Sandia_Natl_Labs_FY19_LDRD_Annual_SAND2020-3752 R_2_S

FY19 ANNUAL REPORT

Characterization of ultralow permeability geomaterials using electrokinetics. A team from Sandia, including collaborators from the University of Illinois at Urbana-Champaign, are extending the hydrogeophysical use of electrokinetics to characterize the permeability of key geomaterials. The effort is developing and refining a laboratory apparatus to estimate permeability for rock cores from

the electroosmotic and streaming potential coupling coefficients. At Sandia, coupled numerical models and analytical solutions were developed using an eigenvalue uncoupling approach to apply these methods at the laboratory and field scale. The team at Illinois produced pore-scale fully coupled electrokinetic simulations of flow around various shapes (see charge and flowlines in images). Pore-scale results are also being fused into lattices and porous media through machine learning.

Prediction and inference of multi-scale electrical properties of geomaterials. A team of Sandia geoscientists and mathematicians developed a pair of breakthrough algorithms for efficiently simulating the geophysical response of complex, multi-scale systems in the Earth’s subsurface critical to Sandia’s ongoing efforts in Nuclear Deterrence as well as Energy and Homeland Security. The problem with such complex systems is that for certain classes of geophysical phenomena (electromagnetic, fluid flow, etc.), the fine-scale detail present in real geologic materials can have a massive effect on the overall geophysical signature, yet capturing all the relevant details in a computer model is practically infeasible. Led by Chester J Weiss, the team proposed and implemented two distinct approaches to overcome this limitation: a hierarchical finite element method (HiFEM) to explicitly capture these details, and a fractional calculus method to accurately capture their macroscopic equivalence. These end-member approaches not only allow for a previously unobtainable set of predictive data, but also allow the team to explore the physics of the intermediate meso-scale where the transition from an exquisitely detailed to a “blurred” but equivalent upscaled model resides. Initial applications of the HiFEM software were extended to also account for the effects of man-made infrastructure in geophysical sensing applications, leading to reduction in computer runtime by a factor of 1000 or more. Details

of the research were published in Geophysics and Geophysical Journal International .

30

LABORATORY DIRECTED RESEARCH & DEVELOPMENT