Evaluating the logistic mixture model on real and simulated TG curves

Abstract

The aim of this paper is to evaluate and explain the fitting of dynamic TG curves by a mixture of logistic functions. This model assumes that more than one physical process may be involved in each mass loss step and that each physical process may extend along all the experiment. One of the main sources of difficulties in TG is that, very often, different stages of decomposition substantially overlap each other. Several real and simulated TG curves were analysed in this paper. An optimal fitting of the TG curves was obtained by a mixture of logistics. In many cases the optimal fitting reproduces accurately the TG curve. Accordingly, the TG curve can be understood as a sum of parallel reactions, where each single reaction is represented by one or a small number of logistic components. Additionally, making use of the analytical derivative of the fitting, a mixture of Arrhenius reaction order equations was applied to the same curves. In all the cases, the fitting obtained with the mixture of Arrhenius was worse than the obtained with the mixture of logistics. A software was developed to automatically perform these tasks. The physical meaning of the fitting was explained.