Effect of Methane Cracking on Carbon Isotope Reversal and the Production of Over-Mature Shale Gas

Abstract

The geochemical statistics indicate that the wetness (C₂~C₅/C₁~C₅) of over-mature shale gas with carbon isotope reversal is less than 1.8%. The magnitude of carbon isotope reversal (δ¹³C₁–δ¹³C₂) increases with decreasing wetness within a wetness range of 0.9~1.8% and then decreases at wetness <0.9%. The experimental result demonstrates that CH₄ polymerization proceeding to CH₄ substantial cracking is an important factor involved in isotope reversal of over-mature shale gas. Moreover, δ¹³C₁–δ¹³C₂ decreases with an increase in experimental temperature prior to CH₄ substantial cracking. The values of δ¹³C₁ and δ¹³C₂ tend to equalize during CH₄ substantial cracking. The δ¹³C₁–δ¹³C₂ of mud gas investigated at different depths during shale gas drilling in the Sichuan Basin increases initially, then decreases with further increase in the depth, and finally tends to zero, with only a trace hydrocarbon gas being detectable. Thus, the approximately equal value between δ¹³C₁ and δ¹³C₂ for over-mature shale gas and very low wetness could potentially serve as useful criteria to screen CH₄ substantial cracking. Two geochemical indices to indicate CH₄ substantial cracking in a geological setting are proposed according to the variation production data with the geochemistry of over-mature shale gas in the Sichuan Basin, China.

1. Introduction

Geological explorations in the 21st century have resulted in the discovery of extensive reserves of shale gas with isotopic reversal [1,2,3]. These expeditions elucidated that shale gas with isotopic reversal developed in areas where the maturity of source rock was much greater than 2.0% R_o [2,4].

In recent years, substantial research has been conducted to unravel the mechanism of isotopic rollover and reversal of over-mature natural gas. However, the specific procedures involved in this phenomenon remain to be not understood completely. Over the past two years, a novel approach was proposed which implied that isotopic reversal of over-mature shale gas was related to CH₄ polymerization occurring in the early stage of CH₄ cracking [5,6]. Chen, B. et al. [5] proved that shale gas with carbon isotope reversal could be produced via a CH₄ pyrolysis experiment conducted in a closed gold tube system. Furthermore, Mi, J.K. et al. [6] also demonstrated that the CH₄ polymerization could form a heavy hydrocarbon gas (C₂~C₅) with depleted δ¹³C by continuous heating at 450–800 °C, while performing the CH₄ thermal cracking experiment at a heating rate of 20 °C per hour. They further hypothesized that the combination of methyl formed in the early CH₄ cracking stage (termed CH₄ polymerization) was a crucial factor for isotope reversal in over-mature shale gas and that the magnitude of carbon isotope reversal (δ¹³C₁–δ¹³C₂) was inversely proportional to the degree of CH₄ cracking. In other words, the δ¹³C₁–δ¹³C₂ would mirror the degree of over-mature natural gas cracking to some extent. Zumberge, J. et al. [1] found that the high production of shale gas occurs only at low wetness values (<5%), corresponding to isotopic reversal, and that the biggest producers tend to have more negative ethane carbon isotope compositions. If the hypothesis that carbon isotopic reversal of over-mature gas is relative to CH₄ cracking, there should be a genetic relationship between CH₄ cracking and the δ¹³C₁–δ¹³C₂ in over-mature shale gas.

