Impact of Microstructure on Dielectric Nanocomposites With High-k Interfacial Layers

Current dielectric nanopowder research in wide bandgap insulators has revealed an order of magnitude increase in dielectric permittivities over bulk values [1]- [3]. Researchers in this area postulate that interfacial layers at material boundaries have enhanced polarizability because of local relaxation mechanisms, such as rotational dipoles or localized space charge polarization [1], [3] . Equivalent macroscopic electromagnetic models that have low-k grain/particle cores with average diameters ranging from 5 to 40 nm and high-k thin encapsulating layers with thicknesses from 0.8 to 1.0 nm are simulated in this paper using finite-element method and control-volume approach. The purpose of this investigation is to explore the impact of the microstructure on the overall electrical properties of these types of materials. To help validate this simulation approach, laminate geometries of silicon dioxide and silicon nitride are fabricated and tested by the authors. From these test samples, an effective permittivity model of interfacial layers for simple laminate geometries is extracted and is used to predict the macroscopic behavior of more complicated nanocomposites. Furthermore, the impact of the conductivity of low-k cores is also explored with the assumption that the high-k interfacial boundaries remain perfectly insulating. For low-conductivity cores, it is shown that the real part of the permittivity decreases with increasing particulate/grain size that is generally consistent with nanopowder research in [3] . However, at higher core conductivities, this dependence completely changes such that the real permittivity increases with increasing particulate/grain size. This change in dependence is due to charges accumulating at core boundaries producing a Maxwell-Wagner effect.