NITROGEN has long been regarded as an element possessed of very feeble affinities, and consequently the recent remarkable experiments of Strutt and collaborators on "Chemically Active Nitrogen" (Proc. Roy. Soc., 1911, A, lxxxv., 219, 377; lxxxvi., 56, 262; lxxxvii., 179; Nature, May 15, 1913, p. 283), which revealed the possibility of nitrogen possessing strong affinities, came as a very surprising result to many chemists.

It seems, however, to be quite overlooked that the present writer no less than eight years ago in his book "Researches on the Affinities of the Elements" (published by Churchill, 1905), showed from a study of the affinities of the successive elements of the periodic system that nitrogen, owing to its position in the periodic system, must possess very strong affinities.

This result was arrived at by studying the manner in which the affinities of the successive elements altered as we went down the series Li, Be, B, C, N, O, F, and the writer then finally concluded that it is quite incorrect to speak of chemically active and inactive elements, but that practically all the elements investigated have their strong and weak affinities, only sometimes these strong affinities are exerted on well known elements-when we call the element "chemically active "—and sometimes on rare elements-when we call the element "inactive."

As the result of the study of the compounds of nitrogen and the compounds of neighbouring elements the writer showed that nitrogen, just like sodium or chlorine, possesses very strong affinities for some elements and very weak ones for other elements, but that its affinities were masked by three causes:

(1) The elements for which it exhibits its strongest affinities are those which have the power of selfcombination most strongly developed.

(2) That nitrogen itself has the power of self-combination strongly developed.

(3) That the elements for which nitrogen exhibits its maximum affinities are all somewhat uncommon in a chemical laboratory, and so the chemical activities of nitrogen are not brought prominently before the eyes of chemists.

All these causes lower the apparent power of entering into chemical combination of nitrogen, and have led chemists to almost universally regard nitrogen as very inert chemically.

Strutt's experiments revealing the chemical activity of nitrogen are, in fact, a remarkable fresh confirmation of the properties that I showed nitrogen to possess some eight years ago, and as Dr. Strutt refers back to Faraday (Nature, May 15, 1913, p. 285) for a vague prophesy in this direction, and appears to be quite unaware of my own results-arrived at by means of a laborious study of known compounds-I think that I can justly claim that I was the first who showed that from its position in the periodic system nitrogen must be an element possessing strong affinities, and also simultaneously clearly indicating what elements these affinities would be most strongly exerted upon.

As these results of mine appear to have been quite overlooked by current writers, and have considerable interest at the present time when atmospheric nitrogen is being fixed on an enormous scale commercially, I venture to bring my results once more before the chemical world.

The following is my method of investigating the magni

elements of the periodic system. First of all I construct the periodic system of elements in the usual way, confining myself, however, solely to the elements of the even series, thus:Li Be Na Mg

:

XERZ

B

с

N

Ne

ΑΙ

Si

P

S

CI

Cu

Zn Ga

Ge

As

Se

Br

Xe

Pb Bi

Cd In Au Hg ΤΙ

FIG. 1.-AFFINITY SURFACE FOR NITROGEN. Showing at a glance how the Affinities of Nitrogen for the various Elements of the Periodic System vary. (Taken from the writer's book, "Researches on the Affinities of the Elements").

such a comparative study it is possible to arrive at a very good idea of the relative strengths of the attractive forces which the element A exerts on each of the above elements.

Next from each of the elements of the above table I erect a perpendicular proportional to the chemical attrac tion or affinity which it possesses for the given element A.

A surface is then described through the summits of these perpendiculars, and we get what I call the "affinity surface" of the element A, in respect to the other elements of the periodic system.

The shape of this surface shows at a glance the relative magnitude of the affinities exerted on the different elements of the periodic system.

The details of constructing these surfaces, together with all the data on which the results are founded, is collected together in the writer's book, "Researches on the Affinities of the Elements," and I do not propose to go further into

this matter here, because a detailed study of the individual compounds formed by each of the thirty-seven elements of the above table with the other elements requires a very large space for discussion, and indeed a volume of some hundreds of pages was necessary for the purpose. As this data is collected together in the above-mentioned book, I will here only apply the results obtained from a study of the affinity surfaces of the successive elements of the first two cycles of the periodic system to nitrogen so as to show that we should expect nitrogen to exhibit very strong affinities for some elements and weak ones for other elements, just as the other elements of the periodic system do.

