Microbial diagenesis of evaporitic gypsum

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/236216057

Microbial diagenesis of evaporitic gypsum ARTICLE in GEOCHIMICA ET COSMOCHIMICA ACTA · JUNE 2009 Impact Factor: 4.33

READS

33

7 AUTHORS, INCLUDING: Christophe Lécuyer

M.-M. Blanc-Valleron

Claude Bernard University Lyon 1

Muséum National d'Histoire Naturelle

224 PUBLICATIONS 4,785 CITATIONS

77 PUBLICATIONS 1,250 CITATIONS

SEE PROFILE

SEE PROFILE

Mahmoud A. Aref

Vincent Grossi

King Abdulaziz University

Claude Bernard University Lyon 1

32 PUBLICATIONS 197 CITATIONS

76 PUBLICATIONS 1,134 CITATIONS

SEE PROFILE

SEE PROFILE

Available from: Vincent Grossi Retrieved on: 09 February 2016

Chemical Geology 347 (2013) 199–207

Contents lists available at SciVerse ScienceDirect

Chemical Geology journal homepage: www.elsevier.com/locate/chemgeo

Biomarker and isotope evidence for microbially-mediated carbonate formation from gypsum and petroleum hydrocarbons G. Aloisi a,⁎, M. Baudrand b, C. Lécuyer b, J.-M. Rouchy c, R.D. Pancost d, M.A.M. Aref e, f, V. Grossi b a

LOCEAN UMR 7159, CNRS-UPMC-IRD-MNHN, Univ. Paris 6, 75252 Paris, France Laboratoire de Géologie de Lyon, Université Claude Bernard Lyon 1, CNRS, 69622 Villeurbanne, France Muséum National d’Histoire Naturelle, Département Histoire de la Terre, 75005 Paris, France d Organic Geochemistry Unit, Bristol Biogeochemistry Research Center, University of Bristol, United Kingdom e Geology Department, Faculty of Science, Cairo University, Giza, Egypt f Department of Petroleum Geology and Sedimentology, Faculty of Earth Sciences, King Abdulaziz University, Jeddah, Saudi Arabia b c

a r t i c l e

i n f o

Article history: Received 25 September 2012 Received in revised form 28 February 2013 Accepted 15 March 2013 Available online 27 March 2013 Editor: D.R. Hilton Keywords: Petroleum Gypsum Sulfate-reducing bacteria Biomarkers Authigenic carbonates Gulf of Suez

a b s t r a c t Along the western coast of the Gulf of Suez large amounts of evaporitic gypsum of Miocene age have been microbially transformed into carbonates and elemental sulfur in the presence of petroleum. Similar diagenetic transformations have been described from numerous sites worldwide but the role of petroleum, specifically as a carbon source for the sulfate-reducing microbial community, remains elusive. We carried out a geochemical investigation of microbial carbonates from the Gulf of Suez that suggests the presence of a community of sulfate-reducing bacteria thriving on carbon substrates contained in petroleum. Specifically, a set of non-isoprenoidal macrocyclic glycerol diethers (McGDs), that we tentatively ascribe to sulfate-reducing bacteria, have a stable carbon isotope composition close to that of petroleum n-alkanes associated with the carbonates. The presence of archaeol that is 13C-enriched relative to bacterial lipids suggests that Archaea are present but either indirectly involved or not involved in the transformation of petroleum-derived carbon. The lipid biomarker pattern we observe is distinct from those observed in settings where sulfate reduction is coupled to the anaerobic oxidation of methane. Our results suggest that petroleum migration has triggered the microbial transformation of gypsum into carbonates in the Gulf of Suez. By extension, the involvement of petroleum in the microbial transformation of gypsum into carbonates in other settings, which was suggested by more indirect, geological and inorganic geochemical evidence, seems very likely. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Sulfate reduction is one of the most widespread microbial processes in marine sediments (Jørgensen, 1982). In organic matter (OM) rich sediments, it typically involves the reduction of seawater sulfate by microbial communities that use OM or methane as reducing agents (Canfield et al., 1993; Hinrichs and Boetius, 2002). OM and methane-based sulfate reduction have been investigated extensively from the biogeochemical standpoint, with thorough microbiological and organic geochemical characterization of the microbial communities involved (e.g. Kasten and Jørgensen, 2000; Knittel and Boetius, 2009; Teske, 2010 and references therein). The microbial reduction of gypsum, however, has received less attention although gypsum formed in evaporitic environments provides an alternative source

⁎ Corresponding author. Tel.: +33 1 44 27 70 72; fax: +33 1 44 27 38 05. E-mail addresses: [email protected] (G. Aloisi), [email protected] (C. Lécuyer), [email protected] (J.-M. Rouchy), [email protected] (R.D. Pancost), [email protected] (M.A.M. Aref), [email protected] (V. Grossi). 0009-2541/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.chemgeo.2013.03.007

of sulfate when it dissolves. In OM or methane-rich environments, gypsum can be reduced by microbes leading to the formation of diagenetic carbonates and elemental sulfur deposits (upon re-oxidation of the H2S formed by sulfate reduction). Examples of this microbiallydriven transformation are the carbonate-sulfur mineralizations of the Zechstein Formation (Upper Permian, Germany) (Peckmann et al., 1999), the Calcare di Base (Upper Miocene, Sicily) (Rouchy, 1981; McKenzie, 1985; Ziegenbalg et al., 2010, 2012) and the diagenetic carbonates of the Lorca Basin (Upper Miocene, Spain) (Rouchy et al., 1998). In an additional biogeochemical setting, evaporitic gypsum is thought to be reduced by organisms that use petroleum hydrocarbons, rather than recent organic matter or methane, as a source of carbon. This hypothesis is supported by the existence of sulfur-bearing authigenic carbonates in places where petroleum comes in contact with gypsum deposits. Examples of this association are common worldwide, in the cap rock of salt diapirs in the Gulf of Mexico (Feely and Kulp, 1957; Davis and Kirkland, 1970), in the Permian basin in Texas, as well as in the stratabound mineral deposits in the Fergana and Amudaria depression of Central Asia, the Mesopotamian basin in Iraq and the Cis-Carpathian trough in Poland and Russia (Davis and Kirkland, 1970; Barker et al., 1979; Ruckmick et al., 1979).

