SR 176

Special Report 176

A METHOD FORbe. CONCENTRATING AND DETERMINING

TRACE ORGANIC COMPOUNDSIN THE ATMOSPHERE

D. C. Leggett, R. P. Murrmann,

T. J. Jenkins and R.Barriera

n.rl

l,,JUL 20 'rJune 1972

CORPS OF ENGINEERS, U.S. ARMY

COLD REGIONS RESEARCH AND ENGINEERING LABORATORYHANOVER. NEW HAMPSHIRE

RoplOdu.(d I,,

NATIONAL TECHNICALINFORMATION SERVICE jU S epo f 0 o1lnf'~Ce

S1,,-gel.d VA •2|3APPROVED FOIR PUBLIC RELEASE. DISTRIBUTION UNLIMITED

VA.

the findingz in this report are not to be

construed as at, .Ifc:ai Department o!the Army position unless so desigmatedby other authorized documients.

" ttt~~tTa SECUlONt•

.. o.o. oo..o.... ................... ... .......

.•[............ ........ '" 11

11 LIAi¥1. isa! S6CIA.L

AVAIL _ Ot

5-.-7 IM,-,-

UnclassifiedSecutifi Classification

DOCUMENT CONTROL DATA. R & D(Securit lasific~ation.I, of WisI. body of abstract and indexi1ng annoation mustwa be entered when the overall reporl Is .I...IIled)

V. ORI$;IN ATIN40 ACTIVITY (Corporate author) Z*. RE1PORT SECURITY CLASSIFICATION

U.S. Army Cold Regions Research and UcasfeEngineering Laboratory 26.GRUI Hanover, New Hampshire 03755

REPORT TITLE

A METHOD FOR CONCENTRATING AND DETERMINING TRACE ORGANIC COMPOUNDS INTHE ATMOSPHERE

4ý OCACRiPTive NOTES (rype of report and Inclusive dates)

5 AUTH ORISI (/fit* et .sm. mWdle Initial, fasilnameo)

D.C. Leggett, R.P. Murrmann, T.F. Jenkins and Re Barriera

6 REPORT nA-TE 78. TOTAL NO. OP PAGES .No. OF REars

June_1972124Is. C ON TRAC T OR GRAN I NO. Oe RIGINATOR'S REPORT NU&IUERIS)

b.voayNo Special Report 176DA-Task A)J,22000002

C.S. OTHER REP'ORT NOIS) (Any other nussbto "~t may be oeslieodtils "portl)

d.

10 COISTR11BUTION STATEMENT

Approved for public release; distribution unlimited.

11. SUPPLEMENTARY NOTES fIs. SPONSORING MILITARY ACTIVITY

U.S. Army Mobility Equipment Researchjand Development CommandIs ASISTRACT

*Determination of subpart-per-billion (sub-ppb) levels of volatile organiccompounds in the atmosphere by flame ionization gas chromatography requires

* ~sample sizes of at least several hundred mnilliters of air. Cryogenic methodsof concentrating trace compounds before analysis have the disadvantage of

* also concentrating large amounts of water, a serious problem in gas-liquidchromatography. A simple method was developed for sample collection andconcentration using porous polymer adsorbants with the unique properties ofhigh capacity for retention of organic compounds and minimal capacity for

*retention of water. This technique was used co determine sub-ppb levels ofvolatile organic compounds in a typical rural atmosphere. Probable sourcesof these organic compounds were vehicle exhaust, biological processes,natural gas leaks, and industrial chemicals.

14. Key wordsAir analysis Land mine detectionAir pollution Organic vaporsAtmospheric analysis Rural atmospheresChromatographic analysis Trace gasExplosives detection

Few FRAUgsf. J AW W.MCHE isDD Nove. 4 I SAM ~ UAAMYUS Unclassified

lacsidty GOesdNCOU.,

A METHOD FORCONCENTRATING AND DETERMINING

TRACE ORGANIC COMPOUNDSIN THE ATMOSPHERE

D. C. Leggett, R. P. Murrmann,

T. J. Jenkins and R.Barriera

June 1972

CONDUCTED FOR

U.S. ARMY MOBILITY EQUIPMENT RESEARCH AND DEVELOPMENT COMMAND

DA TASK A1J22000002B3Y"

CORPS OF ENGINEERS, U.S. ARMY

COLD REGIONS RESEARCH AND ENGINEERING LABORATORYHANOVER, NEW HAMPSHIRE

APPROVED FOR PUBLIC RELEASE; DISTRIBUTION UNLIMITED.

