二氧化碳捕获

of4-Diethylamino-2-butanol,Monoethanolamine,AbdulazizNaami,MohamedEdali,TeerawatSema,RaphaelIdem,*andPaitoonTontiwachwuthikul

InternationalTestCentreforCO2Capture,FacultyofEngineering,UniversityofRegina,Regina,Saskatchewan,Canada,S4S0A2ABSTRACT:ThemasstransferperformanceoftheabsorptionofCOabsorberpackedwithhighefficiency2inanaqueoussolutionofmonoethanolaminewasevaluatedexperimentallyinalab-scaleDXstructuredpackingandcomparedwiththatofmethyldiethanolamine(MDEA)aswellasthatofanewlydevelopedtertiaryaminoalcohol,4-diethylamino-2-butanol(DEAB).Theabsorptionexperimentswereconductedatatmosphericpressure,usingafeedgasmixturecontaining14.9%COanabsorptioncolumncontainingDXstructuredpacking.Theabsorptionperformancewaspresentedinterms2and85.1%nitrogeninoftheCO2removalefficiency,absorberheightrequirement,effectiveinterfacialareaformasstransfer,andoverallmass-transfercoefficient(KGav).Inparticular,theeffectsofparameterssuchasinertgasflowrateandliquidflowratewerecomparedforbothDEABandMDEA.TheresultsshowthattheDEABhasamuchhigherremovalefficiencyforCOthanthatforMDEA.Forallthesolvents,2alongtheheightofthecolumnthanMDEA.Also,theKGavofDEABwasmuchhighertheKtransfercoefficientfortheCOGavincreasedastheliquidflowratewasincreased.Anempiricalcorrelationforthemassafunctionoftheprocessparameters.Intermsofcomparison,theresultsshowthattheDEAB2-DEABsystemhasbeendevelopedassystemprovidedanexcellentoverallmasstransfercoefficient,whichishigherthanthatoftheMDEAsystembutlessthanthatofMEA.1.INTRODUCTION

appropriateplacementofthesubstituent,especiallyhydroxylCarbondioxideisagreenhousegasandsubstantiallyfunction,relativetothepositionoftheaminogroup,inordertocontributestoglobalwarmingandclimatechange.OneoptionpromoteCO2captureperformance.

forreducingCO2emissionsispostcombustioncapturefromManeeintretal.,3,4haveshownthatthenewaminoalcoholspowerplantfluegases.ManycountrieshaveagreedtoreducehavemuchhigherCO2absorptionandcycliccapacitiestheiremissionsofgreenhousegasesintotheatmospheretohelpcomparedtothatofconventionalamine,MEA.Detailsofthepreventdangerousalterationstotheclimatesystem.SeveralsolubilityandphysicalandtransportpropertiesofoneofthematuretechnologiesareavailableforCOnewsolvents,DEAB,canbefoundintheliterature.5−7Itiswell-cryogenics,membranetechnologies,2capture,includingadsorption,andabsorption.1knownthattheoverallperformanceofanysolventdependsnotAccordingtoAstarita,2absorptionisthemostcommonlyusedonlyonitsabsorptioncapacityandenergyefficiency,butalso,processwhenitcomestogastreating.Gasabsorptionbyonthemasstransfercharacteristicsandthekinetics.ThechemicalsolventssuchasaqueoussolutionsofalkanolaminesispresentworkfocusesonmasstransferandcomparestheoneofthemosteffectivemethodsforCOperformanceofthepotentialnewsolventwiththeconventionalusedinindustryforover2removal.Thistechnologyhasbeenhalfacentury.solventsPresently,themostcommonlyusedchemicalsolventsareflthatcanbeusedforthecaptureofCO2fromalkanolamines,andthesealkanolaminescanbeclassifiedinto(MEA))uegas.Consequently,comparedwithaaprimaryindustrialnewchemicalamine(monoethanolaminesolvent,anaminothreechemicalcategories:primary,secondary,andtertiaryalcohol(4-diethylamino-2-butanol(DEAB)),andatertiaryamines.Alkanolaminescommonlyusedaremonoethanolamineamine(methyldiethanolamine(MDEA))wereselectedto(MEA)andmethyldiethanolamine,(MDEA).MEAishighlycomprehensivelystudythenewsolvent’smasstransferreactivewithCO2buthasthelimitationoflowCOcharacteristicsinapacked-bedabsorptioncolumn.

capacity(anequilibriumabsorptionof0.5molCO2absorptionTheprocessofCOMEA)andhighheatofregeneration;ontheother2/molofhand,thecontact2absorptioninapackedtowerdependsmainlyonbetweenthefluegasandtheliquidMDEAhashighCO2absorptioncapacity,andlowheatofsolvent.Inthelast35years,traycolumnshavebeenreplacedinregeneration,butitislimitedbyslowkinetics.Themainlargepartbypackedcolumns.8,9Thishasbeenduetothechallengesuccessfuldevelopmentofefficientpackingthatprovidesahighflabsorptioncapacityperunitvolumeofpacking.Insidetheregeneration.uegasesisfortheCO2postcombustioncapturefrompowerplantThereductionrequiredofreductiontheenergycanrequirementonlybeachievedforsolventbyabsorptioncolumn,thepackingprovidesacontactarea,whichapplicationofnewsolvents.Recently,theInternationalTestseparatestheliquidflowintodroplets.Thisincreasesthe

CentreforCarbonDioxideCapture(ITC),Saskatchewan,Canada,hasdevelopednewchemicalsolventsbasedonaminoalcoholsforCOReceived:April18,20112capturetoovercomethedrawbacksoftheseRevised:March20,2012conventionalamines.ThesynthesisinvolvesasystematicAccepted:April16,2012modificationofthestructureofaminoalcoholsbyan

Published:

April16,2012

©2012AmericanChemicalSociety

70

dx.doi.org/10.1021/ie2008357|Ind.Eng.Chem.Res.2012,51,70−79

Industrial&EngineeringChemistryResearchArticleamountofliquidsurfaceareathatisexposedtothegasphase,Then,Kenhancingabsorptionrates.Goodpackingsshouldoffercharacter-Gcanbeexpressedasfollows:

isticssuchaslargesurfaceareatovolumeratio,lowpressuredropacrosstheabsorber,andgooduniformdistributionbetweenthegasK⎛G=⎜⎜NA

⎞andliquidphasesthroughthecolumn.Onthebasisofthese⎝[P(yA−y*⎟A)]⎟⎠(6)

characteristics,Aroonwilas9anddeMontigny10haveshownthatInagasabsorptionapparatussuchasapackedcolumn,itis

structuredpackingsofferasuperiorperformance11whencomparedmoreusefultorepresentratesofabsorptionintermsoftorandompackings.Fernandesetal.studiedstructuredpackingvolumetricoverallmasstransfercoefficients,representedbytheandfoundthatthehighsurfaceareaofthispackingincreasesthetermKmasstransferefficiencies.

