Identification and Dynamic Activation of Gas Capillary Sealing in Caprocks: A Case Study from the Junggar Basin, China Authors: Yijin Qiang, Jiangxiu Qu, Hailei Liu, Yi Huang, Xiujian Ding, Ablimit Yiming, Keshun Liu, Ming Zha Published: 2026 ABSTRACT Caprock sealing mechanisms critically influence hydrocarbon accumulation. Gas capillary sealing, a specialized type, occurs in seals of interbedded coarse-fine sediments. Free gas accumulating in coarse-grained pore throats forms a microscopic “gas cap”, preventing upward migration of hydrocarbons from the underlying strata. Though recognized historically, prior studies overlooked the impact of formation pore gas. Experimental evidence confirms gas significantly enhances sealing efficiency. Sediment size contrast is essential: gas accumulation at coarse-fine interfaces strengthens sealing. This mechanism is documented globally, including the Anadarko Basin, ODP Site 975, and Junggar Basin. In the Penyijingxi sag periphery in Junggar Basin, conventional seals are suboptimal. The lithology is predominantly characterized by sand-mud interbeds. The direct cap rocks are widely distributed across multiple stratigraphic horizons, yet overpressure exists with reduced hydrocarbon dissipation postgas charging. Thus, gas capillary sealing is identified as the primary caprock mechanism here. It accounts for gas accumulation despite poor seals, advancing the understanding of sealing mechanisms in Junggar Basin. Introduction Cap rocks are essential for the formation and preservation of hydrocarbon reservoirs. Research into their sealing mechanisms forms the basis for assessing their sealing capacity and the preservation conditions of reservoirs. Since the concept of cap rock and fault sealing was introduced, petroleum geologists have begun to study the sealing properties of cap rocks and the preservation of hydrocarbons after migration and accumulation. At a dedicated AAPG seminar on cap rocks and hydrocarbon preservation in 1983, Downey proposed that cap rocks should be studied by integrating both microscopic and macroscopic aspects. Building upon foundational contributions from earlier researchers such as Magara and subsequent studies, Watts was the first to propose the capillary and hydraulic sealing mechanisms, marking the beginning of systematic, qualitative research into cap rock and fault sealing. It is currently accepted that cap rocks primarily exhibit three sealing mechanisms: physical property (capillary), overpressure, and hydrocarbon concentration sealing. The capillary sealing mechanism is the most widespread and common one. − Capillary sealing of natural gas differs significantly from the traditional capillary sealing mechanism. The core distinction lies in the presence of a “gas cap” at the interface between interbedded coarse-grained and fine-grained sediments, which results in the superposition effect of natural gas capillary pressure. Its core principle lies in the superposition of capillary pressure at the interfaces between coarse-grained and fine-grained sediments. Specifically, when free gas accumulates in the pores of coarse-grained sediments (e.g., sandstone), forming isolated “gas caps”, the capillary force at the interface is directed toward the coarse-grained layer, creating a localized barrier. If these gas caps within the seal do not form a three-dimensionally connected network, the capillary forces from individual interfaces can undergo linear superposition, significantly enhancing the overall sealing capacity. Since its conceptualization based on multiphase fluid behavior, , the gas capillary sealing mechanism has been widely validated through physical simulations , and field-scale applications in specific overpressured basins. ,, However, the dynamic activation process of this mechanism within the complex burial and multiphase hydrocarbon charge history of actual sedimentary basins remains poorly constrained. Hao first introduced the natural gas capillary sealing mechanism into the study of overpressured basins, indicating that the sealing capacity of unconsolidated sediment cap rocks originates from the capillary forces at coarse-fine grain interfaces. Furthermore, Zou Huayao et al. proposed that the sealing efficiency of this mechanism is superior to that of diagenetic tightness. They pointed out that the total hydrocarbon content beneath a natural gas capillary seal is higher than that beneath a low-permeability seal formed by diagenesis, indicating that the capillary sealing of natural gas is more efficient than sealing achieved through diagenetic densification. While these classical basin-scale studies firmly established the theoretical framework for conventional hydrocarbon preservation, the urgent demands of the modern energy transition have catalyzed a paradigm shift in caprock evaluation. Consequently, the connotation of the gas capillary sealing theory has undergone a significant expansion in both its application scenarios and methodologies. At the application level, the research focus has extended beyond traditional petroleum systems to emerging environmental fields, particularly geological CO2 storage (CCUS) and underground hydrogen storage (UHS). The more stringent requirements for caprock integrity and long-term mechanical stability in these applications have driven the development of multiphysics coupled evaluation models. , At the methodological level, the rise of artificial intelligence and machine learning technologies has provided revolutionary tools for quantitatively predicting key parameters controlling sealing capacity from vast geological data sets. For instance, recent studies have successfully applied algorithms like XGBoost and ensemble learning to achieve high-precision prediction of diagenetic facies in deep-water low-permeability reservoirs and physical properties in tight reservoirs. , These advances indicate that the data-driven research paradigm is deeply integrating with traditional theory, making dynamic and quantitative evaluation of caprock sealing capacity feasible. The cap rocks in the hydrocarbon-rich uplifts surrounding the Penyijingxi sag in the Junggar Basin are predominantly characterized by sand-mud interbedding with individual thin layers and a high mud-to-gross ratio. Their poor physical properties typically preclude the formation of an effective seal by traditional diagenetic means. However, recent exploration breakthroughs have revealed abundant petroleum resources in this specific study area [62]. Production tests have confirmed that these seemingly poor sand-mud interbeds provide excellent dynamic sealing capacity for the underlying hydrocarbons, yielding substantial industrial oil and gas flows in key exploratory wells such as Shixi-16, Shixi-18, and Xiayan-26 (Figure ). Schematic diagram of the natural gas capillary sealing model (modified from ref Copyright (2002), John Wiley and Sons.). (a) Pressure distribution within the tube; (b) comparison of pressure between the downstream and upstream end of the tube.