In this work, a thermodynamically consistent fluid-driven thermo–hydro–mechanical–fatigue phase-field model is developed for dynamic mixed-mode fracture propagation under pulse fracturing conditions. The phase-field evolution equation is derived from the total Helmholtz free energy and the microforce balance, with the normalized crack-driving term decomposed into mode I and mode II contributions to distinguish the deformation mechanisms, critical energy release rates, and fatigue degradation behaviors associated with the two fracture modes. The mode I contribution mainly accounts for tensile volumetric deformation, fluid pressure, and thermal effects, whereas the mode II contribution is mainly associated with deviatoric deformation and is corrected by considering frictional resistance under compression. Moreover, mode-dependent fatigue degradation functions are introduced for the mode I and mode II critical energy release rates to characterize cumulative fatigue damage under cyclic fluid loading. The accuracy and applicability of the proposed model are validated through hydraulic fracturing experiments, single-edge-notched fatigue tests, and a ceramic plate quenching test under thermal shock. Finally, the mechanisms underlying complex fracture-network formation during pulse fracturing are systematically analyzed. The effects of injection rate and reservoir temperature on fracture-network morphology and reservoir stimulation performance in heterogeneous shale reservoirs containing natural fractures are investigated, together with the influence of cluster spacing on fracture interaction and stimulation performance during multicluster pulse fracturing.

To elucidate the mechanism and energy conversion efficiency of CO2 as a thermal fluid in in-situ oil shale development, this study employed ReaxFF molecular dynamics simulations to compare the pyrolysis behavior of kerogen under vacuum, Ar, and CO2 atmospheres, revealing the mechanism by which CO2 regulates the kerogen pyrolysis network. Furthermore, going beyond the limitations of conventional ex-situ uniform heating models, we simulated, for the first time, the in-situ CO2 injection process under simulated formation conditions, characterized the subsurface heat-transfer behavior of CO2, and quantitatively compared, at the molecular scale, the influence of subsurface non-uniform heat transfer on product conversion. Results show that pyrolysis under a CO2 atmosphere exhibits distinctive kinetic features of early initiation, deep cracking, and strong coke suppression, featuring the lowest onset temperature (1500 K) and the minimum coke residue (3.3 wt%). Unlike the vigorous secondary reactions in vacuum and the physical caging effect in Ar, CO2 can instantly capture highly active radicals and small fragments, suppressing heteroatom-involved cyclization and polycondensation to improve oil and gas quality. In-situ heating simulations further show that, due to thermal-front propagation, local thermal lag, and non-uniform temperature distribution, the conversion of valuable oil and gas under in-situ heating reaches 67.85 %, lower than the 77.95 % obtained under ideal ex-situ uniform heating. This study not only clarifies the molecular mechanism of radical regulation by CO2, but also provides a molecular-level basis for understanding the influence of non-uniform in-situ heat transfer on oil and gas conversion.

Understanding the priority of chemical bond cleavage during kerogen pyrolysis is important for improving hydrocarbon conversion efficiency and pollutant control in oil shale. In this work, ReaxFF molecular dynamics (ReaxFF MD) simulations were employed to investigate the product evolution and bond cleavage sequences of Type I kerogen under non-isothermal and isothermal pyrolysis conditions. The results showed that the heating rate regulated reaction pathways and influenced hydrocarbon conversion, while bond cleavage was directly controlled by temperature. Product distribution and bond cleavage calculations confirmed that lower heating rates favored gaseous product generation, whereas high-temperature C–C main-chain cleavage dominated framework decomposition and promoted hydrocarbon release. High-temperature isothermal treatment further promoted the cleavage of key bonds, including C–S, C–N, and C–C, facilitating the release of light gaseous products. Both simulations revealed significant stage-specific characteristics of bond cleavage. In the early stage, low bond energy sites near aliphatic chains, side-chain termini, and functional groups acted as reaction hotspots, while free radical accumulation promoted continuous bond cleavage, supporting kerogen conversion into favorable oil and gas resources. This study elucidates the controlling factors of pyrolysis reactions from the perspective of microscopic bond evolution and provides a theoretical basis for pyrolysis parameter optimization, process design, and pollutant control in oil shale utilization.