CH₄ is the most stable hydrocarbon component and the chief composition in natural gas [7]. CH₄ thermal stability or the natural gas geochemistry corresponding to CH₄ substantial cracking in a geological setting is very important to determine the largest depth of natural gas exploration [8]. The present study investigated the geochemical data of several hundred isotopically “abnormal” shale gas samples. Then, a CH₄ pyrolysis experiment was conducted to investigate the carbon isotope variation of hydrocarbon gas generated in the CH₄ cracking process. The similar variations in δ¹³C₁ and δ¹³C₂ of mud gas with increasing depth in three shale gas production wells and in that of gaseous products with increasing experimental temperature in the CH₄ cracking experiment indicate that CH₄ polymerization and CH₄ substantial cracking would happen in a geological setting. Finally, two geochemical indices to indicate CH₄ substantial cracking are proposed according to the experimental results of the CH₄ pyrolysis experiment, δ¹³C₁ and δ¹³C₂ variations of mud gas, with increasing depth and the variation of shale gas production as a function of geochemical composition in the Sichuan Basin. The scope of this work is to offer new insights into geochemistry to aid the exploration of over-mature shale gas.

2. Isotope Reversal Range Variation of Over-Mature Natural Gas with the Wetness

Previous exploration has proven that shale gas with isotope reversal possesses very low wetness [2,3,8]. The relationship between wetness and δ¹³C₁–δ¹³C₂ for isotopic reversal natural gas obtained from gas fields in the US and China is shown in Figure 1. It is evident that δ¹³C₁–δ¹³C₂ variation with decreasing wetness of over-mature shale gas can be represented by a parabolic curve. The wetness of shale gas with isotope reversal is less than 1.8%. The δ¹³C₁–δ¹³C₂ increases initially, with decreasing wetness within a wetness range of 0.9~1.8%, and then decreases at wetness <0.9%.

Several researchers have attempted to unravel the mechanism underlying the isotopic reversal of over-mature natural gas since it was extensively encountered in shale gas. Their interpretations can be classified into four types. (1) The isotopic reversal of shale gas at the over-mature stage is caused by the mixing of primary gas from kerogen cracking and secondary gas from the cracking of remained hydrocarbons [2,3,9]. (2) The redox reactions between water and CH₄ at 250–300 °C generate isotopically light carbon dioxide and hydrogen, which further interact and form isotopically light ethane (C₂H₆) and finally cause the isotopic reversal of the shale gas [1,8]. (3) The isotope fractionation mechanism during gas desorption from depressurized late-mature shale leads to isotope reversal in the residual gas produced from shale formation [4]. (4) CH₄ polymerization prior to its substantial cracking causes the isotopic reversal of over-mature natural gas [5,6].

(1) Mixing mechanism: In all mixing models, one end member employed in the mixing process was characterized by high wetness (wetness > 20% generally), while another one was a dry gas (wetness < 5%). For instance, the wetness of the two end members that were selected as primary gas and residual cracking gas by Xia X.Y. et al. [9] for their mixing model were 2.44% and 63.6%, respectively. The 63.6% wetness of residual cracking gas was so high that it rarely exists naturally in geological settings, especially in a shale system with over maturity. Oil-prone kerogen pyrolysis experiment results under hydrous and anhydrous conditions in a closed system illustrated that mixing between primary and cracking gases would not yield isotopic reversal, only isotopic rollover [10,11]. While the wetness of two end members involved in a mixing model is fixed, the geochemistry of gases involved in a mixing process is variable in a geological setting. Thus, the interpretation of isotope reversal of over-mature shale gas by a mixing mechanism does not hold true.

(2) Mechanism of oxidation-reduction reactions between CH₄ and H₂O: In geological settings, as the geological temperature attains the range at which CH₄ can react with water (i.e., 250–300 °C), C₂H₆ reacts with water prior to CH₄ owing to the comparatively lower activation energy. The reaction between C₂H₆ and water would result in the residual C₂H₆ possessing a heavier carbon isotope. Thus, the model of carbon isotopic reversal will not occur.

