Source rock and shale oil potential of the Pabdeh Formation (Middle–Late Eocene) in the Dezful Embayment, southwest Iran

 B. Alizadeh, A. Opera, M. Kalani, M. Alipour, 2020 CC BY-SA B . A l i z a d e h e t a l . G e o l o g i c a A c t a , 1 8 . 1 5 , 1 2 2 , I I V ( 2 0 2 0 ) D O I : 1 0 . 1 3 4 4 / G e o l o g i c a A c t a 2 0 2 0 . 1 8 . 1 5 Organic geochemistry and mineralogy of the Pabdeh Fm. 2 oil potentials associated with a petroleum system (Gross et al., 2015; Jarvie, 2012, 2014; Li et al., 2018; Permanyer et al., 2016; Song et al., 2017; Uffmann et al., 2012; Zhao et al., 2014). The Middle Cretaceous–Early Miocene Petroleum System (Fig. 1), is one of the five petroleum systems recognized in the Zagros Fold Belt and the adjacent Persian Gulf Basin (Bordenave and Hegre, 2010; Bordenave, 2014). This petroleum system comprises two active source rocks (i.e. the Kazhdumi and Pabdeh formations) with two reservoirs (i.e. the Asmari Formation and the Bangestan Group). The Pabdeh Formation is recognized as the only active hydrocarbon generating Paleogene source rock (Fig. 1), and is a secondary source for the Middle CretaceousEarly Miocene petroleum system in the Dezful Embayment (Alizadeh et al., 2012; Bordenave and Hegre, 2010; Opera et al., 2013). In contrast, the Albian Kazhdumi Formation has had substantial contribution to the Asmari reservoir by vertical migration through faults and fracture networks during the Zagros orogeny. Nevertheless, contributions from both source intervals continue until the present (Bordenave and Burwood, 1990; Bordenave and Hegre, 2010). Several studies have investigated the petroleum systems and organic geochemistry of oils and source rocks in the Zagros Fold Belt (Ala et al., 1980; Asadi Mehmandosti et al., 2015; Baniasad et al., 2016; Bordenave and Hegre, 2010; Mashhadi et al., 2015; Opera et al., 2013; Rabbani et al., 2014; Soleimani and Zamani, 2015). In addition, many studies have characterized the Kazhdumi source rock and associated hydrocarbons in the Dezful Embayment (e.g. Alizadeh et al., 2017, 2018a; Baniasad et al., 2019; Sfidari et al., 2016). However, the hydrocarbon potential of the Pabdeh Formation is not fully addressed in the Dezful Embayment and, except for some preliminary and local geochemical studies, little information exists about its organic geochemistry and generation potential (e.g. Alizadeh et al., 2018b, 2019; Bordenave, 2014; Hatampour, 2014; Karimi et al., 2016a, b). Based on previous studies, the Pabdeh Formation has TOC (2-5wt.%) and HI values (300600mg HC/g TOC) typical of Type II organic matter with low thermal maturity (RO≤0.6%). Existing oil-source rock correlation studies have shown that the Pabdeh Formation becomes an important source rock for the Asmari reservoirs in the NE parts of the Dezful Embayment (i.e. where the Kazhdumi Formation constitutes a non-source limestone facies). By contrast, in the central and SW parts of the Dezful Embayment, the main source rock is the Kazhdumi Formation, while the Pabdeh Formation does not reach sufficient maturity for hydrocarbon generation (immature to early mature) (Bordenave and Hegre, 2010). The Brown Shale Unit (BSU) constitutes the middle part of the Pabdeh Formation throughout the Dezful Embayment. However, the exact geographic distribution of this unit and its extension into the adjacent basins (Lurestan and Fars) are still unclear. Recent drillings in the Dezful Embayment have indicated that the BSU, aside from being a significant source rock for the Asmari reservoir can have considerable potential for shale oil resource development, thereby opening a new exploration doorway. The objectives of this study are to comprehensively investigate the mineralogical and organic geochemical characteristics of the BSU in order to shed more light on the possible shale oil potential in the Dezful Embayment. We build upon fresh analytical data obtained from drill cores and cutting samples from four major oilfields located to the south of the Dezful Embayment (Fig. 2). Our results provide valuable insights into the dynamic evolution of the Middle CretaceousEarly Miocene petroleum system in the southern Dezful Embayment and have significant implications for future exploration and production activities.