In this study, a novel gradient enhanced nonlocal damage framework is proposed to model fluiddriven fracture in an isotropic poroelastic medium. The framework incorporates tensile and shear energy limiters to characterize mixed-mode fracture behaviors, with driving forces for tensile and shear damage derived independently through volumetric-deviatoric decomposition strategy. The governing equations couple rock deformation and fluid flow through the effective stress principle and Darcy's law, while mechanical and hydraulic property degradation evolves in accordance with damage accumulation. The above hydro-mechanical-damage coupled system is solved via a robust staggered finite element algorithm with Newton-Raphson nonlinear iteration. Model accuracy is rigorously validated against laboratory experiment, analytical solutions, and established numerical benchmark, collectively confirming framework reliability across distinct physical regimes. Spatial and temporal discretization independence is further demonstrated through comprehensive sensitivity analyses, providing robust assurance of numerical fidelity. The validated framework is systematically applied to investigate fluid-driven fracture propagation under diverse scenarios. Critically, quantitative comparison between mixed-mode and tensile-mode formulations reveals mechanistically distinct fracture behaviors: tensile-mode fractures exhibit greater deflection angles and demand substantially higher injection pressures relative to mixed-mode fractures under identical conditions - demonstrating that neglecting shear damage systematically overestimates propagation pressure and mischaracterizes fracture geometry, underscoring the necessity of mixed-mode frameworks for predictive hydraulic fracturing modeling.

Hydraulic fracturing in cleat-developed coal seams is a multiscale problem involving both the opening behavior of cleats at the fine scale and the hydraulic fractures propagation at the macro scale. Traditional damage mechanics models fail to simultaneously describe fracture evolution across different scales. To address this limitation, this study proposes an adaptive cohesive phase-field model based on the multiscale finite element framework. Multiscale basis functions for the field variables are individually constructed by solving local linear boundary value problems, serving to transfer the degrees of freedom between the fine and coarse meshes. These basis functions from all coarse elements are then assembled into the global stiffness matrix and load vector, similar to the traditional finite element method, enabling multiscale computation on the coarse mesh. To alleviate the computational burden associated with the multiscale damage model, an adaptive mesh refinement technique is proposed. This technique utilizes the phase-field variable and its time derivative as refinement criteria to refine the mesh in the primary fracture region and the impending fracture region, respectively. The required refinement level is determined through a benchmark test, which demonstrates its effectiveness in significantly enhancing computational efficiency. Finally, the fracture initiation mechanisms and propagation patterns in cleat-developed coal seams are investigated from the fine scale and the macro scale, respectively.

