Dielectric behaviour of silicon nitride epoxy nanocomposites

Abstract

Nanodielectric materials have potential to meet the requirements of next generation insulation technology by offering attractive electrical, mechanical and/or thermal properties. A precise understanding of the factors and mechanisms that control the properties of nanocomposites could enable the tailoring and engineering of these composites to meet the requirements of a particular application. The surface chemistry of nanoparticles and their interactions with the base material are very important factors, among others, such as moisture adsorption and particle dispersion, which affect the electrical performance of nanodielectrics. This study examines factors that affect the electrical behaviour of silicon nitride/epoxy nanocomposites. First, a critical review of the models that have been proposed in the literature to describe the electrical behaviour of nanocomposites was conducted. Based on this review, a new model, namely the particle interphase model, was devised. The main proposition of this model is that nanoparticles contain a high concentration of defects close to their surfaces, which could be due to foreign atoms, surface geometrical irregularities or coordinative unsaturation, and these defects can perturb the electronic states in the outer layer of the particles which, consequently, may have a critical impact on the electrical behaviour of the bulk material. Therefore, unlike the existing models, which broadly are based on the proposition of a polymeric interphase layer around the nanoparticles, this model proposed the presence of a thin interphase layer inside the boundaries of the particles themselves that, from an electrical perspective, might have a more profound impact on the performance of nanodielectrics.
The experimental investigation started by characterising the surface chemistry of the nanofiller, where it was identified that the silicon nitride nanoparticles are covered with amine groups on their surface. Since these surface amine groups are the same as the amine groups in the hardener, this led to that these surface groups can react with the resin’s epoxy groups and, thus, affect the resin/hardener stoichiometry. Consequently, the influence of the nanofiller on the epoxy matrix might be related to a commensurate change in the matrix stoichiometry, rather than being directly associated with the presence of the nanofiller. To investigate this hypothesis, the effect of changing the resin/hardener stoichiometry was studied.
By changing the ratio of resin to hardener, a better understanding of the electrical behaviour and its relationship with the structure, dynamics and chemistry of the considered epoxy networks was achieved. Detailed electrical characterization showed that, in the glassy state, electrical properties of the studied epoxy networks are sensitive to the network’s chemical content, rather than to variations in the network’s structure or dynamics. Using formulations that contain an excess of hardener has a detrimental impact on DC conductivity, DC breakdown strength and water uptake of the resulting networks, whereas, decreasing the hardener content leads to enhancements in these properties. Conversely, AC breakdown results indicated that this parameter does not vary on changing the stoichiometry, which suggests that the AC and DC breakdown strengths are controlled by different mechanisms. A tentative explanation for the AC breakdown behaviour is suggested.
For silicon nitride filled systems, based on differential scanning calorimetry results, it was estimated that the inclusion of 2 wt% and 5 wt% of the silicon nitride nanofiller displaces the resin/hardener stoichiometry by ~6.5 % and ~18 %, respectively. This finding was further corroborated by dielectric spectroscopy and water absorption results. Therefore, this study renders conclusive evidence that nanofillers can directly and significantly affect the curing process of an epoxy network and, thus, this parameter should always be
considered when introducing nanofillers into thermosetting matrices. Such a finding implies the presence of covalent bonding between the nanoparticles and the surrounding polymer and, therefore, offered an opportunity to question what is usually conjectured in literature that strong filler/polymer interactions can affect or confine the molecular dynamics of the polymer layer around the particles and also lead to better particle dispersion. The results indicate that while this chemical bonding leads to good nanoparticle dispersion, it does not have an appreciable influence on the segmental dynamics of the polymer. According to the uncertainties of the experimental technique used here, any affected polymeric layer around the particles should not have a thickness greater than 0.8 nm.
While the nanofiller stoichiometric effect explains many aspects of the electrical behaviour of the considered nanocomposites, it cannot account alone for the whole picture. The obtained data indicate the presence of additional effect that superimposes on the filler stoichiometric effect in influencing the electrical behaviour. Modifying the particle surface chemistry, via heat treatment at 1050 ^oC, showed that this additional effect is related to the particle interphase characteristics. The results demonstrate the crucial impact of the particle interphase and thus provide experimental credence to the proposed model.

More information

Identifiers

ORCID for Thomas Andritsch: orcid.org/0000-0002-3462-022X

Catalogue record

Export record