Februaey 24, 1922] 



SCIENCE 



211 



and the radiating potential at 10.14 volts, it 

 seemed that it should be possible to maintain 

 an are at 13.52 volts or even as low as 10.14 

 volts. The failure to maintain an arc at these 

 potentials was ascribed to the insufficient 

 amount of monatomic hydrogen in the tube. 

 During the course of this investigation, Pro- 

 fessor K. T. Compton suggested that it might 

 be possible to dissociate hydrogen by means 

 of a cylindrical tungsten furnace which could 

 also be used as one of the electrodes for the 

 arc. The writer undertook the investigation 

 of the possibility of this and computed the 

 per cent, of monatomic hydrogen which would 

 be in equilibrium with the diatomic gas on the 

 basis of Nernst's equation of the "reaction- 

 isobar."^. Taking the heat of dissociation to 

 be 84,000 calories per giam, j8 = 0.000225, 

 and the chemical constants for the diatomic 

 and the monatomic hydrogen to be^ -3.4 and 

 -1.6 respectively, the per cent, of monatomic 

 hydrogen in equilibrium with diatomic hydro- 

 gen is indicated in the following table: 



1000° 1500° 2000° 2500° 3000° 

 Pressure K K K K K 



0.5 mm. .005 2.36 61.5 Complete 



1.0 mm. .004 1.69 40.0 98.8 



5.0 mm. .002 0.74 26.7 90.4 Dissociation 



As it is possible to obtain a temperature of 

 more than 2000°K in a tungsten furnace, it 

 seemed that a sufficient amount of monatomic 

 liydrogen could be obtained to maintain the are 

 at the lower potentials. 



The furnace consisted of a cylinder of thin 

 sheet tungsten, furnished by the General Elec- 

 tric Company, mounted on water-cooled leads 

 and heated by means of an electric current. 

 A tungsten filament ran axially through the 

 furnace and was also heated by a current. The 

 fall of potential in the furnace and that in the 

 filament were in the same direction and of 

 nearly the same amount. A potential was 

 applied between the furnace and the fila- 

 ment, and was tried in both directions. The 

 potential of the arc was corrected to the amount 



1 Nernst : Theoretical Chemistry. 

 2Eeiehe: Ann. d Physik, 58, p. 657, 1919, and 

 Shames: Phys. Zeits., 21, p. 41, 1920. 



between the middle of the two electrodes. Gas 

 pressures of 0.6, 0.8 mm, 1 mm^ and 2 mm 

 were used. 



When the furnace was not heated the arc 

 could not be maintained lielow the 16.4 volt 

 point. When the temperature of the furnace 

 was raised, a point was reached at which the arc 

 would strike at about 16.6 volts and break at 

 about 14 volts, indicating that the increased 

 dissociation in the are raised the percentage of 

 monatomic gas sufficiently so that the are could 

 be maintained to approximately the ionizing 

 potential of the atom. At still higher furnace 

 temperatures the arc could be made to strike 

 and break at about 13.5 volts and the results 

 when plotted showed also unmistakable evi- 

 dences of ionization at about 10.3 volts. 

 Curves were obtained showing three sharp 

 breaks in the neighborhood of 10.3, 13.2, and 

 16.2 volts. With the furnace at a very high 

 temperature the arc would strike at about 14 

 volts and break at 11 volts. It seems certain 

 that the arc struck at the ionizing potential 

 of the atom and was maintained as low as the 

 resonance potential of the atom. There was a 

 considerable amount of tungsten "sputtered" 

 on the walls of the tube, and from this it was 

 concluded that the temperature of the furnace 

 must have been 2000° to 2500° K. The results 

 seem to indicate that the percentages of disso- 

 ciated hydrogen calculated aljove are approxi- 

 mately correct. 



These results constitute, it is believed, the 

 first direct experimental proof of the correct- 

 ness of the values of the radiating and ionizing 

 potentials predicted by Bohr's theory for the 

 hydrogen atom, and of the interpretation of the 

 ionizing potential of the molecule as due to its 

 dissociation plus the ionization of one of the 

 atoms. 



A complete report of these experiments will 

 be published later. The apparatus will also 

 be used to study the ares in other gases and 

 for investigating the excitation of the spectra 

 of substances at high temperatures. 



0. s. dufpenback 

 Palmer Physical Labobatosy, 

 Princeton, New Jersey, 

 January 26, 1922 



