A machine learning approach to light-induced order-disorder phase transitions: large-scale long-time simulations with it ab initio accuracy

While machine learning excels in simulating material thermal properties, its application to order-disorder non-thermal phase transitions induced by visible light has been limited by challenges in accurately describing potential energy surfaces, forces, and vibrational properties in the presence of a photoexcited electron-hole plasma. Here, we present a novel approach that combines constrained density functional theory with machine learning, yielding highly reliable interatomic potentials capable of capturing electron-hole plasma effects on structural properties. Applied to photoexcited silicon, our potential accurately reproduces the phonon dispersion of the crystal phase and allows for molecular dynamics simulations of tens of thousands of atoms. We show that, at low enough temperatures, the non-thermal melting transition is driven by a soft phonon and the formation of a double-well potential, at odds with thermal melting being strictly first order. Our method paves the way to large-scale, long-time simulations of light-induced order-disorder phase transitions with ab initio accuracy.