Gathev,insteadoftheonebasedontheinterfacialareaunitbecausegas−liquidinterfacialareacannotbemeasuredTheobjectiveofthisworkwastoperformacomparativeaccurately.

studyofthemasstransferperformanceforCOTherefore,itismoreusefultopresentthemass-transferamine,MEA,MDEA,andthe2absorptioninconventionalnewchemicalcoefficientbasedontheunitvolumeoftheabsorptioncolumnsolventDEABusingamasstransferexperimentallaboratory-ratherthanbasedontheinterfacialareaunitasfollows:

scalecolumnwithDXstructuredpacking(27.5mm)ID(fromSulzerChemtechCanada,Inc.)forthethreesolventsundersimilarconditions.ThemasstransferperformanceofthethreeK⎛NGav=⎜⎜Aav

⎞solventsispresentedintermsoftheoverallmasstransfer⎝[P(yA−y*A

)]⎟⎟

⎠(7)

coefficient.

ThetermNAavcanbedeterminedfromtheabsorption2.DETERMINATIONOFTHEOVERALLMASS

experimentsinpackedcolumnswheretheconcentrationprofileTRANSFERCOEFFICIENT

oftheabsorbedcomponentAinthegasphasecanbemeasuredalongthecolumnheight,andthisallowedustoevaluatetheMasstransferoccurswhenacomponentAinagasphaseKGavvalue.

transfersacrossagas−liquidinterfaceintoaliquidphase.TheConsideringanelementoftheabsorptioncolumnwithcomponentAistransferredfromthegasphaseintotheliquidheightdZ,inFigure1,themassbalanceofcomponentAcanbebecauseoftheconcentrationgradientinthedirectionofmassgivenasfollows:

transferwithineachphase.ThemassfluxofcomponentA(Nacrossthegas−liquidinterfaceatsteadystatecanberepresentedA)intermsoftheN⎛yAavdZ=GId⎜⎜A

⎞−ygas-sidemass-transfercoefficient(kasfollows:

G)anddrivingforce(y⎝1−y⎟A⎟⎠(8)A,GA,i)NKA=KGP(yGavP(yA,G−yA,i)(1)

A,G−y*A

)dZ=GIdYA,G(9)

wherePrepresentsthetotalpressureofthesystemandywhereYA,GrepresentsthemoleratioofcomponentAintheofcomponentAingasbulk.A,Gand

ybulkgasandGA,irepresentthemolefractionTheIrepresentsthemolarflowrateoftotalgasgasmasstransfercoefficientkwithoutcomponentA.

GisdifficulttomeasurebecauseofthechangeininterfacialareawithvaryinggasflowratesintheThefinalequationthatdeterminestheoverallmasstransferpackedcolumns.Instead,theoverallmasstransfercoefficientforcoefficient,KGav,canthenbedefinedasfollows:

thegasphase(KHenryG)isoftenused,12anditcanbepresentedintermsofthe’slawconstant(H)asfollows:

K⎛=⎜G⎞⎛dY⎞

Gav⎜I

⎟⎝P(y⎜A,G⎟NA,G−yA)⎟

⎠⎝dZ⎠

A=KG(PyA,G−HCA,L)(2)

(10)

ThisapproachhasbeenStrigle13describedtherelationshipbetweentheoverallmass9usedsuccessfullybyAroonwilasand

Tontiwachwuthikul,deMontigny,10andManeeintr7fortransfercoefficientsandtheindividualmasstransfercoefficientsanalyzingpackedcolumns.Inthisstudy,theexperimentsofasfollows:

CO11Hstructured2absorptionwerecarriedoutintestcolumnspackedwithpackings.Thevariablespresentedineq10canbeK=+obtainedfromtheabsorptionexperiments.TheinertgasflowGkGkL

(3)

rate(GI)isanoperatingconditionoftheexperiment,whichisTheoverallmasstransfercoefficientsforchemicalabsorptioncanconductedatatmosphericpressure(P).TheCObeexpressedasafunctionofenhancementfactor(I):

thegasphase(yalong2concen-trationinA,G)canbemeasuredthelength1ofthepackedcolumn.Theyinequilibriumwiththebulk*AconcentrationistheconcentrationCofsoluteAK=1+HA,L.ThemeasuredGkGIkL0(4)

COTheconcentrationgradients(yY2concentrationcanbeconvertedintomoleratiovalues(A,G)andplottedagainsttheheightofthecolumntoobtaindistances,whichmakesA,G−itdiyA,ffii)takeplaceoverextremelysmallculttomeasurethethesolutemoleratioconcentrationgradient(dYconcentrationofcomponentAatthegas−liquidinterface.UnderFigure1,andtheadvantageofthistechniqueA,G/dZ),asshowninisthatitthissituation,itismorepracticaltoexpressthemassfluxinallowsthecalculationoftheKtermsoftheoverallmasstransfercoefficientandthemolealongthecolumn.WhenGavvalueatanyspecificmoleratiovaluescomparingtheKfractionofcomponentAinthegasphase(yfortwosolvents,itisimportanttomakethecomparisonGavvaluewithwiththeconcentrationofAinthebulkliquid*Aas)infollows:

equilibriumsimilarexperimentalconditions.Forexample,tocomparetheeffectoftheliquidflowrateonKNA=KGP(yexperiments,thesolutionconcentration,Gathevvaluesfortwomoleratio,andA,G−y*A

)(5)

CO2loadingconditioninthepackedcolumnshouldbethe

71

dx.doi.org/10.1021/ie2008357|Ind.Eng.Chem.Res.2012,51,70−79

Industrial&EngineeringChemistryResearchArticleFigure1.MasstransferprocessExperimentalschematicdiagram.

same,oratleastinthesamerange,toensurethattheGcalculationofKI=inertgasflowrateandyGavvaluesarebeingcomparedfairly.ForistheerrorA=gasphaseCOandmass.error%intheexperiment.