INTRODUCTION
The most critical element for the existence of a petroleum system is the presence of a source rock capable of generating adequate amounts of hydrocarbons (Magoon and Dow, 1994). Although the source rocks were traditionally regarded to account for the conventional resources, organic-rich source rocks can contain huge amounts of unconventional hydrocarbon resources (Curiale and Curtis, 2016). Therefore, recent trends in source rock evaluation projects mostly recommend a thorough investigation about the unconventional shale gas and shale The Pabdeh Brown Shale Unit (BSU) is an organic-rich calcareous mudstone within the Paleogene Pabdeh Formation, which has not yet been investigated in detail. A total of 166 core and cutting samples were selected from four wells in the Dezful Embayment to investigate the organic geochemical and mineralogical compositions, as well as the shale oil potential of the BSU. X-Ray Diffraction (XRD) results show that it is mainly comprised of calcite (53wt.%), clay minerals (25wt.%) and quartz (14wt.%). Total Organic Content (TOC) values generally range from 1 to 9wt.% (avg. 4.2, 2.9, 5.2 and 3.3wt.%, for GS, KR, RR and RS wells, respectively) with Hydrogen Index (HI) values ranging between 400 and 650mg HC/g TOC. Based on average values of T max and vitrinite reflectance, as well as saturate biomarker ratios, the BSU is immature at wells RR and RS (ranging from 0.3 to 0.53%) and its maturity increases northward at wells KR and GS (ranging from 0.5% to 0.67%). The organic matter is dominated by Type ΙΙ kerogen and is generally composed of liptinite and amorphous material with minor terrestrial input. Based on various biomarker parameters, the organic matter was most likely deposited under anoxic marine conditions. The favorable mineralogical composition (i.e. presence of brittle minerals) and organic geochemical properties (i.e. TOC>2wt% and Type II kerogen) support the conclusion that the Pabdeh BSU displays a considerable shale oil potential where it attains appropriate thermal maturity.
Several studies have investigated the petroleum systems and organic geochemistry of oils and source rocks in the Zagros Fold Belt (Ala et al., 1980;Asadi Mehmandosti et al., 2015;Baniasad et al., 2016;Bordenave and Hegre, 2010;Mashhadi et al., 2015;Opera et al., 2013;Rabbani et al., 2014;Soleimani and Zamani, 2015). In addition, many studies have characterized the Kazhdumi source rock and associated hydrocarbons in the Dezful Embayment (e.g. Alizadeh et al., 2017Alizadeh et al., , 2018aBaniasad et al., 2019;Sfidari et al., 2016). However, the hydrocarbon potential of the Pabdeh Formation is not fully addressed in the Dezful Embayment and, except for some preliminary and local geochemical studies, little information exists about its organic geochemistry and generation potential (e.g. Alizadeh et al., 2018bAlizadeh et al., , 2019Bordenave, 2014;Hatampour, 2014;Karimi et al., 2016a, b). Based on previous studies, the Pabdeh Formation has TOC (2-5wt.%) and HI values (300-600mg HC/g TOC) typical of Type II organic matter with low thermal maturity (R O ≤0.6%). Existing oil-source rock correlation studies have shown that the Pabdeh Formation becomes an important source rock for the Asmari reservoirs in the NE parts of the Dezful Embayment (i.e. where the Kazhdumi Formation constitutes a non-source limestone facies). By contrast, in the central and SW parts of the Dezful Embayment, the main source rock is the Kazhdumi Formation, while the Pabdeh Formation does not reach sufficient maturity for hydrocarbon generation (immature to early mature) (Bordenave and Hegre, 2010).
The Brown Shale Unit (BSU) constitutes the middle part of the Pabdeh Formation throughout the Dezful Embayment. However, the exact geographic distribution of this unit and its extension into the adjacent basins (Lurestan and Fars) are still unclear. Recent drillings in the Dezful Embayment have indicated that the BSU, aside from being a significant source rock for the Asmari reservoir can have considerable potential for shale oil resource development, thereby opening a new exploration doorway. The objectives of this study are to comprehensively investigate the mineralogical and organic geochemical characteristics of the BSU in order to shed more light on the possible shale oil potential in the Dezful Embayment. We build upon fresh analytical data obtained from drill cores and cutting samples from four major oilfields located to the south of the Dezful Embayment (Fig. 2). Our results provide valuable insights into the dynamic evolution of the Middle Cretaceous-Early Miocene petroleum system in the southern Dezful Embayment and have significant implications for future exploration and production activities.

GEOLOGICAL SETTING
The Zagros Fold Belt is part of the Alpine-Himalayan orogenic system that resulted from the closure of the Neo-Tethys ocean between the Arabian and European plates during Cenozoic times (Alavi and Mahdavi, 1994;Homke et al., 2004). The Zagros Fold Belt, characterized by NW-SE trending folds and thrust faults, extends over 2000km from eastern Turkey and Kurdistan region in north Iraq to southern Iran (Bahroudi and Talbot, 2003;Sherkati et al., 2006). According to the structural and the stratigraphic features several main domains are defined from NW to SE including the Lurestan area, the Dezful Embayment, the Izeh Zone and the Fars Province (Sepehr and Cosgrove, 2004) (Fig. 2). The Dezful Embayment, which contains the most important oil reservoirs in the Zagros Fold Belt is bounded on the northwest by the transverse strike-slip Balarud Fault Zone, on the northeast by the Izeh Fault Zone, on the north by the Mountain Front Fault, and on the east by the Kazerun Fault Zone (Mouthereau et al., 2012;Sepehr and Cosgrove, 2004) (Fig. 2).
From the stratigraphic point of view, a thick sedimentary package exists in the Dezful Embayment, which comprises 9-12km-thick sediments formed under various tectonic settings, i.e. passive margin during Paleozoic and Mesozoic and active compressional setting during Cenozoic. Chronostratigraphic correlations of rock units indicate continuous sedimentation with no main uplift/erosion events from Jurassic to Early Miocene (Fig. 1). A brief episode of instability existed at the end of Cenomanian; this influenced the depositional environments, the source rocks formation and the reservoirs in the Dezful Embayment (Bordenave and Hegre, 2005).
Organic geochemistry and mineralogy of the Pabdeh Fm.

