A Study on the Basic Properties of Oil Shale and Its Oxidative Pyrolysis Kinetic Characteristics in an Air Atmosphere

Abstract

The in situ conversion of oil shale with air injection has the advantage of self-generated heat. The fragmentation degree of oil shale affects the oxidative pyrolysis process. In this paper, the basic properties of oil shale were analyzed, and weight loss observation and high-pressure TGA-DSC (thermogravimetric analysis and differential scanning calorimetry) tests in an air atmosphere were conducted using the cores and particles. The oil shale’s oxidative pyrolysis characteristics and the effect of its particle sizes were evaluated. The results show that the porosity and permeability conditions, TOC (total organic carbon), and inorganic mineral composition of oil shale are highly heterogeneous, with higher permeability and greater TOC along the bedding direction. The derivative of the TGA curve shows a single peak, and the heat flow curve shows a double peak that can be used to determine the oil shale’s oxidation type. The oxidative pyrolysis stage of organic matter can be divided into three temperature ranges, of which the medium temperature range is where the most combustion weight loss and heat release occurs. The activation energy of oxidative pyrolysis, which is affected by factors such as particle size, organic matter content, and pyrolysis temperature, is 46.92–248.11 kJ/mol, indicating the varying degrees of difficulty in initiating the reaction under different conditions. The pre-exponential factor is 3.15 × 10²–6.27 × 10¹¹ 1/s, and the enthalpy value is 2.575–4.045 kJ/g. The combustion indexes and reaction enthalpy under different particle sizes are more correlated with their own organic matter content. As oil shale particle size decreases, the variation law of the activation energy and pre-exponential factor changes with temperature from an initial continuous increase to a decrease, then increases again with the smallest kinetic parameters in the medium temperature zone. A small particle size, high organic matter content, and high pressure are more conducive to initiating the oxidative pyrolysis reaction to achieve in situ conversion of organic matter.

1. Introduction

China has abundant and widely distributed oil shale resources. The effective development of oil shale relies on artificial heating to promote the thermal cracking of kerogen in oil shale and to generate hydrocarbons. Conventional surface retorting technology generates a large amount of three types of waste, posing a huge threat to the ecological environment. Therefore, in situ conversion technology for oil shale has been proposed, which involves heating underground oil shale to create a temperature window for kerogen thermal cracking and the generation of oil and gas [1,2,3,4]. According to the difference in heat sources, the existing technical system can be divided into four major categories: electric heating, convection heating, radiation heating, and combustion heating [5]. Currently, methods other than electric heating technology, which is relatively well-established, still require in-depth research and field trials. In situ electric heating conversion technology has been piloted in the Green River oil shale block in the United States, yielding greater knowledge of the construction process and providing over 25 years of laboratory and shallow field experiments in oil shale, employing deeper and longer heating intervals [6]. However, the single wellbore electric heating mode has disadvantages such as a slow heat conduction rate, a small heating range, high energy consumption, and a low conversion efficiency, which limit the potential to develop this technology further. Compared with simple electric heating, the gas-assisted in situ conversion process has significant advantages, such as a greater heat transfer efficiency, enhanced formation energy, the gas drive effect, and reduced thermal cracking temperature. The injectable gases that can be selected include high-temperature N₂, CO₂, hydrocarbon gases, water vapor, and air [5]. Through oil shale reservoir stimulation, its porosity and permeability can be improved to promote the seepage of injected fluid, facilitate heat transfer, increase the contact area with organic matter, and enable the convenient production of cracked oil and gas.

Air-assisted in situ conversion technology has the unique advantage of self-generated heat for thermal cracking in oil shale formation, which is worthy of in-depth study and field applications. Regarding the characteristics of air injection for in situ combustion (high-temperature oxidation, HTO) in heavy oil reservoirs and low-temperature oxidation (LTO) processes in light oil reservoirs, the oxygen in the injected air can react with organic matter in oil shale to generate heat, which helps to increase the formation temperature, achieving in situ conversion. The generated flue gas can increase the formation energy and exhibits a gas drive effect. In practice, the near-well oil shale is first preheated to the ignition temperature through electric heating or hot fluid injection; then, the air is injected to react with the produced coke and release heat [7]. Successful ignition, controlling the pyrolysis temperature, maintaining the combustion front stability, and preventing gas channeling to reach a high oil and gas production rate are the key aims. It should be noted that air injection technology has been widely applied in oil reservoirs [8,9,10,11,12,13]. The research methods used in this field are relatively well-established, and experimental and numerical simulation techniques can provide a good basis for the study of air injection technology in oil shale.

Currently, only a handful of studies on in situ conversion of oil shale with air injection are available, and most of these studies are laboratory-based [14,15,16,17,18,19]. Regarding oil shale’s oxidative pyrolysis characteristics, Syed et al. (2011) conducted TG (thermogravimetric) experiments on oil shale samples in a nitrogen atmosphere at different heating rates; divided the pyrolysis process into moisture release, devolatization, and evolution of fixed carbon/char; and obtained the kinetic parameters [20]. Bai et al. (2015) studied the thermokinetic characteristics of the pyrolysis and combustion of oil shale from Huadian, Fushun, and Nong’an in Northeast China, finding that the diversity of organic and mineral components in oil shale significantly affects the thermal behavior and kinetic characteristics of the respective samples [21]. El-Rub et al. (2020) conducted laboratory experiments on Omari shale in Jordan, performing TGA and DSC tests at different heating rates, divided the pyrolysis stages, and obtained the activation energy and frequency factor, which can be employed in developing pyrolysis reactor models [22]. Zhao et al. (2021) conducted high-pressure DSC tests to study the oxidation characteristics of shale oil, and evaluated the effects of adding quartz sand or shale debris and pressure. The results suggested that shale detritus generated a strong catalytic effect on reducing combustion difficulty, and spontaneous ignition may be achieved during shale oil oxidation [23]. Ifticene et al. (2022) conducted TGA tests combined with Fourier transform infrared spectroscopy (FT-IR) to investigate the conversion/combustion behavior, kinetic characteristics, and differences in oil shale and its components (asphaltene, kerogen, and oil shale without asphaltene) under an air atmosphere, to gain a deeper understanding of the combustion/conversion mechanism of oil shale [24]. Zhao et al. (2024) studied the oxidation characteristics of shale oil through DSC tests and then used accelerated rate calorimetry (ARC) to assess the self-ignition potential of shale oil under different oxygen concentrations, finding that the activation energy of LTO is lower and that of HTO is higher under high-pressure compared to low-pressure conditions. An increase in the oxygen concentration is beneficial for shortening the oxidation induction period and enhancing the combustion intensity [25]. These studies have investigated the pyrolysis characteristics of different oil shales in N₂ and air atmospheres through TGA and DSC tests. The effects of the heating rate, oxygen concentration, and pressure were mainly evaluated. However, a comprehensive understanding has not yet been achieved, for example, on the influence of the oil shale’s particle size, due to a lack of research. The in situ conversion of oil shale with air injection should be conducted employing formation fracturing or crushing. The oil shale’s particle size affects the contact area between air and oil shale as well as the diffusion time of air within the oil shale, thereby influencing its oxidative pyrolysis characteristics. Deepening the understanding of the oxidative pyrolysis mechanism can provide a solid foundation for research on the exothermic oxidation reaction that occurs under air injection, thermal convection caused by gas flow, porosity variation due to organic matter conversion, and their coupling effects.

