LCA 2010 AluminumAssoc PDF

May21,2010
FinalReport
LifeCycleImpactAssessmentof
AluminumBeverageCans
Preparedfor:
AluminumAssociation,Inc.
Washington,D.C.
Preparedby:
PEAmericas
Contact:
PEAmericas
NunodaSilva
344BoylstonStreet
NeildSouza
Boston,MA02116,USA
MarcBinder
Phone
Fax
+1[617]2474477
+1[617]2362033
Email
n.daSilva@peinternational.com
Internet
www.peamericas.com
FinalReport:CanLCA
PEAmericas
TABLE OF CONTENTS
TABLEOFCONTENTS ..................................................................................................................... III
TABLEOFFIGURES .........................................................................................................................VI
LISTOFTABLES ............................................................................................................................VIII
ACRONYMS .................................................................................................................................VIII
EXECUTIVESUMMARY...................................................................................................................XI
1
INTRODUCTION ....................................................................................................................... 1
GOALANDSCOPEOFTHESTUDY ............................................................................................. 2
2.1
GOAL ......................................................................................................................................... 2
2.2
SCOPE ........................................................................................................................................ 3
2.2.1 ProductSystem(s)Boundaries............................................................................................ 3
2.2.2 DataCollection,SoftwareandDatabases .......................................................................... 7
2.2.3 DataCalculation ................................................................................................................. 8
2.2.4 FunctionalUnit ................................................................................................................... 9
2.2.5 CutoffCriteria ................................................................................................................... 9
2.2.6 Allocation.......................................................................................................................... 10
2.2.7 LifeCycleImpactAssessmentMethodology&ImpactCategoriesConsidered ................ 10
2.2.8 DataQualityRequirements .............................................................................................. 10
2.2.9 CriticalReview .................................................................................................................. 12
PRIMARYALUMINUMPRODUCTION...................................................................................... 13
3.1
PROCESSDESCRIPTION................................................................................................................. 13
3.1.1 BauxiteMining ................................................................................................................. 14

3.1.2 AluminaProduction .......................................................................................................... 17
3.1.3 AnodeProduction ............................................................................................................. 23
3.1.4 AluminumSmelting .......................................................................................................... 25
3.1.5 PrimaryIngotCasting(CastHouse).................................................................................. 34
3.1.6 ElectricalPowerMixModelforElectrolysis...................................................................... 38
3.2
LCIRESULTS.............................................................................................................................. 41
3.2.1 PrimaryEnergyDemand................................................................................................... 41
3.2.2 CarbondioxideEmissions ................................................................................................. 43
3.3
LCIARESULTS ........................................................................................................................... 44
3.3.1 AcidificationPotential ...................................................................................................... 44
Chapter: Table of Contents

3.3.2 EutrophicationPotential ...................................................................................................45
3.3.3 GlobalWarmingPotential(100years)..............................................................................46
3.3.4 PhotochemicalOzoneCreationPotential..........................................................................49
4
BEVERAGECANPRODUCTION ............................................................................................... 51
4.1
PROCESSDESCRIPTION .................................................................................................................51
4.1.1 CanSheet .......................................................................................................................... 51

4.1.2 CanManufacturing ...........................................................................................................58
4.1.3 SecondaryAluminumIngot/Recycling.............................................................................63
4.2
DATACOLLECTIONANDMODELING ................................................................................................69
4.2.1 CanSheet .......................................................................................................................... 69

4.3
LCIRESULTS(AVERAGEDGATETOGATEINVENTORIES)...................................................................73
4.3.1 BeverageCanSheet ..........................................................................................................73

4.3.3 SecondaryAluminumIngot/Recycling.............................................................................75
4.4
LCIANDLCIARESULTSPER1000CANS ..........................................................................................77
4.4.1 ClosedLoopApproach.......................................................................................................77
4.4.2 Recycledcontentapproach ...............................................................................................81
4.5
5
UBCRECYCLINGSCENARIOS ..........................................................................................................84
CONCLUSIONS....................................................................................................................... 86
BIBLIOGRAPHY ............................................................................................................................. 87
APPENDIXA:ESTIMATIONOFTOTALPRIMARYENERGYDEMAND ................................................ 89
APPENDIXB:TRANSPORTATIONDATAFORRAWMATERIALSANDFINISHEDPRODUCTS .............. 94
APPENDIXC:IMPACTINDICATORS ............................................................................................... 95
PRIMARYENERGYCONSUMPTION ..............................................................................................................95
GLOBALWARMINGPOTENTIAL .................................................................................................................96
ACIDIFICATIONPOTENTIAL ........................................................................................................................97
EUTROPHICATIONPOTENTIAL ....................................................................................................................98
PHOTOCHEMICALOZONECREATIONPOTENTIAL(SMOG)................................................................................99
OZONEDEPLETIONPOTENTIAL ..................................................................................................................99
APPENDIXD:DATAQUALITYEVALUATION ..................................................................................101
APPENDIXE:CRITICALREVIEWREPORT .......................................................................................102
5.1
ALUMINUMBEVERAGECANLCIREPORTFEBRUARY2010 ...............................................................103
REVIEWERCOMMENTSFEBRUARY26,2010 .............................................................................................103
FinalReport:CanLCA
PEAmericas
5.2
ALUMINUMBEVERAGECANLCIREPORTOCTOBER2009................................................................ 104
REVIEWERCOMMENTSDECEMBER4,2009 ............................................................................................. 104
Chapter: Table of Contents
APPENDIXF:CRITICALREVIEWERSBIO....................................................................................... 107
TABLE OF FIGURES
Figure1:Processflowchartforaluminumbeveragecanunderclosedlooprecycling......................... 5
Figure2:Processflowchartforaluminumbeveragecanunderrecycledcontentconditions.............. 6
Figure3:IllustrationoftheVerticalaveragingmethod. ........................................................................ 8
Figure4:IllustrationoftheHorizontalaveragingmethod..................................................................... 9
Figure5:Processchainfortheproductionof1000kgofprimaryaluminum. .................................... 13
Figure6:BauxiteimportsintoNorthAmerica. .................................................................................... 14
Figure7:BauxitemininginBrazil. ........................................................................................................ 17
Figure8:ThetransportationofBauxiteimportedtotheUS ............................................................... 18
Figure9:Alumina(Al2O3)productionprocessintheU.S.in2005...................................................... 21
Figure10:BreakdownofaluminaimportsanddomesticallyproducedaluminaintheU.S................ 22
Figure11:IllustrationoftheAnodeproductionprocess. .................................................................... 27
Figure12:ElectrolysismixrepresentativefortheU.S.in2005. .......................................................... 29
Figure13:Electrolysisprocessmodelforcrudealuminumproduction.............................................. 30
Figure14:ElectrolysisprocessmodelforcrudealuminumimportedtotheU.S.. .............................. 32
Figure15:Primaryingotcastingprocessmodelforprimaryaluminumproduction ........................... 37
Figure16:Electricalpowermixforelectrolysis(domesticmix)........................................................... 38
Figure17:Electricalpowermixforelectrolysis(imports).................................................................... 38
Figure18:EuropeanpowermixusingRussiaastherepresentativecountry. ..................................... 39
Figure19:Primaryenergydemandforprimaryaluminumingotproduction ..................................... 42
Figure20:Carbondioxideemissionsduringprimaryaluminumingotproduction ............................. 43
Figure21:Acidificationpotentialresultsforprimaryaluminumingotproduction. ............................ 45
Figure22:Eutrophicationpotentialresultsforprimaryaluminumproduction .................................. 46
Figure23:Globalwarmingpotentialresultsforprimaryaluminumingotproduction........................ 47
Figure24:POCPresultsforprimaryaluminumingotproduction........................................................ 50
Figure25:Unitprocesschainrepresentativeforoneparticularrollingmillsite................................. 52
Figure26:IllustrationoftheCanmakingprocess................................................................................ 59
Figure27:Unitprocesschainatoneselectedsecondaryaluminumproductionsite ......................... 64
Figure28:PowerMixProcessfortheCanManufacturingModel ....................................................... 72
Figure29:Massflowsofthebeveragecanunderclosedloopapproach............................................ 79
Figure30:Primaryenergydemandofthecanunderclosedloopapproach....................................... 81
FinalReport:CanLCA
PEAmericas
Figure31:Massflowsofthebeveragecanunderrecycledcontentapproach................................... 82
Figure32:Primaryenergydemandofthecanunderrecycledcontentapproach.............................. 84
Figure33.PrimaryenergydemandunderdifferentUBCrecyclingratescenarios ............................. 85
Figure34:Totalprimaryenergydemandfor1kWhgenerationofelectricityintheU.S. .................. 90
Figure35:DirectenergyandPrimaryenergydemand(for1kgofaluminum(liquid)intheU.S. ...... 92
Figure36:Greenhouseeffect ................................................................................................................ 1
Figure37:AcidificationPotential ........................................................................................................... 1
Figure38:EutrophicationPotential ....................................................................................................... 1
Figure39:PhotochemicalOzoneCreationPotential ............................................................................. 1
Figure40:OzoneDepletionPotential................................................................................................ 100
Chapter: Table of Figures
LIST OF TABLES
Table1:Summaryofsystemboundaries ............................................................................................... 3
Table2:Keycharacteristicsoftheclosedloopandrecycledcontentapproaches................................ 7
Table3Inputsandoutputsforthebauxiteminingprocess ................................................................ 15
Table4:BreakdownofbauxiteimportstotheU.S.in2005 ................................................................ 16
Table5:InputsandoutputsfortheAluminaproductionunitprocess................................................ 18
Table6:CountrywisebreakdownofaluminaimportstotheU.S.in2005 ......................................... 23
Table7:Inputsandoutputsfortheanodeproductionunitprocess ................................................... 24
Table8:CrudealuminumconsumptionintheU.S.in2005(USGS,2006)........................................... 28
Table9:InputsandoutputsforthedomesticcomponentofElectrolysisunitprocess. ..................... 30
Table10:InputsandoutputsfortheimportedcomponentofElectrolysisunitprocess .................... 33
Table11:Inputsandoutputsforprimaryingotcastingunitprocess .................................................. 36
Table12:Breakdownofelectricalpower(MJ)consumedinelectrolysisbyenergysource ............... 40
Table13:Relativecontributionofelectricalpowerconsumedinelectrolysisbyenergysource........ 40
Table14:LCIparametersfortheproductionof1000kgofprimaryaluminumingotinNA ............... 42
Table15:Scope1,2,and3CO2Equivalentemissionsforprimaryaluminumingotproduction........ 48
Table16:LCIAresultsforproductionof1000kgofprimaryaluminumingotinNorthAmerica. ....... 50
Table17:InputandOutputflowsforthecansheetmakingunitprocess........................................... 52
Table18:InputandOutputflowsforthecansheetmakingunitprocess........................................... 55
Table19:Inputsandoutputsforthecanmakingunitprocess............................................................ 60
Table20:Inputandoutputflowsforscrappreparationunitprocesses.............................................. 65
Table21:Inputandoutputflowsforremeltingandcastingunitprocesses. ...................................... 66
Table22:Distributionofthecanproductionvolumebycansizein2006........................................... 70
Table23:AveragedLCIdataforcansheet(mix)productionprocess.................................................. 73
Table24:AveragedLCIdataoncanmanufacturingprocess ............................................................... 74
Table25:AveragedLCIdataforscrappreparation............................................................................. 75
Table26:AveragedLCIdataforremeltingandcasting....................................................................... 76
Table27:SelectedLCI/LCIAresultsper1000cansunderclosedloopapproach................................. 80
Table28:SelectedLCIandLCIAresultsper1000cansundertherecycledcontentapproach ........... 83
Table29:Totalprimaryenergydemandfor1kWhgenerationofelectricityintheU.S. .................... 89
Table30:Totalprimaryenergydemandfor1kWhgenerationofelectricityinCanada..................... 90
FinalReport:CanLCA
PEAmericas
Table31:Transportationdataforrawmaterialsandfinishedproducts............................................. 94
Table32:Dataqualityevaluationmatrix ........................................................................................... 101
Table33:Resultsdataqualityevaluation .......................................................................................... 101
Chapter: List of Tables
ACRONYMS
AA
AP
CMI
CML
EoL
EP
GaBi
GHG
GWP
IAI
ISO
LCA
LCI
LCIA
NMVOC
ODP
PEA
POCP
RSI
UBC
USGS
VOC
AluminumAssociation
AcidificationPotential
CanManufacturersInstitute
CentreofEnvironmentalScienceatLeiden
EndofLife
EutrophicationPotential
GanzheitlicheBilanzierung(Germanforholisticbalancing)
GreenhouseGas
GlobalWarmingPotential
InternationalAluminumInstitute
InternationalOrganizationforStandardization
LifeCycleAssessment
LifeCycleInventory
LifeCycleImpactAssessment
Nonmethanevolatileorganiccompound
OzoneDepletionPotential
PEAmericas
PhotochemicalOzoneCreationPotential
Recycledscrapingot
UsedBeverageCan
UnitedStatesGeologicalSurvey
Volatileorganiccompound
FinalReport:CanLCA
PEAmericas
EXECUTIVE SUMMARY
Thisreportoutlinesthelifecycle inventoryofanaluminumbeveragecanproducedinthe
UnitedStates.ThestudywascommissionedbytheAluminumAssociation(AA)torespondappropri
ately to the increasing market demand for life cycle data on the environmental performance of
products. A life cycle inventory of a product quantifies all material and energy use and environ
mentalreleasesoveritsentirelifecyclefromrawmaterialacquisitionthroughtoultimatedisposal.
ThegoalofthisstudyistoprovidetheAluminumAssociation,concernedaluminumindustry
stakeholders, and life cycle assessment practitioners with uptodate life cycle inventory data for
beveragecansofthefollowingsizesasproducedinNorthAmerica:8oz.,12oz.,16oz.,24oz.and
32oz.Thescopeofthestudyiscradletograve,i.e.startingwiththeextractionofthebauxiteore
and including all processes required to produce an aluminum ingot, manufacture the aluminum
beveragecan,andrecoverandrecyclesecondaryrawmaterialfromtheusedbeveragecan(UBC).
Twoapproachestomodeltherecyclingsystem(closedlooprecyclingapproachandrecycled
contentapproach)wereconsideredtoensurethatthestudysresultscanbedirectlycomparedwith
theresultsofotherstudies,whichmightuseeitheroftheseapproaches.
Bothapproachesusedacommonsetofbaselineassumptions,asdefinedbelow:
Averagecanweightof13.34kgper1000cans
Theusedbeveragecanrecyclingrateintheyear2006is51.6%
TherecycledcontentofthebeveragecanintheU.S.in2007is67.8%
Post production and runaround scrap from rolling mills is completely recycled in a
closed loop. The production of secondary aluminum ingot from post production and
runaroundscrapdoesnotrequireanyinputofprimaryaluminum.
CanmanufacturingscrapisdealtwithinthesamemannerasUBCscrapandincludedin
therecycledcontentofthecan.
At the end of life of its useful functions, UBCs are collected, shipped and recycled back to
newaluminummetal.AproportionoftheUBCs,however,arelosttolandfillsorothertypesofsinks
duetoconsumerbehaviors.
In modeling both the primary aluminum and secondary aluminum production processes, a func
tionalunitof1000kgaluminumingotisused.SeparatemodelsforNorthAmerica(USandCanada)
andglobalproductionwereusedtoreflectthemixtureofprimaryaluminumconsumptionintheUS
inthetargetyear.Theproductionofsecondaryaluminumstartswithscrapcollectionandendswith
secondarymetalprocessing.Thecastingofaluminumforsubsequentcansheetproductioniscon
sideredasoneprocess.
In producing 1000 kg of primary aluminum, 5246 kg of mined and processed bauxite was
usedtoproduce1915kgofalumina.1018kgofliquidmetalwasgeneratedoutoftheelectrolysis
Chapter: Executive Summary
Aluminumbeveragecansaremanufacturedthroughthefollowingmajorprocesses(FigureA):pri
maryaluminumproduction(includingbauxiteminingandaluminarefining),secondaryaluminum
production;aluminumcansheetproduction,andcanmanufacturing.TheseareshowninFigureA.

reductionprocess,1000kgofingotwascastedintheend,andtherestofthemetalwasrecycled
backinaclosedloop.
FigureA:Processesinthelifecycleofaluminumcans(recycledcontentperspective)
Processbasedprimaryenergydemand,whichincludesprimaryenergyusedforfuelproduc
tion, transportation, and electricity generation and transmission, for 1000 kg primary aluminum
(bauxitemining,aluminarefining,electrolysis,andingotcasting)isshowninthetablebelow.
ProcessBasedPrimaryEnergyDemandfor1000kgPrimaryAluminumProduction
Bauxite
Mining
Alumina Electrolysis IngotCasting

Refining
Primary Energy Demand for fuel pro 1.02GJ

duction, transportation, & electricity
generationandtransmission
27.36GJ
124.27GJ
2.36GJ
Energy from nonfossil fuel sources, 0.02GJ

fromabovenumbers
0.28GJ
49.75GJ
0.23GJ
FinalReport:CanLCA
PEAmericas

Primaryenergydemandfor1000kgsecondaryaluminumproductionwas6909MJ(includ
ingscrappreparation)whichrequired1059kgofunprocessedscrapasinput.
Inmakingcansheets,primaryaluminum,secondaryaluminum,andcanmakingscrap(the
skeleton,orwebscrap,usuallysoldbycanmakers)weremixedandmeltedinthesecondarymelting
furnaces,and/orintheprimarycasthousetoproduceingotsforrolling.Therolledcansheets,in
cluding can body sheets and can lid sheets, were shipped to can makers to manufacture the final
products.
Undertheclosedloop/endoflifeapproach,theproductionof1000cansatacanmanufac
turingplantrequires16.78kgofaluminumsheet(inclusiveofcanbodyandlid).Thecanmanufac
turingprocessyields3.447kgofpostindustrialscrap(theskeleton)andisshippedbacktothere
melting plants. The rolling mills generates 6.501 kg of scrap metal and it is immediately recycled
backtotherollingprocess.Thisistreatedasaninternalrunaroundscrap.
Inordertoproduce16.78kgofaluminumsheet(for1000cans),thetotalamountofalumi
numingotinputrequiredfortherollingprocessis23.31kg.Thisisproducedfrom:
8.736kgofsecondaryaluminumproducedfromUBCscrap,noncanpostconsumerscrap,
andnoncanpostindustrialscrap
3.447kgofaluminumproducedfromcanpostindustrialscrap(skeletonfromcanmanufac
turingplants)
5.8kgofprimaryaluminum,and
6.501kgofaluminumfromrollingprocesswhichistreatedasainternalrunaroundscrap
AttheEndofLife(EoL),6.984kgofUBCscrap(51.6%oftheaveragecanweightof13.34kg
per1000cans)arerecovered.ThecanproductsystemintheU.S.fortheyear2006requires12.23kg
ofscrapfortheproductionofsecondaryingot.Anadditional1.797kgofscrapisthereforerequired
to make up this deficit. In the recycled content approach, this additional scrap is sourced from a
scrapsinkandthereisnonetburdengiventotheproductsystem.Thescrapsinkcanbecon
sideredasastockpileofsecondaryrawmaterialwhichisnotutilizedforfurtherprocessingandre
coveryofthematerial.
In the recycled content system, the total primary energy demand per 1000 cans with a
51.6% UBC recycling rate is 1692 MJ. The share of total primary energy demand over all life cycle
stagesofthecanisshowninFigureB.
FigureB:PEbreakupforcanlifecycle(recycledcontent)
Thesolidcolorportionofeachbarrepresentsthenonrenewablefractionofprimaryenergy
andthelightshadedportionrepresentstherenewablefractionofprimaryenergy.
Theenvironmentalfootprintof1000cansundereachapproach(closedloopandrecycled
content)islistedinthetablebelow:
TableA.EnvironmentalFootprintof1000Cans
SelectedLCIParameters
ClosedLoop
RecycledContent Unit
Primaryenergydemand
1943
1692
MJ
Nonrenewableenergyresources
1540
1374
MJ
Renewableenergyresources
403.3
318.3
MJ
16.78
16.78
Kg
Carbondioxide
121.6
105.9
Kg
Carbonmonoxide
0.047
0.041
Kg
INPUTS
Cansheet
OUTPUTS
FinalReport:CanLCA
PEAmericas
Nitrogenoxides
0.24
0.211
Kg
Sulphurdioxide
0.436
0.376
Kg
VOCs
0.209
0.188
Kg
SelectedLCIAParameters(CML2001)
Globalwarmingpotential(GWP100years)
131.5
113.8
kgCO2Eq.
Acidificationpotential(AP)
0.613
0.532
kgSO2Eq.
1.565
1.561
kg Phosphate
Eq.
Photochemical Ozone creation potential 0.051

