Catalytic Hydrogenation of Post-Mature Hydrocarbon Source Rocks Under Deep-Derived Fluids: An Example of Early Cambrian Yurtus Formation, Tarim Basin, NW China

As a link between the internal and external basin, the deep derived fluids play a key role during the processes of hydrocarbon (HC) formation and accumulation in the form of organic-inorganic interaction. Two questions remain to be answered: How do deep-derived fluids affect HC generation in source rocks by carrying a large amount of matter and energy, especially in post-mature source rocks with weak HC generation capability? Can hydrogen and catalysts from deep sources significantly increase the HC generation potential of the source rock? In this study, we selected the post-mature kerogen samples of the early Cambrian Yurtus Formation in the Tarim Basin of China. Under the catalytic environment of ZnCl2 and MoS2, closed system gold tube thermal simulation experiments were conducted to quantitatively verify the contribution of catalytic hydrogenation to "HC promotion" by adding H2. The catalytic hydrogenation increased the kerogen HC generation capacity by 1.4–2.1 times. The catalytic hydrogenation intensity reaction increased with temperature. The drying coefficient of the generated gas decreased significantly as the increasing yield of heavy HC gas. In the simulation experiment, alkane δ13C becomes lighter after the catalytic hydrogenation experiment, while δ13CCO2 becomes heavier. In the process of catalytic hydrogenation, the number of gaseous products catalyzed by ZnCl2 is higher than that catalyzed by MoS2 under the same conditions, indicating that ZnCl2 is a better catalyst for the generation of gaseous yield. Meanwhile, Fischer-Tropsch synthesis (FFT) reaction was happened in the catalytic hydrogenation process. The simulation experiment demonstrates that hydrogen-rich components and metal elements in deep-derived fluids have significant catalytic hydrogenation effects on organic-rich matter, which improved the HC generation efficiency of post-mature source rocks.


INTRODUCTION
The deep-derived fluids refer to the mantle source volatile fluid below the basement of the sedimentary basin, which is originated from the dehydrating fluid generated in the process of plate subduction or deep metamorphism, and the deep circulating fluid driven by the heat from the mantle or the deep shell source (Jin et al., 2002;Jin et al., 2004;Jin et al., 2007;. A large amount of gas components carried by the deep-derived fluids contain volatile materials, such as H 2 , CH 4 , H 2 S, CO 2 , N 2 , and He. As one of the main volatile fluids, the distribution of geologic H 2 (molecular hydrogen) is widespread in ocean ridges, active tectonic belts, and the Precambrian or Cenozoic strata distributed on land (Jeffrey and Kaplan, 1998;Allen and Seyfried, 2004;Newell et al., 2007;Tivey, 2007;Bradley and Summons, 2010;Sherwood-Lollar et al., 2014;Meng et al., 2015;Etiope, 2017;Guélard et al., 2017;Etiope and Whiticar, 2019;Bougault et al., 2019;Klein et al., 2019;Klein et al., 2020). H 2 has a variety of genetic types, including the release of deep magma degassing, the serpentinization of ultrabasic-basic rocks (Sherwood-Lollar et al., 2014;Etiope, 2017;Klein et al., 2019;Klein et al., 2020), radiolysis of water due to the presence of radioactive minerals (Sherwood-Lollar et al., 2014;Milesi et al., 2016), mechanical cracking of Si-bearing rock (Hirose et al., 2011;Telling et al., 2015), and the reduction of organic matter by microorganisms (Shuai, et al., 2010). The deepderived fluids not only carry energy and hydrogen-rich matter, but a large amount of metal catalysts and other elements such as Mg, Fe, Mn, Ni, Zn, Mo, and Cu (Coveney, 1987;Pinto et al., 1999;Tivey, 2007;Proskurowski et al., 2008;Resing et al., 2015). Meanwhile, in sedimentary basins the enrichment of organic matter is normally accompanied by abundant metal elements (e.g., Fe, Mn, Zn, and Mo) (Lv et al., 2018).
