ACTIVATION OF ELECTRICAL CONTACTS BY ORGANIC VAPORS 795 



bonaceous material lying upon the other. If one calculates striking field 

 by dividing the potential by the separation between the metal electrodes, 

 a very low value is obtained, but this is the field at which electrostatic 

 forces cause movement of carbon particles to decrease the separation; 

 the true field at which the arc finally strikes between carbonaceous mate- 

 rial and the opposing electrode is not significantly lower than the striking 

 field for arcs at clean surfaces. Some or all of the local carbonaceous 

 material is burned up by the arc, and metal vaporized from one of the 

 electrodes is soon fed into the arc so that for most of its life the ions of 

 the arc are metal ions supplied by atoms from one or the other of the 

 electrodes. This is true for even the most heavily carbonized electrodes. 



There is a critical electrode separation, characteristic of the metal of 

 the electrodes, which determines whether the arc is an anode type of 

 arc with metal supplied by the anode or an arc of the cathode type with 

 metal supplied by the cathode. If the separation is greater than this 

 critical value the arc is a cathode arc, and less than this value an anode 

 arc. This critical distance is about 0.5 X 10"'* cm for palladium electrodes 

 and of the order of 3 or 4 X 10~* cm for electrodes of silver. The ratio 

 of these distances is somewhat greater than the ratio of the square roots 

 of the electrical conductivities of the metals at their melting points. The 

 critical distance for palladium is so small that all arcs at active palladium 

 surfaces are cathode arcs. For silver, on the other hand, the critical dis- 

 tance is so large that most arcs at low voltages at silver surfaces are 

 anode arcs. In any practical application of silver electrodes, the car- 

 bonaceous material formed is rarely or never in a sufficiently thick layer 

 to result in cathode arcs for closure at low voltages. In the case of sepa- 

 rating silver electrodes, an active arc may last until the electrode separa- 

 tion is beyond the critical distance for silver; the erosion occurring after 

 this distance is reached is predominantly from the cathode, and the 

 larger net loss may, on occasion, be from the cathode. 



The erosion resulting from repeated arcing at active surfaces is differ- 

 ent in character from that produced by inactive arcs. Inactive arcs give 

 rise to a crater on one electrode and a matching mound on the other, 

 with most of the metal from the crater transferred to the mound. Active 

 arcs, on the other hand, produce smooth erosion without craters and 

 mounds, often with considerable net loss of metal which appears mixed 

 with carbon as a black powder. This smooth erosion is accounted for by 

 the striking of each new arc on carbon formed by preceding arcs, to- 

 gether with the burning off of carbon at the center of each arc and the 

 formation of new carbon around its periphery. 



