U.S.EPA
                 Electronic Industries Alliance

-------

-------
                                      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, 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 on to determine applicable regulatory requirements.
               i                For More Information

       To learn more about the Design for the Environment (DfE) Computer Display Project or
the DfE Program, please visit the DfE Program web site at:

                                    www.epa.gov/dfe

The DfE web site also contains the document, Computer Display Industry and Technology
Profile, which contains an overview of the cathode ray tube (CRT) and liquid crystal display
(LCD) computer monitor markets, and explains the basic operation and manufacturing of CRTs
and thin-film transistor LCDs.

       To obtain copies of DfE Program technical reports, pollution prevention case studies, and
project summaries, please contact:

                   Pollution Prevention Information Clearinghouse (PPIC)
               :            U.S. Environmental Protection Agency
               ;            1200 Pennsylvania Aye., N.W. (7407)
                                 Washington, DC 20460
                                 Phone: (202) 260-1023
                                   Fax:(202)260-4659
                                  E-mail: ppic@epa.gov
                    Web site: www.epa.gov/opptintr/library/ppicdist.htm

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

               !                 - www.cleanproducts.org

-------
                                Acknowledgments

       This life-cycle assessment (LCA) was prepared under 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), by the University of Tennessee (UT) Knoxville Center for Clean Products
and Clean Technologies, with assistance from numerous UT students and staff. The authors
would like to acknowledge the outstanding contributions of Rajive Dhingra, UT Center for
Clean Projects and Clean Technologies, who researched display end-of-life issues for this
project; Mary Swanson, UT Center for Clean Products and Clean Technologies, who provided
guidance on life-cycle impact assessment methodology and assisted with the Qualitative Risk
Screening chapter of this document; Tom Barnett, Philip Gillenwater, Brent Haywood, and
Chip Lasher, UT graduate research assistants who helped perform the technical work; and
Louis Allen, Debbie Bower, Tina Cordy, Sherry Estep, Gail Farris, Margaret Goergen, and
Claire VanRiper-Geibig, who assisted with document production.

       This document was produced as part of the DfE Computer Display Project, 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 and Heather Bowman, Electronic
Industries Alliance, Core Group Co-Chairs; Dipti Singh, Technical Lead and Technical
Workgroup Co-Chair, U.S. EPA, OPPT, DfE Branch; Frank Marella, Sharp Electronics Corp.,
Technical Workgroup Co-Chair; Maria Socolof and Lori Kincaid, University of Tennessee
Center for Clean Products and Clean Technologies; John Lott, Dupont Electronic Materials;
Bob Pinnel, U.S. Display Consortium; Greg Pitts, Ecolibrium; Doug Smith, Sony Electronics
Inc.; Ted Smith, Silicon Valley Toxics Coalition; David Thompson, Matsushita Electronic
Corporation of America; and Dani Tsuda, Apple Computer, Inc. Many thanks also to the
industry representatives and other interested parties who participated in the project's Technical
Workgroup, especially Ross Young of Display Search, and representatives from Sharp
Electronics in Japan.

       We would like to acknowledge Kazuhiro Goto, David Kahaner, and Wally Lopez
 of the Asian Technology Information Program for their work in collecting life-cycle inventory
data from manufacturers in Japan and Korea. Much of the project's success is due to their
diligence.

-------
                EPA Design for the Environment Workgroup
      We would like to express appreciation to the U.S. EPA Office of Pollution Prevention
and Toxics Design for the Environment Workgroup, who provided valuable expertise in the
development of this LCA and reviewed all draft documents.
            . Andrea Blaschka
             Susan Dillman
             Conrad Flessner
             FranklynHall
             Susan Krueger
Tim Lehman
FredMetz
Dave Monroe
Jerry Smrchek
                            Participating Companies

      We would like to thank the companies that provided life-cycle inventory data to the
project. The LCA could not have been completed without their participation.
      American Video Glass Company
      T. Chatani and Co., Ltd.
      Chisso Corporation
      Eizo Nariao Corporation
      Harison Electric Co., Ltd.
      Hoshideh and Philips Display Corporation
      Hyundai|Electronics Industries Co., Ltd.
      liyama Electric Co., Ltd.
      Matsushita Electric Industrial Co., Ltd.
      Merck Japan Ltd.
      Mitsubishi Electronic Co., Ltd.
Nippon Denyo Co., Ltd.
Nippon Electric Glass Co., Ltd.
Polaroid Corporation
Samsung Electronics
Sharp Corporation
Sony Corporation (Japan)
Sony Electronics Inc. (U.S.)
Stanley Electric Co., Ltd.
Techneglas
Toppan Printing Co., Ltd.
Toshiba Display Technology Co., Ltd.

-------

-------
                                      Contents

EXECUTIVE SUMMARY	ES-1

                 i
CHAPTER 1: GOAL DEFINITION AND SCOPE
       1.1    Background	1-1
       1.2    Introduction	1-4
             1.2.1!   Purpose	1-4
             1.2.2   Previous Research	1-4
             1.2.3   Market Trends  	1-5
             1.2.4   Needfor the Project		..1-6
             1.2.5   Targeted Audience and Use of the Study	1-6
       1.3    Product System 	1-7
             1.3dj   Functional Unit	1-7
             1.3.2i   Cathode Ray Tube	1-8
                 1   1.3.2.1    CRT technology	1-8
                    1.3.2.2    CRT manufacturing	 1-11
             1.3.3!   Liquid Crystal Display	1-13
                 1   1.3.3.1    LCD technology	1-13
                    1.3.3.2    LCD manufacturing	1-18
       1.4    Assessment Boundaries	1-22
             1.4.1   Life-Cycle Stages and Unit Processes	1-22
             1.4.2   Spatial and Temporal Boundaries	1-22
             1.4.3   General Exclusions	1-25
References	,	1-26


CHAPTER 2: LIFE-CYCLE INVENTORY
       2.1    General Methodology	2-4
             2.1.1   Data Categories	2-4
             2.1.2   Decision Rules			2-6
             2.1.3   Data Collection andData Sources  	2-12
                 i
             2.1.4   Allocation Procedures	 2-13
             2.1.5   Data Management and Analysis Software	2-14
             2.1.6   Data Quality	2-15
             2.1.7   Critical Review	2-16
       2.2   Materials Extraction and Materials Processing (Upstream Life-Cycle Stages) 2-17
             2.2.1   Methodology	2-17
                    2.2.1.1    Upstream materials processes	 2-17
                    2.2.1.2   Electric grids	2-18
                    2.2.1.3   Fuels	2-20
             2.2.2  Data Sources andData Quality	2-20
                 ;   2.2.2.1    Upstream material and fuel processes	2-20
                    2.2.2.2   Electric grids	 2-22

-------
              2.2.3   Limitations and Uncertainties	2-22
                      2.2.3.1    Upstream material and fuel processes	2-23
                      2.2.3.2    Electricity generation data	2-23
       2.3    Product Manufacturing	2-24
              2.3.1   Methodology	2-24
                      2.3.1.1    Identification of processes and manufacturers	2-24
                      2.3.1.2    Data collection questionnaires	2-26
                      2.3.1.3    Allocation	;	2-27
                      2.3.1.4    Aggregating manufacturing data	,	2-27
              2.3.2   Data Sources and Data Quality	2-28
              2.3.3   Limitations and Uncertainties	„. 2-31
                      2.3.3.1    Product system boundary uncertainties	2-31
                      2.3.3.2    Data collection process uncertainties	 2-32
                      2.3.3.3    Data uncertainties	„. 2-32
       2.4    Product Use	„. 2-34
              2.4.1   Methodology	 2-34
                      2.4.1.1    Energy use rate	2-34
                      2.4.1.2    Effective life (baseline lifespan calculation)	 2-35
                      2.4.1.3    Manufactured life (alternative lifespan calculation) 	2-38
                      2.4.1.4    Effective life versus manufactured life 	2-40
              2.4.2   Data Sources and Data Quality	2-41
              2.4.3.   Limitations and Uncertainties	2-41
       2.5    End-of-Life	2-42
              2.5.1   Methodology	2-42
                      2.5.1.1    EOL disposition options	2-42
                      2.5.1.2    Data collection	2-43
                      2.5.1.3    Assumptions	2-44
              2.5.2   Data Sources andData Quality	2-45
              2.5.3   Limitations and Uncertainties	2-45
       2.6    Transportation	2-46
              2.6.1   Methodology	2-46
              2.6.2   Questionnaire Results	2-46
              2.6.3   Data Sources and Data Quality	2-48
              2.6.4   Limitations and Uncertainties	2-48
       2.7    Summary of Life-Cycle Inventory Results	2-49
              2.7.1   Baseline LCI	2-49
                      2.7.1.1    CRT inventory results 	2-52
                      2.7.1.2    LCD inventory results	2-101
              2.7.2   Relative Data Quality	2-155
              2.7.3   Sensitivity Analyses	,  2-155
                      2.7.3.1    Manufactured life scenario	2-156
                      2.7.3.2    Modified glass energy scenario	  2-157
                      2.7.3.3    Modified LCD module energy scenario	2-158
                      2.7.3.4    LCD End-of-life dispositions	2-158
References	2-160
                                            11

-------
CHAPTER 3:  LIFE-CYCLE IMPACT ASSESSMENT
       3.1     Methodology	,-. .,*,	3-1
              3.1.1   Classification  	3-2
              3.1.2   Characterization	3-6
                 :    3.1.2.1   Renewable and nonrenewable resource use	3-8
                 ;    3.1.2.2   Energy use	3-9
                     3.1.2.3   Landfill use  	3-9
                     3.1.2.4   Global warming impacts  	3-10
                     3.1.2.5   Stratospheric ozone depletion 	3-11
                 \    3.1.2.6   Photochemical smog  	3-11
                 ;    3.1.2.7   Acidification  		3-12
                 l    3.1.2.8   Air particulates	3-12
                     3.1.2.9   Water eutrophication	..........	3-13
                 ;    3.1.2.10  Water quality	  3-13
                     3.1.2.11  Radioactivity  	3-14
                 I    3.1.2.12  Potential human health impacts	3-14
                     3.1.2.13  Ecotoxicity	3-19
                 •    3.1.2.14  Summary of impact score equations 	3-20
                 i    3.1.2.15  Aggregation of impact scores	3-23
              3.1.3   Data Sources and Data Quality	3-25
              3.1.4   Limitations and Uncertainties	3-25
       3.2     Data Management and Analysis Software	3-28
       3.3     Baseline LCIA Results ...		3-29
              3.3.1   Summary of Baseline LCIA Results	3-29
              3.3.2   Renewable and Nonrenewable Resource Use 	3-31
                 !    3.3.2.1   Renewable resource use	3-31
                     3.3.2.2   Nonrenewable resource use	3-33,
                 i    3.3.2.3   Limitations and uncertainties	3-35
              3.3.3   Energy Use	  3.37
                 ;    3.3.3.1   Major contributors to the CRT energy use results	3-37
                     3.3.3.2   Major contributors to the LCD energy use results	3-38
                     3.3.3.3   Limitations and uncertainties	3-39
              3.3.4   Landfill Use	3-40
                     3.3.4.1   Solid waste landfill use	3-40
                 '    3.3.4.2   Hazardous waste landfill use	  3-42
                 i    3.3.4.3   Radioactive waste landfill use	3-44
                     3.3.4.4   Limitations and uncertainties	3-46
              3.3.5   Global Warming		3-46
                     3.3.5.1   Major contributors to the CRT global warming results ...  3-47
                     3.3.5.2   Major contributors to the LCD global warming results ...  3-48
                 ;    3.3.5.3   Limitations and uncertainties	3-49
              3.3.6   Stratospheric Ozone Depletion	3-50
                     3.3.6.1   Major contributors to the CRT ozone depletion results ...  3-51
                     3.3.6.2   Major contributors to the LCD ozone depletion results ...  3-52
                 :    3.3.6.3   Limitations and uncertainties	3-53
                                           ni

-------
3.3.7   Photochemical Smog*	 3-55
        3.3.7.1   Major contributors to the CRT photochemical smog
                 results		3-55
        3.3.7.2   Major contributors to the LCD photochemical smog
                 results	3-57
        3.3.7.3   Limitations and uncertainties	3-58
3.3.8   Acidification	3-59
        3.3.8.1   Major contributors to the CRT acidification results  	3-59
        3.3.8.2   Major contributors to the LCD acidification impact
                 results	3-60
        3.3.8.3   Limitations and uncertainties	3-61
3.3.9   Air Particulates		3-62
        3.3.9.1   Major contributors to the CRT air particulates impact
                 results	3-63
        3.3.9.2   Major contributors to the LCD air particulates impact
                 results	3-63
        3.3.9.3   Limitations and uncertainties	3-64
3.3.10  Water Eutrophication	3-65
        3.3.10.1  Major contributors to the CRT water eutrophication
                 impact results	3-65
        3.3.10.2  Major contributors to the LCD water eutrophication
                 impact results	3-66
        3.3.10.3  Limitations and uncertainties	3-66
3.3.11  Water Quality	3-67
        3.3.1L1  Biological oxygen demand (BOD)	3-67
        3.3.11.2  Total suspended solids (TSS)	,.. 3-68
        3.3.11.3  Limitations and uncertainties	3-70
3.3.12  Radioactivity	3-70
3.3.13  Potential Human Health Impacts	3-73
        3.3.13.1  Chronic occupational health effects	3-73
        3.3.13.2  Chronic public health effects	3-76
        3.3.13.3  Chronic public health effect scores modified to exclude
                 sulfur dioxide	3-79
        3.3.13.4  Limitations and uncertainties: chronic human health
                 effects			3-81
        3.3.13.5  Aesthetic impacts (odor)	3-83
3.3.14  Ecotoxicity	3-85
        3.3.14.1  Aquatic toxicity	3-86
        3.3.14.2  Terrestrial ecotoxicity 	3-89
        3.3.14.3  Terrestrial toxicity impact scores modified to exclude
                 sulfur dioxide	3-91
        3.3.14.4  Limitations and uncertainties	3-94
3.3.15  Summary of Top Contributors by Impact Category	3-95
                              IV

-------
       3.4    Sensitivity Analyses	...;	 .5*	3-98
             3.4.1   Manufactured Life Scenario	3-98
             3.4.2   Modified Glass Energy Scenario	3-102
             3.4.3   Modified LCD Module Energy Scenario	3-105
             3.4.4   Modified LCD EOL Dispositions Scenario	3-108
             3.4.5   Summary of CRT and LCD Sensitivity Analysis Results  ..	3-108
References	3-111


CHAPTER 4: QUALITATIVE RISK SCREENING OF SELECTED CHEMICALS
       4.1    Lead		.4-4
             4.1.11   Lead in Computer Displays  	,	4-4
             4.1.2   Life-Cycle Inventory Inputs and Outputs of Lead for Computer
                    Displays	4-5
             4.1.3   Computer Display Life-Cycle Impacts for Lead	4-8
             4.1.4   Exposure Summary	4-12
                 ;   4.1.4.1   Occupational exposures	4-12
                 i   4.1.4.2   General population	4-14
                    4.1.4.3 ]  Ecological populations	4-16
             4.1.5   Human Health Effects  	4-16
                    4.1.5.1   Chronic effects (noncancer)	4-16
                    4.1.5.2   Carcinogenicity  	4-18
             4.1.6   Environmental Regulations for Lead	4-18
             4.1:7   Alternatives to Lead Use in Computer Displays	4-19
       4,2    Mercury			.'	4-21
             4.2.1   Mercury in Computer Displays  	4-21
             4.2;2   Life-Cycle Inputs and Emissions of Mercury for Computer
                 '   Displays	4-21
             4.23   Computer Display Life-Cycle Impacts for Mercury	4-23
             4.2.4   Exposure Summary  	4-27
                 '.   4.2.4.1   Occupational exposures	 4-27
                 \   4.2.4.2   General population exposures	 4-28

               • • '   4.2.4.3   Exposure and effects to ecological populations	 4-29
             4.2.5   Human Health Effects  	4-30
                    4.2.5.1   Chronic effects (noncancer)	4-31
                 I   4.2.5.2   Carcinogenicity	4-31
             4.2.6   Environmental Regulations for Mercury	4-32
             4.2.7   Alternatives to Mercury Use in LCDs 	,	4-33
                                          v

-------
       4.3   Liquid Crystals	4_35
             4.3.1   Liquid Crystals in Computer Displays	4-35
             4.3.2   Life-Cycle Inputs and Outputs of Liquid Crystals for Computer
                     Displays	4.35
             4.3.3   LCD Life-Cycle Impacts of Liquid Crystals	4-36
             4.3.4   Exposure Summary	4.37
                     4.3.4.1   Occupational exposures	4-37
                     4.3.4.2   General population	4-38
                     4.3.4.3   Ecological population	4-38
             4.3.5   Human Health Effects 	4-38
                     4.3.5.1   Chronic effects (noncancer)	4-39
                     4.3.5.2   Carcinogenicity and mutagenicity	4-39
             4.3.6   Environmental Regulations for Liquid Crystals		4-39
       4.4   Conclusions	4_40
             4.4.1   Lead	4_40
             4.4.2   Mercury  	4_41
             4.4.3   Liquid Crystals	 4.42
References	4.44


CHAPTERS: SUMMARY AND CONCLUSIONS
       5.1    LCI vs. LCIA	5_1
       5.2   LCIA Results	5.4
             5.2.1   CRT andLCD Baseline Results	5.4
             5.2.2   CRTResults	5.5
             5.2.3   LCD Results	5.7
             5.2.4   CRT vs. LCD Sensitivity Analysis Results  	5-8
       5.3    Uncertainties	5.9
       5.4   Cost and Performance Considerations	5-10
       5.5    Improvement Assessment Opportunities and Targeted Audience
             Uses of Report	5.13
       5.6    Suggestions for Future Research	5-14
APPENDICES (See Volume 2)
                                          VI

-------
                                        Tables
Table 1-1:    Desktop display markets—actual for 1998 and projected for 2001 ...
Table 1-2:    Functional unit specifications  	
Table 1-3:    Preliminary list of CRT display components and materials	
Table 1-4:    LCD components and materials	
Table 1-5:    Geographic coverage for each life-cycle stage	
Table 2-1:    LCI data categories 		
Table 2-2:    Bill of primary material inputs for a 17" CRT monitor	
Table 2-3:    Bill bf primary material inputs for a 15" LCD monitor	
Table 2-4:    Decision rule priorities and scope of analysis for the primary material
             inputs of the 17" CRT monitor  	
Table 2-5:    Decision rule priorities and scope of analysis for the primary material
             inputs of the 15" LCD monitor	
Table 2-6:    Data! types by life-cycle stage	
Table 2-7:    Materials having upstream processes included in the CDP LCA 	
Table 2-8:    Net electricity generation by fuel type	
Table 2-9:    Data sources and data quality for the Ecobilan inventories	
Table 2-10:   Data sources and data .quality indicators for the electric generation
             inventories	
Table 2-11:   Companies that provided primary manufacturing data for the CRT
             and/0r LCD	...
Table 2-12:   Location of companies and number of process data sets	
Table 2-13:   Applicable years of primary data sets	
Table 2-14:   Data! quality indicator percentages for the CRT processes  	
Table 2-15:   Data quality indicator percentages for the LCD processes	
Table 2-16:   Average energy use rates  :	
Table 2-17:   Effective life values		
Table 2-18:   Effective life electricity consumption  	
Table 2-19:   Manufactured life values	
Table 2-20:   Manufactured life electricity consumption  	
Table 2-21:   Distribution of EOL disposition assumptions for the CRT and LCD  .
Table 2-22:   Percent contribution of major materials in the final product	
Table 2-23:   Distribution of transport modes and total distances per mode	
Table 2-24:   Total life-cycle inventory summary - baseline analysis	
Table 2-25:   CRT inventory by life-cycle  stage	
Table 2-26:   Top 199% of CRT primary materials inputs (kg/functional unit)	
Table 2-27:   Top 99% of CRT ancillary materials inputs (kg/functional unit) 	
Table 2-28:   CRT utility inputs	
Table 2-29:   Top '99.99% of CRT air pollutant emissions (kg/functional unit)	
Table 2-30:   Top '99.9% of CRT water pollutant outputs (kg/functional unit)	
Table 2-31:   Topl99.9% of CRT hazardous waste outputs (kg/functional unit) ...
Table 2-32:   Top 99.9% of CRT solid waste outputs (kg/functional unit)	
Table 2-33:   CRT radioactive waste outputs (kg/functional unit) 	
Table 2-34:   Top 99.99% of CRT radioactivity outputs (Bq/functional unit) 	
Table 2-35:   CRT process groups	
.  1-5
.  1-7
.  1-9
1-15
1-22
.  2-5
.  2-7
.  2-8
2-11

2-12
2-13
2-18
2-19
2-21

2-22

2-25
2-26
2-28
2-29
2-30
2-35
2-37
2-38
2-39
2-40
2-43
2-44
2-47
2-50
2-52
2-60
2-60
2-62
2-63
2-64
2-65
2-66
2-67
2-67
2-68
                                           VII

-------
Table 2-36:   CRT manufacturing stage primary material inputs 	2-69
Table 2-37:   CRT manufacturing stage ancillary material inputs	2-71
Table 2-38:   CRT manufacturing stage utility inputs	2-74
Table 2-39:   CRT manufacturing stage air emissions	2-76
Table 2-40:   CRT manufacturing stage water outputs (wastewater and pollutants)	2-85
Table 2-41:   CRT manufacturing stage hazardous waste outputs (kg/functional unit)  	2-88
Table 2-42:   CRT manufacturing stage solid waste outputs (kg/functional unit)  ...	2-90
Table 2-43:   CRT manufacturing stage radioactive waste outputs	2-92
Table 2-44:   CRT manufacturing stage radioactivity outputs  	-.	2-93
Table 2-45:   LCD inventory by life-cycle stage	2-101
Table 2-46:   Top 99.9% of LCD primary material inputs (kg/functional unit)	2-109
Table 2-47:   Top 99% of LCD ancillary material inputs (kg/functional unit)	2-110
Table 2-48:   LCD utility inputs	2-111
Table 2-49:   Top 99.99% of LCD air emissions (kg/functional unit)	2-112
Table 2-50:   Top 99% of LCD water pollutant outputs (kg/functional unit)	2-113
Table 2-51:   Top 99% of LCD hazardous waste outputs (kg/functional unit)	2-114
Table 2-52:   Top 99% of LCD solid waste outputs (kg/functional unit)	 2-115
Table 2-53:   LCD radioactive waste outputs (kg/functional unit)	2-116
Table 2-54:   Top 99.9% of LCD radioactivity outputs (Bq/functional unit)	2-116
Table 2-55:   LCD process groups	2-117
Table 2-56:   LCD manufacturing stage primary material inputs		2-118
Table 2-57:   LCD manufacturing stage ancillary material inputs  	2-121
Table 2-58:   LCD manufacturing stage utility inputs	2-125
Table 2-59:   LCD manufacturing stage air outputs  	2-127
Table 2-60:   LCD manufacturing stage water outputs	2-137
Table 2-61:   LCD manufacturing stage hazardous waste outputs	2-141
Table 2-62:   LCD manufacturing stage solid waste outputs	2-143
Table 2-63:   LCD manufacturing stage radioactive waste outputs	, 2-145
Table 2-64:   LCD manufacturing stage radioactivity outputs  	2-146
Table 2-65:   Relative data quality	2-155
Table 2-66:   List of sensitivity analysis scenarios  	2-156
Table 2-67:   Energy inputs to the LCD monitor/module manufacturing process
             group under the baseline and modified energy scenarios	 2-159
Table 2-68:   Distribution of LCD EOL dispositions in the baseline and modified
             EOL scenarios	2-159
Table 3-1:    Inventory types and properties for classifying inventory items
             into impact categories	3-5
Table 3-2:    LCIA characterization approaches for the CDP	3-7
Table 3-3:    Hazard values for carcinogenicity weight-of-evidence
             if no slope factor is available	3-16
Table 3-4:    Summary of natural resources impact scoring	3-21
Table 3-5:    Summary of atmospheric resource impact scoring 	3-21
Table 3-6:    Summary of water resource impact scoring	3-22
Table 3-7:    Summary of human health and ecotoxicity impact scoring	3-22
Table 3-8:    Toxicity data used in example calculations	3-23
Table 3-9:    Data sources for equivalency factors and hazard values	,.. 3-26
                                         vm

-------
Table 3-10:   Baseline life-cycle impact category indicators	3-30
Table 3-11:   Top 99% of the CRT renewable resource use impact score 	3-32
Table 3-12:   Top 99% of the LCD renewable resource use impact score 	3-33
Table 3-13:   Top. 99% of the CRT nonrenewable resource use impact score  	3-34
Table 3-14:   Top 99%. of the LCD nonrenewable resource use impact score  	3-35
Table 3-15:   LPG and natural gas production geographic and temporal boundaries	3-36
Table 3-16:   Top; 99% of the CRT energy use impact score	3-38
Table 3-17:   Top; 99% of the LCD energy use impact score  			3-39
Table 3-18:   Top|99% of the CRT solid waste landfill use impact score	3-41
Table 3-19:   Topi 99% of the LCD solid waste landfill use impact score	 3-42
Table 3-20:   Top; 99% of the CRT hazardous waste landfill use impact score	 3-43
Table 3-21:   Top;99% of the LCD hazardous waste landfill use impact score  	3-43
Table 3-22:   Top 99% of the CRT radioactive waste landfill use impact score		3-45
Table 3-23:   Top'99% of the LCD radioactive waste disposal impact score	3-45
Table 3-24:   Top ^9% of the CRT global warming impact score	3-48
Table 3-25:   Top|99% of the LCD global warming impact score  	3-49
Table 3-26:   Top:99% of the CRT stratospheric ozone  depletion impact score	3-52
Table 3-27:   Top;99% of the LCD stratospheric ozone  depletion use impact score  	3-53
Table 3-28:   Geographic and temporal boundaries of inventories contributing to the
             CRT and/or LCD ozone depletion indicator results	3-54
Table 3-29:   Top;99% of the CRT photochemical smog impact score  .	„	3-56
Table 3-30:   Top;99% of the LCD photochemical smog impact score  	3-57
Table 3-31:   Top|99% of the CRT acidification impact score	3-60
Table 3-32:   Top [99% to the LCD acidification impact score	 3-61
Table 3-33:   Top!99% of the CRT air particulates impact score	3-63
Table 3-34:   Top'99% of the LCD air particulates impact score	3-64
Table 3-35:   Topl99% of the CRT water eutrophication impact score	3-66
Table 3-36:   Top|99% to the LCD water eutrophication impact score	3-66
Table 3-37:   Top'99% of the CRT BOD impact score	3-68
Table 3-38:   Top|99% of the LCD BOD impact score	3-68
Table 3-39:   Top 99% of the CRT TSS impact score	3-69
Table 3-40:   Top>99% of the LCD TSS impact score  	3-70
Table 3-41:   Top;99% of the CRT radioactivity impact score			3-72
Table 3-42:   Top 199% of the LCD radioactivity impact score	 3-72
Table 3-43:   Top 199% of the CRT chronic occupational health effects score	3-74
Table 3-44:   TopJ99% of the LCD chronic occupational health effects score	3-75
Table 3-45:   Top'99% of the CRT chronic public health effects score  	3-77
Table 3-46:   Top 99% of the LCD chronic public health effects score	3-78
Table 3-47:   Materials contributing greater than 1% of the CRT chronic public health
             effects score (without SO2)	3-80
Table 3-48:   Materials contributing greater than 1% of the LCD chronic public health
             effects score (without SO2)	3-81
Table 3-49:   Top [99% of the CRT aesthetic (odor) impacts score	3-84
Table 3-50:   Top £9% of the LCD aesthetic (odor) impacts score	3-84
Table 3-51:   Top!99% of the CRT aquatic toxicity impact score	3-87
Table 3-52:   Top'99% of the LCD aquatic toxicity impact score	3-88
                                         IX

-------
Table 3-53:   Top 99% of the CRT terrestrial toxicity impact score	3-90
Table 3-54:   Top 99% of the LCD terrestrial toxicity impact score	3-91
Table 3-55:   Materials contributing greater than 1% of the CRT terrestrial toxicity
             impact score (without SO2)  	...	3-93
Table 3-56:   Materials contributing greater than 1% of the LCD terrestrial toxicity
             impact score (without SO^  	3-94
Table 3-57:   Summary of top contributors to CRT impacts by impact category	3-96
Table 3-58:   Summary of top contributors to LCD impacts by impact category	3-97
Table 3-59:   Baseline and sensitivity analysis results—manufactured life  	3-100
Table 3-60:   Baseline and sensitivity analysis results—modified glass energy	3-103
Table 3-61:   Baseline and sensitivity analysis results—LCD modified module energy  .. 3-106
Table 3-62:   Summary of CRT and LCD LCIA comparative results	3-109
Table 4-1:    Computer display parts that contain lead	4-5
Table4-2:    CRT life-cycle stage lead inputs	 4-5
Table 4-3:    LCD life-cycle stage lead inputs  	4-6
Table 4-4:    Total life-cycle lead outputs  to the environment from CRTs	 4-7
Table 4-5:    Total life-cycle lead outputs  to the environment from LCDs  	4-8
Table 4-6:    Summary of lead impact scores for CRTs	4-9
Table 4-7:    Summary of lead impact scores for LCDs	4-10
Table 4-8:    Summary of percent contributions from lead-based materials to
             Individual impact categories	4-12
Table 4-9:    Potential occupational exposure pathways for lead over the life cycle
             of a.computer display	4-13
Table 4-10:   Human health effect regulations and guidelines for lead	4-17
Table 4-11:   Life-cycle stage mercury inputs for LCDs	4-22
Table 4-12:   Life-cycle stage mercury outputs from CRTs  	4-22
Table 4-13:   Life-cycle stage mercury outputs from LCDs  	4-23
Table 4-14:   Summary of mercury-based impact scores by impact category for CRTs .... 4-24
Table 4-15:   Summary of mercury-based impact scores by impact category for LCDs .... 4-25
Table 4-16:   Summary of percent contributions from mercury-based materials to
             individual impact categories	4-26
Table 4-17:   EPA water quality criteria for mercury 	4-30
Table 4-18:   Chronic toxicity reference values for mercury and mercury compounds  .... 4-31
Table 4-19:   Life-cycle stage liquid crystal inputs	4-36
Table 4-20:   Summary of LCD liquid crystal impacts	4-37
Table 5-1:    Baseline LCI vs. LCIA: monitor with greater inventory amount and impact .. 5-2
Table 5-2:    Monitor type with greatest impacts for each life-cycle stage and impact
             category (baseline scenario)	5-5
Table 5-3:    Primary cost and performance  data collected from manufacturers for
             the CDP	5-10
Table 5-4:    Cost and performance data for some currently selling CRTs and LCDs	5-11
Table 5-5:    Life-cycle electricity costs (baseline  scenario)	5-11
                                          x

-------
                                       Figures

Figure 1-1:   Lifejcycle stages of a computer display ....	1-3
Figure 1-2:   CR1< manufacturing process	1-12
Figure 1-3:   Traditional twisted nematic AMLCD manufacturing process	1-19
Figure 1-4:   Front glass AMLCD manufacturing (photolithography)	1-20
Figure 1-5    Rear glass AMLCD manufacturing (photolithography)	 1-21
Figure 1-6:   CRT life cycle 	:	1-23
Figure 1-7:   LCD1 life cycle 			1-24
Figure 2-1:   Unilprocess inventory conceptual diagram	2-1
Figure 2-2:   CRt linked processes		2-2
Figure 2-3:   LCD linked processes	2-3
Figure 2-4: .  Decision rule guidelines		2-10
Figure 2-5:   Mass-based life-cycle inputs	2-50
Figure 2-6:   Energy-based life-cycle inputs	 2-51
Figure 2-7:   Mass-based life-cycle outputs (excluding total wastewater)	 2-51
Figure 2-8:   CRT mass-based material inputs by life-cycle stage	.....	2-53
Figure 2-9:   CRT; energy-based inputs by life-cycle stage		2-53
Figure 2-10:  CRT1 mass-based outputs by life-cycle stage		2-53
Figure 2-11:  CRT primary material inputs by life-cycle stage	2-55
Figure 2-12:  CRT ancillary material inputs by life-cycle stage	2-55
Figure 2-13:  CRT fuel inputs by life-cycle stage	2-55
Figure 2-14:  CRT water inputs by life-cycle stage	2-56
Figure 2-15:  CRT electricity inputs by life-cycle stage	2-56
Figure 2-16:  CRT energy inputs by life-cycle stage	2-56
Figure 2-17:  CRT air outputs by life-cycle stage	 2-57
Figure 2-18:  CRT wastewater outputs by life-cycle stage	2-57
Figure 2-19:  CRT waste pollutant outputs by life-cycle stage	2-57
Figure 2-20:  CRT hazardous waste outputs by life-cycle stage	 2-58
Figure 2-21:  CRT solid waste outputs by life-cycle stage  	2-58
Figure 2-22:  CRT radioactive waste outputs by life-cycle stage 	2-58
Figure 2-23:  CRT radioactivity outputs by life-cycle stage	2-59
Figure 2-24:  CRT manufacturing primary inputs	2-97
Figure 2-25:  CRT manufacturing ancillary inputs	2-97
Figure 2-26:  CRT manufacturing fuel inputs	2-97
Figure 2-27:  CRT manufacturing electricity inputs 	2-98
Figure 2-28:  CRT manufacturing water inputs	2-98
Figure 2-29:  CRT manufacturing energy inputs	2-98
Figure 2-30:  CRT manufacturing air emissions	2-99
Figure 2-31:  CRT manufacturing wastewater outputs	2-99
Figure 2-32:  CRT manufacturing water pollutant outputs	2-99
Figure 2-33:  CRT manufacturing hazardous waste outputs	2-100
Figure 2-34:  CRT manufacturing solid waste outputs	2-100
Figure 3-35:  LCD| mass-based material inputs by life-cycle stage	2-102
Figure 2-36:  LCD energy-based inputs by life-cycle stage	2-102
Figure 2-37:  LCD mass-based material outputs  by life-cycle stage	 2-102
                                           XI

-------
Figure 2-38:   LCD primary material inputs by life-cycle stage	:	2-104
Figure 2-39:   LCD ancillary material inputs by life-cycle stage	2-104
Figure 2-40:   LCD fuel inputs by life-cycle stage	 2-105
Figure 2-41:   LCD electricity inputs by life-cycle stage	2-105
Figure 2-42:   LCD energy inputs by life-cycle stage	2-105
Figure 2-43:   LCD water inputs by life-cycle stage	2-106
Figure 2-44:   LCD air outputs by life-cycle stage	2-106
Figure 2-45:   LCD wastewater outputs by life-cycle stage	2-106
Figure 2-46:   LCD water pollutant outputs by life-cycle stage	„ 2-107
Figure 2-47:   LCD hazardous waste outputs by life-cycle stage	2-107
Figure 2-48:   LCD solid waste outputs by lifercycle stage	2-108
Figure 2-49:   LCD radioactive waste outputs by life-cycle stage	2-108
Figure 2-50:   LCD radioactivity outputs by life-cycle stage"'"..		2-108
Figure 2-51:   LCD manufacturing primary materials inputs	2-150
Figure 2-52:   LCD manufacturing ancillary material inputs	2-150
Figure 2-53:   LCD manufacturing fuel inputs	 2-150
Figure 2-54:   LCD manufacturing electricity inputs	2-151
Figure 2-55:   LCD manufacturing water inputs	2-151
Figure 2-56:   LCD manufacturing energy inputs	2-151
Figure 2-57:   LCD manufacturing air pollutant outputs  	2-152
Figure 2-58:   LCD manufacturing wastewater outputs	 2-152
Figure 2-59:   LCD manufacturing water pollutant outputs	2-152
Figure 2-60:   LCD manufacturing hazardous waste outputs	:. 2-153
Figure 2-61:   LCD manufacturing solid waste outputs	 2-153
Figure 3-1:    Impact classification conceptual model	3-4
Figure 3-2:    Renewable resource use impacts by life-cycle stage	3-31
Figure 3-3:    Nonrenewable resource use impacts by life-cycle stage	 3-33
Figure 3-4:    Energy impacts by life-cycle stage	 3-37
Figure 3-5:    Solid waste impacts by life-cycle stage	3-40
Figure 3-6:    Hazardous waste impacts by life-cycle stage	3-43
Figure 3-7:    Radioactive waste impacts by life-cycle stage	 3-44
Figure 3-8:    Global warming impacts by life-cycle stage  	3-47
Figure 3-9:    Ozone depletion impacts by life-cycle stage	 3-51
Figure 3-10:   Smog formation impacts by life-cycle stage	 3-55
Figure 3-11:   Acidification impacts by life-cycle stage	3-59
Figure 3-12:   Particulates impacts by life-cycle stage	3-62
Figure 3-13:   Eutrophication impacts by life-cycle stage  	3-65
Figure 3-14:   BOD impacts by life-cycle stage	3-67
Figure 3-15:   TSS impacts by life-cycle stage	3-69
Figure 3-16:   Radioactivity impacts by life-cycle stage	3-71
Figure 3-17:   Chronic occupational impacts by life-cycle stage	3-73
Figure 3-18:   Chronic public health impacts by life-cycle stage	3-76
Figure 3-19:   Chronic public health impacts (without SO2) by life-cycle stage  	3-79
Figure 3-20:   Odor impacts by life-cycle stage	3-83
Figure 3-21:   Aquatic ecotoxicity impacts by life-cycle stage	3-86
Figure 3-22:   Terrestrial ecotoxicity impacts by life-cycle stage	3-89
                                           Xll

-------
Figure 3-23:   Terrestrial ecotoxicity impacts (without SO2) by life-cycle	3-92
Figure 3-24:   Energy use: baseline and manufactured life sensitivity results   	3-101
Figure 3-25:   Chronic public health effects: baseline and manufactured life
              sensitivity results	...	  3-102
Figure 3-26:   Energy impacts: baseline and modified glass energy
              sensitivity results	  3-104
Figure 3-27:   Global warming:  baseline and modified glass energy
              sensitivity results	3-105
Figure 3-28:   Energy use: LCD baseline and modified module energy
              sensitivity results	3-107
Figure 3-29:   Chronic occupational effects: LCD baseline and modified
              module energy sensitivity results	3-107
                                            xiu

-------
                          Acronyms & Abbreviations

ABS         acrylonitrile-butadiene-styrene
AMLCD     active matrix Kquid crystal display
AP          acidification potential
ATEP        Asian Technology Information Program
a:Si          amorphous silicon
BOD        biological oxygen demand
BOM        bill of materials
Bq          Bequerel
CAAA       Clean Air Act Amendments
CAS         Chemical Abstract Service
CCEL        cold cathode fluorescent lamp
CDP         Computer Display Project
CFC         chlorofluorocarbon
CFR         Code of Federal Regulations
CHEMS      Chemical Hazard Evaluation for Management Strategies
COD        chemical oxygen demand
CO2          carbon dioxide
CPU         central processing unit
CRT         cathode ray tube
CTSA        Cleaner Technologies Substitutes Assessment
CWA        Clean Water Act
D           density
DEAM       Database for Environmental Analysis and Management
DfE          Design for the Environment
DOE        Department of Energy
DQI          data quality indicator
EF          equivalency factor
EIA          Energy Information Administration
ELF          extremely low frequency
EIME        Environmental Information and Management Explorer
EMF        electric and magnetic field
EOL         end-of-life
EP          eutrophication potential / electroplating
EPA         Environmental Protection Agency
EU          European Union
Fe           iron
FGD         flue gas desulfurization
FPD          flat panel display
GWP        global warming potential
H           heat value
HAP         hazardous air pollutant
HC          hydrocarbon
HCFC        hydrochlorofluorocarbon
HDPE        high density polyethylene
                                        xiv

-------
HEAST      Health Effects Assessment Summary Tables
HF          hydrofluoric acid
HEPS        high-impact polystyrene
HSDB       Hazardous Substances Data Bank
HV          hazard value
HWL        hazardous waste landfill
IARC        International Agency for Research on Cancer
1C           integrated circuit
IPCC        Intergovernmental Panel on Climate Change
IQR         interquartile range
IPS          in-plane switching
IRIS         Integrated Risk Information System
IS           impact score
ISO          International Standards Organization
ITO          indium-tin oxide
kg           kilogram
kWh         Mlowatthour
L            literi
LC          lethal concentration
LCA         life-jsycle assessment
LCD         liquid crystal display
LCI          life-pycle inventory
LCIA        life-pycle impact assessment
LD          lethal dose
LDPE        low density polyethylene
LNG         liquified natural  gas
LOAEL      lowest observed adverse effect levels
LPG         liquified petroleum gas
m3           cubic meter
MACT       maximum achievable control technology
MCC        Microelectronics and Computer Technology Corporation
Mfg         manufacturing
MJ          megajoule
MTBF       meah-time-before-failure
N           nitrogen
NESHAP     National Emission Standards for Hazardous Air Pollutants
Ni           nickel
NB5HS       National Institute of Environmental Health Sciences
NIOSH      National Institute of Occupational Safety and Health
NOAEL      no observed adverse effect levels
NO2         nitrogen dioxide
NOX         nitrogen oxides
NPDES      National Pollutant Discharge Elimination System
NRR        nonrenewable resource -
NSC         National Safety Council
ODP         ozone depletion  potential
                                         xv

-------
OEM
ORNL
OSHA
OTV
P
Pb
PbO
PC
PECVD
PEE
PEL
PET
PM
PM10
PMLCD
PMMA
POCP
POTW
PS
PVA
PWB
QSAR
RAIS
RC
RCRA
REL
RfC
RfD
RIE
RR
RTECS
RWL
SAR
SD
SDWA
SETAC
SF
Sn
SO2
sox
SWL
TCLP
TFT
TORNRC
original equipment manufacturer
Oak Ridge National Laboratory
Occupational Safety and Health Administration
odor threshold value
phosphorus
lead
lead oxide
polycarbonate / personal computer
plasma-enhanced chemical vapor deposition
polyethylene ether
permissible exposure limit
polyethylene terephthalate
particulate matter
particulate matter with an aerodynamic diameter less than 10 micrometers
passive matrix liquid crystal display
poly(methyl methacrylate)
phosphate
photochemical oxidant creation potential
publicly owned treatment works
polystyrene
polyvinyl alcohol
printed wiring board
quantitative structure-activity relationship
Risk Assessment Information System
recycled content
Resource Conservation and Recovery Act
recommended exposure limit
reference concentration
reference dose
reactive ion etching
renewable resource
Registry of Toxic Effects of Chemical Substances
radioactive waste landfill
structure-activity relationship
standard deviation
Safe Drinking Water Act
Society of Environmental Toxicology and Chemistry
slope factor
sulfur hexafluoride
tin                                  .
sulfur dioxide
sulfur oxides
solid waste landfill
Toxicity Characteristic Leachate Procedure
thin-film transistor
The Oak Ridge National Recycle Center
                                         xvi

-------
TN          twisted nematic
tox-kg       toxic-kilograms
TRI         toxip release inventory
TSP         total suspended paniculate
TSS         total suspended solids
U           uranium
                " i
UT          University of Tennessee
UWR        Universal Waste Rule
VDU        video display unit
VOC        volatile organic compounds
WOE  _      weight of evidence
WWT       waste water treatment
                                         xvii

-------
this page intentionally left blank

-------
                                                                 EXECUTIVE SUMMARY
                              EXECUTIVE SUMMARY
                 I                                    J                     .    '
       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 model and assess the life-cycle environmental
impacts of flat panel display (FPD) and cathode ray tube (CRT) technologies that can be used for
desktop computer displays. The DfE Computer Display Project (CDP) report provides a baseline
analysis and the opportunity to use the model as a stepping stone for further analyses and
improvement assessments for these technologies.
       The DfE CDP uses life-cycle assessment  (LCA) as an environmental evaluation tool that
looks 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, there
are four major components of an LCA study:  goal definition and scoping, life-cycle inventory,
impact assessment, iand improvement assessment. The more recent International Standards
Organizations definition of LCA includes the same first three components, but replaces the
improvement assessment component of LCA with a life-cycle interpretation component. LCAs
are generally global and non-site specific in scope.
       The DfE CDP analysis also incorporates some elements of the Cleaner Technologies
Substitutes Assessment (CTSA) methodology (Kincaid et al, 1996), which was developed under
the DfE Program to help businesses make environmentally informed choices and design for the
environment. The CTS A process involves comparative evaluations of the relative human and
ecological risk, energy and natural resource use, performance, and cost of substitute technologies,
processes, products;, or materials.
       This project focuses on the LCA, while including some CTSA-related analyses. It
performs the broad analysis of the LCA, which also incorporates many of the CTSA components
(e.g., risk, energy impacts, natural resource use) into the impact assessment. The analysis also
assesses more specific impacts for selected materials and acknowledges product cost and
performance, typical of a CTSA. As only selected materials are qualitatively evaluated for the
CTSA, this project [is an LCA with a streamlined CTSA component.
       LCAs evaluate the environmental impacts from each of the following major life-cycle
stages: raw materials extraction/acquisition; materials processing; product manufacture; product
use, maintenance, aind repair; and final disposition/end-of-life.  The inputs (e.g., resources and
energy) and outputs^ (e.g., products, emissions, and waste) within each life-cycle stage, as well as
the interaction between each stage (e.g., transportation) are evaluated to determine the
environmental impacts.                                           ,
       In this study and project report, the goal and scope of the CDP are the subject of
Chapter 1. The life-cycle inventory (LCI), which involves the quantification of raw material and
fuel inputs, and 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 the  subject of Chapter 3.  The
improvement assessment or life-cycle interpretation is left to the electronics industry given the
                                         ES-l

-------
 EXECUTIVE SUMMARY
results of this study. The report also includes a qualitative risk screening of selected materials to
represent the CTSA component of the report in Chapter 4. The summary and conclusions are
presented in Chapter 5.

I. GOAL DEFINITION AND SCOPE

Purpose and Need

       The purpose of this study is two-fold: (1) to establish a scientific baseline that evaluates
the life-cycle environmental impacts of active matrix liquid crystal display (LCD) and cathode
ray tube (CRT) technologies for desktop computers, by combining LCA and CTSA
methodologies; and (2) to develop a model that can be used with updated data for future life-
cycle analyses. This study is designed to provide the electronics industry with information
needed to improve the environmental attributes of desktop computer displays. The evaluation
considers impacts related to material consumption, energy, air resources, water resources,
landfills, human toxicity, and ecological toxicity. 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 will also provide the industry and
consumers with valuable information to make environmentally informed decisions regarding
display technologies, and enable them to consider the relative environmental merits of a
technology along with its performance and cost. While there has been some work done on the
life-cycle environmental impacts of either CRTs or LCDs, there has not been a quantitative LCA
of both CRTs and LCDs.
       At present, computer displays using CRTs dominate worldwide markets.  The LCD, first
used predominately in notebook computers, is now moving into the desktop computer market.
CRTs use larger amounts of energy to operate than LCDs, and are associated with disposal
concerns due to leaded glass in the displays. LCDs may consume more energy during
manufacturing and contain small amounts of mercury. Given the expected market growth of
LCDs for computer displays, the various environmental concerns throughout the life cycle of the
computer displays, and the fact that the relative life-cycle environmental impacts  of LCDs and
CRTs have not been scientifically established to date, there is a need for an environmental life-
cycle assessment of both of these types of desktop computer display technologies.

Targeted Audience and Use of the Study

       The electronics industry is expected to be one of the primary users of the study results.
The study is intended to provide industry with an analysis that evaluates the life-cycle
environmental impacts of selected computer display technologies. Another result of the study is
an accounting of the relative environmental impacts of various components of the computer
displays, thus identifying opportunities for product improvements to reduce potential adverse
environmental impacts and costs.  Since this study incorporates a more detailed health effects
component than in traditional LCAs, the electronics industry can use the tools and data to
evaluate the health, environmental, and energy implications of the technologies. With this
evaluation, the U.S. electronics industry may be more prepared to meet the demands of extended
product responsibility that are growing in popularity in the global marketplace, and better able to
                                         ES-2

-------
                                                                  EXECUTIVE SUMMARY
meet competitive challenges in the world market. In addition, the results and model in this study
will provide a baseline LCA upon which alternative technologies can be evaluated. This will
allow for more expedited display-related LCA studies, which are growing in popularity by
industry and may be demanded by original equipment manufacturers (OEMs) or international
organizations.    ;                                                           • .     .
       EPA and interested members of the public can also benefit from the results of the project.
The project has provided a forum for industry and public stakeholders to work cooperatively, and
the results can be used by stakeholders as a scientific reference for the evaluated display
technologies. The .results of the project could also be of value to other industries involved in
designing environmental improvements into the life cycle of consumer products.

Product System

       The product system being analyzed in this study is a standard desktop computer display
that functions as a |graphical interface between computer processing units and users. Besides the
CRT display, several FPD technologies were considered for inclusion in this study. Among the
FPD technologies that exist,  the amorphous silicon (a:Si) thin-film transistor- (TFT) active
matrix LCD technology meets the requirements of the functional unit within the parameters of
this analysis and is assessed in this study.
       The product system is the computer display itself and does not include the central
processing  unit (CPU) of the computer that sends signals to operate  the display.  It is assumed
that the LCDs operate with an analog interface, and therefore are compatible with current CRT
CPUs as plug-and^play alternatives.
       In an LCA,; product systems are evaluated on a functionally equivalent basis. The
functional unit is used as the basis for the inventory and impact assessment to provide a reference
to which the inputs and outputs are related. For this project, the functional unit is one desktop   .
computer display over its lifespan, which meets the functional unit specifications presented in
Table ES-1. The CRT technology is the current industry standard for this product system.

                        Table ES-1. Functional unit specifications
Specification
display size8 '
diagonal viewing area a
viewing area dimensions
resolution !
brightness
contrast ratio i
color i
Measure
17" (CRT); 15" (LCD)
15.9" (CRT); 15" (LCD)
12.8" x 9.5" (122 in2) (CRT); 12" x 9"
(108 in2) (LCD)
1024 x 768 color pixels
200cd/m2
100:1
262,000 colors
  An LCD is manufactured such that its nearest equivalent to the 17" CRT display is the 15" LCD. This is because
the viewing area of a 17" CRT is about 15.9 inches and the viewing area of a 15" LCD is 15 inches. LCDs are not
manufactured to be exactly equivalent to the viewing area of the CRT.
                                          ES-3

-------
EXECUTIVE SUMMARY
Assessment Boundaries

       In a comprehensive cradle-to-grave analysis, the display system includes five life-cycle
stages:  (1) raw materials extraction/acquisition; (2) materials processing; (3) product
manufacture; (4) product use, maintenance and repair; and (5) final disposition/end-of-life.  Also
included are the activities that are required to affect movement between the stages (e.g.,
transportation).
       The geographic boundaries of this assessment depend on the life-cycle stage. This LCA
focuses on the U.S. display market; therefore, the geographic boundary for the use and
disposition stages of displays is limited to the United States. The geographic boundaries for raw
material extraction, material processing, and product manufacture are worldwide (although actual
product manufacturing data were only collected from the United States, Japan, and Korea,
described in Chapter 2 of the report). While the geographic boundaries show where impacts
might occur for various life-cycle stages, traditional LCAs do not provide an actual spatial
relationship of impacts.  That is, particular impacts cannot be attributed to a specific location.
Rather, impacts are generally presented on a global or regional scale.
       Considering the temporal boundaries,, this study addresses impacts from the life cycle of a
desktop computer display manufactured using 1997-2000 technology.  The use and disposition
stages cover a period that represents the life of a display. The lifespan, labeled as the "effective"
life, is defined as the period of time the display is in use by primary, secondary, or even tertiary
users before reaching its final disposition.  The effective life, used as the baseline scenario, is
estimated based on past and current use patterns of displays and represents a realistic estimate of
the lifespan. As the effective life is subject to many variables, including fluctuating market
trends, an alternative lifespan is presented in a sensitivity analysis. The alternative lifespan, or
"manufactured" life, defined as the designed durability of a display (e.g., the time a display or
key display component will operate before failing), is approximated based on the manufacturer's
estimated durability of the display.
       Impacts from the infrastructure needed to support the manufacturing facilities (e.g.,
maintenance of manufacturing plants) are beyond the scope of this study. However, maintenance
of clean rooms used in the manufacturing of LCDs (and other components), which require
substantial amounts of energy, are considered part of the manufacturing process.
       Impacts from the transportation and distribution of materials, products, and wastes
throughout the life-cycle of a display were  originally included in the scope of the CDP LCA.
However, only a small part of the overall transport in the life of a monitor was either reported in
primary data collected for this project or available in secondary data. Inconsistencies between
primary and secondary transportation data sources and the overall poor quality of transport data
prevented an accurate assessment of the transportation inventory and impacts. Therefore,
transportation impacts were excluded from the analysis. Section 2.6 describes transport data
limitations and uncertainties in detail.
                                          ES-4

-------
                                                                EXECUTIVE SUMMARY
II.  LIFE-CYCLE INVENTORY (LCI)             ^

General Methodology

       An 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 DfE CDP, LCI inputs include materials used in the computer display product itself, ancillary
materials used in processing and manufacturing the displays, and energy and other resources
consumed in the manufacturing, use, or final disposition of the displays. Outputs include
products, air emissions, water effluents, and releases to land. Figures ES-1 and ES-2 show the
unit processes that are included in the scope of this project for the CRT and LCD life cycles,
respectively.                         -  ~,
                                        ES-5

-------
  EXECUTIVE SUMMARY
CO


|



O

0
^
%
        CO
        CD
          •g
  Z&S
  § •§ -a
  "n. s> ?*

  lit
T3 -= -c C
•c o> co o>
DJ JQ O >

*d ® m r-
•B S S o
3 S 2 -s
      CO
      =>

   . o 5
   oil
   § 8 .2
     O 05
   o o

  11

la
C CD
8"°
CD
CO

|-§
Q.






1
                                                                                          OS
                                                                                          41
                                                                                          a
                                                                                          •a

                                                                                          |
                                                                                          H
                                                                                          2
                                              ES-6

-------
                                                             EXECUTIVE SUMMARY
ui


I
UJ
u.

3!
9
a
UJ


1

a
a

t

fo
o «
. £
-„ O)
CO c
UI -c
<3 3

2"S

c^£
g S

II
        •o
        d>
        CO
        CD
        O
        U
        Q.«3
      •
      ill
       "8£

        8 •=


        Q. ®

        D) ®





        |


        CO


        o

        T3
        0)
                                                         D)


                                                         1

                                                         O
                                                                i i
                                                                I-If
                                                                ^. 



                                                               CO •>

                                                               "ra-H"

                                                               Q E

                                                               O
0)
E
O




™




AS
2 E

^ CO
O CO
CO
S3


^

V~v_
r^-.
0)
E
h-
UI
0.



-------
EXECUTIVE SUMMARY
       Data were also collected on the final disposition of emissions outputs, such as whether
outputs are released directly to the environment, recycled, treated, and/or disposed.  This
information helps 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, based on the mass, environmental, energy, and functional
significance, 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.  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 the
CDP LCI.  In general, greater emphasis was placed on collecting data and/or developing models
for the product manufacturing, use, and end-of-life life-cycle stages.

                        Table ES-2. Datatypes by life-cycle stage
Life-cycle stage
Upstream
(materials extraction and processing)
Product and component manufacturing
Use
Final disposition
(recycling and/or disposal)
Packaging, transportation, distribution
Data types
Secondary data.
Primary data, except secondary data used for frit.
Modeled using secondary data; maintenance and repair are not
included in the analysis.
Modeled using secondary data plus primary data from CRT
recycling facilities.
Not included.
       In the CDP LCI, data were allocated to the functional unit (i.e., a desktop computer
display over its lifetime) as appropriate.  The data that were collected for this study were either
obtained from questionnaires developed for this project (i.e., primary data) or from existing
databases (i.e., secondary data). LCI data were imported into a Life-Cycle Design Software Tool
developed by the UT Center for Clean Products and Clean Technologies with funding from the
EPA Office of Research and Development and Saturn Corporation. The UT Life-Cycle Design
Software Tool organizes data in such a way that each process inventory is independent.
Customized "profiles" (e.g., the manufacture of a CRT or the whole life-cycle of an LCD) can be
developed by linking processes.
       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 CDP 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, and government, including
EPA's DfE Work Group, 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. The Technical Work Group and EPA's DfE Work Group provided technical
                                          ES-8

-------
                                                                 EXECUTIVE SUMMARY
guidance and were given the opportunity to review all major project deliverables, including the
final LCA report.  '.  •
                 i
Upstream Life-cycle Stage Methodology

       The materials extraction and processing inventories for key materials were obtained from
a secondary LCI database developed by Ecobilan (1999).  The U.S. electric grid inventory was
developed from secondary sources by UT. The U.S. electric grid inventory was then modified,
based on the distribution of fuels used in Japan, to develop the Japanese electric grid, which was
used where manufacturing occurs in Asia. Electricity consumed in the life-cycles of the monitors
was linked to the inventory of inputs and outputs from the U.S. or Japanese electric grid
inventories, as appropriate.                                   . _  	

Manufacturing Stage Methodology
                • i
       The inventories for the product manufacturing life-cycle stage were developed from
primary data collected from manufacturers in Asia and the United States.  The manufacturing
processes included in the study, as well as the number of data sets for each process and the
country of origin of; the data, are presented in Table ES-3. A total of 27 product manufacturing
questionnaires were collected for 11 different processes. Allocation of data to the functional unit
was conducted as necessary. Processes for which we collected more than one company's data
were averaged together.
       The quality bf the manufacturing stage data can be evaluated against two factors:  (1) the
date of the data; and (2) the type of data (i.e., measured, calculated or estimated).  The data
collection phase of ibis project began in 1997 and extended through 2000. Some processes are
more sensitive to production dates than others. Most processes included in this analysis are
mature technologies  and are not expected to differ significantly between the years 1997 and
2000. However, an exception is LCD panel/module manufacturing, which is an evolving and
rapidly advancing process and has seen changes between these years. For the LCD panel and
module manufacturing process, most data were from  1998 and 1999.
       Data quality indicators were developed based on whether data were measured, calculated,
or estimated, as reported in company data questionnaires.  The weighted average of data
collected and their associated DQIs are as follows:  for the CRT, 43% of the data were measured,
34% calculated, 13% estimated, and 10% were not classified. For the LCD, a similar distribution
shows 33% measured, 30% calculated, 23% estimated, and 14% not classified.
                 I
Product Use Life-cycle Stage Methodology

       The baselin^ analysis in this project employs an effective life use stage scenario (the
actual amount of time a monitor is used, by one or multiple users, before it is disposed of,
recycled, or re-manufactured). A manufactured life scenario (the amount  of time either an entire
monitor or a single Component will last before reaching a point where the equipment no longer
functions, independent of user choice) is evaluated in a sensitivity analysis.
                                         ES-9

-------
EXECUTIVE SUMMARY
           Table ES-3. Location of companies and number of process data sets
Process
CRT monitor assembly
CRT (tube) manufacturing
CRT leaded glass manufacturing
CRT frit manufacturing
LCD monitor assembly
LCD panel and module manufacturing "
LCD - glass manufacturing
LCD - color filter patterning on front glass
LCD - liquid crystal manufacturing
LCD - polarizer manufacturing
LCD - backlight unit assembly
LCD - backlight light guide manufacturing
LCD - cold cathode fluorescent lamp (CCFL) manufacturing
PWB manufacturing (for CRT and LCD monitors)
Country of origin of data
(# of data sets)
Japan (2), U.S. (1)
Japan (2), U.S. (1)
Japan (1), U.S. (2)
generic secondary data from the U.S.
Japan (2)
Japan (5), Korea (2)
Japan and U.S. (l)b
Japan (1)
Japan (2)
Japan (1)
Japan (3)
Japan (1)
Japan (1)
generic secondary data from the U.S.
a The LCD panel consists of two glass panels patterned with transistors and color filters, liquid crystals inserted
between the panels, and associated row and column drivers. The LCD module consists of the panel, backlight unit,
and associated electronic boards for the entire panel, and the backlight.
b The average of three data sets for CRT leaded glass manufacturing was modified to remove lead from the
inventory.

       To develop the use stage inventory, energy use rates [e.g., kilowatts (kW)] were
combined with the time a desktop monitor is on during its lifespan (hours/life) to calculate the
total quantity of electrical energy consumed during the use life-cycle stage [e.g., kilowatthours
(kWhyiife]. This was then combined with the electric grid inventory of inputs and outputs per
kWh to make up the use stage inventory per monitor. The effective life scenario models the
actual quantity of hours that an average monitor spends in each of the two primary power
consumption modes (full-on and a lower power state) during its lifetime. Assumptions used to
calculate the kilowatthours per effective life are detailed in Section 2.4.  The total electricity
consumption for each monitor was calculated to be 634 kWh/effective life (2,282 Ml/effective
life) for the CRT and 237 kWh/effective life (853 MJ/effective life) for the LCD.

End-of-Iife (EOL) Methodology

       For the EOL analysis, a monitor is assumed to have reached EOL status when:

•      it has served its useful life;
•      is no longer functional; and/or
•      is rendered unusable due to technological obsolescence.
                                           ES-10

-------
                                                                  EXECUTIVE SUMMARY
       The major EOL dispositions considered in this analysis are as follows:
•      recycling - including disassembly and materials recovery;
•      landfilling - including hazardous [Resource Conservation and Recovery Act (RCRA),
       Subtitle C] .and non-hazardous (RCRA, Subtitle D) landfills;
•      remanufacturing - including refurbishing or reconditioning (to make usable again); 'and
•      incineration - waste to energy incineration.

       The functional unit in this analysis is one monitor; therefore, the different EOL
dispositions were allocated as a probability of one monitor going to a certain EOL disposition.
Data were somewhat scarce on the percent of monitors going to each disposition, especially for
LCD monitors, which  have not as yet reached EOL. After literature research and consultation
with the project^Technical Work Group, as well as various other industry experts, project
partners chose best estimates of disposition distributions (Table ES-4).

     Table ES-4. • Distribution of EOL disposition assumptions for the CRT and LCD
Disposition
Incineration ;
Recycling
Remanufacturing '
Hazardous waste landfill
Solid waste landfill
CRT
15%
11%
3%
46%
25%
LCD
15%
15%
15%
5%
50%
 Sources: NSC, 1999; EPA, 1998; CIA, 1997; EIA, 1999; Vorhees, 2000; TORNRC, 2001
       Primary data were collected for CRT recycling from three companies. The data from
these companies represent facility operations ranging from October 1999 to February 2000.  In
the absence of actual data for LCD recycling, data on a CRT shredding-and-materials-recpvery
process was used to model LCD recycling.
       Hazardous/solid waste landfilling and incineration were developed from secondary data
obtained from Ecbbilan. Although data specific to landfilling and incineration operations for
monitors alone were not available, existing inventories were available for landfilling and
incinerating the following major monitor materials (by weight): steel, glass, plastic, and
aluminum. These inventories were combined, based on the approximate proportion of each
material in a CRT and an LCD, to create individual processes for landfilling and for incineration
(for each monitor type). The majority of the assembled monitors by weight is accounted for in
the overall incineration and landfilling processes.
       Remanufacjoiring data were excluded from the assessment because no single set of
operations could be identified to adequately represent remanufacturing activities that could be
incorporated in our; model.
                                         ES-11

-------
EXECUTIVE SUMMARY
LCI Limitations and Uncertainties

       Several factors contribute to the overall quality of data for each life-cycle stage. For
example, the manufacturing stage includes several different processes that were collected from
several different companies. The quality of one data set from one company may be very different
from that of another company. Relative data quality estimates have been made for each life-cycle
stage, including electricity generation, which is included in the results of more than one life-cycle
stage (Table ES-7).  The table also lists the major limitations associated with each life-cycle
stage.

                Table ES-7. Relative data quality and major limitations
Life-cycle stage
Upstream
Manufacturing
Use
EOL
Electricity generation
Relative data quality
Moderate
Moderate to high
Moderate to high
Low to moderate
High
Major limitations
Used only secondary data, which has undetermined
quality and not originally collected for the purpose of
theCDP.
A few data points remain in question.
Assumptions regarding use patterns were made.
Used only secondary data for incineration and
landfilling processes; no data available for
remanufacturing process.
Used secondary data, however it was collected and
modeled for the CDP, resulting in a higher quality
rating despite use of secondary data.
       Although the manufacturing stage was rated in Table ES-7 as having moderate to high
data quality, some of the few data points that remained in question had large effects on the results
and are therefore described below. Of the data collected from manufacturers, several attempts
were made to verify or eliminate outliers in the data; however, uncertainty in some data remained
due to large data ranges and outliers. Specific data with the greatest uncertainty include:  (1) LCD
glass manufacturing data; (2) CRT and LCD glass manufacturing energy inputs; (3) the
distribution and amount of fuel/electricity inputs for LCD module manufacturing; and (4) the use
of a large amount of liquified natural gas (LNG) as an "ancillary material" in LCD module
manufacturing and not as a fuel.
       Uncertainties in the LCD glass manufacturing data stem from the fact that no LCD glass
manufacturers were willing to supply inventory data.  Therefore, the LCD glass manufacturing
inventory was derived from the CRT glass manufacturing data modified to exclude leaded
compounds from the inventory. Thus, the baseline analysis in this study assumes the energy use
per kilogram of CRT glass and LCD glass are equivalent, which is uncertain. In addition to the
uncertainty in the difference between energy used to manufacture CRT glass and LCD glass, the
energy reported to produce a kilogram of CRT glass varied greatly between the three data sets
received for this project, with the highest total energy value being about 150 times that of the
smallest value.  Due to this large discrepancy and because there were not enough data sets to
evaluate the data for outliers, the glass energy data were evaluated in a sensitivity analysis.  The
high glass energy use values were mostly a function of liquefied petroleum gas (LPG) used as a
fuel.                                           .
                                         ES-12

-------
                                                                 EXECUTIVE SUMMARY
       Other data for which large ranges ,were reported, andswhich could be important to the
results, are energy data from LCD panel/module manufacturing. Energy data provided by six
LCD panel/module^ manufacturers were highly variable in both the distribution of energy sources
and the total energy required to produce one LCD panel.  The percent of energy from electricity
ranged from approximately 3% to 87%, and the total energy per panel ranged from 440 MJ to
7,000 MJ. The aveirage energy use per panel was approximately 2,270 MJ, and the standard
deviation was about 2,910 MJ.
       Given the wide variability in the data and large standard deviation, CDP researchers
evaluated these datk for outliers. One data set was found to be a minor outlier and another was
found to be a major outlier. These outliers were excluded from the averages used in the baseline
analysis, but included in the averages used in a sensitivity analysis (see Section 2.7.3.3).
       Finally, a large amount of LNG (194 kg, on average) was reported to be used as an
ancillary material (riot a fuel) in LCD panel/module manufacturing.  CDP researchers confirmed
this application of LNG and the amount with the company providing the data, but it is still
uncertain due to problems in communication (e.g., the language barrier). This data point
remained in the inventory data set for LCD manufacturing, and was assumed to indeed be an
ancillary material, and not a fuel. Keeping the LNG ancillary material in the inventory will not
affect the energy impact results, since LNG used as an ancillary material is only linked to the
production of that material, and not to the use of it as a fuel.

Baseline LCI Results
                                                >
                 i
       Tables ES-5 and ES-6 present the total quantity of inputs and outputs for each life-cycle
stage of the CRT arid LCD based on input and output types.
                                         ES-13

-------
EXECUTIVE SUMMARY
                      Table ES-5.  CRT inventory by life-cycle stage
Inventory type
Upstream
Mfg
Use
EOL
Total
Units3
Inputs
Primary materials
Ancillary materials
Water
Fuels
Electricity
Total energy
1.58e+01
2.11e+00
5.54e+02
8.00e+00
7.32e+01
3.66e+02
4.21e+02
3.54e+00
1.14e+04
4.28e+02
1.29e+02
1.83e+04
2.19e+02
3.47e+00
1.14e+03
0
2.29e+03
2.29e+03
-3.32e+00
1.076+01
-2.736+01
-2.956+00
2.29e-01
-1.28e+02
6.53e+02
1.986+01
1.31e+04
4.33e+02
2.49e+03
2.08e+04
kg
kg
kg(orL)
kg
MJb
MJb
Outputs
Air pollutants
Wastewater
Water pollutants
Hazardous waste
Solid waste
Radioactive waste
Radioactivity
3.00e+01
1.70e+01
8.12e-01
4.89e+02
9.55e+00
4.396-04
3.80e+07
1.83e+02
1.51e+03
2.01e+01
1.13e+02
8.12e+01
1.80e-04
3.78e+06
4.49e+02
0
7.02e-02
0
8.33e+01
2.28e-03
4.80e+07
2.47e+00
-3.65e+00
-6.18e-02
8.28e+00
-1.666+00
2.29e-07
4.80e+03
6.64e+02
1.52e+03
2.09e+01
9.46e+00
1.72e+02
2.90e-03
8.98e+07
kg
kg(orL)
kg
kg
kg
kg
Bq
1 Per functional unit (i.e., one CRT monitor over its effective life).
b3.6MJ = lkWh
                      Table ES-6. LCD inventory by life-cycle stage
Inventory type
Upstream
Mfg
Use
EOL
Total
Units3
Inputs
Primary materials
Ancillary materials
Water
Fuels
Electricity
Total energy
2.35e+02
1.06e+00
2.63e+02
1.47e+01
3.46e+01
6.33e+02
4.92e+01
2.04e+02
2.15e+03
2.58e+01
3.16e+02
1.44e+03
S.Ole+Ol
1.29e+00
4.25e+02
0
8.53e+02
8.53e+02
-2.19e+00
2.11e+00
-1.80e+01
-1.95e+00
1.626-01
-8.44e+01
3.62e+02
2.08e+02
2.826+03
3.86e+01
1.20e+03
2.84e+03
kg
kg
kg
kg(orL)
MJb
MJb
Outputs
Air pollutants
Wastewater
Water pollutants
Hazardous waste
Solid waste
Radioactive waste
Radioactivity
1.12e+02
8.57e+00
4.60e-01
6.72e-03
1.31e+01
2.21e+01
1.20e+07
6.48e+01
3.12e+03
1.23e+00
4.64e+00
1.26e+01
3.14e+03
1.02e+07
1.68e+02
0
2.62e-02
0
S.lle+Ol
3.11e+01
1.79e+07
1.30e+00
-2.41e+00
-4.096-02
1.64e+00
-4.42e+00
-5.23e+00
3.40e+03
3.46e+02
3.13e+03
1.68e+00
6.29e+00
5.23e+01
3.19e+03
4.01e+07
kg
kg
kg(orL)
kg
kg
kg
Bq
 " Per functional unit (i.e., one LCD monitor over its effective life).
 b3.6MJ = lkWh
                                          ES-14

-------
                                                                   EXECUTIVE SUMMARY
       The total inventory results for life-cycle inputs reyeaLthat more primary materials,1 water,
fuels, electricity, arid total energy (i.e., fuel energy plus electricity) are used throughout the CRT
life-cycle, while more ancillary materials are used throughout the LCD life-cycle. For the life-
cycle outputs, the CRT releases more air emissions; water pollutants; hazardous, solid, and
radioactive waste; and radioactivity than the LCD. The LCD releases more total wastewater than
the CRT.  Complete inventory tables for each input and output type by life-cycle stage for the
CRT and LCD are provided in Appendix J.
       For the CRT (Table ES-5), of the inputs measured in mass, the water inputs in the
manufacturing life-Cycle stage constitute the majority of the inputs for the entire life cycle.
Water inputs from the LPG production process constitute almost 80% of the water inputs for all
life-cycle stages.  In this inventory, the LPG is used in large quantities as a fuel in CRT glass
manufacturing. When considering which life-cycle stage contributes most to an inventory
category, the manufacturing stage has the largest inventory by mass for primary materials,
ancillary materials,; water inputs,  and fuel inputs. This is also due to the production of LPG as
needed for CRT glass production. Fuel inputs are dominated by the manufacturing stage and
electricity inputs are dominated by the use stage. The total energy (which is calculated by
converting the mass of the fuel into units of energy and combining the fuel energy with the
electrical energy) is  dominated by the manufacturing life-cycle stage, again mostly due to the
large LPG fuel energy used in CRT glass production.
       CRT outputs measured in mass include air emissions, wastewater, water pollutants, and
hazardous, solid, and radioactive waste. Wastewater, by mass (or volume), constitutes the
greatest output; however, total wastewater volume is not used to calculate water-related impacts.
Instead, individual water pollutants are used to calculate water-related impacts. Of the remaining
outputs measured in mass, which are used to calculate impacts (i.e., air emissions, and hazardous,
solid and radioactive waste), air emissions are the greatest contributor to outputs in mass. Note
that radioactivity is! measured in Bequerels (Bq) and cannot be compared on the same scale.
       Considering each CRT inventory type and their contributions by life-cycle stage, the mass
of wastewater and water pollutants are greatest in the manufacturing life-cycle stage (again due to
LPG consumption)!  The outputs of air emissions, hazardous waste, solid waste, radioactive
waste, and radioactivity all have the greatest contribution from the use stage.
      For the CRT outputs, all the totals represented in Table ES-5 include outputs to all
dispositions.  For example, water outputs sent offsite to treatment as well as those directly
discharged to surface waters are all included. Similarly, hazardous, solid and radioactive waste
outputs may be landfilled, treated, or recycled. The inventory shows these as totals; however,
when impacts are calculated, the dispositions dictate which inventory items will be used to
calculate impacts (Chapter 3).
      For the LCD (Table ES-6), of the inputs measured in mass, the water inputs constitute the
majority of the inputs for the entire life cycle, and most of the water inputs are in the
manufacturing life-cycle stage. When considering which life-cycle stage contributes  most to an
inventory category,; the manufacturing stage has the largest inventory by mass for ancillary
materials, fuels, and water inputs. Primary material inputs are dominated by the upstream stages,
         Note that the total mass of primary materials includes the inputs to each process, which may duplicate
materials used in processes subsequent to other processes. For example, the primary materials used in steel
production are added to the steel used as a primary material for monitor assembly.
                                          ES-15

-------
EXECUTIVE SUMMARY
while electricity inputs are dominated by the use stage. The total energy is dominated by the
manufacturing life-cycle stage. Note that LPG production from glass manufacturing does not
dominate much of the LCD inventory as it did for the CRT, because of the smaller amount of
glass used in the LCD compared to the CRT.
       Of the LCD outputs measured in mass (air emissions, wastewater, water pollutants and
hazardous, solid, and radioactive waste), wastewater constitutes the greatest output; however,
total wastewater volume alone is not used to calculate impacts. Of the remaining outputs
measured in mass, which are used to calculate impacts (i.e., air emissions, water pollutants, and
hazardous, solid and radioactive waste), air emissions are the greatest contributor to the outputs.
Note again, as mentioned for the CRT, that radioactivity is measured in Bequerels (Bq) and
cannot be compared on the same scale.
       Considering each LCD output type and then- contributions by life-cycle stage, the mass of
water pollutants is greatest in the manufacturing life-cycle stage, due to the fuel production
processes that support fuel consumption in the manufacturing processes being included in the
manufacturing life-cycle stage.  Wastewater and hazardous waste outputs are greatest in the
manufacturing stage;  air emissions, solid waste, radioactive waste, and radioactivity have the
greatest contribution from the use stage. As with the CRT, all the output totals represented in
Table ES-6 include outputs to all dispositions.

HI. 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
well defined and continue to evolve (Barnthouse et al, 1997; Fava et al, 1993). For toxicity
impacts in particular,  there are some methods being applied in practice (e.g., toxicity potentials,
critical volume, and direct valuation) (Guinee et al, 1996; ILSI, 1996; Curran, 1996), while
others are in development. However, 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 project partner's interest 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 CDP LCIA are  as
                                         ES-16

-------
                                                                 EXECUTIVE SUMMARY
                 i
follows: renewable resource use, nonrenewable materials use/depletion, energy use, solid waste
landfill use, hazardous waste landfill use, radioactive waste landfill use, 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]), radioactivity, chronic human health effects
(occupational and public), aesthetic impacts (odor), aquatic ecotoxicity, and terrestrial
ecotoxicity.
       Classification of an inventory item into impact categories depends on whether the
inventory item is ap 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 CDP:

•      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 (COj)].
              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).
                i \               •
       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 et al. (1997).
The scoring method provides a hazard value  (HV) for each potentially toxic material, which is
then multiplied by £he inventory amount to calculate the toxicity impact score.
       Using the various approaches, the UT LCIA method calculates impact scores for each
inventory item for each applicable impact category. Impact scores are therefore based on either a
direct measure of the inventory amount or some modification (e.g.,  equivalency or scoring) of
that amount based ion 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.
                                          ES-17

-------
 EXECUTIVE SUMMARY
 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 is also limited by the availability and
 quality of the inventory data. Data collection can be very 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 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. 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. However,
 it should be noted that in LCA, chemical toxicity is often not considered at all.  This
 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 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

       Table ES-8 presents the baseline CRT and LCD LCIA indicator results for each impact
category. Appendix M presents complete LCIA results by material, process, and life-cycle stage.
The indicator results presented in Table ES-8 are the result of the characterization step of LCIA
methodology  where LCI results are converted to common units and aggregated within an impact
                                         ES-18

-------
                                                                EXECUTIVE SUMMARY
category. Note that the impact category indicator results are in a number of different units and
therefore can not be summed or compared across impact categories.
       As shown in the table, under the baseline conditions the CRT indicators are greater than
the LCD indicators in the following categories: renewable resource use, rionrenewable resource
use, energy use, solid waste landfill use, hazardous waste landfill use, radioactive waste landfill
use, global warming, ozone depletion, photochemical smog, acidification, air particulates,
biological oxygen demand (BOD), total suspended solids (TSS), radioactivity, chronic public
health effects, chronic occupational health effects, aesthetics, and terrestrial toxicity. The LCD
indicators are greater than the CRT indicators in the following categories: water eutrophication
and aquatic toxicity.  In addition, as noted in Table ES-8, if phased-out substances are removed
from the CRT and LCD inventories, the LCD ozone depletion indicator would exceed that of the
CRT. Details of each impact category and major contributors to the impacts in those categories
are presented in Chapter 3.

Summary of Top Contributors by Impact Category
                 i
       Tables ES-9 and ES-10 summarize the top contributors to CRT and LCD life-cycle
impacts by impact category. As shown in Table ES-9, CRT impacts are largely driven by two
factors: (1) the large amount of LPG fuel used in CRT glass/frit manufacturing, and (2) the
relatively large amount of electricity consumed during the use stage.  The LPG production
process yields the CRT's top contributor in eight of 20 impact categories. Most of this LPG is
used as a fuel source in CRT glass manufacturing in the glass/frit process group, which, in turn,
produces the top contributor to two of 20  impact categories. Thus, LPG used in the glass/frit
process group (primarily CRT glass manufacturing) is ultimately the key driver for CRT impacts
in ten categories. Similarly, outputs from electricity generation during the use stage result in the

               Table ES-8.
Impact category
Renewable resource use
Nonrenewable resource use
Energy use :
Solid waste landfill use
Hazardous waste landfill use
Radioactive waste landfill use
Global wanning i
Ozone depletion
Photochemical smog ;
Acidification :
Air particulates i
Water eutrophication
Water quality, BOD
Water quality, TSS '.
Radioactivity •
Units per monitor
kg
kg
MJ
m3
m3
m3
kg-CO2 equivalents
kg-CFC-11 equivalents
kg-ethene equivalents
kg-SO2 equivalents
kg
kg-phosphate equivalents
kg
kg
Bq
CRT
1.31E+04
6.68E+02
2.0SE+04
1.67E-01
1.6SE-02
1.81E-04
6.95E+02
2.05E-05b-c
1.71E-01
5.25E+00
3.01E-01
4.82E-02
1.95E-01
8.74E-01
3.85E+07d
LCD
2.80E+03
3.64E+02
2.84E+03
5.43E-02
3.61E-03
9.22E-05
5.93E+02
1.37E-05b
1.41E-01
2.96E+00
1.15E-01
4.96E-02
2.83E-02
6.15E-02
1.22E+07d
                                        ES-19

-------
EXECUTIVE SUMMARY
                 Table ES-8. Baseline life-cycle impact category indicators3
Impact category
Chronic health effects, occupational
Chronic health effects, public
Aesthetics (odor)
Aquatic toxicity
Terrestrial toxicity
Units per monitor
tox-kg
tox-kg
m3
tox-kg
tox-kg
CRT
9.34E+02
1.98E+03
7.58E+06
2.25E-01
1.97E+03
LCD
6.96E+02
9.02E+02
5.04E+06
5.19E+00
8.94E+02
a Bold indicates the larger value within an impact category when comparing the CRT and LCD.
 Several of the substances included in this category were phased out of production by January 1,1996. Excluding
phased out substances decreases the CRT ozone depletion indicator to 1.09E-05 kg CFC-11 equivalents per monitor
and the LCD ozone depletion indicator to 1.18E-05 kg CFC-11 equivalents per monitor. These ozone depletion
indicators are probably more representative of the CDP temporal boundaries and current operating practices. See
section 3.3.6 for details.
c Although the CRT indicator appears larger than the LCD indicator, uncertainties in the inventory make it difficult
to determine which monitor has the greater value. Therefore, this value is not shown in bold.
d Radioactivity impacts are being driven by radioactive releases from nuclear fuel reprocessing in France, which are
included in the electricity data in some of the upstream, materials processing data sets. See section 3.3.12 for
details.
top contributor to seven CRT impact categories.  Note that in 14 of the 20 impact categories, the
top contributor to CRT impacts is responsible for more than 50% of impacts.
       LCD impacts are not as dominated by a few data points, but a few processes (LCD
monitor/module manufacturing and electricity generation in the use stage) are responsible for a
large percent of the impacts. As shown in Table ES-10, both of these processes result in the top
contributors to six LCD impact categories each. In addition, the process to produce LNG used as
an ancillary material in LCD monitor/module manufacturing is the top contributor to an
additional impact category (photochemical smog).  Note that in 11 of the 20 impact categories,
the top contributor to LCD impacts is responsible for more than 50% of impacts.
       As a number of the impact results for both monitor types, and for the CRT in particular,
are being driven by a few data points with relatively high uncertainty, sensitivity analyses of the
baseline results were also conducted.
       Table ES-9.  Summary of top contributors to CRT impacts by impact category
Impact category
Renewable resource
use
Nonrenewable
resource use
Energy use
Solid waste landfill
use
Top contributors
Life-cycle
stage
Manufacturing
Manufacturing
Manufacturing
Use
Process group
LPG production
LPG production
CRT glass/frit mfg.
U.S. electric grid
Material
water
Petroleum (in ground)
Liquefied petroleum
gas
Coal waste
Contribution
to impact
score
79%
56%
72%
38%
                                            ES-20

-------
                                                    EXECUTIVE SUMMARY
Table ES-9. Summary of top contributors to CRT impacts by impact category
Impact category
Hazardous waste
landfill use
Radioactive waste
landfill use
Global wanning
Ozone depletion
Photochemical smog
Acidification
Air participates
Water eutrophication
Water quality, BOD
Water quality, TSS
Radioactivity •
Chronic health effects,
occupational
Chronic health effects,
public
Aesthetics (odor)
Aquatic toxicity
Terrestrial toxicity
Top contributors
Life-cycle
stage
End-of-life
Use
Use
Use
Manufacturing
Use
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Materials
Processing
Manufacturing
Use
Manufacturing
Manufacturing
Use
Process group
CRT landfffling
U.S. electric grid
U.S. electric grid
U.S. electric grid
LPG production
U.S. electric grid
LPG production
LPG production
LPG production
LPG production
Steel production, cold-
rolled, semi-finished
CRT glass/frit
manufacturing
U.S. electric grid
LPG production
CRT tube
manufacturing
U.S. electric srid
Material
EOL CRT monitor,
landfilled
Low-level radioactive
waste
Carbon dioxide
Bromomethane
Hydrocarbons,
unspeciated
Sulfur dioxide
PM
COD
BOD
Suspended solids
Plutonium-241
(isotope)
Liquefied petroleum
gas
Sulfur dioxide
Hydrogen sulfide
Phosphorus
(yellow or white)
Sulfur dioxide
Contribution
to impact
score
91%
61%
64%
49%
36%
47%
43%
72%
96%
97%
62%
78%
83%
94%
26%
83%
Table ES-10. Summary of top contributors to LCD impacts by impact category
Impact category
Renewable resource ''
use
Nonrenewable resourpe
use :
Energy use
Solid waste landfill use
Hazardous waste
landfill use :
Radioactive waste ;
landfill use :
Top contributors
Life-cycle stage
Manufacturing
Materials
processing
Use
Use
End-of-life
Use
Process group
LCD monitor/module mfg.
Natural gas production
LCD monitor use
U.S. electric grid
LCD landfilling
U.S. electric grid
Material
Water
Natural gas
(in ground)
Electricity
Coal waste
EOL LCD
monitor,
landfilled
Low-level
radioactive waste
Contribution
to impact
score
38%
65%
30%
44%
97%
44%
                               ES-21

-------
EXECUTIVE SUMMARY
      Table ES-10. Summary of top contributors to LCD impacts by impact category
Impact category
Global warming
Ozone depletion
Photochemical smog
Acidification
Air particulates
Water eutrophication
Water quality, BOD
Water quality, TSS
Radioactivity
Chronic health effects,
occupational
Chronic health effects,
public
Aesthetics (odor)
Aquatic toxicity
Terrestrial toxicitv
Top contributors
Life-cycle stage
Manufacturing
Manufacturing
Materials
processing
Use
Materials
processing
Manufacturing
Manufacturing
Manufacturing
Materials
processing
Manufacturing
Use
Manufacturing
Manufacturing
Use
Process group
LCD monitor/module mfg.
LCD panel components
manufacturing
Natural gas production
U.S. electric grid
Steel production, cold-
rolled, semi-finished
LCD monitor/module mfg.
LCD monitor/module mfg.
LPG production
Steel production, cold-
rolled, semi-finished
LCD monitor/module mfg.
U.S. electric grid
LPG production
LCD monitor/module mfg.
U.S. electric end
Material
Sulfur
hexafluoride
HCFC-225cb
Nonmethane
hydrocarbons,
unspeciated
Sulfur dioxide
PM
Nitrogen
BOD
Suspended solids
Plutonium-241
(isotope)
Liquefied natural
gas
Sulfur dioxide
Hydrogen sulfide
Phosphorus
(yellow or white)
Sulfur dioxide
Contribution
to impact
score
29%
34%
45%
31%
45%
67%
61%
66%
96%
58%
68%
94%
98%
68%
Sensitivity Analyses

       Due to assumptions and uncertainties in this LCA, as in any LCA, the following
sensitivity analyses of the baseline results were conducted: use stage manufactured life scenario,
modified glass energy assumptions, modified LCD module manufacturing energy assumptions,
and modified LCD EOL distribution assumptions. The sensitivity analyses were chosen because
they evaluated data with either the greatest uncertainties or with large uncertainty and major
contributors to the inventory results. Table ES-11 shows the different sensitivity analyses or
scenarios that are considered in the impact assessment results.
                                        ES-22

-------
                                                                 EXECUTIVE SUMMARY
                    Table ES-11. List of sensitivity analysis scenarios
Monitor
type
Sensitivity analysis scenario
Baseline analyses (for reference)
CRT
LCD
Effective life scenario with average glass energy inputs (all glass manufacturing energy data used)
Effective life scenario with average glass energy inputs (all glass manufacturing energy data used)
and outliers in the LCD module manufacturing energy data removed
Sensitivity analyses
CRT
LCD
CRT
LCD
LCD
LCD
Manufactured life scenario same as baseline except lifespan is based on manufactured life instead of
effective ;life, which results in some revised functional equivalency calculations
Manufactured life scenario same as baseline except lifespan is based on manufactured life, which
results in some revised functional equivalency calculations
Modified glass energy scenario same as baseline except comparatively high glass manufacturing
energy inputs are removed
Modified glass energy scenario same as baseline except comparatively high glass manufacturing
energy inputs are removed
Modified LCD module energy scenario same as baseline except LCD monitor/ module
manufacturing energy outliers are included in the average
Modified LCD EOL scenario same as baseline except LCD EOL dispositions are modified
       Based on the sensitivity analyses, it appears that CRT life-cycle impacts are highly
sensitive to the glass energy data, and less sensitive to the lifespan assumptions (lifespan
assumptions greatly affect the magnitude of CRT life-cycle impacts, but they do not greatly affect
the distribution of impacts among life-cycle stages). LCD impacts are less sensitive to the glass
energy data and in fact are not greatly affected by any of the sensitivity analysis scenarios, except
the longer lifespan under the manufactured life scenario.
       Sensitivity results are also useful to interested members of the public who may be
evaluating the relative impacts of different monitor types and are interested in whether the CRT
or LCD has greater life-cycle impacts in any given impact category. Table ES-12 presents the
monitor type with greatest impacts by impact category and by scenario. This information helps
us determine whether major assumptions (e.g., the monitor lifespan and LCD EOL distribution
assumptions) or uncertain data (e.g., glass energy data and LCD monitor manufacturing energy)
are driving results, j As shown in the table, the modified glass energy scenario is the only scenario
that significantly Changes from the baseline. Under this scenario, life-cycle impact results in
seven categories reverse direction from the baseline assessment, such that the LCD has greater
impacts than the CRT. Therefore, under this scenario, a total of nine out of 20 categories are
greater for the LCD than the CRT, compared to two out of 20 categories under the baseline
scenario. The only! other scenario that affects these results is the manufactured life scenario,
when impacts in the water eutrophication category are greater for the CRT than the LCD.
                                         ES-23

-------
EXECUTIVE SUMMARY
                 Table ES-12. Summary of CRT and LCD LCIA results
Impact category
Renewable resource use
Nonrenewable resource use
Energy use
Solid waste landfill use
Hazardous waste landfill use
Radioactive waste landfill use
Global wanning
Ozone depletion
Photochemical smog
Acidification
Airparticulates
Water eutropMcation
Water quality, BOD
Water quality, TSS
Radioactivity
Chronic health effects, occupational
Chronic health effects, public
Aesthetics (odor)
Aquatic toxicity
Terrestrial toxicity
Monitor type with greatest impacts by scenario
Baseline
CRT
CRT
CRT
CRT
CRT
CRT
CRT
b
CRT
CRT
CRT

CRT
CRT
CRT
CRT
CRT
CRT

CRT
Manu-
factured
life
CRT
CRT
CRT
CRT
CRT
CRT
CRT
b
CRT
CRT
CRT
CRT
CRT
CRT
CRT
CRT
CRT
CRT

CRT
Modified
glass energy
CRT

CRT
CRT
CRT
CRT

b
;.IW?p¥f;fSl
*3S&y%¥Xi&l&iat
CRT
CRT



CRT

CRT


CRT
Modified
LCD module
energy
CRT
CRT
CRT
CRT
CRT
CRT
CRT
b
CRT
CRT
CRT

CRT
CRT
CRT
CRT
CRT
CRT

CRT
Modified
LCD EOL
distribution2
CRT
CRT
CRT
CRT
CRT
CRT
CRT
b
CRT
CRT
CRT

CRT
CRT
CRT
CRT
CRT
CRT

CRT
* Based on a qualitative evaluation, not quantitative results.
b CRT impacts are greater than LCD impacts in this category when all data are included hi the inventories, including
data for substances that have been phased out. However, LCD impacts are greater than CRT impacts when phased
out substances are removed from the inventories (see Section 3.3.6).
IV. QUALITATIVE RISK SCREENING OF SELECTED CHEMICALS

       The scope of the DfE CDP included a streamlined Cleaner Technologies Substitutes
Assessment (CTSA) component to perform a qualitative risk screening of specific materials or
processes. Traditionally, the DfE Program has conducted CTSAs that perform detailed risk
characterizations of alternative chemical processes. The streamlined CTS A for the CDP takes a
more detailed look than the LCA at the toxic effects of chemicals used in a process, without
conducting a complete risk characterization typical of past CTSAs.
       Within the human and environmental health effects impact categories of the LCIA, the
input and output amounts are used as surrogates for exposure.  The additional CTS A-related
analyses are intended to better understand the potential exposures to those materials, during any
processes that use those materials, in order to try to better understand potential chemical risks.
       Lead, mercury, and liquid crystals were selected by the CDP Core Group for further
analysis. These materials were selected for their known or suspected toxicity to humans and the
                                         ES-24

-------
                                                                 EXECUTIVE SUMMARY
                 i                                                .
environment, or because they are of particular interest tojndustry or the U.S. EPA. The analysis
of each material summarized or evaluated the following key areas:

•     Use of the materials in computer displays;
•     Life-cycle inputs and outputs of the materials from computer displays;
•     Life-cycle impacts associated with the material inputs and outputs;
•     Potential exposures to the material including occupational, public, and ecological  .
      exposures; I
•     Potential human health effects;
•     U.S. environmental regulations for the material; and
•     Alternatives to reduce the use of the material in computer displays.            .  .-   .   .
                 i
The following are the conclusions drawn from the analyses of lead, mercury, and liquid crystal
use in the life cycle of both CRTs and LCDs.

Lead
       Lead is found in glass components of CRTs, as well as in electronics components (printed
wiring boards and their components) of both CRTs and LCDs. It is also a top priority toxic
material at the U.S. EPA and the subject of electronics industry efforts to reduce or eliminate its
use.  The following conclusions were drawn from a focused look at lead's role in the life cycle of
the computer display, and its effects on human health and the environment:
                 !
•      Due to the much greater quantity of lead in the CRT,than the LCD, lead-based life-cycle
       impacts from the CRT ranged from moderately to significantly greater than those from
       the LCD in bvery category, with the  exception of solid waste landfill use. The most
       significant difference was in non-renewable resource consumption^ where the CRT
       consumed over 40 thousand times the mass of non-renewable resources attributable to
       lead over the course of its life cycle than those consumed by the LCD. Other categories
       where CRTs had notably greater differences in impacts occurred in hazardous waste
       landfill use,j chronic public health effects, and terrestrial toxicity.
•      Contributions of lead-based impacts are small relative to the total life-cycle impacts from
       other materials in the CRT (e.g., glass, copper wire), with the greatest impacts from lead-
       based CRT butputs  occurring in the  categories of non-renewable resources, aquatic
       toxicity, and chronic public health effects (ranging from 0.1 to 0.2% of the overall impact
       scores in each category).
•      For workers, inhalation is the most likely route of exposure to lead which may result  in
       health concerns.  General population exposure to lead is most likely to come from
       incidental ingestion of lead in the soil, or ingestion of lead brought into the household on
       workers clothing or on shoes. Studies have discovered potentially high concentrations of
       lead in households within close proximity to certain facilities that use lead.
•      Significant worker exposures to lead have been documented by existing studies of several
       processes which contribute to the life-cycle of the computer displays (e.g., lead smelting).
       These exposures have been as high as 90 times the OSHA recommended safety levels for
       exposure to; workers at lead smelters. The resulting occupational chronic health effects to
                                         ES-25

-------
EXECUTIVE SUMMARY
       workers from lead exposure likely have been underestimated by the CDP LCIA
       methodology, which uses material inputs, and not outputs, as surrogates for exposure.
•      Lead and lead compounds pose serious chronic health hazards to humans who may
       become over-exposed either in the workplace, or through the ambient environment.  Lead
       exposure is associated with a range of adverse human health effects, including effects on
       the nervous system, reproductive and developmental problems, and cancer. Lead persists
       in the environment, but is relatively immobile in water under most surface and
       groundwater conditions.
•      Alternatives are being developed, such as lead-free solders and glass components, that
       will potentially minimize the future lead content in both CRTs and LCDs.

Mercury

       Mercury is contained within the fluorescent tubes that provide the source of light in the
LCD. Mercury is also emitted from some fuel combustion processes, such as coal-fired
electricity generation processes, which contribute to the life-cycle impacts of both CRTs and
LCDs. EPA's concern with mercury and the potential for exposure during manufacturing and
end-of-life processes warranted a more detailed analysis of mercury in the CDP. The following
conclusions were drawn from a focused look at mercury's role in the life cycle of the computer
display, and its effects on human health and the environment:

•      The mercury emitted from the generation of power consumed by the CRT (7.75 mg)
       exceeds the entire amount of mercury emissions from the LCD, including both the
       mercury used in LCD backlights (3,99 mg) and the mercury emissions from electricity
       generation (3.22 mg). Although this  was not expected because mercury is used
       intentionally in an LCD, but not in a CRT, the results are not surprising since mercury
       emissions from coal-fired power plants are known to be one of the largest anthropogenic
       sources of mercury in the United States. Because the CRT consumes significantly more
       electricity in the use stage than the LCD, its use stage emissions of mercury are
       proportionately higher than those of the LCD.
•      Contributions from mercury-based impacts are not significant relative to the total life-
       cycle impacts from other materials (e.g., glass, copper wire) in the CRT or LCD, with the
       greatest impacts from mercury-based outputs occurring in the aquatic toxicity category
       (0.4% for CRTs, 0.01% for LCDs).
•      Possible pathways of worker exposure during backlight fabrication include inhalation of
       mercury vapors, and dermal exposure or ingestion of mercury on skin. The most likely
       pathway for general population exposure is inhalation of mercury released into the air.
•      Exposure data relevant to the manufacturing of mercury backlights were not available,
       therefore specific conclusions about the potential magnitude of worker exposures could
       not be made. Occupational chronic health effects to workers from mercury exposures
       calculated during the impact assessment (3.99e-06 tox-kg for LCD, none for CRT) likely
       have been underestimated by the CDP LCIA methodology, which uses material inputs as
       surrogates for exposure.
•      Mercury and mercury compounds pose serious chronic health hazards to humans who are
       exposed. EPA has determined that mercury chloride and methyl mercury are possible
                                        ES-26

-------
                                                                 EXECUTIVE SUMMARY
       human carqinogens. Mercury poses serious chronic health hazards to humans, affecting
       the nervous, system, brain, and kidneys.
•      Alternative! backlights have been developed that not only eliminate mercury from the
       light, but also improve, on many of the optical characteristics of the displays. Current
       development is focused on improving the energy efficiency of the alternative lights.

Liquid Crystals

       Liquid crysjtals (LCs) are organic compounds responsible for generating the image in an
LCD. LCs are not present in CRTs.  The toxicity of the LCs in LCDs has been alluded to in the
literature, yet there; is very little known about the toxicity of these materials. By including LCs in
a more detailed analysis, this section attempted to better characterize any potential hazard and/or
potential exposure >of LCs from the manufacturing, use, and disposal of LCD monitors. The
following conclusions were drawn from a focused look at LC's role in the life cycle of the
computer display, and its effects on human health and the environment.
                 I
•      LCs are combined into mixtures of as many as 20 or more compounds selected from
       hundreds pf potential liquid crystal compounds. Because of the possible variations in
       mixtures and the sheer number of compounds available, a select number of liquid crystals
       were used to assess potential human health hazards.
•      LCs do not! appear to contribute significantly to any of the impact categories defined for
       this study; The total score for LCD occupational impacts based on potential worker
       exposure tq LCs of 4.18 tox-grams, calculated using default toxicity values, represents
       less than 0.01% of the total overall chronic occupational health effects impact score  of
       898 tox-kg |for the functional unit of one LCD.
•      Impacts were not calculated for LC releases in the CDP LCIA because data regarding LC
       outputs were not available to the project. LCs are not used to fabricate CRTs and so
       have no environmental impacts in the CRT life cycle.
•      Occupational exposures to LCs during the fabrication of the LCD panels are not expected
       to be significant. The enclosed nature of the chamber in which the LCDs are assembled,
       combined with the equipment (e.g., gloves, aprons) worn by workers in a clean room
       environment, are both expected to act to minimize exposures. Other occupational
       exposures may exist that have not been identified.
•      Toxicologibal testing by a manufacturer of LC substances and mixtures showed that
       95.6% (562 of 588) of the liquid crystals tested displayed no acute toxic potential to
       humans. Twenty-five of the remaining twenty-six chemicals had the potential to exhibit
       harmful effects to humans, while the remaining crystal was classified as toxic (EU
       classification) and thus was discontinued. An EPA review of toxicity data for the
       confidential LC compounds was unable to identify any relevant toxicity information.
       Insufficient toxicity data exist to assess the toxicity of specific LC compounds.
•      Testing for mutagenic and carcinogenic effects by the supplier showed that 99.9% (614
       out of 615) of the liquid crystal compounds tested displayed no mutagenic effects. The
       remaining chemical that showed mutagenic potential was excluded from further
       development. Additionally, mutagenicity testing of ten LC substances using mammalian
       cells showed no suspicion of mutagenic potential.
                                         ES-27

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

       The purpose of the CDP, as stated in Chapter 1, is to provide a scientific baseline of life-
cycle environmental impacts of CRTs and LCDs, and to develop a life-cycle model for future
analyses. The primary targeted audience is the electronics industry, for whom results may
provide insight into improvement opportunities in the life cycle of CRTs and/or LCDs. In
addition, the general public may also find results useful when considering environmental impacts
of each display type. This report, however, does not include direct comparative assertions or
improvement assessments based on the results. Alternatively, results and conclusions are
described in terms of the overall LCI versus the LCIA, and details of the impact assessment,
including the additional assessments of lead, mercury and liquid crystals, and the sensitivity
analyses. Major uncertainties, cost and performance considerations, suggestions for
improvement opportunities, and suggestions for further research are. also provided.

LCI vs. LCIA

       Inventory data provide information on how much material is being consumed in the life
cycle (i.e., inputs) and how much material is generated/released (i.e., outputs).  The LCI alone,
however, does not always translate directly into impact categories that may be of interest.
Impacts are sometimes driven by materials other than the top inventory contributors. For
example, the top air emission for LCDs is carbon dioxide, however the greatest global wanning
impact score is from SF6 in the LCD monitor/module manufacturing process.
       Some impact categories associated with ancillary materials and water pollutant inventory
types had different outcomes in the LCI versus the LCIA. For example, the three impact
categories affected by the ancillary materials inventory had greater impacts for the CRT (Table
ES-8), although the ancillary material inventory had greater amounts of inputs for the LCD (see
Tables ES-5 and ES-6).  In this case, both primary and ancillary materials contribute to the
impact categories, contributing to the differing results.
       In addition, the LCD had greater inventory amounts of wastewater outputs than the CRT;
however, the impacts related to water releases are in some cases greater for the CRT than the
LCD. In the LCIA, the LCD has greater impacts for water eutrophication and aquatic toxicity,
but not for the two water quality categories (BOD and TSS), chronic health effects to the public,
nor terrestrial toxicity, all of which include water emissions in calculating the impact score.
These results show that the inventory results may not directly  translate into impact results.

CRT and LCD Baseline LCIA Results

       The LCIA results showed that the CRT has greater total life-cycle impact indicators than
the LCD in most of the impact categories (see Table ES-8). In the baseline scenario, the CRT
has greater impacts than the LCD in all but two impact categories (eutrophication and aquatic
toxicity). However, note that for the ozone depletion category, the LCIs for both the CRT and
LCD contain data for substances that were phased out of production by 1996 due to their ozone
depletion potential. Whether these emissions still occur in countries that were signatories to the
Montreal Protocol and its Amendments and Adjustments (such as the United States and Japan) is
not known, but considered to be unlikely. When phased-out substances are included in the
                                        ES-28

-------
                                                                 EXECUTIVE SUMMARY
inventory, the CRT has greater ozone depletion impacts than the LCD.  However, if phased-out
substances are removed from the inventories, the results are switched, with the LCD having
greater impacts.
       When considering which life-cycle stage has greater impacts, the LCIA results showed
that the manufacturing life-cycle stage dominates impacts for most impact categories for both the
CRT and LCD.  Table ES-13 lists the number of impact categories with the greatest impacts by
life-cycle stage.   \          .    '
       A more detailed evaluation of lead, mercury and liquid crystals was completed in
Chapter 4. As expected, the CRT, which has lead in the glass, frit, and printed wiring boards
(PWBs), had greater impacts from lead than did the LCD, which only has lead in the PWBs.
Regarding mercury' there were greater inventories of mercury in the CRT life cycle than in the
LCD life cycle, despite the fact that only the LCD has mercury directly in the product. The
greater amount of mercury in the CRT life cycle is from the release of mercury and mercury
compounds from the generation of electricity. The CRT consumes significantly more electricity
in the use stage than the LCD. Liquid crystals are only found in LCDs, and therefore, there are
no associated impacts for the CRT. Little conclusive information was available on the liquid
crystal materials. A detailed literature search was conducted, however very little data were
available on the toxicity of these materials. Based on the limited toxicity data obtained, liquid
crystals currently do not appear to be a significant human health or environmental hazard in the
LCD life cycle.  However, there were insufficient toxicity data available to make a definitive
conclusion about LC toxicity.
                 !
  Table ES-13. Number of impact categories in each life-cycle stage with greatest impacts
                       among life-cycle stages (baseline scenario)
Monitor type \
1
CRT ;
LCD :
# of impact categories with greatest impacts among life-cycle stages
Upstream
3
'-' 3
Manufacturing
9
11
Use
6
4
EOL
2
2
CRT Results
                1 !
       For the CRT, many of the impacts were driven by a single material in the inventory.  As
shown in Table ES-9, in 14 of the 20 impact categories, the top individual contributor to the
impacts was responsible for greater than 50 percent of the impacts. This shows that the CRT
data are highly sensitive to a few data points. Major conclusions from the CRT LCIA are as
follows:          :

•      Energy used in glass manufacturing and associated production of LPG are driving the
       baseline CRT results (they dominate ten impact categories, including overall life-cycle
       energy use).
•      The large amounts of fuel used as energy sources are driving occupational health effects.
       Occupational impacts are calculated from inventory input amounts, and therefore there
       may or may; not actually be exposure to these fuels (e.g., they may be contained);
                                         ES-29

-------
EXECUTIVE SUMMARY
       however, the results illustrate the potential for health effects, especially under spill or
       upset conditions.
•      The generation of electricity for the use stage dominates seven impact categories.
•      Air emissions of sulfur dioxide from electricity generation (for the use life-cycle stage)
       drive chronic public health effects, acidification, and terrestrial toxicity impacts. This
       may be a concern, for example, in areas in nonattainment of regulated levels of sulfur
       dioxide in the United States.

       The use of LPG fuel in glass manufacturing dominated ten impact categories: two
directly from the LPG used in glass/frit manufacturing (energy use impacts and chronic
occupational health effects) and eight from LPG production (renewable resource use,
nonrenewable resource use, photochemical smog, air particulates, water eutrophication, BOD
water quality, TSS water quality, and aesthetics).  In addition, impacts from the generation of
electricity during the use stage dominated seven impact categories: solid waste landfill use,
radioactive waste landfill use, global warming, ozone depletion, acidification, chronic public
health, and terrestrial toxicity. The CRT tube manufacturing process, which represents the most
functionally and physically (by mass) significant component of the CRT monitor, only dominated
one impact category (aquatic toxicity).  Twenty-six percent of the aquatic toxicity score was from
phosphorus outputs from tube manufacturing, while most of the rest were from the materials
processing life-cycle stage. The remaining two impact categories (hazardous waste landfill use
and radioactivity) had greatest impacts from the landfilling of the assumed hazardous proportion
of CRT monitors, and the release of Plutonium-241 in steel production, respectively (Table
ES-9). The radioactivity impacts are driven by the radionuclide Pu-241, due to the electric grid
inventory included in the steel production secondary data set, which includes nuclear fuel
reprocessing.
       The large amount of LPG reported for glass manufacturing was originally questioned
during the data collection and verification stage of this project. While no compelling reason
could justify removing the LPG data in the baseline case, a sensitivity analysis was conducted in
which the glass energy data were modified.  Other sensitivity analyses were also  conducted (i.e.,
manufactured life, modified LCD monitor manufacturing energy, and modified LCD EOL
distributions). However, the only scenario that substantially altered results was the modified
glass energy scenario (see Table ES-12). It is likely that the actual energy inputs to the glass
manufacturing process is somewhere between the baseline and modified glass scenarios. More
information is needed on energy used in glass manufacturing, which is driving CRT baseline
results.
       The additional analyses for the CRT of lead and mercury also revealed that the use of lead
could present health risks, but the CDP method for calculating occupational impacts uses only
process inputs (not outputs) and may not adequately represent occupational exposures and risks.
Further refinement of the occupational impact analysis may be warranted.
       Although there is no mercury in the CRT monitor, mercury emissions from electricity
generation in the CRT life cycle were greater (in mass) than the mercury used in  the LCD.
Therefore, to reduce mercury emissions from the CRT life cycle,, efforts to reduce electricity
consumption could be taken. Additionally, changes to the electric grid could also reduce
mercury emissions from the CRT life-cycle.
                                         ES-30

-------
                                                                 EXECUTIVE SUMMARY
LCD Results                        _   „           ,  ^

       The LCD impact results were less sensitive to an individual input or output than the CRT
results, although iri 11 of the 20 impact categories an individual input was still responsible for
greater than 50% of the total impacts.  In general, the LCD results are less uncertain than the
CRT results. This |is because most of the CRT results are being driven by either glass input data
or data from secondary sources, while LCD impacts are being driven more by data from primary
sources. Some resiilts to note are as follows:

•      The LCD monitor/module manufacturing process group had greatest impacts in six
       impact categories (Table 3-58).
•      Although the top contributor to the energy impact category was electricity consumed in
       the use stage (30%), the overall energy impacts were greater from the manufacturing
       stage than the use stage.
•      In the glass; energy sensitivity scenario, the use stage had greatest energy impacts,
       although only by a small margin over the manufacturing stage (see Figure 3-26).
•      Sulfur dioxide [emitted from electricity generation in the use stage, and constituting only
       0.37% of the air emission inventory (see Table 2-49)] dominates the acidification, chronic
       public health, and terrestrial toxicity impact categories (Table 3-58). The high public
       health and terrestrial toxicity scores are due to its low non-cancer toxicity value and
       resulting high hazard value (HV).
•      Sulfur hexaifluoride (SF6) from LCD monitor/module manufacturing was the single
       greatest contributor to the global warming impact score; however, carbon dioxide from
       the use stage and the materials processing stage  also contributed  significantly to the
       global warming impacts (Table 3-25).
•      The glass energy inputs did not directly dominate any impact categories, as they did for
       the CRT (due to the smaller mass of glass in the LCD);  however, LPG production
       (required for the glass energy fuel) dominated two categories: TSS water quality and
       aesthetics;(Table 3-58).
•      LNG as an ancillary inventory material was questionably very large and had greatest
       impacts in two categories: nonrenewable resource use and photochemical smog (Table 3-
       58; shown there as "Natural Gas Production" due to that process being used as a surrogate
       for LNG production).

