Experimental and numerical investigations of shale gas permeability

Abstract

Shale gas has become an essential source of energy supply in the United States and the world. It is important to analyze shale gas flow behavior and the associated production behavior from experimental and simulation perspectives. The specific objectives in this study are: (1) analyze shale gas flow behavior in the laboratory using pulse-decay experiments in both un-fractured and fractured cores, which will be accomplished in chapter 2 and chapter 3, respectively (2) propose a workflow to estimate shale gas permeability evolution taking into account multiple physics in the organic-rich matrix, which will be accomplished in chapter 4 (3) perform shale gas production simulations to justify the importance of these physics for shale gas production volume, which will be accomplished in chapter 5. Pulse-decay experiments using adsorptive gas (nitrogen and carbon dioxide) and non-adsorptive gas relate complex relationships between gas apparent porosity and adsorption (in forms of Gibbs/excess and absolute), and gas apparent permeability and adsorption. Among the three types of gas, helium apparent permeability is the highest. The ratio between apparent porosity and intron porosity of tight porous media is the summation of unity and the density ratio between the adsorbed phase and the free gas phase. A coefficient different from the density ratio is involved in the flow governing equation of adsorptive gas, which is in accordance with the adsorption contribution to the gas apparent permeability. For nitrogen, the coefficient is positive; and for carbon dioxide, the coefficient can be positive or negative below and above the phase change pressure, respectively. Virtual experiment simulations by discrete-fracture modeling of the pulse-decay systems reveal that the effective permeability of the tight porous media with complex configurations during the transient flow process is different from that during the steady-state flow process. History matchings of pulse-decay experiments on one naturally fractured core under different pore pressure and effective stress reveal that, matrix permeability is more sensitive to effective stress than matrix porosity, fracture porosity, and fracture permeability. It also implies that gas flow under a higher pressure is more likely in the homogeneous porous media as the matching errors show a decreasing trend with the pore pressure. The workflow of estimating the shale gas permeability evolution curve reveals that, slip flow/pore diffusion and surface diffusion in shale matrix largely mitigate the permeability decrease by the effective stress during the gas depletion process. Shale gas production simulations considering dynamic properties of both the matrix and fractures indicates that, the natural fracture spacing is the most important to shale gas recovery. By contrast, slip flow and pore diffusion are less important. Surface diffusion might be essential to shale gas recovery depending on the surface diffusivity value: the 5-year forecast error of the gas recovery factor without considering surface diffusion is higher than 5% when the surface diffusivity is higher than 1E-6 m2/s.