloys is because SnPb
uses more secondary alloy than BSA. In addition, although less secondary alloy is used for
manufacturing SAC and SABC, the impacts are affected by the difference in melting
temperatures (e.g., SAC and SABC have higher melting temperatures and consume more
resources per unit of secondary alloy produced).
Table 3-7. RR use impacts by life-cycle stage and process group (paste solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
BSA
Score* %
SABC
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Bi production
Total
3.68E-02 0.0001
9.59E+01 0.248
N/A N/A
N/A N/A
N/A N/A
9.60E+01 0.248
5.38E-02 0.0001
N/A N/A
2.03E+03 5.26
3.56E+00 0.0092
N/A N/A
2.04E+03 5.27
2.76E-02 0.0001
N/A N/A
6.07E+02 2.08
N/A N/A
3.95E+02 1.35
l.OOE+03 3.43
5.44E-02 0.0001
N/A N/A
1.31E+03 3.43
2.98E+00 0.0078
5.97E+00 0.0157
1.32E+03 3.46
MANUFACTURING
Solder
manufacturing
Post-industrial
recycling
Total
1.22E+02 0.316
2.48E+02 0.641
3.70E+02 0.957
2.06E+02 0.532
1.92E+02 0.498
3.98E+02 1.03
1.24E+02 0.424
1.01E+02 0.347
2.25E+02 0.770
2.06E+02 0.542
1.92E+02 0.504
3.98E+02 1.05
USE/APPLICATION
Solder application
Total
3.81E+04 98.7
3.81E+04 98.7
3.62E+04 93.6
3.62E+04 93.6
2.80E+04 95.8
2.80E+04 95.8
3.63E+04 95.4
3.63E+04 95.4
END-OF-LIFE
Landfill
Incineration
Demanufacturing
Cu smelting
Unregulated
Total
GRAND TOTAL
9.84E-02 0.0003
-1.77E-02 -0.00005
3.37E+00 0.0087
2.41E+01 0.0624
O.OOE+00 0.00
2.75E+01 0.0713
3.86E+04 100
8.52E-02 0.0002
-1.53E-02 -0.00004
2.92E+00 0.0076
2.08E+01 0.0539
O.OOE+00 0.00
2.38E+01 0.0617
3.87E+04 100
1.05E-01 0.0004
-1.89E-02 -0.0001
3.44E+00 0.0118
N/A N/A
O.OOE+00 0.00
3.52E+00 0.0121
2.92E+04 100
8.55E-02 0.0002
-1.54E-02 -0.00004
2.93E+00 0.0077
2.09E+01 0.0549
O.OOE+00 0.00
2.39E+01 0.0628
3.81E+04 100
*The impact scores are in units of kilograms of resources/1,000 cc of solder applied to a printed wiring board.
N/A=not applicable
EOL processes contribute less than 0.08 percent of life-cycle RR impacts for any of the
solders, with the majority of SnPb, SAC, and SABC impacts coming from the smelting processes
used to recover copper and other valuable metals from waste electronics. As noted previously,
the copper smelting process is not included in the BSA inventory. Negative impacts from
incineration are due to the energy credit for incineration with energy recovery. No resource
impacts are shown for unregulated disposal as the inventory for this process did not include any
resource inputs. Some energy is consumed, however, when waste PWBs are heated to recover
3-16
-------
solder and components. The amount of energy and associated resources consumed in this
process are not known, but they are expected to be small compared to other processes.
Top Contributors to Resource Use Impacts (Paste Solder)
Table 3-8 presents the specific materials or flows contributing greater than or equal to 1
percent of NRR use impacts by solder. As expected from the results presented above, the
materials used to produce electricity in the use/application stage are the top contributors to
overall NRR impacts, with inert rock being the single greatest contributor for all of the solders
and hard coal being the second greatest for all alloys, except BSA. Copper ore from bismuth
production is the flow with the second greatest contribution to BSA impacts at 24 percent. In
addition to resources used to generate electricity in the use/application stage and bismuth
production for BSA, input flows from silver production are major contributors to NRR impacts
for the lead-free alloys.
Table 3-8. Top contributors to NRR use impacts (paste solder)
Solder
SnPb
SAC
BSA
SABC
Life-Cycle Stage
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Upstream
Upstream
Use/application
Use/application
Upstream
Upstream
Use/application
Upstream
Use/application
Upstream
Use/application
Upstream
Use/application
Upstream
Upstream
Use/application
Use/application
Upstream
Upstream
Use/application
Use/application
Process
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Silver Production
Silver Production
Electricity generation
Electricity generation
Silver Production
Silver Production
Electricity generation
Bismuth Production
Electricity generation
Silver Production
Electricity generation
Silver Production
Electricity generation
Bismuth Production
Bismuth Production
Electricity generation
Electricity generation
Silver Production
Silver Production
Electricity generation
Electricity generation
Flow
Inert rock
Hard coal (resource)
Lignite (resource)
Natural gas (resource)
Inert rock
Hard coal (resource)
Zinc-lead-copper ore (12%-3%-2%)
Inert rock
Lignite (resource)
Natural gas (resource)
Limestone (calcium carbonate)
Hard coal (resource)
Inert rock
Copper ore (0.14%)
Hard coal (resource)
Zinc - lead - copper ore (12%-3%-2%)
Lignite (resource)
Inert rock
Natural gas (resource)
Zinc - copper ore (4.07%-2.59%)
Lead - zinc ore (4.6%-0.6%)
Inert rock
Hard coal (resource)
Zinc - lead - copper ore (12%-3%-2%)
Inert rock
Lignite (resource)
Natural gas (resource)
% Contribution
76.8
13.4
2.72
2.11
64.1
11.2
7.61
5.15
2.27
1.76
1.27
1.00
51.7
24.4
9.01
2.34
1.83
1.59
1.42
1.33
1.02
67.9
11.8
5.17
3.50
2.40
1.86
Table 3-9 presents the specific materials or flows contributing greater than or equal to
3-17
-------
percent of RR use impacts by solder. The top RRs are water and air. As expected from the RR
results presented above, resources from electricity production in the use/application stage are the
top contributors to overall RR impacts. Water is the single greatest contributor for all of the
solders ranging from 84 to 89 percent of all impacts for each alloy. Water consumed in silver
and bismuth production also is a top contributor for the lead-free alloys, but the contribution to
total impacts for any alloy is less than 6 percent.
Table 3-9. Top contributors to RR use impacts (paste solder)
Solder
SnPb
SAC
BSA
SABC
Life-Cycle Stage
Use/application
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
Upstream
Upstream
Use/application
Use/application
Upstream
Process
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Silver Production
Electricity generation
Electricity generation
Silver Production
Bismuth Production
Electricity generation
Electricity generation
Silver Production
Flow
Water
Air
Water
Air
water
Water
Air
Water
Water
Water
Air
Water
% Contribution
88.8
9.79
83.7
9.22
5.33
85.9
9.46
2.09
1.21
85.5
9.42
3.49
3.2.1.3 Bar solder results
Total Resource Use Impacts by Life-Cycle Stage (Bar Solder)
Table 3-10 and Figure 3-4 present the bar solder results for NRR use impacts by life-
cycle stage. Table 3-11 and Figure 3-5 present the bar solder results for RR use impacts by life-
cycle stage. The tables list the impact scores per functional unit for the life-cycle stages of each
alloy, as well as the percent contribution of each life-cycle stage to the total impacts for each
alloy.
Table 3-10. NRR use impacts by life-cycle stage (bar solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
4.46E+01 14.2
2.40E+01 7.63
2.45E+02 77.8
1.38E+00 0.438
3.15E-KJ2 100
SAC
Score* %
5.08E+02 66.1
1.16E+01 1.50
2.48E+02 32.2
1.21E+00 0.157
7.68E+02 100
SnCu
Score* %
4.70E+01 15.1
1.63E+01 5.23
2.48E+02 79.3
1.20E+00 0.384
3.12E+02 100
*The impact scores are in units of kilograms of resources/1,000 cc of solder applied to a printed wiring board.
3-18
-------
Qnn
800 -
7DD
4-1
i 600 -
15
o 500 -
•&
o
5 Ann
2
O)
^ 200 -
100 -
n
^=
SnPb SAC SnCu
D End-of-life
D Use/application
• Manufacturing
n Upstream
Figure 3-4. Bar Solder Total Life-Cycle Impacts: NRR Use
As was found with the paste solder results, SAC bar solder has the greatest impact
category indicator for NRR use. The SAC NRR indicator score is 768 kg of NRR per functional
unit, followed by SnPb and SnCu at 315 and 312 kg of NRR per functional unit, respectively2.
As shown in the table and figure, the upstream stage dominates NRR use impacts for SAC (66
percent), while the use/application stage dominates impacts for SnPb and SnCu. An interesting
note is that the use/application stage scores are nearly the same for all three alloys; however, the
greatest difference in the total impacts is due to the large impact from the upstream stage for
SAC.
Table 3-11 and Figure 3-5, which present RR use impacts, show a similar trend as the
NRR impacts in that SAC has the greatest impacts; however, for all three alloys, the
use/application stage dominates impacts (ranging from 63 to 94 percent), while the upstream
stage is an important contributor to the SAC total impact score (35 percent). As with the NRR
use impacts, the use/application stage scores are similar among the three alloys. The upstream
impacts from SAC result in a distinguishably greater total impact score compared to SnPb and
SnCu (i.e., 45 to 50 percent greater). The differences in absolute scores are 2,730 to 2,930 kg
per 1,000 cc of solder applied. To place this in perspective, it is equivalent to 721 to 744 gallons
of water (although the impacts are not comprised solely of water).
The difference between SAC and SnPb is 453 kg of NRR per 1,000 cc of solder applied. If this were all
automotive gasoline, this difference is equivalent to 162 gallons of gasoline. Assuming a driver consumes 20
gallons per week, this is also equivalent to approximately 8 weeks of driving.
3-19
-------
Table 3-11. RR use impacts by life-cycle stage (bar solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
8.56E+01 1.42
4.85E+02 8.04
5.43E+03 90.0
3.06E+01 0.507
6.03E+03 100
SAC
Score* %
3.02E+03 34.5
2.23E+02 2.55
5.49E+03 62.7
2.68E+01 0.305
8.76E+03 100
SnCu
Score* %
5.90E+00 0.101
3.06E+02 5.24
5.49E+03 94.2
2.66E+01 0.456
5.83E+03 100
*The impact scores are in units of kg of resources/1,000 cc of solder applied to a printed wiring board.
10 000
q nnn
8,000
- 7,000
c
ro 6,000
c
s 5,000
c
£; 4,000
ra 3,000
J£
9 nnn
1 000
0
=J
SnPb
_
—
SAC SnCu
n End-of-life
D Use/application
H Manufacturing
D Upstream
Figure 3-5. Bar Solder Total Life-Cycle Impacts: RR Use
Resource Use Impacts by Process Group (Bar Solder)
Table 3-12 lists the NRR use impacts for the process groups in the life-cycle of a solder.
In addition to production processes typically associated with solder manufacturing, process
groups include fuel or energy production associated with a particular process (Table 3-3).
Impacts from the use/application stage, which is the dominant stage contributing to the life-cycle
impacts, are due entirely to the production of electricity for the bar solder application process.
Upstream impacts arise from the materials consumed in the extraction and processing of
the various metals present in the alloys. Silver production contributes significantly to the
upstream impacts for SAC, causing this alloy to have distinguishably greater total impacts than
SnPb and SnCu. Silver processing in SAC dominates the upstream impacts, even though silver
comprises a much smaller percentage of the overall alloy content than tin. For example, SAC is
3-20
-------
95.5 percent tin and only 3.9 percent silver, yet its impacts from silver production are far greater
than those from tin production (59 percent of total NRR impacts for silver versus 6 percent for
tin). This illustrates the relatively high resource consumption of silver extraction and processing
compared to the other solder metals.
As with the paste solder results, manufacturing impacts are small compared to the
upstream and use/application life-cycle stages, and are nearly evenly distributed between solder
manufacturing and post-industrial recycling for SAC. SnPb and SnCu, on the other hand,
consume more NRR in post-industrial recycling than in solder manufacturing. The differences
in the distribution of impacts between solder manufacturing and post-industrial recycling among
the alloys are due to two factors: (1) there are varying amounts of secondary alloy used in
manufacturing each of the alloys, and (2) the alloys have different melting temperatures that
affect their relative resource use. SnPb has the greatest amount of secondary alloy used in
manufacturing and requires more post-industrial recycling than the lead-free alloys; however,
SAC and SnCu have higher melting points and, therefore, require more resources per unit of
secondary alloy produced.
Table 3-12. NRR use impacts by life-cycle stage and process group (bar solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
SnCu
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Total
MANUFACTURING
Solder manufacturing
Post-industrial recycling
Total
2.28E+01 7.23
2.19E+01 6.94
N/A N/A
N/A N/A
4.46E+01 14.2
3.60E+00 1.14
2.04E+01 6.49
2.40E+01 7.63
4.82E+01 6.27
N/A N/A
4.49E+02 58.5
l.OOE+01 1.30
5.08E+02 66.1
5.47E+00 0.713
6.08E+00 0.792
1.16E+01 1.50
3.72E+01 11.9
N/A N/A
N/A N/A
9.83E+00 N/A
4.70E+01 15.1
5.89E+00 1.89
1.04E+01 3.34
1.63E+01 5.23
USE/APPLICATION
Wave application
Total
2.45E+02 77.8
2.45E+02 77.8
2.48E+02 32.2
2.48E+02 32.2
2.48E+02 79.3
2.48E+02 79.3
END-OF-LIFE
Landfill
Incineration
Demanufacture
Cu smelting
Unregulated
Total
GRAND TOTAL
5.51E-02 0.0175
-2.38E-01 -0.0755
1.69E-01 0.0537
1.39E+00 0.443
O.OOE+00 0.0000
1.38E+00 0.438
3.15E+02 100
4.83E-02 0.0063
-2.08E-01 -0.0271
1.48E-01 0.0192
1.22E+00 0.159
O.OOE+00 0.0000
1.21E+00 0.157
7.68E+02 100
4.79E-02 0.0153
-2.07E-01 -0.0662
1.47E-01 0.0470
1.21E+00 0.388
O.OOE+00 0.0000
1.20E+00 0.384
3.12E+02 100
*The impact scores are in units of kg resources/1,000 cc of solder applied to a printed wiring board.
N/A=not applicable
3-21
-------
As discussed with the paste solder results, EOL processes contribute a very small percent
(less than 0.6 percent) of life-cycle NRR impacts for all of the solders, with the majority of the
EOL impact scores coming from smelting processes to recover copper and other valuable metals
from waste electronics.
Table 3-13 lists the RR use impacts for the process groups in the life-cycle of a solder.
Impacts from the use/application stage dominate the life-cycle impacts and are due entirely to the
production of electricity consumed during the wave solder application process.
Table 3-13. RR use impacts by life-cycle stage and process group (bar solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
SnCu
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Total
3.58E-02 0.0006
8.56E+01 1.42
N/A N/A
N/A N/A
8.56E-K)! 1.42
7.57E-02 0.0009
N/A N/A
3.02E+03 34.4
5.95E+00 0.0679
3.02E+03 34.5
5.84E-02 0.0010
N/A N/A
N/A N/A
5.84E+00 N/A
5.90E+00 0.101
MANUFACTURING
Solder manufacturing
Post-industrial recycling
Total
6.80E+01 1.13
4.17E+02 6.92
4.85E+02 8.04
1.07E+02 1.22
1.16E+02 1.33
2.23E+02 2.55
1.06E+02 1.82
2.00E+02 3.42
3.06E+02 5.24
USE/APPLICATION
Wave application
Total
5.43E+03 90.0297
5.43E+03 90.0
5.49E+03 62.6721
5.49E+03 62.7
5.49E+03 94.1992
5.49E+03 94.2
END-OF-LIFE
Landfill
Incineration
Demanufacture
Cu smelting
Unregulated
Total
GRAND TOTAL
1.09E-01 0.0018
-1.86E-02 -0.0003
3.75E+00 0.0621
2.68E+01 0.4437
O.OOE+00 0.0000
3.06E-K)! 0.507
6.03E+03 100
9.57E-02 0.0011
-1.63E-02 -0.0002
3.28E+00 0.0374
2.34E+01 0.2672
O.OOE+00 0.0000
2.68E+01 0.305
8.76E+03 100
9.50E-02 0.0016
-1.62E-02 -0.0003
3.26E+00 0.0558
2.33E+01 0.3987
O.OOE+00 0.0000
2.66E+01 0.456
5.83E+03 100
*The impact scores are in units of kg resources/1,000 cc of solder applied to a printed wiring board.
N/A=not applicable
Upstream impacts arise from the materials consumed in the extraction and processing of
the various metals present in the alloys. Similar to the NRR results, silver production for SAC,
which constitutes 34 percent of total RR impacts, dominates the upstream impacts despite its
relatively low silver content.
Manufacturing impacts are small compared to the upstream and use/application life-cycle
stages, and are nearly evenly distributed between solder manufacturing and post-industrial
recycling for SAC. For SnPb and SnCu, the impacts are greater from post-industrial recycling
than they are from bar solder manufacturing.
EOL processes contribute less than 0.6 percent of life-cycle RR impacts for all of the
solders, with the majority of impacts coming from the smelting processes used to recover copper
3-22
-------
and other valuable metals from waste electronics (see the earlier discussion for paste and NRR
impacts, Section 3.2.1).
Top Contributors to Resource Use Impacts (Bar Solder)
Table 3-14 presents the specific materials or flows contributing greater than or equal to 1
percent of NRR use impacts by solder. As expected from the results presented above, the
materials used to produce electricity in the use/application stage are the top contributors to
overall NRR impacts for SnPb and SnCu, with inert rock being the single greatest contributor for
all of the solders and hard coal being the second greatest. The top two contributors to the SAC
impacts are zinc-lead-copper ore from silver production (27 percent) and inert rock from
electricity generation in the use/application stage (26 percent).
Table 3-14 Top contributors to NRR use impacts (bar solder)
Solder
SnPb
SAC
Life-Cycle Stage
Use/application
Use/application
Manufacturing
Upstream
Upstream
Use/application
Upstream
Use/application
Upstream
Upstream
Upstream
Upstream
Use/Application
Upstream
Use/Application
Upstream
Upstream
Upstream
Upstream
Upstream
Upstream
Upstream
Upstream
Upstream
Upstream
Process
Electricity generation
Electricity generation
Electricity generation for post-
industrial recycling
Lead production
Tin production
Electricity generation for solder
application
Tin production
Electricity generation for solder
application
Tin production
Tin production
Lead production
Silver production
Electricity generation
Silver production
Electricity generation
Silver production
Silver production
Silver production
Tin production
Tin production
Tin production
Silver production
Copper production
Silver production
Tin production
Flow
Inert rock
Hard coal (resource)
Inert rock
Lead - zinc ore (4.6%-0.6%)
Hard coal (resource)
Lignite (resource)
Natural gas (resource)
Natural gas (resource)
Crude oil (resource)
Tin ore
Inert rock
Zinc - lead - copper ore
Inert rock
Inert rock
Hard coal (resource)
Limestone (calcium
carbonate)
Hard coal (resource)
Quartz sand (silica sand;
silicon dioxide)
Hard coal (resource)
Natural gas (resource)
Crude oil (resource)
Crude oil (resource)
Copper ore (0.14%)
Soil
Tin ore
% Contribution
62.3
10.9
4.71
4.45
2.59
2.20
1.79
1.71
1.69
1.15
1.06
26.7
25.8
18.1
4.51
4.47
3.52
2.50
2.25
.56
.47
.32
.17
.09
.00
3-23
-------
Table 3-14 Top contributors to NRR use impacts (bar solder)
Solder
SnCu
Life-Cycle Stage
Use/application
Use/application
Upstream
Upstream
Upstream
Upstream
Use/application
Manufacturing
Upstream
Use/application
Manufacturing
Process
Electricity generation
Electricity generation
Tin production
Tin production
Copper production
Tin production
Electricity generation
Electricity generation for post-
industrial recycling
Tin production
Electricity generation
Electricity generation for solder
manufacturing
Flow
Inert rock
Hard coal (resource)
Hard coal (resource)
Natural gas (resource)
Copper ore (0.14%)
Crude oil (resource)
Lignite (resource)
Inert rock
Tin ore
Natural gas (resource)
Inert rock
% Contribution
63.5
11.1
4.26
2.95
2.83
2.78
2.25
2.25
1.90
1.74
1.12
Table 3-15 presents the specific materials or flows contributing greater than 1
percent of RR use impacts by solder. The top RRs are water and air. As expected from the RR
results presented above, resources from electricity production in the use/application stage are the
top contributors to overall RR impacts. Water from electricity generation for wave application is
the single greatest contributor for all of the solders ranging from 57 to 85 percent of all impacts
for each alloy. Water consumed in silver production also is a top contributor for SAC (31
percent).
Table 3-15. Top contributors to RR use impacts (bar solder)
Solder
SnPb
SAC
SnCu
Life-Cycle Stage
Use/application
Use/application
Manufacturing
Use/application
Upstream
Use/application
Upstream
Manufacturing
Manufacturing
Use/application
Use/application
Manufacturing
Manufacturing
Process
Electricity generation
Electricity generation
Electricity generation for post-industrial recycling
Electricity generation
Silver production
Electricity generation
Silver production
Electricity generation for post-industrial recycling
Solder manufacturing
Electricity generation
Electricity generation
Electricity generation for post-industrial recycling
Solder manufacturing
Flow
Water
Air
Water
Water
Water
Air
Air
Water
Water
Water
Air
Water
Water
% Contribution
81.1
8.94
6.13
56.5
31.3
6.22
3.13
1.16
1.00
84.8
9.35
3.00
1.50
3-24
-------
3.2.1.4 Limitations and uncertainties
The renewable and non-renewable resource use results presented here are based on the
mass of a material consumed. Depletion of renewable materials, which results from the
extraction of RRs faster than they are renewed may occur, but is not specifically modeled or
identified in the RR use impact scores. For the NRR use category, depletion occurs from the
extraction of these NRRs; however, the impact scores do not relate consumption rates to the
Earth's ability to sustain that consumption.
In the paste solder results, the SnPb and lead-free alloy impact scores for both NRR and
RR use are being driven by the electricity consumed to power a reflow solder oven in the
use/application stage. Electricity consumption data are based on the average of two
experimental reflow application runs conducted by the LFSP. The first experimental run was
conducted using a 1998 model reflow oven, which is less energy efficient than the 2002 model
oven used in the second run. These are primary data collected for the purposes of the LFSP
under controlled conditions and are considered to be of good quality. There is considerable
variation in the two data points (from 8,170 to 17,100 MJ per functional unit for SnPb, for
example), which introduces some uncertainty into the average value used in the LCIA. In
addition, while these two data points represent reasonable high and low values, the data are
limited. Section 3.3 presents the results of sensitivity analyses of the high and low electricity
consumption values for each alloy. Chapter 2 describes limitations and uncertainties in the
reflow electricity consumption data in more detail.
In the bar solder results, the energy from wave application also is a major contributor for
all alloys; however, silver production for SAC is another major contributor. The energy data
from wave application are primary data collected for this study and are expected to be
representative of general wave applications, although they are only from one data set. Another
source of uncertainty is that the electricity generation process used in this study is from
secondary data provided in the GaBi database. Data quality of the electricity generation
inventory, as determined by GaBi, is considered "good." In addition, an average U.S. electric
grid mix was selected for use in this study to conform and with the data collected from the solder
application process (all from the U.S.) and with the geographic boundaries of this study. As a
result, use of a secondary data set for electricity generation is not expected to be a large source of
uncertainty.
Finally, the secondary data used for silver production is another source of uncertainty.
This silver production process is a mix of global data from GaBi, and the data quality is
described as "good." Another available data set for silver production (Ecobilan, 1999) suggests
possibly significant variations between the two inventories. GaBi data were chosen for this
study in part because they were considered of good quality, are representative of relatively recent
data (1994-1995), were from the same source as most of the other upstream data sets used in this
study, and were from a company that could be easily contacted for questions regarding the data.
See Chapter 2 for the discussion on upstream inventory data. Because life-cycle impacts in this
and several other impact categories are largely being driven by the inventory for silver
production, the DEAM data are used in an alternate analysis to determine the sensitivity of
overall LCIA results to the silver inventory. This is discussed further in Section 3.3.
3-25
-------
3.2.2 Energy Use
3.2.2.1 Characterization
General energy consumption is used as an indicator of potential environmental impacts
from the entire energy generation cycle. Energy use impact scores are based on both feel and
electricity flows. The impact category indicator is the sum of electrical energy inputs and fuel
energy inputs. Fuel inputs are converted from mass to energy units using the fuel's heat value
(H) and the density (D), presented in Appendix G. The impact score is calculated by:
(IS,), =(AmtE), or [Amtpx(H/D)]t
where:
ISE equals the impact score for energy use (MJ) per functional unit;
AmtE equals the inventory input amount of electrical energy used (MJ) per functional
unit;
AmtF equals the inventory input amount of fuel used (kg) per functional unit;
H equals the heat value of fuel /' (MJ/L); and
D equals the density of fuel / (kg/L).
This category addresses energy use only. The emissions from energy production are
outputs from the energy production process and are classified to applicable impact categories,
depending on the disposition and chemical properties of the outputs (see Classification Section
3.1.1).
3.2.2.2 Paste solder results
Total Energy Use Impacts by Life-Cycle Stage (Paste Solder)
Table 3-16 presents the solder paste results for energy use impacts by life-cycle stage,
based on the impact assessment methodology presented above. Figure 3-6 presents the results
in a stacked bar chart. General energy consumption is used as an indicator of potential
environmental impacts from the entire energy generation cycle.
Table 3-16. Energy use impacts by life-cycle stage (paste solder)
Life-Cycle Stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
8.67E+02 6.94
2.13E+02 1.70
1.14E+04 91.2
1.68E+01 0.135
1.25E+04 100
SAC
Score* %
2.61E+03 19.3
2.29E+02 1.69
1.07E+04 78.9
1.46E+01 0.107
1.36E+04 100
BSA
Score* %
1.25E+03 12.8
1.29E+02 1.33
8.37E+03 85.8
2.49E+00 0.0255
9.76E+03 100
SABC
Score* %
2.12E+03 16.2
2.29E+02 1.75
1.07E+04 82.0
1.46E+01 0.112
1.31E+04 100
*The impact scores are in units of megajoules/1,000 cc of solder applied to a printed wiring board.
3-26
-------
SAC solder paste has the greatest impact category indicator for energy use at 13,600 MJ
per functional unit, closely followed by SABC at 13,100 MJ, and SnPb at 12,500 MJ. BSA is
the only solder paste that consumes substantially less energy (9,760 MJ per functional unit),
primarily due to its lower melting temperature that significantly reduces energy consumption
during solder application. For a relative comparison, the average U.S. household consumes
approximately 9,244 MJ of energy per month (DOE, 2003). As shown in the table and figure, the
use/application stage dominates energy use impacts for all of the solders, accounting for 79 to 91
percent of energy use depending on the alloy.
16 000
14 000
19 nnn
4-1
'c 10 000
15
5 a nnn
t;
E c nnn
^
A nnn
9 nnn
n
SnPb SAC BSA SABC
• End-of-lif e
n Use/application
H Manufacturing
H Upstream
Figure 3-6. Paste Solder Total Life-Cycle Impacts: Energy Use
SnPb, which has a higher melting temperature than BSA but a lower melting temperature
than SAC and SABC, requires the most energy in the use/application stage (11,400
MJ/functional unit). This phenomenon is due to the greater density of the alloy. Although SAC
and SABC have higher melting temperatures and require more energy per unit mass of solder,
the higher density of SnPb requires more energy per unit of volume, causing the use/application
stage energy impacts on a functional unit basis to be slightly greater for SnPb than for the higher
melting temperature alloys. Total energy consumption for SnPb, however, is less than that of
SAC and SABC because SnPb upstream processes are less energy-intensive. SnPb upstream
processes (e.g., ME&P) consume 867 MJ/functional unit compared to 2,610 MJ/functional unit
for SAC, 1,250 MJ/functional unit for BSA, and 2,120 MJ/functional unit for SABC. Solder
manufacturing and EOL processes combined consume less than two percent of the life-cycle
energy of any of the solders.
3-27
-------
Energy Use Impacts by Process Group (Paste Solder)
Table 3-17 lists the energy use impacts of each of the processes in the life-cycle of a
solder. Energy impacts in the use/application stage are due entirely to electricity consumed in
the solder reflow process. Upstream energy impacts, on the other hand, arise from the energy
consumed in the extraction and processing of the various metals present in the alloys. Of note is
that energy impacts from silver processing approach impacts from tin processing in solders that
contain both metals, even though the silver content of the alloys is much less than the tin content.
For example, SAC is 95.5 percent Sn and only 3.9 percent Ag, yet its impacts from silver
production are nearly as great as those from tin production. This illustrates the relatively high
energy intensity of silver extraction and processing compared to the other solder metals.
Table 3-17. Energy use impacts by life-cycle stage and process group (paste solder)
Life-Cycle Stage
Process group
SnPb
Score* %
SAC
Score* %
BSA
Score* %
SABC
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Bi production
Total
5.84E+01 0.467
8.09E+02 6.47
N/A N/A
N/A N/A
N/A N/A
8.67E+02 6.94
1.18E+03 8.73
N/A N/A
1.42E+03 10.5
1.94E+00 0.0143
N/A N/A
2.61E+03 19.3
6.06E+02 6.21
N/A N/A
4.25E+02 4.36
N/A N/A
2.21E+02 2.26
1.25E+03 12.8
1.19E+03 9.11
N/A N/A
9.17E+02 7.00
1.62E+00 0.0124
3.34E+00 0.0255
2.12E+03 16.2
MANUFACTURING
Solder
manufacturing
Post-industrial
recycling
Total
9.52E+01 0.762
1.18E+02 0.942
2.13E+02 1.70
1.14E+02 0.840
1.15E+02 0.851
2.29E+02 1.69
8.11E+01 0.832
4.82E+01 0.494
1.29E+02 1.33
1.14E+02 0.873
1.15E+02 0.878
2.29E+02 1.75
USE/APPLICATION
Reflow
application
Total
1.14E+04 91.2
1.14E+04 91.2%
1.07E+04 78.9
1.07E+04 78.9
8.37E+03 85.8
8.37E+03 85.8
1.07E+04 82.0
1.07E+04 82.0
END-OF-LIFE
Landfill
Incineration
Demanufacturing
Cu smelting
Unregulated
Total
GRAND
TOTAL
1.67E+00 0.0134
-4.10E-01 -0.0033
1.12E+00 0.0090
1.44E+01 0.116
O.OOE+00 0.00
1.68E-K)! 0.135
1.25E+04 100
1.45E+00 0.0107
-3.55E-01 -0.0026
9.69E-01 0.0072
1.25E+01 0.0922
O.OOE+00 0.00
1.46E+01 0.107
1.36E+04 100
1.79E+00 0.0183
-4.39E-01 -0.0045
1.14E+00 0.0117
N/A N/A
O.OOE+00 0.00
2.49E+00 0.0255
9.76E+03 100
1.45E+00 0.0111
-3.57E-01 -0.0027
9.73E-01 0.0074
1.25E+01 0.0957
O.OOE+00 0.00
1.46E+01 0.112
1.31E+04 100
*The impact scores are in units of megajoules/1,000 cc of solder applied to a printed wiring board.
N/A=not applicable
3-28
-------
Manufacturing energy impacts are small compared to the upstream and use/application
life-cycle stages, and are almost evenly distributed between solder manufacturing and post-
industrial recycling. An exception is BSA, which consumes less energy in post-industrial
processing (recycling) of the secondary alloy than in solder manufacturing. As discussed in
Section 3.2.2.1, less secondary BSA is used in solder manufacturing, and as a result less post-
industrial processing occurs. Therefore, the BSA solder manufacturing process is a greater
contributor to the BSA manufacturing stage score than is post-industrial recycling. The
difference is ostensibly made up by the increase in primary production of the metals in BSA
(e.g., upstream impacts). SAC and SABC also have less secondary metals production than SnPb,
but they consume nearly as much energy in post-industrial recycling as SnPb due to their higher
melting temperatures.
EOL processes contribute less than 0.2 percent of life-cycle energy impacts for any of the
solders, with the majority of SnPb, SAC, and SABC impacts at EOL coming from smelting
processes to recover copper and other valuable metals from waste electronics. As noted
previously, a copper smelter process is not included in the BSA inventory due to its high bismuth
content, which is unacceptable to copper smelters. Negative energy impacts from incineration
are due to an energy credit for incineration with energy recovery. No energy impacts are shown
for unregulated disposal, as the inventory for this process did not include any resource inputs.
Some energy is consumed, however, when waste PWBs are heated to recover solder and
components. The amount of energy and associated resources consumed in this process are not
known, but they are expected to be small compared to other processes.
Top Contributors to Energy Use Impacts (Paste Solder)
Table 3-18 presents the specific materials or flows contributing greater than or equal to 1
percent of the total energy impact category indicators by solder. As expected from the results
presented above, the fuels used to produce electricity in the use/application stage are the top
contributors to overall energy impacts, with hard coal being the single greatest contributor for all
of the solders. Per the GaBi inventory employed in this study for electricity generation, coal is
the primary fuel used in the U.S. electric grid, accounting for 52 percent of electricity generation
(PE & IKP, 2000). Uranium used to generate nuclear power in the use/application stage is the
next largest contributor for all solders, again because uranium is the next largest fuel in the U.S.
electric grid (23 percent of the U.S. power grid is from nuclear fuel). In addition to fuels used to
generate electricity in the use/application stage, other major contributors to energy impacts
include fuels used in tin and silver extraction and processing. The extraction and processing
inventories are from secondary data sources that do not distinguish whether these fuels are used
to produce electricity consumed during extraction and processing or used directly in these
processes.
3-29
-------
Table 3-18. Top contributors to energy use impacts (paste solder)
Solder
SnPb
SAC
BSA
SABC
Life-Cycle Stage
Use/application
Use/application
Use/application
Use/application
Use/application
Upstream
Upstream
Upstream
Use/application
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
Upstream
Upstream
Upstream
Upstream
Upstream
Upstream
Use/application
Upstream
Use/application
Use/application
Use/application
Use/application
Use/application
Upstream
Upstream
Upstream
Upstream
Use/application
Upstream
Use/application
Use/application
Use/application
Use/application
Use/application
Upstream
Upstream
Upstream
Upstream
Upstream
Upstream
Use/application
Upstream
Process
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Tin production
Tin production
Tin production
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Silver production
Electricity generation
Electricity generation
Silver production
Tin production
Tin production
Tin production
Silver production
Silver production
Electricity generation
Tin production
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Tin production
Tin production
Tin production
Silver production
Electricity generation
Silver production
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Tin production
Tin production
Tin production
Silver production
Silver production
Silver production
Electricity generation
Tin production
Flow
Hard coal (resource)
Uranium (resource)
Natural gas (resource)
Crude oil (resource)
Lignite (resource)
Hard coal (resource)
Natural gas (resource)
Crude oil (resource)
Primary energy from hydro power
Hard coal (resource)
Uranium (resource)
Natural gas (resource)
Hard coal (resource)
Crude oil (resource)
Lignite (resource)
Uranium (resource)
Hard coal (resource)
Natural gas (resource)
Crude oil (resource)
Crude oil (resource)
Primary energy from hydro power
Primary energy from hydro power
Uranium (resource)
Hard coal (resource)
Uranium (resource)
Natural gas (resource)
Crude oil (resource)
Lignite (resource)
Hard coal (resource)
Natural gas (resource)
Crude oil (resource)
Hard coal (resource)
Primary energy from hydro power
Uranium (resource)
Hard coal (resource)
Uranium (resource)
Natural gas (resource)
Crude oil (resource)
Lignite (resource)
Hard coal (resource)
Natural gas (resource)
Crude oil (resource)
Hard coal (resource)
Uranium (resource)
Crude oil (resource)
Primary energy from hydro power
Uranium (resource)
% Contribution
46.8
23.6
11.9
4.14
3.29
1.95
1.91
1.87
1.50
40.5
20.4
10.3
3.80
3.58
2.85
2.63
2.63
2.58
2.53
2.13
1.47
1.29
1.00
44.0
22.2
11.2
3.90
3.10
1.87
1.84
1.80
1.58
1.41
1.09
42.0
21.2
10.7
3.72
2.96
2.75
2.69
2.64
2.53
1.75
1.42
1.34
1.04
5-30
-------
3.2.2.3 Bar solder results
Total Energy Use Impacts by Life-Cycle Stage (Bar Solder)
Table 3-19 presents the bar solder results for energy use impacts by life-cycle stage,
based on the impact assessment methodology presented above. Figure 3-7 presents the results
in a stacked bar chart. General energy consumption is used as an indicator of potential
environmental impacts from the entire energy generation cycle.
Table 3-19. Energy use impacts by life-cycle stage (bar solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
8.38E+02 28.8
2.48E+02 8.52
1.80E+03 62.0
1.87E+01 0.644
2.91E+03 100
SAC
Score* %
3.78E+03 65.5
1.47E+02 2.55
1.82E+03 31.6
1.64E+01 0.284
5.77E+03 100
SnCu
Score* %
1.29E+03 37.7
2.86E+02 8.39
1.82E+03 53.4
1.63E+01 0.476
3.41E+03 100
*The impact scores are in units of megajoules/1,000 cc of solder applied to a printed wiring board.
7 nnn
6 000
5 000
4-1
E
= 4 000
ra ^>"""
c
o
o •a nnn
€
S 2 000
1 000
o
I — I
~
SnPb
SAC
SnCu
D End-of-life
n Use/application
• Manufacturing
D Upstream
Figure 3-7. Bar Solder Total Life-Cycle Impacts: Energy Use
SAC solder paste has the greatest impact category indicator for energy use at 5,770 MJ
per functional unit, followed by SnCu at 3,410 MJ, and SnPb at 2,910 MJ. The ME&P
(upstream) life-cycle stage drives the SAC energy results (contributing 66 percent) and causes it
to dominate over the other two alloys. The use/application stage energy is the top contributor to
SnPb and SnCu energy impacts and the second greatest contributor to SAC energy impacts.
SAC and SnCu wave application energy are equal to one another and SnPb application energy is
slightly less. The lower wave application energy for SnPb is due to its lower melting
3-31
-------
temperature; however, it is only slightly lower due to SnPb's higher density than SAC and SnCu.
Solder manufacturing consumes between 3 and 9 percent of the life-cycle energy; and EOL
processes consume less than 1 percent of the life-cycle energy of any of the solders.
Energy Use Impacts by Process Group (Bar Solder)
Table 3-20 lists the energy use impacts of each of the process groups in the life-cycle of a
solder. Upstream energy impacts arise from the energy consumed in the extraction and
processing of the various metals present in the alloys. Energy impacts from tin and silver
processing are the largest upstream contributing processes. For SAC, energy impacts from silver
processing are greater than impacts from tin processing, even though the silver content of the
alloys is much less than that of the tin. That is, SAC is 95.5 percent tin and only 3.9 percent
silver, yet its impacts from silver production are greater than those from tin production. This
illustrates the relatively high energy intensity of silver extraction and processing compared to the
other solder metals. Energy impacts in the use/application stage are due entirely to electricity
consumed in the wave solder process.
Table 3-20. Energy use impacts by life-cycle stage and process group (bar solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
SnCu
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Total
7.86E+02 27.0
5.21E+01 1.79
N/A N/A
N/A N/A
8.38E+02 28.8
1.66E+03 28.8
N/A N/A
2.11E+03 36.7
3.23E+00 0.0560
3.78E+03 65.5
1.28E+03 37.6
N/A N/A
N/A N/A
3.17E+00 N/A
1.29E+03 37.7
MANUFACTURING
Solder manufacturing
Post-industrial recycling
Total
4.94E+01 1.70
1.98E+02 6.82
2.48E+02 8.52
7.74E+01 1.34
6.98E+01 1.21
1.47E+02 2.55
9.60E+01 2.81
1.90E+02 5.58
2.86E+02 8.39
USE/APPLICATION
Solder application
Total
1.80E+03 62.0
1.80E+03 62.0
1.82E+03 31.6
1.82E+03 31.6
1.82E+03 53.4
1.82E+03 53.4
END-OF-LIFE
Landfill
Incineration
Demanufacturing
Cu smelting
Unregulated
Total
GRAND TOTAL
1.86E+00 0.0640
-4.32E-01 -0.0148
1.24E+00 0.0428
1.60E+01 0.552
O.OOE+00 0.0000
1.87E+01 0.644
2.91E+03 100
1.63E+00 0.0282
-3.78E-01 -0.0066
1.09E+00 0.0189
1.40E+01 0.243
O.OOE+00 0.0000
1.64E+01 0.284
5.77E+03 100
1.62E+00 0.0473
-3.75E-01 -0.0110
1.08E+00 0.0317
1.39E+01 0.408
O.OOE+00 0.0000
1.63E+01 0.476
3.41E+03 100
*The impact scores are in units of megajoules/1,000 cc of solder applied to a printed wiring board.
N/A=not applicable
3-32
-------
Manufacturing energy impacts are relatively small compared to the upstream and
use/application life-cycle stages. Of the two process groups in the manufacturing stage, post-
industrial recycling impacts are greater than the solder manufacturing process group for SnPb
and SnCu. The SnPb post-industrial recycling process group contribution is four times (400
percent) greater than the SnPb solder manufacturing group; and the SnCu post-industrial
recycling process group contribution is 25 percent greater than the SnCu solder manufacturing
process group. For SAC, the post-industrial recycling process group contributes approximately
11 percent less than that from solder manufacturing. The reason SnPb and SnCu have greater
post-industrial impacts than solder manufacturing (as compared to SAC) is because SnPb and
SnCu are assumed to have greater recycled content (coming from post-industrial recycling). The
recycled content for individual solders is based on averages taken from primary data collected
from solder manufacturers. SnPb has the greatest recycled content percent of all three alloys,
which explains the larger difference between PI recycling and solder manufacturing for SnPb
compared to the other alloys. In the cases where there is less secondary (recycled) metal, and
thus more primary (virgin) metal, there is more primary production of the metals, which
translates into impacts in the upstream life-cycle stage.
EOL processes contribute less than 0.6 percent of life-cycle energy impacts for any of the
solders, with the majority of SnPb, SAC, and SABC impacts at EOL coming from smelting
processes to recover copper and other valuable metals from waste electronics. Negative energy
impacts from incineration are due to an energy credit for incineration with energy recovery. No
energy impacts are shown for unregulated disposal, as the inventory for this process did not
include any resource inputs. Some energy is consumed, however, when waste PWBs are heated
to recover solder and components. The amount of energy and associated resources consumed in
this process are not quantitatively known, but they are expected to be small compared to other
processes.
Top Contributors to Energy Use Impacts (Bar Solder)
Table 3-21 presents the specific materials or flows contributing greater than or equal to 1
percent of the total energy impact category indicators by bar solder. As expected from the
results presented above, the fuels used to produce electricity in the use/application stage are the
top contributors to overall energy impacts, with hard coal being the single greatest contributor
for all of the solders. As described under the paste solder results, per the GaBi inventory
employed in this study for electricity generation, coal is the primary fuel used in the U.S. electric
grid. In addition to fuels used to generate electricity in the use/application stage, other major
contributors to energy impacts include fuels used in silver and tin extraction and processing. The
extraction and processing inventories are from secondary data sources that do not distinguish
whether these fuels are used to produce electricity consumed during extraction and processing or
used directly in these processes.
3-33
-------
Table 3-21. Top contributors to energy use impacts (bar solder)
Solder
SnPb
SAC
SnCu
Life-Cycle Stage
Use/application
Use/application
Upstream
Use/application
Upstream
Upstream
Upstream
Use/application
Manufacturing
Use/application
Manufacturing
Manufacturing
Use/application
Use/application
Upstream
Upstream
Upstream
Upstream
Upstream
Use/application
Upstream
Upstream
Use/application
Upstream
Use/application
Upstream
Use/application
Use/application
Use/application
Upstream
Upstream
Upstream
Use/application
Upstream
Use/application
Use/application
Manufacturing
Manufacturing
Process
Electricity generation
Electricity generation
Tin production
Electricity generation
Tin production
Tin production
Tin production
Electricity generation
Electricity generation for
post-industrial recycling
Electricity generation
Heavy fuel oil (#6)
production for post-
industrial recycling
Electricity generation for
post-industrial recycling
Electricity generation
Electricity generation
Silver production
Silver production
Tin production
Tin production
Tin production
Electricity generation
Silver production
Silver production
Electricity generation
Tin production
Electricity generation
Silver production
Electricity generation
Electricity generation
Electricity generation
Tin production
Tin production
Tin production
Electricity generation
Tin production
Electricity generation
Electricity generation
Heavy fuel oil (#6)
production for post-
industrial recycling
Natural gas production for
solder manufacturing
Flow
Hard coal (resource)
Uranium (resource)
Hard coal (resource)
Natural gas (resource)
Natural gas (resource)
Crude oil (resource)
Uranium (resource)
Crude oil (resource)
Hard coal (resource)
Lignite (resource)
Crude oil (resource)
Uranium (resource)
Primary energy from hydro power
Hard coal (resource)
Hard coal (resource)
Uranium (resource)
Hard coal (resource)
Natural gas (resource)
Crude oil (resource)
Uranium (resource)
Crude oil (resource)
Primary energy from hydro power
Natural gas (resource)
Uranium (resource)
Crude oil (resource)
Natural gas (resource)
Lignite (resource)
Hard coal (resource)
Uranium (resource)
Hard coal (resource)
Natural gas (resource)
Crude oil (resource)
Natural gas (resource)
Uranium (resource)
Crude oil (resource)
Lignite (resource)
Crude oil (resource)
Natural gas (resource)
% Contribution
31.8
16.0
8.14
8.12
7.98
7.82
3.08
2.82
2.41
2.24
1.70
1.21
1.02
16.2
13.3
9.19
8.69
8.52
8.35
8.19
7.43
5.11
4.14
3.29
1.44
1.16
1.14
28.0
14.1
11.6
11.4
11.1
7.15
4.39
2.48
1.97
1.40
1.37
3-34
-------
3.2.2.4 Limitations and uncertainties
The major contributors to energy impacts are from electricity generation used during the
use/application stage (particularly for paste solders) and from upstream materials extraction
processes (particularly for SAC bar solder). Similar to the discussion in Section 3.2.1, where
electricity generation for reflow application is concerned, the same uncertainties apply: (1) the
number of data points used to estimate reflow electricity consumption are limited and cover a
large range, and (2) electricity production data are from a secondary source. With regard to the
first source of uncertainty, the amount of electricity consumed during reflow was measured
during reflow testing conducted by the LFSP. These are primary data collected under controlled
conditions to meet the goals and objectives of this study and represent good high and low
estimates of wave electricity consumption; however, because the value used in this baseline
analysis is averaged from a limited amount of data (two data points for each solder), a sensitivity
analysis was performed using the high and low values (see Section 3.3). On the other hand,
uncertainties from the use of secondary data for electricity generation are not considered large
enough to warrant a separate sensitivity analysis.
For wave application results, primary data were also collected for the solder application
process through a controlled testing protocol. Although data from only one test run were used,
these data were compared to other known testing data and are expected to be representative of
typical wave operations, thus introducing little uncertainty. The use of the secondary data for the
electricity generation data was discussed above in the preceding paragraph.
Uncertainties related to the use of upstream data were discussed in Section 3.2.1 and also
apply here, particularly to the silver production data for the SAC bar solder results. GaBi gives
the silver production data "good" quality rating; however, due to its large impact on the life-
cycle of the bar solder results, sensitivity analyses using an alternative data set were conducted
(see Section 3.3).
3-35
-------
3.2.3 Landfill Space Use Impacts
3.2.3.1 Characterization
Landfill impacts are calculated using solid and hazardous waste flows to land as the
volume of landfill space is consumed. This category includes both solid waste and hazardous
waste landfill use. For solid waste landfill use, this category pertains to the use of suitable and
designated landfill space as a natural resource where municipal waste or construction debris is
accepted. For hazardous waste landfill use, this category pertains to the use of suitable and
designated landfill space as a natural resource where hazardous waste, as designated and
regulated under the Resource Conservation and Recovery Act (RCRA), is accepted. For non-
11 S. activities, equivalent hazardous or special waste landfills are considered for this impact
category. Impact scores are characterized from solid and hazardous waste outputs with a
disposition of landfill. Impact characterization is based on the volume of waste, determined
from the inventory mass amount of waste and material density of each specific hazardous waste
type:
where:
IS,
Amtw
D
(IS,), = (Amtw/D),
equals the impact score for landfill (L) use for waste / cubic meters (m3) per
functional unit;
equals the inventory output amount of solid waste /' (kg) per functional unit; and
equals density of waste /' (kg/m3).
3.2.3.2 Paste solder results
Total Landfill Space Use Impacts by Life-Cycle Stage (Paste Solder)
Table 3-22 presents the solder paste results for landfill space use impacts by life-cycle
stage, based on the impact assessment methodology presented above. This impact category
includes both hazardous and non-hazardous waste landfills. The table lists the impact scores per
functional unit, as well as the percent contribution of each life-cycle stage to the total impacts for
each alloy. Figure 3-8 presents the results in a stacked bar chart.
Table 3-22. Landfill space use impacts by life-cycle stage (paste solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
4.20E-05 1.53%
7.68E-05 2.79%
1.81E-03 65.8%
8.23E-04 29.9%
2.75E-03 100%
SAC
Score* %
1.36E-02 83.9%
9.02E-05 0.558%
1.70E-03 10.5%
8.13E-04 5.03%
1.62E-02 100%
BSA
Score* %
4.37E-03 66.6%
4.21E-05 0.642%
1.33E-03 20.3%
8.24E-04 12.5%
6.57E-03 100%
SABC
Score* %
8.73E-03 77.0%
9.01E-05 0.795%
1.71E-03 15.1%
8.12E-04 7.16%
1.13E-02 100%
*The impact scores are in units of cubic meters of landfill space/1,000 cc of solder applied to a printed wiring
board.
3-36
-------
0 018
0 016
— 0 014
3
"ro 0 012
~ 0 01
O
an nns
In 0 006
n 0 004
E
0 00?
n
••
^^™
SnPb
™
SAC
^B
-
BSA
PB
SABC
• End nf life
Q Use/application
I Manufacturing
• Upstream
Figure 3-8. Paste Solder Total Life-Cycle Impacts: Landfill Space Use
SAC solder paste has the greatest impact category indicator for landfill space use at
0.0162 m3 per functional unit, followed by SABC at 0.0113 m3, BSA at
0.00657 m3, and SnPb at 0.00275 m3 per functional unit. The upstream life-cycle stage
dominates the total landfill space scores of the lead-free alloys, accounting for 67 to 84 percent
of the totals. SnPb landfill space impacts, on the other hand, are dominated by the
use/application stage at 66 percent of its total score, followed by the EOL stage at 30 percent.
The use/application stage is the second greatest contributor for the lead-free alloys, followed by
the EOL stage. The solder manufacturing stage contributes less then 3 percent for any of the
solder alloys.
To put these volumes of landfill space into perspective, in 2001, U.S. residents,
businesses, and institutions produced more than 229 million tons of municipal solid waste, which
is approximately 4.4 pounds (2 kg) per person per day (EPA, 2004). Assuming an average bulk
density of 445 kg/m3 (Franklin Associates, 1999), this equates to approximately 0.0045 m3 of
landfill space. This value falls between the life-cycle landfill space impacts per functional unit
for SnPb and BSA.
Landfill Space Use Impacts by Process Group (Paste Solder)
Table 3-23 lists the landfill space use impacts of each of the process groups in the life-
cycle of a solder paste. Landfill space use impacts are driven by the upstream processes for the
lead-free alloys that alone exceed the total impacts from SnPb. The silver production process
contributes between 60 and 83 percent of the total life-cycle landfill space use impacts. This is
of interest as the composition of silver in those alloys is relatively small (between 1 and 3.9
percent), suggesting that the silver production process generates more landfilled waste per unit of
3-37
-------
metal produced than the other metals. For the SnPb alloy, the upstream processes contribute
only about 1.4 percent to the total impacts, while it is the reflow application process group (e.g.,
reflow application and associated electricity generation) that contributes the most to total
impacts.
Table 3-23. Landfill space use impacts by life-cycle stage and process
group (paste solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
BSA
Score* %
SABC
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Bi production
Total
5.16E-06 0.175
3.68E-05 1.25
N/A N/A
N/A N/A
N/A N/A
4.20E-05 1.42
7.55E-06 0.0461
N/A N/A
1.35E-02 82.7
3.54E-06 0.0216
N/A N/A
1.36E-02 82.8
3.87E-06 0.0576
N/A N/A
4.04E-03 60.2
N/A N/A
3.25E-04 4.84
4.37E-03 65.1
7.62E-06 0.0660
N/A N/A
8.72E-03 75.4
2.96E-06 0.0256
4.92E-06 0.0426
8.73E-03 75.6
MANUFACTURING
Solder
manufacturing
Post-industrial
recycling
Total
2.81E-05 0.951
4.87E-05 1.65
7.68E-05 2.60
2.90E-05 0.177
6.12E-05 0.374
9.02E-05 0.551
2.22E-05 0.331
1.99E-05 0.297
4.21E-05 0.627
2.91E-05 0.252
6.10E-05 0.528
9.01E-05 0.780
USE/APPLICATION
Reflow application
Total
2.01E-03 68.1
2.01E-03 68.1
1.91E-03 11.7
1.91E-03 11.7
1.48E-03 22.0
1.48E-03 22.0
1.92E-03 16.6
1.92E-03 16.6
END-OF-LIFE
Landfill
Incineration
Demanufacturing
Cu smelting
Unregulated
Total
GRAND TOTAL
6.49E-04 22.0
1.65E-04 5.60
1.78E-07 0.0060
8.06E-06 0.273
O.OOE+00 0.00
8.23E-04 27.9
2.95E-03 100
6.42E-04 3.92
1.63E-04 1.00
1.54E-07 0.0009
7.15E-06 0.0437
O.OOE+00 0.00
8.13E-04 4.97
1.64E-02 100
6.50E-04 9.67
1.74E-04 2.59
1.82E-07 0.003
N/A N/A
O.OOE+00 0.00
8.24E-04 12.3
6.72E-03 100
6.42E-04 5.56
1.63E-04 1.41
1.55E-07 0.0013
7.17E-06 0.0621
O.OOE+00 0.00
8.12E-04 7.03
1.16E-02 100
*The impact scores are in units of cubic meters (m3) of landfill space/1,000 cc of solder applied to a printed wiring
board.
N/A=not applicable
Of the four solder paste alloys, EOL processes contribute 5 to 28 percent of total landfill
space use impacts, with the majority coming from the landfill process group itself. This process
group contributes from 4 (for SAC) to 22 (for SnPb) percent of the total impacts, depending on
the alloy, but the actual scores from the landfill process group for each alloy are essentially the
same. Incineration, which produces ash that is landfilled, is the next greatest EOL contributor at
1 to 5.6 percent. Copper smelting also yields ash that requires a small amount of landfill space.
The alloys that are sent to copper smelting have a small proportion of their impact scores from
copper smelting, and an even smaller proportion from demanufacturing. Due to its high bismuth
content, the BSA alloy is assumed to bypass the copper smelting process and go directly to
3-38
-------
landfilling and incineration from demanufacturing; therefore, there is no contribution from
copper smelting for BSA, but it has a larger contribution from demanufacturing than the other
alloys.
For the landfill space impact category, there are no negative impacts from incineration as
there are with other impact categories. (Negative impacts arise from an energy credit for natural
gas used in incineration with energy recovery). This is because the incineration process itself
generates more landfilled waste than would be given credit from the natural gas savings from
incineration with energy recovery. No landfill space use impacts are shown for unregulated
disposal, as this process does not include disposal in a regulated landfill.
Landfill space use impacts from manufacturing are small compared to the upstream,
use/application, and EOL life-cycle stages; these impacts are driven by both solder
manufacturing and post-industrial recycling. For SnPb, SAC, and SABC, the post-industrial
recycling impacts are greater than those from solder manufacturing (e.g., SAC post-industrial
recycling is 6.12 x 10"5 m3 per functional unit, while SAC solder manufacturing is 2.90 x 10"5
m3/functional unit). For BSA, on the other hand, post-industrial recycling contributes less. The
distribution of impacts in the manufacturing life-cycle stage is influenced by a combination of
several factors including: landfilled waste generated during the post-industrial recycling process
is greater than the solder manufacturing process, where much of the waste is sent to recycling;
different melting points of the alloys, which affects the amount of energy used to melt the alloys
and, therefore, the amount of waste from energy production; and varied secondary alloy content
among the alloys.
Top Contributors to Landfill Space Use Impacts (Paste Solder)
Table 3-24 presents the specific materials or flows contributing greater than or equal to 1
percent of landfill space use impacts by solder paste. Slag from silver production is the top
contributor for the three lead-free alloys that all contain silver in varying amounts. Landfilled
slag from silver production contributes from 57 to 78 percent of the total landfill impact scores
depending on the alloy. Sludge from silver production also contributes 4 to 6 percent to total
impacts depending on the alloy. For the SnPb alloy, which does not contain silver in its
composition, the top contributor at 65 percent is sludge from the U.S. electric grid which
supplies electricity to the reflow application process in the use/application life-cycle stage. For
the silver-containing alloys (e.g., the three lead-free alternatives), sludge from electricity
supplied to the use/application stage is the second greatest contributor (10 to 20 percent of total
impacts).
Landfilling of the alloy on a PWB at EOL is the next greatest contributor for each alloy,
contributing from 4 to 24 percent of total impacts. As noted in the process group discussion
above, the actual impact scores from this flow are essentially the same for each alloy. Smaller
contributors include metals in ash from incineration sent to landfills (contributing 1 to 4 percent),
and in the case of BSA, sludge from bismuth production (contributing approximately 4.6 percent
to the BSA landfill impacts).
3-39
-------
Table 3-24. Top contributors to landfill space use impacts (paste solder)
Solder
SnPb
SAC
BSA
SABC
Life-Cycle Stage
Use/application
End-of-life
End-of-life
End-of-life
Upstream
Upstream
Use/application
Upstream
End-of-life
Upstream
Use/application
End-of-life
Upstream
Upstream
End-of-life
End-of-life
Upstream
Use/application
End-of-life
Upstream
End-of-life
Process
Electricity generation
Landfilling (SnPb)
Solder incineration (SnPb)
Solder incineration (SnPb)
Lead production
Silver production
Electricity generation
Silver production
Landfilling (SAC)
Silver production
Electricity generation
Landfilling (BSA)
Bismuth production
Silver production
Solder incineration (BSA)
Solder incineration (BSA)
Silver production
Electricity generation
Landfillling (SABC)
Silver production
Solder incineration (SABC)
Flow
Sludge (hazardous waste)
Sn-Pb solder to landfill
Tin in ash to landfill
Lead in ash to landfill
Sludge (hazardous waste)
Slag (hazardous waste)
Sludge (hazardous waste)
Sludge (hazardous waste)
SAC solder to landfill
Slag (hazardous waste)
Sludge (hazardous waste)
BSA solder to landfill
Sludge (hazardous waste)
Sludge (hazardous waste)
Bismuth in ash to landfill
Tin in ash to landfill
Slag (hazardous waste)
Sludge (hazardous waste)
SABC solder to landfill
Sludge (hazardous waste)
Tin in ash to landfill
% Contribution
64.8
23.5
4.45
1.67
1.16
77.8
10.4
5.72
3.97
57.1
20.0
9.86
4.55
4.20
1.29
1.27
71.3
14.9
5.65
5.24
1.38
3.2.3.3 Bar solder results
Total Landfill Space Use Impacts by Life-Cycle Stage (Bar Solder)
Table 3-25 presents the solder paste results for landfill space use impacts by life-cycle
stage, based on the impact assessment methodology presented above. This impact category
includes both hazardous and non-hazardous waste landfills. The table lists the impact scores per
functional unit, as well as the percent contribution of each life-cycle stage to the total impacts for
each alloy. Figure 3-9 presents the results in a stacked bar chart.
Table 3-25. Landfill space use impacts by life-cycle stage (bar solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
3.79E-05 2.83
1.07E-04 8.02
2.87E-04 21.5
9.05E-04 67.7
1.34E-03 100
SAC
Score* %
2.01E-02 94.0
9.34E-05 0.436
2.90E-04 1.36
9.03E-04 4.22
2.14E-02 100
SnCu
Score* %
1.40E-05 1.05
1.26E-04 9.45
2.90E-04 21.7
9.04E-04 67.8
1.33E-03 100
*The impact scores are in units of m3/l,000 cc of solder applied to a printed wiring board.
SAC solder paste has the greatest impact category indicator for landfill space use at
0.0214 m3 per functional unit, followed by SnPb at 0.00134 m3, and SnCu at
3-40
-------
0.00133 m3 per functional unit. The upstream life-cycle stage dominates the total landfill space
score for SAC, accounting for 94 percent of the totals. On the other hand, SnPb and SnCu
landfill space impacts are dominated by the EOL stage, each at approximately 68 percent of their
total scores. The use/application stage is the second greatest contributor for SnPb and SnCu,
followed by the manufacturing stage. The upstream stage contributes less then 3 percent for
SnPb and SnCu. The EOL stage is the second greatest life-cycle stage for SAC (4 percent),
followed by the use/application and manufacturing stages.
0 095
4-1
'c
D o ois
"(5
o
7=
O
£ o 01
,= u-ul
to
0 005
n
<
n
SnPb SAC SnCu
n End-of-life
n Use/application
• Manufacturing
D Upstream
Figure 3-9. Bar Solder Total Life-Cycle Impacts: Landfill Space Use
Landfill Space Use Impacts by Process Group (Bar Solder)
Table 3-26 lists the landfill space use impacts of each of the process groups in the life-
cycle of a solder paste. Landfill space use impacts are driven by the upstream processes for SAC
that alone exceeds the total impacts from SnPb and SnCu. The silver production process
contributes 94 percent of the total life-cycle landfill space use impacts. As stated under the paste
solder results, this is of interest because the percent composition of silver is relatively small (3.9
percent), suggesting that the silver production process generates more landfilled waste per unit of
metal produced than the other metals. For the SnPb and SnCu alloys, the upstream processes
contribute only about 1 and 3 percent, respectively, to the total impacts, while it is the landfilling
of process group (e.g., landfilling and associated diesel fuel production) that contributes the most
to total impacts.
3-41
-------
Table 3-26. Landfill space use impacts by life-cycle stage and process group (bar solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
SnCu
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Total
5.01E-06 0.375
3.29E-05 2.46
N/A N/A
N/A N/A
3.79E-05 2.83
1.06E-05 0.0496
N/A N/A
2.01E-02 93.9
5.91E-06 0.0276
2.01E-02 94.0
8.19E-06 0.614
N/A N/A
N/A N/A
5.80E-06 0.434
1.40E-05 1.05
MANUFACTURING
Solder manufacturing
Post-industrial recycling
Total
2.52E-05 1.89
8.20E-05 6.13
1.07E-04 8.02
5.63E-05 0.263
3.70E-05 0.173
9.34E-05 0.436
6.26E-05 4.69
6.35E-05 4.76
1.26E-04 9.45
USE/APPLICATION
Solder application
Total
2.87E-04 21.5
2.87E-04 21.5
2.90E-04 1.36
2.90E-04 1.36
2.90E-04 21.7
2.90E-04 21.7
END-OF-LIFE
Landfill
Incineration
Demanufacturing
Cu smelting
Unregulated
Total
GRAND TOTAL
7.22E-04 54.0
1.74E-04 13.0
1.98E-07 0.0148
8.95E-06 0.670
O.OOE+00 0.0000
9.05E-04 67.7
1.34E-03 100
7.21E-04 3.37
1.74E-04 0.812
1.73E-07 0.0008
8.03E-06 0.0375
O.OOE+00 0.0000
9.03E-04 4.22
2.14E-02 100
7.21E-04 54.1
1.75E-04 13.1
1.72E-07 0.0129
7.97E-06 0.597
O.OOE+00 0.0000
9.04E-04 67.8
1.33E-03 100
*The impact scores are in units of m 71,000 cc of solder applied to a printed wiring board.
N/A=not applicable
Of the three bar solder alloys, EOL processes contribute 4 to 68 percent of total landfill
space use impacts, with the majority coming from the landfill process group itself. This process
group contributes from 3 (for SAC) to 54 (for SnPb) percent of the total impacts, depending on
the alloy, but the actual scores from the landfill process group for each alloy are essentially the
same. As with the paste results, incineration, which produces ash that is landfilled, is the next
greatest EOL contributor (1 to 13 percent of total impacts). Copper smelting also yields ash that
requires a small amount of landfill space, thus, the alloys that are sent to copper smelting have a
small proportion of their impact scores from copper smelting, and an even smaller proportion
from demanufacturing.
For the landfill space impact category, there are no negative impacts from incineration as
there are with other impact categories. (Negative impacts arise from an energy credit for natural
gas used in incineration with energy recovery). This is because the incineration process itself
generates more landfilled waste than would be given credit from the natural gas savings from
incineration with energy recovery. No landfill space use impacts are shown for unregulated
disposal as this process does not include disposal in a regulated landfill.
Landfill space use impacts from manufacturing are small compared to the upstream,
use/application, and EOL life-cycle stages; these impacts are driven more or less by either solder
3-42
-------
manufacturing and post-industrial recycling, depending on the alloy and, particularly the amount
of recycled versus virgin material used in manufacturing (discussed in earlier sections).
Top Contributors to Landfill Space Use Impacts (Bar Solder)
Table 3-27 presents the specific materials or flows contributing greater than or equal to 1
percent of landfill space use impacts by solder paste. For SnPb and SnCu, the solder on the
PWB going to landfill is the top contributor to landfill space use (each is 54 percent of total
impacts). For SAC, slag from silver production is the top contributor (87 percent of the total
landfill impact score). Sludge from silver production also contributes 6 percent to total impacts
depending on the alloy. For SnPb and SnCu, which do not contain silver, the second top
contributor (at 21 percent) is sludge from the U.S. electric grid that supplies electricity to the
wave application process in the use/application life-cycle stage. For SAC, sludge from
electricity supplied to the use/application stage contributes only 1 percent of total impacts since
slag and sludge from silver production dominate SAC's impacts.
Table 3-27. Top contributors to landfill space use impacts (bar solder)
Solder
SnPb
SAC
SnCu
Life-Cycle
Stage
End-of-life
Use/application
End-of-life
End-of-life
Manufacturing
Upstream
Manufacturing
Upstream
Upstream
End-of-life
Use/application
End-of-life
Use/application
End-of-life
Manufacturing
Manufacturing
Manufacturing
Process
Landfilling (SnPb)
Electricity generation
Solder incineration (SnPb)
Solder incineration (SnPb)
Heavy fuel oil (#6) for post-
industrial recycling
Lead prodution
Electricity generation for post-
industrial recycling
Silver production
Silver production
Landfilling
Electricity generation
Landfilling
Electricity generation
Incineration
Heavy fuel oil (#6) production for
post-industrial recycling
LPG production for solder
manufacturing
Natural gas production for solder
manufacturing
Flow
SnPb solder on PWB to landfill
Sludge (hazardous waste)
Tin in ash to landfill
Lead in ash to landfill
Sludge (hazardous waste)
Sludge (hazardous waste)
Sludge (hazardous waste)
Slag (hazardous waste)
Sludge (hazardous waste)
SAC solder on PWB to landfill
Sludge (hazardous waste)
SnCu solder on PWB to landfill
Sludge (hazardous waste)
Tin in ash to landfill
Sludge (hazardous waste)
Slags and ash (hazardous waste)
Sludge (hazardous waste)
%
Contribution
53.7
21.1
9.63
3.62
3.37
2.12
1.60
87.2
6.41
3.36
1.34
53.8
21.4
13.2
3.19
2.26
1.19
3-43
-------
3.2.3.4 Limitations and uncertainties
Landfill use pertains to the use of suitable and designated landfill space as a natural
resource where the specified type of waste (solid or hazardous) is accepted. Landfill use impacts
are characterized from solid or hazardous waste outputs with a disposition of landfill. Impact
characterization is based on the volume of waste determined from the inventory mass amount of
waste and materials density of each specific waste.
A limitation in the impact characterization method is that it only addresses the volume of
landfill space used and not the type of materials in the landfilled waste. Toxic materials that are
landfilled, and potentially leach from the landfill, are captured in other impact categories (e.g.,
public health and aquatic ecotoxicity impact categories). In addition, this impact category does
not distinguish between hazardous and non-hazardous landfill space, and does not include
radioactive waste landfill space. The radioactive waste landfill space would be directly
proportional to the amount of electricity consumed in the life-cycle across all alloy alternatives
and, as a boundary-setting decision, it was excluded from the scope in the goals and scoping
phase of this LCA.
Limitations and uncertainties in the LCI data for top contributors to landfill space
impacts also contribute to overall LCIA limitations and uncertainties. SnPb paste and bar
impacts, as well as SnCu bar impacts, are driven by the use/application and EOL life-cycle
stages, while the silver-bearing alloys (both paste and bar) are driven by silver production in the
upstream life-cycle stage, and to a lesser degree, use/application and EOL. The major source of
uncertainty in silver-bearing alternative alloys is the secondary data set used for silver
production. As discussed in Section 3.2.1.4, although this process is considered of "good"
quality per GaBi, an alternate analysis using another silver data set was conducted because life-
cycle impacts in this and several other impact categories were largely being driven by the
inventory for silver production (see Section 3.3).
The second greatest contributor to lead-free paste impact scores, and the greatest
contributor to SnPb paste, is electricity generation from the reflow application of solder.
Uncertainties in these data arise from the fact that (1) an average value from limited data
representing high and low electricity consumption values was used for reflow electricity
consumption, and (2) electricity production data are from a secondary source. A sensitivity
analysis addressing the former source of uncertainty is presented in Section 3.3, but the latter is
not considered large enough to warrant any further analysis.
Primary uncertainty in the EOL scores is related to the assumptions about the disposition
of waste electronics. For example, we assumed that 72 percent of waste electronics is landfilled,
based on the percent of waste electronics destined for recycling and the distribution of U.S.
municipal solid waste between landfilling and incineration (EPA, 2002). The assumption about
the percent of electronic waste currently being recycled is the best available information from
EPA (described in Chapter 2); however, determining the fraction of that waste being diverted to
unregulated recycling or the actual amount of electronics that are destined for landfills or other
dispositions remains difficult.
Another source of uncertainty in EOL impacts is due to the fact that the volume of solder
metals in incinerator ash was estimated based on the scientific literature for metals partitioning
from incineration processes (see Chapter 2). These estimates were done specifically for this
3-44
-------
analysis and are not expected to be a large source of uncertainty. Uncertainty remains, however,
because the data were for incineration of municipal waste, only a portion of which contained
waste electronics. These data were compared against data measured from the incineration of
selected computer equipment and were found to be comparable.
Finally, another limitation as it pertains to the disposal of waste electronics themselves
(and not the disposal of waste from the extraction of fuels used to process waste electronics, for
example) is that the EOL analysis only evaluates metal outputs from PWBs and waste
electronics. This allows the analysis to focus on the metal alloys themselves, but does not
include by-product outputs that might occur during EOL processes (e.g., volume of waste PWBs
that are landfilled). If a separate analysis of EOL were done, and the actual outputs from the
entire process of disposing or recycling waste electronics were considered, the results might be
different.
3-45
-------
3.2.4 Global Warming Impacts
3.2.4.1 Characterization
The build up of carbon dioxide (CO2), and other greenhouse gases, in the atmosphere
may generate a "greenhouse effect" of rising temperature and climate change. GWP refers to the
warming, relative to CO2, that chemicals contribute to this effect by trapping the Earth's heat.
The impact scores for the effects of global warming and climate change are calculated using the
mass of a global warming gas released to air, modified by a GWP equivalency factor. The GWP
equivalency factor is an estimate of a chemical's atmospheric lifetime and radiative forcing that
may contribute to global climate change compared to the reference chemical CO2; therefore,
GWPs are in units of CO2 equivalents. GWPs have been published for known global warming
chemicals within differing time horizons. The LCIA methodology employed in the LFSP uses
GWPs having effects in the 100-year time horizon. Although LCA does not necessarily include
a temporal component of the inventory, impacts from releases during the life-cycle of solder are
expected to be within the 100-year time frame. Appendix D presents a current list of GWPs as
identified by the Intergovernmental Panel on Climate Change (IPCC, 2001). Global warming
impact scores are calculated for any chemicals in the LFSP LCI that are found on the list. The
equation to calculate the impact score for an individual chemical is as follows:
(ISGW\ = (EFGWP xAmtaa),
where:
ISGW equals the global warming impact score for greenhouse gas chemical / (kg CO2
equivalents) per functional unit;
EFGWP equals the GWP equivalency factor for greenhouse gas chemical /' (CO2
equivalents, 100-year time horizon) (Appendix D); and
AmtQQ equals the inventory amount of greenhouse gas chemical /' released to air (kg) per
functional unit.
3.2.4.2 Paste solder results
Total Global Warming Impacts by Life-Cycle Stage (Paste Solder)
Table 3-28 presents the solder paste results for global warming impacts by life-cycle
stage, based on the impact assessment methodology. The table lists the global warming impact
scores per functional unit for the life-cycle stages of each solder paste alloy, as well as the
percent contribution of each life-cycle stage to the total impacts. Figure 3-10 presents the results
in a stacked bar chart.
3-46
-------
Table 3-28. Global warming impacts by life-cycle stage (paste solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
5.92E+01 7.24
8.58E+00 1.05
7.49E+02 91.6
6.18E-01 0.0756
8.17E+02 100
SAC
Score* %
1.60E+02 18.4
9.28E+00 1.06
7.03E+02 80.5
5.35E-01 0.0612
8.73E+02 100
BSA
Score* %
7.58E+01 12.0
5.21E+00 0.825
5.50E+02 87.2
4.49E-02 0.0071
6.31E+02 100
SABC
Score* %
1.33E+02 15.7
9.28E+00 1.09
7.06E+02 83.2
5.37E-01 0.0633
8.49E+02 100
*The impact scores are in units of CO2-equivalents/l,000 cubic centimeters of solder applied to a printed wiring
board.
1 nnn -,
Qnn -
4_i
3
"m vnn
0
^ pnn
C
i2 c;nn
>
D A_r\r\
sr
CNI ^nn
Q ODD -
o
_ onn
100 -
n
SnPb
SAC
BSA
SABC
• hna-oT-iiTe
D Use/application
• Manufacturing
H Upstream
Figure 3-10. Solder Paste Total Life-Cycle Impacts: Global Warming
Global warming impacts follow the same pattern as energy use impacts. This is not
unexpected as large amounts of electrical energy are used in the life-cycle of these alloys, and
electricity generation produces considerable amounts of the global warming gas, CO2. SAC
solder paste has the greatest impact category indicator for global warming at 873 kg of CO2-
equivalents per functional unit, closely followed by SABC at 849 kg CO2-equivalents, and SnPb
at 817 kg CO2-equivalents. BSA is the only solder with a substantially lower global warming
impact (631 kg CO2-equivalents per functional unit). This is due mostly to its lower melting
temperature, and accordingly, its reduced energy requirements during reflow application (see
discussion in Section 3.2.2.2). As shown in the table and figure, the use/application stage
dominates global warming impacts for all of the solders, accounting for 81 to 92 percent of
impacts depending on the alloy. Global warming impacts from Sn/Pb upstream processes (e.g.,
materials extraction and processing) are 59.2 kg of CO2-equivalents/l,000 cc of solder compared
3-47
-------
to SAC for 160 kg CO2-equivalents, BSA for 75.8 kg CCyequivalents, and SABC for 133 kg
CO2-equivalents. The upstream life-cycle stages contribute about 7 to 18 percent of the total life-
cycle impacts depending on the alloy. Solder manufacturing and EOL processes combined
contribute less than 1.2 percent of the life-cycle global warming impacts of any of the solders.
Global Warming Impacts by Process Group (Paste Solder)
Table 3-29 lists the global warming impacts of each of the processes in the life-cycle of
solder paste. Global warming impacts in the use/application stage are due entirely to electricity
consumed in the solder reflow process. Conversely, upstream global warming impacts arise
from the emissions associated with the extraction and processing of the various metals present in
the alloys. The magnitude of global warming scores from silver processing approach those from
tin processing in solders that contain both metals, even though the silver content of the alloys is
much less than the tin content. For example, SAC is 95.5 percent tin and only 3.9 percent silver,
yet SAC impacts from silver production (79.2 kg CO2-equivalents) almost equal those from tin
production (80.9 kg CO2-equivalents). This is due to the relatively high energy intensity of
silver extraction and processing compared to the other solder metals.
Table 3-29. Global warming impacts by life-cycle stage and process group (paste solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
BSA
Score* %
SABC
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Bi production
Total
5.53E+01 6.14
3.89E+00 0.432
N/A N/A
N/A N/A
N/A N/A
5.92E-K)! 6.57
8.09E+01 8.43
N/A N/A
7.92E+01 8.25
7.80E-02 0.0081
N/A N/A
1.60E+02 16.7
4.14E+01 5.99
N/A N/A
2.37E+01 3.42
N/A N/A
1.07E+01 1.54
7.58E+01 10.9
8.17E+01 8.73
N/A N/A
5.10E+01 5.45
6.53E-02 0.0070
1.62E-01 0.0173
1.33E+02 14.2
MANUFACTURING
Solder
manufacturing
Post-industrial
recycling
Total
2.92E+00 0.325
5.66E+00 0.629
8.58E+00 0.953
4.70E+00 0.490
4.57E+00 0.477
9.28E+00 0.966
2.89E+00 0.418
2.32E+00 0.334
5.21E+00 0.752
4.72E+00 0.504
4.56E+00 0.487
9.28E+00 0.992
USE/APPLICATION
Reflow
application
Total
8.32E+02 92.4
8.32E+02 92.4
7.90E+02 82.3
7.90E+02 82.3
6.11E+02 88.3
6.11E+02 88.3
7.93E+02 84.8
7.93E+02 84.8
END-OF-LIFE
Landfill
Incineration
Demanufacture
Cu smelting
Unregulated
Total
1.43E-02 0.0016
-4.25E-02 -0.0047
7.36E-02 0.0082
5.72E-01 0.0636
O.OOE+00 0.00
6.18E-01 0.0686
1.24E-02 0.0013
-3.67E-02 -0.0038
6.37E-02 0.0066
4.95E-01 0.0516
O.OOE+00 0.00
5.35E-01 0.0557
1.53E-02 0.0022
-4.54E-02 -0.0066
7.50E-02 0.0108
N/A N/A
O.OOE+00 0.00
4.49E-02 0.0065
1.24E-02 0.0013
-3.69E-02 -0.0039
6.40E-02 0.0068
4.97E-01 0.0531
O.OOE+00 0.00
5.37E-01 0.0574
3-48
-------
Table 3-29. Global warming impacts by life-cycle stage and process group (paste solder)
Life-cycle stage
Process group
GRAND
TOTAL
SnPb
Score* %
9.00E+02 100
SAC
Score* %
9.60E+02 100
BSA
Score* %
6.92E+02 100
SABC
Score* %
9.36E+02 100
*The impact scores are in units of CO2-equivalents/l,000 cubic centimeters of solder applied to a printed wiring
board.
N/A=not applicable
Global warming impacts from the manufacturing life-cycle stage are small compared to
the upstream and use/application life-cycle stages and are nearly evenly distributed between
solder manufacturing and post-industrial recycling, with the exception of BSA. EOL processes
contribute less than 0.07 percent of life-cycle global warming impacts for any of the solders,
with the majority coming from smelting processes that recover copper and other valuable metals
from waste electronics. Negative global warming impacts from incineration are due to the
energy credit for incineration with energy recovery. No global warming impacts are shown for
unregulated disposal as the inventory for this process does not include any global warming gas
emissions or energy sources as inputs. Some energy is consumed, however, when waste PWBs
are heated to recover solder and valuable components. The amount of energy consumed and the
resulting global warming gases emitted in this process are not known, but are expected to be
relatively small.
Top Contributors to Global Warming Impacts (Paste Solder)
Table 3-30 presents the specific materials or flows contributing at least 1 percent of the
global warming impacts by solder. As expected from the results presented above, global
warming gases generated from the production of electricity in the use/application stage are the
top contributors to overall global warming impacts, with CO2 being the single greatest
contributor for all of the solders (ranging from 77 to 88 percent). CO2 is primarily emitted from
coal-fired power generation; coal is the primary fuel used to generate electricity in the U.S.
electric grid. Electricity generated for the use/application stage also emits methane and nitrous
oxide as top contributors to the overall global warming impacts. In addition to emissions from
electricity generation in the use/application stage, other major contributors to global warming
impacts include CO2 from tin, silver, and bismuth production, depending on the alloy. The
extraction and processing inventories are from secondary data sources that do not distinguish
whether global warming gases are emitted from electric power plants producing electricity for
the metals production processes or emitted directly during extraction and processing.
3-49
-------
Table 3-30. Top contributors to global warming impacts (paste solder)
Solder
SnPb
SAC
BSA
SABC
Life-Cycle Stage
Use/application
Upstream
Use/application
Use/application
Use/application
Upstream
Upstream
Use/application
Use/application
Upstream
Upstream
Use/application
Upstream
Use/application
Upstream
Upstream
Use/application
Process
Electricity generation
Tin production
Electricity generation
Electricity generation
Electricity generation
Tin production
Silver production
Electricity generation
Electricity generation
Tin production
Silver production
Electricity generation
Bismuth production
Electricity generation
Tin production
Silver production
Electricity generation
Flow
Carbon dioxide
Carbon dioxide
Methane
Nitrous oxide (laughing gas)
Carbon dioxide
Carbon dioxide
Carbon dioxide
Methane
Carbon dioxide
Carbon dioxide
Carbon dioxide
Methane
Carbon dioxide
Carbon dioxide
Carbon dioxide
Carbon dioxide
Methane
% Contribution
87.7
6.77
2.84
1.00
77.1
9.27
8.59
2.49
83.4
6.57
3.55
2.70
1.61
79.6
9.62
5.69
2.58
3.2.4.3 Bar solder results
Total Global Warming Impacts by Life-Cycle Stage (Bar Solder)
Table 3-31 presents the global warming impacts by life-cycle stage for bar solder based
on the impact assessment methodology. The table lists the global warming impact scores per
functional unit for the life-cycle stages of each bar solder alloy, as well as the percent
contribution of each life-cycle stage to the total impacts. Figure 3-11 presents the results in a
stacked bar chart.
Table 3-31. Global warming impacts by life-cycle stage (bar solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
5.72E+01 30.5
1.11E+01 5.92
1.19E+02 63.2
6.89E-01 0.368
1.87E+02 100
SAC
Score* %
2.31E+02 64.8
5.19E+00 1.45
1.20E+02 33.6
6.03E-01 0.169
3.57E-KJ2 100
SnCu
Score* %
8.79E+01 40.8
7.15E+00 3.32
1.20E+02 55.6
5.99E-01 0.278
2.16E+02 100
*The impact scores are in units of CO2-equivalents/l,000 cubic centimeters of bar solder applied to a printed wiring
board.
Global warming impacts for bar solder, much like solder paste, have a similar distribution
as that for energy use impacts, due to the large amounts of electrical energy used over the life-
cycle of these alloys. As mentioned before, electricity generation produces considerable
amounts of the global warming gas, CO2. SAC bar solder has the greatest impact category
indicator for global warming at 357 kg of CO2-equivalents per functional unit, followed by SnCu
3-50
-------
at 216 kg CO2-equivalents, and SnPb at 187 kg CO2-equivalents. Unlike the paste solders where
the global warming impacts were dominated by the use/application stage, both the upstream and
use/application stages contributed significantly to the global warming impacts for each of the bar
solders. Global warming impacts from upstream processes (e.g., ME&P) for SAC are 231 kg of
CO2-equivalents/l,000 cc of solder compared to 87.9 kg CO2-equivalents for SnCu and 57.2 kg
CO2-equivalents for SnPb. The upstream life-cycle stages contribute from
31 to 65 percent of the overall global warming impacts for any bar solder.
400
T^n
'E
D
75 300 -
o
° 250
9DD
-150
?
'" 100 -
o
o
S ^n
n
SnPb
SAC
SnCu
• Manufacturing
n Upstream
Figure 3-11. Bar Solder Total Life-Cycle Impacts: Global Warming
Though the impacts resulting from upstream processes varied greatly, global warming
impacts resulting from the use/application stage were nearly identical for each of the solders,
ranging from 119 to 120 kg CO2-equivalents (see Chapter 2 for bar solder energy consumption
details). Solder manufacturing and EOL processes combined contribute less than 6.3 percent of
the life-cycle global warming impacts of any of the solders.
Global Warming Impacts by Process Group (Bar Solder)
Table 3-32 lists the global warming impacts resulting from each of the processes in the
life-cycle of bar solder alloys. Upstream global warming impacts arise from the emissions
associated with the extraction and processing of the various metals present in the alloys. The
magnitude of global warming scores from silver processing (118 kg CO2-equivalents) exceed
those from tin processing (114 kg CO2-equivalents) in the SAC alloy, even though the silver
content of the alloys (0.6 percent) is much less than the tin content (95.5 percent). This is due to
the relatively high energy intensity of silver extraction and processing compared to the other
solder metals. Tin production accounts for the majority of the upstream impacts for the
3-51
-------
remaining solders which do not contain silver and have a tin content of at least 67 percent.
Table 3-32. Global warming impacts by life-cycle stage and
process group (bar solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
SnCu
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Total
5.37E+01 28.6
3.47E+00 1.85
N/A N/A
N/A N/A
5.72E+01 30.5
1.14E+02 31.8
N/A N/A
1.18E+02 32.9
1.30E-01 0.0365
2.31E+02 64.8
8.78E+01 40.7
N/A N/A
N/A N/A
1.28E-01 0.0593
8.79E+01 40.8
MANUFACTURING
Solder manufacturing
Post-industrial
recycling
Total
1.58E+00 0.840
9.53E+00 5.08
1.11E+01 5.92
2.42E+00 0.677
2.77E+00 0.775
5.19E+00 1.45
2.40E+00 1.11
4.75E+00 2.20
7.15E+00 3.32
USE/APPLICATION
Wave solder
application
Total
1.19E+02 63.2
1.19E+02 63.2
1.20E+02 33.6
1.20E+02 33.6
1.20E+02 55.6
1.20E+02 55.6
END-OF-LIFE
Landfill
Incineration
Demanufacturing
Cu smelting
Unregulated
Total
GRAND TOTAL
1.59E-02 0.0085
-4.47E-02 -0.0238
8.18E-02 0.0436
6.36E-01 0.339
O.OOE+00 0.00
6.89E-01 0.37
1.87E+02 100
1.39E-02 0.0013
-3.91E-02 -0.0038
7.16E-02 0.0066
5.57E-01 0.0516
O.OOE+00 0.00
6.03E-01 0.0557
3.57E+02 100
1.38E-02 0.0064
-3.88E-02 -0.0180
7.11E-02 0.0330
5.53E-01 0.256
O.OOE+00 0.00
5.99E-01 0.28
2.16E+02 100
*The impact scores are in units of CO2-equivalents/l,000 cubic centimeters of solder applied to a printed wiring
board.
N/A=not applicable
Global warming impacts from the manufacturing life-cycle stage are small compared to
the upstream and use/application life-cycle stages and are nearly evenly distributed between
solder manufacturing and post-industrial recycling, with the exception of SnPb. Global warming
impacts from the use/application stage are due entirely to the electricity consumed in the wave
solder application process. These impacts are less dominant for bar solders than for the solder
pastes, due to the reduced energy consumption per functional unit required by the wave process
when compared to reflow assembly. For example, the global warming impacts for SnPb solder
paste of 832 kg CO2-equivalents greatly exceed the 119 kg CO2-equivalents of global warming
impacts for the wave application of SnPb bar solders.
EOL processes contribute less than 0.37 percent of life-cycle global warming impacts for
any of the solders, with the majority coming from smelting processes that recover copper and
other valuable metals from waste electronics. Negative global warming impacts from
incineration are due to the energy credit for incineration with energy recovery. No global
3-52
-------
warming impacts are shown for unregulated disposal as the inventory for this process does not
include any global warming gas emissions or energy sources as inputs. Some energy is
consumed, however, when waste PWBs are heated to recover solder and valuable components.
The amount of energy consumed, and the resulting global warming gases emitted in this process
are not known, but are expected to be relatively small.
Top Contributors to Global Warming Impacts (Bar Solder)
Table 3-33 presents the specific materials or flows contributing at least 1 percent of the
global warming impacts by solder. Consistent with the results presented above, global warming
gases generated from the production of electricity in the use/application stage, along with those
generated from the upstream extraction and processing of the metals, are the top contributors to
overall global warming impacts. Carbon dioxide is the single greatest contributor for all of the
solders, comprising at least 95 percent of the global warming releases. CO2 is primarily emitted
from coal-fired power generation (coal is the primary fuel used to generate electricity in the U.S.
electric grid), but also is emitted during various upstream metal production processes. Methane
is the only other listed contributor to global warming, resulting from the silver production
process or from the generation of electricity used during the use/application stage. The
extraction and processing inventories are from secondary data sources that do not distinguish
whether global warming gases are emitted from electric power plants producing electricity for
the metals production processes or emitted directly during extraction and processing.
Table 3-33. Top contributors to global warming impacts (bar solder)
Solder
SnPb
SAC
SnCu
Life-Cycle Stage
Use/application
Upstream
Manufacturing
Use/application
Upstream
Use/application
Upstream
Upstream
Upstream
Use/application
Use/application
Upstream
Manufacturing
Use/application
Process
Electricity generation
Tin production
Electricity generation for
post-industrial recycling
Electricity generation
Lead production
Electricity generation
Tin production
Silver production
Silver production
Electricity generation
Electricity generation
Tin production
Electricity generation for
post-industrial recycling
Electricity generation
Flow
Carbon dioxide
Carbon dioxide
Carbon dioxide
Methane
Carbon dioxide
Carbon dioxide
Carbon dioxide
Carbon dioxide
Methane
Methane
Carbon dioxide
Carbon dioxide
Carbon dioxide
Methane
% Contribution
60.5
28.6
4.58
1.96
1.74
32.1
31.8
31.2
1.61
1.04
53.3
40.7
1.88
1.72
3-53
-------
3.2.4.4 Limitations and uncertainties
Similar to the resource and energy impacts presented in Sections 3.2.1 and 3.2.2,
respectively, the generation of electricity for the use/application stage is a major contributor to
global warming impacts. As a result the same sources of uncertainty from the inventory apply:
(1) reflow energy during application is based on a limited number of data points that cover a
wide range, and (2) electricity production data are from secondary sources. Uncertainties in the
reflow energy data are evaluated in a sensitivity analysis (see Section 3.3), but uncertainties in
the electricity production data are not considered large enough to warrant any further analysis.
Limitations to this impact category also arise from aspects of the LCIA methodology.
GWP refers to the warming that emissions of certain gases—by building up in the atmosphere
and trapping the Earth's heat—may contribute. The LCIA methodology for global warming
impacts uses published GWP equivalency factors having effects in the 100-year time horizon.
These effects are expected to be far enough into the future that releases occurring throughout the
life-cycle of solder on a PWB would be within the 100-year time frame.
The effects of the buildup of global warming gases in the atmosphere may still be the
subject of scientific debate, but in 1995, the IPCC, representing the consensus of most climate
scientists worldwide, concluded that "...the balance of evidence...suggests that there is a
discernable human influence on global climate (IPCC, 1995)." As discussed above, other than
the limitations and uncertainties inherent in predicting future effects, most of the limitations and
uncertainties in the global warming results have to do with the LCI data on greenhouse gas
emissions that occur primarily from electricity generation processes.
3-54
-------
3.2.5 Stratospheric Ozone Depletion Impacts
3.2.5.1 Characterization
The stratospheric ozone layer filters out harmful ultraviolet radiation from the sun.
Chemicals such as chlorofluorocarbons, if released to the atmosphere, may result in ozone-
destroying chemical reactions. Stratospheric ozone depletion refers to the release of chemicals
that may contribute to this effect. Impact scores are based on the identity and amount of ozone
depleting chemicals released to air. Currently identified ozone depleting chemicals are those
with ozone depletion potential (ODP), which measure the change in the ozone column in the
equilibrium state of a substance compared to the reference chemical chlorofluorocarbon (CFC),
CFC-11 (trichlorofluromethane) (Heijungs etal., 1992; CAAA, 1990). The list of ODPs that are
used in this methodology are provided in Appendix D. The individual chemical impact score for
stratospheric ozone depletion is based on the ODP and inventory amount of the chemical:
(ISOD), = (EFoopxAmtooc),
where:
ISOD equals the ozone depletion (OD) impact score for chemical /' (kg CFC-11
equivalents) per functional unit;
EFODP equals the ODP equivalency factor for chemical / (CFC-11 equivalents)
(Appendix D); and
AmtODC equals the amount of ozone depleting chemical / released to air (kg) per
functional unit.
3.2.5.2 Paste solder results
Total Stratospheric Ozone Depletion Impacts by Life-Cycle Stage (Paste Solder)
Table 3-34 presents the solder paste results for stratospheric ozone depletion impacts by
life-cycle stage, based on the impact assessment methodology presented above. The table lists
the stratospheric ozone depletion impact scores per functional unit for the life-cycle stages of
each solder paste alloy, as well as the percent contribution of each life-cycle stage to the total
impacts. Figure 3-12 presents the results in a stacked bar chart.
3-55
-------
Table 3-34. Stratospheric ozone depletion impacts by life-cycle stage (paste solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
4.85E-07 0.488
1.88E-06 1.89
9.69E-05 97.4
2.47E-07 0.248
9.95E-05 100
SAC
Score* %
1.64E-05 14.9
2.28E-06 2.08
9.10E-05 82.8
2.13E-07 0.194
1.10E-04 100
BSA
Score* %
7.58E-06 9.50
1.01E-06 1.26
7.12E-05 89.2
4.83E-08 0.0605
7.98E-05 100
SABC
Score* %
1.06E-05 10.1
2.28E-06 2.18
9.13E-05 87.5
2.14E-07 0.205
1.04E-04 100
*The impact scores are in units of kilograms CFC-11 -equivalents/1,000 cubic centimeters of solder applied to a
printed wiring board.
Following a pattern similar to energy and global warming impacts, the reflow of SAC
solder has the greatest impact category indicator for stratospheric ozone depletion at 0.00011 kg
of CFC-11-equivalents per functional unit, closely followed by SABC at 0.000104 kg of CFC-
11-equivalents, and SnPb at 0.0000995 kg of CFC-11-equivalents. BSA results are substantially
lower at 0.0000798 kg of CFC-11-equivalents per functional unit. It should be noted, that all of
the materials contributing to this impact category are listed as Class I ozone depleting substances
in Title VI of the 1990 Clean Air Act Amendments (CAAA), and, therefore, were phased-out of
U.S. production as of January 1, 1996, with the exception of methyl bromide, which will be
mainly phased-out by 2005. Production of these substances also was phased-out in other
developed countries under the Montreal Protocol and its Amendments and Adjustments, but is
permitted in developing countries until 2010 or 2015, depending on the substance. The
uncertainties associated with having phased-out substances in the inventory and, therefore, in the
LCIA results, are discussed further below.
1 onp r\A
*j
'c
D
•= -I nnp r\A
0
+J
o
c & nnp n^
S=
J2
S K nnp nR
ro
_>
'D
o" A nnp n^
i
fr* 9 OOF-05 -
O
S
O.OOE+00 -
^^=
SnPb
SAC BSA SABC
• End-of-life
n Use/application
• Manufacturing
n Upstream
Figure 3-12. Solder Paste Total Life-Cycle Impacts: Stratospheric Ozone Depletion
3-56
-------
As shown in the table and figure, the use/application stage dominates ozone depletion
impacts for all of the solders, accounting for 83 to 97 percent of impacts depending on the alloy.
