Degree of covalency

For other compounds, the agreement is not always so good. The assumption that the lattice is always wholly ionic is not always true there may be some degree of covalent bonding or (where the ions are very large and easily distorted) some appreciable van der Waals forces between the ions (p.47). 

The ionization and direct displacement mechanisms can be viewed as the extremes of a mechanistic continuum. At the 8 1 extreme, there is no covalent interaction between the reactant and the nucleophile in the transition state for cleavage of the bond to the leaving group. At the 8 2 extreme, the bond formation to the nucleophile is concerted with the bondbreaking step. In between these two limiting cases lies the borderline area, in which the degree of covalent interaction between the nucleophile and the reactant is intermediate between the two limiting cases. The concept of ion pairs is important in the consideration of... 

A common interpretation of the interaction of chalcogens with nucleophiles considers donation of electron density from a lone pair on the donor atom into the o- (E-X) orbital (Figure 15.1). As the degree of covalency increases, a hypervalent three-centre four-electron bond is formed. Real systems fall somewhere between secondary interactions and hypervalent (three centre - four electron) bonds. The two extremes can be distinguished by the correlation of X-E and E D distances.In the hypervalent case both bond distances decrease simultaneously, whereas in the secondary bond the distances are anticorrelated. This concept has been applied in a study of selenoquinones 15.17 (R = Ph, Me) with short Se 0 contacts,for... 

With the increase in electronegativity of the element M the degree of covalence of the bonds M —O and M—0 should increase, as a result of which an increase in electron density on the ion M can be expected. As in the formation of the ir-bond with olefin the ir-backbonding mechanism plays a large role, that should result in an increase in the ir-complex stability. 

These high affinity sites determine most of the selectivity of the ion exchange. This is deduced from Fig. 6 in which the selectivity of the whole ion exchange (In Kn) is plotted as a function of g n, a magnetic parameter of cupric ions adsorbed on high affinity sites. This parameter is particularly sensitive to the degree of covalence of the bound between copper and its... 

The absorption band shifts to lower energy from top to bottom in the table. Roughly speaking the degree of covalency increases also in this sequence. Therefore it may be thought that we are dealing with one and the same transition. This, however, is unlikely. The literature contains many different, and sometimes not firmly based, assignments. 

The shortest cation-anion distance in an ionic compound corresponds to the sum of the ionic radii. This distance can be determined experimentally. However, there is no straightforward way to obtain values for the radii themselves. Data taken from carefully performed X-ray diffraction experiments allow the calculation of the electron density in the crystal the point having the minimum electron density along the connection line between a cation and an adjacent anion can be taken as the contact point of the ions. As shown in the example of sodium fluoride in Fig. 6.1, the ions in the crystal show certain deviations from spherical shape, i.e. the electron shell is polarized. This indicates the presence of some degree of covalent bonding, which can be interpreted as a partial backflow of electron density from the anion to the cation. The electron density minimum therefore does not necessarily represent the ideal place for the limit between cation and anion. 

The effective ionic radii of Shannon and Prewitt (1969) can frequently be used to predict average interatomic distances and to correlate unit cell volumes of series of isostructural oxides and fluorides. However, some systematic discrepancies were recently found in tetrahedral oxy-anion distances and in the unit cell volumes of certain series of fluoride compounds. It was pointed out by Banks, Greenblatt, and Post (1970) that the observed V—0 distances in Ca2VC>4Cl are smaller than those predicted by the effective ionic radii. Subsequently, the discrepancies in Ca2VC>4Cl and other tetrahedral oxy-anion distances were attributed to covalency effects (Shannon, 1971, and Shannon and Cairo, 1972) in which bonds exhibiting a greater degree of covalency were assumed to shorten. 

Biggar (1969) calculated unit cell volume ratios for isotypic Ni and Mg compounds in order to revise the Mg2+ and Ni2+ ionic radii of Ahrens (1952) and Pauling (1960). He concluded that r(Ni2+) =0.97 x r(Mg2+) in oxides and halides. It is interesting to note that Biggar found ratios for the halides F, Cl, Br, and I in agreement with the electronegativity dependence in this paper (Rv = 1.060,0.917,0.919, and 0.836 respectively) but he assumed that the deviations from 0.970 were the result of faulty data. We believe that these deviations are real and are caused by different degrees of covalence. 

Electron spin resonance, nuclear magnetic resonance, and neutron diffraction methods allow a quantitative determination of the degree of covalence. The reasonance methods utilize the hyperfine interaction between the spin of the transferred electrons and the nuclear spin of the ligands (Stevens, 1953), whereas the neutron diffraction methods use the reduction of spin of the metallic ion as well as the expansion of the form factor [Hubbard and Marshall, 1965). The Mossbauer isomer shift which depends on the total electron density of the nucleus (Walker et al., 1961 Danon, 1966) can be used in the case of Fe. It will be particularly influenced by transfer to the empty 4 s orbitals, but transfer to 3 d orbitals will indirectly influence the 1 s, 2 s, and 3 s electron density at the nucleus. 

A somewhat different situation is found in the type of point defect known as a Frenkel defect. In this case, an atom or ion is found in an interstitial position rather than in a normal lattice site as is shown in Figure 7.17. In order to position an atom or ion in an interstitial position, it must be possible for it to be close to other lattice members. This is facilitated when there is some degree of covalence in the bonding as is the case for silver halides and metals. Accordingly, Frenkel defects are the dominant type of defect in these types of solids. 

Lithium bromide and iodide probably have some degree of covalency but this does not affect the general conclusion. 

---