Chemical Properties of Argon

Argon is generally believed to be chemically inert, but it may be of interest to give an account, as was done in the case of helium, of the experimental evidence upon which this belief rests.

Argon is not acted upon by any of the agents - magnesium, calcium, lithium, calcium carbide, oxygen, potash, copper oxide, etc. - used in removing nitrogen and oxygen from air and in purifying the residual gas. This is evidenced by the fact that the proportion of argon in air determined by processes involving the use of these reagents agrees with that found by other methods. Moreover, it does not combine with titanium, boron, or uranium when heated in contact with these elements, or with fluorine at the ordinary temperature or under the influence of the electric spark.

Rayleigh and Ramsay attempted to cause argon to combine with the many reagents, but in no case was there any absorption of the gas.

Other experiments have shown that argon is unaffected by a carbon arc maintained in the gas for several hours, or by the silent electric discharge in presence of the vapour of carbon tetrachloride. By the action of producer gas (which of course contains argon) on a mixture of barium carbonate and carbon at the temperature of the electric arc, a product rich in barium cyanide is obtained, but when a sample of this material was completely decomposed according to Dumas' method the nitrogen recovered was found to contain no argon. Similarly, it has been proved that argon has no action on magnesium, as the magnesium nitride obtained by heating the metal in air, on treatment with water yielded ammonia and a small amount (50 c.c. from 500 gm. nitride) of gas which was shown to contain no argon. Argon does not pass through platinum, palladium, or iron at 900°-950°, under a pressure of 1 atmosphere, and therefore forms no compound or solid solution with these metals. It has been shown independently that argon has no appreciable solubility in solid or liquid copper, silver, gold, nickel, iron, palladium, aluminium, magnesium, uranium, or tantalum.

It has been suggested that this inertness might be purely a matter of temperature; and argon at ordinary temperatures has been compared with mercury vapour, which at 800° C. would appear to be incapable of combining with any other element. This analogy is obviously imperfect, and argon should be compared rather with non-metals of low atomic weight and density than with mercury.

It seemed likely that any compound of argon that might be produced would be endothermic, and, very possibly, unstable: if this were the case it might be possible to cause its formation at a high temperature and preserve it from decomposition by rapidly cooling it to a low temperature. Fischer, therefore, devised an apparatus in which an arc or spark discharge could be maintained between metallic electrodes submerged in liquid argon. Various metals were used - cadmium, titanium, tin, lead, antimony, and bismuth - and in some cases small amounts of metallic nitride were formed from traces of nitrogen present in the argon, but in no case was any evidence obtained of combination between the metal and argon.

Such are the negative results hitherto obtained: it is, however, necessary to give a brief resume of those cases which have been put forward as positive evidence of the formation or existence of compounds of argon.

It is a fact that if a powerful discharge be sent through argon in a vacuum tube having platinum or magnesium electrodes, the spectrum of the gas gradually diminishes in intensity and finally vanishes. This phenomenon at first received the obvious explanation - that the argon was absorbed or occluded by the electrodes; but it is now known that the inert gases, when freed from ordinary diatomic gases by combination of the latter with the electrodes, may become fluorescent and even non-conducting at pressures measurable on a mercury barometer. On the other hand, attention has been called by Ramsay to the fact that electrodes of platinum, magnesium, aluminium, zinc, cadmium, antimony, and mercury always exhibit more "spluttering" in inert gases than in hydrogen, oxygen, nitrogen, etc.; and that vacuum tubes filled with inert gases that have been confined over mercury more frequently show the spectrum of mercury than do tubes similarly filled with ordinary gases.

Again, Cooke has stated that the vapour-density of zinc is 12 per cent, higher in argon than in nitrogen, and this has been put forward as evidence of a tendency to form a compound. But it has since been shown that Cooke's results are untrustworthy, and the vapour densities of zinc, aluminium, and magnesium are unaffected by the presence of argon: this argument therefore falls to the ground.

A very extensive series of experiments was made by Berthelot in order to ascertain the effect of the silent electric discharge on mixtures of argon with the vapours of volatile organic compounds - both fatty and aromatic - and of carbon disulphide, in presence of mercury vapour. With the aromatic compounds (but not with the aliphatic compounds) an absorption of argon was recorded in every case. The diminution in volume of the gas varied from 1 per cent, to 8 per cent, in various experiments, and was accompanied by a green luminosity the spectrum of which showed the lines of argon, mercury, carbon, and hydrogen.

At one time it seemed probable that sufficient care had not been taken to ensure the removal of nitrogen from the gas, and that the phenomena observed were due to the presence of that gas; but Ramsay has repeated these experiments and has found that while both argon and nitrogen are present, the ordinary violet glow is observed, but when all nitrogen has been removed, the tube gives a brilliant green glow which shows the mercury spectrum. The contemplation of these results has a very disquieting effect upon the mind accustomed to regard all these gases as invincibly indifferent to all other elements, and further experiment seems to be needed.

Villard states that when pure argon is compressed at 150 atmospheres in the presence of water at 0°, local supercooling causes the formation of a solid, crystalline hydrate. The dissociation pressure of this body is given as 105 atmospheres at 0° and 210 atmospheres at 10°.

The detection or estimation of argon in a mixture of gases is carried out by methods essentially the same as those used for its preparation. For example, in detecting the argon present in air, oxygen is absorbed by metallic copper, and nitrogen by heated lithium, calcium, or Maquenne's mixture, and the residual gas is examined spectroscopically. Fairly complete removal of nitrogen is necessary, as its spectrum readily masks that of argon: even 37 per cent, of argon in nitrogen can scarcely be detected spectroscopically. To estimate the amount of argon present the same procedure is followed, except that special precautions are taken to prevent any leakage of gas, and the last traces of oxygen, nitrogen, and hydrogen are absorbed by red-hot metallic calcium. The method of using lithium as an absorbent for nitrogen, etc., has been elaborated and used more especially by Schloesing.

The purity of argon may be tested by a determination of its dielectric cohesion: 1 per cent, of air or other diatomic gas multiplies it about 2½ times.