       The additional analyses of lead, mercury, and liquid crystals showed that the LCIA alone
is not adequate enough to determine all the potential impacts within the life-cycle of the LCD
monitors. Further,; the LCIA method in this LCA used only process inputs as surrogates for
occupational exposure.  If the occupational impacts methodology were refined, outputs into the
occupational environment should also be considered.
       For mercury, which is found in the backlights of the LCD monitors, there is nearly the
same amount of mercury by mass emitted to the air during electricity generation as there is
mercury used to make the backlight unit. The mass of mercury input for backlights is only about
20% greater than the mercury air emissions from electricity generation (across all life-cycle
stages):          '
                                         ES-31

-------
EXECUTIVE SUMMARY
       Liquid crystals were also identified by the CDP Core Group as a material for which
additional information would be reviewed. The LCIA did not find the liquid crystals to be
significant contributors to any impact categories; however, this could partially be due to the lack
of information on them. The additional analysis revealed limited information; however,
qualitatively, it did not show significant potential risk.

CRT vs. LCD Sensitivity Analysis Results

       The only sensitivity analysis to show significant difference in the results was the modified
glass energy scenario. In comparing the CRT and LCD, the CRT baseline scenario had greater
impacts than the LCD in all but two impact categories (eutrophication and aquatic toxicity) and
possibly three (ozone depletion). In the modified glass energy scenario, nine of the 20 categories
had greater impacts from the LCD life-cycle than the CRT. Energy use remained greater for the
CRT; however, nonrenewable resource use, global warming, photochemical smog,
eutrophication, BOD and TSS water quality, chronic occupational health effects, and aesthetics
all reversed such that the LCD had greater impacts than the CRT (Table ES-12). As stated
above, it is believed that a more true representation of the monitor life cycles lies somewhere
between the baseline and modified glass energy scenario. Further work is recommended in
clarifying and refining glass energy input information.

Uncertainties

       As with any LCA, it is not uncommon for there to be uncertainty associated with such a
large data collection effort. The limitations and uncertainties associated with this LCA and
LCAs in general have been discussed elsewhere in the executive summary. Two of the largest
sources of uncertainty in this LCA that have a significant effect on the results are as follows:

•      CRT and LCD glass manufacturing energy inputs (from primary data):  The larger
       amount of glass used in CRTs than LCDs results in the CRT having greater associated
       uncertainty than the LCD results.
«      Secondary data for upstream and fuel production processes: When any one material is
       used in the life-cycle of either monitor in large quantities, the impacts associated with the
       inputs and outputs from the production of that material may become significant. For
       example, LPG and LNG production were both used in significant enough amounts to
       influence some impact categories. Therefore, the uncertainty in the secondary data
       becomes important. This highlights the need for a consistent, national (or international)
       LCI database that is updated regularly.

       Other uncertainties associated with individual data points had less effect on the overall
results than the uncertainties mentioned above. For manufacturers interested in conducting
improvement assessments, closer review of such uncertainties may be warranted.
                                        ES-32

-------
                                                                 EXECUTIVE SUMMARY
                 I
       Another point that should be recognized in the 9verall LCA of CRTs and LCDs is that
 CRTs are a more mature technology than LCDs. Changes in LCD manufacturing processes have
 likely occurred during the development and publication of this report. Therefore, conclusions
 must be carefully drawn when evaluating the mature CRT compared to the newer LCD
 technology.

 Cost and Performance Considerations

       The focus of this study has been on the environmental effects associated with CRTs and
 LCDs. The environmental attributes or burdens of a product are not expected to be considered
 alone when evaluating the marketability and commercial success of a product.  The cost and
 performance of each monitor type are obviously critical components to a company's or
 consumer's decisions of whether to produce or purchase a product.  The report briefly addresses
 direct retail costs o|f the monitors and electricity costs associated with the monitors; however, a
 complete cost analysis, including all direct costs (e.g., material costs) and indirect costs (e.g.,'
 environmental costs to society) is beyond the scope of this report.
       The average retail price of 1997-2000 model year monitors, collected from the
 manufacturers who supplied data for this project, is $541 for the CRT and $1,450 for the LCD.
 From these data, the LCD is approximately 2.7 times more costly.  More recent data show that
 prices have come down, and the difference in prices between the CRT and LCD has also been
 reduced.
       The costs from the use stage can be represented by the use stage electricity costs. Based
 on the average cost of residential and commercial electricity in the United States, and the amount
 of energy consumed per functional unit in the use stage (baseline scenario), the electricity costs
 to consumers over me life of the monitors during the use stage are $48 for the CRT  and $18 for
 the LCD.  In addition, the upstream and manufacturing costs of electricity in the baseline
 scenario for the CRT are approximately $1.3/functional unit and $1.5/functional unit,
 respectively; for the LCD they are $0.10/functional unit and $3.4/functional unit, respectively.
       The LCA is defined such that the monitor assessments are performed on a functionally
 equivalent basis.  To the extent possible, data were collected on functionally equivalent monitors.
 When companies were approached to participate in the study, they were informed of the
 performance specification parameters within which the study boundaries were defined.
Therefore, it is assumed that they meet the specifications as presented in Table ES-1, and that
 they perform relatively equivalently. From the primary data, the reported brightness of the CRT
 was less than the LCD, otherwise, they are functionally similar.

 Improvement Assessment Opportunities and Targeted Audience Uses  of Report
                 i
       To meet the primary objective of providing the display industry with data to  perform
improvement assessments, the industry should look at the manufacturing life-cycle stage, while
recognizing the influences of the other stages. CRT improvement opportunities could include
improved energy efficiency during glass manufacturing and display use, as well as reductions in
lead content. LCD; improvement opportunities could also include improved energy efficiency,
especially during manufacturing.  Certain materials, such as SF6 and its contribution to global
warming, may also; be of concern and an area to focus on in future improvement assessments.
                                        ES-33

-------
EXECUTIVE SUMMARY
       In addition, any improvement assessment should consider how changes in one life-cycle
stage will affect impacts in other stages. For example, in Chapter 4 we saw that the mercury
inputs and outputs from the intentional use of mercury in an LCD backlight are less (by mass)
than the mercury emissions from the CRT use stage, due to the relative energy usage by the CRT
and the emissions of mercury from electricity generation.  In this example, we can see that on a
pure mass basis, a product's energy efficiency is a key consideration and any changes in
manufacturing should consider if it will affect changes in use.
       Another objective of this study was to provide an LCA model for future analyses.
Companies or individuals who have more current data for the CRT or LCD can apply them to the
model presented here. For example, changes in an individual process can be identified and
incorporated into the model. The other processes that are not expected to change significantly
can be left unchanged, and only limited data would need to be altered. This would reduce the
time and resources that would normally be required for a complete analysis.
       Finally, those interested in comparing the results of the two monitors can apply their own
set of importance weights to each impact category to determine their individual decision. For
example, if energy impacts are much more important than aesthetics to a particular person, they
can weigh energy more heavily in  concluding which monitor may have fewer environmental
impacts, while keeping in mind the data limitations and uncertainties, as well as cost and
performance considerations.

Suggestions for Future Research

       Areas where future research could be conducted to refine and/or continue the use of the
results in this study are as follows:

•      gather more information on energy use in glass manufacturing;
•      develop consistent materials and fuel processing data in a national (or international) LCI
       database that is updated regularly;
•      refine and/or update some  of the LCD manufacturing data (e.g., LNG data);
•      collect more complete EOL data (e.g., remanufacturing data, and primary data for
       incineration and landfilling) to determine better representation of the EOL impacts;
•      conduct more research on the EOL options for LCDs;
•      collect more detailed data on landfilling and other treatment processes, such as water
       treatment where no impacts were calculated;
•      update manufacturing data to meet more recent monitor model years;
•      conduct a more focused analysis on selected areas for detailed improvement assessments;
       and
•      evaluate process changes or other alternatives against an "average 1997-2000 model year"
       to evaluate impacts of changes or improvements over time.
                                         ES-34

-------
 REFERENCES
                                   REFERENCES
                 i
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. Sopiety of Environmental Toxicology and Chemistry, Pensacola, FL.
                 !
CIA (Computer Industry Almanac, Inc.). 1997. Computer Industry Almanac. Arlington Heights,
       U-*.

Curran, M.A. 1996. Environmental Life Cycle Assessment. McGraw-Hill: New York

EIA (Energy Information Administration). 1999. 1997 Residential Energy Consumption Survey,
       Housing Characteristics report. Web site available at:
       http://wwwJeia.doe.gov/emeu/consUmption/reports.html (under "Residential Households').

Ecobilan. 1999. DRAM, life-cycle inventory database developed by the Ecobilan Group,
       copyright 1999.
                 I                                  •
EPA (Environmentlal Protection Agency). 1998.  "Characterization of Municipal Solid Waste in
       the United States: 1997 update." Report No. EPA530-R-98-007.
                 i
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 land Chemistry (SETAC) and SETAC Foundation for Environmental
       Education.  Pensacola, FL.
                 i

Guinee, J., R. Heijijngs, L. van Oers, D. van de Meent, T. Vermeire, M. Rikken. 1996. LCA
       Impact Assessment of Toxic  Releases. The Hague, The Netherlands.
                 i

ESI (International Iron and Steel Institute). 2001. World Steel Trends and Statistics. Website
       accessed at http://www.worldsteel.org/trends_indicators/figures_3.html, on July 31, 2001.

Kincaid, L. et al.  1996. Cleaner Technologies Substitutes Assessment - A Methodology &
       Resource Guide. U.S. Environmental Protection Agency, Office of Pollution Prevention
       and Toxics.! EPA 744-R-95-002. Washington, DC.

NSC (National Safety Council). 1999.  Electronic Product Recovery and Recycling Baseline
       Report: Recycling of Selected Electronic Products in the United States, Washington,
       D.C.  May.
                                        ES-35

-------
REFERENCES
Swanson, M.B., G.A. Davis, L.E. Kincaid, T.W. Schultz, J.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.

TORNRC (The Oak Ridge National Recycle Center). 2001. Personal communication with D.
       McFarland and A. Bradley, TORNRC, and R. Dhingra, University of Tennessee, Center
       for Clean Products and Clean Technologies.  April.

Vorhees, G. 2000.  Personal communication with G. Vorhees, Envirocycle, and M. L. Socolof,
       University of Tennessee, Center for Clean Products and Clean Technologies. July.
                                        ES-36

-------
                                                                      1.1 BACKGROUND
                                       Chapter 1

                           GOAL DEFINITION AND SCOPE
1.1    BACKGROUND
       This report; presents the results of a voluntary, cooperative project among the Design for
the Environment (DfE) Program in 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 model
and assess the life-;cycle environmental impacts of flat panel display (FPD) and cathode ray tube
(CRT) technologies that can be used for desktop computers.  The DfE Computer Display Project
(CDP) analysis provides a baseline report and the opportunity to use the model as a stepping
stone for further analyses and improvement assessments for these technologies.
       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. DfE
industry projects ate cooperative, joint efforts with trade associations, businesses, public-interest
groups, and acaderiiia to assist businesses in specific industries to identify and evaluate more
environmentally sound products, processes, and technologies. The DfE CDP partnership consists
of members of electronic industry trade associations, computer monitor and component
manufacturers, suppliers to the electronics industry, academic institutions, EPA, and a public
interest group. The direction and focus of this project'was chosen by the project partners.
       The DfE CDP uses life-cycle assessment (LCA) as an environmental evaluation tool,
which is increasingly being used by industry. LCA can be used to evaluate the environmental
effects of a product, process, or activity.  An LCA looks at the full life cycle of the product from
materials acquisition to manufacturing, use, and final disposition. It is a comprehensive method
for evaluating the full environmental consequences of a product system. There are four major
components of an LCA study:  goal definition and scoping, life-cycle inventory, impact
assessment, and improvement  assessment.  LCAs are generally global and non-site specific.
       Under the DfE Program, the Cleaner Technologies Substitutes Assessment (CTSA)
methodology (Kineaid et al, 1996) was developed to generate information needed by businesses
to make environmentally informed choices and to design for the environment. The CTS A
process involves comparative evaluations of substitute technologies, processes, products, or
materials.  Impact areas that are evaluated include human and ecological risk, energy and natural
resource use, performance, and cost.
       Both evaluation tools have similar objectives; however, their applications  generally
differ.  A CTSA islmore site specific and evaluates actual (predicted) impacts. For example,
techniques such as!health risk assessment are incorporated into a CTSA. An LCA is more global
and generic in natuje and generally would not incorporate site-specific parameters when
evaluating impacts.  The LCA may also use surrogate measures to represent impacts instead of
predicting or measuring actual impacts.
       This project focuses on the LCA, while including some CTSA-related analyses. It
performs the broad analysis of the LCA, which also incorporates many of the CTSA components
(e.g., risk, energy impacts, natural resource use) into the impact assessment. The analysis also
                                          1-1

-------
1.1 BACKGROUND
assesses more specific impacts for selected materials and acknowledges product cost and
performance, typical of a CTSA. Because both.methodologies require intensive data gathering
efforts and can be extensive undertakings, the scope must be carefully and clearly defined. As
only selected materials are evaluated for the CTSA, this project could be considered an LCA with
a streamlined CTSA component.
      Life-cycle assessment (LCA) is a comprehensive method for evaluating the full
environmental consequences of a product system. Another related assessment strategy is life-
cycle design, which is a systems-oriented approach for designing more ecologically and
economically sustainable product systems.  It integrates environmental requirements into the
earliest stages of design so total impacts caused by product systems can be reduced.
Environmental, performance, cost, cultural, and legal requirements are balanced (Curran, 1996).
Environmental impacts and health risks caused by product development are intended to be
reduced.  This is very similar to design for environment (DfE) programs where environmental,
issues are incorporated into a product system design process. DfE and life-cycle design have
similar objectives, although their origins differ. DfE was developed as an off-shoot of design for
X, where X could be any number of criteria (e.g., manufacturability, testability, reliability,
recyclability). In DfE, X is environmental protection and sustainability.
      The EPA DfE Program, where X is also environmental protection and sustainability,
promotes risk reduction, pollution prevention, energy efficiency, and other resource conservation
measures through process choices at a facility level. EPA's DfE CTSA process also includes an
analysis of performance and cost. Typically, EPA's DfE Program focuses less on the entire life
cycle and more on evaluating technology or material substitutes to reduce environmental impacts.
This project combines the DfE Program's CTSA process (Kincaid et al., 1996) and the LCA
process, and thus resembles a life-cycle design approach.
      LCAs evaluate the environmental impacts from each of the following major life-cycle
stages:

•     Raw materials extraction/acquisition;
•     Materials processing;
•     Product manufacture;
•     Product use, maintenance, and repair; and
•     Final disposition/end-of-life.

Figure 1-1 briefly describes each of these stages for a computer display product system.  The
inputs (e.g., resources and energy) and outputs (e.g., product and waste) within each life-cycle
stage, as well as the interaction between each stage (e.g., transportation) are evaluated to
determine the environmental impacts.
      As defined by the Society of Environmental Toxicology and Chemistry (SETAC), 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, the international
standard, ISO 14040: Environmental Management—Lifecycle Assessment—Principles and
Framework,  has defined the four major components of ah LCA as: (1) goal and scope; (2)
inventory analysis; (3) impact assessment; and (4) interpretation of results.  The SETAC and
International Standards Organization (ISO) framework are essentially synonymous with respect
to the first three components, but differ somewhat with respect to the fourth component,
                                          1-2

-------
                                                                        1.1 BACKGROUND
improvement assessment or life cycle interpretation. Improvement assessment is the systematic
evaluation of opportunities for reducing the environmental impacts of a product, process, or
activity.  Interpretajtion is the phase of LCA in which the findings from the inventory analyses
and the impact assessment are combined together, consistent with the defined goal and scope in
order to reach conclusions and recommendations. In this study and project report, the goal and
scope are the subject of Chapter 1.  The inventory analysis and impact assessment are the
subjects of Chapters 2 and 3 respectively; and the improvement assessment or life-cycle
interpretation are left to the electronics industry given the results of this study. The life-cycle
inventory and impact assessment strategies are briefly described below.
   INPUTS
                   LIFE-CYCLE STAGES
                                                                             OUTPUTS
 Materials
 Energy
 Resources
                 j   RAW MATERIALS EXTRACTION/ACQUISITION
             \ctiyities related to the acquisition of natural resources, including mining
                non-renewable material, harvesting biomass, and transporting raw
               	i	materials to processing facilities.	
                MATERIALS PROCESSING
 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 and assembling component parts to make a
  _	computer display.
   |    PRODUCT USE, MAINTENANCE, AND REPAIR
 Displays are transported to and used by customers. Maintenance and
repair may be conducted either at the customer's location or taken back
   I	to a service center or manufacturing facility.	
                 !                FINAL DISPOSITION
             At the end of its useful life, the display is retired. If reuse and recycle of
              usable parts is feasible, the product can be transported to an appropriate
              facility and disassembled. Parts and materials mat are not recoverable
               are then transported to appropriate facilities and treated (if required
                 1             or necessary) and/or disposed of.
                                                                                  Wastes
                                                                                 Products
              Boundary

                    Figure 1-1. Life-cycle stages of a computer display

       The life-cycle inventory (LCI) involves the quantification of raw material and fuel inputs,
and solid, liquid, and gaseous emissions and effluents. Traditional LCIs quantify pollutant
categories (e.g., volatile organic compounds [VOCs]) rather than particular chemicals.  This
project also includes a more detailed evaluation of a few specific chemicals found in computer
displays (lead, mercury, and liquid crystals) to enable a more thorough evaluation of risk from
chemical exposure: The approach to the LCI in this study involves defining product components,
developing a bill of materials (BOM), and collecting inventory data on each life-cycle stage for
computer displays. Details of the LCI data gathering activities are provided in Chapter 2.
                                           1-3

-------
1.2 INTRODUCTION
       The life-cycle impact assessment (LCIA) involves the translation of the environmental
burdens identified in the LCI into environmental impacts. LCIA is typically a semi-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. Details of
the LCIA methodology and results are presented in Chapter 3.  This project has furthered the
development of LCIA methodology by including health effect concerns into the LCIA. This
study also qualitatively assesses exposure and chemical risk of selected chemicals in the life
cycles of the computer displays (Chapter 4).

1.2    INTRODUCTION

       Goal and scope definition is the first phase of LCA and is important to the CTSA as well.
This phase is important because it determines why the LCA or CTSA is being conducted and its
intended use, as well as the system and data categories to be studied.
       This chapter presents the goal and scope of the DfE CDP, including its purpose and goals,
previous research and market trends, descriptions of the product systems being evaluated, and the
boundaries used in this study. It incorporates scoping as it is recommended in both the LCA
(e.g.} Curran, 1996; Fava el al, 1991; ISO, 1996) and CTSA processes (Kincaid et al., 1996).

1.2.1  Purpose

       The purpose of this study is two-fold: (1) to establish a scientific baseline that evaluates
the life-cycle environmental impacts of flat panel displays (FPDs)  and cathode ray tubes (CRTs)
for desktop computers, by combining LCA and CTSA methodologies; and (2) to develop a
model that can be used with updated data for future analyses. This study evaluates the newer
active matrix liquid crystal display (LCD) and the more traditional CRT technologies. The
evaluation considers impacts related to material consumption, energy, air resources, water
resources, landfills, human toxicity, and ecological toxicity.  This study is designed to provide
the electronics industry with information needed to improve the environmental attributes of
desktop computer displays.  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 will also provide the industry and consumers with  valuable
information to make environmentally informed decisions regarding display technologies, and
enable them to consider the relative environmental merits of a technology along with its
performance and cost.

1.2.2  Previous Research

       While there has been some work done on the life-cycle environmental impacts of either
CRTs or LCDs, there has not been a quantitative LCA addressing both CRTs and LCDs. For
example, Microelectronics and Computer Technology Corporation (MCC) published an
Electronics Industry Environmental Roadmap (1994) that qualitatively discussed environmental
issues and priority needs for reducing impacts from computer CRTs and FPDs, but this project
did not, nor was it intended to, focus on all aspects of the displays'life cycles.
                                          1-4

-------
                                                                    1.2 INTRODUCTION
       Some of the environmental impacts of CRTs (e.g;., energy use, disposal of lead, and other
end-of-life issues such as recycling) have been identified but not quantified in previous work,
such as that done by EPA's Common Sense Initiative Computer and Electronics Subcommittee.
Atlantic Consulting completed a draft LCA of the personal computer (including a 15" monitor)
for the European Union's (EU) Eco-Label program (Atlantic Consulting and IPU, 1998).
Further, the New Jersey Institute of Technology conducted an LCA of television CRTs, which
have the same technology as the computer monitor CRT.
       Studies on the environmental impacts of FPDs are much less prevalent. A University of
Michigan master's'thesis (Koch, 1996) evaluated the environmental performance of an LCD
manufacturer. The thesis did not quantitatively assess environmental impacts from all life-cycle
stages, as would be done in an LCA.  The EU Eco-Label Program also evaluated a portable
computer with a 13.3" LCD screen (Orango AB and Atlantic Consulting, 1999). The scope was
limited to energy consumption for the impact assessment, and it provided some inventories for
raw materials production. Human and ecological toxicity were excluded from the scope of the
portable computer analysis.

1.2.3   Market Trends

       At present, computer displays using CRTs dominate worldwide markets. They provide a
rich, high-resolution display well suited to a wide range of user requirements.  However, CRT
displays are bulky, use larger amounts of energy to operate than LCDs, and are associated with
disposal concerns clue to the presence of lead in the glass. Color CRT monitors contain lead to
help shield the users from x-rays (x-ray attenuation) and can, under some circumstances, be
classified as a hazardous waste when disposed of.  Newer technologies, collectively referred to as
FPDs, have captured significant market segments. FPDs exhibit desirable qualities such as
reduced size and wjeight and greater portability. Environmentally, they are expected to consume
less energy during use and do not use leaded glass.  However, they may consume more energy
during manufacturing, contain small amounts of mercury, are more costly,  and in the past have
had lower resolutiqn and image quality than the CRT. The LCD, first used predominately in
notebook computers, is now moving into the desktop computer market. The 1998 worldwide
market for desktop'computer monitors was 90 million units.  The 1998 actual and 2001 projected
markets for desktop computer CRTs and LCDs are presented in Table 1-1.  Market projections
anticipate that LCD^ s will capture sizable market share for desktop computer displays.
                 i
       Table 1-1.! Desktop display  markets - actual for 1998 and projected for 2001
1
;
Technology
CRT
Worldwide
North America
LCD
Worldwide
: North America
Number of displays (thousands of units)
1998
88,600
33,801
1,300
229
2001
119,100
42,609
14,3.00
3,787
               Sources: DisplaySearch, 2001.
                                          1-5

-------
1.2 INTRODUCTION
1.2.4  Need for the Project

       Given the expected market growth of LCDs for computer displays, the various
environmental concerns throughout the life cycle of the computer displays, and the fact that the
relative life-cycle environmental impacts of LCDs and CRTs have not been scientifically
established to date, there is a need for a quantitative environmental life-cycle analysis of desktop
computer display technologies. Manufacturers can use these results or the model developed here
to identify areas for improvement concerning the environmental burdens. Further, as companies
or consumers are considering investing in certain displays, they can refer to the results of lids
study to assist them in making environmentally informed decisions.

1.2.5  Targeted Audience and Use of the Study

       The electronics industry is expected to be one of the primary users of the study results.
The study is  intended to provide industry with an analysis that evaluates the life-cycle
environmental impacts of selected computer display technologies.  Scientific verification of the
relative environmental 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 releases of toxic chemicals and reduce risks to
health and the environment. Given the results, the industry can then perform an improvement
assessment of the display technologies. This also allows the electronics industry to make
environmentally informed choices about display technologies when assessing and implementing
improvements such as changes in product, process, and activity design; raw material use;
industrial processing; consumer use; and waste management.
       Another result of the study is an accounting of the relative environmental impacts of
various components of the computer displays, thus identifying opportunities for product
improvements to reduce potential adverse environmental impacts and costs.  Identification of
impacts from the computer display technologies  can also encourage industry to implement
pollution prevention options, such as development and demonstration projects, and technical
assistance and training. Since this study incorporates a more detailed health effects component
than in traditional LCAs, the electronics industry can use the tools and data to evaluate the
health, environmental, and energy implications of the technologies. With this evaluation,  the
U.S. electronics industry may be more prepared to meet the demands of extended product
responsibility that are growing in popularity in the global marketplace, and better able to meet
competitive  challenges in the world market. In addition, the results and model in this study will
provide a baseline LCA upon which alternative technologies can be evaluated.  This will allow
for more expedited display-related LCA studies, which are growing in popularity by industry and
may be demanded by original equipment manufacturers (OEMs) or international organizations.
       EPA and interested members of the public can also benefit from the results of the project.
The project has provided a forum for industry and. public stakeholders to work cooperatively, and
the results can be used by stakeholders as a scientific reference for the evaluated display
technologies. The results of the project could also be of value to other industries involved in
designing environmental improvements into the  life cycle of consumer products.
                                          i-6

-------
                                                                    1.3 PRODUCT SYSTEM
 1.3    PRODUCT SYSTEM         *.  •*          I

 1.3.1  Functional Unit

       The product system being analyzed in this study is a standard desktop computer display
 that functions as a graphical interface between computer processing units and users. The desktop
 market was chosenjto evaluate CRTs and FPDs because it represents a large market for CRTs
 and an anticipated large market for some FPDs.  Also, there are a limited number of technologies
 that meet desktop specifications.  Therefore, focusing on the desktop market will affect a
 significant number ;of products, while keeping the scope of this project manageable.
       The product system is the computer display itself and does not include the central
 processing unit (CPU) of the computer that sends signals to operate the display.  It is assumed
 that the LCDs operate with an analog interface, and therefore are compatible with current CRT
 CPUs as plug-and-play alternatives.
       In an LCA, product systems are evaluated on a functionally equivalent basis. The
 functional unit is used as the basis for the inventory and impact assessment to provide a reference
 to which the inputs! and outputs are related. For this project, the functional unit is one desktop
 computer display over its lifespan. Data collected in this project have been normalized to a
 display that meets the functional unit specifications, which are presented in Table 1-2. These
 product performance specifications are assumed to meet the requirements of the system
 functional unit in the predictable future (i.e., computer technology as predicted through the year
 2001). The CRT technology is the current industry standard for mis product system.
Table 1-2. Functional unit specifications
Specification
display size a [
diagonal viewing area>a
viewing area dimensions
resolution
brightness :
contrast ratio
color
Measure
17" (CRT); 15" (LCD)
15.9" (CRT); 15" (LCD)
12.8" x 9.5" (122 in2) (CRT); 12" x 9" (108 in2) (LCD)
1024x768 color pixels
200 cd/m2
100:1
262,000 colors
the viewing area of a 17" CRT is about 15.9 inches and the viewing area of a 15" LCD is 15 inches. LCDs are not -
manufactured to be exactly equivalent to the viewing area of the CRT.

       Besides the CRT display, several FPD technologies were considered for inclusion in this
study.  Among the FPD technologies that exist, the amorphous silicon (a:Si) thin-film transistor-
(TFT) active matrix LCD technology meets the requirements of the functional unit within the
parameters of this analysis and is assessed in this study.  Section 1.3.3.1 describes the LCD
technology further, land Appendix A briefly describes several FPD technologies and explains why
the non-LCD technologies are not considered for standard desktop computer uses as defined for
this study. The following subsections briefly describe both CRT and LCD technologies.
                                          1-7

-------
13 PRODUCT SYSTEM
1.3.2   Cathode Ray Tube

1.3.2.1 CRT technology

       CRT monitors are a mature technology and are the current industry standard for desktop
computer displays. The technology is the same as that for a television. CRT displays use high
voltages to accelerate electrons toward a luminescent material (phosphor) that is deposited on a
faceplate. The phosphor converts the kinetic energy of the electrons into light.  In color CRTs,
the phosphors are patterned in dots or stripes of red, green, and blue phosphors. The electrons
are emitted from three cathodes in an electron gun assembly and pass through an apertured metal
"shadow mask" during their passage to the phosphor. Electrons from each cathode that are
directed at the wrong color phosphor are absorbed by the shadow mask. Pictures are created by
first focusing the electron beams into tiny spots, which are moved by deflecting the electron
beams electromagnetically with the "yoke." This system is extremely efficient in that it only
requires three video drivers and connections instead of the 2000 or so in the most common FPD
(MCC, 1994).                                             .      .  '
       The high voltages used to accelerate the electrons must be insulated from the external
surfaces of the tube and the CRT must have excellent electrical insulating properties. The
decelerating electrons produce x-rays and the CRT must also be a good x-ray absorber. Leaded
glass surrounds the cathode ray tube to absorb the x-rays.
       The major parts of the CRT display are the faceplate (glass panel), shadow mask (also
referred to as the aperture mask), a leaded glass funnel, and the electron gun with the deflection
yoke. Various connectors, wiring, an implosion band, printed wiring boards (PWBs), and the
casing comprise most of the rest of the display.  Table 1-3 presents a more comprehensive list of
the CRT components and a list of the component materials identified from disassembling a
monitor, and additional research to identify the material makeup of some components (MCC,
1998).
                                           1-8

-------
                     1.3 PRODUCT SYSTEM





aterials
&
«
•*2
e
41
c
0
A
S
o
i

H
K

.^
o
•4«*
.22
-3. Preliminary!
<•«
H










Materials


















Component parts








j.

ro
05
Glass (1-2.5% PbO alkali/alkalim
aluminosilicate)

A
1



x-s
£
fs

»
T
CO
O,
5














3

W
„
ZnS, Y2O2S (powders): Sn, Si


!






52
i
I



















Polyvinyl alcohol


*






.CO
1
1





£
CT
1
1
1












4
•a
i









a.
'

*:
'

P




4
^








1



'*
I


'


3
a
-^
¥
i
S -a
«•§


3






1"
O
s
o
2



CO

*r{
Mixture of alcohol and plas


•





C?
03
8
S
a



















'


1




bQ
1
CS
O
O
G
<


















*


















ternal electron shield
S











g
co

i
i







1







hadow mask assembl]
CO











CrNiFe and NiFe

1
i






13
1
co















,-v

Bi
^o
Lead oxide, zinc borate (-709




















1
*•
(lead solde
4->
.£
I





O
£
1.
CO
CO
C3
•a
J


















Glass funnel






o
a


"a
•0
«
Aquadag** (may also add iron <















i


Conductive coating


QJ
1
§
u





a
*z
"S
i
(L
'S
t
i


















I

>"
C
C

I
)
)
'




















|

i
j
i
>



U
Crt
S
U
tf
T3
O
1


















S
"S
w







1
•I

ti
S
aded glass (30% PbO alkali/alkaline
5

















Neck glass












i
I
r*
C


















C
,c
1












i
:
:
*


















t












0.
1
'%
1


















ase & too neck, rines
ra











•
•
<
»
•^


















)
^












C/
o
£


















Brass rins. brackets



































i








i
(5


















£3
|
§
1
1































)












CO
J3


















Screws, washers












1
co


















Neck clamp












Polysulphone


















Insulating rings










c3
o
Misc. electronics and resin b















i


NeckPWB












1
co




















•§
ca
1
w
^O
H


1-9

-------
13 PRODUCT SYSTEM





^2
CB
*S
«
TJ
03
1
C
o

5
o
u
1
-3


&
l^^
^^
o
f
C3

*3
i
rA
O
1












Materials

















.2
C3
0.
"S
o
§
I











5
300/400 series steels fFe. Ni.


















y
*c
,u
"a
u
c
'c
ca
j:
V.
•a
*S
C
_c;
t!
«•
er
£
C.
^a
5
H*"









i-





















)
«









^
s
8
1 (coated with mixtures of Ba, Sr, C
a


i
!
i















Cathodes












Borosilicate class





















|
Z
«
C












}





















,












"c
c
V
C
|
i
E-
e
.**
"c
1
Jg
t
1























\
1


•*
4
U
L,
.

!
;
1
!
)
4
a
U
"c
«£
U9
1
1
0.
•r
1
b!
a:
t:




















fc
<£
V
1
c
El












)
4
i





















•»












^

























^
.£
*5
K
C





>
)

























1
'!
i :
'1
S





ca
|





















V
1
ff















































1
(2












"








1
g
PC












•»
\


































£
a
1
|





















(/
!
X












-





















_
.
>
>
4
;












'





















Connectors












1
CO





















CO



0
'§
a3
1





"<
i
c/




















c
ackets, anode connectio
PQ



<2
a
c.





3
5




















^-N
A
f
¥
Is
J3
co











Silicone rubber















1





Rubber feet












1





















Anode cap












1





















Insulator pad







                                   1-10

-------
                                                                   1.3 PRODUCT SYSTEM
1.3.2.2 CRT manufacturing           .   ,            ,   ,
                             -       f   i          "'• -   i
                 I                   .              -

       The manufacturing process of the CRT (Figure 1-2) involves first preparing the glass
panel and shadow mask. The shadow mask is a steel panel with a mask pattern applied through
photolithography. ;The color phosphors and a black matrix coating (aquadag) are applied to the
faceplate, also using photolithography, which involves several steps and several chemicals that
etch the material into specific patterns.  A lacquer coating is applied to the phosphor-coated glass
to smooth and seal! the inside surface of the screen, and an aluminum coating is added to enhance
brightness and as a conductor to allow the use of voltages over 12 kilovolts without charging of
the phosphor screep. An electron shield (typically aluminum) is attached to the shadow mask
assembly to prevent stray electrons from reaching outside the screen area  (EPA, 1995; MCC,
1994). This then comprises the faceplate assembly. The two major remaining parts are the
funnel and electron gun.
                                         1-11

-------
13 PRODUCT SYSTEM
o

o
CD
CD
15
to
A


•a
.2
CO


+

^
o
CD
1



™

*-"
CD i.
_ _
CD CO 3
CO -0."-
co g03
.cfl
' t '
assemble
faceplate (glass
andshadow
mask)
A

f

neck assembly
preparation
LA


Pi
aluminiz<
A
i
to
o

1
j

CD
i
(
T
3
co



•g =
CO 0
shadow m
preparati



k.


_^
.u O
3 =
O) (j)

-------
                                                                   1.3 PRODUCT SYSTEM
       The leaded-glass funnel is washed ajid coated with a'black coating (aquadag), which is a
good electrical conductor and a non-reflective surface.  The funnel and faceplate assembly are
joined using frit (solder glass) that is approximately 70% lead oxide. The electron gun is an
assembly of glass aind several metal parts that are heated to embed the metal in the glass. The
electron gun is then fused to the funnel neck. Finishing steps are conducted to complete the
manufacturing of the CRT (or "tube") (e.g., attaching a metal implosion band for implosion
protection and safety) and then the entire monitor is assembled with other necessary parts (e.g.,
the main and neck PWBs, power cord, casing).
                 ! ,

1.3.3  Liquid Crystal Display

       LCDs are used for various applications, with their largest market currently in notebook
computers. LCDs have been gaining a presence in the desktop computer display market and are
expected to continue to do so. Compared to the CRT, the LCD provides a more compact display,
as well as being of higher contrast, sunlight readable, more reliable, and more durable (i.e.,
requiring much less maintenance) (Koch and Keoleian, 1995). In general, the major functional
disadvantages haves been that the resolution and quality of the image have not matched that of
CRTs. However, emerging technologies are expected to meet user requirements.
       The two most common types of LCDs are passive matrix (PMLCD) and active matrix
(AMLCD). Brief descriptions of these and other LCD technologies are presented in Appendix A,
Table A-l. In 1998, AMLCDs constituted approximately 81% and PMLCDs 19% of computer
monitor LCD production (MCC, 1998).  PMLCDs are used primarily for low-end products (e.g.,
cannot perform video applications); AMLCDs are used for high-end multimedia products and
better meet the specifications for standard desktop computers. PMLCDs are forecasted to decline
to less than one percent of the LCD  desktop display market by 2002 (Young, 1998). Therefore,
this project's focus;is on AMLCDs.

1.3.3.1 LCD technology

       In general, an LCD is comprised of two glass plates surrounding a liquid crystal material
that filters external i light. LCDs control the color and brightness of each pixel (picture element)
individually, rather! than from one source, such as the electron gun in the CRT.
       The most cbmmon type of AMLCD, and the one that meets the functional unit
specifications of this project, is the a:Si TFT AMLCD (see Appendix A, Table A-l for
descriptions of various types of AMLCDs). AMLCDs consist of driver tabs along the columns
and rows of the display glass and transparent parallel electrical lines across the glass arranged to
form a matrix. Each intersection of the matrix forms a pixel. The TFT AMLCD has a transistor
at each pixel which functions as an electronic switch to activate an individual pixel. This active
addressing technique allows for high contrast between the on and off states of a pixel.
       Operation 6f the AMLCD is determined by how the liquid crystals are aligned when
activated by an electrical current.  Traditional AMLCDs use a twisted nematic (TN) operating
principle of the liquid crystal, which is evaluated in this study. The orientation of the liquid
crystal molecules either allows or does not allow light from a backlight source to pass through the
display cell. When no electrical current is present, the liquid crystals align themselves parallel to
a polyimide orientation film (alignment layer) on the glass. When a current is applied, the liquid
                 i
                 I.               __

-------
13 PRODUCT SYSTEM
crystals turn perpendicular to the glass. The combination of the alignment layer, electrical charge,
and polarizers that are laminated to the glass panels affect an on or off state of the LCD cell (see
Appendix B for further explanation; also see Castellano [1992] and OTA [1995]). The backlight
supplies the light source for the display and generally has four cold cathode fluorescent tubes that
contain small amounts of mercury vapor. Because the LCD technology essentially regulates
passage of a backlight through the display, LCDs are considered non-emitting display
technologies. CRTs, on the other hand, are emitting displays which emit electrons to illuminate
appropriate phosphors.
       There is a variation of the a:Si TFT AMLCD technology called in-plane switching (IPS).
Compared to the traditional TN mode, IPS TFT allows for a wider viewing angle that is
comparable to the CRT.  The traditional TN mechanism vertically twists the liquid crystals by
sending voltage through the display from electrodes that are on both the front and rear glass
panels.  IPS mode twists the liquid crystals horizontally in response to a voltage applied by
electrodes on the rear glass only. The TN TFT uses indium-tin oxide (TTO) as the transparent
electrode on the front and rear glass panels. For the IPS TFT, no ITO is needed on the front glass
and the electrode on the rear glass is made of any of a number of other materials (e.g., Mo, Ta,
Al/Cr, MoW). Therefore, no  ITO is used for the IPS mode. However, the IPS TFT display
demands an increase in the number of backlights to meet the brightness requirements for desktop
applications (DisplaySearch,  1998). Although this technology may be produced for displays that
meet this study's functional unit, the manufacturers of 15" AMLCDs that supplied data for this
study provided data only for traditional a:Si TFT AMLCDs. IPS was forecasted to have a 35%
market share of LCD desktop monitors in 2000  (Young, 1998) and therefore, studying IPS
technology may be an area for future research.
       Based on the disassembly of an LCD monitor conducted by MCC, a summary of the
materials that are in an AMLCD are presented in Table 1-4 (MCC, 1998). The major components
of the AMLCD are the glass panel (which includes the transistors, electrodes, liquid crystals,
orientation film, polarizers, and row and column drivers), the backlight assembly (including the
cold cathode fluorescent tube, light guide, and associated electronics), other electronics (main
LCD controller), and the stand and cover. The remaining components and materials are listed in
Table 1-4. Tin this  report we will also refer to the LCD "module" as a component. This is
comprised of the LCD panel,  backlight unit, and main LCD controller. Module manuf acturing as
a process modeled in this study includes the major process in LCD manufacturing, which is panel
manufacturing described below.
                                          1-14

-------
                      1.3 PRODUCT SYSTEM






an
CQ
-

1
g
1
1
i
"«r
1
O
u
Q
U

•f

1












Materials












h.
a
•*•*
e
a
S
LCD compi












Soda lime or borosilicate


1




«,
1



















*
u
1
._r
bb
d

I
*



J$
"53
in film tran
e



















Indium-tin oxide (ITO)







1
1







r










. i
_O '
>3 \
ine, cellulose triacetate-acr
"etc.
3:

i



i
I '
1;




i
3;






.
.
.
: "
[ h



|
i


Polyimide


i
I

i

i
tation film i
g
'§









7!
IA^L» glass pant
assembly





A
e
, phenylcyclohexanes biphe
bp
1
1



J

1
3












t







.S

i
t
t



52
i~*
*B
8'
6




















0.
"e
c
c
Is
^











1
^
1
1
fa
Row/colu








i
'
I


*
L,
4

























>>
B
A
§
3


3
U











CO
O-i
1
¥
CJ
S
Row/colun












(L
1
1
•6
E
£
g
a
C












X
£
1

1












*
4





























|
a
V

V
$
1


1
Si
1





















J



N
! G
= ?
(




























r
i
o
3
























GO
1





































1
ft


















fli
T— j

wer cord recent
«s




























4







Aluminum





















Heat sink







4



















P9
C;
ower supply P\^
Ft,




1-15

-------
13 PRODUCT SYSTEM
g



1
Is
S
8
C3
.2
&
a
a


s
8
•







Materials |















component parts
O



























tl
*o5
i


i
*<
£




















4


?
^
*
3




















I
c/:


























W
1
PQ

















st
c
•g
£
cr
•s
T
cc
|
•*-
c/
0
(S























I
:
t
t


1
I




















1
c:
C























1
1
:






















_«.
1
c
1
t?
c
i


c
c
1




















:
>
i


3
>>
r


1

















"o
o
s


^
"e
S
s



















^
'c
\
s
a.
C


)




















>
4
Ll
1






















fc
•V
£
•5
a
c
a
C


j
0
<
1




















r
u


^
•t



















I-
1
i
i






















.
•
>
,
•
:
•4
J
|
C
Polvcarbonate 1




















02
C
'5
*.
JC
.tt
"^
S:
2
•a
a
*




















u
u
)
3
I
1




















OJ
C
£
1
t


)




















•


bJ
tr
1
\
I
a














?
3
\
1


























Insulated Cu 1







j
j



I


Cable assembly







'a?
fe
— -








i
i

i


S "c
-Is
O-i 55 p
C^ *c
1










"












^




]
J
1
5
1


_o















Cable clamp


»-»
3
2
§
2
1
3
§


Polycarbonate : 1















1
_t
1
S







,>























C
I
s















rs

-------
                       1.3 PRODUCT SYSTEM















M
'1
I
e
CC

c
CU
8
e
ft
|
8




^t*
•
^^
4)
2
H











Materials





















fl
!3



S
g
0
a
S
S
Q
J













































Cfl
S
co








/•
fe
C































-











&
$
1
0


































1
"c
1
































'










*








c
a.
PC


































1 BeCufineers
S
8
§
*




:
•


































)
»

<
O
0







































Cloth mesh
»»
i
i
0
3
















































5
1














.



















1 Insulator
























,






.



.
















£7
cc
<
_c
V
R
S

































1 Rear cover








N
!
/






























'














































S2
1 Brackets & washe























































































1 Axle & sorine








§
55




























.
1

.


>
4





















.





















Q
1
V.











































D




















-






















C
§
n
^

c







































4
*
3
t>
^
%
Q




"S
i
Cf
V
V
£
1=
"5
£




-





•





















_
Swivel bearine 1
>.
1
g








































>





^*
"o
•*r
a
c.
v2

«.
1
!
f
>
i
T»
^
4






























3








w
PC
<
_L
c/
R
£










-






















I








s
/







































<

O
c
"5
1
V
.1
"c
|
C
^
K
V
p





























1
'I




5
3
9
§
b
!
!
i
i,


































)
4








c
1


































£l
1








i
•*








i

























D








I
r
a.
i^
V.

































Rubber feet








!
i
;
«
I
!
1













































































1
co








c











































I
I
/)

































h
Power suDolv cov«











































•»











































Front bezel

















































...





































1








CO
ff



































c
1
OJ
5







































Power switch

1
C





































-




1
i
±





i
cc
1
1
1

































O
c
"S
•*-»
-C
.£








;
!


































ri
«(
)







^
~t
z.
t.
^G

































PH
1
S







V
I
i
i
i
i

































^








r
1


































Cable clamo








•
•*


































4








I


































Insulator











*aj
o
c
a
1
CO
r
J5
1
CO
co
r?
v^
CO
'fc?
u
U-t
00 °
a\ 5
5 1
^1
i) 'S



O »t3
co .2
S

g
g
*8
O

1




1-17

-------
13 PRODUCT SYSTEM
1.3.3.2 LCD manufacturing

       LCD technology uses a glass substrate (e.g., soda lime or borosilicate glass).  Once the
glass is acquired, it must be cleaned, which is a critical step in reliable manufacturing.  The
manufacturing techniques of the LCD are similar to the production of semiconductor chips,
which require energy intensive "clean rooms" for manufacturing. A conductor or semiconductor
is deposited on the glass substrate. Most FPDs require a transparent conductor (electrode) such
as ITO, which is usually deposited by a sputtering method.  TFT devices in AMLCDs use
semiconducting materials (e.g., silicon, cadmium selenide) for transistors at each pixel. These
semiconducting materials can be prepared by vacuum evaporation, using either electron beam
evaporation, sputtering, or chemical vapor deposition.  Electrode patterns are then formed by a
photolithographic process that begins with the coating of photoresist on the ITO or metal-coated
substrate.
       The photoresist (photosensitive organic polymer) is  then "developed" using liquid
organic chemicals. This is similar to the manufacture of silicon integrated circuits (ICs). The
ITO or metal is then etched to form electrodes. FPD manufacturing employs etchants (e.g.,
H3PO4/HNO3, HF/HC1, HC1O4, CF4/SFg) that differ from those used in silicon 1C manufacturing.
LCDs use alignment layers and iodine/dye based polarizers. AMLCDs use amorphous silicon,
silicon nitride, various oxides and metals. The silicon 1C manufacturing process of plasma-
enhanced chemical vapor deposition (PECVD) is used for some of these layers, as are more
conventional vacuum deposition techniques. Finally, various organic liquids (the liquid crystals)
are injected between the top and bottom substrate of LCDs (MCC, 1993).
       The general process flow of AMLCD manufacturing is presented in Figure 1-3. The
photolithographic subprocess steps for the front and rear glass panels are presented in Figures 1-4
and 1-5. Various processes include sputtering, PECVD, photolithography, wet etching, reactive
ion etching (RJE), in-process testing, liquid crystal processing, and lamination among others.
The manufacturing process of IPS differs from traditional TN in that there are fewer
photolithography steps required since no ITO is patterned onto the front or rear glass plates.
       Once the photolithographic processes are completed, the polyamide orientation film is
applied and rubbed on the glass. The front and rear glass substrates are merged and the liquid
crystals are inserted. Polarizers are laminated to the front and rear panels, completing the
"AMLCD cell" of the LCD glass panel assembly (see Table 1-4).  The electronic components
that operate the AMLCD cell (i.e., the column  and row driver tabs and wiring boards) are men
attached to the glass. This completes the LCD glass panel assembly.  The other major
components of the LCD are then assembled. Adding the controller PWB and the backlight
assembly make up what may also be called the "LCD module." Finally, the power supply
assembly and the plastic cover and stand are added to make an assembled monitor.
       In total, there are six PWBs needed for the LCD: controller, row driver, column driver,
backlight, power supply, and adjustment knob  PWB.  The major ones by size and. function are
the larger controller and backlight PWBs, and the smaller column and row PWBs. These are
compared to the two major PWBs in the CRT: the main and neck PWBs (see Table 1-3).
                                          1-18

-------
                      1.3 PRODUCT SYSTEM
1-19

-------
13 PRODUCT SYSTEM
                                  1-20

-------
                                                           1.3 PRODUCT SYSTEM
  u
  CO
.§•














X
Q.
OTOLITHOGRA
x
Q.

T) FORMAT IO*
p
Si-
nr
O
ILM TRANSIST
u.
•
Z
fE






























,
'


1

! "

S
i w
;; 1

1 JL.
. [
"*"* *••»
(w 
(0 T-

to alignment
frafar bank fr




















                                    1-21

-------
13 PRODUCT SYSTEM
1.4    ASSESSMENT BOUNDARIES

1.4.1  Life-Cycle Stages and Unit Processes                    -    ..

       In a comprehensive cradle-to-grave analysis, the display system includes five life-cycle
stages: (1) raw materials  acquisition; (2) materials processing; (3) product manufacture; (4)
product use, maintenance and repair; and (5) final disposition/end-of-life. 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 CRTs and LCDs, which are modeled in this study, are
depicted in Figures 1-6 and 1-7, respectively. Details on collecting data for these processes are
presented in Chapter 2.

1.4.2  Spatial and Temporal Boundaries

       The geographic boundaries of this assessment are shown in Table 1-5. This LCA will
focus on the U.S. display  market; therefore, the geographic boundary for the use and disposition
stages of displays is limited to the United States. Raw materials acquisition and material
processing for materials used in the manufacture of computer displays are done throughout the
world. Product manufacturing is done predominately in Asia, although there are foreign-owned
desktop display manufacturers operating plants in the United States and other countries.
Therefore, for purposes of this study, the geographic boundaries for raw material extraction,
material processing, and product manufacture are worldwide.

                 Table 1-5. Geographic coverage for each, life-cycle stage
Life-cycle stage
Raw materials acquisition
Material processing
Product manufacture
Use
Disposition
Geographic coverage
worldwide1
worldwide3
worldwide*
United States
United States
                 alh this study, worldwide boundaries were considered; however,
                 the actual geographic locations for LCI data are presented in
                 Chapter 2 (Sections 2.2 and 2.3).
       While the geographic boundaries show where impacts might occur for various life-cycle
stages, traditional LCAs do not provide an actual spatial relationship of impacts. That is,
particular impacts cannot be attributed to a specific location.  Rather, impacts are generally
presented on a global or regional scale.
                                           1-22

-------
                                       1.4 ASSESSMENT BOUNDARIES
 W
 1
 o
og
gg
coO
°1
sS
IB
si
Inci
                      *i
                      s—
                      ll
                     o
                        *1
                        ll
                             *
         ll-
         J*3
                    .•
                          I
l-li'l-g;
M £3 O •*-* OQ
*«-( S *S  S
o ^ p ^g «


.'M *^ Ca *^3 w
S .2 » » S§

§-||lg

^ 9 "S o
                                                         3
                                                       I
                                                        ;«
                                                        S
                                                       si £
                                                        S sS
                                                        f-s

                                                        ll
                                                        S-3
                                                        ?
-------
1.4 ASSESSMENT BOUNDARIES
   ~v
         §
SPOS
         1
PRODUI


                                                               ~s
                          •§
         s§
         gp
         Hoa

         S|
                                •043

                             ^Sl-S-
                                  1-24

-------
                                                            1.4 ASSESSMENT BOUNDARIES
       This study addresses impacts from the life cycle 5f a desktop computer display
manufactured using 1997-2000 technology. The use and disposition stages cover a period that
represents the life qf a display.  Two lifespans are considered: (1) the "effective" life, defined as
the period of time the display is in use by primary, secondary, or even tertiary users before
reaching its final disposition; and (2) the "manufactured" life, defined as the designed durability
of a display. The effective life is estimated based on past and current use patterns of displays and
represents a realistic estimate of the lifespan. Because the effective life is subject to many
variables, including fluctuating market trends, it is also necessary to evaluate the .displays over
their manufactured jlife. The manufactured life is estimated based on the manufacturer's
estimated durability of the display.. Because of quickly changing technologies in this industry,
the effective life has been shorter than the manufactured life. The effective life, which is
currently the more realistic scenario is used as the primary scenario in the final results presented
in this study. However, the manufactured life is presented as an alternative scenario.
       It is assumed that parameters that may change with time, such as available landfill space,
will remain constant throughout the lifespan of the product system. If the lifespan is relatively
short (i.e., within aitimeframe where significant changes in landfill space would not occur), the
preceding assumptipn is reasonable. If resources become more scarce within the lifespan, this
assumption could Underestimate the impacts.

1.4.3  General Exclusions

       Impacts frbpi the infrastructure needed to support the manufacturing facilities (e.g.,
maintenance of manufacturing plants) are beyond the scope of this study. However, maintenance
of clean rooms used in the manufacturing of LCDs (and other components), which require
substantial amounts of energy, are considered part of the manufacturing process.
                                          1-25

-------
REFERENCES
                                   REFERENCES
Atlantic Consulting and IPU.  1998. LCA Study of the Product Group Personal Computers in the
       EUEcolabel Scheme.  LCAStudyVersionl.il.  January.

Castellano, J. 1992. Handbook of Display Technology.  Stanford Resources, Inc., San Jose, CA.

Curran, M.A. 1996. Environmental Life-Cycle Assessment. McGraw-Hill, New York, NY.

DisplaySearch. 2001. Personal communications with R. Young, DisplaySearch, and J.G.
       Overly, University of Tennessee, Center for Clean Products and Clean Technologies.
       July.

DisplaySearch. 1998. DisplaySearch FPD Equipment and Materials Analysis and Forecast.
       Austin, TX.

EPA (Environmental Protection Agency). 1995. EPA Office of Compliance Sector Notebook
       Project - Profile of the Electronics and Computer Industry. EPA310-R-95-002. Office
       of Enforcement and Compliance Assurance. Washington, DC. September.

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 & SETAC Foundation for Environmental Education, Inc. Washington,
       DC.

FRC (Fuji Chimera Research). 1996.  "The Future of Liquid Crystal and Related Display
       Materials." Originally published in Japan by FRC and translated by InterLingua
       Linguistic Services, Inc. Redondo Beach, CA. 1997.

ISO (International Standards Organization). 1996. ISO 14040, Environmental Management -
       Life-cycle Assessment Principles and Framework. TC2071 SC5N77. International
       Standards Organization, Paris.

Kincaid, L. et al. 1996. Cleaner Technologies Substitutes Assessment - A Methodology &
       Resource Guide. U.S. Environmental Protection  Agency, Office of Pollution Prevention
       and Toxics. EPA 744-R-95-002.  Washington, DC.

Koch, J.  1996. Evaluating Environmental Performance: A Case Study in the Flat Panel Display
       Industry. University of Michigan, Natural Resources and Environment, Ann Arbor, ML
       April.

Koch, J. and G. Keoleian. 1995.  "Evaluating Environmental Performance: A Case Study in the
       Flat-Panel Display Industry." Proceedings of IEEE International Symposium on
       Electronics and Environment, Orlando, May 1-3, pp. 158-165.
                                         1-26

-------
                                                                        REFERENCES
MCC (Microelectronics and Computer Technology Corporation). 1993. Environmental
       Consciousness: A Strategic Competitiveness Issue for the Electronics and Computer
       Industry. Austin, TX.
                |                       -           /
MCC (Microelectronics and Computer Technology Corporation). 1994. Environmental Industry
       Environmental Roadmap.  Austin, TX.

MCC (Microelectronics and Computer Technology Corporation). 1998. Computer Display
       Industry arid Technology Profile. EPA 744-R-98-005. December.

NEC Corporation. >1998.  "The LC Monitor Market as Predicted by a Monitor Manufacturer."
       NEC Home| Electronics, Inc. In Flat Panel Display 1998 Yearbook.  Nikkei
       Microdevices. Originally published by Nikkei Business Publicationa and translated by
       InterLingua Linguistic Services, Inc., Redondo Beach, CA.  1997.

OTA (Office of Technology Assessment).  1995. Flat Panel Displays in Perspective. OTA-
       ITC-631. U.S. Government Printing Office, Washington, DC. September.

Orango A.B. and Atlantic Consulting.  1999. "EU Eco-Label for Portable Computers-Suggested
       Criteria." Version 1.12.  Gotebprg, Sweden. May 19,1999.

SRI (Stanford Resources, Inc.).  1998.  Web site available at:
       http://wwwistanford.resources.com/sr/main/homepage. April.

Young, R. 1998. Personal communications with R.Young of DisplaySearch and M.L. Socolof,
       University of Tennessee Center for Clean Products and Clean Technologies, June.
                                         1-27

-------
this page intentionally left blank

-------
                                                             2. LIFE-CYCLE INVENTORY
                 i                      Chapter 2

                              LIFE-CYCLE INVENTORY

       A life-cycle inventory (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 (Figure 2-1). For the Design for the Environment (DfE) Computer Display
Project (CDP), LCjI inputs include materials used in the computer display product itself, ancillary
materials used in processing and manufacturing of the displays, and energy and other resources
consumed in the manufacturing, use, or final disposition of the displays.  Outputs include
primary products, Co-products, air emissions, water effluents, and releases to land.  Specific unit
processes for CRTJs and LCDs are represented by the boxes in Figures 1-6 and 1-7, and each unit
process has inputs |and outputs particular to that process. Figures 2-2 and 2-3 also show each unit
process for both the CRT and LCD life cycles, and graphically displays how they are linked to
subsequent processes. This figure will be referred to throughout this chapter in the discussion of
each life-cycle stage.
           INPUTS:


      Display materials
      Ancillary materials'

       Energy/resources
OUTPUTS:
Unit Process
> Water effluent
^ Releases to land
^ Primary products
                 i Figure 2-1. Unit process inventory conceptual diagram

       This chapter describes the methods for collecting LCI data in the DfE CDP, and presents
LCI results. Section 2.1 describes the general methodology for LCI data collection, while
Sections 2.2 through 2.6 present the specific methodologies, data sources, data quality,
limitations and uncertainties for each life-cycle stage.  Section 2.7 then concludes with the
combined LCI data for each monitor type.
       More specifically, Section 2.2 presents the LCI methodology for the materials extraction
and materials processing (i.e., "upstream") life-cycle stages, including electricity generation.
Electricity is used in several processes throughout each monitor's life-cycle, and the electricity
generating processes linked to the processes that use electricity.  As a consequence, the inventory
results from electricity generation are reported as part of the associated life-cycle stage for the
process to which it is linked.  For example, electricity used in manufacturing a product
component is included in the  manufacturing stage inventory, while electricity used during the use
stage is included in the use stage inventory.
                                          2-1

-------
2. LIFE-CYCLE INVENTORY
UI

I
CO

I
il
o
a
in



8
CO
UI
CO
•£ 5




«
i|
1)

11
Q.








4t
O
D-
_l














                                       2-2

-------
                2. LIFE-CYCLE INVENTORY
Ill
9
<
55

j
iL
o

111



Ul
3
CO
w
CO
3














„ O)
CO c
U -c
§3
S <*
6^
Si
< s
,
fe CD ?;
o « -e
£§!
« 2 c
•O O Q. _
'§,8^ 0
01 o> O **
o o uj 2
i 2 -^. CD
g fc-tg
0 H ~2 »
w^St
3°3 g£
CD ™ t3
11!
T-*J CD 0)
1 8S
= 8 ®





•o
CD
TJ
T3
CD
J3

1
• s:
CO
CD
y,
8
2 "co
a^
E"E
co ?*
•o S o
•C C J2
0> 0) =
0 Q. S
B = S
8 g c
o o .9
CD S «
S-0 g
gc®
S.S S
CO CO ><
"* 8~
O "o
Q. ®
0,®
!
2
CO
E
0
•a
JS
.S
:=,




_|>







-»•










J








->•




^
























O *
— CO
a

._
o *
CD =it:
=>

CO *
38
CO 0>
c












(



















^


^

4
1














3
J



















•j





P
S
O
2
o





ra
E
|
Q.



k














t
11



r
b>
E
^.
S
0


%rn
"£



















































"o
r
c
Tj
c
?





















•--.




^










•8
S

-" co

\
\
NT

-





E
3 en
c .g1
E E
"co


D,
E
1


^^

	 -^
^

CO •>
i= CO
3 co
CO O
c


1
i
m

CD
1*0,
J?!



•5
D)

f-


1

X








^


\
V*
\
1


— • -^

-->
X^

13
=>




/^
c
c
£
c,
•C



Q
y



^
\
\











t







c
^

1


4
_
.
/
5 =
5 a
3 1
> £
1 O
J-









>

I
j
•'i
S
i



2
5
=
1
3
j



3
O
E
3
^




_«
E
3
^
z
S
u
0
D
^
*
*




k.


r^





|


k
>



s.
X3
m
S
CO
L



•tt
Ol
E

J




V.
!

3)
3.
2

/




s























k

1








s

3 -*i c
^~ of?
'*= SJ
2 -1
i1 r^
c co
— i( "o
2 s / -c
•3 — 3? 	
/
'->S
/~






1
^-^^ O
fc^^ QL



\ -g o>

il
/ 'C iJ
^/ a.




E -2 *S3

-------
2.1 GENERAL METHODOLOGY	

       Section 2.3 presents the LCI methods for the product manufacturing life-cycle inventory,
which was developed from primary data collected through questionnaires designed and sent to
manufacturers for this study.  Sections 2.4 and 2.5 present the methods for developing the use
and end-of-life (EOL) life-cycle stage inventories, respectively. Transportation is another
important aspect of a product life-cycle that can cause environmental impacts. Information
related to the transport of materials, products, and wastes is presented in Section 2.6. As the final
section of Chapter 2, Section 2.7, summarizes and discusses the entire inventory data over the
life-cycles of the CRT and LCD computer monitors.  Sensitivity analyses are also discussed in
Section 2.7.
2.1    GENERAL METHODOLOGY

       This section describes the data categories evaluated in the CDP LCI, the decision rules
used to determine which materials extraction and materials processing life-cycle stages (e.g.,
"upstream" processes) 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 (e.g., a
display or display component) 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 inputs, energy inputs, natural resource inputs, emission outputs, and product outputs.
Inventory data were normalized to mass per functional unit (in the case of material and resource
inputs  and emission or material outputs), megajoules (MJ) per functional unit (in the case of
energy inputs), or the number of components per functional unit (in the case of display
components).  As discussed in Section 1.3, the functional unit is one desktop computer display
over its lifespan.
       Data that reflected production for one year of continuous processes were scaled to one
functional unit. Thus, excessive material or energy associated with startups, shutdowns, and
changeovers were assumed to be distributed over time.  Consequently, any environmental and
exposure modeling associated with the impact assessment reflects continuous emissions such that
equilibrium concentrations may be assumed. If the reporting year was less than one year for any
inventory item, the analysis was adjusted as appropriate.
                                          2-4

-------
                           2.1 GENERAL METHODOLOGY
Table 2-1. LCI data categories
Data Category ;
Description
Material inputs (kg per functional unit)
Primary materials
Ancillary (process)
materials \
Actual materials that make up the final product for a particular process. These can be
individual materials or a combination of materials that comprise a component part.
Materials that are used in the processing of a product for a particular process. Process
materials from monitor manufacturing could include, for example, etchants used during
photolithography which are washed away and not part of the final product, but are
necessary to manufacture the product.
Energy inputs (MJ per functional unit)
Process energy :
Precombustion energy
Transportation energy
Energy consumed by any process in the life-cycle.
The energy expended to extract, process, refine, and deliver a usable fuel for combustion.
Energy consumed in the transportation of the materials or products in the life cycle.
Natural resource inputs (kg per functional unit)
Non-renewable
resources ;
Renewable resources
(e.g., -water)
Materials extracted from the ground that are non-renewable, or stock, resources (e.g.,
coal).
Water or other renewable, or flow, resources (e.g., limestone) are included in the analysis.
Renewable resource data values are presented in mass of water consumed for a particular
process.
Emissions outputs (kg per functional unit)
Air ;
Water |
[
1
Solid wastes ;
!
Mass of a product or material that is considered a pollutant within each life-cycle stage.
Air outputs represent actual gaseous or participate 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 discharges to either surface or groundwater from point or
diffuse sources, after passing through any water treatment devices.
Mass of a product or material that is deposited 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.
Products (kg of material or number of components per functional unit)
Primary products
Co-products
Material or component outputs from a process that are received as input by a subsequent
unit process within the display 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.
            2-5

-------
2.1 GENERAL METHODOLOGY
       Data were also collected on the final disposition of emissions outputs, such as whether
outputs are recycled, treated, and/or disposed. This information helps determine what impacts
will be calculated for a particular inventory item.  Methods for calculating impacts are discussed
in Chapter 3, Life-Cycle Impact Assessment. The dispositions used for this project are as
follows:

•      air,
•      surface water,
       landfill,
•      land (other than landfill),
•      treatment,              '
•      recycling/reuse, and
•      deep well injection.

       Given the enormous amount of data involved in inventorying all of the inputs and outputs
for a product system, life-cycle assessment (LCA) practitioners typically employ decision rules to
make the data collection manageable and representative of the product system and its impacts.
Section 2.1.2 discusses the decision rules used in^the CDP LCI.

2.1.2  Decision Rules

       In an LCA, the materials extraction and materials processing life-cycle stages (referred to
as "upstream" life-cycle stages) include processes for extracting raw materials from the earth and
processing those raw materials into the materials used in the manufacture of the product of
interest. Examples of upstream processes include the mining of iron ore and its processing with
other materials into steel sheet or the extraction of petroleum from underground reserves and its
conversion into plastic pellets. A continuing challenge for LCA practitioners is to collect all the
appropriate data for a product system, including data for these upstream processes as well as data
for product manufacturing, use, and disposal processes.  In this project, decision rules as to what
inventories should be included as upstream processes in the overall modeled life-cycle are based
on the materials used to manufacture the computer monitors.  Also, which component parts to
include in the model depends on these decision rules. In considering upstream materials, a
combination of several factors, including availability of existing data were considered.  For
considering which component manufacturing processes to include, the decision rules, plus
manufacturers willingness to participate, factored into our overall scope of what was included in
the analysis.
       To help determine which upstream processes to include in the CDP LCI, first the bill of
materials (BOM) of the component parts and the  materials that make up those parts (Tables 1-3
and 1-4) was reviewed. The quantities of those materials identified by MCC can be found in
Industry Profile Document (MCC, 1998). Using  the MCC BOM allowed work to begin on
selecting and collecting upstream data before the  actual BOM from the manufacturing stage was
obtained from the project's primary data collection effort. Final decisions on which upstream
processes to include were based on the BOM developed from data collected from manufacturers.
Tables 2-2 and 2-3 list the BOMs of the primary material inputs from the manufacturing of the
CRT and LCD, respectively. The mass quantities given in the tables are the primary material
                                           2-6

-------
                                                              2.1 GENERAL METHODOLOGY
inputs to the manufacturing process and Would be equivalent to the amount of material in the
final product plus excess or waste materials. While this is not exactly equivalent to the mass of
each material that makes up a finished product, it does represent the mass of primary materials
used to manufacture the finished product at participating facilities. Details of how these data
were obtained are presented in Section 2.3, which describes the manufacturing stage inventory.
Table 2-2. Bill of primary material inputs for a 17" CRT monitor
Material/Component
| Sub-component
Lead oxide glass \

Steel
Plastics








Printed wiring
Lead
I
!
Polycarbonate (PC)
Styrene-butadiene co-polymer
Polyethylene ether (PEE)
Acrylpnitrile-butadiene-styrene (ABS)
High-impact polystyrene (HIPS)
Triphenyl phosphate
Tricresyl phosphate
Phosphate ester
boards |(PWB) and components
Cables/wires :
Aluminum (heat sink) '
Nickel alloy (invar)
CRT shield assembly !
Fenite '
Deflection yoke assembly
Demagnen'c coil
Video cable assembly
Power cord assembly i
Electron gun |
CRT magnet assembly
Audio cable assembly
Frit
Solder !
Phosphors
Aquadag
Other (misc.)
TOTAL .;
Mass(

9.76

5.16
3.04








0.85
0.45
0.27
0.27
0.24
0.17
0.15
0.13
0.11
0.11
0.10
0.08
0.07
0.07
0.03
0.02
0.02
0.06
21.16
kg)"

0.45


0.92
0.83
0.74
0.32
0.15
0.05
0.02
0.01



















weight % of total inputs *
46.1%

24.4%
14.4%








4.00%
2.13%
1.29%
1.29%
1.14%
0.80%
0.71%
0.60%
0.54%
0.54%
0.47%
0.36%
0.34%
0.32%
0.13%
0.08%
0.07%
0.30%
100%

2.1%


4.36%
3.91%
3.47%
1.52%
0.71%
0.25%
0.11%
0.04%


















X
• Based on the primary material inputs to the manufacturing process, including material in the final product plus excess or waste
materials.          >
                                           2-7

-------
2.1 GENERAL METHODOLOGY
             Table 2-3. Bill of primary material inputs for a 15" LCD monitor
Material/Com

Steel
Plastics






onent
Subcomponent '.


Polycarbonate (PC)
Poly(methyl methacrylate) (PMMA)
Styrene-butadiene copolymer
Polyethylene ether (PEE)
Triphenyl phosphate
Polyethylene terephthalate (PET)
Glass
Printed wiring boards (PWB) and components
Cables/wires
Aluminum (heat sink, transistor)
Solder (60% tin, 40% lead)
Color filter pigment
Polyvinyl alcohol (PVA) (for polarizer)
Liquid crystals, for 15" LCD, unspecified b
Backlight lamp (cold cathode fluorescent lamp, CCFL)
1 Mercury
Transistor metals, other (e.g., Mo, Ti, MoW)
Indium tin oxide (TTO) (electrode)
Polyimide alignment layer
Other (e.g., adhesives, spacers, misc.) ;
TOTAL !
Mass (kg) a
2.53
1.78






0.59
0.37
0.23
0.13
0.04
0.04
0.01
0.0023
0.0019

0.0019
0.0005
0.0005
0.0031
5.73


0.52
0.45
0.36
0.30
0.09
0.06









3.99E-06





weight % of total inputs *
44.12%
30.98%




-

10.31%
6.52%
4.08%
2.34%
0.66%
0.65%
0.15%
0.04%
0.03%

0.03%
0.01%
0.01%
0.05%
100%


9.00%
7.80%
6.31%
5.23%
1.61%
1.03%









0.0001%





 * Based on the primary material inputs to the manufacturing process, including material in the final product plus excess or waste
 materials. '                             .
 b This does not include all liquid crystals, as those specified as individual chemicals are in very small amounts and included in the
 "other" category.                                 :

       The decision rule process begins by assessing the materials and components in Tables 2-2
and 2-3 for the following attributes:

1.      The mass (M) contribution of each component and material in the display.  The mass is
        important in order to account for the majority of materials and components  that make up a
        display, but also because the more significant the material or component by mass, the
        more materials and resources may be required to manufacture the material or component
        and thus it may have a significant environmental impact.
                                             2-8

-------
                                                           2.1 GENERAL METHODOLOGY
2.     Materials that are of known or suspected environmental (Env) significance (e.g., toxic).
       As this is an environmental life-cycle assessment, consideration of materials or
       components that are known to or are suspected to exhibit an environmental hazard are
       also included to the extent feasible.

3.     Materials that are known or suspected to have a large energy (E) contribution to the
       systems energy requirements.  Energy impacts are of great interest to the use and
       manufacture of display monitors and, therefore, priorities were given to including
       materials or components that are known to or suspected to consume large amounts of
       energy.    ;

4.     Materials or components that are functionally (F) significant to the display.
       "Functionally significant" is defined as important to the technically successful operation
       of the display. For example, the liquid crystals in an AMLCD would be "functionally
       significant"; while screws, gaskets, or the plastic cover would not be.
                 i
5.     Materials or components that are physically (P) unique in the CRT as compared to the
       LCD and vice versa. The physical uniqueness of a material or component could be
       identified by chemical makeup or by size. An example of the latter would be if the
       plastic casing for the CRT and LCDs were made of the same material, but the CRT casing
       had substantially more material by weight.
                 I     ' -
       The priority scheme depicted in Figure 2-4 provides guidelines for applying the CDP
decision rules. Material or component inputs that account for more than five percent of the total
mass of a display technology were given top priority for data collection, as were those of known
or suspected environmental or energy significance  and those that are functionally significant or
physically unique. |Of less emphasis in trying to obtain data, but still included if possible, were
materials or components that are functionally significant but physically similar to those in the
other technologies, and those that were between 1 and 5% of the total mass of the display.
Recognizing the limitations of project resources, materials or components that account for less
than one percent of total mass or are not otherwise significant or unique were excluded a priori.
       Based on this hierarchy and on review of the preliminary BOMs, Tables 2-4 and 2-5
present how components were rated based on the priority scheme, and which ones were included
in the analysis. If a. material or component was included in the analysis as a separate process, it is
listed in the last column of Tables 2-4 and 2-5 by what life-cycle stage that process is in (i.e.,
upstream or manufacturing process).  If a material or component was only included as part of
another process, anil not as a separate process in the profile, the process in which that material or
component is found is provided in the last column.
                                          2-9

-------
2.1 GENERAL METHODOLOGY
                      MATERIAL/COMPONENT INPUTS:
                             - >5% of total mass
                            environmental concern
                              - energy concern
                            functionally significant
                             - physically unique
                            - 1-5% of total mass
                                    of total mass
                                - not otherwise
                                  significant  /Excluded from
                       Figure 2-4. Decision rule guidelines
*e.g., materials are excluded if they are not of known environmental significance (for example, toxic) or are not
physically unique.                       -
                                     2-10

-------
                                                                 2.1 GENERAL METHODOLOGY
                  i
  Table 2-4. Decision rule priorities and scope of analysis for the primary material inputs
                                   of the 17" CRT monitor
Material/Component

Sub-component
Lead oxide glass J

Steel
Plastics








Lead;
!
;
Polycarbonate (PC)
Styrerie-butadiene co-polymer
Polyethylene ether (PEE)
Acrylpnitrile-butadiene-styrene (ABS)
High-impact polystyrene (HEPS)
Triphenyl phosphate
Tricresyl phosphate
Phosphate ester
Printed wiring boards (PWB) and components
Cables/wires
Aluminum (heat sink)
Nickel alloy (invar)
CRT shield assembly i
Ferrite 1
Deflection yoke assembly
Demagnetic coil :
Video cable assembly;
Power cord assembly
Electron gun
CRT magnet assembly
Audio cable assembly
Frit
Solder
Phosphors
Aquadag
Other (misc.) |
Decision rule
M, F, P, E, Env
P,Env
M
—
M
" .M' .
M •"-
M
none
none
P
P
M,E,Env
none
M(<5%),E
M(<5%),P
none
P
F,P
F,P
none
none
F,P
P
none
F, P, E, Env
Env
F,P
F.P

Included in analysis as:
manufacturing process
upstream process
upstream process
	
upstream process
upstream process
part of monitor assy, process
upstream process
upstream process
part of monitor assy, process
part of monitor assy, process
part of monitor assy, process
manufacturing process
part of monitor assy, process
upstream process
upstream process
part of monitor assy, process
upstream process
part of monitor assy, process
part of monitor assy, process
part of monitor assy, process
part of monitor assy, process
part of tube mfg. process
part of monitor assy, process
part of monitor assy, process
manufacturing process
part of monitor assy, process
part of tube mfg. process
part of tube mfg. process
miscellaneous
M = mass is greater than 1% of the total display weight; Env = environmental/toxic concern; E = energy concern;
F = functional (technological) importance; P = physically unique.
Note: The CRT processes included in the CDP LCA were presented in Figures 1-6 and 2-2.
                                             2-11

-------
2.1 GENERAL METHODOLOGY
  Table 2-5.  Decision rule priorities and scope of analysis for the primary material inputs
                                 of the 15" LCD monitor
Material/Comi

Steel
Plastics






onent
Sub-component


Polycarbonate (PC)
Poly(methyl methacrylate) (PMMA)
Styrene-butadiene co-polymer
Polyethylene ether (PEE)
Triphenyl phosphate
Polyethylene terephthalate (PET)
Glass
Printed wiring boards (PWB) and components
Cables/wires
Aluminum (heat sink, transistor)
Solder (60% tin, 40% lead)
Color filter pigment
Polyvinyl alcohol (PVA) (for polarizer)
Liquid crystals, for 15" LCD
Backlight lamp (cold cathode fluorescent lamp)
1 Mercury
Transistor metals, other (e.g., Mo, Ti, MoW)
Indium tin oxide (TTO) (electrode)
Polyimide alignment layer
Other (e.g., adhesives, spacers, misc.)
Decision rule
• M
— •
M
M,P
M
M
M (<5%)
; M(<5%),P
M
M,E,Env

M(<5%),E
Env
P
none
F,P,Env
F, P, Env
Env
F,P
F,P
l: F,P

Included in analysis as:
upstream process
	
upstream process
upstream process
upstream process
part of monitor assy, process
part of monitor assy, process
upstream process
manufacturing process
manufacturing process
part of monitor assy, process
upstream process
part of monitor assy. & module infg.
processes
part of color filter patterning
part of polarizer mfg. process
manufacturing process
manufacturing process
part of backlight lamp process
part of module mfg. process
part of module mfg. process
part of module mfg. process
miscellaneous
 M = mass is greater than 1% of the total display weight; Env = environmental/toxic concern; E = energy concern;
 F = functional (technological) importance; P = physically unique.
 Note: The LCD processes included in the CDP LCA were presented in Figures 1-7 and 2-3.
2.13  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 2-6 lists the types of data (primary or secondary) used for each life-cycle stage in
the CDP LCI. In general, greater emphasis was placed on collecting data and/or developing
models for product manufacturing, use, and end-of-life. Primary data were collected from
product and component manufacturers (in the U.S., Japan, and Korea), and CRT recyclers who
voluntarily agreed to participate in the project. When proprietary data were involved, the
University of Tennessee (UT) Center for Clean Products and Clean Technologies entered into
                                            2-12

-------
                                                           2.1 GENERAL METHODOLOGY
 confidentiality agreements with the affected company. In addition, to both protect confidentiality
 and better represent the various manufacturing processes, data for particular processes that were
 collected from more than one company, where possible, and aggregated. Attempts were made to
 get at least two companies to contribute data for each particular process in the manufacturing
 life-cycle stage, which resulted in some process datasets being the aggregate of multiple (2-7)
 companies data.  However, this was not feasible in every case and some datasets were simply the
 data of one company. Details of the data aggregation methods are provided in Section 2.3.
       Data for the use stage were modeled specifically for this project by UT researchers, but
 were based on secondary data (i.e., secondary data were built upon to create the data used in the
 inventory for the use life-cycle  stage). Data associated with the electricity generation were also
 based on secondary data, but modeled for this project. Transportation information (e.g.,
 transportation mode and distances) were collected from the manufacturers that provided primary
 data.  These data were linked to secondary data inventories of fuel inputs and emissions outputs
 for various types of transport vessels.  Transportation data cover movement of materials and
 components both into and out of a facility, but do not include transportation of packaging or
 distribution of the finished display to the consumer. Finally, secondary data were used for
 upstream processes; More details on each of these data collection efforts are provided in
 subsequent sections of this chapter.

                         Table 2-6. Data types by life-cycle stage
Life-cycle stage
Upstream ;
(materials extraction arid processing)
Product and component manufacturing
Use !
Final disposition
(recycling and/or disposal)
Packaging, transportation, distribution
Data types
Secondary data.
Primary data, except secondary data used for frit.
Modeled using secondary data; maintenance and repair are not
included in the analysis.
Modeled using secondary data plus primary data from CRT
recycling facilities.
Primary data from product and component manufacturers for
transport mode and distance; secondary data for fuel inputs and
emissions outputs for the transport vessel. Packaging and
distribution not included.
       In some instances, neither primary nor secondary data were available. For example,
CRTs are a much more mature technology than the LCD, and end-of-life (EOL) data are much
less prevalent for the LCD than for the CRT. Where primary and secondary data are lacking,
various assumption? and modeling serve as defaults.
                  i
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 CDP LCI
allocation procedures are used when processes or services associated with the functional unit
(e.g., a desktop computer display over its lifetime) are used in more than one product line at the
                                          2-13

-------
2.1 GENERAL METHODOLOGY
same facility (e.g., notebook computers, televisions). For example, transistors are used in LCD
desktop computer displays, but also in other LCD technologies, such as notebook computer
displays. If a facility uses a single process to manufacture transistors for both desktop and
notebook computer display, inputs and outputs are allocated among the product lines to avoid
over-estimating the environmental burdens associated with the product under evaluation.
       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 of transistors supplied only desktop
computer LCD manufacturers, no allocation would be necessary from that manufacturer.
However, it is more likely that the transistor manufacturer would have a larger customer base,
including manufacturers of various products other than liquid crystal desktop computer displays.
This requires allocation of flows from the manufacturing of transistors for several products to
those associated only with desktop computer displays.  As suggested by ISO, if sub-processes
within the transistor facility can be identified that distinguish between transistors manufactured
for LCDs and for other products, the latter sub-processes can be eliminated from the analysis,
thus reducing allocation procedures.
       In this study allocation procedures are 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 display or display component according to the
       actual or estimated electricity consumed during the manufacture, use, or final disposition
       of the product. Similarly, fuel inputs and emission outputs from commercial transport of
       a display component to a display assembler are allocated to the display component
       according to the mass of the component, the distance traveled, and the fraction of the
       transport vehicle's capacity occupied by the number of components shipped.
•      Where a unit process produces co-products, the burdens associated with the unit process
       are allocated to the co-product on a mass basis. In the transistor example above, burdens
       are allocated according to the total mass of transistors used in desktop displays and the
       mass used in notebook computers. Total mass can be calculated from sales records which
       document the number of transistors delivered to different customers and measured mass
       of a set number of transistors.    -                                  ...__..

Allocation is also necessary when a single process produces both energy and products. In this
case, the inputs are partitioned among the energy and products, as appropriate, to avoid allocating
inapplicable chemical burdens to energy production. However, this scenario was not
encountered in the CDP LCI.

2.1.5  Data Management and Analysis Software

       The data that were collected for this study were either obtained from questionnaires
developed for this project, from existing databases, or from primary or secondary data collected
by the UT Center for Clean Products and Clean Technologies. All these data were transferred to
spreadsheets, which were then imported into a Life-Cycle Design Software Tool developed, by
                                           2-14

-------
                                                            2.1 GENERAL METHODOLOGY
                 !                        . T
the Center for Clean Products and Clean Technologies with funding from the EPA Office of
Research and Development and Saturn Corporation.  The software tool was developed to store
and organize life-cycle inventory data and to calculate life-cycle impacts for a product profile.
Written using Microsoft FoxPro programming software, the tool is designed to allow flexibility
in conducting life-cycle design and life-cycle assessment functions. It provides the means to
organize inventory;data, investigate alternative scenarios, evaluate impacts, and assess data
quality.
       The UT Life-Cycle Design Software Tool organizes data in such a way that each process
inventory is independent. Customized "profiles" (e.g., the manufacture of a CRT or the whole
life-cycle of an LCD) can be developed by linking processes together. The tool has the flexibility
to modify or replacb any particular process within a profile to evaluate potential alternatives.  The
data provided in this study may serve as a baseline to compare alternatives or modifications to
particular processes. The models developed for the life-cycles of the CRT and LCD in this study
can remain useful as many of the individual processes in the CRT and LCD life-cycles will likely
remain constant (eJg., steel manufacturing, plastics manufacturing). Changes to specific
processes can be made to conduct analyses of current or emerging process or technology changes.
Relatively quick life-cycle analyses can be conducted on future product or process
improvements, given the baseline data already available through this study.
                • i              '                                •            •
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 (i.e.j 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). CDP data quality for each life-cycle stage is discussed
in detail in Sections 2.3 through 2.6 and summarized below.
       For the pririiary data collected in this project, participating companies reported the
method in which the data were obtained and the time period for which the data are representative.
Data from the 1997-2000 time period were sought, with the most recent data preferred.
Similarly, the time [period of secondary data and method in which the data were originally
obtained was also recorded, where available.  Secondary data cover a broader time period, with
data for most materials from the 1997 to 1998 time period, and data for most fuels from the 1983
to 1993 time period
       Anomalies ;and missing data are common hurdles in any data collecting exercise.
Anomalies are extrbme values within a given data set. Any anomaly identified during the course
of this project that is germane to project results was highlighted for the project team  and
investigated to determine its source (e.g., mis-reported values). If the anomaly could be traced to
an event inherently related to the process, it was left in the data set.  If, however, the anomaly
could not be accounted for, it was removed from the data set. Specific anomalies highlighted by
the project team are discussed in Section 2.7, Summary of Life-Cycle Inventory Results.
       We attempted to account for missing data by replacing it hierarchically. That is, if
specific primary data were missing, secondary data were used.  Where neither primary nor
secondary data were available, such as data on the percent of LCD desktop displays recycled or
remanufactured, assumptions were made and a sensitivity analysis was performed. In the cases
                                           2-15

-------
2.1 GENERAL METHODOLOGY
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.
Further, any averaged process data obtained from fewer than three companies are also aggregated
to avoid revealing individual inventory items from individual companies.

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 CDP LCA to help ensure that the following
criteria were met:

•      the methods used to carry out assessments are consistent with the EPA, SETAC, and ISO
       assessment guidelines;
• •     the methods used to carry out assessments are scientifically  and technically valid within
       the LCA framework;
•      the data used are appropriate and reasonable in relation to the goals of the study;
•      the interpretations reflect the limitations identified and the goals of the study; and
•      the study results are transparent and consistent.

       A project Core Group and Technical Work Group, both consisting of representatives
from industry, academia, and government, and EPA's DfE Work Group provided critical reviews
of the assessment. Members of these groups are listed in Appendix C, Critical Review. The Core
Group served as the project steering committee and was responsible for approving all major
scoping assumptions and decisions. The Technical Work Group and the DfE Work Group
provided technical guidance and reviews of all major project deliverables including the final
LCA report.
       In addition to the critical review process, primary data collected were double-checked
with the original source to ensure that their data are presented accurately.  Additional details on
the data verification process for primary data are presented in Sections 2.3 and 2.5.
                                          2T16

-------
                                2.2 MATERIALS EXTRACTION & MATERIALS PROCESSING
2.2    MATERIALS EXTRACTION ANf) MATERIALS PROCESSING (UPSTREAM
       LIFE-CYCLE STAGES)

2.2.1  Methodology
                 i
       The inventories included in the materials extraction and materials processing (upstream)
life-cycle stages are those of the major primary materials found in the CRT or LCD monitor, as
well as major ancillary materials required to manufacture the monitors.  Inventories for electricity
generation and fuels, which may be used in several life-cycle stages (e.g., to manufacture the
products or to use the products), are also presented.
       The inventories for extraction and processing of major materials were obtained from
existing LCI databases.  Electricity generation inventories were developed for this project from
secondary sources that describe the distribution of fuels for different electric grids and the fuel
inputs and emission outputs associated with different fuel types. The methodologies for
developing these inventories are summarized below.

2.2.1.1 Upstream materials processes

       Materials fqr which upstream processes were included in the CRT and LCD life-cycles
were selected baseci on the decision rules described in Section 2.1.2, as well as the availability of
secondary data for those materials.  An attempt was made to include materials with a mass
greater than or equal to one percent-of the overall inputs to the product manufacture, or materials
that may contribute to a large amount of energy use or have environmental concern. Tables 2-4
and 2-5 in Section 2.1.2 listed the decision criteria for primary materials/components that are
included either as separate processes or as part of another process. The upstream materials
processes for which inventory data were obtained for this project are presented in Table 2-7.
       To determine which source or sources of secondary data to use, nine LCI databases were  .
evaluated against lil selection criteria in a technical memorandum presented to the project review
teams (see Appendix D). Based on this analysis the project team chose the Environmental
Information and Management Explorer (EIME) database and the Database for Environmental
Analysis and Management (DEAM), two life-cycle inventory databases developed by the
Ecobilan (Ecobalarice) Group (Ecobilan, 1999). EIME was developed by Ecobilan specifically
for electronics and the electronics industry and covers many of the materials specific to the CDP,
while the DEAM database includes materials not covered by EIME.  Combined, these databases
contain detailed inventories of materials extraction and processing activities for most of the
materials of interest in this project.
       In the Ecobilan inventory for a particular material, the functional unit is a set mass of the
material. Inputs and outputs are therefore given in terms of mass or other appropriate unit per
unit mass of material. These data were imported into the UT Life-Cycle Design Software Tool
discussed in Section 2.1.5, where they were linked to the mass of material used in the
manufacture of a display monitor to develop inventories specific to the CDP. The associated
amounts of each material were presented in Tables 2-2 and 2-3 in Section 2.1.2. How these
materials are linked to other processes in the CRT and LCD profiles  are shown in Figures 2-2
and 2-3 at the beginning of this chapter.
                                         2-17

-------
2.2 MATERIALS EXTRACTION & MATERIALS PROCESSING
       The materials inventories from Ecobilan also contain data for electricity generation, as
appropriate, and in some cases, for transportation.  Electricity generation and transportation data
that were included in the Ecobilan inventories could not be separated from the inventory data for
materials extraction and conversion processes.  Therefore, when electricity generation and
transport data were included in the Ecobilan inventories, the electricity and transport data
collected specifically for this project were not used with the upstream process data.
                                            i                                     '  •

               Table 2-7.  Materials having upstream processes included in
                                     tfeeCDPLCA
Material
CRT
LCD
METALS .. 1 .
aluminum
ferrite ;
lead
nickel alloy (invar)
steel
/
^
/
^
/
/



/
POLYMERS
acrylonitrile-butadiene styrene (ABS) i
high impact polystyrene (HIPS)
polycarbonate (PC) ;
polyethylene terphthalate (PET) \
poly(methyl methacrylate) (PMMA)
styrene-butadiene co-polymer
^
/
/


/


/
/
^
S
ANCILLARY MATERIALS
natural gas (used to represent LNG)

S
2.2.1.2 Electric grids

       Electricity is used in several processes throughout the life-cycle of the CRT and LCD
monitors, andin some instances in large amounts.  Therefore, the inventory for electricity
generation is included in the scope of this project.
       As described in Chapter 1, Section 1.4.2, the geographic boundaries of this project are
worldwide for upstream and product manufacturing processes and limited to the United States
for the use and end-of-life stages. In addition, most CRT and LCD manufacturing is done in Asia
and most product manufacturing data collected in this study were from the United States or
Japan, except for two LCD manufacturing data sets collected from Korean manufacturers.
Therefore, the inventory associated with electricity generation during manufacturing was based
on either the Japanese or U.S. electric grids, depending on the particular process or component
being manufactured.  Where data were obtained from more than one country for the same
process, only one electric grid inventory could be used for a single process. In these cases, the
Japanese electric grid was used since the majority of manufacturing data are from Japanese
companies. The inventory for electricity generated during use and EOL processing was based on
the U.S. electric grid.
                                          2-18

-------
                                  2.2 MATERIALS EXTRACTION & MATERIALS PROCESSING
       The methodology and results for ihe electricity generation inventories are detailed in
 Appendix E, which presents the electric grid technical memorandum prepared for this project.
 The inventories were developed by first compiling U.S. inventory data for each of the major
 generation categories or fuel types, including electricity generated from coal, gas, petroleum, and
 nuclear fuels. These inventories were then combined with data on net electricity generation by
 fuel type in the U.S. and Japanese electric grids (Table 2-8) to develop the country-wide electric
 generation inventories. The inputs and outputs for the electricity generation inventories are
 presented in terms of mass or radioactivity per kWh of electricity generated. Inventories were
 not included for hydroelectric and renewable energy generation categories due to the scarcity of
 data on inputs and outputs for these categories. In addition, renewables account for only a small
 fraction of total U.S. electricity generation.

                     Table 2-8.  Net electricity generation by fuel type
Fuel
Coal :
Gas
Petroleum
Nuclear ;
Hydro
Other r
Net electricity generation
United States (percent)
57
9
• • 3
20
11
<1
Japan (percent)
18
20
21
31
9
1
       Note that the Japanese grid inventory is based on the same fuel-specific inventories
developed for the U.S. grid, but uses the average distribution of fuels for the Japanese grid (EIA,
1997, EEPC, 1996)1 This introduces some uncertainty into the Japanese electric grid since
Japanese technologies, efficiencies, and pollution control techniques are likely to differ
somewhat from their U.S. counterparts. However, the U.S. fuel-specific inventories were used to
conserve project resources, rather than expending considerable effort on collecting inventory data
from Japanese utilities.
       The electricity generation inventories presented in Appendix E are shown as separate
process inventories' However, in the overall analysis, they are linked to the manufacturing, use
and EOL life-cycle stages, as appropriate. That is, where electricity is used in a process in any of
those life-cycle stages, the inputs and outputs from generating the amount of electricity needed is
allocated to that process. Note that the U.S.  and Japanese electricity generation inventories
developed for this project are not linked to the  upstream life-cycle stages for materials used to
manufacture CRT apd LCD desktop computer  displays. Electricity generation data were already
included in the upstream material inventories received from Ecobilan.
                                          2-19

-------
2.2 MATERIALS EXTRACTION & MATERIALS PROCESSING	

2.2.1.3 Fuels                                :

Several fuels are used in manufacturing and end-of-life processes during.the life-cycle of the
CRT and LCD monitors, and in some instances in large amounts.1 Therefore, fuel production
inventories are included in the scope of this project These inventories are included in the life-
cycle stage in which the fuels are actually consumed (e.g., product manufacturing or end-of-life)
instead of in the upstream (materials processing) life-cycle stage. The following fuel inventories
are included in both the CRT and LCD LCIs:   '

•      natural gas (also used to represent LNG),
•      liquified petroleum gas (LPG),
       fuel oil #2 (distillate),                 !
•      fuel oil #6 (residual), and
•      fuel oil #4 (average of residual and distillate).
                                            i

       Fuel inventories were obtained from Ecobilan. In the Ecobilan inventories, the functional
unit is a set mass of the material, with inputs and outputs given in terms of mass or other
appropriate unit per mass of material (product) produced. These data were imported into the UT
Life-Cycle Design software tool where they were linked to the mass of fuel used in different
processes in the life-cycle of a display. How the fuel processes are linked to other processes in
the CRT and LCD profiles was shown in Figures 2-2 and 2-3.

2.2.2  Data Sources and Data Quality

2.2.2,1 Upstream material and fuel processes

       Table 2-9 summarizes data source and data quality information for the data received from
Ecobilan, which are all secondary data for the purposes of the CDP.  In addition to information
about CDP data quality indicators (e.g., original  source of data, year of data, method in which
data were obtained, and geographic boundaries), the table lists whether or not electricity
generation or transport data were included in the inventories. This information is important
because: (1) electricity generation and transportation data are not from the same sources and
therefore are not necessarily consistent among data sets; and (2) transportation data are not
included in several of the upstream processes, and are thus a data gap for those processes.
       As revealed in Table 2-9, the Ecobilan data were derived from various sources, including
European data sources and U.S. data sources. In addition, the temporal boundaries of the data
vary, with some data being as recent as 1998 but others being from as early as 1975. Electricity
generation data are included in all of the inventories, but transportation data are only included in
six of 16 data sets. All of these factors create some inconsistencies among the data sets and
reduce the data quality when used for the purposes of the CDP.  However, this is a common
difficulty with LCA, which often uses data from secondary sources for upstream processes to
avoid the tremendous amount of time and resources required to collect all the needed data.
         Fuels are also used in the materials processing life-cycle stage, but fuel production processes should
 already be accounted for in the materials inventories obtained from Ecobilan.
                                           __         ,                    -    ,

-------
                                          2.2  MATERIALS EXTRACTION & MATERIALS PROCESSING
Table 2-9. Data sources and data quality for the Ecobilan inventories a
Material
i
Electricity
generation
included?
Transport
included?
Year of
data
METALS
aluminum
ferrite
lead
nickel-alloy (invar) c
steel
Y
Y
Y
Y
Y
-
-
-
Y (nickel)
-
—
— •
Unknown
1991 (nickel)
1975-1990 d
POLYMERS
acrylonitrile-butadiene styrene (ABS)
high impact polystyrene (HIPS)
polycarbonate (PC) ,
polyethylene terphthalate (PET)
poly(methyl methacrylate) (PMMA)
styrene-butadiene co-polymer f
Y
Y
Y
Y
Y
Y
ye
ye
Ye
Ye
ye
Ye
1997
1997
1997
1998
1997
1997
FUELS
natural gas j
liquified petroleum gas (LPG)
fuel oil #2
fuel oil #6 ,
fuel oil #4 B ;
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
1987-98
1983-93
1983-93
1983-93
1983-93
Original source b

ETH
not provided
ETH
ETH
FOEFL, others d

Boustead
Boustead
Boustead
Boustead
Boustead
Boustead

six sources cited
seven sources cited
seven sources cited
seven sources cited
seven sources cited
 * • j-wy uivJ.u.uX'U AH iiiv^ULUl^r.
- : not included in inventory.
0 In general, the Ecobilan inventories provide descriptions of data quality but often do not report how data were collected (e.g.,
measured, estimated, etc,). Therefore, information on the data collection method is not presented here.
b Sources: ETH (Eidgen0ssische Technische Hochschule): Data from the ETH (Swiss Federal Institute of Technology)
(Ecobilan, 1999).  FOEI^L (Swiss Federal Office of Environment, Forests and Landscape):  Data from the Eco-inventory of
Packaging published by the Swiss FOEFL; FOEFL is also known as BUWAL, the acronym in German (Ecobilan, 1999).
Boustead: LCI database developed by Boustead Consulting (Ecobilan, 1999).
  In general, the geographic boundaries for different sources are as follows: (1) ETH and FOEFL data are from Europe; (2)
Boustead data are from Europe and/or the United States; and (3) miscellaneous sources may be European or U.S. data.  '
c The invar inventory is a combination of 36% of the nickel inventory and 64% of the ferrite inventory.
d The steel inventory was originally provided by Ecobilan without detailed documentation; however, DEAM data that were
received later have inventories for several steel production processes. The sources listed here are for the DEAM data.
" Boustead addresses transportation; however, the extent to which it is included in a particular process inventory is uncertain.
  The styrene-butadiene process is the 50/50 average of the styrene and butadiene processes.
g The fuel oil #4 processes the 50/50 average of the fuel oil #2 and fuel oil #6 processes.
                                                     2-21

-------
2.2 MATERIALS EXTRACTION & MATERIALS PROCESSING
2.2.2.2 Electric grids

       Several sources of data were consulted to generate the fuel-specific inventories that were
used to create the overall electricity generation inventories. Table 2-10 summarizes the data
sources and some of the data quality indicators for these inventories. Appendix E discusses the
data sources and data quality in detail.          \
       As shown in Table 2-10, the electricity generation data were obtained primarily from
secondary sources and include data from the mid-1990s as well as data from an unknown time
frame. Most are based on measured data collected by the original source, although some are
estimated or the data collection method is unknown. Finally, most of the fuel-specific
inventories are based on U.S. data, indicating these data are probably less representative and thus
of lower quality when applied to the Japanese electric grid.

               Table 2-10. Data sources and data quality indicators for the
                               electric generation inventories
Type of data
Net electricity
generation by fuel
Primary inputs
(Fuel)
Ancillary inputs
Air emissions
Water releases
Radioactive air and
water releases
Solid wastes
Source
U.S. Energy Information
Administration (EIA)
EIA
EIA, California Energy
Commission, utility contacts
Primarily AP-42 plus Other
sources b
Oak Ridge National
Laboratory (ORNL) c
ORNLd
ORNLe
Year of publication
: 1997
1997
Varies
Varies, but mostly
AP-42 data from 1995
1994
1995
1994
Data collection
method
Measured
Measured
Varies3
Varies2
Unknown
Measured
Unknown
Geographic
boundaries
U.S. and
Japan
U.S.
U.S.
U.S.
U.S. .
U.S.
U.S.
  Includes emission factors from measured and estimated data, plus data where data collection methods were not reported.
 b AP-42 is the U.S. EPA's compilation of Air Pollutant Emissions Factors (EPA, 1996).
 c Coal-fired water release data from ORNL report addressing the externalities of coal fuel cycles (ORNL, 1994).
 d From ORNL report addressing the externalities of nuclear fuel cycles (ORNL, 1995).
 • Coal-fired solid waste data from ORNL report addressing the| externalities of coal fuel cycles (ORNL, 1994); radioactive solid
 waste data from ORNL report addressing the externalities of nuclear fuel cycles (ORNL, 1995).
 2.23  Limitations And Uncertainties

       The limitations and uncertainties associated with the upstream materials, fuels, and
 electricity generation inventories are primarily due to the fact that these inventories were derived
 from secondary sources and thus are not tailored to the specific goals and boundaries of the CDP.
 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 or of
 industry practices in the geographic and temporal boundaries defined for the CDP (see
                                            2-22

-------
                                 2.2 MATERIALS EXTRACTION & MATERIALS PROCESSING
                 i
 Section 1.4).  These are common limitations; arid uncertainties of 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 prpduct life cycle.  Despite these limitations and uncertainties inherent in LCA
 methodology itself LCA remains useful and, indeed is increasingly used by industry,
 governments, and other stakeholders as part of a comprehensive decision-making process or to
 understand broad or general environmental trade-offs.

 2.2.3.1  Upstream material and fuel processes

       Because they are derived from secondary sources, the upstream materials and fuels
 inventories used in the CDP do not precisely meet the geographic and temporal boundaries
 outlined for the CDP. Some data are from Europe, some from the United States, and some a
 combination of both. The manufacturing data for this project were collected from companies in
 the United States, Japan, and Korea  (see Section 2.3),  and the available upstream data may not
 represent the exact Jocation or type of processing represented in the upstream data inventories.
 This is a limitation to using secondary data; however, the upstream data are only one portion of
 the overall inventory of the product systems being evaluated, and the project partners chose to
 focus on collecting primary data for  the product manufacturing life-cycle stage, since those data
 had not been previously compiled.
       Another limitation of the upstream inventory data is the lack of transportation data for
 some processes. 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 material tends to have high environmental impacts. However, the
 original data source's used in the Ecobilan inventories (see Table 2-9) are among the most used
LCI databases in the world (Ecobilan, 1999), which suggests the lack of transportation data for
upstream processesiis not unique to the CDP LCI, but a common limitation of other LCIs as well.

2.2.3.2 Electricity generation data
                • i
       The limitations to the electricity generation inventory data are provided in Appendix E,
Section 6. As another limitation, the Japanese grid was chosen when manufacturing data were
from more than one country. Appendix E also describes how the U.S. fuel-specific inventories
are applied to the Japanese grid, although technologies, efficiencies, etc. used in Japan are likely
to differ from those in the United States. Furthermore, U.S. fuel-specific inventories were
derived from secondary sources which did not necessarily meet our temporal boundaries, but did
meet geographic boundaries  for the U.S. inventory.
                                         2-23

-------
2.3 PRODUCT MANUFACTURING               	.	

2.3    PRODUCT MANUFACTURING

2.3.1   Methodology

2.3.1.1 Identification of processes and manufacturers

       Through literature research and contacts with industry experts, the manufacturing
processes and the component parts of a CRT and an LCD computer monitor were identified. The
major manufacturing processes and components, in terms of resources used, and potential
importance to environmental impacts, were selected for inclusion in our primary data collection
effort.
       Once the components and processes were chosen, companies who might supply
manufacturing data for the project needed to be identified. In order to identify those
manufacturers, the DfE project's Core and Technical Work Groups were consulted.  These
groups consist of parties interested in the project results and willing to provide technical
assistance throughout the project, including identifying contacts in manufacturing facilities..
Manufacturing of CRTs, LCDs, and their component parts is done all over the world. Some
manufacturers are in the United States; however, most manufacturers of desktop computer
monitors and their components are in Asia. Where available, U.S. industry partners provided
contacts at U.S.  as well as some Japanese manufacturing facilities and questionnaires were sent
to those contacts.                           !
       To assist in the collection of data in Asia, UT subcontracted with the Asian Technology
Information Program (ATIP) to identify company contacts, and to distribute and collect
questionnaires from Asian manufacturers. ATIP acted as a liaison between UT and Japanese and
Korean companies that participated in the study.
       Participation in the study was completely voluntary. Companies were provided with the
goals of the study and the potential benefits of their participation. Once a company chose to
participate, they were sent data collection questionnaires to complete information about their
manufacturing process and to provide their inventory of process inputs and outputs.  A copy of
the manufacturing data collection questionnaire Ithat was developed for and used in this study is
provided in Appendix F.
       The manufacturing processes for which primary data were collected are listed below. In
parenthesis are the number of individual data sets collected for each process:

CRT monitor:
•      CRT monitor assembly (3)
•      CRT (tube) manufacturing (3)
•      CRT leaded glass manufacturing (3)
                                         2-24

-------
                                                         2.3 PRODUCT MANUFACTURING
 LCD monitor:                        •,   3          ^  -••
 •      LCD monitor assembly (2)
 •      LCD paneliand module manufacturing (7)
 •      LCD glass manufacturing (I)2                               '
 •      color filter patterning on front glass (1)
 •      liquid crystal manufacturing (2)
 •      polarizer manufacturing (1)
       backlight unit assembly (3)
       backlight light guide manufacturing (1)
 •      cold cathode fluorescent lamp (CCFL) manufacturing (1)

 How these processes are linked to one another in the CRT and LCD life-cycles is presented in
 Figures 2-2 and 2-3, respectively. The companies that provided data for the foregoing processes
 are listed in Table 2-11.

            Table 2-11. Companies that provided primary manufacturing data
                               for the CRT and/or LCD
Company
American Video Glass Company
T. Chatam and Co., Ltd.
Chisso Corporation ; ,
Eizo Nanao Corporation
Hanson Electric Co., Ltd.
Hoshiden and Philips Display
Corporation
Hyundai Electronics Industries Co.,
Ltd.
liyama Electric Co., Ltd.
Matsushita Electric Industrial Co.,
Ltd.
Merck Japan Ltd.
Mitsubishi Electronic Co., Ltd.
Technology
CRT
LCD
LCD
CRT, LCD
LCD
LCD
LCD
CRT, LCD
LCD
LCD
LCD
Company
Nippon Denyo Co., Ltd.
Nippon Electric Glass Co., Ltd.
Polaroid Corporation
Samsung Electronics
Sharp Corporation
Sony Corporation (Japan)
Sony Electronics Inc. (U.S.)
Stanley Electric Co., Ltd.
Techneglas
Toppan Printing Co., Ltd.
Toshiba Display Technology Co., Ltd.
Technology
LCD
CRT
LCD
LCD
LCD
CRT
CRT
LCD
CRT
LCD
CRT, LCD
       Another process that was included in the CRT manufacturing stage data was frit
manufacturing. We were not able to obtain primary data for this process; therefore, secondary
data were collected, from EPA documentation and personal contacts (see Appendix G). An
inventory for printed wiring boards (PWBs) was also developed and included in the
manufacturing stag£ analysis. The PWB inventory is based on manufacturing of the electronic
boards, and does not include the components on the PWBs. The PWB data were obtained from
an industry representative who was able to provide general data not necessarily from one facility,
       2          ,
        LCD glass npnufacturing data were derived from the three sets of CRT leaded glass manufacturing data
(modified to remove lead from the inventory).
                                         2-25

-------
2.3 PRODUCT MANUFACTURING
but from a combination of facilities, based on his experience (Sharp, 2000). More details about
PWB data collection are provided in Appendix G.
       The manufacture of some components were not included in the scope of this study
because they were either deemed to be of less significance to the overall product inventories, or
data could not be obtained. However, all components were included as part of the final
assembled monitor even when individual manufacturing inventories were not. For the CRT, the
manufacture of the electron gun, deflection yoke^ and phosphors were not included as separate
processes, and for the LCD, the transistor metals/materials, spacers, drivers/driver ICs, and color
filters were not included.

2.3.1.2 Data collection questionnaires

       Data collection questionnaires were developed by the UT research team and approved by
the Technical Work Group to most efficiently collect inventory data needed for the LCA.
Appendix F provides a copy of the questionnaire given to product and component manufacturers.
 The data that were collected include 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 were also reported.  Details of the transportation analysis for this project
are presented in Section 2.6.
     .  A total of 27 product manufacturing questionnaires were collected for 11 different
processes. The corresponding countries and the number of data sets from each country are listed
in Table 2-12.

           Table 2-12.  Location of companies and number of process data sets
Process
CRT monitor assembly
CRT (tube) manufacturing
CRT leaded glass manufacturing
CRT frit manufacturing
LCD monitor assembly ;
LCD panel and module manufacturing .
LCD - glass manufacturing
LCD - color filter patterning on front glass
LCD - liquid crystal manufacturing !
LCD - polarizer manufacturing
LCD - backlight unit assembly
LCD - backlight light guide manufacturing
LCD - cold cathode fluorescent lamp (CCFL) manufacturing
PWB manufacturing (for CRT and LCD monitors)
Country of origin of data
(# of data sets)
Japan (2), U.S. (1)
Japan (2), U.S. (1)
Japan (1), U.S. (2)
generic secondary data from the U.S.
Japan (2)
Japan (5), Korea (2)
Japan and U.S. (1)*
Japan (1)
Japan (2)
Japan (1)
Japan (3)
Japan (1)
Japan (1)
generic secondary data from the U.S.
 * Average of three data sets for CRT leaded glass manufacturing modified to remove lead from the inventory.
                                           2-26

-------
                                                          23 PRODUCT MANUFACTURING
 2.3.1.3 Allocation                        ~          f   ,

       Data provided by manufacturers may need to be allocated to the products of interest (i.e.,
 17" CRT or 15" LCD) in three situations:

 •      data are provided for more than the defined functional unit;
 •      data are provided on a rate basis instead of per functional unit (product); and
 •      data are provided for monitors/components of more than one size (i.e., not only 17" CRTs
       or 15" LCDs).
                  i
 In some cases, allocation was not required, as the inventory data collected were for one unit of
 the product, defined as the functional unit. The three cases where allocation was necessary are
 briefly described below.
       In the first c^se, simple scaling was required when data were provided for all 17" CRT or
 15" LCD monitors produced at a plant, as opposed to only one monitor.  This simply requires
 dividing the inventory mass by the number of monitors produced.
       In the second case, data were provided over a certain amount of time. The inventory data
 were then scaled to represent the functional unit.  For example, if it was reported that x kilograms
 of a material are us^dper year to produce one 17" or 15" monitor, and y number of products are
 produced per year, then the amount of that material per functional unit is x/y.
       In the third situation, allocation was also necessary for a company that manufactured
 more than just the product or component of interest for this study. For example, a monitor
 manufacturer may ^assemble various sized monitors, in addition to 15" LCDs or 17" CRTs.
 Therefore it was necessary to allocate the process inventory to only our product of interest. We
 used the difference in mass between the product of interest and other co-products and the
 difference in the number of each product produced to allocate the inventory to the functional unit.

