Section 9.4.3. Interpretation of frequency distributions

It is obvious that the results have to be interpreted carefully when they are applied to crystal-structure discussions. In performing the analysis, the frequency distribution was inspected for each pair of ions. If all values were distributed around a single maximum, then and were set equal to zero and 500 pm, respectively. If there were two or more maxima, one was carefully selected and and were set to the left and right in such a way that the frequency was zero at both limits. In combinations with oxygen, so many distances were available for the most probable maximum that the program could select and automatically.

Maxima outside the selected range may come from errors in data or distances to ions in the second coordination sphere [e.g. Mg²⁺–Cl⁻ in Mg(H₂O)₆Cl₂]. Generally, different oxidation states give rise to different maxima, which therefore have been tabulated separately (e.g. Cr²⁺, Cr³⁺, Cr⁴⁺, Cr⁵⁺, Cr⁶⁺ in combination with O²⁻). In some typical cases, oxidation states cannot be clearly defined. Then the oxidation states have been omitted (e.g. Os—F, Rh—Br, N—S). Nevertheless, sometimes one oxidation state can be separated (e.g. W—Cl and W⁶⁺—Cl¹⁻). Atomic distances between equally charged ions will be contact distances and vary over a wide range (e.g. O²⁻—O²⁻ distances within ions and between such ions).