1. Carbon Capture and Sequestration
Carbon capture and sequestration (CCS) is a process in which carbon dioxide is captured before their emission into the atmosphere and buried in a safe underground reservoir. Because carbon dioxide is a greenhouse gas that exacerbates global warming, it is important to reduce the amount of carbon dioxide in the atmosphere. In our research group, we search for optimal materials that can selectively capture carbon dioxide gas molecules from point contact sources such as the power plant flue gas. Because the number of potential candidates are too large (over billion), computational tools and techniques are necessary to quickly and to accurately screen/characterize a large database of materials.
2. Methane/Hydrogen Storage
The need for alternative fuel source is greater than ever with decreasing amount of oil. To this end, we look into novel porous materials that have large internal surface area that allows significantly large amount of methane/hydrogen storage. Because these porous materials are highly tunable, we look into various strategies to design these materials in silico (inside a computer) that can optimize the gas uptake of methane/hydrogen.
3. Next-generation Materials Genome
The basic molecular building blocks (e.g. metal atoms, organic linkers) can be combined in many different ways to form crystalline structures with different chemistry and topology. The Materials Genome project seeks to enumerate a large database of porous materials and map their functionalities to different sets of applications that can be used in the future.
In practice, it is difficult to correctly model the interactions between guest molecules and framework materials. In many cases, off-the-shelf force field such as UFF provides poor description of the particle interactions and subsequently leads to wrong predictions of adsorption/diffusion properties. To accurately model the system, one can in principle conduct thousands of density functional theory (DFT) calculations to get the adsorption properties at large computational cost. We look for novel solutions in which we can minimize the number of expensive DFT calculations while still getting accurately predictions for different types of porous materials.
5. High-performance Computing Methods for Large-scale Screening/Characterization of Porous Materials
Our in-house developed GPU code can characterize both adsorption and diffusion properties of porous materials in a very efficient manner that allows large-scale screening of many different structures. We are continuing to add features to our code to enhance its applicability. Moreover, we seek to release a public version of the code soon so many other people can use them for their own applications. On top of this, our research group looks into other advancements in computing and seek to find their usage for scientfic applications.