Geochimica et Cosmochimica Acta, Vol. 69, No. 4, pp. 1073–1084, 2005Copyright © 2005 Elsevier Ltd

Printed in the USA. All rights reserved

doi:10.1016/j.gca.2004.07.030

Molecular and compound-specific isotopic characterization of monocarboxylic acids incarbonaceous meteorites

YONGSONG HUANG,1,* YI WANG,1 MARCELO R. ALEXANDRE,1 TAEWOO LEE,2 CHRISTOPH ROSE-PETRUCK,2 MEGAN FULLER,1

and SANDRA PIZZARELLO3

1Department of Geological Sciences, Brown University, Providence, RI 02912, USA2Department of Chemistry, Brown University, Providence, RI 02912, USA

3Department of Chemistry & Biochemistry, Arizona State University, Tempe, AZ 85287, USA

(Received April 13, 2004; accepted in revised form July 26, 2004)

Abstract—Low molecular weight monocarboxylic acids are the most abundant water soluble organiccompounds in the Murchison and many other CM type carbonaceous chondrites. In this study, we examinedthe monocarboxylic acids in Murchison and EET96029.20 carbonaceous meteorites using a new samplepreparation and introduction technique for gas chromatograph recently developed for volatile, water-solubleorganic compounds: solid phase micro-extraction (SPME). We identified more than 50 monocarboxylic acidsfrom Murchison compared with the 18 compounds reported previously. Formic acid, a known interstellarmolecule, has been fully analyzed in these carbonaceous meteorites, with its �D value suggesting aninterstellar origin. We determined both carbon and hydrogen isotopic ratios of individual monocarboxylicacids in Murchison, to better define the origins and genetic relationships of these compounds. The compound-specific isotopic data reveal a large enrichment in 13C (�13C up to � 32.5‰) and particularly D (�D up to� 2024‰). The branched acids are substantially enriched in both 13C and D relative to the straight chain acids,with those branched acids containing a quaternary carbon showing the greatest isotopic enrichment. Theisotopic difference may be attributed to variations in the different synthetic regimes or terrestrial input of

0016-7037/05 $30.00 � .00

straight chain acids. Copyright © 2005 Elsevier Ltd

1. INTRODUCTION

Isotopic studies of organic matter in carbonaceous meteoritesprovide important insights into the origins and synthetic path-ways of meteoritic organic compounds. Carbon, nitrogen andparticularly hydrogen isotopic ratios of bulk organic substancesin “pristine” meteorites such as Murchison generally fall out-side the range observed for terrestrial organic compounds (Ker-ridge et al., 1987; Epstein et al., 1987; Krishnamurthy et al.,1992; Cronin and Chang, 1993; Cooper et al., 1997), indicatingtheir extraterrestrial origins. Advances in isotopic techniques inthe last decade allow compound-specific isotopic analysis(Hayes et al., 1990; Burgoyne and Hayes, 1998; Hilkert et al.,1999; Huang et al., 1999a, 1999b, 2000; Wang and Huang,2001, 2003), presenting new opportunities to better define theorigins and formation pathways of meteoritic organic com-pounds. However, so far only a few compound classes inMurchison have been subjected to compound-specific isotopicstudy, including carbon and nitrogen isotopic ratios (Engel etal., 1990; Engel and Macko, 1997; Pizzarello et al., in press) orhydrogen isotopic ratios (Pizzarello and Huang, in press) ofamino acids; C and H isotopic ratios of dicarboxylic acids(Pizzarello et al., 2001; Pizzarello and Huang, 2002), and Cisotopic ratios of n-alkanes (Sephton et al., 2001) and freearomatics and pyrolysis products of macromolecules (Sephtonet al., 1998, 2000).

Monocarboxylic acids (MCA) are the most abundant solubleorganic compounds in the Murchison meteorite (10 to 100times more abundant than amino acids) (Yuen and Kvenvolden,

* Author to whom correspondence should be addressed.(Yongsong_Huang@brown.edu).

1073

1973; Lawless and Yuen, 1979). However, characterization ofmolecular distributions of MCAs in Murchison was conductedin 1970s using gas chromatography equipped with stainlesssteel capillary GC columns. Only 18 C2 to C8 MCAs werereported in Murchison (Yuen and Kvenvolden, 1973). Thestainless steel capillary column has a limited resolution com-pared with modern fused silica capillary GC column, especiallyfor the complex mixture of straight chain and branched MCAsin the Murchison meteorite. Indeed, later studies of Antarcticcarbonaceous meteorites using fused silica capillary GC col-umn revealed a much larger suite of more than 35 C2 to C12

MCAs, with multiple methyl and ethyl branched acids(Naraoka et al., 1999). Given that Murchison has been demon-strated to contain the most extensive series of extraterrestrialorganic compounds (Cronin and Chang, 1993), it is surprisingthat the composition of the MCAs in Murchison would havebeen relatively simple. There are also concerns from the ex-perimental procedures used to isolate MCAs for GC andGCMS analyses in the early studies on the Murchison. Theprocedure involved a water-CH2Cl2 partitioning step to extractthe MCAs, followed by solvent evaporation before GC andGCMS analyses. Because the low molecular weight acids aremiscible with water and highly volatile, the procedure is likelyto cause significant loss of MCAs and results in bias in molec-ular distributions. Later study using cryogenic distillation andion chromatography (Lawless and Yuen, 1979; Yuen et al.,1984) indeed demonstrated that a major loss in acetic acid inthe earlier report must have occurred.

The bulk �13C and �D values of the MCA fraction (measuredas Na salts) in Murchison have been reported, with the formerranging from �3 to � 6‰, and the later from � 377 to �

697‰ (Epstein et al., 1987; Krishnamurthy et al., 1992). These

1074 Y. Huang et al.

bulk isotopic values are significantly less enriched in 13C and Dthan the amino acids in Murchison (Epstein et al., 1987; Piz-zarello et al., 1991). The �13C values of six (C2 to C5) indi-vidual MCAs have been determined following a lengthy pro-cedure to isolate the individual compounds using cryogenicdistillation, ion chromatography and GC-combustion systemequipped with a packed GC column (Yuen et al., 1984). How-ever, the majority of the individual acids have not been char-acterized isotopically. No compound-specific hydrogen isoto-pic analyses of MCAs have been conducted.

The goals of the present study are: 1) to re-examine theMCAs in Murchison and an Antarctic carbonaceous meteorite(EET96029.20) using an improved and direct sample introduc-tion technique, solid phase microextraction (SPME), which hasrecently been developed for water soluble organic compounds(Lord and Pawliszyn, 2000), and modern GC capillary columnfor GC and GCMS analyses, to eliminate possible proceduralcompound losses and resolve complex structural isomers; 2) todetermine the carbon and hydrogen isotopic ratios of individualMCAs in the Murchison and Antarctic carbonaceous meteoritesusing SPME coupled with GCIRMS (Dias and Freeman, 1997),to better define the origin, genetic relationship and syntheticpaths of these important compounds.

