Natural fracturing induced by hydrocarbon-generation-overpressurization represents a typical mode of rock failure, especially in low-permeability rocks such as shales. By synthesizing published data of rock mechanics, geochemistry, and petroleum geology, we delineate the full lifecycle of hydrocarbon-generation-induced fractures (HGIFs) and propose a dynamic evolution model that encompasses four sequential and interrelated evolution stages: (I) embrittlement and mechanical stratification stage, (II) bedding-parallel fracturing stage, (III) bedding-perpendicular fracturing stage, and (IV) attenuation stage. Stage I is associated with the transformation of clay minerals and authigenic cementation during early diagenesis and burial (vitrinite reflectance Ro < 0.5%). These processes fundamentally alter the mechanical properties of the source rocks and transform muddy sediments into brittle rock masses with pronounced mechanical stratification. Stage II is related to fluid volumetric expansion and overpressure driven by normal oil generation (0.5% < Ro < 1.0%). The early-stage mechanical stratification causes fractures to preferentially propagate along bedding planes. Stage III is characterized by continuous overpressurization driven by the generation of light oil and the cracking of oil to gas (1.0% < Ro < 3.5%). The increase in fluid pressure eventually exceeded the vertical confining pressure, allowing fractures to propagate across bedding planes and form a connected network with the existing bedding-parallel fractures. Stage IV is featured by the waning of fracture development at ultra-high thermal maturity (Ro > 3.5%). The diminished hydrocarbon generation capacity and graphitization of shale together inhibit further fracture formation. Therefore, the formation of HGIFs is governed by the interplay of diagenetic and hydrocarbon generation evolution. The four-stage evolutionary model proposed in this study provides key insights into natural fracture assessment in global basins, and offers critical support for sweet spot prediction and hydraulic fracturing design.

Studies on shale reservoir spaces mostly focus on matrix pores,as manifested by a lack of studies on microfractures'role and on the quantitative characterization of multi-scale pore-fracture structures as a whole.This study delves into the shales of varying lithofacies from the Jurassic Lianggaoshan Formation,northeastern Sichuan Basin,using field emission scanning electron microscopy(FE-SEM)aided by a random forest algorithm for image processing,and achieves the automatic identification and quantitative characterization of organic and inorganic pores and fractures in the shales.Furthermore,by combining the large field-of-view(FoV)image stitching for SEM(SEM-Maps),we comprehensively characterize the distribution characteristics of multi-scale pores and fractures in the shales.The results indicate that the random forest model can effectively distinguish organic and inorganic pores and fractures.The SEM-Maps images at the FoV scale of 300 µm×300 µm are representative and can exclude the effects of heterogeneity.In addition,the shales with various lithofacies exhibit significantly different characteristics of pore-fracture structures.In detail,the medium-high organic matter lamellar felsic shales have a total pore-fracture areal porosity of up to 2.66%,significantly higher than that of low-organic-matter laminated felsic shales(1.66%)and low-organic-matter massive silty to fine-grained sandstones(0.99%).In contrast,the shales with various lithofacies exhibit similar distributions of pore and micro-fracture scales,with the pore scales primarily falling in the range of 20 to 1000 nm,and the microfracture scales predominantly between 200 nm and 5 000 nm.Calculations based on single-component pore-fracture development coefficients,indicate that clay minerals are easier to form pores and fractures compared with felsic minerals in the Lianggaoshan Formation.This study not only reveals the differences in the multi-scale pore-fracture structures of shales with various lithofacies in the Lianggaoshan Formation,but also identifies the shale lithofacies with a high pore-fracture areal porosity and favorable pore-fracture types.These findings provide an important scientific basis for the precise assessment of sweet spots and efficient shale oil development.