(Adapted from ref Copyright (2001), American Association of Petroleum Geologists.). (a) Relationship between total capillary pressure and the number of sedimentary units; (b) relationship between average capillary pressure per unit and temperature. (Data sourced from from ref Copyright (2001), American Association of Petroleum Geologists.). Well log data from the Anadarko Basin: (a) gamma ray from 3430 m depth to 3465 m depth provide controls on lithology; (b) porosity inferred from density log­(DPRO) and Neutron log­(NPRL); (c) the porosity values (shaded) in areas saturated with gas that provided from DPRO minus NPRL. (Adapted from ref Copyright (2002), John Wiley and Sons.). Well log from the ODP 975 site:(a) porosity inferred from density log; (b) gamma ray from 2475 m depth to 2725 m depth provide controls on lithology; (c) the fluid overpressure. (Adapted from ref Copyright (1998), John Wiley and Sons.). (a) Structural units, well location in the study area; (b) stratum, lithology and petroleum system elements. Lithology of cap rocks and oil-testing results in the Penyijingxi sag and surrounding uplifts, Junggar Basin. To resolve this contradiction, this study hypothesizes that the dynamic gas capillary sealing mechanism serves as the primary containment factor in this region. Utilizing original high-resolution X-ray diffraction (XRD), static gas breakthrough pressure tests, and quantitative grain fluorescence on extract (QGF-E) analyses, we systematically characterize the petrophysical deficiencies of the caprocks and reconstruct the multiphase hydrocarbon charge history. However, explicitly verifying this dynamic mechanism within the complex geological framework of the Junggar Basin presents three distinct challenges: (1) the microscale identification of transient “gas caps” within highly heterogeneous, high-frequency sand-mud interbeds; (2) the precise chronological matching between multiphase hydrocarbon charge-dissipation events and the transient activation of the capillary barrier; and (3) assessing the coupled effects of extreme deep-basin overpressure and elevated temperatures on the gas–water interfacial tension. To resolve these complex boundary conditions, this study establishes a targeted, multiproxy evaluation framework. First, quantitative grain fluorescence on extract is utilized to distinguish fluid phase differentiation between the reservoir and the caprock. This serves as a critical temporal tracer to reconstruct the chronological shift from early liquid dissipation to late gas sealing. Second, high-resolution gas logging ratios are employed to mathematically differentiate macroscopic advective bulk leakage from microscale concentration-driven Fickian diffusion, directly tracking the real-time sealing efficacy. Finally, static breakthrough pressure tests are strictly coupled with in situ pressure coefficients to physically validate whether the dynamic superposition of capillary resistance strictly matches the magnitude of the measured extreme overpressure. The primary objective of this study is to explicitly verify the existence of the gas capillary barrier in the Junggar Basin, and to demonstrate how the dynamic superposition of capillary forcestriggered by late-stage massive gas chargingexponentially enhances the macroscopic sealing capacity of the caprock, thereby successfully halting upward leakage and preserving the underlying massive hydrocarbon accumulations. Theoretical Background Genesis of the Concept Capillary forces, serving as the fundamental mechanism that inhibits the migration of nonwetting phase fluids, have their theoretical roots in the study of unsaturated zone water movement and capillary barriers within soil physics and hydrology. In the field of petroleum geology, this principle was systematically applied to the study of caprock sealing, culminating in the classical theory of capillary sealing. The review works by Berg Schowalter and Berg were particularly foundational, with the theory being further developed and refined by scholars such as Downey and Watts. In recent years, emerging fields such as carbon dioxide geological storage (CCS) and hydrogen underground storage (UHS) have provided novel perspectives for testing and refining this traditional theory. In CCS, the lower interfacial tension of the CO2-brine system reduces the capillary sealing pressure, while long-term geochemical processes can dynamically alter caprock properties. In UHS, the molecular characteristics of hydrogen and its potential physicochemical interactions with caprock present new challenges to conventional evaluation methods. Research indicates that hydrogen exposure may alter the porosity and mechanical integrity of shales. In a study on preventing contaminated leachate from permeating the soil, Ross investigated the feasibility of utilizing capillary forces as a barrier. The analysis revealed that when fine-grained sediments (e.g., mudstone) overlie coarse-grained sediments (e.g., sandstone), infiltrating water is prevented from flowing downward into the underlying coarse layer due to capillary pressure. Instead, the underlying coarse-grained sediment acts as a barrier, retaining water within the pores of the overlying fine-grained sediment. This discovery of the capillary barrier effect in interbedded fine- and coarse-grained soils marked the inception of research into the natural gas capillary sealing mechanism. Iverson et al. investigated the sealing mechanisms of two-phase fluids under overpressure conditions in the Powder River Basin, Wyoming. They found that the presence of a single-phase fluid could not sustain overpressure sealing for an extended period, and that a long-lasting overpressure compartment could only exist in the presence of multiphase flow. Subsequently, Benzing and Shook, through their study on the role of formation pressure sealing in hydrocarbon migration and entrapment within basins, discovered that as formation temperature changes, dissolved gas is released from pore fluids. The presence of this gas then restricts fluid passage, enabling the sedimentary layer to form a seal that impedes the upward migration of buoyant hydrocarbons. Based on this, they proposed the “vapor-lock” concept, positing that the presence of pore gas can restrict fluid flow through the sediments, thereby forming an effective seal. This work underscored the critical role of gas in capillary sealing. At ODP Site 975, Revil et al. observed that the sediments exhibited high porosity and strong permeability. They reasoned that conventional causes of overpressure, such as disequilibrium compaction, were insufficient to sustain overpressure in such a basin. By proposing a novel sealing mechanism, they demonstrated the practical feasibility of capillary sealing and, for the first time, explicitly formulated the concept of natural gas capillary sealing. They defined it as follows: in a seal composed of interbedded coarse- and fine-grained sediments, when free gas accumulates in the pore throats of coarse-grained layers, forming minute “gas caps”, it generates a capillary force directed toward the coarse-grained sediment. This force blocks the flow of water and gas. Crucially, the capillary forces thus generated are capable of superposition. The mechanism of natural gas capillary sealing can be explained by the fact that the pressure changes across interfaces between layers of different grain sizes are additive, a phenomenon that has been demonstrated experimentally and can be theoretically explained. In the first model (Figure a), the pores are fully water-saturated. Initially, the system predominantly contains water, but as gas enters the larger pores of the coarse-grained sediment (sandstone), water is displaced. For the gas to invade the fine-grained sediment, the capillary entry pressure must be exceeded. Once the gas displaces the water in the first small pore throat, it can access the entire system. The total sealing capacity of this system equals the capillary pressure of the smallest pores; the capillary pressures are not additive between different layers. The second model (Figure b) illustrates how capillary pressures become additive in interbedded coarse and fine-grained sediments. In this model, each interface between coarse and fine-grained sediment contains an accumulation of free gas, forming a gas bubble. For the underlying hydrocarbons to invade the fine-grained sediment and displace the formation water, they must overcome both the capillary force and the resistance posed by these gas bubbles. To displace water from all pores simultaneously, the total force required is effectively the sum of the forces needed to displace water from each individual pore. Therefore, the total pressure difference required equals the sum of the pressure differences across each individual coarse-fine interface. Reprinted (Adapted or Reprinted in part) with permission from [Deming, D., Cranganu, C., & Lee, Y. (2002). Self-sealing in sedimentary basins. Journal of Geophysical Research: Solid Earth, 107­(B12), ETG 2-1-ETG 2-9]. Copyright [2002] [John Wiley and Sons]. Theoretical Derivation of Gas Capillary Seal A gas capillary seal forms when free gas accumulates and creates a “gas cap” within coarse-grained sediments. For this gas to invade the overlying fine-grained sediment, it