Addition and Ring-Cleavage Rearrangements

In 1950, Smith and McKenzie reported an attempt to prepare cyclo-propylacetic acid (65). They carbonated the Grignard reagent derived from a bromide made from cyclopropylmethanol and PBrs. However, the product turned out instead to be allylacetic acid [Eq. (2)]. Since 

Equations (5) and (6) presume that the cyclic Grignards 3 and 4 are intermediates in the formal 1,2-vinyl migrations. The subsequent finding (52) that the cyclopropylmethyl Grignard is preparable, though rapidly rearranging, supports this presumption. Further proof of its intermediacy is derived from experiments by Maercker (50, 50a) and by Hill (33), shown in Eqs. (7) and (8) respectively. In Eq. (7), cis-trans equilibration 

Richey and Veale (55) have reported a kinetic study of the cyclization of 5-hexen-l-yl organomagnesium solutions in THF [Eq. (9)]. Rates for 

Richey and Veale have also reported a study of the effect of phenyl substitution on the rate of Grignard cyclizations to yield three- and five-membered rings (56). The phenyl substitution in Grignard reagents 5 and 6 led to rates of deuterium label equilibration [analogous to Eq. (6)] 

The most impressive array of new Grignard cyclization-cleavage rearrangement examples has come from studies on the intermolecular addition of Grignard reagents to alkenes. In 1975, Lehmkuhl and coworkers published extensive descriptions of their work in this area (46, 47). If an allylic Grignard reagent adds to an alkene, the addition product 

---