A Review of the Heterogeneity of Organic-Matter-Hosted Pores in Shale Reservoirs

Abstract

Organic-matter-hosted pores are fundamental pore spaces in shale reservoirs, which largely control the expulsion and storage of oil and gas. However, the strong heterogeneity of organic pores greatly increases the complexity of the pore network in shale reservoirs, which make shale reservoir evaluation challenging. The heterogeneity of organic pores in shale reservoirs has been one of the hottest topics of recent years. In this review, the heterogeneity of organic pores in shale reservoirs and their controlling factors are systematically summarized. The formation and evolution of organic pores in shale reservoirs are directly linked to hydrocarbon generation and expulsion, and the heterogeneity of organic pores is a result of various geological and geochemical factors. The development and heterogeneity of organic pores are basically controlled by the differences in kerogen types and maceral compositions of shale deposits, which are mainly attributed to the differences in hydrocarbon generation capacity of different maceral compositions. Thermal maturation of organic matter is responsible for the formation and evolutionary mechanisms of organic pores and their heterogeneities. With increasing maturity, the increasing trend of pore volume and porosity diminishes. The reduction in macropore volume first appears, and the collapse of macropores could lead to an increase in micropore and mesopore volumes. An important turning point for the thermal evolution of OM is 3.5% R_o. At an R_o greater than 3.5%, the chemical structure of OM is transformed from amorphous carbon to crystalline graphite, and the hydrocarbon generation capacity of the OM has been exhausted, thus, resulting in the destruction of OM-hosted pores. The TOC content and mineral compositions of shale reservoirs affect the development and preservation of organic pores, but enhanced TOC content and brittle minerals may work against the development and preservation of organic porosity. Geological factors, e.g., compaction, diagenesis, pore fluid pressure and tectonic deformation, can also affect the organic pore structure of shale reservoirs to some extent, and their differences can enhance the heterogeneity of organic pores. On the basis of the above-mentioned understandings, this review also puts forward and discusses the problems existing in the current study of organic pore and its heterogeneity of shale reservoirs, and points out further research directions.

1. Introduction

A pore in a shale reservoir is a reservoir space for shale oil and gas, and the porosity and pore structure directly affect the quantity and occurrence of retained oil and gas in shale reservoirs. The pores of shale reservoirs include inorganic and organic pores at various scales [1], mainly nanoscale, which are different from the pores of tight oil and gas reservoirs. Pores of shale reservoirs, especially the organic matter (OM)-hosted nanopores, have been extensively studied by previous authors [2,3,4]. In recent years, many methods and techniques have been applied to pore description and characterization in shale reservoirs, including direct observation methods (e.g., scanning electron microscopy (SEM) and nano-CT) [5,6,7], fluid-injection methods (e.g., mercury intrusion porosimetry, N₂ and CO₂ adsorption, He pycnometry) [8,9,10], and physical methods (e.g., nuclear magnetic resonance (NMR), small-angle neutron scattering) [11,12,13]. Combined with the geochemical characteristics of shale (e.g., organic and inorganic compositions, OM maturity, and diagenesis), the results obtained from the above-mentioned methods can be used for the interpretation of the formation and evolution of pores as well as their controlling factors, thus, strongly supporting the evaluation, exploration and development of shale oil and gas [14,15,16,17].

Organic pores play an important role in the pore network system of shale reservoirs [1,5,18]. As for highly-mature and over-mature shale gas reservoirs, organic pores are always the dominant pore type [19,20,21,22]. The characteristics of organic pores, including their development degree, become one of the most important factors affecting shale reservoir properties [18,23,24,25,26].

Strong heterogeneity is another characteristic of pores of shale reservoirs, which leads to a high complexity of the pore system and increases the difficulty of shale reservoir evaluation and prediction. The pore heterogeneity of shale reservoirs is not only related to inorganic pores, but also linked to organic pores. SEM observations have revealed that the heterogeneity of organic pores was mainly manifested in the differences in pore quantity, morphology and distribution. The heterogeneity of organic pores can be found not only in different OM particles (e.g., different types of macerals), but also in different parts of the same OM macerals. Even in the same maceral, there may be obvious differences in the development degree of pores. These pore characteristics have been widely reported in shale reservoirs of different regions and ages [1,20,27,28,29].

Organic pores of shale reservoirs can be divided into primary pores and secondary pores [30,31,32]. Primary organic pores are mainly found in kerogen, and are usually observed in immature and low-maturity shale [33,34,35]. However, secondary organic pores are formed during thermal degradation and are closely related to hydrocarbon generation and expulsion of shale [36,37,38,39], which are the dominant types of shale organic pores [32,40,41,42,43]. In this study, OM-hosted pore generally refers to the OM secondary pore, if it is not specifically indicated.

The heterogeneity of organic pores is an important factor affecting the heterogeneity of pores in shale reservoirs. Therefore, the heterogeneity of organic pores in shale reservoirs may be influenced by the factors related to the formation, evolution and preservation of organic pores, e.g., the original textural characteristics of shale, the buried depth of shale, the organic porosity developed within solid bitumen and the mineral compositions. This study reviews the research progress and existing problems of the major influencing factors of the heterogeneity of organic pores in shale reservoirs. The major purpose of this paper is to enhance the understanding of the physical properties of shale reservoirs and shed light on opportunities for further research.

