796 THE BELL SYSTEM TECHNICAL JOURNAL, MAY 1957 



Part II — Activating Carbon 



4. COMPOSITION OF ACTIVATING POWDER AND RATE OF PRODUCTION 



Experiments have been carried out designed to discover the chemical 

 composition of the carbonaceous material responsible for activation, 

 how much is made per unit of energy in an arc, and where it is made. 

 Now it has been pointed out above that the reason for the uniform erosion 

 in an active arc is the burning off of this black powder by arcs and the 

 consequent continual wandering of successive arcs always to neighboring 

 spots from which the powder has not been burned. This burning off of 

 black powder makes quantitative measurements in air of its rate of 

 formation quite impractical. Activation in vacuum avoids the destruc- 

 tion by burning, and makes possible direct measures of rate of forma- 

 tion; in these tests, the chemical composition of the powder can be found 

 also. All of the quantitative studies of activation in vacuum were made 

 by P. Kisliuk, but the results have not been previously pu))lished. 



In Kisliuk's experiments two electrodes, which were of platinum, were 

 mounted in a glass chamber so they could be operated by the magnetic 

 field of a coil placed outside the chamber, in the manner of a dry reed 

 switch. The electric circuit was arranged to discharge on each closure a 

 capacitor charged to a fixed voltage, with no current flowing in the cir- 

 cuit as the platinum contacts are separated. Air was pumped out and 

 the contacts operated in benzene vapor at a constant rate, discharging 

 the capacitor a convenient number of times per minute. Every experi- 

 ment consisted of measuring the pressure in the system, from which 

 was deduced the rate of disappearance of benzene and the rate of evolu- 

 tion of hydrogen resulting from its decomposition, hydrogen being dis- 

 tinguished from benzene by freezing out the latter in liquid nitrogen. 

 The pressures were measured by an RCA thermocouple gauge (1946) 

 which was sho^^^^ in control tests not to produce benzene decomposition. 

 The benzene, which had been distilled repeatedly to remove water vapor, 

 was used at initial pressures not to exceed 10~- mm Hg determined by 

 a dry ice-acetone bath. The experimental arrangement is sho^Mi in 

 Fig. 9. 



4.1 Composition 



In the first experiments with this system it was found, as had been 

 expected, that with continued operation of the contacts in benzene vapor, 

 the pressure rose steadily, although benzene continued to disappear. 

 The pressure changes corresponded to the evolution of 3.2 ± 0.6 mole- 

 cules of H2 for each vanishing molecule of benzene, agreeing well with 



\ 



A 



