Molecule hypervalent

So far our discussion of hybridization has extended only to period 2 elements, specifically carbon, nitrogen, and oxygen. The elements of period 3 and beyond introduce a new consideration because in many of their compounds these elements are hypervalent—they have more than an octet of electrons around the central atom, oco (Section 8.7) We saw in Section 9.2 that the VSEPR model works well to predict the geometries of hypervalent molecules such as PCI5, SFs, or BrFj. But can we extend the use of hybrid orbitals to describe the bonding in these molecules In short, the answer to this question is that it is best not to use hybrid orbitals for hypervalent molecules, as we now briefly discuss. 

The valence-bond model we developed for period 2 elements works well for compounds of period 3 elements so long as we have no more than an octet of electrons in 

Which of the p orbitals do you think contributes the most in the mixing that leads to the right-most sp hybrid orbital 

Does it matter which of the two sp hybrid orbitals are used to hold the two nonbonding electron pairs  

This discussion reminds us that models in science are not reality but rather are our attempts to describe aspects of reality that we have been able to measure, such as bond distances, bond energies, molecular geometries, and so on. A model may work well up to a certain point but not beyond it, as is the case for hybrid orbitals. The hybrid orbital model for period 2 elements has proven very useful and is an essential part of any modern discussion of bonding and molecular geometry in organic chemistry. When it comes to molecules such as SFg, however, we encounter the limitations of the model. 

First we generate the levels of an octahedral AIl molecule since it illustrates 258 

FIGURE 14.1. /v.sscinbly of the molecular orbital dia.uram of an oclahcdral AH molecule from the orbitals of A and of ll. d orbitals arc not included on A. 

FIGURE 14.2. FitVci on the All L nci.ey levels nf ineluclin.efi i rbiials on A. The orbital, nonbondiiig in 1 iinirc 14.1 ish wereti in enersv. 

3d orbitals is so immense that the higher energy orbitals arc of no importance at all. (In any case, any inclusion of d orbitals will slightly stabilize the e r set and will not alter our conclusions.) Wc will proceed in this chapter without the use of d orbitals and, therefore, force a delocalized description of the electronic structure in several places. 

---

A variation on MNDO is MNDO/d. This is an equivalent formulation including d orbitals. This improves predicted geometry of hypervalent molecules. This method is sometimes used for modeling transition metal systems, but its accuracy is highly dependent on the individual system being studied. There is also a MNDOC method that includes electron correlation. 

Hypervalent molecules incorporate elements with more than a normal complement of eight valence electrons (an octet). 

Examine the charge on the methylidene group, as well as the magnitude and direction of the molecule s dipole moment. Are they consistent with representation of the ylide as a hypervalent molecule or as a zwitterion ... 

Three basis sets (minimal s-p, extended s-p and minimal s-p with d functions on the second row atoms) are used to calculate geometries and binding energies of 24 molecules containing second row atoms, d functions are found to be essential in the description of both properties for hypervalent molecules and to be important in the calculations of two-heavy-atom bond lengths even for molecules of normal valence. 

G1 theory does badly with ionic molecules, with triplet-state molecules such as O2 and S2 and with hypervalent molecules. Gaussian-2 (G2) theory eliminates some of these difficulties by making the following three changes ... 

Hypervalent molecules, like sulfoxides and sulfones, are too unstable. 

If only one set of polarization functions is used, an alternative notation in terms of is also widely used. The 6-31G=i basis is identical to 6-31G(d), and b-SlG ts is identical to 6-31G(d,p). A special note should be made for the 3-21G basis. The 3-21G basis is basiciy too small to support polarization functions (it becomes unbalanced). However, the 3-21G basis by itself performs poorly for hypervalent molecules, such as sulfoxides and sulfones. This can be substantially improved by adding a set of d-functions. The 3-2IG basis has only d-functions on second row elements (it is sometimes denoted 3-21G(=f=) to indicate this), and should not be considered a polarized basis. Rather, the addition of a set of d-functions should be considered an ad hoc repair of a known flaw. 

The foregoing discussion indicates that while there are difficulties in the way of a bonding role for 3d orbitals, for certain situations at least it is possible to conceive of ways in which these difficulties may be overcome. However, it is necessary to say that even for hypervalent molecules such as SF6 which seem to require the use of d orbitals, there are molecular orbital treatments not involving the use of d orbitals. In fact, as shown by Bent in an elegant exposition12, the MO model of SF6 involving the use of d orbitals is only one of several possibilities. The octahedral stereochemistry of SF6, traditionally explained in... 

It is interesting to note that many of the techniques developed in phosphorus chemistry are npw being routinely applied to hypervalent molecules of other elements. For instance, Martin et al. have studied the pseudorotational (Berry) mechanism for the inversion of 10-Si-5-siliconates (1) by 19F n.m.r. and demonstrated a linear correlation between AG for inversion at silicon and the a values of the variable ligand, Y The energy barriers for... 