The upstream processes contribute a larger portion of the total impacts for lead-free alternatives
than they do for SnPb. In fact, for SAC and SABC, the scores for the upstream processes are
high enough to cause the total impacts from these alternatives to exceed those from SnPb,
despite the fact that SnPb use/application impacts are the greatest of all the alloys (6.1 percent
higher than SABC, 6.5 percent higher than SAC). The upstream life-cycle stage for SnPb
contributes less than 1 percent, while the upstream impacts for the three alternatives contribute 9
to 15 percent of the total life-cycle impacts. Solder manufacturing contributes 1 to 2 percent of
the total stratospheric ozone depletion impacts, and EOL processes contribute less than 0.3
percent for all alloys.
Stratospheric Ozone Depletion Impacts by Process Group (Paste Solder)
Table 3-35 lists the stratospheric ozone depletion impacts of each of the processes in the
life-cycle of a solder. Ozone depletion impacts in the use/application stage are due entirely to
electricity consumed in the solder reflow process. Upstream ozone depletion impacts, on the
other hand, arise from emissions from the extraction and processing of the various metals present
in the alloys. It is noteworthy that there are no impacts from Sn production, despite the fact that
tin is the largest or second largest metal component in each of the alloys. There is a small
contribution to the impact category from lead processing for the SnPb alloy (4.85 x 10"7 kg CFC-
11-equivalents per functional unit), with silver being the largest contributor for the lead-free
alloys (e.g. 1.63 x 10"5 kg CFC-11-equivalents for SAC). Bismuth also is a significant
contributor to the BSA upstream impacts (2.70 x 10"6 kg CFC-11-equivalents per functional
unit).
Ozone depletion impacts from the manufacturing life-cycle stage are small compared to
the use/application life-cycle stage. Manufacturing impacts are from energy consumed in solder
manufacturing and post-industrial recycling. The distribution of the manufacturing impacts
between these two processes is similar to that found for energy and global warming impacts, as
discussed in Sections 3.2.2 and 3.2.4. EOL processes contribute less than 0.3 percent of total
stratospheric ozone depletion impacts for any of the solders, with the majority coming from
smelting processes used to recover copper and other valuable metals from waste electronics.
The landfilling process group, which includes diesel fuel production, is the second greatest
contributor to EOL impacts. There are no ozone depletion impacts from incineration or
unregulated disposal as no ozone-depleting substances are emitted from these processes.
3-57
-------
Table 3-35. Stratospheric ozone depletion impacts by life-cycle
stage and process group (paste solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
BSA
Score* %
SABC
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Bi production
Total
O.OOE+00 0.00
4.85E-07 0.440
N/A N/A
N/A N/A
N/A N/A
4.85E-07 0.440
O.OOE+00 0.00
N/A N/A
1.63E-05 13.5
2.68E-08 0.0222
N/A N/A
1.64E-05 13.5
O.OOE+00 0.00
N/A N/A
4.88E-06 5.56
N/A N/A
2.70E-06 3.08
7.58E-06 8.64
O.OOE+00 0.00
N/A N/A
1.05E-05 9.09
2.24E-08 0.0194
4.08E-08 0.0353
1.06E-05 9.15
MANUFACTURING
Solder
manufacturing
Post-industrial
recycling
Total
4.52E-07 0.410
1.43E-06 1.29
1.88E-06 1.71
6.75E-07 0.557
1.61E-06 1.33
2.28E-06 1.88
4.23E-07 0.482
5.84E-07 0.666
1.01E-06 1.15
6.77E-07 0.585
1.60E-06 1.38
2.28E-06 1.97
USE/APPLICATION
Reflow application
Total
1.08E-04 97.6
1.08E-04 97.6
1.02E-04 84.4
1.02E-04 84.4
7.91E-05 90.2
7.91E-05 90.2
1.03E-04 88.7
1.03E-04 88.7
END-OF-LIFE
Landfill
Incineration
Demanufacture
Cu smelting
Unregulated
Total
GRAND TOTAL
3.61E-08 0.0327
O.OOE+00 0.00
9.52E-09 0.0086
2.01E-07 0.182
O.OOE+00 0.00
2.47E-07 0.224
1.10E-04 100
3.12E-08 0.0258
O.OOE+00 0.00
8.24E-09 0.0068
1.74E-07 0.144
O.OOE+00 0.00
2.13E-07 0.176
1.21E-04 100
3.86E-08 0.0440
O.OOE+00 0.00
9.71E-09 0.0111
N/A N/A
O.OOE+00 0.00
4.83E-08 0.0550
8.77E-05 100
3.13E-08 0.0271
O.OOE+00 0.00
8.27E-09 0.0072
1.75E-07 0.151
O.OOE+00 0.00
2.14E-07 0.185
1.16E-04 100
*The impact scores are in units of kilograms CFC-11 -equivalents/1,000 cubic centimeters of solder applied to a
printed wiring board.
N/A=not applicable
Top Contributors to Stratospheric Ozone Depletion Impacts (Paste Solder)
Table 3-36 presents the specific materials or flows contributing at least 1 percent of
ozone depletion impacts by solder. As expected from the results presented above, ozone-
depleting substances emitted during the production of electricity in the use/application stage are
the top contributors to overall ozone depletion impacts, with CFC-114
(dichlorotetrafluoroethane) and CFC-1 l(trichlorofluoromethane) being the two greatest
contributors for each of the solders. Other top contributors include CFC-12
(dichlorodifluoromethane), Hal on-1301, and CFC-13 (chlorotrifluorom ethane), which are
released from either electricity generation, silver production, or bismuth production. The
extraction and processing inventories are from secondary data sources that do not distinguish
whether the ozone-depleting substances are emitted from electric power used or directly emitted
during extraction and processing.
3-58
-------
Table 3-36. Top contributors to stratospheric ozone depletion impacts (paste solder)
Solder
SnPb
SAC
BSA
SABC
Life-Cycle Stage
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Upstream
Use/application
Upstream
Upstream
Use/application
Upstream
Upstream
Use/application
Upstream
Upstream
Upstream
Upstream
Upstream
Use/application
Use/application
Use/application
Use/application
Use/application
Upstream
Upstream
Upstream
Process
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Silver production
Electricity generation
Silver production
Silver production
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Silver production
Silver production
Silver production
Bismuth production
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Silver production
Silver production
Silver production
Flow
CFC-114
CFC-11
CFC-12
Halon(1301)
CFC-13
CFC-114
CFC-11
CFC-12
Halon(1301)
Halon(1301)
CFC-13
CFC-114
CFC-11
CFC-114
CFC-11
CFC-12
Halon(1301)
CFC-13
Halon(1301)
CFC-114
CFC-11
CFC-114
CFC-114
CFC-11
CFC-12
Halon(1301)
CFC-13
Halon(1301)
CFC-114
CFC-11
% Contribution
39.3
38.4
8.25
6.30
5.18
33.4
32.6
7.02
5.36
5.16
4.41
4.19
4.09
36.0
35.0
7.55
5.77
4.74
2.12
1.72
1.53
1.10
35.3
34.5
7.41
5.66
4.65
3.49
2.84
2.77
CFC-114 (dichlorotetrafluoroethane); CFC-11 (trichlorofluoromethane);
CFC-12 (dichlorodifluoromethane); CFC-13 (chlorotrifluoromethane)
While the top contributing flows to ozone depletion impacts result from three different
processes—electricity, silver production, and bismuth production—there are a total of nine
processes for all of the solder paste alloys within the life-cycle that emit ozone depleting
substances (shown in the tables in Appendix D). These include electricity generation, selected
fuel production (heavy fuel oil/#6, light fuel oil/#2, LPG, and diesel fuel), and selected ME&P
(lead, silver, copper, and bismuth). The inventories for all these processes are from secondary
data sources.
Table 3-37 lists the ozone-depleting substances released in the LFSP and their status
under the U.S. CAAA and the Montreal Protocol. In addition to the five top contributors to total
ozone depletion impacts shown in Table 3-36, two additional substances are relatively minor
contributors to the results: methyl bromide and 1,1,1-trichloroethane. As shown in the table and
discussed previously, all of these substances are Class I ozone depleting substances that were
phased-out of production in the U.S. and developed countries as of 1996. An exception is
5-59
-------
methyl bromide, which is designated for phase-out in 2005, except for certain critical uses. All
of these substances are still permitted in developing countries, but will be phased-out by 2010 or
2015, depending on the substance. The presence of phased-out substances in the inventories
makes ozone depletion results highly uncertain, since it is unlikely they are still in use in areas
covered by the geographic boundaries of the LFSP inventories. For example, most of the
greatest ozone depletion impacts occur from U.S. electricity generation, yet it is unlikely U.S.
power manufacturers continue to use these substances in routine operations. The implications of
these uncertainties are discussed further below in Section 3.2.5.4.
Table 3-37. Ozone-depleting substances in the LFSP inventories
Substance
Methyl bromide
Halon(1301)
Trichloroethane, 111- (methyl
chloroform)
CFC-13 (chlorotrifluoromethane)
CFC-12
(dichlorodifluoromethane)
CFC-114
(dichlorotetrafluoroethane)
CFC-11 (trichlorofluoromethane)
Associated process(es)3
LPG production
All processes
LPG production
All processes except LPG
production
All processes except LPG
production
All processes except LPG
production
All processes except LPG
production
CAA"
Class I
Class I
Class I
Class I
Class I
Class I
Class I
Montreal Protocol0
Total phase out for all but certain
critical uses by 2005 or 2015
Phased out by end of 1993 or 2010
Phased out by end of 1995 or 2015
Phased out by end of 1995 or 2010
Phased out by end of 1995 or 2010
Phased out by end of 1995 or 2010
Phased out by end of 1995 or 2010
a Processes in LFSP that emit ozone-depleting substances are as follows: electricity generation, heavy fuel oil/#6,
light fuel oil/#2, LPG, diesel fuel, lead, silver, copper, and bismuth.
b U.S. EPA regulations required the phase-out of Class I ozone-depleting substances, as listed in Title VI of the
U.S. CAAA,asof 1996.
0 Montreal Protocol phase outs for ozone-depleting substances differ for developed and developing countries; the
earlier dates refer to developed countries and the later dates refer to developing countries.
3.2.5.3 Bar solder results
Total Stratospheric Ozone Depletion Impacts by Life-Cycle Stage (Bar Solder)
Table 3-38 presents the bar solder results for stratospheric ozone depletion impacts by
life-cycle stage, based on the impact assessment methodology presented above. The table lists
the stratospheric ozone depletion impact scores per functional unit for the life-cycle stages of
each solder paste alloy, as well as the percent contribution of each life-cycle stage to the total
impacts. Figure 3-13 presents the results in a stacked bar chart.
3-60
-------
Table 3-38. Stratospheric ozone depletion impacts by life-cycle stage (bar solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
4.33E-07 2.32
2.63E-06 14.1
1.53E-05 82.1
2.74E-07 1.47
1.87E-05 100
SAC
Score* %
2.43E-05 58.8
1.29E-06 3.11
1.55E-05 37.5
2.40E-07 0.58
4.13E-05 100
SnCu
Score* %
4.40E-08 0.25
1.98E-06 11.1
1.55E-05 87.3
2.38E-07 1.34
1.78E-05 100
*The impact scores are in units of kilograms CFC-11 -equivalents/1,000 cubic centimeters of bar solder applied to a
printed wiring board.
0 000045
0 00004
'E
D
•5 0 000035
o
t5 0 00003
c
a
"S n nnnno^
c
Q 000005
n
SnPb
SAC
SnCu
D End-of-life
n Use/application
• Manufacturing
n Upstream
Figure 3-13. Bar Solder Total Life-Cycle Impacts: Stratospheric Ozone Depletion
SAC bar solder with 0.0000413 kg CFC-11 equivalents per functional unit had more than
two times the number of ozone depletion impacts as the other bar solders. SnPb and SnCu
follow with 0.0000187 and 0.0000178 kg CFC-11 equivalents per functional unit respectively.
Unlike the solder pastes, this pattern differs slightly from the energy use and global warming
impacts, where SnCu had slightly greater impacts than the baseline SnPb bar solder; however, it
should again be noted that all of the materials contributing to this impact category are listed as
Class I ozone depleting substances in Title VI of the 1990 CAAA and, therefore, were phased-
out of U.S. production as of January 1, 1996, with the exception of methyl bromide, which will
be mainly phased-out by 2005. Production of these substances also was phased-out in other
developed countries under the Montreal Protocol and its Amendments and Adjustments, but is
permitted in developing countries until 2010 or 2015, depending on the substance. The
3-61
-------
uncertainties associated with having phased-out substances in the inventory, and therefore, in the
LCIA results, are further discussed below.
As shown in the table and figure, the ozone depletion impacts from the use/application
stage dominate for the SnCu and SnPb solders, accounting for 87 and 82 percent respectively.
Despite the use/application stage impact scores for the solders being virtually identical, ranging
from 1.53 x 10"5 to 1.55 xlO"5 kg CFC-11 equivalents per functional unit, the use/application
stage accounted for just 38 percent of the overall ozone depletion impacts for the SAC alloy. The
upstream stage impacts for SAC totaled 0.0000243 kg CFC-11 equivalents, or nearly 59 percent
of the ozone depletion impact score. Upstream impacts for SnPb and SnCu accounted for less
than 2.3 percent of the total impacts scores for these alloys. Manufacturing processes accounted
for only 3.1 percent of the impacts for SAC, but ranged from 11 to 14 percent of the impacts of
the non-silver containing solders. End-of-life impacts for all 3 bar solders contributed less than
1.5 percent of the overall impact scores.
Stratospheric Ozone Depletion Impacts by Process Group (Bar Solder)
Table 3-39 lists the stratospheric ozone depletion impacts of each of the processes in the
life-cycle of a solder. Ozone depletion impacts in the use/application stage are due entirely to
electricity consumed in the solder wave process. Upstream ozone depletion impacts, on the
other hand, arise from emissions from the extraction and processing of the various metals present
in the alloys. It is noteworthy that there are no impacts from tin production, despite the fact that
tin is the largest or second largest metal component in each of the alloys. There is a small
contribution to the impact category from silver processing for the SnPb alloy (4.33 x 10"7 kg
CFC-11-equivalents per functional unit), with silver being the largest contributor for SAC
(e.g. 2.43 x 10"5 kg CFC-11-equivalents for SAC). Copper production makes a minimal
contribution to the overall ozone depletion impact score.
Ozone depletion impacts from the manufacturing life-cycle stage are small compared to
the use/application life-cycle stage, though they contribute more than 11 percent of the overall
impact score for the non-silver alloys. Manufacturing impacts are from energy consumed in
solder manufacturing and post-industrial recycling, with post-industrial recycling accounting for
the majority of the impacts. The distribution of the manufacturing impacts between these two
processes is similar to that found for energy and global warming impacts, discussed in Sections
3.2.2 and 3.2.4. EOL processes contribute less than 1.5 percent of total stratospheric ozone
depletion impacts for any of the solders, with the majority coming from smelting processes used
to recover copper and other valuable metals from waste electronics. The landfilling process
group, which includes diesel fuel production, is the second greatest contributor to EOL impacts.
There are no ozone depletion impacts from incineration or unregulated disposal as no ozone-
depleting substances are emitted from these processes.
3-62
-------
Table 3-39. Stratospheric ozone depletion impacts by life-cycle
stage and process group (bar solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
SnCu
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Total
O.OOE+00 0.00
4.33E-07 2.32
N/A N/A
N/A N/A
4.33E-07 2.32
O.OOE+00 0.00
N/A N/A
2.43E-05 58.7
4.48E-08 0.108
2.43E-05 58.8
O.OOE+00 0.00
N/A N/A
N/A N/A
4.40E-08 0.247
4.40E-08 0.247
MANUFACTURING
Solder manufacturing
Post-industrial
recycling
Total
2.27E-07 1.21
2.40E-06 12.9
2.63E-06 14.1
3.14E-07 0.759
9.72E-07 2.35
1.29E-06 3.11
3.12E-07 1.75
1.674E-06 9.38
1.98E-06 11.1
USE/APPLICATION
Reflow application
Total
1.53E-05 82.1
1.53E-05 82.1
1.55E-05 37.5
1.55E-05 37.5
1.55E-05 87.3
1.55E-05 87.3
END-OF-LIFE
Landfill
Incineration
Demanufacturing
Cu smelting
Unregulated
Total
GRAND TOTAL
4.01E-08 0.215
O.OOE+00 0.00
1.06E-08 0.057
2.23E-07 1.20
O.OOE+00 0.00
2.74E-07 1.47
1.87E-05 100
3.51E-08 0.0848
O.OOE+00 0.00
9.26E-09 0.0224
1.95E-07 0.473
O.OOE+00 0.00
2.40E-07 0.580
4.13E-05 100
3.48E-08 0.196
O.OOE+00 0.00
9.20E-09 0.0517
1.94E-07 1.09
O.OOE+00 0.00
2.38E-07 1.34
1.78E-05 100
*The impact scores are in units of kilograms CFC-11 -equivalents/1,000 cubic centimeters of bar solder applied to a
printed wiring board.
N/A=not applicable
Top Contributors to Stratospheric Ozone Depletion Impacts (Bar Solder)
Table 3-40 presents the specific materials or flows contributing at least 1 percent of
ozone depletion impacts by solder. As expected from the results presented above, ozone-
depleting substances emitted during the production of electricity in the use/application stage are
the top contributors to overall ozone depletion impacts, with CFC-114 and CFC-11 being the
two greatest contributors for each of the solders. Other top contributors include CFC-12, Halon-
1301, and CFC-13, which are released from electricity generation, silver production, or the
production of heavy fuel oil used in post-industrial recycling. The extraction and processing
inventories are from secondary data sources that do not distinguish whether the ozone-depleting
substances are emitted from electric power used or directly emitted during extraction and
processing.
3-63
-------
Table 3-40. Top contributors to stratospheric ozone depletion impacts (bar solder)
Solder
SnPb
SAC
SnCu
Life-Cycle Stage
Use/application
Use/application
Use/application
Manufacturing
Use/application
Use/application
Manufacturing
Manufacturing
Upstream
Upstream
Upstream
Use/application
Use/application
Upstream
Use/application
Use/application
Upstream
Use/application
Manufacturing
Use/application
Use/application
Use/application
Manufacturing
Use/application
Use/application
Manufacturing
Manufacturing
Process
Electricity generation
Electricity generation
Electricity generation
Heavy fuel oil (#6)
production, post-industrial
recycling
Electricity generation
Electricity generation
Electricity generation, post-
industrial recycling
Electricity generation, post-
industrial recycling
Silver production
Silver production
Silver production
Electricity generation
Electricity generation
Silver production
Electricity generation
Electricity generation
Silver production
Electricity generation
Heavy fuel oil (#6)
production, post-industrial
recycling
Electricity generation
Electricity generation
Electricity generation
Heavy fuel oil (#6)
production, post-industrial
recycling
Electricity generation
Electricity generation
Electricity generation, post-
industrial recycling
Electricity generation, post-
industrial recycling
Flow
CFC-114
CFC-11
CFC-12
Halon(1301)
Halon(1301)
CFC-13
CFC-114
CFC-11
Halon(1301)
CFC-114
CFC-11
CFC-114
CFC-11
CFC-12
CFC-12
Halon(1301)
CFC-13
CFC-13
Halon(1301)
CFC-114
CFC-11
CFC-12
Halon(1301)
Halon(1301)
CFC-13
CFC-114
CFC-11
% Contribution
33.1
32.4
6.96
5.98
5.31
4.37
2.51
2.45
20.3
16.5
16.1
15.1
14.8
3.47
3.18
2.43
2.18
2.00
1.49
35.2
34.4
7.39
5.94
5.65
4.64
1.24
1.22
While the top contributing flows to ozone depletion impacts result from three different
processes—electricity, silver production, and heavy fuel oil production—there are a total of nine
processes for all of the solder paste alloys within the life-cycle that emit ozone depleting
substances (shown in the tables in Appendix D). These include electricity generation, selected
fuel production (heavy fuel oil/#6, light fuel oil/#2, LPG, and diesel fuel), and selected ME&P
(lead, silver, copper, and bismuth). The inventories for all these processes are from secondary
data sources.
In addition to the top contributing ozone depleting substances presented above, two other
substances, methyl bromide and trichloroethane- 1,1,1, also are emitted from bar solder life-
cycle processes. All of these substances either have been designated or already have been
3-64
-------
phased out in the U.S. Please refer to the paste solder section above (Section 3.2.5.3) and for
further discussion of this issue and the potential limitations and uncertainties.
3.2.5.4 Limitations and uncertainties
The major contributors to stratospheric ozone depletion impacts are from the generation
of electricity for the use/application stage and from silver production. These contributors,
therefore, are subject to the same sources of uncertainty in the use/application stage inventory:
(1) reflow energy consumption during application/use is based on a limited number of data
points that cover a wide range, and (2) electricity production data are from a secondary source.
Uncertainties in the reflow energy data are the subject of a sensitivity analysis (see Section 3.3),
but uncertainties in the electricity production data are considered relatively minor.
The silver inventory, which contributes significantly to the ozone depletion impact score
for SAC, warrants discussion here. Uncertainties related to the silver inventory are described in
Section 3.2.2.3, and have to do with the fact that two alternate silver inventories available to the
LFSP vary significantly in the magnitude of flows from silver production. Section 3.2.2.3
concludes that although the GaBi data set used in this analysis is considered "good' by GaBi,
there remains enough uncertainty to perform an additional analysis using the alternate inventory
from the DEAM database. Results of the alternate analysis are presented in Section 3.3.
The principle difference between paste and bar solder are the manufacturing of the solder
and the manner in which it is applied. For bar solder, the wave application data are expected to
be representative of general wave operations of good quality. The remaining uncertainty,
although expected to be small, is that the electricity production data used for the wave operations
are derived from secondary data.
Perhaps the most significant source of uncertainty in the ozone depletion results is the
presence of phased-out substances in the inventory. In order to better assess these uncertainties,
Table 3-41 lists the geographic and temporal boundaries for the life-cycle inventories of the
processes that emit ozone-depleting substances. As shown in the table, these processes contain
data from developed countries and from dates that precede the phase-out dates; therefore, if it is
assumed that these substances were indeed phased out as required, only methyl bromide would
be included in the inventory.
Figure 3-14 presents ozone depletion impact results for solder paste if only methyl
bromide were in the inventory. Methyl bromide emissions result from the production of LPG,
which is used in post-industrial recycling (manufacturing life-cycle stage) and copper smelting
(EOL life-cycle stage). The figure shows that only upstream and EOL life-cycle stages
contribute to these results. This is in contrast to the results presented in Figure 3-12, which are
based on the inventory using the phased-out substances.
3-65
-------
Table 3-41. Geographic and temporal boundaries of inventories
contributing to the ozone depletion results
Process
Electricity generation
Heavy fuel oil/#6
Lishtfueloil/#2
LPG production
Diesel fuel production
Lead production
Silver production
Copper production
Bismuth production
Geographic boundaries
United States
Germany
Germany
Mainly United States
Germany
Germany
"Global" (Canada, Sweden)
Germany
Germany
Temporal boundaries
1995
1994
1994
1980-1993
1994
1995
1995
1994-1996
1994-1996
1 40F-10
*;
£ 1 90F 10
15
c
.0 -| noF 10
o
c
a
"»*; ft nnp 1 1
"c
_o
> fi OOF 11
D
o-
9
T- A nnp -\ 1
o
LL.
O o nop 1 1
u>
^£
n nop+oo
•
SnPb
^^H
I
SAC
BSA
^^^m
"
3ABC
• EOL
• Manufacturing
Figure 3-14. Ozone Depletion Impacts with Methyl Bromide Only (Paste Solder)
The results in Figure 3-14, compared to those presented in Figure 3-12 show that SAC
still has the greatest impact score, followed by SABC. SnPb has the third greatest impact score,
as shown in Figure 3-12. Adjustment of the inventory to exclude materials due to their expected
phase-out has resulted in an even greater gap between BSA and their other solders. As expected,
the total impact scores for stratospheric ozone depletion are much less (ranging from about
3.43 x 10"11 to 1.22 x 10"10 kg CFC-11- equivalents/functional unit) compared to the results in
Figure 3-12, which range from 8.77 x 10"5 to 1.21 x 10"4 kg CFC-11-equivalents/functional unit;
however, it should be noted that even these results are uncertain since the schedule for methyl
bromide phase-out required a 25 percent reduction in 1999 and a 70 percent reduction in 2003.
3-66
-------
Given the phase-out schedule, and the fact that many manufacturers have actively pursued
alternatives for non-critical uses of methyl bromide, it is entirely possible that methyl bromide is
no longer used in LPG production.
In conclusion, the major limitation to the ozone depletion results is that many of the
flows contributing to ozone depletion impacts have been theoretically phased-out. Lending to
the uncertainty is the fact that if the ozone-depleting substances have indeed been phased-out,
any substitute materials have not been inventoried in this study.
3-67
-------
3.2.6 Photochemical Smog Impacts
3.2.6.1 Characterization
Photochemical oxidants are produced in the atmosphere from sunlight reacting with
hydrocarbons and nitrogen oxides. At higher concentrations they may cause or aggravate health
problems, plant toxicity, and deterioration of certain materials. Photochemical oxidant creation
potential (POCP) refers to the release of chemicals that contribute to this effect. The POCP is
based on simulated trajectories of tropospheric ozone production both with and without volatile
organic carbons (VOCs) present. The POCP is a measure of a specific chemical compared to the
reference chemical ethene (Heijungs et al., 1992). The list of chemicals with POCPs used in this
methodology is presented in Appendix D. As shown in Table 3-42, photochemical smog
impacts are based on partial equivalency because some chemicals cannot be converted into
POCP equivalency factors. For example, nitrogen oxides do not have a POCP; however, VOCs
are assumed to be the limiting factor, and if VOCs are present there is a potential impact. Impact
scores are based on the identity and amount of chemicals with POCP equivalency factors
released to the air and the chemical-specific equivalency factor:
POCP
where:
IS POCp equals the photochemical smog (POCP) impact score for chemical /' (kg ethene
equivalents) per functional unit;
EFPOCp equals the POCP equivalency factor for chemical /' (ethene equivalents)
(Appendix D); and
Amtpoc equals the amount of photochemical smog-creating oxidant /' released to the air
(kg) per functional unit.
3.2.6.2 Paste solder results
Total Photochemical Smog Impacts by Life-Cycle Stage (Paste Solder)
Table 3-42 presents the solder paste results for photochemical smog impacts by life-cycle
stage based on the impact assessment methodology. The table lists the photochemical smog
impact scores per functional unit for the life-cycle stages of each alloy, as well as the percent
contribution of each life-cycle stage to the total impacts. Figure 3-15 shows the results in a
stacked bar chart.
3-68
-------
Table 3-42. Photochemical smog impacts by life-cycle stage (paste solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
1.56E-02 4.98
6.28E-03 2.00
2.91E-01 92.8
6.34E-04 0.202
3.13E-01 100
SAC
Score* %
3.37E-01 54.5
7.38E-03 1.19
2.73E-01 44.2
5.49E-04 0.0888
6.18E-01 100
BSA
Score* %
1.44E-01 39.9
3.47E-03 0.961
2.14E-01 59.2
2.70E-05 0.0075
3.61E-01 100
SABC
Score* %
2.23E-01 44.2
7.38E-03 1.46
2.74E-01 54.3
5.51E-04 0.109
5.05E-01 100
*The impact scores are in units of ethene-equivalents/1,000 cubic centimeters of solder applied to a printed wiring
board.
Figure 3-15. Solder Paste Total Life-Cycle Impacts: Photochemical Smog
07
. /
+j
'E
3 n R
re
•2 n*
-Q °-5
c
^ n 4
(/) U.'f
+-
0)
re n ^
> U.O
'5
a-
a> n o
i U./i
0)
-------
Photochemical Smog Impacts by Process Group (Paste Solder)
Table 3-43 lists the photochemical smog impact scores for each of the processes in the
life-cycle of a solder paste. As with other impact categories, impacts from the use/application
life-cycle stage are entirely from the solder reflow process group. For the lead-free alloys, smog
impacts from upstream processes are due primarily to the silver production process, even though
silver is only a small proportion of the alloy composition. For example, silver production
contributes 25 to 49 percent of the total smog impacts for the lead-free solder alternatives while
the percent composition of silver in those alloys range from 1 to 3.9 percent. For BSA, which is
composed of 57 percent bismuth, only 11 percent of smog impacts are due to bismuth
production.
Table 3-43. Photochemical smog impacts by life-cycle stage
and process group (paste solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
BSA
Score* %
SABC
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Bi production
Total
1.02E-02 2.96
5.37E-03 1.55
N/A N/A
N/A N/A
N/A N/A
1.56E-02 4.51
1.50E-02 2.30
N/A N/A
3.22E-01 49.3
4.27E-04 0.0655
N/A N/A
3.37E-01 51.7
7.67E-03 1.99
N/A N/A
9.61E-02 25.0
N/A N/A
4.03E-02 10.5
1.44E-01 37.4
1.51E-02 2.80
N/A N/A
2.07E-01 38.4
3.57E-04 0.0663
6.09E-04 0.113
2.23E-01 41.4
MANUFACTURING
Solder
manufacturing
Post-industrial
recycling
Total
1.94E-03 0.560
4.34E-03 1.26
6.28E-03 1.82
2.50E-03 0.384
4.88E-03 0.749
7.38E-03 1.13
1.69E-03 0.440
1.78E-03 0.461
3.47E-03 0.901
2.51E-03 0.466
4.87E-03 0.903
7.38E-03 1.37
USE/APPLICATION
Reflow
application
Total
3.23E-01 93.5
3.23E-01 93.5
3.07E-01 47.1
3.07E-01 47.1
2.37E-01 61.7
2.37E-01 61.7
3.08E-01 57.1
3.08E-01 57.1
END-OF-LIFE
Landfill
Incineration
Demanufacturing
Cu smelting
Unregulated
Total
GRAND
TOTAL
1.20E-04 0.0348
-1.23E-04 -0.0355
3.18E-05 0.0092
6.75E-04 0.195
O.OOE+00 0.00
7.05E-04 0.204%
3.46E-01 100%
1.05E-04 0.0162
-1.07E-04 -0.0165
2.78E-05 0.0043
5.91E-04 0.0906
O.OOE+00 0.00
6.17E-04 0.0946
6.52E-01 100
1.29E-04 0.0335
-1.31E-04 -0.0341
3.24E-05 0.0084
N/A N/A
O.OOE+00 0.00
3.01E-05 0.0078
3.85E-01 100
1.06E-04 0.0196
-1.08E-04 -0.0200
2.79E-05 0.0052
5.93E-04 0.110
O.OOE+00 0.00
6.19E-04 0.115
5.39E-01 100
*The impact scores are in units of ethene-equivalents/1,000 cubic centimeter of solder applied to a printed wiring
board.
N/A=not applicable
3-70
-------
Within the manufacturing life-cycle stage, the post-industrial recycling process is a
greater contributor than solder manufacturing for all solder paste alloys except BSA. The
distribution of the manufacturing impacts between these two processes is similar to those found
for energy, and is discussed in Section 3.2.2; however, the manufacturing stage is a small
contributor overall.
EOL processes contribute less than 0.3 percent of total photochemical smog impacts for
any of the solders, with the majority coming from smelting processes used to recover copper and
other valuable metals from waste electronics. The landfilling process group, which includes
diesel fuel production, is the second greatest contributor to EOL impacts. Demanufacturing
contributes less than 0.01 percent for each alloy, and incineration results in a credit based on the
surplus energy generated during the incineration of electronics at EOL.
Top Contributors to Photochemical Smog Impacts (Paste Solder)
Table 3-44 presents the specific materials or flows contributing at least 1 percent of
photochemical smog impacts by solder. As expected from the results above, all the top
contributors are from either the use/application stage or the upstream life-cycle stage. Sulphur
dioxide is the largest contributing individual flow and is emitted during either electricity
production or silver production, depending on the alloy.
For SnPb, sulphur dioxide from the generation of electricity used to reflow solder
contributes about 65 percent to the total smog impact score. Other flows from the
use/application stage for electricity generation, such as unspecified non-methane volatile organic
compounds (NMVOCs), carbon monoxide, xylene, ethane, and methane, all contribute at least 1
percent each to the total smog impact score for SnPb. Other flows for SnPb presented in the
table include sulphur dioxide from tin production (3 percent) and sulphur dioxide from lead
production (1 percent).
Sulphur dioxide resulting from the electricity used in both solder application and silver
production also is the greatest contributor for the silver-containing alloys. The percent
contribution from sulphur dioxide, from both electricity generation for the use/application stage
and silver production combined, range from 66 percent to 79 percent for the lead-free solders.
Others, including unspecified NMVOCs, carbon monoxide, xylene, and methane, contribute at
least 1 percent each of the total impacts per alloy. These flows all result from the production of
the metals required to manufacture the solder paste. The extraction and processing inventories
are from secondary data sources that do not distinguish whether the smog-inducing substances
are emitted from electric power used or directly released during extraction and processing.
3-71
-------
Table 3-44. Top contributors to photochemical smog impacts (paste solder)
Solder
SnPb
SAC
BSA
SABC
Life-Cycle Stage
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
Use/application
Upstream
Upstream
Use/application
Use/application
Upstream
Upstream
Use/application
Use/application
Use/application
Upstream
Use/application
Upstream
Use/application
Upstream
Upstream
Use/application
Use/application
Upstream
Use/application
Upstream
Use/application
Upstream
Upstream
Use/application
Use/application
Use/application
Process
Electricity generation
Electricity generation
Electricity generation
Tin production
Electricity generation
Electricity generation
Electricity generation
Lead production
Silver production
Electricity generation
Electricity generation
Silver production
Tin production
Electricity generation
Electricity generation
Electricity generation
Silver production
Electricity generation
Bismuth production
Electricity generation
Tin production
Silver production
Electricity generation
Electricity generation
Bismuth production
Electricity generation
Silver production
Electricity generation
Tin production
Silver production
Electricity generation
Electricity generation
Electricity generation
Flow
Sulphur dioxide
NMVOC (unspecified)
Carbon monoxide
Sulphur dioxide
Xylene (dimethyl benzene)
Methane
Ethane
Sulphur dioxide
Sulphur dioxide
Sulphur dioxide
NMVOC (unspecified)
NMVOC (unspecified)
Sulphur dioxide
Carbon monoxide
Xylene (dimethyl benzene)
Sulphur dioxide
Sulphur dioxide
NMVOC (unspecified)
Sulphur dioxide
Carbon monoxide
Sulphur dioxide
NMVOC (unspecified)
Xylene (dimethyl benzene)
Methane
NMVOC (unspecified)
Sulphur dioxide
Sulphur dioxide
NMVOC (unspecified)
Sulphur dioxide
NMVOC (unspecified)
Carbon monoxide
Xylene (dimethyl benzene)
Methane
% Contribution
65.1
15.3
4.37
3.08
2.47
1.93
1.38
1.27
47.9
31.0
7.28
3.36
2.29
2.08
1.17
41.5
24.5
9.75
9.65
2.79
2.00
1.72
1.57
1.23
1.17
38.1
37.7
8.95
2.82
2.65
2.56
1.44
1.13
3.2.6.3 Bar solder results
Total Photochemical Smog Impacts by Life-Cycle Stage (Bar Solder)
Table 3-45 presents the bar solder results for photochemical smog impacts by life-cycle
stage, based on the impact assessment methodology presented above (Section 3.2.6.1). The table
lists the photochemical smog impact scores per functional unit for the life-cycle stages of each
alloy, as well as the percent contribution of each life-cycle stage to the total impacts. Figure 3-
16 shows the results in a stacked bar chart.
3-72
-------
Table 3-45. Photochemical smog impacts by life-cycle stage (bar solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
1.47E-02 21.1
8.32E-03 11.9
4.60E-02 65.9
7.11E-04 1.02
6.98E-02 100
SAC
Score* %
4.99E-01 90.6
4.36E-03 0.792
4.66E-02 8.45
6.22E-04 0.113
5.51E-01 100
SnCu
Score* %
1.70E-02 24.0
6.46E-03 9.15
4.66E-02 66.0
6.18E-04 0.876
7.06E-02 100
*The impact scores are in units of kg ethene-equivalents/1,000 cubic centimeters of solder applied to a printed
wiring board.
0 fi
f n ^
C U'° "
D
15
c
° 0 4
*j
o
c
,g
'D n?
o- u'^ "
9
-------
Photochemical Smog Impacts by Process Group (Bar Solder)
Table 3-46 lists the photochemical smog impact scores for each of the processes in the
life-cycle of a bar solder. For SAC, smog impacts from upstream processes are due primarily to
the silver production process, even though silver is only a small proportion of the alloy
composition. For example, silver production contributes 87 percent of the total smog impacts for
SAC, while the percent composition of silver is only 3.9 percent. For SnPb, which is composed
of 63 percent tin, only 14 percent of smog impacts are due to tin production. For SnPb and
SnCu, there is a greater percentage of impacts from tin, which is greater by mass than either lead
or Copper.
Table 3-46. Photochemical smog impacts by
life-cycle stage and process group (bar solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
SnCu
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Total
9.95E-03 14.2
4.79E-03 6.86
N/A N/A
N/A N/A
1.47E-02 21.1
2.11E-02 3.82
N/A N/A
4.77E-01 86.7
7.13E-04 0.130
4.99E-01 90.6
1.63E-02 23.0
N/A N/A
N/A N/A
7.00E-04 0.991
1.70E-02 24.0
MANUFACTURING
Solder manufacturing
Post-industrial recycling
Total
1.02E-03 1.46
7.31E-03 10.5
8.32E-03 11.9
1.41E-03 0.255
2.95E-03 0.536
4.36E-03 0.792
1.40E-03 1.98
5.06E-03 7.17
6.46E-03 9.15
USE/APPLICATION
Solder application
Total
4.60E-02 65.9
4.60E-02 65.9
4.66E-02 8.45
4.66E-02 8.45
4.66E-02 66.0
4.66E-02 66.0
END-OF-LIFE
Landfill
Incineration
Demanufacturing
Cu smelting
Unregulated
Total
GRAND TOTAL
1.20E-04 0.173
-1.16E-04 -0.166
3.18E-05 0.0455
6.75E-04 0.968
O.OOE+00 0.0000
7.11E-04 1.02
6.98E-02 100
1.05E-04 0.0191
-1.02E-04 -0.0184
2.78E-05 0.0050
5.91E-04 0.107
O.OOE+00 0.0000
6.22E-04 0.113
5.51E-01 100
1.05E-04 0.148
-1.01E-04 -0.143
2.76E-05 0.0391
5.87E-04 0.831
O.OOE+00 0.0000
6.18E-04 0.876
7.06E-02 100
*The impact scores are in units of ethene-equivalents/1,000 cubic centimeter of solder applied to a printed wiring
board.
N/A=not applicable
As with other impact categories, impacts from the use/application life-cycle stage are
entirely from the solder reflow process group. Within the manufacturing life-cycle stage, the
post-industrial recycling process is a greater contributor than solder manufacturing for all bar
solder alloys, and varies among solder alloys depending on the percent of metals recycled.
3-74
-------
EOL processes contribute 1 percent or less of the total photochemical smog impacts for
any of the solders, with the majority coming from smelting processes used to recover copper and
other valuable metals from waste electronics. The landfilling process group, which includes
diesel fuel production, is the second greatest contributor to EOL impacts. Demanufacturing
contributes less than 0.05 percent for each alloy, and incineration results in a credit based on the
surplus energy generated during the incineration of electronics at EOL.
Top Contributors to Photochemical Smog Impacts (Bar Solder)
Table 3-47 presents the specific materials or flows contributing at least 1 percent of
photochemical smog impacts by solder. The results show that most of the top contributors are
from either the use/application stage or the upstream life-cycle stage. Sulphur dioxide is the
largest contributing individual flow, and is emitted in largely contributing quantities during
electricity production and metals production.
Table 3-47. Top contributors to photochemical smog impacts (bar solder)
Solder
SnPb
SAC
SnCu
Life-Cycle Stage
Use/application
Upstream
Use/application
Upstream
Manufacturing
Manufacturing
Use/application
Use/application
Use/application
Upstream
Use/application
Upstream
Upstream
Use/application
Use/application
Upstream
Use/application
Use/application
Use/application
Manufacturing
Use/application
Upstream
Process
Electricity generation
Tin production
Electricity generation
Lead production
Heavy fuel oil (#6) for post-industrial
recycling
Electricity generation for post-industrial
recycling
Electricity generation
Electricity generation
Electricity generation
Silver production
Electricity generation
Silver production
Tin production
Electricity generation
Electricity generation
Tin production
Electricity generation
Electricity generation
Electricity generation
Electricity generation for post-industrial
recycling
Electricity generation
Tin production
Flow
Sulphur dioxide
Sulphur dioxide
NMVOC (unspecified)
Sulphur dioxide
NMVOC (unspecified)
Sulphur dioxide
Carbon monoxide
Xylene (dimethyl
benzene)
Methane
Sulphur dioxide
Sulphur dioxide
NMVOC (unspecified)
Sulphur dioxide
NMVOC (unspecified)
Sulphur dioxide
Sulphur dioxide
NMVOC (unspecified)
Carbon monoxide
Xylene (dimethyl
benzene)
Sulphur dioxide
Methane
Carbon monoxide
% Contribution
46.3
13.4
10.9
5.07
4.70
3.50
3.11
1.75
1.37
79.9
5.93
5.60
3.61
1.39
46.3
21.7
10.9
3.11
1.75
1.64
1.37
1.13
3-75
-------
For SnPb, sulphur dioxide from the generation of electricity used in wave soldering
contributes about 46 percent to the total smog impact score. Other flows from the
use/application stage for electricity generation, such as unspecified NMVOCs, carbon monoxide,
xylene, and methane, all contribute at least 1 percent each to the total smog impact score for
SnPb. Other flows for SnPb that are from metals production include sulphur dioxide from tin
production (13 percent) and sulphur dioxide from lead production (5 percent). The
manufacturing stage also contributes 4.7 percent from unspecified NMVOCs and 3.5 percent
from sulphur dioxide, both emitted during post-industrial recycling. The top contributors to the
SnCu alloy are similar to those from SnPb, except that there are no contributions from the lead
production process.
Sulphur dioxide resulting from the electricity used in both solder application and silver
production also is the greatest contributor for the SAC alloy. The percent contribution from
sulphur dioxide from both electricity generation for the use/application stage and silver
production combined is approximately 86 percent. Unspecified NMVOCs also contribute at
least 1 percent from both silver production and electricity generation during application. For the
extraction and processing inventories (e.g., silver production), the secondary data sources do not
distinguish whether the smog-inducing substances are emitted from electric power used or
directly released during extraction and processing.
3.2.6.4 Limitations and uncertainties
For the paste solder results, the two processes that have the top contribution to
photochemical smog impacts are electricity generation for solder reflow application (for all
alloys) and silver production (for the lead-free alloys). As presented earlier, the same sources of
uncertainty from the use/application stage inventory apply: (1) energy consumed during
application/use of the solder paste is based on a limited number of data points that cover a wide
range, and (2) electricity production data were from a secondary source. Energy consumption
during reflow is the subject of a sensitivity analysis in Section 3.3.
For the bar solder results, the wave application data are expected to be representative of
general wave operations and are of good quality. The remaining uncertainty, again not expected
to be too large, is that the electricity production data that are linked to the wave operations are
from secondary data.
Uncertainties related to the silver inventory are described earlier in Section 3.2.1.4,
which concludes that although the GaBi inventory used in this analysis is considered "good" by
GaBi, there remains enough uncertainty that it is the subject of a sensitivity analysis presented
in Section 3.3.
Uncertainty in the smog results also is derived from the impact assessment methodology,
which uses the mass of a chemical released to air per functional unit and the chemical-specific
partial equivalency factor. The equivalency factor is a measure of a chemical's POCP compared
to the reference chemical ethene. As noted in Section 3.1.2, photochemical smog impacts are
based on partial equivalency because some chemicals cannot be converted into POCP
equivalency factors (e.g., nitrogen oxide). The inability to develop equivalency factors for some
chemicals is a limitation of the photochemical smog impact assessment methodology.
3-76
-------
3.2.7 Acidification Impacts
3.2.7.1 Characterization
Acidification impacts refer to the release of chemicals that may contribute to the
formation of acid precipitation. Impact characterization is based on the amount of a chemical
released to air that would cause acidification and the acidification potentials (AP) equivalency
factor for that chemical. The AP equivalency factor is the number of hydrogen ions that can
theoretically be formed per mass unit of the pollutant being released compared to sulfur dioxide
(SO2) (Heijungs et al, 1992; Hauschild and Wenzel, 1997). Appendix D lists the AP values that
were used as the basis of calculating acidification impacts. The impact score is calculated by:
where:
ISAP
EF.
AP
Amt
•AC
(ISAP), = (EFApXAmtAC>l
equals the impact score for acidification for chemical /' (kg SO2 equivalents) per
functional unit;
equals the AP equivalency factor for chemical /' (SO2 equivalents) (Appendix D);
and
equals the amount of acidification chemical / released to the air (kg) per
functional unit.
3.2.7.2 Paste solder results
Total Acidification Impacts by Life-Cycle Stage (Paste Solder)
Table 3-48 presents the solder paste results for acidification impacts by life-cycle stage,
based on the impact assessment methodology presented above. The table lists the acidification
impact scores per functional unit for the life-cycle stages of each solder paste alloy, as well as
the percent contribution of each life-cycle stage to the total impacts. Figure 3-17 presents the
results in a stacked bar chart.
Table 3-48. Acidification impacts by life-cycle stage (paste solder)
_ _ O _T*1_ O A *~1 T»O A
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
3.94E-01 6.06
7.13E-02 1.10
6.03E+00 92.8
4.33E-03 0.0666
6.50E+00 100
SAC
Score* %
6.74E+00 54.0
7.59E-02 0.608
5.66E+00 45.4
3.75E-03 0.0300
1.25E+01 100
BSA
Score* %
2.85E+00 38.9
4.35E-02 0.594
4.43E+00 60.5
-1.40E-04 -0.0019
7.32E+00 100
SABC
Score* %
4.51E+00 43.9
7.59E-02 0.739
5.68E+00 55.3
3.76E-03 0.0366
1.03E+01 100
*The impact scores are in units of kilograms SO2-equivalents/l,000 cubic centimeters of solder applied to a printed
wiring board.
3-77
-------
TO
C
g
"o
O
O)
1°
m
6
n
o
SnPb SAC BSA SABC
• End-of-lif e
n Use/application
n Manufacturing
H Upstream
Figure 3-17. Solder Paste Total Life-Cycle Impacts: Acidification
As shown in the table and figure, SAC solder has the greatest impact category indicator
for acidification with 12.5 kg of SO2-equivalents/functional unit, followed by SABC at 10.3 kg
SO2-equivalents, BSA at 7.32 kg SO2-equivalents/functional unit, and SnPb with the lowest
indicator at 6.50 kg SO2-equivalents/functional unit. Approximately 93 percent of the SnPb life-
cycle acidification impacts are driven by the use/application stage, while the lead-free impacts
are driven by both the upstream and use/application stages. Contributions from solder
manufacturing (less than 1.5 percent of the total life cycle impacts) and EOL processes (less than
0.07 percent) were minimal for all alloys.
Acidification Impacts by Process Group (Paste Solder)
Table 3-49 lists the acidification impacts of each of the processes in the life-cycle of the
solder pastes. The production of energy consumed during the reflow of each of the alloys is the
single greatest contributor for all of the alloys. For the lead-free alloys, upstream processes are
also large contributors, mainly from the silver production process, even though silver comprises
only a small proportion of their compositions. For example, silver production contributes 26 to
50 percent of the total acidification impact scores for the lead-free solder alternatives, while the
percent composition of silver ranges from only 1 to 3.9 percent. For BSA, which is composed of
57 percent bismuth, about 10 percent of acidification impacts are due to bismuth production.
3-78
-------
Table
3-49. Acidification impacts by life-cycle stage and process group (paste solder)
.,*..„,. c., m, c A/-< DC A CAD/-"
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
BSA
Score* %
SABC
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Bi production
Total
3.06E-01 4.71
8.77E-02 1.35
N/A N/A
N/A N/A
N/A N/A
3.94E-01 6.06
4.48E-01 3.59
N/A N/A
6.28E+00 50.3
8.07E-03 0.0647
N/A N/A
6.74E+00 54.0
2.30E-01 3.14
N/A N/A
1.88E+00 25.6
N/A N/A
7.45E-01 10.17
2.85E+00 38.9
4.52E-01 4.40
N/A N/A
4.04E+00 39.3
6.75E-03 0.0657
1.13E-02 0.110
4.51E+00 43.9
MANUFACTURING
Solder
manufacturing
Post-industrial
recycling
Total
2.56E-02 0.394
4.57E-02 0.704
7.13E-02 1.10
3.96E-02 0.317
3.63E-02 0.291
7.59E-02 0.608
2.48E-02 0.339
1.87E-02 0.255
4.35E-02 0.594
3.97E-02 0.387
3.62E-02 0.352
7.59E-02 0.739
USE/APPLICATION
Reflow
application
Total
6.03E+00 92.8
6.03E+00 92.8
5.66E+00 45.4
5.66E+00 45.4
4.43E+00 60.5
4.43E+00 60.5
5.68E+00 55.3
5.68E+00 55.3
END-OF-LIFE
Landfill
Incineration
Demanufacturing
Cu smelting
Unregulated
Total
GRAND
TOTAL
7.51E-05 0.0012
-7.70E-04 -0.0119
5.93E-04 0.0091
4.43E-03 0.0682
O.OOE+00 0.00
4.33E-03 0.0666
6.50E+00 100
6.50E-05 0.0005
-6.67E-04 -0.0053
5.13E-04 0.0041
3.84E-03 0.0307
O.OOE+00 0.00
3.75E-03 0.0300
1.25E+01 100
8.03E-05 0.0011
-8.24E-04 -0.0113
6.04E-04 0.0082
N/A N/A
O.OOE+00 0.00
-1.40E-04 -0.0019
7.32E+00 100
6.52E-05 0.0006
-6.69E-04 -0.0065
5.15E-04 0.0050
3.85E-03 0.0375
O.OOE+00 0.00
3.76E-03 0.0366
1.03E+01 100
*The impact scores are in units of kilograms SO2-equivalents/l,000 cubic centimeters of solder applied to a printed
wiring board.
N/A=not applicable
Within the manufacturing life-cycle stage, the post-industrial recycling process is a
greater contributor than solder manufacturing for all solder paste alloys except BSA. The
distribution of the manufacturing impacts between these two processes is similar to that found
for energy, and is discussed in Section 3.2.2. The manufacturing stage is a small contributor
overall. Likewise, EOL processes do not add significantly to acidification, contributing no more
than 0.07 percent of the total acidification impact score for any solder alloy. The majority of
EOL acidification impacts come from smelting processes used to recover copper and other
valuable metals from waste electronics (contributions range from 0.031 to 0.037 percent of
overall impacts for solders containing copper).
3-79
-------
Top Contributors to Acidification Impacts (Paste Solder)
Table 3-50 presents the specific materials or flows contributing a minimum of 1 percent
of acidification impacts by solder. As expected from the results above, all the top contributors
are from either the use/application stage or the upstream life-cycle stage. Only three materials
contribute greater than 1 percent: sulphur dioxide, nitrogen oxides, and hydrogen chloride
(hydrochloric acid). Sulphur dioxide is the largest contributor for all of the alloys, mostly from
electricity generation in the use/application stage and silver production (for alloys containing
silver). Nitrogen oxides are the second greatest contributor, mostly from electricity in the
use/application stage.
Table 3-50. Top contributors to acidification impacts (paste solder)
Solder
SnPb
SAC
BSA
SABC
Life-Cycle Stage
Use/application
Use/application
Upstream
Use/application
Upstream
Upstream
Upstream
Use/application
Use/application
Upstream
Upstream
Use/application
Upstream
Use/application
Upstream
Upstream
Upstream
Use/application
Use/application
Upstream
Use/application
Upstream
Upstream
Process
Electricity generation
Electricity generation
Tin production
Electricity generation
Tin production
Lead production
Silver production
Electricity generation
Electricity generation
Tin production
Tin production
Electricity generation
Silver production
Electricity generation
Bismuth production
Tin production
Tin production
Electricity generation
Electricity generation
Silver production
Electricity generation
Tin production
Tin production
Flow
Sulphur dioxide
Nitrogen oxides
Sulphur dioxide
Hydrogen chloride
Nitrogen oxides
Sulphur dioxide
Sulphur dioxide
Sulphur dioxide
Nitrogen oxides
Sulphur dioxide
Nitrogen oxides
Sulphur dioxide
Sulphur dioxide
Nitrogen oxides
Sulphur dioxide
Sulphur dioxide
Nitrogen oxides
Hydrogen chloride
Sulphur dioxide
Sulphur dioxide
Nitrogen oxides
Sulphur dioxide
Nitrogen oxides
% Contribution
65.4
24.4
3.10
1.64
1.62
1.27
49.5
32.0
11.9
2.36
1.23
42.7
25.2
15.9
9.91
2.06
1.08
1.07
39.0
38.7
14.5
2.89
1.51
For SnPb solder, sulphur dioxide and nitrogen oxides from electricity produced for the
use/application stage contribute approximately 66 and 25 percent to the total SnPb acidification
impacts, respectively. Other individual flows from the upstream processes for SnPb contribute
less than 3 percent each.
For the lead-free solders, the percent contribution of sulphur dioxide from both electricity
generation (for the use/application stage) and silver production combined ranges from 68 to 82
percent. Nitrogen oxides from electricity generation in the use/application stage are the second
greatest contributors for the lead-free alloys, accounting for about 12 to 16 percent of total
impacts. Flows of sulfur dioxide and nitrogen oxides from tin production contribute about 3
percent or less to acidification impacts for the different alloys, while flows from bismuth
3-80
-------
production contribute about 10 percent of BSA's acidification impacts. BSA has the highest
bismuth content of all the alloys at 57 percent. The extraction and processing inventories are
from secondary data sources that do not distinguish whether the acidification-inducing
substances are emitted during electricity generation or emitted directly during extraction and
processing itself.
3.2.7.3 Bar solder results
Total Acidification Impacts by Life-Cycle Stage (Bar Solder)
Table 3-51 presents the solder paste results for acidification impacts by life-cycle stage,
based on the impact assessment methodology presented in Sect 3.2.7.1. The table lists the
acidification impact scores per functional unit for the life-cycle stages of each solder paste alloy,
as well as the percent contribution of each life-cycle stage to the total impacts. Figure 3-18
presents the results in a stacked bar chart.
Table 3-51. Acidification impacts by life-cycle stage (bar solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
3.76E-01 26.3
9.22E-02 6.46
9.54E-01 66.9
4.86E-03 0.340
1.43E+00 100
SAC
Score* %
9.97E+00 90.8
4.39E-02 0.400
9.65E-01 8.79
4.25E-03 0.0387
1.10E+01 100
SnCu
Score* %
5.00E-01 32.7
5.95E-02 3.89
9.65E-01 63.2
4.22E-03 0.276
1.53E+00 100
*The impact scores are in units of kg SO2-equivalents/1,000 cc of solder applied to a printed wiring board.
3-81
-------
19
- 10
c lu
D
"(5
0 o
1
|
<2 R
c b
3 4
D- 4
t
o1
W 2
0) ^
0
SnPb
SAC
SnCu
n End-of-lif e
D Use/application
H Manufacturing
D Upstream
Figure 3-18. Bar Solder Total Life-Cycle Impacts: Acidification
As shown in the table and figure, SAC solder has the greatest impact category indicator
for acidification with 11 kg of SO2-equivalents/functional unit, followed by SnCu at 1.5 kg
SO2-equivalents and SnPb at 1.4 kg SO2-equivalents/functional unit. Nearly 91 percent of the
SAC life-cycle acidification impacts are driven by the upstream stage. The SnCu impacts are
only slightly higher (approximately 7 percent higher) than SnPb. The use/application stage
scores are approximately equal for each alloy; however, this stage contributes a greater percent
to the total SnPb and SnCu impacts due to the much lower impacts from the upstream stage.
Contributions from solder manufacturing (less than 7 percent of the total life cycle impacts) and
EOL processes (less than 0.4 percent) were small to minimal for all alloys.
Acidification Impacts by Process Group (Bar Solder)
Table 3-52 lists the acidification impacts of each of the processes in the life-cycle of the
bar solders. The production of energy consumed during wave solder application is the single
greatest contributor for the SnPb and SnCu alloys (67 and 63 percent, respectively). For SAC,
silver production in the upstream life-cycle stage is the largest contributor (85 percent) to all
SAC impacts, even though silver comprises only a small proportion of its composition. For
SnPb and SnCu, tin production is the second greatest contributor to total impacts (21 and 32
percent, respectively).
Within the manufacturing life-cycle stage, the post-industrial recycling process is a
greater contributor than solder manufacturing for SnPb and SnCu, while it is equal for SAC. The
distribution of the manufacturing impacts between these two processes depends mostly on the
different melting points of the alloys and varying secondary alloy content among the alloys,
3-82
-------
which are discussed in Chapter 2. The manufacturing stage is a small contributor overall.
Likewise, EOL processes do not add significantly to acidification, contributing no more
than 0.34 percent of the total acidification impact score for any solder alloy. The majority of
EOL acidification impacts come from smelting processes used to recover copper and other
valuable metals from waste electronics.
Table 3-52. Acidification impacts by life-cycle stage and process group (bar solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
SnCu
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Total
2.98E-01 20.9
7.83E-02 5.49
N/A N/A
N/A N/A
3.76E-01 26.3
6.30E-01 5.74
N/A N/A
9.32E+00 84.9
1.35E-02 0.123
9.97E+00 90.8
4.86E-01 31.8
N/A N/A
N/A N/A
1.32E-02 0.865
5.00E-01 32.7
MANUFACTURING
Solder manufacturing
Post-industrial recycling
Total
1.52E-02 1.07
7.70E-02 5.40
9.22E-02 6.46
2.20E-02 0.200
2.20E-02 0.200
4.39E-02 0.400
2.18E-02 1.43
3.76E-02 2.46
5.95E-02 3.89
USE/APPLICATION
Solder application
Total
9.54E-01 66.9
9.54E-01 66.9
9.65E-01 8.79
9.65E-01 8.79
9.65E-01 63.2
9.65E-01 63.2
END-OF-LIFE
Landfill
Incineration
Demanufacturing
Cu smelting
Unregulated
Total
GRAND TOTAL
8.34E-05 0.0058
-8.11E-04 -0.0568
6.58E-04 0.0461
4.93E-03 0.345
O.OOE+00 0.0000
4.86E-03 0.340
1.43E+00 100
7.30E-05 0.0007
-7.10E-04 -0.0065
5.76E-04 0.0052
4.31E-03 0.0393
O.OOE+00 0.0000
4.25E-03 0.0387
1.10E+01 100
7.25E-05 0.0047
-7.05E-04 -0.0461
5.72E-04 0.0374
4.28E-03 0.280
O.OOE+00 0.0000
4.22E-03 0.276
1.53E+00 100
*The impact scores are in units of kg SO2-equivalents/1,000 cc of solder applied to a printed wiring board.
N/A=not applicable
Top Contributors to Acidification Impacts (Bar Solder)
Table 3-53 presents the specific materials or flows contributing a minimum of 1 percent
of acidification impacts by solder. As expected from the results above, nearly all the top
contributors are from either the use/application stage or the upstream life-cycle stage. Outputs
from post-industrial recycling from the manufacturing stage also contribute greater than 1
percent to total impacts. Only these materials contribute greater than 1 percent: sulphur dioxide,
nitrogen oxides, and hydrogen chloride (hydrochloric acid). Sulphur dioxide is the largest
contributor for all of the alloys, mostly from electricity generation in the use/application stage or
silver production (for SAC). Nitrogen oxides are the second greatest contributor, mostly from
electricity in the use/application stage.
3-83
-------
Table 3-53. Top contributors to acidification impacts (bar solder)
Solder
SnPb
SAC
SnCu
Life-Cycle Stage
Use/application
Use/application
Upstream
Upstream
Manufacturing
Manufacturing
Use/application
Upstream
Use/application
Upstream
Use/application
Upstream
Upstream
Use/application
Upstream
Use/application
Upstream
Manufacturing
Use/application
Process
Electricity generation
Electricity generation
Tin production
Tin production
Electricity generation for post-industrial recycling
Electricity generation for post-industrial recycling
Electricity generation
Silver production
Electricity generation
Tin production
Electricity generation
Tin production
Silver production
Electricity generation
Tin production
Electricity generation
Tin production
Electricity generation for post-industrial recycling
Electricity generation
Flow
Sulphur dioxide
Nitrogen oxides
Sulphur dioxide
Nitrogen oxides
Sulphur dioxide
Nitrogen oxides
Hydrogen
chloride
Sulphur dioxide
Sulphur dioxide
Sulphur dioxide
Nitrogen oxides
Nitrogen oxides
Nitrogen oxides
Sulphur dioxide
Sulphur dioxide
Nitrogen oxides
Nitrogen oxides
Sulphur dioxide
Hydrogen
chloride
%
Contribution
47.2
17.6
13.7
7.16
3.57
1.33
1.18
83.5
6.20
3.77
2.31
1.97
1.34
44.5
20.9
16.6
10.9
1.57
1.12
For SnPb solder, sulphur dioxide and nitrogen oxides from electricity produced for the
use/application stage contribute approximately 47 and 18 percent to the total SnPb acidification
impacts, respectively. Other individual flows from the upstream and manufacturing processes
for SnPb contribute 7 percent or lower. The top contributors to SnCu are similar to SnPb.
For SAC, on the other hand, the percent contribution of sulphur dioxide from silver
production is the top contributor at approximately 84 percent. Sulphur dioxide and nitrogen
oxides from electricity generation (for the use/application stage) and from tin and silver
production also are in the top contributors list (6 percent and less). The ME&P inventories are
from secondary data sources that do not distinguish whether the acidification-inducing
substances are emitted during electricity generation or emitted directly during extraction and
processing itself.
3.2.7.4 Limitations and uncertainties
For the paste solder results, the two processes with the greatest contribution to
acidification impacts are electricity generation for the reflow application of solder (for all alloys)
and silver production (for the lead-free alloys). Similarly, for the wave solder results, wave
application (for SnPb and SnCu) and silver production (for SAC) are the top contributors to
acidification impacts. Acidification LCIA results are subject to the same sources of uncertainty
in the use/application stage inventory and silver production inventory as discussed previously.
For reflow solders, the greatest uncertainties are related to (1) reflow energy during
application/use is based on a limited number of data points that cover a wide range, (2)
3-84
-------
electricity production data employed in the use/application stage are from a secondary source,
and (3) the magnitude of many of the flows in the GaBi silver inventory used in this analysis
varies considerably from those in an alternate inventory available from DEAM. Energy
consumed during the reflow process is the subject of a sensitivity analysis in Section 3.3.
Section 3.3 also presents an alternate analysis using the DEAM silver inventory. The same
uncertainties associated with electricity production as a secondary source and the silver
inventory apply to the wave solder results. As discussed in previous sections, there is less
uncertainty associated with the wave application data than with the reflow application data.
Uncertainty in the acidification results also is derived from the impact assessment
methodology. Acidification impact characterization is a function of the mass of an acid-forming
chemical emitted to air and the AP equivalency factor for that chemical. The AP equivalency
factor is the number of hydrogen ions that can theoretically be formed per unit mass of the
pollutant being released compared to SO2. This is a full equivalency approach to impact
characterization where all substances are addressed in a unified, technical model that lends more
certainty to the characterization results than partial equivalency factors discussed with regard to
photochemical smog (Section 3.2.6).
3-85
-------
3.2.8 Air Particulate Impacts
3.2.8.1 Characterization
Air paniculate impacts refers to the release and build up of paniculate matter primarily
from combustion processes. Impact scores are based on the amount released to the air of
paniculate matter with average aerodynamic diameter less than 10 micrometers (PM10), the size
of paniculate matter that is most damaging to the respiratory system. Impact characterization is
simply based on the inventory amount of particulates released to air. This loading impact score
is calculated by:
ISPM = Amt
'PM •rL'"lPM
where:
ISPM equals the impact score for paniculate (kg PM10) per functional unit, and
AmtPM equals the inventory amount of paniculate release (PM10) to the air (kg) per
functional unit.
In this equation, PM10 is used to estimate impacts; however, if only TSP data are
available, these data are used. Using TSP data is an overestimation of PM10 which only refers to
the fraction of parti culates in the size range below 10 micrometers. A common conversion factor
(TSP to PM10) is not available because the fraction of PM10 varies depending on the type of
particulates. The paniculate matter impact category not only serves to represent potential health
effects associated with particulates (e.g., respiratory impacts), but also winter smog which
consists partially of suspended particulate matter or fine dust and soot particles. Winter smog is
distinguished from summer smog (e.g., photochemical smog, which is the build up of
tropospheric ozone concentrations due to VOCs and nitrogen oxides in the presence of sunlight).
Winter smog is a problem that occurs mainly in Eastern Europe and has been the cause of health-
related deaths in the past (Goedkoop, 1995).
3.2.8.2 Paste solder results
Total Air Particulate Impacts by Life-Cycle Stage (Paste Solder)
Table 3-54 presents the solder paste results for air particulate impacts by life-cycle stage,
based on the impact assessment methodology presented in above. The table lists the air
particulate impact scores per functional unit for the life-cycle stages of each solder paste alloy, as
well as the percent contribution of each life-cycle stage to the total impacts. Figure 3-19
presents the results in a stacked bar chart.
3-86
-------
Table 3-54. Air particulate impacts by life-cycle stage (paste solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
8.78E-02 19.4
6.28E-03 1.39
3.58E-01 79.1
3.08E-04 0.0682
4.52E-01 100
SAC
Score* %
9.57E-01 73.7
6.23E-03 0.480
3.36E-01 25.8
2.67E-04 0.0205
1.30E+00 100
BSA
Score* %
3.18E-01 54.3
4.15E-03 0.710
2.63E-01 45.0
3.76E-05 0.0064
5.85E-01 100
SABC
Score* %
6.62E-01 65.8
6.24E-03 0.620
3.37E-01 33.5
2.68E-04 0.027
1.01E+00 100
*The impact scores are in units of kilograms of particulate matter/1,000 cubic centimeters of solder applied to a
printed wiring board.
1 4
4-1
= 12
ns '^
c
o
*J "I
O '
c
s
•C 08-
n
—
SnPt
SAC
BSA
—
SABC
• End-of-life
n Use/application
H Manufacturing
B Upstream
Figure 3-19. Solder Paste Total Life-Cycle Impacts: Air Particulates
As shown in the table and figure, SAC solder has the greatest impact category indicator
for air particulates (1.30 kg particulate matter/functional unit), followed by SABC at 1.01 kg
particulate matter/functional unit. BSA and SnPb results are much lower with impact category
indicators of about 0.58 and 0.45 kg particulate matter/functional unit, respectively. For the
SnPb alloy, approximately 79 percent of the life-cycle air particulate impact score is driven by
the use/application stage, while 19 percent results from upstream processes. Unlike SnPb, the
lead-free alternatives receive greater contributions from the upstream stage than from the
use/application stage. Of the lead-free alternatives, SAC receives the greatest contribution from
upstream impacts at 74 percent, while BSA receives the lowest at 54 percent. The
use/application stage constitutes nearly all the remaining impacts for each lead-free alloy. Solder
manufacturing contributes less than 1.4 percent of the total air particulate impacts, while EOL
processes contribute 0.07 percent or less for any of the individual solder paste alloys.
3-87
-------
Air Particulate Impacts by Process Group (Paste Solder)
Table 3-55 lists the air particulate impact scores for each of the processes in the life-cycle
of the solder pastes. For SAC and SABC, silver production is the greatest contributor to total air
particulate impacts, while electricity generation in the use stage is the greatest contributor for the
SnPb and BSA alloys. As expected, given their greater silver content, impacts from silver
production are greater for SAC and SABC than for BSA. As with other impact categories,
however, the limited silver content of all the silver-bearing alloys results in disproportionately
high impacts from silver production compared to the other metals. For example, silver
production contributes 42 to 64 percent of the total air particulate impacts for the lead-free solder
alternatives, while the percent composition of silver in those alloys never exceeds 3.9 percent.
Table 3-55. Air particulate impacts by life-cycle stage and process group (paste solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
BSA
Score* %
SABC
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Bi production
Total
8.63E-02 19.1
1.49E-03 0.329
N/A N/A
N/A N/A
N/A N/A
8.78E-02 19.4
1.26E-01 9.72
N/A N/A
8.31E-01 63.9
3.93E-05 0.0030
N/A N/A
9.57E-01 73.7
6.47E-02 11.1
N/A N/A
2.48E-01 42.4
N/A N/A
4.85E-03 0.830
3.18E-01 54.3
1.27E-01 12.7
N/A N/A
5.35E-01 53.2
3.29E-05 0.0033
7.34E-05 0.0073
6.62E-01 65.8
MANUFACTURING
Solder
manufacturing
Post-industrial
recycling
Total
2.62E-03 0.580
3.66E-03 0.809
6.28E-03 1.39
3.32E-03 0.256
2.91E-03 0.224
6.23E-03 0.480
2.66E-03 0.454
1.50E-03 0.256
4.15E-03 0.710
3.34E-03 0.332
2.90E-03 0.288
6.24E-03 0.620
USE/APPLICATION
Reflow application
Total
3.58E-01 79.1
3.58E-01 79.1
3.36E-01 25.8
3.36E-01 25.8
2.63E-01 45.0
2.63E-01 45.0
3.37E-01 33.5
3.37E-01 33.5
END-OF-LIFE
Landfill
Incineration
Demanufacturing
Cu smelting
Unregulated
Total
GRAND TOTAL
6.52E-06 0.0014
-4.91E-06 -0.0011
3.52E-05 0.0078
2.72E-04 0.0601
O.OOE+00 0.00
3.08E-04 0.0682
4.52E-01 100
5.64E-06 0.0004
-4.25E-06 -0.0003
3.04E-05 0.0023
2.35E-04 0.0181
O.OOE+00 0.00
2.67E-04 0.0205
1.30E+00 100
6.97E-06 0.0012
-5.25E-06 -0.0009
3.58E-05 0.0061
N/A N/A
O.OOE+00 0.00
3.76E-05 0.0064
5.85E-01 100
5.67E-06 0.0006
-4.26E-06 -0.0004
3.06E-05 0.0030
2.36E-04 0.0235
O.OOE+00 0.00
2.68E-04 0.0266
1.01E+00 100
*The impact scores are in units of kilograms of particulate matter/1,000 cubic centimeters of solder applied to a
printed wiring board.
N/A = not applicable
-------
Tin, which has the greatest percent of the total composition in all the alloys except BS A,
contributes between 10 and 19 percent to impacts for all alloys. Although BSA has a higher
bismuth content (57 percent) than tin (42 percent), and the tin amount in BSA is less than the tin
in the other alloys (ranging from 63 to 95.5 percent), tin still contributes approximately 11
percent to the total impacts, while bismuth contributes less than 1 percent. This indicates that tin
has greater air paniculate emissions than bismuth per unit of metal produced.
Emissions from the production of energy consumed during the reflow of each of the
alloys contribute about 26 to 80 percent of the total air particulates score, depending on the alloy.
The percent contribution of the use stage to SnPb impacts is up to 54 percent higher than its
percent contribution to other alloys, even though the actual scores only differ by up to 26
percent. This is because SnPb upstream processes emit considerably less air particulates than
those of the silver-containing alloys.
The manufacturing stage is a small contributor overall. SnPb, SAC, and SABC have
nearly the same total manufacturing impact scores (approximately 0.006 kg particulate
matter/functional unit), all of which are greater than the impacts from BSA (0.004 kg particulate
matter/functional unit). Despite the similar total manufacturing impacts for SnPb, SAC, and
SABC, there are differing contributions from the solder manufacturing and the post-industrial
recycling processes. SnPb has more impacts from post-industrial recycling (0.0037 kg
particulate matter/functional unit) than SAC and SABC (both at approximately 0.0029 kg
particulate matter/functional unit). This is due to the fact that more secondary SnPb is used and
generated from the post-industrial recycling process. SAC and SABC have lower secondary
alloy content in the solder manufacturing process and, therefore, have lower post-industrial
recycling impacts. The higher impacts from post-industrial recycling for SnPb are counter-
balanced by the greater upstream impacts for the lead-free alternatives, which have greater virgin
content in the alloys.
EOL processes are even smaller contributors to air particulates, accounting for no more
than 0.07 percent of the total air particulates impact indicator for any solder alloy. The largest
contributions result from smelting processes that recover copper and other valuable metals from
waste electronics (percent contributions range from about 0.020 to 0.061 percent, for solders
containing copper). The demanufacturing process group that includes electricity generation is
the second greatest contributor to EOL impacts with between 0.0025 and 0.0079 percent
contribution to total air particulate impacts. Landfilling is a very small contributor to air
particulate impacts, less than 0.0015 percent for all alloys, and incineration results in a credit
based on the surplus energy generated during energy incineration.
Top Contributors to Air Particulate Impacts (Paste Solder)
Table 3-56 presents the specific materials or flows contributing greater than 1 percent to
air particulate impacts by solder. The only materials in the inventory that contribute to this
impact category are unspecified dust and PM10, and only unspecified dust contributes greater
thanl percent. As expected from the results above, all the top contributors are from either the
use/application stage or the upstream life-cycle stage.
3-89
-------
For SnPb, dust emitted from electricity produced for the use/application stage contributes
about 81 percent of total particulate impacts, and dust from tin production in the upstream stage
contributes about 18 percent. The two lead-free alternative solders with the higher silver
content, SAC and SABC, have the greatest impacts from dust emitted from the silver production
process, 62 and 51 percent, respectively. BSA has the lowest silver content of the lead-free
alternative solders. The life-cycle impacts of BSA are greatest from electricity generation from
solder reflow application (48 percent), followed by silver production (40 percent), and tin
production (11 percent). Tin production for all the alloys contributes between 9 and 18 percent.
The ME&P inventories are from secondary data sources that do not distinguish whether the
particulate matter is emitted from electric power used or directly released during extraction and
processing.
Table 3-56. Top contributors to air particulate impacts (paste solder)
Solder
SnPb
SAC
BSA
SABC
Life-Cycle Stage
Use/application
Upstream
Upstream
Use/application
Upstream
Use/application
Upstream
Upstream
Upstream
Use/application
Upstream
Process
Electricity production
Tin production
Silver production
Electricity production
Tin production
Electricity production
Silver production
Tin production
Silver production
Electricity production
Tin production
Flow
Dust (unspecified)
Dust (unspecified)
Dust (unspecified)
Dust (unspecified)
Dust (unspecified)
Dust (unspecified)
Dust (unspecified)
Dust (unspecified)
Dust (unspecified)
Dust (unspecified)
Dust (unspecified)
% Contribution
79.1
19.1
63.9
25.8
9.72
45.0
42.4
11.1
53.2
33.5
12.7
3.2.8.3 Bar solder results
Total Air Particulate Impacts by Life-Cycle Stage (Bar Solder)
Table 3-57 presents the bar solder results for air particulate impacts by life-cycle stage,
based on the impact assessment methodology presented above in Section 3.2.8.1. The table lists
the air particulate impact scores per functional unit for the life-cycle stages of each solder paste
alloy, as well as the percent contribution of each life-cycle stage to the total impacts. Figure 3-
20 presents the results in a stacked bar chart.
Table 3-57. Air particulate impacts by life-cycle stage (bar solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
8.52E-02 57.1
6.94E-03 4.66
5.66E-02 38.0
3.43E-04 0.230
1.49E-01 100
SAC
Score* %
1.41E+00 95.9
2.95E-03 0.201
5.73E-02 3.89
3.00E-04 0.0204
1.47E+00 100
SnCu
Score* %
1.37E-01 68.9
4.20E-03 2.11
5.73E-02 28.8
2.98E-04 0.150
1.99E-01 100
*The impact scores are in units of kilograms of particulate matter/1,000 cubic centimeters of solder applied to a
printed wiring board.
3-90
-------
1 6
1 A
*J l-t
'c
D
"(512
o
•a
o -i .
c '
a
* 0 8
ro
m D R
•s
•"04
&
•g 0 2
D)
n
SnPb
SAC
SnCu
._ . I- |-r
n Use/application
• Manufacturing
rj Upstream
Figure 3-20. Bar Solder Total Life-Cycle Impacts: Air Particulates
As shown in the table and figure, SAC solder has the greatest impact category indicator
for air particulates at 1.47 kg participate matter/functional unit, followed by SnCu and SnPb at
0.199 and 0.149 kg particulate matter/functional unit, respectively. For the SnPb alloy,
approximately 57 percent of the life-cycle air particulate impact score is driven by the upstream
stage, while 38 percent results from the use/application stage. SnCu has greater impacts than
SnPb from the upstream processes, which contribute approximately 69 percent to total SnCu
impacts. The use/application stage for SnCu contributes nearly 29 percent. As with SnPb and
SnCu, SAC receives its greatest contribution from the upstream stage, however, at a much higher
percentage (96 percent). The use/application stage constitutes nearly all the remaining impacts
for SAC. Solder manufacturing and EOL processes contribute small amounts to the overall air
particulate impacts.
Air Particulate Impacts by Process Group (Bar Solder)
Table 3-58 lists the air particulate impact scores for each of the processes in the life-cycle
of the bar solder. For SAC, silver production is the greatest contributor to total air particulate
impacts (84 percent), while tin production is the greatest contributor for the SnPb and SnCu
alloys (56 and 69 percent, respectively). Tin production might be expected to have a larger
impact as it is the largest proportion of the alloy by composition. Silver, on the other hand, is
only a small amount by composition in SAC (3.9 percent by weight); however, its production
dominates the air particulate impacts, while tin production is only 12 percent of total impacts.
This suggests that silver has much greater air particulate emissions than tin per unit of metal
produced.
3-91
-------
Table 3-58. Air particulate impacts by life-cycle stage and process group (bar solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
SnCu
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Total
8.39E-02 56.3
1.33E-03 0.890
N/A N/A
N/A N/A
8.52E-02 57.1
1.78E-01 12.1
N/A N/A
1.23E+00 83.8
6.57E-05 0.0045
1.41E-K)0 95.9
1.37E-01 68.9
N/A N/A
N/A N/A
6.44E-05 0.0324
1.37E-01 68.9
MANUFACTURING
Solder manufacturing
Post-industrial recycling
Total
7.79E-04 0.522
6.16E-03 4.13
6.94E-03 4.66
1.19E-03 0.0811
1.76E-03 0.1197
2.95E-03 0.201
1.18E-03 0.596
3.02E-03 1.52
4.20E-03 2.11
USE/APPLICATION
Solder application
Total
5.66E-02 38.0
5.66E-02 38.0
5.73E-02 3.89
5.73E-02 3.89
5.73E-02 28.8
5.73E-02 28.8
END-OF-LIFE
Landfill
Incineration
Demanufacturing
Cu smelting
Unregulated
Total
GRAND TOTAL
7.25E-06 0.0049
-5.17E-06 -0.0035
3.91E-05 0.0262
3.02E-04 0.203
O.OOE+00 0.0000
3.43E-04 0.230
1.49E-01 100
6.34E-06 0.0004
-4.52E-06 -0.0003
3.42E-05 0.0023
2.64E-04 0.0180
O.OOE+00 0.0000
3.00E-04 0.0204
1.47E+00 100
6.30E-06 0.0032
-4.49E-06 -0.0023
3.40E-05 0.0171
2.62E-04 0.132
O.OOE+00 0.0000
2.98E-04 0.150
1.99E-01 100
*The impact scores are in units of kilograms of particulate matter/1,000 cubic centimeters of solder applied to a
printed wiring board.
N/A=not applicable
Emissions from the production of energy consumed during wave solder application
contribute about 38 and 29 percent of the total air particulates score for SnPb and SnCu,
respectively. The wave application process group for SAC contributes much less on a
percentage basis (3.9 percent), although the absolute quantities for all three alloys are very
similar, ranging from 0.0566 to 0.0573 kg of particulate matter per functional unit.
The manufacturing stage is a small contributor overall, ranging from 0.20 to 4.7 percent.
All three alloys have more impacts from post-industrial recycling than from solder
manufacturing itself. This is due to the fact that more secondary SnPb, compared to secondary
SAC and SnCu, is used and generated from the post-industrial recycling process. As SAC and
SnCu have lower secondary alloy content in the solder manufacturing process, they have lower
post-industrial recycling impacts. The higher impacts from post-industrial recycling for SnPb
are counter-balanced by the greater upstream impacts for the lead-free alternatives, which have
greater virgin content in the alloys.
EOL processes are even smaller contributors to air particulates, accounting for no more
than 0.23 percent of the total air particulates impact indicator for any solder alloy. The largest
contributions result from smelting processes that recover copper and other valuable metals from
waste electronics (percent contributions range from 0.02 to 0.2 percent). The demanufacturing
process group that includes electricity generation is the second greatest contributor to EOL
3-92
-------
impacts, with between 0.0023 and 0.026 percent contribution to total air participate impacts.
Landfilling and incineration are very small contributors to air particulate impacts, and the lack of
particulate emissions from unregulated recycling and disposal result in no impacts associated
with unregulated recycling and disposal.
Top Contributors to Air Particulate Impacts (Bar Solder)
Table 3-59 presents the specific materials or flows contributing greater than 1 percent to
air particulate impacts by solder. The only materials in the inventory that contribute to this
impact category are unspecified dust and PM10. As expected from the results above, all the top
contributors are from the upstream and use/application stages, or to a lesser degree, from the
manufacturing life-cycle stage. Dust from tin production for each alloy is a top contributor.
For SnPb, dust emitted from tin production in the upstream stage contributes about 53
percent of total particulate impacts, and dust from electricity produced for the use/application
stage contributes about 18 percent. Dust from electricity generation from post-industrial
recycling, as well as the post-industrial recycling process itself, contributes less than 4 percent
combined. SAC is dominated by dust from silver production (84 percent), followed by tin
production (12 percent), and electricity generation during wave application (4 percent).
Dust as top contributor to SnCu is from tin production (69 percent), electricity generation
from wave application (29 percent), and electricity from post-industrial recycling (1 percent).
The ME&P inventories are from secondary data sources that do not distinguish whether the
particulate matter is emitted from electric power used or directly released during extraction and
processing.
Table 3-59. Top contributors to air particulate impacts (bar solder)
Solder
SnPb
SAC
SnCu
Life-Cycle Stage
Upstream
Use/application
Manufacturing
Manufacturing
Upstream
Upstream
Use/application
Upstream
Use/application
Manufacturing
Process
Tin production
Electricity generation
Electricity generation for
post-industrial recycling
Post-Industrial SnPb recycling
Silver production
Tin production
Electricity generation
Tin production
Electricity generation
Electricity generation for post-
industrial recycling
Flow
Dust (unspecified)
Dust (unspecified)
Dust (unspecified)
Particulate matter (PM-10)
Dust (unspecified)
Dust (unspecified)
Dust (unspecified)
Dust (unspecified)
Dust (unspecified)
Dust (unspecified)
% Contribution
56.3
38.0
2.87
1.09
83.8
12.1
3.89
68.9
28.8
1.02
3-93
-------
3.2.8.4 Limitations and uncertainties
For paste solders, the three processes with the greatest contribution to air particulate
impacts are electricity generation from solder reflow application and tin production (for all
alloys), and silver production (for the lead-free alloys). Similarly, for bar solders, the processes
with the greatest contribution are silver production, tin production, and wave application. For
the paste solders, sources of uncertainty in the use/application stage inventory have been
discussed previously (e.g., 3.2.1.4) and include the following: (1) reflow energy is based on a
limited number of data points that cover a wide range, and (2) electricity production data are
from a secondary source. Energy consumed during the reflow process is the subject of a
sensitivity analysis presented in Section 3.3. For bar solders, the uncertainty in the use stage is
related to the secondary data of the electricity production inventory, as described above;
however, the wave application data are expected to be a good representation of the process and
the same uncertainties described for reflow application of paste solders does not apply.
Uncertainties related to the silver inventory are described in Section 3.2.1.4 and are
related to the fact that two of the silver inventories available to the LFSP vary considerably in
the magnitude of flows from silver production. Section 3.2.1.4 concludes that although the GaBi
data set used in this analysis is considered "good' by GaBi, and was the preferred inventory for
this study, there remains enough uncertainty to perform an additional analysis using the alternate
inventory from the BEAM database. Results of the alternate analysis are presented in
Section 3.3.
The quality of tin production inventory data is deemed of average reliability and average
completeness from IDEMAT (Delft University of Technology), the original source of the data
supplied through Ecobilan (described in Section 2.2). The data used in the tin production
inventory are from data sources dated 1983 and 1989. As a consequence, the tin production
data, as used in the LFSP, are considered to be of moderate quality.
The impacts from air particulates are calculated as a direct measure of the inventory,
therefore, no direct additional uncertainty is introduced into the results from the characterization
calculations. The impact characterization is intended to be based on PM10 that is in the
respirable range and considered more damaging to the respiratory system than larger particles
when considering the effects of particulate matter on human health. Because most of the
inventory for this category is catalogued as unspecified dust, it is not known if these are PM10
particles. If the dust includes a broader class of particulate emissions, it is likely that the results
are somewhat overstated if they are to represent PM10 only.
3-94
-------
3.2.9 Water Eutrophication Impacts
3.2.9.1 Characterization
Eutrophication (nutrient enrichment) impacts to water are based on the identity and
concentrations of eutrophi cation chemicals released to surface water after treatment.
Equivalency factors for eutrophication have been developed assuming nitrogen (N) and
phosphorus (P) are the two major limiting nutrients. Therefore, the partial equivalencies are
based on the ratio of N to P in the average composition of algae (C106H263O110N16P) compared to
the reference compound phosphate (PO43") (Heijungs et a/., 1992; Lindfors et a/., 1995). If the
wastewater stream is first sent to a publicly-owned treatment works (POTW), treatment is
considered as a separate process, and the impact score would be based on releases from the
POTW to surface waters. Impact characterization is based on eutrophication potentials (EP)
(Appendix D) and the inventory amount:
(ISEUTR
where:
I^EUTR equals the impact score for regional water quality impacts from chemical / (kg
phosphate equivalents) per functional unit;
EFEP equals the EP equivalency factor for chemical /' (phosphate equivalents)
(Appendix D); and
AmtEC equals the inventory mass (kg) of chemical /' per functional unit of eutrophication
chemical in a wastewater stream released to surface water after any treatment, if
applicable.
3.2.9.2 Paste solder results
Total Water Eutrophication Impacts by Life-Cycle Stage (Paste Solder)
Table 3-60 presents the solder paste results for water eutrophication impacts by life-cycle
stage, based on the impact assessment methodology presented above. The table lists the water
eutrophication impact scores per functional unit for the life-cycle stages of each solder paste
alloy, as well as the percent contribution of each life-cycle stage to the total impacts. Figure 3-
21 presents the results in a stacked bar chart.
3-95
-------
Table 3-60. Water eutrophication impacts by life-cycle stage (
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
1.27E-04 0.104
1.60E-03 1.31
1.20E-01 98.5
9.72E-05 0.0800
1.22E-01 100
SAC
Score* %
3.70E-03 3.14
1.63E-03 1.39
1.12E-01 95.4
8.41E-05 0.0714
1.18E-01 100
BSA
Score* %
1.72E-03 1.89
9.32E-04 1.03
8.79E-02 97.1
1.22E-05 0.0134
9.06E-02 100
>aste solder)
SABC
Score* %
2.39E-03 2.04
1.63E-03 1.40
1.13E-01 96.5
8.45E-05 0.0722
1.17E-01 100
*The impact scores are in units of kilograms phosphate-equivalents/1,000 cubic centimeters of solder paste applied
to a printed wiring board.
014
4-1
'E
3 n 1 9
ro
c
° n 1
•a u- !
o
c
2 n OR
.>
'5-
2 n DR
5! u-uo
£
ro n n4
Q.
-------
Water Eutrophication Impacts by Process Group (Paste Solder)
Table 3-61 lists the water eutrophication impacts of each of the processes in the life-cycle
of the solder pastes. Releases associated with the generation of the energy required during
reflow assembly dominate the water eutrophi cation impact score for each of the solder alloys.
Compared to the use/application stage, the manufacturing stage is a small contributor
overall, with SnPb, SAC, and SABC having nearly the same total manufacturing impacts
(approximately 0.0016 kg phosphate-equivalents/functional unit). The impacts from BSA are
lower (0.0009 kg phosphate equivalents/functional units).
Despite the similar total manufacturing impacts for SnPb, SAC, and SABC, the
distribution of impacts between manufacturing processes differs. SnPb has more impact from
post-industrial recycling (0.00113 kg phosphate-equivalents/functional unit) than SAC and
SABC (0.000882 and 0.000880 kg phosphate-equivalents/functional unit, respectively). This is
due to the fact that more secondary SnPb is used and generated from the post-industrial recycling
process. SAC and SABC have lower secondary alloy content in the solder manufacturing, and
thus have lower post-industrial recycling impacts. The greater impacts from post-industrial
recycling for SnPb are counter-balanced by the greater upstream impacts for the lead-free
alternatives that have a larger virgin content in the alloys. See Section 3.2.2.2 for a more
complete discussion of this trade-off. Upstream and EOL processes also are both small
contributors to the eutrophi cation impacts. Upstream process impact scores are dominated by
the silver production process with the overall impacts ranging from approximately 1 to 3 percent
for the lead-free alternatives. By contrast, bismuth production for the BSA alloy contributes
about 0.7 percent to the total BSA life-cycle eutrophi cation impacts.
Table 3-61. Water eutrophication impacts by
life-cycle stage and process group (paste solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
BSA
Score* %
SABC
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Bi production
Total
6.06E-08 0.00005
1.27E-04 0.104
N/A N/A
N/A N/A
N/A N/A
1.27E-04 0.104
8.87E-08 0.0001
N/A N/A
3.69E-03 3.13
5.86E-06 0.0050
N/A N/A
3.70E-03 3.14
4.55E-08 0.00005
N/A N/A
1.10E-03 1.22
N/A N/A
6.14E-04 0.677
1.72E-03 1.89
8.96E-08 0.0001
N/A N/A
2.38E-03 2.03
4.91E-06 0.0042
9.28E-06 0.0079
2.39E-03 2.04
MANUFACTURING
Solder
manufacturing
Post-industrial
recycling
Total
4.63E-04 0.381
1.13E-03 0.932
1.60E-03 1.31
7.50E-04 0.636
8.82E-04 0.749
1.63E-03 1.39
4.69E-04 0.518
4.63E-04 0.511
9.32E-04 1.03
7.53E-04 0.644
8.80E-04 0.752
1.63E-03 1.40
USE/APPLICATION
Reflow application
Total
1.20E-01 98.5
1.20E-01 98.5
1.12E-01 95.4
1.12E-01 95.4
8.79E-02 97.1
8.79E-02 97.1
1.13E-01 96.5
1.13E-01 96.5
3-97
-------
Table 3-61. Water eutrophication impacts by
life-cycle stage and process group (paste solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
BSA
Score* %
SABC
Score* %
END-OF-LIFE
Landfill
Incineration
Demanufacture
Cu smelting
Unregulated
Total
GRAND TOTAL
6.41E-07 0.0005
-4.74E-07 -0.0004
1.18E-05 0.0097
8.53E-05 0.0702
O.OOE+00 0.00
9.72E-05 0.0800
1.22E-01 100
5.55E-07 0.0005
-4.10E-07 -0.0003
1.02E-05 0.0086
7.38E-05 0.0626
O.OOE+00 0.00
8.41E-05 0.0714
1.18E-01 100
6.86E-07 0.0008
-5.07E-07 -0.0006
1.20E-05 0.0132
N/A N/A
O.OOE+00 0.00
1.22E-05 0.0134
9.06E-02 100
5.57E-07 0.0005
-4.12E-07 -0.0004
1.02E-05 0.0087
7.41E-05 0.0633
O.OOE+00 0.00
8.45E-05 0.0722
1.17E-01 100
*The impact scores are in units of kilograms phosphate-equivalents/1,000 cubic centimeters of solder paste applied
to a printed wiring board.
N/A=not applicable
Top Contributors to Eutrophication Impacts (Paste Solder)
Table 3-62 presents the specific materials or flows contributing at least 1 percent of
eutrophication impact scores by solder. The only material that meets this criterion is chemical
oxygen demand (COD) in flows from electricity generation processes and from silver
production (for the silver-containing alloys). Other flows in the LFSP inventory that contribute
to the eutrophication impacts include ammonia/ammonium, phosphate, and nitrate, each
contributing less than 1 percent of the overall impacts for a specific solder. As expected from
the results above, COD from the use/application stage is the top contributor to total
eutrophication impacts, ranging from 94 to 97 percent of total impacts depending on the solder.
Flows of COD from silver production contribute from about 1 to 3 percent. The silver extraction
and processing inventory is from a secondary data source that does not distinguish whether the
eutrophication-causing substances are released from the generation of electric power used or are
directly released during extraction and processing.
Table 3-62. Top contributors to water eutrophication impacts (paste solder)
Solder
SnPb
SAC
BSA
SABC
Life-Cycle Stage
Use/application
Use/application
Upstream
Use/application
Upstream
Use/application
Upstream
Process
Electricity generation
Electricity generation
Silver production
Electricity generation
Silver production
Electricity generation
Silver production
Flow
COD
COD
COD
COD
COD
COD
COD
% Contribution
97.1
94.1
2.93
95.7
1.14
95.1
1.90
3-98
-------
3.2.9.3 Bar solder results
Total Water Eutrophication Impacts by Life-Cycle Stage (Bar Solder)
Table 3-63 presents the bar solder results for water eutrophication impacts by life-cycle
stage, based on the impact assessment methodology presented above. The table lists the water
eutrophication impact scores per functional unit for the life-cycle stages of each bar solder alloy,
as well as the percent contribution of each life-cycle stage to the total impacts. Figure 3-22
presents the results in a stacked bar chart.
Table 3-63. Water eutrophication impacts by life-cycle stage (bar solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
1.13E-04 0.529
2.22E-03 10.4
1.89E-02 88.6
1.08E-04 0.505
2.14E-02 100
SAC
Score* %
5.49E-03 21.3
9.75E-04 3.79
1.92E-02 74.5
9.45E-05 0.368
2.57E-02 100
SnCu
Score* %
9.70E-06 0.047
1.35E-03 6.56
1.92E-02 92.9
9.39E-05 0.455
2.06E-02 100
*The impact scores are in units of kilograms phosphate-equivalents/1,000 cubic centimeters of bar solder applied
to a printed wiring board.
0 03
+J
E n no^
15
o
^5 n no
a
2
Sn ni^
CO
>
?
n
SnPb
SAC
SnCu
D End-of-life
n Use/application
• Manufacturing
n Upstream
Figure 3-22. Bar Solder Total Life-Cycle Impacts: Water Eutrophication
3-99
-------
As shown in the table and figure, SAC has the greatest impact indicator for water
eutrophication at 0.0257 kg phosphate-equivalents/functional unit, followed closely by the SnPb
and SnCu bar solders at 0.0214 and 0.0206 kg phosphate-equivalents/functional unit,
respectively. The use/application life-cycle stage is by far the dominant contributing life-cycle
stage, accounting for at least 75 percent of the total water eutrophi cation impacts of each of the
solder alloys and ranging as high as 93 percent for the SnCu alloy. Impacts from upstream
processes are significant for the SAC alloy, accounting for nearly 23 percent of the overall
impacts, but are not a factor for the non-silver alloys contributing less than one percent of their
overall impact scores. The manufacturing life-cycle stage impacts range from roughly 4 percent
for SAC up to a high of 10 percent for SnPB. EOL processes contribute relatively little to total
impacts, accounting for 0.505 percent or less of the total water eutrophication impacts for each
solder type.
