Problems with the determination of activation energy as function of the reacted fraction from thermoanalytical experiments

The so-called compensation effect is well known between the activation energy, E , and the pre-exponential factor, A . The present work shows by examples that much higher compensation effects may arise when E and A vary with the reacted fraction. For this purpose, a set of five simulated experiments were constructed by first-order kinetics with E = 200 kJ mol −1 at a wide range of heating rates. These data were evaluated by the method of least squares assuming E and A as functions of the reacted fraction. Such E functions were found which highly differed from a constant E while described well the evaluated data. They included a linearly increasing E and several parabolic E functions. The observed effects may contribute to the contradictory kinetic parameters that were reported in the literature of the isoconversional (“model-free”) studies. It was found that the compensation effects between E and A functions can be 8–11 times higher than between E and A values .