Time-Resolved EPR: A Novel Method For Studying Living Chains

Despite its great industrial importance, many aspects of free radical polymerization (FRP)1 remain poorly understood. One of the few methods that can directly probe propagating radicals, i.e., living chains which are at the heart of FRP, is electron paramagnetic resonance (EPR), a very sensitive and nondestructive technique that provides information on the local structure and dynamics and the number density of the propagating radical centers.2,3 Unfortunately, in many cases the concentration of the FRP living chains is too low during the early stages (low conversions) to allow for good EPR signals. This results in poorly resolved, noisy spectra that are prone to misinterpretations. A second problem is the measurement of many different living chain lengths simultaneously, leading to speculations that are difficult to test.4 The field would be significantly advanced if one could obtain spectra unambiguously corresponding to a single living chain length. In this article we demonstrate how the time-resolved EPR (TR-EPR) technique5-7 can be used to overcome these problems. Using TR-EPR in well characterized systems, we have studied the FRP of methyl methacrylate (MMA) and its analogues and obtained well-resolved, intense spectra of monodisperse living chains. To the best of our knowledge there has been only one other study where TREPR was applied to study FRP.8 The work reported here is also one of the few systematic exploitations of the transfer of “spin polarization” via chemical reactions8-11 and the application of TR-EPR in polymer research in general.8,10,12 In conventional EPR, time scales for data acquisition are slow (typically seconds to minutes) compared to the time to add a monomer to a living chain, 1/(kp [M]) (typically milliseconds13). This usually limits conventional EPR measurements to steady state FRPs, where inevitably a broad distribution of living chain lengths exists. Nevertheless there has been some clever efforts to measure conventional EPR of single length living chains. Fischer and Giacometti14 obtained well-resolved spectra of 1-mer living chains of methacrylic acid by creating a stationary concentration of these in the EPR cavity using a flow system. More recently, Matsumoto and Giese15 succeeded in measuring 1-, 2-, and 3-mer living chains of MMA and related monomers, by reduction of the corresponding bromide precursors in an inert solvent. In contrast, TR-EPR probes submicrosecond processes, much faster than 1/(kp[M]); thus, taking a “snapshot” of living chains is possible by this method. These chains are created by very fast addition of a primary radical (created by a short laser pulse) to monomer. This first addition step is complete within 100-300 ns. Subsequent addition rates are much slower (on the order of 1 ms) due to the lower reactivity of living chains. Our TR-EPR observation times (typically 1-2 μs following the laser pulse) fall in a range well after the creation of living chains, yet well before these can react further: only a single living chain length is monitored. To study spectra of living chains as a function of their degree of polymerization, N, one can use photoinitiators which give primary radicals attached to progressively longer polymeric chains. Here we report results for the shortest living chains, N ) 1, created using small molecule photoinitiators. Now, the matrix (consisting of monomer and pre-dissolved polymer) remains fixed during the TR-EPR observation times, allowing one to study the effect of polymer concentration on the spectra. This is not easy to do in conventional EPR in the early stages of FRP where one often needs to accumulate a large number of acquisitions to improve signals. During these times conversion (roughly equal to polymer volume fraction accumulated in the matrix) may change considerably. The conventional EPR spectra of poly(methyl methacrylate) (PMMA) radicals in both glassy and nonglassy matrices have generated considerable controversy since the first observations in the 1950s3 of a 9-line spectrum with an unusual intensity distribution. Studying oligomeric living chains of methacrylic acid Fischer16 observed 16 lines in nine main groups, and interpreted these in terms of nonequivalent â-protons of the CH2 group due to hindered rotation. More recently, Shen et al.17 and Zhu et al.18 observed a “13-line” EPR spectrum at low conversions (j20%) in the bulk FRP of MMA, which gradually turned into a “9-line” spectrum that persisted as the mixture became glassy. Zhu et al.18 suggested that the 9-line signal is due to living chains “trapped” in microscopic glassy domains. Finally, Gilbert and co-workers4 argued, based on computer simulations, that the 13-line signal must be due to oligomeric radicals (N j 5), which must have been abundant in the experiments of Shen et al.17 and Zhu et al.18 where unusually high rates of initiation were used. Resolving these various interpretations based on conventional EPR alone has clearly proven to be very difficult. The EPR signal intensity is proportional to the difference in the populations of the energy levels involved in the observed transitions.2,3 This difference is typically very small in conventional EPR where spin systems are in thermal equilibrium (EZeeman , kT at room temperature). However, just after their creation, free radicals may be in a far-from-equilibrium spin state (i.e. in a “spin polarized” state) where the difference in energy level populations is far greater than that in thermal equilibrium. Hence one may observe greatly enhanced signal intensities by monitoring EPR transitions before the spin system relaxes to equilibrium, typically within microseconds. That is the essence of * To whom correspondence should be addressed at the Department of Chemistry. 7992 Macromolecules 1998, 31, 7992-7995