PREFACE

This repoit was prepared by Mr. D.C. Leggett (Research Cheniist), Dt. R.P. Murrmann(Reseai Clihemist), Mr. T.F. Jenkins, Jr. (Reseatuh Chemst), and SI3 R. Barriera of theResearch Division, U.S. Army Cold Regions Research and Engineering Laboratory (USACRREL).

The work was supported by U.S. Army Mobility Equipment Research and Development

Command, under DA Task A1J22000002.

This report was technically reviewed by Captain P. Hunt, Mr. J. Cragin, andSP M. Hlerron.

The contelts of this report are not to be used for advertising, publlication, or pro-motional ptuposes. Citation of trade names does not constitute an official endorsement orappioval of the use of such commercial pioduets.

CONTENTS

PageIntroduction ............................................................ IExperiniental ........................................................... 2

Materials and equipment ............................................. 2Pteparation of collection tubes ....................................... 2Sample collection ................................................... 3Analytical procedures ............................................... 4

Results and discussion .................................................. 4Summary ............................................................... 6Literature cited ......................................................... 7Abstract ............................................................... 9

ILLUSTRATION

Figure1. Schematic diagram of sample elution and injection system .............. 3

TABLES

TableI. Chromatogram of 10-liter air sample, Grafton County, New Hampshire, 19 May

1971 ........................................................... 511. Chromatogram of 10-liter air samaple, Grafton County, New Hampshire. 9 June

1971 ........................................................... 6III. Electron capturing substances identified in atmospheric samples from rural

Grafton County, New Hampshire, May and June 1971 ................. 7

J

A METHOD FOR CONCENTRATING AND DETERMININGTRACE ORGANIC COMPOUNDS IN THE ATMOSPHERE

by

D.C. Leggett, R.P. Murrmann, T.F. Jenkins, and R. Barriera

INTRODUCTION

The development of trace gas sensors for detecting explosives and mines is of militarysignificance. The development of these devices requires knowledge of the types of probablebackground contaminants that can interfere with the detection. Toward this end the authorshave undertaken the development of methods for the analysis of trace organic compounds inthe atmosphere, since this is the medium in which these detection devices are required towork.

Few data are available on the distribution of trace organic compounds in the unpollutedatmosphere, primarily because detection and measurement of these compounds at the levels found inunpolluted atmospheres requires sensitivity beyond the range of most commercially available in-strumentation. For example, direct analysis of organic compounds by gas chromatography usingflame ionization detectors is sensitive only to about 1 ppb.

Thus, the analysis of remote atmospheres for trace organic compounds required special inletsystems that enabled the compounds to be concentrated from several hundred milliliters of air beforeanalysis.$ Even with prior concentration the levels observed were barely detectable. In fact, con-centration techniques have generally been needed to measure hydrocarbons even in relativelypolluted urban environments.' 2, I, tS 20,

Typical concentration techniques have used cryogenic trapping of the trace organic compoundsby a solid or liquid-coated chromatographic support.' s ,0 21 However, water vapor is also retainedby this procedure. This poses a problem for chromatographic analysis and generally restricts thesize of the air sample to a few hundred milliliters.

Procedures used in this laboratory for analyzing air samples in the presence of large amountsof water were described in earlier reports."' 22 2 The trace organic compounds in 3 liters of air weretrapped cryogenically on glass beads and eluted later at elevated temperature onto a Porapak Qcolumn. Water did not interfere in this process but resolution of the complex mixture of trace com-pounds found in the air samples was inadequate.

