1 866.] On the Absorption and Dialytic Separation of Gases. 223 



VI. On the Absorption and Dialytic Separation of Gases by Colloid 

 Septa." By Thomas Graham, F.R.S., Master of the Mint. 

 Received June 20, 1866. 



(Abstract.) 



It appears that a thin film of caoutchouc, such as is furnished by var- 

 nished silk or the transparent little balloons of india-rubber, has no poro- 

 sity, and is really impervious to air as gas. But the same film is capable of 

 liquefying the individual gases of which air is composed, while oxygen and 

 nitrogen, in the liquid form, are capable of penetrating the substance of the 

 membrane (as ether or naphtha do), and may again evaporate into a 

 vacuum and appear as gases. This penetrating power of air becomes 

 more interesting from the fact that the gases are unequally absorbed and 

 condensed by rubber, oxygen 2J times more abundantly than nitrogen, and 

 that they penetrate the rubber in the same proportion. Hence the rubber- 

 film may be used as a dialytic sieve for atmospheric air, and allows very 

 constantly 41-6 per cent, of oxygen to pass through, instead of the 21 per 

 cent, usually present in air. The septum keeps back, in fact, one-half of 

 the nitrogen, and allows the other half to pass through with all the oxygen. 

 This dialysed air rekindles wood burning without flame, and is, in fact, 

 exactly intermediate between air and pure oxygen gas in relation to com- 

 bustion. 



One side of the rubber-film must be freely exposed to the atmosphere, 

 and the other side be under the influence of a vacuum at the same time. 

 The vacuum may be established within a bag of varnished silk, or in a 

 little balloon, the sides being prevented from collapsing, by interposing 

 a thickness of felted carpeting between the sides of the varnished cloth, 

 and by filling the balloon with sifted sawdust. For commanding a 

 vacuum in such experiments, the air-exhauster of Dr. Hermann Sprengel* 

 is admirably adapted. It possesses the advantage that the gas drawn from 

 the vacuum can also be delivered by the instrument into a gas-receiver 

 placed over water or mercury. The " fall-tube" has merely to be bent at 

 the lower end. 



The surprising penetration of platinum and iron tubes by hydrogen gas, 

 discovered by MM. H. Sainte-Claire Deville and Troost, appears to be con- 

 nected with a power resident in the same and certain other metals, to 

 liquefy and absorb hydrogen, possibly in its character as a metallic vapour. 

 Platinum, in the form of wire or plate, at a low red heat may take up and 

 hold 3-8 volumes of hydrogen, measured cold ; but it is by palladium that 

 the property in question appears to be possessed in the highest degree. 

 Palladium foil from the hammered metal, condensed so much as 643 times 

 its volume-of hydrogen, at a temperature under 100° C. The same metal 

 had not the slightest absorbent power for either oxygen or nitrogen. The 

 capacity of fused palladium (as also of fused platinum) is considerably 

 * Chemical Society's Journal, ser. 2, vol. iii. p. 9 (1865). 



VOL. XV. U 



