| Abstract: | Defect fluctuations in highly distorted polymer chains were simulated by Monte Carlo calculation. The NMR autocorrelation function was derived and described by the superposition of three exponential functions with time constants spread over two orders of magnitude. As a consequence of defect diffusion, longitudinal chain diffusion (reptation) can be expected in polymer melts. By simulating the mean-square displacement of a segment, it was found that after sufficiently long times, compared with defect density correlation times, a linear relationship holds fairly well. As a rule of thumb, it can be stated that the linear Einstein equation is valid for times much greater than 103 mean step times in practical cases (chain length: several thousand segments, defect concentration: 10–20%), or, in other words, for mean-square displacements greater than a few diffusion step lengths. A long-time chain-diffusion coefficient depending on the molecular weight and on the defect concentration could be derived. Effects on the low-field NMR relaxation behavior are derived and discussed. |
