To investigate the tectono-thermal history and denudation thickness of the Qinshui Basin, we collected the vitrinite reflectance data of the Qincan 1 well from PetroChina Huabei Oilfield Company, the vitrinite reflectance was tested with the same methods as described in Yu et al. (2020a) (Table 3); note that the abnormal high reflectance of vitrinite near the unconformity of the Carboniferous and Ordovician strata has been eliminated.

We also collected mudstones and shale stones from the western edge of the basin to the center of the basin to purify illite (Figures 2, 3). We sampled unaltered outcrops of fresh rocks in the elevated area to maximize the relief wherever lithology and access were permitted. All samples collected in this work are Permian–Jurassic mudstones at an elevation ranging from 956 to 1,680 m (Supplementary Table S1; Figure 2). All the samples were analyzed for XRF to calculate the illite crystallinity and relative content of clay minerals and analyzed at the Exploration and Development Research Institute of Daqing Oilfield (China National Petroleum Corporation), China.

Polished thin sections were made from representative samples for microscopic observation. To further identify the clay minerals and composition of the illite structure, scanning electron microscopy (SEM) observations and energy dispersive spectrometry (EDS) analyses were carried out with an acceleration voltage of 15 kV and a working distance of 10 cm at the Electron Microprobe Analysis and Scanning Electron Microscope Laboratory of the Institute of Geology and Geophysics, Chinese Academy of Sciences (IGGCAS). To obtain high-quality images, the polished samples were coated with 8 nm carbon before analysis.

To achieve illite XRD analysis tests, the mudstone, siltstone, and calcareous mudstone samples needed to be ∼150 g with more than 40% clay content. We chose clay-rich samples (argillaceous siltstone, silty mudstone, mudstone, etc.,.) and crushed them to the diameter of 0.5–1 cm with a hammer; we subsequently used a DF-4 hammer crusher to crush the samples for less than 30 s. We used the Stock method to collect 40 mg of particles having diameter <2 μm and removed the liquid in suspension using an LXJ-64-1 centrifuge (Kisch, 1987; H.J.; Kisch, 1990).

We prepared the Oriented Clay Tablets by Precipitation Method with density >3 mg/cm² and dried the tablets naturally at room temperature (Kisch, 1987; H.J.; Kisch, 1990). We carried out the ethylene glycol expansion experiment at 50°C, then according to the clay minerals contained in the sample, we heated the sample at 500°C, the precision on the temperature of the muffle furnace used for heating is ±2°C. The experimental instrument is the Rigaku D/max-2200X diffractometer, with the following measurement parameters: 40 kV voltage, 40 mA current, Cu target, 1/2° emitting slit, 5.5 mm receiving slit, 0.04° Soller slit, scanning step length of 0.017°Δ2θ, and scan time of the 20 s by X’ Celerator detector. The scanning range for mineral phase and polytype identification is 4–60°Δ2θ. The test range of crystallinity of illite and chlorite for oriented flake is 2.6–15°Δ2θ.

Through the X-ray diffraction analysis of the sample, we can determine the dominant authigenic clay mineral assemblages in each stratigraphic unit (Supplementary Table S2). The clay minerals in the Triassic, Jurassic, Permian, and Cambrian strata in the central area of the Qinshui Basin are dominated by illite, and the combination of clay minerals is I + C + K. The interlayer ratio of I/S disorder is high, and illite accounts for 70–90%. The clay mineral assemblage in the Carboniferous strata is dominated by kaolinite, and the clay mineral assemblage is I + C + K with the high interlayer ratio of I/S disorder, and illite accounts for 80–90% (Figure 4; Figure 5A).

According to Figure 5, there is a high illite content in each layer, most samples account for 50–80% of the total content of clay minerals (Figure 5A), and the illite proportion in the shallow layer is higher than that the deep layer (Figure 5B), which may be caused by the sample difference. Kaolinite and the I/S mixed layer account for less than 30% of clay minerals. As the burial depth increases, kaolinite will turn into chlorite when the medium changes from acidic to alkaline (Gao et al., 2020; Zhu et al., 2021). The I/S trend of different layers stays stable, but the proportion of illite is high, indicating the high TTI (Time-Temperature Indicator) of the study area (Figure 5B) (Ramseyer and Boles, 1986). The distribution of clay mineral content shows that the basin has experienced a high paleogeothermal evolution process.

