6533b828fe1ef96bd1287b30

RESEARCH PRODUCT

On some issues in the computational modelling of spacer-filled channels for membrane distillation

Giorgio MicaleMichele CiofaloMariagiorgia La CervaAndrea CipollinaLuigi GurreriAlessandro Tamburini

subject

Settore ING-IND/26 - Teoria Dello Sviluppo Dei Processi ChimiciGeneral Chemical EngineeringSettore ING-IND/25 - Impianti ChimiciThermodynamicsMembrane distillation02 engineering and technologyMembrane distillationComputational fluid dynamics; Heat transfer; Mass transfer; Membrane distillation; Spacer filled channel; Chemistry (all); Chemical Engineering (all); Materials Science (all); Water Science and Technology; Mechanical Engineeringsymbols.namesakeThermal conductivity020401 chemical engineeringComputational fluid dynamicMass transferHeat transferGeneral Materials ScienceMass transferChemical Engineering (all)0204 chemical engineeringSettore ING-IND/19 - Impianti NucleariWater Science and TechnologyFinite volume methodChemistryMechanical EngineeringChemistry (all)Reynolds numberGeneral ChemistrySpacer filled channel021001 nanoscience & nanotechnologyThermal conductionHeat transfersymbolsMaterials Science (all)0210 nano-technologyDimensionless quantity

description

Abstract This study addresses issues which arise in the computational and experimental modelling of flow and heat/mass transfer in membrane distillation and other processes adopting spacer-filled channels (either planar or spiral wound), but have not been sufficiently clarified in the literature so far. Most of the argumentations presented are based on original computational results obtained by the authors by finite volume simulations; some literature results are also considered. The questions addressed regard the choice of scales for the reduction of data and the definition of dimensionless numbers ( Re , f , Nu , Sh ); the definition of average heat or mass transfer coefficients; the combined effects of the parameters that characterize the process (spacer pitch to channel height ratio l / H , flow attack angle γ and Reynolds number Re ) and the applicability of simple correlations; the influence of the spacer's thermal conductivity. In regard to the influence of the parameters, Re, l / H and γ were found to interact heavily, making a separate-effect analysis impossible and power-law friction or heat/mass transfer correlations of little use. Thermal conduction in the spacer, even for low-conductivity polymeric spacers ( λ  ≈ 0.15 Wm − 1  K − 1 ), was found to be responsible for up to 10% of the total heat transfer.

10.1016/j.desal.2017.02.016http://dx.doi.org/10.1016/j.desal.2017.02.016