High-Temperature Microfluidics for Space Propulsion

In this thesis, microfabrication methods and tools for analysis of heated cold-gas microthrusters are presented, with the aim of improving their reliability and performance. Cold-gas thrusters operate by accelerating pressurized gas through a nozzle. These thruster systems are very straightforward in both design and operation, relying on little more than a pressurized tank, a valve, and a nozzle. This makes them suitable for miniaturization, enabling their use on very small spacecraft. However, an inherent drawback with cold-gas thrusters is their low propellant efficiency – in thrusters known as specific impulse, or Isp. This is compounded by the fact that when reducing length, the volume, e.g., that of the propellant tank, reduces with the cube of the length, meaning that the maximum amount of storable fuel reduces quickly. Hence, maximizing fuel efficiency is even more important in miniaturized systems. Still, because of their other advantages, they remain suitable for many missions. Schlieren imaging – a method of visualizing differences in refractive index – was used thrughout this thesis to visualize exhaust jets from microthrusters, and to find leaks in the components. It was found that effects of the processing of conventionally fabricated silicon nozzles, resulted in a misalignment of up to 3° from the intended thrust vector, increasing propellant consumption by up to 5%, and potentially causing unintended off-axis acceleration of the spacecraft. Schlieren imaging was also used to verify that the exhaust from thrusters fabricated with close to circular cross-sections was well behaved. These nozzles did not suffer from the previous misalignment issue, and the shape of the cross-section decreased viscous losses. For applications requiring higher temperatures, a microthruster nozzle with an integrated flow sensor was fabricated from tape cast yttria stabilized zirconia. The ceramic substrate enabled heater temperatures of the nozzle exceeding 1000 °C, resulting in an increase in Isp of 7.5%. Integration of a flow sensor allowed the elimination of couplings and reduced the number of interfaces, thereby reducing the overall risk of failure. Close integration of the sensor allowed moving the point of measurement closer to the nozzle, enabling improved reliability of the measurements of the propellant consumption. The temperature of the heater, in combination with the ion conductive properties of the substrate proved to be a limiting factor in this design. Two routes were explored to overcome these problems. One was to use the temperature dependence of the ion conductivity as a sensing principle, thereby demonstrating a completely new flow sensor principle, which results in better calibration, tighter integration, and 9 orders of magnitude stronger signal. The other was using hafnium oxide, or hafnia, as a structural material for high-temperature micro-electromechanical systems. This involved developing a recipe for casting hafnia ceramic powder, and determining the Young's modulus and thermal shock resistance of the cast samples, as well as studying the minimum feature size and maximum aspect ratio of cast microstructures