1. Introduction
Global energy production is currently undergoing a crucial transition from fossil to non-fossil fuels [
1]. Hydrogen, a widely available, clean, low-carbon, and efficient fuel source, holds a strategic position in clean energy substitution. The development of hydrogen fuel has gradually become a significant direction for the energy technology revolution to pursue and represents a crucial pathway for achieving energy decarbonisation [
2,
3,
4,
5]. Hydrogen is a diatomic element with a molecular weight of 2.01588 [
6]. At room temperature and pressure, it is a colourless, odourless gas. Furthermore, it is highly combustible and poorly soluble in water and has a density of 0.089 g/L (101.325 kPa, 0 °C): the lowest known gas density [
7].
Recently discovered natural hydrogen, also known as ‘golden hydrogen’ or ‘white hydrogen’, is generated by underground geological processes. Compared to artificially produced hydrogen, natural hydrogen is a genuine zero-carbon and renewable primary energy source [
8]. It can aid in addressing energy demand gaps, optimising the hydrogen energy industry structure, and facilitating clean energy substitution [
9]. With respect to the backdrop of global efforts to achieve energy decarbonisation and net-zero emissions, natural hydrogen has sparked widespread research and exploration interest, with multiple countries developing plans for producing and utilising natural hydrogen [
10]. A surge in the exploration for hydrogen has been observed, with anomalous hydrogen found in different basins in China, such as the Sebei gas field in the Qaidam Basin (hydrogen content: 99%); the Songke Well 2 (SK2) in the Songliao Basin (highest hydrogen content: 26.89%); the Sulige gas field in the Ordos Basin (2.1%); multiple gas fields in the eastern Sichuan Basin, Bohai Bay Basin and Jimo and Dongying Sag areas (1–22.8%); Changbai Mountain (1.24%); Tengchong (1–5.15%) and the Zhengye-1 well in northern Guizhou (24.7–36.98%) [
11,
12,
13,
14].
However, current research on natural hydrogen primarily focuses on its genesis, and there is limited understanding of its occurrence. Studying the occurrence of natural hydrogen is crucial for the exploration and development of hydrogen reservoirs. In previous research, the present authors investigated the adsorption of hydrogen onto pure clay minerals [
15]. Common clay minerals in geological formations include montmorillonite, kaolinite, illite, chlorite, an illite/smectite-mixed layer (I/S), and chlorite/smectite-mixed layer [
16]. These clay minerals have complex morphological structures, distributions, and compositions and have undergone varied transformation processes; therefore, they represent physical, chemical, and biological information about the tectonic history, source input, sedimentary system, and diagenetic evolution of basins. In the 1920s, the introduction of X-ray analysis techniques lead to the determination of the characteristics and nature of clay minerals [
17]. A series of works, including the “Formation and Occurrence of Clay Minerals” and “Clay Mineralogy”, laid the foundation for developing clay mineralogy [
18]. By 1980, the crystalline chemical classification table proposed by the International Clay Society for layer-type clay minerals had attracted widespread attention from mineralogists and petroleum geologists worldwide. Recently, research on reservoir clay minerals has brought new benefits, with an increasing number of new technologies, methods, applications, and studies on the research and application of clay minerals in domestic oil and gas exploration. Widely distributed in oil and gas basins, clay minerals are not only the main mineral components of mud shale but also the most important pore-filling materials in sandstone reservoirs, wherein their type, content, and occurrence have a crucial impact on the pore-throat structure and storage permeability of sandstone.
To further investigate the hydrogen adsorption capacity of clay minerals in geological samples, the present authors studied the adsorption of natural hydrogen in Donglouku and Yingcheng formations in SK2, which are rich in hydrogen-bearing layers based on geological conditions [
19]. However, because the clay mineral content in SK2 Donglouku and Yingcheng formation mudstones is minimal (2–8% chlorite), their hydrogen adsorption capacity is not very high. Therefore, to further investigate the adsorption hydrogen capacity of clay minerals in geological samples, the present authors selected the clay-rich Chang 7 member of the Yanchang formation shale in the Ordos Basin as our study object.
