Efficient conformational sampling by Monte Carlo Basin Paving method: Distribution of minima on the energy surface of (H2O)20 and (H2O)50

The efficiency of Basin Paving method [Phys. Rev. E 73 (2006) 015701(R)], a Monte Carlo based optimization algorithm, is investigated by applying it to (H2O)20 and (H2O)50. It is found that Basin Paving does a better sampling of low energy minima compared to the well known Basin Hopping method. The large number of minima obtained from this work is used to form a network of water cluster minima. From the properties of the network, it is found that (H2O)50 minima are more uniformly distributed in the conformational space than that of (H2O)20.