2IO STUDIES IN LUMINESCENCE. 



and the total intensity is 



where 



I = I + j = a j, ( l __| I \ 



i i . a , a 



fli= , "2= -7= 61= , 02 = ; 



It thus appears that the form of the decay curve is the same as though 

 the substance possessed two bands, coincident as regards wave-length, but 

 differing in rate of decay. 



It is not at all unlikely that a condition approaching that assumed in 

 this illustrative case actually exists in most phosphorescent substances. 

 Upon considering the method used in preparing the phosphorescent sul- 

 phides, for example, it seems probable that the distribution of the active 

 substance will be far from uniform. When the mixture is first prepared, 

 and before calcination, the active material is unquestionably in the form 

 of small discrete masses distributed irregularly through the mixture. Upon 

 heating to redness diffusion will occur to a greater or less extent, depending 

 upon the temperature and the duration of heating. But even at high temper- 

 atures this will be a slow process, and considerable variations in concen- 

 tration are likely to remain even after prolonged heating. It is to be 

 expected, therefore, that the phosphorescent sulphides will contain numer- 

 ous nuclei of high concentration surrounded in each case by a region where 

 the concentration is relatively small. 1 



After calcination it is often noticed also that the phosphorescence is very 

 far from being uniform throughout the mass. Owing probably to accidental 

 differences in concentration, or to differences in the heat treatment, the 

 phosphorescence often differs greatly in intensity and even in color in 

 different parts of the same mass. A phosphorescent powder made from 

 such a mass, while presenting the appearance of homogeneity to the un- 

 aided eye, would differ greatly from point to point in the concentration of 

 the active material. 2 Even if the mixture were so perfect that no irregu- 

 larities could be detected with the microscope, a wide deviation from line- 

 arity in the decay curve is to be expected. It is a significant fact that 

 with one exception all decay curves thus far recorded have been determined 

 with powders prepared in practically the same way that the phosphorescent 

 sulphides are prepared. In fact, most of the substances tested were sul- 

 phides. The exception noted above was natural willemite, 3 in which case 

 the variation in brightness over the surface tested was plainly visible. 



'Since a more complete diffusion of the active material will result from prolonged heating it is to be 

 expected, other conditions being the same, that the duration of phosphorescence will be prolonged by in- 

 creasing the time of heating. This agrees with the facts observed in the preparation of the phosphorescent 

 sulphides. 



2 While we are chiefly concerned at present with the influence of lack of homogeneity upon the decay curve, 

 it can scarcely be doubted that the effect upon the phosphorescence spectrum is fully as important. For 

 some reason different parts of the mass are differently affected by the process of calcination. This may be 

 due to differences in concentration throughout the mass, or to the fact that the effect of the surrounding 

 gas varies from point to point. When such a mass is powdered and mixed the effect is the same as though we 

 were to make an intimate mixture of several entirely different phosphorescent substances. Each constituent 

 has its own band or group of bands, characterized by definite wave-lengths and periods of decay, which 

 differ according to the conditions of preparation. It is not surprising that the phosphorescence spectrum 

 of such a material is complex; and we can scarcely expect simple laws to apply to any of the phenomena 

 exhibited by such a mixture. 



The decay of phosphorescence in a specimen of willemite possessing a long-time phosphorescence has 

 been studied by Nichols and Merritt, Physical Review, XXIII, p. 52, Fig. 53. Willemite whose phosphor- 

 escence dies out with great rapidity has been studied by Waggoner, Physical Review, XXVH, p. 209. See 

 also Chapters, IV and VII. 



