Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | er- ff erent ff C-values for Techniques 13 Discussions Atmospheric δ Measurement 53.6‰), bromoform erent sampling sites, − ff 1‰ and agreed well with previously . 1 ± erences were observed for chlorodifluo- ff 9 . 42 2162 2161 − C-range of almost 100‰ making isotope ratios a 13 δ erent sources cover a ff C by almost 10‰. Less pronounced di 13 13.5‰) and other halocarbons. In order to determine the carbon isotopic composition of C1 and C2 halocarbons ment Techniques (AMT). Please refer toif the available. corresponding final paper in AMT This discussion paper is/has been under review for the journal Atmospheric Measure- − 1998; Thompson, 2002; Goldstein and2005; Shaw, Redeker, 2003, 2007; and Mead, references 2008). therein; Archbold, thermore we report first( carbon isotope ratios for iodomethane ( 1 Introduction Compound specific isotope ratio mass spectrometryorganic (CSIRMS) of compounds non methane (NMVOCs) volatile sources emerged and as to a provide powerful information on tool sinks to (Rudolph distinguish et di al., 1997, 2000; Tsunogai, bromomethane at the urbanpublished site results. were Butin at the coastalromethane, site 1,1,1-trichloroethane bromomethane and was chloromethane.ences substantially are enriched related We to suggest the that turnover these of di these compounds in ocean surface waters. Fur- very promising toolthe for determination studying of the thechloromethane biogeochemistry isotopic is hampered composition of by of these their C1 low compounds. mixing and ratios. C2 So halocarbonswith others far, mixing than ratios assystem low and as (ii) 1 a pptv chromatographic (i)veloped set and a up validated. field for The suitable processing sampling cryogenic system thesean was samples high urban tested have volume at and been two sampling de- a di coastal location in Northern Germany. The average The isotopic composition ofinformation volatile on organic compounds their (VOCs) sourcesticular can and the provide fate reported valuable not carbon stable deduciblefrom isotope from di ratios mixing of ratios chloromethane alone. and bromomethane In par- Abstract A high volume sampling systemisotope for determination of volatile halocarbons and hydrocarbons E. Bahlmann, I. Weinberg, R. Seifert, C.Institute Tubbesing, and for Biogeochemistry W. Michaelis and Marine Chemistry, Hamburg, Germany Received: 15 March 2011 – Accepted:Correspondence 17 to: March 2011 E. – Bahlmann Published: ([email protected]) 8 AprilPublished 2011 by Copernicus Publications on behalf of the European Geosciences Union. Atmos. Meas. Tech. Discuss., 4,www.atmos-meas-tech-discuss.net/4/2161/2011/ 2161–2188, 2011 doi:10.5194/amtd-4-2161-2011 © Author(s) 2011. CC Attribution 3.0 License. 5 25 15 20 10 Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | c ffi I), dibromomethane 3 40‰ for anthropogenic W) between 29 August − ◦ cient for measurements of ffi N, 8.4396 ◦ C values from W), from 19 to 24 September 2010. The ◦ 13 δ 2164 2163 N, 9.9736 ◦ ). Accompanying determination of the carbon isotope 3 147‰ for chloromethane synthesized from pectin (cf. Keppler − ) and bromoform (CHBr 2 Br C and the analytes were transferred to an adsorption tube for storage and analy- ◦ 2 Three ambient air samples were taken at the institute building in the center of Ham- In addition, three marine influenced air samples were taken at the AWI Wadden Sea To the best of our knowledge, there is only one study reporting carbon isotope ratios The main objective of this study was to develop a simple and robust sampling system For determining carbon isotope ratio of atmospheric VOCs both, air sampling canis- In particular, stable carbon isotope ratios have been proposed for constraining the and 5 September. The sampling system was placed directly at the coastline 200 m 100 ppm in nitrogen) were used. burg, Northern Germany (53.5686 sampling system inlet wasbetween placed 10:00 at a.m. 30 m andexhaust, above 02:00 emissions the p.m. from ground. local the harborvicinity. time. Sampling area times Possible and local volatiles varied from sources laboratory include activities tra in the Station in List/Sylt, Northern Germany (55.0226 2.2 Standards and samples A Scott EPA TOwas 15/17 used standard containing asof 65 a chloromethane, compounds (1 working bromomethane, ppm iodomethane standard. each and in nitrogen) dichlorodifluoromethane For (each further tests single component standards Briefly, up to 500 L of125 air were drawn throughsis. a A cryo direct analysis trap. of The thedue trap isotopic to composition was of multiple then the interferences heated target from compoundspre-separated to other is on not compounds. a possible Therefore, GasPro the column samplescryo and were the traps. first target compounds For were carbonPorabondQ recollected column on isotope and two determination analysed on each a fraction GC-C-IRMS/MS was system. then separated on a 2 Methods 2.1 Overview (Hamburg, Germany) and a coastal site (Sylt, Germany). and report first results from its application within a survey on air samples from an urban et al., 2008) but notiodomethane for with bromomethane typical and mixing other ratios halocarbons between such 0.5 as and bromoform 10 and pptv. for atmospheric bromomethane (Bill et al.,was 2004) used. in which a cryogenic sampling system appropriate for field work enablingcarbons the isotope in ratio the determination low ofHere, pptv atmospheric we range, halo- describe notably the bromomethane, configuration bromoform and and validation iodomethane. of the developed sampling system been used. Given acarbon minimum isotope carbon analysis, amount thesechloromethane between sampling (Rudolph 0.25 methods ng et are and al., su Archbold 1 1997; ng et al., required Tsunogai 2005; for et Redekerrocarbons al., et and 1999; hydrofluorocarbons al., (Thompson 2006; Thompson et Mead al., et 2002; et al., Redeker al., et 2002; 2008) al., and 2006; Mead several chlorofluo- analysis of dissolved halocarbons (Auerapproach et to al., short 2006), which(CH lived now halocarbons allows such to extend asratios this of iodomethane these (CH compoundsbe could derived provide from valuable the determination additional of information mixing that ratios. canters not (Rudolph et al.,al., 1997; 2005; Tsunogai Redeker et et al., al., 1999; 2006) Thompson and et adsorbent al., tubes 2002; (Mead, Archbold 2008) et have successfully origin and fate of atmospheric chloromethaneand (Harper bromomethane et al., (Mc 2003; Cauly Kepplerthe et et known al., al., 2005) sources 1999); cover a asand broad marine the range sources reported of to carbonet isotope al., ratios 2005). of Tremendous progress has recently been made in the carbon isotopic 5 5 20 15 10 25 15 20 10 Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | . C 1 at ◦ − 1 − C reducing the ◦ (Christoph et al., 3 ) for 30 min at a temperature 1 − C using a Peltier cooler to prevent ◦ ) or pre cleaned ambient air. Before 5 1 were rapidly released into the gas − − 2 2166 2165 and N 2 C were completely transferred from the cryo trap to ◦ ), a restriction capillary (restriction flow 50 mL min 1 − C until analysis. ◦ 80 C provided by a dry shipper. ◦ ≤ − 100 mL min > 110 C. ≤ − ◦ ect the recovery of high boiling compounds such as CHBr ff by a Nafion dryertransfer (Perma the Pure adsorption Inc.NJ, tube USA) wasbreakthrough placed cooled of in to the silica analytes. gel. Underboiling During points these sample as conditions high non-polar as compoundsthe 150 with adsorption tube, while polaraldehydes, water were soluble fractionated compounds, between suchthe the as adsorption alcohols water tube. and After remaining theand in transfer stored the of at samples cryo the trap adsorption and tube wasAfter closed sample transfer, valve 3 wasthe closed cryo and trap valve with 5 a was stream opened of for nitrogen conditioning (1000 mL min After sampling, valve 1was and opened. 