Numerical modelling of water vapor sorption hysteresis in porous sorbents

The shelf life of humidity-sensitive foods and other products can be estimated using numerical models. In certain products, the vapor sorption is subject to hysteresis in changing humidity. Building on a quasi-first-order sorption model, we present an implementation of a domain model developed for soil science for numerical shelf-life estimation. The implemented model requires no further parameterization beyond the standard isotherms. Using silica gel as a well-controlled model system for dry, porous foods, we present an application of the model against an experiment designed to probe the hysteresis behavior for samples preconditioned at high and low humidity, respectively. The agreement between model and experiment is slightly better for the sample preconditioned at high humidity, compared to the one preconditioned at low humidity, due to the stronger influence of hysteresis in the former scenario. Since the isotherms of different samples can vary, the agreement can be improved by choosing the isotherm. Finally, the modelling approach is applied to hazelnuts to showcase the transferability to food products. The proposed approach unifies hysteresis and kinetics in a singular model which can be parameterized with low experimental effort and requires no assumptions on the sorption mechanism.