In shale gas hydraulic fracturing, the effective placement and structural integrity of proppants are essential for maintaining the long-term conductivity of hydraulic fractures. However, under the severe conditions of high temperature, high pressure, and elevated in-situ stress in deep reservoirs, strong inter-particle contact forces and localized stress concentration render them highly susceptible to deformation and even fragmentation, resulting in a substantial decline in fracture conductivity. Constrained by the inherent limitations of conventional laboratory experiments in fully capturing the dynamic evolution mechanisms of microstructures during particle breakage, this study integrates laboratory experiments with the discrete element method (DEM) to systematically reveal the microscopic damage mechanisms and macroscopic mechanical responses of breakage for ceramic proppants and quartz sand proppants of varying sizes. A particle replacement model (PRM) incorporating Apollonian packing and equal-volume replacement was developed to simulate assembly-scale breakage and particle size distribution (PSD) evolution under closure stresses representative of deep shale gas reservoirs. The results indicate that at the single-particle scale, the breakage modes of proppants can be primarily classified into two categories: typical brittle fracture and complex multi-crack network propagation. The compressive strength of individual particles is predominantly governed by microscopic bond parameters and the friction coefficient, whereas the stiffness ratio exerts a negligible influence. At the particle assembly scale, particle breakage exhibits pronounced stress dependence and size effects. Increasing closure stress promotes progressive fragmentation, force-chain redistribution, and PSD reconstruction. Furthermore, under high closure stress, the breakage ratio of ceramic proppant remains remarkably low (ranging from 2.24% to 8.08%), which is substantially lower than that of quartz sand under identical conditions. The proposed multiscale DEM framework links microscopic crack evolution with macroscopic gradation reconstruction, providing a useful approach for evaluating proppant breakage and long-term fracture conductivity in deep shale gas reservoirs.

Natural gas hydrate-bearing sediments are weakly consolidated, time-dependent, and prone to sand production. During depressurization-driven production, coupled reservoir creep and formation sanding can rapidly impair propped fracture conductivity. This study develops a dynamically coupled analytical framework for evaluating long-term fracture performance by integrating a fractional-order rheological constitutive model with a physics-based sedimentation impairment model. Based on damage mechanics, the model describes the nonlinear evolution of fracture flow pathways governed by proppant–matrix interaction, proppant embedment, and fines deposition. Hydrate-bearing sediment specimens were synthesized by artificial reconstitution to reproduce the particle-size distribution of the target reservoir. Tests were conducted under low-temperature and high-pressure conditions representative of in situ hydrate reservoirs. Because continuous sanding is difficult to reproduce experimentally, silt layers of 0, 1.5, and 7.5 mm were pre-placed at the fracture base to represent different sanding intensities. Results indicate that hydrate-bearing sediments exhibit a three-stage creep response controlled by confining pressure. Reduced lateral confinement weakens skeletal stiffness and accelerates the transition to accelerated creep. Conductivity degradation is jointly governed by creep-induced proppant embedment and pore-throat plugging by deposited fines, while greater sediment accumulation and higher closure stress intensify nonlinear conductivity loss. Once the applied stress exceeds the skeletal yield strength, conductivity drops abruptly, indicating a distinct threshold effect. Parameter analysis shows that damage factor α controls microdamage accumulation, whereas fractional order γ characterizes delayed viscous dissipation before structural collapse. This study provides a theoretical basis for fracturing design, sand-control optimization, and productivity evaluation in gas hydrate reservoirs.

