semivolatiledieselexhaustcomponentsGutzwilleretal2002canbeneglectedKleffmannetal2002InadditionsamplingartefactssuchasheterogeneousHONOformationinsamplinglinesasanexampleseeZhouetal2002b ID: 522122
Download Pdf The PPT/PDF document "F.Rohreretal.:Characterisationofthephoto..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
F.Rohreretal.:CharacterisationofthephotolyticHONO-source2191 semi-volatiledieselexhaustcomponents(Gutzwilleretal.,2002)canbeneglected(Kleffmannetal.,2002).Inaddition,samplingartefacts,suchasheterogeneousHONOformationinsamplinglines(asanexampleseeZhouetal.,2002b),areminimisedbytheuseofanexternalsamplingunitinwhichthetwostrippingcoilsaremountedandwhichcanbeplaceddirectlyintheatmosphereofinterest.Inarecentin-tercomparisoncampaignwithaDOASinstrumentinwhichthesameairmasseswereanalysedforthersttime(Kleff-mannetal.,2005 2 )anexcellentagreementwasobtainedalsoforlowconcentrationsofHONOduringdaytime.Thisisincontrasttointercomparisonstudiesofotherchemicaldetec-torswiththeDOAStechnique(Appeletal.,1990;Coeetal.,1997;Feboetal.,1996;M¨ulleretal.,1999),whichshowlargediscrepanciesduringthedayprobablycausedbyun-knowninterferences.Themeasuredandcorrectedinterfer-encesoftheLOPAPinstrumentcanaccountforupto40%atlowHONOconcentrations(Kleffmannetal.,20052).FortherstmeasurementstheexternalsamplingunitoftheLOPAPinstrumentwasdirectlyinstalledintheSAPHIRchamber.Formostoftheexperimentshowever,theinletoftheexternalsamplingunitwasconnectedtothenortherncor-nerofthechamberbyashortPFAtubing(10cm,4mmi.d.).Incontrasttocomplexatmosphericmixtures(Kleffmannetal.,2002),samplingartefactswerefoundtobesmallforthelowtracegasandrelativehighHONOconcentrationsduringtheexperimentsintheSAPHIRchamber.NOandNO2measurementswereperformedwithachemi-luminescenceanalyser(ECOPHYSICSTR480)equippedwithaphotolyticconverter(ECOPHYSICSPLC760).TheNOxdatawereanalysedasdescribedbyRohreretal.(1998).Foratimeresolutionof90sthedetectionlimitswere5and10pptVandtheaccuracies5and10%forNOandNO2,re-spectively.SinceHONOmixingratioswereoftenhigherthanNO2mixingratiosinthechamber,theHONOinter-ferenceofthephotolyticconversionsystemforNO2isanimportantfactor(forexampleseeFig.1).Thisinterferenceof15%hasbeendeterminedexperimentallybyusingapureHONOsourcesimilartothatdescribedbyTairaandKanda(1990)quantiedbytheLOPAPinstrument.Ozonewasmeasuredwithatimeresolutionof90sbyanUVabsorp-tionspectrometer(ANSYCOmodelO341M).Thedetectionlimitandaccuracywas0.5ppbVand5%,respectively.Pho-tolysisfrequenciesweredeterminedwithaspectroradiome-ter(asdescribedbyHofzumahausetal.,1999)ontheroofofthebuildingbesidetheSAPHIRchamberwithatimereso-lutionof2min.Toaccountforshadowingeffectsofstruc-turalelementsofthechamber,thephotolysisfrequencydatawerecorrectedwiththehelpofnumericalcalculations(BohnandZilken,2005;Bohnetal.,2005),leadingtoanaccuracy 