Thermal transport in nanofluids: boiling heat transfer

This thesis is constructed around the topic of thermal transport in nanofluids, with special emphasis of boiling heat transfer. Nanofluids boiling, as it is popularly known, have been researched nearly for two decades. While some controversies surrounding the boiling mechanisms had been sorted out, others still remained. The aim of this thesis is to address the remaining concerns. For the best treatment of the research problem, the experimental work was divided into three segments.

Time-resolved small angle x-ray scattering studies of nanofluids were focused to examine the nanoparticle aggregation in liquid media. These were conducted at two national synchrotron facilities located at Daresbury Laboratory and Diamond Light source in Oxford, UK. In-situ experiments were conducted with static and convective nanofluidic samples. Types of nanofluids and the experimental conditions were chosen to cover a broad range of practical applications. Water based nanofluids of spherical particles of aluminium oxide (Al2O3) and titanium dioxide (TiO2) and acicular particles of aluminium oxide (Al2O3) were exposed to 1Ă wavelength x-ray beam for varying duration of times at a frame rate of 10miliseconds. Data analysis was conducted using the Dream and SAXS Utilities software. Signs of change in particle size were discovered with the near-IEP nanofluids. For further clarification of these SAXS observations, microscopic and photography studies were conducted in the laboratory. SEM studies were supported by optical microscopy. It helped to estimate the aggregate sizes and porosity within aggregates. The settling rates were determined by still photography, which were subsequently compared with the prediction of Stoke’s settling theory. At the end of data and image analysis, it was discovered that the x-ray beam had successfully predicted the settling rates of nanoparticle aggregates. Although SAXS has long been used to analyse particulate systems, for the best of the knowledge of this author, this is the first time its capability as a tool to estimate particle settling rates in nanofluids has been showcased. Furthermore, by fine tuning the present methodology, it seems possible to determine the nanoparticle aggregation rates in a nanofluid.

Saturated pool boiling of nanofluids was experimentally investigated under the atmospheric pressure. A boiling test rig was designed and constructed for this purpose in the Leeds University. Water based and water-ethylene glycol (WEG) based nanofluids were examined for boiling heat transfer on flat copper substrates. The substrates were resistively heated from the bottom, providing surface heat fluxes up to 189kW/m2. Boiling heat transfer coefficients were calculated using the measured temperature differences between substrate and the boiling liquid at each surface heat flux. All nanofluids in general displayed deterioration in boiling heat transfer. Moreover all substrates were found fouled with nanoparticles after boiling. SEM observation on fouled substrates revealed the presence of structures consisting of sub-micron size cavities and pores, which are possibly interconnected by sub-surface network of channels. By measuring their surface roughness, it was further understood that the degree of change of roughness due to boiling depended upon its initial roughness, the particle concentration in the nanofluid, as well as the shape of nanoparticles. This study also points to an interval of roughnesses that gives optimum boiling heat transfer performance. Further experiments are recommended to focus on this aspect. Also to avoid in future is the bubble nucleation in the periphery of the copper substrates that became a major obstacle to visualise bubbles in the middle.

The need to explore the bubble nucleation phenomena on at sub-micron size cavities was inspired by the presence of such cavities on nanofouled substrates. Moreover in literature sometimes the inconsistencies on the degree of enhancement or deterioration were attributed to hitherto-unknown boiling phenomena at these length scales. Two key challenges were to create very small cavities on a smooth substrate and to conduct phenomenally clean boiling experiments on them. In principle there should not be a foreign particle inside the boiler which is larger than the cavity mouth. The biggest challenge however was to find a technique to measure the temperature of the liquid layer on top of the cavities. The infrared thermometry facility at the Nuclear Science and Engineering Department of the Massachusetts Institute of Technology (MIT) was used as a part of research collaboration. Tiny cavities were machined on ultra smooth silicon substrate using the focus ion beam (FIB) technology at the University of Leeds and at Harvard Centre for Nanoscale Systems. The mouth diameters of conical cavities were ranging from 0.6µm to 4.5µm. A boiling test rig was simultaneously developed at MIT. Heating to the liquid was provided by a halogen spot heater. The cleanliness of the test rig was successfully proved by reaching the heterogeneous nucleation superheat of liquid methanol on a silicon wafer. Water and a water based dilute SiO2 nanofluid were boiled in this novel test rig. Temperature profiles of bubble evolution were captured using the IR thermometry. Also the superheated liquid layer temperatures were measured. It was found that the measured values were in good agreement with Young-Laplace theory. Moreover the SiO2 0.01wt%-water nanofluid in most cases demonstrated boiling heat transfer enhancement up to 40% above water. With this work, for the first time the classical Young-Laplace theory was proved for sub-micron cavities. It further removed the suspicion that there might be a different phenomenon governing the bubble nucleation on nanofouled substrates.