Journal article

Publication year: 1998

Pages: 1877-1885

Journal: Journal of Inorganic and General Chemistry

Volume number: 624

Issue number: 11

ISSN: 0044-2313

Publisher: Wiley

Abstract: 
There is still no appropriate way to order quantitatively what we believe to know about structures of solids. Overall formulae like Mg2O4Si (Mg-2[SiO4] not equal (OMg4).[Si2O7]) do not work. Traditional terms like coordination number and polyhedron, ligand or central ion are here, dealing with collectives, only definable arbitrarily. Intimate transitions between chemical compounds (and other reasons too) omit basically the use of a quality like symmetry as an ordering principle. But, as shown here for the first time, quantities like lattice constants and positional parameters lead (using an appropiate crystallographic description) to exactly defined physically useful relations between the lattices of the components (K.G.) to calculate, to analyze and to characterize these quantitatively by numbers between 0 and 400. Thereby it is possible to avoid rigid group like dividions (e. g., sulfate or borate/double oxides) and yet to order chemically useful. Each lattice of the components is additionally geometrically characterized by Q(i) (50-150%). This enables us to compare quite different structures in new ways. Examples given here are those where differentiation of cat- and an-ions is customary. The same treatment seems possible for intermetallic compounds too, for example Zintl phases.