2Kleffmann,J.,L¨orzer,J.C.,Wiesen,P.,Trick,S.,Volkamer,R.,Rodenas,M.,andWirtz,K.:IntercomparisonsoftheDOASandLOPAPTechniquesfortheDetectionofNitrousAcid(HONO)intheAtmosphere,manuscriptinpreparation,2005. Fig.1.HONO,NOx,andozoneformationinSAPHIRilluminatedwithsunlightonAugust08,2001(experimenttypeB,seeTable1).Inthetimeintervalbetween07:00and08:30UT,smallamountsofHONOandNO2wereushedintothechamberduringthehumid-icationprocess.Bluesymbolsmarkobservations,redandgreenlinesshowtheresultofmodelcalculations(seetext).Theshadedareasmarkthetimeperiodswerethechamberwasnotilluminated. of10%.Humiditywasdeterminedwithafrostpointhy-grometer(GeneralEasternmodelHygroM4)andairtemper-aturebyanultrasonicanemometer(MetekUSA-1,accuracy0.3K).Severalcrosschecksoftheinstrumentationwereper-formedtoexaminetheconsistencyofthedatasets.ThepureHONOsourcewasanalysedbytheLOPAPinstrumentandbytheNOxinstrumentationbyusingacatalyticconversionsystemforNOyasdescribedbyFaheyetal.(1985).Bothin-strumentsshowedagreementwithin5%.Asecondtestwasperformedbyintroducing50ppbVofNO2intothechamberwithclosedshutters.Afterexposuretosunlightandphotol-ysisofNO2yieldingequalamountsofNOandozone,theNOandozoneinstrumentsshowedconsistencywithin2%.Moreover,thephotostationarystateofNO,ozone,andNO2reachedafterexposurewithlightwasconsistentwithin10%withthemeasuredphotolysisfrequencyJ(NO2)andtherec-ommendedliteraturevaluefortherateconstantofthereac-tionofNOwithozone(Sanderetal.,2003). www.atmos-chem-phys.org/acp/5/2189/Atmos.Chem.Phys.,5,2189– 2201 ,2005 F.Rohreretal.:CharacterisationofthephotolyticHONO-source2193 Table1.OverviewaboutthedifferenttypesofHONOcharacterisationexperiments. ExperimentTypeRelativeHumidityCOInitialNO2FilterfoilNumberofanalysedexperiments A1%00–11B&#x]TJ/;༶ ;.97;&#x Tf ; 0 ;&#xTd[0;10%00–9C&#x]TJ/;༶ ;.97;&#x Tf ; 0 ;&#xTd[0;10%550ppmV0–2D&#x]TJ/;༶ ;.97;&#x Tf ; 0 ;&#xTd[0;10%550ppmV0+2E30%035ppbV–1 Table2.DetailedexperimentalconditionsforHONOcharacterisationexperiments.HONOconcentrationsaretakenfrommodelcalculationssincenotallexperimentsareaccompaniedbyLOPAPHONOmeasurementsduetolimitedavailability. ExperimentTminTmaxRHmaxHONOmaxS(HONO)maxJ(NO2/maxdate typeCC%ppbppb/h10�3s�1 250701 A30.034.010.0250.1049.7300701 A30.034.010.0180.0779.6020801 A28.032.010.0140.0519.6070801 A18.022.010.0170.07510.2080801 B18.721.389.90.7252.6247.4040901 A31.834.60.40.0110.0447.6020202 B12.419.982.50.3370.7043.3270302 A12.016.010.0120.0336.5040202 E10.012.236.70.3200.1563.6280302 A12.016.010.0060.0166.8010702 C15.816.967.20.4020.6652.5020702 D13.120.276.90.5320.3912.3040702 B15.625.672.70.3790.9667.0050702 B22.627.260.70.5252.2256.6080702 D21.937.860.60.9132.6482.2090702 B23.232.157.90.4260.7265.3130802 A22.028.010.0270.1343.8300902 B13.723.841.60.3491.0394.5260503 A17.027.00.60.1100.2107.9270503 B19.929.248.20.4701.6686.1070603 A21.232.90.70.0800.2005.9120603 A18.332.60.90.0750.1846.2110703 B22.033.235.10.2240.7206.4190903 B22.133.246.60.5421.4564.2220903 C23.736.127.90.1590.4134.4 