The shallow marine shale gas reservoirs in the Upper Ordovician Wufeng Formation and the Lower Silu-rian Longmaxi Formation of residual synclines in northern Guizhou have experienced episodic tectonic superim-posed reformation.As a result,reservoir-preserving matching and reservoir formation and gas accumulation mecha-nisms are complex,leading to significant differences in development efficiency among different sub-layers.There-fore,it is of great significance to investigate the differential characteristics of interconnected pore-fracture sys-tems in different shale sub-layers of various syncline units and gas-bearing intervals for better understanding the reservoir-forming mechanisms and optimizing reservoir targets.Taking the shallow marine shales from the Wufeng-Longmaxi formations in northern Guizhou as the research object,various methods were comprehensively applied including dynamic reconstruction of tectonic burial history and pressure evolution,whole rock X-ray diffraction(XRD),field emission scanning electron microscopy(FE-SEM)imaging,image extraction using Image Pro Plus/Image J,high-pressure mercury intrusion,and directional spontaneous imbibition experiments.Besides,using key parameters such as the timing and amplitude of the last tectonic uplift and the average depressurization rate,the study validated the relatively superior tectonic preservation conditions of the gas-bearing unit in the Shixi syncline.On this basis,the preservation of organic matter pores and the genetic mechanism of interconnected pore-fracture systems in different sub-layers of the first member of the Wufeng-Longmaxi formations in the Shixi area were further elucidated.Specifically,in well SX1 of the Longmaxi Formation,organic matter pores were best developed in sub-layer ⑤,followed by sub-layers ③ and ④,and were least developed in sub-layers ①+②.Moreover,the proportion of rigid minerals-organic matter-clay composites in the microscopic field of view was rela-tively larger in sub-layers ⑤ and ③,followed by sub-layers ④ and ①+②.The spontaneous imbibition slopes parallel to beddings were 0.234 and 0.122 for sub-layers ⑤ and ③,respectively,higher than 0.090 and 0.053 for sub-layers ④ and ①+②.The average mercury withdrawal efficiency values of high-pressure mercury intru-sion were 37.047%and 31.912%for sub-layers ⑤ and ③,respectively,higher than 29.48%and 27.55%for sub-layers ④ and ①+②.The findings implied that the composites contributed significantly to organic matter pore preservation and pore-fracture system connectivity.Based on these findings,the study establishes a classification scheme for the genesis of microscopic pore-fracture systems in the shallow marine shale reservoirs from the Wufeng-Longmaxi formations in northern Guizhou and clarifies their reservoir-controlling mechanisms under the coupling constraints of dynamic and static gas accumulation elements.Three genetic models of pore-fracture systems were proposed,including pore preservation and fracture development with rigid minerals-organic matter-clay composites,relative pore preservation and fracture development with rigid minerals-organic matter-clay composites,and pore collapse type with the enrichment of rigid minerals parallel to shale beddings.

The Qiongzhusi Formation represents the second strata in China to achieve commercial shale gas development, following the Longmaxi Formation. However, a limited understanding persists regarding the pore structure of its shales, particularly concerning the distinctions between its shale and silty shale. This study comprehensively analyzed 25 samples of shale and silty shale from the Sichuan Basin. These analyses included total organic carbon content, X-ray diffraction, porosity measurements, field emission scanning electron microscopy, low-temperature CO2 adsorption, low-pressure N2 adsorption, and high-pressure mercury intrusion capillary pressure. Utilizing multifractal theory, the study investigated the pore structure differences between Qiongzhusi Formation shales and silty shales and their implications for shale gas storage and migration within the formation. The results indicated that the pore volume and specific surface area distribution trends for both shale and silty shale were generally consistent, with mesopores being the primary contributors to both pore volume and specific surface area. However, the shales exhibited superior characteristics, with pore volume of 0.0139 cm3/g and specific surface area of 10.305 m2/g, compared to silty shales, which showed pore volume of 0.0126 cm3/g and specific surface area of 8.184 m2/g. Moreover, the pore size distributions of both shale and silty shale displayed multifractal characteristics. In micropores and mesopores, the shales exhibited weaker heterogeneity in pore size distribution than the silty shales, while both possess higher correlation dimension (D2) and Hurst exponent (H). Conversely, regarding macropores, the silty shales demonstrated weaker heterogeneity in pore size distribution compared to the shales, but the former’s D2 and H were higher than those of the latter. This suggests that micropores and mesopores in the shales are clustered and have better connectivity than those in the silty shales. Conversely, macropores in the silty shales are densely clustered and show better connectivity than those in the shales. Based on the analysis of pore structure, organic geochemical, and mineralogical characteristics, the shales demonstrated better gas-bearing potential. Therefore, shale gas exploration and development in the Qiongzhusi Formation should prioritize shale reservoirs.

