Sequential treatment of diphenylbis((phenylthio)methyl)silane with lithium naphthalenide, hexafluoroacetone, and then aqueous NH 4 Cl gave the (β-hydroxyalkyl)diphenyl((phenylthio)methyl)silane 8, Ph 2 Si(CH 2 SPh)CH 2 C(CF 3 ) 2 OH, a hydroxyl group of which was protected with methoxymethyl to afford the corresponding methoxymethyl ether 9a . Similarly, the reaction of 9a with hexafluoroacetone followed by deprotection of the methoxymethyl group yielded the diphenylbis(β-hydroxyalkyl)silane 11, Ph 2 Si[CH 2 C(CF 3 ) 2 OH] 2 . Sequential treatment of 11 with trifluorormethanesulfonic acid in CDCl 3 and with 2 mol equiv of 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) gave almost quantitatively [DBU·H] +, phenylbis[1,1,1-trifluoro-2-(trifluoromethyl)-2-propanolate(2−)- C 3, O ]silicate(−1) ( 6 ) along with [DBU·H] + triflate at low temperature (−20 °C). The 29 Si NMR, which showed a signal at δ Si −61.75, as well as other NMR analyses indicate that 6 is a pentacoordinate silicate with two 1,2-oxasiletanide rings. Compound 6 decomposed at room temperature to give almost quantitatively MeC(CF 3 ) 2 OH after treatment with water instead of a Peterson reaction product, CH 2 C(CF 3 ) 2 . Taking into consideration the results of a control experiment, it is strongly suggested that a novel type of silicate such as 6 can be regarded as an intermediate of a homo-Brook rearrangement.
No takes yet. Share an insight, caveat, or question.
Kawashima et al. (1998) studied this question.