2. SAMPLES AND ANALYTICAL METHODS

2.1. Meteorite Extraction

A 11.3 g Murchison meteorite, obtained from the Smithsonian In-stitution, and a 4.0 g Antarctic carbonaceous meteorite (EET96029.20,type C2) were used for this study. After removing the outer-most layerof the stone using a clean scalpel and rinsing with double distilledwater, the meteorite samples were crushed in a clean agate mortar. Thepowdered samples were extracted with double-distilled water for 24 hat 110°C under vacuum (Cronin and Pizzarello, 1990; Pizzarello andHuang, 2002). A blank extraction was carried out at the same time andprocessed through the same steps as the authentic samples. The solu-tions were adjusted to pH �10 using 6M NaOH solution and rotaryevaporated to reduce the water to less than 1 mL, which is thentransferred into a 2 mL vial, acidified to pH � 2 using 6M HCl (thespectroscopic grade HCl was redistilled before use), and tightly cappedwith septum (Viton) cap (to allow inserting SPME syringe needle-fiberassembly). In basic conditions, the low molecular weight acids are insalt-forms. No loss of low molecular weight acids were found duringrotary evaporation when standard acids were tested. The acidifiedsolutions were then ready for GC, GCMS and GCIRMS analyses usingsolid phase micro extraction (SPME) as the direct sample introductionapproach. The blank extraction contained no detectable monocarboxy-lic acids.

2.2. GC-FID and GC-MS Analyses

The MCAs were quantified using SPME on an HP6890� GCequipped with a flame ionization detector (FID). We used a 70 mmCarbowaxTM/DVB StableFlexTM SPME fiber which was held by anSPME fiber assembly (Supelco). To obtain the optimum conditions forthe extraction efficiency, we tested a series of experimental conditionsfor adsorption and desorption using an aqueous solution containing 6C2 to C6 monocarboxylic acid standards (concentrations � 1mmol/L inwater). The fiber was immersed in the aqueous standard solution for 5,10, 15 and 20 mins with magnetic stirring, and later desorbed in the GCinjection port for 1, 2, 5, 10 and 15 mins, at 200, 210, and 220°C,respectively. The optimal condition (the condition that yielded thelargest GC peaks, hence highest sampling efficiency within the shortesttime) was found to be: 15 min adsorption with magnetic stirring,followed by 5 min desorption in the GC injection port maintained at

210°C. The GC column used was a 30 m � 0.25 mm � 0.25 �m withNUKOL (Supelco). The oven temperature programming was 35 to 135

°C @ 25°C/min – 185°C @ 1.5°C/min and hold for 10 min. The peakareas and concentration of 12 C2 to C10 standard acids were used tocalculate the partition coefficients based on the equation: n � KC°,where n is the mass of the specific compound removed by SPME filmfrom solution and C° was the concentration of the compound in thesolution (Dias and Freeman, 1997; Lord and Pawliszyn, 2000). Themasses of the individual monocarboxylic acids were then calculatedbased on the K values and peak areas for each compound, to correct forthe different adsorption efficiency for different compounds. Thestraight chain and branched acids with the same carbon number werefound to have very similar K values (difference �10%). Hence, the Kvalues from standard acids with the same carbon number were used tocompute the concentrations when standard acids were not available.

MCAs in Murchison were identified by GC co-injection with thoseof 12 monoacid standards, and/or by using an HP6890� GC interfacedto 5973 N MSD/DS. �18 monoacids yield mass spectra that arevirtually identical to those found the NIST mass spectral library and theidentification of the compounds were further confirmed by comparisonwith published GC retention time (Yuen and Kvenvolden, 1973;Naraoka et al., 1999). Additional 20 acids were identified by interpre-tation of the mass spectra, which was relatively straight forward forthese simple compounds, and by comparing GC retention times withthose with more firm identifications. In total, we recognized 50 MCAsin Murchison.

2.3. Compound-Specific Carbon and HydrogenIsotopic Analysis

Carbon isotopic analysis of individual low molecular weight mono-carboxylic acids using SPME coupled with GCIRMS has been dem-onstrated previously (Dias and Freeman, 1997; Dias et al., 2002).Carbon isotope analyses of individual compounds were performedusing a gas chromatography—combustion—isotope ratio mass spec-trometer (Thermo Finnigan). An HP 6890 GC was connected to aFinnigan MAT Delta�-XL mass spectrometer via a combustion inter-face. Helium (UHP 5.5 grade) was used as the carrier gas, operating atconstant flow mode with a rate of 1.1 mL min�1. The temperatureprogram and capillary column was identical to those for GC and GCMSanalysis. Compounds separated by GC column were converted to CO2

and H2O through a combustion furnace (0.5 mm ID � 1.5 mm OD �34 cm) operated at 940°C and loaded with CuO and Pt wires as oxidantand catalyst, respectively. Six pulses of CO2 reference gas with known�13C values were injected via the interface to the IRMS, for thecomputation of �13C values of sample compounds. Samples wereanalyzed in triplicates, with standard deviation smaller than � 0.3‰.The �13C values were reported relative to the VPDB standard.

Hydrogen isotope analyses of individual compounds were performedusing a gas chromatography—high temperature conversion—isotoperatio mass spectrometer. The same GC system and temperature pro-gramming were used, except for replacing the oxidation reactor with apyrolysis reactor (Hilkert et al., 1999). Compounds separated by GCwere converted to H2 through a pyrolysis furnace (0.5 mm ID � 1.5mm OD � 34 cm) operated at 1445°C (Quantitative conversion can beachieved by pyrolysis at �1400°C (Burgoyne and Hayes, 1998). Sixpulses of hydrogen reference gas with known �D values were injectedvia the interface to the IRMS, for the computation of �D values ofsample compounds. Typical standard deviation of triplicate analyseswas smaller than � 2‰. The natural abundance of deuterium in organiccompounds is significantly lower than the C-13. To deliver sufficientquantities of compounds for hydrogen isotopic analyses, two to threeconsecutive SPME adsorption-desorption processes were carried outfor the same sample while maintaining the GC oven temperature at30°C. After consecutive SPME desorptions, the GC oven temperatureprogram (same as for GC analyses) was activated. While the peakheights of individual monocarboxylic acids approximately doubled (ortripled), we did not observe discernable peak broadening by this mul-tiple loading process, indicating the compounds were adsorbed in thehead of the GC column stationary phase until GC oven temperature wasraised sufficiently. �D values are reported relative to theVSMOW standard.

Since the carboxyl hydrogen in the monocarboxylic acids is ex-

changeable with water and hydrogens in the stationary phase of the GCcolumn, an isotopic correction is performed to eliminate the contribu-

1075Monocarboxylic acids in carbonaceous meteorites

tion from the carboxyl hydrogen (Sauer et al., 2001; Sessions et al.,2002; Huang et al., 2002). To do so, we converted all the 12 mono-carboxylic acids into the sodium salts, and then determined the �Dvalues of the carbon-bound hydrogens on the sodium salts of the 12monocarboxylic acids, using a thermo-conversion elemental analyzer(TCEA) coupled to the IRMS (Hilkert et al., 1999). Three solidstandards NBS-30 (�D � �66‰), and two pure n-alkanes (C21,�76‰, C25 �126‰, measured using offline combustion) were used tocorrect the TCEA measured values to authentic values, as is routinelydone for water samples (Huang et al., 2002). The values of the mono-acid salts were then compared with �D values of the standard mono-carboxylic acids determined using SPME-IRMS, to correct the D/Hexchange of the carboxyl hydrogen. An excellent correlation betweenthe �D values of acid salts (with �D values ranging from �251 to 79‰)and the �D values measured using SPME-GCIRMS is observed for the6 monoacids: �DSPME � 1.0562 �DTCEA � 16.283 with a R2 � 0.9935,with the offset reflecting the hydrogen isotopic shift due to isotopicexchange. This correlation was used to correct the �D values ofmeteorite MCAs measured by SPME, to obtain the �D values of thecarbon-bound hydrogen. For the MCA standards, �D values aftercorrecting SPME determined values using above equation show excel-lent agreement with those determined after converting to sodium salt,with Y � 1.0001X �0.0105, R � 0.9942, where Y axis is correctedSPME �D values of monoacids, and X axis is the TCEA measured �Dvalues of monoacid salts. Since we used identical conditions (water,GC column) for monoacid standards and meteorite acids, we correctedour SPME measured �D values of Murchison monoacids using theabove relationship (�DSPME � 1.0562 �DC-bound H � 16.283) derivedfrom acid standards.