3
The Paleocene-Eocene transgression resulted in the deposition of neritic to basinal marls and argillaceous limestone of the Pabdeh Formation in a narrow foredeep basin (Bordenave and Hegre, 2010). This Formation encompasses the entire Zagros Fold Belt, extends widely from Lurestan to Fars, which contains planktonic fauna throughout the basin (Habibi et al., 2017;James and Wynd, 1965). At the base, the Pabdeh Formation overlies disconformably the Gurpi Formation, whereas its upper boundary is transitional with the Asmari Formation (Fig. 1). Throughout the Dezful Embayment, the Pabdeh Formation is divided into three informal units. Organic geochemistry and mineralogy of the Pabdeh Fm.
4 and upper units were deposited under oxic conditions and have low organic contents. However, the middle unit formed under anoxic conditions and has a relatively higher organic content (Bordenave, 2014).

Conventional well logs including Spectral and
Computed Gamma-Ray (SGR and CGR), sonic, neutron and density porosities, and resistivity were used for interval recognitions and petrophysical investigation of the Pabdeh BSU in the studied wells. In this study, about 324m of core samples (GS: 114m; KR: 30m; RR: 92m and RS: 88m) were studied macroscopically prior to sample selection. Eventually, a total of 143 core and 23 cutting samples of the BSU were obtained from 4 wells drilled in the southern Dezful Embayment (Fig. 2).
A subset of 34 core and cutting samples were selected for X-Ray Diffraction (XRD) analysis on both bulk sample and clay fractions. X-ray clay fraction analysis were carried out using Mg-saturated air-dried samples which were subsequently treated with ethylene glycol and heated to 550ºC as suggested by Poppe et al. (2001). The XRD patterns of the samples were acquired using a PANalytical XPert Pro Multi-Purpose XRD System with CuKα radiation, and PIXcel0D detector, which provides high signal to noise ratio and allows high data collection speeds. The datasets were collected from 2 to 65°2θ for bulk samples, and 2-35°2θ for clay samples. The BGMN software (Bergmann et al., 1998) was used to quantify the mineralogical phases in the bulk samples. With the BGMN approach, quantification is based on a Rietveld refinements of simulated patterns of mineral phases. Structure files were available directly from the BGMN setup package and/or the BGMN website (Bergmann et al., 2014) or works by the BGMN Group. The Scanning Electron Microscopy (SEM) observation was performed on these samples as a complementary analysis for the mineralogical characterization. The secondary electron and backscattered electron images were obtained using a Hitachi XL30 Scanning Electron Microscope equipped with an Energy Dispersive X-ray (EDAX) spectrometer to identify the minerals in the samples.
In order to assess the quantity, quality and maturity of the organic matter contained within the Pabdeh BSU, a total of 166 core and cutting samples were analyzed applying the Rock-Eval 6 pyrolysis technique. Following the standard procedure described by Behar et al. (2001), aliquots of pulverized samples (70-80mg) were loaded into crucibles and various parameters (e.g. S 1 , S 2 , S 3 and T max ) were measured. Additional parameters including the total organic carbon (TOC), hydrogen index (HI), oxygen index (OI) and Production Index (PI) were calculated from these measurements.
For petrographic analyses and vitrinite reflectance measurement, a subset of 32 polished mounts were provided from core and cutting chips embedded in epoxy  Organic geochemistry and mineralogy of the Pabdeh Fm. 5 resin. Polished samples were prepared according to standard procedures described by Taylor et al. (1998). Optical microscopy was performed using a Zeiss Axioplan II polarizing microscope, equipped with J&M photomultiplier. Maceral analysis was conducted under incident white light and blue-light excitation (fluorescence mode). Standards (%Ro: 0.43, 0.56 and 0.95) were used to calibrate the photometer. Reflectance was measured on randomly oriented vitrinite macerals under oil immersion with 100× magnification objective at a wavelength of 546nm.
Nineteen core samples were selected for biomarker analysis from the studied shale unit. The indigenous bitumen was extracted from the samples using a Soxhlet apparatus with a mixture of dichloromethane (DCM) and methanol (CH 3 OH) (90:10 v:v). After asphaltene precipitation (by adding excess amounts of n-heptane), extracts were fractionated into saturate, aromatic and polar compounds by liquid column chromatography. The saturate hydrocarbon fractions were analyzed using a gas chromatograph equipped with a 50m DB-1 fused silica capillary column (internal diameter of 0.25mm and film thickness of 0.25µm) and was coupled to a Finnigan MAT GCQ ion trap mass spectrometer. The gas chromatograph was programmed to operate with an oven temperature of 70ºC, ramped to 300ºC at 4ºC/ min, and held under isothermal conditions for 15 minutes. The sample was injected in splitless mode with the injector temperature at 275ºC. Helium was used as a carrier gas and a mass range between m/z 50 to 650 was scanned by the spectrometer. The relative abundance of individual compounds was calculated by measuring peak areas in relation to the internal standard (deuterated n-tetracosane); then various components were recognized following published works (Peters et al., 2005;Waples and Machihara, 1992).

