Oxidative degradation of a novel AMP/AEP blend designed for CO2 capture based on partial oxy-combustion technology

OxidativedegradationofanovelAMP/AEPblenddesignedforCO2 capturebasedonpartialoxy‐combustiontechnology F.Vegaa,M.Canoa,A.Sannab,J.M.Infantesa,M.M.Maroto‐Valerb,B.Navarretea aChemicalandEnvironmentalEngineeringDepartment,SchoolofEngineering,UniversityofSeville,C/Caminodelos Descubrimientoss/n41092Sevilla,Spain,Phone:954481397,[email protected] bResearchCentreforCarbonSolutions(RCCS),SchoolofEngineering&PhysicalSciences,Heriot‐WattUniversity, Edinburgh,EH144AS,UK Abstract Solventdegradationandvolatilecompoundemissionsaretwoofthemajorconcernsaboutthe deploymentofcarboncapturetechnologiesbasedonchemicalabsorption.Inthiscontext, partialoxy‐combustionmightreducethesolventdegradationduetotheuseofahigherCO2 concentratedfluegas.ThisworkevaluatestheoxidativedegradationofanovelAMP/AEPblend, namelyPOS#1,underpartialoxy‐combustionconditions.Theeffectsoftemperatureandflue gascompositionwereevaluatedintermsofsolventloss,degradationrates,NH3emissionsand degradationproducts.Theexperimentsweresetattemperaturesupto70˚Candtwolevelsof O2concentration–3%v/vand6%v/v.TheCO2concentrationofthefluegasrangedbetween 15%v/vand60%v/vCO2. ThenovelsolventPOS#1showedhighresistancetodegradeandresultedinlowerdegradation ratesthanMEAinalltheoperatingconditionsevaluatedinthiswork.Themaximumdegradation ofAEPandAMPwas24%and19%,respectively.MEAdegradedalmostdoubleunderthesame conditions.TemperatureandO2concentrationenhancedtheoxidativedegradationofPOS#1. However,theuseofhigherCO2concentrationinthefluegasledtolowerdegradationratesof AEPandAMPandhenceoxidativedegradationwaspartiallyinhibitedunderpartialoxycombustionconditions.ThepresenceofhigherCO2contentinthefluegasdecreasedtheNH3 productionanda70%reductionofitsemissionswasachievedastheCO2concentrationshifted from15%v/vto60%v/v.Othermajordegradationcompoundssuchasformateand2,4‐lutidine werealsodecreased.Newdegradationproductswerenotidentifiedsothatthesuggested degradationpathwaysproposedintheliteraturewerenotinfluencedbythepresenceofhigher CO2concentrations. Keywords:degradation;partialoxy‐combustion;solvent;CCS;CO2capture 1. Introduction Thetransitiontowardsalow‐carbonenergysystemreliesonimprovingprocessefficiencyand thedeploymentofrenewableenergyproduction[1].Theenergysectorcontributeswithtwothirdofthegreenhousegas(GHG)emissionsworldwideandtheprimaryenergydemandis expectedtoincrease30%by2040[2].Inthisframework,carboncaptureandstorage(CCS) technologiesareconsideredafrontlineoptiontomeettheEuropeanCommission(EC)2030 strategyaims(40%reductionofGHGemissionsin2030comparedtothe1990emissionlevels) andthereductiontargetsproposedbytheParisAgreementinlastConferenceofParties(COP‐ 21)[3]. AmongCO2capturetechnologies,partialoxy‐combustionisemergingasapromisingapproach thatmightreducetheenergypenaltiesrelatedtotheoverallCO2captureprocess[4].Itisbased ontheuseofoxygen‐enrichedairasanoxidizertoproduceahigherCO2concentratedfluegas thatshouldenhancetheCO2separationprocess,typicallybasedonchemicalabsorption[5]. Partialoxy‐combustioncanpotentiallyreducethecostsassociatedtoboththehighpurityO2 productionrequiredfromoxy‐combustiontechnologyandtheCO2separationprocessrelated topost‐combustiontechnology[6].AccordingtoFavreetal.,partialoxy‐combustionmight