High-throughput density functional calculations to optimize properties and interfacial chemistry of piezoelectric materials

Abstract

High-throughput density functional theory calculations are conducted to search through 1572 $AB{\mathrm{O}}_{3}$ compounds to find a potential replacement material for lead zirconate titanate (PZT) that exhibits the same excellent piezoelectric properties as PZT and lacks both its use of the toxic element lead (Pb) and the formation of secondary alloy phases with platinum (Pt) electrodes. The first screening criterion employed a search through the Materials Project database to find $A\ensuremath{-}B$ combinations that do not form ternary compounds with Pt. The second screening criterion aimed to eliminate potential candidates through first-principles calculations of their electronic structure, in which compounds with a band gap of 0.25 eV or higher were retained. Third, thermodynamic stability calculations were used to compare the candidates in a Pt environment to compounds already calculated to be stable within the Materials Project. Formation energies below or equal to 100 meV/atom were considered to be thermodynamically stable. The fourth screening criterion employed lattice misfit to identify those candidate perovskites that have low misfit with the Pt electrode and high misfit of potential secondary phases that can be formed when Pt alloys with the different $A$ and $B$ components. To aid in the final analysis, dynamic stability calculations were used to determine those perovskites that have dynamic instabilities that favor the ferroelectric distortion. Analysis of the data finds three perovskites warranting further investigation: $\mathrm{CsNb}{\mathrm{O}}_{3}, \mathrm{RbNb}{\mathrm{O}}_{3}$, and $\mathrm{CsTa}{\mathrm{O}}_{3}$.