The stereoscopic well pattern development model serves as a key method for enhancing resource recovery efficiency in shale reservoirs. With the ongoing trend of reduced horizontal well spacing, hydraulic fracturing operations face challenges due to heterogeneous in situ stresses caused by adjacent well production and the presence of natural fracture zones. These factors can lead to excessive fracture propagation, raising the risk of fracture interference and casing deformation, which represents a critical and complex issue in fracturing. To address this challenge, in this study, the deep shale reservoir of the Longmaxi Formation in the Luzhou block, Sichuan Basin, is selected as the target reservoir. Considering the typical geological and engineering characteristics of this area, including great burial depth, high horizontal stress difference, and well-developed natural fracture zones, an integrated fluid–geo-mechanical coupling-based 3D fracture propagation simulator was developed. This model incorporates an interaction criterion between hydraulic and natural fractures, accounts for multi-scale production-induced stress variations, and integrates multi-physics coupling effects, such as fluid leak-off, adsorption-induced deformation, stress sensitivity, and gas–liquid two-phase flow. This study systematically investigates the influence of two main factors on fracture behavior: the natural fracture zones characteristics and heterogeneous in situ stress disturbance caused by adjacent well production, with each case highlighting key phenomena. In the analysis of natural fracture zones, the impact of zone location and width on fracture propagation is examined. For the stress interference scenario, the effects of well spacing, production time, and the number of perforation clusters are evaluated. By utilizing this integrated model and field data, the mechanisms underlying these phenomena are revealed. Key findings include: (i) Compared with conventional 2D models, macroscopic natural fracture zones with an approach angle ≤ 40° exhibit a significant blocking effect on hydraulic fractures, due to the normal stress exerted on the fracture walls. (ii) Wider low-angle natural fracture zones (≥ 40 m) and greater distances (≥ 60 m) result in more severe barrier effects, increasing the risk of uncontrolled fracture propagation on the side without natural fracture zones and inducing inter-well frac hits. (iii) In the context of the deep shale reservoir in the Luzhou block, at well spacings ≤ 310 m, stress disturbance from adjacent well production creates a heterogeneous low in situ stress region between the two wells. This causes hydraulic fractures to extend sharply in the low-stress region, which can directly reach the production wellbore under some conditions. (iv) Smaller well spacing and longer production time lead to higher inter-well frac hit risks under production-induced stress disturbance. Simulation results based on the model developed in this study indicate that these adverse effects of production interference can be mitigated by increasing well spacing (> 310 m) and the number of perforation clusters (6 9 clusters), thereby limiting fracture length in the stress interference region. This study provides valuable insights for resolving the contradiction between maximizing the stimulated reservoir volume (SRV) of fracturing wells and avoiding inter-well fracture connectivity regarding fracturing scale, shedding light on the challenges and opportunities for maximizing oil and gas recovery in shale gas stereoscopic well pattern development.

Hydraulic fractures in unconventional reservoirs typically exhibit narrow widths and pronounced surface roughness. Under such complex geometries, slurry flow and pressure-drop behavior deviate significantly from the laminar-flow assumption. Existing models cannot consistently describe the coupled effects of fracture roughness, proppant diameter, and fracture width on the pressure-drop-velocity relationship. Based on computational fluid dynamics-discrete element method simulations of slurry flow in rough fractures, this study extends the classical Forchheimer law. A Taylor expansion is applied to the viscous term, and the inertial term is generalized using a power-law formulation. The resulting model explicitly unifies the linear viscous, nonlinear viscous, and inertial contributions and captures the coupled influence of fracture roughness, proppant diameter, and width in shaping the nonlinear pressure-drop-velocity relationship. The results indicate that rough fractures operate in a nonlinear flow regime under most conditions. Asperity-scale roughness and particle confinement reduce the effective fracture width and enhance both viscous and inertial dissipation, causing slurry flow to depart markedly from that predicted by Poiseuille's law even at relatively low Reynolds numbers. Based on these findings, this study proposes an explicit nonlinear pressure-drop correction function Phi(p) to quantify the deviation of the actual pressure gradient from that predicted by Poiseuille's law. This correction mitigates the limited applicability of Poiseuille's law when describing flow in rough fractures. The proposed framework provides a transparent, transferable basis for interpreting nonlinear slurry flow in rough fractures and for implementing nonlinear pressure-drop corrections in field-scale slurry flow simulators.

In unconventional reservoirs, hydraulic fracturing generally creates a network of narrow fractures characterized by complex connectivity and rough surfaces. This unique fracture morphology directly affects the flow behavior of slurry and proppant. In this study, a friction-collision constitutive model is developed within the Euler-Euler framework by incorporating collision-induced stress into the classical model of Dontsov and Peirce (2014) to describe proppant-fluid interactions in rough fractures. The proposed model captures the slurry flow and proppant transport dynamics in rough fractures through exact integral formulations, and provides approximate solutions that facilitate coupling with 2D fracture propagation models. Based on this framework, the effects of proppant size, density, and fracture roughness on the spatial distribution of proppant are quantitatively analyzed. The results indicate that: (a) Energy dissipation within fractures is primarily governed by particle-particle collisions, whereas particle-wall collisions are negligible. As fracture roughness increases, collisional dissipation intensifies, leading to a significantly reduction in the particles' kinetic energy; (b) Increasing fracture roughness decreases the effective transport width and shortens the horizontal transport distance of proppant. The use of smaller and lower-density proppants alleviates the hindering effect of fracture roughness on proppant transport; (c) When fracture roughness is below 5, its impact on proppant propped area and propped ratio is minimal; however, it becomes significant when the roughness exceeds 5.