 2.3.1.4 Aggregating manufacturing data

       After one set of data from one company is allocated to one monitor, processes for which
 we collected more than one company's data were averaged together. Once the inventory data for
 a process were averaged, the electricity consumption from that process was linked to the
 appropriate electric grid inventory (i.e., Japanese or U.S.).  All the manufacturing processes were
 linked to the Japanese grid, with the exception of frit and PWB manufacturing, both of which
 were based on data collected in the United States.  Where process data were represented by
 companies in more than one country, the countries in which the majority of facilities were
 located was used for the basis of which electric grid to use. An exception is the polarizer data,
 which was from the 'United States, but the manufacturing was only a pilot plant and not
producing a product! in the open market. Therefore, it was assumed that polarizer manufacturing,
 as with most other L!CD components, was done in Japan. Once each manufacturing process
inventory for each monitor type was complete (i.e., an averaged inventory with an associated
 electric grid), each was aggregated with the rest of the manufacturing stage processes to comprise
the inventory for the manufacturing stage for a monitor. This manufacturing stage inventory was
then combined with the other life-cycle stages to represent the full LCI for each monitor type.
                                          2-27

-------
2.3 PRODUCT MANUFACTURING
2.3.2  Data Sources and Data Quality

       While manufacturers worldwide were offered the opportunity to participate in this
project, only manufacturing data from Japan, Korea, and the United States were collected. The
quality of the data can be evaluated against two factors:  (1) the date of the data; and (2) the type
of data (i.e., measured, calculated or estimated). An understanding of how data were collected
and data verification steps should also be considered when evaluating the data quality.  The data
collection phase of this project began in 1997 and extended through 2000. Some processes are
more sensitive to production dates than others. Most processes included in this analysis are
mature technologies and are not expected to differ significantly between the years 1997 and
2000.  However, an exception is LCD panel manufacturing, which is an evolving and rapidly
advancing process and has seen changes between these years. The countries and dates from
which data for each process were obtained are presented in Table 2-13. For the LCD panel and
module manufacturing process, most data were from 1998 and 1999.
       In the data collection questionnaires, companies identified whether the quantity of each
inventory item was a measured, calculated, or estimated value. These identifiers were referred to
as the "data quality indicator" (DQI) in the manufacturing questionnaire.  The breakdown of
DQIs for the inventory items in the CRT and LCD processes are presented in Tables 2-14 and 2-
15, respectively. The last line in each table shows overall averages weighted by the number of
inventory items in each data set. For the CRT, 43% of the data were measured, 34% calculated,
13% estimated, and 10% were not classified. For the LCD, a similar distribution shows 33%
measured, 30% calculated, 23% estimated, and. 14% not classified.
                                           i
                                           i
                    Table 2-13. Applicable years of primary data sets
Process
CRT monitor assembly
CRT (tube) manufacturing
CRT leaded glass manufacturing
LCD monitor assembly
LCD panel and module manufacturing
LCD - color filter patterning on front glass
LCD - glass manufacturing*
LCD - liquid crystal manufacturing
LCD - polarizer manufacturing
LCD - backlight unit assembly
LCD - backlight light guide manufacturing
LCD - cold cathode fluorescent lamp (CCEL)
manufacturing
# of data sets
3
3
3
: 2
7
1
1
2
1
3
1
: i
i1 '
Dates of inventory for each data set
1997, 1998-9, 1999
1997, 1998, 1998
1998, 2000, 2000
1998, 1999
1997-8, 1998, 1998, 1998-9, 1999, 1999,
1999-2000
1998
1998-2000
1998, 1998
1997
1998,1999,1999
1999
1998-9
 * Primary data, but developed from the CRT glass manufacturing data.
                                           2-28

-------
                                                         2.3 PRODUCT MANUFACTURING
       Data quality can also be reviewed in terms of data collection methods. Much effort was
given to collecting primary data from manufacturers in this study. Questionnaires were sent out
and follow-up communication was conducted to verify data gaps or discrepancies. Twelve
companies were visited directly to clarify data and telephone or electronic communication
followed-up those land the remaining companies that were providing data for the study. Where
data could not be confirmed, additional literature research and discussions with other industry
experts were conducted.

          Table 2-14. Data quality indicator percentages for the
Process ;
CRT monitor assembly .
Data set 1 (22 inventory items)
Data set 2 (11 inventory items)
Data set 3 (33 inventory items)
total inventory items = 66
CRT (tube) manufacturing
Data set 1 (69 inventory items)
Data set 2 (51 inventory items)
Data set 3 (83 inventory items)
total inventory items = 203
CRT leaded glass manufacturing
Data set 1. (43 inventory items)
Data set 2 (45 inventory items)
Data set 3 (2 inventory items)
total inventory items = 90
Overall weighted average for CRT
(359 items) |
% of inventory items that are:
Measured
9%
9%
3%
wt. avg = 6%
91%
45%
21%
wt. avg = 51%
9%
91%
100%
wt. avg = 52%
43%
Calculated
64%
83%
Q%
wt. avg = 35%
1%
55%
78%
wt. avg = 46%
3%
7%
0%
wt avg = 5%
34%
Estimated
27%
0%
Q%
wt. avg = 9%
4%
0%
1%
wt. avg = 2%
86%
0%
0%
wt. avg = 41%
13%
Not reported
0%
9%
97%
wt. avg - 50%
3%
0%
Q%
wt. avg - 1%
2%
2%
0%
wt. avg - 2%
10%
                                         2-29

-------
23 PRODUCT MANUFACTURING
          Table 2-15.  Data quality indicator percentages for the LCD processes

Process
LCD monitor assembly
Data set 1 (36 inventory items)
Data set 2 (10 inventory items)
total inventory items = 46
LCD panel and module manufacturing
Data set 1 (71 inventory items)
Data set 2 (39 inventory items)
Data set 3 (45 inventory items)
Data set 4 (139 inventory items)
Data set 5 (53 inventory items)
Data set 6 (86 inventory items)
Data set 7 (32 inventory items)
total inventory items = 465
LCD - glass manufacturing*
(83 inventory items)
LCD - color filter patterning on front glass
(29 inventory items)
LCD - liquid crystal manufacturing
Data set 1 (41 inventory items)
Data set 2 (6 inventory items)
total inventory items = 47
LCD - polarizer manufacturing
(30 inventory items)
LCD - backlight unit assembly
Data set 1 (20 inventory items)
Data set 2 (12 inventory items)
Data set 3 (12 inventory items)
total inventory items = 44
LCD - backlight light guide manufacturing
(5 inventory items)
LCD - cold cathode fluorescent lamp (CCFL)
manufacturing (36 inventory items)
Overall average for LCD (785 items)
% of inventory items that are:
Measured
5%
10%
wt. avg = 6%
80%
0%
40%
55%
43%
0%
0%
wt. avg = 37%
0%
97<&
22%
0%
wt. avg = 19%
57%
0%
50%
0%
wt. avg = 14%
i
404
1
58%
33%
Calculated
78%
80%
wt. avg = 78%
17%
100%
32%
38%
0%
0%
0%
wt. avg = 26%
0%
0%
64%
0%
wt. avg = 56%
17%
90%
0%
92%
wt. avg = 66%
40%
39%
30%
Estimated
17%
10%
wt. avg = 16%
0%
0%
17%
0%
36%
24%
97%
wt. avg = 17%
100%
3%
7%
0%
wt. avg = 6%
20%
0%
8%
8%
wt. avg = 4%
20%
0%
23%
Not reported
0%
0%.
wt. avg = 0%
3%
0%
11%
7%
21%
16%
•3%
wt. avg == 20%
0%
0%
7%
100%
wt. avg= 19%
6%
10%
42%
Q%
wt avg = 16%
0%
3%
14%
 * Data based on CRT leaded glass manufacturing data.
                                          2-30

-------
                                                        2.3 PRODUCT MANUFACTURING
 2.3.3  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.

2.3.3.1 Product system boundary uncertainties
                  i
       The scope of the analysis included the major monitor components; however, it excluded
certain components, such as column and row driver ICs for the LCD and the electron gun for the
CRT. The components that were thought to possibly have an effect on the inventory were the
column and row drivers, as 1C manufacturing is known to be energy intensive and use various
process chemicals.  Based on some back-of-the-envelope calculations, the exclusion of the
manufacturing of thp column and row driver ICs is not expected to have a large impact on the
inventory or impact results due to the small size of the drivers. Therefore, it is assumed that the
exclusion of the column and row drivers will not have a significant impact on the study results.
       The scope of the analysis in this study was also dependent on whether companies were
willing to provide d^ta. LCD glass manufacturing data from a primary source are not included in
the inventory because no companies were willing to supply the data. For example, one
manufacturer chose hot to provide data because LCD glass manufacturing technology is still
developing and is expected to improve from current low yields and high waste generation.
However, because glass is an important component by weight of the LCD, we chose to modify
the CRT glass manufacturing data to represent LCD glass manufacturing.
       Both CRT arid LCD glass are considered to be "specialty glasses" in the glass industry
and a limited number of companies produce these products. Consequently, there are limited
public data available on the production of these glasses. (One major difference between the CRT
glass and the LCD glass is that CRT glass contains lead oxide while LCD glass does not.)
Therefore, the primary  data collected for this study for CRT glass manufacturing was modified
by removing inputs  and outputs containing lead, and used to represent LCD glass manufacturing.
The remaining inputs and outputs were assumed to be the same per kilogram of glass produced.
Further research was conducted to confirm whether this was a valid assumption for the energy
used in production.  ^Consultation with experts in the field revealed differing estimates between
energy used to produce a kilogram of CRT glass and a kilogram  of LCD glass. Estimates ranged
from an equal amount of energy per kilogram for CRT and LCD glass production, to twice as
much energy per kilogram of LCD glass compared to CRT glass. With an assumption of equal
or greater energy use for the LCD (call that quantity of energy X), the proportion of that energy
(X) that is electrical ;energy and fuel energy was assumed to be the same as that given in the
primary data for CRT glass production (if the portion of energy X for the CRT was 30%
electrical and 70% fuel energy, those same proportions were used for the LCD's breakdown of
energy X).         i
                                         2-31

-------
2.3 PRODUCT MANUFACTURING	

       Uncertainty in the differences between energy used for CRT versus LCD glass production
are related to production yield and melting point. The melting point of LCD glass is greater than
that of CRT glass; however, the difference in the production yield is uncertain compared to the
difference in melting point. The production yield for CRT and LCD depends on whether one
considers the surface area or the volume of the glass, LCD glass is a flat glass product and
excess glass is cut off the ends to obtain the final product. CRT glass manufacturing drops
molten glass into a mold and glass is not cut off of a flat piece of glass as in LCD production, but
excess glass may be produced during the molding process. Another factor in the difference in
energy used for CRT and LCD glass is that the LCD glass must meet high specification standards
for use as a substrate for the transistors to be patterned on the glass. Additional finishing steps
are required and the process is conducted in clean rooms  (described in Section 1.3.3.2, LCD
manufacturing). The assumption that the same amount of energy is used for CRT glass and LCD
glass production takes into account each of these factors. The baseline analysis in this study
assumes the energy use per kilogram of LCD glass is equivalent to that of CRT  glass. The
uncertainty associated with the glass manufacturing data is a limitation to the manufacturing
inventory data set. Additional limitations from the glass manufacturing data are discussed below
with other "data uncertainties" in Section 2.3.3.3.
                                            i
2.3.3.2 Data collection process uncertainties   •
                                            i
       Limitations and uncertainties related to the data collection process include the fact that
companies were self-selected, which could lead to selection bias (e.g., those companies that: are
more advanced in terms of environmental protection might be more willing to supply data than
those that are less progressive). Also, the data were supplied by companies whose vested interest
is to have their product look more desirable, which could result in biased data being provided.
However, multiple sets of data were obtained for this project, where possible, so that average
processes could be developed in an attempt to avoid biased data.  The peer review process and
employment of the Core and Technical Work Groups as reviewers in this project is intended to
help reduce or identify any such bias.  Further, several companies were visited and contacted for
verification of data.                           \
       Other data collection-process limitations resulted from the difficulty in obtaining and
verifying data over long distances (i.e., Japan and Korea to United States) as well as from the
language barrier. The use of ATIP as the Asian Liaison aided in reducing this limitation;
however, there were still language barriers that had to be overcome with ATIP,  as well as
through direct communication with several companies.

2.3.3.3 Data uncertainties                    ;
                                            i
       Additional limitations to the manufacturing stage inventory are related to the data
themselves. Several attempts were made to verify or eliminate outliers in the data; however,
uncertainty in some data remained due to large data ranges and outliers. Specific data with the
greatest uncertainty include:  (1) CRT glass manufacturing energy inputs (mentioned above in
Section 2.3.3.1); (2) the distribution of fuel/electricity inputs for LCD module manufacturing;
and (3) the use of a large amount of liquified natural gas  (LNG) as an "ancillary material" and not
a fuel.
                                          2-32

-------
                                                          2.3 PRODUCT MANUFACTURING
       In addition to the uncertainty in the difference between energy used to manufacture CRT
 glass and LCD glass, the energy reported to produce a kilogram of CRT glass varied greatly
 between the data sets. Consultation with experts in the glass industry confirmed that the average
 energy consumption derived from the primary data sets, although it appeared to be high, could be
 possible. The total energy inputs per kilogram of glass from the primary data sets used in the
 analysis, ranged over a factor of approximately 150 (i.e., the largest total energy value in the data
 was about 150 times that of the smallest value). Due to this large discrepancy, the glass energy
 data is the subject of sensitivity analyses in this study. The high energy use values were mostly a
 function of liquified petroleum gas (LPG) used as a fuel.
       Other data for which large ranges were found, and which could be important to the
 results, are the energy use from LCD panel/module manufacturing. Energy data were provided
 by six of the seven ^companies supplying LCD panel/module data.  The percent of energy from
 electricity ranged from approximately 3% to 87%. Three of the companies had the electrical
 energy component contributing greater than 50% of the total energy use, and the other three
 companies listed other fuels [e.g., LPG, LNG] as contributing greater than 50% to the total
 energy use. Another large discrepancy was the total energy use for panel/module manufacturing.
 Four of the six companies had energy use per panel between approximately 440 MJ/panel and
 940 MJ/panel, while the two remaining companies had approximately 4,100 and 7,000 MJ/panel.
 The average per panel was approximately 2,270 MJ/panel and the standard deviation was about
 2,910 MJ/panel.  ;
       Given the Wide variability in the data and large standard deviation, CDP researchers
 evaluated the data for outliers by breaking the total energy data points into quartile ranges.
 Minor outliers are then those within a certain range of multipliers beyond the middle 50% of the
 distribution. That is, the interquartile range (IQR) (i.e., the range of values representing the
 middle 50%) multiplied by 1.5 is the lower bound of the minor outlier and the IQR multiplied by
 three is the upper bound. Anything beyond the IQR times three is a major outlier. Using this
 approach, one data set was found to be a minor outlier and another was found to be a major
 outlier.  These outliers were excluded from the averages used in the baseline analysis, but
 included in the averages used in a sensitivity analysis (see Section 2.7.3.3).
       Finally, the average amount of LNG used as an ancillary material (not a fuel) in LCD
panel/module manufacturing was reported as 194 kg per functional unit. This data point
remained in the inventory data set for LCD manufacturing, and was assumed to indeed be an
 ancillary material, and not a fuel. LNG was also reported as a fuel as a separate input
 (approximately 3.22 kg/functional unit). Keeping the LNG  ancillary material in the inventory
 will not affect the energy impact results, since LNG used as an ancillary material is only linked to
the production of that material, and not to the use of it as a fuel. It will, however, affect upstream
impacts from the production of the material.  Note that the natural gas process was used as a
 surrogate for the production of LNG, as inventory data was not available for the latter.
                                          2-33

-------
2.4 PRODUCT USE
2.4    PRODUCT USE                     j
                                           i
2.4.1   Methodology

       The methods for developing the use stage inventory are presented in Appendix H and
summarized here. CRTs and LCDs use different' mechanisms to produce images on screen,
which result in different energy use rates. These energy use rates (e.g., kW) can be combined
with the time a desktop monitor is on during its lifespan (hours/life) to calculate the total quantity
of electrical energy consumed during the use life-cycle stage (e.g., kWh/life). In this project, two
lifespan scenarios are considered:

•      Effective life - the actual amount of time a monitor is used, by one or multiple users,
       before it is disposed of, recycled, or re-manufactured. Reuse of a monitor by a
       subsequent user is considered part of its effective life. Recycling, on the other hand, is
       the reuse of parts or materials that require additional processing after disassembly, and it
       is not considered part of the use stage.   |
•      Manufactured life - the amount of time either an entire monitor or a single component
       will last before reaching a point where thfc equipment no longer functions, independent of
       user choices.

       These two scenarios are considered in this project in order to account for potential
differences between how consumers currently use the equipment and how consumers could use
the equipment. Currently, consumers often replace monitors before they physically break down.
This behavior results in a lifespan that is not solely dependent on the monitor technology itself.
The manufactured life, on the other hand, is based on the technology and represents how
consumers could potentially use the equipment. If the lifespans are significantly different, the
difference could have a large impact on how the use stage compares to the  other life-cycle stages
in this study. The baseline analyses in this project use the effective life scenario and the
manufactured life will be part of the sensitivity analyses.

2.4.1.1 Energy use rate

       Most desktop monitors manufactured today are built to use several  different power
consumption modes during normal operation.  There are often up to four different power
consumption modes that can be used by a monitor in going from a state of  active use to a state of
almost complete shut-down. These four modes, from greatest power consumption to least, are
typically entitled "full-on" or active use, "standby," "suspend," and "active-off." For this report,
manufacturers' data on these power modes were| collected from company contacts and Internet
sites for 35 different 17" CRT monitors and 12 different 15" LCD monitors. The complete list of
these data is presented in Appendix H, Attachment A, Table Al.
       For the purposes of this study, the power! consumption modes have been categorized into
two modes: "full-on" and "low." The "low" power mode is an average of the three low power
modes typically provided by the manufacturers (i.e., standby, suspend, and active-off). These
three categories were averaged to create one "low" power consumption mode because hours per
use data (needed for calculations in this study) are only available for a "full-on" and a reduced
power mode. The low mode value for the CRT |s the average of the three modal averages of
                             ——,,_

-------
                                                                       2.4 PRODUCT USE
standby, suspend, and active-off. For the LCD, data on only two low-power modes (standby and
active-off) were provided by manufacturers (see Appendix H, Attachment A, Table A2), and
therefore, the low mode value is an average of those two modal averages. Table 2-16 presents
the average values for full-on and low power modes that were used for subsequent calculations in
this analysis.
                         Table 2-16. Average energy use rates
Monitor type ',
17" CRT i
15" LCD ;
Full-on power mode
[kilowatts (std. dev.)]
0.113 (0.015 SD)
0.040 (0.007 SD)
Low power mode b
[kilowatts (std. dev.)]
0.013 (0.005 SD)
0.006 (0.003 SD)
 • See Appendix H, Attachment A, Table Al for source data.
 b An average of company-reported values for standby, suspend and active-off (see Appendix H, Attachment A, Table Al).
 Note: IkW =1000 Watts = 1000 J/sec.
2.4.1.2 Effective life (baseline lifespan calculation')
                 F
       The effective life scenario attempts to model the actual quantity of hours that an average
monitor spends in e|ach of the two primary power consumption modes (full-on and a lower power
state) during its lifetime. The effective life of an average monitor is based on the following
information:
                 i

•      the proportion of computers that are used in an office environment versus a home
       environment, to account for different use rates in these two basic user environments;
•      the amount of time in a year a typical monitor spends in full-on power mode and in a
       lower power-consuming mode for both office and home environments; and
•      the number of years a typical monitor is used during its lifespan for both office and home
       environments, not including years in storage before a monitor is replaced or discarded (as
       it is not consuming power during storage).

       Under the effective life scenario, we assume there is no difference in the amount of time a
CRT or LCD monitor is operating.  That is, the hours per life for the effective life calculation is
not technology-dependent.  Therefore, the same set of hours-per-life values are used to calculate
the kWhs used per effective lifetime for a CRT and an LCD. The remainder of this section
discusses the data and methods used to calculate the hours-per-life values used in the effective
life scenario. More details are also provided in Appendix H.
                 1
Percentages of Office- and Home-Environment Users

       Home and office users of computer equipment do not follow the same use patterns. Thus,
data are needed on the percent of users in each environment to determine the use pattern of an
"average" computer monitor. It is assumed that 65% of computers are in office environments
and 35% in home environments, based on data available through the Computer Industry Almanac
(CIA, 1997) and the Energy Information Administration (EIA, 1999b) (see Appendix H, Section
2.2.2.1 for more details).
                                         2-35

-------
2.4 PRODUCT USE
       Note that an "office" environment may be a school, hospital, or other commercial
environment, and the computers they use may follow widely varying degrees of use. For;
example, computers (and thus monitors) in a school may only be used a few hours in a day, while
hospitals might operate theirs nearly constantly J For this study, it is assumed that on average,
typical office use patterns (to be presented below, and in Appendix H, Section 2.2.2.2) are
representative of all non-home environment users.

Operating Pattern (average hours in use per year)

       In order to determine the amount of electricity consumed during a monitor's effective life,
we need to know the use operating patterns for both the office and home environments. The
"operating pattern" is defined here as the number of hours per year spent in each power mode.
The average number of hours per mode per year will be the weighted average of the two  user
environments (i.e., 65% office, 35% home).
       A literature search for computer monitor operating patterns was conducted for both office
and home environments and a summary of literature reviewed is presented in Appendix H,
Attachment A, Table A3. Based on the literature (Nordman et al, 1996, Fanara, 1999, EIA,
1999b) and other assumptions presented in Appendix H, Section 2.2.2.2, we assume the number
of hours per year that office and home monitors are used in each mode are as follows:
       Office:
             Full-on power mode:
             Low power mode:
             Total:
1,095 hrs/yr
2,263 hrs/yr
3,358 hrs/yr
       Home:
             Full-on power mode:
             Low power .mode:
             Total:
  522 hrs/hr
  793 hrs/yr
1,315 hrs/yr
Average Years Per Life
       The number of years per life, multiplied by the operating patterns in hours per year (listed
above), will result in the hours per effective life: A monitor may be reused in multiple "lives"
before reaching its end-of-life.  The end-of-life is defined as the point at which the monitor is no
longer used for its intended purpose in the physical form in which it was originally manufactured.
End-of-life options include indefinite storage (in which case it is not reused after storage),
de-manufacturing, recycling, or disposal. A monitor may be stored before being reused;
however, this storage time will not affect our use calculations since no electricity is required to
operate the monitor during this storage.  After its first life as used by the original owner, a
monitor might be used by different people and with different PC systems in subsequent lives.
       For data on the number of years of use that are in a monitor's lifetime, several sources of
information were reviewed (see Appendix H).  Based on a recent study by the National Safety
Council (NSC, 1999), we assumed that a monitor is used for 4 years in its first life and 2.5 years
for its second or subsequent lives. The operating patterns (in hours/year) presented above are
                                          __                            .

-------
                                                                         2.4 PRODUCT USE
assumed to be the same for all of the 6.5 years of the total effective life. However, in the lives
subsequent to the first life, the hours per year values are reduced by the fraction of monitors
assumed to be reused. Matthews et al (1997) estimated that 45% of PCs are reused after a first
life; thus, the effective life operating pattern values in years of life after the first life are 45% of
the values in the first life.
       Lifespan estimates from the National Safety Council (NSC, 1999) were specific to CRT
monitors; however; they were not specific to LCD desktop monitors. The NSC data did contain
estimates of a "Notebook PC,"  which were two to three years for the first life and one to two
years for the remaining lives; however, we expect that desktop LCD monitors will more closely
mirror the lifetime estimates of a desktop CRT monitor than that of a notebook PC.
Consequently, it was assumed that LCD desktop monitors also spend four years in their first life
and 2.5 years in their subsequent lives. Additionally, the NSC document did not attempt to
separate those computer systems or monitors that are used in an office versus a home
environment.  Thus, it was assumed that the same years per life are realized for office and home
environments.     '

Effective Life Estimates (hours per life)

       The data presented above are summarized in Table 2-17 and used to estimate the total
hours per effective life.  The values for hours per year per power mode are assumed to be the
operating pattern throughout the first life (first four years).  In the remaining lives, the annual
operating hours decrease to 45% of the hours in operation during each year  in the first life, with
the remaining lives lasting a total of 2.5 years.  Table 2-17 also presents the total hours per
effective life per mode, based on percentage in office and home environments.  These values  are
in bold in Table 2-l|7 (4,586 and 8,961 hrs per effective life) and will be multiplied by the energy
use rates per mode (presented in Table 2-16), to calculate the total energy consumption per
effective life for each monitor type.

                              Table 2-17. Effective life values
User
environment
Office
(65%)
Home
(35%)
Weighted
average c
Power
mode
. i
Full-on
Low
Full-on
Low
Full-pn
Low
First life
(4 years)
Operating
pattern (hr/yr)
1,095
2,263
523
793
—
—
Total
(hr/4yrs)
4,380
9,052
2,092
3,172
—
—
Remaining lives
(2.5 years)
Operating
pattern (hr/yr) a
493
1,018
235
357
—
—
Total
(hrs/2.5 yrs)
1,233
2,545
588
893
—
—
Model totals b
(hr/effective life)
5,613
11,597
2,680
4,065
4,586
8,961
  	o     r •—^ f —.»««.» .»» *~r iu v*. j.4-Lut. UA.W vsfjwj.u.uug, ^scivii'i.ij.) ucuvu \jn •*•./ /v ui iUUllllwla UlaL cUC 1CLLSCU
(Matthews etal, 1997). .     .
b Modal totals calculated as [(Total for first 4 years) + (Total for remaining 2.5 years)].
° The weighted averages shown for full-on and low power modes are based on the assumption that 65% of users operate in an
office environment and 35% operate in a home environment.
                                           2-37

-------
2.4 PRODUCT USE
Effective Life Total Energy Consumption (kWh/life)

       In order to calculate the total kWhs consumed, first, the energy use rates (kW) were
multiplied by the lifespans (hours per life) for each mode and each monitor type. They were then
summed for the two power modes to obtain a total kWh/life for each monitor type (Table 2-18).

                    Table 2-18. Effective life electricity consumption
Monitor
type
17" CRT
15" LCD
Power
mode
Full-on
Low
Total
Full-on
Low
Total
Energy use rate
(kW)
0.113
0.013
—
0.040
0.006
—
EL calculated lifespan
! (hours/life)
j 4,586
8,961
I 13,547
! 4,586
; 8,961
; 13,547
EL energy consumption
(kWh/life)
518
116
634
183
54
237
2.4.1.3 Manufactured life (alternative lifespan calculation)

       Due to the uncertainty and assumptions associated with the effective life scenario, an
alternative scenario is also considered. The manufactured life is defined here as the length of
time a monitor is designed to operate effectively for the user. It is the number of hours a monitor
would function as manufactured, and is independent of user choices or actions.  One way to
estimate this manufactured life is to use the meari-time-before-failure (MTBF) specification of a
monitor or its components. The CRT MTBF specification dictates the amount of time the
display must operate before it reaches its brightness "half-life," or the ability to  produce 50% of,
its initial, maximum brightness. The MTBF value, generally provided in total hours per life of a
monitor, is what most final manufacturers or assemblers of personal computer (PC) equipment,
including monitor assemblers, typically specify fpr a component. To meet the specification,
suppliers typically calculate the MTBF (a military-based specification) based on component, data.
Suppliers' test results are usually called the "calculated" MTBF. The MTBF value also depends
on which combination of power modes are used during testing, which is referred to as the "duty
cycle" and each supplier may use a different duty cycle to test their component.
       Additionally, monitor assemblers will often perform their own testing, typically entitled
"demonstrated" MTBF.  The testing includes sequences where the monitor is "stressed" by
quickly switching back and forth from an all black picture to an all white one, or quickly
switching individual pixels either on and off or through multiple colors or black and white.
Manufacturers typically find that their demonstrated MTBF is on the order of twice as long as the
calculated MTBF (McConnaughey, 1999; Douglas, 1999). However, it should  be noted that the
demonstrated MTBF is not a real-time testing method, as the testing data is used in a complex
equation to calculate that "demonstrated" value.
       From review of the information obtained :on CRT-based monitors  (see Appendix H,
Attachment A, Table A2), it appears that the CRT (the tube) itself is the limiting component, or
the component that 99% of the time determines whether the entire monitor has reached its
end-of-life. Thus, from the limited information that was obtained on CRTs, and the limited
                                          2-38

-------
                                                                       2.4 PRODUCT USE
 confidence that can be instilled in those data, ;an average of the two ranges obtained on the
 estimated lifetime of CRTs (10,000 and 15,000 hours) was used as the CRT manufactured
 lifetime (12,500 hours) (Goldwassar, 1999; Douglas, 1999).
       For active Beatrix LCDs, the components that have the greatest potential to fail first are
 the display panel itself (including the liquid crystals and thin-film transistors), backlights, driver
 integrated circuit (1C) tabs, and other smaller components. The backlights and driver 1C tabs can
 be field-replaced, tlius their failure does not necessarily represent the end of the monitor's life.
 However, failure of the liquid crystals or transistors, which would require replacement of the
 display panel itself,; would most likely mean that the monitor cannot be cost-effectively repaired.
 The MTBFs of all these components appear to have a broad range. For example, different
 backlight manufacturers reported from as few as 15,000 hours to as many as 50,000 hours
 (Douglas, 1999; Tsuda, 1999; VP150,1999). However, it appears that those components that are
 not field-replaceable (e.g., the LCD panel) have MTBFs in the range of 40,000 to 50,000 hours
 (Tsuda, 1999; Young, 1999). Thus in this study, the amount of time an LCD monitor would
 operate during its manufactured life is assumed to be the average of the two non field-replaceable
 values, or 45,000 hours. In order for a monitor to operate for 45,000 hours, any major
 field-replaceable parts that have MTBFs less than 45,000 hours will need to be accounted for in
 this LCA project. For example, assuming the backlights last on average 32,500 hours (the
 average of the values obtained for backlights), more than one (approximately 1.4, on average)
 would be needed for every panel during its  lifetime. Therefore, in the final  CDP LCA, the
 manufacturing of extra backlights would need to be included in the inventory.
       Little information is available on the duty cycles that component manufacturers use to test
 components.  Thus,; it is assumed that the average  duty cycle used in testing components is 50%
 of the time tested in full-on mode and 50% in a lower power mode. Table 2-19 shows the values
 that are used in this study for the hours per manufactured life for the CRT and LCD. Some
 sensitivity analyses were done and presented in Socolof et al. (2000).
                  \

                          Table 2-19. Manufactured life values
Monitor
type

15" LCD
Total hours
(hours/life)
12,500
!
45,000
Mode
Full-on
Low
Full-on
Low
Duty cycle
(% time spent in each mode during testing)
50%
50%
50%
50%
Hours per mode
(hours/life)
6,250
6,250
22,500
22,500
       To calculate: the manufactured life electricity consumption (kWh/life), the energy use rate
(kW) is multiplied by the lifespan (hours/life) for each monitor in each power mode (Table 2-20).
The LCD manufactured life (45,000 hours) is 3.6 times greater than the CRT manufactured life
(12,500 hours).  In an LCA, comparisons are made based on functional equivalency.  Therefore,
if one monitor will operate for a longer period of time than another, impacts should be based on
an equivalent use. Therefore, based on equivalent use periods, 3.6  CRTs would need to be
manufactured for every LCD. This will be incorporated into the profile analysis for the
comparative manufactured life LCA.
                                         2-39

-------
2.4 PRODUCT USE
                  Table 2-20. Manufactured life electricity consumption
Monitor
type
17" CRT
15" LCD
Power
mode
Full-on
Low
Total
Full-on "
Low
Total
Energy use rate
(kW)
0.113
0.013
—
0.040
0.006
—
ML calculated lifespan
(hours/life)
6,250
6,250
12,500
22,500
22,500
45,000
ML energy consumption
(kWMife)
706
81
787
900
135
1,035
2.4.1.4 Effective life versus manufactured life
                                            i
                                            i
       For the CRT monitor, the effective life total hours are 13,547 versus the manufactured
life total of 12,500. While this suggests that a CRT can be used longer than is physically
possible, it simply reveals our low confidence in these numbers and some of their supporting
values, with less confidence in the manufactured life data than the effective life estimates. A
more complete discussion of the data quality is presented in Section 2.4.2 and Appendix H.
       Assumptions were required several times that could bias these numbers in either
direction; however, it is thought that the manufactured life estimate is most likely low based on
the other estimates for the overall CRT monitor (see Appendix H, Attachment A, Table A2).
However, there was no sound basis for assuming a lower value and thus the above hours per life
values were used. It should also be stated that while these numbers are different, they are within
an 8% error range of one another, and can be taken to be a near 1:1 ratio, indicating a .similar
potential lifespan.                            I
       For LCDs, the comparison across lifespan scenarios is more consistent with what one
would expect, with the manufactured life value of 45,000 hours per life being much greater than
the effective life value of 13,547 hours per life. ;The effective life value reflects the assumption
that a user's use habits are not technology-dependent, and would seem to reveal that LCDs are
not being used as long as they can physically be|(less than a third as long).
       The difference between the effective and manufactured lives are important when
evaluating all the life-cycle stages for a particular monitor type. If the manufactured life is
significantly greater than the effective fife, the use stage will have greater impacts, as compared
to other life-cycle stages.  Therefore, it is important to focus on the lifetime scenario that is most
realistic, while still recognizing the potential imjpacts from another feasible lifespan scenario.
       In this project, we will use the effective life as the primary basis for the use stage
inventory due to the fact that the effective life data are attempting to obtain a more realistic value
for electricity consumed per lifetime, and that we currently have greater confidence in those data
versus the manufactured life data. The manufactured life data will be used in one sense as a
sensitivity analysis and to discuss potential differences in the use stage impacts based on this
alternative lifetime scenario.
                                           2-40

-------
                                                                         2.4 PRODUCT USE
2.4.2  Data Sources and Data Quality              ;
                  i                     '               .
       Source and quality information for the use stage data are detailed in Appendix H,
Table 11. We assigned four categories of data quality ratings: excellent," average, poor, and
unknown. In general, data assigned higher quality ratings were directly measured and represent
1998 data. As data required more calculation or estimation, or were found from a previous year,
the data quality rating was reduced.
       The overall level of use stage data quality is between average and excellent (Appendix H,
Table 11). However, a distinct difference can be seen in the average data quality ratings given to
manufactured life data (average) and the effective life data (excellent). This implies that greater
confidence can be placed in the effective life data than in the manufactured life data.
Additionally, the energy use rate data appears to be of average quality.

2.4.3  Limitations and Uncertainties

       Details of the limitations and uncertainties associated with the energy use rate, the
effective life, and the manufactured life estimates are presented in Appendix H, Section 5.
                                           2-41

-------
2.5END-OF-LIFE
2.5    END-OF-LIFE
2.5.1  Methodology

       A Technical Memorandum, attached as Appendix I, was prepared for this project that
provides background on the EOL issues for the CRT and LCD.  It also provides details on the
methodology used in this project for the EOL life-cycle inventory. This section summarizes the
salient points from that memorandum needed to understand the EOL methodology and results.
For the EOL analysis, a monitor is assumed to have reached EOL status when:

•      it has served its useful life;
•      is no longer functional; and/or          !
•      is rendered unusable due to technological obsolescence.

Each of these situations is addressed by either the manufactured or effective lives (defined in
Section 2.4).
                                           I
2.5.1.1  EOL disposition options

       The major EOL dispositions considered in this analysis are as follows:

•      recycling - including disassembly and materials recovery;
•      landfilling - including hazardous (Subtitle C) and non-hazardous (Subtitle D) landfills;
•      remanufacturing - including refurbishing! or reconditioning (to make usable again); and
•      incineration - waste to energy incineration.

See Appendix I for further descriptions of these dispositions.  Note that reuse is considered part
of the use stage and not included as an EOL disposition.
       The functional unit in this analysis is one monitor; therefore, the different EOL
dispositions were allocated as a probability of one monitor going to a certain EOL disposition.
Data were somewhat scarce on the percent of monitors going to each disposition, especially for
the LCD monitors. After literature research and;consultation with the project's Technical Work
Group, as well as various other industry experts,; project partners chose best estimates of
disposition distributions. Table 2-21 presents the assumptions used for the EOL life-cycle stage
dispositions for the CRT and LCD, respectively J An explanation of the assumptions and the
sources of the data are presented in Appendix I. j
       The values in Table 2-21 have been used in the baseline scenarios for the CRT and LCD
LCIs.  To address the uncertainty in the LCD estimates, a sensitivity analysis was conducted
(which is discussed in Section 2.7.3).          !
                                          2-42

-------
                                                                         2.5 END-OF-LIFE
      Table 2-21. Distribution of EOL disposition assumptions for the CRT and LCD
Disposition
Incineration •
Recycling . i
Remanufacturing ;
Hazardous waste landfill
Solid waste landfill
s
CRT
15%
11%
3%
46%
25%
LCD
15%
15%
15%
5%
50%
 Sources: NSC, 1999; EPA, 1998; Vorhees, 2000; CIA, 1997; EIA, 1999b

2.5.1.2 Data collection

       Inventory data were needed for each of .the EOL dispositions to be included in the life-
cycle profiles. Primary data were collected for CRT recycling from three companies.
Hazardous/solid waste landfilling and incineration were developed from secondary data obtained
from Ecobilan. In attempts made to obtain remanufacturing data, it was found that
remanufacturing prpcesses span a wide range of activities, from as little as replacing button tops
to as extensive as testing and replacing PWBs or transformers. Given the broad range of
possibilities, no single set of operations could be identified to adequately represent
remanufacturing activities that could be incorporated in our model. Remanufacturing data were,
therefore, excluded ;from the assessment.

Recycling

       Companies willing to provide CRT recycling data were given EOL questionnaires, which
are similar to the manufacturing questionnaires, but modified as appropriate for the EOL life-
cycle stage (see Appendix I). The questionnaires were used as a guide for collecting inventory
data. The companies agreed to provide inventory data through personal meetings and telephone
conversations rather than completing the detailed questionnaire. As a result, the most critical
data were identified by the research team to prioritize  data needs, and all the details in the
questionnaire may riot have been provided.
       The three companies contacted were: (1) DMC Recycling; (2) A & B Recycling; and (3)
The  Oak Ridge National Recycle Center (TORNRC).  DMC shreds the complete monitor up and
separates the recovered materials into three major material streams: ferrous, silica-based, and
copper-based.  The ferrous metals are sent to steel mills for recycling, while the other streams are
sent to lead and copper smelters, respectively. A & B Recycling performs a partial disassembly
of the casing and other materials outside the CRT.  These materials (namely, HIPS, steel,
aluminum, and copper wiring) are sent for recycling, while the CRT itself is shipped to
Envirocycle (a CRT recycler in Pennsylvania), where the glass is recovered and sent for recycling
back into CRT glass, and the other materials are also recovered for recycling. TORNRC
conducts complete monitor disassembly (which includes the CRT recycling process similar to the
one performed at Envirocycle), and recovers the individual materials for subsequent recycling.
       None of the recycling companies contacted have yet encountered end-of-life LCDs in .any
appreciable quantities that would justify the development of a separate recycling process for
them. Whatever sporadic quantities of LCDs that they do receive (mainly notebook computer
                                          2-43

-------
 2.5END-OF-LIFE
 displays) are either sent for refurbishing/resale or are processed along with other electronic
 equipment, by recovering different materials from them, such as metals, glass, plastics, etc. In
 the absence of actual data for LCD recycling, the shredding-and-materials-recovery process
 followed by DMC Recycling for CRTs was assumed to be suitable for recovering materials from
 LCDs as well, and was therefore used to model LCD recycling.
                                            i
 Landfilling and Incineration
                                            i
       Generic secondary data were used for the incineration and landfilling processes because
 when monitors are disposed of, they are combined with multiple waste streams, and data for
 monitors alone are not readily available. Although data specific to landfilling and incineration
 operations for monitors alone were not available, DEAM inventories from Ecobilan were
 available for landfilling and incinerating the following major monitor materials (by weight):
 steel, glass, plastic, and aluminum. These inventories were combined, based on the approximate
 proportion of each material in a CRT and an LCD, to create individual processes for landfilling
 and for incineration (for each monitor type). The proportions of these materials in a CRT and an
 LCD, presented in Table 2-22, are estimates of the final assembled monitor based on the
 manufacturing inventory data.  Note that these proportions are slightly different from the
 proportion of total inputs per functional unit presented in Section 2.1.2, Tables 2-2 and 2-3
 because materials efficiency during production is not accounted for here (in Table 2-22). The
 majority of the assembled monitors by weight is accounted for in the overall incineration and
• landfilling processes, as seen in the totals in Table 2-22.
       It should be noted that some of the DEAM inventories associated with incineration or
 landfilling are for generic materials (i.e., glass, plastic), and may not accurately represent the
 makeup of the material used in the monitors. For example, the glass is not leaded glass, and the
 plastics may not represent the exact breakdown of plastics in the monitors being modeled in this
 study (see Section 2.5.3,  Limitations and Uncertainties for further discussion).
                                            i             •           •
                                            i
              Table 2-22. Percent contribution of major materials in the final
                                         product
Material
Glass
Steel
Plastic
Aluminum
Total
CRT
43% (9.48 kg)
30% (6.61 kg)
17% (3.75 kg)
2% (0.441 kg)
92% (22.043 kg)
LCD
9% (0.585 kg)
47% (3.055 kg)
40% (2.60 kg)
1% (0.065 kg)
97% (6.5 kg)

 2.5.1.3 Assumptions

        The assumptions used in the EOL life-cycle stage include the percentage breakdown of
 each EOL disposition option (see Table 2-21), the breakdown of materials for the incineration
 and landfilling inventories (Table 2-22), as well as those listed in Section 2.2 of Appendix I.
                                           2-44

-------
                                                                        2.5 END-OF-LIFE
2.5.2  Data Sourqes and Data Quality
                  i     .
       Primary CRT recycling data collected were collected from three companies: A & B
Recycling, DMC Recycling, and TORNRC.  Efforts were made to collect all data in the
questionnaires; however, priority was given to obtaining the inventory data. The companies,
preferring to provide data over the phone or during personal meetings, were able to provide the
inventory data required for the analyses in this study.
       Specific DQIs, such as those reported for the manufacturing data (see Section 2.3.2,
Tables 2-14 and 2-15), were not obtained. The data from these companies represent facility
operations ranging from October 1999 to February 2000. Also, while the data obtained are from
three recycling facilities that may have different operations, the averaged inventory data are
intended to be representative of various recycling activities in the industry.
       Data for the EOL life-cycle stage are a combination of primary data, for which we do not
have specific DQI, and secondary data, with limited data quality information. The overall data
quality for this life-cycle stage may therefore be limited, and in relative terms, is lower quality
than the manufacturing stage data.
       Sources of data for EOL distribution assumptions include the National Safety Council
(NSC, 1999), EPA (1998), CIA (1997), EIA (1999b) and Vorhees (2000). These are discussed in
Appendix I.                               ...
                  !
2.5.3  Limitations and Uncertainties
                  i
       Assumptions of the disposition percentages for CRTs and LCDs may not be truly
representative of actual dispositions. Recycling technologies are not yet standardized for the
sorting, separation, and processing of different types of CRT glass, metals, and plastics. The
methods currently employed by a few large-volume recyclers who have been in the CRT
recycling business for some years were used in this study and represent "state-of-the-art" in the
CRT recycling industry. The LCD recycling process used was based on a CRT recycling process
employed by one of :the recycling companies  contacted, as it was considered general enough to be
applicable to LCDs.; In the future, when greater numbers of LCDs begin to arrive at recycling
facilities, more standardized processes for handling LCDs specifically might be developed.
      Limitations for the incineration and landfilling inventories are that incineration and
landfilling of the materials were for generic materials and not specific to actually incinerating or
landfilling a CRT or LCD monitor. For the CRT, the glass incineration portion of the monitor is
for generic, non-leaded glass. The plastics are also generic plastics (mainly those used for
packaging that ultimately end up in municipal solid waste, such as HOPE, LDPE, and PET) and
may not account for! the flame retardants that might be in the plastics, for example. Also, only a
few of the major materials by weight are included in the modeled CRT and LCD that are
incinerated (listed in Table 2-22).
                                         2-45

-------
2.6 TRANSPORTATION
2.6    TRANSPORTATION

2.6.1  Methodology

       Transportation of materials, products, and wastes throughout a product's life-cycle has
environmental impacts and should be included in a comprehensive LCA.  However, complete
transportation data for all life-cycle stages is often difficult to obtain. Only six of the 16
upstream data sets used for this study explicitly stated that transportation was included. An
additional six upstream processes are assumed to have considered transportation (Table 2-9). In
the manufacturing stage, transportation data were collected in the questionnaires that were
distributed to manufacturers. Some transportation data were provided by manufacturers on
materials received by their facility and products iand wastes shipped from their facility. Data
were not obtained on transportation during the use stage because distributing questionnaires and
collecting primary data for the use stage were not within the boundaries and scope of the LCA.
Consequently, individual consumer transport to pick up purchased monitors and to send to a
secondary user or to a recycling/disposal facility were not accounted for.  Similarly, EOL data,
either from CRT recyclers, or secondary data for incineration and recycling did not include
transportation data.  Therefore, the transportation data collected in this study may only represent
a small portion of the overall transport in the life of a monitor.
       The manufacturing data collection questionnaires (Appendix F) provided space for
companies to identify transportation information for each material input, product output, or waste
output. The questionnaire asked for the distance traveled, mode of transport (i.e., vehicle type),
number of trips per year, and percent capacity of the vehicle containing a particular material of
interest. Given this  information, the project researchers calculated the total distance traveled for
a transportation mode per functional unit.      ;
       In order to determine the environmental effects of transport, the total distance traveled
must be linked to an inventory associated with a transport mode on a per-distance-traveled basis.
Ecobilan 's DEAM data provided inventories for several vehicles either on a per-distance-
traveled basis or a per mass-load, per-distance-traveled basis. Given the maximum load of the
vehicle, the latter figures can also be used with the transportation data collected in the
manufacturing questionnaires to estimate the total distance traveled per mode per functional unit.

2.6.2  Questionnaire Results

       In many cases, companies completing  the questionnaires provided partial transportation
data. For the CRT processes, manufacturers supplied adequate transportation data for 66% of the
materials that are expected to have transportation data.  For the LCD processes, 73% of the
materials had adequate transportation data to determine the total distance traveled per mode per
functional unit. Of the transportation data that were provided, Table 2-23 lists the distribution of
transport modes for the CRT and LCD manufacturers. To complete the LCI, transport data from
the questionnaires would have to be linked to  vehicle inventories, which were available through
the DEAM data. However, the vehicle inventories were not linked to the questionnaire data
because the questionnaire asked for percent capacity, but did not couple that with the load
capacity of the vehicle. As a result, the data were inconsistent and could not accurately be used
in the overall product LCI.
                                          2-46

-------
                                                                    2.6 TRANSPORTATION
         Table 2-23. Distribution of transport modes and total distances per mode
Transport mode
Large diesel truck i
Small diesel or gas truck
Ocean t
Rail I
Air ; . ,
Total
CRT
Distribution
of modes
61%
21%
16%
2%
0%
100%
Approx. normalized a
distance traveled per
functional unit
3km

-------
2.6 TRANSPORTATION
assumed by the manufacturer. Although the transport data represent transport of several
materials, products, and wastes into and out of manufacturing facilities, the majority of the
distances traveled are from the transport of the final assembled monitors. Of all the reported
transportation for the manufacturing stage, the distance traveled of a final assembled CRT
monitor via the ocean represents 80% of the total distance traveled. For the LCD, 90% of all
reported kilometers traveled are for transport of the final assembled LCD monitor via airplane.
Transport of the final assembled CRT monitor for all transport modes is 86% of all kilometers
traveled per functional unit. Similarly, for the final assembled LCD monitor, approximately 92%
of all manufacturing transportation reported is for the final assembled monitor.

2.6.3  Data Sources and Data Quality

       Primary data were derived from manufacturing questionnaires and inventory data for
transport vehicles were available through DEAM data. However, inconsistencies between the
data made it impossible to accurately apply the DEAM inventories to the questionnaire data.
Therefore, the data quality is very low and complete transportation inventory data are excluded
from the analysis results.                      I

2.6.4  Limitations and Uncertainties

       Inconsistencies between data collected in, questionnaires and DEAM data made it
impossible to use the transportation inventory data as part of the overall life-cycle. From rough
estimates based on data received, it is possible that the transportation impacts are not driving
overall life-cycle impacts, however, this would need to be investigated further to confirm such a
conclusion.
                                          2-48

-------
                                       2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
2.7    SUMMARY OF LIFE-CYCLE INVENTORY RESULTS

2.7.1  Baseline LCI

       This section presents the baseline inventory data for the life-cycles of the CRT and LCD
monitors. The baseline scenario meets the following conditions:
                  !
•      uses the effective life use stage scenario;
•      uses the average value of all the energy inputs from the primary data for glass
       manufacturing;
•      assumes LCD glass manufacturing processes use the same amounts of energy as CRT
       glass manufacturing per kilogram of glass produced;
•      excludes twp outliers from the average of the energy inputs in the LCD panel/module
       manufacturing inventory;
•      excludes transportation in the manufacturing stage, but includes any transportation
       embedded in upstream data sets; and,
•  .    includes the| manufacturing process of materials used as fuels (e.g., natural gas, fuel oils)
       in the life-cycle stage in which they are consumed instead of in the materials processing
       stage. In cases where materials normally considered to be fuels are used as ancillary
       materials their manufacturing processes are included with other upstream processes.

Inventory data presented here are used to calculate impacts in the impact assessment (Chapter 3),
which translates inventory items into impacts. Note that only limited conclusions can be made
based on the inventory alone.
       Table 2-24 presents the total quantity of inputs and outputs for the entire life-cycles of the
CRT and LCD base;d on input and output types. Definitions of the input and output types were
presented in Table 2-1 in Section 2.1.1.  Graphs depicting selected input and output types,
derived from the values in Table 2-24, are in Figures 2-5,2-6, and 2-7.  Complete inventory
tables for each input and output type by life-cycle stage for the CRT and LCD are provided in
Appendix J. The inventories presented in Appendix J list each individual input or output
alphabetically for a particular input or output type. The individual inputs or outputs may be the
sum of that material for several processes.
       The total inventory results for life-cycle inputs reveal that more primary materials,3 water,
fuels, electricity, and total energy (i.e., fuel energy plus electricity) are used throughout the CRT
life-cycle while more ancillary materials are used throughout the LCD life-cycle. For the life-
cycle outputs, the CRT releases more air emissions; water pollutants; hazardous, solid, and
radioactive waste; and radioactivity than the LCD. The LCD releases more total wastewater than
the CRT. The data that comprise the inventory totals presented in Table 2-24 are listed in
Appendix J and broken down by life-cycle stage.  Further details on the inventory are provided
for each monitor type below.
         Note that the total mass of primary materials includes the inputs to each process, which may duplicate
materials used in processes subsequent to other processes. For example, the primary materials used in steel
production are added to the steel used as a primary material for monitor assembly.

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
             Table 2-24. Total life-cycle inventory summary - baseline analysis
Inputs
Primary materials
Ancillary materials
Water
Fuels
Electricity
Total energy
Outputs
Air pollutants
Wastewater
Water pollutants
Hazardous waste
Solid waste
Radioactive waste
Radioactivity
CRT
6.536+02
1.98e+01
1.31e+04
4.336+02
2.49e+03
2.08e+04
LCD
3.63e+02
2.08e+02
2.82e+03
3.86e+01
1.20e+03
2.84e+03
Units
kg/functional unit
kg/functional unit
kg (or L)/functional unit
kg/functional unit
MJ*/functional unit
MJ*/functional unit

6.64e+02
1.52E+03
2.09E+01
9.46e+00
1.72e+02
2.90e-03
8.98e+07
3.46e+02
3.13e+03
1.68e+00
6.29e+00
5.23e+01
1.48e-03
4.01e+07
kg/functional unit
kg (or L)/functional unit
kg/functional unit
kg/functional unit
kg/functional unit
kg/functional unit
Bq/functional unit
 *3.6MJ=lkWh                                 ;
 Note: Bold indicates the larger value when comparing the CRT and LCD.
                                      Material Inputs
                             B Primary materials 0 Ancillary materials u Water • Fuels
         ts
         I

         I
            15000n
            10000-
5000-
                     13,100
           653
                 20
                            433
           2,820
363   208         39
                                CRT                            LCD

                           Figure 2-5. Mass-based life-cycle inputs
                                            2-50

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-26. Top 99% of CRT
Material
Petroleum (in grpund)
Coal, average (in ground)
Assembled CRT monitor
Natural gas
Cathode ray tubei(CRT)
CRT glass, unspecified
Iron (Fe, ore)
Steel
Natural gas (in ground)
Sand
Recycled CRT Glass
Bauxite (A12O3, ore)
Iron scrap
Polycarbonate resin
PWB-laminate
Printed wiring board (PWB)
Styrene-butadiene copolymers
PPE
Upstream
1.32E+00
3.57E+00
0
0
0
0
6.90E+00
2.48E-06
8.41E-01
0
0
1.37E+00
9.46E-01
0
0
0
0
0
primary materials in
Mfg
3.72E+02
5.15E+00
0
1.47E+00
1.07E+01
9.76E+00
0
5.16E+00
3.27E+00
2.40E+00
2.06E+00
0
0
9.23E-01
8.47E-01
8.47E-01
8.27E-01
7.35E-01
Use
3.80E+00
1.79E+02
2.20E+01
1.40E+01
0
0
0
0
0
0
0
0
0
0
0
0
0
0
puts (kg/fur
EOL
-1.52E+00
1.79E-02
0
-8.88E-02
0
0
0
0
-1.64E+00
0
0
0
0
0
0
0
0
0
ictional un
Total
3.75E+02
1.88E+02
2.20E+01
1.54E+01
1.07E+01
9.76E+00
6.90E+QO
5.16E+00
2.47E+00
2.40E+00
2.06E+00
1.37E+00
9.46E-01
9.23E-01
8.47E-01
8.47E-01
8.27E-01
7.35E-01
iti
% of total
54.17%
27.15%
3.38%
2.36%
1.64%
1.50%
1.06%
0.79%
0.38%
0.37%
0.32%
0.21%
0.14%
0.14%
0.13%
0.13%
0.13%
0.11%
See Appendix J for complete inventory table.
       Table 2-27. Top 99% of CRT ancillary materials inputs (kg/functional unit)
Material
Clay (in ground);
Sand (in ground)
Limestone
Limestone (CaCO3, in ground)
Lime
Sodium chloride ;(NaCl, in ground or in
sea)
Sulfuric acid
Hydrochloric acid
Sodium hydroxide
Pyrite (FeS2, ore)
Nitric acid
Ferric chloride
Calcium Chloride
Calcium hydroxide
Hydrofluoric acid
Hydrogen peroxide
Ammonium hydroxide
Pumice
Ammonium chloride
Upstream
4.49e-03
5.85e-02
0
8.60e-01
0
7.61e-01
0
0
0
1.94e-01
0
0
0
0
0
0
0
0
0
Mfg
0
2.74e-02
6.91e-02
1.08e+00
3.04e-02
1.26e-02
2.38e-01
2.36e-01
1.98e-01
0
1.44e-01
1.37e-01
1.27e-01
9.54e-02
8.65e-02
8.45e-02
7.90e-02
7.86e-02
7.76e-02
Use
0
0
2.41e+00
0
1.06e+00
0
0
0
0
0
0
0
0
0
0
0
0
0
0
EOL
8.19e+00
2.71e+00
2.41e-04
-2.39e-01
1.06e-04
-3.07e-05
0
0
0
0
0
0
0
0
0
0
0
0
0
Total
8.19e+00
2.80e+00
2.48e+00
1.70e+00
1.09e+00
7.73e-01
2.38e-01
2.36e-01
1.98e-01
1.94e-01
1.44e-01
1.37e-01
1.27e-01
9.54e-02
8.65e-02
8.45e-02
7.90e-02
7.86e-02
7.76e-02
% of total
41.35%
14.13%
12.51%
8.58%
5.49%
3.90%
1.20%
1.19%
1.00%
0.98%
0.73%
0.69%
0.64%
0.48%
0.44%
0.43%
0.40%
0.40%
0.39%
                                          2-60

-------
                                      2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
e 6.00&07 1
•& 4.00&Q7 -
0 2.00&07 -
*•«
C O.OOBOO •
sg. -2.00Ef07 •
37,900,00
i
CRT- Radioactivity
•TolandfilligTotreatmentnTosurfacewaterpToair " :
0 ' 47,600,000
144,000 3,750,000 8 4,760
4200 oo,uuu , &^ 436000 41
' 0 , 	 . 9 > r> SSa 0 n n
Upstream Mfg '' Use EOL
Life-cycle stage
Figure 2-23. CRT radioactivity outputs by life-cycle stage
CRT Primary Inputs
                                              i
       Bediming with the primary data inputs, Figure 2-11 shows that most primary materials
are from the manufacturing and use life-cycle stages. | To better understand what some of the top
contributing materials are to those life-cycle stage totals, Table 2-26 shows the top 99% of the
materials contributing to the total CRT primary input inventory.  As shown in Table 2-26, !the
largest material contributor is petroleum, which is abput 54% of all the primary CRT inputs. The
petroleum is mostly (>98%) from the LPG production process, which relates back to the LPG
needed as a fuel in glass production. The other major contributor to primary material inputs is
coal (~27%) which is used to produce electricity consumed in the use stage. More detail on the
processes that contribute greatest within the manufacturing stage will be presented below after
brief discussions of the life-cycle stage breakdowns for each inventory type. For the complete
list of primary materials in the CRT inventory, the total mass, and the mass contribution of each
life-cycle stage, see Appendix J-l, Table J-l.       |

CRT Ancillary Inputs
                                              i
       Observing Figure 2-12, the mass of ancillary CRT inputs in the EOL life-cycle  stage was
greatest (11 kg/functional unit).  The upstream stages|had the lowest mass of ancillary inputs
compared to the other life-cycle stages. To better understand the materials contributing to,those
totals, Table 2-27 shows that clay is the greatest contributor by mass at 41% of the total CRT
ancillary inputs.  Clay is used predominately during EOL incineration and landfilling. See Table;
J-2 in Appendix J for the complete list of ancillary materials in the CRT inventory.
                                        2-59

-------
                                       2.f SUMMAR¥ bf LIFE-CYCLE INVENTORY RESULTS
Table 2-27. Top 99% of CRT ancillary materials inputs (kg/functional unit)
Material ,
Alkali cleaning agent
Iron (Fe, ore)
Potassium peroxymonosulfate
Sulfuric acid, aluminum salt
Alkali soda (to neutralize acid waste
water)
Polyethylene glycol
Bauxite (A12O3, ore)
Nitrogen
PWB-solder mask solids
Potassium hydroxide
Lubricant (unspecified)
Chlorine
Zinc (Zn, ore)
Aluminum Oxide
Oil (in ground)
Sodium Carbonate
Tin (Sn, ore)
Upstream
0
7.23e-02
0
0
0
0
1.10e-03
0
0
0
4.11e-02
0
3.79e-02
0
0
0
2.43e-02
Mfg
7.72e-02
0
7.06e-02
6.75e-02
5.45e-02
5.04e-02
4.476-02
4.576-02
4.37e-02
4.27e-02
0
4.03e-02
0
3.37e-02
0
3.22e-02
0
Use
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
EOL
. 0
3.41e-03
0
0
0
0
-1.14e-04
0
0
0
0
0
0
0
3.35e-02
0
0
Total
• 7.72e-02
7.57e-02
7.06e-02
6.75e-02
5.45e-02
5.04e-02
4.57e-02
4.57e-02
4.37e-02
4.27e-02
4.11e-02
4.03e-02
3.79e-02
3.37e-02
3.35e-02
3.22e-02
2.43e-02
% of total
0.39%
0.38%
0.36%
0.34%
0.28%
0.25%
0.23%
0.23%
0.22%
0.22%
0.21%
0.20%
0.19%
0.17%
0.17%
0.16%
0.12%
 See Appendix J for complete inventory table.
CRT Utility Inputs

       Utility inputs in the CRT life-cycle are presented in the inventory in Table 2-28 and
include fuel (kg/functional unit), electricity (MJ/functional unit), and water (kg or L/functional
unit) inputs. Figures 2-13, 2-14, and 2-15 show the total fuels, water, and electricity inputs,
respectively.  The fuel and electricity inputs have also been combined into a total energy input
category, shown in Figure 2-16. This is also considered one of the impact categories of the LCIA
that will be presented in Chapter 3. Therefore, more details on how it is calculated are available
in Chapter 3.  Briefly, the mass of the fuels are converted to units of energy and added to the
electrical energy quantities (in units of MJ).
                                           2-61

-------
             2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
                                         Table 2-28. CRT utility inputs
Material
Upstream
Mfs
Use
EOL
Total :
% of total
Fuels (kg/functional unit):
LPG
Natural gas (in ground)
Coal, average (in ground)
Petroleum (in ground)
Fuel oil #6
Fuel Oil #2
Natural gas
Coal, lignite (in ground)
LNG
Uranium (U, ore)
Fuel oil #4
Total fuels
0
2.76E+00
2.25E+00
2.02E+00
0
0
0
9.73E-01
0
1.21E-04
0
8.00E+00
3.51E+p2
4.56E+01
1.36E+01
9.71E+00
3.68E+00
1.16E+00
2.44E+60
!o
3.35E-01
2.29E-04
1.37E-01
4.28E+02
0
0
0
0
0
0
0
0
0
0
0
0
3.03E-03
-2.09E-01
-1.16E-02
-5.77E-02
0
0
-1.30E+00
0
0
-1.99E-07
-1.38E+00
-2.95E+00
3.51E+02
4.82E+01
1.58E+01
1.17E+01
3.68E+00
1.16E+00
1.14E+00
9.73E-01
3.35E-01
3.49E-04
-1.24E+00
4.33E+02
81.10%
11.14%
3.66%
2.70%
0.85%
0.27%
0.26%
0.22%
0.08%
<0.01%
-0.29%
1130.00%
Electricity (Ml/functional unit):
Electricity
7.32E+01
1.29E+02
2.29E+03
2.29E-01
2.49E+03

Water (ke or L/functional unit): '
Water
5.54E+02
1.14E+04
1.14E+03
-2.73E+01
1.31E+04 '

Total energy (fuels and electricity. Ml/functional unit):
Energy
3.66E+02
1.83E+04
2.29E+03
-1.28E+02
2.08E+04

                   Table 2-28 shows that LPG used in the manufacturing stage dominates the fuel inputs.
            LPG from the manufacturing stage is equal to about 81% of all the fuel inputs in the CRT life-
            cycle. More detail into the process-specific contributions within the manufacturing stage will be
            presented below at the end of this section. Electricity inputs, however, are dominated by the use
            stage (~92% of all electricity throughout the CRT life-cycle).  When fuel energy and electrical
            energy are combined into a total energy input value, the overall energy from manufacturing
            greatly exceeds that from the use stage (18,300 MJ/functional unit versus 2,290 MJ/functional
            unit).  This is depicted in Figure 2-16.         I
                   The other utility listed in Table 2-28 is water. Nearly 87% of the water inputs in the CRT
            life-cycle are from the manufacturing processes; land nearly 80% are from LPG production alone.
            The life-cycle stage contributing the next most is the use stage at 8.7%.  This is from the water
            used to generate electricity used during the use stage. The upstream stages only contribute about
            4% to the total water inputs for the CRT life-cycle. Table J-3 in Appendix J provides the
            complete list of inventory items for the CRT.
                                                       !                '               '  '
            CRT Air Outputs                           \                    '             ,

                   Air emissions from the CRT life cycle are dominated from the use stage as seen in
            Figure 2-17. This indicates that most air emissions by mass are from the generation of electricity
            used by consumers of the monitors. Nearly 68% of the total life-cycle air emissions by mass (or
            450 kg/functional unit) are from the use stage. Carbon dioxide (CO2) alone constitutes 445
_
                                                      2-62

-------
                                        2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
kg/functional unit (or about 66% of all air emissions by mass in the life-cycle and nearly 99% of
the use stage air emissions). Table 2-29 reveals the individual contribution of CO2 and other
inventory items that contribute to the top 99.99% of air emissions. (See Appendix J for complete
inventory table.) These are organized from the air emissions that are the largest contributors to
those that are the smaller contributors. Table J-4 in Appendix J shows the contribution of every
air emission in the inventory, organized alphabetically.  The next largest air emissions, by life-
cycle stage, are emitted during the manufacturing stage, which contribute about 28% to the total
life-cycle air emissions. Almost 85% of that is air emissions from the LPG production process.
All the air emissions in the CRT inventory are designated as direct emissions to the ambient
environment.5
Table 2-29. Top 99.99% of CRT air pollutant emissions (kg/functional unit)
Material
Carbon dioxide
Sulfur dioxide
Nitrogen oxides
Methane
Sulfur oxides
Carbon monoxide
PM
Nonmethane hydrocarbons, remaining
unspeciated
Hydrocarbons, remaining unspeciated
Hydrochloric acid .
Other organics
PM-10
Nitrogen dioxide
Upstream
2.92e+01
3.37e-01
6.99e-03
6.40e-02
5.71e-03
4.18e-02
1.28e-01
9.97e-02
1.28e-02
2.39e-03
5.60e-04
0
5.76e-02
Mfg
1.79e+02
1.26e-01
6.95e-01
9.08e-01
8.20e-01
4.58e-01
1.31e-01
1.10e-01
1.58e-01
1.12e-02
7.83e-02
3.15e-03
0
Use
4.45e+02
2.49e+00
1.18e+00
6.45e-01
0
8.09e-02
0
0
0
l.OSe-Ol
0
5.78e-02
0
EOL
2.59e+00
8.30e-04
-1.90e-02
^.30e-02
-2.97e-02
-4.176-03
-1.88e-02
-1.916-03
-6.12e-04
-1.046-03
-3.65e-03
4.78e-06
1.85e-03
Total
6.55e+02
2.96e+00
1.86e+00
1.57e+00
7.96e-01
5.76e-01
2.40e-01
2.08e-01
1.70e-01
1.206-01
7.52e-02
6.09e-02
5.95e-02
% of total
98.68%
0.45%
0.28%
0.24%
0.12%
0.09%
0.04%
0.03%
0.03%
0.02%
0.01%
0.01%
0.01%
CRT Water Outputs

       The volume (or mass) of wastewater released throughout the CRT life-cycle is
approximately 1,520 L (kg) per functional unit.  Approximately 6% of that is sent to treatment as
opposed to direct discharge to surface water (Figure 2-18).  The mass of chemical pollutants
within the wastewater streams was calculated separately. The total mass of these water
pollutants released, presented by life-cycle stage, is shown in Figure 2-19.  The manufacturing
life-cycle stage contributes the greatest mass of water pollutants with approximately 20 kg per
functional unit. This is about 96% of all the water pollutants for the entire life-cycle. The
upstream stages have the second greatest mass of water pollutants at nearly 1 kg/functional unit
(just under 4%). The use and EOL stages are small contributors, with the EOL being negative
         Note that some companies may not have reported inventory items associated with all output dispositions,
as only some dispositions are used for impact calculations. For example, outputs that are treated or recycled and not
directly released to the environment are not used in calculating impacts and may not have been reported. This could
be applicable to all output inventories.
                                           2-63

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
due to recovery processes within the EOL stagej Table 2-30 shows the major contributors to the
water pollutant quantities, and reveals that sodnjim ion, chloride ions and dissolved solids
contribute nearly 85% of all the water pollutants to the entire life-cycle. Greater than 95% of the
sodium ion outputs are from LPG production and greater than 78% of the chloride ions are from
LPG production. As with other input/output types, LPG production used in glass manufacturing
has a large impact on the CRT inventory.  For the complete inventory, listing water pollutaints
alphabetically and subtotaled for each life-cycle, stage, see Appendix J, Table J-5. Further details
on the manufacturing stage will be provided later.                               '
Table 2-30. Top 99.9% of CRT water pollutant outputs (kg/functional unit)
Material
Sodium (+1)
Chloride ions
Dissolved solids
COD
Suspended solids
BOD
Waste oil
Dissolved solids
Sulfate ion (-4)
Sulfateion(-4)
Ammonia ions
Metals, remaining
unspeciated
COD
Oil & grease
Nitrogen
Calcium (+2)
Carbonate ion
Phenol
Fluoride
Salts (unspecified)
Suspended solids
Fluorides (F-)
Disposition
surface water
surface water
surface water
surface water
surface water
surface water
surface water
treatment
treatment
surface water
surface water
surface water
treatment
surface water
surface water
surface water
surface water
surface water
surface water
surface water
treatment
surface water
Upstream
2.90e-01
4.29e-01
5.30e-03
l.OOe-02
7.72e-03
3.93e-04
3.65e-03
. 0
0
3.75e-02
3.54e-06
6.74e-04
0
0
4.46e-05
4.96e-03
4.83e-03
6.25e-05
1.89e-05
1.71e-03
0
9.63e-05
Mfg.
7.04e+00
6.48e+00
3.62e+00
1.60e+00
8.69e-01
1.95e-01
l.Ole-01
8.01e-02
1.09e-03
9.61e-04
1 2.76e-02
9.75e-03
8.33e-03
7.46e-03
7.18e-03
0
0
3.63e-03
3.45e-03
1.62e-03
1.29e-03
2.97e-03
Use
0
0
0
0
0
0
0
0
6.84e-02
0
0
0
0
0
0
0
0
0
0
0
1.786-03
0
EOL
-3.08e-02
-2.39e-02
-8.77e-05
-3.94e-03
-2.116-03
-4.65e-04
-3.136-04
0
6.84e-06
-1.85e-06
-6.63e-05
-3.42e-05
0
0
0
0
0
-9.41e-06
0
-4.21e-06
1.78e-07
-7.72e-06
Total
7.30e+00
6.88e+00
3.62e+00
1.61e+00
8.74e-01
1.95e-01
1.04e-01
8.01e-02
6.95e-02
3.84e-02
2.75e-02
1.04e-02
8.33e-03
7.46e-03
7.23e-03
4.96e-03
4.83e-03
3.68e-03
3.47e-03
3.33e-03
3.07e403
3.06e-K)3
% of total
34.94%
32.95%
17.36%
7.71%
4.19%
0.93%
0.50%
0.38%
0.33%
0.18%
0.13%
0.05%
0.04%
0.04%
0.03%
0.02%
0.02%
0.02%
0.02%
0.02%
0.01%
0.01%
CRT Hazardous Waste Outputs

       The total mass of hazardous waste generated throughout the life-cycle of the CRT (Figure
2-20) is mostly from the amount of the monitor that is assumed to be placed in a hazardous waste
landfill (Table 2-31). This 7.2 kg is based on the proportion of monitors assumed to be
hazardous waste as determined in Section 2.5 (EOL) and is approximately 87% of the hazardous
waste generated in the EOL stage. Compared to the total mass of hazardous wastes produced
throughout the CRT life-cycle, the EOL stage contributes about 88%. The disposition of the
waste will be used to determine how impacts are calculated in Chapter 3.  Figure 2-20 shows
what portion of hazardous wastes are reported as being landfilled, recycled/reused, or treated.
                                         2-64

-------
                                      2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
The amount of hazardous waste from the jupsjtfearn and manufacturing stages are negligible, by
comparison. Table 2-31 and Table J-6 in Appendix J list the hazardous wastes and where each
hazardous waste is disposed.

       Table 2-31. Top 99.9% of CRT hazardous waste outputs (kg/functional unit)
Material
EOL CRT monitor, landfffled
Hazardous waste
CRT glass, cullet
CRT glass, funnel
Transformer
PWB-waste cupric etchant
Printed wiring board (PWB)
General hazardous waste
PWB-solder dross
General hazardous waste
PWB-decontaminating debris
PWB-route dust
PWB-lead- contaminated waste
oil
Chrome liquid waste (D007
waste)
Cinders from CRT glass mfg
(70%PbO)
Unspecified sludge
Unspecified sludge
CRT glass funnel EP dust (Pb)
(D008 waste)
Waste acid (mostly 3% HC1
solution)
Frit
Slag and ash
Broken CRT glass
Hydrofluoric acid
Disposition
landfill
landfill
R/R
R/R
MR
R/R
R/R
treatment
R/R
landfill
treatment
R/R
treatment
R/R
landfill
R/R
landfill
R/R
R/R
landfill
landfill
landfill
landfill
Upstream
0
3.85E-04
0
0
0
0
0
0
0
4.85E-02
0
0
0
0
0
0
0
0
0
0
0
0
0
Mfg
0
6.15E-01
0
0
0
2.25E-01
0
1.24E-01
6.70E-02
0
1.55E-02
1.20E-02
1.16E-02
9.80E-03
8.26E-03
5.56E-03
5.22E-03
5.01E-03
3.93E-03
2.99E-03
2.47E-03
1.88E-03
1.78E-03
Use
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
EOL
7.20E+00
-1.50E-03
4.84E-01
2.29E-01
2.28E-01
0
1.46E-01
0
0
-9.61E-05
0
0
0
0
0
0
0
0
0
0
0
0
0
Total
7.20E+00
6.14E-01
4.84E-01
2.29E-01
2.28E-01
2.25E-01
1.46E-01
1.24E-01
6.70E-02
4.84E-02
1.55E-02
1.20E-02
1.16E-02
9.80E-03
8.26E-03
5.56E-03
5.22E-03
5.01E-03
3.93E-03
2.99E-03
2.47E-03
1.88E-03
1.78E-03
% of total
76.1%
6.49%
5.12%
2.42%
2.41%
2.38%
1.54%
1.31%
0.71%
0.51%
0.16%
0.13%
0.12%
0.10%
0.09%
0.06%
0.06%
0.05%
0.04%
0.03%
0.03%
0.02%
0.02%
R/R: recycling/reuse.
See Appendix J for complete inventory table.

CRT Solid Waste Outputs

       Figure 2-21 shows that both the manufacturing and use stages contribute significant
amounts of solid waste by mass to the CRT life-cycle. The majority of the solid waste is
landfilled. In terms of mass, the greatest contributor to the solid waste outputs for the CRT life-
cycle is coal waste that is a result of generating electricity (Table 2-32).  Therefore, coal waste is
predominately in the use stage, which uses the most electricity, but also in the manufacturing
stage, and to a much lesser degree in the EOL stage. Note that the electricity generation
processes that support the secondary data used were derived from a different source (i.e.,
Ecobilan) and do not include coal waste as an output; however, the equally large amount of solid
waste generated from those processes is listed as "slag and ash" in the upstream and
                                          2-65

-------
 2.7 SUMMARY OF LEFE-CYCLE INVENTORY RESULTS
 manufacturing inventories. Overall, the top 80% of solid waste generated in the CRT life-cycle
 is from coal waste, slag and ash, dust/sludge, and fly/bottom ash. Note that different inventories
 used in this project have varying nomenclature ^and some of these solid wastes may indeed
 overlap. Note also that the mass of a CRT monitor that is assumed to be landfilled at the EOL
 (3.9 kg/functional unit) is only approximately 2% of the total mass of solid waste in the CRT
 life-cycle.  See Appendix J, Table J-7 for the complete CRT solid waste inventory.
Table 2-32. To
Material
Coal waste
Slag and ash
Dust/sludge
Fly/bottom ash
Unspecified solid waste
EOL CRT Monitor, landfilled
Unspecified solid waste
Unspecified solid waste
Unspecified waste
EOL CRT Monitor, incinerated
Iron scrap
EOL CRT Monitor, recycled
Broken CRT glass
Mixed industrial (waste)
Slag and ash
EOL CRT Monitor,
remanufactured
Mining waste
Mineral waste
Carbon Steel Scrap
flame retardant high-impact
polystyrene (HIPS)
Waste water treatment (WWT)
sludge
Feme chloride
CRT glass, faceplate
p 99% of CRT solid waste out
Disposition
landfill
landfill
landfill
landfill
landfill
landfill
treatment
recycle/reuse
landfill
treatment
recycle/reuse
recycle/reuse
recycle/reuse
landfill
recycle/reuse
recycle/reuse
landfill
landfill
recycle/reuse
recycle/reuse
recycle/reuse
recycle/reuse
recycle/reuse
Upstream
i 0
9.65E-02
i 0
0
4.94E+00
0
: o
3.07E+00
0
0
3.43E-01
0
0
4.87E-02
i 0
! °
4.48&01
4.42E-01
0
o
! 0
0
0
Mfg
1.46E+00
6.66E+01
5.64E-01
3.65E-01
0
0
3.66E+00
4.33E-01
3.38E+00
0
0
0
1.08E+00
l.OOE+00
6.85E-01
0
0
2.61E-03
0
0
3.72E-01
3.69E-01
0
puts (kg/functional unit)
Use
5.09E+01
0
1.97E+01
1.27E+01
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
EOL
5.09E-03
-1.49E+01
1.97E-03
1.27E-03
-7.86E-01
3.91E+00
0
0
-1.49E-02
3.31E+00
2.50E+00
2.42E+00
0
-5.12E-04
-3.01E-03
6.60E-01
-1.90E-06
-6.76E-06
4.10E-01
4.03E-01
0
0
3.54E-01
Total
5.23E+01
5.18E+01
2.02E+01
1.31E+01
4.15E+00
3.91E+00
3.66E+00
3.50E+00
3.36E+00
3.31E+00
2.85E+00
2.42E+00
1.08E+00
1.05E+00
6.82E-01
6.60E-01
4.48E-01
4.44E-01
4.10E-01
4.03E-01
3.72E-01
3.69E-01
3.54E-01
%of
total
30.37%
30.06%
11.75%
7.59%
2.41%
2.27%
2.12%
2.03%
1.95%
1.92%
1.65%
1.40%
0.62%
0.61%
0.40%
0.38%
0.26%
0.26%
0.24%
0.23%
0.22%
0.21%
0.21%
CRT Radioactive Waste Outputs

       Radioactive waste outputs in the CRT inventory are found only in the electricity
generation and cold-rolled steel production process. Therefore, radioactive wastes will be found
wherever electricity is used in a process in the CRT life-cycle.  Only very small amounts
(approximately 0.003 kg/functional unit) of radioactive waste are generated over the entire life-
cycle of the CRT (Figure 2-22 and Table 2-33).  As expected, the majority of this is linked to the
use stage, where most electricity is used in the CRT life-cycle.  Low-level radioactive waste
(79%) and depleted uranium (20%) are most of the waste, with very small amounts of highly
radioactive waste and some unspecified radioactive waste in the inventory. The inventory of
radioactive waste outputs is small, and therefore, Table 2-33 lists all material outputs associated
                                          2-66

-------
                                       2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
with radioactive waste, in descending order Q£ quantity. Tabje J-8 in Appendix J lists these in
alphabetical order.

              Table 2-33.  CRT radioactive waste outputs (kg/functional unit)
Material
Low-level radioactive waste
Uranium, depleted
Radioactive waste (unspecified)
Highly radioactive waste (Class C)
Total radioactive wastes
Disposition
landfill
landfill
landfill
landfill

Upstream
4.11E-04
0
1.88E-05
8.65E-06
4.39E-04
Mfg
1.38E-04
4.15E-05
0
0
1.80E-04
Use
1.76E-03
5.27E-04
0
0
2.28E-03
EOL
1.76E-07
5.27E-08
0
0
2.29E-07
Total
2.31E-03
5.69E-04
1.88E-05
8.65E-06
2.90E-03
%of
total
79.5%
19.6%
0.6%
0.3%
100.0%
CRT Radioactivity Outputs

       Radioactivity is also inventoried in this project as isotopes that are released to the
environment. Radipactivity is measured in Bequerels and may be released to air, water, or land.
The quantity of radioactivity for each life-cycle stage and different dispositions is presented in
Figure 2-23. Table 2-34 shows the top 99.9% of the radioactivity outputs.
         Table 2-34. Top 99.9% of CRT radioactivity outputs (Bq
Material
Molybdenum-99 (isotope)
Plutonium-241 (isotope)
Xenon-133 (isotope)
Tritium-3 (isotope)
Plutonium-240 (isotope)
Cesium-135 (isotope)
Radon-222 (isotope)
Plutonium-239 (isotope)
Xenon-133 (isotope)
Tritium-3 (isotope)
Xenon-133M (isotope)
Krypton-85 (isotope)
Disposition
treatment
landfill
air
treatment
landfill
landfill
air
landfill
treatment
air
air
air
Upstrea
m
0
3.74e+07
2.43e+03
0
1.62e+05
1.46e+05
1.37e+05
1.14e+05
0
3.47e+02
0
1.73e+02
Mfg
3.72e+06
0
6.28e+03
2.20e+04
0
0
0
0
3.48e+03
2.95e+03
1.99e+04
2.08e+03
Use
4.73e+07
0
3.12e+05
2.80e+05
0
0
0
0
4.43e+04
3.74e+04
2.07e+04
2.65e+04
EOL
4.73e+03
0
3.12e+01
2.80e+01
0
0
0
0
4.43e+00
3.75e+00
2.07e+00
2.65e+00
Total
5.10e+07
3.74e+07
3.21e+05
3.02e+05
1.62e+05
1.46e+05
1.37e+05
1.14e+05
4.78e+04
4.07e+04
4.06e+04
2.87e+04
%of
total
56.75%
41.67%
0.36%
0.34%
0.18%
0.16%
0.15%
0.13%
0.05%
0.05%
0.05%
0.03%
/functional unit)
 See Appendix J for complete inventory table.

Radioactivity outputs are related to the generation of electricity and therefore the greatest
quantity of radioactivity is from the use stage, as expected.  Table J-9 in Appendix J lists the
complete inventory.

CRT Manufacturing Stage

       The inventory tables that show the specific materials (i.e., those in Appendix J and Tables
2-26 through 2-34) are the sums of the materials from one or more processes within a life-cycle
stage.  To burrow down deeper into the data, the manufacturing stage inventory data are broken
                                           2-67

-------
 2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
                                           i                                :
 down by process or group of processes. Groupsj of processes were combined when fewer than
 three companies provided data for a process or when confidentiality agreements precluded
 presenting individual process data. The manufacturing process groups are presented in Table
 2-35. Also for confidentiality purposes, upstream and EOL data (derived from Ecobilan 's data)
 were not broken down by process. Burrowing further into the contributing processes or process
 groups is necessary for future manufacturing improvement assessments. Burrowing further into
 process-specific data at the use stage is not necessary because electricity generation is the only
 process in the use stage.                                                      ',
                                           i
                            Table 2-35.  CRT process groups
Process group
Monitor assembly
Tube
Glass/frit
PWB
Japanese grid
U.S. grid
Fuels
Processes) included
. | monitor assembly
CRT (tube) manufacturing
CRT glass manufacturing, frit manufacturing
PWB manufacturing
electricity generation - Japanese electric grid !
electricity generation - U.S. electric grid
production of fuel oils #2, #4 and #6, LPG, and natural
gas
                                           I                        .
       Tables 2-36 through 2-44 list the specific inventories for each process group in the
manufacturing stage for each input and output type.  Figures 2-24 through 2-34 graph the total
inventories for each process group for each inpui and output type. It should be noted that the
input/output type that had the greatest contribution in the manufacturing stage compared to other
stages was fuel inputs, which also translated into total energy inputs being greatest in the
manufacturing stage. Nonetheless, for purposes 'of showing more detail in the manufacturing
stage and allowing for improvement assessments for manufacturers, the individual material
contributions for each manufacturing process group are presented below.
       Of the total 421 kg of primary materials per functional unit in the manufacturing stage,
fuels production contributes the greatest (374 kg/functional unit), followed by monitor assembly
(20.1 kg/functional unit), and then tube manufacturing (11.9 kg/functional unit) (Figure 2-24).
The specific material contributions are presented in Table 2-36. Only small amountsiof ancillary
materials are used in the CRT life cycle and the manufacturing stage only contributed a small
percentage of the overall ancillary materials in the life-cycle.  However, within the manufacturing
life-cycle stage, fuels production and PWB manufacturing had the greatest amount of ancillary
materials (1.16 kg/functional unit each) (Figure 2-25).                            «
                                         2-68

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-36. CRT manufacturing stage primary material inputs
Material
Process group
Monitor assembly
ABS resin
Aluminum (elemental)
Audio cable assembly
Cables/wires
Cables/wires
Cathode ray tube (CRT)
Connector
CRT magnet assembly
CRT shield assembly - ASTM A366/CC#2
Deflection Yoke assembly
Demagnetic coil - PU cbated paper
Fertile
Phosphate ester
Polycarbonate resin
Polystyrene (PS, high impact)
Power cord assembly
PPE
Printed wiring board (PWB)
Solder, unspecified
Steel
Styrene-butadiene copolymers
Tricresyl phosphate ' .
Triphenyl phosphate
Video cable assembly
Total
Tube
Amyl acetate (mixed isomers)
Aquadag
Blue Phosphor (ZnS)
Blue Phosphor (ZnS.Ag.Al)
CRT glass, unspecified
Electron gun
Frit
Green Phosphor (ZnS)
Green Phosphor (ZnS.Cu.Al)
Nickel Alloy (invar)
Red Phosphor (Y2O2S)
Red Phosphor (Y2O2S:Eu)
Steel
Total
Quantity (kg/
(functional unit)


4.24e-01
3.60e-01
9.45e-02
6.12e-02
3.33e-01
1.07e+01
5.67e-02
7.56e-02
2.42e-01
1.51e-01
1.26e-01
1.70e-01
8.31e-03
9.23e-01
1.51e-01
1.13e-01
7.35e-01
8:47e-01
2.67e-02
3.45e+00
8.27e-01
2.30e-02
5.29e-02
1.13e-01
2.01e+01

1.20e-03
2.06e-02
3.84e-03
1.67e-03
9.76e+00
l.Ole-01
6.67e-02
3.34e-03
1.34e-03
2.726-01
4.65e-03
1.33e-03
1.71e+00
1.19e+01
% of process
group total


2.11%
1.79%
0.47%
0.31%
1.66%
53.30%
0.28%
0.38%
1.21%
0.75%
0.63%
0.85%
0.04%
4.60%
0.75%
0.57%
3.66%
4.22%
0.13%
17.21%
4.13%
0.11%
0.26%
0.57%
100.00%

0.01%
0.17%
0.03%
0.01%
81.70%
0.84%
0.56%
0.03%
0.01%
2.28%
0.04%
0.01%
14.30%
100.00%
% of grand total


























4.76%














2.84%
   2-69

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-36. CRT manufacturing stage primary material inputs
Material
Process group
Glass/frit
Barium Carbonate
Glass, unspecified
Lead
Potassium Carbonate
Recycled CRT Glass
Sand
Sodium Carbonate
Strontium Carbonate
Zircon Sand
Borax
Lead
Silica
Total
PWB
PWB-laminate
Solder (63% tin; 37% lead)
Total
Japanese grid
Coal, average (in ground)
Natural gas
Petroleum (in ground)
Uranium, yellowcake
Total
U.S. grid
Coal, average (in ground)
Natural gas
Petroleum (in ground)
Uranium, yellowcake
Total
Fuels
Petroleum (in ground)
Natural gas (in ground)
Total
Grand Total
Quantity (kg/
(functional unit)
!
'
i 2.97&-01
' 4.91e-02
' 4.47e-01
i 3.78e-01
'• 2.06e+00
2.40e+00
4.88e-01
! 3.31e-01
5.43e-02
i 8.00e-03
' 4.67e-02
5.33e-03
6.56e+00

8.47e-01
1 5.08e-02
8.98e-01

2.28e+00
1.25e+00
! 1.29e+00
3.04e-04
4.82e+00

2.86e+00
. 2.23e-01
6.07e-02
7.74e-05
3.15e+00

3.70e+02
3.27e+00
3.74e+02
! 4.21e+02
% of process
group total
. •"

4.52%
0.75%
6.82%
5.76%
31.37%
36.57%
7.43%
5.05%
0.83%
0.12%
0.71%
0.08%
100.00%

94.35%
5.66%
100.00%

47.41%
25.89%
26.69%
0.01%
100.00%

90.97%
7.10%
1.93%
<0.01%
100.00%

99.12%
0.88%
100.00%

% of grand total
:













1.56%



0.21%





1.14%



,

> 0.75%



188.74%
1(90.00%
                                   2-70

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-37. CRT manufacturing stage ancillary material in
Material
Process group
Monitor assembly
2,2,4-trimethylpentane
Cyclohexane
Fluorocarbon resin
Isopropyl alcohol
Surfactant, unspecified
Synthetic resin, unspecified
Total
Tube
Acetone
Acrylic Polymer, unspecified
Alkali cleaning agent
Alkali soda (to neutralize acid waste water)
Ammonia
Ammonium bifluoride
Ammonium Bichromate
Ammonium fluoride
Ammonium hydroxide
Ammonium Oxalate
Ammonium Oxalate Monohydrate
Boric acid
Calcium Chloride
Calcium hydroxide
Chlorine
Chromium (VI)
Dimethyl Formamide
Ferric chloride
HV Carbon (paste)
Hydrochloric acid
Hydrofluoric acid
Hydrogen peroxide
Isopentylacetate
Muratic Acid (drum)
Nitric acid
Nitrogen
Oxalic acid
Oxygen (Liquid)
Periodic Acid
Polyvinyl alcohol
Polyvinyl Pyrrolidone (PVP)
Sodium Bichromate
Quantity (kg/
(functional unit)


1.50e-04
1.88e-04
3.75e-05
1.94e-02
1.42e-04
8.53e-04
2.08e-02

3.17e-04
9.13e-03
7.72e-02
5.45e-02
1.19e-04
2.04e-03
3.50e-05
8.91e-04
1.41e-03
8.92e-05
3.16e-04
4.73e-03
1.27e-01
9.54e-02
4.03e-02
7.63e-05
4.36e-05
1.37e-01
1.14e-05
4.39e-02
7.39e-03
5.34e-02
1.74e-03
1.87e-03
8.17e-03
4.57e-02
5.35e-05
7.57e-03
2.26e-04
8.11e-03
2.41e-02
1.05e-04
% of process
group total


0.72%
0.90%
0.18%
93.40%
0.68%
4.10%
100.00%

0.04%
1.02%
8.61%
6.08%
0.01%
0.23%
<0.01%
0.10%
0.16%
0.01%
0.04%
0.53%
14.18%
10.64%
4.50%
0.01%
<0.01%
15.32%
<0.01%
4.89%
0.82%
5.96%
0.19%
0.21%
0.91%
5.10%
0.01%
0.84%
0.03%
0.90%
2.69%
0.01%
puts
% of grand total








0.59%

































   2-71

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-37. CRT manufacturing stage ancillary material in]
Material
Process group
Sodium Dichromate Dihydrate (VI)
Sodium hydroxide
Sodium Hypochlorite
Sodium Metabisulfite
Sodium Persulfate
Sulfuric acid
Sulfuric acid, aluminum salt
Toluene
unspecified CRT process material
Xylene (mixed isomers)
Total
Glass/frit
Aluminum Oxide
Cerium Oxide
Chromium Oxide
Hydrofluoric acid
Pumice
Total
PWB
Ammonium chloride
Ammonium hydroxide
Formaldehyde
Glycol ethers
Hydrochloric acid
Hydrogen peroxide
Nitric acid
Polyethylene glycol
Potassium hydroxide
Potassium permanganate
Potassium peroxymonosulfate
PWB-solder mask solids
Sodium Carbonate
Sodium hydroxide
Sulfuric acid
Total
Japanese grid
Lime
Limestone
Total
U.S.grid
Lime
Quantity (kg/
(functional unit)

3.10e-05
3.52e-03
' 9.25e-05
4.67e-03
i 3.54e-04
5.58e-02
6.75e-02
! 4.80e-03
i 5.77e-03
! 4.80e-04
8.97e-01
i
! 3.37e-02
3.28e-03
i 5.62e-05
i 7.91e-02
! 7.86e-02
1.95e-01
!
7.76e-02
7.76e-02
6.60e-03
' 2.35e-02
1.92e-01
3.10e-02
; 1.36e-01
! 5.04e-02
4.27e-02
i 1.16e-03
i 7.06e-02
! 4.37e-02
3.22e-02
; 1.94e-01
1 1.83e-01
1.16e+00

i 1.35e-02
i 3.06e-02
4.41e-02

1.69e-02
% of process
group total

<0.01%
0.39%
0.01%
0.52%
0.04%
6.23%
7.53%
0.54%
0.64%
0.05%
100.00%

17.30%
1.68%
0.03%
40.61%
40.37%
100.00%

6.68%
6.68%
0.57%
2.03%
16.51%
2.67%
11.70%
4.34%
3.68%
0.10%
6.08%
3.76%
2.77%
16.71%
15.72%
100.00%

30.57%
69.43%
100.00%

30.52%
puts
% of grand total







•



25.36%






5.51%



i






j





; 32.84%



1.25%


                                   2-72

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-37. CRT manufacturing stage ancillary material in
Material
Process group
Limestone
Total
Fuels
Bauxite (A12O3, ore)
Limestone (CaCOS, in ground)
Sand (in ground)
Sodium chloride (NaCl, in ground or in sea)
Total
Grand Total
Quantity (kg/
(functional unit)

3.85e-02
5.53e-02

4.47e-02
1.08e+00
2.74e-02
1.26e-02
1.16e+00
3.54e+00
% of process
group total

69.48%
100.00%

3.85%
92.71%
2.36%
1.08%
100.00%

puts
% of grand total


1.57%





32.89%
100.00%
   2-73

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
                 Table 2-38. CRT manufacturing stage utility inputs
Material
Process group
Fuels (kg/functional unit):
Monitor assembly
Fuel oil #4
Tube
Fuel oil #6
LNG
Natural gas
Total
Glass/frit
Fuel oil #2
Liquified petroleum gas (LPG)
Natural gas
Total
PWB
Natural gas
Fuels
Coal, average (in ground)
Natural gas (in ground)
Petroleum (in ground)
Uranium (U, ore)
Total
Grand Total
Electricity (MJ/functional unit):
Monitor assembly
Tube
Glass/frit
PWB
Total
Water (kg or L/functional unit):
Monitor assembly
Tube
Glass/frit
PWB
Japanese grid
U.S.grid
Fuels
Total
Quantity
i
'_
,
1.37e-01
[
3.68e+00
3.35e-01
8.37e-01
4.86e+00

: 1.16e+00
' 3.51e+02
; 1.21e-HH)
3.53e+02

3.91e-01

1.36e+01
, 4.56e+01
9.71e+00
! 2.29e-04
6.89e+01
4.28e+02
L
1.33e+01
3.19e+01
7.40e+01
l.OOe+01
1.29e+02

3.51e+01
S.lle+02
0
4.22e+01
4.43e+01
1.82e+01
' 1.05e+04
1.14e+04
% of process
group total





75.86%
6.91%
17.23%
100.00%

0.33%
99.33%
0.34%
100.00%



19.71%
66.20%
14.08%
<0.01%
100.00%






/









% of grand total



0.03%




1.14%


;

82.62%

0.09%





16.12%
1100.00%

10.27%
24.68%
57.27%
7.77%
100.00%

0.31%
7.09%
0.00%
- 0.37%
0.39%
0.16%
91.69%
100.00%
Total energy (fuels and electricity, MJ/functional unit):
Monitor assembly
Tube
1.90e+01
I 2.37e+02


0.10%
1.29%
                                     2-74

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Material
Process group
Glass/frit
PWB
Fuels
Total
Quantity

1.52e+04
2.74e+01
2.79e+03
1.83e+04
% of process
group total





% of grand total

83.23%
0.15%
15.22%
100.00%
   2-75

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
                 Table 2-39. CRT manufacturing stage air emissions
Material
Process group
Tube
Carbon monoxide
Dimethyl Formamide
Nitrogen oxides
Nonmethane hydrocarbons, remaining unspeciated
Sulfur oxides
Toluene
Xylene (mixed isomers)
Total
Glass/frit
Barium
Carbon dioxide
Carbon monoxide
Chromium
Cobalt
Copper
Fluorides (F-)
Lead
Manganese
Nickel
Nitrogen oxides
PM
Sulfur oxides
Zinc (elemental)
Total
PWB
Formaldehyde
Japanese grid
1,1,1-Trichloroethane
1,2-Dichloroethane
2,3,7,8-TCDD
2,3,7,8-TCDF
2,4-Dinitrotoluene
2-Chloroacetophenone
2-Methylnaphthalene
5-Methyl chrysene
Acenaphthene
Acenaphthylene
Acetaldehyde
Acetophenone
Acrolein
Quantity (kg/
(functional unit)
i.
i
1 1.70E-02
3.49E-05
2.17E-03
1 1.59E-04
9.96E-03
3.84E-03
3.84E-04
! 3.35E-02
<
i 9.33E-10
! 2.85E+00
1.64E-04
1.39E-07
1.43E-10
i 6.33E-10
i 2.93E-05
' 3.22E-07
4.67E-10
i 5.33E-10
i 4.47E-02
1.10E-04
5.08E-05
4.67E-09
2.90E+00
f
3.88E-05

i 6.46E-08
4.57E-08
1.68E-14
5.84E-14
3.19E-10
8.00E-09
2.38E-10
2.51E-11
4.32E-09
3.29E-10
6.52E-07
j 1.71E-08
3.31E-07
% of process group
total


50.61%
0.10%
6.48%
0.47%
29.73%
11.46%
1.15%
100.00%

<0.01%
98.45%
0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
1.54%
<0.01%
<0.01%
<0.01%
100.00%



<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
% of grand total









0.02%

:













1.59%

0.00%














                                      2-76

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-39. CRT manufacturing stage air emissions
Material
Process group
Anthracene
Antimony
Arsenic
Barium
Benzene
Benzo[a]anthracene
Benzo[a]pyrene
Benzo[b j Jc]fluoranthene
Benzo[g,h,i]perylene
Benzyl chloride
Berylh'um
Biphenyl
Bromoform
Bromomethane
Cadmium
Carbon dioxide
Carbon disulfide
Carbon monoxide
Chloride ions
Chlorobenzene
Chloroform
Chromium (ill)
Chromium (VI)
Chrysene
Cobalt
Copper
Cumene hydroperoxide
Cyanide (-1)
Di(2-ethylhexyl)phthalate
Dibenzo[a,h]anthracene
Dichloromethane
Dimethyl sulfate
Dioxins, remaining unspeciated
Ethyl Chloride
Ethylbenzene
Ethylene dibromide
Fluoranthene :
Fluorene
Fluorides (F-)
Formaldehyde
Furans, remaining unspeciated
Quantity (kg/
(functional unit)

4.55E-10
9.50E-07
7.01E-07
5.18E-07
1.52E-06
8.00E-10
4.35E-11
3.89E-10
4.30E-10
8.00E-07
3.23E-08
1.94E-09
4.46E-08
1.83E-07
1.41E-07
1.54E+01
1.49E-07
2.80E-03
6.14E-05
2.51E-08
6.74E-08
5.52E-07
1.34E-07
5.35E-10
1.18E-06
3.18E-07
6.06E-09
2.86E-06
8.34E-08
2.96E-10
3.31E-07
5.49E-08
7.46E-13
4.80E-08
1.19E-07
1.37E-09
1.75E-09
1.83E-09
6.60E-06
1.02E-05
1.19E-12
% of process grouj
total

<0.019"

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
                Table 2-39. CRT manufacturing stage air emissions
Material
Process group
Hexane
Hydrochloric acid
Efydrofluoric acid
iideno(l,2,3-cd)pyrene
sophorone
Lead (Pb, ore)
vlagnesium
Manganese (Mn, ore)
Mercury
vlethane
Methyl chloride
Methyl ethyl ketone
Methyl hydrazine
Methyl methacrylate
Methyl tert-butyl ether
Molybdenum
Naphthalene
Nickel
Slitrogen oxides
Citrous oxide
o-xylene
Phenanthrene
Phenol
PM-10
Propionaldehyde
Pyrene
Selenium
Styrene
Sulfur dioxide
Tetrachloroethylene
TOCs, remaining unspeciated
Toluene
Vanadium
Vinyl acetate
Xylene (mixed isomers)
Zinc (elemental)
Total
U.S. grid
1,1,1-Trichloroethane
1 ,2-Dichloroethane
2,3,7,8-TCDD
Quantity (kg/
(functional unit)

! 7.66E-08
1.37E-03
1.71E-04
> , 4.48E-10
6.63E-07
4.41E-07
; 1.26E-05
1.09E-06
; 1.18E-07
8.13E-05
6.06E-07
4.46E-07
•; 1.94E-07
; 2.29E-08
3.99E-08
1.55E-07
2.21E-07
1.47E-05
; 4.07E-02
1.12E-04
1.93E-08
5.21E-09
1.83E-08
: 2.00E-03
4.34E-07
1.26E-09
1 1.61E-06
; 2.86E-08
8.62E-02
4.92E-08
i 1.97E-04
i 1.43E-06
5.71E-06
8.67E-09
i 4.23E-08
! 5.15E-06
1.55E+01
!
! 3.06E-08
5.73E-08
i 2.05E-14
% of process group
total

<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.26%
<0.01%
<0.01%
<0.01%
<0.01%
0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.56%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
100.00%

<0.01%
<0.01%
<0.01%
% of grand total




'











i
f



















8.50%
',



                                     2-78

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-39. CRT manufacturing stage air emissions
Material
Process group
2,3,7,8-TCDF
2,4-Dinitrotoluene
2-Chloroacetophenone
2-Methylnaphthalene
5-Methyl chrysene
Acenaphthene
Acenaphthylene
Acetaldehyde
Acetophenone
Acrolein
Anthracene
Antimony
Arsenic
Barium
Benzene
Benzo[a]anthracene
Benzo[a]pyrene
Benzo[b,j,k]flubranthene
Benzo[g,h,i]perylene
Benzyl chloride
Beryllium
Biphenyl
Bromoform
Bromomethane
Cadmium
Carbon dioxide
Carbon disulfide
Carbon monoxide
Chloride ions
Chlorobenzene
Chloroform
Chromium (3H)
Chromium (VI)
Chrysene
Cobalt
Copper
Cumene
Cyanide (-1)
Di(2-ethylhexyl)phthalate
Dibenzo[a,h]anthracene
Dichloromethane
Quantity (kg/
(functional unit)

7.30E-14
4.01E-10
l.OOE-08
4.26E-11
3.15E-11
9.06E-10
3.60E-10
8.16E-07
2.15E-08
4.15E-07
3.11E-10
6.96E-08
5.99E-07
3.28E-08
1.86E-06
1.48E-10
5.44E-11
1.70E-10
5.75E-11
l.OOE-06
3.05E-08
2.43E-09
5.58E-08
2.29E-07
7.69E-08
7.10E+00
1.86E-07
1.29E-03
2.90E-06
3.15E-08
8.45E-08
3.88E-07
1.15E-07
1.63E-10
1.94E-07
1.59E-08
7.59E-09
3.58E-06
1.04E-07
1.40E-11
4.15E-07
% of process group
total

<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
98.98%
<0.01%
0.02%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
% of grand total










































   2-79

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
                 Table J2-39. CRT manufacturing stage air emissions
Material
Process srowo
Dimethyl sulfate
Dioxins, remaining unspeciated
Ethyl Chloride
Ethylbenzene
Ethylene dibromide
Fluoranthene
Fluorene
Fluoride
Formaldehyde
Furans, remaining unspeciated
Hexane
Hydrochloric acid
Hydrofluoric acid
Indeno(l,2,3-cd)pyrene
Isophorone
Lead
Magnesium
Manganese
Mercury
Methane
Methyl chloride
Methyl ethyl ketone
Methyl hydrazine
Methyl methacrylate
Methyl tert-butyl ether
Molybdenum
Naphthalene
Nickel
Nitrogen oxides
Nitrous oxide
o-xylene
Phenanthrene
Phenol
Phosphorus (yellow or white)
PM-10
Propionaldehyde
Pyrene
Selenium
Styrene
Sulfur dioxide
Tetrachloroethylene
Quantity (kg/
(functional unit)

6.87E-08
9.33E-13
;• 6.01E-08
! 1.35E-07
! 1.72E-09
: 1.07E-09
1.34E-09
3.12E-07
1.35E-06
1.49E-12
9.59E-08
: 1.72E-03
2.15E-04
1.05E-10
8.30E-07
2.04E-07
1.57E-05
' 7.28E-07
:. 1.20E-07
1.03E-02
7.59E-07
5.58E-07
! 2.43E-07
! 2.86E-08
! 5.01E-08
! 9.32E-09
! 2.92E-08
1.09E-06
1.88E-02
5.42E-05
! 9.11E-10
4.00E-09
2.29E-08
! 7.91E-08
! 9.22E-04
1 5.44E-07
! 5.32E-10
i 1.87E-06
I 3.58E-08
• 3.98E-02
1 6.16E-08
% of process group
total
-
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.02%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.14%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.26%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.56%
<0.01%
% of grand total

1





,
1









i























i
2-80
i
i
1

-------
                 2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-39. CRT manufacturing stage air emissions
Material
Process group
TOCs, remaining unspeciated
Toluene
Vanadium
Vinyl acetate
Xylene (mixed isomers)
Zinc (elemental)
Total
Fuels
1 , 1 , 1 -Trichloroethane
1,2-Dichloroethane
1 ,4-Dichlorobenzene
2,4-Dinitrotoluene
2-Chloroacetophenone
2-Methylnaphthalene
3-Methylcholanthrene
5-Methyl chrysene . ;
Acenaphthene .
Acenaphthylene
Acetaldehyde
Acetophenone
Acrolein
Aldehydes
Aluminum (elemental)
Ammonia
Anthracene
Antimony
Aromatic hydrocarbons
Arsenic
Barium
Benzene
Benzo[a]anthracene
Benzo[a]pyrene
Benzo[b,j,k]fluoranthene
Benzo[b]fluoranthene
Benzo[g,h,i]perylene
Benzo[k]fluoranthene
Benzyl chloride
Beryllium
Biphenyl
Bromoform
Bromomethane
Quantity (kg/
(functional unit)

9.19E-05
4.06E-07
2.81E-07
1.09E-08
5.30E-08
2.43E-07
7.17E+00

1.36E-07
2.71E-07
3.06E-07
1.90E-09
4.75E-08
6.11E-09
4.59E-10
1.49E-10
5.31E-09
2.17E-09
3.86E-06
1.02E-07
1.97E-06
1.52E-03
1.98E-05
2.35E-03
2.12E-09
6.36E-07
5.29E-08
1.41E-05
3.33E-07
1.57E-02
1.27E-09
. 6.94E-10
7.46E-10
5.07E-10
5.63E-10
5.07E-10
4.75E-06
1.42E-06
1.15E-08
2.64E-07
1.08E-06
% of process group
total

<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
100 00%
-----
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
                Table 2-39. CRT manufacturing stage air emissions
Material
Process group
Butane
Cadmium
Calcium
Carbon dioxide
Carbon disulfide
Carbon monoxide
Chloride ions
Chlorine
Chlorobenzene
Chloroform
Chromium CUT)
Chromium (VI)
Chrysene
Cobalt
Copper
Cumene
Cyanide (-1)
Di(2-ethylhexyl)phthalate
Dibenzo[a,h]anthracene
Dichloromethane
Dimethyl sulfate
Dimethylbenzantbracene
Dioxins, remaining unspeciated
Ethane
Ethyl Chloride
Ethylbenzene
Ethylene dibromide
Fluoranthene
Fluorene
Fluorides (F-)
Formaldehyde
Furans, remaining unspeciated
Halogenated hydrocarbons (unspecified)
HALON-1301
Hexane
Hydrocarbons, remaining unspeciated
Hydrochloric acid
Hydrofluoric acid
Hydrogen sulfide
Indeno(l,2,3-cd)pyrene
Iron
Quantity (kg/
(functional unit)

5.35E-04
8.20E-07
1.72E-05
1.54E+02
8.81E-07
4.36E-01
3.39E-05
5.84E-09
1.49E-07
4.00E-07
2.13E-05
2.13E-05
1.22E-09
i 2.54E-06
1.57E-06
3.59E-08
1.69E-05
4.95E-07
3.61E-10
1.97E-06
3.25E-07
3.82E-09
1.10E-10
7.90E-04
2.85E-07
6.44E-07
i 8.14E-09
5.74E-09
! 7.03E-09
3.84E-06
1.19E-03
5.11E-10
2.94E-13
5.11E-10
4.59E-04
[ 1.58E-01
; 8.14E-03
1.02E-03
! 3.11E-03
9.43E-10
3.83E-05
% of process group
total

<0.01%
<0.01%
<0.01%
97.92%
<0.01%
0.28%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.10%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
% of grand total
\
(






I






1












1


1





i





                                    2-82

-------
                 2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-39.  CRT manufacturing stage air emissions
Material
Process group
Isophorone
Lead
Magnesium
Manganese
Mercury
Metals, remaining unspeciated
Methane
Methyl chloride
Methyl ethyl ketone
Methyl hydrazine
Methyl methacrylate
Methyl tert-butyl ether
Molybdenum
Naphthalene
Nickel
Nitrogen oxides
Nitrous oxide
Nonmethane hydrocarbons, remaining unspeciated
n-Propane
Other organics
o-xylene
Pentane
Phenanthrene
Phenol
Phosphorus (yellow or white)
PM
PM-10
Polycyclic aromatic hydrocarbons
Propionaldehyde
Pyrene
Selenium
Silicon
Sodium
Styrene
Sulfur oxides
Tetrachloroethylene
Toluene
Vanadium
Vinyl acetate ' . •
Zinc (elemental)
Total
Quantity (kg/
(functional unit)

3.93E-06
1.25E-05
7.46E-05
2.26E-05
8.81E-07
3.16E-07
8.98E-01
3.59E-06
2.64E-06
1.15E-06
1.36E-07
2.37E-07
1.97E-06
3.54E-07
1.24E-04
5.88E-01
1.64E-02
1.10E-01
1.69E-06
7.83E-02
1.11E-06
6.62E-04
2.30E-08
1.08E-07
1.25E-05
1.31E-01
2.28E-04
5.87E-11
2.58E-06
3.65E-09
9.47E-06
1.72E-05
1.02E-04
1.69E-07
8.10E-01
2.92E-07
3.81E-06
2.68E-04
5.15E-08
1.02E-05
1.57E+02
% of process group
total

<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.57%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.37%
0.01%
0.07%
<0.01%
0.05%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.08%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.52%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
100.00%
% of grand total









































85.96%
                    2-83

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Material
Process group
Grand Total
Quantity (kg/
(functional unit)
i
1.83E+02
% of process group
total


% of grand total

100.00%
                                   2-84

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-40. CRT manufacturing stage water outputs (wastewaters and pollutants)
Material
Process group
Disposition

Quantity (kg/
functional unit)

% of process group
total

% of grand
total

WASTEWATER STREAMS
Tube
Wastewater stream
Wastewater stream
Total
Glass/flit
Wastewater stream
PWB
Wastewater stream
Fuels
Wastewater stream
Grand Total

treatment
surface water

.
surface water

treatment

surface water


5.26e+01
4.81e+02
5.33e+02

3.62e+01

4.22e+01

8.94e+02
1.51e+03

9.87%
90.13%
100.00%







WASTEWATER POLLUTANTS
Tube
BOD
Chromium ore
Chromium ore
COD
COD
Copper
Cyanide (-1)
Dissolved solids .
Fluoride
Fluoride
Iron
Lead
Lead
Manganese
Molybdenum
Nickel
Nitrogen
Oil & grease
Phosphate as P2O5
Phosphorus (yellow or white)
Suspended solids
Suspended solids
Zinc (elemental)
Zinc (elemental)
Total
Glass/frit
BOD

surface water
surface water
treatment
surface water
treatment
surface water
surface water
treatment
surface water
treatment
surface water
surface water
treatment
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
treatment
surface water
treatment


surface water

6.39e-03
1.02e-05
1.03e-06
7.22e-03
8.33e-03
1.80e-06
6.06e-07
8.01e-02
3.45e-03
3.51e-04
1.65e-04
3.01e-06
1.03e-06
3.60e-06
1.20e-07
7.93e-05
7.18e-03
2.41e-04
1.21e-06
5.05e-05
4.63e-03
1.28e-03
1.39e-05
1.03e-06
1.20e-01

8.20e-06

5.34%
<0.01%
<0.01%
6.04%
6.97%
<0.01%
<0.01%
67.03%
2.89%
0.29%
0.14%
<0.01%
<0.01%
<0.01%
<0.01%
0.07%
6.01%
0.20%
<0.01%
0.04%
3.87%
1.07%
0.01%
<0.01%
100 00%

<0.01%





2.40%

280%

59.38%



























251%


   2-85

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
    Table 2-40. CRT manufacturing stage water outputs (wastewaters and pollutants)
Material
Process group
Chloride ions
Chromium
COD
Dissolved solids
Fluorides (F-)
Iron
Lead
Nickel
Nitrates/nitrites
Oil & grease
Suspended solids
Total
PWB
Copper (+1& +2)
Leadcmpds
Total
Japanese grid
Sulfate ion (-4)
Suspended solids
Total
U.S. grid
Sulfate ion (-4)
Suspended solids
Total
Fuels
Acids (H-f)
Adsorbable organic halides
Aluminum (+3)
Ammonia ions
Aromatic hydrocarbons
Barium cmpds
BOD
Cadmium cmpds
Chloride ions
Chromium (IH)
Chromium (VI)
COD
Copper (+1 & +2)
Cyanide (-1)
Dissolved organics
Dissolved solids
Disposition
i
surface water !
surface water
surface water
surface water
surface water
surface water ,
surface water
surface water
surface water
surface water '
surface water

!.
treatment i
treatment


surface water '
surface water


treatment
treatment \
\
i1
surface water ;
surface water !
surface water i
surface water
surface water ;
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
Quantify (kg/
functional unit)

l.Ole+00
8.20e-08
8.20e-06
3.62e+00
2.93e-03
2.77e-03
4.34e-05
8.20e-08
3.95e-06
7.22e-03
7.23e-03
4.65e+00

9.71e-05
1.62e-05
1.13e-04

8.72e-04
2.27e-05
8.94e-04

1.09e-03
2.84e-05
1.12e-03

2.50e-09
2.27e-15
8.62e-10
l.Ole-07
5.33e-13
1.71e-12
1.21e-06
1.78e-15
3.56e-05
3.31e-10
3.31e-10
9.80e-06
3.55e-14
2.49e-15
6.62e-09
2.21e-07
% of process group
total

21.73%
<0.01%
<0.01%
77.84%
0.06%
0.06%
<0.01%
<0.01%
<0.01%
0.16%
0.16%
100.00%

85.71%
14.29%
100.00%

97.46%
2.54%
100.00%

97.46%
2.54%
100.00%

<0.01%
<0.01%
<0.01%
0.10%
<0.01%
<0.01%
1.22%
<0.01%
36.07%
<0.01%
<0.01%
9.92%
<0.01%
<0.01%
<0.01%
0.22%
% of grand
total












97.45%


i
0.00%



0.02%



0.02%

















                                     2-86

-------
             .
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-40. CRT manufacturing stage water outputs (wastewaters and nollutants)
Material
Process group
Fluorides (F-)
Halogenated matter (organic)
Hydrocarbons, remaining unspeciated
Iron (+2 & +3)
Lead cmpds
Mercury compounds
Metals, remaining unspeciated
Nickel cmpds
Nitrate
Other nitrogen
Phenol
Phosphates
Polycyclic aromatic hydrocarbons :
Salts (unspecified)
Sodium (+1)
Sulfate ion (-4)
Sulfide
Suspended solids
rocs
Toluene
Waste oil
Zinc (+2)
Total
Grand Total
Disposition

surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
urface water


Quantity (kg/
functional unit)

1.80e-08
7.11e-16
2.61e-09
3.79e-ll
7.11e-15
8.17e-18
6.27e-08
3.55e-15
4.53e-09
9.59e-14
2.21e-08
2.02e-ll
8.53e-15
9.83e-09
4.59e-05
4.26e-09
1.38e-09
5.19e-06
5.33e-12
7.82e-14
6.26e-07
1.58e-10
9.88e-05
4.77e+00
% of process group
total

0.02%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.06%
<0.01%
<0.01%
<0.01%
0.02%
<0.01%
<0.01%
<0.01%
46.44%
<0.01%
<0.01%
5.26%
<0.01%
<0.01%
0.63%
<0.01%


% of grand
total
























100.00%
  2-87

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
   Table 2-41. CRT manufacturing stage hazardous waste outputs (kg/functional unit)
Material
Process group
Tube
Frit
Lead sulfate cake
Silica coat waste
Slag and ash
Slurry scrap (chromium-based)
Spent solvent, unspecified
Jnspecified sludge
Unspecified sludge
Waste oxygenated solvents
Waste water treatment (WWT) filters
Total
Glass/fiit
Barium debris (D008 waste)
Broken CRT glass
Chrome debris (D007 waste)
Chrome liquid waste (D007 waste)
cinders from CRT glass mfg (70% PbO)
CRT glass faceplate EP dust (Pb) (D008
waste)
CRT glass funnel EP dust (Pb) (D008
waste)
Hazardous sludge (Pb) (D008)
Hydrofluoric acid
Lead contaminated grit (D008 waste)
Lead debris (D008 waste)
sludge from CRT glass mfg (1% PbO)
Waste acid (mostly 3% HC1 solution)
Waste Batch (Ba, Pb) (D008 waste)
Waste finishing sludge (Pb) (D008
waste)
Total
PWB
General Hazardous Waste
PWB-Decontaminating debris
PWB-Lead contaminated waste oil
PWB-Route dust
PWB-Solder dross
PWB-Waste cupric etchant
Total
Disposition


andfill i'
andfffl
treatment :
andfffl :
andfffl
treatment
andfffl i
ecycling/reuse
treatment i
andfffl ;


andfffl
andfill
treatment
recycling/reuse
andfill
andfffl ;
r
recycling/reuse
landfill
landfill
landfill
landfill
landfill
recycling/reuse
landfill
landfill i


treatment
treatment
treatment
recycling/reuse
recycling/reuse
recycling/reuse

Quantity (kg/
functional
unit)


2.99e-03
2.67e-05
2.86e-04
2.47e-03
8.62e-04
2.75e-04
5.22e-03
5.56e-03
9.48e-05
3.40e-04
1.92e-02

2.14e-04
1.88e-03
1.47e-04
9.80e-03
8.26e-03
1.03e-03
5.01e-03
1.52e-03
1.78e-03
3.46e-05
2.14e-04
8.77e-04
3.93e-03
1.41e-03
2.56e-04
3.64e-02

1.24e-01
1.55e-02
1.16e-02
1.20e-02
6.70e-02
2.25e-01
4.5Se-01
% of process
group total


15.58%
0.14%
1.49%
12.90%
4.50%
1.43%
27.26%
29.03%
0.49%
1.78%
100.00%

0.59%
5.17%
0.41%
26.95%
22.71%
2.83%
13.78%
4.17%
4.89%
0.10%
0.59%
2.41%
10.81%
3.89%
0.70%
100.00%

27.26%
3.41%
2.56%
2.64%
14.72%
49.42%
100.00%
% of grand
total






•





1.70%
















3.23%







40.45%
                                      2-88

-------
                               2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-41. CRT manufacturing stage hazardous waste outputs (kg/functional unit)
Material
Process group
Fuels
Hazardous waste
Grand Total
Disposition


landfill

Quantity (kg/
functional
unit)


6.15e-01
1.13e+00
% of process
group total




% of grand
total


54.62%
100.00%
                                  2-89

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
      Table 2-42. CRT manufacturing stage solid waste outputs (kg/functional unit)
Material
Process group
Monitor Assembly
Broken CRT glass
Cables/wires
Printed wiring board (PWB)
Waste plastics from CRT monitor
Total
Tube
Broken CRT glass
Ferric chloride
Sludge (aquadag)
Sludge (phosphor)
Spent solvents
(toluene,xylene,dimethyl
formamide,isopropyl alcohol)
Unspecified sludge
Waste alkali (cleaning caustic and
alkali soda effluent)
Waste metals, unspecified
Waste oil
Waste oil
Waste Plastic (packing material)
Waste Plastic (styrene foam)
Waste water treatment (WWT) sludge
Waste water treatment (WWT) sludge
Wastepaper
Wood, average
Total
Glass/frit
abrasive sludge
acid absorbent
blasting media
Cobalt nitrate
CRT glass, faceplate
Diesel fuel
Dust
Nickel nitrate
Oily rags & filter media
Oily rags & filter media
parts cleaner solvent
Plating process sludge
Potassium Carbonate
Disposition


recycling/reuse
recycling/reuse
recycling/reuse
recycling/reuse


recycling/reuse
recycling/reuse
landfill
landfill
recycling/reuse
recycling/reuse
recycling/reuse
recycling/reuse
recycling/reuse
treatment
treatment
recycling/reuse
landfill
recycling/reuse
recycling/reuse
landfill


recycling/reuse
landfill
landfill
treatment
landfill
treatment
treatment
treatment
landfill
recycling/reuse
recycling/reuse
landfill
landfill
Quantity (kg/
functional unit)


3.82e-01
8.86e-03
3.70e-02
8.09e-02
5.09e-01

6.94e-01
3.69e-01
2.22e-03
4.31e-03
4.17e-02
1.26e-01
2.12e-02
8.79e-02
1.43e-03
2.55e-03
3.01e-02
3.77e-03
8.43e-02
3.72e-01
8.34e-03
4.94e-03
1.85e+00

4.21e-02
8.13e-05
3.66e-04
6.10e-05
2.43e-02
4.07e-05
3.43e-03
6.10e-05
3.25e-04
4.07e-05
8.13e-05
3.28e-04
3.30e-03
% of process group
total


75.07%
1.74%
7.28%
15.91%
100.00%

37.43%
19.93%
0.12%
0.23%
2.25%
6.78%
1.15%
4.74%
0.08%
0.14%
1.63%
0.20%
4.55%
20.06%
0.45%
0.27%
100.00%

11.34%
0.02%
0.10%
0.02%
6.54%
0.01%
0.92%
0.02%
0.09%
0.01%
0.02%
0.09%
0.89%
% of grand
total
i

;


1
0.63%



|

;

;


,






2.28%


i











                                      2-90

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-42. CRT manufacturing stage solid waste outputs (kg/functional unit)
Material
Process group
sludge (calcium fluoride, CaF2)
Sodium Carbonate
Unspecified sludge
Waste alkali, unspecified
Waste oil
Waste refractory
Waste water treatment (WWT) sludge
PM
Total
PWB
PWB-Drill dust
Unspecified solid waste
Unspecified solid waste
Total
Japanese grid
Coal waste
Dust/sludge
Fly/bottom ash
Total
U.S. grid
Coal waste
Dust/sludge
Fly/bottom ash
Total
Fuels
Aluminum scrap
Aluminum scrap, Wabash 319
Bauxite residues
FGD sludge
Mineral waste
Mixed industrial (waste)
Non toxic chemical waste
(unspecified)
Slag and ash
Slag and ash
Unspecified solid waste (incinerated)
Unspecified waste
Total
Grand Total
Disposition

recycling/reuse
landfill
landfill
treatment
treatment
landfill
landfill
landfill


landfill
recycling/reuse
treatment


landfill
landfill
landfill


landfill
landfill
landfill


recycling/reuse
recycling/reuse
landfill
landfill
landfill
landfill
landfill
landfill
recycling/reuse
treatment
landfill


Quantity (kg/
functional unit)

1.75e-02
3.29e-03
7.69e-03
4.21e-05
6.54e-03
2.44e-03
2.59e-01
5.33e-04
3.72e-01

1.49e-02
4.33e-01
3.66e+00
4.11e+00

6.48e-01
2.50e-01
1.62e-01
1.06e+00

8.12e-01
3.14e-01
2.03e-01
1.33e+00

1.82e-04
5.08e-07
1.21e-02
2.14e-01
2.61e-03
l.OOe+00
6.11e-04
6.66e+01
6.85e-01
1.33e-02
3.38e+00
7.19e+01
8.12e+01
% of process group
total

4.72%
0.89%
2.07%
0.01%
1.76%
0.66%
69.68%
0.14%
100.00%

0.36%
10.53%
89.11%
100.00%

61.12%
23.59%
15.28%
100.00%

61.10%
23.63%
15.27%
100.00%

<0.01%
<0.01%
0.02%
0.30%
<0.01%
1.39%
<0.01%
92.62%
0.95%
0.02%
4.70%
100.00%

% of grand
total









0.46%




5.06%




1.31%




1.64%












88.62%
100.00%
   2-91

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
           Table 2-43. CRTi
Material
Process group
Japanese grid
Low-level radioactive waste
Uranium, depleted
Total
U.S. grid
Low-level radioactive waste
Uranium, depleted
Total
Grand Total
Disposition


landfill
landfill


landfill
landfill


Quantity (kg/
functional unit)


1.10e-04
: 3.31e-05
1.43e-04
.
2.80e-05
8.41e-06
3.65e-05
1.80e-04
% of process
group total


76.93%
23.07%
100.00%

76.93%
23.07%
100.00%

% of grand total
!



