Restricted ensemble Monte Carlo

To obtain the complete phase behavior of supercooled silicon Vasisht et al. analyzed the interplay of various loci of extremal behavior, namely, the spinodal, temperature of density extrema, and temperature of compressibility extrema. Loci of temperature of maximum and minimum density (TMD and TMinD), temperature of maximum and minimum compressibility TMC and TMinC), and spinodal were evaluated by employing, in addition to the MD simulations, parallel tempering (PT) Monte Carlo simulations (at low temperature and pressures) and restricted ensemble Monte Carlo (REMC) simulations [95] (for locating the... 

Orkoulas G and Panagiotopoulos A Z 1999 Phase behavior of the restricted primitive model and square-well fluids from Monte Carlo simulations in the grand canonical ensemble J. Chem. Phys. 110 1581... 

There are basically two different computer simulation techniques known as molecular dynamics (MD) and Monte Carlo (MC) simulation. In MD molecular trajectories are computed by solving an equation of motion for equilibrium or nonequilibrium situations. Since the MD time scale is a physical one, this method permits investigations of time-dependent phenomena like, for example, transport processes [25,61-63]. In MC, on the other hand, trajectories are generated by a (biased) random walk in configuration space and, therefore, do not per se permit investigations of processes on a physical time scale (with the dynamics of spin lattices as an exception [64]). However, MC has the advantage that it can easily be applied to virtually all statistical-physical ensembles, which is of particular interest in the context of this chapter. On account of limitations of space and because excellent texts exist for the MD method [25,61-63,65], the present discussion will be restricted to the MC technique with particular emphasis on mixed stress-strain ensembles. 

There are two basic approaches to the computer simulation of liquid crystals, the Monte Carlo method and the method known as molecular dynamics. We will first discuss the basis of the Monte Carlo method. As is the case with both these methods, a small number (of the order hundreds) of molecules is considered and the difficulties introduced by this restriction are, at least in part, removed by the use of artful boundary conditions which will be discussed below. This relatively small assembly of molecules is treated by a method based on the canonical partition function approach. That is to say, the energy which appears in the Boltzman factor is the total energy of the assembly and such factors are assumed summed over an ensemble of assemblies. The summation ranges over all the coordinates and momenta which describe the assemblies. As a classical approach is taken to the problem, the summation is replaced by an integration over all these coordinates though, in the final computation, a return to a summation has to be made. If one wishes to find the probable value of some particular physical quantity, A, which is a function of the coordinates just referred to, then statistical mechanics teaches that this quantity is given by... 

Monte Carlo C) simulations [91] are employed to equilibrate the system at very low temperature and high negative pressme (while deducing the temperature of minimum density) and restricted ensemble MC simulations for locating the spinodal at low temperatures. 

If bottlenecks restrict intramolecular vibrational energy redistribution," the unimolecular dissociation is not random and not in accord with equation (4). There is considerable interest in identifying which unimolecular reactions do not obey equation (4). In this section Monte Carlo sampling schemes are described for exciting A randomly with a micro-canonical ensemble of states and nonrandomly with mode selective excitation. For pedagogical purposes, selecting a microcanonical ensemble for a normal mode Hamiltonian is described first. 

This selection process is then iterated, beginning from an initial state of the system, as defined by species populations, to simulate a chemical evolution. A statistical ensemble is generated by repeated simulation of the chemical evolution using different sequences of random numbers in the Monte Carlo selection process. Within limits imposed by computer time restrictions, ensemble population averages and relevant statistical information can be evaluated to any desired degree of accuracy. In particular, reliable values for the first several moments of the distribution can be obtained both inexpensively and efficiently via a computer algorithm which is incredibly easy to implement (21, 22), especially in comparison to now-standard techniques foF soTving the stiff ordinary differential equations (48, 49) which may arise in the deterministic description of chemical kinetics (53). Now consider briefly the essential features of a simple chemical model which illustrates well the attributes of stochastic chemical simulations. 

---