December ISj 1914] 



SCIENCE 



901 



man company, however, is planning to make 

 large quantities of ammonia by this process. 



The nitride (including the Serpek) processes 

 have not as yet proven to be successful fromi 

 the commercial point of view. It is quite 

 possible that these methods may be used in 

 connection with the manufacture of alumin- 

 ium and other metals with which these chem- 

 ical methods are intimately connected. 

 The Electrical Methods for Fixing Nitrogen 



Several electrical methods are used for oxi- 

 dizing the nitrogen of the air into nitric acid 

 and various salts of nitrogen. These methods 

 all produce chemical reactions between gaseous 

 oxygen and nitrogen in intense electric fields. 

 Potential differences of thousands of volts are 

 used and in the arc methods large currents 

 and .high temperatures accompany the use of 

 intense electric fields. In all these methods 

 the aim is to have the electrical discharge take 

 place in the gaseous oxygen and nitrogen and 

 to eliminate as much as possible the effect of 

 the metallic electrodes. Large arcs are there- 

 fore necessary when the electric current is 

 large. In the Birkeland-Eyde method the arc 

 is drawn out by a magnet; in the Schonherr 

 process by a helical current of gas and in the 

 Pauling process by horn electrodes and cur- 

 rents of gas. In the author's method a corona 

 current is used and this seems to give the most 

 perfect type of a purely gaseous discharge. 



The various electrical processes give about 

 the same order of efficiency when this is meas- 

 ured by the number of grams of nitric acid 

 produced per kilowatt hour of consumption of 

 electrical energy. About 60 to 80 gm. of 

 nitric acid are formed per hour per kilowatt 

 of electrical energy. 



The Complexity of Chemical Reactions 

 Although single atoms, ions and possibly 

 molecules have been isolated, the condition 

 under which the isolation takes place is en- 

 tirely unique, the particles traveling with a 

 very great velocity. In general chemical re- 

 actions will not take place under these con- 

 ditions in any way that they can be studied 

 individually. Our knowledge of chemical re- 



actions is therefore entirely statistical and 

 our laws apply to a very large number of re- 

 actions. There are numerous instances 

 where experimental evidence indicates that 

 the chemical reactions are frequently com- 

 plex. The speaker's work on the absorption 

 spectra of uranyl and uranous salts indicated 

 the possible existence of various intermedi- 

 ate compounds in chemical reactions in solu- 

 tions. 



In gases chemical reactions are undoubtedly 

 much less complex than they are in solutions, 

 although here the reactions may not be as 

 simple as they are sometimes represented. 

 The spectroscope is beginning to show indica- 

 tions that the light centers are more numer- 

 ous than the possible number of atom, ion 

 and molecule types. In the case of nitrogen 

 we have various types of line spectra and 

 quite recently Grotrian and Eunge^ have 

 made convincing claims that the so-called 

 cyanogen spectrum is due to nitrogen. 

 (These experimenters worked with large 

 Schonherr arcs about a meter in length.) 



Chemical Reaction Centers 

 Under conditions such as exist in the arc, 

 spark or whenever the temperature is high, 

 many kinds of " centers " may exist. These 

 " centers " may be the sources of light and 

 heat emission or absorption, the ions that 

 show defiections by electric and magnetic 

 fields, and the particles that take part in chem- 

 ical reactions. It must not necessarily be as- 

 sumed that the " centers " of the various phys- 

 ical phenomena are the same. They may be 

 widely different. 



Among the centers which may exist in arcs 

 and sparks and which have been shown to 

 exist in vacuum tubes are 



O2, 0, 6, 63, 6a, O, O, Oj, 



N, N, ii., N3, ""n, N, N,. 



Negative electrons also exist in comparatively 

 large numbers. 



The formation of nitric oxide in the electric 

 discharge may take place in a large number of 



■i-Fhys. Zeit., June 1, 1914. 



