Introduction

The Khami and Bangestan Groups are the main Cretaceous oil producers in southwestern Iran. Among the Cretaceous succession, the Kazhdumi, Gadvan and Garau formations have source rock characteristics and contributed to the generation of most of the oil and associated gas reserves already discovered. The Albian Kazhdumi Formation has been investigated in a number of previous studies (Bordenave and Burwood 1990; Bordenave and Huc 1995; Rabbani and Bagheri Tirtashi 2010; Alizadeh et al. 2012; Mashhadi et al. 2014; Mashhadi and Rabbani 2015). Early Cretaceous source intervals in southern Iran comprise the Gadvan and Garau formations, which are the main subject of this study. There are some previous studies on the Gadvan and Garau formations (or their time equivalent) in the Zagros Basin (Abeed et al. 2011; Rabbani et al. 2013), but all of them have been done outside of the studied area.

The purpose of present study is to investigate the potential source rock characteristics of the Early Cretaceous succession in the oilfields of southwestern Iran. In order to characterize the quantity, quality, and maturity of organic matter in the studied intervals, Rock–Eval pyrolysis and vitrinite reflectance were conducted, and additional data, such as GC, GC–MS and isotope analysis, were used to interpret the depositional environment and oil–source correlation.

Geological setting

The available data suggest that the central Iranian block was part of the Arabian Platform and belonged to the Gondwana supercontinent during the Precambrian and early Paleozoic (Stocklin 1974). The breakup between the Arabian platform and the Central Iranian block at the end of the Triassic caused the formation of the Neo-Tethys Ocean, which expanded during the Jurassic and the Early Cretaceous as a result of the seafloor spreading (Stocklin 1974). The northeast margin of the Arabian platform was passive, with the seafloor spreading during the end of the Triassic to end of the Middle Cretaceous period. During the same period, deep marine sediments, i.e., radiolarite or micritic limestone, were deposited at the northeast edge of the Arabian platform (Muris 1980). After the collision of the Arabian and Eurasian plates (including the Iranian block), the Neo-Tethys Ocean shrank as a result of a complex succession of subductions (Sepehr and Cosgrove 2004). This tectonic change caused the area to become active, and compressional tectonic events occurred from the Cenomanian.

The Dezful Embayment, which belongs geologically to the northern part of the Arabian plate is a tectonic feature at the southwestern edge of the Zagros Basin (Fig. 1) and contains Iran’s major oil fields. The Dezful Embayment is bounded by three main tectonic elements, namely the E–W trending flexure zone (the Balaroud fault) and a “mountain front” flexure with a NW–SE trend in the northern part. To the east and southeast, the Dezful Embayment is bounded by a complex zone of faults and flexures having a N–S trend (Kazerun-Qatar fault) (Falcon 1967; Motiei 1995; Alavi 2007).

Fig. 1
figure 1

Modified after Abdollahie Fard et al. (2006), Bordenave and Hegre (2010), Abeed et al. (2011), Zeinalzadeh et al. (2015)

Location map of the studied area and oil fields in the Zagros Basin with main tectonic boundaries. Studied samples are indicated by different colors.

The Abadan Plain is a structural unit limited to the NE by the Susangerd–Ab-e Teymur–Mansuri Zagros trend, to the south by the Persian Gulf, and to the west by the Iran–Iraq border (Fig. 1). The generally N–S trending pre-Zagros anticlines of the Abadan Plain are not exposed in outcrop, but occur in the subsurface. The Abadan Plain was part of the Mesopotamian Depression during the Jurassic and Early Cretaceous (Jassim and Goff 2006; Abeed et al. 2011; Aqrawi et al. 2010).

A global sea-level rise followed the end of Jurassic regression during the Valanginian. As the sea-level rose, shallow-water limestone was widely deposited on the platform area during the Early Cretaceous, i.e., southwest Lurestan, southern Iraq, the Abadan Plain, and the wide Fars area. The Early Cretaceous morphology of the sedimentary basin was formed after the deposition of the Upper Jurassic evaporites in the Gotnia Formation.

Subsequent to the deposition of these evaporites, the southern part of Khuzestan was covered by a deeper water depression that extended in the Lurestan. In this depression, sedimentation of fine carbonate mud, together with organic matter, prevailed (Fig. 2). Euxinic conditions favorable for the preservation of these organic-rich sediments were also present, making way for the development of a suitable source rock (James and Wynd 1965).

