Molecular dynamics study of the effects of voids and pressure in defect-nucleated melting simulations

Abstract

The defect-nucleated melting of Ar has been simulated by the gradual heating of lattices that contain voids using isobaric molecular dynamics. The criterion given by Solca et al. [Chem. Phys. 224, 253 (1997)] has been used to determine the melting point from the transition temperature versus void size curve. A crystal containing a single void created by the removal of an atom and its nearest (n−1) neighboring atoms was found to give almost the same melting temperature as a crystal containing n randomly distributed single-atom voids. The melting temperature is insensitive to the shape of the void. The critical void size, beyond which there is a sudden drop in the melting temperature, decreases with pressure. At various values of pressures ranging from 0.094 to 531.6 kbar the melting points are found to be in good agreement with the experimental results and with thermodynamic results using the same exp-6 potential. The results are consistent with the Lindemann criterion of melting and in better agreement with the Lindemann criterion than are the thermodynamic results.

A comparison of methods for melting point calculation using molecular dynamics simulations

“…This is about 7 K higher than the experimental value and the values calculated using other methods but is consistent with previous results using the same method. 13 As summarized in Table I, for this simple system, the computed melting points using all four methods agree very well with each other and with the experimental value 65…”

“…Previous work has shown that the way the voids are created does not affect the calculated melting point. 13 Constant pressure simulations were carried out for each system with the temperature linearly increased from 65 K to 100 K during a 2 ns trajectory by rescaling atomic velocities every 100 steps. System properties such as density, per molecule energy, and other well defined quantities can be used to monitor the evolution of the system.…”

“…These methods can be categorized into two groups. The first group, which will be referred to as "direct" methods, include the hysteresis method, 3,4,7,8 the voids method, [9][10][11][12][13][14][15] and solid-liquid interface-based methods. [16][17][18][19][20][21][22] These approaches all involve the direct simulation of the melting process in a dynamical manner.…”

“…However, the existence of superheating, or the so-called hysteresis phenomenon, results in a significant overestimation of the melting points even for simple monatomic molecules. 13 Similarly, when a liquid is cooled down, the phase change temperature is underestimated due to the existence of supercooling. Based on the homogeneous nucleation melting theory, the hysteresis method was developed, a) Author to whom correspondence should be addressed.…”

Calculation of Phase Transitions of Uranium Dioxide Using Structure Factor in Molecular Dynamics

“…This hardware-software appliance allowed to perform computations for simulation periods of 10 ns and more. The potential was set as a file with coefficients from the Table 1 and calculated using (5).…”