CSST published in “Adavnced Materials” the first framework of first-principles multiscale modeling with machine learning

Previously, our preliminary works for 2D materials has explored promising photocatalysis in two-dimensional MoSi2N4 family (Nano Energy, 82, 105716). In this work, We propose the robust concept of first-principles multiscale modeling of mechanical properties based on machine-learning interatomic potentials, conveniently and rapidly trainable over short ab-initio datasets. We show that mechanical/failure responses of complex nanostructures at continuum scale can now be explored with the precision of sophisticated first-principles calculations, affordable computational cost, and without the need for empirical data. Such an approach shows great potential to develop fully automated and coupled platforms, to design, optimize and explore various properties of materials at continuum level, with taking into account atomistic effects and inherent precision of first-principles calculations.