Study of Slip Flow in Shale and Tight Gas Reservoir Rocks Using Lattice Boltzmann Method

Abstract

Abstract Development and production from unconventional resources require understanding of flow mechanisms and rock/fluid interactions, which are different from those in conventional reservoirs. Several flow regimes including continuum, slip and transition can dominate the fluid flow in these reservoirs due to the presence of various pore sizes from micro to nanoscale. Accurate prediction of flow behaviour requires suitable simulation techniques for modelling the flow regimes active in unconventional reservoir rocks. Recently, the lattice Boltzmann method (LBM) has received great attention as a well-accepted method for the simulation of fluid flow in nanoscale systems. However, the impact of different boundary conditions to model such fluid-rock interactions have not been investigated so far. In addition, the LBM simulation results have rarely been compared with the experimental data. In this study, the two-relaxation-time (TRT) based Lattice Boltzmann approach was adopted to simulate the gas flow in nanoscale single pipe. Different boundary conditions including bounce back, specular reflection, diffusive reflective and bounce back-specular reflection (BSR) were used to capture the gas slippage at the wall surface. For validation, the simulation results were compared with the results of other simulation techniques such as Direct Simulation Monte Carlo and Information Preservation methods reported in the literature. After validation, the impacts of different boundary conditions and various tangential momentum accommodation coefficients (TMAC) were studied. In addition, gas flow in a simplified porous medium (a system of pore body/throat) was modelled and the simulation results were compared with those obtained for gas flow in a single channel. Furthermore, the simulation results of gas flow in a single channel and in the simplified porous medium were compared with the scaled experimental data measured on three shale rock samples. The results show that, among different boundary conditions, BSR is the most suitable one for gas flow simulations in shale rock samples. However, the LBM simulation with the BSR boundary condition using the TMAC literature value of 0.8 underestimated the permeability enhancement (due to gas slippage) in these shale rocks. It was found that a TMAC value of 0.6 could better estimate the permeability enhancement. It was also shown that the characteristic length of porous media could be better described by the average of "pore throat" rather than "pore body" sizes. Moreover, the permeability was overestimated for Kn>0.1, when the gas flow was simulated in a single micropipe or microchannel rather than the considered porous medium. The results obtained in this study can be used for more realistic predictions of shale matrix permeability, when the slip or transition flow regime is dominant. It also improves our understanding of using LBM for simulation of fluid flow in unconventional reservoir rocks.