Proppant transport in the wellbore-multi-cluster hydraulic fracture system is a multi-factor coupled competitive flow process, with unclear transport mechanisms and distribution uniformity. Therefore, an engineering-scale MP-PIC model was developed to couple a horizontal wellbore with multi-cluster hydraulic fractures, and numerical simulations were conducted to analyze proppant transport from the wellbore into multiple fracture clusters under varying injection time, cluster number, spacing, approach angle, and fracture width. The coefficient of variation (CV) was introduced to quantitatively evaluate the uniformity of proppant distribution among fractures. Results show that the proppant transport process can be divided into four stages: initial injection, early particle settling, proximal fracture filling, and final equilibrium. Proppant distribution uniformity is negatively correlated with cluster number and positively correlated with cluster spacing, while it first increases and then decreases with increasing fracture width and is weakly affected by approach angle. Increasing cluster number raises the CV by 0.75, whereas increasing cluster spacing and fracture width reduces the CV by 0.66 and 0.47, respectively. A smaller approach angle increases flow resistance, promotes vertical proppant accumulation, and shortens transport distance. Optimizing cluster number, spacing, and fracture width can therefore improve proppant distribution uniformity and conductivity among multiple fractures. (c) 2026 Published by Elsevier B.V. on behalf of The Society of Powder Technology Japan. All rights are reserved, including those for text and data mining, AI training, and similar technologies.

Traditional thermodynamic models of natural gas hydrate formation posit that nucleation occurs only under high-pressure and low-temperature conditions. These thermodynamic models overlook liquid-solid interfacial adsorption of gas molecules, which creates high-density surface nanobubbles in gas-liquid-solid systems, leading to significant local gas enrichment even at low pressures. This study reveals a unique nucleation pathway for methane hydrate from the three-phase contact lines of the surface methane nanobubbles under atmospheric pressure and 277.15 K. High-resolution atomic force microscopy shows hydrate crystals growing radially from nanobubbles on highly oriented pyrolytic graphite, while Raman spectroscopy confirms that the crystal is structure H (sH). This phenomenon is attributed to the high local methane concentration in the environment surrounding surface nanobubbles and a reduced nucleation energy barrier at the contact line. Our findings redefine the spatial and thermodynamic boundaries of hydrate stability, suggesting that hydrates can occur as widespread nanoscale deposits within natural reservoirs, which raises concerns about their susceptibility to releasing methane under global warming.

In this study, a coupled modeling framework is developed by integrating an embedded discrete fracture-finite volume poroelastic geomechanics model with a finite-discrete element fracture propagation model. This coupled system enables a rigorous investigation of the spatialtemporal evolution of the in-situ stress field, fracture propagation behavior, and interwell frac-hit mechanisms under various geological conditions. The proposed framework is validated against analytical solutions and commercial simulators. Results show that with increasing production time, the maximum and minimum principal stresses initially decrease and subsequently recover, whereas the shear stress and the rotation angle of the maximum principal stress first increase and then diminish. Higher reservoir permeability and a larger horizontal stress difference accelerate stress reversal, enlarge the reversal zone, and increase the rate of stress evolution. As parent-well production continues, infill-well hydraulic fractures tend to deflect toward the parent well, enhancing longitudinal asymmetry and increasing frac-hit risks, which are reflected by larger pressure increment magnitudes and shorter response times. These trends remain consistent across different stress-difference and permeability conditions. During the early production stage, lower permeability or a smaller stress difference leads to slower stress reorientation and smaller fracture-path deviation. During mid-production, pronounced stress reversal restricts vertical fracture propagation and suppresses direct interwell connection. In late production, frac-hits becomes inevitable, especially in high-permeability or high-stress-difference reservoirs, where longitudinal fracture asymmetry intensifies and pressure increments peak. Consequently, an optimal stimulation timing window for infill wells is identified, within which reservoir stimulation is maximized while mitigating frac-hit risks. Higher permeability shifts the optimal fracturing time earlier, whereas a smaller stress difference delays stress restoration and shifts the optimal timing later.

