Volatilizable Biogenic Organic Compounds PDF
Preview Volatilizable Biogenic Organic Compounds
Atmos. Meas. Tech.,5,345–361,2012 Atmospheric www.atmos-meas-tech.net/5/345/2012/ Measurement doi:10.5194/amt-5-345-2012 ©Author(s)2012. CCAttribution3.0License. Techniques Volatilizable Biogenic Organic Compounds (VBOCs) with two dimensional Gas Chromatography-Time of Flight Mass Spectrometry (GC×GC-TOFMS): sampling methods, VBOC complexity, and chromatographic retention data J.F.Pankow1,2,W.Luo1,2,A.N.Melnychenko3,K.C.Barsanti2,L.M.Isabelle1,2,C.Chen1,A.B.Guenther4,and T.N.Rosenstiel3 1DepartmentofChemistry,PortlandStateUniversity,Portland,OR97207,USA 2DepartmentofCivil&EnvironmentalEngineering,PortlandStateUniversity,Portland,OR97207,USA 3DepartmentofBiology,PortlandStateUniversity,Portland,OR97207,USA 4AtmosphericChemistryDivision,NationalCenterforAtmosphericResearch,Boulder,CO80305,USA Correspondenceto: J.F.Pankow([email protected]) Received: 13May2011–PublishedinAtmos. Meas. Tech.Discuss.: 10June2011 Revised: 9December2011–Accepted: 4January2012–Published: 14February2012 Abstract.Twodimensionalgaschromatography(GC×GC) columnareprovidedfor21standardcompoundsandfor417 withdetectionbytime-of-flightmassspectrometry(TOFMS) tentatively identified VBOCs. 19 of the 21 authentic stan- was applied in the rapid analysis of air samples contain- dard compounds were found in one of the Cedrus atlantica ing highly complex mixtures of volatilizable biogenic or- SPME samples. In addition, easily quantifiable levels of at ganic compounds (VBOCs). VBOC analytical method- least 13 sesquiterpenes were found in an ATD sample ob- ologies are briefly reviewed, and optimal conditions are tained from a branch enclosure of Calycolpus moritzianus. discussed for sampling with both adsorption/thermal des- Overall,theresultsobtainedviaGC×GC-TOFMShighlight orption (ATD) cartridges and solid-phase microextraction anextreme,andlargelyuncharacterizeddiversityofVBOCs, (SPME) fibers. Air samples containing VBOC emissions consistentwiththehypothesisthatsesquiterpenesandother fromleavesoftwotreespecies(CedrusatlanticaandCaly- compounds beyond the current list of typically determined colpusmoritzianus)wereobtainedbybothATDandSPME. VBOCanalytesmaywellbeimportantcontributorstoglobal The optimized gas chromatographic conditions utilized a atmosphericlevelsoforganicparticulatematter. 45m,0.25mmI.D.low-polarityprimarycolumn(DB-VRX, 1.4µmfilm)anda1.5m, 0.25mmI.D.polarsecondarycol- umn(StabilwaxTM,0.25µmfilm). Excellentseparationwas 1 Introduction achieved in a 36min temperature programmed GC×GC chromatogram. Thousandsof VBOCpeakswere presentin Organic compounds volatilize to the atmosphere from both thesamplechromatograms;hundredsoftentativeidentifica- anthropogenic and biogenic sources. Anthropogenic emis- tionsbyNISTmassspectralmatchingareprovided.Veryfew sions of non-methane volatile organic compounds (VOCs) of the tentatively identified compounds are currently avail- have been estimated at 110 to 150TgCyr−1 (Muller, 1992; able as authentic standards. Minimum detection limit val- uesfora5lATDsamplewere3.5pptv(10ngm−3)foriso- Piccotetal.,1992). Incontrast,theglobalinputtotheatmo- sphereofnon-methanebiogenicvolatileorganiccompounds prene,methylvinylketone,andmethacrolein,and∼1.5pptv (∼10ngm−3)formonoterpenesandsesquiterpenes. Kovats- (BVOCs) has been estimated at ∼1100TgCyr−1 (Guenther et al., 1995), much of that amount being plant related. The typechromatographicretentionindexvaluesontheprimary non-methane BVOCs are: (1) important in the geochemi- column and relative retention time values on the secondary cal cycling of carbon (Guenther, 2002); (2) have significant PublishedbyCopernicusPublicationsonbehalfoftheEuropeanGeosciencesUnion. 346 J.F.Pankowetal.: VBOCswithtwodimensionalGC×GC-TOFMS effects on tropospheric ozone levels (Williams et al., 1997; al., 2010). The complexity of the VBOC group is therefore Starnetal.,1998);(3)affecthydroxylradicalconcentrations duebothto thevarietyofchemicalsub-classesrepresented, (Tan et al., 2001; Lelieveld et al., 2008) and thus the life- and to the substantial numbers of compounds in many of times of ozone-depleting and greenhouse gases (Kaplan et the sub-classes. Another complicating factor is that the re- al., 2006); and (4) oxidized in the atmosphere to products activities of individual VBOCs vary widely. With regard to thatcancondense(Haagen-Smit,1952)andtherebyformat- functionalizedcompounds,thetermVBOC,asusedhere,ex- mospheric organic particulate matter (OPM) that can affect plicitly includes early oxidation products of plant-produced theradiativeandcloudnucleationpropertiesofatmospheric compounds (caryophyllene oxide is an example of such an particles(Goldsteinetal.,2009;Po¨schletal.,2010). oxidationproduct). Thechemicalvariety,numbers,andreactivitiesofVBOCs European Union Directive 