must overcome the capillary pressure difference, ΔP c, generated by the seal. This capillary pressure difference is related to the difference in the pore-throat radii between the fine-grained and coarse-grained sediments, and is expressed as 1 δPc=2σ(1rf−1rc) In this expression, r f is the pore-throat radius of the fine-grained sediment, r c is the pore-throat radius of the coarse-grained sediment, and δP c is the capillary pressure generated by the natural gas capillary seal. According to the pore-network models and experimental studies by Lenormand et al., the essence of capillary sealing lies in the competition between the capillary pressure and the driving pressure. Flow stagnates when the local driving pressure is insufficient to overcome the capillary barrier. The flow behavior between two adjacent pores can be described by the following relationship 2 q={πr48aμ(δp−δPc),δp>δPc0,δp≤δPc In this relationship, α is a dimensionless shape parameter accounting for the noncylindrical geometry of actual pore throats, and q is the fluid flux between pores; r is the throat radius; δp is the fluid driving pressure; δP c is the natural gas capillary pressure; and μ is the fluid viscosity. In this equation, fluid flow occurs and the seal is breached when δp > δP c; conversely, fluid is trapped between pores and flow stagnates when δp < δP c. The numerical experiments by Rothman provided key evidence for this behavior. To upscale this microscopic mechanism to the macroscopic formation scale, Darcy’s law must be modified to incorporate the capillary pressure gradient effect. Under this framework, the natural gas capillary sealing can be described by 3 qf={−kkfμf(d(δp)dz−dPcdz),d(δp)dz>dPcdz0,d(δp)dz≤dPcdz Here, q f represents the volumetric flow through the cap rock containing free gas, k is the absolute rock permeability, k f is the relative permeability,μf is the viscosity of the nonwetting phase, and z denotes depth. In this formulation, an effective seal is established and natural gas flow ceases when the capillary pressure generated by the natural gas capillary sealing exceeds the driving pressure. Conversely, seal leakage occurs when the capillary pressure is less than the driving pressure. Based on compiled data, Schowalter, in his comprehensive review and systematic analysis, summarized the effects of temperature and pressure on interfacial tension based on previously published data. His work was pivotal in interpreting the significance of these parameters for capillary sealing. The results indicate that the gas–water interfacial tension is significantly higher than the oil–water interfacial tension. At shallow depths, under conditions of 20 °C and 0.1 MPa, the gas–water interfacial tension is approximately 0.07 N/m, while the oil–water interfacial tension is about 0.025 N/m, making the gas–water interfacial tension 2.8 times that of the oil–water interface. To evaluate the variation of interfacial tension with depth, the geothermal gradient must be considered. Assuming a representative geothermal gradient of 18.9 °C/km, concomitant rises in pressure and temperature with increasing burial depth lead to a substantial reduction in both gas–water and oil–water interfacial tensions. Notably, at a depth of 4000 m, the gas–water interfacial tension becomes approximately 7 times greater than that of the oil–water interface. Consequently, natural gas accumulating at coarse-fine sediment interfaces can form “gas caps”, thereby enhancing the capillary sealing capacity of the formation. In contrast, the oil phase experiences lower capillary resistance, making it more prone to migration compared to natural gas and less effective at contributing to the formation’s sealing capacity. Experimental Validation of Gas Capillary Seal The natural gas capillary sealing mechanism was not initially recognized as an effective means for large-scale fluid containment in sedimentary basins. Shosa and Cathles constructed an experimental flow model using layered porous media (core tubes filled with sediments of different grain sizes), providing the first experimental demonstration that the introduction of a nonwetting phase fluid could significantly reduce the permeability of a layered system through capillary effects, thereby visually revealing the dynamic formation process of this mechanism in sedimentary basins. Their experiments demonstrated that an interbedded coarse-fine sediment structure is essential, wherein the fine-grained layers (with pore-throat radii of approximately 2 μm) constitute the key capillary barrier units. The simultaneous presence of both a wetting phase (water) and a nonwetting phase (e.g., natural gas) is required, and the Laplace pressure difference generated by their interfacial tension forms the physical basis for the sealing effect. In the physical simulation experiments conducted by Shosa and Cathles, they investigated capillary sealing in two-phase flow through alternating coarse-fine sedimentary layers by constructing a layered porous media flow system. The system utilized a high-performance liquid chromatography (HPLC) pump to drive fluids through a 0.5 m long core tube with an inner diameter of 12.7 mm (Figure a). The tube was packed with a sedimentary unit sequence consisting of a 45 μm coarse-grained quartz layer (coarse silt grade), a 2 μm fine-grained quartz layer (very fine silt to clay grade), and another 45 μm coarse-grained quartz layer. Prior to experimentation, residual gases were eliminated through CO2 flushing (to displace pore air) followed by vacuum-assisted water saturation. In the single-phase flow experiment, a NaHCO3 solution was injected into the flow tube containing five sedimentary units. The pressure distribution was measured at a flow of 0.0835 mD, yielding values for porosity (0.46–0.48) and permeability (7.4 × 10–3 μm2) (Figure a). In the two-phase flow experiment, the system pressure was reduced to 2.0 MPa (below the CO2 dissolution pressure) using a back-pressure controller, triggering gas exsolution and the formation of gas–liquid two-phase flow. During pressure reduction, pressure fluctuations persisted for 30 min at the downstream end of the tube, after which the upstream end of the tube pressure stabilized at 1.12 MPa with no fluid flow through the tube (Figure b). This pressure drop remained stable throughout the 4 h observation period, indicating the successful establishment of a natural gas capillary seal. Reprinted (Adapted or Reprinted in part) with permission from [Shosa, J. D., & Cathles, L. M. (2001). Experimental investigation of capillary blockage of two phase flow in layered porous media. In: Petroleum Systems of Deep-Water Basins: Global and Gulf of Mexico Experience. GCSSEPM Foundation 21st Annual Research Conference, 791–813.]. Copyright [2001] [Society for Sedimentary Geology]. Their comparative experiments utilizing 2, 5, and 8 sedimentary units revealed a linear relationship between the total pressure drop and the number of units. At 20 °C, the total capillary pressure was approximately 0.43 MPa for 2 units, 1.12 MPa for 5 units, and 1.97 MPa for 8 units, demonstrating that the capillary sealing pressure increases with the number of sedimentary units (Figure a). Furthermore, a modified Laplace equation (incorporating interfacial tension and pore-throat radius parameters) accurately predicted the capillary sealing pressure of the sedimentary units and its temperature dependence at 20 °C, 50 °C, 75 °C, and 100 °C. For instance, at 50 °C, the total capillary pressure was about 1.08 MPa for 5 units and 1.90 MPa for 8 units. These results indicate that temperature also influences the capillary sealing pressure, with increasing temperature leading to a decrease in pressure. The average capillary pressure per sedimentary unit was obtained by dividing the total capillary pressure by the number of units. Calculations show that at 20 °C, the average capillary pressure per unit was approximately 0.24 MPa. This average pressure is also affected by temperature; however, the magnitude of this temperature effect diminishes as temperature increases (Figure b). The experiments demonstrate that each gas–water interface