Conventional fracturing fluids often suffer from severe viscosity degradation under ultra-high temperature (>240 degrees C) reservoir stimulation conditions. This is primarily caused by thermal-oxidative degradation and high-shear forces. To address this challenge, a water-soluble reticulated nanocomposite polymer (HT-PVSDA) was developed by integrating nanofluidic topological modulation and free-radical-trapping. This material utilizes trifunctionalized nano-SiO2 as chemical crosslinking nodes and a hydrophobic monomer containing a quaternary ammonium imidazole group as a multifunctional radical trapping agent. The topological network architecture, synergistically reinforced by covalent bonding and self-assembly, confers significantly superior thermal stability upon it. Experimental results demonstrate that the viscosity remained at 112 mPas after 120 min of shearing at 240 degrees C and 170 s(-1). Scanning Electron Microscopy images distinctly reveal a robust spatial framework supported by SiO2 nodes. Density Functional Theory calculations indicate that the radical trap preferentially captures center dot OH radicals (reaction activation energy as low as 5.20 Kcal mol(-1)), subsequently forming a stable product requires only 5.48 kcal mol(-1) of activation energy, lower than the 67.11 Kcal mol(-1) and 51.29 Kcal mol(-1) required for center dot OH to propagate in the polymer backbone. This effectively interrupts the radical chain reaction propagation along the polymer backbone, thereby inhibiting thermal-oxidative decomposition. Furthermore, the polymer exhibits an intelligent temperature-responsive reinforcement mechanism: supramolecular interactions dominate at low-to-medium temperatures, while metal crosslinking and antioxidative functional groups activate within the ultra-high temperature regime. Notably, this system achieves high-strength crosslinking within 5 min at 180 degrees C, further improving heat resistance. This research offers a novel solution for surpassing the performance limits of ultra-high-temperature fracturing fluid materials, demonstrating significant engineering application potential for the development of ultra-deep hydrocarbon resources beyond 10,000 m.

ABSTRACT Nowadays, the exploration of unconventional reservoirs commonly faces the challenges brought by harsh natural reservoir conditions such as high temperature and high salinity. This research proposes and tests a novel thermoviscosifying polymer‐based fracturing fluid (HTEP) through combining natural reservoir conditions with polymer structural characteristics. HTEP is highly sensitive to environmental salinity that the existence of salt would initiate a transition in the self‐assembly mode between polymer molecules. This unique feature not only helps to adjust the viscosity of the fracturing fluid, making it easier to pump, but also enhances the thermal‐thickening effect, improving the fracturing performance of the fracturing fluid. Rheological tests reveal that HTEP possesses excellent shear‐thinning properties with the occurrence of stress decay, while thermos‐shear measurements show that at 100 s −1 and 100°C, the viscosity retention rate of HTEP solution is 80.49%, exhibiting high thermos‐shear stability and thermos‐thickening property. Molecular dynamics simulations reveal that with increasing system temperature, the associative interactions among polymers are enhanced, explaining the thermoviscosifying behavior of HTEP and its stability at elevated temperatures. Therefore, taking into account the above characteristics, this HTEP fracturing fluid has good potential for field applications.

Driven by the dual objectives of the global energy industry—substantially enhancing hydrocarbon recovery and achieving long-term CO2 geological sequestration—this paper systematically reviews the development history and research status of integrated CO2 fracturing-enhanced recovery-storage technology, with a focus on the experimental investigations of CO2-water-rock interactions and the associated numerical simulations. Dynamic and static experiments collectively reveal the coupled dissolution-precipitation effects of CO2-water-rock reactions on reservoir properties under different temperature, pressure and time scales. Nevertheless, several limitations persist, including a scarcity of dynamic reaction equipment and corresponding data, insufficient investigation into micromechanical behaviors, and significant scale-dependent variations in mineral reaction rates. These limitations hinder the accurate prediction of porosity and permeability evolution over geological timescales. Regarding numerical simulation, existing studies have preliminarily modeled CO2 fracture propagation, multiphase flow, and storage behavior, with increasing use of thermo-hydro-mechanical (THM) coupling models and microscale approaches such as molecular dynamics. Nevertheless, current models exhibit notable shortcomings, particularly in coupling chemical mechanisms, characterizing microscale transport-reaction processes, and simulating the integrated fracturing-enhanced recovery-storage process. These shortcomings limit the ability to accurately predict how these reactions influence fracture growth and storage efficiency. Finally, this paper identifies persistent challenges, including the complex coupling of multiple physicochemical processes and the difficulty associated with achieving integrated full-process simulation. Future research should strengthen the integration of experimental and simulation studies, develop full-process, multi-field coupled numerical models, and optimize collaborative design and real-time monitoring systems. These advancements are essential to propel this technology toward large-scale industrial application.