3. RESULTS AND DISCUSSIONS

Since carbon-based organic compounds are the buildingblocks of life on earth, a critical question in studying the originof life is how such compounds could have originated on Earth.A number of hypotheses have been put forward, such as deliv-ery of organics to earth by carbonaceous meteorites, synthesisof organics by meteorite impact, synthesis of organic com-pounds by hydrothermo-reactions in the hydrothermal-ventsaround the ocean ridges (see a review by Chyba and Sagan,1992). The meteorite delivery hypothesis is supported by thediscovery of a large array of organic compounds that are foundin terrestrial biology in various carbonaceous meteorites, in-cluding amino acids and carbohydrates (Cronin and Chang,1993 and references therein; Cooper et al., 2001; Pizzarello etal., 2001; Pizzarello and Huang, 2002). MCAs are the mostabundant organic compounds (10 to 100 times more abundantthan amino acids) in Murchison and Murray meteorites. There-fore, any theory addressing the origin of organic compounds incarbonaceous meteorites must explain the characteristics ofMCAs.

3.1. Molecular Distributions and Concentrations

Overall distribution and comparison with previous data:Yuen and Kvenvolden (1973) and Lawless and Yuen (1979)were the first to characterize and quantify MCAs from aqueousextracts of the Murchison meteorite. The procedure has alsobeen adopted by later workers (e.g., Shimoyama et al., 1989;Naraoka et al., 1999). However, the procedure involves awater-solvent extraction (with DCM) step to acquire MCAsfrom the acidified aqueous solution, followed by evaporativeconcentration of solvent from 5 mL to a small volume conve-nient for GC injection. Because low molecular weight acids(especially C to C acids) are highly soluble in water and

2 4

volatile, significant amount of acids may be lost during the

extraction (e.g., at 25°C, the Kow for acetic acid is 0.68) andsolvent concentrating steps. Differences in octanol-water par-titioning coefficients (Kow) between different acids affect thecompound distributions (Shimoyama et al., 1989).

The distributions of MCAs in our Murchison sample (Fig. 1)show similarities as well as differences compared with pub-lished data by Yuen and Kvenvolden (1973). While all the 18monoacids found in Yuen and Kvenvolden (1973) are alsoidentified, many more (�50) MCAs are found in our analyses.The straight chain acids extend to C10, instead of C6. Peakresolution in Figure 1 is much higher than published chromato-grams by Yuen and Kvenvolden (1973), which display clearpeak tailings. The difference in peak resolution can be attrib-uted to the better capillary GC column and better stationaryphase used in this study than that used previously (stainlesssteel capillary column).

The acetic acid in our Murchison sample shows significantlyhigher abundance than the propanoic acid and other longerchain acids (Fig. 1; Table 1). In contrast, the acetic acidabundance is lower than longer chain acids in Yuen and Kv-envolden (1973) and Lawless and Yuen (1979). This differencecan be attributed to the different procedures used. SPME elim-inates the need to recover the highly water soluble MCAs fromwater using water-solvent extraction, and to evaporate thesolvent to concentrate the acids for injection into GC (Eisertand Levsen, 1996; Lord and Pawliszyn, 2000). A more quan-titative recovery of the Murchison MCAs was conducted byusing an ion chromatograph, showing that acetic acid wasindeed about three times more abundant than the propanoicacid (Yuen et al., 1984), similar to our acetic/propanoic acidratio (Table 1). Similarly, Naraoka et al. (1999) found over 30(C2 to C12) monocarboxylic acids from an Antarctic carbona-ceous meteorites of Japanese collection (A-881458), in whichthe acetic acid is also more abundant than the propanoic acid.

Formic acid was not reported in Murchison in publishedstudies (Yuen and Kvenvolden, 1973; Lawless and Yuen, 1979;Yuen et al., 1984), probably due to compound loss duringsample preparation. A later study using ion chromatographfound formic acid (Briscoe, 1994), but was inconclusive of itsextraterrestrial origin and never formally published. Formicacid is found in our extract of Murchison sample (Fig. 1; Table1). Formic acid is known to be an interstellar molecule (Turner,1989). The identification of the compound is confirmed bycoeluting with the authentic standard on GC, and by carefullyexamining the mass spectrum. Notably, formic acid elutes onthe GC later than the acetic acid, despite the boiling point offormic acid is 100 to 101°C, lower than acetic acid (117 to 118°C). This reversal is apparently because of the molecular inter-actions with the stationary phase, with formic acid havingstronger affinity than acetic acid. Unfortunately, the low abun-dance of formic acid in Murchison precluded compound-spe-cific isotopic analyses. To further verify the origin of the formicacid, we analyzed an Antarctic carbonaceous meteorite,EET96029.20, a type C2 chondrite, using the same procedure.A much higher abundance of formic acid is found inEET96029.20 (Fig. 2), with its �D values suggesting interstel-lar origin (more discussion later).

Branched acids: A large suite of branched MCAs werefound in our analyses (Fig. 1; Table 1). In fact, virtually all the

peaks eluting between the straight chain acids (peaks 22, 35,

1076 Y. Huang et al.

42, 46, 49 and 50) are branched MCAs. Our procedure involveda step to evaporate water (used to extract soluble compounds)in basic conditions. This step would have lead to the volatil-ization of low molecular weight aldehydes, ketones, alcoholsand amines that are know to be present in Murchison (Croninand Chang, 1993). However, because the acids are in the saltform in the basic conditions, they are not affected (see Samplesand Analytical Methods). The mass spectra of the branchedacids display several characteristics that make spectral inter-pretation relatively straight forward: 1) a clear loss of proton,M�-1; 2) a strong M�-15, some times followed by a M�-29 in

Fig. 1. (a) A gas chromatogram of monocarboxylic acshown in Table 1; (b) Normalized relative abundance of smonocarboxylic acids (B-Cn) in Murchison.

�C6 acids; 3) either side of the branch point yield strong and

characteristic mass ions (Fig. 3). The identification of thebranched acids is further assisted by comparing GC retentiontime with our standards and published retention time data (e.g.,Naraoka et al., 1999). The branched structures appear to behighly random, with alkyl substitutions occurring in nearly allpossible positions. With increasing carbon numbers (e.g.,�C10), there are more possible positions to be substituted onthe alkyl chain or branches and possible ways of substitutions(methyl, ethyl, propyl- etc.), resulting in more possiblebranched acid isomers. The well resolved GC peaks at thelower carbon number acids gradually give way to the humps of

he Murchison meteorite. Peak numbers denote 50 acidshain (Cn, where n denotes carbon number) and branched

ids in ttraight c

poorly resolved peaks that are virtually all branched acids (Fig.

ctions.

1077Monocarboxylic acids in carbonaceous meteorites

1). Recognizing this large diversity of nearly randomly substi-tuted branched MCAs is more important for the purpose of ourdiscussion than further attempts to precisely construct the struc-tures of all the branched acids.