Fig. 1 shows the form that the affinity surface assumes for nitrogen.

Description of the Affinity Surface for Nitrogen. The numbers given refer to the heights of the perpendiculars erected from the various elements below mentioned as proportional to the affinity that nitrogen possesses for them.

Cross Sections.

Li, Be, B, C, N, O, F.-The affinity for nitrogen rises from Li to B, then falls, attaining a minimum at C, rises again at N, and then falls, becoming very small at F. Li 41, Be 5, B 8, C o'5, N 4, O 0.2, F 0.1. Na, Mg, Al, Si, P, S, Cl.-The affinity for N is very small at Na, rises steadily, attaining a maximum at P, then falls again, becoming very small at Cl.

Na 1, Mg 6, Al 7, Si 9, P 11, S 0'2, Cl o'I.

Cu, Zn, Ga, Ge, As, Se, Br.-The attraction is very small for Cu, rising as we pass from Cu to Zn, and it remains appreciable until As is reached, then_falling it becomes very small for Se, and still smaller for Br.

Cu o 4, Zn 3, Ga 3, Ge 3, As 3, Se 0'2, Br o'I. Ag, Cd, In, Sn, Sb, Te, I.-The affinity is small for Ag and Cd. It is unknown for In, Sn, Sb, but probably rises from Ag to Sb; for Sb it is very small, and is still smaller

for 1.

Ag o'3, Cd 2.5, In 2.5, Sn 2.5, Sb 10, Te 0.2, I o.15. Au, Hg, Ti, Pb, Bi.-The attraction for nitrogen is very small for these elements.

Au o'1, Hg 0'2, Tl o'2, Pb o'2, Bi 0.2.

Longitudinal Sections.

H, Li, Na, Cl, Au.-The attraction for N increases from H to Li and then steadily decreases until Au is reached.

H1, Li 41, Na 10, Cu o'4, Ag o'3, Au 0'2.

Be, Mg, Zn, Cd, Hg.-The attraction for N in the case of Be is unknown, but probably it is greater than in the case of Mg; the affinity probably reaches a maximum in the case of Mg and then sinks, becoming very small for Hg.

Be 7, Mg 6, Zn 3, Cd 5.5, Tl o'2.

B, Al, Ga, In, T.-The attraction for N at B is very great, and sinks thence to Al; the affinity of N for the other elements is unknown, but probably sinks steadily from B to TI.

B 8, Al 7, Ga 3, In 25, Tl o'2.

C, Si, Ge, Sn, Pb.-The affinity for N is very feeble at C, but it rises greatly as we pass from C to N; for the other elements it is unknown, but probably sinks rapidly from C to Pb.

Co5, Si g'o, Ge 3'0, Sn 2.5, Pb 0.2.

N, P, As, Sb, Bi.-The attraction for N rises rapidly from N to P; the attraction for N of the other elements is unknown, but probably is feeble, falling from As to Bi. N 4, P 8, As 3, Sb 1, Bi 0.5.

O, S, Se, Te-The affinity for N is small all through the series, sinking from O to S, remaining about the same for Se as for S, then rising somewhat for Te.

O o 2, S o'2, Se o 2, Te o'3.

F, Cl, Br, I.-The attraction for N is very small for all these elements, being greatest for I and least for F. Fo1, Cl o'1, Br o'1, I o.15.

Now let us compare these affinities of nitrogen with those of neighbouring elements. If we arrange in order the affinity surfaces of the first series of elements of the periodic system (viz., Li, Be, B, C, N, O, F), a remarkable fact becomes apparent, which is illustrated in Fig. 2. The affinity surfaces of the successive elements assume the appearance of successive positions of an advancing wave, whose crest appears on the extreme right hand side of the diagram in the case of Li, and then sweeps from right to left as we pass from Li towards F, until at F the crest is on the extreme left-hand side (see Fig. 2).

