Unorthodox properties of critical clusters

TL;DR

This paper explores the unconventional properties of critical clusters in thermal systems, revealing their nonextensive and noncanonical behavior, and linking static and dynamic features to q-statistics and aging phenomena.

Contribution

It demonstrates that critical cluster properties deviate from Boltzmann-Gibbs statistics, showing nonextensivity, intermittent dynamics, and aging, thus extending the understanding of critical phenomena beyond traditional frameworks.

Findings

Growth of order parameter suggests nonextensivity of BG entropy

Order parameter dynamics exhibit intermittency and aging

Crossover from canonical to q-statistics near criticality

Abstract

We look at the properties of clusters of order parameter at critical points in thermal systems and consider their significance to statistical-mechanical ground rules. These properties have been previously obtained through the saddle-point approximation in a coarse-grained partition function. We examine both static and dynamical aspects of a single large cluster and indicate that these properties fall outside the canonical Boltzmann-Gibbs (BG) scheme. Specifically: 1) The faster than exponential growth with cluster size of the space-integrated order parameter suggests nonextensivity of the BG entropy but extensivity of a q-entropy expression. 2) The finding that the time evolution of the order parameter is described by the dynamics of an intermittent nonlinear map implies an atypical sensitivity to initial conditions compatible with q-statistics and displays an 'aging' scaling property. 3) Both, the approach to criticality and the infinite-size cluster limit at criticality manifest through a crossover from canonical to q-statistics and we discuss the nonuniform convergence associated to these features. Key words: critical clusters, saddle-point approximation, q-entropy, intermittency, aging