(POCP)
0.046
OzoneLayerdepletionpotential(OPD)
3.61E06
Eutrophicationpotential(EP)
3.76E06
kgEtheneEq.
kgR11Eq.
Undertheclosedloopapproach,the massflowsper1000cansaresimilartotherecycled
contentapproachflowsexceptforthefactthatthedeficitscrapmetalwhichisimportedintothe
systemisassignedanetburdenwhichisthesumoftheenvironmentalburdensassociatedwith
theamountofprimaryaluminumingotrequiredtoreplacethisdeficitminusthoseassociatedwith
theremeltingofsurplusscraptoproducesecondaryaluminumingot.Therefore,theenvironmental
footprintofthecansintherecycledcontentapproachappearstobelowerthanintheclosedloop
consideration.
ToexaminetheinfluenceoftheUBCrecyclingratesontheenvironmentalperformanceof
the beverage can product system, additional scenarios are evaluated based on the following UBC
recyclingrates:
30%HypotheticallowUBCrecyclingrate
45.1%UBCrecyclingrateestimatedbyContainerRecyclingInstitute(ca.2004)(CRI,2008)
62%PeakUBCrecyclingrateachievedduringthemid1990s
75%AluminumAssociationfuturegoalforUBCrecyclingrate
The variation in total primary energy demand per 1000 cans under different scenarios is
showninFigureCandtheresultsarepresentedforbothclosedloop/endoflifeandrecycledcon
tentapproaches.
FigureC:Variationofprimaryenergywithrecyclingratesforclosedloopandrecycledcontent
approaches
AtahigherUBCrecyclingrate(i.e.75%),theclosedloopapproachseemsmorefavorable,
whereasatalowerrecyclingrate(i.e.30%),therecycledcontentapproachseemsmorefavorable,
as the product system would still try to maintain the recycled content of the can and sources of
scrap might be from product systems other than the can system. The reason is that the recycled
contentapproachcutsoffboth,surplusUBCscrapaswellasanunsaturatedscrapdemandincase
therecyclingrateislowerthantherecycledcontentofthecan..Asaresult,theprimaryenergyde
mand under the recycled constant approach does not change as the recycling rate is varied. It is
likely thatrecycled contentofthe canwillincreaseastheUBC recyclingrateincreases,under the
assumptionthatmoreUBCscrapwillbeavailableinthemarketforcanmaking.However,thereare
othermarketforces(e.g.exportofUBCscraptoothercountries,useofUBCscrapforautomotive
sheetproduction)whichcaninfluencetheavailabilityofUBCscraptosecondaryaluminumproduc
ers. The influence of various factors which determine the recycled content of the can in United
Statesisbeyondthescopeofthisstudy.Therefore,inthestudytherecycledcontentofUBCsincans
waskeptconstantat67.8%andonlytheUBCrecyclingratewasvariedinscenarioanalysis.
Detaileddiscussionsofthestudysresultsareextensivelydocumentedinthisreport,butthe
followingmainconclusionscanbedrawnfromtheresultsofthestudy:
The environmental profile of primary aluminum production has a significant influence

ontheoverallenvironmentalburdenresultsforthemanufactureofabeveragecan.The
contribution is around 53% in the case of the recycled content approach and 46% for
closedloopapproach.
FinalReport:CanLCA
PEAmericas
Theelectrolysisunitprocessofprimaryaluminumproductionaccountsforbetween72
and79%ofthecansenvironmentalimpacts.However,itistheupstreamenvironmental
burdensassociatedwiththeelectricityusedforelectrolysisthatisthemaincontributor
totheseimpacts1.Canmanufacturingandrollingoperationsalsocontributeconsidera
blytotheenvironmentalburdensassociatedwithproductionofbeveragecans.
Atpresent,withtheusedbeveragecan(UBC)recyclingrateof51.6%,thedifferenceof
250MJper1000cansbetweenenvironmentalburdensunderclosedloopandrecycled
contentapproachesissignificantandinfavoroftherecycledcontentapproach.Thisis
duetothefactthattheamountofmetalneededtosatisfytherecycledcontentofthe
can sheet is greater than the amount of metal collected. In the closed loop approach,
this deficit is sourced from primary metal while in the recycled content approach, the
additionalscrapisimportedintothesystemburdenfree.However,astheAluminumAs
sociation achieves higher UBC recycling rates, the closed loop approach will produce
morefavorableresults.
Powerproductionandconsumptionassociatedwiththoseupstreamprocesseshoweverareoutofcontrolofthealuminum
industryandthuscannotbedirectlyinfluencedbythealuminumindustry.
1 INTRODUCTION
Life cycle assessment (LCA) is a standardized scientific method for systematic analysis of
flows (e.g. mass and energy) associated with the life cycle of a specified product, a technology, a
serviceormanufacturingprocesssystems(ISO,2006a).Theapproachinprincipleaimsataholistic
andcomprehensiveanalysisoftheaboveitemsincludingrawmaterialsacquisition,manufacturing
as well as use and Endoflife (EoL) management. According to the International Organization for
Standardization(ISO)14040/44standards,anLCAstudyconsistsoffourphases:(1)goalandscope
(frameworkandobjectiveofthestudy);(2)lifecycleinventory(input/outputanalysisofmassand
energy flows from operations along the aluminum products value chain); (3) life cycle impact as
sessment(evaluationofenvironmentalrelevance,e.g.globalwarmingpotential);and(4)interpreta
tion(e.g.optimizationpotential)(ISO,2006a;ISO,2006b).
Thegoalandscopestageoutlinestherationaleofthestudy,theanticipateduseofthere
sultsofthestudy,theboundaryconditions,thedatarequirementsandtheassumptionstoanalyze
theproductsystemunderconsideration,andothersimilartechnicalspecificationsforthestudy.The
goalofthestudyistoanswerthespecificquestionswhichhavebeenraisedbythetargetaudience
andthestakeholdersinvolved,whileconsideringpotentialusesofthestudysresults.Thescopeof
thestudydefinesthesystemsboundaryintermsoftechnological,geographical,andtemporalcov
erageofthestudy,attributesoftheproductsystem,andthelevelofdetailandthecomplexityad
dressedbythestudy.
The life cycle inventory (LCI) stage documents qualitatively and quantitatively analyze the
materialsandenergyused(inputs)aswellastheproductsandbyproductsgeneratedandtheenvi
ronmentalreleasesintermsofnonretainedemissionstotheenvironmentalcompartmentsandthe
wastestobetreated(outputs)fortheproductsystembeingstudied.TheLCIdatacanbeusedonits
own to: understand total emissions, wastes and resourceuse associated with the material or the
product being studied; improve production or product performance; or be further analyzed and
interpretedtoprovideinsightsintothepotentialenvironmentalimpactsfromthesystem(lifecycle
impactassessmentandinterpretation,LCIA).
Theindustryaverageinformationonenvironmentalperformanceofthethreeparticipating
companieswithrespecttotheiroperationsonprimaryaluminumingotproduction,secon
daryaluminumingotproduction,andaluminumcansheetproduction(Alcoa,AlcanandRey
nolds(acquiredbyAlcoain2000)),
Thecumulativeresourcesandenvironmentalburdensassociatedwiththemanufacturingof
1,000cansincludingbauxitemining,aluminarefining,smelting,ingotcasting,cansheetfab
rication, can making, canfilling (including the burdens of the beverage as well), consumer
useandrecoveryofusedbeveragecans(UBC),andrecycling,
Theactualindustryaverageoperationsanda62.4%postconsumerrecyclingraterepresen
tativeoftheNorthAmericanaluminumbeveragecanindustryin1991.
Thestudyquantifiedallsignificantinputsandoutputsofthebeveragecansystem.Theinput
categories included 99.0% of the mass of raw materials identified in the system and 99.6% of the
energy consumed. The environmental releases were quantified for air, water and solid waste and
subdividedintoprocessrelated,fuelrelatedandtransportationrelateddatacategories.Morede
tailed information and results are available in the study report of Pomper, Hauser and Richards
Chapter: Introduction
In1991,Alcoa,Alcan,andReynoldscommissionedaLCIanalysisofthe12oz.NorthAmeri
canaluminumbeveragecan.Theresultsofthispreviousstudyrepresented:

(1994).Thecompanieswerecompetitorsinthebeveragecanmarketandthusprohibitedtheshar
ing of market relevant data under antitrust regulations. Therefore, the information generated by
thestudywasconsolidatedbytheLCIpractitioner.Theinformationderivedfromthestudywaspre
sented in an aggregated form as a weighted averaged of the three individual companys data and
thuswasinlinewiththeantitrustregulations.
Inordertoconformwiththeincreasingpressurefromproductmanufacturersandthecon
sumerretailmarketstoselectenvironmentallypreferablepackagingoptions,AluminumAssociation
(AA)engagedPEAmericas(PEA)tohelpupdatethealuminumbeveragecanLCItothemostupto
dateindustrydataavailable.
Theaimofthestudyistogeneratehighquality,uptodatedataontheenvironmentalper
formanceofaluminumbeveragecanproductionincludingtheflowofsecondarymaterialsfromEnd
ofLifebackintobeveragecans.WithsuchanLCIdatabase,theAluminumAssociationanditsmem
ber companies can assist other organizations to understand and communicate the environmental
benefitsofmanufacturingwithaluminumratherthanothermaterialswithsimilarphysicalproper
ties.Atthesametime,thisdatabasehelpssitelevelpersonnelunderstandtheenvironmentalper
formance of their manufacturing processes, and the potential environmental benefits of process
improvements.Beyondtheoperationsofasinglemanufacturingsite,theLCIdatabaseevaluatesthe
environmental performance of beverage cans throughout their entire life cycle. It provides useful
insightsfordifferentstakeholdergroups,suchasprimaryorsecondaryaluminumproducers,alumi
num users, waste recyclers, government agencies, nongovernmental organizations, LCA practitio
ners,andmedia.
Forthisstudy,acoreprojectteamwasestablishedtodirect,review,andcoordinatetheac
tivities associated with methodological agreement, collection of data, modeling, presentation and
dissemination of the LCI results. The members of the core group for this project are: Craig Covert
[Alcoa], Geoff Cullen [Can Manufacturers Institute], Jim Fava [PE Americas & Five Winds Interna
tional],KenMartchek[Alcoa],RobertStrieter[AluminumAssociation]andGerriWalsh[Ball].
Different companies participated in the study to provide data for the various production
stages of the beverage can include: Mill Rolling (Alcoa, Logan Aluminum, and Wise Alloys), Can
Manufacturing (Can Manufacturing Institute) and other Secondary Aluminum Producers (Aleris,
Novelis).
2 GOAL AND SCOPE OF THE STUDY

2.1 GOAL
The goalofthisLCAstudyistoprovidetheAluminumAssociation,thealuminumindustry
stakeholders,andtheLCApractitionerswithuptodateLCIdataforbeveragecansofthefollowing
sizes2:8oz.,12oz.,16oz.,24oz.and32oz.
The update of the beverage can LCI became necessary as the original LCI became increas
inglyoutofdatepartlyduetotechnologicalandtechnicalchangesandpartlyduetothefactthat
TheLCIdataforthedifferentsizesrepresentweightedaverages.
FinalReport:CanLCA
PEAmericas
theoriginalLCIcouldnolongermeettheinformationneedsoftoday.TheupdatedLCIshallreflect
thecurrenttechnologicalsituationaswellastheNorthAmericanMarketsituation.
TheintendedaudienceforthisstudyistheAluminumAssociationitself,thepotentialcus
tomers and decision makers in industry, as well as the general public. The Aluminum Association
experts will use the information from this study in an aggregated manner for public communica
tions,todevelopmarketingmaterialsforpotentialcustomersandtoprovidedatatocustomersfor
thepurposeofdevelopingLCIswithintheirownapplications.
ThepresentupdateofthebeveragecanLCIisintendedtobeusedforcomparativeasser
tionstobedisclosedtothepublic,andisthereforesubjecttoexternalcriticalreviewaccordingto
ISO14044guidelines.
2.2 SCOPE
ThescopeofthestudycomprisesacradletograveLCI,startingwiththeextractionofthe
bauxiteoreatthemine,includingtheproductionofaluminumingotandthemanufacturingofthe
aluminum beverage can, and ending after the recovery and recycling of the UBC. Table 1 summa
rizesthesystemboundarieswithregardtothegeneralprocesses/quantitiesthatareconsideredin
thestudy.TheLCIprofileresultsareprovidedforeachoftwodifferentrecyclingapproaches:
Closedlooprecycling
Recycledcontent
2.2.1
Product System(s) Boundaries
The product being examined is an average beverage can made of aluminum. Its content,
manufacture, and impact represent the current technological and technical situation in the North
Americanmarket.Thefunctionofthebeveragecanistoserveascontainerforbeveragessuchas
beerorsoftdrinks.Therearenootherfunctionsoftheanalyzedaveragebeveragecanconsideredin
thisstudy.
Included
Excluded
Rawmaterialsextraction
Capitalequipmentandmaintenance
Energyandfuelinputs
Further processing materials

(e.g.chemicals,solvents,etc.)
Overhead(heating,lighting)ofmanufacturingfacilities
wheneasilydifferentiated
Maintenanceandoperationofequipment
Manufactureofanybeverageanditsfillingincans
Internaltransportationofmaterials
Useofproduct
Packagingofcansfordistributiontoconsumers
Humanlabor
Productdisposal(i.e.landfilling)
Processing of raw materials

andsemifinishedproducts
Transportation of raw and

processedmaterials
Productrecycling
Chapter: Goal and Scope of the Study
Table1:Summaryofsystemboundaries

2.2.1.1 Closed loop recycling approach
The system flow chart for closed loop conditions is presented below in Figure 1. The flow
charthighlightstherecyclingofpostindustrialcleancanscrap(fromrollingandcanmanufacturing)
aswellastherecyclingofpostconsumerscrap(UBCfromtheEoLphase).
Theclosedloopapproachisbasedonaproductlifecycleandmaterialstewardshipperspec
tive.Itconsidersthefateofproductsaftertheirusephaseandtheresultantmaterialoutputflows.
Inevaluatingtheenvironmentalimpactsofaproductsystemusingthisapproach,theEoLmanage
mentoftheproductisalsotakenintoaccountandtherefore,possiblechangestoimprovethesys
temcanbeconsidered.Thespecificoriginofinputmaterial(whetherprimaryorrecycled)isirrele
vantastypicallythenetconservationofmaterialiswhatminimizesthetotalenvironmentalimpacts.
Underthisframework,theproductbeingexaminedisconsideredtobecompletelyrecycled
onceitreachestheEoLphase.MateriallossesduringthecollectionandprocessingofUBCaswellas
those associated with the production of secondary material feedstock out of UBC are taken into
account.ConsistentwithISO14044,theselossesareconsideredtobereplacedbyprimarymaterial
feedstock.Foragivenproductsystem,acertainamountofsecondaryrawmaterialfeedstockisre
quired(determinedbyexistingmarketconditions).Itissourcedfrombothpostindustrialandpost
consumerscrap.Inmostproductsystems,postindustrialscrapisrecycledinaclosedloopandpar
tiallysatisfiesthedemandforsecondaryrawmaterialfeedstock(Figure1).Theremainingdemand
forsecondarymaterialfeedstockismetwithpostconsumerscrap,i.e.UBC.
IftheamountofUBCscrapgeneratedislessthanwhattheproductsystemrequires,then
theenvironmentalburdensassociatedwithmeetingthesecondaryrawmaterialfeedstockdemand
areincludedinthisclosedloopmodel,i.e.theenvironmentalimplicationsassociatedwithsupple
mentary raw material production is considered (Table 2). If, however, the amount of UBC scrap is
largerthanwhattheproductsystemrequires,thentheproductsystemreceivesanetcredit,equiva
lenttothesumofenvironmentalimpactsfromprimarymaterialextractionandremeltingofsurplus
postconsumerscrap.
A designer using this approach focuses on optimizing product recovery and material recy
clability.Byfacilitatinggreaterendofliferecycling,thedecisionmakermitigatesthelossofmate
rial after product use. This approach assesses the consequences at the endoflife of the product
based on established technical practices, and supports decisions for an efficient market. This con
ceptallowsdesignforrecycling.
FinalReport:CanLCA
PEAmericas
2.2.1.2 Recycled content approach

ThesystemflowchartfortherecycledcontentapproachispresentedbelowinFigure2.The
recycled content approach adopts a waste management perspective, where the sourcing of raw
materialfeedstock(bothprimaryandsecondaryrawmaterials)isconsideredtobeofimportance.It
assumesthattheuseofrecycledmaterialisagoodindicatorofenvironmentalbenefit.Inthecaseof
Aluminum for example, the benefits of recycling (primary material production and waste disposal
avoidance)outweightheburdensofpostconsumercollectionandprocessing.Therecycledcontent
approachaimsatpromotingamarketforrecycledmaterialsthatisotherwiselimited,uneconomic
orimmature.Itmayalsoserveasagoodmetricformaterialsthatwouldotherwisebeincineratedor
landfilled as waste (assuming that these waste management treatment processes would result in
higherenvironmentalimpactsthanthematerialsrecycling).
Figure 1: Process flow chart indicating the system boundary for aluminum beverage can under
closedlooprecyclingconditions.Theproductbeingexaminedisconsideredtobecompletelyrecy
cledonceitreachestheEoLphase.Materiallossesduringboththecollectionandtheprocessing
of UBC as well as losses associated with the production of secondary material feedstock out of
UBC are taken into account. In the case of excess UBC scrap, the product system receives a net
creditequivalenttothesumofprimarymaterialavoided.Theflowcharthighlightstherecyclingof
postindustrial can scrap (from rolling and can manufacturing) as well as the recycling of post
consumer(UBC)scrap(fromtheEoLphase).
Figure 2: Process flow chart indicating the system boundary for aluminum beverage can under
recycled content conditions. This approach excludes consideration of actual collection and recy
clingefficiencies,ratherconsiderssnapshotsofcurrentmaterialfeedsituationintermsofpri
maryandsecondarymaterials.IftheamountofUBCscrapisgreaterthanwhattheproductsys
tem requires, then the surplus UBC scrap leaves the system without any consideration of envi
ronmentalburdensassociated.
Intherecycledcontentapproach,theamountofsecondaryrawmaterialfeedstockinthefi
nal product is usually specified. The product system then meets the specified recycled content by
sourcing secondary raw materials from postindustrial scrap and from UBC. This approach differs
from the closed loop approach in its accounting for environmental impacts when the amount of
aluminum available from postindustrial scrap and UBC recovery does not meet the specified
amountofrecycledcontent.IftheamountofpostindustrialscrapandUBCrecyclingislessthanthe
productsystemrequires,thenadditionalsecondaryaluminumisimportedburdenfree,i.e.theenvi
ronmentalburdensassociatedwithmeetingthesecondaryrawmaterialfeedstockdemandareex
cluded (Figure 2). If, however, the amount of UBC scrap is greater than what the product system
requires, then the surplus UBC scrap leaves the system without receiving credit for avoided envi
ronmental burdens.Inthismannerthesystemboundaryisstill constructed tofairlyallocateenvi
ronmentalburdensinasystemwithrecycledcontent.Itshouldbenotedthattherecyclingofpost
FinalReport:CanLCA
PEAmericas
industrial rolling scrap is still considered part of a closed loop, which is generally true of current
manufacturingpractices.
Unfortunately,theapplicationoftherecycledcontentapproachmaycreatemarketdistor
tions and environmental inefficiencies. If a designer specifies a high recycled content in a well
meaningefforttoincreasetheoverallenvironmentalperformanceofaproduct,itmaystimulatethe
markettodirectrecycledfeedstocktowardsdesignatedproductsandawayfromproductionwhere
recycling is most economical. For metals, where there is a limited supply of recycled feedstocks,
marketstimulationisineffectiveandmayresultininefficientprocessingandunnecessarytranspor
tation.
ThekeycharacteristicsofthetwoapproachesaredescribedinTable2.
Table2:Keycharacteristicsoftheclosedloopandrecycledcontentapproaches
Closedloopapproach
Recycledcontentapproach
Ifthecanproductsystemforagivensystem
requires more secondary raw materials from
UBC than it generates, then additional burdens
associated with the production of primary raw
materialreplacingthemissingamountofsecon
daryrawmaterialfromUBCscrapareincluded
2.2.2
Data Collection, Software and Databases
Thestudyincludeddatacollectioninthefollowingcategoriesforrolling,canmanufacturing
andsecondaryaluminumproduction:
Fuelandenergyuse,
Useofrawmaterials,ancillarymaterials,
Products,coproducts,
Emissionstoair,waterandsoil,
Adopts a product life cycle and material Adoptsawastemanagementperspective

stewardshipperspective
Excludes the consideration of actual collection and
EoL material is recycled and fed back in a recycling efficiencies, rather considers snap shots of
closedloop(cantocan)production
current material feed situation in terms of primary and
secondaryrawmaterials
Metal losses have to be replaced by virgin
material (losses during remelting of production Surplus secondary material in form of UBC scrap
and EoL scrap, collection and processing of leaves the can production system burden free (i.e. no
UBCs)
burdens associated with remelting of surplus UBC scrap
andnocreditfortheequivalentamountofprimaryalu
IncasetheamountofUBCscrapcollectedin
minumavoidedisassigned)
theEoLphaseexceedstheamountofUBCscrap
consumed (as secondary raw material) for can Ifthecanproductsystemforagivensystemrequires
production in a given system, the product sys more UBC scrap than it generates, then additional bur
tem under consideration receives a net credit dens of importing UBC scrap are excluded(UBC scrap is
equivalent to the sum of primary aluminum importedwithoutconsiderationofremeltingefforts).
avoidedandtheremeltingofsurplusUBCscrap
Wastes.
Primary data collection of information, which is representative for specific production op

erations,wasaccomplishedbydistributingdocumentedquestionnairesintheformofEXCELspread
sheets to all participating companies. Wherever possible, this study is based on primary data col
lected from the participating companies and their respective production sites. In cases where pri
marydatawasnotavailable,secondarydatareadilyavailablefromliterature,previousLCIstudies,
andlifecycledatabaseswasusedfortheanalysis.Thesourcesforsecondarydataaredocumented
inthisstudyreport.Intheabsenceofsecondarydata,expertapproximationswereusedtoclosethe
datagaps.
The LCA model was created using the GaBi 4 software system (current release GaBi 4.3,
http://www.gabisoftware.com) for life cycle assessment, developed by PE INTERNATIONAL. The
databases contained in the GaBi software provide the LCI data of the raw and process materials
usedinthebackgroundsystem.
2.2.3
Data Calculation
Thisstudyutilizedacombinationofverticalandhorizontalaveragingmethodstoderivethe
meanvalueoftheprimarydata.Theverticalmethod(seeFigure3)wasappliedconsistentlytoall
thecompaniesasthismethodismorerepresentativeofactualindustrialprocesses.However,inthe
caseofidenticalprocesses,e.g.rolling,thehorizontalaveragingmethod(seeFigure4)wasused.
Figure 3: Illustration of the Vertical averaging method. The final average is calculated from a
weightedmeanofthesumofalltheoperationsofeachcompany.Intermediateaveragesmayalso
becalculatedaftereachoperation.(ECOBILAN,2001).
FinalReport:CanLCA
PEAmericas
Figure4:IllustrationoftheHorizontalaveragingmethod.Averagesarecalculatedaftereachop
eration and are used as the inputs for the next operation. In our context, this method is used
when a particular company does not produce an intermediate product or when there is insuffi
cientdataforanintermediateoperation(ECOBILAN,2001).
2.2.4
Functional Unit
2.2.5
Cut- off Criteria
The following cutoff criteria were used to ensure that all relevant environmental impacts
wererepresentedinthestudy:
MassIfaflowislessthan1%ofthecumulativemassofalltheinputsandoutputsdepend
ingonthetypeofflow)oftheLCImodel,itmaybeexcluded,provideditsenvironmentalre
levanceisnotaconcern.
EnergyIfaflowislessthan1%ofthecumulativeenergyofalltheinputsandoutputsde
pending on the type of flow) of the LCI model, it may be excluded, provided its environ
mentalrelevanceisnotaconcern.
Thefunctionalunitofthestudyistoproduceonethousandaluminumbeveragecansrepre
sentingweightedaveragesofthecansizesdefinedabove.
EnvironmentalrelevanceIfaflowmeetstheabovecriteriaforexclusion,yetisthoughtto
potentially have a significant environmental impact, it will be included. All material flows
whichleavethesystem(emissions)andwhoseenvironmentalimpactishigherthan1%of
thewholeimpactofanimpactcategorythathasbeenconsideredintheassessment,shall
becovered.
The sum of the neglected material flows shall not exceed 5% of mass, energy or environ
mentalrelevance.
2.2.6
Allocation
Inthisstudy,noallocationwasappliedasallcoproductsareconsideredineachoftherecy
clingmodelsdescribedabove(ISO,2006b).Anyrecoveredmaterialsrecycledintootherproductsare
treatedaswaste,therebyprovidingamaximumallocationofburdenstoaluminumcanproduction.
2.2.7
Life Cycle Impact Assessment Methodology & Impact Categories Considered
Itwasdeterminedduringthescopedevelopmentprocessthatacomprehensivesetofenvi
ronmentalimpactcategoriesweretobeinvestigated.Forthepurposesofsuccinctcommunication
ofthestudyresults,thefollowingimpactcategoriesweredeterminedtobestrepresenttheAlumi
numAssociationsprioritiesinissuesrelatedtosustainability:
GlobalWarmingPotential(GWP)(100years;includescarbondioxide(CO2)andothergreen
housegas(GHG)relevantemissions),
AcidificationPotential(AP),
PhotochemicalOzoneCreationPotential(POCP)alsocalledSummerSmog,
EutrophicationPotential(EP),and
OzoneDepletionPotential(ODP).
ThemeaningandsignificanceoftheseimpactcategoriesisdiscussedindetailinAppendixC:
ImpactIndicatorsofthisreport.Theimpactassessmentresultswerecalculatedusingcharacteriza
tionfactorspublishedbytheCentreofEnvironmentalScienceatLeidenUniversity(CML),theNeth
erlands (CML, 2001). The CML 2001 methodology is the most widely applied impact assessment
methodinLCAstudiesaroundtheworld.
2.2.8
Data Quality Requirements
Thefollowingparagraphsdocumentthecomprehensivedataqualityrequirementsadopted
accordingtoISO14044(ISO,2006b).
2.2.8.1 Temporal Coverage
Primary data collected from the participating companies and associations for their opera
tionalactivitiesarerepresentativefortheyear2006(referenceyear).Additionaldatanecessaryto
modelbasematerialproductionandenergygeneration,etc.wereadoptedfromtheGaBi4software
systemdatabase.ThemostrecentNorthAmericanLCIsubdatasetfromtheInternationalAluminum
Institute(IAI)LCIDatasetforprimaryaluminumdatesbackto2005.Thisdeviationfromthedefined
referenceyearhasbeentakenintoaccountasitisbeingassumedthattherearenoradicalchanges
inthetechnologyforrollingoperationsandcanmanufacturingfromtheyear2005to2006.
FinalReport:CanLCA
PEAmericas
2.2.8.2 Geographical Coverage