2. Pore Characteristics of Different Types of Kerogen and Macerals

OM in shale is basically composed of kerogen and bitumen. The difference between kerogen and bitumen in organic geochemistry is whether they are soluble within organic solvents [44]. As for organic petrology, bitumen is generally regarded as secondary maceral, because it is the secondary OM formed via kerogen transformation [45]. When the vitrinite reflectance (R_o) value is greater than 0.80%, bitumen becomes the dominant type of OM in shale [46]. With an increasing R_o value, the insoluble part of bitumen gradually increases [47], which could be cracked into pyrobitumen. Mastalerz et al. (2018) defined the boundary of bitumen and pyrobitumen at the bitumen reflectance (BR_o) of 1.5% [44]. Many authors have suggested that bitumen and pyrobitumen could be collectively referred to as solid bitumen (SB) [41,48,49,50]. The kerogen of high-quality shale deposits (kerogen type I and II_a) is mainly composed of amorphous OM. During the highly-mature to over-mature stage, some of amorphous OMs display granular structure, known as micrinite. In this study, they are collectively defined as amorphous OM. In addition, some identifiable macerals can be found in shale, such as vitrinite or vitrinite-like maceral (VLM), alginate, exinite, inertinite, and organic zooclasts (e.g., graptolite and chitinozoan). Due to different sources and properties of these OM particles, different types of OM are the basic factors controlling the development of organic pores and their heterogeneities in shale reservoirs [27,38,51,52,53].

2.1. Pore Characteristics of Different Types of Kerogen

The kerogen of shale deposits can generally be divided into three or four types, i.e., type I, type II and type III, of which type II can be further subdivided into II_a (or II₁) and II_b (or II₂) [54]. The kerogen type of shale determines oil and gas generation potentials, thus, affecting the development of organic pores [38,52,55]. Chalmers and Bustin (2007) found that organic pores were more likely to be developed in type I and II kerogens, while they were poorly developed in type III kerogen [56]. Chen et al. (2015) suggested that kerogen type exerted stronger influences on organic pores than maturity, within a certain range of R_o values [52]. The pore-forming ability of type III kerogen was found to be very poor during thermal evolution, which is related to its low hydrocarbon generation ability and higher potential of gaseous hydrocarbon generation [57,58,59]. Yang et al. (2019) believed that type I kerogen had chemically unstable structures, whereas type III kerogen had chemically relatively inert structures; thus, organic pores could easily develop in type I kerogen while a large number of organic pores would be difficult to form within type III kerogen [60]. However, some authors have different understandings on the formation mechanism of pores in kerogen. Löhr et al. (2015) found that some organic pores in gas shale with an R_o value greater than 1.50% may be inherited from primary pores, which were not formed via thermal degradation [30]. Modica et al. (2012) suggested that the previously-existing hydrocarbons filling the pores of kerogen in shale would be converted into oil and then expelled during thermal maturation, and the remaining spaces would be organic pores within kerogen [61]. Although organic pores are easily developed in type I and II kerogens relative to type III kerogen, these pores are difficult to effectively identify due to the swelling effect during the oil generation stage. With a further increase in thermal evolution, the differences of pore development within different types of kerogen will be obvious.

The effect of kerogen type on organic porosity can obviously be displayed in shale deposits with different sedimentary facies. Organic pores are generally well developed in marine highly-mature to over-mature shale [36,62,63], while they are poorly developed in the continental low-mature to medium-mature shale and marine–continental transitional medium-mature to highly-mature shale [64,65,66,67]. These characteristics may not only be related to maturity. More importantly, the kerogen of marine shale is generally classified as type I-II_a [60,68,69], while the kerogen of continental shale and marine–continental transitional shale is basically classified as type III [70,71,72,73].

Optical observations revealed that amorphous OM in shale generally occurs as fine-grained and dispersed forms at several microns to several nanometers, which accounts for most OM components in highly-mature to over-mature shale. SEM observations revealed that amorphous OM can present in various forms, such as matrix, granular, and even spherical [50,74], and the pore characteristics can be greatly varied in amorphous OM particles. Overall, the pore development degree of amorphous OM in marine shale with type I-II kerogen is obviously higher than that of marine–continental transitional shale with type III kerogen (Figure 1a–e). Strong heterogeneity of pores usually occurs in amorphous OM, which can be observed not only in different OM particles (Figure 1f,g) but also in different parts of the same OM particles (Figure 1h). For example, spherical OM particles display different pore characteristics; the rim of the OM particle is porous, while its core seems to have no SEM-visible pores (Figure 1i).

2.2. Pore Characteristics of Solid Bitumen

Compared with kerogen, solid bitumen (SB) contains a large volume of organic pores, mostly sponge-like and bubble-like [20,27,44,74,78] (Figure 2a–c). In addition, these pores display complex structures. With the increase in size and connectivity of individual pores, some pores (e.g., sponge-like pores) may be replaced by irregularly shaped organic pores. These complex pores are considered to be formed via the interconnection of several previously-formed pores during thermal evolution [5,79] (Figure 2c). Milliken et al. (2012) suggested that organic pores of shale mainly occurred in the SB, rather than in kerogen [80]. Suárez-Ruiz et al. (2016) also found that nanopore volumes displayed a good positive correlation with the TOC content of SB-rich shale, while no correlation was exhibited in the kerogen-rich shale [81]. Loucks et al. (2009) revealed that an organic pore network occurred in shale reservoirs, and suggested that organic pores were easily developed in SB, thus, improving the quality of shale reservoirs [1]. Curtis et al. (2012) found SB networks in the Barnett, Haynesville and Horn River shale reservoirs [27], which laid a foundation for the formation of organic pore networks.

Compared with the North American marine shale, the South China marine shale displays relatively high degrees of thermal maturation, and the majority of shale reservoirs are at the highly-mature to over-mature stage. In recent years, the Wufeng–Longmaxi Formation shale reservoirs in the Sichuan Basin have been widely studied, and porous SB can usually be observed (Figure 2a–c), leading to the formation of an organic pore network in shale [31,46,82,83]. Xie et al. (2021) found that the amount of SB in the Wufeng–Longmaxi Formation shale in the Sichuan Basin can indirectly indicate the enrichment degree of shale gas [84].

Figure 2. SEM photographs showing pore characteristics of solid bitumen. (Explanation: SB = solid bitumen.) (a,b) Nanopores are widely developed in the SB of the Longmaxi Formation shale from Well QJ-1 in the Northern Guizhou area (unpublished data). (c) Pores in the SB of Longmaxi Formation shale from the Sichuan Basin show complex morphologies [50]. (d) Pores are poorly developed in the SB of Lower Cambrian shale from Well HY-1 (unpublished data). (e) Pores are poorly developed in the SB of Niutitang Formation shale from Well CY-2 in the southeastern Chongqing area [85]. (f) No SEM-visible pores can be observed in the SB of New Albany shale [86].