Gas content and its occurrence characteristics are two key factors affecting the evaluation of shale gas reservoirs' exploration and development potential. Conventional analysis methods, however, are often costly and generate only discrete data. To address this, this study highlights the Qiongzhusi Formation shale in the Sichuan Basin, developing logging-based prediction models targeting adsorbed and free gas content by integrating well logging data with results from 30 methane isothermal adsorption tests and 3 variable-temperature adsorption experiments. These models were used to characterize the formation's gas content. According to the results, the model-predicted total gas content deviates slightly from measurements via preserved core testing, with an average error rate of 11.89%. Logging interpretation indicates the Qiongzhusi Formation shale has a gas content of 1.56 to 18.55 cm(3)/g, dominated by free gas. The ratio of free to adsorbed gas varies from 0.001 to 8.55, averaging 3.78. By layer, gas content ranks as: (5) > (4) > (7) > (3) > (6) > (1) > (2) > (8) . Notably, layer (4) has the highest free-to-adsorbed gas ratio, suggesting exploration should not be limited to layers (5) and (6) -layers (4) , (7), and (3) , especially layer (4) , also have significant potential.

To elucidate the role of effective pore preservation in controlling shale gas enrichment and high yield. This study focuses on fine-grained sedimentary shale of the Qiongzhusi Formation in Sichuan Basin, sampled from different tectonic settings. Rock facies were classified based on total organic carbon (TOC) measurements, X-ray diffraction (XRD), and cast thin section observations. The same lithofacies shale was selected for multi-scale pore characterization and field emission scanning electron microscopy (FE-SEM) to identify pore types and pore structures. By integrating actual drilling and logging data, paleopressure reconstruction via methane inclusion Raman spectroscopy, and burial history analysis of gas reservoirs, the preservation conditions of gas reservoirs are clarified. This study systematically reveals the mechanisms of shale pore preservation at different tectonic settings under the coupled effects of mineral composition, sealing systems, and fluid overpressure. A conceptual model for pore preservation in the Qiongzhusi Formation shales is further proposed. The results show that the deep shelf shales within the rift trough is characterized by high TOC, a rigid mineral skeleton, and overpressure storage. These shales develop an organic-inorganic composite pore network, exhibiting significantly higher pore volume and gas yield compared to shales outside the trough. At the trough margin, a semi-closed system with plastic mineral framework and moderate overpressure supports medium porosity. In contrast, shales outside the rift trough exhibit the weakest pore preservation capacity due to an open system, normal pressure, and plastic mineral framework. This study proposes a ternary synergistic pore preservation model of "rigid skeleton-closed system-hydrocarbon fluid overpressure", which provides critical geological insight for the exploration and development of deep fine-grained sedimentary shale gas. (c) 2026 Publishing services by Elsevier B.V. on behalf of KeAi Communications Co. Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Oil control and oil enrichment associated with strike-slip faults are an important feature of oil and gas accumulation in petroliferous basins in China. Accurate prediction and evaluation of the oil enrichment potential of strike-slip faults have become one of the core objectives of oil and gas exploration. Using statistical analogy, representative fault dissection, comprehensive analysis, and hierarchical evaluation methods, the study took the Halahatang buried hill area of the Tarim Basin as the study area and conducted systematic research. The results indicated that the oil enrichment potential of strike-slip faults was jointly determined by lateral and vertical oil-controlling effects. The lateral controlling effect constrained the degree of oil and gas migration and accumulation along the fault strike (lateral direction), and the vertical controlling effect constrained the characteristics of oil and gas migration and accumulation along the fault dip direction (vertically upward). The core factors influencing the vertical and lateral oil-controlling effects of strike-slip faults included fault properties, the degree of lateral connectivity between fault strike and source rocks, the effectiveness of fault-fractured reservoirs in forming fracture-vuggy traps, the angle between fault strike during the accumulation period and the regional structural ridge and tectonic stress field of the target layer, and the degree of fault-induced damage to caprock. These factors were attributed to the four core characteristics of faults, i.e., oil supply capacity, connectivity and trap-forming capacity, reservoir-forming capacity, and caprock-preservation capacity, as well as their spatiotemporal matching relationships with fault formation, evolution, and distribution. Previous evaluations of oil enrichment potential in strike-slip faults mostly focused only on oil and gas vertical migration and accumulation along the fault dip direction. This paper supplements that approach by incorporating evaluation concepts and methods for lateral migration and accumulation along the fault strike, and improves the evaluation index system for the controlling effects of vertical migration and accumulation. It proposes a "vertical-lateral, dynamic-static, multi-factor coupled evaluation framework for oil enrichment potential in strike-slip faults". Based on this framework, a comprehensive evaluation of the oil enrichment potential of major faults in the Halahatang buried hill area was conducted. It was clarified that the NWW-near EW-trending faults in the central and southern parts of the buried hill area are the most favorable oil and gas enrichment zones, and the conjugate NNE and NNW-trending strike-slip fault zones in the eastern part are relatively favorable zones. The proposed evaluation framework improves the evaluation concept of oil enrichment potential in areas developed with strike-slip faults, and it has demonstration and referential significance for achieving breakthroughs in hydrocarbon exploration in strike-slip fault belts within petroliferous basins.

