THE GEOCHEMISTRY OF FATTY ACIDS IN RECENT MARINE SEDIMENTS

by

RICHARD FRANCIS LEO A.B.

THESIS

Presented to the Faculty of the Graduate School of The University of Texas in Partial Fulfillment of the Requirements .

For the Degree of

MASTER OF ARTS

THE UNIVERSITY OF TEXAS

Austin, Texas

January, 1966

PREFACE

I am deeply grateful to Dr. Patrick L. Parker for his supervision of this work. Special thanks are extended to Dr. Thomas C. Hoering of the Carnegie Institution of Washington, D.C. and to Dr. E. William Behrens and Dr. Edward C. Jonas of the University of Texas for providing samples for study and for offering helpful suggestions.

December 11, 1965

TABLE OF CONTENTS

INTRODUCTION p. 1 MATERIALS & METHODS P- 4 RESULTS p. 16 DISCUSSION p. 32 SUMMARY p. 40 BIBLIOGRAPHY p. 41

LIST OF TABLES

Table 1. Fatty Acids in Recent Sediments: Acid p. 20 concentrations in samples from (a) the Gulf of Mexico, (b) Baffin Bay, Texas, and (c) Bahia Salada, Mexico. Table 2. Fatty Acids in Recent Sediments: p. 23 Relative amounts of acids, Gulf samples. Table 3. Fatty Acids in Ancient Sediments: The p. 24 Green River and Thermopolis oil shales. Table 4. Fatty Acids in Irish bog butter. p. 25

LIST OF FIGURES

Fig. 1. Gas-liquid chromatogram (DEGS substrate) p. 27 of a fatty acid extract from sediment sample 3-2. Fig. 2. Linear log plot of retention values vs. p. 29 carbon numbers constructed from the chromatogram reproduced in Figure 1. Fig. 3 Infrared spectra of three fatty acids. p. 31

INTRODUCTION

Carbon-containing compounds, in trace quantities, are of widespread occurrence in nature. Organic molecules have even been isolated from such unlikely sources as meteorites. The great bulk of organic matter in the geosphere is found in the sediments, however. Most sediments contain a few per cent organic matter although some are as rich as thirty per cent. Over ninety-five per cent of this organic matter is accounted for as the rather abstruse kerogen.*

The remaining solvent extractable five per cent of the organic matter in sediments is complex in composition. It includes various carbohydrates, fatty acids, and a number of amino acids - the three main components of living organisms. Such natural products as porphyrins and carotenoids, purines and pyrimidines, and hydrocarbons, terpenes, and steroids have also been extracted and identified in sediments. In the past, research activity directed toward investigating the occurrence and distribution of organic constituents in nature centered largely on the species which could be isolated from coal tars and crude oils which represent but a small fraction of the organic material in nature - though, admittedly, of considerable economic importance. It is now realized that organic substances other than the petroleum-related species are not so ’’perishable’* or unstable in the natural environment and their presence poses many interesting problems to the geochemist.

The purpose of this work was to study the geochemical fate of fatty acids in the sedimentary environment. This problem was approached by investigating the fatty acid composition of sediments of recent and ancient geological age. Most of the samples analyzed were of recent age as the emphasis was on the geologically early changes that the fatty acid fraction of cellular debris has undergone. Samples of 400 year old bog butter were also examined as they afforded a particularly suitable means to study the relative stabilities of the various kinds of fatty acids.

Since fats consitute one of the three main groups of organic substances in living organisms, their supply to the sediments is relatively large® In addition, fatty acids are among the more stable organic species in sediments. In principle, fatty acids can survive at low temperatures for billions of years and therefore, might be found in very old sedimentary rocks of mild metamorphic history. But other factors, such as biological activity and chemical interactions within the sediment, tend to diminish this possibility. None of the sediment samples examined in this study were devoid of fatty acids. The oldest of these samples, the 70 million year old Thermopolis shale, was relatively rich in fatty acids.

In this study, the qualitative and quantitative nature of the major fatty acids in typical marine sediments have been established. The relationship between the quantity of fatty acids and the total organic matter in sediment has been noted. The occurrence and distribu tion of several acids heretofore unreported in sediments have been observed.

* The term kerogen is operationally defined as that organic material in sediments which cannot be removed by solvent extraction. It is of variable composition, probably consisting of the polymerized products condensed from cellular debris entrapped in the sediments.

MATERIALS AND METHODS

A. SAMPLES.

1. Recent Sediment

Sediment cores were taken from three separate marine localities. The cores were taken directly to the laboratory still sealed in the coring device. Upon extrusion, portions of each core were sectioned as samples for analysis in this study and then stored, until use, at -3°C. to minimize bacterial activity. The depth location of these samples within their parent cores appears in Table 1. These samples of sediment and their source areas are described in the following three paragraphs.

Seven samples (series 5: nos. 1 through 7) were selected from a two meter core of sediment taken by means of a lucite-lined, Phlegler-type coring device in August 1965 from the Inner Shelf of the Gulf of Mexico. The site was located twelve nautical miles SE of Port Aransas, Texas, at a water depth of 25 meters. The seven samples were sectioned from various depths within the core on the basis of their homogeneity in appearance. Except for sample 5-3 which seems to differ only in its brownish color, the series consists of well-mixed, greenish-gray muds.*

The rate at which the sediment accumulated has been judged to be about 3 cm. per 100 yr. (Behrens, personal communication). This implies that the deepest sample of the series, 5-7, is almost 7000 years old.