The geological kerogen catalytic hydrogenation reaction is a reduction reaction that generates new products and is accompanied by energy conversion. To study the contribution of material conditions in organic-inorganic interactions to hydrocarbon (HC) generation from organic matter, previous studies examined different external hydrogen sources and catalysts (Hawkes, 1972;Jin et al., 2002;Jin et al, 2004). As the energy of the H-H bond is theoretically 436 kJ mol −1 , which is less than the H-OH bond energy of 497 kJ mol −1 (Luo, 2004), H 2 is easier to hydrogenate organic matter than H 2 O under simulated experimental conditions (Jin et al., 2004). Lewan et al. (1979), Lewan, (1997)and Seewald (2003) confirmed that the contribution of exogenous H plays a key role in HC generation and distribution of underground HC components. Mango et al. (1996Mango et al. ( , 1997 showed in an open system that the reaction of exogenous hydrogen with organic matter promotes the rate of HC generation under the action of transition metals. Meng et al. (2015) confirmed that deep hydrogenrich fluids can activate and increase the HC regeneration of ancient source rocks through experiments. Liu et al. (2016) showed that there is Fischer-Tropsch-type (FTT) interaction between H 2 and CO 2 produced by the thermal alteration of source rocks in the geological environment to synthesize HCs.
These deep-derived fluids can hydrogenate post-mature kerogen, thereby activating kerogen's ability to regenerate HCs, by activating the original inert carbon in kerogen and generating HCs (Meng et al., 2015). The post-mature kerogen is rich in polycyclic aromatic HC compounds and polymer groups and has a low H/C ratio. H is a hydrocarbongenerating restrictive reactant for high-maturity kerogen. The hydrocracking reaction can hydro-isomerize and crack (including ring-opening) aromatic compounds and macromolecular groups, i.e., break long-chain alkanes into short-chain alkanes, so that aromatic or naphthenic HCs can be opened to form long-chain alkanes. The hydrocracking reaction can further break the chain (Han, 2001) to promote the reactivation of kerogen to generate HCs Liu et al., 2019). In order to simulate the geological regularities of catalytic hydrogenation more accurately in a more consistent environment with deep-derived fluids, on the basis of previous studies, we further optimized the hydrogen quantification in the simulation experiment. The main objectives of this work are to examine the potential of HC generation in high-maturity marine I-type kerogen in the quantitative hydrogenation process, which involves different catalysts under closed system conditions. Meanwhile, it provided to understand the characteristics of HC isotope fractionation of alkane gas under the influence of deepderived fluids; and to provide theoretical support for the evaluation of organic HC generation.

Experimental Samples
In this study, kerogen was selected from the argillaceous source rocks of the early Cambrian Yurtus Formation in the Dongergou section of the Tarim Basin. Kerogen (Ker) was type I with the vitrinite reflectance (Ro) of 1.91%. The experimental samples were approximately 200 meshes. Total organic carbon (TOC) content of kerogen is 64.0% with purification, atomic ratio H/C 0.87, T max 530°C, pyrolysis parameter S 1 1.43 mg HC/g ker, S 2 10.95 mg HC/g ker, S 3 16.17 mg CO 2 /g ker, hydrocarbon index (HI) 18, oxygen index (OI) 28. The catalysts used in this study, ZnCl 2 and MoS 2 , had a particle size of 200 mesh and met the experimental analytical purity requirements. Deionized water was prepared in the laboratory, and H 2 was high-purity hydrogen gas containing 5% internal standard helium (He).

Experimental Method
The simulation experiment was conducted in the State Key Laboratory of Organic Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences in a closed system composed of gold tubes. During the sample loading process, the loading method of solid-liquid samples described in detail by Liu and Tang (1998) was used. In short, we added the weighed kerogen powder, deionized water, and catalyst powder (10:10: 1, wt: wt: wt) into a gold tube (6 mm inner diameter and 70 mm length), which was welded and sealed for cleaning, placed the tube in an argon atmosphere for 25 min to replace the air in the tube, pinched the wall of the gold tube with manual pliers, and sealed the top port with argon arc welding to complete the loading steps. For the hydrogenation sample loading, we made sure precise quantitative volume H 2 injected into a larger gold tube. The tube (10 mm inner diameter, 120 mm long) was used. The pre-solid and liquid reactant injection method was the same as above: when quantitatively injecting H 2 (containing 5% He as the internal standard gas), temperature control was carried out under room temperature (25°C) with two high-sensitivity pressure gauges connected in series to ensure the amount of H 2 injection using the principle of hydrogen gas replacement to effectively eliminate air interference. The mass of the reactants added to the tube was 40 mg kerogen powder, 40 mg deionized water, 4 mg catalyst powder, and 0.65 mg H 2 . Then we pressed the top port of the gold tube with a pressure clamp to seal the hydrogen component inside the gold tube and flattened the port with the pressure clamp after the wall of the gold tube was welded by argon arc welding. Next, the gold tube was placed in hot water (>60°C) without beadshaped bubbles overflowing to ensure gas tightness in the tube.