functions as an independent sealing element, and flow can only occur when the capillary sealing pressure of all such interfaces is exceeded. Since the sealing capacity of multiple gas–water interfaces is additive, layered, gas-bearing sediments can collectively form a substantial seal. Furthermore, the results reveal three critical prerequisites for the existence of the natural gas capillary sealing mechanism: first, the seal must be composed of interbedded coarse- and fine-grained sediments (sand-mud interbeds); second, the sealing interval must exhibit a pressure compartment, i.e., overpressure; and finally, free natural gas must be present within the interbedded coarse- and fine-grained cap rock. The presence of natural gas capillary sealing significantly enhances the sealing capacity of the cap rock, providing a plausible explanation for dynamic hydrocarbon sealing. Moreover, owing to the long-term stability of grain size distribution in sedimentary formations during diagenesis and the widespread occurrence of gas–water two-phase fluid systems in sedimentary basins, natural gas capillary sealing can be maintained over geological time scales. While an initial closed compartment provides the fundamental spatial boundary for gas entrapment, it is the dynamic activation of the superimposed capillary barrier that physically halts subsequent fluid dissipation. Consequently, the capillary seal acts as the decisive hydraulic mechanism that maintains, preserves, and amplifies the extreme overpressure generated by late-stage massive hydrocarbon charging or mechanical compaction. Reprinted (Adapted or Reprinted in part) with permission from [Shosa, J. D., & Cathles, L. M. (2001). Experimental investigation of capillary blockage of two phase flow in layered porous media. In: Petroleum Systems of Deep-Water Basins: Global and Gulf of Mexico Experience. GCSSEPM Foundation 21st Annual Research Conference, 791–813.]. Copyright [2001] [Society for Sedimentary Geology]. While classical macroscopic core-flooding experiments established the foundational framework of gas capillary sealing, the past decade has witnessed a paradigm shift in validating this mechanism through high-resolution, pore-scale methodologies. Recent microfluidic “lab-on-a-chip” experiments and in situ X-ray microcomputed tomography (micro-CT) have provided direct visual evidence of the dynamic formation of localized gas caps. These real-time observations confirm that nonwetting phase trapping is critically governed by snap-off events and interfacial pinning at coarse-fine structural boundaries. , Furthermore, advancements in 3D pore network modeling (PNM) and lattice Boltzmann method (LBM) simulations have enabled the quantitative calculation of nonlinear capillary pressure superposition in highly heterogeneous pore throats. These modern international advancements not only corroborate the historical hypotheses but also extend the quantitative application of capillary dynamic barriers to contemporary subsurface energy systems, proving essential for evaluating long-term containment integrity in supercritical CO2 geological storage and underground hydrogen storage). Sedimentary Case Studies To establish a robust analogical foundation for investigating the complex Junggar Basin, the Anadarko Basin and ODP Site 975 are selected as comparative field-scale case studies. The Anadarko Basin serves as a classical end-member of severe deep-basin overpressure in ancient cratonic settings, whereas ODP Site 975 represents the opposite end-member of shallow, unconsolidated modern deep-sea sediments. As established previously, natural gas capillary sealing can significantly enhance the sealing capacity of cap rocks. Following the proposal of this mechanism, Deming applied it to the Anadarko Basin, calculating and explaining how natural gas capillary sealing could enhance the cap rock’s effectiveness to the extent of sealing overpressure. An analysis of well logs from over 100 deep boreholes in the basin revealed the presence of multiple, discontinuous thin gas layers in more than half of the wells. Neutron porosity logs measure hydrogen ion concentration, which is significantly higher in water than in gas. Consequently, the presence of gas results in lower readings on neutron porosity logs. Conversely, the low density of gas causes density porosity logs to exhibit higher porosity readings. Therefore, when gas is present, the value obtained from subtracting the neutron porosity from the density porosity is positive (Figure b). The well log from a specific borehole in the basin, shown in Figure c, identifies 9–10 gas-saturated layers based on the crossover between density and neutron porosity curves. Reprinted (Adapted or Reprinted in part) with permission from [Deming, D., Cranganu, C., & Lee, Y. (2002). Self‐sealing in sedimentary basins. Journal of Geophysical Research: Solid Earth, 107­(B12), ETG 2-1-ETG 2-9]. Copyright [2002] [John Wiley and Sons]. A straightforward calculation demonstrates the efficacy of the capillary sealing mechanism. In interbedded sandstone and mudstone sequences, the pore throats in sandstone are typically at least ten times larger than those in mudstone. Using eq provided earlier, the calculated average pore-throat radius is 1.6 × 10–8 m, which is consistent with the average shale pore-throat radius of 2.5 × 10–8 m reported by Krushin. , Under in situ conditions, with a gas–water interfacial tension of 2.5 × 10–2 N/m (kg·s–2), the pressure drop across a single gas–water interface can reach 3 × 106 Pa. Within a 30 m interval, identification of 9 to 10 gas-saturated layers via the crossover of density and neutron porosity logs implies the presence of 18 to 20 individual natural gas capillary sealing interfaces. If the shale layers lack hydraulic connectivity in three-dimensional space, the capillary pressures across these interfaces are additive. Consequently, the cumulative sealing strength potentially achieved by the natural gas capillary seal could exceed 54 MPa. This strength is sufficient to seal the observed 50 MPa overpressure in the Anadarko Basin. This calculation confirms that the natural gas capillary sealing mechanism can provide adequate sealing capacity even where the conventional cap rock properties appear deficient. Ocean Drilling Program (ODP) Site 975 is located on the South Balearic margin, between the Balearic Strait and the Algerian Basin, at a water depth of 2426 m. In situ fluid pressure variations indicate the presence of under-compacted sediments at shallow depths, primarily composed of Pliocene-Pleistocene unconsolidated to weakly consolidated clays, silts, and pelagic oozes. The terrigenous sediment fraction includes clay minerals such as Illite, kaolinite, and smectite, as well as quartz and minor amounts of feldspar and other minerals. Sandstone and mudstone interbeds alternate with dark clays in layers of varying thickness. The well log data from ODP Site 975 are shown in Figure , including porosity variation, gamma-ray curve, and fluid overpressure. The fluid pressure in the formation was calculated using a variant of the fluid mass-balance equation, represented by dots in the figure. The lithostatic pressure was derived directly from the integration of density log values, shown as a solid line (Figure c). Below 2596 m, the fluid pressure increases, paralleling the lithostatic pressure, with a fluid overpressure of approximately 2 MPa. To maintain a fluid overpressure equivalent to the lithostatic pressure, the required permeability would be 4 × 10–7 mD (approximately 4 × 10–10 μm2). This represents an anomalously low permeability for an under-compacted formation. The permeability of the shallow, unconsolidated sediments at ODP Site 975 is approximately 5 orders of magnitude higher than this value. Consequently, traditional sealing mechanisms cannot explain such high sealing capacity in unconsolidated sediments. As indicated by the gamma-ray curve (Figure b), the sediments at Site 975 are not pure shale. Although the grain size distribution is unknown, the capillary entry pressure measured in the aforementioned experiment for artificial quartz sand with grain diameters of 2 and 45 μm and a porosity of 45% was 0.24 MPa per interface. These two values are nearly identical. Approximately ten gas peaks can be observed in the gamma-ray log within the 2566–2596 m interval, implying the presence of about ten natural gas capillary barriers. Using the calculated pressure difference of 0.24 MPa per interface, ten barriers would contribute a total of 2.4 MPa. This result confirms that the natural gas capillary sealing mechanism provides a robust explanation for the fluid overpressure observed at ODP Site 975. Reprinted (Adapted or Reprinted in part) with permission from [Revil, A., L. M. Cathles III, J. D. Shosa, P. A. Pezard, and F. D. de Larouziere, Capillary sealing in sedimentary basins: A clear field example, Geophys. Res. Lett., 25, 389–392, 1998.]