Rock Pore Structure Research Articles

Study on the pore structure and permeability evolution of tight sandstone under liquid nitrogen freezing‐thawing cycles based on NMR technology

To further raise the gas extraction efficiency of the tight sandstone, the liquid nitrogen (LN2) freezing-thawing cycles method can be employed to improve the permeability of the low-permeability reservoirs. Permeability is generally regarded as a macroscopic description of the pore structure and usually has functional relationship to pore structure properties. The permeability of the rock is closely related to the change of microscopic pore structure. The permeability of rock depends on how the subzero temperatures changed the microscopic pore structure of rock, but it has not yet been confirmed obviously. In this study, the nuclear magnetic resonance (NMR) technique was adopted to investigate the pore structure evolution law and permeability of the tight sandstone with different LN2 freezing-thawing cycles. The results demonstrate that the LN2 freezing-thawing cycles promotes pore development and micro-fracture connection, and enhances the pore connectivity. The proportion of meso-pores, macro-pores and micro-fractures in the sandstone samples increases significantly, which provides a channel for the sandstone gas flow and extraction. Total porosity and effective porosity of the samples present a trend of rapid increase as the number of LN2 freezing-thawing cycles increasing, while the residual porosity decreases as the number of LN2 freezing-thawing cycles increasing. Coates model, SDR model (mean T2 model) and PP model were used to calculate and evaluate the permeability of the samples subjected to different LN2 freezing-thawing cycles. Furthermore, PP model can provide a better permeability estimate than the classical Coates and SDR model.

Exploring the Impact of Mineralogy on Pore Structure and Fluid Dynamics in Tight Reservoir Rocks: Insights for Enhanced Oil Recovery and Gas Storage

Exploring the Impact of Mineralogy on Pore Structure and Fluid Dynamics in Tight Reservoir Rocks: Insights for Enhanced Oil Recovery and Gas Storage

Experimental study on the influence of rock pore structure on pressure stimulated voltage variations based on nuclear magnetic resonance

Since rocks will generate voltage under load, studying their voltage characteristics is of prime importance for the prevention and control of mine dynamic disasters and the corresponding secondary disasters. In this study, a pressure stimulated voltage (PSV) test system for rock materials under uniaxial compression was constructed to explore the law of PSV variations of rocks. Meanwhile, a nuclear magnetic resonance test system was employed for investigating the influence mechanism of pore structure changes on PSV variations. The following beneficial results were obtained. A “double-peak” phenomenon is observed on the PSV curves of granite and sandstone, whereas, for marble, the phenomenon only appears under high loading rates. The T2 spectra of different types of rock differ greatly. After granite fractures under load, some primary micro-pores are converted into meso-pores and macro-pores, accompanied by the generation of substantial new micro-pores. These micro-pores activate more rock defects (dislocation and grain boundary), resulting in a higher average PSV and peak PSV. After marble fractures, numerous primary micro-pores are transformed into meso-pores and macro-pores, and the proportion of new micro-pores falls. Consequently, its electricity generation capacity weakens. In contrast, sandstone contains a higher proportion of micro-pores. After it fractures, despite the conversion of some micro-pores into meso-pores and macro-pores, abundant micro-pores are generated again, bringing about a relatively high voltage. In short, the changes in overall porosity cannot represent the electricity generation capacity of rock, and the changes in bound cracks exert a profoundly influence on it. The key to the electricity generation capacity of rock lies in the increase and connection of micro-cracks.

Effects of rock pore and micromorphology on electromagnetic radiation characteristics

Electromagnetic radiation (EMR) is a crucial tool for monitoring and early warning of underground engineering disasters. Investigating the inherent pore characteristics of rocks is essential for a comprehensive understanding of the EMR phenomenon. The EMR was monitored during various types of rock splitting failures. The pore structure and micromorphology of rocks are studied using quantitative methods such as mercury intrusion porosimetry, fractal analysis, and the Gray Level Co-occurrence Matrix (GLCM). Results indicate that the fractal dimension of red sandstone is significantly lower than the other three rocks. The fractal complexity increases sequentially from red sandstone to marble, granite, and limestone. As the fractal dimension decreases, the signal waveform characteristics of the four rocks become more complex before the main fracture, with a significant increase in signals during the compaction and elasticity stages. Higher fractal dimensions lead to a shift in energy and count from elasticity stage to the post-peak stage. The main fracture amplitudes of the four rocks generally exhibited a consistent pattern, following the sequence of granite > marble > limestone > red sandstone. The main fracture amplitude decreases with increasing complexity of the rock's pore micromorphology. Rock pore characteristics affect frequency domain characteristics by influencing rock strength and crack expansion. An increase in the average pore diameter tends to decrease both the main and center frequencies.

