A statistical physics of stationary and metastable states: description of the plasma column experimental data

TL;DR

This paper develops a modified statistical mechanics framework incorporating an additional conserved constraint to describe stationary and metastable states, successfully applied to plasma column data and predicting density profiles with high accuracy.

Contribution

It introduces a new approach with a conserved constraint function to extend statistical mechanics to metastable states, applied to plasma data.

Findings

Accurately predicts plasma density profiles at all radial positions.

Successfully models the smooth tail of the experimental distribution.

Avoids non-analyticities present in previous methods.

Abstract

We propose a statistical mechanics for a general class of stationary and metastable equilibrium states. For this purpose, the Gibbs extremal conditions are slightly modified in order to be applied to a wide class of non-equilibrium states. As usual, it is assumed that the system maximizes the entropy functional $S$, subjected to the standard conditions; i.e., constant energy and normalization of the probability distribution. However, an extra conserved constraint function $F$ is also assumed to exist, which forces the system to remain in the metastable configuration. Further, after assuming additivity for two quasi-independent subsystems, and that the new constraint commutes with density matrix $\rho$, it is argued that F should be an homogeneous function of the density matrix, at least for systems in which the spectrum is sufficiently dense to be considered as continuous. The explicit form of $F$ turns to be $F(p_{i})=p_{i}^{q}$, where $p_i$ are the eigenvalues of the density matrix and $q$ is a real number to be determined. This $q$ number appears as a kind of Tsallis parameter having the interpretation of the order of homogeneity of the constraint $F$. The procedure is applied to describe the results of the plasma experiment of Huang and Driscoll. The experimentally measured density is predicted with a similar precision as it is done with the use of the extremum of the enstrophy and Tsallis procedures. However, the present results define the density at all the radial positions. In particular, the smooth tail shown by the experimental distribution turns to be predicted by the procedure. In this way, the scheme avoids the non-analyticity of the density profile at large distances arising in both of the mentioned alternative procedures.