Water Eutrophication Impacts by Process Group (Bar Solder)
Table 3-64 lists the water eutrophication impacts of each of the processes in the life-cycle
of the bar solder alloys. Releases associated with the generation of the energy required during
wave assembly dominate the water eutrophication impact score for each of the solder alloys.
As mentioned previously, SAC had the highest eutrophication impact score, nearly 20
percent higher than both the SnPb and SnCu solders. The difference is due mostly to the impacts
associated with the mining and extraction of the silver content in the SAC alloy, which
comprises only 3.9 percent of the alloy. Impacts from silver mining are a minimum of 3 orders
of magnitude higher than the impacts associated with the mining of the other metals, including
tin, which makes up 95.5 percent of the solder alloy.
As seen with the paste solders, impacts associated with the use/application stage once
again dominate the overall water eutrophication impacts, ranging from 89 to 93 percent of the
overall impacts for the non-silver alloys. These impacts result from the generation of energy
required for the wave application of solder to PWBs during the assembly process. Despite
having nearly identical impact scores (0.0189- 0.0192 kg phosphate equivalent per 1,000 cubic
centimeters of solder) for all of the alloys, impacts from wave soldering account for only 75
percent of the eutrophication impacts for the SAC alloy, again due to the additional impacts from
the mining and extraction of silver.
For the non-silver containing alloys of SnPb and SnCu, the manufacturing life-cycle
stage processes make up the majority of the remainder of the impacts. Post-industrial recycling
of the solder makes the only other significant contribution to eutrophication impacts, ranging
from 4.4 to 9 percent. Solder manufacturing accounts for no more than 2.1 percent of the overall
eutrophication impacts, while the other remaining life-cycle processes make minimal overall
contributions to eutrophication impacts.
3-100
-------
Table 3-64. Water eutrophication impacts by
life-cycle stage and process group (bar solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
SnCu
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Total
5.89E-08 0.0003
1.13E-04 0.529
N/A N/A
N/A N/A
1.13E-04 0.529
1.25E-07 0.0005
N/A N/A
5.48E-03 21.3
9.79E-06 0.0381
5.49E-03 21.3
9.63E-08 0.0005
N/A N/A
N/A N/A
9.61E-06 0.0466
9.70E-06 0.0
MANUFACTURING
Solder manufacturing
Post-industrial recycling
Total
3.09E-04 1.44
1.91E-03 8.92
2.22E-03 10.4
4.41E-04 1.71
5.34E-04 2.08
9.75E-04 3.79
4.38E-04 2.12
9.15E-04 4.44
1.35E-03 6.56
USE/APPLICATION
Wave application
Total
1.89E-02 88.6
1.89E-02 88.6
1.92E-02 74.5
1.92E-02 74.5
1.92E-02 92.9
1.92E-02 92.9
END-OF-LIFE
Landfill
Incineration
Demanufacture
Cu smelting
Unregulated
Total
GRAND TOTAL
7.12E-07 0.0033
-4.99E-07 -0.0023
1.31E-05 0.0611
9.47E-05 0.443
O.OOE+00 0.00
1.08E-04 0.51
2.14E-02 100
6.23E-07 0.0024
-4.37E-07 -0.0017
1.14E-05 0.0445
8.29E-05 0.322
O.OOE+00 0.00
9.45E-05 0.368
2.57E-02 100
6.19E-07 0.0030
-4.34E-07 -0.0021
1.14E-05 0.0551
8.23E-05 0.399
O.OOE+00 0.00
9.39E-05 0.46
2.06E-02 100
*The impact scores are in units of kilograms phosphate-equivalents/1,000 cubic centimeters of bar solder applied
to a printed wiring board.
N/A=not applicable
Top Contributors to Eutrophication Impacts (Bar Solder)
Table 3-65 presents the specific materials or flows contributing at least 1 percent of
eutrophication impact scores by bar solder alloy. Ammonia and COD are the only materials in
the life-cycle inventory that meet this criterion.
COD releases during the generation of electricity used within the life-cycle of bar solders
are the top contributors to water eutrophication. Electricity generation for the use/application of
solder during the wave assembly process results in the largest COD loading, contributing from
74 to 92 percent of the water eutrophication impact score. The generation of electricity for other
uses, such as post-industrial recycling and manufacturing of the solder alloy also contribute to
the overall COD releases (between 2.8 and 7.7 percent to the total impacts).
Flows of COD from silver production contribute from nearly 20 percent for the SAC
alloy; however, the silver extraction and processing inventory is from a secondary data source
that does not distinguish whether the eutrophication-causing substances are released from the
generation of electric power used or directly released during extraction and processing.
3-101
-------
Other flows in the LFSP inventory that contribute to the eutrophication impacts include
ammonia/ammonium, phosphate, and nitrate, each contributing less than one percent of the
overall impacts for any solder. Ammonia released during the post-industrial recycling of the
SnPb and SnCu alloys accounts for a small percentage of the overall eutrophi cation scores for
each alloy.
Table 3-65. Top contributors to water eutrophication impacts (bar solder)
Solder
SnPb
SAC
SnCu
Life-Cycle Stage
Use/application
Manufacturing
Manufacturing
Manufacturing
Use/application
Upstream
Manufacturing
Manufacturing
Use/application
Manufacturing
Manufacturing
Manufacturing
Process
Electricity generation
Electricity generation for
post-industrial recycling
Post-industrial SnPb
recycling
SnPb bar solder
manufacturing
Electricity generation
Silver production
Electricity generation for
post-industrial recycling
Electricity generation for
solder manufacturing
Electricity generation
Electricity generation for
post-industrial recycling
Electricity generation for
solder manufacturing
Post-industrial SnCu
recycling
Flow
COD
COD
Ammonia
COD
COD
COD
COD
COD
COD
COD
COD
Ammonia
% Contribution
87.4
6.61
2.06
1.04
73.5
19.9
1.51
1.30
91.6
3.24
1.61
1.01
3.2.9.4 Limitations and uncertainties
The major contributors to energy impacts are from electricity generation used during the
use/application stage (particularly for paste solders) and from upstream materials extraction
processes (particularly for SAC bar solder). Similar to the discussion in Section 3.2.1, where
electricity generation for reflow application is concerned, the same uncertainties apply: (1) the
number of data points used to estimate reflow electricity consumption are limited and cover a
large range, and (2) electricity production data are from a secondary source. With regard to the
first source of uncertainty, the amount of electricity consumed during reflow was measured
during reflow testing conducted by the LFSP. These are primary data collected under controlled
conditions to meet the goals and objectives of this study and represent good high and low
estimates of wave electricity consumption; however, because the value used in this baseline
analysis is averaged from a limited amount of data (two data points for each solder), a sensitivity
analysis was performed using the high and low values (see Section 3.3). On the other hand,
uncertainties from the use of secondary data for electricity generation are not considered large
enough to warrant any further analysis.
For wave application results, primary data also were collected for the solder application
3-102
-------
process through a controlled testing protocol. Although data from only one test run were used,
these data were compared to other known testing data and are expected to be representative of
typical wave operations, thus introducing little uncertainty. The use of the secondary data for the
electricity generation data was discussed above.
Uncertainty in the eutrophication results also is derived from the impact assessment
methodology. Eutrophication impacts are calculated from the mass of a chemical released
directly to surface water and the chemical's EP. The EP is a partial equivalency factor derived
from the ratio of nitrogen and phosphorus in the average composition of algae compared to the
reference compound phosphate. As a partial equivalency approach, only a subset of substances
can be converted into equivalency factors, which is a limitation of this LCIA methodology. The
methodology, however, does take into account nitrogen and phosphorus, which are two major
limiting nutrients of importance to eutrophication.
3-103
-------
3.2.10 Water Quality Impacts
3.2.10.1 Characterization
Water quality impacts are characterized as surface water impacts due to releases of
wastes causing oxygen depletion and increased turbidity. Two water quality impact scores are
calculated based on the BOD and TSS in the wastewater streams released to surface water. The
impact scores are based on releases to surface water following any treatment. Using a loading
characterization approach, impact characterization is based on the amount of BOD and TSS in a
wastewater stream. The water quality score equations for each are presented below:
(ISBon>, = (AmtBOD),
and
(ISTSS), = (AmtTSS),
where:
ISBOD equals the impact score for BOD water quality impacts for waste stream / (kg) per
functional unit;
AmtBOD equals the inventory amount of BOD in wastewater stream / released to surface
waters (kg) per functional unit;
IS TSS equals the impact score for TSS water quality impacts for waste stream /' (kg) per
functional unit; and
AmtTSS equals the inventory amount of TSS in wastewater stream /' released to surface
waters (kg) per functional unit.
3.2.10.2 Paste solder results
Total Water Quality Impacts by Life-Cycle Stage (Paste Solder)
Table 3-66 presents the solder paste results for water quality impacts by life-cycle stage,
based on the impact assessment methodology presented in above. This impact category
characterized the impacts on water quality based on the mass loading of BOD and total solids
released to surface water. The table lists the water quality impact scores per functional unit for
the life-cycle stages of each solder paste alloy, as well as the percent contribution of each life-
cycle stage to the total impacts. Figure 3-23 presents the results in a stacked bar chart.
3-104
-------
Table 3-66. Water quality impacts by life-cycle stage (paste solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
2.10E-03 1.17
6.58E-03 3.67
1.70E-01 94.7
8.15E-04 0.455
1.79E-01 100
SAC
Score* %
5.82E-02 25.8
7.70E-03 3.41
1.59E-01 70.5
7.05E-04 0.312
2.26E-01 100
BSA
Score* %
3.59E-02 21.9
3.17E-03 1.94
1.25E-01 76.0
1.64E-04 0.100
1.64E-01 100
SABC
Score* %
3.78E-02 18.3
7.69E-03 3.73
1.60E-01 77.6
7.08E-04 0.343
2.06E-01 100
*The impact scores are in units of kilograms BOD & solids/1,000 cc of solder paste applied to a printed wiring
board.
0 9E>
*;
c
3 09-
03
C
o
2 n 1=;
.2
(0
;o
"7^ n i
in
ofl
Q
O n QE;
CQ u.uo
U)
J£
n
^=
SnPb
—
SAC
BSA
c
••I
5ABC
-\
• FnH nf lifp
n Use/application
I Manufacturing
• Upstream
Figure 3-23. Solder Paste Total Life-Cycle Impacts: Water Quality (BOD & Solids)
As shown in the table and figure, SAC solder paste has the greatest impact indicator for
water quality (0.226 kg BOD & solids/functional unit); followed by SABC at 0.206 kg BOD &
solids/functional unit; SnPb is next with 0.179 kg; and BSA follows with 0.164 BOD &
solids/functional unit. Water quality impacts are driven in large part by contributions from the
use/application stage, which range from 71 to 95 percent, depending on the solder alloy. While
nearly all of the water quality impacts for SnPb result from use/application stage, upstream
processes contribute substantially to the water quality, with impacts ranging from 18 to 26
percent. SAC has the greatest upstream impacts at 0.0582 kg, followed by SABC and BSA with
0.378 and 0.359 kg BOD & solids/functional unit each.
Solder manufacturing impacts for the solders contribute between about 1.9 and 3.7
percent of the total life cycle impacts. SAC and SABC have the highest impacts from
manufacturing (both at about 0.0077 kg BOD & solids/functional unit), followed closely by
3-105
-------
SnPb (0.00658 kg/functional unit). BSA has the least amount of manufacturing impacts
(0.00317 kg/functional unit). EOL processes contribute less than 0.5 percent to total impacts for
each alloy.
Water Quality Impacts by Process Group (Paste Solder)
Table 3-67 lists the water quality impacts of each of the processes in the life-cycle of the
solders. The production of the energy consumed during the reflow assembly of the solders is the
single greatest contributor to the water quality impact score. For the lead-free alloys, upstream
processes also are significant. Within the upstream stage, silver production for SAC and SABC
contribute 26 and 18 percent respectively. As with other impact categories, impacts from silver
production are large and disproportionate to the silver content of the alloys (ranging from 1 to
3.9 percent), demonstrating that water quality is affected more from silver by mass than from
other metals. BSA water quality impacts are more evenly distributed between bismuth (11.3
percent) and silver (10.6 percent) production processes, despite bismuth comprising a much
greater percentage of the solder alloy than silver (57 percent bismuth to 1 percent silver).
The manufacturing stage is a relatively small contributor to the overall water quality
impact scores for the solder alloys. Within the manufacturing stage, the post-industrial recycling
process is a greater contributor than solder manufacturing. Post-industrial recycling contributes
between 1.4 and 3.1 percent, while the solder manufacturing process group contributes 0.7
percent or less for each of the alloys. The distribution of the manufacturing impacts between
these two processes is similar to that found in other impact categories discussed earlier.
Likewise, EOL processes do not add substantially to water quality impacts, contributing
no more than 0.5 percent of the total water quality impact score. The majority of the impacts
come from smelting processes used to recover copper and other valuable metals from waste
electronics, contributions range from 0.253 percent to 0.370 percent, except for BSA which does
not include copper smelting. There are no BOD or solids emissions assumed in the unregulated
recycling and disposal process, and no associated impacts in this impact category.
3-106
-------
Table 3-67. Water quality impacts by life-cycle stage and process group (paste solder)
Life-cycle stage
Process group
SnPb 1 SAC
Score* % | Score* %
BSA
Score* %
SABC
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Bi production
Total
8.84E-07 0.0005
2.10E-03 1.17
N/A N/A
N/A N/A
N/A N/A
2.10E-03 1.17
1.29E-06 0.0006
N/A N/A
5.80E-02 25.7
2.03E-04 0.0898
N/A N/A
5.82E-02 25.8
6.63E-07 0.0004
N/A N/A
1.73E-02 10.6
N/A N/A
1.86E-02 11.3
3.59E-02 21.9
1.31E-06 0.0006
N/A N/A
3.74E-02 18.1
1.70E-04 0.0823
2.81E-04 0.136
3.78E-02 18.3
MANUFACTURING
Solder
manufacturing
Post-industrial
recycling
Total
1.03E-03 0.577
5.55E-03 3.10
6.58E-03 3.67
1.39E-03 0.616
6.31E-03 2.79
7.70E-03 3.41
9.05E-04 0.552
2.27E-03 1.38
3.17E-03 1.936
1.40E-03 0.678
6.29E-03 3.05
7.69E-03 3.73
USE/APPLICATION
Reflow
application
Total
1.70E-01 94.7
1.70E-01 94.7
1.59E-01 70.5
1.59E-01 70.5
1.25E-01 76.0
1.25E-01 76.0
1.60E-01 77.6
1.60E-01 77.6
END-OF-LIFE
Landfill
Incineration
Demanufacturing
Cu smelting
Unregulated
Total
GRAND TOTAL
1.40E-04 0.0780
-1.92E-06 -0.0011
1.67E-05 0.0093
6.61E-04 0.369
O.OOE+00 0.00
8.15E-04 0.455
1.79E-01 100
1.21E-04 0.0535
-1.66E-06 -0.0007
1.44E-05 0.0064
5.72E-04 0.253
O.OOE+00 0.00
7.05E-04 0.312
2.26E-01 100
1.49E-04 0.0911
-2.05E-06 -0.0013
1.70E-05 0.0104
N/A N/A
O.OOE+00 0.00
1.64E-04 0.100
1.64E-01 100
1.21E-04 0.0589
-1.67E-06 -0.0008
1.45E-05 0.0070
5.74E-04 0.278
O.OOE+00 0.00
7.08E-04 0.343
2.06E-01 100
*The impact scores are in units of kilograms BOD & solids/1,000 cc of solder applied to a printed wiring board.
N/A=not applicable
Top Contributors to Water Quality Impacts (Paste Solder)
Table 3-68 presents the specific materials or flows contributing greater than 1 percent of
water quality impacts by solder. As expected from the results above, the majority of the top
contributors are from the upstream and the use/application stages, with the manufacturing stage
also making a contribution. By definition, this section characterizes the water quality based on
BOD and total solids, therefore, the flows presented in Table 3-68 are limited to BOD,
suspended solids, and dissolved solids. Suspended solids are the majority of water quality
impacts for all of the solders, accounting for 89 to 92 percent of the total impact scores, with the
largest individual contributions resulting from electricity generation during the use/application
stage. Other suspended solids flows include those from the upstream metal production processes
as well as heavy fuel oil production. BOD and dissolved solids from electricity production for
the use/application stage combine to account for 6 to 8 percent of the water quality impact
3-107
-------
scores, depending on the solder alloy. Inventories from the extraction and processing of metals,
as well as from fuel production, are from secondary data sources that do not distinguish whether
the emissions are from electric power used or directly released during extraction, processing, or
production.
Table 3-68. Top contributors to water quality impacts (paste solder)
Solder
SnPb
SAC
BSA
SABC
Life-Cycle Stage
Use/application
Use/application
Use/application
Manufacturing
Upstream
Use/application
Upstream
Use/application
Use/application
Use/application
Upstream
Upstream
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
Manufacturing
Process
Electricity generation
Electricity generation
Electricity generation
Heavy fuel oil (#6) production
for post-industrial recycling
Lead production
Electricity generation
Silver production
Electricity generation
Electricity generation
Electricity generation
Bismuth production
Silver production
Electricity generation
Electricity generation
Electricity generation
Silver production
Electricity generation
Electricity generation
Heavy fuel oil (#6) production
for post-industrial recycling
Flow
Solids (suspended)
BOD
Solids (dissolved)
Solids (suspended)
Solids (suspended)
Solids (suspended)
Solids (suspended)
BOD
Solids (dissolved)
Solids (suspended)
Solids (suspended)
Solids (suspended)
BOD
Solids (dissolved)
Solids (suspended)
Solids (suspended)
BOD
Solids (dissolved)
Solids (suspended)
% Contribution
86.9
4.19
3.63
1.42
1.13
64.7
24.9
3.12
2.70
69.8
11.1
10.3
3.37
2.92
71.2
17.6
3.44
2.98
1.89
3.2.10.3 Bar solder results
Total Water Quality Impacts by Life-Cycle Stage (Bar Solder)
Table 3-69 presents the solder paste results for water quality impacts by life-cycle stage,
based on the impact assessment methodology presented above. This impact category
characterized the impacts on water quality based on the mass loading of BOD and total solids
released to surface water. The table lists the water quality impact scores per functional unit for
the life-cycle stages of each solder paste alloy, as well as the percent contribution of each life-
cycle stage to the total impacts. Figure 3-24 presents the results in a stacked bar chart.
3-108
-------
Table 3-69. Water quality impacts by life-cycle stage (bar solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
1.88E-03 4.72
1.01E-02 25.5
2.69E-02 67.5
9.06E-04 2.28
3.98E-02 100
SAC
Score* %
8.65E-02 72.2
5.37E-03 4.48
2.72E-02 22.7
7.93E-04 0.662
1.20E-01 100
SnCu
Score* %
3.34E-04 0.917
8.09E-03 22.2
2.72E-02 74.7
7.87E-04 2.16
3.64E-02 100
*The impact scores are in units of kilograms BOD & solids/1,000 cc of solder paste applied to a printed wiring
board.
0 14
0 12 -
+J
'c
= 01 -
ro al
o
t5 0 08
i
•D 0 06
"o
Q 0 04 -
O
m
^ 0 02
n
SnPb
SAC
SnCu
n End-of-life
n Use/application
• Manufacturing
n Upstream
Figure 3-24. Bar Solder Total Life-Cycle Impacts: Water Quality (BOD & Solids)
As shown in the table and figure, SAC solder paste has the greatest impact indicator for
water quality (0.226 kg BOD & solids/functional unit); followed by SABC at 0.206 kg BOD &
solids/functional unit; SnPb is next with 0.179 kg; and BSA follows with 0.164 BOD &
solids/functional unit. Water quality impacts are driven in large part by the contributions from
the use/application stage, which range from 71 to 95 percent, depending on the solder alloy.
While nearly all of the water quality impacts for SnPb result from use/application stage,
upstream processes contribute substantially to the water quality, with impacts ranging from 18 to
26 percent. SAC has the greatest upstream impacts at 0.0582 kg, followed by SABC and BSA
with 0.378 and 0.359 kg BOD & solids/functional unit each.
Solder manufacturing impacts for the solders contribute between about 1.9 and 3.7
percent of the total life cycle impacts. SAC and SABC have the highest impacts from
manufacturing (both at about 0.0077 kg BOD & solids/functional unit), followed closely by
SnPb (0.00658 kg/functional unit). BSA has the least amount of manufacturing impacts
3-109
-------
(0.00317 kg/functional unit). EOL processes contribute less than 0.5 percent to total impacts for
each alloy.
Water Quality Impacts by Process Group (Bar Solder)
Table 3-70 lists the water quality impacts of each of the processes in the life-cycle of the
solders. The production of the energy consumed during the reflow assembly of the solders is the
single greatest contributor to the water quality impact score. For the lead-free alloys, upstream
processes also are significant. Within the upstream stage, silver production for SAC and SABC
contribute 26 and 18 percent, respectively. As with other impact categories, impacts from silver
production are large and disproportionate to the silver content of the alloys (ranging from 1 to
3.9 percent), demonstrating that water quality is affected more from silver by mass than from
other metals. BSA water quality impacts are more evenly distributed between bismuth (11.3
percent) and silver (10.6 percent) production processes, despite bismuth comprising a much
greater percentage of the solder alloy than silver (57 percent bismuth to 1 percent silver).
Table 3-70. Water quality impacts by life-cycle stage
and process group (bar solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
SnCu
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Total
8.59E-07 0.0022
1.88E-03 4.72
N/A N/A
N/A N/A
1.88E-03 4.72
1.82E-06 0.0015
N/A N/A
8.61E-02 71.9
3.39E-04 0.283
8.65E-02 72.2
1.40E-06 0.0039
N/A N/A
N/A N/A
3.32E-04 0.914
3.34E-04 0.9
MANUFACTURING
Solder manufacturing
Post-industrial recycling
Total
7.84E-04 1.97
9.34E-03 23.5
1.01E-02 25.5
1.55E-03 1.29
3.82E-03 3.19
5.37E-03 4.48
1.54E-03 4.22
6.55E-03 18.0
8.09E-03 22.2
USE/APPLICATION
Wave application
Total
2.69E-02 67.5
2.69E-02 67.5
2.72E-02 22.7
2.72E-02 22.7
2.72E-02 74.7
2.72E-02 74.7
END-OF-LIFE
Landfill
Incineration
Demanufacturing
Cu smelting
Unregulated
Total
GRAND TOTAL
1.55E-04 0.390
-2.02E-06 -0.0051
1.85E-05 0.047
7.34E-04 1.85
O.OOE+00 0.00
9.06E-04 2.28
3.98E-02 100
1.36E-04 0.1134
-1.77E-06 -0.0015
1.62E-05 0.0135
6.42E-04 0.536
O.OOE+00 0.00
7.93E-04 0.662
1.20E-01 100
1.35E-04 0.371
-1.75E-06 -0.0048
1.61E-05 0.044
6.38E-04 1.75
O.OOE+00 0.00
7.87E-04 2.16
3.64E-02 100
*The impact scores are in units of kilograms BOD & solids/1,000 cc of solder applied to a printed wiring board.
N/A=not applicable
3-110
-------
The manufacturing stage is a relatively small contributor to the overall water quality
impact scores for the solder alloys. Within the manufacturing stage, the post-industrial recycling
process is a greater contributor than solder manufacturing. Post-industrial recycling contributes
between 1.4 and 3.1 percent, while the solder manufacturing process group contributes 0.7
percent or less for each of the alloys. The distribution of the manufacturing impacts between
these two processes is similar to that found in other impact categories discussed earlier.
Likewise, EOL processes do not add substantially to water quality impacts, contributing
no more than 0.5 percent of the total water quality impact score. The majority of the impacts
come from smelting processes used to recover copper and other valuable metals from waste
electronics (contributions range from 0.253 percent to 0.370 percent, except for BSA which does
not include copper smelting). There are no BOD or solids emissions assumed in the unregulated
recycling and disposal process, and no associated impacts in this impact category.
Top Contributors to Water Quality Impacts (Bar Solder)
Table 3-71 presents the specific materials or flows contributing greater than 1 percent of
water quality impacts by solder. As expected from the results above, the majority of the top
contributors are from the upstream and the use/application stages, with the manufacturing stage
also making a contribution. By definition, this section characterizes the water quality based on
BOD and total solids, therefore, the flows presented in Table 3-71 are limited to BOD,
suspended solids, and dissolved solids. Suspended solids constitute the majority of water quality
impacts for all of the solders, accounting for 89 to 92 percent of the total impact scores, with the
largest individual contributions resulting from electricity generation during the use/application
stage. Other suspended solids flows include those from the upstream metal production processes
as well as heavy fuel oil production. BOD and dissolved solids from electricity production for
the use/application stage combine to account for 6 to 8 percent of the water quality impact
scores, depending on the solder alloy. Inventories from the extraction and processing of metals,
as well as from fuel production are from secondary data sources that do not distinguish whether
the emissions are from electric power used or directly released during extraction, processing, or
production.
3-111
-------
Table 3-71. Top contributors to water quality impacts (bar solder)
Solder
SnPb
SAC
SnCu
Life-Cycle Stage
Use/application
Manufacturing
Upstream
Manufacturing
Use/application
Use/application
Manufacturing
End-of-life
Upstream
Use/application
Manufacturing
Upstream
Use/application
Use/application
Manufacturing
Use/application
Use/application
Manufacturing
Manufacturing
Manufacturing
End-of-life
Manufacturing
Manufacturing
Process
Electricity generation
Electricity generation for
post-industrial recycling
Lead production
Post-Industrial SnPb recycling
Electricity generation
Electricity generation
Post-Industrial SnPb recycling
Heavy fuel oil #6 production for Cu
smelting
Silver production
Electricity generation
Heavy fuel oil #6 post-industrial
recycling
Silver production
Electricity generation
Electricity generation
Heavy fuel oil #6 post-industrial
recycling
Electricity generation
Electricity generation
Electricity generation for
post-industrial recycling
LPG production for solder
manufacturing
Post-industrial SnCu recycling
Heavy fuel oil #6 production for Cu
smelting
Post-industrial SnCu recycling
Electricity generation for solder
manufacturing
Flow
Solids (suspended)
Solids (suspended)
Solids (suspended)
Solids (suspended)
BOD
Solids (dissolved)
BOD
Solids (suspended)
Solids (suspended)
Solids (suspended)
Solids (suspended)
BOD
BOD
Solids (suspended)
Solids (suspended)
BOD
Solids (dissolved)
Solids (suspended)
Solids (suspended)
Solids (suspended)
Solids (suspended)
BOD
Solids (suspended)
% Contribution
62.0
4.69
4.53
3.23
2.99
2.59
2.46
1.37
69.8
20.8
1.98
1.18
1.00
68.5
11.2
3.31
2.86
2.42
2.04
1.67
1.30
1.27
1.21
3.2.10.4 Limitations and uncertainties
The processes that contribute the greatest to the water quality impacts are electricity
generation for the reflow application of solder, as well as the upstream metal production
processes for the lead-free alloys. Sources of uncertainty in the use/application stage inventory
were discussed in Section 3.2.2.1 and include the following: (1) reflow energy is based on a
limited number of data points that cover a wide range, and (2) electricity production data are
from a secondary source. Energy consumed during the reflow process is the subject of a
sensitivity analysis presented in Section 3.3, but uncertainties in the electricity generation
inventory were not considered significant. For a more detailed discussion, see Section 3.2.2.1.
Uncertainties related to the silver inventory are described in Section 3.2.2 and have to do
with the fact that two alternate silver inventories available to the LFSP vary considerably in the
magnitude of flows from silver production. Section 3.2.2 concludes that although the GaBi data
3-112
-------
set used in this analysis is considered "good" by GaBi, there remains enough uncertainty to
perform an additional analysis using the alternate inventory from the DEAM database. Results
of the alternate analysis are presented in Section 3.3.
Tin production inventory data quality is deemed of average reliability and average
completeness from IDEMAT (Delft University of Technology), the original source of the data
supplied through Ecobilan (described in Section 2.2). The data used in the tin production
inventory are from data sources dated 1983 and 1989. As a consequence, the tin production
data, as used in the LFSP, are considered to be of moderate quality.
Uncertainty in the water quality results is derived from the impact assessment
methodology. Water quality impacts are calculated using a loading approach based on the mass
of BOD and total solids released directly to surface water; therefore, these results are sensitive to
the quality of the inventory data, which are discussed above.
3-113
-------
3.2.11 Occupational Human Health Impacts
This section presents the LCIA characterization methodology and the LCIA results for
the occupational human health impact category; however, some of the discussions relate to all of
the toxicity impact categories in general (e.g., occupational human health, public human health,
and ecotoxicity). The occupational human health impact results presented in this section include
two impact categories: occupational non-cancer impacts and occupational cancer impacts. The
results for these categories are provided within each of the subsections below.
3.2.11.1 Characterization
Potential Human Health Impacts
Human health impacts are defined in the context of life-cycle assessment as relative
measures of potential adverse health effects to humans. Human health impact categories
included in the scope of this LFSP LCA are chronic (repeated dose) effects, including non-
carcinogenic and carcinogenic effects. Chronic human health effects to both workers and the
public are considered. This section presents the potential occupational health impacts, and
Section 3.2.12 presents the potential public health impacts. It was assumed that there is no direct
consumer contact with the solder on PWBs, therefore, quantitative measures of consumer
impacts are not included in the LCIA methodology.
The chemical characteristic that classifies inventory items to the human health effects
(and ecotoxicity) categories is toxicity. Toxic chemicals were identified by searching lists of
toxic chemicals (e.g., Toxic Release Inventory [TRI]) and, if needed, toxicity databases (e.g.,
Hazardous Substances Data Bank [HSDB]), and Registry of Toxic Effects of Chemical
Substances (RTECS), and other literature (see Appendix E). The review was done by the DfE
Workgroup for the DfE Computer Display Project (Socolof etal., 2001), and remains applicable
to the LFSP. Several materials in the LFSP inventory were excluded from the toxic list if they
were generally accepted as non-toxic. The EPA DfE Workgroup also reviewed the list of
chemicals that were included in this project as potentially toxic. The list of potentially toxic
chemicals is provided in Appendix E, and chemicals that were excluded from the toxic list that
appear in the LFSP inventory also are presented in Appendix E.
Human (and ecological) toxicity impact scores are calculated based on a chemical
scoring method modified from the CHEMS-1 that is found in Swanson et al. (1997). To
calculate impact scores, chemical-specific inventory data are required. Any chemical that is
assumed to be potentially toxic is given a toxicity impact score. This involves collecting toxicity
data (described in Appendix E). If toxicity data are unavailable for a chemical, a mean default
toxicity score is given. This is described in detail below. Ecological toxicity is presented in
Section 3.2.13.
Chronic human health effects are potential human health effects occurring from repeated
exposure to toxic agents over a relatively long period of time (i.e., years). These effects could
include carcinogenicity, reproductive toxicity, developmental effects, neurotoxicity,
immunotoxicity, behavioral effects, sensitization, radiation effects, and chronic effects to other
3-114
-------
specific organs or body systems (e.g., blood, cardiovascular, respiratory, kidney and liver
effects). Impact categories for chronic health effects are divided into cancer and non-cancer
effects for both worker and public impacts. Occupational impact scores are based on inventory
inputs; public impact scores are based on inventory outputs.
This section addresses chronic occupational health effects, which refer to potential health
effects to workers, including cancer, from long-term repeated exposure to toxic or carcinogenic
agents in an occupational setting. For possible occupational impacts, the identity and amounts of
materials/constituents as input to a process are used. The inputs represent potential exposures. It
could be assumed that a worker would continue to work at a facility and incur exposures over
time, however, the inventory is based on manufacturing one unit volume of solder as applied to a
particular PWB design and does not truly represent chronic exposure; therefore, the chronic
health effects impact score is more of a ranking of the potential of a chemical to cause chronic
effects than a prediction of actual effects.
Chronic occupational health effects scores are based on the identity of toxic chemicals
(or chemical ingredients) found in inputs from all of the life-cycle stages. The distinction
between pure chemicals and mixtures is made, if possible, by specifying component ingredients
of mixtures in the inventory.
The chronic human health impact scores are calculated using hazard values (HVs) for
carcinogenic and non-carcinogenic effects. Calculation of the occupational non-cancer and
cancer HVs are described below, and the public non-cancer and cancer HV calculations are
described in Section 3.2.12.1. Appendix H provides example calculations of toxicity impacts for
two sample chemicals.
Occupational Human Health Characterization: Non-Cancer
The non-carcinogen HV is based on either no-observed-adverse-effect levels (NOAELs)
or lowest-observed-adverse-effect levels (LOAELs). The non-carcinogen HV is the greater of
the oral and inhalation HV:
\KirihalNOAEL)
inhalation: (HVNr ),. =
V jVL-,.M|.,,f _,;„/ *
\/(inhalNOAELmeJ
V(oralNOAEL,)
oral: (HV
\l(oralNOAELJ
3-115
-------
where:
HVNC oml equals the non-carcinogen oral hazard value for chemical / (unitless);
oralNOAEL t equals the oral NOAEL for chemical / (mg/kg-day);
oral NOAEL mean equals the geometric mean oral NOAEL of all available oral NOAELs
(Appendix E) [12.6 mg/kg-day];
HVNC ^^1^^ equals the non-carcinogen inhalation hazard value for chemical /
(unitless);
inhal NOAEL t equals the inhalation NOAEL for chemical /' (mg/m3); and
inhal NOAEL mean equals the geometric mean inhalation NOAEL of all available inhalation
NOAELs (Appendix E) [68.7 mg/m3].
The oral and inhalation NOAEL mean values are the geometric means of a set of
chemical data presented in Appendix E. If LOAEL data are available, instead of NOAEL data,
the LOAEL, divided by 10, is used to substitute for the NOAEL. The most sensitive endpoint is
used if there are multiple data for one chemical.
The non-carcinogen HVs for a particular chemical are multiplied by the applicable
inventory input to calculate the impact score for non-cancer effects:
(IScHo-Nc), = (HVNCxAmtTCmpJ1
where:
IScHo-Nc equals the impact score for chronic occupational non-cancer health effects for
chemical /' (kg noncancer-toxequivalent) per functional unit;
HVNC equals the hazard value for chronic non-cancer effects for chemical /'; and
AmtTCinput equals the amount of toxic inventory input (kg) per functional unit for chemical /'.
Occupational Human Health Characterization: Cancer
The cancer HV uses cancer slope factors or cancer weight of evidence (WOE)
classifications assigned by EPA or the International Agency for Research on Cancer (IARC). If
both an oral and inhalation slope factor exist, the slope factor representing the larger hazard is
chosen; thus, given that there is a cancer slope factor (SF) for a chemical, the cancer HV for
chronic occupational health effects is the greater of the following:
oralSF,
oral: (HVCA\ =
oralSFmean
inhalation SF;
inhalation: (HVC. ), = -
I-4" tfiltfilfitinvi ' *
inhalation SFmean
3-116
-------
where:
oral SFt
oralSFmean
T-fV
•" ' CA inhalation
inhalation SFi
inhalation SF'
equals the cancer oral hazard value for chemical /' (unitless);
equals the cancer oral slope factor for chemical /' (mg/kg-day)"1;
equals the geometric mean cancer slope factor of all available slope
factors (Appendix E) [0.71 (mg/kg-day)"1];
equals the cancer inhalation hazard value for chemical /' (unitless);
equals the cancer inhalation slope factor for chemical /' (mg/kg-day)"1; and
equals the geometric mean cancer inhalation slope factor of all available
inhalation slope factors (Appendix E) [1.70 (mg/kg-day)"1].
The oral and inhalation slope factor mean values are the geometric means of a set of
chemical data presented in Appendix E.
Where no slope factor is available for a chemical, but there is a WOE classification, the
WOE is used to designate default hazard values as follows: EPA WOE Groups D (not
classifiable) and E (non-carcinogen) and IARC Groups 3 (not classifiable) and 4 (probably not
carcinogenic) are given a hazard value of zero. All other WOE classifications (known, probable,
and possible human carcinogen) are given a default HV of 1 (representative of a mean slope
factor) (Table 3-72). Similarly, materials for which no cancer data exist, but are designated as
potentially toxic, are also given a default value of 1.
Table 3-72. Hazard values for carcinogenicity WOE if no slope factor is available
EPA
classification
Group A
Group Bl
Group B2
Group C
Group D
Group E
IARC
classification
Group 1
Group 2A
N/A
Group 2B
Group 3
Group 4
Description
Known human carcinogen
Probable human carcinogen (limited human data)
Probable human carcinogen (from animal data)
Possible human carcinogen
Not classifiable
Non-carcinogenic or probably not carcinogenic
Hazard
value
1
1
1
1
0
0
N/A=not applicable
The cancer HV for a particular chemical, whether it is from a slope factor or WOE, is
then multiplied by the applicable inventory amount to calculate the impact score for cancer
effects:
3-117
-------
where:
IS,
'CHO-CA
HV
nv (
Amt
CA
TC input
equals the impact score for chronic occupational cancer health effects for
chemical /' ( kg cancertox-equivalents) per functional unit;
equals the hazard value for carcinogenicity for chemical /'; and
equals the amount of toxic inventory input (kg) per functional unit for
chemical /'.
3.2.11.2 Paste solder results
Total Occupational Impacts by Life-Cycle Stage (Paste Solder)
Table 3-73 presents the paste solder results for occupational non-cancer impacts by life-
cycle stage, based on the impact assessment methodology presented above. The table below lists
the occupational non-cancer impact scores per functional unit for the life-cycle stages of each
solder paste alloy, as well as the percent contribution of each life-cycle stage to the total impacts.
Figure 3-25 shows the results in a stacked bar chart.
Table 3-73. Occupational non-cancer impacts by life-cycle stage (paste solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
6.03E+00 0.0011
2.03E+05 36.2
1.75E+05 31.2
1.82E+05 32.6
5.60E+05 100
SAC
Score* %
9.59E+00 0.118
2.84E+03 35.0
2.59E+03 31.9
2.67E+03 32.9
8.12E+03 100
BSA
Score* %
5.24E+00 0.224
7.31E+02 31.3
7.95E+02 34.0
8.05E+02 34.4
2.34E+03 100
SABC
Score* %
9.29E+00 0.177
1.83E+03 34.9
1.69E+03 32.1
1.72E+03 32.8
5.25E+03 100
*The impact scores are in units of kilograms noncancertox-equivalents/1,000 cc of solder applied to a printed
wiring board.
3-118
-------
600,000
SnPb
SAC
BSA
SABC
Figure 3-25. Solder Paste Total Life-Cycle Impacts: Occupational Non-Cancer
Occupational impact scores are based on the potential toxicity of material inputs to each
process. This characterization method does not necessarily indicate where actual exposure is
occurring; instead, it uses the inputs of potentially toxic materials as surrogates for exposure.
While this methodology introduces some uncertainties into the occupational health impact
results, discussed further below, it is an improvement over former LCIA methodologies that do
not evaluate occupational health impacts.
As shown in the figure, the occupational non-cancer impact score for SnPb (560,000 kg
noncancertox-equivalents/functional unit) is far greater than the scores for other solder alloys
(ranging from 2,340 to 8,120 kg noncancertox-equivalents/functional unit). Because SnPb has a
higher toxicity compared to the other alloys, its impacts are larger. Note that the HVs of the
solders are assumed to be the weighted averages of the HVs of the individual metals and fluxes
(when applicable) that make up the alloys.
Three life-cycle stages largely contribute to total impacts, regardless of the solder type:
manufacturing, use/application, and EOL. The EOL stage (34.4 percent) was the largest
contributor for BSA, slightly exceeding the contributions of the use/application stage (34.0
percent) and manufacturing stage (31.3 percent). For the remaining alloys—SnPb, SAC, and
SABC—the solder manufacturing stage accounts for the largest portion of the total occupational
non-cancer impacts score, with values ranging from 35 to 36 percent; however, both the EOL
and use/application stages also make substantial contributions to the impact score, accounting
for a minimum of 31 percent of the overall scores each. For each of the paste solder alloys, the
upstream life-cycle stages did not contribute significantly, accounting for less than 0.3 percent of
the occupational non-cancer life-cycle impacts.
3-119
-------
To help put the scores for occupational non-cancer impacts in perspective, the
occupational non-cancer toxicity score associated with using enough electricity to power a
60-watt bulb for one year is 20,677 kg noncancertox-equivalents. The difference between the
SnPb and SAC results presented above (i.e., 552,000 kg noncancertox-equivalents) is equivalent
to the toxicity impacts associated with continuously running a 60-watt bulb for approximately 27
years. The differences among the lead-free alloys are much smaller; SAC as compared to BSA
is equivalent to running a 60-watt bulb for 143 days, which represents a greater difference than
many of the other impact categories when compared to electricity used to power a lightbulb.
Most of the other impact categories have relative differences on the order of operating a
lightbulb for hours to days. These results could indicate either that there are fewer toxic
materials used in electricity generation than are used in the solder life-cycle or that the quantities
of toxic materials are much greater in the solder life-cycles than for electricity to power a
lightbulb.
Table 3-74 presents the solder paste results for occupational human health cancer
impacts by life-cycle stage, based on the impact assessment methodology presented above. The
table lists the occupational cancer impact scores per functional unit for the life-cycle stages of
each solder paste, as well as the percent contribution of each life-cycle stage to the total impacts.
Figure 3-26 presents the results in a stacked bar chart.
Table 3-74. Occupational cancer impacts by life-cycle stage (paste solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
6.03E+00 7.90
2.07E+01 27.2
4.14E+01 54.3
8.11E+00 10.6
7.62E+01 100
SAC
Score* %
9.43E+00 13.1
1.79E+01 24.8
3.80E+01 52.8
6.71E+00 9.31
7.20E+01 100
BSA
Score* %
5.18E+00 8.17
1.75E+01 27.6
3.27E+01 51.6
7.98E+00 12.6
6.34E+01 100
SABC
Score* %
9.18E+00 12.7
1.80E+01 24.9
3.83E+01 52.9
6.84E+00 9.45
7.23E401 100
*The impact scores are in units of kilograms cancertox-equivalents/1,000 cc of solder paste applied to a printed
wiring board.
3-120
-------
10
SnPb
SAC
BSA
SABC
Figure 3-26. Solder Paste Total Life-Cycle Impacts: Occupational Cancer
As shown in the preceding table and figure, SnPb has the greatest occupational cancer
impact score (76.2 kg cancertox-equivalents/functional unit), but its score is not much higher
than those for SABC and SAC (72.3 and 72.0 kg cancertox-equivalents/functional unit,
respectively). In fact, the results for these three alloys may be indistinguishable given the
uncertainties in the data. BSA has the lowest total impact score at 63.4 kg cancertox-
equivalents/functional unit.
Unlike several other impact categories previously described, the occupational cancer
impacts are not completely dominated by one, or even two, life-cycle stages. For all the solders,
the use/application stage is the greatest contributor to total occupational cancer impacts, ranging
from 52 to 54 percent; however, the manufacturing stage, as well as the EOL and upstream
stages, contribute to a large extent. Potential impacts from the manufacturing stage range from
25 to 28 percent, while EOL stage impacts range from 9 to 13 percent depending on the alloy.
The contributions of upstream life-cycle stages range from 8 to 13 percent.
In comparison to the occupational non-cancer impacts in which SnPb has substantially
greater impacts than the other solders, the total cancer impacts are much closer in magnitude to
one another. This is primarily due to a lack of carcinogenicity data for the solder metals, and
may not be an accurate reflection of the potential occupational cancer impacts of the different
alloys. For example, lead is the only solder metal that has been classified as a probable human
carcinogen (EPA and IARC carcinogenic WOE classifications of B2 and 2A, respectively);
however, since no slope factor is available for lead, it receives the same HV (HV=1,
representative of an average HV) as tin and bismuth, two solder metals that have not been
classified as to carcinogencity. (Average hazard values are assigned to materials that have not
been classified to minimize the bias that typically favors materials with little or no toxicity data.)
Identical mass inputs of these metals will receive identical occupational cancer scores, even
3-121
-------
though their relative carcinogenicity is not known. A lack of carcinogenicity data is one of the
major limitations and uncertainties in the occupational cancer characterization method, and is
discussed further below.
Occupational Impacts by Process Group (Paste Solder)
Table 3-75 lists the occupational non-cancer impacts of each of the process groups in the
life-cycle of the solders. As noted above, the manufacturing, use/application, and EOL stages all
largely contribute to occupational non-cancer impacts for all of the paste solder alloys. The
manufacturing stage is made up of two process groups: solder manufacturing and post-industrial
recycling, both of which include the fuel production of any associated fuels used during
operation. The impacts from solder manufacturing are greater than post-industrial recycling,
accounting for 31 to 36 percent of total impacts for all alloys, compared to less than 0.2 percent
for post-industrial recycling. This is because the major contributors to the manufacturing
impacts are the metals inputs used in production of the alloys (discussed below under the "Top
Contributors" section), and the non-cancer hazard values of some of those metals (e.g., lead and
silver) are very high. On the other hand, the inputs to the post-industrial recycling processes
(e.g., dross inputs, which are outputs from the solder manufacturing process) do not have
associated toxicity data to develop a hazard value, so the default hazard value is used, which is
far below that of lead and silver. Solder manufacturing is the greatest contributor to
occupational non-cancer impacts because it has the greatest quantity of solder inputs, and
because occupational impacts are based on the quantity and potential toxicity of those inputs.
The reflow application process group within the use/application stage is comprised of the
solder reflow process and associated electricity generation. Use/application impacts for
occupational non-cancer, therefore, are from the inputs to the reflow process itself, as well as
inputs to the electricity generation process.
Landfilling is the greatest contributor to EOL occupational non-cancer impacts (24 to 25
percent of total impacts) for all of the alloys, followed by incineration (6 to 17 percent of total
impacts). Demanufacturing, copper smelting, and unregulated recycling/disposal each contribute
approximately 1 percent to the total occupational non-cancer impacts for SnPb, SAC, and SABC.
These processes make equal contributions to the impacts of each solder alloy since they were
assumed to receive equal amounts of waste electronics and, therefore solder, at EOL. Copper
smelting is not included in the BSA EOL model.
Like the solder manufacturing process group discussed above, landfilling and
incineration dominate occupational non-cancer health impacts at EOL because these dispositions
have the greatest inputs of EOL solder, the toxicity and overall quantity of which contribute to
the determination of the overall impact score. Furthermore, the LCIA methodology uses input
quantities as surrogates for exposure in lieu of incorporating an exposure model as would be
done in a chemical risk assessment. For example, within an alloy life-cycle, at this time most
electronics are destined for landfilling (at least 72 percent) as modeled in the LFSP and, as a
result, the LCIA methodology assumes most occupational exposure to solders occur during
landfilling. As a result, the landfilling impacts dominate EOL within each alloy life-cycle. This
occurs despite the fact that there may actually be less true occupational exposure to a landfill
3-122
-------
worker than to a demanufacuturer or copper smelter worker. Given the screening nature of the
LCIA occupational impact category method, the process with the greatest quantities of
potentially toxic materials would tend to have the greatest impacts for a given set of similar
materials. For this reason, the scores for demanufacturing and unregulated recycling/disposal are
identical because the LFSP model assumes that equal amounts of EOL solder go to both those
dispositions. No mass is assumed to be lost between demanufacturing inputs and copper
smelting inputs. The occupational non-cancer impacts from demanufacturing and copper
smelting, therefore, are the same because they have the same mass of solder inputs.
Table 3-75. Occupational non-cancer impacts by life-cycle stage and
process group (paste solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
BSA
Score* %
SABC
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Bi production
Total
5.81E+00 0.0010
2.25E-01 0.00004
N/A N/A
N/A N/A
N/A N/A
6.03E+00 0.0011
8.50E+00 0.105
N/A N/A
1.09E+00 0.0134
3.80E-03 0.00005
N/A N/A
9.59E+00 0.118
4.35E+00 0.186
N/A N/A
3.25E-01 0.0139
N/A N/A
5.62E-01 0.0240
5.24E+00 0.224
8.58E+00 0.163
N/A N/A
7.01E-01 0.0133
3.18E-03 0.0001
8.50E-03 0.0002
9.29E+00 0.177
MANUFACTURING
Solder
manufacturing
Post-industrial
recycling
Total
2.03E+05 36.2
1.07E+01 0.002
2.03E+05 36.2
2.83E+03 34.9
8.79E+00 0.108
2.84E+03 35.0
7.27E+02 31.1
4.38E+00 0.187
7.31E+02 31.3
1.83E+03 34.8
8.77E+00 0.167
1.83E+03 34.9
USE/APPLICATION
Reflow
application
Total
1.75E+05 31.2
1.75E+05 31.2
2.59E+03 31.9
2.59E+03 31.9
7.95E+02 34.0
7.95E+02 34.0
1.69E+03 32.1
1.69E+03 32.1
END-OF-LIFE
Landfill
Incineration
Demanufacturing
Cu smelting
Unregulated
Total
GRAND
TOTAL
1.26E+05 22.4
3.32E+04 5.92
7.86E+03 1.40
7.86E+03 1.40
7.86E+03 1.40
1.82E+05 32.6
5.60E+05 100
1.84E+03 22.7
4.86E+02 5.99
1.15E+02 1.42
1.15E+02 1.42
1.15E+02 1.42
2.67E+03 32.9
8.12E+03 100
5.82E+02 24.9
1.54E+02 6.57
3.47E+01 1.48
N/A N/A
3.47E+01 1.48
8.05E+02 34.4
2.34E+03 100
1.19E+03 22.6
3.13E+02 5.96
7.42E+01 1.41
7.43E+01 1.42
7.42E+01 1.41
1.72E+03 32.8
5.25E+03 100
*The impact scores are in units of kilograms noncancertox-equivalents/1,000 cc of solder paste applied to a printed
wiring board.
N/A=not applicable
Differences in impacts beyond differences in the inventory do arise when evaluating the
solder paste alloys against one another. For example, SnPb has the greatest impacts versus the
other alloys because the toxicity of lead is greater than the toxicity of the materials in the other
3-123
-------
alloys. This is discussed in the subsection below.
Upstream occupational non-cancer impacts arise from the inputs to the extraction and
processing of the various metals present in the alloys. These impacts are small compared to the
total life-cycle impacts. When evaluating the upstream impacts alone, tin production is the
greatest contributor to the upstream impacts for all alloys, but is still a small percentage of total
life-cycle impacts (e.g., from 0.001 to 0.19 percent). For SAC and SABC, silver production is
the second greatest upstream contributor (0.013 percent). For BSA, bismuth production is the
second greatest contributor at 0.024 percent, followed by silver at 0.014 percent.
Table 3-76 lists the occupational cancer impacts of each of the processes in the life-cycle
of the solders. The use/application stage is the greatest contributor to occupational cancer
impacts for the solders. The reflow solder process is the only process group within this stage,
and the only two inputs modeled in the reflow process are solder paste and electricity. Cancer
impacts from the use/application stage, therefore, are based on the carcinogenic potential of the
solder paste and any potentially carcinogenic inputs to the electricity generation process. The
impacts from the use/application stage alone follow the same trend as the total impacts. That is,
SnPb has the greatest occupational cancer impact score (41.4 kg cancertox-equivalents/
functional unit), followed closely by SABC (38.3 kg cancertox-equivalents/functional unit),
which is only slightly above SAC (38.0 kg cancertox-equivalents/functional unit). BSA has the
lowest impacts from the use/application stage at 32.7 kg cancertox-equivalents/functional unit.
BSA impacts are expected to be somewhat lower since less electricity is used for reflowing BSA
than for the other alloys, primarily due to BSA's lower melting temperature.
Within the manufacturing stage, which is the second greatest contributor to occupational
impacts, the solder manufacturing process group impacts are greater than the post-industrial
process group impacts for all the solders. The solder manufacturing process group accounts for
19 to 25 percent and post-industrial recycling accounts for 3 to 6 percent of total impacts for all
alloys.
Within the EOL stage, the landfilling process group is the greatest contributor (about 6 to
9 percent of total impacts), followed by incineration (about 1.7 to 2.4 percent of total impacts).
Demanufacturing, copper smelting, and unregulated recycling/disposal are smaller contributors
to the total occupational cancer impacts for all alloys (about 0.7 percent or less each). Similar to
the occupational non-cancer impacts discussed above, landfilling and incineration dominate
impacts for this category because, instead of an exposure model, the impacts are based on the
quantity of inputs to each process that have the potential to be toxic (carcinogenic, in this case).
The demanufacturing, copper smelting, and unregulated impacts are not all equal, as they were
for occupational non-cancer impacts, because other input materials in the fuel production
processes weigh into the impact scores. This did not occur for non-cancer impacts because the
extremely high non-cancer HVs of some of the solder metals (e.g., lead) overshadowed any
impacts from other processes, such as fuel production.
3-124
-------
Table 3-76. Occupational cancer impacts by life-cycle stage and
process group (paste solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
BSA
Score* %
SABC
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Bi production
Total
5.81E+00 7.62
2.16E-01 0.284
N/A N/A
N/A N/A
N/A N/A
6.03E+00 7.90
8.50E+00 11.8
N/A N/A
9.23E-01 1.28
3.77E-03 0.0052
N/A N/A
9.43E+00 13.1
4.35E+00 6.87
N/A N/A
2.75E-01 0.435
N/A N/A
5.49E-01 0.866
5.18E+00 8.17
8.58E+00 11.9
N/A N/A
5.94E-01 0.821
3.15E-03 0.0044
8.30E-03 0.0115
9.18E+00 12.7
MANUFACTURING
Solder
manufacturing
Post-industrial
recycling
Total
1.60E+01 21.1
4.66E+00 6.12
2.07E+01 27.2
1.37E+01 19.0
4.15E+00 5.77
1.79E+01 24.8
1.56E+01 24.6
1.91E+00 3.01
1.75E+01 27.6
1.39E+01 19.2
4.14E+00 5.72
1.80E+01 24.9
USE/APPLICATION
Reflow application
Total
4.14E+01 54.3
4.14E+01 54.3
3.80E+01 52.8
3.80E+01 52.8
3.27E+01 51.6
3.27E+01 51.6
3.83E+01 52.9
3.83E+01 52.9
END-OF-LIFE
Landfill
Incineration
Demanufacturing
Cu smelting
Unregulated
Total
GRAND TOTAL
5.48E+00 7.19
1.43E+00 1.87
3.43E-01 0.451
5.15E-01 0.675
3.40E-01 0.446
8.11E-HM) 10.6
7.62E+01 100
4.53E+00 6.29
1.18E+00 1.64
2.84E-01 0.394
4.32E-01 0.600
2.81E-01 0.390
6.71E+00 9.31
7.20E+01 100
5.78E+00 9.12
1.51E+00 2.38
3.47E-01 0.547
N/A N/A
3.43E-01 0.541
7.98E+00 12.6
6.34E-K)! 100
4.62E+00 6.39
1.20E+00 1.66
2.90E-01 0.400
4.38E-01 0.606
2.87E-01 0.396
6.84E-K)0 9.45
7.23E+01 100
*The impact scores are in units of kilograms cancertox-equivalents/1,000 cc of solder paste applied to a printed
wiring board.
N/A=not applicable
Upstream occupational cancer impacts arise from the inputs to the extraction and
processing of the various metals present in the alloys. When evaluating the upstream impacts
alone, the tin production process group is the greatest contributor for all alloys, responsible for
about 7 to 12 percent of total impacts. For SAC and SABC, silver production is the second
greatest upstream contributor (1.3 and 0.82 percent, respectively). For BSA, bismuth production
is the second greatest contributor at 0.87 percent, followed by silver production at 0.44 percent.
Top Contributors to Occupational Impacts (Paste Solder)
Table 3-77 presents the specific materials or flows contributing at least 1 percent of
occupational non-cancer impacts by solder. The top contributors are driven by inputs in the
use/application stage, manufacturing stage, and EOL stage. Solder paste inputs to reflow
application are the top contributors for each solder paste, accounting for 31 to 33 percent of total
3-125
-------
impacts, depending on the alloy. The next greatest contributors are primary lead or silver used in
paste manufacturing (25 to 26 percent), and solder on PWBs going to landfilling (22 to 23
percent). Secondary (i.e., recycled) alloys used in solder manufacturing contribute between 4
and 11 percent to total occupational non-cancer impacts. Smaller contributors to total
occupational non-cancer impacts are solder on PWBs going to incineration (contributing about 6
percent), copper smelting (1 percent), unregulated recycling/disposal (1 percent), and
demanufacturing (1 percent).
To better understand how the impact scores are derived and why lead-based impacts are
far greater than other impacts in this impact category, an example from the solder manufacturing
process is presented here. The quantity of primary and secondary lead in the input inventory for
SnPb solder manufacturing is 2.3 kg per functional unit. This quantity is then multiplied by a
toxicity HV to provide a toxicity equivalency for each potentially toxic chemical. For lead, the
non-cancer HV is high (e.g., about 62,400, which is a unitless, relative value based on the
quotient of the mean inhalation NOAEL for 84 chemicals of 69 mg/m3 and a lead inhalation
NOAEL value of 0.0011 mg/m3). Lead's high HV gives it a very high relative toxicity compared
to other toxic materials, which causes the occupational non-cancer impacts from lead to be far
greater than those from other chemicals in the input inventory, especially when combined with
lead's relatively high input amount. In addition, this high score for lead causes the SnPb alloy
impacts to be far greater than those from the other alloys that do not contain lead.
For the lead-free alloys, silver has the highest non-cancer toxicity of the constituent
metals, although the toxicity is not as great as that of lead. For example, in solder manufacturing
the inventory input quantities of silver for the three lead-free alloys range from 0.061 to 0.21
kg/functional unit, and the silver non-cancer HV is 10,000 (unitless), based on an oral LOAEL.
Although the relative toxicity is less than that of lead, the silver toxicity (indicated by the HV) is
large and causes the manufacturing impacts for the lead-free solders to be driven by silver. This
is true even though, compared to the other metals, the relative quantity of silver in the alloys is
small and the actual inventory amount is small. Similarly, silver-bearing alloys at the EOL
contribute significantly to the total impacts for the lead-free alloys. Again, this is because the
HVs for the alloys are a weighted average of the HVs of the constituent metals, and the non-
cancer HV for silver is 10,000 (unitless), compared to those of tin, copper, and bismuth, which
are 1, 26, and 0.0043, respectively.
3-126
-------
Table 3-77. To
Solder
SnPb
SAC
BSA
SABC
Life-Cycle Stage
Use/application
Manufacturing
End-of-life
Manufacturing
End-of-life
End-of-life
End-of-life
End-of-life
Manufacturing
Use/application
Manufacturing
End-of-life
Manufacturing
End-of-life
End-of-life
End-of-life
End-of-life
Use/application
Manufacturing
End-of-life
End-of-life
Manufacturing
End-of-life
End-of-life
Use/application
Manufacturing
End-of-life
Manufacturing
End-of-life
End-of-life
End-of-life
End-of-life
p contributors to occupational non-cancer impacts (paste solder)
Process
SnPb (paste) reflow application
SnPb paste manufacturing
Solder landfilling (SnPb)
SnPb paste manufacturing
Solder incineration (SnPb)
Post-consumer copper smelting
(SnPb)
Demanufacturing- SnPb
Unregulated recycling and disposal
(SnPb)
Sn-Pb paste manufacturing
SAC (paste) reflow application
SAC paste manufacturing
Solder landfilling (SAC)
SAC paste manufacturing
Solder incineration (SnAgCu)
Unregulated recycling and disposal
(SAC)
Demanufacturing-SAC
Post-consumer copper smelting
(SAC)
BSA (paste) reflow application
BSA paste manufacturing
Solder landfilling (BSA)
Solder incineration (BSA)
BSA paste manufacturing
Unregulated recycling and disposal
(BSA)
Demfg-BSA
SABC (paste) reflow application
SABC paste manufacturing
Solder landfilling (SABC)
SABC paste manufacturing
Solder incineration (SABC)
Post-consumer copper smelting
(SABC)
Unregulated recycling and disposal
(SABC)
Demanufacturing-SABC
Flow
Sn-Pb solder paste
Lead (99.995%)
Sn-Pb solder on PWB to landfill
Sn-Pb alloy secondary
Sn-Pb solder on PWB to
incineration
Sn-Pb solder on shredded PWB
Sn-Pb solder on PWB to recycling
Sn-Pb solder to unregulated
recycling
Lead secondary
SAC solder paste
Silver
SAC solder on PWB to landfill
SAC alloy secondary
SAC solder on PWB to
incineration
SAC solder to unregulated
recycling
SAC solder on PWB to recycling
SAC solder on shredded PWB
BSA solder paste
Silver
BSA solder on PWB to landfill
BSA solder on PWB to
incineration
BSA alloy secondary
BSA solder to unregulated
recycling
BSA solder on PWB to recycling
SABC solder paste
Silver
SABC solder on PWB to landfill
SABC alloy secondary
SABC solder on PWB to
incineration
SABC solder on shredded PWB
SABC solder to unregulated
recycling
SABC solder on PWB to recycling
%
Contribution
31.2
24.5
22.4
10.6
5.92
1.40
1.40
1.40
1.18
31.5
25.3
22.7
9.49
5.99
1.42
1.42
1.42
32.5
25.8
23.4
6.17
4.43
1.46
1.46
31.5
25.1
22.6
9.39
5.96
1.41
1.41
1.41
3-127
-------
Table 3-78 presents the specific materials or flows contributing at least 1 percent of
occupational cancer impacts by solder. Natural gas from electricity generation needed for
reflow application is the greatest contributor to occupational cancer impacts for all solder paste
alloys, ranging from 38 to 43 percent contribution of total impacts depending on the solder. The
high impact score for natural gas is primarily due to the large amount of natural gas inputs to the
electricity generation process. No cancer WOE classification or slope factor was available for
natural gas. Consequently, it was assigned a default cancer HV of 1, representative of a mean
HV. The remaining top contributors shown in Table 3-78 include several different flows, all of
which contribute approximately 13 percent or less. These include solder paste used in reflow
application processes, natural gas used in tin production, tin used in solder paste manufacturing,
lead used in solder paste manufacturing, and solder on PWBs going to landfills. One particular
input, "casting process additive," is labeled as such to protect the confidentiality of the material.
Flux materials used in production of the paste constitute greater than 1 percent of total
occupational cancer impacts when they are taken together as a whole. None of the individual
flux components, however, account for at least 1 percent of the total impacts and, as such, are not
presented in the table.
Table 3-78. Top contributors to occupational cancer impacts (paste solder)
Solder
SnPb
SAC
Life-Cycle
Stage
Use/application
Use/application
Upstream
End-of-life
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
End-of-life
Manufacturing
Manufacturing
Manufacturing
Use/application
Upstream
Use/application
Manufacturing
End-of-life
Manufacturing
Manufacturing
Manufacturing
End-of-life
Process
Electricity generation for (paste)
reflow application
SnPb (paste) reflow application
Tin production
Solder landfilling (SnPb)
SnPb paste manufacturing
SnPb paste manufacturing
Post-industrial SnPb recycling
SnPb paste manufacturing
SnPb paste manufacturing
Solder incineration (SnPb)
Natural gas production for paste
manufacturing
SnPb paste manufacturing
SnPb paste manufacturing
Electricity generation
Tin production-DEAM
SAC (paste) reflow application
SAC paste manufacturing
Solder landfilling (SAC)
SAC paste manufacturing
Post-industrial SAC recycling
SAC paste manufacturing
Solder incineration (SAC)
Flow
Natural gas (resource)
SnPb solder paste
Natural gas (resource)
SnPb solder on PWB to landfill
Casting process additive
Tin
Dross
SnPb alloy secondary
Lead (99.995%)
SnPb solder on PWB to
incineration
Natural gas (resource)
Natural gas free customer USA
LFSP fluxes *
Natural gas (resource)
Natural gas (resource)
SAC solder paste
Tin
SAC solder on PWB to landfill
Casting process additive
Dross
SAC alloy secondary
SAC solder on PWB to
%
Contribution
43.2
10.9
7.60
7.12
4.95
4.89
4.64
3.36
2.87
1.88
1.47
1.41
1.22
43.0
11.8
9.71
7.58
6.23
4.58
3.77
2.61
1.64
3-128
-------
Table 3-78. Top contributors to occupational cancer impacts (paste solder)
Solder
BSA
SABC
Life-Cycle
Stage
Manufacturing
Manufacturing
Upstream
Manufacturing
Manufacturing
Use/application
Use/application
End-of-life
Manufacturing
Upstream
Manufacturing
Manufacturing
End-of-life
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Use/application
Upstream
Use/application
Manufacturing
End-of-Life
Manufacturing
End-of-Life
Manufacturing
End-of-Life
End-of-Life
Manufacturing
Manufacturing
Manufacturing
Process
Post-industrial SAC recycling
Natural gas production for solder
manufacturing
Silver production
SAC paste manufacturing
SAC paste manufacturing
Electricity generation for (paste)
reflow application
BSA (paste) reflow application
Solder landfilling (BSA)
BSA paste manufacturing
Tin production
BSA paste manufacturing
BSA paste manufacturing
Solder incineration (BSA)
Post-industrial BSA recycling
BSA paste manufacturing
BSA paste manufacturing
Natural gas production for solder
manufacturing
BSA paste manufacturing
Electricity generation
Tin production
SABC (paste) reflow application
SABC paste manufacturing
Solder landfilling (SABC)
SABC paste manufacturing
Post-industrial SABC recycling
SABC paste manufacturing
Solder incineration (SABC)
Post-industrial SABC recycling
Natural gas production for solder
manufacturing
SABC paste manufacturing
SABC paste manufacturing
Flow
incineration
Heavy fuel oil
Natural gas (resource)
Natural gas (resource)
Natural gas free customer USA
LFSP fluxes *
Natural gas (resource)
BSA solder paste
BSA solder on PWB to landfill
Bismuth (co-mined from Pb, Cu)
Natural gas (resource)
Casting process additive
Tin
BSA solder on PWB to
incineration
Dross
BSA alloy secondary
LFSP fluxes *
Natural gas (resource)
Natural gas free customer USA
Natural gas (resource)
Natural gas (resource)
SABC solder paste
Tin
SABC solder on PWB to landfill
Casting process additive
Dross
SABC alloy secondary
SABC solder on PWB to
incineration
Heavy fuel oil
Natural gas (resource)
Natural gas free customer USA
LFSP fluxes *
%
Contribution
1.45
1.29
1.28
1.24
1.13
37.9
13.2
8.58
7.88
6.80
6.02
4.38
2.27
2.27
1.63
1.48
1.30
1.24
42.9
11.8
9.85
7.61
6.33
4.58
3.74
2.63
1.67
1.44
1.29
1.24
1.13
* The fluxes have been combined together to represent one flow. Taken individually, the fluxes do not contribute
at least 1 percent of the total occupational cancer impact score.
Of note is that none of the top material contributors to the occupational cancer impacts
are known or suspected human carcinogens with slope factors that would give a hazard value
other than one or zero. They either have a cancer WOE classification that results in a cancer HV
3-129
-------
of either zero or one, or they lack data and are given a cancer HV of one. For example, based on
their respective WOE designations, lead has a cancer HV equal to one and silver has a cancer
HV equal to zero. The solder paste and solders on the PWBs at EOL have cancer HVs slightly
below one because they are the weighted average of the individual metals' HVs that are a
combination of one and zero values. This indicates that all the top contributors to this impact
category are used in large enough quantities in the inventory to make them top contributors, but
their carcinogenicity is largely unknown. The occupational cancer impacts, therefore, represent
a lack of data rather than known carcinogenic hazards.
3.2.11.3 Bar solder results
Total Occupational Impacts by Life-Cycle Stage (Bar Solder)
Table 3-79 presents the bar solder results for occupational non-cancer impacts by life-
cycle stage, based on the impact assessment methodology presented above. The table below lists
the occupational non-cancer impact scores per functional unit for the life-cycle stages of each bar
solder alloy, as well as the percent contribution of each life-cycle stage to the total impacts.
Figure 3-27 shows the results in a stacked bar chart.
Table 3-79. Occupational non-cancer impacts by life-cycle stage (bar solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
5.84E+00 0.0008
2.22E+05 31.1
2.13E+05 29.9
2.79E+05 39.1
7.15E+05 100
SAC
Score* %
1.36E+01 0.125
3.53E+03 32.5
3.17E+03 29.2
4.14E+03 38.1
1.09E+04 100
SnCu
Score* %
9.23E+00 14.1
2.07E+01 31.7
2.25E+01 34.5
1.28E+01 19.7
6.53E+01 100
*The impact scores are in units of kg noncancertox-equivalents/1,000 cubic centimeters of solder applied to a
printed wiring board.
3-130
-------
onn nnn
4-1
'E
3 ynn nnn
ro
o
•" 600 000 -
c
,3
"Si 500 000 -
'c
0)
ra 4nn nnn
'5
o-
9 300 000
X
%
o 200 000 -
c
ro
c 1 00 000 -
o
c
o) n
ji u ^
SnPb
SAC SnCu
c . f ..,
• Manufacturing
Figure 3-27. Bar Solder Total Life-Cycle Impacts: Occupational Non-Cancer
As described with the paste solder results, occupational impact scores are based on the
potential toxicity of material inputs to each process. As mentioned above, this characterization
method does not necessarily indicate where actual exposure is occurring; instead, it uses the
inputs of potentially toxic materials as surrogates for potential exposure.
As shown in the figure, the occupational non-cancer impact score for SnPb (715,000 kg
noncancertox-equivalents/functional unit) is far greater than the scores for the other solder alloys
(10,900 and 65.3 kg noncancertox-equivalents/functional unit). Because SnPb has a higher
inherent toxicity compared to the other alloys (based on the toxicity of the constituent metals),
its potential impacts are larger.
Three life-cycle stages largely contribute to total impacts, regardless of the solder type:
manufacturing, use/application, and EOL. The EOL stage was the largest contributor for SnPb
(39 percent) and SAC (38 percent), followed by the manufacturing stage (31 and 33 percent),
and the use/application stage (30 and 29 percent). Upstream impacts for SnPb and SAC are
nominal (0.0008 and 0.125 percent). For SnCu, the same three life-cycle stages dominate,
however, the use/application stage is the top contributor at nearly 35 percent, followed by the
manufacturing stage (32 percent), and the EOL stage (20 percent). The upstream impacts are a
larger percent (14 percent) of the total impacts for SnCu than it is for the other alloys. SnCu is
different from SnPb and SAC since it does not contain the highly toxic lead or silver, thus, the
overall distribution of impacts among life-cycle stages is different. SnCu is more driven by the
quantity of materials with more modest toxicities rather than very high toxicities of a few
materials.
Table 3-80 presents the bar solder results for occupational human health cancer impacts
by life-cycle stage, based on the impact assessment methodology presented above. The table
3-131
-------
lists the occupational cancer impact scores per functional unit for the life-cycle stages of each
bar solder, as well as the percent contribution of each life-cycle stage to the total impacts.
Figure 3-28 presents the results in a stacked bar chart.
Table 3-80. Occupational cancer impacts by life-cycle stage (bar solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
5.84E+00 9.83
1.89E+01 31.8
2.23E+01 37.6
1.23E+01 20.8
5.94E-K)! 100
SAC
Score* %
1.33E+01 23.2
1.30E+01 22.6
2.09E+01 36.3
1.03E+01 17.9
5.75E+01 100
SnCu
Score* %
9.23E+00 16.8
1.39E+01 25.4
2.11E+01 38.4
1.06E+01 19.4
5.49E-K)! 100
*The impact scores are in units of kg cancertox-equivalents/1,000 cubic centimeters of solder applied to a printed
wiring board.
70
+j
3 fin
CB
o
•n £>r>
t5 ou
"Si An
C
a>
ro
•* ^n
5 JU "
o-
01
X 9D
o ^u
1_
Ol
m 10 -
O
O)
J£
n
SnPb
SAC
SnCu
n End-of-life
n Use/application
• Manufacturing
n Upstream
Figure 3-28. Bar Solder Total Life-Cycle Impacts: Occupational Cancer
As shown in the preceding table and figure, SnPb has the greatest occupational cancer
impact score (59.4 kg cancertox-equivalents/functional unit), but its score is not significantly
higher than those for SAC and SnCu (57.5 and 54.9 kg cancertox-equivalents/functional unit,
respectively). In fact, the results for these three alloys may be indistinguishable given the
uncertainties in the data.
Similar to the paste results, the bar solder occupational cancer scores are impacted largely
by each of the four life-cycle stages. For all three bar solders, the use/application stage is the
greatest contributor to total occupational cancer impacts, ranging from 36 to 38 percent.
Potential impacts from the manufacturing stage range from 23 to 32 percent, while EOL stage
3-132
-------
impacts range from 18 to 21 percent depending on the alloy. Contributions from the upstream
life-cycle stage range from 10 to 23 percent.
As discussed in the paste results for occupational cancer toxicity, very few chemicals in
the inventory are known carcinogens or have some quantitative measure of carcinogenicity. The
lack of carcinogenicity data is one of the major limitations and uncertainties in the occupational
cancer characterization method and is addressed further in Section 3.2.11.4 (Limitations and
Uncertainties).
Occupational Impacts by Process Group (Bar Solder)
Table 3-81 lists the occupational non-cancer impacts of each of the process groups in the
life-cycle of the bar solders. As noted above for non-cancer impacts, the manufacturing,
use/application, and EOL stages all largely contribute to occupational non-cancer impacts for all
of the solder alloys. Within the manufacturing stage, the impacts from solder manufacturing are
greater than post-industrial recycling, accounting for 18 to 33 percent of total impacts for all
alloys, compared to less than 0.2 percent for post-industrial recycling. This is because the major
contributors to the manufacturing impacts are the metals inputs used in production of the alloys
(discussed below in the "Top Contributors" section), and the non-cancer hazard values of some
of those metals (e.g., lead and silver) are very high. On the other hand, the inputs to the post-
industrial recycling processes (e.g., dross inputs, which are outputs from the solder
manufacturing process) do not have associated toxicity data to develop a hazard value, so the
default hazard value is used, which is far below that of lead and silver. Solder manufacturing is
the greatest contributor to occupational non-cancer impacts because it has the greatest quantity
of solder inputs, and because occupational impacts are based on the quantity and potential
toxicity of those inputs.
The wave application process group within the use/application stage is comprised of the
wave soldering process and associated electricity generation. Use/application impacts for
occupational non-cancer, therefore, are from the inputs to the wave solder process itself, as well
as inputs to the electricity generation process.
Landfilling is the greatest contributor to EOL occupational non-cancer impacts (10 to 20
percent of total impacts) for all of the alloys, followed by unregulated recycling/disposal (6 to 12
percent of total impacts. Incineration contributes between 2 and 5 percent of total impacts, while
demanufacturing and copper smelting each contribute approximately 1 percent or less to the total
occupational non-cancer impacts for all bar solder alloys.
Like the solder manufacturing process group discussed above, landfilling and
incineration dominate occupational non-cancer health impacts at EOL because these dispositions
have the greatest inputs of EOL solder, the toxicity and overall quantity of which contribute to
the determination of the overall impact score. Furthermore, the LCIA methodology uses input
quantities as surrogates for exposure, in lieu of incorporating an exposure model as would be
done in a chemical risk assessment. For example, within an alloy life-cycle, at this time most
electronics are destined for landfilling (at least 72 percent) as modeled in the LFSP and, as a
result, the LCIA methodology assumes most occupational exposure to solders occurs during
landfilling. The landfilling impacts dominate EOL within each alloy life-cycle. This occurs
3-133
-------
despite the fact that there may actually be less true occupational exposure to a landfill worker
than to a demanufacuturer or copper smelter worker. Given the screening nature of the LCIA
occupational impact category method, the process with the greatest quantities of potentially toxic
materials would tend to have the greatest impacts for a given set of similar materials. For this
reason, the scores for demanufacturing and unregulated recycling/disposal are identical because
the LFSP model assumes that equal amounts of EOL solder go to both of those dispositions. No
mass is assumed to be lost between demanufacturing inputs and copper smelting inputs. The
occupational non-cancer impacts from demanufacturing and copper smelting are the same
because they have the same mass of solder inputs. They are not the same for SnCu because other
inputs from fuel production processes affect the scores, which are not overshadowed by lead or
silver toxicity as is the case with SnPb and SAC.
Table 3-81. Occupational non-cancer impacts by life-cycle stage and
process group (bar solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
SnCu
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Total
5.64E+00 0.0008
2.01E-01 0.00003
N/A N/A
N/A N/A
5.84E+00 0.0008
1.19E+01 0.110
N/A N/A
1.62E+00 0.0149
6.35E-03 0.0001
1.36E+01 0.125
9.22E+00 14.1
N/A N/A
N/A N/A
6.23E-03 0.0095
9.23E+00 14.1
MANUFACTURING
Solder manufacturing
Post-industrial recycling
Total
2.22E+05 31.1
1.80E+01 0.0025
2.22E+05 31.1
3.52E+03 32.5
5.32E+00 0.0490
3.53E+03 32.5
1.16E+01 17.7
9.12E+00 14.0
2.07E+01 31.7
USE/APPLICATION
Solder application
Total
2.13E+05 29.9
2.13E+05 29.9
3.17E+03 29.2
3.17E+03 29.2
2.25E+01 34.5
2.25E+01 34.5
END-OF-LIFE
Landfill
Incineration
Demanufacture
Cu smelting
Unregulated
Total
GRAND TOTAL
1.40E+05 19.5
3.49E+04 4.88
8.73E+03 1.22
8.73E+03 1.22
8.73E+04 12.2
2.79E+05 39.1
7.15E405 100
2.07E+03 19.1
5.17E+02 4.77
1.29E+02 1.19
1.29E+02 1.19
1.29E+03 11.9
4.14E+03 38.1
1.09E+04 100
6.36E+00 9.74
1.57E+00 2.41
3.99E-01 0.611
5.64E-01 0.865
3.95E+00 6.06
1.28E+01 19.7
6.53E+01 100
*The impact scores are in units of kg noncancertox-equivalents/1,000 cc of solder applied to a printed wiring
board.
N/A=not applicable
When evaluating the bar solder alloys against one another, SnPb has the greatest potential
impacts versus the other alloys because the toxicity of lead is greater than the toxicity of the
materials in the other alloys. These potential impacts are based only on the inherent toxicity of
the materials and not their actual fate, transport, and final exposure.
3-134
-------
Upstream occupational non-cancer impacts arise from the inputs to the extraction and
processing of the various metals present in the alloys. Particularly for SnPb and SAC, the
upstream impacts are very small compared to the total life-cycle impacts. Unlike SnPb and
SAC, SnCu does not have toxic metals in its alloy composition (i.e., lead or silver), therefore, the
impacts across the life-cycle are more evenly spread. Nonetheless, when evaluating the
upstream impacts alone, tin production is the greatest contributor to the upstream impacts for all
alloys. For SAC, the silver production process group is the second greatest upstream contributor
(0.015 percent of total impacts).
Table 3-82 lists the occupational cancer impacts of each of the processes in the life-cycle
of the solders. The use/application stage is the greatest contributor to occupational cancer
impacts for the solders. The wave soldering process is the only process group within this stage;
the only inputs modeled in the wave solder process are bar solder, flux, and electricity. Cancer
impacts from the use/application stage, therefore, are based on the carcinogenic potential of the
bar solder, flux, and any potentially carcinogenic inputs to the electricity generation process.
When comparing alloys, the impacts from the use/application stage alone are all very close in
magnitude with SnPb at 22.3 kg cancertox-equivalents/functional unit, followed closely by SnCu
at 21.1 kg cancertox-equivalents/functional unit, and SAC at 20.9 kg cancertox-
equivalents/functional unit.
Within the manufacturing stage, which is the second greatest contributor to occupational
cancer impacts, the solder manufacturing process group impacts are greater than the post-
industrial process group impacts for each solder. The solder manufacturing process group
accounts for 18 to 19 percent and post-industrial recycling accounts for 4 to 13 percent of total
impacts for all alloys.
Within the EOL stage, landfilling is the greatest contributor (about 9 to 10 percent of
total impacts), followed by unregulated recycling/disposal (about 6 percent), and incineration
(about 2 to 3 percent of total impacts). Demanufacturing and copper smelting are smaller
contributors to the total occupational cancer impacts for all alloys (each less than 1 percent).
Similar to the occupational non-cancer impacts discussed above, landfilling and incineration
dominate impacts for this category because, instead of an exposure model, the impacts are based
on the quantity of inputs to each process that have the potential to be toxic (carcinogenic, in this
case). For example, within an alloy life-cycle, most electronics are destined for landfilling (at
least 72 percent), as modeled in the LFSP, indicating that landfills have the greatest inputs of
solder paste at EOL and, therefore, have the greatest EOL occupational cancer impacts. This is
true despite the fact that there may actually be less occupational exposure to a landfill worker
than to a demanufacuturer or copper smelter worker. Given the screening nature of the LCIA
occupational impact category method, the process with the greatest quantities of potentially toxic
materials would tend to have the greatest impacts for a given set of similar materials.
Table 3-82. Occupational cancer impacts by life-cycle stage
and process group (bar solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
SnCu
Score* %
UPSTREAM
3-135
-------
Table 3-82. Occupational cancer
and process group
impacts by life-cycle stage
(bar solder)
Life-cycle stage
Process group
Sn production
Pb production
Ag production
Cu production
Total
SnPb
Score* %
5.64E+00 9.51
1.93E-01 0.325
N/A N/A
N/A N/A
5.84E+00 9.83
SAC
Score* %
1.19E+01 20.8
N/A N/A
1.37E+00 2.38
6.29E-03 0.0109
1.33E-K)! 23.2
SnCu
Score* %
9.22E+00 16.8
N/A N/A
N/A N/A
6.17E-03 0.0112
9.23E-K)0 16.8
MANUFACTURING
Solder manufacturing
Post-industrial recycling
Total
1.11E+01 18.6
7.85E+00 13.2
1.89E+01 31.8
1.05E+01 18.3
2.51E+00 4.37
1.30E-K)! 22.6
9.64E+00 17.6
4.31E+00 7.85
1.39E-K)! 25.4
USE/APPLICATION
Solder application
Total
2.23E+01 37.6
2.23E+01 37.6
2.09E+01 36.3
2.09E+01 36.3
2.11E+01 38.4
2.11E+01 38.4
END-OF-LIFE
Landfill
Incineration
Demanufacture
Cu smelting
Unregulated
Total
GRAND TOTAL
6.09E+00 10.3
1.50E+00 2.53
3.82E-01 0.643
5.72E-01 0.963
3.78E+00 6.37
1.23E+01 20.8
5.94E+01 100
5.09E+00 8.85
1.26E+00 2.18
3.19E-01 0.555
4.85E-01 0.844
3.16E+00 5.49
1.03E-K)! 17.9
5.75E+01 100
5.25E+00 9.56
1.30E+00 2.36
3.29E-01 0.600
4.94E-01 0.901
3.26E+00 5.94
1.06E-K)! 19.4
5.49E+01 100
*The impact scores are in units of kg cancertox-equivalents/1,000 cubic centimeter of solder applied to a printed
wiring board.
N/A=not applicable
Upstream occupational cancer impacts arise from the inputs to the extraction and
processing of the various metals present in the alloys. When evaluating the upstream impacts
alone, tin production is the greatest contributor for all alloys, responsible for about 10 to 21
percent of total impacts. For SAC, silver production is the second greatest upstream contributor
(2.4 percent).
Top Contributors to Occupational Impacts (Bar Solder)
Table 3-83 presents the specific materials or flows contributing at least 1 percent of
occupational non-cancer impacts by solder. The top contributors are driven by inputs in the
use/application stage, manufacturing stage, and EOL stage for all three alloys, as well as the
upstream stage for SnCu. Bar solder inputs to the wave application process are the top
contributors for each bar solder alloy, accounting for approximately 15 to 30 percent of total
impacts, depending on the alloy. There are several other top contributors depending on the alloy,
including primary lead, silver, or copper used in paste manufacturing (9 to 28 percent), and
solder on PWBs going to landfilling (10 to 20 percent). Solder sent to unregulated
recycling/disposal contributes between 6 and 12 percent, and secondary (i.e., recycled) alloys
3-136
-------
used in solder manufacturing contribute between 4 and 14 percent to total occupational non-
cancer impacts. SnCu does not have impacts from silver or lead; however, SnPb and SAC both
have high relative toxicities. There are other materials that contribute greater than 1 percent to
SnCu impacts that do not appear in the top contributors for SnPb and SAC. For example, flux
materials contribute between 1 and 3 percent to total impacts for SnCu. As discussed in the
paste solder results, the SnPb impacts are far greater than SAC and SnCu due to the high relative
toxicity of lead.
Table 3-83. Top contributors to occupational non-cancer impacts (bar solder)
Solder
SnPb
SAC
SnCu
Life-Cycle
Stage
Use/application
End-of-life
Manufacturing
Manufacturing
End-of-life
End-of-life
End-of-life
End-of-life
Use/application
Manufacturing
End-of-life
End-of-life
End-of-life
Manufacturing
End-of-life
End-of-life
Use/application
Upstream
End-of-life
Manufacturing
Use/application
End-of-life
Manufacturing
Manufacturing
Manufacturing
Use/application
Use/application
Use/application
End-of-life
Process
SnPb (bar) wave application
Solder landfilling (SnPb)
SnPb bar manufacturing
SnPb bar manufacturing
Unregulated recycling and disposal
(SnPb)
Solder incineration (SnPb)
Post-consumer copper smelting
(SnPb)
Demanufacturing-SnPb
SAC (bar) wave application
SAC bar manufacturing
Solder landfilling (SAC)
Unregulated recycling and disposal
(SAC)
Solder incineration (SAC)
SAC bar manufacturing
Post-consumer copper smelting
(SAC)
Demanufacturing-SAC
SnCu (bar) wave application
Tin production
Solder landfilling (SnCu)
SnCu bar manufacturing
Electricity generation
Unregulated recycling and disposal
(SnCu)
Post-Industrial SnCu recycling
Post-Industrial SnCu recycling
SnCu bar manufacturing
SnCu (bar) wave application
SnCu (bar) wave application
SnCu (bar) wave application
Solder incineration (SnCu)
Flow
SnPb solder bar
SnPb solder on PWB to landfill
Lead
SnPb alloy secondary
SnPb solder to unregulated
recycling
SnPb solder on PWB to
incineration
SnPb solder on shredded PWB
SnPb solder on PWB to recycling
SAC solder bar
Silver
SAC solder on PWB to landfill
SAC Solder to unregulated
recycling
SAC solder on PWB to
incineration
SAC alloy secondary
SAC solder on shredded PWB
SAC solder on PWB to recycling
SnCu solder bar
Natural gas (resource)
SnCu solder on PWB to landfill
Tin
Natural gas (resource)
SnCu solder to unregulated
recycling
Fluorosilicic acid
Dross
Sn-Cu alloy secondary
FluxC*
FluxD*
FluxF*
Sn-Cu solder on PWB to
incineration
% Contribution
29.8
19.5
17.1
14.0
12.2
4.88
1.22
1.22
29.1
28.1
19.1
11.9
4.77
4.30
1.19
1.19
14.8
14.1
9.66
9.06
8.33
6.04
5.14
4.31
3.76
3.12
2.60
2.60
2.42
3-137
-------
Table 3-83. Top contributors to occupational non-cancer impacts (bar solder)
Solder
Life-Cycle
Stage
Manufacturing
Manufacturing
Manufacturing
Use/application
Use/application
Process
Post-Industrial SnCu recycling
SnCu bar manufacturing
Post-industrial SnCu recycling
SnCu (bar) wave application
SnCu (bar) wave application
Flow
Fluoroboric acid
Copper
Heavy fuel oil
FluxE*
Flux A*
% Contribution
2.25
2.09
1.66
1.56
1.04
The chemical names of the fluxes have been withheld to protect confidentiality.
Table 3-84 presents the specific materials or flows contributing at least 1 percent of
occupational cancer impacts by solder. The top contributors to the SnPb impacts are bar solder
from wave application, solder on a PWB going to a landfill, and dross inputs to post-industrial
recycling. For SAC and SnCu, the top contributors are natural gas from tin production, bar
solder from wave application, and tin from bar manufacturing. As explained under the paste
solder results, the high impact score for natural gas is primarily due to the relatively large
amount of natural gas inputs to the associated processes. No cancer WOE classification or slope
factor was available for natural gas. Consequently, it was assigned a default cancer HV of 1,
representative of a mean HV. The remaining top contributors shown in Table 3-84 include
several different flows, all of which contribute approximately 10 percent or less.
Table 3-84. Top contributors to occupational cancer impacts (bar solder)
Solder
SnPb
SAC
Life-Cycle Stage
Use/application
End-of-life
Manufacturing
Upstream
Use/application
Manufacturing
End-of-life
Manufacturing
Use/application
Manufacturing
Use/application
Use/application
End-of-life
Manufacturing
Use/application
Use/application
Use/application
Upstream
Use/application
Process
SnPb (bar) wave application
Solder landfilling (SnPb)
Post-industrial SnPb recycling
Tin production
Electricity generation
SnPb bar manufacturing
Unregulated recycling and disposal
(SnPb)
SnPb bar manufacturing
SnPb (bar) wave application
SnPb bar manufacturing
SnPb (bar) wave application
SnPb (bar) wave application
Solder incineration (SnPb)
Post-industrial SnPb recycling
SnPb (bar) wave application
SnPb (bar) wave application
SnPb (bar) wave application
Tin production
SAC (bar) wave application
Flow
SnPb solder bar
SnPb solder on PWB to landfill
Dross
Natural gas (resource)
Natural gas (resource)
SnPb alloy secondary
SnPb solder to unregulated
recycling
Tin
FluxC*
Lead
FluxD*
FluxF
SnPb solder on PWB to incineration
Heavy fuel oil
Flux E *
Flux A *
Flux B *
Natural gas (resource)
SAC solder bar
%
Contribution
15.5
10.1
10.0
9.47
8.77
7.26
6.34
6.09
3.43
3.29
2.86
2.86
2.54
1.92
1.72
1.14
1.14
20.7
13.4
5-138
-------
Table 3-84. Top contributors to occupational cancer impacts (bar solder)
Solder
SnCu
Life-Cycle Stage
Manufacturing
Use/application
End-of-life
End-of-life
Use/application
Use/application
Use/application
Manufacturing
Upstream
End-of-life
Manufacturing
Use/application
Use/application
Use/application
Manufacturing
Manufacturing
Manufacturing
Upstream
Use/application
Manufacturing
Use/application
End-of-life
End-of-life
Manufacturing
Use/application
Manufacturing
Use/application
Use/application
End-of-life
Manufacturing
Manufacturing
Manufacturing
Use/application
Use/application
Use/application
Manufacturing
Process
SAC bar manufacturing
Electricity generation
Solder landfilling (SAC)
Unregulated recycling and disposal
(SAC)
SAC (bar) wave application
SAC (bar) wave application
SAC (bar) wave application
Post-industrial SAC recycling
Silver production
Solder incineration (SAC)
SAC bar manufacturing
SAC (bar) wave application
SAC (bar) wave application
SAC (bar) wave application
Post-industrial SAC recycling
Natural gas production in solder
manufacturing
SAC bar manufacturing
Tin production
SnCu (bar) wave application
SnCu bar manufacturing
Electricity generation
Solder landfilling (SnCu)
Unregulated recycling and disposal
(SnCu)
Post-industrial SnCu recycling
SnCu (bar) wave application
SnCu bar manufacturing
SnCu (bar) wave application
SnCu (bar) wave application
Solder incineration (SnCu)
Post-industrial SnCu recycling
Post-industrial SnCu recycling
Natural gas production for solder
manufacturing
SnCu (bar) wave application
SnCu (bar) wave application
SnCu (bar) wave application
SnCu bar manufacturing
Flow
Tin
Natural gas (resource)
SAC solder on PWB to landfill
SAC Solder to unregulated
recycling
FluxC*
FluxD*
FluxF*
Dross
Natural gas (resource)
SAC solder on PWB to incineration
SAC alloy secondary
FluxE*
Flux A*
FluxB*
Heavy fuel oil
Natural gas (resource)
Natural gas products
Natural gas (resource)
SnCu solder bar
Tin
Natural gas (resource)
SnCu solder on PWB to landfill
SnCu solder to unregulated
recycling
Dross
Flux C *
Sn-Cu alloy secondary
Flux D *
Flux F *
Sn-Cu solder on PWB to
incineration
Crude oil products
Heavy fuel oil
Natural gas (resource)
FluxE*
Flux A *
Flux B *
Natural gas products
%
Contribution
13.3
9.16
8.76
5.47
3.54
2.96
2.96
2.85
2.37
2.19
1.97
1.77
1.18
1.18
1.10
1.06
1.02
16.7
14.5
10.8
9.60
9.47
5.92
5.12
3.71
3.68
3.09
3.09
2.37
2.18
1.97
1.86
1.23
1.23
1.11
1.06
* The chemical names of the fluxes have been withheld to protect confidentiality.
As discussed with the paste results, none of the top material contributors to the
occupational cancer impacts are known or suspected human carcinogens with slope factors that
5-139
-------
would give a hazard value other than one or zero. They either have a cancer WOE classification
that results in a cancer HV of either one or zero, or they lack data and are given a cancer HV of
one. Thus, all the top contributors to this impact category are used in large enough quantities in
the inventory to make them top contributors, but their carcinogenicity is largely unknown. The
occupational cancer impacts, therefore, represent a lack of data rather than known carcinogenic
hazards.
3.2.11.4 Limitations and uncertainties
Most of the limitations and uncertainties associated with the chronic human health results
presented here and in Section 3.2.12 can be grouped into three categories:
1. Structural or modeling limitations and uncertainties associated with the accuracy of the
toxic chemical classification method and the chemical scoring approach used to
characterize human health effects.
2. Toxicity data limitations and uncertainties associated with the availability and accuracy
of toxicity data to represent potential human health effects.
3. LCI data limitations and uncertainties associated with the accuracy and
representativeness of the inventory data.
Each of these is discussed below:
Structural or modeling limitations and uncertainties. The chemical scoring method used
in the human health effects impact characterization is a screening tool to identify chemicals of
potential concern, not to predict actual effects or characterize risk. A major limitation in the
method is that it only measures relative toxicity combined with inventory amount. It does not
take chemical fate, transportation, or degradation into account. In addition, it uses a simple
surrogate value (e.g., inventory amount) to evaluate the potential for exposure, when actual
exposure potential involves many more factors, some of which are chemical-specific. The LCIA
method for toxicity impacts also takes the most toxic endpoint to calculate a hazard value,
regardless of the route of exposure (e.g., inhalation or ingestion); therefore, this approach does
not model true potential exposures, but rather the relative toxicity as compared to other
chemicals, to compare life-cycle results among alloys. This is addressed further in Section
3.2.12.4 with respect to public health impacts.
Other sources of uncertainty include possible omissions by the LFSP researchers in the
impact classification process (e.g., potentially toxic chemicals not classified as such) or
misrepresentation of chemicals in the impact characterization method itself (e.g.,
misrepresenting a chemical as a small contributor to total impacts, because of missing or
inaccurate toxicity data). Some of these limitations and uncertainties also may be considered
limits in the toxicity data which are discussed further below.
It should be noted, however, that because LCA involves analyzing many processes over
the entire life-cycle of a product, a comprehensive, quantitative risk assessment of each chemical
input or output cannot be done. Rather, LCA develops relative impacts that often lack temporal
3-140
-------
or spatial specificity, but can be used to identify materials for more detailed evaluation.
Toxicity data limitations and uncertainties. Major uncertainties in the impact assessment
for potentially toxic chemicals result from missing toxicity data and from limitations of the
available toxicity data. Uncertainties in the human health hazard data (as typically encountered
in a hazard assessment) include the following:
• Using dose-response data from laboratory animals to represent potential effects in
humans.
• Using data from homogenous populations of laboratory animals or healthy human
populations to represent the potential effects on the general human populations with a
wide range of sensitivities.
Using dose-response data from high dose toxicity studies to represent potential effects
that may occur at low levels.
Using data from short-term studies to represent the potential effects of long-term
exposures.
