Thermal and Electrical Properties of Nanocomposites, Including Material Properties

The research described in this thesis is part of a state-funded IOP-EMVT project in cooperation with industrial companies, aiming at the design, assessment and implementation of new, environmental friendly (e.g. oil and SF6 - free) solid dielectric materials. A large disadvantage of solid polymer dielectrics is their relatively low thermal conductivity. Therefore, the focus in this thesis is on if and how nanotechnology can improve the thermal conductivity without deteriorating existing electrical properties. Epoxy resin, which is very common polymer material in the electrical and power industry, has been used as a host to create new insulating materials: nanocomposites. In order to improve the thermal conductivity of epoxy resin, thermally conducting but electrically insulating nanofillers, such as aluminum and magnesium oxides (Al2O3 and MgO), silicon dioxide (SiO2), boron and aluminum nitrides (BN and AlN) were used to dope the polymer matrix. Good compatibility and adhesion was achieved by surface modification of the nanoparticles, using a silane coupling agent. Proper dispersion of nanoparticles is a vital factor for the final properties of nanocomposites. Good and stable dispersion of nanoparticles in polymer matrices have been achieved by mechanical mixing and ultrasonic vibration. The quality of the dispersion of nanoparticles was satisfactory for most of the nanocomposite samples. The fabricated composites were classified into three types, according to the average particle size and the extent of agglomerates observed inside the polymer matrix. Dielectric spectroscopy revealed that the relative permittivity of many nanocomposites is lower than that of the pure epoxy. This surprises, since the relative permittivity of the bulk materials of the fillers used is higher than that of the epoxy. The anomalous dielectric behaviour of nanocomposites was explained by the existence of an interface layer between polymer matrix and inorganic filler, and its influence on the macroscopic properties of the composite. The dielectric spectroscopy investigations demonstrated a reduction of the real and imaginary parts of the complex permittivity for all samples after subjecting the samples to postcuring. The postcuring process leads to evaporation of absorbed water and finalizes the process of epoxy curing. It was postulated that the interface polymer volume, which is affected by the alignment of polymer chains around surface treated nanoparticles, conducts the heat much better than an amorphous polymer that is not altered by nanoparticles. We proposed a three-phase Lewis-Nielsen model to fit the thermal conductivity behaviour of nanocomposites, which have a third phase of aligned polymer layers. The model fits the experimental data very well and takes the thermal resistance of the interface into account. Besides the interfacial layer and its nature, the size of the particles, their aspect ratio, crystal structure and alignment inside the polymer as well as surface modification are important aspects in determining the thermal conductivity of composites. Several ways are proposed to optimize the nanocomposite processing to enable scaling up to large industrial volumes. Finally, possible harmful effects of nanoparticles on health and required precautions for the workplace are discussed in the course of this thesis