A relativistic canonical symplectic particlein-cell method for energetic plasma analysis

A relativistic canonical symplectic particle-in-cell(RCSPIC)method for simulating energetic plasma processes is established.By use of the Hamiltonian for the relativistic Vlasov-Maxwell system,we obtain a discrete relativistic canonical Hamiltonian dynamical system,based on which the RCSPIC method is constructed by applying the symplectic temporal discrete method.Through a 106-step numerical test,the RCSPIC method is proven to possess long-term energy stability.The ability to calculate energetic plasma processes is shown by simulations of the reflection processes of a high-energy laser(1?×?1020 W cm-2)on the plasma edge.

Sorption and permeation of gaseous molecules in amorphous and crystalline PPX C membranes: molecular dynamics and grand canonical Monte Carlo simulation studies

Amorphous and crystalline poly （chloro-p-xytylene） （PPX C） membranes are constructed by using a novel com- putational technique, that is, a combined method of NVT＋NPT-molecular dynamics （MD） and gradually reducing the size （GRS） methods. The related free volumes are defined as homology clusters. Then the sorption and the permeation of gases in PPX C polymers are studied using grand canonical Monte Carlo （GCMC） and NVT-MD methods. The results show that the crystalline PPX C membranes provide smaller free volumes for absorbing or transferring gases relative to the amorphous PPX C area. The gas sorption in PPX C membranes mainly belongs to the physical one, and H bonds can appear obviously in the amorphous area. By cluster analyzing on the mean square displacement of gases, we find that gases walk along the x axis in the crystalline area and walk randomly in the amorphous area. The calculated permeability coefficients are close to the experimental data.

Multiscale Simulation Starting at the Molecular Level for Adsorption Process Development

This article presents a multiscale simulation approach starting at the molecular level for the adsorption process development. A grand canonical Monte Carlo method is used for the prediction of adsorption isotherms of methanol on an activated carbon at the molecular level. The adsorption isotherms obtained in the linear region （or adsorption constant） are exploited as a model parameter required for the adsorption process simulation. The adsorption process model described by a set of partial differential equations （PDEs） is solved by using the conservation element and solution element method, which produces a fast and an accurate numerical solution to PDEs. The simulation results obtained from the adsorption constant estimated at the molecular level are in good agreement with the experimental results of the pulse response. The systematical multiscale simulation approach addressed in this study may be useful to accelerate the adsorption process development by reducing the number of experiments.

Improved Canonical Quantization Method of Self Dual Field

In this paper,the improved canonical quantization method of the self dual field is given in order to overcome linear combination problem about the second class constraint and the first class constraint number maximization problem in the Dirac method.In the improved canonical quantization method,there are no artificial linear combination and the first class constraint number maximization problems,at the same time,the stability of the system is considered.Therefore,the improved canonical quantization method is more natural and easier accepted by people than the usual Dirac method.We use the improved canonical quantization method to realize the canonical quantization of the self dual field,which has relation with string theory successfully and the results are equal to the results by using the Dirac method.

Stability and Convergence of the Canonical Euler Splitting Method for Nonlinear Composite Stiff Functional Differential-Algebraic Equations

A novel canonical Euler splitting method is proposed for nonlinear compositestiff functional differential-algebraic equations, the stability and convergence of themethod is evidenced, theoretical results are further confirmed by some numerical experiments.Especially, the numerical method and its theories can be applied to specialcases, such as delay differential-algebraic equations and integral differential-algebraicequations.