Energy, temperature, and heat capacity in discrete classical dynamics

Søren Toxvaerd
DOI: https://doi.org/10.1103/physreve.109.015306
IF: 2.707
2024-01-25
Physical Review E
Abstract:Simulations of objects with classical dynamics are in fact a particular version of discrete dynamics, since almost all the classical dynamics simulations in natural science are performed with the use of the simple "leapfrog" or "Verlet" algorithm. It was, however, Newton who in Principia , Proposition I in 1687 first formulated the discrete algorithm, which much later in 1967 was rederived by L. Verlet. Verlet also formulated a first-order approximation for the velocity v(t) at time t , which has been used in simulations since then. The approximated expressions for v(t) and the kinetic energy lead to severe errors in the thermodynamics at high densities, temperatures, strong repulsive forces, or for large discrete time increments used in discrete "molecular dynamics" (MD) simulations. Here we derive the exact expressions for the discrete dynamics, and show by simulations of a Lennard-Jones system that these expressions now result in equality between temperatures determined from the kinetic energies and the corresponding configurational temperatures determined from the expresssion of Landau and Lifshitz, derived from the forces. https://doi.org/10.1103/PhysRevE.109.015306 ©2024 American Physical Society
physics, fluids & plasmas, mathematical
What problem does this paper attempt to address?