CHEMICAL NEWS,}

Nov. 17,

235

"about 7, 20, 33, 50, 70, and 80 per cent SO3." Hence I take it that 50 cc. of fuming sulphuric acid (20 per cent

C. Those made by mixing phenol and concentrated sulphuric acid and adding thereto a little concentrated hydrochloric acid (Johnson, CHEMICAL NEWS, 1890, lxi., 15, &c.).

D. Those made by heating a mixture of phenol and concentrated sulphuric acid (Gill, &c.).

This indicates clearly enough that, according to the authors, I advocated simply mixing the reagents, whereas, as a matter of fact, the main object of my communication on that occasion was to draw attention to the primary importance, when making this reagent, not only of heating the mixture of phenol and sulphuric acid, but of heating it for a sufficient length of time. I had found some time previous to 1890 that even four hours' digestion was insuffi cient, and that to get a really satisfactory reagent—that is, one that could be depended upon to give a pure yellow colour with nitrates-about eight hours' digestion was necessary. I have met with several chemists who have been troubled with a phenolsulphonic acid solution that gave a bright green colour with nitrates, and, I may add, I once (previous to 1890) made one myself. The paper under discussion gives abundant evidence of the fact that this is quite a common experience, and the principal reason why treatment of the nitrate residue in the cold has been advocated appears to be that the reagent used would not stand moderate heating with a nitrate without developing a more or less pronounced green tinge. The reagent produced when my directions for making it are carefully followed will be found to give a more or less deep red colour, according to the amount of nitrate present, when the beaker is allowed to stand on the top of the wateroven for fifteen minutes; it never develops a green colour. Before fixing on the period of eight hours, I used at one time to withdraw, by means of a long graduated 5 cc. pipette, 1 cc. of the hot liquid, dilute it with 1 cc. of water, and add cc. HCl, mix, and add 1 cc. of the reagent so obtained to a dry residue from 1 cc. of the strong standard KNO3 solution in a small beaker on the water-oven. After fifteen minutes' standing this was then diluted, ammonia (or potash) added in excess, and the colour of the solution, diluted to 100 cc. in a Nessler glass, observed. If a green tinge was produced, the digestion was continued for another hour, when another similar test was made. In this way I found that eight hours' digestion was always more than sufficient. Moreover, by this prolonged digestion a newly made solution will give the same results as one that has been made for several months. I should like to add that I cc. of my reagent suffices to determine nitric nitrogen up to about 15 parts per 100,000. Above that amount 5 cc. or less of the water under examination should be taken for the determination.

The authors direct that their reagent be made with "fuming sulphuric acid (13 per cent SO3)." I have not been able to find a source of an acid of this composition. Kahlbaum supplies fuming sulphuric acid containing

25 grms. of pure phenol, might be used for making the reagent recommended, in place of the quantities given by the authors. This reagent requires to be brought into intimate contact with the nitrate residue by rubbing with a glass rod. This procedure, however, is totally unnecessary with my reagent, and I have never used it. Probably the water and hydrochloric acid present help materially to bring about quick solution of the residue. I do not know whether the authors "add slowly a strong solution of potassium hydroxide (10 to 12 normal) until the maximum colour is developed" to each residue they are dealing with; but I should have thought that the best plan would be to make a special KOH solution containing, say, 600 grms. per litre, add some gradually from a measuring cylinder to a diluted residue of 1 cc. standard KNO, solution, after the phenolsulphonic acid has been added, and note the amount required to give the maximum colour. Then mark the bottle with that amount, and always add the same measured quantity in each case.

in the addition of a three-way capillary tap. In Fig. 2 the general arrangement of the apparatus adopted when it is desired to demonstrate the proportions in which various gases unite by volume is shown.

Rubber rings are placed round the eudiometer in convenient positions, the tap is opened, and the air expelled by sinking the tube in the mercury; the tap is then closed, and the tube raised until the upper ring is at the level of the mercury in the cylinder. The tube A is then con

nected to a gas generator or holder, a Kipp's apparatus in which hydrogen is being formed, for example; the tap is turned so that A and B communicate, the gas is allowed to stream through в until the air is displaced from the connecting tubes; the tap is then turned, and gas admitted to the eudiometer until the desired volume has been collected. The eudiometer is again raised until the second ring is level with the mercury in the reservoir, after which the second gas is admitted.

Should any gas remain after the explosion it can easily be obtained and examined by opening the tap, and sinking the eudiometer.

Though primarily designed for lecture demonstrations the apparatus may be used for exact work, graphite being used as a lubricant in place of vaseline, &c.

The eudiometer has the following advantages over the ordinary form :

1. The tube can be completely filled with mercury, the mercury drives the air out in front of it, and no bubbles stick to the sides.

2. The volumes of the gases used can be exactly controlled.

3. The residual gas, if any, can be collected in a pure state and tested.

4. Each experiment occupies five or six minutes only, so that

5. A series of experiments can be carried out during the same lecture, allowing of the demonstration of the validity of Gay Lussac's Law, whatever the proportions of the gases used, and the damping effect of excess of either gas on the violence of the explosion.

Working Men's College, Melbourne, Victoria.

