Tertiary organolithium

The carbolithiation of six- to nine-membered 3-methylenecycloalka-l,4-dienes has been investigated and shown to be an exceptionally facile and general process.69 Primary, secondary, and tertiary organolithium reagents may be employed for carbolithiation of cyclic trienes with uniform efficiency, generating cyclic pentadienyl carbanions. The six-electron pentadienyl systems display unique reactivity as a function of ring size. 

Secondary organolithiums 1 add to ethylene about one million times as fast as primary organolithiums (tertiary organolithiums even faster still),4 so it is easy to control the addition of r-BuLi, s-BuLi, z-PrLi etc. to give a primary organolithium product 2. The reaction is quantitative in Et20 at -25 °C.7 8... 

In contrast with unreactive, unfunctionalised terminal alkenes, allylic and homoallylic ethers (22, 24) and alcohols (20) from which the product organolithiums (21, 23, 25) can be chelated in a (preferably) five-membered, oxygen-containing ring, carbolithiate rapidly and cleanly.23 Coordination overrides any preference for the lithium to be bonded to the primary carbon, but cannot overcome the unfavourability of forming a tertiary organolithium - 26 gives 27, but 28 cannot be carbolithiated. Coordination to sulfur in similar thioethers 29 works too. 

It is possible to form, and cyclise, tertiary alkyllithiums, provided they are benzylic or allylic, by starting with a selenide. Krief has used selenium acetals to construct the starting materials 249 and 252, and on treatment with n-BuLi an extremely rapid (less than 20 min even at -110 °C effectively instantaneous at -78 °C) selenium-lithium exchange ensues to give tertiary organolithiums 250 and 253. Cyclisation to give 251 or 254 takes half an hour at -78 °C, and... 

Similar tertiary organolithiums have also been formed, and cyclised (but less stereoselectively) by decarboxylation 135... 

When the alkene trap is 1,1-disubstituted, cyclisation of a tertiary benzylic organolithium gives a product with two adjacent quaternary centres. Krief applied this type of cyclisation to the synthesis of cuparene 244136 using a different disconnection from the one used by Bailey (above). Though it is irrelevant to the synthesis of cuparene, the cyclisation of 257 to 258 is also stereoselective and produces a single stereoisomer of 259 on carbonation of the cyclised organolithium. The tertiary organolithium is too basic to cyclise in THF and in this solvent... 

In each case, stereoselectivity can be explained by assuming that the reaction proceeds through a chair-like transition state, in which there are interactions between the two ends of the C-Li and C=C bonds, and in which the substituents all occupy pseudo-equatorial positions (structures 338, 339 and 340). The same transition state model suffices to explain the stereoselectivities of the furan and pyrrolidine forming reactions above. Stereoselectivity in the cyclisations of the selenide-derived tertiary organolithiums would arise from a conformation with a precedented pseudo-axial phenyl ring. 

The discovery, at the end of the 1970s [24], of a remarkably unreactive solution of a tertiary organolithium compound was the start of an interesting expansion of Li-Barbier chemistry. 

---