Origin and Characteristics of the Crude Oils and Condensates in the Callovian-Oxfordian Carbonate Reservoirs of the Amu Darya Right Bank Block, Turkmenistan

The Amu Darya Right Bank Block (ADRBB) covers approximately $1.86×104km2$ and is located northeast of the Amu Darya basin, a large hydrocarbon-rich depositional basin on the southeastern Turan platform (Figures 1 and 2). The Callovian-Oxfordian carbonate reservoirs in the ADRBB are the major reservoirs for natural gas and condensate accumulation [1]. However, since 2020, the Wes-2 well, located in the middle section of the ADRBB, has begun producing crude oils [2]. The appearance of crude oils indicates that the formation process of hydrocarbon reservoirs in the ADRBB is different, and further geochemical investigations are needed to confirm the parent materials of crude oil, natural gas, and condensate. Predecessors have done a lot of studies on the geological settings of hydrocarbon reservoirs in the ADRBB [3–6]; only a small number of geochemical studies have been reported, and these geochemical data are mainly used to describe the petrological characteristics of the reservoirs [7–9]. The oil-source correlation cannot be performed, and the hydrocarbon migration pathways cannot be described in geochemistry due to the lack of comprehensive geochemical analysis of hydrocarbon fluids accumulated in the reservoirs and source rocks in the basin.

The primary purpose of this study is to decipher the origin of the condensates and the newly found crude oils in the ADRBB. To achieve this, the geochemical indexes used for oil source correlation in scientific research have been used as thoroughly as possible. The parameters include isoprenoids, n-alkanes, steranes, terpanes, carbon isotope, and slope factor of alkane. The acyclic isoprenoids, pristane (Pr), and phytane (Ph), which occur together in ancient sediments and their oil products, are believed to have been formed from chlorophyll by the thermal evolution process [10]. The ratio of pristane to phytane (Pr/Ph) can usually indicate the redox conditions of the sedimentary environment [11, 12], while the ratio of pristane to nC₁₇ and phytane to nC₁₈ can distinguish the type, thermal maturity, and biodegradation of sedimentary organic matter and oils [13–15]. Moreover, the values of mole percent n-alkanes can indicate the maturity and biodegradation of oil [16]. The main peak carbon of low-maturity oil is large, usually around C₁₅ or even larger. At the same time, the chromatographic curve of the low-maturity oil shows serrated, with the carbon predominance index (CPI) and odd-even predominance (OEP) being deviated from 1. On the contrary, the main peak carbon of high maturity oil is small, commonly C₅ or C₆. The chromatographic curve is relatively smooth, with the CPI and OEP close to 1 [17–22]. Furthermore, the baseline bump of the chromatographic curve represents the existence of unresolved complex mixtures (UCM), which are formed by the strong biodegradation of oil in geological history [23–27]. Thompson [28] reported that the hybrid condensates may have an excess of C₇ components, exceeding the projection of the exponential trend established in the C₁₀+ region and showing an imperfect exponential regression. In addition, steranes originate from sterols in eukaryotic cell membrane lipid, which occur in algae and higher plants [29, 30]. Tricyclic terpanes originate from prokaryotic cell membranes [31]. It has also been suggested that the enrichment of tricyclic terpanes is related to the input of marine algae [32–35]. The relative proportions of regular steranes [36–41] and tricyclic terpanes [42–47] in living organisms are related to specific environments. It is believed that steranes and tricyclic terpanes in sediments might provide valuable paleoenvironmental information. For some compounds in steranes or terpanes, such as C₂₉ regular steranes, 18α-trisnorneohopane, and C₃₂ hopanes (the latter two are pentacyclic triterpenes), the relative content of thermally stable geological molecular configuration can indicate the maturity of organic matter, which is widely used to indirectly calculate the maturity of oil [14, 47–51]. Gammacerane (C₃₀H₅₂) is a special pentacyclic triterpene with a six-membered ring E, while its isomer, C₃₀ hopanes (C₃₀H₅₂), which also belong to pentacyclic triterpenes, possesses a five-membered ring E [52]. The gammacerane index, the ratio of gammacerane to C₃₀ hopanes, was initially associated with hypersaline paleoenvironments but more recently has been extended and established as marker for water-column stratification [53–57]. In fact, the essence of water-column stratification is the different water density, which can be affected by salinity, so the gammacerane index that can characterize water-column stratification has a certain relationship with water salinity. From the perspective of elements, stable carbon isotopic is of great importance in the analysis of the oil source since the formation process of oil and gas is the process of kerogen cracking. Kinetic cracking, which preferentially breaks ¹²C–¹²C bonds from kerogen to pyrolysate compared with ¹²C–¹³C bonds and ¹³C–¹³C bonds, could be a plausible explanation of isotopic fractionation in geological history. Therefore, kerogen pyrolysis products from the same source, such as oil, gas, and pyrobitumen, are isotopically comparable, which can provide clues for oil-source correlation [58–62].

The core of the correct interpretation of hydrocarbon migration and accumulation is accurate oil-source correlation which depends on the proper parameter selection. Then, the formation model of hydrocarbon reservoirs can be accurately restored by combining the correlation results with the forming period of hydrocarbon reservoirs and evolution process of traps [2, 63]. The exact reservoir formation model is conducive to identifying and predicting the direction of hydrocarbon migration and accumulation, locating the distribution range of potential and prospective hydrocarbon reservoirs, and guiding the exploration direction, which can provide a highly necessary theoretical support for oil and gas exploration and development.

