Fu, Rong
(2017)
An investigation of magnetic nanofluids for various thermal applications.
PhD thesis, University of Nottingham.
Abstract
Magnetic nanofluid (MNF) is one special kind of nanofluid which possesses both magnetic and fluid properties. Nowadays, extensive attention has been focussed on development of thermal applications. Investigations of magnetic hyperthermia are emerging as a new frontier in studies of cancer therapy. The theory of treatment is based on the fact that magnetic nanoparticles produce heat under an AC magnetic field via a mechanism called magnetic losses. Facing with the present technical limits and growing demands for safe treatment, researchers have realized the advantage of assembling superparamagnetic nanoparticles (SMNP) into colloidal clusters for effective heating at low field intensity and frequency. In contrary to the isolated particles, the magnetic losses of the clusters are affected by inter-particle dipole interactions. The role of dipole interactions is complex and contradictory findings have been reported. Understanding the role of dipole interactions is the key to optimizing the clusters for efficient hyperthermia heating.
Magnetic nanofluids have also been proven to be a highly thermally conductive working fluid. The dispersed SMNPs enable control over the fluid’s thermal physical properties, flow and heat transfer processes via an external magnetic field. The main challenges include how to improve the applicability of theoretical models on predicting thermal physical properties and interpreting the role of particle migration during a convective heat transfer process. Numerous results suggested that their anomalous physical properties should be attributed to particle aggregation since it changes the effective particle concentration and generates thermal percolation paths. Also, the rate of particle migration is heavily dependent of the size of aggregates. Therefore, it is necessary to study the effect of colloidal stability on thermal physical properties and convective heat transfer enhancement of the magnetic nanofluid.
At the beginning of this doctoral research project, we investigated the effect of dipole interactions on hyperthermia heating cluster composed of multi SMNPs by time-quantified Monte Carlo simulation. The cluster’s shape is characterized by treating it as an equivalent ellipsoid. When the shape is highly anisotropic such as in chain and cylinder, dipole interactions not only facilitate the magnetization process but also impede the demagnetization process by aligning the individual moments to the cluster’s morphology anisotropy axis. Thus, the heating capability of chain and cylinder clusters are superior to non-interacting particles at the most angles between the field direction and morphology anisotropy axis. At high field intensity, the influence of dipole interactions on magnetic losses will be reduced to a minimum once the cluster loses its morphology anisotropy (i.e. cube or sphere); the probability to obtain improved heating becomes very low.
Then, experimental and theoretical works were conducted together to find out how to improve the heating ability of anisotropic-less clusters at lower field intensity and frequency. Hydrophobic Fe3O4 nanoparticles were assembled into sphere-like clusters using the emulsion droplet evaporation method. The hydrodynamic size of the cluster was controlled within the range of 70 – 140 nm. An induction heating system equipped with an Optic-fiber thermometer was set up to test the heating efficiency of as - prepared Fe3O4 clusters with different size. Meanwhile, standard Monte Carlo simulation was performed to study the contribution of dipole interactions at different sizes. The findings suggested that if one expects anisotropic-less clusters to heat better, he should reduce the cluster’s size so that the clusters are in forms of dimer and/or trimers or use SMNP with high magnetization and magneto-crystalline anisotropy.
Finally, a stable and surfactant-free magnetic nanofluid was prepared for study of convective heat transfer enhancement. Ethylene glycol and water mixture was selected as the base liquid, which is often used for cooing an automotive engine. The surfaces of Fe3O4 nanoparticles were modified with citric acid to make the colloidal stability sensitive to the pH of the particle suspension. It was found that the density and specific heat of obtained MNF can be interpreted well by mixing theory and thermal equilibrium model respectively. After the colloidally stability was optimized, the MNF exhibited Newtonian behavior. The viscosity barely changed with the shear rate despite variances in particle concentration and temperature. A modified Krieger and Dougherty model was used to explain the relationship between the size of aggregates and viscosity. Meanwhile, we found that the thermal conductivity can be predicted by the Maxwell model, which presumes the nanofluid has common features with a solid–liquid mixture. At last, it was demonstrated that the convective heat transfer coefficient of our MNF was 10 % higher than that of base liquid at transition and turbulent flow.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
![[thumbnail of PhD thesis.pdf]](https://eprints.nottingham.ac.uk/style/images/fileicons/application_pdf.png "PhD thesis.pdf")