Spin coating of passive electroactive ceramic devices

This thesis reports an extensive body of research undertaken to provide information relating to the potential integration of several passive electronic components, namely, multilayer ceramic capacitors (MLCC), ferrite inductors and thick film resistors. The specific materials concerned are barium-titanate based dielectrics, a ferrite inductor paste and a ruthenium-based resistor paste. The central objective is to investigate the potential for spin coating of standard and modified slip/paste formulations for use in the production of well defined layers of the dielectric and ferrite materials. Aspects of this technology, which might restrict co-deposition of these systems, have been addressed. In addition, their potential integration with ruthenium oxide resistor films has been explored. Layers of the main materials, obtained by standard commercial processing methods of screen printing and doctor blading, have been used for direct comparison. Extensive characterisation has been carried out on the materials in the powder form, both before and after thermal processing. These data then act as a benchmark for the key materials properties in their subsequent analysis in thin film layer form. The analytical techniques used include: Thermogravimetric Analysis (TGA), Scanning Electron Microscopy (SEM), Energy-Dispersive X-ray Analysis (EDX), X-ray Diffraction (XRD) and X-ray Photoelectron Spectroscopy (XPS). Characterisation of the barium titanate based dielectric ceramic and ferrite inductor paste materials as powders before and after thermal processing in the range 150, 500, 850 and 1150 deg C indicates a high carbon content in the surface region. By comparison, spin coated layers of each of these systems on alumina substrates before and after heating in the same temperature range as that used for the residual powders, showed a marked decrease in the carbon content in the surface region. In addition, deposition of the dielectric onto a ferrite surface which itself had been spin coated on alumina indicated no merging of the different layers. Thermal processing of this bi-layer system up to 850 deg C provides evidence that there may be some channelling of the bismuth photoelectrons through suitably sized pores in the dielectric layer. The commercial doctor blading method has also been used to produce dielectric layers and these are shown to be somewhat different to those produced by spin coating. In particular, the doctor bladed material contains more carbonaceous material in the surface region compared to the spin coated material and the additives are not detected in the former but are in the latter. The resistor material, in screen printed form, has been characterised before and after thermal processing up to a temperature of 850 deg C. The data obtained clearly indicates the presence of a lead silicate glass frit on the surface of this material after thermal processing to 850 deg C. This has clear implications for the deposition of layers of this material directly on to dielectric or ferrite surfaces. Finally, a series of MLCC devices have been fabricated using spin coating to produce barium titanate based dielectric layers from both standard and modified ceramic slip formulations. In all cases the standard commercial screen printing technology has been used for the production of the associated electrode arrays. The structural and electrical properties of the model MLCC devices produced by this spin coating/screen printing technological approach have been shown to be of extremely high quality. In terms of the normal specifications of capacitance (11020 pF), dissipation factor (1.31 %), monitored insulation resistance (5.37 x 1010 Ω), voltage breakdown (1412 V) and porosity (1.61 %) they compared very favourably with a nominally equivalent set produced from the commercial doctor blading/screen printing technique. Hence, this body of work has shown that spin coating can be used very effectively to produce inter-layers of materials in the fabrication of passive electronic devices. Moreover, important information has been acquired concerning the potential of the integration of these three commercially important materials. Further work is now required to optimise the technology such that interconnection between the passive devices concerned can be reduced. This should then lead to significant savings on space in future generations of many electronic devices. (author)