The Amu Darya basin, which covers an area of 42.7 km² located in Turkmenistan (65%) and Uzbekistan (25%), extends southwestward into Iran and southeastward into Afghanistan (Figure 1). According to lithology, structural characteristics, and sedimentary period, the Amu Darya basin can be divided into three units upwards: basement, intermediate transition layer, and overlying cover where the petroleum systems are developed [1]. The basement is mainly composed of commonly metamorphosed Paleozoic rocks. The intermediate transition layer consists of Permian-Triassic rocks, which are strongly compacted and diagenetically altered. The overlying cover includes Lower to Middle Jurassic coal-bearing mudstone, Middle to Upper Jurassic carbonates, Upper Jurassic evaporites, Cretaceous-Paleogene marine clastic rocks with carbonate intervals, and Neogene continental clastics [64]. The Amu Darya basin can be divided into some first-order structural units, such as Kopet-Dagh foredeep (southwest of the basin), Murgab depression (southeast of the basin), Khiva-Zaunguz depression (northwest of the basin), and Chardzhou stop (northeast of the basin) based on the morphology of the basement and structural fluctuation [1, 65]. The ADRBB, which is surrounded by the Amu Darya River and the country boundary of Turkmenistan-Uzbekistan-Afghanistan, is located on the southeast of the Chardzhou step (Figure 2). The block is divided into three sections: west, middle, and east, which contain six secondary structural units. The western section is composed of the Chardzhou uplift and Kennedykiddskurt uplift. The middle section includes Carlabekaul depression, Sandykly uplift, and Bieshikent depression. The eastern section consists of the Gissar piedmont thrust belt [66, 67].

The thick Lower to Middle Jurassic, largely continental, coal-bearing rocks are the main source rocks in the Amu Darya basin (the Lower to Middle Jurassic source rocks are denoted by J_1-2 coal-bearing rocks). And the overlying Callovian-Oxfordian carbonates, which are denoted by J₂k-J₃o, are the major gas-paying beds conformably overlain by thick gypsum-salt rocks of the Gordak Formation, which were deposited on the Kimmeridgian-Tithonian time (Figure 3) [69–73]. The Callovian-Oxfordian carbonates contain a certain proportion of argillaceous limestone with high organic matter content, which is another set of source rocks in the Amu Darya basin [74]. Furthermore, at the end of the Oxfordian period, a set of black mudstones with high gamma values developed in the east-central part of the basin, which is the secondary source rocks in the Amu Darya basin. In the ADRBB, the black mudstone layer, also known as the gap layer, is developed only in the middle-east of the block, with a small thickness, mostly between 5 and 20 m [66, 67]. From bottom to top, the Lower to Middle Jurassic coal-bearing mudstone, the Callovian-Oxfordian carbonates, the gap layer, and the Kimmeridgian-Tithonian gypsum-salt layer form a set of the good reservoir forming assemblage, which provides a material basis for the formation of large gas fields on the ADRBB [75–77].

The ADRBB has experienced the evolution process of continental-marine-lagoon facies. After being uplifted in the Late Triassic, the ADRBB reentered a subsidence stage in the Early-Middle Jurassic (J_1-2). The depositional environment of the ADRBB gradually evolved from delta to marine facies [1, 2, 78, 79]. The ADRBB experienced a large-scale transgression in the Callovian period of the Middle-Late Jurassic (J₂k), and a Callovian ramp carbonate platform developed on the basis of the formed accommodation space in this period. During the Oxfordian period (J₃o), the ADRBB experienced another transgression. The carbonate platform was generally submerged and retreated and then evolved to a rimmed shelf-type carbonate platform with marginal reefs (Figure 4) [64, 66]. In the Late Jurassic Kimmeridgian period, the ADRBB began to enter the gypsum deposition stage and the carbonate platform gradually degenerated into an evaporitic platform [80]. The ultrathick gypsum-salt layer developed then, widely protecting oil and gas reservoirs.

Core samples for source rock evaluation and oil correlation were collected from Lower to Middle Jurassic coal-bearing rocks at wells Msa-1, Mc-1, Nye-1, and Wb-1 (3416 m–4130 m; the depth range refers to the range from the shallowest depth to the deepest depth of the wells sampled in this layer, the same as follows). The Callovian-Oxfordian carbonate source rocks were taken wells Msa-1, Sa-2, Mc-1, Soj-1, Nye-1, Wb-1, Wb-16, and Ejb-1 (3170 m–4320 m). The gap source rocks, which only developed in the middle-east of the ADRBB, were collected at wells Msa-1, Mc-1, Nye-1, Wb-1, and Ejb-1 (3026 m–4065 m).

The crude oil and condensate samples were collected from the middle part of the block. The condensate came from well Wb-92D while the crude oil came from well Wes-2. The samples were stored in Schlumberger single-phase-sampling cylinders from the holding tank associated with each well. Crude oil and condensate samples were analyzed immediately after sampling to avoid volatilization of light hydrocarbon and contamination.

The total organic carbon content (TOC) of source rocks, which can characterize the organic enrichment, was measured by the LECO CS-230 Carbon/Sulfur analyzer. The Rock-Eval-6 pyrolysis instrument determined the pyrolysis data of source rocks, which can characterize the abundance, type, and maturity of organic matter. The S₁ peak of residual hydrocarbon was analyzed at 300°C for 3 min. The S₂ peak of newly generated hydrocarbon is analyzed at the temperature range of 300°C–650°C with a heating rate of 25°C/min. Mean vitrinite reflectance (⁠$Ro$⁠) is based on the measurement of over fifty random-orientation vitrinite particles in each sample to ensure accuracy.

A YSB-automatic multifunction extractor recovered the extractable organic matter (EOM) in source rocks. The EOM solution was evaporated using rotary-evaporation to a small amount for geochemical analysis.