2concentration,example,iftwoexperimentshadthesamemoleratiovalue,butIntheliterature,9,10itisassumedthatthetermofyoneexperimenthadasolutionCO2loadingofα=0.15moltozerobecausethecompoundAisquicklyconsumed*AisinequaltheCO2/molamineandthesecondavalueofα=0.36molCOfastchemicalreactionandisdifficulttomeasure,butinthismolamine,thecomparisonwouldnotbefair,sincethesecond2/workwemeasurethetermofy*A.

experimenthadareducedcapacityforabsorptionduetoitshighloadingandrelevantlowfree-amineconcentration.To3.THERMODYNAMICSFORCO2ABSORPTION

determinetheloadingataspecificheightofthecolumnandKnowingthethermodynamicpropertiesofaminesolution,CO2concentration,weusedeq11:

suchasthephysicalsolubilityofCOCO2ortheCO2Henry’sconstant,physicaldiffusivityofα=αlean.loading+((((D1−D2)×A×time)CO2,equilibriumconstantsfor2-aminesystem,viscosity,anddensity,isrequiredinunderstandingCO/(M×L))×mass.error%)(11)

Solubility2absorption.

3.1.andPhysicalproperties.Inthisstudy,thesolubilityofCO2inaqueousaminesolutionwasusedtowheremeasurethetermy*Aasshownineq12.

ααistheCO2loadingatanyheightofthecolumn,

lean.loadingistheloadingofsolventcominginatthetopofthecolumn,Misthesolventmolarity,Listheliquidflowrate,Ay*iscrosssectionalareaofcolumn,and,D=(GA

=PCO2

I×yA)where

HeCO2‐amine(12)

72

dx.doi.org/10.1021/ie2008357|Ind.Eng.Chem.Res.2012,51,70−79

Industrial&EngineeringChemistryResearchArticlewhereinterface,y*AHeistheequilibriumconcentrationofgasatthewasstirredat0°Cfor4h.Aqueous1MNaOH(10mL)wasCO2−addedslowlyat0°C.Aftergasevolutionhadsubsided,excess/kmol,amineisthesolubilityofCO2inaqueousaminesolution,kPam3andPCO2isthepartialpressureofCO2solidNaClwasaddedtothemixture.ThemixturewasinkPa.

thoroughlyextractedwithCH2Cl2(2×100mL),theorganicThetermofHeCO2‑aminecanbedeterminedfromeq13:13extractsweredried(Na2SO4),filtered,andthefiltratewasdistilledatatmosphericpressuretoremoveCHHe=He⎛HeCOusingaviqreuxcolumn2Clresultingoilwasthenfractionated2.Thetoaf-CO2‐amineN2‐H2

O⎞

2O‐amine⎜⎜⎝He⎟N⎟

ford4-(diethylamino)-2-butanol(48.3g,49%):bp96−111°C2O‐H2O⎠

(13)

(22mmHg).IR_max:3075−3600cm−1;1HNMR,ı(200ThesolubilityofpureDEABsolventcanbedeterminedfromMHz,CDClthesolubilitiesofCOMeCHO),3):0.73(t,6H,J=7Hz,MeCH1.10−1.30(m,2H,CH2),1.952),0.84(d,3H,J=6Hz,−2.15(m,2H,VersteegandvanSwaaij.2and14NThe2OforwaterbasedontheworkofcorrelationsaredescribedineqsNCH2),2.20−2.45(m,4H,N(CH14and15below:

OHwasnotobserved.2Me)Fractional2),3.52−3.65(m,1H,MeCHO);distillationwasusedtopurifythechemical,theidentityofwhichwaslaterHe⎛2284⎞Nverifiedbyusingthenuclearmagneticresonance(NMR)2O‐H2O=8.55×106exp⎜⎝−T⎟⎠(14)

technique.Approximately,thepurityofDEABwas94.5%andtheimpuritieswere4-methoxy-2-butanol,2.3%;2-butanol,He⎛⎜2044⎝−

⎞

CO1.9%;anddiethylamine1,0.5%.Atthispoint,DEABis2‐H2O=2.82×106expT⎟⎠

(15)

synthesizedin-houseonasmallscale.MEAandMDEAwereTheTineqs14and15canbeusedfromtheexperimentaldataobtainedfromFisherScientificwithapurityof99+%.NitrogenofthepackedcolumnbydeterminingthesolubilityofCOandCO2weresuppliedbyPraxairInc.,withpuritiesof99.9%.thetermofy2inaqueousaminesolution;then,Allmaterialsinthisstudywereusedasreceivedwithoutfurtherineq12.

*Acanbecalculatedpurification.

Theabsorptionperformanceoftheamineswasevaluatedbyy*=PCO2conductingexperimentsinabench-scaleabsorptionunit,ofA

HeCOwhichasimplifiedflowdiagramisgiveninFigure1.Theunit2‐amine

(12)

consistedofaglassabsorptioncolumn(27.5×10−2minFromeq12,wefoundfromourcalculationsthatydiameterand2.15minheight)packedwith37elementsofstain-2.219×10−8

.Thisisverycloseto0.Therefore,practically,A*equals

thelesssteelstructuredpacking(SulzerDX)and(900m2/m3).COThecolumnwasdesignedforacounter-currentmodeofThe2−DEABphysicalreactionpropertiescanbebasedtakenontobetheinstantaneous.

molarityofamineoperationinwhichaliquidsolutionwasintroducedtothesolutionwerederivedfromtheworkofManeeintretal.6incolumnatthetopwhileagasmixtureenteredthecolumnwhichthephysicalandtransportpropertiesofaqueousaminesbelowthepackingsection.AseriesofgassamplingpointsandaminoalcoholsolutionsforCO2captureweremeasured.werealsoinstalledatregularintervalsalongthesidesoftheTheymeasureddensities,viscosities,andrefractiveindicesofcolumntoallowmeasurementsofthegas-phaseCOvariousconcentrationsof4-diethylamino-2-butanol+waterconcentrationduringexperiments,andaseriesofthermo-2mixtures.

couplepointswasalsoinstalledalongthesidesofthecolumntoallowmeasurementsofthetemperatureofliquidduringthe4.EXPERIMENTALSECTION

experiments.

4-Diethylamino-2-butanol(DEAB)withstructuralformulaasTheabsorptionunitwasalsocomposedof(i)two35-LgiveninScheme1wassynthesizedintheCO2laboratoryatthesolutiontanks,(ii)threecalibratedmassflowmeters,(iii)threeneedlevalves,(iv)avariable-speedgearpump,and(v)anScheme1.StructuralFormulaof4-Diethylamino-2-butanolinfrared(IR)gasanalyzer.Thesolutiontanksservedas(DEAB)

reservoirsforsupplyingandreceivingtheliquidsolutionusedintheexperiments.TwomassflowmetersfromAalborgInstruments&ControlsInc.(modelGFM17)wereusedtomeasuretheflowratesforNthethirdmassflowratefrom2andCOBiosInternational2streams,respectively,andCorp.withhighefficiencywasusedtomeasurethemixedgasstreambeforeenteringthecolumn.Thegearpump(Cole-Parmer)wasusedtodrivetheliquidsolutiontothetopofthecolumn.TheIRgasanalyzer(model301D,NovaAnalyticalSystems,Inc.)wasUniversityofRegina,Regina,Canada.Detailsofthesynthesisoperatedduringtheexperimentstowithdrawsamplesandmeasureprocedurearegivenelsewhere.4AbriefdescriptionofthetheCOprocedureishowevergivenasfollows:todiethylamine(49.7g,2concentrationofthegasmixtureinsidethecolumn.ThisanalyzercanmeasureCO0.68mol)wasaddedmethylvinylketone(51.4g,0.73mol)experiments,2concentrationsupto20%.