thespectralrangeoftheradiationinsidethechamberwasvaried.ForthispurposeHONOformationwasstudiedunderhumidconditionswhenlightatshortwavelengths(370nm)wasabsorbedoutsidetheTeonchamberbytheuseofthelterfoil.ThephotolysisfrequencyofNO2,J(NO2),de-creasedbyafactorof3bythelterfoil,J(HONO)byafactorof10,andthephotolysisfrequencyofozone,J(O1D),decreasedbyafactorof100.Nevertheless,whenthelterfoilwasused,stillasignicantHONOformationwasob-servedunderirradiation(seeFig.2).Butthereisasignif-icantdifferencebetweentheexperimentswithandwithoutthelterfoilinFigs.1and2.Sinceboth,theproductionofHONO(whichisproportionaltoJ(NO2))andthedestruc-tionofHONOviaJ(HONO)arereducedbythelterfoil,theslopeoftheHONOincreaseisverymuchreducedintheex-perimentshowninFig.2althoughthemaximumHONOcon-centrationsaresimilarinbothexperiments.Themaindiffer-enceisthereforethemaximumNOxconcentrationwhichisverylowfortheexperimentinFig.2withreducedJ(HONO).4.2ModelcalculationsInordertoquantifytheHONOproductionrateS(HONO)SAPHIRundervariousconditionsmodel www.atmos-chem-phys.org/acp/5/2189/Atmos.Chem.Phys.,5,2189– 2201 ,2005 2194F.Rohreretal.:CharacterisationofthephotolyticHONO-source Table3.Reactionschemeusedforthemodelcalculationwithre-actionconstantstakenfromSanderetal.(2003). O1D+O2�!O3NO+O3�!NO2O1D+N2�!O3NO+OH�!HNO2O1D+H2O�!2OHNO2+NO3�!N2O52HO2�!H2O2NO2+O3�!NO32HO2+H2O�!H2O2NO2+OH�!HNO3CO+OH�!CO2+HO2NO3+OH�!HO2+NO2H2+OH�!HO2O3+OH�!HO2H2O2+OH�!H2O+HO2HNO2+h�!NO+OHHCHO+OH�!CO+HO2H2O2+h�!2OHHNO2+OH�!H2O+NO2HCHO+h�!CO+2HO2HNO3+OH�!H2O+NO3HCHO+h�!CO+H2HO2+NO�!NO2+OHNO2+h�!NO+O3HO2+NO3�!NO2+OHNO3+h�!NOHO2+O3�!OHNO3+h�!NO2+O3HO2+OH�!H2OO3+h�!O1DN2O5�!NO2+NO3 calculationswereperformedusingthephotochemicalreactionschemeoutlinedinTable3describingasimpleNOx/HONO/CO/HCHOchemistrywithreactionconstantstakenfromSanderetal.(2003).Inaddition,Reactions(R3)(seeintroduction),(R4),(R5)and(R6)wereintroducedintothemodeltoaccountforseveralphenomenaobservedinSAPHIR: OHCX�!HO2k=k.COCOH/ (R4) alltracers�!k=ow=volume (R5) Y.HCHO/Ch�!HCHOrateV0�0:2ppb=h (R6) Reaction(R3)describestheobservationofphotolyticallyin-ducedHONOformation.Inaddition,tosimulatethesmallincreasesofNO2andHONOduringsomehumidicationprocessesinthedarkchamber,appropriateamountsofNO2andHONOwereintroducedintothemodelcalculationsalso(seeFig.1).ReactionR4accountsforthephenomenonthatevenwithverycleanstartingconditionswhenallmeasured Table4.ResultofttingtheobservedtimeseriesofNOandNO2fortheexperimentsmentionedinTables1and2tomodelcalcula-tionsusingtheparameterisation(1). TimeperiodaiRH0T0 07/2001–07/20024.71013cm�311.6%3950K08/2002–12/20038.51013cm�311.6%3950K NMHCsarebelowtheirdetectionlimitimmediateozonefor-mationisobservedwhenthechamberisilluminated.AllmeasuredspecieslikeCOandNMHCsarebelowtheirde-tectionlimitsatthebeginningoftheexperiments.ForthisreasonanunknownspeciesXwasintroducedinthechemi-calmechanismwhichcanreactwithOHgivingHO2.Sub-sequentreactionofHO2withNOgivesthedesiredozoneformation.Tofacilitatecomparisons,thereactionconstantofCOwithOHfromSanderetal.