Bulk molecular characterization for whole oils and EOMs was performed using gas chromatography (GC) methods. Biomarkers of aliphatic, aromatic, and N–S–O compounds in whole oils and EOMs were determined using the gas chromatography–mass spectrum (GC–MS). Before GC–MS analysis, the whole oil and EOMs were complexated to improve the signal-to-noise ratio of the biomarkers. Gas chromatography with flame ionization detector (GC–FID) was equipped with an elastic quartz capillary column DB-5 with a length of 30 m and an inner diameter of 0.25 mm, and the carrier gas was He at a constant flow of 1.0 mL/min. The oven temperature was programmed from 80°C to 310°C, with a 6°C/min heating rate. The GC–MS test runs were based on a Trace GC Ultra system coupled to a dual-stage quadrupole mass spectrometer. The GC was equipped with a programmed temperature vaporizer (PTV) injection system and a DB-5MSUI elastic quartz capillary column (⁠$60m×0.25mm×0.25mm$⁠), and the flow of carrier gas (He) was at a constant rate of 1.0 mL/min. The GC oven was programmed as follows: hold 100°C for 5 min, heating to 220°C at a rate of 4°C/min, then, continue to rise to 320°C at a rate of 2°C/min, and finally, hold at 320°C for 20 min. The mass spectrometer was operated in both full-scan and selected ion monitoring (SIM) modes with an ionization energy of 70 eV.

The interpreted 3D seismic volume was acquired from the China National Petroleum Corp.; the profiles discussed in this paper were extracted from those 3D seismic data. The data undergo correction, stacking, inverse filtering, migration, and imaging processes. Landmark seismic interpretation system provided the propriety migrated profiles and Society of Exploration Geophysicists (SGY) format files. The software was used to visualize the SGY files and convert vertical time axes to depth. Several drill holes, such as wells Wb-92D, Wes-2, and Wb-1, allow a direct correlation to be made between the seismic and geological units. Balancing and restoration of seismic sections were conducted by stripping off (back-stripping) each sequence and restoring faults and folds by using MOVE® 2016.1 software so that the paleotopography or specific period surfaces were restored to continuous layers with topography [82]. Faults were restored by using the incline-shear algorithm, which assumes that the hanging wall deforms by simple shear and that the footwall remains undeformed by compression. Next, the cross-section was restored by using the flexural-slip unfolding algorithm. After unfolding, the new sequence was backstripped and the same procedure was repeated until only the deepest layer remained.

The quality of source rocks was evaluated from the organic matter’s abundance, type, and maturity. The maturity of source rocks will be discussed in Section 5.2 together with the maturity of crude oils and condensates. Figure 5 is a crossplot of TOC and hydrocarbon generation potential (S₁ + S₂) data of source rocks listed in Table 1. Since TOC values and hydrocarbon generation potential values of all samples may differ by up to two orders of magnitude, a double-logarithmic coordinate system is adopted in Figure 5. The TOC percentage varies between 1.93% and 2.84% (average 2.32%) in J_1-2 coal-bearing mudstone, and that of gap layer mudstone is higher than J_1-2 coal-bearing mudstone with the range 4.65%–6.73% (average 5.63%). The TOC percentage of J₂k-J₃o carbonate ranges from 0.42% to 0.88% (average 0.62%). Similar to TOC percentage, the hydrocarbon generation potential of mudstone source rocks is higher than that of carbonate rocks. The hydrocarbon generation potential of J_1-2 coal-bearing mudstone varies between 1.72 and 3.51 mg/g (average 2.34 mg/g), and those of gap layer mudstone and J₂k-J₃o carbonate are 12.11–17.36 mg/g (average 14.99 mg/g) and 0.77–2.19 mg/g, respectively. According to Peters’ [83] and Jarvie’s [84] theories, the evaluation criteria of mudstone and carbonate source rocks are listed in Table 2 and plotted in Figure 5. The zone framed by the orange dashed line in Figure 5 is suitable for evaluating the hydrocarbon generation capacity of mudstones such as J_1-2 coal-bearing mudstone and gap layer mudstone, while the zone framed by the green dashed line is adopted to evaluate that of carbonate rocks such as J₂k-J₃o carbonate. TOC is generally linearly and positively correlated with hydrocarbon generation potential, so the level of source rocks can be determined by the higher one of the two parameters when they are not in the same level. The J_1-2 coal-bearing mudstone belongs to level III (good), while the J₂k-J₃o carbonate and the gap layer belong to levels II–III (very good-good) and levels I–II (excellent-very good), respectively. In other words, the hydrocarbon generation capacities of the three sets of source rocks display a general upward increasing trend with the increase in organic matter abundance.

The source rocks corresponding to the initial definition of kerogen types I–III are ranged in the order of decreasing petroleum potential and thus of the H/C ratio [85]. Generally, the atomic H/C ratio of kerogen type I is higher than 1.5 and that of kerogen type III is lower than 0.8. The average atomic H/C ratios of the kerogens of the J₁-₂ coal-bearing mudstone, the J₂k-J₃o carbonate, and the gap layer mudstone are 0.43–0.81 (average 0.62), 0.72–1.15 (average 0.92), and 1.33–1.47 (average 1.38), respectively (Table 1). In the Van Krevelen diagram (Figure 6) using the element ratio parameters of kerogen in Table 1, the kerogens of source rocks upwards are divided into type III, mixed type II/III, and type II. Moreover, three carbonate samples are plotted in the area of type III kerogen, showing that although the sedimentary environment of J₂k-J₃o carbonate in the ADRBB is marine facies, and there is still some input of terrestrial organic matter in the process of deposition. These organic matter abundance and kerogen types can serve as liquid petroleum and gas sources.