1999/13/EC defines a VOC as anyorganiccompoundhavingavaporpressureof≥10−2kPa have posed considerable challenges in efforts designed to (10−4atm)at293.15K.Isopreneasapureliquidatthistem- develop quantitative understandings of important processes perature has a vapor pressure of po=70kPa (0.7atm). A governing the VBOCs. First, large uncertainties remain re- L gardingsimplytheoverallmagnitudeofthetotalannualmass semivolatileorganiccompound(SVOC)hasbeendefinedas a compound with a po value in the range from 10−2kPa emissionsofVBOCsandwhetherunidentifiedorundetected L (10−4atm)downto10−9kPa(10−11atm)(Bidleman,1988). compoundscontributesignificantlytothoseemissions(e.g., GoldsteinandGalbally,2007).EvenwhenthelistofVBOCs By definition, an SVOC has significant affinity for con- considered is relatively “comprehensive” (e.g., isoprene, α- densed phases, yet is sufficiently volatile that it can parti- /β-pinene,α-phellandrene,camphene,13-carene,limonene, tionsignificantlytotheatmosphere,particularlywhenother myrcene, α-/γ-terpinene, terpinolene, linalool, nopinone, mechanisms (e.g., oxidation) continually remove the com- methyl-chavicol, α-bergamotene, β-caryophyllene, α-/β- pound from the gas phase and thus maintain the driving farnesene,longifolene),themeasuredfluxesandatmospheric force for volatilization. De Gouw et al. (2011) have re- concentrationsdonotappeartoaccountfortotalVBOCmass ported observations that suggest that SVOCs can evaporate emissions. For example, measured fluxes have not always from spilled crude oil, be oxidized in the atmosphere to been consistent with observed levels of OH reactivity (Di lowervaporpressurecompounds,andtherebyleadtoforma- tion of atmospheric OPM. At po≈10−3kPa (10−5atm) at Carlo et al., 2004), observed O3 fluxes (Goldstein et al., L 2004), or emissions data from co-located enclosure sam- 293K,sesquiterpenes(e.g.,farnesene)arerelativelyvolatile ples (Goldstein et al., 2004; Bouvier-Brown et al., 2009a). SVOCs,andareofsignificantinterestherefortheircontribu- Moreover, observations of higher than expected secondary tiontobothgas-andparticle-phaseprocesses. Therefore,in organicaerosollevelshavebeencitedasevidenceof“miss- placeofthetermBVOCs,weusethetermvolatilizablebio- ing” VBOCs (e.g., Goldstein and Gallbally, 2007). There genicorganiccompounds(VBOCs)torefertothefullspan are numerous difficulties that complicate the acquisition of ofbiogeniccompoundsofinteresthere. (TheVBOCgroup accurateVBOCemissionestimates. First,massemissiones- isnotintendedheretoincludemethane.) timatesaredifficultforreactiveterpenoids,eveninlocations Plants release VBOCs simply because in plant tissues wheretheemissionratesarerelativelyhigh(Bouvier-Brown these compounds tend to be lost to the surrounding envi- etal.,2009a,b;Ortegaetal.,2007). Second,largeuncertain- ronmentbydiffusivetransportmechanisms;acceleratedloss tiesremainregardingthevariabilitiesinVBOCemissionsby can occur because of mechanical/herbivore wounding (e.g., plantspecies,geographiclocation,andseason(e.g.,Helmig Heiden et al., 2003). Of particular interest is the fact that et al., 2007). For example, annual emissions in the US of increased emissions of certain VBOCs are associated with isoprene and a number of monoterpenes may be changing plantsexperiencingenvironmentalstress(e.g.,Heidenetal., by amounts and for reasons that are not adequately under- 2003). Stress indicator compounds include ocimene, farne- stood(Purvesetal.,2004),andtheresponsesofVBOCemis- sene, methyl salicylate, salicylic acid, jasmonic acid, and a sions to changing climate parameters remain highly uncer- group of C aldehydes, alcohols, and esters referred to as 6 tain (Rosenstiel et al., 2003; Guenther et al., 2006; Chen et “green-leaf volatiles” (Heil and Ton, 2008; Ka¨nnaste et al., al.,2009;PenuelasandStaudt,2010). Third,almostnothing 2008;StaudtandBertin;1998). isknownconcerningeithertherelativeimportanceofchiral Overall, VBOCs are highly diverse and include: (1) nu- variants (e.g. (+)- vs. (−)-α-pinene, (+)- vs. (−)-β-pinene, merous terpenes including isoprene (C H ) which is etc.),orhowmuchimportantbiogeochemicalinformationis 5 8 a hemiterpene, monoterpenes (C H ), sesquiterpenes heldwithinchiralpatternsofVBOCemissions(Williamset 10 16 (C H ), and diterpenes (C H ); (2) terpenes that are al., 2007; Yassaaetal., 2010). Overall, despitethefactthat 15 24 20 32 functionalized(e.g,oxidized)innumerouswaysandatava- manyBVOCemissionstudiescurrentlyappeareachyearin riety of positions; (3) alkanes and alkenes; (4) alkyl alde- the scientific literature, most have targeted only a few com- hydes and ketones; (5) alkyl alcohols, ethers, acids, and es- poundsasanalytes. Incontrast,investigatorsmakingtheef- ters; and 6) chiral variants such as (±)-α-pinene, (±)-β- fort to look for other compounds (e.g., Jardine et al., 2010) pinene, and (±)-limonene (Williams et