• Assuming a linear dose-response relationship.
• Possibly increased or decreased toxicity resulting from chemical interactions.
Uncertainty is associated with using a default HV (i.e., assuming average toxicity for that
measure when a chemical could be either more or less toxic than average) for missing toxicity
data; however, the use of neutral default values for missing data reduces the bias that typically
favors chemicals with little available information. Use of a data-neutral default value to fill data
gaps is consistent with principles for chemical ranking and scoring (Swanson and Socha, 1997).
Of the 177 chemicals classified as potentially toxic in this LFSP LCA, 81 (46 percent) had no
toxicity data for non-carcinogenic effects and 88 (50 percent) had no toxicity data for
carcinogenic effects (e.g., WOE classification or slope factor). Sixty chemicals (34 percent) had
no human health toxicity date whatsoever.
Specific to carcinogenic effects, the lack of measured carcinogenicity data is a major
uncertainty in the occupational cancer results. The 88 potentially toxic chemicals with no
carcinogenic toxicity data receive a median HV (HV=1), which is equal to the HV assigned to
known or suspected carcinogens with no slope factor. Of the 89 chemicals that have cancer data,
30 received an HV of zero because they have WOE classifications of D or E or IARC
classifications of 3 or 4 (i.e., not classifiable, non-carcinogenic, or probably not carcinogenic).
Of the remaining 59 known or suspected carcinogens, 25 have the slope factors needed to
calculate a hazard value other than 1, and none of the top material contributors to the
occupational cancer impacts that are known or suspected human carcinogens have slope factors.
The occupational cancer impacts, therefore, are largely distributed among the material inputs
used in the greatest quantity in the solder life-cycle, but the relative carcinogenicity of these
materials is uncertain.
For the solder alloys, either in paste or solid form, direct toxicity data are not available;
however, instead of being given default HVs, they are given HVs based on the weighted average
of the HVs of the constituent metals and fluxes (when applicable). Although the resulting HVs
3-141
-------
are not known to be completely representative of an appropriate HV for a solder, they are
assumed to be the best estimates for this screening methodology given the available data. This
introduces uncertainty only for the occupational impacts as the solders themselves are inputs to
given processes and it is the inputs that are the basis for the impact characterization for
occupational impacts. (Note that because the solders are given toxicity HVs does not mean that
they are designated RCRA toxic wastes by the U.S. EPA; it only indicates that there is a
potential for exposure to potentially toxic materials.) For the public health impacts, scores are
based on outputs, which are the environmental releases of the individual metals when the solders
break down and do not include the solders as a whole. The uncertainty in estimating an HV for
an alloy using a weighted average of the constituent metals does not affect the public health
impact categories. Instead, for the public health impacts which are based on outputs, there is
uncertainty associated with predictions of how the metal constituents are partitioned and released
to the environment, which is related to limitations in the inventory (discussed below).
LCI data limitations and uncertainties. For both paste and bar solders, the majority of non-
cancer occupational impacts are spread out among three stages: manufacturing, EOL, and
application stages. In most cases, the greatest impacts are from lead, silver, or secondary alloy
inputs in manufacturing; solder used in application; and solder on PWBs at EOL. The quantities
of these materials in the inventory represent surrogates for exposure. As a result, the potential
relative toxicity of each alloy across their life-cycles is affected by (1) the amount of lead and
silver inputs, which is closely related to the percent composition of those metals in the alloys; (2)
the amount of paste or bar solder used in the application process, which is related to the volume
of paste used, as determined with the functional unit definition; and, (3) the solder on a board at
EOL, which is based on the functional unit definition. The lead and silver inputs from solder
paste manufacturing data were collected as primary data for this project from three major
manufacturers and averaged together. These data are considered to be of good quality as
discussed in Chapter 2 and, therefore, the inventory uncertainty and limitations associated with
the occupational non-cancer impacts from manufacturing are not anticipated to be too great. The
impacts from application and EOL are based on the volume of solder applied to a board, which is
the defined functional unit. This is based on the physical densities of the individual solders and
is not expected to be a source of uncertainty in the inventory; however, there are EOL
uncertainties related to the assumptions about EOL dispositions (e.g., 72 percent of solder goes
directly to landfilling for SnPb, SAC, SABC, and SnCu) which determines the relative amount of
solder in a functional unit assumed to be sent to each disposition. These are discussed in greater
detail in Chapter 2, limitations and uncertainties in the EOL inventory.
The LCI data limitations for occupational cancer results also are similar to those for
occupational non-cancer results; however, because the top contributing impacts in this impact
category are from all life-cycle stages, the limitations and uncertainties are related to all life-
cycle stages. In summary, the use/application limitations and uncertainties related to electricity
inputs arise from the following: (1) for reflow soldering, reflow energy is based on a limited
number of data points that cover a wide range, and (2) for reflow and wave soldering, electricity
production data are from a secondary source. The reflow energy data are the subject of a
sensitivity analysis in Section 3.3, but issues associated with electricity production data are not
considered to be significant.
3-142
-------
Uncertainties and limitations from the solder inputs in the use/application stage, the metal
inputs in the solder manufacturing processes, and the solders on PWBs at EOL are related to the
functional unit definition. Data on these solder inputs are from primary data collected for this
project and are considered to be of good quality with no major limitations or uncertainties. EOL
uncertainties, as mentioned above, are related to the assumptions about the percent of solder
going to the various EOL dispositions. Limitations and uncertainties from the upstream life-
cycle stage arise from the fact that the upstream metals production data are from secondary
sources.
3-143
-------
3.2.12 Public Human Health Impacts
This section presents the LCIA characterization methodology and the LCIA results for
the public human health impact category. General information that is common to all the toxicity
impact categories (i.e., occupational human health, public human health, and ecological toxicity)
was presented in Section 3.2.1 1 and is applicable to this section. For chronic public health
effects, the impact scores represent surrogates for potential health effects to residents living near
a facility from long-term repeated exposure to toxic or carcinogenic agents. Impact scores are
calculated for both cancer and non-cancer effects, and are based on the identity and amount of
toxic chemical outputs with dispositions to air, soil and water.1 As stated previously, inventory
items do not truly represent long-term exposure, instead impacts are relative toxicity weightings
of the inventory.
The scores for impacts to the public differ from occupational impacts in that inventory
outputs are used as opposed to inventory inputs. This basic screening level scoring does not
incorporate the fate and transport of the chemicals. The public human health impact results
presented in this section include two impact categories: public non-cancer impacts and public
cancer impacts.
3.2.12.1 Characterization
Section 3.2.11.1 {Potential Human Health Impacts) provides a general discussion of the
human health characterization approach in this LCIA. Below are the specific equations used to
calculate impact scores for potential public non-cancer and cancer impacts.
Public Human Health Characterization: Non-Cancer
The chronic public health effects impact score for non-cancer effects is calculated by:
where:
IScHp-Nc equals the impact score for chronic non-cancer effects to the public for chemical /'
(kg non-cancertox-equivalent) per functional unit;
HVNC equals the hazard value for chronic non-cancer effects for chemical / (based on
either inhalation or oral toxicity, see Section 3.2.11.1); and
AmtTCoutput equals the amount of toxic inventory output of chemical /' to air, water, and soil
(kg) per functional unit.
More detail on the HVNC is provided in Section 3.2.11.1.
1 Disposition to soil includes direct, uncontained releases to soil as could occur from unregulated disposal.
It does not include solid or hazardous waste disposal in a regulated landfill. Disposition to water, however, could
include groundwater if a landfill model shows releases to groundwater, for example.
3-144
-------
Public Human Health Characterization: Cancer
The chronic public health effects impact score for cancer effects is calculated as follows:
where:
ISr
JCHP-CA
HV
nv CA
AmtTC output
x Amt
TCoutpuP
equals the impact score for chronic cancer health effects to the public for
chemical /' (kg cancertox-equivalent) per functional unit;
equals the hazard value for carcinogenicity for chemical /' (based on either
inhalation or oral carcinogenicity, see Section 3.2.11.1); and
equals the amount of toxic inventory output of chemical / to air, water, and soil
(kg) per functional unit.
3.2.12.2 Paste solder results
Total Public Health Impacts by Life-Cycle Stage (Paste Solder)
Table 3-85 presents the solder paste results for public human health non-cancer impacts
by life-cycle stage, based on the impact assessment methodology presented above. The table
lists the public non-cancer impact scores per functional unit for the life-cycle stages of each
paste solder alloy, as well as the percent contribution of each life-cycle stage to the total impacts.
Figure 3-29 presents the results in a stacked bar chart.
Table 3-85. Public non-cancer impacts by life-cycle stage (paste solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
1.95E+02 0.222
4.74E+01 0.0538
2.86E+03 3.25
8.49E+04 96.5
8.80E+04 100
SAC
Score* %
7.80E+03 74.0
3.50E+01 0.333
2.68E+03 25.5
1.74E+01 0.165
1.05E-K)4 100
BSA
Score* %
2.88E+03 57.4
2.02E+01 0.404
2.10E+03 41.9
1.62E+01 0.324
5.01E-KJ3 100
SABC
Score* %
5.10E+03 65.0
3.51E+01 0.447
2.69E+03 34.3
1.64E+01 0.209
7.84E+03 100
*The impact scores are in units of kilograms noncancertox-equivalents/1,000 cc of solder paste applied to a printed
wiring board.
3-145
-------
• End-of-life
D Use/application
I Manufacturing
• Upstream
SnPb
SAC
BSA
SABC
Figure 3-29. Solder Paste Total Life-Cycle Impacts: Public Non-Cancer
The public non-cancer impacts for SnPb (88,000 kg noncancertox-equivalents/functional
unit) are far greater than the other alloys (ranging from 5,010 to 10,500 kg noncancertox-
equivalents/functional unit for BSA and SAC, respectively). The EOL stage dominates impacts
for SnPb, contributing nearly 97 percent to the total SnPb public non-cancer impacts. The EOL
impacts for the other alloys contribute only about 0.2 to 0.3 percent of total impacts. EOL public
non-cancer impacts are much greater for SnPb than the other solders due to lead's high HV
combined with its greater teachability as determined by TCLP testing (see Chapter 2 and
Appendix C), which is discussed further below.
For the lead-free alternatives, the upstream life-cycle stage is the greatest contributor to
overall public non-cancer impacts. SAC has the greatest upstream public non-cancer impacts at
7,800 kg noncancertox-equivalents/functional unit, which is 74 percent of total SAC public non-
cancer impacts. SABC has 5,100 kg noncancertox-equivalents/functional unit or 65 percent
contribution to total SABC impacts. BSA has fewer upstream public non-cancer impacts with
2,880 kg noncancertox-equivalents/functional unit, a 57 percent contribution.
The use/application stage, which is made up of the reflow soldering process group, is the
second greatest contributor for all alloys. Impacts from this life-cycle stage are associated with
outputs from the generation of electricity used to power the reflow ovens and are greatest for the
alloys that consume the most energy during use. For this stage, SnPb has the greatest impacts
(2,860 kg noncancertox-equivalents/functional unit), followed by SABC, SAC, and BSA (2,690,
2,680, and 2,100 kg noncancertox-equivalents/functional unit, respectively). The percent
3-146
-------
contribution of the use/application stage to SnPb total impacts is relatively small (3 percent)
compared to the lead-free alloys (about 26 to 42 percent for SAC and BSA, respectively). This
is due to lead's high HV which causes its impact scores at EOL to be much greater than SnPb
impact scores from solder reflow (e.g., from outputs from electricity generation). Life-cycle
public non-cancer impacts from the manufacturing stage are relatively small for all of the solder
paste alloys, ranging from 20.2 to 47.4 kg noncancertox-equivalents/functional unit or 0.05 to
0.4 percent of total impacts.
Table 3-86 presents the paste solder results for public human health cancer impacts by
life-cycle stage, based on the impact assessment methodology presented above in Section
3.2.12.1. The table lists the public cancer impact scores per functional unit for the life-cycle
stages of each solder paste alloy, as well as the percent contribution of each life-cycle stage to
the total impacts. Figure 3-30 presents the results in a stacked bar chart.
Table 3-86. Public cancer impacts by life-cycle stage (paste solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
3.00E-01 4.31
1.16E-01 1.67
5.09E+00 73.2
1.45E+00 20.8
6.96E+00 100
SAC
Score* %
2.01E+00 28.4
1.36E-01 1.93
4.80E+00 68.1
1.10E-01 1.56
7.05E-K)0 100
BSA
Score* %
7.65E-01 14.9
6.09E-02 1.18
3.97E+00 77.0
3.56E-01 6.92
5.15E+00 100
SABC
Score* %
1.43E+00 22.0
1.36E-01 2.08
4.82E+00 74.1
1.20E-01 1.85
6.51E+00 100
*The impact scores are in units of kilograms cancertox-equivalents/1,000 cc of solder applied to a printed wiring
board.
re
o
tj
3
x
o
r
o
o
O)
7 -
5
4
3 4
• End-of-life
D Use/application
• Manufacturing
B Upstream
SnPb
SAC
BSA
SABC
Figure 3-30. Solder Paste Total Life-Cycle Impacts: Public Cancer
3-147
-------
The total public cancer impact scores for SAC and SnPb are very close at 7.05 and 6.96
kg cancertox-equivalents/functional unit, respectively, the distribution of their impacts across the
solder life-cycle varies; that is, the use/application stage is the greatest contributor for both
alloys. For SnPb, however, the EOL is the second greatest contributor by life-cycle stage, while,
for SAC, the upstream life-cycle stage is the second greatest stage. The alloy with the next
greatest public cancer impact score is SABC at 6.51 kg cancertox-equivalents/functional unit,
while BSA has the lowest total score at 5.15 kg cancertox-equivalents/functional unit. The
use/application stage dominates impacts for all solder alloys, ranging from 68 to 77 percent of
total impacts.
While the EOL stage is the second greatest contributor to the SnPb total impact score at
21 percent of total impacts, it only contributes about 1.6 to 6.9 percent of the total scores of the
lead-free alloys. For these alloys, the upstream life-cycle stage is the second greatest
contributor, ranging from 15 to 28 percent. For SnPb, upstream processes contribute only about
4.3 percent of the total impacts. The manufacturing stage impacts are small for all the solder
paste alloys, ranging from 1.7 to 2.0 percent, depending on the alloy.
To help put the public health impact scores into perspective, they are compared to
impacts from burning a 60-watt lightbulb. The public health toxicity impacts associated with the
electricity used to burn a 60-watt bulb for one day is 4,729 kg noncancertox-equivalents. The
difference between the public health impacts for SnPb and SAC is 77,500 kg noncancertox-
equivalents, which is equivalent to the public health impacts that would be associated with
burning a 60-watt bulb for approximately 16 days straight.
For the cancer impacts, the small difference between SnPb and SAC (i.e., 0.09 kg
cancertox-equivalents) is equivalent to the cancer impacts associated with burning a 60-watt
lightbulb for approximately 18 minutes. The difference between the SnPb and SABC cancer
scores (i.e., 1.18 kg cancertox-equivalents) is equivalent to running a 60-watt bulb continuously
for 4 hours.
Public Health Impacts by Process Group (Paste Solder)
Table 3-87 lists the public non-cancer impacts of each of the process groups in the life-
cycle of the solders. Within the EOL stage of the SnPb life-cycle, landfilling is the greatest
contributor to total impacts (73 percent of total public non-cancer impacts), followed by
incineration (19 percent), and unregulated recycling/disposal (4.5 percent). Copper smelting and
demanufacturing are very small contributors to the total SnPb public non-cancer toxicity impacts
(0.006 and 0.0003 percent, respectively).
EOL processes are much less significant to total public non-cancer impacts for the lead-
free alloys. When evaluating these alloys alone, unregulated recycling and disposal is the
greatest contributor to EOL impacts, with scores of 14.5, 14.8, and 13.1 kg noncancertox-
equivalents/functional unit for SAC, BSA, and SABC, respectively. This process group only
contributes approximately 0.1 to 0.3 percent to the total scores.
For the lead-free solders, the silver production process in the upstream life-cycle stage is
the process group with the greatest contribution to public non-cancer impacts, accounting for 45
to 72 percent of total impacts. The next greatest contributor within the upstream life-cycle stage
3-148
-------
for SAC and SABC is tin production (1.9 and 2.5 percent contribution), while bismuth
production is the next largest contributor for BSA at 10 percent, followed by tin production at
2 percent.
As noted previously, the second greatest contributor to lead-free solder public non-cancer
impacts within all the life-cycle stages is the reflow solder application process, contributing 26 to
42 percent to the total public non-cancer impacts. The solder application process is the fourth
largest contributor to SnPb public non-cancer impacts.
Table 3-87 also shows the contribution of the two process groups—solder manufacturing
and post-industrial recycling—within the manufacturing stage which contribute a small
proportion to the overall impacts for all of the solders. SnPb, SAC, and SABC have similar
impact scores for solder manufacturing (20.5, 18.7, and 18.8 kg noncancertox-
equivalents/functional unit, respectively), while the BSA score is lower (11.7 kg noncancertox-
equivalents/functional unit). For the post-industrial recycling process group, impacts are greatest
for SnPb (26.8 kg noncancertox-equivalents/functional unit), equal for SAC and SABC (16.3 kg
noncancertox-equivalents/functional unit for both), and lowest for BSA (8.52 kg noncancertox-
equivalents/functional unit). Total manufacturing impacts follow the same trend as the total life-
cycle impacts with SnPb being greatest, SAC and SABC being approximately equal, and BSA
being the lowest.
Table 3-87. Public non-cancer impacts by life-cycle stage
and process group (paste solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
BSA
Score* %
SABC
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Bi production
Total
1.33E+02 0.151
6.18E+01 0.0703
N/A N/A
N/A N/A
N/A N/A
1.95E-KJ2 0.222
1.95E+02 1.85
N/A N/A
7.60E+03 72.1
5.47E+00 0.0519
N/A N/A
7.80E+03 74.0
9.97E+01 1.99
N/A N/A
2.27E+03 45.3
N/A N/A
5.08E+02 10.1
2.88E+03 57.4
1.97E+02 2.51
N/A N/A
4.89E+03 62.3
4.58E+00 0.0583
7.68E+00 0.0979
5.10E+03 65.0
MANUFACTURING
Solder
manufacturing
Post-industrial
recycling
Total
2.05E+01 0.0233
2.68E+01 0.0305
4.74E-K)! 0.0538
1.87E+01 0.178
1.63E+01 0.155
3.50E+01 0.333
1.17E+01 0.234
8.52E+00 0.170
2.02E+01 0.404
1.88E+01 0.240
1.63E+01 0.208
3.51E+01 0.447
USE/APPLICATION
Reflow
application
Total
2.86E+03 3.25
2.86E-KJ3 3.25
2.68E+03 25.5
2.68E+03 25.5
2.10E+03 41.9
2.10E+03 41.9
2.69E+03 34.3
2.69E+03 34.3
END-OF-LIFE
3-149
-------
Table 3-87. Public non-cancer impacts by life-cycle stage
and process group (paste solder)
Life-cycle stage
Process group
Landfill
Incineration
Demanufacture
Cu smelting
Unregulated
Total
GRAND
TOTAL
SnPb
Score* %
6.39E+04 72.6
1.71E+04 19.4
2.81E-01 0.0003
4.95E+00 0.0056
3.93E+03 4.47
8.49E+04 96.5
8.80E+04 100
SAC
Score* %
8.95E-01 0.0085
-6.30E-02 -0.0006
2.43E-01 0.0023
1.77E+00 0.0168
1.45E+01 0.138
1.74E+01 0.165
1.05E+04 100
BSA
Score* %
1.21E+00 0.0241
-7.30E-02 -0.0015
2.86E-01 0.0057
N/A N/A
1.48E+01 0.295
1.62E+01 0.324
5.01E+03 100
SABC
Score* %
1.15E+00 0.0147
1.17E-01 0.0015
2.44E-01 0.0031
1.78E+00 0.0226
1.31E+01 0.167
1.64E+01 0.209
7.84E+03 100
*The impact scores are in units of kilograms noncancertox-equivalents/1,000 cc of solder paste applied to a printed
wiring board.
N/A=not applicable
Table 3-88 lists the public cancer impacts of each of the process groups in the life-cycle
of the solders. The impact scores from the use/application stage that dominate the scores for all
alloys are predominately due to potentially carcinogenic outputs from electricity generation in
the reflow application process group. Other contributing outputs are the flux materials released
from the paste during solder reflow. EOL impacts arise from output flows of potentially
carcinogenic materials released from the various EOL processes. Within the SnPb life-cycle,
landfilling is the greatest process group contributor to EOL impacts (15 percent of total public
cancer impacts), followed by incineration (4 percent), and unregulated recycling/disposal (2
percent). Copper smelting and demanufacturing are small contributors to the total SnPb public
cancer impact scores (0.18 and 0.0061 percent, respectively). For SAC and SABC, unregulated
disposal has the highest EOL impact score, albeit a small proportion of total impacts (1.3 and 1.4
percent, respectively). BSA has the most EOL impacts from landfilling, as well as unregulated
recycling and disposal (both about 2.8 percent of the BSA total public cancer impact score),
because it has a different EOL scenario than the other alloys (i.e., after demanufacturing, solder
on boards is not sent to copper smelting, but instead either landfilled or incinerated). Other
processes which contribute include incineration at 1.3 percent of the total impacts and
demanufacturing at 0.01 percent.
Potential upstream impacts arise from outputs of potentially carcinogenic materials in the
extraction and processing of the various metals present in the alloys. In the SnPb life-cycle, the
public cancer impact scores from tin extraction and processing comprise about 3.8 percent of the
total compared to about 0.53 percent for lead extraction and processing. For the lead-free alloys,
silver production dominates upstream impacts, contributing about 9 to 23 percent of the total
score depending on the alloy. Tin production, which is the second greatest contributor to
upstream impacts for the lead-free alloys, accounts for about 4 to 6 percent of the total public
cancer scores. Public cancer impacts from silver processing exceed impacts from tin processing
in solders that contain both metals, even though the silver content of the alloys is much less than
the tin content. For example, SAC is 95.5 percent tin and only 3.9 percent silver, yet its impacts
from silver production are greater than those from tin production. This indicates that potential
3-150
-------
cancer impacts from silver extraction and processing outputs are disproportionately high
compared to the other solder metals.
Table 3-88. Public cancer impacts by life-cycle stage and process group (paste solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
BSA
Score* %
SABC
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Bi production
Total
2.63E-01 3.78
3.71E-02 0.534
N/A N/A
N/A N/A
N/A N/A
3.00E-01 4.31
3.84E-01 5.45
N/A N/A
1.62E+00 23.0
5.25E-04 0.0074
N/A N/A
2.01E+00 28.4
1.97E-01 3.82
N/A N/A
4.84E-01 9.40
N/A N/A
8.46E-02 1.64
7.65E-01 14.9
3.88E-01 5.96
N/A N/A
1.04E+00 16.0
4.39E-04 0.0067
1.28E-03 0.0197
1.43E-K)0 22.0
MANUFACTURING
Solder
manufacturing
Post-industrial
recycling
Total
2.66E-02 0.382
8.98E-02 1.29
1.16E-01 1.67
3.52E-02 0.500
1.01E-01 1.43
1.36E-01 1.93
2.42E-02 0.471
3.67E-02 0.712
6.09E-02 1.18
3.54E-02 0.543
l.OOE-01 1.54
1.36E-01 2.08
USE/APPLICATION
Reflow application
Total
5.09E+00 73.2
5.09E+00 73.2
4.80E+00 68.1
4.80E+00 68.1
3.97E+00 77.0
3.97E+00 77.0
4.82E+00 74.1
4.82E-K)0 74.1
END-OF-LIFE
Landfill
Incineration
Demanufacture
Cu smelting
Unregulated
Total
GRAND TOTAL
1.02E+00 14.7
2.81E-01 4.04
4.23E-04 0.0061
1.23E-02 0.177
1.30E-01 1.87
1.45E+00 20.8
6.96E+00 100
4.82E-04 0.0068
1.05E-02 0.149
3.66E-04 0.0052
1.06E-02 0.150
8.79E-02 1.25
1.10E-01 1.56
7.05E+00 100
1.43E-01 2.77
6.91E-02 1.34
4.31E-04 0.0084
N/A N/A
1.44E-01 2.80
3.56E-01 6.92
5.15E-HM) 100
4.81E-03 0.0739
1.43E-02 0.220
3.67E-04 0.0056
1.08E-02 0.166
9.01E-02 1.38
1.20E-01 1.85
6.51E-K)0 100
*The impact scores are in units of kilograms cancertox-equivalents/1,000 cc of solder applied to a printed wiring
board.
N/A=not applicable
In the manufacturing life-cycle stage, post-industrial recycling contributes more to total
impacts than solder manufacturing. For all alloys, post-industrial recycling contributes between
0.71 and 1.5 percent; and solder manufacturing contributes between 0.38 and 0.54 percent of
total impacts depending on the alloy.
Top Contributors to Public Health Impacts (Paste Solder)
Table 3-89 presents the specific materials or flows contributing greater than one percent
of public non-cancer impacts by solder. As presented above, the SnPb impacts are dominated by
the EOL stage. In particular, lead emissions to water, from both landfilling and incineration at
the EOL stage, constitute about 91 percent of total SnPb life-cycle impacts combined. For both
3-151
-------
of these processes, lead emissions to water occur from landfill leachate (e.g., from leaching of
waste electronics or incinerator ash). Sulphur dioxide emissions from electricity generation in
the use/application stage are the next greatest contributors to SnPb public non-cancer impacts at
about 3 percent, followed by lead emissions to air, water, and soil from unregulated recycling
and disposal which all contribute less than 2 percent.
While the SnPb public health non-cancer impacts are dominated by EOL lead emissions,
the lead-free alternatives are largely influenced by upstream metals production processes (e.g.,
silver, tin, and bismuth production) and electricity generation for reflow soldering. Specific
flows that contribute greatly to impact scores include the following: sulphur dioxide from silver
production (24 to 39 percent contribution); sulphur dioxide from electricity production for reflow
soldering (25 to 41 percent contribution); and lead emissions to soil from silver production (18 to
29 percent). Smaller contributors are lead emissions to air from silver production, arsenic
emissions to soil from silver production, and sulphur dioxide emissions from tin and bismuth
production.
Table 3-89. Top contributors to public non-cancer impacts (paste solder)
Solder
SnPb
SAC
BSA
SABC
Life-Cycle Stage
End-of-life
End-of-life
Use/application
End-of-life
End-of-life
End-of-life
Upstream
Upstream
Use/application
Upstream
Upstream
Upstream
Use/application
Upstream
Upstream
Upstream
Upstream
Upstream
Use/application
Upstream
Upstream
Upstream
Upstream
Process
Solder landfilling (SnPb)
Solder incineration (SnPb)
Electricity generation
Unregulated recycling and disposal (SnPb)
Unregulated recycling and disposal (SnPb)
Unregulated recycling and disposal (SnPb)
Silver production
Silver production
Electricity generation
Silver production
Tin production
Silver production
Electricity generation
Silver production
Silver production
Bismuth production
Tin production
Silver production
Electricity generation
Silver production
Silver production
Tin production
Silver production
Flow
Lead emissions to water
Lead emissions to water
Sulphur dioxide
Lead emissions to air
Lead emissions to soil
Lead emissions to water
Sulphur dioxide
Lead emissions to soil
Sulphur dioxide
Lead emissions to air
Sulphur dioxide
Arsenic emissions to soil
Sulphur dioxide
Sulphur dioxide
Lead emissions to soil
Sulphur dioxide
Sulphur dioxide
Lead emissions to air
Sulphur dioxide
Sulphur dioxide
Lead emissions to soil
Sulphur dioxide
Lead emissions to air
%
Contribution
72.6
18.8
3.19
1.67
1.67
1.12
38.7
28.5
25.0
2.05
1.85
1.11
41.2
24.3
17.9
9.56
1.99
1.29
33.7
33.5
24.6
2.50
1.78
As discussed in detail in Section 3.2.11.2 (Top Contributors to Occupational Impacts
section), human health impacts are derived from multiplying the inventory amount by the HV for
a particular material. Lead has a high non-cancer toxicity HV (62,400), indicating that emissions
of lead will have a higher non-cancer impact score than emissions of a less toxic substance when
3-152
-------
the output amount is the same. Further, lead has higher leachability than the other solder metals
as evidenced by TCLP testing conducted in support of the LFSP. For example, the fraction of
lead in SnPb that was found to leach is approximately 0.19, compared to a fraction of 0.000019
silver in SAC, and 0.000013 of copper in SAC (see Chapter 2 and Appendix C). These two
factors are responsible for the SnPb impacts at EOL being far greater than the impacts from the
other alloys.
The public non-cancer impact scores of the lead-free paste solders, on the other hand, are
dominated somewhat by sulfur dioxide emissions (HV=660), and to a lesser extent by lead
emissions from silver production. None of the lead-free solder metals themselves are top
contributors to public non-cancer impacts, even though silver, with the second highest HV of any
of the solder metals behind lead, has a relatively high HV of 10,000. This reveals that sulfur
dioxide, which has a lower HV than silver, has a greater inventory amount than silver, and the
metals in the lead-free solders are either not of high enough toxicity or enough quantity to be top
contributors to the total impacts. The relatively high percent contributions of lead emissions
from silver production to the total impacts of the lead-free solders are primarily due to lead's
high HV, rather than a large inventory amount.
Table 3-90 presents the specific materials or flows contributing at least 1 percent of
public cancer impacts by solder. Nitrogen oxides from electricity generation needed for reflow
application are the greatest contributors to public cancer impacts, ranging from 30 to 33 percent
contribution to total impacts depending on the solder. Methane from electricity generation in the
use/application stage also is a large contributor, ranging between about 14 and 15 percent. The
relatively high public cancer impact scores for nitrogen oxides and methane are primarily due to
their relatively large output flows from the extraction, processing, and consumption of fossil
fuels to generate electricity. Since no cancer toxicity data were available for either of these
materials, they were both assigned a default cancer HV of 1.
Table 3-90. Top contributors to public cancer impacts (paste solder)
Solder
SnPb
Life-Cycle Stage
Use/application
End-of-life
Use/application
Use/application
Use/application
End-of-life
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
Upstream
Use/application
Process
Electricity generation
Solder landfilling (SnPb)
Electricity generation
Electricity generation
Electricity generation
Solder incineration (SnPb)
SnPb (paste) reflow application
SnPb (paste) reflow application
SnPb (paste) reflow application
Tin production
Electricity generation
SnPb (paste) reflow application
Tin production
SnPb (paste) reflow application
Flow
Nitrogen oxides
Lead emissions to water
Methane to air
Carbon monoxide
Dust (unspecified) to air
Lead emissions to water
Flux material C *
Flux material F *
Flux material D *
Nitrogen oxides
NMVOC (unspecified) to air
Flux material E *
Dust (unspecified) to air
Flux material A *
%
Contribution
32.8
14.8
14.6
5.52
5.18
3.84
3.17
2.64
2.64
2.18
1.67
1.59
1.25
1.06
3-153
-------
Table 3-90. Top contributors to public cancer impacts (paste solder)
Solder
SAC
BSA
SABC
Life-Cycle Stage
Use/application
Use/application
Upstream
Use/application
Use/application
Upstream
Use/application
Upstream
Use/application
Use/application
Upstream
Upstream
Upstream
Use/application
Use/application
Upstream
Use/application
Use/application
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
Use/application
End-of-life
Upstream
Use/application
Use/application
Use/application
End-of-life
Upstream
Upstream
End-of-life
Use/application
Use/application
Upstream
Use/application
Use/application
Upstream
Use/application
Use/application
Use/application
Upstream
Upstream
Process
Electricity generation
Electricity generation
Silver production
Electricity generation
Electricity generation
Tin production
SAC (paste) reflow application
Silver production
SAC (paste) reflow application
SAC (paste) reflow application
Silver production
Silver production
Tin production
Electricity generation
SAC (paste) reflow application
Silver production
SAC (paste) reflow application
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Silver production
BSA (paste) reflow application
BSA (paste) reflow application
BSA (paste) reflow application
Solder landfilling (BSA)
Tin production
BSA (paste) reflow application
Electricity generation
BSA (paste) reflow application
Unregulated recycling and
disposal (BSA)
Tin production
Silver production
Solder incineration (BSA)
Electricity generation
Electricity generation
Silver production
Electricity generation
Electricity generation
Tin production
SABC (paste) reflow application
SABC (paste) reflow application
SABC (paste) reflow application
Tin production
Silver production
Flow
Nitrogen oxides
Methane to air
Dust (unspecified) to air
Carbon monoxide
Dust (unspecified) to air
Nitrogen oxides
Flux material C *
Arsenic emissions to soil
Flux material F *
Flux material D *
Methane to air
Nitrogen oxides
Dust (unspecified) to air
NMVOC (unspecified) to air
Flux material E *
Arsenic emissions to air
Flux material A *
Nitrogen oxides
Methane to air
Carbon monoxide
Dust (unspecified) to air
Dust (unspecified) to air
Flux material C *
Flux material F *
Flux material D *
Bismuth emissions to water
Nitrogen oxides
Flux material E *
NMVOC (unspecified) to air
Flux material A *
Bismuth emissions to air
Dust (unspecified) to air
Arsenic emissions to soil
Bismuth emissions to water
Nitrogen oxides
Methane to air
Dust (unspecified) to air
Carbon monoxide
Dust (unspecified) to air
Nitrogen oxides
Flux material C *
Flux material F *
Flux material D *
Dust (unspecified) to air
Arsenic emissions to soil
%
Contribution
30.4
13.6
11.9
5.12
4.81
3.15
3.02
2.82
2.52
2.52
2.41
2.03
1.81
1.55
1.51
1.36
1.01
32.4
14.4
5.45
5.12
4.84
4.35
3.63
3.63
2.58
2.20
2.18
1.65
1.45
1.44
1.26
1.15
1.04
O O '
JJ._
14.8
8.30
5.57
5.23
3.45
3.28
2.74
2.74
1.98
1.97
3-154
-------
Table 3-90. Top contributors to public cancer impacts (paste solder)
Solder
Life-Cycle Stage
Use/application
Upstream
Use/application
Upstream
Use/application
Process
Electricity generation
Silver production
SABC (paste) reflow application
Silver production
SABC (paste) reflow application
Flow
NMVOC (unspecified) to air
Methane to air
Flux material E *
Nitrogen oxides
Flux material A *
%
Contribution
.69
.68
.65
.42
.10
For the SnPb alloy, lead outputs from landfilling contribute 15 percent of the total public
cancer impact score for SnPb. The relatively high impact score for this flow is due to the fact
that lead was found to leach substantially more than metals in the other alloys. The remaining
top contributors for any of the alloys shown in Table 3-80 include several different flows, all of
which contribute approximately 12 percent or less. These include carbon monoxide, dust, flux
materials, arsenic, NMVOCs, and bismuth emissions. These emissions are from various
processes and life-cycle stages.
Of interest is that arsenic is the only top material contributor to the public cancer impacts
that is a known human carcinogen (cancer HV=29). The only other material that has been
classified by EPA or IARC as to carcinogenicity is lead, which is a "probable human
carcinogen." As discussed previously (Section 3.2.11.1), theLFSP LCIA methodology assigns
chemicals with a positive WOE classification, but no slope factor, a HV equal to 1, which is
representative of an average HV. The methodology also assigns chemicals with no cancer
toxicity data a HV of 1 to avoid the bias that typically favors chemicals with missing data. This
was the case with all of the other top contributors to solder paste public cancer impacts, which
were all assigned a HV of 1 due to missing data; therefore, much of the public cancer impacts are
driven by a lack of data, rather than known carcinogenic hazards. This is particularly true for the
lead-free alloys that are not affected by lead emissions. For SnPb, on the other hand, lead
outputs contribute about 18.6 percent to the total impacts (for landfilling and incineration
combined), and the lead HV is based on some carcinogenic rating, although the potential potency
of lead as a carcinogen is not known. SnPb is less driven by a lack of data than the lead-free
alloys; however, it is still highly driven by a lack of data given that all the remaining top
contributors, aside from lead emissions, have no applicable carcinogenic data.
3.2.12.3 Bar solder results
Total Public Health Impacts by Life-Cycle Stage (Bar Solder)
Table 3-91 presents the bar solder results for public human health non-cancer impacts by
life-cycle stage, based on the impact assessment methodology presented above. The table lists
the public non-cancer impact scores per functional unit for the life-cycle stages of each bar
solder alloy, as well as the percent contribution of each life-cycle stage to the total impacts.
Figure 3-31 presents the results in a stacked bar chart.
3-155
-------
Table 3-91. Public non-cancer impacts by life-cycle stage (bar solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
1.84E+02 0.138
5.28E+01 0.0394
4.55E+02 0.339
1.33E+05 99.5
1.34E+05 100
SAC
Score* %
1.16E+04 94.7
1.95E+01 0.160
4.65E+02 3.81
1.66E+02 1.36
1.22E+04 100
SnCu
Score* %
2.20E+02 30.0
2.65E+01 3.62
4.65E+02 63.4
2.16E+01 2.94
7.33E+02 100
*The impact scores are in units of kg noncancertox-equivalents/1,000 cc of solder applied to a printed wiring
board.
1 4n nnn -,
4-1
c
3
— 1 9n nnn
o
•&
° 1 nn nnn
,3
? Rn nnn
0)
ro
.>
3 fin nnn
9
X
C An nnn
01
o
c
CO
O9n nnn
o
c
o) n
ji U H
SnPb
SAC SnCu
D End-of-life
n Use/application
• Manufacturing
n Upstream
Figure 3-31. Bar Solder Total Life-Cycle Impacts: Public Non-Cancer
The public non-cancer impacts for SnPb (134,000 kg noncancertox-
equivalents/functional unit) are far greater than the other alloys (12,200 and 733 kg
noncancertox-equivalents/functional unit for SAC and SnCu, respectively). The EOL stage
dominates impacts for SnPb, contributing 99.5 percent to the total SnPb public non-cancer
impacts. The EOL impacts for the other alloys contribute only about 1 to 3 percent of total
impacts. EOL public non-cancer impacts are much greater for SnPb than the other solders due to
lead's high HV combined with its greater teachability as determined by TCLP testing (see
Chapter 2 and Appendix C), which was discussed above in the paste solder results (3.2.12.2).
For the lead-free alternatives, the upstream life-cycle stage is the greatest contributor to
overall public non-cancer impacts. SAC has the greatest upstream public non-cancer impacts at
11,600 kg noncancertox-equivalents/functional unit, which is 95 percent of total SAC public
non-cancer impacts. SnCu has the greatest proportion of its impacts from the wave soldering
use/application stage at 465 kg noncancertox-equivalents/functional unit or 63 percent
3-156
-------
contributions to total SnCu impacts. The upstream public non-cancer impacts for SnCu are 220
kg noncancertox-equivalents/functional unit or a contribution of 30 percent.
The use/application stage, which is made up of the wave soldering process group, is the
second greatest contributor for SnPb and SAC and the greatest contributor for SnCu. Impacts
from this life-cycle stage are associated with outputs from wave soldering (e.g., flux releases)
and from the generation of electricity used to melt the bar solder for wave application, and are
greatest for the alloys that consume the most energy during use. SAC and SnCu have slightly
greater impacts from the use/application stage, both at 465 kg noncancertox-
equivalents/functional unit, than does SnPb, 455 kg noncancertox-eqivalents/functional unit.
The percent contribution of the use/application stage to SnPb total impacts is relatively small (3
percent) compared to the lead-free alloys (about 26 to 42 percent for SAC and BSA,
respectively). This is due to lead's high HV which causes its impact scores at EOL to be much
greater than SnPb impact scores from solder reflow (e.g., from outputs from electricity
generation). Life-cycle public non-cancer impacts from the manufacturing stage are relatively
small for all of the bar solder alloys, ranging from 19.5 to 52.8 kg noncancertox-
equivalents/functional unit or about 0.04 to 4 percent of total impacts.
Table 3-92 presents the bar solder results for public human health cancer impacts by life-
cycle stage, based on the impact assessment methodology presented above in Section 3.2.12.1.
The table lists the public cancer impact scores per functional unit for the life-cycle stages of each
bar solder alloy, as well as the percent contribution of each life-cycle stage to the total impacts.
Figure 3-32 presents the results in a stacked bar chart.
Table 3-92. Public cancer impacts by life-cycle stage (bar solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
2.88E-01 4.20
2.66E-01 3.87
3.41E+00 49.7
2.90E+00 42.3
6.87E+00 100
SAC
Score* %
2.95E+00 23.7
3.91E-01 3.15
8.08E+00 65.0
1.01E+00 8.13
1.24E+01 100
SnCu
Score* %
4.18E-01 4.20
4.32E-01 4.34
8.07E-01 81.0
1.04E+00 10.5
9.96E+00 100
*The impact scores are in units of kg cancertox-equivalents/1,000 cc of solder applied to a printed wiring board.
3-157
-------
14
+j
1 12
15
c
o
= m
•5 IU
1 8
c
^t
ro
> 6
._ 0
o-
?
S 4
•i
-------
noncancertox-equivalents/functional unit). These process groups only contribute approximately
1.3 and 2.7 percent to the total scores, respectively. SnCu has the lowest unregulated recycling
and disposal score because neither lead nor silver are in the alloy. The high toxicity values of
lead and silver cause the unregulated recycling and disposal impacts for SnPb and SAC to be
greater than those from SnCu.
For SAC, the silver production process in the upstream life-cycle stage is the process
group with the greatest contribution to public non-cancer impacts, accounting for 92 percent of
total impacts. The next greatest contributor within the upstream life-cycle stage for SAC is tin
production (2.2 percent contribution). For SnCu, as expected based on mass composition, tin
production is greatest (29 percent), followed by copper production (1.2 percent).
As noted previously, the wave solder application process is either the first or second
greatest contributor to lead-free solder public non-cancer impacts within all the life-cycle stages,
and there is only one process group within this life-cycle stage.
Table 3-93 also shows the contribution of two process groups—solder manufacturing
and post-industrial recycling—within the manufacturing stage, which contribute a small
proportion to the overall impacts for all of the solders. SAC and SnCu have similar impact
scores for solder manufacturing (9.7 and 9.6 kg noncancertox-equivalents/functional unit,
respectively), while the SnPb score is slightly lower (7.6 kg noncancertox-equivalents/functional
unit). For the post-industrial recycling process group, impacts are greatest for SnPb (45 kg
noncancertox-equivalents/functional unit), followed by SnCu and SAC (16.9 and 9.88 kg
noncancertox-equivalents/functional unit, respectively). In each case, post-industrial recycling
impacts are greater than solder manufacturing inputs which are driven by the post-industrial
recycling process, as well as the secondary metal content of each alloy.
Table 3-94 lists the public cancer impacts of each of the process groups in the life-cycle
of the bar solders. The impact scores from the use/application stage are predominately due to the
flux materials released during wave soldering. Other contributions are from potentially
carcinogenic outputs from electricity generation in the wave soldering process group.
EOL impacts arise from output flows of potentially carcinogenic materials released from
the various EOL processes. For all the bar alloys, unregulated recycling and disposal has the
highest EOL impact score, ranging from about 8 to 21 percent of total life-cycle impacts. For
SnPb, landfilling is the second greatest contributing process group (about 17 percent of the total
public cancer impact score). Incineration contributes 4 percent to the total SnPb public cancer
impact score. For SAC and SnCu, the other processes aside from unregulated recycling and
disposal contribute small proportions to the total impact scores.
3-159
-------
Table 3-93. Public non-cancer impacts by life-cycle stage and process group (bar solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
SnCu
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Total
1.29E+02 0.0964
5.52E+01 0.0412
N/A N/A
N/A N/A
1.84E+02 0.138
2.74E+02 2.24
N/A N/A
1.13E+04 92.4
9.13E+00 0.0749
1.16E-K)4 94.7
2.11E+02 28.8
N/A N/A
N/A N/A
8.96E+00 1.22
2.20E+02 30.3
MANUFACTURING
Solder manufacturing
Post-industrial recycling
Total
7.63E+00 0.0057
4.52E+01 0.0337
5.28E+01 0.0394
9.67E+00 0.0792
9.88E+00 0.0809
1.95E-K)! 0.160
9.58E+00 1.31
1.69E+01 2.31
2.65E-K)! 3.62
USE/APPLICATION
Solder application
Total
4.55E+02 0.339
4.55E+02 0.339
4.65E+02 3.81
4.65E-KJ2 3.81
4.65E+02 63.4
4.65E-KJ2 63.4
END-OF-LIFE
Landfill
Incineration
Demanufacture
Cu smelting
Unregulated
Total
GRAND TOTAL
7.10E+04 53.5
1.80E+04 13.4
3.12E-01 0.0002
5.50E+00 0.0041
4.37E+04 32.6
1.33E+05 99.5
1.34E+05 100
1.01E+00 0.0082
-6.70E-02 -0.0005
2.73E-01 0.0022
1.99E+00 0.0163
1.63E+02 1.34
1.66E+02 1.36
1.22E+04 100
2.32E-02 0.0032
-3.17E-01 -0.0433
2.71E-01 0.0370
1.97E+01 2.68
1.93E+00 0.264
2.16E-K)! 2.94
7.33E+02 100
*The impact scores are in units of kg noncancertox-equivalents/1,000 cc of solder applied to a printed wiring
board.
N/A=not applicable
Potential upstream impacts arise from outputs of potentially carcinogenic materials in the
extraction and processing of the various metals present in the alloys. With the bar solder alloys,
the public cancer impact scores from tin extraction and processing comprise between about 3.7
and 4.2 percent of the total; for SnPb, about 0.48 percent of impacts are from lead extraction and
processing; and for SnCu about 0.0086 percent for copper extraction and processing. For SAC,
silver production dominates upstream impacts, contributing about 19 percent to the total score.
Public cancer impacts from silver processing exceed impacts from tin processing in solders that
contain both metals, even though the silver content of the alloys is much less than the tin content.
As described in earlier sections, SAC is 95.5 percent tin and only 3.9 percent silver, yet its
impacts from silver production are greater than those from tin production. This indicates that
potential cancer impacts from silver extraction and processing outputs are disproportionately
high compared to the other solder metals.
In the manufacturing life-cycle stage of SnPb, post-industrial recycling contributes
slightly more to total impacts than does solder manufacturing. For the lead-free alloys, solder
manufacturing contributes more than does post-industrial recycling. SAC solder manufacturing
contributes 2.7 percent compared to SAC post-industrial recycling, which contributes 0.49
3-160
-------
percent. SnCu solder manufacturing contributes about 3.3 percent compared to about 1.1 percent
for post-industrial recycling. This is because there is more recycled metal content in SnPb than
for SAC and SnCu.
Table 3-94. Public cancer impacts by life-cycle stage and process
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
?roup (bar solder)
SnCu
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Total
2.55E-01 3.72
3.31E-02 0.483
N/A N/A
N/A N/A
2.88E-01 4.20
5.40E-01 4.35
N/A N/A
2.41E+00 19.3
8.76E-04 0.0070
2.95E+00 23.7
4.17E-01 4.19
N/A N/A
N/A N/A
8.60E-04 0.0086
4.18E-01 4.20
MANUFACTURING
Solder manufacturing
Post-industrial recycling
Total
1.14E-01 1.67
1.51E-01 2.20
2.66E-01 3.87
USE/APPLICATION
Solder application
Total
3.41E+00 49.7
3.41E+00 49.7
3.30E-01 2.66
6.08E-02 0.489
3.91E-01 3.15
8.08E+00 65.0
8.08E+00 65.0
3.28E-01 3.29
1.04E-01 1.05
4.32E-01 4.34
8.07E+00 81.0
8.07E+00 81.0
END-OF-LIFE
Landfill
Incineration
Demanufacture
Cu smelting
Unregulated
Total
GRAND TOTAL
1.15E+00 16.7
2.96E-01 4.31
4.70E-04 0.0068
1.37E-02 0.199
1.44E+00 21.0
2.90E+00 42.3
6.87E+00 100
5.41E-04 0.0044
1.12E-02 0.0903
4.11E-04 0.0033
1.19E-02 0.0958
9.87E-01 7.94
1.01E+00 8.13
1.24E+01 100
5.65E-04 0.0057
1.16E-02 0.117
4.08E-04 0.0041
1.18E-02 0.119
1.02E+00 10.2
1.04E+00 10.5
9.96E-K)0 100
*The impact scores are in units of kg cancertox-equivalents/1,000 cc of solder applied to a printed wiring board.
N/A=not applicable
Top Contributors to Public Health Impacts (Bar Solder)
Table 3-95 presents the specific materials or flows contributing greater than 1 percent of
public non-cancer impacts by bar solder. As presented above, the SnPb impacts are dominated
by the EOL stage. In particular, lead emissions to water, from both landfilling and incineration
at EOL constitute about 66 percent of the total SnPb life-cycle impacts combined. For both of
these processes, lead emissions to water occur from landfill leachate (i.e., from leaching of waste
electronics or incinerator ash). Lead emissions to air, soil, and water from unregulated recycling
and disposal are the next greatest contributors (12, 12, and 8.2 percent, respectively); these are
from direct releases to the environment.
3-161
-------
Table 3-95. Top contributors to public non-cancer impacts (bar solder)
Solder
SnPb
SAC
SnCu
Life-Cycle Stage
End-of-life
End-of-life
End-of-life
End-of-life
End-of-life
Upstream
Upstream
Use/application
Upstream
Upstream
Upstream
End-of-life
Use/application
Upstream
Manufacturing
End-of-life
End-of-life
Upstream
Manufacturing
Process
Solder landfilling (SnPb)
Solder incineration (SnPb)
Unregulated recycling and disposal (SnPb)
Unregulated recycling and disposal (SnPb)
Unregulated recycling and disposal (SnPb)
Silver production
Silver production
Electricity generation
Silver production
Tin production
Silver production
Unregulated recycling and disposal (SAC)
Electricity generation
Tin production
Electricity generation for post-industrial
recycling
Post-consumer copper smelting (SnCu)
Post-consumer copper smelting (SnCu)
Copper production
Electricity generation for solder
manufacturing
Flow
Lead to water
Lead to water
Lead to air
Lead to soil
Lead to water
Sulphur dioxide to air
Lead to soil
Sulphur dioxide to air
Lead to air
Sulphur dioxide to air
Arsenic to soil
Silver to water
Sulphur dioxide to air
Sulphur dioxide to air
Sulphur dioxide to air
Copper to air
Copper to soil
Sulphur dioxide to air
Sulphur dioxide to air
% Contribution
53.3
13.1
12.3
12.3
8.19
49.6
36.5
3.68
2.63
2.24
1.43
1.32
61.9
29.1
2.19
1.20
1.20
1.18
1.09
While the SnPb public health non-cancer impacts are dominated by EOL lead emissions,
SAC is largely influenced by upstream metals production processes. Specific flows that
contribute greatly to the SAC impact scores include sulphur dioxide from silver production (50
percent contribution) and lead emissions to soil from silver production (about 37 percent).
Smaller contributors are sulphur dioxide from electricity generation for wave soldering, lead
emissions to air from silver production, sulphur dioxide emissions from tin production, arsenic
emissions to soil from silver production, and silver emissions to water from unregulated
recycling and disposal.
Sulphur dioxide emissions from electricity generation in the use/application stage are the
top contributor to SnCu public non-cancer impacts at about 62 percent, followed by sulphur
dioxide emissions to air from tin production at 29 percent. Other top contributors are sulphur
dioxide and copper to air from various processes in a mix of life-cycle stages.
As discussed in detail in Section 3.2.11.2 in the Top Contributors to Public Health
Impacts, human health impacts are derived from multiplying the inventory amount by the HV for
a particular material. Lead has a high non-cancer toxicity HV (about 62,400), indicating that
emissions of lead will have a higher non-cancer impact score than emissions of a less toxic
substance when the output amount is the same. Lead has higher teachability than the other
solder metals as well, as evidenced by TCLP testing conducted in support of the LFSP. For
example, a fraction of lead in SnPb that was found to leach is 0.19, compared to a fraction of
0.000019 silver in SAC, and a fraction of 0.000013 copper in SAC (see Chapter 2 and Appendix
C). These two factors are responsible for the SnPb impacts at EOL being far greater than the
3-162
-------
impacts from the other alloys.
The public non-cancer impact scores of the lead-free bar solders, on the other hand, are
dominated somewhat by sulphur dioxide emissions (HV=660), and to a lesser extent (for SAC),
by lead emissions from silver production. The results suggest that sulphur dioxide, which has a
lower HV than silver's 10,000 HV, has a greater inventory amount than silver, and that the
metals in the lead-free solders are either not of a high enough toxicity or not enough quantity to
exceed the impacts from sulphur dioxide. The relatively high percent contributions of lead
emissions from silver production to the total impacts of SAC are primarily due to lead's high
HV, rather than a large inventory amount.
Table 3-96 presents the specific materials or flows contributing at least 1 percent of
public cancer impacts by bar solder. The top contributor to public cancer impacts for each bar
alloy is flux from wave application in the use/application stage. For the SnPb bar solder alloy,
flux material "F" contributes approximately 26 percent of total public impact score. (Note,
letters are used in place of flux chemical names to protect confidentiality of companies that
supplied the data.) The second greatest contributor to SnPb impacts are lead outputs to water
from landfilling that contribute 17 percent to total public cancer impacts. The relatively high
impact score for this flow is due to the fact that lead was found to leach substantially more than
the metals in the other alloys (see Chapter 2 and Appendix C). Flux E from the wave application
is the third greatest contributor to SnPb cancer impacts at 15 percent. The remaining top
contributors for SnPb shown in Table 3-86 include several different flows, all of which
contribute approximately 7 percent or less. These include tin to water, nitrogen oxides to air,
lead to air and soil, methane to air, and dust to air. These emissions are from various processes
and life-cycle stages as shown in the table.
Table 3-96. Top contributors to public cancer impacts (bar solder)
Solder
SnPb
SAC
Life-Cycle Stage
Use/application
End-of-life
Use/application
End-of-life
Use/application
End-of-life
End-of-life
End-of-life
End-of-life
End-of-life
Use/application
Upstream
Upstream
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
End-of-life
Process
SnPb (bar) wave application
Solder landfilling (SnPb)
SnPb (bar) wave application
Unregulated recycling and disposal (SnPb)
Electricity generation
Unregulated recycling and disposal (SnPb)
Solder incineration (SnPb)
Unregulated recycling and disposal (SnPb)
Unregulated recycling and disposal (SnPb)
Unregulated recycling and disposal (SnPb)
Electricity generation
Tin production
Tin production
SAC (bar) wave application
SAC (bar) wave application
SAC (bar) wave application
Silver production
SAC (bar) wave application
SAC (bar) wave application
Unregulated recycling and disposal (SnAgCu)
Flow
Flux material F *
Lead to water
Flux material E *
Tin to air
Nitrogen oxides to air
Tin to water
Lead to water
Lead to air
Lead to soil
Lead to water
Methane to air
Nitrogen oxides to air
Dust (unspecified) to air
Flux material C *
Flux material F *
Flux material D *
Dust (unspecified) to air
Flux material E *
Flux material A *
Tin to air
%
Contribution
25.5
17.2
15.3
6.68
5.36
4.45
4.18
3.92
3.92
2.62
2.39
2.18
1.25
16.9
14.1
14.1
10.2
8.46
5.66
4.90
3-163
-------
Table 3-96. Top contributors to public cancer impacts (bar solder)
Solder
SnCu
Life-Cycle Stage
End-of-life
Use/application
Upstream
Upstream
Upstream
Use/application
Upstream
Upstream
Use/application
Upstream
Use/application
Use/application
Use/application
Use/application
Use/application
End-of-life
End-of-life
Use/application
Upstream
Use/application
Use/application
Upstream
Process
Unregulated recycling and disposal (SnAgCu)
Electricity generation
Tin production
Silver production
Silver production
SAC (bar) wave application
Silver production
Tin production
Electricity generation
Silver production
SnCu (bar) wave application
SnCu (bar) wave application
SnCu (bar) wave application
SnCu (bar) wave application
SnCu (bar) wave application
Unregulated recycling and disposal (SnCu)
Unregulated recycling and disposal (SnCu)
Electricity generation
Tin production
SnCu (bar) wave application
Electricity generation
Tin production
Flow
Tin to water
Nitrogen oxides to air
Nitrogen oxides to air
Arsenic to soil
Methane to air
Flux material B *
Nitrogen oxides to air
Dust (unspecified) to air
Methane to air
Arsenic to air
Flux material C *
Flux material D *
Flux material F *
Flux material E *
Flux material A *
Tin to air
Tin to water
Nitrogen oxides to air
Nitrogen oxides to air
Flux material B *
Methane to air
Dust (unspecified) to air
%
Contribution
3.27
3.00
2.56
2.42
2.07
1.94
1.74
1.47
1.34
1.17
21.3
17.7
17.7
10.7
7.08
6.38
4.25
3.78
2.49
2.43
1.68
1.43
* Flux names have been removed to protect confidentiality.
The top three contributors to bar SAC cancer impacts are three fluxes from wave
application that, when combined, constitute about 45 percent of the total public cancer impacts
for SAC. The fourth top contributor is dust from silver production (10 percent). The remaining
top contributors each contribute 8 percent or less. The top five contributors to bar SnCu cancer
impacts are fluxes from wave application, which combined constitute 74 percent of total impacts.
The remaining individual contributors contribute 6 percent or less, and are from various
processes and life-cycle stages as shown in the table.
Arsenic is the only top material contributor to the public cancer impacts that is a known
human carcinogen (cancer HV=29). The only other material that has been classified by EPA or
IARC as to carcinogenicity is lead, which is a probable human carcinogen. As discussed
previously, the LFSP LCIA methodology assigns chemicals with a positive WOE classification,
but no slope factor, a HV equal to 1, which is representative of an average HV. The
methodology also assigns chemicals with no cancer toxicity data a HV of 1 to avoid the bias that
typically favors chemicals with missing data. This was the case with all of the other top
contributors to solder paste public cancer impacts, which were all assigned a HV of 1 due to
missing data; therefore, much of the public cancer impacts are driven by a lack of data, rather
than known carcinogenic hazards. This is particularly true for the lead-free alloys, which are not
affected by lead emissions. For SnPb, on the other hand, of the top contributors in Table 3-96,
lead outputs contribute about 32 percent to the total impacts (for landfilling, incineration, and
unregulated recycling/disposal combined), and the lead HV is based on some carcinogenic
3-164
-------
rating, although the potential potency of lead as a carcinogen is not known; therefore, SnPb is
less driven by a lack of data than the lead-free alloys. It is still highly driven by a lack of data
given that all of the remaining top contributors, aside from lead emissions, have no applicable
carcinogenic data.
3.2.12.4 Limitations and uncertainties
This section summarizes the limitations and uncertainties associated with public non-
cancer and cancer health impacts. The public health LCIA limitations and uncertainties that
address (1) structural or modeling limitations and (2) toxicity data limitations, are identical to
those for occupational health impacts. For a detailed discussion, refer to Section 3.2.11.4. For
example, much of the public cancer impact results are driven by a lack of toxicity data, rather
than known carcinogenic hazards. In addition, the LCI data limitations for public health impacts
in many cases are similar to those described in Section 3.2.11.4. LCI data limitations pertinent
to public health impacts are summarized below.
For SnPb, the EOL impacts dominate non-cancer total impacts for both paste and bar
solder results, and cancer impacts also are somewhat influenced by EOL. The limitations and
uncertainties for SnPb are most influenced by the EOL uncertainties and limitations. Public
health impacts are based on process outputs as opposed to occupational impacts that are based on
process inputs. The EOL outputs have uncertainties associated with the inventory quantities as
they were based on assumptions about partitioning of the metals to various media, depending on
the EOL process. Details of the limitations and uncertainties for outputs from each of the EOL
processes are presented in Chapter 2, which provides limitations and uncertainties in the EOL
inventory.
To summarize, for landfilling there is relatively low uncertainty associated with the
teachability testing data used to calculate metal outputs from the landfill process, which are
primary data collected for the purposes of the LFSP. Uncertainties do exist and are associated
with (1) the TCLP test method itself and its representativeness of actual landfill conditions, and
(2) the analytical method (for example, limitations in analytical detection limits and quality
uncertainties associated with laboratory blanks). These limitations and uncertainties are
discussed in more detail in Chapter 2, which summarizes the teachability results, and in
Appendix C, which presents the teachability report. To address concerns that the TCLP test
method is not representative of actual landfill conditions (i.e., it overstates the teachability of
lead), a bounding analysis has been conducted that uses a lower bound of lead teachability to
help determine the sensitivity of the results to the teachability data (see Section 3.3).
For incineration, secondary literature was reviewed to make assumptions about metal
releases and partitioning to various environmental media. This introduced slightly more
uncertainty into the incineration outputs than is expected with the landfilling data. Uncertainties
associated with unregulated recycling and disposal are due to the almost complete absence of
analytical data on the partitioning of metals among environmental media for these processes.
EPA is currently conducting trials to assess metal emissions from open burning of electronics
waste. These data could be used later to reassess the assumptions used here for unregulated
recycling and disposal processes.
3-165
-------
Uncertainties from copper smelting have less effect on the results as this process
contributes small proportions to the total impacts. Nonetheless, uncertainties associated with
copper smelting arise from the inability of the researchers to get direct quantitative data from
primary data sources. Conversations with primary data suppliers and literature reviews, led to
assumptions that are believed to be reasonable to predict outputs; therefore, uncertainty is
considered to be acceptable for copper smelting outputs.
In addition to metal output uncertainties, there are EOL uncertainties related to the
assumptions about EOL dispositions (e.g., 72 percent of solder goes directly to landfilling for
SnPb, SAC, SABC, and SnCu). These are discussed in greater detail in Chapter 2.
Public health impacts of the lead-free alloys are generally dominated by the upstream and
use/application life-cycle stages. The uncertainties associated with these stages affect the
uncertainties for these alloys more so than the EOL uncertainties discussed above. Upstream
uncertainties stem from the use of secondary data sources. Silver production, which accounts for
a large proportion of the total public non-cancer impacts for the silver-bearing solders, has
associated uncertainties that are described in Section 3.2.1.4. As presented in that section,
although the secondary silver data set are considered "good" by GaBi, an alternate silver
inventory (from DEAM) is used to assess the sensitivity of LCIA results to silver production
data (see Section 3.3).
The use/application limitations and uncertainties related to electricity generation for paste
reflow soldering outputs arise from two issues: (1) electricity generation outputs are based on
the amount of electrical power used in the reflow solder process that was determined based on
two primary data points for reflow energy covering a large range in energy, and (2) electricity
production data are from a secondary source. Electricity consumption in the use/application
stage is evaluated in a sensitivity analysis for paste results (see Section 3.3). For a more detailed
discussion, refer to Section 3.2.1.4. Uncertainties from electricity use during bar solder wave
application relate to the use of secondary electricity generation data, but the reflow energy
uncertainty mentioned above does not apply.
Other uncertainties related to wave and reflow application relate to the assumption that
all the flux materials, either in the paste or as applied during wave soldering, are volatilized and
released to the environment. Primary data were not available on the capture of these materials or
on the actual releases to the environment; therefore, the assumption that all the flux materials are
released into the environment is an upper bound estimate for flux emissions to air, and a source
of uncertainty in the application processes. This is mostly relevant to the human health cancer
impacts for bar solders, each of which have a flux as their highest top contributor to total cancer
impacts.
Given that the lead toxicity is such an important driver of public non-cancer impacts,
further investigation into the impact score results has been done. For non-cancer impacts, the
LCIA methodology employed in this study calculates HVs based on either inhalation or oral
NOAELs or LOAELs. For chemicals that do not have NOAELs, LOAELs are used as the basis
of the toxicity hazard value. If a chemical has both an inhalation and oral NOAEL, the toxicity
value that results in the higher toxicity is chosen. This is a simple screening methodology that
allows for many chemicals through various transport and exposure pathways to be considered in
an analysis. The disadvantage of such a screening method is that it is applied to a variety of
3-166
-------
chemicals with various potential exposure scenarios (as is the case in an LCA), and the actual
toxicity and exposure for any one particular chemical in a particular process may not be
accurately represented. This method simply identifies chemicals of concern based on the most
toxic exposure pathway for that chemical without regard to the specific pathway in a particular
process. The reason this method uses either inhalation or oral toxicity data is because it is far too
cumbersome to select a particular route of exposure for every chemical in every process in the
life-cycle analysis; however, given that the lead toxicity is such an important driver of public
non-cancer health impacts, further understanding and resolution of the data is warranted.
The lead non-cancer HV in the LCIA methodology employed in this study is based on an
inhalation LOAEL. Of the top contributors in Table 3-95 and Table 3-96, 93 percent of the
paste results and 75 percent of the bar results were from lead emissions to water. To identify
what the results might look like if an ora/NOAEL were used, an alternate analysis is presented
here. Note, however, that this is not consistent with the methodology employed throughout all
the life-cycle stages, which uses the most toxic NOAEL or LOAEL, regardless of the route of
exposure. While an oral NOAEL might represent a more accurate exposure pathway for most of
the EOL releases, it may not do so for other processes represented in the analysis. Because lead
released to water is a large proportion of impacts, it seems worthy to estimate the sensitivity of
the results to the inhalation NOAEL by conducting the analysis with an oral NOAEL for lead.
In the baseline case, the non-cancer HV for lead is 62,427, which is based on an
inhalation LOAEL of 0.011 mg/m3 (ATSDR, 1999), which is calculated to be equivalent to a
NOAEL of 0.0011 mg/m3. In the alternative case, the non-cancer HV is 10,000, based on an oral
NOAEL of 0.0015 mg/kg-day (ATSDR, 1999). The HVs are calculated using equations
presented in Section 3.2.11.1. Figures 3-33 and 3-34 show the comparative results for the non-
cancer impacts using different lead toxicity non-cancer HVs for both the paste and bar solders,
respectively. The results from the alternate analysis, which is based on the oral NOAEL, have
the same conclusions for both the paste and bar analyses as they had for the baseline analyses;
that is, SnPb remains the highest impact score, by a much smaller margin for the alternative
analysis compared to the baseline. For the paste results, SnPb impacts were 7.6 times greater
than SAC in the baseline case, and SnPb was 2.1 times greater than SAC in the alternative case.
In both cases, EOL remained the top contributor, but to a lesser degree in the alternate case. In
the baseline case, EOL contributed 96 percent to total impacts. With the oral-based HV, EOL
comprised 82 percent of the total impacts.
For bar solder results, SnPb was 10.9 times greater than SAC in the baseline case
(inhalation-based HV) and 2.7 times greater than SAC in the alternate case. In the baseline,
EOL contributed 99 percent to total impacts, and in the alternate case, EOL contributed 97
percent to total impacts.
It is important to reiterate that by changing only the lead HV, we are not being consistent
in how other chemicals are treated; therefore, this analysis should not be construed as a
reasonable analysis to replace the baseline analysis. It is simply conducted to determine how the
results are impacted given a change in only the lead non-cancer HV.
3-167
-------
Public non-cancer
9 OOE+04
SClClF+ClA
'E
3
— 7 OOE+04
c
o
c
3
•* c; nnF+DA
'3
"V 4 OOE+04
0
Q o nnF+nA
c
co
A 9 DDF+DA
o
c
^ 1 OOE+04
0 OOE+00
impacts (paste solder)
Baseline lead HV analysis
I 1 1
DSnPb
• SAC
DBSA
DSABC
Alternate lead HV analysis
Figure 3-33. Comparative lead HV analysis (paste solder)
Public non-cancer impacts (bar solder)
1 4F+05
4-1
'c
1 9F+n^
03
0
"o i op+OS
1
c « nF+n/i
1
3 R nF+n4
X
"C 4 np+n4
o
c
CO
V o nc4.n/i
0
O)
j£ n np+nn
1
,
rjSnPb
• SAC
QSnCu
Baseline lead HV analysis Alternate lead HV analysis
Figure 3-34. Comparative lead HV analysis (bar solder)
3-168
-------
3.2.13 Aquatic Ecotoxicity Impacts
3.2.13.1 Characterization
Ecotoxicity refers to effects of chemical outputs on non-human living organisms.
Impact categories could include both ecotoxicity impacts to aquatic and terrestrial ecosystems.
The method for calculating terrestrial toxicity, however, would be the same as for the chronic,
non-cancer public toxicity impacts described above, which are based on mammalian toxicity
data. As the relative ranking approach of the LCIA toxicity method does not modify the toxicity
data for different species or for fate and transport, both human and terrestrial LCIA impacts are
the same; therefore, only aquatic toxicity, which uses a different methodology, is presented
below.
Toxicity measures for fish are used to represent potential adverse effects to organisms
living in the aquatic environment from exposure to a toxic chemical. Impact scores are based on
the identity and amount of toxic chemicals as outputs to surface water. Impact characterization
is based on CHEMS-1 acute and chronic hazard values for fish (Swanson etal., 1997) combined
with the inventory amount. Both acute and chronic impacts comprise the aquatic ecotoxicity
term. The HVs for acute and chronic toxicity are based on LC50 (the lethal concentration to 50
percent of the exposed fish population) and NOEL (no-observed-effect level) (orNOEC [no-
observed-effect concentration]) toxicity data, respectively, mostly from toxicity tests in fathead
minnows (Pimephales promelas) (Swanson et a/., 1997). The acute fish HV is calculated by:
(HVFA\ =
FA'
where:
HVPA equals the hazard value for acute fish toxicity for chemical / (unitless);
LC50 equals the lethal concentration to 50 percent of the exposed fish population for
chemical /'; and
LC50 mean equals the geometric mean LC50 of available fish LC50 values in Appendix E
(24.6 mg/L).
The chronic fish HV is calculated by:
I/NOEL
(HVFC\ =
3-169
-------
where:
HVPC equals the hazard value for chronic fish toxicity for chemical /';
NOEL equals the no-observed-effect level for fish for chemical /'; and
NOEL mean equals the geometric mean NOEL of available fish NOEL values in
Appendix D (3.9 mg/L).
For chemicals that do not have chronic fish toxicity data available, but do have LC50 data,
the LC50 and the log Kow of the chemical are used to estimate the NOEL. Based on studies
comparing the LC50 to the NOEL (Kenega, 1982; Jones and Schultz, 1995, and Call et al, 1985)
as reported in Swanson et al. (1997), NOEL values for organic chemicals within a certain range
of log Kow values are calculated using the following continuous linear function:
For organics with 2 • log Kow < 5:
NOEL = LC50/(5.3 x log Kow - 6.6)
Organic chemicals with high log Kow values (i.e., greater than 5) are generally more toxic
to fish and are not expected to follow a continuous linear function with Kow, thus, they are
estimated directly from the LC50. In addition, inorganic chemicals are poorly fat soluble and
their fish toxicity does not correlate to log Kow. The NOEL values of the inorganic chemicals
were, therefore, also based on the fish LC50 values.
For inorganics or organics with log Kow • 5:
NOEL = 0.05 x (LC50)
For organics with log Kow<2, which are poorly fat soluble but assumed to have a higher
NOEL value than those with higher Kow values or than inorganics, the NOEL is estimated as
follows:
For organics with log Kow <2:
NOEL = 0.25 x (LC50)
Once the HVs are calculated, whether from NOEL data or estimated from the LC50 and
the Kow, the aquatic toxicity impact score is calculated as follows:
(ISAQ),
3-170
-------
where:
IS
'AQ
FA
HV,
HVPC
Ami
TC output,water
equals the impact score for aquatic ecotoxicity for chemical /' (kg aquatictox-
equivalent) per functional unit;
equals the hazard value for acute fish toxicity for chemical /' (unitless);
equals the hazard value for chronic fish toxicity for chemical /'; and,
equals the toxic inventory output amount of chemical / to water (kg) per
functional unit.
3.2.13.2 Paste solder results
Total Aquatic Ecotoxicity Impacts by Life-Cycle Stage (Paste Solder)
Table 3-97 presents the solder paste results for aquatic ecotoxicity impacts by life-cycle
stage, based on the impact assessment methodology presented above. The table lists the aquatic
ecotoxicity impact scores per functional unit for the life-cycle stages of each solder paste alloy,
as well as the percent contribution of each life-cycle stage to the total impacts. Figure 3-35
presents the results in a stacked bar chart.
Table 3-97. Aquatic ecotoxicity impacts by life-cycle stage (paste solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
1.07E-01 0.0084
1.61E-01 0.0126
1.49E+00 0.117
1.27E+03 99.9
1.27E+03 100
SAC
Score* %
1.85E+01 50.9
5.88E-02 0.162
1.40E+00 3.84
1.64E+01 45.1
3.64E+01 100
BSA
Score* %
5.96E+00 25.5
3.40E-02 0.145
1.09E+00 4.68
1.63E+01 69.7
2.34E+01 100
SABC
Score* %
1.19E+01 31.0
5.90E-02 0.153
1.40E+00 3.65
2.51E+01 65.2
3.85E+01 100
*The impact scores are in units of kilograms aquatictox-equivalents/1,000 cc of solder applied to a printed wiring
board.
3-171
-------
1,400
f 1,200
ro
SnPb
SAC
BSA
SABC
Figure 3-35. Solder Paste Total Life-Cycle Impacts: Aquatic Ecotoxicity
The total aquatic ecotoxicity impact score for SnPb (1,270 kg aquatictox-
equivalents/functional unit) is far greater than the other solder paste alloys. SABC has the next
greatest impact score (38.5 kg aquatictox-equivalents/functional unit), which is only slightly
greater than that of SAC (36.4 kg aquatictox-equivalents/functional unit). BSA has the lowest
aquatic ecotoxicity score of all the alloys (23.4 kg aquatictox-equivalents/functional unit).
The EOL stage accounts for nearly all of the SnPb impacts, contributing 99.9 percent to
the total aquatic ecotoxicity impact score; however, EOL only accounts for about 45 to 70
percent of total impacts for the lead-free solders. For these alloys, the upstream life-cycle stage
also is substantial contributor to total impacts (26 to 51 percent). SAC has the greatest upstream
aquatic ecotoxicity impact score at 18.5 kg aquatictox-equivalents/functional unit, which is
51 percent the of total SAC aquatic ecotoxicity impacts. SABC has an upstream aquatic
ecotoxicity impact score of 11.9 kg aquatictox-equivalents/functional unit, which contributes 31
percent of SABC's total impacts. BSA has a smaller upstream aquatic ecotoxicity impact score
of 5.96 kg aquatictox-equivalents/functional unit, which is 26 percent of BSA's total impacts.
The use/application stage, which is comprised of the reflow soldering process and the
associated generation of electricity, is the third greatest contributor for the lead-free alloys.
Their aquatic ecotoxicity impact scores from this stage are all relatively small and close to one
another in magnitude (1.09, 1.40, and 1.40 kg aquatictox-equivalents/functional unit for BSA,
SAC, and SABC, respectively). These scores represent between 3.7 and 4.7 percent of the totals.
Of note is that SnPb has a greater impact score for the use/application stage than the lead-free
alloys, but the SnPb use/application score only contributes 0.12 percent to SnPb total impacts.
This is due to SnPb's high impact score at EOL. Impacts from the manufacturing stage are
small, ranging from 0.013 to 0.16 kg aquatictox-equivalents/functional unit for SnPb and BSA,
respectively. The manufacturing impacts for each alloy are less than 0.2 percent of total impacts
3-172
-------
and only 0.01 percent of SnPb impacts.
A benchmark of aquatic ecotoxicity impacts from burning a 60-watt lightbulb is provided
here to help put the magnitude of the impacts into perspective. The difference between the SnPb
and SAC ecotoxicity results is 1,234 kg aquatictox-equivalents/functional unit. The ecotoxicity
impacts associated with burning a 60-watt bulb for one day is 2.48 kg aquatictox-equivalents and
for one year is 905 kg aquatictox-equivalents; therefore, the difference between the SnPb and
SAC results is equivalent to burning a 60-watt bulb for approximately 1 year and 4 months. On
the other hand, the difference between the SAC and BSA results is only 13 kg aquatictox-
equivalents/functional unit, which is equivalent to ecotoxicity impacts associated with burning a
60-watt bulb for about 5.2 days.
Aquatic Ecotoxicity Impacts by Process Group (Paste Solder)
Table 3-98 lists the aquatic ecotoxicity impacts of each of the process groups in the life-
cycle of the solders. Within the EOL stage of the SnPb life-cycle, landfilling is the greatest
contributor to total impacts (78 percent of total aquatic ecotoxicity impacts), followed by
incineration (20 percent), and unregulated recycling/disposal (1.2 percent). Copper smelting and
demanufacturing are small contributors to the total SnPb aquatic ecotoxicity impacts (0.0034 and
0.00001 percent, respectively).
When evaluating the lead-free alloys alone, unregulated recycling and disposal is the
greatest process group contributor to EOL impacts, with scores of 15.0, 14.8, and 22.7 kg
aquatictox-equivalents/functional unit for SAC, BSA, and SABC, respectively (which contribute
41 to 63 percent of the total life-cycle impacts depending on the alloy). The second greatest
contributor to EOL impacts for the lead-free solders is landfilling (accounting for 3 to 5 percent
of total impacts). For the lead-free alloys, unregulated recycling/disposal has far greater aquatic
ecotoxicity impacts than landfilling, despite there being more electronics that are presumed to go
to landfilling (72 percent) than unregulated disposal (4.5 percent). This is because only a small
fraction of each metal in the lead-free alloys (between 0.000013 and 0.024 for all metals) was
found to leach during the project's teachability testing (Chapter 2 and Appendix C), but some
12.5 percent (i.e., a fraction of 0.125) of solder metals sent to unregulated recycling and disposal
are assumed to be released directly to surface waters via surface water runoff from waste
electronics burn piles.
For the lead-free solders, the silver production process is the greatest contributor to
upstream aquatic ecotoxicity impacts, contributing 24 to 51 percent to total impacts. For SAC,
copper production is the next greatest contributor, followed by tin production, but these
contributions are small (0.01 percent or less each). For BSA, after silver production, bismuth
production is the next largest contributor at 1.82 percent, followed by tin production at 0.0016
percent contribution. The second greatest contributor for SABC also is bismuth production,
however the score is only 0.00645 kg aquatictox-equivalents/functional unit, or 0.015 percent of
the total aquatic ecotoxicity impacts. Tin and copper production contribute even less to total
impacts (less than 0.008 percent).
The use/application stage has only one process group contributing to that life-cycle stage:
solder reflow application; thus, no further discussion on the breakdown of this life-cycle stage is
3-173
-------
warranted. Although the manufacturing life-cycle stage contributes a small proportion to the
overall impacts, Table 3-98 shows the contribution of the two process groups—solder
manufacturing and post-industrial recycling—within the manufacturing stage. For all the alloys,
post-industrial recycling has a greater aquatic ecotoxicity impact score than the solder
manufacturing process group.
Table 3-98. Aquatic ecotoxicity impacts by life-cycle stage
and process group (paste solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
BSA
Score* %
SABC
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Bi production
Total
5.06E-04 0.00004
1.07E-01 0.0084
N/A N/A
N/A N/A
N/A N/A
1.07E-01 0.0084
7.41E-04 0.0020
N/A N/A
1.85E+01 50.9
3.80E-03 0.0104
N/A N/A
1.85E-K)! 50.9
3.79E-04 0.0016
N/A N/A
5.53E+00 23.6
N/A N/A
4.26E-01 1.82
5.96E+00 25.5
7.48E-04 0.0019
N/A N/A
1.19E+01 31.0
3.18E-03 0.0083
6.45E-03 0.0167
1.19E+01 31.0
MANUFACTURING
Solder
manufacturing
Post-industrial
recycling
Total
1.13E-02 0.0009
1.49E-01 0.0117
1.61E-01 0.0126
1.40E-02 0.0386
4.48E-02 0.123
5.88E-02 0.162
1.12E-02 0.0480
2.28E-02 0.0974
3.40E-02 0.145
1.41E-02 0.0366
4.49E-02 0.117
5.90E-02 0.153
USE/APPLICATION
Reflow application
Total
1.49E+00 0.117
1.49E+00 0.117
1.40E+00 3.84
1.40E-K)0 3.84
1.09E+00 4.68
1.09E+00 4.68
1.40E+00 3.65
1.40E+00 3.65
END-OF-LIFE
Landfill
Incineration
Demanufacturing
Cu smelting
Unregulated
Total
GRAND TOTAL
9.99E+02 78.3
2.59E+02 20.3
1.46E-04 0.00001
4.33E-02 0.0034
1.54E+01 1.20
1.27E+03 99.9
1.27E+03 100
1.05E+00 2.89
2.75E-01 0.757
1.27E-04 0.0003
5.03E-02 0.138
1.50E+01 41.3
1.64E+01 45.1
3.64E+01 100
1.19E+00 5.08
3.10E-01 1.33
1.49E-04 0.0006
N/A N/A
1.48E+01 63.3
1.63E+01 69.7
2.34E+01 100
1.60E+00 4.16
6.47E-01 1.68
1.27E-04 0.0003
1.21E-01 0.315
2.27E+01 59.0
2.51E+01 65.2
3.85E+01 100
*The impact scores are in units of kilograms aquatictox-equivalents/1,000 cc of solder applied to a printed wiring
board.
N/A=not applicable
Top Contributors to Aquatic Ecotoxicity Impacts (Paste Solder)
Table 3-99 presents the specific materials or flows contributing at least 1 percent of
aquatic ecotoxicity impacts by solder. As expected from the results presented above, the SnPb
impacts are dominated by the EOL stage. The aquatic ecotoxicity impacts are based on outputs
to water. It is expected that the top contributors are lead emissions to water, mostly from
landfilling, with a significant amount from incineration from leaching of incinerator ash disposed
3-174
-------
in landfills, and a smaller amount from unregulated recycling/disposal. Combined, lead
emissions from these three processes constitute about 99.8 percent of the total life-cycle impacts.
Lead emissions from landfilling alone are the largest contributor to SnPb impacts (78 percent).
Further, lead emissions from landfilling are responsible for the fact that SnPb life-cycle impacts
are far greater than those of the other alloys. This is partly a function of the higher teachability
of lead, compared to the teachability of the other metals. For example, the fraction of lead in the
SnPb alloy that was found to leach was approximately 0.19 (kg of Pb per kg of solder),
compared to the fractions of 0.000019 and 0.000013 of silver and copper, respectively, in SAC
(Chapter 2 and Appendix C).
Table 3-99. Top contributors to aquatic ecotoxicity impacts (paste solder)
Solder
SnPb
SAC
BSA
SABC
Life-Cycle Stage
End-of-life
End-of-life
End-of-life
Upstream
End-of-life
Use/application
End-of-life
Upstream
Upstream
End-of-life
End-of-life
Upstream
End-of-life
Use/application
End-of-life
End-of-life
Upstream
End-of-life
Use/application
End-of-life
Upstream
End-of-life
Upstream
End-of-life
Process
Solder landfilling (SnPb)
Solder incineration (SnPb)
Unregulated recycling and disposal
(SnPb)
Silver production
Unregulated recycling and disposal
(SAC)
Electricity generation
Solder landfilling (SAC)
Silver production
Silver production
Unregulated recycling and disposal
(SAC)
Unregulated recycling and disposal
(BSA)
Silver production
Solder landfilling (BSA)
Electricity generation
Solder incineration (BSA)
Unregulated recycling and disposal
(SABC)
Silver production
Unregulated recycling and disposal
(SABC)
Electricity generation
Solder landfilling (SACB)
Silver production
Solder landfilling (SABC)
Silver production
Solder incineration (SABC)
Flow
Lead emissions to water
Lead emissions to water
Lead emissions to water
Cadmium emissions to water
Silver emissions to water
Chlorine (dissolved) emissions
to water
Silver emissions to water
Lead emissions to water
Zinc emissions to water
Copper emissions to water
Silver emissions to water
Cadmium emissions to water
Silver emissions to water
Chlorine (dissolved) emissions
to water
Silver emissions to water
Silver emissions to water
Cadmium emissions to water
Copper emissions to water
Chlorine (dissolved) emissions
to water
Silver emissions to water
Lead emissions to water
Copper emissions to water
Zinc emissions to water
Silver emissions to water
%
Contribution
78.3
20.3
1.20
45.7
39.6
3.29
2.42
2.13
1.88
1.66
63.3
21.2
4.84
4.01
1.26
32.8
27.8
26.2
3.13
2.95
1.30
1.21
1.14
1.05
3-175
-------
Another contributing factor leading to lead driving impacts, in addition to the teachability
of lead, is that it has a relatively high aquatic toxicity measure (discussed below); however, lead
does not have the highest relative aquatic toxicity compared to the other metals as it did for
human health non-cancer toxicity.
Among the lead-free alloys, silver, cadmium, and copper emissions to water are the
greatest contributors to aquatic ecotoxicity impacts. For SAC, cadmium emissions from silver
production contribute 46 percent, and silver emissions from unregulated recycling and disposal
contribute 40 percent. The remaining flows—chlorine emissions from reflow application, silver
emissions from landfilling, lead and zinc emissions from silver production, and copper emissions
from unregulated recycling and disposal—all contribute under 4 percent each to the total SAC
ecotoxicity impacts.
For BSA, silver emissions from unregulated recycling and disposal contribute 63 percent,
and cadmium emissions from silver production contribute nearly 20 percent to total aquatic
ecotoxicity impacts. Silver emissions from landfilling, chlorine from electricity generation
during reflow application, and silver emissions from incineration each contribute less than 5
percent.
The three top contributors to the SABC impacts are cadmium emissions from silver
production (about 26 percent); and silver and copper emissions from unregulated recycling and
disposal (27 percent each). The remaining top flows—chlorine from electricity generation for
reflow application, silver and copper emissions from landfilling, lead and zinc emissions from
silver production, and silver emissions from incineration—each contribute less than 4 percent to
total impacts.
To help clarify the results, the aquatic ecotoxicity HVs for the top contributing flows are
listed below in descending order of hazard (HVs for all materials classified as potentially toxic
are presented in Appendix E):
Cadmium: 28,500
Silver: 10,050
Copper: 2,732
Lead: 976
Zinc: 382
Chlorine: 267
The HVs are relative values that rank the aquatic ecotoxicity potential of a chemical as
compared to the average toxicity of many chemicals. The HVs are multiplied by the inventory
output amounts for chemicals with potential aquatic ecotoxicity impacts to derive an impact
score. Of the top contributors documented in Table 3-99, cadmium has the highest aquatic
ecotoxicity HV, followed by silver. This helps explain why most impacts for the lead-free
alternatives are driven by silver (from EOL processes) and cadmium (from silver production).
For SnPb, on the other hand, the HV of lead is lower than cadmium and silver, however, the
EOL
3-176
-------
output flows of silver and cadmium, both of which are a result of the presence of silver in the
lead-free alloys, are not found in the SnPb inventory. Alternatively, the lead at EOL constitutes
nearly all the impacts for SnPb, which are far greater than the total impacts for any of the other
alloys.
3.2.13.3 Bar solder results
Total Aquatic Ecotoxicity Impacts by Life-Cycle Stage (Bar Solder)
Table 3-100 presents the bar solder results for aquatic ecotoxicity impacts by life-cycle
stage, based on the impact assessment methodology presented above. The table lists the aquatic
ecotoxicity impact scores per functional unit for the life-cycle stages of each bar solder alloy, as
well as the percent contribution of each life-cycle stage to the total impacts. Figure 3-36
presents the results in a stacked bar chart.
Table 3-100. Aquatic ecotoxicity impacts by life-cycle stage (bar solder)
Life-cycle stage
Upstream
Manufacturing
Use/application
End-of-life
Total
SnPb
Score* %
9.56E-02 0.0062
2.87E-01 0.0185
2.36E-01 0.0152
1.55E+03 99.96
1.55E+03 100
SAC
Score* %
2.75E+01 13.9
6.83E-02 0.0345
2.39E-01 0.120
1.70E+02 86.0
1.98E+02 100
SnCu
Score* %
7.03E-03 0.0808
6.99E-02 0.804
2.39E-01 2.74
8.38E+00 96.4
8.70E+00 100
*The impact scores are in units of kilograms of aquatictox-equivalents/1,000 cubic centimeters of solder applied to
a printed wiring board.
1800
4-1
'E 1 goo
"ra
c 1400
O !^uu
•&
o
c 19DD
* mnn
01
ro
> snn
3
o-
V 600
x DUU
s
0 400
ro
3
°" 9DD
O)
J£
SnPb
1 '
SAC
' 1
SnCu
D End-of-life
D Use/application
H Manufacturing
fj Upstream
Figure 3-36. Bar Solder Total Life-Cycle Impacts: Aquatic Ecotoxicity
3-177
-------
The total aquatic ecotoxicity impact score for SnPb (1,550 kg aquatictox-
equivalents/functional unit) is far greater than the other bar solder alloys. SAC has the next
greatest impact score (198 kg aquatictox-equivalents/functional unit), followed by SnCu with the
lowest of 8.7 kg aquatictox-equivalents/functional unit.
The EOL stage accounts for nearly all of SnPb impacts, contributing 99.96 percent to the
total aquatic ecotoxicity impact score. For the lead-free bar solder alternatives, the EOL stage is
also the vast majority (96 and 86 percent), although the absolute scores are far lower than that of
SnPb._For SAC, the upstream life-cycle stage contributes 14 percent to the total impacts.
The use/application stage is a small contributor to overall impacts for all three alloys,
although it varies in terms of the percent contribution. Nonetheless, the aquatic ecotoxicity
impact scores for all three alloys from this stage are all relatively small and close to one another
in magnitude (0.236, 0.239, and 0.239 kg aquatictox-equivalents/functional unit for SnPb, SAC,
and SnCu, respectively). Of note is that SnPb has a greater impact score for the use/application
stage than the lead-free alloys, but the SnPb score only contributes 0.12 percent to SnPb total
impacts. This is due to SnPb's high impact score at EOL. Impacts from the manufacturing
stage are small, as are upstream impacts from SnPb and SnCu (all less than 0.3 kg aquatictox-
equivalents/functional unit).
Aquatic Ecotoxicity Impacts by Process Group (Bar Solder)
Table 3-101 lists the aquatic ecotoxicity impacts of each of the process groups in the life-
cycle of the bar solders. Within the EOL stage of the SnPb life-cycle, landfilling is the greatest
contributor to total impacts (71 percent of total aquatic ecotoxicity impacts), followed by
incineration (18 percent), and unregulated recycling/disposal (11 percent). Copper smelting and
demanufacturing are small contributors to the total SnPb aquatic ecotoxicity impacts (0.0031 and
0.00001 percent, respectively).
When evaluating the lead-free alloys alone, unregulated recycling and disposal is the
greatest process group contributor to EOL impacts, with scores of 169 and 7.89 kg aquatictox-
equivalents/functional unit for SAC and SnCu, respectively (which contribute 85 and 91 percent
of the total life-cycle impacts, respectively). The second greatest contributor to EOL impacts for
the lead-free solders is landfilling (accounting for 0.6 or 4 percent of total impacts). For the
lead-free alloys, unregulated recycling/disposal has far greater aquatic ecotoxicity impacts than
landfilling, despite there being more electronics that are presumed to go to landfilling (72
percent) than unregulated disposal (4.5 percent). This is because only a small fraction of each
metal in the lead-free bar alloys (between 0.000013 and 0.000027 for all metals) was found to
leach during the project's teachability testing (Chapter 2 and Appendix C), but some 12.5
percent (i.e., a fraction of 0.125) of solder metals sent to unregulated recycling and disposal are
assumed to be released directly to surface waters via surface water runoff from waste electronics
burn piles.
Within the upstream life-cycle stage, silver production for SAC contributes nearly 14
percent while all the other metals production process groups are negligible contributors to the
overall aquatic ecotoxicity impacts for all alloys. The use/application stage has only one process
group contributing to that life-cycle stage: wave solder application. No further discussion on
the breakdown of this life-cycle stage is warranted. Although the manufacturing life-cycle stage
3-178
-------
contributes a very small proportion to the overall impacts, Table 3-101 shows the contribution of
the two process groups—solder manufacturing and post-industrial recycling—within the
manufacturing stage. For all the alloys, post-industrial recycling has a greater aquatic ecotoxicity
impact score than the solder manufacturing process group.
Table 3-101. Aquatic ecotoxicity impacts by life-cycle stage
and process group (bar solder)
Life-cycle stage
Process group
SnPb
Score* %
SAC
Score* %
SnCu
Score* %
UPSTREAM
Sn production
Pb production
Ag production
Cu production
Total
4.92E-04 0.00003
9.51E-02 0.0061
N/A N/A
N/A N/A
9.56E-02 0.0062
1.04E-03 0.0005
N/A N/A
2.75E+01 13.9
6.35E-03 0.0032
2.75E+01 13.9
8.04E-04 0.0092
N/A N/A
N/A N/A
6.23E-03 0.0716
7.03E-03 0.0808
MANUFACTURING
Solder manufacturing
Post-industrial recycling
Total
3.57E-02 0.0023
2.51E-01 0.0162
2.87E-01 0.0185
4.12E-02 0.0208
2.71E-02 0.0137
6.83E-02 0.0345
2.37E-02 0.272
4.63E-02 0.532
6.99E-02 0.804
USE/APPLICATION
Solder application
Total
2.36E-01 0.0152
2.36E-01 0.015
2.39E-01 0.1204
2.39E-01 0.1204
2.39E-01 2.7426
2.39E-01 2.74
END-OF-LIFE
Landfill
Incineration
Demanufacture
Cu smelting
Unregulated
Total
GRAND TOTAL
1.11E+03 71.4
2.73E+02 17.5
1.63E-04 0.00001
4.81E-02 0.0031
1.71E+02 11.0
1.55E+03 99.96
1.55E+03 100
1.18E+00 0.597
2.93E-01 0.148
1.42E-04 0.0001
1.57E-03 0.0008
1.69E+02 85.2
1.70E+02 86.0
1.98E+02 100
3.91E-01 4.49
9.70E-02 1.12
1.41E-04 0.0016
1.52E-03 0.0175
7.89E+00 90.7
8.38E-K)0 96.4
8.70E-K)0 100
*The impact scores are in units of kg aquatictox-equivalents/1,000 cubic centimeters of solder applied to a
printed wiring board.
N/A=not applicable
Top Contributors to Aquatic Ecotoxicity Impacts (Bar Solder)
Table 3-102 presents the specific materials or flows contributing at least 1 percent of
aquatic ecotoxicity impacts by solder. As expected from the results presented above, the SnPb
impacts are dominated by the EOL stage. The aquatic ecotoxicity impacts are based on outputs
to water. It is expected that the top contributors are lead emissions to water, mostly from
landfilling, with a significant amount from incineration (from leaching of incinerator ash
disposed in landfills), and a smaller amount from unregulated recycling/disposal. Lead
emissions from landfilling alone are the largest contributor to SnPb impacts (71 percent), further,
lead emissions from landfilling are responsible for the fact that SnPb life-cycle impacts are far
greater than those of the other alloys. This is partly a function of the higher teachability of lead
3-179
-------
compared to the leachability of the other metals. For example, the fraction of lead in the SnPb
alloy that was found to leach was approximately 0.19 (kg of Pb per kg of solder), compared to
the fractions of 0.000019 (kg of Pb per kg of solder) and 0.000013 (kg of Pb per kg of solder) of
silver and copper, respectively, in SAC (Chapter 2 and Appendix C).
Table 3-102. Top contributors to aquatic ecotoxicity impacts (bar solder)
Solder
SnPb
SAC
SnCu
Life-Cycle Stage
End-of-life
End-of-life
End-of-life
End-of-life
Upstream
End-of-life
End-of-life
End-of-life
Use/application
End-of-life
Process
Solder landfilling (SnPb)
Solder incineration (SnPb)
Unregulated recycling and disposal (SnPb)
Unregulated recycling and disposal (SAC)
Silver production
Unregulated recycling and disposal (SAC)
Unregulated recycling and disposal (SnCu)
Solder landfilling (SnCu)
Electricity generation
Solder incineration (SnCu)
Flow
Lead to water
Lead to water
Lead to water
Silver to water
Cadmium to water
Copper to water
Copper to water
Copper to water
Chlorine (dissolved) to
water
Copper to water
%
Contribution
71.4
17.6
11.0
81.8
12.5
3.42
90.4
4.49
2.35
1.12
Another contributing factor leading to lead driving impacts, in addition to the leachability
of lead, is that it has a relatively high aquatic toxicity measure (discussed below). Lead does not
have the highest relative aquatic toxicity compared to the other metals as it did for human health
non-cancer toxicity.
