Homoelement bonding

Bonding between atoms of similar type may often be exists when a binary compound shows an apparently anomalous stoichiometry. For instance, the solids with empirical formulae NaO, KO₂, LiS, CaC₂ and NaN₃ consist of the ions O₂^2-, O₂^- , S₂^2-, C₂^2-and N₃^- correspondingly. With a p-block element of intermediate electronegativity the combination of an electropositive metal gives so-called Zintl compounds. Some consist of discrete polyatomic units like Ge₄ tetrahedra in KGe; in others there are continuous bonded networks like Si chains in CaSi, or layers in CaSi₂. Frequently these structures can be understood by isoelectronic analogy with the nonmetallic elements: so Ge₄ ⁴ - (in KGe) has similar valence electron count as P₄; Si^2- (in CaSi) is likely isoelectronic to S, and Si^- (in CaSi₂) to P. Even though this analogy is helpful the ionic formulation may be misleading, like the solids are frequently metallic in appearance and are semiconductors.

The word metal-metal bonding is used when such type of homoelement bonding involves the more electropositive element of a binary pair. Once more, it may sometimes be exist when an unusual oxidation state is found. For instance, HgCl consists of molecular Hg₂Cl₂ units with Hg-Hg bonds and GaS also has Ga-Ga bonds.

Metal-rich compounds are created by early transition metals, with formulae like Sc₂Cl₃ and ZrCl, and structures depicting extensive metal-metal bonding. They are particularly common with elements of the 4d and 5d series and sometimes may not be supposed from the stoichiometry. An instance is MoCl₂, which consists of the cluster [Mo₆Cl₈]⁴⁺ with a metal-metal bonded Mo octahedron. Metal-metal bonding frequently provides rise to anomalous magnetic or other properties but the surest condition is a structural one by metal-metal distances comparable with or shorter than those found within the metallic element.