Quasi-Favorskii rearrangement mechanism

The quasi-Favorskii rearrangement obviously cannot take place by the cyciopropanone mechanism. The mechanism that is generally accepted (called the semi-... 

The mechanism discussed is in accord with all the facts when the halo ketone contains an a hydrogen on the other side of the carbonyl group. However, ketones that do not have a hydrogen there also rearrange to give the same type of product. This is usually called the quasi-Favorskii rearrangement. An example is found in the preparation of Demerol ... 

In order to distinguish between a mechanism proceeding via a symmetrical cyclopropanone intermediate (Favorskii reaction) and a mechanism closely related to the benzilic acid rearrangement and called semibenzilic (or quasi-Favorskii) rearrangement, the ring contraction of 2-bromocyclobutanone was studied in deuterium oxide using sodium carbonate as base (50 C) or in boiling deuterium oxide only. 

The quasi-Favorskii rearrangement has been used often in the synthesis of unnatural, conplex, polycyclic structures. Indeed, this method is uniquely suited for such targets. However, undesired side reactions can occur. An exanple, discussed here within its mechanistic context, is based on a report by Ueda and coworkers. They treated the polycyclic dibrominated diketone 37 with potassium hydroxide in the expectation of obtaining cubane-l,3-dicarboiq lic acid 38. Given what was known about quasi-Favorskii reactions at the time (vide supra), the plan made perfect sense. However, exposure of 37 to 5% KOH for 15 min at 80 °C afforded not 38 but the cyclopropyl lactone 43 fScheme 7.12). A proposed mechanism for the process began with a Haller-Bauer cleavage, always a potential risk in quasi-Favorskii... 

The proposed mechanism of the process includes a quasi-Favorskii rearrangement, as illustrated in Scheme 73 for the transformation of 117 into 118. Attack of bromide on the epoxide is expected to occur in a trans diaxial fashion, in accord with the Fiirst-Plattner rule, at the position of the epoxide distal to the electron-withdrawing hydroxyl group. The immediate result of this ring opening would be 121, which is in rapid conformational equilibrium with 122. Migration of a o bond from this latter conformation with concomitant expulsion of bromide would give 123, which dehydrates to afford the observed product... 

There are three significant variations on the Favorskii rearrangement, the homo-, quasi-, and photo-Favorskii rearrangements. The homo-Favorskii and quasi-Favorskii rearrangements oecur if the precursor does not possess the classic a-hydrogen and a -halide. The photo-Favorskii is a variation involving a light-induced radical mechanism. 

Stevens and coworkers reported on the quasi-Favorskii rearrangement of a norborane derivative. The exo-2-bromo-e /o-2-benzoyl norborane 52 tmdergoes a rearrangement with lithium anilide as the nucleophile to form 53. The concerted semi-benzylic mechanism of this reaction produces only one of the possible isomeric products. A trace amount of the bromo-displaced product 54 was also observed. The rearranged product was... 

If there are no enolizable hydrogens present, the classical Favorskii rearrangement is not possible. Instead, a semi-benzylic mechanism can lead to a rearrangement referred to as quasi-Favorskii. 

Several structural requirements appear to be necessary in order to be able to observe the semibenzilic mechanism. Thus the acidity of the a hydrogen plays an important role. Indeed, it should be weakly acid in order to favor nucleophilic attack at the carbonyl carbon. Of course, this kind of mechanism will be facilitated by the absence of an a hydrogen the rearrangement is then known under the name of quasi-Favorskii. Since both substrates just discussed react via the semibenzilic mechanism, we shall discuss them simultaneously. 

In these constrained molecules, the Favorskii rearrangement leads essentially to the ring-contracted acid or ester in good yields. However, the formation of ring-opened products has been reported when polyhalogenated strained ketones react according to the quasi-Favorskii mechanism. Results using this type of substrate are reported in Table 8. 

---