. Copyright [1998] [John Wiley and Sons]. Geological Setting The Penyijingxi sag and its peripheral uplifts are characterized by abundant hydrocarbon production. However, the cap rocks in this area typically feature sand-mud interbeds with small individual thickness and a lack of stable, thick mudstone layers, which are conventionally considered insufficient for effectively sealing large-scale hydrocarbon accumulations. Nevertheless, exploration results confirm that thin, interbedded mudstones can still provide effective sealing for hydrocarbon reservoirs. Data analysis and experimental results indicate the development of overpressure beneath the Jurassic in the study area. Furthermore, the degree of hydrocarbon dissipation has been observed to decrease after the charge of condensate gas. This study concludes that the primary sealing mechanism for the cap rocks in the Penyijingxi sag and its surrounding uplifts is natural gas capillary sealing. Junggar Basin, located in northwestern China, is one of the country’s significant large composite petroliferous basins, covering an area of approximately 13,600 km2 and exhibiting an overall irregular triangular shape. The basin is primarily divided into six first-order tectonic units: the Luliang Uplift, Western Uplift, Central Depression, Eastern Uplift, Ulungur Depression, and the Northern Tianshan Foreland Thrust Belt. The focus of this study is the Penyijingxi sag and its surrounding uplifts (Figure ). The Penyijingxi sag and its peripheral uplifts formed during the Hercynian tectonic movement, depositing on a Carboniferous volcanic rock basement and experiencing two rift cycles. The Early Permian represented a basin extension stage, followed by rapid subsidence during the Middle Permian. From the Late Permian to Middle Triassic, subsidence rates decreased, marking a foreland basin development stage. , Subsidence rates increased again during the Early to Middle Jurassic until tectonic uplift occurred in the Late Jurassic. The Permian strata, from bottom to top, comprise the Jiamuhe Formation, Fengcheng Formation, Xiazijie Formation, and the Upper and Lower Urho formations. Among these, the Fengcheng Formation and the Lower Urho Formation are two high-quality source rock units, serving as the primary sources for Permian-sourced hydrocarbons in the Penyijingxi sag. , The Fengcheng Formation source rocks contain Type I–II1 kerogen with high organic matter abundance, whereas the Lower Urho Formation source rocks contain Type II2–III kerogen with relatively lower organic matter abundance. Overpressure is widely developed in the study area. According to the overpressure classification scheme proposed by Li Wei et al., strata in the Shixi Uplift fall within the weak overpressure range, while the Dabasong area belongs to the strong overpressure range. , The study area is adjacent to hydrocarbon-generating sags such as the Mahu and Shawan sags, resulting in abundant hydrocarbon production. For instance, drill-stem testing in the Shixi-16 well, located in a structural high on the margin of the sag, was conducted in the Carboniferous interval at depths of 4723–4738 m, yielding production rates of 124.14 t/d of oil, 21,600 m3/d of gas, and 326 m3/d of water. Testing in the deeper interval of 4812–4822 m yielded 57.62 t/d of oil, 8600 m3/d of gas, and 281.2 m3/d of water. Subsequently, the deployed Shixi-18 well achieved production rates of 32.09 t/d of oil, 6000 m3/d of gas, and 287.08 m3/d of water. The Yantan-1 well in the northern part of the study area yielded 1.69 t/d of oil, 1000 m3/d of gas, and 35.95 m3/d of water (Figure ). These drill-stem test results indicate that the reservoirs in the study area are predominantly oil–water and gas–water zones. The widespread presence of water in the oil reservoirs signifies a characteristic coproduction of oil, gas, and water. Materials and Methods X-ray Diffraction (XRD) Analysis To accurately quantify the bulk mineralogy and clay mineral assemblages of the caprocks, X-ray diffraction (XRD) was performed strictly adhering to the Chinese Petroleum Industry Standard SY/T 5163-2018. The core samples were crushed, ultrasonically cleaned to remove drilling fluid contaminants, and ground to a particle size of less than 200 mesh. The powder samples were analyzed using a Bruker D8 Advance X-ray. Scanning was conducted over a range of 10°–65° at a rate of 2°/min. The relative weight percentages of bulk minerals and specific clay species (chlorite, kaolinite, Illite, and smectite) were quantified by calculating the integrated areas of their characteristic basal diffraction peaks using the Rietveld refinement method. Gas Breakthrough Pressure Testing The static macroscopic sealing capacity of the mudstone matrix was evaluated by measuring the gas breakthrough pressure according to the SY/T 5748-2020 standard. Tests were conducted using a custom-built high-pressure relative permeability system equipped with a hydrostatic core holder and Teledyne ISCO syringe pumps. Cylindrical core plugs were fully saturated with simulated formation brine and loaded into the core holder under an in situ effective confining stress to replicate subsurface geomechanical conditions. Nitrogen gas (N2) was utilized as the nonwetting injection phase to simulate natural gas charging. Gas was injected stepwise at the upstream face. To ensure fluid pressure equilibration within the ultralow permeability matrix, each pressure increment was maintained for 24 h while continuously monitoring the downstream flow using a high-precision differential pressure transducer. The breakthrough pressure was recorded at the exact threshold where continuous gas bubbles emerged and a macroscopic gas flow rate was detected at the downstream outlet. Quantitative Grain Fluorescence (QGF) Analysis A total of 105 core samples (74 reservoir rocks and 31 caprocks) were meticulously selected from key exploration wells (e.g., Pendong 1, Shixi 16, Shixi 18, and Shixi 26). To ensure the scientific rigor and comparability of the phase differentiation analyses, the caprock and reservoir samples were strictly paired in three-dimensional space. Specifically, all 31 caprock samples were acquired exclusively from the immediate overlying direct sealing intervals strictly adjacent to the sampled hydrocarbon-bearing reservoir horizons within the identical wellbores. This rigorous spatial constraint guarantees that any observed physicochemical discrepancies inherently reflect the vertical sealing efficacy of the specific caprock matrix. To reconstruct the multiphase hydrocarbon charge history and identify transient fluid leakage, quantitative grain fluorescence (QGF) and QGF on Extract (QGF-E) analyses were conducted on both reservoir and caprock samples following the established protocols by Liu and Eadington [69]. For QGF-E preparation, the residual hydrocarbons adsorbed on the framework grains were extracted using high-purity dichloromethane (DCM) solvent. The fluorescence spectra of the extracts were acquired using a Hitachi F-7000 fluorescence spectrophotometer. The excitation wavelength was fixed at 270 nm, and the emission spectra were recorded continuously from 300 to 500 nm. Key spectral parameters were extracted, including the maximum fluorescence intensity, the corresponding peak wavelength (λmax), and the total scanning fluorescence (TSF) ratio (R 1, defined