achievea25%energyreductionoftheCO2captureprocess[7]. Inaddition,solventdegradation,whichcontributesupto10%ofthetotalCO2capturecosts, constrainsthedevelopmentofchemicalabsorptionanditsapplicationinCCStechnologies[8,9]. Furtherconcernsaboutemissionsofvolatilesolventsandtheirreactiontoformcarcinogens compounds,suchasnitraminesandnitrosaminesintheatmosphere,arecurrentlylimitingtheir useasaCO2mitigationapproach[10].Recentstudiesonmonoethanolamine(MEA)degradation showedthattheoperatingconditionsderivedfrompartialoxy‐combustioncanleadtoa substantialreductionofboththesolventdegradationratesandtheemissionsofvolatile degradationcompoundsandsolventsunderabsorptionconditions[11].Thepresenceof elevatedCO2concentrationinthefluegaslikelylimitstheO2mass‐transferintothebulkliquid andhenceinhibitstheencountersbetweenthesolventandO2[12]. Thesemainfindingsdescribedaboveencourageresearchtodevelopnewsolventformulations andblendsthatcanimprovetheoverallperformanceoftheCO2captureprocessintermsof fasterabsorptionkinetics,higherCO2absorptioncapacity,lowerenergyrequirementsfor solventregenerationandhigherresistancetothermalandoxidativedegradation[12‐15]. Primaryaminessuchasmonoethanolamine(MEA)havebeenproposedtobeblendedwith tertiaryamines,showinglowerenergyrequirementsduringthesolventregenerationstage[16]. Potassiumcarbonatepromotedwithpiperazine(PZ)and2‐amino‐2‐methyl‐1‐propanol (AMP)/PZblendshasalsoshowedfurtherimprovementsontheabsorptionkinetics[17,18].In thisrespect,anewsolventblend,namelypartialoxy‐combustionsolvent(POS#1),resultedina highperformanceforCO2captureunderpartialoxy‐combustionconditions.POS#1combines AMPwithaminoethylpiperazine(AEP)showinganexcellentbehaviourintermsofabsorption kineticsandcyclicCO2capacity,particularlyat40%v/vCO2concentrationinthefluegas[19]. AMPhasshownahighresistancetobedegradedboththermallyandinpresenceofO2[20,21]. Itsresistanceisnotalteredasitisblendedwithotheramino‐basedsolvents[22].Thelow degradationratesshowedathightemperaturemakeAMPanattractivesolvent[23].Itsuse shouldallowtooperatethestrippingstageatelevatedtemperatures.Elevatedstripping pressureleadtohigherstrippingtemperature,whichmightreducefurthertheenergy consumptionduringtheregenerationstage[24].AMPresultedhighlyresistanttodegradein comparisonwithotherprimaryamines.Itisunlikelypossibletoabstractonehydrogenatom fromthetertiarycarbonoftheα‐positionnitrogenatomduethemolecularstructureofAMP andhencetheformationofradicalsviahydrogenabstractionisinhibited[20,21].Table1 summarizesthemostrelevantdegradationcompoundsreportedfromtheliterature.With respecttooxidativedegradationcompounds,Lepaumierexplainedthatmethylationreactions shouldbeaprimarydegradationstepforAMPdegradation[20].Anewdegradationpathway hasbeenproposedbyWangandJensinwhichAMPmaydegradeviaperoxylradicalformation [22]. AEPbelongstopiperazinesthatalsoshowedlowdegradationratesattemperaturesbelow 150C[13].AlthoughAEPprovidedanelevatedresistancetooxidativedegradation,the degradationrateswerehigherthanbothaqueoussolutionofPZandAEP/PZblends[25].The maindegradationcompoundsidentifiedfromAEPdegradationstudiesaresummarizedinTable 2.Mostofthemwereidentifiedfromthermaldegradationexperimentsincombinationwith othersolvents.AlthoughhereisnospecificstudiesonAEPdegradationmechanism,Freeman suggestedthatAEPisasnucleophileasPZandhenceAEPmightattackdifferentprotonatedAEP toformthermaldegradationcompounds.Mostlikely,AEPreactswithprotonatedeithertheαnitrogenornitrogenfromtheprimaryaminogroup[26,27].                  