Petrophysics
The BSU is distinguishable throughout the entire Dezful Embayment with respect to its distinctive petrophysical characteristics. This unit is distinguished from the upper and lower parts of the Pabdeh Formation by the relatively higher CGR and SGR response (up to 35 API and 10 to 150 API, respectively) lower sonic velocities (from approximately 3500 to 5500m/s, avg. 4000m/s), higher resistivity (approximately up to 50ohm.m for RR and RS wells and more than 200ohm.m for GS and KR wells), and higher neutron and density porosities ( Fig. 3A-D). A typical CGR and SGR response for the BSU begins with lower readings at the base, progressively increases to higher readings in the middle (with high frequency variations), and eventually decreases to lower readings at the top ( Fig.  3A-D). Simultaneously, an increase in the CGR and SGR response and the separation between them can indicate linear relationship between concentrations of organic matter and uranium enrichment for the BSU (Fig. 3A-D). As shown in the Organic Petrography chapter, common presence of hydrogen-rich organic matter can be linked with a higher productivity and/or preservation of organic matter due to change from oxic to anoxic conditions (i.e. during sea level rise) (Fang et al., 1993).

Lithology, lithofacies and mineralogy
Lithologically, the BSU mainly contains bituminous marls interbedded with argillaceous limestone in the Dezful Embayment. Generally, these marls and limestones have reflective brownish appearance in drilling core and cuttings, which is probably due to the high abundance of organic matter (Fig. 4). In addition, macroscopic observations reveal that the samples mostly comprise dark grey to brown organic-rich argillaceous limestones alternating with light colored thin carbonatic interbeds (Fig. 4). Relatively higher thermal maturities in KR and GS samples have given rise to prominent a few centimeters-scale oil-stains and solid bitumen (Fig. 4C).
Thin-section and SEM studies show that the Pabdeh BSU consists mainly of fine to medium laminated wackestone and wackestone-packstone lithofacies ( Fig.  5A, B, C). Skeletal planktonic foraminifera, phosphatic and glauconitized fragments, quartz and pyrite are the main constituents visible under the optical microscope. Organic matter is also abundant (Fig. 5D). Quartz commonly fills the skeletal fragments ( Fig. 5E, F). Pyrite framboids become abundant in mud-rich layers (Fig. 5G, H).
X-ray Quantitative Phase Analyses (QPA) reveals that the BSU is mainly composed of calcite, clay minerals, and quartz, with minor amount of ankerite, dolomite and pyrite (Table 1; Fig. 3A-D). Calcite is the dominant mineral ranging from 31 to 71wt.%, with an average of 53wt.%. Clay minerals are the second most common mineral constituent including kaolinite (up to 31wt.%, avg. 11wt.%), illite/smectite mixed layer (5 to 24wt.%, avg. 14wt.%) and some chlorite (up to 5.75wt.%). Smectite is lacking most likely due to increased burial depths since unstable smectite transforms into illite in the form of mixed layer illite/smectite at temperatures higher than 60-80ºC (Bjørlykke, 2010;Nadeau, 2011). The proportion of quartz ranges from 3 to 35wt.% (avg. 14wt.%). The pyrite content ranges up to 2wt.%. Feldspars (K-feldspar and plagioclase) are lacking except for two samples from wells RS and RR (2.4 and 0.8wt.%, respectively). Organic geochemistry and mineralogy of the Pabdeh Fm.  Organic geochemistry and mineralogy of the Pabdeh Fm.  Organic geochemistry and mineralogy of the Pabdeh Fm. 8

Rock-Eval pyrolysis
Rock-Eval pyrolysis data is listed in Table I and plotted against depth for all studied wells in Figure 3. The TOC values for the BSU show a wide range of variation in the sampled wells, with relatively higher average values in GS and RR compared to KR and RS wells (avg. 4.17, 2.9, 5.2 and 3.3wt.%, for GS, KR, RR and RS wells, respectively). Rock-Eval S 1 and S 2 yields of the BSU range from 0.1-6.5 and 3-101mg HC/g rock, respectively (Table I). The HI values vary generally between 400 and 650mg HC/g TOC. The middle part of the BSU displays higher TOC and HI values, which is in compliance with the increase in the gamma-ray readings and the abundance of clay mineral ( Fig. 3A-D). The T max values of studied samples from KR and GS wells are relatively higher (≥430ºC) compared to wells RR and RS (<425ºC).

