800 



THE BELL SYSTEM TECHNICAL JOURNAL, MAY 1957 



it has been noted that area increases less rapidly than linearly with 

 •energy, but it is certain that no universal rule applies in all cases; for 

 example, by restricting the total electrode area, the over-all arc area can 

 be forced to be constant independent of energ3^l° One can conclude only 

 that all of the facts are accounted for by a benzene layer several mole- 

 cules thick on the electrode surfaces with decomposition by each arc of 

 all of the benzene within its over-all area. 



We are now in a position to consider the much lower rate of decompo- 

 sition of benzene during the initial period before the electrodes became 

 active. In Fig. 10, this lower initial rate is 1.1 X 10^^ molecules per arc. 

 This is somewhat less than the number of molecules calculated to lie in 

 a monolaj^er on the surface covered by an arc. The area used in this cal- 

 culation was that of a cathode arc, but it is well kno\vn that before con- 

 tacts become active a large proportion of the arcs are anode arcs which 

 have smaller areas (Reference 9, page 1088). The estimated area may, 

 however, be about correct because in the case of an anode arc, carbon 

 is decomposed by heat over an area larger than that of the arc itself. 



Within the precision of the estimates we are able to make, it can be 

 said that for inactive contacts operating in benzene vapor each arc de- 

 composes a single layer of adsorbed molecules of benzene. After the 

 contacts become active, the amount decomposed by each arc is greater 

 and is the equivalent of several layers of molecules. It was surmised long 

 ago that much of the vapor adsorbed on active contacts is held by carbon 

 already on the surface rather than by the surface metal. The increased 



ylO 



16 



Q 



UJ 



(X. 



o 



z 



O 4 

 CD 



a. 

 < 



LU 



03 



2 



Z 



10 15 20 25 30 



THOUSANDS OF OPERATIONS 



35 



40 



Fig. 10 — Measurements by P. Kisliuk of the amount of carbon formed at arc- 

 ing platinum contacts, each arc 1,250 ergs. The final slope represents the produc- 

 tion of 3.5 X 10~'^ gm of carljon per erg of arc energy, or one benzene molecule 

 decomposed for every 150 electrons flowing in the arc. 



