84 CONTRIBUTIONS TO CHEMISTRY AND MINERALOGY. [bull. 167. 
acid. He afterwards obtained this in larger quantity by treating phos- 
phorus oxychloride with dry ammonia, and assigned to it the formula 
PO (OH) NH 2 
PO (OH) NH 2 ' 
and the name pyrophosphodiamic acid. The formation of such an 
acid can only be due to a deep decomposition of the molecule. 
Although unacted on by water alone or by anhydrous ether alone, I 
have found that a smooth decomposition is easily effected by dissolving 
the chlorophosphuret in alcohol -free ether, aud shaking this solution a 
long time with water, whereby intimate contact is effected. The final 
products are hydrochloric acid and an acid in which the chlorine is 
wholly replaced by oxygen and hydrogen, without further change. I 
have named this substance, P 3 N 3 6 H G , trimetaphosphimic acid. 1 If the 
water contain bases or acetates in solution, the salts are directly 
obtained. As the acid will be described in a separate section, I mention 
here only that it is extremely soluble and totally devoid of crystallizing 
power, but forms some salts of characteristic crystalline form, and bas 
a strong tendency to form double salts, one of these, P 3 N 3 O b H 3 NaBa, 
giving especially fine crystals. 
As intermediate products, chlorhydrines have been observed. Theo- 
retically, five of these are possible, one of which I have isolated, namely : 
Triphosphonitrilic tetrachlorhydrine, P 3 N 3 0l4O 2 H 2 . — One part triphos- 
phonitrilic chloride is dissolved in 10 parts alcohol-free ether, and the 
solution agitated with about one-third its volume of water for six to 
eight hours, best by means of a small turbine. The water then con- 
tains hydrochloric acid and trimetaphosphimic acid, and the ether 
contains the chlorhydrines and unchanged chloride. The ether is 
dried over calcium chloride, and the greater part distilled off in the 
water bath, the latter portions being removed at ordinary tempera- 
ture by a current of dry air. The residue consists of unchanged 
chloride, the chlorhydrine in question, and small quantities of others. 
The greater part of the chloride is removed by a little benzene, and 
the residual chlorhydrine well washed with carbon disulphide. The 
yield depends somewhat on the time the water has acted, but as there 
is a continuous conversion of chloride into chlorhydrine, and of the 
latter into trimetaphosphimic acid, the amount present at any one 
time is not great. Under the above conditions it was about 10 per 
cent of the theoretical, fully one-half the chloride being recovered. 
1 According to Mente (Ann. Chem. (Liebig), Vol. CCXLVIII, pp. 241, 244), Gladstone's pyrophospho- 
PO.OH 
diamic acid is really NH^ /NH, i. e., dimetaphospkimic acid, which has the same empirical composi- 
PO.OH 
tion as trimetaphosphimic acid. Mente did not obtain his acid from triphosphonitrilic chloride, but 
from phosphorus oxychloride, and it is not clear that it is really identical with Gladstone's acid from 
chlorophosphuret of nitrogen. The independent existence and stability of the tetra-acid affords a 
presumption that other members of the series can exist also. 