Numerical simulation to influence of fluid distribution on acoustic attenuation coefficient in gas water two-phase fractured media

Abstract The equivalent fluid model simplifies the actual distribution state of fluid, and the relationship between rock elastic properties and pore fluid parameters established by the equivalent fluid model usually deviates greatly from the experimental results. A random discrete fracture model is established, and the gas-water two-phase distribution in fracture pores is described in detail. Based on the acoustic wave theory, the finite difference numerical simulation of the acoustic wave field of the rock sample model is carried out. The variation law of acoustic attenuation coefficient with saturation under the different pore fluid distribution modes is analyzed. It is of great significance to the prediction of rock pore structure and the identification of reservoir fluid.

Mineral composition, pore structure and mechanical properties of coal measure strata rocks: A case study of Pingdingshan Coalfield

The microstructure and mechanical properties of coal measure rocks (CMR) are critical factors in the successful geological storage of CO2. While previous research has predominantly focused on coal seams, the overlying and underlying bedrock strata are equally significant yet often overlooked. This study addresses this gap by selecting coal and adjacent strata from the Twelve Mine as representative samples. A comprehensive suite of analyses was conducted, including thin section identification, X-ray diffraction (XRD), nuclear magnetic resonance (NMR), uniaxial compression, and acoustic emission tests. These analyses were used to determine the mineralogical composition and pore structure characteristics of the CMR, elucidate the mechanical failure mechanisms of different CMR types, and explore the relationships between rock strength, mineral composition, and pore characteristics. The findings indicate that the primary minerals in coal are clay minerals, while mudstone is composed predominantly of clay minerals and quartz, sandstone is rich in quartz and feldspar, and limestone contains significant amounts of calcite. The porosity of CMR varies between 1.73 % and 10.12 %, with 82.72 % of the pores being micro- to mesopores and the remaining 17.28 % being macropores. The pore structures exhibit fractal characteristics, with the fractal dimension of mesopores ranging from 2.581 to 2.902, and that of macropores from 2.968 to 2.997. Coal predominantly fails through a combination of tensile and shear failure mechanisms, mudstone and limestone through tensile failure, and sandstone through shear failure. Furthermore, the results suggest that rock strength is positively correlated with quartz content and negatively correlated with kaolinite content and porosity. In terms of their influence on coal rock strength, the factors in descending order of impact are kaolinite, quartz, calcite, microporosity, macroporosity, dolomite, and mesoporosity. Thus, while the mechanical properties of these rocks are primarily governed by their mineralogical composition, the pore structure also plays a significant role.

Characterization of radon gas transmission rate in pore structures of different rock layers by radon daughters inversion calculation

Determining the transmission rate of radon gas in overburden strata is crucial for conducting a comprehensive study of radon gas's longitudinal and long-distance migration mechanisms. This study investigates the mineral components of rocks in the underground strata of the mining area using the X-ray diffraction method. Additionally, it examines the pore structure parameters of the rocks at different depths using the low-temperature nitrogen adsorption method. This research introduces an approach to inversion calculate the radon gas transmission rate through the activity ratio of radon's characteristic daughters based on the decay law and activity balance of 210Po and 210Pb daughters. In addition, it determines the transmission rates of radon gas in overlying strata at various depths through this method. The relationship between the rock's mineral composition and pore structure is investigated, and the effects of pore structure and mineral composition on the radon gas transmission rate are analyzed. The findings indicated that the pore structure exerts a dual impact on radon gas transport: macropores serve as channels for upward radon gas transport, while micropores offer most of the adsorption area. In contrast, the radon gas transmission rate is indirectly influenced by the mineral composition content associated with the medium's adsorption capacity and pore structure. In the studied lithologies, an increase in quartz content promotes radon gas transmission, while an increase in clay mineral content impedes it. Finally, the mechanisms of radon gas transport, daughter adsorption, and the impacts of rock pore structure and mineral composition on the radon transmission rate are discussed.

