Discrete Molecular Dynamics

Abstract

Computer simulation of the time evolution in a classical system is a standard numerical method, used in numerous scientific articles in Natural Science. Almost all the simulations are performed by discrete Molecular Dynamics (MD). The algorithm used in MD was originally formulated by I. Newton at the beginning of his book Principia. Newton׳s discrete dynamics is exact in the same sense as Newton׳s analytic counterpart Classical Mechanics. Both dynamics are time-reversible, symplectic, and have the same dynamic invariances. There is no qualitative difference between the two kinds of dynamics. This is due to the fact, that there exists a ׳׳shadow Hamiltonian׳׳ nearby the Hamiltonian H(q,p) for the analytic dynamics, and where its dynamics can be obtained by an asymptotic expansion from H(q,p), and where the positions generated by MD are located on the analytic trajectories for the shadow Hamiltonian.It is only possible to obtain the solution of Newton׳s classical differential equations for a few simple systems, but the exact discrete Newtonian dynamics can be obtained for almost all complex classical systems. Some examples are given here: The emergence and evolution of a planetary system. The emergence and evolution of planetary systems with inverse forces. The emergence and evolution of galaxies in the expanding Universe.The fact that there exist two equally valid formulations of classical dynamics raises the question: What is the classical limit of quantum mechanics? Discrete molecular dynamics is mathematically different from analytic dynamics. The Heisenberg uncertainty between the concurrent values of positions and momenta is an inherent property of the discrete dynamics, but the analytic quantum electrodynamics (QED) is in all manner fully appropriate, and there is a lack of justification for preferring discrete quantum mechanics.