In this study, oil shale from the Chang 7 section of the Changqing Oilfield was selected for lab experiments. Based on the oil shale’s basic property analysis, weight loss observation experiments and high-pressure TGA-DSC tests in an air atmosphere were conducted using cores and powders with different particle sizes. The weight loss process of the oil shale under the oxidative pyrolysis effect was divided into stages, and the activation energy, pre-exponential factor, and enthalpy value were obtained. The influence of particle size on the oil shale’s oxidative pyrolysis characteristics was evaluated. The results show that a small particle size, high organic matter content, and high pressure are more conducive to initiating the oxidative pyrolysis reaction to achieve in situ conversion of organic matter. These results also provide the impetus for the reservoir stimulation of oil shale, and lay a foundation for the future research to prove the feasibility of air injection development.

2. Experimental Equipment and Methods

2.1. Preparation of Experimental Materials

Laboratory experiments were conducted using outcrop oil shale rock samples from the Chang 7 section of the Changqing Oilfield. The sample processing procedure is shown in Figure 1. Two planes were cut from the rock sample for core drilling. At five positions parallel to the bedding (P1–P5) and five positions perpendicular to the bedding (C1–C5), two standard core columns were taken by drilling at each location, totaling 20 core columns, each with a size of φ25 × 50 mm. To prevent hydraulic cutting from reservoir damage, wire cutting was used throughout the process. At the same time, the debris at these sampling positions was mixed evenly and ground and sieved into five types of particle sizes: <60 mesh, 60–80 mesh, 80–100 mesh, 100–120 mesh, and >120 mesh.

2.2. Analysis of Basic Properties of Oil Shale

Porosity and permeability test method: The porosity and permeability of oil shale core samples were measured using an automatic porosity and permeability measuring instrument. The porosity was determined by helium gas pore volume measurement, and the permeability was measured by nitrogen gas permeation. Before testing, the core samples were pre-treated with oil washing and drying. The tests were conducted at normal temperature and pressure, with a driving pressure difference of 1–1.5 MPa and an effective confining pressure of 4–5 MPa. A total of 20 core samples were tested.

Total organic carbon analysis method: A total organic carbon (TOC) analyzer was used to convert the organic matter in the samples into CO₂ through thermal decomposition or chemical oxidation, and then the amount of generated CO₂ was measured to calculate the organic carbon content in the samples. A total of 5 samples were tested. Each sample was tested three times, and the average value and standard deviation were calculated.

XRD whole rock analysis method: The rock samples were ground to a particle size of 320 mesh or finer, placed in the sample chamber of the X-ray diffractometer (XRD) spectrometer (X’ Pert PRO MPD, PANalytical Company in Almelo of the Netherlands), and the instrument parameters were set to automatically scan and obtain the diffraction pattern. For clay mineral testing, the clay fraction was extracted from the oil shale powder for clay mineral composition determination. Different minerals have distinct crystal structures, which show different 2θ angles on XRD patterns, and by identifying the 2θ angles of these characteristic peaks, the mineral types in the sample can be quickly determined. The Rietveld full-profile refinement method was used for quantitative phase analysis. Based on the crystal structure model, a “theoretical” diffraction pattern was calculated, and parameters such as scale factors, unit cell parameters, and peak shape parameters were continuously adjusted through the least squares method to achieve the best fit between the theoretical and experimental patterns, thereby determining the precise content of each mineral. A total of 5 samples were tested, and to ensure the reliability of the results, each test was conducted three times to guarantee consistent outcomes.

SEM observation method: A Zeiss EVO LS 15 scanning electron microscope (ZEISS Group Global in Oberkochen, Germany) was used to observe the morphology of kerogen and micro-fractures in oil shale at an acceleration voltage of 6 kV and a magnification of 500 times. The morphology of typical minerals in oil shale was observed at an acceleration voltage of 10–15 kV and a magnification of 9–15 k. The spatial resolution was 3.0 nm @ 30 kV. During the experiment, a thin layer of conductive material was sprayed onto the surface of the rock sample, which was then placed in the sample chamber of the scanning electron microscope (SEM) and evacuated. The instrument was turned on and aligned, and an optical microscope was used to select the observation area and gradually increase the magnification. The target area was locked and images were captured for microstructure analysis. While observing the samples through SEM, EDS (Energy-Dispersive X-ray Spectroscopy, ZEISS Group Global in Oberkochen, Germany) analyses were also conducted. Since SEM observation usually requires the surface of rock samples to be sputtered with conductive substances for a clearer view of the microstructure, this interferes with EDS analysis. Therefore, no conductive substances were sputtered when conducting the EDS analysis on typical minerals, so the corresponding SEM images are somewhat blurry.