(3) Isotope fractionation mechanism during the uplift and depressurization of shale: The initial gas generated in the shale exhibits a normal isotope trend, with δ¹³C₁ < δ¹³C₂. After the generation of the new gas essentially ceases at high maturities (VR > 2%) and the uplift as well as depressurization are initiated, methane enriched in ¹²C desorbs from the source rock more rapidly than CH₄ enriched in ¹³C. Consequently, the residual CH₄ becomes enriched in ¹³C more rapidly than the residual C₂H₆, which leads to isotope reversal of the residual gas present in rocks [4]. Thus, the gas stored in conventional reservoirs should exhibit a normal isotope trend in view of this fractionation mechanism. However, the natural gases with isotopic reversal were found both in the marine conventional gas carbonate reservoir in the Sichuan Basin, China [12] and in the coal-derived tight reservoirs [13,14,15]. Therefore, the isotope fractionation mechanism of shale gas might just be a speculation and requires more concrete evidence.

(4) CH₄ polymerization mechanism: In theory, the ultimate products of CH₄ thermal cracking are C and H₂. However, CH₄ does not crack into carbon and H₂ directly during the cracking process, but does in the gradual dehydrogenation process [5,16,17]. In the CH₄ cracking process, a methane molecular dehydrogenates and forms a methyl radical (CH₃•) and a hydrogen radical (H•) initially. If two H• combine and form H₂, two CH₃• can also combine to produce C₂H₆. There are two kinds of methane molecules (¹²CH₄ and ¹³CH₄) in natural gas. ¹²CH₃–H is characterized by preferential bond cleavage, while ¹³CH₃–H is not. Thus, the preferentially formed CH₃• is depleted in the ¹³C. ¹²CH₃• combination, which leads to depletion of ¹³C in heavy gas, and further causes the carbon isotopic reversal of over-mature gas.

Natural gas with the isotopic reversal feature was originally considered to be of abiogenic origin before shale gas with similar features was identified. The abnormal isotopic feature of abiogenic hydrocarbon gas was thought to be caused by the polymerization of CH₄ formed by Fischer–Tropsch synthesis between H₂ and CO₂ (or CO) as well [18,19,20]. However, the hypothesis that CH₄ polymerization leads to isotope reversal in over-mature gas requires further validation with more experimental data.

3. Carbon Isotopic Variation of Hydrocarbon Gas Products in CH₄ Pyrolysis Experiment

The modern petrochemical industry has demonstrated that higher hydrocarbons can be generated via CH₄ polymerization in the presence of bifunctional catalysts [16,17]. However, in geological settings with sparse metallic catalysts, whether heavy gases depleted in ¹³C and generated by CH₄ polymerization can form naturally or not must be experimentally proven. In this study, a CH₄ pyrolysis experiment was conducted without any catalysts in a gold tube system.

To quantify the CH₄ content in products at different temperatures, a mixture of N₂ and CH₄ was used in the CH₄ cracking experiment. N₂ was selected as the reference gas in this experiment as it is the most stable diatomic molecule with a decomposition temperature greater than 3000 °C, thus ensuring that it could not crack at the highest temperature (800 °C) used in this experiment. The gas mixture was purchased from Zhaoge Gas Company, Beijing, China. The percentages of CH₄ and N₂ (v/v) in the mixing gas were 89.67% and 10.33%, respectively. No other gaseous component was detected in the mixture via gas chromatography (GC). The carbon isotopic composition of CH₄ was -26.07%.

The experiments were conducted in a gold tube system. The experimental system had 20 autoclaves and every one had an independent heating apparatus. The sizes of the tubes used in this experiment were 100 mm in length, 5.5 mm in outer diameter and 0.25 mm in thickness, respectively. One end of the tube had been sealed by argon-arc welding before the gas mixture loading. To load gas mixture into the gold tube, a custom made whifflow was developed, which could connect with a vacuum pump and the open tube mouth. In a typical gas loading procedure, the air in the tube was first removed by the vacuum pump to a pressure less than 1 kPa. Then, the mixing gas was injected into the tube from a gas cylinder from the open mouth of the tube. The pressure of gas injected in the tube was about 4 atm in this experiment, which was controlled via a pressure gauge on the gas cylinder. Finally, the open end of tube was sealed by argon-arc welding while the tube was immersed in ice water. The external fluid pressure of tube was kept a constant value of 50 MPa throughout the experiment. The experimental temperature was programmed as follows: the tube was first heated from 20 °C to 300 °C in 1 h and held for 30 min, then heated at a rate of 20 °C/h till the target temperature was arrived. Subsequently, the autoclave heating was terminated. Finally, the tubes were moved out for the relative analysis as the temperature of autoclave decreased to room temperature.