2 weretrapped The closed cryotrap argon and was and valvephase. carefully 3 traces removed to To of from prevent thea O the analyte adsorption flowrate losses dry tube during shipper.2 bar) sample was transfer Co- placed (usuallyof behind observed highly valve at volatile 2. analytesAfter such 20 Without min, as this valve chloromethane 4 restrictionand and was flushed we opened dichlorodifluoromethane. with observed for either 40 nitrogen losses entering min, (50 mL the the min cryo adsorption trap tube, was heated water to vapour 125 was removed from the gas stream of 125 During sampling thewere valves drawn 1 with a and membranemany) 2 pump through (KNF were Neuberger the open. N86 sampling KNDC, Freiburg, system Ger- 300 with to a 500 L flowrate of between ambient 3 air and 5 Lwas min then directed throughwater a vapour condenser of kept the at air.a approximately It 5 has previously been2002). shown Finally, the that target a compounds condenserture were does of enriched not in the cryotrap at a tempera- The flowrate and the(Omega FMA-1608A, sampling Deckenpfronn, volume Germany). werefilter (Sartorius, monitored The Teflon with membrane air filter, a first diameter: mass passed 45 mm, flowmeter a pore particle size: 0.2 µm) and The sampling consists of three consecutive steps: (1) the cryogenic adsorption of 72 h back trajectories were calculated by Hysplit 4.8 for an arriving height of 30 m 3. 2. 1. systems (Miller et al.,technical infrastructure 2008). limiting the However, use both of approaches such require sampling a systems in well field established campaigns, 2.3.1 Cryofocussing and design of theClassical cryotrap cryogenic sampling systems useextraction of either the liquid target nitrogen compounds from orweg air liquid et (Bill argon et al., al., for 2004; 2011). the Pupek et More al., 2008; recently, Zuider- a cryostat has been applied in cryogenic sampling the target compounds fromtrap ambient to air, an (2) adsorption the tube and transfer (3) of the the conditioning analytes of from the the cryo trap. cryo prevailing marine and coastalcontinental influence influence for for the the samples samples taken taken at at Sylt Hamburg. and2.3 a prevailing Sampling system The sampling system (Fig. 1)of was the designed isotopic with composition respect of to theparts target field compounds. suitability were and To made avoid integrity contamination, of all stainless system steel or silanized glass. 02:00 p.m. local time.salt Potential marshes, local tidal sources flats for and, to halocarbons a include minor emissions degree, from harbor(Hamburg activities samples) in and List. 2System m (GDAS) (Sylt data (Draxler samples) and using Rolph, NCEP’s 2011). Global The backward Data trajectories Assimilation indicate a away from the Wadden Sea Station. Sample times varied between 10:00 a.m. and 5 5 15 10 25 20 15 10 25 20 Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | 2 1 − , 80 mg ® C for flowrates ◦ . and most of the 2 2 110 − inch, 50 cm length) with C. We found this suitable ective temperature inside ◦ 3 / ff 3 inch outer diameter, 7 inch inch stainless steel tube that C and a flow of 100 mL min nity for CO 4 ◦ 4 / ffi / 1 1 , 215 mg Carbopack X 569 ® C for 28 days in the gas phase above the ◦ , such as Ascarite or Soda lime (Rudolph 2 C the space between both tubes is filled with 176 ◦ 2168 2167 content of the samples. The remaining CO − 2 TA in order of the sampling flow direction. The 110 − ® must be removed. Chemical traps frequently used in 2 ¨ uhlheim Germany) at 320 and water 2 ) present in ambient air. Water was removed from the sample 3 − , Gerstel, M 2 / 1 . 1 oad to the cryo trap. With this setup we observed elevated levels of − 1003 and 9 mg Tenax ) further reducing the CO ffl 1 − ® from the cryo trap passes the adsorbent tubes during sample transfer. Prior to 2 Nafion dryers have the potential for artefact formation.During So first far, tests, transformation we of attached the Nafion dryer in front of the cryo trap to reduce Chemical traps such as magnesium perchlorate or phosphorus pentoxide are not The cryo trap consists of an outer stainless steel tube ( To achieve complete trapping at ¨ usseldorf, Germany) as a cooling source. A dry shipper is a dewar that contains an tests revealed that the sulfonic acid groups of the Nafion membrane can catalyze the Nafion dryer placed inmeans silica of gel. a second Nafion Remaining dryer traces during of analysis. watercarbonyl were finally compounds removed to by triple alcohols bonds and have been reactions reported of (Miller et compounds al., containing 2008). the double water or o bromomethane. Parallel runscial with formation of and bromomethane without on the the Nafion Nafion dryer membrane indicated of an up artifi- to 10 pptv. Further at three stages duringsampled sampling air and was reduced analysis. byto means First of prevent the a clogging condenser water kept vapour ofmoisture at pressure 5 the was in cryo frozen the fraction out trap of for in the air the water samples was cryo mobilized up trap. from to the During 500 cryo L. trap. the The sample This remaining water transfer air was only removed a by a small adsorbents used in theCO adsorption tubes have aanalysis, low the a adsorbent tubes20 mL were min flushed withwas helium removed during at the room GC temperature pre-separation by (2 an min, initial flushingsuitable step. for water removal in high-volumewater sampling (up systems to due to 30 mL the m large amount of Before analysis water andtrace CO gas analysis to removeet excessive al., CO 1997;halides Redeker especially et of polybrominated al., compounds 2007) and reduce thus the have been recovery ruled of out. polyhalogenated The methyl 2.3.3 Removal of CO conditioner (TC of nitrogen. glass rings were mainly designated for trapping water2.3.2 vapour and CO Adsorption tubes The adsorption tubes werelength) made and of filled stainless with steelCarboxen 77 ( mg Carboxenfilling 1016 was fixed with stainlessing steel the plugs. adsorbents. Silanized Before glass use, wool the was packed used tubes for were separat- conditioned for 24 h in a tube adsorbents separated by plugs ofbottom precleaned silanized the glass package wool. of Fromraschig the the rings; top trap to 25–35 was the cm as47 silanized cm follows: glass Porapak 0–20 beads N; cm: 60–250 47–50 empty; µm; cm: 20–25 35–41 cm cm empty. silanized Tenax The Ta; 41– empty space at the top and the silanized pling in remote placesthe where stopcock no of liquid the nitrogen drywas is shipper drilled available. was into shortened For the to thethe 10 stopcock sampling cm cryo to and system trap insert a depended hole the onup of cryo to the 20 5 mm flushing trap. L diameter min flow The rate e and was below an air inlet at theis side connected 4 cm to below the the sampling top pump and (Fig. an 1). inner cal cryogenic sampling systems weD employed a dry shipper (Voyageuradsorbent 12, taking Air up Liquide, liquidnitrogen nitrogen. and provides The a temperature Voyageur 12 of adsorbed can liquid adsorb nitrogen. up It to has 14 L an of approval liquid for air transport allowing cryogenic sam- especially for field work in remote areas. In order to overcome the limitations of classi- 5 5 25 15 20 10 25 15 20 10 Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | σ 15% on ± 2‰) via an . 