2.3. Observation of Weight Loss of Heated Oil Shale

Observation of heating oil shale powder: 30 g of oil shale powder (80–100 mesh) was placed into a muffle furnace for heating (the maximum working temperature is 1100 °C, and the temperature control accuracy is 0.1 °C). The color change of the oil shale powder was observed and its mass was weighed at different temperatures (120, 200, 300, 400, 500, 600 °C) after it had been heated for 1–2 h. Then, the weight loss law of the powder was analyzed. Two sets of experiments were conducted simultaneously for comparative analysis.

Observation of heating oil shale core: Core samples P2-1 (parallel to the bedding) and C3-1 (perpendicular to the bedding) were weighed before and after being dried at 80 °C for 3 h. Then, they were placed into a muffle furnace and heated at 450 °C for 6 h for oxidative pyrolysis. After being taken out, they were weighed, and the changes in their appearance and morphology were observed. The mass loss of the two core samples was analyzed.

2.4. TGA and DSC Tests of Oil Shale in Air Atmosphere

The oxidative pyrolysis reaction kinetics characteristics of oil shale in an air atmosphere were studied using a thermal gravimetric analyzer and a differential scanning calorimeter. The thermal gravimetric analyzer (Themy HP TGA with a maximum working temperature of 1100 °C and a maximum working pressure of 15.2 MPa) was used to measure the mass change of the sample during the oxidative pyrolysis process, while a differential scanning calorimeter (PDSC-Q20 with a maximum working temperature of 725 °C and a maximum working pressure of 10 MPa) was used to measure the sample’s heat absorption and heat release under the same conditions. The experimental procedures of the TGA and DSC tests were similar: (1) Turn on the equipment for preheating to ensure that it reaches a stable state. (2) Set parameters such as temperature range, heating rate, gas flow rate, pressure, etc. Select air as the shielding gas and run the blank experiment (empty crucible) under set conditions to subtract background values. (3) Place the sample in the crucible and avoid moving the sample during heating. After confirming that all parameters are set correctly, begin the experiment and closely monitor the operation status and the data collected during the experiment. (4) After the experiment, export the experimental data, including temperature, time, sample mass, and heat flow rate, for processing and analysis.

Five different oil shale particle sizes were evaluated in the experiment, including <60 mesh, 60–80 mesh, 80–100 mesh, 100–120 mesh, and >120 mesh. When conducting TGA and DSC tests, using an insufficient amount of sample can result in weak signals, while using too much can cause reaction lag and reduced resolution. Generally, the sample amount should not exceed 10–15 mg, but for heterogeneous samples, it is recommended to use more than 20 mg. Therefore, in this experiment, 40 mg was used for TGA testing and 30 mg for DSC testing. Using an appropriate flow rate is crucial for prompt removal of decomposition products and for maintaining a stable testing environment. Too low of a flow rate can cause decomposition products to accumulate in the sample chamber, affecting heat transfer and the reaction equilibrium, while too high of a flow rate can cause unnecessary disturbance to the microbalance. A standard flow rate of 50 mL/min was chosen for the actual test, which can effectively purge the reaction chamber and maintain system stability. The test pressure was set at 3 MPa, mainly considering the formation pressure exerted by the sample and the equipment’s pressure resistance. The test temperature range should cover the entire thermal decomposition weight loss process of the sample until a stable state is reached. The final determined test temperature ranges for TGA and DSC were 50–600 °C and −50–550 °C, respectively, and the heating rate was set at 10 °C/min. A rate that is too fast may cause thermal lag and overlapping decomposition peaks, while a rate that is too slow will prolong the test time. The oil shale samples used in this study showed significant differences in TOC depending on the sampling direction. Although the rock sample amount was increased in TGA and DSC tests, the test results may still have been affected by the heterogeneous TOC. To ensure representativeness, rock cuttings from different drill holes were thoroughly mixed and ground, and experiments were repeated. Each test was repeated at least three times for each particle size, and the most representative results were selected for analysis.

3. Experimental Results and Analysis

3.1. Basic Properties of Oil Shale

3.1.1. Porosity and Permeability

The measured porosity and permeability of the 20 core samples are shown in Figure 2a. For the cores drilled parallel to the bedding (with the prefix “P”), the porosity ranges from 2.395% to 5.909%, and the permeability values of six of the cores are within the range 0.00022 to 0.0028 md, confirming a good exponential relationship between porosity and permeability (Figure 2b). The permeability of cores P3-1 and P5-3 is relatively high, ranging from 0.022 to 0.12642 md. Cores P4-1 and P4-3 have micro-fractures due to the geological structure of their sampling location, with permeability values reaching 3.40198 and 5.05102 md, respectively. For the cores drilled perpendicular to the bedding (with the prefix “C”), the porosity ranged from 0.751% to 5.291%, but the permeability was extremely low and could not be measured under the conditions utilized herein. It can be concluded that the permeability of oil shale is significantly affected by the bedding direction, mainly showing greater permeability along the natural bedding direction. The heterogeneity in both the plane and the vertical direction is significant; the porosity and permeability of the cores obtained from the same borehole (vertically) and adjacent boreholes (horizontally) show obvious differences. The development of micro-fractures in local areas further exacerbates the heterogeneity of the reservoir.

3.1.2. TOC

The TOC of the oil shale samples from the five boreholes is shown in Figure 3, ranging from 9.70 wt% to 16.06 wt%, with an average of 12.43 wt% (the standard deviation ranges from 0.2256% to 0.9888% and errors range from −1.09% to 1.31%). The TOC is far greater than 2 wt%, indicating that the rock source is of a high quality. The TOC at boreholes P1 and P2 (15.94–16.06 wt%) is significantly higher than that at boreholes C3-C5 (9.70–11.02 wt%). This is because the lighter organic matter particles tend to accumulate parallel to the sedimentary layer, resulting in the enrichment of organic matter along the layer’s direction during the original sedimentation. Organic matter and inorganic minerals alternate in enrichment on the bedding plane, forming a clearly visible bedding structure in the oil shale. This sample has a medium-level TOC. The TOC of oil shale in China mainly ranges from 5 wt% to 20 wt%, but some oil shales such as those in the Huadian region, Jilin Province, China, have TOCs as high as 33.31 wt% [5].

