Abstract :
[en] By means of first-principles calculations for the SbSI semiconductor, we show that bare density functional theory fails to reproduce the experimentally observed ferroelectric phase, whereas a more advanced approach, based on hybrid functionals, correctly works. When comparing the paraelectric and ferroelectric phases, our results show polar displacements along the c direction of the Sb and S sublattices with respect to the iodine framework, leading to a predicted spontaneous polarization of P similar or equal to 20 mu C/cm(2), in good agreement with experiments. In the ferroelectric phase, the semiconducting behavior of SbSI is confirmed by relatively large values for the indirect and direct gaps (similar or equal to 2.15 eV and 2.3 eV respectively). An analysis of the electronic structure, in terms of density of states, charge density distribution, and anomalies in the Born effective charges, reveals (i) the clear presence of a Sb(III) lone pair and (ii) a large covalency in the SbSI bonding, based on the hybridization between Sb and S ions, in turn more ionically bonded to iodine anions. Finally, the interplay between ferroelectricity and spin-orbit coupling reveals a coexistence of Dresselhaus and Rashba relativistic effects and a spin texture that can be reversed by switching the polarization, of potential appeal in electrically controlled spintronics.
Scopus citations®
without self-citations
25