Fluid dynamics simulations for an open-sorption heat storage drum reactor based on thermophysical kinetics and experimental observations

To gain insight into the involved thermodynamic processes of open sorption systems and aid in layout and development of future designs, two different methodologies are presented for the numerical description of adsorption reactors using moving beds. In accordance to earlier laboratory measurements, the molecular sieve Köstrolith 4AK was selected as sorption material and thermo-physical measurements using thermogravimetry with simultaneous differential scanning calorimetry were presented to extract the necessary interpolation functions for modelling the kinetic behaviour.

Using discrete particle models in a Navier Stokes CFD Solver and particle simulations based on LIGGGHTS, the mixing characteristics of the rotating drum setup were accessed. Adsorption based on the thermophysical measurements was implemented into the discrete particle model. Finally, a parametric study with different temperatures and water content in the inflow air was performed using a transient porous volume approach based on the adsorption implementation and different mixing algorithms.

Whereas in a simple adsorption kinetic implementation, the volume averaged temperature of the reactor was already significantly reduced after 1.5h, more realistic implementations showed a prolonged reaction time with a temperature peak at around 15min. The temperature gap between the temperature of the particles and the usable temperature level for the energy extraction was reduced by introducing a mixing algorithm.

In the simulations, zeolite temperature yields could be reached between 15K and 28K, corresponding to air temperatures above the material between 21K and 33K, which compared well to the experimental observations, where temperature shifts of the process air of up to 36K were reported. The presented simulation methodology is able to identify partly unused areas in reactors.

Numerical optimisation of the flow field and enhancing the particle mixing lead to improved reactor solutions.