al., 2007; Yassaa et havefoundawiderangeofVBOCs. Atmos. Meas. Tech.,5,345–361,2012 www.atmos-meas-tech.net/5/345/2012/ J.F.Pankowetal.: VBOCswithtwodimensionalGC×GC-TOFMS 347 Given the significant impact of VBOC emissions on e.g., cannot distinguish individual monterpenes, individual biosphere-atmosphereinteractionsandairchemistry,thereis sesquiterpenes,etc.) a great need for development and application of new ana- lyticalmethodologiescapableofcharacterizingthecomplex 2.2 Determinationinthelaboratory nature of VBOC emissions. Development of new method- ologiescanbeobtainedby: (1)applicationofsuitablesam- ATD cartridges are easily transported to and from the field, ple collection methods; (2) application of high-separation and provide a simple, sensitive, and quantitative approach gas chromatography (GC) methods that can adequately ad- for determining a wide range of VOCs at ambient atmo- dress VBOC complexity; and (3) accumulation of GC re- sphericlevels(Pankowetal.,1998,2003). Laboratory-based tention index information based on as many authentic stan- ATD determinations of VBOCs in field samples have uti- dard compounds as possible (many VBOCs of interest are lized GC/FID, GC/MS, and GC×GC-TOFMS (Saxton et noteasilyobtainedinpureform). Herewedescribethecol- al., 2007). (FID=flame ionization detector.) Several con- lection,tentativeidentificationanddetermination,andchro- siderationsareimportantintheimplementationofATDwith matographic characterization of VBOCs in highly complex VBOCs. First, losses of reactive VBOCs may need to be samples using two-dimensional gas chromatography/time- prevented by removal of oxidants, particularly ozone, prior of-flight mass spectrometry (GC×GC/TOFMS), the latter to passage of the sample air through the cartridge (Cao and being the most powerful separation+detection methodology Hewitt, 1994; Calogirou et al., 1996). Numerous different currentlyavailableforVBOCs.Weprefacethedescriptionof ozoneremovalmethodshavebeenused(Helmig,1997;Fick thelaboratorymeasurementswithabriefreviewofavailable et al., 2001; Pollmann et al., 2005). These have involved methodsforsamplingandanalysis. sodiumthiosulfateonfilters(Helmigetal.,1998),potassium iodideonglasswoolorfilters(HelmigandGreenberg,1994), manganesedioxideonporousnetsorcopperscreens(Hoff- 2 Availablesamplingandanalysismethods mann, 1995; Calogirou et al., 1996), and titration of ozone withnitrogenmonoxide(Komendaetal.,2003). Second,be- 2.1 Determinationwithfield-deployedinstrumentation causeatmosphericconcentrationsdecreasestronglywithde- creasing VBOC volatility (viz., vapor pressure), if the goal Analyticalinstrumentshaveoftenbeendeployedtothefield is the simultaneous quantitation of a wide range of com- instudiesofplant-derivedVBOCsinambientairandasem- pounds, it may be necessary to use a large sample volume. anatingfromplantswithinexperimentalenclosures. Though That choice usually leads to the need for multiple sorbents oftencostlyandlogisticallydifficult,fielddeploymentofin- in the bed: weaker sorbent first, then the stronger sorbent. strumentsisadvantageouswhenplantemissionsandambient Thelowpo compoundsareretainedontheweakersorbent, L concentrations are subject to short time variations: VBOCs andthehigherpo compoundspassontoandareretainedby L thataremoderately-to-highlyreactiveareofparticularinter- thestrongersorbent. Aftersampling,thethermaldesorption est in this regard. Field-deployed analytical approaches are flowoccursinabackflushdirectionsothatthelowerpo com- L discussedinarecentreview(OrtegaandHelmig,2008)and pounds are not exposed to the stronger sorbent. Given the include: (1) chemiluminescence for detection of isoprene largedynamicrangeinconcentrationusuallyaccompanying (Guentheretal.,1996;SingsaasandSharkey,2000);(2)pro- such samples, currently it can be helpful to employ a unit ton transfer reaction mass spectrometry (PTR-MS) for the such as the TurboMatrix 650 ATD (PerkinElmer, Waltham, direct simultaneous detection of multiple compounds (Karl MA)forthethermaldesorption. WiththeATD,thedesorp- et al., 2001; Bamberger et al., 2010; Mielke et al., 2010); tionflowcanbedividedsothatpart(e.g.,30%)goestothe (3) solid phase microextraction (SPME) fiber collection of GC and the remainder goes to a pre-cleaned cartridge to be analytesfollowedbythermaldesorptiontoafieldGCinstru- heldforapossiblesecondGCrun. ForthefirstGCrun,the ment;and(4)adsorption/thermaldesorption(ATD)cartridge injector split ratio might be 10:1. If the injector split ratio sampling of a known air volume (e.g., 1–10 l) followed by in the second GC run is lower (e.g., 5:1), then the method thermaldesorptionoftheanalytestoafieldGCinstrument. sensitivity in the second run can be higher than in the first PTR-MShasbeenemployedwhenVBOCconcentrationsare run(e.g. (70/30)(0.2/0.1)≈4×). Thiscanallowlessabun- subjecttohightemporalvariability(e.g.,minutes)asinstud- dantcompoundstobecomequantifiable,thoughsomeofthe