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

Download:

Current browse context:

cond-mat

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: Percolation in the Harmonic Crystal and Voter Model in three dimensions

Abstract: We investigate the site percolation transition in two strongly correlated systems in three dimensions: the massless harmonic crystal and the voter model. In the first case we start with a Gibbs measure for the potential, $U=\frac{J}{2} \sum_{<x,y>} (\phi(x) - \phi(y))^2$, $x,y \in \mathbb{Z}^3$, $J > 0$ and $\phi(x) \in \mathbb{R}$, a scalar height variable, and define occupation variables $\rho_h(x) =1,(0)$ for $\phi(x) > h (<h)$. The probability $p$ of a site being occupied, is then a function of $h$. In the voter model we consider the stationary measure, in which each site is either occupied or empty, with probability $p$. In both cases the truncated pair correlation of the occupation variables, $G(x-y)$, decays asymptotically like $|x-y|^{-1}$. Using some novel Monte Carlo simulation methods and finite size scaling we find accurate values of $p_c$ as well as the critical exponents for these systems. The latter are different from that of independent percolation in $d=3$, as expected from the work of Weinrib and Halperin [WH] for the percolation transition of systems with $G(r) \sim r^{-a}$ [A. Weinrib and B. Halperin, Phys. Rev. B 27, 413 (1983)]. In particular the correlation length exponent $\nu$ is very close to the predicted value of 2 supporting the conjecture by WH that $\nu= \frac{2}{a}$ is exact.
Comments: 8 figures. new version significantly different from the old one, includes new results, figures etc
Subjects: Statistical Mechanics (cond-mat.stat-mech)
DOI: 10.1103/PhysRevE.74.031120
Cite as: arXiv:cond-mat/0601108 [cond-mat.stat-mech]
  (or arXiv:cond-mat/0601108v3 [cond-mat.stat-mech] for this version)

Submission history

From: Vesselin Marinov [view email]
[v1] Fri, 6 Jan 2006 07:31:30 GMT (24kb)
[v2] Sat, 7 Jan 2006 20:31:23 GMT (25kb)
[v3] Wed, 19 Jul 2006 20:52:41 GMT (28kb)

Link back to: arXiv, form interface, contact.