The unusual thermodynamic properties of the Ising antiferromagnet supplemented with a ferromagnetic, mean-field term are outlined. This simple model is inspired by more realistic models of spin-crossover materials. The phase diagram is estimated usin
g Metropolis Monte Carlo methods, and differences with preliminary Wang-Landau Monte Carlo results for small systems are noted.
The dynamics of desorption from a submonolayer of adsorbed atoms or ions are significantly influenced by the absence or presence of lateral diffusion of the adsorbed particles. When diffusion is present, the adsorbate configuration is simultaneously
changed by two distinct processes, proceeding in parallel: adsorption/desorption, which changes the total adsorbate coverage, and lateral diffusion, which is coverage conserving. Inspired by experimental results, we here study the effects of these competing processes by kinetic Monte Carlo simulations of a simple lattice-gas model. In order to untangle the various effects, we perform large-scale simulations, in which we monitor coverage, correlation length, and cluster-size distributions, as well as the behavior of representative individual clusters, during desorption. For each initial adsorbate configuration, we perform multiple, independent simulations, without and with diffusion, respectively. We find that, compared to desorption without diffusion, the coverage-conserving diffusion process produces two competing effects: a retardation of the desorption rate, which is associated with a coarsening of the adsorbate configuration, and an acceleration due to desorption of monomers evaporated from the cluster perimeters. The balance between these two effects is governed by the structure of the adsorbate layer at the beginning of the desorption process. Deceleration and coarsening are predominant for configurations dominated by monomers and small clusters, while acceleration is predominant for configurations dominated by large clusters.
Many partitioning methods may be used to partition a network into smaller clusters while minimizing the number of cuts needed. However, other considerations must also be taken into account when a network represents a real system such as a power grid.
In this paper we use a simulated annealing Monte Carlo (MC) method to optimize initial clusters on the Florida high-voltage power-grid network that were formed by associating each load with its closest generator. The clusters are optimized to maximize internal connectivity within the individual clusters and minimize the power deficiency or surplus that clusters may otherwise have.