Among the lead-free bar alloys, silver, cadmium, copper, and chlorine emissions to water
are top contributors to aquatic ecotoxicity impacts. For SAC, silver emissions from unregulated
recycling and disposal contribute about 82 percent, cadmium emissions from silver production
contribute about 13 percent, and copper emissions from unregulated recycling and disposal
contribute 3 percent.
For SnCu, copper from unregulated recycling and disposal contributes the greatest at 90
percent. Copper emissions from landfilling and incineration, as well as chlorine from wave
application, each contribute less than 5 percent to the total aquatic ecotoxicity impact scores.
As described earlier in Section 3.2.13.2, the aquatic ecotoxicity HVs for the top
contributing flows for the bar solders are listed below in descending order of hazard (HVs for all
materials classified as potentially toxic are presented in Appendix E):
Cadmium: 28,500
Silver: 10,050
Copper: 2,732
Lead: 976
Chlorine: 267
To reiterate from previous sections, the HVs are relative values that rank the aquatic
ecotoxicity potential of a chemical as compared to the average toxicity of many chemicals. The
HVs are multiplied by the inventory output amounts for chemicals with potential aquatic
3-180
-------
ecotoxicity impacts to derive an impact score. Of the top contributors documented in
Table 3-102, cadmium has the highest aquatic ecotoxicity HV, followed by silver, and then
copper. This helps explain why most impacts for SAC are driven by silver from EOL processes,
cadmium from silver production, and copper from EOL processes. For SnCu, copper emissions
from EOL processes dominate impacts, and for SnPb, lead emissions dominate impacts. The
large impact score for SnPb also is a function of the higher teachability of lead, as discussed
above.
3.2.13.4 Limitations and uncertainties
The LCIA methodology for aquatic ecotoxicity impacts is subject to the same structural
or modeling limitations and toxicity data limitations discussed previously for the occupational
and public health impact categories. For a detailed discussion, refer to the Limitations and
Uncertainties subsection of Section 3.2.11.4. One important distinction is that more toxicity
data tend to be available for aquatic effects than for human carcinogenic effects, for example. Of
the 178 chemicals classified as potentially toxic in this LFSP LCA, 53 had outputs to water that
should be considered in the aquatic ecotoxicity impact category. Of these, 41 had aquatic
ecotoxicity data suitable for inclusion in the LCIA
The LCI data limitations also are similar to those described in preceding sections. For
SnPb, EOL processes dominate total impacts. As a result, the limitations and uncertainties for
SnPb are most influenced by the EOL limitations and uncertainties. Most of the SnPb impacts
are from outputs to water from landfilling or incineration processes as derived from teachability
testing associated with this project (see Appendix C). As primary data collected for the purposes
of the LFSP, the teachability data are considered to be of relatively low uncertainty; however,
further information about their limitations and uncertainties was presented in Section 3.2.12.4
and is applicable here.
The lead-free alloy results for both paste and bar solders, on the other hand, are more
influenced by limitations and uncertainties in the unregulated recycling/disposal inventory.
(Emissions from landfilling also are among the top contributors to lead-free impacts in some
cases and, thus, are subject to the limitations and uncertainties described for lead outputs from
landfilling.) Unregulated recycling/disposal uncertainties are greater than those associated with
landfill outputs due to the almost complete absence of analytical data on the partitioning of
metals among environmental media for unregulated recycling and disposal processes. Data from
EPA trials currently underway to assess metal emissions from open burning of electronics waste
could be used later to reassess the assumptions used in this LCA for unregulated recycling and
disposal processes.
For the other EOL processes, there also are uncertainties associated with the inventory
quantities as they were based on assumptions about partitioning of the metals to various media,
depending on the EOL process. For incineration, secondary literature was reviewed to make
assumptions about metal releases and partitioning to various environmental media. This
introduced slightly more uncertainty into the incineration outputs than is expected with the
landfilling data. Uncertainties from copper smelting and unregulated recycling/disposal have
less effect on the results as they both contribute small proportions to total impacts. Nonetheless,
uncertainties associated with copper smelting arise from the inability of the researchers to obtain
3-181
-------
direct quantitative data from primary data sources, as was discussed previously.
In addition to metal output uncertainties from landfilling and incineration, there are EOL
uncertainties related to the assumptions about EOL dispositions to each EOL process (e.g., 72
percent of solder goes directly to landfilling for SnPb, SAC, SABC, and SnCu). These are
discussed in greater detail in Chapter 2, limitations and uncertainties in the EOL inventory).
In addition to the EOL stage, the aquatic ecotoxicity impact scores of the silver-bearing
alloys are largely influenced by the upstream life-cycle stage. Upstream uncertainties have been
discussed in previous sections and relate to the fact that the data are from secondary data
sources. Silver production, which accounts for large amounts of the total aquatic ecotoxicity
impacts for most of the lead-free solders, has associated uncertainties that are described in
Section 3.2.1.4. As presented in that section, although the secondary silver data set from GaBi is
considered "good," it is addressed with an alternate analyses in Section 3.3.
The use/application stage has a relatively small influence on the results. Nonetheless, the
limitations and uncertainties related to electricity consumption and generation described
previously apply here.
3-182
-------
3.3 ALTERNATE ANALYSES
3.3.1 Reflow Application Energy Analysis
The energy requirements for the reflow application process are based on primary data
collected from two facilities where test runs were conducted (described in Section 2.4). The two
ovens in which these tests were performed represent different technologies resulting in a large
range in energy consumption rates due to the difference in the efficiencies of the ovens. In the
baseline analysis, an average energy consumption value from these two test runs was used in the
determination of the life-cycle impacts reported earlier in Chapter 3. Table 3-103 shows the
baseline energy consumption average and the low and high individual data points that were used
to calculate the average. The low estimates are either 27 or 35 percent lower than the baseline
and the high estimates are either 27 or 35 percent higher than the baseline.
Table 3-103. Energy estimates for the reflow ap
Alloy
SnPb
SAC
BSA
SABC
Baseline
energy*
115
124
82.4
124
Low
energy*
73.9
80.6
60.1
80.6
Percent change
from baseline
-35
-35
-27
-35
plication process
High
energy*
155
168
105
168
Percent change
from baseline
35
35
27
35
* Units are in kWh/kg of solder applied to a printed wiring board. (Note: This unit is different from the impact
results which are presented per unit volume of solder on a printed wiring board.)
For many of the impact categories evaluated, impacts from energy used in the
use/application life-cycle stage constituted a majority of impacts. For paste solder, nearly all of
the use/application energy consumption occurs during the reflow soldering process. Table 3-104
lists the impact categories, and the alloys within each category, for which a majority of the
impacts resulted from the energy consumed during reflow. The only categories in which none
of the alloys had a majority of their impacts from energy used during reflow application were
occupational non-cancer, occupational cancer, public non-cancer, and aquatic ecotoxicity.
The analyses determine the sensitivity of the baseline impact results to the selection of a
value for the energy used during reflow. To demonstrate the sensitivity, results of the baseline
analysis were re-evaluated using the range of energy consumption values shown in Table 3-103
for the energy use impact category only. This category was selected as an example of the
potential sensitivity because a large percentage (between about 81 and 92 percent) of the of the
baseline impacts in this category for all four alloys resulted from the energy consumed during
reflow.
3-183
-------
Table 3-104. Impact categories and alloys with majority of impacts from energy
used in reflow application of paste solders
Impact Category
Non-renewable resource use
Renewable resource use
Energy use
Landfill space use
Global warming
Ozone depletion
Photochemical smog
Air acidification
Air paniculate matter
Water eutrophication
Water quality
Public human health — cancer
Alloy(s)
SnPb, SAC, BSA, SABC
SnPb, SAC, BSA, SABC
SnPb, SAC, BSA, SABC
SnPb
SnPb, SAC, BSA, SABC
SnPb, SAC, BSA, SABC
SnPb, BSA, SABC
SnPb, BSA, SABC
SnPb
SnPb, SAC, BSA, SABC
SnPb, SAC, BSA, SABC
SnPb, SAC, BSA, SABC
When the low and high energy data points are used to generate life-cycle impact results
for each type of solder paste, the magnitude of the impact scores change; however, the relative
comparison among alloys remains the same. As shown in Figure 3-37, for all three scenarios
(low energy, baseline, and high energy), SAC has the highest impacts, followed by SABC, SnPb,
and finally BSA.
When considering the contributions of individual life-cycle stages to the energy use
impact category (Section 3.2.2), the portion of the total life-cycle energy use impacts attributable
to the energy use during the use/application stage remain substantial, even when the low energy
data are used. This is illustrated in Table 3-105, which shows the percent contribution of the
use/application stage for the low energy, the baseline average, and the high energy data. The
table shows that even using the low energy values (i.e., a 27 to 35 percent decrease in energy use
in reflow application depending on the alloy), the energy impact results remain driven by the
use/application stage (73 to 88 percent) compared to the baseline where 82 to 91 percent of
impacts are from the use/application stage.
Although only the energy use impact category was re-evaluated using the alternate data,
it is not necessary to re-evaluate the other impact categories. None of the other categories had a
higher percentage of their impacts attributable to the reflow energy consumption as the energy
use impact category and are unlikely to be as affected by a change in the reflow data. Overall,
3-184
-------
the analyses suggest that the relative results between solders and the overall conclusions of the
study are not too sensitive to the variations in the reflow energy data (assuming the range used in
this sensitivity analysis represents a true or realistic range of the energy estimates for reflow
applications process).
Low energy
High energy
Figure 3-37. Sensitivity Analysis of Energy Consumption
During Reflow Solder Application
Table 3-105. Use/application energy sensitivity analysis:
percent contribution of use/application stage to energy impacts
Energy estimate
Low energy
Baseline
High energy
Percent Contribution
SnPb
88.2
91.2
94.0
SAC
73.2
78.9
85.1
BSA
83.1
85.8
89.5
SABC
76.8
82.0
87.4
3-185
-------
3.3.2 Alternate Silver Inventory Analysis
Upstream silver production was the greatest contributing process group for many of the
impact categories of the lead-free solder pastes in the baseline LCA. For SAC, six impact
categories were dominated by the silver production process, including landfill space use,
photochemical smog, air acidification, air particulates, public non-cancer, and aquatic
ecotoxicity (presented in Table 3-120). For BSA, the landfill space use impact category had
silver production as the top contributing process group; and for SABC, the landfill space use and
the air particulate matter impact categories had silver production as the top contributing process
group (see
Tables 3-121 and 3-122). As expected, SAC is more influenced by the silver production process
group than the other alloys because of its greater silver content. In addition, the silver process
contributed significantly to many other categories for each of the alloys, though it may not have
been the dominant contributor.
Due to the large influence that silver production had on many of the impact categories, an
alternate analysis to the baseline was performed by substituting an alternate silver data set
(DEAM) for the GaBi silver mix data set used to calculate the baseline results. For a discussion
of the GaBi data set and an explanation of why that data set was used for the baseline, please
refer to Section 2.2. Tables 3-106 and 3-108 show the results of the alternate analyses for paste
and bar solders respectively, as compared to the baseline. In the tables, bold entries indicate the
highest impact score (i.e., the greatest environmental impacts) among the alloys within each
impact category, while the shaded entries indicate the lowest impact score among alloys within
each category.
The results of the alternate analysis are dramatic and can be readily observed in
Table 3-123, which compares the baseline results for paste solders with those developed using
the alternate DEAM silver data set. For the baseline analysis, SnPb had the highest impacts in
six impact categories while SAC had the higher impacts in the remaining ten categories. Neither
BSA nor SABC had impacts that were the highest impact score in any category; however, when
results were generated using the DEAM data set, SnPb had the highest impacts in fourteen of the
sixteen impact categories, with SAC (particulate matter) and BSA (NRRuse) leading in one
category each. In many cases, SAC was only slightly less than SnPb, and most likely within the
error range of the data. Nonetheless, the analysis resulted in a noticeable change in relative
results between SnPb and SAC. Likewise, SnPb had the lowest impact scores—indicating it was
the best performer of the alloys evaluated—in five impact categories using the GaBi mixed silver
data set, but did not register the lowest score in any impact category during the alternate
analysis. BSA accounted for the lowest impact score in fifteen of the sixteen impact categories.
These results indicate the high sensitivity of the overall life-cycle results for paste solders to the
silver data set, and suggest that additional effort to further resolve the silver mining and
extraction data would be well spent.
A comparison of the baseline and alternate analyses for bar solders is shown in
Table 3-109. For the baseline analysis using the GaBi data set, SAC had highest life-cycle
impacts in twelve impact categories while SnPb had highest impacts in the remaining four
categories; however, results from the alternate analysis indicate that SAC had highest impacts in
3-186
-------
only seven impact categories and SnPb had highest impacts in nine impact categories. This is
not as dramatic a change as was seen with the paste results; however, several impact-specific
conclusions were altered. In addition, while SAC was not the lowest score for any impact
categories in the baseline, it was the lowest in five impact categories in the alternate analysis.
Again, this shows the importance of the silver inventory on results and the variability among
different silver production data sets. The baseline is expected to be of good quality and is
believed to be of greater quality than the BEAM data, but regardless of the relative quality of
each data set, these results show the possible variability and sensitivity of the results to the silver
inventory data.
3-187
-------
4*
o
tsi
4*
«
=
w
S
•a
o
a
CO
ro
4>
•53
_o
o
= *
a.
-g
'a
W W p.
rn oo Tt ^j J-H
•-! rn .-! ^ oo
fN in ^H _
9999
+ rTl rTl +
W
Q 9 9 9 ?
W W W f5
oo
oo
ON
IH ts os
W W
—! ts' os ^°
? g ?
o o o o
««^s
o o
in ^4 ^4
n
•—i ITS
9 9
W M
^ o
• ^o
^H m
O O
+ +
W W
OO Tt
•n oq
t-' t-'
o o
W W
^H O
O O
W W
"* O
? ?
u u
TT O
r-; oo
r~~' oo'
3 .e
w
bo bo
5-.6
|3
o-KJI
O O "S co
bo bo bo bo
4
X
o
bo bo bo
O
«
i^. (D
€ 1
8 |
I S-
bo bo
O •—i
W W
w w
t- OS
^o t-
W W
•* OS
"*. ^
OS
ve
0 ^H
W W
•n oq
vo m'
O ^H
w w
in •*
in ts'
o •—i
W W
ITS •*
o vq
W W
^o t~-
OS fs(
V 1 V
bo bo
11
^
i-H ^ rt
|§
•I S
2 a
Shaded
3-188
-------
Table 3-107. Comparison of baseline and alternate LCA analysis (paste solders)
SnPb
SAC
BSA
SABC
Solder
Alloy
Baseline
High
6
10
0
0
Low
5
0
11
0
Alternate
High
14
1
1
0
Low
0
1
15
0
Table 3-108. Alternative silver production analysis (bar solders)
Impact Category
NRRuse
RRuse
Energy use
Landfill
Global warming
Ozone depletion
Photochemical smog
Acidification
Particulate matter
Eutrophication
Water quality
Occ non-cancer
Occ cancer
Public non-cancer
Public cancer
Aquatic toxicity
unit
per functional unit*
kg
kg
MJ
m3
kg CO2-Equiv.
kgCFC-11-equiv.
kg ethene-equiv.
kg SO2-equiv.
kg
kg phosphate-equiv.
kg
kg noncancertox-
equiv.
kg cancertox-equiv.
kg noncancertox-
equiv.
kg cancertox-equiv.
kg aquatictox-equiv.
Baseline
SnPb SAC SnCu
3.15E+02 7.68E+02 3.12E+02
6.03E+03 8.76E+03 5.83E+03
2.91E+03 5.77E+03 3.40E+03
1.34E-03 2.14E-02 1.33E-03
1.87E+02 3.57E+02 2.16E+02
1.87E-05 4.13E-05 1.78E-05
6.98E-02 5.51E-01 7.06E-02
1.43E+00 1.10E+01 1.53E+00
1.49E-01 1.47E+00 1.99E-01
2.14E-02 2.57E-02 2.06E-02
3.98E-02 1.20E-01 3.64E-02
7.15E+05 1.09E+04 6.53E+01
5.94E+01 5.75E+01 5.49E+01
1.34E+05 1.22E+04 7.33E+02
6.87E+00 1.24E+01 9.96E+00
1.55E+03 1.98E+02 8.70E+00
Alternate silver process
SnPb SAC SnCu
3.15E+02 3.29E+02 3.12E+02
6.03E+03 5.75E+03 5.83E+03
2.91E+03 4.04E+03 3.32E+03
1.34E-03 1.31E-03 1.33E-03
1.87E+02 2.71E+02 2.16E+02
1.87E-05 1.71E-05 1.78E-05
6.98E-02 7.88E-02 7.06E-02
1.43E+00 1.81E+00 1.53E+00
1.49E-01 2.78E-01 1.99E-01
2.14E-02 2.02E-02 2.06E-02
3.98E-02 3.37E-02 3.64E-02
7.15E+05 1.39E+04 6.53E+01
5.94E+01 5.90E+01 5.49E+01
1.34E+05 1.01E+03 7.33E+02
6.87E+00 1.01E+02 9.96E+00
1.55E+03 1.71E+02 8.70E+00
*The functional unit is 1,000 cc of solder applied to a printed wiring board.
Notes: Bold impact scores indicate the alloy with the highest score for an impact category.
Shaded impact scores indicate the alloy with the lowest score for an impact category.
Table 3-109. Comparison of baseline and alternate LCA analysis (bar solders)
Solder
Alloy
SnPb
SAC
SnCu
Baseline
High
4
12
0
Low
6
0
10
Alternate
High
9
7
0
Low
6
5
5
3-189
-------
3.3.3 Alternate Leachate Analysis
The leachability study conducted for this project was used to estimate the outputs of
metals from landfilling PWB waste or residual metals in ash. Lead was found to leach to a much
greater extent than the other metals in the solders being analyzed in this study. These
leachability results contributed to the large public non-cancer and aquatic ecotoxicity impacts for
the SnPb as compared to the other alloys for both the paste and the bar solder results (see
Sections 3.2.12 and 3.2.13). Two major contributors to these high SnPb results were the high
leachability of lead and the fact that the lead has a very high relative toxicity. The TCLP
leachability study conducted to determine the landfilling outputs is based on standard EPA
TCLP test protocol using acetic acid, a substance known to readily leach lead. It is unknown to
what extent these test conditions represent actual landfill conditions, which can vary
dramatically over the lifetime of a landfill. It should be noted that only two impact categories
(public non-cancer and aquatic ecotoxicity) were largely influenced by the EOL landfilling
process, with the SnPb alloy particularly affected in both cases. To determine the sensitivity of
the results to the lead leachability data, this section presents the results of an alternate analysis
using the detection limit of lead as a lower bound of possible lead leachability during the TCLP
study.
For the alternate analysis, the measured fraction of lead detected in the leachate during
leachability testing of 0.19 (the baseline analysis) was replaced with the fraction of 0.000021
based on the TCLP detection limit for lead (0.01 Pb). The life-cycle impacts for both the public
non-cancer and the aquatic ecotoxicity categories were then recalculated.
Tables 3-110 and 3-111 present the paste and bar results, respectively, for both the
baseline analysis and the alternate lead leachate analysis. As shown in the tables, even with the
assumption that lead essentially does not leach (i.e., assuming the study detection limit for the
leachability of lead), the SnPb alloy impacts scores are still at least 2.5 times higher than the
score of the next closest alloy for public non-cancer impacts and a full order of magnitude higher
for aquatic ecotoxicity; however, the relative differences between SnPb and the lead-free alloys
are far less than in the baseline analysis.
Table 3-110. Alternative lead leachate analysis for selected impact
categories in the paste solder results
Impact
category
Public non-
cancer
Aquatic
ecotoxicity
Unit per
functional
unit (b)
kg
noncancertox-
equiv.
kg aquatictox-
equiv.
Baseline
SnPb SAC BSA SABC
8.80E-H)4 1.05E+04 5.01E+03 7.84E+03
1.27E403 3.64E+01 2.34E+01 3.85E+01
Alternate lead leachate data
SnPb SAC BSA SABC
2.41E+04 1.05e+04 5.01E+03 7.84E+03
2.76E-K)2 3.64E+01 2.34E+01 3.85E+01
(a) Impact categories selected are those that were highly impacted by the leachate data in the baseline analysis.
(b) The functional unit is 1,000 cc of solder on a printed wiring board.
Notes: Bold impact scores indicate the alloy with the highest score for an impact category.
Shaded impact scores indicate the alloy with the lowest score for an impact category.
3-190
-------
Table 3-111. Alternative lead leachate analysis for selected impact
categories in the bar solder results
Impact category
Public non-cancer
Aquatic
ecotoxicity
Unit per
functional unit (b)
kg noncancertox-
equiv.
kg aquatictox-
equiv.
Baseline
SnPb SAC SnCu
1.33E+05 1.22E+04 7.26E+02
1.55E+03 1.98E+02 8.70E+00
Alternate lead leachate data
SnPb SAC SnCu
6.23E+04 1.22E+04 7.26E+02
4.44E+02 1.98E+02 8.69E+00
(a) Impact categories selected are those that were highly impacted by the leachate data in the baseline analysis.
(b) The functional unit is 1,000 cc of solder on a printed wiring board.
Notes: Bold impact scores indicate the alloy with the highest score for an impact category.
Shaded impact scores indicate the alloy with the lowest score for an impact category.
These results are not completely unexpected given the high toxicity of lead compared to
the other metals. This analysis suggests that any elevation of the teachability data for SnPb due
to the aggressive nature of acetic acid towards the lead-based solder was unlikely to have
changed the overall impacts for SnPb relative to the other solders. The SnPb alloy would still
have the higher potential impacts for both public non-cancer and aquatic ecotoxicity than the
other solder alloys, based primarily on its relative toxicity.
3-191
-------
3.4 SUMMARY OF LIFE-CYCLE IMPACT ANALYSIS CHARACTERIZATION
AND RESULTS
This section presents an overview of the characterization methods and the life-cycle
impact results for the paste and bar solder alloys. Section 3.4.1 provides the equations for each
impact category that are used to calculate impact scores. Section 3.4.2 describes the LCIA data
sources and data quality. For both paste and bar solders, respectively, Sections 3.4.3 and 3.4.4
provide the total life-cycle impact category indicator scores for each alloy for each of the sixteen
impact categories evaluated in this study.
The LFSP LCIA methodology does not perform the optional LCIA steps of normalization
(calculating the magnitude of category indicator results relative to a reference value), grouping
(scoring and possibly ranking of indicators across categories), or weighting (converting indicator
results based on importance and possibly aggregating them across impact categories). Grouping
and weighting, in particular, are subjective steps that depend on the values of different
individuals, organizations, or societies performing the analysis. Since the LFSP involves a
variety of stakeholders from different geographic regions and with different values, these more
subjective steps were intentionally excluded from the LFSP LCIA methodology. Normalization
also was intentionally not included as there are not universally accepted normalization reference
values for all the impact categories included in this study. Furthermore, one of the primary
purposes of this research is to identify the relative differences in the potential impacts among
alloys, and normalization within impact categories would not affect the relative differences
among alloys within the impact categories.
Section 3.4.5 summarizes the limitations and uncertainties associated with the LCIA
methodology as well as the general limitations and uncertainties associated with the results.
3.4.1 Impact Score Equations
Table 3-112 summarizes the impact categories, associated impact score equations, and
the input or output data required for calculating natural resource impacts. Each of these
characterization equations are loading estimates. For a more detailed discussion of loading
estimates, refer to Section 3.1.
3-192
-------
Table 3-112. Summary of natural resources impact scoring
Impact category
Use of renewable
resources
Use/depletion of
non-renewable
resources
Energy use, general
energy consumption
Landfill space use
Impact score approach
ISRR= AmtRRxa-RC)
ISNRR= AmtNRRx(l-RC)
ISE = AmtE or (AmtF x H/D)
ISL = Amtw / D
Data required from inventory
(per functional unit)
Inputs
Material mass (kg)
(e.g., water)
Material mass (kg)
Energy (MJ)
(electricity, fuel)
None
Outputs
None
None
None
Mass of waste (hazardous and
solid waste combined) (kg) and
density (e.g., volume, m3)
Abbreviations: RC=recycled content; H=heat value of fuel /'; D=density of fuel /'.
The term abiotic ecosystem refers to the nonliving environment that supports living
systems. Table 3-113 presents the impact categories, impact score equations, and inventory data
requirements for abiotic environmental impacts to atmospheric resources.
Table 3-113. Summary of atmospheric resource impact scoring
Impact category
Global warming
Stratospheric ozone
depletion
Photochemical smog
Acidification
Air quality (paniculate
matter)
Impact score approach
ISGW = EFGWP x AmtoG
I^OD ~~ E-TODP x AmtoDc
J-Spocp "E-Tpocp x Amtpoc
ISAP = EFAp x AmtAC
ISPM = AmtpM
Data required from inventory
(per functional unit)
Inputs
None
None
None
None
None
Outputs
Amount of each greenhouse gas
chemical released to air
Amount of each ozone depleting
chemical released to air
Amount of each smog-creating
chemical released to air
Amount of each acidification chemical
released to air
Amount of particulates: PM10 or TSP
released to air a
a Assumes PM10 and TSP are equal; however, using TSP will overestimate PM10.
Table 3-114 presents the impact categories, impact score equations, and required
inventory data for abiotic environmental impacts to water resources.
3-193
-------
Table 3-114. Summary of water resource impact scoring
Impact category
Water eutrophication
Water quality (BOD and
TSS)
Water quality (TSS)
Impact score approach
l^EUTR ~~ ErEp X A.mtEC
ISWQ = AmtBOD + AmtBOD
I^TSS ~~ AmtTSS
Data required from inventory
(per functional unit)
Inputs
None
None
None
Outputs
Amount of each eutrophication chemical
released to water
Amount of BOD and suspended solids
(TSS) in each wastewater stream released to
surface water
Amount of suspended solids (TSS) in each
wastewater stream released to surface water
Table 3-115 summarizes the human health and ecotoxicity impact scoring approaches.
The impact categories, impact score equations, the type of inventory data, and the chemical
properties required to calculate impact scores are presented. The human health effects and
ecotoxicity impact scores are based on the scoring of inherent properties approach to
characterization. For a more detailed discussion of characterization methods, refer to
Section 3.1.
Table 3-115. Summary of human health and ecotoxicity impact scoring
Impact category
Chronic human
health effects —
occupational,
cancer
Chronic human
health effects —
occupational,
noncancer
Chronic human
health effects —
public, cancer
Chronic human
health effects —
public, noncancer
Aquatic
ecotoxicity
Impact score equations
IScHO-CA ~~ H • CA X AmtTcinput
IScHo-Nc — -HVNC x AmtTCinput
IScHp-cA — -HVCA x AmtTCoutput
IScHp-Nc — -HVNC x AmtTCoutput
ISAQ =(HVFA + HVFC)x
-^rntTCoutputiWater
Data required from inventory
(per functional unit)
Inputs
Mass of each
primary and
ancillary toxic
chemical
Mass of each
primary and
ancillary toxic
chemical
None
None
None
Outputs
None
None
Mass of each toxic
chemical released to air
and surface water
Mass of each toxic
chemical released to air
and surface water
Mass of each toxic
chemical released to
surface water
Chemical
properties data
required
WOE or SF
Mammal NOAEL
or LOAEL
WOE or SF
Mammal NOAEL
or LOAEL
Fish LC50 and/or
fish NOEL
3-194
-------
Individual impact scores are calculated for inventory items for a certain impact category
and can be aggregated by inventory item (e.g., a certain chemical), process, life-cycle stage, or
entire product profile. For example, global warming impacts can be calculated for one inventory
item (e.g., CO2 releases), for one process that could include contributions from several inventory
items (e.g., electricity generation), for a life-cycle stage that may consist of several process steps
(e.g., product manufacturing), or for an entire profile (e.g., a functional unit of a solder).
3.4.2 LCIA Data Sources and Data Quality
Data that are used to calculate impacts come from: (1) equivalency factors or other
parameters used to identify hazard values; and (2) LCI items. Equivalency factors and data used
to develop hazard values presented in this methodology include GWP, ODP, POCP, AP, EP,
WOE, SF, mammalian LOAEL/NOAEL, fish LC50, and fish NOEL. Published lists of the
chemical-specific parameter values exist for GWP, ODP, POCP, AP, and EP (see Appendix D).
The other parameters may exist for a large number of chemicals, and several data sources must
be searched to identify the appropriate parameter values. Priority is given to peer-reviewed
databases (e.g., Health Effects Assessment Summary Tables [HEAST], Integrated Risk
Information System [IRIS], Hazardous Substances Data Bank [HSDB]), next other databases
(e.g., Registry of Toxic Effects of Chemical Substances [RTECS]), then other studies or
literature, and finally estimation methods (e.g., structure-activity relationships [SARs] or
quantitative structure-activity relationships [QSARs]). The specific toxicity data that are used in
the LFSP are presented in Appendix E.
The sources of each parameter presented in this report and the basis for their values are
presented in Table 3-116. Data quality is affected by the data source itself, the type of data
source (e.g., primary versus secondary data), the currency of the data, and the accuracy and
precision of the data. The sources and quality of the LCI data used to calculate impact scores
were discussed in Chapter 2. Data sources and data quality for each impact category are
discussed further in Section 3.2, LCIA Results.
Table 3-116. Data sources for equivalency factors and hazard values
Parameter
Global warming potential
Ozone depletion potential
Photochemical oxidant creation
potential
Acidification potential
Nutrient enrichment/eutrophication
potential
Basis of parameter values
Atmospheric lifetimes and radiative forcing
compared to CO2
The change in the ozone column in the
equilibrium state of a substance compared to
CFC-11
Simulated trajectories of ozone production
with and without VOCs present compared to
ethene
Number of hydrogen ions that can
theoretically be formed per mass unit of the
pollutant being released compared to SO2
Ratio of N to P in the average composition of
algae (C106H263O110N1(iP) compared to
phosphate (PO43')
Source
IPCC, 2001
UNEP, 2003; WMO 1999
Heijungse/a/., 1992;
El, 1999
Heijungse/a/., 1992;
Hauschild and Wenzel,
1997
Heijungse/a/., 1992;
Lindforse/a/., 1995
3-195
-------
Table 3-116. Data sources for equivalency factors and hazard values
Parameter
Weight-of-evidence
Slope factor
Mammalian: LOAEL/NOAEL
Fish lethal concentration to 50
percent of the exposed population
(LC50)
Fish NOEL
Basis of parameter values
Classification of carcinogenicity by EPA or
IARC based on human and/or animal toxicity
data
Measure of an individual's excess risk or
increased likelihood of developing cancer if
exposed to a chemical, based on dose-response
data
Mammalian (primarily rodent) toxicity studies
Fish (primarily fathead minnow) toxicity
studies
Fish (primarily fathead minnow) toxicity
studies
Source
EPA, 1999; IARC, 1998
IRIS and HEAST as cited
in RAIS online database
IRIS, HEAST and various
literature sources provided
by EPA and/or UT
contractor
Various literature sources
and Ecotox database
Literature sources and
Ecotox database
3.4.3 Paste Solder Results Summary
The indicator results presented throughout the remainder of this section are the result of
the characterization step of the LCIA methodology where LCI results are converted to common
units and aggregated within an impact category. Results are expressed in units specific to an
individual impact category and, therefore, cannot be summed or compared across impact
categories.
Table 3-117 presents a summary of the paste solder results for each impact category
calculated using the impact assessment methodology presented in previous subsections of
Section 3.2. Impact scores shown in bold indicate the alloy with the highest impact score in an
impact category, while shaded scores indicate the alloy with the lowest impact score. SnPb has
the greatest impact category indicator in six impact categories, including eutrophication, RR use,
and four toxicity-related categories—public non-cancer, occupational non-cancer, occupational
cancer, and aquatic ecotoxicity. SAC has the highest impact category indicator in the remaining
ten impact categories: NRR use, energy use, landfill space use, global warming, ozone depletion,
photochemical smog, acidification, particulate matter, water quality, and public cancer. SnPb
has the lowest impact category indicator among the alloys in five impact categories: NRR use,
landfill space use, photochemical smog, acidification, and particulate matter. BSA has the
lowest indicators in the remaining eleven categories.
When evaluating the lead-free alternatives alone, without considering SnPb, BSA has the
lowest life-cycle impact score in all categories and SAC has the highest in all categories, except
aquatic ecotoxicity and occupational cancer, for which SABC has the highest impact scores.
Both impacts scores, however, are not much greater than those for SAC, and all the lead-free
alloys have substantially lower aquatic ecotoxicity impacts than SnPb. These scores only
indicate the relative or incremental differences among the alloys and do not necessarily indicate
any level of concern. The LCIA is not intended to quantify the significance of any particular
3-196
-------
impact score, but instead it shows the relative difference among the alloys within a particular
impact category; however, for some impact categories, especially the toxicity categories, results
are not necessarily linear. In other words, a score often does not mean potential impacts are ten
times worse than a score of one. Detailed discussions of the results of each impact category,
along with the associated uncertainties, are presented in Section 3.2.2.
Table 3-117. Paste solder LCIA results
Impact category
Non-renewable resource use
Renewable resource use
Energy use
Landfill space
Global warming
Ozone depletion
Photochemical Smog
Acidification
Paniculate matter
Eutrophication
Water quality
Occupational non-cancer
Occupational cancer
Public non-cancer
Public cancer
Aquatic ecotoxicity
Units per
functional unit*
kg
kg
MJ
m3
kg CO2-equiv.
kgCFC-11-equiv.
kg ethene-equiv.
kg SO2-equiv.
kg
kg phosphate-equiv.
kg
kg noncancertox-equiv.
kg cancertox-equiv.
kg noncancertox-equiv.
kg cancertox-equiv.
kg aquatictox-equiv.
Quality
rating**
M-H
M-H
H
M-H
H
L-M
M-H
M-H
M-H
H
H
M-H
L-M
M-H
L-M
M-H
SnPb
1.61E+03
3.48E+04
1.25E+04
2.75E-03
8.17E+02
9.95E-05
3.13E-01
6.50E+00
4.52E-01
1.22E-01
1.79E-01
5.60E+05
7.62E+01
8.80E+04
6.96E+00
1.27E+03
SAC
1.82E+03
3.47E+04
1.36E+04
1.62E-02
8.73E-K)2
1.10E-04
6.18E-01
1.25E-K)!
1.30E+00
1.18E-01
2.26E-01
8.12E+03
7.20E+01
1.05E+04
7.05E-H)0
3.64E+01
BSA
1.76E+03
2.64E+04
9.76E+03
6.57E-03
6.31E+02
7.98E-05
3.61E-01
7.32E+00
5.85E-01
9.06E-02
1.64E-01
2.34E+03
6.34E+01
5.01E+03
5.15E+00
2.34E+01
SABC
1.72E+03
3.41E+04
1.31E+04
1.13E-02
8.49E+02
1.04E-04
5.05E-01
1.03E+01
1.01E+00
1.17E-01
2.06E-01
5.25E+03
7.23E+01
7.84E+03
6.51E+00
3.85E+01
* The functional unit is 1,000 cc of solder applied to a printed wiring board.
** Quality rating summarizes the overall relative data quality associated with each impact category: high (H),
medium (M), or low (L). Further explanation is provided in section 3.2.1.3.
Notes: Bold impact scores indicate the alloy with the highest score for an impact category.
Shaded impact scores indicate the alloy with the lowest score for an impact category.
Table 3-118 summarizes the top contributing life-cycle stages for each alloy by impact
category. The life-cycle stage or stages that contribute fifty percent or more to impacts in each
impact category are listed in the table. In cases where an individual life-cycle stage did not
constitute a majority, the top stages that together exceed fifty percent are listed. In these cases,
the life-cycle stage listed first represents the one with a greater percentage of impacts attributable
to that impact category.
As shown in the table, the use/application life-cycle stage dominates much of the
impacts. For SnPb, thirteen out of sixteen impact categories have the majority of their impacts
from the use/application stage. The manufacturing stage dominates in one category:
occupational non-cancer, although it is not a majority by itself. The EOL stage is a top
contributor to occupational non-cancer and a majority for two other toxicity-related impact
categories, public non-cancer and aquatic ecotoxicity. The EOL impacts affected by outputs are
based on the metal constituents of the solders and not other materials in a PWB or the product
which houses the PWB; that is, outputs from incineration include only the solder metals and not
combustion products of the PWB itself. An analysis of an entire PWB assembly would likely
result in differing impacts than shown in this analysis.
3-197
-------
Table 3-118. Solder paste life-cycle stages contributing a majority of impacts
Impact category
Non-renewable resource use
Renewable resource use
Energy use
Landfill space use
Global warming
Ozone depletion
Photochemical smog
Air acidification
Air particulates
Water eutrophication
Water quality
Occupational health — non-cancer
Occupational health — cancer
Public human health — non-cancer
Public human health — cancer
Aquatic ecotoxicity
SnPb
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Manufacturing,
End-of-life
Use/application
End-of-life
Use/application
End-of-life
SAC
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
Upstream
Upstream
Upstream
Use/application
Use/application
Manufacturing,
End-of-life
Use/application
Upstream
Use/application
Upstream
BSA
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
End-of-life,
Use/application
Use/application
Upstream
Use/application
End-of-life
SABC
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
Manufacturing,
End-of-life
Use/application
Upstream
Use/application
End-of-life
For the lead-free alternatives, the upstream life-cycle stage plays a more important role
than it does for SnPb. SAC has nine impact categories where the use/application stage is the
majority contributor and six categories in which the upstream stage provides the majority of
impacts. Manufacturing and EOL are top contributors to only one impact category:
occupational non-cancer. The BSA impacts are driven by the use/application stage in eleven
categories, the upstream stage in three categories, and the EOL in two categories.
Manufacturing, along with EOL, contributes to the majority of impacts in the occupational non-
cancer impact category. The impact categories for SABC are driven by the same stages as BSA,
with the exception of the occupational non-cancer impact category. SABC occupational non-
cancer impacts are driven by the manufacturing and EOL stages, as is the case for SnPb and
SAC.
For all categories that are dominated by the use/application stage, except occupational
non-cancer, impacts are from the electricity generation for the reflow application process. For
occupational non-cancer, the use/application stage dominates from the actual reflow application
process. In most cases where the upstream stage dominates impacts in a category, it is silver
3-198
-------
production that is responsible for the high impacts, as is illustrated in the tables that follow. In
the manufacturing stage, which contributes significantly to occupational non-cancer for SnPb,
SAC, and SABC, it is the solder manufacturing process that is the source.
As stated in the previous sections, because the use/application stage is so dominant, a
sensitivity analysis of the use/application energy is provided in Section 3.3. Additionally,
alternative analyses are conducted with (1) alternative silver production process data, and (2) the
results of the less aggressive teachability study for EOL processes. These are also presented in
Section 3.3.
Table 3-119 through 3-122 list the top contributing flows and their associated processes
and life-cycle stages for each impact category for each of the solders. The tables show that for
each alloy nearly all impact categories are driven by a different flow. For example, in the SnPb
life-cycle, hard coal is the top contributor to energy impacts, sulphur dioxide is the top
contributor to photochemical smog, and COD is the top contributor to water eutrophication (e.g.,
nutrient enrichment).
There are some flows that are top contributors to more than one impact category. For
example, sulphur dioxide that drives photochemical smog and air acidification in the SnPb life-
cycle is from electricity generation associated with reflow application. In the lead-free solder
life-cycles, sulphur dioxide is the top contributor to three categories: photochemical smog, air
acidification, and public human health (non-cancer); however, in these cases, the sulphur dioxide
is from silver production in the upstream life-cycle stage, as opposed to electricity generation for
reflow application in the case of SnPb.
Another top flow in the SnPb life-cycle that contributes to more than one category is lead
emissions to water from landfilling. This is essentially the leachate from landfilling the SnPb
alloy. Lead emissions to water contribute 72.6 percent to the public health (non-cancer) impact
category and 78.3 percent to the aquatic ecotoxicity impact category.
In several instances, the top contributing individual flows comprise a large majority of
the total contribution to the alloy's life-cycle impacts within a category. For example, COD
constitutes 97.1 percent of the total water eutrophi cation impacts. As there are not a large
amount of chemicals for which there are eutrophi cation potentials, and the inventory in this
project only has a few water eutrophying chemicals, it is understandable that one material might
greatly dominate impacts. This is true for COD, despite its relatively low eutrophi cation
potential (see Appendix D).
Many top contributors constitute a majority of the total impacts within a category. In the
SnPb results, eleven of the sixteen impact categories had top flows representing a majority of
total impacts.
By contrast, for lead-free solders, only seven of the sixteen categories had flows
contributing fifty percent or more. For each alloy, however, they were not always the same
impact categories that contribute greater than fifty percent. For example, with aquatic
ecotoxicity, silver emissions to water from unregulated recycling/disposal of BSA (Table 3-121)
contribute sixty-three percent, while cadmium emissions to water from silver production for
SAC (Table 3-120) are only forty-six percent of total aquatic ecotoxicity impacts.
3-199
-------
Table 3-119. Top contributing flows to SnPb solder paste impacts
Impact category
Non-renewable resource use
Renewable resource use
Energy
Landfill space use
Global warming
Ozone depletion
Photochemical smog
Air acidification
Air particulates
Water eutrophication
Water quality
Occupational
health — non-cancer
Occupational
health — cancer
Public human
health — non-cancer
Public human health — cancer
Aquatic ecotoxicity
Life-cycle stage
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
End-of-life
Use/application
End-of-life
Process
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
SnPb reflow application
Electricity generation
Solder landfilling (SnPb)
Electricity generation
Solder landfilling (SnPb)
Flow
Inert rock
Water
Hard coal
(resource)
Sludge (hazardous
waste)
Carbon dioxide
CFC-114
Sulphur dioxide
Sulphur dioxide
Dust (unspecified)
Chemical oxygen
demand
Solids (suspended)
SnPb solder paste
Natural gas
Lead emissions to
water
Nitrogen oxides
Lead emissions to
water
%
Contrib.
76.8
88.8
46.8
64.8
87.7
39.3
65.1
65.4
79.1
97.1
86.9
31.2
43.2
72.6
32.8
78.3
5-200
-------
Table 3-120. Top contributing flows to SAC solder paste impacts
Impact category
Non-renewable resource
use
Renewable resource use
Energy
Landfill space use
Global warming
Ozone depletion
Photochemical smog
Air acidification
Air particulates
Water eutrophication
Water quality
Occupational
health — non-cancer
Occupational
health — cancer
Public human
health — non-cancer
Public human
health — cancer
Aquatic ecotoxicity
Life-cycle stage
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
Upstream
Upstream
Upstream
Use/application
Use/application
Use/application
Use/application
Upstream
Use/application
Upstream
Process
Electricity generation
Electricity generation
Electricity generation
Silver production
Electricity generation
Electricity generation
Silver production
Silver production
Silver production
Electricity generation
Electricity generation
SAC reflow
application
Electricity generation
Silver production
Electricity generation
Silver production
Flow
Inert rock
Water
Hard coal (resource)
Slag (hazardous
waste)
Carbon dioxide
CFC-114
Sulphur dioxide
Sulphur dioxide
Dust (unspecified)
Chemical oxygen
demand
Solids (suspended)
SAC solder paste
Natural gas
(resource)
Sulphur dioxide
Nitrogen oxides
Cadmium emissions
to water
%
Contrib.
64.1
83.7
40.5
77.8
77.1
33.4
47.9
49.5
63.9
94.1
64.7
31.5
43.0
38.7
30.4
45.7
5-201
-------
Table 3-121. Top contributing flows to BSA solder paste impacts
Impact category
Non-renewable resource
use
Renewable resource use
Energy
Landfill space use
Global warming
Ozone depletion
Photochemical smog
Air acidification
Air particulates
Water eutrophication
Water quality
Occupational
health — non-cancer
Occupational
health — cancer
Public human
health — non-cancer
Public human
health — cancer
Aquatic ecotoxicity
Life-cycle stage
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
End-of-life
Process
Electricity generation
Electricity generation
Electricity generation
Silver production
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Electricity generation
BSAreflow
application
Electricity generation
Electricity generation
Electricity generation
Unregulated recycling
and disposal (BSA)
Flow
Inert rock
Water
Hard coal
Slag (hazardous
waste)
Carbon dioxide
CFC-114
Sulphur dioxide
Sulphur dioxide
Dust (unspecified)
Chemical oxygen
demand
Solids (suspended)
BSA solder paste
Natural gas
(resource)
Sulphur dioxide
Nitrogen oxides
Silver emissions to
water
%
Contrib.
51.7
85.9
44.0
57.1
83.4
36.0
41.5
42.7
45.0
95.7
69.8
32.5
37.9
41.2
32.4
63.3
3-202
-------
Table 3-122. Top contributing flows to SABC solder paste impacts
Impact category
Non-renewable resource use
Renewable resource use
Energy
Landfill space use
Global warming
Ozone depletion
Photochemical smog
Air acidification
Air particulates
Water eutrophication
Water quality
Occupational
health — non-cancer
Occupational
health — cancer
Public human
health — non-cancer
Public human health — cancer
Aquatic ecotoxicity
Life-cycle stage
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
Use/application
Use/application
Use/application
Use/application
End-of-life
Process
Electricity generation
Electricity generation
Electricity generation
Silver production
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Silver production
Electricity generation
Electricity generation
SABC reflow
application
Electricity generation
Electricity generation
Electricity generation
Unregulated recycling
and disposal (SABC)
Flow
Inert rock
water
Hard coal
Slag (hazardous
waste)
Carbon dioxide
CFC-114
Sulphur dioxide
Sulphur dioxide
Dust
(unspecified)
Chemical oxygen
demand
Solids
(suspended)
SABC solder
paste
Natural gas
(resource)
Sulphur dioxide
Nitrogen oxides
Silver emissions
to water
%
Contrib.
67.9
85.5
42.0
71.3
79.6
34.5
38.1
39.0
53.2
95.1
71.2
31.5
42.9
33.7
33.1
32.8
3-203
-------
3.4.4 Bar Solder Results Summary
Table 3-123 presents a summary of the bar solder results for each impact category
calculated using the impact assessment methodology presented in previous subsections of
Section 3.2. Impact scores shown in bold indicate the alloy with the highest impact score in an
impact category, while shaded scores indicate the alloy with the lowest impact score. SnPb has
the greatest impact category indicator in four impact categories, all of which are toxicity-related
categories—public non-cancer, occupational non-cancer, occupational cancer, and aquatic
ecotoxicity. SAC has the highest impact category indicator in the remaining twelve impact
categories. SnPb has the lowest impact category indicator among the alloys in five impact
categories: energy use, global warming, photochemical smog, acidification, and particulate
matter. BSA has the lowest indicators in the remaining eleven categories.
When evaluating the lead-free alternatives alone, without considering SnPb, SAC has the
highest impact score in all sixteen of the categories evaluated. Conversely, SnCu had the lowest
indicator scores. These scores only indicate the relative or incremental differences among the
alloys and do not necessarily indicate any level of concern. The LCIA is not intended to
quantify the significance of any particular impact score, but instead it shows the relative
difference among the alloys within a particular impact category. Detailed discussions of the
results of each impact category, along with the associated uncertainties, are presented in Section
5.2.2.
Table 3-123. Bar solder LCIA results
Impact category
Non-renewable resource use
Renewable resource use
Energy use
Landfill space
Global warming
Ozone depletion
Photochemical smog
Acidification
Particulate matter
Eutrophication
Water quality
Occupational non-cancer
Occupational cancer
Public non-cancer
Public cancer
Aquatic ecotoxicity
Units per
functional unit*
kg
kg
MJ
m3
kg CO2-equiv.
kgCFC-11-equiv.
kg ethene-equiv.
kg SO2-equiv.
kg
kg phosphate-equiv.
kg
kg noncancertox-equiv.
kg cancertox-equiv.
kg noncancertox-equiv.
kg cancertox-equiv.
kg aquatictox-equiv.
Quality
rating**
M-H
M-H
H
M-H
H
L-M
M-H
M-H
M-H
H
H
M-H
L-M
M-H
L-M
M-H
SnPb
3.15E+02
6.03E+03
2.91E+03
1.34E-03
1.87E+02
1.87E-05
6.98E-02
1.43E+00
1.49E-01
2.14E-02
3.98E-02
7.15E+05
5.94E+01
1.33E+05
4.13E+00
1.55E+03
SAC
7.68E+02
8.76E+03
5.77E+03
2.14E-02
3.57E+02
4.13E-05
5.51E-01
1.10E+01
1.47E+00
2.57E-02
1.20E-01
1.09E+04
5.75E+01
1.22E+04
5.04E+00
1.98E+02
SnCu
3.12E+02
5.83E+03
3.40E+03
1.33E-03
2.16E+02
1.78E-05
7.06E-02
1.53E+00
1.99E-01
2.06E-02
3.64E-02
6.53E+01
5.49E+01
7.26E+02
2.58E+00
8.70E+00
* The functional unit is 1,000 cc of solder applied to a printed wiring board.
** Quality summarizes the overall relative data quality associated with each impact category: high (H), medium
(M), or low (L). Further explanation is provided in Section 3.2.1.3
Notes: Bold impact scores indicate the alloy with the highest score for an impact category.
Shaded impact scores indicate the alloy with the lowest score for an impact category.
Table 3-124 summarizes the top contributing life-cycle stages for each alloy by impact
category. The life-cycle stage or stages that contribute fifty percent or more to impacts in each
3-204
-------
impact category are listed in the table. In cases where an individual life-cycle stage did not
constitute a majority, the top stages that together exceed fifty percent are listed. In these cases,
the life-cycle stage listed first represents the one with a greater percentage of impacts attributable
to that impact category.
Table 3-124. Bar solder life-cycle stages contributing a majority of impacts
Impact category
Non-renewable resource use
Renewable resource use
Energy use
Landfill space use
Global warming
Ozone depletion
Photochemical smog
Air acidification
Air particulates
Water eutrophication
Water quality
Occupational
health — non-cancer
Occupational
health — cancer
Public human
health — non-cancer
Public human
health — cancer
Aquatic ecotoxicity
SnPb
Use/application
Use/application
Use/application
End-of-life
Use/application
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
End-of-life,
Manufacturing
Use/application,
Manufacturing
End-of-life
Use/application
End-of-life
SAC
Upstream
Use/application
Upstream
Upstream
Upstream
Upstream
Upstream
Upstream
Upstream
Use/application
Upstream
End-of-life,
Manufacturing
Use/applications,
Upstream
Upstream
Use/application
End-of-life
SnCu
Use/application
Use/application
Use/application
End-of-life
Use/application
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
Use/application,
Manufacturing
Use/application,
Manufacturing
Use/application
Use/application
End-of-life
As shown in the table, the use/application life-cycle stage dominates the impacts. For
SnPb, eleven of the sixteen impact categories are driven by contributions from the
use/application stage, with end-of-life processes dominating four other impact categories.
Similarly, the use/application stage is the major contributor to thirteen of the impact categories
for the SnCu alloy. Upstream and end-of-life processes contribute the majority of the impacts in
the remaining SnCu impact categories. The manufacturing stage dominates in one category:
occupational non-cancer, although it is not a majority by itself. The EOL impacts affected by
outputs are based on the metal constituents of the solders and not other materials in a PWB or the
3-205
-------
product which houses the PWB; that is, outputs from incineration include only the solder metals
and not combustion products of the PWB itself. An analysis of an entire PWB assembly would
likely result in differing impacts than shown in this analysis.
For the lead-free solder alternative, SAC, the upstream life-cycle stage plays a more
important role than it does for SnPb. SAC has ten impact categories where the upstream stage is
the majority contributor, while the use/applications stage dominates another four categories.
Like the other two solders, the end-of-life stage drives the aquatic ecotoxicity impact category.
Table 3-125 through 3-127 list the top contributing flows and their associated processes
and life-cycle stages for each impact category for each of the solders. For all categories that are
dominated by the use/application stage, except for occupational and public health categories,
impacts result from the electricity generation for the wave application process. For the public
and occupational health categories, the use/application stage dominates from the actual wave
application process. As stated in the previous sections, because the use/application stage is so
dominant, a sensitivity analysis of the use/application energy is provided in Section 3.3.
Additionally, alternative analyses are conducted with (1) alternative silver production process
data, and (2) the results of the less aggressive teachability study for EOL processes. These are
also presented in Section 3.3.
The tables show that for each alloy nearly all impact categories are driven by a different
flow. Silver production is the primary process driving many of the upstream impacts for SAC,
yet as many as six different material flows resulting from silver production are responsible for
being the major contributor in any one impact category. For example, suspended solids from
silver production drive the water quality impacts, while halon (1301) is the largest contributor to
ozone depletion. Only the release of sulfur dioxide to air during extraction and processing of
silver is the major contributor in more than one impact category driven by silver production. For
SnCu and SnPb bar solders, natural gas and dust releases to air from tin production are the only
releases from upstream processes that make up a majority contribution to the impact categories.
There are some flows that are top contributors to more than one impact category, though
they may originate from separate processes. For example, sulphur dioxide that drives
photochemical smog and air acidification in the SnPb life-cycle is from electricity generation
associated with reflow application. In the SAC solder life-cycle, sulphur dioxide is the top
contributor to three categories: photochemical smog, air acidification, and public human health
(non-cancer). In these cases, however, the sulphur dioxide is from silver production in the
upstream life-cycle stage, as opposed to electricity generation for the wave application in the
case of SnPb.
Another top flow in the SnPb life-cycle that contributes to more than one category is lead
emissions to water from landfilling. This is essentially the leachate from landfilling the SnPb
alloy. Lead emissions to water contribute 53.3 percent to the public health (non-cancer) impact
category and 71.4 percent to the aquatic ecotoxicity impact category. As mentioned above, refer
to Section 3.3 for an alternate analysis of these impacts using a less aggressive teachability test
method.
In several instances, the top contributing individual flows comprise a large majority of
the total contribution to the alloy's life-cycle impacts within a category. For example, COD
constitutes 87.4 percent of the total water eutrophication impacts from SnPb bar solder. As there
is not a large amount of chemicals for which there are eutrophi cation potentials and the
3-206
-------
inventory in this project only has a few water eutrophying chemicals, it is understandable that
one material might greatly dominate impacts. This is true for COD, despite its relatively low
eutrophication potential (see Appendix D).
Many top contributors constitute a majority of the total impacts within a category, though
the bar solder results are dominated by one flow less than the paste solders. For SnPb solder
paste, eleven of the sixteen impact categories had top flows representing a majority of total
impacts, while only eight of the sixteen categories for bar solder had a leading contributor of
more than fifty percent. SAC and SnCu solders had contributions greater than fifty percent in
eight and nine categories respectively.
3-207
-------
Table 3-125. Top contributing flows to SnPb bar solder impacts
Impact category
Non-renewable resource use
Renewable resource use
Energy
Landfill space use
Global warming
Ozone depletion
Photochemical smog
Air acidification
Air particulates
Water eutrophication
Water quality
Occupational
health — non-cancer
Occupational
health — cancer
Public human
health — non-cancer
Public human health — cancer
Aquatic ecotoxicity
Life-cycle stage
Use/application
Use/application
Use/application
End-of-life
Use/application
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
Use/application
Use/application
End-of-life
Use/application
End-of-life
Process
Electricity generation
Electricity generation
Electricity generation
Landfilling
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Tin production
Electricity generation
Electricity generation
SnPb wave application
SnPb wave application
Solder landfilling (SnPb)
SnPb wave application
Solder landfilling (SnPb)
Flow
Inert rock
Water
Hard coal
(resource)
SnPb solder to
landfill
Carbon dioxide
CFC-114
Sulphur dioxide
Sulphur dioxide
Dust (unspecified)
Chemical oxygen
demand
Solids (suspended)
SnPb bar solder
SnPb bar solder
Lead emissions to
water
Flux material F
Lead emissions to
water
%
Contrib.
62.3
81.1
31.8
53.7
60.5
33.1
46.3
47.2
56.3
87.4
62.0
29.8
15.5
53.3
25.5
71.4
5-208
-------
Table 3-126. Top contributing flows to SAC bar solder impacts
Impact category
Non-renewable resource
use
Renewable resource use
Energy
Landfill space use
Global warming
Ozone depletion
Photochemical smog
Air acidification
Air particulates
Water eutrophication
Water quality
Occupational
health — non-cancer
Occupational
health — cancer
Public human
health — non-cancer
Public human
health — cancer
Aquatic ecotoxicity
Life-cycle stage
Upstream
Use/application
Use/application
Upstream
Use/application
Upstream
Upstream
Upstream
Upstream
Use/application
Upstream
Use/application
Upstream
Upstream
Use/application
End-of-life
Process
Silver production
Electricity generation
Electricity generation
Silver production
Electricity generation
Silver production
Silver production
Silver production
Silver production
Electricity generation
Silver production
SAC wave application
Tin production
Silver production
SAC wave application
Unregulated recycling
and disposal (SAC)
Flow
Zinc-Pb-Cu Ore
Water
Hard coal (resource)
Slag (hazardous
waste)
Carbon dioxide
Halon(1301)
Sulphur dioxide
Sulphur dioxide
Dust (unspecified)
Chemical oxygen
demand
Solids (suspended)
SAC bar solder
Natural gas
(resource)
Sulphur dioxide
Flux material C
Silver emissions to
water
%
Contrib.
26.7
56.5
16.2
87.2
32.1
20.3
79.9
83.5
83.8
73.5
69.8
29.1
20.7
49.6
16.9
81.8
5-209
-------
Table 3-127. Top contributing flows to SnCu bar solder impacts
Impact category
Non-renewable resource
use
Renewable resource use
Energy
Landfill space use
Global warming
Ozone depletion
Photochemical smog
Air acidification
Air particulates
Water eutrophication
Water quality
Occupational health — non-
cancer
Occupational
health — cancer
Public human
health — non-cancer
Public human
health — cancer
Aquatic ecotoxicity
Life-cycle stage
Use/application
Use/application
Use/application
End-of-life
Use/application
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
Use/application
Upstream
Use/application
Use/application
End-of-life
Process
Electricity generation
Electricity generation
Electricity generation
Landfilling
Electricity generation
Electricity generation
Electricity generation
Electricity generation
Tin production
Electricity generation
Electricity generation
SnCu wave
application
Tin production
Electricity generation
SnCu wave
application
Unregulated recycling
and disposal (SnCu)
Flow
Inert rock
Water
Hard coal (resource)
SnCu solder to
landfill
Carbon dioxide
CFC-114
Sulphur dioxide
Sulphur dioxide
Dust (unspecified)
Chemical oxygen
demand
Solids (suspended)
SnCu bar solder
Natural gas
(resource)
Sulphur dioxide
Flux material C
Copper emissions to
water
%
Contrib.
63.5
84.8
28.0
53.8
53.3
35.2
46.3
44.5
68.9
91.6
68.5
14.8
16.7
61.9
21.3
90.4
3-210
-------
3.4.5 Limitations and Uncertainties
3.4.5.1 General LCIA methodology limitations and uncertainties
This section summarizes some of the limitations and uncertainties in the LCIA
methodology in general. Specific limitations and uncertainties in each impact category are
discussed in Sections 3.2.2 through 3.2.13 with the LCIA results for the LFSP.
The purpose of an LCIA is to evaluate the relative potential impacts of a product system
for various impact categories. There is no intent to measure the actual impacts or to provide
spatial or temporal relationships linking the inventory to specific impacts. The LCIA is intended
to provide a screening-level evaluation of impacts.
In addition to lacking temporal or spatial relationships and providing only relative
impacts, LCA also is limited by the availability and quality of the inventory data. Data
collection can be time-consuming and expensive, and confidentiality issues may inhibit the
availability of primary data.
Uncertainties are inherent in each parameter described in Table 3-112 through 3-115._
For example, toxicity data require extrapolations from animals to humans and from high to low
doses (for chronic effects), resulting in a high degree of uncertainty. Sources for each type of
data should be consulted for more information on uncertainties specific to each parameter.
Uncertainties exist in chemical ranking and scoring systems, such as the scoring of
inherent properties approach used for human health and ecotoxicity effects. In particular,
systems that do not consider the fate and transport of chemicals in the environment can
contribute to misclassifications of chemicals with respect to risk. Uncertainty is introduced
where it was assumed that all chronic endpoints are equivalent, which is likely not the case. In
addition, when LOAELs were not available but NOAELs were, a factor often was applied to the
NOAEL to estimate the LOAEL, thus introducing uncertainty. The human health and
ecotoxicity impact characterization methods presented in the LFSP LCIA are screening tools that
cannot substitute for more detailed risk characterization methods; however, the methodology is
an attempt to consider chemical toxicity at a screening level for potentially toxic materials in the
inventory.
Uncertainty in the inventory data depends on the responses to the data collection
questionnaires and other limitations identified during inventory data collection. These
uncertainties are carried into the impact assessment. Uncertainties in the inventory data include,
but are not limited to, the following:
•• missing individual inventory items;
•• missing processes or sets of data;
•• measurement uncertainty;
•• estimation uncertainty;
•• allocation uncertainty/working with aggregated data; and
•• unspeciated chemical data.
3-211
-------
The goal definition and scoping process helped reduce the uncertainty from missing data,
although it is assured that some missing data still exist. The remaining uncertainties were
reduced primarily through quality assurance/quality control measures (e.g., performing
systematic double-checks of all calculations on manipulated data). The limitations and
uncertainties in the inventory data were discussed further in Chapter 2.
3.4.5.2 General limitations and uncertainties of results
Limitations and uncertainties in LFSP LCIA results are due to limitations and
uncertainties inherent in LCIA methodology itself, as well as limitations and uncertainties in the
project LCI data. General limitations and uncertainties in the LCIA methodology were discussed
above, and limitations and uncertainties in the project inventory were discussed in Chapter 2. In
addition, particular limitations and uncertainties as they pertain to individual impact category
results are presented in Sections 3.2.2 through 3.2.13.
The overall limitations and uncertainties associated with the results of each impact
category are summarized in Tables 3-117 and 3-123 as relative DQIs. The DQI are qualitative
indicators representing a high (H), medium (M), or low (L) level of overall quality, or some
combination thereof.
For example, most categories in the paste solder results presented in Table 3-117 are
given a medium-to-high relative DQI. Those with lower DQIs include ozone depletion,
occupational cancer, and public cancer. Listed below by impact category are the relative DQI
measures (in parentheses) and the major sources of uncertainty for those categories:
•• Non-renewable and renewable resource use (M-H)—reflow application energy variability
and the use of secondary electricity generation data;
•• Energy use (H)—reflow application energy variability;
•• Landfill space use (M-H)—the use of secondary upstream data;
•• Global warming (H)—reflow application energy variability;
•• Ozone depletion (L-M)—several ozone depleting chemicals in the inventories (from
secondary data sources) are scheduled to have been phased out;
•• Photochemical smog, acidification, and air particulates (M-H)—depends somewhat on
secondary upstream data;
•• Eutrophication and water quality (H)—the use of secondary electricity generation data;
•• Occupational and public non-cancer and aquatic ecotoxicity (M-H)—uncertainty in the
EOL leachate study; and
• • Occupational and public cancer (L-M)—lack of carcinogenicity data for most chemicals.
Details of the uncertainties that contribute to the overall data quality for each impact category are
presented in Sections 3.2.2 through 3.2.13.
3-212
-------
REFERENCES
ATSDR. 1999. Toxicological Profile for Lead (Update). U.S. Department of Health and
Human Services, Agency for Toxic Substances and Disease Registry. Atlanta, Georgia.
July.
Bare, Jane. 1999. Slide Presentation: "Consistent Science for Environmental Decision-making."
Tools for Sustainability Workshop, February 18-19. Accessed at
http://www.epa.gov/ORD/NRMRL/std/mtb/PDONE/TFS6.PDF on August 8, 2001.
Barnthouse, L., J. Fava, K. Humphreys, R. Hunt, L. Laibson, S. Noesen, J. Owens, J. Todd, B.
Vigon, K. Weitz, J. Young (Eds.). 1997. Life-Cycle Impact Assessment: The State-of-
the-Art. Society of Environmental Toxicology and Chemistry, Pensacola, Florida.
CAAA (Clean Air Act Amendments). 1990. Title VI (42 U.S. Code Annotated, Section 7671a;
and associated regulations: 40 CFR 82, App. F, Subpart A and App. I, Subpart A).
Call, D.J., L.T. Brooke, M.L. Knuth, S.H. Porier, and M.D. Hoglund. 1985. "Fish subchronic
toxicity prediction model for industrial organic chemicals that produce narcosis."
Environmental Toxicology and Chemistry 4:335-341.
Curran, M.A. 1996. Environmental Life-Cycle Assessment. McGraw-Hill, New York, NY.
DOE. 2003. 2002 Buildings Energy Databook. U.S. Department of Energy:
http://btscoredatabook.eren.doc.gov. [Converted 105.2 mill BTU/yr for average
household use to 9,244 MJ/month]
Ecobilan. 1999. DEAM, Life-Cycle Inventory Database - Silver Data. Developed by Ecobilan
Group, copyright 1999.
El (Eco-Indicator). 1999. Eco-indicator 99 Methodology Report. Available at:
http://www.pre.nl/eco-indicator99/ei99-reports.htm.
EPA (Environmental Protection Agency). 2004. U.S. EPA Office of Solid Waste Emergency
Response: http://www.epa.gov/epaoswer/non-hw/muncpl/facts.htm.
EPA. 2004. "Harmonizing the Clean Air Act and Montreal Protocol Methyl Bromide
Phaseouts." Accessed at http://www.epa.gov/spdpublc/mbr/harmoniz.html on March 12,
2004.
EPA. 2002. Municipal Solid Waste in the United States: 2000 Facts and Figures,
EPA530-R-02-001, Office of Solid Waste and Emergency Response, June.
3-213
-------
EPA. 1999. Integrated Risk Information System (IRIS). EPA Office of Research and
Development, National Center for Environmental Assessment. Database available at
http ://www. epa.gov/iris/
Fava, J.; F. Consoli, R. Denison, K. Dickson, T. Mohin, and B. Vigon (Eds). 1993. A
Conceptual Framework for Life-Cycle Impact Assessment. Society of Environmental
Toxicology and Chemistry (SETAC)-USA, and SETAC Foundation for Environmental
Education. Pensacola, Florida.
Franklin Associates. 1999. "Characterization OF Municipal Solid Waste in the United States:
1998 Update." July 1999. Prairie Village, Kansas (update to EPA/530-R-94-042,
November 1994).
Goedkoop, M., ""The Ecolndicator 95"", Final Report, NOH Report 9523, Free consultants,
Amersfoort, The Netherlands, 1995.
Guinee, J., R. Heijungs, L. van Oers, D. van de Meent, T. Vermeire, M. Rikken. 1996. LCA
Impact Assessment of Toxic Releases. The Hague, The Netherlands.
Hauschild, M. andH. Wenzel. 1997. Environmental Assessment of Products, Volume 2:
Scientific Backgrounds. Chapman and Hall, London.
Heijungs, R., J.B. Guinee, G. Huppes, R.M. Lankreijuer, H.A. Udo de Haes, A. Wegener
Sleeswijk, A.M.M. Ansems, P.G. Eggels, R. van Duin, and H.P. De Goede. 1992.
Environmental Life Cycle Assessment of Products—Guide. CML Center for
Environmental Studies (Centrum voor Milieukunde), Leiden University, Leiden,
Netherlands.
IARC (International Agency for Research on Cancer). 1998. IARC Monographs. Database on
risks to humans available at http://www.iarc.fr.
ILSI (International Iron and Steel Institute). 2001. World Steel Trends and Statistics. Website
available at http:// www.worldsteel.org/trends_indicators/figures_3.html.
ILSI (International Life Sciences Institute). 1996. Human Health Impact Assessment in Life
Cycle Assessment: Analysis by an Expert Panel. Washington, DC.
IPCC (Intergovernmental Panel on Climate Change). 2001. Climate Change 2001: The
Scientific Basis. Albritton, D.L., L.G. Meiro Filho. Available at:
www.ipcc.ch/pub/wglTARtechsum.pdf
IPCC. 1995. "IPCC Second Assessment Synthesis of Scientific-Technical Information Relevant
to Interpreting Article 2 of the UN Framework Convention on Climate Change."
3-214
-------
Available at web site http://www.ipcc.ch/pub/sarsyn.htm, accessed on August 23, 2001.
ISO (International Standards Organization) 14042. 2001. " ISO 14042: 2001- Environmental
Management- Life Cycle Assessment- Life-Cycle Impact Assessment." Available at the
web site http://www.iso.ch/iso/en/.
Jones, S.L. and T.W. Schultz. 1995. "Quantitative Structure-activity Relationaships for
Estimating the No-observable-effects Concentration in Fathead Minnows (Pimephales
promelas)." Quality Assurance. 4:187-203.
Kenega, E.E. 1982. "Predictability of Chronic Toxicity from Acute Toxicity of Chemicals in
Fish and Aquatic Invertebrates." Environmental Toxicology and Chemistry. 1:347-358.
Lindfors L. G., K. Christiansen, L. Hoffman, Y . Virtanen, V. Juntilla, O-J Hanssen, A.
Roenning, T. Ekval. 1995. Nordic Guidelines on Life-Cycle Assessment. Report #Nord
1995:20. Nordic Council of Ministers, Copenhagen.
PE & IKP. 2000. GaBi 3. The software system for Life Cycle Engineering. Product
Engineering (PE) GmbH and Institut fur Klingststoffkunde und Kunststoffprufung (IKP),
University of Stuttgart, Germany.
SET AC (Society of Environmental Toxicology and Chemistry). 1994. Life-Cycle Assessment
Data Quality: A Conceptual Framework. SET AC and SET AC Foundation for
Environmental Education, Inc., Washington, DC.
Socolof M.L., J.G. Overly, L.E. Kincaid, J.R. Geibig. 2001. Desktop Computer Displays: A
Life-Cycle Assessment, Volumes 1 and2. U.S. Environmental Protection Agency, EPA
744-R-01-004a,b, 2001. Available at: http://www.epa.gov/dfe/pubs/comp-
lic/lca/tocl.pdf.
Swanson, M.B., G.A. Davis, L.E. Kincaid, T.W. Schultz, I.E. Bartmess, S.L. Jones, E.L. George.
1997. "A Screening Method for Ranking and Scoring Chemicals by Potential Human
Health and Environmental Impacts." Environmental Toxicology and Chemistry.
16(2): 372-383.
Swanson. M.B. and Adam C. Socha (eds.). 1997. Chemical Ranking and Scoring: Guidelines
for Relative Assessments of Chemicals. SET AC Press, Pensacola, Florida.
UNEP (United Nations Environment Programme), 2003. Montreal Protocol / UNEP. Available
at: www.uneptie.org/ozonaction/compliance/protocol/ods.html.
WMO (World Meteorological Organisation), 1999. Scientific Assessment of Ozone Depletion:
1998. Global Ozone Research and Monitoring Project-Report no. 44. Geneva.
3-215
-------
3-216
-------
APPENDIX A:
LCI DATA COLLECTION FORMS
Solder Manufacturing Data Collection Form A-l
End-of-Life/Post-Industrial Recycling Data Collection Form A-ll
-------
™
DESIGN FOR THE ENVIRONMENT
LEAD-FREE SOLDER PROJECT
Life-Cycle Inventory (LCI) Data Collection Form
**For Solder Manufacturers** y o
Introduction
The Design for the Environment (DfE) Program in the U.S. Environmental Protection Agency's (EPA) Office of Pollution Prevention and Toxics has begun a voluntary,
cooperative project with the electronics industry to assess the life-cycle environmental impacts of solder alternatives. The DfE Program conducts comparative analyses of
alternative products or processes to provide businesses with data to make environmentally informed choices about product or process improvements. The DfE Program
has no regulatory or enforcement agenda and was established to act as a partner with industry to promote pollution prevention. This environmental life-cycle assessment
will address human and environmental impacts (e.g., energy, natural resource use, global warming, chronic toxicity) of various solders. The University of Tennessee
(UT) Center for Clean Products and Clean Technologies is conducting the life-cycle inventory (LCI), which is the data collection phase of a life-cycle assessment, with
technical assistance from the Electronic Industries Alliance (EIA), IPC - Association Connecting Electronics Industries, and other partners.
Boundaries
A life-cycle assessment considers impacts from materials acquisition, material manufacturing, product manufacturing, use, and final disposition of a product. The LCI
data are intended to be used to evaluate relative environmental impacts over the entire life-cycle of a product. In this project, the product is a type of solder. Therefore,
data associated with the materials and processes used directly in the manufacturing, use, and disposition of the product are relevant to the LCI and requested in this form.
You will not need to include materials or energy not directly used in the production of the solder (e.g., general building heating and air conditioning).
Product focus
This project will evaluate tin-lead solder (for wave and reflow operations)
and consider the following lead-free alternatives:
- Sn/Cu (wave)
~ Sn/Ag/Cu (wave and reflow)
- Sn/Ag/Bi or Sn/Ag/Cu/Bi (reflow)
Most recent (or projected) production data are desired.
Inventory data
We are asking for data on one or multiple "product(s) of interest" that you manufacture, which may be one as defined above under Product Focus. The inputs and outputs
data (Fig. 1) that you provide will be aggregated in the LCI to quantify the overall inputs and outputs of a solder alternative over its life-cycle. A separate form should be
completed for each solder of interest.
p. i
LFSP Solder Manufacturing Stage - Data Collection Form Final version2, 9/17/02
-------
Data sources
Much of the requested information can be drawn from existing sources, including, but not limited to the following:
1. Purchase and production records 5. Audit and analysis results (e.g., wastewater discharge analyses)
2. Bills and invoices 6. Local, state, and federal reporting forms (e.g., hazardous waste manifests)
3. Material Safety Data Sheets (MSDS) 7. Local, state, and federal permits
4. Toxic Release Inventory (TRI) forms 8. Monthly utility billing records
How the data will be used
UT will collect inventory data and tally the inputs and outputs for the different solders. Information gathered by this form will be used to develop environmental profiles
based on inputs and outputs for the manufacturing stage of the solders. The profiles will be used to evaluate environmental impacts from each product. The
environmental profiles can be used to encourage product design changes for product improvement. UT will aggregate data and ensure that data associated with particular
companies remain anonymous to the EPA. UT can enter into confidentiality agreements where proprietary data are concerned. Please understand that accurate and
representative information from you is critical for the success of this project.
Results of project
The results are intended to provide industry with an analysis of the life-cycle environmental impacts and an analysis of end-of-life issues (e.g.,
recyclability and leachability) of leaded and lead-free solders. Results will help identify areas for product and process improvement as related to risk and
environmental impact (e.g., identifying material use inefficiencies) and will identify impacts from various life-cycle stages of the solders. Use of the
results will also help meet growing global demands of extended product responsibility.
Benefits of involvement
As a provider of data, you will be invited to be a member of the project's Technical Workgroup, which reviews interim project reports and is informed of on-going project
status. This will allow for your interests to be considered in project development and data collection. By supplying data, the results will partially reflect your operations
and, therefore, the results will be directly relevant to your interests. The project will allow you to directly apply results to your manufacturing process and identify areas
for improvement and may directly affect industry selection of alterantive solders. You will also be recognized as working voluntarily and cooperatively with the U.S.
EPA.
Deadline
The data collection time frame for this project is June 2002 to October 2002. Submission of forms are encouraged as soon as possible; however, we are attempting to
obtain all completed forms before October 21, 2002.
Your cooperation and assistance are greatly appreciated.
For any questions, please contact Maria Leet Socolof at 865-974-9526, or Jack Geibig at 865-974-6513, and/or the Draft Final Goal Definition and Scoping Document.
P. 11
LFSP Solder Manufacturing Stage - Data Collection Form Final version2, 9/17/02
-------
INSTRUCTIONS
1. Please be sure to read the introductory text on each page before filling out the form.
2. The data you supply in the tables should represent inputs and outputs associated only with the "product of interest" (i.e., a solder as defined in the introduction
under Product Focus, and what you specify in Table 2, #1). If quantities provided are not specific to the "product of interest," please explain how they differ
in the comments section at the bottom of the appropriate table. The ultimate goal is to quantify the amount of inputs and outputs per unit (e.g., kg) of solder manufactured.
3. Where supporting information is available as independent documents, reports or calculations, please provide them as attachments with reference to the associated
table(s) in this form.
4. If you have more than one product of interest to this project, please duplicate this form and fill out one form for each product.
5. If there is not adequate room on a page to supply your data (including comments), please copy the appropriate page and attach it to this packet.
6. The ensuing pages refer to the following indices to detail specifics about the data. Additional information is provided below as required.
Data Quality Indicators Index: These indicators will be used to assess the level of data quality in this form. Please report a DQI for the numerical value
requested in each table on the following pages. The first category, Measured, pertains to a value that is a directly measured quantity. The second category,
Calculated, refers to a value that required one or more calculations to obtain. The third category, Estimated, refers to a value that required a knowledgable employee's
professional judgement to estimate. Lastly, the fourth category, Assumed, should be used only when a number had to be speculatively estimated.
Hazardous and Nonhazardous Waste Management Methods Index: These methods are applicable to both hazardous and nonhazardous wastes (Tables 7a and 7b).
Please give the appropriate abbreviation in the Management Method column on p. 7 where requested. Depending on whether the management method is on or offsite,
please indicate by specifying "on" or "off in the appropriate column on p. 7.
For Tables 3 -6:
Data Quality Indicators Index
M -Measured
C -Calculated
E - Estimated
A - Assumed
For Tables 6a and 6b:
For Tables 7a and 7b (also provided on page 7):
Wastewater Treatment/Disposal Methods Index
A - Direct discharge to surface water
B - Discharge to offsite wastewater treatment facility
C - Underground injection
D - Surface impoundment (e.g., settling pond)
E - Direct discharge to land
F - Other (please specify in comments section)
Waste Management Methods Index
RU - Reused
R - Recycled
L - Landfilled
S - Solidified/stabilized
Iv - Incinerated - volume reduction
le - Incinerated - energy conversion
D - Deep well injected
O - Other (please specify in comments section)
IF YOU HAVE QUESTIONS, PLEASE CONTACT EITHER:
Maria L. Socolof:
Phone: 865-974-9526
Email: socolofml(q)utk.edu
OR
Jack Geibig:
Phone: 865-974-3625
Email: jgeibig@utk.edu
p. m
LFSP Solder Manufacturing Stage - Data Collection Form
Final version2, 9/17/02
-------
1. FACILITY & CONTACT INFORMATION
Table 1. Facility Information Contact Information
1. Company name: 4a. Prepared by: Date:
2. Facility name: 4b. Title:
5. Major products manufactured onsite and their % of your total production (by weight or volume—and please specify):
3. Facility address (location): 4c. Phone number: Ext.:
4d. Fax number:
4e. Email address:
LFSP Solder Manufacturing Stage - Data Collection Form - p. 1 of 7 Final version2, 9/17/02
-------
2. PRODUCT OF INTEREST INFORMATION
Table 2.
1. Solder of interest (please check one alloy, provide its composition, and complete the form for this alloy).
Note, if more than one solder listed below is manufactured, please provide a separate form (Tables 2-7) for each solder of interest
| |Sn/Ag/Cu
IIsn/Pb I"
Sn/Cu
[bar]
JSn/Ag/Bi
]sn/Ag/Cu/Bi
[paste]
[paste]
2. Solder type (please check):
4. Solder melting point:
Bar
Paste
3. Solder density:
6. Year (or period of time) for which data are
supplied (past, current, or projected):
5. Annual production (past, current, or projected) (e.g., units, kg, Ibs):
7. Facility's percent global market share
for solder of interest (optional):
8. Brief description of the main operations/subprocesses
required to manufacture the product of interest:
9. From where (what countries) are your base metals supplied (company names optional)
and what percent does each location contribute to your supply of each metal?
10. Please descnbe any recommended assembly pro tiles for your customers for this solder:
11. What % of your solder from your manufacturing process is recycled?
a. If recycled on-site, how?
If recycled, (please check):
ON-SITE
OFF-SITE
b. If recycled off-site, where? (please provide facility name and location if possible):
12. Do you accept customer's solder dross for recycling? YES NO
13. Do you accept back other contaminated waste forms specifically to recycle the solder?
14. Have you conducted or do you have any leachability studies on the solder of interest?
JYES
IYES
ING If so, what?
[NO If yes, please pro vide a copy.
LFSP Solder Manufacturing Stage - Data Collection Form - p. 2 of 7
Final version2, 9/17/02
-------
3. PRIMARY & ANCILLARY INPUTS
1. Primary & Ancillary Materials: Primary materials are defined as those materials that become part of the final product. Ancillary materials are those material inputs that assist production,
yet do not become part of the final product (e.g., cleaning materials). Please include the trade name and the generic name of each material where applicable.
2. CAS # orMSDS: Please include either the CAS (Chemical Abstract Service) number of each material (fill in the blank with the number), or state "MSDS" and append a copy to this document.
3. Annual quantity/units & Density/units: Please specify the annual amount of material consumed in the year of interest (as specified in Table 2, #6). Please use the units of mass-per-year
(e.g., kg/yr, Ib/yr). If you specify units of volume in lieu of mass, please provide the density. If annual quantities are not available, provide applicable units (e.g., kg/1000 kg of product).
4. Data quality indicators: See the Data Quality Indicators Index on p. iii for abbreviations. Please supply the DQI for the annual quantity value given.
5. Recycled content: Please specify the recycled content of each material identified. For example, 60/40/0 would represent a material that has 60% virgin material, 40% pre-consumer
recycled and 0% post-consumer recycled content. Enter N/A (not applicable) for all components that are assemblies.
Table 3a.
Primary Materials
EXAMPLE: GRTX resin (polypropylene resin)
1.
2.
3.
4.
5.
6.
7.
Primary material comments:
CAS#
or MSDS2
MSDS
Annual
Quantity
450,000
Units
kg/yr
Density
Units
—
DQI4
M
Recycled
Content
60/40/0
Table 3b.
Ancillary Materials
EXAMPLE: Petroleum naphtha (cleaning solvent)
1.
2.
3.
4.
5.
6.
7.
Ancillary material comments:
CAS#
or MSDS2
8032-32-4
Annual
Quantity
920
Units
liters/yr
Density
0.96
Units
kg/liter
DQI4
C
Recycled
Content
100/0/0
LFSP Solder Manufacturing Stage - Data Collection Form - p. 3 of 7
Final version2, 9/17/02
-------
4. UTILITY INPUTS
1. Annual quantity/units: Please specify the amount of the utility consumed in year of interest (as sepcified in Table 2, #6). If possible, please exclude nonprocess-related consumption.
If this is not possible, please include a comment that nonprocess-related consumption is included. If annual quantities are not available, provide applicable units
(e.g., kg/1000 kg of product).
2. Data quality indicators: See the Data Quality Indicators Index on p. iii for abbreviations. Please supply the DQI for the annual quantity value given.
3. Individual Utility Notes:
Electricity:
The quantity of electricity should reflect only that used toward manufacturing the product of interest (identified on p. 2). One approach would be to start with your facility's total annual
electrical energy consumption, remove nonprocess-related consumption, then estimate what portion of the remaining consumption is related to the specific operations of interest.
Please include consumption in all systems that use electricity for process-related purposes. Some examples include compressed air, chilled water, water deionization and HVAC
consumption where clean or controlled environments are utilized.
Natural gas and LNG:
Please exclude all use for space heating or other nonprocess-related uses. If you choose to use units other than MCF (thousand cubic feet), please utilize only units of energy
content or volume (e.g., mmBTU, therm, CCF).
Fuel oils:
Please use units of either volume or energy content (e.g., liters, mmBTU, MJ). Additionally, if the fuel oil is not delivered by underground pipeline, please include the associated
transportation information.
All waters (e.g., DI, city):
Please include all waters received onsite. Please indicate consumption in units of mass or volume.
Table 4.
Utilities3
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
Electricity
Natural gas
Liquified natural gas (LNG)
Fuel oil - type #2 (includes distillate and diesel)
Fuel oil - type #4
Fuel oil - type #6 (includes residual)
Other petroleum-based fuel
Water
Utility comments:
Annual
Quantity
Units
MJ
MCF
MCF
liters
liters
liters
liters
liters
DQI2
LFSP Solder Manufacturing Stage - Data Collection Form - p. 4 of 7
Final version2, 9/17/02
-------
5. AIR EMISSIONS
1. Air emissions: The emissions listed in the table below are some of the more common ones found in air release inventories; if you have information on other specific emissions, please
provide them in the space provided. If you have any reporting forms or other air emission records for applicable year, please attach copies to this form. Also, if you have
information on stack as well as fugitive emissions, please copy this page and place each set of emissions on a different page. The energy consumed in any equipment used onsite to treat
air emissions should be included in the utilities values on p. 4.
2. Annual quantity/units: Please specify the amount of air emissions generated and released to the environment in the year of interest (as specified in Table 2, #6). If the emissions data
are for a different year, please specify the year in the comments section below. Please use units of mass-per-year (e.g., kg/yr, Ib/yr). If annual quantities are not available, provide applicable
units (e.g., kg/1000 kg of product).
3. Data quality indicators: See the Data Quality Indicators Index on p. iii for abbreviations. Please supply the DQI for the annual quantity value given.
Table 5.
Air Emissions
Total particulates
Particulates < 10 microns (PM-10)
Sulfur oxides (SOx)
Nitrogen oxides (NOx)
Carbon monoxide
Carbon dioxide
Methane
Benzene
Toluene
Xylenes
Naphthalene
Total nonmethane VOCs
Other speciated hydrocarbon emissions:
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
CAS
number
630-08-0
124-38-9
74-82-8
71-43-2
108-88-3
1330-20-7
91-20-3
Annual
Quantity
Units
DQI
3
Table 5 (continued).
Air Emissions
Ammonia
Arsenic
Chromium
Copper
Lead
Manganese
Mercury
Nickel
Other emissions:
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Air emission comments:
CAS
number
7664-41-7
7440-38-2
7440-47-3
7440-50-8
7439-92-1
7439-96-5
7439-98-7
7440-02-0
Annual
Quantity
Units
DQI
3
LFSP Solder Manufacturing Stage - Data Collection Form - p. 5 of 7
Final version2, 9/17/02
-------
6. WASTEWATER RELEASES & CONSTITUENTS
1. Annual quantity/units: Please specify the amount of wastewater(s) generated in the year of interest (as specified in Table 2, #6). Please use units of mass-per-year (e.g., kg/yr, Ib/yr).
If multiple streams exist, please copy this page and fill it out for each stream. If annual quantities are not available, provide applicable units (e.g., kg/1000 kg of product).
2. Data quality indicators: See the Data Quality Indicators Index on p. iii for abbreviations. Please include one DQI for the annual wastewater stream quantity value supplied, and one DQI
for the wastewater constituents information supplied. If more than one DQI is applicable to the wastewater constituents data, please clarify this in the comment section.
3. Wastewater constituents: Please let us know what type of values you are supplying (e.g., daily maximums, monthly averages, annual averages). Additionally, if you have any reporting
forms of other wastewater constituent records for the year of interest, please attach them to this form. The energy consumed in any equipment used onsite to treat wastewater
releases should be included in the utilities values on p. 4.
4. Concentration/units: Please specify the concentration of wastewater constituents generated in the year of interest. Please use units of mass-per-volume (e.g., mg/liter, Ib/gal).
5. Wastewater treatment/disposal method: See the Wastewater Treatment/Disposal Methods Index on p. iii for method abbreviations.
Table 6a.
Wastewater Stream
Annual
Quantity
Units
Treatment/Disposal
Method5
DQI for
Annual Quantity
DQI for
Constituents below
Table 6b.
Wastewater Constituents
Dissolved solids
Suspended solids
Carbonaceous Oxygen Demand (COD)
Biological Oxygen Demand (BOD)
Oil & grease
Hydrochloric acid
Sulfuric acid
Other acids (please specify):
1.
2.
Phosphorus
Phosphates
Sulfates
Fluorides
Cyanide
Chloride
Chromium
Aluminum
Nickel
CAS
number
7647-01-0
7664-93-9
Concentration4
Units
Table 6b (continued).
Wastewater Constituents
Mercury
Lead
Nitrogen
Zinc
Tin
Ferrous sulfate
Ammonia
Nitrates
Pesticides
Other speciated constituents:
1.
2.
3.
4.
5.
6.
Wastewater comments:
CAS
number
Concentration4
Units
LFSP Solder Manufacturing Stage - Data Collection Form - p. 6 of 7
Final version2, 9/17/02
-------
7. HAZARDOUS & NONHAZARDOUS WASTES
1. Hazardous wastes and EPA hazardous waste numbers: Please list your waste streams that are considered hazardous by the U.S. EPA. Include the hazardous waste codes for any
hazardous waste you include.
2. Annual quantity/units & Density/units: Please specify the amount of waste generated in the year of interest (as specified in Table 2, #6). Use units of mass-per-year (e.g., kg/yr, Ib/yr).
Please also provide the density for each waste. If annual quantities are not available, provide applicable units (e.g., kg/1000 kg of product).
3. Data quality indicators: See the Data Quality Indicators Index on p. iii for abbreviations. Please supply the DQI for the annual quantity value given.
4. Management method: See key to right of tables for Management Methods Index. If none are applicable, please indicate other and use the comments section to expound.
Table 7a.
Hazardous Wastes
EXAMPLE: Spent solvent (toluene)
1.
2.
3.
4.
5.
6.
7.
8.
Hazardous waste comments:
EPA Haz.
Waste #'
F005
Annual
Quantity
20,000
Units
kg/yr
Density
0.9
Units
kg/liter
DQI3
M
Mgmt.
method
le
On or
offsite?
off
Table 7b.
Nonhazardous Wastes
EXAMPLE: Waste metal chips
1.
2.
3.
4.
5.
6.
7.
Nonhazardous waste comments:
Annual
Quantity
22,000
Units
kg/yr
Density
1,000
Units
kg/m3
DQI3
C
Mgmt.
method
R
On or
offsite?
off
Management Methods Index
RU Reused
R Recycled
L Landfilled
S Solidified/stabilized
Iv Incinerated-volume reduction
le Incinerated-energy conversion
D Deep well injected
O Other (specify in comments)
LFSP Solder Manufacturing Stage - Data Collection Form - p. 7 of 7
Final version2, 9/17/02
-------
DESIGN FOR THE ENVIRONMENT
LEAD-FREE SOLDER PROJECT
Life-Cycle Inventory (LCI) Data Collection Form
**For Solder Recycling Operations**
Introduction U.S. EPA
The Design for the Environment (DfE) Program in the U. S. Environmental Protection Agency's (EPA) Office of Pollution Prevention and Toxics has begun a voluntary,
cooperative project with the electronics industry to assess the life-cycle environmental impacts of solder alternatives. The DfE Program conducts comparative analyses of
alternative products or processes to provide businesses with data to make environmentally informed choices about product or process improvements. The DfE Program has
no regulatory or enforcement agenda and was established to act as a partner with industry to promote pollution prevention. This environmental life-cycle assessment will
address human and environmental impacts (e.g., energy, natural resource use, global warming, chronic toxicity) of various solders. The University of Tennessee (UT) Center
for Clean Products and Clean Technologies is conducting the life-cycle inventory (LCI), which is the data collection phase of a life-cycle assessment, with technical
assistance from IPC — Association Connecting Electronics Industries, the Electronic Industries Alliance (EIA), and other partners.
Boundaries
A life-cycle assessment considers impacts from materials acquisition, material manufacturing, product manufacturing, use, and final disposition of a product. The LCI data
are intended to be used to evaluate relative environmental impacts over the entire life-cycle of a product. In this project, the product is a type of solder. Therefore, data
associated with the materials and processes used directly in the manufacturing, use, and disposition of the product are relevant to the LCI and requested in this form. You
will not need to include materials or energy not directly used in the production of the solder (e.g., general building heating and air conditioning).
Product focus
This project will evaluate tin-lead solder (for wave and reflow operations) and consider
the following lead-free alternatives:
— Sn/Cu (wave)
— Sn/Ag/Cu (wave and reflow)
- Sn/Ag/Bi or Sn/Ag/Cu/Bi (reflow)
Most recent production data are desired (2001 or 2002).
Inventory data
We are asking for data on one or multiple "product(s) of interest" that you manufacture, which may be one as defined above under Product Focus. The inputs and outputs
data (Fig. 1) that you provide will be aggregated in the LCI to quantify the overall inputs and outputs of a solder alternative over its life-cycle. A separate form should be
completed for each different type of solder of interest recycled.
p. i
LFSP EOL Stage - Data Collection Form version 9/18/02
-------
Data sources
Much of the requested information can be drawn from existing sources, including, but not limited to the following:
1. Purchase and production records 5. Audit and analysis results (e.g., wastewater discharge analyses)
2. Bills and invoices 6. Local, state, and federal reporting forms (e.g., hazardous waste manifests)
3. Material Safety Data Sheets (MSDS) 7. Local, state, and federal permits
4. Toxic Release Inventory (TRI) forms 8. Monthly utility billing records
How the data will be used
UT will collect inventory data and tally the inputs and outputs for the different solders. Information gathered by this form will be used to develop environmental profiles
based on inputs and outputs for the end-of-life stage of the solders. The profiles will be used to evaluate environmental impacts from each product. The environmental
profiles can be used to encourage product design changes for product improvement. UT will aggregate data and ensure that data associated with particular companies remain
anonymous to the EPA. UT can enter into confidentiality agreements where proprietary data are concerned. Please understand that accurate and representative information
from you is critical for the success of this project.
Results of project
The results are intended to provide industry with an analysis of the life-cycle environmental impacts and an analysis of end-of-life issues (e.g., recyclability and leachability)
of leaded and lead-free solders. Results will help identify areas for product and process improvement as related to risk and environmental impact (e.g., identifying material
use inefficiencies) and will identify impacts from various life-cycle stages of the solders. Use of the results will also help meet growing global demands of extended product
responsibility.
Benefits of involvement
As a provider of data, you will be invited to be a member of the project's Technical Workgroup , which reviews interim project reports and is informed of on-going project
status. This will allow for your interests to be considered in project development and data collection. By supplying data, the results will partially reflect your operations
and, therefore, the results will be directly relevant to your interests. The project will allow you to directly apply results to your manufacturing process and identify areas for
improvement and may directly affect industry selection of alternative solders. You will also be recognized as working voluntarily and cooperatively with the U.S. EPA.
Deadline
The data collection time frame for this project is May 2002 to November 2002. Submission of forms are encouraged as soon as possible; however, we are attempting to
obtain all completed forms before October 21, 2002.
Your cooperation and assistance are greatly appreciated.
For any questions, please contact Maria Leet Socolofat 865-974-9526, or Jack Geibig at 865-974-6513 , at the University of
Tennessee, 311 Conference Center Bldg., Knoxville, TN37996-4134. Fax: 865-974-1838.
For more project details, see the Project Fact Sheet, DfE Website < WWW. epa.gov/dfe >, or the Draft Final Goal Definition and Scoping Document.
p. ii
LFSP EOL Stage - Data Collection Form version 9/18/02
-------
INSTRUCTIONS
1. Please be sure to read the introductory text on each page before filling out the form.
2. The data you supply in the tables should represent inputs and outputs associated only with the "product of interest" (i.e., a solder as defined in the introduction
under Product Focus, and what you specify in Table 2, #1). If quantities provided are not specific to the "product of interest," please explain how they differ
in the comments section at the bottom of the appropriate table.
3. Where supporting information is available as independent documents, reports or calculations, please provide them as attachments with reference to the associated
table(s) in this form.
4. If you have more than one product of interest to this project, please duplicate this form and fill out one form for each product.
5. If there is not adequate room on a page to supply your data (including comments), please copy the appropriate page and attach it to this packet.
6. The ensuing pages refer to the following indices to detail specifics about the data. Additional information is provided below as required.
