Experimental and numerical heat transfer studies of nanofluids with an emphasis on nuclear fusion applications

Abstract

A nanofluid is a mixture of a low concentration of solid particles (10-100nm in size at concentrations below 10%vol.) and a carrier fluid (usually conventional coolants). These novel fluids exhibit anomalous heat transfer phenomena which cannot be explained using classical thermodynamic models. The fluids can be designed to offer unsurpassed heat transfer rates for heat transfer related applications at low costs of manufacturing. This PhD thesis describes the efforts to test whether these fluids can be utilised for high heat flux applications (similar to those encountered in proposed future fusion reactors) and also to discover the mechanisms which give rise to the phenomenal heat transfer enhancements observed. A broad metadata statistical analysis was performed on published literature which provided qualitative results regarding the heat transfer enhancement to be expected from nanofluids, indicated trends connecting by part mixture properties and heat transfer enhancement values exhibited and provided probable explanations of the heat transfer mechanisms involved. This study was performed to tackle the novelty and scientific uncertainty issues encountered in the field. Optical laser diagnostics experiments were performed on a high heat flux device (HyperVapotron) in isothermal conditions. The study provided extensive information regarding the flow structures formed inside the device using conventional coolants and nanofluids. This helped to both, understand the conventional operation of the device as well as review probable suitable geometries for the utilisation of the device using nanofluids. Finally, a Molecular Dynamics Simulation code was composed to model heat conduction through a basic nanofluid. The code results suggested the formulation of a new type of complex heat transfer mechanism that might explain the augmentation of heat transfer encountered experimentally. A new low cost high throughput platform (HTCondor®) has been used to run the code in order to demonstrate the capabilities of the system for less financially able institutions.