Total organic carbon (TOC) is a key parameter for geological sweet spot optimization and resource evaluation. However, accurate TOC prediction remains challenging under small-sample conditions because conventional physical and machine learning methods are commonly limited by insufficient labeled data. To address this problem, this study proposes a hybrid deep learning framework that integrates a convolutional autoencoder with a BP neural network (CAE-BPNN). In this framework, the CAE is first used to learn representative feature expressions from abundant unlabeled logging data, and the learned encoder is then transferred to the supervised prediction stage, where labeled samples are used for TOC prediction through the BPNN. The dataset includes 177 measured TOC samples and 1388 unlabeled logging samples from the Fengcheng Formation shale in Well MY1, in the Mahu Sag, and data from Well MY2 are used for independent validation. Model performance is evaluated using five-fold cross-validation with the coefficient of determination (R2) and root mean square error (RMSE) as metrics. For Well MY1, CAE-BPNN achieves the best performance, with R2 = 0.89 and RMSE = 0.061, outperforming CNN (R2 = 0.85, RMSE = 0.075), GBDT (R2 = 0.83, RMSE = 0.076), RF (R2 = 0.81, RMSE = 0.082), and BPNN (R2 = 0.77, RMSE = 0.085). In the independent validation using Well MY2, CAE-BPNN also shows superior predictive performance, with R2 = 0.81 and RMSE = 0.367. These results indicate that unlabeled logging data can effectively enhance feature representation and improve TOC prediction accuracy under limited labeled-sample conditions. The proposed method provides an effective solution for small-sample TOC prediction and offers a reliable basis for movable oil evaluation using the oil saturation index (OSI), as well as a reference for predicting other geological parameters such as S1, S2, and porosity.

The Fengcheng Formation(P1f) in the Mahu Sag of the Junggar Basin is a typical alkaline lake deposit rich in shale oil. The strong heterogeneity poses significant challenges for oil-bearing evaluation, severely restricting exploration decision-making and favourable interval prediction. Therefore, the 283 m shale sedimentary sequence in the P1f section of Well Maye one was studied. Source rocks of the P1f geological and geochemical characteristics, hydrocarbon-generating source materials, and sedimentary paleoenvironment were investigated using extensive organic and inorganic geochemical analyses. An improved three-dimensional(3D) oil content evaluation model based on pyrolysis parameters and oil saturation index (OSI) was used to classify the oil content of different types of organic matter and predict the shale dessert section in P1f. The source rock in P1f is developed in the saline and anoxic environment of sedimentary water stratification. The overall characteristics of the source composition of parent material are that the abundance of bacteria and algae is higher than that of terrestrial plants, and the abundance of bacteria is higher than that of algae, and cyanobacteria account for a certain proportion. The hydrocarbon generation potential index of P1f decreases with the transition of kerogen type from I to III, and the OSI value shows a trend of first increasing and then decreasing from Type I to Type II1 and then to Type III. The shale oil resources in the P1f shale were classified into four categories: enriched, moderately enriched, less efficient, and invalid resources. The enriched and moderately enriched resources are mainly Type II1 kerogen. Six favorable shale intervals were predicted vertically in the P1f of Well Maye 1. The 3D model used in this study can more accurately evaluate the quality of shale oil reservoirs in Mahu Sag. The research results provide favorable guidance for exploring and developing strongly heterogeneous alkaline shale oil in Mahu Sag and other regions in the world.

