but I doubt not that all will be prepared in course of time. Eight of the nine oxygen derivatives, which are theoretically possible in the ethylic hydride group, are known, whereas only seventeen of the thirty-seven compounds which theory predicts for the propylic hydride group, have, up to the present time, been described.

CH3.CH.CH, is distinguished from CH3.CH, by the group CH2 which it contains, in addition to CH3.CH,. The introduction of this group impresses on the oxygen derivatives of the hydrocarbon C, H, some features that are wanting in those of the homologous body C2H6.

8

If an atom of hydrogen in the group CH2 is replaced by the water-residue Ĥ, pseudo-propylic alcohol CH3.CHH.CH, results. There can be no doubt that the abnormal properties of pseudoamylic alcohol are due to the same cause. Amylene possesses probably the formula CH3.2CH2.CH.CH2, and pseudo-amylic alcohol would be accordingly CH3.2 CH2.CHH.CH,, whereas amylic alcohol must be represented thus:-CH3.3 CH2.CH2Ĥ. Oxidising agents attack, as a rule, first the hydrogen in that group in which the water-residue is contained. Therefore, it is clear that to a pseudo-alcohol no normal aldehyde, nor an acid with the same number of carbon-atoms as the alcohol, can correspond, because the group CHH cannot be converted into COH or COH without the destruction of the molecule. In the case of pseudopropylic alcohol, acetone CH,.CO.CH,, and in that of pseudoamylic alcohol, the ketone CH,.2CH2.CO.CH, must be formed. Kolbe appears to have obtained this body. If the action of oxidising agents is continued, the ketones are by degrees broken up into bodies of simpler composition, until at last only carbonic acid and water remain. Wurtz obtained from pseudo-amylic alcohol, pseudo-butylic alcohol, acetone, acetic acid, and carbonic acid.*

Benzhydrolet and B-hexylic alcoholt are also pseudo-alcohols. Analogy justifies the conclusion that the hydrocarbons, homologous with propylene and higher up in the series, are represented by the general formula CH3.nCH2.CH.CH2, and furnish, like propylene and amylene, pseudo-alcohols, the water-residue attaching itself to the group CH.

* Ann. Ch. Pharm. Oct. 1864, 105. Ibid. 132.

Ann. Ch. Pharm. Jan., 1865, 11.

Ann. Ch. Pharm. Aug., 1865, 146.

290

XXVI.-Notes on Pyrophosphodiamic Acid.

By J. H. GLADSTONE, Ph.D., F.R.S.

SINCE the paper of the late Mr. Holmes and myself, in which the formula of Pyrophosphodiamic acid was established,* I have made several new observations on this body. I propose throwing them together under the three headings of Characteristic Test, Methods of Preparation, and Compounds.

Characteristic Test.

The ready formation of ferric pyrophosphamate from an acid solution of a pyrophosphodiamate affords a very good test for this substance. If it be an insoluble salt that is to be examined, it should be dissolved or decomposed by cold dilute sulphuric acid; and, if a solution, it should be rendered strongly acid, with sulphuric acid if possible, but other acids will answer the purpose though they are more likely to destroy the compound sought for. To this acid solution a few drops of a ferric salt should be added; sufficient to colour it distinctly red. On heating this mixture the presence of pyrophosphodiamic acid is shown by the appearance of a haziness in the warmer currents, which soon extends itself over the whole liquid, and the ferric pyrophosphamate separates in small white flocculent masses. If the acid be present in large quantity, the liquid will become gelatinous. As far as I know, no other ferric compound of a similar appearance can be produced by heating any very acid solution, but the fact of the precipitate being really the pyrophosphamate may be further proved by its solubility in ammonia. Occasionally the precipitate will not form till the mixture is fairly boiling. The reaction which takes place may be thus expressed :

P2N2H¿O+Fе2Cl2+2H2O = P2NH2Fе2O.H2O+NH1Cl+2HCl

6

But the question arises, Is there no other compound that will give the pyrophosphamate under similar circumstances? There are amides of pyrophosphoric acid which contain a larger amount of the elements of ammonia, and these do give the same reaction; but

*Chem. Soc. J. xvii, p. 225.