Figure 1.17 Some examples of hypervalent molecules that have more than eight electrons in the valence shell of the central atom.

The best Lewis-type representation of the bonding in OCF3 would therefore appear to be as in 4, even though the carbon atom does not obey the octet rule. This molecule can be considered to be a hypervalent molecule of carbon just like the hypervalent molecules of the period 3 elements, such as SFfi. We introduced the atom hypervalent in Chapter 2 and we discuss it in more detail in Chapter 9. But it is important to emphasize that the bonds are very polar. In short, OCF3 has one very polar CO double bond and three very polar CF single bonds. A serious limitation of Lewis structures is that they do not give any indication of the polarity of the bonds, and much of the discussion about the nature of the bonding in this molecule has resulted from a lack of appreciation of this limitation. 

Because they do not obey the octet rule, hypervalent molecules have often been thought to involve some type of bonding that is not found in period 2 molecules. Ideas concerning the nature of this bonding have developed along a somewhat tortuous path that it is interesting and instructive to follow. We will in the end conclude that the nature of the bonding in these molecules is not different in type from that in related period 2 molecules and that there is therefore little justification for the continued use of this concept. 

Lewis considered covalent and ionic bonds to be two extremes of the same general type of bond in which an electron pair is shared between two atoms contributing to the valence shell of both the bonded atoms. In other words, in writing his structures Lewis took no account of the polarity of bonds. As we will see much of the subsequent controversy concerning hypervalent molecules has arisen because of attempts to describe polar bonds in terms of Lewis structures. 

It is difficult to give a localized orbital description of the bonding in a period 3 hypervalent molecule that is based only on the central atom 3s and 3p orbitals and the ligand orbitals, that is, a description that is consistent with the octet rule. One attempt to do this postulated a new type of bond called a three-center, four-electron (3c,4e) bond. We discuss this type of bond in Box 9.2, where we show that it is not a particularly useful concept. Pauling introduced another way to describe the bonding in these molecules, namely, in terms of resonance structures such as 3 and 4 in which there are only four covalent bonds. The implication of this description is that since there are only four cova-... 

Hybrid Orbital Descriptions of the Bonding in Hypervalent Molecules... 

A satisfactory description of the bonding in hypervalent molecules can also be given in terms of molecular orbitals but this does not directly correspond to the very useful picture of five or more localized bonds (see, for example, Mingos, 1998, p. 250). 

The use of resonance structures such as 7 and 8 to describe bond polarity led to a subtle change in the meaning of the octet rule, namely, that an atom obeys the octet rule if it does not have more than eight electrons in its valence shell. As a result, resonance structures such as 7 and 8 are considered to be consistent with the octet rule. However, this is not the sense in which Lewis used the octet rule. According to Lewis, a structure such as 7 would not obey the octet rule because there are only three pairs of electrons in the valence shell of carbon, just as BF3 does not obey the octet rule for the same reason. Clearly the octet rule as defined by Lewis is not valid for hypervalent molecules, which do, indeed, have more than four pairs of shared electrons in the valence shell of the central atom. 

We can summarize the foregoing discussion of the octet rule and hypervalent molecules as follows ... 

Hypervalency is not a consequence of some special type of the bonding. The bonds in hypervalent molecules are similar to those in any other molecules and may range from predominately ionic to predominately covalent. 

The cu-bonding model provides a more complete and fundamental description of hypervalent molecules that are often interpreted in terms of the VSEPR model.144 In the present section we examine some MX species that are commonly used to illustrate VSEPR principles, comparing and contrasting the VSEPR mnemonic with general Bent s rule, hybridization, and donor-acceptor concepts for rationalizing molecular geometry. Tables 3.32 and 3.33 summarize geometrical and NBO/NRT descriptors for a variety of normal-valent and hypervalent second-row fluorides to be discussed below, and Fig. 3.87 shows optimized structures of the hypervalent MF species (M = P, S, Cl n = 3-6). 

Figure 3.87 Structural depictions of hypervalent molecules (see Tables 3.32 and 3.33).

The deviations between theory and experiment normally tend to be somewhat smaller for first-row than for second-row compounds. Among the established semiempirical methods, PM3 seems to be the best for the first-row compounds, but the OM1 and OM2 approaches with orthogo-nalization corrections [23-25] perform even better, with mean absolute deviations being around 3-5 kcal/mol. For the second-row compounds, MNDO/d is currently the most accurate among the semiempirical methods considered. This performance has been attributed [16-18] to the use of an spd basis which allows a balanced description of normalvalent and hypervalent molecules. OM1 and OM2 have not yet been parameterized for second-row elements. 

---