Data-driven modeling of the constitutive equations of materials

Historically, constitutive laws are constructed by phenomenological observations or first-principle derivations. With atomistic simulations, we now have mechanistic insights into the fundamental physical processes, but scaling these atomistic processes up to larger-scale models remains a fundamental challenge. In this project, we use a data-driven method to discover the constitutive equations by learning from atomistic simulations. Enabled by this method, we can construct the constitutive relations for complex systems, which are challenging to elucidate or characterize by experiments. Currently, we focus on the constitutive PDEs for energy storage materials and stimuli-responsive polymers. 

Mechanics of energy storage materials

All-solid-state lithium battery is a promising candidate for high-capacity rechargeable batteries: Li anodes can provide the highest energy density and the lowest electrochemical potential among all the known anode materials; garnet-type solid electrolytes such as LLZO can stabilize electrochemical reactions at the Li metal anode and enhance the safety of Li metal batteries. However, new problems emerge. This research integrates computations and experiments to elucidate the mechanics of Li plating/stripping at the Li-solid electrolyte interface across different scales and design anode configurations with improved cycling performance and mechanical integrity.