Machine learning reveals memory of the parent phases in ferroelectric relaxors Ba(Ti$_{1-x}$,Zr$_x$)O$_3$

TL;DR

This study employs unsupervised machine learning on first-principles simulations to uncover phase behaviors and structural memory effects in ferroelectric relaxors Ba(Ti$_{1-x}$,Zr$_x$)O$_3$, revealing insights into their electromechanical properties.

Contribution

The paper introduces a novel machine learning workflow to analyze phase transitions and structural origins in ferroelectric relaxors, extending understanding beyond traditional methods.

Findings

Memory of BaTiO$_3$ phases in relaxors

Detection of nanodomain precursors and polar nanoregions

Identification of nanodomain phases at high Zr concentrations

Abstract

Machine learning has been establishing its potential in multiple areas of condensed matter physics and materials science. Here we develop and use an unsupervised machine learning workflow within a framework of first-principles-based atomistic simulations to investigate phases, phase transitions, and their structural origins in ferroelectric relaxors, Ba(Ti$_{1-x}$,Zr$_x$)O$_3$. We first demonstrate the applicability of the workflow to identify phases and phase transitions in the parent compound, a prototypical ferroelectric BaTiO$_3$. We then apply the workflow on Ba(Ti$_{1-x}$,Zr$_x$)O$_3$, with $x\leq0.25$ to reveal (i) that some of the compounds bear a subtle memory of BaTiO$_3$, phases beyond the point of the pinched phase transition, which could contribute to their enhanced electromechanical response; (ii) the existence of peculiar phases with delocalized precursors of nanodomains -- likely candidates for the controversial polar nanoregions; and (iii) nanodomain phases for the largest concentrations of $x$