High-Frequency Response of Josephson Point Contacts

Abstract

For a current‐driven Josephson junction shunted by an Ohmic resistance, the dc voltage response and impedance to external high‐frequency currents is calculated with a second‐order perturbation method based on the unperturbed solution for the time evolution of the voltage as given by Aslamazov and Larkin. The response is proportional to signal power and has three characteristic‐frequency regions depending on whether the internal self‐generated Josephson frequency, ω₀, is larger than, equal to, or smaller than the signal frequency, ω. For ω₀< or >ω the response could be expressed by easily observable parameters of the dc voltage‐current characteristics. This permits comparison of the predictions with experimental results obtained on point‐contact junctions whose dc characteristics were only approximately represented by the simple model chosen for analysis. For ω₀≪ω the response is found to be proportional to (i) the slope of the dc characteristic, (ii) the inverse square of the applied frequency, (iii) the inverse battery current in terms of the critical current, and (iv) the voltage amplitude of the inherent Josephson oscillations. For ω₀≫ω ``classical'' detection proportional to curvature is obtained. For ω₀∼ω resonance detection occurs, which depends to a considerable degree on fluctuations of the average contact voltage. These predictions are in reasonable agreement with experiments in which point‐contact junctions were illuminated with klystron radiation at 10 and 90 GHz. Evidence is presented that the voltage fluctuations limiting the resonance response are the result of noise generated in the junction. Comparison of the noise‐equivalent power expected from the simple model with measured results suggests that parasitic elements, such as shunt capacity, are not negligible at 90 GHz.