Edward Schofield

Non-classical free-radical polymerization: 5. The purification and polymerization of 1-vinyl-2-pyrrolidone

Abstract The impurities in 1-vinyl-2-pyrrolidone (NVP) have been investigated by g.c./m.s. and 13 C n.m.r. The inadequacy of normal purification procedures is pointed out, and a simple technique for reducing the impurity-level to undetectable values is described. Sources of non-ideality in free-radical polymerization are classified and it is suggested that a plot of ωj −0.5 vs. [M] could be particularly valuable in revealing the origins of non-ideal behaviour, including the presence of retarders. (ω, j are the rates of polymerization and initiation, respectively, and M represents monomer). Extensive data on NVP polymerization in ethanol solution at 60°C with azo-bis-isobutyronitrile as initiator have been obtained by dilatometric studies; for this purpose redetermination of the polymer density proved necessary. The data have been analysed as outlined above. The efficacy of the purification procedure in removing ‘active’ impurities has thus been demonstrated and it is concluded that, as already suggested by others, the polymerization under the prevailing conditions is subject to a medium effect operating principally on the propagation process. This is discussed briefly.

Non-classical free-radical polymerization. 6. Degradative addition in the polymerization of acenaphthylene: a treatment including size-dependent termination coefficients

The structure of acenaphthylene suggests that degradative addition to monomer may occur during free-radical polymerization. We present an analysis, based on the kinetic treatment in an earlier paper, of the kinetic data of Romani and Weale1 on the polymerization of acenaphthylene in toluene at 1 atm pressure. Two cases are considered: (i) where the termination coefficient is constant, and (ii) where kt is size-dependent. The latter case is appropriate because small radicals are involved; analysis is facilitated by introduction of the concept of a ‘group-termination-coefficient’. Both treatments lead to results in good agreement with the experimental data, which are, therefore, consistent with the occurrence of degradative addition in this polymerization. The estimated kinetic parameters have different values in (i) and (ii), notably in that reinitiation is important in the latter, but insignificant in the former. Possible techniques for more detailed examination of re-initiation are outlined. Molecular weight distributions for the polymerization with degradative addition to monomer with constant kt have been calculated.