The suite of branched acids we found in Murchison ap-pears to be more extensive than those observed in theAntarctic meteorite A-881458 (Naraoka et al., 1999). Therelative abundance ratio of branched over straight MCAs ishigher in our sample of Murchison than in A-881458 (Fig.1a,b). Given that Murchison is renowned for containing thelargest suite of functionalized organic compounds (Cronin

Table 1. �D and �13C values (Mean � S.D.*)

No. Name Co

1 Acetic2 Formic3 Propanoic4 2-methylpropanoic5 2,2-dimethylpropanoic6 Butanoic7 2 and 3-methylbutanoic8 3,3-dimethylbutanoic9 2,2-dimethylbutanoic

10 Pentanoic11 2,3-dimethylbutanoic12 2-ethylbutanoic acid13 2-methylpentanoic14 2,2-dimethylpentanoic15 3-methylpentanoic16 4-methylpentanoic17 2,3-dimethylpentanoic18 2-ethyl,3-methylbutanoic19 2,4-dimethylpentanoic20 2-ethyl,2-methylbutanoic21 3,4-dimethylpentanoic22 Hexanoic23 2,2-dimethylhexanoic24 2,3-dimethylhexanoic25 2-methylhexanoic26 2-ethylpentanoic27 3-methylhexanoic28 2-ethyl, 4-methylpentanoic29 3-ethylpentanoic30 4-methylhexanoic31 3,5-dimethylhexanoic32 3,4-dimethylhexanoic33 2-propylpentanoic34 2-ethylhexanoic35 Heptanoic36 2-methylheptanoic37 3,6-dimethylheptanoic38 3-methylheptanoic39 2,3-dimethylheptanoic40 4-methylheptanoic41 2-ethylhexanoic42 Octanoic43 2-methyloctanoic44 3-methyloctanoic45 4-methyloctanoic46 Nonanoic47 2-methylnonanoic48 Caprolactam49 Decanoic50 Benzoic

* S.D., standard deviation of three sequential inje

and Chang, 1993), it is not surprising that MCAs in Murchi-

son should be so abundant and diverse. The complex mixtureof the branched MCAs is consistent with an origin throughrandom gas phase reactions involving radicals, ions andmolecules. These reactions are exothermic and thermody-namically favored with minimal activation energies that canproceed at very low temperature (10 to 100 K) (Herbst,1995). The radicals could allow formation of branches atvirtually any point on the alkyl chain of the carboxylic acids.The random nature of the monocarboxylic acids observed inour study clearly supports the indigenous, abiotic origin ofthese MCAs in Murchison rather than terrestrial contami-

ocarboxylic acids in the Murchison meteorite.

ol/g) �D � S.D. �13C � S.D.

19.1 � 14.2 �7.7 � 0.2

681.7 � 31.3 7.3 � 1.61516.7 � 76.3 6.3 � 1.2

620.8 � 32.2 0 � 0.41390.6 � 54.0 7.7 � 0.61778.7 � 43.3 3.2 � 1.52024.0 � 98.9 6.1 � 1.3774.7 � 13.5 �10.4 � 0.3

1160.8 � 42.1 3.0 � 2.69.0 � 0.9

1115.8 � 72.1 10.7 � 1.132.5 � 0.7

1615.6 � 40.0 5.4 � 0.21740.8 � 31.2 7.5 � 1.2

6.6 � 2.61.3 � 0.19.0 � 1.23.7 � 2.43.4 � 0.9

192.4 � 11.8 �18.1 � 0.320.4 � 9.215.2 � 2.8

1594.0 � 97.1 11.3 � 1.3�2.4 � 4.3

1463.8 � 44.7 2.7 � 1.24.8 � 1.1

14.0 � 1.71449.1 � 44.1 4.3 � 0.3

�21.7 � 3.64.4 � 2.90.6 � 0.7

1742.9 � 98.4 �1.3 � 0.5100.5 � 15.0 �18.2 � 0.4

1271.4 � 25.1 �8.5 � 0.3�4.7 � 3.7

1358.6 � 11.7 5.1 � 1.61610.3 � 76.6 �3.4 � 2.51128.2 � 87.5 5.4 � 0.61624.7 � 22.9 5.3 � 1.9115.6 � 1.70 �25.9 � 0.3

1378.9 � 64.6 2.8 � 1.51334.4 � 34.0 �7.1 � 0.11916.2 � 46.7 14.2 � 4.0�76.5 � 16.2 �24.7 � 0.6103.5 � 37.7 �5.9 � 0.5386.1 � 22.4 �31.8 � 1.5

�61.7 � 12.9 �22.6 � 0.8399.9 � 15.8 2.8 � 1.0

of Mon

nc. (nm

770.142.0

226.685.775.8

141.1229.1

56.071.7

185.977.837.498.374.3

115.382.685.764.360.260.657.1

217.762.8

193.199.7

107.5148.4

71.592.1

132.462.3

109.867.1

127.4168.8140.6

84.4136.6

97.598.1

105.6106.7111.0

93.4100.3181.9

59.6123.9

53.4137.1

nants.

cids (C

1078 Y. Huang et al.

3.2. Carbon Isotopic Ratios of Individual MCAs

Carbon isotopic analyses of 6 MCAs in Murchison werepreviously conducted after isolation of individual acids bycryogenic transfer and ion chromatography, followed by furtherseparation and conversion using a GC combustion system(Yuen et al., 1984), which gave more quantitative recovery ofvolatile acids than the one used in Yuen and Kvenvolden(1973) and Lawless and Yuen (1979). The Yuen et al. (1984)procedure using an ion chromatograph reduced loss of volatileacids during evaporative concentration of solvents which canlead to large evaporative isotopic fractionation for low molec-ular weight compounds (Wang and Huang, 2001; 2003). How-ever, the lengthy procedure clearly limited the number ofanalyzable compounds. Only a few major components (6 acids,C2 to C5) were isolated (purity of the isolates were not reported,though) and measured for carbon isotopic compositions (Yuenet al., 1984).

Coupling SPME with GCIRMS: Because the sample intro-duction into the GC injection port using SPME does not in-volve a solvent concentration step, we avoid preferential loss oflow molecular weight acids and possible evaporative isotopicfractionation. A previous study (Dias and Freeman, 1997) hasshown that sample introduction by SPME does not cause car-bon isotopic fractionation of monocarboxylic acids. Zwank etal. (2003) also demonstrate that SPME extraction does notaffect �13C values of petroleum and chlorinated hydrocarbons,with the only exception for CCl4 which shows an unexplainable7‰ deviation.

Fig. 2. The lower molecular weight monocarboxylic ameteorites and their �D values.

