I. Fleming et al., THE PHENYLDIMETHYLSILYL GROUP AS A MASKED HYDROXY GROUP, Journal of the Chemical Society. Perkin transactions. I, (4), 1995, pp. 317-337
A phenyldimethylsilyl group attached to carbon can be converted into a
hydroxy group 1-->5, with retention of configuration at the migrating
carbon, by any of three main methods. The first involves protodesilyl
ation, to remove the phenyl ring from the silicon atom. followed by ox
idation of the resulting functionalised silicon atom using peracid or
hydrogen peroxide. The second uses mercuric acetate for the same purpo
se, and can be combined in one pot with the oxidative step using perac
etic acid. This method has a variant in which the mercuric ion is comb
ined with palladium(ii) acetate, both in less than stoichiometric amou
nts. The third uses bromine, which can also be used in one pot in conj
unction with peracetic acid. In this method, but not in the method bas
ed on mercuric acetate, the peracetic acid may be buffered with sodium
acetate. The method using bromine as the electrophile for removing th
e benzene ring has a more agreeable variant in which it is administere
d in the form of potassium bromide, which is oxidised to bromine by th
e peracetic acid. The scope and limitations of each of these methods a
re reported with a range of examples possessing between them many of t
he common functional groups. Simple benzene rings, alcohols, ethers, e
sters, amides and nitriles are compatible with all three methods, and
ketones do not undergo Baeyer-Villiger reaction under any of the condi
tions. Amines, however. are oxidised to amine oxides. Ketones may be b
rominated in the third of the three main recipes. The absence of acid
in the third method makes it especially valuable when the phenyldimeth
ylsilyl group has a neighbouring nucleofugal group such as hydroxy or
acetoxy. Carbon-carbon double bonds are incompatible with the methods,
except for terminal monosubstituted double bonds, which can survive t
he conditions used in the first of the three methods.