Modeling carbon dioxide and methane adsorption on illite and calcite: enhancing the simplified local density model through crystal structure modifications.

The simplified local density (SLD) model has been frequently utilized to describe gas adsorption in porous media. The common assumptions associated with the SLD model are the graphene slit and uniform distribution of atoms. However, these fail to depict the heterogeneous surface of minerals. In this study, the original SLD model was modified by substituting mineral crystal structures for the homogeneous carbon layer when building the slit. The modified model may capture the heterogeneity-induced variation of the fluid-pore interaction potential and adsorbed phase density near the mineral surface. The calculated adsorption isotherms of methane and carbon dioxide on illite and calcite surfaces at 330.15 K were compared with literature experiment data to validate the modified SLD model. For the simulation of gas adsorption isotherms, the modified model predictions agree reasonably well with the previously reported experiment results using the gravimetric method. The calculated density profile in the slit indicates the monolayer adsorption behavior of CH4 and CO2. Based on more precise interaction potentials, the number of regression parameters was reduced to two, and the physical meaning of the model parameters was clarified. Therefore, for estimating hydrocarbon storage in reservoirs, the modified SLD model could be used as an efficient alternative to time-consuming molecular simulation methods.

Effects of Supercritical CO2 on the Pore Structure Complexity of High-Rank Coal with Water Participation and the Implications for CO2 ECBM.

To reveal how mineral changes affect a coal pore structure in the presence of water, an autoclave was used to carry out the supercritical CO2 (ScCO2)-H2O-coal interaction process. To reveal the changes in pore complexity, mercury intrusion capillary pressure (MICP), low-pressure nitrogen adsorption, CO2 adsorption, and field emission scanning electron microscopy (FESEM) experiments were combined with fractal theory. The experimental data of MICP show that the MICP data are meaningful only for the pore fractal dimension with pore sizes >150 nm. Therefore, the pores were classified into the classes >150, 2-150, and <2 nm. The results show that the pore volume and specific surface area of the coal increased significantly after the reaction. ScCO2-H2O can cause the formation of many new pores and fractures in the coal. The presence of H2O may increase the potential for the injection of CO2 into the coal seam. The complete dissolution of calcite surfaces caused a significant increase in the pore volume and specific surface area of the pores >150 nm. The morphologies of these pores are controlled by the morphologies of the complete dissolution carbonate particles. The pore morphologies were relatively uniform, and the fractal dimensions decreased. However, the incomplete dissolution of calcite leads to irregular variations in the morphologies for the pores in the 2-150 nm pore size range. The pore morphologies that are produced by incompletely dissolved calcite particles are more complex, which increases the fractal dimensions after the reaction. The fractal dimensions of the pores <2 nm decreased after the reaction, indicating that the newly generated micropores were more uniform and had regular pore morphologies.

Estimating Lost Gas Content for Shales Considering Real Boundary Conditions during the Core Recovery Process.

Shale gas has become an important natural gas resource in recent years as the conventional oil and gas resources are depleting. Shale gas content is one of the most important parameters for reserve calculation and sweet-spot prediction. The traditional core recovery method is widely used to determine gas content. However, the estimation of lost gas content is the main factor of error and difficulty. Large errors and uncertainties occur when using the widely used methods, such as the United States Bureau of Mines (USBM) method. Hence, a more accurate method is required. In this work, a full-process model is developed in COMSOL Multiphysics to describe the lost gas with time during the core recovery process as well as the desorption stage after the core is covered. In this method, by setting the initial gas pressure and flow parameters and matching the desorbed gas volume and considering variable diffusivity with respect to temperature, the initial gas content and the gas lost with respect to time are calculated. Overall, 10 field data are tested using this full-process model, and the USBM method is also applied to compare the results. It is found that if the ratio of lost gas volume estimated using the USBM method to the desorbed gas volume of the field data is lower than 2.0, the USBM method underestimates the lost gas compared to the full-process method; if the ratio is about 2.0, the results from the USBM and the full-process methods are comparable; and if the ratio is close to 3.0, the USBM method tends to overestimate the lost gas. The modeling results indicate that this proposed full-process method is more theoretically sound than the USBM method, which has high uncertainties depending on the number of desorbed gas data points used. Nevertheless, this proposed method requires a large number of parameters, leading to the difficulty in finding true parameters. Therefore, an optimization algorithm is required. In summary, this study provides theoretical support and a mathematical model for the inversion calculation of lost gas during shale core recovery. It is helpful to evaluate the resource potential and development economics of shale gas more accurately.