200

G. Aloisi et al. / Chemical Geology 347 (2013) 199–207

Most of the evidence for the microbial nature of the gypsumcarbonate transformation is indirect and comes from i) the 13C-depletion of authigenic carbonates that incorporate dissolved inorganic carbon 13 (DIC) from the microbial oxidation of C-depleted organic carbon and, 34 ii) the S-depletion of elemental sulfur which originates from the fractionation of sulfur isotopes during microbial sulfate reduction (Pierre and Rouchy, 1988; Aref, 1998; Ziegenbalg et al., 2010). Recently, the microbial communities that carried out the transformation of gypsum into carbonates and sulfur in the Calcare di Base have been characterized using lipid biomarkers (Ziegenbalg et al., 2012). Biomarkers point to a microbial community very similar to that carrying out the anaerobic oxidation of methane (AOM) coupled to sulfate reduction at recent marine cold seeps (Niemann and Elvert, 2008). In settings where oil is providing the source of carbon, however, the link between reactants (gypsum, oil) and products (carbonates, elemental sulfur) relies solely on the indirect evidence introduced above. Since microorganisms capable of reducing sulfate using non-methane hydrocarbons are known to exist (Widdel and Rabus, 2001; Kniemeyer et al., 2007), we investigated the lipid biomarker content of diagenetic carbonates from the Gulf of Suez (Pierre and Rouchy, 1988; Aref, 1998). These carbonates are formed via reduction of Middle Miocene evaporitic gypsum deposits and often occur in association with petroleum residues. Our main goal was to look for the molecular and the isotopic signatures of the microbial community suspected of mediating this biogeochemical process. 2. Study areas and samples Diagenetic carbonates were sampled from the Gebel El Zeit and Gemsa areas on the western coast of the Gulf of Suez, Egypt (Fig. 1, Table 1). These areas are well known for the occurrence of diagenetic carbonates and sulfur formed as end products of the microbial reduction of evaporitic gypsum (Pierre and Rouchy, 1988; Wali et al., 1989; Aref, 1998). Outcrops of diagenetic carbonates occur within a sequence of gypsum-anhydrite evaporites of Middle Miocene age that lies unconformably on the Precambrian basement. Authigenic carbonates in the Gebel el Zeit area occur generally as lenticular or irregular stratiform bodies ranging from a few centimeters to a few kilometers in length (Aref, 1998). These stratiform bodies are isolated within the sulfate layers and cannot be correlated with each other. In the Gemsa area the diagenetic carbonates are concentrated in the cap rock of a diapir, where they are associated with native sulfur deposits. Evidence that gypsum has been diagenetically transformed into carbonates in our study area comes from the petrography and stable isotope geochemistry of diagenetic carbonates and sulfur (Pierre and Rouchy, 1988; Philip et al., 1994; Aref, 1998): 1) the carbonates are generally depleted in 13C relative to seawater DIC implying that they contain inorganic carbon produced during the biodegradation of 13C-poor OM; 2) the elemental sulfur is depleted in 34S relative to seawater indicating that the reduction of sulfate is microbially-mediated; 3) carbonate pseudomorphs after gypsum–carbonate crystals having the crystal habit of the pre-existing gypsum crystals — are abundant; and 4) secondary porosity or brecciation, formed due to the reduction in volume accompanying the mineral transformation, is observed. Based on the oxygen isotopes of the carbonates, Pierre and Rouchy (1988) showed that the gypsum-carbonate transformation initiated early in a hypersaline solution during the deposition of gypsum and continued as the Miocene evaporites were uplifted and flushed by meteoric waters, inducing contact with petroleum migrating from deeper strata. In fact, both in the Gebel El Zeit and Gemsa areas petroleum seeps often occur in correspondence to the diagenetic carbonate outcrops. The petroleum source rocks in the Gulf of Suez are Upper Cretaceous Shales, lower Miocene marls and upper Miocene intragypsiferous diatomites. The Miocene evaporites make up the seal of this petroleum system and cap the reservoir rocks that are Paleozoic, Cretaceous and lower Miocene sandstones, fissured and cavernous

limestones of Late Cretaceous, Eocene and Miocene age, and porous basement rocks (Aref, 1998). The macroscopic appearance of the diagenetic carbonates we sampled (Fig. 2A–D) strongly resembles that of diagenetic carbonates described by Pierre and Rouchy (1988) and Aref (1998) from the same outcrops. Samples EG-06-11 and EG-06-17 are gray to dark gray, hard, porous carbonates with vugs up to about 2 mm in diameter. This porosity is interpreted as the result of the reduction in volume occurring during the transformation of sulfate minerals (gypsum or anhydrite) into carbonates (Aref, 1998). In addition to a strong oil smell, all samples contain light brown to black bitumen stains or oil microseeps that appear when breaking the sample mechanically. Samples EG-06-13, EG-06-14, EG-06-15 and EG-06-16 also contain macroscopic crystals of elemental sulfur, while sample EG-06-21 contains gypsum crystals. 3. Methods 3.1. Carbonate content and mineralogical composition The carbonate content of diagenetic carbonates was measured twice on 100 mg of powder using a Mélières manocalcimeter (MMC) accounting for the different stoichiometries of calcium carbonate (calcite or aragonite) and dolomite (Baudrand et al., 2012). The bulk mineralogy was determined by X-Ray diffractometry using a Siemens D-500 instrument (Ni filtered Cu Kα radiation) scanning 2°–64° (2θ) at a rate of 0.02° 2θ per second. The MacDiff program by Rainer Petschik (servermac.geologie.unifrankfurt.de/Rainer.html) was used to quantify the relative proportions of calcite and dolomite in carbonate mixtures from X-ray diffractograms using the peak surface method. The relative intensities were corrected using correction factors (I/Icor) from the PDF2 database of International Centre for Diffraction Data. Standard forms used are 05-0586 for calcite and 73-2324 for dolomite. The carbonate content measured by MMC was then used to calculate absolute concentrations of calcite and dolomite in the sample from the relative concentrations obtained by XRD. More details on the carbonate content and X-Ray diffractometry are presented in Baudrand et al. (2012). 3.2. Carbon and oxygen stable isotope compositions The authigenic carbonates we investigated contain mixtures of calcium carbonate and dolomite, which potentially have distinct diagenetic origins (Pierre and Rouchy, 1988). The carbon and oxygen stable isotope compositions of these different carbonate phases were determined according to the semi-automatic on-line method of Baudrand et al. (2012). Briefly, the stable isotope composition of dolomite is calculated after having measured that of calcite and bulk sample with an auto sampler MultiPrep™ system coupled to a Dual-Inlet GV-Instruments Isoprime™ isotope ratio mass spectrometer (irms). Results are expressed in ‰ relatively to Vienna Pee Dee Belemnite (VPDB) standard. The analytical precision depends on the proportions of calcite and dolomite in the sample (Baudrand et al., 2012). 3.3. Total organic carbon (TOC) and lipid biomarker analysis The amount of TOC in each carbonate sample was determined by Rock-eval-Pyrolysis (Espitalié et al., 1985–86) using subsamples devoid of hydrocarbon stains. Samples with the highest amount of non-petroleum TOC (selected by visual inspection) were analyzed for their lipid content. The carbon stable isotope composition of TOC (δ 13CTOC) was measured on hydrocarbon stained carbonate subsamples following dissolution of the carbonate in 0.1 M HCl at ambient temperature. δ 13CTOC values were obtained using a Eurovector Elemental Analyzer (EuroEA3028-HT) connected to a GV instrument

