1. 0·1532 gram gave 0·4146 gram CO2, and 0·0652 gram H2O.

II. 0·1109 gram gave 144 c.c. N at 206°, and 768 mm. bar. = 0.0166 gram N.

The substance dissolves in aniline with a reddish-violet colour, which becomes bluer on cooling. Its xylene-solution is also violet. It is practically insoluble in alcohol and glacial acetic acid, but dissolves in hot alcoholic potash with a bluish-violet colour. With concentrated sulphuric acid it gives a dull greenish-blue, becoming bright blue on dilution, and finally violet, and precipitating. It is transformed by the action of fuming sulphuric acid into a sulphonic acid which dyes silk and wool of a dull violet colour.

ACTION OF DIAZOTISED PARANITRANILINE UPON SECONDARY Mon

AMINES.

Action upon Diphenylamine.

The starting point of this series of compounds was the body already described by Roussin and Poirrier (Dingl. polyt. J., 1879, p. 423; Chem. Industrie, 1879, 2, p. 293), obtained by the action of diazoparanitrobenzene upon diphenylamine.

On mixing an aqueous solution containing one molecule of diazotised paranitraniline with an alcoholic solution of diphenylamine (one molecule), a deep reddish-violet colour is produced, and the mixture soon becomes semi-solid from the separation of a dense brown precipitate. The latter was collected, washed with water, and basified by ammonium carbonate. A dull red substance was obtained which after crystallisation from dilute alcohol, formed brown leaflets having a melting point of 151°.

is

Paranitrobenzene- azodiphenylamine, C.HN N.C,H1.NH.CH, NO2 readily soluble in alcohol with an orange colour, which changes to violet on the addition of hydrochloric acid, and the hydrochloride separates out in clusters of small needles having a beautiful violet reflection. The salt is extremely unstable, being instantly decomposed by water or by exposure to the air. The base dissolves in strong sulphuric acid with a violet colour.

NH2

Paraamidobenzene-azodiphenylamine, C.H.NN.C.H1.NH.CH.—

The reduction of the nitro-base was effected by adding ammonium sulphide to the cold alcoholic solution. In the course of a few minutes the colour of the solution changed from orange to yellow, and on diluting with water an orange crystalline substance slowly separated out. The amido-base thus obtained was mixed with a small quantity of some secondary product containing sulphur, from which it could not be entirely freed even by repeated solution in dilute HCl, filtration, and reprecipitation by ammonia. It was found, however, that this impurity was removed by the diazotising process, so that no further attempts at purification were made, and the following remarks apply to the crude substance.

The melting point of the crude base is 90-91° C.. Its salts dissolve readily in water with a magenta colour. The base itself is but slightly soluble in boiling water, but dissolves readily in alcohol, acetone, chloroform, and the benzene hydrocarbons with a yellow colour. adding hydrochloric acid to the alcoholic solution, a green colour is first produced, and this on the further addition of acid passes into red. The solution in concentrated sulphuric acid is violet passing into red on dilution. The salts act as dye-stuffs, imparting a bright yellowishorange shade to silk and wool.

B-Naphtholparaazobenzene-azodiphenylamine,

In order to diazotise the amido-base, C.H,<NHCH,.N<TH;

it was

found most advantageous to dissolve it in alcohol with the addition of hydrochloric acid, and then add the theoretical quantity of NaNO2 dissolved in water little by little to the well cooled solution. The orange-coloured solution was then gradually mixed with an aqueous solution of sodium ß-naphtholate. A reddish precipitate was obtained which was collected, washed and dried. Purification was effected by dissolving the substance in alcoholic soda, filtering, and precipitating by an acid. This operation was repeated several times, and finally a violet precipitate was obtained which, after being collected and washed with water till free from saline impurity, dried to a dull bronzy powder. After crystallisation from hot glacial acetic acid it was ob tained in the form of warty scales having a dull bronzy lustre and melting at 203-204° C.

I. 0.3301 gram gave 0·9128 gram CO2, and 0·1471 gram H2O. II. 0·1558 gram gave 21-2 c.c. N at 22° and 7671 mm. bar = 0.0242 gram N.

This new secondary azo-compound is characterised by the following properties :-It dissolves slightly in boiling alcohol and readily in benzene with a red colour. It is dissolved by glacial acetic acid, the solution being red when hot, and reddish-violet when cold. It dissolves in strong sulphuric acid with a greenish-blue colour, changing to reddish-blue on standing. Alcoholic potash or soda dissolves the substance with great facility, the solution being red. A most characteristic property of this compound is the fine blue coloration produced by adding hydrochloric acid to its dilute solution in alcoholic potash.

These researches have been conducted in the course of technological studies carried on in the laboratory of Messrs. Brooke, Simpson, and Spiller at the Atlas Works, Hackney Wick; and it is with great pleasure that I record my thanks to this firm. The investigations will be extended to the action of diazoparanitrobenzene upon tertiary monamines and phenols, and the corresponding series of reactions with the isomeric nitranilines and their homologues.