Evolution of Pore Structure and Fractal Characteristics in Red Sandstone under Cyclic Impact Loading

Fatigue damage can occur in surface rock engineering due to various factors, including earthquakes, blasting, and impacts. The underlying cause for the variations in physical and mechanical properties of the rock resulting from impact loading is the alteration in the internal pore structure. To investigate the evolution characteristics of the pore structure under impact fatigue damage, red sandstone subjected to cyclic impact compression by split Hopkinson pressure bar (SHPB) was analyzed using nuclear magnetic resonance (NMR) technology. The parameters describing the evolution of pore structure were obtained and quantified using fractal methods. The development of the pore structure in rocks subjected to cyclic impact was quantitatively analyzed, and two fractal evolution models based on pore size and pore connectivity were constructed. The results indicate that with an increasing number of impact loading cycles, the porosity of the red sandstone gradually increases, the T2 cutoff (T2c) value decreases, the most probable gray value of magnetic resonance imaging (MRI) increases, the pores’ connectivity is enhanced, and the fractal dimension decreases gradually. Moreover, the pore distribution space tends to transition from three-dimensional to two-dimensional, suggesting the expansion of dominant pores into clusters, forming microfractures or even macroscopic fissures. The findings provide valuable insights into the impact fatigue characteristics of rocks from a microscopic perspective and contribute to the evaluation of time-varying stability and the assessment of progressive damage in rock engineering.

Numerical simulation of residual oil distribution characteristic of carbonate reservoir after water flooding

Carbonate reservoirs are characterized by abundant reserves and are currently focal points for development in oil and gas producing regions such as the Ahdab oilfield, Tarim Basin, Sichuan Basin, and Ordos Basin. The primary method for exploiting carbonate reservoirs is waterflooding. However, due to the complex pore structure and pronounced heterogeneity of carbonate rocks, the waterflooding process often leads to an unclear distribution of remaining oil and low waterflooding recovery efficiency, significantly impacting the stable and high production of carbonate reservoirs. This paper presents a two-phase flow model of oil and water in distinct pore structures by integrating fluid flow equations and interface tracking equations. It visually represents the waterflooding process at the pore scale, elucidates the distribution and formation mechanism of remaining oil, and discusses the mechanism of microscopic displacement efficiency change. The study reveals that: 1) After waterflooding, the distribution patterns of remaining oil can be categorized into dead-end remaining oil, pressure balance remaining oil, wall-bound remaining oil, Jamin effect remaining oil, and water-encapsulating remaining oil, which are governed by microscopic pore structure, wettability, and preferential flow paths; 2) From the perspective of actual reservoir displacement efficiency, intergranular pores > intergranular dissolved pores > visceral foramen > mould pore, with this trend being more pronounced under hydrophilic wetting conditions; 3) Given the oil-wet to strong oil-wet wettability characteristics of these carbonate rocks, capillary forces pose significant resistance during waterflooding. The conclusion underscores the importance of leveraging the reservoir’s microscopic pore structure and wettability characteristics for actual oil wells, elucidating the evolutionary law of the mechanical mechanism of oil-water interface advancement, clarifying oil-water percolation characteristics at the pore scale, and understanding the microscopic displacement physical mechanism, all of which are crucial for guiding the design of schemes aimed at enhancing reservoir recovery efficiency.

Control of Physical Properties by the Matching between Rock Components and Pore Structure in Shale Reservoirs of the Permian Lucaogou Formation, Jimusar Sag

Control of Physical Properties by the Matching between Rock Components and Pore Structure in Shale Reservoirs of the Permian Lucaogou Formation, Jimusar Sag

Experimental study on the dynamic evolution of rock pore structure in real-time monitoring based on nuclear magnetic resonance

The rock structure determines the stability and reliability of underground engineering. Great achievements have been made in the study of rock structural damage, but there are few studies on real-time monitoring of rock structural changes under different increasing and unloading decreasing pressures. In this paper, the dynamic evolution rule of pore structure in rock under the combined action of water pressure and different confining pressure of increasing and decreasing is monitored by NMR technique. The porosity and permeability of rock under different confining pressures are consistent. The increase of confining pressure leads to the closure of pores, which leads to the decrease of porosity and permeability, but the amplitude and time of change are different. During the process of decreasing confining pressure, the T2 spectral curve, porosity and permeability of the rock did not fully recover due to the reduction of confining pressure, indicating that plastic deformation occurs inside the rock at this time. The mathematical relationship between porosity and permeability is established, and the permeability of rock is not only related to the size of porosity, but also closely related to the pore structure of rock.