Through SEM observation (Figure 6 and Supplementary Material), except for the upper Paleozoic Carboniferous Hutian section, the interlayer illite of mudstone and sandy mudstone in other layers is the main clay mineral of most samples. Illite associated with illite/smectite (I/S) and illite without I/S can be observed in different horizons, we can identify illite in four forms according to microstructure: ①. Strip illite associated with I/S (Figure 6A). ②. Feather and filamentous illite associated with I/S (Figure 6D). ③. Feather and Bridge illite (Figure 6E). ④. Rosette illite (Figure 6I).

The illite energy spectrum is characterized by Al/Si ratio is close to 0.8 and high K content, and the main elements are O, Al, Si, and K, and the K/Al ratio is close to 0.4 (Uysal, 2000; Uysal et al., 2000, 2001, 2006; Zhu et al., 2021). I/S (R = 3) mainly appears in the form of cotton floc (Figures 6A,G), and its energy spectrum is characterized by K/Ca content close to 0.5 and Al/Si ratio close to 0.5 (More details in the Supplementary material). Chlorite mainly occurs in the Shihezi Formation of Upper Paleozoic and Heshanggou formation of Lower Triassic of Mesozoic. Leaf-like (Figure 6C) and needle/layered liked chlorite can be observed, and its energy spectrum is characterized by high Fe content, and the Fe/Si ratio is greater than 0.5 (Figure 6F). Kaolinite is mainly observed in the maroon iron clay rock of the upper series of the Hutian section of the Carboniferous in the upper Paleozoic, and it is hard to find kaolinite in the other horizons, which is in the form of vermicular porous (Figure 6B), and its energy spectrum characteristics show that the elements only contain O, Al, and Si, with Si/O ∼1, the Si/Al ∼0.8 (More details in the Supplementary material). It can be observed that kaolinite rarely contains other clay mineral components (Uysal, 2000; Uysal et al., 2000, 2001, 2006; Zhu et al., 2021).

In the SEM experimental observation, it is obvious that illite is the main component of sandstone interlayer clay minerals, and the other clay mineral components are relatively rare, indicating that the higher paleogeothermal temperature is conducive to the formation of illite in the process of basin evolution. Illite associated with I/S may be authigenic illite, but it may be affected by grinding flakes resulting in the formation of imperfect plates. However, it can be seen that many illite crystals growing alone have good crystal forms, showing rose and feather crystal forms (Figure 6I). Generally, illite with good crystal form represents the authigenic illite, but the crystalline morphology of terrigenous clastic illite is generally poor and is easy to be damaged in the process of long-distance transportation.

Based on the reconstruction of the maximum paleotemperature of the basin, the tectono-thermal history of well Qincan 1 was simulated by the software Basin Mod 1D (Figure 10). The software includes several models, such as burial history, tectono-thermal history, hydrocarbon generation history, migration, and accumulation history. In this study, we mainly use the Basin Mod 1D module to simulate the burial history and tectono-thermal history of the Qinshui basin.

Firstly, the relationship between vitrinite reflectance and depth is used to calculate the denudation thickness since the Early Cretaceous thermal event (Figure 9), estimated at 3,269.32 m. Secondly, combined with geological parameters such as stratum stratification data of the Qincan 1 well and the geological background in the middle of Qinshui Basin, the burial history curve of the Qincan 1 well is established. According to the apatite fission-track (AFT) study results (Zhu et al., 2014), the AFT data (Zhu, 2013) reflect the uplift process since the Early Cretaceous. Finally, by adjusting the amount of strata denudation and geothermal gradient, the simulated Vitrinite reflectance should be consistent with the measured Vitrinite reflectance, and the accurate tectono-thermal history evolution path can be obtained.

From Figure 10, it appears clear that the change of geothermal flow can be divided into two stages. The first stage is characterized by a large geothermal flow of ∼68 Mw/m², in the Early Carboniferous, and slowly decreasing to 60 mW/m² in the Late Triassic; the second stage corresponds to a geothermal flow gradually rising, reaching a peak value of 84.8 mW/m² in the Early Cretaceous, and then decreasing to 61.69 mW/m² at the present-day.

