Determination of the free energy of water at room temperature by parallel grand canonical Monte Carlo

Due to the practical importance of the free energy and entropy of a system, many attempts have been made to calculate them using molecular dynamics or Monte Carlo simulations. In particular, by applying the standard Grand Canonical Monte Carlo simulation with the Histogram Reweighting technique it is possible, in principle, to estimate the gland partition function and hence the statistical properties of the system. Nevertheless, due to the hydrogen-bonded structure of water at room temperature, the probability of adding and removing molecules, becomes prohibitively small. Even with the use of standard bias methods, such as excluded volume map sampling and rotational bias, the acceptance probabilities remain slight. It is, therefore, difficult to obtain reasonable density fluctuations and hence to apply the histogram analysis. In this work we use the Parallel Grand Canonical Monte Carlo method to simulate water at room temperature. The fact that the addition/deletion of molecules in this technique is not done physically makes it possible to get very high acceptance ratios and therefore to generate suitable density fluctuations at room temperature for water. We discuss the results compared to other methods and analyze its relative performance.