Physico-thermal properties of TiO2 nanoparticles using molecular dynamics simulations with relevance to thermal conductance of nanofluids

Nanoparticles have attracted the attention of researchers in a number of multidisciplinary fields as they possess enhanced structural and physical properties, which make them desirable to a wide range of industries. These enhancements have mostly been attributed to nanoparticles possessing large surface area-to-volume ratio. The concept of engineered nanofluids refers to making stable suspension of nanoparticles in base fluids to provide more efficient and thermally enhanced effective thermal conductance in a number of engineering applications involving heat transfer processes. Although experimental and computational studies have shown significant enhancement of thermal conductivity of nanofluids compared to the conventional heat transfer fluids, the underlying mechanism behind this enhancement is still not well understood. Out of the several mechanisms which have been proposed in the literature to be behind this enhancement of thermal conductivity, nanoparticle clustering seems to evolve as a relatively more significant factor for the enhancement of effective conductance of nanofluids and hence requires more detailed investigation in the future. In this study, initially, numerical investigations are conducted to study the effect of particle clustering on thermal conductivity of nanofluids. The degree of thermal enhancement is analysed for different factors that capture particle clustering, such as aggregate size, particle concentration, interfacial thermal resistance, and fractal and chemical dimensions. This analysis is conducted for three model water-based nanofluids of aluminium oxide (Al2O3), copper oxide (CuO) and titanium dioxide (TiO2) nanoparticles in which particle concentrations are varied up to 4 volume %. Results from the numerical work are validated using available experimental data. Studies showed that predicted and experimental data for thermal conductivity enhancement are in good agreement as particle clustering is seen to influence effective thermal conductance of the model nanofluids. Considering this observation and to further understand the mechanism behind particle clustering, particle scale properties such as thermodynamic, structural, surface and interaction force properties are investigated in vacuum and liquid environments at different temperatures, using Molecular Dynamics (MD) simulations. These simulations are carried out on TiO2 nanoparticles (with particle sizes ranging between 2 – 6nm) as a model system owing to their wide industrial applications, with a particular focus on two mostly encountered forms of TiO2 namely; anatase and rutile. From MD simulations, the Radial Distribution Function (RDF) plots of the particles in vacuum revealed much higher crystallinity with peaks of almost an order higher than that of the nanoparticles in water. For anatase, surface energy of the particles in vacuum is seen to be higher than that of the particles in water by about 100% for the smaller size particles (especially for 2 and 3nm), and about 60% for relatively larger particles (especially for 4 to 6 nm). In both environments, surface energy of anatase nanoparticles is also seen to increase with particle size to a maximum value of critical particle size (i.e. 4 nm) after which no further significant increase in surface energy is observed. Finally, force – displacement analysis which captures interaction force relations is conducted for different particle sizes in vacuum and water environments, using MD simulations. This analysis provides interaction force information which is vital in studying particle interactions. In this study, the nanoparticles are seen to be more attractive at a smaller size and the attraction increases as the particle size increases. Particles are also seen to become repulsive at initial contact. Overall, this thesis presents vital information on particle scale properties of TiO2 nanoparticles (as a model system) under different environments. These properties are known to influence clustering of nanoparticles which in turn influence thermal enhancement of nanofluids. In the near future, knowledge of these particle scale properties will aid in the development of more enhanced nano and bulk systems with better physico-thermal properties, to meet the ever growing requirement for technological advancement.