Hydrocarbon source rock evaluation of the Lower Cretaceous system in the Baibei Depression, Erlian Basin

Original Article Hydrocarbon source rock evaluation of the Lower Cretaceous system in the Baibei Depression, Erlian Basin Energy Exploration & Exploitation 2018, Vol. 36(3) 355 372! The Author(s) 2017 DOI: 10.1177/0144598717748761 journals.sagepub.com/home/eea Hua Liu, 1 Jinglun Ren, 2 Jianfei Lyu, 2 Xueying Lyu 1 and Yuelin Feng 1 Abstract The K 1 s, K 1 d, K 1 t, and K 1 a Formations are potential source rock intervals for hydrocarbon formation, all of which are part of the Lower Cretaceous system in the Baibei Depression in the Erlian Basin in China. However, no well has found oil flow because the hydrocarbon-generating potential of the source rocks has not been comprehensively evaluated. Based on organic geochemical and petrological analyses, all the source rocks possess highly variable total organic carbon and S 1 +S 2 contents. Total organic carbon and S 1 +S 2 contents indicate that the K 1 a 2 Formation through the K 1 d 1 Formation are source rocks that have fair to good generative potential and the K 1 d 2 Formation through the K 1 s Formation are source rocks that have good to very good generative potential. The organic matter in the K 1 a 2 Formation is dominated by Type I and II kerogen; thus, it is considered to be oil prone based on H/C versus O/C plots. Most of the analyzed samples were deposited in reducing environments and sourced from marine algae; thus, they are oil prone. However, only two source rock intervals were thermally mature with vitrinite reflectance values in the required range. Hydrocarbon-generating histories show that the K 1 t and K 1 a 2 intervals began to generate hydrocarbons during the depositional period of the K 1 d 2 and K 1 d 3 Formations, respectively, and stopped generating hydrocarbons at the end of the depositional period of the late Cretaceous. Therefore, the main stage of hydrocarbon migration and accumulation was between the depositional period of the K 1 d 2 and K 1 s Formations, and the critical moment was the depositional period of the late K 1 s Formation. The generation conversion efficiency reached approximately 55% in the K 1 a 2 Formation and 18% in the K 1 t Formation at 1 School of Geosciences, China University of Petroleum, Qingdao, China 2 Beijing Ultra Do Resources Technology Inc., Beijing, China Corresponding author: Hua Liu, School of Geosciences, China University of Petroleum, 66 Changjiang West Road, Qingdao, Shandong 266580, China. Email: liuhua77@upc.edu.cn Creative Commons CC-BY: This article is distributed under the terms of the Creative Commons Attribution 4.0 License (http://www.creativecommons.org/licenses/by/4.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage).

356 Energy Exploration & Exploitation 36(3) the end of the Cretaceous sedimentary stage. In general, the effective oil traps are those reservoirs that are near the active source rock in the generating sags in the Baibei Depression. Keywords Source rock evaluation, hydrocarbon potential, active source rock, Lower Cretaceous system, Baibei Depression, Erlian Basin Introduction The Baibei Depression, which is located in the western margin of the Erlian Basin, is a secondary structural unit in the eastern part of the Chuanjing Sub-basin (Du, 2003; Zhao et al., 2010; Figure 1). The depression is part of the Early Cretaceous rift group, which was formed and developed on the Hercynian Fold Basement (Wei and Xu, 1994; Zhai et al., 2010). Previous work in the region has shown that four potential source rocks were developed: the Sahantala Formation (K 1 s), the Duhongmu Formation (K 1 d 3,K 1 d 2, and K 1 d 1 ), the Tengge er Formation (K 1 t), and the A ershan Formation (K 1 a 2 and K 1 a 1 ) (Ding et al., 2015, 2016; Gao et al., 2015; Wei and Xu, 1994; Zhai et al., 2010). However, previous studies on the source rock in the Baibei Depression have been limited to a single well (well Sc1), and only the A ershan Formation has been found to be thermally mature, with vitrinite reflectance values in the range of 0.6 0.75%R O (Gao et al., 2015). Despite continued interest and substantial exploration activities, only two out of six wildcat wells drilled in the Baibei Depression have shown fluorescence levels of oil and gas, and no well has experienced breakthrough oil flow (Gao et al., 2015). The fact that the distribution and petroleumgenerating potential of active source rocks have not been evaluated has seriously affected the exploration of hydrocarbons in the Baibei Depression. To improve this situation, we Figure 1. Tectonic location of the Baibei Depression (modified from Gao et al. (2015)).