(2003)wasusedasapa-rameterforReaction(R4).Reactionssimilarto(R5)wereintroducedforalltracespeciesinthereactionmechanismtodescribedilutionbythereplenishmentowasrstorderlossreactions.Thereactionconstantwascalculatedfromthevol-umeofthechamberandthemeasuredreplenishmentow.Reaction(R6)isnecessarytofollowtheobservedHCHOproductionwhenthechamberisilluminated.TheHCHOfor-mationwaslinearwithtimeanddependingonhumidity,lightintensityandtemperature.TherateofHCHOformationusedinthemodelwasadjustedtothemeasuredrate.However,thisreactionhasonlymarginalinuenceontheHOxbudgetinthechamberandisgivenhereonlyforcompleteness.ThecharacterisationoftheHCHOsourceintheSAPHIRcham-berwillbedescribedinaforthcomingpaper.SubsequentmodelcalculationsshowedthatthefollowingempiricallyderivedparameterisationfortheproductionrateofHONOdescribestheobservedtimeseriesofNOandNO2: S.HONO/SAPHIR;iDaiJ.NO2/.1C.RH=RH0/2/e�T0=T Tcm�3s�1Ui=1;2 (1) TrepresentsthetemperatureinK,RHtherelativehumidityin%,J(NO2)thephotolysisfrequencyofNO2ins�1andai,RH0andT0arettingparametersspeciedinTable4.Equa-tion(1)isanempiricalfunctionwithonlythreeparameterswhichisabletodescribetheHONOformationinSAPHIRwithgoodprecisionforabroadbandofboundaryconditions.ThisisasignicantstepforwardintheprocessofusingthesimulationchamberSAPHIRasatoolforinvestigationofatmosphericchemistry.Equation(1)isnotbasedonaphys-icalmodeloftheprocessescontrollingHONOformationinSAPHIR.TondtheoptimumsetofparametersforEq.(1),humid-ity,temperature,andJ(NO2)wereusedasobserved.TheconcentrationsoftheunknownspeciesXforReaction(R4) Atmos.Chem.Phys.,5,2189– 2201 ,2005www.atmos-chem-phys.org/acp/5/2189/ 2196F.Rohreretal.:CharacterisationofthephotolyticHONO-source 5DiscussionInotherstudies,aphotoenhancedbackgroundreactivitywasproposedtoexplainelevatedreactivityinsimulationcham-bersunderirradiation(Akimotoetal.,1987;GlassonandDunker,1989;KillusandWhitten,1990;SakamakiandAki-moto,1988;Wangetal.,2000).However,incontrasttoallknownstudiesofthebackgroundreactivityofsimulationchambers,HONOwasunequivocallyidentiedunderillu-minatedconditionsinthepresentstudyforthersttime.WeobservedthattheadditionofCOhadnoeffectontheHONOproduction.SincetheadditionofCOdecreasedtheOHradicalconcentrationbyatleastthreeordersofmag-nitude,formationofHONObythegasphasereactionofNOandOH(seeTable2)isnotofimportanceunderourexperimentalconditions.ThephotolyticHONOsourceinSAPHIRwasfoundtobeproportionaltothephotolysisfre-quencyofNO2,whichisingoodagreementwithparame-terisationsofthebackgroundreactivitymadeinthestudyofWangetal.