Table1.MaindegradationproductsidentifiedfromAMPdegradationstudies SolventMoleculeStructureMolecular WeightCASnumberReference   2‐amino‐2‐methyl‐1‐propanol    89124‐68‐5‐ DegradationcompoundMoleculeStructureMolecular WeightCASnumberReference Ammonia  177664‐41‐7[20,21,23] Formate  45‐[20,21,23,28] Acetone  5867‐64‐1[21,23] Acetoneoxime  73127‐06‐0[23] Oxalate  88‐ [20,21,23] 4,4‐dimethyl‐2‐oxazoline  9930093‐99‐3[23] 2,4‐Lutidine  107108‐47‐4[21,23] 4,4‐dimethyl‐2‐oxazolidinone  11526654‐39‐7[20,21,23] Foroxidativedegradationpathways,hydrogenabstractionofoneofthemethylanecarbonin theringissuggestedtobetheinitialsteptoproduceoxidativedegradationofAEP.Peroxyl radicalmightreactinalkalinemediumtoproduceperoxideviaintermolecularhydrogen abstraction[27]. Table2.MaindegradationproductsidentifiedfromAEPdegradationstudies SolventMoleculeStructureMolecular WeightCASnumberReference   1‐(2‐aminomethyl)piperazine    129140‐31‐8‐ DegradationcompoundMoleculeStructureMolecular WeightCASnumberReference Ammonia  177664‐41‐7[13,27] Formate  45‐ [13] Piperazine  86110‐85‐0[27,29] N‐formylpiperazine  1147755‐92‐2[13] 1‐[2‐[(2‐ aminoethyl)amino]ethyl]piperazine*  129140‐31‐8[13,29] 1,4‐piperazinediethanamine*  1726531‐38‐0[13,27,29] 1,1'‐(1,2‐ethanediyl)bis‐piperazine  19819479‐83‐5[13] PolyAEP  ‐ ‐ [27] ThisworkaimsatevaluatingthedegradationofPOS#1underpartialoxy‐combustionconditions inordertoconfirmboththeresistanceoftheAMP/AEPblendunderthesenewoperating conditionsandthecapacitiesofthistechnologytodecreasesolventdegradation.Theeffectsof keyoperatingparameterssuchastemperatureandoxygenconcentrationarealsostudied.The degradationofAMPandAEParefurtherdiscussedintermsofsolventloss,maindegradation compounds,volatilecompoundemissionsanddegradationmechanisms. Althoughdegradationpathwaysandthemostrelevantdegradationcompoundsfromoxidative degradationofAMPandAEPhavebeenstudiedatdifferentO2partialpressureand temperatures,theeffectofCO2concentrationinthefluegasisstillunknown.Inthiswork,the influenceofhigherCO2contentinthebulkgasonthedegradationproductformationwillbe evaluatedandnewinsightsaboutmajordegradationproductsandtheirpossibledegradation pathwayswillbefurtherdiscussed. 2. Experimentalprocedure  AEPandAMPsolventsweresuppliedbyAcrosOrganicswithpurityover99vol.%.Thesolution werepreparedusingdeionizedwater.ThesyntheticfluegaswithdifferentCO2/O2compositions wereprovidedin50L‐cylindersbyLinde.Thecertificateanalysisofthecylinders(DINENISO 6141)guaranteesalevelofuncertaintyofthewithin+/‐1%.Thesulphuricacidrequiredfor ammoniaemissiondeterminationwasprovidedina0.1NaqueoussolutionformatbyPanreac withalevelofuncertaintywithin+/‐1%. Inthiswork,operatingparameterssuchastheabsorptiontemperature,theCO2andtheO2 concentrationofthesyntheticfluegaswerestudiedinordertoevaluatetheireffectsonthe oxidativedegradationbehaviourofPOS#1solvent‐AEP/AMPblend‐.Asummaryoftheranges evaluatedinthisworkisreportedinTable3.VariationsoftheCO2compositionofthesynthetic fluegassimulatethedifferentlevelofoxygen‐enrichedairusedinpartialoxy‐combustion capture.Conventionalpost‐combustioncapturerepresentsa15%v/vCO2–typicallyCO2 compositionfromair‐firingcoalcombustion[30]–whereashigherCO2concentrationsimplya