Sun, Qiangqiang
(2021)
Micro- and nano-scale fluid flow and heat transfer with phase change in a channel.
PhD thesis, University of Nottingham.
Abstract
Micro/ Nanoscale fluid flow and heat transfer with phase change in a channel widely exist in film cooling of aircrafts and electronic devices, etc. Compared with their macroscale counterpart, Micro/ Nano-scale fluid flow and heat transfer show some new features such as velocity slip, temperature jump, density layering and non-uniform viscosity phenomena. Besides, some factors (i.e., axial conduction, viscous dissipation and rarefied effect) that are normally neglected in macroscale flow and heat transfer become dominant at Micro/ Nano-scale counterparts. In this thesis, Micro/ Nanoscale fluid flow and heat transfer are thoroughly investigated by various approaches. The two-dimensional energy equation with a first-order velocity slip model and a temperature jump model is solved analytically, and impacts of viscous dissipation, axial conduction and rarefied effect on the local Nusselt number, the asymptotic Nusselt number and the bulk temperature profile of fluid are clarified. The second law analysis method based on the local entropy generation rate is then proposed to compute and visualize the resistance of flow at low Reynolds number featuring velocity slip after implementing the linear Navier velocity slip boundary condition in OpenFOAM through three schemes. Then, classical molecular dynamics simulations are performed to reveal the features of Micro/ Nanoscale flow and heat transfer (i.e., velocity and temperature slip, non-uniform density and viscosity profiles). This study is followed by the study of the phase change (i.e., thermal de-icing and the icephobic coatings) at the molecular level.
It is found that the asymptotic dimensionless bulk temperature of fluid converges to a constant value that is higher than the wall temperature at a given set of Brinkman number, P´eclet number and Knudsen number regardless of the inlet conditions. When neglecting axial conduction and the rarefied effect, the asymptotic Nusselt number with or without viscous dissipation is 17.5 or 7.54, respectively. Effects of axial conduction on the asymptotic Nusselt number are negligible when the P´eclet number is greater than 10, while its influence on the non-dimensional bulk temperature of fluid and local Nusselt number can be neglected only when P e > 100. The flow resistance is caused by the wall drag force work (equal to 0 for no-slip flow) and viscous dissipation in the flow field related to the local entropy generation rate. The fully developed bulk temperature obtained from molecular dynamics simulation agrees with the continuum based solution of the analytical energy equation at channel height 24 nm, while this agreement reduces with the decrease of the height due to the nanoscale features. At height 6 nm, velocity slip exists around the hydrophobic wall, and enhanced near-wall viscosity of liquid and reduced velocity slip length are observed at larger fluid-wall interaction strength. A region around 0.2 nm wide without liquid atoms is formed at the hydrophilic wall, leading to a zero velocity in this hollow domain and a no-slip boundary condition. Most importantly, the thermal slip length is remarkably dependent on the liquid density layering in the proximity of the wall and inversely proportional to the first peak value of liquid adjacent to the interface. The energy consumption of thermal de-icing can be modelled as a bilinear function of wall temperature (Tw) and ice thickness (H). The melting time is almost bilinear with respect to H and 1/(Tw − 273.15 K), and converges to a constant value at Tw ≥ 313.15K. When adopting a hydrophobic surface, which is generally considered as icephobic and prevents ice accretion, more time and almost constant energy are required to melt the ice. The ice adhesion stresses on wall substrates can be reduced significantly due to the geometrical structure, buckling phenomenon and hydrophobicity of the graphene-carbon nanotube junctions, respectively, and the reduction can be further strengthened when there is a water lubrication layer.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
![[thumbnail of Revised thesis approved by internal examiner for final submission]](https://eprints.nottingham.ac.uk/style/images/fileicons/application_pdf.png "Revised thesis approved by internal examiner for final submission")