79.72%
>
:

20.28%
100.00%
                                   2-92

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-44. CRT manufacturing stage radioactivity outputs
Material
Process group
Japanese grid
Antimony-124 (isotope)
Antimony-125 (isotope)
Argon-41 (isotope)
Barium- 140 (isotope)
Bromine-89 (isotope)
Bromine-90 (isotope)
Cesium-134 (isotope)
Cesium-134 (isotope)
Cesium-137 (isotope)
Cesium-137 (isotope)
Chromium-51 (isotope)
Chromium-51 (isotope)
Cobalt-57 (isotope)
Cobalt-57 (isotope)
Cobalt-58 (isotope)
Cobalt-58 (isotope)
Cobalt-60 (isotope)
Cobalt-80 (isotope)
Iodine-131 (isotope)
Iodine-131 (isotope)
Iodine-132 (isotope)
Iodine-132 (isotope)
Iodine-133 (isotope)
Iodine-133 (isotope)
Iodine-134 (isotope)
Iodine-135 (isotope)
Iodine-135 (isotope)
Iron-55 (isotope)
Lron-59 (isotope)
Krypton-85 (isotope)
Krypton-85M (isotope)
Krypton-85M (isotope)
Krypton-87 (isotope)
Krypton-88 (isotope)
Lanthanum-140 (isotope)
Manganese-54 (isotope)
Manganese-54 (isotope)
Molybdenum-99 (isotope)
Niobium-95 (isotope)
Niobium-95 (isotope)
Disposition


treatment
treatment
air
treatment
air
air
air
reatment
air
treatment
air
treatment
air
treatment
air
treatment
ak
treatment
air
treatment
air
treatment
air
treatment
air
air
treatment
treatment
treatment
air
air
treatment
air
air
treatment
air
treatment
treatment
air
treatment
Quantity (Bq/
functional unit)


4.95e-01
1.97e+00
l.OOe+03
3.67e-02
1.16e-04
4.72e-05
3.18e-03
1.32e+00
2.40e-02
1.99e+00
6.29e-02
2.39e+00
1.69e-04
5.78e-02
2.16e-03
2.35e+01
1.62e-02
6.17e+00
7.58e-02
1.10e+00
1.54e-02
4.17e-01
7.03e+01
4.72e-01
7.98e-02
4.01e-03
3.38e-01
5.62e+00
2.88e-01
1.66e+03
8.06e+01
1.49e+00
3.00e+01
1.41e+02
3.93e-02
8.92e-04
1.57e+00
2.97e+06
3.54e-05
4.056-01
% of process group
total


<0.01%
<0.01%
0.03%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.06%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
98.42%
<0.01%
<0.01%
% of grand
total










































   2-93

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
             Table 2-44. CRT manufacturing stage radioactivity outputs
Material
Process group
Rubidium-88 (isotope)
Ruthenium-103 (isotope)
Silver-llOM (isotope)
Silver-llOM (isotope)
Sodium-24 (isotope)
Strontium-89 (isotope)
Strontium-90 (isotope)
Strontium-95 (isotope)
Sulfur-136 (isotope)
Technetium-99M (isotope)
Technetium-99M (isotope)
Tin-113 (isotope)
Tritium-3 (isotope)
Tritium-3 (isotope)
Xenon-131M (isotope)
Xenon-131M (isotope)
Xenon-133 (isotope)
Xenon-133 (isotope)
Xenon-133M (isotope)
Xenon-133M (isotope)
Xenon-135 (isotope)
Xenon-135 (isotope)
Xenon-135M (isotope)
Xenon-138 (isotope)
Zinc-85 (isotope)
Zirconium-95 (isotope)
Total
U.S.grid
Antimony-124 (isotope)
Antimony-125 (isotope)
Argon-41 (isotope)
Barium-140 (isotope)
Bromine-89 (isotope)
Bromine-90 (isotope)
Cesium-134 (isotope)
Cesium-134 (isotope)
Cesium-136 (isotope)
Cesium-137 (isotope)
Cesium-137 (isotope)
Chromium-51 (isotope)
Chromium-51 (isotope)
Disposition

air
treatment
air
treatment
treatment
treatment
treatment
treatment
treatment
air
treatment :
treatment
air
treatment
air
treatment ;
air
treatment
air
treatment ;
air :
treatment
air
air
treatment
air


treatment
treatment ;
air
treatment
air
air
air :
treatment
treatment
air
treatment
air
treatment
Quantity (Bq/
functional unit)

3.29e-01
4.95e-02
1.06e-06
5.78e-01
8.80e-02
9.51e-02
2.24e-02
2.46e-01
5.30e-02
4.75e-06
3.45e-02
5.46e-02
2.35e+03
1.76e+04
1.36e+02
1.81e+01
1.30e+03
2.78e403
1.96e+04
2.28e+01
7.39e+02
2.07e+01
1.41e+01
4.68e+01
2.65e-02
9.16e-05
3.01e+06

1.26e-01
5.02e-01
2.55e+02
9.33e-03
2.95e-05
1.20e-05
8.09e-04
3.37e-01
1.44e-02
e.lle-03
5.06e-01
1.60e-02
6.07e-01
% of process group
total

<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.08%
0.58%
<0.01%
<0.01%
0.04%
0.09%
0.65%
<0.01%
0.02%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
100.00%

<0.01%
<0.01%
0.03%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
; % of grand
total



























79.70%












i

                                     2-94

-------
, 2;?>SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-44. CRT manufacturing stage radioactivity outputs
Material
Process group
Cobalt-57 (isotope)
Cobalt-57 (isotope)
Cobalt-58 (isotope)
Cobalt-58 (isotope)
Cobalt-60 (isotope)
Cobalt-80 (isotope)
Iodine-131 (isotope)
Iodine-131 (isotope)
Iodine-132 (isotope)
Iodine-132 (isotope)
Iodine-133 (isotope)
Iodine-133 (isotope)
Iodine-134 (isotope)
Iodine-135 (isotope)
Iodine-135 (isotope)
Iron-55 (isotope)
Iron-59 (isotope)
Krypton-85 (isotope)
Krypton-85M (isotope)
Krypton-85M (isotope)
Krypton-87 (isotope)
Krypton-88 (isotope)
Lanthanum-140 (isotope)
Manganese-54 (isotope)
Manganese-54 (isotope)
Molybdenum-99 (isotope)
Niobium-95 (isotope)
Niobium-95 (isotope)
Rubidium-88 (isotope)
Ruthenium- 103 (isotope)
Silver-llOM (isotope)
Silver-llOM (isotope)
Sodium-24 (isotope)
Strontium-89 (isotope)
Strontium-90 (isotope)
Strontium-95 (isotope)
Sulfur-136 (isotope)
Technetium-99M (isotope)
Technetium-99M (isotope)
Tin-1 13 (isotope)
Tritium-3 (isotope)
Disposition

air
treatment
air
treatment
air
treatment
air
treatment
air
treatment
air
treatment
air
air
treatment
treatment
treatment
air
air
treatment
air
air
treatment
air
treatment
treatment
air
treatment
air
treatment
air
treatment
treatment
treatment
treatment
treatment
treatment
air
treatment
treatment
air
Quantity (Bq/
functional unit)

4.30e-05
1.47e-02
5.49e+01
5.98e+00
4.13e-03
1.57e+00
1.93e-02
2.80e-01
3.92e-03
1.06e-01
1.79e+01
1.20e-01
2.03e-02
1.02e-03
8.60e-02
1.43e+00
7.34e-02
4.23e+02
2.05e+01
3.78e-01
7.62e+00
3.58e+01
9.99e-03
2.27e-04
4.00e-01
7.55e+05
9.01e-06
1.03e-01
8.37e-02
1.26e-02
2.69e-07
1.47e-01
2.24e-02
2.42e-02
5.69e-03
6.27e-02
1.35e-02
1.21e-06
8.77e-03
1.39e-02
5.98e+02
% of process group
total

<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.06%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
98.41%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.08%
% of grand
total










































    2-95

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
             Table 2-44. CRT manufacturing stage radioactivity outputs
Material
Process group
Tritium-3 (isotope)
Xenon-131M (isotope)
Xenon-131M (isotope)
Xenon-133 (isotope)
Xenon-133 (isotope)
Xenon-133M (isotope)
Xenon-133M (isotope)
Xenon-135 (isotope)
Xenon-135M (isotope)
Xenon-138 (isotope)
Zinc-85 (isotope)
Zirconium-95 (isotope)
Total
Fuels
Radioactive substance (unspecified)
Radioactive substance (unspecified)
Total
Grand Total
Disposition
:
treatment ;
air i
treatment :
air
treatment !
air
treatment
treatment
air
air
treatment
air


air
surface water .
,

Quantity (Bq/
functional unit)

4.47e+03
3.45e+01
4.60e+00
4.98e+03
7.07e+02
3.31e+02
5.79e+00
5.27e+00
3.59e+00
1.19e+01
6.75e-03
2.33e-05
7.67e+05

9.19e+02
8.52e+00
9.27e+02
3.78e+06
% of process group
total

0.58%
<0.01%
<0.01%
0.65%
0.09%
0.04%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
100.00%

99.08%
0.92%
100.00%

% of grand
total






'



i


20.27%



0.02%
100.00%
                                     2-96

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS




























•|f 400
3 350
•= 300
g 250
3 200
•5 150
i 100
fii 50






C 2-,

S
1"
* °..





*. 400
kg/functional un





CRT Manufacturing - Primary Materials Inputs

20'1 ' 11-9 7.0 0.9 4.8 3.1
Monitor Tube Glass/frit PWB Japanese US grid
assembly grid
CRT mfg process groups
Figure 2-24. CRT manufacturing primary inputs

CRT Manufacturing - Ancillary Materials Inputs


.1.16
- • r. 1 - ««
Monitor Tube ~ Glass/frit FWB Japanese US grid
asserrbly grid
CRT mfg process groups
Figure 2-25. CRT manufacturing ancillary inputs

CRT Manufacturing - Fuel Inputs
353
1
0.1 4.9 ^Hj 0.4 0.0 0.0
Monitor Tube Glass/frit FWB Japanese US grid
assembly grid
CRT mfg process groups
Figure 2-26. CRT manufacturing fuel inputs


374
1
-i 	 ™^" 	 1
Fuels •







1.16
I
Fuels
-





68.9
Fuels





























   2-97

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
 Monitor
assembly
                              CRT Manufacturing - Electricity Inputs
                                        74
                                                          Japanese    US grid
                                                            grid
                    Tube     Glass/frit   !   FWB
                                      ;

                               CRT mfg process groups

                   Figure 2-27. CRT manufacturing electricity inputs
                                                 Fuels
CRT Manufacturing - Water inputs
- 12000 -,
kg/functional un
_&
35 ^^ 0 42 44
, Monitor Tube . Glass/frit • FWB Japanese
assembly • grid
CRT mfg process groups
Figure 2-28. CRT manufacturing water

10,500
1
US grid Fuels

l
inputs
a 16000
§ 14000
•« 1200°
e 10000
5  8000
o  6000
3  4000
S  2000
s     o
                               CRT Manufacturing - Total Energy
                                        15,200  -
                     19
                               237
                    Manitor
                   assembly
                                                   27
                      Tube     Glass/frit
                    FWB
                                                  Japanese
                                                     grid
                                       US grid

          CRT mfg process groups

Figure 2-29.  CRT manufacturing energy inputs
                                                                                2,790
                                                                Fuels
                                              2-98

-------












CRT Manufacturing
as 200 n
c
.2 100-
0
§ 50-
^ 0-


0 0.03 2.9
.Man'rtor Tube Glass/frit
asserrbly
CRTmfg

,«,
- Air Emissions



3.9&05 16 72.
	 , 	 ^ 	 , 	 asm 	 ,
FWB Japanese US grid
grid
process groups



157
••••
•
A
Fuels

Figure 2-30. CRT manufacturing air emissions

CRT Manufacturing - Wastewater Outputs
32 1000 -,
kg/functional ui
533
0 H 36
	 ^^B .»__*
Manitor Tube Glass/frit
asserrbly
CRT mfg

42 0 0
•^^^^
FWB Japanese US grid
grid
process groups
894
Fuels

Figure 2-31. CRT manufacturing wastewater outputs











kg/functional unit
5-1
4-
3-
2-
1 -
0-
CRT Manufacturing - Water Pollutant Outputs
4.7&00
O.OEtOO 1-2E01 BB 1.1E-04 8.9E-04 1.1 &03
Wbnitor Tube Qass/frit FWB Japanese US grid
asserrbly grid
CRT mfg process groups
Figure 2-32. CRT manufacturing water pollutant outputs
9.9E-05
Fuels
2-99

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
                        CRT Manufacturing - Hazardous Waste Outputs
                                            : 4.6&01
                                                                          6.2E-01
                                                       O.O&OO
                                     O.O&OO
                 Monitor
                assembly
Tube
Glass/frit
                           Japanese
                             grid
                                          US grid

              CRT mfg process groups

Figure 2-33. CRT manufacturing hazardous waste outputs
                                        Fuels
kg/functional unit
80-,
70-
60-
50-
40-
30-
20-
10-
CRT Manufacturing - Solid Waste Outputs
i
0.5 1-9 0.4 ' 4-1 1.1 1.3
^^ 	 I ____
Monitor Tube Glass/frit FWB Japanese US grid
assembly | grid
CRT mfg process groups
Figure 2-34. CRT manufacturing solid waste outputs
71.9
Fuels
       Among the utility inputs, both fuels and electricity were greatest in the glass/frit
manufacturing processes (Figures 2-26 and 2-27). The fuels, especially, are dominated by the
glass/frit process group, representing 83% of the mass of all the fuels in the manufacturing life-
cycle stage. It is LPG in this inventory that clearly dominates the fuel inputs at about 351
kg/functional unit (99% of the glass/frit fuel inputs) (Table 2-38).  Water inputs are greatest in
the fuel production processes, contributing 1,050 kg (or liter)/functional unit (Figure 2-28). Total
energy use from manufacturing is shown in Figure 2-29. The glass/frit manufacturing process
group contributes the greatest to the total energy impacts in the manufacturing stage.* A
sensitivity analysis will be conducted on the glass data and more details are provided in
Section 2.7.3.
       For outputs from the manufacturing stage, the mass of air emissions are dominated by
fuel production (Figure 2-30).  Individual materiial (pollutant) contributions for each process
group are presented in Table 2-39.  Wastewater outputs (i.e., the volume or mass of wastewater
released) are also greatest from the fuel production processes (Figure 2-31; Table 2-40);
however, the mass of chemical pollutants in the wastewater is greatest from the glass/frit
                                          2-100

-------
                                      2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
manufacturing process group (4.65 kg/functional unit) out of the total manufacturing stage mass
of pollutants, which is 4.77 kg/functional unit (Figure 2-32; Table 2-40).
       Hazardous wastes from the CRT manufacturing stage were a small portion of the overall
hazardous wastes generated by mass. Nonetheless, for purposes of future manufacturing stage
improvement assessments, Table 2-41 presents the individual material contributions for each
manufacturing process group; and Figure 2-33 shows that fuels production contributes the most
(0.62 kg/functional unit) hazardous waste to the manufacturing stage. Manufacturing solid
wastes are not as small a portion of the total mass of solid wastes throughout the CRT life-cycle
(42%) as hazardous wastes are, as was depicted in Figure 2-21.  Fuels production is the greatest
contributor to manufacturing-generated solid wastes  (71.9 kg/functional unit), followed by PWB
manufacturing (4.1 kg/functional unit) (Figure 2-34 and Table 2-43).
       Radioactive waste and radioactivity are directly related to the electricity generation
process and therefore, only the Japanese and U.S. electric grid processes generate these outputs
(some small radioactivity outputs are generated by fuels production processes) in the
manufacturing stage. (These outputs also occur in upstream processes that have an electric grid
included in the inventory.) Tables 2-44 and 2-45 show that more radioactive wastes and
radioactivity are from the Japanese grid. This is a result of more manufacturing processes as
modeled in this project being in Japan.

2.7.1.2 LCD inventory results

       The LCD inventory is presented similar to the CRT inventory above. The total LCD
inventory presented in Table 2-24 and Figures 2-5,2-6, and 2-7  shows the inventory from all life-
cycle stages combined.  The totals by life-cycle stage are presented in Table 2-45, Figure 2-35,
Figure 2-36, and Figure 2-37.

                      Table 2-45.  LCD inventory by life-cycle stage
Inventory type
Upstream
Mfg
Use
EOL
Total
Units*
Inputs
Primary materials
Ancillary materials
Water
Fuels
Electricity
Total energy
2.35e+02
1.06e+00
2.63e+02
1.47e+01
3.46e+01
6.33e+02
4.92e+01
2.04e+02
2.15e+03
2.58e+01
3.16e+02
1.44e+03
8.01e+01
1.29e+00
4.25e+02
0
8.53e+02
8.53e+02
-2.19e+00
2.11e+00
-l.SOe+01
-1.95e+00
1.62e-01
-8.446+01
3.62e+02
2.08e+02
2.82e+03
3.86e+01
1.20e+03
2.84e+03
kg
kg
kg
kg(orL)
MJ
MJ
Outputs
Air pollutants
Wastewater
Water pollutants
Hazardous waste
Solid waste
Radioactive waste
Radioactivity
1.12e+02
8.57e+00
4.60e-01
6.72e-03
1.31e+01
1.37e-04
1.20e+07
6.48e+01
3.12e+03
1.23e+00
4.64e+00
1.266+01
4.87e-04
1.02e+07
1.68e+02
0
2.62e-02
0
S.lle+Ol
8.52e-04
1.79e+07
1.30e+00
-2.41e+00
-4.09e-02
1.64e+00
-4.42e+00
1.62e-07
3.40e+03
3.46e+02
3.13e+03
1.68e+00
6.29e+00
5.236+01
1.48e-03
4.01e+07
kg
kg
kg(orL)
kg
kg
kg
Bq
 *Per functional unit (i.e.* one LCD monitor over its effective life)
                                         2-101

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
LCD Materials Inputs ;
... 2500-
§ 2000-
1 1500-
§ 1000-
o 500-
j|> -500 -
• Primary materials • Ancillary materials n Fuels Q Water
2,15
235 263 .204
., .,14.7 49.7 25.£
3
13 425
80-1 o.on ™
2.1
Upstream Mfg Use BOL ' -18.0
Life-cycle stage
Rgure 2-35. LCD mass-based material inputs by life-cycle stage
r f

LCD Energy Inputs
H Bectrfcity oTotal energy
*• 2000-,
= .cnn 1,440
f 1500- 	
•f 1000- 633
IT ~":' — '•
853 853
H 0.2
jg -500 J Upstream Mfg Use BDL
Life-cycle stage
Figure 2-36. LCD energy-based inputs by life-cycle stage
— i 	 1
-84.4

LCD Inventory Outputs
• Air pollutants • Water pollutants n Hazardous w aste 0 Solid w aste • Radioactive w aste
-: 200., ! 1
f 150- 112 |
1 1°°- • 13.1 "" 12-6 1
1 "• 1^00^04 m*"-™ \
jh -50 J Upstream Mfg
Life-cycle s
Figure 2-37. LCD mass-based mater

38
loo 1 •? 1-6
m.02S Fiil 8.5E-04 1 '^0.041
1.6E-07
Use BX-4.4
tage
ial outputs by life-cycle stage



                                   2-102

-------
                                      2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
       Considering inputs, Figure 2-35 shows that of the inputs measured in mass, the water
inputs constitute the majority of the inputs by mass for the entire life cycle, and most of the water
inputs are in the manufacturing life-cycle stage.  Details on each inventory type are provided
below. When considering which life-cycle stage contributes most to an inventory category, the
manufacturing stage has the largest inventory by mass for ancillary materials, fuels, and water
inputs. Primary material inputs are dominated by the upstream stages while electricity inputs are
dominated by the use stage. The total energy is dominated by the manufacturing life-cycle stage
(Figure 2-36). Note that LPG production from glass manufacturing does not dominate much of
the LCD inventory as it did for the CRT because of the smaller amount of glass used in the LCD
compared to the CRT.
       Of the outputs measured in mass (air emissions, wastewater, water pollutants and
hazardous, solid, and radioactive waste), wastewater constitutes the greatest output (Table 2-46);
however, wastewater alone is not used to  calculate impacts. Instead, water pollutants are used to
calculate impacts and therefore listed separately in the inventory. Of the remaining outputs
measured in mass (i.e., air emissions, water pollutants and hazardous, solid and radioactive
waste), which are shown in Figure 2-37, air emissions are the greatest contributor to the outputs.
Note again, as mentioned for the CRT, that radioactivity is measured in Bequerels (Bq) and
cannot be compared on the same scale.
       Considering each output type and  their contributions by life-cycle stage, the mass of water
pollutants is greatest in the manufacturing life-cycle stage, due to the fuel production processes
that support fuel consumption in the manufacturing processes being included in the
manufacturing life-cycle stage.  Wastewater and hazardous waste outputs are greatest in the
manufacturing stage; air emissions, solid waste, radioactive waste, and radioactivity have the
greatest contribution from the use stage. As with the CRT, all the output totals represented in
Table 2-45 include outputs to all dispositions.
       The tables and figures discussed above show the total inventories for particular input or
output types by life-cycle stage. Tables in Appendix J list each material that contributes to those
totals.  Figures 2-38 through 2-50 show the total contribution by life-cycle stage, based on the
entire input/output type-specific tables in  Appendix J. Summary tables for the LCD (Tables 2-46
through 2-54), developed from the Tables in Appendix J, show the top contributing inventory
items to each input or output type. Note that Table 2-48 includes input/output types that are
classified together as utilities: water, fuel, electricity and total energy.

LCD Primary Inputs

       Figure 2-38 shows that most LCD primary materials by mass are from the upstream life-
cycle stages.  The top 99.9% of the materials contributing to the total LCD primary input
inventory are shown in Table 2-46.  The largest material contributors are natural gas, coal and
petroleum (a combined 89% of all the primary LCD inputs), which are used to generate
electricity consumed throughout the life-cycle of the monitor.  Most of the electricity consumed
in the LCD life-cycle is in the manufacturing and use stages, as was seen in Figure 2-36.
However, most of the natural gas primary material reported in the materials processing stage (229
kg/functional unit) is not used to generate electricity, but is an ancillary material in the LCD
monitor/module manufacturing process. More detail on the processes that contribute greatest
within the manufacturing stage will  be presented after brief discussions of the life-cycle stage
breakdowns for each inventory type. For  the complete list of primary materials in the LCD

                                          2-103

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
inventory, the total mass, and the mass contribution of each life-cycle stage, see Appendix J,

Table J-10.                                !
         250
                                 LCD - Primary Materials
                                                                       -2.2
                 Upstream           Mfg  t            Use              EOL

                                        Life-cycle stage



                    Figure 2-38. LCD primary material inputs by life-cycle stage
       1250,
       *  2c
       Bl
       §  150-
       ti  100-


       I5:
1.1
                  Upstream
                                LCD - Ancillary Materials
                204
                                 1.3
                Mfg             Use

                   Life-cycle stage
                                                 2.1
EOL
                    Figure 2-39. LCD ancillary material inputs by life-cycle stage
                                         2-104

-------
                              2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
        Upstream
                              LCD - Fuels
                            26
                                             0.0
            Mfjg              Use
               Life-cycle stage

 Figure 2-40. LCD fuel inputs by life-cycle stage
                            LCD - Electricity
                                               853
                                                                0.2
                             Mfg               Use              EOL
                                Life-cycle stage

               Figure 2-41. LCD electricity inputs by life-cycle stage
-500-I
                           LCD - Total Energy
                             1,440
                                               853
             Mfg              Use
                Life-cycle stage

Figure 2-42. LCD energy inputs by life-cycle stage
                                 2-105

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
                                   LCD - Water Inputs
                 Upstream
                              Mijg              Use
                                 Life-cycle stage
                                                                       -18
                  EOL
                         Figure 2-43.  LCD water inputs by life-cycle stage
C 200-1
=1
« 150-J
£ 100
1  50
                       112
                   0.0
                                    LCD-Air Pollutants
                                                         168
                                         65
                          1.4E-04
0.0
                                                                      0.0
1.3
                  Upstream            Mfg              Use
                                         Life-cycle stage
                                          i
                          Figure 2-44. LCD air outputs by life-cycle stage
                                                                EOL
LCD - Wastewater Outputs
?= 3000-]
jf 2500-
g 2000-
5 1500-
^ 1000-
^ 500-
**T 0 -
2,524 ,
^H 597
8.6 0 Hi~ — I
j? -500 J Upstream Mig !
Life-cycle
Figure 2-45. LCD wastewater
0 0
H To surface w ater
E3 To treatment
-2.4 Q
Use EOL
stage
outputs by life-cycle stage
                                          2-106

-------
                                   2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
 *-  2
   -1 J
                              LCD - Water Pollutants
                              1.11
                                                     HTo surface water

                                                     E To treatment
                                                0.00  0.03
           Upstream
                        Mlg               Use


                           Life-cycle stage


      Figure 2-46. LCD water pollutant outputs by life-cycle stage
                                                         -0.04   4.97E-06
                       EOL
"5 3-

I 2

I 0
                              LCD - Hazardous Waste
                   To landfill uTo recycling/reuse aTo treatment uTo land (other)
8
   o  o  p
o  o  o  o
q  q  q  q
o"  o  o"  o
           Upstream
                       Mfg               Use

                           Life-cycle stage
                       EOL
                Figure 2-47. LCD hazardous waste outputs by life-cycle stage
                                      2-107

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
                                      LCD - Solid Waste
                                                    31.1
                                            • To landfill   ;
                                            E]To recycling/reuse
                                            • To treatment
                                                         025 0-030     0.3     1.95
                                                                         -6.6
        1
           -10 J
                Mfg   ;           Use
                   Life-cycle stage

Figure 2-48. LCD solid waste outputs by life-cycle stage
                                  LCD - Radioactive Waste
                                       (all to landfill)
                                                        8.5E-04
                                                                         1.6E-07
                   'Upstream    "        Mfg               Use              EOL
                                           Life-cycle stage

                      Figure 2-49.  LCD radioactive waste outputs by life-cycle stage
LCD - Radioactivity £
^
• 2.Et07-i *? £

§ 2.&07 -
"co
| 1.&07-
1 5.&06-
t 0 &00 -
WaCTW
-5.&06-
UJ H
f T
UJ
00
) T^

0> UJ
" ^3 '
•HH T—
^^^B o co ^ o
^H o o o o
^^H Hi uu uj LLI
^H o co ^r o
^^ o T-" -^-" o
Upstream

Figure 2-50. LCD
S^
?
UJ , UJ UJ
i- , CM O
i- OJ O
•* i o> o







IB To landfill
QTo treatment
QTo surface water

D To air

o m o co 
-------
                                      2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-46. Top 99.9% of LCD
Material
Natural gas (in ground)
Coal, average (in ground)
Petroleum (in ground)
Natural gas
Assembled LCD monitor
Iron (Fe, ore)
Steel
Assembled 15" LCD backlight unit
LCD module
Polycarbonate resin
Bauxite (A12O3, ore)
Iron scrap
LCD glass
Poly(methyl methacrylate)
15" LCD light guide
PWB-laminate
Printed wiring board (PWB)
Styrene-butadiene copolymers
PPE
Cables/wires
LCD front glass (with color filters)
Aluminum (elemental)
Sand
Recycled LCD glass
Upstream
2.29E+02
1.72E+00
7.09E-01
0
0
3.26E+00
0
0
0
0
0
4.63E-01
0
0
0
0
0
0
0
0
0
0
0
0
primary material inputs (kg/functional unit)
Mfg
5.16E+00
8.03E+00
2.23E+01
4.22E+00
0
0
2.53E+00
1.48E+00
1.18E+00
5.16E-01
5.09E-01
0
4.52E-01
3.83E-01
3.74E-01
3.74E-01
3.74E-01
3.62E-01
3.00E-01
2.34E-01
1.78E-01
1.34E-01
1.11E-01
9.54E-02
Use
0
6.69E+01
1.42E+00
5.22E+00
6.50E+00
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
i EOL
i -1.08E+00
; 1.27E-02
1 -l.OOE+00
i -5.75E-02
! o
: 0
! o
0
! 0
: o
: o
i o
' o
i 0
; o
0
o
! 0
i o
i o
0
0
: o
! 0
Total
. 2.33E+02
7.67E+01
2.34E+01
9.39E+00
6.50E+00
3.26E+00
2.53E+00
1.48E+00
1.18E+00
5.16E-01
5.09E-01
4.63E-01
4.52E-01
3.83E-01
3.74E-01
3.74E-01
3.74E-01
3.62E-01
3.00E-01
2.34E-01
1.78E-01
1.34E-01
1.11E-01
9.54E-02
% of total
64.25%
21.15%
6.45%
2.59%
1.79%
0.90%
0.70%
0.41%
0.33%
0.14%
0.14%
0.13%
0.12%
0.11%
0.10%
0.10%
0.10%
0.10%
0.08%
0.06%
0.05%
0.04%
0.03%
0.03%
LCD Ancillary Inputs                                      \

       As presented in Figure 2-39, the greatest mass of ancillary LCD inputs is in the
manufacturing life-cycle stage at approximately 204 kg/functional unit. Table 2-47 shows that
liquified natural gas (LNG) contributes about 93% to this total. It is in the LCD module/monitor
manufacturing process where this large amount of LNG was repbrted as an ancillary material (to
be discussed below). Note that this is separate from LNG reported as a fuel, and LNG as an
ancillary material is not used to calculate energy impacts in the LCIA. Following LNG is
nitrogen at about 3% and clay at less than 1% of the total ancillary materials by mass. Excluding
LNG from the inventory, nitrogen constitutes about 50% and clay 14% of the total ancillary
materials in the LCD life-cycle. The contributions from the manufacturing stage will be
discussed in further detail below.  See Table J-ll in Appendix J|for the complete list of ancillary
materials in the LCD inventory.                             ;
                                         2-109

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
        Table 2-47. Top 99% of LCD ancillary material inputs (kg/functional unit)
Material
LNG
Nitrogen
Clay (in ground)
Limestone
Sand (in ground)
Sodium hydroxide
Hydrogen
Lime
Sodium chloride (NaCl, in ground or sea)
Limestone (CaCOS, in ground)
Isopropyl alcohol
Sulfuric acid
Upstream
0
0
1.30e-03
0
9.55e-03
0
0
0
4.37e-01
5.07e-01
0
0
Mfg
1.94e+02
6.02e+00
0
l.OSe-01
1.326-03
4.45e-01
4.44e-01
4.74e-02
6.08e-04
5.49e-02
3.496-01
3.25e-01
Use
0
0
0
8.996-01
0
0
0
3.95e-01
0
0
0
0
EOL
-0
0
1.69e+00
1.71e-04
5.60e-01
0
0
7.49e-05
1.08e-05
-1.556-01
0
0
Total
1.94e+02
6.02e+00
1.69e+00
l.Ole+00
5.716-01
4.45e-0i
4.446-01
4.42e-0l
4.38e-01
4.06e-01
3.49e-01
3.25e-01
% of total
93.19%
2.89%
0.81%
0.48%
0.27%
0.21%
0.21%
0.21%
0.21%
0.20%
0.17%
0.16%
LCD Utility Inputs                         \

       Utility inputs in the LCD life-cycle are presented in Table 2-48 and include fuel
(kg/functional unit), electricity (Mr/functional unit), water inputs (kg or L/functional unit).., and
total energy (Ml/functional unit; a combination of fuel and electricity inputs). Table; 2-48 and
Figure 2-40 show that most fuels (26 kg/functional unit) are used in the manufacturing stage.
This represents 67% of the total fuels. LPG (16.8 kg/functional unit) dominates the total fuel
inputs at 44% of all the fuels in the LCD life-cycle.  More detail as to the breakdown by process
within the manufacturing stage will be presented below after each input/output type is discussed.
       Electricity inputs are dominated by the use stage (853 Ml/functional unit), followed by
the manufacturing stage (316 MJ/functional unit) (see Figure 2-41).  When fuel energy and
electrical energy are combined into a total energy input value, the overall energy from
manufacturing exceeds that from the use stage (1,440 MJ/functional unit versus 853
Ml/functional unit). This is also depicted in Figure 2-42.
       The other utility considered in Table 2-48 is water (Figure 2-43).  Approximately 76%
(2,150 L/functional unit) of the water inputs in the LCD life-cycle are from the manufacturing
processes. The life-cycle stage contributing the next most to water inputs is the use stage at 15%
(425 L/functional unit). The upstream contributes about 9% (263 L/functional unit). Table J-12
in Appendix J provides the complete list of inventory items for the LCD.
                                         2illO

-------
                                       2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
.- 	 	 Table 2-48. f ,CD ntilitv in&ifs
Material
Fuels (kg/functional unit):
LPG
Natural gas (in ground)
LNG
Coal, average (in ground)
Petroleum (in ground)
Kerosene
Coal, lignite (in ground)
Natural gas
Steam
Fuel oil #6
Fuel oil #2
Jranium (U, ore)
Fuel oil #4
Total fuels
Upstream

0
1.03E+01
0
2.49E+00
1.52E+00
0
4.10E-01
0
0
0
0
, 7.86E-05
0
1.47E+01
Mfg

1.68E+01
2.41E+00
3.22E+00
6.86E-01
4.84E-01
4.65E-01
0
1.16E+00
1.45E-01
1.25E-01
5.42E-02
1.15E-05
2.11E-01
2.58e+01
Use

0
0
0
o'
0
0
0
0
0
0
0
0
0
0
EOL

1.38E-03
-1.38E-01
0
-7.66e-03
-3.81e-02
0
0
-8.61E-01
0
0
0
0
-9.096-01
-1.95E+00
Total

1.68E+01
1.26E+01
3.22E+00
3.17E+00
1.96e+00
4.65e-01
4.10e-01
3X)le-01
1.45e-01
1.25e-01
5.42e-02
9.01e-05
-6.99e-01
3.86e+01
% of total

43.63%
32.66%
8.34%
8.21%
5.08%
1.21%
1.06%
0.78%
0.37%
0.33%
0.14%
<0.01%
-1.81%
100.00%
Electricity (Ml/functional unit):
Electricity
3.46e+01
3.16e+02
8.53e+02
1.62e-01
1.20e+03

Water (kg or L/functional unit):
Water
2.63E+02
2.15E+03
4.25E+02
-1.80E+01
2.82e+03

Total energy (fuels and electricity. Ml/functional unit):
Energy
6.33E+02
1.44E+03
8.53E+02
-8.44E+01
2.84E+03

LCD Air Outputs

       Air emissions from the LCD life-cycle are greatest (by mass) in the use stage as seen in
Figure 2-44. This indicates that most air emissions by mass are from the generation of electricity
used by consumers of the monitors. Forty-nine percent of the total life-cycle air emissions by
mass (or about 168 kg/functional unit) are from the use stage. Carbon dioxide (COj) emissions
from the use stage alone constitute about 166 kg/functional unit, or almost 48% of all air
emissions by mass in the life-cycle and nearly 99% of the use stage air emissions.  The remaining
air emissions that contribute to the top 99.99% of air emissions are presented in Table 2-49 and
the complete list of air emissions are presented alphabetically in Table J-13 in Appendix J.  The
appendix also provides life-cycle stage subtotals. The next largest air emissions, by life-cycle
stage, are emitted during the upstream stages, which contribute about 32% to the total life-cycle
air emissions. All the air emissions in the inventory except for ethylacetate and methyl ethyl
ketone from the manufacturing stage (a combined 1.36 x 10"4 kg/functional unit) were reported as
being emitted directly to the air (see Appendix J, Table J-13). Only those materials directly
released to the air are used to calculate impacts.  This will be discussed further in Chapter 3.
                                         2-111

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
            Table 2-49. Top 99.99% of LCDl air emissions (kg/functiomai unit)
Material
Carbon dioxide
Methane
Nitrogen oxides
Sulfur dioxide
Tetramethyl ammonium
Other organics
Carbon monoxide
Nitrogen fluoride
Nonmethane hydrocarbons
Hydrochloric acid
Benzene
PM
Ammonia
Phosphine
Hydrofluoric acid
Sulfur oxides
Unspecified LCD process
Cr-etchant, unspecified
Nitrogen dioxide
PM-10
Isopropyl alcohol
Hydrocarbons, remaining
Al-etchant, unspecified
Disposition
Air
Air
Air
Air
Air
Air
Air
Adr
Air
Air
Air
Air
Air
Air
Air
Air
Ak
Air
Air
Air
Air
Ak
Air
Upstream
1.07e+02
3.546+00
6.56e-01
4.63e-02
0
4.45e-01
3.74e-01
0
2.07e-01
1.50e-03
8.85e-02
9.16e-02
1.12e-02
!o
2.27e-04
2.57e-02
0
0
3.086-02
3.45e-07
0
9.30e-03
0
Mfg
6.22e+01
1.22e-01
7.62e-01
2.95e-01
6.43e-01
1.35e-02
3.85e-02
2.45e-01
8.87e-03
6.58e-02
2.70e-03
6.99e-03
6.26e-02
6.26e-02
5.27e-02
4.07e-02
4.49e-02
4.12e-02
0
6.85e-03
1.78e-02
7.75e-03
1.37e-02
Use
1.66e+02
2.41e-01
4.39e-01
9.30e-01
0
0
3.02e-02
0
0
4.02e-02
4.36e-05
0
0
0
5.02e-03
0
0
0
0
2.16e-02
0
0
0
EOL
1.39e+00
-2.82e-02
-1.36e-02
2.96e-04
0
-2.41e-03
-3.456-03
0
-1.26e-03
-6.886-04
-4.82e-04
-1.24e-02
-6.95e-05
0
-1.47e-04
-1.93e-02
0
0
4.46e-04
3.43e-06
0
-6.51e-04
0
Total
3.36e+02
3.876+00
1.84e+00
1.27e+00
6.43e-01
4.56e-01
4.39e-01
2.45e-01
2.15e-01
1.07e-01
9.07e-02
8.62e-02
7.37e-02
6.26e-02
5.78e-02
4.71e-02
4.49e-02
4.12e-02
3.126-02
2.84e-02
1.78e-02
1.64e-02
1.37e-02
% of total
97.18%
1.12%
0.53%
0.37%
0.19%
0.13%
0.13%
0.07%
0.06%
0.03%
0.03%
0.02%
0.02%
0.02%
0.02%
0.01%
0.01%
0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
LCD Water Outputs

       The volume (or mass) of wastewater released throughout the LCD life-cycle is
approximately 3,128 L (kg) per functional unit.  Approximately 19% of that is sent to treatment
as opposed to direct discharge to surface water (81%; Figure 2-45).  The mass of chemical
pollutants within the wastewater streams were calculated separately from the total wastewater
volume.  The total mass of water pollutants released, presented by life-cycle stage are shov/n in
Figure 2-46. Of the small amount of water pollutants released, the manufacturing life-cycle stage
contributes the greatest with approximately 1.23 kg per functional unit. This is about 73% of all
the water pollutants for the entire life-cycle. The upstream stages have the second greatest mass
of water pollutants at nearly 0.46 kg/functional unit (27%). The use and EOL stages are small
contributors, with the EOL being negative due to recovery processes within the EOL stage. To
see the top 99% contributors to the water pollutant quantities, Table 2-50 reveals that chloride
and sodium ions contribute nearly 61% to all the water pollutants in the life-cycle, mostly from
the manufacturing and upstream stages. For the complete inventory, listing water pollutants
alphabetically and by life-cycle stage, see Appeiidix J, Table J-14. Further details on the
manufacturing stage will be provided later.
                                          2-112

-------
                                       2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-50. To
Material
Chloride ions
Sodium (+1)
Dissolved solids
COD
Nitrogen
Suspended solids
BOD
COD
BOD
Sulfate ion (-4)
Sulfate ion (-4)
Fluorides (F-)
Nitrogen
Phosphorus (yellow or white)
Waste oil
Suspended solids
Phosphorus (yellow or white)
Colon bacillus (bacteria in
large intestine)
Oil & grease
p 99% of LCD water pollutant outputs (kg/functional unit)
Disposition
surface water
surface water
surface water
surface water
surface water
surface water
treatment
treatment
surface water
surface water
treatment
surface water
treatment
treatment
surface water
treatment
surface water
surface water
treatment
Upstream
2.33e-01
1.73e-01
3.21e-03
7.67e-03
1.44e-05
4.98e-03
0
0
7.72e-04
2.40e-02
0
5.14e-05
0
0
1.75e-03
0
1.92e-06
0
0
Mfg
3.12e-01
3.41e-01
1.75e-01
8.'20e-02
7.98e-02
5.80e-02
5.74e-02
3.90e-02
2.79e-02
2.94e-03
1.32e-04
1.29e-02
1.26e-02
6.91e-03
4.876-03
5.60e-03
4.33e-03
3.89e-03
3.61e-03
Use
0
0
0
0
0
0
0
0
0
0
2.55e-02
0
0
0
0
6.65e-04
0
0
0
EOL
-1.58e-02
-2.03e-02
-5.69e-05
-2.69e-03
0
-1.44e-03
0
0
-3.18e-04
-1.20e-06
4.84e-06
-5.01e-06
0
0
-2.06e-04
1.26e-07
0
0
0
Total
5.29e-01
4.94e-01
1.78e-01
8.70e-02
7.98e-02
6.15e-02
5.74e-02
3.90e-02
2.83e-02
2.69e-02
2.57e-02
1.30e-02
1.26e-02
6.91e-03
6.41e-03
6.26e-03
4.33e-03
3.89e-03
3.61e-03
% of total
31.53%
29.42%
10.63%
5.18%
4.76%
3.66%
3.42%
2.33%
1.69%
1.60%
1.53%
0.77%
0.75%
0.41%
0.38%
0.37%
0.26%
0.23%
0.21%
LCD Hazardous Waste Outputs

       The total mass of hazardous waste generated throughout the life-cycle of the LCD is
about 6.29 kg/functional unit. Figure 2-47 shows that this is mostly from the manufacturing
stage, which contributes 4.64 kg/functional unit, or almost 74%.  The EOL stage hazardous waste
outputs equal 1.64 kg/functional unit, or 26%.  The disposition of the waste will be used to
determine how impacts are calculated in Chapter 3.  Only hazardous wastes sent to landfills are
directly calculated as impacts, which will be presented and discussed in Chapter 3. Figure 2-47
shows what portion of hazardous wastes are landfilled, recycled/reused, treated or otherwise
land-applied. Nearly all of the hazardous waste in the manufacturing stage (~99%) is
recycled/reused or treated. Less than 1% of the hazardous waste from the manufacturing stage is
landfilled. Nearly all the hazardous waste from the EOL stage is landfilled. Table 2-51 shows
the top contributors to the LCD life-cycle. Note that multiple entries of a material are due to
different dispositions for that material.  See Table J-15 in Appendix J for the complete
alphabetical inventory of hazardous waste outputs.  Additional detail on the manufacturing stage
are presented below.
                                          2-113

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
        Table 2-51. Top 99% of LCD hazardous waste outputs (kg/functional unit)
Material
Isopropyl alcohol
EOLLCD Monitor, landfiUed
Waste acid (mainly HF)
Thinner, unspecified
Remover, unspecified
Sodium sulfate
Isopropyl alcohol
Tetramethyl ammonium
hydroxide
Waste acid (mainly HF)
PWB-Waste cupric etchant
Remover, unspecified
Hazardous waste, unspecified
Rinse, unspecified
Spentsolvent(non-halogenated)
Hazardous waste, unspecified
Waste acids, unspecified
Unspecified sludge
PWB-Solder dross
Acetone
Waste solvent (photoresist)
Waste solvent (photoresist)
Spent solvent (with halogenated
materials)
Phosphoric acid
Disposition
treatment
landfill
recycling/reuse
treatment
treatment
recycling/reuse
recycling/reuse
recycling/reuse
treatment
recycling/reuse
recycling/reuse
treatment
recycling/reuse
treatment
landfill
recycling/reuse
land (other than
landfill)
recycling/reuse
treatment
treatment
recycling/reuse
treatment
landfill
Upstream
0
0
0
0
0
0
0
0
0
0
o
0
0
0
6.72E-03
0
0
0
0
0
0
0
0
Mfg
1.91E+00
0
5.69E-01
5.40E-01
3.03E-01
2.44E-01
1.69E-01
1.42E-01
1.36E-01
9.93E-02
8.84E-02
6.16E-02
4.67E-02
4.66E-02
2.97E-02
3.24E-02
3.09E-02
2.96E-02
2.77E-02
2.17E-02
2.05E-02
1.55E-02
1.44E-02
Use
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
, 0
0
0
0
0
0
EOL
0
1.64E+00
0
0
0
0
0
0
0
0
0
0
0
0
-1.05E-03
0
0
0
0
0
0
0
0
Total
1.91E+00
1.64E+00
5.69E-01
5.40E-01
3.03E-01
2.44E-01
1.69E-01
1.42E-01
1.36E-01
9.93E-02
8.84E-02
6.16E-02
4.67E-02
4.66E-02
3.54E-02
3.24E-02
3.09E-02
2.96E-02
2.77E-02
2.17E-02
2.05E-02
1.55E-02
1.44E-02
%of
total
30.41%
26.13%
9.04%
8.57%
4.81%
3.89%
2.69%
2.26%
2.15%
1.58%
1.40%
0.98%
0.74%
0.74%
0.56%
0.52%
0.49%
0.47%
0.44%
0.35%
0.33%
0.25%
0.23%
LCD Solid Waste Outputs   .                ;

       Figure 2-48 shows that the use stage contributes the most amount (31 kg/functional unit)
of solid waste by mass to the LCD life-cycle, and 100% of that waste is landfilled. The
manufacturing stage contributes 12.6 kg/functional unit. In terms of mass, the greatest material
contributors to the solid waste outputs for the LCD life-cycle are coal waste (-41%), followed by
dust/sludge (16%). Most of this is from the generation of electricity in the use stage (Table 2-
52). Note also that the mass of an LCD monitor that is assumed to be landfilled (0.89
kg/functional unit) is 1.7% of the total mass of solid waste in the LCD life-cycle. See Appendix
J, Table J-16 for the complete LCD solid waste inventory. The manufacturing stage breakdown
will be discussed at the end of this section.
                                         2-114

-------
                                      2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
           Table 2-52. Top 99% of LCJ) solid waste outputs (kg/functional unit)
Material
Coal waste
Dust/sludge
Fly/bottom ash
Unspecified solid waste
Slag and ash
Unspecified solid waste
Unspecified solid waste
Iron scrap
EOL LCD Monitor,
incinerated
EOL LCD Monitor, recycled
EOL LCD Monitor, .
remanufactured
EOL LCD Monitor, landfilled
Unspecified sludge
Waste LCD glass
Unspecified waste
CARBON STEEL SCRAP
Waste plastic from LCD
modules
Polycarbonate
Waste alkali, unspecified
Waste acid (containing F and
detergents)
Waste LCD glass
Mineral waste
Mining waste
Slag and ash
Waste acids, unspecified
Mixed industrial (waste)
Waste alkali (color filter
developer, unspecified)
Disposition
landfill
landfill
landfill
landfill
recycle/reuse
recycle/reuse
treatment
recycle/reuse
treatment
recycle/reuse
recycle/reuse
landfill
recycle/reuse
recycle/reuse
recycle/reuse
recycle/reuse
treatment
recycle/reuse
recycle/reuse
landfill
landfill
landfill
landfill
landfill
treatment
landfill
recycle/reuse
Upstream
0
0
0
2.40E+00
8.02E+00
1.50E+00
0
1.67E-01
0
0
0
0
0
0
4.05E-01
0
0
0
0
0
0
2.20E-01
1.41E-01
8.19E-02
0
4.34E-02
0
Mfg
2.28E+00
8.80E-01
5.70E-01
0
3.40E+00
2.11E-01
1.63E+00
0
0
0
0
0
8.46E-01
7.20E-01
1.72E-01
0
4.03E-01
0
3.23E-01
2.70E-01
2.63E-01
1.26E-04
0
3.49E-02
1.05E-01
4.83E-02
8.91E-02
Use
1.90E+01
7.34E+00
4.75E+00
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
EOL
3.60E-03
1.39E-03
9.01E-04
-5.10E-01
-9.67E+00
0
0
1.10E+00
9.75E-01
9.75E-01
9J5E-01
8.94E-01
0
0
-9.83E-03
4.58E-01
0
3.90E-01
0
0
0
-4.46E-06
-1.23E-06
-1.99E-03
0
-1.35E-03
0
Total
2.13E+01
8.23E+00
5.32E+00
1.89E+00
1.75E+00
1.71E+00
1.63E+00
1.27E+00
9.75E-01
9.75E-01
9.75E-01
8.94E-01
8.46E-01
7.20E-01
5.67E-01
4.58E-01
4.03E-01
3.90E-01
3.23E-01
2.70E-01
2.63E-01
2.21E-01
1.41E-01
1.15E-01
1.05E-01
9.04E-02
8.91E-02
%of
total
40.64%
15.72%
10.16%
3.62%
3.35%
3.27%
3.11%
2.42%
1.86%
1.86%
1.86%
1.71%
1.62%
1.38%
1.08%
0.88%
0.77%
0.75%
0.62%
0.52%
0.50%
0.42%
0.27%
0.22%
0.20%
0.17%
0.17%
LCD Radioactive Waste Outputs

       Radioactive waste outputs in the LCD inventory are limited to the electricity generation
and steel production processes, with steel production processes accounting for only about 9% of
the total. Therefore, radioactive wastes will be found wherever electricity is used in a process in
the LCD life-cycle. Only very small amounts (approximately 0.0015 kg/functional unit) of
radioactive waste are generated over the entire life-cycle of the LCD (Figure 2-49 and Table 2-
53).  As expected, the majority of this is linked to the use stage, where most electricity is used in
the LCD life-cycle, followed by the manufacturing stage. Low-level radioactive waste (78%) and
depleted uranium (21%) are most of the waste, with negligible amounts of highly radioactive
waste and unspecified radioactive waste.  The inventory of radioactive waste outputs is small,
                                         2-115

-------
             2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
             and therefore, Table 2-53 lists all material outputs associated with radioactive waste, in
             descending order of quantity. Table J-17 in Appendix J lists these in alphabetical order.
Table 2-53. LCD radioactive waste outputs (kg/functional unit) i
Material
Low-level radioactive waste
Uranium, depleted
Radioactive waste (unspecified)
Highly radioactive waste (Class C)
Total radioactive wastes
Disposition
landfill
landfill
landfill
landfill

Upstream
1.28E-04
0
5.77E-06
2.72E-06
1.37E-04
Mfg
3.74E-04
1.12E-04
0
0
4.87E-04
Use
6.56E-04
1.97E-04
0
0
8.52E-04
EOL
1.24E-07
3.73E-08
0
0
1.62E-07
Total
1.16E-03
3.09E-04
5.77E-06
2.72E-06
1.48E-03
%of
total
78.49%
20.93%
0.39%
0.18%
100.00%
             LCD Radioactivity Outputs

                    Radioactivity is also inventoried in this project as isotopes that are released to the
             environment. Radioactivity is measured in Bequerels and may be released to air, water, or land,
             or may also be treated. The quantity of radioactivity for each life-cycle stage and different
             dispositions are presented in Figure 2-50.  Table 2-54 shows the top contributors to the total
             radioactivity outputs.  Radioactivity outputs are associated with the generation of electricity and,
             therefore, the greatest quantity of radioactivity is from the use stage. Appendix J, Table J-18 lists
             the complete inventory.

                      Table 2-54. Top 99.9% of LCD radioactivity outputs (Bq/functional unit)
Material
Molybdenum-99 (isotope)
Plutonium-241 (isotope)
Tritium-3 (isotope)
Xenon-133 (isotope)
Xenon-133M (isotope)
Plutonium-240 (isotope)
Cesium-135 (isotope)
Radon-222 (isotope)
Plutonium-239 (isotope)
Xenon-133 (isotope)
Tritium-3 (isotope)
Krypton-85 (isotope)
Disposition
treatment
landfill
treatment
air
air
landfill
landfill
air
landfill
treatment
air
air
Upstream
0
1.18e+07
0
7.63e+02
0
5.08e+04
4.59e+04
4.30e+04
3.57e+04
0
1.09e+02
5.45e+01
Mfg.
l.Ole+07
0
5.96e+04
4.98e+03
6.59e+04
0
0
0
0
9.43e+03
7.98e+03
5.64e+03
Use
1.76e+07
0
1.04e+05
1.16e+05
7.73e+03
0
0
0
0
1.65e+04
1.40e+04
9.88e+03
EOL
3.35e+03
0
1.98e+01
2.21e+01
1.47e+00
0
0
0
0
3.13e+00
2.65e+00
1.88e+00
Total
2.77e+07
1.18e+07
1.64e+05
1.22e+05
7.36e+04
5.08e+04
4.59e+04
4.30e+04
3.57e+04
2.60e+04
2.21e+04
1.56e+04
% of total
69.09%
29.33%
0.41%
0.30%
0.18%
0.13%
0.11%
0.11%
0.09%
0.06%
0.05%
0.04%
-
                                                       2-116

-------
                                      2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
LCD Manufacturing Stage             «

       The inventory tables that show the specific materials in each inventory (i.e., those in
Appendix J and Tables 2-46 through 2-54) are the sums of the materials from one or more
processes within a life-cycle stage. To burrow down deeper into the data, the manufacturing
stage inventory data are broken down by process or group of processes.  Similar to the CRT
analysis, groups of processes were combined where fewer than three companies provided data for
a process or where confidentiality agreements precluded presenting individual process data
(Table 2-55).  Burrowing further into the contributing processes or process groups is necessary
for future manufacturing improvement assessments.

                            Table 2-55. LCD process groups
Process group
Monitor/module
Panel components
LCD glass
Backlight
PWB
Japanese grid
U.S. grid
Fuels
Process(es) included
panel/module manufacturing, monitor assembly
polarizer manufacturing, patterning color filters on glass, liquid crystal manufacturing
LCD glass manufacturing
backlight unit assembly, backlight light guide, cold cathode fluorescent lamp manufacturing
PWB manufacturing
electricity generation - Japanese electric grid
electricity generation - U.S. electric grid
production of fuel oils #2, #4 and #6, LPG, and natural gas
       Tables 2-56 through 2-64 list the specific inventories for each process group in the
manufacturing stage for each input and output type. Figures 2-51 through 2-61 graph the total
inventories for each process group for each input and output type. As revealed in Table 2-45,
ancillary material inputs, fuels, water and total energy inputs all were greatest in the
manufacturing stage.  Similarly, wastewater, water pollutant and hazardous waste outputs were
also greatest in the manufacturing stage. Similar to the discussion for CRTs, the manufacturing
stage inventories by process group, for all input and output types, are presented here to reveal
more specifics about the inventory and to allow manufacturers to conduct improvement
assessments.
                                         2-117

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
            Table 2-56. LCD manufacturing stage primary material Inputs
Material
Process Group
Module/Monitor
1,4-butanolide
l-memyl-2-pyrrolidinone
2-(2-butoxyethoxy)-ethanol acetate
AINd
Aluminum (elemental)
Assembled 15" LCD backlight unit
Cables/wires
Glycol ethers
Indium tin oxide
LCD front glass (with color filters)
LCD glass
LCD material (confidential)
LCD module
LCD spacers, unspecified
Liquid crystals, for 15" LCD
Mild fiber
Molybdenum
MoW
Polarizer
Polycarbonate resin
Polyimide alignment layer, unspecified
PPE
Printed wiring board (PWB)
Solder (60% tin, 40% lead)
Steel
Styrene-butadiene copolymers
Titanium
Triallyl isocyanurate
Triphenyl phosphate
Unspecified LCD material
Total
Panel Components
3,4,5-trifluorobromobenzene
3,4-difluorobromobenzene
4-4(-propylcyclohexyl)cyclohexanone
4-bromophenol
4-ethylphenol
4-pentylphenol
4-propionyIphenol
LCD glass
Quantify (kg/
functional unit)


4.06e-04
;• 4.06e-04
8.08e-06
i 2.97e-05
l.Ole-01
1.48e+00
2.30e-01
4.06e-04
5.26e-04
1.78e-01
; 2.16e-01
; 3.11e-04
1.18e+00
1.69e-05
1.24e-03
7.34e-07
5 1.78e-04
9.09e-04
; 4.07e-02
4.01e-01
4.86e-04
3.00e-01
! 3.74e-01
3.81e-02
2.50e+00
3.62e-01
1.33e-04
1.54e-05
9.25e-02
[ 1.19e-04
7.50e+00

'-, 2.64e-04
^ 3.65e-04
2.18e-04
3.27e-04
7.006-05
: 3.42e-04
1.94e-04
2.36e-01
% of process
group total


0.01%
0.04%
<0.01%
<0.01%
1.34%
19.67%
3.07%
0.01%
0.01%
2.38%
2.88%
<0.01%
15.78%
<0.01%
0.02%
<0.01%
<0.01%
0.01%
0.54%
5.35%
0.01%
4.00%
4.98%
0.51%
33.38%
4.82%
<0.01%
<0.01%
1.23%
<0.01%
100.00%

0.08%
0.04%
0.07%
0.10%
0.02%
0.11%
0.06%
74.75%
% of grand total
>



f



























15.26%









                                    2-J118

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-56. LCD manufacturing stage primary material inputs
Material
Process Group
Pigment color resist, unspecified
Polyester adhesive
Polyethylene terephthalate
Polyvinyl alcohol
Total
Glass mfg
Barium Carbonate
Glass, unspecified
Potassium Carbonate
Recycled LCD glass
Sand
Sodium Carbonate
Strontium Carbonate
Zircon Sand
Total
Backlight
15" LCD light guide
Aluminum (elemental)
Argon
Backlight lamp (CCFL)
Cables/wires
Glass, unspecified
Mercury
Metals, remaining unspeciated
Neon
Poly(methyl methacrylate)
Polycarbonate resin
Polyethylene terephthalate
Rubber, unspecified
Steel
Total
PWB
PWB-laminate
Solder (63% tin; 37% lead)
Total
Japanese Grid
Coal, average (in ground)
Natural gas
Petroleum (in ground)
Uranium, yellowcake
Total
Quantity (kg/
functional unit)

3.72e-02
6.25e-04
3.14e-02
8.61e-03
3.16e-01

1.37e-02
2.28e-03
1.75e-02
9.54e-02
l.lle-01
2.26e-02
1.53e-02
2.51e-03
2.81e-01

3.74e-01
3.35e-02
3.53e-05
1.94e-03
3.43e-03
4.14e-02
3.99e-06
6.81e-04
6.31e-05
3.83e-01
1.14e-01
2.74e-02
6.01e-04
2.52e-02
l.Ole+00

3.74e-01
2.24e-02
3.96e-01

7.69e+00
4.20e+00
4.33e+00
1.02e-03
1.62e+01
% of process
group total

11.80%
0.20%
9.96%
2.73%
100.00%

4.90%
0.23%
6.24%
34.00%
39.64%
8.05%
5.47%
0.90%
99.42%

37.18%
3.35%
<0.01%
0.19%
0.34%
4.11%
<0.01%
0.07%
0.01%
38.12%
11.38%
2.72%
0.06%
2.50%
100.02%

94.35%
5.66%
100.00%

47.41%
25.89%
26.69%
0.01%
100.00%
% of grand tota































2.04%



0.81%





32.96%
  2-119

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
            Table 2-56. LCD manufacturing stage primary material inputs
Material
Process Group
U.S. Grid
Coal, average (in ground)
Natural gas
Petroleum (in ground)
Uranium, yellowcake
Total
Fuel Production
Natural gas (in ground)
Petroleum (in ground)
Total
Grand Total
Quantity (kg/
functional unit)


3.47e-01
2.71e-02
! 7.36e-03
9.39e-06
3.82e-01

5.16e+00
1.79e+01
2.31e+01
4.92e+01
% of process
group total


90.97%
7.10%
1.93%
<0.01%
100.00%

22.33%
77.67%
100.00%

% of grand total






0.78%



46.94%
100.00%
                                    2-il20

-------
                      2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-57. LCD manufacturing stage ancillary material inputs
Material
Process Group
Module/Monitor
1,4-butanolide
l-Methoxy-2-propanol
2-(2-butoxyethoxy)-ethanol
2,2,4-trimethylpentane
2-ethoxyl ethylacetate
Acetic acid
Acetone
Al-etchant, unspecified
Aluminum sulfate
Ammonia
Ammonium bifluoride
Ammonium fluoride
Ammonium hydroxide
Argon
Calcium hydroxide
Carbon dioxide
Chlorine
Cleaner, unspecified
Cresol-formaldehyde resin
Cr-etchant, unspecified
Cyclohexane
Dimethylsulfoxide
Ethanol
Ethanol amine
Ferric chloride
Fluorocarbon resin
Flux, unspecified
Glycol ethers
Helium
Hexamethyldisilizane
Hydrochloric acid
Hydrofluoric acid
Hydrogen
Hydrogen peroxide
Isopropyl alcohol
no etchant, unspecified
Krypton
LNG
Methyl ethyl ketone
Monosilane
Quantity (kg/
functional unit)


4.04e-05
1.10e-02
3.04e-02
1.52e-05
. 1.78e-03
6.40e-03
1.03e-02
5.88e-03
1.05e-01
1.55e-02
2.36e-03
1.14e-02
5.15e-06
7.87e-03
1.39e-01
3.74e-05
1.55e-02
1.47e-04
8.29e-04
1.77e-02
2.03e-05
6.63e-02
1.35e-02
7.85e-02
8.92e-03
3.386-06
7.35e-05
2.12e-02
6.18e-04
2.58e-04
4.316-02
4.21e-02
4.44e-01
1.47e-04
3.49e-01
2.94e-03
2.58e-05
1.94e+02
7.35e-06
1.12e-03
% of process
group total


6.91%
100.00%
0.02%
<0.01%
<0.01%
<0.01%
0.01%
<0.01%
0.05%
0.01%
<0.01%
0.01%
<0.01%
<0.01%
0.07%
<0.01%
0.0.1%
<0.01%
<0.01%
0.01%
<0.01%
0.03%
0.01%
0.04%
<0.01%
<0.01%
<0.01%
0.01%
<0.01%
<0.01%
0.02%
0.02%
0.22%
<0.01%
0.17%
<0.01%
<0.01%
95.80%
<0.01%
<0.01%
% of grand total










































                         2-121

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
            Table 2-57. LCD manufacturing stage ancillary material inputs
Material
Process Group
N-Butylacetate <
Nitric acid
Nitrogen i
Nitrogen fluoride
Nitrous oxide
Oxygen
Perfluoromethane
Phosphine
Phosphoric acid
Photoresist, unspecified i
Polyaluminum chloride ',
Polyethylene mono(nonylphenyl) ether glycol
Polyimide, unspecified
Propylene glycol
Propylene glycol monomethyl ether acetate
Rinse, unspecified
Sodium dihydrogen phosphate dihydrate
Sodium hydroxide
Solder, unspecified
Sulfur hexafluoride ;
Sulfuric acid
Surfactant, unspecified >
Synthetic resin, unspecified
Tetramethyl ammonium hydroxide
Unspecified LCD process material
Water
Xylene (mixed isomers)
Total
Panel Components '
Acetone '
Borax
Carbon dioxide
Cyclohexane
Developing solution, unspecified
Diluent, unspecified
Ethanol
Ethylacetate
Exfoliation liquid, unspecified
HCFC-225ca :
HCFC-225cb ;
Heptane
Quantity (kg/
functional unit)

3.83e-02
1.24e-02
5.90e+00
1.08e-01
1.36e-03
7.75e-03
1.29e-03
2.69e-02
3.95e-02
1.38e-02
6.40e-03
3.40e-04
2.94e-05
4.46e-03
1.56e-02
5.27e-02
4.06e-06
3.59e-01
7.35e-05
1.62e-02
2.29e-01
1.09e-04
6.57e-04
1.29e-01
2.58e-02
6.88e-02
1.57e-03
2.03e+02

1.03e-02
9.13e-05
4.82e-03
3.89e-03
4.00e-02
8.27e-03
1.17e-02
9.68e-04
1.43e-02
1.37e-04
1.37e-04
1.03e-02
% of process
group total

0.02%
0.01%
2.91%
0.05%
<0.01%
<0.01%
<0.01%
0.01%
0.02%
0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.01%
0.03%
<0.01%
0.18%
<0.01%
0.01%
0.11%
<0.01%
<0.01%
0.06%
0.01%
0.03%
<0.01%
206.91%

3.19%
0.01%
1.50%
1.21%
12.45%
2.57%
3.65%
0.30%
4.44%
0.04%
0.04%
3.19%
% of grand total




























99.48%







i





                                     2-122

-------
                      2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-57. LCD manufacturing stage ancillary material inputs
Material
Process Group
Hydrochloric acid
Hydrogen
Methyl ethyl ketone ;
Nitric acid second cerium ammonium
Nitrogen
Orthoboric acid
Perchloric acid
Photoresist, unspecified
Polyethylene terephthalate
Sulfunc acid
Tetrahydrofuran
Toluene
Total
LCD Glass
Aluminum Oxide
Cerium Oxide
Chromium Oxide
Hydrofluoric acid
Pumice
Total
Backlight
Diethyl ether
Ethanol
Process material for backlight assembly
Unspecified ancillary material
Total
PWB
Ammonium chloride
Ammonium hydroxide
Formaldehyde
Glycol ethers
Hydrochloric acid
Hydrogen peroxide
Nitric acid
Polyethylene glycol
Potassium hydroxide
Potassium permanganate
Potassium peroxymonosulfate
PWB-solder mask solids
Sodium Carbonate
Sodium hydroxide
Quantity (kg/
functional unit)

1.74e-03
3.14e-06
4.84e-04
1.13e-02
1.17e-01
7.30e-04
3.82e-03
2.94e-03
3.20e-02
1.58e-02
3.82e-03
2.75e-02
3,22e-01

1.56e-03
1.52e-04
2.60e-06
3.66e-03
3.64e-03
9.02e-03

9.28e-05
4.63e-05
7.03e-05
4.19e-04
6.28e-04

3.42e-02
3.42e-02
2.91e-03
1.04e-02
8.46e-02
1.37e-02
5.99e-02
2.23e-02
1.88e-02
5.14e-04
3.12e-02
1.93e-02
1.42e-02
8.56e-02
% of process
group total
-
0.54%
<0.01%
0.15%
3.51%
36.27%
0.23%
1.19%
0.91%
9.93%
4.91%
1.19%
8.56%
99.98%

17.30%
1.68%
0.03%
40.61%
40.37%
100.00%

14.78%
7.36%
11.19%
66.67%
100.00%

6.68%
3.42%
0.57%
2.03%
16.51%
2.67%
11.70%
4.34%
3.68%
0.10%
6.08%
3.76%
2.77%
16.71%
% of grand total













0.16%






0.00%





0.00%















                         2-123

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
            Table 2-57. LCD manufacturing stage ancillary material inputs
Material
Process Group
Sulfuric acid '<
Total
Japanese Grid
Lime '•
Limestone ;
Total
U.S. Grid
Lime
Limestone
Total
Fuel Production
Bauxite (A12O3, ore) ',
Limestone (CaCOS, in ground) I
Sand (in ground) !
Sodium chloride (NaCl, in ground or in sea)
Total
Grand Total
Quantity (kg/
functional unit)

8.05e-02
5.12e-01

4.53e-02
1.03e-01
1.48e-01

2.05e-03
4.66e-03
6.71e-03

2.16e-03
5.49e-02
1.32e-03
6.08e-04
5.90e-02
. 2.04e+02
% of process
group total

15.72%
96.74%

30.57%
69.43%
100.00%

30.52%
69.48%
100.00%

3.66%
93.07%
2.24%
1.03%
100.00%

% of grand total


0.25%
!


0.07%
|


0.00%

i



0.03%
1190.00%
                                     2-124

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-58. LCD manufacturing Stage utility inputs
Material
Process group
Fuels (kg/functional unit):
Monitor/Module
Fuel on #4
Kerosene
LNG
Liquified petroleum gas (LPG)
Natural gas
Total
Panel Components
Kerosene
Natural gas
Steam (100 psig)
Fuel oil #2
Fuel oil #6
Natural gas
Total
LCD glass
Fueloil#2
Liquified petroleum gas (LPG)
Natural gas
Total
Backlight
LNG
Natural gas
Total
PWB
Natural gas
Fuels
Coal, average (in ground)
Natural gas (in ground)
Petroleum (in ground)
Uranium (U, ore)
Total
Grand Total
Electricity (MJ/functional unit):
Monitor/Module
Panel Components
-LCD glass
Backlight
PWB
Quantity



2.11e-01
2.98e-01
3.22e+00
5.83e-01
8.48e-01
5.16e+00

1.68e-01
1.18e-07
1.45e-01
4.07e-04
1.25e-01
8.64e-02
5.25e-01

5.38e-02
1.62e+01
5.63e-02
1.64e+01

4.17e-06
4.17e-06
8.33e-06

1.72e-01

6.86e-01
2.41e+00
4.84e-01
1.15e-05
3.58e+00
2.58e+01

2.59e+02
4.64e+01
2.20e+00
4.46e+00
4.43e+00
% of process
grouD total



4.09%
5.77%
62.40%
11.30%
16.44%
100.00%

31.97%
<0.01%
27.56%
0.08%
23.91%
16.47%
100.00%

0.33%
99.33%
0.34%
100.00%

50.00%
50.00%
100.00%



19.19%
67.27%
13.54%
<0.01%
100.00%




-


% of grand total








19.99%







203%




6344%



000%

0.67%





13.87%
100.00%

81.80%
14.70%
0.70%
1.41%
1.40%
   2-125

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
                 Table 2-58. LCD manufacturing stage utility inputs
Material
Process group
Total
Water (kg or L/functional unit): ;
Monitor/Module
Panel Components
LCD glass
Backlight
PWB
Japanese electric grid
U.S. electric grid
Fuels
Total
Quantify

3.16e+02

1.08e+03
2.08e+02
1.62e+00
1.92e+02
1.86e+01
1.49e+02
2.20e+00
5.07e+02
2.15e+03
Total energy (fuels and electricity, MJ/functional unit):
Monitor/Module
Panel Components
LCDglass
Backlight
PWB
Fuels
Total
5.08e+02
6.29e+01
7.05e+02
4.46e+00
. 1.21e+01
1.456+02
1.44e+03
% of process
sroup total




















% of grand total
:'
100.00%

49.96%
9.66%
0.08%
8.91%
0.86%
6.91%
0.10%
i 23.53%
100.00%

35.36%
4.38%
49.03%
0.31%
0.84%
10.09%
BO.00%
                                     2-126

-------
                2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-59. LCD manufacturing stage air outputs
Material
Process Group
Module/Monitor
Acetic acid
Acetone
Al-etchant, unspecified
Ammonia
Argon
Carbon dioxide
Cr-etchant, unspecified
Cyclohexane
Diethylene glycol
Hexamethyldisilizane
Hydrochloric acid
Hydrofluoric acid
Hydrogen
Isopropyl alcohol
ITO etchant, unspecified
Monosilane
N-bromoacetamide
Nitric acid
Nitrogen fluoride
Nitrogen oxides
Phosphine
Phosphoric acid
PM
Polyimide, unspecified
Sulfur hexafluoride
Sulfur oxides
Tetramethyl ammonium hydroxide
Unspecified LCD process material
Total
Panel Components
Carbon dioxide
Ethylacetate
HCFC-225ca
HCFC-225cb
Heptane
Hydrochloric acid
Methyl ethyl ketone
Nitrogen oxides
Nonmethane hydrocarbons, remaining unspeciated
PM
Quantity (kg/
functional unit)


1.36e-03
1.86e-04
1.37e-02
6.23e-02
5.80e-03
2.16e-03
4.12e-02
4.85e-05
9.69e-05
1.37e-06
6.06e-02
5.21e-02
1.33e-04
1.78e-02
6.86e-03
1.54e-03
9.18e-03
2.69e-04
2.45e-01
5.48e-01
6.26e-02
4.85e-05
1.10e-05
1.40e-04
7.30e-03
1.12e-03
6.43e-01
4.49e-02
1.83e+00

4.82e-03
2.44e-06
1.40e-04
1.40e-04
7.77e-05
7.32e-06
1.35e-04
4.11e-04
7.77e-05
2.74e-05
% of process
group total


0.07%
0.02%
0.75%
3.41%
0.32%
0.12%
2.25%
<0.01%
0.01%
<0.01%
3.31%
2.85%
0.01%
0.97%
0.38%
0.08%
0.50%
0.01%
13.43%
30.00%
3.43%
<0.01%
<0.01%
0.01%
0.40%
0.06%
35.16%
2.46%
100.00%

81.79%
<0.01%
2.37%
2.37%
1.32%
0.12%
2.29%
6.98%
1.32%
0.47%
% of grand
total






























2.82%


•








                   2-127

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
                 Table 2-59. LCD manufacturing stage air outputs
Material
Process Group
Toluene
Total
LCD Glass
Carbon dioxide
Carbon monoxide
Chromium
Nitrogen oxides
PM
Sulfur oxides
Total
Backlight
Diethyl ether
Ethanol
Nitrogen oxides
Process material for backlight assembly
Total
PWB
Formaldehyde
Japanese Grid
1,1,1-Trichloroethane
1,2-Dichloroethane
2,3,7,8-TCDD
2,3,7,8-TCDF
2,4-Dinitrotoluene
2-Chloroacetophenone
2-Methylnaphthalene .
5-Methyl chrysene
Acenaphthene
Acenaphthylene
Acetaldehyde
Acetophenone
Acrolein
Anthracene
Antimony
Arsenic
Barium
Benzene
Benzora]anthracene
Benzo[a]pyrene
Benzo[bjjk]fluoranthene
BenzoFgJulperylene
Quantity (kg/
functional unit)

5.44e-05
5.89e-03

1.30e-01
2.07e-07
6.40e-09
2.04e-03
5.06e-06
2.36e-06
1.32e-01
i
9.26e-05
4.63e-05
2.95e-02
7.03e-05
2.97e-02

1.71e-05

1 2.17e-07
1.54e-07
5.65e-14
1.96e-13
1.07e-09
; 2.69e-08
r 8.00e-10
8.46e-ll
1.45e-08
l.lle-09
t 2.19e-06
5.776-08
l.lle-06
; . 1.53e-09
3.20e-06
2.36e-06
1.74e-06
5.13e-06
', 2.69e-09
> 1.46e-10
1.31e-09
i 1.45e-09
% of process
group total

0.92%
100.00%

98.45%
<0.01%
<0.01%
1.55%
<0.01%
<0.01%
100.00%

0.31%
0.16%
99.29%
0.24%
100.00%



<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
% of grand
total


001%







0.20%





0.05%

0.00%























                                   2-128

-------
                2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-59. LCD manufacturing stage air outputs
Material
Process Group
Benzyl chloride
Beryllium
Biphenyl
Bromoform
Bromomethane
Cadmium
Carbon dioxide
Carbon disulfide
Carbon monoxide
Chloride ions
Chlorobenzene
Chloroform
Chromium (111)
Chromium (VI)
Chrysene
Cobalt
Copper
Cumene hydroperoxide
Cyanide (-1)
Di(2-ethylhexyl)phthalate
Dibenzo[a,h]anthracene
Dichloromethane
Dimethyl sulfate
Dioxins, remaining unspeciated
Ethyl Chloride
Ethylbenzene
Ethylene dibromide
Fluoranthene
Fluorene
Fluorides (F-)
Formaldehyde
Furans, remaining unspeciated
Hexane ,
Hydrochloric acid
Hydrofluoric acid
Lndeno(l,2,3-cd)pyrene
Isophorone
Lead (Pb, ore)
Magnesium
Manganese (Mn, ore)
Quantity (kg/
functional unit)

2.69e-06
1.09e-07
6.53e-09
1.50e-07
6.15e-07
4.74e-07
5.18e+01
5.00e-07
9.43e-03
2.07e-04
8.46e-08
2.27e-07
1.86e-06
4.51e-07
1.80e-09
3.98e-06
1.07e-06
2.04e-08
9.61e-06
2.81e-07
9.95e-10
l.lle-06
1.85e-07
2.51e-12
1.61e-07
3.99e-07
4.61e-09
5.88e-09
6.15e-09
2.22e-05
3.43e-05
4.00e-12
2.58e-07
4.61e-03
5.77e-04
1.51e-09
2.23e-06
1.48e-06
4.23e-05
3.67e-06
% of process
group total

<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
99.14%
<0.01%
0.02%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
% of grand
total









































                   2-129

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
                 Table 2-59. LCD manufacturing stage air outputs
Material
Process Group
Mercury
Methane
Methyl chloride
Methyl ethyl ketone
Methyl hydrazine
Methyl methacrylate
Methyl tert-butyl ether
Molybdenum
Naphthalene
Nickel
Nitrogen oxides
Nitrous oxide'
o-xylene
Phenanthrene
Phenol
PM-10
Propionaldehyde
Pyrene
Selenium
Styrene
Sulfur dioxide
Tetrachloroethylene
TOCs, remaining unspeciated
Toluene
Vanadium
Vinyl acetate
Xylene (mixed isomers)
Zinc (elemental)
Total
U.S. Grid
1,1,1-Trichloroethane
1,2-Dichloroethane
2,3,7,8-TCDD
2,3,7,8-TCDF
2,4-Dinitrotoluene
2-Chloroacetophenone
2-Methylnaphthalene
5-Methyl chrysene
Quantity (kg/
functional unit)

3.97e-07
2.73e-04
2.04e-06
1.50e-06
6.54e-07
7.69e-08
1.34e-07
5.20e-07
7.44e-07
4.95e-05
1.37e-01
3.76e-04
6.49e-08
1.75e-08
6.15e-08
6.72e-03
1.46e-06
4.24e-09
5.40e-06
9.61e-08
2.90e-01
1.65e-07
6.63e-04
4.81e-06
1.92e-05
2.92e-08
1.42e-07
1.73e-05
5.22e+01

3.71e-09
6.95e-09
2.48e-15
8.86e-15
4.86e-ll
1.22e-09
5.16e-12
3.82e-12
% of process
group total

<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.26%
<0.01%
<0.01%
<0.01%
<0.01%
0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.56%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
100.00%

<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
% of grand
total

-



























80.58%









                                    2430

-------
                2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-59. LCD manufacturing stage air outputs
Material
Process Group
Acenaphthene
Acenaphthylene
Acetaldehyde
Acetophenone
Acrolein
Anthracene
Antimony
Arsenic
Barium
Benzene
Benzo[a]anthracene
Benzo[a]pyrene ~ •
Benzo[bj,k]fluoranthene
Benzo[g,h,i]perylene
Benzyl chloride
Beryllium
Biphenyl
Bromoform
Bromomethane
Cadmium
Carbon dioxide
Carbon disulfide
Carbon monoxide
Chloride ions
Chlorobenzene
Chloroform
Chromium (DI)
Chromium (VT)
Chrysene
Cobalt
Copper
Cumene
Cyanide (-1)
Di(2-ethylhexyl)phthalate
Dibenzo[a,h]anthracene
Dichloromethane
Dimethyl sulfate
Dioxins, remaining unspeciated
Quantity (kg/
functional unit)

1.10e-10
4.37e-ll
9.90e-08
2.60e-09
,5.04e-08
3.77e-ll
8.45e-09
7.26e-08
3.98e-09
2.26e-07
1.80e-ll
6.60e-12
2.06e-ll
6.98e-12
1.22e-07
3.69e-09
2.95e-10
6.77e-09
2.78e-08
9.33e-09
8.61e-01
2.26e-08
1.57e-04
3.52e-07
3.82e-09
1.02e-08
4.71e-08
1.40e-08
1.98e-ll
2.35e-08
1.93e-09
9.20e-10
4.34e-07
1.276-08
1.696-12
5.04e-08
8.33e-09
1.13e-13
% of process
group total

<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
98.98%
<0.01%
0.02%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
% of grand
total







































                   2-131

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
                 Table 2-59. LCD manufacturing stage air outputs
Material
Process Group
Ethyl Chloride
Ethylbenzene
Ethylene dibromide
Fluoranthene
Fluorene
Fluoride
Formaldehyde
Furans, remaining unspeciated
Hexane
Hydrochloric acid
Hydrofluoric acid
Indeno(l,2,3-cd)pyrene
Isophorone
Lead
Magnesium
Manganese
Mercury
Methane
Methyl chloride
Methyl ethyl ketone
Methyl hydrazine
Methyl methacrylate
Methyl tert-butyl ether
Molybdenum
Naphthalene
Nickel
Nitrogen oxides
Nitrous oxide
o-xylene
Phenanthrene
Phenol
Phosphorus (yellow or white)
PM-10
Propionaldehyde
Pyrene
Selenium
Styrene
Sulfur dioxide
Quantity (kg/
functional unit)

7.29e-09
1.64e-08
2.08e-10
1.30e-10
1.63e-10
3.78e-08
1.64e-07
1.80e-13
1.16e-08
! 2.08e-04
i 2.60e-05
1.28e-ll
l.Ole-07
2.47e-08
1.91e-06
8.83e-08
1.45e-08
; 1.25e-03
9.20e-08
; 6.77e-08
: 2.95e-08
i 3.47e-09
6.08e-09
1.13e-09
! 3.54e-09
1.33e-07
2.28e-03
i 6.58e-06
! 1.10e-10
4.85e-10
2.78e-09
9.59e-09
: 1.12e-04
i 6.60e-08
6.45e-ll
2.26e-07
i 4.34e-09
4.83e-03
% of process
group total

<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.02%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.14%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.26%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.56%
% of grand
total







































                                    2-132

-------
                2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-59. LCD manufacturing stage air outputs
Material
Process Group
Tetrachloroethylene
TOCs, remaining unspeciated
Toluene
Vanadium
Vinyl acetate
Xylene (mixed isomers)
Zinc (elemental)
Total
Fuel Production
1,1.1-Trichloroethane
1,2-Dichloroethane
1,4-Dichlorobenzene
2,4-Dinitrotoluene
2-Chloroacetophenone
3-Methylcholanthrene
5-Methyl chrysene
Acenaphthene
Acenaphthylene
Acetaldehyde
Acetophenone
Acrolein
Aldehydes
Aluminum (elemental)
Ammonia
Anthracene
Antimony
Aromatic hydrocarbons — •-— "
Arsenic
Barium
Benzene
Benzo[a]anthracene
Benzo[a]pyrene
Benzo[bj,k]fluoranthene
Benzo[b]fluoranthene
Benzo[g,h,i]perylene
Benzo[k]fluoranthene
Benzyl chloride
Beryllium
Quantity (kg/
functional unit)

7.47e-09
1.12e-05
4.92e-08
3.41e-08
1.32e-09
6.42e-09
2.95e-08
8.70e-01

6.91e-09
1.38e-08
1.89e-08
9.68e-ll
2.42e-09
2.84e-ll
7.60e-12
2.77e-10
1.16e-10
1.97e-07
5.18e-09
l.OOe-07
8.98e-05
9.59e-07
3.53e-04
1.15e-10
1.15e-10
2.55e-09
7.13e-07
1.77e-08
2.70e-03
6.97e-ll
3.84e-ll
3.80e-ll
3.09e-ll
3.21e-ll
3.09e-ll
2.42e-07
7.31e-08
% of process
group total

<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
100.00%

<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.03%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
% of grand
total








134%






























                   2-133

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
                 Table 2-59. LCD manufacturing stage air outputs
Material
Process Group
Biphenyl
Bromofonn
Bromomethane
Butane
Cadmium
Calcium
Carbon dioxide
Carbon disulfide
Carbon monoxide ;
Chloride ions |
Chlorine
Chlorobenzene
Chloroform ;
Chromium (ID)
Chromium (VT)
Chrysene
Cobalt
Copper
Cumene
Cyanide (-1)
Di(2-ethylhexyl)phthalate
Dibenzo[a,h]anthracene
Dichloromethane
Dimethyl sulfate
Dimethylbenzanthracene
Dioxins, remaining unspeciated
Ethane ;
Ethyl Chloride
Ethylbenzene
Ethylene dibromide
Fluoranthene
Fluorene
Fluorides (F-)
Formaldehyde
Furans, remaining unspeciated
Halogenated hydrocarbons (unspecified)
HALON-1301
Hexane
Quantity (kg/
functional unit)

5.88e-10
1.35e-08
5.53e-08
3.31e-05
4.18e-08
8.31e-07
9.44e+00
4.49e-08
2.89e-02
1.75e-06
l.Ole-09
7.60e-09
2.04e-08
1.05e-06
1.05e-06
6.70e-ll
1.27e-07
7.98e-08
1.83e-09
8.64e-07
2.52e-08
2.18e-ll
l.OOe-07
l,66e-08
2.36e-10
5.60e-12
4.88e-05
1.45e-08
3.28e-08
4.15e-10
S.Ole-10
3.66e-10
1.98e-07
5.78e-05
2.606-11
1.42e-14
2.476-11
2.84e-05
% of process
group total

<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
97.19%
<0.01%
0.30%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
% of grand
total







































                                    2-134

-------
                2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-59. LCD manufacturing stage air outputs
Material
Process Group
Hydrocarbons, remaining unspeciated
Hydrochloric acid
Hydrofluoric acid
Hydrogen sulfide
Indeno(l ,2,3-cd)pyrene
Iron
Isophorone
Lead
Magnesium
Manganese
Mercury
Metals, remaining unspeciated
Methane
Methyl chloride
Methyl ethyl ketpne
Methyl hydrazine
Methyl methacrylate
Methyl tert-butyl ether
Molybdenum
Naphthalene
Nickel
Nitrogen oxides
Nitrous oxide
Nonmethane hydrocarbons, remaining unspeciated
n-Propane
Other organics
o-xylene
Pentane
Phenanthrene
Phenol
Phosphorus (yellow or white)
PM
PM-10
Polycyclic aromatic hydrocarbons
Propionaldehyde
Pyrene
Selenium
Silicon
Quantity (kg/
functional unit)

7.75e-03
4.15e-04
5.18e-05
1.51e-04
5.31e-ll
1.85e-06
2.00e-07
6.39e-07
3.80e-06
1.15e-06
4.55e-08
1.60e-08
1.20e-01
1.83e-07
1.35e-07
5.88e-08
6.91e-09
1.21e-08
9.92e-08
1.98e-08
6.06e-06
4'.27e-02
8.l3e-04
8.79e-03
8.67e-08
1.35e-02
1.59e-07
4.10e-05
1.22e-09
5.53e-09
6.10e-07
6.95e-03
l.lle-05
2.84e-12
1.31e-07
2.00e-10
4.81e-07
8.31e-07
% of process
group total -

0.08%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
1.24%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.44%
0.01%
0.09%
<0.01%
0.14%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.07%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
% of grand
total







































                   2-135

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
                  Table 2-59. LCD manufacturing stage air outputs
Material
Process Group
Sodium
Styrene !
Sulfur oxides
Tetrachloroethylene
Toluene
Vanadium
Vinyl acetate
Zinc (elemental)
Total
Grand Total
Quantity (kg/
functional unit)

4.92e-06
8.64e-09
3.96e-02
1.49e-08
2.63e-07
1.30e-05
2.63e-09
6.04e-07
9.72e+00
6.48e+01
% of process
group total

<0.01%
<0.01%
0.41%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
100.00%

% of grand
total









15.00%
100.00%
                                    2-136

-------
                 2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-60. LCD manufacturing stage water outputs
Material
Process Group
WASTEWATER STREAMS
Module/monitor
Panel components
LCD glass
Backlight
PWB
Fuels
Total
WATER POLLUTANTS
Module/monitor
1,1,1-Trichloroethane
Antimony
Arsenic
BOD
BOD
Boron
Cadmium
Chromium
Chromium (VI)
COD
COD
Colon bacillus (bacteria in large intestine)
Copper
Cyanide (-1)
Cyanide (-1)
Dissolved solids
Fluorides (F-)
Fluorides (F-)
Hexane
Iron
Lead
Manganese
Mercury
Nickel
Nitrogen
Nitrogen
Oil & grease
Oil & grease
Organic phosphorus, unspecified
Phenol
Phosphorus (yellow or white)
Disposition


both
both
surface water
both
treatment
surface water



surface water
surface water
surface water
surface water
treatment
surface water
surface water
surface water
surface water
surface water
treatment
surface water
surface water
surface water
treatment
surface water
surface water
treatment
surface water
surface water
surface water
surface water
surface water
surface water
surface water
treatment
surface water
treatment
surface water
surface water
surface water
Quantify (kg/
functional unit)


2.87e+03
3.53e-01
1.67e+00
1.92e+02
1.86e+01
4.33e+01
3.12e+03


2.29e-08
1.14e-07
1.14e-07
1.74e-02
5.05e-02
4.58e-06
1.14e-07
8.84e-06
2.29e-07
2.68e-03
3.90e-02
3.89e-03
9.18e-07
3.66e-06
6.67e-07
7.55e-03
1.28e-02
2.40e-04
5.88e-04
2.63e-06
6.17e-06
2.29e-07
9.69e-08
2.29e-07
7.93e-02
1.16e-02
2.02e-04
3.53e-03
2.29e-07
2.29e-07
4.31e-03
% of process
group total











<0.01%
<0.01%
<0.01%
6.69%
19.41%
<0.01%
<0.01%
<0.01%
<0.01%
1.03%
14.99%
1.50%
<0.01%
<0.01%
<0.01%
2.90%
4.91%
0.09%
0.23%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
30.45%
4.45%
0.08%
1.35%
<0.01%
<0.01%
1.65%
% of grand total


91.80%
0.01%
0.05%
6.15%
0.60%
1.39%
100.00%

































                    2-137

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-60. LCD manufacturing stage water outputs
Material
Process Group
Phosphorus (yellow or white)
Polychlorinated biphenyls
Suspended solids
Suspended solids
Tetrachloroethylene
Tin
Trichloroethylene
Zinc (elemental)
Total
Panel components
BOD
BOD
Borax
COD
Hydrochloric acid
Nitrogen
Orthoboric acid
Phosphorus (yellow or white)
Suspended solids
Total
LCD glass
BOD
Chloride ions
Chromium
COD
Dissolved solids
Fluorides (F-)
Iron
Lead
Nickel
Nitrate
Oil & grease
Suspended solids
Total
Backlight
BOD
Iron
Lead
Mercury
Nickel
Nitrogen
Disposition

treatment
surface water
surface water
treatment
surface water
surface water
surface water
surface water


surface water
treatment
treatment
surface water
treatment
surface water
treatment
surface water
surface water


surface water
surface water
surface water
surface water
surface water
surface water
surface water
urface water
urface water
urface water
urface water
urface water


treatment
treatment
treatment
treatment
treatment
treatment
Quantity (kg/
functional unit)

6.91e-03
1.14e-08
1.55e-02
4.26e-03
2.29e-08
4.58e-07
2.29e-08
2.63e-06
2.60e-01

1.34e-03
3.89e-03
1.31e-06
2.21e-03
3.29e-06
5.71e-04
1.31e-06
2.48e-05
6.46e-04
8.69e-03

3.80e-07
4.68e-02
3.80e-09
3.80e-07
1.68e-01
1.36e-04
1.28e-04
2.01e-06
3.80e-09
1.83e-07
3.34e-04
3.35e-04
2.15e-01

3.00e-03
8.33e-05
8.33e-07
8.33e-08
3.33e-06
l.OOe-03
% of process
group total

2.65%
<0.01%
5.96%
1.64-%
<0.01%
<0.01%
<0.01%
<0.01%
100 00%

15.44%
44.73%
0.02%
25.45%
0.04%
6.58%
0.02%
0.29%
7.43%
100 00%

<0.01%
21.73%
<0.01%
<0.01%
77.84%
0.06%
0.06%
<0.01%
<0.01%
<0.01%
0.16%
0.16%
100.00%

54.50%
1.51%
0.02%
<0.01%
0.06%
18.17%
% of grand total




















070%

'











17 46%

1





                                  2438

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-60. LCD manufacturing stage water outputs
Material
Process Group
Oil & grease
Suspended solids
Total
PWB
Copper (+1 & +2)
Lead cmpds
Total
Japanese grid
Sulfate ion (-4)
Suspended solids
Total
U.S. grid
Sulfate ion (-4)
Suspended solids
Total
Fuels
Acids (H+)
Adsorbable organic halides
Aluminum (+3)
Ammonia ions
Aromatic hydrocarbons
Barium cmpds
BOD
Cadmium cmpds
Chloride ions
Chromium (HI)
Chromium (VI)
COD
Copper (+1 & +2)
Cyanide (-1)
Dissolved organics
Dissolved solids
Fluorides (F-)
Halogenated matter (organic)
Hydrocarbons, remaining unspeciated
Iron (+2 & 4-3)
Lead cmpds
Mercury compounds
Metals, remaining unspeciated
Nickel cmpds
Nitrate
Disposition

treatment
treatment


treatment
treatment


surface water
surface water


treatment
treatment


surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
Quantity (kg/
functional unit)

8.33e-05
1.338-03
5.50e-03

4.28e-05
7.14e-06
5.00e-05

2.93e-03
7.63e-05
3.01e-03

1.32e-04
3.45e-06
1.36e-04

1.76e-08
1.82e-ll
6.88e-06
1.33e-03
4.25e-09
1.36e-08
9.12e-03
1.42e-ll
2.65e-01
3.49e-09
3.49e-09
7.71e-02
2.84e-10
1.99e-ll
4.67e-08
2.01e-05
1.76e-06
5.67e-12
1.70e-05
1.73e-08
5.67e-ll
6.52e-14
4.72e-04
2.84e-ll
2.97e-06
% of process
group total

1.51%
24.22%
100.00%

85.71%
14.29%
100.00%

97.46%
2.54%
100.00%

97.46%
2.54%
100.00%

<0.01%
<0.01%
<0.01%
0.18%
<0.01%
<0.01%
1.23%
<0.01%
35.77%
<0.01%
<0.01%
10.42%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.06%
<0.01%
<0.01%
% of grand total



045%



000%



024%



001%


























   2-139

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
                Table 2-60. LCD manufacturing stage water outputs
Material
Process Group
Other nitrogen
Phenol
Phosphates
Polycyclic aromatic hydrocarbons
Salts (unspecified)
Sodium (+1)
Sulfateion(-4)
Sulflde
Suspended solids
TOCs
Toluene
Waste oil
Zone (+2)
Total
Grand Total
Disposition

surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water
surface water


Quantify (kg/
functional unit)

7.66e-10
1.75e-04
2.83e-08
6.81e-ll
7.84e-05
3.41e-01
4.36e-06
9.86e-09
4.14e-02
4.25e-08
6.24e-10
4.87e-03
1.40e-09
7.40e-01
1.23e+00
% of process
group total

<0.01%
0.02%
<0.01%
<0.01%
0.01%
46.05%
<0.01%
<0.01%
5.59%
<0.01%
<0.01%
0.66%
<0.01%
100.00%

% of grand total














(50.02%
1(90.00%
                                    2-140

-------
                      2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-61. LCD manufacturing stage hazardous waste outputs
Material
Process Group
Monitor/module
Mercury
Waste metals, unspecified
Waste acids, unspecified
Waste acid (mainly HF)
Waste acid (mainly HF)
Unspecified sludge
Thinner, unspecified
Tetramethyl ammonium hydroxide
Sodium sulfate
Rinse, unspecified
Remover, unspecified
Remover, unspecified
Phosphoric acid
Nitric acid
Isopropyl alcohol
Isopropyl alcohol
Ferric chloride
Acetone
Acetic acid
Total
Panel components
Spent solvents (F003 waste)
Flammable liquids (F003 waste)
Acid waste (D002 waste)
Spent solvent (with halogenated
materials)
Spent solvent (non-halogenated)
HCFC-225cb
HCFC-225ca
Waste solvent (photoresist)
Waste solvent (photoresist)
Waste acid (chrome mixed acid)
Total
LCD glass
Waste Batch (Ba, Pb) (D008 waste)
Waste acid (mostly 3% HC1 solution)
Hydrofluoric acid
Chrome liquid waste (D007 waste)
Chrome debris (D007 waste)
Barium debris (D008 waste)
Disposition


recycling/reuse
recycling/reuse
recycling/reuse
recycling/reuse
treatment
land (other than landfill)
treatment
recycling/reuse
recycling/reuse
recycling/reuse
recycling/reuse
treatment
landfill
landfill
recycling/reuse
treatment
recycling/reuse
treatment
landfill


treatment
treatment
treatment
treatment
treatment
recycling/reuse
recycling/reuse
recycling/reuse
treatment
recycling/reuse


landfill
recycling/reuse
landfill
recycling/reuse
treatment
landfill
Quantity (kg/
functional unit)


2.00e-06
1.17e-03
3.24e-02
5.69e-01
1.36e-01
3.09e-02
5.40e-01
1.42e-01
2.44e-01
4.67e-02
8.84e-02
3.03e-01
1.44e-02
3.43e-04
1.69e-01
1.91e+00
1.37e-02
2.77e-02
4.46e-03
4.28e+00

2.74e-04
9.13e-04
1.19e-03
1.55e-02
4.66e-02
3.11e-05
3.11e-05
2.05e-02
2.176-02
7.18e-03
1.14e-01

6.55e-05
1.82e-04
8.24e-05
4.54e-04
6.83e-06
9.91e-06
% of process
group total


<0.01%
0.03%
0.76%
13.30%
3.17%
0.72%
12.62%
3.33%
5.72%
1.09%
2.07%
7.08%
0.34%
0.01%
3.95%
44.75%
0.32%
0.65%
0.10%
100.00%

0.24%
0.80%
1.04%
13.63%
40.89%
0.03%
0.03%
18.00%
19.05%
6.29%
100.00%

' 8.18%
- 22.74%
10.29%
56.70%
0.85%
1.24%
% of grand
total





















92.11%











2.46%







                         2-141

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-61. LCD manufacturing stage hazardous waste outputs
Material
Process Group
Total
Backlight
Hazardous waste, unspecified
Hazardous waste, unspecified
Waste glass, with mercury
Waste CCEL, with mercury
Waste CCFL, with lead
Silver '
Chromium
Total
PWB
PWB-Waste cupric etchant
PWB-Solder dross
PWB-Routedust
PWB-Lead contaminated waste oil
?WB-Decontaminating debris
Hazardous waste, unspecified
Total
Fuels
Hazardous waste, unspecified
Grand Total
Disposition



recycling/reuse .
treatment
andfill
treatment
treatment
andfill
landfill


recycling/reuse
recycling/reuse
recycling/reuse
treatment
treatment ;
treatment


andfill

Quantity (kg/
functional unit)



1.42e-02
6.80e-03
1.05e-10
8.17e-10
8.17e-08
2.72e-09
1.52e-06
2.11e-02

9.93e-02
2.96e-02
5.31e-03
5.14e-03
6.85e-03
5.48e-02
2.01e-01

2.97e-02
4.64e+00
% of process
group total



67.68%
32.31%
<0.01%
<0.01%
<0.01%
<0.01%
0.01%
100.00%

49.42%
14.72%
2.64%
2.56%
3.41%
27.26%
10000%

'

% of grand
total





















100.00%
                                  2-142

-------
                    2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-62. LCD manufacturing stage solid waste outputs
Material
Process Group
Monitor/module
Isopropyl alcohol
LCD panel waste
Printed wiring board (PWB)
Remover, unspecified
Unspecified sludge
Unspecified sludge
Unspecified solid waste
Waste acid (containing F and detergents)
Waste acids, unspecified
Waste alkali, unspecified
Waste LCD glass
Waste LCD glass
Waste metals, unspecified
Waste oil
Waste plastic from LCD modules
Waste plastic from LCD modules
Waste plastics from LCD monitor
Total
Panel components
Isopropyl alcohol
Polyester resin
Unspecified solid waste
Used silica gel
Waste alkali (color filter developer,
unspecified)
Waste LCD glass
Total
LCD glass
abrasive sludge
acid absorbent
blasting media
Cinders from LCD glass mfg
Cobalt nitrate
Diesel fuel
Dust
LCD glass EP dust '
LCD glass EP dust
LCD glass, unspecified
Nickel nitrate
Oily rags & filter media
Disposition


treatment
landfill
landfill
treatment
recycling/reuse
treatment
recycling/reuse
landfill
treatment
recycling/reuse
landfill
recycling/reuse
recycling/reuse
treatment
recycling/reuse
treatment
landfill


recycling/reuse
recycling/reuse
treatment
landfill
recycling/reuse
landfill


recycling/reuse
landfill
landfill
landfill
treatment
treatment
treatment
landfill
recycling/reuse
landfill
treatment
landfill
Quantity (kg/
functional unit)


1.03e-02
2.43e-02
7.50e-03
3.09e-02
8.46e-01
5.73e-02
2.02e-02
2.70e-01
1.05e-01
3.23e-01
2.06e-01
7.20e-01
2.93e-03
1.61e-02
7.40e-02
4.03e-01
4.05e-02
3.16e+00

2.53e-02
3.20e-02
1.10e-02
6.22e-04
8.91e-02
5.74e-02
2.15e-01

1.95e-03
3.77e-06
1.70e-05
3.83e-04
2.83e-06
1.88e-06
1.59e-04
4.77e-05
2.32e-04
1.13e-03
2.83e-06
1.51e-05
% of process
group total


0.33%
0.77%
0.24%
0.98%
26.79%
1.82%
0.64%
8.56%
3.32%
10.24%
6.52%
22.80%
0.09%
0.51%
2.35%
12.77%
1.28%
100.00%

11.75%
14.84%
5.09%
0.29%
41.37%
26.66%
100.00%

32.61%
0.06%
0.28%
6.40%
0.05%
0.03%
2.65%
0.80%
3.88%
18.83%
0.05%
0.25%
% of grand total



















25 07%







1.71%













                      2-143

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-62. LCD manufacturing stage solid waste outputs
Material
Process Group
Oily rags & filter media
parts cleaner solvent '
Plating process sludge
Potassium Carbonate
sludge (calcium fluoride, CaF2)
Sludge from LCD glass mfg
Sodium Carbonate
Unspecified sludge
Waste alkali, unspecified
Waste LCD glass
Waste oil
Waste refractory
Total
Backlight
Broken CCFL
Cardboard
Polyethylene, foamed
Polyethylene/polypropylene waste
Unspecified nonhazardous waste
Waste backlight casing (PC)
Waste backlight light guide (PMMA)
Total
PWB
PWB-Drill dust
Unspecified solid waste
Unspecified solid waste
Total
Japanese grid
Coal waste
Dust/sludge
Fly/bottom ash
Total
U.S. grid
Coal waste
Dust/sludge
Fly/bottom ash
Total
Fuels
Aluminum scrap
Aluminum scrap, Wabash 319
Bauxite residues
Disposition

recycling/reuse
recycling/reuse
landfill
landfill
recycling/reuse
andfill
landfill
landfill
treatment
landfill
treatment
landfill


andfill
treatment
treatment
treatment
recycling/reuse
andfill
andfill


andfill i
recycling/reuse
treatment


andfill
andfill [
andfill i


andfill
andfill
andfill


ecycling/reuse
ecycling/reuse
andfill
Quantity (kg/
functional unit)

1.88e-06
3.77e-06
1.52e-05
1.53e-04
8.13e-04
4.06e-05
1.53e-04
3.56e-04
1.95e-06
8.70e-05
3.03e-04
1.13e-04
5.98e-03

2.69e-07
1.82e-05
9.99e-04
2.72e-03
1.26e-04
1.46e-05
1.52e-03
5.40e-03

6.59e-03
1.91e-01
1.62e+00
l.Sle+00

2.18e+00
8.42e-01
5.45e-01
3.57e+00

9.85e-02
3.81e-02
2.46e-02
1.61e-01

8.77e-06
2.37e-08
5.87e-04
% of process
group total

0.03%
0.06%
0.25%
2.55%
13.59%
0.68%
2.55%
5.95%
0.03%
1.45%
5.07%
1.89%
100.00%

<0.01%
0.34%
18.50%
50.45%
2.32%
0.27%
28.12%
100.00%

0.36%
10.53%
89.11%
100.00%

61.12%
23.59%
15.28%
100.00%

61.10%
23.63%
15.27%
100.00%

<0.01%
<0.01%
0.02%
%; of grand total












i

!