3.1.3. Inorganic Mineral Composition

The total rock composition of the oil shale samples from five drill holes is shown in Figure 4a. The results indicate that the samples are from a low-energy environment with mixed organic-rich sources, possibly from deep lake or deep sea sedimentary environments. The content of clay minerals, which are the main components of the shale, ranges from 41 wt% to 56.7 wt%, with an average of 47.38 wt%. Among them, illite accounts for 53 wt%, the illite–smectite mixed layer accounts for 38 wt%, and there are also small amounts of chlorite and kaolinite (the standard deviation ranges from 1.013 wt% to 2.546 wt% and error ranges from −3% to 4%) (Figure 4b). The content of pyrite, a sulfur-containing iron mineral, ranges from 12.5 wt% to 17.7 wt%, with an average of 15.3 wt%, indicating that there were abundant organic matter and reducing conditions in the sedimentary environment that promoted the deposition of sulfide iron minerals. The quartz content ranges from 11.0 wt% to 16.5 wt%, with an average of 13.54 wt%, which is much lower than the quartz content of most shales (30 wt% to 40 wt%), suggesting that the samples may have come from a relatively still water sedimentary environment, such as a lake or slow-flowing river. The average contents of carbonate minerals such as calcite, siderite, and ankerite are 4.62 wt%, 12.76 wt%, and 2.8 wt%, respectively, with a total of 20.18 wt%, indicating that the oil shale formed in an anoxic, iron-rich reducing environment and underwent moderate diagenesis, increasing the brittleness of the rock. The EDS test results show that the rock sample contains a large amount of elements such as Al, Si, and O, a medium amount of elements such as Fe and S, and a small amount of elements such as Ca and Mg. This is in good agreement with the XRD analysis results suggesting that the rock sample contains a large amount of clay minerals and a medium amount of FeS and carbonate minerals.

3.1.4. Microscopic Morphology

The microscopic morphology of the oil shale is shown in Figure 5. The main features are as follows: layered clay minerals (such as illite and illite–smectite mixed layer) form the base; high-brightness pyrite “strawberry balls” are scattered among them; the gray blocks are organic matter adhering to the mineral surface, with dark nano-scale pores distributed on their surfaces or inside them; the angular particles are dispersed quartz, feldspar, and other detrital particles; and the black gaps are intergranular pores and micro-fractures between the mineral particles or at the edges of the organic matter. The EDS analysis indicates that the rock sample is mainly composed of 43.09% O, 30.56% C, 12.66% Si, 5.01% Al, 3.35% Fe, 2.39% S, 0.49% Mg, and 0.28% Ca. The observed microscopic mineral types are consistent with its elemental composition and are also in good agreement with the XRD analysis results.

3.2. Weight Loss Phenomenon of Heated Oil Shale

The weight loss rate and color changes of the two oil shale powder samples during the heating process are shown in

Table 1. Weight loss rate of oil shale powder during heating.

Temperature, °C	Heating Time, h	Sample A			Sample B
		Mass, g	Stage Weight Loss, %	Cumulative Weight Loss, %	Mass, g	Stage Weight Loss, %	Cumulative Weight Loss, %
Room	/	31.02	0	0	30.02	0	0
120	1	30.63	1.26	1.26	29.61	1.37	1.37
200	2	30.6	0.10	1.35	29.53	0.27	1.63
300	2	30.27	1.06	2.42	29.01	1.73	3.36
400	2	27.61	8.58	10.99	26.21	9.33	12.69
500	2	25.7	6.16	17.15	24.45	5.86	18.55
600	2	24.9	2.58	19.73	23.71	2.47	21.02

and Figure 6. The samples lost 1.26–1.37 wt% of their weight when heated at 120 °C for 1 h, mainly due to the loss of moisture. At 200 °C, the mass reduction was the smallest, at only 0.1–0.27 wt%. At 400 °C, the mass reduction was the largest, at up to 8.58–9.33 wt%. After heating at 600 °C, the total mass reduction was 19.73–21.02 wt%, averaging around 20 wt%. The oil shale was dark brown at room temperature, but when heated at 300 °C and 400 °C, the color turned to dark gray. At 500 °C, the color became grayish white, and at 600 °C, it changed to brick red because pyrite underwent an oxidation reaction to form hematite, namely, 4FeS₂ + 11O₂ → 2Fe₂O₃ + 8SO₂.

The weight loss rate and color change of two oil shale cores during the heating process are shown in

Table 2. Weight loss rate of oil shale cores during heating.

Sample Code	Sampling Direction	Diameter, cm	Length, cm	Porosity, %	Permeability, md	Original, g	After Drying, g	After Heating, g	Weight Loss Rate, %
P2-1	Parallel to bedding plane	2.483	5.064	5.587	0.0028	47.067	45.893	33.775	26.4
C3-1	Perpendicular to bedding plane	2.480	5.002	4.731	0	47.696	46.257	31.826	31.2

and Figure 7. After being heated at 450 °C for 6 h, the core as a whole turned red-brown, and the light-colored minerals remained unchanged. After oxidation, the core is prone to fragmentation and cannot maintain its integrity, so no porosity and permeability measurements were conducted. Research indicates that both the combustion of organic matter and the decomposition of inorganic matter lead to material loss and pore formation. The uneven composition causes uneven thermal expansion, and when the thermal stress exceeds the cementation strength of the matrix or that of the mineral grains themselves, transgranular and intergranular fractures occur, that is, micro-fractures are formed. Eventually, a pore communication network is created, and the new pores are mainly medium- and large-sized ones [7,26]. Core C3-1 was taken perpendicularly to the bedding, and during the heating process, a large number of bedding fractures appeared on the side of the core, which was conducive to the entry of air that reacted with the organic matter. The weight loss rate of this core reached 31.2 wt%, which was higher than that of core P2-1, at 26.4 wt%. The weight loss rates of both cores exceeded the organic matter content (about 16 wt%). Compared with the powder sample experiments, it was found that the smaller surface area of the core was not conducive to the volatilization of the cracked oil and gas that were retained for combustion, and the actual combustion temperature might have reached over 600 °C, causing the loss of adsorbed water and interlayer water and inorganic minerals (including carbonates and clay minerals) to decompose, resulting in a significant increase in the weight loss rate. According to the mineral composition, the weight loss ratio of the three carbonate minerals during pyrolysis ranges from 31.03% to 44%. Considering the content of these carbonate minerals in oil shale, the maximum weight loss due to their pyrolysis can reach up to 9.80%. Additionally, the weight of the clay minerals also decreases at high temperatures due to dehydration and dehydrogenation.

