Here we suggest that integral isoconversional method, when applied in a mathematically correct way, can lead to satisfactory results with the least number of adjustable parameters. Differential and incremental methods are used in cases when the apparent activation energy, E, varies with the degree of conversion, α. However, in some cases the observed E(α) dependence can spuriously be induced by small variations in α(T) curves and there is only little to no benefit gained from allowing arbitrary change of E between adjacent conversion levels. As a result, the E(α) dependences are highly “fragile” and subject to minor variations in the experimental data. On the other hand, when the activation energy is optimized globally for all isoconversional levels, a significantly more robust estimate is obtained and the agreement between the experimental and simulated data is still plausible. The approach is demonstrated on two datasets which were evaluated with both variable E(α) dependence and with constant value of E.
Read full abstract