Thermal energy or heat storage systems using chemical reactions to store and release energy operate in charging and discharging phases. The charging phase in this work is a dehydration process with constant heating rate decomposing salt hydrates as chemical components resulting in the obtention of a less hydrated or anhydrous form and, at the same time, storing the released heat (energy storage). Latest research on thermal decomposition of several salt-hydrates concerned experimental and numerical investigations (Huang et al., 2010; Sugimoto et al., 2007). A mathematical model of heat and mass transfer in a fixed-bed reactor for heat storage is proposed on the basis of a set of partial differential equations (PDEs) controlling the balances of mass, conversion, and energy in the bed and the reactor. These PDEs are numerically solved by means of the finite element method using Comsol Multiphysics 4.3a. The objective of this paper is to describe an adaptive modelling approach and establish a correct set of PDEs describing the physical system and appropriate parameters for simulating the thermal decomposition process. Thus it could help in the design of thermal energy storage system. The recommendations the International Confederation for Thermal Analysis and Calorimetry (Vyazovkin et al., 2011) on kinetic behaviour were used to explain transport phenomena and reactions mechanism in gas and solid phases. The generalized Prout–Tompkins equation was therefore adopted with some modifications based on thermal analysis experiments and literature. The experimental data from the TGA–DSC measurements are then used to validate the kinetic model. This latter result after validation is used in the Comsol model to simulate the lab-scale reactor in charging mode (thermal decomposition).