362

MOLECULAR HEAT OF COMPOunds.

[173.

These quotients are not uniform in amount, as would be the case if the atomic heats of all the elements were alike, and if the molecular heat of a compound were represented by the sum of the atomic heats of its constituent elements; but it may be easily seen that the number in the different members of the same class of compounds is nearly alike.

For example, assuming that the molecule of the chlorides of the alkaline metals is represented by the formula MCI (a diatomic formula), the molecular heat of this class, 12.88, divided by 2, is 6'44, which agrees very well with the ordinary number for the atomic heat of the elements. Again, if the molecule of the chlorides of the metals of the alkaline earths, and most of the strongly basic metals, iron, cobalt, zinc, nickel, &c., be represented by the formula N'Cl, (a triatomic formula), the molecular heat 1865, divided by 3, gives 6:22 as the result, again agreeing with the ordinary numbers for the atomic heat of the elements. Similar remarks are applicable in the case of the bromides and the iodides of these two classes of metallic elements, as will be obvious on inspecting the table opposite.

This correspondence between these two sets of chlorides, bromides, and iodides, however, disappears, if it be supposed, as till recently was generally admitted, that the chlorides, bromides, and iodides are all diatomic. If the chlorides of the second class represented in the table as consisting each of 3 atoms-such as chlorides of calcium, barium, magnesium, &c., be supposed to be formed upon the type MCI, or to contain two atoms only in their molecules their molecular heat will be 9:32, and dividing by 2, the number will be 4'66.* The general conclusion deducible from these experiments is that, whilst the equivalents of the halogens and of the alkaline metals, including also thallium and silver, are truly their atomic weights, the equivalents of the

*This may, perhaps, be rendered more clear by an example:-The specific heat of baric chloride is, o'08957; if it be represented by the formula BaCl, with an atomic weight 104'0, its atomic heat will be 0.08957 × 104 = 9'315; whereas if it be represented as Ba"Cl,, with an atomic weight 208, its atomic heat will be just double, or 18.63. In the first case, the molecule of the salt is supposed to be diatomic, or to contain 2 atoms of its constituents; in the second it is represented as triatomic, or as containing 3 atoms. The quotient obtained by dividing the atomic heat by the number of atoms in the molecule will of course be different in the two cases :

*N" representing a dyad, R"" a triad, Xiv a tetrad, and Yvi a hexad element.

Kopp explains the difference from one another in the quotients given in the last column of the foregoing table, by the fact that the atomic heats of many of the non-metallic elements are not identical with those of the metals, and he still maintains that the molecular heat of a compound is always really the sum of the atomic heats of its constituent elements. From his own experiments he concludes that the number which represents the atomic heat of sulphur and of phosphorus is 5'4, or a little lower than the values assigned to them by Regnault, and this number, he

364

SPECIFIC HEAT OF ORGANIC SOLIDS.

[173.

states, is confirmed by deducting from the experimental number obtained as the molecular heat of the sulphides and phosphides, the atomic heat of the metals which enter into their formation. By a similar process of calculation when applied to the oxides, he infers that the atomic heat of oxygen is 4, that of fluorine 5, and that of hydrogen 2:3; whilst he adopts 18 for that of carbon, from Regnault's experiments upon the diamond.

Amongst the most severe tests of this view are the results obtained from the atomic heats of organic compounds, some of which are given in the following table :

With the exception of the first two on the list, together with those of potassic ferrocyanide, cane sugar, and mannite, the results, calculated from the numbers assumed by Kopp as the specific heats of oxygen, hydrogen, and carbon, accord with those obtained by experiment very satisfactorily.

The specific heat of organic liquids has been studied as yet but imperfectly. It is, however, clear that the specific heat of liquids rises very rapidly with rise of temperature, and this rise stands in no simple relation to the amount of expansion which the liquid experiences. It is not therefore surprising that in the case of liquids, even when elementary, no approximation to the law of Dulong and Petit for the elements in their solid form (that the specific heat is inversely as the atomic weight) has been ascertained to exist. The specific heats of a few liquids, and

their variations at different temperatures, have already been given on Regnault's authority (p. 328). In the above table, the numbers for ether, ethal, and oil of turpentin were determined by Favre and Silbermann. The remaining results are those obtained by Kopp (Pogg. Ann. 1848, lxxv. 98).

From these experiments, scanty and few in number though they are, it appears that the specific heat of equal masses of organic liquids decreases as the molecular weight of the substance increases; further, that when the products obtained by multiplying the specific heat into the molecular numbers are compared, this product generally increases as the molecular weight increases, and in the homologous series of the alcohols, the fatty acids, and the ethers, the increase is about 7 for each addition of CH, in the molecule of the compound. In the case of some metameric bodies, such as methyl acetate and ethyl formiate, the numbers for their atomic heats are alike; and ether, which has the same molecular weight as the compounds just mentioned, has the same atomic heat, though it is not metameric with them.

Latent Heat.

(174) Disappearance of Heat during Liquefaction.-When matter passes from the solid into the liquid state, or from the

366

LATENT HEAT OF LIQUIDS.

[174

liquid into the aëriform state, heat in large quantity disappears, and ceases for the time to affect the thermometer; hence, this modification of heat is called latent heat. For example, when

a lump of ice at o° C. is brought into a warm room, it gradually thaws and is converted into water; but neither the ice, nor the water in contact with it rises in temperature. So long as any portion of the ice remains unmelted, the water continues to indicate the temperature of o°, as does also the ice. Again, a pound of water at 100° C. mixed with a pound of water at o°, gives two pounds of water at 50°, which is the mean temperature; but a pound of ice at o° mixed with a pound of water at 100°, gives two pounds of water, of which the temperature is only 10°5.

In this case the water has lost 89°5, whilst the ice has gained only 10°5; so that 79° have disappeared, or have become latent. Hence, in order to convert a pound of ice at o° into water at o, heat sufficient to raise 79 lb. of water from o° to 1° C. is needed; that is, 79 units of heat. This heat, however, is not lost, for if the progressive cooling of water be observed in an atmosphere many degrees below the freezing-point, it will be found that the temperature of the liquid sinks regularly until it reaches o°, when it becomes stationary, and freezing begins; the heat being supplied from that which is latent in the water. As soon as the whole has become solid, the thermometer again shows that the temperature of the mass sinks, until at length it reaches that of the surrounding air. Some idea of the quantity of heat that is required to convert ice into water, without any apparent rise in temperature, may be formed from the fact that the simple conversion of a cube of ice three feet in the side into water, also at o, would absorb the whole heat emitted during the combustion of a bushel of coal. (Faraday.) Pouillet has calculated that the whole of the heat of the sun's rays which fall upon the surface of the earth in the course of twelve months, would be expended in melting a layer of ice which covered the entire surface of the globe for a thickness of 1013 feet (30.89 metres).

This 1: rge amount of heat latent in water, which is given forth as it freezes, furnishes a source of heat of the greatest value in mitigating the severity of any sudden setting in of frost, as the very act of freezing moderates the effect of the depression of temperature on surrounding objects, and renders the transition from heat to cold, and of course the converse gradual and uniform. Another ve by this gradual liquefaction of

accumulated during a long w