August 30, 1906] 



NA TURE 



445 



amount." In order to give this increase, about 1-5 

 cwt. of nitrate of soda would be required to bo applied 

 annually to each acre, that is to say, 12,000,000 tons 

 would be needed. ."Ks at present situated the world 

 is not in a position to supply this vast amount of 

 nitrogenous product. Since Crookes sounded this note 

 i)f warning- many attempts have been made to o.\idise 

 atmospheric nitrogen on a commercial scale, but until 

 within the last fifteen months no process based upon 

 iloctrical oxidation has been an actual commerci.il 

 success. It yet remains to see whether the process of 

 I)rs. Caro and Frank, which depends upon the forma- 

 tion of calcium cyanamide, will be able to compete in 

 Ihe first place with natural nitrates and ammoniacal 

 [)roducts, and, secondly, with the electric process of 

 Hirkeland and E\'de, which, as we will sliortly show, 

 appears to have solved the problem of the fi.\ation of 

 atmospheric nitrogen. Drs. Caro and Frank found 

 that by passing nitrogen over heated barium carbide 

 bariinn cyanide was produced thus, 



BaC, + N„ = Ba(CN)„ 



but that when calcium carbide was employed calcium 

 cyanamide and not calcium cyanide is formed, 



CaC, + N, = CaCN,-l-(:. 

 When the cyanamide is heated with water under 

 pressure it is decomposed with formation of ammonia 

 and calcium carbonate, 



CaCN, + 3H,0 = CaC03 + 2NH3. 



This process is stated to take place slowly when the 

 cyanamide is distributed on the soil. Although the 

 manufacturers state that cyanamide is stable and does 

 not deteriorate on keeping, some at least of the users 

 say it is unstable and deteriorates considerably as a 

 fertiliser when kept. 



.•\n electrical process — that of Bradley and Lovejoy 

 — which was almost a success, was worked for about 

 eighteen months at Niagara. They employed a con- 

 tinuous current with a potential of 10,000 volts. .As 

 it is very diflicult to keep steady discharges at this 

 high voltage, a slowly rotating framework with pro- 

 jecting electrodes was employed. As it rotated, the 

 electrodes, which were of platinum, approached other 

 projecting electrodes ; discharges were thus provoked, 

 but immediately interrupted. In an apparatus of only 

 5 kilowatts as many as 414,000 arcs were produced per 

 minute. The working of such an apparatus on a 

 technical scale was, as might be supposed, of great 

 difficulty, and although considerable quantities of nitric 

 acid were produced per kilowatt year, it did not prove 

 commercially successful. 



In May, 1905, a factory was started at Notodden, in 

 Norway, for the manufacture of calcium nitrate from 

 air and limestone by means of the electric arc flames. 

 -A photograph of the factory as it is at present is 

 shown in Fig. i. In the Birkeland-Eyde process, 

 which is w-orked at Notodden, a high-tension arc flame 

 is produced between two pointed copper electrodes. 

 The electrodes are attached to a high-tension alter- 

 nator, and are placed equatorially between the poles of 

 a powerful electromagnet, so that the terminals of the 

 electrodes are in the middle of the magnetic field. 

 .An electric disc flame is thus produced which is shown 

 diagrammatically in Fig. 2, and a photograph of the 

 actual flame in Fig. 3. The photographed flame, 

 which represented about 250 h.p., was produced 

 between water-cooled electrodes made of copper 

 tubing. 



The working potential employed is 5000 volts, the 



current is an alternating one of 50 periods per second, 



and the distance of the terminals apart is about 8 mm. 



.As already mentioned, the electrodes are copper tubes 



NO. 1922, VOL. 74] 



which are water-cooled, and are in the form of a 

 narrow, elongated U ; even with flames of 750 kilo- 

 watts at 5000 volts the same form of electrode can be 

 employed. By cooling the electrodes, about 7-5 per 



I 



-View of nitric acid factory at Notodden. 



cent, of the electric energy employed between the 

 electrodes is removed as heat by the water. 



Prof. Birkeland explains the formation of the disc- 

 flames in the following way : — " At the terminals of the 

 closely adjacent electrodes, a short arc is formed, esta- 

 blishing an easily movable and ductile current con- 

 ductor in a strong and extensive magnetic field, i.e. 

 from 4000 to 5000 lines of force per sq. cm. in the 

 centre. The arc then moves in a direction perpendi- 

 cular to the lines of force, at first with an enormous 

 velocity which subsequently diminishes ; and the ex- 

 tremities of the arc retire from the terminals of the 

 electrodes. As the length of the arc increases, its 

 electrical resistance also increases, so that the tension 

 is increased until it becomes sufficient to create a new- 

 arc at the points of the electrodes. The resistance of 

 this short arc is very small, the tension of the elec- 

 trodes therefore sinks suddenly, with the consequence 

 that the outer long arc is extinguished. ... In an 



T-n-r-r-r-r-TT-'-.'— -y-i '' -.v- 



Fig. 2.— Diagram of electric arc flame. 



alternating current all the arcs with a positive direc- 

 tion' of current run one way, while all with a negative 

 direction run the opposite way (see Fig. 2), presuppos- 

 ing the magnetising being effected by direct currents. 

 In this manner a complete luminous circular disc is 

 presented to the eye." 