Figure 2. SEM photographs showing pore characteristics of solid bitumen. (Explanation: SB = solid bitumen.) (a,b) Nanopores are widely developed in the SB of the Longmaxi Formation shale from Well QJ-1 in the Northern Guizhou area (unpublished data). (c) Pores in the SB of Longmaxi Formation shale from the Sichuan Basin show complex morphologies [50]. (d) Pores are poorly developed in the SB of Lower Cambrian shale from Well HY-1 (unpublished data). (e) Pores are poorly developed in the SB of Niutitang Formation shale from Well CY-2 in the southeastern Chongqing area [85]. (f) No SEM-visible pores can be observed in the SB of New Albany shale [86].

However, pores are poorly developed in some SB particles [85,87,88,89,90] (Figure 2d–f), indicating that organic-matter-hosted pores may not only be controlled by maturity. Even at the highly-mature to over-mature stage, pores could not be observed in some SB particles. Wei et al. (2015) found that organic pores were poorly developed in smaller volumes of SB of the Wufeng–Longmaxi Formation shale in the Sichuan Basin [90]. Wang et al. (2020) found that pores were rarely developed in some SB particles of the Niutitang Formation shale in the southeastern Chongqing area [85]. Liu et al. (2019) found that micro-cracks were generally developed on the margins or cores of non-porous SB in the Lower Cambrian shale in the upper Yangtze area. They believed that the lack of pores in the SB may have resulted from the escape of trapped gas via micro-cracks, leading to the collapse of the organic pores [91]. However, micro-cracks had not developed in the margins or cores of each non-porous SB (Figure 2d,e), indicating that the absence of OM-hosted pores is sometimes not related to micro-cracks.

On the other hand, the occurrence of SB in shale may reduce the porosity of kerogen. Löhr et al. (2015) suggested that SB filling in primary pores of kerogen could occupy the pore spaces of kerogen [30]. Furmann et al. (2016) also found that liquid petroleum generated simultaneously with bitumen could block organic pores, and suggested that the number of organic pores formed at the “oil window” stage should be much larger than the number of SEM-visible organic pores [92]. Such an explanation could account for the truth that organic pores were more developed at the “gas window” stage than at the “oil window” stage [93].

In summary, although SB is usually regarded as the most porous maceral, pore characteristics of SB particles in different shale formations are still different.

2.3. Pore Characteristics of Different Macerals

Compared with SB, other macerals (e.g., alginite, inertinite, graptolite and vitrinite-like maceral) of marine shale display more complex morphologies and origins of their pores [5,30,94,95,96]. Hu et al. (2020) found that the residual alginate particles in the Longmaxi Formation shale usually had two occurrence forms. Type I alginate displays lenticular or nearly circular shapes, and type II alginate exhibits granular or aggregated shapes. Elliptical or elongated organic pores are developed in the former, which are distributed along the long axis of minerals, probably reflecting the original biological structures (Figure 3a). Secondary pores with strong heterogeneity are developed in the latter (Figure 3b) [20]. Jiao et al. (2019) also suggested that the pore morphology of the alginite of the Wufeng–Longmaxi Formation shale was complex, which may be related to the inherited primary pores [97]. Liu et al. (2017) found that organic pores had rarely developed in the alginite of New Albany immature shale in the Illinois basin (Figure 3c). However, the shale had evolved in the “oil window”, and secondary organic pores had developed in the alginite (Figure 3d). The researchers believed that the formation of these secondary organic pores seemed to be related to the catalysis of surrounding minerals [38].

The pore-forming ability of inertinite is very poor during thermal evolution [86,89]. Even at very high thermal maturity, pores could not be found in inertinite (Figure 3e) [38]. Li et al. (2020) performed a comparison study of pore structure parameters of the isolated kerogen samples from Lower Silurian and Upper Permian shale deposits in the Yangtze area, and found that the volume and specific surface area of organic pores of the former were six, and three and a half times higher, than those of the latter, respectively. Such a difference may be induced from the occurrence of vitrinite and inertinite with very stable chemical structures in the Upper Permian shale, thus, resulting in a weak ability of organic pore formation [98]. Although secondary pores never developed in the inertinite, primary pores inherited from plant tissues could be preserved [33,38,86,99] (Figure 3f). Therefore, in addition to hydrocarbon generation potential, the properties and structure of the original OM are the additional factors controlling the development of organic pores in shale reservoirs [100].

The OM of the graptolite in shale deposits also shows the heterogeneity of pores. It is generally believed that SEM-visible pores rarely developed in graptolite [50,57,101,102]. However, few organic pores can be found in graptolite. Teng et al. (2022) found that organic pores failed to be observed in non-granular graptolite of the Longmaxi Formation shale, but nanoscale organic pores could be found in the granular graptolite (Figure 3g,h) [31]. Ma et al. (2016) found that pore characteristics of different tissues of graptolite were very different. There were basically no pores in the graptolite periderm perpendicular to bedding, but nanoscale pores were observed in the swirling cortical fibers at the edge of the periderm wall of the graptolite (Figure 3i,j) [103]. Tenger et al. (2017) conducted a comparison study of pore characteristics of different types of macerals in the Wufeng–Longmaxi Formation shale of Well JY1 in the Sichuan Basin. They found that a small volume of micropores could be developed in the graptolite, but the development degree was relatively poor, lower than in SB and amorphous OM [104]. Graptolite is a type of hydrogen-poor maceral which has weak hydrocarbon generation capacity. Organic pores may not originate from graptolite itself, but may be inherited from primary pores. Due to the occurrence of lipids in the graptolite, graptolite has certain hydrocarbon potentials, and organic pores could be formed during hydrocarbon generation [101,105].

