Fig. 31.—Cavendish's apparatus for exploding detonating gas. The bell jar standing in the bath is filled with a mixture of two volumes of hydrogen and one volume of oxygen, and the thick glass vessel A is then screwed on to it. The air is first pumped out of this vessel, so that when the stopcock C is opened, it becomes filled with detonating gas. The stopcock is then re-closed, and the explosion produced by means of a spark from a Leyden jar. After the explosion has taken place the stopcock is again opened, and the water rises into the vessel A.
Fig. 32.—Eudiometer
We have already mentioned in the last [chapter] that the combination of these gases, or their explosion, may be brought about by the action of an electric spark, because the spark heats the space through which it passes, and acts consequently in a manner similar to ignition by means of contact with an incandescent or burning substance.[29 bis] Cavendish made this experiment on the ignition of detonating gas, at the end of the last century, in the apparatus shown in fig. [31]. Ignition by the aid of the electric spark is convenient, for the reason that it may then be brought about in a closed vessel, and hence chemists still employ this method when it is required to ignite a mixture of oxygen with a combustible gas in a closed vessel. For this purpose, especially since Bunsen's time,[30] an eudiometer is employed. It consists of a thick glass tube graduated along its length in millimetres (for indicating the height of the mercury column), and calibrated for a definite volume (weight of mercury). Two platinum wires are fused into the upper closed end of the tube, as shown in fig. [32].[31] By the aid of the eudiometer we may not only determine the volumetric composition of water,[32] and the quantitative contents of oxygen in air,[33] but also make a number of experiments explaining the phenomenon of combustion.
Thus, for example, it may be demonstrated, by the aid of the eudiometer, that for the ignition of detonating gas, a definite temperature is required. If the temperature be below that required, combination will not take place, but if at any spot within the tube it rises to the temperature of inflammation, then combination will ensue at that spot, and evolve enough heat for the ignition of the adjacent portions of the detonating mixture. If to 1 volume of detonating gas there be added 10 volumes of oxygen, or 4 volumes of hydrogen, or 3 volumes of carbonic anhydride, then we shall not obtain an explosion by passing a spark through the diluted mixture. This depends on the fact that the temperature falls with the dilution of the detonating gas by another gas, because the heat evolved by the combination of the small quantity of hydrogen and oxygen brought to incandescence by the spark is not only transmitted to the water proceeding from the combination, but also to the foreign substance mixed with the detonating gas.[34] The necessity of a definite temperature for the ignition of detonating gas is also seen from the fact that pure detonating gas explodes in the presence of a red-hot iron wire, or of charcoal heated to 275°, but with a lower degree of incandescence there is not any explosion. It may also be brought about by rapid compression, when, as is known, heat is evolved.[35] Experiments made in the eudiometer showed that the ignition of detonating gas takes place at a temperature between 450° and 560°.[36]
The combination of hydrogen with oxygen is accompanied by the evolution of a very considerable amount of heat; according to the determinations of Favre and Silbermann,[37] 1 part by weight of hydrogen in forming water evolves 34,462 units of heat. Many of the most recent determinations are very close to this figure, so that it may be taken that in the formation of 18 parts of water (H2O) there are evolved 69 major calories, or 69,000 units of heat.[38] If the specific heat of aqueous vapour (0·48) remained constant from the ordinary temperature to that at which the combustion of detonating gas takes place (but there is now no doubt that it increases), were the combustion concentrated at one point[39] (but it occurs in the whole region of a flame), were there no loss from radiation and heat conduction, and did dissociation not take place—that is, did not a state of equilibrium between the hydrogen, oxygen, and water come about—then it would be possible to calculate the temperature of the flame of detonating gas. It would then be 8,000°.[40] In reality it is very much lower, but it is nevertheless higher than the temperature attained in furnaces and flames, and is as high as 2,000°. The explosion of detonating gas is explained by this high temperature, because the aqueous vapour formed must occupy a volume at least 5 times greater than that occupied by the detonating gas at the ordinary temperature. Detonating gas emits a sound, not only as a consequence of the commotion which occurs from the rapid expansion of the heated vapour, but also because it is immediately followed by a cooling effect, the conversion of the vapour into water, and a rapid contraction.[41]