Torquoselection

The formation of cyclic nitrones (150) from pericyclic mechanism. Kinetic and computational studies have provided evidence for the involvement of a novel pseudo-pericyclic electrocyclization in the conversion of o-vinylphenyl isocyanates into quinolin-2-ones. " Such reactions have also provided evidence of torquoselectivity in a 6jt system. Hash vacuum thermolysis of triazoles (151) has been found to afford dihydroquinolines (155), presumably by generation of a-oxoketenimines (152) which can undergo a [1,5]-hydrogen shift to the o-quinoid imines (153)7(154) and subsequent electrocyclization (see Scheme 57). 

Benzocyclobutenone and derivatives have been shown to react with diazomethylene anions to give 2,3-benzodiazepines under very mild conditions. A mechanism has been proposed which involves four-electron electrocyclic ring opening of the initial alkoxide, with important acceleration by the alkoxide anion. Torquoselectivity to set up an eight-electron electrocyclic ring closure to the benzodiazepine enolate has been attributed to the strong preference of the alkoxide anion for outward rotation. This mechanism has been supported by isolation of one of the /S-diazo alcohols, and its conversion to a benzodiazepine on LDA treatment.84... 

Trialkylsilyl vinyl ketenes (72) have been shown to react stereoselectively with a-benzotriazolyl organolithium species to give highly substituted cyclopentenones. The selectivity was found to be kinetic, not thermodynamic, in origin. Several possible mechanisms have been proposed (Scheme 10). It has been suggested that the observed stereoselectivity may result from torquoselectivity in a concerted reaction, or from stereospecific conrotatory cyclization of cation (73), formed stereoselectively because of the interaction shown between the electron-withdrawing group Z and the metal ion.75... 

Electron-accepting R groups are best represented by two low-lying orbitals, one filled and one empty. Figure 6.1 shows that the interaction of the HOMOs will inhibit the in pathway, whereas the HOMO (cyclobutene)-LUMO (R) interaction will promote it. The LUMO (cyclobutene)-HOMO (R) also disfavors the in pathway. These results are summarized in Figure 6.3 interaction (1) favors the in mode and interactions (2) and (3) impede it. This means that the energy gap between the cyclobutene HOMO and the R LUMO must be as small as possible for an in reaction to occur. In other words, R must be a powerful acceptor. However, Rondan and Houk believe that even then, interaction (3) may largely cancel interaction (1) and the torquoselectivity will not be very pronounced. 

A priori, we would expect disrotatory reactions to show poorer torquoselectivity than conrotatory reactions for two reasons. Consider, for example, the hexatriene cyclohexadiene interconversion. On the one hand, the overlap between R and the distal carbon C6 is similar for the in and out pathways, as in the in mode, the major lobe at C6 is oriented away from R ... 

On the other hand, hexatriene must take a helical structure.6 By analogy with the octatetraene case, it may be inferred that electronic effects will have a smaller influence and steric effects a larger influence than in the cyclobutene torquoselectivity. This is indeed what was found by Houk and co-workers.7... 

The Rondan-Houk theory rationalizes many experimental results and leads to several predictions which have since been confirmed. Exceptions have been explained by Houk and co-workers, using numerical calculations. In a sense, the problem of electrocyclic reaction torquoselectivity may be considered solved. 

Explain the torquoselectivity in the following cyclobutene ring-opening reactions ... 

Photoirradiation at 300nm of iV-alkoxycarbonyl-l,2-dihydropyridines results in ring closure and the formation of 2-azabicyclo[2.2.0]hex-5-enes. Substituents are tolerated at the 2-, 3-, and 4-positions however, the yields are significantly lower than those for unsubstituted dihydropyridines. Irradiation of 2-substituted-l,2-dihydropyridines 112 proceeds via a torquoselective process to give only the endo-product 113 (Equation 5) <2001JOC1805, 2000T9227>. 

Like Roald Hoffmann and Ken s own Ph.D. adviser, R. B. Woodward, Ken seems to enjoy making up erudite-sounding names for new phenomena that he discovers. In addition to periselectivity and torquoselectivity , Ken has added theozyme to the chemical lexicon. 

Irradiation of 2-substituted-l,2-dihydropyridines 546 proceeds via a torquoselective process to give only the endo product 547 <2001JOC1805, 2000T9227>. 

In a more polar solvent, Favorskii reactions cease to be stereospecific, and presumably take place by ionisation of the chloride to give the same cation from each diastereoisomer. Whether the reaction takes place by way of the cation or with concerted loss of the chloride ion, this reaction presented a serious puzzle before its pericyclic nature was recognised. The a overlap of the p orbital on C-2 of the enolate with the p orbital at the other end of the allyl cation 6.340 or with the orbital of the C—Cl bond 6.341 looked forbiddingly unlikely—it is 3-endo-trig at C-2. It is made possible by its pericyclic nature, where the tilt of the orbitals can begin to sense the development of overlap. The torquoselectivity in the development of overlap 6.341, however improbable it looks, corresponds to inversion of configuration at the carbon atom from which the chloride departs. 

---