004%




1440%




28 34%




1 28%




                                  -2-144

-------
                       2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
   Table 2-62. LCD manufacturing stage solid waste outputs
Material
Process Group
FGD sludge
Mineral waste
Mixed industrial (waste)
Non toxic chemical waste (unspecified)
Slag and ash
Slag and ash
Unspecified solid waste (incinerated)
Unspecified waste
Total
Grand Total
Disposition

landfill
landfill
landfill
landfill
landfill
recycling/reuse
treatment
landfill


Quantity (kg/
functional unit)

1.09e-02
1.26e-04
4.83e-02
2.95e-05
3.40e+00
3.49e-02
6.44e-04
1.72e-01
3.66e+00
1.26e+01
% of process
group total

0.30%
<0.01%
1.32%
<0.01%
92.69%
0.95%
0.02%
4.70%
100.00%

% of grand total









29.10%
100.00%
Table 2-63. LCD manufacturing stage radioactive waste outputs
Material
Process Group
Japanese grid
Low-level radioactive waste
Uranium, depleted
Total
U.S. grid
Low-level radioactive waste
Uranium, depleted
Total
Grand Total
Disposition


landfill
landfill


landfill
landfill


Quantity (kg/
functional unit)


3.71e-04
l.lle-04
4.82e-04

3.40e-06
1.02e-06
4.42e-06
4.87e-04
% of process
group total


76.93%
23.07%
100.00%

76.93%
23.07%
100.00%

% of grand total




99.09%



0.91%
100.00%
                          2-145

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-64. LCD manufacturing stage radioactivity outputs
Material
Process Group
Japanese grid
Antimony-124 (isotope)
Antimony-125 (isotope)
Argon-41 (isotope)
Barium-140 (isotope)
Bromine-89 (isotope)
Bromine-90 (isotope)
Cesium-134 (isotope)
Cesium-134 (isotope)
Cesium-137 (isotope)
Cesium-137 (isotope)
Chromium-51 (isotope)
Chromium-Si (isotope)
Cobalt-57 (isotope)
Cobalt-57 (isotope)
Cobalt-58 (isotope)
Cobalt-58 (isotope)
Cobalt-60 (isotope)
Cobalt-80 (isotope)
Iodine-131 (isotope)
Iodine-131 (isotope)
Iodine-132 (isotope)
Iodine-132 (isotope)
Iodine-133 (isotope)
Iodine-133 (isotope)
Iodine-134 (isotope)
Iodine-135 (isotope)
Iodine-135 (isotope)
Iron-55 (isotope)
Iron-59 (isotope)
Krypton-85 (isotope)
Krypton-85M (isotope)
Krypton-85M (isotope)
Krypton-87 (isotope)
Krypton-88 (isotope)
Lanthanum-140 (isotope)
Manganese-54 (isotope)
Manganese-54 (isotope)
Molybdenum-99 (isotope)
Niobium-95 (isotope)
Niobium-95 (isotope)
Disposition


treatment
treatment
air
treatment
air
air
air,
treatment
air ;
treatment
air
treatment
air
treatment
air
treatment
air
treatment
air
treatment
air
treatment
air
treatment
air
air
treatment
treatment
treatment
air
air
treatment
air
air
treatment
air
treatment
treatment
air
treatment
Quantity (Bq/
functional unit)


1.67e+00
6.63e+00
3.37e+03
1.23e-01
3.90e-04
1.59e-04
1.07e-02
4.45e+00
8.08e-02
6.69e+00
2.11e-01
8.02e+00
5.68e-04
1.94e-01
7.26e-03
7.90e+01
5.46e-02
2.07e+01
2.55e-01
3.70e+00
5.18e-02
1.40e+00
2.37e+02
1.59e+00
2.68e-01
1.35e-02
1.14e+00
1.89e+01
9.70e-01
5.59e+03
2.71e+02
5.00e+00
l.Ole+02
4.73e+02
1.32e-01
3.00e-03
5.29e+00
9.98e+06
1.19e-04
1.36e+00
% of process
group total


<0.01%
<0.01%
0.03%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.06%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
98.42%
<0.01%
<0.01%
% of grand total








!











:





















                                   2-146

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-64. LCD manufacturing stagfe radioactivity outputs
Material
Process Group
Rubidium-88 (isotope)
Ruthenium-103 (isotope)
Silver-llOM (isotope)
Silver-llOM (isotope)
Sodium-24 (isotope)
Strontium-89 (isotope)
Strontium-90 (isotope)
Strontium-95 (isotope)
Sulfur-136 (isotope)
Technetium-99M (isotope)
Technetium-99M (isotope)
Tin-113 (isotope)
Tritium-3 (isotope)
Tritium-3 (isotope)
Xenon-131M (isotope)
Xenon-131M (isotope)
Xenon-133 (isotope)
Xenon-133- (isotope)
Xenon-133M (isotope)
Xenon-133M (isotope)
Xenon-135 (isotope)
Xenon-135 (isotope)
Xenon-135M (isotope)
Xenon-138 (isotope)
Zinc-85 (isotope)
Zirconium-95 (isotope)
Total
U.S. grid
Antimony-124 (isotope)
Antimony-125 (isotope)
Argon-41 (isotope)
Barium-140 (isotope)
Bromine-89 (isotope)
Bromine-90 (isotope)
Cesium-134 (isotope)
Cesium-134 (isotope)
Cesium-136 (isotope)
Cesium-137 (isotope)
Cesium-137 (isotope)
Chromium-51 (isotope)
Chromium-51 (isotope)
Disposition

air
treatment
air
treatment
treatment
treatment
treatment
treatment
treatment
air
treatment
treatment
air
treatment
air
treatment
air
treatment
air
treatment
air
treatment
air
air
treatment
air


treatment
treatment
air
treatment
air
air
air
treatment
treatment
air
treatment
air
treatment
Quantity (Bq/
functional unit)

l.lle+00
1.67e-01
3.56e-06
1.94e+00
2.96e-01
3.20e-01
7:52e-02
8.29e-01
1.78e-01
1.60e-05
1.16e-01
. 1.84e-01
7.90e+03
5.91e+04
4.56e+02
6.08e+01
4.37e+03
9.34e+03
6.58e+04
7.65e+01
2.48e+03
6.97e+01
4.74e+01
1.57e+02
8.92e-02
3.08e-04
l.Ole+07

1.53e-02
6.09e-02
3.09e+01
1.13e-03
3.58e-06
1.45e-06
9.82e-05
4.09e-02
1.75e-03
7.41e-04
6.13e-02
1.94e-03
7.36e-02
% of process
group total

<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.08%
0.58%
<0.01%
<0.01%
0.04%
0.09%
0.65%
<0.01%
0.02%
<0.01%
• 
-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
             Table 2-64. LCD manufacturing stage radioactivity outputs
Material
Process Group
Cobalt-57 (isotope)
Cobalt-57 (isotope)
Cobalt-58 (isotope)
Cobalt-58 (isotope)
Cobalt-60 (isotope)
Cobalt-80 (isotope)
Iodine-131 (isotope)
Iodine-131 (isotope)
Iodine-132 (isotope)
Iodine-132 (isotope)
Iodine-133 (isotope)
Iodine-133 (isotope)
Iodine-134 (isotope)
Iodine-135 (isotope)
Iodine-135 (isotope)
Iron-55 (isotope)
fron-59 (isotope)
Krypton-85 (isotope)
Krypton-85M (isotope)
Krypton-85M (isotope)
Krypton-87 (isotope)
Krypton-88 (isotope)
Lanthanum-140 (isotope)
Manganese-54 (isotope)
Manganese-54 (isotope)
Molybdenum-99 (isotope)
Niobium-95 (isotope)
Niobium-95 (isotope)
Rubidium-88 (isotope)
Ruthenium-103 (isotope)
Silver-llOM (isotope)
Silver-llOM (isotope)
Sodium-24 (isotope)
Strontium-89 (isotope)
Strontiuin-90 (isotope)
Strontium-95 (isotope)
Sulfur-136 (isotope)
Technetium-99M (isotope)
Technetium-99M (isotope)
Tin-113 (isotope)
Tritium-3 (isotope)
Disposition

air
treatment
air
treatment
air
treatment
air
treatment
air
treatment
air
treatment
air
air
treatment
treatment
treatment
air
air
treatment
air
air
treatment
air
treatment
treatment
air
treatment
air
treatment
air
treatment
treatment
treatment
Teatment
treatment
treatment
air
treatment
treatment
air
Quantity (Bq/
functional unit)

5.21e-06
1.78e-03
6.65e+00
7.25e-01
5.01e-04
1.90e-01
2.34e-03
3.39e-02
4.75e-04
1.28e-02
2.17e+00
1.45e-02
2.46e-03
1.24e-04
1.04e-02
1.73e-01
8.90e-03
S.lSe+Ol
2.48e+00
4.58e-02
9.25e-01
4.34e+00
1.21e-03
2.75e-05
4.85e-02
9.15e+04
1.09e-06
1.25e-02
1.02e-02
1.53e-03
3.26e-08
1.78e-02
2.72e-03
2.93e-03
6.90e-04
7.60e-03
1.64e-03
1.47e-07
1.06e-03
1.68e-03
7.25e+01
% of process
group total

<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.06%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
98.41%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
0.08%
% of grand total


























'















                                     2-148

-------
                     2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-64.  LCD manufacturing stag! radioactivity outputs
Material
Process Group
Tritium-3 (isotope)
Xenon-131M (isotope)
Xenon-131M (isotope)
Xenon-133 (isotope)
Xenon-133 (isotope)
Xenon-133M (isotope)
Xenon-133M (isotope)
Xenon-135 (isotope)
Xenon-135 (isotope)
Xenon-135M (isotope)
Xenon-138 (isotope)
Zinc-85 (isotope)
Zirconium-95 (isotope)
Total
Fuels
Radioactive substance (unspecified)
Radioactive substance (unspecified)
Total
Grand Total
Disposition

treatment
air
treatment
air
treatment
air
treatment
air
treatment
air
air
treatment
air


air
surface water


Quantity (Bq/
functional unit)

5.42e+02
4.18e+00
5.58e-01
6.04e+02
8.57e+01
4.01e+01
7.02e-01
2.286+01
6.39e-01
4.35e-01
1.44e+00
8.18e-04
2.82e-06
9.30e+04

4.44e+01
4.11e-01
4.48e+01
1.02e+07
% of process
group total

0.58%
<0.01%
<0.01%
0.65%
0.09%
0.04%
<0.01%
0.02%
<0.01%
<0.01%
<0.01%
<0.01%
<0.01%
100.00%

99.08%
0.92%
100.00%

% of grand total














091%



0.00%
100.00%
                        2-149

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
LCD manufacturing - primary material inputs




                                 16.2
                       o.32
                                0.28
                                 01

                                 *
                                 D)
                                    LCD mfg. process group



                      Figure 2-51. LCD manufacturing primary material inputs
                                                                          23.1
LCD manufacturing - ancillary material inputs
•£ 250-n
3 200-
"3 150-
0 100-
"'§ 50 •
3 0-
203
^m
•
•• 0.322 0.009 0.001
t: m M co -*±
1 1 _ § g i
S E o 2 O S
°-f 3
LCD mfg.



0.512 0.148 0.007 0.059
CD O T3 CO
| l-o '§> §
|'& "
process group
Figure 2-52. LCD manufacturing ancillary material inputs
LCD manufacturing - fuel inputs
§20-i 16.4 j
J 10. H i
1 5' K 0.53 H 8-3&06
^i o4 — ^^ — i — •— — i — ^* — i — • —
]JS ~£ £ — w £
J"J ^ C (TJ O)
§ o g S "a 2
i E « S o S
13 EL §• O CD
o
C3
LCD mfg.

0.17


m
i




process group

0.00 0.00


(D "a
M '=
W -Q O)
S -c eo
Q. °> ~>
3




3.58
•an '
. • -.
0)

-------
                               2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
LCD manufacturing
±! 300-j 259
3 ^n
— BUB
c 20°" H
~ BiBi
1 10°- 1 46.4
^ • •• 2.2
«= Q MB m
"g •£ -2 S
S 3 _ C Jg
c ^ « g "o)
2 £ . « S Q
Q- p O
o ~*
o
Figure 2-54.




4.5


£
f
CO
m
- electricity inputs





4.4 o.O

CQ 
-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS

M
c
3
1
C
O
i
LCD
60-i
50 -
40 -
30 -
20 -
10-



1-8 5.9S03
li i
Ii 1|
1
manufacturing - air emissions



52.2
^^_
^H
0.03 ^H
0.13 1.7E-05 ^H 0.87
LCD glass
Figure 2-57.
•"= S S) §
= ? <£ 13 O)
1 !» §
CD «
LCD mfg. process group
LCD manufacturing air pollutant outputs



9.7
ID
£

LCD manufacturing - wastewater outputs
C 3500 -,
£ 3000-
c 2500-
g 2000-
c 1500-
5 1000-
3; SOD-
O> o -
2,870

•
•
•
^H 192
^1 0.35 1.7 18.6 0.0 0.0
^H «_.
"S  5S 'S>
7=T3 TnCD -ss. = n ST3 a>
§0 gc 0) -g 0- S-g OT
ie|g. g § I01 =
O
LCD mfg. process group
Figure 2-58. LCD manufacturing wastewater outputs

1





43.3

' J2
CD
U.




                        LCD manufacturing -water pollutant outputs
         8.E-01
                                 2.2E-01
                                                                            7.4E-01
                                         5.5E-03
                                                  5.0E-05    3.0E-03   1.4E-04
                            S         LCD mfg. process group



                        Figure 2-59. LCD manufacturing water pollutant outputs
                                          2-152

-------
                          2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
4.3&-00
           LCD manufacturing - hazardous waste outputs
         1.1E-01    8.0E.Q4
                           2.1 E-02
                2.0E-01   o.OEt-00    O.O&OO    3.0E-02
                  ^m
                  CO
O>
Q
O
                             OQ
0.
OS
-3
                                                         T3


                                                         CO
                                                                  U.
                        LCD mfg. process group

        Figure 2-60. LCD manufacturing hazardous waste outputs
              LCD manufacturing - solid waste outputs
                                               3.57
                                      1.80
                            0.005
                   0.006
                        LCD mfg. process group

           Figure 2-61.  LCD manufacturing solid waste outputs
                                                                 3.66
                             2-153

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS	   '

       Of the total 49 kg of primary materials per functional unit in the manufacturing stage, the
fuels production contributes the greatest (23.1 kg/functional unit), followed by the Japanese
electric grid (16.2 kg/functional unit) (Figure 2^51 and Table 2-56). Nearly all (203 out of 204
kg/functional unit) of the ancillary materials used during manufacturing are from the
monitor/module process group (Figure 2-52 and Table 2-57).  liquified natural gas (LNG)
constitutes about 96% (194 kg/functional unit) of the total ancillary materials in the
monitor/module process group. As stated earlier, this is not used to calculate energy impacts.
       Of the utility inputs, electricity and water inputs were greatest in the monitor/module
manufacturing processes; fuels were greatest in the LCD glass manufacturing process group
(Figures 2-53, 2-54, and 2-55). Sixty-three percent of the fuel inputs are from the LCD glass
manufacturing process group. Within that group, the use of LPG clearly dominates at about 16.2
kg/functional unit (over 99% of the LCD glass fuel inputs) (Table 2-58).5 The monitor/module
manufacturing fuel inputs are about 20% of the total manufacturing fuel inputs (5.2 kg/functional
unit). About 259 MJ/functional unit of electricity in the LCD manufacturing stage are from the
monitor/module processes7, or 82% of all manufacturing electricity (Figure 2-54). The total
energy, which converts fuel mass into energy and adds that to the electrical energy, is greatest in
the LCD glass manufacturing processes and contributes 705 MJ/functional unit to the 1,440
MJ/ftinctional unit in the manufacturing stage (Figure 2-56). Water inputs are most significant in
the monitor/module manufacturing process, contributing 1,080 kg (or liter)/functional unit
(Figure 2-55), which is 50% of all the manufacturing water inputs. The fuels production and
panel components process groups contribute about 24% and 10% to the water manufacturing
inputs, respectively.
       For outputs from the manufacturing stage, the mass of air emissions are dominated by the
generation of electricity (Figure 2-57). Individual material (pollutant) contributions for each
process group are presented in Table 2-59. Wastewater outputs (i.e., the volume or mass of
wastewater released) are greatest for the monitor/module manufacturing processes (92%)
(Figure 2-58), but only 21% of the chemical pollutants in the wastewater streams come from
those processes (Figure 2-59). Table 2-60 shows the individual contributions from each material.
       Hazardous wastes from the LCD manufacturing stage dominated over other life-cycle
stages, and within the manufacturing stage, the monitor/module processes had the greatest
hazardous waste outputs by mass (4.3 kg/functional unit) (Figure 2-60). The greatest
contributors by mass are isopropyl alcohol (a total of 2.1 kg/functional unit, or 49% of all wastes
from the monitor/module manufacturing processes)  (Table 2-61). These wastes, however, are
recycled and although they  are a large portion  of the inventory, they will not affect the impact
assessment (to be presented in Chapter 3) as they are not directly released to the environment.
       Solid wastes generated during the manufacturing stage were only about 21% of the
overall solid wastes generated throughout the LCD life-cycle, as was shown  earlier in
         Note: An industry participant questioned the, large fuel contribution reported here; however, further
discussions with industry supported that no valid reason could justify removing these data. Glass energy inputs are
evaluated in a sensitivity analysis (see Section 2.7.3 and 3.4).

       7 This amount of electricity is consistent with the industry participant that expressed doubt in the large fuel
energy contribution and subsequent overall energy use amount in module manufacturing.

-------
                                       2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Figure 2-48. Within the manufacturing stagey the fuels production, Japanese grid and
monitor/module manufacturing process groups are all major contributors to the solid waste
outputs (Figure 2-61). The individual material contributions are provided in Table 2-62.
       Radioactive waste and radioactivity are directly related to the electricity generation
process and therefore, only the Japanese and U.S. electric grid processes generate these outputs in
the manufacturing stage. Tables 2-63 and 2-64 show that more radioactive wastes and
radioactivity are from the Japanese electric grid. This is a result of more manufacturing
processes being in Japan, as modeled in this project, as well as the greater fraction of nuclear
power in the Japanese electric grid.

2.7.2  Relative Data Quality

       Sections 2.2 through 2.6 (and associated appendices) discuss the data quality and data
limitations for each life-cycle stage.  Several factors contribute to the overall quality for an entire
life-cycle stage. For example, the manufacturing stage includes several different processes that
were collected from several different companies.  The quality of one data set from one company
may be very different from that of another company. Relative data quality estimates have been
made for each life-cycle stage, including electricity generation, which is included in more than
one life-cycle stage (Table 2-65). In addition, transportation data quality is listed in Table 2-65,
although it has been excluded from the analysis due to the very low data quality.

                           Table 2-65. Relative data quality
Life-cycle stage
Upstream
Manufacturing
Use
EOL
Electricity generation
Transportation
Relative data quality
Moderate
Moderate to high
Moderate to high
Low to moderate
High
Very low
2.7.3   Sensitivity Analyses

       The inventory results presented above in Section 2.7.1 are the "baseline" results in this
study.  The baseline scenario includes the parameters/assumptions presented in the
methodologies for the effective life scenario. However, due to assumptions and uncertainties in
this LCA, as in any LCA, sensitivity analyses on the baseline results have been conducted.  Four
areas have been identified where sensitivity analyses were most warranted:

•      use stage lifespan assumptions;
•      glass manufacturing energy inputs;
•      LCD monitor/module manufacturing energy inputs, and
•      LCD EOL disposition assumptions.
                                          2-155

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Selected sensitivity analyses were chosen based on the data with either the greatest uncertainties
or with a large uncertainty and a major contributor to the inventory results. The matrix in Table
2-66 shows the different sensitivity analyses or scenarios that are considered in the impact
assessment results. Discussions of the sensitivity analyses for manufactured life (use stage),
glass manufacturing energy inputs, LCD monitor manufacturing energy inputs, and LCD EOL
inventories follow in this section.  Complete inventories of each sensitivity analysis scenario are
not presented; however, the effects determined in the LCIA results of the sensitivity analyses are
shown in Chapter 3 (see Section 3.4).
                     Table 2-66. List of sensitivity analysis scenarios
Monitor
type
Sensitivity analysis scenario
Baseline analyses (for reference)
CRT
LCD
Effective life scenario with average glass energy inputs (all glass manufacturing energy data used)
Effective life scenario with average glass energy inputs (all glass manufacturing energy data used)
and outliers in the LCD module manufacturing energy data removed
Sensitivity analyses
CRT
LCD
CRT
LCD
LCD
LCD
Manufactured life scenario same as baseline except lifespan is based on manufactured life instead of
effective life, which results in some revised functional equivalency calculations (see Section 2.7.3.1
below) :
Manufactured life scenario same as baseline except lifespan is based on manufactured life, which
results in some revised functional equivalency calculations (see Section 2.7.3.1 below)
Modified glass energy scenario same as baseline except comparatively high glass manufacturing
energy inputs are removed
Modified glass energy scenario same as baseline except comparatively high glass manufacturing
energy inputs are removed
Modified LCD module energy scenario same as baseline except LCD monitor/ module
manufacturing energy outliers are included in the average
Modifed LCD EOL scenario same as baseline except LCD EOL dispositions are modified
2.7.3.1 Manufactured life scenario

       To address uncertainties in the use stage lifespan assumptions, we applied the
manufacturing life scenario to the CRT and LCD life-cycle profiles. (See Section 2.4 for a
discussion of the product use stage and the differences in the "effective life" versus
"manufactured life" life span assumptions.) Recall that the LCD manufactured life (45,000
hours) is 3.6 times greater than the CRT manufactured life (12,500 hours). In an LCA,
comparisons are made based on functional equiyalency. Therefore, if one monitor will operate
for a longer period of time than another, impacts should be based on an equivalent use. Thus,
                                         2-156

-------
                                      2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
based on equivalent use periods, under the manufactured life'scenario 3.6 CRTs would need to
be manufactured for every LCD. This was incorporated into the profile analysis for the
comparative manufactured life LCA.  Similarly, on average, 1.4 LCD backlights (which can be
cost-effectively replaced) will be needed during the manufactured lifetime of an LCD monitor.
This was also incorporated into the profile.  Thus, the following modifications were made:

       change the CRT electricity input in the use stage from 635 kWh (2,286 MJ) to 788 kWh
       (2,837 MJ);8
       change the LCD electricity input in the use stage from 237 kWh (853 MJ) to 1,035 kWh
       (3,726 MJ);
•      increase the manufacturing of CRTs by a factor of 3.6 to account for the functional
       equivalency of CRTs and LCDs. This was done by increasing the functional unit (22 kg
       CRT monitor) by a factor of 3.6, which equates to manufacturing 3.6 times more CRTs
       than in the baseline case; and
       increase the manufacturing of the LCD backlight lamp by a factor of 1.4 to account for
       the functional equivalency of LCDs and CRTs. This was done by increasing the
       backlight lamp mass (0.0023 kg), which is an input to the backlight unit assembly
       process, by a factor of 1.4.

Note that functional equivalency modification requires that the manufactured life scenario results
be used only when comparing the CRT and LCD. These results cannot be accurately used to
compare EL to ML for CRT or LCD.  LCIA results of the sensitivity analysis are presented in
Chapter 3.

2.7.3.2 Modified glass energy scenario

       In the second case, the energy input values for CRT glass manufacturing (and
consequently LCD glass manufacturing) were considered uncertain due to the large discrepancy
in fuel and electricity values among the individual data sets. The baseline case uses the average
of the data supplied and confirmed by the companies who supplied the data. However, because
one set of data was significantly higher, that one set of data was removed from the profile for the
sensitivity analysis. (A statistical evaluation of the glass manufacturing data for outliers could
not be conducted because there were not enough data sets.)
       In the baseline scenario, the averaged primary data from manufacturers of total energy to
produce a kilogram of CRT or LCD glass was 1,560 MJ (433 kWh) of energy, with only 0.3% of
that as electrical energy.  The sensitivity analysis scenario assumes 16.3 MJ (4.5 kWh) per
kilogram of glass produced, with approximately 30% as electrical energy. The majority of the
fuel energy in the baseline scenario was from LPG.  The energy consumption values can be
compared to estimates for the entire glass industry, which includes the more prevalent general
flat glass, as well as speciality glasses, such as CRT and LCD glass.  In a report of the Glass
Technology Roadmap Workshop (Energetics Inc. 1997), it was estimated that in practice, about
1.1 MJ of energy are required to melt  a kilogram of glass; and electrical energy contributes
        This represents the electricity use for a 12,500 hour life span. This figure is then multiplied by a factor of
3.6 in the functional equivalency calculations (see third bullet, below).

                                        2-157                                        "

-------
2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
approximately 13% of the total process energy requirements from glass production, as reported in
1994. Although this does not translate into energy requirements for CRT or LCD glass, it
suggests the baseline data collected for this analysis may be inflated. Therefore, the sensitivity
analysis uses revised energy input values for glass production, but it is also not known whether
these values represent the true energy requirements for CRT and LCD glass production. The
sensitivity analysis is considered a lower bound of energy requirements for monitor glass
production.
                                          i
                                          I
2.7.3.3 Modified LCD module energy scenario

       LCD monitor/module manufacturing energy was another area of relatively large
uncertainty and variability in the inventory data] The CDP received seven sets of LCD
monitor/module manufacturing data from five companies in Japan and two in Korea. Of these,
the manufacturing energy data from one company in Korea was incomplete and could not be   .
used. For the remaining six data sets, total energy inputs ranged from 330 MJ to 7,310 MJ, with
a mean and standard deviation of 2,269 MJ and 2,906 MJ, respectively. Given the wide
variability in the data and large standard deviation, CDP researchers evaluated the data for
outliers by breaking the total energy data points into quartile ranges. Minor outliers are then
those within a certain range of multipliers beyond the middle 50 percent of the distribution. That
is, the interquartile range (IQR) (i.e., the range of values representing the middle 50 percent)
multiplied by 1.5 is the lower bound of the minor outlier and the IQR times three is the upper
bound of the minor outlier. Anything beyond IQR times three is a major outlier.  Using this
approach, one data set was found to be a minor outlier and another was found to be a major
outlier.  These outliers were excluded from the averages used in the baseline analysis, but
included in the averages used in the LCD monitor/module manufacturing energy sensitivity
analysis.
       Table 2-67 summarizes the energy inputs for the LCD monitor/module manufacturing
process group under the baseline and modified LCD module energy scenarios. Note that total
energy inputs are approximately 4.5 times lower under the baseline scenario. However, because
of the different types of energy (fuel and electricity) employed by different manufacturers, the
mean electric energy is higher for the baseline than the modified energy scenario.

2.7.3.4 LCD End-of-life dispositions         i

       Finally, because very few desktop LCDs have reached their end of life, and usually only if
they have been damaged in some way, very little is known about the percentage of LCDs that are
remanufactured, recycled, landfilled or incinerated.  In the baseline scenario, it was assumed that
a certain proportion of monitors go to each EOL disposition. As the functional unit in this study
is one monitor, we used those proportions to represent the probability that one monitor would go
to the respective disposition. To address uncertainties in the allocation of disposition
percentages, a sensitivity analysis was conducted with a different set of final disposition
numbers. Details and assumptions for the sensitivity analysis are provided in Appendix I.  Table
2-68 presents the distribution of LCD EOL dispositions assumed under the baseline and modifed
EOL dispositions scenarios. LCIA results for trie sensitivity analysis are presented in
Section 3.4.
                                         2-158

-------
                               2.7 SUMMARY OF LIFE-CYCLE INVENTORY RESULTS
Table 2-67. Energy inputs to the LCD monitor/module manufacturing process group

Total energy
(MJper
monitor)
Electric energy
(MJper
monitor)
Fuel energy
(MJper
monitor)
% electric
energy
% fuel energy
Baseline (excludes two outlier data sets from the means used in the inventory)
Range
Mean
Standard
deviation
333 to 934
508
284
199 to 359
259
68
48 to 695
249
300
25 to 88
60
27
12 to 75
40
27
Modified Energy (includes two outlier data sets in the means used in the inventory)
Range
Mean
Standard
deviation
333 to 7,317
2,274
2,906
125 to 359
222
79
48 to 7,146
2052
2,956
2 to 88
10
36
12 to 98
90
36
      Table 2-68. Distribution of LCD EOL dispositions in the baseline and
                         modified EOL scenarios
Disposition
Incineration
Recycling
Remanufacturing
Hazardous waste landfill
Solid waste landfill
Baseline
15%
15%
15%
5%
50%
Modified
15%
0%
40%
5%
40%
                                  2-159

-------
REFERENCES
                                   REFERENCES
Bocko, P. 2000. Personal communication with P. Bocko, Corning Incorporated and M.L.
       Socolof, University of Tennessee Center for Clean Products and Clean Technologies.
       October 5.

CIA (Computer Industry Almanac, Inc.). 1997. Computer Industry Almanac. Arlington
       Heights, EL.

Douglas, D. 1999.  Personal communication with D. Douglas and E. Korman, Dell, and J.G.
       Overly, University of Tennessee Center for Clean Products and Clean Technologies.
       March.

Ecobilan. 1999. DEAM, life-cycle inventory database developed by the Ecobilan Group,
       copyright 1999.

EIA (Energy Information Administration). 1997.  Country Energy Data Report for 1997.  Web
       site available at http://www.eia.doe.govj

EIA. 1999a. Electric Power Monthly, February 1999 (With Data for November 1998).
       DOE/EIA-0226(99/02).

EIA. 1999b. "1997 Residential Energy Consumption Survey," Housing Characteristics report.
       Web site available at http://www.eia.doe.gov/emeu/consumption/reports.html (under
       "Residential Households').

Energetics, Incorporated. 1997. Report of the Glass Technology Roadmap Workshop, April 24-
       25,1997, Alexandria, Virginia.  Sponsored by the U.S. Department of Energy, Corning,
       Incorporated, Fort Motor Company, Pilkington libbey-Owens-Ford, Praxair, foe., and
       Techneglas, Incorporated. Columbia, Maryland.

EPA (Environmental Protection Agency). 1996.  Compilation of Air Pollutant Emission Factors
       (AP-42), Fifth Edition, Volume I. AP-42.  January.

EPA. 1998.  "Characterization of Municipal Solid Waste in the United States: 1997 Update."
       Report No. EPA530-R-98-007.

Fanara, A. 1999. Personal communication with A. Fanara, U.S. Environmental Protection
       Agency (EPA) Energy Star Program, and  J.G. Overly, University of Tennessee Center
       for Clean Products and Clean Technologies.  July.

FEPC (Federation of Electric Power Companies).  1996.  "Electricity Review Japan: The FEPC."
       Web site available at http://www.fepc of jp/erj/data 03.html (1996 data).
                                        2-160

-------
         	REFERENCES

Goldwasser, S.M. 1999. "Notes on the Troubleshooting ;and Repair of Computer and Video
       Monitors." Web site available at
       http://plop.phys.cwru.edu/repairfaq/REPAIR/F_monfaq.html.

ISO (International Standards Organization). 1996. ISO 14040, Environmental Management -
       Life-cycle Assessment Principles and Framework.  TC 2071 SC 5N 77. International
       Standards Organization, Paris.

Lapp, J.C. 2000. Personal communication with J.C. Lapp, Corning Incorporated, and M.L.
       Socolof, University of Tennessee, Center for Clean Products and Clean Technologies.
       October 11 and October 23.

Matthews, H. S., et al.  1997. "Disposition and End-of-Life Options for Personal Computers."
       Green Design Initiative Technical Report #97-10.  Carnegie Mellon University,
       Pittsburgh, PA. My.

McConnaughey, M.  1999. Personal communication with M. McConnaughey, ViewSonic, and
       J.G. Overly, University of Tennessee, Center for Clean Products and Clean Technologies.
       March.

MCC (Microelectronics and Computer Technology Corporation).  1998.  Computer Display
       Industry and Technology Profile.  EPA 744-R-98-005. December.

Nordman, B., M. A. Piette and K. Kinney. 1996. "Measured Energy Savings and Performance of
       Power-Managed Personal Computers and Monitors." Presented at the American Council
       for an Energy Efficient Economy Summer Study, Pacific Grove, CA. LBL-38057,
       UC-1600. August.

NSC (National Safety Council). 1999. Electronic Product Recovery and Recycling Baseline
       Report:  Recycling of Selected Electronic Products in the United States. Washington,
       DC.  May.

ORNL (Oak Ridge National Laboratory). 1994. Report 3: Estimating Externalities of Coal
       Fuel Cycles.  Oak Ridge, TN.  September.

ORNL. 1995. ReportS: Estimating Externalities of Nuclear Fuel Cycles.  Oak Ridge, TN.
       April.

SET AC (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.

Sharp, J. 2000. Personal communication with J. Sharp, Teredyne Corp., J.G. Overly, University
       of Tennessee, Center for Clean Products and Clean Technologies, and M.L. Socolof,
       University of Tennessee, Center for Clean Products and Clean Technologies.  August
                                        2-161

-------
REFERENCES
Socolof, M.L, et al 2000. "Preliminary Life-cycle Assessment Results for the Design for title
       Environment Computer Display Project." International Symposium on Electronics and
       the Environment, The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
       Computer Society Conference Record, May 8-10, San Francisco, CA.

TORNRC (The Oak Ridge National Recycle Center). 2001. Personal communication with D.
       McFarland and A. Bradley, TORNRC, and R. Dhingra, University of Tennessee, Center
       for Clean Products and Clean Technologies. April.

Tsuda, D. 1999. Personal communication with D. Tsuda, Environmental Technologies and
       Strategies Department, Apple Computer, Inc., and M.L. Socolof, University of
       Tennessee, Center for Clean Products and Clean Technologies.  July.

Vorhees, G. 2000.  Personal communication with G. Vorhees, Envirocycle, and M.L; Socolof,
       University of Tennessee, Center for Clean Products and Clean Technologies. July.

Young, R.  1999. Personal communication with R. Young, Display Search, and M.L: Socolof,
       University of Tennessee, Center for Clean Products and Clean Technologies. August.

VP150. 1998. Viewsonic VP150 17" monitor specifications sheet.

Wood, S,P. 2000.  Personal communication with S.P. Wood, Techneglas, and M.L. Socolof,
       University of Tennessee, Center for Clean Products and Clean Technologies. October 27.
                                        2-162

-------
                                                                    3.1 METHODOLOGY
                                      Chapters   .

                        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 al, 1997; Fava et al, 1993). For toxicity
impacts in particular, there are some methods being applied in practice (e.g., toxicity potentials,
critical volume, and direct valuation) (Guinee et al., 1996; ILSI, 1996; Curran, 1996), while
others are in development. However, there is currently no general consensus among the LCA
community as to one method over another. LCIA sophistication has also been discussed in
efforts to determine the appropriate level of analytical sophistication for various types of decision
making requirements (Bare et al, 1999) or one that adequately addresses toxicity impacts.
       Section 3.1 of this chapter presents the University of Tennessee (UT) LCIA methodology,
which takes a more detailed approach to chemical toxicity impacts than some methods currently
being used. Section 3.1 also discusses data sources, data quality, and the limitations and
uncertainties in the LCIA methodology. The UT methodology calculates life-cycle impact
category indicators for a number of impact categories, including several traditional LCA impact
categories  (e.g., 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
interest in human and ecological toxicity and to fill a common gap in LCIAs. Work conducted
for Saturn  Corporation and the EPA Office of Research and Development by the UT Center for
Clean Products and Clean Technologies has provided the basis for much of this methodology
(Swanson, 2001).
       Section 3.2 of this chapter describes the data management and analysis software used to
calculate LCIA results. Section 3.3 presents the baseline LCIA results for both the CRT and the
LCD. Baseline results are presented by impact category and include a discussion of the specific
limitations and uncertainties in each category.  Section 3.4 presents sensitivity analyses of the
baseline results.

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 then 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 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.
       Conceptually, there are three major phases  of LCIA, as defined by the Society of
Environmental Toxicology and Chemistry (SETAC) (Fava et al, 1993):
                                         3-1

-------
3.1 METHODOLOGY
•      Classification - The process of assignment and initial aggregation of data from inventory
       studies to impact categories (e.g., greenhouse gases or ozone depletion compounds).
•      Characterization - The analysis and estimation of the magnitudes of potential impacts
       for each impact category, derived through application of specific impact assessment tools.
       In the CDP, "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 or weighting is
an optional element to be included depending on the goals and scope of the study. The CDP
addresses the first two LCIA steps and leaves the valuation step to industry or others. In
addition, further qualitative risk screening of selected materials is conducted beyond the
traditional LCIA "characterization" phase in Chapter 4.  The methodologies for life-cycle impact
classification and characterization are described in Sections 3.1.1 and 3.1.2, respectively.
Sections 3.1.3 and 3.1.4 address data sources and data quality, and limitations and uncertainties
associated with the LCIA methodology.

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 to be included in the CDP LCIA are listed below:

•      Natural Resource Impacts
             renewable resource use
             nonrenewable materials use/depletion
             energy use
             solid waste landfill use
             hazardous waste landfill use
             radioactive 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])
             radioactivity
                                         3-2

-------
                                                                      3.1 METHODOLOGY
•      Potential Human Health and Ecotoxicity Impacts
              chronic human health effects (occupational and public)
              aesthetic impacts (odor)
              aquatic ecotoxicity
              terrestrial ecotoxicity

       The second step of classification is assigning inventory inputs or outputs to applicable
impact categories. Classification 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 for a particular
inventory item. Figure 3-1 shows a conceptual model of classification for the CDP. 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 both a global warming gas and has the
potential to create photochemical oxidants (smog formation).
       Output inventory items from a process may have varying dispositions, such 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.
                                           3-3

-------
3.1 METHODOLOGY
  Inventory
  data type
 Disposition
  General
  impact
  categories
  Chemical
  properties
Specific
impact •
categories*
                                                                                Landfill    Treatment
 Resource/
energy use
   Certain chemi
     (e.g., green louse
:al pro Derties indicate Which specific infipact
      gases will have a
                   categories
global warminb impact score)
 will be calculated
(see Table 3-1).
* Equations for calculating impact scores for each impact category are provided in Sect. 3.1.2.
** Excluded from the scope of the GDP; however, included in the UT Life-Cycle Design Toolkit.
Note, radioactivity (not depicted in this figure) is classified for radioactive isotope outputs to air, water or landfill.

                        Figure 3-1. Impact classification conceptual model
                                              3-4

-------
                                                                          3.1 METHODOLOGY
          Table 3-1. Inventory types and properties for classifying inventory items
Inventory Type
Input
Output
Chemical/Material Properties

Impact Category

Natural Resource Impacts
material, water
material, fuel
electricity, fuel
—
— -
—
—
—
— -
solid waste to
landfill
hazardous waste to
landfill
radioactive waste
to landfill
renewable
nonrenewable
energy
RCRA a - defined nonhazardous waste (or
other country-specific definitions)
RCRA a - defined hazardous waste"
(or other country-specific definitions)
radioactive waste
renewable resource use
nonrenewable resource
use/depletion
energy use
solid waste landfill use
hazardous waste landfill use
radioactive waste landfill use
Abiotic Ecosystem Impacts
—
—
—
—
—
• —
—
— •
—
air
air
air
air
air
water
water
water
radioactivity to air,
water, or land
global warming gases
ozone depleting substances
substances that can be photochemically
oxidized
substances that react to form hydrogen
ions (H+)
air particulates (PM10, TSP) »
substances that contain available nitrogen
or phosphorus
BOD a
TSSa
radioactive substance (isotope)
global warming
stratospheric ozone depletion
photochemical smog
acidification
air quality (air particulates)
water eutrophication (nutrient
enrichment)
water quality: BOD
water quality: TSS
radioactivity
Human Health and Ecotoxicity
material
—
—
—
—
—
air, water
air
water
air, water
toxic material
toxic material
odorous material
toxic material
toxic material
chronic human health effects -
occupational
chronic human health effects -
public
aesthetic impacts (odor)
aquatic ecotoxicity
terrestrial ecotoxicity
  Acronyms:  Resource Conservation and Recovery Act (RCRA); 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 METHODOLOGY
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 CDP:

•      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. Two
examples are nonrenewable resource depletion and landfill use.  Use of nonrenewable resources
are directly estimated as the mass (loading) of that material consumed (input amount). Use of
landfill space applies the mass loading of an output of hazardous, nonhazarclous, or radioactive
waste and converts that loading into a volume to estimate the amount of landfill space consumed.
       The equivalency method uses equivalency factors that exist for certain impact categories.
Equivalency factors are values that provide a relative measure or weighting that relate an
inventory output amount to some impact category relative to a certain chemical.  For example, to
relate an atmospheric release to the global warming impact category, chemical-specific global
warming potential (GWP) equivalency factors are used.  GWPs are a measure of the possible
warming effect on the earth's surface arising from the emission of a gas relative to carbon
dioxide (CO2). They are based on atmospheric  lifetimes and radiative forcing of different
greenhouse gases.
       The scoring of inherent properties method 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 CHEMS-1 method described by Swanson et
al. (1997), and presented below. 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. The aesthetics category directly applies an inherent chemical property (i.e.,
odor threshold concentration), but does not convert that value into a relative score, or HV.
       Using the various approaches, the UT LCIA method calculates impact scores for each
inventory item for each applicable impact category. Impact scores are therefore based on either a
direct measure of the inventory amount or somd modification (e.g., equivalency or scoring) of
that amount based on the potential  effect the inventory item may have on a particular impact
                                        3-6

-------
                                                                     3.1 METHODOLOGY
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 then used to calculate
impact scores. For each inventory item, an individual score is calculated for each applicable
impact category. The equations presented in the subsections that follow 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 CDP
Impact Category
Characterization Approach
Natural Resource Impacts
Renewable resource use
Nonrenewable materials use/depletion
Energy use
Solid waste landfill use
Hazardous waste landfill use
Radioactive waste landfill use
loading
loading
loading
loading
loading
loading
Abiotic Ecosystem Impacts
Global warming
Stratospheric ozone depletion
Photochemical smog
Acidification
Air quality (paniculate matter)
Water eutrophication (nutrient enrichment)
Water quality (BOD, TSS)
Radioactivity
equivalency (full)
equivalency (full)
equivalency (partial)
equivalency (full)
loading
equivalency (partial)
loading
loading
Human Health and Ecotoxicity
Chronic human health effects - occupational
Chronic human health effects - public
Aesthetic impacts (odor)
Aquatic ecotoxicity
Terrestrial ecotoxicity
scoring of inherent properties
scoring of inherent properties
application of inherent properties
scoring of inherent properties
scoring of inherent properties
                                           3-7

-------
3.1 METHODOLOGY
3.1.2.1  Renewable and nonrenewable resource use

       Natural resources are materials that are found in nature in their basic form rather than
being manufactured (e.g., water, minerals, petroleum and wood). Renewable (or flow) resources,
which are those that can be regenerated, are typically biotic resources (e.g., forest products, other
plants or animals) and water. Nonrenewable (or stock) resources are abiotic, such as mineral ore
or fossil fuels. Both of these natural resource impacts are calculated using the loading approach.
Renewable and nonrenewable resource consumption impacts use direct consumption values (i.e.,
material mass) from the inventory.
       Renewable resource impact scores are based on the following process inputs in the LCI:
primary, ancillary, water, and fuel inputs of renewable materials. To calculate the loading-based
impact scores, the following equation is used:
where:
RC
equals the impact score for use of renewable resource / (kg) per functional unit;
equals the inventory input amount of renewable resource i (kg) per functional
unit; and                                         ......
equals the fraction recycled content (post industrial and post consumer) of
resource i.
       In the CDP LCI, most manufacturers that provided primary data did not report recycled
content nor was the recycled content available for material inventories from secondary sources.
Therefore, to calculate the impact score for use of renewable resources the recycled content (RC)
was assumed to be zero.
       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. For the nonrenewable materials use/depletion category, depletion of
materials results from the extraction of nonrenewable resources. Nonrenewable resource impact
scores are based on the amount of primary, ancillary, and fuel inputs of nonrenewable materials.
To calculate the loading-based impact scores the following equation is used:
where:
IS,
  'NRR
Amt
    •NRR
RC
                                              x (1 -
equals the impact score for use of nonrenewable resource i (NRR) (kg) per
functional unit;
equals the inventory input amount of nonrenewable resource i (kg) per functional
unit; and
equals the fraction recycled content (post industrial and post consumer) of
resource i.
Due to the lack of data on the recycled content of nonrenewable resources, RC was assumed to
be zero.
                                        3-8

-------
                                                                     3.1 METHODOLOGY
3.1.2.2  Energy use                       ;;          =;   .

       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 fuel and
electricity inputs. 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 K, Table K-l. The impact score is calculated by:

                            (ISE)t = AmtEi or [AmtFx(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;
Amtp         equals the inventory input amount of fuel used (kg) per functional unit;
H            equals the heat value of fuel i (MJ/L); and
D            equals the density of fuel i (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).

3.1.2.3  Landfill use

       Landfill impacts are calculated using solid, hazardous, or radioactive waste outputs to
land as volume of landfill space consumed. Solid waste landfill use pertains to the use of
suitable and designated landfill space as a natural resource where municipal waste or
construction debris is accepted. A solid waste landfill impact score is calculated using solid
waste outputs disposed of in a solid waste (nonhazardous). landfill. Impact characterization is
based on the volume of solid waste, determined from the inventory mass amount of waste and
material density of each specific solid waste type:                   .    -
                                        t  = (Amtsw/D)f
where:
Amt.
    sw
equals the impact score for solid waste landfill (SWL) use for waste i (m3) per
functional unit;
equals the inventory output amount of solid waste-i (kg) per functional unit; and
equals density of waste i (kg/m3).
       Hazardous waste landfill use 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, is accepted. For non-US activities, equivalent hazardous or
special waste landfills are considered for this impact category. Impact scores are characterized
from hazardous waste outputs with a disposition of landfill. Impact characterization is based on
                                           3-9

-------
3.1 METHODOLOGY
the volume of hazardous waste, determined from the inventory mass amount of waste and
material density of each specific hazardous waste type:
                                       i  -•(AmtBW/D)i
where:
D
equals the impact score for hazardous waste landfill (HWL) use for waste i (m3)
per functional unit;
equals the inventory output amount of hazardous waste i (kg) per functional unit;
and
equals density of waste i (kg/m3),
       Radioactive waste pertains to the suitable and designated landfill space as a natural
resource mat accepts radioactive waste. Impacts are characterized from radioactive waste outputs
with a disposition of landfill. Impact characterization is based on the volume of radioactive
waste, determined from the inventory mass amount of waste and material density of each specific
waste.
where:
ISRWL        equals the impact score for radioactive waste landfill (RWL) use for waste i (m3)
             per functional unit;
AmtRW       equals the inventory output amount of radioactive waste i (leg) per functional unit;
             and
D           equals density of waste i (kg/m3).

3.1.2.4 Global warming impacts

       The build up CO2 and other greenhouse gases in the atmosphere may generate a
"greenhouse effect" of rising temperature and climate change. Global warming potential (GWP)
refers to the warming (relative to CO2) that chemicals contribute to this effect by trapping the
earth's heat. The impact scores for global warming (global climate change) effects 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 being presented in
this memorandum uses GWPs having effects in the 100-year time horizon. Although LCA does
not necessarily have a temporal component of the inventory, these impacts are expected to be far
enough into the future that releases occurring throughout the life cycle of a computerlmonitor
would be within the 100-year time frame. Appendix K, Table K-2 presents a current list of
GWPs as identified by the Intergovernmental Panel on Climate Change (IPCC) (Houghton et al.,
1996). Global warming impact scores are calculated for any chemicals  in the LCI that are found
in Appendix K, Table K-2. The equation to calculate the impact score for an individual chemical
is as follows:
                                        3-10

-------
                                                                    3.1 METHODOLOGY
where:
EF
Z-'* /
   GWP
Amtr.
                                                xAmtGG\
equals the global warming impact score for greenhouse gas chemical i (kg CO2
equivalents) per functional unit;
equals the GWP equivalency factor for greenhouse gas chemical i (CO2
equivalents, 100 year time horizon) (Appendix K, Table K-2); and
equals the inventory output amount of greenhouse gas chemical i released to air
(kg) per functional unit.
3.1.2.5 Stratospheric ozone depletion

       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 potentials (ODPs), which measure the change in the ozone column in the
equilibrium state of a substance compared to the reference chemical chlorofluorocarbon (CFC)-
11 (Heijungs et al, 1992; CAAA, 1990).  The list of ODPs that are used in this methodology are
provided in Appendix K, Table K-3.  The individual chemical impact score for stratospheric
ozone depletion impacts is based on the ODP and inventory amount of the chemical:
where:
EF,
   OOP
Amt,
    ODC
                               (ISOD)t  =
equals the ozone depletion impact score for chemical i (kg CFC-11 equivalents)
per functional unit;
equals the ODP equivalency factor for chemical i (CFC-11 equivalents)
(Appendix K, Table K-3); and
equals the amount of ozone depleting chemical i released to air (kg) per functional
unit.
3.1.2.6 Photochemical smog

       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 may contribute to this effect. The POCP
is based on simulated trajectories of tropospheric ozone production 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 to be used
in this methodology are presented in Appendix K, Table K-4. As shown in Table 3-2,
photochemical smog impacts are based on partial equivalency because some chemicals cannot be
                                          3-11

-------
3.1 METHODOLOGY
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:
ISpocp
EF,
   POCP
Amt,
    •POC
equals the photochemical smog impact score for chemical i (kg ethene
equivalents) per functional unit;
equals the POCP equivalency factor for chemical i (ethene equivalents)
(Appendix K, Table K-4); and
equals the amount of smog-creating chemical i released to the air (kg) per
functional unit.
3.1.2.7  Acidification

       This refers to the release of chemicals that may contribute to the formation of acid
precipitation. Impact characterization is based 0n 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 K, Table K-5 lists the AP values that will
be used as the basis of calculating acidification impacts. The impact score is calculated by:
where:
Amt
    AC
equals the impact score for acidification for chemical i (kg SO2 equivalents) per
functional unit;
equals the AP equivalency factor for chemical i (SO2 equivalents) (Appendix K,
Table K-5); and
equals the amount of acidification chemical i released to the air (kg) per functional
unit.
3.1.2.8  Air particulates

       This refers to the release and build up of particulate matter primarily from combustion
processes. Impact scores are based on particulate release amounts [particulate matter with
average aerodynamic diameter less than 10 micrometers (PM10)] to the air. This size of
particulate matter 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:
                                         3-12

-------
                                                                     3.1 METHODOLOGY
                                     ISPM '= Amtp
                                                PM
where:

AmtPM
              equals impact score for particulates (kg PM10) per functional unit, and
              equals the inventory output amount of particulate release (PM10) to the air (kg) per
              functional unit.
In this equation, PM10 is used to estimate impacts. However, if only total suspended particulates
(TSP) data are available, these data may be used. Note that using TSP data is an overestimation
of PM10 which only refers to the fraction of particulates 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.

3.1.2.9 Water eutrophication

       Eutrophication (nutrient enrichment) impacts to water are based on the identity and
concentrations of eutrophication 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 of importance to eutrophication. 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 al.,  1992;
Lindfors et al., 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 K, Table K-6) and the inventory amount:
                               (ISEVTR)I =
                                                x AmtEC)i
where:
EFEP
Amt
    EC
              equals the impact score for regional water quality impacts from chemical i (kg
              phosphate equivalents) per functional unit;
              equals the EP equivalency factor for chemical i (phosphate equivalents)
              (Appendix K, Table K-6); and
              equals the inventory output mass (kg) of chemical i per functional unit of
              eutrophication chemical in a wastewater stream released to surface water after any
              treatment, if applicable.

3.1.2.10 Water quality

       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 biological oxygen demand (BOD) and total suspended solids (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
                                          3-13

-------
3.1 METHODOLOGY
characterization is based on the amount of BOD and TSS in a wastewater stream. The water
quality score equations for each are presented below:

                                  (!SBOD)i  = (Amtgoo) i
                                          and
                                   (ISTSS)i  = (AmtTSS)i
where:
Amt,
    •BOD
IS-
  TSS
Amt
    TSS
             equals the impact score for BOD water quality impacts for waste stream i (kg) per
             functional unit;
             equals the inventory output amount of BOD in wastewater stream i released to
             surface waters (kg) per functional unit;
             equals the impact score for TSS water quality impacts for waste stream i (kg) per
             functional unit; and
             equals the inventory amount of TSS in wastewater stream i released to surface
             waters (kg) per functional unit.
3.1.2.11 Radioactivity

       Radioactivity inventoried as the quantity of an isotope released to the environment is
considered in the radioactivity impact category. These outputs, such as those from the generation
of nuclear energy, can be air, water, or land releases.  The radioactivity impact is a direct loading
score measured in Bequerels of radioactivity, and calculated as follows:
where:
                                   flS-A  =

             equals the impact score for radioactivity of isotope i (Bq) per functional unit; and
             equals the inventory amount of radioactivity of isotope i (Bq) per functional unit.

While this impact category uses a loading approach, further refinement of this impact score
calculation in the future could use radioactivity dose conversion factors, which convert
radioactivity quantities (e.g., Bq) into human doses equivalents (e.g., sievert or rem).

3.1.2.12 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 LCA are chronic (repeated dose) effects, which include
noncarcinogenic and carcinogenic effects, and aesthetics (although not a health effect per se,
aesthetics pertains to human welfare). Chronic human health effects to both workers and the  -
public are considered.  Quantitative measures of consumer impacts are not included in this LCIA
methodology because there are no direct outputs quantified in the LCI from the use stage of a
computer monitor.  The CDP does, however, quantify indirect outputs from energy consumption
(i.e., pollutants released from energy production). In addition, Appendix L qualitatively
discusses direct  consumer impacts, such as electromagnetic radiation or eye strain.
                                        3-14

-------
                                                                      3.1 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), or other literature. Upon review by the EPA DfE Workgroup (see
Appendix C), several materials in the CDP inventory were excluded from the toxic list if they are
generally accepted as nontoxic. 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 K, Table K-8, and chemicals that were excluded from the toxic list that
appear in the CDP inventory are presented in Appendix K, Table K-9.
       Human (and ecological) toxicity impact scores are calculated based on a chemical scoring
method modified from CHEMS-1 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.  If toxicity data are unavailable for a chemical, a mean
default toxicity score is given. This is described in further detail below. Ecological toxicity is  ~
presented in Section 3.1.2.13.                                                          •  •--•
       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, chronic effectsto other
specific organs or body systems (e.g.,  blood, cardiovascular, respiratory, kidney and4iver
effects). Impact categories for chronic health effects are divided into worker and public impacts:
Occupational impact scores are based on inventory inputs and public impact scores are based on
inventory outputs.

Chronic occupational health effects

       This refers 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 and we could assume that a worker would
continue to work at a facility and incur exposures over time.  However, the inventory is based on
manufacturing one monitor and does not truly represent chronic exposure. Therefore, the chronic
health effects impact score is more 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 primary and ancillary inputs from materials processing and
manufacturing life-cycle stages.  The distinction between  pure chemicals and mixtures is made
implicitly, 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 for noncarcinogenic effects. The former HV uses cancer slope factors or cancer
weight of evidence (WOE) classifications assigned by EPA and/or the International Agency for
Research on Cancer (IARC) when no  slope factor exists.  If both an oral "and inhalation slope
factor exist, the slope factor representing the larger hazard is  chosen. Where no slope factor is
available for a chemical, but there is a WOE classification, the WOE is used to designate default
                                          3-15

-------
3.1 METHODOLOGY
hazard values as follows: EPA WOE Groups D (not classifiable) and E (noncarcinogen) 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 3-3). Similarly, materials
for which no cancer data exist, but are designated as potentially toxic, are also given a default
value of 1.

             Table 3-3. Hazard values for carcinogenicity weight-of-evidence
                               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 • 1
Probable human carcinogen (limited human data)
Probable human carcinogen (from animal data)
Possible human carcinogen
Not classifiable
Noncarcinogenic or probably not carcinogenic
Hazard
Value
1
1
1
1
0
0
N/A: not applicable.

       The cancer hazard value for chronic occupational health effects is the greater of the
following:
                              oral:   (HVr
                                                   oral SF
                       inhalation:   (HV.
                                        CA
                                         •inhalation
                               inhalation SF{
                          '   inhalation SF___
where:
oral
oralSFm
inhalation SFS
inhalation SFm
equals the cancer oral hazard value for chemical i (unitless);
equals the cancer oral slope factor for chemical i (mg/kg-day);
equals the geometric mean cancer slope factor of all available slope factors
(0.71 mg/kg-day);
equals the cancer inhalation hazard value for chemical i (unitless);
equals the cancer inhalation slope factor for chemical i (mg/kg-day)"1; and
equals the geometric mean cancer inhalation slope factor of all available
inhalation slope factors (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 K, Table K-10.
                                         3-16

-------
                                                                  3.1 METHODOLOGY
       The noncarcinogen HV is based on either no-observed-adverse-effect levels (NOAELs)
or lowest-observed-adverse-effect levels (LOAELs). The noncarcinogen HV is the greater of the
inhalation and oral HV:
            oral:
                                            ll(oral NOAEL)
                                         -
                    inhalation:  (HV,
                                    NCit
                                  ll(inhal NOAEL)
                                 IKinhalNOAEL^J
where:
oralNQAELi
oralNOAELm,
inhalNOAEL;
inhalNOAELm.
      equals the noncarcinogen oral hazard value for chemical i (unitless);
      equals the oral NOAEL for chemical i (mg/kg-day);
      equals the geometric mean oral NOAEL of all available oral NOAELs
      (11.88 mg/kg-day);
      equals the noncarcinogen inhalation hazard value for chemical i (unitless);
      equals the inhalation NOAEL for chemical i (mg/m3); and
      equals the geometric mean inhalation NOAEL of all available inhalation
      NOAELs (68.67 mg/kg-day).
The oral and inhalation NOAEL mean values are the geometric means of a set of chemical data
presented in Appendix K, Table K-8. 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 sum of the carcinogen and noncarcinogen HVs for a particular chemical is multiplied
by the applicable inventory input to calculate the impact score:
where:
HVNC
Amt
    TCinput
                         (ISCHO)i = f(HVCA + HVNO
equals the impact score for chronic occupational health effects for chemical i (tox-
kg) per functional unit;
equals the hazard value for carcinogenicity for chemical i;
equals the hazard value for chronic noncancer effects for chemical z; and
equals the amount of toxic inventory input (kg) per functional unit for chemical i.
                                         3-17

-------
3.1 METHODOLOGY
Chronic public health effects

       For chronic public health effects, the impact score represents a surrogate for potential
health effects to residents living near a facility, including cancer, from long-term repeated
exposure to toxic or carcinogenic agents.  Impact scores are based on the identity and amount of
toxic chemical outputs with dispositions to air and water.1  As .stated above, 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 the occupational impacts in that inventory
outputs are used as opposed to inventory inputsl Note that this basic screening level scoring does
not incorporate the fate and transport of the chemicals. The chronic public health effects impact
score is calculated as follows:
where:
I$CHP
Amt
    TCmaput
equals the impact score for chronic human health effects to the public for
chemical i (tox-kg) per functional unit; and
equals the amount of toxic inventory output of chemical i to air and water (kg) per
functional unit.
Aesthetic impacts (odor)

       This refers to impacts that detract from the quality of the local environment from a human
perspective. Characterization in this project is based on odor.  Impact scores are based on the
identity and amount of odor-causing chemicals (Heijungs et al, 1992; EPA 1992), released to the
air and their odor threshold value (OTV) (Heijungs et al, 1992) (Appendix K, Table K-7). This
approach does not score chemicals as is done for toxic chemicals. The OTV is specific to a
chemical, but does not use an equivalency factor that is based on a reference chemical or a hazard
value based on a mean OTV. In this case, the OTV is a concentration which, when divided into
the mass output of a chemical, results in an impact score in units of volume of malodorous air:
where:
KM

Amtoc

OTV
equals the aesthetics impact score for chemical i (m3 malodorous air) per
functional unit;
equals the amount of odor-causing output for chemical i released to air (mg) per
functional unit; and
equals the odor threshold value for chemical i (mg/m3) (Appendix K, Table K-10).
       1 Disposition could be to groundwater. For example, a landfill model could have releases mat go to
groundwater.
                                       -—                  .__    ,       .

-------
                                                                     3.1 METHODOLOGY
Note that this impact assessment methodology determines the volume of malodorous air created
if there is no dilution. In reality, many of the air releases reported in the LCI may occur at
concentrations below the chemical's odor threshold.

3.1.2.13  Ecotoxicity

       Ecotoxicity refers to effects of chemical outputs on nonhuman living organisms.  Impact
categories include ecotoxicity impacts to aquatic and terrestrial ecosystems.

Aquatic toxicity

       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 et al, 1997) combined
with the inventory amount. Both acute and chronic impacts are combined into the aquatic
toxicity term. The hazard values (HVs) for acute and chronic toxicity are based on LC50 and
NOAEL toxicity data, respectively, mostly from toxicity tests in fathead minnows (Pimephales
promelas) (Swanson et al.,  1997). The acute fish HV is calculated by:
where:
HVFA        equals the hazard value for acute fish toxicity for chemical i (unitless);
LCSOi        equals the lethal concentration to 50% of the exposed fish population for
             chemical i; and
             equals the geometric mean LC50 of available fish LC50 values in Appendix K,
             Table K-8 (23.45 mg/L).
The chronic fish HV is calculated by:
                                            HNOAELi
                                          IINOAEL_._
where:
HVFC
NOAEL;
NOAELm
equals the hazard value for chronic fish toxicity for chemical i;
equals the no observed adverse affect level for fish for chemical i; and
equals the geometric mean NOAEL of available fish NOAEL values in
Appendix K, Table K-7 (3.90 mg/L).
                                          3-19

-------
3.1 METHODOLOGY
The aquatic toxicity impact score is calculated as follows:
where:
IS*,,
              equals the impact score for aquatic ecotoxicity for chemical i (tox-kg) per
              functional unit; and
              equals the toxic inventory output amount of chemical i to water (kg) per
              functional unit.
Terrestrial toxicity

       Toxicity measures for mammals (primarily rodents) are used to represent potential
adverse effects to organisms living in the terrestrial environment from exposure to a toxic
chemical.  Impact scores are based on the identity and amount of toxic chemicals as outputs to
air and surface water. Impact characterization is based on chronic toxicity hazard values
combined with the inventory amount. The terrestrial toxicity impact score is based on the same
noncancer chronic data used for human health because underlying data are from the same
mammal studies (see Section 2.1.2.12 for the HVNC term). The cancer hazard value was not
included in the terrestrial impact score as it is based on ranking for potential human
carcinogenicity. The terrestrial toxicity impact score is as follows:
where:
Amt
    'TCoutput
              equals the impact score for terrestrial toxicity for chemical i (tox-kg) per
              functional unit; and
              equals the toxic inventory output amount of chemical i (kg) per functional unit.
3.1.2.14 Summary of impact score equations

       Table 3-4 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.
                                         3-20

-------
                                                                     3.1 METHODOLOGY
                 Table 3-4.  Summary of natural resources impact scoring
Impact Category

Use of renewable
resources
Use/depletion of
nonrenewable
materials
Energy use, general
energy consumption
Solid waste landfill
use
Hazardous waste
landfill use
Radioactive waste
landfill use
Impact Score Approach

KRR = AmtRR X (1 - RG)
ENRR- AmtNRRx(l-RC)
ISE = AmtE or (AmtF x H/D)
ISSWL = Amtsw/D
^SHWL =AmtHW/D
ISRWL = AmtRW / D
Data Required from Inventory
(per functional unit)
Inputs
Material mass (kg)
(e.g., water)
Material mass (kg)
Energy (MJ)
(electricity, fuel)
none
none
none
Outputs
none
none
none
solid waste mass (kg) and
density (i.e., volume, m3)
hazardous waste mass (kg)
and density (i.e., volume,
m3)
radioactive waste mass (kg)
and density (i.e., volume,
m3)
Abbreviations: RC = recycled content; H = heat value of fuel i; D = density of fuel i.

       The term abiotic ecosystem refers to the nonliving environment that supports living
systems. Table 3-5 presents the impact categories, impact score equations, and inventory data
requirements for abiotic environmental impacts to atmospheric resources.

               Table 3-5. Summary of atmospheric resource impact scoring
Impact Category

Global warming
Stratospheric ozone
depletion
Photochemical smog
Acidification
Air quality (particulate
matter)
Impact Score Approach

ISGw = EFow x Amtoo
IS0D = EFODP x AmtoDC
ISpocp =EFPOCP x AmtpOC
KAP = EFfj, 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 PMjo and TSP are equal; however, using TSP will overestimate PM
                                                            10.
       Table 3-6 presents the impact categories, impact score equations, and required inventory
data for abiotic environmental impacts to water resources.
                                           3-21

-------
             3.1 METHODOLOGY
                              Table 3-6. Summary of water resource impact scoring
Impact Category

Water eutrophication
Water quality (BOD)
Water quality (TSS)
Impact Score Approach

KEUTR = EFEP x AmtEC
ISBOD = AmtBOD
•l^Tss = Amtrss
Data Required from Inventory
(per functional unit)
Inputs
none
none
none
Outputs
amount of each eutrophication chemical
released to water
amount of BOD in each wastewater
stream released to surface water
amount of suspended solids (TSS) in
each wastewater stream released to
surface water
                   Table 3-7 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.

                       Table 3-7. Summary of human health and ecotoxicity impact scoring
Impact
Category
Chronic human
health effects -
occupational
Chronic human
health effects -
public
Aesthetic
impacts (odor)
Aquatic toxicity
Terrestrial
toxicity
Impact Score Equations
18^0 = (HV^ + HV^x
Amtronpat
Kan, = (HVcA + HV^x
Amtrcomput
KAS = Amtoc/OTV
ISAQ=(HVFA + HVFC)x
AmtTCou^)UtiWaler
ISjER = HVNC x Amtj-coutput
Data Required from Inventory
(per functional unit)
Inputs
mass of each
primary and
ancillary toxic
chemical
none
none
none
none
Outputs
none
mass of each toxic
chemical released to
air and surface water
mass of odorous
chemicals released to
air
mass of each toxic
chemical released to
surface water
mass of each toxic
chemical
released to air or
surface water
Chemical
Properties Data
Required

WOEorSF
and/or mammal
NO AEL or
LOAEL
WOE or SF
and/or mammal
NOAELor
LOAEL
human odor
threshold values
fish LC50 and/or
fishNOAEL
mammal
NOAEL
.
                                                   3-22

-------
                                                                       3.1 METHODOLOGY
3.1.2.15 Aggregation of impact scores     ,

       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 CRT desktop monitor over its life).
       The following example illustrates how impacts are calculated. 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;
•      Aquatic toxicity; and
•      Terrestrial toxicity.

       Despite the output types being waterborne releases, the water eutrophication 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:
— (-H *CA:toluene ~*~ "•*NOtoluene) X •"Jn^TCoutput:toluene
= (HVCA:benzo(a)pyrene ^
        JCHP:benzo(a)pyrene
       Table 3-8 presents toxicity data for the example chemicals from Appendix K, Table K-8.
Using benzo(a)pyrene as an example, the hazard values are calculated as follows:
Table 3-8. 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)a
none
3.1"
7.3C
Chronic noncancer effects
Oral
(mg/kg-day)
100b
no data
Inhalation
(mg/m3)
411.1"
no data
1 inhalation SF
"NOAEL
c oral SF
                                           3-23

-------
3.1 METHODOLOGY
Cancer effects:
                               oral:
                                             oralSFt
                                           oralSF ,_
HV
   CAoral:benzo{a)pyrene
                      ; 7.3 (mg/kg-day)-1 -i- 0.71 (mg/kg-day)'1
                      : 10.3
                                                    inhalation SF.
                         inhalation:   (HVr.  ). = —
                                           <-i,_i'i
                                                  inhalation SF_
TTV
•"•v CAinhalation:benzo(a)pjTene
                     = 3.1 (mg/kg-day)'1 -r 1.7 (mg/kg-day)"1
                     = 1.82
Thus, the cancer HV is 10.3, the greater of the two values.
Noncancer effects:

       Since no data are available for noncancer effects, a default HV of one is assigned,
representative of mean toxicity.

TotalHV:
       Thus the total hazard value for benzo(a)pyrene is given by:
       HV,
           bcnzo(a)pyrene
                     = HVCA + HVNC
                     = 10.3 +1
                     = 11.3
Similarly, the HV for toluene is found to be 0.12.  Given the following hypothetical output
amounts:
                             = 1.3 kg of toluene per functional unit
       •^tTC-o:BENZO(A)PYRENE  = 0. 1 kg of benzo(a)pyrene per functional unit

the resulting impact scores are as follows:
!S
  CHP-W:TOLUENE
                             = 0.12 x 1.3   = 0.16 tox-kg of toluene per functional unit
                             = 1 1-3 x 0.1   =1.13 tox-kg of benzo(a)pyrene per functional unit
                                           3-24

-------
                                                                      3.1 METHODOLOGY
If these were the only outputs from Process A relevant to chronic public health effects, the total
impact score for this impact category for Process A would be:
       •••^
         CHP:PROCESS_A
                           = 0.16 + 1.13
                           = 1.29 tox-kg per functional unit for Process A.

If the product system Y contained three processes altogether (Processes A, B, and C), and the
impact scores for Process B and C were 2.5 and 3.0, respectively, impact scores would be added
together to yield a total impact score for the product system relevant to, chronic public health
effects:
       IS
         CHP:PROHLE_Y
IS
— IScHP:PROCESS_A T A>JCHPJROCESS_B
= 1.29 + 2.5 + 3.0
                                                            'CHP:PROCESS_C

                           = 6.8 tox-kg 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.

3.1.3  Data Sources and Data Quality

       Data that are used to calculate impacts are from:  (1) equivalency factors or parameters
used to identify hazard values; and (2) LCI items. Equivalency factors and data used to develop
hazard values, which have been presented in this methodology, include GWP, ODP, POCP, AP,
EP, WOE, SF, mammalian LOAEL/NOAEL, OTV, fish LC50, and fish NOAEL. Published lists
of the chemical-specific parameter values exist for GWP, ODP, POCP, AP, EP and OTV (see
Appendix K).  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., HEAST, IRIS, HSDB), then other databases (e.g., RTECS), 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 CDP are presented in Appendix K, Table K-8.  The sources of each parameter presented in
this report, and the basis for their values, are presented in Table 3-9. Data quality is affected by
the type of data source (e.g., primary versus secondary data), the currency of the data, and the
accuracy and precision of the data, and will depend on the source. 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 3.3, Baseline LCIA Results.

3.1.4  Limitations and Uncertainties

       This section summarizes some of the limitations and uncertainties in LCIA methodology,
in general.  Specific limitations and uncertainties in each impact category are discussed in
Section 3.3 with the baseline LCIA results.
       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
                                           3-25

-------
3.1 METHODOLOGY
or temporal relationships linking the inventory to specific impacts. The LCIA is intended to
provide a screening-level evaluation of impacts. More detailed characterization of exposure and
toxicity has been conducted on selected materials for the CDP in Chapter 4.

            Table 3-9. Data sources for equivalency factors and hazard values
Parameter
Global wanning potential (GWP)
Ozone depletion potential (ODP)
Photochemical oxidant creation
potential (POCP)
Acidification potential (AP)
Nutrient enrichment/eutrophication
potential (EP)
Weight-of-evidence (WOE)
Slope factor (SF)
Mammalian: Lowest observed
adverse effect level / No observed
adverse effect level
(LOAEL/NOAEL)
Fish lethal concentration to 50% of
the exposed population (LC50)
EishNOAEL
Odor threshold value (OTV)
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 (C10£L2iaOii
-------
                                                                      3.1 METHODOLOGY
       Uncertainties also are inherent in chemical ranking and scoring systems, such as the
scoring of inherent properties approach used lor human nealth 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.  In addition, when LOAELs were not available but NOAELs were, a factor of ten was
applied to the NO AEL to estimate the LOAEL, introducing uncertainty. The human health and
ecotoxicity impact characterization methods presented here are screening tools that cannot
substitute for more detailed risk characterization methods. However, it should be noted that in
LCA, chemical toxicity is often not considered at all. This 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 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).  The limitations
and uncertainties in the inventory data were discussed further in Chapter 2.
                                          3-27

-------
3.2 DATA MANAGEMENT AND ANALYSIS SOFTWARE
3.2   DATA MANAGEMENT AND ANALYSIS SOFTWARE

      The inventory and chemical characteristics data for the CDP are stored in a database
within a software package developed by UT, using the Microsoft Visual FoxPro application
programming language, under a cooperative agreement with the EPA Office of Research and
Development. The software package calculates impact scores based on the stored inventory and
chemical data and on the appropriate formulas for each impact category, as presented in
Section 3.1.                              ;
                                      3-28

-------
                                                            3.3 BASELINE LCIA RESULTS
3.3    BASELINE LCIA RESULTS

       This section presents the baseline LCIA results calculated using the impact assessment
methodology presented in Section 3.1. As noted in the section on baseline LCI results (Section
2.7.1), the baseline scenario meets the following conditions:

•      uses the effective life use stage scenario (e.g., use stage calculations are based on the
       actual amount of time a monitor is used by one or multiple users before it reaches its final
       disposition);                 .
•      uses the average value of all the energy inputs from the primary data for glass
       manufacturing;
•      removes two outliers from the primary data for energy inputs during LCD panel/module
       manufacturing and then uses the average of the remaining energy inputs;
•      excludes transportation in the manufacturing stage, but includes any transportation
       embedded in upstream data sets;
•      includes the manufacturing processes of materials used as fuels (e.g., natural gas, fuel oil)
       in the manufacturing stage instead of in  the upstream, materials processing stage. In
       cases where materials normally considered to be fuels are used as ancillary materials,
       their manufacturing processes are included with other upstream processes; and
•      assumes LCD glass manufacturing processes use the same amounts of energy as CRT
       glass manufacturing per kilogram of glass produced.

       Section 3.3.1 summarizes the baseline life cycle impact category indicators for both the
CRT and LCD. Sections 3.3.2 through 3.3.14 present a breakdown of the impact category
indicators by life-cycle stage,  list the materials that contribute 99% of the total for both monitor
types, and discuss limitations  and uncertainties in each impact category. Each of the tables in
this report shows the top contributors to the impacts because the complete tables, which are
provided in Appendix J, are often lengthy.  Section 3.3.15 summarizes the top contributors to
each impact category, and Appendix M presents complete LCIA results.

3.3.1   Summary of Baseline LCIA Results

       Table 3-10 presents the baseline CRT and LCD LCIA indicator results for each impact
category, calculated using the impact assessment methodology presented in Section 3.1. The
indicator results presented in the table are the result of the characterization step of LCIA
methodology where LCI results are converted to common units and aggregated within an impact
category. Note that the impact category indicator results are in a number of different units and
therefore can not be summed or compared across impact categories. Note also that the CDP
LCIA methodology does not perform the optional LCIA steps of normalization (calculating the
magnitude of category indicator results relative  to a reference value), grouping (sorting and
possibly ranking of indicators), or weighting (converting and possibly aggregating indicator
results across impact categories). Ranking and weighting, in particular, are subjective steps that
depend on the values of the different individuals, organizations, or societies performing the
analysis. Since the CDP involves a variety of stakeholders from different geographic regions and
with different values, these more subjective steps were intentionally excluded from the CDP
LCIA methodology.

-------
33 BASELINE LCIA RESULTS
                 Table 3-10. Baseline life-cycle impact category indicators3
Impact category
Renewable resource use
Nonrenewable resource use
Energy use
Solid waste landfill use
Hazardous waste landfill use
Radioactive waste landfill use
Global wanning
Ozone depletion
Photochemical smog
Acidification
Air particulates
Water eutrophication
Water quality, BOD
Water quality, TSS
Radioactivity
Chronic health effects, occupational
Chronic health effects, public
Aesthetics (odor)
Aquatic toxicity
Terrestrial toxicity
Units per monitor
kg
kg
MJ
m3
m3
m3
kg-CO2 equivalents
kg-CFC-11 equivalents
kg-ethene equivalents
kg-SO2 equivalents
kg
kg-phosphate equivalents
kg
kg
Bq
tox-kg
tox-kg
m3
tox-kg
tox-kg
CRT
1.31E+04
6.68E+02
2.08E+04
1.67E-01
1.68E-02
1.81E-04
6.95E+02
2.05E-05biC
1.71E-01
S.25E+00
3.01E-01
4.82E-02
1.95E-01
8.74E-01
3.85E+07d
9.34E+02
1.98E+03
7.58E+06
2.25E-01
1.97E+03
LCD
2.80E+03
3.64E+02
2.84E+03
5.43E-02
3.61E-03
9.22E-05
5.93E+02
1.37E-05b
1.41E-01
2.96E+00
1.15E-01
4.96E-02
2.83E-02
6.15E-02
1.22E+07d
6.96E+02
9.02E+02
5.04E+06
5.19E+00
8.94E+02
a Bold indicates the larger value within an impact category when comparing the CRT and LCD.
b Several of the substances included in this category were phased out of production by January 1,1996. Excluding
phased out substances decreases the CRT ozone depletion indicator to 1.09E-05 kg CFC-11 equivalents per monitor
and the LCD ozone depletion indicator to 1.18E-05 kg CFC-11 equivalents per monitor. These Ozone depletion
indicators are probably more representative of the CDP temporal boundaries and current operating practices.  See
Section 3.3.6 for details.
c Although the CRT indicator appears larger than the LCD indicator, uncertainties in the inventory make it difficult
to determine which monitor has the greater value. Therefore, this value is not shown in bold.
d Radioactivity impacts are being driven by radioactive releases from nuclear fuel reprocessing in France, which are
included in the electricity data in some of the upstream, materials processing data sets. See Section 3.3.12 for
details.
       As shown in the table, under the baseline conditions the CRT indicators are greater than
the LCD indicators in the following categories: renewable resource use, nonrenewable resource
use, energy use, solid waste landfill use, hazardous waste landfill use, radioactive waste landfill
use, global warming, photochemical smog, acidification, air particulates, biological oxygen
demand (BOD), total suspended solids (TSS), radioactivity, chronic public health effects, chronic
occupational health effects, aesthetics, and terrestrial toxicity. The LCD indicators are greater
than the CRT indicators in the following categories: water eutrophication and aquatic toxicity. In
addition, as noted in Table 3-10, the CRT ozonp depletion indicator is greater than that of 'the
LCD when phased out substances are left in the CRT and LCD inventories. However, if phased
                                            3-30

-------
                                                             3.3 BASELINE LCIA RESULTS
out substances are removed from the CRT and LCD inventories, the LCD ozone depletion
indicator would exceed that of the CRT.             h
       A number of the impact results for both monitor types, and for the CRT hi particular, are
being driven by a few data points with relatively high uncertainty. Therefore, sensitivity analyses
of the baseline results are presented in Section 3.4.

3.3.2  Renewable and Nonrenewable Resource Use

3.3.2.1 Renewable resource use

       Figure 3-2 presents the CRT and LCD impact category indicators for renewable resource
use by life-cycle stage, based on the impact assessment methodology presented in Section
3.1.2.1. Tables M-l and M-2 in Appendix M present complete renewable resource results for the
CRT and LCD, respectively.  A renewable resource is one that is being replenished at a rate
greater than or equal to its rate of depletion. Note that several of the resources listed in the
Appendix and in the tables that follow are not renewable or can not be replenished, per se, but
are considered renewable since they can be restored or are present in nearly infinite, non-
depletable amounts. For example, water is typically considered a renewable resource since it can
be restored to potable quality and is therefore being "replenished" at a rate greater than or equal
to its rate of depletion.  However, current trends toward shortages of potable water suggest that
water might be more appropriately classified as a nonrenewable resource.
                                Renewable resource use
                                 11,500
                                       -17
         -2,000 -I
CRT
Monitor type
     2,130
264 ^BB  426


         LCD
                                                                        -16
                   Figure 3-2.  Renewable resource use impacts by life-cycle stage
                                          3-31
                                                                       DEOL
                                                      250
                                       -3.5
                                                                  74
                             CRT
                                                               LCD
                                          Monitor type
                    Figure 3-3. Nonrenewable resource use impacts by life-cycle
                                             stage
                                          3-33

-------
33 BASELINE LCIA RESULTS
       As shown in Figure 3-2, the baseline life-cycle impact category indicator for renewable
resource use is 13,100 kg per monitor for the CRT and 2,800 kg per monitor for the LCD. Both
the CRT and LCD renewable resource use results are dominated by the manufacturing life-cycle
stage, with manufacturing accounting for 87% and 76% of the CRT and LCD totals, respectively.
       Table 3-11 presents the life-cycle inventory items that contribute to the top 99% of the
CRT renewable resource use total. It also lists the LCI data type (primary, secondary, or
model/secondary).  As shown in Table 3-11, water used in the production of LPG clearly
dominates the CRT renewable resource use impact score.  LPG is primarily used as an energy
source in CRT glass manufacturing, indicating that the glass/frit process group is ultimately the
greatest contributor to the CRT renewable resource use impact score. Other significant
contributors include water used to produce electricity in the United States during the use of the
monitor, water used in CRT tube manufacturing, and water used in the production of steel. The
LCI data.for T Pfr nrnrhirrinn anrl gtt»^T rnfrniifor'tnri-nrr t»tv» frwm
3.3 BASELINE LCIA RESULTS
the end of its effective life.  The latter value is based on the assumption that 50% of LCDs are
sent to a solid waste landfill at the end of their effective lives.

          Table 3-19.  Top 99% of the LCD isolid waste landfill use impact score
Life-cycle stage
Use
Use
Use
Materials processing
Materials processing
Manufacturing
Manufacturing
End-of-Life
Process group
U.S. electric grid
U.S. electric grid
U.S. electric grid
Steel production, cold- '
rolled, semi-finished
Natural gas production
LCD monitor/module
mfg.
Japanese electric grid
LCD landfilling
Material
Coal waste
Dust/sludge
: Fly/bottom ash
Unspecified solid waste
Slag and ash
Waste water treatment
sludge
1 Coal waste
EOL LCD monitor,
1 landfilled
LCI data
type
Model/secondary
Model/secondary
Model/secondary
Secondary
Secondary
Primary
Model/secondary
Primary
Contribution
to impact
score*
44%
13%
11%
9.9%
; 7.7%
• 5.6%
5.0%
3.5%
*Column adds to greater than 100% due to a credit from incineration with energy recovery during the EOL life-
cycle stage.                                 I

       LCI data for LCD monitor/module manufacturing were collected by the CDP, and LCD
solid waste disposal volumes were estimated by the CDP based on the amounts and density of
LCD materials assumed to be disposed of in a solid waste landfill. Like the CRT, data for other
process groups either were modeled by the CDP from secondary data sources or came from
secondary sources.  As discussed above for the CRT, the materials processing inventories from
secondary sources (i.e., Ecobilan) include electricity generation within the materials
manufacturing processes.  The different inventories used in this project have varying
nomenclature and some of the solid waste materials listed in the table may indeed overlap.
                                          i
3.3.4.2 Hazardous  waste landfill use         \
                                          \
       Figure 3-6 presents the CRT and LCD LCIA results for the hazardous waste landfill use
impact category, based on the impact assessment methodology presented in Section 3.1.2.3.
Tables M-9 and M-10 (see Appendix M) present complete hazardous waste landfill use results
for the CRT and LCD, respectively.  Hazardous waste landfill use impacts are characterized from
hazardous waste outputs with a disposition of landfill, which includes about 83% of the 9.46 kg
of hazardous waste/functional unit generated by the CRT life cycle and about 27% of the 6.3
kg/functional unit generated by the LCD life cycle. This consumes approximately 0.017 m3 of
hazardous waste landfill space for the CRT and 0.036 m3 for the LCD, based on the .mass and
densities of the various materials. The results for both monitor types are dominated by monitor
disposal at the end  of its effective life. Approximately 46% of CRTs and 5% of LCDs are
assumed to be landfilled as hazardous waste (see Section 2.5 and  Appendix I).
                                        3-42

-------
                                                             3.3 BASELINE LCIA RESULTS
out substances are removed from the CRT and LCD inventories, the LCD ozone depletion
indicator would exceed that of the CRT.
       A number of the impact results for both monitor types, and for the CRT in particular, are
being driven by a few data points with relatively high uncertainty. Therefore, sensitivity analyses
of the baseline results are presented in Section 3.4.

3.3.2  Renewable and Nonrenewable Resource Use

3.3.2.1 Renewable resource use

       Figure 3-2 presents the CRT and LCD impact category indicators for renewable resource
use by life-cycle stage, based on the impact assessment methodology presented in Section
3.1.2.1. Tables M-l and M-2 in Appendix M present complete renewable resource results for the
CRT and LCD, respectively. A renewable resource is one that is being replenished at a rate
greater than or equal to its rate of depletion.  Note that several of the resources listed in the
Appendix and in the tables that follow are not renewable or can not be replenished, per se, but
are considered renewable since they can be restored or are present in nearly infinite, non-
depletable amounts. For example, water is typically considered a renewable resource since it can
be restored to potable quality and is therefore being "replenished" at a rate greater than or equal
to its rate of depletion.  However, current trends toward shortages of potable water suggest that
water might be more appropriately classified as a nonrenewable resource.
                                Renewable resource use
                                 11,500
                                       -17
         -2,000 J
CRT
Monitor type
     2,130
264 jgf*  426
   ^^^^^H

         LCD
                                                                        -16
                   Figure 3-2.  Renewable resource use impacts by life-cycle stage
                                          3-31



-100 -I



Figure

CRT
Monitor type
3-3. Nonrenewable resource use

stage
LCD
impacts by


life -cycle




                                          3-33

-------
 33 BASELINE LCIA RESULTS
        As shown in Figure 3-2, the baseline life-cycle impact category indicator for renewable
 resource use is 13,100 kg per monitor for the CRT and 2,800 kg per monitor for the LCD. Both
 the CRT and LCD renewable resource use results are dominated by the manufacturing life-cycle
 stage, with manufacturing accounting for 87% and 76% of the CRT and LCD totals, respectively.
        Table 3-11 presents the life-cycle inventory items that contribute to the top 99% of the
 CRT renewable resource use total. It also lists the LCI data type (primary, secondary, or
.-..Tnodel/secnndanA _AajihnsxmJLnJI!ahL= 3-11
3.3 BASELINE LCIA RESULTS
the end of its effective life. The latter value is based on the assumption that 50% of LCDs are
sent to a solid waste landfill at the end of their effective lives.

           Table 3-19. Top 99% of the LCD solid waste landfill use impact score
Life-cycle stage
Use
Use
Use
Materials processing
Materials processing
Manufacturing
Manufacturing
End-of-Life
Process group
U.S. electric grid
U.S. electric grid
U.S. electric grid
Steel production, cold- '
rolled, semi-finished
Natural gas production
LCD monitor/module
mfg.
Japanese electric grid
LCD landfilling
Material
1
\ Coal waste
Dust/sludge
;Fly/bottom ash
Unspecified solid waste
Slag and ash
Waste water treatment
: sludge
: Coal waste
EOL LCD monitor,
landfilled
LCI data
type
Model/secondary
Model/secondary
Model/secondary
Secondary
Secondary
Primary
Model/secondary
Primary
Contribution
to impact
score*
44-%
13%
11%
9.9%
, 7.7%
, 5.6%
5.0%
3.5%
 *Column adds to greater than 100% due to a credit from incineration with energy recovery during the EOL life-
 cycle stage.                                                           -     .  ;

       LCI data for LCD monitor/module manufacturing were collected by the CDP, and'LCD
 solid waste disposal volumes were estimated by the CDP based on the amounts and density of
 LCD materials assumed to be disposed of in a sblid waste landfill. Like the CRT, data for other
 process groups either were modeled by the CDP from secondary data sources or came from
 secondary sources. As discussed above for the CRT, the materials processing inventories from
 secondary sources (i.e., Ecobilari) include electricity generation within the materials;
 manufacturing processes. The different inventories used in this project have varying
 nomenclature and some of the solid waste materials listed in the table may indeed overlap.

 3.3.4.2 Hazardous waste landfill use

       Figure 3-6 presents the CRT and LCD LCIA results for the hazardous waste landfill use
 impact category, based on the impact assessment methodology presented in Section 3.1.2.3.
 Tables M-9 and M-10 (see Appendix M) present complete hazardous waste landfill use results
 for the CRT and LCD, respectively. Hazardous^ waste landfill use impacts are characterized from
 hazardous waste outputs with a disposition of landfill, which includes about 83% of the 9.46 kg
 of hazardous waste/functional unit generated by the CRT life cycle and about 27% of the 6.3
 kg/functional unit generated by the LCD life cycle. This consumes approximately 0.017 m3 of
 hazardous waste landfill space for the CRT and 0.036 m3 for the LCD, based on the mass and
 densities of the various materials. The results for both monitor types are dominated by monitor
 disposal at the end of its effective life. Approximately 46% of CRTs and 5% of LCDs are
 assumed to be landfilled as hazardous waste (see Section 2.5 and Appendix I).
                                         3-42

-------
                                                             3.3 BASELINE LCIA RESULTS
out substances are removed from the CRT and LCD inventories, the LCD ozone depletion
indicator would exceed that of the CRT.
       A number of the impact results for both monitor types, and for the CRT in particular, are
being driven by a few data points with relatively high uncertainty. Therefore, sensitivity analyses
of the baseline results are presented in Section 3.4.

3.3.2  Renewable and Nonrenewable Resource Use

3.3.2.1 Renewable resource use

       Figure 3-2 presents the CRT and LCD impact category indicators for renewable resource
use by life-cycle stage, based on the impact assessment methodology presented in Section
3.1.2.1. Tables M-l and M-2 in Appendix M present complete renewable resource results for the
CRT and LCD, respectively.  A renewable resource is one that is being replenished at a rate
greater than or equal to its rate of depletion. Note that several of the resources listed in the
Appendix and in the tables that follow are not renewable or can not be replenished, per se, but
are considered renewable since they can be restored or are present in nearly infinite, non-
depletable amounts. For example, water is typically considered a renewable resource since it can
be restored to potable quality and is therefore being "replenished" at a rate greater than or equal
to its rate of depletion.  However, current trends toward shortages of potable water suggest that
water might be more appropriately classified as a nonrenewable resource.
                                Renewable resource use
                                 11,500
                                       -17
                                                           2,130
                                                      264 fl^H  426
                                                                        -16
                              CRT         Monitor type          LCD

                   Figure 3-2.  Renewable resource use impacts by life-cycle stage
                                          3-31

-------
33 BASELINE LCIA RESULTS
       As shown in Figure 3-2, the baseline life-cycle impact category indicator for renewable
resource use is 13,100 kg per monitor for the CRT and 2,800 kg per monitor for the LCD.  Both
the CRT and LCD renewable resource use results are dominated by the manufacturing life-cycle
stage, with manufacturing accounting for 87% and 76% of the CRT and LCD totals, respectively.
       Table 3-11 presents the life-cycle inventory items that contribute to the top 99% of the
CRT renewable resource use total. It also lists the LCI data type (primary, secondary, or
model/secondary). As shown in Table 3-11, water used in the production of LPG clearly
dominates the CRT renewable resource use impact score.  LPG is primarily used as an energy
source in CRT glass manufacturing, indicating that the glass/frit process group is ultimately the
greatest contributor to the CRT renewable resource use impact score.  Other significant
contributors include water used to produce electricity in the United States during the use of the
monitor, water used in CRT tube manufacturing, and water used in the production of steel. The
LCI data for LPG production and steel manufacturing are from secondary sources, while the LCI
data for the U.S. electric grid are based on the model developed by the CDP for the amount of
electricity consumed by a CRT during use combined with data from secondary sources on the
inputs and outputs from U.S. power plants. CRT tube manufacturing LCI data are primary data
collected by the CDP.

          Table 3-11. Top 99% of the CRT renewable resource use impact score
Life-cycle stage
Manufacturing
Use
Manufacturing
Materials processing
Manufacturing
Manufacturing
Process group
LPG production
U.S. electric grid
CRT tube manufacturing
Steel production, cold-rolled,
semi-finished
Japanese electric grid
PWB manufacturing
Material
Water
Water
Water
Water
Water
Water
LCI data
type
Secondary
Model/secondary
Primary
Secondary
Model/secondary
Primary
Contribution to
impact score*
79%
8.7%
6.2%
3.6%
0.34%
0.32%
* Column may not add to 99% due to rounding.

       Table 3-12 presents the inventory items contributing to the top 99% of the LCD
renewable resource use total and the LCI data types (primary, secondary, or model/secondary).
As shown in the table, water used in LCD module/monitor manufacturing is the greatest
contributor to the LCD renewable resource use. impact score.  Other significant contributors
include water used in the production of LPG, water used by the U.S. electric grid during the use
life-cycle stage, and water used in steel production. It is LCD glass manufacturing that consumes
the LPG responsible for the high LCD renewable resource use score. The LCI data for LCD
module manufacturing are primary data collected by the CDP. LPG production and steel
manufacturing are from secondary sources, while the LCI data for the U.S electric grid are based
on the model developed by the CDP for the amount of electricity consumed by an LCD during
the use stage combined with data from secondary sources.
                                       3-32

-------
                                                           3.3 BASELINE LCIA RESULTS
Table 3-12. Top 99% of the LCD renewable resource use impact score
Life-cycle stage
Manufacturing
Manufacturing
Use
Materials processing
Manufacturing
Manufacturing "
Manufacturing
Manufacturing
Process group
LCD module/monitor mfg.
LPG production
U.S. electric grid
Steel production
(cold-rolled, semi-finished)
LCD panel components
Backlight
Japanese electric grid
PWB Manufacturing
Material
Water .
Water
Water
Water
Water
Water
Water
Water
LCI data
type
Primary-
Secondary
, Model/secondary
Secondary
Primary
Primary
Model/secondary
Primary
Contribution to
impact score*
38%
18%
15%
8.2%
6.4%
6.8%
5.3%
0.66%
* Column may not add to 99% due to rounding.
3.3.2.2 Nonrenewable resource use

      Figure 3-3 presents the CRT and LCD impact category indicators for nonrenewable
resource use by life-cycle stage, based on the impact assessment methodology presented in  ,
Section 3.1.2.1. Tables M-3 and M-4 in Appendix M present complete nonrenewable resource
results for the CRT and LCD, respectively. The total nonrenewable resource use indicator was
668 kg per monitor for the CRT and 364 kg per monitor for the LCD.  As shown in Figure 3-3,
the CRT nonrenewable resource use results are dominated by the manufacturing life-cycle stage,
which contributed 68% of the total. The LCD nonrenewable resource use score is dominated by
the upstream materials processing stages, which contributed 69% of the total.
                              Nonrenewable resource use
                                 451
                                      -3.5
                             CRT
LCD
                                         Monitor type
                   Figure 3-3.  Nonrenewable resource use impacts by life-cycle
                                         3-33

-------
33 BASELINE LCIA RESULTS
       Table 3-13 presents the inventory items contributing to the top 99% of the CRT
nonrenewable resource use impact score. It also lists the LCI data type (primary, secondary, or
model/secondary). Similar to the renewable resource use LCIA results, the LPG production
process, which mainly supports the CRT glass manufacturing process, clearly dominates the CRT
nonrenewable resource use impact score. Petroleum used to make LPG is the non-renewable
resource being consumed by the LPG production process in the greatest amounts, followed by
natural gas, and coal.  Note that the LPG actually consumed during CRT glass manufacturing
does not appear in the nonrenewable resource use results. This is because it was accounted for in
the nonrenewable resource use score for the LPG production process when it was extracted from
the ground.
       Fuels (coal and natural gas) consumed by the U.S. electric grid during monitor use are
also among the greatest contributors to the CRT nonrenewable resource use impact scores. The
LCI data for LPG production are from secondary sources, while the LCI data for the U.S. electric
grid are based on the model developed by the CDP for the amount of electricity consumed by a
CRT during use combined with data from secondary sources on the inputs and outputs from U.S.
power plants.

        Table 3-13.  Top 99% of the CRT nonrenewable resource use impact score
Life-cycle stage
Manufacturing
Use
Manufacturing
Use
Manufacturing
Materials processing
Materials processing
Manufacturing
Use
Manufacturing
Manufacturing
Manufacturing
Materials processing
Manufacturing
Materials processing
Manufacturing
Process group
LPG production
U.S. electric grid
LPG production
U.S. electric grid
LPG production
Steel production, cold-
rolled, semi-finished
Steel production, cold-
rolled, semi-finished
Fuel oil #6 production
U.S. electric grid
Natural gas production
U.S. electric grid
Japanese electric grid
Aluminum production
Japanese electric grid
Polycarbonate
production
Japanese electric grid
Material*
Petroleum (in ground)
Coal, average (in ground)
Natural gas (in ground)
Natural gas
Coal, average (in ground)
Iron Ore
Coal, average (in ground)
Petroleum (in ground)
Petroleum (in ground)
Natural gas (in ground)
Coal, average (in ground)
Coal, average (in ground)
Bauxite
Petroleum (in ground)
Natural gas (in ground)
Natural gas
LCI data
type
Secondary
Model/secondary
Secondary
Model/secondary
Secondary
Secondary
Secondary
Secondary
Model/secondary
Secondary
Model/secondary
Model/secondary
Secondary
Model/secondary
Secondary
Model/secondary
Contribution
to impact
score*
56%
27%
6.7%
2.1%
2.0%
0.99%
0.60%
0.58%
0.57%
0.51%
0.43%
0.34%
0.20%
0.19%
0.19%
0.19%
 * Column may not add to 99% due to rounding.
                                       3-34

-------
                                                            3.3 BASELINE LCIA RESULTS
       Table 3-14 presents the inventory items contributing to the top 99% of the LCD non-
renewable resource use impact score. In this case, the impact score is dominated by the natural
gas extracted to produce natural gas in the upstream, materials processing life-cycle stage.
Liquified natural gas (LNG) from this production process is used as an ancillary material in the
LCD module/monitor manufacturing process group, indicating LCD module/monitor
manufacturing is ultimately responsible for this non-renewable resource use. However, only one
of the seven companies that provided data for the LCD module/monitor manufacturing process
group reported this use of LNG.  Note that the actual use of LNG in the LCD
module/manufacturing process group.does not appear in the nonrenewable resource results.
Similar to the LPG results discussed above for the CRT, this is because it has been accounted for
in the natural gas production process results.
       Other primary contributors to this impact score include coal used to produce electricity
for the U.S. electric grid, and petroleum used to produce LPG. The LCI data for all of the
primary contributors to the LCD non-renewable resource use score were either from secondary
sources or CDP models combined with secondary sources.
Table 3-14. Top 99% of the LCD nonrenewable resource use impact score
Life-cycle stage
Materials processing
Use ..
Manufacturing
Manufacturing
Manufacturing
Use
Manufacturing
Manufacturing
Materials processing
Manufacturing
Materials processing
Materials processing
Use
Process group
Natural gas production
U.S. electric grid
LPG production
Japanese electric grid
Natural gas production
U.S. electric grid
Japanese electric grid
Japanese electric grid
Steel production (cold-
rolled, semi-finished)
LPG production
Steel production (cold-
rolled, semi-finished)
Natural gas production
U.S. electric grid
Material
Natural gas (in ground)
Coal, average (in ground)
Petroleum (in ground)
Coal, average (in ground)
Natural gas (in ground)
Natural gas
Petroleum (in ground)
Natural gas
Iron ore
Natural gas (in ground)
Coal (in ground)
Coal (in ground)
Petroleum (in ground)
LCI data
type
Secondary
Model/secondary
Secondary
Model/secondary
Secondary
Model/secondary
Model/secondary
Model/secondary
Secondary
Secondary
Secondary
Secondary
Model/secondary
Contribution
to impact
score*
65%
18%
4.9%
2.1%
1.5%
1.4%
1.2%
1.2%
0.89%
0.59%
0.54%
0.45%
0.39%
 *Column may not add to 99% due to rounding

3.3.2.3 Limitations and uncertainties

       The renewable and nonrenewable resource use results presented here are based on the
mass of a material consumed. Depletion of renewable 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 scores. This may be particularly
important for water, which, while considered a renewable resource, is in shorter and shorter
supply as world population grows and more of the world's water resources become degraded.
                                          3-35

-------
33 BASELINE LCIA RESULTS
For the nonrenewable materials use category, depletion of materials results from the extraction of
nonrenewable resources. However, the impact scores do not directly relate consumption rates to
the earth's ability to sustain that consumption.
       The CRT and LCD impact scores for renewable resource use, and the CRT impact score
for nonrenewable resources use, are being driven by the fuels consumed during CRT or LCD
glass manufacturing. However, as discussed in Section 2.3.3.3, there is a high degree of
variability in the three sets of CRT glass manufacturing energy data received by the CDP.
Furthermore, as discussed in Section 2.3.3.1, LCD glass manufacturing data were developed
from the CRT data because no companies were willing to supply the LCD data. Therefore, glass
energy use inputs are uncertain for both the CRT and the LCD and were the subject of a
sensitivity analysis, discussed in Section 3.4.
       The LCD impact score for nonrenewable resource use is being driven by LNG used as an
ancillary material during LCD module/monitor manufacturing.  However, only one LCD
module/monitor manufacturer reported using LNG as an  ancillary material, which was confirmed
by CDP researchers in follow-up communications. Given the fact that only one of seven
manufacturers reported the ancillary use of LNG, the LCD nonrenewable resource use indicator
may not be representative of the industry as a whole. If we remove this application of LNG from
the LCD inventory, the LCD nonrenewable resource result is reduced by 66%, from 364 kg per
monitor to 125 kg per monitor.
       Inventory data for most of the materials contributing 99% of the CRT and LCD impact
scores come from  secondary sources, and were not developed specifically for the CDP.  The
limitations and uncertainties associated with secondary data sources are summarized in Section
2.2.2.  Table 3-15 looks more closely at the LPG and natural gas production geographic and
temporal boundaries. These are the production processes that are driving a large part of the CRT
and LCD resources use indicators. As shown in the table, most of the LPG and natural gas
production data are for the United States, although the LPG data set includes some data from
other countries. Both data sets rely on several different sources and have different temporal
boundaries. In particular, LPG production data are less recent, and may not accurately reflect
current production practices. All of these factors create some inconsistencies among the data sets
and reduce the data quality when  used for the purposes of the CDP. However, this is a common
difficulty with LCA, which often uses data from secondary sources to avoid, the tremendous
amount of time and resources required to collect all  the needed data.