G. Aloisi et al. / Chemical Geology 347 (2013) 199–207

201

MEDITERRANEAN SEA

0

20

40 km 30°

Gu lf

28°

RE

EGYPT

DS

of

EA

26°

Su ez 24°

32°

34°

N Gebel El Zeit lf

Gu

EG-06-11/13 EG-06-14/15/16

of

EG-06-17

eit

lZ

lE

ez

be

Su

Ge

Gemsa Gu

lf

of

Su

ez

Oil Field

Gemsa Field

Miocene evaporites

EG-06-21 Fig. 1. Location map showing the sampling sites of authigenic carbonates in the Gulf of Suez (modified after (1998)).

Table 1 Mineralogy, isotopes geochemistry and lipid biomarker content of authigenic carbonates from the Gulf of Suez. Sample

Location

MCM

wt.%

Mineralogy Calcite

Dolomite

wt.%

wt.%

EG11

Ras Dib

96.4

91.3

EG13

Ras Dib

96.4

79.8

4.8

EG14 EG15 EG16 EG17

Gebel el Zeit Gebel el Zeit Gebel el Zeit Ras Zeit

EG21

Gemsa

15.5 93.5 92.1 91.7

100 100 99.5 100

56.7

93.4

91.5

41.3

0.9 Notes. n-Al: n-alkanes; McGDs: macrocyclic glycerol diethers; ARC: archaeol.

δ13CCO3

δ18OCO3

TOC

δ13CTOC

δ13Cn-Al

δ13CMcGDs

δ13CARC

‰PDB

‰PDB

wt-%

‰PDB

‰PDB

‰PDB

‰PDB

−24.5

−3.7

0.39

−31.7

−29 −33

n.d.

−31

−10.7 −15.8 −16.6 −12.6 −13.7 −13.0 −15.4

3.1 1.3 4.0 2.6 3.3 3.6 −2.3

0

n.d.

0.46 0.1 0.15 0.13

−28.7 −24.9 −26.4 −26.4

−10.6 −16.6

3.1 −9.6

0.11

−28.0

−28 −30

−24

n.d.

n.d.

−24 −31 −24 −26

202

G. Aloisi et al. / Chemical Geology 347 (2013) 199–207

Fig. 2. Macroscopic appearance (A-D, scale bar 1 cm) and thins section micrographs (E-H, scale bar 1 mm) of a selection of authigenic carbonates formed by microbial reduction of evaporitic gypsum in the Gulf of Suez.

Isoprime™ stable isotope-ratio mass spectrometer. Results are standardized using TPAC (Phenyl/Amine; C = 88.2%; δ 13C = -28.73‰) and are expressed in ‰ relatively to VPDB standard. Values are average of triplicate analyses.

Carbonate samples selected for lipid analysis were cleaned by scraping the surfaces and ground in a manual mortar. Lipids were extracted from the total carbonates by way of sonication [MeOH 2x, dichloromethane (DCM):MeOH (1:1 v/v) 2x and DCM 2x]. Following

