Control of the steric course of the Wittig reaction

The modern interpretation of the Wittig reaction mechanism

“…Where such data is available, 36,53,54,55,56,57,58 quantitative comparison of ylide reactivities is difficult, since the Wittig reactions in different studies were often carried out under different conditions and, most significantly, with different 70 solvents. However, qualitatively, it is known that non-stabilised ylides react essentially instantaneously (giving stable oxaphosphetane -abbreviated OPA -at low temeprature), while reactions of semi-stabilised ylides are generally complete in a matter of seconds or minutes (especially with aldehydes), even at 75 low temperature. Some triphenylphosphine-derived stabilised ylides do not undergo Wittig reactions or require heating to effect the reaction, 3 but we have observed methyldiphenylphosphinedrived stabilised ylides to react rapidly with aldehydes even at -78 °C.…”

“…The experimental observations informing the betaine rationale, the alternative explanations for these observations, and the evidence that indicates the non-involvement of betaines is summarised below. (i) Observation indicating the involvement of betaine: The 70 most significant evidence came from the Wittig reactions of several ylides (generated using phenyllithium from the parent phosphonium bromide salt) with carbonyl compounds at low temperature, which gave a precipitate that on warming yielded alkene and phosphine oxide but if treated with HBr gave β- 75 hydroxyphosphonium salt (β-HPS, an example of which is species 7 in Scheme 3). Subsequently the precipitate was isolated and shown to be a betaine-LiBr complex 64 (which has since been detected spectroscopically 65 ).…”

“…38 Thus, the betaine mechanism is unable to account for moderate to low Z-selectivity under kinetic control in Wittig reactions of non-stabilised ylides. Furthermore, the reactions of certain non-stabilised ylides (primarily ethylides -see later) are the only ones that are know not to be irreverisble 75 under Li salt-free conditions. It is also pertinent that OPA is the only observable intermediate in reactions of non-stabilised ylides, and that betaines have been shown conclusively not to be involved in the Wittig reactions of dibenzophosphole-derived non-stabilised ylides.…”

“…before OPA decomposition occurs) to the Wittig reactions of non-stabilised ylides causes very high E-selectivity in the reactions, which in the absence of methanol (or if it is added after warming to room temperature) would show high Z-selectivity. 22 75 …”

“…4 One of these is reproduced here. The unimolecular rate constants for ring opening of cyclopropylcarbinyl radicals are extremely large (Scheme 75 10(b)). 83 By analogy, if the formation of radical ions from the Wittig reactants shown in Scheme 10(a) was occurring (see path B) then exceptionally fast ring cleavage would occur.…”