Figure 7 shows plots of mole% alkanes (⁠$y$⁠) versus carbon number (⁠$x$⁠) in log-linear axes for the condensate from well Wb-92D and crude oil from well Wes-2 and linear-linear for the crude oil from well Wes-2. The condensate from well Wb-92D ranges from C₆ to C₂₉. The C₆₊ components in the condensate from well Wb-92D present an exponential trend (⁠$y=Ae−αx$⁠) and high slope factor (⁠$SF=eα$⁠), indicating thermogenic origins and negligible alterations. However, the exponential progression’s correlation coefficient (⁠$R2=0.913$⁠) is relatively lower than 0.99, indicating that the condensates are formed by a mixing source [13, 28]. The admixture of two petroleum fluids possessing different but perfect exponential progressions among their components results in a nonexponential product. And the concentration of C₇ exceeds C₆, which also indicates that it is formed by a mixed source or evaporative fractionation. However, the evaporative fractionation generally leads to a minor exceeding [28]. The significantly exceeding indicates that evaporation fractionation has a limited effect on the condensate formation from well Wb-92D.

The $x$-axis value of each point represents the sum of normal and branched alkanes with this carbon number. The $y$-axis on the left is a logarithmic coordinate used for exponential regression of the mole percentage of condensates and crude oils. The $y$-axis on the right is a linear coordinate used to observe the predominance of even C-numbers in crude oils (modified from Thompson [28]).

The crude oil from well Wes-2 ranges from C₁₀ to C₄₀. Although the main peak carbon is C₁₄, with the content of 4.98%, there is a minor difference between the content of other carbon numbers and that of C₁₄, resulting in a low exponential regression correlation coefficient (Figure 7). The evolution process of oil is perceived as a series of increasing from C₄₀ to C₁₀ in view of the thermal origin, the conversion of heavier to lighter molecules. However, there is a predominance of even C-numbers in C₁₂–C₁₆ and C₂₂–C₂₆ with $OEP=0.9$ (odd-even predominance), making the curve show a sawtooth shape in the rectangular coordinate system (Figure 7) and indicating that the crude oil from well Wes-2 is in a low-maturity state. The predominance of even C-numbers in C₁₂–C₁₆ may be attributed to rocky coals with the presence of unsaturated fatty acids and specific inputs from organisms, such as bacteria and diatoms [91–94], while the predominance of even C-numbers in C₂₂–C₂₆ may relate to reductive processes in highly saline, carbonate environments that were generated from the remains of algal communities [95]. The wide carbon number range, predominance of even C-numbers, and no obvious main peak carbon number imply that the crude oil from well Wes-2 is not the product of single-source rock but mixed and low-maturity source origin.

Representative steranes and terpane distributions in the condensates and the crude oils from the ADRBB are illustrated using the $m/z$ 217 and $m/z$ 191 mass chromatograms in Figure 8. The baselines of the gas chromatograms are relatively flat without the unresolved complex mixture (UCM) hump, indicating that neither condensates nor crude oils have undergone strong biodegradation. The condensate from well Wb-92D has more abundant 5α(H),14α(H),17α(H) C₂₉ steranes than 5α(H),14α(H),17α(H) C₂₇ steranes, while the latter is more enriched in the crude oil from well Wes-2. Similarly, the fraction of the diasteranes is also higher in the condensate. The distribution patterns of terpanes are also significantly different. The Wb-92D condensate has more concentration of Ts than C₂₃TT in the $m/z$ 191 mass chromatogram, with C₂₃TT being the highest peak in tricyclic terpane homologues. Compared with gammacerane, the predominance of C₃₀ hopane is more obvious. The concentration of C₃₁ to C₃₅ homohopanes decreases slowly with the increase of the carbon number. In Wes-2 crude oil, C₂₃TT is dominant in tricyclic terpane homologues and its concentration is higher than that of Ts. The predominance of C₃₀ hopane over gammacerane is smaller than that in condensate. The partial reversal of the homohopanes series occurred with C₃₅>C₃₃ and C₃₄. The ratios or relative concentration parameters of isoprenoids, steranes, and terpanes derived from GC/GC–MS, containing various oil-source correlation information, are listed in Table 3 and analyzed in Section 5.

The shape and trend of the stable carbon isotope type curve can be used to identify the correlation between petroleum and kerogen [96, 97]. The isotopic composition of each fraction will be differentiated during the pyrolysate process of kerogen due to isotopic fractionation, and the carbon isotope will increase with the increasing component polarity and boiling point [98–100]. The carbon isotope sequence of each pyrolysis product approximately shows the following trend [61, 101–104]: saturated hydrocarbon < whole oil < aromatic hydrocarbon < polar hydrocarbon < asphaltene < kerogen.

The δ¹³C values of the condensates and crude oils are −24.4‰ and−27.1‰, respectively (Table 4). The condensates are isotopically heavier than the crude oils, with a difference of 2.7‰. The δ¹³C values of EOMs of J_1-2 coal-bearing mudstone vary from −26.3‰ to −25.2‰ (average −25.7‰), while that of the J₂k-J₃o carbonate and gap layer are −27.9‰ to −26.3‰ (average −27.1‰) and −28.2‰ to −26.8‰ (average −27.3‰), respectively. The carbon isotopes of EOMs showed a trend of becoming lighter upward, which also occurred in the kerogens. The δ¹³C values of J_1-2, J₂k-J₃o, and gap layer kerogen are −25.3‰ to −24.0‰ (average−24.6‰), −27.0‰ to −25.2‰ (average−26.2‰), and−27.4‰ to −26.4‰ (average−26.8‰), respectively. In the source rocks, the J_1-2 coal-bearing mudstone shows the typical characteristics of continental organic matter; the carbon isotope of a component in the J_1-2 coal-bearing mudstone is isotopically heavier than that of the corresponding component in the gap layer and J₂k-J₃o carbonate rocks.

