Physics-Based Feature Makes Machine Learning Cognizing Crystal Properties Simple.

Machine learning (ML) accelerates the rational design and discovery of materials, where the feature plays a critical role in the ML model training. We propose a low-cost electron probability waves (EPW) descriptor based on electronic structures, which is extracted from high-symmetry points in the Brillouin zone. In the task of distinguishing ferromagnetic or antiferromagnetic material, it achieves an accuracy (ACC) at 0.

First-principles prediction of a room-temperature ferromagnetic and ferroelastic 2D multiferroic MnNX (X = F, Cl, Br, and I).

Two-dimensional (2D) multiferroic materials have great potential applications in multifunctional nanoelectronics devices. Here, we construct a series of stable and isolated monolayers as 2D manganese nitrohalides MnNX (X = F, Cl, Br, and I) and systematically investigate the structural, electronic and magnetic properties using first-principles and Monte Carlo simulations. We find that ground states simultaneously show in-plane ferroelasticity and room-temperature ferromagnetic properties.

Huge Piezoelectric Response of LaN-based Superlattices.

We construct LaN-based artificial superlattices to investigate the ferroelectricity and piezoelectricity using the volume matching conditions of the parent components that soften the elastic constant and increase the piezoelectric modulus . The proposed superlattice consists of LaN and YN (or LaN and ScN) buckled monolayers alternately arranged along the crystallographic -direction. The structure of polar wurtzite (w-LaYN/w-LaScN) is both mechanically and dynamically stable, and the computed energy barrier makes the ferroelectric polarization switching possible.