Removal of water by passing the air samples over desiccants before cryogenically trapping thetrace compounds is a useful technique" 20 but is not applicable to some classes of compounds be-cause of adsorptive losses,"'

Sample collec ion in remote environments is also a problem because it is often inconvenient toperform on-site analysis. 'T1he collection of samples in re Mote areas for later laboratory analysis

•- ,-•:o•, .. .. • .... . , . .. • ' • •" • • . . • . .. . •'' "' •.• .. °' 1' • '' • . .[ ...- *"". *," i'" • • -. .. . ... ...-- - -5 ,'-,

`zfl7;r7,W

ITTIT

2 CONCENTRATING AND DETERMINING TRACE ORGANIC COMPOUNDS IN ATMOSPHEREj

etqtnres a soul e of refrigerant if colt('etrat joi is to be done in Ili- 1t.eld. or some type of con-taitlet it' Wholva l samiples tire to he retturned to the lalborato01y. Ini any case,* care Lutust be takeni to

preserve sample integm ity (hiring the t ime between collectioni of the samlpe and analysis.

Porous polymers such as the Porapaks and Chromosorb 100 series apparently have not been

despitesal atiiounqu sitablt.Il nlss of wae eane teadorbanataed chlleeydicreoved byi alonshinatied

(hollPst o t102 wits usdt elmitor cneantalyit ftl ape nbing výltile tro rganlic copud impu ed toslphtericltomples. det wert ed b o quas lritative apl. without ititerfrtc rn wtr ial. lhuhti

I(Thisd 15or designed s to dethrodio-acurn ievnenrt trace gaeorgtaatioplOitwl ndobedyb copunseinlie-moteatmyotpher es ao T hailvis prtob sleus. rtdb h eemnto f vltl rai oioi~

Mnaterials and equlpuieat;uppleesBydaigiaisapetoghaw cniig

SoanplkQSte ollenctn tubeund iwerre vuaefoto -1 men :316r ateacnlescteel twithotuing andfvalesrtantsTh ben smalamonedo witer acetoinead by hieootorb. ahe tu omesewely removed Wit y 2. fluWangther soits100-etio tu0beswih Porpa he0-S bdsotreit anayisd the valvles weeonalngcedt the tubes wraith Swgcompud

atoehedeiesminedbygaschrombtogeraphd afertheyt inerfrne fille wit athi . Fdsoraltly. bakthngh themat10swita ielidis eigne tlow deftermi'zine tiact ilatey wteatopere shwnt wil cle ndbyoablaekdly on asu ciromarahmai thea flameonizaitaion petr opleraemts."ai/ul cl.F~t-ili or a eeal

cialmplet sysletemsown she., vmatialdnFgre1 h ape loop, an 8-in, by :3i6 stainless steel tbn n avstaU-tubee plartiayfled with 100-tone n 12-ehlAnakromi ABS tueoate iled with 2.0 g of-Waterilisoneaoil,

place btweeti tehePraa U-t tube bnt and the mir valve s so hate loop culd to0 isltedhe tuenwtirae iletnuystend Texept for thles sTiple lopakn was renlosed byiilanie olsswoetie.uevav

The tunbles used lene thfis studafethy were baedoitllnedwithl befoe aobeny aing used and were anlyewith ~iathelylil flwfte the smplesl nilte were sollentedoTbiepreautionas tlanke itoinimize the possihgrlitofthantanmina ionisariousl afeciting theraesut at 8 li e1reel aflow sclevel Foftih volatiles gompounds

Reuet worktially filledlt a with 10-ti 2methe tubeso Abecometslightly c5it1miClated selicthoughil.lwas aronnetd the tub fitwi-was canrbe dtcted. Theisio Smaybptlieg rsulto egratin oailry diffusionrplac.sed btocr ng witlem tUletubesadtemcrvles that areno loo unsool eisltd. Tthy lekeaondtire fittig

Prp arathion of cgllectitt ont llla ~ltnls rc i osaetknt rtc ~ tubes rt h tuisTiber Albs, austd be taken stod perevbaedut theimperiatel btefotrte tbesnfruseriaingwereuanaabove

of cot)adminto srolyafcing th hraayi: teiesut sgiiant ~tmeoextrmllolels of cvotalaans til e rodpiunds

being....................

