Fresh water scarcity is on one of the key challenge for this century. Although the planet is 70\% covered by water, the access to fresh water suitable for life is highly unequal. The challenge of water desalination using renewable energy requires lower cost solution than is currently available. In the context of solar energy powered desalination, thermal processes offer a promising way to provide fresh water at various scales. The use of a direct contact evaporator and condenser in place of traditional shell and tube heat exchangers greatly enhances the efficiency of heat transfer. Thus the understanding of the thermal and hydrodynamic features of the flow in such heat exchangers plays a fundamental role in their design. The current experimental techniques are still inadequate to obtain a full picture of small scale transport phenomena taking place locally at liquid-vapor interfaces, while the emergence and improvement of multiphase CFD techniques provides powerful tools to investigate two-phase flows at small length scale.The goal of this work is to develop a robust CFD framework to study diffusion driven evaporation as it typically occurs in the evaporator, and implement it in a commercial CFD code, ANSYS Fluent. The completion of this framework allows a better understanding of the hydrodynamic as well as the heat and mass transfer for various operating and system conditions. The use of the single fluid approach, Volume Of Fluid method, that describes each phase by the means of a scalar function that is advected using a transport equation allows an efficient means to solve the two-phase flow problem. Nonetheless, the implementation of an in-house algorithm to model diffusion driven evaporation as well as accurately computing the interfacial area is necessary since this not available on the default solver. The developed algorithm is validated using multiple benchmarks. The framework developed is applied toward the modeling of a direct contact evaporator for a counter-current Humidification-Dehumidification desalination system. The computational results show adequate agreement with multiple benchmarks. The study uses several boundary conditions, and shows a strong dependence between the packed column performance and the water distribution while the gas distribution has little effect for the conditions studied. Finally, the study takes interest into understanding the blockage or local flooding phenomenon observed both experimentally and numerically. The numerical calculations applied to a Representative Elementary Unit (REU) consider flow pattern, geometry, and wettability as parameters. The results show the geometry and wettability to be the key factors responsible of the blockage instability for the conditions studied.
Read