Thegeographicalcoverageforthisstudyisasfollows:
PrimaryAluminumProductionNorthAmerica,
Datasource:IAI,representativeforNorthAmericansituation
AluminumSheetRollingUnitedStates
3majorsheetproducers(4sites)intheUSaccountingforabout1.7milliontonsofsheet
producedintheUS
CanManufacturingUnitedStates
Datasource:CMI,representativeforoperationsintheU.S
SecondaryAluminumProductionUnitedStates
Datasource:4ofthelargestproducers(5sites)ofsecondaryaluminumintheUScover
ingabout2millionmetrictonsofsecondaryaluminiumproduction.
2.2.8.3 Technological Coverage

In this study, sitespecific data representing the current technology mix for aluminum can
sheet production, can manufacture, and secondary aluminum production was collected. Primary
aluminumproductiondata(includingbauxitemining,aluminarefining,aluminumsmelting,andpri
mary ingot production) in North America were provided by the IAI. The scope of the IAI data on
primaryaluminumproductioninNorthAmericaisasfollows:
Bauxiteminingglobalaverage,
Aluminarefiningglobalaverage,
ElectrolysisNorthAmerica,
PrimaryingotcastingNorthAmerica.
Ancillaryandprocessmaterialdata,suchastheproductionofchemicals,fuels,energyand
power,wasadoptedasaverageindustrymixesfromtheGaBi4softwaresystemdatabase(current
releaseGaBi4.3,http://www.gabisoftware.com).
Duringthedatacollection,eachproductionfacilitywasrequiredtoassignadataqualityin
dicatortothedatareported.Thedataqualityindicatorsareclassifiedasfollows:
MeasuredIfthedatavaluereportedisbaseduponcontinuousphysicalmeasurement(e.g.
electricityorwaterconsumptionisbaseduponmeasurementofrespectivemetersatthefa
cility),
Calculated If the data reported has been calculated using a certain empirical formula or
factors(e.g.emissionfactorforCO2orSO2dependinguponthetypeoffuelusedandproc
esstechnology),
Averaged If the data reported is based upon an average of a number of values or meas
urements(e.g.theamountofsmeltingslaggenerated),
2.2.8.4 Completeness
EstimatedIfthedatareportedisbaseduponinformedestimatesorexpertjudgment(e.g.
thedistancetherawmaterialsaretransportedbeforetheyarriveatthefacility).
Althoughitisdifficulttoconductacomprehensivedataqualityandreliabilitycheckonthe
data reported from several production sites, consistency and quality checks for mass and energy
balanceresultswereconductedandresultscomparedwithpublisheddataparticularlyprocessand
flow data in previous LCI studies. The checks showed that the reported data was in the range re
portedbysimilarstudieswithincomparableboundaryconditions.Inadditionthedataqualitywas
evaluatedusingtheWeidemamethodologyonLCAdataquality(Weidemaetal.,InternationalJour
nalofLCA3(5)page259265;1998).AppendixDshowstheresultsoftheevaluationaswellasthe
appliedevaluationcriteriafordataquality,rangingforonetofive,whereoneisthebestqualityand
fivethemostuncertain.Theevaluationshowedthedataconformedtocriteriavaluesofoneortwo
(seeAppendixD).
This quality assurance (QA) process was performed at different stages of the project. The
objectiveoftheQAprocesswastoensurethatthedatacollection,thedevelopmentoftheLCImod
el,andthefinalresultsareconsistent with thescopeofthestudyand thatthestudydeliversthe
requiredinformation.
2.2.9
Critical Review
TheresultsoftheLCAstudyareintendedtosupportexternalcommunication;thereforein
ordertobecompliantwithISO14044,acriticalreviewofthestudywasconducted.
ThegoalandscopeofthecriticalreviewisdefinedinaccordancewithISO14044,paragraph
6.1.FollowingISO14044,thecriticalreviewprocessshallensurethat(ISO,2006b):
themethodsusedtocarryouttheLCAareconsistentwiththisInternationalStandard
themethodsusedtocarryouttheLCAarescientificallyandtechnicallyvalid
thedatausedareappropriateandreasonableinrelationtothegoalofthestudy
theinterpretationsreflectthelimitationsidentifiedandthegoalofthestudy
thestudyreportistransparentandconsistent
ThereviewofthisstudywasdonebytheLCIcoregroupmembersandanindependentre
viewpanelcomprisedofthefollowingmembers:
ToddBoggess[Secat],Chair
MaryAnnCurran[UnitedStatesEnvironmentalProtectionAgency]
BruceVigon[SETAC]
A dialogue between the critical reviewers and the project team allowed continual integra
tionofcriticalreviewfeedbackintothestructureofthestudy,andthedraftingofthisfinalreport.
FinalReport:CanLCA
PEAmericas
3 PRIMARY ALUMINUM PRODUCTION

Thischapterdescribestheunitprocessesassociatedwithprimaryaluminumingotproduc
tion.Thefollowingsectionscoveradescriptionoftheprocessbeingmodeled(Chapter3.1),presen
tationoftheLCIresults(Chapter3.2)andpresentationoftheLCIAResults(Chapter3.3).Therele
vantsupportinginformationusedtomodeltheLCIprofileofaprimaryaluminumingotproducedin
North America is noted where relevant to the portions of this section (e.g. IAI, 2005; USGS, 2006,
2007).
3.1 PROCESS DESCRIPTION
Figure5:Processchainfortheproductionof1000kgofprimaryaluminum.
Inthisstudy,theprimaryaluminumproductionintheU.S.ismodeledbasedupontheLCI
dataobtainedfromIAIandstatisticsonbauxite,aluminaandaluminumpublishedannuallybythe
Chapter: Primary Aluminum Production
Thelifecyclestagesofprimaryaluminummaterialprocessingincludesthecomponentproc
esses of bauxite mining, alumina refining, electrolysis (including smelting and anode production),
andprimaryingotcasting.AprocessflowdiagramisshowninFigure5.Theinitialrawmaterialisthe
bauxiteoreandfinalproductistheprimaryaluminumingotwithintermediateproductsofalumina
(aluminumoxide)andaluminum(liquid)metal.

U.S. Geological Survey (USGS). Overall, the data quality and consistency of the IAI data has been
identifiedtobeofhighquality.
NorthAmericaimportsbauxiteandalumina,bothofwhichareintermediatesofthesame
ingredientinthecanmakingprocess.Tobeabletorepresentthesimpleproductionprocessshown
inFigure5,amorecomplexmodelhadtobeimplemented,whereinthealuminaimportswerecon
vertedintoequivalentamountsofbauxitetobeminedusingtheIAIglobalaverageof2.739tonsof
bauxite per metric ton of alumina produced. The additional amount of bauxite to be mined was
addedtothebauxiteimportsmodelasshowninFigure6.
IAI_NA: Bauxite mining 2005

G aBi 4 process plan: M ass [kg]
IAI_NA: Bauxite
imports
IAI_NA: Bauxite X
for alumina imports
3502.1 kg
1744 kg
Bauxite
5246.1 kg
Bauxite [hp]
Figure 6: Bauxite imports into North America. IAI_NA: Bauxite for alumina imports represents
the bauxiteequivalents of the amount of alumina imported into NA. Figures are representative
fortheproductionof1000kgofprimaryaluminum.
Thefollowingsubsectionscharacterizethegenericprocessdescriptionforeachoftheunit
processesassociatedwithprimaryaluminumproductionandarebasedontheprofileofthealumi
numindustryoutlinedinpreviousstudies(AA,1998;IAI,2005).
3.1.1
Bauxite Mining
Bauxiteoreistheprimaryrawmaterialsourcefor thealuminumproduction. Aluminum is

almost exclusively produced from Bauxite. This ore consists primarily of the minerals gibbsite
Al(OH)3, boehmite, and diaspore AlOOH, together with minor fractions of the iron oxides goethite
and hematite, the clay mineral kaolin and small amounts of TiO2. Bauxite is typically found at a
depthof0to600feetbeneaththeearthcrust,withanaveragedepthof80feet.Itisminedinopen
pitminesbyremovingtheoverburden.Theremovedmaterialisstockpiledforuseinrestoringthe
siteafterthebauxitehasbeenexcavated.Thebauxitedepositisloosenedbymeansofexplosives,
depending on its hardness and other local conditions. In some cases the bauxite is crushed in a
grinding process using dust control equipment to prevent from excessive dust emission, and/or
treatedwithwatertoremoveimpuritiesbeforeitisshipped.Thiswashingprocessiscalledbenefici
ation.Beneficiatedbauxitewilltypicallybedriedpriortoshipmenttotherefinery.Thewastewater
fromwashingisnormallyretainedinasettlingpondandrecycledforcontinualuse.
FinalReport:CanLCA
PEAmericas
This bauxite mining unit process begins with the extraction and processing of the bauxite
oreanditendswiththeoutputofbeneficiatedbauxitetoberefinedinthesubsequentprocessto
producealumina.Theoperationsassociatedwiththisunitprocessinclude(AA,1999;IAI,2005):
Theextractionofbauxiterichmineralsonsite,
Beneficiationactivitiessuchasgrinding,washing,screening,anddrying,
Treatmentofminingsiteresiduesandwaste,and
Restorationactivitiessuchasgrading,dressing,andplanting.
UnitprocessinputsandoutputsareshowinTable3.
Flow
Unit
Amount
Inputs
Diesel[Crudeoilproducts]
kg
6.0436
Power[Electricpower]
MJ
36.21
Thermalenergy(heavyfueloil)[Thermalenergy]
MJ
52.89
Thermalenergy(naturalgas)[Thermalenergy]
MJ
0.07
Bauxite[Nonrenewableresources]
kg
5775.8
Water(seawater)[Water]
kg
302
Water(surfacewater)[Water]
kg
2633.7
Outputs
Bauxite[Inorganicintermediateproducts]
kg
5246.2
Dust(unspecified)[Particlestoair]
kg
5.1
Overburden(deposited)[Stockpilegoods]
kg
529.6
Steam[Inorganicemissionstoair]
kg
143.6
kg
261.6
Water(surfacewater)[Water]
kg
2514.9
The domesticbauxite productionintheUSislessthan1%of thetotal demandofbauxite

(USGS, 2007). Most of the domestically produced bauxite is utilized for nonmetallurgical applica
tionssuchasabrasives,chemical,refractorymaterials(USGS,2007).Therefore,thebauxitedemand
formetallurgicalpurposes(i.e.toproducemetallic aluminum)intheU.S.isprimarilymet through
Table3Inputsandoutputsforthebauxiteminingprocess.Flowsarerepresentativefor1000kg
primaryaluminumproduction

imports.ThebauxiteimportsbasedontheircountryoforigintotheU.S.in2005isgiveninTable4.
Brazilisthelargestexporterofbauxite(accountingforapproximately33%ofthetotalbauxiteim
portstotheU.S.)followedbyGuineaandJamaica.ThegroupofcountriesincludedintheOthers
categoryareChina,India,theDominicanRepublicandVenezuela.
ThebauxiteminingmodelisbasedonglobalaverageIAIdataexceptforenergysourcesand
fuelswherecountryspecificdatawasused.TheU.S.scenario(basicallyamixofbauxiteimporting
sourcesandquantities)wasusedtorepresentthatofCanadaalso.Itwasthereforechosenasrepre
sentative for the entire North American region. This assumption was based on the fact that the
overallcontributionofthebauxiteminingprocesstotheenvironmentalimpactsissmallcompared
withotherprocessesconsidered.Thereforetheimpactofsuchanassumptionwouldbenegligiblein
the final results. As an illustration of the processes considered in the mining process, the bauxite
miningmodelforBrazilisshowninFigure7onthefollowingpage.
Table4:BreakdownofbauxiteimportstotheU.S.in2005[USGS,2007]byCountry.Brazilisthe
largestexporterofBauxitetotheU.S.accountingfornearlyonethirdofthetotalimportstothe
U.S.
Country
Bauxiteimport(thousandmetrictons)
Australia
232
Brazil
3210
Guinea
2460
Guyana
1280
Jamaica
1760
Others
916
Total
9858
ThetransportationofbauxitefromeachindividualcountrytotheU.S.isconsideredinthe
model to create the bauxite import mix model as shown in Figure 8. It is assumed that bauxite is
transportedthroughabulkcommodityoceancarrierutilizingheavyfueloilasitsenergysource.The
transportation distance was approximated based upon the average nautical distance between a
majorportineachofthebauxiteexportingcountriesandNewOrleansintheU.S.Thetransporta
tion distance for bauxite imports from Other countries was based upon the weighted average
distancefromthecountriescomprisingtheOtherscategory.Thetransportationdistancewases
timatedusingawebbasedcalculator(WorldPortsDistances,2007).
FinalReport:CanLCA
PEAmericas
Figure7:BauxitemininginBrazil.Figureusedasanexampletoillustratetheminingprocessmod
eled for the LCI. Flows are representative for the production of 1000 kg of primary aluminum
(scaledtotheshareofbauxiteimportsfromBrazil).
3.1.2
Alumina Production
Thisstepofmanufacturingbeginswiththeprocessingofbeneficiatedbauxiteandendswith
theoutputofaluminatobesubsequentlyprocessedinthesmelter.Theoperationsassociatedwith
thisunitprocessinclude(AA,1999;IAI,2005):
bauxitegrinding,digestion,andprocessingofliquors,
aluminaprecipitationandcalcination,
maintenanceandrepairofplantsandequipment,and
treatmentofprocessair,liquids,andsolids.
TheunitprocessinputsandoutputsarelistedinTable5.
AluminarefiningcomprisestheconversionofbauxitetoaluminumoxideAl2O3(alumina)us
ing the Bayer process (e.g. Chin 1988; Droy and Michaux 2003; Mylona et al. 2003; Frank et al.
2008).Mostrefineriesuseamixtureofblendedbauxitetoprovidefeedstockwithconsistentprop
erties.Themixtureisgroundandblendedwithrecycledplantliquor.Thisliquorcontainsdissolved
sodium carbonate and sodium hydroxide recovered from previous extraction cycles plus super
natantliquorrecycledfromredmudholdingponds.Theslurryisheatedandpumpedtodigesters,
which are heated in pressure tanks. In digestion, iron and silicon impurities form insoluble oxides
calledred mud.Thered mudsettles outandarichconcentrationofsodiumaluminatesisfiltered
and seeded to form hydrate alumina crystals in precipitators. These crystals are then heated in a
calcinationprocess.Theheatinthecalcinersdrivesoffcombinedwaterleavingalumina.
IAI_NA: Bauxite mining (Bauxite imports)

G aBi 4 process plan: M ass [kg]
AU: Bauxite
p
mining with transport 82.42 kg
Bauxite from
Australia
BR: Bauxite
p
Bauxite from Brazil
GN: Bauxite
p
Bauxite from Guinea
GY: Bauxite
p
Bauxite from Guyana
JA: Bauxite
p
Bauxite from
Jamaica
GLO: Bauxite
p
Bauxite from other

countries
Bauxite imports p X
mixer [hp]
3502.1 kg
Product
Bauxite
Figure 8: The transportation of Bauxite imported to the US is included in each countryspecific

model.Prefixesforlabelsarestandardcountrycodesandrepresentcountryspecificminingmod
els already discussed. GLO represents the global average mining model applied to mining from
otherminorexportingcountries.Flowquantitiesarerepresentativeof1000kgofprimaryalumi
numproduction(scaledtotheshareofbauxiteimports).
Table5:InputsandoutputsfortheAluminaproductionunitprocess.Flowsarerepresentativefor
1000kgprimaryaluminumproduction.
Flows
Unit
Amount
Inputs
Inorganicintermediateproducts
Bauxite
kg
5246.2
Sodiumhydroxide(50%;causticsoda)
kg
172
Limequicklime(lumpy)[Minerals]
kg
75.5
FinalReport:CanLCA
PEAmericas
Energyandfuelinputs
Diesel
kg
1.31
MJ
865.6
Thermalenergyfromhardcoal
MJ
3060.2
Thermalenergyfromheavyfueloil
MJ
7851.7
Thermalenergyfromnaturalgas
MJ
7909.6
Water
Seawater
kg
201.2
Surfacewater
kg
15190
Outputs
Products
Aluminumoxide(alumina)
kg
1915.4
Wasteforrecovery
Lubricant(grease)[Hazardouswasteforrecovery]
kg
0.91
Wastefordisposal
Redmud(dry)[Hazardouswastefordisposal]
kg
2187
Industrialwasteformunicipaldisposal
kg
29.3
Waste(solid)
kg
47.1
Sand(Aluminaproduction)
kg
58.3
Emissionstoair
Steam
kg
4989.7
Mercury(+II)
kg
4.02E04
Emissionstowater
Solids(suspended)
kg
0.091
Water(seawater)
kg
241.4
Water(surfacewater)
kg
10160.5
AccordingtotheIAI,theproductionof1metrictonofaluminarequiresapproximately2.739
metric tons of bauxite (taking into account the purity of bauxite and losses during processing and
transportation)(IAI,2005).Thisisaglobalrepresentativeaveragethathasbeenadoptedtomodel
the alumina production process in North America. The alumina production process is shown in
Figure9.
FinalReport:CanLCA
PEAmericas
Figure9:Alumina(Al2O3)productionprocessintheU.S.in2005.Flowquantitiesarerepresentativeof1000kgofprimaryaluminumproduction
(scaledtotheshareofdirectaluminaproduction).
AccordingtotheUSGS,theU.S.imported1,866,000metrictonsofaluminaanddomestically
produced4,724,000metrictonsofaluminain2005(USGS,2007).Thefractionofaluminaimportsto
totalaluminaconsumedforprimaryaluminumproductionisapproximately40%(Figure10).
IAI_NA: Alumina (Production + Imports)

GaBi 4 process plan: Mass [kg]
IAI_NA:
ALUMINA (USA)
IAI_NA: ALUMINA
Imports to USA
1160,7 kg
754,66 kg
Aluminum oxide
(alumina)
1915,4 kg
Alumina <hp>
Figure10:BreakdownofaluminaimportsanddomesticallyproducedaluminaintheU.S.in2005.
Flowsarerepresentativeforproducing1000kgofprimaryaluminum.
TheimportsofaluminafromAustraliaandSurinameaccountedforapproximately80%ofthetotal
aluminaimportstotheU.S.ThecountryspecificbreakdownofaluminaimportstotheU.S.in2005is
giveninTable6.ThetransportationofaluminaimportedtotheUSisincludedineachcountry
specificmodel.
FinalReport:CanLCA
PEAmericas
Table6:CountrywisebreakdownofaluminaimportstotheU.S.in2005[USGS,2007]
Country
Aluminaimports(thousandmetrictons)
Australia
932
Brazil
17
Canada
103
China
10
France
16
Germany
49
Jamaica
116
Japan
Suriname
560
Venezuela
27
Others
31
Total
1866
Aswiththebauxiteminingprocess,itwasassumedthatthealuminaproductionintheU.Sis
representativeoftheentireNorthAmericanregion.Asthecontributionoftheprocessesinvolvedin
producingaluminafrombauxiteplaysasignificantroleintheenvironmentalimpactsoftheentire
canmaking process, this assumption was based upon two key factors: (1) The fact that the most
significant environmental factor in alumina production is energy derived directly from fossil fuels
(seeFigure9)andtherefore,anydifferencesintheresultswouldbeminorandonlycontributedby
thedifferenceinefficienciesoftheenergyconverters.(2)Thedifferenceinenvironmentalimpacts
(e.g.GWP100)betweenimportedanddomesticallyproducedaluminaislessthan1%andtherefore
differences in this ratio for Canada do not affect results significantly. A simple sensitivity analysis
wasusedtoconfirmtheseassertions.
3.1.3
Anode Production
Therearetwogenerictypesofreductioncells:prebakeandSderberg(Anseen,Okstad,In
nvar,&Olsen,1979;Bergsdal,Strmann,&Hertwich,2004;IAI,2005).TheSderbergdesignhasa
singleanodewhichcoversmostofthetopsurfaceofthereductioncell.Anodepaste(briquettes)is
fedtothetopoftheanodeandastheanodeisconsumedintheprocess,thepastefeedsdownward
bygravity.Heatfromthepotbakesthepasteintoamonolithicmassbeforeitgetstotheelectrolytic
bathinterface.

Theprebakedesignhasprefiredblocksofsolidcarbonsuspendedfromaxialbusbars.The
busbarsbothholdtheanodesinplaceandcarrythecurrentrequiredforelectrolysis.
Theprocessformakingtheaggregateforbriquettesorprebakeblocksisidentical.Cokeis
calcined,groundandblendedwithpitchtoformapastethatissubsequentlyextrudedintoblocksor
briquettesandallowedtocool.Whilethebriquettesaresentdirectlytothepotsforconsumption,
theblocksarethensenttoaseparatebakingfurnace.
Bakingfurnacetechnologyhasevolvedfromsimplepitsthatdischargedvolatilestotheat
mosphere during the baking cycle to closed loop type designs that convert the caloric heat of the
volatileintoaprocessfuelthatreducesnetenergyconsumption.TheIAIdataforelectrolysisrepre
sents85%ofproductionfromprebakefacilitiesandtheremaining15%fromSderbergfacilities(IAI,
2005).
Theoperationsassociatedwithanodeproductioninclude(AA,1999;IAI,2005):
recoveryofspentanodematerials,
anodemixpreparation,blockorbriquetteformingandbaking,
roddingofbakedanodes,
maintenanceandrepairofplantandequipment,and
An illustration of the anode production process implemented in the model is shown in

Figure11.TheunitprocessinputsandoutputsarelistedinTable7.Notethatforallunitprocesses,
only reported values of carbon dioxide are listed here. These (when shown) are direct emissions
fromfuelconsumption.Note:forthecalculationoftheLCIandLCIAresults,thesereportedvalues
wereexcludedandstandardGaBiprocesseswereusedtocalculateemissions.Thisapproachguar
anteesthatallemissionsareaccountedforcorrectlyandaccurately.
Table7:Inputs andoutputsfor theanodeproduction unitprocess.Flowsarerepresentativefor

theproductionof1000kgofprimaryaluminum.
Flow
Unit
Amount
Inputs
Materials
Coke(Ccarrier)
kg
345.2
HardcoalpitchMix(CTrger)
kg
92.32
Refractory[Minerals]
kg
5.819
Steelsheetpart(St)[Metalparts]
kg
1.706
Coolingwater
kg
200.7
EnergyandFuels
FinalReport:CanLCA
PEAmericas
Diesel
kg
0.401
MJ
213.13
MJ
45.16
Thermalenergyfromheavyfueloil
MJ
93.32
MJ
1035.6
Outputs
Products
Anode(Ccarrier)
kg
437.5
Emissionstoair
Carbondioxide
kg
177.6
Fluorides
kg
0.004
Benzo{a}pyrene[GroupPAHtoair]
kg
3.48E05
Polycyclicaromatichydrocarbons[GroupPAHtoair]
kg
0.0281
Emissionstowater
Fluoride
kg
0.0002
Oil(unspecified)
kg
5.02E05
Polycyclicaromatichydrocarbons(PAH)
kg
5.52E05
Solids(suspended)
kg
0.0005
Aluminum Smelting
Molten aluminum is produced from alumina by the HallHeroult electrolytic process (e.g.
Frank, et al., 2008; Grjotheim & Kvande, 1993). This involves two steps: dissolving the alumina
(Al2O3) produced in the preceding aluminum refining step in a molten cryolitic bath, and passing
electriccurrentthroughthissolution,therebydecomposingthealuminaintoaluminumandoxygen.
Aluminumistappedoutofthereductioncell(pot)atdailyintervalsandtheoxygenbondswiththe
carbontoformcarbondioxideandcarbonmonoxide.
Aluminumsmelterstypicallyuseairpollutioncontrolsystemstomonitorandreduceemis
sions.Theprimarysystemistypicallyascrubber.Someplantsusedryscrubberswithaluminaasthe
absorbent that is subsequently fed to the pots and allows for the recovery of scrubbed materials.
Otherplantsusewetscrubbers,whichrecirculateanalkalinesolutiontoabsorbemissions.Unlike
dryscrubbers,wetscrubbersabsorbcarbondioxide,nitrogenoxideandsulfurdioxidethatareen
trainedinthewastewaterliquor.
3.1.4

Thisunitprocessbeginswiththeprocessingofaluminaandendswiththeoutputofmolten
primaryaluminumtobesubsequentlycastintoprimaryingotinthecastingprocess.Theoperations
associatedwithelectrolysisinclude(AA,1999;IAI,2005):
recovery,preparation,andhandlingofprocessmaterials,
manufactureofmajorprocessequipment(e.g.,cathodeshells),
processofcontrolactivities(metal,bath,heat),
maintenanceandrepairofplantandequipment,and
ThetotalcrudealuminumconsumptionintheU.S.intheyear2005isshownintheTable8.
The domestic production of primary aluminum in the U.S. and imports from Canada account for
approximately42%and32%,respectively,ofthetotalcrudealuminumconsumption.
FinalReport:CanLCA
PEAmericas
Figure11:IllustrationoftheAnodeproductionprocess.Allflowsarerepresentativefor1000kgprimaryAl.production(domesticcomponent).

