In this study, core samples from the oil shale geological survey boreholes ZK1306, ZK1802 and ZK1803 were collected for both bulk TOC and Rock-eval screening (Figure 1). In the following, 12 samples were selected as key representatives according to their lithology and TOC content for detailed investigations by means of biomarker identification and quantification.

TOC and Rock-eval data as the organic geochemical parameters, can provide basic information about the organic richness, maturity and quality of organic matter in potential source rock. All the Yan’an Formation samples show TOC content above 1%. The coal and oil shale samples contain TOC above 10% (Table 1), which indicate very good hydrocarbon generation potential as defined by Peters and Cassa (1994). Most samples show HI value below 300 mg/g TOC except one outlier shows HI value up to 700 mg/g TOC (Figure 3), which indicate the samples contain mainly type II2 and type III kerogen (Espitalie et al., 1977). All the samples characterized by Tmax value below 435°C that their thermal maturity is very low. It can be observed that the coal and oil shale samples show much higher S2 value, lower oxygen index (OI = S3/TOC*100%) and higher sulfur content compared to the mudstone samples (Table 1).

Based on the TOC value and lithology, four mudstone and two coal samples selected from the Well ZK1306, four oil shales from the Well ZK1802 and two oil shales from the Well ZK1803 of Yan ’an Formation were extracted with Soxhlet extractor. From the extraction results, it can be observed that the extracted organic matter (EOM) of the coal samples (ave. 0.84 mg/g) is much higher than that of the oil shales (ave. 0.22 mg/g) and the mudstones (ave. 0.18 mg/g), which might due to the high TOC content of the coal samples (Table 2). With MPLC fractionation, EOM of the samples can be divided into saturate hydrocarbon (saturates), aromatic hydrocarbon (aromatics), resins, and asphaltenes (SARA) (Jewell et al., 1974; Jewell et al., 1972). All the coal and mudstone samples show lower amount of saturates and aromatics as well as high content of resins and asphaltenes, while the maltenes of the four oil shales are dominated by saturates and aromatics. The relatively higher amount of saturates and aromatics suggest a relatively higher maturity of the oil shale samples (Peters et al., 2005).

The n-alkane and isoprenoid parameters derived from the gas chromatograms are contained in Table 3. The Upper Triassic Yan ’an Formation source rocks show that they contain well-developed n-alkanes and exhibit unimodal characteristics (Figure 4). The dominant peak carbon numbers are nC₂₃, nC₂₅, nC₂₇ for different samples. A clear odd/even preference (OEP) and Carbon Preference Index (CPI) in their n-alkane distributions can be observed, which indicate the input of higher terrestrial plants (Peters et al., 2004). It is obvious that the OEP and CPI values of the oil shale samples are much lower compared to that of the coals and mudstones, which might indicate the higher thermal maturity of the oil shales. It can be confirmed by the higher Tmax value of the oil shale samples (Table 1).

The pristane/phytane ratio (Pr/Ph) has been widely used to characterize the depositional environment (Chandra et al., 1994; Large and Giże, 1996). The Pr/Ph ratios below 1.0 usually indicate anoxic, marine carbonate lithology and values from 1.0 to 1.5 marine shale lithology. Values higher than 2 indicate deltaic shales or terrestrial environments (Peters et al., 2005). Most of the samples show high Pr/Ph ratios above 3.0, indicating the terrestrial swamp depositional condition.

The m/z 191 Chromatograms with peaks related to different terpane compounds are shown in Figure 5 and the biomarker ratios calculated from the chromograms are given in Table 3. It is obvious that the relative peak heights of the tricyclic terpanes are extremely low for all the samples. Most samples show C₃₀ hopane as the dominant component and much low amount of gammacerane. The diahopane are nearly undetectable. All the samples show significant higher peak of 22,29,30, 18a-trisnorneohopane (Ts) compared to the peak of 22,29,30, 17a-trisnorhopane (Tm). The homohopanes also show very low peak heights.

