Routes to carboxylic acid functional acrylonitrile copolymers via N ‐ tert ‐butyl‐ N ‐(1‐diethylphosphono‐2,2‐dimethylpropyl) free nitroxide based nitroxide‐mediated polymerization

Styrene (S)/acrylonitrile [AN; initial acrylonitrile molar feed compositions ( f AN,0 's) = 0.10–0.86) and tert ‐butyl methacrylate ( t BMA)/AN ( f AN,0 = 0.10–0.80) copolymers were synthesized at 90°C in 50 wt % 1,4‐dioxane solutions with a unimolecular initiator, N ‐(2‐methylpropyl)‐ N ‐(1‐diethylphosphono‐2,2‐dimethylpropyl)‐ O ‐(2‐carboxylprop‐2‐yl) hydroxylamine [BlocBuilder (BB)]. In the t BMA/AN copolymerizations, 8.0–8.5 mol % N ‐ tert ‐butyl‐ N ‐(1‐diethylphosphono‐2,2‐dimethylpropyl) free nitroxide relative to BB was added. The S/AN copolymers exhibited narrow, monomodal molecular weight distributions (MWDs) with low polydispersities [weight‐average molecular weight ( M w )/number‐average molecular weight ( M n ) = 1.14–1.26], and the M n versus monomer conversion ( X ) plots were relatively linear ( M n = 18.1 kg/mol, X ≈ 0.7); this suggested that pseudo‐living behavior was approached. AN proved to be an effective controlling comonomer for t BMA because the t BMA/AN copolymers exhibited narrow monomodal MWDs with M w / M n = 1.17–1.50 and relatively linear M n versus X plots to reasonably high X values ( M n = 15.6 kg/mol, X ≈ 0.6). The AN and S monomer reactivity ratios were r AN = 0.07 ± 0.01 and r S = 0.27 ± 0.02 (Fineman–Ross) and r AN = 0.10 ± 0.01 and r S = 0.28 ± 0.02 (Kelen–Tüdos), respectively; these values were in good agreement with conventional free‐radical polymerization. Error‐in‐variables model (EVM) analysis indicated that the use of cumulative composition S/AN data was more effective than typical approaches using low‐ X data with the Mayo–Lewis model. The AN and t BMA reactivity ratios [ r AN = 0.07 ± 0.01 and r t BMA = 1.24 ± 0.20 (Fineman–Ross) and r AN = 0.14 ± 0.01 and r t BMA = 0.89 ± 0.19 (Kelen–Tüdos)] were similar to those reported for related alkyl methacrylate/AN conventional radical copolymerizations. EVM analysis suggested significant experimental error was associated with the t BMA/AN system, and this warrants further investigation. © 2012 Wiley Periodicals, Inc. J Appl Polym Sci, 2012