Mathematical methods have been developed to evaluate solid-state kinetics. They generally fall into two categories: model-fitting or model-free (isoconversional). Modelfitting methods determine the kinetic triplet (model, frequency factor and activation energy) whereas isoconversional (model-free) methods generate the activation energy as a function of reaction progress without modelistic assumptions.This work investigated the relationship between calculation methods and artifactual variation in activation energy. Variable activation energy, often reported by isoconversional methods, could be an artifact due to experimental errors. This can lead to erroneous mechanistic conclusions about a reaction being complex.In this work, a new approach is proposed for obtaining the kinetic triplet that combines the advantages of isoconversional and model-fitting methods. An isoconversional method was used to select the reaction model by comparing calculated activation energies to that predicted by modelistic approaches. The selected model was the one that had an activation energy closest to the isoconversional value and the frequency factor was then obtained from the selected model. This complementary approach was used to evaluate simulated and real experimental data.The desolvation kinetics of several structurally related solvates of sulfameter (5methoxysulfadiazine) were evaluated both isothermally and nonisothermally. Calculated desolvation kinetic parameters were compared and related to the crystal structure of the solvates. A relationship was observed between calculated activation energy and solvent size. The larger the solvent molecule, the higher its solvate's desolvation activation energy. The solid-state reaction models selected were rationalized on the basis of crystal structures of the solvates where the solvent molecules were in cavities in the crystal.Finally, it was found that kinetic parameters obtained isothermally and nonisothermally were not in agreement. It was concluded that kinetic results from isothermal experiments may not be extended to nonisothermal or vice versa without confirmation of their equivalence.Abstract Approved: ____________________________________ Thesis Supervisor ____________________________________