Attempt to Separate the Isotopes of Chlorine. 133 



or when I is large and the second term in both the numerator 

 and the denominator can be neglected, # 



E _ 2e 2klx ~ mx ~. 



Differentiating and equating to zero, we find that #=(1/3) 

 for maximum efficiency when I is large, which means that 

 if pure samples o£ the isotopes m x and m 3 were available, they 

 should be mixed in the proportion of two parts of m 1 to 

 one part of m 2 in order to make the most efficient filter for 

 the separation of m 2 ; provided that the atoms redistribute 

 themselves to form the molecules m^m^ m 1 m 2 , and m 2 m 2 in 

 the proportions shown above, and that the thickness of the 

 filter were such that the proportion of light which it absorbed 

 was large. The efficiency can, of course, be increased inde- 

 finitely by increasing the length of the filter, but a practical 

 limit is set by the slowness of the reaction if the filter 

 is made too long. It should be pointed out that the above 

 calculation depends on the assumption that there is no 

 appreciable overlapping of the absorption lines due to the 

 different kinds of molecule. Continuous absorption super- 

 posed on the line absorption, if it were effective in promoting 

 the reaction, would also reduce the efficiency. 



'Experimental^ 



Mixtures of hydrogen and chlorine in equivalent ■ pro- 

 portions were exposed to the action of light which had 

 been filtered through a layer of chlorine, and alter an 

 appropriate time mercury was added to the reaction mixture 

 to remove all the free chlorine before the mixture was 

 exposed to unfiltered light. The remaining hydrogen and 

 hydrogen chloride were bubbled through conductivity water, 

 and the hydrochloric acid was converted into sodium chloride, 

 the equivalent of which was then determined. In the 

 first series of experiments the mixture of hydrogen and 

 chlorine was made by the electrolysis of pure hydrochloric 

 acid containing 30 per cent, of acid in a voltameter with 

 arc carbons as electrodes passing through a rubber plug 

 which was protected from the action of the chlorine by 

 a layer of hard paraffin-wax. The mixed gases were bubbled 

 through a wash-bottle containing 66 per cent, sulphuric acid 

 in order to remove spray and hydrogen-chloride vapour. 

 The gases then passed into the reaction vessel of 1850 c.c. 

 capacity which had been previously exhausted by means of 

 a Graede rotary oil-pump to a pressure of about 0*3 mm. 

 When the reaction vessel had been filled to atmospheric 



Phil. Mag. S. 6. Vol. 43. No. 255. March 1922. 2 F 