Organic petrography
Microscopic examinations indicate that the BSU contains mainly liptinite (alginite) and Amorphous Organic Matter (AOM) with limited input of terrestrial vitrinite and inertinite macerals in agreement with previous studies (e.g. Opera et al., 2013). Liptinite macerals are morphologically similar to telalginite and lamalginite, which are present in most of the studied shale samples. These alginite materials show bright yellow to yellowish fluorescent under blue-light excitation ( Fig.  6A-C). The vitrinite and inertinite macerals are small in size and have low abundances in the studied samples ( Fig. 6D-F). Some samples at well RR indicate relatively higher concentration of vitrinite particles. This data is also in agreement with the relatively lower HI values calculated for the samples from this well (Table I) and probably indicates relatively higher contribution of terrigenous organic matter in the mentioned location (Hunt, 1996). Petrographic observations by incident white light indicate that some samples have bitumen staining and solid bitumen, particularly in KR well ( Fig. 6G-H). The representation of bitumen staining and solid bitumen indicates kerogen conversion from these shale samples upon reaching maturity corresponding to oil generation window. High content of algal-derived macerals and amorphous organic matter in the organic-rich shales suggest that the BSU contain Type II organic matter deposited under open marine settings. This is consistent with results obtained from Rock-Eval pyrolysis data (Table I). Petrographic inspections indicate that framboidal pyrite crystals are abundant in different sizes in all studied samples (Fig. 5). This observation is also supported by SEM microscope (Fig. 5G, H). The presence of framboidal pyrite could indicate a reducing depositional conditions for the studied wells (Zhao et al., 2014).
Vitrinite reflectance measurements were used to assess the maturity level of the BSU. The results of mean random Vitrinite Reflectance measurements (VRr) for the studied wells are given in Table I. Most of the measurements show a low standard deviation and a unimodal pattern, indicating   the presence of autochthonous vitrinite particles. Vitrinite reflectance show low variation, ranging from 0.3 to 0.53% for wells RR and RS and from 0.5% to 0.67% for wells GS and KR (Table I), indicate a relatively low level of thermal maturity (immature to early stage of oil window). This is in agreement with the thermal maturity trend inferred from Rock-Eval T max values.

Molecular composition
Biomarker parameters have been used effectively for assessing the thermal maturity of crude oil and source rock as well as characterizing the depositional environment and source input of organic matter (Peters et al., 2005). Isoprenoids, n-alkanes and saturate biomarker parameters (sterane and hopane) are presented in Table 2 for samples from the BSU. Total Ion Chromatograms (TIC) for the samples indicate a unimodal distribution of n-alkanes with a clear dominance of short-chain n-alkanes (C 15 to C 20 ) relative to long-chain n-alkanes (Fig. 7). The Carbon Preference Index (CPI) values of the studied samples range from 0.7 to 1.3, indicating slightly even-carbon predominance (CPI<1). The acyclic isoprenoid pristane/phytane ratios vary between 0.5 and 1, with an exception for KR (0.32) and RR (1.3)   Organic geochemistry and mineralogy of the Pabdeh Fm.
12 samples. The ratios of Pr/n-C 17 and Ph/n-C 18 range from 0.4 to 1.9 and from 0.42 to 1.6, respectively (Table 2).
Terpane distributions are characterized by a high content of pentacyclic terpanes compared to tricyclic and tetracyclic terpanes (Fig. 8). The most abundant hopane peak is C 30hopane in all the studied samples. The relative abundance of C 29 to C 30 hopane is generally similar in all samples with values between 0.4 to 0.7. The Ts/(Ts+Tm) ratio for all samples is below 0.44 (Table 2). Oleanane was detected in all the samples at varying concentration (Fig. 8). The oleanane index varies between 10% to 48%, with maximum values for GS and RR samples (Table 2). Gammacerane is also traced, although in low abundance (gammacerane/hopane ratio ranges from 0.04 to 0.26) ( Table 2). Generally, the studied samples show relatively similar profiles of regular sterane distributions (Fig. 9). Diasteranes are present at low quantities in all the analyzed samples (Table 2).

Hydrocarbon generation potential and organic matter type
The amount of hydrocarbons produced during pyrolysis (S 2 ) is an important parameter to evaluate the generation potential of source rocks (Peters and Casa, 1994). Higher S 2 values are in accordance with higher HI values and TOC contents in the studied samples. Plots of the generation potential (S I + S 2 ) versus TOC indicate that the entire sections in all four wells have good to excellent oil generation potential (Fig. 10).
The organic matter type could be characterized by HI values (Hunt, 1996). Values higher than 300mg HC/g TOC could indicate Type I and II kerogen, which are derived from lacustrine or marine source rocks (Dembicki, 2009;Hunt, 1996). Samples of the BSU indicate values between 400 to 650mg HC/g TOC (Table I), consistent with Type II kerogen. Furthermore, the plotting data of S 2 versus TOC could also supports that the BSU predominantly contains Type II organic matter (Fig. 11A). The plot of HI versus T max is used to determine the kerogen type and maturity (Peters and Casa, 1994;Tissot and Welte, 1984). Based on this diagram, samples of the BSU contain Type II organic matter, although a few samples are located in Type II-III kerogen area for well RR (Fig. 11B). The lower HI value for some samples of RR well implicates the contribution of terrigenous organic matter. This conclusion is further supported by the organic petrographic results indicating a relatively higher contribution of vitrinite/intertinite macerals in well RR relative to other wells.

