Accelerated search for BaTiO3-based piezoelectrics with vertical morphotropic phase boundary using Bayesian learning

Dezhen Xue,Prasanna V Balachandran,Ruihao Yuan,Tao Hu,Xiaoning Qian,Edward R Dougherty,Turab Lookman
DOI: https://doi.org/10.1073/pnas.1607412113
2016-11-22
Abstract:An outstanding challenge in the nascent field of materials informatics is to incorporate materials knowledge in a robust Bayesian approach to guide the discovery of new materials. Utilizing inputs from known phase diagrams, features or material descriptors that are known to affect the ferroelectric response, and Landau-Devonshire theory, we demonstrate our approach for BaTiO3-based piezoelectrics with the desired target of a vertical morphotropic phase boundary. We predict, synthesize, and characterize a solid solution, (Ba0.5Ca0.5)TiO3-Ba(Ti0.7Zr0.3)O3, with piezoelectric properties that show better temperature reliability than other BaTiO3-based piezoelectrics in our initial training data.
What problem does this paper attempt to address?