A series of five samples (Nos. 4-1 through 4-5) were obtained from a five meter core of sediment removed in June 1965 from Baffin Bay, Texas, about 1000 meters NNW of ship channel marker number 36 and at a water depth of approximately three meters. The coring device consisted of a length of 7*3 cm. diameter aluminum irrigation tubing and was manipulated from a barge especially constructed for that purpose. Baffin Bay is a shallow, hypersaline, estuarine bay of some 35 sq. mi. in area. It is located within the property of the King Ranch adjoining Laguna Madre in south-eastern Texas. The bay water is about twice as saline as the open ocean or Gulf (approx. 35 ppt.). Although the upper water layers of the bay appear to be well oxygenated, the sediments below the bay center are reduced and somewhat acidic (Rusnak I 960). These surface sediments seem to be barren of life and the distinctive odor of hydrogen sulfide is easily detected. The rate of sedimentation is given by G. Rusnak (I 960 as 12 cm./100 yrs The muds in samples 4-1 through 4-4 are alike in texture and color (various shades of dark gray) and have a discernable mahogany-type horizontal layering or laminated struc-

ture. Clay-sized grains are dominant. Sample 4-5 consists largely of sand grains (diameter range of 0.0625 to 2.0 mm.) coated with a dark material, the identity of which is uncertain.

A third core of some three meters in length, from which four samples (3-1 through 3-4) were sectioned, was bored November 1964 in about a meter depth of water in Bahia Salada, Tamaulipas, Mexico. The core site was approximately 200 meters SW of the northern peninsula at the mouth of the bay (24 deg., 30 min. N. Lat.; 97 deg., 45 min. W. Long.). The coring operation was accomplished with a pulley-rigged metal tripod fixed directly on the sediment surface in the bay. The tripod served to position the adjustable piston insert within the aluminum coring tube and to facilitate withdrawal of the coring tube. Geographically, Bahia Salada is somewhat similar to Baffin Bay, i.e., a shallow marginal or coastal bay, located in a region of semi-arid climate and adjoining lagoonal waters impounded by a barrier island. The salinity or salt content of the water, however, has become unusually high in recent times - over ten times that of the Gulf of Mexico at the time of coring. The sedimentation of the mud collected was fairly regular as evidenced by laminae appearing throughout the core. The mud was comparatively more heterogeneous that that collected in Baffin Bay or

in the Gulf, however. Bands of coarser and shellier material (probably of eolian transport) and mottling, the effects of burrowing organisms earlier in the sediment f s history, disrupt the regular pattern of the sediment at various places within the core. The mud varied in color from light to dark gray. Shell fragments at depths of 75 cm. and 150 cm. in the core below the sediment-water interface were dated by J. Pearson as being 900 - 75 and 1750 ± 80 years old, respectively, and from these values, a sedimentation rate of 8 cm./100 yrs. was calculated. Information gained from powder diffraction patterns indicate the proportions of the clay minerals, illite, montmorillonite, and kaolinite, in samples from Bahia Salada and Baffin Bay are roughly equivalent and, together, compose approximately 25% of the total material in the clay fraction of these samples. 2. Ancient Sediment and Bog Butter

Two samples of organic-rich shales, the Green River and Thermopolis shales, were analysed for fatty acids to compare the abundance and distribution in these older sediments with the recent. In an effort to gain further information as to the age effect or relative stabilities and possible chemical transformations of fatty acids with time, two samples of bog butter and one sample of freshly-made butter were also analysed.

The Green River shale was formed in the Eocene of

the Tertiary period, some 40 million years ago, and extends through three western states in the Colorado Plateau. The shale is unusually rich in organic matter. The sample used in this study was 15.3% organic carbon. It was collected by Dr. T. C. Hoering near Rifle, Colorado. In preparation for extraction and analysis, the rock sample was pulverized to a powder fine enough to pass through a NBS 100 sieve.

A ball-milled sample of the Thermopolis oil shale (obtained from the Socony Mobil Oil Co.) was also investigated in this study. It came from Bighorn County, Wyoming and is around 70 million years old (Cretaceous period). It contained 1.6% organic carbon.

Butter was once preserved in Ireland by putting it into a wooden tub and burying it in a bog for a period of years. With time, the original fat undergoes partial saponification and the water-soluble fragments and the more volatile, lower molecular weight components are gradually removed. The residual components, referred to as bog butter, assumes the consistency of a tallow or cheese-like material.

Large quantities of this material must have been forgotten or abandoned by their owners and, since the beginning of the 19th century, many tubs, some as heavy as 100 lbs, have been retrieved from old solid bogs at depths ranging from 10 to 12 feet (Bergmann, 1963)•

Two samples of this bog butter, one from Galway County and another from Monaghan County, were obtained from the Irish National Museum (Dublin). Carbon-14 age determinations show them to be 400 years old. Elemental analyses at two independent laboratories - Huffman Laboratories (Wheatridge, Colo.) and F & M Scientific Co (Avondale, Pa.), gave the following composition: 75% carbon, 13% hydrogen, 12% oxygen, and less than 0.3% nitrogen.

One sample of recent butter, freshly-made by Mrs. 0. C. Lee of Rockport, Texas purposely for this work, was also investigated.

B. ISOLATION OF FATTY ACIDS FROM SEDIMENT.

The chemical procedure employed in isolating fatty acids from sediment and in preparing their methyl esters for chromatographic identification is similar to that described earlier (Parker and Leo, 1965). It was designed to minimize decomposition of the unsaturated acids and does not exclude any type of fatty acid. It does not distinguish fatty acids in the free state from those bound as esters, however. This procedure is as follows. The frozen sample was treated with 6N hydrochloric acid to remove all carbonate from the sediment. After filtration, the moist sample was treated with methanol for fifteen minutes while being mechanically

stirred and subjected to ultrasonic vibrations (Branson, LS-75). The methanol was recovered by filtration. The sample was then treated with chloroform and again stirred and sonified for 15 minutes. The chloroform was removed by filtration and combined with the methanol fraction recovered in the preceding step. The sediment was dried and weighed to the nearest tenth of a gram. The combined solvent mixture was taken to near dryness with a rotating evaporator, taken up in chloroform, and washed with IN hydrochloric acid in a separatory funnel to remove any inorganic salts. The chloroform extract was then taken to dryness under a stream of dry nitrogen gas on a steam bath The residue was saponified with 0.5 N methanoic potassium hydroxide solution for 30 minutes on a steam bath. Hydrocarbons and other non-acidic organic impurities were removed by extracting the alkaline solution with n-heptane. Chloroform was then added to the aqueous extract and the two phase mixture was acidified with 6N hydrochloric acid. The organic phase was then taken to dryness and boiled briefly with 10% boron trifluoride in methanol to prepare the methyl esters (Metcalfe, 1961). Finally, the methyl esters were extracted into a known quantity (1 to 4 ml.) of n-hexane, following destruction of any excess boron trifluoride gas with water.

