Metavalent multicenter bonding in pnictogens and chalcogens: nature and mechanism of formation

Due to the exceptional property portfolio and technological applications of phase change materials, mostly chalcogens related to IV-VI and V2-VI3 families, which are in turn related to pnictogens (group V or 15) and chalcogens (group VI or 16), the nature of the unconventional chemical bonding in these materials has been debated for almost 70 years. This unconventional bond, which has been quoted in the literature as resonant, hypervalent, electron-rich multicenter, three-center-four-electron (3c-4e), and metavalent, is believed to be responsible for the exceptional properties of phase change materials. In the last decade, two bonding models, the metavalent and the electron-rich multicenter models, have competed to explain the nature of this unconventional bond, which we have here renamed as metavalent multicenter bond (MMB) for the sake of clarity. In this comprehensive work, we address the nature of MMB and propose that MMB is an electron-deficient multicenter bond (EDMB), related to the threecenter-two-electron (3c-2e) bond. For that purpose, we explore the pressure-induced mechanism of MMB formation in the some of the simplest possible systems, pnictogens (As, Sb, Bi) and chalcogens (Se, Te, Po), with density-functional theory calculations. In the way, we find that polonium is the only element among chalcogens and pnictogens with crystalline α and β structures already exhibiting MMBs at RP. We find that the mechanism of MMB formation in pnictogens (chalcogens) is comprised of three (two) stages, is similar to that of the EDMB formation in B2H6, in some Zintl phases, intermetallics, and cluster compounds, and in atomic/polymeric nitrogen and hydrogen at high pressures. On the other hand, the mechanism of EDMB formation is completely different from that of the 3c-4e bond formation in molecules. Finally, we propose the simplest geometries of EDMBs that can be found in solids along one, two, and three dimensions and comment on the validity of the doublet/octet rules in the hypercoordinated multicenter units with EDMBs.