Table 8: Crude aluminum consumption in the U.S. (domestic production plus imports) in 2005
(USGS,2006)
Region
Africa
NorthAmerica
LatinAmerica
Asia
Europe
Country
Aluminumcrude
Percentageshare(%)
Quantity(metrictons)
77700
1.31%
SouthAfrica
77700
4401000
74.34%
USA
2481000
41.91%
Canada
1920000
32.43%
230136
3.89%
Argentina
65600
Brazil
19400
Mexico
97
Panama
39
Venezuela
145000
297623
5.03%
Bahrain
22700
China
68200
Japan
536
Korea
87
Tajikistan
127000
UnitedArabEmirates
79100
850157
14.36%
Belgium
36
France
789
Germany
2400
Italy
757
FinalReport:CanLCA
PEAmericas
Oceania
Total
Netherlands
436
Norway
4060
Spain
79
UK
22600
Russia
819000
63200
1.07%
Australia
63200
5919816
100%
TheimportsofcrudealuminumtotheU.S.(excludingCanada)accountfor26%ofthetotal
consumptionandtherefore,theNorthAmericanshare(U.S.andCanada)is74%.Thebreakdownof
theelectrolysismixintoimportedanddomesticcomponentsishighlightedinFigure12.
TheelectrolysisprocessmodelsfordomesticproductionandimportmixareshowninFigure
13andFigure14,respectively.Table9andTable10listtheinputsandoutputsofthesetwoproc
esses.
Figure12:ElectrolysismixrepresentativefortheU.S.in2005.Allflowsarerepresentativeforpro
ducing1000kgofprimaryaluminum.
Figure13:ElectrolysisprocessmodelforcrudealuminumproductionrepresentativefortheU.S.in
2005.Allflowsarerepresentativeforproducing1000kgofprimaryaluminum.
Table9:InputsandoutputsforthedomesticcomponentofElectrolysisunitprocess.
Flow
Unit
Amount
Inputs
Materials
Aluminumfluoride
kg
11.94
kg
1420.3
FinalReport:CanLCA
PEAmericas
Anode(Ccarrier)
kg
325.2
Blastingabrasive
kg
0.104
Cathode
kg
7.61
Steelsheetpart(St)
kg
5.97
Water
kg
9026
Water(seawater)
kg
74.59
Energyandfuels
MJ
41762
Outputs
Products
Aluminum(liquidmetal)
kg
757.12
Wasteforrecovery
kg
2.91
Refractory
kg
5.82
Smelterrecyclingbyproduct
kg
6.12
Emissionstoair
Carbondioxide
kg
1181.6
Fluorides
kg
0.589
kg
0.002
Polycyclicaromatichydrocarbons[GroupPAHtoair]
kg
0.216
Tetrafluoromethane(CF4)
kg
0.112
Hexafluoroethane(C2F6;R116)
kg
0.0112
Emissionstowater
Fluoride
kg
0.037
Oil(unspecified)
kg
0.005
Solids(suspended)
kg
0.067

Polycyclicaromatichydrocarbons(PAH,unspec.)
kg
1.79E04
Hazardouswaste
Carbon(unspecified)
kg
5.45
Refractory
kg
10.67
Sludge
kg
12.83
Water[Water]
kg
11860
kg
74.6
IAI_NA-2005-Electrolysis (imports)
GaBi 4 process plan:Reference quantities
Alumina
495,1 kg
Aluminum oxide
(alumina)
Aluminum fluoride
4,223 kg
Cast iron part (sand

casting)
1,699 kg
Petrol coke at
refinery
2,06 kg
IAI_NA: Anode
production 2005
p
112 kg
IAI_NA-2005-Energ
y model Power Mix
Aluminum fluoride
GLO: Electrolysis X
2005 (US imports) [b]
<b>
Cast iron part
Cathode
Anode (C carrier)
1,417E004 MJ
Power
261,3 kg
Aluminum (liquid)
Aluminium (liquid
metal)
Figure14:ElectrolysisprocessmodelforcrudealuminumimportedtotheU.S.asrepresentative
for2005.Allflowsarerepresentativeforproducing1000kgofprimaryaluminum.
FinalReport:CanLCA
PEAmericas
In the electrolysis process, electrical energy is the primary energy resource. The electrical
powermixmodelforelectrolysisisdiscussedbelowinSection3.1.6.Inthisstudy,thepowercon
sumption of electrolysis is found to be 15552 kWh per metric ton of primary aluminum. Overall,
accordingtoIAI,theworldaveragepowerconsumptionoftheelectrolysisprocessis15289kWhper
metrictonofprimaryaluminum.
Flow
Unit
Amount
Inputs
Materials
Aluminumfluoride
kg
4.22
kg
495.1
Anode(Ccarrier)
kg
112
Cathode
kg
2.059
Steelsheetpart(St)
kg
1.699
Water
kg
2755
Water(seawater)
kg
4531.5
Energyandfuels
MJ
14169
Outputs
Products
kg
261.3
Emissionstoair
Carbondioxide
kg
401
Fluorides
kg
0.268
Tetrafluoromethane(CF4)
kg
0.039
Hexafluoroethane(C2F6;R116)
kg
0.004
Polycyclicaromatichydrocarbons(PAH)[GroupPAHtoair]
kg
0.075
kg
6.69E04
Table10:InputsandoutputsfortheimportedcomponentofElectrolysisunitprocess

Emissionstowater
Solids(suspended)
kg
0.052
Fluoride
kg
0.082
Polycyclicaromatichydrocarbons(PAH,unspec.)
kg
4.22E04
Oil(unspecified)
kg
0.002
Hazardouswaste
Refractory
kg
3.527
Carbon(unspecified)
kg
3.012
Sludge
kg
1.210
Wasteforrecovery
Refractory
kg
1.622
Smelterrecyclingbyproduct
kg
2.292
kg
0.669
Water[Water]
kg
2626.3
kg
4.532
PFC(Perfluorocarbon)generationinAluminumSmelting
PFCemissionsinthealuminumsmeltingprocessarelistedinTable9andTable10(asHexaf
louroethane and Tetrafluoroethane) for domestic and imported components respectively. It is
estimatedthat165.5gofPFCsareemittedpertonofprimaryaluminumproduction(IAI,2005).CO2
equivalentsarecalculatedbasedonCML20013valuesof5700(kg)forCF4 and11900(kg)forC2F6
andarelistedinTable15.
3.1.5
Primary Ingot Casting (Cast House)
Moltenmetalsiphonedfromthepotsissenttoaresidentcasthousefoundineachsmelter.
In some cases, due to proximity, molten metal is transported directly to a shape casting foundry.
Moltenmetalisthentransferredtoaholdingfurnacewherethecompositionisadjustedtothespe
cificalloyrequestedbyacustomer.Insomeinstances,dependingontheapplicationandthebath
compositioninthepots,someinitialhotmetaltreatmenttoremoveimpuritiesmaybedone.
TheUSEPAandothergovernmentsarestillusingtheIPCC2ndAssessmentvaluesof6,500
forCF4and9,200forC2F6.
FinalReport:CanLCA
PEAmericas
Whenthealloyingiscomplete,themeltisfluxedtoremoveimpuritiesandreducegascon
tent. The fluxing consists of slowly bubbling a combination of nitrogen and chlorine or of carbon
monoxide,argon,andchlorinethroughthemetal.Fluxingmayalsobeaccomplishedwithaninline
degassingtechnologywhichperformsthesamefunctioninaspecializeddegassingunit.
Fluxingremovesentrainedgasesandinorganicparticulatesbyflotationtothesurfaceofthe
metal. These impurities (typically called dross) are skimmed off. The skimming process also takes
somealuminumanddrossesarenormallyfurtherprocessedtorecoverthealuminumcontentand
tomakeproductsusedintheabrasivesandinsulationindustries.
Dependingontheapplication,metalisthenprocessedthroughaninlinefiltertoremoveany
oxidesthatmayhaveformed.Subsequently,metaliscastintoingotsinavarietyofmethods:open
molds(typicallyforremeltingot),throughdirectchillmoldsforvariousfabricationshapes,electro
magneticmoldsforsomesheetingots,andthroughcontinuouscastersforaluminumcoils.
Thisunitprocessbeginswiththeprocessingofmoltenprimaryaluminumandendswiththe
outputofsheetingotsuitableforrolling,extruding,orshapecasting.Thevariousoperationscarried
outinthecasthouseinclude(AA,1999;IAI,2005):
Pretreatmentofhotmetal(cleaningandauxiliaryheating);
Recoveryandhandlingofinternalprocessscrap;
Batching,metaltreatment,andcastingoperations;
Homogenizing,sawing,andpackagingandcastingoperations;
Maintenanceandrepairofplantandequipment;and
Treatmentofprocessair,liquids,andsolids.
ThemodelfortheprimaryingotcastingprocessisshowninFigure15.FollowingtheIAIin
ventorydata(IAI,2005),alloyingmaterialsarenotconsideredinthemodel.Theingotisconsidered
tobecomposedof100%aluminumascanbeseeninthefigurewherethealloyingprocessboxis
usedonlytoreintroducerecycleddrossbackintothecastingprocess.Forthepurposeofsimplify
ingthemodel,thedeficitfromtheotherwiseexpected15kginputofalloyingmaterialisobtained
fromthealuminumDCcastingot.

Table11:Inputsandoutputsforprimaryingotcastingunitprocess.Flowsarerepresentativefor
theproductionof1000kgofprimaryaluminum.
Flow
Unit
Amount
Inputs
Materials
Alloycomponents(Aluminum)
kg
15.05
kg
1018.5
Chlorine
kg
0.055
Coolingwater
kg
100.34
Energyandfuels
Diesel
kg
3.31
MJ
252.86
MJ
81.3
MJ
1213.7
Outputs
Products
AluminumDCcastingot
kg
1003.4
Wasteforrecovery
Dross
kg
18.36
Filterdust
kg
0.030
Steelscrap(St)
kg
17.26
Wastefordisposal
Dross(Fines)
kg
0.803
Waste(solid)
kg
0.2
Refractory
kg
2.91
Emissionstoair
FinalReport:CanLCA
PEAmericas
Hydrogenchloride
kg
0.016
Emissionstowater
Oil(unspecified)
kg
0.010
Solids(suspended)
kg
0.020
Figure15:Primaryingotcastingprocessmodelforprimaryaluminumproduction.Flowsarerepre
sentativefor1000kgofprimaryaluminumproduction.

3.1.6
Electrical Power Mix Model for Electrolysis
Inordertomodeltheelectricityconsumptionforelectrolysis(domesticproductionandim
ports),amodelforelectricitysupplyhasbeendevelopedwhichisbasedupontheelectricitymixof
thealuminumindustry.TheunitprocessmodelforelectrolysisitselfwasbasedonNorthAmerican
practices,andisassumedtobegloballythesame.Thedirectelectricityinputduringelectrolysisisa
criticalLCIparameterthatcansignificantlyinfluencetheenvironmentalburdenoftheoverallbever
agecanmanufacturingprocess.Therefore,separateelectricalpowermixmodelswerecreatedfor
North America (includes the U.S. and Canada and representing domestic production) and Other
regions(representingimports)basedontheconsumptionofthealuminumindustry.Therespective
flowchartsareshowninFigure16andFigure17,respectively.Theelectricityconsumptionforpri
maryaluminumproductionisbasedonIAIstatistics(IAI,2007).
Figure 16: Electrical power mix for electrolysis (domestic mix). Flows are representative for pro
ductionofthedomesticcomponentof1000kgoftotalprimaryaluminum(scaledtotheshareof
domesticproduction).
Figure17:Electricalpowermixforelectrolysis(imports).Flowsarerepresentativeforproduction
oftheimportedcomponentof1000kgtotalprimaryaluminum(scaledtotheshareofimports).
FinalReport:CanLCA
PEAmericas
Thecompositionofeachofthecontinentalorregionalpowermixeswascompiledaccording
totherespectivesharesofdifferentpowerproductiontechnologies.Arepresentativecountryfrom
each region having a significant share of the regional power consumption for electrolysis was se
lectedtomodelthepowermixforeachregion,e.g.RussiaforEurope(seeFigure18),Australiafor
Oceania,etc.(IAI,2007).
Figure18:EuropeanpowermixusingRussiaastherepresentativecountry.Flowsarerepresenta
tive for production of the European imports component of 1000 kg of total primary aluminum
(scaledtotheshareofimports).
According to the IAI statistics (IAI, 2007), the power mixes representative for the U.S. and
Canadaareofthesamecomposition,astheybothbelongtothecategoryNorthAmerica(referto
Area2inTable12andTable13).Thebreakdownofelectricalpowerbyenergysourceconsumed
intheelectrolysisprocessforthedifferentregionsisshowninTable12whileTable13providesthe
relative contributions of electrical power by energy source consumed in the electrolysis for each
regionisillustrated.Itcanbeobservedthathydropowerhasthelargestshare(approximately69%)
intermsoftotalelectricityinputforelectrolysisfortheU.S.andCanada.

Table12:Breakdownofelectricalpower(MJ)consumedinelectrolysisbyenergysourceforeach
region[IAI,2007]
ReportedElectricalPowerConsumed(Gigawatthours)
Energy
Source
Area1
Area2
Area3
Area4/5 Area6A/6B Area7
Africa
North
America
Latin
America
Asia
Europe
Oceania
Hydro
5808
55249
33195
3868
88228
7659
194187
Coal
17992
23702
14182
13480
25618
95334
Oil
599
1722
2327
Naturalgas
76
233
1412
21572
6704
29997
Nuclear
951
473
15697
17121
Total
24827
79841
34607
40221
126191
33279
338966
Total
Table13:Relativecontributionofelectricalpowerconsumedinelectrolysisbyenergysourcefor
eachregion[IAI,2007]
ReportedElectricalPowerConsumed(%)
Energy
Source
Area1
Area2
Area3
Area4/5
Area6A/6B Area7
Africa
North
America
Latin
America
Asia
Europe
Oceania
Hydro
23.4%
69.2%
95.9%
9.6%
69.9%
23.0%
57.3%
Coal
72.5%
29.7%
0.0%
35.3%
10.7%
77.0%
28.1%
Oil
0.0%
0.0%
0.0%
1.5%
1.4%
0.0%
0.7%
Naturalgas
0.3%
0.3%
4.1%
53.6%
5.3%
0.0%
8.8%
Nuclear
3.8%
0.6%
0.0%
0.0%
12.4%
0.0%
5.1%
Total
100.0%
100.0%
100.0%
100.0%
100.0%
100.0%
100.0%
FinalReport:CanLCA
PEAmericas
Total
3.2 LCI RESULTS

Inthischapter,themostimportantLCIresultsarerepresentedfortheproductionof1met
rictonofprimaryaluminumingotintheNorthAmericaregion.
3.2.1
Primary Energy Demand
Theprimaryenergydemandisameasureofthetotalamountofprimaryenergyextracted
fromthe earth,including bothnonrenewable (i.e.fossilfuels)andrenewable(hydropower,wind,
solar, etc.) resources, taking into account the efficiency of electric power generation and heating
methods.
Itisnecessarytodistinguishtheprimaryenergydemandandtheenergythatisreceivedas
socalledendenergyattheoperationsite.Anexamplemayhelptomakethispointclear:
1. Primaryenergywouldbetheamountofhydropowerenergygeneratedatthegen
erationsite(attheresourcessites,forinstanceinhydropowerplants)
2. Hydropower in the power grid mix indicates the energy in the distribution net
workthatisreadytobeusedattheoperationsite.
The energy efficiency coefficient indicates the efficiency of the energy conversion (and its
transmission, if applicable) system, and relates the primary energy demand and the end energy
throughthefollowingequation:
Theenergyrequiredforelectrolysisisprimarilyprovidedaselectricalpower.IntheU.S.andCanada,
hydroelectricgenerationprovidesthemajorityofelectricalpower.However,thisisnotthecasefor
othercountries,e.g.Australia,whereelectricalpowergeneratedfromcoalisthedominantsourceof
energyconsumedintheelectrolysisprocess.
TheprimaryenergydemandresultsareillustratedinFigure19,includingabreakdownbe
tween nonrenewableandrenewable resources.Theproductionof1metric tonofprimaryalumi
num ingot representative for North American conditions requires 105 GJ of energy from non
renewablesourcesand50GJfromrenewablesources.Theelectrolysisprocessaccountsfor80%of
the total primary energy demand. The electrolysis process and the anode production (anode pro
duction contributes approximately 14% of the primary energy demand for electrolysis) are highly
energy intensive processes compared to other unit processes. As a result of the different power
production efficiencies, the overall nonrenewable fraction of primary energy for electrolysis is
greaterthan therenewablefraction.TheprimaryenergydemandresultsaresummarizedinTable
14
below.
More detailed information on the methodology to estimate the primary energy demand
fromdirectenergyusedataisdocumentedinAppendixA.

Table14:LCIparametersfortheproductionof1000kgofprimaryaluminumingotinNorthAmer
ica
Inventory
Unit
Bauxite
mining
Alumina
Refining
Electrolysis
Casthouse Total
GJ/ton
1.02
27.36
124.27
2.36
155.00
Nonrenewable
GJ/ton
1.00
27.08
74.52
2.12
104.73
Renewable
GJ/ton
0.02
0.28
49.75
0.23
50.27
CO2emissions
ton CO2/ 0.07

ton
2.03
7.47
0.14
9.70
parameter
PrimaryEnergy
Demand
Primary Energy Demand (GJ per metric ton of primary

aluminum ingot)
180
160
Renewable
Non-renewable
140
120
100
80
60
40
20
0
Bauxite Mining
Alumina
Production
Electrolysis
Casting
Total
Unit Process
Figure 19: Primary energy demand from renewable and nonrenewable sources for primary alu
minumingotproductionperunitprocessandintotal.Electrolysisaccountsforthelargestprimary
energy demand (80%) of which 40% comes from renewable sources, while alumina production,
which accounts for only 17.6% of total primary energy demand utilizes 26% of the total non
renewableenergy.
FinalReport:CanLCA
PEAmericas

3.2.2
Carbon dioxide Emissions
Carbondioxideisoneofthegreenhousegasesthatcontributestotheglobalwarmingphe
nomenon.Carbondioxideemissionsaremainlyassociatedwiththeconversionoffossilenergycar
riers(e.g.lignite,crudeoil,naturalgas)intothermaland/ormechanicalenergybymeansofburning
andareexpressedinkilogramsofCO2.ThecarbondioxideemissionresultsoftheLCIareillustrated
inFigure20.Itiscalculatedthatabout9.7metrictonsofcarbondioxideisemittedpermetrictonof
primaryaluminumingotproduced.Thecarbondioxideresultsarecloselylinkedtotheprimaryen
ergy demand results and their graphs have much the same shape. The electrolysis process is the
largestcontributor,producingalmost7.5metrictonsofcarbondioxideforeachtonofprimaryalu
minumingotproduced.Theupstreamemissionsassociatedwiththeelectricitysupplychainforelec
trolysisaccountfor70%ofthetotal7.5metrictonsCO2forelectrolysis.Thecarbondioxideresults
aresummarizedinTable14.
10000
kg CO2/metric ton primary aluminum ingot
9000
8000
7000
6000
5000
4000
3000
2000
1000
0
Bauxite Mining
Alumina Production
Electrolysis
Casting
Total
Figure20:Carbondioxideemissionsduringprimaryaluminumingotproductionperunitprocessin
total.Electrolysisisthemajorcontributor tothetotalCO2emissionsof which70%arefromthe
electricitysupplychain.
Unit Process
3.3 LCIA RESULTS

