734
CHEMISTRY
occurrence of change, evidence that moisture exercises a determining influence is afforded by the observation that the rate of change is affected by the withdrawal of moisture. Thus Pringsheim, Dixon, and Baker have all found that the rate at which hydrogen and chlorine interact is diminished by carefully drying the gases. Experiments made by Y. Meyer, in conjunction with Krause and Askenasy {Liebig's Ann. 264, p. 85; 269, p. 49), show that the same mixture of hydrogen and oxygen sealed up in a number of bulbs and heated to various temperatures behaved irregularly, interaction occurring in some to a far greater extent than in others placed under apparently the same conditions. Such results may be regarded as proof that even the interaction of hydrogen and oxygen is determined by a third substance. But there are a number of cases in which the attempt has been unsuccessfully made to check the occurrence of change by removing moisture (cf. Cowper, Apparent ^ . jj g Baker, Trans. Chem. Soc. 1894, exceptions. ^ 611); of none of these can it be asserted, however, that the precautions taken have been sufficient to preclude the assumption that with still greater care positive results would be obtained. Among the most, indeed the most, carefully conducted experiments of the kind are those of Shenstone on the production of ozone, and on the interaction of mercury and halogens {Trans. Chem. Soc. 1897, p. 471). According to this observer “well-dried oxygen ozonizes exceedingly badly,” whence it may fairly be assumed that dry oxygen would not ozonize. On the other hand, when the well-dried gas was ozonized the ozone produced appeared to undergo decomposition very rapidly, while that produced when moist oxygen was used seemed to be stable. Shenstone therefore assumes that water vapour, at the ordinary temperature, retards the reversion of ozone to oxygen in a very remarkable degree. That this should be the case appears highly improbable, and it is necessary to seek for an explanation of the anomaly in some other direction. It is conceivable that a minute amount of “ acid ” impurity may intervene in the case of the dried gas and determine its decomposition, but that in the wet gas this impurity is rendered more or less harmless by condensation. Such impurity might be derived from the interaction of nitrogen and oxygen under the influence of the discharge, for it is known that “oxides of nitrogen” determine the decomposition of ozone. As regards the interaction of halogens and mercury, Shenstone was unable to detect any alteration in their activity consequent on drying. But when the complex character of his apparatus is taken into account, as well as the fact that it was not all made of hard glass, and that it could not be dried at a high temperature, it is more than probable that he was not able to carry the drying sufficiently far, especially when it is remembered how extraordinarily minute, according to H. B. Baker’s later observations, is the amount of moisture required to induce decomposition in the case of mercurous chloride. Chlorine being a far more active agent than oxygen under ordinary conditions, it is to be expected that it will be far more difficult to check the occurrence of change when it is used than is generally the case with oxygen. There can be little doubt that the evidence is sufficient to justify the conclusion that chemical interactions are not generally of the simple form AB + CD = AC + BD, and that their occurrence is, as a rule, dependent on the introduction of some “ third component ” into the system. It is even probable that this is a rule absolute. If the attempt be made to determine the conditions which
must prevail in order that an interchange may occur, with a view of determining the nature of the Mature “ third component,” the only conclusion at pre- of third sent possible is that they are essentially those substance. which obtain in a voltaic circuit; in fact, if Faraday’s doctrine of the unity of chemical and electrolytic change be accepted it follows that both are determined by like conditions. On this assumption, a’circuit of change must comprise three distinct terms or components, one of which must be an electrolyte. On this view, it is easy to understand the occurrence of change in many cases on the introduction of moisture. It is only necessary to bear in mind that as we invariably operate in glass vessels which are to some extent soiled, it is impossible to avoid the presence of traces of acid or “salts,” which render the water an electrolyte, and therefore that the introduction of moisture means the introduction of an electrolyte. Thus, in the case of the combustion of carbonic oxide, the system within which change occurs may be regarded as consisting of oxygen, carbonic oxide, and impure, i.e. conducting water, which conjoined form a conducting circuit within which the interaction occurs, viz. :— H.,0 CO oh2 o_co 2 CO on. o coI h;o As early as 1830 it was shown by de la Rive that the readiness with which zinc is attacked by an aqueous solution of sulphuric acid depends upon the amount of impurity it contains, and that the acid which acts most readily is that which has the highest electric conducting power. When purified as far as possible zinc is all but insoluble, and therefore it may fairly be supposed that ideal pure zinc would be insoluble in diluted sulphuric acid. It at once dissolves when coupled up with any relatively electronegative conductor at a rate depending on the resistance in the circuit in which it is arranged and on the electromotive force developed in the interaction. The conditions which determine the dissolution of zinc may therefore be said to be that it forms one of the three components in a conducting circuit comprising itself, a relatively electronegative chemically inert conductor, and an electrolyte which can exchange its electro-positive constituent for the zinc. The apparent inactivity of “pure” zinc has been ascribed to the circumstance that when placed in acid it immediately becomes coated with a protecting Voltaic film of hydrogen—varnished, as it were—and and that the part played by the impurity is that of chemical depolarizer : that, in fact, it conditions the escape change correlaof the hydrogen. That the impurity, to some tive. extent, exercises this latter function, there can be no doubt; but, on the other hand, there is every reason to believe that three “components” are required to form a voltaic circuit. It is well known that the electromotive force of a zinc-sulphuric acidcouple (or triplet, to use a far more suitable expression) is a fluctuating value, and that the nature of the conductor is of some importance. Thus a cell containing a polished silver plate as positive element is less effective than one in which the plate is coated with finely divided platinum. It is not difficult to understand this, as the extent to which hydrogen escapes from the positive is much influenced by the character of its surface, and it is well known that finely divided platinum promotes the escape of gases from solution. Therefore when finely divided platinum is thus used, not only is the effect diminished which retention of hydrogen on the surface of the plate would have in increasing the resistance in the circuit, but the extent to which “polarization” or reversal takes place is also considerably diminished. In fact, it must not be forgotten that the action is a reversible one, i.e.,