CONCENTRATING AND DETERMINING TRACE ORGANIC COMPOUNDS IN ATMOSPHERE 3d.

2"PositionVolvo

IOven(llO°C) 2 -t Chrornatograph

a. To VentC a.FlHw L-Pattern ValvesMe irs

b.Filter

a. ubei Dewar

LiquidNitrogen

a. Sample elution.

To GasI I ChromatogrophI 2-

Sample Tube To VentL_ _

GlassBeads Hooting(190Oc) Mantle

b. Sample injection.

Figure 1. Schematic diagram of sample elution and injection system. a. Helium"zero" gas cylinder. b. Filter conetaining molecular sieve and silica jel. c. Sam-

ple collection tube containing Porapak Q-S adsorbant. d. 2 -position microvolumevalve (Carle). e. L,-pattern microvalve. f. U-tube containing 15% DC 200 silicone

oil on Anakrom ABS (100- to 120-mesh).

Sample collection

Samples were collected in the field by drawing air through the collection tubes with WelchScientific Co. bicycle-type hand pumps having a displacement of about 400 ml. The pumps werecalibrated before field use by water displacement. In tho laboratory, samples were generally takenwith a Cast portable electrical vacuum pump. The rate of the pump was calibrated with a soap-bubble flow meter.

4 CONCENTRATING AND DETERMINING TRACE ORGANIC COMPOUNDS IN ATMOSPHERE

Varying the rate of sample collection from 100 to 1000 ml/min did not significantly change thecollection efficiency and gave the same results as the hand pump. Therefore, collection efficiencywas considered to be complete up to the point at which a given compound began to elute (break-through volume). This volume varies according to the retention characteristic of the compound. Ingeneral, compounds of C, and larger were quantitatively adsorbed from a 10-liter air sample takenat 251C. The breakthrough volume for compounds eluting earlier was determined by preparing acurve of observed concentration versus sample size while sampling the air in a large room, a rela-tively constant source.

Analytical procedures

After the sample was collected, the tubes were connected to a regulated source of helium andflushed in the direction in which they were collected for 30 min at a flow of 50 ml/min at ambienttemperature (- 250C). This procedure removed residual water as well as early-eluting compoundsincluding Cl-C3 hydrocarbons. The sample was eluted into the sample loop by backflushing the tubewith helium at a flow of 50 ml/min for 30 min. During the elution cycle the temperature of the ovenwas maintained at 1100C and the sample loop was cooled with liquid nitrogen. After the samplewas eluted the loop was isolated and heated to 1900C; the two-way valve was rotated to the "inject"position and the microvalves were opened in sequence, sweeping the contents onto a chromatographiccolumn.

The column chosen for this work was a 9-ft x %-in. stainless steel tube filled with WatersAssociates' Durapak, Carbowax 400 chemically bound to Porasil C. The column material was washedwith acetone and dried in a vacuum at 800C before the column was filled. The washing was importantto obtain maximum resolution of sample components and minimal column bleed. However, this proce-dure somewhat changed the retention characteristics of the column. It is thought that some of theCarbowax was removed by this washing, exposing more adsorption sites on the solid. (Acetoneelutes after toluene on this column.) The analytical conditions are given in Tables I ani II.

The effluent from the column was either fed into a flame ionization detector or split between theflame and a pulsed electron capture detector. In the latter case the signals were recorded simulta-neously on a Leeds and Northrup dual-channel millivolt recorder with a disk integrator. Peaks fromthe flame detector were integrated and converted to concentratioi, using weighted response factors."The response of the flame detector was determined by periodic injections of 0.5 pl of n-heptane.Sample components were identified by comparing their Metention times with those of known com-pounds. Electron capture peaks were not quantitated.

