222 



Dr. A. W. Hofmanu on the Phosphorus Bases. [Mar. 21, 



I. Formation of Alcohol Phosphines by means of Phosphonium Iodide. 

 The starting-point of the new series of researches was a lecture experi- 

 ment. Wishing to exhibit to my class the decomposition of phosphoretted 

 hydrogen by the spark-current of an induction-coil, I was unable to procure, 

 by any of the methods hitherto described, phosphoretted hydrogen of 

 sufficient purity for this experiment. I was thus led to select a rather 

 unusual substance as a source for phosphoretted hydrogen, viz. the beau- 

 tiful compound of the latter with hydriodic acid, generally designated as 

 phosphonium iodide. This substance, formerly accessible only with diffi- 

 culty, may now be easily prepared in any quantity. If a slow stream of 

 water, or better of potash or soda, be allowed, by means of a dropping-tube, 

 to flow into a small vessel containing phosphonium iodide, a regular current 

 of perfectly pure phosphoretted hydrogen is evolved, which may at once be 

 introduced into a eudiometer provided with spark-wires, and be submitted 

 to experiment. "With the first spark that passes, a brown cloud of finely 

 divided phosphorus appears in the eudiometer, lining gradually the in- 

 side of the tube. After the lapse of five minutes, two volumes of phos- 

 phoretted hydrogen have become expanded into three volumes of pure 

 hydrogen gas. 



The facility with which phosphonium iodide is thus seen to split up 

 into its constituents, hydriodic acid and phosphoretted hydrogen, led me 

 to think that this body might be made available for the preparation of the 

 compounds I had so long endeavoured to obtain. Two different pro- 

 cesses suggested themselves, both aiming at a reproduction of 'he con- 

 ditions under which, as I have shown now more than twenty years ago*, 

 the alcohol derivatives of ammonia are readily obtained. For this purpose 

 it was necessary to disengage phosphoretted hydrogen in the presence of 

 an alcohol iodide under pressure. This could be easily accomplished by 

 submitting a mixture of an alcohol iodide and phosphonium iodide in 

 sealed vessels to the action of an agent (such as water or a metallic oxide* 

 zinc oxide for instance), slowly liberating the phosphoretted hydrogen from 

 the iodide. But this process appeared to be capable of a further simplifica* 

 tion. Instead of withdrawing the hydriodic acid in the phosphonium iodide 

 from the reaction bymeans of water or a metallic oxide,it seemed worth trying; 

 to utilize this acid in the production of the very alcohol iodide to be acted 

 upon by phosphoretted hydrogen ; and the question arose whether this 

 result might not be readily attained by decomposing under appropriate- 

 circumstances the phosphonium iodide by the alcohols themselves. 



These several anticipations have been fully confirmed by experiment. 

 Both processes yield alcoholic phosphines easily and copiously ; and, re- 

 markably enough, whilst the former (action of alcohol iodide upon phos- 

 phonium iodide) gives rise to the formation of exclusively the primary and 

 secondary phosphines I had so long endeavoured to produce, the secondary 

 process (action of the alcohols upon phosphonium iodide) furnishes only the 

 * Phil. Trans. 1850, p. 93 ; 1851, p. 357. 