Liu et al. 357 conducted organic petrological and geochemical analyses to study the source rock potential for hydrocarbon generation in the Baibei Depression. The essential properties for source rock evaluation described here are the quantity, quality, and thermal maturity of the organic matter, as identified by Tissot and Welte (1984). This study evaluates the source rock potential from wells Sc1, Ym2, Ym4, and Ym5. A total of 51 new core samples were studied, which belonged to the Duhongmu, Tengge er, and A ershan Formations (Table 1). Geological setting The Baibei Depression is a long and narrow EW-trending depression that covers an area of 1080 km 2 (Gao et al., 2015; Figure 1). In the west, it is adjacent to the Baiyinchagan Depression; in the south, it is adjacent to the Sanggendalai Depression and the Baiyanhua Arch; and it is bounded in the north by the Xibai Fault and is adjacent to the Bayinbaolige Uplift (Gao et al., 2015; Figure 1). The Baibei Depression is a Mesozoic sedimentary rift zone controlled by two first-order sequence boundaries, which has developed Archean-Paleozoic, Mesozoic, and Cenozoic strata (Bai et al., 2016). The generalized stratigraphy for the Baibei Depression includes deposits from the following (in descending order): the Neogene; the Paleogene; the Upper Cretaceous Erliandabusu Formation; the Lower Cretaceous Saihantala (K 1 s), Duhongmu (K 1 d), Tengge er (K 1 t), and A ershan (K 1 a) Formations; and the Paleozoic Formation (Zhao et al., 2010; Figure 2). The Lower Cretaceous system includes the major sedimentary cap and the main oil-bearing series in the area (Wang et al., 2012). The potential source rocks are mainly distributed in the K 1 s, K 1 d, K 1 t, and K 1 a 2 Formations. Methodology Dispersed organic matter in sediments is the material source of hydrocarbon generation, and organic matter richness is an important factor in determining the hydrocarbon-generating potential (Hunt, 1995; Mustapha and Abdullah, 2013). Total organic carbon (TOC), chloroform asphalt A total hydrocarbon content (HC), and hydrocarbon-generating potential (S 1 +S 2 ) are the main geochemical indicators for the evaluation of organic matter richness (Tissot and Welte, 1984). They were quantified on all samples consisting of mudstones and limy mudstones in the K 1 s, K 1 a, K 1 t, and K 1 d Formations. Different types of kerogen will produce different hydrocarbons. Generally, Type I and II kerogen commonly derived from lacustrine and marine lower plankton are the best kerogen and are capable of generating liquid hydrocarbons (Hakimi et al., 2012). On the other hand, Type III kerogen composed of terrestrial higher plants has potential to generate gas (Behar et al., 2003; Ruble et al., 2001). In this study, the type of organic matter in the analyzed source rocks is classified from the Rock-Eval pyrolysis data and petrographic analysis according to the standard of SY/T 5735-1995. Hydrocarbons are sourced from rocks only when the thermal maturity of the organic matter reaches the generation or expulsion threshold. Therefore, the thermal maturity of source rocks is an essential parameter to evaluate their hydrocarbon-generating potential. Sweeney and Burnham (1990) suggested that a source rock becomes mature when it enters the range of 0.55 1.25%R O, while Tissot and Welte (1984) used the Tmax metric to evaluate the thermal maturity of organic matter, and indicated that the top of the hydrocarbon window is when Tmax values ranged from 430 to 470 C. The thermal maturation parameter

