Numerical and analytical study of nanofluids thermal and hydrodynamic characteristics.

摘要

The mechanisms of nanofluids heat transfer enhancement are still unclear. Previous studies about nanofluids have tried to solve some of many challenges about the thermal and hydrodynamic performance of nanofluids and their properties; however still there are many problems unsolved and questions without a certain answer. Hence, more studies are necessary, which can be experimental, numerical and theoretical. In the present study, nanofluids are investigated intensively using numerical and analytical approaches.;The numerical part consists of three chapters and covers a wide range of heat transfer problems, including; laminar and turbulent, single-phase and two-phase as well as mixed convection and forced convection flows. Several particle volume fractions and a large number of Reynolds numbers are considered. Chapter two is dedicated to laminar mixed convection flow of Al2O 3-water nanofluid inside a horizontal tube. Uniform heat flux is applied at the wall. Two Reynolds numbers and three particle volume fractions are used, and finally the thermal and hydrodynamic numerical results from three different two-phase models and the single phase model are compared with experimental data. It is shown that the predictions of these different approaches are extremely different. For a laminar mixed convection flow, two-phase models are in better agreement with a given experimental data. The two-phase models predictions are close but far from single-phase.;Chapter three evaluates the sensitivity of the laminar formulation on selected combinations of models for the conductivity and viscosity of nanofluids. Two models for the conductivity and three for the viscosity are chosen, which make six combinations. These choices are found to have very important effects on the final results. Therefore, every numerical study should first justify their choice of viscosity-conductivity correlations. Also, a list of the most important models for the conductivity and viscosity of nanofluids are gathered and included in this chapter.;Chapter four evaluates the predictions of single-phase and three different two-phase models for turbulent forced convection inside a horizontal tube. Uniform heat flux is applied at the wall. Realizable k-ϵ turbulent model is used, which is a two-equation model. Two sets of experimental data for different nanofluids (Al2O3-water and Cu-water) are used, which cover a wide range of volume fractions and Reynolds numbers. The single-phase results accuracy is confirmed with an appropriate selection of conductivity-viscosity combination. The results from different two-phase models are found to be very close; however, they were too far from the single-phase predictions and the experimental data. Two-phase models could not satisfy the experimental data for turbulent forced convection flow of two different nanofluids from different experimental studies, while single-phase approach does it well.;In the analytical part of the study, new models for the thermal conductivity of nanofluids and the Nusselt number of the flow around the nanoparticles are derived. These models take into account the effect of Brownian motion, interfacial thermal resistance, particles clustering, clusters size distribution and micro-convection as well as particles concentration, particles size and temperature. The clusters size and size distribution are analyzed based on the fractal theory. The proposed model for the conductivity of nanofluids is compared with experimental data from several studies for five different nanofluids and various magnitudes of volume fractions. This model is also compared with two similar models. It shows very good agreement with experiment and better performance compared to those selected models.;Keywords: Nanofluid, Thermal conductivity, Dynamic viscosity, Nusselt number, Convection, Turbulent flow, Two-Phase, Fractal theory.