The oxidative pyrolysis of oil shale has an impact on its physical and mechanical properties. Some studies have shown that as the pyrolysis temperature increases, the permeability in the direction perpendicular to the bedding can increase by 1–4 orders of magnitude, and the permeability in the direction parallel to the bedding can reach 2–4 orders of magnitude higher than this [27,28]. The compressive strength and elastic modulus of oil shale in both directions first decrease and then increase with the rise in temperature, reaching their minimum values at 400 °C. The compressive strength in the direction perpendicular to the bedding is higher than that parallel to the bedding, while the elastic modulus shows the opposite trend [28]. Research shows that below 400 °C, as the temperature rises, organic matter pyrolyzes, porosity increases, and cementation strength decreases. However, at 400–600 °C, clay minerals will sinter, making them stronger. This is the key reason enabling the oil shale’s compressive strength to be recovered [29].

3.3. Kinetic Characteristics of Oxidative Pyrolysis Reaction of Oil Shale

3.3.1. Weight Loss and Heat Release Characteristics of Oil Shale

The TGA and DSC curves of oil shale powder with different particle sizes in an air atmosphere are shown in Figure 8a,b. The TGA shows the change in weight loss ω, which is the change in the ratio of the shale sample’s weight m to its initial weight m_o with temperature. The characteristics of the weight loss and heat flow of oil shale are very different from those of conventional crude oil. The organic matter in oil shale, kerogen, is solid and difficult to separate from inorganic minerals. Therefore, oil shale is processed into a powder to examine its oxidative pyrolysis characteristics. The weight loss curve (ω) with temperature first decreases then increases, followed by a sharp and rapid decrease, and finally becomes flat (the derivative dω/dT mainly shows only one weight loss peak), while the heat flow curve has a double-peak feature. The weight loss and heat release characteristics at different temperature stages and the corresponding physical and chemical reactions are shown in

Table 3. General characteristics of weight loss and heat release of oil shale with temperature increase.

Stage		Temperature Range	TGA Signal	DSC Signal	Corresponding Response
Dehydration		25–200 °C	Weight declines	Endothermic peak	Physical dehydration
Oxidative pyrolysis of organic matter	LTO	200–300 °C	Weight rises first and then falls	Exothermic peak	Oxygen addition and carbon removal reaction of kerogen and simultaneous polymerization to form oxidized coke
	HTO	300–500 °C	Significant weight loss	Strong exothermic peak	Kerogen pyrolysis generates hydrocarbons and coke and combusts
Oxidative pyrolysis of inorganic matter	Oxidation of pyrite	500–600 °C	Weight declines	Exothermic peak	4FeS2 + 11O2 → 2Fe2O3 + 8SO2
	Carbonate decomposition	600–800 °C	Weight declines	Endothermic peak	Such as CaCO3 → CaO + CO2↑(gas)

. Generally, from room temperature to 200 °C, physical dehydration mainly occurs, with a low weight loss rate, showing an endothermic peak. From 200 °C to 500 °C, the oxidative pyrolysis of organic matter mainly occurs; from 200 °C to 300 °C, the LTO reaction mainly occurs (the oxygen addition and carbon removal reaction of kerogen can result in the sample mass first increasing then decreasing, and simultaneously polymerizing to form some oxidized coke), and from 300 °C to 500 °C, the HTO reaction mainly occurs (direct kerogen pyrolysis generates hydrocarbons and coke and combusts, resulting in a significant decrease in sample mass), corresponding to two heat release peaks. The division of pyrolysis into the LTO and HTO stages can also be supplemented by gas analysis. During the HTO stage, a large amount of CO₂ is produced; future research should take this into account. During the HTO stage, the combustion of kerogen leads to significant weight loss and heat release, and is the main weight loss stage of the entire oxidative pyrolysis process, consistent with the results of numerous studies [30]. When the temperature exceeds 500 °C, the oxidative pyrolysis of inorganic matter mainly occurs, with a slow decrease in mass; from 500 °C to 600 °C, pyrite oxidation mainly occurs, reflected as a small heat release peak, and from 600 °C to 800 °C, carbonate decomposition mainly occurs, showing an endothermic peak. Finally, due to the large amount of residual inorganic minerals remaining after the experiment, the weight loss rate cannot reach 100% [30,31,32]. In contrast, the weight loss curve of crude oil mainly shows two stages of decline, with the derivative showing two weight loss rate peaks, corresponding to two heat release peaks. Between the two peaks, the coke formed by the coking reaction temporarily decreases the contact area between air and the internal crude oil, resulting in a slowdown in weight loss and heat release. The entire process mainly consists of four stages: distillation, LTO, medium-temperature oxidation (MTO), and HTO (

Table 4. General characteristics of weight loss and heat release of crude oil with temperature increase.

Stage	Temperature Range	TGA Signal	DSC Signal	Corresponding Response
Distillation	≤120 °C	Weight declines	Endothermic peak	Light hydrocarbon volatilization
LTO	120–200 °C	Weight declines	Exothermic peak	Oxygenation and decarboxylation reaction, releasing CO2
MTO	200–400 °C	Significant weight loss	Strong exothermic peak	Heavy component bond breaking and polycondensation reaction generates light oil, oxidized coke, and CO2
HTO	≥400 °C	Significant weight loss	Strong exothermic peak	Thermal cracking of resins and asphaltenes generates hydrocarbons and coke that burn

). At the end of the experiment, the weight loss rate can reach or approach 100% [33].