The formation of pores in vitrinite during thermal evolution is dependent on the type and maturity of the vitrinite. In the low-mature to medium-mature shale, pores, in general, rarely develop in vitrinite [106,107] (Figure 3k), but micro-cracks can be observed [108] (Figure 3l). In the highly-mature to over-mature shale, smaller pores can be observed in vitrinite, which show uneven distribution (Figure 3m). Xu et al. (2022) found that organic pores had not developed in the structured vitrinite of the Triassic shale in the Sichuan Basin, but a small amount of round-elliptic organic pores could be observed on the surface of unstructured vitrinite [48]. Similar to vitrinite, vitrinite-like maceral in marine shale displays no pore characteristics (Figure 3n,o). The poor pore-forming ability of vitrinite and vitrinite-like maceral during thermal evolution should be related to its low hydrocarbon generation potential and gas-prone nature.

In summary, the difference of maceral types in shale is the basic factor controlling the heterogeneity of organic pores, and the difference of pore characteristics in different macerals may be the result of the interaction of primary pores and secondary pores. However, the identification of different types of macerals is still challenging under SEM conditions [74,109,110,111]. In particular, the identification of macerals will be difficult in the highly-mature to over-mature shale. Moreover, the heterogeneity of organic pores caused by the difference of macerals will be more obvious.

Figure 3. SEM photographs showing pores characteristics of different types of macerals. (Explanation: AOM = amorphous organic matter; SB = solid bitumen; Q = quartz; VLP = vitrinite-like particle.) (a) The pores in the alginite of Longmaxi Formation shale in the Sichuan Basin show an orientated distribution [20]. (b) Secondary pores were developed in the alginite of Longmaxi Formation shale in the Sichuan Basin, and displayed the characteristics of heterogeneity [20]. (c) No SEM-visible pores in the alginite of New Albany immature shale [38]. (d) A large number of secondary pores were developed in the alginite of New Albany shale at the “oil window” stage [38]. (e) Pores were lacking in the inertinite of New Albany shale at the highly-mature stage [38]. (f) Honeycomb primary pores were developed in the inertinite of New Albany shale in Daviess County, which were filled with authigenic quartz [86]. (g) Graptolite in Longmaxi Formation shale of the Sichuan Basin [31]. (h) Close-up of the area delimited by the dashed red rectangle in (g) showing the scattered pores [31]. (i) Swirling cortical fibers in the graptolite of Longmaxi Formation shale [103]. (j) Nanoscale pores were developed between the laminated cortical fibers in Longmaxi Formation shale [103]. (k) Pores were widely developed in the SB, while they were rarely developed in the vitrinite of Lower Jurassic Ziliujing Formation shale in the northeastern Sichuan Basin [107]. (l) Micro-cracks were developed in the vitrinite of Middle–Upper Permian shale in the Lower Yangtze region [108]. (m) Nanopores were developed in the vitrinite of Lower Permian shale in the Ordos Basin [75]. (n) Vitrinite-like maceral in Longmaxi Formation shale of the Sichuan Basin [31]. (o) Close-up of the area delimited by the dashed red rectangle in (n) showing no pores in the vitrinite-like maceral [31]. Part (h) and (o) are the close-up views of the red dashed areas in figures (g) and (n), respectively.

Figure 3. SEM photographs showing pores characteristics of different types of macerals. (Explanation: AOM = amorphous organic matter; SB = solid bitumen; Q = quartz; VLP = vitrinite-like particle.) (a) The pores in the alginite of Longmaxi Formation shale in the Sichuan Basin show an orientated distribution [20]. (b) Secondary pores were developed in the alginite of Longmaxi Formation shale in the Sichuan Basin, and displayed the characteristics of heterogeneity [20]. (c) No SEM-visible pores in the alginite of New Albany immature shale [38]. (d) A large number of secondary pores were developed in the alginite of New Albany shale at the “oil window” stage [38]. (e) Pores were lacking in the inertinite of New Albany shale at the highly-mature stage [38]. (f) Honeycomb primary pores were developed in the inertinite of New Albany shale in Daviess County, which were filled with authigenic quartz [86]. (g) Graptolite in Longmaxi Formation shale of the Sichuan Basin [31]. (h) Close-up of the area delimited by the dashed red rectangle in (g) showing the scattered pores [31]. (i) Swirling cortical fibers in the graptolite of Longmaxi Formation shale [103]. (j) Nanoscale pores were developed between the laminated cortical fibers in Longmaxi Formation shale [103]. (k) Pores were widely developed in the SB, while they were rarely developed in the vitrinite of Lower Jurassic Ziliujing Formation shale in the northeastern Sichuan Basin [107]. (l) Micro-cracks were developed in the vitrinite of Middle–Upper Permian shale in the Lower Yangtze region [108]. (m) Nanopores were developed in the vitrinite of Lower Permian shale in the Ordos Basin [75]. (n) Vitrinite-like maceral in Longmaxi Formation shale of the Sichuan Basin [31]. (o) Close-up of the area delimited by the dashed red rectangle in (n) showing no pores in the vitrinite-like maceral [31]. Part (h) and (o) are the close-up views of the red dashed areas in figures (g) and (n), respectively.

