Many natural and artificial processes involve the simultaneous flow of two or three fluid phases (such as water, oil, and air) through porous media. However, the prediction of how fast the phases move and whether they can be retrieved is not trivial. It requires knowledge of the topology of different fluids in a complex three-dimensional space that ranges from simple soil to root-soil systems, and from a uniform sandstone to fractured carbonate rocks. Both porous media (soil or rock) and fluid properties (i.e. which fluid is preferentially wetting the rock surface) determine the final spatial distribution when multiple fluids are competing within pores. Furthermore, minuscule rock or soil pores often control which fluids become trapped and where. Quantifying the spatial arrangement of fluids in pores of complex geometry and varying wettability remains an unsolved fundamental problem. The project objective is to explore and quantify dynamic spatial fluid arrangements from first principles and their control on flow complexity of porous media. To address the objective, the project will develop a flexible, multiscale numerical method to describe two- and three-fluid phase flow configurations in porous materials of heterogeneous wettability. Through cooperation, detailed 3D experimental data imaged with x-ray microtomography of two and three-phase flow configurations will be available for both input and verification of the numerical method. Results will enable the quantification and upscaling of rock surfaces of different wettability, porous media roughness and the ultimate arrangement of fluids to larger spatial regions of interest.
One example region of application is the Edwards Aquifer in Texas, which provides water for approximately two million people in the Austin and San Antonio metropolitan areas. The aquifer is formed by heterogeneous limestone, inter-bedded and overlain with sand/gravel sediments. Additional flow-path heterogeneity originates from faults and fractures: 95% of the flow occurs in a highly conductive fracture network, whereas 95% of water storage is in the rock matrix. The difference between fracture network conductivity and matrix storage, combined with highly erratic rainfall, are at the core of water management issues in central Texas. When it rains and water moves through fractures and out of the system exceedingly fast (thereby causing flash flooding); after a dry period, due to wettability changes and water/air competition in this heterogeneous reservoir, the storage matrix might be 90% full, but wells often become dry. Controlling a contaminant spill in such a heterogeneous medium (also known as karst) introduces the competition of three fluid phases in the pore space. Regardless of this specific example, the findings of the proposed research will become relevant to many different fields of research such as carbon dioxide sequestration in depleted oil reservoirs and in remediation and enhanced oil recovery efforts. Results will be of significant use in the extrapolation of existing theory and models to mixed wet, two- and three-fluid phase systems. In addition, a digital image repository for porous materials initiated along with this proposal will enhance research infrastructure and benchmarking of porous media theory, models and experiments. Finally, karst formations and water management issues are part of the everyday experience in Texas and they are ideal to educate a broad audience on geology, flow physics and chemistry, engineering as well as applied mathematics and computer science. Results from this project will be tested by teachers at professional development projects funded by the NSF and DOE, and then integrated into accessible online modules and included in an interactive exhibit in the Austin's Children Museum.
