Data-Mining Driven Design for Novel Perovskite-type Piezoceramics

Abstract

Materials Genome Initiative is envisioning the discovery, development, manufacturing and deployment of advanced materials twice as fast and at a fraction of cost. High throughput computation and experimentation will generate big data, which underscores the emergence of the fourth paradigm data science. In contrast to machine-learning needing big-data, data-mining assisted by domain knowledge and expertise works well with a limited number of data. In this presentation, data-mining based on material genome approach were performed in field of perovskite-type oxides. New ferroelectric ceramics based on BiFeO3 for high temperature piezoelectric applications are realized with piezoresponse of 1.5~4.0 times the present commercial non-perovskite counterpart. Our essay demonstrates data-mining driven searching sure able to reduce time-to-insight and human effort on synthesization, accelerating new materials discovery and deployment.