Theoretical Prediction of the Occurrence of Gas Hydrate Stability Zones: A Case Study of the Mohe Basin, Northeast China.

Source rocks of the Mohe Basin, Northeast China are gas-prone and the organic matter has advanced to late oil-generation stages, producing condensate and natural gas. This provides suitable conditions for the Mohe Basin to become one of the most prolific terrestrial natural gas hydrate (NGH)-bearing areas in China. Knowing this, here we predict the depth and thickness of pure methane hydrate stability zones (HSZs) and gas hydrate stability zones (GHSZs) via simulating the hydrate-phase equilibrium and other formation P-T conditions. Furthermore, factors that have a major impact on the occurrence of HSZs are discussed. Results showed that the composition of gas (guest) molecules and the geothermal gradient are the two most controlling factors on HSZs. Moreover, it was found that a pure methane HSZ with a thickness of about 255 m can form in areas with a geothermal gradient of <1.5 °C/100 m, with top and bottom depth limits less than 493 m and greater than 748 m, respectively. In contrast, pure methane hydrates have difficulty forming, while hydrates from wet gas can form where there is a geothermal gradient of >1.6 °C/100 m. Furthermore, a wet gas HSZ with a thickness of at least 735 m can be expected when the geothermal gradient reaches 2.3 °C/100 m, with top and bottom depth limits at 115 and 850 m, respectively. Ultimately, a pure methane HSZ can still form in the abnormally high-pressured areas when the geothermal gradient is up to 2.0 °C/100 m. Overall, HSZs can occur due to the combined effect of formation temperature, pressure, and gas composition. Finally, based on the results from this study and drilling data, future successful hydrate drilling schemes can be implemented in the Mohe Basin and similar terrestrial areas.

New Adsorption Models for Entirely Describing the Adsorption Isotherm and Heat of Methane in Heterogeneous Nanopore Structures of Coal.

To understand the adsorption mechanism of methane in heterogeneous nanopore structures of coal, integral adsorption models based on linear, exponential, hyperbolic and quadratic energy distribution functions are established. The adsorption energy domain of the new models is assumed to be a finite interval. These new adsorption models can describe both the adsorption isotherm and the adsorption heat. A volumetric method of adsorption with a microcalorimetry system is used to measure the adsorption isotherms and integral heat, and then the parameters of the new models are obtained by fitting the experimental data. Since the adsorption heat can be different for different adsorption models, it is necessary to fit the adsorption isotherms and heat simultaneously. The fitting results of the adsorption isotherms and heat show that the new models are able to describe the experimental data better than the Langmuir model. By comparing the fitting results and the effective range of adsorption energy of the different adsorption models, it is shown that the exponential energy distribution function is the most reasonable model for methane adsorption in coals, which can be used to evaluate the energetic heterogeneity of nanopores in coal samples. The decreasing exponential energy distributions of three coal samples indicate that a larger adsorption energy corresponds to fewer adsorption sites in the coal samples. The proportion of high adsorption energy is related to the micro-nanopore volume in the coal samples.

Variation in Permeability during CO2-CH4 Displacement in Coal Seams. Part 2: Modeling and Simulation.