The tricyclic terpanes are thought to derived from microorganisms and algae, or they can also be transformed from hopanoid at high level of maturity (Seifert and Michael Moldowan, 1978). The extremely low ratios of the tricyclic to pentacyclic terpanes might indicate low maturity of the samples (Figure 5).The ratio of Ts/(Ts + Tm) varies with organic matter input and thermal maturity (Peters and Moldowan, 1993), which is strongly influenced by the transformation from Tm to Ts due to increasing maturity. Slightly high values of Ts/(Ts + Tm) can be observed for the two oil shale samples from Well ZK1803, indicating the higher maturity of the two samples. The C₃₁ homohopane isomerization indexes and C₃₂ homohopane isomerization indexes are also closely linked with the thermal maturity. The indexes will reach a final ratio of about 60% with increasing maturity as the hopane 22R can generally be transformed to hopane 22S (Peters and Moldowan, 1993). The oil shale samples show much higher C₃₁ homohopane isomerization indexes and C₃₂ homohopane isomerization indexes compare to the coal and the mudstone samples, which indicate the higher maturity of the oil shale samples.

The gammacerane index (Ga/C₃₀H) is generally considered to be an indicator of restricted depositional environment that is characterized by a stratified saline and hypersaline water column (Philp et al., 1991; Sinninghe Damsté et al., 1995; Summons et al., 2008). The Ga/C₃₀H values of all the source rock samples are ≤0.1, which indicates the freshwater depositional condition of Yan ’an Formation in this area. However, the slightly higher Ga/C₃₀H values of the mudstone and coal samples indicate the higher salinity of their depositional water column.

The C₃₅ homohopanes can be observed in all the source rock samples, indicating at least suboxic condition. The calculated ratios of C₃₅ Hopane/(C₃₁-C₃₅ homohopanes) are all very low, which indicate the depositional environment was not strongly anoxic. As the oil shale samples show much higher C₃₅ hopane/(C₃₁-C₃₅ homohopanes) ratios compared to the coals and mudstones (Table 3), which indicate the more anoxic depositional environment of the oil shale samples.

The steranes chromatograms with m/z 217 for the source rock samples in this study are show in Figure 6. The C₂₇, C₂₈, C₂₉ regular steranes occur in all the extracts and the C₂₉ ααα20R is the prominent component. The C₂₁, C₂₂ pregnanes are hardly detectable, and the diasteranes also show very low relative peak heights.

The biomarker ratios calculated from the steranes are shown in Table 3. It is generally believed that the C₂₇ steranes related to the input of algae and aquatic organisms (Huang and Meinschein, 1979), the C₂₈ steranes originated from the phytoplankton (Seifert and Michael Moldowan, 1978) and the C₂₉ steranes derived from terrestrial high plants (Volkman, 1986). All the source rock samples are enriched in C₂₉ steranes, suggesting the input of organic matter is mostly of terrigenous higher plants.

As with increasing maturity, the relative high abundance of C₂₉ sterane 20S and C₂₉ sterane αα can be transformed to C₂₉ sterane 20R and C₂₉ sterane ββ, respectively. The calculated C₂₉αββ (20S + R)/(ααα (20S + R) +αββ (20S + R)) and C₂₉ααα20S/(S + R) ratios can be used to evaluate the thermal maturity. The oil shale samples show much higher values of C₂₉αββ (20S + R)/ααα (20S + R) +αββ (20S + R) ratios compared to the mudstone and coal samples, indicating the higher thermal maturity of the oil shale samples.

The formation of diasteranes might result from the clay-catalyzed rearrangement or thermal maturity induced transformation of regular steranes (Rubinstein et al., 1975). The two coal samples have higher diasterane/regular sterane ratios (0.26 and 0.31) compared to the oil shale and mudstone samples (around 0.1) indicating the active clays in the coal samples.

The quantity and quality of the hydrocarbons that can be generated from source rock are mainly controlled by the abundance of organic matter, the type of the organic matter input, and the thermal maturity (Vandenbroucke and Largeau, 2007). It seems that one parameter alone is generally insufficient to yield high potential oil or gas prone sediments (Peters et al., 2004; Yunker et al., 2005).

The average TOC amounts of the mudstone, oil shale and the coal samples are 3.11%, 14.68% and 22.90%, respectively. The values point to the good and very good TOC content (above 1%). Petroleum generation potential (S1 + S2) is another important parameter to evaluate the hydrocarbon generation potential of kerogen. It has been suggested that the S1 + S2 value above 6 is related to good and very good hydrocarbon generation potential and the value below 6 indicates very poor to fair source rock (Tissot and Welte, 1979). According to Figure 7 the oil shale and coal samples show good to very good potential while most of the mudstones have poor to fair potential.