0 ± 8 . 26 − . Water was further removed by a 2 C, hold for 10 min. ◦ . The oven temperature program was as 1 − to 240 2170 2169 1 − ), directed through a Nafion dryer and refocused 1 . of the flow were transferred to an DeltaV isotope − 1 − C min ◦ C and separated on a GasPro column (Agilent, 30 m, ◦ reference gas (Air Liquide, Germany, 2 intensities against chloromethane and bromomethane as in- 2 C (15 min, 25 mL min ◦ C, hold for 5 min.; 6 ◦ and then ramped to 2.7 mL min 2 level. σ All isotope values are reported relative to the Vienna Pee Dee Belemnite (VPDB) Prior to sample analysis, the performance of the IRMS was evaluated by repeated About 20% of the sample were directed into the MS via a split for monitoring the ternal standards. Thethe overall 1 uncertainty of this procedure is estimated to level. 2.4.3 Identification, quantification andThe purity analytes control of were VOCs identifiedwith by known comparison standards. of theirtained Further mass retention compounds spectra time were with and identified thetification mass by Nist was spectra mass comparison done spectral of on databaseTOC the the version EPA ob- 15/17 2.0. Agilent standard. Primary systemon quan- Compounds used the quantified for IRMS against pre-separation via a against the standard CO the were Scotty quantified to monitor the repeatability. Resultsquality are criteria: only reported (i) forvalley. peak peaks purity that met better the than following 90% (ii)scale peak and separation within better the lab than uncertainties, 90% unless otherwise stated, are reported on the 1 spectrometer (Thermo Finnigan DSQ II)80% for were monitoring directed to the a peakThermo commercial purity. combustion Finnigan) The interface remaining converting (GC-combustion interface the III; Nafion analytes dryer to and about CO 0.5 mLratio min mass spectrometer via an open split. injections of a certified CO open split. The Scott TOC EPA 15/17 standard was used as a daily working standard For the isotope ratiogan GC-C-IRMS/MS determination system each (Trace GC fractionDSQ II) II; was system combustion using analysed Interface a CP-PorabondQ III; onaration column of DeltaV (Varian, a the 25 IRMS m, analytes. Thermo 250 and Again, µm Finni- i.d.) 20% of for final the sep- sample were directed to a Quadrupole mass 2.4.2 Isotope ratio determination 0.32 µm i.d.) with heliumthe as a CO carrier. The flowratefollows: was 40 set to 5 mL for 4 min to remove fractionation. Thevalve remaining (Vici, 80% Schenkon, of Swiss)0.32 the id., and sample 25 cm recollected were length), onthe one directed remaining two containing components. to the cryo target a traps compounds Valco (Silica and the six steel other port 1/16#, containing The pre-separation was performedGermany). on an The 6890N/5975B analytes GC-MScarrier (Agilent, were at Waldbronn, 350 desorbed fromon the a quartz adsorbent capillary tube (0.32 i.d,analytes into 60 cm were a length) desorbed immersed helium in at liquid gas nitrogen. 25 Afterwards the A direct analysis of thedue isotopic to composition of multiple the interferences targetpre-separated from compounds on other was not a compounds. possible Gas-Protwo Therefore, cryo column the traps. and sample For the was carbona first target isotope PorabondQ determination compounds column each were and fractionanalytical recollected was analysed set then on up on separated is on a shown GC-C-IRMS/MS in Fig. system. 2. 2.4.1 A scheme of Pre-separation the Nevertheless, with the Nafion dryercial behind formation the of cryo bromomethane. trap we did not observe2.4 any artifi- Carbon isotope ratio determination nucleophilic substitution of methanol to bromomethane in the presence of bromide. 5 5 25 15 20 10 20 15 10 Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | 6) = 004). I and n . 3 0 = p 025) and for . Br, CH 6; 0 3 = = n 05) revealed usu- C. However, there 5%, it is expected p -compounds were . ◦ 3 ± 0 6, 6; . 0 = p < ± n ect of the pre-separation 3 . ff 8, . 22 2 − ± 8 . 33 − -compounds and C 2 2172 2171 ciency of the sampling system 0‰; pre-separation: ffi for 100 min (total volume 500 L). Extraction was done ± ciency was tested for the entire sampling procedure 1 erences between the recovery rates for direct injection 0 -compounds, C . 6; pre-separation: ffi − ff 1 = 24 n 6) ranged from 0.5‰ for propene and benzene to 2.8‰ for − = n 1‰; . erence between both procedures has been observed for bromo- 3 ff ± 6 10 pptv), we determined the linearity and reproducibility of the carbon . < 38 − (usually 3 Standard deviations ( The results of these tests are given in Table 1. The recovery rates for the entire erage standard deviation for C 1.5‰, 1.1‰ and 0.9‰,ally respectively. A no Mann-Whitney-U-test significant ( influenceever, of a the significant di pre-separation onform (direct: the carbon isotope1,1-dichloroethane ratios. (direct: How- to the isotope ratiostopic obtained composition without pre-separation. could onlyand Without be pre-seperation peak determined the overlaps. iso- fordetermined. 7 After The compounds pre-separation, results because for isotope these of ratios compounds coelutions are for displayed 31 inbromoform Table compounds 1. and generally could decreased be with increasing numbers of carbon atoms. The av- 3.2.2 Reproducibility of the analytical system Since the sampling procedure showedto an be excellent free of recovery analytical of artefacts.termination 98 Thus, was the only reproducibility of tested the forwas carbon the tested isotope ratio on analytical de- the system. 0.2standard The nmole were level e with injected into the the Scottcarbon GC-MS TOC isotope EPA system, 15/17 pre-separated ratios standard. and as 5 analysed mL described for of stable in the Sects. 