wengroup

THE WEN GROUP @ U HOUSTON

Materials Design with AI, Simulation, and Data

ABOUT

We are a computational group working in the broad area of chemical and materials sciences at the University of Houston. An overarching theme of our group is to leverage artificial intelligence, high-performance computing, and atomistic molecular simulations to understand and design new materials for renewable energy and healthcare applications. We develop state-of-the-art machine learning models, cutting-edge automation workflows, and advanced reliability analysis frameworks, and apply them to study Li-ion and solid-state batteries, zeolite catalysts, and thermal-photonic devices.

news
book

PEOPLE

icon
icon
icon

Principal Investigator

Mingjian Wen

Mingjian Wen

Assistant Professor, Chemical & Biomolecular Engineering

mjwen@uh.edu

CV

Mingjian tracks how atoms move and interact with each other on weekdays and cheers for soccer games on weekends. He seems to be tied to spherical objects.

Postdocs

Santosh Adhikari

Santosh Adhikari

sadhik22@central.uh.edu

Santosh received his PhD in Physics (spring, 2021) from Temple University. He is interested in machine learning-assisted understanding and discovery of novel materials.

Jin Dai

Jin Dai

jdai9@central.uh.edu

Jin received her PhD in Materials Science Engineering from Michigan State Univerisy in 2023. Her research focuses on exploring the interfacial properties between electrodes and solid electrolytes using simulation techniques.

Graduate Students

Wei-Fan Huang

Wei-Fan Huang

whuang25@uh.edu

Wei-Fan received his B.S. and M.S. in Chemical Engineering from National Cheng Kung University in Taiwan in 2018 and 2021. His research focuses on developing machine learning models and their applications in molecular simulation of battery materials.

Emmanuel A. Olanrewaju

Emmanuel A. Olanrewaju

eolanrew@uh.edu

Emmanuel loves computational work to understand and design materials for renewable energy applications. He currently works on data-driven understanding of the stability of zeolites.

Dale E. Green

Dale E. Green

degreen@uh.edu

Dale received his B.S. in Chemical Engineering from Texas Tech in 2001 and M.S. ChE from Kansas State in 2012. He has worked in various manufacturing and technology roles for Dow Chemical and Olin since 2001. His research interests are centered around computational catalyst design and optimization with application in epoxy technology. (Co-advised with Prof. Lars Grabow.)

You

You

<you>@uh.edu

The headshot image is generated with the text "A graduate student doing AI for science research." by DALL·E, an AI image generation tool. Does this describe you? Come and join us.

Undergraduate Students

Bobby Brown

Bobby Brown

Bobby is a fellow of the Energy Scholars Program, an extension of SURF.

Heer Loungani

Heer Loungani

Heer is an intern from HCC, currently pursuing a B.S. degree in Computer Science.

Chris Mobley

Chris Mobley

A PURS research assistant, Chris is pursuing a B.S. in Chemical Engineering and a minor in Data Science.

Lawrence Smith

Lawrence Smith

Lawrence is a Chemical Engineering student at UH.

Catalina Campos

Catalina Campos

Catalina is an intern from HCC, currently pursuing a B.S. degree in Computer Science.

Abby Zhao

Abby Zhao

Abby is a high school student interested in chemical science.

RESEARCH

icon
icon
icon

    Machine Learning Chemical Reactions

    Machine learning methods, especially deep learning, have significantly expanded a chemist's toolbox, enabling the construction of quantitatively predictive models directly from data. These models make it possible to explore the gigantic chemical space to make chemical discoveries.

    Our group focus on the chemical reaction space. We develop novel graph neural networks that are able to represent any chemical reactions with bond alterations and apply them to model reaction properties such as reaction energy, reaction type, and activation energy. These models have been used to investigate the reaction pathways in battery electrolytes.


    Accelerated Atomistic Molecular Simulations

    Molecular simulations are a powerful computational technique for exploring material behavior and properties based on an understanding of the physics of bonding at the atomic scale. At the core of any molecular simulation lies a description of the interactions between atoms that produces the forces governing the atomic motion. In classical molecular simulations, such interactions are modeled via interatomic potentials, which make it possible to simulate a large number of atoms for extended periods of time.

    We develop both physics-based and machine learning interatomic potentials and apply them to study the energetic, structural, thermal, and mechanical properties of materials and devices for renewable energy applications. We also develop model analysis frameworks (e.g. uncertainty quantification methods), aiming at making classical molecular simulations more reliable, reproducible, and accessible.


    High-throughput Materials Discovery

    Quantum chemical theory and computations are powerful tools for understanding and designing materials. Conventional approaches that manually perform the calculations are difficult to manage complex materials science workflows and difficult to fully utilize modern high-performance computing resources.

    Building on top of the software stack that powers the Materials Project, we develop high-throughput materials discovery recipes for density functional theory (DFT) and apply them to discover new materials with unique mechanical and photonic properties with applications in solid-state batteries and thermal-photonic devices.


CODES

icon
icon
icon
  • MatTen

    An equivariant graph neural model for predicting tensorial properties of crystals such as the elasticity tensor.

    GitHub

    Documentation

  • KLIFF

    A fitting package that can be used to develop both physics-based and machine learning interatomic potentials.

    GitHub

    Documentation

  • BonDNet

    A graph neural network machine learning model for the prediction of bond dissociation energies.

    GitHub

    Documentation

  • RxnRep

    Contrastive pretraining to learn chemical reaction representations (RxnRep) for downstream tasks.

    GitHub

    Documentation

OPENINGS

icon
icon
icon
  • Postdocs

    There is currently no postdoc opening.

  • Graduate Students

    We take a couple of new graduate students each year. Prospective students are encouraged to take a look at the department website for information on how to apply for admission. Students who have already been admitted please send Dr. Wen an email if you are interested in joining the group.

  • Undergraduate Students

    We welcome undergraduate students to join our group to conduct research. Please contact us if you are interested.