NEW ORGANIC COMPOUNDS OF NITROGEN.* By Prof. MARTIN O. FORSTER, D.Sc., Ph.D., F.R.S.

Ir may be stated without fear of contradiction that the most versatile form of elemental matter is nitrogen. Rivalled only by argon and its associates in reluctance to take part in chemical action, its entrance into combination with other elements leads to interesting forms of activity in great profusion. Union with hydrogen in different proportions, for example, produces ammonia, hydrazine, and hydrazoic acid, three highly reactive substances having characteristics which stand in marked contrast with one another. If oxygen be brought into the system, hydroxylamine, nitrous acid, and nitric acid may be mentioned as typical materials capable of entering into chemical changes of the most diverse order.

Organic derivatives of nitrogen, however, present an even greater display of individuality. Prussic acid, the alkaloids, nitroglycerine, gun cotton, celluloid, artificial musk, lyddite, indigo, the azo-dyes, hæmoglobin, and the enzymes are a few of the conspicuous nitrogen compounds which suggest themselves in this connection, and a survey of their activities would justify a reference to nitrogen as

"An element so various it seems to be

Not one, but all Hermetic Art's epitome."

It is not the occasion, however, to discuss the foregoing materials, my present purpose being rather to deal with some new organic derivatives of nitrogen which, although not associated with any important industrial development, nevertheless display properties of considerable interest to

chemists.

The extraordinary inertness of elemental nitrogen has been already mentioned, and the underlying cause of this feature is the tenacity with which two atoms of the substance remain in combination with each other. But in circumstances which will be explained later, three atoms of nitrogen may be brought into and maintained in com

* A Discourse delivered before the Royal Institution, May 5, 1911.

bination, the resulting complex being known as the triazogroup. It is a remarkable fact that although, as has just been seen, nitrogen is distributed among naturally occurring substances almost as widely as carbon itself, there is no recorded example of a triazo-compound being obtained from natural sources. Clearly then, the union of atoms in the triazo-group is not among those marriages which are made in Heaven.

That being the case, it becomes necessary to consider the materials and processes which underlie the synthetical production of triazo-compounds. Proceeding from ammonia, and replacing one of the hydrogen atoms by an amino-group, diamide, or hydrazine, is produced, but although discovered by Curtius in 1887, the method by which he prepared it is of historical interest only, it having been left to the ingenuity of Raschig, as recently as 1907, to accomplish the production of this substance by a simple and inexpensive process. This consists in first replacing an atom of hydrogen in ammonia by an atom of chlorine, and then, by direct action of more ammonia on the resulting chloramine, replacing the halogen by an amino-group:— NH3+NaOCl = NH2Cl + NaOH NH3+NH,C1 = NH2NH2,HCI.

The new process illustrates in remarkable fashion the effect which physical condition may exert on chemical change. Although chloramine and ammonia may act together as already shown in a constructive direction, an alternative, destructive course is open to them, which is accelerated if the viscosity of the liquid is diminished :

2NH3+3NH2Cl = N2 + 3NH4Cl.

Thus Raschig was able to increase the yield of hydrazine by increasing the viscosity of the solution with addition of glue, whilst the undesirable effect, namely, liberation of nitrogen, was shown to follow the addition of acetone.

Just as ammonia may be doubled on itself to produce diamide or hydrazine, the latter might be expected to allow one of its hydrogen atoms to be replaced by another aminogroup and furnish triamide, or triazane. This, however, has not been accomplished, and there is evidence to suggest that such a substance would be most unstable. Whenever attempts are made to add another nitrogen link to the chain of two atoms in hydrazine, the terminal atoms of the three-link chain are found to have combined with one another, forming an enclosed association or ring of atoms. This is the triazo-group, and its simplest known form is the compound with hydrogen called azoimide, or hydrazoic acid, HN3. The latter name emphasises the first point of interest in connection with the triazo-group. Whereas the simpler compounds of nitrogen with hydrogen, namely, ammonia and hydrazine, are both strong bases, forming very stable salts with acids, azoimide is a well-defined acid, forming salts with bases including ammonia and hydrazine themselves, the products then having composi tion expressed by the curious formula, NH, and N5H5. Another salt, that with ferric iron, provides a very delicate test for hydrazoic acid, the red colour being noticeable in solutions containing only one part per million.

Hydrazoic acid, discovered in 1890, also by Curtius, may be shown by experiment to follow an attempt to add another atom of nitrogen to hydrazine. Whilst the action of nitrous acid on ammonia causes both atoms of nitrogen to be liberated in elemental form, HNO2+NH3 = N2+2H2O, hydrazine is converted by the same agent into hydrazoic acid, HNO2+ NH2 NH2=HN3+2H20. This process, however, is not adapted to the production of hydrazoic acid or its salts in large quantities. Two methods are available for this purpose. One of these, due to W. Wislicenus (1892), consists in passing a current of dry nitrous oxide over shallow layers of powdered sodamide at about 200°, and by means of certain modifications good yields of sodium azide may be obtained:

NaNH2 + N2O = NaN3+ H2O.

A later method, elaborated by Stollé and Thiele working independently (1908), is based on the experiment just