Data Quality Indicators Index: These indicators will be used to assess the level of data quality in this form. Please report a DQI for the numerical value
requested in each table on the following pages. The first category, Measured, pertains to a value that is a directly measured quantity. The second category,
Calculated, refers to value that required one or more calculations to obtain. The third category, Estimated, refers to a value that required a knowledgable employee's
professional judgement to estimate. Lastly, the fourth category, Assumed, should be used only when a number had to be speculatively estimated.
Hazardous and Nonhazardous Waste Management Methods Index: These methods are applicable to both hazardous and nonhazardous wastes (Tables 7a and 7b).
Please give the appropriate abbreviation in the Management Method column on p. 7 where requested. Depending on whether the management method is on or offsite,
please indicate by specifying "on" or "off in the appropriate column on p. 7.
For Tables 3 -6:
Data Quality Indicators Index
M -Measured
C -Calculated
E - Estimated
A - Assumed
For Tables 6a and 6b:
For Tables 7a and 7b (also provided on page 7):
Wastewater Treatment/Disposal Methods Index
A - Direct discharge to surface water
B - Discharge to offsite wastewater treatment facility
C - Underground injection
D - Surface impoundment (e.g., settling pond)
E - Direct discharge to land
F - Other (please specify in comments section)
Waste Management Methods Index
RU - Reused
R - Recycled
L - Landfilled
S - Solidified/stabilized
Iv - Incinerated - volume reduction
le - Incinerated - energy conversion
D - Deep well injected
O - Other (please specify in comments section)
IF YOU HAVE QUESTIONS, PLEASE CONTACT EITHER:
Maria L. Socolof:
Phone: 865-974-9526
Email: socolofml(q)utk.edu
OR
Jack Geibig:
Phone: 865-974-3625
Email: jgeibig@utk.edu
p. in
LFSP EOL Stage - Data Collection Form
version 9/18/02
-------
1. FACILITY & CONTACT INFORMATION
Table 1. Facility Information Contact Information
1. Company/Facility name: 4a. Prepared by: Date:
2. Facility address (location): 4b. Title:
4c. Phone number: Ext.:
4d. Fax number:
4e. Email address:
3. Products produced onsite (e.g., secondary lead, recycled Sn/Pb):
LFPS EOLStage - Data Collection Form - p. 1 of 7 version 9/18/02
-------
2. PRODUCT OF INTEREST INFORMATION
Table 2.
NOTE: If more than one solder listed in #3 is processed, please provide a separate form for each alloy, if possible
1. What is your major recycled product (e.g., lead, tin, copper)?
2. Do you accept: | |post industrial waste (e.g., dross from printed wiring board assemblers)
| |post consumer waste (e.g., printed wiring boards from disassembled consumer products)
3. What waste solder alloys do you recieve for recycling [check the applicable alloy(s) and provide composition]:
| |Sn/Pb | |Sn/Ag/Cu
| |Sn/Cu | |Sn/Ag/Bi
| |Sn/Ag/Cu/Bi
4. What is your annual production of recycled solder metal (past, current, or projected) (e.g., units, kg, Ibs).
Specify each solder metal that is recycled and the production associated with each metal:
5. What percent of your operations are associated with processing electronics scrap only?
6. Year (or period of time) for which data are 7. Facility's percent global market share
supplied (past, current, or projected): for solder of interest (optional):
8. Briefly describe the main operations/subprocesses
required to process the waste solder:
9. What by-products are produced?
10. If you are processing lead-free solders in your recycling operations, briefly describe how operations differ from processing Sn-Pb (e.g., greater energy demands, greater time,
more refining steps): Note, if you are processing lead-free solders separately, please provide all separate tables in this form for each different alloy processed.
LFSP EOL Stage - Data Collection Form - p. 2 of 7 version 9/18/02
-------
3. PRIMARY & ANCILLARY INPUTS
| Data for
alloy
1. Primary & Ancillary Materials: Primary materials are defined as those materials that become part of the final product. Ancillary materials are those material inputs that assist production,
yet do not become part of the final product (e.g., cleaning materials). Please include the trade name and the generic name of each material where applicable.
2. CAS # orMSDS: Please include either the CAS (Chemical Abstract Service) number of each material (fill in the blank with the number), or state "MSDS" and append a copy to this document.
3. Annual quantity/units & Density/units: Please specify the annual amount of material consumed in the year of interest (as specified in Table 2, #6). Please use the units of mass-per-year
(e.g., kg/yr, Ib/yr). If you specify units of volume in lieu of mass, please provide the density. If annual quantities are not available, provide applicable units (e.g., kg/1000 kg of product).
4. Data quality indicators: See the Data Quality Indicators Index on p. iii for abbreviations. Please supply the DQI for the annual quantity value given.
5. Recycled content: Please specify the recycled content of each material identified. For example, 60/40/0 would represent a material that has 60% virgin material, 40% pre-consumer
recycled and 0% post-consumer recycled content. Enter N/A (not applicable) for all components that are assemblies.
Table 3a.
Primary Materials
EXAMPLE: GRTX resin (polypropylene resin)
1.
2.
3.
4.
5.
6.
7.
Primary material comments:
CAS#
or MSDS2
MSDS
Annual
Quantity
450,000
Units
kg/yr
Density
Units
—
DQI4
M
Recycled
Content
60/40/0
Table 3b.
Ancillary Materials
EXAMPLE: Petroleum naphtha (cleaning solvent)
1.
2.
3.
4.
5.
6.
7.
Ancillary material comments:
CAS#
or MSDS2
8032-32-4
Annual
Quantity
920
Units
liters/yr
Density
0.96
Units
kg/liter
DQI4
C
Recycled
Content
100/0/0
LFSP EOL Stage - Data Collection Form - p. 3 of 7
version 9/18/02
-------
4. UTILITY INPUTS
| Data for
alloy
1. Annual quantity/units: Please specify the amount of the utility consumed in year of interest (as sepcified in Table 2, #6). If possible, please exclude nonprocess-related consumption.
If this is not possible, please include a comment that nonprocess-related consumption is included. If annual quantities are not available, provide applicable units
(e.g., kg/1000 kg of product).
2. Data quality indicators: See the Data Quality Indicators Index on p. iii for abbreviations. Please supply the DQI for the annual quantity value given.
3. Individual Utility Notes:
Electricity:
The quantity of electricity should reflect only that used toward manufacturing the product of interest (identified on p. 2). One approach would be to start with your facility's total annual
electrical energy consumption, remove nonprocess-related consumption, then estimate what portion of the remaining consumption is related to the specific operations of interest.
Please include consumption in all systems that use electricity for process-related purposes. Some examples include compressed air, chilled water, water deionization and HVAC
consumption where clean or controlled environments are utilized.
Natural gas and LNG:
Please exclude all use for space heating or other nonprocess-related uses. If you choose to use units other than MCF (thousand cubic feet), please utilize only units of energy
content or volume (e.g., mmBTU, therm, CCF).
Fuel oils:
Please use units of either volume or energy content (e.g., liters, mmBTU, MJ). Additionally, if the fuel oil is not delivered by underground pipeline, please include the associated
transportation information.
All waters (e.g., DI, city):
Please include all waters received onsite. Please indicate consumption in units of mass or volume.
Table 4.
Utilities3
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
Electricity
Natural gas
Liquified natural gas (LNG)
Fuel oil - type #2 (includes distillate and diesel)
Fuel oil - type #4
Fuel oil - type #6 (includes residual)
Other petroleum-based fuel
Water
Utility comments:
Annual
Quantity
Units
MJ
MCF
MCF
liters
liters
liters
liters
liters
DQI2
LFPS EOL Stage - Data Collection Form - p. 4 of 7
version 9/18/02
-------
5. AIR EMISSIONS
| Data for
alloy
1. Air emissions: The emissions listed in the table below are some of the more common ones found in air release inventories; if you have information on other specific emissions, please
provide them in the space provided. If you have any reporting forms or other air emission records for applicable year, please attach copies to this form. Also, if you have
information on stack as well as fugitive emissions, please copy this page and place each set of emissions on a different page. The energy consumed in any equipment used onsite to treat
air emissions should be included in the utilities values on p. 4.
2. Annual quantity/units: Please specify the amount of air emissions generated and released to the environment in the year of interest (as specified in Table 2, #6). If the emissions data
are for a different year, please specify the year in the comments section below. Please use units of mass-per-year (e.g., kg/yr, Ib/yr). If annual quantities are not available, provide applicable
units (e.g., kg/1000 kg of product).
3. Data quality indicators: See the Data Quality Indicators Index on p. iii for abbreviations. Please supply the DQI for the annual quantity value given.
Table 5.
Air Emissions
Total particulates
Particulates < 10 microns (PM-10)
Sulfur oxides (SOx)
Nitrogen oxides (NOx)
Carbon monoxide
Carbon dioxide
Methane
Benzene
Toluene
Xylenes
Naphthalene
Total nonmethane VOCs
Other speciated hydrocarbon emissions:
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
CAS
number
630-08-0
124-38-9
74-82-8
71-43-2
108-88-3
1330-20-7
91-20-3
Annual
Quantity
Units
DQI
3
Table 5 (continued).
Air Emissions
Ammonia
Arsenic
Chromium
Copper
Lead
Manganese
Mercury
Nickel
Other emissions:
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
Air emission comments:
CAS
number
7664-41-7
7440-38-2
7440-47-3
7440-50-8
7439-92-1
7439-96-5
7439-98-7
7440-02-0
Annual
Quantity
Units
DQI
3
LFSP EOL Stage - Data Collection Form - p. 5 of 7
version 9/18/02
-------
6. WASTEWATER RELEASES & CONSTITUENTS
| Data for
alloy
1. Annual quantity/units: Please specify the amount of wastewater(s) generated in the year of interest (as specified in Table 2, #6). Please use units of mass-per-year (e.g., kg/yr, Ib/yr).
If multiple streams exist, please copy this page and fill it out for each stream. If annual quantities are not available, provide applicable units (e.g., kg/1000 kg of product).
2. Data quality indicators: See the Data Quality Indicators Index on p. iii for abbreviations. Please include one DQI for the annual wastewater stream quantity value supplied, and one DQI
for the wastewater constituents information supplied. If more than one DQI is applicable to the wastewater constituents data, please clarify this in the comment section.
3. Wastewater constituents: Please let us know what type of values you are supplying (e.g., daily maximums, monthly averages, annual averages). Additionally, if you have any reporting
forms of other wastewater constituent records for the year of interest, please attach them to this form. The energy consumed in any equipment used onsite to treat wastewater
releases should be included in the utilities values on p. 4.
4. Concentration/units: Please specify the concentration of wastewater constituents generated in the year of interest. Please use units of mass-per-volume (e.g., mg/liter, Ib/gal).
5. Wastewater treatment/disposal method: See the Wastewater Treatment/Disposal Methods Index on p. iii for method abbreviations.
Table 6a.
Wastewater Stream
Annual
Quantity
Units
Treatment/Disposal
Method5
DQI for
Annual Quantity
DQI for
Constituents below
Table 6b.
Wastewater Constituents
Dissolved solids
Suspended solids
Carbonaceous Oxygen Demand (COD)
Biological Oxygen Demand (BOD)
Oil & grease
Hydrochloric acid
Sulfuric acid
Other acids (please specify):
1.
2.
Phosphorus
Phosphates
Sulfates
Fluorides
Cyanide
Chloride
Chromium
Aluminum
Nickel
CAS
number
7647-01-0
7664-93-9
Concentration4
Units
Table 6b (continued).
Wastewater Constituents
Mercury
Lead
Nitrogen
Zinc
Tin
Ferrous sulfate
Ammonia
Nitrates
Pesticides
Other speciated constituents:
1.
2.
3.
4.
5.
6.
Wastewater comments:
CAS
number
Concentration4
Units
LFSP EOL Stage - Data Collection Form - p. 6 of 7
version 9/18/02
-------
7. HAZARDOUS & NONHAZARDOUS WASTES
1. Hazardous wastes and EPA hazardous waste numbers: Please list your waste streams that are considered hazardous by the U.S. EPA. Inclu
hazardous waste you include.
2. Annual quantity/units & Density/units: Please specify the amount of waste generated in the year of interest (as specified in Table 2, #6). U
Please also provide the density for each waste. If annual quantities are not available, provide applicable units (e.g., kg/1000 kg of product
3. Data quality indicators: See the Data Quality Indicators Index on p. iii for abbreviations. Please supply the DQI for the annual quantity va
4. Management method: See key to right of tables for Management Methods Index. If none are applicable, please indicate other and use the c
Table 7a.
Hazardous Wastes
EXAMPLE: Spent solvent (toluene)
1.
2.
3.
4.
5.
6.
7.
8.
Hazardous waste comments:
EPA Haz.
Waste #'
F005
Annual
Quantity
20,000
Units
kg/yr
Density
0.9
Units
kg/liter
DQI3
M
Mgmt.
method
le
Data for alloy
de the hazardous waste codes for any
se units of mass-per-year (e.g., kg/yr, Ib/yr).
).
lue given.
omments section to expound.
On or
offsite?
off
Table 7b.
nhazardous Wastes
EXAMPLE: Waste metal chips
1.
2.
3.
4.
5.
6.
7.
Nonhazardous waste comments:
Annual
Quantity
22,000
Units
kg/yr
Density
1,000
Units
kg/m3
DQI3
C
Mgmt.
method
R
On or
offsite?
off
Management Methods Index
RU Reused
R Recycled
L Landfilled
S Solidified/stabilized
Iv Incinerated-volume reduction
le Incinerated-energy conversion
D Deep well injected
O Other (specify in comments)
LFSP EOL Stage - Data Collection Form - p. 7 of 7
version 9/18/02
-------
APPENDIX B:
USE/APPLICATION ENERGY TESTING
Geibig, J., M. Socolof, P. Paulraj, and T. Brady. "Life-Cycle Impacts of Energy
Consumption during Reflow Assembly of Electronics using Lead-Free Solders," IPC
APEX 2003, Anaheim, California.
-------
Life-Cycle Impacts of Energy Consumption during Reflow
Assembly of Electronics Using Lead-Free Solders
Jack Geibig
University of Tennessee
Knoxville, TN
Jgeibig(@,utk.edu
Prawin Paulraj
Intel Corporation
Hillsboro, OR
Prawiapaulrai (@,intel. com
Maria Socolof
University of Tennessee
Knoxville, TN
Msoc(@,utk.edu
Todd Brady
Intel Corporation
Chandler, AZ
Todd.a.bradv(@,intel.com
Abstract— The energy consumed during the reflow
assembly of printed wiring board assemblies is expected to
be environmentally significant within the solder product
life-cycle. Wide differences in the melting temperatures of
lead and lead-free solders alternatives suggests that there
may be large and important tradeoffs associated with the
selection of solder and its ultimate impact on the
environment. Preliminary results of testing, conducted as
part of an overall life-cycle assessment of lead and lead free
solders, are presented in this paper and then compared to
previously conducted studies. Life-cycle impacts associated
the test data are also presented.
Testing results indicate that energy consumption can vary
by as much as 40 percent across alternative solders, with the
National Electronics Manufacturing Initiative (NEMI)
recommended Sn/Ag/Cu alloy consuming eight percent
more energy than eutectic Sn/Pb, and the Sn/Ag/Bi alloy
consuming as much as 32 percent less energy. Although
absolute energy consumption values during this test were
higher than other studies, relative energy differences
between solder types strongly agreed with those of previous
studies. Finally, the environmental impacts associated with
the energy consumed during reflow assembly were
demonstrated to be significant when compared energy use
in upstream life-cycle processes.
INTRODUCTION
Adoption of lead-free solders for the manufacturing of
electronics presents the industry with many challenges.
One such challenge results from the elevated melting points
of the leading solder alternatives and the changes required
in the associated assembly profiles. More energy is likely
required to maintain the higher oven temperatures required
to melt and then reflow these solders during assembly,
resulting in increased costs to assemblers and potential
environmental impacts [1,2].
The University of Tennessee has partnered with the US
EPA Design for the Environment Program, non-government
organizations, and members of the electronics industry to
evaluate the life-cycle environmental and human health
impacts of lead and lead-free solder use in the electronics
industry. The primary goal of the project is to conduct a
detailed life-cycle assessment (LCA) of leading solder
alternatives that considers the impacts associated with the
entire product system. For solder, the product system life-
cycle stages include materials extraction and processing of
the metal ore, manufacturing of the solder, application of
the solder during assembly, and the final disposition of the
solder as part of waste electronics.
Primary life-cycle impacts occurring during the solder
application life-cycle stage are expected to result from the
energy consumed during the reflow assembly process [2, 3].
To assess the environmental consequences of a change in
solders during reflow, project partners conducted testing at
an Intel facility to estimate the energy consumed during the
reflow assembly of printed wiring boards (PWBs) using
select lead and lead-free solders. This paper presents the
findings of the testing and compares the results to the
energy consumed from other upstream life-cycle processes.
SOLDER REFLOW TEST METHODOLOGY
Development of a testing protocol was performed in
cooperation with a group of industry experts knowledgeable
about reflow assembly as well as the overall goals of the
LCA project. The advisory group included representatives
from solder suppliers, equipment manufacturers, and
electronics manufacturers with in-house assembly
capability. The developed protocol balanced the need to
collect data in a timely and cost efficient manner with the
desire to capture the primary factors of power consumption
during assembly; namely, the shape of the oven temperature
profile, conveyor speed, oven loading, and the overall mass
of the printed wiring board (PWB) assembly. In order to
evaluate the power consumption under typical operating
conditions, it was assumed that the ovens would be
operating continuously throughout the day or that work
would be scheduled to minimize cost of operation.
Therefore, testing was confined to the measurement of
power consumption during periods of steady-state
B-l
-------
operation, neglecting the preheat cycle.
Solders for evaluation were selected with the overall
objectives of the LCA study in mind, and include the
solders selected for evaluation in the larger LCA study.
Solder alloys compositions evaluated during the testing
include:
• Sn/Pb - 63/37
• Sn/Ag/Bi (SAB) - 42/1/57
• Sn/Ag/Cu (SAC) - 95.5/3.9/0.6
As a result of prior testing at Intel, assembly profiles
describing the rate and duration of the incremental
temperature changes the assembly must undergo to obtain a
functioning solder joint were already available for all but
the bismuth-containing solder. A suggested profile for the
bismuth-containing solder was obtained from Hewlett
Packard and used by Intel to develop an appropriate reflow
profile. The suggested profile was adjusted using a set of
thermocouples attached to the surface of the panel. The
panel was then passed repeatedly through the temperature
zones of the reflow oven while the profile was adjusted
until the surface temperature of the panel met the minimum
peak melting temperature of the solder. The resulting
profile for each solder is depicted in Figure 1.
Figure 1. Solder Reflow Profiles
For comparison purposes, each profile was developed using
a constant conveyor speed across profiles to ensure a
constant and comparable oven loading during periods of
energy measurement. Characteristics of the solder profiles
are presented in Table 1.
Table 1. Reflow Profile Specifications
Solder
Sn/Ag/Bi
Sn/Pb
Sn/Ag/Cu
Peak Temperature
(range)
160.2-170.1C
204.4-219. 1C
235.2-248.8C
TAL
(average)
65 sees
51 sees
65 sees
5 Temp
9.9C
14.7C
13.6C
An Intel micro ATX motherboard that had been previously
assembled was selected as the test assembly for this testing.
The motherboard was selected as a baseline for testing
because it is at the upper end of applications typical for the
consumer electronics market in terms of size, mass and
complexity.
Because solder reflow occurs once the joint reaches the
minimus temperature required for the particular solder, and
because the scope of our testing was limited to energy
consumption and not joint testing, preassembled boards
were used to limit the cost of the testing. A photo of the test
board is shown in Figure 2. Specifications for the test
assembly are presented in Table 2.
Figure 2. Reflow test PWB assembly
Table 2. Test Vehicle Specifications
PWB Type
Length
Width
Mass of Assembly
Mass of Solder
(estimated)
Micro ATX Motherboard
9.6 inches
9.6 inches
225 grams
2.5 grams/board
Testing was conducted at the Intel facility in Hillsboro,
Oregon using a ten zone forced convection reflow oven
with an attached water-cooled chiller unit to cool the
assemblies following reflow. Energy measurements were
taken at the main power feeds to both the oven and chiller
using appropriately sized transducers and a data logger.
Assemblies were fed into the oven at a controlled rate of
35.5 inches per minute until the oven achieved a fully
loaded condition under the design profile. Energy
measurements were taken from the time the first assembly
entered the oven until the final assembly exited the chiller, a
test run duration of thirteen minutes. Assemblies exiting the
oven were allowed to reach room temperature before being
reintroduced to the oven for the next test run.
TESTING RESULTS
Results from the reflow testing are presented in Table 3
below, along with the results from a similar study conducted
by the National Electronics Manufacturing Initiative
(NEMI) [4].
Table 3. Energy Consumption during Reflow Testing
Solder
Sn/Ag/Bi
UT/
Intel
(kW)
15.7
% Change
from
Baseline
-32.5
NEMI
(kW)
N/A
% Change
from
Baseline
—
B-2
-------
Sn/Pb
Sn/Ag/Cu
23.3
25.2
—
8.3
14.8
16.5
—
11.5%
Testing results indicate that there are significant differences
in the amounts of energy required to reflow the various
solders under our test conditions. For example, as
compared to eutectic Sn/Pb solder, the SAC alloy consumed
8.3 percent more energy over the same period of process
operation. This is mostly due to the elevated melting point
(218 °C) of the SAC alloy, which is a full 35 °C higher
than that of the eutectic Sn/Pb alloy. The increased
temperature not only results in higher energy consumption
during reflow, but also requires the re-engineering of most
PWB surface components which can fail under the higher
temperature reflow cycle.
By contrast, the SAB alloy consumed nearly 33 percent less
energy over the same test period. This is largely due to the
influence of the high concentration of bismuth in the solder
that acts to reduce the overall melting point of the alloy to
138 °C, a full 45 °C less than the melting point for the Sn/Pb
eutectic. Still, the results relative to the other solders are
somewhat lower than can be attributed simply to the
decreased melting point. The larger decrease may also
involve other factors such as higher oven efficiency at the
lower temperature, and less energy loss from the oven due
to PWB throughput. In addition, the peak reflow
temperature of 179 °C for the high bismuth alloy does not
approach the typical reflow temperatures used for Sn/Pb,
making the full range of currently approved components
available for assembly without concern for increased
component failure rates.
Results from similar testing conducted as part of the
research activities of the NEMI Lead-Free Component team
have also been displayed in Table 4 for comparison
purposes. As shown in the table, the data presented in this
paper are higher than those reported by the NEMI group [4].
Other studies, both published and unpublished confirm this
disparity [5, 6, 7]. However, while the absolute energy
values are higher, the relative energy consumption among
the different solder alloys reported in this work agrees very
well with that of the other studies. At the time of this
writing, the authors are investigating the source of disparity
between the reported data sets, but are uncertain as to the
cause due to our unfamiliarity with the other studies.
Possible sources of disparity may include the use of less
efficient, older reflow equipment, testing protocols, and
differences in the conditions under which testing occurred
(e.g. reflow profiles). The NEMI study did not include the
SAB alloy so no comparison can be made to the data
collected in this study for that alloy.
An attempt was made to characterize the magnitude of
energy loss to the system attributable to the mass of PWB
assembly passing through the reflow zone. This 'heat sink'
affect is not solely attributable to the mass of the solder, but
rather is related to the mass of the overall assemblies and
the individual characteristics of the materials involved. By
comparing the energy consumption of the reflow ovens
under loaded and unloaded conditions, the amount of
additional energy required due to the work being passed
through the system is estimated and presented in Table 4
below.
Table 4. Baseline Reflow Oven Power Consumption
Solder
Sn/Ag/Bi
Sn/Pb
Sn/Ag/Cu
Unloaded
(kW)
15
20.9
22.2
Loaded
(kW)
15.7
23.3
25.2
% of Total
Energy Due to
Loading
4.5
10.3
11.9
These results apply only to the PWB assembly used in this
testing. However, they also provide a snapshot against
which other board designs and configurations may be
compared to assess the potential magnitude of their
respective energy consumption and the potential range of
values possible.
LIFE CYCLE COMPARISON
Results from the energy consumption testing reported in the
previous section were combined with energy data collected
from other life-cycle stages to assess the impacts of energy
use within the product life-cycle. Sources of energy
included in this evaluation were electricity from the US
power grid, heavy fuel oil, and natural gas. Energy values
within each life-cycle stage were converted to a common
value of megajoules (MJ) and then combined to obtain an
energy use for the entire life-cycle stage. To facilitate
comparison of the energy use across life-cycle stages and
for different solders, a functional unit based on the volume
of solder was used to normalize all data. The volumes were
converted to mass using the density of the solder alloys, and
all data adjusted and reported in energy use per mass of
solder processed.
Life-cycle impacts resulting from energy use were
calculated and presented for the materials extraction &
processing (e.g. diesel to power mining equipment), solder
manufacturing (e.g. natural gas to fire the refining pots),
and solder application life-cycle stages. Since end-of-life
(EOL) energy use data (e.g. electricity to power shredders)
are not yet completely collected and aggregated, impacts
from end-of-life were not included in this evaluation. The
resulting data by life-cycle stage for each solder are
presented in Figure 3.
The figure shows that the energy consumed during the
application and assembly of the PWB's dominate, with
results ranging from ranging from 91-96 percent of the
overall life-cycle energy, depending on the solder type.
Unlike with the other life-cycle stages where the energy
consumption is tightly linked to the mass of solder
B-3
-------
produced, the energy consumed during the reflow
application stage is a function of the physical characteristics
of the solder alloy, and only minutely affected by the mass
of solder processed. The differences in energy consumption
between life-cycle stages become magnified after the data
are normalized by the mass of solder produced.
Energy impacts for reflow solders
(excluding EOL)
2500-
2000-
_
01 -2 1000-
g 500-
Figure 3. Life-Cycle Energy Use (excluding EOL)
Our testing found a higher rate of energy use during reflow
than other reported data. For purposes of comparison, the
NEMI data were substituted for the project test data and the
life-cycle energy use was recalculated. The results are
shown in Figure 4.
E
2500-
| 2000-
D
1 1500-
E
S
| 1000-
0)
1 500-
nergy impacts for reflow solders (excluding
EOL): NEMI use data
DUSE
• MFG
^ ^
1278
,11
v ji U Mb&K
I
> — ~A
1 330
859
L/J15 Ji5
SnPb SAC SAB
Figure 4. Life-Cycle Energy Use (excluding EOL)
using NEMI Testing Data
As can be seen, while the overall values dropped
considerably, the energy use during reflow still dominated
the energy use category. Values ranged from a low of 86
percent for SAC to a high of 93 percent for Sn/Pb solder.
An analysis was conducted on the data to determine the
sensitivity of the energy use data to variations in the mass of
solder applied to the PWB. It was determined that PWB's
would have to contain nearly 27 grams of SAC or over 60
grams of Sn/Pb solder per PWB assembled for the
normalized energy use from reflow soldering to approach
that of the other life-cycle stages. Since the mass of solder
applied to a typical PWB in the consumer market ranges
from 1-3 grams [3, 8], the energy consumption from the
application stage appears to dominate within the range of
typical assembly conditions.
Energy use has several environmental consequences, among
them global warming. As an example of the importance of
energy consumption within the life-cycle, the global
warming impacts from energy consumption were calculated
and presented here. Global warming results from a build-up
of CO2 and other greenhouse gases that are emitted to the
atmosphere, some during the production of electricity and
other energy sources. Global warming impacts are
calculated using the mass of greenhouse gases released to
the atmosphere, which are then modified using a global
warming potential equivalency factor. The equivalency
factor is an estimate of the chemical's atmospheric lifetime
and radiative forcing referenced to a common chemical, in
this case CO2 [9].
Global warming impacts were calculated for the life-cycle
energy use excluding EOL energy consumption, and
presented in Figure 5. The results indicate that the energy
consumed during reflow assembly of the solder is the
primary influence on global warming impacts. The reflow
application of solder is responsible for from 91-96 percent
of the global warming impacts, depending on the solder
type. While the results are preliminary and do not include
the EOL processes, it is expected that this trend will hold
once EOL is included in the data set, due to the enormous
amount of energy required during assembly as compared to
the other life-cycle stages.
DISCUSSION OF RESULTS
Energy use during the reflow process was demonstrated by
this research to be a critical factor in the assessment of the
overall environmental footprint of the solder product
system. The test data indicate that the energy use during
solder reflow assembly, once normalized for mass of solder
processed, accounts for as much as 96 percent of the total
energy consumed over the entire life-cycle, excluding EOL.
Energy consumption was found to vary significantly
between solder alloys, primarily due to the difference in
melting points and the corresponding changes in the reflow
profile design parameters. Testing indicated that soldering
with the SAC alloy would result in an 11 percent increase in
reflow energy use and an overall increase in life-cycle
energy consumption of 13.8 percent when compared to
Sn/Pb. Conversely, soldering with the SAB alloy would
result in a reduction in energy use of nearly 30 percent over
that of Sn/Pb over the same life-cycle stages.
B-4
-------
•if
c- E
QI ,_
II
w
-------
APPENDIX C:
SOLDER LEACHABILITY TESTING
Townsend, T. "Leachability of Printed Wiring Boards Containing Leaded and Lead-Free
Solder." Report prepared for Abt Associates in support of the Lead-Free Solder Project,
March, 2005.
-------
teachability of Printed Wire Boards Containing Leaded and
Lead-Free Solder
Report Presented To:
Abt Associates Inc.
Project Manager: Cheryl Keenan
Report Prepared By:
Timothy G. Townsend
Associate Professor
Department of Environmental Engineering Sciences
University of Florida
May 30, 2003
(Updated March 8, 2005)
-------
Table of Contents
1.0 Introduction 2
2.0 Background 3
3.0 Methods 5
4.0 Results 9
5.0 Observations 14
Appendix C.I Quality Assurance Results 16
Appendix C.2 Location of Samples from Large Board 18
-------
1.0 Introduction
Major components of electronic devices are printed wiring boards (PWBs).
Metallic solder is a major component of a printed wiring assembly (PWA), which
is the PWB populated with components. The prevalent solder type used on most
PWBs is a tin-lead solder. The presence of lead raises several environmental
concerns, including the fate of the lead upon disposal of the discarded
electronic device. Alternative solder types are available. Examples include tin-
copper and tin-silver-copper. The U.S. EPA's Design for the Environment Program
has worked with stakeholders to examine the life-cycle environmental impacts
of tin-lead and lead-free solders. As part of this effort, a life-cycle assessment
(LCA) is being conducted by the University of Tennessee. The impact and fate
of the chemicals in the different solder types upon landfill disposal is an
important consideration in the LCA.
To support the PWB solder LCA, laboratories at the University of Florida were
contracted to conduct regulatory leaching tests on PWBs manufactured with
five alternative solder types. The two leaching tests performed were the toxicity
characteristic leaching procedure (TCLP) and the synthetic precipitation
leaching procedure (SPLP). Both tests were developed by the U.S.
Environmental Protection Agency and are often used in waste management
decision making. The application and limitations of these tests are discussed.
The five solder types investigated include:
• 63% Sn/ 37%Pb,
• 99.3% Sn/0.7% Cu,
• 95.5% Sn/ 4.0% Ag / 0.5% Cu,
• 96.0% Sn/ 2.5% Ag / 1.0% Bi / 0.5% Cu,
• 42.0% Sn/ 1.0% Ag / 57.0% Bi.
TCLP and SPLP tests were conducted on four different PWB sections, each with a
unique configuration and solder density.
-------
2.0 Background
2.1 The Motivation for Leaching Tests
Toxic heavy metals in process waste or discarded product have the potential to
impact human health and the environment when those materials are managed
improperly. The potential risk posed, however, cannot always be simply judged
by the total amount of metals that are present. For some wastes, the heavy
metals may be bound or encapsulated in such a fashion that they do not
migrate from the waste when disposed. Leaching tests are typically used to
assess the potential for heavy metals (or other chemicals) to migrate or leach
from a solid waste in different disposal scenarios.
2.2 Considerations in Selecting Leaching Test Methodology
Several different leaching test methodologies have been developed by
regulatory or testing agencies, or have been described in published literature.
Some leaching methods are relatively simple and rapid. In these tests, wastes
are exposed to a leaching solution in a laboratory container under a prescribed
set of conditions and the concentrations of metals in the solution are measured
after a specified time of exposure. Others evaluate the leaching of metals from
wastes by constructing simulated disposal environments (such as a landfill), and
observing the concentrations of the metals of concern over time.
The selection of an appropriate leaching test depends on several
considerations. The objective of the leaching test is a paramount consideration.
The specific use of a particular leaching test also may be required as part of a
regulatory application (see the discussion of the TCLP below). For assessing the
possible impact from co-disposal of a waste on landfill leachate concentrations,
simple laboratory tests provide an adequate indication of how metals might
leach from the waste. However, since so many factors impact metal
leachability from a waste (e.g., pH, oxidation reduction potential), simple tests
cannot account for all conditions that occur in a landfill. More elaborate testing
protocols (e.g., lab testing under multiple testing conditions, simulated landfill
experiments) may be required. Cost and time are also a major consideration in
leach testing. While more elaborate testing requirements may provide more
realistic results, they are more expensive and may be much more time-
consuming.
Two relatively simple leaching tests are the TCLP and SPLP. The procedures are
similar with the exception of the leaching fluid used. They are described in
greater detail in the following sections. The rationale for selecting these tests is
also discussed.
-------
2.3 TCLP
The TCLP, EPA Method 1311, uses an acetic acid solution to simulate conditions
in a municipal waste landfill where organic acids are produced as a result of
waste decomposition. The TCLP requires 100 g of material for the test and the
material must be size-reduced prior to leaching. Leaching takes place at a 20:1
liquid to solid ratio in a rotary extractor at 30 rpm for 18 hours. The leachates are
then filtered and analyzed for the chemicals of concern.
2.4 SPLP
The SPLP, EPA Method 1312 is similar in nature to the TCLP, but utilizing a leaching
fluid designed to simulate acid rainfall. It contains trace amounts of nitric and
sulfuric acids. The TCLP is used to make hazardous waste determinations. The
SPLP is frequently used to assess risk from environments where large amounts of
organic acids are not expected to be produced (beneficial use through land
application, near surface soil leachate).
2.5 Rationale for Selection of Leaching Experiments
The objective of the research was to evaluate the extent to which metals leach
from PWBs assembled with different solder types. Data developed during testing
would then be used to inform the LCA on potential end-of-life releases from
PWBs disposed by landfilling. However, only minimal data regarding leaching of
metals from PWBs with different solder types have been reported previously, the
TCLP and SPLP were selected to provide a means of leaching a large number of
samples over a range of conditions. The TCLP and SPLP have been found in
many cases to bracket the range of leaching concentrations encountered
when wastes are leached with actual landfill leachate.
-------
3.0 Methods
3.1 Materials Tested
To assess the effects of PWB configuration on leachability, three different PWB
types were selected from boards donated by industry based on their varied
specifications. Solder was applied to the PWBs prior to their shipment to the
University of Florida by passing the unpopulated boards through the appropriate
assembly process. Unpopulated boards were used to prevent metal
contamination from components and to ensure that the results reflect only the
contributions from the applied solder. The PWB types selected for leachability
testing are described as follows:
• A large multi-layer PWB with a variable surface circuit density (designated
board type AB)
• A small PWB with a uniform high solder population density (designated as
board type C)
• A small PWB with a uniform low population solder density (designated as
board type D)
Solders applied to each of the PWB types to be tested included:
• 63% Sn / 37% Pb
• 57.0% Bi/ 42.0% Sn/ 1.0% Ag
• 95.5% Sn/ 4.0% Ag / 0.5% Cu
• 96.0% Sn/ 2.5% Ag / 1.0% Bi / 0.5% Cu
• 99.3% Sn/0.7% Cu
One PWB of each type was also provided with no solder applied to the surface.
These unsoldered PWBs were used as sample "blanks." For boards C and D, the
board types were slightly different for the Sn-Pb and Bi/Sn/Ag solder as
compared to the other three solder types. The difference was minor but was
observed in the weights of the populated and blank boards.
As will be discussed below, the TCLP and SPLP each require 100 g of sample.
One hundred-gram sections of board type C and board type D were identified
and used as samples. Two different 100-g sections of board type AB were
-------
identified and used as samples (designated as board samples A and B). Thus a
total of 4 different board samples were tested (A, B, C and D).
3.2 Sample Processing
The TCLP and SPLP require that samples be size-reduced to less than 0.95 cm.
Size-reduction of the PWBs was performed using an industrial metal press. The
dimensions and weights of each board were measured upon receipt. The
weight data were used to estimate the board-solder density. One hundred-
gram board sections were identified and these were used as the actual
samples. These samples were cut into small squares to meet the size reduction
requirement. To protect against contamination of the samples, the surface and
blade of the metal press were washed with nitric acid before and during the
cutting process.
In the case of samples C and D, the initial weight of each board type was
slightly over 100 g, the size requirement for the TCLP and SPLP. Thus only a small
piece on the edge of the board was identified and removed to bring the
weight of the boards to approximately 100 g. The same piece was removed in
each case. The remainders of the C and D boards were then size-reduced to
meet the requirements of the leaching tests.
The AB boards weighed several times more than 100 g. Thus, two target areas
were identified based on a visual inspection and the overall density of the
boards. Board sample A was selected from a section of the board with a higher
solder density than board sample B (based on visual inspection). Appendix C.2
presents a photo with the approximate location of each section of the board
noted. The same area was cut from each board so that the same architecture
was captured for each sample (i.e. the same amount of solder points were
captured).
In both cases, because of a slight variability among the densities of each board,
the final weights of each sample differed slightly. This was accounted for in the
later testing by maintaining the liquid to solid ratio of 20:1 as required by the
leaching tests.
3.3 Leaching Tests
The TCLP and SPLP are similar, but use different leaching fluids. The TCLP
extraction solution was prepared by diluting a mixture of 11.4 mL of acetic acid
(CH3COOH) and 128.6 mL of 1 N sodium hydroxide (NaOH) to two liters using
reagent water. The final pH of the solution was 4.93 + 0.05. The SPLP leaching
solution was prepared by mixing 60 g of sulfuric acid with 40 g of nitric acid. The
SPLP extraction fluid was prepared by adding between 0.4 and 0.5 mL of the
sulfuric acid / nitric acid mixture to a 2 L volumetric flask and diluting it to volume
with reagent water. The resultant pH was 4.22 +/- 0.05. The leaching tests
-------
involved placing 100 g of reduced size PWB into a 2.2-liter extraction vessel,
adding two liters of leaching solution to the vessel, tumbling for 18 + 2 hrs, and
filtering the extract using a pressurized filtration apparatus with a 0.7-|im
borosilicate glass fiber filter (Environmental Express TCLP filters).
3.4 Leachate Analysis
After filtration, the extract was digested (U.S. EPA Method 3020A). The
digestates were first analyzed for Pb, Ag, Cu, and Sn using a Thermo Jarrell Ash
ICAP 61 E Tracy Analyzer. This instrument was not, however, equipped to analyze
for bismuth. Thus, the digestates were analyzed a second time using flame
atomic absorption (FLAA) spectrometry using a Perkin-Elmer5100 Atomic
Absorption Spectrophotometer. While the detection limits for each element
were below the RCRA toxicity characteristic concentration (TC) limit (for
determining whether a solid waste is a TC hazardous waste), many of the initial
results were below detection limit, even for samples where the elements were
known to be a part of the solder. Thus, many of the samples were re-digested
for analysis using a graphite furnace, and were reanalyzed using this more
sensitive technique (the Perking Elmer 5100 Atomic Absorption
Spectrophotometer). Laboratory blanks, sample spikes, field duplicates, and
calibration check samples were performed as appropriate.
3.5 Estimation of Solder Density
The UF labs were asked to estimate the solder density of the various samples
tested (solder density being defined as the percent of board by weight
consisting of solder). The first attempt to do this was conducted by weighing
each board as received, weighing the blank boards, and then subtracting the
weights to determine solder weight. This method was found to be unsatisfactory
for samples A and B. This resulted from the relatively small weight of solder on
the boards (relative to the boards themselves) and because of inherent weight
differences even between like boards. Solder density estimates for samples C
and D represent the solder density over the entire PWB since the PWBs
themselves weighed only slightly more than the 100 g required for the leaching
tests. Even the results of the D board tests, however, were questionable
because of the relatively small fraction of solder contained. Inherent differences
in overall board weight could have an impact on accuracy of measurements of
small solder weights.
In an effort to get a more accurate estimate of the solder densities of boards A,
B, and D, sections of these boards from extra samples were digested in acid and
the metal content was measured. Specifically, Bi-Sn-Ag board samples were
digested and the mass of solder was estimated based on the amount of bismuth
measured in the digestate. The volume of solder required to assemble a PWB is
a function of both the PWB design and the geometry of the solder connections
7
-------
required. Therefore, the mass of each of the solders for each PWB sample type
were estimated using the ratio of the appropriate solder density (i.e. the density
for the type of solder the PWB was assembled with) to the density of the Bi-Sn-Ag
solder. The solder mass for each sample was then used to calculate the
percentage of the overall sample weight (roughly 100 g) that was comprised of
solder. These estimated densities for PWB samples undergoing leachability
testing are presented in Table 1. Within each solder type (i.e. each column of
the table) the board type with a higher solder density would be expected to
leach more metal because of the higher concentration of metal in the given
100-g sample size consistent across PWB types. It is noted that there is no
standardized digestion procedure for digesting whole boards.
Table 1. Estimated Solder Densities of PWB Samples
(units % by weight of solder on the boards)
Board
Type
A
B
C
D
Sn-Pb
1.7%
0.66%
5.9%
1.0%
Sn-Ag-Bi
1.8%
0.68%
6.4%
1.0%
Sn-Ag-Cu
1.5%
0.58%
5.3%
0.87%
Sn-Ag-Bi-Cu
1.5%
0.58%
5.6%
0.88%
Sn-Cu
1.5%
0.58%
5.3%
0.87%
Notes:
Board types A, B, and D were determined by acid digestion of a sample from the Sn-Ag-
Bi board, followed by analysis of Bi.
Board type C was determined by difference in weight between blank boards and
populated boards.
-------
4.0 Results
The results of the leaching tests are provided in Tables 2-5. Each table presents
the duplicate results and the calculated mean for the TCLP and SPLP performed
on each sample. In cases where one of the replicate measurements was below
the detection limit and the other was not, the average was calculated by
setting the non-detected sample concentration as the detection limit
concentration. This provides a more conservative (higher) mean concentration.
Values in the tables listed as 'less than' a number (e.g., <2.0) indicates the value
was not detected above the detection limit.
-------
Table 2. TCLP and SPLP Results for Sample A
Solder
Type
SPLP A
(mg/L)
SPLPB
(mg/L)
Average
SPLP
(mg/L)
TCLP A
(mg/L)
TCLPB
(mg/L)
Average
TCLP
(mg/L)
63% Sn -- 37% Pb
Ag
Bi
Cu
Pb
Sn
<0.02
O.76
0.05
2.82
<0.02
<0.02
<0.76
0.73
3.61
<0.02
O.02
O.76
0.39
3.21
O.02
O.02
O.76
2.36
162
O.02
O.02
O.76
2.17
153
0.027
O.02
O.76
2.27
157
0.024
57%B/-42%Sn- l%Ag
Ag
Bi
Cu
Pb
Sn
<0.02
<0.02
0.76
0.017
<0.02
0.02
0.022
0.78
0.013
O.02
0.02
0.021
0.77
0.015
O.02
0.02
21.5
29.8
0.51
0.045
0.02
20.7
31.3
0.468
O.02
0.02
21.1
30.6
0.490
0.033
95.5% Sn - 4.0% Ag - 0.5% Cu
Ag
Bi
Cu
Pb
Sn
<0.02
<0.76
1.94
0.01
<0.02
0.02
O.76
1.29
0.01
0.02
0.02
O.76
1.62
0.01
0.02
0.02
O.76
29.7
0.01
0.02
0.02
O.76
28.3
0.015
0.02
0.02
O.76
29.0
0.013
0.02
96% Sn - 2.5% Ag -0.5% Cu-1% Bi
Ag
Bi
Cu
Pb
Sn
<0.02
<0.02
0.79
O.01
0.34
0.02
0.02
1.2
O.01
0.028
0.02
0.02
1.0
O.01
0.184
0.02
0.02
34.5
O.01
0.02
0.02
0.02
27.5
0.048
0.02
0.02
0.02
31.0
0.029
0.02
99.3%Sn-0.7%Cu
Ag
Bi
Cu
Pb
Sn
<0.02
<0.76
0.94
O.01
0.38
O.02
O.76
0.73
O.01
0.45
O.02
O.76
0.84
O.01
0.42
O.02
O.76
35.7
O.01
0.02
O.02
O.76
38.4
0.026
0.02
O.02
O.76
37.0
O.018
0.02
Blank Boards
Ag
Bi
Cu
Pb
Sn
<0.02
<0.76
1.01
O.01
<0.02
O.02
O.76
0.81
O.01
O.02
O.02
O.76
0.91
O.01
O.02
O.02
O.76
29.2
0.017
O.02
O.02
O.76
35.8
O.01
O.02
O.02
O.76
32.5
0.014
O.02
10
-------
Table 3. TCLP and SPLP Results for Sample B
Solder
Type
SPLP A
(mg/L)
SPLPB
(mg/L)
Average
SPLP
(mg/L)
TCLP A
(mg/L)
TCLPB
(mg/L)
Average
TCLP
(mg/L)
63% Sn -- 37% Pb
Ag
Bi
Cu
Pb
Sn
<0.02
O.76
0.07
1.78
<0.02
O.02
O.76
0.06
1.59
O.02
O.02
O.76
0.065
1.68
O.02
O.02
O.76
38.9
68.1
O.02
O.02
O.76
27.7
57.7
O.02
O.02
O.76
33.3
62.9
O.02
57%B/-42%Sn- l%Ag
Ag
Bi
Cu
Pb
Sn
<0.02
<0.02
1.03
0.01
<0.02
0.02
O.02
1.01
0.01
O.02
0.02
O.02
1.02
0.01
O.02
0.02
7.54
32.8
0.122
0.047
0.02
8.99
62.1
0.12
O.02
0.02
8.27
47.5
0.121
0.34
95.5% Sn - 4.0% Ag - 0.5% Cu
Ag
Bi
Cu
Pb
Sn
<0.02
<0.76
1.96
0.01
<0.02
0.02
O.76
1.27
0.01
0.02
0.02
O.76
1.61
0.01
0.02
0.02
O.76
49.7
0.01
0.02
0.02
O.76
50.5
0.01
0.02
0.02
O.76
50.1
0.01
0.02
96% Sn - 2.5% Ag -0.5% Cu-1% Bi
Ag
Bi
Cu
Pb
Sn
<0.02
<0.02
1.31
O.01
0.02
0.02
0.02
1.36
O.01
0.03
0.02
0.02
1.34
O.01
0.03
0.02
0.02
56.3
O.01
0.02
0.02
0.02
47.0
O.01
0.02
0.02
0.02
51.7
O.01
0.02
99.3%Sn-0.7%Cu
Ag
Bi
Cu
Pb
Sn
O.02
O.76
1.49
O.01
0.068
O.02
O.76
1.34
O.01
0.033
O.02
O.76
1.41
O.01
0.051
O.02
O.76
56.4
O.01
0.02
O.02
O.76
44.0
O.01
0.02
O.02
O.76
50.2
O.01
0.02
Blank Boards
Ag
Bi
Cu
Pb
Sn
O.02
O.76
1.05
O.01
O.02
O.02
O.76
0.87
O.01
O.02
O.02
O.76
0.96
O.01
O.02
O.02
O.76
23.9
0.026
O.02
O.02
O.76
48.0
O.01
O.02
O.02
O.76
36.0
0.018
O.02
11
-------
Table 4. TCLP and SPLP Results for Sample C
Solder
Type
SPLP A
(mg/L)
SPLPB
(mg/L)
Average
SPLP
(mg/L)
TCLP A
(mg/L)
TCLPB
(mg/L)
Average
TCLP
(mg/L)
63% Sn -- 37% Pb
Ag
Bi
Cu
Pb
Sn
<0.02
O.76
0.02
2.33
<0.02
<0.02
<0.76
0.11
2.66
<0.02
O.02
O.76
.065
2.50
O.02
O.02
O.76
0.021
54.5
0.13
O.02
O.76
0.02
51.4
0.044
O.02
O.76
0.021
52.9
0.087
57%B/-42%Sn- l%Ag
Ag
Bi
Cu
Pb
Sn
<0.02
<0.02
0.060
0.04
<0.02
0.02
O.02
0.065
0.01
O.02
0.02
O.02
0.063
0.025
O.02
0.02
18.0
1.06
0.44
O.02
0.02
17.8
1.47
0.91
.024
0.02
17.9
1.27
0.67
0.22
95.5% Sn - 4.0% Ag - 0.5% Cu
Ag
Bi
Cu
Pb
Sn
<0.02
O.76
<0.02
0.01
<0.02
0.02
O.76
O.02
0.01
0.02
0.02
O.76
O.02
0.01
0.02
0.02
O.76
O.02
0.01
0.032
0.02
O.76
O.02
0.01
0.052
0.02
O.76
O.02
0.01
0.042
96% Sn - 2.5% Ag -0.5% Cu-1% Bi
Ag
Bi
Cu
Pb
Sn
<0.02
0.031
<0.02
O.01
<0.02
0.02
0.02
O.02
O.01
0.02
0.02
0.026
O.02
O.01
0.02
0.02
0.02
O.02
O.01
0.02
0.02
0.02
O.02
O.01
0.031
0.02
0.02
O.02
O.01
0.026
99.3%Sn-0.7%Cu
Ag
Bi
Cu
Pb
Sn
<0.02
<0.76
<0.02
O.01
<0.02
O.02
O.76
0.02
O.01
0.036
O.02
O.76
0.02
O.01
0.028
O.02
O.76
0.02
O.01
0.14
O.02
O.76
0.02
O.01
0.088
O.02
O.76
0.02
O.01
0.114
Blank Boards
Ag
Bi
Cu
Pb
Sn
<0.02
<0.76
0.068
O.01
<0.02
O.02
O.76
0.054
O.01
0.041
O.02
O.76
0.061
O.01
0.031
O.02
O.76
3.13
O.01
O.02
O.02
O.76
2.07
O.01
O.02
O.02
O.76
2.60
O.01
O.02
12
-------
Table 5. TCLP and SPLP Results for Sample D
Solder
Type
SPLP A
(mg/L)
SPLPB
(mg/L)
Average
SPLP
(mg/L)
TCLP A
(mg/L)
TCLPB
(mg/L)
Average
TCLP
(mg/L)
63% Sn -- 37% Pb
Ag
Bi
Cu
Pb
Sn
<0.02
O.76
0.031
2.25
<0.02
<0.02
<0.76
0.038
2.44
<0.02
O.02
O.76
0.034
2.34
O.02
O.02
O.76
0.155
18.4
0.021
O.02
O.76
0.119
16.1
O.02
O.02
O.76
0.137
17.2
0.21
57%B/-42%Sn- l%Ag
Ag
Bi
Cu
Pb
Sn
<0.02
<0.02
0.503
0.01
<0.02
0.02
O.02
0.021
0.01
O.02
0.02
O.02
0.262
0.01
O.02
0.02
12.3
0.444
0.092
0.073
0.02
8.21
0.361
0.078
O.02
0.02
10.3
0.387
0.085
0.047
95.5% Sn - 4.0% Ag - 0.5% Cu
Ag
Bi
Cu
Pb
Sn
<0.02
<0.76
<0.02
0.07
<0.02
0.02
O.76
0.03
0.01
0.02
0.02
O.76
0.025
0.04
0.02
0.02
O.76
O.02
0.01
0.035
0.02
O.76
O.02
0.01
0.045
0.02
O.76
O.02
0.01
0.040
96% Sn - 2.5% Ag -0.5% Cu-1% Bi
Ag
Bi
Cu
Pb
Sn
<0.02
<0.02
<0.02
O.01
<0.02
0.02
0.02
O.02
O.01
0.02
0.02
0.02
O.02
O.01
0.02
0.02
0.02
O.02
O.01
0.02
0.02
0.02
0.039
O.01
0.02
0.02
0.02
0.03
O.01
0.02
99.3%Sn-0.7%Cu
Ag
Bi
Cu
Pb
Sn
<0.02
<0.76
<0.02
O.01
<0.02
O.02
O.76
0.02
O.01
0.02
O.02
O.76
0.02
O.01
0.02
O.02
O.76
0.02
O.01
0.02
O.02
O.76
0.023
O.01
0.076
O.02
O.76
0.022
O.01
0.048
Blank Boards
Ag
Bi
Cu
Pb
Sn
<0.02
<0.76
<0.02
O.01
<0.02
O.02
O.76
0.02
O.01
O.02
O.02
O.76
0.02
O.01
O.02
O.02
O.76
0.478
O.01
O.02
O.02
O.76
0.657
O.01
O.02
O.02
O.76
0.568
O.01
O.02
13
-------
5.0 Observations
Several observations are noted regarding the leaching results.
• Only two of the metals in the solder types are regulated as toxicity
characteristic (TC) metals and thus capable of causing the boards to be
RCRA hazardous wastes: Pb and Ag. The TCLP results found lead from the
SnPb board to leach at concentrations greater than the RCRA TC limit (5
mg/L). Silver did not leach to concentrations greater than its TC limit (5
mg/L), and was in fact rarely encountered above the detection limit.
• The fact that silver did not leach is contradictory to some of the limited
previous research regarding silver. Most of this previous research,
however, was conducted on solder alone, and not as part of a PWB. It is
clear, as evidenced by the silver results, and others discussed below, that
the other metals present on the PWB and in solution play a large role on
the relative leachability of a given metal.
• Copper was routinely measured in all of the samples. This was a result of
the copper contained in the boards themselves (with no solder). The AB
board leached more copper than the C and D boards. This is likely a
result of the multi-layer configuration of the AB board. More of the surface
was exposed for the copper to leach. Thus, the average leachate
concentration from samples C and D were used to estimate copper
leaching in order to minimize the effect of copper leaching from the
board itself rather than the solder.
• Copper leaching was suppressed somewhat in the tin-lead solder board.
This follows expected electrochemical behavior between lead and
copper.
• Lead, copper and bismuth all leached greater in the TCLP relative to the
SPLP. This has been observed for lead and copper in other research. The
acetic acid used as part of the TCLP acts to complex with some metals
and thus increases the amount that can be leached. The marked
difference between TCLP and SPLP was not noted for silver and tin; both
of these metals, however, were in most cases below the detection limit.
• The Bi-Sn-Ag solder appeared to contain small levels of lead, as it was
observed to leach in the TCLP for all of the board types.
14
-------
• The relationship between solder density (percent solder by weight on a
board) and the metal leachability was examined. Only lead and bismuth
provided a clear relationship of the impact of solder density. Tin and silver
were not detected routinely enough to make such comparisons. Since
copper came from the boards themselves, a comparison of solder density
impacts could also not be made. When comparing the leachability
between samples A and B and between samples C and D, both lead and
bismuth showed increased concentrations for the samples with the large
solder weight. This was most evident in the TCLP results (the bismuth
samples were typically below detection in the SPLP samples). While earlier
drafts of this document reported a mathematical equation related the
solder density to the measured leachate concentrations, such equations
are omitted from this version because the relationship did not hold
between the A/B samples and C/D samples. It is hypothesized that the
particular configuration of the A/B samples allowed more leaching of
lead and bismuth to occur per mass of solder when compared to the C/D
boards. Thus, even though sample C contained more solder than sample
A, sample A leached more. This could be the result of different board
architecture and the fact that AB was a multi-layer board. For use in the
life-cycle analysis, the average of the TCLP samples from A and B were
used to estimate leaching of lead, tin, silver, and bismuth. As stated
earlier, copper leachability estimates used the average of TCLP samples C
and D. Samples were chosen for their greater reliability for each metal
type. The measured leachate concentrations were converted to mass of
metal leached per unit mass of solder using the density of the solder on
the board.
• Caution should be taken when applying the TCLP results too broadly. The
TCLP was designed to be a rapid test for determining whether a solid
waste should be a hazardous waste because of the presence of certain
toxic elements. It was designed to simulate plausible worst case leaching
conditions that might be encountered in a municipal solid waste (MSW)
landfill. Recent research has found that lead leachability is less in typical
landfill leachate relative to the TCLP (Jang, Y.; Townsend, T. Environ. Sci. Tech.
2003, 37, 4778-4784). Other metals may actually leach more in MSW
leachate. Valuable future tests would include leaching different PWBs in
actual landfill leachates and to construct simulated landfills for assessing
leachability in more realistic environments.
15
-------
Appendix C.I. Quality Assurance Results
Quality assurance results are presented in the following tables.
Table. C.1.1. Measured concentration (mg/L) of Blank QA Samples.
QA Set 1
QA Set II
QA Set III
QA Set IV
QA Set V
QA Set VI
Ag Cone.
(mg/L)
<0.02
0.02
0.02
O.02
O.02
0.02
Cu Cone.
(mg/L)
O.02
0.02
0.02
O.02
O.02
0.02
Pb Cone.
(mg/L)
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
Sn Cone.
(mg/L)
O.01
0.01
0.01
O.01
O.01
0.01
Bi Cone.
(mg/L)
O.76
O.76
O.76
O.76
O.76
O.76
Table C.I .2. QA Recovery Results for Blank Spiked Samples
QA Set 1
QA Set II
QA Set III
QA Set IV
QA Set V
QA Set VI
%Ag
Recovery
94.6%
93%
106.1%
114.1%
96.9%
93.7%
%Cu
Recovery
102.6%
94.6 %
91.7%
88.8%
93.7%
94.7 %
%Pb
Recovery
103.6%
101.2%
95.5 %
94%
100.2%
98.2%
%Sn
Recovery
108.3%
105.7%
96.7 %
92.4%
94.8 %
102.3%
%Bi
Recovery
107%
115.7%
109.6%
110.3%
99.7%
1 1 5.4%
16
-------
Table C.I .3. QA Recovery Results for Blank Spiked Samples.
QA Set 1
QA Set II
QA Set III
QA Set IV
QA Set V
QA Set VI
%Ag
Recovery
96.5%
101.2%
81 .2%
94.2%
108.1%
89.2%
%Cu
Recovery
91.3%
91.5%
96.8%
118.6%
93.6%
93.7%
%Pb
Recovery
98.4%
94.8%
98.2%
96.6%
97.7%
97.5%
%Sn
Recovery
88.1%
97.9%
100.3%
106.2%
106.8%
103.4%
%Bi
Recovery*
1 04.2%
96.5%
87.2%
95.5%
* Only four QA data points were needed for Bi analysis because of the limited sample set
number analyzed.
Table C.I.4. Mean concentrations for all TCLP and SPLP Reagent Blank samples.
TCLP Blanks
SPLP Blanks
Ag Cone.
(mg/L)
0.02
0.02
Cu Cone.
(mg/L)
0.02
0.02
Pb Cone.
(mg/L)
0.01
0.01
Sn Cone.
(mg/L)
0.02
0.02
Bi Cone.
(mg/L)
<0.76
<0.76
17
-------
Appendix C.2 Location of Sample on Board AB
Board AB consisted on one large multi-layer board. Two approximately
100-g areas were identified and cut from each board sample for leach testing.
To minimize the number of cuts performed, two side-by-side locations were
selected in long strips. The following figure illustrates the approximate location of
these two samples, identified as A and B. The A sample visually contained a
greater density of solder than the B sample.
18
-------
APPENDIX D:
LIFE-CYCLE IMPACT ASSESSMENT
SUPPORT DATA (NON-TOXICITY)
Global Warming Potentials D-l
Ozone Depletion Potentials D-3
Photochemical Oxidation Creation Potentials D-6
Acidification Potentials D-ll
Water Eutrophication Potentials D-12
-------
Global warming potentials
Flow
CF3I
Carbon dioxide [Inorganic emissions to air]
Ch2Br2
ChSBr
Dichloromethane (methylene chloride) [Halogenated organic emissions to air]
HFC-161 CH3CH2F
CH3CI
Methane [Organic emissions to air (group VOC)]
Trichloromethane (chloroform) [Halogenated organic emissions to air]
HFC-152 CH2FCH2F
HFC-41 Methyl fluoride 593-53-3
HCFC 123 (dichlorotrifluoroethane) [Halogenated organic emissions to air]
HFC 152a (difluoroethane) [Halogenated organic emissions to air]
Trichloroethane [Halogenated organic emissions to air]
HCFC 225ca (dichloropentafluoropropane) [Halogenated organic emissions to air]
HCFC-21 CHCI2F
Nitrous oxide (laughing gas) [Inorganic emissions to air]
HFC 143 (trifluoroethane) [Halogenated organic emissions to air]
ChBrF2
HFC-32 Difluoromethane 75-10-5
HCFC 124 (chlorotetrafluoroethane) [Halogenated organic emissions to air]
HCFC 225cb (dichloropentafluoropentane) [Halogenated organic emissions to air]
HFC 245ca (pentafluoropropane) [Halogenated organic emissions to air]
HCFC 141b (dichloro-1-fluoroethane) [Halogenated organic emissions to air]
HFC-365mfc CF3CH2CF2CH3
HFC-245fa CHF2CH2CF3
HFC-134 1,1,2,2-tetrafluoro-1,2-diiodoethane 359-35-3
HFC-236ea CHF2CHFCF2
HFC 134a (tetrafluoroethane) [Halogenated organic emissions to air] '
HFC-236cb CH2FCF2CF3
Halon (1211)
HFC 43-10 (decaf luo rope ntane) [Halogenated organic emissions to air]
CFC (soft)
HCFC 22 (chlorodifluoromethane) [Halogenated organic emissions to air]
Carbon tetrachloride (tetrachloromethane) [Halogenated organic emissions to air]
HCFC 142b (chlorodifluoroethane) [Halogenated organic emissions to air]
HFC 125 (pentafluoroethane) [Halogenated organic emissions to air]
HFC 227ea (heptafluoropropane) [Halogenated organic emissions to air]
HFC 143a (trifluoroethane) [Halogenated organic emissions to air]
CFC 11 (trichlorofluoromethane) [Halogenated organic emissions to air]
Tetrafluoromethane [Halogenated organic emissions to air]
CFC 113 (trichlorofluoroethane) [Halogenated organic emissions to air]
Halon (1301) [Halogenated organic emissions to air]
Global warming
potentials (100-
year CO2-
equivalents)
<1
1
1
5
10
12
16
23
30
43
97
120
120
140
180
210
296
330
470
550
620
620
640
700
890
950
1100
1200
1300
1300
1300
1500
1600
1700
1800
2400
3400
3500
4300
4600
5700
6000
6900
Sources
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
b
b
b
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
d
d
d*
D-l
-------
Flow
CFC (hard)
CFC 1 1 5 (chloropentafluoroethane) [Halogenated organic emissions to air]
Octafluoropropane perfluoropropane 76-19-7
Decafluorobutane perfluorobutane 355-25-9
Cyclooctafluorobutane perfluorocyclobutane 115-25-3
Dodecafluoro-pentane perfluoropentane 678-26-2
Tetradecafluorhexane perfluorohexane 355-42-0
HFC 236fa (hexafluoropropane) [Halogenated organic emissions to air]
CFC 114 (dichlorotetrafluoroethane) [Halogenated organic emissions to air]
CFC 12 (dichlorodifluoromethane) [Halogenated organic emissions to air]
CFC 116 (hexafluoroethane) [Halogenated organic emissions to air]
HFC 23 (trifluoromethane) [Halogenated organic emissions to air]
CFC 13 (chlorotrifluoromethane) [Halogenated organic emissions to air]
Sulphur hexafluoride [Inorganic emissions to air]
Global warming
potentials (100-
year CO2-
equivalents)
7100
7200
8600
8600
8700
8900
9000
9400
9800
10600
11900
12000
14000
22200
Sources
a
a
a
a
a
a
a
a
a
a
a
a
b
c
c
c
c
c
c
c
c
c
d
Sources:
(a) IPPC 2001 Report: IPCC - Albritton, D.L. ; Meiro Filho, L.G.. www.ipcc.ch/pub/wg1TARtechsum.pdf.
(b) Eco-lndicator 1995.
(c) WMO 98 report: The Scientific Assessment of Ozone Depletion,1998. World Meteorological
Organisation, Global Ozone Research and Monitoring Project. Report No. 44. 100 year.
(d) LCA Handbook: Houghton et al., 1994 & 1996; GWP values for the substances marked with * are
1994.
D-2
-------
Ozone depletion potentials
Flow
HCFC-151 C2H4FCI
HCFC-251 C3H4FCI3
HCFC-123
Methyl Chloride
HCFC-31 CH2FCL
HCFC-261 C3H5FCI2
HCFC-262 C3H5F2CI
HCFC-124
HCFC-225ca
HCFC-253 C3H4F3CI
HCFC-271 C3H6FCI
HCFC-225cb
HCFC-21 CHFCI2
HCFC-121 C2HFCI4
HCFC-252 C3H4F2CI2
HCFC-131 C2H2FCI3
HCFC-132 C2H2F2CI2
HCFC 22 (chlorodifluoromethane)
CFC (soft)
HCFC-133 C2H2F3CI
HCFC 142b (chlorodifluoroethane)
HCFC-141 C2H3FCI2
HCFC-142 C2H3F2CI
HCFC - 221 C3HFCI6
HCFC-225 C3HF5CI2
HCFC-122 C2HF2CI3
HCFC-223 C3HF3CI4
HCFC-222 C3HF2CI5
HCFC-224 C3HF4CI3
HCFC-231 C3H2FCI5
HCFC-241 C3H3FCI3
Trichloroethane [Halogenated organic emissions to air]
HCFC-226 C3HF6CI
HCFC-232 C3H2F2CI4
C2H4FBr
HCFC 141b (dichloro-1-fluoroethane) [Halogenated organic emissions to
air]
HCFC-243 C3H3F3CI2
HCFC-242 C3H3F2CI3
Halon-2311
HBFC-2311
HCFC-244 C3H3F4CI
HCFC-233 C3H2F3CI3
Halon-2401
HBFC-2401
HCFC- 234 C3H2F4CI2
Ozone depletion
potential (OOP)
(CFC-1 1
equivalents)
0.005
0.01
0.02
0.02
0.02
0.02
0.02
0.022
0.025
0.03
0.03
0.033
0.04
0.04
0.04
0.05
0.05
0.055
0.055
0.06
0.065
0.07
0.07
0.07
0.07
0.08
0.08
0.09
0.09
0.09
0.09
0.1
0.1
0.1
0.1
0.11
0.12
0.13
0.14
0.14
0.14
0.23
0.25
0.25
0.28
Sources
a*
a*
a'*
a*
a*
a*
a'*
a'*
a*
a*
a'*
a'*
a*
a*
a*
a*
a*
a*
a*
a*
a*
a*
a*
a*
a*
a*
a*
a*
a*
a**
a*
a*
a*
a*
a*
a*
a*
a*
a*
b
b*
c
c
c
c
c
c
c
c
c
c
d
d
d
e
D-3
-------
Flow
C3H4FBr3
C3H5FBr2
HCFC-235 C3H2F5CI
air]
Methyl bromide [Halogenated organic emissions to air]
C3H6FBr
CH2FBr
CHF2Br HBFC-22B1; bromodifluoromethane
CFC 113 (trichlorofluoroethane) [Halogenated organic emissions to air]
C2HFBr4
C3H4F3Br
C3H5F2Br
CFC 114 (dichlorotetrafluoroethane) [Halogenated organic emissions to
CFC 13 (chlorotrifluoromethane) [Halogenated organic emissions to air]
CFC 12 (dichlorodifluoromethane) [Halogenated organic emissions to air]
CFC 11 (trichlorofluoromethane) [Halogenated organic emissions to air]
CFC-1 1 1 pentachlorofluoroethane
CFC- 112 Tetrachlorodifluoroethane
CFC-211 heptachlorofluoropropane
CFC-212 hexachlorotrifluoropropane
CFC-21 3 pentachlorotrifluoropropane
CFC-214 Tetrachlorotetrafluoropropane
CFC-21 5 trichloropentafluoropropane
CFC-21 6 dichlorohexafluoropropane
CFC-21 7 monochloroheptafluoropropane
CFC (hard)
CHFBr2
C3H4F2Br2
Carbon tetrachloride (tetrachloromethane) [Halogenated organic emissions
to air]
C2H2FBr3
C2H3F2Br
C2HF4Br
Halon-1202
Halon-1201
HBFC-1201
C2H2F2Br2
C3HFBr6
C2HF3Br2
C2H2F3Br
C2H3FBr2
C2HF2Br3
C3HF3Br4
C3HF2Br5
C3H2FBr5
C3H3FBr4
Ozone depletion
potential (OOP)
(CFC-1 1
equivalents)
0.3
0.4
0.52
0.6
0.6
0.7
0.73
0.74
0.8
0.8
0.8
0.8
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1.1
1.1
1.1
1.2
1.25
1.4
1.4
1.5
1.5
1.6
1.6
1.7
1.8
1.8
1.9
1.9
1.9
Sources
a*
a
a*
a"
a
a*
a*
a"
a"
a*
a*
a*
a"
a
a"
a"
a
a
a
a
a
a
a
a
a
a*
a*
a
a*
a*
a*
a*
a*
a*
a*
a*
a*
a*
a*
a*
a*
b
b*
b*
c
c
c
c
c
c
d
d
d
e
e
e
e
e
e
e
f
D-4
-------
Flow
C3HF5Br2
C3H2F2Br4
C3HF4Br3
C3H3F3Br2
Halon (1211) [Halogenated organic emissions to air]
C3H3F2Br3
C3HF6Br
C3H3F4Br
C3H2F3Br3
Halon (2404) [Halogenated organic emissions to air]
Halon 2402 dibromotetrafluoroethane 124-73-2
C3H2F4Br2
Halon (1301) [Halogenated organic emissions to air]
C3H2F5Br
Ozone depletion
potential (OOP)
(CFC-1 1
equivalents)
2
2.1
2.2
2.5
3
3.1
3.3
4.4
5.6
6
6
7.5
10
14
Sources
a*
a*
a*
a*
a"
a*
a*
a*
a*
a*
a*
a"
a*
e
e
e
Sources:
(a) Montreal Protocol / UNEP (www.uneptie.org/ozonaction/compliance/protocol/ods.html).
a" These values are estimates and will be revised periodically.
a** This formula does not refer to 1,1,2-trichloroethane.
a' Identifies the most commercially viable substances with OOP values listed against them to be used for
the purposes of the Protocol.
a* Where a range of ODPs is indicated, the highest value in that range shall be used for the purposes of
the Protocol.
(b) WMO (World Meteorological Organisation), 1999. Scientific Assessment of Ozone Depletion: 1998.
Global Ozone Research and Monitoring project - Report no. 44. Geneva, in Guinee, 2002: LCA Handbook,
Institute of Environmental Sciences, The Netherlands.
b* WMO (World Meteorological Organisation), 1992. Scientific Assessment of Ozone Depletion: 1991.
Global Ozone Research and Monitoring Project - Report no. 25. Geneva, in Guinee, 2002: LCA Handbook,
Institute of Environmental Sciences, The Netherlands.
Solomon, S. and Wuebbles, D.J. (1995) Ozone Depletion Potentials, Global Warming Potentials and Future
Chlorine/Bromine Loading, in Scientific Assessment of Ozone Depletion: 1994 (Assessment Co-Chairs D.L.
Albritton, R.T. Watson and P.J. Aucamp), World Meteorological Organisation, Global Ozone Research and
Monitoring Project, Report No. 37, World Meteorological Organisation, Geneva.
(c) Heijungs et al. (1992) and The Eco-lndicator-Final Report. NOH. 1995.
(d) Hauschild 1998 and Eco-lndicator 1999.
(e) The Scientific Assessment of Ozone Depletion,1998. World Meteorological Organisation, Global Ozone
Research and Monitoring Project. Report No. 44. In GaBiS (GaBi, 2000).
(f) Solomon, S. and Albritton, D.L. (1992) Time-Dependent Ozone Depletion Potentials for Short and Long-
Term Forecasts. Nature, 357, 33-37. In Wenzel and Hauschild, 1995.
D-5
-------
Photochemical oxidant potential
Flow
Chloromethane (methyl chloride) [Halogenated organic emissions to air]
Methane [Organic emissions to air (group VOC)] (alkane)
Trichloroethane [Halogenated organic emissions to air] (Methyl
chloroform)
Carbon tetrachloride (tetrachloromethane) [Halogenated organic
emissions to air]
Polychlorinated dibenzo-p-dioxins (2,3,7,8 - TCDD) [Halogenated organic
emissions to air]
Polychlorinated dibenzo-p-furans (2,3,7,8 - TCDD) [Halogenated organic
emissions to air]
Dichloroethane (isomers) [Group NMVOC to air]
Tetrafluoromethane [Halogenated organic emissions to air]
air]
Dichlorobenzene (p-DCB; 1,4-dichlorobenzene) [Halogenated organic
emissions to air]
Chlorobenzene [Halogenated organic emissions to air]
CFC 113 (trichlorofluoroethane) [Halogenated organic emissions to air]
Vinyl chloride (VCM; chloroethene) [Halogenated organic emissions to air]
air]
Polychlorinated biphenyls (PCB unspecified) [Halogenated organic
emissions to air]
CFC 22 (chlorodifluoromethane) [Halogenated organic emissions to air]
air]
CFC 142b (chlorodifluoroethane) [Halogenated organic emissions to air]
CFC 134a (tetrafluoroethane) [Halogenated organic emissions to air]
CFC 13 (chlorotrifluoromethane) [Halogenated organic emissions to air]
CFC 125 (pentafluoroethane) [Halogenated organic emissions to air]
CFC 12 (dichlorodifluoromethane) [Halogenated organic emissions to air]
Dichlorobenzene (o-DCB; 1,2-dichlorobenzene) [Halogenated organic
emissions to air]
CFC 116 (hexafluoroethane) [Halogenated organic emissions to air]
air]
CFC 11 (trichlorofluoromethane) [Halogenated organic emissions to air]
CxHy Chloro
Trichloromethane (chloroform) [Halogenated organic emissions to air]
dimethyl carbonate
Methyl Formate
Nitrogen Dioxide
Tetrachloroethene (perchloroethylene) [Halogenated organic emissions to
air] (tetrachloroethylene)
Formic acid
Carbon monoxide [Inorganic emissions to air]
Sulphur Dioxide
Tertiary - Butyl Acetate
Methyl acetate [Group NMVOC to air] (esters)
Photochemical
oxidant
potential
0.005
0.006
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.021
0.023
0.025
0.027
0.028
0.029
0.032
0.036
0.048
0.053
0.059
Sources
a
a
a
a*
a*
a*"
a
a
a*
a*
a*
b
c
c
c
c
c
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
f
D-6
-------
Flow
Dichloromethane (methylene chloride) [Halogenated organic emissions to
air]
Ethene (acetylene) [Group NMVOC to air] (alkyne)
Tertiary Butanol
Ethane [Group NMVOC to air] (alkane)
Methanol [Group NMVOC to air] (alcohol)
Styrene [Group NMVOC to air]
2-methyl 2-butanol
Propionic acid ( 79-09-4)
Dimethoxy methane (Methylal)
Neopentane (dimethylpropane)
Methyl tert-Butyl Ether
Propane [Group NMVOC to air](a//
-------
Flow
Methyl mercaptan
Ethane diol
n-decane (alkanes)
n-undecane (alkanes)
Decane
Dichloroethene (trans)
Pentane (n-pentane) [Group NMVOC to air] (alkanes)
Crude Oil
CxHy Hydrocarbons
Petrol
Diisopropyl ether
Ethanol (ethyl alcohol) [Group NMVOC to air] (alcohol)
2-methylnonane (alkanes)
Butanol (alcohol)
Isobutyl Acetate
isopentane CH2CH(CH3)C2H5 (alkanes)
2-methyl 1-butanol
Butane (n-butane) [Group NMVOC to air](a//
-------
Flow
n-Propanol
2-pentanone
Methyl propyl Ketone
Butyraldehyde (n-; iso-butanal) [Group NMVOC to air]
Hexa-2-one
Hexa-3-one
Methylcyclohexane (alkanes)
n-butanol
Isobutene (alkene) isobutylene
Methyl propene
n-propylbenzene (aromatic)
Toluene (methyl benzene) [Group NMVOC to air] (aromatic)
Acetaldehyde (aldehyde)
3-methylbut-1-ene (alkene)
Allyl chloride
Ethyl benzene [Group NMVOC to air] (aromatic)
Polycyclic aromatic hydrocarbons (PAH) [Group PAH to air]
Benzo{a}pyrene [Group PAH to air]
Cyclohexane (hexahydro benzene) [Group NMVOC to air]
Phenol (hydroxy benzene) [Group NMVOC to air]
Cyclopentanone [Group NMVOC to air]
Caprolactam
Chlorophenols
CxHy Aromatic
Diphenyl
Hexachlorobiphenyl
Naphthalene
Phthalic acid anhydride
Valeraldehyde (aldehyde) fpentanaldehyde)
Pentanal
2-methylbut-1-ene (alkene)
Xylene (dimethyl benzene) [Group NMVOC to air]
Propionaldehyde Propanol (aldehyde)
Acrolein (aldehyde)
2-methylbut-2-ene (alkene)
1 ,3 - butadiene (look at 74)
1-hexene
o-ethyltoluene (aromatic) (2-ethyltoluene)
Butadiene [Group NMVOC to air]
p-ethyltoluene (aromatic) (4-ethyltoluene)
Butene (vinyl acetylene) [Group NMVOC to air]
1-pentene (alkene)
Ethene (ethylene) [Group NMVOC to a\r](alkenes)
p-xylene (aromatic)
m-ethyltoluene (aromatic) (3-ethyltoluene)
o-xylene (Aromatic)
Trans-2-hexene
cis 2-hexene
Photochemical
oxidant
potential
0.543
0.548
0.548
0.568
0.572
0.599
0.6
0.612
0.627
0.627
0.636
0.637
0.641
0.671
0.7
0.73
0.76098
0.761
0.761
0.761
0.761
0.761
0.761
0.761
0.761
0.761
0.761
0.761
0.765
0.765
0.771
0.777
0.798
0.8
0.842
0.851
0.874
0.898
0.906
0.906
0.959
0.977
1
1.01
1.02
1.05
1.07
1.07
Sources
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
c
c
c
c
c
c
c
c
c
d
d
e
e
e
e
f
f
f
f
f
f
f
f
f
f
D-9
-------
Flow
1-butene (alkene)
Isoprene (alkene)
1 ,2,5-trimethylbenzene (aromatic)
m-xylene (aromatic)
Propene (propylene) [Group NMVOC to a\r](alkene)
2-pentene (trans) (alkene)
cis 2-penene
2-butene (trans) (alkene)
cis 2-butene
1 ,2,3- Trimethylbenzene (aromatic)
1 ,2,4- trimethylbenzene (aromatic)
3,5 dimethyl toluene
3,5 dimethyl ethyl benzene
1 ,3,5 - trimethyl benzene
Photochemical
oxidant
potential
1.08
1.09
1.1
1.11
1.12
1.12
1.12
1.13
1.15
1.27
1.28
1.3
1.32
1.38
Sources
a
a
a
a
a
b
b
b
b
b
b
b
b
b
b
b
b
b
e
Sources:
(a) LCA Handbook: Derwent, R.G., M.E. Jenkin, S.M. Saunders & M.J. Piling, 1998. Photochemical Ozone
Creation Potentials for Organic Compounds in Northwest Europe Calculated with a Master Chemical
Mechanism. Atmos. Environ. 32 (14-15): 2429-2441.
* updated from Jenkin, M.E. & G.D. Hayman, 1999. Photochemical Ozone Creation Potentials for
Oxygenated Volatile Organic Compounds: Sensitivity to Variations in Kinetic and Mechanistic Parameters.
** value for inorganic substances from Derwent, R.G., M.E. Jenkin & S.M. Saunders, 1996. Photochemical
Ozone Creation Potentials fora Large Number of Reactive Hydrocarbons under European Conditions. Atmos.
(b) Eco-lndicator 1999.
(c) Eco-lndicator 1995. j ( )
Photochemical Ozone Creation Potentials: A Study of Different Concepts. J. Air Waste Manage. Assoc. 42(9),
1152-1158.
(e) High NO x: Wenzel and Hauschild: Anderson- Skold, Y., Grennfelt, P. and Pleijel, K. (1992)
Photochemical Ozone Creation Potentials: A Study of Different Concepts. J. Air Waste Manage. Assoc. 42(9),
(f) GaBiS (PE & IKP, 2000).
D-10
-------
Acidification potentials
Flow
Tetrachloroethene (perchloroethylene) [Halogenated organic emissions to air]
Hydrogen bromine (hydrobromic acid) [Inorganic emissions to air]
Nitric acid [Inorganic emissions to air]
Chloromethane (methyl chloride) [Halogenated organic emissions to air]
Vinyl chloride (VCM; chloroethene) [Halogenated organic emissions to air]
Sulphuric acid [Inorganic emissions to air]
Nitrogen Dioxides
Nitrogen oxides [Inorganic emissions to air] (NOx)
Trichloroethane [Halogenated organic emissions to air]
Trichloroethene (isomers) [Halogenated organic emissions to air]
Dichloromethane (methylene chloride) [Halogenated organic emissions to air]
Sulfur Trioxide
Trichloromethane (chloroform) [Halogenated organic emissions to air]
Carbon tetrachloride (tetrachloromethane) [Halogenated organic emissions to air
Hydrochloric Acid
Hydrogen chloride [Inorganic emissions to air]
Phosphoric Acid
Sulfur Oxides
Sulphur dioxide [Inorganic emissions to air]
Nitric Oxide
Nitrogen monoxide
Hydrogen cyanide (prussic acid) [Inorganic emissions to air]
Hydrofluoric acid
Hydrogen fluoride [Inorganic emissions to air]
Ammonia [Inorganic emissions to air]
Hydrogen sulphide [Inorganic emissions to air]
Acidification
potential (S02-
equivalents)
0.19
0.396
0.508
0.634
0.634
0.653
0.7
0.7
0.72
0.72
0.744
0.8
0.803
0.83
0.88
0.88
0.98
1
1
1.07
1.07
1.185
1.6
1.6
1.88
1.88
Sources
a
a
a
a
a
a
a
a
a
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
c
c
c
c
c
c
c
c
c
c
d
d
d
d
d
d
d
d
Sources:
(a) LCA Handbook: Heijungs, R., J.B. Guinee, G. Huppes, R.M. Lankreijer, H.A. Udo de Haes, A. Wegener
Sleeswijk, A.M.M. Ansems, P.G. Eggels, R. van Duin, and H.P. de Goede. 1992. Environmental Life-Cycle
Assesment of Products. Vol. I: Guide, and Vol II: Backgrounds. Leiden: CML Center for Environmental Studies,
Leiden University.
(b) GaBiS (PE & IKP, 2000).
(c) Hauschild and Wenzel - Hauschilld, M.Z. and Wenzel, H. Acidification as Assessment Criterion in the
Environmental Assessment of Products, in: Scientific Background for Environmental Assessment of Products
(eds M. Hauschild and H. Wenzel), Chapman & Hall, London. 1997.
(d) Eco-lndicator 1995.
D-ll
-------
Eutrophication potentials of material flows to water
Flow
Chemical oxygen demand (COD) [Analytical measures to water]
Nitrate [Inorganic emissions to water]
Nitric Acid
Nitrogen dioxide
Nitrogen Monoxide
Ammonium [Inorganic emissions to water]
Ammonia [Inorganic emissions to water]
Total Nitrogen
Phosphoric acid
Phosphate [Inorganic emissions to water]
Phosphorous oxide
Total Phosphorus
Eutrophication
potential (phosphate-
equivalents)
0.022
0.1
0.1
0.13
0.2
0.33
0.35
0.42
0.97
1
1.34
3.06
Sources
a
a
a
a
a
a
a
a
a
a
a
a
b
b
b
b
Sources:
(a) LCA Handbook (2001): Based on Heijungs et al. (1992) with some modifications.
(b) GaBiS (PE & IKP 2000).
D-12
-------
APPENDIX E:
LIFE-CYCLE IMPACT ASSESSMENT
SUPPORT DATA (TOXICITY)
Supporting Toxicity Data E-l
Toxicity Data for Potentially Toxic LFSP Chemicals E-l2
Toxicity Hazard Values for Potentially Toxic LFSP Chemicals E-17
Materials Excluded from Toxic Classification E-21
Final Toxicity Data Selections For Use in LCA E-22
Human Health Toxicity Data Collection E-24
Aquatic Toxicity Data Collection E-25
Other Toxicity-Related Data E-26
Slope Factors E-27
Oral NOAEL Data E-31
Inhalation NOAEL Data E-35
Fish Lethal Concentration E-37
Fish NOEL Data E-42
Geometric Means Summary Table E-47
-------
APPENDIX E:
SUPPORTING TOXICITY DATA
E.I TOXICITY DATA COLLECTION
Background:
In the Lead Free Solder Project (LFSP), human and ecological toxicity impacts are
calculated by using a chemical ranking method (described in Chapter 3, Sections 3.2.11 through
3.2.13). This method was originally developed for a life-cycle assessment (LCA) done with
support from the EPA Office of Research and Development (ORD) and Saturn Corporation. It
was updated for the EPA's Design for the Environment (DfE) Program Computer Display
Project (CDP) in consultation with ORD. The final CDP method was reviewed by ORD as well
as EPA's Office of Pollution Prevention and Toxics Risk Assessment Division (RAD) prior to
publication (Socolof et al., 2001). Other minor updates have been made for this LFSP, which
include (1) separating chronic heath impacts into cancer impacts and chronic non-cancer impacts
(for both public and occupational impacts) and (2) removing the presentation of the terrestrial
ecotoxicity impact category.
Separating the chronic human impacts into two separate categories was done because the
hazard values (HVs) calculated for each of these two impact categories are calculated based on
geometric means for different endpoints. For cancer impacts, the HV is based on the geometric
mean of cancer slope factors. The geometric mean for cancer slope factors are largely
influenced by the slope factors for dioxins, which are very high. Thus the associated hazard
values of most cancer impacts have numerically small HVs (since the HV is calculated by
dividing the chemical-specific slope factor by the geometric mean). Compared to the non-cancer
HVs, the cancer HVs are generally much smaller numbers. Therefore, combining the two impact
scores into one impact category causes the non-cancer impacts to overshadow the cancer
impacts. Therefore, to observe any real resolution in the cancer impact category, the cancer and
non-cancer impact categories have been separated for the LFSP.
The other change from the CDP was to remove the terrestrial toxicity impact category as
being presented independently, because the chronic non-cancer impacts presented alone are
calculated the same way as the terrestrial ecotoxicity impacts. Thus, the terrestrial ecotoxicity
impacts are represented by the non-cancer impacts and thus are not presented separately in the
LFSP.
In the LCA, there is no intent to conduct a full risk assessment or even a screening level
risk assessment, given that there are no real spatial or temporal boundaries to this global,
industry-wide LCA. In order to provide some weighting of the inventory data to represent
potential toxicity, basic toxicity data (e.g., a no observable adverse effect level [NOAEL] for
chronic, non-carcinogenic effects) are used. The intent is to modify the inventory data by the
inherent toxicity of the material to provide a relative toxicity measure.
Table E-l lists the toxicity data used for potentially toxic chemicals in the LFSP
E-l
-------
inventory, and Table E-2 lists the associated HVs calculated per the methodologies described in
Section 3.2.11 through 3.2.13. To save project resources, toxicity data that had been collected
for previous DfE projects were used in the LFSP. Toxicity data used prior to this project were
collected by Syracuse Research Corporation (SRC) (under contract with EPA) and EPA's RAD.
Chemicals identified in the LFSP inventory, for which toxicity data had not been previously
collected, were collected by the Toxicity and Hazard Assessment Group in the Life Sciences
Division at the Oak Ridge National Laboratory (ORNL). ORNL conducted their search in April,
2003, and the data were subsequently reviewed and/or supplemented by EPA's RAD. The
description below presents the method used to collect the LFSP toxicity data.
Data Collection Approach:
Once inventory data are collected for the project, the inventory flows are checked to
determine if they are potentially toxic. The lists of potentially toxic and non-toxic chemicals
were reviewed by EPA. Those excluded from the toxicity list, and assumed to be non-toxic are
provided in Table E-3. The chemicals then deemed potentially toxic are assembled for toxicity
data collection. The data are first checked for correct chemical name and Chemical Abstracts
Service (CAS) registry number, and the associated inventory disposition (e.g., release to water)
is identified to help determine classification into different toxicity impact categories.
Classification helps determine what toxicity data need to be collected. For example, if an
inventory flow is released to water, it will require aquatic toxicity data.
For most of the chemicals identified in the inventory of the life-cycle of the solder
alternatives being evaluated, toxicity data were collected for the CDP. For these chemicals, data
from the CDP were used. For new chemicals identified in this LCA, chronic human toxicity
endpoints and both acute and chronic aquatic toxicity endpoints were searched. The following
specific endpoints are used for calculating human toxicity scores:
• inhalation or oral NOAEL (or inhalation or oral LOAEL),
cancer slope factors, and
• cancer weight of evidence (WOE).
For ecological toxicity, the following endpoints are used for calculating aquatic toxicity:
fish LC50, and
fish NOEL.
In some cases, all endpoints needed to be searched, and in others, only aquatic toxicity endpoints
need to be searched. This simply depended on what data were already available from the
previous studies.
EPA's RAD provided guidance for collecting toxicity data for DfE Cleaner Technologies
Substitutes Assessments. This served as the basis for data collection for this LCA; however, it
was modified as applicable to an LCA. As stated in the RAD guidance, when searching for the
toxicity endpoints, the first sources to be reviewed were to be:
E-2
-------
• EPA's Integrated Risk Information System (IRIS) (http://www.epa.gov/iris/),
• Agency for Toxic Substances and Disease Registry (ATSDR) toxicological profiles,
• EPA's High Production Volume (HPV) Challenge robust summaries and supporting
documents, and
Organization for Economic Cooperation and Development's (OECD's) Screening
Information Data Set (SIDS) robust summaries and supporting documents.
If endpoints from these sources were found, and did not conflict with other sources from this list,
those data were chosen. Applicable data were included in a matrix of the chemicals and
endpoints of interest and provided to UT by ORNL. If more than one value was found for an
endpoint, decisions of what data to use were discussed between ORNL and UT and then UT and
EPA.
If endpoints were not found from the above sources, the following databases were to be
searched:
• Toxline,
• Medline (as appropriate, depending on the toxicity endpoint or endpoints for which data
are being sought), and
TSCATS (Toxic Substances Control Act Test Submissions)-the EPA database that holds
data submitted to the Agency under TSCA sections 4 and 8). Although data in TSCATS
may be unpublished and, therefore, not subjected to peer review by the editors of a
journal, the data may provide useful information on particular chemicals and can be
considered for preparation of robust summaries if the TSCATS data meet Agency
standards for data quality/data adequacy.
For studies providing endpoint data found in these or other alternative sources, ORNL was
instructed to prepare brief summaries of the studies (following the format of a robust summary to
the extent possible, see www.epa.gov/chemrtk/robusumgd.htm). ORNL would then document
which value was chosen and explain why. Consideration of EPA's criteria for data quality/data
adequacy would also be incorporated into the explanation
(www.epa.gov/opptintr/chemrtk/datadfin.htm).
Toxicity Data:
Table E-4 presents the final chosen toxicity data and, where necessary, provides
comments on the selection process. Tables E-5, E-6, and E-7 provide the supporting toxicity
data collected for the LFSP project by ORNL. The data in Tables E-5, E-6, and E-7 were
reviewed by UT. The chosen data were then reviewed by EPA and the actual data points used in
the LFSP life-cycle impact assessment (LCIA) are also provided in Table E-4.
The LCIA methodology is similar to that which was used for the CDP, and is described
in Section 3.1 of this report. The toxicity data required for the LCIA, and what was requested
from ORNL, are as follows:
E-3
-------
• Cancer (mammalian toxicity)
-D oral SF
-D inhalation SF
-D WOE
• Non-cancer (mammalian toxicity)
-D oral NOAEL (or LOAEL)
-D inhalation NOAEL (or LOAEL)
• Aquatic ecotoxicity
-D LC50
-D NOEL
In the cases where chronic ecotoxicity (e.g., no observable effect level [NOEL]) data are not
available, thelogKow and the LC50 are used to predict the NOEL (described in Section 3.1.2.13).
The log Kow values were determined using the LOGKOW/KOWWIN Program found at the
following address: http://esc.syrres.com/interkow/interkow.exe. Table E-5 provides the human
health data and Table E-6 presents the aquatic toxicity data. When other data related to the
toxicity of a chemical were readily available, such data were also reported as "other" toxicity
values, which are provided in Table E-7.
For the LFSP, there were 11 chemicals for which both mammalian toxicity and aquatic
ecotoxicity data were needed and seven chemicals for which only aquatic ecotoxicity data were
needed (mammalian toxicity data were already available from previous projects for those seven
chemicals). The remaining chemical inventory for the LFSP constitutes approximately 150
chemicals. Toxicity data from previous projects (e.g., the CDP) were used for those chemicals.
The toxicity data used for all potentially toxic chemicals in the LFSP are presented in Table E-l.
Per guidance provided by RAD, ORNL was asked to first search the following sources
for toxicity data: IRIS, ATSDR toxicological profiles, HPV challenge robust summaries and
supporting documents, and SIDS robust summaries and supporting documents. If data were not
found in these sources, Toxline, Medline and TSCATS were to be searched, also per RAD
guidance. If data were used from these latter sources, ORNL was instructed to provide robust
summaries for data. No data were used from these sources, thus no robust summaries were
prepared by ORNL.
In cases where there was more than one data point, ORNL selected a data point based on
the applicability of the study to the endpoint of interest and the robustness of the study (as best
could be determined from the available data). In many cases, the original sources were not
reviewed, but information from secondary sources (e.g., EPA's ECOTOXicology Data Base
System [U.S. EPA, 2002]) on the test type and duration were considered. The following
hierarchy offish studies, based on Swanson et al. 1997, was employed to choose LC50
ecotoxicity data in order of preference:
(1) fathead minnow 96-h flow-through test
(2) 96-h flow-through test for another freshwater fish, excluding trout
(3) fathead minnow 96-h static test
(4) 96-h static test for another freshwater fish, excluding trout
E-4
-------
If the only adequate data were for trout, they would also be used. In cases where multiple data
points (with equivalent quality, test type, and species type) were available, an average of those
data was taken as the data point of interest. This was preferred over taking the most toxic
response, as these data are used in relative ranking of chemicals and not to serve as protective
exposure limits.
Other aquatic species (e.g., daphnia, algae) were not used in the original methodology
used to develop the LCIA toxicity method used in this study (i.e., CHEMS-1, Swanson et al.,
1997); however, this does not preclude future versions of this methodology from using other
species besides fish, which would represent lower trophic levels (e.g., daphnia or algae).
E.2 GEOMETRIC MEAN DATA FOR CALCULATING TOXICITY HAZARD VALUES
Tables E-8 through E-12 provide the chemical-specific toxicity data used to calculate the
geometric means for each toxicity endpoint. Table E-13 provides a summary of the geometric
means of each endpoint. The data contributing to the geometric mean calculations were used for
previous projects and this project did not attempt to verify each data point. The geometric means
are used as the comparative basis for calculating the HVs as described in Sections 3.2.11 through
3.2.13.
E.3 REFERENCES
ACC, et al. 2002. (American Chemistry Council; Propylene Glycol Ether Panel Toxicology
Research Task Groups, and CEFIC/Oxygenated Solvents Producers Assoc. Toxicology Support
Group). High Production Volume Submission: Test Plan and Robust Studies Summary for
Propylene Glycol Ethers Category. December 30, 2002. Submitted to U.S. Environmental
Protection Agency.