iesofforestair(e.g.,Mielkeetal.,2010),fluxesbyeddyco- early-elutingcompoundsmaythenbeoverloaded. variance measurements (e.g., Karl et al., 2001; Bamberger CommonATDadsorbentsforVBOCsincludetheporous et al., 2010), and emission processes in dynamic enclosure organic polymer TenaxTM TA and several more strongly studies (e.g., Grabmer et al., 2006; Bouvier-Brown et al., sorbing carbon-based materials (Table 1). Attractive char- 2007).Goodagreementhasbeenobtainedfortotalsesquiter- acteristics of these adsorbents are their good abilities to re- penesinfield-deployedPTR-MSvs.SPME(Bouvier-Brown versibly sorb analytes, low water affinities (i.e., low break- et al., 2007). (PTR-MS cannot distinguish different struc- throughvolume(BV)valuesforwater),andcomplementary tural isomers because it only measures molecular masses, sorption strengths (Dettmer and Engewald, 2002; Arnts et www.atmos-meas-tech.net/5/345/2012/ Atmos. Meas. Tech.,5,345–361,2012 348 J.F.Pankowetal.: VBOCswithtwodimensionalGC×GC-TOFMS Table 1. Properties of adsorbent materials used in collection and analysis of volatilizable biogenic organic compounds (VBOCs) using adsorption/thermaldesorption(ATD). Sorbent Surfacearea Maximum Type Class Range Water Breakthroughvolume(BV)at20◦C(lg−1) (m2g−1) temp.(◦C) capacity water pentane benzene limonene (mgg−1)a TenaxTMTA 35 350 porouspolymer weak C6-C30 <3.3b 0.039c 5d 70d 12000d 0.065d TenaxTMGR 24 350 porouspolymer+ weak C6-C30 <2.0b 0.092c 1.7e 9e graphitizedcarbon CarbotrapTMB 100 >400 graphitizedcarbon medium/weak C5-C12 <1.2b 5.9f 11.7f CarbopackTMB 100 >400 graphitizedcarbon medium/weak C5-C12 Carbograph1TDTM 100 >400 graphitizedcarbon medium/weak C5-C12 CarbotrapTMX 240 >400 graphitizedcarbon medium/strong C3-C9 CarbopackTMX 240 >400 graphitizedcarbon medium/strong C3-C9 Carbograph5TDTM 560 >400 graphitizedcarbon medium/strong C3-C7 24g CarboxenTM569 485 >400 carbonmol.sieve strong C2-C5 403b 0.257c 200h 85h 16000h CarboxenTM1000 1200 >400 carbonmol.sieve verystrong C2-C5 445g 0.418c CarbosieveTMSIII 975 >400 carbonmol.sieve verystrong C2-C5 395b 0.378c 600i aWatersorptioncapacitiesat20◦Candrelativehumidity=95–100%. bHelmig,D.andVierling,L.: WateradsorptioncapacityofthesolidadsorbentsTenaxTA,TenaxGR,Carbotrap,CarbotrapC,CarbosieveSIII,andCarboxen569andwater managementtechniquesfortheatmosphericsamplingofvolatileorganictracegases,Anal.Chem.,67,4380–4386,1995. cDettmer,K.andEngewald,W.:Adsorbentmaterialscommonlyusedinairanalysisforadsorptiveenrichmentandthermaldesorptionofvolatileorganiccompounds,Anal.Bioanal. Chem.,373,490–500,2002. dhttp://www.sisweb.com/index/referenc/tenaxta.htm,lastaccess:6March2011. ehttp://www.sisweb.com/index/referenc/tenaxgr.htm,lastaccess:6March2011. fhttp://www.sigmaaldrich.com/etc/medialib/docs/Supelco/Bulletin/4501.Par.0001.File.tmp/4501.pdf,lastaccess:6March2011. gFastyn,P.,Kornacki,W.,Gierczak,T.,Gawłowski,J.,andNiedzielski,J.: Adsorptionofwatervapourfromhumidairbyselectedcarbonadsorbents,J.Chromatogr.A,1078, 7–12,2005. hhttp://www.sisweb.com/index/referenc/carbo569.htm,lastaccess:6March2011. ihttp://www.sisweb.com/index/referenc/carbs111.htm,lastaccess:6March2011. al.,2010). BVvaluesformonoterpenesonTenaxTM TAare Super-Q and HayeSep-Q adsorbent cartridges (Papiez et high(Table1)sothatforsamplevolumesofafewlitersand al., 2009; Ormeno et al., 2010; Heuskin et al., 2012) and ∼100mg of TenaxTM TA, monoterpenes are quantitatively SPMEfibers(Bouvier-Brownetal.,2009b)providealterna- retained. Thatisnotthecaseforisopreneandothervolatile tives to ATD cartridges. SPME, however, can be consider- compounds, retentionofwhichrequirestheadditionalpres- ably less sensitive than ATD (as with highly volatile ana- ence of a carbon-based sorbent. Many current ATD appli- lytesthatareweaklysorbedand/orwithlowvolatilitycom- cations utilize 0.25 inch O.D.×3.5 inch cartridges packed pounds at low concentrations), does not provide as much with ∼100mg of TenaxTM TA followed by ∼100mg of ei- overall sample stability (Bouvier-Brown et al., 2007; Baker therCarbotrapTM BorCarbographTM 1TD(Komendaetal., andSinnott,2009),andislaborintensivetocalibrateinquan- 2001; Sartin et al., 2001; Hakola et al., 2006; Helmig et titative applications. Nevertheless, SPME becomes exceed- al.,2007;OrtegaandHelmig,2008;Haapanalaetal.,2009; ingly attractive for its convenience when the analyte con- Geron and Arnts, 2010). The TenaxTM TA prevents com- centrations are high, the work is proceeding in/near a lab- pounds like the monoterpenes from substantively reaching oratory, and/or the desired determinations are only qualita- the carbon-based sorbent whereon they would not only be tive or semi-quantitative in nature. SPME has been used in strongly retained but also possibly chemically altered. Sig- laboratory-based measurements with GC-FID to