We have determined carbon isotopic ratios of 48 individual

MCAs from Murchison meteorite using the SPME-GCIRMSapproach (Table 1). The �13C values of individual monocar-boxylic acids range from �31.8 to � 32.5‰. Notably, thestraight chain acids have lower �13C values. Except for pro-panoic and butanoic acids with values � 7.3 and 0‰, respec-tively, all other straight chain MCAs have negative �13C values(�7.7 to �25.9‰), and there is an apparently trend of lower�13C values with increasing chain length, except for the aceticacid (Fig. 4 and 5). The low �13C value for acetic acid (�7.7‰)is surprising, since previous data (Yuen et al., 1984) indicateacetic acid has the highest �13C values (�22.7‰) among the 6MCAs measured. The likely source for the low �13C value ofacetic acid is the presence of acetic acid contamination in ourspecimen of Murchison sample, which had been stored on ashelf at Smithsonian Institution at room temperature for the last34 yr. Acetic acid is one of most common products of microbialdegradation of organic matter and is in high abundance in theair. The sample used by Yuen et al. (1984) could have beenmuch better protected from organic contamination than oursample. In addition, the Murchison samples are known to behighly inhomogeneous, with significant differences in molecu-lar and isotopic compositions in different stones of the mete-orite (Cronin and Chang, 1993; Pizzarello et al., 2003). Forexample, the �13C values of different preparations of MCAsfrom Murchison by the same research group gave �13C valuesof �3, 0, �1, �6.7, and �D values � 377, �581, �679, and697‰, respectively (Epstein et al., 1987; Krishnamurthy et al.,1992). The maximum difference of �13C values between dif-

1-C6) in the Murchison and EET96029.20 carbonaceous

ferent preparations is as large as 9.7‰ (and �D 320‰).

gment (

1079Monocarboxylic acids in carbonaceous meteorites

Chain length and �13C values: The decreasing �13C valueswith carbon chain length for straight chain MCAs (except forthe possibly contaminated acetic acid) confirm the previousobservation by Yuen et al. (1984). In addition, we also ob-

Fig. 3. Mass spectra of monocarboxylic acids. Mass fraor quaternary carbon.

13

Fig. 4. Distribution of � C and �D values of straight and brancmeteorite, where n � normal, M � methyl, DM � dimethyl.

served �13C decrease trends (though less conspicuous than thestraight chain acids) for 2-methyl, 3-methyl and dimethyl sub-stituted branched mono-acids (Fig. 4). The data are consistentwith a kinetically controlled carbon addition reaction during

m/z) signals are enhanced at positions adjacent to tertiary

h chain monocarboxylic acids (C3-C10) in the Murchison

meteord by Cr

1080 Y. Huang et al.

synthesis of these compounds as has been suggested by Yuen etal., (1984). Because ions or radicals containing C-12 reactfaster than those containing C-13, during carbon chain growthfrom lower carbon number acids, the longer chain acids wouldbe more enriched in C-12 than the shorter chain counterparts.

Straight chain vs. branched acids: Yuen et al. (1984) ob-served a � 4‰ carbon isotopic enrichment in the two branchedC4 and C5 acids relative to the C4 and C5 n-acids, but did notelaborate the results because of the limited data. Our datademonstrate that the �13C values of the branched MCAs areconsistently higher than those of the straight chain counterparts(Table 1; Fig. 4 and 5). The 13C enrichment ranges from a fewper mil to over 40 per mil (obtained by subtracting the �13Cvalues of straight chain acids from the correpondent branchedacids) and is of much larger scale than observed previously.The �13C values of the 2, 2-dimethyl monocarboxylic acids(i.e., those containing a quaternary carbon) appear to be thecompounds most enriched in 13C (Fig. 4).

The origin of these isotopic distinctions may come fromseveral sources, 1) molecular precursors with different carbonisotopic ratios; 2) different extent of carbon isotopic fraction-ation during the synthesis of the straight chain and branchedMCAs; and 3) straight chain and branched acids may haveformed in different regions of molecular clouds with differentconditions (e.g., temperature, density); and 4) straight chainacids may have different sources with lower �13C values. If theassumption is correct that a kinetically controlled carbon addi-tion reaction was responsible for the synthesis of both straightchain and branched compounds (Yuen et al., 1984), carbonaddition at either terminal points or branched positions wouldhave derived from the same molecular precursors. Branchedacids are thermodynamically more stable and have lower free

Fig. 5. The �13C vs. �D plot for 50 acids in Murchisonmonocarboxylic, dicarboxylic, and hydroxy acids reporte

energy of formation (Gfo) than the straight chain acids (Stull et

al., 1969) (Fig. 6). Therefore, forming branched acids couldhave been a favored and faster process than forming straightchain acids and may have been less selective of 12C and 13Cprecursor species, resulting in smaller carbon isotopic fraction-ation.

It is also possible that branched acids may have been selec-tively formed in different environments or regions of densemolecular clouds from those forming predominantly straightchain acids. To explore this possibility, the temperature-depen-dence of the relative equilibrium populations of linear andcorresponding branched acids was calculated with the Jaguarsoftware package. (Jaguar 3.5, 1998, Schroedinger, Inc., Port-land, OR) A B3LYP model with 6–31G* basis set was used inall DFT (Density Functional Theory) calculations for monocar-boxylic acids. Pulay et al.’s Modified Scaled Quantum Me-chanical Force Fields (SQM) method (Baker et al., 1998) wasused for scaling of vibration frequencies to improve calcula-tions of thermochemical properties which were computed in therange from 30.0 K to 300.0 K. The absolute values of thecalculated Gibbs free energies for each of the acids are notsignificant because only the difference in Gibbs free energy isused for the calculations of the equilibrium constant shown inFigure 6. Furthermore, the temperature-dependence of the cal-culated equilibrium constants for the conversion betweenbranched and straight chain acids is primarily caused by theacids’ vibrational spectra. Vibrational modes with less the 10cm�1 were ignored. These calculations demonstrate that excessbranched acids produced relative to straight chain acids shoulddecrease as temperature decreases (Fig. 6). Therefore, thehigher �13C values of branched acids relative to straight chainacids cannot be explained by preferentially increased produc-tion of branched acids at lower temperature. Alternatively,

ite, whereas A denotes a �13C vs. �D plot for Murchisononin and Chang (1993).

straight chain acids may have multiple sources. For example,

1081Monocarboxylic acids in carbonaceous meteorites

straight chain acids are more likely to receive terrestrial con-tamination with low �13C values, since branched low molecularweight MCAs are rare or absent in terrestrial biology. Ourcurrent data do not allow us to differentiate these possiblesources for the isotopic distinction between the straight chainand branched acids. To better understand this phenomenon,several studies may be important: 1) analyzing more pristinecarbonaceous meteorites; 2) conducting experiments to synthe-size straight chain and branched MCAs in conditions similar tothe interstellar medium, and determining the carbon isotopicvalues of the acids.

3.3. Hydrogen Isotopic Ratios of MCAs

Our previous study has demonstrated that C-bound hydrogenin dicarboxylic acids (C3 to C5) does not exchange significantlywith water during the hot water extraction at 110°C (Fuller andHuang, 2003a). The C-bound �-protons in MCAs are lessacidic than those in dicarboxylic acids, and therefore, even lesslikely to exchange with water. This ensures that the �D valueswe determined from carboxylic acids in Murchison are indig-enous.

We determined �D values for 32 MCAs in Murchison (Table

Fig. 6. Equilibrium constants for the conversion between branchedand straight chain acids as based on equilibrium thermodynamics:2-MPA � 2 methyl-propanoic acid; BA � Butanoic acid; 2,2-DMPA� 2,2-dimethyl-propanoic acid; PA � pentanoic acid. The right scalesdenote by how much the concentration of straight chain acids is largerthan that of branched acids.

1). Except for the C9 and C10 straight chain mono acids, all the

acids have positive �D values which are unlikely to be found incompounds of living organisms on Earth. Organisms obtaintheir hydrogen primarily from the water they use, and oftenstrongly select H over D during biosynthesis, the �D values oftheir compounds are lower than the water they use (Epstein etal., 1987). Except for some rare cases (e.g., lakes that haveundergone extensive evaporation), the water on earth have 0‰(ocean water) or negative �D values, and hence the organiccompounds produced from such waters would have even morenegative �D values. The hydrogen isotopic data thus furthersupport the extraterrestrial origin of the MCAs in Murchison.

