810 THE BELL SYSTEM TECHNICAL JOURNAL, MAY 1957 



be expected to be formed by each arc. The model is not sufficiently well 

 defined to permit any more exact conclusions. 



In another experiment, silver electrodes, which had first been com- 

 pletely activated, were operated for a long period at a greatly reduced 

 benzene vapor pressure. It was found that they remained completely 

 active unless the pressure was very much less than the minimum of 0.7 

 mm Hg at w^hich activation could be produced. In repeated tests at a 

 variety of low benzene vapor pressures, the number of operations re- 

 quired for the contacts to become inactive was recorded. This number 

 was found to increase very abruptly with increasing vapor pressure, and 

 above about 0.02 mm Hg the contacts remained active indefinitely. 

 This result must again be related to the capacity of a mass of spongy 

 carbon to hold a great amount of adsorbed benzene. Very probably the 

 upper limit of pressure below which contacts cannot be deactivated, 

 depends upon the thickness of the carbon layer produced before the 

 benzene pressure is lowered. 



Similar but less extensive experiments were carried out with palladium 

 electrodes. 



These hysteresis effects observed in the activation and deactivation 

 of contacts seem capable of explaining the erratic observations that had 

 been made previously. If the immediate history of contacts is sufficiently 

 well known, behavior can perhaps be predicted fairly well for various 

 experimental conditions. 



7. BROWN DEPOSIT 



Closely related to the activation of relay contacts is the formation of 

 polymerized layers of organic material upon contact surfaces as a result 

 of friction. This material, which is commonly kno\^^l as "bro\Aai deposit", 

 is produced at contacts which do not make or break current. Its mode 

 of formation is thus entirely different from that of the carbon which is 

 the cause of activation. Both have, however, a common origin in layers 

 of organic molecules adsorbed upon surfaces. Discovery of broAMi deposit 

 and most of the investigation of it were carried out elsewhere (Ref. 20), 

 but some discussion of brown deposit is appropriate here because of its 

 relation to the carbon of activation and because of a study of its forma- 

 tion by P. Kisliuk. 



7.1 Composition 



The composition of brown deposit was determined by Kisliuk in an 

 apparatus similar to that used to investigate the carbonaceous material 

 responsible for activation, Fig. 9, and by the same analytical procedure. 

 The apparatus was modified so that a palladium or platinum electrode 



I 



