The Behavior Of Halogens At High Temperature And High Pressure: A Monte Carlo Study

A diatomic molecular crystal loses its molecular character exhibits a transition to a metallic atomic phase under high pressure when the distances between non-bonded atoms become comparable with the molecular bond length. For dense solid Hydrogen this transition was theoretically predicted by Wigner and Huntington in 1935. Many experiments have tried to achieve this, but up to now no experimental evidence has been reported in the solid under the pressures as high as 340GPa. Parallel to these experimental efforts, a large number of theoretical works has shown that the picture of the transition must be more complex than what originally proposed. As a matter of fact there has been a remarkable progress in studies of the similar transition of other diatomic molecular solids upon strong compression, especially for halogens which have metallic transition, molecular dissociation and atomic phase at relatively low pressure。The solid halogens chlorine, bromine, and iodine are among the simplest molecular solids. Along with increasing pressure, the consistent viewpoint of phase-transition order that happened both in solid bromine and iodine is from a molecular phase (phase I), passing through an incommensurate phase (phase V), finally, to an atomic phase (phase II). Recently, a study of solid iodine based on first-principles calculations and later corroborated by x-ray diffraction (XRD) experiments indicate there may be a new phase between phase I and V. This phase has two different covalent intramolecular bonds in molecular solid iodine (phase I'), and it exists before the onset of phase V [15]. An X-ray absorption spectroscopy experiments indicated the intramolecular distance of solid bromine first increases, it abruptly begins to decrease when reaching its maximum value at 25±5GPa, proving there may be phase transformation at 25±5GPa which is not observed previously. A maximum variation of 0.08 ?, is obtained at 65±5GPa where again a phase transition occurs [16]. It is obvious that the behaviors of intramolecular distances of solid halogens under high pressure, before molecular dissociation, become a very interesting study.The resemblant metallic transitions have been observed both in fluid iodine and hydrogen, and the transition pressures are much lower in the fluid phase than in the solid phase. The mechanism of metallization in the liquid phase at lower pressure (lower density) is still obscure. The same as in solid iodine, the change of intramolecular distance in liquid iodine before molecular dissociation also draws a lot interests. An X-ray absorption spectroscopy experiment measured the bond length of I2 in the room-temperature solid phase and in the liquid phase just above the melting curve, with increasing pressure. Their results show that in the solid phase it is nearly constant, while in the liquid it distinctly increases and becomes larger than in the solid at the pressure higher than 2.1GPa. As the case stands, the bond length in liquid phase is shorter than in the solid phase at ambient pressure.We have performed constant pressure-constant temperature Monte Carlo simulations of chlorine, bromine and iodine with the anisotropic atom-atom effective pair potentials presented by Rodger et al. in 1988. The equations of state, pair distribution functions, and melting curves of the halogens have been calculated.In this paper we present a new method which combines Monte Carlo simulation with fist-principles to optimize the bond length. We optimize the bond lengths of I2 in the liquid phase just above the melting curve and in room-temperature solid phase with this method, as a function of pressure. Climbing along the melting curve, the bond length in the liquid phase undergoes a much faster expansion than in the solid. As a matter of fact while at ambient pressure the bond length in the liquid phase is known to be shorter than in the solid phase. In the 0-2GPa pressure range the bond length in the solid is nearly constant while in the liquid phase it increases appreciably and eventually the order is reversed at pressure higher than 2GPa. The agreements between simulation results and experiment phenomena indicate that there dose be an anomalous band length expansion in liquid at low pressure.Optimizing the bond length of iodine in room-temperature solid phase with the same method, we obtain a similar anomalous band length changed phase transition to solid bromine under high pressure. In the 0-6GPa the intramolecular distance first decreases appreciably, then it increases reaching its maximum value at 9GPa. From this value the intramolecular distance abruptly begins to decrease evidencing a nonpreviously observed phase transformation taking place at 9GPa. A maximum variation of 0.008 ? is observed at 11GPa where again a phase transition occurs. Our results shows the new phase transition of solid iodine at 9GPa which closes to the experiments prognosticate at 8-10GPa. Our simulation confirms there may be a similar phase transition of solid iodine to solid bromine at high pressure.