Continental shale oil reservoirs in China exhibit strong heterogeneity, with complex fluid occurrence and distribution patterns that lead to low shale oil and gas recovery rates. Therefore, CO2 injection for reservoir stimulation is a highly promising key technology for shale oil development. This study conducted supercritical carbon dioxide (SC–CO2) soaking experiments under high-temperature and pressure conditions (100 °C, 25 MPa) and characterized the pre- and post-soaking reservoir changes using methods such as X-ray diffraction (XRD), gas adsorption, and high-pressure mercury intrusion. It systematically analyzed the reservoir stimulation mechanisms of SC-CO2 in four lithofacies shales of the Funing Formation in the Suibei Basin—felsic-clayey, felsic-dolomitic, felsic, and dolomitic shales. These findings indicate that after SC-CO2 soaking, carbonate minerals and organic matter dissolve preferentially, resulting in a passive increase in the relative contents of quartz and clay minerals. The resulting reconstruction of shale mineral composition alters the pore structure, whereas adsorption-induced swelling of clay minerals reduces micropore volume. Carbonate dissolution generates secondary pores, significantly increasing throat radius. However, secondary mineral precipitation clogs pore throats, intensifying reservoir heterogeneity and reducing mesopore volume. Owing to differences in mineral composition, the stimulation effects of SC-CO2 vary among the four lithofacies. Felsic shales, with high quartz content, exhibit strong resistance to dissolution and are less affected by SC-CO2. Dolomitic and felsic-dolomitic shales, with contain high carbonate content minerals, are highly sensitive to SC-CO2 because of intense dissolution. Felsic-clayey shales, which are rich in clay minerals, exhibit complex pore dynamic equilibrium constrained by mineral transformation and particle migration. This study provides a theoretical basis for CO2 injection to enhance oil recovery while achieving carbon sequestration in shale oil reservoirs.

The primary sedimentary fabrics that trigger the formation and development of soft-sediment deformation structures worldwide are still not fully understood and even questionable. Little specialized investigation has been conducted on saline-lacustrine depositional settings characterized by multisource mixed sedimentary sequences displaying strong heterogeneity in fabrics and textures, lithofacies associations, and sedimentary structures. This work presents a systematic observation, classification, and interpretation of soft-sediment deformation structures developed in the saline-lacustrine fine-grained mixed sedimentary sequences of the middle Permian Lucaogou Formation. Here, seven soft-sediment deformation structure categories are depicted and explained in detail according to quantitative sedimentologic and petrographic constraints, including ductile (minor folds and convolute lamination, as well as load structures), brittle (minor faults, lithic breccias and debris-flow structures, boudinage structures, and sedimentary dikes), and hybrid brittle-ductile structures. Relationships were documented between soft-sediment deformation structures and lithofacies categories, emphasizing that dolomicrites and dolarenites are favorable for the formation of minor faults characterized by obviously higher average slip and dip angles of similar to 0.30 cm and similar to 64.0 degrees and similar to 0.20 cm and similar to 50.6 degrees, respectively, and dolomitic siltstones and mudstones are conducive to the development of both sedimentary dikes/clastic sills and also boudinage structures characterized by significantly higher averages of maximum width of 2.04 cm and 0.83 cm and 0.84 cm and 0.65 cm, respectively. Furthermore, both the feldspathic lithic silt to fine-grained sandstones and dolomitic lithofacies contribute significantly to the wide distribution of lithic breccias and debris-flow structures, yielding the highest maximum width of individual structures >0.45 cm. Potential triggering mechanisms for the differential development of soft-sediment deformation structures can be identified mainly through correlations between critical petrographic parameters (e.g., minor fault slip and dip angle, average maximum width of individual structures, etc.) and sandstone (argillaceous rock)/stratum ratios. This implies that there are crucial effects from terrigenous sandy input, argillaceous concentration, and brittle-ductile mechanical interaction on seismically induced liquefaction or fluidization. Finally, a conceptual model is proposed to illustrate the distribution pattern of soft-sediment deformation structures in the depositional architecture of saline-lacustrine fine-grained mixed sedimentary sequences, demonstrating that in the spatial variation from brittle to hybrid brittle-ductile and then to ductile types, the latter coincides well with decreases in both brittleness and arkosic concentration. Our findings demonstrate the importance of better understanding of the heterogeneous physical properties of saline-lacustrine shale reservoirs and their importance for guiding identification and fracturing development of oil resources, and this study provides a unique opportunity to explicitly highlight the broader scientific issues, including recognizing soft-sediment deformation structures as (1) geologic archives of paleoearthquake events, (2) seepage channels and accumulation spaces of strategic metals and mineral-metallogenic fluids, and (3) a unique mechanism for tracking continental tectono-dynamic and regional deformation processes occurring in multiple phases.

