The random cluster model on a general graph and a phase transition characterization of nonamenability

The random cluster model on a general infinite graph with bounded degree wired at infinity is considered and a "ghost vertex" method is introduced in order to explicitly construct random cluster measures satisfying the Dobrushin-Lanford-Ruelle condition for q[greater-or-equal, slanted]1. It is proved that on a regular nonamenable graph there is a q0 such that for q[greater-or-equal, slanted]q0 there is a phase transition for an entire interval of values of p, whereas on a quasi-transitive amenable graph there is a phase transition for at most a countable number of values of p. In particular, a transitive graph is nonamenable if and only if there is a phase transition for an entire interval of p-values for large enough q. It is also observed that these results have a Potts model interpretation. In particular, a transitive graph is nonamenable if and only if the q-state Potts model on that graph has the property that for q large enough there is an entire interval of temperatures for which the free Gibbs state is not a convex combination of the q Gibbs states obtained from one-spin boundary conditions. It is also proved that on the regular tree, , with q[greater-or-equal, slanted]1 and p close enough to 1, there is unique random cluster measure despite the presence of more than one infinite cluster. This partly proves Conjecture 1.9 of H.

[1]  T. Liggett.  Interacting Particle Systems , 1985 .

[2]  M. Aizenman,et al.  Sharpness of the phase transition in percolation models , 1987 .

[3]  Harry Kesten,et al.  Full Banach Mean Values on Countable groups. , 1959 .

[4]  H. Kesten.  Percolation theory for mathematicians , 1982 .

[5]  R. Ellis,et al.  Large deviations and statistical mechanics , 1985 .

[6]  R. Burton,et al.  Density and uniqueness in percolation , 1989 .

[7]  Russell Lyons,et al.  Group-invariant Percolation on Graphs , 1999 .

[8]  R. Holley.  Remarks on the FKG inequalities , 1974 .

[9]  Scaling properties for the unscreened surfaces of fractal patterns. , 1987, Physical review letters.

[10]  C. Fortuin,et al.  On the random-cluster model II. The percolation model , 1972 .

[11]  David Ruelle,et al.  On the use of “small external fields” in the problem of symmetry breakdown in statistical mechanics☆ , 1972 .

[12]  J. Steif,et al.  Amenability and Phase Transition in the Ising Model , 1997 .

[13]  C. Wu.  Ising models on hyperbolic graphs , 1996 .

[14]  Robijn Bruinsma,et al.  Soft order in physical systems , 1994 .

[15]  Olle Häggström,et al.  Infinite clusters in dependent automorphism invariant percolation on trees , 1997 .

[16]  Olle Häggström,et al.  Monotonicity of uniqueness for percolation on Cayley graphs: all infinite clusters are born simultaneously , 1999 .

[17]  I. Benjamini,et al.  Percolation Beyond $Z^d$, Many Questions And a Few Answers , 1996 .

[18]  C. Fortuin,et al.  On the random-cluster model: I. Introduction and relation to other models , 1972 .

[19]  A. Martin-Löf,et al.  On the uniqueness of the equilibrium state for Ising spin systems , 1972 .

[20]  Charles M. Newman,et al.  Percolation in ∞ + 1 dimensions , 1990 .

[21]  Peter Gerl.  Random walks on graphs with a strong isoperimetric property , 1988 .

[22]  R. Ellis,et al.  Entropy, large deviations, and statistical mechanics , 1985 .

[23]  Harry Kesten,et al.  Symmetric random walks on groups , 1959 .

[24]  T. Lindvall.  Lectures on the Coupling Method , 1992 .

[25]  Wang,et al.  Nonuniversal critical dynamics in Monte Carlo simulations. , 1987, Physical review letters.

[26]  Olle Häggström,et al.  The random-cluster model on a homogeneous tree , 1996 .

[27]  G. Grimmett.  The Stochastic Random-Cluster Process and the Uniqueness of Random-Cluster Measures , 1995 .