The maturity assessment is conducive to unravel the impact of maturity on oil-source correlations. By the quantitative analysis of the diverse molecular biomarkers, the source rocks, condensates, and crude oils of the ADRBB can be distinguished evidently. In this study, two pairs of indicators are selected to assess thermal maturity: C₂₇ diasteranes/C₂₇ steranes-Ts/Tm [105] and the isomerization index of the C₂₉ steranes (Figures 9 and 10). The C₂₇ DiaSt/C₂₇ St has high sensitivity from early mature to early overmature. The proportion of diasteranes will gradually increase with the increase of maturity [16, 106]. However, it must be pointed out that lithology may affect this parameter [107, 108]. Acidic catalysis is essential in converting sterenes to diasterenes which are the precursors of diasteranes. Therefore, the C₂₇ DiaSt/C₂₇ St ratio is usually relatively low in carbonate rocks and their generated unmixed petroleum. Bennett and Olsen [109] used C₂₇ DiaSt/C₂₇ St as an indicator of source rock lithology and confirmed this conclusion.

Figure 9 shows that the C₂₇ DiaSt/C₂₇ St ratio of J₂k-J₃o carbonate rocks is the lowest (0.04–0.15) due to lithology (Table 3). The C₂₇ DiaSt/C₂₇ St ratio of the gap layer and J_1-2 mudstone positively correlates with Ts/Tm; the deeper the buried depth is, the larger the C₂₇ DiaSt/C₂₇ St ratios and Ts/Tm ratios are. The C₂₇ DiaSt/C₂₇ St ratio of the crude oil from well Wes-2 is slightly larger than that of J₂k-J₃o carbonate rocks and marginally smaller than that of the gap layer. Simultaneously, its Ts/Tm value is near the median of the two types of rocks, showing a close genetic relationship with the two sets of source rocks. The two ratios of the condensate from Wb-92D are considerably larger than those of the source rocks in the ADRBB, showing that it is significantly maturer than the source rocks in the ADRBB. This suggests that the high-maturity condensate from well Wb-92D is not indigenous but migrated products from much deeper levels.

In addition to the C₂₇ DiaSt/C₂₇ St-Ts/Tm crossplot, the isomerization index of C₂₉ steranes can evidently distinguish the source rocks and petroleum [16, 48, 49, 110, 111] (Table 3). The isomerization of C₂₉ steranes at the C-14 and C-17 positions makes the C₂₉ββ/(ββ + αα) ratio increase from close to 0 to 0.7 with the increasing maturity (the equilibrium value is 0.67–0.71), while the isomerization of C₂₉α, β, β steranes at C-20 makes the C₂₉ S/(S + R) ratio increase from 0 to about 0.5 with the increasing of maturity (equilibrating at 0.52–0.55) [16]. It can be seen in Figure 10 that the maturity relationship of the three sets of source rocks approximately presents the trend of decreasing upward, which is similar to the trend of Ts/Tm in Figure 9. The J₂k-J₃o carbonate rocks and the gap layer are mixed together with the crude oil from Wes-2 in the figure, showing a relatively low-maturity state and genetic relationship. However, the C₂₉ββ/(ββ + αα) ratio and the C₂₉ S/(S + R) ratio of the condensate from well Wb-92D are significantly higher than those of the source rocks in the ADRBB, indicating that the condensate originates from source rocks with higher maturity.

The result of maturity calculation shows that the condensate from well Wb-92D originated from the source kitchen with higher maturity outside the ADRBB. That is to say, under the premise that the strata are not uplifted and eroded to a large extent and the geothermal gradient is the same, the source kitchen should be the tectonic region in the Amu Darya basin deeper than the ADRBB. According to the existing reports, the source kitchen of the condensates is the Murgab depression. In contrast, the crude oil from well Wes-2 is mainly from the local source rocks of the ADRBB, which is different from the origin of the condensates.

After the discussion in Section 5.1, it is confirmed that the source rocks in the middle section of the ADRBB are in a low-maturity stage, indicating that the organic matter is relatively close to the original state of sediment. For example, the C₂₉ββ/(ββ + αα) ratio and the C₂₉ S/(S + R) ratio of the source rocks have not reached the equilibrium value, which make it possible to identify the sedimentary environment by biomarkers, since the thermal evolution process of organic matter will differentially deplete biomarkers and change the mole percent of components that can indicate the sedimentary environment, leading to the wrong conclusions.

Figure 11 shows a crossplot of ratios of isoprenoids to normal hydrocarbons in Table 3, which validates the sedimentary environment of source rocks and petroleum fluids and determines their maturity [112–114]. The organic matter of J_1-2 coal-bearing mudstone is mainly type III kerogen formed by terrestrial organic matter deposition. With the progress of transgression, the seawater continued to be deepening. The J₂k-J₃o carbonate formed mixed type II/III kerogen from terrestrial and marine organic matter, while the gap layer deposited in the deepest seawater is type II kerogen formed by marine organic matter. According to the analysis in Section 4.2, the two petroleum fluids may be formed by mixed sources. Figure 11 shows that the condensate from well Wb-92D is mostly mixed type II/III, which may come from J_1-2 coal-bearing mudstone and J₂k-J₃o carbonate, and the crude oil from well Wes-2 is primarily marine, which may originate from the gap layer and J₂k-J₃o carbonate. However, note that the evolution degree of the condensate is separated from the source rocks in the ADRBB, while the evolution degree of the crude oil is well comparable with the samples. This may imply that the condensate migrated from outside the block, and the crude oil is the product of hydrocarbon generation and expulsion of local source rocks.