While CO2 huff-n-puff (CO2 HnP) is a promising technique for shale oil recovery, the characteristics and controlling factors of microscopically movable oil in lacustrine argillaceous-rich shales remain poorly understood. Shale samples from the Qingshankou Formation in the Songliao Basin were collected, and a series of experiments, including low-pressure N-2 adsorption, mercury injection porosimetry, and nuclear magnetic resonance, were conducted. High-pressure and high-temperature CO2 HnP experiments were then conducted to investigate the effects of cycle number, soaking time and changes in pore structure on movable oil distribution. The shales exhibit multi-scale pores and lamellar fractures containing substantial residual oil (41.33-52.16% saturation). CO2 HnP effectively mobilizes oil from macropores (50-1000 nm) and fractures (>1000 nm), with a limited effect in micro-mesopores (<50 nm). Three CO2 HnP cycles were optimal for movable oil extraction. Extending the soaking time increased movable oil by similar to 4%, primarily from macropores and fractures (5.59-6.05%), with minimal improvement in smaller pores. A combination of CO2 flooding followed by CO2 HnP increased total movable oil by 4.83-7.26%, significantly enhancing recovery from micropores (7.26%) and macropores (9.21%). This study clarifies the pore size distribution and mobilization constraints of movable oil in argillaceous-rich shales. The integrated CO2 flooding and HnP strategy proves to be highly effective, especially for movable oil in micro-mesopores. This study is the first to investigate pore-scale movable oil in lacustrine argillaceous-rich shales during CO2 huff-n-puff under in situ reservoir conditions, and could provide critical insights for optimizing shale oil recovery in the Songliao Basin and similar lacustrine reservoirs.

In the exploration and development of marine shale gas in southern China, traditional lithofacies classification does not fully account for the effect of thermal maturity on reservoir space, resulting in suboptimal development of thermally mature marine shales. Therefore, a reservoir facies classification method based on thermal maturity, organic matter content, and mineral composition was proposed to improve the accuracy of reservoir evaluation and identify optimal reservoir facies types. Thin section observation, scanning electron microscopy (SEM), and various experimental tests were conducted to systematically analyze the reservoir facies characteristics of the Lower Silurian Longmaxi Formation shale in the southern Sichuan Basin, and a comprehensive reservoir evaluation was conducted. The research results showed that the reservoir facies of the Longmaxi Formation shale in southern Sichuan were mainly overmature organic-rich siliceous shale (OR-S) and overmature organic-rich mixed shale (OR-M). Vertically, the reservoir facies showed a transition from overmature organic-rich siliceous shale (OR-S) at the bottom to overmature organic-rich mixed shale (OR-M) at the top. Laterally, the reservoir facies exhibited significant heterogeneity, and the siliceous mineral content gradually decreased from southwest to northeast, showing a transition from overmature organic-rich siliceous shale (OR-S) to overmature organic-rich mixed shale (OR-M) and overmature organic-rich argillaceous shale (OR-A). Through grey correlation analysis, a reservoir facies index grading standard was established. Total organic carbon (TOC) content, gas content, porosity, reservoir facies thickness ratio, siliceous mineral content, and clay mineral content were selected as key evaluation indicators, identifying overmature organic-rich siliceous shale (OR-S) as the optimal reservoir facies. This reservoir facies is characterized by moderate thermal maturity, high TOC content, high gas content, large porosity, high siliceous mineral content, and large reservoir facies thickness, exhibiting superior reservoir performance.

The development of continental shale oil in the Fengcheng Formation, Mahu Sag, is hindered by complex lithofacies and strong reservoir heterogeneity, leading to unclear lithofacies distribution patterns in individual wells and ambiguous mechanisms governing oil-bearing property variations. To address these challenges, this study innovatively integrates artificial intelligence (AI) with multifractal theory. Leveraging scanning electron microscopy (SEM), nuclear magnetic resonance (NMR), multi-stage pyrolysis, and oil saturation data, we developed an intelligent lithofacies identification model to elucidate the differential oil-bearing mechanisms across lithofacies. By correlating logging parameters with fractal dimensions, we established a Full well section "lithofacies-pore structure-oil content" characterization model. Key findings include: 1. An ensemble learning model, optimized via Bayesian optimization and an improved grey wolf optimizer (IGWO), achieves continuous and accurate lithofacies identification in single wells. 2. Mineral composition and reservoir heterogeneity jointly control storage capacity. Carbonate minerals enhance oil-bearing properties through dissolution-induced pore enlargement, whereas quartz and feldspar exhibit dual roles-suppressing storage at low concentrations (<30 %) but optimizing pore networks via rigid scaffolding and dissolution at higher concentrations. 3. Multifractal analysis quantifies the "macropores govern free oil, micropores dominate adsorbed oil" mechanism. 4. A well-scale lithofacies-oil content model is established by integrating logging responses with fractal parameters. This study constructs an intelligent characterization framework for the shale oil, providing a standardized AI-driven solution for sweet-spot prediction and economic development.