Well drilling relies on the rate of penetration (ROP) as a core indicator of efficiency. Accurate ROP prediction is vital for enhancing the extraction efficiency, optimizing the drilling process and reducing the production costs. However, drilling in deep and ultra-deep reservoirs faces three major challenges: (1) the available data are sparse, (2) the dynamic drilling characteristics exhibit complicated nonlinear behavior and strong coupling among drilling variables, and (3) existing models often show limited generalization ability and lack interpretability. To overcome these limitations, we propose a Bayesian-optimized meta-learning framework that couples an LSTM-Transformer base learner with Model-Agnostic Meta-Learning (MAML) and Bayesian Optimization (BO). First, the time series data from multiple drilling wells are organized as related tasks, and an LSTM-Transformer hybrid network is constructed to capture short-term temporal variations together with global sequence-level interactions among drilling parameters. Subsequently, MAML is employed to learn a task-agnostic initialization across wells, enabling rapid adaptation of the base learner to a previously unseen well through several gradient steps. Then, BO is integrated to systematically tune key hyperparameters of both the base learner (e.g., LSTM units, Transformer heads) and the meta-learning process (e.g., inner/out-loop learning rate). Finally, Shapley Additive exPlanations (SHAP) are used to quantify the contributions of operational and geological/logging variables to the predicted ROP, providing both global and local interpretability. The proposed framework is experimentally validated on field drilling datasets from multiple oil wells, including experiments with noisy data and an out-of-domain dataset to assess cross-well generalization. The findings show that the Bayesian-optimized MAML-LSTM-Transformer consistently outperforms conventional learning baselines, achieving on the test set an R2 of 0.876, an RMSE of 1.687 m/h, an MAE of 1.324 m/h, and a MAPE of 9.056%. Overall, the results support the proposed framework as an effective and practically deployable solution for small-sample, cross-well ROP prediction.

Carbon capture, utilization and storage (CCUS) has become an important approach for reducing carbon emissions while enabling the resource utilization of CO2. CO2 fracturing as a representative application of CCUS in unconventional reservoir development, provides multiple advantages. However, the drag mechanism governing proppant particles in thickened CO₂ within fractures remains insufficiently understood. This limits the accurate description of particle transport and the optimization of engineering parameters. In this study, a three-dimensional numerical model for interphase drag was developed and numerically solved using the finite volume method. The model was then validated through high-temperature and high-pressure particle settling experiments. On this basis, the effects of temperature, pressure, interphase velocity, particle size, fracture wall, and particle concentration on interphase drag were investigated. The interphase drag mechanism is essentially controlled by the coupling between non-Newtonian rheology and flow restructuring. Compared with water and water-based fracturing fluids, the overall wall effect in thickened CO2 is much weaker. Unlike the models developed from particle beds and porous media, volumetric drag in thickened CO2 does not exhibit strong exponential amplification, because shear thinning weakens the viscous effect and the three-dimensional fracture geometry allows momentum recovery along the fracture. Furthermore, a single-particle drag coefficient model, a wall effect factor model, and a volumetric drag model applicable to thickened CO2 systems were established. The results provide both theoretical support and practical models for predicting proppant transport, improving flow conductivity, and optimizing operational parameters in CO2 fracturing.