Although paleosalinity [115] and maturity [116] may have a certain impact on pristane and phytane concentrations, the pristane/phytane (Pr/Ph) ratio is still a helpful parameter to differentiate oxic from anoxic/dysoxic sedimentary environments [16]. It is generally believed that $Pr/Ph<1.0$ represents the deep and anoxic reducing environment and $Pr/Ph>2.0$ indicates the shallow and oxygen-enriched oxidizing environment, while Pr/Ph between 1.0 and 2.0 reflects the transitional weak oxidizing-weak reducing environment. Therefore, Pr/Ph negatively correlates with water depth [16]. Figure 12 shows that the Pr/Ph of J_1-2 coal-bearing mudstone is between 1.0 and 2.0, that of J₂k-J₃o carbonate is between 0.5 and 1.5, and that of the gap layer is between 0 and 1.0. The water gradually deepens in the evolution process from the sedimentary environment of J_1-2 coal-bearing mudstone to the gap layer. The crude oil from well Wes-2 has a relatively low Pr/Ph, which is consistent with the previous understanding that the petroleum fluids with a predominance of even C-numbers have a low Pr/Ph [117], while the condensate from well Wb-92D has a high Pr/Ph, indicating that its parent material is formed in an oxic environment.

The gammacerane index (gammacerane/17α,21β[H]C₃₀ hopane) and extended tricyclic terpane ratio (⁠$ETR=C28+C29/Ts$⁠) can be used to reflect changes in water salinity/alkalinity through geologic time [54, 118] (Table 3, Figure 13). The gammacerane is believed to be formed by reduction of tetrahymanol originated from bacterivorous ciliates, implying a strong reduction and high-salinity environment [119, 120]. The ETR has been taken to distinguish Jurassic reservoired oil from Triassic as an age-related parameter [121, 122]. ETR is relatively less affected by thermal maturity and biodegradation due to the high thermal stability of C₂₈TT, C₂₉TT, and Ts to ensure a more accurate interpretation of the parent source. Holba et al. [121] found that the ETR of Triassic oils exceeds 2.0, whereas the Jurassic ETR has lower ratios corresponding to major mass extinction that occurred at the end of the Triassic, which may have altered the principal biological sources of tricyclic terpanes, mostly less than 1.2. Hao et al. [54] observed a relationship between source rocks deposited in deep lake environments and the relative abundance of several terpanes. They pointed out that the high gammacerane index and ETR values are well correlated with source rocks deposited in relatively high salinity zones.

The gammacerane index of the J_1-2 coal-bearing mudstone is 0.15–0.24, that of the J₂k-J₃o carbonate rocks is 0.19–0.28, and that of the gap layer is 0.25–0.41 (Figure 13). It shows that during the deposition period of the gap layer, the environmental water body of the ADRBB was deep, anoxic, and saline, while the gammacerane index of the J_1-2 coal-bearing mudstone and the J₂k-J₃o carbonate rocks is relatively low, showing a shallow, oxic, and freshwater body, which is consistent with the conclusions of isoprenoids/normal hydrocarbons in Figure 11 and pristane/phytane in Figure 12. The ETR of the J_1-2 coal-bearing mudstone is 0.23–0.69, that of the J₂k-J₃o carbonate rocks is 0.27–1.13, and that of the gap layer is 0.69–1.27 (Figure 13). The ETR of all source rocks is less than 1.2, except for one sample of the gap layer. Generally, the ETR of the J_1-2 coal-bearing mudstone is the lowest. With the gradual prosperity of organisms after the Triassic mass extinction [123], the ETR of the J₂k-J₃o carbonate rocks and gap layer rise again and increase upward. The observed upward increase in the gammacerane index, which is accompanied by an upward increase in ETR and an upward decrease in Pr/Ph ratios, therefore, suggests an evolution of the ADRBB from a shallow-oxic freshwater delta to a deep-anoxic saline marine [2].

Figure 13 shows that the condensate from well Wb-92D has low gammacerane index and medium ETR, which is distributed between the J_1-2 coal-bearing mudstone and the J₂k-J₃o rocks, indicating that its sedimentary environment is the same as these two types of source rocks. The crude oil from well Wes-2 has a gammacerane index as high as that of the gap layer, indicating that the reducing environment plays a controlling role in the formation of the crude oils. The lower ETR value of the crude oils may be caused by mixed source effect.

Steranes originate from sterols in eukaryotes, which occur in algae and higher plants [29, 30]. Either C₂₇ steranes or C₂₉ steranes usually have the highest relative proportion in the C₂₇–C₂₈–C₂₉ sterane (5α,14α,17α[H] 20R) system due to different sedimentary environments and organic matter input. Generally, C₂₇ steranes are derived primarily from marine zooplankton and algae [16, 39, 124] and C₂₉ steranes are mainly from terrigenous higher plants, freshwater microalgae, Chlorophyta, and Cyanobacteria [36–38, 125, 126], whereas C₂₈ originates from chlorophyll C-containing phytoplankton, including diatoms, coccolithophores, and dinoflagellates [40, 127, 128]. Furthermore, tricyclic terpanes, which are relatively less affected by maturity and biodegradation, can also be used to identify paleoenvironments. The precursors of tricyclic terpanes are thought to originate mainly from prokaryotic cell membranes [31]. It has also been suggested that the enrichment of tricyclic terpanes is related to the input of marine algae [32–35]. It is generally believed that tricyclic terpanes with low carbon numbers such as C₁₉ and C₂₀ tricyclic terpanes may come from diterpene precursors, which reflect the input of higher plants which represent terrestrial facies [42, 43]. The sedimentary environment in shallow water may be conducive to the formation and distribution of C₁₉ and C₂₀ tricyclic terpanes [43]. Typical coal and coal-derived petroleum are rich in C₁₉ tricyclic terpanes. The predominance of C₂₃ tricyclic terpane concentration is obvious in marine source rocks and oil [44, 129], while the concentration of C₂₁ tricyclic terpane is dominant in freshwater lacustrine source rocks and oil [45, 46].

