Numerical studies for an ab initio investigation into the Boltzmann prescription in statistical mechanics of large systems

TL;DR

This paper uses large-scale numerical simulations to examine whether the Boltzmann entropy is extensive in large systems, providing initial insights into non-extensive statistical mechanics through ab initio methods.

Contribution

It introduces a novel numerical approach using Monte Carlo algorithms to investigate the extensivity of Boltzmann entropy in large Hamiltonian systems from first principles.

Findings

Simulations considered over a billion states but are still too small to definitively settle the question.

The approach offers a new way to explore non-extensive statistical mechanics.

Provides theoretical context to support numerical findings.

Abstract

We present numerical investigations into the question of the validity of the Boltzmann prescription in Statistical Mechanics for large systems, addressing the issue of whether extensivity of energy implies the extensivity of the Boltzmann entropy. The importance of the question stems from the fact that it is currently considered open by some investigators but quite settled by others. We report ab initio results for gas-like Hamiltonian systems with long-range as well as short-range interactions, based on simulations that explicitly consider more than $2^{30} \approx 10^9$ states of the full Hilbert space. The basis of the technique is Monte Carlo algorithms. Despite the largeness of the numbers used, careful inspection shows that the systems studied are still too small to settle uniquely the issues raised. Therefore, the new approach outlined represents a first step in addressing on first principles the question of non-extensive statistical mechanics. General theoretical comments are also supplied to supplement the numerical investigations.