Chemiluminescence 449 



(1880) , which could be oxidized with light emission in solution, 

 were oils in alkaline solution. The luminous phenomena accom- 

 panying vaporized oils should be reinvestigated; as it is quite pos- 

 sible that chemiluminescence is involved. 



Another luminescence, somewhat akin to that of phosphorus, may 

 be observed with metals. After the preparation of pure alkali metals 

 by Sir Humphry Davy (1778-1829) around 1807, it was recorded by 

 W. Petrie (1850) and also by E. Linnemann (1858) and H. Baum- 

 hauer (1867), that fresh-cut surfaces of potassium and sodium are 

 luminous and remain so for an hour or so until they become covered 

 with a film of hydroxide. According to Phipson (1859) , the light is 

 especially bright at 70° C. Petrie realized that the light was of 

 chemical origin. At that time the light of most gases and metals in 

 flames Avas attributed to increase in temperature, although J. W. 

 Draper (1848) had investigated hot flames and entitled his paper, 

 " On the Production of Light by Chemical Action." 



Certain types of " cold " flames involving chemiluminescence can 

 be obtained with metals. In the vapor state sodium will luminesce, 

 emitting the yellow D lines in contact with halogens,^^ with hydro- 

 gen, or with mercuric halides at a relatively low temperature. A 

 number of other metals behave in a similar manner. For example, 

 mercury vapor coming in contact with chlorine, luminesces green; 

 with bromine, a yellow, and with iodine, an orange-red light ap- 

 pears. These and allied chemiluminescences emitted during a reac- 

 tion between simple gas molecules, have been much studied in the 

 twentieth century. 



The light which appears when " active nitrogen " or " active 

 hydrogen " (nitrogen and hydrogen atoms) are formed after an 

 electrical discharge and the two atoms recombine to the molecule 

 can be considered a chemiluminescence provided the recombination 

 explanation of mechanism is the correct one. On the other hand, 

 if the light comes from " metastable " molecules of nitrogen of long 

 life, formed during the electrical discharge, the after-glow should 

 not be designated chemiluminescence. 



This yellow after-glow of nitrogen is a striking phenomenon. It 

 has been found that pure nitrogen shows a very weak after-glow; a 

 small amount of oxygen or some other gas is necessary for an easily 

 visible light. Although " phosphorescence " of gases after passage of 

 electricity was noticed by E. Becquerel (1859) and received sporadic 



^^ The relation of quantum energy in the light emitted to the heat of reaction was 

 first considered by F. Haber and W. Zisch {Ztschr. fur Physik 1: 302-326, 1922) in 

 the case of the yellow luminescence observed when chlorine and sodium vapor are 

 mixed. 



