Computational Discovery of Materials for CO2 Capture

There has been a rapid increase in atmospheric CO2 levels over the past century, sparking significant concerns about global warming. Metal-organic frameworks (MOFs) are a promising class of “spongy” porous materials for capturing CO2 from the atmosphere or point emission sources. Their modular synthesis nature allows for targeted tuning of material properties and for virtually unlimited variations in the material's structure and chemistry.

We are looking for a highly self-driven undergraduate student to test a new molecular simulation method for modeling gas adsorption in MOFs. The student will play a major role in testing this new simulation method on characteristic MOF materials and evaluating the computational efficiency and accuracy of the method. The student will also have the chance to dive into the development of new simulation algorithms. The ultimate goal is to apply this new molecular simulation method to efficiently search the large materials space for promising MOFs that can be eventually deployed in practical chemical processes for CO2 mitigation. This is a chance to learn how to simulate molecular behavior in silico and how machine learning can help solve global challenges.

Join our team and let’s discover new materials to save the planet! 

The student will gain valuable skills and knowledge in molecular modeling, machine learning, and computational materials, in the context of gas adsorption in nanoporous materials. They will also develop strong communication skills through presentations in group meetings and one-on-one meetings with the advisor. If the project is successful, the student will have the chance to publish their research outcome in professional journals and present their work at local symposiums or national conferences. This hands-on experience will empower the student with practical insights into computational material discovery, preparing them for future endeavors in sustainable chemical and materials engineering

• Prof. Kaihang Shi
  
• Digital Porous Materials Lab 
  

1. Familiar with the basics of molecular simulations for gas adsorption, especially grand canonical Monte Carlo simulation:

Exploring the Structural, Dynamic, and Functional Properties of Metal-Organic Frameworks through Molecular Modeling (https://onlinelibrary.wiley.com/doi/full/10.1002/adfm.202308130)

Using molecular simulation to characterize metal–organic frameworks for adsorption applications (https://onlinelibrary.wiley.com/doi/full/10.1002/adfm.202308130)

Using Molecular Simulation to Predict Adsorption Properties by Randy Snurr (https://www.youtube.com/watch?v=bzmXUN1CGvo)

2. Familiar with machine learning force field or machine learning potential. Some useful references are:

Exploring the Structural, Dynamic, and Functional Properties of Metal-Organic Frameworks through Molecular Modeling (https://onlinelibrary.wiley.com/doi/full/10.1002/adfm.202308130)

Learning local equivariant representations for large-scale atomistic dynamics (https://www.nature.com/articles/s41467-023-36329-y)

Machine Learning Force Fields (https://pubs.acs.org/doi/10.1021/acs.chemrev.0c01111)

To understand how machine learning force field works requires the student to be familiar with molecular simulation and supervised machine learning techniques.

3. If the student has no previous experience with Python, they should also be familiar with the basics of Python coding and command-line Linux.

More relevant learning resources are available in the Shi group handbook:

https://shigrouphandbook.notion.site/Learning-Resources-56606605cb9447f5a9c22174960c1fc1

Department of Chemical and Biological Engineering, Computational Materials Design, Artificial Intelligence, Machine Learning, Molecular Simulation, Global Warming, Computational Materials Discovery, Computational Science and Engineering