The same phenomenon appears with the elements of the second (Na, Mg, Al, Si, P, S, Cl, Ar) cycle, and later cycles of the periodic law, only in most cases the wave form seems to pass more or less discontinuously from right left, and there often seems to be a depression at the elements of the fourth group, brobably an apparent effect due to the abnormally developed power of self-combination possessed by these elements. For example, when we come to compare the heats of formation of the two oxides of carbon, we have, according to Thomsen, the following (C, O)=29,000 cal., (CO, 0) = 67,960 cal.,

numbers :

so that the addition of the second atom of oxygen to the carbon atom is attended with the liberation of a far greater I quantity of visible heat than the addition of the first atom. This is a remarkable result, quite at variance with the usual rule (see "Affinities of the Elements," p. 241 et seq.) which holds in chemistry, and the explanation is, probably, that the small quantity of heat evolved in the formation of CO is due to the fact that before an oxygen atom can combine at all with a carbon atom it has to separate it from its combinations with other carbon atoms. The carbon atoms attract each other very strongly indeed, as is shown not only by their great power of self-combina. tion, but also by the great hardness and involatility of certain forms of carbon. Consequently this initial separation of the carbon atoms absorbs a very large quantity of heat, and this lowers the apparent heat of formation of CO to much below its real value.

In gaseous CO, however, the carbon atoms are already separated, and so no further great quantity of heat is absorbed before the additional O can be added on to form CO2. It is very probable, therefore, that the apparent heats of formation and stability of the oxides, and indeed of all other compounds of carbon, are very much below their real values on account of the exceptionally large quantities of heat absorbed in separating the carbon atoms from each other. This is probably the reason why the heat of formation of CH, is less than the heat of formation of SiH4, although CH, is a much more stable body than SiH4.

Silicon is the element next to carbon which possesses to the greatest extent the power of self-combination. The

apparent heat of formation of its oxide is also, therefore, probably much less than its true value. In the case of certain monatomic atoms, such as those of sodium, mercury, &c., and of elements in which the power of selfcombination is but slightly developed, the apparent heats of formation probably correspond fairly closely with the real heats of formation.

We should therefore expect that when we come to compare the thermal data and stability of the compounds of the various elements we will find that there is an apparent general depression very much below their true values as we approach such elements as carbon and silicon which lie in the middle of a cycle of elements of the periodic system where the power of self-combination is largely developed; whereas there should be nearly correct values for such elements as Na, Hg, &c., which lie on the extreme borders of the cycle of elements, and where the power of self-combination is but feebly developed. A glance of the relative magnitudes of the affinity surfaces of the successive elements exhibited in Fig. 2 shows that this is actually the case.

Applying these results to the affinity surface of nitrogen we will now show that the chemical inactivity of nitrogen is to a great extent only an apparent effect.

A study of the affinity surface for nitrogen shows that the point of maximum chemical affinity lies over boron.

When we contrast the nitrogen affinity surface with those of Na or Cl, it becomes manifest that while Na expends its energies upon elements belonging to the later groups of the periodic system (viz., O, S, Se, F, Cl, Br, I), and has little affinity for elements of the earlier groups (such as the alkali metals and alkali earth metals), and that while chlorine expends its energies principally upon the elements of the early groups of the periodic system (such as the alkali metals and alkaline earth metals), but has little affinity for elements of later groups (such as the halogens and oxygen group of elements) that, in contradistinction to these, nitrogen expends its energies on elements of intermediate groups-principally on those of Groups III., IV., and V.-but has little affinity for those of earlier or later groups (such as the alkali elements and the halogens). For example, its combinations with boron, silicon, and phosphorus are quite remarkable by reason of their great stability. The phosphide P3N2 must be heated in oxygen gas to a temperature above that at

members is double the atomic weight of the means, and are all multiples of seven. Triads of atomic weights have been fully recognised by Dumas, Faraday, and other philosophical chemists, as indubitable evidence of community of origin, of transmutation, and important factors in the classification of elementary substances. Radium (as was indicated in Dr. Wilde's tables of elements some years previous to its discovery) is one of the synthetic transformations of helium, and is the next higher member of the series of barium, as was since confirmed by Mme. Curie. Helium is also shown in the author's table of 1878, as the analytic transformation ultimate of radium and other members of the second series of elements.

The compounds of O with Cl, for example, are characterised by their explosive properties, just as are so many nitrogen compounds.

Moreover, the nitrogen atoms themselves have a very considerable power of self-attraction, as is shown by the stability of the molecule N2 in gaseous nitrogen, and by the occurrence of the diazo group-N-N-in organic chemistry, and this circumstance again lowers the ap parent affinities of nitrogen for other elements in the way previously explained at length, for this attraction of nitrogen for itself has to be overcome before nitrogen will enter into combination with other elements.