Both the chloroform and methyl alcohol solvents used in this procedure were reagent grade and were redistilled before use. Only distilled water was used. The hexane and heptane solvents employed in the extractions were of spectrochemical quality.

The simultaneous use of blanks during several of the sediment preparations showed that the amount of fatty acid introduced into the final sample extract from foreign sources was negligible (less than 5%).

0. IDENTIFICATION & MEASUREMENT OF FATTY ACIDS.

The methyl esters of the fatty acids were identified and measured by means of gas-liquid chromatography. The instrument used was the F & M Biomedical Gas Chromatograph, model 400, with a hydrogen flame ionization detector. It was operated isothermally and an eight foot stainless steel column, packed with 10% diethylene glycol succinate coating on 60/30 mesh, Chrom P-AW support, was employed (Lipsky et al., 1959)* The column oven temperature was kept at 210°C. and the carrier gas (helium) flow rate was approximately 30 cc./min. Fatty acid ester standards of even and odd carbon-numbered saturates from 011:0 to 020:0,* of even carbon-numbered monounsaturates from 014:1 to 013:1, and of several methyl-branched isomers, i-014, i-Cl6, a-Cl7, and i-Cl3, were purchased from Applied Science

* acid code explained in Table 1

Laboratories, State College, Pa. and used in standardizing the chromatographs, qualitatively through retention time values and quantitatively by peak areas. Peak areas were measured by triangulation.

Under the conditions described, fatty acids from CIO to C2O were easily identified and measured. Acids containing from 14 to 13 carbon atoms in their molecular structure include the most abundant fatty acids in sediment.

In order to verify the identification of the acid species in sediment established initially with the polar chromatographic column, further information was gathered by the following means:

1. Gas-liquid chromatography, non-polar column: The sample extracts of fatty acid esters were rechromatographed, using a non-polar column packing in place of the polar DEGS. Once again, retention values were compared with those for known standards* The column consisted of a six foot glass tube (3 mm. dia.) packed with Gas-Chrom Z (Applied Science), mesh, coated with 10% Apiezon L grease. The column oven temperature was maintained at 195°C. and helium flowed through the column at the rate of 50 cc./min. On occasion the column oven temperature was programmed at 3°C. per min. 2. Graphical representation of retention value data: Logarithms of retention values were plotted against corresponding carbon numbers for selected chromatograms. This

method yields sets of curves which are linear for a given homologous series of acids, i.e., the straight-chain monounsaturates, the straight-chain saturates, the isomethyl-branched saturates, and the anteiso-methyl-branched saturates will form four separate but parallel lines on the graph (Woodford & Van Gent, I 960).

3. Internal standards: Appropriate members of the aforementioned fatty acid standards were added to selected sample extracts and chromatographed. By comparing retention values, peak areas, and peak symmetry of particular peaks of interest on the chromatogram of the extract alone with that of the same extract plus standard(s), the identities of the fatty acids in question can be made more definite.

4. Subtraction method for unsaturated acids: By bromination of the extract with 2 per cent bromine in ether at low temperatures, the unsaturates can be distinguished by their diminished peak area, or entire disappearance, upon re-chromatographing the mixture (Burchfield & Storrs, 1962).

5. Preparative GLC: Samples of fatty acid extracts from sediment were combined and individual acids isolated chromatographically for later confirmatory work with mass spectrometry and infrared spectrophotometry. This was accomplished with the Aerograph 202 Gas Chromatograph,

utilizing a well-conditioned DEGS column. No significant bleed-off was detected. Operating conditions were similar to those described previously for the same type column. The individual acids were collected in pyrex tubes, then rinsed into vials with petroleum ether. Small aliquots of each of the collected peaks were rechromatographed on the more sensitive F & M instrument and the purity of the collected material established.

6. Mass spectrometry: Through the mass spectrometry laboratory of the University of Texas, further confirming data was obtained. Measurements were made with the CEO 21-102 mass spectrometer with heated inlet system. The ionizing current and the ionizing voltage at the time of operation were 25 microamperes and 70 volts, respectively. The reservoir temperature was 25O°C. Mass spectra of standards of the same fatty acids believed to be contained in the chromatographically isolated peak fractions selected for further identification by this method were also obtained on the same instrument for comparison or M finger-print” purposes.

7* Infrared spectrophotometry: The absorption spectra of certain fatty acids isolated by preparative GLC were taken. Both KBr pellets and solutions were used. The instrument employed was the Perkin-Elmer Model 237-B Grating Infrared Spectrophotometer.

E. PER CENT ORGANIC CARBON

The carbonate-free sediment sample was dried, pulverized, weighed, and combusted with copper oxide and one third of an atmosphere of pure oxygen on a vacuum line. Water vapor was removed by means of a dry ice-isopropanol mixture as the gases resulting from combustion were recirculated through the vacuum line furnace several times by means of a Toepier pump. Upon passage to the other side of the pump, the carbon dioxide gas was frozen out with liquid air and the excess oxygen pumped off. It was then allowed to expand into a small evacuated line against a calibrated manometer. Pressure readings were made with a cathetometer.

