Radical Ring-Opening Polymerization of Phosphorus Heterocycles: Computational Design of Suitable Phosphetane Monomers

High-level ab initio calculations have been used to determine the propensities of various phosphetanes towards radical ring-opening polymerization. At the G3(MP2)-RAD level of theory, the propagation rate constants of 1-methylphosphetane (7.5 × 104 L mol–1 s–1), 1-phenylphosphetane (4.6 × 105 L mol–1 s–1), cis,cis-2,4-dichloro-1-phenylphosphetane (3.8 × 107 L mol–1 s–1), cis,cis-2,4-difluoro-1-phenylphosphetane (3.0 × 107 L mol–1 s–1), and 1-phenyl-3-oxaphosphetane (4.0 × 106 L mol–1 s–1) are very high, rendering them unsuitable for copolymerization with common alkenes. In contrast, the propagation rate constants of 1-tert-butylphosphetane (1.7 × 103 L mol–1 s–1) and cis,cis-2,4-dimethyl-1-phenylphosphetane (9.2 × 102 L mol–1 s–1) indicate that either incorporation of a t-butyl substituent at phosphorus or alkylation at the 2- and/or 4-positions will produce monomers with more compatible reactivities for copolymerization with alkenes. In the case of 1-tert-butylphosphetane, however, homolytic substitution of the propagating radical with the t-butyl substituent at P will be competitive with the propagation step and could affect the microstructure of the polymer. The borane adduct and the oxide of 1-phenylphosphetane were both found to be unreactive towards radical ring-opening. The calculations suggest that, for chiral phosphetanes, the ring-opening reaction is enantioselective at phosphorus and the resulting polymer will be isotactic.