Nanoporous Materials for Gas Capture and Storage
Metal−organic frameworks (MOFs) consist of metal clusters or ions that are joined by organic linkers to form porous network solids with large surface areas, high crystallinity, and, in some cases, redox-active open metal sites. MOFs are promising for gas storage and separation applications, particularly in the area of carbon capture and ammonia and hydrogen storages. Our goal is to understand adsorption properties of MOFs and to design high performance MOFs for gas capture and storage using first-principles density functional theory (DFT) calculations and machine-learning (ML) techniques. We also closely work with several experimental groups.
Chemiresistive Gas Sensors
Chemiresistive gas sensors, also known as chemiresistors, are a type of chemical sensor that detect gases and volatile organic compounds (VOCs) by measuring changes in electrical resistance. Chemiresistive sensors are popular because they are easy to make, customizable, flexible, and have a fast response time. However, their performance can be significantly compromised in the presence of water. Therefore, we aim to design chemiresistive gas sensor materials with enhanced water stability using DFT calculations.
Halide Perovskites for Solar Cells
Halide perovskites have been intensively investigated for photovoltaic applications because of their good optoelectronic (or bandgap) properties and low cost. Meanwhile, various high-pressure experiments have shown that the se materials generally undergo reversible phase transitions between different crystalline phases as well as between crystalline and amorphous phases under external pressure. Considering this, our goal is to understand the origins of their good bandgaps and pressure-induced phases transitions using DFT calculations.
Methods
To understand fundamental properties of materials, we use first-principles density functional theory (DFT) and many-body perturbation theory (MBPT) methods. Also, we seek to pursue an AI-assisted materials design strategy to efficiently develop high-performance materials for given applications using (DFT-generated) high-quality materials databases. We believe that this approach will significantly accelerate materials developments.
Developments of New Computational Methods
We are developing new methodologies to accurately predict fundamental properties of materials and maximize the efficiency of computational methods. These methodologies not only allow for efficient and accurate calculations of material properties but also play a crucial role in designing new materials:
(i) We are conducting research to develop van der Waals density functionals (vdW-DFs) in accurately describing non-local correlation effects between electrons.
(ii) We are developing an algorithm that can deterministically find commensurate heterostructures, enhancing the efficiency of interface structure simulations using DFT calculations and predicting various interface structures that have not been reported so far.
(iii) We are developing machine learning potentials (MLPs) based on artificial neural networks to secure the time-scale of MD and the accuracy of DFT, thereby conducting research to predict fundamental properties of material systems that conventional computational methods have not been able to handle until now.