* The term mud is generally used to describe a sediment in which the dominant or major fraction of constituent particles have diameters which fall within the range of 0.0002 to 0.0625 mm., i.e., mud is a mixture of claysized grains (0.0002-0.0040 mm.) and silt-sized grains (0.0040-0.0625 mm.).

RESULTS

The possibility of artificial contamination of the sediment samples from such sources as sewage, industrial wastes, refuse or oil and grease from ships is rather unlikely, particularly in the case of the bays. They are located in rather remote, undeveloped geographical regions. Bahia Salada is especially isolated. It is situated some 50 miles from the nearest Mexican village (which has no industry) and there are no paved roads leading to the bay. Contamination in the collection and of the samples is unlikely due to the care with which they were handled. One cannot be as sure of uncontaminated samples in the case of the ancient sediments. In this case the geochemist can only use care in selecting his material and move forward.

The experimental results are presented in Tables 1 through 4* Figure 1 is a typical chromatograph and represents the type of day to day measurement made in the study. The fatty acids identified in recent sediments were (1) straight-chain saturates, both even and odd carbon-numbered, from C12:0 to C18:0, (2) normal monounsaturates from C;1 to Cl£:l, (3) iso-methyl branched saturates from i-Cl2 to i-Cl6, and (4) the anteiso-

methyl branched saturate; a-Cl5. Species less than 12 carbon atoms (e.g. 011:0) or more than 13 (e.g. 020:0) were detected but were not extensively investigated. It is felt that the acids of intermediate weight account for the greater fraction of the fatty acids which can be extracted from these sediments.

Before discussing the analytical results and relating them to other work it would be appropriate to state how each acid was identified and this identity confirmed by one or more independent measurements. Identification was based primarily on GLC evidence using the BEGS column (Fig. 1). The agreement of retention values given by fatty acids in the extracts with those obtained from standards and those predicted from the literature (James, I 960) was excellent.

Following this first identification on the polar DEGS column every sample was run on the nonpolar Apiezon L column. In this way a second set of retention values were obtained. Considering that the chemical procedure itself is fairly specific, the two GLC retention values are adequate proof of the identity of the major peaks such as 016:0.

The peaks corresponding to the unsaturated acids could be removed by bromination of the sample. This was taken as confirming evidence for the presence of the

unsaturated acids. It was noted, however, that in a few samples the 16:1 peak could not be completely removed by bromination. Thus, there may be a trace of an unidentified acid present in this region.

The iso and anteiso acids had not been reported for sediments prior to this work. Therefore a special effort was made to confirm their presence. The i-Cl4, i-Cl5, a-Cl5 and i-Cl6 peaks were collected using the Aerograph 200 GLC. Mass spectra of the collected i-Cl4 and i-Cl6 peaks were taken and compared with the mass spectra obtained from standards of i-Cl4 and i-Cl6. The mass spectra of the standards and samples were identical. The parent peak and the carbomethoxy fragments were the most salient features for diagnostic purposes. Unfortunately with the instrument used the iso and anteiso acids do not have a cracking pattern different from the normal acid of the same molecular weight. However the normal acid had been removed by GLC.

Infrared spectra of the combined collected iso and anteiso Cl 5 peaks were taken in carbon tetrachloride. The results are shown in Fig. 2. The strong absorption band at 1750 cm’ 1 due to the carboxyl group is present. The band at cm"* 1 is characteristic of iso and anteiso methyl branching. Iso acids show a doublet in this region which serves to distinguish them from normal and anteiso forms.

As a final proof of identification, a linear log plot of the retention value data of a sediment sample was constructed (Fig. 2). Graphical presentation of the data in this manner results in a series of straight and parallel lines. Each line is representative of a separate homologous series of fatty acids. There is a slight deviation from linearity in the lower molecular weight region of the plot but this was expected (Ackman et al., 1963) and is not critical.

In order to evaluate the error due to the chemical procedure a uniform sample of recent sediment was analyzed in triplicate. For C14:0 the results were 0.42, 0.35 and 0.31 ug/g 0 This and other evidence indicates that the day to day reproducibility is between 30 and 50 per cent. The concentrations of fatty acids in sediments reported are minimum values. There is no way to know what fraction of the acid present in sediment is extracted. Carbon-14 labelled C16:0 which has been well mixed with fresh sediment cannot be quantitatively recovered, 50% recovery being high o

of acid code: The number following capital letter C designates the number of carbon atoms in the molecule. The number following the colon refers to the number of double bonds present, if any. The acid is the normal, straight-chain form unless preceded by a lower case i- (or a-) which designates the iso- (or anteiso-) methyl-branched, saturated acid.

*The per cent organic carbon in the GRS and TS samples were 15-3 and 1.6, respectively.

*Both the Monaghan County (M) and the Galway County (G) bog butter samples contained trace amounts of methylbranched saturates from Cl 2 to Cl 7, odd carbon-numbered saturates from Cll to Cl 7, and the unsaturates, 014:1, C16:l, and CIS:2. The recent butter sample contained methyl branched fatty acids of 13, 14, 15, 16, and 17 carbon atoms, and all the straight-chain saturates from 9 carbon atoms to 20.

Figure 1. Gas-liquid chromatogram of a fatty acid extract (sample 3-2) on a DEGS substrate.

Operating conditions: temperature 210°C. sample injection 0.010 ml. helium flow 30 cc/min.

Key to peak identities: A i-012 F i-CU K 015:0 Q C17:0 B C12:0 G 014:0 L 015:1 R 017:1 C i-013 H 014:1 M i~CI6 S 013:0 D 013:0 I i-015 N 016:0 T 013:1 E 013:1 J a-015 0 016:1

Code explained in footnote, Table 1.

Figure 2. Linear log plot of retention value data obtained from the chromatogram reproduced in figure 1. The plotted points for n-saturates are enclosed in triangles, those from monounsaturates in circles, iso-methyl branched saturates in squares, and the lone anteisomethyl branched saturate (J) in a hexagon.