According to simulations, the evolution of the paleogeotemperature gradient is also divided into two stages. The first stage, from the Early Carboniferous to the Late Triassic, is relatively stable at 2.9°C/100 m, and a second stage, taking place after the Late Triassic, where the paleogeotemperature gradient increased to a maximum value of 6.51°C/100 m during the Early Cretaceous (Figure 10). According to the burial history (Figure 10), the basin reached its maximum depth at the Early-Middle Cretaceous, and gradually uplifted from the Middle-Late Cretaceous. The relationship between the simulated vitrinite reflectance curve and the measured vitrinite reflectance curve fit well, indicating that the tectono-thermal history recovery is reliable.

If the separation of illite is not so pure, the determined value of the crystallinity of illite may include some bias due to the presence of detrital illite. Detrital illite refers here to terrigenous illite detritus deposited concomitantly with sediments, mainly from metamorphic illite originating from the study area. Consequently, the paleo-geotemperature reflected by the crystallinity of illite may not reflect the paleo-geotemperature of the source strata, but the provenance. Therefore, the separation and purification of authigenic illite is an important step to avoid the presence of detrital illite. To determine the crystallinity of authigenic illite, we must separate the authigenic illite from sandstone, mudstone, fault gouge, and other types of rock samples first, then remove the non-authigenic illite components from the rock samples and finally extract the authigenic illite.

As explained, the separation of authigenic illite will have a direct influence on the quality of the testing results; separation and purification of illite are essential to maximize the complete removal of detrital K-bearing minerals such as detrital K-feldspar and detrital illite. Moreover, the authigenic illite should also be enriched to the greatest extent, whereas the content of kaolinite, chlorite, smectite, and other non-authigenic illite components should be reduced to the greatest extent. Judging from the current situation, the only effective way to achieve this goal is to extract more fine-grained clay components.

The separation of illite is quite challenging for two reasons. Firstly, the separation of authigenic illite requires a fine particle size. Conventional clay separation generally requires the extraction of <2 μm components, while the separation of authigenic illite requires at least particle size of <0.3 μm; Secondly, the separation of authigenic illite requires successive separation steps to extract different continuous fractions simultaneously, e.g., 1–0.5 μm, 0.5–0.3 μm, 0.3–0.15 μm, and <0.15 μm. The separation and purification of illite are generally limited by instruments and equipment, and the separation was carried out to fraction sizes of 0.5 μm. Therefore, the clay minerals finally separated and purified may contain some detrital illite, chlorite, kaolinite, and other clay minerals, which might affect the test results. According to the XRD patterns (Figure 6 and Figure 4), we can observe diffraction patterns characteristics of chlorite and kaolinite, which therefore indicate that the separation of early authigenic illite is not completely pure. Consequently, the calculated illite crystallinity might include detrital illite crystallinity, which might have an impact on the paleotemperature inversion calculation. But through the SEM (Supplementary Material, Figure 6), Most of the clay minerals in the sandstone interlayer are illite with a good crystal shape (Figure 6), and some illites are associated with I/S (Figure 6A), indicating that there are many authigenic illites to some extent.

According to the calculation results of the illite crystallinity and the vitrinite reflectance of well Qincan 1, a strong tectono-thermal event occurred in the basin. As a result, the maximum ancient geothermal gradient is 6.5°C/100 m, which is larger than the current geothermal gradient of 2.4°C/100 m, and the maximum paleo-geotemperature was around 180–190°C as recorded at the Carboniferous strata.

