Development of Non-empirical Force Field for Molecular Simulation

In the molecular simulation study, the force field is of great importance. The force field should be accurate and non-empirical in order to make object interpretation and prediction. Meanwhile, the transferability of the force field should also be guaranteed for wide applications. In this case, we have proposed a systematic method to generate non-empirical force field for different purposes. In the calculation of vibrational spectroscopy, we developed a polarizable force field based on charge response kernel (CRK) theory and applied the model to calculate IR, Raman and Sum frequency generation (SFG) spectra of various organic molecules. This force field, for the first time, extends the computational study of SFG spectra to complicated organic molecules, especially in C=O, C-H stretch region. The extension of the force field to large organic molecules, such as polymers and proteins, are in progress. On the other hand, in the calculation of adsorption properties in porous materials, we developed a a systematic way to generate the intermolecular force field through ab initio calculations. All these non-empirical force field allows us to make objective interpretation of experimental phenomenon.

Microscopic Investigation of the Electrode-electrolyte Interface

In modern battery systems, the interface structures play an important role when intercalation/deintercalation process of ions occur at electrode-electrolyte interfaces. However, the detection of such buried interface is quite difficult. Recent sum frequency generation (SFG) spectroscopy has been proved to be able to detect the interface structure due to its high surface sensitivity and selectivity. However, the observed SFG spectra are often difficult to be interpreted in order to extract the detailed structure information of interface. In this case, we performed molecular dynamics (MD) simulation and collaborated with the surface sensitive SFG spectroscopy to investigate electrode-electrolyte interfaces. The experimental SFG spectra provide valuable information to validate our molecular model. Meanwhile, the MD simulation results enable us to analyze the detailed microscopic information of interface, including the layer structure of electrolyte, the molecular orientation of each layer and so on. This work will generate a detailed understanding of electrode-electrolyte interface in a molecular point of view.

Computational Study of Organic Interface and Monolayer Interface

Sum frequency generation (SFG) spectroscopy has been used to investigate the organic interfaces, such as Langmuir monolayer systems. The interpretation of C-H stretch region provides rich information of interface structures of organic molecules, as C-H stretch widely exists in organic systems. However, the interpretations of observed C-H stretch SFG spectra are often confused, because of the crowded vibrational bands in this region. On the other hand, the theoretical study of structure and SFG spectra at organic interface is still less of investigation due to a lack of an accurate molecular model, especially for C-H stretching region. How to generate an accurate and polarizable force field is the key problem. For these purposes, we developed a general polarizable force field and applied it to investigate various organic interface, such as ethanol, acetonitrile and Langmuir monolayer. This work aims at generate quantitative relationship between observed SFG spectra and the analyzed molecular orientation through molecular simulations. This work will realize the theoretical analysis of complicated C-H stretching vibrational spectra and further guide to the investigation of polymer and membrane systems.

Prediction of Physical Chemistry Properties of Porous Materials

The porous materials, such as metal organic frameworks (MOFs), covalent organic frameworks (COFs) and zeolite, have been widely applied in industrial adsorption, separation and catalysis processes. How to design the materials with desired properties is the key problem in applications. Using molecular simulation method, it is possible to theoretically predict various properties of functional materials before synthesis. The results will help to provide guidance and reduce the cost in the development of new materials. In order to perform objective prediction, the accurate and non-empirical molecular modeling is of great importance. For this purpose, a systematic method to predict the adsorption based on ab initio force field was proposed. The proposed method has been successfully applied to predict the adsorption, separation and ion exchange properties of porous materials, such as MOFs, COFs, and zeolite. Meanwhile, we are now in collaboration with experimental studies to developed new functional COF materials with potential applications in adsorption and supercapacitor.