3. Effect of Thermal Evolution on Organic Pore Heterogeneity

In recent years, a large number of studies have been carried out on the correlations between porosity/organic porosity and maturity of shale reservoirs. It is widely accepted that the development and evolution of pores (especially organic pores) were influenced by the degree of thermal evolution and were closely related to the hydrocarbon generation process [1,80,112,113]. Jarvie et al. (2007) found that the porosity of the Barnett shale at the “oil window” stage was positively correlated with maturity and the amount of hydrocarbon generation [36]. Cander (2012) also showed that organic porosity of shale increased with the increase in burial depth at the “oil window”–wet gas stage [114], and the formation of secondary organic pores mainly accounted for the increase in porosity [115]. Wang et al. (2020) conducted a comparative study of the Triassic Yanchang Formation, Lower Silurian Longmaxi Formation and Lower Cambrian Niutitang Formation shale reservoirs, and showed that the development of organic pores was mainly influenced by maturity. As for shale with low maturity (R_o < 1.5%), organic pores were poorly developed. For the shale with high-maturity to over-maturity (3.0% < R_o < 4.0%), organic pores would be destroyed, thus, leading to decreased porosity. The stable development of organic pores can be only guaranteed in shale reservoirs at suitable maturity (1.5% < R_o < 3.0%) [116]. Chen and Xiao (2014) proposed three stages for the evolution of organic pores, i.e., pore formation stage (0.6% < R_o < 2.0%), pore development stage (2.0% < R_o < 3.5%) and pore destruction stage (R_o > 3.5%) [117]. Ding et al. (2019) proposed four stages for the evolution of organic pores, i.e., no pore formation stage (R_o < 0.7%), pore formation stage (0.70% ≤ R_o < 1.3%), pore development stage (1.3% ≤ R_o < 3.5%) and pore destruction stage (R_o ≥ 3.5%) [118]. Wang et al. (2020) established a conceptual model for organic porosity evolution of the shale reservoirs, and suggested that organic porosity first increased, and then decreased, during thermal evolution. The turning point corresponding to R_o value was about 3.0% [116]. Although different authors had different opinions on the stages of organic pore evolution, the trend was very similar. The organic pore of the shale was formed at the beginning of hydrocarbon generation, and ceased at the end of hydrocarbon generation. During the stage of the “oil window”, a large amount of liquid hydrocarbons were generated from the kerogen, which could be expelled from the kerogen due to hydrocarbon-generation pressurization, thus, resulting in the formation of organic pores within the kerogen [118]. With the increasing thermal maturity, the retained liquid hydrocarbons were cracked into solid bitumen and hydrocarbon gases, and the quantity of organic pores rapidly increased [78]. With the further increase in maturity, the organic porosity of shale reservoirs would decrease, and the chemical structure of the OM was destroyed to some extent. The decrease in organic porosity of shale reservoirs at the over-mature stage could be explained by the following two aspects: first, that OM begins to be carbonized, resulting in the destruction of pore structure; second, that strong compaction leads to the collapse of the organic pores [58,119,120,121].

It is well accepted that the chemical structure of kerogen changes continuously during thermal evolution [122,123,124,125] (

Table 1. Relationships between thermal evolution, chemical composition, chemical structure and pore evolution of OM in source rocks (modified from [112,118,122,127,128,129]).

Ro/%	Thermal Evolution Stage	Hydrocabon Type	Chemical Composition and Structure of OM	Evolution of OM-Hosted Pores
<0.5	Immature	Biogenic gas	Drop of oxygen-containing functional groups, short aliphatic side chains, and low-carbon-number aromatic rings from kerogen.	Non-porous
0.5–1.0	Mature	Liquid oil	Long aliphatic side chains and cross-linked chains were broken from kerogen. With the increasing aromaticity, the amount of small-scale aromatic clusters decreased, but the average interlayer spacing remained almost unchanged.	Pore formation
1.0–1.35		Condensate and wet gas	With the increasing aromaticity, aromatic compounds predominated over aliphatic compounds in the kerogen.
1.35–2.0	Highly-mature to over-mature			Pore development
		Thermogenic dry gas	The aromatization of aliphatic compounds yielded the naphthalenes and 2 × 2 aromatic rings, and the average interlayer spacing decreased.
2.0–3.5			Non-aromatic functional groups were almost diminished. The condensation of small-scale aromatic rings resulted in the increase in 3 × 3 aromatic rings. Moreover, the average interlayer spacing of aromatic rings remained constant.
>3.5			The condensation of 3 × 3 aromatic rings led to the increased sizes of aromatic rings, characterized by the enhanced proportion of aromatic rings larger than 3 × 3. Moreover, the aromatic rings displayed a better arrangement. The average interlayer spacing decreased rapidly with increasing Ro values.	Pore destruction

). With the increase in thermal maturity, the kerogen tends to have a stable and orderly chemical structure (i.e., graphitization), which is characterized by the gradual dropping of aliphatic chains and the increase in aromaticity [126,127,128,129]. Although the aromaticity of the OM has been suggested to increase with increasing R_o values, the detailed reasons still remain controversial [127,129,130,131]. Some authors have suggested that the condensation of aromatic rings resulted in the increase in aromaticity [122,129,131]. However, Hou et al. (2020) pointed out that the increase in aromaticity might be induced from the aromatization or dehydrogenation of hydroaromatic structures [127]. Hou et al. (2019) found that an important turning point of the OM evolution occurred at R_o = 3.5%, which could be interpreted as the critical point of the OM chemical structure from amorphous carbon to crystalline graphite [126]. With the increasing degree of graphitization, the hydrocarbon generation capacity of shale decreased rapidly [132,133]. Moreover, the graphitization would lead to chemical structure rearrangement of the OM [128]. In addition, the development of organic pores is closely related to the chemical composition and spatial arrangement of the OM [3,128,134]. Although oxygen-containing functional groups played a positive role in the development of micropores [135], the aliphatic compounds played an overwhelming role in the formation of micropores [136,137]. Liu et al. (2018) found that the evolution of organic micropores was controlled by the variation of OM chemical compositions at different thermal maturities. At the R_o ranging from 0.5% to 1.4%, micropore volume was mainly controlled by the aliphatic parts of the chemical structure, which decreased with a decrease in aliphatic parts. At the R_o ranging from 1.4% to 4.0%, micropore volume was controlled by the number of aromatic rings, which increased with the increasing content of aromatic carbon [137]. In addition, Zhang et al. (2022) proposed that the development of mesopores was related to chemical compositions of the OM and their spatial arrangement. Moreover, the variation of chemical compositions could influence the heterogeneity of the pore surfaces and structures via the alteration of surface chemical properties and pore geometry, while the variation of spatial arrangement could only alter the distribution of pore spaces, thus, influencing the heterogeneity of pore structures [3]. As for the aromatic compounds, it was found that with the increase in thermal maturity, the proportion of aromatic fringes below 2 × 2 decreased, while that of aromatic fringes above 3 × 3 increased; such a change was not conducive to the development of micropores. The separation and tortuosity of aromatic fringes were closely related to the Langmuir volume and pressure parameters [138]. Overall, the thermal evolution of OM can alter the chemical compositions (e.g., aliphatic compounds, oxygen containing groups and aromatic compounds) and structures (e.g., length, spacing and orientation of aromatic fringes) of the OM, thus, influencing the structure of OM-hosted pores and the adsorption capacity of methane.

