P H Y S Microwave Disk Resonator

In order to test the microwave properties of superconductors, we are developing a disk resonator, to be operated in the TM010 mode at 3.36 GHz. The goal is a measurement of microwave surface resistance at both low and high power, initially for niobium thin films, but eventually for other thin film superconductors. The currents in a disk resonator flow in and out radially. Edge currents are zero. Other types of resonators suffer from “current crowding” at the edges, with the result that the intrinsic material properties are difficult to deduce from the experimental results. CNF facilities are used for this project, since there is a need for precise dimensions, including the input and output couplers with micron-level features. Summary of Research: The resonator is, in principle, extremely simple. It consists of a 500 nm niobium film ground plane, sputtered at CNF on a 50.8 mm diameter single crystal sapphire disk which is 330 μm thick. This is separated from a smaller niobium disk 3.4 cm in diameter sputtered on a second sapphire disk substrate. These two disks are spatially separated by a third sapphire disk which acts as the low-loss dielectric. Figure 1 shows the general layout. The resonator patterns were laid out with L-Edit. Masks were made with the Heidelberg DWL laser pattern generator and direct writer. Photoresist SPR 220-3.0 was selected after experimenting to determine the best results. We used a reversal process, with an exposure of 16 seconds in the HTG contact aligner, followed by 90 minutes in the YES-58SM image reversal oven. After a 60 second flood exposure in the contact aligner, we developed the patterns in 321 developer. This proved satisfactory for the disk and the ground plane, but the lift off would not reproduce the small features of the coupling capacitor well enough for satisfactory high power operation. Now we are experimenting with other techniques, both lift-off and etching. Figure 2, taken with a Zeiss secondary emission microscope (SEM) from a niobium lift-off pattern made by the Schwab group [1]. It illustrates some of the problems which can be encountered with niobium lift-off. It also shows the grain size of the niobium which was sputtered in the CVC sputter deposition system. The small 100 nm niobium grains will degrade high power performance [2]. Coupling to the resonator depends on the type of measurement. Low power and weak coupling allows determination of the superconducting surface resistance Figure 1: Layout of disk resonator (plan view).