G. Aloisi et al. / Chemical Geology 347 (2013) 199–207

removal of elemental sulfur with activated Cu, lipids were separated into an apolar and a polar fraction using chromatography over a wet packed column of inactivated (4% H2O) silica, with hexane:DCM (9:1 v/v) and DCM:MeOH (1:1 v/v) as eluents, respectively. Polar lipids were silylated [pyridine/N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA), 2:1 v/v; 1 h at 60 °C] and all lipid fractions were analysed by gas chromatography (GC), GC-mass spectrometry (GC-MS) and GC-isotope mass spectrometry (GC-irms) as described by Baudrand et al. (2010). Carbon isotope measurements on organic compounds have an error of less than ± 1.0‰. 4. Results and Discussion 4.1. Mineralogy and Petrography The carbonate content of the diagenetic carbonates is higher than 93 wt-% (Table 1). X-Ray diffraction analyses show that in samples EG-06-11, EG-06-13 and EG-06-21 calcite is very abundant (>79 wt-%), while dolomite is present only as a minor constituent (b15.5 wt-%). In sample EG-06-17 both calcite and dolomite are abundant. The carbonate fraction in samples EG-06-14, EG-06-15 and EG-06-16 is composed exclusively of dolomite. XRD analysis reveals the presence of traces of gypsum in samples EG-06-13 (wt-% = 1.9) and EG-06-21 (wt-% = 0.2), while traces of elemental sulfur are detected in samples EG-06-14 (wt-% = 0.1) and EG-06-15 (wt-% = 0.7). Petrographic thin sections (Fig. 2E-H) provide evidence that the carbonates were formed by substitution of pre-existing gypsum. Calcite pseudomorphs after gypsum are present in all samples. Fig. 2H shows acicular calcite crystals formed by substitution of gypsum needles. Calcite pseudomorphs are embedded in a fine-grained matrix where carbonate peloids are abundant. Sparitic calcite cements with individual crystals from 250 to 950 μm in size are also visible (Fig. 2G). This microfacies is devoid of matrix or peloids, suggesting that it represents the substitution of massive gypsum levels by carbonates. Pseudomorphoses are rarely visible in these sparitic intervals because the pre-existing gypsum crystals were probably in contact with one another in the absence of matrix to highlight their outline. Another common feature is the presence of millimeter-sized voids partially filled with sparitic calcite (Fig. 2F). This is the microscopic manifestation of the decrease in volume of the solid phase during the gypsum to carbonate transformation (Aref, 1998). In sample EG-06-13, gypsum crystals are visible and in contact with calcite (Fig. 2E). The pore-filling character of the gypsum crystals suggests that it is secondary gypsum formed by oxidation of elemental sulfur in oxic conditions at the outcrop, rather than remains of original gypsum before the diagenetic transformation into carbonate occurred. 4.2. Carbon and oxygen stable isotope composition The carbon and oxygen stable isotope ratios of the calcites and dolomites we investigated fall in the range of values previously observed by Pierre and Rouchy (1988) for diagenetic carbonates from the Gebel El Zeit and Gemsa areas (Fig. 3). With the exception of calcite in sample EG-06-11 (δ 13C = -24.5‰), the carbon stable isotope compositions of our carbonates fall in a limited interval of values ranging from -16.6‰ to -10.7‰. These values indicate that both calcite and dolomite contain carbon from a relatively 13C-poor organic source. In this particular diagenetic environment, the obvious process producing depleted DIC is microbial sulfate reduction; the question remains, however, of the carbon source that fuels this process. The extent of the 13C-depletion in microbial carbonates can be used as a tracer of the carbon source that fuels microbial processes. The main difference in isotopic signature is between methane (-110 b δ 13C‰ PDB b -30) and organic matter or petroleum (δ 13C ~ -25‰) (Whiticar, 1999; Killops and Killops, 2005). Thus, extremely 13 C-depleted carbonates (δ 13C b -30‰) can be safely ascribed to

203

methane oxidation (Ritger et al., 1987; Sakai et al., 1992; Bohrmann et al., 1998). For isotopically heavier carbonates, including the present ones, the interpretation is more complicated because either a formation from oxidation of organic matter or petroleum, or a mixture of DIC from methane oxidation and a heavier source, could explain their δ 13C. For example, an isotopically heavy source of DIC (δ 13C as high as + 30‰; (Claypool and Kaplan, 1974)) is produced during bacterial methanogenesis (Whiticar, 1999). When this heavy DIC source mixes with DIC from methane oxidation at modern cold seeps, diagenetic fluids with intermediate δ 13C values are produced (Chatterjee et al., 2011). Although the setting we are investigating is likely hydrologically different from that of cold seeps, this example shows that intermediate isotope DIC signatures (δ 13C ~ -20‰) do not imply the oxidation of single carbon source with intermetiate isotope composition. Moreover, a heavy DIC source has been detected in authigenic carbonates from our study area (Pierre and Rouchy, 1988), suggesting that controls on the δ 13C of DIC are multiple in our setting. We therefore cannot discriminate between methane and petroleumorganic matter as carbon sources for sulfate reduction based on the carbon isotopes of the carbonates alone. The δ18O value of authigenic carbonate is a function of the δ18O composition of the diagenetic fluid and of the temperature-dependent oxygen isotope fractionation between water and calcium carbonate. Calcites (-9.6‰ b δ18O b 1.3‰) and dolomites (2.6‰ b δ18O b 4.0‰) in the present natural carbonate mixtures have distinct isotopic signatures, implying that they precipitated in distinct diagenetic settings. Furthermore, the oxygen isotope composition of the seven dolomites shows little variation, suggesting that all dolomites originate from the same diagenetic fluid. The much larger range of δ18O values observed for calcite implies some variability in the δ18O composition of the precipitating fluid and/or of the temperature of precipitation. We used the relations of O'Neil et al. (1969) for calcite and of Fritz and Smith (1970) for dolomite to test salinity-temperature scenarios for the diagenetic environments in which the Gebel El Zeit and Gemsa carbonates formed. Calcite in sample EG11-06-21 has a δ 18O of -9.6‰, suggesting precipitation from a fresh or poorly saline 18O-depleted fluid. Supposing that the precipitating fluid is isotopically similar to the aquifers of the eastern Gulf of Suez with δ 18O as low as -6.6‰ vs SMOW (Mazor et al., 1973; Gat and Issar, 1974), the temperature of formation of this calcite can be estimated to have been 27 °C. Lower precipitation temperatures would imply a greater 18O-depletion of the diagenetic fluid. A higher temperature (about 65 °C) and an isotopically heavier seawater-like fluid (δ18O = 0‰) would produce an isotopically similar calcite but would imply some degree of burial. However, geological evidence for extensive burial of the evaporite formation is lacking (Pierre and Rouchy, 1988). Isotopically heavier calcites in samples EG-06-11 (δ18O = -3.7‰) and EG-06-17 (δ18O = -2.3‰) are consistent with precipitation from a seawater-like fluid (δ18O = 0‰) at temperatures of 22 °C and 30 °C respectively, or with higher salinity fluids at higher temperature. These δ18O values are not consistent, however, with precipitation from a meteoric fluid with δ18O = -6.6‰ because they would imply a temperature approaching 0 °C, which is not realistic in this diagenetic scenario. Supposing a temperature of 15 °C, the δ18O value of the diagenetic fluid for the calcite in sample EG-06-13 (δ18O = 1.3‰ vs PDB) would be 3.3‰ vs SMOW. Since the temperature is likely to have been at least 15 °C, it is safe to say that this calcite precipitated from a fluid more 18O-rich and saline than seawater. Similarly, by supposing a minimum precipitation temperature of 15 °C, the range of δ 18O values of the dolomites implies precipitation from diagenetic fluids with a δ 18O range from 4.2 to 5.7‰ vs SMOW. Again, higher temperatures are likely in this evaporitic environment, implying that dolomites precipitated from a saline 18O-rich diagenetic fluid. To summarize, oxygen isotopes suggest that diagenetic calcites precipitated from fluids with a wide range of salinity (higher than seawater for EG-06-13 to fresh water for EG-06-21), while dolomites precipitated from fluids more saline than seawater.