     Table 3-15. LPG and natural gas production geographic and temporal boundaries
Production Process
LPG production
Natural gas production
Location
Mainly U.S., but includes some
other countries
U.S.
Source
Seven sources cited
Six sources cited
Year
1983 to 1993
1987 to 1998
                                       3-36

-------
                                                            3.3 BASELINE LCIA RESULTS
 3.3.3  Energy Use
                                    - '"•  ''?'-.-         **•"-" "iri|>

       Figure 3-4 presents the CRT and LCD impact results for energy use by life-cycle stage,
 based on the impact assessment methodology presented in Section 3.1.2.2.  Tables M-5 and M-6
 in Appendix M list complete energy use results for the CRT and LCD, respectively.  The total
 indicator for this impact category was 20,800 MJ per monitor for the CRT,  and 2,840 MJ per
 monitor for the LCD.
       CRTs generally are assumed to have greater life-cycle energy use impacts than the LCDs
 due to the high energy requirements in the use stage. This is borne out by the results in
 Figure 3-4, which show that CRT energy consumption during use is roughly 2.7 times that of the
 LCD. However, contrary to expectations, CRT energy use impacts are driven by the
 manufacturing life-cycle stage, which contributes about 88% of the total score. The use stage,
 which was expected to be responsible for a large amount of energy consumption impacts, only
 contributes about 11% of the total score. LCD energy consumption impacts are also largest in
 the manufacturing life-cycle stage which accounts for almost 51% of the impacts in this category.
 Both the use and upstream (materials processing) life-cycle stages are also significant
 contributors to LCD life-cycle energy use, accounting for 30 and 22%, respectively. Note that the
 sum of the upstream, manufacturing, and use life-cycle stages is greater than 100% due to an
 energy credit for incineration with energy recovery at the end of a monitor's useful life.
                                       Energy
                                                              LCD
                                         Monitor type

                          Figure 3-4.  Energy impacts by life-cycle stage
3.3.3.1 Major contributors to the CRT energy use results

       Table 3-16 presents the life-cycle inventory items contributing to the top 99% of the CRT
energy use results and the LCI data type (primary, secondary, or model/secondary).  As shown in
the table, LPG used in the glass/frit process group, primarily from CRT glass manufacturing,
clearly dominates the CRT energy use result, followed by electricity consumed during use of a
CRT monitor, and natural gas, petroleum, and coal consumed during LPG production. Since
LPG is used primarily as an energy source during CRT glass manufacturing, most of the sum of
the glass/frit manufacturing and LPG production energy use impacts—roughly 87% of the CRT
life-cycle energy use impacts—can be attributed to the CRT glass manufacturing process.
                                         3-37

-------
33 BASEUGNE LCIA RESULTS
               Table 3-16. Top 99% off the CRT energy use impact score
Life-cycle stage
Manufacturing
Use
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Materials processing
Manufacturing
Manufacturing
Manufacturing
Process group
Glass/frit manufacturing
CRT monitor use
LPG production
LPG production
LPG production
CRT tube manufacturing
Glass/frit manufacturing
Glass/frit manufacturing
Glass/frit manufacturing
CRT tube manufacturing
Polycarbonate production
LPG production
CRT tube manufacturing
Glass/frit manufacturing
Material
Liquified petroleum gas
Electricity
Natural gas (in ground)
Petroleum (in ground)
Coal, average (in ground)
Fuel oil #6
Natural gas
Fuel oil #2
Electricity
Natural gas
Natural gas (in ground)
Uranium (U, ore)
Electricity
Electricity
LCI data
type
Primary
Model/secondary
Secondary
Secondary
Secondary
Primary
Primary
Primary
Primary
Primary
Secondary '
Secondary
Primary
Primary
Contribution
to impact
score*
72%
11%
10%
2.0%
1.4%
0.72%
0.26%
0.24%
0.23%
0.18%
0.16%
0.16%
0.15%
0.13% -
 *Column may not add to 99% due to rounding.
3.3.3.2 Major contributors to the LCD energy use results

      Table 3-17 lists the inventory items contributing to the top 99% of the LCD life-cycle
energy use results. Electricity consumed during use of the LCD monitor by the consumer is the
single largest contributor to LCD energy use impacts, closely followed by LPG utilized to
produce LCD glass. Other major contributors include natural gas consumed during natural gas
production, electricity and LNG used as a fuel during LCD monitor/module manufacturing, and
natural gas consumed during LPG production. Note that the LNG used as an ancillary material in
LCD module/monitor manufacturing is not included in the LCD energy use impact calculations
since it is not used as a source of energy. However, natural gas used as an energy source to
produce the LNG is included (the third item listed in the table).
                                       3-38

-------
                                                            3.3 BASELINE LCIA RESULTS
                Table 3-17.  Top 99% of the LCD energy use impact score
Life-cycle stage
Use
Manufacturing
Materials processing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Materials processing
Materials processing
Manufacturing
Manufacturing
Materials processing
Manufacturing
Manufacturing
Materials processing
Materials processing
Materials processing
Materials processing
Manufacturing
Materials processing
Materials processing
Manufacturing
Materials processing
Materials processing
Process group
LCD monitor use
LCD glass manufacturing
Natural gas production
LCD module/monitor mfg.
LCD module/monitor mfg.
- LPG production
LCD panel components
LCD module/monitor mfg.
Natural gas production
Natural gas production
LCD module/monitor mfg.
LPG production
Polycarbonate production
LPG production
LCD module/monitor mfg.
PMMA sheet production
PMMA sheet production
Steel production
(cold-rolled, semi-finished)
Steel production
(cold-rolled, semi-finished)
Natural gas production
PMMA sheet production
Steel production
(cold-rolled, semi-finished)
LCD module/monitor mfg.
Styrene-butadiene copolymer
production
Aluminum production
Material
Electricity
Liquified petroleum gas
Natural gas (in ground)
Electricity
Liquified natural gas
Natural gas (in ground)
Electricity
Natural gas
Coal (in ground)
Petroleum (in ground)
Liquified petroleum gas
Petroleum (in ground)
Natural gas (in ground)
Coal (in ground)
Kerosene
Petroleum (in ground)
Natural gas (in ground)
Petroleum (in ground)
Natural gas (in ground)
Natural gas (in ground)
Electricity
Electricity
Fuel oil #4
Natural gas (in ground)
Coal (in ground)
LCI data
type
Model/
secondary
Primary
Secondary
Primary
Primary
Secondary
Primary
Primary
Secondary
Secondary
Primary
Secondary
Secondary
Secondary
Primary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Primary
Secondary
Secondary
Contribution
to impact
score*
30%
25%
14%
8.9%
5.8%
3.4%
1.4%
1.3%
1.3%
0.89%
0.88%
0.69%
0.67%
0.51%
0.46%
0.42%
0.41%
0.36%
0.35%
0.32%
0.32%
0.31%
0.31%
0.29%
0.29%
*Column may not add to 99% due to rounding.

3.3.3.3 Limitations and uncertainties

       Some of the limitations and uncertainties in the energy use indicators are similar to those
in the renewable and nonrenewable resource use categories. First, as discussed in Section
3.3.1.1, the energy data for both CRT and LCD glass manufacturing are uncertain due to the
variability in the primary glass data received by the CDP from three glass manufacturers.  Glass
manufacturing energy data are the subject of a sensitivity analysis in Section 3.4.  Second, data
for LPG production and natural gas production, which are among the largest contributors  to the
                                          3-39

-------
33 BASELINE LCIA RESULTS	

energy use indicators, are from secondary sources and are therefore subject to the limitations and
uncertainties associated with secondary data (see Section 2.2.2 and 3.3.1.1). Not counting the
use stage data, note that about 14% of the CRT energy use impacts shown in Table 3-16, above,
are from secondary sources, compared to about 24% of the LCD energy use impacts
(Table 3-17).
       The amount of electricity consumed during use of a monitor was modeled by the CDP
from secondary sources on the amount of electricity consumed during different power modes and
the amount of time a monitor spends in each mode. Data  quality for the effective life scenario
(the baseline scenario presented here) is considered to be excellent, based on the source and
quality information detailed in Appendix H and discussed in Sections 2.4.2 and 2.4.3.

33.4  Landfill Use

3.3.4.1 Solid waste landfill use

       Figure 3-5 presents the CRT and LCD LCIA results for the solid waste landfill use impact
category, based on the impact assessment methodology presented in Section 3.1.2.3.  Tables
M-7 and M-8 in Appendix M present complete results for the CRT and LCD, respectively. Life-
cycle solid waste landfill use was 0.17 m3 for the CRT and 0.054 m3 for the LCD. The solid
waste landfill indicators for both monitor types are dominated by waste disposal during the use
stage—which contributes 59% of the total for the CRT and 68% for the LCD—primarily from
wastes generated as a by-product of electricity production. Both monitor types have negative
solid waste impact scores during the end-of-life stage. This is due to an energy credit from
incineration processes, which offsets some of the solid waste impacts from electricity generation.
c 8.00&02 -
"3
ts 3.00E-02 -
-2.00E-02-
£
9.91 E-02
solid waste H Upstream
• Mfg
nUse
QEOL
3.70E-02
1. 075-02
1.10E-02
-1.32E-03 HiBill^^^H -4.35E-03
CRT LCD
Monitor type
Figure 3-5. Solid waste impacts by life-cycle stage
       Table 3-18 presents the materials that contribute to the top 99% of the CRT solid waste
landfill use impact score.  Note that the material contributions actually add to greater than 100%
due to the energy credit from incineration processes, discussed above.  Coal waste from U.S.
electricity production is the single largest contributor to CRT impacts in this impact category,
followed by slag and ash from LPG production, and dust/sludge and fly bottom ash from U.S.
electricity production. Electricity is used to power the monitor during the use stage, while LPG
is primarily used in the production of CRT glass during manufacturing. LPG production also
                                        3-40

-------
                                                             3.3 BASELINE LCIA RESULTS
 results in an unspecified solid waste that contributes 4.596 of the CRT solid waste impact score,
 while the CRT glass/frit process group generates a wastewater treatment sludge that contributes
 5% of the score.  Thus, the CRT glass/frit process group contributes about 30% of the CRT solid
 waste impact score, either directly (as a result of the glass manufacturing process itself) or
 indirectly (from LPG production). Other processes that are significant contributors include steel
 production and the landfilling of a CRT at the end of its effective life. The latter value is based
 on the assumption that 25% of CRTs that have reached their end of life are disposed of in a solid
 waste landfill (see Section 2.5 and Appendix I).
Table 3-18. Top 99% of the CRT solid waste landfill use imoact score
Life-cycle stage
Use
Manufacturing
Use
Use
Materials processing
End-of-Life
Manufacturing
Manufacturing
Process group
U.S. electric grid
LPG production
U.S. electric grid
U.S. electric grid
Steel production, cold-
rolled, semi-finished
CRT landfilling
LPG production
CRT glass/frit mfg.
Material
Coal waste
Slag and ash
Dust/sludge
Fly bottom ash
Unspecified solid waste
EOL CRT monitor, landfffled
Unspecified waste
Waste water treatment sludge
LCI data
type
Model/secondary
Secondary

Model/secondary
Secondary
Primary
Secondary
Primary
Contribution
to impact
score*
38%
21%
12%
10%
6.6%
5.0%
4.5%
4.4%
•••Column adds to greater than 100% due to a credit from incineration with energy recovery during the EOL life-
cycle stage.
       CRT glass manufacturing data were collected specifically for the CDP, while data for
other process groups were either modeled by the CDP from secondary sources (e.g., U.S. electric
grid data) or are entirely from secondary sources (e.g., LPG and steel production data). The mass
and volume of CRT materials that are landfilled were developed for the CDP based on the mass
reported in each inventory data set (collected as primary data) and the density of CRT materials
assumed to be disposed of in a solid waste landfill. Note that the upstream inventories, which
were derived from secondary sources (i.e., Ecobilan), include electricity generation within the
materials manufacturing processes. These inventories do not include coal waste as an output, but
list "slag and ash" as an output. The different inventories used in this project have varying
nomenclature and some of the solid waste materials listed in the table may indeed overlap.
       Table 3-19 presents the materials that contribute to the top 99% of the LCD solid waste
landfill use results. Like the CRT solid waste results, the material contributions in the table are
actually greater than 100% due to some negative values at end-of-life from an incinerator energy
credit.  Coal waste, dust/sludge, and fly/bottom ash from U.S. electricity during the use stage
dominate the LCD solid waste impacts, contributing 68% of the impact score. Other significant
contributors include the following: (1) an unspecified solid waste from producing steel used in
the manufacture of the monitor, (2) slag and ash generated during the production of natural gas
which is then used by one LCD module/monitor manufacturer as an ancillary material (LNG)
during LCD module/monitor manufacture, (3) a wastewater treatment sludge from LCD
module/monitor manufacturing, (4) coal waste from the generation of electricity in Japan during
manufacturing, and (5) landfilling of non-hazardous or non-recovered components of the LCD at
                                          3-41

-------
3.3 BASELINE LCIA RESULTS
the end of its effective life.  The latter value is based on the assumption that 50% of LCDs are
sent to a solid waste landfill at the end of their effective lives.

          Table 3-19.  Top 99% of the LCD solid waste landfill use impact score
Life-cyde stage
Use
Use
Use
Materials processing
Materials processing
Manufacturing
Manufacturing
End-of-Life
Process group
U.S. electric grid
U.S. electric grid
U.S. electric grid
Steel production, cold- :
rolled, semi-finished
Natural gas production
LCD monitor/module
mfg.
Japanese electric grid
LCD landfilling
Material
Coal waste
Dust/sludge
Fly/bottom ash
Unspecified solid waste
Slag and ash
Waste water treatment
sludge
Coal waste
EOL LCD monitor,
landfilled
LCI data
type
Model/secondary
Model/secondary
Model/secondary
Secondary
Secondary
Primary
Model/secondary
Primary
Contribution
to impact
score*
44%
13%
11%
9.9%
7.7%
5.6%
5.0%
3.5%
*Column adds to greater than 100% due to a credit from incineration with energy recovery during the EOL life-
cycle stage.

       LCI data for LCD monitor/module manufacturing were collected by the CDP, and-LCD
solid waste disposal volumes were estimated by the CDP based on the amounts and density of
LCD materials assumed to be disposed of in a solid waste landfill. Like the CRT, data for other
process groups either were modeled by the CDP from secondary data sources or came from
secondary sources. As discussed above for the CRT, the materials processing inventories from
secondary sources (i.e., Ecobilari) include electricity generation within the materials
manufacturing processes.  The different inventories used in this project have varying
nomenclature and some of the solid waste materials listed in the table may indeed overlap.

3.3.4.2 Hazardous waste landfill use

       Figure 3-6 presents the CRT and LCD LCIA results for the hazardous waste landfill use
impact category, based on the impact assessment methodology presented in Section 3.1.2.3.
Tables M-9 and M-10 (see Appendix M) present complete hazardous waste landfill use results
for the CRT and LCD, respectively. Hazardous waste landfill use impacts are characterized from
hazardous waste outputs with a disposition of landfill, which includes about 83% of the 9.46 kg
of hazardous waste/functional unit generated by the CRT life cycle and about 27% of the 6.3
kg/functional unit generated by the LCD life cycle. This consumes approximately 0.017 m3 of
hazardous waste landfill space for the CRT and 0.036 m3 for the LCD, based on the mass and
densities of the various materials. The results for both monitor types are dominated by monitor
disposal at the end of its effective life. Approximately 46% of CRTs and 5% of LCDs are
assumed to be landfilled as hazardous waste (see Section 2.5 and Appendix I).
                                        3-42

-------
                                                             3.3 BASELINE LCIA RESULTS
~  1.80E-02
3

1  1.20E-02 -
          6.00E-03 -
   O.OOEfOO
                                    Hazardous waste
                                       1.53&02
                           1.41&03
                   1.09E-04          O.OOE|00


                               CRT
                                                     1-07B04
                                                                LCD
                                            Monitor type

                        Figure 3-6. Hazardous waste impacts by life-cycle stage
       Table 3-20 presents the materials that contribute to the top 99% of the CRT hazardous
 waste landfill use results. About 91% of the total hazardous waste landfill space consumed
 throughout the life-cycle of the CRT is from the amount of the monitor that is assumed to be
 disposed of as hazardous waste. The next largest contributor is an unspecified hazardous waste
 from LPG production. Most of this LPG is used to manufacture CRT glass. CRT outputs to a
 hazardous waste landfill at the end-of-life were estimated by the CDP. The LPG inventory is
 from secondary sources.
	 Table 3-20. Top 99% of the CRT hazardous waste landfill „«, ™r^t wrr
Life-cycle stage
End-of-Life
Manufacturing
Process group
CRT landfillins
LPG production
Material
EOL CRT monitor, landfffled
Hazardous waste
LCI data
type
Primary
Secondary
Contribution to
impact score
91%
8.1%
       Table 3-21 lists the top contributors to the LCD hazardous waste landfill use results
LCD results are also dominated by landfilling of the LCD monitor at the end of its effective life
even though only 5% of LCDs are assumed to be landfilled.  Other significant contributors
include an unspecified hazardous waste from LPG production, and acetic acid from LCD
monitor/module manufacturing. LPG is used in the manufacture of LCD glass LCD outputs to
a hazardous waste landfill at the end-of-life were estimated by the CDP. The LPG inventory is
from secondary sources.
Table
Life-cycle
stage
End-of-Life
Manufacturing
Manufacturing
3-21. Top 99% of the LCD hazardous waste landfill use im
Process group
LCD landfilling
LPG production
LCD monitor/module mfg.
Material
EOL LCD monitor, landfilled
Hazardous waste
Acetic acid
LCI data
type
Primary
Secondary
Primary
pact score
Contribution to
impact score
97%
1.8%
0.88%
                                         3-43

-------
33 BASELINE LCIA RESULTS
3.3,4.3 Radioactive waste landfill use

       Figure 3-7 presents the CRT and LCD LCIA results for the radioactive waste landfill use
impact category, based on the impact assessment methodology presented in Section 3.1.2.3.
Tables M-l 1 and M-12 in Appendix M present complete results for the CRT and LCD,
respectively. Life-cycle radioactive waste landfill use indicators for the CRT are 1.81E-04 rh3
per monitor for the CRT and 9.22E-05 m3 per monitor for the LCD. As shown in the figure,
CRT radioactive waste landfill impacts are dominated by radioactive waste disposal in the use
stage, which contributes about 79% of the total impacts. This result is to be expected, given the
relatively large amount of electricity consumed by a CRT during use and the associated
radioactive waste from nuclear power plants. The use stage also contributes the greatest amount
of LCD impacts (58%), but the manufacturing stage is also a significant contributor (33%) due to
electricity consumed during manufacturing. LCD manufacturing electricity is linked to the
Japanese electric grid, which derives 31% of its power from nuclear sources.  By comparison,
about 20% of the U.S. electric grid is powered by nuclear sources.
^ 1.50&04n
3
"g 1.00&04-
          5.00E-05-
                              Radioactive waste

                           1.43&04
             2.74&05  L12E-Q5
                                                                           @ Upstream
                                                                           • Mfg
                                                                           rjEOL
                                                                    5.32E-05
                                                             3.04&I
                                                    8.55E-06
                                                                          1.01&08
                          CRT
                                                                 LCD
                                             Monitor type

                        Figure 3-7. Radioactive waste impacts by life-cycle stage
        Table 3-22 lists the materials that contribute.to the top 99% of the CRT radioactive waste
 landfill use score and the LCI data type. Note that the LCI data for all of the materials in the table
 are from secondary sources or models.  Low-level radioactive waste and depleted uranium from
 the U.S. electric grid are the radioactive materials being landfilled in the greatest quantities,
 followed by low-level radioactive waste from the production of steel used in the monitor. The
 latter radioactive waste is a byproduct of electricity production used in the manufacture of steel.
 It should be noted that the electricity generation data utilized in the steel inventory are from
 France (Glazebrook, 2001), where a large percentage of electricity is derived from nuclear
 sources. Therefore, these emissions may not be representative of emissions from steel production
 in some parts of Asia or in the United States. This issue is discussed further in the section on
 limitations and uncertainties, below.
                                          3-44

-------
                                                             3.3 BASELINE LCIA RESULTS
Table 3-22. Top 99% of the CRT radioactive waste landfill use imnact score
Life-cycle stage
Use
Use
Materials processing
Manufacturing
Materials processing
Materials processing
Manufacturing
Manufacturing
Process group
U.S. electric grid
U.S. electric grid
Steel production, cold-
rolled, semi-finished
Japanese electric grid
Invar
Ferrite manufacturing
Japanese electric grid
U.S. electric grid
Material
Low-level radioactive waste
Uranium, depleted
Low-level radioactive waste
Low-level radioactive waste
Low-level radioactive waste
Low-level radioactive waste
Uranium, depleted
Low-level radioactive waste
LCI data
type
Model/secondary
Model/secondary
Secondary
Model/secondary
Secondary
Secondary
Model/secondary
Model/secondary
^Column may not add to 99% due to rounding.
Contribution
to impact
61%
18%
9.0%
3.8%
2.6%
2.5%
1.1%
0.97%

       Other significant contributors to the CRT radioactive waste score include low-level
radioactive waste from electricity used in Japan during manufacturing, and low-level radioactive
waste from invar and ferrite manufacturing. Invar is an alloy of nickel and iron. Like the steel
data discussed above, the invar and ferrite manufacturing data also include emissions from
electricity production. Finally, low-level radioactive waste from U.S. electricity consumed during
the manufacturing stage contributes slightly less than 1% of the CRT radioactive waste landfill
use impacts. The frit and PWB manufacturing processes consume this electricity. These are the
only CRT components linked to the U.S. grid.
       Table 3-23 lists the materials that contribute to the top 99% of the LCD radioactive waste
landfill use results and the LCI data type. Note that LCI data for all of the primary contributors
to this impact category are from secondary sources or modeled from secondary sources.
Together, low-level radioactive waste and depleted uranium disposal from electricity consumed
during use of the monitor account for about 57% of the LCD radioactive waste landfill use
indicator, followed by low-level radioactive waste and depleted uranium disposal from electricity
used during manufacturing (roughly 32%). Waste disposal from steel production is also  a
significant contributor at 8.7%. Like the CRT data discussed above, these emissions occur from
electricity production in France and may not be representative of U.S. or some Asian practices.
Table 3-23. Top 99% of the LCD radioactive waste disposal impact score
Life-cycle stage
Use
Manufacturing
Use
Materials processing
Manufacturing
Process group
U.S. electric grid
Japanese electric grid
U.S. electric grid
Steel production, cold-
rolled, semi-finished
Japanese electric grid
Material
Low-level radioactive waste
Low-level radioactive waste
Uranium, depleted
Low-level radioactive waste
Uranium, depleted
LCI data
type
Model/secondary
Model/secondary
Model/secondary
Secondary
Model/secondary
^Column may not add to 99% due to rounding.
Contribution
to impact
score*
44%
25%
13%
8.7%
7.5%

                                         3-45

-------
3.3 BASELINE LCIA RESULTS
3.3.4.4 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, hazardous, or radioactive) is accepted.  Landfill
use impacts are characterized from solid, hazardous, or radioactive 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. Note that
different countries may have different landfill designations for the final disposition of similar .
waste streams (e.g., a waste considered hazardous in the U.S. may be accepted in a solid waste
landfill elsewhere). However, where possible, equivalent landfills (e.g., special waste landfills
and hazardous waste landfills) were considered for these impact categories.
       CRT and LCD impact results for the solid and radioactive waste landfill use categories
were driven almost entirely by waste outputs reported in inventories from secondary sources.
These inventories were not developed specifically for the CDP and therefore are subject to the
limitations and uncertainties associated with secondary data (see Section 2.2.2 and 3.3.2). La
particular, radioactive waste disposal from some of the upstream materials processing data may
not be representative of conditions in the U.S. or parts of Asia. These data include emissions
from electricity production within the materials inventory, which are the primary source of
radioactive waste streams.  For example, steel production data include the French electric grid,
where a large percentage of the power supply comes from nuclear power plants. In addition,
some of the upstream data may not be representative of current conditions, with steel production
data covering the period from 1975 to 1990 and invar production data being from 1991.
       CRT and LCD impact results for the hazardous waste landfill use category were
dominated by monitor disposal at the end of their effective lives. Hazardous waste landfill
disposal volumes were estimated based on the percent of monitors with hazardous waste
landfilling as their final disposition, the monitor mass, and the material densities. However, data
on the percentage of CRTs that are landfilled are not separated into hazardous and non-hazardous
landfilling processes. Therefore, these percentages were estimated by the CDP, as described in
Appendix I. Even less is known about the final disposition of LCDs, particulary since very few
LCD desktop monitors have reached the end of their effective lives (and then, only if they have
been damaged in some way). Therefore, the effect of different LCD EOL dispositions was
evaluated in a sensitivity analysis (see Section 3.4.)

3.3.5  Global Wanning

       Figure 3-8 presents the CRT and LCD LCIA results for the global warming impact
category, based on the impact assessment methodology presented in Section 3.1.  Tables  M-13
and M-14 in Appendix M list complete  global warming results for the CRT and LCD,
respectively.  The life-cycle global warming indicators for the CRT and LCD were 695 and 593
kg of CO2 equivalents per monitor, respectively. The CRT global warming indicators are (driven
by the use life-cycle stage, which contributes about 66% of the total. The manufacturing stage,
which contributed 88% of the CRT energy consumption impacts, only contributes about 29% of
the total global warming score. LCD global warming impacts, on the other hand, have the
greatest contribution from the manufacturing life-cycle stage, which accounts for about 40% of
the potential impacts in this category. Both the upstream (materials processing) and use
                                        3-46

-------
                                                            3.3 BASELINE LCIA RESULTS
 life-cycle stages are also significant contributors to the LCD global warming results, accounting
 for 31% and 29% of the total, respectively.
                                      Global warming
                                   459
                                         1.3
                                                        182
                               CRT                              LCD
                                            Monitor type

                       Figure 3-8. Global warming impacts by life-cycle stage
       One might expect the distribution of global warming impacts across life-cycle stages to
mirror those of energy consumption, as CO2 is generally a large emission from electricity
generation. However, as discussed in Section 3.3.3, CRT energy impacts are greatest in the
manufacturing stage due to the large amounts of energy used to manufacture glass. Since the
energy used in glass manufacturing is not only from electricity, but more so from other fuels
(LPG, natural gas, and fuel oil), there is not a direct correlation between CRT global warming
impacts and CRT energy impacts.
      The distribution of LCD global warming impacts across life-cycle stages does mirror the
distribution of LCD energy use impacts discussed in Section 3.3.3. However, as discussed
below, the manufacturing stage global warming impacts for the LCD are being driven more by
the use of sulfur hexafluoride (SF6) in LCD monitor/module manufacturing than by the use of
electricity.

3.3.5.1 Major contributors to the CRT global warming results

      Table 3-24 presents the life-cycle inventory items contributing to the top 99% of the CRT
global warming results and the LCI data type (primary, secondary, or model/secondary).  As
shown in the table, CRT global warming impacts are dominated by CO2 emissions from
electricity generation during use of the monitor, followed by CO2 and methane emissions from
producing LPG used as fuel in the CRT glass/frit process group. Together these three emissions
contribute almost 89% of the CRT life-cycle global warming score.  Carbon dioxide and methane
emissions from a number of other processes also add to the CRT global warming score, as does
nitrous oxide emissions from the LPG production process. It is likely that most of the CO2
emissions from the materials processing life-cycle stage can be attributed to emissions from
electricity generation or fuel combustion. As discussed in Section 2.2.2.1, the upstream materials
processing inventories used in this study include data from electricity generation.
                                         3-47

-------
33 BASELINE LCIA RESULTS
       Note that almost all of the LCI data for global warming emissions are from secondary
sources. This is because the CRT global warming results are dominated by CO2 emissions from
electricity generation, and electric grid data were either developed by the CDP from secondary
sources or already included in the upstream, materials processing inventories from secondary
sources.

             Table 3-24. Top 99% of the CRT global warming impact score
Life-cycle stage
Use
Manufacturing
Manufacturing
Manufacturing
Use
Materials processing
Manufacturing
Manufacturing
Materials processing
Materials processing
End-of-Life
Manufacturing
Materials processing
Materials processing
Process group
U.S. electric grid
LPG production
LPG production
Japanese electric grid
U.S. electric grid
Steel production, cold-rolled,
semi-finished
U.S. electric grid
LPG production
Polycarbonate production
Aluminum production
CRT incineration
CRT glass/frit mfg.
Invar
Styrene-butadiene copolymer
production
Material
Carbon dioxide
Carbon dioxide
Methane
Carbon dioxide
Methane
Carbon dioxide
Carbon dioxide
Nitrous oxide
Carbon dioxide
Carbon dioxide
Carbon dioxide
Carbon dioxide
Carbon dioxide
Carbon dioxide
LCI data
type
. Model/secondary
Secondary
Secondary
Model/secondary
Model/secondary
Secondary
Model/secondary
Secondary
Secondary
Secondary
Secondary
Primary
Secondary
Secondary
Contribution
to impact
score*
64%
22%
2.5%
2.2%
1.9%
1.9%
1.0%
0.72%
0.66%
0.52%
0.51%
0.40%
0.33%
0.24%
*Column may not add to 99% due to rounding.
3.3.5.2 Major contributors to the LCD global warming results

       Table 3-25 presents the life-cycle inventory items contributing to the top 99% of the LCD
global warming results and the LCI data type (primary, secondary, or model/secondary). Sulfur
hexafluoride used in LCD module manufacturing is the single largest contributor to LCD global
warming impacts, followed by CO2 emissions from electricity generation during the use stage,
CO2 and methane emissions from natural gas production, and CO2 emissions from the generation
of electricity used during manufacturing in Japan.
                                       3-48

-------
                                                            3.3 BASELINE LCIA RESULTS
Table 3-25. Top 99% of the LCD global warming impact score
Life-cycle stage
Manufacturing
Use
Materials processing
Materials processing
Manufacturing
Manufacturing
Materials processing
Use
Materials processing
Materials processing
Manufacturing
End-of-Life
Process group
LCD monitor/module mfg.
U.S. electric grid
Natural gas production
Natural gas production
Japanese electric grid
LPG production
Steel production, cold-
rolled, semi-finished
U.S. electric grid
PMMA sheet production
Polycarbonate production
Natural gas production
LCD incineration
Material
Sulfur hexafluoride
Carbon dioxide
Carbon dioxide
Methane
Carbon dioxide
Carbon dioxide
Carbon dioxide
Methane
Carbon dioxide
Carbon dioxide
Carbon dioxide
Carbon dioxide
LCI data
type
Primary
Model/secondary
Secondary
Secondary
Model/secondary
Secondary
Secondary
Model/secondary
Secondary
Secondary
Secondary
Secondary
Contribution to
impact score*
29%
28%
16%
12%
8.7%
1.2%
1.1%
0.85%
0.45%
0.43%
0.35%
0.35%
*Column may not add to 99% due to rounding.
       Sulfur hexafluoride is a potent global warming gas, with a global warming potential
(GWP) equivalancy factor of 23,900 CO2 equivalents (see Table K-2 in Appendix K).  It is used
as an etchant in a dry-etching process of amorphus silicon and SiNx films.  The CO2 and methane
emissions from natural gas production can be attributed to the use of LNG as an ancillary
material in LCD monitor/module manufacturing. However, as discussed in the section on non-
renewable resource use (Section 3.3.2), only one LCD module manufacturer reported this use of
LNG.
       Carbon dioxide emissions (and, in one case, methane emissions) round out the remainder
of the primary contributors to the LCD global warming indicator.  Most of the carbon dioxide
emissions occur from upstream processes and are due to electricity generation.  With the
exception of the SF6 data, the LCI data for all of the top LCD global warming emissions are from
secondary sources.  Sulfur hexaflouride emissions data were developed by the CDP based on an
emissions factor (0.45) applied to SF6 inputs reported by LCD monitor/module manufacturers.
The emissions factor is from the Intergovernmental Panel on Climate Change publication, Good
Practice Guidance and Uncertainty Management in National Greenhouse Gas Inventories
(Penman et al, 2000.)

3.3.5.3 Limitations and uncertainties

       Global warming potential (GWP) refers to the warming that emissions of certain gases
may contribute by building up in the atmosphere and trapping the earth's heat. As discussed in
Section 3.1.2.4, 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 a computer monitor
would be within the 100-year timeframe.
                                         3-49

-------
3.3 BASELINE LCIA RESULTS
       The effects of the buildup of global wanning gases in the atmosphere is still the subject of
scientific debate, but in 1995 the Intergovernmental Panel on Climate Change (IPCC),
representing the consensus of most climate scientists worldwide, concluded that"... the balance
of evidence—suggests that there is a discernible human influence on global climate (IPCC,
1995)." Other than the limitations and uncertainties  inherent in predicting future effects, most of
the limitations and uncertainties in the CRT and LCD global warming results have to do with the
LCI data on greenhouse gas emissions, which occur primarily from electricity generation
processes.
       As noted above, the U.S. and Japan electric grid inventories used in the CDP were
developed by the CDP, and electric grids used with upstream processes are embedded in the
upstream inventories. U.S. electric grid emissions of CO2 are based on data in the EPA
publication, National Air Quality and Emissions Trends Report, 1997 (EPA, 1998), which were
the best data available when the electric grid inventory data was developed and are expected to
be reasonably accurate. However, the Japanese electric grid inventory was derived from the U.S.
inventory based on the mix of fuels used in Japan. Because Japanese power plants may employ
different pollution control devices, use fuels of different quality, or other factors, their
greenhouse gas emissions could actually be higher or lower than those reported in the inventory.
       Similarly, the electric grid inventories embedded in upstream, materials processing
inventories may have differing geographic and temporal boundaries or may be representative of
older technologies. Therefore, actual emissions of greenhouse gases could be higher or lower
than reported. This is a common limitation of LCAs, which must often rely on secondary data
sources to avoid the considerable time and resources required to collect primary data for every
process.

3.3.6  Stratospheric Ozone Depletion

       Figure 3-9 presents, the CRT and LCD LCIA results for the stratospheric ozone depletion
impact category by life-cycle stage, based on the impact assessment methodology presented in
Section 3.1.2.5. Note that most of the CRT ozone depletion impacts occur from the use stage
(50%) and the upstream, materials processing stages (45%), while most of the LCD. ozone
depletion impacts occur from the manufacturing stage (63%). As will be shown later, this is
important because upstream and use stage data are primarily from secondary data sources,
whereas manufacturing data were collected by the CDP.  Tables M-15 and M-16 in
Appendix M list complete stratospheric ozone depletion results for the CRT and LCD,
respectively.
                                        3-50

-------
                                                           3.3 BASELINE LCIA RESULTS

3.00E-05 -i
.> *•
§. '| 2.00E-05 -

-------
33 BASELINE LCIA RESULTS	

for the glass/Mt process group), 1,1,1-trichloroethane emissions from electricity generation in the
use stage, and bromomethane emissions from electricity used in manufacturing.

       Table 3-26.  Top 99% of the CRT stratospheric ozone depletion impact score
Life-cycle stage
Use
Materials processing
Materials processing
Materials processing
Manufacturing
Materials processing
Materials processing
Use
Manufacturing
Process group
U.S. electric grid
ABS production
Aluminum production
Invar
LPG production
Lead
Steel production, cold-
rolled, semi-finished
U.S. electric grid
U.S. electric grid
Material
. Bromomethaneb
HALON-1301b
HALON-1301b
HALON-1301b
Bromomethaneb
HALON-1301b
HALON-1301b
1,1, l-Trichloroethaneb
Bromomethaneb
LCI data
type
Model/secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Model/secondary
Model/secondary
Contribution to
impact score3
49%
20%
14%
5.9%
3.7%
2.6%
2.0%
1.1%
0.78%
* Column may not add to 99% due to rounding.
b Class I substance as listed in Tide VI of the Clean Air Act Amendments.

       Note that all of the materials listed in Table 3-26 are Class I ozone depleting substances.
As discussed above, all of these substances except bromomethane were phased out of production
by January 1,1996. Note also that all of the LCI data for the materials in the table are from
secondary sources. For both of these reasons, the LCI data for the materials in the table are   .
highly uncertain. This is discussed further under limitations and uncertainties, below.

3.3.6.2 Major contributors to the LCD ozone depletion results

       Table 3-27 lists the top contributors to the LCD life-cycle stratospheric ozone depletion
indicator and the LCI data type.  Together HCFC-225cb and HCFC 225ca used in the LCD panel
components process group account for 59% of the LCD ozone depletion indicator.  Note that
HCFC 225cb and HCFC 225ca are Class n ozone depleting substances that are not scheduled for
phaseout until 2015.  [Under U.S. regulations and the Montreal Protocol and its Amendments and
Adjustments these substances can not be produced or imported after 2015, except for use as
refrigerants in equipment manufactured before January 1,2020 (EPA, 200 lb).]  Also note that
the impact scores for these materials are based on primary LCI data collected from
manufacturers. Therefore, these data are considered to be  more reliable than data for Phase I
substances from secondary sources.
                                        3-52

-------
                                                              33 BASELINE LCIA RESULTS
Table 3-27. Top 99% of the LCD stratospheric ozone depletion II«P ™pa^ c™r^
Life-cycle stage
Manufacturing
Use
Manufacturing
Materials processing
Manufacturing
Materials processing
Process group
LCD panel components
U.S. electric grid
LCD panel components
Aluminum production
(virgin)
Japanese electric grid
Steel production, cold-
rolled, semi-finished
Material
HCFC-225cbb
Bromomethane0
HCFC-225cab
HALON-1301C
Bromomethane0
HALON-1301C
LCI data
type
Primary-
Model/secondary
Primary
Secondary
Model/secondary
Secondary
Contribution to
impact score9
34%
27%
25%
7.8%
3.1%
1.4%
b Class H substance as listed in Title VI of the Clean Air Act Amendments.
c Class I substance as listed in Title VI of the Clean Air Act Amendments.
      i

       Other significant contributors to the LCD ozone depletion score include bromomethane
emissions from electricity generation in the United States and Japan, and halon emissions from
upstream, materials processing stages. As noted above in the discussion of CRT ozone depletion
results, the halons were phased out of production in 1996, which suggests that these data are not
representative of current conditions. Bromomethane is still being produced, but bromomethane
emissions data are also uncertain as will be discussed in the section on limitations and
uncertainties, below.

3.3.6.3 Limitations and uncertainties

       Both the CRT and LCD life-cycle stratospheric ozone depletion results are highly
uncertain due to the inclusion of a number of Class I ozone depleting substances in inventories
from secondary sources.  As discussed above, except for bromomethane, developed countries
that are parties to the Montreal Protocol and its Amendments and Adjustments phased out the
production of Class I substances  by 1996.  To better assess the uncertainties in these results,
Table 3-28 lists the geographic and temporal boundaries for the life-cycle inventories of the
process groups listed in tables 3-26 and 3-27, above.
                                         3-53

-------
33 BASELINE LCIA RESULTS
 Table 3-28. Geographic and temporal boundaries of inventories contributing to the CRT
                      and/or LCD ozone depletion indicator results
Process group
ABS production
Aluminum production
Invar production
Japanese electric grid
LCD panel components
Lead production
LPG production
Steel production
U.S. electric grid
Geographic boundaries
Germany, Italy, Netherlands
Not provided
Multiple countries
U.S. and Japan"
Japan
Not provided
Mainly U.S.
Multiple countries
U.S.
Temporal boundaries
1997
Not provided
1991 (nickel), Not provided (lead)
1993b
1998
Not provided
1983-1993
1975-1990
1993b
1 Based on the U.S. electric grid inventory modified to account for the fuel mix used in Japan.
b Date of stack tests from which bromomethane emission factor was developed (from EPA Web site: Emission
Factor Documentation for AP-42 Section 1.1: Bituminous and Subbituminous Coal Combustion.
http^/www.epa.gov/ttn/chief/ap42/ch01/bgdocs/b01s01.pdf.)

       The most recent data are for LCD panel components manufacturing, which are primary
data collected from manufacturers in Japan by the CDP and expected to be of better quality than
older data from secondary sources. Data for ABS production are also fairly recent, dating from
1997. However, the temporal boundaries for most of the data are either not listed in the
inventories, or pre-date the Class I substance production phase out. In addition, most of the data
are from Europe and/or the United States, where very few Class I ozone depleting substances are
currently used.  Thus, we suspect that emissions of Class I substances reported in the inventories
no longer occur, indicating that the CRT and LCD life-cycle impact results should be reduced to
2.05E-05 kg of CFC-11 equivalents per monitor for the CRT, and 1.37E-05 kg of CFC-11
equivalents per monitor for the LCD.
       Bromomethane is a Class I ozone depleting substance that has not yet been phased out of
production and is emitted during coal combustion to produce electricity. Bromoethane emissions
from electricity production are estimated from an emission factor reported in AP-42, the EPA
compilation of air pollutant emission factors (EPA, 1996). EPA (1996) provides an emission
factor rating that is, "an overall assessment of how good a factor is, based on both the quality of
the test(s) or information that is the source of the factor and on how well the factor represents the
emission source." The bromomethane emissions factor rating is "D," or below average,
indicating CDP data quality for bromomethane emissions from electricity generation is also
below average.
        In conclusion, it appears that most of the Class I substance emissions data are highly
 uncertain or of below average quality.  Manufacturing data collected by the CDP, which includes
 emissions of Class IE substances, are of better quality and expected to be more representative of
 current conditions.  When all data are included in the ozone depletion calculations, the CRT has
 a greater ozone depletion impact score than  the LCD.  However, if we remove phased out
 substances from the inventory, the results are switched with the LCD having a greater score in
 this category than the CRT.
                                         3-54

-------
                                                             3.3 BASELINE LCIA RESULTS
 3.3.7  Photochemical Smog

        Figure 3-10 presents the CRT and LCD LCIA results for the photochemical smog impact
 category by life-cycle stage. These results were calculated using the impact assessment
 methodology presented in Section 3.1.2.6. Tables M-17 and M-18 in Appendix M list complete
 results for the CRT and LCD, respectively. One 17" CRT monitor produces 0.171 kg of ethene
 equivalents throughout its life cycle, while a functionally equivalent 15" LCD monitor produces
 0.141 of ethene equivalents. The CRT photochemical smog impact score is dominated by
 emissions during the manufacturing stage (71% of total); the LCD impact score is dominated by
 emissions during the upstream, materials processing stages (91% of total). However, as
 discussed below, it is fossil fuel production processes that emit the majority of smog'forming
 emissions during the life-cycle of either monitor type. Both the CRT and LCD receive a slight
 credit on emissions of smog forming chemicals at the end of their effective lives due to energy
 recovery from incineration processes.
                                    Smog formation
                                                               LCD
                                           Monitor type

                        Figure 3-10. Smog formation impacts by life-cycle stage
3.3.7.1  Maior contributors to the CRT photochemical smog results

       Table 3-29 lists the materials that contribute to the top 99% of the CRT photochemical
smog indicator result.  The LPG production process alone, which emits various unspeciated
hydrocarbons, benzene, aldehydes, ethane, and formaldehyde, accounts for almost 67% of CRT
photochemical smog impacts. As noted earlier in the discussion of other impact category
indicators, most of this LPG is used as a fuel source in CRT glass manufacturing.  However,
CRT glass energy data reported in the three data sets received by the CDP were highly variable
and therefore the subject of a sensitivity analysis (see Section 3.4).
       Other materials responsible for more than  1% of the CRT photochemical smog score
include the following:  (1) hydrocarbon (methane  and nonmethane) emissions associated with
steel production, (2) methane emissions from electricity generation during the use stage,
(3) toluene emissions from CRT tube manufacturing, (4) nonmethane hydrocarbon emissions
associated with ABS production, and (5) nonmethane hydrocarbon emissions associated with
polycarbonate production. Note that the inventory for each upstream material (e.g., steel
                                         3-55

-------
3.3 BASELINE LCIA RESULTS
production, ABS production, etc.) contains data from the raw materials extraction, materials
manufacture, and (usually) electricity generation processes. Therefore, hydrocarbon emissions
associated with steel production, for example, could be from one of many individual processes,
such as the steel production process itself, from fuels consumed during the mining of ore, or from
the combustion of fuels as an energy source during steel manufacturing.

           Table 3-29.  Top 99% of the CRT photochemical smog impact score
Life-cycle stage
Manufacturing
Manufacturing
Materials processing
Manufacturing
Use
Materials processing
Manufacturing
Manufacturing
Materials processing
Materials processing
Materials processing
Manufacturing
Materials processing
Manufacturing
Materials processing
Materials processing
Manufacturing
Manufacturing
Materials processing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Process group
LPG production
LPG production
Steel production, cold-
rolled, semi-finished
LPG production
U.S. electric grid
Steel production, cold-
rolled, semi-finished
LPG production
CRT tube mfg.
ABS production
Polycarbonate
production
Aluminum production
Natural gas production
ABS production
LPG production
Stryene-butadiene
copolymer production
Invar
Fuel oil #6 production
LPG production
Lead
Fuel oil #6 production
Natural gas production
LPG production
CRT tube mfg.
Material
Hydrocarbons, unspeciated
Nonmethane hydrocarbons,
unspeciated
Nonmethane hydrocarbons,
unspeciated
Methane
Methane
Hydrocarbons, unspeciated
Benzene
Toluene
Nonmethane hydrocarbons,
unspeciated
Nonmethane hydrocarbons,
unspeciated
Nonmethane hydrocarbons,
unspeciated
Nonmethane hydrocarbons,
unspeciated
Nonmethane hydrocarbons,
unspeciated
Aldehydes
Hydrocarbons, remaining
unspeciated
Nonmethane hydrocarbons,
unspeciated
Hydrocarbons, unspeciated
Formaldehyde
Nonmethane hydrocarbons,
unspeciated
Nonmethane hydrocarbons,
unspeciated
Methane
Ethane
Xylene (mixed isomers)
LCI data
type
Secondary
Secondary
Secondary
Secondary
Model/secondary
Secondary
Secondary
Primary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Primary
Contribution
to impact
score*
36%
25%
19%
3.4%
2.6%
2.0%
1.6%
1.3%
1.3%
1.1%
0.86%
0.53%
0.41%
0.39%
0.39%
0.32%
0.31%
0.29%
0.21%
0.21%
0.21%
0.18%
0.17%
                                        3-56

-------
                                                            3.3 BASELINE LCIA RESULTS
Table 3-29. Top 99% of the CRT photochemical smog impact score
Life-cycle stage
Manufacturing
Manufacturing
Process group
LPG production
Natural gas production
Material
Pentane
Benzene
•'•Column may not add to 99% due to rounding.
LCI data
type
Secondary
Secondary

Contribution
to impact
score*
0.16%
0.14%

3.3.7.2 Major contributors to the LCD photochemical smog results

       Table 3-30 lists the materials that contribute to the top 99% of the LCD life-cycle
photochemical smog results. LCD results are dominated by unspeciated hydrocarbon emissions
methane emissions, and benzene emissions from natural gas production in the materials
processing life-cycle stage, which together account for about 75% of the total. This natural gas is
used as an ancillary material by one LCD monitor/module manufacturer. Other LCD
monitor/module manufacturers reported using LNG as a fuel, but not as an ancillary number.  A
number of other materials and processes contribute more than 1% of the total LCD
photochemical smog score, most notably unspeciated, nonmethane hydrocarbon emissions
associated with steel production. As noted above under the CRT results, the latter hydrocarbon
emissions could occur from any one of various processes (e.g., ore mining, steel production,
electricity generation, etc.) wrapped into the steel production inventory.
Table 3-30. Top 99% of the LCD photochemical smog impact score
Life-cycle stage
Materials processing
Materials processing
Materials processing

Manufacturing
Manufacturing

Materials processing
Materials processing
Use
Manufacturing

Process group
Natural gas production
Natural gas production
Natural gas production
Steel production, cold-
rolled, semi-finished
LCD monitor/module
mfg.
LPG production
LPG production
Natural gas production
Steel production, cold-
rolled, semi-finished
U.S. electric grid
Natural gas production
PMMA sheet production
Material
Nonmethane
hydrocarbons, unspeciated
Methane
Benzene
Nonmethane
hydrocarbons, unspeciated
Isopropyl alcohol
Hydrocarbons, unspeciated
Nonmethane
hydrocarbons, unspeciated
Hydrocarbons, unspeciated
Hydrocarbons, unspeciated
Methane
Nonmethane
hydrocarbons, unspeciated
Nonmethane
hydrocarbons, unspeciated
LCI data
type
Secondary
Secondary
Secondary
Secondary
Primary
Secondary
Secondary
Secondary
Secondary
Model/secondary
Secondary
Secondary
Contribution
to impact
score*
45%
17%
12%
11%
2.5%
2.1%
1.4%
1.2%
1.2%
1.2%
1.0%
0.87%
                                        3-57

-------
33 BASELINE LCIA RESULTS
           Table 3-30. Top 99% of the LCD photochemical smog impact score
Life-cycle stage
Materials processing
Manufacturing
Materials processing
Materials processing
Process group
Polycarbonate production
Natural gas production
Aluminum production
PET resin production
Material
Nonmethane
hydrocarbons, unspeciated
Methane
Nonmethane
hydrocarbons, unspeciated
Nonmethane
hydrocarbons, unspeciated
LCI data
type
Secondary
Secondary
Secondary
Secondary
Contribution
to impact
score*
0.73%
0.39%
0.39%
0.36%
*Column may not add to 99% due to rounding.

3.3.7.3 Limitations and uncertainties

    •   Photochemical smog indicators are calculated using 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 the chemical's photochemical oxidant creation potential (POCP) compared to the
reference chemical ethylene.  As noted in Section 3.1.2.6, 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.
       The CRT impact score for photochemical smog formation is being driven by the process
for producing the large amount of LPG used in CRT glass manufacture.. However, as discussed
in Section 2.3.3.3 and previous subsections of this Section 3.3, the three sets of glass
manufacturing energy data received by the CDP were highly variable, making the average glass
energy inputs used in the baseline analysis uncertain. Therefore, the emissions of smog forming
chemicals from LPG production, which are based on the glass LPG inputs, are also uncertain.
CRT glass energy inputs were subjected to a sensitivity analysis. Sensitivity results are discussed
in Section 3.4.
       The LCD impact score for photochemical smog formation is being driven by the natural
 gas production process to produce the large amount of LNG used as an ancillary material during
 LCD monitor/module manufacturing. However, as discussed earlier, only one LCD
 monitor/module manufacturer reported this use of LNG, which indicates the average LNG inputs
 used in the LCD inventory may be unduly high.  Therefore, the mass of smog forming chemical
 emissions from the natural gas production process, which are based on the amount of ancillary
 LNG inputs, may also by unduly high.
        The majority of the CRT and LCD photochemical smog results are based on life-cycle
 inventories from secondary sources and are therefore subject to the limitations and uncertainties
 associated with secondary data,  discussed previously. In particular, see Section 3.3.1.1 for a
 detailed discussion of limitations and uncertainties in the LPG and natural gas production data.
                                         3-58

-------
                                                            3.3 BASELINE LCIA RESULTS
 3.3.8  Acidification

       Figure 3.11 presents the CRT and LCD LCIA results for the acidification impact
 category, based on the impact assessment methodology presented in Section 3.1.2.7.  Tables M-
 19 and M-20 in Appendix M list complete results for the CRT and LCD, respectively. The life-
 cycle acidification impact indicator result is 5.25 kg of SO2 equivalents per monitor for the CRT
 and 2.96 kg of SO2 equivalents per monitor for the LCD. As might be expected, acidification
 impacts are greatest in the use stage for both monitor types, due to the emissions of SOX and NOX
 from U.S. power plants. Use stage impacts are more dominant for the CRT (65% of life-cycle
 acidification impacts) than the LCD (43%) due to the relatively large amount of power consumed
 during use by the less energy-efficient CRT.

-* a- 5.000 -
> c
'5 =
er —
o g 3.000 -
8-2
w g 1.000 -
-1.000 -
Acidification
H Upstream
• Mfg
DUse
3.44 QEOL
xm — ^M^I-
CRT LCD
Monitor type
Figure 3-11. Acidification impacts by life-cycle stage




       Emissions from manufacturing processes account for 28% of CRT acidification impacts
and 38% of LCD acidification impacts. Material processing is responsible for about 7 and 19%
of CRT and LCD impacts in this category, respectively. Both technologies receive a slight credit
on acidification impacts during the EOL stage due to an energy credit from incineration
processes with energy recovery, which offset some of the impacts from electricity production.

3.3.8.1 Major contributors to the CRT acidification results

       Table 3-31 lists the materials that contribute to the top 99% of the CRT acidification
impacts score and their LCI data type. SO2 and NOX emissions from the U.S. electric grid are the
two largest contributors to the CRT acidification indicator, together accounting for 63% of the
total.  SOX and NOX emissions from the LPG production process are also significant, contributing
about 23%. As noted previously, most of the LPG from this production process is used to
manufacture CRT glass, but the mass of LPG inputs is uncertain and the basis of a sensitivity
analysis later in this chapter (see Section 3.4).
       Other materials that contribute more than one percent of the CRT acidification indicator
are SO2 emissions from invar production, hydrochloric acid emissions from electricity generation
during the use stage, and sulfur dioxide emissions from electricity used in manufacturing. Note
that most of the LCI data for materials in Table 3-31 are from secondary sources.
                                         3-59

-------
33 BASELINE LCIA RESULTS
               Table 3-31. Top 99% of the CRT acidification impact score
Life-cycle stage
Use
Use
Manufacturing
Manufacturing
Materials processing
Use
Manufacturing
Manufacturing
Materials processing
Manufacturing
Manufacturing
Use
Materials processing
Materials processing
Manufacturing
Materials processing
Manufacturing
Process group
U.S. electric grid
U.S. electric grid
LPG production
LPG production
Invar
U.S. electric grid
Japanese electric grid
U.S. electric grid
Steel production, cold-
rolled, semi-finished
CRT glass/frit
manufacturing
Japanese electric grid
U.S. electric grid
Aluminum production
Polycarbonate production
U.S. electric grid
Polycarbonate production
LPG production
Material
Sulfur dioxide
Nitrogen oxides
Sulfur oxides
Nitrogen oxides
Sulfur dioxide '
Hydrochloric acid
Sulfur dioxide
Sulfur dioxide
Sulfur dioxide
Nitrogen oxides
Nitrogen oxides
Hydrofluoric acid
Sulfur dioxide
Nitrogen dioxide
Nitrogen oxides
Sulfur dioxide
Nitrous oxide
LCI data
type
Model/secondary
Model/secondary
Secondary
Secondary
Secondary
Model/secondary
Secondary
Model/secondary
Secondary
Primary
Model/secondary
Model/secondary
Secondary
Secondary
Model/secondary
Secondary
Secondary
Contribution
to impact
score*
47%
16%
15%
7-6%
4.8%
1.8%
1.6%
0.76%
0.73%
0.59%
0.54%
0.41%
0.40%
0.26%
0.25%
0.23%
0.22%
*Column may not add to 99% due to rounding.

3.3.8.2 Major contributors to the LCD acidification impact results

       Table 3-32 lists the materials responsible for the top 99% of the LCD acidification impact
results and the LCI data type.  Sulfur dioxide emissions from the U.S. electric grid during the use
stage are the greatest contributor at 31%, followed by NOX from natural gas production in the
materials processing stage. The latter process produces natural gas used by one LCD
monitor/module manufacturer as an ancillary material, indicating the LCD monitor/module
manufacturing process group is ultimately responsible for this contribution to the impact score.
However, as noted previously, only one LCD monitor/module manufacturer reported the
ancillary use of LNG. Other LCD monitor/module manufacturers reported using LNG as a fuel,
but not as an ancillary material.  NOX, ammonia, hydrofluoric acid, and hydrochloric acid
emissions from LCD monitor/module manufacturing contribute another 22% of the LCD
acidification impact score. LCD monitor/module manufacturing data were collected directly by
the CDP from manufacturers in Asia.
       NOX emissions from the U.S. electric grid during the use stage, and SOX and NOX
emissions from the Japanese electric grid during manufacturing are also among the top
contributors to the LCD acidification results. LCI data for these process groups were developed  •
by the CDP from secondary sources.
                                        3-60

-------
                                                            3.3 BASELINE LCIA RESULTS
Table 3-32. Top 99% to the LCD acidification impact score
Life-cycle stage
Use
Materials processing
Manufacturing
Use
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Use
Manufacturing
Materials processing
Materials processing
Manufacturing
Materials processing
Materials processing
Manufacturing
Use
Process group
U.S. electric grid
Natural gas prod.
LCD monitor/module mfg.
U.S. electric grid
Japanese electric grid
LCD monitor/module mfg.
Japanese electric grid
LCD monitor/module mfg.
LCD monitor/module mfg.
LPG production
U.S. electric grid
LCD backlight
Natural gas production
Natural gas production
LPG production
Steel production, cold-rolled,
semi-finished
PMMA sheet production
Natural gas production
U.S. electric gird
Material
Sulfur dioxide
Nitrogen oxides
Nitrogen oxides
Nitrogen oxides
Sulfur dioxide
Ammonia
Nitrogen oxides
Hydrofluoric acid
Hydrochloric acid
Sulfur oxides
Hydrochloric acid
Nitrogen oxides
Ammonia
Sulfur oxides
Nitrogen oxides
Sulfur oxides
Sulfur oxides
Nitrogen oxides
Hydrofluoric acid
LCI data
type
Model/secondary
Secondary
Primary
Model/secondary
Model/secondary
Primary
Model/secondary
Primary
Primary
Secondary
Model/secondary
Primary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Model/secondary
Contribution
to impact
score*
31%
15%
13%
10%
9.8%
4.0%
3.2%
• 2.8%
1.8%
1.3%
1.2%
0.70%
0.69%
0.65%
0.65%
0.64%
0.44%.
0.35%
0.27%
""Column may not add to 99% due to rounding.
3.3.8.3 Limitations and uncertainties

       Acidification impact characterization is a function of the mass of an acid-forming
chemical emitted to air and the acidification potential (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 SO2. This is a full equivalency approach to
impact characterization where all substances are addressed in a unified, technical model, which
lends more certainty to the characterization results than partial equivalency factors discussed with
regard to photochemical smog (Section 3.3.7).
       For the CRT, and less so for the LCD, impact results are being driven primarily by SO2
and NOX emissions from U.S. power plants during use of the monitor by the consumer.  As
discussed in Section 3.3.5  and noted above, the U.S. and Japanese electric grid inventories were
developed by the CDP from secondary sources.  U.S. electric grid emissions of the criteria
pollutants, including SO2 and NOX, are based on data in the EPA publication, National Air
Quality and Emissions Trends Report, 1997 (EPA, 1998), which were the best data available
when the electric grid inventory data was developed and are expected to be reasonably accurate.
However, the Japanese electric grid inventory was derived from the U.S. inventory based on the
mix of fuels used in Japan. Because Japanese power plants may employ different pollution
                                         3-61

-------
3.3 BASELINE LCIA RESULTS
control devices, use fuels of different quality, or other factors, their emissions could actually be
higher or lower than those reported in the inventory.
      Ld data for many of the other primary contributors to the CRT acidification impact
category are from existing LCI databases. The limitations and uncertainties associated with these
data have been discussed extensively in other subsections of this chapter and pertain here. On
the other hand, LCI data for many of the other primary contributors to the LCD acidification
indicator results were collected directly by the CDP from manufacturers in Asia. These data are
considered to be of better quality since  they were collected to meet the goals, objectives arid
temporal and spatial boundaries of the CDP.

3.3.9  AirParticuIates

      Figure 3-12 presents the CRT and LCD LCIA results for the air particulates impact
category by life-cycle stage, based on the impact assessment methodology presented in Section
3.1.2.8.  Tables M-21 and M-22 in Appendix M list complete air particulates results for the CRT
and LCD, respectively.
                                      Particulates
        -0.040-1
     CRT     _0019                    LCD     -0.012

                  Monitor type

Figure 3-12. Particulates impacts by life-cycle stage
       The life-cycle air particulates indicator is 0.30 kg of air particulates per monitor for the
CRT and 0.115 kg of air particulates per monitor for the LCD. Recall from Section 3.1.2.8 that
air particulates impact results are ideally based on release amounts of paniculate matter with
average aerodynamic diameter less than 10 micrometers (PM10) to the air. This is the size of
particulate matter that is most damaging to the respiratory system. However, as will be shown
later in this section, a significant portion of the particulate emissions data for both monitor types
do not specify a particulate size.  This makes it more difficult to draw conclusions about the
relative life-cycle air particulate impacts of the CRT and LCD.
       The manufacturing and upstream materials processing stages have almost equal
contribution to CRT air particulate impacts, at 45% of the total for the manufacturing stage and
43% of the total for the upstream stages. LCD impacts, on the other hand, are dominated by
particulate emissions during the upstream, materials processing stages, which contribute 80% of
the total score. Both technologies receive a substantial reduction in life-cycle air particulate
impacts at EOL due to an energy credit from incineration with energy recovery.  The energy
                                         3-62

-------
                                                             3.3 BASELINE LCIA RESULTS
credit, which is from incineration with energy recovery, is applied to electric power production
where it offsets some particulate emissions that would otherwise occur from electrical power
production.

3.3.9.1 Major contributors to the CRT air particulates impact results

       Table 3-33 lists the materials that contribute to the top 99% of the CRT air particulates
impact score and their LCI data type.  PM emissions from LPG production are the single largest
contributor to the overall score, at 43% of the total. This LPG is primarily an energy source in
CRT glass manufacturing, indicating the glass/frit process group is the ultimate source of these
air particulate emissions. As noted previously, CRT glass energy inputs are the subject of a
sensitivity analysis, discussed in Section 3.4.
       Other major contributors to the CRT air particulates impact results are PM emissions
from the steel production process group, PM10 emissions from the U.S. electric grid during the
use stage, and PM emissions from aluminum production processes. Note that the inventories for
steel and aluminum production combine data from the raw materials extraction, materials
manufacture, and electricity generation processes. The PM emissions reported for the material
production process group could be from any one of these individual processes, but particulate
matter emissions are most often associated with combustion processes.

              Table 3-33. Top 99% of the CRT air particulates impact score
Life-cycle stage
Manufacturing
Materials processing
Use
Materials processing
Process group
LPG production
Steel production, cold-rolled,
semi-finished
U.S. electric grid
Aluminum production
Material
>,
PM
PM
PM-10
PM
LCI data
type
Secondary
Secondary
Model/secondary
Secondary
Contribution to
impact score*
43%
35%
19%
3.0%
*Column adds to greater than 99% due to an offset of emissions from incineration with energy recovery at EOL

       As shown in Table 3-33, the impact scores associated with the LPG, steel, and aluminum
production process groups—82% of CRT air particulates impacts—are based on emissions of
PM instead of emissions of PM10. This could be a matter of different terminology used in the
secondary data sets for these process groups (that is, PM is used to represent PMi0), or it could
represent a broader class of particulate emissions, of which PM10 emissions would be a subset. If
the latter case is true, it is likely that CRT air particulate impacts are overstated.

3.3.9.2 Major contributors to the LCD air particulates impact results

       Table 3-34 lists the materials that contribute to the top 99% of the LCD air particulates
impact score and their LCI data type. PM emissions from steel production are the largest
contributor to the overall score, followed by PM emissions from natural gas production. Natural
gas from this process supplies the LNG used as an ancillary material by one LCD
monitor/module manufacturer.
       Other major contributors to the LCD air particulates impact results are PMi0 emissions
from the U.S. electric grid during the use stage and from the Japanese electric grid during
                                          3-63

-------
33 BASELINE LCIA RESULTS	.	

manufacturing, and PM emissions from LPG production. LPG from the latter process supplies
energy to the LCD glass manufacturing process.

             Table 3-34. Top 99% of the LCD air particulates impact score
Life-cycle stage
Materials processing
Materials processing
Use
Manufacturing
Manufacturing
Process group
Steel production, cold-rolled,
semi-finished
Natural gas production
U.S. electric grid
Japanese electric grid
LPG production
Material
PM
PM
PM-10
PM-10
PM
LCI data
type
Secondary
Secondary
Model/secondary
Model/secondary
Secondary
Contribution to
impact score*
45%
25%
19%
5.9%
5.4%
*Column adds to greater than 99% due to an offset of emissions from incineration with energy recovery at EOL.

       As shown in Table 3-34, the impact scores associated with the steel, natural gas, and LPG
production process groups—roughly 75% of LCD air particulates impacts—are based on
emissions of PM instead of emissions of PM10. As with the CRT, air particulate impacts should
be based on PM10  emissions, indicating LCD air particulate impacts may be overstated.

3,3.9.3 Limitations and uncertainties

       The CDP LCIA methodology for air particulates is based on emissions of PMi0to air,
which is the size of particulate matter that is most damaging to the respiratory system.  However,
as noted in Tables 3-33 and 3-34, the majority of the CRT and LCD impacts were calculated
from emissions of 'TM" rather than PM10. This could be a matter of different terminology used
in the secondary data sets for these process groups, or it could represent a broader class of
particulate emissions, of which PM10 emissions would be a subset.  If the latter case is true, it is
likely that both the CRT and LCD air particulate impacts are overstated.
      . The LCI data for all of the major contributors to both the CRT and LCD were either
developed by the CDP from secondary sources (e.g., the U.S. and Japanese electric grids) or are
from secondary LCI data sets (e.g., the fuel and upstream materials production processes).  The
limitations and uncertainties associated with these data have been discussed in other subsections
of this chapter and pertain here. Note that U.S. electric grid emissions of the criteria pollutant
PMj0 are based on data in the EPA publication, National Air Quality and Emissions Trends
Report, 1997 (EPA, December, 1998, EPA/454/R-98-016), and are expected.to be reasonably
accurate.
       Finally, the amount of LPG used to produce CRT glass, which is ultimately driving the
CRT air particulates results, is also uncertain due to the large variability in CRT glass energy
inputs received from glass manufacturers.  See Section 3.4 for a sensitivity analysis of CRT glass
energy inputs.
                                        3-64

-------
                                                             33 BASELINE LCIA RESULTS
 3.3.10 Water Eutrophication

       Figure 3-13 presents the CRT and LCD LCIA results for the water eutrophication impact
 category by life-cycle stage, based on the impact assessment methodology presented in Section
 3.1.2.9. Tables M-23 and M-24 in Appendix M are complete results for the CRT and LCD,
 respectively.
                                     Eutrophication
-s    5.00E-02 i

3..-S  4.00E-02-
CT j;
? =  3.00E-02 -

jf g  2.00&02 -

w" ?  1.00E-02 -

Q. 5  O.OOE+00
&   -1.00&02 J
                            4.76&02
                                                            4.90&02
                                 CRT
                                         -1.10E-04
LCD
                                                                         -7.51 E-05
                                            Monitor type

                         Figure 3-13. Eutrophication impacts by life-cycle stage
       The life-cycle water eutrophication indicators are 0.048 kg of phosphate equivalents for
the CRT and 0.050 kg of phosphate equivalents for the LCD. Results for both the CRT and LCD
are completely dominated by emissions from the manufacturing stage, which accounts for 99%
of the indicator for both technologies.  Both technologies have negative scores at the end-of-life
due to incineration with energy recovery.  The energy recovery offsets some of the water
emissions from the electricity generation inventory included in the incineration data set.

3.3.10.1 Maior contributors to the CRT water eutrophication impact results

       Table 3-35 lists the materials that contribute to the top 99% of the CRT water
eutrophication results. Together, chemical oxygen demand (COD) and ammonia ions from the
LPG production process group account for about 91% of the total score. Most of the LPG from
this process is used as an energy source in CRT glass manufacturing (see Section 3.4 for the
sensitivity analysis of CRT glass energy inputs). Emissions of nitrogen, COD and phosphorus
from the CRT tube manufacturing process group contribute about seven percent of the CRT
water eutrophication impacts. COD and other nitrogen emissions from steel production are the
remaining top contributors to the CRT eutrophication score. LPG and steel production data are
from secondary sources, while the CRT tube manufacturing outputs are primary data collected by
the CDP.
                                         3-65

-------
33 BASELINE LCIA RESULTS
           Table 3-35. Top 99% of the CRT water eutrophication impact score
Life-cycle stage
Manufacturing
Manufacturing
Manufacturing
Materials processing
Materials processing
Manufacturing
Manufacturing
Process group
LPG production
LPG production
CRT tube manufacturing
Steel production, cold-rolled,
semi-finished
Steel production, cold-rolled,
semi-finished
CRT tube manufacturing
CRT tube manufacturing
Material
COD
Ammonia ions
Nitrogen
Other nitrogen
COD
COD
Phosphorus (yellow
or white)
LCI data
type
Secondary
Secondary
Primary
Secondary
Secondary
Primary
Primary
Contribution to
impact score*
72%
19%
6.3%
0.37%
0.33%
0.33%
0.32%
*Column may not add to 99% due to rounding.

3.3.10.2 Major contributors to the LCD water eutrophication impact results

       Table 3-36 lists the materials that contribute to the top 99% of the LCD water
eutrophication results.  Like the CRT, the LCD water eutrophication indicator is driven by
emissions from a single process group, in this case, LCD monitor/module manufacturing.
Together, emissions of nitrogen and phosphorus from that process account for 94% of the total
score.  The LPG production process is the next largest contributor to the, LCD water
eutrophication score, where water releases of COD and ammonia ions account for more than 4%
of the total.

           Table 3-36. Top 99% to the LCD water eutrophication impact score
Life-cycle stage
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Materials processing
Process group
LCD monitor/module mfg.
LCD monitor/module mfg.
LPG production
LPG production
LCD panel components
PMMA sheet production
Material
Nitrogen
Phosphorus (yellow or
white)
COD
Ammonia ions
Phosphorus (yellow or
white)
Ammonia
LCI data
type
Primary
Primary
Secondary
Secondary
Primary
Secondary
Contribution
to impact
score*
67%
27%
3.4%
0.88%
0.48%
0.40%
*Column may not add to 99% due to rounding.
3.3.10.3  Limitations and uncertainties

       Eutrophication (nutrient enrichment) impacts are calculated from the mass of a chemical
released directly to surface water and the chemical's eutrophication potential (EP). The EP is a
partial equivalency factor derived from the ratio of nitrogen and phosphorus in the average
                                        3-66

-------
                                                             3.3 BASELINE LCIA RESULTS
composition of algae compared to the reference compound phosphate (see Section 3.1.2.9). As a
partial equivalency approach, only a subset of substances can be converted into equivalency
factors, which is a limitation of this LCIA methodology.  However, the methodology does take
into account nitrogen and phosphorus, which are the two major limiting nutrients of importance
to eutrophication.
       CRT water eutrophication results are dominated by LCI data from secondary sources, and
are therefore subject to the limitations and uncertainties associated with secondary data.
Furthermore, these results are ultimately due to the large amount of LPG reported to be used as a
fuel in LPG, glass production. Because of the large degree of variability in glass energy data
received from three CRT glass manufacturers, CRT glass energy inputs are also uncertain and the
subject of a sensitivity analysis (see Section 3.4).  LCD results, on the other hand, are driven
almost entirely by primary LCI data from the manufacturing life-cycle stage, which were
collected to meet the goals, objectives, and temporal and geographic boundaries of the  CDP and
are therefore considered to be of better quality.

3.3.11 Water Quality

3.3.11.1 Biological oxygen demand (BOD')

       Figure 3-14 presents the CRT and LCD LCIA results for the BOD water quality impacts
category by life-cycle stage, based on the impact assessment methodology presented in Section
3.1.2.10.  Complete results are listed in Tables M-25 (CRT) and M-26 (LCD) in Appendix M.
                                        BOD
                          1.95E-01
                                                          2.79E-02
                                                  7.72E-04  ^^^ O.OOE+00
        -5.00E-02 J
LCD
            -4.65E-04

               Monitor type

Figure 3-14.  BOD impacts by life-cycle stage
                                                                       -3.18E-04
       During the life-cycle of a 17" CRT monitor, 0.195 kg of BOD are released to surface
water.  The life-cycle of a functionally equivalent 15" LCD results in 0.0283 kg of BOD surface
water releases. As shown in Figure 3-14, BOD impacts for both monitor types are driven by
surface water releases in the manufacturing stage, which contribute 100% of CRT impacts and
99% of LCD impacts.  Note that small BOD impacts also occur in the upstream, materials
processing life-cycle stage for both monitor types. These are almost entirely offset by negative
BOD values at end of life due to the offset of electric grid emissions when the monitors are
incinerated with energy recovery.  Note also that there are no BOD emissions from the U.S.
                                          3-67

-------
33 BASELINE LCIA RESULTS
electric grid during the use stage. The incineration inventory is a secondary data set that contains
a different electric grid inventory than the U.S. electric grid inventory developed by the CDP.
      Table 3-37 lists the materials responsible for the top 99% of the CRT BOD impacts.
CRT impacts in this category are driven by BOD releases from the LPG production process, most
of which is used to make LPG employed as fuel in CRT glass manufacturing. BOD releases
from CRT tube manufacturing also contribute a small percentage to the total score.

                  Table 3-37. Top 99% of the CRT BOD impact score
Life-cycle stage
Manufacturing
Manufacturing
Process group
LPG production
CRT tube manufacturing
Material
BOD
BOD
LCI data
type
Secondary
Primary
Contribution to
impact score
96%
3.3%
      Table 3-38 lists the materials that contribute to the top 99% of the LCD BOD impacts. As
shown in the table, LCD impacts are slightly more distributed among processes than CRT
impacts, with BOD releases from four processes or process groups making up the list of top
contributors. Note that BOD releases from LPG production (most of which is used to make LPG
for LCD glass manufacturing) are much less for the LCD than the CRT, even though the LCD
glass manufacturing inventory was derived from the CRT glass manufacturing inventory. This is
because the CRT contains approximately ten times more glass than the LCD.

                  Table 3-38. Top 99% of the LCD BOD impact score
Life-cycle stage
Manufacturing
Manufacturing
Manufacturing
Materials processing
Process group
LCD monitor/module mfg.
LPG production
LCD panel components
Natural gas production
Material
BOD
BOD
BOD
BOD
LCI data
type
Primary
Secondary
Primary
Secondary
Contribution to
impact score
61%
32%
4.7%
0.99%
3.3.11.2 Total suspended solids (TSS1

       Figure 3-15 presents the CRT and LCD LCIA results for the TSS water quality impacts
category by life-cycle stage, based on the impact assessment methodology presented in Section
3.1.2.10.  Tables M-27 and M-28 in Appendix M list complete results for the CRT and LCD,
respectively.
       The life-cycle TSS impact indicator is 0.874 kg of TSS for the CRT and 0.0615 kg of
TSS for the LCD.  TSS impacts for both monitor types are driven by the manufacturing stage,
where 99 and 94% of impacts occur for the CRT and LCD, respectively. TSS impacts also occur
in the upstream, materials processing life-cycle stage for both monitor types. Both technologies
receive a credit on TSS impacts at EOL due to an offset of electric grid emissions when the
monitors are incinerated with energy recovery.
                                       3-68

-------
                                                           3.3 BASELINE LCIA RESULTS


TSS EUps
• Mfg
** 1.00E+00 -|
§ 8.00E-01 -

"5 6.00E-01 -

.0 4.00E-01 •
C 2.00E-01 -
j5 O.OOE+00 •
•§* -2.00E-01 -
8.69E-01 ~ O Use
HBH
^^H^ll . ri EOL
^^^^^^Hj
^^^^^H
^^^H
7.72E-03 ^^B O.OOE+00 4.98E-03 O.OOE+00

tream








CRT -2'115-03 LCD -1'44&03~~1
Monitor type
Figure 3-15. TSS impacts by life-cycle stage


      Table 3-39 presents the major contributors to the CRT TSS indicator and lists the LCI
data type. As with many other impact categories, the LPG production process is the single
largest contributor to the CRT TSS indicator, accounting for 97% of the total score. Most of the
LPG from this process is used as a fuel to produce CRT glass, but CRT energy inputs are
uncertain and evaluated in a sensitivity analysis in Section 3.4. TSS surface water releases from
the CRT glass/frit process group, CRT tube manufacturing, and fuel oil #6 production are also
top contributors to the CRT TSS score.  However, the contribution of these processes or process
groups is small compared to that of the LPG production process.

                   Table 3-39. Top 99% of the CRT TSS impact score
Life-cycle stage
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Process group
LPG production
CRT glass/frit mfg.
CRT tube manufacturing
Fuel oil # 6 production
' Material
Suspended solids
Suspended solids
Suspended solids
Suspended solids
LCI data
type
Secondary
Primary
Primary.
Secondary
Contribution to
impact score
97%
0.83%
0.53%
0.33%
      Table 3-40 presents the top contributors to the LCD TSS impact score. Like the CRT
results discussed above, TSS surface water releases from the LPG production process are
responsible for the majority of LCD TSS impacts. LPG from this production process is used to
produce LCD glass. Note that the actual mass of TSS releases from the LCD-related process is
much smaller than those from the CRT-related process.  This is because the LCD only uses about
10% as much glass as the CRT. TSS releases from the LCD monitor/module process group also
account for a sizeable percentage of LCD TSS impacts.
                                        3-69

-------
33 BASELINE LCIA RESULTS
                   Table 3-40.  Top 99% of the LCD TSS impact score
Life-cycle stage -.
Manufacturing
Manufacturing
Materials processing
Materials processing
Materials processing
Materials processing
Manufacturing
Process group
LPG production
LCD monitor/module mfg.
PMMA sheet production
Natural gas production
Steel production, cold-rolled,
semi-finished
Aluminum production
(all virgin)
LCD panel components
Material
Suspended solids
Suspended solids
Suspended solids
Suspended solids
Suspended solids
Suspended solids
Suspended solids
LCI data
type
Secondary
Primary
Secondary
Secondary
Secondary
Secondary
Primary
Contribution to
impact score*
66%
25%
2.2%
2.0%
1.6%
1.1%
1.0%
""Column may not add to 99% due to rounding.
3.3.11.3 Limitations and uncertainties

       Both BOD and TSS indicators are calculated using a loading approach (i.e., the impact
score is based on the inventory amounts) and are therefore highly sensitive to inventory date
quality. CRT impact results are driven almost entirely by the LPG production inventory (a
secondary data set) and are therefore subject to the limitations and uncertainties associated with
secondary data. In particular, see Section 3.3.2.3 for a detailed discussion of LPG production
data quality. In addition, note that LPG production impacts are almost all due to the large
amount of LPG reported to be used as a fuel in CRT glass manufacturing. As note previously,
CRT glass energy inputs are uncertain and are evaluated in a sensitivity analysis (see
Section 3.4).
       LCD impact results, on the other hand, are driven by LCI data from both primary and
secondary sources and are therefore considered to be of somewhat better quality than the CRT
results. However, a significant percentage of LCD water quality impacts  also come from the
large amount of LPG used as a fuel input to LCD glass manufacturing. These energy inputs are
also uncertain and are evaluated in the sensitivity analysis in Section 3.4.

3.3.12 Radioactivity

       Figure 3-16 presents the CRT and LCD LCIA results for the radioactivity impact category
by life-cycle stage, based on the impact assessment methodology presented in Section 3.1.2.11.
Complete CRT and LCD results are presented in Tables M-29 and M-30 in Appendix M,
respectively.
       The life-cycle radioactivity indicator is 38.5 million Bequerels (Bq) for the CRT and 12.2
million Bq for the LCD.  Radioactivity impacts are driven by radioactive  emissions from the
upstream, materials processing stage for both monitor types, which contributes 99% of CRT life-
cycle impacts and 98% of LCD life-cycle impacts.  This result was unforeseen, since one might
expect the majority of radioactive emissions to occur from the use stage, due to electricity
generation at nuclear power plants. As it turns out, radioactivity impacts are being driven by data
for nuclear fuel reprocessing that are included in the electric grid inventories for the steel, invar
(an alloy of nickel and ferrite), and ferrite production process groups.
                                        3-70

-------
                                                             3.3 BASELINE LCIA RESULTS
                                      Radioactivity
      a 40,000,000-,
      3
      •5 30,000,000 •
      c
      .2 20,000,000 •
      u
      I 10,000,000-
     m
               0
                     38,000,000
12,000,000
                            35,000   436'000   41
          92.2.
                                                                            29
                                CRT
                                                                LCD
                                            Monitor type

                         Figure 3-16. Radioactivity impacts by life-cycle stage
       The LCIs for steel, invar, and fertile were obtained from two databases developed by the
former Ecobilan Group, an LCA consulting firm that was previously headquartered in France
(see Section 2.2.1.1 for a discussion of how these databases were selected). Per Ecobilan, the
fertile inventory contains older data thai may include radioactive emissions from electricity use.
For the steel and nickel inventories, the source of both is site data in Europe, with the radioactive
emissions coming from electricity in Europe, where nuclear fuel is reprocessed. In fact, the
electricity data are from France for both materials, and France is one of the few countries
(including Japan and the United Kingdom) that reprocesses nuclear fuel (Glazebrook, 2001).
Therefore, the radioactivity impacts calculated from Ihese inventories are more representative of
impacts from countries lhal reprocess nuclear fuel lhan impacts from countiies lhal do nol.
       To further illustrate Ihis poinl, Tables 3-41 and 3-42 lisls Ihe materials and process groups
thai contribute to the lop 99% of the CRT and LCD radioactivity indicator results, respectively.
As shown in the lables, radioactivity impacls for bolh monitor types are driven by releases of
plulonium-241 from steel (bolh monitor types), invar (CRT), and fertile (CRT) production.
Plulonium-241 is a byproducl of fuel reprocessing. Xenon -133 releases from the U.S. electric
grid contribute slighlly to Ihe LCD radioactivity impacls and to a lesser degree to Ihe CRT tola!
impacts.  Note thai Ihe aclual amount of radioactivity from Xenon-133 is greater for the CRT
lhan the LCD, bul conttibules a smaller percenl of lolal impacls.
                                          3-71

-------
3.3 BASELINE LCIA. RESULTS
               Table 3-41.  Top 99% of the CRT radioactivity impact score
Life-cycle stage
Materials processing
Materials processing
Materials processing
Use
Materials processing
Materials processing
Process group
Steel production, cold-
rolled, semi-finished
Invar
Ferrite
U.S. electric grid
Steel production, cold-
rolled, semi-finished
Steel production, cold-
rolled, semi-finished
Material
Plutonium-241 (isotope)
Plutonium-241 (isotope)
Plutonium-241 (isotope)
Xenon-133 (isotope)
Plutonium-240 (isotope)
Cesium-135 (isotope)
LCI data
type
Secondary
Secondary
Secondary
Model/secondary
Secondary
Secondary
Contribution
to impact
score*
62%
18%
17%
0.81%
0.27%
0.24%
*Column may not add to 99% due to rounding.
               Table 3-42.  Top 99% of the LCD radioactivity impact score
Life-cycle stage
Materials processing
Use
Manufacturing
Materials processing
Materials processing
Process group
Steel production, cold-
rolled, semi-finished
U.S. electric grid
Japanese electric grid
Steel production, cold-
rolled, semi-finished
Steel production, cold-
rolled, semi-finished
Material
Plutonium-241 (isotope)
Xenon-133 (isotope)
Xenon-133M (isotope)
Plutonium-240 (isotope)
Cesium-135 (isotope)
LCI data
type
Secondary
Model/secondary
Secondary
Secondary
Secondary
Contribution
to impact
score*
96%
0.95%
0.54%
0.42%
0.38%
*Column may not add to 99% due to rounding.

       Most of the radioactivity impacts are based on LCI data from secondary sources and are
therefore subject to the limitations and uncertainties in secondary data, discussed previously. In
addition, because radioactivity impacts are being driven by radioactive emissions from fuel
reprocessing in France, they may not be representative of radioactivity impacts elsewhere.
However, most of the CRT and LCD primary manufacturing data were collected from companies
in Japan, where fuel reprocessing also occurs.  For example, if Japanese CRT and LCD monitor
and/or components manufacturers purchase steel from Japanese steel mills, the radioactivity
emissions from electricity used to manufacture the steel could be similar. Japan ranked second in
worldwide steel production in 2000 behind Mainland China, and third in 1999 behind Mainland
China and the United States (ESI, 2001).
       Note that the Japanese electric grid, which is linked to CRT and LCD production
inventories, was developed from the U.S. electric grid inventory and therefore does not account
for radioactive emissions from fuel reprocessing. This means that radioactive impacts from
Japanese manufacturing processes that consume electricity are understated. For example,
electricity used in the CRT glass/frit process group was the ninth largest contributor to the CRT
energy use score, but the inventory for this process group does not account for fuel reprocessing
emissions.                                                                       .
                                        3-72

-------
                                                             3.3 BASELINE LCIA RESULTS
3.3.13 Potential Human Health Impacts                        '     ,

       As discussed in Section 3.1.2.12, human health impacts includedjn the scope of this LCA
are chronic (repeated dose) effects, including non-carcinogenic and carcinogenic effects to both
workers and the public, and aesthetics. (Although not a health effector se, aesthetics pertains to
human welfare.)
       Chronic health effect (cancer and noncancer) impacts are calculated using the scoring of
inherent properties approach where an impact score is based on the inventory amount weighed by
a hazard value (HV). The HV represents the chronic toxicity of a specific material (see Table
K-8 in Appendix K for a list of toxicity values used to calculate hazard values). In this manner
the inventory amount (the toxic chemical input amount for occupational health effects, and the
output amount for public health effects) is  used as a surrogate for exposure, while the hazard
value represents the inherent toxicity of the chemical for chronic exposure.
       The CDP human health effects LCIA methodology does not consider the fate and
transport of a toxic chemical in the environment, nor does it evaluate the potential for actual
exposures to occur.  LCI data do not have the temporal and spatial specificity needed to estimate
potential dose rates, for example, nor do they contain information on engineering controls used in
an occupational setting to reduce exposure. [It should be noted that more sophisticated models
for evaluating human health effects in an LCA framework are being developed that use a
multimedia fate, multi-pathway human exposure, and lexicological potency approach (Bare,
1999). However, such models are less comprehensive in terms of the number of chemicals for
which there are data.] The limitations and  uncertainties in the health effects scores are discussed
further below, following the presentation of results.
                                                                   "ass.*:- - S'rsis.
3.3.13.1 Chronic occupational health effects

      Figure 3-17 presents the CRT and LCD LCIA results by life-cycle stage for the chronic   '
occupational health effects impact category, based on the impact assessment methodology
presented in Section 3.1.2.12. Complete CRT and LCD results are presented in Tables M-31 and
M-32 in Appendix M.
       *•  9.00E+02 n
        g  6.00&-02 -
       _o
        O  3.00E+02 -
        c
       C  O.OOE+00
        O -3.00E+02-J
                                  Chronic occupational

                            9.00E+02
                                                          6.84E+02
             3.56E+01
                   -2.83E+00  3.59E-01
                                                                        -1.94E+00 '
           CRT                           LCD

                      Monitor type

Figure 3-17.  Chronic occupational impacts by life-cycle stage
                                          3-73

-------
33 BASELINE LCIA RESULTS
       The life-cycle chronic occupational health effects indicator is 934 tox-kg per functional
unit for the CRT, and 696 tox-kg per functional unit for the LCD. As shown in the figure, the
total score is dominated by toxic chemical inputs to the manufacturing stage, which account for
98% and 96% of CRT and LCD impacts in this category, respectively. This result was expected
since inputs to the other life-cycle stages tend to be raw materials (e.g., ores, coal, etc., for the
materials processing and use life-cycle stages) or finished products (e.g., the monitors themselves
for the EOL stage) that are not classified as toxic materials (see Table K-9 in Appendix K for a 1
list of materials excluded from the toxic classification). Both the CRT and LCD receive negative
chronic occupational health effects scores at end of life due to the offset of electric grid
 emissions when the monitors are incinerated with energy recovery.
       Table 3-43 lists the materials responsible for the top 99% of the CRT chronic
occupational health effects score and the LCI data type. LCI data for most of the top contributors
are primary data collected from manufacturers by the CDP. In general, these data are expected
to be of better quality (for the purposes of the CDP) than data from secondary sources, since they
were collected to meet the goals and scope of the CDP.

        Table 3-43.  Top 99% of the CRT chronic occupational health effects score
Life-cycle stage
Manufacturing
Manufacturing
Manufacturing
Use
Manufacturing
Use
Manufacturing
Process group
CRT glass/frit mfg.
PWB manufacturing
CRT tube manufacturing
U.S. electric grid
CRT glass/frit mfg.
U.S. electric grid
CRT tube manufacturing
Material
Liquified petroleum gas
Sulfuric acid
Sulfuric acid
Natural gas
Barium carbonate
Petroleum (in ground)
• Fuel oil #6
LCI data
type
Primary
Primary
Primary
Model/secondary
Primary
Model/secondary
Primary
Contribution
to impact
score*
•75%
13%
4.1%
3.0%
1.8%
0.81%
0.79%
*Column may not add to 99% due to rounding.

       LPG inputs to the glass/frit process group, primarily from CRT glass manufacturing,
contribute 75% of the CRT impacts in this category. The high impact score for LPG is mainly
due to the large amount of LPG inputs to the glass/frit process group (351 kg/functional unit),
which results in a high score when multiplied by the HV. No toxicity data were available for
LPG. Therefore, it was assigned default HVs of one for both cancer and noncancer effects (total
HV=2), representative of mean cancer and noncancer toxicity values.  As noted previously, glass
manufacturing energy data are uncertain and therefore evaluated in a sensitivity analysis (see
Section 3.4).
       Sulfuric acid used in PWB manufacturing is the next greatest contributor to the CRT
chronic occupational health effects results (13%) followed by sulfuric acid used in CRT tube
manufacturing (4.1%). The sulfuric acid HV is based on ah inhalation NOAEL of 0.1 mg/m3,
which is significantly lower (and therefore more toxic) than the geometric mean inhalation   -
NOAEL of 68.7 mg/m3. Consequently, sulfuric acid impacts are driven more by its inherent
toxicity for noncancer effects than the input amounts (0.18 kg per functional unit for PWB
manufacturing and 0.056 kg per functional unit for CRT tube manufacturing). Sulfuric acid has
                                        3-74

-------
                                                            3.3 BASELINE LCIA RESULTS
no cancer slope factor and an IARC weight of evidence (WOE) classification of 3 (not
classifiable for carcinogenicity), and therefore received an HV of zero for cancer effects.
       Natural gas and petroleum used as fuels in the U.S. electric grid, barium carbonate used
in the CRT glass/frit process group and fuel oil #6 used to manufacture the CRT tube round out
the top contributors to the CRT chronic occupational health effects score. Barium carbonate has
no cancer slope factor and an EPA cancer WOE of D (not classifiable), and therefore has an HV
of zero for cancer effects. However, its oral NOAEL is 0.21 mg/kg-day compared to the
geometric mean oral NOAEL of 11.9 mg/kg-day, which results in an HV of 57 for noncancer
effects.  Therefore, like sulfuric acid, the barium carbonate impacts are driven more by its
inherent toxicity than the input amount (0.297 kg per functional unit). No specific toxicity data
were available for natural gas, petroleum, or fuel oil #6; consequently they were assigned a
default HV of one for both cancer and noncancer effects (total HV=2).
       Table 3-44 lists the materials that contribute to the top 99% of the LCD chronic
occupational health effects score.  Like the CRT, LCI data for most of the top contributors are
primary data collected from manufacturers by the CDP, and are therefore expected to be of
generally better quality (for the purposes  of the CDP) than data from secondary sources.

        Table 3-44.  Top 99% of the
Life-cycle stage
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Use
Manufacturing
Manufacturing
Process group
LCD monitor/module mfg.
LCD monitor/module mfg.
PWB manufacturing
LCD glass manufacturing
LCD monitor/module mfg.
LCD panel components
U.S. electric grid
LCD monitor/module mfg.
LCD monitor/module mfg.
Material
Liquified natural gas
Sulfuric acid
Sulfuric acid
Liquified petroleum gas
Phosphme
Sulfuric acid
Natural gas
Dimethylsulfoxide
Ethanolamine
LCI data
type
Primary
Primary
Primary
Primary
Primary
Primary
Model/secondary
Primary
Primary
Contribution to
impact score*
. 57%
23%
8.0%
4.7%
1.8%
1.6%
1.5%
1.1%
0.62%
*Column may not add to 99% due to rounding

       As shown in the table, LCD impacts in this category are dominated by LNG used in LCD
monitor/module manufacturing. The high impact score for LNG is primarily due to the large
amount of ancillary LNG inputs (194 kg/functional unit) in the LCD monitor/module
manufacturing inventory, which results in a high score when multiplied by the HV. No toxicity
data were available for LNG.  Therefore, it was assigned a default, mean HV of one for both
cancer and noncancer effects (total HV=2). As noted previously, only one of seven LCD
module/monitor manufacturers reported using LNG as an ancillary material. Therefore, the total
score for LCD chronic occupational health effects may not be representative of the industry as a
whole. If we remove this application of LNG from the LCD inventory, the LCD occupational
health effects result is reduced by 58 percent, from 683 tox-kg per monitor to 288 tox-kg per
monitor.  Note, however that other LCD monitor/module manufacturers did report using LNG as
a fuel.
                                         3-75

-------
3.3 BASELINE LC1A RESULTS
       Sulfiiiic acid used in three process groups (LCD module/monitor manufacturing, PWB
manufacturing, and LCD panel components) accounts for another 33% of the LCD chronic
occupational health effects score. As discussed above for the CRT, sulfuric acid has a relatively
low toxicity value, and therefore a high HV, which results in a high impact score for a small
input amount.  The LCD module/monitor manufacturing process group has the highest impact
score for sulfuric acid because it has the greatest input amount (0.229 kg per functional unit).
       The remaining top contributors to the LCD chronic occupational health effects score are
LPG used in LCD glass manufacturing; and phosphine, dimethylsulfoxide, and ethanolamine
used in LCD monitor/module manufacturing; and natural gas used as a fuel by the U.S. electric
grid. As discussed earlier, LCD glass energy inputs are uncertain and evaluated in a sensitivity
analysis in Section 3.4. The LPG score is based on default HVs (representative of the geometric
mean toxicity values) for both cancer and noncancer effects, since no toxicity data were available
for LPG.
       The phosphine score is driven by its low oral NOAEL value (0.026 mg/kg-day), which is
significantly lower than the geometric mean value of 68.7 mg-kg-day. Thus, due to its inherently
high toxicity, a relatively small input of phosphine (in this case, 0.027 kg per functional unit)
results in a relatively high chronic occupational health effects score.  No slope factors or cancer
WOE classifications were found for phosphine, indicating a default HV of one was used, which
is far outweighed by the non-cancer hazard value.
       Dimethylsulfoxide is less toxic than phosphine (oral LOAEL =1.0 mg/kg-day), but has a
greater input amount (0.066 kg per functional unit). However, phosphine's greater toxicity
outweighs the greater input amount for dimethylsulfoxide, resulting in a higher impact score for
phosphine.
       No specific toxicity data were available for natural gas; consequently it was assigned a
default HV of one for both cancer and noncancer effects (total HV=2).

3.3.13.2  Chronic public health effects

       Figure 3-18 presents the CRT and LCD LCIA scores by life-cycle stage for the chronic
public health effects category, based on the impact assessment methodology presented in Section
3.1.2.12. Complete results are presented in Tables M-33 and M-34 in Appendix M, respectively.
c
         1.20&03 -
      I
         6.00&02-
         O.OOE+00
                                  Chronic public health

                                    1.65E+03
                   2.28E+02
                         9.77E+01
                                                                   6.17E+02
                                                  2.35E+02
                                  1.74E-01    5.01 E+01
                               CRT
                                                         LCD
                                                    -6.30E-02
                                          LUD
                      Monitor type

Figure 3-18. Chronic public health impacts by life-cycle stage
                                        3-76

-------
                                                             3.3 BASELINE LCIA RESULTS
        The life-cycle chronic public health Effects score is 1,980 tox-kg per functional unit for
 the CRT and 902 tox-kg per functional unit for the LCD. As shown in the figure, the CRT score
 is  dominated by toxic chemical outputs from electricity generation in the use stage, which
 account for almost 84% of CRT impacts in this category. To a lesser degree, LCD chronic public
 health effect impacts are also driven by emissions from electricity generation in the use stage,
 which account for more than 68% of the total. Note that the ratio of CRT to LCD use stage  '
 public health impacts is the same as the ratio of CRT to LCD use stage electricity consumption
 (634 kWh/life for the CRT to 237 kWh/life for the LCD).
        The materials processing stage contributes almost 12% of CRT chronic public health
 effect impacts and almost six percent of LCD impacts.  The manufacturing life-cycle stage is
 responsible for five and 26% of CRT and LCD impacts in this category, respectively. Both
 monitors receive very small public chronic health effects scores at end of life. This is because
 most public health effect impacts from CRT and LCD recycling and disposal processes are offset
 by a credit on electric grid emissions when the monitors are incinerated with energy recovery.
        Table 3-45 presents the materials that contribute to the top 99% of the CRT chronic
 public health effects score. As  shown in the table, SO2 emissions from a number of different
 process groups almost completely dominate CRT impacts in this category, accounting for more
 than 98% of the total. All of the SO2 LCI data shown in the table are either from secondary data
 sets not developed specifically for the CDP, or from the electric grid inventories developed from
 secondary sources for this project. Sulfur dioxide has a relatively high HV based on an
 inhalation NOAEL of 0.104 mg/m3, compared to the geometric mean inhalation NOAEL of 68.7
 mg/m3. In addition, from a mass loading perspective, SO2 emissions were the second largest
 contributor to CRT life-cycle air pollutant emissions, exceeded only by emissions of carbon
 dioxide (CO2). Carbon dioxide is not classified as toxic, and therefore does not contribute to the
 human health effects scores.
Table 3-45. Top 99% of the CRT chronic public health effects score
Life-cycle stage
Use
Materials processing
Manufacturing
Manufacturing
Materials processing
Materials processing
Materials processing
Manufacturing
Materials processing
Process group
U.S. electric grid
Invar
Japanese electric grid
U.S. electric grid
Steel production, cold-
rolled, semi-finished
Aluminum production
Polycarbonate production
LPG production
Lead production
Material
Sulfur dioxide
Sulfur dioxide
Sulfur dioxide
Sulfur dioxide
Sulfur dioxide
Sulfur dioxide
Sulfur dioxide
Carbon monoxide
Sulfur dioxide
LCI data
type
Model/secondary
Secondary
Model/secondary
Model/secondary
Secondary
Secondary
Secondary
Secondary

Contribution to
impact score*
83%
8.3%
2.9%
1.3%
1.3%
0.70%
0.40%
0.29%
0.23%
                '. to 99% due to rounding.

       Most of the sulfur dioxide emissions that contribute to the CRT chronic public health
effects score are from the combustion of fossil fuels used to generate electricity. For example,
the electricity required to power the monitor during the use stage accounts for the vast majority
of SO2 emissions and 83% of the CRT chronic public health effects score. Sulfur dioxide
                                         3-77

-------
3.3 BASELINE LCIA RESULTS
emissions from electricity consumed in the United States and Japan during the manufacturing
life-cycle stage account for another 4.2% of the total score.  Much of the SO2 emissions reported
in secondary data sets for the materials processing life-cycle stage may also be from electricity
generation since many of these data also contain an electric grid inventory.
       Table 3-46 lists the materials that contribute to the top 99% of the LCD chronic public
health effects score and the LCI data type. LCD impacts in this category are also dominated by
SO2 emissions, which are responsible for roughly 93% of impacts. Like the CRT, most of these
emissions occur from electricity generation, either during the use stage (68%) or manufacturing
(21%). As noted previously, SO2 has a relatively high HV due to its low toxicity value
(inhalation NOAEL = 0.104 mg/m3). Similar to the CRT, from a mass loading perspective, SO2
emissions were the second largest contributor to LCD life-cycle air pollutant emissions, exceeded
only by emissions of CO2.

           Table 3-46. Top 99% of the LCD chronic public health effects score
Life-cycle stage
Use
Manufacturing
Manufacturing
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Manufacturing
Manufacturing
Process group
U.S. electric grid
Japanese electric grid
LCD monitor/module mfg.
Steel production, cold-
rolled, semi-finished
PMMA sheet production
Natural gas production
Natural gas production
Aluminum production
Natural gas production
Polycarbonate production
LCD monitor/module mfg.
U.S. electric grid
Material
Sulfur dioxide
Sulfur dioxide
Phosphine
Sulfur dioxide
Sulfur dioxide
Methane
Benzene
Sulfur dioxide
Carbon monoxide
Sulfur dioxide
Phosphorus
(yellow or white)
Sulfur dioxide
LCI data
type
Model/secondary
Model/secondary
Primary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Primary
Model/secondary
Contribution to
impact score*
68%
21%
3.2%
1.4%
0.96%
0.78%
0.59%
0.57%
0.53%
0.49%
0.38%
0.35%
*Column may not add to 99% due to rounding.

       Other top contributors to the LCD chronic public health effects score include phosphine
and phosphorus from LCD monitor/module manufacturing, and methane, benzene, and carbon
monoxide from natural gas production. As noted above in the section on chronic occupational
health effects, the phosphine score is driven by its low oral NOAEL value (0.026 mg/kg-day),
which is significantly lower than the geometric mean value of 68.7 mg-kg-day, resulting in a high
HV. Thus, a relatively small output of phosphine (in this case, air emissions of 0.063 kg per
functional unit) results in a relatively high chronic public health effects score.
       The benzene and phosphorus chronic health effects scores are also driven more by their
toxicity than the output amounts.  Benzene is a known human carcinogen (EPA WOE Class A)
that also causes noncancer health effects. The HV for benzene is based on its oral slope factor
[0.055 (mg/kg-day)"1] and its inhalation NOAEL for noncancer effects (1.15 mg/m3), which
                                        3-78

-------
                                                            3.3 BASELINE LCIA RESULTS
 together result in a high HV. (Benzene ftsblias an mhaWtiffh slope factor and an oral NOAEL
 value, but these yield lower hazard values when compared to the geometric mean values.)
 Phosphorus has an EPA WOE classification of D (not classifiable as to human carcinogenicity),
 but has a low oral NOAEL value (0.015 mg/kg-day), which also gives it-a high HV.
       No toxicity data were available for methane. Therefore, it received a default HV of one
 for both cancer and noncancer effects (HV=2 total). The HV for carbon monoxide is based on an
 inhalation LOAEL of 55 mg/m3.

 3.3.13.3 Chronic public health effect scores modified to exclude sulfur dioxide

       Because the chronic public health effects scores for both the CRT and the LCD are
 dominated by SO2 emissions, a secondary analysis was run to identify the top contributors to
 public health impacts when SO2 emissions are excluded from the inventories. Results of this
 analysis may be more useful to manufacturers seeking to identify problematic toxic chemicals
 within their own manufacturing processes.
       Figure 3-19 presents the CRT and LCD chronic public health effects scores by life-cycle
 stage when SO2 emissions are excluded from the inventories. Under this scenario, the CRT
 score is reduced almost 99% from 1980 tox-kg to 26 tox-kg per functional unit, and the LCD
 score is reduced about 93% from 902 tox-kg to 61 tox-kg per functional unit.  Note that these
 scores should not be used to evaluate which monitor type has higher overall impacts in this
 category, but they are useful for identifying life-cycle improvement opportunities that were
 previously  obscured by SO2 impacts.
                             Chronic public health (without SO2)
      *•  60.0
      c

      «  40.0 -
      §
      o  20.0 -

      *§>   0.0
       X
       3 -20.0 J
5.6
      15
            6.1
                  -0.37
         CRT
                                           LCD
                                          Monitor type
                      Figure 3-19.  Chronic public health impacts (without SO2)
                                        by life-cycle stage
       With SO2 emissions removed from the inventories, chronic public health effect impacts
are highest in the manufacturing life-cycle stage for both the CRT (56% of impacts) and the LCD
(65% of impacts). The use stage is the next largest contributor for the CRT (22%), and the
materials processing stage in the next largest contributor for the LCD (32%). As will be shown
below, use stage impacts are significant for the CRT, even when SO2 emissions are excluded,
because of the CRT's relatively high electricity consumption during use by the consumer and the
associated emissions of pollutants from U.S. power plants.
                                         3-79

-------
33 BASELINE LCIA RESULTS	;	

       Table 3-47 presents the materials that contribute greater than one percent of CRT impacts
when SO2 emissions are excluded from the CRT inventory. Under this scenario, CRT chronic
public health impacts are still being driven by emissions of criteria air pollutants,2 including
carbon monoxide, nitrogen oxides, and sulfur oxides (assuming that SO2-emissions comprise a
large part of the sulfur oxide emissions shown in the table).  As shown in the table, emissions of
these three pollutants or pollutant categories are responsible for some 48% of CRT chronic
public health impacts when pure SO2 emissions are excluded from the CRT inventory. Note that
the majority of these emissions occur from the LPG production process, and most of this LPG is
used as a fuel in CRT glass manufacturing. CRT glass manufacturing energy inputs are uncertain
and evaluated in a sensitivity analysis (See Section 3.4). Other significant contributors include
arsenic from lead production, methane from LPG production and the U.S. electric grid inventory,
vanadium and benzene from LPG production, and  titanium tetrachloride from aluminum
production.

         Table 3-47. Materials contributing greater than 1% of the CRT chronic
                        public health effects score (without SO^
Life-cycle stage
Manufacturing
Use
Materials processing
Manufacturing
Manufacturing
Use
Manufacturing
Use
Manufacturing
Manufacturing
Materials processing
Manufacturing
Use
Use
Materials processing
Process group
LPG production
U.S. electric grid
Lead
LPG production
LPG production
U.S. electric grid
LPG production
U.S. electric grid
LPG production
LPG production
Aluminum production
(virgin)
CRT glass/frit mfg.
U.S. electric grid
U.S. electric grid
Steel Prod., cold-rolled,
semi-finished
Material
Carbon monoxide
Nitrogen oxides
Arsenic
Methane
Sulfur oxides
Methane
Nitrogen oxides
Carbon monoxide
Vanadium
Benzene
Titanium tetrachloride
Fluorides (F-)
Arsenic
Hydrochloric acid
Carbon monoxide
LCI data
type
Secondary
Modeled/secondary
Secondary
Secondary
Secondary
Modeled/secondary
Secondary
Modeled/secondary
Secondary
Secondary
Secondary
Primary
Modeled/secondary
Modeled/secondary
Secondary
Contribution
to impact
score
22.35%
,9.12%
8.55%
6.50%
6.21%
4.99%
4.44%
4.23%
4.05%
3.32%
2.79%
- 2.27%
2.16%
1.91%
1.16%
       Table 3-48 presents the materials that contribute greater than one percent of LCD impacts
when SO2 emissions are excluded from the LCD inventory. Under this scenario, phosphine
emissions from LCD monitor/module manufacturing are the dominant factor in the LCD chronic
public health effects score, contributing 47% of the total. Other significant contributors include
methane, benzene, carbon monoxide, and nitrogen oxides from natural gas production, and
        The criteria air pollutants are those for which U.S. National Ambient Air Quality Standards have been
adopted. They are carbon monoxide, lead, nitrogen oxides, ozone, particulate matter, and sulfur dioxide.
                                        3-80

-------
                                                             3.3 BASELINE LCIA RESULTS
 phosphorus, fluorides, tetramethyl ammonium hydroxide, anji nitrogen oxides from LCD
 monitor/module manufacturing. Recall that fhe LCD monitor/module manufacturing process
 consumes the majority of the natural gas made in the natural gas production process, where LNG
 is used as an ancillary material. However, only one of the seven LCD monitor/module
 manufacturers that provided inventory data to the CDP reported the ancillary use of LNG.  Other
 LCD monitor/module manufactures did report the use of LNG as a fuel.

         Table 3-48. Materials contributing greater than 1% of the LCD chronic
Life-cycle stage
Manufacturing
Materials processing
Materials processing
Materials processing
Manufacturing
Manufacturing
Materials processing
Manufacturing
Manufacturing
Use
Process group
LCD monitor/module mfg.
Natural gas production
Natural gas production
Natural gas production
LCD monitor/module mfg.
LCD monitor/module mfg.
Natural gas production
LCD monitor/module mfg.
LCD monitor/module mfg.
U.S. electric arid
Material
Phosphine
Methane
Benzene
Carbon monoxide
Phosphorus (yellow or
white)
Fluorides (F-)
Nitrogen oxides
Tetramethyl
ammonium hydroxide
- Nitrogen oxides
Nitrogen oxides
LCI data
type
Primary
Secondary
Secondary
Secondary
Primary

Secondary
Primary
Primary

Contribution
to impact
score
46.7%
11.4%
8.61%
7.81%
5.56%
4.15%
2.12%
2.09%
1.78%
1.43%
3.3.13.4 Limitations and uncertainties: chronic human health effects

       Most of the limitations and uncertainties in the chronic human health effects results
presented here 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 are discussed below.

       Structural or modeling limitations and uncertainty. 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, transformation, or degradation into account.  In addition, it uses a simple
surrogate value (i.e., inventory amount) to evaluate the potential for exposure, when actual
                                         3-81

-------
3.3 BASELINE LCIA RESULTS
exposure potential involves many more factors, some of which are chemical-specific. Other
sources of uncertainty include possible omissions by the CDP 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 may also be considered limits in the toxicity data
which are discussed further below.
       It should also be noted, however, that because LCA involves analyzing many processes
over the entire life cycle of a product, a comprehensive, quantitative risk assessment pf each
chemical input or output can not be done. Rather, LCA develops relative impacts that often lack
temporal or spatial specificity, but can be used to identify materials for more detailed evaluation.
More detailed assessments of the toxicity and potential exposures to selected materials are
performed in Chapter 4.

       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 homogeneous 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.

       Regarding uncertainties resulting from missing toxicity data, there is uncertainty
associated with using a default HV (i.e.Lassuming average toxicity for that measure when a
chemical could be either more or less toxic than average). 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 273 chemicals classified as
potentially toxic in the CDP LCA, 156 (57%) had no toxicity data for carcinogenic effects and
128 (47%) had no data for noncarcinogenic effects. Ninety-seven chemicals (36%) had no
human health toxicity data whatsoever.

       LCI data limitations and uncertainty. Limitations and uncertainties in the LCI data have
been discussed previously and are generally related to:  (1) uncertainties in data from secondary
sources that may not be representative of the geographic and temporal boundaries of this LCA,
and (2) uncertainties in a few of the primary data points collected specifically for this project.
With regard to the latter, glass manufacturing energy inputs are particularly uncertain despite
                                         __                                   _____    _

-------
                                                             3.3 BASELINE LCIA RESULTS
 numerous attempts to resolve the uncertainty, but are responsible for a significant portion of CRT
 human health impacts. Glass manufacturing energy inpiits aire evaluated in a sensitivity analysis
 in Section 3.4.  The amount of LNG used as an ancillary material in LCD monitor/module
 manufacturing is also uncertain (also despite attempts to resolve questions regarding the data),
 but this material contributes a significant portion of LCD occupational health impacts. As noted
 previously, removing this application of LNG from the LCD monitor/module manufacturing
 inventory would reduce the LCD chronic occupational health effects score by 68%.