The plot of Pr/n-C17 versus Ph/n-C18 can be used to assess the organic matter input of source rocks (Hanson et al., 2000; Ramanampisoa and Radke, 1995). As shown in Figure 8, all the sample’s plot within the region of Type III kerogen. The organic matter of type III kerogen deposited in terrestrial environment is consistent with previous investigation based on the major and trace elements analysis (He et al., 2022). Another common way to distinguish the sedimentary environment and the main source of organic matter for the kerogen is to use the C₂₇-C₂₈-C₂₉ ternary plots (Peters and Moldowan, 1993). The relatively higher C₂₉ regular sterane contents also indicate the terrestrial depositional environment and the mainly higher plant input for all the samples (Figure 9).

The isomerization values of steranes are usually used to measure the thermal maturity of organic matter. With increasing maturity, isomerization at C-20 in the C₂₉ααα sterane lead to the ratio of 20S/(20S + 20R) increase from 0 to about 0.5 (equilibrium = 0.52–0.55) (Seifert and Michael Moldowan, 1978). Moreover, the C₂₉17α(H), 21α(H) sterane will also be transformed to the C₂₉17β(H), 21β(H) sterane, and the equilibrium value is about 0.57–0.62 (Peters and Cassa, 1994). Based on the cross plot of C₂₉ sterane ββ/(ββ+αα) versus C₂₉ sterane 20S/(20S + 20R) ratio (Figure 10), the analyzed source rock samples are immature to low mature and most of the oil shale samples reached a thermal maturity level equivalent to oil generation window. The biomarker results are consistent with the low Tmax values of the samples. All the samples show Tmax values below 435°C, the low Tmax values might be suppressed by the hydrogen-rich macerals as the oil shale and coal samples show much higher HI values (Dewing and Sanei, 2009; Yang and Horsfield, 2020). It has been reported that the vitrinite reflectance of the oil shale samples is between 0.44% and 0.57% with average value of 0.51% (Yang et al., 2024). Thus, the samples are within maturity level of immature to low mature. The higher C₃₁ homohopane isomerization indexes, C₃₂ homohopane isomerization indexes of the oil shale samples also confirmed that oil shales show much higher thermal maturity compared to the mudstones and coal samples (Table 3).

In summary, the oil shale samples with much higher TOC content, higher thermal maturity compared to the mudstones and show much higher potential for hydrocarbon exploration.

The OI values might somehow indicate the variation of redox condition of the samples at low thermal maturity (Peters et al., 2005). The samples deposited in the reducing environment normally contain lower oxygen content. Figure 11 shows the correlation between the TOC and OI for the source rock samples. It can be observed that the oil shale and coal samples are deposited in a more reducing condition compared to the mudstones. It can also be confirmed by the relatively high sulfur content of the coal and oil shale samples as sulfur content is very sensitive to the redox condition.

The ratios of C₃₅ hopane/(C₃₁-C₃₅ homohopanes), Pr/n-C17 and Ph/n-C18 can also be used to analyzed the depositional environment (Peters and Moldowan, 1993). The higher C₃₅ hopane/(C₃₁-C₃₅ homohopanes) values of the oil shale samples reflect their relatively reducing environment compared to the coals and mudstones. Surprisingly, the ratios of pristane/n-C₁₇ and phytane/n-C₁₈ plotted in the oxidizing area in Figure 8. As the source rock samples contains high TOC amount, the depositional environment should not be oxidizing. The high Pr/n-C₁₇ values might be explained by the kerogen sulfurization and associated chemical reactions at early diagenesis that lead to the elevation of pristane (Ghassal et al., 2016).

The co-deposition of oil shale and coal indicate both the similarity and variability of the depositional condition. The development of coal and oil shales generally requires relatively stable structures, warm and humic climate as well as abundant organic matter input (Wang et al., 2016). The difference is that the coals developed in a shallow swampy environment, while the water column of the oil shales are deeper.

The Ga/C₃₀ hopane values of the oil shale samples are slightly lower compared to the coals (Table 3), reflecting the lower salinity of the depositional water column. The low salinity might indicate the deep-water column of the oil shale formation and less influence from the adjacent gypsum rock (Xian et al., 2023). The deduction is also consistent with the redox environment of the samples. The oil shale developed in the deeper and more reducing environment and the coal developed in the shallow water environment with more higher plant organic matter input.