Two gold tubes were loaded into the autoclave at every temperature experiment in case of tube breakage. Moreover, duplicate tubes were also used as replicates for gas analysis and quantification.

Identification and quantification of individual gas components were carried out using a Wasson-Agilent 7890 GC. The heating program for the GC oven was heating from 20 °C to 68 °C (held isothermal for 7 min), then to 90 °C at a rate of 10 °C/min (held isothermal for 1.5 min), and finally to 175 °C at a rate of 15 °C/min (held isothermal for 1.5 min). An external standard with 12 gas components was used for the chromatographic calibration. Certified gas standards were prepared at a precision of better than ± 0.1 mol% for each component made by Zhaoge Gas Company, Beijing, China.

The stable carbon isotopes of the hydrocarbon gases were measured by using an Isochrom II GC-IRMS. The analysis temperature was programmed as follows: heating started from an initial temperature of 30 °C (isothermal for 3 min), then heating at 15 °C/min to 150 °C, and held isothermal for 8 min. The δ¹³C measurement for each component was repeated at least twice to ensure the analytical error of each component was less than 0.3‰. The δ¹³C values of each component presented in this work were the average values of two measurements.

The geochemistry of gaseous products of CH₄ cracking experiments at different temperatures is presented in

Table 1. Components and carbon isotopes of gaseous products at different temperatures.

Temp (°C)	Contents (% Vol)									δ13C (‰)
	CH4	C2H6 (×10−1)	C2H4 (×10−3)	C3H8 (×10−2)	C3H6 (×10−3)	iC4H10 (×10−3)	nC4H10 (×10−3)	H2 (×10−1)	N2	CH4	C2H6
450	89.65	0.01	0	0	0	0	0	0	10.35	-26.08	/
475	89.55	0.01	0	0.03	0	0	0	0	10.45	-25.94	/
500	89.53	0.01	0	0.04	0	0	0	0.01	10.47	-25.71	/
525	89.48	0.01	0	0.04	0	0	0	0.06	10.51	-25.71	-33.78
550	89.44	0.02	0	0.04	0	0	0	0.15	10.54	-25.84	-33.51
575	89.31	0.15	0	0.11	0	0	0	0.21	10.66	-25.74	-33.22
600	89.21	0.39	0	0.2	0	0	0	0.35	10.71	-26.01	-32.74
625	89.12	0.69	0.14	0.37	0.31	0	0.27	0.7	10.74	-25.69	-32.47
650	89.06	1.08	0.19	0.56	0.38	0.24	0.37	1.07	10.72	-25.97	-30.26
675	88.66	3.08	2.33	5.19	2.89	1.68	2.56	2.26	10.75	-25.81	-27.47
700	88.41	4.50	1.14	3.97	0.96	0.97	1.62	3.38	10.76	-25.83	-26.02
725	87.25	9.53	1.05	3.75	0.83	0.9	1.39	5.03	11.25	-25.66	-25.11
750	86.25	10.11	0.86	3.41	0.61	0.62	0.98	7.02	12	-25.28	-24.82
775	85.02	8.6	0.58	2.63	0.36	0.45	0.68	8.12	13.28	-25.04	-24.68
800	82.34	7.2	0.22	0.75	0.14	0.24	0.3	9.1	16.02	-24.60	-24.50

/: no detected.