Priortotheanaqueoussolutionoftheaminedropwisewithstirringoveraperiodof1h.Thereactionwaspreparedinthefeedtankbydilutingtheconcentratedtemperaturewasmaintainedbelow0°Cusinganice-saltbath.aminewithdeionizedwatertoagivenconcentration.ThefeedAftertheadditionwascomplete,thecoolingbathwasremoved,tankisconnectedtonitrogenballoonstokeepthesolventsandthemixturewasstirredatroomtemperature.Thentheunderablanketofnitrogen.TheactualamineconcentrationreactionmixturewasdissolvedinMeOH(20mL).Thewasdeterminedbytitrationwithastandard1.0kmol/m3solutionwascooledto0°CandNaBHhydrochloricacid(HCl)solutionusingmethylorangeportionwise,tothestirredreaction4(30.7g,0.81mol)wasadded,mixture.Themixture

indicator.Horwitz(1975)15presentedtheofficialanalytical

73

dx.doi.org/10.1021/ie2008357|Ind.Eng.Chem.Res.2012,51,70−79

Industrial&EngineeringChemistryResearchArticlechemistrytechniquetodeterminedthesolutionCOwereconductedtotesttheworkingorderofthepackedcolumn.Byaddingexcess1.0NHCIacidtothesample,allof2loading.theCOVerificationoftheapparatuswasperformedbycomparingresultsevolvedintoagasburetformeasurement.

2EachexperimentalrunbeganbyintroducingNwiththeworkdonebydeMontigny,16whichdetailedtheabsorption2andCOgasesfromcylindersthroughmassflowmetersatdesiredflow2ofCO2fromasimulatedfluegasstream(air+COratestoproduceaCOincreasingliquid2)usingaMEAsolution.Theresultsindicatedthatflowrate2−N2gasmixture,whichwasfedtothebottomofthecolumn.TheconcentrationofCOimprovestheKGavvalue,asshowninFigure2.TheresultsservedascheckedbytheIRgasanalyzerandadjusted2inthefeedgaswasuntilthethebasisforcomparisonwiththenewabsorptionexperiment.desiredvaluewasobtained.ThepreparedsolventwasthenTheresultsofthecomparisonbetweenthetwostudiesarepumpedatagivenflowratetothecolumntop,andaneedleshowninFigure2.Thenewdatafollowsthesametrendand

valvewasinstalledonthetopofthecolumntocontroltheliquidflowratesoastocreatecountercurrentcontactbetweengasandliquid.AfterabsorbingCOwas2andtravelingthroughthecolumn,theCO2-richsolutioncollectedcontinuouslyintheliquidreceivingtank.Thisoperationwascontinuedforatleast30mintoallowthesystemtoreachsteadystateconditions.Atthispoint,gasphase-COpositionsalongthecolumnwere2concentrationsatdifferentmeasuredthroughaseriesofsamplingpointsusingtheIRgasanalyzer.Atthesametime,liquidsamplesweretakenfromthebottomofthecolumnandanalyzedfortheirconcentrationsandCO2loading.

5.RESULTSANDDISCUSSION

Theexperimentsformasstransferinapackedcolumnweredividedintothreeparts:absorptionintoMEA,MDEA,andDEAB.Theabsorptionprocesswasconductedinacounter-currentmodeatpresetoperatingconditions.AtsteadystateFigure2.Verifyingthepackedcolumnperformancefor2MMEA

operation,thegasconcentration,andthetemperatureprofiles(bothstudiedatY=0.09(molCO2/molair).

alongthecolumnweremeasuredandrecorded.Aswell,theoutletliquidcompositionwasanalyzedforitsCOfallsinthesamerangeasthedatapublishedbydeMontigny.16transfercoefficientcanbecalculated2loading.Theoverallmassusingeq10.ThereareafewminordifferencesbetweentheexperimentsThetotalnumberofmeasureddatapointsfrom27experimentalfromthetwostudies.Thenewexperimenthadlowergasandrunswas540,whichincludes270measuredpointsofCOliquidflowratethandeMontigny’s16work.However,sincetheconcentrationand270measuredpointsoftemperature.

2gasresultsareinthesamerangeandfollowasimilartrend,theFromthesedata,theconcentrationandtemperaturechangepackedcolumnwasdeterminedtobeingoodworkingorder.alongthecolumnsweremeasured.Theeffectsofliquidflow5.2.EffectofLiquidFlowRateinthePackedColumn.rate,solutionconcentration,andtemperatureonKLiquidflowratehasagreatimpactontheCOexperimentalequipmentusedGaforvwerethenexamined.ThethisoftheDXstructuredpacking.Asshown2absorptionefficiencyinFigure3,

studywasverifiedbycomparingtheresultsfor2MMEAwithresultsfromasimilarsetup16for2MMEAreportedintheliteraturebydeMontigny.Amassbalancecalculationusingeq16wasperformedattheendofeachexperimentinordertoconfirmthevalidityoftherun.ThecalculationcomparedtheamountofCOmeasuredbyIR,withtheamount2removedfromthegasphase,asofCOtheliquidphase,asmeasuredbythesolutionCO2addedintoTheoretically,thesetwovaluesshouldbeequal,2loadingapparatus.butbecauseoftheerrorsinequipment,suchastheIRanalyzer,COloadingmeasurementapparatus,andthelarge,elongatedliquid2trap,themassbalanceerrorobtainedinthisstudywasabout2−6%andthepercentaverageabsolutedeviation(AAD)formassbalancewas14.7%whichisacceptableformasstransferstudies.

mass balance error(%)

=⎛⎜absorbed−removedCO2⎞⎝

absorbedCO⎟×100

2⎠(16)

Figure3.CO2concentrationprofilesatdifferentliquidflowrates3(m325.1.VerificationofthePackedColumn.Maneeintretal.817.85·h))forMEA,DEAB,andMDEA(amine=2.0(kmol/m/(mhadmoreliquidchannelingintheir1-inchcolumnsetup,sothe23.2°(kmol/(m2C;inletliquid·h));Tα==23.60.14°C).