3.41 and 3.42. Results were compared 1 nmole the standard deviationfor increased and bromomethane was for 2.6‰ 0.02al. for nmole chloromethane (1997) each. and and 1.8‰ Redekeramounts This et of al. corroborates carbon. (2007) the who study observed of similar Rudolph deviations for et comparable for carbon amounts between 20 and 1 nmole were usually below 0.3‰. However, below 1000 pptv for asystematic 500 L bias sample. for carbon The amounts carbon down isotope to ratio 0.02 nmole. determination Standard was deviations free ( of a 3.2.1 Reproducibility versus concentration Due to theCHBr low mixing ratiosisotope of ratios some for target low levelsinjecting compounds of chloromethane such carbon and in as the bromomethaneamounts CH sub-nmole into ranged range. the from This GC-IRMS. 0.02 was The to performed by 20 injected nmol carbon corresponding to mixing ratios between 1 and increasing boiling points for compounds withare boiling no points statistically above 80 significant di on the adsorptionconclude tubes that and no significant those losses for of the the target entire analytes3.2 occur sampling during process. sampling. Reproducibility of the Therefore, carbon we isotope determination iodomethane standard into a streamthe of cryo nitrogen trap and at pre-concentrating a the flowas of analytes 5 described on L min above andof the the recovery standard rates mixturequantification. onto were the calculated Agilent against GC-MS direct system injections used forsampling pre-separation procedure and averaged 97.8% (range fromranged 87% from to 1% 110%) and to the 7%. reproducibility Our data indicate a slight decrease of the recovery rates with 3.1 Trapping and desorption e The trapping and desorptionon e a 0.2 nmole level by injecting 5 mL of the Scott TOC EPA 15/17 and 50 µL of the 3 Results and discussion 5 5 15 20 10 25 20 15 10 Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | ± C- 4 . 13 30% δ ± 25 erence − ff 3‰ and . 0 erences in ff ± 3‰). Furane C value was C values. As . 1 . 2 13 13 δ ± 18 δ 2 C by 3.0‰ relative − . 13 17 − 57.1‰. 87 pptv for the coastal site) − 8‰ in the urban samples and . ± C-values of 0 erences in the 13 ff ± δ 1 . C-values. Vinylchloride was strongly 34 erences. Even though the average mix- 13 − C value of ff δ 13 δ 2174 2173 : The average mixing ratios of dichlorodifluo- C value of 0.5‰. In contrast, hexafluoropropene 3‰. DME was depleted relative to the saturated C-value of 13 . 1 δ 13 ered by roughly one order of magnitude (urban site: ff δ ± 0 . C with average 29 erence between the sites. The average ff − 13 60 pptv for the urban site and 554 25.0‰ with individual standard deviations between 0.8‰ to 2.2‰. − ± C with an average ns, which is conform to previous studies (Redeker et al., 2007). Propa- C values of 13 ffi 13 2‰, respectively. Propene, the only unsaturated hydrocarbon that could be . δ 8 c related sources at both sites and therefore obtain similar isotopic signatures. 31.8‰ to ± ffi C thus points towards local sources rather than towards propene transported over − 6 37.2‰ at the coastal site. . 13 C-values neither between the alkanes nor between the two sites. This is in line with − ected the carbon isotope ratio determination of butane. However, the carbon isotope Dichlorodifluoromethane (CFC-12) We also could determine carbon isotope ratios of several oxygenated compounds On average, the alkene, propene and 2-butene were enriched in δ 6‰) were mainly in the same range as those of the alkenes with the exception of 13 13 . ff romethane (614 showed no significant di 3.3.2 Organohalogens The mixing ratios of the long-livedof CFCs and the chloromethane generally atmospheric feltorganohalogens within background covered levels. a broadenriched in range The of averagewas carbon strongly depleted isotope in ratios of the 1 propenal that was stronglyshowed enriched in the urbanhydrocarbons air showing samples a ( mean of determined at both sites,the showed atmospheric degradation no of propene site by OH-radicalsfractionation specific is factor di assigned of with 11.7‰ a (Rudolph considerable etin al., 2000), the lack ofa a long site distance specific from di urban to coastal areas. including dimethylether (DME), furane,based propanal on and their propenal, mass-spectra. which were The identified carbon isotope ratios of the aldehydes ( cal tra to the para diene and 2-butyne were− even more enriched with OH-radicals (Rudolph et al., 2000). Furthermore, the alkanes most likely stem from lo- δ the small fractionation factors (1.41–3.44‰) reported for the reaction of alkanes with Except for butane thatpounds co-eluted were with well 1,2-dichloro-1,1,2,2-tetrafluoroethane, separated. thethe com- The butane portion carbon of wassamples 1.2-dichloro-1.1.2.2-tetrafluoroethane less from to the than coastal 2%a site in this urban portion samples amountedratios and from to both therewith 5 sites negligible. to showeding no 20% significant ratios In and di between thus both may304–1620 sites pptv; have di coastal site: 11–193 pptv), we observed no significant di The carbon isotope ratios obtainedvious for studies the (Rudolph et hydrocarbons al., agreevalues 1997; with Tsunogai the of et results al., the of 1999; alkanesfrom Redeker pre- et propane, al., 2007). butane, isobutane, pentane and isopentane ranged determine carbon isotope ratios offor 37 chlorobenzene compounds and with up mixingbons to ratios in between 1600 pptv urban 0.3 pptv for airvented propane. the samples carbon caused The isotope stronglychloroethane, high determination iodomethane, tailing amounts of chloroform, peaks and several of bromoform organohalogens. in hydrocar- ride in all the and This and 1,1,2-trichloro-1,2,2-triflouorethane applied IRMS for in for carbon and some tetrachlo- of thus the pre- urban air3.3.1 samples. Hydrocarbons and oxygenated VOCs The carbon isotoperesults ratios from and previous studies mixing arepounds ratios presented which in of Table could 2. the betheir Values ambient mass are clearly spectra air only identified and given samples which for either met com- as by the well quality comparison criteria as with outlined standards above. In or total, by we could 3.3 Ambient air samples 5 5 25 15 20 10 20 25 15 10 Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | ± C 0 13 . 2‰; . δ ( C val- 0 38 C value ± 13 − C values C values 2 δ 13 . 13 13 C value of δ 3 pptv at the δ δ 41 13 C determina- ± − δ 13 cient for the car- δ ffi 24‰) (Auer et al., − 9‰) and the coastal . Fucus serratus 1 ± C of 36 pptv for the urban and 9 . 13 C values and of mixing ra- erence in the mixing ratios ± ff δ 13 39 ( C values was only possible in δ − 13 δ 3‰. This isotopic enrichment by . 0 18.2‰). The isotope ratios reported ± − 0 36‰ for the marine background in the . erent halophytes which correspond with 9‰ from a coastal and an urban area in − . ff 31 3 − C values of bromoform in the coastal sam- ± C value average over both sites was 13 C of 1 2176 2175 . level. This is close to the natural variability in δ 13 13 δ 37 σ δ − 13.8‰ and Dunaliella tertiolecta − C values at the urban site (coastal site: 40.8‰. Given the respective mixing ratios of 1.9, 1.0 13 − δ C-values of bromomethane in the urban samples were erence between the urban ( 13 ff 22.9‰, δ C value of − 13 C-values averaged 4.1‰ on the 1 8‰). δ . 13 ± 1‰). Our data are in the range reported by Redeker et al. (2007), . 7‰ from a suburban site in Berkely, USA (Bill et al., 2004). At 0 δ 6‰ ( . 4) being in excellent agreement with the only previous reported 1 . : The : Chloromethane mixing ratios were 524 1 ± 4 3‰ were reported by Redeker et al. (2007) from Belfast, Ireland. ± = . 