The relationships between organic pore structure parameters and maturity are shown in Figure 4, based on previous works [16,20,22,87,98,139,140,141,142,143,144,145,146,147]. The organic micropore volume of shale samples showed an increasing trend at R_o values < 2.5% with increasing maturity, but decreased at R_o value > 2.5%. It is noteworthy that the type of kerogen affects the correlation between maturity and organic micropore volume (Figure 4a), showing that micropores were poorly developed in type III kerogen. The evolution of organic non-microporous volume was different from that of organic microporous volume. The organic mesopore volume reached a maximum value at R_o = 3.2–3.5%, then decreased sharply with further increase in R_o (Figure 4b). However, the turning point of organic macropore volume corresponded to a lower maturity (R_o = 1.5–1.8%) (Figure 4c), probably indicating that thermal evolution exerted a more obvious effect on large-scaled pores. At the highly- and over-mature stages, macropores are probably converted to micropores and mesopores due to the collapse of larger pores, resulting in an increase in micropore and mesopore volumes. In addition to the kerogen type, some outliers may be induced from rigid matrix minerals [148]. For example, high quartz content is conducive to the preservation of macropores (Figure 4c). At R_o < 3.0%, there is a positive correlation between organic porosity and R_o values, but such an increasing trend would be diminished at R_o exceeding 3.0% (Figure 4d).

The fractal dimension of low pressure N₂ adsorption data is widely used to quantitatively characterize the heterogeneity of pores in shale reservoirs. This method has been widely applied in whole-rock shale samples [62,98,149,150]. Some authors have also studied the heterogeneity of organic pores in shale reservoirs [4,141,145,151]. For example, Wang et al. (2021) isolated the kerogen fraction of Wufeng–Longmaxi Formation shale with various maturities (EqVR_o = 2.61–3.23%) and measured pore structure parameters of bulk shale samples and corresponding kerogen fractions. They found that the fractal dimensions D₁ (representing the roughness of pore surface) and D₂ (representing the irregularity of pore structure) of kerogen fractions were significantly lower than those of bulk shale samples, which increased with the increase in the EqVR_o value (Figure 5) [145]. This seems to reveal that the heterogeneity of pore structure in the mineral matrix of shale samples is stronger than that in the OM, but the OM also shows obvious heterogeneity of pores.

In summary, the heterogeneity of organic pores in shale reservoirs will increase with increasing maturity. The study of the evolution of organic pores needs to comprehensively consider the influences of TOC, OM type, diagenesis and other factors.

4. Effect of TOC on Organic Pore Heterogeneity

TOC content is one of major factors influencing the development of organic pores, since it directly reflects the amount of OM in shale. However, the effect of TOC on the heterogeneity of organic pores remains controversial. Curtis et al. (2012) studied four shale formations (Barnett, Haynesville, Horn River and Kimmeridge shale) and found that organic pores were barely developed in the Kimmeridge shale with the highest TOC content (TOC = 51.4%), while organic pores were widely developed in the Barnett and Horn River shale samples with lower TOC content [27]. Milliken et al. (2013) found that the TOC content of Marcellus shale had a positive correlation with porosity when the TOC content was less than 5.6%. However, TOC content was negatively correlated with porosity when the TOC content was greater than 5.6%. They believed that TOC could not always control and predict the development of organic pores, but that TOC is a major factor controlling the development of organic pores within a specified TOC range [29]. Similarly, Wood et al. (2015) also found that shale samples with a porosity greater than 7.5% all had low TOC content (<1.0 wt.%) [152], which was attributed to the strong sealing capacity of shale formations. With increasing TOC content, the gas adsorption capacity of shale samples increased and the effective pore throat narrowed, leading to a reduction in permeability and, thus, the increased sealing capacity of shale formations. During the primary migration, only a small amount of liquid hydrocarbons was expelled and a large amount of liquid hydrocarbons retained in the shale formations as the fillings within the mineral-matrix pores [153]. Although OM-hosted pores are the most important part of the pore system of shale reservoirs, SB sometimes blocks the larger pores, and plays a detrimental role in reservoir quality [152]. Yang et al. (2018) found a positive correlation between pore structure parameters and TOC content of Lower Cambrian Qiongzhusi Formation shale in the southern Sichuan Basin, and a large volume of micropores had developed in shale samples with higher TOC content. Thus, they suggested that TOC content was still a major factor controlling the pore structure of shale [154]. Tenger et al. (2021) found a positive correlation between TOC content and organic porosity of Lower Cambrian Niutitang Formation shale at a TOC content of less than 6%. However, a negative correlation was observed at a TOC content greater than 6%. They explained that high OM content would increase the plasticity of shale, thus, reducing its compaction-resistant ability [83].