(2000).Inaddition,thephotolyticHONOsourceincreasedwiththesquareofrelativehumidityandexponen-tiallywithtemperature.Awaterdependenceoftheback-groundreactivitywasalsoobservedinmostotherstudies(Akimotoetal.,1987;KillusandWhitten,1990;SakamakiandAkimoto,1988).Theexcellentagreementbetweenex-perimentalresultsandmodelcalculationclearlyshowsthatthephotolyticHONOsourceisthedominantNOxsourceinthechamber,ingoodagreementwithsuggestionsofKillusandWhitten(1990).Incontrast,adirectphotolyticNOxsourcewhichhasrecentlybeproposedforsnow(Davisetal.,2001;Honrathetal.,1999,2000;Jonesetal.,2000,2001)andglasssurfaces(Zhouetal.,2002b,2003)bythephotol-ysisofnitratecanbeexcluded,sincethemodelledHONOconcentrationwouldbesignicantlyhigherthanthemea-suredoneinthiscase.Intwoolderstudies(Akimotoetal.,1987;SakamakiandAkimoto,1988),theelevatedreactivityinsimulationcham-bersunderirradiationwasexplainedbyaphotoenhance-mentofReaction(2),2NO2+H2O,sincetheradicalsourcestrengthincreasedwithincreasinghumidity,radiationandNO2concentration.Inaddition,theinvolvementofNO2inthephotoenhancedHONOformationwasveryrecentlyob-servedonorganicsubstrates(Georgeetal.,2005).However,aphotoenhancementofaNO2reactionisconsideredunlikelyforSAPHIRbasedontheresultsofthepresentstudy,sincethemajorityofexperimentsstartedwithverylowNO2con-centrationsontheorderof20pptVorless.Duringthecourseoftheseexperiments,NO2increasedtoseveral100pptVwithoutinuenceonthephotolyticHONOformation(seeFigs.1and2forexample).Inadedicatedexperiment(ex-perimenttypeE,Table1),35ppbVofNO2at30%relativehumiditywasusedasstartingconditionsagainshowingnoenhancementonthephotolyticHONOproduction.SinceamonolayeradsorptionofNO2ontheTeonsurfacecannotbeexpectedforafewpptVofNO2inthegasphase,anin-creaseoftheNO2concentrationbythreeordersofmagnitudeshouldclearlyhaveaninuenceonanyNO2surfacereac-tion.Sincethiswasnotobserved,theseexperimentsshowedthataphotoenhancementofthereactionofNO2withwatervapour(ReactionR2)orofNO2withadsorbedorganiccom-poundsarenotofimportanceinSAPHIR.Inaddition,ad-sorptionofHONOontheTeonsurface,formedduringpriorexperimentsbyanyNO2reactions,canalsobeexcluded.Inthiscase,HONOshoulddesorbfromthewallswhenthehu-midityisincreasedinthedark,asobservedinothercham-bers(e.g.SyominandFinlayson-Pitts,2003),incontrasttotheSAPHIRchamber(seeFig.2).InthestudyofKillusandWhitten(1990)thephotoen-hancementofthebackgroundreactivityinTeonchamberswasexplainedbythephotolysisofnitrate,sinceelevatedre-activitywasobservedafterexperimentsinwhichhighnitricacidconcentrationswereused.ThephotolysisofnitrateasasourceofHONOwasrecentlyalsoproposedintheatmo-sphereoversnow(Beineetal.,2001,2002;Dibbetal.,2002;Honrathetal.,2002;Zhouetal.,2001),groundandvegeta-tionsurfaces(Zhouetal.,2002a),toexplainhighday-timeconcentrationsofHONO.Inaddition,aphotolyticHONOsourcebyphotolysisofnitratewasproposedforglasssur-faces(Zhouetal.,2002b,2003).However,basedontheresultsfromthepresentstudy,thephotolysisofnitratecanbeexcluded.AsignicantHONOformationwasalsoob-servedinSAPHIRwhenlightwithwavelengths370nmwasblockedbythelterfoil.Sincetheweakabsorptionbandofnitrateat300nm("D7lmol�1cm�1,MeyersteinandTreinin,1961)wasfoundtoberesponsiblefornitriteandNO2formationinsolution(Wagneretal.,1980),andsincenitrateabsorptiondoesnotextendtowavelengths&#x]TJ/;༶ ; .96;&#x Tf ;.77;&#x 0 T; [00;370nm,nosignicantphotolyticHONOformationwouldhavebeenexpectedintheexperimentswiththelterfoil.InthepresentstudyagoodcorrelationofthephotolyticHONOformationwiththephotolysisfrequencyofNO2wasfound.Thiscanbedemonstratedbytheexcellentagreementbetweenmea-surementsandmodelcalculationusingEq.