(molCO);GI=2/molamine);inletgasT=newsetuphasmanyimprovementsonthe1-inchcolumnexperiment,asshowninFigure1,andeliminatedchannellinginincreasingliquidflowratecausedareductioninCOthecolumn.Toverifytheexperimentalequipment,afewtrialruns

concentrationofgas-phase,indicatinghigherabsorption

274

dx.doi.org/10.1021/ie2008357|Ind.Eng.Chem.Res.2012,51,70−79

Industrial&EngineeringChemistryResearchArticleefficiency.Thiswasaresultofthebasicphenomenatakingpackingcolumnsoflessthan50mmdiameter,sothemaximumplace,soanincreaseinliquidflowrateledtoagreaterdegreeofliquidflowrateshouldbe5m3/(m2wettedpackingsurfaceandincreasedtheefficiencyofthemass-5.3.EffectofLoading.Thevalue·h).

ofKastheCOGavdecreaseswithtransfer.

COamount2loading,whichisbecauseTheeffectoftheliquidflowrateonKGavvaluesareshowninofactiveaminesdecreases,2loadingincreases,thecausingtheKFigure4forMDEA,DEAB,andMEA.Again,itisevidentthat

betweensystemefficiencyGavtodecrease.Therelationshipandconcentrationofactiveabsorbent,whichisrepresentedintermsofCO2loading,isshowninFigure5.ThehigherCO2

FigureFigure(amine4.=E2.0ffect(kmol/mofliquid3);flGowrateinthepackedcolumnonKGav(amine5.=2.0Effect(kmol/moflean3);GloadingonKGavinthepackedcolumn(molCO(molCOI=17.85(kmol/(m2·h));YMEA=0.09CO2/molair);YMDEA,DEAB=0.14(molCO2/molair);α=0.142/molamine);inletgasT=24.6°C;inletliquidT=23.2axes:RHSrefersinlettogasMEA;T=LHS24.6I=17.85(kmol/(m2°refersC;inlet·h));Y=0.13(mol°C)2/molair);toliquidMDEATand=23.2DEAB).

°C)(vertical(verticalaxes:RHSreferstoMEA;LHSreferstoMDEAandDEAB).

theKGavvaluesforMEAarehigherthanthevaluesforMDEAloading,reflectingthelowerabsorbentconcentration,gavetheandDEABundercorrespondingoperatingconditions,andthislowermass-transfercoefficient(KisbecauseMEAisaprimaryaminewhereasDEABisatertiary0.09Gav)forthesystem.ByincreasingtheCOaminoalcoholandMDEAatertiaryamine.Thenewchemical2loadingfromto0.25mol/mol,theKaminoalcoholDEABprovidesagreaterKGavvaluewasreducedby20%forMDEAandDEABandforMEAby50%,asshowninFigure5.

by30%andthisledtoincreasedefficiencyGaofvvaluethanMDEAthecolumn.Also,TheexperimentalresultsgiveninFigure5showthatCOtheresultsshowthattheliquidflowratehasaninfluenceontheloadingofthefeedsolutionhadanapparentinfluenceonthe2valueofKKKGav(i.e.,anincreaseinliquidloadgenerallyyieldsagreaterGavvalue.AsCO2loadingincreased,theKGavvalue).ThepossiblereasonforthisbehaviorisresultinginaloweroverallmasstransferGacoevvaluedeclined,fficient.Thisthatahigherliquidloadleadstothefollowing:(i)agreaterillustratesareductionintheefficiencyofthecolumns.

liquidsidemasstransfercoefficient(kL),whichisdirectly5.4.EffectofConcentrationonLoadingCapacityofproportionaltotheoverallKDEAB.TheconcentrationofDEABhasagreateffectoncyclictransfer;(ii)aGagreatervinthecaseofliquid-phasecontrolledmasseffectivearea,whichiscapacity.AsshowninFigure6,decreasedconcentrationof

causedbymoreliquidspreadingonthepackingsurface;and(iii)theamountoffreeaminemoleculesinthesystembecominglargerorthesystemhavingmorecapacitytoabsorbCO2fromthegasphase,therebyenhancingtheKGavvalues.ThisisevidencethatDEABisnotanunusualsolventbutfollowstrendssimilartothoseexhibitedbyothersolventsasafunctionofthesementionedparameters.TheunusualresultsarethespecificattributesofDEABcomparedtoMDEA(bothtertiaryamines).Asindicatedbefore,theunexpectedresultsarethatDEABhaslargercyclicandabsorptioncapacitiesthanMDEA,whileatthesametime,itsmasstransfercoefficientislargerthanthatofMDEA.

However,theincreaseinsolutionflowrateleadstohighercirculationandregenerationcosts,and,thus,mightnotimprovetheoverallsystemefficiency.Maximizingliquidflowratesmightnotleadtooptimumoperatingconditions.Therefore,anoptimumFigurefl17.856.EffectofDEABconcentrationonloadingcapacity(GI=thanowratewhenonflowthetheratehastobedetermined.TheeffectofliquidliquidKGavvaluefrom4.0to5.0m3/(m2flowrateincreasesfrom5.0·h)tois7.0greaterm3liquid(kmol/(m2T=22.8°·C).

h));L=5m3/(m2·h);inletgasT=24.0°C;inlet2rate·h),forandthisthetypereasonofpackingforthisIDisisthat5.0them3/(mmaximum/(m2liquidflowDEABfrom2to1.1molledtoincreaseinrichloadingfromspecificationsfromSulzerforoperatinglaboratory·h),thestructured

technical0.26to0.65mol/mol,respectively.Thisincreaseofrichloading

75

dx.doi.org/10.1021/ie2008357|Ind.Eng.Chem.Res.2012,51,70−79

Industrial&EngineeringChemistryResearchArticleledtoanincreaseofcycliccapacityby60%,whichwouldofferedahigherperformanceof65%.Thehighestremovalreducethecostofregeneration.Regenerationhasalarge(100%)wasachievedbyMEA,which,obviously,reachedinfluenceon2thecapitalcostofgastreatingplants,asindicatedinAstarita.completeremovalefficiency.IfwereducetheconcentrationofThecycliccapacity(Cthoseaminesto9%,theactualperformanceofthesethreeC)canbeexpressedmathematicallyasfollows:

aminescanbedifferentiatedbyconsideringthecolumnheightrequiredtoreach9%removal.AccordingtoFigure7,COCC=(mole of CO2abs−mole of CO2regen)/VsolutionabsorptionintoMEAsolutiontookplacewithinonly0.4mfrom2thecolumnbottom,whileDEABandMDEArequiredashighCC=(loading of CO2abs−loading of CO2regen)(M)(17)

as1.25mand2.15m,respectively,tomeetthesameremovalwhereM=concentrationofsolution(mol/L).

