|
Internet Electronic Journal of Molecular Design - IEJMD, ISSN 1538-6414, CODEN IEJMAT
| ABSTRACT - Internet Electron. J. Mol. Des. November 2002, Volume 1, Number 11, 583-592 |
An Application of the Multicanonical Monte Carlo Method to the Bulk
Water System
Chizuru Muguruma, Yuko Okamoto, and Masuhiro Mikami
Internet Electron. J. Mol. Des. 2002, 1, 583-592
|
Abstract:
Powerful Monte Carlo algorithm may solve the multiple-minima
problem for the bulk water system. The multicanonical algorithm is
based on a non-Boltzman weight factor and produces flat probability
distribution of potential energy artificially. The method allows the
system to rove through the complex potential energy surface without
getting trapped in a local minimum state, and has been proven to be
efficient for studying first-order phase transitions of complex systems
such as spin glasses and proteins. One of the features of the method is
that the expectation values of thermodynamic properties can be
calculated as a function of temperature by applying the
histogram-reweighting techniques to the results of one long production run.
In the present study, we determined the multicanonical weight factor that can
produce flat probability distribution of potential energy corresponding
to the temperature range from 170 to 630 K. From the peak of the heat
capacity, we found a phase transition at 190 K. The lower energy
structures and oxygen-oxygen radial distribution functions imply that
the structure at lower temperatures is irregular. However, the average
number of hydrogen bonds per water molecule is nearly equal to four
at low temperatures, which suggests the formation of amorphous ice.
We conclude that the phase transition we found in the present study is
the one between liquid water and amorphous ice. In order to study
first-order phase transition between water and crystalline ice with the
multicanonical algorithm, we have to obtain more precise
multicanonical weight factor in the low energy region.
|