|
Commercialisation of High Temperature Superconducting microwave circuits to communication systems depends on losses of HTS films exhibited at high frequencies. The losses must be much lower that those of conventional metallic films to compensate for the change of technology and need for cooling. Hence accurate measurements of surface resistance have been of primary importance almost since the discovery of High Temperature Superconductors in 1987. Resonant and direct transmission methods, used in the past for metals and Low Temperature Superconductors, have been employed for the microwave characterisation of YBaCuO, B1SrCaCuO and ThBaCaCuO films. After a copper cavity, a parallel plate resonator and a confocal resonator, recently a dielectric resonator has become the most popular technique for loss measurements of HTS films at microwave frequencies. The application of dielectric resonators (DRs) to characterisation of HTS materials is due to possibility of high sensitivity, resolution and big range of surface resistances that can be measured. This application followed use of dielectric resonators for measurements of surface resistance of metals [1-3] and of dielectric materials, benefiting from developments in Low Phase Noise Stable Oscillators and Filters. The first use of DRs to microwave characterisation of HTS materials was proposed (to the knowledge of authors) in 1989 by Fieduszko and Heideman [4]. Since then several groups [5-16] have developed variety of designs for this purpose. For the dielectric resonator technique the surface resistance of superconducting films, Rs, is calculated from the loss equation, similarly as for any other resonator technique, namely from (1) Q Q Qd Qrad where Q0 is the unloaded quality factor and Qd and Qrad are the quality factors related to conductor, dielectric and radiation losses respectively. O-8194-2071-9/96/$6.OO SP!E Vol. 2697 / 77 Conductor losses are directly determined by the surface resistance of superconducting parts and of metallic parts, and dielectric losses by the loss tangent of the dielectric rod according to: ._!:._ = + Rsmet Q A Amet —=pdtanö (2) where A, Amet and Pd are geometrical factors of the superconducting parts, the metallic parts and of the dielectric rod, tanö is the loss tangent of the dielectric and Rsmet is the surface resistance of the metallic parts. The factors A and Amet depend on the design and dimensions of the resonator. The constant Pd for resonators in the trapped state is approximately equal to unity, only weekly depending on dimmensions. The unloaded Quality factor Q0 is determined from the loaded quality factor, calculated from the transmission or reflection coefficients measured around the resonance. In general to achieve high sensitivity and high resolution of the dielectric resonator technique for measurements of surface resistance of HTS films, radiation and dielectric losses of the resonator should be small as compared to losses in conducting parts. Also losses in HTS films under test should be large as compared to losses in metallic parts of the resonator. To ensure small errors in the surface resistance values, Qfactors should be measured with high accuracy and geometrical factors should be known exactly. It is however often not feasible to fulfil all these requirements in practice. Usually a resonator's design, choice of dielectric or computational models used for calculation of geometrical factors may impose significant limitations on accuracy of a system for characterisation of HTS materials. This paper (not pretending to cover fully this complicated issue) discusses some problems related to errors which may be encountered when dielectric resonators are used for surface resistance measurements of superconducting films.
|
