A trireticulate crystal structure: trihydrogen cobalticyanide and trisilver cobalticyanide.
P. Pauling,L. Pauling
DOI: https://doi.org/10.1073/PNAS.60.2.362
1968-06-01
Abstract:Many reticulate (framework) crystal structures are known, in which the atoms are attached to one another by essentially covalent bonds to form a three-dimensional framework. A simple example is diamond; a more complex example is Prussian blue, KFeFe(CN)6-H20, in which iron atoms at the lattice points of a simple cubic lattice are joined together by cyanide groups extending along the edges of the cubes, with the potassium ions and water molecules occupying alternate positions in the centers of the cubes.1 Cuprous oxide, Cu2O, and argentous oxide, Ag2O, have a bireticulate structure. Each copper or silver atom forms covalent bonds, at 1800 with one another, to two oxygen atoms, and each oxygen atom is bonded to four copper or silver atoms that surround it tetrahedrally, to form a tetrahedral framework like that of diamond; there are, however, two interpenetrating frameworks, not bonded to one another, each of which occupies the interstices of the other. We have now found that trihydrogen cobalticyanide (trihydrogen cobalt(III) hexacyanide, H3Co(CN)6) and trisilver cobalticyanide (trisilver cobalt(III) hexacyanide, Ag3Co(CN)o) have a structure in which the atoms are bonded together to form three interpenetrating frameworks. Crystals of these two substances have been described by Ludi, Gildel, and Dvorak2 as hexagonal with a = 6.431 4i 0.004 A and c = 5.695 :I 0.004 A for the acid and a = 7.030 + 0.009 A and c = 7.127 4 0.004 A for the silver salt. The hexagonal unit of structure of each contains one formula. The authors report the spacings and observed intensities for 25 X-ray powder diffraction lines for each substance. They suggest a structure in which there are octahedral cobalticyanide groups at each point of the hexagonal lattice, with each group bonded to the group directly above and below it along the hexagonal axis by hydrogen bonds between the nitrogen atoms or by silver atoms midway between the nitrogen atoms. These structures, which were not supported by a comparison of calculated and observed intensities of the X-ray diffraction maxima, are, however, stereochemically unreasonable. Because of the large amount of triple-bond character of the carbon-nitrogen bond in cyanides, as shown by the carbon-nitrogen distance 1.15 A, a single bond formed by the nitrogen atom would be expected to make an angle of 1800 with the carbon-nitrogen axis, as in the Prussian-blue structure. With the hydrogen atoms and silver atoms located in the positions assigned to them by Ludi et al., the corresponding angle is 1000 for trihydrogen cobalticyanide and 1200 for the silver salt. In addition, the nitrogen-hydrogen-nitrogen distance corresponding to their structure (given by them as 3.23 A) is much larger than would be expected by comparison with tetrahydrogen ferrocyanide,3 in which the corresponding distances have the values 2.68 and 2.88 A.
Medicine,Materials Science,Chemistry