40.4‰, and 5 1 pptv at the coastal site. However this di ± : Iodomethane coeluted with carbon disulfide, dichloromethane and . 9 ± 2 n − . 7‰). The values mirror previously published results. Tsunogai (1999) 0 ± . : A reliable determination of bromoform 3 . ± . 33 0 37 − 43 ± − 18 39 7) with a slight di 3‰ ( − 6‰ from a remote site in the arctic (Alert, Canada). Slightly more depleted C values were determined in more coastal influenced air masses. A potential 2 . − − . . = 11‰ for iodomethane emitted from di 3 1 13 79.8‰, n ± δ 36 ± ± 30 pptv for the coastal site. The − − 5 4 15‰) and the planktonic algae . . 47 ± − − Bromoform Iodomethane Iodomethane belongs to the few compounds revealing a strong within site variation Bromomethane Chloromethane 1‰ ( 41 39.6‰ with slightly enriched 36 . of ples was here were corrected forDue the to isotopic this shift correction observed wetion for estimated of the the bromoform standard overall to (see reproducibilitythese Sect. for samples 4.2.2). the and itues thus remains for unresolved bromoform weather reflectsAs the the for variability iodomethane, natural of there variability the areof or no atmospheric simply literature bromoform. the data analyticalform However, available similar on uncertainty. produced the isotope in isotopic ratios composition incubation were reported experiments for by bromo- the brown algae can currently not substantiateof the the reasons carbon for isotope the ratios. observed large within sitethe variation coastal samples. In theeluting urban C8 air hydrocarbons. samples the The determination average was hampered by co- ratios of atmospheric iodomethanedata available from to incubation experiments compare andof with. greenhouse experiments But revealed ownthe unpublished more enriched atmospheric valueswas in determined our in study. The airriched strongly masses depleted coming fromimportant the source open in open North oceansof Sea currently iodomethane under while in discussion the the is the sea more photolytic surface en- formation layer (Moore and Zafirou, 1994). Nevertheless, we and 4.9 pptv no cleartios relation becomes between evident. variations ofa Iodomethane the has few a days relative (Harpercarbon short and isotope values, atmospheric Hamilton, more lifetime likely 2003).may of provide explain a only Therefore the snapshot lack both, than of an the any integrated correlation. mixing signal, which There ratios are and no the literature data on carbon isotope were of the carbon isotope ratios. For the three coastal samples determined, is mainly driven bycannot one state urban a sample systematic showingratios. relation an between elevated the mixing carbon ratio isotope and ratios thus and we propenal. the Although the mixing mixing ratiosbon in all isotope samples ratio were determination generally su of we iodomethane were in only the able coastalfor to the samples determine fractionation after carbon of a isotope the careful samples. ratios adjustment of the time intervals − values of the coastal site the 10‰ is accompanied by a decreaseurban of site the to average 7 mixing ratios from 10 620 4 site ( and co-workers reported ansubtropical average Pacific.− Thompson et al.values (2002) of determined an average − urban site: who gave an average Ireland, but are depletedBristol, in UK comparison ( to those reported by Mead et al. (2008) from 5 5 25 15 20 10 25 15 20 10 Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | of ε 6‰) . 4 of 21.5 C value ± ε 13 2 . δ C values of 44.8‰. The 53 − 13 − δ 1‰ it was strongly . C of erence between the 2 13 30.1‰ for industrially C value for 1-chloro- ff ± δ − 3). Anyhow the mixing 13 5 . 7‰) were comparable to 33.1‰ and δ = . 0 8 for the coastal site. This − 1 n − ± ± 5 3‰. The isotopic signal in the . . 5 36 1 pptv, C value of ± 33.8‰, . − 44.8‰ was observed in westerly air erent soils indicate a large 5 2 − 13 C value of carbontetrachloride was . ff − C between the sites were observed. δ ± 13 C value of 36 13 0 δ 6‰ for the coastal site. As for dichlorod- . 13 δ − . : The average δ 3‰). Our values from the coastal site re- 2 . C at the urban site ( ± 3 C is surprising as it was not accompanied by C values ( 5 13 28‰. 2178 2177 ± . ) of this reaction has not been determined yet. 13 − 13 2 C value of ε . 31 δ 13 − 44 8‰ without any significant di C value of − erences in . ff C value of dissolved bromoform in a sea water sample 2 C value of vinylchloride may result from its rapid atmo- 13 δ 13 ± 13 6‰ for Belfast (Ireland). More enriched δ . δ 6 . 5 : The isotope ratios of trichlorodifluoromethane were 4‰ given for chloroform in that study. In contrast, the car- : In contrast to dichlorodifluoromethane, chlorodifluo- . The average found ± 24 C values analysed were 6 4‰. 9 C. − 13 ± . . 7 pptv and the 4 13 δ 4 . C values of chloroethane could only be determined in the sam- . 42 0 ± 13 5 pptv. With an average − . 3 37 ± 3) and average mixing ratios were 104 Vinyl chloride was only detectable in the urban air samples with mix- δ . 3 0 − = : . ± 27 n C for chloroform occurred not in the same sample as the depletion in : Chloroform stable isotope ratios could only be determined in the 3 − C as compared to the other chlorinated compounds. To the best of our erences in the mixing ratios (10 . 13 ff 2‰. The average mixing ratio of carbontetrachloride in the urban samples 13 . 1 9‰ ( 0‰, have been reported from Bristol, UK. (Mead et al., 2009). Interestingly, 3‰ for the urban site and 25 pptv with a mean . . . ± 1 1 5 C value of ± 1 . ± ± ± 13 9 5 9 C for chlorodifluoromethane in northerly air masses as compared to westerly air δ . . . 27 C value of 13 − Trichlorofluoromethane 1-Chloro-1,1,difluoroethane (HFC-142b) Chlorodifluoromethane Carbontetrachloride: Chloroform Vinyl chloride: Chloroethane C for iodomethane. This depletion in 33 29 28 13 1,1,difluoroethane was Redeker et al. (2007) observedof a slight although statistically notmasses significant and enrichment air masses from Europe. − ifluoromethane, no significant di Our values corroborate theδ results of Redecker et al. (2007) who gave an average romethane was significantly depletedas in compared to thesemble coastal those site reported ( inage the study of− Redeker et al. (2007) who provided an aver- − agrees well with the resultsof of Mead et al. (2008)was who 92 reported an average urban air samples mightin be the influenced target by fraction incomplete (85 and recovery 90%) of and carbontetrachloride thus has to be taken with great care. ratios found here forthose chloroform reported were by by Redekerinto almost et the two al. range orders (2007) of of butbon isotope magnitude the ratios lower carbon reported than isotope forthan chloroform ratios 10‰ by found Mead enriched et here in al. fell (2008) were on average more significant di depletion in 13 spheric degradation and/or evasion ofsuch isotopically as enriched landfills. vinylchloride from sources coastal samples due toin strong the chromatographic urban interferences samples. from 2-methyl-1-butene relative enriched isotope ratios wereEast observed respectively in while air the masses depleted frommasses the that North passed and along North the Dutch and the German coast. It is noteworthy that this As the OH-radical attacks11.4‰ the reported double for bound the atmospheric it2000). degradation may In of addition, be propene incubation in by experimentsto OH-radicals the with (Rudolph, 26.0‰ di same for order the asThus, this the microbial extraordinary degradation high of vinylchloride in soils (Bloom et al., 2000). ing ratios were 2 those of chloromethane. ing ratios of 5 enriched in knowledge, no isotopic data foron atmospheric our vinylchloride reference are standard soproduced one far vinyl might published. chloride. assume Based In a theOH-radicals. atmosphere, vinylchloride The is fractionation rapidly factor degraded ( mainly by 2006). In the same studyfrom the the Baltic Sea was determined to ples from theimpeded coastal a site. reliable carbon isotope In ratio the determination of urban this air compound. Average samples mix- the tailing of the butane peak 5 5 25 15 20 10 25 15 20 10 Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | 3 C 13 3‰) . C en- 3 13 ± 3‰) and 2 . . 