Modeling sub‐resolution porosity of a heterogeneous carbonate rock sample

AbstractAccurately estimating the petrophysical properties of heterogeneous carbonate rocks across various scales poses significant challenges, particularly within the context of water and hydrocarbon reservoir studies. Digital rock analysis techniques, such as X‐ray computed microtomography and synchrotron‐light‐based imaging, are increasingly employed to study the complex pore structure of carbonate rocks. However, several technical limitations remain, notably the need to balance the volume of interest with the maximum achievable resolution, which is influenced by geometric properties of the source–detector distance in each apparatus. Typically, higher resolutions necessitate smaller sample volumes, leading to a portion of the pore structure (the sub‐resolution or unresolved porosity), that remain undetected. In this study, X‐ray microtomography is used to infer the fluid flow properties of a carbonate rock sample having a substantial fraction of porosity below the imaging resolution. The existence of unresolved porosity is verified by comparisons with nuclear magnetic resonance (NMR) data. We introduce a methodology for modeling the sub‐resolution pore structure within images by accounting for unresolved pore bodies and pore throats derived from a predetermined distribution of pore throat radii. The process identifies preferential pathways between visible pores using the shortest distance and establishes connections between these pores by allocating pore bodies and throats along these paths, while ensuring compatibility with the NMR measurements. Single‐phase flow simulations are conducted on the full volume of a selected heterogeneous rock sample by using the developed pore network model. Results are then compared with petrophysical data obtained from laboratory measurements.

Evolution of pore structure and analysis of freeze damage in granite during cyclic freeze-thaw using NMR technique

Understanding the changes in a rock's pore structure and unfrozen water content during the freeze-thaw (F-T) process is crucial for unraveling the F-T damage mechanism of rock masses. This study employs nuclear magnetic resonance (NMR) tests to resolve the pore structure characteristics of granite specimens subjected to varying numbers of F-T cycles (0, 50, 100, and 200 cycles). Throughout the F-T process, unfrozen water (comprising bound water and free water) is quantified through T2 distribution curves, and the freeze-induced strain at different subzero temperatures is investigated. The results reveal that F-T cycles lead to the expansion of pore sizes from small to large, reducing the complexity of the pore structure. Moreover, an increased proportion of meso- and macro-pores accelerates the rate of water-ice phase transition in granite specimens, resulting in intensified freeze damage and greater maximum freeze-induced strain. Additionally, when previously damaged granite is refrozen, the presence of free water induces higher levels of freeze-induced strain. This research comprehensively analyzes the F-T damage process in granite, laying a crucial theoretical foundation for understanding freeze damage mechanisms in rock engineering.

Complexity upgrade and triggering mechanism of mixed-wettability: Comparative study of CO2 displacement in different phases

Complex pore structures and wettability affect CO2 storage in underground reservoirs. Revealing the flow law of CO2 in complex reservoirs is crucial for evaluating storage stability and security. Considering the complexity of the pore structure and wettability of rock, a complex mixed-wettability characterization method based on a complex pore structure is proposed. Matrix numerical models for two-dimensional heterogeneous complex mixed-wettability rock were developed to simulate CO2 displacement in different phases. A comparative study found that, under single-phase wettability, the weaker the wettability, the better the effect of gaseous CO2 displacement. Moreover, the stronger the wettability, the better the displacement effect of ScCO2. The wetting phase of CO2 was transformed from the non-wetting phase. At the late stage of gaseous CO2 displacement, when the contact angles were 30° and 60°, the displaced water phase was absorbed back into the matrix owing to imbibition, resulting in the formation of a peak in the displacement efficiency curve. Compared to that at a contact angle of 60°, the displacement efficiency at a contact angle of 30° decreased more significantly after the peaks, while the suck-back phenomenon did not occur when the contact angle was 90°. Under mixed-wettability conditions, the flow law of CO2 in different phases is directly affected by the contact angle of the pore walls. The displacement efficiencies of gaseous CO2 and ScCO2 increased by 9.58% and 17.96%, respectively, owing to the dual driving force caused by imbibition. The complexity of mixed-wettability is upgraded by providing pore walls with different proportions of contact angles (six types). The flow law of different phases of CO2 showed that, the higher the proportion of walls with small contact angles (3:2:1), the worse the displacement effect of gaseous CO2 and the better the displacement effect of ScCO2. Compared with single-phase wettability and mixed-wettability, the displacement efficiency of gaseous CO2 and ScCO2 under complex mixed-wettability decreased by 4.26%, 1.00% and 20.41%, 3.37%, respectively. Therefore, ignoring complex mixed-wettability often leads to a higher displacement efficiency.