On all these grounds, therefore, we must come to the conclusion that the current belief that nitrogen is an element characterised by its chemical inertness, and by the feebleness of its chemical affinities, is therefore hardly correct. Nitrogen possesses chemical affinities of a strength almost comparable with those of chlorine or sodium, only it differs from these elements as regards the elements it exerts on them.

ON SOME NEW RELATIONS OF ATOMIC WEIGHTS, AND TRANSFORMATIONS OF NEON AND HELIUM.

By Dr. HENRY WILDE, F.R.S.

AT an extraordinary general meeting of the Manchester Literary and Philosophical Society, held on July 22nd, a paper was read by Dr. Henry Wilde, F.R.S., on "Some New Multiple Relations of the Atomic Weights of Elementary Substances, and on the Classification and Transformations of Neon and Helium." In several of the author's papers on the "Origin of Elementary Substances," published by the Society, 1878-1906 (see CHEMICAL NEWS, 1878, xxxviii., p. 66 et seq.), special attention was directed to the seventh series of his classification, on account of the magnitude and importance of its primary members in the economy of nature, viz., nitrogen, silicon, iron, and gold.

Silicon in combination with oxygen constitutes more than half the weight of the earth's crust, and is the principal constituent of glass for all the purposes of civilised life.

The arbitrary policy of several writers in doubling the atomic weights of four of the gaseous members of this series, viz., neon, argon, krypton, and xenon, induced the author to review the multiple relations of the seventh series with the important result (1) that six triads are formed out of the eight principal members of the series, in which the sum of the atomic weights of the extreme

The positions of helium and neon, as the transformation ultimates of the second and seventh series respectively, are further interesting in connection with the recent announcements that these elements have been found in glass vessels and tubes in which they had no previous existence. Assuming the reality of these observations, the phenomena not only admit of explication from Dr. Wilde's classifications, but also account for the disthe research. One of the investigators could only find neon, while others, working independently, found helium alone, and in other cases a mixture of both gases. results were of sufficient interest to induce the author to ascertain the composition of various glasses used in the arts.

These

The principal and most important constituent of the glasses tabulated by Dr. Wilde is silicon, the transformation ultimate of which is neon. The next important constituents of the glasses are barium, calcium, and lead; all members of the second series of elements, the transformation ultimate of which is helium. The alkali metals, sodium and potassium, are constituents of nearly all glasses, and their transformation ultimates (with others of the first series) will be hydrogen and neon, but without helium.

All the silicates of the first and second, and some of other series, are easily vitrified in small quantities in laboratory crucibles. Their spectra can then be examined during electrification in tubes (under suitable conditions of temperature and pressure) for the discovery of new elements, and the identification of those already known.

ANALYSIS OF FERRO - TITANIUM.

By A. R. SCOTT.

THE following method for the determination of silicon, titanium, aluminium, iron, and manganese will be found very simple and accurate, especially so with regard to the complete separation of titanium and aluminium.

The method of estimating the aluminium by the long fusion with sodium carbonate, &c., is a long and tedious process, and the results obtained are not at all reliable.

Method.-Weigh out o'25 grm. of the finely powdered alloy, and fuse with about 10 grms. of potassium bisulphate in a deep platinum basin. Fuse until no grit can be felt with a platinum rod. Cool, and put basin and contents into a porcelain evaporating basin. Extract with water, and add 10 cc. of concentrated H2SO4. Allow to stand on edge of hot plate for an hour. Filter into 600 cc. beaker. Wash with hot water, and ignite residue in platinum crucible. Fuse contents of crucible with a little potassium bisulphate, and extract as before with water and a few drops of H2SO4. Filter, and add filtrate to main filtrate. Ignite residue in a tared platinum crucible, cool and weigh. The increase in weight is SiO2, with a little Fe2O3 and TiO2. H.F. the residue, and weigh; the loss is SiO2-calculate to silicon. Dissolve the residue in a little HC1; transfer to small beaker. Add NH4OH, boil and filter; ignite in platinum crucible and weigh as

Aug. 1, 1913

Old Hall Road, Gatley.

PREPARATION OF THE NITRITES OF THE

2. 3R3N.HNO2=2R3N + R3NO3 + 2NO + H2O. The nitrite is formed in each case at a low temperature, which decomposes at a higher temperature into the endproducts of the respective reactions, viz., alcohol and in the case of secondary amines, and the free amine in the nitrogen in the case of primary amines, nitroso-compound amines, however, the corresponding nitrates are also case of the tertiary amines. In the case of the tertiary formed as a result of the decomposition of the nitrites.

to

This explanation does not, as a matter of course, apply the purely aromatic amines which yield diazocompounds.

