Multi-objective Optimization as a Tool for Material Design

Nonempirical Definition of the Mendeleev Numbers: Organizing the Chemical Space

“…The database contains the crystal structure information for 1591 binary and 80 unary systems, which includes all elements except noble gases (Ar, Xe, Rn), lanthanoids from Ce to Lu, and all short-lived elements heavier than Bk. Of these, only 446 systems have the magnetic information obtained from several multiobjective evolutionary searches for low-energy and highly magnetized phases, as implemented in the USPEX algorithm . The hardness of all crystal structures in this database was initially computed using the Lyakhov–Oganov model .…”

“…Of these, only 446 systems have the magnetic information obtained from several multiobjective evolutionary searches for low-energy and highly magnetized phases, as implemented in the USPEX algorithm. 15 The hardness of all crystal structures in this database was initially computed using the Lyakhov−Oganov model. 16 The database is fully consistent because all crystal structures were relaxed and their energies computed with the same settings using density functional theory with the projector-augmented wave method (PAW) and PBE 17 functional as implemented in the VASP code.…”

Coevolutionary search for optimal materials in the space of all possible compounds

“…What is the nature of the chemical scale (or the Mendeleev number, which is an integer showing of the sequence of elements on the chemical scale)? Pettifor derived these quantities empirically, while we redefined them 12 using a non-empirical (thus, more universal) way, which clarifies their physical meaning. In redefining these quantities, we used the most important chemical properties of an atom (size R and electronegativity χ) in which the combination of these properties can be used as a single parameter succinctly characterizing chemistry of the element.…”

Computational prediction of novel MAB phases

“…The fitness of a generated structure generally reflects on the energy and its evolution. Recent advances in the USPEX code lets users specify additional properties to include within the fitness parameter which allows multiple parameters to be optimized simultaneously [82]. Examples of such parameters are volume, hardness, elasticity properties, bandgap, density, and structural order.…”