Useful method to analyze data on overall transformation kinetics

The aim of this article is to demonstrate the important source of errors when overall crystallization kinetics data are plotted in coordinates ln [ − ln (1 − α)] against lnt (in isothermal case), respectively ln [ − ln (1 − α(T))] against ln q (in the case of linear temperature change at a rate q). Due to the specific properties of the logarithmic function, (in particular because ln 0 → − ∞), this plot exaggerates the role of both the initial stage (α → 0) and the stage near the end of the process (α → 1). Unfortunately, these are just the ranges where most grave experimental errors appear. In case the double logarithmic scale is used, data outside the limits − 2 < ln ( − ln (1 − α)) < 1, respectively 0.1 < α < 0.9 are not reliable and should not be taken into account. Instead, we propose suitable coordinates for presentation of experimental data, so that the power n in the Kolmogorov–Johnson–Mehl–Avrami equation is determined in a more reliable manner.