Figure 3. Infrared Spectra of Fatty Acids (esters of), (top) methyl isotetradecanoate (i-Cl4), standard, (middle) methyl isopentadecanoate (i-Cl5), peak I. (bottom) methyl pentadecanoate (C15:0), standard.

Note: All of the above species were in solution in carbon tetrachloride. The substance whose spectra appears in the middle figure was not pure but contained peak J (a-Cl5) as well. Iso acids can be distinguished from their corresponding anteiso and normal forms in that they show a doublet in the range from 1360 to cm" 1 .



FIGURE I



FIGURE 2



FIGURE 3

5-1 5-2 5-3 5-4 5-5 5-6 5-7 acid* micrograms of acid ester per gram of sediment C12;0 0.7 0.9 1.1 0.4 1.0 0.3 0.6 i-C14 0.5 0.4 0.7 0.2 0.6 0.2 0.4 014:0 3.1 2.7 4 • 4 1.8 4.3 1.4 2.6 014:1 1.7 1.1 2.1 0.7 2.0 0.7 1.3 i-C15 0.3 0.2 0.4 0.2 0.5 0.1 0.3 a-C15 0.3 0.2 0.4 0.2 0.5 0.1 0.3 015:0 2.1 1.7 2.9 1.3 2.9 1.0 1.7 1-C16 0.4 0.3 0.5 0.2 1.0 0.1 0.4 016:0 16.3 15.3 20.9 9.1 23.0 6.9 12.9 016:1 6.3 5.5 11.0 3.9 10.3 3.0 5.7 017:0 0.9 0.9 1.5 0.5 1.7 0.4 0.8 018:0 3.7 3.6 4 • 4 2.0 5.0 1.8 4.2 018:1 6.7 per 7.2 13.4 3.9 cent organic carbon 10.1 2.9 4.3 0.5 0.4 0.3 0.3 0.2 0.2 0.3 centimeters below the sediment-water interface 10- 19- 60- 71- 155- 172- 198- 19 30 71 115 172 198 208

Table 1. Fatty Acids in Recent Sediments. a) Samples from the Gulf of Mexico: Series 5.

4-1 4-2 4-3 4-4 4-5 acid micrograms of acid ester per gram of sediment 012:0 1.1 0.8 1.1 1.4 1.5 i-CU 2.4 1.9 3.1 0.3 0.2 014:0 3.8 2.9 2.9 3.2 1.6 014:1 0.8 0.3 0.6 0.7 0.3 i-C15 0.4 0.5 0.5 0.9 0.5 a-C15 0.3 0.4 0.4 0.9 0.5 015:0 2.0 1.4 4.9 1.4 0.5 i-C16 3.8 2.3 1.1 0.5 0.2 016:0 7.8 11.9 10.9 13.9 8.0 016:1 6.4 3.7 4.1 4.2 1.5 017:0 1.5 1.2 2.3 0.8 0.6 C18:O 5.1 4.2 6.3 8.1 8.7 018:1 6.3 per 3-9 5.6 3.5 cent organic carbon 1.6 3.2 2.6 5.2 2.1 0.1 centimeters below water interface the sediment- 157- 262- 348- 405 434- 162 268 353 410 435

Table 1. Fatty Acids in Recent Sediments. b) Samples from Baffin Bay, Texas: Series 4.

3-1 3-2 3-3 3-4 acid micrograms of acid ester per gram of sediment C12:0 0.4 0.5 0.4 0.9 i-C14 0.3 0.7 0.4 0.9 014:0 1.2 2.1 0.9 2.8 014:1 0.3 0.6 0.2 0.7 i-C15 0.3 0.6 0.2 0.3 a-C15 0.3 0.6 0.2 0.3 015:0 0.6 1.1 0.5 1.6 i-C16 0.4 0.7 0.7 1.2 016:0 5.1 9.1 4.3 9.9 016:1 2.4 3.6 1.7 5.0 017:0 0.5 0.6 0.5 0.7 018:0 2.1 4.2 1.9 2.3 018:1 5.0 3.3 1.8 4.5 percent organic carbon 0.6 0.5 0.7 0.8 centimeters below water interface the sediment- 6- 19 79- 105 177- 194 254- 278

Table 1. Fatty Acids in Recent Sediments. c) Samples from Bahia Salada, Mexico: Series 3

acid 5-1 Relat 5-2 ;ive val 5-3 .ues bas 5-4 >ed on 5-5 C16:0 as 5-6 ten ur 5-7 tits. 012:0 0.4 0.6 0.5 0.4 0.4 0.4 0.5 i—C14 0.3 0.3 0.3 0.2 0.3 0.3 0.3 014:0 1.9 1.8 2.1 2.0 1.9 2.0 2.0 C14:l 1.0 0.7 1.0 0.8 0.9 1.0 1.0 Clf>:0 1.3 1.1 1.4 1.4 1.3 1.4 1.3 i-C16 0.3 0.2 0.2 0.2 0.4 0.1 0.3 C16:O 10.0 10.0 10.0 10.0 10.0 10.0 10.0 016:1 3.9 3.6 5.3 4.3 4*4 4.3 4.4 017:0 0.6 0.6 0.7 0.6 0.7 0.6 0.6 C18:O 2.3 2.4 2.1 2.2 2.2 2.6 3.3 018:1 4.1 4.7 6.4 4.3 4.4 4.2 3.3

Table 2. Fatty Acids in Recent Sediments. Samples from the Gulf of Mexico: Series 5.

acid GRS TS (1) (2)_ (1) (2) 011:0 0.3 2 tr 012:0 1.3 9 0.8 53 013:0 2.3 15 0.2 13 i-CU 0.6 4 tr 014:0 3.4 22 1.1 73 014:1 1.1 7 tr 015:0 2.0 13 0.8 53 1-C16 0.5 3 tr C16:O 12.2 so 3.8 253 016:1 1.9 12 0.8 53 017:0 1.6 10 1.5 100 018:0 8.4 55 2.9 193 018:1 3.7 24 0.8 53 019:0 1.4 9 1.9 127 020:0 2.6 17 2.3 153 021.0 2.0 13 4.4 293 022:0 3.6 24 2.0 133 023:0 2.1 11. 1.0 67 024:0 2.9 19 1.5 100

Table 3. Fatty Acids in the Green River (GRS) and Thermopolis (TS) Oil Shales.* (1) ug. F.A. per g. of sample (2) ug. F.A. per g. of organic 0.