Fig. 2
figure 2

After Bordenave and Hegre (2010)

Stratigraphic chart of the studied area in southwest Iran. The main sources are indicated by two stars and the marginal ones by one star.

The Garau Formation consists of five units in the Abadan plain area. The bottom unit consists of dark gray to black carbonaceous shale and dark argillaceous pyritic limestone (James and Wynd 1965). Overlying this is dark gray limestone with carbonaceous shale. Above this is gray to brown shale, and above that is limestone with black nodules. The upper part of the formation is alternating gray shale with shaley limestone, and glauconitic sandy limestone and sandy marls in the uppermost part. Almost all of the Garau Formation is considered to have been deposited under a low-energy anoxic environment, and it has excellent source rock potential, mostly in the lower part. The age of the Garau Formation ranges from Valanginian to Coniacian (Fig. 2). Limestone sequences of the Sulaiy and Yamama formations in Saudi Arabia/Iraq, the Minagish Formation in Kuwait are the time equivalents of the Garau Formation (James and Wynd 1965; Aqrawi et al. 2010).

The Gadvan Formation (Fig. 2) was deposited during the Barremian in generally shallow oxic environment (James and Wynd 1965). However, the environment remained dis-oxic in the Lurestan and S–W Dezful Embayment (Bordenave and Huc 1995). In the Abadan Plain and in most of the Dezful Embayment, the Gadvan Formation is dominated by brownish-yellow marl and argillaceous limestone (Kent et al. 1951; Fakour et al. 2010).

Method and materials

During this study, 140 core and cutting samples from Early Cretaceous sediments were obtained from 24 wells in 12 oil fields from northern Dezful Embayment and Abadan Plain (locations are shown in Fig. 1). The samples were selected from the Gadvan (Barremian) and Garau (Early Cretaceous) formations using gamma ray logs and binocular microscopes to avoid caving and contamination (Bordenave and Huc 1995).

Rock–Eval pyrolysis and vitrinite reflectance

The first step in the geochemical study of petroleum systems in an area is to characterize the quantity, quality, and maturity of the source rocks. The most widely used method to identify the generation potential of source rock is Rock–Eval pyrolysis (Espitalié et al. 1977). After some treatment to remove synthetic contamination such as mica, iron junk, nut shell, and drilling mud additives, samples were ground and homogenized.

Rock–Eval apparatus yields the following information for rock samples: Total organic carbon (TOC), S1 hydrocarbon already present in the sample, S2 remaining potential, S3 oxygen content, and T max related to the maturity (Espitalié et al. 1977; Tissot and Welte 1984; Espitalié and Bordenave 1993; Peters and Cassa 1994). Table 1 shows the Rock–Eval measures and calculation parameters for studied samples.

Table 1 Rock–Eval pyrolysis result of the studied samples

Twenty-eight samples were selected for measuring vitrinite reflectance according to the Rock–Eval pyrolysis results. Reflectance was measured using Zeiss Axioplan II microscope in oil immersion following the method described by Taylor et al. (1998) and Alizadeh et al. (2012). The results are exhibited in Table 2.

Table 2 Vitrinite reflectance measurement results of studied samples

Complementary analysis

According to the Rock–Eval pyrolysis results, 25 rock samples were selected for extraction, liquid and gas chromatography experiments. Gas chromatography mass spectrometry (GC–MS) and isotopic analyses were performed on 14 and 11 of the selected samples, respectively. Compound-specific isotope measurement (GC-IRMS) was carried out on four samples from both the Gadvan and Garau formations.

Ground rock samples were extracted by Soxhlet using dichloromethane (DCM) for 24 h; the solvent was removed by centrifuge evaporator. To fractionate the extract, a liquid column of silica gel activated by alumina was employed. Saturate, aromatic, and polar compounds were fractionated by n-hexane, benzene, and chloroform, respectively.