After the sample loading of gold tubes was completed, the tubes were placed in the high-temperature reaction kettle set for heating. The temperature was first raised in 2 h to reach each predetermined temperature point, and then constant heating was performed for 72 h. There were five groups of experimental samples with each set at six temperature points, 350°C, 375°C, 400°C, 450°C, 500°C, and 550°C. Group I is kerogen + deionized water, group II is kerogen + deionized water + ZnCl 2 , group III is kerogen + deionized water + MoS 2 , group IV is kerogen + deionized water + ZnCl 2 +H 2 , group V is kerogen + deionized water + MoS 2 +H 2 . The simulated experimental pressure was constant at 500 bar, and the abnormal pressure change in the kettle did not exceed 1 bar.

Experimental Analysis
Following the research method of Pan et al. (2006) and Jia et al. (2014), the gold tube was removed and carefully placed into a custom-made vacuum glass tube. The vacuum tube was connected to an Agilent 6890 N full-component gas chromatograph (GC), which was modified by Wasson ECE Instrumentation for determination of gas molecular composition. The GC contains three detection channels, one flame ionization detector (FID), and two thermal conductivity detectors (TCD). N 2 was the carrier gases for the organic detection channel (FID) and inorganic channel (TCD1), and He was the carrier gas for hydrogen and helium detection channel (TCD2). The detection column was a Paraplot Q-type capillary column. The vacuum environment of the system was maintained by connecting a vacuum pump in the glass tube and piercing the gold tube inside to release the gas component. There are two valves connected in series to the vacuum tube and the chromatograph, which are used to measure the pressure change of the gas component and release part of the gas into the chromatograph to measure its molecular composition. We determined the type and composition of the gas products through different detection response values and corresponding time points. The method used the peak areas of different detection values of FID and TCD to complete the quantification of different gaseous products generated by the peak area ratio of the known quantitative standard gas (Liu and Tang, 1998).
After the GC detection analysis, a gas-tight syringe was used to extract a small amount of gas was extracted from the vacuum tube using a compact gas sampling needle for the carbon and hydrogen isotope analysis. The ratios of carbon and hydrogen isotopes refer to the Vienna Peedee belemnite standard (VPDB). The carbon isotope ratio detection and analysis were performed using a GV Isoprime IRMS interfaced with an Agilent 6890 N-Isoprime 100 instrument. The detection value of each sample was measured two to three times. If there was no obvious difference between the previous two detection values, the average value was taken as the same position value of the sample. If the difference between the first two test values was large, the third test was required, and the average value of the two adjacent values in three tests was taken as the final sample isotope composition. The carbon isotope analysis accuracy of this equipment was within ±0.4‰. Hydrogen isotope ratio detection was performed using the Thermo Trace GC 1310-Delta V Advantage instrument. The detection method was consistent with the carbon isotope analysis, and the hydrogen isotope accuracy was within ±3‰. During the detection process, the pre-determined δ 13 C value calibration gases were periodically analyzed to check the accuracy of the measured value.

Total Gas Yield
The total gas yield of different simulated experimental samples showed an upward trend with increasing temperature ( Table 1). The total gas yield of group I was measured at the initial temperature of 350°C at 56.6 m 3 /t Ker, and the yield increased substantially after 400°C. At the highest temperature of 550°C, the yield reached a maximum value of 241.8 m 3 /t Ker. After the catalysts ZnCl 2 and MoS 2 were added to groups II and III, the total gas yield increased from the lowest values of 70.0 m 3 /t Ker (350°C) and 70.5 m 3 /t Ker (350°C), respectively, and began to increase significantly at 400°C. The maximum yields were 513.4 m 3 /t Ker and 347.5 m 3 /t Ker, which were 2.1 to 1.4 times the maximum value in group I, respectively. The addition of ZnCl 2 and MoS 2 significantly promoted the reaction, and the yield increased more by the addition of ZnCl 2 than MoS 2 , and the gas increase was mainly from CH 4 and CO 2 . In groups IV and V, the total gas yields (without accounting for H 2 yield) were 55.7 m 3 /t Ker (350°C) and 29.2 m 3 /t Ker (350°C), respectively, and then the yield increased greatly at 375°C. The maximum yields were 506.5 m 3 /t Ker and 368.4 m 3 /t Ker at 550°C, respectively. Compared with groups II and III, the groups IV and V showed little change in the total yield. With the addition of H 2 reaction resulted in the alkane yield increased, but the CO 2 yield decreased, resulting in an insignificant change in the total gas yield overall.