. Besides the original components (CH₄ and N₂), heavy hydrocarbon gases (including alkane and alkene) as well as H₂ were generated during the experiment, at different temperatures. The heavy hydrocarbon gases were formed solely via the CH₄ polymerization, as no heavy hydrocarbon gases were detected in the original mixing gas. Overall, CH₄ concentration decreased with increasing temperature (Figure 2a). However, a slow decreasing in CH₄ content appeared below 700 °C, and a steep decline in CH₄ percentage occurred above 700 °C. The contents of C₂H₆ and H₂, the highest percentage gases in the newly formed components, were in the same order of magnitude (×10⁻¹%). The contents of all heavy gases increased initially and then decreased with experimental temperature increasing (Figure 2b,c). The corresponding temperatures at which the contents of C₂H₆ and rest heavy gases reached their maximum values were 750 °C and 675 °C, respectively. The highest concentration of total heavy gas was 1.05% at 750 °C.

The variation in C₂H₆ content with increasing temperature could be classified under three increasing segments (slow, moderate, and intense) as well as a decreasing segment. The slow increase in C₂H₆ content below 575 °C (Figure 2b) indicated that CH₄ polymerization was weak. Similarly, the moderate increase in C₂H₆ content within a temperature range of 575 °C to 650 °C suggested that CH₄ polymerization increased, while the intense increase in the C₂H₆ content within a temperature range of 650 °C to 750 °C indicated that CH₄ polymerization increased sharply. The decrease in C₂H₆ content above 750 °C (Figure 2b) was relative to the cracking of C₂H₆ formed by CH₄ polymerization.

The temperature of 750 °C, at which C₂H₆ content began to decrease, was higher than that of 700 °C, at which CH₄ content began to decrease rapidly. This appeared to conflict with the general knowledge in chemistry that the thermal stability of C₂H₆ is lower than that of CH₄. Shuai et al. (2006) observed via a coal pyrolysis experiment in a confined system that the generation and cracking of C₂H₆ could co-exist at high temperatures [21]. Their experiment results indicated that C₂H₆ cracking began at 500 °C in a coal pyrolysis experiment with the same heating rate of 20 °C/h as that in our experiment. In theory, the ultimate products of CH₄ cracking are C and H₂. However, CH₄ does not produce C and H₂ directly in the cracking process, but rather in the gradual dehydrogenation process. The reactions involved in the process include CH₄ polymerization to form heavy hydrocarbon gas, dehydrogenation of the heavy hydrocarbon gas to produce unsaturated heavy gas and the further dehydrogenation of the unsaturated heavy hydrocarbon gas to generate C and H₂ [5,16,17]. Therefore, all the aforementioned reactions (including CH₄ polymerization and heavy gas further dehydrogenation or cracking) could not stop as CH₄ does not convert into C and H₂ completely. Although the content of CH₄ decreases obviously at the highest experimental temperature of 800 °C, C₂H₆ does not disappear in experimental products. In other words, C₂H₆ cracking and C₂H₆ formation vs. CH₄ polymerization could co-exist at the high temperatures of our experiment. In the temperature range of 700–750 °C, the ratio of C₂H₆ formation vs. CH₄ polymerization was greater than that of C₂H₆ cracking, which was attributed to the increase of C₂H₆ content in this temperature range. Correspondingly, the higher ratio of C₂H₆ cracking vs. that of C₂H₆ formation via CH₄ polymerization caused the decrease in the content above 750 °C. In this study, the whole process of CH₄ cracking is divided into two stages: the early cracking stage and the substantial cracking stage. The early CH₄ cracking stage includes all the reactions prior to solid C formation. Whereas the substantial CH₄ cracking stage refers to the change from solid C appearance to CH₄ complete conversion into solid C.

The decrease in content of other heavy gases above 675 °C (Figure 2c) was also caused by their cracking. The same cracking temperature of 675 °C for C₃H₈ and C₄H₁₀ did not mean that they shared the same thermal stability as well. Instead, the temperature interval of 650–675 °C used in this experiment might cover the temperature point at which C₄H₁₀ contents began to decrease.