ACGUL 2002. (American Conference of Governmental Industrial Hygienists). Threshold limit
values and biological exposure indices. ACGIH, Cincinnati, Ohio.
Akzo Nobel. 2002a. MSDS (Material Safety Data Sheet) Ethoduomeen OV/13. Akzo Nobel
Surface Chemistry AB, Stenungsund, Sweden, http://www.surfactantseurope.akzonobel.com.
Retrieved April, 2003.
Akzo Nobel. 2002b. MSDS (Material Safety Data Sheet) Ethoduomeen T/22. Akzo Nobel
Surface Chemistry AB, Stenungsund, Sweden, http: //www. surf actantseur ope. akzonob el. com.
Retrieved April, 2003.
Arena, J.M. 1970. Poisoning: Toxicology, symptoms, treatments. 2nd edition. Springfield, 111:
C.C. Thomas.
Birge, W.J., J.A. Black, and A.G. Westerman. 1979. Evaluation of aquatic pollutants using fish
E-5
-------
and amphibian eggs as bioassay organisms. In: Sympos. Animals Monitors Environ. Pollut. 12:
108-118. Nielsen, Migaki, and Scarpelli, (eds.). Storrs, CT. (as cited inECOTOX).
Camargo, J.A., and J.V. Tarazona, 1991. Short-Term Toxicity of Fluoride Ion (F-) in Soft Water
to Rainbow Trout (Salmo gairdneri) and Brown Trout (Salmo trutta fario). Fluoride 24(2):76-83
(as cited in ECOTOX).
Cardwell, R.D., D.G. Foreman, T.R. Payne, and DJ. Wilbur. 1976. Acute toxicity of selected
toxicants to six species offish. EPA-600/3 -76-008, U.S. Environmental Protection Agency,
Duluth, MN.
Curtis, M. W., T. Copeland, and C. Ward, 1979. Acute Toxicity of 12 Industrial Chemicals to
Freshwater and Saltwater Organisms. Water Resources 13(2): 137-141 (as cited in ECOTOX).
Dawson, G.W., A.L. Jennings, D. Drozdowski, and E. Rider. 1977. The acute toxicity of 47
industrial chemicals to fresh and saltwater fishes. Jour. Hazard. Mater. 1: 303-318. (as cited in
ECOTOX).
Erten-Unal, M., B. Wixson, N. Gale, and J.L. Pitt. 1998. Evaluation of toxicity, bioavailability
and speciation of lead, zinc and cadmium in mine/mill wastewaters. Chemical Speciation and
Bioavailability 10: 37-46.
Ewell, W.S., J.W. Gorsuch, R.O. Kringle, K.A. Robillard, and R.C. Spiegel. 1986.
Simultaneous evaluation of the acute effects of chemicals on seven aquatic species. Environ.
Toxicol. Chem. 5: 831-840. (as cited inECOTOX).
Friberg, L., G.F. Nordberg, E. Kessler, and V.B. Vouk (eds.) 1986. Handbook of the
Toxicology of Metals. 2nd. Ed. Vols. I and II. Amsterdam: Elsevier Science Publishers B.V.
Goettl, J.P. Jr., P.H. Davies, and J.R. Sinley. 1976. Water Pollution Studies. In: Colorado Fish
Res. Rev. 1972-1975, D.B. Cope, (Ed.). DOW-R-R-F72-75, Colorado Div. Of Wildlife,
Boulder, CO. pp. 68-75. (as cited inECOTOX).
Hagan, E.G., W.H. Hansen, O. Fitzhugh, P. Jenner, et al. 1967. Food flavorings and
compounds of related structure. II. Subacute and chronic toxicity. Jour. Food Cosmet. Toxicol.
5:141-147. (as cited in WHO, 1999).
Holtze, K.E. 1983. Effects of pH and ionic strength on aluminum toxicity to early
developmental stages of rainbow trout {Salmo gairdneri Richardson). Res. Rep. Ontario
Ministry of the Environment, Rexdale, Ont, Canada: 39. (as cited in ECOTOX).
HSDB. 2003. (Hazardous Substances Databank). Bismuth, elemental. TOXNET, National
Library of Medicine, National Institute of Health, http://toxnet.nlm.nih.gov. Retrieved April 15,
2003.
E-6
-------
Kimball,G. no date given. The effects of lesser known metals and one organic to fathead
minnows \Pimephalespromelas] and Daphnia magna. U.S. Environmental Protection Agency,
Duluth, MN. (cited in Suter and Tsao, 1996).
Konradova, V. and V. Bencko. 1975. Mechanical damage to rabbit tracheal epithelium from
inhaling inert pyrite dust of needle-like structure. Part I. Jour. Hyg. Microbiol. Epidemiol.
Immunol. 19:279-285.
Ku, A.Y. and J.M. Schoenung. 2002. Toxicity, Availability and Extraction of the Metals Used
in Pb-Free Solders. Presented at UC SMART Workshop, Pb-Free Solder for Electronic, Optical,
and MEMS Packaging Manufacturing. University of California, Los Angeles, September 5,
2002.
Landry, T.D. and B.L. Yano. 1984. Dipropylene glycol monomethyl ether: A 13 week
inhalation toxicity study in rats and rabbits. Fundam. Appl. Toxicol. 4: 612-617.
Mallinckrodt Baker, Inc. 2001. MSDS (Material Safety Data Sheet) Terpineol. J.T. Baker,
from Mallinckrodt Baker, Inc., Phillipsburg, NJ.
NIOSH. 1997. Pocket Guide to Chemical Hazards. U.S. Dept. of Health and Human Services,
Public Health Service, Centers for Disease Control and Prevention, National Institute for
Occupational Safety and Health. DHHS Publication No. 97-140. U.S. Govt. Printing Office,
Washington, D.C.
Perstorp Specialty Chemicals, 2000. MSDS (Material Safety Data Sheet) Dimethylolpropionic
acid. Perstorp Specialty Chemicals AB, Perstorp, Sweden.
Pimentel, R. and R. Bulkley. 1983. Influence of water hardness on fluoride toxicity to rainbow
trout. Environ. Toxicol. Chem. 2: 381-386. (as cited inECOTOX).
Rowe, V.K., D.D. McCollister, H.C. Spencer, et al. 1954. AMA Arch Ind Hyg Occup Med 9:
509-525.
RTECS. 2003a. (Registry of Toxic Effects of Chemical Substances). Zinc Sulfate. RTECS
database originally prepared by the National Institute for Occupational Safety and Health
(NIOSH), now produced and distributed by MDL Information Systems, Inc., under authority of
the U.S. Government.
RTECS. 2003b. (Registry of Toxic Effects of Chemical Substances). Benzo(k)fluoranthene.
RTECS database originally prepared by the National Institute for Occupational Safety and
Health (NIOSH), now produced and distributed by MDL Information Systems, Inc., under
authority of the U.S. Government.
E-7
-------
RTECS. 2003c. (Registry of Toxic Effects of Chemical Substances). Fluosilicic Acid. RTECS
database originally prepared by the National Institute for Occupational Safety and Health
(NIOSH), now produced and distributed by MDL Information Systems, Inc., under authority of
the U.S. Government.
Socolof, M.L. J.G. Overly, L.E. Kincaid, J.R. Geibig. 2001. Desktop Computer Displays: A
Life-Cycle Assessment, Volumes 1 and 2. U.S. Environmental Protection Agency, EPA 744-R-
01-004a,b, 2001. Available at: http://www.epa.gov/oppt/dfe/pubs/comp-dic/lca/).
Smith, L. T. Holsen, N. Ibay, R. Block, and A. De Leon. 1985. Studies on the acute toxicity of
fluoride ion to stickleback, fathead minnow, and rainbow trout. Chemosphere 14: 1383-1389.
(as cited in ECOTOX).
Spehar, R.L. 1986. Criteria Document Data. Memorandum to DJ. Call, U.S. EPA, Duluth,
MN/Center for Lake Superior Environmental Studies, University of Wisconsin-Superior.
September 16, 1986. (as cited in ECOTOX).
Stauffer Chemical Co. 1992. Initial Submission: Toxicology lab report on fluosilicic acid with
cover letter dated 10/27/92. EPA/OTS; Doc. #88-920010075. (as cited in RTECS, 2000)
Suter, G.W. and C.L. Tsao. 1996. Toxicological benchmarks for screening potential
contaminants of concern for effects on aquatic biota: 1996 revision. Risk Assessment Program,
Health Sciences Research Division, Oak Ridge National Laboratory. ES/ER/TM-96/R2.
Swanson, M.B., G.A. Davis, L.E. Kincaid, T.W. Schultz, I.E. Bartmess, et al. 1997. "A
Screening Method for Ranking and Scoring Chemicals by Potential Human Health and
Environmental Impacts," Environmental Toxicology and Chemistry, Vol. 16, No. 2, pp. 372-
383, SET AC Press.
U.S. CFR. 1994. United States Code of Federal Regulations 29, 51910.1000. OSHA
(Occupational Safety and Health Administration), Dept. of Labor. Occupational safety and
health standards.
U.S. CFR. 2002. United States Code of Federal Regulations 40, Part 141.61. National Primary
Drinking Water Regulations. Maximum Contaminant Levels for organic contaminants.
Environmental Protection Agency, Water Programs.
U.S. Coast Guard, 1984-85. CHRIS - Hazardous Chemical Data. Vol. II. United States Coast
Guard, Dept. of Transportation. Washington D.C.: U.S. Govt. Printing Office.
U.S.DHHS. 1995. (United States Department of Health and Human Services). Toxicological
profile for poly cyclic aromatic hydrocarbons (PAHS), update. Agency for Toxic Substances and
Disease Registry, Public Health Service, U.S. DHHS.
E-S
-------
U.S. EPA, 2003a. ECOSAR Program. Risk Assessment Division (7403). EPIWINv. 3.10
(Estimations Programs Interface for Windows), Pollution Prevention Framework.
http://www.epa.gov/oppt/p2framework/docs/epiwin.htm. Run on line April, 2003 by Oak Ridge
National Laboratory.
U.S. EPA, 2003b. ECOSAR Program. Risk Assessment Division (7403). EPIWINv. 3.10
(Estimations Programs Interface for Windows), Pollution Prevention Framework.
http://www.epa.gov/oppt/p2framework/docs/epiwin.htm. Run on line September, 2003 by U.S.
EPA.
U.S. EPA, 2002. ECOTOX User Guide: ECOTOXicology Database System. Version 3.0.
Available: http:/www.epa.gov/ecotox/. Retrieved online April, 2003.
U.S. EPA, 1998a. Printed Wiring Board Cleaner Technologies Substitutes Assessment: Making
Holes Conductive. Chapter Three. Design for the Environment. EPA 744R-98-004a and EPA
744R-98-004b.
U.S. EPA, 1998b. Integrated Risk Information System (IRIS). Zinc and Compounds. Office of
Research and Development, National Center for Environmental Assessment, U.S. Environmental
Protection Agency, http://www.epa.gov/iris/. Retrieved online April, 2003.
U.S. EPA, 1997. Integrated Risk Information System (IRIS). Benzo[k]fluoranthene. Office of
Research and Development, National Center for Environmental Assessment, U.S. Environmental
Protection Agency, http://www.epa.gov/iris/. Retrieved online April, 2003.
U.S.EPA, 1996. Cleaner Technologies Substitutes Assessment, A Methodology and Resource
Guide. Chapter 5. Chemical and process information. Environmental hazards summary, p. 5-59.
Pollution Prevention Information Clearinghouse (PPIC), U.S. Environmental Protection Agency,
Washington, D.C., Decmeber.
U.S. EPA, 1994. Estimated Aquatic Toxicity Values, Screen Reclamation Chemicals. Draft
Cleaner Technologies Substitutes Assessment (CTSA): Screen Reclamation, Chapter Two.
Design for the Environment. EPA 744R-94-005a, September 1994.
U.S. EPA, 1993. Provisional Guidance for Quantitative Risk Assessment of Poly cyclic
Aromatic Hydrocarbons. EPA/600/R-93/089. United States Environmental Protection Agency.
USGS/CERC. 2003. (U.S. Geological Survey; Columbia Environmental Research Center,
Columbia Missouri). Acute toxicity database. Biological Resources Division, Central Region,
Columbia Environmental Research Center.
LINK"http://sss.cerc.usgs.gov/"http://www.cerc.usgs.gov/. Retrieved online April, 2003.
Van der Hoeven, J.C., and G.T. Welboren. 1987. Assessment of the acute toxicity of Dowanol-
DPnB in Poecilia reticulata. NOTOX Report No. not reported. July 1987. Unpublished report.
E-9
-------
Weiss, G. 1980. Hazardous Chemicals Data Book. Fluosilicic Acid. G. Weiss (ed.) Park
Ridge, NJ: Noyes Data Corporation, p. 474.
WHO, 1999. WHO Food Additives Series: 42. Aliphatic acyclic and alicyclic terpenoid tertiary
alcohols and structurally related substances. International Programme on Chemical Safety,
World Health Organization, Safety Evaluation of Certain Food Additives. Prepared by the Fifty-
first meeting of the Joint FAO/WHO Expert Committee on Food Additives (JECFA). Geneva,
Switzerland.
Wilde, E.W., R. Soracco, L. Mayack, R. Shealy, et al. 1983a. Comparison of chlorine and
chlorine dioxide toxicity to fathead minnows and bluegill. Water Res. 17: 1327-1331. (as cited
in ECOTOX).
Wilde, E.W., R. Soracco, L. Mayack, R. Shealy, and T. Broadwell, 1983b. Acute Toxicity of
Chlorine and Bromine to Fathead Minnows and Bluegills. Bull. Environ. Contam. Toxicol.
31(3):309-314 (as cited in ECTOX).
E.4 GLOSSARY OF TOXICITY COMPARISON TERMS
CC (Concentration of concern)
Calculated aquatic toxicity value derived by dividing the lowest chronic value in mg/L by ten.
EC50 (Effective Concentration 50)
A calculated dose of a substance which is expected to cause an effect on 50% of a defined
animal population.
LDLo (Lethal Dose Low)
The lowest dose (other than LD50) of a substance introduced by any route, other than inhalation,
over any given period of time in one or more divided portions and reported to have caused death
in humans or animals.
LD50 (Lethal Dose 50)
A calculated dose of a substance which is expected to cause the death of 50% of a defined
experimental animal population.
LC50 (Lethal Concentration 50)
A calculated concentration of a substance in air or water, which is expected to cause the death of
50% of a defined experimental animal population.
LOAEL (Lowest observable adverse effect level)
MCL (Maximum Contaminant Level)
E-10
-------
The highest level of a contaminant that is allowed in drinking water. It is a national primary
drinking water regulation established by EPA.
NOEL (No observable effect level)
NOAEL (No observable adverse effect level)
OEL (Occupational exposure limit)
The concentration of a substance in air, that a worker may safely be exposed to on a regular
basis, usually for an 8 hour workday.
PEF (Potency equivalency factor)
A calculated carcinogenicity comparison of a substance, relative to (in this case benzo(a)pyrene)
another substance.
PEL (Permissible exposure limit)
The 8-hour time weighted average for the concentration of a substance in air that must not be
exceeded during any 8-hour workshift of a 40 hour work week.
TDLo (Toxic Dose Low)
The lowest dose of a substance reported to produce any toxic effect in humans or tumorigenic,
reproductive, or multiple effects in animals.
TLm (Median tolerance limit)
A calculated dose which is expected to cause an effect (includes death) in 50% of a test
population.
WOE (Weight of evidence)
Classification of relevance and quality of studies used to make a determination of
carcinogenicity.
E-ll
-------
Table E-1. Toxicity data for potentially toxic LFSP chemicals
Cas#
1746-01-6
51207-31-9
121-14-2
91-57-6
56-49-5
3697-24-3
83-32-9
208-96-8
75-07-0
64-19-7
67-64-1
98-86-2
107-02-8
No CAS #
7429-90-5
7664-41-7
6484-52-2
120-12-7
7440-36-0
7440-38-2
7440-39-3
20-02-0
71-43-2
56-55-3
50-32-8
56832-73-6
205-99-2
191-24-2
207-08-9
100-44-7
7440-41-7
117-81-7
7440-69-9
1303-96-4
No CAS #
7440-42-8
Material (flow)
2,3,7,8-TCDD (2,3,7,8-Tetrachlorodibenzo-p-Dioxin)
2,3,7,8-TCDF (2,3,7,8-Tetrachlorodibenzo Furan)
2,4-Dinitrotoluene
2-Methylnaphthalene
3-Methylcholanthrene
5-Methyl chrysene (category: PAH)
Acenaphthene (category: PAH)
Acenaphthylene (category: PAH)
Ethanal (Acetaldehyde)
Acetic acid
Acetone
Acetophenone
Acrolein
Aluminium (AI3+)
Aluminum (Al)
Ammonia
Ammonium nitrate
Anthracene (category: PAH)
Antimony (Sb)
Arsenic (As)
Barium (Ba)
Barium compounds [Barium (Ba++)]
Benzene
Benzo{a}anthracene (category: PAH)
Benzo{a}pyrene
Benzo{b,j,k}fluoranthene (category: PAH)
Benzo{b}fluoranthene
Benzo{g,h,l}perylene (category: PAH)
Benzo{k}fluoranthene
Benzyl chloride
Beryllium (Be)
Bis(2-ethylhexyl)phthalate[Di(2-ethylhexyl)phthalate]
Bismuth
Borax
Boron (B III)
Boron (B)
oral SF
(mg/kg- day)
1
1 .50E+05
1 .50E+04
0.68
--
--
X
--
X
X
--
X
--
X
--
X
--
X
--
1.5
--
X
0.055
0.73
7.3
X
0.73
X
-
0.17
4.3
X
-
--
--
--
inhalSF
(mg/kg- day)"1
1 .50E+05
1 .50E+04
X
--
--
--
--
--
7.70E-03
--
--
--
--
--
--
--
--
--
50
--
--
0.029
0.31
3.1
--
0.31
--
-
X
8.4
X
-
--
--
--
WOE
(EPA&
IARC) (a)
1
3
B2
--
--
2B
--
D
2B
--
D
--
C,3
--
SARD
--
SAR1
--
A
--
D
A,1
B2
B2.2A
B2
B2
D
B2
B2,3
X
B2.2B
-
--
--
--
oral NOAEL
(b) (mg/kg-
day)
9.00E-08
--
0.2
--
X
--
175
--
125
195
100
423
--
--
60
34
1000
X
8.00E-04
0.21
0.21
1
--
--
--
--
--
-
--
X
50
3,243
--
8.8
8.8
inhal
NOAEL (b)
(mg/m3)
X
--
X
--
X
--
X
--
300
X
X
X
--
--
X
40
X
X
X
X
X
1.15
--
--
--
--
--
-
--
X
50
-
--
X
X
oral
LOAEL
(b,c)
(mg/kg-
day)
X
--
X
--
2.86
--
350
--
X
X
X
X
--
--
X
X
X
0.35
X
X
X
10
--
--
--
--
--
-
--
X
X
-
--
X
X
inhal
LOAEL
(b,c)
(mg/m3)
X
--
X
--
X
--
X
--
X
X
X
X
--
--
X
X
X
X
X
X
X
98
--
--
--
--
--
-
--
5.50E-04
X
-
--
X
X
fish LC50
(mg/L)
XX
XX
24
XX
XX
XX
XX
XX
34
XX
720
XX
XX
3.6
11
2
0.01
14.4
14.4
580
200
19
XX
XX
XX
XX
XX
1000
XX
2
1
5
XX
113
113
fish NOEL
(mg/L)
XX
XX
6
XX
XX
XX
XX
XX
9
XX
180
XX
XX
0.36
3.3
9.00E-02
--
1.6
2.1
50
10
4
XX
XX
XX
XX
XX
0.006
XX
0.2
0.08
0.5
XX
27
27
E-12
-------
Table E-1. Toxicity data for potentially toxic LFSP chemicals
Cas#
7726-95-6
75-25-2
7440-43-9
20-04-2
75-15-0
630-08-0
75-69-4
76-14-2
75-71-8
75-72-9
7782-50-5
1341-24-8
108-90-7
16065-83-1
7440-47-3
18540-29-9
218-01-9
7440-48-4
7440-50-8
No CAS #
98-82-8
57-12-5
53-70-3
25321-22-6
107-06-2
75-09-2
77-78-1
57-97-6
74-84-0
75-00-3
100-41-4
106-93-4
206-44-0
86-73-7
16984-48-8
No CAS #
Material (flow)
Bromine
Bromoform
Cadmium (Cd)
Cadmium cmpds (as CdCI2) [Cadmium (Cd++)]
Carbon disulfide
Carbon monoxide (CO)
CFC 1 1 (Trichlorofluoromethane)
CFC 114 (1 ,2-dichlorotetrafluoroethane)
CFC 12 (Dichlorodifluoromethane)
CFC 13 (Dichlorotrifluoromethane)
Chlorine (CI2)
Chloroacetophenone
Chlorobenzene
Chromium (Cr III)
Chromium (Cr)
Chromium, hexavalent (CrVI)
Chrysene (category: PAH)
Cobalt (Co)
Copper (Cu)
Copper (Cu+, Cu++)
Cumene
Cyanide (CN)
Dibenzo{a,h}anthracene
Dichlorobenzene (mixed isomers)
Ethylene dichloride (Dichloroethane)
Dichloromethane (Methylene chloride)
Dimethyl sulfate
Dimethylbenzanthracene
Ethane
Ethyl chloride
Ethylbenzene
Ethylene dibromide
Fluoranthene (category: PAH)
Fluorene (category: PAH)
Fluoride
Fluorides (F-)
oral SF
(mg/kg- day)
1
--
7.90E-03
X
X
--
--
--
--
--
--
--
--
X
X
X
X
7.30E-03
--
X
--
X
X
7.3
X
9.10E-02
7.50E-03
X
--
--
X
X
85
X
X
--
--
inhalSF
(mg/kg- day)"1
--
3.90E-03
6.1
--
--
--
--
--
--
--
--
--
--
--
--
41
3.10E-03
--
--
--
--
--
3.1
X
9.10E-02
1 .65E-03
X
--
--
X
X
7.60E-01
X
X
--
--
WOE
(EPA&
IARC) (a)
--
B2
B1,1
B1.2A
--
--
--
--
--
--
--
--
SARD
D
1
A,1
X
--
D
--
SARD
D
B2
SARD
B2.2B
B2.2B
B1.2A
--
--
3
SARD
B2
D
D
--
--
oral NOAEL
(b) (mg/kg-
day)
--
17.9
X
5.00E-03
X
X
X
2.73E+02
15
--
14
--
12.5
1468
--
2.5
--
--
5.30E-01
5.30E-01
154
10.8
--
X
18
155
--
X
--
X
136
--
125
125
--
6.00E-02
inhal
NOAEL (b)
(mg/m3)
--
X
X
X
10
114.5
X
X
X
--
X
--
377
X
--
X
--
--
X
X
537
X
--
610.4
221
796
--
X
--
3600
2370
--
X
X
--
X
oral
LOAEL
(b,c)
(mg/kg-
day)
--
X
4.00E-02
X
X
X
349
X
X
--
X
--
X
X
--
X
--
--
X
X
X
X
--
X
X
X
--
X
--
X
X
--
X
X
--
X
inhal
LOAEL
(b,c)
(mg/m3)
--
X
2.20E-02
X
X
55
X
X
X
--
X
--
X
X
--
X
--
--
X
X
X
X
--
X
X
X
--
1 .40E-02
--
X
X
--
X
X
--
X
fish LC50
(mg/L)
XX
XX
0.001
0.1
694
XX
XX
XX
XX
XX
0.34
XX
17
3.3
52
22.6
XX
XX
1 .40E-02
1 .40E-02
6
56
XX
1
136
330
XX
XX
XX
16
11
XX
XX
XX
--
--
fish NOEL
(mg/L)
XX
XX
0.001
--
174
XX
XX
XX
XX
XX
0.02
XX
2
0.33
5.2
2.23
XX
XX
4.00E-03
4.00E-03
0.49
5.7
XX
0.05
34
83
XX
XX
XX
4
1
XX
XX
XX
--
--
E-13
-------
Table E-1. Toxicity data for potentially toxic LFSP chemicals
Cas#
7782-41-4
16872-11-0
16961-83-4
(d)
(d)
(d)
(d)
(d)
(d)
50-00-0
No CAS #
75-63-8
75-45-6
110-54-3
7647-01-0
7664-39-3
74-90-8
7783-06-4
193-39-5
1309-36-0
78-59-1
67-63-0
7439-92-1
20-11-1
NA
7439-96-5
7439-97-6
no CAS#
74-82-8
67-56-1
74-83-9
74-87-3
78-93-3
60-34-4
Material (flow)
Fluorine (F2)
Fluoroboric acid
Fluorosilicic acid
Flux A (d)
Flux B (d)
Flux C (d)
Flux D (d)
Flux E (d)
Flux F (d)
Formaldehyde (CH2O)
Light Fuel Oil (#2, distillate and diesel)
Halon 1301
HCFC 22 (Chlorodifluoromethane)
Hexane
Hydrochloric acid
Hydrofluoric acid (Hydrogen fluoride)
Hydrogen Cyanide
Hydrogen Sulfide
lndeno{1,2,3-cd}pyrene (category: PAH)
Iron pyrite
Isophorone
Isopropyl alcohol
Lead (Pb)
Lead compounds (as PbCI2) [Lead (Pb++, Pb4+)]
Liquified petroleum gas (LPG)
Manganese
Mercaptan
Mercury (Hg)
Mercury cmpds (as HgCI2) [Mercury (Hg+, Hg++)]
Metals, unspecified
Methane (natural gas)
Methanol
Methyl bromide (bromomethane)
Methyl chloride (Chloromethane)
Methyl ethyl ketone
Methyl hydrazine
oral SF
(mg/kg- day)
1
--
-
-
-
-
--
-
-
-
X
--
--
--
--
X
--
X
--
7.30E-01
-
9.50E-04
X
X
X
--
X
X
X
--
X
X
1 .30E-02
X
3
inhalSF
(mg/kg- day)"1
--
-
-
-
-
--
-
-
-
4.50E-02
--
--
--
--
X
--
X
--
3.10E-01
-
X
X
X
X
--
X
X
X
--
X
--
6.30E-03
X
17.2
WOE
(EPA&
IARC) (a)
--
-
-
-
-
--
-
-
-
B1.2A
--
--
--
--
3
--
SARD
--
B2
-
C
1
B2.2B
B2.2B
--
D
D,3
C
--
SARD
C,3
C,3
D
A3
oral NOAEL
(b) (mg/kg-
day)
6.00E-02
-
-
-
450
--
-
-
-
15
--
--
X
X
X
--
10.8
3.1
--
-
150
230
--
0.14
X
X
--
500
0.4
X
125
--
inhal
NOAEL (b)
(mg/m3)
X
-
-
-
200
--
-
-
-
0.6
--
--
5,260
X
15
--
X
X
--
1
X
268.3
--
--
X
6.00E-03
X
--
130
4.3
1138.4
8047
--
oral
LOAEL
(b,c)
(mg/kg-
day)
X
0.77
0.77
-
1000
--
-
-
-
X
--
--
X
X
X
--
30
X
--
-
X
X
0.014
0.014
--
X
X
0.226
--
X
X
X
X
--
inhal
LOAEL
(b,c)
(mg/m3)
X
-
-
-
810
--
-
-
-
X
--
X
73
X
--
7.07
15
--
-
X
X
0.011
0.011
--
0.15
9.00E-03
X
--
X
X
1550
X
--
fish LC50
(mg/L)
100
1000
100
900
930
XX
0.5
1000
1000
24
XX
XX
XX
2.5
19
265
1,385
XX
XX
1000
XX
8,623
31.5
5
2600
--
0.155
0.155
XX
29,400
11
550
3,220
XX
fish NOEL
(mg/L)
10
20
10
90
100
XX
0.05
100
100
6
XX
XX
XX
0.25
0.95
13
346
XX
XX
-
XX
2,156
0.004
0.26
260
--
0.005
0.005
XX
7,350
3
138
805
XX
E-14
-------
Table E-1. Toxicity data for potentially toxic LFSP chemicals
Cas#
80-62-6
1634-04-4
7439-98-7
91-20-3
7440-02-0
20-14-4
14797-55-8
no CAS#
10024-97-2
NA
NA
109-66-0
85-01-8
108-95-2
7723-14-0
123-38-6
115-07-1
129-00-0
7440-20-2
7782-49-2
7440-21-3
7440-22-4
7681-52-9
7440-24-6
100-42-5
7446-09-5
no CAS#
7664-93-9
127-18-4
7440-28-0
7440-31-5
7440-32-6
108-88-3
71-55-6
67-66-3
7440-62-2
Material (flow)
Methyl methacrylate
Methyl tert butyl ether (MTBE)
Molybdenum (Mo)
Naphthalene
Nickel (Ni)
Nickel cmpds (as NICI2) [Nickel (Ni++, Ni3+)]
Nitrates
Nitrogen Oxides (NOx)
Nitrous oxide
Particulate matter (PM-10) [Particulates < 10 microns]
Particulate matter, total (PM)
Pentane
Phenanthrene (category: PAH)
Phenol
Phosphorus
Propionaldehyde
Propylene (Propene)
Pyrene (category: PAH)
Scandium (Sc)
Selenium (Se)
Silicon (Si)
Silver
Sodium Hypochlorite
Strontium (Sr)
Styrene
Sulfur dioxide
Sulfur oxides (SOx)
Sulfuric acid
Tetrachloroethylene (Perchloroethylene)
Thallium (Tl)
Tin (Sn)
Titanium
Toluene
Trichloroethane (1 ,1 ,1-trichloroethane)
Trichloromethane (Chloroform)
Vanadium (V)
oral SF
(mg/kg- day)
1
X
X
--
X
X
X
--
--
--
--
--
X
X
X
X
X
X
X
--
X
--
X
X
--
X
X
--
X
5.20E-02
--
--
X
X
--
6.10E-03
--
inhalSF
(mg/kg- day)"1
X
X
--
X
X
X
--
--
--
--
--
X
X
X
X
X
X
X
--
X
--
X
X
--
X
X
--
X
2.00E-03
--
--
X
X
--
8.10E-02
--
WOE
(EPA&
IARC) (a)
SARD
SARD
--
C
A
A,1
--
--
--
--
--
D
D
D,3
D
SAR3
SARD
D
--
D
--
D
3
--
C,2B
3
--
1
B2.2B
--
--
C
D,3
--
B2.2B
--
oral NOAEL
(b) (mg/kg-
day)
7.5
100
X
71
5
--
1.6
--
--
--
--
--
--
60
1 .50E-02
X
X
75
--
1 .50E-02
--
X
2.1
190
100
X
--
X
14
--
--
X
100
2.50E+02
X
3.00E-03
inhal
NOAEL (b)
(mg/m3)
111.7
2880
X
X
X
--
X
--
--
--
--
--
--
X
X
200
9375
X
--
X
--
X
X
X
565
0.104
--
0.1
740.2
--
--
0.8
411.1
1.21E+03
X
X
oral
LOAEL
(b,c)
(mg/kg-
day)
X
X
0.14
X
X
--
X
--
--
--
--
--
--
X
X
X
X
X
--
X
--
1 .40E-02
X
X
X
X
--
X
X
--
--
1146
X
X
12.9
X
inhal
LOAEL
(b,c)
(mg/m3)
X
X
X
9.3
X
--
X
--
--
--
--
--
--
X
X
X
X
X
--
X
--
X
X
X
X
X
--
X
X
--
--
X
X
X
X
X
fish LC50
(mg/L)
259
786
157
6
2.48
27
2,213
XX
XX
XX
XX
XX
XX
34
0.02
44
5
XX
XX
4.9
XX
4.00E-03
0.53
210
4
XX
XX
31
17
XX
626
--
34
48
71
4
fish NOEL
(mg/L)
65
197
0.125
0.59
0.09
1
213
XX
XX
XX
XX
XX
XX
8
--
11
1
XX
XX
0.1
XX
0.001
0.05
20
0.44
XX
XX
2
2
XX
62.6
--
4
7
18
0.67
E-15
-------
Table E-1. Toxicity data for potentially toxic LFSP chemicals
Cas#
108-05-4
1330-20-7
7440-66-6
No CAS #
7733-02-0
Key:
Material (flow)
Vinyl acetate
Xylene (C24H30) [mixed isomers]
Zinc (Zn)
Zinc (Zn++)
Zinc sulfate
oral SF
(mg/kg- day)
1
X
X
X
--
-
(a)=See Table 3-72 in Section 3.2.1 1.1 for a description of WOE classifications.
inhalSF
(mg/kg- day)"1
X
X
X
--
-
(b)=Only lowest value of the NOAEL (or LOAEL/10) is used to calculate chronic, non-cancer effects.
(c)=LOAEL only needed if no NOAEL found.
(d)=Flux material names and CAS#s have been withheld to protect confidentiality.
WOE
(EPA&
IARC) (a)
SARD
D
D
--
D
oral NOAEL
(b) (mg/kg-
day)
100
179
0.9
--
-
XX=Aquatic toxicity data not needed because there are no waterborne releases of this chemical in the LFSP inventories.
inhal
NOAEL (b)
(mg/m3)
176
X
X
--
-
oral
LOAEL
(b,c)
(mg/kg-
day)
X
X
1
--
1
inhal
LOAEL
(b,c)
(mg/m3)
X
X
X
--
-
X=Data not needed because other data are provided to calculate impact score (e.g., LOAEL not needed if NOAEL provided, and WOE used if SF not available).
SARO=Not a probable carcinogen based on structure-activity relationship (SAR) evaluation.
SAR1=Possible carcinogen based on SAR evaluation.
- - =No data available, defaulted to mean hazard value (see Section 3.1 .2.12 for an explanation of hazard values).
Sources:
fish LC50
(mg/L)
100
13
9.00E-02
14
1.27
Oral and inhalation slope factors (SF): Integrated Risk Information System (IRIS) or Health Effects Assessment Summary Tables (HEAST) (EPA, 1994) as cited in Risk
Assessment Information System (RAIS): http://risk.lsd. ornl.gov/rap_hp.shtml.
Weight of Evidence (WOE): IRIS Web site (http://www.epa.gov/IRIS).
Oral no observable adverse effect level (NOAEL), inhalation NOAEL, oral lowest obserable adverse effect level (LOAEL) and inhalation LOAEL:
IUCLID, 1996; HEAST, 1994; Kincaid and Geibig, 1998; EPA, 2000a; SRC, 2000; EPA, 2000b; Geibig and Swanson, 2000; Sax and Lewis, 1987; NIOSH, 1978; EPA, 1984;
and EPA, 1987.
Fish LC50 and fish NOAEL: EPA, 2001 ; HSDB; Davis et al. 1994, Appendix E; and Geiger et al., 1984, 1985, 1986, 1988, 1990.
Sources associated with data collected from ORNL (May, 2003) are listed in this Appendix under the References Section E.3.
fish NOEL
(mg/L)
25
1
0.036
0.8
-
E-16
-------
Table E-2. Toxicity hazard values (HV) for potentially toxic chemicals in the LFSP
CAS#
1746-01-6
51207-31-9
121-14-2
91-57-6
56-49-5
3697-24-3
83-32-9
208-96-8
75-07-0
64-19-7
67-64-1
98-86-2
107-02-8
No CAS #
7429-90-5
7664-41-7
120-12-7
7440-36-0
7440-38-2
7440-39-3
20-02-0
71-43-2
56-55-3
50-32-8
56832-73-6
205-99-2
191-24-2
207-08-9
100-44-7
7440-41-7
117-81-7
7440-69-9
1303-96-4
No CAS #
7440-42-8
7726-95-6
75-25-2
No CAS #
7440-43-9
20-04-2
75-15-0
630-08-0
75-69-4
76-14-2
75-71-8
75-72-9
Material (flow)
2,3,7,8-TCDD (2,3,7,8-Tetrachlorodibenzo-p-Dioxin)
2,3,7,8-TCDF (2,3,7,8-Tetrachlorodibenzo Furan)
2,4-Dinitrotoluene
2-Methylnaphthalene
3-Methylcholanthrene
5-Methyl chrysene (category: PAH)
Acenaphthene (category: PAH)
Acenaphthylene (category: PAH)
Ethanal (Acetaldehyde)
Acetic acid
Acetone
Acetophenone
Acrolein
Aluminium (AI3+)
Aluminum (Al)
Ammonia
Anthracene (category: PAH)
Antimony (Sb)
Arsenic (As)
Barium (Ba)
Barium compounds [Barium (Ba++)]
Benzene
Benzo{a}anthracene (category: PAH)
Benzo{a}pyrene
Benzo{bj,k}fluoranthene (category: PAH)
Benzo{b}fluoranthene
Benzo{g,h,l}perylene (category: PAH)
benzo{k}fluoranthene
Benzyl chloride
Beryllium (Be)
Bis(2-ethylhexyl)phthalate[Di(2-ethylhexyl)phthalate]
Bismuth
Borax
Boron (BIN)
Boron (B)
Bromine
Bromoform
BSA (bismuth-tin-silver) alloy*
Cadmium (Cd)
Cadmium cmpds (as CdCI2) [Cadmium (Cd++)]
Carbon disulfide
Carbon monoxide (CO)
CFC 1 1 (Trichlorofluoromethane)
CFC 1 14 (1 ,2-dichlorotetrafluoroethane)
CFC 12 (Dichlorodifluoromethane)
CFC 13 (Dichlorotrifluoromethane)
Cancer HV
2.11E+05
2.11E+04
9.58E-01
1 .OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
O.OOE+00
4.53E-03
1 .OOE+00
O.OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
O.OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
2.94E+01
1 .OOE+00
O.OOE+00
7.75E-02
1 .03E+00
1 .03E+01
1 .OOE+00
1 .03E+00
O.OOE+00
1 .OOE+00
2.39E-01
6.06E+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
1.11E-02
9.90E-01
3.59E+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
Non-cancer
HV
1 .56E+08
1 .OOE+00
7.00E+01
1 .OOE+00
4.90E+01
1 .OOE+00
8.00E-02
1 .OOE+00
2.29E-01
7.18E-02
1 .40E-01
3.31 E-02
1 .OOE+00
1 .OOE+00
2.33E-01
1 .72E+00
1 .40E-02
4.00E+02
1 .75E+04
6.67E+01
6.67E+01
5.97E+01
1 .OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
1 .25E+06
1 .37E+00
4.32E-03
1 .OOE+00
1 .59E+00
1 .59E+00
1 .OOE+00
7.82E-01
1 .OOE+02
3.12E+04
2.80E+03
6.87E+00
6.00E-01
4.01 E-01
5.13E-02
9.33E-01
1 .OOE+00
Aquatic
ecotoxicity
HV
not searched
not searched
1 .68E+00
not searched
not searched
not searched
not searched
not searched
1.16E+00
not searched
5.58E-02
not searched
not searched
1 .77E+01
3.42E+00
5.56E+01
8.88E+03
4.15E+00
3.57E+00
1 .20E-01
5.13E-01
2.27E+00
not searched
not searched
not searched
not searched
not searched
6.50E+02
not searched
3.18E+01
7.34E+01
1 .27E+01
not searched
3.62E-01
3.62E-01
not searched
not searched
not searched
2.85E+04
2.47E+02
5.79E-02
not searched
not searched
not searched
not searched
not searched
E- 17
-------
Table E-2. Toxicity hazard values (HV) for potentially toxic chemicals in the LFSP
CAS#
7782-50-5
1341-24-8
108-90-7
16065-83-1
7440-47-3
18540-29-9
218-01-9
7440-48-4
7440-50-8
No CAS #
98-82-8
57-12-5
53-70-3
25321-22-6
107-06-2
75-09-2
77-78-1
57-97-6
74-84-0
75-00-3
100-41-4
106-93-4
206-44-0
86-73-7
16984-48-8
No CAS #
7782-41-4
16872-11-0
16961-83-4
(d)
(d)
(d)
(d)
(d)
(d)
50-00-0
No CAS #
75-63-8
75-45-6
110-54-3
7647-01-0
7664-39-3
74-90-8
7783-06-4
193-39-5
1309-36-0
Material (flow)
Chlorine (CI2)
Chloroacetophenone
Chlorobenzene
Chromium (Cr III)
Chromium (Cr)
Chromium, hexavalent (Cr VI)
Chrysene (category: PAH)
Cobalt (Co)
Copper (Cu)
Copper (Cu+, Cu++)
Cumene
Cyanide (CN)
Dibenzo{a,h}anthracene
Dichlorobenzene (mixed isomers)
Ethylene dichloride (Dichloroethane)
Dichloromethane (Methylene chloride)
Dimethyl sulfate
Dimethylbenzanthracene
Ethane
Ethyl chloride
Ethylbenzene
Ethylene dibromide
Fluoranthene (category: PAH)
Fluorene (category: PAH)
Fluoride
Fluorides (F-)
Fluorine (F2)
Fluoroboric acid
Fluorosilicic acid
Flux A (d)
Flux B (d)
Flux C (d)
Flux D (d)
Flux E (d)
Flux F (d)
Formaldehyde (CH2O)
Fuel Oil, light (#2, distillate and diesel)
Halon 1301
HCFC 22 (Chlorodifluoromethane)
Hexane
Hydrochloric acid
Hydrofluoric acid (Hydrogen fluoride)
Hydrogen Cyanide
Hydrogen Sulfide
lndeno{1,2,3-cd}pyrene (category: PAH)
Iron pyrite
Cancer HV
1 .OOE+OO
1 .OOE+OO
O.OOE+00
O.OOE+00
1 .OOE+OO
2.41 E+01
1 .03E-02
1 .OOE+OO
O.OOE+00
1 .OOE+OO
O.OOE+00
O.OOE+00
1 .03E+01
O.OOE+00
1.28E-01
1 .06E-02
1 .OOE+OO
1 .OOE+OO
1 .OOE+OO
O.OOE+00
O.OOE+00
1 .20E+02
O.OOE+00
O.OOE+00
1 .OOE+OO
1 .OOE+OO
1 .OOE+OO
1 .OOE+OO
1 .OOE+OO
1 .OOE+OO
1 .OOE+OO
1 .OOE+OO
1 .OOE+OO
1 .OOE+OO
1 .OOE+OO
2.65E-02
1 .OOE+OO
1 .OOE+OO
1 .OOE+OO
1 .OOE+OO
O.OOE+00
1 .OOE+OO
O.OOE+00
1 .OOE+OO
1 .03E+00
1 .OOE+OO
Non-cancer
HV
1 .OOE+OO
1 .OOE+OO
1.12E+00
9.54E-03
1 .OOE+OO
5.60E+00
1 .OOE+OO
1 .OOE+OO
2.64E+01
2.64E+01
1 .28E-01
1 .30E+00
1 .OOE+OO
1.13E-01
7.78E-01
9.03E-02
1 .OOE+OO
4.91 E+04
1 .OOE+OO
1 .91 E-02
1 .03E-01
1 .OOE+OO
1.12E-01
1.12E-01
1 .OOE+OO
2.33E+02
2.33E+02
1 .82E+02
1 .82E+02
1 .OOE+OO
3.43E-01
1 .OOE+OO
1 .OOE+OO
1 .OOE+OO
1 .OOE+OO
1.14E+02
1 .OOE+OO
1 .OOE+OO
1.31 E-02
9.41 E+00
4.58E+00
1 .OOE+OO
1 .30E+00
4.52E+00
1 .OOE+OO
6.87E+01
Aquatic
ecotoxicity
HV
2.67E+02
not searched
3.40E+00
1 .93E+01
1 .22E+00
2.84E+00
not searched
not searched
2.73E+03
2.73E+03
1 .21 E+01
1.12E+00
not searched
1 .03E+02
2.96E-01
1.22E-01
not searched
not searched
not searched
2.51 E+00
6.14E+00
not searched
not searched
not searched
2.00E+00
2.00E+00
6.36E-01
2.20E-01
6.36E-01
7.07E-02
6.55E-02
1 .OOE+OO
1 .27E+02
6.36E-02
6.36E-02
1 .68E+00
not searched
not searched
not searched
2.54E+01
5.40E+00
3.93E-01
2.90E-02
not searched
not searched
1 .03E-01
E- 18
-------
Table E-2. Toxicity hazard values (HV) for potentially toxic chemicals in the LFSP
CAS#
78-59-1
67-63-0
7439-92-1
20-11-1
No CAS #
7439-96-5
7439-97-6
no CAS#
74-82-8
67-56-1
74-83-9
74-87-3
78-93-3
60-34-4
80-62-6
1634-04-4
7439-98-7
91-20-3
7440-02-0
20-14-4
14797-55-8
no CAS#
10024-97-2
NA
NA
109-66-0
85-01-8
108-95-2
7723-14-0
123-38-6
115-07-1
129-00-0
7440-20-2
7782-49-2
7440-21-3
7440-22-4
no CAS #
no CAS #
no CAS #
no CAS #
7681-52-9
7440-24-6
100-42-5
7446-09-5
no CAS#
7664-93-9
Material (flow)
Isophorone
Isopropyl alcohol
Lead (Pb)
Lead compounds (as PbCI2) [Lead (Pb++, Pb4+)]
Liquified petroleum gas (LPG)
Manganese
Mercury (Hg)
Mercury cmpds (as HgCI2) [Mercury (Hg+, Hg++)]
Methane (natural gas)
Methanol
Methyl bromide (bromomethane)
Methyl chloride (Chloromethane)
Methyl ethyl ketone
Methyl hydrazine
Methyl methacrylate
Methyl tert butyl ether (MTBE)
Molybdenum (Mo)
Naphthalene
Nickel (Ni)
Nickel cmpds (as NICI2) [Nickel (Ni++, Ni3+)]
Nitrates
Nitrogen Oxides (NOx)
Nitrous oxide
Particulate matter (PM-10) [Particulates < 10 microns]
Particulate matter, total (PM)
Pentane
Phenanthrene (category: PAH)
Phenol
Phosphorus
Propionaldehyde
Propylene (Propene)
Pyrene (category: PAH)
Scandium (Sc)
Selenium (Se)
Silicon (Si)
Silver
SAC (tin-silver-copper) alloy*
SABC (tin-silver-bismuth-copper) alloy*
SnCu (in-copper) alloy*
SnPb (tin-lead) alloy*
Sodium Hypochlorite
Strontium (Sr)
Styrene
Sulfur dioxide
Sulfur oxides (SOx)
Sulfuric acid
Cancer HV
1 .34E-03
1 .OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
O.OOE+00
O.OOE+00
1 .OOE+00
1 .OOE+00
O.OOE+00
1 .OOE+00
1 .83E-02
O.OOE+00
1 .01 E+01
O.OOE+00
O.OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
1 .OOE+00
O.OOE+00
O.OOE+00
1 .OOE+00
O.OOE+00
1 .OOE+00
O.OOE+00
9.55E-01
9.70E-01
9.92E-01
1 .OOE+00
O.OOE+00
1 .OOE+00
1 .OOE+00
O.OOE+00
1 .OOE+00
1 .OOE+00
Non-cancer
HV
9.33E-02
2.56E-01
6.24E+04
6.24E+04
1 .OOE+00
1 .OOE+02
1.14E+04
6.19E+02
1 .OOE+00
5.28E-01
3.50E+01
6.03E-02
1.12E-01
1 .OOE+00
1 .87E+00
1 .40E-01
1 .OOE+03
1 .97E-01
2.80E+00
1 .OOE+00
8.75E+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
1 .OOE+00
2.33E-01
9.33E+02
3.43E-01
7.32E-03
1 .87E-01
1 .OOE+00
9.33E+02
1 .OOE+00
1 .OOE+04
3.91 E+02
2.51 E+02
1 .20E+00
2.31 E+04
6.67E+00
7.37E-02
1 .40E-01
6.60E+02
1 .OOE+00
6.87E+02
Aquatic
ecotoxicity
HV
not searched
4.66E-03
9.76E+02
1 .99E+01
2.45E-02
2.00E+00
9.39E+02
9.39E+02
not searched
1 .37E-03
3.54E+00
7.30E-02
1 .25E-02
not searched
1 .55E-01
5.11E-02
3.14E+01
1 .07E+01
5.33E+01
4.81 E+00
2.94E-02
not searched
not searched
not searched
not searched
not searched
not searched
1 .21 E+00
5.13E+03
9.14E-01
8.82E+00
not searched
not searched
4.40E+01
not searched
1 .01 E+04
not searched
not searched
not searched
not searched
1 .24E+02
3.12E-01
1 .50E+01
not searched
not searched
2.74E+00
E- 19
-------
Table E-2. Toxicity hazard values (HV) for potentially toxic chemicals in the LFSP
CAS#
127-18-4
7440-28-0
7440-31-5
7440-32-6
108-88-3
71-55-6
67-66-3
7440-62-2
108-05-4
1330-20-7
7440-66-6
No CAS #
7733-02-0
Key:
Material (flow)
Tetrachloroethylene (Tetrachloroethene, Perchloroethyli
Thallium (Tl)
Tin (Sn)
Titanium
Toluene
Trichloroethane (1 ,1 ,1-trichloroethane)
Trichloromethane (Chloroform)
Vanadium (V)
Vinyl acetate
Xylene (C24H30) [mixed isomers]
Zinc (Zn)
Zinc (Zn++)
Zinc sulfate
CAS=Chemical Abstracts Service.
Cancer HV
7.32E-02
1 .OOE+00
1 .OOE+00
1 .OOE+00
O.OOE+00
1 .OOE+00
4.76E-02
1 .OOE+00
O.OOE+00
O.OOE+00
O.OOE+00
1 .OOE+00
O.OOE+00
Non-cancer
HV
1 .OOE+00
1 .OOE+00
1 .OOE+00
8.58E+01
1 .67E-01
5.68E-02
1 .09E+01
4.67E+03
3.90E-01
7.82E-02
1 .56E+01
1 .OOE+00
1 .40E+02
HV=Hazard value. The methodologies for calculating the HVs are in Sections 3.2.11 through 3.2.13.
Aquatic
ecotoxicity
HV
3.40E+00
not searched
1 .02E-01
2.00E+00
1 .70E+00
1 .07E+00
5.63E-01
1 .20E+01
4.02E-01
5.79E+00
3.82E+02
6.63E+00
8.08E+01
not searched=aquatic ecotoxicity HV was not needed for the LFSP and thus toxicity data were not collected.
*HVs for each solder alloy were calculated as a weighted average of the HV for each comoponent metal in the alloy.
E-20
-------
Table E-3. Materials excluded from toxic classification
CAS#
NA
106-97-8
7440-70-2
124-38-9
NA
NA
NA
16887-00-6
NA
64-17-5
7440-59-7
7439-89-6
NA
8008-20-6
7727-37-9
74-98-6
NA
79-09-4
NA
NA
7440-23-5
NA
497-19-8
1310-73-2
14808-79-8
18496-25-8
14265-45-3
7704-34-9
NA
NA
Material (flow)
BOD (Biological Oxygen Demand)
Butane (n-C4H10)
Calcium (Ca)
Carbon Dioxide (CO2)
Carbonate ion [Carbonates (CO3~, HCO3-, CO2]
Charcoal
COD (Chemical Oxygen Demand)
Chloride (C1-)
Dissolved solids
Ethanol (Ethyl Alcohol)
Helium (He)
Iron (Fe)
Iron (Fe++, Fe3+)
Kerosene
Nitrogen
n-Propane [Propane (C3H8)]
Phosphates (PO4-3)
Propionic Acid
Salts (unspecified)
Sawdust
Sodium (Na)
Sodium (Na+)
Sodium carbonate (Na2CO3, soda ash)
Sodium hydroxide (NaOH)
Sulfates (SO4~)
Sulfides (S~)
Sulfites (SO3~)
Sulfur
Suspended Solids
TOCs (Total organic compounds)
Reason for exclusion
judgement
GRAS
judgement
judgement
judgement
judgement
judgement
judgement
judgement
GRAS
GRAS
judgement
judgement
judgement
GRAS
GRAS
judgement
GRAS
judgement
judgement
judgement
judgement
judgement
judgement
judgement
judgement
judgement
judgement
judgement
judgement
CAS#=Chemical Abstracts Service Registry Number
NA=not applicable
GRAS="Generally Regarded as Safe" according to the U.S. Food and Drug Admimistration
E-21
-------
Table E-4. FINAL TOXICITY DATA SELECTIONS FOR USE IN THE LCIA
Cas#
Material
For human and ecological endpoints:
207-08-9
16872-11-0
16961-83-4
(b)
(b)
(b)
(b)
(b)
1309-36-0
7733-02-0
7440-69-9
Benzo(k)fluoranthene
Fluoroboric acid
Fluorosilicic acid
Flux A
FluxB
FluxD
FluxE
FluxF
Iron pyrite
Zinc sulfate
Bismuth
Selection comments by UT
inhalation NOAEL not used as it is an occupational limit,
presumably including safety and/or uncertainty factors,
thus not consistent with a NOAEL; no supporting NOAEL
or LOAEL found; therefore, assume "no data"
inhalation NOAEL not used as it is an occupational limit,
presumably including safety and/or uncertainty factors;
therefore, not consistent with a NOAEL; no supporting
NOAEL or LOAEL found; therefore, assume "no data."
The NOAEL is actually a dermal "NOAEL/LOAEL" as
reported in the PWB CTSA (USEPA 1998a)
inhalation NOAEL not used as it is an occupational limit,
presumably including safety and/or uncertainty factors;
therefore, not consistent with a NOAEL; no supporting
NOAEL or LOAEL found; therefore, assume "no data."
The NOAEL is actually a dermal "NOAEL/LOAEL" as
reported in the PWB CTSA (USEPA 1998a)
the fish LC50 is based on same chemical name, but with
a different CAS# than we were provided
since the source of the LC50 data does not supply the
original data source of the toxicity value, we chose to use
the ECOSAR estimate
inhalation NOAEL not used as it is an occupational limit,
presumably including safety and/or uncertainty factors,
thus not consistent with a NOAEL; no supporting NOAEL
or LOAEL found; therefore, assume "no data"
chose fathead minnow data (1 .27 mg/L) instead of
rainbow trout data; based on our methodology (i.e.,
exclude trout data due to species sensitivity) (Swanson et
al. 1997)
for oral NOAEL, converted 227 g/d using 70 kg body
weight; didn't use inhalation NOAEL as it is a PEL
(occupational limit) which incorporates time-weighted
exposure and possibly safety and/or uncertainty factors
and thus not consistent with a NOAEL
For fish LC50 and fish NOEL endpoints only:
oral or inhal SF
-
-
..
-
WOE (EPA &
IARC)
B2
-
-
..
-
D
3243
oral
NOAEL
(mg/kg-
day)
-
450
..
-
500
inhal
NOAEL
(mg/m3)
-
200
..
-
oral LOAEL
(a) (mg/kg-
day)
0.77
0.77
-
1000
..
-
1
inhal
LOAEL
(a)
(mg/m3)
-
810
..
-
fish LC50(mg/L)
>1000
>100
900
930
<=0.5
>1000
5.4
14
5
fish NOEL
(mg/L)
0.006
>=20
>10
90
100
<=0.05
>100
0.87
0.8
0.5
E-22
-------
Table E-4. FINAL TOXICITY DATA SELECTIONS FOR USE IN THE LCIA
Cas#
7429-90-5
7440-41-7
7782-41-4
7782-49-2
7681-52-9
7440-24-6
7440-62-2
Material
Aluminum
Beryllium
Fluorine
Selenium
Sodium Hypochlorite
Strontium
Vanadium
Selection comments by UT
took average of LCSOs
took average of LCSOs
took average of LCSOs
took average of LCSOs
took average of LCSOs
used rainbow trout listed in fish LC50 column, as fathead
minnow data source had no date and did not provide time
period of the test
oral or inhal SF
WOE (EPA &
IARC)
oral
NOAEL
(mg/kg-
day)
inhal
NOAEL
(mg/m3)
oral LOAEL
(a) (mg/kg-
day)
inhal
LOAEL
(a)
(mg/m3)
fish LC50(mg/L)
11
2
>100
4.9
0.530 (measured)
210
4
fish NOEL
(mg/L)
3.3
0.2
>10
0.1
<=0.05
20
0.67
Notes:
Dark shading indicates data are not needed
(a) LOAEL only needed if no NOAEL found (LOAEL/10 will be used to represent NOAEL)
(b) Flux material names and CAS#s have been withheld to protect confidentiality
— = no data
** = low toxicity
E-23
-------
Table E-5. HUMAN HEALTH TOXICITY DATA COLLECTION
Cas#
Material
oral SF
(mg/kg-
day)-1
inhal SF
(mg/kg-
day)-1
Searched for human and ecological toxicity endpoints:
207-08-9
16872-11-0
16961-83-4
(b)
(b)
(b)
(b)
(b)
1309-36-0
7733-02-0
7440-69-9
Benzo(k)fluoranthene
Fluoroboric acid
Fluorosilicicacid
Flux A
FluxB
FluxD
FluxE
FluxF
Iron pyrite
Zinc sulfate
Bismuth
N/A
N/A
N/A
N/A
N/A
N/A
WOE (EPA
& IARC)
B2
N/A
N/A
Source*
U.S. EPA,
1997
oral NOAEL
(mg/kg-day)
N/A
N/A
N/A
Source*
inhal NOAEL
(mg/m3)
0.04 (Norway
DEL, human)
2.5 as F (human,
8-10 hr/day, 5
d/wk)
2.5 as F (human,
8-10 hr/day, 5
d/wk)
Source*
RTECS, 2003b
U.S. CFR,
1994, NIOSH,
1997
U.S. CFR,
1994, NIOSH,
1997
oral LOAEL (a)
(mg/kg-day)
N/A
0.77 (for fluorides;
human; 2 yr; bone,
joint and G.I.
effects)
0.77 (for fluorides;
human; 2 yr; bone,
joint and G.I.
effects)
Source*
U.S.
EPA,
1998a
U.S.
EPA,
1998a
inhal LOAEL
(a) (mg/m3)
N/A
N/A
N/A
Source*
data withheld for confidentiality
data withheld for confidentiality
data withheld for confidentiality
data withheld for confidentiality
data withheld for confidentiality
N/A
N/A
N/A
N/A
N/A
N/A
N/A
D
N/A
U.S. EPA
1998b
N/A
N/A
227g/d
(human, 3
wk)
HSDB, 2003
1.0 (for iron salts,
soluble as iron,
human, 8 hr/day,
5 d/wk
N/A
2.5 (PEL for 8 hr
day, 5 d/wk, for
bismuth fluoride)
ACGIH, 2002
U.S. CFR,
1994
N/A
1.0 (human; zinc
cmpds. as zinc)
221 mg/kg (LDLo,
human)
U.S.
EPA
1998b
Arena,
1970
N/A
N/A
N/A
Notes:
(a) LOAEL only needed if no NOAEL found (LOAEL/10 will be used to represent NOAEL)
Cancer WOE B2 = Probable human carcinogen
Cancer WOE D = Not classifiable as to human carcinogenicity
* Full citations of sources are provided in the References section of this Appendix (E.3)
(b) Flux material names and CAS#s have been withheld to protect confidentiality
BOLD indicates values used for LFSP
E-24
-------
Table E-6. AQUATIC TOXICITY DATA COLLECTION
Cas#
207-08-9
16872-11-0
16961-83-4
(a)
(a)
(a)
(a)
(a)
1309-36-0
7733-02-0
7440-69-9
7429-90-5
7440-41-7
7782-41-4
7782-49-2
7681-52-9
7440-24-6
7440-62-2
Material
Benzo(k)fluoranthene
Fluoroboric acid
Fluorosilicic acid
Flux A
FluxB
FluxD
FluxE
FluxF
Iron pyrite
Zinc sulfate
Bismuth
Aluminum
Beryllium
Fluorine
Selenium
Sodium Hypochlorite
Strontium
Vanadium
fish LC50(mg/L)
0.026 (96 hr predicted value for fish
exceeds water solubility)
NA
49 (as sodium fluorosilicate, bluegill, 96
hr)
Source*
U.S.EPA, 2003a
Dawson etal., 1977
fish NOEL (mg/L)
NA
N/A
N/A
Source*
ECOSAR
LC50 mg/L
(predicted
96-hr)
..
>1000
>100
ECOSAR
Chronic mg/L
(predicted)
0.006
>=20
>10
Source*
U.S.EPA, 2003b
U.S.EPA, 2003b
U.S.EPA, 2003b
data withheld for confidentiality
data withheld for confidentiality
data withheld for confidentiality
data withheld for confidentiality
data withheld for confidentiality
6746.128 (96 hr predicted LC50); report
as>1000
1.27 (fathead minnow, 96 hr LC50)
N/A
0.12,0.16,0.31 (rainbow trout; static, 96
hr)
37.9 (fathead minnow; time not given)
51, 128, 140, 193, 107.5, 200 (as sodium
fluoride, rainbow trout, 96 hr static)
11.5, 12.5, 45, 48 (rainbow trout, 96 hr)
0.08,5.9,1.56,0.44,1.37,0.39,0.58,0.18,0.
1 7, 0.79, 0. 1 4, 0. 72, 0. 35, 1 0(fathead
minnow, 96 hr)
>0.17-<15.61 (rainbow trout, 28 day)
0.16 (rainbow trout, 28 day)
U.S. EPA, 2002
Erten-Unal, etal., 1998
Holtze, 1983
Cardwelletal., 1976
Pimentel & Bulkley, 1983; Smith etal.,
1985, Camargo and Tarazona, 1991
Goettl et al., 1976; Spehar 1986
Ewell, etal., 1986; Wilde, etal.,
1983a, Wilde, etal., 1983b, Curtis et
al., 1979
Birgeetal., 1979
Birgeetal., 1979
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
..
14.0
5.0
11.0
2.0
>100
4.9
0.530
(measured)
210
4.0
..
0.800
0.500
3.3
0.200
>10
0.100
<=0.05
20.0
0.670
U.S.EPA, 2003b
U.S.EPA, 2003b
U.S.EPA, 2003b
U.S.EPA, 2003b
U.S.EPA, 2003b
U.S.EPA, 2003b
U.S.EPA, 2003b
U.S.EPA, 2003b
U.S.EPA, 2003b
U.S.EPA, 2003b
Notes:
(a) Flux material names and CAS#s have been withheld to protect confidentiality
* Full citations of sources are provided in the References section of this Appendix (E.3)
ECOSAR data in last columns were done by EPA after ORNL's search
where >/<, used absolute values
** = low toxicity
BOLD indicates values used for LFSP
E-25
-------
Table E-7. OTHER TOXICITY-RELATED DATA
Cas#
Material
For human and ecological endpoints:
207-08-9
16872-11-0
16961-83-4
(a)
(a)
(a)
(a)
(a)
1309-36-0
7733-02-0
7440-69-9
Benzo(k)fluoranthene
Fluoroboric acid
Fluorosilicic acid
Flux A
FluxB
FluxD
FluxE
FluxF
Iron pyrite
Zinc sulfate
Bismuth
For fish LC50 and fish NOEL endpoints only:
7429-90-5
7440-41-7
7782-41-4
7782-49-2
7681-52-9
7440-24-6
7440-62-2
Aluminum
Beryllium
Fluorine
Selenium
Sodium Hypochlorite
Strontium
Vanadium
Other Mammalian Toxicity value
0.0002 mg/L (MCL established
for PAH's)
430 mg/kg (oral LD50, rat)
Source*
U.S.CFR, 2002
RTECS 2003c
Other Aquatic Toxicity Value
(mg/L)
0.001 (13hrLT50, Daphnia
magna)
0.125 mg/L (aquatic
concentration of concern, CC)
N/A
Source*
U.S. EPA, 2002
U.S.EPA1998a
Other Cancer Data
0.01 (PEF; potency
equivalency factor)
Source*
U.S.EPA, 1993
data withheld for confidentiality
data withheld for confidentiality
data withheld for confidentiality
data withheld for confidentiality
data withheld for confidentiality
49.7 mg/m3 (rabbits exhib.
damaged tracheal epithelium
after 0.5-8 hours inhalation
exposure)
14.29 mg/kg (oral TDLo for
zinc & compounds, human)
0.05 mg/L (0.0014 mg/kg/day
in drinking water for 70 kg
human)
Konradova and
Bencko, 1975
RTECS 2003a
Ku & Schoenung,
2002
N/A
4.6 ppm (rainbow trout, 96 hr)
N/A
N/A
N/A
2.3 ppm (TLm for trout, time
not specified)
<1 .7 mg/L (fish acute toxicity
value; <0.02 mg/L CC
N/A
1 .8-1 .9 (LC50, fathead
minnow)
U.S. Coast Guard,
1984-85
Weiss 1 980
U.S. EPA, 1996
Kimball, n.d.
3.625 mg/kg (5 day, subcutan;
equivocal tumorigenic agent,
rabbit)
RTECS 2003a
Note:
Dark shading indicates data are not needed
* Full citations of sources are provided in the References section of this Appendix (E .3)
(a) Flux material names and CAS#s have been withheld to protect confidentiality
E-26
-------
E-8 SlopeFactors
Table E-8. Chemicals used to calculate geometric mean slope factor values for carcinogenic hazard
value
Chemical
Acephate
Acetaldehyde
Acrylamide
Acrylonitrile
Alachlor
Aldrin
Aniline
Aramite
Aroclor1016
Aroclor1016
Aroclor1221
Aroclor1221
Aroclor1232
Aroclor1232
Aroclor1242
Aroclor1242
Aroclor1248
Aroclor1248
Aroclor1254
Aroclor1254
Aroclor1260
Aroclor1260
Arsenic, Inorganic
Atrazine
Azobenzene
Benz[a]anthracene
Benzene
Benzidine
Benzo[a]pyrene
Benzo[b]fluoranthene
Benzo[k]fluoranthene
Benzotrichloride
Benzyl Chloride
Beryllium and compounds
Bis(2-chloro-1 -methylethyl)ether (Technical)
Bis(2-chloroethyl)ether
Bis(2-ethylhexyl)phthalate
Bis(chloromethyl)ether
Bromodichloro methane
Bromoform
Butadiene, 1,3-
Cadmium (Diet)
Cadmium (Water)
Captafol
Captan
Carbazole
Carbon Tetrachloride
Chloranil
Chlordane
Chloro-2-methylaniline HCI, 4-
Chloro-2-methylaniline, 4-
CAS # Oral Slope Factor Inhalation Slope Factor
(mg/kg-day)-1 (mg/kg-day)-1
30560-19-1
75-07-0
79-06-1
107-13-1
15972-60-8
309-00-2
62-53-3
140-57-8
12674-11-2
12674-11-2
11104-28-2
11104-28-2
11141-16-5
11141-16-5
53469-21-9
53469-21-9
12672-29-6
12672-29-6
11097-69-1
11097-69-1
11096-82-5
11096-82-5
7440-38-2
1912-24-9
103-33-3
56-55-3
71-43-2
92-87-5
50-32-8
205-99-2
207-08-9
98-07-7
100-44-7
7440-41-7
108-60-1
111-44-4
117-81-7
542-88-1
75-27-4
75-25-2
106-99-0
7440-43-9
7440-43-9
2425-06-1
133-06-2
86-74-8
56-23-5
118-75-2
057-74-9
3165-93-3
95-69-2
8.70E-03
4.50E+00
5.40E-01
8.00E-02
1.70E+01
5.70E-03
2.50E-02
4.00E-01
2.00E+00
4.00E-01
2.00E+00
4.00E-01
2.00E+00
4.00E-01
2.00E+00
4.00E-01
2.00E+00
4.00E-01
4.00E-01
2.00E+00
1.50E+00
2.22E-01
1.10E-01
7.30E-01
5.50E-02
2.30E+02
7.30E+00
7.30E-01
7.30E-02
1.30E+01
1.70E-01
4.30E+00
7.00E-02
1.10E+00
1 .40E-02
2.20E+02
6.20E-02
7.90E-03
8.60E-03
3.50E-03
2.00E-02
1.30E-01
4.03E-01
3.50E-01
4.60E-01
5.80E-01
7.70E-03
4.50E+00
2.40E-01
1.70E+01
2.50E-02
4.00E-01
2.00E+00
4.00E-01
2.00E+00
4.00E-01
2.00E+00
4.00E-01
2.00E+00
4.00E-01
2.00E+00
4.00E-01
2.00E+00
4.00E-01
2.00E+00
5.00E+01
1.10E-01
3.10E-01
2.90E-02
2.30E+02
3.10E+00
3.10E-01
3.10E-02
8.40E+00
3.50E-02
1.10E+00
2.20E+02
3.90E-03
1.80E+00
6.10E+00
6.10E+00
5.30E-02
1.30E+00
E-27
-------
E-8 SlopeFactors
Table E-8. Chemicals used to calculate geometric mean slope factor values for carcinogenic hazard
value
Chemical
Chlorobenzilate
Chlorodibromoethane
Chloroform
Chloromethane
Chloronitrobenzene, o-
Chloronitrobenzene, p-
Chlorothalonil
Chromium VI (chromic acid mists)
Chromium VI (particulates)
Chrysene
Coke Oven Emissions
Crotonaldehyde, trans-
Cyanazine
Cyclohexane, 1 , 2,3,4, 5-pentabromo-6-chloro-
DDD
DDE
DDT
Di(2-ethylhexyl)adipate
Diallate
Dibenz[a,h]anthracene
Dibromo-3-chloropropane, 1 ,2-
Dibromochloromethane
Dibromoethane, 1,2-
Dichloro-2-butene, 1 ,4-
Dichlorobenzene, 1 ,4-
Dichlorobenzidine, 3,3'-
Dichloroethane, 1,2-
Dichloroethylene, 1,1-
Dichloropropane, 1,2-
Dichloropropene, 1,3-
Dichlorvos
Dieldrin
Diethylstilbesterol
Dimethoxybenzidine, 3,3'-
Dimethylaniline HCI, 2,4-
Dimethylaniline, 2,4-
Dimethylbenzidine, 3,3'-
Dimethylhydrazine, 1,1-
Dinitrotoluene Mixture, 2,4/2,6-
Dinitrotoluene, 2,4-
Dinitrotoluene, 2,6-
Dioxane, 1 ,4-
Diphenylhydrazine, 1,2-
Direct Black 38
Direct Blue 6
Direct Brown 95
Epichlorohydrin
Ethyl Acrylate
Ethylbenzene
Ethylene Oxide
Ethylene Thiourea
CAS # Oral Slope Factor Inhalation Slope Factor
(mg/kg-day)-1 (mg/kg-day)-1
510-15-6
73506-94-2
67-66-3
74-87-3
88-73-3
121-73-3
1897-45-6
18540-29-9
18540-29-9
218-01-9
8007-45-2
123-73-9
21725-46-2
87-84-3
72-54-8
72-55-9
50-29-3
103-23-1
2303-16-4
53-70-3
96-12-8
124-48-1
106-93-4
764-41-0
106-46-7
91-94-1
107-06-2
75-35-4
78-87-5
542-75-6
62-73-7
60-57-1
56-53-1
119-90-4
21436-96-4
095-68-1
119-93-7
57-14-7
25321-14-6
121-14-2
606-20-2
123-91-1
122-66-7
1937-37-7
2602-46-2
16071-86-6
106-89-8
140-88-5
100-41-4
75-21-8
96-45-7
2.70E-01
8.40E-02
6.10E-03
1.30E-02
2.50E-02
1.80E-02
1.10E-02
7.30E-03
1.90E+00
8.40E-01
2.30E-02
2.40E-01
3.40E-01
3.40E-01
1.20E-03
6.10E-02
7.30E+00
1.40E+00
8.40E-02
8.50E+01
2.40E-02
4.50E-01
9.10E-02
6.00E-01
6.80E-02
1.00E-01
2.90E-01
1.60E+01
4.70E+03
1 .40E-02
5.80E-01
7.50E-01
9.20E+00
3.00E+00
6.80E-01
6.80E-01
6.80E-01
1.10E-02
8.00E-01
8.60E+00
8.10E+00
9.30E+00
9.90E-03
4.80E-02
1 .02E+00
1.10E-01
2.70E-01
8.10E-02
6.30E-03
4.10E+01
4.10E+01
3.10E-03
2.20E+00
3.40E-01
3.10E+00
2.40E-03
7.60E-01
9.30E+00
9.10E-02
1.20E+00
1 .40E-02
1.60E+01
4.90E+02
1.72E+01
8.00E-01
4.20E-03
3.85E-03
3.50E-01
E-28
-------
E-8 SlopeFactors
Table E-8. Chemicals used to calculate geometric mean slope factor values for carcinogenic hazard
value
Chemical
Folpet
Fomesafen
Formaldehyde
Furazolidone
Furium
Furmecyclox
Heptachlor
Heptachlor Epoxide
Hexachlorobenzene
Hexachlorobutadiene
Hexachlorocyclohexane, Alpha-
Hexachlorocyclohexane, Beta-
Hexachlorocyclohexane, Gamma-
Hexachlorocyclohexane, Technical
Hexachlorodibenzo-p-dioxin, Mixture
Hexachloroethane
Hexahydro-1 ,3,5-trinitro-1 ,3,5-triazine (RDX)
HpCDD, 2,3,7,8-
HpCDF, 2,3,7,8-
HxCDD, 2,3,7,8-
HxCDF, 2,3,7,8-
Hydrazine
Hydrazine Sulfate
lndeno[1 ,2,3-cd]pyrene
Isophorone
Methoxy-5-nitroaniline, 2-
Methyl Hydrazine
Methyl-5-Nitroaniline, 2-
Methylaniline Hydrochloride, 2-
Methylene Chloride
Methylene-bis(2-chloroaniline), 4,4'-
Methylene-bis(N,N-dimethyl) Aniline, 4,4'-
Methylenebisbenzenamine, 4,4'-
Mirex
Nickel Refinery Dust
Nickel Subsulfide
Nitrofurazone
Nitropropane, 2-
Nitrosodiethanolamine, N-
Nitrosodiethylamine, N-
Nitrosodimethylamine, N-
Nitroso-di-N-butylamine, N-
Nitroso-di-N-propylamine, N-
Nitrosodiphenylamine, N-
Nitrosomethylethylamine, N-
Nitroso-N-ethylurea, N-
Nitrosopyrrolidine, N-
OCDD
OCDF
PeCDD, 2,3,7,8-
PeCDF, 1,2,3,7,8-
CAS # Oral Slope Factor Inhalation Slope Factor
(mg/kg-day)-1 (mg/kg-day)-1
133-07-3
72178-02-0
50-00-0
67-45-8
531-82-8
60568-05-0
76-44-8
1024-57-3
118-74-1
87-68-3
319-84-6
319-85-7
58-89-9
608-73-1
19408-74-3
67-72-1
121-82-4
37871-00-4
38998-75-3
34465-46-8
55684-94-1
302-01-2
10034-93-2
193-39-5
78-59-1
99-59-2
60-34-4
99-55-8
636-21-5
75-09-2
101-14-4
101-61-1
101-77-9
2385-85-5
NA
12035-72-2
59-87-0
79-46-9
1116-54-7
55-18-5
62-75-9
924-16-3
621-64-7
86-30-6
10595-95-6
759-73-9
930-55-2
3268-87-9
39001-02-0
36088-22-9
57117-41-6
3.50E-03
1.90E-01
3.80E+00
5.00E+01
3.00E-02
4.50E+00
9.10E+00
1.60E+00
7.80E-02
6.30E+00
1.80E+00
1.30E+00
1.80E+00
6.20E+03
1 .40E-02
1.10E-01
1 .50E+03
1 .50E+03
1 .50E+04
1 .50E+04
3.00E+00
3.00E+00
7.30E-01
9.50E-04
4.60E-02
3.00E+00
3.30E-02
1.80E-01
7.50E-03
1.30E-01
4.60E-02
2.50E-01
1 .80E+00
1 .50E+00
9.50E+00
2.80E+00
1.50E+02
5.10E+01
5.40E+00
7.00E+00
4.90E-03
2.20E+01
1.40E+02
2.10E+00
1 .50E+02
1 .50E+02
7.50E+04
7.50E+04
4.50E-02
4.50E+00
9.10E+00
1.60E+00
7.80E-02
6.30E+00
1.80E+00
1.80E+00
4.55E+03
1 .40E-02
1.50E+03
1.50E+03
1.50E+04
1.50E+04
1.70E+01
1.70E+01
3.10E-01
1.72E+01
1 .65E-03
1.30E-01
8.40E-01
1.70E+00
9.40E+00
1.50E+02
5.10E+01
5.40E+00
2.10E+00
1.50E+02
1.50E+02
7.50E+04
7.50E+04
E-29
-------
E-8 SlopeFactors
Table E-8. Chemicals used to calculate geometric mean slope factor values for carcinogenic hazard
value
Chemical
CAS # Oral Slope Factor Inhalation Slope Factor
(mg/kg-day)-1 (mg/kg-day)-1
PeCDF, 2,3,4,7,8-
Pentachloronitrobenzene
Pentachlorophenol
Phenylenediamine, o-
Phenylphenol, 2-
Polybrominated Biphenyls
Polychlorinated Biphenyls (high risk)
Polychlorinated Biphenyls (low risk)
Polychlorinated Biphenyls (lowest risk)
Prochloraz
Propylene Oxide
Quinoline
Simazine
Sodium Diethyldithiocarbamate
Stirofos (Tetrachlorovinphos)
TCDD, 2,3,7,8-
TCDF, 2,3,7,8-
Tetrachloroethane, 1,1,1,2-
Tetrachloroethane, 1,1,2,2-
Tetrachloroethylene
Tetrachlorotoluene, p- alpha, alpha, alpha-
Toluene-2,4-diamine
Toluidine, o- (Methylaniline, 2-)
Toluidine, p-
Toxaphene
Trichloroaniline HCI, 2,4,6-
Trichloroaniline, 2,4,6-
Trichloroethane, 1,1,2-
Trichloroethylene
Trichlorophenol, 2,4,6-
Trichloropropane, 1,2,3-
Trifluralin
Trimethyl Phosphate
Trinitrotoluene, 2,4,6-
Vinyl Bromide
Vinyl Chloride
geometric mean
count (n)
min
max
57117-31-4
82-68-8
87-86-5
95-54-5
90-43-7
59536-65-1
1336-36-3
1336-36-3
1336-36-3
67747-09-5
75-56-9
91-22-5
122-34-9
148-18-5
961-11-5
1746-01-6
51207-31-9
630-20-6
79-34-5
127-18-4
5216-25-1
95-80-7
95-53-4
106-49-0
8001-35-2
33663-50-2
634-93-5
79-00-5
79-01-6
88-06-2
96-18-4
1582-09-8
512-56-1
118-96-7
593-60-2
75-01-4
7.50E+03
2.60E-01
1.20E-01
4.70E-02
1.94E-03
8.90E+00
2.00E+00
4.00E-01
7.00E-02
1.50E-01
2.40E-01
1.20E+01
1 .20E-01
2.70E-01
2.40E-02
1 .50E+05
1 .50E+04
2.60E-02
2.00E-01
5.20E-02
2.00E+01
3.20E+00
2.40E-01
1.90E-01
1.10E+00
2.90E-02
3.40E-02
5.70E-02
1.10E-02
1.10E-02
7.00E+00
7.70E-03
3.70E-02
3.00E-02
1.40E+00
0.71
175
0.00095
150000
7.50E+03
2.00E+00
4.00E-01
1 .30E-02
1.50E+05
1.50E+04
2.60E-02
2.00E-01
2.00E-03
1.10E+00
5.70E-02
6.00E-03
1 .OOE-02
1.10E-01
3.08E-02
1.70
105
0.00165
150000
blank=no data
Source: Risk Assessment Information System (RAIS), http://risk.lsd.ornl.gov/cgi-bin/tox/TOX_9801
(downloaded 11/00): IRIS/HEAST Slope Factors.
E-30
-------
Table E-9. Oral No Observable Adverse Effect Level (NOAEL) data
Chemical
2,3,7,8-TCDD
Arsenic
Terbufos
Vanadium
Cadmium cmpds
Manganese oxide
Polychlorinated biphenyls
Phosphorus (yellow or white)
Selenium
Phosphine
Chloropyrifos
Ammonium bifluoride
Fluorine
Acrylamide
1,3-Dichloropropene
Manganese
2,4-Dinitrotoluene
Hexachloro-1 ,3-butadiene
Uranium
Barium
Barium carbonate
Barium cmpds
Barium sulfate
Bromomethane
Nitrobenzene
Hexachlorobenzene
Copper
Lead
Cyanazine
Trifluralin
Zinc (elemental)
Acrylonitrile
Alachlor
Benzene
Carbon tetrachloride
Decabromodiphenyl oxide
Hexachloroethane
Nitrites
Pyridine
Chlorothalonil
Nitrate
Nitrates/nitrites
Sodium hypochlorite
Chromium (VI)
Ethyl dipropylthiocarbamate
Methyl parathion
Pentachlorophenol
Hydrogen sulfide
CAS#
1746-01-6
7440-38-2
13071-79-9
7440-62-2
20-04-2
1313-13-9
1336-36-3
7723-14-0
7782-49-2
7803-51-2
2921-88-2
1341-49-7
7782-41-4
79-06-1
542-75-6
7439-96-5
121-14-2
87-68-3
7440-61-6
7440-39-3
513-77-9
20-02-0
7727-43-7
74-83-9
98-95-3
118-74-1
7440-50-8
7439-92-1
21725-46-2
1582-09-8
7440-66-6
107-13-1
15972-60-8
71-43-2
56-23-5
1163-19-5
67-72-1
14797-65-0
110-86-1
1897-45-6
14797-55-8
7681-52-9
18540-29-9
759-94-4
298-00-0
87-86-5
7783-06-4
Value
9E-08
0.0008
0.0025
0.003
0.005
0.005
0.007
0.015
0.015
0.026
0.03
0.05
0.06
0.1
0.125
0.14
0.2
0.2
0.2
0.21
0.21
0.21
0.21
0.4
0.46
0.5
0.53
0.57
0.625
0.75
0.9
1
1
1
1
1
1
1
1
1.5
1.6
1.6
2.1
2.5
2.5
2.5
3
3.1
unit
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
E-31
-------
Table E-9. Oral No Observable Adverse Effect Level (NOAEL) data
Chemical
4,4'-Methylenedianiline
Atrazine
1,1,2-Trichloroethane
Butylate
Hydroquinone
Nickel
Nickel chloride
Methyl methacrylate
1 ,2,4-Trichloro benzene
Boron
Carbaryl
1.4-Dichlorobenzene
Maleic anhydride
Cyanide (-1)
Hydrogen cyanide
Captan
Chlorobenzene
Chlorine
Tetrachloroethylene
2,4-D
Dichlorodifluoromethane
Formaldehyde
Bromoform
1,2-Dichloroethane
1,2-Dichlorobenzene
Aluminum hydroxide
Trichloroethylene
Maneb
Ethylene oxide
N,N-dimethylaniline
Ammonia
2-methoxyethanol
Acetonitrile
Biphenyl
Chlorophenols [o]
Di (2-ethylhexyl) phthalate
Methyl isobutyl ketone
P-cresol
Aluminum (elemental)
Phenol
Boric acid
Orthoboric acid
4-Nitrophenol
Coolant
Ethylene glycol
Naphthalene
Butyraldehyde
Diethanolamine
CAS#
101-77-9
1912-24-9
79-00-5
2008-41-5
123-31-9
7440-02-0
7718-54-9
80-62-6
120-82-1
7440-42-8
63-25-2
106-46-7
108-31-6
57-12-5
74-90-8
133-06-2
108-90-7
7782-50-5
127-18-4
94-75-7
75-71-8
50-00-0
75-25-2
107-06-2
95-50-1
21645-51-2
79-01-6
12427-38-2
75-21-8
121-69-7
7664-41-7
109-86-4
75-05-8
92-52-4
20-05-3
117-81-7
108-10-1
106-44-5
7429-90-5
108-95-2
11113-50-1
10043-35-3
100-02-7
not available
107-21-1
91-20-3
123-72-8
111-42-2
Value
3.2
3.5
3.9
5
5
5
5
7.5
7.8
8.8
9.6
10
10
10.8
10.8
12.5
12.5
14
14
15
15
15
17.9
18
18.8
23
24
25
30
32
34
50
50
50
50
50
50
50
60
60
67
67
70
71
71
71
75
75
unit
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
E-32
-------
Table E-9. Oral No Observable Adverse Effect Level (NOAEL) data
Chemical
Pyrene
Acrylic acid
Butyl acrylate
Acetone
Methyl tert-butyl ether
Nickel cmpds
Styrene
Toluene
Vinyl acetate
Acetaldehyde
Dibutyl phthalate
Fluoranthene
Fluorene
Methyl ethyl ketone
N-butyl alcohol
Ethylbenzene
Benzaldehyde
Diethyl phthalate
Isophorone
Butyl benzyl phthalate
Cumene
Dichloromethane
Acenaphthene
m, p-xylene
o-xylene
Xylene (mixed isomers)
Strontium
Strontium carbonate
Acetic acid
Dioctyl sebacate
Propylene oxide
Glycol ethers
Isopropyl alcohol
1,1,1-Trichloroethane
1,2-Dichloropropane
2-ethoxyethanol
m-xylene
1 ,2-Dichlorotetrafluoroethane
Freon 113
Metolachlor
Ethanol amine
Acetophenone
Di propylene glycol butyl ether
4,4'-lsopropylidenediphenol
Diethyl ether
Methanol
Terephthalic acid
Polyvinyl pyrrolidone (PVP)
CAS#
129-00-0
79-10-7
141-32-2
67-64-1
1634-04-4
20-14-4
100-42-5
108-88-3
108-05-4
75-07-0
84-74-2
206-44-0
86-73-7
78-93-3
71-36-3
100-41-4
100-52-7
84-66-2
78-59-1
85-68-7
98-82-8
75-09-2
83-32-9
1330-20-7
95-47-6
1330-20-7
7440-24-6
1633-05-2
64-19-7
122-62-3
75-56-9
111-76-2
67-63-0
71-55-6
78-87-5
110-80-5
108-38-3
76-14-2
76-13-1
51218-45-2
141-43-5
98-86-2
29911-28-2
80-05-7
60-29-7
67-56-1
100-21-0
9003-39-8
Value
75
83
84
100
100
100
100
100
100
125
125
125
125
125
125
136
143
150
150
151
154
155
175
179
179
179
190
190
195
200
200
203
230
250
250
250
250
273
273
300
320
423
450
500
500
500
500
550
unit
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
E-33
-------
Table E-9. Oral No Observable Adverse Effect Level (NOAEL) data
Chemical
Bis (2-ethylhexyl) adipate
Tetrahydrofuran
Glyphosate
2-(2-butoxyethoxy)-ethanol acetate
Anthracene
Dimethyl phthalate
Heptane
p-xylene
Phosphate ester
Polyethylene mono (nonylphenyl) ether glycol
Diethylene glycol
Chromium (III)
Chromium trioxide
Tert-butyl alcohol
Bismuth
Zirconium
CAS#
103-23-1
109-99-9
1071-83-6
124-17-4
120-12-7
131-11-3
142-82-5
106-42-3
57583-54-7
9016-45-9
111-46-6
16065-83-1
1333-82-0
75-65-0
7440-69-9
7440-67-7
Count n=
geometric mean=
minumum=
maximum=
Value
610
782
800
1000
1000
1000
1000
1000
1000
1000
1250
1468
1468
1599
3243
3494
160
13.987
9E-08
3494
unit
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
Mg/KgDay
E-34
-------
Table E-10. Inhalation No Observable Adverse Effect Level (NOAEL)
Chemical
1,1,1-Trichloroethane
1 ,2,4-Trichloro benzene
1,2-Dichloroethane
1,2-Dichloropropane
1,3-Butadiene
1,3-Dichloropropene
1,4-Dichlorobenzene
1 ,4-Dioxane
1 -Methoxy-2-propanol
2-Ethoxyethanol
2-Methoxyethanol
4,4'-lsopropylidenediphenol
4-Nitrophenol
Acetaldehyde
Acetonitrile
Acrylic acid
Allyl chloride
Ammonia
Ammonium nitrate (solution)
Aniline
Benzene
Bromomethane
Butyl acrylate
Butyl benzyl phthalate
Butyraldehyde
Carbon disulfide
Carbon monoxide
Carbon tetrachloride
Chlorobenzene
Coolant
Cumene
Cumene hydroperoxide
Cyclohexane
Di (2-ethylhexyl) phthalate
Dichlorobenzene (mixed isomers)
Dichloromethane
Diethanolamine
Epichlorohydrin
Ethyl chloride
Ethylbenzene
Ethylene
Ethylene glycol
Ethylene oxide
Formaldehyde
Glycol ethers
HCFC-22
Hexachloro-1 ,3-butadiene
HFC-125
Hydrochloric acid
Isopropyl alcohol
Maneb
CAS#
71-55-6
120-82-1
1 07-06-2
78-87-5
1 06-99-0
542-75-6
106-46-7
123-91-1
1 07-98-2
110-80-5
1 09-86-4
80-05-7
1 00-02-7
75-07-0
75-05-8
79-10-7
107-05-1
7664-41-7
6484-52-2
62-53-3
71-43-2
74-83-9
141-32-2
85-68-7
123-72-8
75-15-0
630-08-0
56-23-5
1 08-90-7
not available
98-82-8
80-15-9
110-82-7
117-81-7
25321-22-6
75-09-2
111-42-2
1 06-89-8
75-00-3
100-41-4
74-85-1
107-21-1
75-21-8
50-00-0
111-76-2
75-45-6
87-68-3
354-33-6
7647-01-0
67-63-0
12427-38-2
Value
1214.9
24.3
221
710
2800
49.6
75
360
658
7480
93.3
10
30
300
91.5
74
68.3
40
185
19
1.15
4.3
120
144
3200
10
114.5
34.3
377
10
537
31
1500
50
610.4
796
0.27
20.7
3600
2370
11600
10
18
0.6
121
5260
58.2
245000
15
268.3
10
unit
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
E-35
-------
Table E-10. Inhalation No Observable Adverse Effect Level (NOAEL)
Chemical
Mercury
Methanol
Methyl chloride
Methyl ethyl ketone
Methyl isobutyl ketone
Methyl methacrylate
Metyl tert-butyl ether
N,N-Dimethylaniline
N-butyl alcohol
Nitrobenzene
p-cresol
p-xylene
Phosphine
Phosphoric acid
Propionaldehyde
Propylene
Propylene glycol
Propylene oxide
Sec-butyl alcohol
Styrene
Sulfur dioxide
Sulfuric acid
Terephthalic acid
Tetrachloroethylene
Tetrahydrofuran
Titanium
Titanium tetrachloride
Toluene
Toluene-2,4-diisocyanate
Trichloroethylene
Vinyl acetate
Vinyl chloride
Vinylidene chloride
CAS#
7439-97-6
67-56-1
74-87-3
78-93-3
108-10-1
80-62-6
1634-04-4
121-69-7
71-36-3
98-95-3
1 06-44-5
1 06-42-3
7803-51-2
7664-38-2
123-38-6
115-07-1
57-55-6
75-56-9
78-92-2
1 00-42-5
7446-09-5
7664-93-9
100-21-0
127-18-4
1 09-99-9
7440-32-6
7550-45-0
1 08-88-3
584-84-9
79-01-6
1 08-05-4
75-01-4
75-35-4
Count n=
Geometric mean=
minimum=
maximum=
Value
0.006
130
1138.4
8047
224
111.7
2880
0.006
0.1
27.5
10
5812.6
0.25
50
200
9375
170
237
8270
565
0.104
0.1
3
740.2
0.2
0.8
0.009
411.1
0.03
586.6
176
69754.5
120
84
68.6653
0.006
245000
unit
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
mg/m3
E-36
-------
Table E-11. Fish Lethal Concentration to 50 percent of exposed population (LC50)
Chemical
1,1,1-Trichloroethane
1,1,2-Trichloroethane
1 ,2,3,5-Tetrachlorobenzene
1 ,2,4-Trichloro benzene
1 ,2,4-Trimethylbenzene
1,2-Dichlorobenzene
1,2-Dichloroethane
1,2-Dichloropropane
1,3-Butadiene
1,3-Dichloropropene
1,4-Dichlorobenzene
1 ,4-Dioxane
1 -Methylphenanthrene
2,4,5-Trichlorotoluene
2,4,6-Trichlorophenol
2,4-D
2,4-Dinitrophenol
2,4-Dinitrotoluene
2-Ethoxyethanol
2-Methoxyethanol
2-Nitropropane
3,4-Dinitrotoluene
4,4'-lsopropylidenediphenol
4,4'-Methylenedianiline
4-Nitrophenol
Acetaldehyde
Acetone
Acetonitrile
Acrylamide
Acrylic acid
Acrylonitrile
Alachlor
Allyl chloride
Aluminum
Aluminum (+3)
Ammonia
Ammonium nitrate (solution)
Ammonium sulfate (solution)
Aniline
Anthracene
Antimony
Antimony cmpds
Arsenic
Arsenic cmpds
Atrazine
Barium
Barium cmpds
Benzaldehyde
Benzene
Benzoyl chloride
CAS#
71-55-6
79-00-5
634-90-2
120-82-1
95-63-6
95-50-1
1 07-06-2
78-87-5
1 06-99-0
542-75-6
106-46-7
123-91-1
832-69-9
6639-30-1
88-06-2
94-75-7
51-28-5
121-14-2
110-80-5
1 09-86-4
79-46-9
610-39-9
80-05-7
101-77-9
1 00-02-7
75-07-0
67-64-1
75-05-8
79-06-1
79-10-7
107-13-1
1 5972-60-8
107-05-1
7429-90-5
7664-41-7
6484-52-2
7783-20-2
62-53-3
120-12-7
7440-36-0
20-00-8
7440-38-2
20-01-9
1912-24-9
7440-39-3
20-02-0
1 00-52-7
71-43-2
98-88-4
Value
48
82
4
3
8
1
136
127
4
0.24
34
9850
1
1
3
71
11
24
16305
22655
5
2
5
45
41
34
7200
1640
109
186
10
5
72
11
3.6
2
800
4000
108
0.01
14.4
833
14.4
32
16
580
200
27
19
35
unit
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
E-37
-------
Table E-11. Fish Lethal Concentration to 50 percent of exposed population (LC50)
Chemical
Beryllium
Biphenyl
Bis(2-ethylhexyl) adipate
Boron
Boron (B III)
Bromomethane
Butyl benzyl phthalate
Butylate
Butyraldehyde
Cadmium
Cadmium cmpds
Caffeine
Captan
Carbaryl
Carbon disulfide
Carbon tetrachloride
Carbonyl sulfide
Catechol
Chlorine
Chlorine dioxide
Chlorobenzene
Chloroform
Chlorophenols [o]
Chloroprene
Chlorothalonil
Chlorpyrifos
Chromium
Chromium (VI)
Chromium cmpds
Chromium III
Cobalt cmpds
Coolant
Copper
Copper (+1 & +2)
Copper cmpds
Cresol (mixed isomers)
Cumene
Cumene hydroperoxide
Cyanazine
Cyanide (-1)
Cyclohexane
Cyclohexanone
Cyclohexylamine
Decabromodiphenyl oxide
Di (2-ethylhexyl)phthalate
Diaminotoluene (mixed isomers)
Dibutyl phthalate
Dichlorobenzene (mixed isomers)
Dichloromethane
Diethanolamine
CAS#
7440-90-5
92-52-4
1 03-23-1
7440-42-8
74-83-9
85-68-7
2008-41-5
123-72-8
7440-43-9
20-04-2
58-08-2
1 33-06-2
63-25-2
79-15-0
56-23-5
463-58-1
120-80-9
7782-50-5
10049-04-4
1 08-90-7
67-66-3
20-05-3
126-99-8
1 897-45-6
2921-88-2
7440-47-3
1 8540-29-9
20-06-4
16065-83-1
20-07-5
7440-50-8
20-08-6
1319-77-3
98-82-8
80-15-9
21725-46-2
57-12-5
110-82-7
1 08-94-1
108-91-8
1163-19-5
117-81-7
25376-45-8
84-74-2
25321-22-6
75-09-2
111-42-2
Value
2
2
0.35
113
113
11
43
7
32
0.001
0.1
151
0.2
8
694
41
2685
9
0.34
0.17
17
71
19
2
0.05
2.4
52
22.6
33
3.3
0.38
227634
0.014
0.014
0.33
13
6
62
18
56
5
630
222
0.06
1
37
1
1
330
4710
unit
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
E-38
-------
Table E-11. Fish Lethal Concentration to 50 percent of exposed population (LC50)
Chemical
Diethyl phthalate
Dimethyl phthalate
Di-n-octyl phthalate
Edetic acid (EDTA)
Epichlorohydrin
Ethyl chloride
Ethyl dipropylthiocarbamate
Ethylbenzene
Ethylene
Ethylene glycol
Ethylene oxide
Fluorine
Formaldeyde
Freon 113
Glycol ethers
Glyphosate
Hexachloro-1 ,3-butadiene
Hexachlorobenzene
Hexachlorocyclopentadiene
Hexachloroethane
Hexane
Hydrazine
Hydrochloric acid
Hydrofluoric acid
Hydrogen cyanide
Hydroquinone
Isobutyraldehyde
Isopropyl alcohol
Lead
Lead cmpds
Lead sulfate cake
Lithium salts
M,p-xylene
Malathion
Maleic anhydride
Maneb
Manganese cmpds
Mercury
Mercury cmpds
Metam sodium
Methanol
Methl mercury
Methyl chloride
Methyl ethyl ketone
Methyl isobutyl ketone
Methyl methacrylate
Methyl parathion
Methyl tert-butyl ether
Methylenebis (phenylisocyanate)
Metolachlor
CAS#
84-66-2
131-11-3
117-84-0
60-00-4
1 06-89-8
75-00-3
759-94-4
100-41-4
74-85-1
107-21-1
75-21-8
7782-49-2
50-00-0
76-13-1
111-76-2
1071-83-6
87-68-3
1 1 8-74-1
77-47-4
67-72-1
110-54-3
302-01-2
7647-01-0
7664-39-3
74-90-8
123-31-9
78-84-2
67-63-0
7439-92-1
20-11-1
7446-14-2
121-75-5
108-31-6
12427-38-2
20-12-2
7439-97-6
1 37-42-8
67-56-1
115-09-3
74-87-3
78-93-3
108-10-1
80-62-6
298-00-0
1634-04-4
101-68-8
51218-45-2
Value
32
121
1
473
35
16
27
11
14
227634
84
100
24
290
1490
600
0.09
22
0.007
1
2.5
4.83
19
265
1385
141
41
8623
31.5
5
60.8
2600
13
0.1
2963
2
150
0.155
0.155
0.39
29400
0.09
550
3220
505
259
9
786
15
unit
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
E-39
-------
Table E-11. Fish Lethal Concentration to 50 percent of exposed population (LC50)
Chemical
Metribuzin
Molybdenum
Molybdenum (Mo II, Mo III, Mo IV, Mo V, Mo VI)
Molybdenum trioxide
m-xylene
N, N-Demethylaniline
Naphthalene
N-butyl alcohol
Nickel
Nickel cmpds
Nitrate
Nitrates/nitrites
Nitric acid
Nitrites
Nitrobenzene
Nitrogen dioxide
N-nitrosodiphenylamine
o-xylene
p-cresol
Phenol
Phosphoric acid
Phosphorus (yellow or white)
Phthalic anhydride
Picric acid
Polychlorinated biphenyls
Propionaldehyde
Propylene
Propylene oxide
p-xylene
Pyridine
Sec-butyl alcohol
Selenium
Silver
Silver cmpds
Silvex
Sodium Hypochlorite
Strontium
Styrene
Sulfuric acid
Terbufos
Terephthalic acid
Tert-butyl alcohol
Tetrachloroethylene
Tin
Tin (Sn++, Sn4+)
Titanium tetrachloride
Toluene
Toluene-2,4-diisocyanate
Trans-1 ,2-dichloroethylene
Trichloroethylene
CAS#
21087-64-9
7439-98-7
1313-27-5
1 08-38-3
121-69-7
91-20-3
71-36-3
7440-02-0
20-14-4
14797-55-8
7697-37-2
14797-65-0
98-95-3
10102-44-0
86-30-6
95-47-6
1 06-44-5
1 08-95-2
7664-38-2
7723-14-0
85-44-9
88-89-1
1336-36-3
123-38-6
115-07-1
75-56-9
1 06-42-3
110-86-1
78-92-2
7782-49-2
7440-22-4
93-72-1
7681-52-9
7440-24-6
1 00-42-5
7664-93-9
13071-79-9
100-21-0
75-65-0
127-18-4
7440-31-5
7550-45-0
1 08-88-3
584-84-9
156-60-5
79-01-6
Value
80
157
157
370
16
65
6
1860
2.48
27
2213
2213
26
225
119
196
1
16
25
34
70
0.02
364
170
3
44
5
306
2
100
3670
4.9
0.004
12
13
0.53
210
4
31
0.01
29
1954
17
626
626
25
34
53
45
44
unit
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
E-40
-------
Table E-11. Fish Lethal Concentration to 50 percent of exposed population (LC50)
Chemical
Trichlorofluoromethane
Triethylene glycol
Trifluralin
Vanadium
Vinyl acetate
Vinyl chloride
Vinylidene chloride
Xylene (mixed isomers)
Zinc (+2)
Zinc (elemental)
Zinc cmpds
Benzo(k)fluoranthene
Beta terpineol
Di propylene glycol butyl ether
2,2-Dimethylolpropionic acid
Ethoduomeen
Fluoroboric acid
Fluorosilicic acid
Iron pyrite
Tri propylene glycol butyl ether
Zinc sulfate
Bismuth
CAS#
75-69-4
112-27-6
1 582-09-8
7440-62-2
1 08-05-4
75-01-4
75-35-4
1 330-20-7
7440-66-6
20-19-9
207-08-9
138-87-4
29911-28-2
4767-03-7
53127-17-6
16872-11-0
16961-83-4
1309-36-0
55934-93-5
7733-02-0
7440-69-9
Count n=
Geometric mean=
minimum=
maximum=
Value
114
88100
0.11
4
100
143
108
13
0.09
0.09
17
1000
5.4
930
1000
0.5
1000
100
1000
900
14
5
221
24.592
0.001
227634
unit
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
E-41
-------
Table E-12. Fish No Observed Effect Level (NOEL)
Chemical
1,1,1-Trichloroethane
1,1,2-Trichloroethane
1 ,2,4-Trichloro benzene
1 ,2,4-Trimethylbenzene
1,2-Dichlorobenzene
1,2-Dichloroethane
1,2-Dichloropropane
1,3-Butadiene
1,3-Dichloropropene
1,4-Dichlorobenzene
1 ,4-Dioxane
2,4-D
2,4-Dinitrophenol
2,4-Dinitrotoluene
2-Ethoxyethanol
2-Methoxyethanol
2-Nitropropane
4,4'-lsopropylidenediphenol
4,4'-Methylenedianiline
4-Nitrophenol
Acetaldehyde
Acetone
Acetonitrile
Acrylamide
Acrylic acid
Acrylonitrile
Alachlor
Allyl chloride
Aluminum (+3)
Ammonia
Ammonium nitrate (solution)
Ammonium sulfate (solution)
Aniline
Antimony
Antimony cmpds
Arsenic
Arsenic cmpds
Atrazine
Barium
Barium cmpds
Benzene
Benzoyl chloride
Biphenyl
Bis (2-ethylhexyl)adipate
Boron
Boron (B III)
Bromomethane
Butyl acrylate
Butyl benzyl phthalate
CAS#
71-55-6
79-00-5
120-82-1
95-63-6
95-50-1
1 07-06-2
78-87-5
1 06-99-0
542-75-6
106-46-7
123-91-1
94-75-7
51-28-5
121-14-2
110-80-5
1 09-86-4
79-46-9
80-05-7
101-77-9
1 00-02-7
75-07-0
67-64-1
75-05-8
79-06-1
79-10-7
107-13-1
1 5972-60-8
107-05-1
7664-41-7
6484-52-2
7783-20-2
62-53-3
7440-36-0
20-00-8
7440-38-2
20-01-9
1912-24-9
7440-39-3
20-02-0
71-43-2
98-88-4
92-52-4
1 03-23-1
7440-42-8
74-83-9
141-32-2
85-68-7
Value
7
1
0.2
0.68
0.05
34
23
1
0.06
3
2588
6
3
6
4076
5664
1
0.42
11
10
9
1800
410
27
47
3
0.51
18
0.36
0.09
40
200
27
1.6
42
2.1
2
3
50
10
4
9
0.12
0.09
27
27
3
0.31
2
unit
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
E-42
-------
Table E-12. Fish No Observed Effect Level (NOEL)
Chemical
Butylate
Butyraldehyde
Cadmium
Captan
Carbaryl
Carbon disulfide
Carbon tetrachloride
Carbonyl sulfide
Catechol
Chlorine
Chlorine dioxide
Chlorobenzene
Chloroform
Chlorophenols [o]
Chloroprene
Chlorothalonil
Chlorpyrifos
Chromium
Chromium III
Chromium VI
Chromium cmpds
Cobalt cmpds
Coolant
Copper
Copper (+1 & +2)
Copper cmpds
Cresol (mixed isomers)
Cumene
Cumene hydroperoxide
Cyanazine
Cyanide (-1)
Cyclohexane
Di (2-ethylhexyl) phthalate
Di-n-octyl phthalate
Diaminotoluene (mixed isomers)
Dibutyl phthalate
Dichlorobenzene (mixed isomers)
Dichloromethane
Diethanolamine
Diethyl phthalate
Dimethyl phthalate
Edetic acid (EDTA)
Epichlorohydrin
Ethoduomeen
Ethyl chloride
Ethyl dipropylthiocarbamate
Ethylbenzene
Ethylene
Ethylene glycol
CAS#
2008-41-5
123-72-8
7440-43-9
1 33-06-2
63-25-2
75-15-0
56-23-5
463-58-1
120-80-9
7782-50-5
10049-04-4
1 08-90-7
67-66-3
20-05-3
126-99-8
1 897-45-6
2921-88-2
7440-47-3
16065-83-1
1 8540-29-9
20-06-4
20-07-5
not available
7440-50-8
20-08-6
1319-77-3
92-82-8
80-15-9
21725-46-2
57-12-5
110-82-7
117-81-7
117-84-0
25376-45-8
84-74-2
25321-22-6
75-09-2
111-42-2
84-66-2
131-11-3
60-00-4
1 06-89-8
53127-17-6
75-00-3
759-94-4
100-41-4
74-85-1
107-21-1
Value
2
8
0.001
0.05
1
174
5
671
2
0.02
0.01
2
18
3
0.56
0.01
0.12
5.2
0.33
2.23
2
0.02
56909
0.004
0.004
0.02
3
0.49
16
5
5.7
0.39
0.08
0.05
9
0.05
0.05
83
1178
5
30
240
9
0.05
4
3
1
3
56909
unit
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
E-43
-------
Table E-12. Fish No Observed Effect Level (NOEL)
Chemical
Ethylene oxide
Formaldehyde
Freon 113
Glycol ethers
Glyphosate
Hexachlorobenzene
Hexachloroethane
Hexane
Hydrazine
Hydrochloric acid
Hydrofluoric acid
Hydrogen cyanide
Hydroquinone
Isobutyraldehyde
Isopropyl alcohol
Lead
Lead cmpds
Lead sulfate cake
Lithium salts
M,p-xylene
m-xylene
Malathion
Maleic anhydride
Maneb
Manganese cmpds
Mercury
Mercury cmpds
Metam sodium
Methanol
Methyl chloride
Methyl ethyl ketone
Methyl isobutyl ketone
Methyl methacrylate
Methyl parathion
Methyl tert-butyl ether
Methylenebis (phenylisocyanate)
Metolachlor
Metribuzin
Molybdenum
CAS#
75-21-8
50-00-0
76-13-1
111-76-2
1071-83-6
1 1 8-74-1
67-72-1
110-54-3
302-01-2
7647-01-0
7664-39-3
74-90-8
123-31-9
78-84-2
67-63-0
7439-92-1
20-11-1
7446-14-2
1 330-20-7
1 08-38-3
121-75-5
108-31-6
12427-38-2
20-12-2
7439-97-6
not applicable
1 37-42-8
67-56-1
74-87-3
78-93-3
108-10-1
80-62-6
298-00-0
1634-04-4
101-68-8
51218-45-2
21087-64-9
7439-98-7
Molybdenum (Mo II, Mo III, Mo IV, Mo V, Mo VI)
Molybdenum trioxide
N,N-Dimethylaniline
N-butyl alcohol
N-nitrosodiphenylamine
Naphthalene
Nickel
Nickel cmpds
Nitrate
Nitrates/nitrites
1313-27-5
121-69-7
71-36-3
86-30-6
91-20-3
7440-02-0
20-14-4
14797-55-8
Value
118
6
73
373
150
1
0.35
0.25
0.48
0.95
13
346
35
10
2156
0.004
0.26
6.08
260
1
2
0.01
741
0.09
8
0.005
0.005
0.1
7350
138
805
126
65
0.88
197
16
1
20
0.125
0.125
19
12
465
0.13
0.59
0.09
1
213
213
unit
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
E-44
-------
Table E-12. Fish No Observed Effect Level (NOEL)
Chemical
Nitric acid
Nitrobenzene
Nitrogen dioxide
o-xylene
p-cresol
p-xylene
Phenol
Phosphoric acid
Phthalic anhydride
Picric acid
Polychlorinated biphenyls
Propionaldehyde
Propylene
Propylene oxide
Pyridine
Sec-butyl alcohol
Silver
Silver cmpds
Styrene
Sulfuric acid
Terephthalic acid
Tert-butyl alcohol
Tetrachloroethylene
Tin
Tin (Sn++, Sn4+)
Titanium tetrachloride
Toluene
Toluene-2,4-diisocyanate
Trichloroethylene
Triethylene glycol
Trifluralin
Vinyl acetate
Vinyl chloride
Vinylidene chloride
Xylene (mixed isomers)
Zinc (+2)
Zinc (elemental)
Benzo(k)fluoranthene
Beta terpineol
Di propylene glycol butyl ether
2,2-Dimethylolpropionic acid (DMPA)
Fluorosilicic acid
Iron pyrite
Tri propylene glycol butyl ether
Zinc sulfate
Fluoroboric acid
Bismuth
Aluminum
Beryllium
Fluorine
CAS#
7697-37-2
98-95-3
10102-44-0
95-47-6
1 06-44-5
1 06-42-3
1 08-95-2
7664-38-2
85-44-9
88-89-1
1336-36-3
123-38-6
115-07-1
75-56-9
110-86-1
78-92-2
7440-22-4
1 00-42-5
7664-93-9
100-21-0
75-65-0
127-18-4
7440-31-5
7550-45-0
1 08-88-3
584-84-9
79-01-6
112-27-6
1 582-09-8
1 08-05-4
75-01-4
75-35-4
1 330-20-7
7440-66-6
207-08-9
138-87-4
29911-28-2
4767-03-7
16961-83-4
1309-36-0
55934-93-5
7733-02-0
16872-11-0
7440-69-9
7429-90-5
7440-41-7
7782-41-4
Value
1
30
19.6
2
6
0.2
8
4
91
41
0.14
11
1
77
25
918
0.001
0.001
0.44
2
7
488
2
62.6
62.6
1
4
13
8
8810
0.01
25
36
27
1
0.036
0.036
0.006
0.87
100
100
10
100
90
0.8
20
0.5
3.3
0.2
10
unit
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
E-45
-------
Table E-12. Fish No Observed Effect Level (NOEL)
Chemical
Selenium
Sodium hypochlorite
Strontium
Vanadium
CAS#
7782-49-2
7681-52-9
7440-24-6
7440-62-2
Count n=
Geometric mean=
minimum=
maximum=
Value
0.1
0.05
20
0.67
199
3.9012
0.001
56909
unit
mg/L
mg/L
mg/L
mg/L
E-46
-------
Table E-1 3. Geometric means used to calculate toxicity hazard values a
Parameter
Oral SF
Inhalation SF
Oral NOAEL
Inhalation NOAEL
Fish LC50
Fish NOEL
n
175
105
160
84
221
199
min
0.00095
0.00165
9E-08
0.006
0.001
0.001
max
150000
150000
3494
245000
227634
56909
Geometric mean
0.707
1.70
14.0
68.7
24.6
3.90
3 The chemical data used to generate the geometric means are listed in Tables E-X
through E-X.