RESULTS -AND DISCUSSION

"Tables I and II show the results of representative analyses of 10-liter air samples from ruralGrafton County, near Lyme, New Hampshire, taken in May and June 1971. Identifications aretentative since they were determined on only one column. Other liquid phases could have been usedto confirm the presence of particular compounds, but the main purpose of this study was to evaluatea technique and not to rigorously identify individual compounds. However, most of the organiccompounds tentatively identified have been reported in atmospheric analyses made by other investi-gators., 1 1 *I to It 20 at as

Most of these compounds are found in vehicle exhaust, which undoubtedly was a major sourceot atmospheric contaminants otserved in this atudy. Otl'er sources of atmospheric hydrocarbonsinclude soils and vegetation4 IT 26 " and natural gas and gasoline vapors.,' 1 "' , Organic compoundscontaining oxygen are probably derived largely from natural sources such as vegetation in remoteareas.' In more polluted areas, vehicle emissions may contribute significant amounts of oxygen-ates.'' 'o Volcanic emanations have been found to contain hydrocarbons and oxygenates as well ashalogenated cormpounds, presumably derived from pyrolysis of o.git:,c matter at the source."

CONCENTRATING AND DETERMINING TRACE ORGANIC COMPOUNDS IN ATMOSPHERE 5

Table 1. Cbromatogram ao 10.liter air sample,* Graftoa County, Now Hampshire, 19 May 1971.

Retention time Concentratiol.Compound (min) (ppb)

isobutane 6.3 0.10unknown 6.6n-butane 6.7 0.531-butene and isobutene 7.2 0.122-butene, trans 7.4 0.151, 3.butadiene 7.9 0.08butyne and dimethylpropane 8.3 0.012-methylbutane 9.1 0.42n-pentane 9.5 0.292-methyl-l-butene 9.7 0.011-pentene 9.9 0.012-pentene, trans 10.1 0.032-pentene, cis 10.4 0.032-methyl-1,3-hutadiene 10.6 0.012,£.dimethylbutane 11.2 0.02.-methylpentane 11.7 0.30

8-methylpentane and methylcyclopentane 11.8 0.21n-hexane 12.0 0.25unknown 12.4acetaldehyde 13.1 Iunknown 13.42.2-dimethylpentane 13.7 0.01benzene 13.9 0,53vichloroethylene 14.1 1n-heptane 14.7 0.02unknown 15.0unknown 15.2unknown 15.7unknown 16.1unknown 16.3toluene 16.8 0.54unknown 17.3acetone 18.1 2.5ethylbenzene 20.6 0.10m and p-zylene 21.3 0.24o-zylene 22.3 0.34unknown 23.6

*Column: 9-ft x %-in. Carbowax 400 Durapak: temperature programmed from-15 0C to 1300 C at 100/m; carrier gas: helium at 20 ml/min; detector: flameionization.tNot determined because of high background interference.

Electron capturing compounds, again identified by their retention times on the Carbowax column,are listed in Table Ill. Except for carbon disulfide, these compounds are halogenated hydrocarbons,all of which are commonly used industrial chemicals: this probably accounts for their presence in theatmosphere.:' " These compounds would not have been observed if flame lonization detection alonehad been used, because the chromatographic peaks were obscured by the flame response to ",ydro-carbons eluting at the same time as these compounds or were below the detection limits of the flame(aWout I part in 10" or loss in this work). The advantage of using the electron capture detector is

6 CONCENTRATING AND DETERMINING TRACE ORGANIC COMPOUNDS IN ATMOSPHERE

Table IU. Chromatogram of 10-liter air sample,* (kafton County, New Hampshire, 9 Jam 1971.

Retention time ConcentrationCompound (mm) (pph)

Isohatalie 6.6 0.01n-butuie 6.9 u.ll2-butene. trans 7.6 0.092-.iethylinitane and Freon 11 9.3 0.:151ii-pentane 9.6 0.182-ilethyl- -butene 9.8 0.003I-pentene 10.0 0 022-pentenetlirmis 10.7 0.082-inethyl1-2-1)11ten e 10.3• 0.000}52-pentene, cii 10.5 0.042-Ilnethy I .l.-I.X utit hdeue 10.7 0. 19

