We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cond-mat.stat-mech

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo

Condensed Matter > Statistical Mechanics

Title: Examination of Boltzmann's H-Function: Dimensionality and Interaction Sensitivity Dependence, and a comment on his H-Theorem

Abstract: Boltzmann's H-Theorem, formulated 150 years ago in terms of H-function that also bears his name, is one of the most celebrated theorems of science and paved the way for the development of nonequilibrium statistical mechanics. Nevertheless, quantitative studies of the H-function, denoted by H(t), in realistic systems are relatively scarce because of the difficulty of obtaining the time-dependent momentum distribution analytically. Also, the earlier attempts proceeded through the solution of Boltzmann's kinetic equation, which was hard. Here we investigate, by direct molecular dynamics simulations and analytic theory, the time dependence of H(t). We probe the sensitivity of nonequilibrium relaxation to interaction potential and dimensionality by using the H-function H(t). We evaluate H(t) for three different potentials in all three dimensions and find that it exhibits surprisingly strong sensitivity to these factors. The relaxation of H(t) is long in 1D, but short in 3D. We obtain, for the first time, a closed-form analytic expression for H(t) using the solution of the Fokker-Planck equation for the velocity space probability distribution and compare its predictions with the simulation results. Interestingly, H(t) is found to exhibit linear response when vastly different initial nonequilibrium conditions are employed. The oft-quoted relation of H-function with Clausius's entropy theorem is discussed.
Subjects: Statistical Mechanics (cond-mat.stat-mech)
Cite as: arXiv:2207.00805 [cond-mat.stat-mech]
  (or arXiv:2207.00805v1 [cond-mat.stat-mech] for this version)

Submission history

From: Biman Bagchi - [view email]
[v1] Sat, 2 Jul 2022 11:37:01 GMT (1330kb)

Link back to: arXiv, form interface, contact.