358 Energy Exploration & Exploitation 36(3) Table 1. Results of Rock-Eval pyrolysis and TOC analysis. Sample well Lithology Form. S 1 +S 2 (mg/g) PI Tmax ( C) S3 (mg/g) PC (%) RC (%) TOC (wt%) HI Sc1 Limy mudstone K 1 d 2 1.46 0.03 430 0.88 0.17 1.22 1.39 102 Sc1 Limy mudstone K 1 d 2 0.82 0.02 427 0.53 0.11 0.96 1.07 75 Sc1 Limy mudstone K 1 d 2 1.75 0.02 429 0.58 0.20 1.17 1.37 126 Sc1 Mudstone K 1 d 1 0.27 0.01 439 0.70 0.07 0.59 0.66 41 Sc1 Mudstone K 1 d 1 0.43 0.03 441 0.70 0.09 0.94 1.03 41 Sc1 Limy mudstone K 1 d 1 0.91 0.02 435 0.57 0.10 0.56 0.66 135 Sc1 Limy mudstone K 1 d 1 1.74 0.02 443 0.69 0.18 1.04 1.22 140 Sc1 Mudstone K 1 d 1 0.00 0.00 431 0.41 0.01 0.15 0.16 0 Sc1 Mudstone K 1 d 1 0.01 0.07 433 0.36 0.01 0.17 0.18 6 Sc1 Mudstone K 1 d 1 0.31 0.03 427 0.95 0.06 0.50 0.56 54 Sc1 Mudstone K 1 t 0.14 0.09 436 0.25 0.02 0.44 0.46 28 Sc1 Mudstone K 1 t 0.09 0.04 430 0.63 0.03 0.27 0.30 30 Sc1 Mudstone K 1 t 15.70 0.02 437 0.59 1.34 1.48 2.82 546 Sc1 Limy mudstone K 1 t 0.80 0.02 438 0.60 0.09 0.41 0.50 158 Sc1 Mudstone K 1 t 0.28 0.09 442 0.35 0.04 0.56 0.60 42 Sc1 Mudstone K 1 t 0.00 0.79 436 0.61 0.02 0.20 0.22 0 Sc1 Mudstone K 1 t 0.02 0.17 421 0.24 0.01 0.32 0.33 6 Sc1 Mudstone K 1 t 0.00 0.02 442 0.69 0.02 0.17 0.19 0 Sc1 Mudstone K 1 t 0.00 0.00 / 0.80 0.02 0.18 0.20 0 Sc1 Mudstone K 1 t 2.64 0.06 436 0.21 0.23 0.89 1.12 221 Sc1 Mudstone K 1 t 0.07 0.09 437 0.31 0.01 0.21 0.22 27 Sc1 Mudstone K 1 t 0.01 0.05 439 0.47 0.01 0.14 0.15 7 Sc1 Mudstone K 1 a 2 1.08 0.06 437 0.54 0.12 0.68 0.80 126 Sc1 Limy mudstone K 1 a 2 5.40 0.08 442 0.67 0.49 0.89 1.38 360 Sc1 Limy mudstone K 1 a 2 1.80 0.12 438 0.77 0.18 0.88 1.06 149 Sc1 Limy mudstone K 1 a 2 3.42 0.07 442 0.77 0.33 1.00 1.33 241 Sc1 Limy mudstone K 1 a 2 0.28 0.10 437 0.78 0.05 0.39 0.44 57 Sc1 Limy mudstone K 1 a 2 1.56 0.10 436 0.55 0.16 0.85 1.01 139 Sc1 Limy mudstone K 1 a 2 11.49 0.09 440 0.48 1.00 1.03 2.03 517 Sc1 Limy mudstone K 1 a 2 0.81 0.07 442 0.75 0.10 0.64 0.74 101 Sc1 Limy mudstone K 1 a 2 3.22 0.10 438 0.67 0.31 1.00 1.31 222 Sc1 Limy mudstone K 1 a 2 0.81 0.08 437 0.76 0.10 0.49 0.59 127 Sc1 Limy mudstone K 1 a 2 3.49 0.11 438 0.50 0.32 0.85 1.17 267 Sc1 Limy mudstone K 1 a 2 1.94 0.09 438 0.69 0.20 0.42 0.62 284 Sc1 Limy mudstone K 1 a 2 10.71 0.06 439 0.52 0.94 1.01 1.95 515 Sc1 Limy mudstone K 1 a 2 6.54 0.09 441 0.58 0.58 0.53 1.11 538 Sc1 Mudstone K 1 a 2 0.07 0.39 444 0.50 0.03 0.37 0.40 10 Sc1 Mudstone K 1 a 2 0.00 1.00 / 0.31 0.01 0.19 0.20 0 Sc1 Mudstone K 1 a 2 0.00 0.62 436 0.32 0.01 0.21 0.22 0 Ym2 Limy mudstone K 1 a 2 16.69 0.01 437 1.11 1.45 1.46 2.91 567 Ym2 Limy mudstone K 1 a 2 4.94 0.01 437 1.16 0.48 0.80 1.28 383 Ym4 Mudstone K 1 a 2 16.72 0.02 426 1.19 1.46 1.71 3.17 519 Ym5 Mudstone K 1 a 2 0.00 0.00 409 1.06 0.03 0.13 0.16 0 Ym5 Mudstone K 1 a 2 0.00 0.00 515 1.07 0.03 0.14 0.17 0 Ym5 Mudstone K 1 a 2 0.00 0.00 / 0.73 0.02 0.03 0.05 0 (continued)