The co-occurrence of marine gas hydrates, shallow gas, and deep oil and gas has been widely recognized, offering favorable conditions for integrated resource development. This study aims to explore the coupled production dynamics of shallow gas hydrates and underlying free gas to support the industrialization of marine natural gas hydrate resources. A productivity prediction model for the combined extraction of hydrate and underlying gas is developed, integrating multi-physics coupling mechanisms. Using logging data from two test wells in the Qiongdongnan Basin, a short-term physical model is constructed to simulate and predict production performance. Results show that: (1) The average wellhead gas production rate increases significantly with a linear relationship to the production pressure differential; (2) pressure propagation differs notably between hydrate and free gas layers, with a distinct low-temperature zone forming near the wellbore in the hydrate-bearing layer; (3) free gas migration results in a high-saturation gas zone near the hydrate-gas interface, and elevated pressure differentials can trigger secondary hydrate formation; and (4) during short-term testing, hydrate decomposition is limited, with most gas production sourced from the underlying gas layer. By adjusting production pressure or controlling wellbore temperature, hydrate decomposition and reformation can be balanced to optimize recovery efficiency. This study highlights a novel multi-gas co-production approach and provides a theoretical basis for sustainable deepwater hydrate development.

At present， the CO2 huff and puff method is mainly used in international oilfields to improve the recovery of shale reservoirs. However， challenges arise during the development stage of high-cycle huff and puff， including poor boundary physical properties， limited regional conventional huff and puff effectiveness， and usability difficulties. Supercritical carbon dioxide （SC-CO2） has the characteristics of low viscosity， high density， and strong solubility. In order to reveal the huff and puff effect of SC-CO2， the CO2 huff and puff test was carried out on the core after SC-CO2 immersion treatment. The influence of the huff and puff period， the shut-in time， and the injection pressure on the enhanced oil recovery （EOR） of SC-CO2 was analyzed. The shale before and after the huff and puff process was scanned using low-field nuclear magnetic resonance technology to analyze the distribution and extraction of oil from the shale matrix pores. The results show that the cumulative recovery of the shale samples after SC-CO2 immersion treatment increases with the increase in huff and puff cycles， while the single-cycle recovery decreases with the increase in huff and puff cycles. The core permeability after immersion increases by 1-2 orders of magnitude， and the porosity increases by 3-4 times； the shale surface becomes more CO2-absorbent. The extraction capacity of SC-CO2 can be increased by extending the shut-in from 6 hours to 24 hours， resulting in an approximately 10% increase in the recovery. The cumulative recovery and the single-cycle recovery of shale increase with the rise in injection pressure. The recovery can increase by 8%-17% When the pressure increases from 6 MPa to 15 MPa. The saturated oil in shale primarily exists in micropores and mesopores in the shale matrix， and the maximum amount of oil is produced within the first cycle of huff and puff.

Controlling wettability at mineral–fluid interfaces via external electric fields is a problem of fundamental interest in interfacial fluid physics, with relevance to technologies such as electrical enhanced oil recovery. In this work, molecular dynamics simulations are employed to elucidate how a direct-current electric field reorganizes the solid–liquid interface of a model calcite–brine–oil system. The oil phase contains n-octane, hexanoic acid, benzothiophene, and indole, which competitively adsorb onto the calcite {104} surface. In the absence of a field, polar oil species penetrate the hydration film and anchor directly to the lattice, giving rise to an intrinsic mixed-wettability state. Application of an electric field profoundly restructures the electric double layer: it induces differential migration of inorganic ions and strong orientational polarization of interfacial water. These effects transform a fragmented water film into a continuous, three-dimensional hydrogen-bonded network with an appreciably larger thickness. The reconstructed hydration film is consistent with an increased relative energetic favorability of the aqueous phase, reduces direct oil–calcite contact, and facilitates the outward displacement of the polar oil components. This interfacial reconstruction decreases the equilibrium contact angles of the polar-oil droplets from 110°–114° to 85°–87°, corresponding to a transition from oil-wet to weakly water-wet conditions. The mechanism is interpreted as an electrostatically controlled competitive adsorption phenomenon analogous to electrocapillarity on a mineral surface, in which field-induced interfacial-water reorganization makes a substantial contribution. These findings provide molecular-level insights into electric-field-tuned wetting and solid–liquid interfacial regulation, extending the physics of electrocapillarity to complex, multicomponent mineral–brine–oil systems.