In this chapter, the Life Cycle Impact Assessment (LCIA) results are presented for the pro
ductionof1metrictonofprimaryaluminumingotinNorthAmerica.UnliketheLifeCycleInventory,
whichonlyreportssumsforindividualemissions,theLCIAincludesmethodologiesforweightingand
combiningdifferentemissionsintoametricforsignificanttheLifeCycleInventory,aLifeCycleIm
pactAssessment(LCIA).
AsdescribedinSection2.2.7ofthisreport,theimpactassessmentresultswerecalculated
usingcharacterizationfactorspublishedbytheCentreofEnvironmentalScienceatLeidenUniversity
(CML), Netherlands (CML, 2001). The CML 2001 methodology is the most widely applied impact
assessmentmethodinLCAstudiesaroundtheworld.Additionalinformationonthesignificanceof
theseimpactcategoriesisincludedinAppendixC:ImpactIndicatorsofthisreport.
3.3.1
Acidification Potential
Theacidificationpotentialisameasureofemissionsthatcauseacidifyingeffectstotheenvi
ronmentandisexpressedaskilogramSO2Equivalent.
Themajoracidifyingemissionsarenitrogenoxides(NOx)andsulfurdioxide(SO2),aswellas
ammoniaemissionsthatleadtoammoniumdeposition.Theacidificationpotentialrelatedto1000
kg of primary aluminum ingot production in North America amounts to 50.43 kg SO2 equivalent
(Figure21;Table16).TherelativeshareofthisacidificationpotentialindicatorfromSO2emissionsto
airis74%,andfromNOxemissionstoairis25%.
Breakingtheemissionsdownbyproductionstagesshowsthattheelectrolysisprocessisre
sponsiblefor77%ofthetotalacidificationpotentialresult;followedbyaluminarefiningwhichhas
an18%contribution.
Wealsonotethat74%oftheacidificationimpactsareassociatedwithupstreamemissions
duringelectricityproduction.
FinalReport:CanLCA
PEAmericas
kg SO2 Equiv./metric ton of primary aluminum ingot
60
50
40
30
20
10
0
Bauxite Mining
Alumina Production
Electrolysis
Casting
Total
Unit Process
Figure21:Acidificationpotentialresultsforprimaryaluminumingotproduction.77%ofthetotal
acidification potential result is attributed to the process of electrolysis and upstream emissions
fromelectricitygenerationaccountfor74%ofthetotalAP.
3.3.2
Eutrophication Potential
Theeutrophicationpotentialrelatedtothemanufactureof1metrictonofprimaryalumi
numingotinNorthAmericaamountstoalmost2.35kgPhosphateequivalent(Figure22;Table16).
Theeutrophicationpotentialfromemissionstoair(mainlyNOxemissions)contributesto95%ofthe
totalimpacts.Theremaining4%oftheeutrophicationpotentialisduetoemissionstowater(mainly
fromnitrateemissions,chemicaloxygendemandCODandNOxreleasestowater).Theremainderof
the eutrophication impact indicator comes from emissions to soil, which represents 0.7% (mainly
fromammoniaandphosphorusreleases).
Breaking the impact down by contributions from different production stages, Figure 22
showsthatthealuminarefiningandelectrolysisprocessestogetherareresponsiblefor91%ofthe
eutrophicationimpactsresult,withindividualcontributionsof72%and19%,respectively.Emissions
toairfromupstreamprocesses(suchaselectricityproduction)accountforapproximatelytwothirds
(67%)ofthetotaleutrophicationpotentialresult.
Theeutrophicationpotentialisameasureofemissionsthatcauseeutrophyingeffectstothe
environment and is expressed as kilogram of Phosphate Equivalent. The eutrophication of aquatic
systems is primarily caused by excessive inputs of nitrogen and phosphorus (mostly as a result of
overfertilization).
kg Phosphate Equiv/metric ton of primary aluminum

ingot
3.0
2.5
2.0
1.5
1.0
0.5
0.0
Bauxite Mining
Alumina Production
Electrolysis
Unit Process
Casting
Total
Figure22:Eutrophicationpotentialresultsforprimaryaluminumproduction.Aluminaproduction
andelectrolysistogetheraccountforabout91%ofthetotalEPofwhichindirectemissionstoair
accountfor67%.
3.3.3
Global Warming Potential (100 years)
The Global Warming Potential (GWP) is a measure of the emission of greenhouse gases
(GHG)suchasCO2andmethane(CH4)andisexpressedaskilogramofCO2equivalents.Greenhouse
gasemissionsarefoundtocauseanincreaseintheabsorptionofradiationemittedbythesunand
reflectedbytheearth,magnifyingthenaturalgreenhouseeffect.
Thetotalglobalwarmingpotential(GWP)relatedtotheproductionof1metrictonofpri
maryaluminumingotinNorthAmericais11068kgCO2equivalent.AbreakdownoftheGWPimpact
bycomponentgreenhousegasesshowsthatalmost85.8%ofthenetGWPcomesfromCO2,9.79%
from Tetrafluoromethane, 2.56% from CH4, 1.6% from Hexafluoroethane, and 0.3% from nitrous
oxide(N2O).
AbreakdownoftheresultsbyindividualproductionstagesisshowninFigure23andshows
that79%oftheglobalwarmingimpactscomefromtheelectrolysisprocess.Aluminarefiningisnext
largestcontributorwitha19%shareofnetglobalwarmingpotential.
The share of global warming potential from direct greenhouse gas emissions is approxi
mately 25% of net GWP impact, while indirect CO2 emissions (mainly from electricity production)
accountforanother50%ofnetGWPimpact.
FinalReport:CanLCA
PEAmericas
kg CO2 Equiv/metric ton of primary aluminum ingot
12000
10000
8000
6000
4000
2000
0
Bauxite Mining
Alumina Production
Electrolysis
Casting
Total
Unit Process
Figure23:Globalwarmingpotentialresultsforprimaryaluminumingotproduction.Theelectroly
sisprocessisresponsiblefor79%oftheglobalwarmingimpacts,ofwhich85.8%areduetoCO2
emissions.
BasedupontherequestoftheAluminumAssociationandthecoregroupmembers,theGHG
emissionresultsfortheprimaryaluminumingotproductionwerefurthercategorizedapplyingthe
concept of scopes as outlined in the Greenhouse Gas (GHG) Protocol (WRI and WBCSD, 2004). As
theGHGProtocolwasnotdesignedtobeappliedtoproducts4,theresultscategorizationwasper
formed as closely as possible to the requirements of the GHG Protocol. Following the concept of
scopes,thebreakdownoftheGHGemissionsasdeterminedincompliancewiththeISO14044stan
dard(ISO,2006b)isprovidedforScope1(directGHGemissions),Scope2(indirectGHGemissions
attributable to energy conversion processes) and Scope 3 (further GHG emissions from the supply
chain)5.TheresultsareillustratedinTable15.
Scope1:DirectGHGemissionsoccurfromsourcesthatareownedorcontrolledbythecompany,for
example,emissionsfromcombustioninownedorcontrolledboilers,furnaces,vehicles,etc.;emis
sionsfromchemicalproductioninownedorcontrolledprocessequipment.
Scope2:IndirectGHGemissionsfromelectricityarecomprisedofGHGemissionsfromthegenera
tionofpurchasedelectricityconsumedbythecompany.Purchasedelectricityisdefinedaselectric
TheGHGprotocolisapplicabletothecompaniesonly.
Detailedinformationaboutthestandardandisapplicationareavailablefromwww.ghgprotocol.org.
GHGanalysisandbreakdownintoscope1,2and3

itythatispurchasedorotherwisebroughtintotheorganizationalboundaryofthecompany.Scope2
emissionsphysicallyoccuratthefacilitywhereelectricityisgenerated.
Scope3:OtherindirectGHGemissionsareanoptionalreportingcategorythatallowsforthetreat
ment of all other indirect emissions. Scope 3 emissions are a consequence of the activities of the
company, but occur from sources not owned or controlled by the company. Some examples of
Scope3activitiesareextractionandproductionofpurchasedmaterials;transportationofpurchased
fuels;anduseofsoldproductsandservices.
We recommend that sustainability issues be assessed and addressed by taking the whole
supplychain(orratherthesupplyweb)intoaccount.ThisallowsconsiderationofbothdirectGHG
emissionsaswellasemissionsindirectlyassociatedwiththerespectiveproductionprocesses.
Table15:Scope1,2,and3CO2Equivalentemissionsforprimaryaluminumingotproduction.
kgCO2equiv./tonAluminum
Bauxite
mining
Alumina
refining
Anode pro smelting

duction
casting
Total
AA
IAI
AA
IAI
AA
IAI
AA
IAI
AA
IAI
AA
IAI
Process
178
177
1584
1584
1762
1761
Electricity
11
170
122
19
30
5152
4922 40
36
5392
5115
Fossilfuels
58
23
1409 1346 95
69
84
1657
1522
PFC
1037
1159
1037
1159
Fuelsupplychain 8
238
10
406
15
677
Auxiliarymaterial
279
227
37
543
totalscope1+2
28
1579 1468 292
276
7773
7665 135
120 9848
9558
2096
8216
69
total scope 1+2 77

+3
529
95
150
11068
Scope 1 refers to GHG emissions from the process itself (1762 kg CO2 equivalent), from
burningthefossilfuelsforthermalenergydemandatthefactory(1657kgCO2equivalent)andspe
cificemissionsfromthesmeltingstage(1037kgCO2equivalent).
Scope 2 refers to GHG emissions from upstream processes such as electricity production,
whichcontribute5392kgCO2equivalent.
Scope3referstoGHGemissionscomingfromthesupplychains(677kgCO2equivalent)and
auxiliarymaterials(543kgCO2equivalent),foratotalof1221kgCO2equivalent.
ThesumofScope1and2emissionsindicatestheGHGemissionswhicharetheresponsibil
ityofthefactory,whichforproductionofonemetrictonofprimaryaluminumingotis9848kgCO2
FinalReport:CanLCA
PEAmericas
equivalent6.Thisvalueisinthesamerangeasthe9558kgCO2equivalentfigurementionedinthe
IAI report. However, this figure for Scope 1 and 2 emissions does not reflect the full life cycle im
pacts,andthe1221kgCO2equivalentScope3emissionsmustbeincludedtofindthatthefulllife
cycleglobalwarmingpotentialfromproducingonemetrictonofprimaryaluminumingotis11068
kgofCO2equivalent.
3.3.4
Photochemical Ozone Creation Potential
ThePhotochemicalOzoneCreationPotential(POCP)measurestheemissionsofprecursors
thatcontributetolowlevelsmog(alsocalledSummerSmog),producedbythereactionofNOxand
volatileorganiccompounds(VOC)undertheinfluenceofultravioletlight.POCPisexpressedaskg
Etheneequivalent.
ThePOCPresultsareillustratedinFigure24aswellasinTable16.ThePOCPrelatedtothe
productionofonemetrictonofprimaryaluminuminNorthAmericais3.06kgEtheneequivalent.
Thebreakdownofemissionstoairwhichcontributetosmogcreationpotentialforprimaryalumi
numproductionis59%fromSO2,19%fromnonmethanevolatileorganiccompounds(NMVOC),17
%fromNOx,and2%frommethane.
Liketheotherlifecycleimpactcategories,theelectrolysisprocessisthelargestcontributor
tosmogcreationimpacts,accountingfor77%ofthetotalPOCP.Thisisfollowedbyaluminarefining
whichisresponsiblefor19%ofthenetsmogcreationimpact.
emissions.
Note:thesefiguresindicatesaCO2equivalentforallGHGemissions(totalscore)andisnottobemistakenforeffectiveCO2
Approximately70%ofthesecontributingemissionsareassociatedwithproductionofelec
tricityrequiredforprimaryaluminumingotproduction.

kg Ethene Equiv/metric ton of primary aluminum ingot
4.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
Bauxite Mining
Alumina Production
Electrolysis
Casting
Total
Unit Process
Figure24:Photochemicalozonecreationpotentialresultsforprimaryaluminumingotproduction.
Electrolysis is the largest contributor to smog creation impacts, accounting for 77% of the total
POCP,ofwhich70%areattributedtoelectricityproduction.
Table16:LCIAresultsforproductionof1000kgofprimaryaluminumingotinNorthAmerica.
Impact Assessment Unit
Category
Alumina
Refining
Electrolysis
Cast
house
Total
ton
CO2 0.05
Equiv./ton
2.12
8.74
0.15
11.06
AcidificationPotential kg
SO2 1.80
Equiv./ton
9.16
38.92
0.55
50.43
Eutrophication
Potential
kg
Phos 0.16
phate
Equiv./ton
0.47
1.79
0.05
2.48
Photochemical Ozone kg Ethene 0.10

CreationPotential
Equiv./ton
0.58
2.35
0.04
3.06
GlobalWarming
Potential
FinalReport:CanLCA
PEAmericas
Bauxite
mining
4 BEVERAGE CAN PRODUCTION

Thischapterdescribesinacomprehensivewaythemanufacturingofbeveragecansasrep
resentativefortheU.S.marketconditions.Itincludessubsectionsontheprocessdescription(Chap
ter4.1),datacollectionandmodeling(Chapter4.2),theLCIresults(Chapter4.3)andtheLCIAresults
(Chapter4.4).
4.1 PROCESS DESCRIPTION

4.1.1
Can Sheet
Themanufacturingofthealuminumbeveragecanbeginswiththeconversionofingotsinto
can stock and lid stock coil, which are subsequently converted into can bodies and lids at the can
manufacturingplant.TheprocesschainatarollingmillisillustratedinFigure25.InHotmillrolling,
aluminumingots(approximately18to26inchesthickandweighingapproximately15to30metric
tons)arepreheatedtoabout1000Fandfedthroughahotreversingmill.Inthereversingmill,the
coilpassesbackandforthbetweenrollersandthethicknessisreducedfromtheinitialthicknessto
between1to2incheswithacorrespondingincreaseinlength.Followingthereversemills,theslabs
arefedtoacontinuoushotmillwherethethicknessisfurtherreducedtolessthatinchinthick
ness.Themetal,calledrerollorhotcoil,isrolledintocoilandreadytobetransferredtothecold
mill.
Priortothecoldmill,thecoilsmaybeannealedtogivethemetaltheworkabilityfordown
stream processing. Some plants have moved towards self annealing which requires no additional
energy investment as the industry has improved their energy management. The coils are then
passed through multiple sets of continuous rollers to reduce the gauge to approximately 0.012
inchesrequiredbythecanmakers.Thecoilsareslittothewidthandcuttothelengthrequiredby
canmanufacturers.Thecoilsarepackagedtopreventdamagetothemetalinshipping.
Chapter: Beverage Can Production
Sheetrollingdiffersslightlybasedonthefinaluseofthecansheetforthebodyofthecan
orthelid. Themain differenceisacoatingstepfor sheetsthatareusedinlidmaking.Inputsand
outputsforthesheetrollingprocessareshowninTable17andTable18forthebodysheetandlid
sheetprocessesrespectively.
Figure25:Unitprocesschainrepresentativeforoneparticularrollingmillsite.Theflowchartde
monstrates the assembly of the single process steps in the GaBi4Software in order to quantify
theaverageLCIdataforacansheet.Cansheetmakingisseparatedintotwoparts,relatedtothe
productionofthea)lidandb)thebodycomponents.Thecansheetmakingprocessdiffersinthe
coating which the lids receive. The ratio of lid to body production (by weight) is approximately
22/78.Thissinglemodelwascustomizedforeachofthesurveyedsitesandresultinginventories
wereweightedandaveraged.
Table17:InputandOutputflowsforthecansheetmakingunitprocess(scaledtothebodycom
ponent). Flows are representative for 1000 kg of can sheet production (780 kg of body compo
nent).
Flows
Units
Amount
INPUTS
EnergyandFuels
Thermalenergy(naturalgas)
MJ
3077
Power
MJ
1025
FinalReport:CanLCA
PEAmericas
Thermalenergy(heavyfueloil)
MJ
36.45
Steam(MJ)
MJ
6.95
Thermalenergy(lightfueloil)
MJ
3.079
Thermalenergy(LPG)
MJ
1.262
Kerosene
kg
0.1
Diesel
kg
0.079
Gasoline(regular)
kg
0.011
Heavyfueloil
kg
0.003
AluminumIngot
kg
1072
Steel
kg
0.13
Processwater
kg
835.6
Coolingwater
kg
473.0
Rollingoil
kg
3.603
Lubricant(unspecified)
kg
1.185
Hydraulicoil
kg
0.811
Filtermedia
kg
0.616
Lubricatingoil
kg
0.391
Solvent
kg
0.006
Woodenpallets(EURO,40%moisture)
kg
2.535
Coatings(unspecified)
kg
2.306
Sulphuricacid(100%)
kg
0.651
Cardboard(packaging)
kg
0.303
Polyethylenefilm(PE)
kg
0.141
PaperPlasticComposite
kg
0.073
Metals
OperatingMaterials
OtherMaterials

Foam
kg
0.036
Phosphoricacid
kg
0.024
Chromicacid
kg
0.009
OUTPUTS
Products
kg
780.0
VOC(unspecified)
kg
1.333
Dust(unspecified)
kg
0.136
Nitrogenoxides
kg
0.117
Carbonmonoxide
kg
0.088
Hydrogenfluoride
kg
0.002
Ethylbenzene
kg
0.001
Sulphurdioxide
kg
0.001
Wastewater
kg
2032
Solids(dissolved)
144.2
Solids(suspended)
24.36
Biologicaloxygendemand(BOD)
12.39
Chloride
9.185
Chemicaloxygendemand(COD)
2.479
Aluminumion(+III)
0.144
Aluminum(+III)
4.73E02
Zinc(+II)
2.2E02
Ammonium/ammonia
1.23E03
Canstockbody
Emissionstoair
Emissionstowater
Otherwastes
FinalReport:CanLCA
PEAmericas
Aluminumscrap
kg
291.6
Nonhazardouswasteforlandfilling
kg
4.113
Nonhazardouswasteforfurtherprocessing
kg
0.949
Scrapmetalforrecycling,excludingaluminum
kg
0.405
Hazardouswasteforfurtherprocessing
kg
0.068
Hazardouswasteforlandfilling
kg
0.022
Flows
Units
Amount
INPUTS
EnergyandFuels
MJ
917.7
Power
MJ
636.5
Thermalenergy(MJ)
MJ
47.87
Steam(MJ)
MJ
41.59
Thermalenergy(heavyfueloil)
MJ
10.06
MJ
0.695
Kerosene
kg
0.186
Diesel
kg
0.107
Gasoline(regular)
kg
0.009
AluminumIngot
kg
316.9
Steel
kg
0.034
Processwater
kg
312.0
Coolingwater
kg
108.3
Metals
OperatingMaterials
Table18:InputandOutputflowsforthecansheetmakingunitprocess(scaledtothelidcompo
nent).Flowsarerepresentativefor1000kgofcansheetproduction(220kglidcomponent).

Solvent
kg
6.712
kg
3.706
Rollingoil
kg
3.354
Hydraulicoil
kg
0.229
Filtermedia
kg
0.214
Lubricatingoil
kg
0.15
Epoxyresin
kg
9.265
Woodenpallets(EURO,40%moisture)
kg
3.851
Vinylchloride(VCM;chloroethene)
kg
0.761
Coatings(unspecified)
kg
0.598
Cardboard(packaging)
kg
0.365
Naphtha
kg
0.350
Sodiumcarbonate(soda)
kg
0.304
Sulphuricacid(100%)
kg
0.28
Polyesterresin(unsaturated;UP)
kg
0.223
Succinicaciddibutyleneester
kg
0.180
Xylene(oxylene;1,2dimethylbenzene)
kg
0.179
Phosphoricacid
kg
0.133
OUTPUTS
Products
Kg
220
VOC(unspecified)
Kg
0.395
Nitrogenoxides
Kg
0.063
Dust(unspecified)
Kg
0.027
OtherMaterials
Canstocklidpackaged
Emissionstoair
FinalReport:CanLCA
PEAmericas
Xylene(dimethylbenzene)
Kg
0.020
Carbonmonoxide
Kg
0.019
Trimethylbenzene
3.01
Toluene(methylbenzene)
2.31
Ammonia
1.389
Ethylbenzene
1.233
Methylisobutylketone
7.76
Sulphurdioxide
4.11
Cumene(isopropylbenzene)
2.89
Wastewater
kg
1063
Solids(dissolved)
kg
0.032
Solids(suspended)
20.05
7.798
7.172
Chloride
2.356
Aluminumion(+III)
0.242
Aluminum(+III)
1.23E02
Chromium(unspecified)
2.37E03
Aluminumscrap
kg
95.72
kg
3.030
kg
1.684
kg
1.093
Hazardouswasteforincineration
kg
1.026
Hazardouswasteforfurtherprocessing
kg
0.323
Emissionstowater
OtherWastes

4.1.2
Can Manufacturing
Aluminum coils are shipped from the rolling mills to can manufacturing plants. Within the
canmanufacturingplants,coilsaresetuprightandmovedintopositiontofeedthecuppingpress.
Thecoilisunwoundandlubricatedatarateneededtofeedthepress.Inthecuppingpress,blanks
ordiscsarestampedandthenpressedintocups.Thisprocessresultsingenerationofmanufacturing
(skeleton)scrapwhichisthenshippedbacktosecondarycastingfacilities.Thecupsthenundergoa
series of forming, ironing, and punching operations to form the final profile of the can as per the
desiredspecifications.Inordertoensureaflattop,thecansaretrimmedatthetop.Thetrimscrap
is also shipped back to secondary casting facilities. After trimming, the cans undergo a series of
washingstepsbeforebeingdriedinanoven.Paintsarethenappliedexternallytothecans,thepaint
actingasalabelforthecanproduct.Thecansarethenfurtherinternallycoatedtoestablishabar
rierbetweenthemetalandbeverage.
Thenextstep,neckingofthecan,reducesthediameteroftheopenendofthecantomatch
the diameter of the lid.The diameter of the lid is smaller than the can diameter, allowing for an
overallreductionintheamountofaluminumusedinafinishedaluminumcan.Afterthediameter
hasbeenreduced,theflangethatformspartofthesealtothelidisformed.Thecansthenundergo
aqualitycontrolprocesstochecktheintegrityofthefinalproductbeforetheyareshippedtofillers.
Thecansarestackedontoplasticpalletsinshipping.
Layers of cans are separated using corrugated paper or plastic sheets. The entire pallet is
thenbandedtogetherwithplasticbandsandinrarecasesiscoveredwithshrinkwraptoprotectthe
cansfromdamageanddirtduringbothshippingandstorage.
Can lids are manufactured from a different alloy than can body. Alloys for can lids have
highermagnesiumcontentinplaceofthemanganeseusedinthebodies,aslidsaredesignedtobe
stifferthanthecanbody.Followingthecoldrolling,thecanstocksforlidsiscleanedandcoatedand
shipped to the manufacturer. Lids can be manufactured from either coils or from scrolled sheets.
Themanufacturingstepsareverysimilarindependentofthetypeoffeedsoonlythecoilfedproc
esswillbedescribedhere.Themajorstepsintheprocessarestampingoutends,curlingtheedges
of the shells, applying the sealing compound, stamping tabs, stamping the end features onto the
ends,andfinallyattachingthetabstotheendstomakeacompletedlid.Anillustrationofthecan
making processisshowninFigure26andTable19liststheinputsandoutputsofthe canmaking
process.
FinalReport:CanLCA
PEAmericas
Figure26:IllustrationoftheCanmakingprocess.Flowsarerepresentativefortheproductionof1000cans.Dataiscontributedbytheweighted
averageofinventoriesinthesurveyedsites(seeDataCollectionandModeling).
Table 19: Inputs and outputs for the can making unit process. Flows are representative for the
manufacturingof1000cansfromaluminumsheet.
Flows
Units
Amount
INPUTS
EnergyandFuels
Power
MJ
77.62
MJ
70.37
Thermalenergy(LPG)
MJ
0.6
Diesel
3.329
kg
16.78
Water
kg
85.77
Coatings(can)
kg
0.916
Hydrogenfluoride
kg
0.225
Sulphuricacidaq.(96%)
kg
0.198
Limequicklime(lumpy)
77.20
Lubricatingoil
39.42
Inks(can)
31.25
Polyethylenepart(PE)
13.61
Polypropylenepart(PP)
17.87
Solvent
8.781
OUTPUTS
Products
pcs.
1000
Metals
Aluminumsheet
OtherMaterials
2PCCan
Emissionstoair
FinalReport:CanLCA
PEAmericas
Alcohols(unspec.)
43.77
Ethyleneglycol
23.9
Ether(unspec.)
4.03
Formaldehyde(methanol)
2.02
Naphtha
0.987
Xylene(dimethylbenzene)
2.36E2
Manganese(+II)
1.13E2
Hydrogenfluoride
9.21E3
Ethylbenzene
6.23E4
Methanol
2.47E4
Emissionstowater
Wastewater
kg
58.63
Phosphorus
kg
0.5
Solids(dissolved)
kg
0.11
Sulphate
kg
0.045
kg
0.026
Calcium(+II)
kg
0.02
3.9
Chloride
3.5
Solids(suspended)
3.5
Totalorganicboundedcarbon
2.4
Sodium(+I)
2.1
Oil(unspecified)
1.4
Nitrogen(astotalN)
0.84
Fluoride
0.45
Aluminum(+III)
0.43
Potassium
0.3
Magnesium(+III)
kg
1.4E4
Silicateparticles
kg
7.6E5
Arsenic(+V)
kg
3E5
Strontium
kg
1.4E5
Manganese(+II)
kg
9.2E6
Phenol(hydroxybenzene)
kg
8.3E6
Sulphide
kg
5.2E6
Iron
kg
3.3E6
Zinc(+II)
kg
3E6
Selenium
kg
2.9E6
Copper(+II)
kg
2.6E6
Barium
kg
1.3E6
Chlorine
kg
1.2E6
Lead(+II)
kg
9.1E7
Chromium(unspecified)
kg
9E7
Cadmium(+II)
kg
8.6E7
Cyanide
kg
5.2E7
Nickel(+II)
kg
3.5E7
Silver
kg
2.9E7
Mercury(+II)
kg
1.1E8
Aluminumscrap
kg
3.447
Totalwasteforincineration
kg
0.225
Sludge
kg
0.174
Waste(recycling)
kg
0.13
OtherWastes
FinalReport:CanLCA
PEAmericas
Waste(incineration)
49.66
Waste(landfill)
44.03
Sludge(fromprocessing)
26.31
Hazardouswasteforincineration
0.752
4.1.3
Secondary Aluminum Ingot / Recycling
Thesecondaryaluminum productionprocessincludestheunitprocessesofshredding,de
coating,remeltingandsecondaryingotcastingasshowninFigure27.Thesecondaryaluminumin
gotisproducedfrompostconsumerscrap(UBC)recoveredfromtheconsumerwastestream.The
UBCs may be collected in a municipal curbside program and dropped off by individuals or groups
interestedingeneratingrevenue.
UBCsmustbetreatedpriortotheirmeltinginafurnace.First,theUBCsareshreddedtore
movetrappedwaterandothercontaminants.Theuniformsizeoftheshredshelpsmaterialflowin
downstreamprocessing.Theshredsarepassedundermagneticseparatorstoremoveferrouscon
tamination.Insomefacilities,airknivesarealsousedtopreventtheinclusionofheavycontamina
tionsuchaslead,stainlesssteel,orzinc.
Themetalleavestheshreddersandpassesintoadecoatingunit.Thisunitheatsthemetal
andcoatings,resultinginthevaporizationandoxidationofthecoatings.Thedecoatingprocessre
sultsinthetransferofthehotmetaltothemeltingfurnace.Primaryaluminummetalisconsumed
tomakeupforsystemmeltlossandsweetenthecompositionifnecessary.Inaddition,alloyingad
ditivesarealsoaddedtothefinalspecificationsoftheingottobeproduced.Thecastingprocessis
similartotheprocessdescribedinthesectiononPrimaryIngotCasting(CastHouse)(Chapter3.1.5).
Thescrappreparationisseparatedfromremeltingandcastingandunitprocessinformation
isshowninTable20andTable21.Itisimportanttonotethatthesenumbersarespecificforalumi
numingotsusedincanmakingandinnowayarerepresentativeofthealuminumingotproduction
ingeneral(AA).
Figure27:Unitprocesschainatoneselectedsecondaryaluminumproductionsiteillustratingthe
attempttoquantifytheLCIdataforsecondaryaluminumingotproduction
FinalReport:CanLCA
PEAmericas
Table20:Inputandoutputflowsforscrappreparationunitprocesses(i.e.shredding&decoating).
The figures indicate the average overall surveyed sites. Flows are representative for 1000 kg of
preparedscrap.
Units
INPUTS
EnergyandFuels
Amount
Thermalenergy(MJ)
MJ
281.22
Power
MJ
30.91
kg
1013
kg
0.656
OUTPUTS
Products
kg
1000
Hydrogenchloride
6.21
Nitrogendioxide
2.071
Sulphurdioxide
kg
8.63E06
VOC(unspecified)
3.667
Otheremissionstoair
kg
3.97E10
Dust(unspecified)
23.49
AluminumFines
kg
5.103
kg
0.1
Baghouselime
kg
0.042
kg
2.791
Metals
Aluminumscrap
OperatingMaterials
Calciumhydroxide
Aluminumscrap(processed)
Emissionstoair
Otherwastes
Flows
UBCContamination
kg
6.57
Table21:Inputandoutputflowsforremeltingandcastingunitprocesses.Thefiguresindicatethe
averageoverallsurveyedsites.Flowsarerepresentativefor1000kgofrollingingot.
Flows
Units
INPUTS
Amount
EnergyandFuels
MJ
1890
Power
MJ
1022
Thermalenergy(LPG)
MJ
15.56
MJ
1.481
Diesel
kg
0.704
Power(Fromwaste)
MJ
0.020
Gasoline(regular)
kg
0.012
Kerosene
1.074
Thermalenergy(propane)
MJ
6.53E04
Aluminum
kg
1046
Alloycomponents
kg
5.669
Coolingwater
kg
848
Filtermedia
kg
0.972
Hydraulicoil
kg
0.823
kg
0.496
Processwater
kg
716.5
Salt(flux)
kg
0.176
Metals
OperatingMaterials
FinalReport:CanLCA
PEAmericas
OtherMaterials
Sodiumchloride(rocksalt)
kg
2.564
Potassiumchloride
kg
2.564
Sodiumchloride(rocksalt)
kg
0.6
Sodiumhydroxide(100%;causticsoda)
kg
0.097
Sulphuricacid(100%)
kg
0.867
Cryolite
kg
0.214
Sodiumhypochlorite
kg
0.014
Refractory
kg
0.007
OUTPUTS
Products
kg
1000
Dust(unspecified)
kg
0.121
Nitrogenoxides
kg
0.112
Hydrogenchloride
kg
0.106
Nitrogendioxide
kg
0.102
Carbonmonoxide
kg
0.083
Sulphurdioxide
4.336
Hydrogenfluoride
3.451
Chlorine
1.62
Otheremissionstoair
1.355
Ammonia
0.444
Lead(+II)
kg
4.51E06
Dioxins(unspec.)
kg
1.5E11
Aluminumrollingingot
Emissionstoair
Emissionstowater
Wastewater
kg
1637
Solids(dissolved)
kg
0.128
kg
0.021
kg
0.013
Chloride
9.47
Suspendedsolids,unspecified
3.014
Oil(unspecified)
2.08
Solids(suspended)
1.755
Aluminum(+III)
0.844
Aluminumion(+III)
0.166
Ammonia
kg
3.61E05
Zincion(+II)
kg
2.15E05
Zinc(+II)
kg
4.51E06
Cyanide
kg
4.51E06
Dross
kg
39.37
kg
28.33
kg
6.922
Coatedscrap(All5XXXseriesalloy)
kg
3.526
Scrapmetalforrecycling,excludingaluminum
kg
1.07
Baghouselime
kg
0.560
AluminumFines
kg
0.275
Otherwastes
Postindustrialscrapormanufacturingscrapisproducedatrollingmills,canmanufacturing
plants,andcanfillers(excludedfromdatacollectioninthisstudy).Intherollingmills,scrapismainly
generatedbytrimmingtheendsandsidesofthesheetduringtheproductionofthecoils.Withincan
manufacturing, skeleton scrap is generated in the stamping process to manufacture the cups. The
scrapfromrollingmillsandcanmanufacturingplantsistreatedinasimilarmannersinceneitherhas
FinalReport:CanLCA
PEAmericas
beencoated(i.e.itiscleanscrap).Intheunderlyingmodel,thisscrapisconsideredtobedirectlyre
meltedintoingotsandfedbackintotherollingprocess.
4.2 DATA COLLECTION AND MODELING