target.ThisisbecauseMEAisaprimaryamine,butMDEAisa5.5.COtertiaryamineandDEABisatertiaryaminoalcohol.Onthe2AbsorptionPerformanceofSingle-AmineSolutions.Figure7showsthegasphaseCObasisofthemasstransferperformancealone,bothDEABand2concentration

MDEAwouldnotbegoodcandidatesassolventsforCOcapture.However,oneoftheotherimportantcriteriafor2solventselectionistheenergyrequiredforsolventregenera-tion.Inthiscase,wehaveshown3,4thatDEABisbyfarbetterthanMEA.Therefore,itisapparentthatDEABasablendedsolventwithsuchaprimaryamineisagoodcandidateforCOcapture.However,beforewemoveintothestudyofthemass2transfercharacteristicsofablendedsolvent,itisveryimportanttodeterminethemasstransfercharacteristicsoftheindividualcomponents(i.e.,aqueousDEABalone)inordertoevaluateboththecontributionsofDEABandtodeterminetheblendratiooftheblendedamines.InthisstudywearedoingasystematicstudyofDEAB(atertiaryamine)sothatwecandeterminethecontributionsofDEABwhenmixedwithaprimaryorsecondaryamineintheoverallperformanceoftheblendedamineinCOCO2capture.

Furthermore,2ismoresolubleFigure7.CO22inDEABthanMEA.Also,2concentrationprofileforsingle-aminesolutions2Monthebasisofourpreviousstudy,itwasshownthatDEABMEA,hasaveryhighCOloadingMDEA,=0.14and(molDEABCOatliquidflowrate,5(m3/(m2·h))andlean2/molamine).

conventionalaminessuch2absorptioncapacitycomparedwithasMEAevenatpracticalpartialpressuresofCOprofilesthatwereobtainedfromabsorptionexperimentsusinginFigure8.17Also,2usedintheCOthemolecular2absorptionprocess,asshownweightofDEABishigh,145threesingle-aminesolutions(i.e.,MEA,MDEA,andDEAB).g/mol.ThisalsoimplieshighviscosityatveryhighDEABDuringthe3experiments,theconcentrationofamineswasconcentrations.Bearinginmindtheviscositylimitation,we2.0kmol/mandtheCO2loadingofthefeedsolutionswasdecidedtoworkwithDEABconcentrationsintherangeof(0.14mol/mol).ConsideringtheCO1−2.0M.Also,intermsofvolatility,ItcanbeseenthatthecanbeseenthattheMDEA2removalefficiencyforeachprofile,itsolventgavethevolatilitylowestCO2absorptionperformanceat45%,whileDEAB

17ofDEABislowerthanMEAandabithigherthanMDEA.However,oneoftheotherimportantcriteriafor

Figure8.EquilibriumsolubilityofCO2inaqueoussolutionsof2MMEA,2MDEA,2MMDEA,2MAMP,2MDEAB,and2MPZbySema17(linesaretrendlinesoftheexperimentalresults(Sema17))3−5.

76

dx.doi.org/10.1021/ie2008357|Ind.Eng.Chem.Res.2012,51,70−79

Industrial&EngineeringChemistryResearchArticlesolventselectionisenergyrequiredforsolventregeneration.InPrepresentspartialpressureofCOthiscase,literaturehaveshown3,4thatDEABisbyfarbettereq20and21andbyreplacingtheko2(kPa).ByLtermwithLx

rearranging

,therelationthanMEA.Therefore,itisapparentthatDEABisagoodbecomesasfollows:

candidateforCO2capture,mostpreferablyasablendedsolventwithaprimaryaminesuchasMEA.

Kx⎡(αeq−α)C⎤

Gav∝L

⎢⎣

⎢

6.MASS-TRANSFERCORRELATIONS

P⎥CO2⎥⎦(22)

Masstransfercoefficientcorrelationsareimportantbecausetheyenablepredictionoftheexperimentalresults.SeveralThisrelationshipcanbesolvedbyplottingthetermof(Kcorrelationshavebeenusedtopredictmasstransfercoefficientsthetermof[(αGaofv/Lx)againsteq−α)C]/PCO2.Inthedevelopmentthisinchemicalengineeringprocesses.Stringle18predictedcorrelation,severalvaluesofxweretriedbutthebestresultswerecorrelationsforoverallmasstransfercoefficientsKobtainedwithx=0.6.TheplotisshowninFigure9.Alinear

follows:

GavasKGav∝LbGc

(18)

whereLrepresentsliquidmassvelocity(lb/(ft2h)),bisthecoefficient,Grepresentsgasmassvelocity(lb/(ft2h)),andcisthecoefficient.Whentheliquidfilmcontrolsthesystem,thevalueofbisbetween0.22and0.38,dependingonthetypeofpacking.Theeffectofliquidfilmongasrateisverysmall,andthevalueofcineq18normallyisonly0.06−0.08.However,thefilmvalue,controlsc,increasesthesystem.tobetween18,190.67and0.80whenthegasKohletal.20predictedthecorrelationforKGavforMEAasfollows:

KGav=F(L/μ)2/3[1+5.7(αeq−α)Ce0.0067T−3.4p](19)

wereFrepresents2thepackingfactor,Listheliquidflowrate(lb/(hft)),μisliquidviscosity(cps),αeqisCOFigure9.Relationshipbetween(KGav/L0.6)and[(αeq−α)C]/PCO2atP2loadingofsolutioninequilibriumwithCOsolventconcentration=2.0(kmol/m3).

2,αisCO2loading,Cisamineconcentration,PrepresentspartialpressureofCOsolution,andTrepresentsthetemperature.

2overtheregressionofthedatapointintheplotproducesanequationoftheTheKGavcorrelationineq19issuitableforanMEAsystemcommonformy=mx+bwheremistheslopeofthelineandbisusingrandompackingintheabsorptioncolumnbutisnottheyaxisintercept,soeq22canberewrittenasfollows:

suitableforotherabsorptionsystems,assuggestedbydeMontigny16andAroonwilasandTontiwachwuthikul21becauseofdifferentpackingcorrelationfactorsthatwillprovideKGav=mLx⎢⎡(αeq−α)CdifferentK⎢⎣P+b⎤⎥CO2⎥⎦(23)

Gavdata.NewcorrelationsweredevelopedbydeMontigny22forMEAsystemstopredictKThiscorrelationfortheKGaforGavvalue.

