NOTES ON THE DEVITRIFICATION OF OLD GLASS 145 



Students at Dalhousie University had occasion to carry out a 

 vapour density determination with a Dumas bulb, and was 

 amazed to find that the bulb became heavily frosted at the 

 temperature of boiling water. It was decided to test Germann's 

 explanation to see if it applied to this case. 



The Dumas bulbs in question were believed to be from 

 twenty to thirty years old. They had not changed in appear- 

 ance except that some of them had small crystals on their inner 

 surface. Germann mentioned that old glass will show efflores- 

 cences of sodium carbonate, and it was found that a bulb with 

 crystals gave an alkaline reaction when washed out with water. 



One of the bulbs was fastened to a Toepler pump, surround- 

 ed by an electric furnace, and, having been exhausted, was heated 

 to about 200°C. Shortly after heating the pressure rose from 

 less than 0.01 mm to 0.2 mm. This was due to water 

 vapour which was pumped off. The temperature was 

 raised to about 450°C. and more gas was given off. The pres- 

 sure finally reached an apparent maximum when the gas was 

 pumped off and analysed. There was about 0.3 cc. and it 

 proved to be practically all carbon dioxide. 



A new bulb was kept in boiling sodium hydroxide solution 

 and then placed in an oven at 120°C. It was thought that 

 this treatment might remove the surface layer and prevent 

 devitrification, but frosting occurred. 



Following the suggestion given by Germann, another 

 bulb was washed with hydrofluric acid and then placed in an 

 oven at 120°C. No devitrification occurred. The inside of 

 the bulb was perfectly clear. According to Germann's theory 

 the devitrification requires previous exposure to water vapour. 

 As all the bulbs were sealed when made the inner surface was 

 not exposed to the atmosphere, and hence no devitrification 

 occurred on the inside. 



A fresh bulb was partially washed with hydrofluoric acid 

 and placed in the oven. It was found that only the parts of the 

 surface which were not washed with the acid were devitrified. 



