580 THE BELL SYSTEM TECHNICAL JOURNAL, MAY 1953 



ture in degrees Kelvin and R the Boltzman constant for one gram mole 

 of the material. R is closely equal to 2 calories per degree K. Hence, if 

 H is expressed in calories per gram mole and the value of K is obtained 

 to fit equation (15), we have 



34,500 



T = 9.2 X 10"'' X e"2^ . 



The constant K is close to that given by the Langmuir-Dushman' theory 



J. _hN _ 6.62 X 10-^^ X 6.06 X 10^^ _ ,5 .... 



^"H 34,500 X 4.187 X 10^ " '^ ^ ^ ' ^^^^ 



where h = Planck's constant equal to 6.62 X 10"^^ ergs, A^" is Avogadro's 

 number equal to 6.06 X 10 and H is the activation energy expressed 

 in ergs. Su K6 has shown that the activation energy for grain boundary 

 slip is essentially the same as for self diffusion and for creep. 



Similar results have been found for a-brass and a-iron. These have 

 activation energies shown by equation (17) 



a-brass 41 kilocalories per mole, 



(17) 

 a-iron 85 kilocalories per mole. 



Although measurements have not been made for copper, the activation 

 energy of self diffusion is about 57.2 kilocalories^ per mole, but 39.9 

 kilocalories for the principle impurity silicon. 



All of these measurements were made for strains under 10~*, and the 

 question arises as to whether these concepts are valid for the much 

 higher strains experienced in the wrapped solderless joint. From the 

 photoelastic pictures. Figs. 16 and 17, it is obvious that the greatest 

 stress inhomogeneity occurs in the neighborhood of the corners and 

 flow will take place in such a way as to relieve the high stress concen- 

 tration. This will have the effect of making the terminal and wire mate 

 even closer and may result in a slight transient lowering of the hoop 

 stress. After the initial formation, however, it will be the long time re- 

 laxation of the hoop stress in the wire that determines the lasting quality 

 of the joint. 



As discussed previously, the twist that the terminal takes is deter- 

 mined by the mean value of the hoop stress in the wire, and any relaxa- 

 tion in this hoop stress can be studied by observing the angle of twist 

 as a function of time and temperature. By using a long terminal wound 



» S. Dushman and I. Langmuir, Phys. Rev. 20, (1922) p. 113, 1922. 

 • Zenner, Elasticity and Anelasticity of Metals. Table 12, p. 98, Chicago Uni- 

 versity Press. 