 3.3.13.5 Aesthetic impacts (odor)

       Figure 3-20 presents the CRT and LCD LCIA results for the aesthetic impacts (odor)
 category, based on the impact assessment methodology presented in Section 3.1.2.12. Complete
 results for the CRT and LCD are presented in Tables M-35 and M-36 in Appendix M,
 respectively. The life-cycle aesthetic (odor) impact result is 7.58 million m3 malodorous air per
 functional unit for the CRT and 5.04 million m3 malodorous air per functional unit for the LCD.
 As shown in the figure, this impact category indicator is dominated by air emissions in the
 manufacturing stage for both the CRT (96% of total) and the LCD (98% of total). Both monitor
 types receive relatively minor contributions in the use and materials processing life-cycle stages,
 and negative values at end of life.  Negative values are due to the offset of electric power plant '
 emissions from incineration with energy recovery.
                                         Odor
      C  7,000,000

      1  5,000,000
      o
      '•g  3,000,000
      C
      jg  1,000,000

      £ -1,000,000 -
                           7,280,000
201,000

   -34,100
61,600
                               CRT
                             LCD
                                            Monitor type

                             Figure 3-20. Odor impacts by life-cycle stage
Major Contributors to the CRT Aesthetics (Odor) Result

       Table 3-49 lists the materials that contribute to the top 99% of the CRT aesthetic impacts
result and the LCI data type.  Air emissions of hydrogen sulfide from LPG production in the
manufacturing life-cycle stage dominate the CRT odor impacts, contributing 94% of the total
score. Hydrogen sulfide impacts are calculated based on an odor threshold value (OTV) of
0.00043 mg/m3.  [See Table K-7 in Appendix K for a list of OTVs used to calculate aesthetic
(odor) impacts.]  As noted previously, most of the LPG produced by this process is used as a fuel
in CRT glass manufacturing,  but glass manufacturing energy inputs are uncertain. The next
largest contributor to the CRT is acetaldehyde emitted from the U.S. electric grid during the use
                                          3-83

-------
33 BASELINE LCIA RESULTS
stage. Acetaldehyde has a lower OTV (0.00027 mg/m3) than LPG, and is also emitted in smaller
quantities. Emissions of hydrogen sulfide from fuel oil #6 production, steel production, and ABS
production are the remaining top contributors to the CRT aesthetic impacts score. LCI data for
all of the top contributors are either from secondary data sets or developed by CDP researchers
from secondary sources.

             Table 3-49. Top 99% of the CRT aesthetic (odor) impacts score
Life-cycle stage
Manufacturing
Use
Manufacturing
Materials processing
Materials processing
Process group
LPG production
U.S. electric grid
Fuel oil #6 production
Steel production, cold-
rolled, semi-finished
ABS production
Material
Hydrogen sulfide
Acetaldehyde
Hydrogen sulfide
Hydrogen sulfide
Hydrogen sulfide
LCI data
type
Secondary
Model/secondary
Secondary
Secondary
Secondary
Contribution to
impact score*
94%
2.5%
0.98%
0.42%
0.31%
*Column may not add to 99% due to rounding.

Major Contributors to the LCD Aesthetics (Odor) Result

       Table 3-50 presents the materials that contribute to the top 99% of the LCD aesthetics
impact score and the LCI data type.  LCD impacts are dominated by air emissions of phosphine
from, LCD monitor/module manufacturing, which contribute 89% of the total score. OTVs
reported for phosphine range from 0.014 to 2.8 mg/m3. The lower, more sensitive value (0.014
mg/m3) was used to calculate impacts.

             Table 3-50. Top 99% of the LCD aesthetic (odor) impacts score
Life-cycle stage
Manufacturing
Manufacturing
Use
Manufacturing
Manufacturing
Process group
LCD monitor/module mfg.
LPG production
U.S. electric grid
LCD monitor/module mfg.
LCD monitor/module mfg.
Material
Phosphine
Hydrogen sulfide
Acetaldehyde
Ammonia
Acetic acid
LCI date
type
Primary
Secondary
Model/secondary
Primary
Primary
Contribution to
impact score*
89%
6.8%
1.4%
1.2%
0.44%
"•Column may not add to 99% due to rounding.

       Other significant contributors include hydrogen sulfide from LPG production,
acetaldehyde from the U.S. electric grid, and ammonia and acetic acid from LCD
module/monitor manufacturing. Most of the LPG made in the LPG production process is used as
a fuel in LCD glass manufacturing, indicating this process is ultimately responsible for LPG
production impacts. However, LCD glass energy inputs are uncertain and evaluated in a
sensitivity analysis (Section 3.4). LCD monitor/module manufacturing data were collected
directly from manufacturers by the CDP, while the LPG production inventory was obtained from
Ecobilan. The U.S. electric grid inventory was developed by CDP researchers from secondary
sources,
                                       3-84

-------
                                                              3.3 BASELINE LCIA RESULTS
Limitations and Uncertainties         ;-   ;           ;..  t

       Aesthetic (odor) impact scores are based on the identity and amount of odor-causing
chemicals (Heijungs et al, 1992; EPA, 1992), released to the air divided-by their chemical-
specific OTVs. An OTV is the lowest concentration of a substance in air that can be smelled
based on a standardized test. Limitations and uncertainties in the aesthetics impact score stem
from structural or model uncertainty (whether or not odor thresholds will actually be exceeded),
OTV data uncertainty (how well published OTVs represent the odor threshold of different
populations), and LCI data uncertainty.
       The aesthetics impact score calculates the mass of malodorous air that could result if a
chemical release occurs in a finite volume of air. It does not predict whether actual odor impacts
will occur. This is because LCI data do not describe the time rate of release or whether dilution
and mixing with ambient air will dilute the concentration of a pollutant to below its odor
threshold. In addition, odor thresholds are highly variable because  of the differing ability of
individuals to detect odors. Therefore, the impact scores may not account for odors perceived by
the most sensitive populations or may overstate impacts perceived by less sensitive populations.
Finally, the aesthetic impact scores are subject to the limitations and uncertainties in the LCI
data, since they are calculated from air emissions data in the inventories. The limitations and
uncertainties in LCI data were discussed in Section 2.2.2.2, and have been discussed extensively
with LCIA results for other impact categories, above.

3.3.14 Ecotoxicity

       Ecbtoxicity refers to effects of chemical outputs on non-human living organisms. As
discussed in Section 3.1.2.13, ecotoxicity impact categories included in the scope of this LCA
include impacts to  aquatic and terrestrial organisms.
       Ecotoxicity impacts are calculated using the scoring of inherent properties approach
where an impact score is based on the inventory amount weighed by a hazard value (HV).  The
HV represents the toxicity of a specific material to aquatic or terrestrial organisms (see
Table K-8 in Appendix K for a list of toxicity values used to calculate hazard values).  Aquatic
HVs are based on acute and chronic toxicity values for fish, while terrestrial HVs are based on
chronic noncancer toxicity values for mammals,  usually rodents. Similar to the chronic human
health impacts discussed in Section 3.3,13,  the inventory amount (the toxic chemical outputs to
water for aquatic toxicity effects, and the outputs to air and water for terrestrial toxicity effects) is
used as a surrogate for exposure, while the hazard value represents the inherent toxicity of the
substance.
       Also like the human health effects methodology, the CDP ecotoxicity LCIA methodology
does not consider the fate and transport of a toxic chemical in the environment, nor does it
evaluate the potential for actual exposures to occur.  In addition, the methodology is limited in
that it does not consider toxicity data from all types of aquatic or terrestrial species, but rather
focuses on a few selected species for which more toxicity data are available. The limitations and
uncertainties in the ecotoxicity scores are discussed further below, following the presentation of
results.
                                          3-85

-------
33 BASELINE LCIA RESULTS
3.3.14.1 Aquatic toxicity

       Figure 3-21 presents the CRT and LCD LCIA results for the aquatic toxicity impact
category, based on the impact assessment methodology presented in Section 3.1.2.13. Complete
results for the CRT and LCD are presented in Tables M-37 and M-38 in Appendix M,
respectively.
                                   Aquatic ecotoxicity
       i 6.00E+00-,
       03
      J 4.00E+00


       § 2.00&00
       £ O.OOEfOO
       o
                                                           5.12E+00
                           8.93E-02
                   1.36E-01          O.OOE+00
                                                   6.19E-02
                                    O.OOE+00
CRT
                                         -2-50&°4
                                                                LCD
                                            Monitor type

                       Figure 3-21. Aquatic ecotoxicity impacts by life-cycle stage
       The life-cycle aquatic toxicity indicator is 0.22 tox-kg per functional unit for the CRT and
5.19 tox-kg per functional unit for the LCD. As shown in the figure, the CRT aquatic toxicity
indicator is driven by water releases in the materials processing stage (64% of total), while 'the
LCD aquatic toxicity indicator is completely dominated by water releases in the manufacturing
stage (99% of total). Both monitor types receive zero scores in the use stage and small, negative
values at end of life. Negative values are due to the offset of electric power plant emissions from
incineration with energy recovery.
       Table 3-51 lists the materials that contribute to the top 99% of the CRT aquatic toxicity
impact score and the LCI data type. As shown in the table, CRT aquatic toxicity impacts are
broadly distributed across a number of different process groups, with most of the top contributors
responsible for less than five percent of the impacts. Most of the LCI data from which the scores
were calculated are from secondary data sets, although a substantial fraction are from primary
data collected to meet the goals and scope of the CDP LCA. Water releases of phosphorus from
CRT tube manufacturing represents the single largest contributor to the CRT aquatic toxicity
score, accounting for 26% of the total. Aquatic toxicity impacts for phosphorus are driven more
by its inherent acute toxicity than the output amount. The phosphorus acute HV is calculated
from a fish LCSO of 0.020 mg/L, which is significantly more toxic than the geometric mean value
of23.5mg/L.
       The only other specific outputs that contribute more than five percent of the CRT aquatic
toxicity score are water discharges of aluminum ions (valence = +3) and copper ions (valence =
-}-l and +2) from aluminum production. The aluminum HV is calculated from an LC50 value of
36 mg/L and a NOAEL value of 3.6 mg/L, which are within an order of magnitude of the
geometric mean values of 23.5 mg/L and 3.9 mg/L. Copper, on the other hand, is much more
toxic to fish, with an LC50 value of 0.014mg/L and a NOAEL value of 0.004 mg/L. The
                                        3-86

-------
                                                            3.3 BASELINE LCIA RESULTS
aluminum aquatic toxicity score exceeds, that of copper because it is discharged in much greater
quantities.
Table 3-51. Top 99% of the CRT aquatic toxicity impact score
Life-cycle stage
Manufacturing
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Manufacturing
Materials processing
Materials processing
Materials processing
Materials processing
Manufacturing
Manufacturing
Manufacturing
Materials processing
Manufacturing
Materials processing
Manufacturing
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Process group
CRT tube manufacturing
Aluminum production
Aluminum production
Invar
Invar
Invar
Lead
CRT tube manufacturing
Ferrite manufacturing
Aluminum production
ABS production
Lead
CRT glass/frit mfg.
CRT tube manufacturing
CRT tube manufacturing
Steel production, cold-
rolled, semi-finished
LPG production
Steel production, cold-
rolled, semi-finished
LPG production
Lead
Polycarbonate production
Steel production, cold-
rolled, semi-finished
Ferrite manufacturing
Aluminum production
ABS production
Invar
Ferrite manufacturing
ABS production
Styrene-butadiene
copolymer production
Aluminum production
Polycarbonate production
Aluminum production
Material
Phosphorus (yellow or white)
Aluminum (+3)
Copper (+1 & +2)
Copper (+1 & +2)
Aluminum (+3)
Zinc (+2)
Aluminum (+3)
Fluoride
Zinc (+2)
Zinc (+2)
Ammonia
Copper (+1 & +2)
Fluorides (F-)
Zinc (elemental)
Copper
Phosphorus (yellow or white)
Phenol
Ammonia
Aluminum (+3)
Zinc (+2)
Copper (+1 & +2)
Copper (+1 & +2)
Aluminum (+3)
Barium sulfate
Aluminum (+3)
Ammonia
Copper (+1 & +2)
Copper (+1 & +2)
Copper (+1 & +2)
Titanium tetrachloride
Mercury compounds
Strontium (Sr IT)
LCI data
type
Primary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Primary
Secondary
Secondary
Secondary
Secondary
Primary
Primary
Primary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Contribution
to impact
score*
26%
12%
9.5%
5.0%
4.4%
4.0%
3.6%
3.1%
3.0%
2.9%
2.7%
2.7%
2.6%
2.3%
2.1%
2.0%
1.9%
1.2%
1.1%
0.82%
0.54%
0.45%
0.43%
0.40%
0.39%
0.36%
0.31%
0.25%
0.24%
0.20%
0.19%
0.14%
                                         3-87

-------
33 BASELINE LCIA RESULTS
             Table 3-51. Top 99% of the CRT aquatic toxicity impact score
Life-cycle stage
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Process group
Steel production, cold-
rolled, semi-finished
Ferrite manufacturing
Lead
Steel production, cold-
rolled, semi-finished
ABS production
Steel production, cold-
rolled, semi-finished
Styrene-butadiene
copolymer production
Steel production, cold-
rolled, semi-finished
Aluminum production
Polycarbonate production
Invar
Material
Aluminum (+3)
Ammonia
Barium sulfate
Nitrogen dixode
Mercury compounds
Zinc (+2)
Mercury compounds
Fluorides (F-)
Lead compounds
Zinc (+2)
Strontium (Sr n)
LCI data
type
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Contribution
to impact
score*
0.12%
0.12%
0.10%
0.088%
0.088%
0.087%
0.086%
0.086%
0.076%
0.076%
0.074%
*Column may not add to 99% due to rounding.

       Table 3-52 lists the materials that contribute to the top 99% of the LCD aquatic toxicity
impact score and the LCI data type. Unlike the CRT, LCD impacts in this category are not
distributed across a number of different process groups, but dominated by phosphorus emissions
from a single process group, LCD monitor/module manufacturing.  Phosphorus releases from
LCD monitor/module manufacturing are several orders of magnitude higher than phosphoius
releases from CRT tube manufacturing (the greatest contributor to the CRT aquatic toxicity
impact score). However, the LCD aquatic toxicity score for phosphorus is still driven by the
inherent acute toxicity of phosphorus, rather than the release amount.
       Other top contributors to LCD impacts in this category include ammonia releases from
PMMA sheet production, and phosphorus emissions from LCD panel components
manufacturing.  No toxicity data were available for ammonia. Consequently, it was assigned a
default HV of two, representative of mean acute and chronic fish toxicity values.
Table 3-52. Top 99% of the LCD aquatic toxicity imi
Life-cycle stage
Manufacturing
Materials processing
Manufacturing
Process group
LCD monitor/module mfg.
PMMA sheet production
LCD panel components
Material
Phosphorus (yellow or white)
Ammonia
Phosphorus (yellow or white)
jact score
LCI data
type
Primary
Secondary
Primary
Contribution
to impact
score
98%
0.63%
0.56%
                                       3-88

-------
                                                             3.3 BASELINE LCIA RESULTS
3.3.14.2 Terrestrial ecotoxicitv
                                                      ..K.
       Figure 3-22 presents the CRT and LCD LCIA results for the terrestrial toxicity impact
category, based on the impact assessment methodology presented in Section 3.1.2.13.  Complete
results for the CRT and LCD are presented in Tables M-39 and M-40 in Appendix M,
respectively.
        1.80Ef03-,
      § 1.20E+03-
      o
     J 6.00E+02 -
      X O.OOE+00
      o
                                 Terrestrial ecotoxicity

                                  1.65E+03
2.28B-02
      9.46Ef01
                                               6.16E+02
                                                        2.33E+02
                                         2.88E-01     4.54E+01
                                                                          1.46E-02
                               CRT
                                              LCD
                                            Monitor type

                     Figure 3-22. Terrestrial ecotoxicity impacts by life-cycle stage
       The life-cycle terrestrial toxicity indicator is 1,970 tox-kg per functional unit for the CRT
and 894 tox-kg per functional unit for the LCD. As shown in the figure, the CRT result is
dominated by toxic chemical outputs from electricity generation in the use stage, which account
for almost 84% of CRT impacts in this category. To a lesser degree, LCD terrestrial toxicity
impacts are also driven by emissions from electricity generation in the use stage, which account
for more than 69% of the total.
       The materials processing stage contributes about 12% of CRT terrestrial toxicity impacts
and about five percent of LCD impacts. The, manufacturing life-cycle stage is responsible for
five and 26% of CRT and LCD impacts in this category, respectively. Both monitors receive
very small terrestrial toxicity scores at end of life.  This is because most terrestrial toxicity
impacts from CRT and LCD recycling and disposal processes are offset by a credit on electric
grid emissions when the monitors are incinerated with energy recovery.
       The terrestrial toxicity impact results are almost identical to the chronic public health
effects results presented previously (see Section 3.3.13). Recall that human health and terrestrial
toxicity impacts are calculated using the same noncancer toxicity values (and the same inventory
data), with the main difference being that toxicity data on carcinogenic effects are excluded from
the terrestrial toxicity impact calculations. However, human health and terrestrial toxicity
impacts are almost identical because:  (1) impacts in both categories are dominated by emissions
of sulfur dioxide from electricity generation (see Tables 3-53 and 3-54 below for top contributors
to the CRT and LCD terrestrial toxicity impacts), and (2) sulfur dioxide has a high hazard value
for noncancer effects and a hazard value of zero for cancer effects.
                                          3-89

-------
33 BASELINE LCIA RESULTS
       Table 3-53 presents the materials that contribute to the top 99% of the CRT terrestrial
toxicity impact score. As already noted, SO2 emissions from a number of different process
groups almost completely dominate CRT impacts in this category, accounting for slightly less
than 99% of the total. Most of these emissions are from the combustion of fossil fuels to
generate electricity. All of the SO2 LCI data are either from secondary data sets not developed
specifically for the CDP or from the electric grid inventories developed from secondary sources
for this project.  Sulfur dioxide has a relatively high HV, based on an inhalation NOAEL of
0.104 mg/m3 and the geometric mean inhalation NOAEL of 68.7 mg/m3. In addition, as noted in
the section on human health effects (3.3.13), from a mass loading perspective (i.e., based on the
inventory alone), SO2 emissions were the second largest contributor to CRT life-cycle air
pollutant emissions, exceeded only by emissions of carbon dioxide (CO2). Carbon dioxide is not
classified as toxic, and therefore did not contribute to the terrestrial toxicity impact category.

            Table 3-53. Top 99% of the CRT terrestrial toxicity impact score
Life-cycle stage
Use
Materials processing
Manufacturing
Manufacturing
Materials processing
Materials processing
Materials processing
Manufacturing
Process group
U.S. electric grid
Invar
Japanese electric grid
U.S. electric grid
Steel production, cold-
rolled, semi-finished
Aluminum production
Polycarbonate production
LPG production
Material
Sulfur dioxide
Sulfur dioxide
Sulfur dioxide
Sulfur dioxide
Sulfur dioxide
Sulfur dioxide
Sulfur dioxide
Carbon monoxide
LCI data
type
Model/secondary
Secondary
Model/secondary
Model/secondary
Secondary
Secondary
Secondary
Secondary
Contribution to
impact score*
83%
8.4%
2.9%
1.3%
1.3%
0.70%
0.40%
0.27%
 ' Column may not add to 99% due to rounding.

       Table 3-54 lists the materials that contribute to the top 99% of the LCD terrestrial toxicity
impact score and the LCI data type. LCD impacts in this category are also dominated by SO2
emissions, which are responsible for roughly 92% of impacts. Like the CRT, most of these
emissions occur from electricity generation, either during the use stage (68%) or manufacturing
(21%). As noted previously, SO2 has a relatively high HV, due to its low toxicity value
(inhalation NOAEL = 0.104 mg/m3).  Similar to the CRT, from a mass loading perspective (i.e.,
based on the inventory alone), SO2 emissions were the second largest contributor to LCD life-
cycle air pollutant emissions, exceeded only by emissions of CO2.
       Other top contributors to the LCD terrestrial toxicity score include phosphine and
phosphorus from LCD monitor/module manufacturing, and carbon monoxide and methane from
natural gas production.  As noted above in the section on chronic occupational health effects, the
phosphine, phosphorus, and benzene scores are driven more by their inherent toxicity than their
output amounts.
                                        3-90

-------
                                                             3.3 BASELINE LCIA RESULTS
Table 3-54. Top 99% of the LCD terrestrial toxicitv impact score
Life-cycle stage
Use
Manufacturing
Manufacturing
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Materials processing
Manufacturing
Process group
U.S. electric grid
Japanese electric grid
LCD monitor/module mfg.
Steel production, cold-
rolled, semi-finished
PMMA sheet production
Natural gas production
Aluminum production
Natural gas production
Polycarbonate production
Natural gas production
LCD monitor/module mfg.
Material
Sulfur dioxide
Sulfur dioxide
Phosphine
Sulfur dioxide
Sulfur dioxide
Benzene
Sulfur dioxide
Carbon monoxide
Sulfur dioxide
Methane
Phosphorus
(yellow or white)
LCI data
type
Model/secondary
Model/secondary
Primary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Secondary
Primary
Contribution
to impact
score
68%
21%
3.2%
1.4%
0.96%
0.59%
0.57%
0.50%
0.50%
0.39%
0.38%
3.3.14.3  Terrestrial toxicitv impact scores modified to exclude sulfur dioxide

       Because the terrestrial toxicity impact scores for both the CRT and the LCD are
dominated by SO2 emissions, a secondary analysis was run to identify the top contributors to
these impacts when SO2 emissions are excluded from the inventories. Results of this analysis
may be more useful to manufacturers seeking to identify problematic toxic chemicals within their
own manufacturing processes.
       Figure 3-23 presents the CRT and LCD terrestrial toxicity impact scores by life-cycle
stage when SO2 emissions are excluded from the inventories.  Under this scenario, the CRT
score is reduced almost 99% from 1,970 tox-kg to 21 tox-kg per functional unit, and the LCD
score is reduced about 94%, from 894 tox-kg to 54 tox-kg per functional unit. Note that these
scores should not be used to evaluate which monitor type has higher overall impacts  in this
category, but they are useful for identifying life-cycle improvement opportunities that were
previously obscured by SO2 impacts.                  '
                                          3-91

-------
33 BASELINE LCIA RESULTS
                           Terrestrial ecotoxicity (without SO2)
       "c  60.0 ->
       3
       "S  40.0 -
p  20.0
           0.0
          -20.0 J
                     5.2
                           11
                                 4.1
                             CRT
                                      -0.26
                                                       LCD
                                                                       -0.18
                                          Monitor type
                      Figure 3-23. Terrestrial ecotoxicity impacts (without SO2)
                                       by life-cycle stage
       With SO2 emissions removed from the inventories, terrestrial toxicity impacts are highest
in the manufacturing life-cycle stage for both the CRT (56% of impacts) and the LCD (70% of
impacts). The materials processing stage is the next largest contributor for both monitor types,
contributing 25% of CRT impacts and 27% of LCD impacts, followed by the use stage, which
contributes 20% of CRT impacts and almost 3% of LCD impacts. Both monitors have slight
negative values at end-of-life, due to the offset of electric grid emissions from incineration with
energy recovery.
       Table 3-55 presents the materials that contribute greater than one percent of CRT
terrestrial toxicity impacts when SO2 emissions are excluded from the CRT inventory. Like the
modified public chronic human health effects results discussed in Section 3.3.13, under this
scenario CRT chronic public health impacts are still being driven by emissions  of criteria air
pollutants, including carbon monoxide, nitrogen oxides, and sulfur oxides (assuming that SO2
emissions comprise a large part of the sulfur oxide emissions shown in the table).  As shown in
the table, emissions of these three pollutants or pollutant categories are responsible for some 46%
of CRT terrestrial toxicity impacts when pure SO2 emissions are excluded from the CRT
inventory. Note that the majority of these emissions occur from the LPG production process, and
most of this LPG is used as a fuel in CRT glass manufacturing. CRT glass manufacturing energy
inputs are uncertain and evaluated in a sensitivity analysis (see Section 3.4). Other significant
contributors include arsenic from lead production, methane from LPG production and the U.S.
electric grid inventory, vanadium and benzene from LPG production, and titanium tetrachloride
from aluminum production.
                                        3-92

-------
                                                            3.3 BASELINE LCIA RESULTS
             Table 3-55. Materials contributing greater than 1% of the CRT
Life-cycle stage
Manufacturing
Materials processing
Use
Manufacturing
Use
Manufacturing
Manufacturing
Manufacturing
Materials processing
Use
Manufacturing
Manufacturing
Use
Use
Materials processing
Materials processing
Manufacturing
Materials processing
Manufacturing
Process group
LPG production
Lead
U.S. electric grid
LPG production
U.S. electric grid
LPG Production
LPG production
LPG production
Aluminum production
(all virgin)
U.S. electric grid
CRT glass/frit mfg.
LPG production
U.S. electric grid
U.S. electric grid
Steel Prod., cold-rolled,
semi-finished
Invar
CRT tube manufacturing
Lead
LPG production
Material
Carbon monoxide
Arsenic
Nitrogen oxides
Vanadium
Carbon monoxide
Benzene
Methane
Sulfur oxides
Titanium tetrachloride
Methane
Fluorides (F-)
Nitrogen oxides
Arsenic
Hydrochloric acid
Carbon monoxide
Titanium tetrachloride
Carbon monoxide
Titanium tetrachloride
Arsenic
LCI data
type
Secondary
Secondary

Secondary
Model/secondary
Secondary
Secondary
Secondary
Secondary

Primary
Secondary


Secondary
Secondary

Secondary
Secondary
Contribution
to impact
score
26%
11%
5.7%
5.1%
49%
4.2%
4.1%
3.9%
3.5%
3.1%
2 3%
28%
27%
2.4%
1.4%
1.2%
1.0%
1.0%
1.0%
       Table 3-56 presents the materials that contribute greater than one percent of LCD
terrestrial toxicity impacts when SO2 emissions are excluded from the LCD inventory. As with
the LCD modified chronic human health results discussed in Section 3.3.13, when SO2 emissions
are excluded, phosphine emissions from LCD monitor/module manufacturing are the dominant
factor in the LCD terrestrial toxicity score, contributing 53% of the total. Other significant
contributors include benzene, carbon monoxide, methane, and nitrogen oxides from natural gas
production, and phosphorus, fluorides, tetramethyl ammonium hydroxide, and nitrogen oxides
from LCD monitor/module manufacturing.  Recall that the LCD monitor/module manufacturing
process consumes the majority of the natural gas made in the natural gas production process,
where LNG is used as an ancillary material. However, only one of the seven LCD
monitor/module manufacturers that provided inventory data to the CDP reported the ancillary use
of LNG. Other manufacturers reported using LNG as a fuel.
                                         3-93

-------
3.3 BASELINE LCIA RESULTS
             Table 3-56. Materials contributing greater than 1% of the LCD
                      terrestrial toxicity impact score (without SO2)
Life-cycle stage
Manufacturing
Materials processing
Materials processing
Materials processing
Manufacturing
Manufacturing
Materials processing
Manufacturing
Manufacturing
Process group
LCD module/monitor mfg.
Natural gas production
Natural gas production
Natural gas production
LCD module/monitor mfg.
LCD module/monitor mfg.
Natural gas production
LCD module/monitor mfg.
LCD module/monitor mfg.
Material
Phosphine
Benzene
Carbon monoxide
Methane
Phosphorus (yellow or white)
Fluorides (F-)
Nitrogen oxides
Tetramethyl ammonium
hydroxide
Nitrogen oxides
LCI data
type
Primary
Secondary
Secondary
Secondary
Primary
Primary
Secondary
Primary
Primary
Contribution
to impact
score
53%
9.8%
8.2%
6.5%
6.3%
4.7%
1.2%
1.2%
1.0%
33.14.4 Limitations and uncertainties

       Most of the limitations and uncertainties in the ecotoxicity results are similar to the
limitations and uncertainties in the human health effects scores. The reader is referred to Section
3.3.13 for a full discussion of these limitation and uncertainties. In summary, they 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 ecotoxicity.
3.     LCI data limitations and uncertainties associated with the accuracy and
       representativeness of the inventory data.

       With regard to toxicity data, other limitations and uncertainties in the ecotoxicity results
are related to the use of surrogates to assess toxicity to all species within an impact category. For
example, the aquatic toxicity category uses fish (usually the fathead minnow) as a surrogate to
assess toxicity to all aquatic organisms, but it has been well-established in the ecotoxicology
literature that fish are not the most sensitive test species to all or most industrial chemicals. In
fact, invertebrates  (daphnids) or algae (Selenastrum) often are more sensitive to particular
chemicals than fish (Smrchek, 1999). Similarly, the terrestrial toxicity category uses mammals,
primarily rodents,  as a surrogate to assess toxicity to all terrestrial organisms. Terrestrial plants
and soil organisms (insects, earthworms, etc.) are not considered, but both of these may be more
sensitive than mammals.
       Because there are difficulties in comparing test endpoints for different types of
organisms, and because there is a very limited toxicity database for some of the other organisms,
the LCIA methodology employed in this study uses fish as a surrogate for aquatic toxicity and
mammals as a surrogate for terrestrial ecotoxicity. This helps to reduce data gaps and the
difficulties in comparing test endpoints for different types of organisms. Furthermore,  we
                                         3-94

-------
                                                            3.3 BASELINE LCIA RESULTS
 believe this approach to be acceptable for a study, such as the CDP LCA, that gives relative
 ranking of impacts from different chemicals 6r process groups instead of absolute values.
 However, it should be noted that this approach can result in an underestimation of the absolute
 ecotoxicity and hazards of chemicals.
       With regard to LCI data limitations not discussed previously, it should be noted that the
 CRT and LCD LCAs do not address spills or other accidental releases that could have significant
 adverse effects on aquatic or terrestrial organisms. This is a common limitation of LCA, which
 is often too labor-intensive to address different operational scenarios across the product life
 cycle.                                                       .

 3.3.15 Summary of Top Contributors by Impact Category

       Tables 3-57 and 3-58 summarize the top contributors to CRT and LCD life-cycle impacts
 by impact category. As shown in Table 3-57, CRT impacts are largely driven by two factors: (1)
 the large amount of LPG fuel used in CRT glass/frit manufacturing, and (2) the relatively large
 amount of electricity consumed during the use stage.  The LPG production process yields the
 CRT's top contributor in eight of 20 impact categories. Most of this LPG is used as a fuel source
 in CRT glass manufacturing in the glass/frit process group, which, in turn, produces the top
 contributor to two of 20 impact categories. Thus, LPG used in the glass/frit process group
 (primarily CRT glass manufacturing) is ultimately the key driver for CRT impacts in ten
 categories. Similarly, outputs from electricity generation during the use stage result in the top
 contributor to seven CRT impact categories. Note that in 14 of the 20 impact categories, the top
 contributor to CRT impacts is responsible for more than 50% of impacts.
       Both the glass manufacturing energy and the use stage lifespan (which determines the
 amount of electricity generated during the use stage) are evaluated in a sensitivity analysis in
 Section 3.4. In the modified glass energy sensitivity analysis, LPG inputs are greatly reduced and
 impacts are therefore reduced, but in the modified lifespan (manufactured life) sensitivity
 analysis, the number of hours a monitor is in use is increased. Thus, CRT impacts are increased.
       LCD impacts are not as dominated by a few data points, but a few processes (LCD
 monitor/module manufacturing and electricity generation in the use stage) are responsible for a
 large percent of the impacts. As shown in Table 3-58, both of these processes result in the top
 contributors to six LCD impact categories each. In addition, the process to produce LNG used as
 an ancillary material in LCD monitor/module manufacturing is the top contributor to an
 additional impact category (photochemical smog). Note that in 11 of the 20 impact categories,
 the top contributor to LCD impacts is responsible for more than 50% of impacts.
       Like the CRT, both the glass energy inputs and use stage lifespan of the LCD are
 evaluated in a sensitivity analysis in Section 3.4. LCD monitor/module manufacturing energy
 and LCD EOL dispositions are also evaluated.  LCD monitor/module manufacturing energy was
 selected for a sensitivity analysis because of the high degree of variability seen in data provided
by manufacturers. The impacts presented in the baseline scenario are calculated with energy
 outliers removed from the average, but the outliers are included in the sensitivity, analysis. This
results in higher electricity consumption but lower fuel consumption, which, in turn, causes
reduced impacts in some categories  and increased impacts in others.  Note that the LNG used as
an ancillary material in LCD monitor/module manufacturing is not affected by the sensitivity
analysis since it only focuses on materials used as an energy source.
                                         3-95

-------
33 BASELINE LCIA. RESULTS
      Table 3-57. Summary of top contributors to CRT impacts by impact category
Impact category
Renewable resource
use
Nonrenewable
resource use
Energy use
Solid waste landfill
use
Hazardous waste
landfill use
Radioactive waste
landfill use
Global warming
Ozone depletion
Photochemical smog
Acidification
Air participates
Water eutrophication
Water quality, BOD
Water quality, TSS
Radioactivity
Chronic health effects,
occupational
Chronic health effects,
public
Aesthetics (odor)
Aquatic toxicity
Terrestrial toxicity
Top contributors
Life-cycle
stage
Manufacturing
Manufacturing
Manufacturing
Use
End-of-life
Use
Use
Use
Manufacturing
Use
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Materials
Processing
Manufacturing
Use
Manufacturing
Manufacturing
Use
Process group
LPG production
LPG production
CRT glass/frit mfg.
U.S. electric grid
CRT landfilling
U.S. electric grid
U.S. electric grid
U.S. electric grid
LPG production
U.S. electric grid
LPG production
LPG production
LPG production
LPG production
Steel production, cold-
rolled, semi-finished
CRT glass/frit
manufacturing
U.S. electric grid
LPG production
CRT tube
manufacturing
U.S. electric grid
Material
water
Petroleum (hi ground)
Liquefied petroleum
gas
Coal waste
EOL CRT monitor,
landfilled
Low-level radioactive
waste
Carbon dioxide
Bromomethane
Hydrocarbons,
unspeciated
Sulfur dioxide
PM
COD
BOD
Suspended solids
Plutonium-241
(isotope)
Liquefied petroleum
gas
Sulfur dioxide
Hydrogen sulf ide
Phosphorus
(yellow or white)
Sulfur dioxide
Contribution
to impact
score
79%
56%
72%
38%
91%
61%
64%
49%
36%
47%
'43%
72%
96%
97%
62%
78%
83%
94%
26%
83%
                                    3-96

-------
                 3.3 BASELINE LCIA RESULTS
Table 3-58. Summary of top contributors to LCD impacts by impact catesorv
Impact category
Renewable resource
use
Nonrenewable resource
use
Energy use
Solid waste landfill use
Hazardous waste
landfill use
Radioactive waste
landfill use
Global warming
Ozone depletion
Photochemical smog
Acidification
Airparticulates
Water eutrophication
Water quality, BOD
Water quality, TSS
Radioactivity
Chronic health effects,
occupational
Chronic health effects,
public
Aesthetics (odor)
Aquatic toxicity
Terrestrial toxicity

Life-cycle stage
Manufacturing
Materials
processing
Use
Use
End-of-Iife
Use
Manufacturing ,
Manufacturing
Materials
processing
Use
Materials
processing
Manufacturing
Manufacturing
Manufacturing
Materials
processing
Manufacturing
Use
Manufacturing
Manufacturing
Use
Top contributors
Process group
LCD monitor/module mfg.
Natural gas production
LCD monitor use
U.S. electric grid
LCD landfffling
U.S. electric grid
LCD monitor/module mfg.
LCD panel components
manufacturing
Natural gas production
U.S. electric grid
Steel production, cold-
rolled, semi-finished
LCD monitor/module mfg.
LCD monitor/module mfg.
LPG production
Steel production, cold-
rolled, semi-finished
LCD monitor/module mfg.
U.S. electric grid
LPG production
LCD monitor/module mfg.
U.S. electric grid
Material
Water
Natural gas
(in ground)
Electricity
Coal waste
EOLLCD
monitor,
landfilled
Low-level
radioactive waste
Sulfur
hexafluoride
HCFC-225cb
Nonmethane
hydrocarbons,
unspeciated
Sulfur dioxide
PM
Nitrogen
BOD
Suspended solids
Plutonium-241
(isotope)
Liquefied natural
Sulfur dioxide
Hydrogen sulfide
Phosphorus -
(yellow or white)
Sulfur dioxide
Contribution
to impact
score
38%
65%
30%
44%
97%
44%
29%
34%
45%
31%
45%
67%
61%
66%
96%
58%
68%
94%
98%
68%
3-97

-------
 3.4 SENSITIVITY ANALYSES
 3.4   SENSITIVITY ANALYSES

       Due to assumptions and uncertainties in this LCA, as in any LCA, several sensitivity
 analyses of the baseline results were conducted.  Section 2.7.3 described how areas for sensitivity
 analyses (scenarios) were selected, and the modifications made to the baseline inventory.
 Sections 3.4.1 through 3.4.4 recap these modifications and present sensitivity analysis results.
 Section 3.4.5 summarizes the effects of different scenarios on CRT and LCD impacts.

 3.4.1  Manufactured Life Scenario

       Due to the uncertainty and assumptions associated with the baseline use stage lifespan
 (effective life) scenario, a "manufactured life" scenario was also considered. Recall that the
 manufactured life is defined as the length of time a monitor is designed to operate effectively,
 while the effective life is defined as the actual amount of time a monitor is used, by one or
 multiple users, before it reaches its final disposition. The manufactured life is the number of
 hours a monitor would function as manufactured, and is independent of user choices or actions.
 Section 2.4.1.3 presented a detailed discussion of how the manufactured life was determined, and
 is summarized below.
       The manufactured life of both monitor types was estimated using the
 mean-time-before-failure (MTBF) specifications of the monitor and its components.  From
 review of MTBF information obtained on CRT-based monitors (see Appendix H, Attachment A,
 Table A2), it appears that the CRT tube itself is the component that 99% of the time determines
 whether the entire monitor has reached its end-of-life. Thus, an average of the two ranges
 obtained on the estimated lifetime of CRT tubes (10,000 and 15,000 hours) was used as the CRT
 manufactured lifetime (12,500 hours).
       For active matrix LCDs, the components that have the greatest potential to fail first are
 the display panel itself (including the liquid crystals and thin-film transistors), backlights, driver
 integrated circuit (1C) tabs, and other smaller components. The backlights and driver 1C tabs can
 be field-replaced, thus their failure does not necessarily represent the end of the monitor's life.
 However, failure of the liquid crystals or transistors, which would require replacement of the
 display panel itself, would most likely mean that the monitor cannot be cost-effectively repaired.
 Thus, in this study, the amount of time an LCD monitor would operate during its manufactured
 life is assumed to be the average of the non field-replaceable values, or 45,000 hours. In order
 for a monitor to operate for 45,000 hours, any major field-replaceable parts that have MTBFs less
' than 45,000 hours are accounted for in the inventory. For example, assuming the backlights last
 on average 32,500 hours (the average of the values obtained for backlights), approximately 1.4
 backlights on average would be needed for every panel during its 45,000 hour lifetime.
       To calculate the manufactured life electricity consumption (kWh/life), the energy use rate
 (kW) was multiplied by the lifespan (hours/life) for each monitor in each power mode (see Table
 2-20 in Section 2.4.1.3).  The LCD manufactured life (45,000 hours) is 3.6 times greater than the
 CRT manufactured life (12,500 hours). In an LCA, comparisons are made based on functional
 equivalency. Therefore, if one monitor will operate for a longer period of time than another,
 impacts should be based on an equivalent use. Thus, based on equivalent use periods, 3.6 CRTs
 would need to be manufactured for every LCD.  This was incorporated into the profile analysis
 for the manufactured life LCA. To apply the manufactured life scenario to the CRT and LCD
                                         __,_.,_

-------
                                                              3.4 SENSITIVITY ANALYSES
life-cycle profiles, the following modifications were made tdithe baseline (effective life)
scenario:

       change the CRT electricity input in the use stage from 635 kWh (2,286 MJ) to 788 kWh
       (2,837 MJ);3
       change the LCD electricity input in the use stage from 237 kWh (853 MJ) to 1,035 kWh
       (3,726 MJ);
•      increase the manufacturing of CRTs by a factor of 3.6 to account for the functional
       equivalency of CRTs and LCDs. This was done by increasing the functional unit by a
       factor of 3.6, which equates to manufacturing, using, and recycling or disposing of 3.6
       times more CRTs than in the baseline case; and
•      increase the manufacturing of the LCD backlight lamp by a factor of 1.4 to  account for
       the functional equivalency of LCDs and CRTs.  This was done by increasing the
       backlight lamp mass (0.0023 kg) by a factor of 1.4, which in turn results in  an increase in
       inputs and outputs associated with manufacturing the backlilght.

       Table 3-59 presents the CRT and LCD life-cycle results by impact category for the
baseline and manufactured life scenarios. It also presents the percent change from the baseline to
the manufactured life scenario for both monitor types. Note that the manufactured  life results are
most useful for evaluating the CRT and LCD together, not for comparing the CRT  or LCD
baseline (effective life)  results to its manufactured life results. This is because the CRT and LCD
manufactured life results are functionally equivalent, but the CRT to CRT and LCD to LCD
effective life and manufactured life scenarios are not. The baseline for both monitor types
represents impacts from manufacture and final disposition of one monitor.  The CRT
manufactured life scenario represents impacts from 3.6 CRT monitors and the LCD
manufactured life scenario represents impacts from one LCD monitor and 1.4 backlights.
However, the percent change figures are presented to better understand how the manufactured
life scenario affects overall results.
       3 This represents the electricity use for a 12,500 hour life span. This figure is then multiplied by a factor of
3.6 in the functional equivalency calculations (see third bullet, below).

-------
3.4 SENSITIVITY ANALYSES
         Table 3-59. Baseline and sensitivity analysis results—manufactured life

Impact category
Renewable resource use
Nonrenewable resource use
Energy use
Solid waste landfill use
Hazardous waste landfill use
Radioactive waste landfill use
Global warming
Ozone depletion
Photochemical smog
Acidification
Airparticulates
Water eutrophication
BOD
TSS
Radioactivity
Chronic health effects,
occupational
Chronic health effects, public
Aesthetics (odor)
Aquatic toxicity
Terrestrial toxicity

Units/ Monitor
kg
kg
MJ
m3
m3
m3
kg-CO2 eq.s
kg-CFC-11 eq.s
kg-ethene eq.s
kg-SO2 eq.s
kg
kg-phosphate
eq.s
kg
kg
Bq
tox-kg
tox-kg
m3
tox-kg
tox-kg
CRT
Baseline
1.31e+04
6.68e+02
2.08e+04
1.67e-01
1.68e-02
2.00e-04
6.95e+02
2.05e-05
1.71e-01
5.25e+00
3.01e-01
4.82e-02
1.95e-01
8.75e-01
3.85e+07
9.34e+02
1.98e+03
7.58e+06
2.25e-01
1.97e+03
Manu-
factured
4.83e+04
2.58e+03
7.70e+04
6.86e-01
6.05e-02
S.OOe-04
2.90e+03
8.27e-05
6.20e-01
2.19e+01
1.13e+00
1.74e-0r
7.02e-01
3.15e+00
1.14e+08
3.396+03
8.56e+03
2.74e+07
S.lOe-Ol
8.54e+03
%
change
268%
286%
270%
312%
260%
328%
317%
304%
262%
317%
277%
260%
260%
260%
197%
263%
333%
262%
260%
333%
LCD
Baseline
2.80e+03
3.64e+02
2.84e+03
5.43e-02
3.60e-03
l.OOe-04
5.93e+02
1.37e-05
1.41e-01
2.96e+00
l.lSe-Ol
4.96e-02
2.83e-02
6.15e-02
1.22e+07
6.96e+02
9.02e+02
5.04e+06
5.19e+00
8.94e+02
Manu-
factured
4.30e+03
6.12e+02
5.71e+03
1.79e-01
3.60e-03
3.006-04
1.17e+03
2.66e-05
1.476-01
7.296+00
1.87e-01
4.96e-02
2.83e-02
6.15e-02
1.28e+07
7.41e+02
2.98e+03
5.30e+06
5.19e+00
2.97e+03
%
change
53.5%
68.0%
101%
230%
0.00%
194%
97.3%
94.1%
4.22%
146%
63.4%
0.02%
0.02%
0.02%
4.49%
6.47%
230%
5.00%
0.02%
232%
* Bold indicates impact category indicator that reversed direction from the baseline scenario such that the CRT
indicator is now greater than the LCD.

       As shown in Table 3-59, under the manufactured life scenario CRT impacts exceed those
of the LCD in every impact category except aquatic toxicity. CRT impacts were expected to be
greater than those of the LCD in most impact categories for the following reasons:

•      Under the baseline scenario CRT impacts exceeded those of the LCD in every category
       but water eutrophication and aquatic toxicity.
•      The manufactured life scenario assumes more CRTs are manufactured than LCDs during
       the manufactured life lifespan, which results in greater impacts.
•      The manufactured life use stage is longer than the baseline, effective life use stage, and
       the CRT consumes more electricity during use than the LCD.

       By looking at the percent change in impact scores from the baseline to manufactured life
for a monitor type we can better understand which aspect of the life-cycle is driving impacts. For
example, CRT impacts increased by roughly 260% in several impact categories, which is the
increase from manufacturing or disposing of an additional 2.6 monitors. Energy impacts
increased by more than 260% due to the additional increase in electricity consumption during
use.  The CRT chronic public human health effects category increased by some 330%. This is
explained by the increase in SO2 emissions in the use stage and the high HV for SO2, which has a
                                        3-100

-------
                                                               3.4 SENSITIVITY ANALYSES
proportionately greater effect on overall impacts than increased outputs of other pollutants with
lower HVs in other life-cycle stages.     !   *'             r
       LCD impacts increased only slightly from the baseline to the manufactured life scenario
in some impact categories, but increased up to 230% in others. Most of the LCD impact
categories with less than one percent increase are for impacts related to water discharges (e.g.,
water eutrophication, aquatic toxicity, etc.). This is because the most significant change to the
LCD inventory from the baseline to the manufactured life scenario was in the use stage,  and few
water discharges are reported in the U.S. electric grid inventory.  On the other hand, the  chronic
public health effects and terrestrial toxicity impact categories show the greatest increase. These
results are driven by air emissions of SO2 from U.S. power production, which increased
significantly with the longer lifespan.
       To further illustrate how the longer lifespan (and additional manufacturing requirements,
mainly for the CRT) in the manufactured life scenario is affecting impacts, Figures 3-24 and 3-25
compare the energy impacts and public chronic health effects, respectively, of both monitor types
under the baseline and manufactured life scenarios. As shown in Figure 3-24, CRT energy
impacts are still dominated by the manufacturing stage in the manufactured life scenario. This is
mainly due to the large amount of LPG used to manufacture 3.6 sets of CRT glass. On the other
hand, LCD energy impacts during the use stage exceeded those in manufacturing by a factor of
about 2.6 in the baseline scenario, but are ten times greater in the manufactured life scenario.
This is due to the longer lifespan for a single LCD in the manufactured life scenario.
     •a
     •3
 7.00E+04 i
 6.00E404 -
 5.00E4O4 -
 4.00E404 -
 3.00E+04 -
 2.00E+04 -
 1.00E404 -
 O.OOE+00
-1.00E404 -I
                                    Energy use impacts
                    Baseline
                         Manu. Life
                                                             Baseline
                                                                  Manu. Life
                                               Scenario
                         CRT                                      LCD
                     Figure 3-24. Energy use: baseline and manufactured life sensitivity results
       Figure 3-25 shows that CRT chronic public health impacts are similarly distributed in the
baseline and manufactured life scenarios.  However, a greater percentage of LCD chronic health
impacts are in the use stage under the manufactured life scenario than the baseline due to the
longer use stage.                               •
                                          3-101

-------
3.4 SENSITIVITY ANALYSES
                                 Chronic public health effects
   <
   "Bb
   ^
 8.00E+03 -,

 6.00E403 -

 4.00E+03 -

 2.00E+03 -

 O.OOE+00

-2.00E+03 -
                                Manu. Life              .     ..  . Baseline        Manu. Life
                         CRT                                         LCD
                                              Scenario
                Figure 3-25. Chronic public health effects: baseline and manufactured life sensitivity
                                               results
3.4.2  Modified Glass Energy Scenario

       One area of relatively large uncertainty and variability in the primary data supplied for the
project was the glass manufacturing data. Both the CRT and LCD glass data were based on data
supplied by CRT lead oxide (PbO) glass manufacturers, since no LCD glass manufacturers were
willing to provide data to the project. To represent an LCD glass inventory, lead (Pb) materials
in the CRT glass inventory were removed. Based on conversations with industry members, it
was assumed that the same amount of energy per kilogram of glass produced is used to generate
LCD and CRT glass. In addition, large variability among the three data sets collected resulted in
a large degree  of uncertainty in the glass inventory. Finally, a number of the CRT impact
category results are being driven by the glass energy data or by the production process for
producing the large amount of LPG used as a fuel in glass manufacturing.  Consequently, the
glass manufacturing inventories for both LCD and CRT glass were modified and life-cycle
impacts were recalculated.
       To conduct the sensitivity analysis, the energy input data for glass manufacturing were
modified by removing, from the average, data that appeared unusually large, and that might be
inconsistent with general industry statistics. However, industry statistics are greatly lacking
when specifically considering specialty glasses such as CRT and LCD glass. Modifying the
energy inputs greatly reduced the  fuel energy amounts reported for the glass production process.
       The baseline scenario, based on averaged primary data from manufacturers, assumed that
the total energy to produce a kilogram of CRT or LCD glass was 1,560 MJ (433 kWh) of energy,
with only 0.3% of that as electrical energy. The sensitivity analysis scenario assumes 16.3 MJ
(4.5 kWh) per kilogram of glass produced, with  approximately 30% as electrical energy. The
majority of the fuel energy in the baseline scenario was from LPG. The actual input amounts are
not presented here to protect the confidentiality of data provided by glass manufacturers.
       Under the sensitivity (modified glass energy) scenario, only the manufacturing stage is
affected, since the production of the fuels used during manufacturing is included in the
manufacturing life-cycle stage. All impact categories, not only energy, are also affected because
                                        3-102

-------
                                                              3.4 SENSITIVITY ANALYSES
the inputs and outputs from fuel production and electricity generation processes affect each of the
impact categories evaluated in this study.
       Table 3-60 shows the baseline impact results and the revised impact results based on the
modified glass energy inputs.  The overall life-cycle impact results are highly sensitive to the
energy consumption values from glass manufacturing.  Under the modified glass energy scenario,
the nonrenewable resource use, global warming, photochemical smog, BOD, TSS, chronic
occupational health effects, and odor impact categories reversed direction such that the LCD had
greater impacts within each impact category than the CRT in the overall life cycle. Note that the
percent change in CRT results in most impact categories is much greater than that of the
corresponding LCD results. This is because the CRT uses approximately ten times more glass
than the LCD and therefore, the CRT results are much more sensitive to the glass manufacturing
data than are the LCD results.
Table 3-60. Baseline and sensitivity analysis results — modified glass energy

Impact category
Renewable resource use
Nonrenewable resource use
Energy use
Solid waste landfill use
Hazardous waste landfill use
Radioactive waste landfill use
Global warming
Ozone depletion1"
Photochemical smog
Acidification
Air participates
Water eutrophication
BOD
TSS
Radioactivity
Chronic health effects, public
Chronic health effects,
occupational
Aesthetics (odor)
Aquatic toxicity
Terrestrial toxicity

Units/Monitor
kg
kg
MJ
m3
m3
m3
kg-CO2 eq.s
kg-CFC-11 eq.s
kg-ethene eq.s
kg-SO2 eq.s
kg
kg-phosphate eq.s
kg
kg
Bq
tox-kg
tox-kg
m3
tox-kg
tox-kg
CRT
Baseline
1.31e+04
6.68e+02
2.08e+04
1.67e-01
1.68e-02
2.00e-04
6.95e+02
2.05e-05
1.71e-01
5.25e+00
S.Ole-01
4.82e-02
1.95e-01
8.75e-01
3.85e+07
1.98e+03
9.34e+02
7.58e+06
2.25e-01
1.97e+03
Glass
energy
2.67e+03
2.35e+02
3.02e+03
1.23e-01
1.54e-02
2.00e-04
5.23e+02
1.97e-05
5.59e-02
4.02e+00
1.72e-01
4.10e-03
7.00e-03
2.06e-02
3.16e+07
1.97e+03
2.30e+02
1.09e+04
2.18e-01
1.97e+03
%
change
-79.6%
-64.8%
-85.5%
-26.0%
-8.13%
0.12%
-24.8%
-3.75%
-67.3%
-23.4%
-42.9%
-91.5%
-96.4%
-97.6%
-17.9%
-0.56%
-75.4%
-99.9%
-3.07%
-0.42%
LCD
Baseline
2.80e+03
3.64e+02
2.84e+03
5.43e-02
3.60e-03
l.OOe-04
5.93e+02
1.37e-05
1.41e-01
2.96e+00
1. 15e-01
4.96e-02
2.83e-02
6.15e-02
1.22e+07
9.02e+02
6.96e+02
5.04e+06
5.19e+00
8.94e+02
Glass
energy8
2.43e+03
3.49e+02
2.04e+03
5.27e-02
3.606-03
l.OOe-04
5.87e+02
1.37e-05
1.37e-01
2.92e+00
1.10e-01
4.80e-02
2.16e-02
3.09e-02
1.22e+07
9.01e+02
6.63e+02
4.79e+06
5.19e+00
8.94e+02
%
change
-13.4%
-4.14%
-28.0%
-2.82%
-1.35%
0.01%
-1.01%
-0.18%
-2.84%
-1.45%
-3.97%
-3.17%
-23.8%
-49.7%
0.00%
-0.02%
-4.74%
-4.99%
0.01%
-0.01%
1 Bold indicates impact category indicator that reversed direction from the baseline scenario such that the LCD
indicator is now greater than the CRT.
b LCD impacts in this category are greater than CRT impacts when phased out substances are removed from the
inventories (see Section 3.3.6).

       The energy impacts for the baseline and modified glass energy scenarios are presented in
Figure 3-26. In the baseline scenario, over 18,000 MJ of energy were consumed per CRT
monitor during manufacturing. Almost 83% of this was from the glass/frit process group, mainly
from glass manufacturing energy alone.  When the glass energy inputs are reduced under the
modified scenario, total energy use in the CRT manufacturing stage decreases some 97% to just
                                         3-103

-------
3.4 SENSITIVITY ANALYSES
under 500 MJ, and the use stage dominates the overall life-cycle energy impacts at approximately
2,300 MJ per functional unit (i.e., per monitor). The 97% decrease in manufacturing stage
energy use is due to the reduced glass manufacturing fuel inputs and the consequent reduction in
energy inputs to the fuel production process.
    •J
                                   Energy use impacts
                                                 13 Up stream
                                                 • Mfg.
                                                 QUse
                                                 QEOL
                               M
                               vo
       -5.00E403 -
                   Baseline
   ModQass Energy

CRT
Baseline     Mod'Qass Energy

        LCD
                                             Scenario
                    figure 3-26. Energy impacts: baseline and modified glass energy
                                         sensitivity results
       The modified glass energy scenario has a lesser, but still significant, effect on the
distribution of LCD energy impacts across life-cycle stages.  Under the sensitivity scenario, the
LCD manufacturing stage energy consumption is reduced 55% from 1,440 MJ per monitor to
about 640 MJ per monitor, and the use stage becomes the biggest energy consumer at about 850
MJ per monitor.
       Global warming is one of the impact categories in which the CRT has the greater impacts
than the LCD under the baseline scenario, but the LCD has the greater impacts under the
sensitivity analysis. Figure 3-27 shows the global warming impacts for both monitor types under
the baseline and modified glass energy scenarios.  Under the latter scenario, CRT global
warming impacts in the manufacturing stage are reduced some 85%, but LCD impacts are only
reduced 2.5%. Again, this illustrates the greater sensitivity of CRT impact results to glass energy
inputs. Also, as discussed in Section 3.3.5, a large part of LCD global warming impacts are
driven by sulfur hexafluoride emissions from LCD monitor/module manufacturing, which are
unaffected by the revised glass energy scenario.
                                        3-104

-------
                                                             3.4 SENSITIVITY ANALYSES
                                    Global warming
                   Baseline    Mod Glass Energy

                         CRT
Scenario
Baseline    . Mod. Glass Energy

        LCD
                 figure 3-27. Global warming: baseline and modified glass energy sensithity
                                            results
3.4.3  Modified LCD Module Energy Scenario

       LCD monitor/module manufacturing energy was another area of relatively large
uncertainty and variability in the inventory data.  As discussed in Section 2.7.3.3, total energy
inputs reported in six data sets received from LCD monitor/module manufacturers in Japan and
Korea ranged from 330 MJ to 7,310 MJ, with a mean and standard deviation of 2,269 MJ and
2,906 MJ, respectively. The manufacturing energy data reported in two of the data sets were
found to be outliers and removed from the averages used in the baseline inventory. However, for
this sensitivity analysis, the outliers were added back in to the averages. Thus, to apply the
modified LCD manufacturing energy scenario to the LCD profile, the following modifications
were made to electricity and fuels:

•      changed the electric energy inputs to the LCD monitor/module manufacturing process
       group from 82.1 kWh (253 MJ) per monitor to 70.1 kWh (217 MJ) per monitor,
•      changed the fuel oil # 4 inputs from 0.25 kg to 0.30 kg per monitor,
•      changed the kerosene inputs from 0.35 kg to 0.23 kg per monitor,
       changed the LPG inputs from 0.68 kg to 0.45 kg per monitor,
•      changed the LNG inputs from 3.8 kg to 45 kg per monitor, and
•      changed the natural gas inputs from 0.99 to 0.70 kg per monitor.
       Note that one LCD monitor/module manufacturer also reported using a large amount of
LNG as an ancillary material. However, this input amount is not affected by the sensitivity
analysis, which only deals with inputs used as an energy source.
       Table 3-61 presents the baseline impact results and the revised LCD impact results based
on the modified LCD module energy scenario. It also shows the baseline CRT results. Under
the modified LCD energy scenario, LCD impacts in ten categories actually decrease slightly, due
to the slight decrease in average electrical energy consumed during LCD monitor/module
manufacturing.  However, impacts in six categories increase slightly, and impacts in four
categories (nonrenewable resource use, energy use, photochemical smog and chronic
                                         3-105

-------
3.4 SENSITIVITY ANALYSES
occupational health effects) increase by more than 10%.  As expected, life-cycle energy impacts
are the most affected by this sensitivity analysis, due to the increased fuel consumption during
manufacturing.  However, under this scenario, none of the impact category results reversed
direction from the baseline such that the LCD now has greater impacts than the CRT or vice
versa, where the baseline LCD impacts were greater than the CRT.

    Table 3-61. Baseline and sensitivity analysis  results—LCD modified module energy
Impact category
Renewable resource use
Nonrenewable resource use
Energy use
Solid waste landfill use
Hazardous waste landfill use
Radioactive waste landfill use
Global wanning
Ozone depletion8
Photochemical smog
Acidification
Air participates
Water eutrophication
BOD
TSS
Radioactivity
Chronic health effects,
occupational
Chronic health effects, public
Aesthetics (odor)
Aquatic toxicity
Terrestrial toxicity
Units/Monitor
kg
kg
MJ
m3
m3
m3
kg-CO2 eq.s .
kg-CFC-lleq.s
kg-ethene eq.s
kg-SO2 eq.s
kg
kg-phosphate eq.s
kg
kg
Bq
tox-kg
tox-kg
m3
tox-kg
tox-kg
CRT
Baseline
1.31e+04
6.68e+02
2.08e+04
1.67e-01
1.68e-02
2.00e-04
6.95e+02
2.056-05
1.71e-01
5.25e+00
3.01e-01
4.82e-02
1.95e-01
8.75e-01
3.85e+07
9.34e+02
1.98e+03
7.58e+06
2.25e-01
1.97e+03
LCD
Baseline
2.80e+03
3.64e+02
2.84e+03
5.43e-02
3.60e-03
l.OOe-04
5.93e+02
1.37e-05
1.41e-01
2.96e+00
1.15e-01
4.96e-02
2.83e-02
6.15e-02
1.22e+07
6.96e+02
9.02e+02
5.04e+06
5.19e+00
8.94e+02
Mod. energy
2.78e+03
4.06e+02
4.68e+03
5.47e-02
3.60e-03
l.OOe-04
6.17e+02
1.37e-05
1.61e-01
3.00e+00
1.19e-01
4.96e-02
2.83e-02
6.13e-02
1.22e+07
7.66e+02
8.82e+02
5.04e+06
5.19e+00
8.74e+02
% change
-0.69%
11.4%
64.9%
0.74%
-0.01%
-3.88%
4.05%
-0.26%
13.7%
1.48%
3.85%
0.00%
-0.12%
-0.30%
-0.09%
10.1%
-2.14%
0.01%
0.02%
-2.25%
" LCD impacts in this category are greater than CRT impacts when phased out substances are removed from the
inventories (see Section 3.3.6).

       Figure 3-28 presents the LCD baseline and sensitivity analysis results for the energy use
impact category, the category with the greatest percent change from the baseline to the modified
LCD module energy scenario. Under this scenario, LCD energy use impacts in the
manufacturing stage increased almost 230% from 1,440 MJ per functional unit to 3,280 MJ per
functional unit. However, total life-cycle energy use impacts increased only 65%. This
sensitivity analysis did not affect consumption rates outside of the manufacturing stage.
                                        3-106

-------
                                                                   3.4 SENSITIVITY ANALYSES
                                     LCD energy use impacts
       3.50E+03 i
       3.00E+03 -
       2.50E+03 -
       2.00E403 -
       1.50E-K)3 -
       1.00E-f03 •
       5.00E-KJ2 -
       O.OOE+00
      -5.00E+02 -
                                              3.28E+03
         1.44E+03  8.53E402
6.33E402
             Baseline
                                                              Modified Module Energy

                   figure 3-28. Energy use: LCD baseline and modified module energy sensitivity
                                                 results
       Figure 3-29 shows the effects of the sensitivity analysis on LCD chronic occupational
health effects, another impact category with a relatively large percentage change.  As shown in
the figure, the manufacturing stage impact score in this category increased about ten percent,
from 684 tox-kg per monitor to 755 tox-kg per monitor, due to the increase in fuel inputs. The
chronic occupational health effect impacts were less sensitive than energy impacts because health
effects results are calculated using a scoring approach that considers the inherent toxicity of a
chemical instead of a simple loading approach (as is used for energy impacts). No toxicity data
were available for LNG, the input with the greatest change in quantity. Therefore, the LNG HV
is representative of a mean toxicity value.
                                 LCD chronic occupational effects
   ^  8.00E-K52 -,
   "3
   ^  6.00E4O2 -
   £  4.00E+02 -

   J  2.00E+02 -
   "Sb
   •¥  O.OOE-t-OO

   *" -2.00E-H32 J
         6.84E+02
                                                                 7.55E-f02
                   3.62E-03
              Baseline
                         -1.94E-HJO
                                         3.59E-0
                                             Modified Module energy
                                                               -1.94E-HHT
                 Figure 3-29. Chronic occupational effects: LCD baseline and modified module energy
                                              sensitivity results
                                            3-107

-------
3.4 SENSITIVITY ANALYSES
3.4.4   Modified LCD EOL Dispositions Scenario

       Finally, because very few desktop LCDs have reached their end of life, and usually only if
they have been damaged in some way, very little is known about the EOL disposition of LCDs.
In the baseline scenario it was assumed that a certain percent of EOL LCDs are incinerated,
recycled, remanufactured, landfilled as solid waste, and landfilled as hazardous waste. (See
Section 2.7.3 and Appendix I for an explanation of how EOL disposition percentages were
determined.) To address uncertainties in the allocation of disposition percentages, this sensitivity
analysis qualitatively evaluates a different set of final disposition numbers, as follows:

•      change percent recycled from 15% to 0%,
•      change percent remanufactured from 15% to 40%,
•      change percent landfilled (solid waste) from 50% to 40%.
•      do not change fraction incinerated (15%) or fraction sent to a hazardous waste landfill
       (5%).

       Thus, under the modified EOL disposition.scenario, recycling and solid waste landfilling
impacts would decrease, remanufacturing impacts would increase, and incineration and
hazardous waste landfilling impacts would not change. However, in attempts made to obtain
remanufacturing data, it was found that remanufacturing processes spanned a wide range of
activities, from as little as replacing button tops to as extensive as testing and replacing PWBs or
transformers.  Given the broad range of possibilities, and because few desktop LCDs have
reached their end of life, no single set of operations could be identified to adequately represent
remanufacturing activities that could be incorporated in our model. Remanufacturing data were,
therefore, excluded from the assessment.
       As shown in the baseline LCIA results (Section 3.3), LCD EOL dispositions have little
effect on overall life-cycle impacts under the baseline scenario. In fact, the only impact
categories in which an EOL process was a top contributor to  overall impacts were the hazardous
waste landfill use impact category, where the portion of a monitor landfilled contributed 97% of
impacts, and the solid waste landfill use category, where the portion of a monitor landfilled
contributed 3.5% of impacts.  As noted above, hazardous waste landfill use impacts would not
change under the modified LCD EOL dispositions scenario, but solid waste landfill impacts
would be expected to decrease slightly.  In a preliminary quantitative analysis of this scenario,
LCD life-cycle solid waste landfill impacts were found to decrease less than one percent, and
life-cycle impacts in other impact categories decreased less than 0.1%. Thus, the modified LCD
EOL dispositions scenario would have only a minor effect on LCD life-cycle impacts and would
not change comparative CRT and LCD results.

3.4.5   Summary of CRT and LCD Sensitivity Analysis Results

       The results of the sensitivity analyses are useful to manufacturers who want to understand
how uncertainty in the inventory affects impacts.  This information can be used to identify areas •
for additional study or potential improvement opportunities.  As discussed in Sections 3.4.1
through 3.4.2, it appears that CRT life-cycle impacts are highly sensitive to the glass energy data,
and less sensitive to the lifespan assumptions (lifespan assumptions greatly affect the magnitude
of CRT life-cycle impacts, but they do not greatly affect the distribution of impacts among life-
                                     —_                                         _

-------
                                                             3.4 SENSITIVITY ANALYSES
cycle stage). LCD impacts are less sensitive to the glass energy data and in fact are not greatly
affected by any of the sensitivity analysis scenarios, except tne longer lifespan under the
manufactured life scenario.
       Sensitivity results are also useful to interested members of the public who may be
evaluating the relative impacts of different monitor types and are interested in whether the CRT
or LCD has greater life-cycle impacts in any given impact category. Table 3-62 presents the
monitor type with greatest impacts by impact category and by scenario.  This information helps
us determine whether major assumptions (e.g., the monitor lifespan and LCD EOL distribution
assumptions) or uncertain data (e.g., glass energy data and LCD monitor manufacturing energy)
are driving results. As shown in the table, the modified glass energy scenario is the only scenario
that significantly changes the results from the baseline CRT and LCD comparative results.
Under this scenario, life-cycle impact results in seven categories reverse direction from the
baseline assessment, such that the LCD has greater impacts than the CRT. Therefore, under this
scenario, a total of nine out of 20 categories are greater for the LCD than the CRT, compared to
two out of 20 categories under the baseline scenario. The only other scenario that affects these
results is the manufactured life scenario, when impacts in the water eutrophication category are
greater for the CRT than the LCD.
                  Table 3-62. Summary of CRT and LCD LCIA results
Impact category
Renewable resource use
Nonrenewable resource use
Energy use
SW landfill use
HW landfill use
RW landfill use
Global warming
Ozone depletion
Photochemical smog
Acidification
Air particulates
Water eutrophication
Water quality, BOD
Water quality, TSS
Radioactivity
Chronic health effects,
occupational
Chronic health effects, public
Aesthetics (odor)
Monitor type with greatest impacts by scenario
Baseline
CRT
CRT
CRT
CRT
CRT
CRT
CRT
b
CRT
CRT
CRT
, LCD
CRT
CRT
CRT
CRT
CRT
CRT
Manu-
factured
life
CRT
CRT
CRT
CRT
CRT
CRT
CRT
b
CRT
CRT
CRT
CRT
CRT
CRT
CRT
CRT
CRT
CRT
Modified
glass
energy
CRT

CRT
CRT
CRT
CRT

b
?'rf*^?v%3J3&g>&!:Z%
CRT
CRT



CRT
''ffSffr'^r^W'
.".;ifcj#Je£?i^*¥;
CRT
^SiMPS
Modified
LCD
module
energy
CRT
CRT
CRT
CRT
CRT
CRT
CRT
b
CRT
CRT
CRT

CRT
CRT
CRT
CRT
CRT
CRT
Modifed
LCD EOL
distribution3
CRT
CRT
CRT
CRT
CRT
CRT
CRT
b
CRT
CRT
CRT

CRT
CRT
CRT
CRT
CRT
CRT
                                         3-109

-------
3.4 SENSITIVITY ANALYSES
                    Table 3-62.  Summary of CRT and LCD LCIA results
Impact category
Aquatic toxicity
Terrestrial toxicity
Monitor type with greatest impacts by scenario
Baseline
;:"Siiiill
CRT
Manu-
factured
life

CRT
Modified
glass
energy

CRT
Modified
LCD
module
energy
y.-g^-^s? •'•fpw^^-gi'i; are g
CRT
Modifed
LCDEOL
distribution"
i^espffspss;
i -laspi'tSjIpSst*
CRT
* Based on a qualitative evaluation, not quantitative results.
b CRT impacts are greater than LCD impacts in this category when all data are included in the inventories, including
data for substances that have been phased out. However, LCD impacts are greater than CRT impacts when phased
out substances are removed from the inventories (see Section 3.3.6).
                                            3-110

-------
                                                                         REFERENCES
                                    REFERENCES

 Bare, J.C., D.W. Penninton, and H.A. Udo de Haes.  1999. Life Cycle Impact Assessment
       Sophistication:  International Workshop. Int. J. LCA. 4(5): 299-306.

 Bare, J.  1999.  Slide Presentation: "Consistent Science for Environmental Decision-making."
       Tools for Sustainability Workshop, February 18-19. Accessed at
       http://www.epa.gov/ORD/NRMRiystd/mtb/PDONE/TFS6.PDF. Accessed August 8
       2001.

 Bamthouse, 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, FL.

 Bintein, S., J. Devillers, and W. Karcher. 1993. "Non-linear Dependence of Fish
       Bioconcentration on n-Octonol/Water Partition Coefficient." SAR QSAR Environ Res
       1:29-39.

 CAAA (Clean Air Act Amendments).  1990. Title VI (42 U.S. Code Annotated, Section 767la;
       and associated regulations: 40CFR82, App. F, Subpart A and App. I, Subpart A).

 Curran, M.A.  1996. Environmental Life Cycle Assessment. McGraw-Hill: New York.

 EPA (Environmental Protection Agency).  1992. Reference Guide to Odor Thresholds for
       Hazardous Air Pollutants listed in the Clean Air Act Amendments of 1990
       EPA/600/R-92/047. March.

 EPA. 1996.  Compilation of Air Pollutant Emission Factors (AP-42), Fifth Edition, Volume I.
       January.

 EPA. 1998.  National Air Quality and Emissions Trends Report, 1997. EPA/454/R-98-016.
       December.

 EPA. 1999.  Integrated Risk Information System (IRIS). Database available at
       http://www.epa.gov/ngispgm3/iris/sybst-fl.htm.

 EPA. 2001a. EPA Web site (a), Methyl Bromide Phase Out information available at
       http://www.epa.gov/docs/ozone/mbr/mbrqa.htmlSq3. Accessed July 17, 2001.

 EPA. 2001b. EPA Web site (b), HCFC phaseout schedule available at
       http://www.epa.gov/ozone/title6/phaseout/hcfc.html. Accessed July 25, 2001.

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) and SETAC Foundation for Environmental
  /    Education. Pensacola, FL.

-              :        -                3-111             '              ~~  ~~

-------
REFERENCES
Glazebrook, B. 2001. Electronic Communication with B. Glazebrook, Ecobilan and L. Kincaid,
      University of Tennessee, Center for Clean Products and Clean Technologies. July 9.

Glazebrook, B. (b). 2001. Electronic Communication with B. Glazebrook, Ecobilan and L.
      Kincaid, University of Tennessee, Center for Clean Products and Clean Technologies.
      July 9.

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.

Harris, J.C. 1981. "Rate of Hydrolysis." In W. J. Lyman, W. F. Reehl and D. H. Rosenblatt
      (Eds.). Research and Development of Methods for Estimating Physicochemical
      Properties in Organic Compounds of Environmental Concern.  Final Report, Phase IE.
      Part I, Chapter 7, pp.1-48. Arthur D. Little, Cambridge, MA.

Hauschild, M. and H. Wenzel. 1997. Environmental Assessment of Products, Volume 2:
      Scientific Backgrounds. London: Chapman and Hall.

Haywood, B. 1999. University of Tennessee Technical Memorandum for Development of the
      Dermal Hazard Value.

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 Assessment of Products.  Vol. I: Guide, and Vol. II:
      Backgrounds.  Leiden: CML Center for Environmental Studies, Leiden University.

Houghton, J. T. et al. (Eds.).  1996.  Climate Change 1995: The Science of Climate Change.
      Contribution of Working Group to the Second Assessment Report of the
      Intergovernmental Panel on Climate Change, Cambridge University Press, New York,
      NY.

Hunter, R. S. and F. D. Culver. 1992. MICROQSAR version 2.0. Institute for Biological  and
      Chemical Process Analysis, Montana State University, Bozeman, MT.

IARC (International Agency for Research on Cancer).  1998. IARC Monographs. Database on
      Risks to Humans available at http://www.iarc.fr.

JJSI (International Iron and Steel Institute). 2001. World Steel Trends and Statistics. Web site
       available at http://www.worldsteel.org/trends_indicators/figures_3.html.  Accessed July
       31.

International Life Sciences Institute (ELSI). 1996. Human Health Impact Assessment in Life
       Cycle Assessment: Analysis by an Expert Panel.  Washington, DC.
                                       3-112

-------
                                                                        REFERENCES
IPCC (Intergovernmental Panel on Climate Change). 1995. "IPCC Second Assessment
       Synthesis of Scientific-Technical Information Relevant to Interpreting Article 2 of the UN
       Framework Convention on Climate Change." Web site available at
       http://www.ipcc.ch/pub/sarsyn.htm. Accessed August 23, 2001. -

IUCLJD (International Uniform Chemical Information Database). 1996. European Chemical
       Bureau, European Commission-JRC, Environment Institute, Ispra, Italy, 1996.

Lindfors, L.G., K. Christiansen, L. Hoffman, Y. Virtanen, V. Juntilla, O-J Hanssen, A. Roenning,
       T. Ekvall. 1995. Nordic Guidelines on Life-Cycle Assessment.  Copenhagen: Nordic
       Council of Ministers, Report* Nord 1995:20.

Niemi, G.J., G.D. Veith, R.R. Regal, and D.D Vaishnav. 1987. "Structural Features Associated
       with Degradable and Persistent Chemicals."  Environ. Toxicol. Chem. 6:515-527.

Penman, J., D. Kruger, I. Galbally, T. Hiraishi, B. Nyenzi, S. Emmanul, L. Buendia, R. Hoppaus,
       T. Martinsen, J. Meijer, K. Miwa, and K. Tanabe (Eds). 2000. Good Practice Guidance
       and Uncertainty Management in Natural Greenhouse Gas Inventories.  IPCC National
       Greenhouse Gas Inventories Programme, Published for the IPCC by the Institute for
       Global Environmental Strategies, Japan.

Smrchek, J. EPA. 1999. Electronic communication with J. Smrchek, EPA, and Dipti Singh,
       EPA. March 29.

Swanson, M.B., G.A. Davis, L.E. Kincaid, T.W. Schultz, J.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.

Swanson, M.B., G.A. Davis, J.G. Overly, R. Dhingra, K. Kelly. 2001.  Life Cycle Design for the
       Automobile: Environmental Technology Draft Report.
                                        3-113

-------
this page intentionally left blank

-------
                              4. QUALITATIVE RISK SCREENING OF SELECTED MATERIALS
                                      Chapter 4

           QUALITATIVE RISK SCREENING OF SELECTED MATERIALS

       The scope of the DfE CDP, as presented in Chapter 1, was to first conduct an
environmental life-cycle assessment (LCA) to evaluate a generic 17" CRT and 15" LCD desktop
computer display, followed by a streamlined Cleaner Technologies Substitutes Assessment
(CTSA), which would target specific materials or processes that warrant further evaluation.
Traditionally, the DfE Program has conducted CTSAs that perform detailed risk
characterizations of alternative chemical processes. The streamlined CTSA for the CDP takes a
more detailed look than the LCA at the toxic effects of chemicals used in a process, without
conducting a complete risk characterization typical of past CTS As.
       In order to provide meaningful and timely results, the CDP Core group agreed to select a
few materials that are of interest to EPA and industry and conduct analyses concurrently with the
LCA, instead of conducting a CTSA on selected materials or processes after the LCA results
were presented. The LCA identifies material inputs and outputs and then characterizes them in a
life-cycle impact assessment (LCIA). In the human and environmental health effects impact
categories, these input and output amounts are used as surrogates for exposure. For the selected
materials, the additional CTS A^elated analyses are intended to better understand the potential
exposures to selected materials, during any processes that use thee materials, in order to try to
better understand potential chemical risks.
       The materials that were selected for further analysis were lead, mercury, and liquid
crystals.  The justification for choosing these materials, and a brief description of the scope are
provided below:

•      Lead: Lead is a top priority toxic material at the U.S. EPA. Lead is found in glass
       components of CRTs, as well as in electronics components (printed wiring boards and
       their components) of both CRTs and LCDs. The electronics industry is also concerned
       with lead and continues to take steps to reduce the use of lead in electronics products.
       Lead has been extensively studied for its toxic effects and has been addressed elsewhere
       with regard to CRTs and electronics (ATSDR, 1999).  Therefore, within the scope of the
       CDP, lead is recognized as a material of concern, but extensive evaluation beyond the
       LCA is not conducted.  Some discussion of the potential for exposure is included in this
       chapter, but it references existing studies for further information.
•      Mercury: Another top priority toxic material at the U.S. EPA is mercury. The
       fluorescent tubes that provide the source of light in the LCD contain mercury. Although
       very small amounts of mercury are found in the LCD backlights, EPA's concern with
       mercury and the potential for exposure during manufacturing and end-of-life processes
       are reasons why a more detailed analysis of mercury is warranted in the CDP. In
       addition, mercury is emitted from some fuel combustion processes, such as coal-fired
       electricity generation processes, and there is interest in the relative magnitude of mercury
       emissions from these sources as compared to the magnitude of mercury emissions from
       its intentional use in LCD backlights.  Another reason for including mercury is to begin to
       do an improvement assessment, because there appear to be potential alternatives to
       backlights with mercury.
                                          4-1

-------
4. QUALITATIVE RISK SCREENING OF SELECTED MATERIALS
•      Liquid crystals:  The toxicity of the liquid crystals in LCDs has been alluded to in the
       literature and there is a need to better understand the toxicity of these materials as well as
       provide the appropriate context of potential exposure and any associated risk. For
       example, during normal use of a display, no exposure would be expected.  Liquid crystals
       are generally organic materials in broad categories such as polycyclic aromatic
       hydrocarbons (e.g., phenylcyclohexanes, biphenyls) (EIAJ, 1996). By including liquid
       crystals in a more detailed analysis, this chapter attempts to better characterize any
       potential hazard and/or potential exposure of liquid crystals from the manufacturing, use,
       and disposal of LCD monitors.

       Choosing these three materials as a priority does not presume that these are the only
materials of importance and worthy of additional analyses. The results of the LCI and LCIA in
Chapters 2 and 3 provide more information on where to focus additional analysis efforts or
improvements.  Chapter 5 also identifies some potential improvement opportunities.
       Sections 4.1 through 4.3 present the qualitative risk screening of lead, mercury, and liquid
crystals, respectively. Subsections in each section briefly describe the following:

•      The use of the materials in computer displays:  These subsections describe how the
       materials are used in computer displays and give information on the mass used in
       particular applications.
•      Life-cycle inputs and outputs of the materials from computer displays: The life-cycle
       assessment approach not only focuses on the material contained within the product, but
       also emphasizes the environmental impacts of material inputs and outputs from every
       life-cycle stage. These subsections summarize the life-cycle inputs and outputs of the
       materials found in the CDP life-cycle inventories (LCIs).
•      Life-cycle impacts associated with the material inputs and outputs: As discussed in
       Chapter 3,  life-cycle impact assessment (LCIA) is a screening-level evaluation of
       potential impacts to any system (e.g., the environment) as a result of some action (e.g., a
       chemical release). In the LCIA, life-cycle inventory data were classified into various
       impact categories (e.g., greenhouse gases or ozone depletion) based on the characteristics
       of the inventory item. Characterization methods were then used to quantify the
       magnitude of the contribution the emission or consumption of the inventory item could
       have in producing the associated impact.  The result is expressed as an impact score
       which has been calculated using specific impact assessment tools. These subsections
       summarize the CDP LCIA results for lead, mercury, and liquid crystal inputs and outputs.
•      Potential exposures to the material including occupational, public, and ecological
       exposures: Toxic materials may pose a threat to human health anytime there is the
       potential for human exposure .throughout the life cycle of a computer display. Exposure
       occurs anytime a chemical or physical agent comes into contact with an organism, be it
       human or ecological.  The magnitude of exposure depends on the concentration of the
       chemical at the contact point, and the duration and frequency of the exposure. The
       concentration of the chemical at the contact point is influenced by several factors,
       including the type, quantity, and disposition (e.g., airborne, surface water) of the initial
       release, and the subsequent environmental fate of the chemical (the ultimate disposition
       of the chemical as it is transported through the environment). Note that exposure is often
       defined in terms of exposure pathways. An exposure pathway describes the route a
                                          _                      :

-------
	4. QUALITATIVE RISK SCREENING OF SELECTED MATERIALS

 contaminant travels from its source to an individual.;,A complete exposure pathway
 consists of the source and mechanism of release, transport medium, point of potential
 human contact, and the exposure route (e.g., inhalation, ingestion, dermal contact). These
 subsections focus on identifying potential exposures for three groups: workers in facilities
 using the chemicals (occupational exposures), the general public, who may be exposed to
 releases of the chemicals into the ambient environment, and ecological populations.
 While a quantitative exposure assessment is beyond the scope of this study, a qualitative
 discussion of potential pathways of exposure is presented for each of the groups listed
 above.
 Potential human health effects: Human health effects can include acute effects from
 short-term exposure, as well as chronic effects from repeated, long-term exposure. To be
 consistent with the scope of the LCIA, these subsections focus on chronic effects,
 including noncancer and cancer effects, unless no chronic toxicity data are available.
 U.S. environmental regulations for the material: These subsections briefly summarize
 U.S. environmental regulations that may affect facilities that manufacture materials for
 the computer display or are otherwise affected by the life cycle of computer displays (e.g.,
 disposal facilities). They do not summarize environmental regulations from other
 countries where displays or display components are manufactured, which may differ
 significantly from U.S. regulations.
 Alternatives to reduce the use of the material in computer displays: These subsections
 identify alternatives that can substitute for a material, or, in some cases, source reduction
 methods to reduce the use of a material.  The discussion of alternatives presented here is
 not a rigorous evaluation of their performance, cost, and environmental attributes, but
 rather a summary of the current knowledge base that may be useful for manufacturers
 seeking to identify improvement opportunities.

 In Section 4.4, summary information  and conclusions are presented.
                                    4-3

-------
4.1 LEAD
4.1    LEAD

       Lead and/or lead compounds are present in both CRTs and LCDs. Because lead and lead-
containing compounds have long been known to pose a threat to both human health and the
environment, this section presents a more detailed look at the use of lead in computer displays
and its effects on human health and the environment.

4.1.1   Lead in Computer Displays

       Lead is a significant material in current CRTs, accounting for up to 8% of the overall
composition of the CRT by weight (Menad, 1999), with a 17" monitor containing as much as
1.12 kg of lead (Monchamp et. al, 2001).  Lead is used in several parts of the CRT monitor,
including the funnel and neck glass, the sealing frit, as solder on printed wiring boards (PWBs)
within the monitor, and sometimes in the front panel glass of the CRT.  Lead is not as prevalent
in LCDs, only being found on PWBs.
       Lead, in the form of lead oxide, lines the inner surface of both the neck and funnel glass
of the CRT, or may in some cases be contained within the glass itself.  The lead oxide layer acts
as, a shield, protecting users from x-ray emissions given off by the electron gun. The lead oxide
layer can comprise as much as 28%-,by weight of the funnel (Lee et al., 2000) and 32% of the
neck (Menad, 1999).           ..,.,,                              .   -
       The sealing frit, which is used to make a vacuum-tight connection between the funnel and
the front panel, is comprised of as much as 80% lead oxide (Busio and Steigelmann, 2000). The
lead oxide is mixed with boric oxide and zinc oxide, along with several other compounds, into a
paste.  The glass frit paste is applied as a bead around the top of edge of the funnel and allowed
to dry. The front panel is then attached to the funnel and fired using a belt furnace for 30 to 60
minutes at typical temperatures of 450°C (Busio and Steigelmann, 2000). The lead, boric, and
zinc oxides devitrify to form large crystals which give strength to the frit seal (Techneglas, 2001).
Recent research has been directed at either reducing the lead content of the frit or reducing the
energy required to fuse the frit.
       Printed wiring boards found in both LCDs and CRTs primarily use a lead-based solder as
a surface finish and to attach electrical components to the circuit board. Solder is typically
comprised of 37-40% lead. Depending on the type of component, parts can be applied using a
solder paste which is  subsequently melted, or by passing the boards over a wave of molten
solder. Data collected for the life-cycle inventory indicates that a 17" monitor has approximately
51 grams of solder, or just under 19.8 grams of lead, while a 15" LCD panel contains nearly 22.4
grams of solder, or just about 8.5.grams of lead. Several lead-free solder alternatives are
currently being tested and are in limited use. Solder is the only significant source of lead in
LCDs.
       The CRT panel glass itself may also contain a small percentage of lead oxide, typically
ranging up to 4% by weight (Lee et al, 2000).  The lead acts as a stabilizer for the glass during
its formation, and also serves to keep the glass from browning.  Lead-free CRT panel glasses are
currently being produced successfully.
       A list of the lead-containing parts that make up a computer display, along with the
quantity of lead and percentage of lead in each part, is presented in Table 4-1.
                                          4-4

-------
                                                                              4.1 LEAD
                   Table 4-1. Computer display parts tnat contain lead
Part
Funnel
Front panel
Neck
Frit
PWBs (total)
PWBs (total)
Display type
CRT
CRT
CRT
CRT
CRT
LCD
Quantity •
(Kg)a
0.91
0.18
0.012
0.026
0.051
0.043
% Lead content of part
(by weight)
22-28% b-c
0-4".'
26-32 b-°
70-80 b'c'd
N/A
N/A
' Quantity of lead in a 17" monitor (Monchamp et. al, 2001).
"Menad, 1999.
c Lee et.aL, 2000.
dBusio and Steigelmann, 2000.
N/A= Not applicable
4.1.2   Life-Cycle Inventory Inputs and Outputs of Lead for Computer Displays

       Data on lead and lead-containing materials were collected and compiled as part of the
life-cycle inventory. Material inputs containing lead included primary materials (e.g., lead-based
solder) which end up as part of the product, as well as from ancillary materials (e.g., lead
consumed during the production of steel used to make CRT parts) which are consumed as part of
the manufacturing process or other supporting processes, such as energy production. The data
were aggregated by material from individual processes and are presented by life-cycle stage for
both CRTs and LCDs in Tables 4-2 and 4-3 below. More detailed material input data, which
include the processes for each input and the quantity of lead released, are included in
Appendix N.

                 Table 4-2. CRT lead-containing inputs by life-cycle stage
Life-cycle stage
Materials processing
Materials processing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Input
Lead (Pb, ore)
Lead (Pb, ore)
Frit
Lead
Printed wiring board (PWB)
Solder (63% tin; 37% lead)
Solder, unspecified - (CRT Assembly)
Quantity
6.50E-05
4.96E-01
6.67E-02
4.94E-01
8.47E-01
5.08E-02
2.67E-02
Units
kg
kg
kg
kg
kg
kg
kg
Type
Ancillary material
Primary material
Primary material
Primary material
Primary material
Primary material
Primary material
                                          4-5

-------
4.1 LEAD
                Table 4-3. LCD lead-containing inputs by life-cycle stage
Life-cycle stage
Materials processing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Input
Lead (Pb, ore)
Printed wiring board (PWB)
Solder (60% tin, 40% lead)
Solder (63% tin; 37% lead)
Solder, unspecified
Quantity
2.47E-05
3.74E-01
3.81E-02
2.24E-02
7.35E-05
Units
'kg
kg
kg
kg
kg
Type
Ancillary material
Primary material
Primary material
Primary material
Ancillary material
       Material inputs can be raw materials such as lead ore, or output materials from a previous
process or life-cycle stage. For example, small quantities of lead extracted from lead ore are
sometimes used as additives in the production of several materials including ferrite, steel, and
invar. Once extracted, lead is an input material to the manufacturing processes of several CRT
components including CRT glass manufacturing and the manufacturing of the sealing frit paste,
which then is an input for the CRT tube manufacturing process. Similarly, lead is used to
produce solder which is then used to produce PWBs used in LCDs and CRTs.
       Releases of lead and lead-based materials into the environment occur throughout the
entire life cycle of the computer display. Environmental releases include airborne, waterborne,
solid waste,  and radioactive emissions of. lead isotopes associated with nuclear fuel reprocessing.
Similar to the inputs, emissions data were,aggregated by the material released from individual
processes and then reported by life-cycle_stage. The lead or lead-based material released, the
quantity of the release, the type of release (e.g., waterborne), and the ultimate disposition of the
release all contribute to the environmental impacts.
       The life-cycle outputs containing lead for both CRTs and LCDs are shown in Tables 4-4
and 4-5, respectively. More detailed data on lead and lead-based outputs for each process are
presented in Appendix N.
                                          4-6

-------
                                    4.1 LEAD

Life-cycle
stage
Materials
processing
Materials
processing
Materials
processing
Materials
processing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Use
End-of-life
End-of-life
End-of-life
Table 4-4. Life-cycle lead outputs to the environment from CRTs
Output
Lead
Lead
Lead compounds
Lead-210 (isotope)
Broken CRT glass
Broken CRT glass
Cinders from CRT glass mfg (70% PbO)
CRT glass faceplate EP dust (Pb) (D008
waste)
CRT glass funnel EP dust (Pb) (D008
waste)
Frit
Hazardous sludge (Pb) (D008)
Lead
Lead
Lead
Lead (Pb, ore)
Lead compounds
Lead compounds
Lead contaminated grit (D008 waste)
Lead debris (D008 waste)
Lead sulfate cake
Printed wiring board (PWB)
PWB-Solder dross
Sludge from CRT glass mfg (!%PbO)
Waste batch (Ba, Pb) (D008 waste)
Waste finishing sludge (Pb) (D008 waste)
Lead
Lead
Lead compounds
Printed wiring board (PWB)
Quantity
1.66E-03
2.29E-08
1.59E-05
1.02E+00
1.88E-03
1.08E+00
8.26E-03
1.03E-03
5.01E-03
2.99E-03
1.52E-03
1.03E-06
1.30E-05
4.64E-05
4.41E-07
1.62E-05
1.17E-09
3.46E-05
2.14E-04
2.67E-05
3.70E-02
6.70E-02
8.77E-04
1.41E-03
2.56E-04
1.27E-05
1.42E-05
1.60E-09
1.46E-01
Units
kg
kg
kg
Bq
kg
kg
kg
kg
kg
kg
kg
kg
kg
kg
kg
kp
kg
kg
kg
kg
kg
kg
kg
kg
kg
kg
kg
kg
kg
Type
Airborne
Solid waste
Waterborne
Radioactivity

Solid waste
Hazardous waste
Hazardous waste
Hazardous waste
Hazardous waste
Hazardous waste
Waterborne

Waterborne

Waterborne

Hazardous waste


Solid waste




Airborne
Airborne

Hazardous waste
R/R = recycling/reuse
Disposition
Air
Landfill
Surface water
Air
Landfill
R/R
Landfill
Landfill
R/R
Landfill
Landfill
Treatment
Air
Surface water
Air
Treatment
Surface water
Landfill
Landfill
Landfill
R/R
R/R
Landfill
Landfill
Landfill
Air
Air
Surface water
R/R

4-7

-------
4.1 LEAD
            Table 4-5. Life-cycle lead outputs to the environment from LCDs
Life-cycle stage
Materials processing
Materials processing
Materials processing
Materials processing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Use
End-of-life
End-of-life
Outputs
Lead
Lead
Lead compounds
Lead-210 (isotope)
Lead
Lead
Lead (Pb, ore)
Lead compounds
Lead compounds
Printed wiring board (PWB)
PWB-Lead contaminated waste oil
PWB-Solder dross
Waste batch (Ba, Pb) (D008 waste) .
Waste CCFL, with lead
Lead
Lead
Lead compounds
Quantity
3.13E-06
5.42E-09
3.68E-06
3.21E-01
8.84E-06
8.33E-07
1.48E-06
5.67E-11
7.14E-06
7.50E-03
5.14E-03
2.96E-02
6.55E-05
8.17E-08
4.76E-06
4.76E-06
4.98E-10
Units
fy
kg
kg
Bq
kg
kg
.kg
kg
kg
kg
kg
kg
k£
k?
kg
kg
kg
Type
Airborne
Solid waste
Waterbome
Radioactivity
Airborne
Waterbome
Airborne
Waterbome
Waterbome
Solid waste
Hazardous waste
Hazardous waste
Hazardous waste
Hazardous waste
Airborne
Airborne
Waterbome
Disposition
Air
Landfill
Surface water
Air
Air
Treatment
Air
Surface water
Treatment
Landfill
Treatment
Recycling/
reuse
Landfill
Treatment
Air
Air
Surface water
4.1.3  Computer Display Life-Cycle Impacts for Lead

       The life-cycle impacts of lead, lead compounds, and materials containing lead (e.g., lead-
based solder on printed wiring boards) calculated for CRTs and LCDs during the LCIA are
summarized in Tables 4-6 and 4-7 respectively. Impact scores in the table are expressed in units
specific to each impact category (see Chapter 3.1 for a discussion of impact category units and
weighting). The total impact score for each category resulting from lead and lead-based
materials is presented at the bottom of each table.
                                          4-8

-------
                                    4.1 LEAD








£
&
u
S
-•
o
|si| 1
U-s'S & §
>>
111
« «

09 3
.|i 1
si
||| ^
J3
, 1 s
g ^ a 'Sb
*-|B ^



Material

0>
Si. a>
5s M
CM co


en
vo

0


O



en

0



0

o


o





o

o


o


0


o



0

o

*— 1
o
VJ
ON



I
1
1
in

»— 4
en

O


9
en

O



o

0


o




Lead compounds
o

o


0


o

o
s
1— <


o

0


o




o
1-H
i

ra 60
•a -S

i2 o


0

0


o


o


o



o

.22E-07
VO

o




Broken CRT glass
o

o
-—

o


0


o



o •

.88E-06
VO

0

o
^"^

Cinders from CRT glass mfg
PbO)
o

o


o


o


o



o

1
CN

O

2"
&

CRT glass faceplate EP dust
(D008 waste)
0

0


o


o


0



0

VO
I
en

O




1
o

o


o


0


o



0

.38E-06
1~l

o


^^
no
Hazardous sludge (Pb) (DOO
.94E-05 1
»O
*O
S
en
ON
»— t
O
00
oo
ON

1
^

O



o

o

-H

3




o\
9
g
1-1
oc
§
en
«S

0


9
£^
e>>
oi

O



o

o


0




Lead compounds
0

o


o


o


0'



o

i
cs

o



oo
Lead contaminated grit (DOC
waste)
o

o


o


0


o



0

oo
o
oo
"""

0




Lead debris (D008 waste)
o

o


0


o


o



o

oo
en
O
en

0




Lead sulfate cake
0

0


o


o


0



o

f-
§
VO

o

tg

J""V
Sludge from CRT glass mfg
PbO)
o

o


o


0


o



0

s
g
ex)
T— *

O




Waste Batch (Ba, Pb) (D008
waste)
bO
0

0


o


o


o



o

1
en
04

0

oo
s
g
H.
Waste finishing sludge (Pb) i
waste)

.§
3 • -
&
3

1
V
o
'—<
o


O


9
w
in
oi

o



0

o


o




•a



o
I—1


9
*~H
o


0


in
oq

O



O

o


o




•3

1

oo
O
en

0


1
en

O



o

o


o




Lead compounds

o
H3
=g
•A -
c
W •
«
1-1
^
rf
o
at
00
o\

^
tr
(—1
•f
*


0

c:
i

^i
i
oe



:ores BY Category
cc
C.)
A
g
^2

O
4-9

-------
4.1 LEAD








w
§
Js
c3
O
u
&2

CJ
«

.g
-B
1
O
trt
i
a
CO
tj*
U
CQ
H






















1
o
u
in
u
a
£














!,&> -a
tn *3 "T
ll 1
V *W ^^^
H
if?
11 1
-
"S! ^
Q ^
* S3 <^
Frt K3 S
!S *o ^^
ll

1 « 1
"2 1§ g ^
J?- ^3 s""*/
1
s s
1 1 1 If
*• B S '""'
gJ IM



|3
S
1






o
g> §>
4 "
nj


VO
O
t— i
CO
o



o
I 6.26E-06
0





0




o



o











•8

0

0



o
o
o





0




o


V,
I
•*
(S





I
o
rC
Si
1
S
w
00
VO
m
9
H
01



o
| 7.36E-06 |
o





o




o



o






CO
•O
1
o.
a
8
•a
3

0

o



o
o
1
C5
c«i



0




O



O






1
1
_eo
s^x
0
cs
T-)


" S
1 s
>§ °
« D.


o
i
00
in
1
r-4


O
i
o





0




o



o











•3
^-H
*— 4
W
VO
in
o\
1
i-H


O
I 1.13E-10 |
o





0




o



o






CO
1
O

g
o
o
1

0

o



o
o
o



vo
O
i

o\


o



o



X— \
1
T3
O
_c
t-4
>

c
s

o

0



0
0
o





o



9
^
VO

g
O
u
•a
,2


id-of Life
fl


0
i
r*
IT
O1
CO


0
1 5.03E-05 1
i—i
o
i
^•4

W

vps
<^
CO

e\

ON
r«.
VO
i/5
V)
c»
Tt
^





i
*•
n
^
pa
8
'CO
1
3
o
H
                                        4-10

-------
                                                                                4.1 LEAD
        Impact scores for some lead-based inputs and outputs shown in Tables 4-2 though 4-5
 were not calculated if the type and disposition of the input of release was not expected to.
 contribute to any of the impact categories. For example, a waterborne release of lead with a
 disposition going to treatment assumed that lead was not yet released to the environment where
 impacts could occur, and therefore no impacts were calculated. However, since inventory data
 for subsequent disposal processes could not be obtained, it was assumed the lead (or other
 inventory item) had been removed to a level such that the subsequent release of treated
 wastewater would not contribute significantly to the impact.  Similarly, impact scores were not
 calculated for releases going to recycling/reuse or for product outputs.
 u    u    ^  Lead-based impacts from the CRT ranged from moderately to significantly greater
 than those from the LCD in every category, with the exception of solid waste landfill use  The
 most significant difference was in non-renewable resource consumption, where the CRT (989
 grams) consumed over 40 thousand times the mass of non-renewable resources over the course
 of its life cycle than those consumed by the LCD (0.025 grams). Hazardous waste landfill use is
 another significant difference, with lead-based life-cycle outputs from  CRTs using over 2 600
 times the space of the lead-based outputs from LCDs. However, the absolute volume of waste
 from the CRT is still a relatively small volume (1.50 cm3). Other categories where CRTs had
 notably greater impacts as a result of lead include the chronic public health effects and terrestrial
 toxicity impact categories.
        Based on the CDP LCIA methodology, chronic occupational health effect impacts were
 only calculated for lead inputs (excluding lead ore) to processes in the computer display life
 cycle. Only the manufacturing life-cycle stage had lead inputs from which impacts were
 calculated as shown in Table 4-6.  The overall impact scores (0.988 tox-kg for CRT, none for
 LCD) likely underestimate the chronic occupational impacts for lead because they do not
 consider chronic occupational impacts from other processes such as the mining, smelting, and
 refining of the lead, which are known to pose potential occupational exposures (see Section
 4.1.4). For a more detailed discussion of how chronic occupational health effect impacts were
 calculated, refer to Section 3.1.2.12.
       The contribution  of lead-based impacts for each computer display technology to the
 overall impacts for each individual impact category is shown in Table 4-8.  Values in the table
 are expressed in the percent contribution the material made to the overall impact score for all
materials (e.g., mercury, fuel oil, glass) for each category. The percent  contributions give an
indication of the importance of lead^based impacts relative to the life-cycle impacts from other
materials or outputs from the computer display.
                                         4-11

-------
4.1 LEAD
         Table 4-8. Summary of percent contributions from lead-based materials
Impact category
Non-renewable resource
Hazardous waste landfill use
Solid waste landfill use
Radioactivity
Chronic health effects- public
Chronic health effects- occupational
Aquatic toxicity
Terrestrial toxicity 	 	
CRT
1.48E-01 %
8.99E-02 %
NA
2.70E-08 %
1.80E-04 %
1.10E-01 %
1.96E-01 %
8.90E-05 %
LCD
6.78E-06 %
1.57E-04 %
1.73E-02 %
2.63E-06 %
5.58E-06 %
N/A
1.71E-03 %
2.82E-06 %
N/A= Not applicable

       It can be seen from Table 4-8 that the contributions of lead-based impacts are not
significant relative to the total impacts from other materials (e.g., glass, copper wire, electronic
components) in each category. Impacts from lead-based CRT outputs in the categories of
nonrenewable resources, aquatic toxicity, and chronic public health effects are all range from 0.1-
0.2% of the overall impact scores in each category.

4.1.4  Exposure Summary

       Lead may pose a threat to human health anytime there is the potential for human exposure
to the lead throughout the life cycle of a computer display. Exposure occurs anytime a chemical
or physical agent, in this case lead or lead compounds, comes into contact with an organism, be it
human or ecological.  This section qualitatively identifies potential exposures for three groups:
 occupational workers in facilities using lead (occupational exposures), the general population
living nearby these facilities which may be exposed to lead releases into the ambient
 environment, and ecological populations in the area surrounding a facility.

 4.1.4.1 Occupational exposures

        Workers are typically exposed to far greater concentrations of chemicals for longer
 periods of time than other populations.  Worker exposures to lead can be especially serious given
 the overall toxicity of lead and lead compounds. As a result, both employers and government
 agencies have adopted recommendations or requirements for employers who wish to limit worker
 exposures.
        Occupational exposures can occur anytime a worker comes into contact with lead,
 whether it be through dermal (skin) contact with a part containing lead (e.g., lead oxide coating
 on glass funnel), through the inhalation of lead particulates dispersed into the air, or through the
 inadvertent ingestion of lead. Many of the primary and support processes required to
 manufacture computer displays have lead in the workplace, and correspondingly, the potential for
 worker exposure. The processes associated with lead inputs and outputs throughout the
 computer displays' life cycles are presented in Tables N-l through N-4 in Appendix N.  It is
 important to note that while this list gives an indication of where likely lead exposures may
 occur, it is not exhaustive. Many processes and subprocesses may be contained within a process
 listed, each of which may pose its own potential for occupational lead exposures.
                                           4-12

-------
 	—	—	4.1 LEAD

        Exposures to lead are more likely^o occur during;the extraction, manufacturing, and
 disposal life-cycle stages of a computer display. During the use of the computer display,
 potential exposures to lead are unlikely as the components containing the lead are contained
 within the outside shell of the computer display, limiting the opportunity-for contact with
 consumers. Table 4-9 presents some typical pathways leading to the occupational exposure of
 workers to lead over the life cycle of a computer display.

           Table 4-9. Potential occupational exposure pathways for lead over the
Exposure route
Inhalation
Dermal
Ingestion
Transport
media
Air"
Air
Air
Direct contact
Direct contact
Air
Example mechanisms of exposure
Lead fumes resulting from the vaporization of lead during smelting
Lead oxide dust released to the air during lead frit manufacturing
Lead aerosols created during the aeration of tin/lead solder plating baths
during PWB production
Handling of leaded CRT glass funnels prior to assembly
Consumption of food eaten with lead-contaminated hands (or drinking,
smoking, etc)
Ingestion of lead contaminated soil particles which become airborne
during lead mining
       Workers may be exposed to airborne lead concentrations through the release of lead dust
fumes, or aerosols into the workplace.  The lead is transported by the air, where it is inhaled into'
the lungs and then absorbed into the bloodstream. The greatest potential for high-level
occupational exposure is during lead smelting and refining, where lead is vaporized during high
temperature heating resulting in the release of lead fumes and small respirable particles of lead
(EPA, 1986). Lead concentrations in air at three primary lead smelters were found to range from
80-2,900 //g/m3, peaking at a level 58 times the OSHA recommended guidance level of 50 //g/m3
(HSDB, 2001). Another study found that during the smelting and refining of lead, mean
concentrations of lead in air reached as high as 4,470 //g/m3, nearly 90 times the OSHA guidance
level (Fu and Bofetta, 1995). Exposures to lead dust may also occur during lead mining, frit
manufacturing, CRT glass manufacturing, or processes in which metallic lead is heated in the
presence of air. Exposures to lead fumes are only possible during high temperature operations
(above 500°C), such as welding or spray coating of metals with molten lead (Sittig, 1985).
       Dermal exposures can take place anytime lead or materials containing lead are physically
handled by workers. Opportunities for dermal exposures to lead are numerous in  processes
throughout the computer display life-cycle, as many processes involve lead or parts containing
lead, especially in CRT manufacturing.  Lead can be transferred to the skin of workers through
contact with lead-containing materials and parts. Dermal exposures may also occur during
cleaning and maintenance of equipment used to smelt, refine, or apply lead in a molten state
(e.g., solder wave machinery for PWBs) or in areas with large airborne lead concentrations that
may settle out onto work surfaces directly contacted by workers.
                                          4-13

-------
4.1 LEAD		

       The contribution of dermal exposures to the overall lead body burden is uncertain.  It is
believed that most forms of lead are unable to readily penetrate the skin, allowing only a small
amount of lead to enter the bloodstream. (ATSDR, 1999). Alkyl lead compounds, which are the
known exception, are primarily used as additives in gasoline and are notTised directly in
computer display manufacturing (Bress andBidanset, 1991; ATSDR, 1999). Therefore, dermal
exposures to inorganic lead compounds are not expected to be as significant as the inhalation or
ingestion routes of exposure (EPA, 1986).
       Along with inhalation, ingestion of lead-bearing dust and fumes is a major route of
exposure in lead smelting and refining industries (EPA, 1986). Airborne dust particles of lead
can eventually settle onto skin, equipment, clothing, and work surfaces, where they may be
subsequently transferred to the mouth and become ingested. Airborne particles may also be
inhaled and swallowed, directly when greater than 5 micrometers in size (ATSDR, 1999).  Once
ingested, the amount of lead that reaches the bloodstream through the stomach depends on a
number of factors, such as the age of the subject, length of time since last meal, and how well the
lead was able to dissolve in the stomach.  Studies have found that roughly 6% of the lead
ingested will absorb into the blood stream of an adult who has recently eaten (within the last
day), while upwards of 60-80% was absorbed in adults who had not recently eaten (ATSDR,
1999).
       Lead exposures of workers are frequently measured by biological testing (e.g., blood lead
levels, urinary lead levels) rather than monitoring the workplace for lead concentrations, making
occupational data on lead exposures often not readily available (EPA, 1986).  For a discussion of
blood lead levels, corresponding effects, and recommended exposure guidelines, refer to Section
4.1.5 of this chapter.
        Blood-lead levels have been reported in studies of workers for several industries relevant
to computer display manufacturing. For example, workers occupationally exposed to lead during
glass production were tested to determine their blood lead levels. Workers were divided into
groups based on work activities and blood samples were collected at the end of each shift.
Concentrations of lead in the blood ranged from 70 to 680 //g/1, with median values ranging from
 170 to 340 Aig/m3, depending on the worker group. Data on types and rates of exposure were not
identified (Ludersdorf et al, 1987). Another study found that workers producing ceramic coated
capacitors and resistors using leaded glass were exposed to occupational lead levels ranging from
61 to 1,700 ytig/m3. Blood lead levels ranged from 16 to 135 jug/dL in these same workers,
greatly exceeding the OSHA recommended level of 50 /zg/m3 (Kaye et al, 1987).
        The presence of lead in the workplace does not mean that occupational exposures are
 unavoidable.  Worker exposures to lead can be reduced or even eliminated through the use of
 personal protective equipment, sound operating practices, or through advanced machinery that
 protects workers from exposure (e.g., an enclosed and vented wave solder machine). To
 determine actual worker exposures to lead, a complete exposure assessment specific to each
 manufacturing process would be required.