Examination of the �D values reveals some interestingtrends, some of which are similar to those observed in carbonisotopic data. Except for the acetic acid (which may have beencontaminated as discussed above), higher carbon acids aremore enriched in light isotope, hydrogen, than the shorter chaincounterparts. This trend is less clear for C3, C4 and C5 acids,but becomes quite consistent from C5 to C10 acids (Fig. 4). Forthe branched acids, the trend is also present but shows largevariations (Fig. 4). These results would also support the kinet-ically controlled one carbon addition reactions as an importantmechanism during the formation of MCAs (Yuen et al., 1984).However, the trend of D isotopic depletion with chain lengthappears to be less conspicuous than for the 13C depletion (Fig.4 and 5). This may reflect the fact that only new carbon-carbonbond is directly involved during chain length growth by addi-tion of one carbon moieties, rather than C-H bond. Although aradical containing D atoms would have been less reactive dueto a greater mass than a radical containing only H atoms, theimpact of D on the isotopic fractionation during the formationof new C-C bond would have been smaller than the impact ofcarbon isotope species which participate directly in the forma-tion of new C-C bonds.

Rather consistent is the finding that the branched MCAs aremore enriched (up to 1500‰) in deuterium than their straightchain counterparts (Fig. 4 and 5). These �D values are substan-tially higher than the bulk �D values of the MCA fractions fromMurchison, but are more similar in range to those observed foramino acids (Epstein et al., 1987; Krishnamurthy et al., 1992).The isotopic distinctions are even more pronounced than thoseobserved in our carbon isotopic data. Stability of the moleculesalone may be difficult to explain the observed hydrogen isoto-pic difference, since H/D atoms are not directly involved in theformation of new C-C bonds during the synthesis of MCAs.Combining the data from carbon and hydrogen isotopic anal-yses, it appears that straight chain and branched acids may havesimply been formed in different interstellar environments withdifferent molecular precursors and conditions. Contaminationof straight chain acids from terrestrial sources cannot be ex-cluded from our data.

The hydrogen isotopic value of formic acid in EET96029.20indicates its extraterrestrial origin. Except for the formic acid,this meteorite contains much lower amounts of MCAs (espe-cially branched MCAs) compared with Murchison. Straightchain MCAs of the same carbon number in EET96029.20 alsohave lower �D values. Since Antarctic samples have been in theEarth environment for an unknown amount of time, contami-nation from terrestrial sources is much more likely than

Murchison. Fortunately, the �D values of the several straight

1082 Y. Huang et al.

chain acids are still high enough for us to recognize the con-tributions of interstellar sources to the MCAs in EET96029.20.

3.4. Comparison with other Compounds or Meteoritesand Implications

Because isotopic compositions of organic compounds pro-vide important clues about their origins and their genetic rela-tionship with other compounds, extensive research has beencarried out on the carbon, hydrogen, nitrogen isotopic ratios oforganic components in carbonaceous meteorites (Robert andEpstein, 1982; Kerridge, 1983,1985; Epstein et al., 1987; Ker-ridge et al., 1987; Engel et al., 1990; Pizzarello et al., 1991,1994; Krishinamurthy et al., 1992; Engel and Macko, 1997;Cooper et al., 1997; Alexander et al., 1998; Sephton et al.,2003). A wide range of carbon and hydrogen isotopic valueshave been found for different organic fractions, with typicallyhigher values for more polar compound classes. There is ageneral trend of correlation between the �D and �13C values,which is thought to indicate that both H and C atoms werederived from a homogeneous isotopic source, and experiencedsimilar process leading to the similar model of isotopic frac-tionation (Cronin and Chang, 1993). The high deuterium andC-13 contents relative to terrestrial values for the organicfractions indicate that the organics may have been formed fromparental interstellar matter at low temperature (Robert andEpstein, 1982; Kerridge, 1983), whereas the large variabilitypoints to heterogeneities in the isotopic compositions in theinterstellar clouds.

A plot of �13C and �D values of individual monocarboxylicacids shows a good correlation between the two differentisotopic values (Fig. 5). Notably, the branched acids are wellseparated from the straight chain acids, with the branched oneshaving consistently higher values plotted on the upper righthand side. In comparison with the plot based on bulk isotopicvalues (Cronin and Chang, 1993), the individual acids aredivided into patches in the new plot (Fig. 5), with the bulkisotopic values fall in between the two end members. The �Dvalues of branched acids are more similar to what was reportedfor the amino acids from bulk fractions (Epstein et al., 1987).

The data from MCAs can be compared with those of dicar-boxylic acids from Murchison and Tagish Lake meteorites(Pizzarello et al., 2001; Pizzarello and Huang, 2002). Dicar-boxylic acids from Tagish Lake sample have �13C valuesranging from � 5.5 to � 22.5‰, �D ranging from � 1116 to� 1322 ‰, whereas Murchison values range from � 21.4 to� 28.1‰ (�13C) and � 389 to � 1550‰ (�D). It is notable thatthe molecular distributions of soluble organic compounds in theTagish lake samples show marked differences from Murchison,with the majority of the compound classes being either absent,e.g., the hydroxyacids, or far less abundant, such as the aminoacids and MCAs (Pizzarello et al., 2001). In addition, westudied carbon isotopic ratios of individual dicarboxylic acidsfrom the Antarctic meteorite ALH 85013.50 (Fuller and Huang,2003b) and obtained dicarboxylic acid distributions that areremarkably similar to the Murchison and Tagish Lake samples.The carbon isotopic values for succinic acid (�25.1), glutaricacid (�23.5) and 2-methyl glutaric acid (�27.5) are also sim-ilar to Tagish Lake meteorite. Clearly, the �13C and �D values

of MCAs in Murchison are in the same range as observed in the

dicarboxylic acids, which would indicate similar conditions/environments for their formation.

Cronin and Chang (1993) suggest that MCAs may be mostlyunaltered interstellar molecules. Our identification and �D anal-ysis of formic acid, the known interstellar MCA (Turner, 1989),in Murchison and EET96029.20 support this possibility. Thealternative possibility is the parent body hydrolysis of alkylnitriles (e.g., methyl cyanide, ethyl cyanide). Interstellar cloudsare known to contain an extensive suite of unsaturated nitrilesand cyanopolynes, extending to C11. Reduction and hydrolysisof these precursors in the parent body could form straight chainMCAs. However, interstellar cyanonitriles are dominated bythose containing an odd number of carbon atoms (HC3N,HC5N, HC7N, HC9N, HC11N), with the only even carbon-number methylcyanonitrile (CH3C3N) whose concentration isone order of magnitude lower than HC5N (Irvine, 1998). As-suming that these compounds would be reduced and hydro-lyzed to MCAs, this would mean that odd-numbered MCAswould be far more abundant in meteorites than even-numberones. The odd carbon dominance is not observed in our results.Contamination by terrestrial even carbon-number MCAs can beexcluded since odd-carbon MCAs do not show systematicallyhigher C and H isotopic ratios than the even-carbon ones (Table1; Fig. 2). Cronin and Change (1993) suggest that, to yield thecomplete diversity of branched structures observed in meteor-ites, the major route of synthesis would have been free radicalreactions rather than reduction of unsaturated precursors. Theradicals such as cyano, amino, hydroxyl, and alkyl groups cancombine on the dust grains (either interstellar or parent bodyenvironments) to form the complete structural diversity. Thispathway of MCA formation is in agreement with our results.