Product Components
In terms of the total yield distribution of gaseous HCs, the C 1-5 yield of group I increased continuously with temperature, and the maximum yield was 99.49 m 3 /t Ker. After the catalysts ZnCl 2 and MoS 2 were added to groups II and III, the maximum yields of C 1-5 were 208.22 m 3 /t Ker and 137.77 m 3 /t Ker, which were Frontiers in Earth Science | www.frontiersin.org March 2021 | Volume 9 | Article 626111 increased 2.1 and 1.4 times as much as group I, respectively. It proved that the addition of ZnCl 2 and MoS 2 significantly promoted the generation of the gas components. In groups IV and V, the maximum yields of C 1-5 were 375.34 m 3 /t Ker and 292.32 m 3 /t Ker, representing 1.8 and 2.1 times increase compared to groups II and III. The catalytic hydrogenation.
The methane (C 1 ) yield increased with temperature and reached the maximum at 550°C. The maximum C 1 yield in group I was 98.55 m 3 /t Ker. When the catalysts ZnCl 2 and MoS 2 were added to groups II and III, the maximum yields of C 1 were 207.56 m 3 /t Ker and 136.89 m 3 /t Ker, which were 2.0 and 1.4 times as much as the yield of group I, respectively. In groups IV and V, the maximum yields of C 1 were 369.82 m 3 /t Ker and 283.63 m 3 /t Ker, respectively, which were 1.8 and 2.1 times the yields of groups II and III (Figure 1).
The change in the ethane (C 2 ) yield of group I with temperature was very small, and the maximum C 2 yield was 0.8 m 3 /t Ker (500°C). After the catalysts ZnCl 2 and MoS 2 were added to groups II and III, the maximum C 2 yields were 1.16 m 3 /t Ker and 0.85 m 3 /t Ker, respectively. The addition of ZnCl 2 and MoS 2 did not increase the C 2 yield significantly. In groups IV and V, the maximum yields of C 2 were 64.76 m 3 /t Ker and 42.59 m 3 /t Ker (both 500°C), which were 80 and 50 times those of groups II and III without H 2 , respectively. Exogenous H 2 reacts more easily than H 2 O to promote the generation of C 2 (Figure 2).
The change in the propane (C 3 ) yield of group I with temperature was close to 0.0 m 3 /t Ker. After adding the catalysts ZnCl 2 and MoS 2 to groups II and III, the change in the C 3 yield was still 0.0 m 3 /t Ker, indicating that the addition of metal salts did not significantly increase the C 3 yield. In groups IV and V, the maximum yields of C 3 were 34.4 m 3 /t Ker and 21.8 m 3 / t Ker (450°C), respectively; the addition of external H 2 and catalysts exceedingly improved the C 3 yield (Figure 3). The comparison of the results of C 1 , C 2 , and C 3 shows that the addition of exogenous H 2 increases the HC generation yield, and the peak temperature of HC generation decreases sequentially with the increasing carbon number.
The catalytic addition of the H cracking reaction had a significant effect on the drying coefficient of natural gas (C 1 / C 1-5 ) ( Table 1). The C 1 / C 1-5 ratio of group I (kerogen + deionized water) increased with temperature, from a minimum of 0.90 (350°C) to 0.99 (550°C). When the catalysts ZnCl 2 and MoS 2 were added to groups II and III, the C 1 / C 1-5 ratio became larger with increasing temperature at the ratios of 0.89-1.00 and 0.87-0.99, respectively. The addition of ZnCl 2 and MoS 2 promoted the formation of C 2+ HCs at relatively low temperatures (≤400°C). After adding H 2 and catalysts ZnCl 2 and MoS 2 , the addition of exogenous H 2 greatly increased the heavy HC (C 2+ ) components, and the C 1 / C 1-5 ratios at 350°C were 0.55 and 0.78, respectively. The C 1 / C 1-5 ratio gradually decreased with the increasing temperature and dropped to the lowest point at 0.48 and 0.50, respectively, at 400°C. After 400°C, C 1 produced by the hydrocracking of kerogen continued to increase, as well as C 2+ HCs in the product. The conversion to C 1 significantly increased the C 1 yield, and the natural gas drying coefficient reached the maximum values of 0.99 and 0.97 at 550°C.
