J 



40 SYNTHESIS BY SUNLIGHT IN RELATIONSHIP 



most effective. The use of quartz flasks as containers for the 

 colloid gives, with the same intensity of insolation, definite results 

 in a much shorter period, and it has been found that a " Uviol " 

 mercury arc, in a " Uviol " glass protecting vessel, immersed in an 

 outer wider cylinder of glass so as to produce a thin layer of the 

 colloid, through which the carbon dioxide is passed, gives quite 

 clearly positive reactions for formaldehyde by all the colour tests, 

 with an exposure to light of only four or five hours. 



The results are obtained either with colloidal hydrated ferric 

 oxide, or colloidal oxide of uranium, in exceedingly dilute solution. 

 For example, in the case of uranium, 0-028 per cent, of the oxide, 

 and, in the case of the iron, 0-113 per cent, of ferric oxide were the 

 concentrations used. 



Controls carried out in the absence of light give no formaldehyde. 

 The same point is shown by our earlier unsuccessful experiments 

 in which the illumination was inadequate. 



Formaldehyde in small amounts has already been synthesised 

 from inorganic sources in several reactions in which hydrogen in 

 the nascent condition, or absorbed in palladium, has been present. 

 But there is here chemical energy as such, presented by the hydrogen, 

 and so far as we are aware there is no case known of utilisation of 

 light energy by an inorganic catalyst which does not itself become 

 altered in the reaction. 



Thus Bach 1 obtained formaldehyde from hydrogen palladium 

 and carbon dioxide, Fenton 2 obtained it by the action of carbon 

 dioxide and water on metallic magnesium, and other observers have 

 obtained it by the interaction of dilute sodium amalgam and moist 

 carbon dioxide. 



The only experiment approaching more closely to our own as 

 recorded below is one devised by Bach, 3 and later repeated with 

 modifications by Euler 4 and by Usher and Priestley. 5 



Bach passed a current of carbon dioxide through a solution of 

 1-5 per cent, crystalloidal uranium acetate, and in the presence of 

 light obtained a precipitate of mixed oxides of uranium, which did 

 not occur when light was excluded during the passage of the carbon 



1 Comptes Eendus, 1898, vol. cxxvi.. p. 79. 



2 Journ. Chem. Soc. Trans., 1907, vol. xci., p. 687. 



3 Comptes Eendus, 1893, vol. cxvi., p. 1145. 



4 Berichte d. deut. chem. Gesellsch., 1904, Jahrang 37 5 vol. ii., p. 3411. 



5 Evj. So-. P/-03., 1903, B, vol. Ixxvii., p. 369; and 1903, B, vol. Ixxviii., 

 p. 322. 