To improve understanding of proppant transport mechanisms in complex fracture networks, a field-scale model based on the Multiphase Particle-in-Cell (MP-PIC) method was developed using fracture geometries derived from real shale outcrops and validated against experimental data. Results indicate that: (1) intense vortex formation during early-to-mid injection stages or at high flow rates exacerbates the longitudinal heterogeneity of proppant distribution. (2) In the near-wellbore zone, the synergistic effects of fracture width variations, high flow velocities, and natural weak planes drive proppants to migrate preferentially along paths aligned with the maximum and minimum principal stress directions, forming a dual-channel transport pattern. (3) The volume of proppant entering secondary fractures decreases with distance from the injection point, and the proppant dune height within dominant channels exhibits stepwise attenuation. Larger intersection angles between secondary and main fractures hinder proppant migration. (4) Smaller proppant size and lower density improve the planar sweep and distribution uniformity coefficients, while increased fracturing fluid viscosity extends the proppant sweep range and further improves uniformity; high injection rates promote long-distance proppant transport and broader coverage but may reduce uniformity, leading to sparse proppant distribution and necking at fracture mouths. These findings provide quantitative guidance for optimizing hydraulic fracturing designs.

In-fracture diverting fracturing is an effective technique for improving the quality of fracture network construction in deep shale gas reservoirs. However, theoretical guidance for the injection of temporary plugging agents (TPAs) remains limited. To investigate the transport and plugging behavior of TPAs within fractures under different injection parameters, a wedge-shaped fracture model considering proppant dune distribution is established, and a CFD-DEM coupling method is employed to perform numerical simulations. The results indicate that gravity and fluid drag play dominant roles in TPA transport within fractures. At the initial stage, particles settle under gravity. With increasing particle accumulation, restricted flow space enhances drag and drives particles toward the fracture end. In narrow fracture sections, particle interactions intensify. Large particles preferentially form bridges, and smaller particles fill the voids, resulting in a stable temporary plugging layer. An appropriate particle size proportion, moderate TPA concentration, increased injection velocity, and low-viscosity fracturing fluid are favorable for achieving efficient in-fracture temporary plugging. These findings provide theoretical support for optimizing TPA injection parameters in in-fracture diverting fracturing of deep shale gas reservoirs.

To study the influence of drilling fluid invasion on the mechanical properties of gas hydrate-bearing sediments, acoustic-electricity-mechanics and displacement experimental devices were used to prepare gas hydrate-bearing sediment samples under different temperature, pressure, and saturation conditions, and undrained triaxial experiments were conducted under an effective confining pressure of 3 MPa. Under the condition of an invasion pressure difference of 2 MPa, the invasion process of water-based drilling fluid into gas hydrate-bearing sediment samples was simulated, and a post-invasion triaxial experiment was conducted. A numerical simulation of drilling fluid invasion was conducted based on experiments, and the evolution laws of parameters such as temperature, pressure, hydrate saturation, stress, strain, and expansion rate were analyzed. The results are shown as follows: (1) Before the invasion of drilling fluid, as the hydrate saturation increases, the gas hydrate-bearing sediment sample changes from strain hardening to strain softening, with a critical saturation range of 15–25

To address the challenges of mechanical failure mechanisms and strength prediction for non-diagenetic hydrate reservoirs, triaxial mechanical tests were performed on non-diagenetic hydrate-bearing argillaceous silt sediments under in-situ static conditions. This study delineated the variation patterns of mechanical properties, including failure strength, volumetric strain, elastic modulus, cohesion, internal friction angle, and stress paths. It further elucidated the mechanical influence mechanism of heterogeneous fracture-filling non-diagenetic hydratebearing sediments (THFHBS-Hetero) on the sediment matrix. The results demonstrated that under 2 MPa effective confining pressure and 40 % hydrate saturation, the failure strength of single-layer THFHBS-Hetero was 31 % higher than that of homogeneous pore-filling non-diagenetic hydrate-bearing sediments (THFHBS-Homo). The reliability of the laboratory-fabricated samples was validated by comparing them with literature data and remolded cores from the Qiongdongnan Area. Drawing on the experimental results, this study innovatively proposes the particle-scale mechanisms and particle movement patterns governing the shear failure of porefilling and fracture-filling non-diagenetic hydrate-bearing sediments (ND-HBS), accounting for the effects of hydrate saturation, effective confining pressure, and the number of hydrate-filled fracture layers. It was determined that the shear failure of the sediments is governed by the detachment, sliding, and rotation of "hydratesediment" cemented aggregates. This research provides a theoretical basis for understanding the variation of mechanical parameters during subsequent well construction and the exploitation of non-diagenetic hydrates.