We collected the data of TOC, organic porosity and organic pore structure parameters of shale reservoirs from previous works [16,20,22,98,101,139,141,142,143,144,145,146,147,155,156,157], and the associated cross-plots are shown in Figure 6. When the TOC content of the shale samples exceeded 10%, the organic micropore volume decreased obviously (Figure 6a). When the TOC content of the shale samples exceeded 5–7%, the organic mesopore and macropore volumes decreased significantly (Figure 6b,c). Thus, it can be inferred that organic pores occurred in the low-TOC shale samples in the forms of micropore, mesopore and macropore. However, the mesopore and macropore of the high-TOC shale samples (TOC > 5–7%) would be compacted, resulting in the predominance of micropore in the OM. Thus, the difference and heterogeneity of organic pores in shale can be deduced from the TOC variations. In addition, organic porosity also displays a strong heterogeneity even in the same shale formation. For example, the TOC content of Longmaxi Formation shale mainly varied from 1 to 5%, but the organic porosity of shale samples with similar TOC content was extremely varied, ranging from 10 to 32% (Figure 6d). Such a difference may have resulted from the differences of OM types, mineral compositions, maturity and tectonic activity in different shale formations or in the same shale formation from different regions.

5. Effect of Mineral Compositions on Organic Pore Heterogeneity

Inorganic minerals in shale, such as quartz, clays, carbonates and pyrite, may have exerted a certain impact on the development and preservation of organic pores, thus, affecting the heterogeneity of organic pores. In general, brittle minerals, such as quartz, have a constructive effect on the formation and preservation of organic pores, while clay minerals with strong plasticity cannot provide conditions for the stable preservation of organic pores (Figure 7a). Organic pores are often poorly developed in clay-rich shale samples with high maturity. Li (2017) studied the Lower Cambrian and Lower Silurian shale reservoirs in south China, and found that different covariations between porosity and mineral compositions, due to the varied compaction-resistant abilities of different minerals. For example, the content of brittle minerals such as quartz and feldspar was positively correlated with porosity, while the content of clay minerals was negatively correlated with porosity [158]. In addition to the weak plasticity of quartz minerals, the aggregate of amorphous microquartz grains plays an important role in the preservation of organic pores (Figure 7b,c) [83,148,159]. Gao et al. (2022) found that a rigid framework of microquartz aggregates had formed in deeply-buried shale of the Wufeng–Longmaxi Formation, and clay minerals within the SB showed directional distribution, revealing the effective compaction-resistant ability (Figure 7d) [159]. However, the pore preservation capacity of quartz of different origins was also different. The biogenic quartz was more effective in preserving porosity than the quartz of other origins (e.g., quartz formed from the transformation of smectite to illite) [148,160]. Biogenic silica could preserve porosity during the deposition of siliceous oozes to very early diagenesis [148]. Moreover, Liu et al. (2019) found that a stable structure could be formed in the mineral–organic aggregations and pyrite framboids of Lower Cambrian shale, which had strong compaction-resistant ability and played a positive role in the preservation of organic pores. They also believed that preservation conditions had a greater influence on the preservation of organic pores than maturity [91].

Since the formation of organic pores is closely related to the hydrocarbon generation process, some minerals play a catalytic role in hydrocarbon generation to some extent, and have exerted a certain impact on the development of organic pores [161,162,163]. For example, it is widely accepted that clay minerals have a catalytic effect on the hydrocarbon generation of OM [164,165], and organic pores are especially well developed in the clay-mineral–organic aggregations [30,166,167,168,169]. On the other hand, authigenic quartz could be formed via the conversion of montmorillonite to illite, which may play a positive role in the preservation of organic pores [170,171]. Ma et al. (2017) believed that some elements of inorganic minerals would migrate during the thermal evolution of shale and could be introduced into the OM margins, thus, forming stable clay coatings. Pores are poorly developed on the OM margins, while pores are widely developed in the core of OM particles, thus, displaying pore heterogeneity [172].

In fact, the relationship between the content of inorganic minerals and organic porosity is very complex. In particular, such a relationship displays an unclear trend based on robust data under the conditions of the occurrence of other factors controlling organic porosity [16,22,98,139,141,142,143,145,157]. The relationships between the content of major minerals (quartz and clay) of shale samples and pore structure parameters of OM are shown in Figure 8. Complex relationships can be observed to have occurred between quartz content and the organic pore volume of shale samples (Figure 8). When the quartz content is low (<40–50%), organic mesopore volume displays an increasing trend with the increasing content of quartz, whereas organic micropore volume shows a decreasing trend. However, a negative correlation between pore volume and quartz content is displayed when the quartz content exceeds 50% (Figure 8a,b). Such relationships could be caused by the following two aspects. The first is that micropores and mesopores connect to each other and combine into macropores due to the influence of thermal evolution, tectonic deformation and other factors. Moreover, the high quartz content avoids the collapse of larger pores, which is favorable for the preservation of macropores (Figure 4c). Secondly, organic pores may be extruded to a certain extent due to the high quartz content, and the high quartz content may also cause the blocking of primary pores, resulting in a decrease in pore volume. These evidence indicate that moderate quartz content is favorable for the preservation of larger pores. The volume of organic micropores and mesopores display a decreasing trend with the increasing content of clay minerals (Figure 8c,d). Increased content of clay minerals lead to poor preservation conditions of organic pores, probably due to the strong plasticity of clay minerals. In short, the effect of minerals on the development and heterogeneity of organic pores seems to be complex.

The effect of mineral compositions on the development of organic pore and its heterogeneity is obviously exhibited in different shale lithofacies. Liu (2019) found that organic pores were widely developed in siliceous shale of the Upper Permian Dalong Formation in the western Hubei area, while organic pores were poorly developed in mixed shale that contained a higher content of carbonates. A large number of organic pores were blocked in the mixed shale, and organic pores displayed flat shapes due to compaction [173]. Yang et al. (2018) also found that organic pores were developed in siliceous shale of the Lower Cambrian Qiongzhusi Formation in the southern Sichuan Basin, while organic pores were poorly developed in argillaceous/siliceous mixed shale. However, organic pores were rarely developed in carbonate-rich shale or clay-rich shale samples [154].