Figures 14 and 15 are ternary plots of the relative proportions of C₂₇–C₂₈–C₂₉ steranes and C₁₉+C₂₀–C₂₁–C₂₃ tricyclic terpanes, respectively (Table 3). There is a predominance of C₂₉ steranes and C₁₉+C₂₀ tricyclic terpanes in J_1-2 coal-bearing mudstone in respective compound systems (C₂₇–C₂₈–C₂₉ steranes & C₁₉+C₂₀–C₂₁–C₂₃ tricyclic terpanes), showing the characteristics of a terrestrial organic matter input. However, the J_1-2 coal-bearing mudstone also contains a certain proportion of C₂₇ steranes and C₂₃ tricyclic terpanes, representing the input of marine organic matter, which is consistent with the sedimentary characteristics of delta in marine-continental transitional facies. On the contrary, the gap layer shows the proportion dominance of C₂₇ steranes and C₂₃ tricyclic terpanes, indicating that the source of organic matter is mainly marine. The J₂k-J₃o carbonate rocks are distributed between the J_1-2 coal-bearing mudstone and the gap layer, showing that the input of continental organic matter is approximately equal to that of marine organic matter and indicating the process of a continuous deepening of seawater with decreasing continental organic matter input and increasing marine organic matter input from the deposition period of J_1-2 coal-bearing mudstone to the deposition period of the gap layer. The projection point of the condensate from well Wb-92D falls in the deltaic-coastal zone with the J_1-2 coal-bearing mudstone and the J₂k-J₃o carbonate samples, whereas the crude oil from well Wes-2 is located in the zone of open marine with gap layer and J₂k-J₃o carbonate samples. It shows that the condensates originate mainly from the J_1-2 coal-bearing mudstone, whereas the crude oils mainly come from the mudstone of the gap layer. And the J₂k-J₃o carbonate rocks, which are both source rocks and reservoirs, contribute to the formation of the condensates and the crude oils.

The EOMs and kerogens are significantly depleted in δ¹³C compared to the condensate from well Wb-92D (Table 4, Figure 16). The EOM can be regarded as the hydrocarbon generated but has not been expulsed from the source rocks, which means that there is no genetic relationship between the condensates and the source rocks of the ADRBB. Using the average value (0.87‰) of carbon isotope $D$ values between all EOMs and kerogens, the carbon isotope of the parent kerogen of the condensates is speculated (⁠$–24.4‰+0.87‰≈–23.5‰$⁠), since the condensates are hybrid products of J_1-2 coal-bearing mudstone, J₂k-J₃o carbonate rocks, and gap layers. Although the condensates show the characteristics of a mixed source, the speculated value of parent material can be representative to a certain extent. According to the isotope fractionation theory, kerogen will be increasingly ¹³C enriched with increasing maturity. Based on the speculated carbon isotope, the condensate originated from the source rocks with higher maturity, which conforms to the judgment of thermal maturity.

The crude oil from well Wes-2 is isotopically lighter than the condensate, indicating that they are not from a common origin. The δ¹³C value of the crude oil is –27.1‰, which is approximately equivalent to that of the EOMs of the gap layer and the J₂k-J₃o source rocks and less than that of the kerogen of the gap layer and the J₂k-J₃o source rocks, indicating that the oil from well Wes-2 is the product of the local gap layer and the J₂k-J₃o carbonate source rocks.

To better understand petroleum generation, expulsion, and migration history in the right bank block, the tectonic evolution processes were restored on the basis of the seismic data and profiles (Figures 17 and 18). The restored results were carefully examined in the context of the regional geological setting. The evolution process of strata and traps as geological elements of the petroleum system can be observed intuitively through the tectonic evolution process. The oil-source correlation results in this study illustrate the origin and migration direction of hydrocarbon. That is, the gas condensates from the highly mature source rocks in the Murgab depression in the southeast of the Amu Darya basin experienced long-distance lateral migration and accumulated in the ADRBB. In contrast, the newly discovered crude oils originate from the local low-maturity source rocks of the ADRBB, which merely experienced short-distance vertical migration and accumulated into J₂k-J₃o carbonate reservoirs. Shan et al. [2] identified three hydrocarbon accumulation periods as critical moments of the petroleum system in the middle section of the ADRBB. The accurate hydrocarbon accumulation model can be established by overlaying the migration direction of hydrocarbons originating from various sources on the tectonic evolution profiles at the critical moments of the petroleum system. The hydrocarbon accumulation processes of the WB gas field and WES gas field are summarized as follows:

The carbonate reservoirs and evaporite strata were deposited on the paleo-uplift in the Jurassic Callovian-Oxfordian and Kimmeridgian-Tithonian stages, respectively, and traps that could capture the passing hydrocarbon were formed at the WB gas field and WES gas field [66, 80].