Crude oil recovery and residual oil distribution at the microscopic pore scale are governed by pore structure, wettability, and imbibition dynamics. However, the evolution of pore structure and wettability during spontaneous imbibition and their impact on oil displacement efficiency remain poorly understood, particularly in saline lacustrine shale reservoirs with complex pore networks and variable wettability. The lack of a systematic framework linking imbibition to oil displacement efficiency hinders a comprehensive understanding of its controlling mechanisms, necessitating further investigation. Therefore, this study selected four typical lithologies from the Lucaogou Formation in the Jimusar Sag. Through a combination of methods, including high-pressure mercury intrusion (MIP), low-temperature N2 adsorption (LTN2A), CO2 adsorption (CO2GA), nuclear magnetic resonance (NMR), computer tomography scanning (CT), focused ion beam scanning electron microscopes (FIB-SEM) and QEMSCAN identification, along with multifractal theory and wettability index, the flow mechanisms of shale oil and the distribution of residual oil during pressurized spontaneous imbibition were investigated. The results: (1) All lithology samples exhibited a nonlinear recovery trend of "increase-decrease-increase" during pressurized spontaneous imbibition. (2) The pore connectivity of micritic dolomite is poor, with high heterogeneity. Dolomitic siltstone and arenaceous dolomite exhibit better mesopore connectivity and higher oil migration efficiency. Feldspathic lithic fine sandstone, with high pore connectivity, shows strong oil and gas aggregation and fluid transport capabilities in macropores. (3) Two typical residual oil distribution patterns and their controlling mechanisms were identified in the Lucaogou Formation shale reservoir: a macropore-dominated residual oil aggregation pattern associated with water-wet feldspathic lithic fine sandstones, and a micropore-mesopore dominated residual oil retention pattern commonly occurring in oil-wet micritic dolomite. This study systematically reveals the flow mechanisms of shale oil and the distribution of residual oil during pressurized spontaneous imbibition, based on the synergistic effects of mineral composition, pore structure, wettability, and capillary forces, providing a theoretical foundation and practical reference for improving shale oil recovery. (c) 2026 Publishing services by Elsevier B.V. on behalf of KeAi Communications Co. Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Accurately delineating the range of the Xiang-E submarine uplifts is the key to the exploration and development of Silurian shale gas in the Western Hunan–Hubei region. Based on the graptolite stratigraphic division of Well JD1 in Jianshi area, Hubei Province, and combined with the GBDB online database (Geobiodiversity Database), the study compared the shale graptolite sequences of the Wufeng Formation and Longmaxi Formation from 23 profile points and 11 wells which cross the Ordovician–Silurian boundary. The range of the Xiang-E submarine uplift was delineated, and its evolution model and formation mechanism at the Ordovician–Silurian transition were discussed. The graptolite stratigraphic correlation results of drillings and profiles confirmed the development of submarine uplifts in the Western Hunan–Hubei region at the Ordovician–Silurian transition–Xiang-E submarine uplift. Under the joint control of the Guangxi movement and the global sea-level variation caused by the condensation and melting of polar glaciers, the overall evolution of the Xiang-E submarine uplift is characterized by continuous uplift from the Katian Age to the early Rhuddanian Age, with the influence gradually expanding, and then gradually shrinking back in the middle and late Rhuddanian Age. The initial form of the Xiang-E submarine uplift may have originated from the Guangxi movement, and the global sea-level variation caused by polar glacier condensation and melting is the main controlling factor for the changes in its influence range. Within the submarine uplifts range, the Wufeng–Longmaxi Formations generally lack at least two graptolite zone organic-rich shales in the WF2-LM4, and the shale gas reservoir has a poor hydrocarbon generation material foundation, posing a high risk for shale gas exploration. The Silurian in Xianfeng, Lichuan, Yichang of Hubei and Wushan of Chongqing has good potential for shale gas exploration and development.

