Apostolopoulou, Maria; (2019) Fluid transport through porous media: A novel application of kinetic Monte Carlo simulations. Doctoral thesis (Ph.D), UCL (University College London).

Preview

Text Thesis_Apostolopoulou.pdf - Accepted Version Download (7MB) | Preview

Abstract

With increasing global energy demands, unconventional formations, such as shale rocks, are becoming an important source of natural gas. Current efforts are focused on understanding fluid dynamics to maximise natural gas yields. Although shale gas is playing an increasingly important role in the global energy industry, our knowledge of the fundamentals of fluid transport through multiscale and heterogeneous porous media is incomplete, as both static and dynamic properties of confined fluids differ tremendously from those at the macroscopic scale. Transport models, derived from atomistic studies, are frequently used to bridge this gap. However, capturing and upscaling the interactions between the pore surface and fluids remains challenging. In this thesis, a computationally efficient stochastic approach is implemented to simulate fluid transport through complex porous media. One-, two-, and three-dimensional kinetic Monte Carlo models were developed to predict methane transport in heterogeneous pore networks consisting of hydrated and water-free micro-, meso-, and macropores, representative of shale rock minerals. Molecular dynamics (MD) simulations, experimental imaging and adsorption data, which describe the surface – fluid interaction and the pore network features respectively were utilised to inform the KMC models. The stochastic approach was used to (1) quantify the effect of the pore network characteristics (pore size, chemistry, connectivity, porosity, and anisotropy) on the transport of supercritical methane, (2) estimate the permeability of an Eagle Ford shale sample and evaluate the effect of proppants on permeability, and (3) to upscale atomistic insights and predict fluid diffusivity through different size pores. The results obtained were consistent with the analytical solutions of the diffusion equation, experimental data, and MD simulations, respectively, demonstrating the effectiveness of the stochastic approach. In addition, the applicability of less computationally intensive deterministic approaches was examined using multiple case studies; recommendations are provided on the optimal conditions under which each method can be used.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as JSONXMLCSV

Downloads by country - last 12 months

Export as JSONXMLCSV

Archive Staff Only

View Item