The weight loss and exothermic characteristics of the samples are sensitive to particle size. Generally speaking, the smaller the particle size of the powder, the more complete the pyrolysis and combustion of oil shale, the greater the weight loss rate, and the more likely it is to show bimodal exothermic features. It has been found that the smaller the particle size, the larger the contact surface area with air, the more easily the air diffuses into the interior of the particles to contact the organic matter, and the more likely the interior temperature of the particles is to approach the heating rate. During the heating process, the LTO reaction is more complete, and the amount of heat released is greater, being even higher than the peak value of the HTO heat released, indicating that the smaller the particle size, the more conducive it is to the initiation of the conversion of organic matter at a lower temperature. Correspondingly, when the particle size of oil shale is larger, for example, less than 60 mesh, it shows only single-peak exothermic features. The main reason for this is that due to the larger particle size, it is more difficult for air to enter the interior of the particles, so the LTO mainly occurs on the particle surface and at an insufficient rate, resulting in lower heat release. As the temperature rises, HTO is triggered, and the liquid or gaseous hydrocarbons produced by the conversion of organic matter and the generated coke are burned, resulting in a significant exothermic phenomenon. It should be noted that the weight loss rate and heat release are not strictly positively correlated with the particle mesh size. There are some abnormal data points that do not conform to this rule, such as the sample with a particle size of 80–100 mesh having the smallest weight loss rate. This may be due to the lower content of organic matter and other pyrolysable inorganic minerals in the sample. Li et al.’s results show that the weight loss rate of different oil shale samples during pyrolysis has good consistency with their TOC [5]. As described previously, the TOC and the oxidizable and pyrolysable inorganic mineral content in oil shale are heterogeneous. If the samples are not mixed thoroughly during preparation, and the amount used in each test is small, differences in the composition of the tested samples will be much more likely.

3.3.2. Division of Oxidative Pyrolysis Process of Oil Shale into Stages

(1) Determination of combustion characteristic parameters

To precisely divide the oxidative pyrolysis of oil shale with different particle sizes into stages, the derivative of the TGA curve (ω) was first calculated to obtain the DTGA curve (dω/dT). Then, the ignition temperature T_i, burnout temperature T_f, maximum derivative of TGA curve (dω/dT)_max and its corresponding temperature T_max, and average derivative of TGA curve (dω/dT)_mean between T_i and T_f were determined using the TGA-DTGA extrapolation method, where (dω/dT)_max and (dω/dT)_mean take the positive value. Taking the TGA curve of particles smaller than 60 mesh in Figure 9 as an example, firstly, (dω/dT)_max was determined to be 0.142%/°C according to the DTGA curve’s peak, and a vertical line was drawn through this point to obtain T_max (370.3 °C) and intersect with the TGA curve. Secondly, a tangent line was drawn at the intersection point on the TGA curve, and this was intersected with the tangent lines before the first inflection point and after the second inflection point of the TGA curve, respectively; then, vertical lines were drawn through these two intersection points to obtain T_i (328.1 °C) and T_f (446.3 °C), respectively. Finally, (dω/dT)_mean was obtained as 0.098%/°C by averaging the dω/dT between T_i and T_f. Further, the combustion indexes C_b, G, and S_N of oil shale can be calculated using the following specific equations [19,28]:

$$C_b={\displaystyle \frac{{(d\omega /dT)}_{max}}{T_i^2}}$$

(1)

$$G={\displaystyle \frac{{(d\omega /dT)}_{max}}{T_i\cdot T_{max}}}$$

(2)

$$S_N={\displaystyle \frac{{(d\omega /dT)}_{max}\cdot {(d\omega /dT)}_{mean}}{T_i^2\cdot T_f}}$$

(3)

In the equations, C_b is the combustibility index, reflecting the difficulty of fuel ignition. G is the combustion stability index, which is a combined reflection of the ignition difficulty and the intensity of combustion. The larger the G value, the more stable the combustion. S_N is the comprehensive combustion characteristic index, which reflects the combined characteristics of fuel ignition and burnout. The higher the S_N value, the better the combustion characteristics of the fuel.

The combustion characteristic parameters of oil shale with different particle sizes are shown in

Table 5. Combustion characteristic parameters and stages of oxidative pyrolysis of oil shale with different particle sizes.

Particle Size, Mesh	Ti, °C	Tf, °C	Tmax, °C	(dω/dT)max, %/°C	(dω/dT)mean, %/°C	Cb	G	SN	Stage I: Dehydration, °C	Stage II: Oxidative Pyrolysis of Organic Matter, °C			Stage III: Oxidative Pyrolysis of Inorganic Matter, °C
										Low Temp.	Medium Temp.	High Temp.
<60	328.1	446.3	370.3	0.142	0.098	1.32 × 10−6	1.17 × 10−6	2.88 × 10−10	<180	180–328	328–446	446–500	>500
60–80	323.6	426.5	365.1	0.136	0.098	1.30 × 10−6	1.15 × 10−6	2.98 × 10−10	<153	153–324	324–427	427–500	>500
80–100	319.1	424.7	377.4	0.100	0.072	9.82 × 10−7	8.30 × 10−7	1.67 × 10−10	<183	183–319	319–425	425–500	>500
100–120	301.1	456.7	379.5	0.123	0.072	1.36 × 10−6	1.08 × 10−6	2.13 × 10−10	<171	171–301	301–457	457–500	>500
>120	312.7	429.1	367.4	0.175	0.111	1.79 × 10−6	1.52 × 10−6	4.61 × 10−10	<161	161–313	313–429	429–500	>500

and Figure 10. It can be seen that T_i with different particle sizes is between 301.1 and 328.1 °C, T_f is between 424.7 and 456.7 °C, and T_max is between 365.1 and 379.5 °C. These key temperatures are relatively stable and are less affected by particle size. For the combustion indexes C_b, G, and S_N, the overall trend is that they do not change much with the decrease in particle size, and only increase significantly when the particle size is greater than 120 mesh. When the particle size is 80–100 mesh, both the (dω/dT)_max and (dω/dT)_mean are small, resulting in small values of the three combustion indexes. When the particle size is 100–120 mesh, the (dω/dT)_mean is small, resulting in a small index S_N. It is found that the three combustion indexes have a notably positive correlation with the weight loss rate or organic matter content with different particle sizes. Therefore, it can be confirmed that the combustion characteristics of oil shale are jointly affected by particle size and organic matter content, but mainly determined by the latter. Oil shale with a small particle size and high organic matter content shows better performance in terms of combustibility, stability, and overall performance.