4. CONCLUSIONS

Our re-examination of monocarboxylic acids in Murchisonrevealed a large suite (�50) of previously unreported com-pounds. The number of MCAs recognized in our study appearsto exceed those observed in Antarctic meteorites and is con-sistent with Murchison containing the most complete suite oforganic molecules studied so far. We have confirmed the in-digenous nature of formic acid in Murchison and an Antarcticcarbonaceous meteorite, EET96029.20, with its �D value con-sistent with an interstellar source. The dominance of branchedover straight chain acids in Murchison can be attributed togreater thermodynamic stability of branched acids relative tothe straight chain counterparts.

SPME has been demonstrated to an excellent sample intro-duction approach to GC analyses for the volatile organic com-pounds such as MCAs in meteorites. The method avoids com-pound losses due to evaporation during solvent concentratingstep, and better retain the molecular and isotopic integrity of thelow molecular weight, water soluble and volatile organic com-pounds such as MCAs. SPME also shortens the sample pro-cessing time by many folds since it is not necessary to performliquid-liquid (water-solvent) extractions. SPME does not causecarbon isotopic fractionation. Our new data further show thatSPME is equally suitable for compound-specific hydrogen iso-tope analyses.

Carbon isotopic ratios, and to a less extent hydrogen isotope

ratios of straight chain monocarboxylic acids show an enrich-

1083Monocarboxylic acids in carbonaceous meteorites

ment of lighter isotopes with increasing chain length, support-ing a kinetically controlled carbon addition reaction involvingradicals. Both carbon and hydrogen isotope ratios show a cleardistinction between the straight chain and branched monocar-boxylic acids, with branched compounds significantly enrich-ment in heavier isotopes. There is a general correlation betweenthe carbon isotopic values and hydrogen isotopic values forindividual acids. The data suggest differences in synthetic pathsduring the formation of normal and branched monocarboxylicaicds. However, the majority of acids have carbon or hydrogenisotopic values that far exceed the values encountered in bio-logic compounds produced on Earth, confirming the indigenousnature of these Murchison compounds.

Synthesis of branched acids may have experienced less ki-netic isotopic fractionation from 13C and D enriched precursorsif branched acids are formed at greater rate than the straightchain acids. This could partially explain the higher �13C and �Dvalues in branched acids. However, other processes may haveinvolved. For example, partial contamination of straight chainacids from terrestrial sources cannot be excluded based on ourdata. Until more pristine meteorites are examined, or moreremote detection of large molecules in the interstellar dustclouds, our current data are inadequate to differentiate theseprocesses.

Acknowledgments—This work was supported by NASA Exobiologygrants NAG5–10665 and NNG04GJ34G to YH. MRA thanks a fel-lowship from CAPES. We thank Dr. Etienne Deloule and Dr. MarkSephton for valuable comments and suggestions that helped improvethe manuscript.

Associate editor: C. Koeberl

REFERENCES

Alexander C. M. O., Russell S. S., Arden J. W., Ash R. D., GradyM. M., and Pillinger C. T. (1998) The origin of chondritic macro-molecular organic matter: a carbon and nitrogen isotope study.Meteor. Planet. Sci. 33, 603–622.

Baker J., Jarzecki A. A., and Pulay P. (1998) Direct Scaling of Prim-itive Valence Force Constants: An Alternative Approach to ScaledQuantum Mechanical Force Fields. J. Phys. Chem. A. 102, 1412–1424.

Briscoe J. F., (1994) The determination of formic and acetic acids inchondritic meteorites using ion exclusion chromatography. Ph.D.thesis, Arizona State University.

Burgoyne T. W. and Hayes J. M. (1998) Quantitative production of H2by pyrolysis of gas chromatographic effluents. Anal. Chem. 70,5136–514.

Chyba C. and Sagan C. (1992) Endogenous production, exogenousdelivery and impact-shock synthesis of organic molecules, an in-ventry for the origins of life. Nature 355, 125–132.

Cooper G. W., Thiemens M. H., Jackson T. L., and Chang S. (1997)Sulfur and hydrogen isotope anomalies in Meteorite sulfonic acids.Science 277, 1072–1074.

Cooper G. W., Kimmich N., Belisle W., Sarinana J., Brabham K., andGarrel L. (2001) Carbonaceous meteorites as a source of sugar-related organic compounds for the early Earth. Nature 414, 879–882.

Cronin J. R. and Chang S. (1993) Organic matter in meteorites:molecular and isotopic analysis of the Murchison meteorite. In TheChemistry of Life’s Origins (eds. J. M. Greenberg, C. X. Mendoza-Gómez and V. Pirronello), pp. 209–258. Kluwer Academic Pub-lishers.

Cronin J. R. and Pizzarello S. (1990) Aliphatic hydrocarbons of theMurchison meteorite. Geochim. Cosmochim. Acta 54, 2859–2868.

Dias R. F. and Freeman K. H. (1997) Carbon isotope analysis ofsemivolatile organic compounds in aquesou media using solid-phase microextraction and isotope ratio monitoring GC/MS. Anal.Chem. 69, 944–950.

Dias R. F., Freeman K. H., Lewan M. D., and Franks S. G. (2002) �13Cof low-molecular-weight organic acids generated by the hydrouspyrolysis of oil-prone source rocks. Geochim. Cosmochim. Acta 66,2755–2769.

Eisert R. and Levsen K. (1996) Solid-phase microextraction coupled togas chromatography: a new method for the analysis of organics inwater. J. Chromat. 733, 143–157.

Engel M. H., Macko S. A., and Silfer J. A. (1990) Carbon isotopecomposition of individual amino acids in the Murchison meteorite.Nature 348, 4719.

Engel M. H. and Macko S. A. (1997) Isoptopic evidence for extrater-restrial non racemic amino acids in the Murchison meteorite. Na-ture 389, 26519.

Epstein S., Krishnamurthy R. V., Cronin J. R., Pizzarello S., and YuenG. U. (1987) Unusual stable isotoe ratios in amino acid and car-boxylic acid extracts from the Murchison meteorite. Nature 326,477–479.

Fuller M. and Huang Y. (2003a) Quantifying hydrogen-deuteriumexchange of meteoritic dicarboxylic acids during aqueous extrac-tion. Meteor. Planet. Sci. 38, 357–363.

Fuller M. and Huang Y. (2003b) Molecular isotopic characterization ofthe ALH 85013.50 meteorite: defining the extraterrestrial organiccompounds. Lunar Planet. Sci. XXXIII, Lunar Planet. Inst., Hous-ton. #1237 (abstr.).

Hayes J. M., Freeman K. H., Popp B. N., and Hoham C. H. (1990)Compound-specific isotopic analysis: A novel tool for reconstruc-tion of ancient biogeochemical processes. Org. Geochem. 16,1115–1128.

Herbt E. (1995) Chemistry in the interstellar medium. Annu. Rev. Phys.Chem. 46, 27–54.

Hilkert A. W., Douthitt C. B., Schlüter H. J., and Brand W. A. (1999)Isotope ratio monitoring gas chromatography/mass spectrometry ofD/H by high temperature conversion isotope ratio mass spectrom-etry. Rapid Comm. Mass Spectr. 13, 1226–1230.

