Research Interests

Our research focuses on accelerating the discovery and optimization of advanced materials by combining graph neural networks (GNNs) with generative artificial intelligence. We develop machine learning models that learn the relationship between atomic structure, chemical composition, and material properties directly from large-scale computational and experimental datasets. GNNs are used to accurately predict properties such as stability, ion transport, electronic behavior, mechanical performance, and catalytic activity, while generative models explore vast chemical design spaces to propose entirely new materials with targeted functionalities. By integrating physics-based simulations, density functional theory, and explainable AI, our models enables rapid screening and inverse design of next-generation materials for applications including energy storage, catalysis, semiconductors, structural materials, and sustainable technologies. This AI-driven approach significantly reduces the time and cost required for traditional trial-and-error materials discovery while providing fundamental insights into the underlying materials physics.

Our research develops high-fidelity computational fluid dynamics (CFD) models to understand complex transport phenomena involving turbulent flows, multiphase systems, and fluid–structure–particle interactions. We combine physics-based simulations, advanced numerical methods, and artificial intelligence to investigate how fluid dynamics influence the transport, deposition, and evolution of particles across engineering and environmental systems. Current research includes an NSF-supported collaborative project led by Dr. Liyuan Hou at Utah State University on microplastic transport in turbulent flows, where we develop computational models to investigate the coupled effects of biofilm growth, hydrodynamic forces, and particle dynamics, with the goal of revealing the mechanisms governing contaminant fate in aquatic environments. More broadly, we aim to develop predictive computational frameworks that enable data-driven analysis and intelligent design of transport processes in environmental, energy, and manufacturing applications.

We develop computational and artificial intelligence approaches to accelerate the discovery, design, and optimization of next-generation energy storage materials. Our research integrates first-principles calculations, multiscale modeling, electrochemical simulations, and machine learning to understand the fundamental mechanisms governing ion transport, interfacial reactions, phase transformations, and materials degradation. Current applications include lithium-ion batteries and aluminum–air batteries, where we investigate electrode materials, solid electrolyte interphases (SEI), catalytic reactions, and electrochemical performance to enable safer, higher-energy-density, and longer-lasting energy storage systems. Our long-term vision is to establish autonomous computational frameworks that combine physics-based modeling with AI to accelerate innovation across a broad range of electrochemical energy technologies.