as I 360 nm/I 320 nm at 270 nm excitation). These parameters were utilized to quantitatively differ entiate the hydrocarbon phase states trapped within the pore network. The QGF-E (quantitative grain fluorescence on extract) spectrum represents the fluorescence characteristics of hydrocarbons adsorbed on the surface of reservoir grains. It can be used to identify current oil layers or residual oil zones in exploration and drilling evaluation. Studies show that samples from current oil layers or residual oil zones exhibit relatively higher fluorescence intensity in their QGF-E spectra. QGF λmax refers to the wavelength corresponding to the maximum fluorescence intensity (Imax) in the Quantitative Grain Fluorescence (QGF) spectrum. Research has found that a larger maximum wavelength (λmax) indicates greater crude oil density and higher viscosity. TSF R 1 is a parameter obtained from total scanning fluorescence (TSF) analysis. It is defined as the ratio of the fluorescence intensity at an emission wavelength of 360 nm to that at 320 nm under an excitation wavelength of 270 nm. This parameter reflects the ratio of tricyclic aromatic hydrocarbons to monocyclic aromatic hydrocarbons in crude oil and can be used to characterize the maturity, density, and aromatic composition of the oil. The samples were collected from the Penyijingxi sag of in the Junggar Basin, China. The target intervals are the Lower and Upper Wuerhe Formations of the Permian system, which constitute the primary caprock distribution zone in the study area. All 31 caprock samples for this study were obtained from conventional cores of exploration wells targeting gas reservoirs (Well IDs: Pendong 1, Shixi 26, Shixi 16, Shixi 18). To ensure sample relevance, strict screening was applied, selecting core segments specifically from the direct caprock overlying proven natural gas accumulations, while intervals associated with known medium and light oil reservoirs were excluded. Results Lithological and Mineralogical Characteristics of Cap Rocks The study area is characterized by relatively poor cap rock conditions, predominantly composed of sand-mud interbeds. This is evident in the Shawan Sag, the western flank (Chepaizi-Zhongguai-Xiayan area) and eastern flank (Mosuowan-Mobei-Shixi area) of the Penyijingxi sag. Specific well data illustrate this lithology: the Zhongjia-1 well exhibits approximately 290 and 90 m thick sand-mud interbeds in the Lower Urho and Fengcheng formations, respectively; the Yantan-1 well has about 570 m of interbeds in the Upper Urho Formation; the Shixi-18 well shows roughly 190 m of interbeds in the Baikouquan and Upper Urho formations; and the Shixi-16 well contains about 140 m of interbeds in the Upper Urho Formation (Figure ). The mud-to-gross ratio of these cap rocks is approximately 50%, indicating an equal proportion of sand and mud. The direct cap rocks are widely distributed: the western part of the study area is covered by the Jiamuhe (P1j), Fengcheng (P1f), Xiazijie (P2x), and Upper Urho (P2w) formations, while the direct cap rocks in the eastern part comprise P1f, P2w, and the Lower Urho (P3w) Formation. Furthermore, the cap rock thickness in the study area is generally small, mostly less than 100 m, with individual layer thickness primarily below 10 m, indicating relatively homogeneous cap rocks. For cap rocks, clay minerals exhibit varying composition, pore structure, consolidation degree, and plasticity at different diagenetic stages. As burial depth increases, clay minerals gradually transform into Illite. During this process, the arrangement of clay minerals becomes denser and more ordered, thereby enhancing the sealing capacity of the cap rock. X-ray diffraction (XRD) analysis of the Permian cap rocks in the study area reveals their mineral composition, as shown in Figure . The results indicate that the clay mineral content in the Permian cap rocks is highly variable, ranging from 6.97% to 44.96%, with an average of 22.04%. Statistically, the cap rocks in the study area have low clay mineral content, dominated by chlorite and kaolinite, resulting in poor physical sealing capacity. The chlorite content is mostly below 30%, and the kaolinite content is predominantly under 15%. The resulting low conversion to Illite fails to enhance the physical sealing capacity of the caprock. Clay mineral content of cap rocks in the Penyijingxi sag and surrounding uplifts, Junggar Basin. Breakthrough Pressure and Overpressure Distribution Breakthrough pressure is the most direct parameter reflecting the physical sealing capacity of a cap rock; a higher value indicates stronger sealing capacity, and vice versa. Following the SY/T 5748-2020 standard method for rock gas breakthrough pressure testing, cap rock samples from the Penyijingxi sag in Junggar Basin were analyzed. The test results show a breakthrough pressure of 4 MPa for mudstone at 4655.42 m depth in the Shixi-16 well, and 2 MPa for mudstone at 4467.94 m depth in the Yanbei-2 well. The breakthrough pressure across major gas-bearing basins in China shows a wide distribution, primarily ranging between 10 and 30 MPa. According to the caprock breakthrough pressure evaluation criteria established by Zhao & Yang, the breakthrough pressure in the study area is less than 5 MPa, classifying it as a caprock with relatively poor sealing capacity. The breakthrough pressures in the study area are significantly lower than this average. In summary, the cap rocks in the Penyijingxi sag of Junggar Basin exhibit poor physical properties and low breakthrough pressures. Overpressure is widely developed in the Jurassic, Triassic, Permian, and Carboniferous strata of the study area. The pressure coefficient, defined as the ratio of the measured formation pore pressure to the hydrostatic pressure at the corresponding depth, serves as a key parameter for quantifying the magnitude of overpressure in the study area. Pressure coefficient data from wells such as Moshen-1 and Mo-23 indicate that the top of the overpressured zone is located at approximately 4300 m depth, within the middle of the J1s Sangonghe Formation. At this top boundary, the pressure coefficient increases from 1.1 to 2.0 (Figure a). Previous studies on overpressure in the deep southeastern part of the Penyijingxi sag suggest a close genetic relationship between hydrocarbon generation, particularly natural gas generation, and the formation of the deep overpressured zone. (a) Pressure coefficient that provided overpressure interface; (b) carbonate minerals increased dramatically within a short period of time; (c) formation temperature. (d) Carbonate cement in the overpressure interface; (e) carbonate cement include iron calcite, asphalt and iron dolomite. The transformation of clay minerals is one of the factors contributing to overpressure generation. This process releases ions such as Si, Ca, and Fe into pore fluids. The released elements can migrate through the system and precipitate in adjacent pores, forming secondary minerals, thus reducing the porosity and permeability of the rock. Geothermal data from the study area show that the temperature at the top of the overpressured zone ranges between 100 and 130 °C (Figure c). At this specific boundary, the carbonate mineral content increases sharply (Figure b). Microscopically, localized secondary cements such as ferrocalcite and ankerite are observed, frequently co-occurring with asphalt (Figure d,e).Given the intrinsic lack of smectite within the cap rock matrix, these iron-rich carbonate cements cannot be the product of in situ massive smectite-to-Illite transformation. Instead, they are geochemically interpreted to precipitate from the upward expulsion of highly mineralized, deep diagenetic fluids originating from the underlying mature Permian source rocks, coupled with the localized dissolution of iron-rich tuffaceous materials and volcanic rock fragments characteristic of the Urho Formation. When these upward-migrating deep fluids encounter the abrupt temperature–pressure gradient at the overpressure boundary, rapid precipitation of metal cations occurs. While this localized cementation reduces porosity at the