Statics and dynamics of the ten-state mean-field Potts glass model: a Monte Carlo study

摘要: We investigate by means of Monte Carlo simulations the fully connected p-state Potts model for different system sizes in order to see how static and dynamic properties a finite compare with the, exactly known, behaviour thermodynamic limit. Using p = 10 we are able study equilibrium dynamics as large N 2560. find that quantities, such energy, entropy, spin glass susceptibility well distribution parameter P(q) show very strong finite-size effects. From calculate fourth-order cumulant g4(N,T) Guerra G(N,T) these quantities cannot be used locate transition temperature investigated. Also spin-autocorrelation function C(t) shows effects it does not plateau even temperatures around dynamical critical TD. dependence on T α-relaxation time can understood scaling ansatz. obey time-temperature superposition principle TD, but so significantly lower T. Finally relaxation individual spins their depends strongly chosen spin, i.e. is dynamically heterogeneous, which explains non-exponentiality C(t).