(1)fortheexper-imentswithandwithoutthelterfoil(seee.g.Figs.1and2).Fromthewavelengthdependenceofthestudiedphotoen-hancedHONOformationintheSAPHIRchamberandfromthetransmissioncharacteristicsofthelterfoil,photochem-icalprocessesatwavelengthsexclusively370nmandex-clusively&#x]TJ/;༶ ; .96;&#x Tf ;.77;&#x 0 T; [00;420nmcanbeexcluded,sinceHONOformationshouldhavebeenreducedbytwoordersofmagnitudeintherstcaseandonlyby15%inthesecondcase,causedbythetransmissionofthelterfoil.InarecentstudyofSalibaetal.(2001)itwasshownbyinfraredspectroscopythatadsorbednitricacidshouldbeal-mostundissociatedonsurfacesuptoawatercoverageofthesurfaceofthreeformalmonolayers.However,photol-ysisofadsorbedHNO3canonlyexplaintheexperimentalobservations,iftherelativeshapeoftheUVabsorptionspec-traofundissociatedadsorbedHNO3issignicantlydiffer-enttothespectraofundissociatedgaseousHNO3(Sanderet Atmos.Chem.Phys.,5,2189– 2201 ,2005www.atmos-chem-phys.org/acp/5/2189/ F.Rohreretal.:CharacterisationofthephotolyticHONO-source2197 al.,2003).TheUVabsorptioncrosssectionofHNO3/H2OlmsonAl2O3surfaceswasrecentlymeasuredbyBerlandetal.(1996).Unfortunately,onlythe!bandat200nmwasinvestigated.Fortheoscillatorstrengthofthebandnosignicantdifferencebetweenthinlmsofnitricacidandgaseousnitricacidwasobserved.Inaddition,thebandwasshiftedbyonly10nmtolongerwavelengthforthethinlm.Accordingly,itcanbeconcludedthatadsorbedundissociatedHNO3hasasimilarUVabsorptionspectrumcomparedtogaseousnitricacid.AsanupperlimitofthelongwavelengthphotolysisofadsorbedHNO3inthespec-tralrangeoftheSAPHIRchamber,itwasassumedthattheabsorptioncrosssectionofadsorbedHNO3issimilartothatofgaseousHNO3givenbySanderetal.(2003)andre-mainsconstantinthewavelengthrange350–420nmwiththelowestvaluegivenforgaseousHNO3for350nm.Underthisassumption,HONOformationbyphotolysisofadsorbedHNO3shouldbereducedbyafactorof�50bytheuseofthelterfoil.However,onlyareductionofafactorof3wasobservedintheexperiments.Inaddition,byusingtheob-viouslyoverestimatedabsorptioncrosssectionsmentionedabove,aHNO3adsorptionof10monolayers,whichisanunrealistichighvaluefortheSAPHIRchamberandaquan-tumyieldof1forHONOformationbyHNO3photolysis,whichistwoordersofmagnitudehigherthantheeffectivequantumyieldfornitriteformationbynitratephotolysisinsolutionatpH4–7(Marketal.,1996),therateofHONOformationinthechamberwouldbestillmorethanoneor-derofmagnitudelowerthanthemeasuredoneintheexperi-mentswiththelterfoil.Anotherargumentagainstthepho-tolysisofadsorbedundissociatedHNO3istheobservation,thatphotolyticHONOformationstillincreasedforrelativehumiditiesof�50%.Forthehighestrelativehumiditiesof80%itcanbeexpected,thatHNO3willdissociatetoni-trate(Salibaetal.,2001;SvenssonandLjungstr¨om,1987),whichwasexcludedasaprecursorofHONOinthecham-ber(seeabove).Accordingly,itisproposedthatadsorbednitricaciddoesnotrepresenttheprecursorofHONOformedduringtheirradiationofthechamber,althoughthiscannotbecompletelyexcluded,sincetheUVabsorptionspectraofadsorbednitricacidonTeonsurfacesisunknown.InthestudyofKillusandWhitten(1990)elevatedback-groundreactivitywasdocumentedforTeonchambersafterexperimentswithhighnitricacidconcentrations.BasedontheseresultsitmightbepossiblethattheprecursorofHONOisformedbyareactionofHNO3withunknowncompoundsinthebrestructureoftheTeonfoil.Candidatesforthesecompoundsmightbehighermolecularorganicsfromthepro-ductionoftheTeonmaterial,whichareoxidisedbyHNO3.Thus,thehypotheticalprecursursmightbeorganicnitrates,nitrites,nitroaromatics,pernitricacidsetc.,formedintheTeonfoil.FromtheexperimentswiththelterfoilitcanbeconcludedthattheprecursorofHONOshouldphotolyseinasimilarwavelengthrangeasNO2.IncontrastitcannotphotolyseinthespectralrangeinwhichnitrateorgaseousHNO3absorb.ThewaterdependenceofthephotolyticHONOsourcewasdescribedbyaquadraticfunction(seeEq.1),whichresem-blesthedependenceofwateruptakeonFEPTeonshowninFig.7ofSvenssonetal.(1987).Basedontheobservedhu-miditydependenceofthephotolyticHONOsourceitcanbespeculatedthateitheronlythedissociatedformoftheprecur-sorphotolysesorthatHONOisformedbyareactionofpho-tolysisproductswithadsorbedwater.OnecandidatemightbeahypotheticalfastreactionofanexcitedNO2molecule,formedinthephotolysisoftheunknownprecursor,withwa-ter.Incontrast,thereactionofgroundstateNO2moleculeswithadsorbedwatercannotexplaintheexperimentalnd-ings,sincenoenhancementofthephotolyticHONOfor-mationwasobservedwithincreasingNO2concentration.SinceitiswellknownthatmoleculesdiffusethroughTeonmaterial,itcanbeexpectedthatHNO3willreactwiththepostulatedorganiccompoundsalsoontheinternalsurfaceoftheTeonfoil.SinceHNO3isformedastheendproductofNOxinmostexperimentsandsinceefcientdepositionofHNO3onthewallcanbeexpected,theconcentrationoftheprecursorandaccordinglythephotolyticHONOsourcewillprobablynotdecreasesignicantly,asobservedoveraperiodofmorethantwoyearsinthechamber.However,thisar-gumentholdsonly,iftheconcentrationofthespeculatedor-ganiccompound,whichformstheprecursorofHONObyre-actionwithHNO3,ishighenough.However,toestimatethelossfractionfromareservoirofaHONOproducingspeciesintheTeonlmwecalculateatypicalHONOproductionof71019molecules(1.7mgN)inone5hexperiment(50%relativehumidity,J(NO2)=510�3s�1/.IfthesourcewereinthebulkoftheFEPwallmaterialthelossofnitrogencor-respondstoonly0.02ppmmassfractionoftheinnerFEPwall(approx.80kg).Inconclusion,althoughseveralpathwaysofthephotolyticHONOformationdiscussedintheliteraturecouldbeex-cluded,e.g.aphotoenhancementofthereactionofNO2andH2Ooraphotolysisofnitrate,theprecursorofHONOformedphotolyticallyintheSAPHIRchamberwasnotiden-tied.Accordinglyfurtherworkisneededtoclarifythispro-cesswhichisofparamountimportancefortheradicalbal-anceofsimulationchambers.ItwasshownthatHONOproductioninthesimulationchamberSAPHIRcanbepredictedwithgoodprecisionforlongertimeintervals.AnunexplainedchangeintheHONOproductionwasobservedinAugust2002,i.e.achangeoftheparametera1inEq.