In-situ hydrogen production via air injection in gas reservoirs is a promising low-carbon technology, but its feasibility is strongly constrained by reservoir physical properties. In this study, the lower limits of porosity, pressure, and gas saturation for effective hydrogen generation were systematically investigated through adiabatic calculations and numerical simulations using a fractured horizontal well model. A total of 159 simulation cases were designed, and the hydrogen mole fraction in the produced gas was used as the evaluation criterion. The results show that to achieve a hydrogen mole fraction above 20 %, the reservoir pressure must be at least 18 MPa, with porosity > 0.05 and gas saturation > 0.5. The main reaction zone is confined to the hydraulic fracture network; the matrix outside the stimulated zone experiences temperatures below 240°C and contributes little to hydrogen production. Low-permeability reservoirs (≤ 0.01 mD) hinder fuel displacement, leading to higher boundary temperatures and more favorable conditions for hydrogen generation. A field case study on Well Ba7 (Sichuan Basin) optimized the huff-and-puff parameters, including pure oxygen injection, an injection rate of 30,000 m3/d, an cumulative injected volume of 1.5 × 106 m3, a soak time of 10 d, and a production rate of 40,000 m3/d. Under these conditions, a single cycle produces 0.673 × 106 m3 of hydrogen and 13.60 × 106 m3 of incremental methane, benefit ranging from 25.1 to 53.86 million RMB. The proposed screening criteria provide a quantitative basis for selecting candidate gas reservoirs and designing field pilots for in-situ hydrogen production.

The subsurface utilization of CO₂ has attracted increasing attention in carbon capture, utilization, and storage (CCUS) and unconventional hydrocarbon recovery. Quasi-dry CO₂ fracturing fluids, which contain only a small amount of water, offer a promising low-water stimulation strategy by mitigating water-sensitive formation damage while retaining the stimulation advantages of CO₂. However, achieving sufficient viscosity and proppant transport under high-CO₂ and low-water conditions remains a major challenge. Herein, an interfacial-stabilization-based dual-thickener strategy was proposed to develop a quasi-dry CO₂ fracturing fluid with improved mixed-phase stability, viscosity, and proppant transport capability. A CO₂-phase thickener (TM) and an aqueous-phase thickener (ACA) were synthesized, and the effects of CO₂:H₂O mass ratio and thickener concentration on system performance were systematically evaluated. The optimized formulation, containing 4.0 wt% TM and 1.8 wt% ACA at a CO₂:H₂O mass ratio of 7:3, achieved a viscosity of 57.41 mPa·s. The proppant settling rate decreased to 0.6784 cm/min, more than two orders of magnitude lower than that in pure SC-CO₂, while the core damage rate remained only 12.62%. Molecular dynamics simulations revealed that TM and ACA synergistically accumulated at the CO₂–H₂O interface and formed a stable interfacial bilayer driven by strong electrostatic interactions. Ion pairs between protonated amine groups and deprotonated carboxyl groups enhanced interfacial film strength, suppressed droplet coalescence, and promoted mixed-phase stability. This work provides a molecular-level strategy for designing high-performance low-water CO₂ fracturing fluids for unconventional reservoirs.