204

G. Aloisi et al. / Chemical Geology 347 (2013) 199–207

Fig. 3. (A) Carbon and oxygen stable isotope composition of authigenic calcite and dolomite from the Gulf of Suez. δ13C and δ18O data are superposed on the data of Pierre and Rouchy (1988). (B) Relation between the δ13C of authigenic calcite and dolomite and δ13C of TOC.

Aref (1998) and Pierre and Rouchy (1988) showed that carbonates from the studied areas formed by the diagenetic substitution of evaporitic gypsum during two distinct diagenetic phases. Syngenetic carbonates formed during early diagenesis, coeval with or shortly following the deposition of gypsum, in marine pore water with different levels of evaporation. Thus, they are relatively 18O-rich. In this case, the microbes that catalyze the reduction of gypsum use substrates contained in organic matter deposits associated with the evaporitic gypsum. All dolomites of the present study and calcite in sample EG-06-13 have the isotopic characteristic of syngenetic carbonates (Aref, 1998). Epigenetic carbonates, on the other hand, are formed in a later diagenetic stage, when gypsum formations are uplifted and exposed to meteoric waters, resulting in fresh or poorly saline diagenetic fluids. Epigenetic carbonates are typically 18O-poor and show petroleum microseepage or bitumen stains. The carbon source for the microbial community reducing mineral gypsum is thought to be contained in petroleum (Pierre and Rouchy, 1988; Aref, 1998). Based on its 18O-depletion, the calcite in samples EG-06-21 is epigenetic. The calcite in samples EG-06-11 and EG-06-17 may represent an intermediate situation between syngenesis and epigenesis. 4.3. Total organic carbon (TOC) and lipid biomarkers The TOC content was measured on portions of carbonates devoid of hydrocarbon stains in order to determine the non-petroleum (i.e. from ancient living biomass) OM content of the carbonates. Consistently, we obtained rather low TOC concentrations (from 0.1% to 0.4%; Table 1). These low TOC contents do not preclude carbon from the carbonate subsamples to be derived from the biodegradation of the surrounding oils since: i) petroleum stains were present in almost all carbonate samples investigated, and ii) fluids associated with any carbonate subsample likely contained oil-derived DIC. The δ 13C values of the OM in the different carbonates varies from −31.7‰ to − 24.9‰. This variability likely reflects the presence of hydrocarbons with distinct δ 13C signatures in the samples used for TOC analyses. In samples where calcite dominates in the carbonate fraction (ca. between 50% and 95% of calcite), the δ 13C of TOC is correlated with the δ 13C of calcite (Fig. 3). This is consistent with a large part of the calcite carbon being derived from hydrocarbons. This is not the case for dolomites, which show rather constant δ 13C values and no correlation with the δ 13C of TOC (Fig. 3). This observation is consistent with the dolomites being formed syngenetically – as suggested by their 18O-isotope composition – by microbial reduction of sedimentary organic matter. Indeed, it is likely that sedimentary organic matter has a more uniform carbon isotope signature within the same basin

(Sackett, 1989), compared to petroleum hydrocarbons that may have migrated from different source areas (Killops and Killops, 2005). The presence of oils having multiple origins is consistent with the distribution of individual hydrocarbons present in the isolated apolar lipid fractions analysed by GC-MS (Fig. 4a). Indeed, each hydrocarbon profile is characterized by specific sterane and hopane distributions. For instance, the proportion of C35 extended hopane relative to total C31 to C35 homohopanes (22S + 22R) is about 21%, 14% and 12% in samples EG-06-21, EG-06-11 and EG-06-17, respectively. The presence of other specific and unique hydrocarbons (e.g., sample EG-06-21 shows a relatively strong proportion of gammacerane that is not present in samples EG-06-11 and EG-06-17) and the different carbon isotopic composition of n-alkanes in samples EG-06-11 and EG-06-17 further support the presence of different oils. Finally, hydrocarbon profiles suggest different degrees of degradation for the different oils, with oil from sample EG-06-21 appearing more heavily degraded than the two other ones (Fig. 4). To further investigate the microbial transformation of gypsum into carbonates, we looked for potential lipid biomarkers of the microbial community that catalyzed this process. GC-MS analysis revealed abundant microbial lipids in the polar lipid fraction of the most degraded sample (EG-06-21; Fig. 4b). Some of these lipids (compounds B1-B3) have been previously identified as a series of non-isoprenoid macrocyclic glycerol diethers (McGDs), and have been tentatively ascribed to bacteria closely involved with sulfur cycling under reducing conditions (Baudrand et al., 2010). These lipids occur together with archaeol, indicating the presence of a microbial community composed of anaerobic bacteria and Archaea (Baudrand et al., 2010). Trace amounts of a non-isoprenoid dialkyl glycerol diether (tentatively identified as di-C17:1 glycerol diether) and of extended archaeol (C25/C20 archaeol) were detected in some samples (Fig. 4b), attesting for the presence of sulfate-reducing bacteria (Pancost and Sinninghe Damsté, 2003) and of halophilic archaea (Teixidor et al., 1993; Ziegenbalg et al., 2012), respectively. The same lipid assemblage was detected in samples EG-06-11 and EG-06-17, although the proportions of the different microbial lipids vary from one sample to the other (Fig. 4b). Interestingly, the abundance of bacterial McDGs seems, to some extent, linked to the level of biodegradation of the hydrocarbons, with the higher amounts of these lipids being detected in the most heavily biodegraded sample EG-06-21. The McGDs present in the most degraded sample (EG-06-21) are 2 to 4‰ depleted in 13C relative to n-alkanes in the same sample. Typically, the carbon isotope composition of prokaryote heterotrophic biomass is similar to that of their carbon source, while the acetogenic