The Ordos Basin, located in the western part of North China’s craton, is one of China’s major oil-bearing basins [
20]. Recently, considerable breakthroughs have been made in the exploration of Triassic shale gas in the Chang 7 member in the southeastern extension of the Ordos Basin. The Chang 7 Shale Member has a high total organic carbon (TOC) content, ranging from 2% to 4% [
20,
21,
22]. The Chang 7 Shale Member contains type I and II kerogens, with a vitrinite reflectance (Ro) ranging from 0.7% to 1.2% [
23]. Clay minerals in Chang 7 mudstones include illite, kaolinite, montmorillonite, I/S, and chlorite. Previous studies have shown that the clay mineral content is generally between 20% and 60% [
24,
25,
26]. In Chang 7 mudstones, the illite content is relatively high, ranging from 11.01% to 38.21%; the kaolinite content is low, not exceeding 6% of all clay minerals; montmorillonite and I/S are heterogeneously distributed with high contents found in a small proportion of mudstone samples, averaging 7.31% and 11.42%, respectively; and the chlorite content is relatively high, ranging from 5.11% to 33.04% [
24,
25,
26]. The present research on the Chang 7 Shale Member focuses on its formation and shale oil and gas formation; no research is available on the adsorption of hydrogen by shale. Studying the adsorption of hydrogen by shale can help clarify the occurrence of natural hydrogen, further explain the impact of clay mineralogy on hydrogen adsorption, and provide insights for the exploration and development of natural hydrogen reservoirs. Utilizing natural hydrogen contributes to the promotion of sustainable energy development and utilisation, reduces reliance on traditional fossil fuels, drives green economic transition, and contributes to the sustainable development of future society and the environment.
2. Geological Background
The Ordos Basin, a large multicycle superimposed basin, is the second-largest sedimentary oil and gas basin in China, and it contains the largest proven resources of natural gas, coalbed methane, and coal reserves in China [
27]. In addition, it contains considerable reserves of petroleum, water, geothermal energy, rock salt, cement limestone, natural alkali, bauxite, and brown iron ore [
28,
29,
30]. The structural history of the Ordos Basin includes the mid-to-late Proterozoic continental rift, an early Palaeozoic North China craton surface sea basin, a late Paleozoic to Middle Triassic intracratonic sag within the North China craton, the development of the Late Triassic to Cretaceous Ordos inland basin and the formation and development of small fault basins around the Ordos Basin in the Cenozoic era [
27]. The structure of the Ordos Basin is characterised by a wide, gently sloping eastern side and a narrow, steeply sloping western side. The basin can be divided into six secondary tectonic units, including the Yishan slope, the Yimeng and Weibei uplifts, and others [
31] (
Figure 1a). During the Mesozoic period, owing to tectonic subsidence, the Ordos Basin became a large lake basin, where the main oil and gas source rock intervals developed [
27,
31].
During the Middle Jurassic period, a series of oil-bearing formations developed in the Ordos Basin, with the Triassic Yanchang formation being the main source rock. This fluvial–lacustrine–deltaic sedimentary unit records the entire evolution of the lake basin [
32]. Based on lithological combinations, sedimentary cycles, and well-log characteristics, the Yanchang formation is divided into 10 segments labelled Chang 10–Chang 1 from bottom to top [
33]. The Yanchang formation is interpreted as a third-order sequence and comprises the maximum flooding surface within the Chang 7 interval [
34]. The Chang 7 Shale Member comprises deep-water black organic-rich shale, turbidite siltstone, and delta-front distributary channel sandstone. It has a thickness of 20–100 m and is found at depths of <3500 m in the central part of the basin [
25,
26,
27,
28,
29,
30]. It represents the Carnian stage of the Late Triassic and is further divided into the following three parts: Chang 7
3, Chang 7
2, and Chang 7
1. Chang 7
3, overlying Chang 8 deltaic sandstone, is dominated by thick black organic-rich shale with occasional interbeds of fine-grained siltstone [
35]. Chang 7
2 and Chang 7
1 comprise interbedded dark grey shale, silt-rich shale, siltstones, and fine-grained sandstone [
35]. The Chang 7 Shale Member types I and II kerogen have a total organic carbon content (TOC) ranging from 0.29% to 24.68% (average 13.75%). The Ro ranges from 0.7% to 1.2% [
36]. The reservoir is primarily composed of siltstones and fine-grained sandstones, with porosities ranging from 4.8% to 12.6% and permeabilities ranging from 0.01 to 1.35 mD [
37]. The Chang 7 Shale Member is known for its high oil content, homogenous distribution, and large reserves. As of 2008, its known reserves exceeded 500 million tonnes, with predicted reserves exceeding 1.3 billion tonnes [
38].