Depositional environment
The distribution of n-alkanes in crude oils and source extracts can provide useful information about the sources  of the organic matter (Peters et al., 2005). Unimodal distribution and predominance of short-chain n-alkanes for all the studied samples (Fig. 7), could imply a marine environment with high contribution of aquatic organic matter (Ghassal et al., 2016;Peters et al., 2005). Moreover, the low abundance of the long-chain n-alkane (n-C 22+ ) and the slight even carbon preference (CPI≤1) indicate the predominance of marine with minor terrigenous organic matter input into depositional environment (Peters et al., 2005). Pristane originates from phytol by oxidation and decarboxylation, and phytane by dehydration and reduction (Peters et al., 2005). Overall, except for one sample from well RR, all samples have pristane/phytane (Pr/Ph) values less than 1, indicating that the organic matter was mainly deposited under anoxic condition (Peters et al., 2005). This is in accordance with the cross plot of the Pr/n-C 17 versus Ph/n-C 18 (Fig. 12), which shows that the studied samples contain Type II organic matter deposited under reducing conditions (Fig. 12).
Oleanane in crude oils and source extracts is an important marker for both source input and geological age. This compound is originated from Cretaceous or younger, higher plants, which is in agreement with the geological age of the studied samples (Peters et al., 2005). The oleanane index for samples from the BSU varies from 10% to 48% (Table 2). The high abundance of oleanane for some samples may be related to periodically high terrigenous input into the environment (Peters et al., 2005). The gammacerane content is an indicator of water salinity and water stratification in a sedimentary environment . The gammacerane/hopane ratio ranges from 0.04 to 0.26, suggesting that the BSU was mainly deposited in a stratified water environment. A relatively high salinity is suggested for the depositional environment by the high pregnane concentrations in the studied extracts (Fig. 9). The C 31 -22R-hopane/C 30 -hopane ratio is generally higher than 0.25 for marine environments, whereas values lower than 0.25 point to lacustrine settings (Peters et al., 2005). For all the samples, the C 31 22R homohopane/C 30 α(H)-hopane ratios is higher than 0.25 (Table 2), indicating that the organic-rich BSU was deposited in a marine environment. Moreover, the C 35 /C 34 homohopane ratios obtained for most of the analyzed samples are above 0.80, suggesting that the organic matter was deposited in a marine environment under anoxic conditions (Mello et al., 1988). Organic geochemistry and mineralogy of the Pabdeh Fm.  Organic geochemistry and mineralogy of the Pabdeh Fm.   l o g i c a A c t a , 1 8 . 1 5 , 1 -2  Organic geochemistry and mineralogy of the Pabdeh Fm.

16
The sterane composition can be used to characterize the depositional environment, organic matter input and the maturation range of potential source rocks (Seifert and Moldowan, 1977). The relative abundance of C 27 -C 29 regular steranes is used to indicate the source of organic matter (Huang and Meinschein, 1979). High relative abundance of C 27 , C 28 and C 29 steranes is due to marine phytoplankton and algae, lacustrine algae and terrigenous higher plants, respectively (Moldowana and Seifert, 1985). A ternary plot of C 27 -C 29 sterane distribution of samples from the BSU shows a slightly higher relative abundance of C 27 and C 29 steranes compared to the C 28 sterane (Fig.  13). This suggests predominantly marine conditions with high algal input and minor contribution of terrigenous higher plants in agreement with the organic petrographic inspections. Diasteranes/steranes ratio is commonly used as an indicator of the redox conditions in the depositional environment (Peters et al., 2005). Samples in this study have low diasterane/sterane ratios (ranging between 0.17 to 0.4), suggesting anoxic clay-poor or carbonate source rock (Peters et al., 2005). Regular steranes/17α-hopanes ratio represents input of eukaryotic (mainly algae and plankton) versus prokaryotic (bacteria) organic matter to the sedimentary environment of source rocks (Peters et al., 2005;Moldowan et al., 1986). Varying sterane/hopane ratios obtained for the studied samples (ranges from 0.4 to 3.18) may indicate periodic variations in marine organic matter input (from planktonic and algae to bacteria). Conclusions from biomarker parameters are consistent with organic petrographic observations stated earlier (i.e. significant amount of liptinite and amorphous organic matter, high abundance of pyrite framboids, phosphatic and glauconite grains). These are in agreement with reducing marine settings during deposition of the Pabdeh BSU.

