Les travaux présentés dans ce manuscrit de doctorat s'inscrivent dans la thématique du stockage inter-saisonnier de l'énergie solaire thermique pour l'habitat et le tertiaire (eau chaude sanitaire et chauffage). Le stockage thermochimique en air humide est une des solutions les plus prometteuses, en particulier avec un réacteur à lit fixe. Le bromure de strontium et l'alun de potassium ont été sélectionnés comme réactifs pour leurs caractéristiques énergétiques lors de réactions d'hydratation et de déshydratation. L'étude est constituée d'avancées théoriques, de nombreuses expérimentations et d'un modèle numérique détaillé. Une étude thermodynamique a démontré l'existence d'une droite de charge qui relie les conditions d'entrée et de sortie de l'air humide au passage du réactif. Les équations régissant les réactions chimiques, les transferts massiques et thermiques et la conservation de la quantité de mouvement ont été établies et un modèle numérique monodimensionnel couplant ces phénomènes a été développé. Des essais sur différents échantillons des deux sels et pour divers conditions opératoires ont été effectués dans le but de comprendre les phénomènes physico-chimiques ainsi que pour valider l'étude théorique et le modèle numérique. / This PhD thesis focuses on seasonal solar thermal energy storage for household applications such as production of heat and domestic hot water. Thermochemical storage was chosen for that purpose. The specific solid/gas reactions with water vapor, also called hydration/dehydration reactions, were used with a multi-scale global approach. The level of the reactor was identified as the critical level of that multi-scale approach. As a consequence, the integrated fixed-bed reactor technology in a moist air open loop system was adopted. A theoretical, experimental and numerical methodology was used for the study where strontium bromide and potassium alum salts were chosen as reactive materials. The corresponding reactions are: + 5 (H2O) ↔ (with Δhr=67.4 kJ/molwater and Δsr=175 J/K.molwater) + 9 (H2O) ↔ < KAl(SO4)2.12H2O > (with Δhr=44.2 kJ/molwater and Δsr=109.8 J/K.molwater) The first salt exhibits very good thermochemical properties. On the other hand, the main advantages of potassium alum are its low cost and the fact that it presents no sanitary risk. More than 30 cycles with 3 different samples of potassium alum and more than 25 cycles with 4 samples of strontium bromide under various stationary and dynamic operating conditions were carried out in order to understand the phenomena. The main experimental results were the following ones: • A very good stability and reproducibility of physical and chemical phenomena was observed for both materials. • A thermal reaction front was also observed. • A thermal hysteresis for both salts was found. • Based on that last observation a theoretical equation named charge-discharge line was developed. Experimental results with both salts validate the charge-discharge line theory. • A correlation between reaction kinetics, temperature rise due to the reaction, power of the reaction and the operating conditions was observed. The criterion for that correlation is the affinity of the reaction. A proportional correlation between affinity and reaction kinetics, temperature rise and power of the reaction was observed. • Spontaneous hydration and over-hydration reactions do not produce any particular difficulties or problems. • Pressure drop through the reactor and evolution of salts volume were also measured. Experimental energy density was measured in the range of 350 kWh/m3 for strontium bromide and 240 kWh/m3 for the potassium alum. • In general, strontium bromide is a very good candidate material for seasonal storage, while potassium alum cannot provide satisfying temperature rise and power. The equations governing those phenomena were also established and used to develop a 1D numerical model with partial differential equations coupling chemical phenomena, mass and thermal transfer phenomena and momentum conservation. Verification, validation and confirmation of this model under a very large range of operating conditions were carried out based on the experimental results of strontium bromide. A total of 19 different test cases were studied in order to validate the numerical model. The effect of humidity, temperature, quantity of reactive material and air flow were studied both for stationary and dynamic conditions. The numerical model was able to provide very satisfying results.