The use of thin-film sensors has several advantages over wire or foil sensors. For example, thin-film sensors do not require special machining of the components on which they are mounted, and, with thicknesses less than 10 μm, they are considerably thinner than wires or foils. The thin-film sensors are thus much less disturbing to the operating environment, and have a minimal impact on the physical characteristics of the supporting component.
Fabricating metal-based sensors directly onto metal components requires the use of an electrically insulating layer between the component and the sensor. Flame-sprayed insulators provide good insulating capability, but the thickness of the coatings (300 μm and greater due to the coating porosity) lessens the advantage of the thin-film sensor. Thus, a much thinner electrical insulating layer (barrier) is highly desirable, and is the motivation behind this work.
The requirements for such an electrical insulator are driven by the environments of aerospace applications, which currently require insulation of 100 kΩ or better to 1,100 °C in oxidizing conditions. For a typical thin-film sensor active area of 1 mm², this electrical insulation thickness is estimated to be at least 3.2 μm of theoretically pure, fully dense aluminum oxide. The insulator must have its surface area free of defects that reduce the thin-film sensor active area or cause an electrical short to the substrate. The defect size should be less than 1 μm² in area with less than 1 per 1 mm² of the typical thin-film sensor active area.
The application of alternative thin-film electrical insulators has been examined to further minimize the insulation thickness. However, ceramics lose their electrical resistance exponentially with increasing temperatures. Short circuit paths at grain boundaries and film defects can further reduce the insulation properties from the bulk values at high temperatures.
The major technical challenge in multilayered electrical insulating films is the optimization of the fabrication process with respect to material composition to achieve a reliable, high-temperature insulating film. The coefficient of thermal expansion of both the substrate material and the insulation material must be taken into account for the formation of defects, as well as surface chemistry that could influence film adhesion. Thus, insulation schemes are specifically tailored to substrate and application.
A method of insulation was needed for large stainless steel components using minimal surface preparation and a single magnetron sputtering source. A thin-film multilayer insulation of Al2O3/SiC/Al2O3 was demonstrated to provide the necessary electrical insulation for thin-film sensors. Test sensors were fabricated on the insulating films on stainless steel substrates, and survived welding lead wires to the sensors without shorting to the substrate. The test sensors also survived annealing to 800 °C for one hour without failure of the sensor films or lead wires.
The optimized series of depositions using a single magnetron sputtering source was 2.6 μm of aluminum oxide on the stainless steel substrate, followed by 60 nm of silicon carbide, followed by 1.3 μm of aluminum oxide. After each deposition, the chamber is vented to room air and the sputtering target changed to the appropriate material.
Though this method of electrically insulating a metal substrate with thin-film multilayers was only demonstrated withAl2O3/SiC/Al2O3 on stainless steel alloy 316, there is no reason that the method could not be used on any metal substrate that forms a passivating oxide coating such as Inconel or TIMETAL, which both are a close match to the CTE (coefficient of thermal expansion) of Al2O3. The use of SiC is preferred due to its ease of use, but ideally any reactive carbide such as Cr3C2, Mn3C, Fe3C, Co3C, and Ni3C could be applicable as well.
The use of Cr3C2 and SiC as intermediate layers with Al2O3 has been shown effective in creating pinhole-free thin-film electrical insulation on stainless steel. It should be noted that the use of the oxides SiO2 and YSZ as intermediate layers has not been found effective.
This work was done by Charles A. Blaha, Gustave C. Fralick, and John Wrbanek of Glenn Research Center. NASA is seeking partners to further develop this technology through joint cooperative research and development. For more information about this technology and to explore opportunities, please contact http://technology.grc.nasa.gov . LEW-19239-1