Liu et al. 359 Table 1. Continued Sample well Lithology Form. S 1 +S 2 (mg/g) PI Tmax ( C) S3 (mg/g) PC (%) RC (%) TOC (wt%) HI Ym5 Mudstone K 1 a 2 0.00 0.11 422 0.75 0.02 0.09 0.11 0 Ym5 Mudstone K 1 a 2 0.00 0.00 / 0.17 0.01 0.04 0.05 0 Ym5 Mudstone K 1 a 2 0.00 0.00 / 0.16 0.00 0.04 0.04 0 Ym5 Mudstone K 1 a 2 0.00 0.00 / 0.13 0.00 0.03 0.03 0 Ym5 Mudstone K 1 a 2 0.00 0.00 / 0.07 0.00 0.01 0.01 0 Ym5 Mudstone K 1 a 2 0.00 1.00 / 0.02 0.00 0.03 0.03 0 Form.: Formation; HC: total hydrocarbon (ppm); HI: hydrogen index; PC: pyrolytic carbon (%); PI: productivity index; RC: residual carbon (%); TOC: total organic carbon (wt%). used in this study was vitrinite reflectance, and the mature threshold of organic matter was identified as 0.5%R O in the Baibei Depression. To evaluate the source rock potential in the Baibei Depression, the geochemical analysis was carried out to determine the TOC content and to provide kerogen analysis, Rock-Eval pyrolysis (Tissot and Welte, 1984). A total of 121 core samples were screened by TOC analysis and Rock-Eval pyrolysis, including 70 test samples provided by the Yanchang Oilfield Company and 51 new testing samples from 2014. TOC and Rock-Eval pyrolysis were performed on 100 mg crushed rock samples, which were heated to 600 C in a helium atmosphere, using a Rock-Eval II instrument equipped with a TOC module. Selected mudstone samples were extracted using the traditional Soxhlet extraction method to determine the percentage content of bitumen and oil. For the microscopic petrography analysis, selected mudstone/limy mudstone samples were crushed and embedded in liquid epoxy resin. Samples were then progressively ground with an abrasive powder and polished to obtain a smooth surface. Microscopic analysis was carried out using a Leica CTR6000 M Photometry Microscope under reflected white light and ultraviolet light. Vitrinite reflectance (R O ) analysis was carried out using the Diskus FOSSIL software with the maceral module under a 50X oil immersion objective. The Schlumberger s PetroMod one-dimensional modeling software was used to reconstruct the thermal histories of the studied wells. The maturity of the source rocks and the timing of hydrocarbon generation in the basin were also modeled according to the workflow from Abdalla et al. (1999). R O is the most widely used indicator for assessing source rock maturity (Allen and Allen, 1990). Hydrocarbon generation modeling is based on the TOC and hydrogen index (HI) of the source rocks and maturity modeling is based on the EASY% R O model created by Sweeney and Burnham (1990). Evaluation of source rocks Organic matter richness, type, and thermal maturity are the three main geochemical indicators for the evaluation of source rocks (Tissot and Welte, 1984). The initial evaluation of the source rock from well Sc1 in 2006 was augmented by 51 new samples from five wells to ensure the reliability of the source rock evaluation.

360 Energy Exploration & Exploitation 36(3) Figure 2. Generalized stratigraphy of the Cretaceous system in the Baibei Depression. Possible source rocks are marked. Form. is an abbreviation of formation. Organic matter richness and source rock quality The results of the evaluation of these criteria for organic matter richness are shown in Table 2. Previous work has shown that the K 1 a, K 1 t, K 1 d, and K 1 s Formations

Liu et al. 361 Table 2. Evaluation criteria of organic matter richness for mudstone source rock. Organic matter richness parameters Hydrocarbon source rock grade TOC (wt%) Chloroform asphalt (wt%) HC (ppm) S 1 +S 2 (mg/g) None <0.4 <0.015 <100 Poor 0.4 0.6 0.015 0.05 100 200 <2.0 Fair 0.6 1.0 0.05 0.1 200 500 2.0 6.0 Good 1.0 2.0 0.1 0.2 500 1000 6.0 20 Very good >2.0 >0.2 >1000 >20 HC: total hydrocarbon (ppm); TOC: total organic carbon (wt%); S 1 : free hydrocarbons (mghc/g rock); S 2 : kerogen generation capability (mghc/g rock). Table 3. Evaluation results of organic matter richness and source rock quality in the Baibei Depression. Formation Sample number TOC (wt%) S 1 +S 2 (mg/g) Range Average Evaluation Range Average Evaluation Result K 1 s 12 1 3 2.11 Very good 2 15 7.9 Good Good to very good K 1 d 3 16 1 7 3.07 Very good 2 40 15.5 Good Good to very good K 1 d 2 9 1.07 7 2.72 Very good 0.82 35 14 Good Good to very good K 1 d 1 13 0.16 4 1.17 Good 0.27 16 3 Fair Fair to good k 1 t 12 0.15 2.82 1.19 Good 2 15.7 3.8 Fair Fair to good K 1 a 32 1 3.17 1.32 Good 2 16 5.1 Fair Fair to good TOC: total organic carbon (wt%); S 1 : free hydrocarbons (mghc/g rock); S 2 : kerogen generation capability (mghc/g rock). (Figure 2) are good source rocks with high organic matter richness (Gao et al., 2015), but the new samples reveal differences between these formations. The ranges of TOC values and S 1 +S 2 values of each formation are shown in Table 3. All of these values indicate source rocks with good to very good generative potential. The K 1 s Formation possesses high TOC values (1 3%) and S 1 +S 2 values (2 15 mg/g), and the TOC and S 1 +S 2 contents in the samples from the K 1 a Formation through the K 1 s Formation show richness in organic matter and a hydrocarbon-generating potential that gradually increases in the sequence. Based on the evaluation criteria of organic matter richness for mudstone source rock (Table 2), the K 1 a 2 Formation through the K 1 d 1 Formation are source rocks with fair to good generative potential, and the K 1 d 2 Formation through the K 1 s Formation are source rocks with good to very good generative potential (Table 3). Figure 3 shows that the source rocks in different formations in the Baibei Depression are heterogeneous in terms of their HI, S 1 +S 2, TOC, and HC contents. The quality of the Cretaceous source rocks increases gradually from bottom to top, but some samples are anomalously poor in organic matter richness, which we assume to be caused by prolonged oxidation near unconformities (Figure 3).