6.1.DevelopmentofCorrelationsDEAB.Inthistovvaluewasplottedagainsttheactualexperimentalvaluesinorderchecktheaccuracyofthemodel.Asstudy,theoverallmasstransfercoefficientcorrelationwasshownineq23,KGavisafunctionofliquidflowrate(L),COdefinedintermsoftheindividualmasstransfercoefficient,asofsolution(α),concentrationisafunction2isafunctionoftheloadingshownineq4below:

oftheloadingofthesolution(C),partialpressureisafunctionof11theloadingofCO2inkPa(PCO2),whilem,xandbareconstantK=+HGkGIkparameters.Inthisstudy,x=0.6yieldedtheleastdeviationofL0(4)

experimentalvaluesfrompredictedvalues,whereasthevaluesforThetermsontheright-handsideoftheequationrepresentthethecoefficientsmandbwereobtainedfromtheexperiment.Thegasmasstransfercorrelationpredicted,theKfipackingswhichwererelativelyingoodGavvaluesforstructuredagreementwiththegaslmandfiresistanceliquidfilm,theequationtolmmassresistancescantransfertobesimpliismassmuchtransfer.fiedlargerWhenas

thanthattheofliquidtheexperimentalvalues.The%AADofthiscorrelationwas14.57%.KBecauseofthenatureofitsderivation,thismodelisprovidedonlyGav∝IkLo

av

(20)

forthepurposeofpredictingunknownKAccordingtoPerry,23thetermofkoLineq20wasshowntovaryexperimentalconditionsforDEABforthispacking.

Gavvaluesbasedonasafunctionofliquidflowrate(L)tothepowerofx.The6.2.RelativeCostIssues.ThecostofbuildingaCOexponentabsorptionplantisdependentonthechemicalsolventandthe22xliesinarangebetween0.3and0.7.Furthermore,Astaritasuggestedthattheenhancementfactor,I,canbeenergythatcanbesavedfromtheregenrationprocess.Theexpressedasfollows:

mostpromisingareasforoperatingcostsavingswiththenewchemicalsolventarefromhighcycliccapacityandlowenergyI≈

(αeq−α)C

ofconsumption.ItisestimatedthatthenewchemicalsolventPCO2

(21)

(DEAB),whichhashighcycliccapacityandlowregenerationwhereαenergy,candecreasetheenergyrequirementby50%.SincetheeqrepresentstheCO2loadingofsolutioninequilib-energyrequirementforregenerationaccountsforabout70%ofriumwithPCO2,αisCO2loading,Cisamineconcentration,and

operatingcosts,andbysplittingcapitalandoperatingcostsfor

77

dx.doi.org/10.1021/ie2008357|Ind.Eng.Chem.Res.2012,51,70−79

Industrial&EngineeringChemistryResearchArticleCO2captureat50%each,itcanbeestimatedthatbyusingc=coefficient

DEAB,theCO2capturecostscanbereducedbyabout35%.CAlso,eventhoughthecostofproducingDEABnowishigh,thisDEABA,L=bulkconcentration(kmol/m3)=4-(diethylamino)-2-butanolissomewhatcompensatedbytheneedtouselowerDEABF=packingfactor

concentrationscomparedwithhigherconcentrationsofMEAG=gasmassvelocity(lb/(ft2toachievethesameremovalefficiency.

G=gasflowrate(kmol/(m2·h))7.CONCLUSIONS

HeI·h))

N2O=solubilityofN2O,kPam3/kmol

ComparativemasstransferstudiesofCOHECO2‑amine=solubilityofCO2inaqueousaminesolution,2absorptioninaqueousMEA,MDEA,andDEABsolutionshavebeenconductedusingkPam3/kmol

structuredpackinginanabsorptioncolumn.TheperformancewasHEN2O‑amine=solubilityofN2Oinaqueousaminesolution,presentedintermsoftheoverallmasstransfercoefficient,KkPam3/kmol

Also,theeffectsofvariousparameters,suchasCOGav.2loadingoftheHeN2O‑DEAB=solubilityofN2OinpureDEAB,kPam3/kmolsolution,flowratesolutiononKconcentrationonloadingcapacity,andliquidHeCO2‑H2O=solubilityofCO2inwater,kPam3/kmolGav,werestudied.

(1)TheresultshaveshownthattheCOHeN2O‑H2O=solubilityofN2Oinwater,kPam3/kmolsolutionandliquidflowratehaveeffectsonK2loadingoftheI=enhancementfactor

increasesasliquidflowrateincreasesandKaGavinthatKGavGvisimprovedbykG=gassidemasstransfercoefficient(kgmol/(m2decreasedK·s·kPa))G=overallmasstransfercoefficient(kgmol/(m2flowrateandCOCO2loadinginthefeedsolution.Also,theliquid2loadinghaveasignificantimpactontheCOK·s·kPa))Gaabsorptionperformanceofstructuredpackings,andanincrease22v=volumetricoverallmasstransfercoefficientintheseparametersgenerallyimprovesthemass-transferLefficiency.However,thereisnochangeintheKμ=·sliquid·kPa))

(kgmol/(mflowrate(m3/(m2resultingfromachangeintheinertgasflowrate.

GavvalueMDEA=liquid·h))=methyldiethanolamineviscosity(cps)

(2)Theconcentrationofanewchemicalsolvent,DEAB,hasMEA=monoethanolamineagreateffectonloadingcapacity;astheconcentrationofDEABNA=massfluxdecreased,theloadingcapacityincreased.Also,theexper-P=pressure,kPa

imentalresultshaveshownthattheDEABsolventhashigherPCO2=partialpressureofCO2,kPaCO2removalefficiencythanMDEA.

T=temperature,C°(3)Thenewsynthesizedaminoalcohol(DEAB)providedYmuchhigherCOA,G=moleratiovalue

2absorptioncapacityandhighercyclicycapacity,andthisledtoreductionofthecirculationrateandA,G,yA,i=molefraction(kgmol/kgmol)yenergyforsolventregeneration.ReductionoftheenergyZ*A==height,gasphase(m)

inequilibriumrequirementandtheheightofthecolumnswouldleadtoreducedcapitalcostandlong-termoperatingcosts.TheDEABαeq=CO2loadingofsolutioninequilibriumwithPCO2(molsystemprovidedanexcellentoverallmasstransfercoefficientCOthatishigherthanthatoftheMDEAsystem.

(4)AnewcorrelationwasusedtopredictoverallmasstransfercoefficientforDEABsolventusingstructuredpackings.Inthisstudy,theK■

α=2/molCOamine)

2loading(molCO2/molamine)

REFERENCES

Gavvaluesareinrelativelygoodagreementwiththeactualvalueandthe%AADofthiscorrelationisStorage(1)IPCC.andNew;CambridgeIPCCSpecialYork,2005.

