412 ANNUAL REPORT OF THE Off. Doc. 



the machinery was too costly to keep up to make it a paying invest- 

 ment, and in 1904 it failed and the attempt was abandoned. Pirot 

 and Gay in France made attempts to bind the atmospheric nitro- 

 gen on a large scale in 1903 but tailed in reaching practical results. 



In Germany about the same time, 1903, the firm of Siemens and 

 Halske in Berlin began operations, adopting Pi'ank's method which 

 consists in allowing pure nitrogen to act on alkaJi earth carbides. 

 Calcium carbide, for instance, when heated moderately absorbs nit- 

 rogen forming a compound known as Calcium Cyanamid, also called 

 "Lime Nitrogen" in commerce. This salt decomposes quickly in the 

 soil to ammonia, etc., and has proven to be a good fertilizer, in some 

 cases equal to Chili saltpeter, but whether by this method it shall 

 be possible to produce the nitrogen compounds at low enough cost 

 to be considered a practical success remains to be seen, or whether 

 the cyanide salt can be used under all conditions, or whether it is 

 harmless to plants at various stages of development. 



Frank and Caro began experiments 1895 for preparation of car- 

 bides with a view to the manufacture of cyanides. First Barium, 

 later Calcium, was used as the base, but instead of cyanide being 

 produced, the product was the barium of calcium cyanamid (written 

 BaCNo and CaCN,) which yielded cyanides on fusion with alkali 

 salts. Heated with water under high pressure calcium carbonate 

 and ammonia were formed according to the following chemical 

 equation: CaCN3+3H,0=CaCO+2 NHg. The better grade 

 amids contained 14 to 32 per cent, nitrogen. By dissolving the cyan- 

 amide in water and crystalizing in the cold a dicyanamide (CN^H^)^ 

 with 66 per cent, nitrogen is formed. This Frank's method has been 

 operated cheapest in Italy where cheap water power is available. 



The latest and undoubtedly the most promising, and practically 

 successful method in operation for oxidizing the atmospheric nitro- 

 gen, is the Birkeland and Eyde method now in operation at Notod- 

 den, Norway. The oxidation of the nitrogen is done in an arc pro- 

 duced by an alternating electric current of a very high voltage — 

 5,000 volts — and a frequency of 50 periods per second. It is the pe- 

 culiar structure or shape of the arc flame which gives it a great pow- 

 er for oxidizing nitrogen, and which phenomenon was observed by 

 Birkeland and Eyde. It is the efficiency due to this peculiar shaped 

 flame which makes it possible to utilize such arcs for the oxidation of 

 the free nitrogen at low enough cost to make it pay commercially. 

 The arc is namely made to flatten out into a thin disc-like flame by 

 passing it through the field of a powerful direct current magnet, and 

 it is the use of this arc flame for chemically uniting nitrogen and 

 oxygen that has been patented and is now successfully used at the 

 place mentioned. The electrodes are thin walled copper tubes about 

 1.5 cm. diameter, and through which flows a current of water. The 

 points of these electrodes come within 6 to 8 m.m. of each other, and 

 the ends of the magnet come perpendicular to the electrodes, and a 

 few inches apart. It is because of the action of the magnets, as well 

 as the water flowing through the electrodes, that the latter are not 

 melted in the high heat formed, but can carry the current for sev- 

 eral days and nights before they need to be renewed. The magnet 

 prevents short circuiting at the electrodes. 



The reason why an arc of this disk-like shape is better and more 

 efficient than the arcs used by Lovejoy and Bradley, and others, is, 

 that it presents more heating surface with the same power than 



