The 63rd JSAP Spring Meeting, 2016

Presentation information

Oral presentation

1 Interdisciplinary Physics and Related Areas of Science and Technology » 1.1 Interdisciplinary and General Physics

[20a-S322-1~8] 1.1 Interdisciplinary and General Physics

Sun. Mar 20, 2016 10:15 AM - 12:15 PM S322 (S3)

Chiemi Fujikawa(Tokai Univ.)

11:15 AM - 11:30 AM

[20a-S322-5] Understanding the Difference in Cohesive Energies between Alpha and Beta Tin

Sergei Manzhos1, Fleur Legrain1, Oleksandr Malyi2, Clas Persson2 (1.Ntl Uni of Singapore, 2.Uni of Oslo)

Keywords:tin,phase transition,ab initio modeling

The transition temperature between the low-temperature alpha phase of tin to beta tin is close to the room temperature (Ttrans =130C), and the difference in cohesive energy of the two phases at 0 K of about dEcoh=0.02 eV/atom is at the limit of the accuracy of DFT (density functional theory) with available exchange-correlation functionals. It is however critically important to model the relative phase energies correctly for any reasonable description of phenomena and technologies involving these phases, for example, the performance of tin electrodes in electrochemical batteries. Here, we show that several commonly used and converged DFT setups using the most practical and widely used PBE functional result in dEcoh≈0.04 eV/atom, with different types of basis sets and with different models of core electrons (all-electron or pseudopotentials of different types), which leads to a significant overestimation of Ttrans. We show that this error is due to the errors in relative positions of s and p –like bands, which, combined with different populations of these bands in alpha and beta Sn, leads to overstabilization of alpha tin. We show that this error can be effectively corrected by applying a Hubbard +U correction to s –like states, whereby correct cohesive energies of both alpha and beta Sn can be obtained with the same computational scheme. We quantify for the first time the effects of anharmonicity on dEcoh and find that it is negligible.