443

LXI.-On the Production of Hydroxylamine from Nitric Acid.

By EDWARD DIVERS, M.D., Principal of the Imperial Japanese College of Engineering.

DURING the last few months I have resumed an investigation of the conversion of free nitric acid into hydroxylamine, which I began in 1872, but the pursuit of which I had soon after to relinquish. So far as I can learn, this field of research has in the interim not been occupied, and I am therefore in a position, with the aid of recent work, to publish something of interest on this subject as yet unrecorded.

Lossen having discovered hydroxylamine în 1865 by reducing ethyl nitrate with tin and hydrochloric acid, Maumené followed by showing that for ethyl nitrate ammonium nitrate might be substituted (Compt. rend., 1870). Having confirmed for my own satisfaction the accuracy of Maumené's observations, I found that sodium nitrate could be equally well employed, and then, proceeding one step further, that nitric acid itself could be used with excellent results. My further work, then and since, forms the subject of this communication.*

Experiments with Tin.

Tin is I believe the only metal which has hitherto been known to produce hydroxylamine, and then only in conjunction with hydrochloric acid. This metal will, however, not only convert nitrates and nitric acid into hydroxylamine when used with hydrochloric acid, but will also form this base by acting alone upon pure dilute nitric acid. In attempting to verify this statement, it will be found very easy to fail. A 3 or 4 per cent. solution of nitric acid, free from other acids (or a false appearance of success will be almost certain), should be left on an abundance of granulated tin for about half an hour, after which

* In Watts's Dictionary, 1st Suppl., 1872, full mention is made that nitric acid yields hydroxylamine when treated with tin and hydrochloric acid, but no authority is given for the statement. Groves, in the account he gives of hydroxylamine in the last edition of Miller's Chemistry, 1878, makes no mention of the reduction of nitric acid itself. The same is true of Roscoe and Schorlemmer in their Treatise on Chemistry, 1878; Acworth and Armstrong in their elaborate paper "On the Reduction of Nitric Acid" (Part I), in this Journal for 1877, recognise the possibility of hydroxylamine being a source of the nitrous oxide produced by nitric acid acting on metals, but do not treat of its formation in this way as an ascertained fact.

So little indeed does this fact of the formation of hydroxylamine by the action of metals upon nitric acid appear to be known, that in all the discussions of the difficulties in estimating nitrates and nitrites as ammonia produced in acid solutions 2 K

VOI. XLIII.

hydroxylamine may generally be detected in the usual way.

Nitric acid, after standing a sufficient time upon tin, no longer yields hydroxylamine. Ammonia is one of the products in all cases, whether the nitric acid is dilute or the strongest that will act upon tin, and at any stage of the action.

Tin and Sulphuric Acid.-In examining the effect of sulphuric acid upon the reaction between nitric acid and tin, two comparative experiments were carried on together. In each experiment about 45 grams of granulated tin were placed in a flask filled with carbon dioxide gas, and a solution poured in, measuring 250 c.c., and containing in one case 5 grams of nitric acid (hydrogen nitrate) alone, in the other case the same quantity of nitric acid together with 20 c.c. of sulphuric acid, about 4 mols., that is, to 1 mol. of nitric acid. The flasks were closed and set aside for two days in very hot weather, there being scarcely any evolution of gas. Much grey and dense stannoso-stannic hydroxide was deposited, about the same quantity in each, and the solutions became yellow. But the solutions differed in composition. That without sulphuric acid contained some stannous salt dissolved and a very little stannic salt, but no hydroxylamine. The sulphuric acid solution contained much hydroxylamine, much stannous salt, and so much stannic salt as to give a copious precipitate with water. These tin solutions proved very unstable, but would no doubt have been much less so in winter, when there would also have been formed much less stannic salt; but these points are here unimportant, except perhaps to those who may repeat the experiments. These parallel experiments demonstrate the great effect of sulphuric acid upon the yield, if not upon the production, of hydroxylamine.

Another experiment was made, in which the sulphuric acid and nitric acid were diluted scarcely more than sufficient to allow of their free action upon the tin; for without some dilution there is hardly any action. In this experiment 5 grams (1 mol.) of nitric acid and 35 grams (4 mols.) of sulphuric acid, diluted to about 62 c.c. only of solution, were poured upon 35 grams of tin contained in a flask filled with carbon dioxide, and having its mouth loosely stopped. Action was at once apparent by the heating up of the mixture and the escape of nitrogen oxides, but it never grew violent, and the amount of gases evolved was not very great. Nitric oxide constituted the bulk of -even as late as that contained in a paper in the Chem. News, 46, 63, by Kinnearno reference is made to it.

* Namely, by treating the liquid with hydrogen sulphide and filtering off the tin sulphide, adding copper sulphate in the least possible excess and filtering off the copper sulphide, and lastly, adding potassium hydroxide in excess to precipitate cuprous oxide by means of the hydroxylamine.

To exclude air.