Recent butter Bog butter Bog butter Recent butter Bog butter Bog butter (M) (G) (M) (G) acid ug F.A. per mg sample relative amts.(016:0 10) 012:0 3.5 5.9 5.5 0.5 0.6 0.5 014:0 19.5 39.1 42.8 2.8 3.6 3.7 015:0 3.6 5.9 5.5 0.5 0.6 0.5 C16:0 69.6 104.3 117.4 10.0 10.0 10.0 C16:l 2.5 0.4 017:0 2,0 3.2 2.3 0.3 0.3 0.2 C18:0 36.2 71.3 52.5 5.2 6.a 4.5 C18:l 58.7 43.1 17.7 8.4 4.1 1.5 C18;2 3.0 0.4 stearic acid/ol eic acid ratio. 0.6 1.7 2.9

Table 4. Fatty Acids in Butter Samples.*

DISCUSSION

Fatty acids may account for ten per cent of the total organic matter of marine organisms. The dominant fatty acids in most marine organisms are those with an even number of carbon atoms arranged in a straight chain. Particularly conspicious are the polyunsaturates — the most abundant of the various kinds of fatty acids. This, then, is the source material for the geochemistry of fatty acids.

The most striking geochemical aspect of the fatty acids in recent sediment is their low concentration. The data in Table 1 show that fatty acids account for only about one per cent of the organic matter in sediments as contrasted to about ten per cent in organisms. The reason for this geologically early disappearance of fatty acids probably relates to the early appearance of kerogen in sediment.

The polyunsaturated acids so abundant in organisms, are absent in the recent sediments (Table 1). The monounsaturates are significantly reduced but still present. Probably, the double bonds serve as reactive sites for reactions which tie up the acids. This again suggests that kerogen may arise from the early reaction of biochemicals.

Abelson, Hoering, and Parker (1962, 1963) studied the thermal stabilities of fatty acids in algae by heating fresh algae at low temperatures (200°C) for varying lengths of time. In the experiments with algae, it was found that the major amount of the organic matter, upon prolonged incubation, is converted to a black, insoluble residue, so that only a small fraction of the fatty acid can be recovered - similar to sedimentary organic matter. They also measured fatty acid distributions in sediments varying in age from the recent to 500 million years and attempted to use the thermal stability information in interpreting the sediment data in terms of relative survivabilities of the different kinds of fatty acids. It was learned that the polyunsaturated species disappear quickly. The monounsaturates and the saturates, however, are unusually resistant, the saturates, in particular, possessing the intrinsic stability to survive for billions of years.

P. M. Williams (1965) investigated fatty acids in sea water, recent sediment from the Pacific Ocean and straits adjoining Vancouver Island, Canada, and in several planktonic species. The sediments were reported to contain from 13 to 67 ug of fatty acid per gram of dry sediment (vs. from 13 to 62 in my Gulf samples); palmitic acid having a mean concentration in the five samples of 10.37 (vs. 13.3 in my Gulf samples). Fatty acid species reported by him were (1) the

normal saturated acids with 14, 16, and 13 carbon atoms, (2) the normal monounsaturates with 14, 16, 13, and 20 carbon atoms, (3) the normal diunsaturates with 14, 16, and 13 carbon atoms in their straight-chains, and (4) the normal triunsaturates of 16 and 13. Identification was based solely on retention value data using a polar polyester column. No chemistry was done to eliminate non-acid lipids prior to esterification. Considering Abelson’s study of the thermal stability of fatty acids and the results of this present study, it seems unlikely that these polyunsaturates do exist in sediments. Williams did not report odd or branched chains which do indeed appear in the sediments used in this study. Several of his species identified as polyunsaturates may very well be odd fatty acids. It is especially likely that his 14:2 is really 15:0. However, for the dominant acids, Williams’ quantitative data agrees well with that reported here.

Cooper (1962) and Cooper and Bray (1963) extracted and identified fatty acids in recent sediments (from the San Nicholas and Santa Barbara basins off the southern California coast), ancient sediments (of Cretaceous and Mississippian age), and waters from petroleum reservoirs. The fatty acid concentrations measured were of the same order of magnitude as those

determined in this study. They reported the presence of normal saturates ranging from C14:0 to C30:0 in all three types of samples. Odd as well as even carbon-numbered acids were detected. Although even acids predominated over odd acids in all cases, they found that the relative abundance of odd acids increases in passing from the recent samples to the ancient. To account for this fatty acid distribution in the recent vs. ancient samples, they proposed a free radical mechanism (similar to the Kolbe synthesis) by which even acids are decarboxylated during diagenesis into odd fatty acids. This is a very interesting proposal and the evidence of the present study for such a reaction will be discussed later.

Lawlor and Robinson (1965) investigating hydrocarbon and fatty acid distribution in the Green River Shale suggested a modification to the mechanism proposed by Cooper. They proposed fatty acid transformation by carbon gain in the region of higher molecular weight (fatty acids containing twenty-three carbon atoms or more) rather than just carbon loss through decarboxylation which they conclude is operative for only the lighter fatty acids.

