A promising concept for seasonal solar heat storage in the built environment is based on the reversible sorption process of water vapor into the crystalline structure of salt hydrates. Understanding of hydration and dehydration processes on molecular level is essential for the success of different materials for thermochemical heat storage. Molecular Dynamics simulations including reactive potentials are used to model the hydration/dehydration process in materials like magnesium sulfate, lithium sulfate, etc. The application of such a model is time consuming and limited to the accuracy of the potential used in the simulations. Our solution to the problem is to gain essential information on different reactive structures by optimizing the reactive potential using DFT calculations and to use them into a stochastic Monte Carlo approach. Our aim for this pilot project is to use DFT calculations in order to compute binding energies for MgSO4*xH20 (x=1..6) structures in order to further optimize the reactive force field.
