|dc.description.abstracteng||This thesis provides new insights into the mechanism and kinetics of reversible addition-fragmentation chain transfer (RAFT) polymerizations. Electron paramagnetic resonance (EPR) spectroscopy experiments to determine the rate coefficients governing the RAFT equilibrium were developed. The equilibrium constant, Keq, is deduced from the concentration ratio of the intermediate species, INT, and propagating radicals, P, via an EPR spectrum taken during stationary RAFT polymerization. Another approach uses highly time-resolved EPR spectroscopy to trace INT and P concentrations in single-pulse laser-initiated RAFT polymerizations (SP-PLP-EPRRAFT). Predici simulations of the experimental data result in rate coefficients for addition, kad, fragmentation, kbeta , and cross-termination, ktcross. Both methods have been applied to xanthate-, trithiocarbonate- and dithiobenzoate-mediated RAFT polymerizations of butyl acrylate at −40 °C. The equilibrium constants, Keq = kad/k , obtained from the stationary approach are in excellent agreement with the ones from SP-PLP-EPR-RAFT, indicating that both experimental approaches provide access to reliable data for RAFT kinetics. Fast fragmentation of INT has been observed in all polymerizations under investigation. The values for the fragmentation rate coefficient, kbeta , are 2.3×10^3 s^−1 for the xanthate, 1.4×10^2 s^−1 and
4.5 × 10^1 s^−1 for the trithiocarbonates, and 4.7 s^−1 for the dithiobenzoate. The corresponding equilibrium constants are 12 L · mol^−1, 2.6×10^4 L · mol^−1, 8×10^4 L · mol^−1, and 3×10^5 L · mol^−1, respectively. Keq is highest for the dithiobenzoate and lowest for the xanthate, which is consistent with the better control of dithiobenzoate-mediated acrylate polymerization as compared with the xanthate. Cross-termination plays a minor role when xanthates or trithiocarbonates are used as the RAFT agent, but is an important reaction step when dithiobenzoates are employed. In the latter case, adopting a chain-length dependent ktcross is necessary to explain the experimental data. To gain further insight into the rate retardation phenomenon observed in some
dithiobenzoate-mediated polymerizations and to evaluate the accuracy of ab initio
calculated Keq values reported by Coote et al., the EPR experiments were carried
out on monomer-free model systems. These systems were composed of a radical, RAFT agent bearing a leaving group, which was identical to the initiator-derived
radical, i. e., a tert-butyl, a cyano-iso-propyl or a phenylethyl group. The corresponding equilibrium constants at 20 °C are between 10^5 and 10^8 L · mol^−1, 53 L · mol^−1 and 2.2 × 10^3 L · mol^−1, respectively.
The trends in Keq for the different model systems correlate with the stability of
the intermediate radical and the stabilization energy of the radical which adds to
the thiocarbonyl bond of the RAFT agent. The theoretical values show the same
trends but are up to six orders of magnitude above the experimentally obtained
equilibrium constants. In addition, ab initio calculations predict a pronounced chainlength dependence of Keq, which was tested using macromolecular RAFT agents for stationary EPR experiments and by comparing the macromolecular systems with monomer-free model systems. Only a minor influence of the chain length was observed. The experimental results thus question ab initio calculations predicting slow fragmentation of INT and a pronounced chain-length dependence of Keq.
In addition, the product mixtures of the model systems were analyzed by nuclear
magnetic resonance (NMR) spectroscopy. The results of EPR and NMR measurements show that cross-termination with subsequent “missing step” reactions of unstable cross-termination products are responsible for the rate retardation observed in dithiobenzoate-mediated polymerizations.||de