G. Aloisi et al. / Chemical Geology 347 (2013) 199–207

A) Apolar lipids

C30 H (-29

C29 H (-28

C22 alk (-27

205

C21 alk (-26 C27 St (-27

EG-21

Ph Ga H H H H H C21 alk (-30

C22 alk (-30

*

C30 H (-29

*

*

*

Sq (-29 *

*

*

*

*

*

EG-11

*

*

Ph

H

*

Relative intensity

* Pr

* *

C21 alk (-25

Pr *

*

*

*

*

*

*

*

*

*

20

25

EG-17

C30 H (-29

*

*

*

*

15

*

C27 St (-26 *

iC16

10

* * H H

C22 alk (-26

*

Ph

iC18

H *

30

*

35

*

*H *

40

H *

* H* 45

*

*

*

*

50

*

*

55

*

* 60

Retention time (min.) Macrocyclic non-isoprenoid DGD (Bacteria)

B) Polar lipids

13C

B1 : - 28

13C

B2 : - 30

13C

B3 : - 30

13C

Ar : - 24

e.g. B1 Archaeol (Archaea) B2

EG-21

B3 C16 FA (-29

C37:2DGD

Ar

Ext Ar

Ar (- 31 C18 FA (-33

C16 FA (-31

EG-11

Relative intensity

Poll. Poll.

B1 B 2

C16 FA (-29 C18 FA (-27

EG-17

10

B1 B

15

20

25

30

35

40

45

50

2

55

Ar

60

65

Retention time (min.) Fig. 4. Total ion chromatrograms of apolar lipid fractions A) and selected ion chromatrograms (m/z 130 + 133 + 145) of polar lipid fractions B) from carbonate samples EG-21, EG-11 and EG-17. Ph = phytane, Pr = pristane, iC16 = C16 isoprenoid, * = n-alkanes, C21 alk = C21 n-alkane, C27 St = cholestane, Sq = squalane, H = a,b-hopane, FA = fatty acid, C37:2-DGD = unidentified di-C17:1 glycerol diether, Ar = archaeol, Ext Ar = extended archaeol. Note: Cont. = contaminant.

206

G. Aloisi et al. / Chemical Geology 347 (2013) 199–207

lipids in these organisms are about 4‰ depleted relative to biomass (Pancost and Pagani, 2006). Taken together, the 13C composition of n-alkanes and of McGDs, and the relationship between their abundance and the level of biodegradation of the oil, suggest that the McGDs have been synthesized by a bacterial population growing heterotrophically on hydrocarbons (Fig. 5). Unfortunately, the low amount of these lipids in the two other samples did not allow the measurement of their 13C composition. The McGDs observed in the present study had been initially related to extreme (hyperthermophilic and/or hyperhalophilic) conditions (Baudrand et al., 2010). Although similar lipid structures have been reported from hydrothermal systems (Pancost and Pagani, 2006; Blumenberg et al., 2007), they have also been described in hypersaline non-hydrothermal settings from the Messinian of Sicily (Ziegenbalg et al., 2012) and in an ancient (Pliocene) coastal marine setting (van Dongen et al., 2007). They have never been observed in any cultured bacteria. The stable oxygen isotopic composition of diagenetic calcite in the present Egyptian samples (Fig. 3) indicates that carbonate precipitation occurred under non-extreme conditions, pointing to an origin of McGDs from mesophilic (likely halophilic), hydrocarbon-degrading sulfate-reducing bacteria. The 13C composition of archaeol could be determined in the two most degraded samples (EG-06-21 and EG-06-11). In both cases, archaeol is 6-7‰ depleted in 13C relative to calcite (Table 1). This isotopic shift corresponds to the equilibrium difference in δ 13C between dissolved CO2 and HCO3- (Zeebe and Wolf-Gladrow, 2001). One possible explanation for these observations is that archaeol is produced by CO2-utilising archaeabacteria and that the 13C difference with diagenetic calcite records the isotopic equilibrium of dissolved CO2 and HCO3-. This seemingly implies a lack of isotopic discrimination during CO2 uptake and biosynthesis of archaeol, which is against theory and environmental evidence (Hayes, 1993). Such apparent lack of isotopic discrimination could be produced by a nearly complete utilization of the CO2 source that would mask the instantaneous isotope effects. In addition, efficient CO2 removal would contribute – alongside sulfate reduction - in producing the thermodynamic conditions for carbonate mineral precipitation. Taken together, the carbon stable isotope signature of carbonates and biomarkers suggests that the DIC in this diagenetic environment is a mixture between DIC produced by microbial degradation of oil and DIC from a more 13C-rich source (Fig. 5). A recent study of the Calcare di Base formation (Miocene, Sicily) provides the first biomarker evidence for the involvement of sulfatereducing bacteria in the transformation of gypsum into carbonates (Ziegenbalg et al., 2010, 2012). The biomarker patterns in the Calcare di Base resemble those observed at modern and ancient methaneseeps, where methane is oxidised anaerobically by a consortium of anaerobic methane-oxidizing archaea (ANME) and sulphate-reducing bacteria (Boetius et al., 2000; Pancost and Sinninghe Damsté, 2003; Knittel and Boetius, 2009). Such consortia are characterized by specific lipids (i.e. archaeal biomarkers such as crocetane, pentamethylicosane,