Previous studies have attempted to determine the occurrence time of tectono-thermal events in Qinshui Basin; the fission-track age of apatite, for example, is estimated at ∼20–40 Ma, and the zircon fission-track age is ∼100–150 Ma (Ren et al., 2005; Zhu et al., 2014). If the fission-track age of zircon happened to be smaller than the age of the formation, it would indicate that the formation temperature was once greater than 250 °C and that the zircon has undergone annealing (Hurford, 1986; Fitzgerald and Gleadow, 1988; Yan et al., 2010; Malusà and Fitzgerald, 2019). Secondly, the isotope age determination results of the igneous rocks around the Qinshui Basin indicate that the time of magma intrusion and eruption is between the Late Jurassic and Early Cretaceous. For example, the Taershan-Erfengshan alkaline complex, which is in the southwest of the Qinshui Basin, has an Ar-Ar age of about 124 ± 0.3 Ma. In addition, the U-Pb age of the Wanrong-Gufeng Mountain in the southern Qinshui Basin ranges between 127 ± 1 Ma∼140.5 ± 5.3 Ma (Yan et al., 1988; Shanxi, 1989; Wu et al., 1996, 2008; Ying et al., 2011; Si, 2015; Huo, 2016; Yang, 2016; Chang et al., 2017; Yang et al., 2017). It shows that the tectono-thermal events mainly occurred in the Late Jurassic to the Early Cretaceous period (More details in Supplementary Table S4 and Supplementary Figure S1).

Among the geological factors, thermal field, tectonic stress field, hydrodynamic field, and formation conditions have an important influence on the formation of coalbed methane (Wu et al., 2014). The coal deposits in the Qinshui Basin began to form in the Late Carboniferous and gradually ceased when the delta deposits were formed in the early Permian (Sun et al., 2005; Ma et al., 2016; Chen et al., 2019; Yu et al., 2020a). The subsidence of the basin lasted until the Late Triassic when the temperature of the coal seam reached 135°C (the burial depth was up to 4 km). Influenced by the increase of temperature, the first stage of methane generation began, and the vitrinite reflectance maturity of the coal seam reached a medium level (R _o ≈ 1.2%) (Zeng et al., 1999; Cai et al., 2011). Then, the subsequent Indosinian orogenic movement caused the uplift and inversion of the basin during the Early Jurassic, and it remained in a state of slow deposition until the Late Jurassic (Zeng et al., 1999; Ma et al., 2016). At the same time, the exposed rocks around the Qinshui Basin, which include Taershan, Huyanshan, Zijinshan, Erfengshan, Gufengshan, etc, started to form (Yan et al., 1988; Shanxi, 1989; Wu et al., 1996, 2008; Ying et al., 2011; Si, 2015; Huo, 2016; Yang, 2016; Chang et al., 2017; Yang et al., 2017). It can be seen that there was a tectono-thermal event represented by intense magmatic activity in the Late Jurassic to Early Cretaceous, which is generally considered to be closely related to the secondary gas peak of coal seams in the basin at the same time (Figure 11) (Zeng et al., 1999; Xu et al., 2004; Jiang et al., 2005; Ren et al., 2005; Sun et al., 2005; Duan et al., 2011; Chen et al., 2019, 2019; Yu et al., 2020a; Gao et al., 2021).

From Late Jurassic to Early Cretaceous, magmatic thermal events caused rapid temperature increase of coal measures. A large amount of hydrocarbon was generated by coal measures, with coal rock as the main source of hydrocarbon generation. The rapid increase of gas led to hydrocarbon generation pressurization in the coal body (Ju et al., 2018); furthermore, the increase of temperature led to the expansion of gas to form abnormal high pressure (Figure 11).

The tectono-thermal event in Qinshui Basin between the Late Jurassic to Early Cretaceous was associated with the large-scale magmatic eruption and the formation of the Ultra-High-Pressure metamorphic belt (Yan et al., 1988; Shanxi, 1989; Wu et al., 1996; Zhang, 1997, 2002; Wu et al., 2008; Ying et al., 2011; Chen and Ding, 2012; Si, 2015; Huo, 2016; Yang, 2016; Chang et al., 2017; Yang et al., 2017; Yu et al., 2020a). Combined with the estimation of the maximum paleotemperature and paleotemperature gradient of illite crystallinity and vitrinite reflectance, the deep mantle of the basin was active during the Late Jurassic—Early Cretaceous with the lithosphere thinned and led to the formation of a high geothermal background value of mantle upwelling, which was the main reason for the tectono-thermal event and uplift denudation event of the basin in this period.