Thermal maturity
Thermal maturity of the organic matter contained within the BSU has been assessed using vitrinite reflectance (%Ro), pyrolysis T max and Production Index (PI) values, and saturate biomarkers (Tables I; 2). The T max values of studied samples indicate that the organic matter at wells RR and RS is immature (i.e. mostly below 425ºC). In contrast, samples from wells KR and GS have higher T max values (i.e. ranging between 425 to 435ºC), which suggest marginally-mature to mature state (Fig. 11B). The pattern of variation in PI data (ranging from 0.011 to 0.3, Table I) are in agreement with the T max values. Thermal maturity trends are also assessed by vitrinite reflectance measurements being considered the most reliable maturity indicator for sedimentary organic matter (Dow, 1977). A thermally immature to early mature state is suggested for the BSU at wells RR and RS (RO values ranging from 0.42 to 0.5%). In contrast, reflectance values range from 0.5 to 0.67% for wells KR and GS ( Table  I). The same results are suggested by vitrinite reflectance estimated from T max (Jarvie et al., 2001), which shows values of 0.33-0.65%Ro consistent with immature to early mature stages. However, one should bear in mind that the true maturity of the studied interval may be slightly higher since both maturity parameters might be suppressed due to the presence of a significant amount of hydrogen-rich liptinite macerals and bitumen impregnation (Hackley and Cardott, 2016;Hutton and Cook, 1980). Several biomarker maturity indicators have been applied to evaluate thermal maturity of the BSU. Generally, the relatively higher abundance of n-alkanes in the biomarker region of obtained gas chromatograms is consistent with lower level of thermal maturation for RR and RS wells (Fig.  7). The C 32 homohopane 22S/(22S+22R) rises up to 0.6, while 0.57 to 0.62 is the equilibrium range during maturation corresponding to the onset of oil generation (Mackenzie et al., 1980;Seifert and Moldowan, 1980). The C 32 homohopane 22S/(22S+22R) ratio of the studied samples from RR and RS wells are low, varying from 0.28 to 0.48 and from 0.43 to 0.55, respectively. Thus, these samples are immature at these locations. In contrast, the studied samples from KR and GS wells have relatively higher C 32 homohopane 22S/(22S+22R) ratios, suggesting that the samples are immature at these locations. In contrast, the BSU samples from KR and GS wells have relatively higher C 32 homohopane 22S/(22S+22R) ratios, suggesting that the organic matter from these wells are thermally mature and have reached the equilibrium stage ( Table 2). The C 29 20S/(20S + 20R) and ββ/(αα + ββ) sterane ratios are particularly effective for thermal maturity of source rocks at the beginning of the oil window (Seifert and Moldowan, 1986). The ratio of C 29 20S/(20S + 20R) rises from 0 to 0.5 (equilibrium= 0.52-0.55) with maturity, while the ratio of ββ/(αα + ββ) increases from near-zero to 0.7 (equilibrium= 0.67-0.7) (Peters et al., 2005). Based on these thresholds, the majority of samples at wells RS and RR do not reach the equilibrium values. Nevertheless, the studied samples from KR and GS wells have near equilibrium values for these parameters and can be considered thermally mature (Fig. 14).
The Ts/(Ts+Tm) ratio should be used with caution because Ts and Tm could be influenced by factors other than maturity such as lithology and organic matter type (Moldowan et al., 1985). Carbonate source rocks have unusually low Ts/(Ts+Tm) ratios compared with shales due to catalyzing effects of clay minerals (Peters et al., 2005). The relatively lower maturity of the BSU and the uniform nature of Ts/(Ts+Tm) values in the studied samples suggest that the Ts/Tm ratio is more influenced by the lithofacies (i.e. carbonate source rock) than the maturity. Moretanes are thermally less stable than hopanes; therefore, the C 30 moretane/C 30 hopane ratio decreases with increasing thermal maturity, from approximately 0.8 in immature sediments to less than 0.15 (minimum 0.05) in mature organic matter (Peters et al., 2005;Mackenzie et al., 1980). The RR and RS samples have higher C 30 moretane/C 30 hopane ratio (mean= 0.25) consistent with the immaturity of these samples. On the other hand, the C 30 moretane/C 30 hopane ratio of studied samples from KR and GS wells have values lower than 0.16, suggesting relatively higher level of maturity.
Maturity interpretations derived from biomarker parameters are also supported by T max and vitrinite reflectance values, consistent with a lower level of maturity at wells RR and RS compared to wells KR and GS. These results are well in agreement with previous modeling studies in the studied area, which indicated the Pabdeh Formation has reached the early stage of thermal maturity (e.g. Alizadeh et al., 2012;Bordenave and Hegre, 2010;Opera et al., 2013). The relatively higher maturity of the BSU at wells KR and GS relative to wells RR and RS seems to be at odds with a shallower present-day burial depth at the two former wells (Fig. 3A-D; Table I). This may be a consequence of the recent uplift at KR and GS localities due to the Zagros orogeny. In contrast, the RR and RS localities have experienced gentle folding at the southwestern margin of the Zagros deformation front. This data is also in agreement with the regional geology of the studied area where the structures become progressively younger from NE to SW (Hessami et al., 2001).

Shale oil potential
Geologically, the fundamental parameters to form an economically recoverable shale oil reserve, are the organic  11. A) TOC vs. S 2 and B) HI vs. T max diagrams, indicating Type II kerogen for the Pabdeh BSU in the studied wells (after Langford and Blanc-Valleron, 1990).

