First-principles study of elastic properties and pressure-induced phase transitions of Sn: LDA versus GGA results

Abstract

The structural, elastic, and high-pressure properties of tin have been studied by pseudopotential density functional theory static calculations. Both local density (LDA) and generalized gradient (GGA) approximations are used to model exchange-correlation effects. Only LDA structural and elastic results are found to be in good agreement with experiment. On the contrary, only GGA calculations are able to accurately reproduce experimental binding energies. Inclusion of nonlinear core corrections in the pseudopotential does not alter these conclusions, although it has some influence on the results at a quantitative level. It is concluded that LDA should be the method of choice to study high-pressure properties of tin. LDA calculations predict the ground state diamond $(\ensuremath{\alpha})$ phase to transform at 0.75 GPa to the $\ensuremath{\beta}\ensuremath{-}\mathrm{Sn}$ structure. This is followed by $\stackrel{\ensuremath{\rightarrow}}{\ensuremath{\beta}}\mathrm{body}\ensuremath{-}\mathrm{centered}\ensuremath{-}\mathrm{tetragonal}$ (bct) and $\mathrm{bct}\ensuremath{\rightarrow}\mathrm{bcc}$ transitions at 15.9 and 42.9 GPa, respectively. At very high pressures, the calculations predict a $\mathrm{bcc}\ensuremath{\rightarrow}\mathrm{hcp}$ phase transformation at 163 GPa. The hcp phase remains the most stable up to 200 GPa.