Molecular dynamics stochastic difference equation

All these different mechanisms of mass transport through a porous medium can be studied experimentally and theoretically through classical models (Darcy s law, Knudsen diffusion, molecular dynamics, Stefan-Maxwell equations, dusty-gas model etc.) which can be coupled or not with the interactions or even reactions between the solid structure and the fluid elements. Another method for the analysis of the species motion inside a porous structure can be based on the observation that the motion occurs as a result of two or more elementary evolutions that are randomly connected. This is the stochastic way for the analysis of species motion inside a porous body. Some examples that will be analysed here by the stochastic method are the result of the particularisations of the cases presented with the development of stochastic models in Sections 4.4 and 4.5. 

Stochastic dynamics The stochastic dynamics (SD) method is a further extension of the original molecular dynamics method. A space-time trajectory of a molecular system is generated by integration of the stochastic Langevin equation which differs from the simple molecular dynamics equation by the addition of a stochastic force R and a frictional force proportional to a friction coefficient g. The SD approach is useful for the description of slow processes such as diffusion, the simulation of electrolyte solutions, and various solvent effects. 

Poisson-Boltzmann equation (PBE), multigrid (MG), algebraic multigrid (AMG), finite difference (FD), finite element (FE), Gauss-Seidel (GS), conjugate gradient (CG), successive over relaxation (SOR), stochastic dynamics (SD), quantum mechanics (QM), molecular mechanics (MM), molecular dynamics u, (MD). 

Since the CW-ESR spectra provides structural information and dynamics at different time scales, proper account of fast and slow motion of the labeled molecules is required for correct reproduction of the spectra. While the fast motion can be derived from a fast-motional perturbative model, in the slow-motion regime the effects on the spin relaxation processes exerted by the molecular motions requires a more sophisticated theoretical approach. The calculation of rotational diffusion in solution can be tackled by solving the stochastic Liouville equation (SLE) or by longtime-scale molecular dynamics simulations [94—96]. 

This paper has focused on two recent computer methods for discrete simulation of chemical kinetics. Beginning with the realization that truly microscopic computer experiments are not at all feasible, I have tried to motivate the development of a hierarchy of simulations in studies of a class of chemical problems which best illustrate the absolute necessity for simulation at levels above molecular dynamics. It is anticipated (optimistically ) that the parallel development of discrete event simulations at different levels of description may ultimately provide a practical interface between microscopic physics and macroscopic chemistry in complex physicochemical systems. With the addition to microscopic molecular dynamics of successively higher-level simulations intermediate between molecular dynamics at one extreme and differential equations at the other, it should be possible to examine explicitly the validity of assumptions invoked at each stage in passing from the molecular level to the stochastic description and finally to the macroscopic formulation of chemical reaction kinetics. 

The pure state of the molecular environment is never precisely known. Since the dynamics of the molecular pure state depends on this unknown environment s (pure) state, one gets a stochastic dynamics for the molecular pure states. For a two-level system-with its pure states describable by a Bloch sphere—the situation is illustrated in Fig. 8. The dynamics of some given molecular initial state is governed not only by the Schrodinger equation, but also by external influence. Depending on the (pure) state of the environment, we reach different final molecular states. Usually only probabilistic predictions can be given (and no information about the precise trajectory of pure states, i.e., the trajectory on the Bloch sphere). 

Note that a classification of the surface reaction mechanisms can be done either on the base of the nature of limiting stages, or on the base of dynamical models of elementary acts. The first way of classification is conditional, depending strongly on the relative values of different terms in the equations of chemical kinetics (6.1.19) or (6.3.1). Classification on the base of dynamical models (non-adiabatic, adiabatic, collineai-, impact, stochastic, etc.) needs the detailed study of the physical nature of reactive interactions. Such study is at the very beginning now both in theoretical and experimental (molecular beams) directions, and it should lead to detailed information on mechanisms of surface reactions. 

---