Experimental.

from Kahlbaum, a few being obtained from E. Merck. The hydrochlorides of the amines were mostly indented They were found to be very pure, and were further purified from traces of ammonium chloride by exhausting them with small quantities of dry absolute alcohol, in which they were found to be almost wholly soluble. The general method of preparing the amine nitrites consists in vacuum distillation in steam. In the case of those nitrites, how

BY THE INTERACTION OF THE HYDROCHLORIDES OF THE BASES AND ALKALI NITRITES. EXPLANATION OF THE ACTION OF NITROUS ACID ON THE AMINES. (PART I).

By PANCHANAN NEOGI, M.A., Senior Professor of Chemistry, Government College, Rajshahi, Bengal, India.

NEOGI and Adhicari (Trans. Chem. Soc., 1911, xcix., 116) have already shown that ammonium nitrite can be prepared in fairly large quantities by subliming in a vacuum a concentrated solution of the mixture of ammonium chloride and alkali nitrites. Neogi (Trans. Chem. Soc., 1912, ci., 1608) has also shown that coniinium nitrite may be prepared by distilling in a current of steam under reduced pressure a mixture of the hydrochloride and sodium or potassium nitrites. In continuation of the work referred to above, a systematic attempt has been made to prepare the amine nitrites by the interaction of the hydrochlorides of the bases and alkali nitrites. Experiments have abundantly shown that this method of preparing the amine nitrites is a general one. The success of the isolation of the amine nitrites from mixtures of the hydrochlorides of the bases and alkali nitrites lies on two facts hitherto unobserved :1. The amine nitrites may be distilled off in steam under reduced pressure in a very pure condition.

2. The amine nitrites, which are comparatively unstable in the solid or liquid condition as the case may be, are, in general, stable in solution, and, in fact, most of them may be concentrated to a small bulk on the water-bath without great loss.

The actual isolation of ammonium nitrite and a large number of amine nitrites affords ample confirmation of the theory already advanced by the author (Neogi, Proc. Chem. Soc., 1911, xxvii., 242; Trans. Chem. Soc., 1912, ci., 1610), that an amine nitrite is an intermediate compound in the well known interaction of amine hydrochlorides and sodium or potassium nitrite, or, in other words, amines and nitrous acid. The following equations show how the reactions really take place in two stages :

were also made to isolate them by vacuum sublimation.

Sublimation in Vacuum.

One or 2 grms. of the hydrochloride were mixed with excess of sodium or potassium nitrite, and dissolved in a very small quantity (10 to 20 cc.) of water. The liquid was evaporated in a flask with a long neck, provided with a "catch" arrangement in order to avoid spirting, at 40° to 50° in the vacuum cf the Töppler pump, and the gases evolved were collected in a Crum nitrometer. In order to drive off the last traces of moisture the entire neck of the flask was immersed in the bath. A capacious test-tube immersed in a beaker of cold water was inserted between the flask and the pump, which served as a condenser. The flask was held in a very slanting position so that particles of the liquid might not spirt into the neck of the flask. After the liquid was completely evaporated to dryness and traces of moisture had been driven off, the neck of the flask was partially raised, and the temperature of the bath was also gradually raised to the temperature at which the nitrite sublimed. When the experiment was completed the bottom of the flask was broken off, and the sublimed salt carefully washed out with water, mixed with the distillate contained in the test-tube, and the solution thus obtained was evaporated in a vacuum desiccator. The crystals were analysed by the "urea" and CrumFrankland methods. As only small quantities of the hydrochloride were available the yield was not very satis factory by this method. The success of the experiment will depend on the success with which the last traces of moisture have been driven off from the flask, as the nitrites are extremely hygroscopic.

Vacuum Sublimation in Steam.

A much better yield, however, of the nitrite was obtained by the process of vacuum distillation in steam. The operation was conducted in two ways:-(1) Distilling the solution in a current of steam whilst a vacuum was maintained, and (2) repeatedly distilling in a vacuum the mixture with fresh çuantities of water. The first of these two processes has the advantage of being continuous, but the bumping is severe. The second process gives the better yield. The best results were obtained in the following manner :-The solution of the mixture was placed in a