determine nificantinterconversionofseveralmonoterpeneshasbeenre- isopreneemissionratesfromenclosedtreeseedlingbranches ported on CarbotrapTM 200 and CarbotrapTM 300 (Green- (Tsui et al., 2009), and with GC/FID, GC/MS and 100 l berg et al., 1999a). Similar results have been reported by Tedlarbagstomeasuresesquiterpeneemissionsfromwhole CaoandHewitt(1993). trees(PinussabinianaandPinusponderosa)(BakerandSin- nott, 2009). Super-Q and HayeSep-Q are most appropriate Atmos. Meas. Tech.,5,345–361,2012 www.atmos-meas-tech.net/5/345/2012/ J.F.Pankowetal.: VBOCswithtwodimensionalGC×GC-TOFMS 349 forlower-volatilityVBOCs(e.g.,mono-andsesquiterpenes); of forest canopy air and air from branch and leaf enclo- samples typically are solvent extracted an analyzed using sures(Saxtonetal.,2007). Quantitativemeasurementswere GC/MS(Papiezetal.,2009;Ormenoetal.,2010;Heuskinet reported for isoprene, α-pinene, β-pinene, camphene and al.,2012). limonene,aswererelativepeakintensitiesfor11tentatively “Whole-air” sampling with canisters or bags provides an identifiedcompounds(Saxtonetal.,2007). alternativetoATDcartridgesandSPMEfibersasameansto collect and transport sample analytes to a conventional lab- 3 Experimental oratory for analysis. Whole air in a canister or bag can be aliquottedwithasampleloop,cryofocused(e.g.,at−150◦C 3.1 Chemicalsandstandardmixtures on a trap containing 60/80 mesh glass beads as in method TO-14; US EPA, 1999a), sampled using ATD (Pressley et 21 VBOC chemicals (≥95% pure) were obtained from al., 2004) or SPME (Bouvier-Brown et al., 2007), or cry- Sigma-Aldrich (St. Louis, MO) (isoprene, α-pinene, cam- ofocussed directly on the GC column (Pankow, 1986). For phene, myrcene, β-pinene, α-phellandrene, 13-carene, canisters, inert internal surfaces are important, and both limonene, terpinolene, p-cymene, nopinone, linalool, 4- Summa-polishedTMstainlesssteelandSilcosteelTMcanisters terpinenol,terpineol,eucalyptol,camphor,estragole(methyl areused(USEPA,1999a,b).However,canistersarestillsub- chavicol), α-cedrene, caryophyllene, aromadendrene, and jecttolossesathighhumidity(Battermanetal., 1998), and α-humulene). Optimally, the list of standard compounds arecostlyandcomplicatedtoclean. Teflonbagssufferfrom would have included many tens of C -and-higher VBOCs. 10 blankproblems,frequentlyleak,andaredifficulttocleanaf- Unfortunately, authentic standard materials are not read- ter use (Greenberget al., 1999a); Tedlar bagsare subject to ily available for the vast majority of such compounds, and similarproblems. Availabledataindicatethatmanyanalytes evenjusthavingdoubledthenumberofstandardcompounds aremorestablewhenstoredonATDcartridgesthaninbags, wouldhaverequireduseofmulti-component“essentialoils”, especiallywhenoxidantspeciesarepresentinthesampleair. with concomitant prior quantitation of the VBOC compo- Evidence of losses in Teflon bags for methacrolein, methyl nents of interest. Such efforts are underway in our lab- vinylketone,andα-pinenehavebeenreported(Greenberget oratories, but were not applied in this study. Standard al.,1999a). Asummaryofminimumdetectionlimit(MDL) mixes of the 21 VBOCs in methanol were prepared at per- values for VBOCs by various methodologies is provided in component concentrations from 0.5ngµl−1 to 20ngµl−1. Table2. A gas standard containing four internal standard (IS) com- pounds (fluorobenzene, toluene-d , 4-bromofluorobenzene, 8 2.3 GC×GC-TOFMS and 1,2-dichlorobenzene-d ) at 80ngml−1 per component 4 was prepared in a stainless steel canister as described else- Applications of GC×GC usually utilize a primary column where(Pankowetal.,1998). TheIScompoundswereused: witheitheranon-polarorlow-polaritystationaryphase. Co- (1)tomonitortheoveralleffectivenessofthethermaltrans- elutionofsomepeaksfromtheprimarycolumnisinevitable ferfromeachsamplingcartridgetotheprimaryGCcolumn; withhighlycomplexsamples. WithGC×GC(aka2DGC), (2)asclearlyidentifiable(e.g.,non-biogenic)retentiontime periodic slices of effluent from the primary column are in- markers;and(3)asourceofconstantreferencesignalsduring dividually: (1) cryofocused at the end of the column; then thedeterminationoftheMDLvaluesreportedhere,asbased (2) thermally desorbed to a secondary column with a more onanalysescarriedoutatvaryingon-cartridgelevelsofiso- polar stationary phase. Separation of compounds that co- preneandtheothertargetanalytes. Isotopically-labelledsur- existinagivenslicecanbeaccomplishedonthesecondary rogatestandardcompoundswerenotnecessaryinthisstudy. columnifthecompoundsdifferinpolaritiesand/orpolariz- abilities. GC×GC-TOFMS has been applied with success 3.2 ATDcartridges–NCARpreparationand in the analysis of diesel fuel (Arey et al., 2005), weathered proceduresforfieldsampling petroleum(Frysingeretal.,2003),urbanaerosolparticulate matter (Hamilton et al., 2004), tobacco smoke particulate Field samples discussed here were collected by the matter (Cochran, 2008), and tissue extracts (Welthagen et Biosphere-Atmosphere Interactions Group of the National al., 2005). GC×GC chromatographic retention index data CenterforAtmosphericResearch(NCAR).Detailsregarding havebeentabulatedfordieselfuelhydrocarbons(Areyetal., sample collection are available (Greenberg et al., 1999a,b, 2005),butnotforVBOCs. 