ACTIVATIOX OF ELECTKICAL COXTACT.S BY ORGANIC VAPORS 811 



could be rubbed back and forth upon another electrode of the same mate- 

 rial. The driving force was a magnet outside the glass apparatus. 



In tests carried out in benzene vapor in the absence of air, it was found 

 that the deposit formed on the electrodes contained 65 per cent as much 

 hydrogen as was in the original benzene, about 2 atoms of hydrogen for 

 every 3 carbon atoms, this figure having a possible experimental error of 

 as much as 20 per cent. The bro^\^l deposit formed by friction thus differs 

 significantly from the pure carbon produced by arcing which is responsi- 

 ble for activation.* The experimentally determined composition of the 

 brown deposit does not, of course, distinguish between hydrogen or 

 benzene simply adsorbed in the deposit and hydrogen existing in it in 

 some combined form, 



7.2 Rate of Production 



In Kjsliuk's experiments, which were carried out in the absence of 

 air, the rate of production of bro^\^l deposit was found to be independent 

 of benzene vapor pressure do^^'^l to 3 X 10~^ mm Hg, which was the 

 lowest pressure tested, just as was the case in the formation of carbon 

 bv arcs. 



When air is present, the rate of formation of bro^^^l deposit maj^ de- 

 pend upon vapor pressure of the organic molecules. Unpublished experi- 

 ments have indicated, furthermore, that there may be a limiting vapor 

 pressure below which the deposit does not form, with this pressure de- 

 pendent upon the idle period between operations.-" 



In some of Kisliuk's vacuum tests a palladium electrode was rubbed 

 back and forth over an area determined microscopically to be about 

 4 X 10~* cm^, and produced the polymerization on each rub of 2.1 X 

 10^" molecules of benzene, or 5 X 10^- molecules per cm- of rub. This is 

 smaller than the nvmiber of molecules in a monolayer (7 X 10^^ per cm^, 

 Reference 16) by a factor of 140. Part of the discrepancy is certainlj^ due 

 to the fact that the true area of contact of the electrodes is less than the 

 apparent area as seen under the microscope. From more careful estimates 

 of area it has been found bj^ other observers that the amount of benzene 

 that is polymerized by friction is, in general, comparable with that ad- 

 sorbed as a monolayer on the rubbing surfaces. 



7.3 Brown Deposit and the Carbon of Activation 



Although both bro^ni deposit and the carbon of activation are pro- 

 duced from the decomposition of adsorbed organic molecules, there are 



* In this connection, it is interestinj^ to point out, however, that any metal 

 surface, ui)on wliich l)ro\vii (le])osit has been prothicetl l)y friction in an appro- 

 priate atmosphere, is found to l)e fully active when tested in a suitable circuit. 

 This activity naturally does not last after the brown deposit has been burned off. 

 In this characteristic, the brown deposit behaves like any foreign more or less 

 insulating laj-er upon a contact surface. 