362 Energy Exploration & Exploitation 36(3) Figure 3. (a) Crossplot diagram of HI versus S 1 +S 2 that shows the hydrocarbon-generating potential of all source rock samples in the Baibei Depression and (b) crossplot diagram of TOC versus HC that shows the hydrocarbon-generating potential of all source rock samples in the Baibei Depression (modified from Li et al. (2016)). Kerogen type The old data showed that the H/C and O/C values of organic matter in well Sc1 were generally high, and the evaluations suggested that the organic matter in the K 1 a 2 Formation was mainly Type I kerogen; the organic matter in the K 1 t, K 1 d, and K 1 s Formations was mainly Type II kerogen; and a small amount of Type III kerogen appeared in the K 1 t Formation (Figure 4). However, the new data indicated that the H/C and O/C values in these source rock intervals were generally lower than in the old data. Based on the evolutionary trend of kerogen shown in the Van Krevelen diagram (Tissot and Welte, 1984), the new samples seem to be more mature than the old ones (Figure 4). The evaluation results showed that the organic matter in the K 1 a 2 Formation is Type I and II kerogen, the organic matter in the K 1 t Formation is mainly Type III, and the organic matter in the K 1 d Formation is mainly Type II. Almost all the organic matter in the samples from well Sc1 were dominated by phytane; they possessed low Pr/Ph values (mostly below 0.8) and contained relatively high concentrations of gammacerane, as indicated by them having g/c 30 H values over 0.2 (Figure 5(a)). From this, we can conclude that most of the analyzed samples were deposited in a saline environment. The samples from the K 1 d 1 and K 1 s Formations were slightly lower in gammacerane content, suggesting a slightly brackish depositional environment. A crossplot diagram of Pr/nC 17 versus Ph/nC 18 shows the sedimentary environment and material source for all the analyzed samples (Figure 5(b)). Most of the samples were deposited in a reducing environment and were sourced from marine algae, making them oil prone, while a few samples were sourced from woody materials and are considered to be gas-prone source rocks. Thermal maturity Results of vitrinite reflectance from the mudstone and coal samples are shown in Figure 6. The vitrinite reflectance value for the K 1 s Formation ranged between 0.4 and 0.5%R O, that

Liu et al. 363 Figure 4. Crossplot diagrams of H/C versus O/C that show the kerogen type of all analyzed source rock samples in the Baibei Depression. Red represents previous data (2006). Green represents new data (2014). Figure 5. Discrimination diagrams for sedimentary environments of source rocks in the Baibei Depression. (a) Crossplot diagram of Pr/Ph versus g/c 30 H(g ¼ Gammacerane, H ¼ Hopane) and (b) crossplot diagram of Pr/nC 17 versus Ph/nC 18 that shows the sedimentary environment of all analyzed samples. for the K 1 d Formation ranged between 0.43 and 0.58%R O, that for the K 1 t Formation ranged between 0.5 and 0.68%R O, and that for the K 1 a 2 Formation was over 0.72%R O. Tmax data were in agreement with these S 1 +S 2 values, as shown in Table 1. All the samples possessed low to high S 1 +S 2 values (0 700 mg/g TOC), low Tmax values (420 448 C), and

364 Energy Exploration & Exploitation 36(3) Figure 6. Depth plots of R O for well Sc1 in the Baibei Depression. quite a narrow range for vitrinite reflectance (0.5 0.9%R O ). Therefore, most of the analyzed samples are interpreted to be partially mature or mature for hydrocarbon generation, but they fail to reach the oil generation peak. The new relationship between the measured vitrinite reflectance and depth for well Sc1 indicates that source rocks from the Saihantala Formation to the K 1 d 3 Formation are immature, those from the K 1 d 2 Formation to the upper Tengge er Formation are partially mature and only generated a small amount of oil, and those from the lower Tengge er Formation are mature and began to generate a large amount of oil. The source rocks of the K 1 a 2 Formation are in the large-scale hydrocarbon generation stage, but it fails to reach the oil generation peak. The thermal maturities of the two evaluations are similar, but the newly measured data change the evolutionary trend of the thermal maturity because most of the new samples were calcareous mudstone, whereas the old samples were mudstone. These results are summarized in a crossplot diagram of the hydrocarbon-generating potential versus the Tmax values, which shows that the source rocks in the study area have entered the mature stage and all of the R O values were less than 1.0 (Figure 6). Figure 7 indicates that the upper Tengge er Formation is immature to partially mature, while the lower Tengge er Formation and the second member of the A ershan Formation are mature and began to generate a large amount of oil. The low maturity of the K 1 a 2 Formation was previously shown by the old data, while the new data show an increase in the thermal maturity of the source rocks, resulting in higher hydrocarbongenerating potential than the previous evaluation. Consequently, we can consider the beginning of the oil window to be 1650 m deep in the Sc1 well based on all the sample values.

