ACTION OF PHOSPHIDE OF SODIUM ON HALOID ETHERS. 189 
Phosphide of Sodium. 
The success attending the use of this reagent for the preparation of phos- 
phines depends entirely on the observance of certain conditions in its manu- 
facture, which we have carefully determined by experiment. 
The most essential of these are, the manner in which the sodium and phos- 
phorus are allowed to combine, and the nature and quantity of the reagent em- 
ployed to prevent the temperature from rising too high during their combination, 
and to protect the resulting phosphide from subsequent oxidation by the air. 
If proper attention be paid to these precautions, phosphide of sodium may 
be prepared in a perfectly safe manner in large quantities and of uniform com- 
position. Whatever this latter may be, the bulk of the phosphide behaves as 
Na,P, as is evident from the nature of the products obtained from it. 
We prepare the phosphide by melting sodium under xylol and adding 
ordinary phosphorus in small pieces,—shaking the vessel in which the mixture is 
made after each addition of the latter substance in order to bring it thoroughly 
into contact with the sodium. As regards the proportions of sodium and phos- 
phorus, the theoretical quantities for the formation of the compound Na.P, are, 
in round numbers, 7 of sodium and 3 of phosphorus ; but our experiments show 
that with these proportions very bad results are obtained, as regards phosphine 
compounds. It appears to be necessary to employ a large excess of phosphorus 
—the proportions which we have found to be most suitable being 19 of that 
body to 20 of sodium—very nearly twice the theoretical quantity. 
Considering that large quantities of phosphorus are converted into the 
amorphous modification, which does not apparently combine with sodium at 
those temperatures at which we operate, a simple explanation is afforded of the 
necessity for employing an excess of phosphorus ; for, otherwise, unchanged 
sodium remains in the product, and, as we shall presently explain, this readily 
decomposes the chief phosphine compound produced when phosphide of sodium 
acts on chloride of benzyl. But, on the other hand, an excess of phosphorus 
over the proportions we have indicated also exercises a prejudicial influence 
on the quantity of phosphine compounds produced, and at present we can only 
explain this on the assumption that higher phosphides of sodium are formed, 
which are inactive. 
The xylol necessary to prevent the reaction of the phosphorus and sodium 
from being too violent must be free from water and high boiling impurities. A 
single distillation of ordinary xylol is sufficient to purify it for this purpose. 
The proportion of xylol employed is also an important condition. This, how- 
ever, we shall best consider when we describe our experiments on the action of 
the phosphide on chloride of benzyl. 
The following is a description of the exact method we employ for preparing a 
batch of phosphide of sodium containing 78 grms. of sodium and phosphorus :— 
VOL. XXX. PART I. 2G 