E-47
-------
APPENDIX F:
SUMMARY OF INDUSTRY PERFORMANCE
TESTING OF SOLDER
Bhatia, G, and J. Siegel. "Summary of Lead-Free Solder Performance Based
on Existing Data Provided by the Electronics Industry." Report prepared for
EPA Design for the Environment Program by Abt Associates, December,
2002.
-------
Appendix F
F.I INTRODUCTION
F.I.I SCOPE
This appendix summarizes existing data on the performance of lead-free solders available in the
electronics industry. In particular, it considers literature that referenced three specific alternative
solder types: tin-copper (Sn-Cu), tin-silver-copper (Sn-Ag-Cu), and tin-silver-copper-bismuth
(Sn-Ag-Cu-Bi). Additionally, it includes performance data for the tin-lead (Sn-Pb) alloy, as
several literature sources compare alternative alloy data with existing tin-lead standards. This
document is intended to provide EPA's Design for the Environment (DfE) Lead-free Solder
Partnership and other interested parties with a consolidated source of key lead-free solder
performance data. It identifies and summarizes existing data as well as documents these sources
for further research.
During a preliminary literature search, lead-free solder performance data available in the
electronics industry were found to be varied; alloy compositions as well as performance tests
carried out on the alternative solders differed. As this appendix intends to be inclusive rather
than overlook key applicable results, it includes summaries of documents that reference alloy
compositions falling within the alloy families considered (for example, Sn-3 Ag-4Cu and Sn-
0.5Ag-4Cu fall under the ternary Sn-Ag-Cu alloy family). However, it should be noted that
multiple sources have illustrated that performance results vary when an alloy's composition was
altered. For example, Lau et al. cite that the elongation of the tin-silver-copper system drops
rapidly with increasing bismuth content until it reaches the 3% level, where the elongation
decreases slowly and later levels off with a further increase in Bi content. As a result,
performance data for alloys were not limited to the compositions as defined by the EPA's DfE
Lead-free Solder Partnership (see Table F. 1.1.1), but included relevant data for alloy
compositions close to the Partnership's selection.
Table F.I.1.1: EPA's DfE Lead-free Solder Partnership's Selection: Alloy Compositions
and Family
DfE Lead-free Solder Partnership Selection, Alloy
Composition
99.2% Tin and 0.8% Copper
95.5% Tin, 3.9% Silver, and 0.6% Copper
96.0% Tin, 2.5% Silver, 0.5% Copper, and 1
0% Bismuth
Alloy Family
Considered
Sn-Cu
Sn-Ag-Cu
Sn-Ag-Cu-Bi
F-l
-------
F.1.2 BACKGROUND
The Japanese Ministry of International Trade and Industry (MITI) proposed take-back legislation
in Japan, requiring consumer and business users to return end-of-life (EOL) equipment to
retailers for recycling, making the manufacturer responsible for the cost of recycling. In response
to this and other proposed legislation, several major Japanese electronics manufacturers initiated
their own roadmaps and publicly announced accelerated plans to eliminate lead-solder from
certain or all products. Companies making this commitment included Matsushita, Sony, Toshiba,
and Hitachi, with others likely to follow. Currently, Matsushita is successfully marketing lead-
free consumer products; Sony has a goal of eliminating lead from products, except for a few
uses, by the end of March 2005; and Toshiba's general policy is that all products are available
lead-free by the end of 2003. Supplementary to this, published on 13 February, 2003, the
European Directive on Waste Electrical and Electronic Equipment (WEEE) requires the
substitution of lead, amongst other listed heavy metals, in new electrical and electronic
equipment. The Directive is to become effective on 13 August, 20051. These changes in
international legislation will potentially eliminate lead from electronic devices produced in the
European Union and by foreign competition, thus, driving the implementation of lead-free
assembly around the world.
As a result of international legislative and market pressures to phase-out the use of tin-lead
solders, the use of lead-free solder alternatives in electronic products manufactured in the U.S.
has also received increasing attention. This worldwide shift to lead-free products gives rise to
several questions, key among them is the performance of alternative solders. In search of a
substitute alloy(s), researchers have conducted numerous performance tests on a host of
alternative alloys.
A large number of the alternative solders being considered as a replacement for Sn-Pb are rich in
tin and coupled with additional elements to enhance alloy characteristics. Solder performance is
determined by testing the alternative solder for characteristics such as joint strength, fatigue
resistance2, and high temperature life. Preliminary literature searches provided some basic
information on the elements considered for lead-free solder alloys. For example, silver is
comparatively available in abundance, however, it is high in cost. Bismuth poses potential
problems with supply as well as embrittlement (as lead contamination drops its melting
temperature causing joint embrittlement). Copper on the other hand, is readily available as well
as soluble in tin. Additionally, copper-containing tin alloys have been used by the industry in
the past.
1 U.K. Department of Trade and Industry, 2005, Sustainable Development and Environment; accessed at:
http://www.dti.gov.uk/sustainabilitv/weee/
2 Fatigue resistance: The maximum stress that a material can endure for a given time without breaking.
F-2
-------
F.2 LITERATURE SUMMARY
Research of alternative solders' performance was found to be taking place on a large scale by
multi-stakeholder partnerships and industry sectors, academia, and non-regulatory federal
agencies (for example, the National Institute of Science and Technology). It was also found that
a large number of studies were ongoing with performance data that is yet to be released. For
example, the High Density Packaging (HDP) User Group International studies regarding solder
reliability characterization was an ongoing research project during the time this appendix was
written; results were later released in 20033.
The studies that were reviewed for this appendix were found difficult to compare; studies
differed in their focus and often considered different alloy combinations and performance tests.
Additionally, resulting data were presented in varying metrics. Such disparities in the available
data hindered the comparability of performance results across sources.
In order to present these data in the most useful format, a summary of each paper is provided in
this section (Section F.2). Select quantitative data from the individual studies have been
presented in Section F.3. Qualitative data have been summarized in Section F.4.
It should be noted that these literature sources often referenced more alloys than those
summarized. In order to remain within the scope of this document, only those alloys of interest
to the Partnership have been presented.
3 Results became available after the research for this appendix concluded. Results were presented in four
papers at the APEX 2003 Conference. The papers presented were: Lead-Free Design, Materials, and Process of High
Density Packages, Joe Smetana, Alcatel; Lead-Free Solder Joint Reliability of High Density Packages - Part I:
Design For Reliability, Walter Dauksher, Ph.D., Agilent; Lead-Free Solder Joint Reliability of High Density
Packages-Part II: Reliability Testing and Data Analysis, John Lau, Ph.D., Agilent Technologies; and Lead-Free
Solder Joint Reliability of High Density Packages - Part III: Failure Analysis, Dongkai Shangguan, Ph.D.,
Flextronics International.
F-3
-------
Table F.2.1: List of the Summarized Literature, Solders Addressed, and the Focus of Each Study
Section
No.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Title
Electronics Manufacturing with
Lead-Free, Halogen Free &
Conductive-Adhesive Materials
Reliability of Solder Joints
Assembled with Lead-Free
Solder
The Solder Programme
Mechanical Properties of Sn-
3 .Omass%Ag-0. 5mass%Cu Alloy
Properties of Lead Free Alloy
and Performance Properties of
Lead Free No-Clean Solder Paste
Lead-FREE Alloys: Fitting the
Square Peg in the Square Hole
Research Update: Lead-Free
Solder Alternatives
AIM: Technical Data Sheet
Materials and Process
Considerations for Lead-Free
Electronics Assembly
Database for Solder Properties
with Emphasis on New Lead- free
Solders
Authors
John H. Lau,
C.P. Wong,
Ning-Cheng Lee,
S.W. Ricky Lee
Masayuki Ochiai,
Toshiya Akamatsu,
Hidefumi Ueda
William J. Plumridge
Yoshiharu Kariya,
William Plumbridge
Quan Sheng,
Sandy Kwiatek
Angela Grusd,
Chris Jorgensen
Jasbir Bath, Carol
Handwerker, Edwin
Bradley
AIM
Karl Seelig and
David Suraski
NIST and CSM
Organization
Agilent Technologies, Inc., Georgia Institute of
Technology, Nin-Cheng Lee, Hong Kong
University of Science and Technology,
respectively
Fujitsu Laboratories Ltd., Japan
The Open University Materials Engineering
Department, UK
The Open University Materials Engineering
Department, UK
OMG Americas
Heraeus Cermalloy, IPC - Association
Connecting Electronics Industries, respectively
National Electronics Manufacturing Initiative
(NEMI)
AIM
AIM
National Institute of Standards & Technology
(NIST) and Colorado School of Mines (CSM)
Solders Addressed
Sn-Cu
/
/
/
/
/
/
/
Sn-Ag-
Cu
/
/
/
/
/
/
/
/
/
/
Sn-Ag-
Cu-Bi
/
/
Study Focus
- Physical properties
- Mechanical properties
- Wetting properties
- Reliability properties
- Mechanical properties at twisting
- Fatigue life subjected to twisting
- Solder ball joints of EGA packages
-Solder joints of QFPs
- Tensile properties
- Fatigue response
- Creep behavior
- Tensile behavior
- Creep behavior
- Mechanical properties
- Creep performance
- Wetting properties
(No-clean solder paste system)
- Physical properties
- Creep/Fatigue
- Wettability
- Physical properties
- Reliability
- Reflow and wave soldering
- Mechanical properties
- Mechanical properties
- Wetting properties
- Fatigue resistance
- Solder joint reliability
- Wave Soldering and SMT applications
- Physical properties
- Mechanical properties
- Wetting properties
- Reliability testing
- Physical properties
- Mechanical properties
- Thermal properties
F-4
-------
F.2.1 Electronics Manufacturing With Lead-Free, Halogen Free & Conductive-Adhesive
Materials
Author(s): John H. Lau, C.P. Wong, Ning-Cheng Lee, S.W. Ricky Lee
Organization: by author: Agilent Technologies, Inc., Georgia Institute of
Technology, Nin-Cheng Lee, Hong Kong University of Science
and Technology, respectively
Publication/Source: McGraw-Hill, Ch. 13: Prevailing Lead-Free Alloys, p. 13.1-13.62
Date: September 2000
DfE Alloys Considered: Sn-Cu, Sn-Ag-Cu, Sn-Ag-Cu-Bi
Summary: This is a comprehensive handbook, covering integrated circuit (1C)
packaging, printed circuit board (PCB)/substrates, assembly of 1C
packages, and novel conductive adhesive materials. Emphasis is on
fundamental principles, engineering data, and manufacturing technologies.
Among others, this source considers the Sn-Cu, Sn-Ag-Cu and Sn-Ag-Cu-
Bi alloys.
Physical properties: Eutectic4 Sn-Cu has the highest melting temperature among prevailing lead-
free solders, suggesting greater difficulty in adopting this alloy. The ternary eutectic composition
(approximately 95.6Sn-3.5Ag-0.9Cu) has a melting point of 217°C, while the melting
temperature for Sn-Ag-Cu-Bi ranges between 207-216°C. Sn-Cu is comparable in surface
tension, electrical resistivity, and density with Sn-Ag, Sn-Ag-Cu and Sn-Ag-Cu-Xdue to the
dominant presence of tin. The hardness however, does vary; that of the ternary alloy is
comparable with Sn-Pb. Bismuth-containing alloys on the other hand exhibit considerably higher
hardness than Sn-Pb due to the precipitation and Bi-dissolution strengthening mechanisms. (For
specific results, see Section F.3, Table F.3.1.a)
Mechanical properties: Eutectic Sn-Cu is lower in tensile strength but higher in elongation than
both eutectic Sn-Ag and Sn-Pb, reflecting its softness and ductility. The tensile strength of Sn-
Ag-Cu is higher than eutectic Sn-Pb. Near the ternary eutectic point, Sn-Ag-Cu alloys are higher
than Sn-Pb in yield strength, shear strength, impact strength, and creep5 resistance. For Sn-Ag-
Cu alloys further away from ternary eutectic composition, the melting temperature (214 to
244°C) increases, as well as the tensile and shear strengths, at the expense of reduction in
elongation. Sn-Ag-Cu-Bi alloys exhibit a higher tensile strength and yield strength, a lower
elongation and a slower creep rate as compared to eutectic Sn-Pb. Shear strength of Sn-Cu is
comparable with Sn-Pb. The creep strength of Sn-Cu is higher than lOOSn, but lower than Sn-
Ag-Cu at both 20 and 100°C. At 25 and 100°C, the time to rupture increases in the following
order: eutectic Sn-Ag, Sn-Ag-Cu < eutectic Sn-Cu < 60Sn-40Pb. The ternary Sn-3.5Ag-0.75Cu
4 Eutectic: having the lowest melting point possible. For Sn-Cu this is implies 99.3% Tin and 0.7% Copper.
5 Creep: under constant load or stress, solder undergoes progressive inelastic deformations over time. This
time dependent deformation is called creep.
F-5
-------
alloy exhibits the longest time to break in creep tests. The tensile strength and creep resistance of
this system increases with an increase in Bi content, then levels off at approximately 7-10% Bi.
Elongation of this system, however, drops rapidly with increasing Bi content until it reaches the
3% level, then it decreases slowly and later levels off with additional Bi content. (For specific
results, see Section F.3, Tables F.3.l.b andF.3.1.c)
Wetting properties: The wetting properties of eutectic Sn-Cu show great potential as
replacements for Sn-Pb in wave and reflow processes. Tests show that the wetting ability of
alloys decreases in the following order: eutectic Sn-Pb > Sn-Ag-Cu > Sn-Ag > Sn-Cu when an
unactivated flux is used. The difference in wetting diminishes when an activated flux is used and
when the wetting time is plotted against superheating. At 260°C, the wetting time descends in
the following order: 96Sn-2.5Ag-lBi-0.5Cu > 96.2Sn-2.5Ag-0.5Sb-0.8Cu > 63Sn-37Pb >
99.3Sn-0.7Cu > 96.5Sn-3.5Ag > 95.5Sn-4Ag-0.5Cu. Wetting time studies conducted by the
meniscograph method presented increasing wetting times for solders in the following order:
63Sn-37Pb < Sn-Ag-Cu-2Bi ~ Sn-Ag-Cu-IBi < Sn-3.5Ag-0.75Cu < Sn-lAg-0.5Cu < Sn-0.7Cu-
0.3Ag < Sn-0.75Cu. However, the wetting time decreases with increasing temperature at a
slightly different rate. Finally, both Sn-Ag-Cu-IBi and Sn-Ag-Cu-2Bi were found to display
wetting behavior that is fairly comparable with 63Sn-37Pb.
The reflow spreading of eutectic Sn-Cu is better than eutectic Sn-Ag, but poorer than eutectic
Sn-Pb. Studies presented the following spreading behavior in decreasing order: 63Sn-37Pb > Sn-
Ag-Cu-4.5Bi, Sn-Ag-Cu-7.5Bi > Sn-3.5Ag-0.75Cu > 99.25Sn-0.75Cu. This source states that
preferably the use of eutectic Sn-Cu should be confined to wave soldering. Varying references
ranged wetting times for the Sn-Ag-Cu alloy from 0.23 to 1.1 seconds, while spreading behavior
ranged between 3.9 to 5 and contact angle ranged between 21 to 47 degrees. The presence of Bi
significantly improves the solder spreading properties of lead-free solders. The Sn-Ag-Cu-Bi
system is outstanding in creep resistance and wetting. (For specific results, see Section F.3,
Table F.S.l.d)
Reliability: The tensile strength of the eutectic Sn-Cu is fairly poor, however its fatigue
resistance is fairly good. One study showed fatigue resistance to increase in the following order:
63Sn-97Pb < 64Sn-36In < 58Bi-42Sn < 50Sn-50In < 99.25Sn-0.75Cu < lOOSn < 96Sn-4Cu.
However, the low-cycle isothermal fatigue (strain 0.2%, 0.1 Hz, ^=0.8, 300 K) performance
shows that the number of cycles to failure for eutectic Sn-Cu is less than one-third of that for
eutectic Sn-Pb, while ternary 95.4Sn-3.1Ag-1.5Cu is significantly greater. For the two cases in
this study which compared Sn-Cu with Sn-Pb, Sn-Cu is consistently better. For a 12-mm, 144-
flexible ball grid array (fleXBGA) assembly at different cycling temperatures, Sn-Ag was the
best, with low or no failure rates. The ternary Sn-4Ag-0.5Cu and Sn-3.4Ag-0.7Cu are similar to
each other and also have better performance than eutectic Sn-Pb. At -40 to 125°C cycling,
however, Sn-Cu performs similarly to Sn-Pb and little improvement is shown for Sn-Ag-Cu over
Sn-Pb. In this range, eutectic Sn-Ag is again the best performer. For temperature cycling
performance in ball grid array (EGA) assembly, eutectic Sn-Ag appears to be superior to Sn-Cu,
F-6
-------
but the opposite is observed for flip-chip assembly. It was reported that the thermal fatigue6 life
for flip-chip assembly descends in the following order: eutectic Sn-Cu > Sn-3.8Ag-0.7Cu,
eutectic Sn-Pb > eutectic Sn-Ag.
The presence of Bi in the lead-free alloys can form a 52Bi-30Pb-18Sn ternary eutectic structure
in the solidified solder joint which has a melting temperature of 96°C. This can be a concern
because the solder joints become weak when subjected to thermal cycling. Lau et al., present
additional data on temperature cycling and heat treatment reliability for Sn-Ag-Cu as well as Sn-
Ag-Cu-Bi. In all the reported results, the Sn-Ag-Cu system is the prevailing alternative to lead-
containing solder. (For specific results, see Section F.3, Tables F.3.1.a to F.3.1.d)
F.2.2 Reliability of Solder Joints Assembled with Lead-Free Solder
Author(s): Masayuki Ochiai, Toshiya Akamatsu, Hidefumi Ueda
Organization: Fujitsu Laboratories Ltd., Atsugi, Japan
Publication/Source: Fujitsu Science Technology Journal, 38, 1, p. 96-101
Date: June 2002
DfE Alloys Considered: Sn-Pb, Sn-Ag-Cu
Summary: The dynamic mechanical properties and reliability of Sn-Ag-Cu were
tested in this study. Compared to the eutectic Sn-Pb solder, the ternary
alloy was found harder to deform and more resistant to hardening, thus
having a longer fatigue life.
Dynamic mechanical properties at twisting, temperature dependence: The shear modulus
(similar to Young's modulus for tension, but indicates the ratio of a shear stress to its resulting
shear strain) of both Sn-Pb and Sn-Ag-Cu decreased with rising temperature. The tin-lead alloy,
however, had a much larger rate of decrease than the ternary alloy, showing that the former
softens faster than the latter with increasing temperatures. It was also found that the Sn-Ag-Cu
solder is more difficult to deform and less likely to harden than the Sn-Pb solder; therefore, it has
a longer fatigue life.
Influence of twisting velocity on dynamic mechanical properties: The tin-lead solder was found
to deform easily at twisting velocities below 1 rad/s (i.e., the range of twisting velocities that
solder joints are subjected to in normal equipment operation). The ternary alloy was shown to be
difficult to deform plastically and thus less likely to harden.
Fatigue life of solders subjected to twisting cycles: The fatigue life of Sn-Ag-Cu solder was
approximately 10,000 cycles, almost twice the fatigue life of the Sn-Pb solder. These results
again indicate that compared to the tin-lead solder, Sn-Ag-Cu is harder to deform plastically and
therefore less likely to harden. This suggests that the ternary alloy has sufficient fatigue
6 Thermal fatigue: premature failure resulting from cycling stresses due to temperature changes.
F-7
-------
resistance for use in electronics assembly.
Solder ball joints of EGA (Ball Grid Array) packages during transition to lead-free soldering:
While in transition, lead and lead-free solders will be used combined in EGA ball joints. Mixing
Sn-Ag-Cu solder with Sn-Pb was found to reduce the fatigue life slightly, maintaining its
superiority to that of the Sn-Pb solder. This suggests sufficient reliability for the mixed solder
joint.
Solder joints ofQFPs (Quad Flat Pack) after transition to lead-free soldering: After the
transition, QFP leads will be plated with lead-free solder, contaminating the joints with lead-free
solder plating. A plating composition of Sn-2Bi presented an approximate 30% reduction in
fatigue life in the Sn-Ag-Cu solder. However, the fatigue life was still superior to that of the Sn-
Pb solder. It was concluded that Sn-Ag-Cu solder joints, with an expected level of bismuth
contamination, will have a fatigue life comparable to current Sn-Pb solder joints. (For specific
results, see Section F.3, Table F.3.2)
F.2.3 The Solder Programme at the Open University Materials Engineering Department:
An Update, 2001
Author(s): William J. Plumbridge
Organization: Materials Engineering Department, The Open University,
Buckinghamshire, U.K.
Publication/Source: Materials Engineering Department, The Open University, UK.
(http://technology.open.ac.uk/materials/mat-hp.htmn
Date: 2001
DfE Alloys Considered: Sn-Pb, Sn-Cu, Sn-Ag-Cu
Summary: The Open University program has been directed towards the testing
performance of solder joints. This source briefly reviews the background
and current status of the research into solder alloys and solder
interconnections for use in electronics. It presents in-depth results for
fatigue, creep and fatigue-creep interactions at high temperatures. It
considers the Sn-0.5Cu and ternary Sn-3.8Ag-0.7Cu alloys.
Tensile Properties: The behavior of the referenced alloys was tested at temperatures between -
10 and 75°C and strain rates between 10"1 and 10'V1. Temperature and strain rate were found to
have a substantial effect on strength. Raising the temperature from -10 to 75°C was found to
reduce the tensile strength by approximately 75% of its value at -10°C (for example, the Sn-Pb
and Sn-Cu alloys fell below 10 MPa at 75°C with a strain rate of 10'V1). Ductility trends with
temperature and strain rate were seen to be small and inconsistent. The Sn-Ag-Cu and Sn-Ag
alloys display the smallest elongation to failure although the ductility values of all the alloys fall
between 20 and 55%. The Sn-0.5Cu solder is usually the weakest and most ductile of the tested
alloys, whereas comparatively, the Sn-Ag-Cu alloy is the strongest (with strength being greatest
at -10°C with the fastest straining rates). This paper finds that the "inter-relationships between
F-8
-------
strength, ductility, temperature and strain rate are complex, and the relative merits of the alloys
may change according to the test conditions."
Fatigue Response: Fatigue tests were carried out at room temperature and at 75°C on Sn-37Pb,
Sn-0.5Cu, and Sn-3.5Ag, exhibiting softening (around 15-20%) when subjected to strain
controlled cycling. The incorporation of a dwell in the strain cycle reduces the number of cycles
to failure in comparison with continuous cycling, irrespective of the dwell location. Generally,
longer dwells result in lower numbers of cycles to failure, with balanced dwells resulting in the
shortest life times.
Creep Behavior: Creep testing was carried out between -50°C and 130°C and times to rupture
were examined up to several thousand hours. The creep behavior of the Sn-0.5Cu alloy is similar
to that of Sn-37Pb at 75°C, while Sn-Ag-Cu exhibits much greater creep resistance that appears
to increase at lower stress levels. Both the silver-containing alloys exhibit a much greater creep
resistance than Sn-37Pb, appearing to increase at lower stress levels. This superior creep
performance is intrinsic to the alloy, as greater life is retained when testing at the same
homologous temperatures to non-silver alloys. At high temperatures (for example, 99°C), the
rupture time of the silver-containing alloys are extremely sensitive to stress, where minor
changes in service conditions could result in profound consequences on creep life. Lead-free
alloys show lower creep ductility as compared with the eutectic Sn-37Pb (approximately 40%) at
75°C. The creep ductility of the silver-containing alloys is the lowest at around 20%, and appears
to be unaffected by applied stress.
Tin Pest: Tin pest can be found in the Sn-0.5Cu alloy when stored for over a year at
temperatures below 13°C. Here white tin transforms to grey tin with a substantial increase in
volume, resulting primarily in surface wart formation and cracking, and finally in complete
disintegration. (For specific results, see Section F.3, Table F.3.3)
F.2.4 Mechanical Properties of Sn-3.0mass%Ag-0.5%mass%Cu Alloy
Author(s): Yoshiharu Kariya and William J. Plumbridge
Organization: Materials Engineering Department, The Open University,
Buckinghamshire, U.K.
Publication/Source: Materials Engineering Department, The Open University, U.K.
Date: Not Provided
DfE Alloys Considered: Sn-Ag-Cu
Summary: This paper investigates the tensile and creep behavior of Sn-3.OAg-0.5Cu
in the rapidly cooled, as-cast state, and compares it with Sn-3.8Ag-0.7Cu
and Sn-3.5Ag. Temperature for the tensile tests ranged between 263K and
398K, and the constant load creep tests were performed at 348K.
The ternary alloys, Sn-3.OAg-0.5Cu and Sn-3.8Ag-0.7Cu, were found to have similar tensile
F-9
-------
strengths, where tensile strength was found to decrease with increasing temperature and with
decreasing strain rate. The tensile strength for the former alloy was 20% higher than Sn-3.5Ag
and double that observed in Sn-0.5Cu at a strain rate of 10'3/s and 348K. Both Sn-3.OAg-0.5Cu
and Sn-3.8Ag-0.7Cu were shown to be superior to the Sn-3.5Ag alloy in this characteristic.
The creep resistance of both the ternary alloys were found to be comparable to each other and
clearly superior to the Sn-Ag alloy. Applied stress had little effect on the creep ductility of the
alloys, with the creep ductility of Sn-3.OAg-0.5Cu being almost equivalent to eutectic Sn-Ag and
the standard Sn-Ag-Cu for this property. (For specific results, see Section F.3, Table F.3.4)
F.2.5 Properties of Lead Free Alloy and Performance Properties of Lead Free No-Clean
Solder Paste
Author(s): Quan Sheng, Charles Bradshaw, Sandy Kwiatek
Organization: OMG Americas, Research Triangle Park, NC
Publication/Source: Presented at IPC SMEMA Council APEX® 2002
(www.goapex.org)
Date: 2002
DfE Alloys Considered: Sn-Pb, Sn-Ag-Cu
Summary: This paper examines the development of a no-clean solder paste system
with the unique needs of the 214-220°C melting point of lead-free alloys.
The properties of the Sn-3.5Ag-0.5Cu no-clean solder paste are compared
to 63Sn-37Pb no-clean solder paste.
Mechanical properties of the two alloys compared favorably, showing slightly lower ultimate
tensile strength and yield strength for the lead-free alloy. Elongation results were inconsistent for
the two alloys. Creep performance of the ternary alloy in bulk was found to be superior to the
63 Sn-Pb alloy. Wetting properties of solder joints made with both pastes were found to be
comparable. Both alloys demonstrated similar static viscosity, dynamic viscosity, tack,
printability, solderability, wide reflow window, and reflow characteristics. Finally, the lead-free
no-clean paste was found to potentially have a longer print life than 63Sn-Pb. From a
performance standpoint, lead-free no-clean Sn-3.5Ag-0.5Cu paste has similar characteristics to
63Sn-Pb, and could be used for PCB applications. (For specific results, see Section F.3, Table
F.3.5)
F-10
-------
F.2.6 Lead-FREE Alloys: Fitting the Square Peg in the Square Hole
Author(s): Angela Grusd and Chris Jorgensen
Organization: Heraeus Cermalloy and IPC - Association Connecting Electronics
Industries
Publication/Source: Circuitree, p. 98-102
Dffi Alloys Considered: September 1999
DfE Alloys Considered: Sn-Cu, Sn-Ag-Cu
Summary: This paper provides an overview of numerous lead-free alloys, examining
temperature ratings, cost, and other factors. It notes that two
alloys-99.3Sn-0.7Cu and 95.5Sn-4.0Ag-0.5Cu-have mid-range melting
temperatures (i.e. between 200°C-230°C), slightly higher than that of tin-
lead, and have been popular choices in the industry, particularly in the
case of reflow soldering.
Tin-Copper: The melting temperature for this alloy (99.3Sn-0.7Cu) is 227°C. This alloy may
prove suitable for high-temperature applications such as those required by the automotive
industry. Testing shows significant improvement in creep/fatigue data over Sn-Pb alloy.
However, the Sn-Ag-Jf alloys are found to perform better in creep testing.
Tin-Silver-Copper: The melting temperature for this alloy (95.5Sn-4.0Ag-0.5Cu) falls between
217-219°C. This temperature range makes it well-suited for high operation temperatures (up to
175°C). The mechanical stability of the joint is degraded when the melting point of the solder is
approached. Thus, elevated temperature cycling produces less damage with higher melting point
solders than it does for Sn-Pb solders (melting point of 183°C). These solders however, do not
wet copper as well as the eutectic Sn-Pb solder using commercial fluxes. However, if the fluxes
are suited for high-temperature use, good fillet formation can be achieved. Wettability can also
be improved using no-clean fluxes when soldering in nitrogen atmosphere. This paper points out
that there are other factors besides performance, such as cost, to consider when selecting a lead-
free alloy. (For specific results, see Section F.3, Table F.3.6)
F.2.7 Research Update: Lead-Free Solder Alternatives
Author(s): Jasbir Bath, Carol Handwerker, Edwin Bradley
Organization: National Electronics Manufacturing Initiative (NEMI)
Publication/Source: Circuits Assembly (www.circuitassembly.com). p. 31-40.
Date: May 2000
DfE Alloys Considered: Sn-Cu, Sn-Ag-Cu
Summary: This paper identifies Sn-3.9Ag-0.6Cu as the recommended choice for
reflow soldering, and Sn-0.7Cu or Sn-3.5Ag as the recommended choices
for wave soldering. It provides an update on current research for lead-free
solder alternatives, and makes note that further investigations are being
conducted on the alternative alloys.
F-ll
-------
Tin-Copper: The eutectic alloy Sn-0.7Cu has a melting temperature of 227°C. Reliability data
indicates it is similar to Sn-37Pb for surface-mount use. Due to a melting temperature 10°C
higher than the ternary Sn-Ag-Cu alloy, Sn-0.7Cu is found undesirable for reflow applications.
This temperature does not present the same concern for wave soldering applications. This paper
makes note of a tendency for fillet lifting when using tin-silver, tin-copper or tin-silver-copper
alloys for wave soldering with lead containing surface finishes, due to the presence of lead. A
significant advantage to using Sn-0.7Cu is the low cost of bar solder.
Tin-Silver-Copper: Alloys within this family with a melting range between 217°C and 222°C are
good substitutes for tin-lead solder. The European IDEALS consortium recommended the Sn-
3.8Ag-0.7Cu alloy as the best lead-free alloy for reflow. Reliability for this alloy composition
was found equivalent to or better than the Sn-Pb and Sn-Pb-Ag alloys.
Within this ternary alloy family, several readily available alloys-Sn-3.5Ag-0.7Cu, Sn-3.6Ag-
0.9Cu, Sn-3.8Ag-0.7Cu, as well as Sn-4Ag-0.5Cu~have melting temperatures near 217°C.
Alloy compositions within the range of Sn-3.5 to 4% (weight) Ag-0.5 to 1% (weight) Cu are
close enough to the eutectic to have similar liquidus7 temperatures, microstructures and
mechanical properties. Bath et al. note that results from literature and solder vendors indicate
that the solderability of the ternary alloy is adequate, however, like all lead-free alloys, worse
than eutectic Sn-Pb.
The NEMI Lead-Free Task Force decided on the Sn-3.9Ag-0.6Cu solder as their
recommendation to the industry for reflow soldering. For wave soldering the recommended
choices are Sn-0.7Cu and Sn-3.5Ag. The NEMI Lead-Free Task Force is continuing to
investigate the performance of these substitutes. Updated information can be found on the NEMI
web page: http://www.nemi.org/newsroom/Presentations/index.html. (For specific results, see
Section F.3, Table F.3.7)
F.2.8 AIM: Technical Data Sheet
Organization: AIM (a global manufacturer of electronics soldering materials)
Publication/Source: AIM: Technical Articles: Lead-free Product Data Sheets
(http://www.aimsolder.eom/leadfreej:dss.cfm?section=assembly)
Dated: Not Provided
DfE Alloys Considered: Sn-Cu, Sn-Ag-Cu
Summary: AIM's Technical Data Sheets present the characteristics of select lead-free
solder alloys. The alloys relevant to the scope of this study are: Sn-0.7Cu,
Sn-3Ag-0.5Cu (LF218™), and Sn-3.8-4.OAg-0.5-0.7Cu (TSC-4).
Tin-Copper: The Sn-0.7Cu alloy is high in purity with a high melting temperature of 227°C.
7 Liquidus: the lowest temperature at which a metal or alloy is completely liquid.
F-12
-------
This eutectic alloy can be used for high temperature lead-free applications.
Tin-Silver-Copper: The Sn-3.8-4.OAg-0.5-0.7Cu alloy has a low melting point of 217-218°C,
good wetting properties, excellent fatigue resistance, excellent solder joint reliability and is
compatible with all flux types. The Sn-3Ag-0.5Cu (LF218™) alloy also has a melting point of
217-218°C and falls under the JEIDA recommendation for lead-free soldering. These two
ternary alloys are near drop-in replacements for eutectic Sn-37Pb in both wave and hand
soldering applications. In wave soldering, both these alloys produce less dross than other solder
alloys, wet well, and provide superior joint strength. In SMT (Surface-Mount Technology)
applications, they produce stronger solder joints, have greater mechanical fatigue resistance, and
are good substitutes for the eutectic tin-lead alloy. Additionally, the Sn-3Ag-0.5Cu and Sn-3.8-
4.0Ag-0.5-0.7Cu no-clean solder pastes pass all Bellcore and IPC specifications. (For specific
results, see Section F.3, Table F.3.8)
F.2.9 Materials and Process Considerations for Lead-Free Electronics Assembly
Author(s): Karl Seelig8 and David Suraski
Organization: AIM
Publication/Source: AIM: Lead-free Articles
(http://www.aimsolder.com/lead_free.cfm?section=articles#2)
Date: Not provided
DfE Alloys Considered: Sn-Cu, Sn-Ag-Cu
Summary: This paper presents analyses of tin-silver, tin-copper, and tin-silver-copper
alloys and compares reliability testing results and process considerations
for them. In order to obtain reliability results, the alloys were subjected to
various thermal and mechanical fatigue tests. The paper also briefly
discusses cost and patent issues related to these solders.
Tin-Copper: While tin-copper solders may be less costly than those containing silver, there are
other issues to consider. The Sn-0.7Cu alloy has a melting temperature of 227°C, prohibiting its
use for many temperature-sensitive applications. It is also a poor wetting alloy compared to other
lead-free solders. This could require the use of nitrogen and aggressive fluxes for many
applications and may result in wetting-related defects. Additionally, Sn-Cu typically has lower
capillary action to draw it into barrels during Plated Through Hole (PTH) Technology and lacks
the fatigue resistance needed for surface mount assembly. Finally, the poor fatigue
characteristics of this alloy may result in field failures, which negates initial cost savings
provided by this less-expensive alloy.
Tin-Silver-Copper: Most of the world seems to be looking to the Sn-Ag-Cu family of alloys as a
Note: Karl Seelig, AIM, has provided a number of technical papers presenting results of lead-free solder
alloys, often presenting overlapping data. It should also be noted that Table F.3.9 combines performance data from
several of these sources (including literature not summarized in this appendix, but listed under References).
F-13
-------
substitute for lead solder alloys. The Sn-4Ag-0.5Cu alloy has a melting point of 218°C and its
base materials are abundantly available. It offers very good fatigue characteristics and good
overall joint strength. Wetting tests demonstrate that alloys with lower silver contents (for
example, Sn-2.5Ag-0.7Cu-0.5Sb) wet stronger and faster than those with higher silver contents
(for example, Sn-4Ag-0.5Cu). However, the silver content of this alloy makes it cost prohibitive
for some applications. Further, silver-containing alloys have experienced failure during fatigue
testing, due to a phase change which causes structural weakness. The low silver alloys can
reduce this problem and offer improved wetting and slightly lower melting temperatures. The
low silver alloys are available worldwide, provide the advantages of the Sn-Ag-Cu family of
alloys, are less cost prohibitive, and avoid the problems associated with Sn-Cu and dual-alloy
processes.
Dual Alloy Assembly: Apart from problems associated with Sn-Cu, intermixing Sn-Ag-Cu and
Sn-Cu solders may result in non-uniformly alloyed solder joints. This may cause the joint to be
susceptible to fatigue failure due to inability to relieve stress and strain. Further, when repairs or
touch-ups are needed, two inventories of alloys are required and operators must be sure not to
mix the alloys.
Reliability - Thermal Cycling Testing: Test boards were built using Sn-0.7Cu and Sn-4Ag-
O.SCu in conjunction with 1206 thin film resistors. The boards were thermally shocked from -40
to 125°C for 300, 400 and 500 15-minute cycles. Post-test inspections show that the Sn-Cu alloy
exhibited some cracked solder joints as a result of poor wetting. In addition, well-formed solder
joints made from the Sn-Cu alloy also showed cracks on the third set of boards cycled to 500
repetitions. The Sn-4Ag-0.5Cu alloy on the other hand, did not show any cracks during testing
up to 500 repetitions, demonstrating that it has significantly superior thermal fatigue resistance
as compared to Sn-Cu. However, it should be noted that the Sn-4Ag-0.5Cu alloy did exhibit
some change in grain structure throughout the joint subsequent to the thermal cycling.
Mechanical Strength-Flex Testing: To test the solders' mechanical strength, test boards were
built using the two alloys in conjunction with 1206 thin film resistors, and were subjected to flex
testing. The test results show that solder joints produced from Sn-0.7Cu cracked during flex
testing, indicating a weak joint that is unable to withstand a wide range of mechanical stresses.
On the contrary, solder joints produced from Sn-4Ag-0.5Cu passed all flex test requirements.
(For specific results, see Section F.3, Table F.3.9)
F-14
-------
F.2.10 Database for Solder Properties with Emphasis on New Lead-free Solders Release
4.0
Organization:
Publication/Source:
Dated:
Alloys Considered:
Summary:
National Institute of Standards & Technology (NIST) and
Colorado School of Mines (CSM)
Properties of Lead-Free Solders
(http://www.boulder.nist.gov/div853/lead%20free/props01.html)
February 11, 2002 (last updated)
Sn-Pb, Sn-Cu, Sn-Ag-Cu, Sn-Ag-Cu-Bi
This database summarizes the mechanical and thermal properties of lead-
free alloys from numerous sources. These data were summarized in a
series of tables. Excerpts of these tables have been presented in Section
F.3, illustrating the properties of the tin-lead solder along with three lead-
free solders in compositions identical or similar to those being examined
by the DfE Partnership.
This source presents data on the shear strengths and wetting angles; mechanical properties such
as ductility, tensile, physical; and thermal properties of multiple solder alloy compositions. (For
specific results, see Section F.3, Tables F.3.10.a. through F.3.10.g)
F-15
-------
F.3 PERFORMANCE TABLES
Table F.3.1.a: Physical Properties of Lead-free Solders, Summary Table
Alloy Family
Alloy Composition
Melting Temperature
(°C)
Surface Tension
(dyne/cm)
Density
(gm/cm3)
Thermal
Conductivity
(W/cm°C)
Electrical Resistivity
(|_iO-cm)
Hardness
(Vickers hardness,
kg/mm2 (HV);
Brinell hardness (BH))
CTE (ppm)
Sn-Pb
63Sn-37Pb
183
380 at 260°C,
417at233°C(air),
464 at 233°C (nitrogen)
8.36, 8.4
0.509 at 30°C,
0.50 at 85°C
14.5,15.0,17
12.8 (HV), 17 (BH)
18.74,25,21,24
Sn-Cu
99.3Sn-0.7Cu
227
491at277°C(air),
461 at277°C
(nitrogen)
7.31
-
10-15
-
-
Sn-Ag-Cu
95.5Sn-
3.8Ag-0.7Cu
217
-
7.5
-
13
15(BH)
14.83
(Sn-3Ag-
4Cu)
95.5Sn-
4Ag-0.5Cu
217-255
-
7.44, 7.39
-
10-15
-
-
95.4Sn-
3.1Ag-1.5Cu
216-217
-
-
-
-
-
Sn-Ag-Cu-Bi
93.3Sn-3.1Ag-
3.1Bi-0.5Cu
209-212
-
7.56
(Sn-2Ag-0.5Cu-7.5Bi)
-
10.6
(Sn-3Ag-3Cu-2Bi)
34.5
(Sn-3Ag-3Cu-2Bi)
-
Source: Lau et al, "Electronics Manufacturing With Lead-Free, Halogen Free & Conductive-Adhesive Materials,'
September 2000.
where alloys had no performance data.
CTE Coefficient of Thermal Expansion
F-16
-------
Table F.3.1.b: Creep Behavior of Lead-free Solders, Summary Table
Alloy Family
Alloy Composition
Creep Strength at
0.1 mm/min
(N/mm2)
Creep at 25°C
(MPa)
20°C
100°C
lOOh to failure
lOOOh to failure
Time to break
(MPa)
Number of Cycles to failure*
Sn-Pb
63Sn-37Pb
-
-
6 (Sn-40Pb)
2.8 (Sn-40Pb)
-
3,650
Sn-Cu
99.3Sn-0.7Cu
8.6
2.1
-
-
-
1,125
Sn-Ag-Cu
95.5Sn-3.8Ag-0.7Cu
13
5
27(Sn-4Ag-0.5Cu)
7.5 (Sn-4Ag-0.5Cu)
323 (Sn-lAg-0.5Cu);
3,849 (Sn-3.5Ag-0.75Cu)
8,936 (95.4Sn-3.1Ag-1.5Cu)
Sn-Ag-Cu-Bi
93.3Sn-3.1Ag-3.1Bi-0.5Cu
-
-
-
-
218(Sn-Ag-Cu-7.5Bi);
1747 (Sn-Ag-Cu-4.5Bi);
2203 (Sn-Ag-Cu-2Bi)
6,522
Source: Lau et al., "Electronics Manufacturing With Lead-Free, Halogen Free & Conductive-Adhesive Materials," September 2000.
* Relative performance in Fatigue Resistance of lead-free solders in low-cycle isothermal fatigue test (strain 0.2%; 0.1 Elz; ,R=0.8; 300K).
Table F.3.1.c: Mechanical Properties of Lead-free Solders, Summary Table
Alloy Family
Alloy Composition
Ultimate Tensile Strength (MPa)
Yield Strength (MPa)
Young's Modulus (GPa)
Elongation (%)
Shear
Strength
(MPa)
at 0.1 mm/min 20°C
100°C
at 0.1 mm/min; 22°C
gap thickness: 76.2|_im;
cooling rate =10°/s 170°C
at 1 mm/min at reflow temperature (RT)
at 1 mm/min at 100°C
By ring-and-plug test
Impact Strength (J/cm2)
Sn-Pb
63Sn-37Pb
19-56*
27.2-37**
38.1(-70°C),
30.2 (20°C),
19.7 (140°C),
32, 33.58, 35,
15.7, 31.03
31-58.87***,
35-176****
23
14
36.5 (Sn-40Pb)
4.5 (Sn-40Pb)
34.5 (Sn-40Pb)
21.6(Sn-40Pb)
40.27
31
Sn-Cu
99.3Sn-0.7Cu
23
37
-
45
20-23
16-21
29.8
10.1
28.5 (Sn-lCu)
21.2(Sn-lCu)
-
-
Sn-Ag-Cu
95.5Sn-3.8Ag-0.7Cu
48;48.5(95.4Sn-
3.1Ag-1.5Cu)
45
-
36.5(95.4Sn-3.1Ag-
l.SCu)
27
17
63.8
25.1
-
-
-
77(Sn-3.5Ag-
0.75Cu)
Sn-Ag-Cu-Bi
93.3Sn-3.1Ag-3.1Bi-0.5Cu
78
85.3 (Sn-2Ag-7.5Bi-0.5Cu)
-
19
-
-
-
-
-
-
-
-
Source: Lau et al., "Electronics Manufacturing With Lead-Free, Halogen Free & Conductive-Adhesive Materials," September 2000.
* The ultimate tensile strength values fall between 19 and 56 MPa (with an average of 39.47 MPa) as per ten references cited by
Lau et al., Table 13.2.
** The yield strength values fall between 27.2 and 37 MPa (with an average of 30.62 MPa) as per four references cited by Lau et
al., Table 13.2.
The elongation values fall between 31 and 52.87% (mean 41.0%) as per six references cited by Lau etal., Table 13.2.
The elongation value according to a reference cited by Lau etal., Table 13.2, ranged between 35-176 percent.
***
****
F-17
-------
Table F.3.1.d: Wetting Properties of Lead-free Solders, Summary Table
Wetting Properties
Alloy Family
Alloy Composition
-
/- * * * i FluxA611,260-280°C
Contact Angle
(degrees) Flux A260HF, 260-280°C
Flux B2508, 260-280°C
Immersion Pb PCB
Immersion Sn PCB
Immersion Ag PCB
Wetting Time
at 260°C NiAu, PCB
(seconds)
OSP1
OSP2
OSP3
A
OSP 3
B
A
Immersion Ag
B
Spread
A
Immersion Pd
B
A
NiAu
B
Sn-Pb
60Sn-40Pb
17
22
32
31
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
62Sn-38Pb
-
-
-
-
0.36(at235°C)
0.27(at235°C)
0.20(at235°C)
0.32(at235°C)
0.20(at235°C)
0.21(at235°C)
0.24(at235°C)
4.55
5
4.7
4.7
4.4
4.7
5
5
Sn-Ag-Cu
95.5Sn-3.8Ag-
0.7Cu
-
-
-
-
0.28
0.23
0.25
0.42
0.26
0.23
0.27
4.2
4.35
4.55
4.8
3.9
3.9
4.4
5
95.5Sn-4.7Ag-
1.7Cu
21
47
45
35
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Sn-Ag-Cu-Bi
Sn-3.3Ag-3Bi-
l.lCu
-
-
-
-
0.24
0.26
0.19
0.44
0.26
0.25
0.27
4
4.45
4.6
4.95
4.4
4.65
4.7
5
Source: Lau et al., "Electronics Manufacturing With Lead-Free, Halogen Free & Conductive-Adhesive Materials," September 2000.
Key:
A Peak 240°C, dwell 60-s for Pb-free, 215°C, 60-s dwell for Sn-Pb-Ag, scale 1 to 5 (best), forced-air convection, air.
B Peak 240°C, dwell 60-s for Pb-free, 215°C, 60-s dwell for Sn-Pb-Ag, scale 1 to 5 (best), 230°C bp VPR.
F-18
-------
Table F.3.2: Tin-Silver-Copper Solder Joint Reliability Compared with Sn-Pb, Summary Table
Alloy
Family
Sn-Pb
Sn-Ag-Cu
Temperature Dependence
Shear Modulus
(at increasing
temperature)
decreases (at a larger
rate than Sn-Ag-Cu)
decreases (at a smaller
rate than Sn-Pb)
Deformation
(twisting velocity
below 1 rad/s)
easily deformed
more difficult to
deform (than Sn-Pb)
Hardening
-
less likely to harden
(than Sn-Pb)
Fatigue life
-
10,000 cycles (almost
twice Sn-Pb)
QFP solder joints
Fatigue life
-
superior (to Sn-Pb) with
controlled Bi-
contamination
Source: Ochiai et al., "Reliability of Solder Joints Assembled with Lead-Free Solder"
Table F.3.3: Tensile, Fatigue, and Creep Properties of Lead-free Alloys, Summary Table
Alloy
Family
Sn-Pb
Sn-Cu
Sn-Ag-Cu
Tensile Properties
Elongation
to Failure
-
-
comparativel
y smallest
Ductility
-
comparatively
most ductile
20-55%
Tensile
Strength***
below lOMPa
below lOMPa
(weakest)
comparatively
strongest
Fatigue Tests*
softening (15-20%)
softening (15-20%)
softening (15-20%)
Creep Properties**
Creep Behavior
(at 75°C)
-
similar to
Sn-37Pb
greater creep
resistance
Time to Rupture
-
lower creep ductility than
Sn-37Pb
extremely sensitive to
stress
Source: William J. Plumbridge, The Solder Programme at the Open University Materials, Engineering Department: An Update, 2001
(http://technologv.open.ac.uk/materials/mat-hp.html')
* At room temperature and at 75°C; subjected to strain controlled cycling.
** Between -50°C and 130°C; times to rupture examined up to several thousand hours.
*** At 75°C with a strain rate of 10V.
Table F.3.4: Tensile and Creep Behavior of Two Sn-Ag-Cu Alloys in the Rapidly Cooled, As-Cast State,
Summary Table
Alloy Composition
Sn-3Ag-0.5Cu
Sn-3.8Ag-0.7Cu
Tensile Strength*
(Strain rate: 10'Vs; 348K)
higher than Sn-3.5Ag (by 20%);
double Sn-0.5Cu
decreases (with increasing temperature &
decreasing strain rate)
Creep Resistance* *
better than Sn-3.5Ag
better than Sn-3.5Ag
Time to Rupture
(Stress component: approx. 14)
similar to Sn-3.8Ag-0.7Cu;
superior than Sn-3.5Ag (x 20)
-
Source: Yoshiharu Kariya and William J. Plumbridge, "Mechanical Properties of Sn-3.0mass%Ag-0.5%mass%Cu Alloy", Materials
Engineering Department, The Open University, U.K.
* Tensile tests ranged between 263K and 3 98K.
** Constant load creep tests were carried out at 348K.
F-19
-------
Table F.3.5: Tin-Silver-Copper Solder Performance Compared with Sn-Pb, Summary Table
Alloy Family
Solder Paste Alloy Composition
Ultimate Tensile Strength
Yield Strength
Creep Performance
Wetting Properties
Viscosity (Static & Dynamic)
Tack
Solderability
Reflow Characteristics
Print life
Sn-Pb
Sn-37Pb
-
-
-
comparable
similar
similar
similar
similar
-
Sn-Ag-Cu
Sn-3.5Ag-0.5Cu
slighter lower than Sn-37Pb
slighter lower than Sn-37Pb
superior to Sn-37Pb
comparable
similar
similar
similar
similar
longer than Sn-37Pb
Source: Quan Sheng, Charles Bradshaw, Sandy Kwiatek, "Properties of Lead Free Alloy and Performance Properties of Lead Free No-
Clean Solder Paste", OMG Americas, 2002
Table F.3.6: Creep Behavior and Wettability of Three Solder Alloys, Summary Table
Alloy
Family
Sn-Pb
Sn-Cu
Sn-Ag-Cu
Alloy Composition
Sn-37Pb
99.3Sn-0.7Cu
95.5Sn-4.0Ag-0.5Cu
Melting Temperature
(°C)
183
227
217-219
Creep / Fatigue
-
superior than Sn-Pb
-
Wettability
-
inferior copper wetting (compared
with eutectic Sn-Pb)
inferior copper wetting (compared
with eutectic Sn-Pb)
Source: Angela Grusd and Chris Jorgensen. "Lead-FREE Alloys: Fitting the Square Peg in the Square Hole", Circuitree, September
1999.
Table F.3.7: Performance of Lead-free Alloys, Summary Table
Alloy
Family
Sn-Cu
Sn-Ag-Cu
Alloy Composition
Sn-0.7Cu
Sn-3.6Ag-0.9Cu
Sn-3.8Ag-0.7Cu
Melting
Range (°C)
227
216-217
-
Liquidus
Temperature
-
-220
-220
Wave
Soldering
optimum
-
-
Reflow
Soldering
undesirable
-
optimum
Reliability*
-
-
similar to / superior than
Sn-Pb and Sn-Pb-Ag
Source: Jasbir Bath et al., "Research Update: Lead-Free Solder Alternatives", National Electronics Manufacturing Initiative (NEMI),
Circuits Assembly (www.circuitassemblv.com). May 2000.
* Reliability testing was carried out from -20 to 125°C for up to 3,000 cycles; and power cycling from 25 to 110°C for 5,000 cycles.
F-20
-------
Table F.3.8: Lead-Free Alloys during Wave Soldering and in SMT Applications, Summary Table
Alloy Family
Alloy Composition
Melting Temperature (°C)
Wave
Soldering
SMT
Applications
Dross Production
Wetting Properties
Joint Reliability
Joint Reliability
Mechanical Fatigue
Resistance
Comments
Sn-Cu
Sn-0.7Cu
227
-
-
-
-
-
used for high
temperature lead-free
applications
Sn-Ag-Cu
Sn-3.8-4.OAg-0.5-0.7Cu
(TSC-4)
217-218
less than other alloys
good
superior
excellent
excellent
no-clean solder pastes pass
Bellcore and IPC
specifications
Sn-3Ag-0.5Cu
(LF218™)
217-218
less than other alloys
good
superior
excellent
excellent
falls under JEIDA recommendation;
no-clean solder pastes pass Bellcore and
IPC specifications
Source: AIM - Technical Data Sheet (http://www.aimsolder.com/leadfree tdss.cfm?section=assembly')
Table F.3.9: Thermal and Mechanical Properties of Lead-free Alloys, Summary Table
Property
Melting Temperature
Relative Wetting Properties
Relative Thermal
Properties**
Joint Strength
Fatigue Resistance
Mechanical Strength -
Flex Testing***
Alloy Composition
Sn-0.7Cu
227
poor
poor
poor
failed
(cracked solder joints)
Sn-4Ag-0.5Cu
218
weaker & slower than CASTIN®*
(lower-Ag content alloy)
good
superior
passed
Source: Karl Seelig and David Suraski, "Materials and Process Considerations for Lead-Free Electronics Assembly"
* The CASTIN® alloy (Sn-2.5Ag-0.8Cu-0.5Sb), consists of the ternary alloy with the addition of a grain -refining and melting
temperature-decreasing dopant.
** Test boards were built using each the alloy in conjunction with 1206 thin film resistors. Thermal shock ranged between -40 to
125°C for 300, 400 and 500 15-minute cycles.
*** Test boards were built using each alloy in conjunction with 1206 thin film resistors and were subjected to flex testing.
F-21
-------
Table F.S.lO.a: Mechanical Properties of Lead-free Alloys Compared With Eutectic Sn-37Pb,
Summary Table
Physical and Mechanical Properties
Alloy Family
Alloy Composition
Density (g/cm3)
Melting Point (°C)
Specific Heat (J/g)
CTE (p-m per m.°C)
Electrical Conductivity (%IACS)*
Electrical Resistivity ((j.O-cm)
Brinell Hardness (HB) or
Vickers Hardness (VHN)
Wettability Ratio
Tensile Strength (20°C)
(N/mm2 at Strain Rate 0.004 s ')
+/- 5 N/mm2
Stress to Rupture
+/- 10 N/mm2
Joint Shear Strength 20°C
(N/mm2 at 0.1 mm/mm)
100°C
Creep Strength 20°C
(N/mm2 at 0.1 mm/min)
100°C
Sn-Pb
Sn-
37Pb
8.4
183
45
19
11.9
14.5
17
(HB)
95,91
40
-
-
23
14
3.3
1.0
Sn-Cu
Sn-
O.VCu
7.3
227-
240
-
-
13
10-15
-
-
-
4,300
1,460
23,20
16,21
8.6
2.1
Sn-Ag-Cu
Sn-3.5Ag-
0.7Cu
-
-
-
-
13
-
-
-
48
-
-
-
-
13
5
Sn-3.8Ag-
0.7Cu
7.5
217
-
-
13
13
15(HB)
-
48
-
-
27
17
13.0
5.0
Sn-4Ag-
O.SCu
7
217-218
-
-
-
-
-
-
-
-
-
-
-
-
-
Sn-Ag-Cu-Bi
Sn-2Ag-
0.5Cu-7.5Bi
8
186-212
-
-
-
-
-
-
-
-
-
-
-
-
-
Sn-3Ag-
3Cu-2Bi
-
-
-
-
-
10.6
34.5
(VHN)
97,96
-
-
-
-
-
-
-
Source: NIST and CSM, "Database for Solder Properties with Emphasis on New Lead-free Solders", February 2002,
(http://www.boulder.nist.gov/div853/lead%20free/props01.html')
CTE: Coefficient of Thermal Expansion
* 100%IACS = 58.00MS/m
F-22
-------
Table F.S.lO.b: Mechanical Properties of Sn-0.7Cu, Sn-3.2Ag-0.8Cu and Eutectic Sn-37Pb,
Summary Table
Alloy
Family
Sn-Pb
Sn-Cu
Sn-Ag-Cu
Alloy
Composition
Sn-37Pb
Sn-0.7Cu
Sn-3.2Ag-0.8Cu
Process*
-
water quenched
(average)
air cooled
water quenched
(average)
air cooled
Yield Strength
(MPa)
27.2
15
16
28
20
Ultimate Tensile
Strength
(MPa)
30.6
19
22
32
30
Uniform
Elongation
(%)
3
5.4
9.1
3.4
6.2
Total
Elongation
(%)
48
20.8
41.2
22.1
26.1
Source: NIST and CSM, "Database for Solder Properties with Emphasis on New Lead-free Solders", February 2002,
(http://www.boulder.nist.gov/div853/lead%20free/props01.html)
* Two processes were carried out: water quenched and air cooled. Four runs were carried out for the water quenched process and
the results were averaged.
Table F.S.lO.c: Strength, Ductility and Tensile Properties of Lead-Free Solder Alloys Compared with
Eutectic Sn-Pb Alloy, Summary Table
Strength, Ductility and Tensile Properties
Alloy Family
Alloy Composition
Elastic Modulus
0.2% Yield Strength
Tensile Strength
Relative Elongation (Total)
Strength Coefficient
GPa
psi
MPa
psi
MPa
%
psi
a
Hardening Exponent
Sn-Pb
Sn-37Pb
15.7
3,950
27.2
4,442
30.6
48
4,917
33.9
0.033
Sn-Cu
Sn-3Cu
-
-
-
6,420
-
-
-
-
-
Sn-Ag-Cu
Sn-0.5Ag-4Cu
-
3,724
25.7
4,312
29.7
27
-
-
-
Sn-3Ag-4Cu
-
6,276
43.3
7,006
48.3
22
-
-
-
Sn-Ag-Bi-Cu
Sn-2Ag-7.5Bi-
O.SCu
-
12,370
85.3
13,440
92.7
12
-
-
-
Sn-2Ag-
46Bi-4Cu
-
9,806
67.6
10,070
69.4
3
-
-
-
Source: NIST and CSM, "Database for Solder Properties with Emphasis on New Lead-free Solders", February 2002,
(http://www.boulder.nist.gov/div853/lead%20free/props01.html)
F-23
-------
Table F.S.lO.d: Shear Strengths, Solidus and Liquidus Temperatures, and Wetting Angles, Summary
Table
Alloy
Family
Sn-Pb
Sn-Cu
Sn-Ag-Cu
Alloy
Composition
Sn-37Pb
Sn-40Pb
Sn-0.7Cu
Sn-3.6Ag-lCu
Sn-3.8Ag-0.7Cu
Sn-4.7Ag-l.7Cu
Shear Strength (MPa)*
Test Temperature
22°C
22°C
170°C
Cooling Rate**
1.5 °/s
-
37.4
_
54
-
47
10°/s
-
36.5
29.8
67
63.8
58
10°/s
-
4.5
10.1
24.4
25.1
21.6
Temperature
Solidus
(°C)
183
183
227
217
217
217
Liquidus (°C)
183
188
_
217.9
-
-
Wetting Angle
(degrees)
-
17
_
-
-
21
Source: NIST and CSM, "Database for Solder Properties with Emphasis on New Lead-free Solders", February 2002,
(http://www.boulder.nist.gov/div853/lead%20free/props01.html')
* Cross-head speed: 0.1 mm/min; gap thickness: 76.2 |am
** Cooling rate in soldering (test) but joints
Table F.3.10.e: Thermal Properties of Candidate Lead-Free Solders, Summary Table
Alloy Family
Sn-Cu
Sn-Ag-Cu
Sn-Ag-Cu-Bi
Alloy Composition
Sn-0.7Cu
Sn-3.2Ag-0.5Cu
Sn-3.5Ag-0.75Cu
Sn-3.8Ag-0.7Cu
Sn-4Ag-0.5Cu
Sn-4Ag-lCu
Sn-3.5Ag-0.7Cu-5Bi
Sn-3.2Ag-l.lCu-3Bi
Liquidus Temperature (°C)
227
218
218
220
-
220
-
240
Reflow Temperature (°C)
245-255
238-248
238-248
238-248
-
238-248
-
230-240
Melting Range (°C)
227
217-218
-
217-220
217-225
217-220
198-213
-
Source: NIST and CSM, "Database for Solder Properties with Emphasis on New Lead-free Solders", February 2002,
(http://www.boulder.nist.gov/div853/lead%20free/props01.html)
F-24
-------
Table F.3.10.f: Ternary Sn-Ag-Cu Elastic Properties vs. Temperature, Summary Table
Elastic Property
Yield Stress
As-Cast
(MPa)
Yield Stress
Aged
(MPa)
Elastic Modulus
As-Cast
(MPa)
Elastic Modulus Aged
(MPa)
Yield Strain
As-Cast
(MPa)
Yield Strain
Aged
(MPa)
Minimum
Mean
Maximum
Minimum
Mean
Maximum
Minimum
Mean
Maximum
Minimum
Mean
Maximum
Minimum
Mean
Maximum
Minimum
Mean
Maximum
Test Temperature (°C)
-25
41.51
41.645
41.78
36.77
38.655
40.54
2863.6
3978.3
5093
3415.9
3495.95
3576
0.011
0.01505
0.0191
0.0165
0.0178
0.0191
25
30.13
31.835
33.54
21.21
21.925
22.64
4956.5
5357.75
5759
3828.7
4312.55
4796.4
0.008
0.00845
0.0089
0.0067
0.00715
0.0076
75
16.45
20.975
25.5
16.97
17.005
17.04
4021.6
4455.5
4889.4
3752.6
4004.8
4257
0.0053
0.0062
0.0071
0.0058
0.00615
0.0065
125
13.47
13.635
13.8
10.71
12.15
13.59
2836.8
3837.25
4837.7
2742.4
3336.3
3930.2
0.0045
0.00565
0.0068
0.0054
0.00555
0.0057
160
9.63
10.19
10.75
10.79
11.35
11.91
2217.3
3309.05
4400.8
2715.7
3663.7
4611.7
0.0047
0.0056
0.0065
0.0049
0.0055
0.0061
Source: NIST and CSM,
(http://www.boulder.nist
"Database for Solder Properties with Emphasis on New Lead-free Solders", February 2002,
gov/div853/lead%20free/props01.htmD
F-25
-------
F.4 QUALITATIVE PERFORMANCE RESULTS
Alloy
Composition*
Comments
Reference
Tin-Copper
Eutectic Sn-Cu
Sn-Cu
Sn-0.7Cu
Sn-0.7Cu
Sn-Cu
Sn-0.7Cu
Sn-3Cu
>• Has the highest melting temperature.
>• Is lower in tensile strength and higher in elongation than Sn-Ag and Sn-Pb.
>• Shear strength is comparable with Sn-Pb.
- Creep strength is higher than lOOSn but lower than Sn-Ag-Cu (at 20 and 100°C).
- Time to rupture is higher than Sn-Ag-Cu but lower than Sn-40Pb (at 25 and 100°C).
>• Wetting properties can potentially replace Sn-Pb in wave and reflow processes.
>• Reflow spreading is better than Sn-Ag but poorer than eutectic Sn-Pb.
>• Is good for wave soldering.
>• Wettability (when using an unactivated flux) is lower than Sn-Pb.
>• Has fairly good fatigue resistance.
>• Tensile strength drops with increasing temperatures.
>• Is weaker and more ductile than Sn-Ag-Cu and Sn-Pb.
>• Creep performance of Sn-0.5Cu is similar to Sn-37Pb and poorer than Sn-Ag-Cu at 75°C.
>• Is suitable for high-temperature applications.
>• Creep/fatigue data is superior to Sn-Pb but inferior to Sn-Ag-X
>• Is the best choice for wave soldering (along with Sn-3.5Ag).
>• Is undesirable for reflow applications.
>• Is similar to eutectic Sn-37Pb for surface-mount use.
>• High melting temperature prohibits alloy use for temperature- sensitive applications.
>• Demonstrates poor wetting alloy (as compared with other lead-free solders).
>• Has a low capillary action to draw it into barrels during PTH technology.
>• Has poor overall fatigue characteristics.
>• Lacks the fatigue resistance needed for surface mount.
>• Cracked during mechanical strength-flex testing indicating a weak joint unable to withstand a wide range of
mechanical stresses.
>• Is cost-effective.
>• Is a good alternative for wave soldering and hand soldering applications.
>• Has poor wetting.
>• Recommended for high-temperature applications only .
Lau et al.
Plumbridge,
William J.
Grusd, Angela and
Chris Jorgensen
Bath et al.
Seelig, Karl and
David Suraski
AIM(a)
AIM(a)
Tin-Silver-Copper
Sn-3.5Ag-0.9Cu
Sn-Ag-Cu
Sn-Ag-Cu
Sn-3Ag-0.5Cu
>• Tensile strength is higher than eutectic Sn-Pb.
>• Is higher than Sn-Pb in yield strength, shear strength, impact strength, and creep resistance (alloys near
eutectic Sn-Ag-Cu).
>• Tensile strength, shear strength, and melting temperature increases while elongation decreases (alloys further
away from eutectic Sn-Ag-Cu).
>• Demonstrates the longest time to break in creep tests (Sn-3.5Ag-0.75Cu).
>• Wettability (when using an unactivated flux) is lower than Sn-Pb but higher than Sn-Cu.
>• Is a prevailing alternative to lead-containing solder.
>• Is difficult to plastically deform and less likely to harden.
>• Fatigue life is longer than Sn-Pb (sufficient fatigue resistance for use in electronics assembly).
>• Displays the smallest elongation to failure.
>• Is stronger than Sn-Cu and Sn-Pb.
>• Has much greater creep resistance than Sn-37Pb.
>• Has lower creep ductility than Sn-37Pb.
>• Potentially the most popular lead-free alloy is Sn-3.8Ag-0.7Cu (patented).
>• Tensile strength decreases with increasing temperature and decreasing strain rate.
>• Tensile strength is similar to Sn-3.8Ag-0.7Cu, and superior than Sn-3.5Ag and Sn-0.5Cu (at 10"3/s and 348K).
>• Creep resistance is comparable to Sn-3.8Ag-0.7Cu and superior to Sn-Ag.
Lau et al.
Ochiai et al.
Plumbridge,
William J.
Kariya, Yoshiharu
and William J.
Plumbridge
F-26
-------
Tin-Silver-Copper (contd.)
Sn-3.5Ag-0.5Cu
Sn-4Ag-0.5Cu
Sn-3.9Ag-0.6Cu
Sn-3Ag-0.5Cu
(LF218™)
Sn-3Ag-0.5Cu
(LF218™)
Sn-3.8-4Ag-0.5-
O.VCu
(TSC-4)
Sn-3.8-4Ag-0.5-
O.VCu
(TSC-4)
Sn-3.5Ag-0.5Cu
Sn-4Ag-0.5Cu
>• Mechanical properties are comparable with Sn-37Pb
>• Has slightly lower ultimate tensile strength and yield strength than Sn-37Pb.
>• Creep performance is superior to Sn-37Pb.
>• Wetting properties is comparable to Sn-37Pb.
>• Has similar static viscosity, dynamic viscosity, tack, printability, solderability, wide reflow window and
reflow characteristics as Sn-37Pb.
>• Has a larger print life than Sn-37Pb.
>• Alloy paste is usable in PCB applications.
>• Is well-suited for high operation temperatures (up to 175°C).
>• Joint mechanical stability degrades when the melting point is approached.
>• Does not wet copper as well as eutectic Sn-Pb when using commercial fluxes.
>• Is the preferred choice for reflow soldering.
>• Demonstrates adequate solderability, yet inferior to Sn-Pb.
>• In line with the International Tin Research Institute alloy range recommendation, thus qualifying for
international standards.
>• Has a low melting point for a lead- free alloy.
>• Lowest cost alloy from the Sn-Ag-Cu family.
>• Best wetting Sn-Ag-Cu alloy.
>• Has excellent solder joint reliability.
>• Is compatible with all flux types.
>• Has excellent mechanical fatigue resistance.
>• Is a virtual drop- in for eutectic Sn-Pb in wave and hand soldering applications.
>• Produces less dross than other solder alloys, wets well, and provides superior joint strength in wave soldering.
>• Produces stronger solder joints, has greater mechanical fatigue resistance, and is a virtual drop-in for the
eutectic Sn-Pb solder in SMT applications.
>• In line with JEIDA recommendation.
>• No-clean solder pastes pass all Bellcore and IPC specifications.
>• In line with JEIDA recommendation.
>• Lowest cost of pure metals for this alloy .
>• Has a low melting point.
>• Demonstrates good wetting.
>• Demonstrates excellent solder joint reliability.
>• Is compatible with all flux types.
>• Demonstrates excellent mechanical fatigue resistance.
>• Is a virtual drop-in for the eutectic Sn-Pb solder in SMT applications.
>• In line with the NEMI recommendation.
>• No-clean solder pastes pass all Bellcore and IPC specifications.
>• Demonstrates similar characteristics as CASTIN® and LF218™.
- Higher cost of metals than CASTIN® and LF218™.
>• Presents a potential silver phase change issues.
>• Has similar characteristics to Sn-3Ag-0.5Cu
>• Is slightly higher cost of metals then Sn-3Ag-0.5Cu.
>• Demonstrates good fatigue characteristics (superior thermal fatigue resistance as compared to Sn-Cu).
>• Has good overall joint strength.
>• Exhibits some change in grain structure during thermal cycling.
>• Passed all mechanical strength-flex test requirements.
>• Sufficient supply of base materials.
Sheng et al.
Grusd, Angela and
Chris Jorgensen
Bath et al.
AIM(b)
AIM(a)
AIM(b)
AIM(a)
AIM(a)
Seelig, Karl and
David Suraski
F-27
-------
Tin-Silver-Copper-Bismuth
Sn-Ag-Cu-Bi
>• Surface tension, electrical resistivity, and density are comparable with Sn-Ag, Sn-Ag-Cu and Sn-Ag-Cu-X
>• Demonstrates superior hardness to Sn-Pb.
>• Has higher tensile and yield strengths, lower elongation, and a slower creep rate than Sn-Pb.
>• Wetting behavior is fairly comparable with Sn-37Pb (with 1 or 2% Bi-content).
>• Outstanding in creep resistance and wetting.
Lau et al.
* Several literature sources cited select characteristics for alloys that differed in composition from that mentioned. Such compositions
have been included in parentheses following the appropriate comment.
F-28
-------
F.5 REFERENCES
AIM(a): "AIM Lead-Free Soldering Guide: Alloys, Chemistries, Data, Experience,
Consultation",
(http://www.aimsolder.com/techarticles/AIM%201ead-free%20guide.pdf?section=assembly)
AIM(b): "Technical Data Sheet: Technical Articles: Lead-free Product Data Sheets",
(http://www.aimsolder.com/leadfree_tdss.cfm?section=assembly)
Bath, Jasbir, Carol Handwerker, Edwin Bradley, May 2000: "Research Update: Lead-Free
Solder
Alternatives", Circuits Assembly, p. 31-40, www.circuitassembly.com.
Grusd, Angela and Chris Jorgensen, September 1999: "Lead-FREE Alloys: Fitting the Square
Peg in the Square Hole", Circuitree, p. 98-102.
Kariya ,Yoshiharu and William J. Plumbridge: "Mechanical Properties of Sn-3.0mass%Ag-
0.5%mass%Cu Alloy", Materials Engineering Department, The Open University, U.K.
Lau, John H., C.P. Wong, Ning-Cheng Lee, and S.W. Ricky Lee, September 2000: "Electronics
Manufacturing With Lead-Free, Halogen Free & Conductive-Adhesive Materials", McGraw-
Hill, Ch. J3: Prevailing Lead-Free Alloys, p. 13.1-13.62.
National Institute of Standards & Technology (NIST) and Colorado School of Mines (CSM),
February 11, 2002 (last updated): "Database for Solder Properties with Emphasis on New
Lead-free Solders Release 4.0", Properties of Lead-Free Solders
(http://www.boulder.nist.gov/div853/lead%20free/props01.htmn.
Ochiai, Masayuki, Toshiya Akamatsu, Hidefumi Ueda, June 2002: "Reliability of Solder Joints
Assembled with Lead-Free Solder", Fujitsu Science Technology Journal, 38, 1, p. 96-101.
Plumbridge, William J., 2001: "The Solder Programme at the Open University Materials
Engineering Department: An Update, 2001", Materials Engineering Department, The Open
University, UK, http://technology.open.ac.uk/materials/mat-hp.html
Quan Sheng, Charles Bradshaw, Sandy Kwiatek, 2002: "Properties of Lead Free Alloy and
Performance Properties of Lead Free No-Clean Solder Paste", Presented at IPC SMEMA
Council APEX® 2002 (www.goapex.org).
Seelig, Karl and David Suraski: "Materials and Process Considerations for Lead-Free Electronics
Assembly", AIM: Lead-free Articles,
(http://www.aimsolder.com/lead_free.cfm?section=articles#2)
F-29
-------
APPENDIX G:
LIFE-CYCLE INVENTORY FUEL DATA
Fuel Conversion Data G-l
-------
Table G-l. Fuel conversion factors
Fuel
Diesel Fuel
Heavy fuel oil #6 (residual)
Light fuel oil #2 (distillate)
Liquified petroleum gas (LPG)
Natural Gas
Heat Value
(H)
(MJ/L)
35.875
38.579
36.739
23.276
0.034
Reference
(1)
(1)
(1)
(1)
(3)
Density (D)
(kg/L)
0.845
0.944
0.843
0.542
7.58x ID'4
Reference
(5)
(2)
(2)
(2)
(4)
References:
1. Davis, S.C. 1999. Transportation Energy Data Book, Edition 19. 1999. Center for Transportation
Analysis, Oak Ridge National Laboratory, ORNL 6958, Appendix B, Table Bl. Oak Ridge, Tennessee,
September.
2. Energy Information Administration (EIA) 1999. International Energy Annual 1997. U.S. Department of
Energy. DOE/EIA 0219 (97), Washington, DC. April.
3. Based on: Wang, M. 1999. The Greenhouse Gases, Regulated Emissions, and Energy Use in
Transportation (GREET) Model, Version 1.5. Argonne National Laboratory, University of Chicago.
4. Calculated from: Perry, R.H. and D. Green (Eds.) 1984. Perry's Chemical Engineer's Handbook, 6th
Edition, page 9-15, Table 9-13, and p. 9-16, Table 9-14. McGraw-Hill, Inc., New York, NY.
5. www.afdc.doe.gov/pdfs/fueltable.pdf. Took average of values provided for diesel fuel at 60 degrees F.
G-l
-------
APPENDIX H:
EXAMPLE TOXICITY CALCULATIONS
Example Toxicity Calculation H-l
-------
APPENDIX H:
EXAMPLE TOXICITY CALCULATION
The following example illustrates how toxicity impacts are calculated. Please refer to
Section 3.2. 1 1 of the LFSP report for descriptions of the methodologies for calculating these
impacts.
If two toxic chemicals (e.g., toluene and benzo(a)pyrene) are included in a waterborne
release to surface water from Process A, impact scores would be calculated for the following
impact categories (based on the classification shown in Table 3-1):
•• Chronic public health effects, cancer and non-cancer; and,
•• Aquatic ecotoxicity.
Despite the output types being waterborne releases, the water eutrophi cation and water
quality impact categories are not applicable here because the chemical properties criteria in
Table 3-1 are not met. That is, these chemicals do not contain N or P and are not themselves
wastewater streams.
Using chronic public health effects as an example, impact scores are then calculated for
each chemical as follows:
Cancer effects:
IS
CHP-CA:toluene
ll3CHP-CA:benzo(a)pyrene
Non-cancer effects:
lS
CHP-NC:toluene
CHP-NC :benzo(a)pyrene
X Amt
TCoutput:toiuene
— U\T
n v NC:benzo(a)pyrene
Table H-l presents toxicity data for the example chemicals from Appendix E. The
hazard values and impact scores are calculated as follows:
Table H-l. Toxicity data used in example calculations
Chemical
Toluene
Benzo(a)pyrene
Cancer
Weight of
evidence
D,3
B2, 2A
Slope factor
(SF)
(mg/kg-day)1
None
7.3 (oral)
3.1 (inhalation)
Chronic non-cancer effects
Oral
(mg/kg-day)
100 (NOAEL)
No data
Inhalation
(mg/m3)
4 11.1 (NOAEL)
No data
-------
Cancer effects:
The cancer HV for benzo(a)pyrene is calculated as follows:
Oral: (HVCAoral), = l/(oral NOAEL.)
l/^mlNOAEL^J
HVCAoral:benzo(a)pyrene = 7.3 (mg/kg-day)-1 - 0.71 (mg/kg-day)-1
= 10.3
Inhalation: (HVCAinl^{= inhalation SFt
inhalation Sfmem
HVCAinhalation:benzo(a)pyrene =3.1 (mg/kg-day)-1 -1.7 (mg/kg-day)-1
= 1.82
Thus, the cancer HV is 10.3, the greater of the two values. The cancer HV for toluene is zero
since it has no slope factor and a WOE classification of D (EPA) and 3 (IARC).
Given a hypothetical waterborne release amount of 0.1 kg of benzo(a)pyrene per
functional unit, the impact score for benzo(a)pyrene cancer effects is given by:
l^CHP-CA,W:benzo(a)pyrene ~~ 10.3 X 0. 1
= 1.03 kg cancertox-equivalents of benzo(a)pyrene
per functional unit
Toluene's impact score for cancer is zero since its HV is zero.
Non-cancer effects:
Since no data are available for non-cancer effects of benzo(a)pyrene, a default HV of one
is assigned, representative of mean toxicity.
The non-cancer HV for toluene is calculated as follows:
Oral: (HVNCoral\ = l/(oral NOAEL,)
H(oralNOAELmeJ
= 1/100 mg/kg-day - 1/14.0 mg/kg-day
= 0.140
-------
Inhalation: (HVNCMalation\ = l/(inhal NOAEL.)
\l(inhalNOAELmeJ
= 1/411.1 mg/m3 + 1/68.7 mg/m3
= 0.167
Thus, the non-cancer HV for toluene is 0.167, the greater of the two values.
Given the following hypothetical output amounts:
AmtTC.0:TOLUENE = 1.3 kg of toluene per functional unit
AmtTC.0.BENZO(A)pYRENE =0.1 kg of benzo(a)pyrene per functional unit
The resulting non-cancer impact scores are as follows:
1^CHP-NC,W:TOLUENE ~~ 0.167 X 1.3
= 0.22 kg non-cancer-equivalents of toluene per functional unit
1^CHP-NC,W:BENZO(A)PYRENE ~~ 1x0.1
= 0.1 kg non-cancer-equivalents of benzo(a)pyrene
per functional unit
If these were the only outputs from Process A relevant to chronic public health effects,
the total non-cancer impact score for this impact category for Process A would be:
ISCHP-NC:PROCESS_A = IScHP-NC-W:TOLUENE + ISCHP_NC -W:BENZO(A)PYRENE
= 0.22 + 0.1
= 0.23 nkg non-cancertox-equivalents per functional unit
for Process A.
If the product system Y contained three processes altogether (Processes A, B, and C), and
the non-cancer impact scores for Process B and C were 0.5 and 1.0, respectively, impact scores
would be added together to yield a total impact score for the product system relevant to chronic
public non-cancer health effects:
TS = TS + TS + TS
1C5CHP-NC:PROFILE_Y 1JCHP-NC:PROCESS_A ^ ll3CHP-NC:PROCESS_B ^ ll3CHP-NC:PROCESS_C
= 0.23+0.5 + 1.0
= 1.73 kg non-cancertox-equivalents per functional unit
for Profile Y.
An environmental profile would then be the sum of all the processes within that profile for each
impact category.
------- |