The CO 2 yield in each group of experiments increased with increasing temperature and reached the maximum value at 550°C. The CO 2 yield of group I gradually increased with increasing temperature, and the maximum CO 2 yield was 142.25 m 3 /t Ker. In groups II and III, with the addition of catalysts ZnCl 2 and MoS 2 , the maximum yields of CO 2 were 280.91 m 3 /t Ker and 192.35 m 3 /t Ker, respectively, which are 2.0 and 1.4 times the maximum yield without the addition of the catalysts. The increase in CO 2 and HC components was similar, indicating that the metal ions mainly catalyze the lysis of kerogen itself rather than decomposing the bond between H and O. The CO 2 yields in  groups IV and V were 0.39-0.40 times those of groups II and III, respectively ( Figure 4). The significant decrease in the yield of CO 2 resulted from the reaction that occurred with the addition of H 2 . The experimental comparison of groups I, II, and III showed that the H 2 yield did not increase significantly with temperature (the highest is 0.02 m 3 /t Ker at 550°C in group I), but the maximum yields after adding ZnCl 2 and MoS 2 were 24.27 m 3 /t Ker and 17.43 m 3 /t Ker at 550°C. The yield increased by one to two orders of magnitude, indicating that the H 2 production was significantly promoted by the addition of catalysts, combined with the C 2 yield. The dominant reason of the H increase is the cracking of organic matter.

Product Component Carbon Isotope Changes
The results show that δ 13 C becomes heavier with increasing temperature. Since kerogen of the Yurtus Formation is at a high maturity stage, only C 1 , C 2 , and CO 2 isotope values can be effectively detected in product components. For the δ 13 C value of gaseous HCs under different control conditions at the same temperature point, the δ 13 C value of each gaseous HCs increased with the intensity of reaction between catalysis and hydrogen addition ( Table 2). The δ 13 C 1 value of group I increased continuously with the increase in temperature from −39.8 to −34.9‰. When ZnCl 2 and MoS 2 were added to groups II and III, the δ 13 C 1 values changed from −44.7 to −36.7‰ and from −43.6 to −36.2‰. In groups IV and V, the δ 13 C 1 values changed from −45.9 to −40.2‰ and from −52.4 to −43.4‰, showing an increase before the 400°C stage, then a decrease in the 400-450°C interval, and finally becoming heavier with increasing temperature. For the C 1 yield after 400°C, catalytic hydrocracking plays a major role in promoting the disproportionation of small low-carbon molecules. Since the bond energy of 12 C-12 C is less than the bond energy of 12 C-13 C, the δ 13 C 1 fractionation follows; as the intensity of catalytic hydrogenation increases, δ 13 C 1 becomes lighter ( Figure 5).
The δ 13 C 2 value increased sequentially with the increasing temperature. In group I, the value of δ 13 C 2 changed from −28.2 to −24.4‰ at 350-400°C. After 400°C, the δ 13 C 2 value only changed from −24.4 to −23.6‰. In groups II and III with catalysts ZnCl 2 and MoS 2 , the δ 13 C 2 value was generally stable, similar to group I, changing from −29.9 to −24.0‰ and from −29.1 to −23.7‰, respectively. The addition of ZnCl 2 and MoS 2 promoted the reaction and the exchange of C elements to a lesser extent. The δ 13 C 2 value changed in groups IV and V from −38.4 to −14.1‰ and from −37.6 to −12.1‰, indicating that catalytic hydrogenation significantly increases the formation of C 2 and the exchange of C elements. At 550°C, the δ 13 C 2 value was the largest because the addition of H 2 promoted the catalytic cracking reaction. When the temperature exceeded the gas generation window (<500°C), a large amount of C 2 cracked to form C 1 , resulting in an abnormally large δ 13 C 2 value ( Figure 6).
As one of the main reaction products, the CO 2 isotope fractionation value is mainly controlled by its own molecular quality. The comparison of the experimental results between the hydrogenated and unhydrogenated groups showed that the δ 13 C CO2 value in the hydrogenated group was larger than the unhydrogenated group, which is negatively correlated with the change in the fractional distillation of the δ 13 C value of alkanes. The δ 13 C CO2 value of CO 2 generated by hydrocracking was relatively stable. The δ 13 C CO2 values of groups I, II, and III were distributed between −35‰ and −33‰. The δ 13 C CO2 values in groups IV and V showed a decreasing trend first and then an increasing trend. The participation of ZnCl 2 catalyst increased the δ 13 C CO2 value of the product. The joint addition of H 2 and the catalyst further increased the fractionation of the δ 13 C CO2 value.  The effect of MoS 2 on the δ 13 C CO2 value was larger at the early stage of the reaction (Figure 7). At the same time, relative to the changes in δ 13 C values for C 1 and C 2 , the range of carbon isotope fractionation of CO 2 was smaller than that of alkanes because of the oxidative decarboxylation of its controlled long-chain HCs, rather than the bond energy difference controlled by the C-C bond leading to preferential cracking (Lu et al., 2010).