 4.1.4.2 General population

        The general population living nearby a manufacturing facility using lead may potentially
 be exposed to lead emissions from the facility into the surrounding ambient environment. The
 likelihood and quantity of the potential exposure is dependent on the type and quantity of release,
 the receiving media, the local environmental conditions, and the fate and transport characteristics
                                           —-

-------
 	:	_		   4.1 LEAD

 of the release.  General population exposure,|o lead is most h'kely to occur through ingestion of
 lead contaminated food, water, and soil, as well  as through inhalation of lead particulates in the
 ambient air (EPA, 1986).
        Lead released into the ambient air will typically be in the form of lead particulate matter,
 which is eventually removed from ambient air through washout by precipitation (rain or snow) or
 through gravitational settling.  Estimates indicate that the majority of lead released into the
 environment is dispersed into the atmosphere (EPA, 1980). With a relatively small mass mean
 diameter of 0.55 ^m (HSDB, 2001), lead-containing particles can stay aloft for up to 64 hours
 and travel 1600 km, though they are more likely to be deposited within 10 km of the emission
 source (HSDB, 2001).  General populations living near a source of lead emissions may encounter
 the lead while it is still airborne,  leading to potential inhalation exposure.  The  direct inhalation
 of lead accounts for only a small part of the overall lead exposure to nearby populations,
 although the reentrainment of lead-contaminated soil is a common route of exposure (ATSDR
 1999).
        Ingestion of lead is the most significant  route of exposure for general populations
 (ATSDR, 1999).  Particulates removed from the  air are deposited into the soil,  surface water, and
 onto local vegetation, where they may be ingested by nearby residents. Grains, vegetables, and
 fruits grown in close proximity to a source of lead emissions may contain lead which has been
 absorbed from contaminated soil through the root system. Lead also has the ability to
 bioaccumulate in  the soft tissues of fish and wildlife, which are then consumed by sportsmen and
 their families.
       Incidental ingestion of soil, which may occur while eating or smoking with soil-coated
 hands or when soil becomes reentrained and swallowed directly, often results in the largest lead
 exposures to residents living near emission sources. Lead-contaminated soil can also enter the
 home by being tracked into the house or carried home from the workplace on clothing, where it
 can come into contact with eating surfaces or food and become ingested. A study measuring lead
 in the home found mean lead levels as high as 22,191 ^g/g in homes located within 1.6 km of a
 lead smelting facility, and mean levels of 2,687 //g/g in homes of workers at the smelting facility,
 irrespective of distance from the plant (ATSDR,  1999). One study found that once lead is
 swallowed, up to 50% of the lead is released from the contaminated  soil into the stomach after
 only 10 minutes (HSDB, 2001).
       Lead may also be released directly to surface water or indirectly to groundwater through
the leaching of lead from landfills. Life-cycle inventory releases to surface water include 464
grams of lead and 159 grams of lead compounds  per CRT.  Surface water may also become
contaminated through soil deposition or through surface water run-off from contaminated soil.
Groundwater lead contamination from the leaching of lead-contaminated debris from solid- or
hazardous waste disposal sites is unlikely to be significant due to the relative insolubility of lead
(HSDB, 2001).  Lead released to both surface waters and groundwater will typically remain
insoluble, forming precipitates and settling into the sediment of the lake or stream. However,
very little lead is typically found in U.S. waters which are used to supply the public with drinking
water, due to strict governmental regulations (0.005 ppm lead) (ATSDR, 1999).
                                          4-15

-------
4.1 LEAD
4.1.4.3 Ecological populations

       Inorganic lead typically does not pose a significant health threat to fish and wildlife
populations, except at extremely high concentrations.  Once introduced into surface waters, the
levels of soluble lead depend on the pH of the water and the dissolved salt content.  At neutral
pH, inorganic lead typically does not remain soluble in water at high concentrations, forming a
precipitate that ultimately deposits in the sediment. However, as the alkalinity and pH decrease,
the relative soluble concentrations of lead may become higher.
       Toxic substances such as lead are capable of concentrating in the tissues of fish and
wildlife. The bioconcentration of lead in fish is low-to-moderate in most species, with a
bioconcentration factor (BCF) of 42 an 45 being reported for two fresh water fish species1.
However, BCFs for certain other species, such as blue mussles (4,985), eastern oysters (1,000+)
and 4 types of fresh water invertebrate species (range of 499 to 1,700), were much higher (EPA,
1999).

4.1.5  Human Health Effects

       Lead has been classified by EPA as a persistent biaccumulative toxic (PBT) chemical
(EPA, 2001b).  PBT pollutants are highly toxic, long-lasting substances that can build up in the
food chain to levels that are harmful to human and ecosystem health. Lead's ability to persist in
the environment without breaking down, along with its tendency to bioaccumulate, poses adverse
health effects to birds and mammals at the top of the food chain, along with anyone who
consumes them for food.  Lead and lead-based compounds have been associated with a range of
adverse human health effects, including effects on the nervous system, reproductive and
developmental problems, and cancer.

4.1.5.1 Chronic effects (noncancer)

       Lead is toxic to human health regardless of the form (Gosselin, 1984). It is one of the
most hazardous of the toxic compounds because the dose of lead is cumulative over a lifetime,
and the health effects are many and severe.  Lead has been known to cause hematological,
gastrointestinal, and neurological dysfunction in adults and children. Chronic exposures have
also caused hypertension and reproductive impairment in both men and women, as well as
slowed development in children (Sittig, 1985).
       Adverse effects, other than cancer or mutations, are generally assumed to have a dose or
exposure threshold. A reference dose (RfD) is an estimate of the daily exposure through
ingestion to the human population that is likely to be without an appreciable risk of noncancer
detrimental effects during a lifetime. Likewise, a reference concentration (RfC) represents an
estimate of the daily inhalation exposure to the human population that is likely to be without an
appreciable risk of noncancer detrimental effects during a lifetime.
       Because of the relative toxicity of lead and the cumulative nature of lead doses, a safe
level of human exposure has yet to be  identified by researchers, preventing EPA from
        1 Bioconcentration is defined by EPA as the non-dietary accumulation of chemicals in aquatic organisms
 (U.S. EPA, 1999).

-------
 		     4.1 LEAD

 establishing a RfD or RfC for inorganic lead (ATSDR, 1999),  Instead, lead exposure is
 determined by using exposure biokinetic models that relate exposure levels to an estimated blood
 lead level, which is then compared to actual blood lead levels where adverse effects are known to
 occur2.  For example, increased blood pressure has been observed in adults with a blood-lead
 level as low as 7 ^g/dL (ACGIH, 1991). Lead concentrations in excess of 60 //g/100g blood
 have been associated with neuropathy, gastrointestinal disturbances, and anemia, while workers
 with blood-lead levels between 50-70 ^g/100 g to have shown  decreased neural response
 (ACGIH, 1991).
        As a guideline, a blood-lead level of concern for adult workers of 30 //g/dL has been  *
 established by both the Occupational Safety and Health Administration (OSHA) and ACGIH. A
 guideline of 10 ,ug/m3 (for a child) has been set by the Center for Disease Control (CDC) for
 general population exposures to lead in the ambient environment.  A summary of human health
 effect guidelines for lead is presented in Table 4-10.
Table 4-10. Human health effect reflations and guidelines for lead
Type
Agency/Category
Workplace exposures to lead
Worker blood-lead
target/action levels
Pregnant worker:
fetal blood-lead
target/action levels
Workplace air
exposure limit
OSHA, Adults who "wish to bear children"
OSHA, Blood-lead level of concern
OSHA, Medical removal
ACGIH, Biological Exposure Index (BEI) Blood-lead level of
concern (ACGIH, 1998)
NIOSH, level to be maintained through air concentrations
OSHA
CDC
OSHA Permissible exposure limit (PEL)
NIOSH Recommended exposure limit (EEL) (NIOSH, 1997)
ACGIH TLV TWA (ACGIH, 1998)
Ambient environment exposures to lead
Blood-lead
target/action levels for
child
CDC
OSHA
World Health Organization blood-lead level of concern
Regulatory level

30/ig/dL
40#g/dL
50ng/dL
30/ig/dL
60 ftg/100 g
30/ig/lOOg
lO/ig/dL8
50pg/nf
100 fig/m3
SOflg/m3

lOfte/m3"
30 /ig/100 g
20ug/dL
         	— —~ • " —*" w»r i M.VVU iw • v* wj. A^UV* ii uiw cu.iAvu.UL \JJL ICOU. Ill U1C LflAJUU lo ill ICooL L\J U.g/QI -.,
Medical evaluation and environmental remediation should be done for all children with blood-lead levels greater
than 20 pg/dL. Medical treatment may be necessary for children with a blood-lead concentration above 45 ug/dL
(RTI, 1999).
Notes:  ACGIH: American Conference of Governmental Industrial Hygienists; NIOSH: National Institute for
Occupational Safety and Health; TWA: Time weighted average; TLV: Threshold limit value
         In order to estimate blood-lead levels, worker exposure levels based on releases reported in the inventory
would be required. However, without information pertaining to the exposure conditions (which is unavailable to
this study) and fate and transport of the releases, worker exposure cannot be calculated.
                                            4-17

-------
4.1 LEAD
4.1.5.2 Carcinogenicity

       The potential for a chemical to cause cancer is evaluated by weight-of-evidence
classifications, specific to the rating organization, which are typically determined by laboratory
or epidemiological studies.  Lead and inorganic lead-based compounds have been classified by
the International Agency for Research on Cancer (IARC) as possible human carcinogens (Group
2B), based on sufficient evidence of carcinogenicity in animals (IARC, 1987). Lead has also
been classified as an A3 carcinogen (confirmed animal carcinogen with unknown relevance to
humans) by the American Conference of Governmental Industrial Hygienists (ACGIH, 1998).
The U.S. EPA has given lead a weight-of-evidence classification of B2, indicating lead is a
probable human carcinogen and a confirmed animal carcinogen (IRIS, 1999).  There is currently
no established cancer slope factor for lead, which could be used to estimate cancer risk from an
exposure amount.

4.1.6  Environmental Regulations for Lead

       Apart from the regulations and recommendations regarding worker safety presented in the
previous section, lead is regulated in a number of ways. This section presents a brief summary of
the U.S. regulations for lead and lead compounds expected to impact facilities that manufacture
materials for the computer display. It should be noted that many of the parts and materials which
go into the manufacture of computer displays are manufactured in countries outside the U.S.,
with their own lead regulations which may differ significantly from those discussed below.
       Air emissions of lead are regulated under the Clean Air Act (CAA) of 1970 and the
amendments to the CAA of 1977 and 1990.  Under the CAA, lead is regulated as a hazardous air
pollutant (HAP), which is by definition a chemical that is generally known or suspected to cause
serious health problems. Stationary source categories involved in the life cycle of a computer
display that must meet new source performance standards include primary and secondary lead
smelters, glass manufacturing plants, and metallic mineral processing plants (EPA, 1977; EPA,
1980a; ATSDR, 1999). A National Ambient Air Quality Standard (NAAQS) was also
established for lead, requiring that the concentration of lead in air that the public breathes be no
higher than 1.5 fj,g/m3  averaged over 3 months [40 CFR 50.12].
       Lead releases to surface water are regulated under the Clean Water and Effluent
Guidelines and Standards promulgated under the Clean Water Act of 1977.  Lead is identified as
a priority pollutant [40 CFR 401.15], requiring the limitation of lead concentrations in pollutant
discharges from point sources.  The regulations also set standards of performance for new point
sources, as well as pretreatment standards for both new and established sources. Regulated point
source categories include lead smelters, steam electric power generation, glass manufacturers,
and aluminum production and others, all of which contribute to the life-cycle impacts of a
computer display. New point sources of lead contamination must also apply for National
Pollution Discharge Elimination System (NPDES) permits which will establish effluent limits for
sources of lead discharge.
        To protect the  population from a contaminated water supply, toxic substances in drinking
water are regulated under the Safe Drinking Water Act of 1986. A federal drinking water
standard of  15 ug/L has been established for lead.
                                          4-18

-------
 	—	:	.	        4.1 LEAD

        EPA also regulates lead content in hazardous and solid wastes under the Resource
 Conservation and Recovery Act (RCRAJ: A solid wasteroriiaining lead or lead compounds may
 be considered a D008 characteristic hazardous waste if, when subjected to a Toxicity
 Characteristic Leachate Procedure (TCLP) test, the extract exceeds 5.0 mg/L [40 CFR 261.24]
 for lead. Other lead-contaminated wastes may be considered hazardous if specifically listed in 40
 CHI 261.30-33, unless specifically excluded. Listed wastes from specific sources which
 contribute to the manufacture of computer displays include emission control dust from steel
 production and from lead smelting (K061 and K069 respectively), waste leaching solution of
 control dust from secondary lead smelting (K100), and spent baths and residues from
 electroplating operations containing cyanide (F006), which are sometimes used in PWB
 manufacturing. Specific sources of hazardous wastes, whether characteristic or listed wastes, are
 subject to handling, storage, and disposal restrictions detailed in the code of federal regulations.
       Manufacturers who emit lead are required to report the quantity of the emissions under
 the Community Right-to-Know Act.  EPA has recently reduced the reportable quantity threshold
 for lead from 10,000 Ibs per year to 100 Ibs per year of lead.

 4.1.7 Alternatives to Lead Use in Computer Displays

       Because of increasing pressure through regulation and market forces, attempts to reduce
 or eliminate lead in electronics have become popular. Several countries are considering or have
 already passed restrictions on the use and disposal of lead, prompting many companies to
 establish aggressive timelines for reducing or eliminating lead in their products. Several
 opportunities to eliminate or reduce the amount of lead used in a computer display are being
 aggressively researched. Two options being researched extensively are the development of a
 reduced lead frit, and lead-free solders for PWB manufacturing and assembly.
       Although not large in mass in a monitor, frit glass is 70 to 80% lead by weight. Lead is
 one component of a mixture that crystalizes under intense heat, providing strength to the
 vacuum-tight frit seal. An alternative lead-free frit glass has been developed that is based on tin
 and zinc oxides, along with phosphate (Busio and Steigelmann, 2000). The lead-free glass is
 inherently mechanically weak, requiring large amounts of ceramic fillers (A12O3) to be added to
 improve the mechanical strength of the seal.  It also requires the addition of vitreous silica
 particles to match the thermal expansion requirements of the CRT glass. The resulting mixture
 requires a firing cycle approaching 450°C, which is typical of frit glasses. A drawback is that the
 frit glasses stay vitreous during the typical 30-60 minute furnace dwell time. However, initial
 evidence suggests that the fired frit seal remains rigid during the pumping step, which occurs at
 350°C. Although comprehensive test results are not yet available, the high-temperature stability
 and rigidity of the lead-free frit glass is currently being tested under vacuum (Busio and
 Steigelmann, 2000).
       Lead-free solders have been the subject of industry research for some time. Driven by
renewed regulatory attention and by recent corporate commitments to reduce or eliminate lead
from their product lines, alternatives to lead-based solder are garnering increased attention.
Alternative lead-free solders include tin in combination with one or more of the following
metals: silver, copper, bismuth, germanium, and antimony. Several companies, including Sony,
Toshiba, Hitachi, and Ford Motor Company, have either already begun to implement electronics
production using a lead-free solder alternative, or have announced plans to do so. Though still a
                                         4-19

-------
4.1 LEAD
relatively new and untested technology, initial testing has shown that alternatives are capable of
producing quality component connections, though they have a narrower operating window and
require higher temperatures to apply (Keenan and Kellett, 2001).
                                           4-20

-------
                                                                           4.2 MERCURY
 4.2    MERCURY

        Mercury is not only used in the manufacturing of LCDs, but also is emitted from a
 number of processes over the life cycle of both LCDs and CRTs. Because of mercury's toxicity
 to both humans and the environment, this section presents a more detailed look at the uses and
 impacts of mercury, and its potential for causing harm as a result of the manufacture, use, and
 disposal of computer displays.

 4.2.1   Mercury in Computer Displays

        Mercury is an important material in the construction of cold cathode fluorescent lamps
 (CCFLs) that are used to backlight the LCD. A typical LCD utilizes a minimum of two CCFLs
 and can use as many as eight in larger displays. The CCFL consists of a glass tube filled with a'
 small amount of rare gas (typically argon) and a few drops of mercury. Metal electrodes built
 into the ends of the tube conduct electric current to the inside gas, vaporizing a portion of the
 mercury which then becomes excited, emitting light in the ultraviolet spectrum. Fluorescent
 phosphors, which coat the inside of the glass tube, convert the ultraviolet emissions from the
 mercury gas into visible white light. The phosphors are responsible for nearly all of the visible
 light from the lamp, with the visible mercury spectrum contributing only a little to the lamps
 output (Srivastava and Sommerer, 1998). No mercury is contained directly within the CRT.
       Although  a great deal of mercury is obtained by CCFLs, there also is a small source of
 mercury generated by the infrequent breakage of mercury lamps in the photolithographic
 exposure systems used to make both the CRT and LCD. Mercury filters are used in some water
 cooled exposure table equipment to catch the mercury from lamp explosions or trap it in water
 cooling baths. In  air cooled lamp usage, the proximity of aluminum metal  gives sites for
 amalgam formation and a number of clean-up procedures are used. Unbroken lamps  are returned
 to the lamp manufacturer for recycling or disposal (Donofrio and Eckel, 1999).
       Mercury may also be emitted from several processes required to manufacture, operate
 and dispose of both CRTs and LCDs.  For example, electricity generated from the combustion of
 coal results in the emission of mercury contained within the fuel.  Other processes potentially
 responsible for mercury emissions include mercury ore processing, non-ferrous metal production,
 and the recycling of LCDs at the end of their useful life. The amount of mercury released
 through these incidental mercury emissions is comparable to the amount of mercury used as a
 direct material in the manufacture of LCD backlights.

 4.2.2  Life-Cycle Inputs and Emissions of Mercury for Computer Displays

       Data on mercury and mercury-containing materials were collected and compiled as part of
 the life-cycle inventory. Material inputs containing mercury include primary materials (e.g.,
 CCFLs) which end up as part of the product, as well as ancillary materials (e.g., fossil fuels'
 containing mercury) which are consumed as part of the manufacturing process or other
 supporting processes (e.g., energy production). The data were aggregated by material from
individual processes and are presented by life-cycle stage for LCDs in Table 4-11. More detailed
material input data, which include the processes that use each input and the quantity of mercury
releases, are included in Appendix N.    ,
                                         4-21

-------
4.2 MERCURY
              Table 4-11.  Life-cycle stage mercury inputs for LCDs
Life-cycle stage
Manufacturing
Manufacturing
Inputs
Mercury
Backlight lamp (CCFL)
Quantity
3.99e-06
1.94E-03"
Units
kg
kg
Type
Primary material
Product
a Quantity of release shown represents entire mass of input material. Mercury may only comprise a fraction of the
total mass of material shown.

       Mercury is not listed as an input in the life-cycle inventory for CRTs. However, mercury
is contained in raw material inputs (e.g., mercury contaminants in fossil fuels burned to produce
energy) into processes such as non-ferrous metal production and energy production for both
LCDs and CRTs. Though mercury inputs exist, they occur in upstream manufacturing processes
where input data did not contain detailed composition data for fuel inputs.
       Releases of mercury and mercury-containing materials into the environment occur
throughout the entire life cycle of the computer display. Environmental releases include
airborne, waterborne, solid waste, and hazardous waste emissions.  Similar to the inputs,
emissions data were aggregated by the material released from individual processes and then
reported by life-cycle stage. The mercury and mercury-containing material released, the quantity
of the release, the type of release (e.g., waterborne), and the ultimate disposition of the release all
affect the nature and type of environmental impacts.
       The total life-cycle outputs/emissions containing mercury for both CRTs and LCDs are
organized by output type and shown in Tables 4-12 and 4-13, respectively. More detailed data
on mercury and mercury-containing outputs for each process are presented in Appendix N.

                 Table 4-12. Life-cycle stage mercury outputs from CRTs
Life-cycle stage
Materials processing
Materials processing
Materials processing
Manufacturing
Manufacturing
Use
End-of-life
End-of-life
Outputs
Mercury
Mercury ....
Mercury compounds
Mercury
Mercury compounds
Mercury
Mercury
Mercury compounds
Quantity
3.00E-06
1.42E-10a
9.68E-07
1.12E-06
1.35E-12
7.51E-06
-1.15E-07
4.33E-11
Units
kg
kg
k*
kg
^
kg
kg
kR
Type
Airborne
Solid waste
Waterborne
Airborne
Waterborne
Airborne
Airborne
Waterborne
Disposition
Air
Landfill
Surface Water
Air
Surface Water
Air
Air
Surface Water
 a Quantity of release shown represents entire mass of waste disposed. Mercury may only comprise a fraction of the
 total mass of waste shown.
                                            4-22

-------
                                                                            4.2 MERCURY
Table 4-13. Life-cycle stage mercury outputs from T.rns
Life-cycle stage
Materials processing
Materials processing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Use
End-of-life
End-of-life
Outputs
Mercury
Mercury compounds
Broken CCFL
Mercury
Mercury compounds
Waste CCFL, with mercury
Waste glass, with mercury
Wastewater stream, from
CCFLmfg.
Mercury
Mercury
Mercury compounds
Quantity
9.44E-07
5.82E-07
2.69E-07"
2.64E-06
6.52E-14
8.17E-103
1.05E-103
167
2.80E-06
-8.64E-08
1.62E-11
Units
kg
kg
kg
kg
kg
kg
kg
kg
kg
kg
kg
Type


Solid waste


Hazardous waste
Hazardous waste
Waterborne

Airborne
Waterborne
Disposition
Air
Surface Water
Landfill
Treatment
Surface Water
Treatment
Landfill
Treatment
Air
Air
Surface Water
 1 Quantity of release shown for solid waste and hazardous waste represents entire mass of waste disposed. Mercury
 may only comprise a fraction of the total mass of waste shown.
       Mercury is released into the environment in many forms, but is most typically an airborne
release. The largest air emissions of mercury result from the generation of electricity from fossil
fuel burning. For LCDs, there is nearly the same amount of mercury emitted to the air from
energy production (3.22 mg) as the mass of mercury used in the fabrication of an LCD (3.99 mg).
In fact, the amount of mercury emitted to the air from electricity generation for CRTs (7.75 mg)
is greater than the entire amount of mercury from both the fabrication and energy production for
an LCD. Other airborne releases include the processing of ores such as lead, and the production
of several raw materials such as aluminum, polycarbonate, and steel.

4.2.3   Computer Display Life-Cycle Impacts for Mercury

       The life-cycle impacts of mercury, mercury-based compounds, and materials containing
mercury (e.g., waste glass from broken CCFLs) calculated for CRTs and LCDs during the LCIA
are summarized in Tables 4-14 and 4-15, respectively. Impact scores in the table are expressed
in units specific to each impact category (see Chapter 3.1 for a discussion of impact category
units and weighting). The total impact score for each category resulting from mercury and
mercury-based materials is presented at the bottom of each table.
                                          4-23

-------
4.2 MERCURY


s
^
o
*»•<
1
1
tJ
OS
>5

09
13
CQ
P
E
•P«
£
|
|
«
E
«t-t
o
e
E
E
f>
CG
i
a
3
C














j>.
w
I
1

s§
u
to
o
ft
«




























"3
« •§ ifc
|| ,§,
H """'
It?
Ill
i "3
S3 ^"^
^ yj O t9l
111 I
3 «
S
'Hi!
o — §* -2
C3
11

S 3
^O ~Jjj sis'
••4 FO ^^
II -

K> i~
s
"2 § * **L^
CS In
Hj W
^•^ s^


•a
*rj
1



-


0)
U a>
CJ S»
i ^
y^ ^
a



VO
en
O



o


1
co


o




o



'



c
a
^
bJ
_e
"5
<
u
O
i
1
b
c
•c
(
c
^
I
10
9.02E-04



0


T-H
«n


o




o



o
C
c
I
t
C
a
s
M
C
"E
C/3
4
C
<
I
c
c
•(
(
0
s
VO
1

o



0


VD
T— H
"


0




o







C
a
s



j

c.
^
i
*
2
0
§
^
o\
1



o


o
T— 1
t-


o


'

o



u
c
c
1
E?
a
s



b
•j

(
^
i
c
S
.51E-06

0



o


VO
Mr]
1-H
in
^


o




o







C
a
S







.
i
^
n
p
O



o


10
1


o




o







c
u
IH
s




jj
•»•
(4
c
n

f
.28E-08
CN
4.04E-08



0


CO
i

<
*T
C

— 1
10
1
OS



o


i
V)


0




0






i
o
w
<
(

1
1
]
c
]
1
»
^

H
                                       4-24

-------
	 	 	 _ 	 ; 	 __^_ 	 4.2 MERCURY




a
.
Z
CM
J,

§
%
•4-
1
^a
O5
pact score
.1
1
^
6
y
o3
s
*o
b

^
cc


*7<

1













BS by category
JJ
§
1













it!
2'x g
« -8 a-
H
'3
*3 /-^>
11
s ^
CP
^

pC
lift
!"is

S w
111
131
si

(U B
S *•* *~*
s§ J-L
"o c
^^

s
•O S " rfT-
S ^ « g

33 w
^




I


"S cu
£> 2?
i •§
eg co
1
^f
o\

r~-
co
*~"




o


9.44E-07



o




o







1"
1


2 "^
JS is.
s
o

^t*
§
IO




o


3.07E-04



0




0





CO
T3
C
Mercury compou


QC
co c
I 8
11
o



0





o


o


*— 1
W
oc
•^



o







Broken CCFL


c
C
3
o
c2
3
B
C3
t
W
>o

• e
c
o
e
•



o
en


i



o




o







J

w
acturin
§
d
«


[I
ex
<




O


i
en



O




O





•3
e
Mercury compoui

M
C
1
4-t
3
O



O





0


0



0




s
n

^"''"





Waste glass, with
mercury

t>0
C
3
a
c
g
CX
C


0





o


1



o




o







1
1



1
oo
c
t
\;
0


O





o


oo
i
OO
I



o




0







1


^§
o
c
d.
V
oe

fVl
d
V
_ I




O


"1
CO



O




O





CO

Mercury compoui


a
"o
1
<
t
^

_
<
&
V




V
1
f


c
H
T-


^
i




V
1
i
£*,

l>





|
"S
1
f*
•w
i
s
i
1
| ,;... -
•*->
1
o
rS d '
^3 S
ra T3
co C
S ^S
B co
« « '
O &
'I co
3 S
§3 *^
'S "o
£ CO
" J
l'|

||

CO cj
g .§
CO M
o *S
1 1
a S
11
1 6

J3 O
•O U
C 60
5 «
•o c •
"3 "

o ^
|5
CO [Q
=2
1
3
1
3
8
OH

4-25

-------
4.2 MERCURY
       Impact scores for some mercury-based inputs and outputs shown in Tables 4-11 through
4-13 were not calculated if the type and disposition of the input or release was not expected to
contribute to any of the impact categories. For example, a waterborne release of mercury with a
disposition going to treatment assumes that the mercury was treated prior to being released to the
environment. However, since inventory data for subsequent treatment/disposal processes could
not be obtained, it was assumed the mercury (or other inventory item) had  been removed to a
level such that the subsequent release of treated wastewater would not contribute significantly to
aquatic toxicity impacts. Similarly, impact scores were not calculated for releases going to
recycling/reuse or for those designated as a product.
       The life-cycle mercury-based outputs from LCDs had a broader affect on the environment
than those from CRTs, impacting a wider group of impact categories. Impacts to both solid and
hazardous waste landfill use, as well as to the chronic health effects of workers, all directly result
from the use of mercury in the LCD backlights. No mercury is required for the fabrication of a
CRT. Although the quantities are not large  (see Tables 4-11 through 4-13), they cannot be
discounted, given the toxicity of mercury to both human health and the environment.
       Chronic occupational toxicity impacts were only calculated for mercury inputs to
processes in the CDP.  The overall impact scores (3.99E-06 tox-kg for LCD, none for CRT)
likely underestimate the chronic occupational impacts for mercury, because they are based on
inputs only and do not consider chronic occupational impacts from outputs in other processes
such as aluminum production or fluorescent lamp recycling, which may result in emissions of
mercury that originate within the workplace.
       The contribution of mercury-based impacts for each computer display technology to the
overall impacts for each individual impact category is shown in Table 4-16.  Values in the table
are expressed in the percent contribution the material made to the overall impact score for all
materials (e.g., liquid crystals, fuel oil, glass) for each category. The percent contributions give
an indication of the importance of mercury-based impacts relative to the life-cycle impacts from
other materials or outputs from the computer display.

      Table 4-16.  Summary of percent contributions from mercury-based materials to
                               individual impact categories
Impact category
Hazardous waste landfill use
Solid waste landfill use
Chronic health effects- public
Chronic health effects- occupational
Aquatic toxicity
Terrestrial toxicity
CRT
N/A
N/A
2.64E-05
N/A
4.01E-01
2.64E-05
LCD
NA
3.65E-08
3.45E-05
5.80E-07
1.05E-02
3.48E-05
 N/A = Not applicable
                                           4-26

-------
                                                                           4.2 MERCURY

        The results from Table 4-14 and 4-lfiJndicate tha| the mercury impacts from a CRT
  exceed the impacts from an LCD in categories common to both technologies.  This was not
  expected, because mercury is used intentionally in an LCD, but not in a CRT.  However the
  results are not surprising because mercury emissions from coal-fired power plants are kriown to
  be one of the largest anthropogenic sources of mercury in the United States. Because the CRT
  consumes significantly more electricity in the use stage than the LCD, its use stage emissions of
  mercury are proportionately higher than those of the LCD. In fact, the mercury emitted from the
  generation of power consumed by the CRT exceeds the entire amount of mercury emissions from
  the LCD, including both the mercury used in LCD backlights and the mercury emissions from
  electricity generation in the use stage that can be attributed to the LCD.
        The impacts resulting from mercury and mercury-based materials do not appear to be
  significant relative to the total impacts of all of the computer display materials (eg liquid
  crystals, lead solder), as shown in Table 4-16. The largest contribution is 0.4% of the total
  aquatic toxicity impacts for CRTs, and 0.01% of the total aquatic toxicity impacts for LCDs
 Impacts to other categories from both LCDs and CRTs were minimal.

 4.2.4   Exposure Summary

       Mercury may pose a threat to human health anytime there is the potential for human
 exposure throughout the life cycle of a computer display. Exposure occurs anytime a chemical or
 physical agent, in this case mercury or mercury compounds, come into contact with an organism
 be it human or ecological. This section qualitatively identifies potential exposures for workers in
 facilities using mercury (occupational exposures), the general public, who may be exposed to
 mercury releases into the ambient environment, and the ecological population.

 4.2.4.1 Occupational exposures

       About 4 mg of elemental mercury (combined total of all mercury contained within the
 backlights) is used to manufacture the fluorescent backlight for the LCD backlight unit assembly
 Workers manufacturing the backlights may be exposed to the mercury used for these lights  This
 study found no information on the specific manufacturing processes that are used to make CCELs
 or the specific worker exposures that could occur, but did find manufacturing information for
 generic fluorescent lamps.  We assume the processes are similar and have included a brief
 discussion of the fluorescent lamp manufacturing process and potential sources of worker
 exposure, below.
       In fluorescent lamp manufacturing, pre-cut bulbs are washed,  dried, and coated with a
 liquid phosphor emulsion that deposits a film on the inside of the bulb.  Mount assemblies are
 then fused to each end of the bulb and the bulb is transferred to an exhaust machine  The bulb is
 then exhausted and mercury is injected into the bulb.  Some of the mercury combines with the
 emulsion on the interior of the bulb where it remains over the life of the bulb. The glass bulb is
 then filled with an inert gas and sealed (EPA, 1997).
       During the lamp manufacturing process, emissions of mercury can occur from transfer
 and parts repair during mercury handling, by the mercury injection operation, and from broken
 lamps, spills, and waste materials. Workers can be exposed to mercury from any of these
 sources, but mercury air levels can be reduced by process modifications, containment, ventilated
mclosures, local exhaust ventilation, and temperature control (EPA, 1997).

                                         4-27      ~                        ~  :—

-------
4.2 MERCURY
       There were no other inputs of mercury reported in the LCD life-cycle inventory, and none
were reported in the CRT life cycle. Workers may also be exposed to inorganic mercury in the
fluorescent backlights when processing LCDs at the end-of-life. Family members of workers
may be exposed as well, from a worker's clothing or shoes if they are contaminated with mercury
and brought home.
       The processes associated with LCI mercury inputs and releases are presented in
supplemental tables in Appendix N. It is important to note that while this list gives an indication
of where likely mercury exposures may occur, it is not exhaustive. Many processes and
subprocesses may be contained within a process listed, each of which may pose its own potential
for occupational mercury exposures.

4.2.4.2 General population exposures

       Mercury is a persistent biaccumulative toxic (PBT) chemical, as designated by EPA
(EPA, 2001b).  PBT pollutants are highly toxic, long-lasting substances that can build-up in the
food chain to levels that are harmful to human and ecosystem health. The general public may be
exposed to metallic mercury that is not safely contained (although it is unlikely that the backlight
would break to release mercury during normal LCD use) or to methylmercury-contaminated
foods. Mercury also can be passed from a pregnant woman to developing child through the
placenta, and from a mother to nursing infant through breast milk.
       Most of the mercury released to the environment throughout the CRT and LCD life cycles
results from electricity generation required for use, manufacturing, and materials processing life-
cycle stages. A total of approximately 4 mg of mercury and mercury compounds are released to
air for an LCD, and approximately 12 mg are released to air for a CRT. Mercury is naturally
present in coal and becomes airborne when coal is burned to generate electricity. Airborne
mercury can stay in the atmosphere for up to a year, and can travel thousands of miles (EPA,
2001a). EPA modeling suggests that "a substantial fraction" of the mercury released to air by
utilities is dispersed "well beyond the local area" due to the fine paniculate nature of the
emissions and tall stacks (EPA, 1998).  Mercury in the atmosphere moves to land and water by
 settling out with particles and being washed out by rain (dry and wet deposition). It may be
 deposited directly to water or be carried by runoff to a lake, stream, or ocean. In the LCIs, in
 addition to the air releases of mercury, 0.6 mg of mercury and mercury compounds are released
 directly to surface water in the LCD life cycle, and 1 mg in the life cycle of the CRT. Ultimately,
 at the end of life, the 4 mg of mercury in the LCD backlight will most likely be released to the
 environment during LCD recycling or disposal processes.
        Surface water is the environmental medium of most concern for mercury.  In a surface
 water environment, inorganic mercury can be transformed into methylmercury, a form which
 readily bioaccumulates in fish (inorganic mercury does not tend to bioaccumulate).  An EPA
 study of mercury supports a "plausible link" between releases of mercury from industrial and
 combustion sources and methylmercury found in fish3 (EPA, 2001c). Methylmercury
        3 The mercury released to the environment from the CRT and LCD life cycles are only part of the overall
 burden of mercury released to the environment from coal-fired power plants for all uses of electricity. The
 proportion of mercury in fish that is due to coal-fired power generation is not known.  In addition, there are other
 natural and anthropogenic sources of mercury to the environment.
                                            4-28

-------
                                                                             4.2 MERCURY
 concentrations at the top of the food chain (such as in predatory fish or fish-eating animals) can
 be thousands or even millions of times greater than that iti the surface water itself (EPA, 2000a).
        The most important mercury exposures to the general public result from eating fish that
 are contaminated with methylmercury. The populations of most concent are children and women
 of child-bearing age (the developing fetus may be the most sensitive to the effects of
 methylmercury). Also of concern are people whose diet largely depends on fish, such as with
 some native cultures. The overall amount of exposure to mercury from eating fish depends on
 both the concentration of mercury in the fish, and on the amount of fish a person regularly eats.
 Fish advisories due to methylmercury contamination have been issued by EPA4,39 states, and
 some tribes, providing consumption limits for certain species of fish (EPA, 2000a, 200 lc).
        Freshwater fish are most affected, but some saltwater fish have also been found to be
 contaminated with methylmercury.  The Food and Drug Administration (FDA) has issued a fish
 consumption advisory for pregnant women, nursing mothers, and small children to avoid eating
 certain large saltwater fish  (shark, swordfish, king mackerel, and tilefish), and to limit overall
 weekly fish consumption, due to methylmercury contamination (FDA, 2001).

 4.2.4.3 Exposure and effects to ecological populations

        In addition to the fish themselves being contaminated, wildlife that eat fish also may be at
 risk from exposure to methylmercury. Species of concern include loons,  eagles, mink, otter,
 wood stork, and the endangered Florida panther. Adverse effects of mercury to wildlife include
 death, reproduced reproductive success, impaired growth and development, and behavioral
 abnormalities. Levels have been measured in some individual wild animals that are comparable
 to those levels seen to cause harmful effects in laboratory tests with the same species (EPA,
 2001c). Ambient water criteria have been developed by EPA under the CWA. Criteria for'the
 protection of aquatic life for mercury are presented in Table 4-17.
        "EPA is issuing a national advisory concerning risks associated with mercury in freshwater fish caught by
friends and family. The groups most vulnerable to the effects of .mercury pollution include: women who are
pregnant or may become pregnant, nursing mothers, and young children. To protect against the risks of mercury in
fish caught in fresh waters, EPA is recommending that these groups limit fish consumption to one meal per week for
adults (6 ounces of cooked fish, 8 ounces of uncooked fish) and one meal per week for young children (2 ounces
cooked fish or 3 ounces of uncooked fish)." - EPA Office of Water, January 2001
                                           4-29

-------
4.2 MERCURY
Type of criteria
Criteria value O^g/L)
Notes
For protection of aquatic life
Freshwater Criteria Maximum Concentration
(CMC)
Freshwater Criterion Continuous
Concentration (CCC)
Saltwater Criteria Maximum Concentration
(CMC)
Saltwater Criterion Continuous Concentration
(CCC)
1.4a
0.77 a
1.8"
0.94 a
the acute limit for the priority
pollutant in freshwater
the chronic limit for the priority
pollutant in freshwater
the acute limit for the priority
pollutant in saltwater
the chronic limit for the priority
pollutant in saltwater
Source: Water Quality Criteria and Standards, EPA Office of Water. Revised: 06/21/2001
http://oaspub.epa.gov/wqsdatabase/epa jep_parameter; report for mercury
* Criteria for metals are expressed in terms of the dissolved metal in the water column. This recommended water
quality criterion was derived from data for inorganic mercury (II), but is applied here to total mercury. If a
substantial portion of the mercury in the water column is methylmercury, this criterion will probably be under-
protective (EPA is updating the ambient water quality criteria on methylmercury.) In addition, even though
inorganic mercury is converted to methylmercury and methylmercury bioaccumulates to a great extent, this criterion
does not account for uptake via the food chain because sufficient data were not available when the criterion was
derived.
 4.2.5  Human Health Effects
       Mercury affects the nervous system, brain, and kidneys.  Effects on the nervous system
 vary depending on the form of mercury. Inorganic mercury salts, for instance, do not enter the
 brain as readily as elemental mercury or methylmercury. Symptoms of mercury effecting the
 brain and nervous system include personality changes, tremors, changes in vision such as
 narrowing of the visual field, deafness, loss of muscle coordination, loss of sensation, and
 problems with memory (ATSDR, 1999).
       The fetus, infants, and young children are especially susceptible to the effects of mercury
 on the nervous system. As mentioned above, mercury (especially methylrnercury in food) can be
 passed from a pregnant woman to the unborn developing child, and from a mother to a nursing
 infant through breast milk. The developmental effects of mercury vary in severity depending on
 the amount of exposure.  Children exposed in this way may show small decreases in IQ or may
 be slower to walk and talk. More severe effects might include brain damage with mental
 retardation, blindness,  muscle weakness or seizures, and inability to speak (ATSDR, 1999).
        Mercury accumulates in the kidneys and all forms of mercury can cause kidney damage at
 higher exposures (ATSDR, 1999). Short term exposure (hours) to high levels of mercury vapor,
 as might occur through an accidental spill in the workplace, include damage to the lining of the
 mouth, irritated lungs and airways, nausea, vomiting, diarrhea, increased blood pressure or heart
 rate, skin rashes, and eye irritation. Skin contact may cause an allergic reaction (skin rashes) in
 some people (ATSDR, 1999).
                                             4-30

-------
                                                                             4.2 MERCUB
 4.2.5.1 Chronic effects (noncancer)

        A reference dose (RfD) is an estimate (with uncertainty spanning perhaps an order of
 magnitude) of the daily exposure through ingestion to the human population (including sensitive
 subgroups) that is likely to be without an appreciable risk of deleterious noncancer effects during
 a lifetime (in mg/kg-day).  Similarly, a reference concentration (RfC) is an estimate (with
 uncertainty spanning perhaps an order of magnitude) of the daily inhalation exposure to the
 human population (including sensitive subgroups) that is likely to be without an appreciable risk
 of deleterious noncancer effects during a lifetime (in mg/m3) (Barnes and Dourson, 1988). RfDs
 and RfCs established by EPA for mercury and mercury compounds are presented in Table 4-18.

     Table 4-18.  Chronic toxicity reference values for mercury and mercury compounds
Form of mercury*
Elemental mercury
(Hg)
Methylmercury
(CHsHg*)
Mercuric chloride
(HgCl)
Ingestion: reference
dose (RfD) (mg/kg-
day)b
not available
0.0001
0.0003
Inhalation: reference
concentration (RfC)
(mg/m3)c .
0.0003
not available
not available
Notes, Source
Based on hand tremor, memory
disturbances, and other effects seen in
human occupational inhalation studies
(IRIS, 2001).
Based on developmental
neuropsychological impairment seen
in human epidemiological studies
(IRIS, 2001).
Based on autoimmune effects in rats
(IRIS, 2001).
b milligrams per kilogram of body weight per day for oral exposure (ingestion)
c milligrams per cubic meter of air, assuming continuous inhalation exposure for a 70-kg adult
4.2.5.2 Carcinogenicity

       EPA has determined that mercury chloride and methylmercury are Possible Human
Carcinogens (cancer weight of evidence [WOE] classification C) based on limited evidence of
carcinogenicity in animals and inadequate or lack of human data. Elemental mercury is classified
by EPA as Not Classifiable as to Human Carcinogenicity (WOE class D) based on inadequate or
no evidence of carcinogenicity (IRIS, 2001).
                                           4-31

-------
4.2 MERCURY
4.2.6  Environmental Regulations for Mercury
                                                  %                                   *
       This section presents a brief summary of the U.S. regulations for mercury and mercury
compounds that may affect facilities that manufacture materials for the computer display or are
otherwise affected by the life cycle of computer displays. It should be noted that many of the
parts that go into the computer display are manufactured in other countries with their own
regulations which may differ significantly from those presented below.
       Air emissions of mercury are regulated under the CAA of 1970 and the amendments of
1977 and 1990. Under the CAA, mercury is regulated as a hazardous air pollutant (HAP), which
is by definition a chemical that is either known or is  suspected to cause serious health problems
for humans. EPA established National Emission Standards for HAPs (NESHAPs) for mercury
emissions based on risk under the pre-1990 version of the Clean Air Act. These NESHAPS [40
CFR 61 Subpart E] cover three source categories: ore processing facilities, mercury cell chlor-
alkali plants, and sewage sludge driers.  Specific source requirements are specified for, among
other things, municipal waste combusters, hazardous waste combusters, and mercury ore
processing facilities. These source requirements could take the form of either a NESHAP or a
maximum achievable control technology (MACT) requirement.
       OSHA has established standards for protecting worker health through the maintenance of
a safe working environment.  The OSHA permissible exposure limit (PEL) for workplace
exposure to mercury is 0.1 mg/m3 (8-hour time weighted average). NIOSH has also established
recommended exposure limits (RELs) for several mercury compounds including a REL of 0.05
mg/m3 for mercury vapor.
       Mercury emissions to surface water are regulated under the CWA, which lists mercury as
a priority pollutant [40CFR 401.15], requiring the limitation of mercury in point source
discharges. For mercury discharges, CWA regulations specify technology-based effluent limits
for classes and categories of industries (see 40 CFR  401,403, Appendix B), and describe the
rights of states to establish effluent limits more stringent than technology-based standards.
Technology-based standards are listed for the following specific industries and point sources
involved in computer display manufacturing:  nonferrous metals production, including primary
precious metals and mercury (40 CFR 250); secondary mercury (40 CFR 421.200); steam electric
power generation (40 CFR 423- Appendix A); and mercury ore mining (40 CFR 440.40). The
CWA also requires that new and existing points sources of mercury obtain a NPDES permit,
which will establish effluent limits for mercury discharges to surface waters.
       To protect human health and preserve the nations drinking water supply, EPA has been
tasked by the SDWA to establish safe drinking water standards for toxic chemicals. In
accordance with the SDWA, EPA has established a safe drinking water standard for mercury of
2/zg/L.
       The release or disposal of solid or hazardous waste containing mercury is regulated under
RCRA, which outlines specific classification and disposal requirements for products and wastes
 that contain mercury.  Mercury is both a characteristic and a listed waste under RCRA. A solid
 waste containing mercury may be considered a D009 characteristic hazardous waste if, when
 subjected to a TCLP test, the extract exceeds 0.2 mg/L [40 CFR 261.24] for mercury.  Other
 mercury-contaminated wastes may be considered hazardous if they are specifically listed in 40
 CFR 261.30-33, unless they are specifically excluded.  Listed wastes for mercury include
 leachate resulting from the disposal of more than one restricted waste classified as hazardous
 (F039), and wastewater treatment sludge and brine purification muds resulting from the mercury
 -                                        —   .

-------
                                                                           4.2 MERCURY
 cell process in chlorine production (KK^andJOm respectively). Hazardous wastes are subject
 to land disposal restrictions requiring that wastes be treated to below regulatory threshold levels
 before they may be land-disposed.
        In order to reduce the amount of hazardous waste in a landfill, EPA established the
 Universal Waste Rule (UWR) in 1995. The rule was intended to encourage the recycling and
 proper disposal  of common hazardous waste components found in municipal waste streams, and
 reduce the regulatory burden on businesses who produce these wastes. The rule allows for less
 stringent standards for storing, transporting, and collecting wastes. However, the waste must
 comply with full hazardous waste requirements for final recycling, treatment, or disposal.
 Batteries and fluorescent lamps are included in the rule.
       EPA also has regulated the air emissions of mercury from hazardous waste combustion
 and from industrial boilers and furnaces under RCRA.  EPA has issued new standards for
 mercury emissions from these sources.   .
       In December 2000, EPA announced plans to require coal-fired power plants to cut their
 emissions of mercury. EPA plans to propose the regulations by 2003, with final rules in place by
 2004 (EPA, 2000a). More recently, in April 2001  the Bush administration sought to dismiss an
 electric industry lawsuit that would stop the EPA from regulating mercury and other toxic air
 pollutants (Doggett, 2001).
       Manufacturer's who emit mercury are required to report the quantity of the emissions
 under the Emergency Planning and Community Right-to Know-Act (EPCRA).  EPA has
 established a reportable quantity threshold of 10 pounds for a facility that manufactures,
 processes, or otherwise uses mercury. A similar threshold of 10 pounds exists for facilities that
 manufacture, process, or otherwise use mercury compounds. Data that has been reported by
 facilities for mercury is made available to the public through the publication of the Toxics
 Release Inventory (TRI).

 4.2.7  Alternatives to Mercury Use in LCDs

       In an effort to minimize or even eliminate the quantity of mercury used during the
 fabrication of LCDs, manufacturers have begun to develop mercury-free alternatives to mercury
 vapor backlights. One such alternative is a flat lamp which is filled with inert gas xenon in place
 of the typical mercury vapor lamps.  The lamp has the appearance of a large white tile about one
 centimeter in thickness. Its dimensions are the exact same as the screen itself, illuminating the
 image evenly, hence eliminating the need for complex optical systems to distribute the light. The
 new lamp is capable of emitting enough light to make the monitor twice as bright, making it
 possible to use the screen during  daylight, while also extending the viewing angle fOSRAM
 1998).                                                                            '
       The new lamps generate light in a fashion similar to conventional backlight lamps. An
 electrical current passed through a gas discharge produces ultra-violet light which is then
 converted to visible light by phosphors. Unlike mercury gas which causes 'greying' in the
 phosphors over time, the xenon gas does not affect the phosphors, extending the life of the lamp.
 The lights have an average lifetime of up to 50,000 hours compared to only about 20,000 hours
 for the conventional mercury backlights, extending the life of computer LCD displays before they
have to be replaced (OSRAM, 1998).
       A drawback of the lamps, and the current subject of research is the reduced energy
efficiency of the new mercury-free lamps. The luminous efficiency of the new mercury-free

                                         4-33

-------
4.2 MERCURY
lamp is only about half of the efficiency of a conventional backlight, due to less efficient
conversion of the UV light to visible light by the phosphors (OSRAM, 1998).
                                           4-34

-------
                                                                   4.3 LIQUID CRYSTALS
 4.3    LIQUID CRYSTALS

       One of the most significant differences between the two computer displays is the use of
 liquid crystals in the LCD to generate an image. The toxicity of liquid crystals in LCDs has been
 alluded to in literature indicating the potential for human health concerns. Because of relative
 lack of information about these compounds, this section provides a more detailed look a liquid
 crystals to better understand their potential impacts on human and ecological populations,

 4.3.1  Liquid Crystals in Computer Displays

       Liquid crystals (LCs) are organic compounds with the optical and structural properties of
 crystals, but with the mechanical features of fluids. There are hundreds of LC compounds that
 may be used in an LCD, each with different physical and optical characteristics.  They are
 typically classified by molecular weight, with low molecular weight LCs typically used for LCD
 computer displays. LCs are not required for the fabrication of CRTs.
       LCs are responsible for forming and transmitting the image produced by an LCD. The
 LC portion of an LCD typically consists of as many as 20 different LC substances, mixed
 together to form a white, opaque liquid that flows easily (Merck, 1999).  The mixture consists of
 elongated molecules that are held together at their ends by polar forces and aligned in the same
 direction. The molecules move together in a series of flexible molecular chains, with each chain
 influencing the alignment of other chains. By exposing the molecular chains to electric fields,
 the alignment of the chains,  and by extension their ability to transmit light, can be manipulated
 (SEMI, 1995).
       LCDs are fabricated  using a complex multi-step process in a clean room (refer to Chapter
 1 for a more detailed description of the fabrication process).  Once the front and back layers have
 been manufactured and assembled into a display cell, the LC is  ready to be added. The
 assembled display cells are placed into a vacuum chamber containing a reservoir of the LC
 mixture, and the chamber is  evacuated. A corner of the empty display cell is lowered into the LC
 mixture using a remote control. Nitrogen gas is then introduced into the evacuated chamber to
 bring the pressure up to approximately 1 atmosphere, exerting pressure on the surface of the LC
 mixture, forcing it up into the display cell (SEMI, 1995). Approximately 0.6 mg of LC is
required for every square centimeter of panel surface (Merck, 1999).  The display cell is sealed
 once the LC fully penetrates the cell.

 4.3.2  Life-Cycle Inputs and Outputs of Liquid Crystals for Computer Displays

       Data on LC compounds were  collected and compiled as part of the life-cycle inventory.
The data were aggregated by material from individual processes and presented by life-cycle stage
for LCDs in Table 4-19.
       The individual LC compounds identified in Table 4-19 formed the ingredients of the
 liquid crystal mixture used in the LCD. These compounds, used in varying quantities, are mixed
together in a liquid crystal manufacturing process to develop a mixture with the desired optical
characteristics for the LCD.  While the above inputs illustrate the quantities of individual LC
compounds found in the LCD evaluated in this LCA, LC mixtures found in other LCDs could be
comprised of up to 20 or more liquid  crystal compounds selected from the hundreds of
                                         4-35

-------
4.3 LIQUID CRYSTALS
compounds currently available for use in LCDs. A small quantity of the liquid crystal mixture
(1.2 grams) is then used in the fabrication of the LCD.

                    Table 4-19. Life-cycle stage liquid crystal inputs
Life-cycle stage
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Inputs a
Liquid crystal A
Liquid crystal B
Liquid crystal C
Liquid crystal D
Liquid crystal E
Liquid crystal F
Liquid crystal G
Liquid crystal mixture, for 15" LCD
Quantity
0.26
0.37
0.22
0.33
0.070
0.34
0.19
1.2
Units
g
g
g
g
g
g
g
s.
Type
Primary material
Primary material
Primary material
Primary material
Primary material
Primary material
Primary material
Primary material
a Identities of liquid crystal compounds have been masked to protect the confidentiality of the compound names.
       Life-cycle inventory data indicate that LCs are primarily released at the time of the
product's final disposition. Although other outputs of liquid crystals surely exist, they are likely
minimal and were not identified in the life-cycle inventory collected for LCDs. Evaporative LC
emissions during the manufacturing and mixture formulation processes are minimal due to the
typically low vapor pressures of LC compounds (Becker, 2001).  Releases resulting from broken
or defective LCDs during manufacture are also likely to be small because of the strong adhesives
forces between LCs and the polymer film covering the outer layer of the displays (Becker, 2001).
However, small amounts of LC compounds were observed to present in the waste of one LCD
manufacturing facility during data collection (Overly, 2001). Because data were not provided
from manufacturers on manufacturing releases, and no data were available for releases at end-of-
life, it is difficult to definitively quantify the releases or their significance to the environment.

433  LCD Life-Cycle Impacts of Liquid Crystals

       The life-cycle impacts of LC compounds calculated for LCDs during the LCIA are
summarized in Tables 4-20. Impact values in the table are expressed in units specific to each
impact category (see Chapter 3.1 for a discussion of impact category units and weighting). The
total impact score for each category from LCs is presented at the bottom of the table.
                                          4-36

-------
                                                                   4.3 LIQUID CRYSTALS
                    Table 4-20. Summary of LCD liquid crystal impacts
Life-cycle
stage
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Manufacturing
Material
Liquid crystal A
Liquid crystal B
Liquid crystal C
Liquid crystal D
Liquid crystal E
Liquid crystal F
Liquid crystal G
Total category impact score
Chronic health effects- occupational
(tox-kg)
5.29E-04
4.35E-04
3.89E-04
1.40E-04
6.84E-04
6.53E-04
7.31E-04
3.56E-03
       Impact scores have been calculated based on the inventory item, release type, and its
 reported disposition.  Occupational impacts to workers as a result of LC inputs are shown in the
 Table 4-20. Impacts were not calculated for LCs which end up as part of the product because
 users of the product are not expected to become exposed to the LC compounds during the typical
 operation of the LCD. In addition, because releases of LCs to the environment were not provided
 by the manufacturers in the LCI, potential impacts resulting from these releases were also not
 assessed in this project. LCs are not used to fabricate CRTs and so have no environmental
 impacts in the CRT life cycle.
       LCs do not appear to contribute significantly to any of the impact categories defined for
 this study.  The total score for occupational impacts based on potential worker exposure to LCs
 of 4.18 tox-grams represents less than 0.01% of the total overall chronic occupational health
 effects impact score of 898 tox-kg for the functional unit of one LCD.

 43.4 Exposure Summary

      Like any materials classified as potentially toxic, LCs have the potential to pose a threat
 to human health anytime there is the potential for human or ecological exposure throughoutthe
 life cycle of a LCD. Exposure occurs anytime a chemical or physical agent, in this case LC
 compounds, come into contact with an organism, be it human of ecological.  This section
 qualitatively identifies potential exposures for workers in facilities using LCs (occupational
 exposures), the general public, who may be exposed to LC releases into the ambient
 environment, and the ecological population.

 4.3.4.1 Occupational exposures

      Occupational exposures to LCs during the fabrication of the LCD panels are not expected
 to be significant.  LCD panels are fabricated in a clean room environment.  The previously
 assembled but empty display cells are placed into a vacuum chamber containing a reservoir of the
LC mixture, and lowered into the LC mixture using a remote control. Approximately 1.2 grams
of LC compounds are used to develop the LCD panel. The enclosed nature of the chamber
combined with the equipment (e.g., gloves, aprons) worn by workers in a clean room
environment may both act to minimize exposures.
                                         4-37

-------
4.3 LIQUID CRYSTALS
       However, the potential for other occupational exposures still exist.  Workers could
become exposed to LCs during other manufacturing process steps, such as  during the formulation
of the LC mixture (approximately 1.8 grams from Table 4-31), the handling and disposal of
broken or defective panels, or during the recycling or disposal of an LCD at the end of its useful
life. Because of the physical nature of the LCs (e.g., they are not volatile),  typical worker
exposures are likely to be through dermal or ingestion (e.g., accidentally ingesting LC present on
a workers hands) routes. Worker exposure to LCs via dermal exposure is expected to be minimal
for workers who wear gloves while handling LCs.

4.3.4.2' General population

       Because of the enclosed nature of the LCD panel, it is unlikely that consumers could
become exposed to LC compounds contained within the display through normal usage. LCs may
be released into the environment should the panel become fractured, either through accident or
through final disposal. No other releases of LC compounds into the environment were identified
by the LCI data collected for the LCD, making it difficult to assess any possible exposures to
nearby populations.

4.3.4.3 Ecological populations

       Potential exposure to ecological populations could typically only occur through the
migration of LCs through the environment after being released during manufacturing or disposal
(either at LCD end-of-life, during disposal of broken or damaged panels during manufacturing, or
during disposal of containers or equipment contaminated with LCs). The potential for transport
of the LC through the environment is dependent on the identity of the chemical compound. The
number of choices of LC compounds along with the unavailability of data on disposal quantities
make it difficult to accurately assess the potential exposures which might result.

4.3.5  Human Health Effects

       There are at least ten meaningful groups of liquid crystal compounds representing several
hundred LC substances. Each of these substances is chemically unique, each having the potential
to  affect human health.  Because of the number of LC compounds potentially present in the LCD,
it is not possible to provide a comprehensive review of the human health effects of the universe
 of liquid crystal substances. However, a review of a small sample of LC compounds, those
 which appeared in the project LCI collected for the LCD, was conducted by EPA and the results
 presented below.
        Toxicological testing of LC substances and mixtures was conducted by three
 manufacturers responsible for roughly 90% of LC production. Testing of their chemical products
 included testing for acute toxicity effects, and effects on skin and eyes. Results of the testing
 indicated that of 588 LC substances tested, only twenty-five LC compounds had an LD505 less
 than 2,000 mg/kg of body weight (classified as exhibiting harmful effects to humans by the
        5 LD50 represents the dose of chemical which is lethal to 50% of the test population. By comparison, the
 LD50 for sodium chloride (table salt) is 3,000 mg/kg of body weight.

-------
                                                                    4.3 LIQUID CRYSTALS
 European Union), and only one had a LD50 of, less than 200 (classified as toxic by the European
 Union). The remaining 562 LC substances did not have any acute toxic potential.  Of the twenty-
 five harmful LCs, only twenty-two substances are produced and all are present at less than 10%
 concentration by weight, meaning that the resulting LC mixtures are not expected to exhibit
 harmful properties to humans (Becker, 2000). The remaining three harmful chemicals along with
 the lone toxic chemical were discontinued and excluded from further development. Several
 compounds, but still a minority, were found to be skin or eye irritants (Merck, 1999).

 4.3.5.1 Chronic effects (noncancert

        EPA conducted a review of existing toxicological data for the LC substances listed in
 Table 4-31. The review failed to identify a RFD, RFC, NOAEL, or LOAEL for any of the LC
 substances shown. This typically indicates that insufficient testing of these chemical compounds
 has been performed to accurately determine their potential for chronic human effects.

 4.3.5.2 Carcinogenicitv and mutagenicitv

        An EPA review of toxicological studies for the liquid crystals identified in the life-cycle
 inventory for LCDs failed to identify an existing slope factor for any of the LC compounds. A
 lack of carcinogenicity data usually does not indicate that a compound is not carcinogenic, but
 only that sufficient testing to ascertain carcinogenicity has yet to be performed.
       Toxicological testing of LC substances for mutagenic effects was conducted by three
 manufacturers responsible for roughly 90% of LC production. In a bacterial mutagenicity test of
 615 LC substances, one LC compound showed mutagenic potential, with the remaining
 compounds displaying no mutagenic effects. The lone chemical that failed the test was excluded
 from further development and was never marketed (Becker, 2000). Additionally, 10 LC
 compounds representing each of the significant groups of LCs, underwent mutagenicity testing
 using mammalian cells.  None of the tests indicated mutagenic activity.  Based on both sets of
 data, the manufacturer concluded that there does not appear to be a suspicion of mutagenic
 potential in the liquid crystals it produces (Merck, 1999).

 4.3.6  Environmental Regulations for Liquid Crystals

       No regulations exist specifically for liquid crystals compounds. However, regulations
may exist for individual liquid crystal compounds. Because of the number of possible LC
compounds available for use in a LCD, a comprehensive review of U.S. regulations could not be
provided.
                                         4-39

-------
4.4 CONCLUSIONS
4.4    CONCLUSIONS

       The purpose of this chapter was to provide a more detailed analysis of a few select
materials of interest to EPA and industry that was intended to better understand the potential
exposures and chemical risks to both human and ecological populations. The materials selected
for further analysis included lead, mercury, and liquid crystals, each selected for its known or
suspected toxicity to humans  and the environment, or because they are of particular interest to
indsutry or the U.S. EPA. The analysis of each material summarized or evaluated the following
key areas:

•      Use of the materials in computer displays;
•      Life-cycle inputs and outputs of the materials from computer displays;
•      Life-cycle impacts associated with the material inputs and outputs;
•      Potential exposures to the material including occupational, public, and ecological
      ' exposures;
•      Potential human health effects;
•      U.S. environmental regulations for the material; and
•      Alternatives to reduce the use of the material in computer displays.

The following are the conclusions drawn from the analyses of lead, mercury, and liquid crystal
use in the life cycle of both CRTs and LCDs.

4.4.1  Lead

       Lead is found in glass components of CRTs, as well as in electronics components (printed
wiring boards and their components) of both CRTs and LCDs.  It is also a top priority toxic
material at the U.S. EPA and the subject of electronics industry efforts to reduce or eliminate its
use. The following conclusions were drawn from a focused look at lead's role in the life cycle of
the computer display, and its effects on human health and the environment:

•      Due to the much greater quantity of lead in the CRT than the LCD, lead-based life-cycle
       impacts from the CRT ranged from moderately to significantly greater than those from
       the LCD in every category, with the exception of solid waste landfill use. The most
       significant difference was in non-renewable resource consumption, where the CRT
       consumed over 40 thousand times the mass of non-renewable resources over the course of
       its life cycle than those consumed by the LCD.  Other categories where CRTs had notably
       greater differences in impacts occurred in hazardous waste landfill use, chronic public
       health effects, and terrestrial toxicity.
•      Contributions of lead-based impacts are small relative to the total life-cycle impacts from
       other materials in the CRT (e.g., glass, copper wire), with the greatest impacts from lead-
       based CRT outputs occurring in the categories of non-renewable resources, aquatic
       toxicity, and chronic public health effects (ranging from 0.1 to 0.2% of the overall impact
       scores in each category).
•      For workers, inhalation is the most likely route of exposure to lead which may result in
       health concerns.  General population exposure to lead is most likely to come from
       incidental ingestion of lead in the soil, or ingestion of lead brought into the household on
                                          __

-------
                                                                       4.4 CONCLUSIONS
       workers clothing or on shoes.  Studies have discovered potentially high concentrations of
       lead in households within close proximity to certain facilities that use lead.
       Significant worker exposures to lead have been documented by existing studies of several
       processes which contribute to the life-cycle of the computer displays (e.g., lead smelting).
       These exposures have been as high as 90 times the OSHA recommended safety levels for
       exposure to workers at lead smelters. The resulting occupational chronic health effects to
       workers from lead exposure likely have been underestimated by the CDP LCIA
       methodology, which uses material inputs, and not outputs, as surrogates for exposure.
       Lead and lead compounds pose serious chronic health hazards to humans who may
       become over-exposed either in the workplace, or through the ambient environment. Lead
       exposure is associated with a range of adverse human health effects, including effects on
       the nervous system, reproductive and developmental problems, and cancer. Lead persists
       in the environment, but is relatively immobile in water under most surface and
       groundwater conditions.
       Alternatives are being developed, such as lead-free solders and glass components, that
       will potentially minimize the future lead content in both CRTs and LCDs.

4.4.2  Mercury

       Mercury is contained within the fluorescent tubes that provide the source of light in the
LCD. Mercury is also emitted from some fuel combustion processes, such as coal-fired
electricity generation processes, which contribute to the life-cycle impacts of both CRTs and
LCDs. EPA's concern with mercury and the potential for exposure during manufacturing and
end-of-life processes warranted a more detailed analysis of mercury in the CDP. The following
conclusions were drawn from a focused look at mercury's role in the life cycle of the computer
display, and its effects on human health and the environment:

•      The mercury emitted from the generation of power consumed by the CRT (7.75 mg)
       exceeds the entire amount of mercury emissions from the LCD, including both the
       mercury used in  LCD backlights (3.99 mg) and the mercury emissions from electricity
       generation (3.22 mg). Although this was not expected because mercury is used
       intentionally in an LCD, but not in a CRT, the results are not surprising since mercury
       emissions from coal-fired power plants are known to be one of the largest anthropogenic
       sources of mercury in the United States. Because the CRT consumes significantly more
       electricity in the  use stage than the LCD, its use stage emissions of mercury are
       proportionately higher than those of the LCD.
       Contributions from mercury-based impacts are not significant relative to the total life-
       cycle impacts from other materials (e.g., glass, copper wire) in the CRT or LCD, with the
       greatest impacts  from mercury-based outputs occurring in the aquatic toxicity category
       (0.4% for CRTs, 0.01% for LCDs)
      Possible pathways of worker exposure during backlight fabrication include inhalation of
      mercury vapors,  and dermal exposure or ingestion of mercury on skin. The most likely
      pathway for general population exposure is inhalation of mercury released into the air.
      Exposure data relevant to  the manufacturing of mercury backlights were not available,
      therefore specific conclusions about the potential magnitude of worker exposures could
      not be made.  Occupational chronic health effects to workers from mercury exposures

                                        4-41

-------
4.4 CONCLUSIONS
       calculated during the impact assessment (3.99e-06 tox-kg for LCD, none for CRT) likely
       have been underestimated by the CDP LCIA methodology, which uses material inputs as
       surrogates for exposure.
       Mercury and mercury compounds pose serious chronic health hazards to humans who are
       exposed. EPA has determined that mercury chloride and methylmercury are possible
       human carcinogens. Mercury poses serious chronic health hazards to humans, affecting
       the nervous system, brain, and kidneys.
       Alternative backlights have been developed that not only eliminate mercury from the
       light, but also improve on many of the optical characteristics of the displays. Current
       development is focused on improving the energy efficiency of the alternative lights.
4.4.3  Liquid Crystals

       Liquid crystals are organic compounds responsible for generating the image in an LCD.
LCs are not present in CRTs. The toxicity of the LCs in LCDs has been alluded to in the
literature, yet there is very little known about the toxicity of these materials. By including LCs in
a more detailed analysis, this section attempted to better characterize any potential hazard and/or
potential exposure of LCs from the manufacturing, use, and disposal of LCD monitors. The
following conclusions were drawn from a focused look at LCs role in the life cycle of the
computer display, and its effects on human health and the environment.

•      LCs are combined into mixtures of as many as 20 or more compounds selected from
       hundreds of potential liquid crystal compounds. Because of the possible variations in
       mixtures and the sheer number of compounds available, a select number of liquid crystals
       were used to assess potential human health hazards.
•      LCs do not appear to contribute significantly to any of the impact categories defined for
       this study. The total score for LCD occupational impacts based on potential worker
       exposure to LCs of 4.18 tox-grams, calculated using default toxicity values, represents
       less than 0.01% of the total overall chronic occupational health effects impact score of
       898 tox-kg for the functional unit of one LCD.
•      Impacts were not calculated for LC releases in the CDP LCIA because data regarding LC
       outputs were not available to the project.  LCs are not used to fabricate CRTs and so
       have no environmental impacts in the CRT life cycle.
•      Occupational exposures to LCs during the fabrication of the LCD panels are not expected
       to be significant. The enclosed nature of the chamber in which the LCDs are assembled,
       combined with the equipment (e.g., gloves, aprons) worn by workers in a clean room
       environment, are both expected to act to minimize exposures.  Other occupational
       exposures may exist that have not been identified.
 •      Toxicological testing by a manufacturer of LC substances and mixtures showed that
       95.6% (562 of 588) of the liquid crystals tested displayed no acute toxic potential to
       humans.  Twenty-five of the remaining twenty-six chemicals had the potential to exhibit
       harmful effects  to humans, while the remaining crystal was classified as toxic (EU
        classification) and thus was discontinued.  An EPA review of toxicity data for the
        confidential LC compounds was unable to identify any relevant toxicity information.
       Insufficient toxicity data exist to assess the toxicity of specific LC compounds.
                                          —            -                          -

-------
                                                               4.4 CONCLUSIONS
Testing for mutagenic and carcinogenic effects by the supplier showed that 99.9% (614
out of 615) of the liquid crystal compounds tested displayed no mutagenic effects. The
remaining chemical that showed mutagenic potential was excluded from further
development. Additionally, mutagenicity testing of ten LC substances using mammalian
cells showed no suspicion of mutagenic potential.
                                  4-43

-------
REFERENCES
                                   REFERENCES

ACGIH (American Conference of Governmental Industrial Hygienists).  1991. Documentation
       of the Threshold Limit Values and Biological Exposure Indices. Sixth edition. ACGIH,
       Cincinnati, OH.

ACGIH. 1998. Threshold Limit Values for Chemical Substances and Physical Agents
       Biological Exposure Indices. ACGIH, Cincinnati, OH.

ATSDR (Agency for Toxic Substances and Disease Registry). 1999. Toxicological Profile for
       Mercury. U.S. Department of Health and Human Services. Web site available at
       http://www.atsdr.cdc.gov/toxprofiles/tp46.html.  March.

ATSDR. 1999. Toxicological Profile for Lead. U.S. Department of Health and Human Services.
       Web site available at: http://www.atsdr.cdc.gov/toxprofiles/tp46.html. March

Barnes, D.G., and M. Dourson.  1988.  "Reference Dose (RfD):  Descriptions and Uses in Health
       Risk Assessments." Regulatory Toxicology and Pharmacology.  8: 471-486.

Bress, W.C., and Bidanset, J.H. 1991. "Percutaneous in vivo and in vitro absorption of lead."
       Veterinary and Human Toxicology. 33: 212-214.

Becker, W.  2000. "Results of the Toxicological Testing of Liquid Crystal Compounds."
       Presented at International Display Works Conference, Kobe, Japan. December.

Becker, W.  2001. Personal communication by e-mail with W. Becker, Merck, Liquid Crystals
       Division, and M. Socolof, University of Tennessee, Center for Clean Products and Clean
       Technologies. December 6.

Busio, M. and Steigelmann, O.  2000. "New Frit Glasses for Displays." Glasstech.  Berlin Glass
       Science and Technology. 73(10): 319-325.

Doggett, T.  2001.  "EPA Seeks to Dismiss Utility Lawsuit over Mercury." Reuters press release
        dated April 12,2001 accessed at
       http://www.planetark.org/dailynewsstory.cfm?newsid=10496&newsdate=12-Apr-2001.
        September 2.

Donofrio, R. and Eckel,  R. 1999. "Arc Lamps for Microelectronics." Presented at Vehicle
        Displays and Microsensors '99 SID Symposium, pp.103-111.

EIAJ (Electronic Industries Association of Japan).  1996.1996 Update to: Research Report on
        the Visions of the Electronic Display Industry to the Year 2000.  Translated and published
        in English by MerLingus.
                                          4-44

-------
                                                                         REFERENCES
 EPA (Environmental Protection Agency). 1986. EPA Air Quality Criteria for Lead. Publication
       No. EPA 600/8-83-028af. U.S. EPA, Office of Research and Development, Office of
       Health and Environmental Assessment.  Research Triangle Park, NC.

 EPA. 1998. Fact Sheet:  "Utility Air Toxics Report to Congress."  U.S. EPA, Office of Air
       Quality, Planning, and Standards. Web site available at:
       http://www.epa.gov/ttn/oarpg/utilfs.html. February 25.

 EPA. 1999. "Proposed Rule Change; Lead and Lead Compounds, Lowering of Reporting
       Thresholds, Community Right-to-Know  Toxic Chemical Release Reporting." Federal
       Register: Volume 64, Number 148. U.S.EPA, Office of Pollution Prevention and
       Toxics, Washington, D.C. August.               '.......

 EPA. 2000a. "Mercury Study Report to Congress: White Paper."  U.S. EPA Office of Air and
       Radiation. Web site available at  http://www.epa.gov/oar/merwhite.html.  (The full eight-
       volume 1997 Mercury Report to Congress is available online at:
       http://www.epa.gov/airprogm/oar/mercury.html).  September 26

 EPA. "2000b. "EPA Decides Mercury Emissions from Power Plants Must be Reduced."  U.S.
       EPA Headquarters Press Release. December 14.

 EPA. 2001a. "Mercury Study Report to Congress: Overview." EPA Office of Air and
       Radiation. Web site available at: http://www.epa.gov/oar/mercover.html. (The full eight-
       volume 1997 Mercury Report to Congress is available online at:
       http://www.epa.gov/airprogm/oar/mercury.html). February 8.

 EPA. 2001b. "Persistent and Bioaccumulative Toxic Chemical Program's Chemical List." U.S.
       EPA, Office of Pollution Prevention and  Toxics. Web site available at:
       http://www.epa.gov/opptintr/pbt/cheminfo.htm. March 8.

 EPA. 200 Ic. "National Advisory on Mercury in Fish Caught by Family and Friends: for Women
       Who Are Pregnant or May Become Pregnant, Nursing Mothers, and Young Children "
       U.S, EPA, Office of Water, EPA-823-F-01-004. January, 2001.

FDA (U.S. Food and Drug Administration.).  2001. "Consumer Advisory: an Important Message
       for Pregnant Women and Women of Childbearing Age Who May Become Pregnant about
       the Risks of Mercury in Fish." U.S. Food and Drug Administration, Center for Food
       Safety and Applied Nutrition. March.

Gosselin, R.E., R.P. Smith, andH.C. Hodge.  1984. Clinical Toxicology of Commercial Products.
       5th ed. Williams and Wilkins, Baltimore, MD.

HSDB (Hazardous Substance Data Base). 2001.  MEDLARS Online Information Retrieval
       System. National Library of Medicine, National Toxicology Program (via TOXNET).
       Bethesda, MD.
                                        4-45

-------
REFERENCES
IARC (International Agency for Research on Cancer). 1987. IARC Monographs on the
       Evaluation of the Carcinogenic Risk of Chemicals to Humans: Overall Evaluations of
       Carcinogenicity. Supplement 7.  World Health Organization, Lyon, France.

IRIS (Integrated Risk Management System). 1999. U.S. Environmental Protection Agency.
       National Center of Environmental Assessment. Office of Research and Development.
       Cincinnati, OH.

Keenan, C. and Kellett, TV 2001. Personal communication with C. Keenan and T. Kellett, Abt
       Associates Incorporated, and K. Hart, U.S. EPA. Background Research on Lead-Free
       Solder Alternatives.

Lee, C., Chang, S., Wang, K., and Wen, L. 2000. "Management of Scrap Computer Recycling in
       Taiwan." Journal of Hazardous Materials.  A73: 209-220.

Menad, N. 1999.  "Cathode Ray Tube Recycling."  Resources, Conservation, and Recycling.
       26: 143-154.

MERCK.  1999.  "Toxicological Investigations of Liquid Crystals." Presented at 28th Freiburger
       Arbeitstagung.  March 24-26.

Monchamp, A., Evans, H., Nardone, J., Wood, S., Proch, E., Wagner, T. 2001. "Cathode Ray
       Tube Manufacturing and Recycling:  Analysis of Industry Survey." Electronic Industries
       Alliance.  Spring.

Musson, S., Jang, Y., Townsend, T., and Chung, L. 2000. "Characterization of Lead Leachability
       from Cathode Ray Tubes Using the Toxicity Characteristic Leachate Procedure."
       Environmental Science and Technology. 34: 4376-4381.

NIOSH (National Institute for Occupational Safety and Health).  1997. NIOSH Pocket Guide to
       Chemical Hazards.  Publication No. 97-140. U.S. Government Printing Office.
       Washington D.C.

OSRAM.  1998. "The World's First Mercury-Free Flat Lamp." Web site available at:
       http://www.osram.com/txt/aktuell/planone.htm.  July 13.

Sittig, M.  1985. Handbook of Toxic and Hazardous Chemicals and Carcinogens. 2nded.
       Noyes Data Corporation, Park Ridge, NJ.

Srivastava, A.M. and Sommerer, T.J. 1998. "Fluorescent Lamp Phosphers." Electrochemical
       Society Interface. Summer.

Techneglas,  Incorporated. "Frit Facts: A Brief Technological Summary of Televison Solder
       Glass for CRT Technicians, Engineers, and Managers", downloadable fact sheet.
       http://www.techneglas.com.
                                         4-46

-------
                                                                          5.1 LCI vs. LCIA
                                        Chapters

                           SUMMARY AND CONCLUSIONS

        The purpose of the CDP, as stated in Chapter 1, is to provide a scientific baseline of life-
 cycle environmental impacts of CRTs and LCDs, help manufacturers identify areas to focus
 improvement assessment activities, and to develop a life-cycle model for future analyses.  The
 primary targeted audience is the electronics industry, for whom results may provide insight into
 improvement opportunities in the life cycle of CRTs and/or LCDs. In addition, the general
 public may also find results useful when considering environmental impacts of each display type.
 This chapter briefly summarizes the results and draws conclusions based on those results.  This
 report, however, does not include direct comparative assertions or improvement assessments
 based on the results.  Alternatively, results and conclusions are described in terms of the overall
 LCI versus the LCIA, and details of the impact assessment, including the additional assessments
 of lead, mercury and liquid crystals, and the sensitivity analyses.  Major uncertainties, cost and
 performance considerations, suggestions for improvement opportunities, and suggestions for
 further research are also provided.

 5.1    LCI vs. LCIA

       In this LCA, a life-cycle inventory (LCI) was compiled from many data sources,
 including both primary and secondary data sources. The primary data were obtained from
 component and monitor manufacturers  of CRTs and LCDs. In an LCA, inventory data provide
 information on how much material is being consumed in the life cycle (i.e., inputs) and how
 much material is generated/released (i.e., outputs). The LCI results of this report are detailed in
 Chapter 2.  The LCI provides inventory data grouped by inventory type (e.g., primary material,
 energy,  air emission, solid waste).
       The LCI alone, however, does not always translate directly into impact categories that
 may be of interest.  That is, a given amount of one material may have different impacts (for a
 certain impact category) than the same amount of another material. Furthermore, some materials
 may affect more than  one impact category. For example, an air emission could affect air
 acidification as well as being toxic to humans breathing it. Therefore, a life-cycle impact
 assessment (LCIA) is conducted to reveal potential impacts in several impact categories. In this
 CDP LCIA, described in detail in Chapter 3, impacts are sometimes driven by materials other
 than the top inventory contributors.  For example, the top air emission for LCDs is carbon
 dioxide (Table 2-49),  however the greatest global warming impact score is from SF6 in the LCD
 monitor/module manufacturing process (Table 3-25).
      To illustrate that the inventory results may not directly translate into impact results, the
 first two columns in Table 5-1 show which monitor has greater inventory amounts for each
 inventory type in the LCI,  and the last two columns show which monitor has greater impact
 scores for each impact category in the LCIA. The impact categories that are affected by each
inventory type are in the same rows as the associated inventory type. As seen in Table 5-1, some
impact categories associated with ancillary material and water pollutant inventory types had
opposing outcomes in the LCI versus the LCIA. For example, the three impact categories
affected by the ancillary material inventory had greater impacts for the CRT, although the
                                          5-1

-------
5.1 LCI vs. LCIA
ancillary material inventory had greater amounts of inputs for the LCD. In this case, both
primary and ancillary materials contribute to the impact categories, causing differing results.
       Considering the wastewater outputs, which are greater for the LCD than the CRT, the
impacts related to water releases are in some cases greater for the CRT than the LCD.  Note that
although the wastewater volume is greater for the LCD, the total mass of water pollutants in the
LCI is greater for the CRT (see Table 2-24). In the LCIA, the LCD has greater impacts for water
eutrophication and aquatic toxicity, but not for the two water quality categories (BOD  and TSS),
chronic health effects to the public, nor terrestrial toxicity, all of which include water emissions
in calculating the impact score.
Table 5-1. Baseline LCI vs. LCIA: monitor with greater inventory amount and impact
Baseline life-cycle inventory (LCI)
Inventory type
Primary materials
Ancillary materials
Water inputs
Fuel inputs
Electricity inputs
Total energy inputs
Air pollutant outputs
Monitor with
greater
inventory
results
CRT

LCD
CRT
CRT
CRT
CRT
CRT
Baseline life-cycle impact assessment (LCIA)
Potential impact category(ies)
associated with inventory type
Renewable resource use
Nonrenewable resource use
.Chronic health effects, occupational
Renewable resource use
Nonrenewable resource use
Chronic health effects, occupational
Renewable resource use
Energy use
Chronic health effects, occupational
Energy use
Energy use
Global warming
Ozone depletion
Photochemical smog
Acidification
Air particulates
Chronic health effects, public
Aesthetics (odor)
Terrestrial toxicity
Monitor with
greater
impact
results
CRT
CRT
CRT
CRT
CRT
CRT
CRT
CRT
CRT
CRT
CRT
CRT
a
CRT
CRT
CRT
CRT
CRT
CRT
                                           5-2

-------
                                                                                  5.1 LCI vs. LCIA
   Table 5-1. Baseline LCI vs. LCIA: monitor, with greater inventory amount and impact
Baseline life-cycle inventory (LCI)
Inventory type
Wastewater outputs
Water pollutant outputs
Hazardous waste outputs
Solid waste outputs
Radioactive waste outputs
Radioactivity outputs
Monitor with
greater
inventory
results
LCD
CRT
CRT
CRT
CRT
CRT
Baseline life-cycle impact assessment (LCIA.)
Potential impact category(ies)
associated with inventory type
none
Water eutrophication
Water quality, BOD
Water quality, TSS
.Chronic health effects, public
Aquatic toxicity
Terrestrial toxicity
Hazardous waste landfill use
Solid waste landfill use
Radioactive waste landfill use.
Radioactivity
Monitor with
greater
impact
results
NA
LCD
CRT
CRT
CRT
LCD
CRT
CRT
CRT
CRT
CRT
 The LCIs for both the CRT and LCD contain data for substances that were phased out of production by 1996 due
to their ozone depletion potential. Whether these emissions still occur in countries that were signatories to the
Montreal Protocol and its Amendments and Adjustments (such as the United States and Japan) is not known, but
considered to be unlikely.  When phased-out substances are included in the inventory, the CRT has greater ozone
depletion impacts than the LCD. However, if phased-out substances are removed from the inventories, the results
are switched, with the LCD having greater impacts.
                                              5-3

-------
5.2 LCIA RESULTS
5.2    LCIA RESULTS
                                    V
5.2.1  CRT and LCD Baseline Results

       The LCIA results, presented in detail in Chapter 3, showed that the CRT has greater total
life-cycle impact indicators in most of the impact categories (see Table 3-10). In the baseline
scenario, the CRT has greater impacts than the LCD in all but two impact categories
(eutrophication and aquatic toxicity). However, note that for the ozone depletion category, the
LCIs for both the CRT and LCD contain data for substances that were phased out of production
by 1996 due to their ozone depletion potential. Whether these emissions still occur in countries
that were signatories to the Montreal Protocol and its Amendments and Adjustments (such as the
United States and Japan) is not known, but considered to be unlikely. When phased-out
substances are included in the inventory, the CRT has greater ozone depletion impacts than the
LCD.  However, if phased-out substances are removed from the inventories, the results are
switched, with the LCD having greater impacts.
       When considering which life-cycle stage has greater impacts, the LCIA results showed
that the rnanufacturing life-cycle stage dominates impacts for most impact categories for both the
CRT and LCD (refer to Section 3.3). Table 5-2 summarizes which life-cycle stages have the
greatest impacts for each impact category for the CRT and LCD.  As shown in Table 5-2, the
CRT has nine and the LCD has 11 impact categories with greatest impacts from the
manufacturing life-cycle stage. Only six categories (solid waste landfill use, global warming,
ozone depletion, acidification, air particulates, and chronic public health) have greatest impacts
from the CRT use stage, and four categories (solid waste landfill use, acidification, air
particulates, and chronic public health) have greatest impacts from the LCD use stage. The CRT
has three categories with greatest impacts from the upstream life-cycle stage and the LCD has
three.  The end-of-life (EOL) life-cycle stage is greatest for the same two impact categories for
both the CRT and LCD (hazardous waste landfill use and radioactive waste landfill use). Note
that the EOL stage impacts are generally very small contributors to the overall impacts. This is
likely because of the small inventories associated with the EOL processes, but also may be a
function of the incomplete and/or secondary data for the EOL (i.e., no remanufacturing data, and
secondary data not completely specific to the monitors evaluated in this study).
       A more detailed evaluation of lead, mercury, and liquid crystals was completed in
Chapter 4. As expected, the CRT, which has lead in the glass, frit, and printed wiring boards
(PWBs), has greater impacts from lead than did the LCD, which only has lead in the PWBs.
Regarding mercury, there were greater inventories of mercury in the CRT life cycle than in the
LCD life cycle, despite the fact that only the LCD has mercury directly in the product. The
greater amount of mercury is from the release of mercury and mercury compounds from the
generation of electricity.  And as the CRT life cycle uses more electricity than the LCD, there
was a greater quantity of mercury releases reported for the CRT than the LCD. Liquid crystals
are only found in LCDs, and therefore, there are no associated impacts for the CRT. Little
conclusive information was available on the liquid crystal materials. A detailed literature search
was conducted, however very little data were available on the toxicity of these materials.  Based
 on the limited toxicity data obtained, liquid crystals currently do not appear to be a significant
human health or environmental hazard in the LCD life cycle. However, there were insufficient
toxicity data available to make a definitive conclusion about liquid crystal toxicity.
                                           5-4

-------
                                                                        tLCIA RESULTS
          Table 5-2. Monitor
                         and
type with greatest impacts for each life-cycle stage
impact category (baseline scenario)
Impact category
Renewable resource use
Nonrenewable resource use
Energy use
Solid waste landfill use
Hazardous waste landfill use
Radioactive waste landfill use
Global warming
Ozone depletion
Photochemical smog
Acidification
Air particulates
Water eutrophication
Water quality, BOD
Water quality, TSS
Radioactivity
Chronic health effects, occupational
Chronic health effects, public
Aesthetics (odor)
Aquatic toxicity
Terrestrial toxicity
TOTALS
Monitor type with greatest impacts
Upstream
LCD




"-


LCD


CRT


CRT, LCD



CRT

CRT=3
LCD=3
Manufacturing
CRT
CRT, LCD
CRT, LCD



LCD
LCD
CRT


LCD
CRT, LCD
CRT, LCD

CRT, LCD

CRT, LCD
LCD
CRT, LCD
CRT=9
LCD=11
Use



CRT, LCD


CRT
CRT

CRT, LCD
CRT, LCD





CRT, LCD



CRT=6
LCD=4
EOL




CRT, LCD
CRT, LCD














CRT=2
LCD=2
5.2.2   CRT Results

       For the CRT, many of the impacts were driven by a single material in the inventory. As
stated in Section 3.3.15 and shown in Table 3-57, in 14 of the 20 impact categories, the top
individual contributor to the impacts was responsible for greater than 50% of the impacts. This
shows that the CRT data are highly sensitive to a few data points. Major conclusions from the
CRT LCIA are as follows:

       Energy used in glass manufacturing and associated production of LPG are driving the
       baseline CRT results (they dominate ten impact categories, including overall life-cycle
       energy use).
                                         5-5

-------
5.2 LCIA RESULTS
•      The large amounts of fuel used as energy sources are driving occupational health effects.
       Occupational impacts are calculated from inventory input amounts, and therefore there
       may or may not actually be exposure to these fuels (e.g., they may be contained);
       however, the results illustrate the potential for health effects, especially under spill or
       upset conditions.
•      The generation of electricity for the use stage dominates seven impact categories.
•      Air emissions of sulfur dioxide from electricity generation (for the use life-cycle stage)
       drive chronic public health effects, acidification, and terrestrial toxicity impacts. This
       may be a concern, for example, in areas in nonattainment of regulated levels of sulfur
       dioxide in the United States.

       The use of LPG fuel in glass manufacturing dominated ten impact categories:  two
directly from the LPG used in glass/frit manufacturing (energy use impacts and chronic
occupational health effects) and eight from LPG production (renewable resource use,
nonrenewable resource use, photochemical smog, air particulates, water eutrophication, BOD
water quality, TSS water quality, and aesthetics).  In addition, impacts from the generation of
electricity during the use stage dominated seven impact categories:  solid waste landfill use,
radioactive waste landfill use, global warming, ozone depletion, acidification, chronic public
health, and terrestrial toxicity.  The CRT tube manufacturing process, which represents the most
functionally and physically (by mass) significant component of the CRT monitor, only dominated
one impact category (aquatic toxicity). Twenty-six percent of the aquatic toxicity score was from
phosphorus outputs from tube manufacturing, while most of the rest were from the materials
processing life-cycle stage. The remaining two impact categories (hazardous waste landfill use
 and radioactivity) had greatest impacts from the landfilling of the assumed hazardous proportion
 of CRT monitors, and the release of Plutonium-241 in steel production, respectively
 (Table 3-57).  The radioactivity impacts are driven by the radionuclide Pu-241, due to the electric
 grid inventory included in the steel production secondary data set, which includes nuclear fuel
 reprocessing.
        The large amount of LPG reported for glass manufacturing was originally questioned
 during the data collection and verification stage of this project. While no compelling reason
 could justify removing the LPG data in the baseline case, a sensitivity analysis was conducted in
 which the glass energy data were modified.  Other sensitivity analyses were also conducted (i.e.,
 manufactured life, modified LCD monitor manufacturing energy, and modified LCD EOL
 distributions). However, the only scenario that substantially altered the comparative results was
 the modified glass energy scenario (see Table 3-62 and Section 3.4.5).
        The overall energy in the baseline scenario was nearly seven times greater than that in the
 modified scenario (from 20,800 MJ to 3,020 MJ), and the amount of LPG dropped to zero, while
 other energy sources increased. The basis for the modified data was removing the energy inputs
 from one suspect data set.  As a result, the CRT modified glass energy scenario had greater
 energy use impacts in the use stage than in the manufacturing stage for the CRT. The amount of
 LPG used in glass manufacturing in the baseline scenario is 351 kg/monitor of LPG, which alone
                                            5-6

-------
                                                                           ', LCIA RESULTS
 costs about $71.' This is a significant amount of the cost.of a complete CRT monitor (the range
 of a few currently selling 17" CRTs is $158-116, and the average cost from primary data
 collected in the CDP was $541, which are presented in Section 5.4).  Therefore, it is likely that
 the actual energy inputs to the glass manufacturing process is somewhere between the baseline
 and modified glass scenarios. In conclusion, more information is needed on energy used in glass
 manufacturing, which is driving CRT baseline results.
        The additional analyses for the CRT of lead and mercury also revealed that the use of lead
 could present health risks, but the method of using only inputs to evaluate occupational impacts
 (see Section 3.1.2.13) may not adequately represent occupational exposures and risks. Further
 refinement of the occupational impact analysis may be warranted.
        Although there is no mercury in the CRT monitor, mercury emissions from electricity
 generation in the CRT life cycle were greater (in mass) than the mercury used in the LCD.
 Therefore, to reduce mercury emissions from the CRT life cycle, efforts to reduce electricity
 consumption could be taken. Additionally, changes to the electric grid could also reduce
 mercury emissions from the CRT life-cycle.

 5.2.3   LCD Results

        The LCD impact results were less sensitive to an individual input or output than the CRT
 results, although in 11 of the 20 impact categories an individual input was still responsible for
 greater than 50% of the total impacts (Table 3-58). In general, the LCD results are less uncertain
 than the CRT results. This is because most of the CRT results are being driven by either glass
 input data or data from secondary sources, while LCD impacts are being driven more by data
 from primary sources. Some results to note are as follows:

        The LCD monitor/module manufacturing process group had greatest impacts in six
        impact categories (Table 3-58).
        Although the top contributor to the energy impact category was electricity consumed in
        the use stage (30%), the overall energy impacts were greater from the manufacturing
        stage than the use stage.
        In the glass energy sensitivity scenario, the use stage had greatest energy impacts,
        although only by a small margin over the manufacturing stage (see Figure 3-26).'
        Sulfur dioxide [emitted from electricity generation  for the use stage, and constituting only
       0.37% of the air emission inventory (see Table 2-49)] dominates the acidification, chronic
       public health, and terrestrial toxicity impact categories (Table 3-58).  The high public
       health and terrestrial toxicity scores are due to its low non-cancer toxicity value and
       resulting high hazard value (HV).
       Sulfur hexafluoride (SF6) from LCD monitor/module manufacturing was the single
       greatest contributor to the global warming impact score; however, carbon dioxide from
       the use stage and the materials processing stage also contributed significantly to the
       global warming impacts (Table 3-25).
         Based on a "daily market price" on August 29,2001, of $0.4160/gallon of LPG
(http://w\vw.americanpowernet.com/pub_energy/futures.html). For 351 kg/functional unit in the CRT
manufacturing life-cycle stage, the cost is about $71 per functional unit (i.e., one monitor), assuming a density of
LPG of 2.053 kg/gallon. For the LCD, the 16.8 kg/functional unit of LPG would cost about $3.40 per monitor

-------
52 LCIA RESULTS
•      The glass energy inputs did not directly dominate any impact categories, as they did for
       the CRT (due to the smaller mass of glass in the LCD); however, LPG production
       (required for the glass energy fuel) dominated two categories: TSS water quality and
       aesthetics (Table 3-58).
•      LNG as an ancillary inventory material was questionably very large and had greatest
       impacts in two categories:  nonrenewable resource use and photochemical smog
       (Table 3-58; shown there as "Natural Gas Production" due to that process being used as a
       surrogate for LNG production).

       The additional analyses of lead, mercury, and liquid crystals showed that the LCIA alone
is not adequate enough to determine all the potential impacts within the life-cycle of the LCD
monitors. Similar to the conclusion for the CRT, lead-based occupational impacts would require
further refinement of the LCIA methodology. The LCIA method in this LCA used inputs as
surrogates for occupational exposure. There are outputs, within the occupational setting, that
should also be considered.
       For mercury, which is found in the backlights of the LCD monitors, there is nearly the
same amount of mercury by mass emitted to the air during electricity generation as there is
mercury used to make the backlight unit. The mass of mercury input for backlights is only about
20% greater than the mercury air emissions from electricity generation (across all life-cycle
stages).
       Liquid crystals were also identified by the CDP Core Group as a material for which
additional information would be reviewed.  The LCIA did not find the liquid crystals to be
significant contributors to any impact categories; however, this could partially be due to the lack
of information on them.  The additional analysis also revealed limited information, but
qualitatively, did not show significant potential risk.

5.2.4  CRT vs. LCD Sensitivity Analysis Results

       The only sensitivity analysis to show significant difference in the results was the modified
 glass energy scenario. In comparing the CRT and LCD, the CRT baseline scenario had greater
 impacts than the LCD in all but two impact categories (eutrophication and aquatic toxicity) and
 possibly three (ozone depletion). In the modified glass energy scenario, nine of the 20 categories
 had greater impacts from the LCD life-cycle than the CRT. Energy use remained greater for the
 CRT; however, nonrenewable resource use, global warming, photochemical smog,
 eutrophication, BOD and TSS water quality, chronic occupational health effects, and aesthetics
 all reversed such that the LCD had greater impacts than the CRT (Table  3-62).  As stated above,
 it is believed that a more true representation of the monitor life  cycles lies somewhere between
 the baseline and modified glass energy scenario. Further work is recommended in clarifying and
 refining glass energy input information.
                                           5-8

-------
                                                                    5.3 UNCERTAINTIES
 5.3    UNCERTAINTIES

       As with any LCA, it is not uncommon for there to be uncertainty associated with such a
 large data collection effort. Two of the largest sources of uncertainty in mis LCA that have a
 significant effect on the results are as follows:

 •      CRT and LCD glass manufacturing energy inputs (from primary data): The larger
       amount of glass used in CRTs than LCDs results in the CRT having greater associated
       uncertainty than the LCD results.
 •      Secondary data for upstream and fuel production processes: When any one material is
       used in the life-cycle of either monitor in large quantities, the impacts associated with the
       inputs and outputs from the production of that material may become significant. For
       example, LPG and LNG production were both used in significant enough amounts to
       influence some impact categories.  Therefore, the uncertainty in the secondary data
       becomes important. This highlights the need for a consistent, national (or international)
       LCI database that is updated regularly.

       Other uncertainties associated with individual data points collected from primary data
 sources may be found in the data for this analysis. However, they had less effect on the overall
 results than the uncertainties mentioned above. For manufacturers interested in conducting
 improvement assessments, closer review of such uncertainties may be warranted.
       Other uncertainties in the LCA pertain to uncertainties inherent in LCIA methodology.
 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. Uncertainties are inherent in
 each impact category, and the reader is referred to the baseline LCIA results in Section 3.3 for a
 detailed discussion of uncertainties by impact category.
       Another point that should be recognized in the overall comparison of CRTs and LCDs is
 that CRTs are a more mature technology than LCDs. Changes in LCD manufacturing processes
have likely occurred during the development and publication of this report. Therefore,
comparisons must be carefully drawn when evaluating the mature CRT to the newer LCD
technology.
                                         5-9

-------
5.4 COST AND PERFORMANCE CONSIDERATIONS
5.4    COST AND PERFORMANCE CONSIDERATIONS

       The focus of this study has been on the environmental effects associated with CRTs and
LCDs. The environmental attributes or burdens of a product are not expected to be considered
alone when evaluating the marketability and commercial success of a product. The cost and
performance of each monitor type are obviously critical components to a company's or
consumer's decisions of whether to produce or purchase a product. This section briefly
addresses a few direct costs associated with the monitors. A complete cost analysis, including all
direct costs (e.g., material costs) and indirect costs (e.g., environmental costs to society) are
beyond the scope of this report.  Direct retail costs of the monitors and electricity costs are
presented herein.
       The average retail price of 1997-2000 model year monitors, collected from the
manufacturers who supplied data for this project, as well as the performance information, are
presented in Table 5-3. Costs collected from current monitors on the market are presented in
Table 5-4. From Table 5-3, which represents primary data collected on the actual monitors
included in this study, the LCD is approximately 2.7 times more costly. More recent data show
that prices have come down, and the difference in prices between the CRT and LCD has also
been reduced.

 Table 5-3. Primary cost and performance data collected from manufacturers for the CDF
Monitor

CRT monitor
(functional unit
aggregate)
LCD monitor
(functional unit
aggregate)
Display
Size
(inches)
17
15
Resolution
(pixels)
1024x768
1024x768
Brightness
range
(cd/m2)
86-154
200-300
Contrast ratio
range

~
200:1-300:1
Number of
Colors

"Full color"
"Full color"
Average cost
from primary
data
(US$)
$541
$1,450
— Not reported or not applicable.

       A complete cost analysis would require assessing the costs from each life-cycle stage.
The costs presented above are retail costs that presumably represent the manufacturing costs, but
probably not external environmental costs, for example. The costs from the use stage can be
represented by the electricity costs during the use stage. The average cost of residential and
commercial electricity in the United States is approximately $0.021/MJ,2 and the CRT and LCD
monitors use about 2,290 and 853 MJ/functional unit, respectively, in the use stage baseline
scenario, which assumes a total of 13,547 hours per life over a period of 6.5 years (see Section
2.4.1.2). Therefore, the electricity costs to consumers during the use stages are $48 for the CRT
and $18 for the LCD. The amount of electricity consumed and the associated cost of that
electricity for each life-cycle stage in the baseline scenario are presented in Table 5-5.
       2 This number was calculated from a value found at the following Web address:
 www.eia.doe.gov/cneaf/electricity/ esr/tl 1 .txt.
                                           5-10

-------
                                                 5.4 COST AND PERFORMANCE CONSIDERATIONS
Table 5-4. Cost and performance data for some currently selling CRTs and T,rr>sa
Monitor

CRTs
Monitor 1
Monitor 2
Monitor 3
LCDs
Monitor 1
Monitor 2 *
Monitor 3
Monitor 4 *
Monitor 5
Display
Size
(inches)

17/16
17/16.1
17/16

15.1
15.1
15.1
15.1
15
Resolution
(pixels)
Brightness
(cd/m2)
Contrast ratio


1280x1040
1280x1040
1600x1200

—
—

"High contrast, anti-static,
anti-glare coating."
—

1024x768
1024x768
1024x768
1024x768
1024x768
—
200
200 c
200
210 c
200:1
250:1
200:1 c
250:1
350:1
Number of
Colors



—


16.7 million
16.7 million
16+ million
16.7 million

2001 Cost
($US)

$158
$171
$316

$349
$400
$439
$499
$554.
 b                                  	», except where noted otherwise.
 8/29/of°m ±G manufacturer's Web site excePt for Prices'which where obtained from http://www.cdw.comon
 c Data are from the manufacturer's Web site.
 — Not reported or not applicable.

Life-cycle stage
Upstream
Manufacturing
Use
EOL
Total
Table 5-5. Life-cycle electricity costs (baseline scenario)

Electricity use
(MJ/fimctional
unit)
(see Table J-3)
73.2
129
2,290
0.229
2,492
CRT
unit cost
($/MJ)
0.012 ° .
0.012°
0.021 *
0.012"

Cost
($US)
$1.3
$1.5
$48
$0.003
$51
LCD
Electricity use
(MJ/functional
unit)
(see Table J-12)
8.55
278
853
0
1,140
unit cost
($/MJ)
0.012°
0.012 °
0.021 b
0.012°

Cost
($US)
$0.10
$3.4
$18
0
$22
               •—            '	j — — -——««.*»** UWI.V.L AO H'v.xf-i—r.wft. *» 11. ,fiooujj3lfl2 3.0 ivjLj/jfctvvn
($0.0443/kWh)/(3.6MJ/kWh) = $0.012/MJ. Source: www.eia.doe.gov/cneaf/electricity/esr/tll.txt. Note that the
use of the U.S. average cost is simply to compare costs among life-cycle stages, although the actual costs would be
mostly from Asia.
b 1999 U.S. average cost of electricity for the residential and commercial sectors is $0.0771/kWh  Assuming 3 6
MJ/kWh, ($0.0771/kWh)/(3.6 MJ/kWh) = $0.021/MJ. Source: www.eia.doe.gov/cneaf/electricity/esr/tll.txt.'
                                                5-11
   impacts, while keeping in mind the data limitations and uncertainties, as well as cost and
   performance considerations.
                                                   5-13

-------
 5.4 COST AND PERFORMANCE CONSIDERATIONS
        The LCA is defined such that the monitor assessments are performed on a functionally
 equivalent basis. To the extent possible, data were collected on functionally equivalent monitors.
 The data presented in Table 5-3 are the range of specifications provided by the monitor
 assemblers, but not necessarily from the manufacturers of all of the component parts. Therefore,
 we are unsure if the specifications provided by the monitor assemblers also represent those of the
 component parts, since the component parts manufacturers did not consistently supply
 performance data as requested in the data collection questionnaire used in this study. However,
 when companies were approached to participate in the study, they were informed of the
 performance specification parameters within which the study boundaries were defined.
 Therefore, it is assumed that they meet the specifications as presented in Chapter 1 (Table 1-2)
 and in Table 5-3 and they perform relatively equivalently. In the primary data, the reported
 brightness of the CRT was less than the LCD, otherwise, they are functionally similar.
5.6 SUGGESTIONS FOR FUTURE RESEARCH
5.6    SUGGESTIONS FOR FUTURE RESEARCH

       Areas where future research could be conducted to refine and/or continue the use of the
results in this study are as follows:

•      gather more information on energy use in glass manufacturing;
•      develop consistent materials and fuel processing data in a national (or international) LCI
       database that is updated regularly;
•      refine and/or update some of the LCD manufacturing data (e.g., LNG data);
«      collect more complete EOL data (e.g., remanufacturing data, and primary data for
       incineration and landfilling) to determine better representation of the EOL impacts;
•      conduct more research on EOL options for LCDs;
"      collect more detailed data on landfilling and other treatment processes, such as water
       treatment where no impacts were calculated;
•      update manufacturing  data to meet more recent monitor model years;
•      conduct a more focused analysis on selected areas for detailed improvement assessments;
       and
•      evaluate process changes or other alternatives against an "average 1997-2000 model year"
       to evaluate impacts of changes or improvements over time.
                                         5-14

-------
                                                5.4 COST AND PERFORMANCE CONSIDERATIONS
Table 5-4. Cost and performance data for sonii currently selling CRTs and LCDs0
Monitor

Display
Size
(inches)
Resolution
(pixels)
Brightness
(cd/m2)
Contrast ratio

Number of
Colors

2001 Cost
($US)
CRTs
Monitor 1
Monitor 2
Monitor 3
17/16
17/16.1
17/16
1280x1040
1280x1040
1600x1200
—
—
—
—
"High contrast, anti-static,
anti-glare coating."
—
	
—
	
$158
$171
$316
LCDs
Monitor 1
Monitor 2 *
Monitor 3
Monitor 4 *
Monitor 5
15.1
15.1
15.1
15.1
15
1024x768
1024x768
1024x768
1024x768
1024x768
—
200
200 c
200
210 c
200:1
250:1
200:1 c
250:1
350:1
16.7 million
16.7 million
16+ million
16.7 million
—
$349
$400
$439
$499
$554.
' All information from Vol. EC23 of the eCOST.com catalog, except where noted otherwise.
6 Data from the manufacturer's Web site except for prices, which where obtained from http://www.cdw com on
8/29/01.
c Data are from the manufacturer's Web site.
— Not reported or not applicable.

Life-cycle stage
Upstream
Manufacturing
Use
EOL
Total
Table 5-5. Life-cycle electricity costs (baseline scenario)
CRT
Electricity use
(MJ/functional
unit)
(see Table J-3)
73.2
129
2,290
0.229
2,492
unit cost
($/MJ)
0.012"
0.012°
0.021 *
0.012"
—
Cost
($US)
$1.3
$1.5
$48
$0.003
$51
LCD
Electricity use
(MJ/functional
unit)
(see Table J-12)
8.55
278
853
0
1,140
unit cost
($/MJ)
0.012°
0.012°
0.021 *
0.012°

Cost
($US)
$0.10
$3.4
$18
0
$22
 11999 U.S. average cost of electricity for the industrial sector is $0.0443/kWh. Assuming 3.6 MJ/kWh,
($0.0443/kWh)/(3.6 MJ/kWh) = $0.012/MJ. Source: www.eia.doe.gov/cneaf/electricity/esr/tll.txt. Note that the
use of the U.S. average cost is simply to compare costs among life-cycle stages, although the actual costs would be
mostly from Asia.
b 1999 U.S. average cost of electricity for the residential and commercial sectors is $0.0771/kWh.  Assuming 3.6
MJ/kWh, ($0.0771/kWh)/(3.6MJ/kWh) = $0.021/MJ. Source:  www.eia.doe.gov/cneaf/electricity/esr/tll.txt.
                                                5-11

-------
5.4 COST AND PERFORMANCE CONSIDERATIONS
       The LCA is defined such that the monitor assessments are performed on a functionally
equivalent basis. To the extent possible, data were collected on functionally equivalent monitors.
The data presented in Table 5-3 are the range of specifications provided by the monitor
assemblers, but not necessarily from the manufacturers of all of the component parts. Therefore,
we are unsure if the specifications provided by the monitor assemblers also represent those of the
component parts, since the component parts manufacturers did not consistently supply
performance data as requested in the data collection questionnaire used in this study. However,
when companies were approached to participate in the study, they were informed of the
performance specification parameters within which the study boundaries were defined.
Therefore, it is assumed that they meet the specifications as presented in Chapter 1 (Table 1-2)
and in Table 5-3 and they perform relatively equivalently. In the primary data, the reported
brightness of the CRT was less than the LCD, otherwise, they are functionally similar.
                                          5-12

-------
 5.5 IMPROVEMENT ASSESSMENT OPPORTUNITIES & TARGETED AUDIENCE USES OF REPORT

5.5    IMPROVEMENT ASSESSMENT OPPORTUNITIES AND TARGETED
       AUDIENCE USES OF REPORT

       To meet the primary objective of providing the display industry with data to perform
improvement assessments, the industry should look at the manufacturing life-cycle stage, while
recognizing the influences of the other stages. CRT improvement opportunities could include
improved energy efficiency during glass manufacturing and display use, as well as reductions in
lead content. LCD improvement opportunities could also include improved energy efficiency,
especially during manufacturing.  Certain materials, such as SF6 and its contribution to global
warming, may also be of concern and an area to focus on in future improvement assessments.
       In addition, any improvement assessment should consider how changes in one life-cycle
stage will affect impacts in other stages. For example, in Chapter 4 we saw that the mercury
inputs and outputs from the intentional use of mercury in an LCD backlight are less than (by
mass) the mercury emissions from the CRT use stage, due to the relatively high energy usage by
the CRT and the emissions of mercury from electricity generation.  In this example, we can see
that on a pure mass basis, a product's energy efficiency is a key consideration and any changes in
manufacturing should consider if it will affect changes in use. However, this project did not
conduct a quantitative risk assessment to determine where the greatest potential for mercury
exposure (and therefore risk) might occur. Consequently, we cannot definitively say whether it is
better to have a potentially less energy-efficient backlight that does not contain mercury or a
more energy-efficient backlight that does. Nonetheless, this analysis highlights the life-cycle
trade-offs that must be considered in an improvement assessment.
       Another objective of this study was to provide an LCA model for future analyses.
Companies or individuals who have more current data for the CRT or LCD can apply them to the
model presented here. For example, changes in an individual process can be identified and
incorporated into the model. The other processes that are not expected to change significantly
can be left unchanged, and only limited data would need to be altered. This would reduce the
time and resources that would normally be required for a complete analysis.
       Finally, those interested in comparing the results of the two monitors can apply their own
set of importance weights to each impact category to determine their individual decision. For
example, if energy impacts are much more important than aesthetics to a particular person, they
can weigh energy more heavily in concluding which monitor may have fewer environmental
impacts, while keeping in mind the data limitations and uncertainties, as well as cost and
performance considerations.
                                         5-13

-------
5.6 SUGGESTIONS FOR FUTURE RESEARCH
5.6    SUGGESTIONS FOR FUTURE RESEARCH

       Areas where future research could be conducted to refine and/or continue the use of the
results in this study are as follows:

•      gather more information on energy use in glass manufacturing;
•      develop consistent materials and fuel processing data in a national (or international) LCI
       database that is updated regularly;
      refine and/or update some of the LCD manufacturing data (e.g., LNG data);
•     collect more complete EOL data (e.g., remanufacturing data, and primary data for
      incineration and landfilling) to determine better representation of the EOL impacts;
•     conduct more research on EOL options for LCDs;
•     collect more detailed data on landfilling and other treatment processes, such as water
      treatment where no impacts were calculated;
•     update manufacturing data to meet more recent monitor model years;
•     conduct a more focused analysis on selected areas for detailed improvement assessments;
      and
      evaluate process changes or other alternatives against an "average 1997-2000 model year"
      to evaluate impacts of changes or improvements over time.
                                       5-14

-------