Gas chromatography (GC) analysis was performed by an instrument equipped with a 30 m DB-1 fused silica capillary column (i.d. 0.25 mm; 0.25 μm film thickness). For mass spectrometry analysis, the GC was coupled to a Finingan MAT GCQ ion trap mass spectrometer. The oven temperature was programmed from 70 to 300 °C at a rate of 4 °C/min, in which 15 min was considered as the isothermal period. The injection temperature was 275 °C in split-less mode, and the carrier gas was helium. For calculating GC and GC–MS, a biomarker ratio peak area for each compound was considered.

For GC-IRMS analysis, in which carbon isotopic ratio is measured for specific saturated hydrocarbon, a trace GC was coupled to a Delta V isotope ratio mass spectrometer via a combustion interface, and a ComFlow IV unit, all ThermoFisher. At the beginning and end of each analysis, CO2 was injected for calibration. Isotopic ratios are reported with respect to the Vienna Pee Dee Belemnite (VPDB) standard.

Result and discussion

Rock–Eval and vitrinite reflectance

The total organic carbon content of the Gadvan Formation varies from 0.25 to 1.6 wt%. The average TOC in this formation is 0.65 wt% (Table 1) which is considered as poor potential source rock (Hunt 1996). Cutting samples from Azadegan show significant pollution from migrating hydrocarbons, as shown by their S1 values which are higher than their S2 values, for a maturity lower than 430 °C T max. A similar conclusion can be made for the Darquain 4205 m sample. Therefore, the results of Rock–Eval analyses of these polluted samples are not discussed further.

In the Garau Formation, present time TOC ranges from 0.27 up to 5.72%, with an average of 1.76% (Table 1). This value is consistent with good to very good source potential (Peters and Cassa 1994). According to paleogeographic maps, the depocenter of the Garau Formation is located in the Dezful Embayment (east of the Ahwaz oilfield) and in the Lurestan Basin (Bordenave and Hegre 2010).

The S2 peak in Rock–Eval indicates the current ability of the sample to produce hydrocarbon. In the studied samples, the present time S2 values of the Gadvan and Garau formations range 0.13–3.75 and 0.97–9.33 mgHC/g Rock, respectively. The average values of the present time hydrogen index (HI) are 122 and 213 mgHC/g TOC in the Gadvan and Garau formations, respectively. Regarding the plot in Fig. 3, it can be concluded that the Gadvan is a marginal source rock. In contrast, it is noteworthy that Garau samples are already close to the peak of oil expulsion window (T max in the 440–450 °C range). Therefore, initial values of both S2 and HI of Garau samples could have been much higher, even double the current values. Then the Garau Formation could be classified as very good to excellent source rock.

Fig. 3
figure 3

Plot of S1 + S2 versus TOC, showing production potential of the samples

The hydrogen index (HI) is plotted versus the oxygen index (OI) and T max values in Fig. 4 in order to identify the kerogen type of the studied samples. According to these diagrams, the kerogen of the Gadvan Formation is Type III, and in contrast, the Garau Formation contains kerogen belonging to Type II. Abeed et al. (2011), using visual kerogen studies together with the elemental analysis, reported kerogen Type II-S for the Sulaiy Formation (time equivalent of Garau Formation).

Fig. 4
figure 4

Plot of HI versus T max (left) and OI (right), showing maturity and kerogen type of studied samples

Rock sample maturities were determined on the basis of T max, PI, and vitrinite reflectance. In the Gadvan Formation, T max values range between 430 and 441 °C, with an average of 433 °C. Vitrinite reflectance was measured for 17 samples and yielded approximately 0.67%. According to plot of PI versus T max in Fig. 5, as well as vitrinite reflectance measurements, the maturity of the organic matter in the Gadvan Formation can be regarded as being, at the most, at the beginning stage of oil generation. In the studied area, toward the northeast and the Dezful Embayment, the T max values and maturity of the Gadvan Formation increase from 430 °C in the Abadan Plain to more than 440 °C in the northern Dezful Embayment, which is consistent with the increasing depth (Fig. 6). It should be considered that maturity measurements have been done on the samples from exploration wells which are drilled at the tops (apexes) of anticlines, so in the neighboring synclines the maturity of this formation will be higher.

Fig. 5
figure 5

Plot of PI versus T max, showing maturity of the studied samples

Fig. 6
figure 6

After Bordenave and Huc (1995)

Depositional environment and average TOC contour map of the Gadvan (left) and Garau (right) formations. Average T max in the well location are indicated for each field.