Liu et al. 365 Figure 7. (a) Depth plots of chloroform asphalt A conversion rates from different evaluations and (b) depth plots of HC conversion rates from different evaluations. HC is the total hydrocarbon content (modified from Li et al. (2016)). Distribution of active source rock Active source rock is the rock that not only generates hydrocarbon but also expulses hydrocarbon, which controls the distribution of hydrocarbon in a basin (Jin, 2001). Identification of active source rock In terms of the relationship between TOC and S 1 +S 2 or HI, the limit of organic content for active source rocks from the Lower Cretaceous system in the Baibei Depression is determined to be 1.0% TOC. Based on this limitation, it is likely that most of the samples from the lower Saihantala, K 1 d 3,K 1 d 2, the middle part of K 1 t, and the upper-middle part of K 1 a 2 Formations could all become potential source rocks, as they contain more than 1.0%TOC (Figure 8). Combined with the vitrinite reflectance values, most of the analyzed samples from the K 1 s and K 1 d Formations are interpreted to be too immature for hydrocarbon generation. Some samples from the middle of the Tengge er Formation show early maturity for hydrocarbon generation, with 0.7%R O, and samples from the K 1 a 2 Formation show full maturity for hydrocarbon generation, with 0.9%R O. The active source rock in our study area is confirmed to be the middle part of the K 1 t and K 1 a 2 Formations. Distribution of active source rock Based on the model of a petroleum system, the distribution of active source rock plays a vital role in the development of oil and gas exploration strategies. The relationship between vitrinite reflectance and burial depth of source rocks in the Erlian Basin was found to be complicated in previous research (Chen, 2011; Gao et al., 2015; Wang et al., 2012; Zhai et al., 2010; Zhao et al., 2015). These difficulties arose because of uplifting and erosion of strata at the end of the lower Cretaceous sedimentation period at almost all locations in the Erlian

366 Energy Exploration & Exploitation 36(3) Figure 8. Synthetic evaluation of geochemical section from well Sc1 in the Baibei Depression. Form. is an abbreviation of formation. Basin (Zhao et al., 2015). The Ro H trend line of well Sc1 data shows that the vitrinite reflectance of the surface was not zero and was higher than 0.2%R O (Figure 6), which indicates that the thickness of rock eroded at the end of the Cretaceous sedimentation period was larger than that of the subsequent deposits. Based on the R O trend back to zero method reported by Burnham and Sweeney (1989), we estimate the eroded thickness in the Baibei Depression to be 600 700 m. The expression of R O H at the maximum burial depth is Ro ¼ 0:2667 Expð0:00046 HmÞ ð1þ Hm ¼ Hp Hh þ He ð2þ

Liu et al. 367 Figure 9. N S-oriented profile over well Sc1 in the Baibei Depression, which shows the Ro values of source rocks of the Lower Cretaceous system. Figure 10. N S-oriented profile over well Ym2 in the Baibei Depression, which shows the Ro values of source rocks of the Lower Cretaceous system. where Hm is the maximum buried depth, Hp is the present burial depth, Hh is the thickness above the K 1 s Formation, and He is the eroded thickness. According to the values of eroded thickness, the maturation of the source rock can be calculated using equations (1) and (2). Figures 9 and 10 show the N S-oriented profile of R O values, which indicates that the K 1 a 2 Formation in the center of Baibei Depression has entered the oil generation peak stage. As can be observed in Figure 10, the largest vitrinite reflectance is above 2.0%R O to the south of well Ym2, and the maximum buried source rocks of the K 1 a Formation have entered the dry gas generation stage. Meanwhile, the source rock in the Tengge er Formation has entered the oil generation peak stage only in the northern part of the profile, and its highest R O value is 1.1%R O, while the southern source rock only qualifies for the low maturation stage (Figure 9). In general, the active source rocks in the Baibei Depression fall along a narrow belt that experienced the necessary oil generation conditions. Hydrocarbon generation In the Baibei Depression, hydrocarbon generation evolution and intensity have been studied based on basin modeling and thermal simulation. The hydrocarbon generation evolution history of well Sc1 is shown in Figure 11, which demonstrates that the maximum burial depth was at the end stage of the Cretaceous sedimentation. The depression was uplifted and eroded during the late Cretaceous, and then entered another subsidence stage before the