5 44 ± − 5 . C was observed 13 : Unusually 10‰ (Miller et al., 2001). Si- ). Interestingly, we observed the ± 3 erent fractions, we obtained an aver- ected by sulphur, we presume these ff ff of 45 ε 46 signal For the coastal samples, where erences between the urban and the coastal 2180 2179 m/z ff 4‰. Again a reanalysis of our data revealed sub- . 0 45 and ± 0 . 0‰. and found no indication for interferences. As CBrF 6‰) as compared to air masses from western Europe . . 46/44 ratio is still a m/z 5 86 10 C at the coastal site were not accompanied by a significant − 2 pptv) this huge discrepancy is rather due to the analytical ± . ± 13 m/z 6 0 . 1 . ± 39 7 46 signal. As both compounds elute shortly after carbonylsulfide 39 . − 3‰). A thorough reanalysis of these data revealed interferences on − . 3 m/z C value of ± 3 13 . δ ect for bromotrifluoromethane, eluting 15.3 s after carbonylsulfide. In the ur- 2‰). As mentioned before, our data from the coastal station ( ff . 8‰ (King and Saltzman, 1997) and the degradation of bromomethane by 5 45 and ± C value of ± and carbonylsulfide were recollected in di 13 5 3 . δ m/z erences are caused by the air sea exchange of these compounds in concert with 42 These enrichments in The 72 h backward trajectories indicate a prevailing marine and coastal influence Pentafluoroethane, norflurane and bromotrifluoromethane of 69 ff − sume the bromomethane emitted back into the atmosphere to be isotopically enriched partial degradation in surface oceansfrom both, altering intrinsic the sources isotopic and signatures.in from marine Bromomethane the surface atmosphere, waters is by knownof biotic to and up be abiotic rapidly to processes degraded 20% withotic overall per degradation degradation rates day due (Kingε to and Saltzman, hydrolysis 1997, andmethylotrophic and transhalogenation bacteria references is is therein). assignedmultaneously, assigned bromomethane The with is with abi- an produced a incomposition large the of surface the water. intrinsic bromomethane Although is the unknown, isotopic it appears reasonable to pre- the marine boundary layersuppose e.g. atmospheric due degradation to processes reactions notferences. to with This be chlorine also the radicals. decisive holds factor trueat Therefore, for for the these we bromomethane, urban dif- as site is the caused elevateddi by average only mixing one ratio single measurement. We assume these isotopic mote sites correspond to previoustherein; studies Redeker (Goldstein et and al., 2007). Shaw, 2003,et Although and al. statistically (2007) references not indicated significant, a the smallthe data enrichment of North for Redeker chlorodifluoromethane Atlantic in air ( ( masses from resemble the latter. In contrast, our data from thedecrease urban site of are the enriched mixing by ratios 9.0‰. which would point towards an enhanced degradation in for the samples takentaken at at Sylt Hamburg. andal. The (2007) a from comparison prevailing Belfast (marine continental ofrichment influenced) influence our of provide these further for data compounds indication the in with forof marine samples an those and/or chloromethane isotopic remote reported en- by air 3‰ masses. by A to Redeker slight 4‰ et enrichment at urban and rural sites compared to marine and re- triflouroethane (6.7‰) intrichloro-1,2,2-trifluoroethane coastal were samples. less enriched (3.6‰ Concurrently, and chloromethane 4.0‰, respectively). and 1,1,2- 3.3.3 Variability of the carbon isotopeOur ratios data reveal considerablesampling isotopic site di for several compounds.for A 1,1-dichloro-1-fluoroethane pronounced (10.0‰) enrichment andmixing in propenal ratios of (7.6‰) both in compoundspared urban were significantly to samples. elevated the at The the urbantowards coastal site a site strong (22.6 as coastal pptv com- or versus marinewas source. 8.0 pptv observed In contrast, and for a 96 pronounced bromomethane pptv enrichment (10.5‰), versus in chlorodifluoromethane 6 pptv) (9.0‰), pointing and 1,1,1- age has an average atmosphericwere lifetime comparable of (3 65interferences years than and to the atmospheric degradation mixing or ratios source-related for processes. both sites interferences to result fromtion the interface formation or of inratio fluoro-sulfur-compounds measurements the in of ion bromotrifluoromethane the (CBrF source. combus- opposite e This isban further air samples supported it by wasan recollected average the in carbon the same isotope stantial fraction interferences as on carbonylsulfide the andCBrF yielded riched carbon isotope ratiosnorflurane were (4 observed forthe pentafluoroethane (16 in an interval where the both sites. 5 5 25 15 20 10 25 15 20 10 Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | erences ff , 2002. 3‰ are in ex- . , 2006. 3 ± , 2004. 5 , 2005. . 41 − http://ready.arl.noaa.gov/HYSPLIT. C Stable Isotope Analysis of Atmo- 13 members Ralf Lendt, Sabine Beckmann δ ff doi:10.1002/rcm.1791 doi:10.1023/a:1015592115435 doi:10.1029/2003gl018639 2182 2181 doi:10.1016/j.chroma.2006.07.043 , 2000. C by about 10‰ in the coastal samples. A similar , 2002. 