422 HEAT EMITTED DURING COMBINATION DEFINITE. [199. In reality, however, it is not so; for, independently of the difficulties which the exact admeasurement of heat always involves, there are others which will be rendered evident by considerations such as the following. Scarcely any molecular change can take place without either evolution or absorption of heat. When a gas or a vapour becomes liquefied or solidified, the change of state is always attended by the evolution of the heat which it previously contained in a latent state (177), and the effect is reversed when a solid passes into the liquid or the aëriform condition, heat being then absorbed (174, 178). Now, the instances in which chemical combination takes place without any alteration in the physical condition of bodies are rare, and the cases in which the product occupies exactly the same volume as the bodies from which it was formed, are still more so. When two gaseous elements, like chlorine and hydrogen, unite and form a compound which is not only gaseous but which occupies the same volume as the bodies did before their combination, the problem is presented in the simplest form: the heat observed in such a case is due solely to the chemical action; but when the products, though gaseous, occupy a smaller volume after they have entered into combination,-as when 2 volumes of carbonic oxide unite with 1 volume of oxygen, and form but 2 volumes of carbonic anhydride, the heat emitted during the act of combination is due partly to chemical action, and partly also to the condensation which the gases have experienced. When the product assumes the liquid state, as occurs in the formation of water during the combustion of hydrogen in oxygen, the quantity of heat emitted owing to this change of state is still more considerable. When, on the other hand, the solid passes into the aëriform state, as when carbon is converted into carbonic anhydride, the heat actually observed is less than that which the combination ought really to produce and the effect is reversed when the solid state is assumed by the product, as when phosphorus becomes oxidized to phosphoric anhydride; in which case the heat evolved exceeds that really due to the act of combination. But even when no change of state is observed, minor disturbing causes are at work. Supposing it were possible to obtain a direct combination of iodine with a metal, such as iron or zinc; even then, though two solids united to form a third solid, it would not necessarily happen that the whole of the heat emitted was due to the chemical action. If zincic iodide, for example, contracted in the act of combination, a small portion of the heat observed would be due to that evolved by the solid in consequence 201.] EXPERIMENTS ON HEAT OF COMBINATION. 423 of its change of volume; whereas, if the iodide occupied a larger space after combination than that of the two elements separately, the heat observed would be less than that resulting from the chemical action and even if no change of volume occurred, it might happen that the compound had a specific heat different from that of the original elements, and in such case a slight elevation or depression of temperature might be occasioned, which was not really the chemical result of the act of combination. From the foregoing statement it is therefore clear that the experimental numbers, however carefully the observations are made, can very rarely yield the actual quantities of heat due to chemical actions: they are compound results from which the true calorific equivalents of the different elements (or heat evolved by the combination of chemical equivalents of the different elements) must be deduced by other means. (200) Early Experiments.-The importance of determining accurately the amount of heat arising from chemical action was first distinctly announced by Lavoisier, who instituted a series of experiments with the view of finding the quantity of heat evolved during the combustion of various substances: his method consisted in ascertaining the quantity of ice that was melted, when known quantities of these bodies were burned in his calorimeter. The first experiments with any claim to accuracy, however, are those of Dulong, which have formed the foundation for all subsequent researches upon the subject. Important additional investigations have since been made, particularly by Despretz, and more recently by Andrews, by Favre and Silbermann, by Berthelot, and by Thomsen. (201) Researches of Andrews.The apparatus employed by Andrews in these experiments (Phil. Mag. 1848 [3], xxxii. 321, 426) was of a simple kind. When the substances to be combined were in the gaseous state, and the products of combustion were also gaseous, the two gases were mixed in the proper proportions, as in the performance of eudiometric experiments, and intro FIG. 156. duced into a vessel of thin sheet copper (a, fig. 156), of a capacity of about 23 cubic inches (380 cubic centim.). It was closed by a 424 EXPERIMENTS OF ANDREWS. [201. screw, the head of which was perforated to admit a cork, through which a silver wire, b, passed; this wire was connected by a thin platinum wire within the vessel to a second silver wire soldered to the screw itself. The platinum wire could be ignited by connecting it for an instant with a small voltaic battery, and thus the gaseous mixture could be detonated at the pleasure of the • operator. The copper vessel containing the mixed gases was then introduced into a larger vessel, c, which was filled up with a known quantity of water: the vessel, c, was suspended in a cylinder, d, provided with a moveable cover, and the whole was enclosed in an outward cylindrical vessel, c, arranged so as to admit of its being made to rotate upon its shorter axis. The apparatus having been mounted, was caused to rotate, in order to bring every part to a uniform temperature: after which the exact amount of this initial temperature was read off by a very sensitive thermometer, capable of indicating differences of temperature of, of a degree Centigrade. The thermometer was then withdrawn, and the gases exploded by igniting the fine platinum wire; the outer vessel of water was closed by a cork, and the apparatus was caused to rotate for thirty-five seconds, in order to establish an equilibrium of temperature in all its parts. The thermometer was again introduced, and the rise of temperature was ascertained. After this observation the apparatus was again made to rotate for thirty-five seconds, and the loss of heat thus occasioned was ascertained. This third rotation was necessary in order to determine the cooling effect of the atmosphere upon the apparatus, during the time that the experiment lasted; in these cases it seldom amounted to more than of the total quantity of heat set free. When solid bodies were burned in oxygen, the form of the apparatus was modified; the combustion was effected in a copper vessel (a, fig. 157), of about 4 litres, or 250 cubic inches in capacity, which was filled with oxygen: and a known quantity of the combustible was supported in a small platinum dish, b; when all was ready, the vessel a having been accurately closed, the combustible was ignited by means of a voltaic current sent through a very fine platinum wire in connexion with the insulated wire f. Previously to this ignition, the vessel a was immersed in a large cylinder c, filled with a known quantity of water, and the whole was surrounded by an outer vessel of tin-plate d, to prevent radiation. The copper vessel could be agitated within the vessel of water by means of the lever e. Particular expedients were required in certain cases to ensure the ignition: for instance, in 202.] EXPERIMENTS OF FAVRE AND SILBERMANN. 425 burning zinc filings and other metals in oxygen, a minute portion of phosphorus was employed to kindle the metal; the weight of this piece of phosphorus being known, the heat which it emitted. was calculated, and deducted from that observed. In some cases these experiments lasted fifteen or sixteen minutes, so that the correction for the cooling effect of the external air acquired considerable importance. When chlorine was used instead of oxygen, it was not necessary to ignite the substance under trial: but in order to prevent the spontaneous ignition of the body, the latter was enclosed in a thin glass bulb, which was broken by agitation of the apparatus at the moment that everything was prepared. The chlorine itself was in most instances contained in a glass vessel, which was filled with the gas by displacement; an excess of the body for combination with the chlorine was always employed, so as to ensure the total absorption of the chlorine. The time allowed for absorption was in each case six minutes and a half. (202) Experiments of Fuvre and Silbermann.-A very extensive series of researches upon the development of heat during molecular and chemical changes were undertaken by Favre and Silber |