Polymerization of glycidol and its derivatives: A new rearrangement polymerization

Anionic (KOH) polymerization of glycidol, or its trimethylsilyl ether (TMSGE) followed by hydrolysis, gives a low molecular weight, largely amorphous polymer that is not the reported 1,3-polyglycidol but, based on ¹³C-NMR, largely a 1,4-poly(3-hydroxyoxetane) with much branching. This result is achieved by a simple rearrangement of the usual, propagating secondary oxyanion to a primary one. Substantial amounts of four dimers (5–10%), four trimers, and some tetramers were also found. One dimer was isolated and shown to be glycidyl glycerin, the usual thermal dimer from glycidol. Possible structures of the other dimers are proposed. The polymerization appears to begin with the rapid formation of the glycidoxy anion , formed by base abstraction of a proton from glycidol and by nucleophilic displacement of the SiMe₃ group from TMSGE. Other bases such as KOtert-Bu give similar 1,4 polymer for glycidol but, with TMSGE, there is considerable 1,3 polymerization. Detailed mechanisms are proposed. The polymer perpared from R-TMSGE with KOH was highly crystalline, high melting (166°C), H₂O soluble, isotactic poly(3-hydroxyoxetane). The cationic polymerization of tert-butyl glycidyl ether (TBGE) and TMSGE gave low molecular weight 1,3 polyethers. The TBGE polymer was all head-to-tail whereas the polyglycidol from TMSGE contained extensive head-to-head chain units with considerable branching. Mechanisms for these interesting differences are proposed.