By virtue of its ability to reduce rock breakdown pressure, displace adsorbed gas, and induce minimal reservoir damage, supercritical CO₂ has led to the development of several new fracturing technologies, such as enhanced fracturing and SC-CO₂ fracturing, which have become important research directions for medium- and deep-shale stimulation. In addition, because CO₂ is the injected working fluid, SC-CO₂ fracturing also has the potential to facilitate CO₂ storage by promoting subsurface retention of a portion of the injected CO₂, thereby offering a pathway toward synergizing reservoir stimulation with carbon sequestration. However, studies on fracture propagation and intercluster interference mechanisms when these new fracturing technologies are combined with multistage, multicluster fracturing are lacking. To address this issue, a thermo–hydro–mechanical–damage (THMD) coupling model considering the intermediate principal stress, reservoir heterogeneity, and natural fracture effects is developed to investigate the impacts of fracturing fluid systems, cluster spacing, and reservoir temperature on the final fracture network morphology. The results show that SC-CO₂ fracturing can activate a greater number of natural fractures and form a complex fracture network; however, its application should account for operating costs and the risk of uncontrolled fracture propagation. Although the stimulation performance of enhanced fracturing is not the most optimal, it can effectively reduce intercluster interference and promote more uniform propagation among multiple clusters. During field operations, excessively small cluster spacing significantly intensifies intercluster interference, restricts the propagation of intermediate clusters, and leads to reduced fracture length and a decrease in the stimulated reservoir area (SRA). Increasing reservoir temperature enhances the cold-shock effect and increases fracture complexity, but it may also intensify intercluster interference to some extent. This study provides theoretical support for optimizing fracturing design in medium- and deep-shale reservoirs.

Deep coal seams are ideal reservoirs for CO2 geological storage. However, their pore-fracture structures are usually saturated with a large amount of formation water after fracturing stimulation, making it crucial to effectively increase the solubility of CO2 in formation water for successful geological sequestration. Therefore, this study employs molecular simulation to investigate the dissolution stability of CO2 in water with the addition of sodium lauroyl sarcosinate (SLS) under the conditions of 60℃ and 20MPa. The results indicate that although CO2 can be captured by micelles in the short term, the system is difficult to mix stably at the macro level. Regardless of the concentration of SLS, micelle aggregation and solubilization resistance both increase the mixing free energy, leading to gas-liquid separation. Thus, the addition of SLS in deep coal seams cannot effectively promote CO2 geological storage. To enable environmentally friendly surfactants with biodegradability to function effectively in CO2 geological sequestration, a synergistic approach involving the addition of co-surfactants, natural polymers, and nanoparticles can be considered.

Due to the notable performance of supercritical carbon dioxide (SC-CO2) fracturing in terms of environmental protection, carbon emission reduction and resource utilization, it has gradually become an important direction for the green transformation of the oil and gas industry. In this paper, a phase-field regularized cohesive zone model is proposed for SC-CO2 fracturing in thermo-poroelastic media. The damage type of the rock during SC-CO2 fracturing is modeled as dynamic quasi-brittle fracture and the heat transfer between SC-CO2 and the rock is considered. Avoiding modeling the matrix and fracture regions separately and individually, the net pressure difference between them is neglected, and the fluid flow in both regions is modeled using a unified governing equation. Since the physical parameters of CO2 vary significantly with temperature and pressure, data on the physical parameters of CO2 at different temperatures (273.15 similar to 423.15 K) and pressures (0 similar to 100 MPa) are exported with the help of REFPROP and coupled with the governing equations of fluid flow and heat transfer by means of interpolating functions. The proposed model is discretized using the multi-field finite element method, and then divided into a thermo-hydro-mechanical coupling subproblem and a damage subproblem, both of which are solved sequentially by the staggered algorithm. The accuracy of the proposed model is verified by a published numerical case and a SC-CO2 fracturing experiment, then the fracturing effects of different fracturing fluids are evaluated, and finally, the influence of different engineering and geological parameters on the fracturing effects of SC-CO2 is discussed.