higheroxygencontentintheoxidizerforpartialoxy‐combustionoperations.ThemaximumCO2 concentrationstudiedinthisworkwas60%v/vCO2,correspondingtoan80%v/voxygenenrichedoxidizerduringthecombustionprocess[11].AccordingtoFavreatal.,highervaluesof theoxygen‐enrichedairshouldnotproducefurtherreductionsontheoverallenergy requirementsoftheCO2captureprocessunderpartialoxy‐combustionconditionsrespectto post‐combustionandoxy‐combustion[7].Inaddition,theO2concentrationvariedfrom3%v/v to6%v/vandtheoperationtemperaturewasevaluatedattheoperatingtemperatureandthe temperaturebulgeinaconventionalCO2chemicalabsorptionprocess‐50°Cand70°C, respectively.Twoexperimentswereconductedat100Ctoevaluatetheoxidativeandthermal degradationatelevatedtemperature.Adetaileddescriptionoftheexperimentalparametersis reportedinpreviouswork[11]. Table3.Summaryoftheexperimentaltestscarriedoutinthiswork ParameterOperatingrange CO2(%v/v)15204060 O2(%v/v)36363636 Temp(°C)50 705070507050705070507050701005070100  Figure1showsaschemeofthelab‐rigusedinthiswork.Adetaileddescriptionofthelab‐rig andtheoperatingprocedurecanbeextractedfrompreviouswork[11].Briefly,thelab‐rig consistsofa250‐mLsemi‐batchstirredreactorwheretheAMP/AEPblendwasplaced.Theinitial solventconcentrationwas15%p/pAMPand20%p/pAEP,whichisunderpatentapplication[31]. AsitcanbeseeninFig.1,fourexperimentscouldbeperformedinparallel.Syntheticfluegas wassuppliedfromcylindersandthedesiredfluegasflow–0.1mL/min–wascontrolledusing mass‐flowcontrollers.Thesyntheticfluegaswassaturatedinwaterattheexperimental temperaturebeforebubblingitintothereactors.Apairofreactorswereplacedintotwothermal bathsinwhichtheexperimentaltemperaturewascontrolledbyPIDcontrollersactingontwo electricalresistancessitedinthethermal‐oilbaths.Thetemperatureoftheelectricalresistances werealsocontrolledtoavoidanyovertemperaturemal‐functionduringthetests.Oilwasalso pumpedfromthethermal‐oilbathsintotheexhaustgasheatexchangerstomaintainthewater balanceinthesystem.   Figure1.Schematicdiagramofthesemi‐batchaminedegradationapparatus  VolatileNH3determinationswereperformedinaccordancewithconditionaltestmethod27 (CM‐27)using0.1sulphuricacidasareagent.Theexhaustgasleavingthereactorswas E‐28 E‐26 %CO2 %N2 3%O2 %CO2 %N2 6%O2  50ºC E‐66 E‐65 E‐74 E‐77 E‐82 E‐80 E‐89E‐90E‐91 E‐97E‐98E‐99  70ºC E‐163 FI F FI F FI F FI F P P P P P‐151 C PSV C PSV 45 C 5 5 C 7 7 77 4 4 5 4 C 8888 6 2 22 2 111 1 33 6 1Pre‐saturatedimpinger 2Reactor 3Thermalbath 4Heatexchange 5Sulphuric acidvessed 6Peristalticpump 7Exitgas 8Liquidsamplingport C permoleofsolvent.AccordingtoDuetal.,CO2loadingsabove0.3moleCO2permolesolvent ledtolowerdegradationratesofPZandthisphenomenonwasalsoobservedinAEPasaPZ‐ derivedcompound[13].  Figure5.AEPdegradationpathwayfromDuetal.[13]andFreemanandRochelle[26,27] 3.2 Effectoftemperature Figures6aand6bsummarizetheaminelossinthesolventblendobtainedalongtheexperiments carriedoutat50˚Cand70˚Cusing6%v/vO2inthefluegas.At50˚C,mostofthereductionof AMPandAEPconcentrationwaslowerthan10%after30days,whileapreviousstudyshowed thatMEAlostupto23%underthesamevaluesofboththeO2concentrationandtemperature [11].Therefore,POS#1providedhighresistancetooxidativedegradationat50˚C,showinga highperformanceintermsofsolventdegradationthanMEA. At50°C,thelowestCO2concentration(15%v/v)resultedinaslightlyloweraminelossin comparisontothehigherCO2concentrationruns(Fig.6a).TheinhibitingeffectofCO2inthe