archaeol or hydroxyarchaeol as well as iso- and anteiso-odd-numbered fatty acids derived from sulfate-reducing bacteria) with highly negative δ13C values (between -50‰ and -110‰). We have not detected most of these biomarkers in our carbonates, and the dominant archaeal lipid we did detect is the taxonomically widespread archaeol which here is characterized by high rather than low δ13C values. Furthermore, we did not find lipids specific of aerobic methanotrophic bacteria (e.g. 13C-depleted hopanoids, (Birgel and Peckmann, 2008; Birgel et al., 2011)), indicating the absence of these organisms. Rather, the δ13C values of the Egyptian carbonates, together with the molecular and the 13C composition of the associated (biodegraded) oils and microbial lipids, suggest that non-methanic hydrocarbons have constituted a major carbon source for the microbial transformation of gypsum into carbonates. Sulfate-reducing bacteria have the ability to degrade petroleum hydrocarbons anaerobically (Widdel and Rabus, 2001; Kniemeyer et al., 2007). This process is thought to trigger the formation of diagenetic carbonates at seafloor oil seeps (Roberts and Aharon, 1994; Naehr et al., 2009; Birgel et al., 2011; Chevalier et al., 2011). However, the lipid signature of bacteria involved in carbonate production at oil seeps is unknown, which may explain why specific biomarkers of oil-degrading bacteria have not been detected in seep carbonates so far (Birgel et al., 2011). On the other hand, the anaerobic metabolic pathways of hydrocarbon oxidation involved during the microbiallymediated carbonate formation from gypsum and oil also remain unknown, and likely deserve further attention as they may yield specific metabolites with a high diagnostic value. 5. Conclusion We provide the first biomarker evidence suggesting that a microbial community dominated by sulfate-reducing bacteria is capable of using petroleum hydrocarbons to reduce gypsum and produce substantial amounts of diagenetic carbonates in the Gebel El Zeit and Gemsa areas. This microbially-mediated carbonate formation from gypsum and petroleum hydrocarbons has been previously hypothesized based on the carbon stable isotope composition of carbonates and on the sulfur stable isotope composition of associated elemental sulfur (Feely and Kulp, 1957; Davis and Kirkland, 1970; Barker et al., 1979; Ruckmick et al., 1979). Our results provide the first direct evidence for this microbial link, including a description of the population’s biomarker profile, and could be broadly relevant to a range of the oil-gypsum-carbonate diagenetic associations that have been described. Acknowledgements The authors thank François Martineau and François Fourel for assistance with the stable isotope analysis of diagenetic carbonates and of TOC. This work has been funded by the French National Research Agency (ANR) through the MICROCARB project (Grant to GA) and by the Institut Universitaire de France (Grant to CL). Jörn Peckmann and an anonymous reviewer provided useful comments that improved the manuscript. References

Fig. 5. Model proposed for the biogeochemical processes involved in the transformation of gypsum to carbonates by sulfate-reducing bacteria.

Aref, M.A.M., 1998. Biogenic carbonates: are they a criterion for underlying hydrocarbon accumulations? An example from the Gulf of Suez region, Egypt. AAPG Bulletin 82 (2), 336–352. Barker, J.M., Cochran, D.E., Semrad, R., 1979. Economic Geology of the Mishraq native sulfur deposit, northern Iraq. Economic Geology 74 (2), 484–495. Baudrand, M., Grossi, V., Pancost, R.D., Aloisi, G., 2010. Non-isoprenoidal macrocyclic glycerol diethers associated with Miocene authigenic carbonates. Organic Geochemistry 41 (12), 1341–1344. Baudrand, M., et al., 2012. Semi-automatic determination of the carbon and oxygen stable isotope compositions of calcite and dolomite in natural mixtures. Applied Geochemistry 27 (1), 257–265.

G. Aloisi et al. / Chemical Geology 347 (2013) 199–207 Birgel, D., Peckmann, J., 2008. Aerobic methanotrophy at ancient marine methane seeps: a synthesis. Organic Geochemistry 39 (12), 1659–1667. Birgel, D., Feng, D., Roberts, H.H., Peckmann, J., 2011. Changing redox conditions at cold seeps as revealed by authigenic carbonates from Alaminos Canyon, northern Gulf of Mexico. Chemical Geology 285 (1–4), 82–96. Blumenberg, M., Seifert, R., Petersen, S., Michaelis, W., 2007. Biosignatures present in a hydrothermal massive sulfide from the Mid-Atlantic Ridge. Geobiology 5, 435–450. Boetius, A., et al., 2000. A marine microbial consortium apparently mediating anaerobic oxidation of methane. Nature 407 (6804), 623–626. Bohrmann, G., Greinert, J., Suess, E., Torres, M., 1998. Authigenic carbonates from the Cascadia subduction zone and their relation to gas hydrate stability. Geology 26 (7), 647. Canfield, D.E., Thamdrup, B., Hansen, J.W., 1993. The anaerobic degradation of organic matter in Danish coastal sediments — iron reduction, manganese reduction and sulfate reduction. Geochimica et Cosmochimica Acta 57 (16), 3867–3883. Chatterjee, S., et al., 2011. Pore water sulfate, alkalinity, and carbon isotope profiles in shallow sediment above marine gas hydrate systems: a numerical modeling perspective. Journal of Geophysical Research 116 (B9). Chevalier, N., et al., 2011. Authigenic carbonates at cold seeps in the Marmara Sea (Turkey): a lipid biomarker and stable carbon and oxygen isotope investigation. Marine Geology 288 (1–4), 112–121. Claypool, G.E., Kaplan, I.R., 1974. The origin and distribution of methane in marine sediments. In: Kaplan, I.R. (Ed.), Natural Gases in Marine Sediments. Plenum, New York, pp. 99–139. Davis, J.B., Kirkland, D.W., 1970. Native sulfur deposition in castile formation, Culberson-County, Texas. Economic Geology 65 (2), 107–121. Espitalié, J., Deroo, G., Marquis, F., 1985–86. La pyrolise Rock Eval et ses applications. Partie 1. Revue de l'Institut Français du Pétrole 40 (5), 563–579. Feely, H.W., Kulp, J.L., 1957. Origin of Gulf Coast salt dome sulfur deposits. American Association of Petroleum Geologists Bulletin 41, 1802–1853. Fritz, P., Smith, D.G.W., 1970. Isotopic composition of secondary dolomites. Geochimica et Cosmochimica Acta 34 (11), 1161–1173. Gat, J.R., Issar, A., 1974. Desert isotope hydrology: water sources of the Sinai Desert. Geochimica et Cosmochimica Acta 38, 1117–1131. Hayes, J.M., 1993. Factors controlling 13C contents of sedimentary organic compounds: Principles and evidence. Marine Geology 113, 111–125. Hinrichs, K.-U., Boetius, A., 2002. The Anaerobic Oxidation of Methane: New insights in Microbial Ecology and Biogeochemistry, Ocean Margin Systems. Springer, Heidelberg, pp. 457–477. Jørgensen, B.B., 1982. Mineralization of organic matter in the sea bed-the role of sulphate reduction. Nature 296, 643–645. Kasten, S., Jørgensen, B.B., 2000. Marine Geochemistry. Springer, Berlin, pp. 263–281. Killops, S.D., Killops, V.J., 2005. Introduction to Organic Geochemistry. Blackwell Publishing, Oxford (393 pp.). Kniemeyer, O., et al., 2007. Anaerobic oxidation of short-chain hydrocarbons by marine sulphate-reducing bacteria. Nature 449 (7164), 898–901. Knittel, K., Boetius, A., 2009. Anaerobic Oxidation of Methane: Progress with an Unknown Process, Annual Review of Microbiology. Annual Review of Microbiology 311–334. Mazor, E., Nadler, A., Molcho, M., 1973. Mineral springs in the Suez Rift Valley — comparison with waters in the Jordan Rift Valley and postulation of marine origin. Journal of Hydrology 20, 289–309. McKenzie, J.A., 1985. Stable-isotope mapping of Messinian evaporative carbonates of cantral Sicily. Geology 13 (12), 851–854. Naehr, T.H., Birgel, D., Bohrmann, G., MacDonald, I.R., Kasten, S., 2009. Biogeochemical controls on authigenic carbonate formation at the Chapopote “asphalt volcano”, Bay of Campeche. Chemical Geology 266 (3–4), 390–402.