whether in doing so they pass through the intermediate stage of the diamic acid it is impossible to say. Thus the weak aqueous solution which is obtained when pyrophosphotriamic acid is treated with hot water gives this reaction; but there is little fear of confounding the triamic with the diamic acid, on account of the almost insolubility of the one, and the extreme solubility of the other. There exist, however, two substances, or two modifications of the same substance, intermediate in composition between the diamic and triamic acids, which yield the ferric pyrophosphamate with the greatest facility. It is necessary, therefore, to have some means of distinguishing these. Their salts seem in general less soluble than the pyrophosphodiamates; but there is a much more trustworthy means of recognition, and one that has the advantage of being applicable to quantitative separation. Pyrophosphodiamic acid is very soluble both in water and in alcohol, but the higher compounds, though water dissolves them freely, are thrown down from their aqueous solution on the addition of common spirits of wine. The alcoholic solution may be tested for the diamic acid by the iron test, but it should be diluted with about an equal volume of water before it is heated, so as to ensure a sufficiently high temperature to effect the transformation into the pyrophosphamate.

Methods of Preparation.

The reactions by which pyrophosphodiamic acid is produced are very numerous; indeed, there would seem to be a great tendency in the elements to arrange themselves in this particular form. Most of the processes, however, give at the same time a larger or smaller quantity of the higher compound which is insoluble in alcohol.

In the papers already referred to the following three methods were pointed out :

1st. By decomposing chlorophosphuret of nitrogen with an alcoholic solution of an alkali, or even by the slow action of

3rd. By acting on oxychloride of phosphorus with dry ammonia at a low temperature. Two molecules of ammonia are absorbed, as is shown by the increase of weight-about 22 per cent.; and I have reason to believe that the resulting white solid mass is a mixture of chloride of ammonium and an amidated oxychloride. The following will be the expression of this first stage of the reaction:

which on the addition of water splits up thus:-
:-

2PNH2C2O + 3H2O = P2N2H605 + 4HCI

The last two methods of producing this acid may be modified in a way which saves time and trouble, namely, by using the strongest aqueous ammonia, sp. gr. 0·880, instead of the dry gas. Thus phosphorus may be burnt in a large vessel containing some ammonia at the bottom. In this case there is a white substance produced which is insoluble in water or acids, and which renders the solution milky, and is very difficult to separate from the liquid containing the pyrophosphodiamate of ammonium. Again, oxychloride of phosphorus may be dropped continuously into aqueous ammonia, when pyrophosphodiamic acid is produced, perhaps without the intervention of the amidated oxychloride

2PC130 + 8NH3 + 3H2O = P2N2H ̧О5 + 6NHC1

Akin to this last method is the following:

4th. By throwing pieces of pentachloride of phosphorus into the strongest aqueous solution of ammonia. The reaction is very violent, and it is desirable to cool the vessel in which it is performed. An insoluble white flocculent substance appears in small quantity, which gives reactions characteristic of pyrophosphotriamate of ammonium, while the solution made decidedly acid, and heated with a ferric salt, gives the ferric pyrophosphamate. This is due to a mixture of pyrophosphodiamic acid and the higher compound. It is hardly credible that no phosphoric acid is produced, but the magnesium test fails to detect any in the tribasic condition.

of

This experiment was performed quantitatively. 0·65 grm. pentachloride of phosphorus gave 0035 grm. of pyrophosphotriamate of ammonium, and 0.305 grm. of ferric pyrophosphamate.

A considerable part of this small loss must be due to the escape of dense fumes on bringing the two bodies into contact. The main reaction seems to be

2PC15 + 12NH3 + 5H2O

=

P,N,H,O, + 10NH,CH

5

If the solution of ammonia be not very strong, phosphate and chloride of ammonium are the only products.

5th. By saturating oxychloride of phosphorus with dry ammonia-gas at 100° C., or thereabouts. It then takes up four molecules, and on the addition of water there is formed a mixture. of the insoluble triamic with the soluble diamic acid, and generally a large quantity of the intermediate compound. The production of the pyrophosphodiamic acid probably takes place in these two stages :

and

PCl,O + 4NH

=

3

2NH CI+PN,H,CIO

2PN2H CIO + 3H2O = P2N2HO + 2NH CI

6th. By boiling for a long time a solution of pyrophosphotriamic acid in water, especially if some hydrochloric acid be present.

This is an intermediate step in the reduction of the amide to phosphoric acid.

7th. By treating the acid intermediate in composition between the triamic and the diamic acid with a large amount of some other acid, even acetic; or by decomposing its salts with a strong acid.

8th. By dissolving pyrophosphotriamic acid or a pyrophosphotriamate in sulphuric acid, either quite strong or mixed with a very small quantity of water. Heat is required to effect the solution, and it must not be continued long, lest the action proceed too far. The acid solution diluted with water and neutralized, gives no precipitate, showing that the pyrophosphotriamic acid is really