The Garau Formation samples are characterized by T max values that range up to 453 °C. Average vitrinite reflectance was measured at approximately 0.82%. The plot in Fig. 5 and vitrinite reflectance measurements in Table 2 indicate organic matter maturities equal to peak oil generation in the Garau Formation, as is obvious in the T max values in Fig. 6. The maturity of the Garau Formation increases toward the east of the northern part of the studied area which is consistent with the reported maturity values of its time equivalent Sulaiy Formation in southern Iraq (Abeed et al. 2011). Again it should be mentioned that the maturity of Garau Formation will be higher in the synclines and kitchen areas, because measurements are derived from samples from wells on the top of anticlines.

Most of the studied samples of the Gadvan Formation are located in the Abadan Plain. However, as shown in Fig. 6, samples located toward the Dezful Embayment show a slight increase in their TOC. Abeed et al. (2011) reported that the TOC content of the Zubair Formation (time equivalent of Gadvan) varies between 0.4 and 1%. Hence, regionally, the organic matter content of the Gadvan Formation decreases from the Dezful Embayment in Iran to the Mesopotamian Basin in Iraq.

The present time TOC content of the Garau Formation increases eastward, as indicated in the average TOC contour map in Fig. 6. In the Mesopotamian Basin west of the studied area, the TOC of the Sulaiy Formation (time equivalent of Garau) has been reported at up to 10%, which means an increasing TOC westward (Abeed et al. 2011).

Gas chromatography

Two typical gas chromatograms for the Gadvan and Garau formations are shown in Fig. 7, and corresponding concentration ratios can be found in Table 3.

Fig. 7
figure 7

GC-SAT chromatograms for the selected studied sample of the Gadvan and Garau formations indicate n-alkanes as well as Pr and Ph. GC profile of the Gadvan Formation shows a binary envelope while in the Garau Formation, an unimodal profile can be seen

Table 3 Result of liquid chromatography and GC-FID for Gadvan and Garau formations

The concentration of long and short chain n-alkane is significant in the Gadvan Formation, as shown in Fig. 7 (Peters and Moldowan 1993). The carbon preference index (CPI) for long chain n-alkanes and Pristane/Phytane ratios is more than 1; Pr/nC17 and Ph/nC18 are less than 1 (Table 3; Fig. 8). Regarding these parameters, the high land plant contribution, low maturity, absences of biodegradation, and anoxic to sub-oxic depositional environment for organic matter in the Gadvan Formation can be inferred (Tissot and Welte 1984; Moldowan et al. 1985).

Fig. 8
figure 8

Plot of Pr/nC17 versus Ph/nC18 from GC-FID, showing maturation, depositional environment, and kerogen type of organic matter

In the Garau Formation, the unimodal GC chromatogram values as well as a high concentration of short chain n-alkanes show an advanced level of maturity. CPI values are close to 1, the Pristane concentration is much lower than the Phytane, and, as Fig. 8 shows, the Pr/nC17 and Ph/nC18 are less than 1. According to the GC parameters, high maturity, lack of biodegradation, and Type II organic matter, a strongly anoxic depositional marine environment for the Garau Formation can be inferred (Tissot and Welte 1984; Moldowan et al. 1985; Peters and Moldowan 1993).

Biomarker study

Because biomarkers are derived from biological precursor molecules in specific organisms, and because each of these organisms lives only under certain environmental conditions, it is logical to use biomarkers as indicators of those conditions (Bjorøy et al. 1992). However, any application of biomarkers for organic-facies interpretation should be made with a full awareness that organic facies (and hence biomarker distributions) can change rapidly, both vertically and laterally (Moldowan et al. 1986; Bjorøy et al. 1992). If one is working with a typical sample of cuttings material, the biomarker signature will be an average for all facies present in the sample, and thus, it will be much harder to interpret (Tables 4, 5).

Table 4 GC–MS fragmentograms peaks identification, peaks are shown in Fig. 9
Table 5 Result isotope analysis in the studied samples

In this study, GC–MS analysis was done on saturates and aromatics fractions in 14 samples in both the Garau and Gadvan formations. Biomarker parameters for the studied samples are shown in Fig. 9 and are summarized in Table 6. Terpanes, steranes, and aromatic parameters were used to investigate the geochemical characteristics of the formations under consideration.

