High-resolution x-ray scattering measurements have revealed a second critical length scale at the cubic-to-tetragonal transition in S1T1O3, corresponding to very large correlation lengths of the order of 2500 A at T c + 1 K. The correlation length becomes substantially larger still after the sample temperature has been cycled once through T c , a novel memory effect in large-length-scale critical behavior of this kind. We discuss the interpretation of these observations in terms of the effect of extended defects close to a weakly first-order transition.
18 22 24 26 Atomic Volume (A ) 28 20 16 18 20 22 Atomic Volume {A3) 24 26 28 30 32 FIG. 1. Total energy vs atomic volume for (a) GaAs, (b) InAs, and (c) A1Sb in the diamond (filled circles), wurtzite (open circles), and SC16 (squares) structures. The lines are polynomial fits to the data. 1995 The American Physical Society 16 936 0163-1829/95/52(23)/16936(2)/$06. 00 52We report revised calculations with improved accuracy to those in the 1994 paper which demonstrated the theoretical stability of the SC16 structure in III-V compounds under pressure, and the kinetic barriers to its experimental formation.In a recent paper by Mujica, Needs, and Munoz, it was claimed that the calculations contained in our original paper for GaAs were not completely converged. We have investigated possible sources of error, and found that for the wurtzite and diamond structures the k-point sampling which we used, while adequate for diamond and lonsdaleite, was insufhcient for the compounds.We have rerun our calculations using grids of 9 X 9 X 9 k points for the zinc-blende structure with two atoms, reduced by symmetry to 35 nonequivalent k points. For the wurtzite structure, a grid of 7X7X4 k points for the larger fouratom cell was used. At this sampling density the energy is converged with respect to k-point density to within 0.002 eV/atom, as compared with single runs of the ground-state structure with even finer grids. For the SC16 the k-point sampling was thoroughly described in the previous paper, and is accurate. The result for GaAs is in good agreement with that found by Mujica, Needs, and Munoz, and our wurtzitezinc-blende differences are now in agreement with the work of Yeh pt gI.Moreover, the difFerence between the c/a ratios and the ideal value is very dependent on k-point sampling. Contrary to what we said in our previous paper, stable wurtzite structure c/a ratios are smaller than ideal (GaAs c/a = 1.6499, InAs c/a =1. 6354, and for A1Sb c/a =1.6442). The energy differences between wurtzite and diamond are also much smaller, of order 0.01 eV/atom.The revised 6gures for the pressure-volume relations are shown in Fig. 1. Also, the value of Vo for BC8-GaAs should be 18.86 A .The sequence of phase stability is unchanged, and the main conclusions of our paper (that SC16 has a range of thermodynamic stability in all compounds) are also unchanged. However, the pressure range of this stability is reduced with predicted transition pressures from diamond to SC16 being increased to 109 kbar (GaAs), 66 kbar (InAs), and 46 kbar (A1Sb).
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.