207

Niemann, H., Elvert, M., 2008. Diagnostic lipid biomarker and stable carbon isotope signatures of microbial communities mediating the anaerobic oxidation of methane with sulphate. Organic Geochemistry 39 (12), 1668–1677. O'Neil, J.R., Clayton, R.N., Mayeda, T.K., 1969. Oxygen isotope fractionation in divalent metal carbonates. Journal of Chemical Physics 51 (12), 5547–5558. Pancost, R.D., Pagani, M., 2006. Controls on the carbon isotopic composition of lipids in marine environments. The Handbook of Environmental Chemistry, pp. 209–249. Pancost, R.D., Sinninghe Damsté, J.S., 2003. Carbon isotopic compositions of prokaryotic lipids as tracers of carbon cycling in diverse settings. Chemical Geology 195 (1–4), 29–58. Peckmann, J., Paul, J., Thiel, V., 1999. Bacterially mediated formation of diagenetic aragonite and native sulfur in Zechstein carbonates (Upper Permian, Central Germany). Sedimentary Geology 126 (1–4), 205–222. Philip, G., Wall, A.M.A., Aref, M.A.M., 1994. On the origin of native sulfur deposits In Gebel-El-Zeit, Gulf Of Suez, Egypt. Carbonates and Evaporites 9 (2), 223–232. Pierre, C., Rouchy, J.M., 1988. Carbonate replacements after sulfate evaporites in the Middle Miocene of Egypt. Journal of Sedimentary Petrology 58 (3), 446–456. Ritger, S., Carson, B., Suess, E., 1987. Methane-derived authigenic carbonates formed by subduction-induced pore-water expulsion along the Oregon/Washington margin. Geological Society of America Bulletin 98 (2), 147. Roberts, H.H., Aharon, P., 1994. Hydrocarbon-derived carbonate buildups of the northern Gulf-of-Mexico continental slope — a review of submersible investigations. Geo-Marine Letters 14 (2–3), 135–148. Rouchy, J.-M., 1981. La Genèse des Evaporites Massiniennes de Méditerranée. Rouchy, J.M., et al., 1998. Sedimentary and diagenetic markers of the restriction in a marine basin: the Lorca Basin (SE Spain) during the Messinian. Sedimentary Geology 121 (1–2), 23–55. Ruckmick, J.C., Wimberly, B.H., Edwards, A.F., 1979. Classification and genesis of biogenic sulfur deposits. Economic Geology 74 (2), 469–474. Sackett, W.M., 1989. Stable carbon isotope studies on organic matter in the marine environment. Handbook of Environmental Isotope Geochemistry. Elsevier, pp. 139–169. Sakai, H., Gamo, T., Ogawa, Y., Boulegue, J., 1992. Stable isotopic ratios and origins of the carbonates associated with cold seepage at the eastern Nankai Trough. Earth and Planetary Science Letters 109, 391–404. Teske, A., 2010. Sulfate-reducing and methanogenic hydrocarbon-oxidizing microbial communities in the marine environment. Part 21: Microbial Communities based on hydrocarbons, oils and fats: Natural habitats. In: Timmis, Kenneth (Ed.), Handbook of Hydrocarbon Microbiology. Springer, pp. 2203–2223. Teixidor, P., Grimalt, J.O., Pueyo, J.J., Rodriguez-Valer, F., 1993. Isopranylglycerol diethers in non-alkaline evaporitic environments. Geochimica et Cosmochimica Acta 57, 4479–4489. van Dongen, B.E., et al., 2007. Formation of iron sulfide nodules during anaerobic oxidation of methane. Geochimica et Cosmochimica Acta 71 (21), 5155–5167. Wali, A.M.A., Eldougdoug, A.A., Aref, M.A.M., 1989. Geology, isotope geochemistry, and the role of salt spines sulfur genesis, Gemsa Area, Red Sea, Egypt. Annals of the geological survey of Egypt XVI, 1–15. Whiticar, M.J., 1999. Carbon and hydrogen isotope systematics of bacterial formation and oxidation of methane. Chemical Geology 161, 291–314. Widdel, F., Rabus, R., 2001. Anaerobic biodegradation of saturated and aromatic hydrocarbons. Current Opinion in Biotechnology 12 (3), 259–276. Zeebe, R.E., Wolf-Gladrow, D., 2001. CO2 in Seawater: equilibrium, kinetics, isotopes. Elsevier Oceanographic Series, Amsterdam (346 pp.). Ziegenbalg, S.B., et al., 2010. Formation of secondary carbonates and native sulphur in sulphate-rich Messinian strata, Sicily. Sedimentary Geology 227 (1–4), 37–50. Ziegenbalg, S.B., et al., 2012. Anaerobic oxidation of methane in hypersaline Messinian environments revealed by 13C-depleted molecular fossils. Chemical Geology 292–293, 140–148.