Ylids, cyclic carbonyl

Sydnones, which contain a cyclic ylid type structure [147], have one nitrogen and two possible oxygen protonation sites. Calculations quoted by Coulson (1952) show the largest negative ch urge on the carbonyl oxygen and protonation at that site has been observed by... 

The physical and chemical properties of the X -phosphorins 118 and 120 are comparable to those of phosphonium ylids which are resonance-stabilized by such electron-pulling groups as carbonyl or nitrile substituents Thus they can be viewed as cyclic resonance-stabilized phosphonium ylids 118 b, c, d). As expected, they do not react with carbonyl compounds giving the Wittig olefin products. However, they do react with dilute aqueous acids to form the protonated salts. Similarly, they are attacked at the C-2 or C-4 positions by alkyl-, acyl- or diazo-nium-ions Heating with water results in hydrolytic P—C cleavage, phosphine oxide and the hydrocarbon being formed. 

As the Wittig reaction forms both tt and cx-bonds, the disconnection is right across the middle of the alkene giving a choice of starting materials. So with the exo-cyclic alkene 26, very difficult to make by elimination methods, we could use formaldehyde or cyclohexanone as the carbonyl component with either phosphonium salt 25 or 28. It is a matter of personal choice whether you draw the ylid, the phosphonium salt or the alkyl halide at this stage. 

At first sight, the synthesis of Z-38 by the Wittig reaction seems too risky. The phosphonium salt 39 has a more acidic proton (C02H) than the one we want to remove to make the ylid, and the aldehyde 40 not only also has an acidic proton (OH), but it prefers to remain as the cyclic hemia-cetal 41 so that there is no carbonyl group at all. 

---