Research

Multiscale simulation is a core skill in the THEMMES group. We are actively investigating various topics relevant to energy materials and systems via multiscale simulation. Simulation scale depends on the system we are interested in. We might study an origin in microscopic scale using quantum mechanics simulation technique, e.g., DFT simulation. It would be interesting to understand the phenomena in molecular scale with the help of statistical level simulation, e.g., MD simulation based on many different levels of force-field. Furthermore, it is to be observed in mesoscale simulation to match experimental findings taking advantage of its approximation, e.g., KMC or microkinetic modelling. In more large scale, we are able to utilise continuum level simulation, with the help of various user-friendly programs, e.g., COMSOL.  

Artificial intelligence is a general approach whilst any other area in our daily life. The THEMMES group is also interested in accelerating our investigation with this technology, e.g., using machine learning force field to construct a potential energy surface or discovering new functional materials via diverse machine learning models based on pre-built or novel database. We are happy to use various techniques from classical machine learning to deep learning. We are not only trying to take advantage of pre-built models and database, but willing to develop new models and construct database. Currently, we are focusing on utilising many databases from simulation. It might lead us to efficiently investigate materials chemistry. 

Basically, the THEMMES group is theoretical energy materials modelling group, so our main focus is to be an expert to conduct multiscale simulation and understand various energy materials and systems. However, it is always a rule of thumb that the model match experimental findings. It motivates us to do experimental validation. We actively collaborate with experimental groups to get this in many research themes. We are also trying to develop our own experimental setups to corroborate our findings efficiently.