boundary, its heterogeneous distribution confirms that it acts merely as a supplementary structural framework, rather than the standalone, basin-wide barrier responsible for the extreme overpressure. Gas Logging (C1/C2) Profiles Natural gas is universally present within the seal layers of Junggar Basin. Studies on the formation mechanism of overpressured seals in the central part of the basin indicate that the overpressure primarily results from the deep burial of Jurassic coal-measure source rocks, which induces diagenesis and gas generation. The lithology of these seals is predominantly characterized by interbedded thick sandstones and mudstones, with intercalations of siltstone. Gas logging, as primary data acquired during drilling, directly reflects the hydrocarbon-bearing properties of formations. Analyzing the variation characteristics of the light-to-heavy component ratio in gas logging data from seal layers allows for inferences regarding the dominant mode of natural gas migration. The differentiation of migration modes relies on the distinct mass transfer kinetics of gas components. Gas diffusion through low-permeability caprocks is driven by concentration gradients according to Fick’s laws. Because methane (C1) possesses a smaller molecular kinetic diameter and higher diffusion coefficient than ethane (C2), significant compositional fractionation occurs along the migration pathway. Consequently, diffusion is characterized by a monotonic, progressive increase in the C1/C2 ratio upward through the strata. Conversely, leakage implies bulk advection driven by pressure gradients through macroscopic pathways such as microfractures. This mass transport occurs rapidly with minimal component fractionation, resulting in a relatively low and stable C1/C2 ratio without a clear vertical gradient. The criteria linking these physicochemical mechanisms to gas logging ratios are well-established in hydrocarbon migration evaluations. Investigation of the C1/C2 variation characteristics within the Permian seal layers of typical wells in the study area provides further insights. Applying these criteria to the Permian seals in the study area: The C1/C2 profile of the Pendong 1 well (Figure a) predominantly exhibits vertical stability with baseline fluctuations (4900–5130 m, 5150–5180 m), indicating bulk advection without fractionation; thus, leakage is the dominant mechanism. Localized intervals (5130–5150 m) display sharp, monotonic increases in C1/C2, reflecting typical diffusion. This suggests significant natural gas escape from the paleo-oil reservoir. The C1/C2 trend in the Moshen 1 well seal (Figure b) also indicates dominant leakage (e.g., interval 6450–6750 m) and localized diffusion (e.g., intervals 6400–6450 m, 6750–6800 m), similarly pointing to gas escape from the paleo-reservoir into the seal. For the Shixi 18 well, which has gas production underlying the seal, the C1/C2 trend within the seal (Figure c) suggests that gas migrated into the seal primarily by diffusion, with no significant escape from the paleo-oil reservoir. C1/C2 gas logging ratio curves in the cap rock of the Penyijingxi sag and surrounding uplifts, Junggar Basin.Light yellow blocks indicate stable baseline fluctuations characteristic of advective bulk leakage, whereas light blue blocks highlight sharp, monotonic increases corresponding to concentration-driven diffusion. Hydrocarbon Phase Differentiation Liu et al. introduced Quantitative Fluorescence Techniques to rapidly detect fluorescence spectral characteristics and intensity information from oil inclusions within grains and adsorbed hydrocarbons on grain surfaces. Grain fluorescence analysis was conducted on six wells, including Pendong 1 and Shixi 26, in the study area. The statistical results reveal (Figure ) that multiple hydrocarbon phases, including medium oil, light oil, and condensate, were charged into the reservoirs of wells such as Pendong 1. Light oil and condensate show high frequency of occurrence, while medium oil is relatively rare. Similarly, fluorescence characteristics of medium oil, light oil, and condensate are also identified within the cap rocks. However, the proportion of medium and light oil signals increases significantly in the cap rocks, whereas the frequency of condensate fluorescence decreases markedly. This indicates that condensate is scarcely observed in the seal layers. Frequency distribution of QGF-E characteristics of oil in reservoir and cap rock of the Penyijingxi sag and surrounding uplifts, Junggar Basin. Cap rocks serve as a critical barrier against the upward leakage of oil and gas, and their physical properties fundamentally determine the preservation or dissipation of hydrocarbons. Quantitative fluorescence analysis of cap rocks and reservoirs in the study area reveals distinct differences in hydrocarbon phases. The reservoirs exhibit fluorescence characteristics of light oil, medium oil, and condensate, whereas the cap rocks show signatures of light oil, medium oil, and only minor amounts of condensate. This indicates that light and medium oils from the reservoir leaked upward through the cap rock, while condensate oil and gas were effectively preserved. Discussion Insufficiency of Static Sealing Mechanisms The integration of petrophysical testing and mineralogical data delineates a pronounced geological paradox in the Penyijingxi sag: the cap rocks are characterized by high-frequency sand-mud interbeds with a remarkably low static gas breakthrough pressure (2.0–4.0 MPa). Physically, this intrinsic macroscopic threshold is severely insufficient to retain the massive underlying condensate reservoirs and the extreme fluid overpressure anomaly. Attributing this hydrocarbon accumulation to conventional static mechanisms, such as diagenetic densification, contradicts the empirical evidence. Although localized carbonate cementation occurs near the top boundary of the overpressured zone, the regional petrophysical densification remains fundamentally limited. The total clay mineral content averages only 22.04% and is dominated by chlorite and kaolinite, lacking the requisite smectite-to-Illite transformation volume to form a basin-wide ductile physical barrier. Furthermore, the maintenance of intense overpressure over geological time scales strictly requires a robust, pre-existing hydraulic barrier. Consequently, the overpressure observed in the Junggar Basin is the manifestation of successful fluid entrapment, rather than the primary cause of the initial seal. This structural contradiction mandates a dynamically activated enhancement mechanism linked directly to evolving fluid phase behaviors. Kinematic Evolution of the Gas Capillary Seal To validate the chronological rationality of this dynamically activated mechanism, the hydrocarbon generation and expulsion history of the primary Permian source rocks (e.g., the Fengcheng and Lower Urho formations) must be superimposed onto the basin’s burial and thermal evolution timeline. Based on this dynamic matching and the distinct hydrocarbon phase differentiation recorded by QGF-E data. During the early to-middle burial stages (Triassic to Early Jurassic), the Permian source rocks entered the conventional oil window, primarily generating and expelling medium and light liquid hydrocarbons. Confronted with the inherently suboptimal static physical properties of the thin sand-mud interbedded cap rocks, these early liquid phases migrated upward and experienced extensive vertical leakage. This early dissipation event is unequivocally captured by the pervasive medium and light oil QGF signatures permanently trapped within the cap rock matrix (Figure ). At this stage along the geological timeline, the static physical barrier completely failed to constrain fluid migration. Subsequently, continuous profound subsidence from the Late Jurassic through the Cretaceous pushed the source rocks into the highly mature and overmature stages. This thermal evolution triggered a massive, late-stage generation of natural gas and condensate. As this nonwetting free gas phase invaded the overlying interbedded strata, driven by buoyancy, it preferentially accumulated in the larger