2.2-dimethylbutane 11.3 0.022-mnethylpentane and Freon 113 11.8 0.12"*3-methylpentane and methylcyclopentane 11.9 0.09n-hexane 12.1 .0OGacetaldehyde and tetrachloromethane 12.9 Ubenzene 14.0 0.13trichloroethylene 14.2 1 tn-heptane 14.9 0. 11tetrachloroethylene 15.3unknown I 5.4unknown 15.7unknown 15.8unknown 16.0unknown 16.4unknown 16.8toluene 16.9 0.09unknown 17.4acetone 17.8 1.7unknown 17.9unknown 19.3

ethylbenzene 2G.6 0.08unknown 21.8unknown 23.2

*Column: 9-ft x %-in. Carbowax 400 Durapak: temperature programmed from-15 0 C to 130 0 C at 100 /m: carrier gas: helium at 20 ml'min; effluent from thecolumn was split (4:1) between a flame and a pulsed electron capture detectorwith purge of 50 mI/mm of argon-5% methane added between column And detector.tAs 2-methylbutane.**As 2-methylpentane.

ttNot determined because of high background interference.**Ele-Itron capture peak not determined.

that it affords a high degree of specificity; that is, it responds to small amounts of certain types ofcompounds such as halogenated hydrocarbons without interference from relatively large amounts ofhydrocarbons and oxygenates (e.g., see ref. 23).

SUMMARY

A method was develr'ped for the detection and estimation of volatile organic compounds in theatmosphere in the sub-ppb range. Trace substances were concentrated from large volumes of air by

,. + • -" . . . • " '+ .. ... . . . . . . . .. .. . ....... .. ..+.. .. ... ++ I

CONCENTRATING AND DETERMINING TRACE ORGANIC COMPOUNDS IN ATMOSPHERE 7

Table Ill. Electron capturing substances identified in atmospheric samples from ruralGrafton County, New Hampshire, May and June 1971.

dichlorodifluoroniethane (Freon 12)

carbon disulfide

fluorotrichloromethano (Freon 11)1,1,2-trichloro- 1, 2 ,2-trifluoroethane (Freon 113)

te ira ehlorosnethane

trichlorot-thylene

tri chloroinethatne

tetra chloroethylene

drawing the sample through a tube containing Porapak Q-S adsorbant. Residual water was removedfrom the sample by flushing the tube with dry helium before backflushing the trace compounds at

Plevated temperature into a cold trap. The trap was then heated and the contents were flushed intoa gas chromatograph. A tube packed with 2 g of 100- to 120-mesh Porapak ".S was sufficient tocompletely adsorb hydrocarbons of C, and larger from a 10-liter sample of air taken at 250C.

This method was applied to the determination of trace compounds in rural atmospheric samples.Probable sources of these compounds included vehicle exhaust, biulogical processes, and industrialchemicals.

LITERATURE CITED

1. Altshaller, A.P. and T.A. Bellar (1963) Gas chromatographic anal.,sis of hydrocarbons inthe Los Angeles atmosphere. Journal of Air Pollution Control Association,vol. 13. no. 2. p. 81-87.

2. Bellar. T.A., M.F. Browm and I.E. Sigsby, Jr. (1963) Detennination of atmospheric pollut-ants in the part-per-billhon range bty gas chromatography. Analytical Chemistry,vW. 35, no. 12, p. 1924-1927.

:1. Bellar. T.A. and .I.E. Sigsby, Jr. (1970) Direct gas chromatographic analysis of lowmolecular weight substituted organic compounds in emissions. EnvironmentalScience and Techno;,;ý:y. vol. 4. no. 2, p. 150.156.

4. Cavanagh. L.A. ( 1968) Development of instrumentation of airborne collection of

atmospheric organic chemicals. Stanford Research Institute, Project PRU-6856,Final Report.

5. Cavanagh. L.A., C.F. Schadt and E. Robinson (1967) Atmospheric hydrocarbon and carbonmnonioxide measurements at Point Barrow, Alaska. Environmental Science andTechnology, vol. 3. no. 3. p. 251-257.

6. Dravnieks, A.. B.K. Krotoszynski, J. Whitfield. J. Burton, A. O'Donnell and T. Burgwald1971) high speed collection of organic vapors front the aunosphere. Environ.

mental Science anti Technology, vol. 5. no. 12, p. 1220.1222.