Previous studies have shown that the thickness of the thermal lithosphere of the present orogenic belt in the central part of the NCC is about 76–90 km (He et al., 2001; Wang et al., 2001; Zhang et al., 2006; Zhang, 2012). Based on their data, these authors estimated that the thickness of the thermal lithosphere of the Qinshui Basin in the Early Cretaceous was about 50–60 km, indicating that the lithosphere of the basin was thinned during the Early Cretaceous. At the same time, previous studies suggest that the NCC experienced strong lithospheric thinning and lithospheric mantle property transformation during the Mesozoic (Menzies et al., 1993; Griffin et al., 1998; Xu, 2001; Wu et al., 2008), accompanied by strong magmatism (F. Wu et al., 2005a; F.-Y. Wu et al., 2005b; Yang et al., 2008). Combined with the magmatism around the Qinshui Basin, we suggest that the lithospheric thinning in the Qinshui Basin is controlled by the thinning and destruction of the NCC.

Previous investigations have been conducted on the destruction of the NCC (Zhu et al., 2012, 2017; Yang et al., 2021b). According to Zhu et al. (2017), at 165 Ma, from the beginning of the Middle Jurassic Yanshanian movement episode A, the West Pacific plate subducted to the East Asian continental margin with a high speed and low angle, the tectonic environment in Qinshui Basin, therefore, corresponds to a compression environment, and angular unconformity events were generally developed. Starting from the Late Jurassic at 155 Ma, the western Pacific plate subducted from a low angle to a high angle subduction toward the edge of the East Asian continent (Zheng et al., 2018). With the rotation and retreat of the subducted plate, the corresponding tectonic environment of the Qinshui Basin became extensional, and magmatic activity began to appear; during this period, the Taihang Mountains began to uplift, and the eastern part of the Qinshui Basin as the western wing of the Taihang Mountains began to fold and uplift (Zhu, 2013). Accompanied by strong magmatism, the Late Paleozoic coal-measure strata experienced a high-temperature heating process. The regional tectonic stress field of the basin during the Middle-Late Jurassic is consistent with that of the whole of eastern China. The NNE trending folds are the most developed in the basin, covering the whole area on a large scale, reflecting the horizontal compressive stress field in the direction of NWW-SEE in the Yanshanian period. Under the action of this stress field, a wide and gently deformed syncline, Qinshui Basin, is formed (Zhu et al., 2015). At 140 Ma, the Yanshanian movement episode B began in the early Cretaceous, and the subduction angle of the Western Pacific plate to the East Asian continental margin began to decrease again. The tectonic environment corresponding to the Qinshui Basin was compressional, and crustal shortening, folding and stratigraphic angular unconformity events were widely developed (Zheng et al., 2018). From 125 Ma in the Early Cretaceous, the subduction angle of the Western Pacific plate to the East Asian continental margin changed to a high angle again, and the rate of subduction, rotation, and retreat reached the maximum, and finally left a retentive body in the mantle transition zone (Zhu et al., 2015). The corresponding area of Qinshui Basin, which was subjected to compression, thrust, and folds uplifting, began to develop structural inversion and normal faults. Since the late Cretaceous, the young and buoyant paleo Western Pacific plate gradually entered the subduction zone, and magmatic activity in the corresponding area of the Qinshui Basin gradually ceased, and the extensional tectonic activity was much weaker than that of the early Cretaceous (Shanxi, 1989). Throughout the Cretaceous period, there was a major structural system transition in eastern China, that is, the early continental collision and compression thrust structures were transformed into extensional structures, and the eastern NCC and Ordos basin became active with the lithosphere thinned (Ren et al., 2002; Chen and Ding, 2012; Yu et al., 2020b).

The Qinshui Basin is a transitional zone between rift basins in the east of the NCC and the Ordos Basin in the West, the thinning of the lithosphere and crustal extension also affected this area, forming a small amount of magmatic activity. The basin was subjected to weak NW-SE extensional tectonism and formed a series of small to normal faults based on the Yanshanian conjugate joints. Previous studies have shown that the deep cause of the early Cretaceous tectono-thermal events and the uplift and denudation events in the Qinshui Basin is mainly due to the subduction of the western Pacific plate into the deep part of the NCC, and the current low geothermal gradient in the Qinshui Basin is also affected by the withdrawal of the West Pacific Plate. Combined with the results of previous studies (Ren et al., 2020), our work suggests that the Qinshui Basin experienced a tectono-thermal event during the Late Jurassic to Early Cretaceous which is in response to deep lithospheric thinning of the NCC.