pore networks of the coarse intervals, forming discrete “gas caps” at multiple lithological interfaces. Crucially, the formation of deep overpressure in the Junggar Basin predominantly coincided with this late-stage massive gas generation. The temporal alignment of gas charging and overpressure build-up confirms that the capillary barrier was dynamically activated exactly when the extreme hydraulic seal was required. Following the Young–Laplace principles, the capillary resistance across these newly established gas–water interfaces underwent a strict linear superposition. This dynamic process exponentially amplified the macroscopic sealing capacity, creating a composite barrier that successfully trapped the contemporaneous overpressure and halted any subsequent fluid leakage. Limitations and Future Work While this study demonstrates the existence and dominant role of the natural gas capillary sealing mechanism in the Junggar Basin through petrophysical, logging, and geochemical (QGF) evidence, a quantitative mathematical closure remains a limitation. Accurately calculating the cumulative capillary sealing pressure (δP c) requires precise in situ parameters, specifically the pore-throat radii of the deep, highly compacted mudstones (r f) and sandstones (r c), as well as the gas–water interfacial tension (σ) under extreme reservoir temperatures and pressures. Obtaining these parameters necessitates specialized high-pressure and high-temperature physical simulations, which were beyond the scope of the current phase of this study. Consequently, calculating the exact proportional contribution of this mechanism to the overpressure anomaly is currently unfeasible without relying on overly generalized assumptions. Future research will focus on acquiring these specific core and fluid parameters to quantitatively model the cumulative capillary sealing capacity and mathematically close the loop on this dynamic sealing model. Conclusions The natural gas capillary sealing theory has played a significant role in guiding practical exploration and development. To elucidate its importance, this paper analyzes multiple case studies and arrives at the following conclusions:1.Both theoretical and experimental studies demonstrate that natural gas can enhance the sealing capacity of cap rocks, forming a novel capillary sealing mechanismnamely, natural gas capillary sealingwhose sealing capacity is additive. The essential prerequisites for this mechanism are first, the cap rock must consist of sand-mud interbeds; second, overpressure must be present; and finally, free gas must exist within the interbedded cap rock.2.A pronounced geological paradox exists within the Permian caprocks surrounding the Penyijingxi sag: the caprocks are macroscopically characterized by high-frequency sand-mud interbeds with severely deficient static breakthrough pressures (merely 2.0–4.0 MPa) and a fundamental lack of ductile clay minerals. Conventional static diagenetic sealing mechanisms fundamentally fail to explain the effective containment of the underlying massive condensate reservoirs and the extreme deep-basin overpressure.3.The dynamic activation of the natural gas capillary sealing mechanism acts as the decisive hydrodynamic switch resolving this paradox. Quantitative grain fluorescence on extract (QGF-E) data provide independent temporal evidence for this dynamic process: the pervasive preservation of early liquid oil signatures within the caprock matrix, contrasted with the abrupt exclusion of late-stage condensate, explicitly confirms an “early liquid dissipation, late gas sealing” kinematic evolution.4.The late-stage massive natural gas charge, synchronous with the deep-basin overpressure build-up, served as the dynamic trigger. The accumulation of the nonwetting free gas phase at multiple coarse-fine lithological interfaces formed transient “gas caps”, triggering the linear superposition of capillary resistance. This dynamic process exponentially amplified the macroscopic sealing capacity of the inherently poor caprocks to tens of MPa, perfectly matching the magnitude of the measured extreme overpressure and physically halting subsequent fluid leakage. 1.Both theoretical and experimental studies demonstrate that natural gas can enhance the sealing capacity of cap rocks, forming a novel capillary sealing mechanismnamely, natural gas capillary sealingwhose sealing capacity is additive. The essential prerequisites for this mechanism are first, the cap rock must consist of sand-mud interbeds; second, overpressure must be present; and finally, free gas must exist within the interbedded cap rock. Both theoretical and experimental studies demonstrate that natural gas can enhance the sealing capacity of cap rocks, forming a novel capillary sealing mechanismnamely, natural gas capillary sealingwhose sealing capacity is additive. The essential prerequisites for this mechanism are first, the cap rock must consist of sand-mud interbeds; second, overpressure must be present; and finally, free gas must exist within the interbedded cap rock. 2.A pronounced geological paradox exists within the Permian caprocks surrounding the Penyijingxi sag: the caprocks are macroscopically characterized by high-frequency sand-mud interbeds with severely deficient static breakthrough pressures (merely 2.0–4.0 MPa) and a fundamental lack of ductile clay minerals. Conventional static diagenetic sealing mechanisms fundamentally fail to explain the effective containment of the underlying massive condensate reservoirs and the extreme deep-basin overpressure. A pronounced geological paradox exists within the Permian caprocks surrounding the Penyijingxi sag: the caprocks are macroscopically characterized by high-frequency sand-mud interbeds with severely deficient static breakthrough pressures (merely 2.0–4.0 MPa) and a fundamental lack of ductile clay minerals. Conventional static diagenetic sealing mechanisms fundamentally fail to explain the effective containment of the underlying massive condensate reservoirs and the extreme deep-basin overpressure. 3.The dynamic activation of the natural gas capillary sealing mechanism acts as the decisive hydrodynamic switch resolving this paradox. Quantitative grain fluorescence on extract (QGF-E) data provide independent temporal evidence for this dynamic process: the pervasive preservation of early liquid oil signatures within the caprock matrix, contrasted with the abrupt exclusion of late-stage condensate, explicitly confirms an “early liquid dissipation, late gas sealing” kinematic evolution. The dynamic activation of the natural gas capillary sealing mechanism acts as the decisive hydrodynamic switch resolving this paradox. Quantitative grain fluorescence on extract (QGF-E) data provide independent temporal evidence for this dynamic process: the pervasive preservation of early liquid oil signatures within the caprock matrix, contrasted with the abrupt exclusion of late-stage condensate, explicitly confirms an “early liquid dissipation, late gas sealing” kinematic evolution. 4.The late-stage massive natural gas charge, synchronous with the deep-basin overpressure build-up, served as the dynamic trigger. The accumulation of the nonwetting free gas phase at multiple coarse-fine lithological interfaces formed transient “gas caps”, triggering the linear superposition of capillary resistance. This dynamic process exponentially amplified the macroscopic sealing capacity of the inherently poor caprocks to tens of MPa, perfectly matching the magnitude of the measured extreme overpressure and physically halting subsequent fluid leakage. The late-stage massive natural gas charge, synchronous with the deep-basin overpressure build-up, served as the dynamic trigger. The accumulation of the nonwetting free gas phase at multiple coarse-fine lithological interfaces formed transient “gas caps”, triggering the linear superposition of capillary resistance. This dynamic process exponentially amplified the macroscopic sealing capacity of the inherently poor caprocks to tens of MPa, perfectly matching the magnitude of the measured extreme overpressure and physically halting subsequent fluid leakage.