Optical Properties of Films of Sodium and Potassium. 91) 



(present paper attention will be drawn to a number of 

 interesting cases. 



As L showed in a paper published in this Journal lor 

 October 1916, there is a certain critical temperature for each 

 metal, and also lor certain other elements, such as iodine and 

 selenium, above which it is impossible to condense them 

 upon the wall of a glass bulb by distillation in vacuo in the 

 form of a homogeneous film. These critical temperatures 

 are surprisingly low. For example, the temperature for 

 cadmium is —90° C. If we cool the wall of the bulb to this 

 temperature the cadmium vapour condenses as a homo- 

 geneous film of a dark blue colour. If the wall is above 

 this temperature we obtain a granular deposit devoid of 

 colour. If, however, the film is started at the low tempera- 

 ture, we can go on building it up as a homogeneous non- 

 granular film even at room temperature. The explanation 

 of the phenomenon is given in the paper referred to. 



This discovery has thrown some light upon the case of the 

 films of the alkali metals, the optical properties of which 

 were studied in 1902. 



It is quite certain, I think, that all data previously 

 published regarding the transmission of light by sodium and 

 potassium films refer only to granular films deposited above 

 the critical temperature, the colour of the transmitted light 

 resulting from the granular structure. I have produced 

 homogeneous films by condensing the metal upon the inner 

 wall of a quartz bulb cooled to the temperature of liquid air. 

 Films of sodium and potassium, prepared in this way, vary 

 from blue-grey to black (in transmitted light) and reflect 

 with the brilliancy of silver. The films are not stable at 

 room temperature, but break up into granules in the course 

 of a few seconds, minutes, or hours, depending on their 

 thickness. They are astonishingly transparent to ultra- 

 violet light, a potassium film, through which a tungsten 

 filament is absolutely invisible, transmitting everything 

 below wave-length 2900 even down to the last cadmium line 

 at wave-length 2147. 



A spherical quartz bulb about 3 cm. in diameter was 

 used in the investigation. It was highly exhausted and 

 contained a minute quantity of sodium which could be 

 driven about over the wall as a coloured deposit by means of 

 a very minute flume burning at the tip of a glass capillary. 

 The film prepared in this way is granular, the colour varying 

 from light blue through green and greenish yellow to a dull 

 red. Films of this nature will be discussed later on. 1{\ 

 .now, the wall on one side of the bulb is cleaned by the 



H2 