In composite wellbore blockages, organic coatings on inorganic deposits can markedly restrict effective contact between conventional acids and the internal inorganic components, thereby reducing the overall removal efficiency of the blockage. Based on the characteristics of gas-well blockages from the SN Gas Field, a W/O microemulsion acid system integrating organic dispersion and inorganic dissolution was developed. A composite blockage model with compaction and cementation states similar to those of the field deposits was constructed, and its unplugging behavior was systematically evaluated and the underlying mechanism was investigated through static dissolution, dynamic displacement, and imbibition experiments. The results showed that the microemulsion acid system achieved a static dissolution percentage of 89.4%, markedly higher than the 68.9% achieved by the conventional acid system. Under dynamic displacement conditions, the permeability-restoration efficiency reached 43.72%, compared with only 21.07% for the conventional acid system. An imbibition kinetic relationship was established based on the mass change during the imbibition experiments. The apparent imbibition coefficient of the microemulsion acid system was 0.87 g·h-1/2, compared with 0.29 g·h-1/2 for the conventional acid system, indicating a stronger spontaneous imbibition capacity and facilitating the entry of the unplugging fluid into the compacted blockage medium. These effects jointly promote blockage disintegration and flow-channel restoration, providing an experimental basis and a technical approach for the integrated treatment of composite wellbore blockages.

Deep carbonate reservoirs are rich in oil and gas reserves, but high-temperature and deep-burial conditions may cause conventional retarded acid to fail, thereby reducing acid stimulation efficiency. To address these challenges, microcapsule acids with high thermal stability, strong acid resistance, and delayed release capability were prepared. Methyl methacrylate and N-methylolacrylamide copolymer were selected as the shell material, and microcapsule acids capable of withstanding temperatures up to 180 degrees C were synthesized via the interfacial polymerization method. Optimized parameters for mini-emulsion preparation and the interfacial polymerization process improved the uniformity of microcapsule particle size. The particle size and its distribution, micromorphology, and shell material properties of the microcapsule acids were characterized by laser particle size analysis, infrared spectroscopy, scanning electron microscopy, differential scanning calorimetry, and thermogravimetric analysis. The experimental results showed that the average particle size of the microcapsule acids prepared by the interfacial polymerization method was 321.7 mu m, with most particles distributed between 200 mu m and 400 mu m, exhibiting uniformity and good dispersibility. Thermal analysis indicated that the shell material maintained its structural integrity under high-temperature conditions of 180 degrees C. Suspension tests demonstrated that the microcapsule acids could achieve long-distance transport in clear water and be evenly distributed in the formation. Acid release experiments confirmed that these microcapsule acids could achieve delayed release. These results further indicate that the microcapsule acids have broad application potential in acid fracturing of deep carbonate reservoirs.

Understanding the molecular mechanism of kerogen pyrolysis under microwave fields is crucial for enhancing the hydrocarbon conversion efficiency of oil shale. In this study, ReaxFF MD simulations were performed under coupled microwave fields to explore bond-breaking sequences and reaction pathways of Type-I kerogen under both microwave and conventional heating. Results show that microwaves significantly reduce the optimal pyrolysis temperature and enhance oil yields. The non-thermal effects of microwaves induce earlier cleavage of key bonds (e.g., Cal-S, Cal-O-Cal), confirmed by quantum chemical calculations showing significant bond elongation under electric fields, which intensifies primary pyrolysis reactions and effectively suppresses secondary cracking of valuable oil and gas molecules. Non-thermal effects also facilitate cyclization of S and N atoms, concentrating heteroatoms in aromatic fractions, aiding downstream purification of aliphatic components. Moreover, adjusting microwave field strength validated its quality-enhancing effect: moderate enhancement accelerates early formation of light oil and gas, while excessive strength intensifies radical and cross-linking reactions, slowing pyrolysis and lowering final yield.