Fig. 9
figure 9

Typical saturate GC–MS fragmentograms of the selected studied samples. Labeled peaks are identified in Table 4

Table 6 Result of GC–MS analysis for Gadvan and Garau formations

First, the biomarkers related to the maturity of the samples will be examined. The value of C31 homohopane 22S/(22S + 22R) increases with maturity, from 0 to 0.61 (Peters et al. 2005). This ratio for the Gadvan and Garau formations ranges between 0.48–0.6 and 0.54–0.65, suggesting an early oil window and peak oil generation, respectively (Table 6). The T s/(T s + T m) ratio in the Gadvan Formation changes from 0.14 to 0.36, while in the Garau Formation this ratio ranges between 0.36 and 0.67, which shows again a low level of maturity in the Gadvan Formation and high level of maturity in the Garau (Peters and Moldowan 1993). In addition to maturity, the lithology of the studied samples can influence the T s/(T s + T m) ratio; for the same maturity, the argillaceous content of the source rocks generally increases the T s/(T s + T m) ratio (Peters et al. 2005). As most of the Garau samples are richer in carbonate, the high T s/(T s + T m) ratios correlate to higher thermal maturity.

The C29 ββ/(ββ + αα) sterane ratio increases from zero to nearly 0.7 (0.61–0.71 equilibrium) (Peters and Moldowan 1993). In the studied samples of the Gadvan and Garau formations, this ratio ranges between 0.3–0.53 and 0.46–0.63, which implies early and peak oil generation window, respectively (Table 6; Fig. 9). The average ratio of 5α(H),14α(H),17α(H) C29 and steranes 20S/(20S + 20R) is 0.4 and 0.45 in the Gadvan and Garau formations, respectively.

Aromatic biomarkers are more resistive to temperature than are saturate ones (Peters et al. 2005). In this study, MPI-1 was measured for the rock samples (Table 6). Peters et al. (2005) proposed a formula for calculating the equivalent vitrinite reflectance of a special sample using MPI-1 (Fig. 10). According to Table 6 and Peter’s formula, the average calculated vitrinite reflectance in the Gadvan and Garau formations are 0.75 and 0.84%, respectively.

Fig. 10
figure 10

Plot of MPI-1 versus calculated vitrinite reflectance in the studied samples (Peters et al. 2005) showing maturity of organic matter in the studied formations

Regarding biomarkers more specifically related to the environment of deposition, Table 6 shows the relative proportion of the C27–C28 and C29 regular steranes. Higher C27 values for the Garau samples indicate a marine depositional environment. By contrast, higher C28 values for the Gadvan Formation mean the input of terrestrial organic matter (Hunt 1996; Mackenzie et al. 1983). Total/NIOC (2003) in a joint study showed that, with the increasing age of source rock, the amount of C28 regular steranes in the samples decreases. They introduced a table to use the amount of C28 regular steranes as indicators of the age of an specific oil in the Persian Gulf. In this study, the C28 regular sterane content of both the Gadvan and the Garau has the similar average values, 27.33 and 27.25%, respectively, which can be used to further source–oil correlation.

A cross-plot of DBT/Phen versus Pr/Ph can be used to detect the depositional environment of organic matter (Hughes et al. 1995). As indicated in Fig. 11, the Gadvan Formation is located in a mixed shale/carbonate environment, and the Garau Formation is located in a marine carbonate depositional environment.

Fig. 11
figure 11

Plot of DBP/Phen versus Pr/Ph, showing a depositional environment of organic matter in the studied samples

Abeed et al. (2011), using visual kerogen interpretation and biomarker data, reported kerogen Type II-S for the Sulaiy Formation (time equivalent of the Garau Formation), which is in agreement with the high concentration of dibenzothiophene (strongly anoxic marine environment) compound in the Garau Formation.

Isotope study

In this study, bulk composition stable carbon isotope analysis was applied to determine the depositional environment of the Gadvan and Garau formations (Table 5). Moreover, individual n-alkanes GC-IRMS were measured to differentiate the source rock for future oil–source correlation. As Fig. 12 shows with a plot of δ13C aromatic versus δ13C saturates (Sofer 1984), the Gadvan samples originated in a marine environment with terrestrial influences, while the Garau samples derived from a marine environment.

