382 
Proceedings of Royal Society of Edinburgh. [sess. 
A New Method for Determining Phosphorus in Organic 
Phosphorus Compounds. By Prof. E. A. Letts and 
R. F. Blake, Esq., Queen's College , Belfast. 
Read July 21, 1889. 
In the phosphines and their derivatives, which we have inves- 
tigated from time to time, considerable uncertainty has always 
attended the determinations of phosphorus by the ordinary methods 
recommended for the purpose. In fact, we never felt any confidence 
in the result, for no matter bow carefully the determinations were 
made, duplicate analyses led to different numbers. 
The uncertainty depends partly upon the difficulty of oxidising 
the phosphorus in such compounds to phosphoric acid. For, as a 
rule, in any dry combustion process which may be employed, vola- 
tile oxidation products, containing phosphorus, are formed of great 
stability, which frequently pass over the red-hot oxidising mixture 
almost unchanged. Moreover, the glass of the tube is attacked by 
the oxidising mixture, and this undoubtedly leads to inaccuracies, 
probably of considerable magnitude. 
If any moist combustion process is resorted to, only a part of the 
phosphorus is converted into phosphoric acid. A more certain and 
trustworthy process than any of those which are in general use was 
therefore desirable, and in some of our investigations, where every- 
thing depended upon a correct estimation of the phosphorus, it was 
essential. It occurred to one of us that the difficulty in finding such 
a process ought not to be so great after all, for by burning a substance 
in the ordinary way with oxide of copper, the phosphorus ought to 
be completely oxidised, and should be found at the end of the opera- 
tion as phosphate of copper, in which it could be estimated without 
much difficulty by the molybdate method. In addition to the sim- 
plicity of such a method, it should also possess the great advantage of 
permitting the simultaneous determination of carbon and hydrogen. 
Our anticipations have been fully realised, and the new process, 
based on the above principle, if somewhat tedious, we believe to 
be accurate, of general application, and easily carried out. We shall 
describe it with the necessary detail, and then give some of our 
results obtained with it. 
