Variable Stoichiometry And Transition State Structure Prediction Methods And Applications

As an important variable about physical properties, chemical composition of material provides a new dimension for the search of new functional materials. The determination of structure is a key problem to develop new materials design, such as tungsten boride. Due to its character of superhard and high melting point, it is becoming a hot spot of research. Theory study has found that it has 6 different components W2B, WB, WB2, WB4, WB3 and W2B5, as volunteers of new functional materials. Due to samples containing different compositions, it is difficult for experiment to determine the structure of the tungsten boride. So theory research is very important. Now a lot of structure prediction methods are developed. But most of the methods can only predict fixed composition structures. It need run many times, with large amount of calculation. So developing variable stoichiometry structure prediction is the key to solve this problem. In addition, the present structure prediction methods often only give the thermodynamic stability of the structure (ground state and metastable structures), without the information of experimental synthesis. So it is a key problem to develop a reaction path search method for the realization of new functional material design. Early reaction path search method mainly focuses on adsorption, molecule and doping, etc. A simple chemical reaction process mainly focuses on reaction path optimization. But this approach requires artificial guess at the beginning of the optimization of reaction path. So a correct simulation need is correct chemical intuition to guess reaction path at the beginning. But for complex system, basic human intuition is impossible to build a correct chemical reaction path. Therefore, search the reaction path is also a problem to be solved. In this paper, on the basis of CALYPSO structure prediction method, we developed variable composition structure prediction method and transition state search method. We tested our method on known and unknown system, and obtained meaningful innovation work:1. Considering the change of the material chemical composition under extreme conditions, we developed the CALYPSO variable stoichiometry structure prediction methods. We found that our method is reliable and stable after testing the system. Because of the uncertainty of the chemical formula, so we need a certain range of the various components structure prediction, in order to find the most stable chemical composition. Due to early CALYPSO algorithm aimed at fixed chemical composition structure prediction, therefore structure evolution and the processing of the results cannot be directly applied to the study of variable stoichiometry structure prediction. We select known lithium hydrogen system as the benchmark. Successful search to the chemical composition and structure of the most stable proves the effectiveness of our method.2. We used CALYPSO variable stoichiometry structure prediction method and studied phase transition of hydrogen and nitrogen system under high pressure. We found that at about 30 GPa ammonia and nitrogen gas is rich in chemical composition. One can synthesize a kind of layered conductive N2H high energy density materials. Energy density of the material has 4.4 KJ/g which similar to the energy density of the TNT. It’s a potential high energy density material.3. With advantages of CALYPSO in the field of structure prediction matrix particle swarm optimization algorithm is applied to the reaction path search. It mainly includes the representation and evolution of the path in the reaction. We put forward the form of matrix for the reaction path in the characterization, and use particle swarm optimization algorithm search to evolving reaction path. At the same time, we put forward another efficient optimize line method to optimize the reaction path. At the same time, we study the phase transition mechanism from graphite into diamond, wurtzite to rocksalt structure transformation of gallium nitride. We explored the silicon phase transition barrier between high pressure phases, and solved the confusion of experiment.