At the end of the early Cretaceous, the hydrocarbon generated in the deeply buried Murgab depression in the southeast of the Amu Darya basin migrated to the Sandykly uplift on the ADRBB along the unconformity between the Jurassic strata and the basement and the fault running through the basement [1]. Then, the of WB gas field traps captured the hydrocarbon from outside the block to form the first-generation oil and gas reservoir (Figure 19(a)). However, due to the lack of basement faults in the WES gas field, there is no effective migration path between the hydrocarbon migrating from the Murgab depression and the reservoir, making it difficult for the traps of the WES gas field to capture the hydrocarbon to form a reservoir. That is, under the background that hydrocarbon from the Murgab depression was massively charging the middle reservoirs on the ADRBB, the reservoir of the WES gas field remained in the state of “empty trap.”

At the end of the late Cretaceous, the J_1-2 coal-bearing source rocks in the Murgab depression entered the condensate window and generated and expelled a large amount of condensate gas which originated from thermal-cracking oil and kerogen [1]. The condensate gas continued to charge the WB gas field. Meanwhile, the J_1-2 coal-bearing mudstone of the ADRBB reached the mature state and entered the oil window. The generated hydrocarbon began to migrate upward into the carbonate reservoirs driven by buoyancy. In the WB gas field, the hydrocarbon generated locally charged the reservoir together with the condensate gas migrating from the Murgab depression to form the second-generation oil and gas reservoir (Figure 19(b)). In the WES gas field, the trap prevented the upward migration of the hydrocarbon from J_1-2 coal-bearing mudstone of the ADRBB and enriched hydrocarbon in situ (Figure 20(a)).

At the end of the Paleogene, the Callovian-Oxfordian carbonate strata and gap layer mudstone of the ADRBB entered the oil window and began hydrocarbon generation. The carbonates were both source rocks and reservoirs whose hydrocarbon production only needed to migrate to the high point of traps along the structural ridge in the carbonate strata to accumulate. The gap layer mudstone was covered by thick and dense evaporite rock series, which blocked the upward migration of the hydrocarbon generated by the gap layer. Driven by the internal pressure, the hydrocarbon was expelled into the lower carbonate rocks. The hydrocarbon generated by the carbonate rocks and gap layer was mixed with the oil and gas that have accumulated into reservoirs, forming the third-generation oil and gas reservoir (Figure 20(b)).

In the Neogene Period, the ADRBB experienced Himalayan orogeny [65], which changed and adjusted the carbonate traps, some were destroyed by faulting, and some were strengthened by the increase in closure height. It can be seen from the profiles that the Himalayan orogeny resulted in the formation of fault blocks in the south of the WB gas field (Figure 19(c)). And the WB gas field and the WES gas field were subjected to tectonic compression of orogeny, which enhanced the amplitude of the anticline and increased the volume of traps to accommodate more oil and gas (Figure 20(c)).

After the Himalayan orogeny, the ADRBB has not experienced large tectonic movement so that the hydrocarbon reservoirs can be well preserved up to now (Figures 19(c) and 20(c)). In the traps with developed basement faults, such as the WB gas field, the geochemical indices show the characteristics of high maturity, indicating that the oil and gas generated in the Murgab depression were the main sources to continuously charge the traps, while the oil and gas generated by the source rocks in the local depression of the ADRBB were the secondary. In the traps with undeveloped basement faults, such as the WES gas field, the traps mainly captured the oil and gas generated by the local depression of the ADRBB.

The comprehensive geochemical index comparison provides insight into the sedimentary environment, organic matter input, and thermal maturity of the ADRBB source rocks and petroleum. It is considered that the source of the condensate from well Wb-92D is mainly J_1-2 coal-bearing mudstone, supplemented by J₂k-J₃o carbonate rocks. However, the maturity of the condensate is significantly higher than that of the local source rocks, indicating that the condensate should originate from the Murgab depression and accumulated in the ADRBB. The recently discovered crude oil from well Wes-2 mainly comes from the gap layer mudstone, followed by the J₂k-J₃o carbonate rocks, and its maturity and carbon isotope are comparable with the local source rocks. The WES hydrocarbon reservoir is typically formed by the local source rocks.

According to the shape and the formation process of the current hydrocarbon reservoirs in the middle section of the ADRBB, the crude oils which are the same as those of well Wes-2 may be accumulated in the traps with undeveloped basement faults. The traps with developed basement faults are charged with condensates and natural gas from Murgab depression. Due to the difference in density, the crude oils generated from local source rocks cannot displace the condensates and natural gas in these traps. For the traps with undeveloped basement faults, the oil and gas from the Murgab depression cannot be charged due to the lack of an effective migration channel. So, the crude oils originating from local source rocks can accumulate in these “empty traps” driven by buoyancy and capillary force. Exploring the high structural position of carbonate reservoirs without basement faults, which are located at the transition position between depression and uplift, the explorers may find the new commercial crude oil reservoirs.

The organic geochemical data used to support the findings of this study are included within the article.

The authors declare that they have no conflicts of interest.

Yunpeng Shan analyzed the geochemical data and wrote the paper; Hui Chai pointed out the demand for geochemical analysis and obtained the permission of Turkmenistan to publish this article; Hongjun Wang provided the samples and designed the study; Liangjie Zhang and Penghui Su polished the English language; Xiangwen Kong drew the figures; Zhenhua Bai searched the literatures; Muwei Cheng and Hongwei Zhang organized the article format.

This study was supported by the National Major Science and Technology Projects of China (Grant no. 2016ZX05029005) and the Technology Research Projects of CNPC (Grant no. 2021DJ33). We thank CNPC (Turkmenistan) Amu Darya gas company for kindly providing samples in this study.