Catalytic Hydrogenation Mechanism
Catalytic hydrogenation showed that the yield and isotopic value of the gaseous product components changed significantly and were also accompanied by FTT synthesis reaction products. The source supply angle mainly depends on the control influence of H 2 and different metal elements in the reaction environment.

Product Component Comparison
The post-mature kerogen of the Yurtus Formation has very weak pyrolysis and HC generation capacity. With the increase in temperature and the catalysis of hydrogenation by different catalysts, the kerogen HC generation capacity significantly improved. As the most stable HC molecule, C 1 had an increased yield. In addition to the catalytic hydrogenation reaction of kerogen, there were other gaseous HCs produced by catalytic hydrocracking. The yield of different catalysts changed significantly. Under the same hydrogenation conditions, the yield of C 1 under the action of ZnCl 2 was greater than that of MoS 2 , with a yield ratio of 1.0-1.4 and average of 1.3. The results of the control experiment between the experimental group with added H 2 and catalyst and the experimental group with catalyst only showed that the addition of H 2 delayed the decomposition of heavy HCs such as C 2 and C 3 , leading to a higher temperature of HC generation peaks. Meanwhile, the heavy HC group bond energy was smaller than the methyl group energy, which led to the preferential participation of heavy HC groups in hydrocracking and thus to a significant increase in the yield of C 2+ HCs and a significant reduction in the drying coefficient of the product. The addition of exogenous H 2 greatly changed the kerogen's own cracking mode, and heavy HC component C 2+ increased significantly. It was well consistent with the results that the drying coefficient of the product gases with the increased intensity of catalytic hydrogenation is significantly lower than that of the ordinary geological environment (Lewan et al., 1979;Mango et al, 1994;Ma et al., 2018).

Change in the i/n-Alkane Ratio
Among gaseous HC components, normal paraffins are formed by free radical reactions, and isoparaffins are derived from free radical cracking on kerogen and asphalt branches under acidic cationic ion reaction (Eisma and Jurg, 1969;Almon and Johns, 1977;Kissin, 1987;Pan et al., 2006). The Yurtus Formation kerogen is high maturity, and the branched chains of its molecular structure is not developed; thus, the main source of isoparaffin is the acidification cation reaction. Thompson and Creath, (1966) showed that in North American industrial HC reservoirs mainly generated by free radical HC generation, the iC 4 /nC 4 ratio is around 0.5, generally not greater than 1.0; the iC 5 /nC 5 ratio is around 1.0 and generally less than 2.0. In this study, the comparison of the same component at the same temperature showed that the value of iC n /nC n under the action of ZnCl 2 was larger than MoS 2 . For the products of butane (C 4 ), from the initial temperature of 350°C-400°C, the iC 4 /nC 4 value of the products produced by the two catalysts was relatively stable, ranging from 0.241 to 0.309 and 0.184 to 0.217, respectively.
The maximum values were 1.045 and 1.042, respectively, at 450°C. For the products of pentane (C 5 ), the temperature changed greatly from the initial temperature of 350°C-500°C; the range of the iC 5 / nC 5 ratio under the action of ZnCl 2 was 2.200-6.252 and under the action of MoS 2 was 0.096-1.409 (Table 1, Figure 8). The results showed that under the action of ZnCl 2 , more HCs were generated by the cationic reaction, which promoted more formation of the isoparaffins. The higher the temperature is, the more significant changes in the cationic reaction are. The results further reveal that the catalytic effect of ZnCl 2 on promoting HC generation by the cation reaction is stronger than that of MoS 2 .

