1102 THE BELL SYSTEM TECHNICAL JOURNAL, SEPTEMBER 1954 



temperature, no adhesion occurred even when the time of loading was 

 very long. It was also found that if the load was removed, even mo- 

 mentarily, before the sample was placed in the oven, no adhesion oc- 

 curred. For a given temperature there is an induction period before the 

 wire adheres to the block, and this induction period increased, in general, 

 as the temperature decreased. This induction period was found to be a 

 time of nucleation. This is shown by the fact that the period increases 

 as the square of the contact dimensions. Since diffusion is a function of 

 x/\/Dt where x is the distance, D the diffusion constant and t the time, 

 this observation shows that nucleation starts at a given point and pro- 

 ceeds for a certain fraction of the contacting surface before fusion 

 strengths are observed. After the induction period, the fusion force 

 — the force required to pull wire and block apart ■ — increased at a 

 rapid rate for the cases of copper, tinned copper, and zinc wire on a 

 nickel silver block. 



In order to account for the effect of fusion due to the increase of con- 

 tact area, the ratio of the fusion force to the contact area was determined, 

 yielding a shearing stress. The area of contact could be easily measured 

 with a microscope since a bright surface was produced hy the shearing 

 process. The shearing stress, Fig. 8, for the case of tinned copper wire 

 on nickel silver is shown to approach a limit of about 3,000 psi which is 

 approximate^ the limiting shearing stress for tin. Hence, it appears 

 that tin diffuses into the copper wire and the nickel silver base. Further- 



TIME IN SECONDS 



Fig. 7 — Diffusion forces for tinned and bare copper and zinc wire on nickel 

 silver base. 