3. Materials and Methods
The samples used in this study were collected from wells near Tongchuan (
Figure 1) covering the interval of the Chang 7 Shale Member. Most samples were black/deep grey shales or silty shales. The samples were cut into 30 μm thick sections, and a polarising microscope (EOS 700D; Canon, Tokyo, Japan) was used to analyse their sedimentary structures and composition. A polarising microscope (DM4500P; Leica, Wetzlar, Germany) with reflected and fluorescent light was used to observe the same spots in the thin sections. Subsequently, scanning electron microscopy (SEM; Gemini 300, Carl Zeiss AG, Oberkochen, Germany) was performed on gold/palladium-coated slices and rock fragments to further identify the microtextures and pore characteristics of the samples. The X-ray diffraction (XRD; Panaco Empyrean; PANalytical B.V., Almelo, The Netherlands) of sample components was conducted with 2θ values ranging from 5° to 90°. To determine the composition of clay minerals in the samples, a suspension centrifugation separation method was used to separate the clay, which was then dried and ground to a fine powder. The samples were then prepared as air-dried-oriented slides, ethylene glycol-saturated slides (8 h ethylene glycol saturation at 55 °C), and high-temperature slides (ethylene glycol-saturated slides treated for 3 h at 500 °C). XRD spectra were obtained to determine the clay mineral types, and peak fitting was performed using the symmetric Gaussian–Lorentzian function theory. The mass percentages of clay minerals were quantitatively calculated, and the K values of standard samples were used for calibration.
In order to obtain the total organic content (TOC), the samples were ground to approximately 100 mesh (<0.15 mm), and a carbon/sulfur analyser (LECO CS230) was used according to the Chinese National Standard (2003) 19145-2003 [
39]. Before the measurement, 100 mg of powdered shale (100 mesh) was treated in a 5% HCl solution at 80 °C to remove inorganic carbonates. Nitrogen adsorption experiments were conducted using an adsorption analyser (ASAP2020; Micromeritics, Norcross, GA, USA) with dried (non-extracted) shales with a 40–60 mesh particle size. The samples were first vacuum-degassed for 12 h at 105 °C, followed by N
2 adsorption measurements at −196 °C in a liquid nitrogen bath at relative pressures (i.e., P/P
0) ranging from 0.001 to 0.995. The data were analysed according to the following physical adsorption theories: Gurvich, Brunauer–Emmett–Teller (specific surface area [SSA]), Barrett–Joyner–Halenda (pore volume distribution and total pore volume [TPV]) and Dubinin–Astakhov (micropore volume) [
40].
High-pressure hydrogen adsorption was conducted at 25 °C, 45 °C, and 65 °C from 0 to 18 MPa using BSD-PH automatic high-pressure gas adsorption instrument (
Figure 2). The sample filling quantity was approximately 30 g to ensure testing accuracy. The volume method was used for testing, and the expansion of the samples was determined using helium. High-precision pressure sensors were used to measure the pressure. Before hydrogen was introduced, the system was evacuated to remove residual gases. For the adsorption measurements, hydrogen gas was continuously transferred from the reference cell to the sample cell. The difference in the transferred gas volume and the free gas volume in the sample cell was the excess hydrogen adsorption. Hydrogen molecules are small, and adsorption equilibrium is achieved quickly, typically within a short period. At each pressure point, the contact duration was approximately 40 min, and the entire adsorption experiment took about 13 h. Regarding the determination of equilibrium, the instrument used in this experiment was the static volumetric method, where equilibrium was judged based on the pressure change within a certain time frame. The equilibrium condition employed in this test was a pressure change per minute less than 0.1% of the current pressure, ensuring that the adsorption error was less than 0.01 mL/g (STP). We supplement relevant details in the methodology section. The specific method can be found in [
15].