Having a clear understanding of the permeability variation mechanism is important for controlling the process of displacement of CH4 with CO2 in deep coal seams. Based on the stress-strain equation of porous elastic media and horizontal strain variations of coal, a mathematical model predicting permeability variation after CO2 injection into gas saturated coal seams was established. The model shows that, during the displacement of CH4 with CO2, the shrinkage strain of the coal matrix increases logarithmically with the decrease of pore pressure. With a decrease in the reservoir pressure, permeability rebound occurs with the influence of matrix shrinkage and gas slippage. Under low confining pressures, the rebounded permeability is high, and its associated rebound pore pressure is also high. For coals with a high cleat compression coefficient, the permeability decreases range is obvious. And permeability rebound only happens under low reservoir pressures. Coal properties, e.g., Poisson's ratio and Langmuir volume, show obvious influences in permeability variation during gas production. The model was also extended to predict permeability variation for a well-control area. During gas drainage process, the permeability in the well-controlled area first increases, then decreases, and then slowly returns to the original state with the lengthening of well-controlled radius. Under high confining pressures, the permeability decline range is more obvious. Also, correspondingly, the attenuation range of permeability increases and the rebound range decreases. The proposed model is beneficial in predicting permeability variations during the displacement of CH4 with CO2, as well as guiding CO2 injection into coal seams.

Atomistic Study of Dynamic Contact Angles in CO2-Water-Silica System.

The dynamic wetting for the CO2-water-silica system occurring in deep reservoirs is complex because of the interactions among multiple phases. This work aims to quantify the contact angle of CO2-water flow in the silica channel at six different flow velocities using molecular dynamics. The dynamic contact angle values at different contact line velocities are obtained for the CO2-water-silica system. By calculating the rates of the adsorption-desorption process of CO2 and water molecules on the silica surface using molecular dynamics simulations, it has been found that the results of the dynamic contact angle can be explained by the molecular kinetic theory and predicted from the equilibrium molecular simulations. Moreover, the capillary pressure at different contact line velocities is predicted according to the Young-Laplace equation. The change in contact angles at different velocities is compared with empirical equations in terms of capillary number. The results of this study can help us better understand the dynamic process of the multiphase flow at the nanoscale under realistic reservoir conditions.

Interactions and exchange of CO2 and H2O in coals: an investigation by low-field NMR relaxation.

The mechanisms by which CO2 and water interact in coal remain unclear and these are key questions for understanding ECBM processes and defining the long-term behaviour of injected CO2. In our experiments, we injected helium/CO2 to displace water in eight water-saturated samples. We used low-field NMR relaxation to investigate CO2 and water interactions in these coals across a variety of time-scales. The injection of helium did not change the T2 spectra of the coals. In contrast, the T2 spectra peaks of micro-capillary water gradually decreased and those of macro-capillary and bulk water increased with time after the injection of CO2. We assume that the CO2 diffuses through and/or dissolves into the capillary water to access the coal matrix interior, which promotes desorption of water molecules from the surfaces of coal micropores and mesopores. The replaced water mass is mainly related to the Langmuir adsorption volume of CO2 and increases as the CO2 adsorption capacity increases. Other factors, such as mineral composition, temperature and pressure, also influence the effective exchange between water and CO2. Finally, we built a quantified model to evaluate the efficiency of water replacement by CO2 injection with respect to temperature and pressure.

Effect of surface energy on carbon dioxide hydrate formation.

In this work, the effect of the surface energy between the hydrate clusters and the aqueous phase on the hydrate formation of carbon dioxide was thoroughly investigated. Our results show that the threshold pressure for hydrate formation is less sensitive to the temperatures if the surface energy is not larger than 7 mJ/m(2). However, the threshold pressure is very sensitive to the temperatures and increases significantly with the surface energy if it is over 7, 9, and 10 mJ/m(2) for the temperatures of 279, 277.45, and 276 K, respectively. The value of the surface energy for the CO(2) hydrate/water system was determined as 9.3 mJ/m(2) by comparing our results with the experimental data for stable CO(2) hydrate formation at the temperature range of 277.45-278.85 K and at pressures of 55-165 bar. This value is very close to the recently reported value of 7.5 ± 1.4 mJ/m(2) from molecular simulation. A theoretical method was proposed for computing the induction time of hydrate formation adopting a composite of the time required for critical nuclei formation and their growth to a detectable size. By using this method, the surface energy was avoided in the induction time calculation and an average crystallite volume of 0.238 mm(3) at the induction time was derived based on the experimental data. It provides an approach for predicting the induction time for gas hydrate formation.