Antarafacial shift

Such a transition state is likely to be highly strained, however, and no such 1,3-antarafacial shifts have actually been observed. A 1,7-thermal antarafacial shift in (36, x = 2), where the T.S. is likely to be much less strained (i.e. able to adopt the required helical geometry) has, however, been observed in the vitamin D series. 

In a [1,7] hydrogen shift, the allowed pathway is an antarafacial shift, in which the hydrogen atom leaves the upper surface at C-l, and arrives on the lower surface at C-7. This can be drawn 5.3 as a [a2s+n6a] process or 5.4 as a [a2a+K6s] process. This time it is structurally an antarafacial shift, but the developing overlap that happens to be illustrated can be described with one suprafacial and one antarafacial component either way round. It is helpful to draw as many suprafacial components as possible, i.e. preferring 5.1 to 5.2, since the structurally suprafacial reaction is then also described with suprafacial overlap developing. Similarly it is helpful to draw 5.3 rather than 5.4, since that makes the antarafacial component the triene system, from one side of which to the other the antarafacial shift of the hydrogen is taking place. 

The stereochemistry has been proved to follow the allowed pathways. Heating the diene 5.5 induces a suprafacial hydrogen shift to give the diene 5.6, a suprafacial deuterium shift then converts this diene into the diene 5.7, and another suprafacial deuterium shift converts it into a fourth isomer 5.8. The major components at equilibrium are the isomers 5.6, with an E trisubstituted double bond and S at the stereogenic centre, and 5.8, with a Z trisubstituted double bond and R at the stereogenic centre. Neither of the other possible isomers is evident, showing that no [1,5] antarafacial shifts had occurred. 

The antarafacial nature of the [1,7] shift was first inferred from the observation that they were known only in open-chain systems like that shown in Chapter 1 as 1.16 — 1.17. More recently it has been proved by equilibrating the triene 5.10 with the two products of [1,7] shifts, 5.9 and 5.11, the former from antarafacial shift of the hydrogen atom and the latter from antarafacial shift of the deuterium. There is no trace of either of the alternative isomers in which hydrogen has shifted to the top surface or deuterium to the bottom of C-7 (C-10 in steroid numbering). 

Fig. 4-16a Frontier orbitals for a [1,3] antarafacial shift with retention of...

The other isomer would come from a [l,7] shift. This would be allowed for an antarafacial shift, but would be impossible geometrically. 

Fig. 6.12 The [l,5]-suprafacial shift of an H atom drawn as [ct2s+j.4s] and [cr2a+7I4a] processes and the [l,7]-antarafacial shift of an H atom drawn as [o2s- -TC6J and [a2 +n6s] processes...

An alternative analysis of sigmatropic reactions involves drawing the basis set atomic orbitals and classifying the resulting system as Htickel or Mobius in character. When this classification has been done, the electrons involved in the process are counted to determine if the TS is aromatic or antiaromatic. The conclusions reached are the same as for the frontier orbital approach. The suprafacial 1,3-shift of hydrogen is forbidden but the suprafacial 1,5-shift is allowed. Analysis of a 1,7-shift of hydrogen shows that the antarafacial shift is allowed. This analysis is illustrated in Figure 10.31. These conclusions based on orbital symmetry considerations are supported by HF/6-31G calculations, which conclude that 1,5-shifts should be suprafacial, whereas... 

For the antarafacial shift, on the other hand, some overlap can be retained, even if the geometry of the transition state is very strained (Fig. 4.28). 

T. S. for [1,7] antarafacial shift, MOblus system, 1 node, 8 electrons, aromatic, A allowed... 

---