2004). Briefly, samples were obtained using stainless The varying polarizabilty/polarity characteristics of the steel, 0.25 inch O.D.×3.5 inch, dual sorbent cartridges VBOCs are imparted by varying numbers of rings, double (TenaxTM TA plus CarbotrapTM B, or TenaxTM GR plus bonds, and polar functionalizations. GC×GC with a less- CarbographTM). Prior to sampling, cartridges were cleaned polar primary column and a more-polar secondary column by heating for 8h at 275◦C with a 50mls−1 flow of ultra- is well suited for separating VBOC mixtures. GC×GC- highpurityN gas. Cleancartridgeswerecappedandstored 2 TOFMShasbeenusedtodetermineVBOCsinATDsamples at 10◦C until used. The apparatus consisted of an O trap 3 www.atmos-meas-tech.net/5/345/2012/ Atmos. Meas. Tech.,5,345–361,2012 350 J.F.Pankowetal.: VBOCswithtwodimensionalGC×GC-TOFMS b a T p S a ptv=partspertrillion IM=selectedionmon methylsalicylatesesquiterpenesmonoterpenesisoprenemethacroleinmethylvinylketone compound MDLdefinitiondetectioncollection ble2.Minimum byvolume;conversion itoringmodeforthem 0.9–1.48–120.7–2.14–123.5103.5103.510 pptvngmb3−(S/N)=10signal/noiseGCxGC-TOFMSATD,5liter ThisWork detectionlimit(M betweenngmandpp3−assspectrometer(i.e.,no 1106209326067200120350 pptvngm3−blanknoise3oftheσPTR-MSdirectinlet (2010)Mielkeetal. DL)valuesreporte tvassumes=293KaT tfullscanningmode). 402262005667020470204 pptvngm3−(S/N)=2signal/noisePTR-MSdirectinlet (2010)Langfordetal. dinthisworkand nd in totalpressure=1atm 213 528 pptvngm3− notstatedPTR-MSdirectinlet (2008)Karletal. otherstudies. . 400340050280100280 pptvngm3−blanknoise1.5oftheσGC/MSATD,1liter (2010)Misztaletal. p 514 ptvngm3− notstatedGC/MSATD,3liter (2007)Saxtonetal. 921 p 79–1111–60132 ptvngm− notstatedGC/MSATD,3liter (2006)Hakolaetal. 3 1613 pptvngm3− notstatedGC/MSSIMaATD,6liter (2004)Greenbergetal. p G 16131313 ptvngm3− notstatedGC-FIDATD,6liter (1999a)reenbergetal. Atmos. Meas. Tech.,5,345–361,2012 www.atmos-meas-tech.net/5/345/2012/ J.F.Pankowetal.: VBOCswithtwodimensionalGC×GC-TOFMS 351 (KI impregnated filter), the sample cartridge, and a flow- controlled pump. Samples were collected at a flow rate of 200cm3min−1 for 30min. After sampling, cartridges were sealed, shipped at ambient temperatures back to the labora- tory (maximum 1 week shipping time), then maintained at 10◦Cuntilanalyzed. Afterfourweeksatambienttempera- tures,losseswere<10%forC toC compoundsofinterest 3 6 (Greenbergetal.,1999a). 3.3 ATDcartridgesandSPMEfibers–PSU preparationandproceduresforsamples obtainedinthelaboratory In laboratory ATD sampling of air from vials holding plant materials,eachglasscartridgecontained100mgofTenaxTM TAplus120mgofCarbotrapTM B.(Priortopacking,5gof TenaxTM TAwasplacedinaglasscolumnandcleanedwith Fig.1. DiagramofsystemusedtoobtainSPMEandATDsamples a 250ml flow of 50/50 (v/v) acetone/hexane.) After pack- ofVBOCsinthelaboratorywithsamplesofCedrusatlantica. ing, each cartridge was conditioned for 1h at 300◦C with 100mlmin−1 of He gas (precleaned using a U-shaped hy- drocarbon trap in liquid N ). Each conditioned cartridge sampling event was carried out for a specific time period 2 wassealedwithbrass Swagelok endcaps thathadbeenpre- (10to60min)under“dynamic”conditions(=continuousair cleanedbyrinsingin50:50acetone/hexanefollowedbybak- flow–50mlmin−1–throughthevialduringfiberexposure). ing (90min, 150◦C). The endcaps were fitted with Teflon SPMEsampleswereanalyzedimmediately. ferrules precleaned with methanol and water. Each sealed cartridge was stored in a clean glass culture tube. Other 3.4 GC×GC-TOFMSmeasurements cartridge handling procedures were as described elsewhere (Pankow et al., 1998). The SPME assembly was obtained AllATDandSPMEsampleswereanalyzedusingaLecoPe- fromSigma-Aldrich(St.Louis,MO)andutilizedwithfibers gasus 4D GC×GC-TOFMS (Leco, St. Joseph, Michigan). coated with polydimethylsiloxane/divinylbenzene (coating Unlike the efforts using ATD, with SPME no calibration thickness65µm). Priortosampling,thefiberswerecleaned standardrunsweremade: theprimarypurposeoftheSPME for 30min at 250◦C in a GC injector through which pre- runs was to provide samples that could easily indicate the cleanedHewasflowingat50mlmin−1. levelofsamplecomplexitythatistobeencounteredinanal- Plant material samples (needles) were collected from a yses of VBOCs, and easily provide needed retention time mature(>20m)CedrusatlanticatreelocatedonthePortland data. For all ATD cartridges (standard or sample or