(2) Temperature range division of oxidative pyrolysis process

Based on the combustion characteristic parameters, the oxidative pyrolysis stages of oil shale can be more precisely divided. The main process is as follows: (a) The upper limit temperature of the dehydration stage is determined by the inflection point where the DSC curve initially changes from endothermic to exothermic and the inflection point where the weight loss curve initially decreases and then increases. (b) The upper limit temperature of the organic matter oxidative pyrolysis stage is determined by the inflection point where the DTGA curve becomes horizontal after reaching its peak. (c) The organic matter oxidative pyrolysis stage can be further divided into three stages for the low-, medium-, and high temperature ranges according to T_i and T_f, so as to obtain the reaction kinetic parameters of these three typical stages in the subsequent process. Among them, the medium temperature range encompasses temperatures between T_i and T_f, denoted the temperature range T_df, at which combustion mainly occurs and the maximum weight loss rate occurs, corresponding to the HTO exothermic peak; in the low temperature range, encompassing temperatures less than T_i, the LTO reaction mainly occurs, and the weight loss curve first increases and then decreases, corresponding to the LTO exothermic peak; in the high temperature range, encompassing temperatures higher than T_f, oxidative pyrolysis of residual organic matter occurs. In addition, the temperature range at which significant weight loss occurs in the organic matter oxidative pyrolysis stage is denoted T_do, which is greater than T_df. The oxidation pyrolysis stage division results are shown in Figure 9 and Table 5. At different particle sizes, the upper limit temperature of the dehydration stage (that is, the lower limit temperature of the organic matter oxidative pyrolysis) is 153–183 °C, and the upper limit temperature of the organic matter’s oxidative pyrolysis stage is 500 °C. Within these, the upper limit of the low temperature range is 301–328 °C, and the upper limit of the medium temperature range is 425–457 °C. When the temperature is greater than 500 °C, inorganic matter enters the oxidative pyrolysis stage. Compared with high pressure, some studies show that the ignition temperature of organic matter under normal pressure and an air atmosphere will increase, and the combustion temperature range will expand [19,31]. High pressure is conducive to ignition and in situ conversion start-up.

3.3.3. Kinetic Parameters of Oxidative Pyrolysis Reaction of Oil Shale

(1) Activation energy and pre-exponential factor

The oxidative pyrolysis of organic matter is the core stage of oil shale’s oxidative pyrolysis, which determines the main generation process of shale oil and gas. Therefore, the reaction kinetics parameters of the oxidative pyrolysis of organic matter were calculated. The kinetic model of the oil shale oxidative pyrolysis reaction is usually based on TGA data and uses the Arrhenius equation as its basis [19]. The Arrhenius equation describes the relationship between the reaction rate of a chemical reaction and temperature, and the equation is as follows:

$$k=A\cdot e^{-{\displaystyle \frac{E_a}{RT}}}$$

(4)

Taking the logarithm on both sides,

$$lnk=lnA-{\displaystyle \frac{E_a}{RT}}.$$

(5)

Then, k can be obtained by using the Friedman method, and the specific equation is as follows:

$$k=\beta {\displaystyle \frac{d\alpha }{dT}}{\displaystyle \frac{1}{{(1-\alpha )}^n}}$$

(6)

where k is the reaction rate, 1/s; A is the pre-exponential factor, 1/s; E_a is the activation energy, J/mol; R is the gas constant, 8.314 J/(mol·K); T is the reaction temperature, K; α is the conversion rate, a decimal, indicating the reaction degree of the sample, α = (m_o − m)/(m_o − m_f), where m_o, m_f, and m are the initial mass, final mass, and mass at a certain temperature of the oil shale sample, respectively, mg; n is the reaction index, which was assigned a value of 1; and β is the heating rate, °C/s. Specifically, the conversion rate of the oil shale was calculated based on the TGA curve; then, the conversion rate of the oxidative pyrolysis stage of organic matter was substituted into Equation (6) to calculate k. According to Equation (5), a scatter diagram of lnk versus 1/T was plotted and fitted to obtain the linear formula y = ax + b; then, −E_a/R = a, lnA = b, and thereby the activation energy E_a and the pre-exponential factor A were obtained.

The conversion rates of organic matter in oil shale with different particle sizes and at different temperature ranges during oxidative pyrolysis, as well as the activation energy and pre-exponential factor, are shown in Figure 11. According to the calculation, the total weight loss rate of samples with different particle sizes is 11.23–16.77 wt%, among which the weight loss rate during the oxidative pyrolysis of organic matter is 9.82–15.73 wt% (consistent with the TOC). In the low temperature range, the weight loss rate is 0.15–0.90 wt%, corresponding to a conversion rate of 1.34–6.51%; in the medium temperature range, the weight loss rate is 8.64–14.82 wt%, corresponding to a conversion rate of 76.94–88.37%; and in the high temperature range, the weight loss rate is 0.55–1.48 wt%, corresponding to a conversion rate of 3.28–9.82 wt%. The reaction activation energy, affected by factors such as particle size, organic matter content, and pyrolysis temperature, is 46.92–248.11 kJ/mol, and the pre-exponential factor is 3.15 × 10²–6.27 × 10¹¹ 1/s. Some studies have shown that the activation energy of the pyrolysis reaction of oil shale in an inert gas atmosphere is generally high (above 200 kJ/mol), and the pre-exponential factor is generally large (at an order of 10¹¹–10¹³ 1/s). Relatively, the activation energy and pre-exponential factor of the oxidative pyrolysis reaction in an air atmosphere are small, with a wide range of variation. The activation energy is in a range from several dozen to one or two hundred kJ/mol, and the pre-exponential factor is generally below the order of 10⁸ 1/s, but sometimes it can be as low as 10³–10⁴ 1/s [30,31,32]. The results of this study are overall consistent with this. Further analysis shows the following:

