Electrical conduction in solids I. Influence of the passage of current on the contact between solids

Abstract

The effect of passing an electric current through the interface between two contacting pieces of gold has been investigated, and it has been shown that the current can cause appreciable changes in the true area of contact between the surfaces. This phenomenon has been studied by measuring the associated alterations in the constriction resistance. (This is the resistance caused by the constriction produced in the current stream as it passes through the tiny areas of contact between the metals.) It is shown that the response of the region of contact may be explained as a result of the heat generated in this resistance by the current. For any given current there is a certain critical degree of constriction through which it will just pass with ­out causing a permanent change in the contact region; if the current flows through a contact area which presents a constriction resistance greater than this critical value, then the heat generated will be sufficient to cause the yield pressure of the metal near the interface to fall, and the area of contact will increase accordingly. The critical constriction resistance associated with any current has been found to be inversely proportional to the magnitude of the current. The response of the contact region to short-duration pulses of current has been studied. The results show that the behaviour is independent of the length of the pulse in the range investigated (10 μ s to 10 ms). They also indicate that when a short pulse of current passes between the pieces of metal mechanical collapse will occur only if the current is sufficiently large to cause melting of the metal near the interface. It is possible to calculate the temperature in the contact region from the potential difference developed across the constriction. Calculations based on the accepted mathematical treatment indicate that mechanical collapse occurred in these experiments when the temperature at the interface was raised to about 950 °C. (This is significantly below the melting point of gold, 1063 °C.) This result was not supported by direct examination of the specimens, which showed clear evidence of melting whenever collapse occurred. It is suggested that the accepted mathematical treatment of constriction resistances may not be valid when the temperature approaches the melting point; and in part II a new treatment, which accounts for the observed results, is derived.