blank, StateUniversity(PSU)campus.Threesamplesof0.5to1.0g NCARorPSU),priortoanalysiseachcartridgewasloaded each were collected in October–December (2010) at ∼2m withtheIScompoundsbyinjecting0.2mloftheISgasstan- above ground level from tips of branches with full-sun ex- dardintoa50mlmin−1,5minflowofprecleanedN2leading posure. Eachsamplewasplacedinanindividualprecleaned to the sample inlet end of the cartridge. For a standard car- 40mlclearglassvialfittedwithaTeflon-linedseptum. Af- tridge,the21targetanalytecompoundswerethenaddedby ter60minofexposureat∼20◦Ctoacoollightsource(air- injecting 4 µl of one of the methanolic standard mix solu- portedhalogenlamp,300watt,1000µmolphotonsm−2s−1), tionsintotheinletofthecartridge;mostofthemethanolwas sampling by SPME or ATD occurred as air was passing thenremovedandtheanalytesweremovedfurtherontothe through the vials (Fig. 1). The air was either cleaned lab- cartridgebya50mlmin−1,5minflowofprecleanedN2. oratoryair(CLA)oruncleanedlaboratoryair(ULA).Clean- TheATD400(Perkin-Elmer,Waltham,MA)thermaldes- ing of the air was accomplished using a Perkin-Elmer hy- orption apparatus was connected to the GC injector by a drocarbon trap. Use of the uncleaned laboratory air greatly fused silica transfer line (220◦C). Each ATD cartridge was increased the complexity of the samples and thereby fa- thermally desorbed (270◦C, 10min) at 40mlmin−1 with cilitated a more rigorous test of the ability of GC×GC- zero split to a focusing trap (45mg of TenaxTM TA, 10◦C) TOFMS to adequately separate and detect VBOCs in the in the ATD 400. The focusing trap was desorbed (290◦C, presenceofmyriadotherVOCs. ATDairsamplingoccurred 3min) with zero split (splitless) to the fused silica transfer at 60mlmin−1 for 60min (3.6l). ATD cartridges were ei- line and thus the GC injector. Flow from the GC injec- ther analyzed immediately or within three weeks after stor- tor to the primary GC column occurred with a 15:1 split. ageat−15◦C;standardcartridgesanalyzedafterthreeweeks (Increased method sensitivity could have been achieved by at−15◦Cshowednoevidenceofanalyteloss. EachSPME reducing the split to as low as 5:1 without reduction of www.atmos-meas-tech.net/5/345/2012/ Atmos. Meas. Tech.,5,345–361,2012 352 J.F.Pankowetal.: VBOCswithtwodimensionalGC×GC-TOFMS Table3.SummaryofGC×GC-TOFMSconditionsusedinVBOCdeterminations. initial improved ◦ ◦ GCinjector 250 C;split20:1forATD 225 C;split15:1forATD,splitlessforSPME columnflow 1.2mlmin−1(massflowcontroller) 1.0mlmin−1(massflowcontroller) primarycolumn non-polar:Rxi-5ms,30m,0.25mmI.D., lowpolarity:DB-VRX,45m,0.25mmI.D., 0.25µmfilm(Restek,Bellefonte,PA) 1.4µmfilm(Agilent,SantaClara,CA) GC×GCmodulation 4speriod,0.8shotpulse 4speriod,0.9shotpulse GC×GCmodulator trapwithcoldgasfromLN2,thenhotpulseat20◦aboveprimaryovenforreleasetosecondarycolumn secondarycolumn polar:BPX-50,1.25m,0.10mmI.D., polar:StabilwaxTM,1.5m,0.25mmI.D., 0.10µmfilm(SGE,Austin,TX) 0.25µmfilm(Restek,Bellefonte,PA). GCprogram 40◦Cfor5min,15◦Cmin−1to300◦C,then 45◦Cfor5min,10◦Cmin−1to175◦C,hold (primaryoven) holdat300◦Cfor5min at175◦Cfor2min,4◦Cmin−1to240◦C, ◦ thenholdat240 Cfor10min ◦ MSsource 200 C,electronimpact(70eV) MSdetector 1550V MSdataacquisition 150spectras−1;35to500amu chromatographicperformance.) SPMEfibersweredesorbed stated otherwise. Figure 2a and b are GC×GC-TOFMS splitless for 3min in the GC injector (225◦C) which con- chromatogramsobtainedbyATDforthe21-compoundstan- tained an SPME liner (Restek, Bellefonte, PA). Thereafter, dardmix. Goodseparationwasachievedforallofthecom- an injector purge flow of 100mlmin−1 was kept open for pounds except for α-cedrene and caryophyllene. The latter 4min. Each GC×GC run was initiated immediately upon two compounds possess sufficiently different EI mass spec- beginningoftheheatingofthefocusingtrap(ATD),orafter tra that they are differentiable even when not fully resolved theSPMEfiberwasinsertedintotheinjector(SPME).Inev- chromatographically. erycase,TOFMSdatacollectionwasdelayed180s. Anini- In sample runs, the peak selection criteria (PSC) used tialsetandanimprovedsetofGCcolumns/conditionswere todecideuponwhichchromatographicpeakswouldreceive used(Table3). further consideration were applied on the Leco data system asfollows: signaltonoiseratio>200,peakarea>100000, and match similarity of >780 with an entry in the NIST 4 Resultsanddiscussion massspectrallibrary. Foreachsample,someofthosepeaks corresponded to target compounds, and eluted at the proper 4.1 Detectionlimits retention times, and so were positively identified. Other peaks passing the PSC but not corresponding to the target The particular MDL values for the 21 target compounds compounds were considered only tentatively identified. Ta- investigated here are provided in Table 4. For isoprene, ble S1 (Supplement) summarizes the information acquired the MDL was 3.5pptv (10ngm−3). For the