13 The authors of this article would like to thank the BMBF for funding this doi:10.1021/es991179k ) for carbon isotope determination on the 0.2 nmole level ranged from σ C values for bromomethane from the urban site of 13 doi:10.1029/2001gl012946 δ ), NOAA Air Resources Laboratory, Silver Spring, MD., 2011. C values were enriched in jectory) Model access viaphp NOAA ARL READY Website ( spheric Oxygenated VolatileMass Organic Spectrometry, Anaytical Compounds Chemistry, by 82, 6797–6806, Gas 2010. Chromatography-Isotope Ratio pean estuaries, Biogeochemistry, 59, 143–160, flow system for the stable carbonin isotope analysis water, of J. volatile Chromatogr. halogenated A, organic 1131, compounds 24–36, methyl bromide, Geophys. Res. Lett., 31, 4, ation during Microbial DechlorinationChloride:? of Implications Trichloroethene, for cis-1,2-Dichloroethene,2768–2772, Assessment and of Vinyl Natural Attenuation, Environ. Sci. Technol., 34, bromide and methyl chloride1045, emitted from a coastal salt marsh, Geophys. Res. Lett., 29, stable isotope analysisRapid of Commun. methyl Mass halides Spectrom., 19, and 337–342, chlorofluorocarbons at pptv concentrations, We reported isotope ratios for a broad range of VOCs in urban and marineThe air in 13 Goldstein, A. H. and Shaw, S. L.: Isotopes of volatile organic compounds: an emerging ap- Draxler, R. R. and Rolph, G. D.: HYSPLIT (HYbrid Single-Particle Lagrangian Integrated Tra- Giebel, B. M., Swart, P. K., and Riemer, D. D.: Christof, O., Seifert, R., and Michaelis, W.: Volatile halogenated organic compounds in Euro- Bill, M., Conrad, M. E., and Goldstein, A. H.: StableBloom, carbon Y., isotope Aravena, composition R., of Hunkeler, atmospheric D., Edwards, E., and Frape, S. K.: Carbon Isotope Fraction- Auer, N. R., Manzke, B. U., and Schulz-Bull, D. E.: Development of a purge and trap continuous Bill, M., Rhew, R. C., Weiss, R. F., and Goldstein, A. H.: Carbon isotope ratios of methyl Archbold, M. E., Redeker, K. R., Davis, S., Elliot, T., and Kalin, R. M.: A method for carbon References work within the framework ofthe Sopran support (BMBF grant of 03F0611E)sampling We Justus also and van gratefully the acknowledge Beusekom supportand from Peggy of Bartsch. the our AWI the Wadden technical Sea sta Station in Sylt during the Acknowledgements. as a promising toolthese for compounds. better constraining the role of the ocean in the global cycle of carbon isotope ratios found here are consistent with those previouslycellent reported. agreement with those reportedδ by Bill et al.but (2004). less In contrast, pronounced bromomethane trend1-chloro-1,1difluoroethane and was chloromethane. observed We for hypothesizeare that trichloroethane, related these to chlorodifluoromethane, di atmosphere-ocean exchangedation with processes fractionating in biotic the and surface ocean abiotic and degra- suggest carbon isotope ratio determination ter than 2‰. Nevertheless, theis determination often of hampered these compounds by in high urban loads air of masses hydrocarbons. Northern Germany. Several compoundscomposition. have not For yet the been organohalogens analyzed having for been their measured isotopic in previous studies the termination of carbon isotope ratiosshipper of has a shown wide to rangeconventional be of dewars VOCs a or was suitable developed. cryostats andtaken The in easy-to-use over dry a various cooling period source applications.lowing of that for sampling two can Up campaigns weeks replace to in without remoteprocedure 30 areas. the ranged samples Recovery need from rates can for of 92 theducibility be any entire to technical (1 sampling 105% infrastructure with al- 0.5‰ low to standard 2.8‰. deviations. Withwith a The typical sampling mixing overall volume ratios repro- between of 1 500 and L, 10 pptv carbon can isotope be determined ratios with of a compounds precision bet- this process. 4 Conclusions In this study a simple and field suitable cryogenic sampling system for subsequent de- considering its rapid degradation and the exceptional strong isotopic fractionation of 5 5 30 25 10 15 20 25 15 20 10 Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | , , 2002. , 1999. doi:10.1002/rcm.1795 , 2008. , 2005. , 2008. doi:10.1029/2001jd001267 doi:10.1029/1999jd900217 , 2005. 2184 2183 doi:10.1071/en08037 , 1994. ect of oceanic uptake on atmospheric lifetimes of se- , 2003. ff , 2011. ¨ ockmann, T.: Analytical system for carbon stable isotope doi:10.1021/ac702084k doi:10.5194/acp-5-2403-2005 , 2003. , 2007. doi:10.1002/rcm.1812 doi:10.1029/94JD00786 doi:10.1007/b10456 doi:10.5194/amtd-4-101-2011 lected trace gases, J. Geophys. Res., 107, ACH1-1-9, measurements of light non-methane133, hydrocarbons, Atmos. Meas. Tech. Discuss., 4, 101– CH3Br, Geophys. Res. Lett., 23, 53–56, 1996. tropical plants, Nature, 416, 163–165, 2002. and methyl chloride inPacific, J. urban, Geophys. Res., coastal, 104, and 16033–16039, maritime atmospheres over the western North air, Geophys. Res. Lett., 24, 659–662, 1997. of selected hydrocarbons withGeophys. OH-radicals Res.-Atmos., 105, and 29329–29346, its 2000. relevance for atmospheric chemistry,tures J. of Background Tropospheric Chloromethane211, and 2002. CFC113, Biogeochemistry, 60, 191– ganic compounds to determine the27, “photochemical 3865–3868, age” of 2000. an air mass, Geophys. Res. Lett., specific determination of delta C-13 in volatile organic compounds at ppt levels in ambient hydrocarbons from Irish urban, rural,doi:10.1029/2006jd007784 and marine locations, J. Geophys. Res., 112, D16307, ninkmeijer, C. A. M.:hydrocarbons Isotope and analysis halocarbons of separatedtrom., hydrocarbons: 19, from 455–460, trapping, ambient recovering air, and Rapid archiving Commun. Mass Spec- carbon subtraction technique for stableride, carbon methyl isotopic bromide, analysis and C2–C5 of saturated atmosphericmass hydrocarbons methyl spectrometry, using chlo- Rapid continuous-flow isotope Commun. ratio Mass Spectrom., 19, 477–483, carbon isotope analysis of selected halocarbonslocation, at Environ. parts Chem., per 5, trillion 340–346, concentration in an urban Oremland, R. S., andoxidation of Goldstein, methyl A. halides by H.: methylotrophic bacteria,2001. Large Proc. Natl. carbon Acad. Sci., isotope 98, fractionation 5833–5837, associated with monds, P. G.: Medusa: AMeasurements Sample of Preconcentration Atmospheric and Trace GC/MSAnalyt. Halocarbons, Detector Chem., Hydrocarbons, System 80, and for 1536–1545, in Sulfur Situ Compounds, Res., 99, D8, 17–41, M.: Thetion distinctive in isotopic constraining signaturedoi:10.1016/s0045-6535(03)00206-6 the of atmospheric plant-derived chloromethane chloromethane: budget, Chemosphere, possible 52, applica- the 433–436, atmospheric chloromethane budgetChem. Phys., gained 5, using 2403–2411, stable carbon isotope ratios, Atmos. 2005. the CH3Br budget of thethe United atmosphere, States Proceedings of of America, the 96, National 10006–10009, Academy 1999. of Sciences of proach for studying atmospheric budgets and2003. chemistry, Chemical reviews, 103, 5025–5048, Zuiderweg, A., Holzinger, R., and R Yvon-Lewis, S. A. and Butler, J. H.: E Yvon, S. A. and Butler, J. H.: An improved estimate of the oceanic lifetime of atmospheric Yokouchi, Y., Ikeda, M., Inuzuka, Y., and Yukawa, T.: Strong emission of methyl chloride from Tsunogai, U., Yoshida, N., and Gamo, T.: Carbon isotopic compositions of C2-C5 hydrocarbons Rudolph, J., Czuba, E., and Huang, L.: The stable carbonThompson, isotope A. fractionation E., for Anderson, reactions R. S., Rudolph, J., and Huang, L.: Stable Carbon Isotope Signa- Rudolph, J., Lowe, D. C., Martin, R. J., and Clarkson, T. S.: A novel method for compound Rudolph, J. and Czuba, E.: On the use of isotopic composition measurements of volatile or- Redeker, K. R., Davis, S., and Kalin, R. M.: Isotope values of atmospheric halocarbons and Pupek, M., Assonov, S. S., Muhle, J., Rhee, T. S., Oram, D., Koeppel, C., Slemr, F., and Bren- Miller, B. R., Weiss, R. F., Salameh, P. K., Tanhua, T., Greally, B. R., Muhle, J., and Sim- Mead, M. I., Khan, M. A. H., Bull, I. D., White, I. R.,Miller, Nickless, L. G., and G., Shallcross, Kalin, D. E.: R. Stable M., McCauley, S. E., Hamilton, J. T. G., Harper, D. B., Millet, D. B., Moore, R. and Zafirou, O.: Photochemical production of methyl iodide in seawater, J. Geophys. Keppler, F., Harper, D. B., Rckmann, T., Moore, R. M., and Hamilton, J. T. G.:Komatsu, D. New D., Tsunogai, insight U., into Yamaguchi, J., and Nakagawa, F.: A selective unsaturated hydro- McCauley, S. E., Goldstein, A. H., and DePaolo, D. J.: An isotopic approach for understanding Harper, D. B. and Hamilton, J. T. G.: The HandbookHarper, of Environmental D. Chemistry B., Vol. 3, Hamilton, Part P, J. T. G., Ducrocq, V., Kennedy, J. T., Downey, A., and Kalin, R. 5 5 30 25 20 15 10 20 30 10 25 15 Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | h 37.4 − ; c 45 c c b e e e e c e 7 8 6 3 5 7 5 − 0 ...... 