368 Energy Exploration & Exploitation 36(3) Figure 11. Hydrocarbon generation conversion efficiency by basin modeling in the Baibei Depression. early Paleogene. The subsequent deposition was less than the eroded thickness of the strata, so the paleogeotemperature was higher than the present geotemperature. The source rock of the K 1 a 2 Formation crossed the threshold for hydrocarbon generation during the deposition of the K 1 d 2 strata, reached the mature stage during the deposition of the upper Cretaceous, and reached its highest R O value during the deposition of the Lower Cretaceous. However, hydrocarbon generation stagnated just when it entered the peak stage of hydrocarbon generation because of tectonic uplift and erosion in the Baibei Depression. As the major oil source rock of the study area, the organic matter of the K 1 a 2 Formation began to generate hydrocarbons during the deposition of the K 1 d 2 strata and its hydrocarbon generation efficiency reached 55% at the end of the Cretaceous deposition stage. While the Tengge er Formation began to generate hydrocarbons during the deposition of the K 1 d 3 strata, with a generation conversion efficiency that reached approximately 18% at the end of the Cretaceous deposition period; thus, it can be considered as a minor oil source rock in the Baibei Depression (Figure 11). Our analysis has revealed that the K 1 a 2 and K 1 t Formations in the Baibei Depression are the two main active source rocks, but their hydrocarbon generation intensities are different. They reached their peak hydrocarbon generation during the deposition of K 1 s, after which their hydrocarbon generation stagnated due to erosion. As a result, the hydrocarbon generation intensity at the end of the Saihantala Formation deposition period was the most significant. According to the results of the basin modeling, the hydrocarbon generation intensity map of the two main source rocks is as illustrated in Figures 12 and 13. Generally, if oil generation intensity is higher than 200 10 4 t/km 2, the oil can be effectively expelled and accumulate into reservoirs to form oil fields. The R O value of the source rocks is closely related to their thickness and is therefore considered to indicate the hydrocarbon-generating sag. Figure 12 shows that there are two oil-generating sags for the K 2 a 2 Formation during the deposition at the end of the K 1 s Formation, which have oil generation

Liu et al. 369 Figure 12. Isogram chart of oil generation intensity for K 1 a 2 source rock interval in the Baibei Depression. Figure 13. Isogram chart of oil generation intensity for K 1 t source rock interval in the Baibei Depression. intensities higher than the 200 10 4 t/km 2 threshold in the Sc1 and Ym4 Sags. However, the oil generation intensity for the K 1 t Formation suggests that the Sc1 Sag reached the criteria to generate oil at the end of the deposition of the K 1 s Formation, while the Ym4 Sag in the eastern depression had a poorer ability to generate oil (Figure 13).