UniversityReportPress:onCambridge,CarbonDioxideUnitedCaptureKingdomand14.57%■comparedwiththe20%valuereportedintheliterature.

AUTHORINFORMATION

Solvents(2)Astarita,;Wiley:G.,NewSavage,York,D.1983

W.,Bisio,A.GasTreatingwithChemicalCorrespondingSynthesis,(3)Maneeint,K.;Idem,R.O.;Tontiwachwuthikul,P.;Wee,*COSolubilities,andGasCyclicStreams.Capacities9thInternationalofAminoConferenceAlcoholsA.G.forH.2CapturefromFlueonRaphael.idem@uregina.ca.

Tel.:+1306585Author

4470.Fax:+13065854855.E-mail:GreenhouseNotes

Nov(4)16Tontiwachwuthikul,−20,Gas2009.

Technologies(GHGT9);WashingtonDC,USA,P.;Wee,A.G.H.;Idem,R.O.;Maneeintr,■Theauthorsdeclarenocompetingfinancialinterest.

K.;MethodFan,G.-J.;forcapturingVeawab,A.;carbonHenni,dioxideA.;Aroonwilas,fromgasA.;streams.Chakma.WorldA.ACKNOWLEDGMENTS

IntellectualA.Naamiwouldliketoacknowledgethescholarshipsupport022437,PropertyOrganizationPatent(WIPO),No.WO/2008/fromtheLibyanHigherEducationalstudiesthroughthe(5)Maneeintr2008.

K.;Idem,R.O.;TontiwachwuthikulP.;Wee,A.G.H.culturalsectionoftheLibyanEmbassyinOttawa,Canada.AlsoSolventthefinancialsupportfromtheInternationalTestCentreforThe2ndDevelopmentInternationalforConferenceCO2CaptureonAdfromVancesIndustrialinPetrochemicalsGasStreams.andCO2Capture,UniversityofRegina,Regina,andNaturalSciencesandEngineeringResearchCouncilofCanadaWee,(6)PolymersManeeintr,(ICAPP2007);A.G.H.PhysicalK.;Henni,andA.;Bangkok,Thailand,Jun25−28,2007.TransportIdem,R.PropertiesO.;Tontiwachwuthikul,ofAqueousAminoP.;■

(NSERC)isappreciated.

AlcoholEnviron.SolutionsProt.2008for,86CO,2912Capture−295.

fromFlueGasStreams.ProcessSaf.NOMENCLATUREb=coefficient

Wee,(7)Maneeintr,C=amineconcentration(mol/molamine)

COA.G.H.ComparativeK.;Henni,A.;AqueousMassIdem,R.O.;Tontiwachwuthikul,P.;SolutionsTransferofDEABPerformanceandMEA.StudiesInd.Eng.ofChem.2AbsorptionintoRes.2010,49(6),2857−2863.

78

dx.doi.org/10.1021/ie2008357|Ind.Eng.Chem.Res.2012,51,70−79

Industrial&EngineeringChemistryResearchfi(8)Aroonwilas,A.;Tontiwachwuthikul,P.MassTransferCoef-propanolcientsand(AMP)CorrelationUsingforStructuredCO2AbsorptionPacking.Ind.intoEng.2-Amino-2-methyl-1-Chem.Res.1998,37Transfer(9),569.

Aroonwilas,CoefficientsA.;ofTontiwachwuthikul,StructuredPackingP.inBehaviorCOoftheMassChemical(10)deMontigny,Reactions.D.;Ind.Tontiwachwuthikul,Eng.Chem.Res.1999P.;Chakma,,38,2Absorptionwith2004.

A.ParametricStudiesMonoethanolamineofCarbonDioxideAbsorption(11)Fernandes,J.;Solutions.Lisboa,PedroCan.F.;J.Simoes,Chem.intoEng.HighlyPedro2001C.;,Concentrated79Mota,,137.JoseP.B.;ExtractionEsteban,(12)Geankoplis,withS.ApplicationStructuredofCFDintheStudyofSupercriticalFluidC.J.Packing.TransportJ.Supercrit.ProcessandFluidSeparation2009,50,61Process−69.Principles(13)Sema,,4thT.,ed.;Edali,PrenticeM.,Naami,Hall:NJ,A.,2003.

Idem,R.Tontiwachwuthikul,P.NButanol2OSolubility(14)Versteeg,Solutions.andG.Ind.N2OF.;vanEng.DiffSwaaij,Chem.usivityW.ResofP..AqueousM.2012Solubility,51,4-(Diethylamino)-2-925and−930.

diffusivityofacidEng.gasesData1988(CO2,,33N,2O)29−in34.

aqueousalkanolaminesolutions.J.Chem.Methods(15)Horwitz,(16)deMontigny,,12thed.;W.GeorgeAssociationD.ComparingBanta:ofOManasha,fficialAnalyticalPackedColumnsWI,1975.

Chemists(AOAC)andMembraneAbsorbersRegina,Saskatchewan,forCO2Capture.Canada,PhD.2004.

Dissertation,UniversityofRegina,Correlations(17)Sema,T.;Naami,A.;Idem,4-(Diethylamino)-2-butanolforEquilibriumSolutions.SolubilityofR.;Ind.CarbonTontiwachwuthikul,Eng.Chem.DioxideRes.in2011AqueousP.,50,14008(18)−Strigle,14015.

R.F.RandomPackingandPackedTowers;GulfPublishingCarbamate(19)Benamar,Company:Houston,TX,1987.

ConcentrationA.;Aroua,inM.DEA,K.MDEAModelingandofTheirCO2MixturesSolubilityUsingandthePublishing(20)DeshmukhKohl,Co:A.−MatherHouston,L.;Riesenfeld,Model.FluidTX,1997.

F.C.PhaseGasPuriEquilib.fication2005,5th,231ed.;,150.Gulfin(21)2-Amino-2-methyl-1-propanolTontiwachwuthikul,P.;Meisen,Solutions.A.;Lim,J.Chem.J.C.SolubilityEng.Dataof1991CO2,36Ultra-highly(22),130.

deMontigny,ConcentratedD.CarbonMonoethanolamineDioxideAbsorptionSolutions.StudiesMasterUsing’sthesis,(23)Perry,UniversityR.H.;ofRegina,Green,D.Regina,Perry’Saskatchewan,sChemicalEngineerCanada,’sHandbook1998.,8thed.;McGranHillBookCompany:NewYork,2008.

79

Articledx.doi.org/10.1021/ie2008357|Ind.Eng.Chem.Res.2012,51,70−79