﻿Hydrogen through a Palladium Septum. 213 



Experiments 17 and 18. — At 280° the pressures with a 

 sample of London coal-gas were 364*4 millim., at a barometric 

 pressure of 760*4 millim. ; and 361*8 millim. at a barometric 

 pressure of 761*5 millim. The ratios are 0*4792 and 0*4751 

 respectively. Supposing the highest attainable ratio to be 

 0*9053, as deduced from the ten experiments with pure 

 hydrogen, the percentage of hydrogen is easily found by the 

 ratios (0*4792 x 100)/0*9053,and (0*4751 x 100)/0*9053. The 

 numbers are 52*94 and 52*49 per cent, respectively. It would 

 be possible thus to have a continuous register of the percentage 

 amount of hydrogen in coal-gas, provided the palladium did 

 not alter its state. As it is, I am afraid that the apparatus 

 will not retain its activity for a sufficiently long time. It 

 should also be remarked that it is necessary to wash the coal- 

 gas with a solution of potassium permanganate, in order to 

 free it from sulphur compounds. If this precaution is not 

 taken, the palladium becomes coated with a layer of sulphide, 

 and it ceases to be permeable to hydrogen. 



Experiments were next made in which the palladium tube 

 contained various gases. 



Experiment 19. — The tube contained carbon dioxide. On 

 exposing it to hydrogen at 280° the pressure rose with great 

 rapidity — much more rapidly than had previously been 

 observed. The rise ceased at 735 millim., the barometric 

 pressure being 764*0. The quotient is 0*9621, a higher 

 number than any yet obtained. 



Experiment 20. — Carbon monoxide was in the interior ; the 

 pressure rose rapidly on admission of hydrogen to 734*5 

 millim., the barometer reading 769*5. The quotient is 0*9545, 

 also a very high number. In both of these experiments the 

 fall on passing air over the exterior of the tube was much 

 slower than the rise with hydrogen. The end point was 

 difficult to note, for the last change of pressure was very 

 slow. It may be that the exterior surface of the palladium 

 being exposed to hydrogen alone, little difficulty was experi- 

 enced in molecules entering ; but that on reversing the 

 action, the contact of hydrogen with the interior walls was 

 interfered with by the indifferent gas present. The same 

 phenomenon had previously been noticed, but in a minor 

 degree. It is noteworthy that no trace of water was formed 

 in either case. Hydrogenized palladium is without action on 

 carbon monoxide or dioxide at 280°. 



Experiment 21. — The tube was filled with cyanogen. The 

 total pressure was 741 millim., the barometer standing at 

 764*5 millim. The quotient is 0*9693, a very high nunfber. 

 On substituting air, the cyanogen exerted finally a pressure of 