370 Energy Exploration & Exploitation 36(3) Discussion In accordance with the analysis of the hydrocarbon-generating ability of source rocks, the main hydrocarbon generation and distribution models for the different structural units of the Baibei Depression are distinct. The Ym4 Sag had one active source rock interval (K 2 a 2 Formation) and only reached a low maturation stage (Figures 12 and 13). Therefore, the main active kitchen exists in the northeastern part of the Baibei Depression, and hydrocarbons migrate a short distance because of the relatively low hydrocarbon generation ability. Therefore, the valid trap is the lithologic reservoirs around or in the active source rock interval. Meanwhile, the Sc1 Sag has two active source rock intervals (K 2 a 2 and K 1 s Formations) and reached a high maturation stage. Therefore, the migration and accumulation of hydrocarbons in the southwestern part of the Baibei Depression is active and plentiful. However, the Baibei Depression was entirely uplifted in the late Mesozoic, which may have resulted in the loss of hydrocarbons in some traps with weak preservation such as faulted traps. However, the lithologic traps in the hydrocarbon-generating sags are good for oil and gas filling and preservation because of their earlier formation time and superior preservation condition with surrounding mudstone. Therefore, those lithologic traps surrounding or in active hydrocarbon generation sags are advantageous to target for oil reservoir exploration, as they were less influenced by later tectonic movement in the Baibei Depression. Conclusions Four source rock intervals developed in the Lower Cretaceous system possess generally fair to occasionally good hydrocarbon-generating potential in the Baibei Depression. The source rock samples possess variable TOC and S 1 +S 2 contents and indicate that the K 1 a 2 Formation through the K 1 d 1 Formation are source rocks that have fair to good generative potential, while the K 1 d 2 Formation through the K 1 s Formation are source rocks that have good to very good generative potential. Most of the analyzed samples were deposited in reducing environments and were sourced from marine algae and are therefore oil prone. Most of the samples in the lower K 1 t and K 1 a 2 Formations are thermally mature but did not reach the oil-generating peak. The source rock began to generate hydrocarbons mainly during the deposition of the K 1 d 2 and K 1 s Formations but stopped generating hydrocarbons at the end of the late Cretaceous deposition. There are two generation sags developed in the Baibei Depression, and the source rock in the Sc1 Sag is more mature. Therefore, the migration and accumulation of hydrocarbon in the southwestern part of the Baibei Depression should be active and plentiful. Acknowledgements We would like to thank the Geological Scientific Research Institute of Huabei Oilfield for their contribution of geological data and valuable discussion. Declaration of conflicting interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Liu et al. 371 Funding The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by a grant by the Important National Science & Technology Specific Projects (Grant No. 2016ZX05006-003) and the Strategic Priority Research Program of the Chinese Academy of Sciences (Grant No. XDA14010305). References Abdalla YM, Michael JP, William AA, et al. (1999) Modeling petroleum generation in the southern Muglad rift basin, Sudan. AAPG Bulletin 83: 1943 1964. Allen PA and Allen JR (1990) Basin analysis principles and applications, Vol. 82. Oxford: Blackwell Scientific Publications. Bai XY, Liu HC, Li KS, et al. (2016) Geologic characteristics and petroleum exploration potential of Baibei Sag in Erlian Basin. Journal of Yanan University (Natural Science Edition) 35(1): 89 92. Behar F, Lewan MD, Lorant F, et al. (2003) Comparison of artificial maturation of lignite in hydrous and nonhydrous conditions. Organic Geochemistry 34: 575 600. Burnham AK and Sweeney JJ (1989) A chemical kinetic model of vitrinite maturation and reflectance. Geochimica et Cosmochimica Acta 53(10): 2649 2657. Chen Y (2011) Research of hydrocarbon accumulation conditions of the Lower Cretaceous Sihantala Sag in Erlian Basin. Master Thesis, China University of Petroleum (Beijing), Beijing, China (in Chinese). Ding XJ, Liu GD, Zha M, et al. (2015) Characteristics and origin of lacustrine source rocks in the Lower Cretaceous, Erlian Basin, northern China Erlian Basin, northern China. Marine and Petroleum Geology 66: 939 955. Ding XJ, Liu GD, Zha M, et al. (2016) Geochemical characterization and depositional environment of source rocks of small fault basin in Erlian Basin, northern China. Marine and Petroleum Geology 69: 231 240. Du JH (2003) Exploration of Concealed Reservoir in Erlian Basin. Beijing: Publishing House of Oil Industry. Gao YW, Chen ZJ, Yu J, et al. (2015) Hydrocarbon accumulation conditions research and favorable zones prediction of Baibei sag in Erlian Basin. Petroleum Geology and Engineering 29(4): 38 41. Hakimi MH, Abdullah WH and Shalaby MR (2012) Geochemical and petrographic characterization of organic matter in the Upper Jurassic Madbi shale succession (Masila Basin, Yemen): Origin, type and preservation. Organic Geochemistry 49: 18 19. Hunt JM (1995) Petroleum Geochemistry and Geology, 2nd ed. San Francisco, CA: Freeman. Jin Q (2001) Importance and study of effective source rocks. Petroleum Geology and Recovery Efficiency 8(1): 1 4. Li JH, Yu L and Shuai J (2016) The geochemical characteristics of source rocks in Baibei Depression of Erlian Basin. Ground Water 38(3): 248 250. Mustapha KA and Abdullah WH (2013) Petroleum source rock evaluation of the Sebahat and Ganduman Formations, Dent Peninsula, Eastern Sabah, Malaysia. Journal of Asian Earth Sciences 76: 346 355. Ruble TE, Lewan MD and Philp RP (2001) New insights on the Green River petroleum system in the Uinta basin from hydrous pyrolysis experiments. AAPG Bulletin 85: 1333 1371. Sweeney JJ and Burnham AK (1990) Evaluation of a simple model of vitrinite reflectance based on chemical kinetics. AAPG Bulletin 74: 1559 1570. Tissot BP and Welte DH (1984) Petroleum Formation and Occurrence. New York: Springer.

372 Energy Exploration & Exploitation 36(3) Wang SX, Zhang ZH, Zhang ZP, et al. (2012) Geochemical characteristics and oil-source correlation of the Jurassic source rocks in the Erlian Basin. Bulletin of Mineralogy, Petrology and Geochemistry 31(4): 396 403. Wei KS and Xu HD (1994) Sequence stratigraphic features of cretaceous nonmarine sediments in Erlian Basin. Earth Science Journal of China University of Geosciences 19(2): 181 193. Zhai ZJ, Wang RT, Xiao LP, et al. (2010) Evaluation of source rocks in the Baiyinchagan sag. Journal of Oil and Gas Technology 32(3): 170 172. Zhao XZ, Jiang SQ, Dan WN, et al. (2010) Petroleum geologic characteristics of Artala Sag in Erlian Basin. Lithologic Reservoirs 22: 12 17. Zhao XZ, Liu GD, Jin FM, et al. (2015) Distribution features and pattern of effective source rock in small faulted lacustrine basin: A case study of the Lower Cretaceous in Erlian Basin. Acta Petroleum Sinica 36(6): 641 652.