Steve WaiChing Sun | Predicting Faulting, Landslides, and Liquefaction of Soil from Motion of Grains at Nanoscale
No matter how large a rock formation is, in its most elementary form it is just a construct of many solid particles, the open spaces or pores of which allow fluid and gas to flow through. Understanding how those solid particles, pores and fluid interact at microscopic scale is important for large-scale engineering applications such as nuclear waste disposal, carbon sequestration and storage, oil drilling, and hydraulic fracturing, and can help predict landslides and slope failures.
Steve WaiChing Sun
Assistant Professor of Civil Engineering and Engineering Mechanics
—Photo by Jane Nisselson
Much progress has been made in developing pore-scale simulations, but these models remain counterproductive and impractical for field applications that span square miles.
Steve WaiChing Sun, assistant professor of civil engineering and engineering mechanics, is improving predictions of large-scale field problems with insight from small-scale observations and simulations. His recent work incorporates 3D tomographic images of core samples to analyze why and how environmental and mechanical conditions result in flow barriers in rock. By combining X-ray imaging techniques with computational models that simulate flow and grain motions, Sun provides evidence that some flow barriers may form not only by compressing pores, but also by making flow channels more torturous and less interconnected. This finding sheds light on understanding the mechanism that leads to leakage of pore fluids, such as wastewater or radioactive materials, from some reservoirs but not others, and was awarded the 2013 Caterpillar Best Paper Prize.
Sun’s research is focused on advancing the understanding of multiphase materials under extreme conditions and enhancing predictive capabilities for related engineering applications. His new computational framework can rapidly extract microstructural attributes inferred from micro-tomographic images and makes it possible to conduct cost-efficient computations to predict large-scale catastrophic events—research that can be used to save lives.
“I grew up in Hong Kong where there are lots of rainfall-induced landslides and slope failures during the summer,” he explains. “Owing to the complexity of the events, it is hard to derive simple analytical models to predict those disasters. The idea of using computer models to predict those events and thus save lives are therefore fascinating to me.”
Along with his work in modeling and homogenization of mechanical and hydraulic properties of porous media from CT images, his research also includes the development of solution techniques for coupled deformation-diffusion in non-isothermal saturated and unsaturated porous media and formulations of stabilized mixed-field finite element models for large deformation multiphysics problems.
“I am particularly interested in advancing the understanding of why and how plastic strain localizes and fracture propagates in porous media that are partially saturated with multiple fluids, such as water, oil, and CO2,” he says. “Due to the recent advancement in micro-CT and digital image correlation techniques, there are tremendous opportunities to incorporate nanoscale and micro-scale information to improve the accuracy and reliability of large-scale models (such as a reservoir simulator) that are often developed to model kilometer-scale problems and contribute to the understanding of a wide spectrum of natural and man-made processes.”
Prior to joining Columbia, Sun was a senior member of the technical staff in the mechanics of materials department at Sandia National Laboratories in Livermore, CA.
BSc, UC Davis, 2005; MSc, Stanford University, 2007; MA, Princeton University, 2008; PhD, Northwestern University, 2011
—by Amy Biemiller