In addition to measuring the n-saturates in the organic solvent extractable fatty acids fraction of three sediments - the San Nicholas basin (recent), the Green River shale (40 m.y.), and the McGinn shale (1600 m.y.), Hoering and Abelson (1965) investigated the n-saturates resulting from the mild chromic acid oxidation of the kerogen material of these sediments. The acid distributions in both fractions were comparable but the amounts of acids in the oxidized kerogen fraction were considerably greater. For example using recent (San Nicholas basin) sediment they recovered only 6 ug of C16:0 per gram of organic carbon by solvent extraction but 126 ug when the sample was oxidized prior to extraction. Thus, it would appear that fatty acids are among the building blocks of kerogen and can be recovered without losing their biochemical identity. Although the samples used in the present study were not the same as those used by Hoering and Abelson it is important to note that the extraction procedure used in the present investigation yielded as much acid as their oxidation. For example, the concentration of 016:0 in sample 5-1 was 3260 ug/g.org.C. This order of magnitude is typical for the other members of the series.

The homologous series based on the saturated iso and anteiso fatty acids (Tables 1-3) have not been reported by other workers. Cooper (1962) would not have seen them because his chemical procedure (formation of urea adducts) excluded them. Since they were present in recent sediments in appreciable quantities, it was decided to look for them in two ancient sediments. They were detectable in both shales but the Green River shale was much richer in them (Table 3). It is interesting to note that the most significant variation in the fatty acid pattern between the Gulf and Baffin Bay is the normal saturated/iso-saturated ratio (e.g. C14:0/i~C14:0). While branched chain acids are rare, they do occur in many organisms. Kaneda (1963) has shown that the branched chain acids are the major acids for the bacteria Bacillus subtilus. It may be that those in sediments represent dead bacteria, but one can not be certain on the basis of this data. These acids certainly should be useful in attempts to characterize sediments and petroleum by their fatty acid pattern.

Although the Tables do not include all fatty acid components present in trace quantities, it was found in almost every sample that all the n-saturates from C12:0 to Cl$:0 have associated with them methyl-branched and monounsaturated isomers. This association is most pronounced in the region of C14:0 to C16:0 (see Fig. 1).

The relative amounts of odd carbon acids found in the two ancient sediments are greater than those found in the recent sediment. However, the data do not clearly support Cooper’s (1962) conclusions concerning the

generation of odds from evens in geological time.

Sample C16:O+C18:O/2xC17:O

Gulf 10

Baffin Bay 9

Bahia Salada 8

Green River 6

Thermopolos 2

Cooper’s recent 16.7

Cooper’s ancient 1.7

Thus, the difference between Cooper’s ancient and recent is an order magnitude while the difference between the ratios of the Bahia Salada and the ancient Green River’s are no greater than those between the Bahia Salada and the recent Gulf. Many more sediments, recent and ancient, must be compared before Cooper’s ideas can be tested adequately.

The data in Tables 1 & 3 are expressed as ug fatty acid per gram of sediment. One can also express the data as ug fatty acid per gram of organic carbon. It is important to be able to do both as the following discussion will illustrate. Four of the Baffin Bay samples were homogeneous (mud) in grain-size composition. By contrast one, 4-5 > was composed largely of sand. This sample, 4-5, contained less fatty acid per gram of sediment than the others (e.g. 41 vs 25 for 4-1). But

if one considers the data expressed as ug acid per gram of organic carbon sample 4-5 stands out (e.g. vs 13x10 for 4-1)• This may be explained as a consequence of secondary enrichment in fatty acids in the sandy layer from adjacent mud layers. It would be most interest ing to know whether hydrocarbons have also started to migrate into these narrow sand lenses.

The Bog Butter (Table 4) was studied because it is such a geologically unique sample. Assuming that butter 400 years ago was the same as it is today, the source material is exactly known. The data in Table 4 serve to illustrate the relative "survivabilities" of the acids. As with the sediment, the polyunsaturated has disappeared and the C1B:1 has decreased.

SUMMARY

In summary, the general characteristics or features of fatty acids in these recent sediments are outlined below

The vertical distribution of fatty acids in the recent sediments is relatively uniform. Apparently any transformations which take place in the sediments do so readily. The fatty acid distribution pattern in recent sediments varies more for different environmental types than for depth within one environment. As in nearly all biological systems, the species present in greatest abundance are the normal saturates with an even number of carbon atoms with palmitic acid being dominant. Normal unsaturates are greatly reduced in quantity in the sediments, with only the monounsaturates surviving in appreciable amounts. In recent marine sediments, it appears that each normal saturate of intermediate molecular weight has associated with it, its methyl-branched and monounsaturated isomers. Odd fatty acids are more abundant than some earlier workers have reported. Detailed quantitative measurements of fatty acids in many other recent environments and in ancient sediments promises to be a useful geochemical tool.

BIBLIOGRAPHY

Abelson, P. H. (1959) "Geochemistry of Organic Substances” in Researches in Geochemistry, edited by P. H. Abelson. Wilby (New York), 79-103.

Abelson, P. H. and Parker, P. L. (1962) "Fatty Acids in Sedimentary Rocks”. Carnegie Inst. Washington, Yearbook 61,

Abelson, P. H., Hoering, T. C. and Parker, P. L. (1963) "Fatty Acids in Sedimentary Rocks” in Advances in Organic Geochemistry, edited by U. Colombo and G. D. Hobson. Ma cMillan~(New York) , 169-174.

Ackman, R. G., Burgher, R. D. and Jangaard, P. M. (1963) "Systematic Identification of Fatty Acids in the Gas- Liquid Chromatography of Fatty Acid Esters: A Preliminary Survey of Seal Oil". Canadian J. Biochem. Physiol., 41 1627-1641.

Ackman, R. G. and Sipos, J. C. (1965) "Isolation of the Saturated Fatty Acids of Some Marine Lipids with Particular Reference to Normal Odd-numbered Fatty Acids and Branched Chain Fatty Acids". Comp. Biochem. Physiol., 15, 445-456.

Bergmann, W. (1963) "Geochemistry of Lipids” in Organic Geochemistry, edited by I. A. Breger. Macmillan (New York), 503-542.