The inner wall of the tubes heated to different temperatures has been depicted in Figure 3. The color of the inner wall of the tube was brown at 725 °C, and became black with increasing temperature. The brown and black solids deposited on the inner walls above 725 °C were identified as carbon via energy spectrum analysis. The temperature at which the solid C formed was consistent with that at which CH₄ began to decrease sharply. Consequently, the appearance of solid carbon on the tube inner wall at 725 °C indicated that the substantial cracking of CH₄ began in the temperature range of 700–725 °C.

Zhang et al. (2013) reported that the lowest temperature of solid C generated by CH₄ cracking is 400 °C in an isothermal Fischer–Tropsch experiment [22]. The lowest temperature (725 °C) of solid C generated in our non-isothermal CH₄ cracking experiment is higher than that of solid C generated in the Zhang et al. (2013) isothermal Fischer–Tropsch experiment. This difference in temperature corresponding to solid C appearance is attributed to the compensation between temperature and isothermal duration as well as the catalyst effect in the Zhang et al. (2013) isothermal experiment. Nevertheless, the hydrocarbon gas shares a feature of carbon isotopic reversal before solid C appearance in the two experiments. This also indicates the CH₄ polymerization happens before CH₄ substantial cracking.

As presented in Table 1 and Figure 4, δ¹³C₁ and δ¹³C₂ generally increased with increasing temperature (Figure 3). The variation in δ¹³C₂ with increasing temperature could be divided into three segments: gradual increase in δ¹³C₂ below 625 °C; rapid increase from 625 to 725 °C; and slow increase above 725 °C. The variation of δ¹³C₁ was not observable within the analytic error (±0.5‰) below 725 °C. This is consistent with the slight decrease in CH₄ content (1.26%) below 725 °C. The distribution between δ¹³C₁ and δ¹³C₂ showed a distinct reversal trend for (δ¹³C₁ > δ¹³C₂) below 725 °C, while it was normal (δ¹³C₁ < δ¹³C₂) above 725 °C.

The experimental CH₄ cracking results indicated that CH₄ polymerization leading to heavy gas formation had occurred at 450 °C. However, the CH₄ polymerization was so weak that no obvious decrease in CH₄ content was observed—and consequently, a trace heavy gas was observed below 600 °C. CH₄ polymerization increased noticeably within the temperature range of 600–750 °C, resulting in increased heavy gas content. The generation of non-saturated heavy gas above 600 °C indicated that saturated heavy gas had begun to undergo cracking and further caused the rapid isotopic reversal of δ¹³C₂. Nevertheless, C₂H₆ content continued to increase, and the carbon isotopic distribution exhibited a reversal below 725 °C (Figure 4). This was attributed to the relatively higher ratio of the CH₄ polymerization to form heavy gas than that of heavy gas cracking, i.e., a high ratio of C₂H₆ cracking compared to C₂H₆ formation, caused by the decrease in C₂H₆ content above 725 °C. The values of δ¹³C₁ and δ¹³C₂ tended toward becoming equal as CH₄ began to substantial crack at 700–725 °C.

The CH₄ cracking experiment elucidated that methane cracking could be divided into two stages, early cracking and substantial cracking. The combination of methyl depleted in δ¹³C, which formed in the early CH₄ cracking stage, and produced heavy hydrocarbon gas with a light isotopic composition. The methyl combination could be looked at as CH₄ polymerization. CH₄ polymerization is an important underlying factor for isotope reversal in over-mature gas. Moreover, δ¹³C₁–δ¹³C₂ decreased with an increase in experimental temperature prior to CH₄ substantial cracking. The values of δ¹³C₁–δ¹³C₂ tended to become zero during the process of CH₄ substantial cracking. Thus, δ¹³C₁–δ¹³C₂ could potentially be interpreted as a factor to predict the degree of CH₄ cracking.

