[ 225 ] 



XVI. Fractional Distillation applied to American Petroleum 

 and Russian Kerosene. By J. Alfred Wanklyn and 

 W. J. Cooper*. 



WE entertain a higher opinion of the possibilities of 

 Fractional Distillation than is admitted by the great 

 majority of chemists. According to our view separations are 

 possible where there is any difference of volatility : and we 

 hold that, unless chemical action of some description frus- 

 trates the separating process, there is no limit to the pos- 

 sibility of separation and purification by fractional distillation 

 in all cases where the components of the mixture differ in vola- 

 tility. By the word volatility we do not mean quite the same 

 thing as vapour-tension. The relative volatility of two liquids 

 (as one of us pointed out more than thirty years ago in a 

 paper in the Proceedings of the Royal Society) is expressed 

 not by the vapour-tension, but by vapour-tension multiplied 

 by vapour-density. There must be instances, but they are 

 very rare, where two liquids of different vapour- tensions have 

 identical volatility. Putting such cases on one side, we hold 

 that, unless chemical action bars the way, there is no limit to 

 purification by fractional distillation. 



Our method of working differs, vitally we are disposed to 

 think, from the ordinary manner in which chemists carry out 

 the process of fractionation. Our process is exhaustive : 

 hitherto the method in common use has not proved to be 

 exhaustive, but has missed out many terms of the series to 

 which it has been applied. 



In our paper published in May 1894, in the Philosophical 

 Magazine, we gave an account of our work on Russian Kero- 

 sene, showing a series of hydrocarbons rising in the scale, not 

 by increments of 14 but by increments of 7. 



We have applied our exhaustive fractionation to the 

 American Petroleum of commerce, and, so far as we have 

 investigated, a precisely similar state of things is found in the 

 American liquid as in the Russian. Restricting ourselves at 

 first to the more volatile portion of the oil, we find an homo- 

 logous series with the common increment of 7 instead of 14. 

 The admirable research of Cahours and Pelouze, which dates 

 back to the vears 1862 and 1863, established that these 

 American hydrocarbons belong to the Marsh-gas family. 

 We have now come across the same hydrocarbons, and we 

 have found other hydrocarbons which fractionation, as prac- 

 tised in the years 1862 and 1863, had missed. 



In our hands fractional distillation furnishes liquids which 



* Communicated by the Authors. 



