Measurements on Superconducting Contacts

The contact resistance between crossed wires has been measured as a function of the current at various temperatures in the liquid helium range. Most contacts were stable enough to establish a "diagram of state, " i. e. , determine curves of constant resistance in I-T space. The following facts have been established: 1. The critical temperature of contacts between clean wires of tin is suppressed by about 0. 2^ due to the pressure on the contact. 2. The addition of a copper layer on one or both of the wires reduces the critical currents, but hardly influences the critical temperature. This can be understood if one assumes that the density of the superconducting electrons decreases in the copper layer, thus producing an increase in the penetration depth. At layer thicknesses of several hundred angstroms the penetration depth becomes large compared to the contact radius, and the critical current vs temperature curve approaches that of a very thin wire. 3. Contacts between tin and copper wires below the critical temperature and at low currents show a constant resistance, which rises sharply at a critical current. Graphs of this quasicritical current as a function of the temperature have been obtained. 4. Clean contacts between tin and indium usually behave as one would expect from the foregoing. In one case out of three, however, the resistance was strongly dependent on the current at a temperature as high as 4. 2^. Plotting the low-current values of the resistance as a function of the temperature showed a behavior as if one of the contact materials had a critical temperature of 5 to 6^. A search of the literature revealed that the Sn-In system has two intermetallic compounds. The compound In₆Sn₂ has been prepared, and was found to have a critical temperature of about 5. 5^. Since the contacts were closed at 4. 2^ and no possibility of transfer of metal from one side to the other existed, it must be assumed that the proximity of the tin to the indium is sufficient to produce partial superconductivity by way of long-range correlation of the electronic wave functions.