Biemann, K. (1965) Mass Spectrometry. McGraw-Hill (New York).

Bordovskiy, 0. K. (1965) "Accumulation and Transformation of Organic Substances in Marine Sediments". Marine Geol., 3, 3-114.

Burchfield, H. P. and Storrs, E. E. (1962) Biochemical Applications of Gas Chromatography. Academic Press (New York).

Chapman, D. (1965) "Infrared Spectroscopy of Lipids". J. Amer. Oil Chemists" Soc., 42, 353-371*

Cooper, J. E. (1962) "Fatty Acids in Recent and Ancient Sediments and Petroleum Reservoir Waters". Nature 193 744-746.

Cooper, J. E. and Bray, E. E. (1963) "A Postulated Role of Fatty Acids in Petroleum Formation". Geochim. et Cosmochim. Acta, 27, 1113-1127.

Crippen, R. C. and Smith, C. E. (1965) "Procedures for the Systematic Identification of Peaks in Gas-Liquid Chromatographic Analysis". J. Gas Chromat., 37-42.

Degens, E. T. (1965) Geochemistry of Sediments. Prentice- Hall (Englewood Cl if J’s, N.J.).

Emery, K. 0. (I 960 The Sea Off Southern California. Wiley (New York).

Hawke, J. C., Hansen, R. P. and Shorland, F. B. (1959) "Gas-Liquid Chromatography: Retention Volumes of the Methyl Esters of Fatty Acids with Special Reference to Normal Odd-numbered, Iso and Anteiso Acids". J. Chromat., 2, 547-551®

Hilditch, T. P. and Williams, P. N. (1964) The Chemical Constitution of Natural Fats. Wiley (New TorkTT””

Hoering, T. C. and Abelson, P. H. (1965) "Fatty Acids from the Oxidation of Kerogen". To be published in Carnegie Inst. Washington, Yearbook 64 (December 1965).

Hoering, T. C. (1965) "The Extractable Organic Matter in Precambrian Rocks and the Problem of Contamination". To be published in Carnegie Inst. Washington, Yearbook 64 (December 1965)•

Horning, E. C., Ahrens, E. H., Lipsky, S. R. et al. (1964) "Quantitative Analysis of Fatty Acids by Gas-Liquid Chromatography”. J. Lipid Res., £, 20-27®

James, A. T. (i 960 "Qualitative and Quantitative Determination of the Fatty Acids by Gas-Liquid Chromatography" in Methods of Biochem. Anal, (vol. VIII), edited by D• Glick. Interscience (New York) 9 1-60.

Jeffrey. L. M., Pasby, B. F., Stevenson, B. and Hood, D. W. (1964) "Lipids of Ocean Water" in Advances in Organic Geochemistry, edited by U. Colombo and G. D. Hobson. Macmillan (New York), 175-197-

Jurg, J. W. and Eisma, E. (1964) "Petroleum Hydrocarbons: Generation from Fatty Acids". Science, 1449 1451-1452,

Kaneda, T. (1963) "Biosynthesis of Branched Chain Fatty Acids". J. Biol. Chern., 235, 1222-1235*

Lawlor, D. L. and Robinson, W. E. (1965) "Fatty Acids in Green River Formation Oil Shale", in Preprints, Division of Petroleum Chemistry, ACS, vol. 10 (no. 1).

Lipsky, S. R., Landowne, R. A. and Lovelock, J. E. (1959) "Separation of Lipides by Gas-Liquid Chromatography". Anal. Chern., $52-$56.

Markley, K. S., editor (1964) Fatty Acids, v. I-111. Interscience (New York).

Mclver, R. D. (1962) "Ultrasonics-A Rapid Method for Removing Soluble Organic Matter from Sediments". Geochim. et Cosmochim. Acta, 26, 343-346.

Metcalfe, L. D. and Schmitz, A. A. (1961) "The Rapid Preparation of Fatty Acid Esters for Gas Chromatographic Analysis". Anal. Chern., 33 j 363-364*

Oppenheimer, C. H. (I 960 "Bacterial Activity in Sediments of Shallow Marine Bays". Geochim. et Cosmochim. Acta, 19, 244-260.

Parker, P. L. and Leo, R. F. (1965) "Fatty Acids in Blue-Green Algal Mat Communities". Science, 14 $, 373-374*

Parsons, R. T. (1963) "Suspended Organic Matter in Sea Water" in Progress in Oceanography, Vol. 1., edited by M. Sears, Macmillan (New York), 203-242.

Rosenberg, E. (1964) "Purines and Pyrimidines in Sediments from the Experimental Mohole". Science, 146 9 16$O-16$l.

Rosenfeld, W. D. (1945) "Fatty Acid Transformations by Anaerobic Bacteria". Arch. Biochem., 16, 263-273*

Rusnak, Go A* (I 960 "Sediments of Laguna Madre, Texas" in Recent Sediments, Northwest Gulf of Mexico, edited by F* P. Shepard. Am. Assoc. Petr.' Geol. [Tulsa) , 153-196.

Shorland, F. B. (1962) "The Comparative Aspects of Fatty Acid Occurrence and Distribution" in Comparative Biochemistry, A Comprehensive Treatise, edited by M. Florkin and H. S. Mason. Academic Press (New York), 1-102.

Slowey, J. F., Jeffrey, L. M. and Hood, D. W. (1962) ”The Fatty Acid Content of Ocean Water”. Geochim. et Cosmochim. Acta, 26, 607-616.

Williams, P, Mo (1965) "Fatty Acids Derived from Lipids of Marine Origin”. J. Fish. Res. Bd. Canada, 22, 1107-1122.

Woodford, F. P. and van Gent, C. M. (I 960 ”Gas-Liquid Chromatography of Fatty Acid Methyl Esters: The "Carbon number” as a Parameter for Comparison of Columns” J. Lipid Res., JL,

The vita has been removed from the digitized version of this document.