Thus the first positive and very important fact respecting argon is that its specific gravity is nearly 20—that is, that it is 20 times heavier than hydrogen, while nitrogen is only 14 times and oxygen 16 times heavier than hydrogen. This explains the difference observed by Rayleigh between the densities of nitrogen obtained from its compounds and from the atmosphere (Chapter V., Note [4 bis]). At 0° and 760 mm. a litre of the former gas weighs 1·2505 grm., while a litre of the latter weighs 1·2572, or taking H = 1, the density of the first = 13·916, and of the latter = 13·991. If the density of argon be taken as 20, it is contained in atmospheric nitrogen to the extent of about 1·23 per cent. by volume, whilst air contains about 0·97 per cent. by volume.
When argon had been isolated the question naturally arose, was it a new homogeneous substance having definite properties or was it a mixture of gases? The former may now be positively asserted, namely, that argon is a peculiar gas previously unknown to chemistry. Such a conviction is in the first place established by the fact that argon has a greater number of negative properties, a smaller capacity for reaction, than any other simple or compound body known. The most inert gas known is nitrogen, but argon far exceeds it in this respect. Thus nitrogen is absorbed at a red heat by many metals, with the formation of nitrides, while argon, as is seen in the mode of its preparation and by direct experiment, does not possess this property. Nitrogen, under the action of electric sparks, combines with hydrogen in the presence of acids and with oxygen in the presence of alkalis, while argon is unable to do so, as is seen from the method of separation from nitrogen. Rayleigh and Ramsay also proved that argon is unable to react with chlorine (dry or moist) either directly or under the action of an electric discharge, or with phosphorus or sulphur, at a red heat. Sodium, potassium, and tellurium may be distilled in an atmosphere of argon without change. Fused caustic soda, incandescent soda-lime, molten nitre, red-hot peroxide of sodium, and the polysulphides of calcium and sodium also do not react with argon. Platinum black does not absorb it, and spongy platinum is unable to excite its reaction with oxygen or chlorine. Aqua regia, bromine water, and a mixture of hydrochloric acid and KMnO4 were also without action upon argon. Besides which it is evident from the method of its preparation that it is not acted upon by red-hot oxide of copper. All these facts exclude any possibility of argon containing any already known body, and prove it to be the most inert of all the gases known. But besides these negative points, the independency of argon is confirmed by four observed positive properties possessed by it, which are:—
1. The spectrum of argon observed by Crookes under a low pressure (in Geissler-Plücker tubes) distinguishes it from other gases.[7] It was proved by this means that the argon obtained by means of magnesium is identical with that which remains after the conversion of the atmospheric nitrogen into nitric acid. Like nitrogen, argon presents two spectra produced at different potentials of the induced current, one being orange-red, the other steel-blue; the latter is obtained under a higher degree of rarefaction and with a battery of Leyden jars. Both the spectra of argon (in contradistinction to those of nitrogen) are distinguished by clearly defined lines.[8] The red (ordinary) spectrum of argon has two particularly brilliant and characteristic red lines (not far from the bright red line of lithium, on the opposite side to the orange band) having wave-lengths 705·64 and 696·56 (see Vol. I., p. [565]). Between these bright lines there are in addition lines with wave lengths 603·8, 565·1, 561·0, 555·7, 518·58, 516·5, 450·95, 420·10, 415·95 and 394·85. Altogether 80 lines have been observed in this spectrum and 119 in the blue spectrum, of which 26 are common to both spectra.[9]
2. According to Rayleigh and Ramsay the solubility of argon in water is approximately 4 volumes in 100 volumes of water at 13°. Thus argon is nearly 2½ times more soluble than nitrogen, and its solubility approaches that of oxygen. Direct experiment proves that nitrogen obtained from air from boiled water is heavier than that obtained straight from the atmosphere. This again is an indirect proof of the presence of argon in air.