Organic-inorganic composite blockages in wellbores and near-wellbore regions severely restrict hydrocarbon production. Conventional acids or organic solvents often fail to remove them effectively because of dense organic coatings, rigid inorganic frameworks, and strong particle cementation. Here, a high-acid-loading bicontinuous microemulsion acid (BC-MEA) was developed for oil-wet composite blockages recovered from the X1 well. Phase behavior, conductivity, Cryo-SEM, interfacial tension, and molecular dynamics simulations were combined to investigate interfacial assembly and acidic-species solubilization and transport, while blockage-removal performance was experimentally evaluated. At a Span 80/APEO mass ratio of 6:4, BC-MEA exhibited interconnected bicontinuous microdomains and an oil–water interfacial tension of 3.47 × 10-4 mN·m−1. Simulations revealed a more continuous interfacial molecular arrangement and superior potential-energy contribution to interfacial assembly for this formulation. Acidic species in BC-MEA exhibited dual-region synergistic solubilization, characterized by CH3COOH interfacial retention and H3O + aqueous-phase confinement. Compared with the surfactant-free oil–water system, their diffusion coefficients decreased by 77.8 % and 64.5 %, respectively, reflecting substantially restricted transport of acidic species within BC-MEA. BC-MEA remained macroscopically homogeneous at 90 °C and 25,000 ppm salinity, with a corrosion inhibition efficiency of 91.89 %. After BC-MEA treatment, the permeability recovery of composite blockages reached 44.63 %, exceeding those of the acidic blockage-removal fluid and the organic solvent by 30.98 and 37.49 percentage points, respectively. The single-end liquid-uptake experiment further indicated pronounced liquid-invasion and sustained-action capabilities, suggesting the potential for deep blockage-removal. Overall, the experimental and simulation results showed that the blockage-removal performance of BC-MEA was closely associated with its microstructural characteristics and interfacial regulation capability.

To develop lightweight ceramic proppants and to promote the high-value utilization of two different forms of perlite, this study employed raw perlite and expanded perlite as partial substitutes for bauxite and fabricated low-density proppants via disk granulation followed by high-temperature sintering. The effects of perlite type, dosage (5-20 wt%), and sintering temperature (1400-1500 degrees C) on the microstructure and properties were systematically investigated. The results reveal that a moderate addition of raw perlite and expanded perlite significantly promotes mullite formation and reduces proppant density. However, excessive addition increases porosity, decreases the aspect ratio of rod-like mullite crystals, and induces the formation of ellipsoidal mullite, thereby elevating the breakage ratio. The optimized expanded perlite proppants sintered at 1450 degrees C exhibited the optimal overall performance, with a bulk density of 1.46 f 0.01 g/cm(3) and an acid solubility of 5.98 f 0.09 %. The breakage ratio of 8.21 f 0.56 % under the industry-standard closure stress of 35 MPa, meeting the requirement of SY/T 51088-2014 (breakage ratio < 9 %). At a higher closure stress of 41.4 MPa, the proppants achieve a conductivity of 29.435 & micro;m(2)& sdot;cm. Notably, Fe3+ incorporation into the mullite structure enhanced structural stability, thereby improving mechanical performance. This work demonstrates a sustainable strategy for fabricating low-density ceramic proppants and highlights the potential of perlite-based materials as an environmentally friendly substitute for bauxite in hydraulic fracturing applications.

Carbon Capture, Utilization, and Storage (CCUS) technology holds significant strategic importance in unconventional oil and gas development, and supercritical carbon dioxide (scCO2), with its low viscosity, high diffusivity, and strong reactivity, has emerged as a promising fracturing fluid. However, most existing studies have focused on the reaction behavior under pure scCO2 conditions, while overlooking the pore-permeability responses induced by dynamic variations in scCO2 concentration during fracturing. In this study, tight sandstones from the S Formation of the SL Basin were investigated through CO2-brine-rock reaction experiments under CO2 concentrations ranging from 0% to 100%. By integrating XRD, contact angle, NMR, and permeability measurements into a synergistic characterization framework, the coupling mechanisms among pore structure, wettability, and flow capacity were systematically revealed. The results demonstrate that an scCO2 concentration of 30%-50% constitutes the optimal window, where permeability enhancement is most pronounced, wettability shifts from hydrophilic to neutral, and pore-throat restructuring improves connectivity. This study elucidates the synergistic effects of pore-enlargement and water-lock removal on the improvement of flow conditions. Poreenlargement primarily enhances the absolute permeability of the core, while water-lock removal primarily enhances the gas-phase relative permeability; however, the net increase in permeability is not a simple sum of the two contributions. These findings provide theoretical support for concentration optimization and sequestration applications in scCO2 fracturing technology.