4. Carbon Isotope Variation of Mud Gas vs. the Depth during Shale Gas Drilling

As mentioned above, the CH₄ cracking experiment proved that δ¹³C₁–δ¹³C₂ in over-mature shale gas could be used as an index to extrapolate the degree of CH₄ cracking. Nevertheless, the experimental conclusion requires further validation of the practical data obtained from shale gas exploration. Generally, shale gas collected from a gas field is often a mixture of the gases obtained at different depths from the entire reservoir. The carbon isotopic composition of shale gas does not display any variations in δ¹³C₁ and δ¹³C₂ with increasing depth. In reality, the mud gas from varying depths during shale drilling is the original shale gas stored in shale. Thus, the carbon composition of mud gas from different depths could reveal whether the trend of δ¹³C₁ and δ¹³C₂ variation with increasing depth—similar to that observed during the CH₄ cracking experiment with increasing temperature—exists in geological settings.

The Sichuan Basin is a petroliferous gas basin located in the southwest of China. A large amount of shale gas has been produced in both the Cambrian Qiongzhusi Formation and the Silurian Longmaxi Formation in the basin. Both these clusters of shale gas attain over maturity [23,24]. Moreover, a majority of the shale gas extracted from these formations bears a feature of isotope reversal [25].

Niu et al. have presented the carbon isotope composition variation with depth of mud gases obtained from the three shale gas wells (DY5, JY10-10, SY3, Figure 5) [26]. In the three wells, the δ¹³C₁ of mud gas increased with increasing depth. Contrastingly, the δ¹³C₂ of mud gas initially decreased with increasing depth, followed by a sharp increase. All the mud gas observed in the interval segment exhibited isotope reversal. With increasing depth, δ¹³C₁–δ¹³C₂ of mud gas increases initially, then decreases and ultimately reaches zero. The vertical variation in δ¹³C₁–δ¹³C₂ for mud gas was fairly consistent with the conclusion derived from the CH₄ cracking experiment that δ¹³C₁–δ¹³C₂ decreased with increasing CH₄ cracking degree.

The CH₄ cracking experiment results indicated that CH₄ polymerization had occurred at 450 °C (Easy R_o = 1.49%). At this maturity level, the C₂H₆, which existed originally in over-mature shale gas, could not undergo substantial cracking in geological settings. With a rise in geological temperature or maturity level, CH₄ polymerization also increases, and more C₂H₆ depleted in δ¹³C is formed. The latter cause depletion of δ¹³C in the original C₂H₆, further resulting in isotope reversal in natural gas. With the increase in C₂H₆ formation via polymerization, the δ¹³C₁–δ¹³C₂ in natural gas escalates as well. The cracking of C₂H₆ formed via CH₄ polymerization occurs at a greater depth or higher temperature and causes a rapid increase in δ¹³C₂. At this stage, CH₄ cracking and polymerization attain a dynamic equilibrium, while the δ¹³C₁ and δ¹³C₂ tend to become equal. Thus, the vertical variation in δ¹³C₁–δ¹³C₂ in mud gases with increasing depth is in concurrence with the conclusions derived from the CH₄ cracking experiment. Moreover, the vertical variations of δ¹³C₁ and δ¹³C₂ with increasing depth is consistent with the results presented in Figure 1, which reflects the relationship between δ¹³C₁–δ¹³C₂ and the wetness of over-mature shale gas.

The quantity of mud gas is very low in the segment of the three wells, where δ¹³C₁ nearly equals δ¹³C₂. In addition, hardly any hydrocarbon gas was detected in the deeper segment below the layer where δ¹³C₁ nearly equals δ¹³C₂. Thus, it is evident that the phenomenon of very low gas content and approximately equivalent δ¹³C₁ and δ¹³C₂ should be caused by CH₄ substantial cracking.

