786 THE BELL SYSTEM TECHNICAL JOURNAL, MAY 1957 



higher minimum arc currents for cathode arcs at clean surfaces is attrib- 

 uted to the higher thermal and electrical conductivities of metals and to 

 the absence of loose material making poor contact with the surface. 

 This picture is supported by the observation that metal contacts are 

 made temporarily active by almost any kind of loose surface particles 

 of very small size (Reference 2, Page 961). 



2.4 Erosion 



2.Jf{a) Palladium and Platinum. Further evidence that an active arc at 

 palladium or platinum surfaces is always a cathode arc is furnished by 

 the fact that the cathode loses much more metal in an active palladium 

 arc than does the anode. (See also Reference 4, Table I). 



The direct way of proving that an active arc at palladium or platinum 

 surfaces is a cathode arc would, of course, be microscopic examination 

 of the contact surfaces after a single arc. This is not practicable because 

 surfaces become active only after repeated arcs, but one can do what is 

 apparently quite equivalent by looking at the damage done by a single 

 arc to surfaces on which small carbon particles have been dusted. Ex- 

 periments by Haworth do indeed prove that arcs at such surfaces are 

 cathode arcs, even at the Ioav striking potential of 50 volts, and when 

 the maximum diameter of the carbon particle is only 1 X 10~^ cm. Fig. 

 7(b) is typical of many examinations by Haworth of palladium cathodes 

 after a single arc at surfaces upon which carbon particles had been de- 

 posited. The striking potential was 50 volts and the capacitance that 

 was discharged was C = 10~^ /, so that the energy C{Vo — v)v was 50 

 ergs. For comparison, photographs are reproduced in Figs. 7(a) and 



.if. 





^ ^ .\^ 



.*»•- 



Fig. 7 — Photomicrographs of palladium cathode surfaces after single cathode 

 arcs. The photograph of (b) was obtained after a 50 erg arc with 50 volt striking 

 potential at a surface upon which carbon particles has been deposited. This sort 

 of cathode damage was observed for all of the different sizes of carbon particles 

 which were tested, even for the smallest having diameters of onlj^ 10"'* cm. The 

 comparison photographs (a) and (c) represent the damage done respectivelj' by 

 40 erg and 80 erg arcs to palladium surfaces without carbon particles, each arc 

 at the striking potential of 400 volts. 



