High-throughput screening of two-dimensional multifunctional Janus M2X2via machine learning force fields

Abstract

Two-dimensional (2D) Janus materials possess unique physical properties due to their broken mirror symmetry, yet their large compositional space makes systematic discovery challenging. Here, we perform a high-throughput, data-driven screening of Janus M₂X₂ monolayers to identify candidate materials with promising optoelectronic and electromechanical properties. From 15 428 designed candidates, a transfer-learning-based ensemble machine-learning force field enables efficient prescreening of structural stability across the chemical space. Stepwise thermodynamic, dynamical, and mechanical filtering narrows the set to 7 monolayers that satisfy the adopted stability criteria. Hybrid-functional calculations show that 6 are semiconductors with band gaps ranging from 1.78 to 3.49 eV. Notably, Al₂TeSe exhibits a large in-plane piezoelectric coefficient (d₁₁ = 8.95 pm V⁻¹), low-barrier sliding ferroelectricity (∼22 meV), and strong second-order nonlinear optical response (up to 1034 pm V⁻¹). In addition, the intrinsic dipole-induced potential difference supports charge separation and a suitable band alignment for photocatalytic water splitting. This work presents a systematic approach for chemical space exploration and identifies Janus M₂X₂ monolayers as candidate multifunctional 2D materials.