bulkliquidwasreducedatlowtemperature.PeroxylformationfromAMPandnucleophile attacksbetweenAEPmoleculesmightbereducedat50˚C.Theinhibitingeffectassociatedto higherCO2loadedsolutionswerealsodecreased.Oppositeeffectswereobservedas temperatureshiftedto70˚C,asFig.6billustrates.Intheseruns,thedegradationofAMPand AEPwassignificantlyenhancedincomparisonwithlowtemperatureexperiments.Moreover,a cleartrendshowsthatatthesameoxygencontentinthefluegas,theincreaseoftheCO2 concentrationresultsinadecreaseofthesolventsdegradation(Fig.6aand6b).Asitcanbeseen inpreviousFig.3,theinhibitingeffectofCO2concentrationshouldstrengthenastemperature increases.HighertemperaturesshouldspeeduptheC‐Hcleavageduetovibratingenergy accumulation,producingmoreAMP‐derivedradicals,whichpromoteAMPdegradation[21,23]. AlthoughmoreAMP‐derivedradicalswerepresentinthebulkliquid,thehigherCO2loaded solutioninhibitedtheformationofperoxylradicals.Itshouldbepointedoutthattheinhibiting effectassociatedtothepresenceofCO2inthefluegasdecreasedastheCO2contentintheflue gasincreased,asitcanbeextrapolatedfromthetrendsobservedinFigures6aand6b.  (a)  (b) Figure6.Solventlosspercentageobtainedfromexperimentsrunusing6%v/vO2andat(a) 50°Cand(b)70°C  Theresultsprovidedfromthehightemperatureexperimentsfittedadequatelytoafirstorder kinetics.Thevaluesoftheapparentfirstorderconstant(k1)arereportedinTable4.Thelarge differencebetweenthekineticconstantdeterminedinourworkandthoseinliteratureisdue theoperatingtemperature.Thereferencesreportkineticconstantofthermaldegradation occurringattemperaturesover135˚Cwhereasthemaximumtemperaturestudiedinourwork was70˚C.Thistemperatureisthetemperaturebulgetypicallyreachedinaconventional absorberforCO2chemicalabsorption.Theincreaseoftemperaturefrom70˚Cto100˚C producedasubstantialincreaseonAEPdegradation,whereasAMPresultedinsimilarvaluesof k1forallthecases,showingahighresistancetothermaldegradation.Thesestatementsare consistentwithotherworksreportedinliterature,inwhichAEPpresentedlowresistanceto degradationattemperaturesover100˚C.Instead,AMPprovidedlowvaluesofk1at temperaturesupto140˚C,asitcanbeseeninTable4. ThedifferenceobservedbetweenthestudiescarriedoutbyFreemanetal.[33]andWangand Jens[34]mightbeexplainedduetothepresenceofMEA.IntheexperimentsrealizedbyWang andJens,theyusedaMEA/AMPblendwhereasFreemanetal.usedAMPinaqueoussolution. ThepresenceofMEAinaMEA/AMPblendpartiallyinhibitstheAMPdegradation.Thesame phenomenawasobservedinthiswork.Thek1valuesobtainedforAMPinAMP/AEPblendswere similartotheseprovidedinAMP/MEAblends[34]. Table4.Summaryoftheexperimentaltestsandkineticconstantscarriedoutinthiswork ReferenceTemp(˚C)[CO2](%v/v)AEPk1/10‐7s‐1AMPk1/10‐7s‐1 Thiswork70151.0420.579 Thiswork70200.8100.694 Thiswork70400.6940.347 Thiswork70600.4630.231 Thiswork100602.3150.926 [33]135NAN**‐ 2100 [33]150NAN**‐ 8600 [34]*100NAN**‐ 1.72 [34]*120NAN**‐ 3.1 [34]*140NAN**‐ 6.37 [13]150236500‐ [13]1752202200‐ *determinedinpresenceof2mMEA **inertN2  3.3 EffectofO2concentration TheO2contentinthefluegasiscriticalfortheoxidativedegradationofsolvents,inparticular forAMP,whichmainlydegradesfollowingtheperoxylradicalmechanism[21.23].Higher amountofO2inthebulkgas(6%v/v)enhancedtheoveralldegradationprocess.Asitcanbe seenintable5,theinitialconcentrationofAEPdecreasedfrom16.58%wtto15.24%wtafter30 daysunder15%v/vCO2and70˚C.AMPconcentrationvariedfrom14.13%wtto12.23%wtunder