Metal intrusion enhanced deep learning-based high temperature deterioration analysis of rock materials

The deep learning-based method demonstrates superior capability in high-temperature deterioration analysis of rock materials through scanning electron microscopy (SEM) characterization. However, the blurred boundaries between rock particles and pore structure presented in SEM images always affect the training efficiency. Hence, in this study, the metal intrusion technology and backscattered electron (BSE) observations are applied to assist a deep learning-based model to analyze the deterioration of the rock materials exposed to different levels of high temperatures. >2000 micro images for each level were inputted and trained to distinguish the features of rock after 5 temperature levels deterioration. The results reveal that the classification accuracy on the heat deterioration of the rock materials achieved 94.3%. The classification accuracy under the optimized observation area (from 240 pixels × 240 pixels to 280 pixels × 280 pixels) is stable over 90%, balancing the high precision and training efficiency. Moreover, the interpretable high-temperature degradation characteristics of the micro-damage on rock particles and pore regions can be extracted, enabling further analysis of the degradation process of high-temperature treated rocks. We expect this work will inspire a high-precision, wide-application and cost-efficient method for the deterioration analysis of rock materials in underground projects.

Effect of pore structure characteristics on gas-water seepage behaviour in deep carbonate gas reservoirs

The study of gas-water two-phase flow behavior has always been a focus and a difficult problem in the field of oil and gas seepage, and it is related to the formulation of reasonable development plans for oil and gas reservoirs. Deep carbonate gas reservoirs (depth>4500m) develop cross-scale porous media, which have extremely strong heterogeneity and very complex seepage characteristics. To the best of our knowledge, there has been no systematic study on the two-phase flow patterns of gas and water in deep carbonate gas reservoirs. This paper uses cast thin sections, scanning electron microscopy (SEM) and nuclear magnetic resonance (NMR) to study the influence of carbonate rock pore structure characteristics on gas storage and seepage capabilities. The results show that the existence of cavities enhances the gas storage capacity of the reservoir and improves the seepage capacity of pore-throat space near the cavity. Fractures can effectively communicate with isolated cavities, enhance reservoir connectivity, and thereby significantly improve the overall seepage capacity of the reservoir. The centrifugal technique was combined with NMR to study the movable fluid characteristics of carbonate reservoirs. The critical flow throat radius and movable fluid saturation of the reservoir reveal that the degree of development of cavities and fractures determines the mobility of the fluid in the reservoir. The fluid mobility in the fracture-cavity-type reservoir is the strongest, while the fluid mobility in the pore-type reservoir is the weakest. Analysis of the normalized curve of gas-water relative permeability under high temperature and pressure reveals that formation water preferentially enters connected pores with low seepage resistance, and the strong storage capacity of cavities can effectively delay the non-uniform intrusion of formation water. The results of water invasion simulation experiments show that the water invasion phenomenon can improve the gas production rate and recovery rate of pore-type reservoirs. However, the formation water flow and channeling phenomena caused by reservoir heterogeneity will cause a large amount of gas to be sealed in the pore space of cavity-type and fracture-cavity-type reservoirs, reducing the recovery rate and economic benefits of these reservoirs. These findings further provide solid fundamentals for efficient and effective exploitation of deep carbonate gas reservoirs.

Deterioration of volcanic tuffs from rock dwellings in Brhlovce (Slovakia) induced by freeze-thaw cycling studied by non-destructive tests and µCT visualization

The aim of this paper is to evaluate the effect of freeze-thaw-induced weathering of volcanic tuffs from the rock dwellings in the village of Brhlovce (Slovakia) based on a non-destructive and experimental study of tuff pore space properties. Traditional techniques for characterization of the porosity and pore size distribution and new experimental techniques like spontaneous imbibition (pore interconnection) or indicative rock pore structure method (pore size distribution) were compared with pore network models acquired by X-ray computed microtomography before and after freeze-thaw cycling in a custom-built thermodilatometer VLAP04. The results offer insight into how frost weathering altered the pore network of volcanic tuffs. Brhlovce tuffs are highly susceptible to frost damage due to their high porosity, pore interconnectivity, and bimodal pore size distribution pattern, where nanometric-sized micropores predominate over larger micrometric-sized capillary pores. Poromechanical principles were applied to confirm that ice crystallization initiates in well-connected micropores with a corresponding crystallization pressure below 10 MPa. After undergoing 100 freeze-thaw cycles with temperature oscillations ranging from − 10 to 10 °C, there was a significant increase in total porosity, pore interconnectivity, and rock permeability. Additionally, the pore size distribution was altered. These parameters influence the transport of fluids and the moisture regime in rock masses, which can serve as a precursor to physical and chemical weathering processes