Reaction Mechanism
HC generation of catalytic hydrogenation is mainly carried out in the form of H radicals and H ions. There are two modes of catalysis: "Lewis acid" catalysis and "Brønsted acid" catalytic processes (Figure 9). When performing Lewis acid site catalytic processes, the main chemical reactions are decarboxylation and C-C bond breaking (Li et al., 2002;He et al., 2011;Ma et al., 2018). Since the Lewis acid site composed of Zn 2+ or Mo 2+ is an empty orbital position with high affinity for electrons, when the organic matter is decarboxylated, Zn 2+ or Mo 2+ at this position gets an electron from the adsorbed organic molecule, and carboxylic acid loses CO 2 . Free radicals further undergo a rearrangement reaction with H radicals, leading to the breakage of the C-C bond and the formation of free HCs with shorter bond lengths. The product of the decarboxylation reaction is mainly CO 2 . The breakage of the C-C bond mainly promotes the generation of free HCs. When performing Brønsted acid catalytic processes, through contact with Zn 2+ and Mo 2+ , H 2 or water molecule lose an electron to generate H ions, promoting the kerogen unsaturated cycloalkane addition reaction. As the hydrogenation reaction continues, lowmolecular-weight saturated chain HCs are eventually formed (Wu et al., 2012;Ma et al., 2018).
The yield results show that Zn 2+ has a better catalytic effect than Mo 2+ because Zn 2+ has a more stable and long-lasting catalytic activity than Mo 2+ , and it is easier to fully contact the reactants. The chemical properties of Cl − are extremely stable. Previous studies suggest that Cl − has a certain inhibitory effect on the catalyzed cracking of kerogen (Li et al., 2002). In the catalytic hydrogenation experiment of H 2 , ZnCl 2 shows a strong catalytic effect on hydrogen generation. S 2− reacts with H 2 to generate part of H 2 S, which has a catalytic effect on HC generation (He et al., 2011), but the product test results show that the H 2 S content is relatively low, and the catalytic contribution is not significant.

Stable Carbon and Hydrogen Isotope Change Mechanism
For the HC isotope composition of oil and natural gas, relative to its parent material, kerogen is lean in 13 C and 2 H. This is because the kerogen 12 C-12 C bond and the C-C bond in 1 H-CC-1 H are more easily broken than the 12 C-13 C bond and the 2 H-CC-1 H CC bond, thereby allowing lighter isotopes to enter HC (Lu et al., 2010).
The change in δ 13 C 1 -δ 13 C 2 values can reflect the catalytic hydrogenation effects effectively. The data of groups I, II and III showed that δ 13 C 1 -δ 13 C 2 has a strong linear relationship ( Figure 10). After adding the catalysts, the change in the δ 13 C 1 -δ 13 C 2 value was decreased significantly and the range changed widely, but C 2 yield changed little (Figure 2). The results showed that C 2 mainly promotes the elemental exchange between carbon isotopes after the catalyst were added and has little effect on the yield reaction equilibrium. The comparison of the experimental results showed that the catalyst addition leads to lighter δ 13 C 1 and δ 13 C 2 . Meanwhile, the Frontiers in Earth Science | www.frontiersin.org March 2021 | Volume 9 | Article 626111 metal ions are involved in enhancing the reaction process and promoting the isotope fractionation level during the reaction. The catalytic hydrogenation extremely promotes the activation of the carbon source in post-mature kerogen. Within a certain temperature range (≤450°C), the distribution of δ 13 C 1 -δ 13 C 2 is clustered and distributed in a stable lighter interval. At 500°C, the δ 13 C 1 /δ 13 C 2 ratio is close to the range with no H 2 added. The C 2 component passed the peak of HC generation because the C 2 component with a smaller δ 13 C value starts to crack into C 1 , which led to the leaving C 2 with heavier δ 13 C. At the same time, C 3 cracked completely due to the δ 13 C kinetic fractionation to promote the formation of lighter δ 13 C 1 and heavier δ 13 C 2 .
Restricted by high temperature and the amount of carbon material, the heavier carbon content retained in generated C 2 increases at 550°C, resulting in larger change of linear variations. It also implies the δ 13 C 1 -δ 13 C 2 parameters can provide supplementary verification to evaluate whether the HC generation process of source rocks was affected by catalytic hydrogenation in sedimentary basins.
