Different computer-based simulation models, able to predict the performance of Reverse ElectroDialysis (RED) systems, are currently used to investigate the potentials of alternative designs, to orient experimental activities and to design/optimize prototypes. The simulation approach described here combines a one-dimensional modelling of a RED stack with a fully three-dimensional finite volume modelling of the electrolyte channels, either planar or equipped with different spacers or profiled membranes. An advanced three-dimensional code was used to provide correlations for the friction coefficient (based on 3-D solutions of the continuity and Navier-Stokes equations) and the Sherwood numbers (based on 3-D solutions of a scalar transport equation), as well as to test simple models for the Ohmic resistances (based on 3-D solutions of a Laplace equation for the electrical potential). These results were integrated with empirical correlations for the transport properties of electrolytes and membranes, and were used as the input for the higher scale model. The overall model was validated by comparison with experimental data obtained in laboratory-scale RED stacks under controlled conditions. This combined approach constitutes a fully predictive, potentially very accurate, and still extremely fast-running, tool for the approximate simulation of all the main variables, suitable for performance prediction and optimization studies

No abstract

No abstract

No abstract

Computational results were obtained for oscillatory flow with zero time mean ͑reciprocating flow͒ in a plane channel using a finite-volume method. A forcing term that varied cosinusoidally in time was imposed, and its frequency and amplitude were made to vary so as to span a range of regimes from purely laminar to fully turbulent. The results were validated against analytical solutions and literature data. In turbulent flow, although the computational grid did not fully resolve all the turbulence scales, the results were judged to be sufficiently accurate to capture all the essential features of the problem. The present computational results confirmed the existence of four main flow regimes ͑laminar, disturbed laminar, intermittently turbulent, and fully turbulent͒, already identified in the previous literature. One of the most interesting results was that the relation between the amplitudes of the forcing term and of the flow rate was found to be approximately linear both in the laminar and in the turbulent regimes; the reasons for this peculiar behavior were investigated and discussed. The influence of oscillation frequency and forcing term amplitude on transition to turbulence was also studied; results were compared with transition criteria proposed in literature, and a flow regime chart was proposed. Finally, the effect of unsteadiness on heat transfer was investigated by imposing different temperatures at the opposite walls of the channel and computing mean and fluctuating temperature distributions and heat transfer rates. The Nusselt number was found to increase significantly even in the disturbed laminar regime and to vary as Re 0.8 ͑as in steady turbulent flow͒ in the turbulent regime.

No abstract

No abstract

Heat transfer in spacer-filled channels of the kind used in Membrane Distillation was studied in the Reynolds number range 100-2000, encompassing both steady laminar and earlyturbulent flow conditions. Experimental data, including distributions of the local heat transfer coefficient h, were obtained by Liquid Crystal Thermography and Digital Image Processing. Alternative turbulence models, both of first order (k-ε, RNG k-ε, k-ω, BSL k-ω, SST k-ω) and of second order (LRR RS, SSG RS, ω RS, BSL RS), were tested for their ability to predict measured distributions and mean values of h. The best agreement with the experimental results was provided by first-order ω-based models able to resolve the viscous/conductive sublayer, while all other models, and particularly ε-based models using wall functions, yielded disappointing predictions.