5. Discussion
Organic matter (kerogen) and clay minerals are the main components that influence gas adsorption in shale. A study on SK2 mudstone, comprising mainly purple and green mudstones and mudstone–siltstones containing almost no organic matter and only one clay mineral (2–8% chlorite), found that under the same temperature and pressure conditions used in this study (i.e., 25 °C, 18 MPa), the hydrogen adsorption capacity of the shale was 2–12 times higher than that of SK2 mudstone. This indicates the importance of organic matter and clay minerals for hydrogen adsorption.
Wang et al. compared the hydrogen absorption capacity of shale with and without organic matter (kerogen) at the same temperature (30 °C) and revealed that samples without organic matter had a considerably lower hydrogen adsorption capacity than those containing organic matter and the adsorption capacity of the sample without organic matter accounted for more than half of the sample with organic matter [
46]. Inorganic components include clay minerals, quartz and feldspar, and clay minerals, which are the main components of hydrogen adsorption. Therefore, their experimental conclusions are consistent with ours, that both organic matter and clay minerals have important effects on hydrogen adsorption capacity.
Organic matter (kerogen) plays a crucial role in the hydrogen adsorption capacity of shale, similar to its impact on methane adsorption. Previous studies on methane adsorption in shale have suggested that organic matter controls the physical and chemical properties of mud shale [
47]. High organic matter content in shale generally corresponds to a higher hydrocarbon generation potential and great adsorption capacity characterised by a high pore volume [
47,
48]. As the organic matter content increases, hydrocarbon generation strength also increases. Simultaneously, the number and types of pores developed in the organic matter vary with the organic matter content, which also affects the adsorption capacity. Many studies have indicated a positive correlation between clay mineral content and pore volume [
49]. For example, using the Chang 7 Shale Member as an example, it was proposed that the clay mineral content controlled mesopore (10–50 nm) development. Similar indirect indications were made for the TPV of <5 nm pores in the samples based on fractal dimensions [
25]. Research on marine Longmaxi formation shale suggests that clay minerals primarily affect the development of pores in the range of 2–5 and 20–100 nm. Studies on different types of pure clay minerals have found that clay minerals primarily control the development of 2–50 nm pores [
50]. Moreover, the transformation of clay minerals can affect pore structures. For instance, the transformation of montmorillonite to illite results in a reduction in pore volume and SSA. This is because montmorillonite has a larger internal SSA and pore volume than illite. A study suggested a strong correlation between the mesopore volume and total clay content [
25]. As the clay mineral content increases, the morphological transition of mesopores from slit- to bottle-shaped occurs. Clay minerals not only control mesopore volume but also influence mesopore morphology, and the total clay content is positively correlated with SSA. In our previous study on the adsorption of hydrogen by clay minerals, montmorillonite exhibited higher adsorption than chlorite, and illite and kaolinite had adsorption capacities lower than the adsorption limit [
15]. The adsorption of hydrogen in clay minerals is positively correlated with the SSA and micropore and mesopore volume [
15]. Therefore, the higher the clay mineral content, the larger the SSA, and the larger the total pore volume, the stronger the hydrogen adsorption capacity of the shale (
Table 5).
Shale is a complex material that contains abundant organic matter in the form of kerogen. Oxygen-containing groups on shale surfaces, such as alcohols and carbonyls, bind to methane more easily than hydrogen, resulting in a lower adsorption capacity of hydrogen than methane in shale [
47]. In addition, hydrogen is a nonpolar gas and, therefore, exhibits weak interactions with shale surfaces, leading to weak physical adsorption dominated by weak van der Waals forces.