Simplified approach to modelling the catalytic degradation of low-density polyethylene (LDPE) by applying catalyst-free LDPE-TG profiles and the Friedman method

The course of the thermogravimetric degradation of LDPE in the presence of different aluminosilicate catalysts was modelled by applying a differential isoconversional Friedman approach. An analysis of catalyst-free PE-TG profiles confirmed that the degradation profiles predicted by various reaction models overlap over the entire conversion range once the data are analysed using a differential isoconversional Friedman approach. The results demonstrate that the catalytic degradation of LDPE can be predicted by a correlation twin, i.e. the two specific functional relations between the activation energy, pre-exponential factor and conversion. The crucial step for ensuring good agreement between the predicted and the measured profiles is to extrapolate the discrete values of the activation energies and pre-exponential factors to the zero conversion. It turns out that linear extrapolation and interpolation from the discrete values outperforms regression functions based on various order polynomials, and that apparent deviations from the global trend at lower conversions are not a consequence of the misinterpretation of the experimental results but are an experimental fact. The assumption about the compensation effect between the pre-exponential factor and activation energy holds within the conversion range from 10 to 90%. However, it is generally unsuitable for modelling purposes due to the uncertain extrapolation of the kinetic parameters to the zero conversion.