(1)byafactorof1.8.Therefore,weconcludethatthechamberrelated(photolytic)HONOsourcemustbequantiedinregularintervalstoconrmitsstability.SinceEq.(1)isanempiricalparameterisationoftheHONOsourceintheSAPHIRchamber,wedonotadvisetheappli-cationofourparametersinotherchambers.However,Eq.(1)canbeusedtocompareHONOformationratesmeasuredinothersimulationchambersbyappropriatelyscalingit,takingintoaccounttheSAPHIRS/Vratioof1m�1.Whencompar- www.atmos-chem-phys.org/acp/5/2189/Atmos.Chem.Phys.,5,2189– 2201 ,2005 F.Rohreretal.:CharacterisationofthephotolyticHONO-source2201 Sander,S.P.,Friedl,R.R.,Golden,D.M.,Kurylo,M.J.,Huie,R.E.,Orkin,V.L.,Moortgat,G.K.,Ravishankara,A.R.,Kolb,C.E.,Molina,M.J.,andFinlayson-Pitts,B.J.:ChemicalKineticsandPhotochemicalDataforUseinStratosphericModeling,JPLPublication02-25,Pasadena,California,2003. Svensson,R.,Ljungstr¨om,E.,andLindqvist,O.:KineticsoftheReactionbetweenNitrogenDioxideandWaterVapour,Atmos.Environ.,21,1529–1539,1987. Syomin,D.A.andFinlayson-Pitts,B.J.:HONODecompositiononBorosilicateGlassSurfaces:ImplicationsforEnvironmen-talChamberStudiesandFieldExperiments,Phys.Chem.Chem.Phys.,5(advancedarticle),doi:10.1039/b309851f,2003. Taira,M.andKanda,Y.:ContinousGenerationSystemforLow-ConcentrationGaseousNitrousAcid,Anal.Chem.,62,630–633,1990. Vogel,B.,Vogel,H.,Kleffmann,J.,andKurtenbach,R.:MeasuredandSimulatedVerticalProlesofNitrousAcid,PartII–ModelSimulationsandIndicationsforaPhotolyticSource,Atmos.En-viron,37,2957–2966,2003. Wagner,I.,Strehlow,H.,andBusse,G.:FlashPhotolysisofNitrateIonsinAqueousSolutions,Z.Phys.Chem.,123,1–33,1980. Wang,L.,Milford,J.B.,andCarter,W.P.L.:ReactivityEsti-matesforAromaticCompounds.Part1.UncertaintyinChamber-DerivedParameters,Atmos.Environ.,34,4337–4348,2000. Zhou,X.,Beine,H.J.,Honrath,R.E.,Fuentes,J.D.,Simpson,W.,Shepson,P.B.,andBottenheim,J.W.:SnowpackPhotochem-icalProductionofHONO:aMajorSourceofOHintheArcticBoundaryLayerinSpringtime,Geophys.Res.Lett.,28,4087–4090,2001. Zhou,X.,Civerolo,K.,Dai,H.,Huang,G.,Schwab,J.,andDe-merjian,K.:SummertimeNitrousAcidChemistryintheAt-mosphericBoundaryLayerataRuralSiteinNewYorkState,J.Geophys.Res.,107(D21),4590,doi:10.1029/2001JD001539,2002a. Zhou,X.,He,Y.,Huang,G.,Thornberry,T.D.,Carroll,M.A.,andBertman,S.B.:PhotochemicalProductionofNitrousAcidonGlassSampleManifoldSurface,Geophys.Res.Lett.,29(14),10.1029/2002GL015080,2002b. Zhou,X.,Gao,H.,He,Y.,Huang,G.,Bertman,S.,Civerolo,K.,andSchwab,J.:NitricAcidPhotolysisonSurfacesinLow-NOxEnvironments:SignicantAtmosphericImplications,Geophys.Res.Lett.,30(23),2217,doi:10.1029/2003GL018620,2003. www.atmos-chem-phys.org/acp/5/2189/Atmos.Chem.Phys.,5,2189– 2201 ,2005