Relationship Between CO 2 and Alkane δ 13 C CO 2 is one of the main by-products of the reaction process. Its δ 13 C fractionation change is opposite to that of alkane δ 13 C. δ 13 C CO2 increases with the enhancing intensity in the reaction (Figures 11, 12). When the catalytic hydrogenation stage is weak, the kerogen HC generation is mainly carried out by selfcracking; thus, the changes in kerogen δ 13 C in the low temperature stage are greatly affected by temperature and the external H 2 participation in the reaction. The δ 13 C CO2 value became larger accompanied by an increasing intensity of the catalytic reaction. In group I, the δ 13 C CO2 change from the large to the small was at the 400°C inflection point. In group II and III, the δ 13 C CO2 change from large to small was at the 450°C inflection point and the HC generation mode of kerogen was affected distinctly by the catalytic elements during pyrolysis reaction. In group IV and V, the addition of H 2 changed the characteristics of the original kerogen evolution process and promoted isotope exchange in the HC generation. The change in δ 13 C CO2 from lighter to heavier was mainly controlled by the dynamic fractionation of carbon isotope of CO 2 . The maximum δ 13 C CO2 temperature under the action of ZnCl 2 is delayed compared to that under the action of MoS 2 . The δ 13 C 2 value under the action of ZnCl 2 is less than that under the action of MoS 2 , which demonstrates that the catalytic effect of ZnCl 2 under the same conditions is stronger than that of MoS 2 .
As one of the main modes of inorganic HC generation, FTT synthesis is well verified in geological discoveries and experiments (Fu et al., 2007;Hosgörmez, 2007;McCollom, 2013;Suda et al., 2014;McCollom, 2016;Etiope, 2017; and greatly affected by the redox conditions of the environment. The catalytic hydrogenation experiment provides a good reduction environment and source supply conditions. As shown in Figure 13, the relationship between CO 2 and CH 4 in the catalytic hydrogenation state of ZnCl 2 was y 1.1359x-17.687, R 2 0.9375. Under the catalytic hydrogenation state of MoS 2 , the relationship between CO 2 and CH 4 changed to y 2.1496x-97.873, R 2 0.9533. Under the stable reaction at >400°C, the relationship between CO 2 and CH 4 changed to y 1.9199x-73.472, R 2 0.9921, indicating that the decrease in the yield of CO 2 in the hydrogenation reaction showed a strong linear relationship with the increase in C 1 . The CO 2 yield decreased rapidly at 375°C owing to the solubility of water at high temperatures. In the high-temperature stage (≥400°C) under the catalytic state, the activation reaction of hydrogen radicals and carbon molecules was accelerated. In the catalytic hydrogenation reaction of CO 2 , a large quantity of HCs and water mainly composed of C 1 are produced in the FTT synthesis reaction accompanying H 2 and CO 2 . The change in the slope of the equation shows that the catalytic effect of ZnCl 2 on kerogen pyrolysis and HC generation is stronger than that of MoS 2 . With H 2 added to the HC generation reaction, the CO 2 yield decreased significantly. However, the CO 2 yield remained stable after falling to a certain value. The yield was stable under the influence of different temperature points. According to the δ 13 C CO2 isotope change, a heavier trend in both H 2 added groups implies that 12 C is more likely to participate in the catalytic hydrogenation response than 13 C. Combined with the experimental results described in Table 1, Figures 4, 13, the addition of H 2 inhibited the CO 2 generation. The FTT synthesis decreased the generated CO 2 and H 2 yield. The results of the control experiment showed that the final CO 2 conversion rate was a constant under a different reaction environment. The amount of activated C with the H 2 addition reaction indicated that activated C reactants would not react in the geological environment.
The catalysis reaction by the addition of H 2 was carried out in a neutral reduction system, and the FTT synthesis accompanied by CO 2 generated HCs, which had a significant effect on the total HC yield. In geological processes, the FTT synthesis reaction plays an important role as a link between the deep hydrogen-rich fluid and the hydrogen generation of post-mature organic matter.

CONCLUSION
For the ancient source rocks of the Yurtus Formation in the Tarim Basin, the catalytic substances and exogenous hydrogen carried under deep-derived fluids can "secondarily generate HCs" for the activation of source rocks. ZnCl 2 and MoS 2 have a strong catalytic effect on the reaction of post-mature kerogen and H 2 . Catalytic addition of H 2 promotes the HC generation capacity of kerogen and the increase in HCs by 0.4-1.1 times, significantly reduces the natural gas drying coefficient, and promotes FTT synthesis of secondary reactions. Isotope fractionation mainly follows the laws of kinetics and shows good identification characteristics. By comparing the composition of gaseous products and isotope characteristics, the catalytic effect of ZnCl 2 is stronger than MoS 2 under the same conditions. Meanwhile, it implies deep-derived hydrogen fluids containing Zn, Mo and other metal elements can make a remarkable contribution to reactivate the post-mature source rocks for HC generation.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding authors.