0 . . 1 0 4 1 6 1 1 1 4 ± ± ± ± ± ± ± ± ± 5 5 σ n 1 7 5 5 1 f c c b b b g b 9 29 to ...... 6 1 b . 7 1 2 . 4 2 9 4 2 . . − ...... σ 7 3.1 6 0.5 6 1.1 6 0.4 6 1.0 6 0.8 6 1 . 22 33 28 27 24 28 27 1 ; 9 1 4 4 1 2 1 ± 21 3 b − − ± − − − − − ± ± ± ± ± ± ± ± 3 ; -33 − ; ; – – – ; ; ; – – – ; ; – – ± . 0 0 b 1 1 3 7 0 2 0 ; . b b a b b b b ...... 2 4 6 b 4 9 . 6 2 9 1 3 ...... 5 ± ± ± ± ± ± ± ± 25 43 5 . 18 27 27 21 28 26 29 mean 6 3 9 2 4 1 1 0 2 − − − − . − − − − − ± ± ± ± ± ± ± ± ± 9 4 2 3 4 9 5 8 . 39 ...... 0 . − 42 37 37 23 27 25 28 29 ; a − − − − − − − − 3 . 13C 38.6 48.6 45.1 24.0 21.5 63.6 0 Rudolph et al. (2002); δ mean − − − − − − ± 2 e . 36 − σ n 5.3 2 4.7 2 5.32.9 3 38.2 2 1.4 3 – 1.2 3 0.3 20.8 32.1 22.3 – 3 – – – 1.1 3 3.33.2 3 3 – 1.1 3 1.8 3 1.3 3 1.9 3 5.3 34.6 3 2.1 3 – 25 ± 2.80.8 6 6 1.1 6 1.4 6 2.5 6 1.4 6 1.1 6 0.80.6 6 6 0.8 6 1.10.9 6 6 0.5 6 0.7 6 1.6 6 0.61.3 6 6 6.6 6 1.1 6 0.9 6 1.4 6 1.4 6 0.8 6 1.91.0 6 1.8 6 0.9 6 6 σ n 1.8 6 0.51.9 6 6 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± C mean 13 0.5 34.4 1 36.5 29.3 29.5 15.7 13.6 28.3 31.7 18.1 34.1 17.2 37.9 41.5 23.5 28.2 28.0 29.4 39.9 35.7 53.2 24.3 − − − − − − − − − − − − − − − − − − − − − − σ δ 21 25 26 36 60 36 25 21 13C ± 0.9 0.2 4.1 3.5 2.7 3.0 1.3 0.4 33.8 27.7 26.4 26.5 21.2 44.3 29.2 50.1 27.3 39.6 41.1 29.9 26.8 26.8 44.3 22.3 24.1 29.1 20.9 21.2 17.0 66.8 27.0 31.0 30.2 61.7 30.9 39.6 27.9 61.3 100 210 167 109 413 δ mean − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − Tsunogai et al. (1999); ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± d 4 3 2 2 2 2 2 2 3 2 1 2 4 5 5 3 4 4 4 5 6 2 2 7 3 σ 5 3 mixing ratio mean ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± After pre-seperation Direct injection Recovery σ n 4.6 3 0.66.8 3 2 1.9 3 92 6.5 3 1.1 3 72 2.63.8 3 3 283.0 1.2 8.0 3 530 1.6 680 1.9 3 87 1.5 31.81.8 3 6.1 22.6 3 3 2.1 3 0.2 2 0.2 30.30.1 3 614 3 10.0 27.3 2.1 3 21 1.81.7 30.9 3 3 503 304 0.8 3 1615 0.7 3 524 1.3 23,27 13.1 31.0 3 222.0 73 10.31.9 2 2 – ± mean 2186 2185 ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± = C mean Mead et al. (2008); intensities on the IRMS against chloromethane and bromomethane [s] [%] [‰] [‰] 13 c RT 18.3 26.8 26.9 29.5 1 28.9 37.2 25.4 31.5 25.7 31.0 29.5 36.5 24.6 29.3 24.3 53.6 25.5 24.9 41.2 31.0 24.6 25.5 29.4 36.5 28.3 28.1 36.2 29.0 44.2 26.3 39.1 57.1 2 − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − − σ δ ± = Rudolph et al. (1997) 0.5 16 0.1 0 2.1 37 14 5 10 22 8 140 0.4 86 2.1 31 3 88 1 2 99 8 0.7 23 75 30 0.3 26 33 0.2 0.1 h ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± Coastal site Urban site Other studies mixing ratio mean = [s] [pptv] [‰] [pptv] [‰] RT Redeker et al. (2007); = b Giebel et al. (2010); g Comparison of the carbon isotope ratios obtained with and without pre-separation for Averaged concentrations and isotopic values for all compounds reported in this paper VinylchlorideBromomethaneChloroethane 75-01-04 74-83-9 1026 1096 75-00-3 107 1196 92 102 Dichlorodifluoromethane 75-71-8 908 101 CompoundPropeneChloromethaneMethanol CAS 74-87-3 115-07-1 744 743 67-56-1 109 790 93 BromoformEthylbenzene 75-25-2 100-41-4 3001 2856 85 87 Chlorobenzene 108-90-7 2777 93 Heptane 142-82-5 2623 98 1,2 Dibromoethane 106-93-4 2613 92 Dibromochloromethane 124-48-1 2584 90 1,1,2 Trichloroethene 79-00-5 2536 95 BromodichloromethaneCyclohexane 75-27-4 2295 98 110-82-7 2297 93 HexaneTrichloroethene 79-01-06 2252 110-54-3 2249 96 Carbontetrachloride 56-23-5 2225 96 Benzene 71-73-2 2216 97 1,2 Dichloroethane 107-06-2 2094 97 Chloroform 67-66-3 1986 100 1,1 Dichloroethane1,1,2 trichloro- 1,2,2 trifluoroethane 76-13-1 1963 100 75-34-3 1901 99 1,1 Dichloroethene 75-35-4 1628 98 1,2 Dichloroethene cis 156-59-2 1884 101 1,2 Dichloroethene trans 156-60-5 1763 100 Trichlorofluoromethane 75-69-4 1608 98 Iodomethane 74-88-4 1548 106 1,2-dichloro-1,1,2,2-tetrafluoroethane 76-14-2 1243 Thompson et al. (2002); Bill et al. (2004); Bromoform 75-25-2 2856 2.4 TolueneChlorobenzene 108-90-7 2777 108-88-3 0.3 2660 34 Trichloroethene 79-01-06 2536 44 1.2-Dichloropropane 78-87-5 2396 1.1 Carbontetrachloride 56-23-5 2225 104 Chloroform 67-66-3 1968 10.0 1.1.2-Trichloro-1.2.2-Triflouroethane 76-13-1 1963 69 Trichlorfluoromethane1.1-Dichloro-1-flouorethane*Pentane* 1717-00-6 1646 75-69-4 22 1608 277 109-66-0 1737 20 Isopentane*2-Butyne* 78-78-4 1694 503-17-3 1736 32 0.4 Dimethylether*VinylchloridePropenal* 115-10-6 893Furane* 10.0 Propanal* 75-01-4Iodomethane 10262-Butene cis 107-02-8 1319 110-009 96 123-38-6 74-88-4 1350 1358 519-18-1 1377 242 1570 2.1 2.6 138 5.3 Propadiene*Cyclopropane*DichlorodifluoromethaneBromomethane1-Chloro-1.1-difluoroethane* 75-78-1 463-49-0 75-19-4 75-68-3 908 831Propene-methyl* 834 1112 554 25 16 74-83-9 1096 7.0 115-11-7 1336 186 10.6 Isobutane*ChloroethaneButane* 75-00-3 75-28-5 1196 1171 106-97-8 2.0 15 1243 52 Propane* 74-98-6 819 193 Chloromethane 74-87-3 744 620 ChlorodifluoromethanePropene 75-45-6 703 231 115-07-1 743 61 Compound1.1.1 Trifluoroethane* 420-46-2 546 CAS 12 Bromotrifluoromethane* 75-63-8 610 3.7 Hexafluoropropene* 116-15-4 687 1.2 as internal standards. a f * Mixing ratios have been calculated from the CO Table 2. from the coastal and the urban sampling site. Table 1. the Scott Speciality Gases500 L. TOC 15/17 Results standard are andpounds. only recovery given rates for for compounds a that sampling were volume free of of interferences from other com- Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | Discussion Paper | sampling pump l/min. 5 ambient air nitrogen or bar 2 activated charcoal valve massflow meter massflow meter fraction residual to IRMS Valco 6-port Fractionation LN target valve4 fraction valve2 combustion Interface GC oven 110°C toMSD dry shipper Voyageur 12 Voyageur GC toMSD carrier carrier 2.7ml/min. Quantification Pre-separation and and Pre-separation filled with filled liquide nitrogen 2188 2187 carrier carrier 1.6 ml/min. valve5 valve valve1 Carbon isotope ratio determination ratio isotope Carbon LN Valco 8-port cryo trap cryo valve3 condenser 5°C cryo trap cryo valve Nafion dryer Nafion in silica gel waste flow restriction flow ml/min. 70 at bar 2 LN Valco 8-port Desorption and and Desorption Pre-concentration inletfilter thermodesorption 330°C,min. 15 cryo trap cryo -110°C waste He 15 ml/min. adsorption tube adsorption Peltier cooler -5°C Scheme of the analytical system. Scheme of the sampling system. rRaschig rings bed glas – 75 µm 250 N Porapak Fig. 2. Fig. 1.