(a) When the particle size is relatively large, as the temperature increases, the activation energy and pre-exponential factor gradually increase, indicating that the higher the temperature, the more difficult it is to initiate the oxidative pyrolysis of residual organic matter (such as semi-coke). As the particle size decreases, the LTO reaction in the low temperature range proceeds to near completion and the oxygen addition reaction leads to a greater increase in sample mass, while the weight loss due to the decarbonization reaction is smaller. This means that the activation energy and pre-exponential factor calculated based on the conversion rate in the low temperature range are higher, and specifically, when the particle size is >120 mesh, it reaches its maximum. Eventually, the activation energy and pre-exponential factor decrease first and then increase with the rise in temperature. In the TGA test, as the temperature rises, the weight loss rate increases. The middle temperature range is where combustion mainly occurs, and in the high temperature zone, the final stage of organic matter oxidation and decomposition occurs. The remaining heavy organic matter is difficult to oxidize and decompose, so the required activation energy for this reaction is the greatest. This is consistent with the conclusion reached by Wang Jiao and Zhao Shuai and others, that, as the conversion rate increases, the activation energy for the oxidation of shale oil gradually increases [19,23].

(b) For the medium temperature range as the main combustion temperature range, the activation energy and pre-exponential factor generally decrease and tend to stabilize as the particle size decreases, indicating that the energy required to initiate the HTO reaction decreases with the reduction in particle size. However, in the low temperature range, the activation energy and pre-exponential factor increase gradually as the particle size decreases, while in the high temperature range, they decrease first and then increase as the particle size decreases.

It should be noted that the current study employs the single heating rate method to determine the activation energy and pre-exponential factor. This method has the advantages of simplicity, rapidity, and low cost, but it can only provide the average value of the activation energy in each temperature range and cannot reflect the variation in the activation energy with the conversion rate during complex reaction processes. The multiple heating rate method, however, can obtain the reaction kinetic parameters at different conversion rates, and the results are more reliable and accurate [19,23]. In future work, we will verify and expand on the findings of this study through multiple heating rate experiments.

(2) Enthalpy of reaction

When using DSC curves to determine reaction enthalpy, the entire DSC curve is first corrected with a linear baseline to facilitate the precise calculation of the integral area of each thermal effect peak. At the same time, the sample mass is normalized. Therefore, the enthalpy value of a certain oxidative pyrolysis stage can be obtained by integrating the heat flow rate of this stage. The formula is as follows:

$$∆H={\int }_{t1}^{t2}{\displaystyle \frac{dH}{dt}}dt/m_o$$

(7)

where ΔH is the reaction enthalpy value, J/g; t₁ and t₂ are the start and end times of heat release during the reaction, s; dH/dt is the reaction heat flow rate, mW; and m_o is the mass of oil shale sample, mg.

The enthalpy values of organic matter oxidative pyrolysis at different temperature ranges with different particle sizes are shown in Figure 12. It can be seen that the enthalpy values in the low temperature range are 0.605–1.550 kJ/g (corresponding to the LTO stage), increasing with the decrease in particle size. The enthalpy values in the medium temperature range are 1.311–2.812 kJ/g (the main heat release stage in HTO), and those in the high temperature range are 0.409–0.557 kJ/g (the residual heat release stage in HTO), both showing a trend of first decreasing and then increasing with the reduction in particle size. The total enthalpy values range from 2.575 to 4.045 kJ/g, also showing a similar trend. Additionally, it is found that the enthalpy values at different particle sizes have a good linear positive correlation with their total weight loss rate, indicating that the enthalpy values are not only affected by particle size but, more importantly, are determined by their weight loss rate or organic matter content.

4. Conclusions

(1) Oil shale has strong heterogeneity in terms of porosity, permeability, TOC, and inorganic mineral composition. Only along the bedding does it display permeability, mainly in the range of 0.00022–0.0028 md, which has a good exponential relationship with porosity. Local micro-fractures also tend to develop along the bedding, reaching 3.40–5.05 md. The TOC along the bedding is 16 wt%, while it is 9.70–11.02 wt% perpendicular to the bedding. The inorganic minerals consist of 47.38 wt% clay minerals, 15.3 wt% pyrite, and 20.18 wt% carbonate minerals.

(2) The oil shale’s derivative of the weight loss curve shows a single peak, and the heat flow curve shows a double peak, successively experiencing dehydration and oxidative pyrolysis of organic matter and inorganic matter. The oxidative pyrolysis stage of organic matter can be precisely divided into three temperature ranges. The medium temperature range is where combustion weight loss and heat release mainly occur, ranging from the ignition temperature of 301.1–328.1 °C to the burnout temperature of 424.7–456.7 °C, with the highest pyrolysis rate at 365.1–379.5 °C. The results of the weight loss observation experiments and the TGA-DSC tests are in good agreement.

(3) The oxidative pyrolysis reaction kinetic parameters of the oil shale are significantly affected by factors such as particle size, organic matter content, and pyrolysis temperature. The activation energy of the organic matter’s oxidative pyrolysis is 46.92–248.11 kJ/mol, the pre-exponential factor is 3.15 × 10²–6.27 × 10¹¹ 1/s, and the enthalpy value is 2.575–4.045 kJ/g. The combustion index and reaction enthalpy of oil shale with different particle sizes are mainly determined by the organic matter content. The smaller the particle size, the more complete the oxygen addition reaction is at low temperatures, resulting in a varying pattern of the activation energy and pre-exponential factor as temperature increases, changing from a continuous increase to a decrease and then increasing again as particle size decreases. A small particle size, high organic matter content, and high pressure are more conducive to initiating the oxidative pyrolysis reaction to achieve in situ conversion of organic matter.