4.2.1
Can Sheet
In total, four sites provided data on can sheet production for this study. Overall, the data
qualityandtheconsistencyofthecollecteddatahavebeenidentifiedtobeofmediumtohighqual
ity.Participatingsitesrepresentthefollowinglistingofcansheetproducers:
AlcoaTwosites
LoganAluminum(ajointventureofNovelisInc.andARCOAluminum)Onesite
WiseAlloysOnesite
Intotal,about1.7millionmetrictonsperyearofcansheetproductionarerepresentedby
theprovideddata.
TheprocessmodelforeachofthereceiveddatasetswascreatedintheGaBi4softwaresys
tem database (release GaBi 4.3, http://www.gabisoftware.com). The following data requirements
forupstreamsupplieshavebeenapplied:
Energysupply,includingelectricityandfuels:Regionalelectricitygridmixes(representative
ofthestate/regioninwhichtherollingmillislocated)andU.S.specificfuelsdataavailablein
GaBi4databasehavebeenused.
Aluminum ingots: For primary aluminum ingots, the inventory represented in section 3.1
whichisbasedontheIAIdata,hasbeenused.Forsecondaryingots,twoseparateLCIpro
fileswereestimated.Thetwoprofilesrepresenttheproductionofsecondaryaluminumin
gotfromUBCscrap.Manufacturingscrapisconsideredtobedirectlyremeltedandfedback
intotherollingprocess.
Processmaterials:DatafromtheGaBi4softwaresystemdatabasehavebeenused.
4.2.2
Can Manufacturing
Inthisstudy,themanufacturingofcanswasmodeledbaseduponthedataprovidedbythe
CanManufacturersInstitute(CMI).CMIprovidedaggregatedindustryaverageinformationonaper
onethousandcanbasis,representativeforoperationsintheU.S.Italsoprovideddataondirectraw
material,ancillary,andenergyinputsaswellasdirectreleasestotheenvironmentfromcanmanu
factureoperations.Overall,thedataqualityandconsistencyoftheprovideddatahasbeenidenti
fiedtobeofhighquality.Thedistributionoftheproductionvolumebydifferentcansizesin2006is
illustrated in Table 22. It can be observed that the 12 oz (including all variants) beverage can ac
countsforapproximately89.37%ofthetotaltwopiecealuminumbeveragecanproductioninthe
U.S.
In addition, transportation burdens (transport of primary and secondary ingots to rolling

mills and transport of aluminum can sheet to can manufacturing plants) were also included. The
data on transportation was based upon information provided by companies and/or informed esti
matesandaregiveninAppendixB
Table22:Distributionofthecanproductionvolumebycansizein2006[Source:CMI]
CanSize
Productionvolume(%)
8oz
3.48
12oz
89.37
16oz
3.96
Others
3.19
Thesystemboundaryformodelingthecanmanufacturinginthisstudywasdefinedfromde
liveryofaluminumsheetstocanmanufacturerstopackagedcanbodiesandlidsleavingthegateof
the can manufacturing facility. The following data requirements for upstream supplies have been
applied:
Energy supply, including electricity and fuels: CMI provided the statewise breakdown of
electricityconsumptiondata.Thestatewiseelectricitydatawasaggregatedtoestimatere
gionalbreakdownsofelectricityconsumption.Aregionalpowermixprocessona1MJbasis
wascreatedutilizingtheregionalgridmixesdatasetsavailableinGaBi4LCIdatabaseforthe
U.S.ThepowermixprocessisillustratedinFigure28.Thefueldatasets(e.g.diesel,natural
gas)fromtheGaBi4databasehavebeenused.
Aluminumsheet:Thealuminumsheetinventorygeneratedwithinthisstudyhasbeenused,
representingthemostuptodateinformationonaluminumsheetproduction.
Coating,Inks,andSolvents:CMIprovidedrepresentativeinformationonthecompositionof
coatings,inksandsolvents.Inviewoftheconfidentialityofthecompositionmix,furtherde
tails cannot be provided in this report. Representative datasets from GaBi 4 were used to
createthecompositionmixforcoatings,inks,andsolvents.
Other process materials: Data from the GaBi 4 databases have been used to characterize
otherancillarymaterialsusedincanmanufacturing.
Intotal,fivesitesprovideddataonsecondaryaluminumproductionforthisstudy.Overall,
thedataqualityandconsistencyofthecollecteddatahasbeenidentifiedtobeofmediumquality.
Participatingsitesrepresentthefollowinglistingofsecondaryaluminumproducers:
AlcoaTwosites
AlerisOnesite
LoganAluminum(ajointventureofNovelisInc.andARCOAluminum)Onesite
NovelisOnesite
Intotal,about2million metrictonsperyearofsecondaryaluminumingotproduction are

representedbytheprovideddata.
FinalReport:CanLCA
PEAmericas
TheprocessmodelforeachofthereceiveddatasetswascreatedintheGaBi4softwaresys
tem database (release GaBi 4.3, http://www.gabisoftware.com). The following data requirements
forupstreamsupplieshavebeenapplied:
Energysupply,includingelectricityandfuels:Regionalelectricitygridmixes(representative
of the state/region in which a secondary aluminum production facility is located) and U.S.
specificfuelsdataavailableintheGaBi4databasehavebeenused.
Aluminumingots:Forrecycledscrapingots(RSI),datasetfromtheGaBi4databasehasbeen
used.
Process materials: Data from the GaBi 4 software system database (release GaBi 4.3,
http://www.gabisoftware.com)havebeenused.
Inaddition,transportationburdens(transportofprimaryingotstothefacility,transportof
manufacturingandUBCscrap)wereincluded.Thedataontransportationwasbaseduponinforma
tionprovidedbythecompaniesand/orinformedestimatesandaregiveninAppendixB.
Figure28:PowerMixProcessfortheCanManufacturingModel.Flowsarerepresentativeforthe
productionof1000cans.
FinalReport:CanLCA
PEAmericas
4.3 LCI RESULTS (AVERAGED GATE-TO-GATE INVENTORIES)

Inthischapter,theimportantLCIresults(gatetogate)arepresentedfortheproduction
ofbeveragecansintheU.S(i.e.provisioningofprimarymetalisnotincludedintheseresults).The
LCIresultsper1000canscanberepresentedonthebasisof1000oz.ofvolumeoffillbydividingthe
resultsbyafactorof12.408.
4.3.1
Beverage Can Sheet
TheaveragedgatetogateLCIofaluminumcansheets(representativeofcanbodyandlid
stock)isgiveninTable23.ThecompanyspecificverticalLCIprofileswereaveragedusingthepro
ductionvolumeastheweightingcriteriatoestimatetheindustryLCIprofileofcansheetproduction
intheU.S.InTable23,onlyselectedinputsandoutputshavebeenhighlighted.Thedominantfuel
used is natural gas (refer to unit process information) and this represents sound environmental
practices since use of other fossil fuels such as coal or fuel oil would result in higher combustion
relatedemissionstoair.
Inventoryresultspertonof
Unit
sheet
Primaryenergydemand
13.95
GJ
13.58
GJ
0.369
GJ
1389
Kg
Carbondioxide
870
Kg
Carbonmonoxide
0.328
Kg
Nitrogenoxides
1.931
Kg
Sulphurdioxide
2.741
Kg
VOCs
3.27
Kg
INPUTS
Aluminumrollingingot
OUTPUTS
Inventoryresultspertonof
Unit
sheet
907.2
kgCO2Eq.
4.118
kgSO2Eq.
Table23:AveragedLCIdataforcansheet(mix)productionprocess(1000kgofaluminumsheet)
representinggatetogateinformation.
0.281
kgPhosphateEq.
Photochemical Ozone creation potential

0.488
(POCP)
kgEtheneEq.
OzoneLayerdepletionpotential(ODP)
kgR11Eq.
3.44E05
4.3.2
Can Manufacturing
Thegatetogateinventoryparametersofthecanmanufacturingmodelareillustratedin
Table24.Thisinventorycharacterizesonlyselecteddirectrawmaterial,ancillary,andenergyinputs
and the direct releases to the environment for the can manufacturing process. The included air
emissions result from the combustion of natural gas and liquefied petroleum gas and the use of
diesel,coatings,inks,andsolventsduringcanmanufacturing.
Theaverageweightof1000canswasestimatedusingtwoapproaches.Inthefirstapproach,
the weight was estimated as the difference of the total aluminum sheet input and the amount of
scrapgeneratedduringcanmanufacturing.Usingthisapproach,theaverageweightper1000cansis
13.3356 kg. Alternatively, the average weight per 1000 cans can be determined by taking the
weightedaverageoftheweightofdifferentcansizes.Theaverageweightfordifferentcansizeswas
providedbyCMI.Usingthesecondapproach,theaverageweightof1000canswasestimatedtobe
13.534kg.Therefore,thedifferenceinweightper 1000cansfortwoapproachesisapproximately
0.1984kg.Thecoregroupoftheprojectrecommendedusingtheestimateof13.3356kgper1000
cansinthisstudy.
Table24:AveragedLCIdataoncanmanufacturingprocess(per1000cans)representinggateto
gateinformation
Inventory results per 1000

Unit
cans
Primaryenergydemand
404.5
MJ
386.3
MJ
18.2
MJ
16.78
Kg
Carbondioxide
24.62
Kg
Carbonmonoxide
8.439
Nitrogenoxides
49.87
INPUTS
Cansheet
OUTPUTS
FinalReport:CanLCA
PEAmericas
Sulphurdioxide
78.36
VOCs
0.1137
Kg
Inventory results per 1000

Unit
cans
25.07
kgCO2Eq.
0.116
kgSO2Eq.
1.538
kgPhosphateEq.
Photochemical Ozone creation potential

0.017
(POCP)
kgEtheneEq.
kgR11Eq.
4.3.3
1.87E06
Secondary Aluminum Ingot / Recycling
TheaveragedgatetogateLCIofsecondaryaluminumproductionisillustratedinTable25
andTable26.ThecompanyspecificverticalLCIprofileswereaveragedusingproductionvolumesas
theweightingcriteriatoestimatetheindustryLCIprofileofsecondaryaluminumingotintheU.S.In
the following tables, only selected inputs and outputs have been compiled and the included air
emissionsarisefromtheconsumptionofnaturalgasanddiesel.Theprimaryenergysourceisnatu
ralgas(refertounitprocessinformation).
Inventory results per ton of

Unit
preparedscrap
Primaryenergydemand
437.4
MJ
435.1
MJ
2.282
MJ
1013
Kg
Carbondioxide
28.5
Kg
Carbonmonoxide
88.46
Nitrogenoxides
52.55
INPUTS
Aluminumscrap
OUTPUTS
Table25:AveragedLCIdataforscrappreparation(1000kgofpreparedscrap)representinggate
togateinformation
Sulphurdioxide
53.40
VOCs
49.38

Unit
preparedscrap
29.49
kgCO2Eq.
97.45
gSO2Eq.
7.364
gPhosphateEq.

(POCP)
gEtheneEq.
kgR11Eq.
9.13E07
Table26:AveragedLCIdataforremeltingandcasting(1000kgofsecondaryaluminumingot)rep
resentinggatetogateinformation

Unit
secondaryingot
Primaryenergydemand
6451
MJ
6279
MJ
172.9
MJ
1046
Kg
Carbondioxide
425.5
Kg
Carbonmonoxide
0.107
kg
Nitrogenoxides
0.984
kg
Sulphurdioxide
1.481
Kg
VOCs
0.655
Kg
INPUTS
Aluminum
OUTPUTS
FinalReport:CanLCA
PEAmericas

Unit
secondaryingot
431.7
kgCO2Eq.
2.348
kgSO2Eq.
0.148
kgPhosphateEq.

(POCP)
kgEtheneEq.
kgR11Eq.
2.05E05
4.4 LCI AND LCIA RESULTS PER 1000 CANS

Thissection presentsthe LCIAresultsforthebeveragecansfor the twoEoL approaches
closedloopandrecycledcontent.Abaselinescenarioisdefinedfortheapproachesandtheparame
tersand/orassumptions:
Averagecanweightof13.34kgper1000cans
TheUBCrecyclingrateintheyear2006is51.6%(AA,2007)
TherecycledcontentofthebeveragecanintheU.S.in2007is67.8%
Postproductionandrunaroundscrapfromrollingmillsiscompletelyrecycledinaclosed
loop(asseeninFigure1).Theproductionofsecondaryaluminumingotfrompostproduc
tionandrunaroundscrapdoesnotrequireanyinputofprimaryaluminum.
CanmanufacturingscrapisdealtwithinthesamemannerasUBCscrapandincludedinthe
recycledcontentofthecan.
TheLCIAresultsper1000canscanberepresentedonthebasisof1000oz.ofvolumeoffill
bydividingtheresultsbyafactorof12.41.
4.4.1
Closed Loop Approach
Underaclosedloopapproach,themassflows(per1000cans)overthecompletelifecycleof
thebeveragecanareshowninFigure29.
The production of 1000 cans at a can manufacturing plant requires 16.78 kg of aluminum
sheet(inclusiveofcanbodyandlid).Thecanmanufacturingprocessyields3.447gofpostproduc
tionscrap.Theamountofpostproductionscrapgeneratedatrollingmillsis6.501kg.
Inordertoproduce16.78kgofaluminumsheet,thetotalamountofaluminumingotinput
requiredfortherollingprocessis23.31kg.Thisconsistsof:
6.218kgofingotproducedfrompostproductionscrap(fromrollingmills)and
Theenergyforprimaryingotproductionrepresentedhereinallconsiderationsincludespart
ofthealuminumoutputasliquidmetal.Thismeansthatnotallaluminumoutputfromthisprocess
isiningotformandtherefore,thesenumberscannotbedirectlycomparedwiththosestatedearlier
(e.g.Table14)forprimaryaluminumproduction.
17.09 kg secondary ingot with 67.8% scrap input (excluding rolling scrap) to the remelting
process
AttheEndofLife(EoL),6.984kgofUBCscrap(51.6%oftheaveragecanweightof13.34kg
per 1000 cans) is recovered. 12.23 kg of scrap is required, of which only 10.433 kg are recovered
fromcanmakersandUBCcollectioncombined.Thedeficitof1.797kginthelatterareassigneda
net burden which amount to the corresponding amount of primary metal needed as a replace
ment,minustheburdensforscrappreparationandremelting.TheselectedLCI/LCIAresultsofthe
beveragecanundertheclosedloopapproachareshowninTable27.
Withregardtotheprimaryenergydemandintheclosedloopsystem,thetotalprimaryen
ergydemandper1000canswitha51.6%UBCrecyclingrateis1,943MJ.Thebreakdownshareof
totalprimaryenergydemandoveralllifecyclestagesofthecanisshowninFigure30.Theprimary
energydemandfortheproductionprocess(includingtheupstreamprocessessuchasrawmateri
alsandenergy)arepresentedforboththeprimaryandthesecondaryaluminumingotproduction.
Similarly,theimpactscoresfortherollingandcanmanufacturingaregiven.Thefractionofrenew
ableenergysourcesisshownaslightshadedbars;thefractionofnonrenewableenergysourcesis
shownasfullbars.
Overall, a majority of the primary energy demand is met with nonrenewable energy re
sources.AsdiscussedpreviouslyinSection3.2.1onprimaryenergydemand,theendenergyusedat
the operation sites is dominated by hydropower (which generates power based upon renewable
resources).However,thesignificantlylowerconversionefficiencyofotherpowergenerationplants
that use nonrenewable energy sources (such as hard coal) causes the impact score to be higher
thanthatassociatedwiththerenewableenergyresources.Thecalculationsaredevelopedindetail
inAppendixA.
ExaminingtheLCIresultsforCO2emissiontoair,itcanbeobservedthattheproductionof
primary aluminum ingot has a significant influence on the overall environmental impact score be
causeofitshighprimaryenergyconsumption.
FinalReport:CanLCA
PEAmericas
Figure 29: Mass flows (per 1000 cans) over the complete life cycle of the beverage can under
closedloopapproach(UBCrecyclingrate51.6%)
Thetotalamountofcarbondioxideemissionsper1000cansisaround122kg.Abreakdown
ofemissionsfromthedifferent canmanufacturingstageswoauldshowthatCO2emissionsclosely
matchtheprimaryenergydemandtrendsshowninFigure30,and,therefore,isnotshownhere.