A computationally efficient modeling of flow in complex porous media by coupling multiscale digital rock physics and deep learning: Improving the tradeoff between resolution and field-of-view

Digital rock physics is at the forefront of characterizing porous media, leveraging advanced tomographic imaging and numerical simulations to extract key rock properties like permeability. However, fully capturing the heterogeneity of natural rocks necessitates imaging increasingly larger sample volumes, presenting a significant challenge. Direct numerical simulations at these scales become either prohibitively expensive or computationally unfeasible due to limitations in resolution and field of view (FOV). This issue is particularly pronounced in carbonate rocks, known for their complex, multiscale pore structures, which exacerbate the resolution-FOV tradeoff. To address this, we introduce a machine learning strategy that merges multiscale imaging data from various resolutions with a 3D convolutional neural network (CNN) model. This approach is innovative in its ability to identify cross-scale correlations, thereby enabling the estimation of transport properties in larger volumes—properties that are difficult to simulate directly—using trainable proxies. The integration of multiscale imaging with deep learning allows for accurate permeability predictions at scales beyond those feasible with traditional direct simulation methods. By employing transfer learning across different scales during the training phase, our multiscale machine learning model achieves robust performance, with an R² exceeding 0.96 when evaluated on diverse lower-resolution domains with larger FOVs. Notably, this method significantly enhances computational efficiency, reducing the computational time by orders of magnitude. Originally developed for the intricate pore structures of carbonate rocks, our approach shows promise for application to a wide range of multiscale porous media, offering a viable solution to the longstanding tradeoff between imaging resolution and FOV in digital rock physics.

Utilizing Differences in Mercury Injection Capillary Pressure and Nuclear Magnetic Resonance Pore Size Distributions for Enhanced Rock Quality Evaluation: A Winland-Style Approach with Physical Meaning

Pore structure is a fundamental parameter in determining the hydrocarbon storage capacity and flow characteristics of a reservoir. Mercury injection capillary pressure (MICP) and nuclear magnetic resonance (NMR) are two commonly utilized techniques for characterizing rock pore structures. However, current studies indicate that disparities in testing methodologies due to distinct physical characteristics lead to a partial misalignment in pore size distributions. We conducted MICP (dynamic) and NMR (static) experiments on eight tight sandstone and eight shale samples and proposed a method to utilize information from the differences in MICP and NMR pore size distributions, aiming to enhance the accuracy of rock quality analysis. We observed that in rock cores where large pores are interconnected with smaller pore throats, MICP tends to overestimate the proportion of these smaller pores and underestimate the larger ones. Furthermore, we integrated information from both dynamic and static experimental processes based on physical significance and found that the fitting accuracy of the newly proposed method is superior to the Winland r35 equation. Compared to the Winland r35 equation, our new method significantly improves fitting accuracy, increasing the R-squared value from 0.46 to 0.93 in sandstones and from 0.80 to 0.87 in shales. This represents a potential high-precision, comprehensive tool for rock quality analysis, offering a new perspective for an in-depth understanding of rock properties.

Molecular Insights of Excessive Water Cut during Cyclic Gas Injection in Liquid-Rich Shale Reservoirs: Contributions of Hydrocarbon Condensation and Water Trapping

Summary In a recent pilot test of cyclic gas injection (huff ‘n’ puff) in a Permian shale reservoir, excessive water product was observed, the reason for which remains unclear. In this work, we analyze the mechanisms of gas huff ‘n’ puff processes using molecular dynamics (MD) simulations and explain the reason for the high water-cut phenomenon. We aim to investigate the hydrocarbon-water-rock interactions during the gas injection as well as production within a shale rock in the pore scale. To mimic the heterogeneous pore structure of the shale rock, we have designed a pore system, including a bulk pore, a pore throat, and a dead-end pore. We simulate the distribution of different fluids during the initial equilibrium stage, the primary depletion stage, and the huff ’n’ puff stage. The results show that an excessive amount of water is trapped by the condensation mechanism in the larger pores during the primary depletion stage. The water is then recovered with the injection of working (lean) gases. Moreover, we have analyzed the effect of different injection gases (IGs) and found that carbon dioxide (CO2) yields a higher water cut compared with methane (C1). Moreover, our findings have revealed the trapping mechanisms of hydrocarbon-water mixtures in shale rocks and have highlighted the impacts of pore structures on the recovery of shale reservoirs. As such, we have provided a potential explanation of the observed phenomenon.