Analysis of Single Phase, Dispersion, Discrete and Mixture Models, Predicting Nanofluid Transport in Forced and Natural Convection and Particle Deposition Through the Porous Media
Abstract
Two different models were analyzed: single phase (homogenous and dispersion) and two-phase (Lagrangian-Eulerian model or discrete phase model and the mixture) with both constant and temperature dependent properties to further investigate and clarify the differences and evaluate the assumption of the single-phase model. The obtained results were subjected to an intensive comparison with the available experimental data and numerical works in the literature. The influence of some important parameters such as, source and sink terms, injected particle mass flow rate, particle diameter, particle type, slip velocity, particle forces, Reynolds and Peclet numbers, wave amplitude, constant or temperature dependent properties and particle concentration on the heat transfer and flow characteristics of nanofluids were determined and discussed in detail. It was observed that the two phase Lagrangian-Eulerian model (DPM) overestimated the heat transfer coefficient values and the results from the mixture model displayed an unrealistic increase in heat transfer particularly for high particle volume fraction. The proposed single-phase approach revealed a very good agreement with the experimental data and the maximum difference in the average heat transfer coefficient between the single-phase and DPM was found to be 5.9%. Particle deposition through porous media was analyzed utilizing the discrete particle model (DPM). The Brinkman-Forchheimer extended Darcy model was used for the flow inside a saturated porous matrix. The effect of porous permeability (Da=10-8-10-4), Reynolds number (Re=500-2000), volume concentration (0%, 0.3% and 3%) and different particle forces on the deposition rate have been documented. The particle adhesion/detachment was solved with respect to the force balance considering drag, Saffman lift, Brownian, thermophoresis, gravity and Van Der Waals. Our results reveal that the mass deposition rate can be omitted when there is no porous media inside the channel. It is found that, the porous permeability has a substantial role on nanoparticle mobility and a critical Reynolds number (500≤Re≤1000) exists where the entrapment rate is maximized. The impact of different pertinent forces on the deposition was also considered, and our results establish that Brownian motion had the most dominant effect on the deposition rate in the presence of a porous medium.