A three-dimensional multiphase particle-in-cell (MP-PIC) method was adopted to establish a liquid-solid two-phase flow model accounting for complex fracture networks. The model was validated using physical experimental data. On this basis, the main factors influencing proppant transport in fracture network were analyzed. The study shows that proppant transport in fracture network can be divided into three stages: initial filling, dominant channel formation and fracture network extension. These correspond to three transport patterns: patch-like accumulation near the wellbore, preferential placement along main fractures, and improved the coverage of planar placement as fluid flows into branch fractures. Higher proppant density, lower fracturing fluid viscosity, lower injection rate, and larger proppant grain size result in shorter proppant transport distance and smaller planar placement coefficient. The use of low-density, small-diameter proppant combined with high-viscosity fracturing fluid and appropriately increased injection rate can effectively enlarge the stimulated volume. A smaller angle between the main fracture and branch fractures leads to longer proppant banks, broader coverage, more uniform distribution, and better stimulation performance in branch fractures. In contrast, a larger angle increases the likelihood of proppant accumulation near the branch fracture entrance and reduces the planar placement coefficient.

Temporary plugging and diverting fracturing enhances unconventional reservoir stimulation by forming a tight sealing layer through the bridging and filling of diverters in fractures. The tightness and pressure bearing capacity of the sealing layer are governed by the clogging efficiency of fine particles and fibers in the bridging region. However, the clogging mechanism within the pores of coarse particles after bridging remains poorly understood. This study employs a hybrid CFD-DEM method to investigate the migration and clogging behavior of diverters in a bridging skeleton. The results indicate that fine particles initially migrate rapidly through dominant flow paths, while fiber movement is restricted by their high aspect ratio and mainly localized at the skeleton entrance. In single-size particle systems, larger fine particles form size-dominated clogs at pore throats, with clogging ratios increasing with both particle size and concentration. Mixed particle systems exhibit a more uniform clogging distribution and reduced sensitivity to concentration, owing to the cooperative filling of pore spaces. Fibers exhibit high clogging ratios under all conditions due to their tendency to attach and entangle on bridging particle surfaces. In single-size systems, particles whose sizes match pore throats and higher concentrations yield denser, less permeable sealing layers. Mixed particle systems result in even lower normalized permeability via synergistic filling. Fiber-induced clogging leads to the lowest permeability, forming continuous covers at pore entrances. For clogging stability, fine particles exhibit a "high contact number-high clogging ratio" pattern, while fibers display a "moderate contact number-high clogging ratio" pattern with fewer contacts required for retention. At low concentration, fine particles exhibit pronounced anisotropy in normal contact forces distribution, while mixed particle systems and fibers both reduce this anisotropy and promote a more uniform force network. This study clarifies multiscale sealing mechanisms and provides a theoretical basis for optimizing temporary plugging and diverting fracturing.

SC-CO₂ fracturing is a promising waterless stimulation technology for tight reservoirs because of its high fracturing efficiency, reduced water-sensitive formation damage, and potential for enhanced oil recovery. However, the intrinsically low viscosity of SC-CO₂ limits its proppant-carrying capacity. Although various CO₂ thickeners have been developed, the influence of thickener molecular architecture on thickening performance remains insufficiently understood. Herein, polysiloxane-based CO₂ thickeners with different architectures, including a linear thickener (FND), a branched thickener (HFND), and a nano-SiO₂-grafted thickener (DFND), were designed and synthesized. Their structures, thermal stability, phase compatibility, and thickening performance in SC-CO₂ without cosolvent were systematically investigated by experimental characterization and molecular dynamics simulations. The results showed that molecular architecture played a critical role in regulating thickening efficiency. Compared with FND, HFND exhibited a higher viscosity, which may be associated with enhanced intermolecular entanglement arising from its branched architecture. DFND achieved the highest viscosity, likely owing to the combined effects of the rigid nano-SiO₂ nodes and the structurally complex grafted architecture, which may facilitate the development of a more interconnected molecular network. At 60 °C, the viscosity of the DFND system reached 3.067 mPa·s, markedly higher than those of FND and HFND. Molecular simulations further demonstrated that DFND had the largest absolute interaction energy with CO₂, and the strongest restriction on CO₂ diffusion. These findings reveal the molecular-level structure–performance relationship of SC-CO₂ thickeners and provide a molecular-architecture-guided strategy for designing efficient CO₂ fracturing fluids.