7. Eggertsen, F.T. and F.M. Nelsen (1958) Gas chromatographic analysis of engine exhaustand atmosphere. Analytical Chemistry, vol. 30, no. 6, -,. 1040.1043.

8. Ellis, C.F.. R.F. Kendall and B.11. Eccleston ( 1965) Identification of some oxygenatesin autonohbile exhausts by combined gas liquid chrontiatograpliv and infraredtechniques. Analytical Chemistry. vol. :37, no. 4. p. 511-516.

"A :..... .•....,..... ..... .. ... -' ' '•W .' ,,.

8 CONCENTRATING AND DETERMINING TRACE ORGANIC COMPOUNDS IN ATMOSPHERE

LITERATURE CITED (Cont'4)

9. Feinland, R., A.J. Andreatch and D.P. Cotrupe (1961) Automotive exhaust gau analysis bygas-liquid chromatography using flame ionization detection. Analytical Chemistry,vol. 33, no. 8, p. 991.994.

10. 'racchia, M.F., F.J. Schuette and P.K. Mueller (1967) A method for sampling and detennina-tion of organic carbonyl compounds in automobile exhaust. Environmental 3cienceand Technology. vol. 1, no. 11, p. 91&.929.

11. Lonneman, W.A., T.A. Bellar and A.P. Altshuller (1968) Aromatic hydrocarbons in theatmosphere of the Los Angeles Basin. Environmental Science and Technology,vol. 2, no. 11, p. 1017.1020.

12. Lovelock, J.E. (1971) Atmospheric fluorine compounds as indicators of air movements.Nature, vol. 230, p. 379.

13. McNair, H.M. and E.J. Bonelli (1967) Basic gas chromatography. Palo Alto: Varian

Associates.

14. Mutrrmann, R.P., D.C. Leggett and T.F. Jenkins (1970) Feasibility of tunnel detection bytrace gas analysis. U.S. Army Cold Regions Research and Engineering Laboratory(USA CRREL) Special Report 148 (AD 708861).

15. Neligan, R.E. (1962) Hydrocarbons in the Los Angeles atmosphere. Archives of EnvironmentalHealth, vol. 5, p. 581.591.

16. Papa, L.J.. D.L. Dinsel and W.C. Harris (1968) Gas chromatographic determinanon of C1 toC12 hydrocarbons in automotive exhaust. Journal of Gas Chromatography, vol. 6,p. 270.279.

17. Rasmussen, R.A. (1970) Isoprene: identified as a forest-type emission to the atmosphere.Environmental Science and Technology, vol. 4. no. 8, p. 6674671.

18. Rasmussen, R.A. and F.W. Went (1965) Volatile organic material of plant origin in theatmosphere. Proceedings of the National Academy of Sciences, p. 215.220.

19. Smith, G.H. and M.M. Ellis (1963) Chromatographic analysis of gaies from soils and vegeta-tion related to geochemical prospecting for petroleum. Bulletin of the AmericanAssociation of Petroleum Geologists, vol. 47, no. 11, p. 1897-1903.

20. Stephens, E.R. and F.R. Burleson (1967) Analysis of the atmosphere for light hydrocarbons.Journal of Air Pollution Control Association. vol. 17, no. 3, p. 147.153.

21. Stephens, E.R. and F.R. Burleson (1969) Distributit,,, of light hydrocarbons in ambient air.Journal of Air Pollution Control Association, vol. 19, no. 12. p. 92D-936.

22. Stoiber, R.E., D.C. Leggett, T.F. Jenkins. R.P. Murrmann and W.I. Rose, Jr. (1971) Organiccompounds in volcanic gas from Santiaguito Volcano, Guatemala. GeologicalSociety of America. Bulletin 82, p. 2299.2302.

23. Williams, l.H. (1965) G&s chromatographic techniques for the identification Of low concentra-tions of atmospheric pollutants. Analytical Chemistry, vol. 37, no. 13, p. 1723-1732.

24. Williams, F.W. and M.E. Umstead (1968) Determination of trace contaminants in air by con-centrating on porous polymer beads. Analytical Chemistry, vol. 40, r.o. 14, p. 2232-2284.