3. The ratio k of the two specific heats (at a constant pressure and at a constant volume) of argon was determined by Rayleigh and Ramsay by the method of the velocity of sound (see Chapter XIV., Note [7] and Chapter VII., Note [26]) and was found to be nearly 1·66, that is greater than for those gases whose molecules contain two atoms (for instance, CO, H2, N2, air, &c., for which k is nearly 1·4) or those whose molecules contain three atoms (for instance, CO2, N2O, &c., for which k is about 1·3), but closely approximate to the ratio of the specific heats of mercury vapour (Kundt and Warburg, k = 1·67). And as the molecule of mercury vapour contains one atom, so it may be said that argon is a simple gaseous body whose molecule contains one atom.[10] A compound body should give a smaller ratio. The experiments upon the liquefaction of argon, which we shall presently describe, speak against the supposition that argon is a mixture of two gases. The importance of the results in question makes one wish that the determinations of the ratio of the specific heats (and other physical properties) might be confirmed with all possible accuracy.[11] If we admit, as we are obliged to do for the present, that argon is a new element, its density shows that its atomic weight must be nearly 40, that is, near to that of K = 39 and Ca = 40, which does not correspond to the existing data respecting the periodicity of the properties of the elements in dependence upon their atomic weights, for there is no reason on the basis of existing data for admitting any intermediate elements between Cl = 35·5 and K = 39, and all the positions above potassium in the periodic system are occupied. This renders it very desirable that the velocity of sound in argon should be re-determined.[12]
4. Argon was liquefied by Professor Olszewsky, who is well known for his classical researches upon liquefied gases. These researches have an especial interest since they show that argon exhibits a perfect constancy in its properties in the liquid and critical states, which almost[13] disposes of the supposition that it contains a mixture of two or more unknown gases. As the first experiments showed, argon remains a gas under a pressure of 100 atmospheres and at a temperature of -90°; this indicated that its critical temperature was probably below this temperature, as was indeed found to be the case when the temperature was lowered to -128°·6[14] by means of liquid ethylene. At this temperature argon easily liquefies to a colourless liquid under 38 atmospheres. The meniscus begins to disappear at between -119°·8 and -121°·6, mean -121° at a pressure of 50·6 atmospheres. The vapour tension of liquid argon at -128°·6, is 38·0 atmospheres, at -187° it is one atmosphere, and at -189°·6 it solidifies to a colourless substance like ice. The specific gravity of liquid argon at about -187° is nearly 1·5, which is far above that of other liquefied gases of very low absolute boiling point.
The discovery of argon is one of the most remarkable chemical acquisitions of recent times, and we trust that Lord Rayleigh and Professor Ramsay, who made this wonderful discovery, will further elucidate the true nature of argon, as this should widen the fundamental principles of chemistry, to which the chemists of Great Britain have from early times made such valuable contributions. It would be premature now to give any definite opinions upon so new a subject. Only one thing can be said; argon is so inert that its rôle in nature cannot be considerable, notwithstanding its presence in the atmosphere. But as the atmosphere itself plays such a vast part in the life of the surface of the earth, every addition to our knowledge of its composition must directly or indirectly react upon the sum total of our knowledge of nature.
Footnotes:
[1] From the Greek Αργὸν—inert.
[1 bis] In Note [16 bis], Chapter V., I mentioned that, judging from the specific gravity of argon, it might possibly be polymerised nitrogen, N3, bearing the same relationship to nitrogen, N2, that ozone, O3, bears to ordinary oxygen. If this idea were confirmed, still one would not imagine that argon was formed from the atmospheric nitrogen by those reactions by which it was obtained by Rayleigh and Ramsay, but rather that it arises from the nitrogen of the atmosphere under natural conditions. Although this proposition is not quite destroyed by the more recent results, still it is contradicted by the fact that the ratio of the specific heats of argon was found to be 1·66, which, as far as is now known, could not be the case for a gas containing 3 atoms in its molecule, since such gases (see Chapter XIV., Note [7]) give the ratio approximately 1·3 (for example, CO2). In abstaining from further conclusions, for they must inevitably be purely conjectural, I consider it advisable to suggest that in conducting further researches upon argon it might be well to subject it to as high a temperature as possible. And the possibility of nitrogen polymerising is all the more admissible from the fact that the aggregation of its atoms in the molecule is not at all unlikely, and that polymerised nitrogen, judging from many examples, might be inert if the polymerisation were accompanied by the evolution of heat. In the following footnotes I frequently return to this hypothesis, not only because I have not yet met any facts definitely contradictory to it, but also because the chief properties of argon agree with it to a certain extent.