Fig. 12
figure 12

Plot of δ13C aromatic versus saturate values to investigate organic matter depositional environment (Sofer 1984)

Regarding Table 5, the canonical variable (CV) confirms the result because, for all the Garau Formation samples, the CV is less than 0.47, while of the 8 analyzed samples from the Gadvan Formation, 5 have CV values more than 0.47 (Sofer 1984).

GC-IRMS analysis, in which carbon isotopic ratios were measured on specific saturated hydrocarbons, is shown in Table 7. As illustrated in Fig. 13, two different n-alkane isotope profiles for the Gadvan and Garau formations emerged, in which the Garau Formation profile is more negative and δ13C for individual n-alkanes is systematically lower than in the Gadvan samples. This obvious difference in the isotope value was not seen in bulk isotope data in Table 5 and Fig. 12. Hayes (1993) stated that more negative isotope values indicate more planktonic algae.

Table 7 Result of GC-MS analysis for Gadvan and Garau formations (see Fig. 13 for n-alkane isotope profile)
Fig. 13
figure 13

Plot of n-alkanes GC-IRMS profile for four samples in the Gadvan and Garau formations; two different profiles are obvious. Also, Garau profiles are lighter in n-alkane isotope value

Conclusion

Results of Rock–Eval analysis indicate that the Gadvan Formation, with an average TOC of 0.65% and HI values of 122 mgHC/g TOC, can be classified as marginal source rock; by contrast, the Garau Formation, with TOC and HI current values up to 5.72% and 413 mgHC/g TOC, respectively, should be considered as having good to excellent source potential. Moreover, it should be recalled that the initial TOC and HI values of the Garau samples were both significantly higher than the current values. The kerogen type in the Gadvan Formation is mainly Type III with some input of Type II, and, as Fig. 4 shows, the kerogen is of Type II in the Garau Formation.

Gadvan samples are situated between 3510 and 4525 m depth. These samples are immature or at the beginning of the oil window, but have not yet reached the expulsion window (average T max = 433 °C, average VR0 = 0.67%). It is understood that in the synclinal part of the Abadan plain, the Gadvan may have reached the expulsion window. However, as the Gadvan Formation contains a limited amount of Type III kerogen, the contribution of the Gadvan on the charge of Abadan Plain hydrocarbon reserves is deemed negligible. By contrast, Garau samples came from a current burial ranging from 4250 to 5239 m (with the exception of Haftkel oil field, where more than 1000 m of Aghajari layers were eroded prior to the Zagros Folding) and are already in the peak of oil expulsion, or already at the beginning of the gas and condensate stage (average T max = 441 °C, average VR0 = 0.82%). Again, maturity is higher in the deeper part of the Abadan Plain. At Haftkel, Garau is shallower, 3300–4250 m, and no vitrinite reflectance were measured. However, the average T max of 441 °C confirms the importance of the erosion of the Aghajari Formation.

GC parameters for the Gadvan Formation indicate high land plant contribution, low maturity, and an oxic to sub-oxic depositional environment of organic matter. In the Garau Formation, an advanced level of thermal maturity, Type II (likely II-S) organic matter, and a marine depositional environment under anoxic condition have been recognized.

Saturate and aromatic biomarker parameters, as well as isotopic analysis, confirm the Rock–Eval and GC results. In the Gadvan Formation, a maturity level of early oil generation (MPI-1 suggests equivalent VR0 ≈ 0.75) and mixed shale/carbonate with terrigenous input depositional environment (CV > 0.47) has been established, while, for the Garau Formation, a high level of maturity (MPI-1 suggests equivalent VR0 ≈ 0.84) and marine carbonate depositional environment (CV < 0.47) has been determined.

GC-IRMS analysis results show two different n-alkane isotope profiles where the Garau Formation profile is more negative, which shows more planktonic algae and higher maturity. These n-alkane isotope profiles can be used for further oil–source correlation in ongoing studies.

The excellent, mature Garau source rocks in the Abadan Plain area have generated huge amounts of oil, and possibly gas and condensate, which were ready to be trapped in the large N–S oriented anticlines, such as Azadegan, Hosseinieh, Darquain and Mahshahr, which were formed well before the Zagros Folding.