5. Production of Shale Gas vs. δ¹³C₁–δ¹³C₂ and Wetness

Since natural gas is primarily composed of CH₄, especially for over-mature gas, the content and production of shale gas would decrease when CH₄ undergoes substantial cracking in geological settings. The experimental result of CH₄ pyrolysis indicates that δ¹³C₁ and δ¹³C₂ trend to be equal as in CH₄ substantial cracking. Based on this assumption, the production of over-mature shale gas would diminish with decreasing δ¹³C₁–δ¹³C₂. Figure 6 depicts the relationship between production and δ¹³C₁–δ¹³C₂ in shale gas sourced from different wells of the Longmaxi Formation in the Sichuan Basin, China. Both the average production per month and the accumulated production of shale gas from different wells generally showed a decrease with the reduction in δ¹³C₁–δ¹³C₂. It also indicates that natural gas (CH₄) cracking impacts the production of shale gas quite significantly. The nonlinear relationship between the production of shale gas and δ¹³C₁–δ¹³C₂ is attributable to the fact that shale gas production is controlled by diverse geological factors and the development of engineering methods, such as the physical properties of shale, pore evolution with increasing maturity, total organic carbon, type of organic matter and fracturing method. Nevertheless, CH₄ cracking is undoubtedly a crucial factor that affects shale gas production, according to the production data of shale gas in the Sichuan Basin. The shale gas from the Longmaxi Formation in the Sichuan basin, China, bears a feature of δ¹³C₁–δ¹³C₂ > 1‰ (Figure 6).

The wetness of the shale gas with isotope reversal in the Sichuan Basin is less than 0.9% (Figure 1). Moreover, there is a positive relationship between the wetness and the δ¹³C₁–δ¹³C₂ of the shale gas found in this location. The relationship between the wetness and the production of shale gas in the Longmaxi Formation of the Sichuan Basin has been presented in Figure 7. Both the average production per month and the accumulated production present a decreasing trend with a decline in the wetness of shale gas from different wells. The statistics in Figure 7 demonstrate that the wetness of the shale gas produced in the Sichuan Basin is more than 0.2%.

6. Conclusions

Geochemical statistics indicate that the wetness of shale gas with carbon isotope reversal is less than 1.8%. Furthermore, δ¹³C₁–δ¹³C₂ presents a parabolic variation with decreasing wetness. It can be seen that δ¹³C₁–δ¹³C₂ increases with decreasing wetness at the wetness range of 0.9~1.8%, and declines with decreasing wetness at wetness <0.9%. The CH₄ experiment cracking suggests that the polymerization occurring in the early CH₄ cracking stage is an important mechanism involved in the isotope reversal in over-mature gas. Moreover, δ¹³C₁–δ¹³C₂ exhibited a decrease with the increase in experimental temperature, prior to CH₄ substantial cracking. The values of δ¹³C₁ and δ¹³C₂ tend to become equal during the process of CH₄ substantial cracking. Thus, δ¹³C₁–δ¹³C₂ can be adopted as an index to interpret the degree of CH₄ cracking. The geochemical data of mud gas obtained during shale gas drilling in Sichuan Basin suggest that the value of δ¹³C₁–δ¹³C₂ increases initially, followed by a decrease with increasing depth and finally tending towards zero, with only a trace or no hydrocarbon gas being identifiable. The δ¹³C₁ and δ¹³C₂ variations with increasing depth of mud gas are consistent with that of gaseous products formed in the CH₄ cracking experiment with increasing experimental temperature. The actual statistics of production and geochemistry of shale gas in the Longmaxi Formation of the Sichuan Basin show that production of shale gas declines with the decreasing δ¹³C₁–δ¹³C₂ value and wetness. According to the results obtained from the CH₄ cracking experiment and geochemical data of over-mature shale gas exploration and development in Sichuan Basin, we suggest that shale gas exploration in deeper layer must be undertaken with great care when the wetness and δ¹³C₁–δ¹³C₂ are less than 0.2% and 1‰, respectively.