6. Other Factors Influencing Organic Pore Heterogeneity

In addition to the OM itself, some geological factors could also influence the development of organic pores, leading to the heterogeneity of organic pores. Diagenesis has exerted a certain influence on the quantity, morphology, preservation condition and connectivity of organic pores [174]. Compaction is one of most important factors controlling the variation of the porosity and pore structure of shale reservoirs. Since the property, morphology and habitat of the OM in shale reservoirs are different, compaction could lead to the heterogeneity of organic pores (Figure 7a) [5,175]. However, it must be noted that the effect of compaction on the development of organic pores may be synchronous with the processes of hydrocarbon generation and expulsion, so these factors should be taken into consideration comprehensively. However, the effect of the compaction on the development of organic pores would be more significant when the burial depth of shale exceeds the lower limit of hydrocarbon generation. It is generally believed that the decrease in porosity in shale reservoirs mainly accounts for the collapse of pores due to compaction at the over-mature stage [28,118,176].

The fluid pressure of shale reservoirs plays a critical role in the preservation of organic pores. The hydrocarbon generation of OM is the main mechanism for the development of organic pores in shale. With increasing amounts of generated hydrocarbon, the fluid pressure increases inside organic pores, and plays an important role in supporting organic pores. Thus, the compaction-resistant ability of organic pores is increased, and they could be well developed and preserved. The study of the Lower Silurian Longmaxi Formation shale in the Sichuan Basin found that fluid overpressure played a constructive role in the preservation of organic pores [177,178,179] and favored the preservation of the shapes of circular organic pores [174]. Guo et al. (2020) suggested that deep and ultra-deep shale gas reservoirs in the Sichuan Basin generally had the characteristics of high fluid pressure and high porosity, and reservoir fluid overpressure was the key factor controlling the development of high porosity in high-quality shale under deep burial conditions [180]. Cao et al. (2017) also found that an undercompacted zone in the matured continental shale formation of the Xihu Sag could be formed due to abnormal overpressure, and can protect the residual primary organic pores of shale [181].

In the strongly-deformed zones, the heterogeneity of organic pores may mainly result from tectonic activities. The deformation of shale would occur due to tectonic stresses, which would also influence the pore characteristics of the shale. In general, stable tectonic conditions favor the preservation of organic pores, while tectonic deformation is not conducive to the preservation of pores [166,182]. Ma et al. (2020) studied two successions of the Lower Cambrian shale in the Chongqing area, which had a similar sedimentary environment and thermal maturity. They found that organic pores of shales in the tectonic non-deformed zone were relatively developed, and the specific surface area and pore volume of non-micropores of non-deformed shale samples were higher than those of deformed shale samples [121]. However, the effect of tectonic stress on the micropores of the deformed shale samples was little [183]. Therefore, tectonic activity leads to the variation of organic pore structure parameters. Li (2017) studied Lower Paleozoic marine shale in the tectonic compression zone of the Yangtze area, and found that strong compression could lead to poor development of organic pores in shale, resulting in lower porosity, pore volume, specific surface area and methane adsorption capacity [184]. In addition, tectonic activity may also enhance OM plasticity, thus, resulting in the destruction and even collapse of organic pores during burial [30,33]. Yao et al. (2020) found that a large amount of micro-cracks were formed within the OM of shale due to multiple extension and compression effects in the Linqing Sag, resulting in the heterogeneity of organic pores [185]. On the other hand, micro-cracks may be formed in shale samples in a tectonic deformation zone. These micro-cracks may lead not only to the increase in reservoir spaces to some extent, but also the leakage of hydrocarbon gases, thus, resulting in a reduction in the fluid pressure coefficient and the compression of organic pores. Therefore, tectonism will not only alter the pore structure characteristics of organic pores and enhance its heterogeneity, but also exert an important impact on the gas content of shale.

7. Conclusions

Organic pores of shale display strong heterogeneity, which are not only controlled by the properties of shale itself, including TOC content, maturity, kerogen types, maceral compositions and mineral compositions, but are also influenced by geological factors, such as fluid pressure, compaction and tectonic deformation. These factors have exerted different controls on the heterogeneity of organic pores in shale, leading to differences in the quantity, distribution and pore structure of organic pores. The kerogen type and maceral compositions of shale are the basic factors influencing the development of organic pores and their heterogeneity. With increasing maturity, organic pores have experienced an evolutionary process from pore formation, to pore development, to pore destruction. A critical point for the destruction of organic pores is 3.5% R_o. When the R_o value exceeds 3.5%, the OM is graphitized and its hydrocarbon generation capacity is almost exhausted. A complex relationship between TOC and organic porosity or pore structure parameters of organic pores would be displayed. In the high-TOC shale deposits, the correlation between TOC and organic pore structure parameters may be poor, or have even disappeared due to compaction, thus, showing obvious heterogeneity. The effect of different types of minerals in shale on the development of organic pores is significantly different. Brittle minerals play a constructive role in the preservation of organic pores. However, clay minerals have a certain catalytic effect on hydrocarbon generation and may have a positive effect on the formation of organic pores, but they are generally not conducive to the preservation of organic pores. In the different shale lithofacies, organic pores show obvious strong heterogeneity due to the variations of mineral compositions. Fluid pressure has great significance for the preservation of organic pores. Diagenesis, especially diagenetic compaction and strong compression caused by tectonic deformation, can alter the pore structure of shale to some extent, enhancing the heterogeneity of organic pores.

The heterogeneity of organic pores in shale has resulted from various factors. However, the major factors controlling heterogeneity still remain unclear, and the formation mechanism of organic pores and its influencing factors should be further studied. There are still some unsolved problems. For example, the formation and evolutionary mechanism of pores in some macerals remain unclear, and, moreover, the identification of different types of macerals is still challenging in highly-mature shale. At the over-mature stage, the development and preservation processes of organic pores in shale are highly complex. The reverse relationship between organic porosity/organic pore structure parameters and TOC content and its mechanism, still need to be studied. Minerals have dual effects on the development of organic pores and their heterogeneity, and the associated mechanisms still need to be explored. These scientific issues are also important research directions for shale reservoir evaluation, in the future.