English

Phase coexistence of gradient Gibbs states

Probability 2011-11-10 v2 Mathematical Physics math.MP

Abstract

We consider the (scalar) gradient fields η=(ηb)\eta=(\eta_b)--with bb denoting the nearest-neighbor edges in Z2\Z^2--that are distributed according to the Gibbs measure proportional to \texteβH(η)ν(\textdη)\texte^{-\beta H(\eta)}\nu(\textd\eta). Here H=bV(ηb)H=\sum_bV(\eta_b) is the Hamiltonian, VV is a symmetric potential, β>0\beta>0 is the inverse temperature, and ν\nu is the Lebesgue measure on the linear space defined by imposing the loop condition ηb1+ηb2=ηb3+ηb4\eta_{b_1}+\eta_{b_2}=\eta_{b_3}+\eta_{b_4} for each plaquette (b1,b2,b3,b4)(b_1,b_2,b_3,b_4) in Z2\Z^2. For convex VV, Funaki and Spohn have shown that ergodic infinite-volume Gibbs measures are characterized by their tilt. We describe a mechanism by which the gradient Gibbs measures with non-convex VV undergo a structural, order-disorder phase transition at some intermediate value of inverse temperature β\beta. At the transition point, there are at least two distinct gradient measures with zero tilt, i.e., Eηb=0E \eta_b=0.

Keywords

Cite

@article{arxiv.math/0512502,
  title  = {Phase coexistence of gradient Gibbs states},
  author = {Marek Biskup and Roman Kotecky},
  journal= {arXiv preprint arXiv:math/0512502},
  year   = {2011}
}

Comments

3 figs, PTRF style files included

R2 v1 2026-07-22T17:29:01.669Z