TheprocessbreakdownshowsthatforbothprimaryenergydemandandCO2emissionsto
air,theproductionofprimaryaluminumisthekeyimpactsource,followedbythecanmanufactur
ingwiththerespectivesharesofapproximately67%and20%.Withregardtothesecondaryalumi
numproduction,itisnoticedthatthisprocessplaysaminorroleincontributingtotheprimaryen
ergydemandandCO2emissions.Adetailedexaminationofenergyconsumptionrequiredforsecon
daryaluminumproductionshowsthatmostenergyrequiredforsecondaryproductiongoestopro
ductionfromUBCscrap(around6%totheoverallresults),andonlyaverysmallsharegoestothe
remeltingofthepostproductionscrap(lessthan1%).
Table27:SelectedLCI/LCIAresultsper1000cansunderclosedloopapproach
Primaryenergydemand
Results
Unit
1943
MJ
1540
MJ
403.3
MJ
16.78
Kg
Carbondioxide
121.6
Kg
Carbonmonoxide
0.047
Kg
Nitrogenoxides
0.24
Kg
Sulphurdioxide
0.436
Kg
VOCs
0.209
Kg
131.5
kgCO2Eq.
0.613
kgSO2Eq.
1.565
kgPhosphateEq.
PhotochemicalOzonecreationpotential(POCP)
0.051
kgEtheneEq.
3.76E06
kgR11Eq.
INPUTS
Cansheet
OUTPUTS
FinalReport:CanLCA
PEAmericas
Figure30:Shareoftotalprimaryenergydemand(MJper1000cans)overthelifecycleofthecan
under closed loop approach. The solid color portion of each bar represents the nonrenewable
fractionofprimaryenergyandthelightshadedportionrepresentstherenewablefractionofpri
maryenergy.Expectedly,primaryaluminumproductionhasthehighestprimaryenergydemand
comparedtoallotherprocesses.
4.4.2
Recycled content approach
Themassflows(per1000cans)overthecomplete lifecycleofthebeverage canmodeled

underarecycledcontentapproachareshowninFigure31.
Examining how this change affects our primary LCI metrics, the total primary energy de
mandisfoundtodecreaseby251MJto1692MJper1000cans,whiletheCO2emissionsdecrease
by16kgto106kgCO2per1000cans.Abreakdownofprimaryenergydemandbyenergysourceand
lifecyclestagesforthissystemmodelisshowninFigure32.ExceptforremeltingofsurplusUBCand
grosscreditvalues,therelativeshareoftotalprimarydemandofotherlifecyclestagesisnotdiffer
entfromtheclosedloopresults.Theprimaryenergydemandfortheproductionprocesses(includ
ing the upstream processes such as raw materials extraction and energy production) are pre
sented for both the primary and the secondary aluminum ingot production. Again, the fraction of
renewableenergysourcesisshownaslightshadedbars,andthefractionofnonrenewableenergy
sourcesareshownasfullbars.
Themassflowsper1000cansaresimilartotheclosedloopapproachflows,exceptforthe
deficitUBCscrapflows.Intherecycledcontentapproach,thesurplusUBCscrapflowisassignedtoa
scrapsinkandthereisnonetburdengiventotheproductsystem.Thescrapsinkcanbecon
sideredasastockpileofsecondaryrawmaterialwhichisnotutilizedforfurtherprocessingandre
coveryofthematerial.Asaresult,theenvironmentalburdensofthecanproductsystemdecreases
comparedwiththeclosedloopsystem.
Figure31:Massflows(per1000cans)overthecompletelifecycleofthebeveragecanunderrecy
cledcontentapproach(UBCrecyclingrate51.6%)
FinalReport:CanLCA
PEAmericas
Table28:SelectedLCIandLCIAresultsper1000cansobtainedwiththerecycledcontentapproach
Primaryenergydemand
Results
Unit
1692
MJ
1374
MJ
318.3
MJ
16.78
Kg
Carbondioxide
105.9
Kg
Carbonmonoxide
0.041
Kg
Nitrogenoxides
0.211
Kg
Sulphurdioxide
0.376
Kg
VOCs
0.188
Kg
113.8
kgCO2Eq.
0.532
kgSO2Eq.
1.561
kgPhosphateEq.
PhotochemicalOzonecreationpotential(POCP)
0.046
kgEtheneEq.
3.61E06
kgR11Eq.
INPUTS
Cansheet
OUTPUTS
Figure32:Relativeshareoftotalprimaryenergydemand(MJper1000cans)overthelifecycleof
thecanunderrecycledcontentapproach.Thesolidcolorportionforeachbarrepresentsthenon
renewablefractionofprimaryenergyandthehashedportionrepresentstherenewablefraction
ofprimaryenergy.
4.5
UBC recycling scenarios
TheinfluenceoftheUBCrecyclingrateontheenvironmentalperformanceofthebeverage
can product system wasevaluated by considering additional scenarios for the following UBC recy
clingrates:
30%HypotheticallowUBCrecyclingrate
45.1%UBCrecyclingrateestimatedbyContainerRecyclingInstitute(ca.2004)(CRI,2008)
62%PeakUBCrecyclingrateachievedduringthemid1990s
75%AluminumAssociationfuturegoalforUBCrecyclingrate
The lower hypothetical recycling rate scenario was selected to interpret the results, when
therecyclingrateislowerthantherecycledcontent.Thevariationintotalprimaryenergydemand
per1000cansunderdifferentscenariosisshowninFigure33andtheresultsarepresentedforboth
closedloopandrecycledcontentapproaches.AtahigherUBCrecyclingrate(i.e.75%),theclosed
loopapproachseemsmorefavorable,whereasatalowerrecyclingrate(i.e.30%),therecycledcon
tentseemsmorefavorable,astheproductsystemwouldstilltrytomaintaintherecycledcontentof
thecan.ThereasonisthattherecycledcontentapproachcutsoffbothsurplusUBCscrap(noaddi
tional benefit or credit) as well as an unsaturated scrap demand in case of a recycling rate lower
thantherecycledcontent(additionalscrapnecessarycomeswithoutburden). Asaresult,thepri
mary energy demand under recycled constant approach does not change as the recycling rate is
varied.ItislikelythatrecycledcontentofthecanwillincreaseastheUBCrecyclingrateincreases,
FinalReport:CanLCA
PEAmericas
undertheassumptionthatmoreUBCscrapwillbeavailableinthemarketforcanmaking.However,
there are other market forces (e.g. export of UBC scrap to other countries, use of UBC scrap for
automotivesheetproduction)whichcaninfluencetheavailabilityofUBCscraptosecondaryalumi
numproducers.Theinfluenceofvariousfactorswhichdeterminetherecycledcontentofthecanin
UnitedStatesisbeyondthescopeofthisstudy.Therefore,inthestudytherecycledcontentofthe
canwaskeptconstantat67.8%andonlytheUBCrecyclingratewasvariedinscenarioanalysis.
Figure33.Primaryenergydemandresultsper1000cansunderdifferentUBCrecyclingratescenar
ios for both closed loop and recycled content approaches. The solid color portion for each bar
representsthenonrenewablefractionofprimaryenergyandthedashedportionrepresentsthe
renewablefractionofprimaryenergy.
5 CONCLUSIONS
ThisstudyprovidestheAluminumAssociationanditsmembercompanieswithanupdated
LCIandLCIAofprimaryaluminumproductioninNorthAmericaandaluminumbeveragecanproduc
tionintheU.S.
Thisstudyquantifiesallthesignificantinputsandoutputstothebeveragecansystemunder
two approaches to modeling endoflife impacts: closed loop and recycled content. The system
boundary of this life cycle assessment for beverage cans includes primary aluminum production,
secondaryaluminumproduction;aluminumcansheetproduction,canmanufacturing,andrecycling
ofUBC.
Information on the primary energy demand for primary aluminum production in North
Americashowsthat67%comesfromnonrenewableresources.Electrolysisaccountsfor80%ofthe
totalenergydemandforprimaryproduction.Itisestimatedthat11.1metrictonsofCO2areemitted
pertonofprimaryaluminumingotproducedofwhich8.7tons(78%)arefromtheelectrolysisproc
ess alone. The carbon dioxide emissions profile is similar to the energy profile, since the greatest
contributiontogreenhousegasesisassociatedwiththecombustionoffossilfuelsforheatatpro
duction sites and upstream environmental burdens associated with generation of electricity. The
electrolysisprocessduringprimaryaluminumproductionintheU.S.consumesapproximatelytwo
thirdsofitselectricitydemandfromhydropower.Thistranslatesintoarelativelylowercarbonfoot
print compared to the use of electricity generated from fossil fuels. However, on account of the
lowerenergyconversionefficiencyoffossilfuelbasedpowergeneration,therenewablefractionof
totalprimaryenergydemandislowerthanthenonrenewablefraction.Afurtheranalysisofgreen
house gas (GHG) emissions was done following the guidelines in the GHG Protocol (WRI and
WBCSD).Scopes1and2(directGHGemissionsandindirectGHGemissionsattributabletoenergy
conversionprocesses)togethercontributeto9,847kg.CO2equivalentsemittedpertonofprimary
aluminum produced while Scope 3 (further GHG emissions from the supply chain) adds another
1,221 kg. CO2 equivalents to these emissions. Scopes 1 and 2 indicate the emissions that are the
directresponsibilityoftheproductionfactoriesandtheresultsfollowcloselywithpreviousstudies
(IAI).
Concerningtheendoflifeconsiderations,theresultsofthestudyindicatethattherawma
terial extraction and processing represent 67% of the total primary energy demand (1943 MJ per
1000 cans) under a closed loop approach, with production of the primary aluminum ingot alone
accountingfor46%ofnetprimaryenergydemand,andproductionofthesecondaryaluminumingot
(incl.scrappreparation)accountingfor8%ofnetprimaryenergydemand.Undertherecycledcon
tent system model, the contribution of raw material acquisition to total primary energy demand
(1692MJper1000cans)decreasesslightlyto62%.Thisisduetothenetburdenwhichisgivento
deficitUBCscrapintheclosedloopsystemmodel.
FinalReport:CanLCA
PEAmericas
BIBLIOGRAPHY
AA. (2007). Aluminum Can Reclamation. Retrieved April 2008, from www.aluminum.org:
http://www.aluminum.org/Content/ContentFolders/AssociationHeadlines/June2007/UBC_Recycling
_Rate_2006.pdf
AA.(1998).LifeCycleInventoryReportfortheNorthAmericanAluminumIndustry.Washing
tonD.C.:TheAluminumAssociation.
Anseen,A.G.,Okstad,S.,Innvar,R.,&Olsen,L.(1979).OperationofSoderbergElectrodes.
ElkemSeminarinSmelting.RiodeJaneiro.
Artherton,J.(2007).C.InternationalJournalofLifeCycleAssessment,12(1),5960.
Bergsdal,H.,Strmann,A.H.,&Hertwich,E.G.(2004).TheAluminiumIndustryEnviron
ment,TechnologyandProduction.NTNUProgramforindustriellkologiRaport.No.:8/2004.
CML.(2001).LifecycleassessmentAnoperationalguidetoISOstandards,Vol1,2and3.
Leiden:CenterforEnvironmentalScience,LeidenUniversity.
CRI.(2008).AluminumRecyclingRates.RetrievedApril2008,fromContainerRecyclingInsti
tute:http://www.containerrecycling.org/alum_rates.htm
Droy,B.,&Michaux,D.(2003).PatentNo.US6,555,076,B1.UnitedStates.
ECOBILAN. (2001). Ecoprofile of high volume commodity phthalate esters
(DEHP/DINP/DIDP).TheEuropeanCouncilforPlasticisersandIntermediates(ECPI).
Frank,W.B.,Haupin,W.E.,Dawless,R.K.,Granger,D.A.,Wei,M.W.,Calhoun,K.J.,etal.
(2008).Aluminum.UlllmannsEncyclopediaofIndustrialChemistry.JohnWiley&Sons,Inc.
Grjotheim, U., & Kvande, H. (1993). Introduction to Aluminum Electrolysis. Understanding
theHallHeroultProcess.AluminumVerlagGmbH,260.
IAI. (2007). Historical IAI Statistics. Retrieved November 2007, from www.world
aluminum.org:http://www.worldaluminum.org/Statistics/Historical+statistics
IAI.(2005).LifecycleassessmentofAluminum:InventorydataforthePrimaryAluminumIn
dustry,Year2005Update.London:InternationalAluminumInstitute.
ISO.(2006a).InternationalStandard,ISO14040,Environmentalmanagementlifecycleas
sessmentprinciplesandframework,2006.Geneva:InternationalStandardOrganization.
Mylona,E.,Kalamboki,T.,&Xenidis,A.(2003).ProcessingofBauxiteOresin:BAUXITEAND
ALUMINAPROCESSINGMETHODANDTAILINGSPRODUCTION.RetrievedfromMineralIndustryRe
searchOrganisation:http://www.miro.co.uk/tailsafeif/restricted/docs/wp1_2_6_bauxite.pdf
USGS. (2006). 2005 Minerals Yearbook Aluminum. Washington, D.C.: United States Geo
logicalSurvey,DepartmentofInterior.
USGS. (2007). 2005 Minerals Yearbook Bauxite and Alumina. Washington, D.C.: United
StatesGeologicalSurvey,DepartmentofInterior.
Chapter: Bibliography
ISO.(2006b).InternationalStandard,ISO,14044,Environmentalmanagementlifecycleas
sessmentrequirementsandguidelines,2006.Geneva:InternationalStandardOrganization.

USGS.(2008).2007MineralsYearbookBauxiteandAlumina[AdvanceRelease].Washington,
D.C:UnitedStatesGeologicalSurvey,DepartmentofInterior.
World Ports Distances. (2007). Retrieved November 2007, from World Ports Distances:
http://www.distances.com
WRI&WBCSD.(2004).TheGreenhouseGasProtocolACorporateAccountingandReport
ing Standard (revised edition). World Resources Institute Washington, D.C. and World Business
CouncilforSustainableDevelopment,Geneva.
FinalReport:CanLCA
PEAmericas
APPENDIX A: ESTIMATION OF TOTAL PRIMARY ENERGY DEMAND FROM DIRECT ENERGY USE
IntheU.S.,duringtheproductionofaluminum(liquid)metalduringtheelectrolysisprocess,
theconsumptionofdirectelectricityismainlyfromhydropower(approximately70%).However,due
tosignificantlylowerenergyconversionefficiencyfromcoal,theenvironmentalimpactsassociated
withthegenerationofendelectricityfromcoal(efficiencyofabout30%)outweightherelativelylow
environmentalimpactsofhydropower(efficiencyofabout80%).Becauseofthis,powerproduction
fromcoaldominatestheenvironmentalimpactsofthetotalelectricityproductionperkgaluminum
(liquid).
Table29:Totalprimaryenergydemandfor1kWhgenerationofelectricityfromdifferentenergy
sources in the U.S. Data source: GaBi 4 software system database (current release GaBi 4.3,
http://www.gabisoftware.com)
Unit
Power
from Power from Power from Power
nuclear power hardcoal
hydropower
from
plant
plant
natural
gas
Primaryenergy
MJ/kWh
14.323
11.298
4.524
11.136
Nonrenewableenergy
resources
MJ/kWh
14.301
11.295
0.022
11.092
MJ/kWh
0.022
0.003
4.502
0.044
IntheU.S.,toproduce1kWhfromhardcoal,itrequiresapproximately11.3MJofprimary
energy,whereastoproduce1kWhfromhydropower,itrequiresonly4.5MJofprimaryenergy.This
difference is on account of variations in the overall efficiency of electricity production as stated
above.ThisisalsographicallyillustratedinFigure34.
Primaryenergydemandisthequantityofenergydirectlywithdrawnfromthehydrosphere,atmosphere,
geosphere,orotherenergysourcewithoutanyanthropogenicchanges
Chapter: Appendix A: Estimation of Total Primary Energy Demand from Direct Energy Use
The total primary energy demand7 required to produce 1 kWh of end electricity (3.6 MJ)
fromeachenergysourceintheU.S.andCanadaisgiveninTable29andTable30,respectively.

16
Total primary energy demand (MJ per kWh)
Renewable
Non-renewable
14
12
10
0
Power from nuclear power
plant
Power from hard coal
Power from hydropower
Power from natural gas
Figure34:Totalprimaryenergydemandfor1kWhgenerationofelectricityfromenergysourcesin
theU.S.(Source:GaBi4.3LCIdatabase)
Table30:Totalprimaryenergydemandfor1kWhgenerationofelectricityfromdifferentenergy
sources in Canada. Data source: GaBi 4 software system database (current release GaBi 4.3,
http://www.gabisoftware.com)
Unit
Power
nuclear
plant
MJ/kWh
14.323
12.179
4.500
10.592
Nonrenewableenergyresources MJ/kWh
14.301
12.162
0.021
10.382
0.022
0.017
4.479
0.210
Primaryenergy
MJ/kWh
from Power from Power

Power
power hardcoal
from hy from natu
droelectric ralgas
plant
Duetothefactthatelectricitygenerationfromfossilfuelsislessefficientthanhydroelectric
power generation, sourcing a unit of energy from fossil fuelbased generation sources creates a
much higher primary energy demand than sourcing the same amount of energy from renewable
resources.Asaresult,thenonrenewablefractionoftotalprimaryenergydemandishigherthanthe
renewablefraction.
In our study, the direct electricity use of 55.16 MJ per kg aluminum (liquid) metal corre
sponds to a total primary energy demand of 102.7 MJ (nonrenewable fraction 55.1 MJ and re
newablefraction47.6MJ).Althoughhydropowerconstitutesapproximately69.4%ofdirectelec
FinalReport:CanLCA
PEAmericas
tricityuseforelectrolysisintheU.S.andCanada,therenewablefractionisonly46.35%ofthetotal
primary energy demand. The estimation of total primary energy demand can be expressed by the
followinggenericformula:
Hydro69.4%
Coal29.7%
Naturalgas0.6%
Nuclear0.3%
Theprimaryenergydemandforelectrolysis(fromelectricityuseonly)intheU.S.andCan
adaarecalculatedasbelow:
U.S.
TheaboveresultsarealsographicallyillustratedinFigure35.
IntheU.S.andCanada,asprovidedbytheIAIdata,itrequiresapproximately55.16MJofdi
rectelectricityusedtoproduce1kgofaluminum(liquid)metal.Thebreakdownforelectricitygen
erationis31.1MJ(56.38%)intheU.S.and24.06MJ(43.62%)inCanada.Therelativefractionsfor
energysourcemixforelectricityproductionforelectrolysisintheU.S.andCanadaaregivenbelow
(assumedsameforbothcountriesasperIAIdata):

60
Energy (MJ per kg aluminum (liquid) metal)
Non renewable
Renewable
50
30.1
40
30
9.5
Coal +
natural gas
+ nuclear
20
21.6
10
Hydro
27.0
0
Direct energy (MJ)
Primary energy demand (MJ)
Figure35:Directenergy(inMJ)andPrimaryenergydemand(inMJ;forelectricityuseonly)for1
kgofaluminum(liquid)intheU.S.
Canada
Total
FinalReport:CanLCA
PEAmericas
APPENDIX B: TRANSPORTATION DATA FOR RAW MATERIALS

AND FINISHED PRODUCTS
Table31:Transportationdataforrawmaterialsandfinishedproducts
Rawmaterial/product
Tonnage
fraction Mode of transpor Distance
transported(%)
tation
(km)
Primaryingottosecondaryfurnaces
100%
Rail
2400
CanBodyStocktoCanManufacturer
100%
Rail
100
Transport of UBC scrap from collec 100%

tion point to secondary aluminum
producers
Truck
160
Transport of manufacturing scrap to 100%

secondaryaluminumproducer
Truck
160
FinalReport:CanLCA
PEAmericas
APPENDIX C: IMPACT INDICATORS

Thefollowingdescribesbrieflythevariousimpactcategoriesincludedinthereport8.
PRIMARY ENERGY CONSUMPTION

Primaryenergydemandisthequantityofenergydirectlywithdrawnfromthehydrosphere,
atmosphere,geosphere,orenergysourcewithoutanyanthropogenicchanges.
Itisameasureoftheresourceusedacrossthelifecycleofaproduct.
Forfossilfuelsanduranium,thiswouldbetheamountofresourcewithdrawnexpressedin
its energy equivalent (i.e. the energy content of the raw material). For renewable resources, the
energycharacterizedamountofbiomassconsumedwouldbedescribed.Forhydropower,itwould
be based on the amount of energy that is gained from the change in the potential energy of the
water (i.e. from the height difference). As aggregated values, the following primary energies are
designated:
ThetotalPrimaryenergyconsumptionnonrenewable,giveninMJ,essentiallycharacter
izesthegainfromtheenergysourcesnaturalgas,crudeoil,lignite,coalanduranium.Naturalgas
andcrudeoilwillbeusedbothforenergyproductionandasmaterialconstituents,e.g.inplastics.
Coalwillprimarilybeusedforenergyproduction.Uraniumwillonlybeusedforelectricityproduc
tioninnuclearpowerstations.
The total Primary energy consumption renewable, given in MJ, is generally accounted
separatelyandcompriseshydropower,windpower,solarenergy,andbiomass.Itisimportantthat
theendenergy(e.g.1kWhofelectricity)andtheprimaryenergyusedarenotmiscalculatedwith
each other; otherwise the efficiency for production or supply of the end energy will not be ac
countedfor.
PleaserefertoCML(2001)andGaBidocumentationathttp://documentation.gabisoftware.comfor
moredetailsontheimpactcategories.
Chapter: Appendix C: Impact Indicators
The energy content of the manufactured products will be considered as feedstock energy
content.Itwillbecharacterizedbythenetcalorificvalueoftheproduct.Itrepresentsthestillusable
energycontent.
GLOBAL WARMING POTENTIAL

Themechanismofthegreenhouseeffectcanbeobservedonasmallscalein,asthename
suggests,greenhouses.Theseeffectsarealsooccurringonaglobalscale.Incidentshortwaveradia
tionfromthesuncomesintocontactwiththeearthssurfaceandispartlyabsorbed(leadingtodi
rectwarming)andpartlyreflectedasinfraredradiation.Thereflectedpartisabsorbedbysocalled
greenhousegasesinthetroposphereandisreradiatedinalldirections,includingbacktoearth.This
resultsinawarmingeffectattheearthssurface.
In addition to the natural mechanism, the greenhouse effect is enhanced by human activities.
Greenhousegasesthatareconsideredtobecaused,orincreased,anthropogenicallyare,forexam
ple,carbondioxide,methane,andCFCs.
Page 25The alongside figure
shows the main processes of the an
thropogenic greenhouse effect. An
analysis of the greenhouse effect
should consider possible long term
global effects. The global warming
potential is calculated in carbon diox
ide equivalents (CO2Eq.). This means
that the greenhouse potential of an
emission is given in relation to CO2.
Sincetheresidencetimeofthegasesin
the atmosphere is incorporated into
the calculation, a time range for the
assessment must also be specified. A
periodof100yearsiscustomary.
FinalReport:CanLCA
PEAmericas
Figure36:
Greenhouseeffect
ACIDIFICATION POTENTIAL
Theacidificationofsoilsandwateroccurspredominantlythroughthetransformationofair
pollutantsintoacids.ThisleadstoadecreaseinthepHvalueofrainwaterandfogfrom5.6to4or
lower.Sulfurdioxideandnitrogenoxideandtheirrespectiveacids(H2SO4andHNO3)producerele
vant contributions to this acidification. This damages ecosystems, whereby forest dieback is the
mostwellknownimpact.
Acidificationhasbothdirectandindirectdamagingeffects(suchasnutrientsbeingwashed
outofsoilsoranincreasedsolubilityofmetalsintosoils).Butevenbuildingsandbuildingmaterials
canbedamaged.Examplesincludemetalsandnaturalstones,whicharecorrodedordisintegrated
atanincreasedrate.
When analyzing acidification, it should be considered that although it is a global problem,

the regional effects of acidification can vary. Figure 37 displays the primary impact pathways of
acidification. The acidification potential is
given in sulfur dioxide equivalents (SO2
Eq.). The acidification potential is de
scribedastheabilityofcertainsubstances
tobuildandreleaseH+ions.Certainemis
sions can also be considered to have an
acidification potential, if the given S, N
and halogen atoms are set in proportion
to the molecular mass of the emission.
Thereferencesubstanceissulfurdioxide.
Figure37:
AcidificationPotential
EUTROPHICATION POTENTIAL
Eutrophication is the enrichment of nutrients in a certain place. Eutrophication can be
aquatic or terrestrial. Air pollutants, wastewater, and fertilization in agriculture all contribute to
eutrophication.
The result in water is an accelerated algae growth, which in turn, prevents sunlight from
reachingthelowerdepths.Thisleadstoadecreaseinphotosynthesisandlessoxygenproduction.In
addition, oxygen is needed for the decomposition of dead algae. Both effects cause a decreased
oxygenconcentrationinthewater,whichcaneventuallyleadtofishdyingandtoanaerobicdecom
position(decompositionwithoutthepresenceofoxygen).Hydrogensulfideandmethanearethere
byproduced.Thiscanlead,amongothers,tothedestructionoftheecosystem.
Ineutrophicatedsoils,anincreasedsusceptibilityofplantstodiseasesandpestsisoftenob
served,asisadegradationofplantstability.Ifthenutrificationlevelexceedstheamountsofnitro
gen necessary for a maximum harvest, it can lead to an enrichment of nitrate. This can cause, by
meansofleaching,increasednitratecontentingroundwater.Nitratealsoendsupindrinkingwater.
Nitrate at low levels is harmless from a toxicological point of view. However, nitrite, a reaction
product of nitrate, is toxic to humans. The
causes of eutrophication are displayed in
Figure38.
The eutrophication potential is calcu

lated in phosphate equivalents (PO4Eq). As
with acidification potential, its important to
remember that the effects of eutrophication
potentialdifferregionally.
FinalReport:CanLCA
PEAmericas
Figure38:
EutrophicationPotential
PHOTOCHEMICAL OZONE CREATION POTENTIAL (SMOG)