The tight sandstone reservoir features the development of micro- and nano-scale pores and micro-fractures, contributing to a complex pore-throat structure. This complexity results in an indistinct charging sequence of oil and gas in different types of storage spaces during the accumulation period, thereby escalating the challenges associated with the exploration and development of tight oil and gas. Focusing on the Fuyu oil reservoir in the northern Songliao Basin, this study integrates large-field-of-view stitched scanning electron microscopy with mineral surface scanning techniques. Innovatively, a micro-nano scale comprehensive reservoir evaluation method, incorporating both pores and micro-fractures, is proposed. Within the study area, three predominant pore-fracture combination types are identified: intergranular pore-clay mineral shrinkage fractures, intergranular pore-brittle mineral intergranular fractures, and intragranular pore-clay mineral shrinkage fractures. Building upon this, Wood's alloy, exhibiting high-temperature rheological properties, is injected into rock cores under various pressure conditions. It is observed that with increasing injection pressure, the alloy injection process demonstrates a structured order, with a clear preference for charging intergranular pore-fractures over clay mineral-related pores. Furthermore, the alloy injection efficiency curve exhibits a distinctive parabolic shape. Based on the characteristic properties of Wood's alloy and crude oil, the injection pressure is equivalently transformed, reconstructing the micro-scale charging process of tight oil under reservoir conditions. Consequently, a sequential charging model for tight reservoirs is established, encompassing micro-nano-scale intergranular pore-fractures, nano-scale clay mineral intragranular pores, and shrinkage fractures. This model considers parameters such as source-reservoir pressure difference, storage space type, fluid properties, etc. From both qualitative and quantitative perspectives, it clarifies the microscopic accumulation sequence of tight reservoirs. This research focuses on the development of a new multi-scale evaluation method for tight reservoir storage spaces, combining fluid injection with visualization technology. The findings are crucial for the microscopic sweet spot evaluation and efficient development of tight reservoirs.

Mobility is a crucial metric for assessing sweet spots of continental shale oil. However, due to the complexity of shale oil reservoirs characteristics and the lack of systematic analyses of factors influencing mobility, the difference in shale oil mobility under multiple lithofacies control remains unclear, causing significant challenges for mobility evaluation and sweet spot prediction. This study examines continental shales of the Fengcheng Formation in the Mahu Sag, employing scanning electron microscopy (SEM), nitrogen adsorption (NA), nuclear magnetic resonance (NMR), spontaneous imbibition (SI), and contact angle measurements (CAM) to investigate the pore structure, connectivity, and wettability properties of different lithofacies shale. Quantitative analyses of shale movable oil content and saturation were conducted using multistep temperature pyrolysis (MTP) and NMR centrifugation techniques. Furthermore, the influence of reservoir characteristics, geochemical characteristics, and lamination development on shale oil mobility were discussed. Results indicate that larger pore diameter, higher imbibition slopes, and lower fractal dimensions of movable fluid pores (D2) correspond to higher movable oil saturation. Organic matter exerts a dual effect on shale movable oil content. When the TOC is below a threshold, the movable oil content gradually increases with TOC. Laminations exhibit favorable reservoir properties and light oil enrichment, enhancing shale oil mobility. Massive siltstone (MS) develops interconnected intergranular pores with the best pore structure and connectivity, the lowest D2 values, and the highest shale oil mobility. Laminated felsic shale (LFS) and laminated calcareous shale (LCS) exhibit moderate mobility, where the development of microfractures enhances fluid flow by connecting isolated pores into pore-fracture networks. In contrast, massive felsic shale (MFS) and bedded felsic shale (BFS) primarily develop intragranular dissolution pores with more complex structures and poorer connectivity, resulting in weaker mobility. A more accurate approach for assessing shale oil mobility has been presented, taking into account both total oil content and movable oil saturation. More importantly, this study establishes a comprehensive conceptual model illustrating the potential relationships among shale lithofacies, reservoir characteristics, and movable oil flow space in the study area. This research not only provides a systematic approach for assessing shale oil mobility but also deepens the understanding of flow mechanisms of continental shale oil, offering theoretical guidance for optimizing sweet spots in the Fengcheng Formation shale oil reservoirs of the Mahu Sag.