Chain-Length-Dependent Termination in n-Butyl Methacrylate and tert-Butyl Methacrylate Bulk Homopolymerizations Studied via SP-PLP-ESR

Abstract

The chain-length dependence of the termination rate coefficient, k t, of bulk homopolymerizations of n-butyl methacrylate (n-BMA) and tert-butyl methacrylate (t-BMA) at ambient pressure and temperatures between −30 and 60 °C has been studied via the single pulse−pulsed laser polymerization−electron spin resonance (SP-PLP-ESR) technique. The decay of radical concentration, c R, after laser SP initiation is monitored with a high time resolution of microseconds by ESR spectroscopy. Radical chain length, i, increases linearly with time t after applying the laser pulse. The experimental k t i,i values refer to rate coefficients for termination of two radicals of identical chain length i. The variation of k t i,i with chain length is adequately represented via the composite model proposed by Smith et al., in which two power-law expressions, k t i,i ∝ i −α, are contained with the exponents αs and αl referring to short-chain and long-chain radicals, respectively. The transition between the two regimes occurs at the crossover chain length, i c. The rate coefficients extrapolated for termination of two radicals of chain length unity, k t 1,1, are almost identical for n-BMA and t-BMA with an activation energy of E A(k t 1,1) ≈ 10 kJ mol−1. The αs values are close to each other: 0.65 ± 0.10 (n-BMA) and 0.56 ± 0.10 (t-BMA). Both αl values are found to be 0.20 ± 0.05, which is close to the theoretical value of αl = 0.16. The crossover chain lengths are i c ≈ 50 for n-BMA and i c ≈ 70 for t-BMA. The minor differences in composite-model parameter values of n-BMA and t-BMA are assigned to differences in chain mobility.

Chain‐Length‐Dependent Termination in Radical Polymerization of Acrylates

“…[13,36] To account for the resulting systematic error in I(0), the highest value of I(0) measured within each series of SP PLP EPR experiments under identical conditions was implemented for data evaluation. This typically results in an upward correction of c R 0 by a factor of two and a great enhancement of the fitting quality.…”

“…For methacrylate polymerizations, representative α s values are 0.63 [30] and 0.65 [45] for MMA (obtained via SP PLP EPR and RAFT CLDT respectively), 0.65 for BMA, [36] 0.50 for benzyl methacrylate [37] and 0.64 for dodecyl methacrylate (DMA). [37] All of these values are remarkably close to those quoted above from diffusion studies, noting that benzyl methacrylate, with its cyclic pendant group, should be like styrene.…”

“…The most obvious explanation for the family trend in i c values is that this quantity is related to glass transition temperature: [36] the T g values of polymethacrylates are typically 50 -100 °C greater than the corresponding polyacrylate values. [51] This means that at a given temperature the acrylates are more flexible, which should allow them to start behaving as long chains at a lower chain length, and hence have a lower i c .…”

“…This too is an attractive hypothesis, i.e., that i c reflects the size at which chains become entangled. [36] The problem with this idea is that under dilute-solution conditions chains are not entangled at all, and yet 'normal' i c values have always been measured under such conditions. A way to explain this is if i c is related to transient entanglement formation as two terminating chains overlap.…”

“…In earlier times it was the same story with propagation rate coefficients, k p : progress in unraveling the cloudy picture at higher temperatures was only made once k p for SPRs was unambiguously measured at the same low temperatures as employed here. [6] The second aspect of the counter argument is that in our earlier study of BA termination in toluene using SP PLP EPR it was found that α s , α l and i c are relatively constant with temperature, [13,29,30,36] and thus the values found at low temperature serve as reasonable estimates for higher temperatures. It was further found that E a (k t 1,1 ) ≈ E a (η -1 ) for BA SPRs, [13] which means that values of k t 1,1 measured at low temperatures may easily be extrapolated to higher temperatures: all one needs to know is the temperature dependence of the fluidity of the reaction medium, which may easily be measured if it is not available in the literature.…”