Investigation on sorption phenomena for thermal storage applications

Energy storage has become a main topic in the last years, growing in importance with the increase of renewables penetration in the energy mix and the new energy efficiency policies. Indeed, a storage permits to absorb the renewable production peaks and release this energy when it is requested by the users, giving flexibility to renewables. In particular, thermal energy storage (TES) technologies can be adopted in a wide range of applications: they can be used for the recovery of waste heat increasing the efficiency of the process involved, or coupled with renewables, mainly solar power, for a daily or seasonal storage. Among TES, thermochemical energy storage (TCES) is getting attention between researches thanks to its high energy density and near-zero thermal losses which are a significant advantage respect to latent and sensible TES. These two particular properties makes TCES a great option for seasonal thermal energy storage. Nowadays, zeolites and silica gels have been mostly investigated for storage purposes. However, these two materials have some significant drawbacks related to their energy density and working temperatures which make them not really efficient for TCES applications. So, other materials are under studying, e.g. AlPOs, SAPOs, MOFs, and hydrated salts. SAPO-34 is a quite interesting material for TES applications due to its hydrothermal stability, low regeneration temperature, and an interesting type IV isotherm. For these reasons this thesis is focused on the study and the characterization of the SAPO-34 by means of experimental tests and numerical modeling. A SAPO-34 sample, produced by Mitsubishi Chemical and named AQSOA-Z02, has been analyzed at CNR-ITAE (Consiglio Nazionale delle Ricerche, Istituto di Tecnologie Avanzate per l'Energia "Nicola Giordano") laboratories. The obtained results have then been employed to validate the model of the system implemented at Politecnico of Torino. The modeling of the adsorption process have been performed through MD and GCMC simulations; a model tuning has been carried out in order to obtain satisfying results and a good matching with experimental data.