B.— CHEMISTRY. 45 



Travancore for the manufacture of thoria and ceria required in the 

 incandescent mantle industry. During this manufacture each gram of 

 sand evolves 1 c.c. of helium at N.T.P., so that 100 tons of sand would 

 discharge into the atmosphere approximately 100,000 litres of the gas. 

 Our requirements of this raw material were entirely met through the kind 

 assistance of the late Mr. Edmund White, formerly managing director 

 of Messrs. Thorium, Limited. 



The gas was liberated by heating the monazite at 1000° in heat- 

 resisting steel pots in a stream of carbon dioxide, and the issuing gas was 

 passed over cupric oxide at 500° to oxidise hydrogen and carbon monoxide. 

 Carbon dioxide was then removed by a(;^ueous caustic soda and the residual 

 gas passed over metallic magnesium a t600° in order to remove nitrogen, 

 and over metallic calcium at 580° to eliminate the remaining impurities. 

 Several hundredweights of sand were thus treated and returned to Messrs. 

 Thorium, Limited, who found that they could still employ the heated 

 material in their process providing that it was mixed with a certain 

 proportion of raw sand. The purified gas containing 99.5 per cent, of 

 helium was compressed into storage cylinders. 



Corrosion Research. 



The researches on corrosion were originally started by a committee of 

 the Institute of Metals in 1916, and after eight years the more scientific 

 developments of these problems were undertaken by the Corrosion Research 

 Committee of the Department of Scientific and Industrial Research, this 

 work being pursued in the Metallurgical Department of the Royal School 

 of Mines until the workers concerned were transferred to Teddington at 

 Easter 1928. 



Corrosion of Immersed Metals. 



Research on the corrosion of immersed metals has been concentrated 

 on an attempt to put the theory of this phenomenon on a secure quanti- 

 tative foundation. For although earlier work in this country and in the 

 United States had furnished a qualitative explanation of corrosive action 

 in water or in salt solutions, yet this description of the process postulated 

 the influence of more than a dozen factors on the corrosive rates of immersed 

 metals. Accordingly, one aim of the present research is to acquire precise 

 information as to the interaction of these factors, and another objective is 

 to ascertain whether the lack of reproducibility in corrosion experiments 

 is inherent in the corrosion process itself or whether it is due to imperfect 

 regulation of all variables. Among these factors are purity of materials : 

 metal, water, salt and atmosphere, constancy of temperature and pressure, 

 and freedom from mechanical agitation. Zinc of a purity of 99.99 per cent., 

 distilled water with an electrical conductivity of 0.058 gemmhos at 20° and 

 purified oxygen were employed, and all experiments were carried out at 

 25° within a temperature range of ±0.02° over long periods of time, 

 sometimes for upwards of six months. 



Measurements of oxygen absorbed, corrected for any hydrogen 

 evolved, made at frequent intervals during the course of such experiments, 

 have enabled one to plot continuous corrosion time curves which are often 

 sufficiently smooth and regular to be investigated mathematically. 



