Single Atom Statistical Model Of Thermal Atomic Reaction

Thermal motions of atoms and molecules may be the most common process of the nature, which dominates the evolution of physical and chemical properties of all the kinds of objects. Therefore, theoretically predicting the evolution rate is of great scientific and technological significance, concerning understanding and designing common (thermal) chemical reactions, the growing process and service process of materials, atomic transport process of livings and the evolution of global climate, and so on. In 1889, according to the experimental phenomena, a Swedish chemist founded the so called Arrhenius law to describe the rate of thermal and chemical reaction, and in the middle of last century scientists established Transition State Theory (TST), which greatly promotes the ability of predicting the reaction process of thermal atoms. However, obvious deviations were observed between theoretical prediction and experimental results in many systems in recent years. Although great efforts have been made to address the problem, not yet comes an ideal theoretical method with simple operation.This thesis takes the lead in confirming that atoms in actual gas and condensed matter have the same Maxwell velocity distribution as in ideal gas. Furthermore, we setup a theoretical model to predict the reaction process of thermal atoms based on single atom statistics. The model works so simply that the prediction can be performed on ab initio level without any experience parameters. In order to compare with the existing theoretical model, the new model was applied to investigate the following processes:1. The self-diffusion of Pt, Cu and Ar on the crystal surface:Firstly, the self-diffusion processes were vast simulated by classical Molecular Dynamics (MD), and the dependence of reaction rate R upon temperature T (R-T curve) was obtained through statistical analysis. The comparison between the R-T curve and the prediction of the theoretical models shows that our new model is significantly superior to the conventional TST and reveals that the reaction barrier in the conventional TST is obvious different from the actual one.2. Bimolecular reaction (H2+CN=> HCN+H) and C60 isomerization:Our new model and the TST, were applied to fit the experimental R-T curve of H2+CN(?)HCN+H reaction, respectively, and the new model produces a reaction barrier of 0.27eV, while the TST produces 0.06eV, which is far from the theoretical one,0.27eV, obtained from quantum mechanics calculation. On the isomerization of C60 molecule, a lot of MD simulations were performed to get the R-T curve, which was compared with the new model and the TST, showing that the new model is much better than the TST.3. Ludwig-Soret effect:Ludwig (in 1856) and Soret (in 1879) found that impurity atoms in solids or in melts tend to gather in low temperature area due to temperature gradient field, and this effect is being widely used in isotope separation, atom transport in living cells and many other research fields. However, the researchers are facing with a series of puzzles because no theoretical model of the effect has been set up on the atomic level. For example, it is even difficult to define the Soret coefficient for a given material. The present work can deduce the Ludwig-Soret law on atomic level from the new model, and the expression of the Soret coefficient is tested by MD simulation of single atom diffusion along a carbon chain with a temperature gradient.