Transition Element Fluorides

Considering the influence of electronic configurations on crystal structures it may be asked, whether certain structure t5rpes are restricted to fluorine compounds of the transition elements. Apart from the structure types distorted by the Jahn-Teller effect such a limitation is not obvious at all. On the contrary quite a number of structure prototypes are represented by compounds of the main group elements. Bonding thus must be similar in both, main group and transition element fluorides, at least as for the factors that influence crystal structmes. 

Low-valent transition-element fluorides, especially those in which carbonyl and organo-ligands are incorporated, require other approaches, which are tailored to the element and the oxidation state. Here an inherent thermodynamic problem is imposed by the high C—F bond energy, which can exceed the M—F bond energy,... 

Structural Features of Binary Transition-Element Fluorides... 

Besides radical additions to unsaturated C—C bonds (Section III.B.l) and sulfene reactions (see above), sulfonyl halides are able to furnish sulfones by nucleophilic substitution of halide by appropriate C-nucleophiles. Undesired radical reactions are suppressed by avoiding heat, irradiation, radical initiators, transition-element ion catalysis, and unsuitable halogens. However, a second type of undesired reaction can occur by transfer of halogen instead of sulfonyl groups283-286 (which becomes the main reaction, e.g. with sulfuryl chloride). Normally, both types of undesired side-reaction can be avoided by utilizing sulfonyl fluorides. 

The magnetic criterion is particularly valuable because it provides a basis for differentiating sharply between essentially ionic and essentially electron-pair bonds Experimental data have as yet been obtained for only a few of the interesting compounds, but these indicate that oxides and fluorides of most metals are ionic. Electron-pair bonds are formed by most of the transition elements with sulfur, selenium, tellurium, phosphorus, arsenic and antimony, as in the sulfide minerals (pyrite, molybdenite, skutterudite, etc.). The halogens other than fluorine form electron-pair bonds with metals of the palladium and platinum groups and sometimes, but not always, with iron-group metals. 

This structure type, that has been attributed to the fluorides KNbFa and KTaFs by Bode and Dohren [39, 40), is restricted in the case of potassium compounds KMeFe to the same transition elements Me + = Nb, Ta, Mo, W, Re (129), the sodium compounds NaMeFe of which adapt the NaSbFs structure. But as all AgMeFe-compounds 187) hitherto known seem to crystallize in the KNbFe-structure, this type is somewhat wider spread. 

The structural information we have of pentafluorides in the solid state is relatively new. The similar melting points (near 100° C and below) and even more so the almost identical boiling points (close to 230°) of the transition metal fluorides MeFs point to similar structures of these compounds. Their high volatility is clearly less than that of the hexafluorides so that one may assume associated aggregates or polymere molecules in the solid state. New structure analyses showed this assumption to be true. There exist at least three structure types within the 12 pentafluorides of d-transition elements hitherto known. Two crystal... 

Ternary fluorides Cs2MeF4 on the other hand, containing the large cesium ions, are known of several transition element ions Me +. They crystallize in rhombohedral structures with large sizes of the unit cells z = 14) 8). Detailed information about the coordination of Me and Cs in these compounds is not yet available. 

The remaining trifluorides of the transition elements known so far apparently crystallize rhombohedrally like VFa. Prior to this structure type the cubic ReOs-type will be discussed briefly, though only oxide-fluorides rather than trifluorides seem to adapt this structure. Yet it is te simple basic type of which the others may be derived. 

The following discussion is restricted to such structural elements already observed in transition metal fluorides, whereas others, examples of which are not known yet, were omitted. Later on influences will be described that may affect crystal structures by means of size, charge, electronic configuration and bonding of the constituent ions of a compound. 

Partial covalency in essentially ionic bonds changes somewhat the distribution of electrons, detectable as electron delocalisation by the modem methods of nuclear magnetic and electron spin resonance (NMR and ESR). Although the interpretations of these measurements widely differ (see 292, 293, 320) they doubtless prove the existence of partial covalency (in the order of magnitude of 10%) even in the most ionic fluorides AMeFg. Little work seems to have been done one fluorides of the heavier transition elements (96), but there is an abundant literature on first transition series fluorides, of which an arbitrary selection is given below for further information. ... 

This sequence is particularly well characterized for fluoride complexes of high-spin cations of the first-series transition elements (Allen and Warren, 1971). Moreover, between successive transition metal series, values of Ac increase by about thirty to fifty per cent. For example, in hydrated cations of the first and second transition series, Ac for [CftHjO) 3 and [Mo(H20)6]3+ are 17,400 cm-1 and 26,110 cm-1, respectively. 

Another simplification consists in the fact that among the various structural types of fluorides, the involved d-transition elements generally possess the coordination number 6. The crystallographic features can be deduced from the arrangement of (MF6) octahedra5,6. Besides in three-dimensional (3-D) networks such as found in perovskite, rutile or pyrochlore types for instance, fluorides crystallyze in two-dimensional (2-D) layer structures, one-dimensional (1-D) chain structures and isolated unit arrangements. 

Most elements (including the /-groups) in the oxidation state z have the property (82) that I of the gaseous or / of the solid fluoride is approximately z eV higher than of the corresponding oxide. This difference is far smaller in the immediate post-transition elements such as Zn(II), Cd(II) and Hg(II). This fact may conceivably be connected with the usually high coordination number N = 6,8 and 8 of these elements. 

S. BAIRD (Texas Instruments) I would like to suggest an experiment. You believe that the iron fluoride complex sets down at a kink and stops its motion. Further insight to this mechanism could be gained by substituting other transition elements that form stable complexes with fluorine perhaps one or more of them will be too big to fit into the kink. Another way would be to use other complexing agents than fluorine for the iron in order to vary the size of the ion setting down at the kink. 

Barton and cowoikers have shown how elemental fluorine can be used in nitrobenzene to obtain a selective fluorination of a steroid, a reaction of importance in drug synthesis (equation 74). A variety of transition metal fluorides, such as C0F3, are milder fluorinating agents for alkanes. 

---