Anionic rearrangement experimental compounds

Ion-pair effects, as in the case of the unsubstituted allyl alkali metal compounds 23a-d, do not markedly influence the rotational barriers of the 1,3-diphenylallyl-lithium species 27a-f, although the reason is different 27a-f are solvent-separated ion pairs 36). Addition of HMPT to the THF solution of 27a raises the AG3 c value by 0.9 kcal/mol which corresponds to a rate retardation of 5-6 times. In the case of the methyl-substituted allyl anion 27b HMPT slows down the rate by a factor of only 2-3. With the 2-cyano anion 27c the AG values of the Li+ compound in THF and the Li +, Na+ andK+ species in dimethyl sulfoxide (DMSO) are the same18). In the case of 27e and 27f the rate of the rearrangement is not affected if HMPT or TMEDA are added to the THF solutions. The observations of some solvent dependence in the case of the sterically less hindered 27a and b, but of no effect with the more crowded 27e and f are in line with the general observation that solvent-separated ion pairs are favored with respect to contact ions pairs by increasing steric hindrance 25). Hence, these experimental results could be interpreted to mean that in the case of 27a and b contact ion pairs participate in the allyl anion rearrangement reaction. 

The hydrazine rearrangement was the first known anionic rearrangement of organosilicon compounds and has been studied more thoroughly than any other. A previous review, now somewhat out-of-date (24), covered the early experimental results on this reaction. 

---