FIXATION OF ATMOSPHERIC NITROGEN 405 



is reversible. It never proceeds beyond the formation of a very 

 small amount of nitric oxide, when a state of equilibrium is 

 produced corresponding to the temperature. It is essential 

 therefore to use as high a temperature as possible to promote 

 combination and to cool the gases as quickly as possible to avoid 

 the dissociation of the nitric oxide. The temperature in the 

 electric flame is said to exceed 3000 C., and the escaping gases 

 are brought down to between 800 and 1000 C. As the gases 

 in the atmosphere occur in the ratio oxygen 21 to nitrogen 79 by 

 volume, and the theoretical proportions required are equal 

 volumes, atmospheric air does not provide by itself the most 

 favourable material. It is, of course, known from experiments 

 on a small scale that the addition of oxygen to the air to be ex- 

 posed to the electric flame greatly enhances the quantity of 

 nitric oxide produced, but the cost of such addition seems to be, 

 for manufacturing purposes at present, prohibitive. Somfe 

 physicists seem to be of opinion that the action of the arc in 

 causing combination between nitrogen and oxygen is not wholly 

 thermal, and since it has been shown that union occurs under 

 the action of the silent electric discharge, which is not hot, it is 

 supposed that the result is brought about more directly by the 

 electric discharge at temperatures below that at which nitric 

 oxide becomes unstable. 



The flame of the electric arc used in the Birkeland and Eyde 

 furnaces is formed between two copper electrodes, which are 

 close together and established in a highly magnetic field produced 

 between the poles of a powerful electro-magnet. The electrodes 

 are made of thick copper tubing through which a stream of water 

 passes for cooling purposes. The chamber in which the flame 

 burns is circular, only a few centimetres in width and about 

 three metres in diameter. 



The production of the flame and its appearance as a disc has 

 oeen explained as follows by Birkeland in a lecture to the 

 Faraday Society in London, 1906 : 



" At the terminals of the closely adjacent electrodes a short 

 arc is formed, thus establishing an easily movable and ductile 

 current-conductor in a strong and extensive magnetic field i.e. 

 4000-5000 lines of force per square centimetre in the centre. The 

 arc thus formed then moves in the direction perpendicular to 

 the lines of force, at first with enormous velocity, which subse- 

 quently diminishes ; and the extremities of the arc retire from 