monoter- for438chromatographicpeaks(including21standardcom- penes, theMDLvalueswereintherange0.7–2.1pptv(4to pounds)thatmetthePSCinthevariouschromatogramspre- 12ngm−3). For the sesquiterpenes, the MDL values were sentedhere. ∼1pptv (∼10ngm−3). No blank problems were expe- rienced for any of the target compounds. As such, all – Calycolpus,ATD,ambientair: MDLvalueswereassessedbyvaryingtheon-cartridgemass Figure 3 is a chromatogram for an ATD sample ob- amounts of the target analytes, and determining which val- tainedduringanNCARenclosureexperimentusingthe uesyieldedaninstrumentsignaltonoiseratioof10:1. MDL lower branch of a tree (Calycolpus moritzianus) grow- valueswerethencalculatedasequalingthemassamountsat ing in a tropical forest in South America (Columbia, 10:1signal-to-noisedividedbythesamplevolumeof5l. 11.0704◦N, 74.0411◦W). For this sample, the initial set of chromatographic conditions was used (Table 3). 4.2 Chromatograms The peaks for 15 of the 21 standard compounds met the PSC and were positively identified based on their Figures2–6showchromatogramsforsamplesrunusingthe chromatographicretentiontimesandmassspectra(Ta- improvedsetofchromatographicconditions(Table3)unless bleS2,Supplement);fourofthese(α-pinene,limonene, Atmos. Meas. Tech.,5,345–361,2012 www.atmos-meas-tech.net/5/345/2012/ J.F.Pankowetal.: VBOCswithtwodimensionalGC×GC-TOFMS 353 9 3. Standard (~1 ng per component on column) MT = monoterpenes OOxxMMTT == ooxxyyggeennaatteedd mmoonnootteerrppeenneess ST = sesquiterpenes OxST = oxygenated sesquiterpenes e m tion t92.9 nti e et y r dar isoprene 1,2-DCB-d4 n eco9 toluene-d8 OxMT s1.1 s) FB t(2 BFB ST MT 9 0. 300 800 1300 1800 2100 t1(s) primary retention time (a) Figure 2.a. 4 2. Standard (~1 ng per component on column) e m ion t1.9 notteeprrippniionnneeeooll OOxxMMTT nti estragole e et 1,2-DCB-d y r 4 44tterpiinenoll ar ST nd BFB camphor cedrene co linalool humulene se.41. MT s) pcymene aromadendrene t(2 campmhyernceene limtoenrpeinneo,le enuecalyptol caryophyllene phellandrene, 3carene pinene -pinene 9 0. 852 1352 1852 t1(s) primary retention time (b) Figure 2.b. Fig. 2. (a) GC×GC-TOFMS chromatogram of a standard containing 21 VBOCs at ∼1ng per component on-column; internal standard compoundsarefluorobenzene(FB),toluene-d8,bromofluorobenzene(BFB),and1,2-DCB-d4(1,2-dichlorobenzene-d4);obtainedusingthe improvedchromatographicconditions(Table3).(b)EnlargedregionofGC×GC-TOFMSchromatogramshowing20VBOCsat∼1ngper componenton-column. α-terpineol,andcaryophyllene)areknownconstituents Supplement) and included four monoterpenes and 10 of C. moritzianus essential oils (Valde´s et al., 2006; sesquiterpenes. In addition, at levels two to five times Diazetal.,2008). Peaksforanothertwostandardcom- larger than blank levels, three of the 127 peaks were pounds were positively identified based on their mass identifiedassalicylates(methylsalicylate,2-ethylhexyl spectralandretentiontimedata,thoughweretoosmall salicylate, and homomenthyl salicylate). Two of the to meet the PSC. The number of tentatively identified 127peaksmatchedwellasisomersofcumeneandtwo peaksmeetingthePSCinFig.3totaled127(TableS2, matchedwellasisomersofaromadendrene. Overall,at www.atmos-meas-tech.net/5/345/2012/ Atmos. Meas. Tech.,5,345–361,2012 354 J.F.Pankowetal.: VBOCswithtwodimensionalGC×GC-TOFMS 6 2 2. ATD –branch enclosure, Calycolpus moritzianus MT = monoterpenes OxMT = oxygenated monoterpenes e ST = sesquiterpenes m tion t61.76 nti e et y r ar d n o sec.261 isopprene MT OOxxMMTT SSTT s) ( t2 aallkkaanneess 6 7 0. 182 682 1182 t (s) primary retention time 1 Figure 3. Fig.3. GC×GC-TOFMSchromatogramofVBOCsfromCalycolpusmoritzianusbyATDusingabranchenclosureinthefield;obtained usingthe“initial”chromatographicconditions(Table3). 9 3. SPME -cleaned lab air (CLA) Cedrus atlantica MT = monoterpenes OxMT = oxygenated monoterpenes e ST = sesquiterpenes m OOxxSSTT = ooxxyyggeennaatteedd sseessqquuiitteerrppeenneess tion t92.9 OxST nti e et ry ar d on OxMT c e9 s1.1 s) isoprene ST ( t2 alkanes MT 9 0. 300 800 1300 1800 2100 t (s) primary retention time 1 Figure 4. Fig. 4. GC×GC-TOFMS chromatogram of VBOCs from Cedrus atlantica by SPME and using cleaned laboratory air (CLA); obtained usingthe“improved”chromatographicconditions(Table3). least 13 sesquiterpene compounds were present at eas- – Cedrus,SPME,laboratoryair(cleaned): ilydetectableconcentrations.Fortheseandthetentative Figure 4 is a chromatogram for VBOCs emitted from identificationsdiscussedbelowforFigs.4–6,confirmed CedrusatlanticaneedlessampledinPortland(Decem- identificationswillawaitacquisitionofsuitableauthen- ber 2010). Sampling took place with SPME using the ticstandards. cleaned laboratory air (CLA) using dynamic sampling conditionsasdescribedabove. Thepeaksfor10ofthe Atmos. Meas. Tech.,5,345–361,2012 www.atmos-meas-tech.net/5/345/2012/
Description:The list of books you might like