Despiteplayingaprotectiveroleinthestratosphere,atgroundlevelozoneisclassifiedasa
damaging trace gas. Photochemical ozone production in the troposphere, also known as summer
smog, is suspected to damage vegetation and material. High concentrations of ozone are toxic to
humans.
Radiationfromthesunandthepresenceofnitrogenoxidesandhydrocarbonscreatecom
plex chemical reactions, producing aggressive reaction products, one of which is ozone. Nitrogen
oxidesalonedonotcausehighozoneconcentrationlevels.
Hydrocarbonemissionsoccurfromincompletecombustion,inconjunctionwithpetroleum
processing(storage,turnover,refuelingetc.),orfromsolvents. Highconcentrationsofozonearise
whenthetemperatureishigh,humidityislow,whenairisrelativelystaticandwhentherearehigh
concentrations of hydrocarbons. Because CO (mostly emitted from vehicles) reduces the accumu
latedozonetoCO2andO2,highconcentrationsofozonedonotoftenoccurnearhydrocarbonemis
sionsources.Higherozoneconcentrationsmorecommonlyariseinareasofcleanair,suchasfor
ests,wherethereislessCO.
InLifeCycle Assessments,photochemicalozonecreationpotential(POCP)isreferredtoin
ethyleneequivalents(C2H4Eq.).Whenanalyzingresults,itisimportanttorememberthattheactual
ozoneconcentrationisstronglyinfluencedbyweatherandbypollutioncharacteristicsofthelocal
conditions.
PhotochemicalOzoneCreationPotential
OZONE DEPLETION POTENTIAL

Most ozone resides in the upper part of the atmosphere. This region, called the strato
sphere,ismorethan10kilometers(6miles)aboveEarthssurface.There,about90%ofatmospheric
ozoneiscontainedintheozonelayer,whichshieldsusfromharmfulultravioletradiationfromthe
Sun.However,itwasdiscoveredinthemid1970sthatsomehumanproducedchemicalscouldde
stroy ozone and deplete the ozone layer. The resulting increase in ultraviolet radiation at Earths
surfacemayincreasetheincidencesofskincancerandeyecataracts.
Humanactivitiescausetheemissionofhalogensourcegasesthatcontainchlorineandbro
mineatoms.Theseemissionsintotheatmosphereultimatelyleadtostratosphericozonedepletion.
Thesourcegasesthatcontainonlycarbon,chlorine,andfluorinearecalledchlorofluorocarbons,
usually abbreviated as CFCs. CFCs, along with carbon tetrachloride (CCl4) and methyl chloroform
(CH3CCl3),historicallyhavebeenthemostimportantchlorinecontaininggasesthatareemittedby
human activities and destroy stratospheric ozone. These and other chlorinecontaining gases have
Figure39:

beenusedinmanyapplications,includingrefrigeration,airconditioning,foamblowing,aerosolpro
pellants,andcleaningofmetalsandelectroniccomponents.
In Life Cycle Assessments, Ozone Depletion Potential (ODP) is referred to in CFC11
equivalents.SincemostozonedepletingcompoundswerephasedoutbytheMontrealProtocolin
1987, the stratospheric ozone layer has been recovering and there are very few ozone depleting
emissions.
UV - radiation
Stratosphere
15 - 50 km
Absorption
Absorption
CFCs
Nitrogen oxide
Figure40:OzoneDepletionPotential
FinalReport:CanLCA
PEAmericas
APPENDIX D: DATA QUALITY EVALUATION

DataqualitywasevaluatedusingtheWeidemamethodologyasdescribedintheInterna
tionalJournalofLCA3(5)page259265;1998,Weidemaetal.;LCAdataquality.Thefollowingta
blesshowtheevaluationmatrixandtheevaluation.
Table32:Dataqualityevaluationmatrix
Chapter: Appendix D: Data quality evaluation
Table33:Resultsdataqualityevaluation
APPENDIX E: CRITICAL REVIEW REPORT
FinalReport:CanLCA
PEAmericas
5.1 ALUMINUM BEVERAGE CAN LCI REPORT FEBRUARY 2010

REVIEWERCOMMENTSFEBRUARY26,2010
Reviewerscomments
Followup
Regardingdataqualityanditsdiscussioninthedocument,
thatstillcouldposesomequestionswhenthereportis
released.Theresponsethatdataqualityisnowmoreuni
formlytreated,thatis,eachprocessoractivitynowhasa
statementregardingthequalityofthedataused,iscor
rect.However,theactualbasisforthosestatementsand
anyimplicationsoftheresultantuseofthedatawiththat
qualityratingisnotprovided.Isitusefultosaythatthe
qualityishighormedium,yes,becauseitgivesageneral
impressionofthedegreeofadherencetothefourquality
categoryparameters?DoIunderstandhowthatrating
wasarrivedatorhow/ifitaffectedtheuseoftheinforma
tionorthedrawnconclusions,no.
AppendixDonDataquality
evaluationadded.
Oneoftheoriginalcommentsinthe4Decemberdocu
ment:Evenfortheseoperationsitisn'tpossibletodiscern
howthedataqualityindicatorswereappliedtoreachthe
conclusionsregardingaggregatedquality."Theresponse
merelysays"Updatedinseveralplaces."However,what
doesitmeantosaythatdatawerefoundtobeofhigh
quality?Howdataqualitywasdeterminedneedstobe
explained
Point1.
Page13mentionsthecriticalreviewbutwherecanreaders AppendixEadded
findthereviewcomments?Aswediscussedtheotherday,
animportantpartofacriticalreviewforpublicassertionis
fulldisclosureofwhatthereviewfoundandhowissues
wereresolved.Thereviewprocessandcommentsshould
beincludedasanappendix.
Specificcomments:
Page3;Shouldthisread"theAluminumAssocia
tionexperts.."notexpects?
WhydoesFigure1haveablueboxaroundthe
canproductionbutFigure2doesnot?
Figure9:Aretheyears(19992005(b))neededin
theAluminabox?Thecaptionsaysthedatareflect2005.[I
realizethefiguresareprobablygeneratedfromGaBiand
maynotberevisable.]
Page74:Onlyselecteddata,andimpacts,are
Formatting/typographical
mistakesfixed.
Chapter: Appendix E: Critical review report
No

showninthischapter.WherecantheentireLCIbefound?
Page97:Page25Thealongsidefigure...??
5.2 ALUMINUM BEVERAGE CAN LCI REPORT OCTOBER 2009

REVIEWERCOMMENTSDECEMBER4,2009
No
Reviewerscomments
Followup
Whileourcommentshavebeenindividuallyaddressed,Ifeel Thereisasectionondatacal
culationwherethemethods
thatsomeoftheresponsesarecursory(someareevenar
foraveragingareexplained.
gumentativewithoutarealbasis).Forexample,PEAsre
sponsetobeingaskedforthemethodthatwillbeusedfor
averagingdata,thereplyis"ThefinalLCIprofilewillrepre
sentindustryaverageofvariousunitprocesses."Thisreply
doesnotprovideadescriptionofthemethod.
Ihavetheimpressionthattheauthorshavediligentlycon
sideredourcomments,evenifthewrittenresponsesare
thin.Iseenoproblemwithmovingaheadwiththenextstep
(conferencecall?)inthereviewprocess
Inthereportanumberofstatementsaremadeinqualitative
ornonnumericalfashionthatdoesnotallowthereaderto
judgethevalidityofthepreparersdecisions.Forexample,
pagexoftheExecutiveSummarycontainsthestatement
Modelingfordomesticproductionwasusedtoreflectthe
factthatthemajorityofsecondaryaluminumconsumedin
thetargetyearwasproducedintheUS.Withoutanumeri
calvalue,itisnotpossibletounderstandifthiswasacorrect
determinationornot.Ifthepercentageis85,thenitisrea
sonable;ifitis51,itmaynotbe.Section2.2.8.2onpage12
isevenlessdescriptiveofwhythegeographiccoveragesare
valid.
OK.Commentaboutdomestic
productionwasremovedasit
wasmisleading.Caningotis
producedinrecyclingfacilities
intheUSusingdomestic
scrap.
Therearesomenumericaldiscrepanciesbetweenthereport
andtheaddendumthatarenotexplained.Forexample,in
thethirdbulletpointonpagexoftheExecutiveSummary
(andelsewhereinthereport)itstatestherecycledcontent
percentageis67.8%,whileonpage5oftheaddendumitis
60.1%forthesamedatayearof2007.
Thisismentionedinthead
dendum.Itisadifferentway
ofcountingrecycledcontent
andtheresultsoftheadden
dumreflectthis.
Therearesomeassumptionsorconditionsinherentinthe
systemaccountingthatmaynotbeunderstandabletoaudi
Additionalcommentaryadded
FinalReport:CanLCA
PEAmericas
encememberswhoarenotthoroughlyfamiliarwithLCIac
countingrules.Forexample,astatementinoneoftheEoL
scenariosthatimportsscrapburdenfreeisnotintuitive.
Specifically,thecommentAnadditional1.797kgofscrapis
thereforerequiredtomakeupthedeficitwhichintherecy
cledcontentapproachisimportedburdenfreeshould
haveadditionalcommentaryastowhythisisreasonable.
Section4.4.2,2ndpar.providesabetterexplanationand
shouldbeconsideredforinsertionintheExecutiveSum
mary.
inseveralplaces.
Inadditiontosomegrammaticalandtypographicissues,
Updatedinseveralplaces.
which,inaminorway,detractfromthereportscredibility,
therearesomeplacewhereeditingwouldimprovetheun
derstandingofthestudycontextandresults.Forexample,
onpage2itstatesTheaimofthestudyistogeneratehigh
quality,uptodatedataontheenvironmentalperformance
ofaluminumbeveragecanproduction.Itissuggestedthat
slightlyamendingthistoaddincludingtheflowofsecon
darymaterialsfromEndofLifebackintobeveragecans.,
wouldmakeitclearthatcanproductionisthefocus,but
sourcing/processingofreusablealuminumisessentialtothis
characterization.
Therationalefortherecycledcontentapproach(page5)
needstobeclearthatinadditiontotheassumptionsstated,
thebenefitsofrecycling(primarymaterialproductionand
wastedisposalavoidance)outweightheburdensofpost
consumercollectionandprocessing.Ascurrentlydescribed
thisnetbenefitrequirementisnotclear.
WhilethisistrueforAlumi
num,itisnotalwaysthecase
withothermaterials.Com
mentadded.
InSection2.2.9,pleasecorrectmyaffiliationtoreflectthe
transitionfromBattelle(initialstudyreview)toSETAC(final
reportandaddendum).
OK;noreferenceinaddendum
InSection3.1.5,itindicatesthatmaterialsfromingotcasting
(e.g.drossandfilterdust)recoveredforuseinsystemsother
thanbeveragecansarebeingtreatedaswastes(i.e.haveno
coproductburdens),eventhoughthetextimpliestheseare
coproducts.Althoughthisdistinctionmaybeinconsequen
tialintermsofflowsorimpacts,itseemsinconsistentwith
thestatementmadeearlierregardingallocation.InSection
2.2.6itstatedInthisstudy,noallocationwasappliedasall
coproductsareconsideredineachoftherecyclingmodels
describedabove(ISO,2006b).Someclarificationtotheef
fectthatanyrecoveredmaterialscycledintootherproducts
aretreatedaswaste,therebyprovidingamaximumalloca
tionofburdenstoaluminumcanproduction,wouldresolve
thisapparentinconsistency.
OK.addedinsection2.2.6
Chapter: Appendix E: Critical review report
10
Thedataqualitydiscussionthroughoutthereportisinconsis
tentandinmostcaseslackingaltogether.Actualdiscussion
ofdataquality,regardlessofspecificity,isprovidedforonly
threesetsofprocesses,primaryAl,cansheet,andcanpro
duction.Evenfortheseoperationsitisnotpossibletodis
cernhowthedataqualityindicatorswereappliedtoreach
thestatedconclusionsregardingaggregatedquality.
Updatedinseveralplaces.
11
InSection4.4.1thefirstparagraphhasaformattingerror
flaggedthathasnotbeenfixed
Shouldbefinenow
12
Ingeneral,theaddendumcommentaryregardingthetwo
EoLoptionsisunderstandableandresponsivetotheearlier
inputfromthereviewpanel.However,thereareanumber
ofplaceswheretheconnectiontothefullstudyreportneeds
tobemoreexplicit.Forexample,nowhereintheaddendum
doesitstatethattheanalysisisfor12ouncecans.Ifitis
possiblefortheadditionalanalysessomehowtobesepa
ratedfromthereport,theaddendumshouldbemoreorless
standalone
Theaddendumisnotfordis
closureandtherefore,not
meanttobeastandalone
document
13
Inasimilarvein,referencestoorsourcesofdataneedtobe
complete.Intheaddendum,eithertheyaremissingalto
gether,forexamplethe2007recycledcontentpercentage,
orarecitedin(source,date)formatwiththeactualcitation
notincluded.Thisisnottransparent
Point12.
14
Thesignificanceorlackthereofforthedifferentresultsbe
tweentheaddendumandthereportshouldbeexplained.
Thetextsimplystatestheyareminorandthefinalpara
graphstatesthereisnosignificancetotheoverallLCAre
sults.Thatmuchisobviousfromthenumericalvalues,but
itwouldbeusefultounderstandwhythedifferencesare
considerednonsignificant.Ifthissimplyduetothesmall
numericalchanges,thedatavariability,orthedifferencein
impactconsequencesthemselves,thereportshouldsostate.
Thisisstatedinthefirstpara
graphitisanotherwayof
countingrecycledcontent.The
reasonforthedifferencesare
explained.
15
Thecommentmadeearlierbythereviewteam,andinprin
cipleagreeduponbytheconsultant,toeitherprovidevol
umebasedresultsoraconversionfactorappearsnotto
havebeenimplementedintheaddendum.Althoughthe
textinSection4.3ofthereportdoesprovideanum
ber:volumeconversionfactor,itwasnotclearwhetherthe
1000canbasisconversiontovolumewasfor12oz.cansora
weightedaverageofallsizes
Theaddendumusesthesame
parametersasthecaseofthe
report.Insection2.1itis
statedthattheLCIdatarepre
senttheweightedaverageof
thedifferentcansizes.
16
Iftheseresults,orthoseoftheoriginalreportforthatmat
ter,aregoingtobereleasedtothirdparties,somedetailson
CMLandGaBiReferences
FinalReport:CanLCA
PEAmericas
theaggregationofinventoryflowsintoLCIAcategoriesand
theconversiontoimpactsneedstobeincluded,evenifthe
actualcomputationsaredoneentirelywithinthesoftware.
Otherwise,thereisalossoftransparencyandconsequen
tiallycredibility.TheAppendixCmaterialdiscussesthisin
generalbutdoesnotprovideamappingofflowstoimpacts,
particularlywherethereisapossibilityofinconsistenciesin
assigninginventoryflowstoimpactcategories
added.
17
Thisistheapproachadopted
Schematiconpage4ofAddendum:Showscanmanufactur
inthemainreport.
ingscrapremeltedwithrollingprocessscrap.Therealityis
thatcanmanufacturingscrapisusuallycoated,decorated
(inks)orcontainscompoundforsealingthecan(lids),there
foreitissentthroughthesecondarysystem,muchlikeUBCs.
Thearrowshouldbeextendedhorizontallytoconnectwith
thatrecycleloop.Idontthinkthisaffectstheanalysis,butit
representswhatactuallyoccurs.
18
Assumptiononpage5ofAddendum:Thecommentaboutno Thisisaccountedforinthe
materialflowsandhasbeen
primaryaluminumrequiredforproductionofRSIiscorrect.
confirmedbyAA.
However,RSImadefromcanlidsandbodiesisgenerally
usedtomakenewcanbodystock.TheRSImadefromthis
mixcontainsmoremagnesiumthanisallowedin3004or
3104alloyspecifications,thereforesomeamountofprimary
(orotherformofpurerscrap)isusedtodilutethemagne
siumpercentage
19
Thereportseemstoexcludethefactsthat
1. UBCsareusedinotheraluminumsheetendprod
ucts,suchasbuildingproductssheet
2. Otherformsofscrapsuchasbuildingproductssheet
areusedtomakeRSIthatisusedintherecycleloop
forcanbodystock.
Basedonespecially(b)above,mybeliefistherecy
cledcontentofcans(fromallformsofscrap)ishigherthan
statedinthereport
APPENDIX F: CRITICAL REVIEWERS BIO
Scrapistreatedasscrapinthe
modelirrespectiveofwhether
itiscanscraporindustrial
scrap(asitdoesnothavean
influenceontheLCA).The
percentageofsecondaryalu
minum(fromAAandproduc
ers)includesallkindsofscrap.
Chapter: Appendix F: Critical reviewers Bio

Dr. Bruce W. Vigon
Dr. Vigon has designed and executed research and development programs ranging from
shorttermfocusedstudiestomultimilliondollar,multiyeareffortsforarangeofcorporateclients
andgovernmentagencies,bothdomesticallyandinternationally.Effortshaveinvolvedorganizations
asdiverseastheUSEnvironmentalProtectionAgency,EnvironmentCanada,USDepartmentofDe
fense,stateandprovincialgovernments,variousacademicinstitutions,theWorldBusinessCouncil
forSustainableDevelopment,WorldWildlifeFundandotherenvironmentaladvocacyorganizations,
andahostofprivatesectorcompaniesrangingacrosstheautomotive,textiles,consumerproducts,
electronics,defense,andpackagingindustries.BruceisamemberofSETACandisactiveinseveral
focusareas,includinglifecycleassessment,environmentalsystemsanalysis,environmentalchemis
try,contaminantfateandeffectsresearch,andriskassessment,since1990.HehasbeenEditorof
theSETACGlobeLCACommunityofPracticenewsletterformorethan18yearsandhasbeenchair
andcriticalreviewpanelistonmorethanahalfdozenLCAstudies,includingseveralhighlycomplex
andcontroversialproductandservicesystems.
Dr.MaryAnnCurran
Dr.CurrandirectstheUSEPASystemAnalysisBranchsLifeCycleAssessment(LCA)research
program which includes the development of LCA methodology, the performance of lifecycle case
studies, lifecycle workshops and conferences, and the development of a life cycle data and re
sourceswebsite(www.epa.gov/ORD/NRMRL/lcaccess).AsarecognizedinternationalexpertinLCA,
Dr.CurranworkscloselywiththeSocietyofEnvironmentalToxicologyandChemistry(SETAC),which
has been instrumental in advancing LCA awareness worldwide, and actively participates in the
UNEP/SETACLifeCycleInitiative.Dr.CurranalsoservesontheSETACNorthAmericaLCAAdvisory
Group. In addition, Dr. Curran is on the editorial boards of the International Journal of Life Cycle
Assessment,ManagementofEnvironmentalQuality,andtheadvisoryboardsoftheonlinejournals
Sustainability and the International Journal of Environmental Research and Public Health. Since
2005, Dr. Curran has served as the Subject Editor (Cleaner Production Tools) for the Journal of
CleanerProduction.SheprovidestechnicalsupporttoEPAprogramofficesindevelopingpolicyand
regulations including guidelines for the federal procurement of environmentallypreferable prod
ucts.Dr.Curranprovidestechnicalreviewandassistancetooutsidegroupsontheapplicationofthe
life cycle concept to areas such as green product and process design and development. She has
participated in the technical peer review of industrysponsored lifecycle studies on various prod
ucts,suchaselectricity,diapers,cleaners,plastics,coalash,buildingmaterials,andpackaging.
ToddBoggess
Todd is a 1998 graduate of the University of Kentucky with a BS in Economics. Todd is a

member of Alpha Kappa Psi, a professional business fraternity. Following graduation, Todd spent
twoyearsasaconsultantwithSystems&ComputerTechnology,whereheprovidedfinancialinput
into the design and update of accounting software for nonprofit organizations. In 2001, Todd ac
ceptedapositionwiththeUniversityofKentuckyintheSponsoredProjectsAccountingdepartment.
There,hemanagedfederalandstatefundedgrantsfortheUniversityofKentuckyResearchFounda
FinalReport:CanLCA
PEAmericas
tion.InDecemberof2001,ToddjoinedSecat,Inc.asBusinessManagerwhereheoverseesallfinan
cialandaccountingresponsibilities.ToddactsasprojectmanagerforUSDepartmentofEnergycon
tractsgrantedtoSecat,Inc.ToddalsoservesasSecretaryfortheBoardofDirectorsforSecat.Secat
providestechnologyandR&Dservicestomaterialscompanies.ToddiscurrentlyenrolledintheMBA
programatWesternKentuckyUniversityandexpectstograduateinJune2011.
Chapter: Appendix F: Critical reviewers Bio

LCA 2010 AluminumAssoc PDF

Caricato da

Informazioni sul documento

Titolo originale

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

LCA 2010 AluminumAssoc PDF

Caricato da

Copyright:

Formati disponibili

May21,2010

3.1.1 BauxiteMining ................................................................................................................. 14

3.3.1 AcidificationPotential ...................................................................................................... 44

Chapter: Table of Contents

4.1.1 CanSheet .......................................................................................................................... 51

4.2.1 CanSheet .......................................................................................................................... 69

4.3.1 BeverageCanSheet ..........................................................................................................73

REVIEWERCOMMENTSDECEMBER4,2009 ............................................................................................. 104

Chapter: Table of Contents

Chapter: Table of Figures

Chapter: List of Tables

Chapter: Executive Summary

Alumina Electrolysis IngotCasting

Primary Energy Demand for fuel pro 1.02GJ

Energy from nonfossil fuel sources, 0.02GJ

Chapter: Executive Summary

Photochemical Ozone creation potential 0.051

Chapter: Executive Summary

The environmental profile of primary aluminum production has a significant influence

Chapter: Executive Summary

2 GOAL AND SCOPE OF THE STUDY

Product System(s) Boundaries

Further processing materials

Processing of raw materials

Transportation of raw and

Chapter: Goal and Scope of the Study

2.2.1.2 Recycled content approach

Chapter: Goal and Scope of the Study

Data Collection, Software and Databases

Chapter: Goal and Scope of the Study

Adopts a product life cycle and material Adoptsawastemanagementperspective

Primary data collection of information, which is representative for specific production op

Cut- off Criteria

Chapter: Goal and Scope of the Study

Life Cycle Impact Assessment Methodology & Impact Categories Considered

Data Quality Requirements

2.2.8.2 Geographical Coverage

2.2.8.3 Technological Coverage

Chapter: Goal and Scope of the Study

3 PRIMARY ALUMINUM PRODUCTION

3.1 PROCESS DESCRIPTION

Chapter: Primary Aluminum Production

IAI_NA: Bauxite mining 2005

Bauxiteoreistheprimaryrawmaterialsourcefor thealuminumproduction. Aluminum is

The domesticbauxite productionintheUSislessthan1%of thetotal demandofbauxite

Chapter: Primary Aluminum Production

Chapter: Primary Aluminum Production

IAI_NA: Bauxite mining (Bauxite imports)

Bauxite from Brazil

Bauxite from Guinea

Bauxite from Guyana

Bauxite from other

Figure 8: The transportation of Bauxite imported to the US is included in each countryspecific

Chapter: Primary Aluminum Production

Chapter: Primary Aluminum Production

IAI_NA: Alumina (Production + Imports)

Chapter: Primary Aluminum Production

An illustration of the anode production process implemented in the model is shown in

Table7:Inputs andoutputsfor theanodeproduction unitprocess.Flowsarerepresentativefor

Chapter: Primary Aluminum Production

Chapter: Primary Aluminum Production

Chapter: Primary Aluminum Production

Chapter: Primary Aluminum Production