thesameexperimentalconditions.AccordingtoWangetal.,theuseofhigherO2partialpressure duringtheexperimentsacceleratedthedegradationratesobservedforAMP[21.23]. TheresistanceofAMPandAEPtobedegradedathighO2contentwasincreasedinthe experimentsoperatedunderpartialoxy‐combustionconditions.Thepresenceof6%v/vO2in thefluegasincreasedtheinhibitingeffectassociatedtohigherCO2concentration.AEPonly degradedupto17.70%wtusing6%v/vO2.60%v/vCO2and70˚C.Thiswasa16%increaseofthe AEPconcentrationafter30daysincomparison,tothepost‐combustionrun(15%v/v)underthe O2concentration(Table5).IntheAMPcase,theconcentrationvariedfrom13.93%wtto 12.23%wtafter30daysfrompartialoxy‐combustion(60%v/v)topost‐combustionconditions (15%v/v),respectively.Therefore,partialoxy‐combustion,whichresultsinhigherCO2 concentrationinthefluegas,shouldenhancethePOS#1degradationresistanceunderoxidative degradationincomparisontoconventionalpost‐combustionconditions.Theinhibitingeffect associatedtothehigherpresenceofCO2wasenhancedatelevatedO2concentrationsintheflue gas. Table5.Solventlossforpost‐combustioncase(15%v/vCO2)andpartialoxy‐combustioncase (60%v/vCO2)after30days TestConditionsAEP concentration after30days (%wt) AMP concentration after30days (%wt) AEP solvent loss(%) AMP solvent loss(%) O2(%v/v)CO2 (%v/v) Temperature (°C) 3 155019.4214.782.901.47 7016.5814.1317.105.80 60 5018.3613.838.207.80 7018.0214.599.902.73 10018.0514.119.755.93 6 155018.4714.067.656.27 7015.2412.2323.8018.47 60 5018.2114.178.955.53 7017.7013.9311.507.13 1009.9911.9850.0520.13  3.4 NH3emissionsanddegradationproducts TheNH3emissionsderivedfromPOS#1degradationweredeterminedatfourtimeintervals duringtheexperiments.Figure7summarizestheaverageNH3emissionmeasuredforeach intervalforalltheexperimentscarriedoutinthiswork.  (a)  (b)  (c)  (d) Figure7.EvolutionoftheaverageNH3emissionsmeasuredinthelab‐rigexhaustgas:(a) 15%v/vCO2.(b)20%v/vCO2.(c)40%v/vCO2and(d)60%v/vCO2.Theexperimentsplottedin thosefiguresweresetat3%v/vO2and50°C(bluebars).6%v/vO2and50°C(greenbars).3%v/v O2and70°C(redbars)and6%v/vO2and70°C(yellowbars)  NH3emissionsweredramaticallyreducedfromdegradedPOS#1incomparisonwithMEA.NH3 emissionsover40ppmweremeasuredduringexperimentscarriedoutusingMEAasasolvent thatwerefourtimethosedeterminedforPOS#1worstcase.POS#1producedlowerNH3 emissionsincomparisonwithMEAunderalltheoperatingconditionsstudiedinthiswork[11]. ItshouldbepointedoutthatNH3emissionswerefurtherenhancedatelevatedtemperature andO2content,whenitdoubledundertheseconditions.Emissionsfromexperimentsrunat50 ˚Cwerebelow3ppmasitisshown(inblueandgreenbars)inFigures7a,b,candd.Similarlow NH3emissionsweredeterminedinotherexperimentswithaqueousAMPconductedat55˚C [36].TheinhibitingeffectofhigherCO2concentrationinthefluegasareclearlyobservedinFig. 7aand7d.TheNH3emissionsdecreasedinalltheexperimentsoperatedunderpartialoxycombustionconditions.A70%reductionofNH3emissionswereachievedinthe60%v/vCO2 cases,incomparisontotheconventionalpost.‐combustioncase.Therefore,theresultsobtained fromNH3emissionmeasurementsconfirmthatamoreCO2concentratedfluegasreducesthe oxidativedegradationofPOS#1–andhenceAMPandAEP‐. ThesolventlossarerepresentedasfunctionofthetotalNH3emissionsobtainedfromeach experimentcarriedoutinthiswork(Fig.8).AsitoccursinMEAexperiments,thedegradationof AMPandAEParecloselyrelatedtotheamountofNH3emittedandtheirmeasurementcanbe usedformonitoringPOS#1degradationinCO2capturebychemicalabsorption[11].The dependenceofthesolventlosswiththetotalNH3productionduringtheexperimentsshoweda linearapproximationwithR2valuesrangingbetween0.75and0.85.  