Huang Y., Freeman K. H., Eglinton T. I., and Street-Perrott F. A.(1999a) �13C analyses of individual lignin phenols in the lacustrineenvironment: a novel proxy for deciphering the past terrestrialvegetation changes. Geology 27, 471–474.

Huang Y., Street-Perrott F. A., Perrott F. A., Metzger P., and EglintonG. (1999b) Glacial—interglacial environmental changes inferredfrom the molecular and compound—specific �13C analyses of sed-iments from Sacred Lake, Mt Kenya. Geochim. Cosmochim. Acta63, 1383–1404.

Huang Y., Dupont L., Sarnthein M., Hayes J. M., and Eglinton G.(2000) Mapping of C4 plant input from North West Africa into theNorth East Atlantic sediments. Geochim. Cosmochim. Acta 64,3505–3513.

Huang Y., Shuman B., Wang Y., and Webb T. III (2002) Hydrogenisotope ratios of palmitic acid in lacustrine sediments record late-Quaternary climate variations. Geology 30, 1103–1106.

Irvine W. M. (1998) Estraterrestrial organic matter: a review. Orig. LifeEvol. Biosph. 28, 365–383.

Kerridge J. F. (1983) Isotopic composition of carbonaceous-chondritekerogen: evidence for an interstellar origin of organic matter inmeteorites. Earth Planet. Sci. Lett. 64, 186–200.

Kerridge J. F. (1985) Carbon, hydrogen and nitrogen in carbonaceouschondrites: abundances and isotopic compositions in bulk samples.Geochim. Cosmochim. Acta 49, 1707–1714.

Kerridge J. F., Chang S., and Shipp R. (1987) Isotopic characterisationof kerogen-like material in the Murchison carbonaceous chondrite.Geochim. Cosmochim. Acta 51, 2527–2540.

Krishnamurthy R. V., Epstein S., Cronin J. R., Pizzarello S., and YuenG. U. (1992) Isotopic and molecular analyses of hydrocarbons andmonocarboxylic acids of the Murchison meteorite. Geochim. Cos-mochim. Acta 56, 4045–4058.

Lawless J. G. and Yuen G. U. (1979) Quantification of monocarboxylicacids in Murchison carbonaceous meteorite. Nature 282, 396–398.

Lord H. and Pawliszyn J. (2000) Evolution of solid-phase microextra-tion technology. J. Chromatogr. A. 885, 153–193.

1084 Y. Huang et al.

Naraoka H., Shimoyama A., and Harada K. (1999) Molecular distri-bution of monocarboxylic acids in Asuka carbonaceous chondritesfrom Antarctica. Orig. Life Evol. Biosph. 29, 187–201.

Pizzarello S., Krishnamurthy R. V., Epstein S., and Cronin J. R. (1991)Isotopic analyses of amino acids from the Murchison meteorite.Geochim. Cosmochim. Acta 55, 905–910.

Pizzarello S., Feng X., Epstein S., and Cronin J. R. (1994) Isotopicanalyses of nitrogenous compounds from the Murchison meteorite:Ammonia, amines, amino acids and polar hydrocarbons. Geochim.Cosmochim. Acta 58, 5579–5587.

Pizzarello S., Huang Y., Becker L., Poreda R. J., Nieman R. A., CooperG., and Williams M. (2001) The Organic Content of the TagishLake Meteorite. Science 293, 2236–2239.

Pizzarello S. and Huang Y. (2002) Molecular and isotopic analyses ofTagish Lake alkyl dicarboxylic acids. Meteor. Planet. Sci. 37,687–696.

Pizzarello S., Zolensky M., and Turk K. A. (2003) Nonracemic isova-line in the Murchison meteorite: Chiral distribution and mineralassociation. Geochim. Cosmochim. Acta 67, 1589–1595.

Pizzarello S., Huang Y., and Fuller M. (2004) The carbon isotopicdistribution of Murchison amino acids. Geochim. Cosmochim. Acta68, 4963-4969.

Pizzarello S. and Huang Y. (2005) The deuterium content of individualMurchison and Murray amino acids: a case for the presolar distri-bution of biomolecular precursors. Geochim. Cosmochim. Acta., inpress.

Robert F. and Epstein S. (1982) The concentration and isotopic com-position of hydrogen, carbon and nitrogen in carbonaceous mete-orites. Geochim. Cosmochim. Acta 46, 81–95.

Sauer P. E., Eglinton T. I., Hayes J. M., Schimmelmann A., andSessions A. (2001) Compound-specific D/H ratios of lipid biomar-kers from sediments as a proxy for environmental and climaticconditions. Geochim. Cosmochim. Acta 65, 213–222.

Sessions A. L., Jahnke L. L., Schimmelmann A., and Hayes J. M.(2002) Hydrogen isotope fractionation in lipids of the methane-oxidizing bacterium Methylococcus capsulatus. Geochim. Cosmo-chim. Acta 66, 3955–3969.

Sephton M. A., Pillinger C. T., and Gilmour I. (1998) �13C of free andmacromolecualr aromatic structures in the Murchison meteorite.

Geochim. Cosmochim. Acta 62, 1821–1828.

Sephton M. A., Pillinger C. T., and Gilmour I. (2000) Aromaticmoieties in meteoritic macromolecular materials: Analyses by hy-drous pyrolysis and �13C of individual compounds. Geochim. Cos-mochim. Acta 64, 321–328.

Sephton M. A., Pillinger C. T., and Gilmour I. (2001) Normal alkanesin meteorites: molecular �13C values indicate an origin by terrestrialcontamination. Precamb. Res. 106, 45–56.

Sephton M. A., Verchovsky A. B., Bland P. A., Gilmour I., GradyM. M., and Wright I. P. (2003) Investigating the variations incarbon and nitrogen isotopes in carbonaceous chondrites. Geochim.Cosmochim. Acta 67, 2093–2108.

Shimoyama A., Naraoka H., Komiya M., and Harada K. (1989) Anal-yses of carboxylic acids and hydrocarbons in Antarctic carbona-ceous chondrites, Yamato-74662 and Yamato-793321. Geochim. J.23, 18119.

Stull D. R., Westrum E. F., Jr., and Sinke G. C. (1969) The chemicalthermodynamics of organic compounds, John Wiley & Sons, NewYork.

Turner B. E. (1989) Recent progress in astrochemistry. Space Sci. Rev.51, 235–337.

Wang Y. and Huang Y. (2001) Hydrogen isotope fractionation of lowmolecular weight n-alkanes during progressive vaporization. Org.Geochem. 32, 991–998.

Wang Y. and Huang Y. (2003) Hydrogen isotopic fractionation ofpetroleum hydrocarbons during vaporization: implications for as-sessing artificial and natural remediation of petroleum contamina-tion. Appl. Geochem. 18, 1641–1651.

Yuen G. and Kvenvolden K. A. (1973) Monocarboxylic aicds inMurray and Murchison carbonaceous meteorites. Nature 246, 301–302.

Yuen G., Blair N., Des Marais D. J., and Chang S. (1984) Carbonisotope composition of low molecular weight hydrocarbons andmonocarboxylic acids from Murchison meteorite. Nature 307, 252–254.

Zwank L., Berg M., Schmidt T. C., and Haderlein S. B. (2003)Compound-specific carbon isotope analysis of volatile organiccompounds in the low-microgram per liter range. Anal. Chem. 75,

5575–5583.