FIGURE 12.
Cross-plot of Pr/n-C 17 versus Ph/n-C 18 showing Type II kerogen for the Pabdeh BSU, deposited under reducing conditions (after Shanmugam, 1985).
The relatively thick (110m, in average) organic-rich mudstone of the BSU can represent a considerable oil potential by showing high TOC contents (ranging mainly between 1-9wt.%, with an average of 3.83wt.%) and organic matter of Type II kerogen (HI ranging mainly between 450-650mg HC/g TOC). Variations in TOC values mimic the vertical trend in natural gamma-ray response in the BSU and the maximum organic richness corresponds with the middle part. This part has a thickness of about 55m in the four studied wells and the average organic content amounts to 5.2wt.%.
A geochemical indicator for evaluation of potentially producible oil from shale plays is provided by the oil crossover effect, which is defined as the Oil Saturation Index (OSI= S 1 /TOC×100) (Jarvie, 2012). Samples with OSI higher than about 100mg HC/g TOC indicate a considerable potential (Jarvie, 2012). Except for a few samples from wells KR and RS, the OSI values of the BSU are mostly below the crossover point (OSI= 100) (Table I; Fig. 15). However, most of KR and GS samples are plotted in an oil show range with OSI values around 70mg HC/g TOC (Fig. 15). Nevertheless, the observed oil contents for studied samples can be influenced to some extents by the evaporative loss of light hydrocarbons during sample storage or preparation (Jarvie, 2012).  Resistivity logs were also used to understand the amount of hydrocarbon saturation in the BSU. Variations in pore network and pore fluid are the major causes for changes in resistivity (Lu et al., 2015). Most detrital minerals (quartz, feldspar, and carbonates) are not conductive and form nonconductive rock frameworks. In contrast, presence of clay minerals could reduce the resistivity, and hence decrease the apparent hydrocarbon saturation (Lu et al., 2015 and references therein). Nevertheless, mineralogy is not likely to have a major control on resistivity logs in the BSU due to its predominantly carbonaceous composition (Table 1). An increase in the resistivity is also expected when the source rock is enough mature, and a part of its organic matter seem to be transformed into hydrocarbons (Kalani et al., 2015;Kinley et al., 2008;Lu et al., 2015;Passey, 1990). This phenomenon explains the very high resistivity at wells KR and GS compared to wells RR and RS in the studied area ( Fig. 3A-D), which is in agreement with maturity information discussed above. In other words, the increase in resistivity at wells KR and GS compared to two other wells can be the result of higher maturity and hydrocarbon generation in the former well locations.
High silica/carbonate and low clay contents lead to shale brittleness (Bowker, 2007). Brittle minerals not only favor formation of tectonic fractures, but also control the fracability of shales (Gottardi and Mason, 2018). The mineralogical composition of the Pabdeh BSU shows that the prevailing brittle phases are typically calcite (53wt.% in average). The very brittle quartz mineral is about 14wt.%, in average, whereas the content of ductile clay minerals varies significantly (26wt.% in average). Generally, results show that the BSU could be brittle except for one sample in well RR (Table 1). Thus, the high calcite and quartz contents as well as presence of natural fractures in the Pabdeh Formation (Khoshbakht et al., 2009), could suggest that the BSU is a potential candidate for unconventional shale oil exploitation applying hydrofracturing.

CONCLUSIONS
This study provides new data and interpretations in organic geochemistry, organic petrography and mineralogy of the BSU within the Pabdeh Formation in the southern Dezful Embayment. It also discusses shale oil potential of the BSU in the study area. The BSU is mainly composed of calcite, clay minerals (kaolinite and illite/smectite) and quartz. Minor amounts of ankerite, dolomite and pyrite are also contained.
Most of the studied samples contain TOC between 1 to 9wt.%, HI between 400-650mg HC/g TOC, and S 2 between 5 to 40mg HC/g rock, indicating that the BSU possess excellent to very good potential for hydrocarbon generation. Organic petrography reveals that organic matter is dominated by AOM and liptinite (alginite), with minor amounts of vitrinite and inertinite, typical of Type II kerogen. Thermal maturity parameters including T max , vitrinite reflectance and biomarker maturity ratios indicate that the BSU is immature at wells RR and RS attains earlymature to mature stages at wells KR and GS.
Based on bulk geochemical data, acyclic isoprenoids and saturate biomarkers, minimal variation in depositional conditions occurred during deposition of the BSU. The studied shale samples were deposited under anoxic marine conditions that received algal/bacterial organic matter. The presence of oleanane in association with sterane/hopane ratios also suggest that some terrigenous inputs took place during deposition of the BSU.
In summary, considering the organic geochemical and mineralogical information presented in this study we can conclude that the BSU can be a good prospect for shale oil exploitation within the Pabdeh Formation in the southern Dezful Embayment. Therefore, beside its potential as a conventional source rock, some parts of the Pabdeh Formation may constitute unconventional shale oil potential which can influence future exploration/production activities in the Zagros Fold Belt of Iran.

ACKNOWLEDGMENTS
The authors gratefully acknowledge the Petroleum Geology and Geochemistry Research Center of Shahid Chamran University of Ahvaz (PGGRC) for laboratory facilities. The    o l o g i c a A c t a , 1 8 . 1 5 , 1 -2  Organic geochemistry and mineralogy of the Pabdeh Fm.   l o g i c a A c t a , 1 8 . 1 5 , 1 -2  Organic geochemistry and mineralogy of the Pabdeh Fm.   Organic geochemistry and mineralogy of the Pabdeh Fm.