Figure8.AEPandAMPlossplottedversusthetotalamountofNH3produced.  Asexpected,themostrelevantdegradationcompoundsidentifiedintheexperimentswere identicaltothosereferencedintheliterature(tables1and2).TheCO2contentinthefluegas hadnoinfluenceonthemaindegradationpathwayssuggestedinpreviousstudies.Formatewas themajordegradationcompoundidentifiedintheliquidsampleswithdrawnduringthe experiments.AsitoccurredtoNH3formation,higheramountsofformatewereformedasO2 concentrationandtemperaturewereincreased,buthigherCO2concentrationsdecreasedits formationmainlyduetotheinhibitingeffectrelatedtoO2mass‐transferintothebulkliquid.In respecttoAMPdegradationcompounds.2.4‐Lutidinewasthemajordegradationproduct, particularlyatelevatedtemperatures.TheinhibitingeffectofCO2wasalsoobservedin2.4‐ Lutidineformation.Othercompoundssuchasacetoneandoxazolidinoneswerealsoidentified. Inparticular.4.4‐dimethyl‐2‐oxazolidine(DMOZD)formationwasincreasedunderhigherCO2 concentrations,probablyduetoringenclosuremechanismsenhancement[21.23]. EthylenediamineandoxopiperazineswerethemajordegradationcompoundderivedfromAEP degradationandtheirproductionwasenhancedwithincreasedtemperatureandCO2 concentration.OthercompoundssuchasPZ,N,N‐dis(2‐aminoethyl)piperazine(DAEP)andpoly‐ AEPswereidentifiedmainlyinsamplesat100˚C.Theconcentrationofthesedegradation compoundswerefoundthedetectionlimitoftheanalyticalmethodusedinthiswork,and therefore,theywerenotdetectedinsampleswithdrawnfromexperimentsconductedat50˚C. Ingeneral.POS#1showedahighresistancetotheoxidativedegradationinbothpostcombustionandpartialoxy‐combustionconditions.AlthoughPOS#1degradedlessin comparisonwithMEAinpost‐combustiontests,theuseofoxy‐combustionconditionsenhanced thePOS#1degradationresistanceandreducedby50%thedegradationratesobservedinthe post‐combustionruns.Thevolatilecompoundsemissionswerealsodramaticallydecreased undertheseconditions,andtherefore.thecostassociatedtosolventmake‐upandcleangas treatmentshouldbefurtherdecreasedusingPOS#1and60%v/vCO2fluegas. 4. Conclusions Theevaluationofthedegradationbehaviourofanovelsolventblendhasbeenperformedunder partialoxy‐combustionconditionsinasemi‐batchreactorfor30days.Thenovelsolvent,namely POS#1,consistsofanAMP/AEPblend.POS#1showedhighresistancetooxidativedegradation atabsorberconditions.AMPandAEPlosswere19%and24%,respectively,almosthalf comparedto38%reductionforMEAatconventionalpost‐combustionconditions.Amongthe twosolvents.AMPshowedahighperformanceprovidingloweraminelossthanAEP,in particularduringhightemperatureexperiments.ExperimentsconductedathigherCO2 concentrationdemonstratedthatpartialoxy‐combustionreducestheoxidativedegradationof POS#1.TheuseofamoreCO2concentratedfluegaspartiallyinhibitstheoxidativedegradation oftheAMP/AEPblend,reducinghalfthedegradationratesobservedunderpost‐combustions conditions.TheincreaseoftheCO2partialpressurecausestheCO2loadingofthesolventblend reacheshighervalues.ThepresenceofmoreAMPcarbamateintothebulkliquiddecreasedthe formationofperoxylradicalfromfreenon‐loadedAMP.InrespecttoAEP,higherCO2inbulkgas producesmoreCO2loadedsolutionsandlimitedthepresenceofprotonatedAEPinthesolution.