Page images

which made it desirable to check the results of observations, the difficulties being principally due to the impurity of the substances employed although subjected to repeated fractional distillation. Benzene constitutes an exception, as it may be purified by fractional solidification. The results of combustion of the successive homologues show such inconstancy of difference that it can only be accounted for by impurity of the hydrocarbon employed. The results are, however, capable of being checked by comparing them with those obtained by calculation from three distinct methods, namely:-(1) Starting from benzene and employing the formula x = H+CH, +156000 cal.; (2) from the mean values obtained for the liquid phenols from the formula CεH = CH, OH + 53600 cal.; and (3) from the mean value of the phenol ethers according to the formula C2H, C,H,O·CH — 121700 cal. The mean results obtained from these calculations agree within 0.6 per cent. with those obtained by experiment.



[blocks in formation]

779.5 cal.

[ocr errors]
[ocr errors]

1091 4


[ocr errors]
[ocr errors]



[ocr errors]
[ocr errors]



[ocr errors]
[ocr errors][merged small]

Heat Equivalent of Ethers of the Phenol Series. By F. STOHMANN, P. RODATZ, and W. HERZBERG (J. pr. Chem. [2], 35, 22-39).-1. Phenol Ethers.-a. Anisoil, OPhMe. This was rectified by a distillation between 152° and 153°, the pressure being about 755 mm. Five determinations were made of the heat of combustion, giving mean values of 8345 cal. per gram, or 901282 cal. per molecular equivalent. b. Phenetoïl, OPhEt, rectified between 166.5° and 167.5°. A mean of nine determinations gave the heat of combustion as 8666 cal. per gram, or 1057225 per molecular equivalent. c. Phenyl propyl ether, OPhPra. This was made by distilling together normal propyl bromide with alcoholic solution of potassium phenoxide, separating the product subsequently from water, washing with aqueous potash, drying, and redistilling. The boiling point was then constant at 185.2 under normal pressure, and the ether so obtained gave 8922 cal. per gram, or 1213426 per molecular equivalent. Experiments were made on higher members of the series, but not with sufficiently concordant results for figures to be given.

2. Cresol Ethers. Two of these were examined, namely, metacresyl methyl ether, C,H-OMe, and paracresyl ethyl ether, C,HOEt, and were found to have heat-equivalents almost identical with their isomerides of the previous group, namely, 8666 and 8920 cal. respectively.

3. Xylenol Ethers.-Two of these were examined, namely, xylyl methyl ether (1 : 3) and xylyl ethyl ether (1: 4).

4. Two thymyl ethers were experimented on, and finally two ethers obtained by substitution of methyl for hydrogen in resorcinol and quinol.

Comparison of the results for this series of ethers shows that the

introduction of a methyl-group corresponds with an increase of 155900 cal., whether the methyl-group enters the phenyl- or the methyl-group. Also the heat of formation is found to rise uniformly by 7100 cal. for each successive homologue.

These results agree closely with those obtained from examination of allied groups, namely, the alcohols and the phenols; the former giving 156163 cal. the latter 156356 cal. as the equivalent of the methyl-group.

From these experiments, calculations were made on the heat value of the methoxyl-group in these ethers (= H + OCH1); thus, for liquid benzene 779530 cal., for phenyl methyl ether 901282, and for dimethylresorcinol 1022966 cal., these corresponding with successive additions of CH2O, of which the equivalent is 121718 cal.

The latter part of the paper is concerned with deductions from the foregoing experimental results. First it is observed that similarly constituted isomeric substances have practically the same heat-equivalent, for instance, resorcinol, catechol, and quinol; or the ortho-, meta-, and para-cresols. Secondly, it is seen that isomerides if belonging to different chemical groups have markedly different heat-equivalents; thus the heat-equivalent of phenyl methyl ether is 901282 cal., whilst that of any one of the metameric cresols is 882288 cal. Similar results were obtained for other metameric bodies.

Finally the heat of formation of the phenyl ethers was calculated by subtracting the sum of the heat-equivalent of one molecule of phenol and of alkyl alcohol from that of the ether. Calculated in this way, the values for the different homologous and isomeric ethers were found to be practically constant, and equal to -6720 cal. Slight differences for two of these ethers are accounted for by the difference falling well within the limits of experimental error, considering the nature of the calorimetric method employed.

The ethers of dihydroxybenzenes give similar results, the calculation, however, being somewhat complicated by the difference of physical condition of the reacting substancee and the product. Dimethylresorcinol gives -6824, and dimethylquinol -3412 cal. for

their heat of formation.

A. H. F.

Relation between the Critical Temperatures of Substances and their Thermal Expansions as Liquids. By T. E. THORPE and A. W. RÜCKER (Phil. Mag. [5], 21, 431-434), and by A. BARTOLI and E. STRACCIATI (ibid., 533-534).-Controversial papers in relation to Mendéleff's formula (compare Thorpe and Rücker, Trans., 1884, 135).

Specific Heats of the Vapours of Acetic Acid and Nitrogen Tetroxide. By R. THRELFALL (Phil. Mag. [5], 23, 223-224).The author contends that the values obtained by Berthelot and Ogier for the specific heats of vapours of acetic acid and nitrogen tetroxide (Abstr., 1883, 6) support the theory of the dissociation of these gases. He further compares the specific heats of these vapours where the temperature coefficient is large with those of nitrous oxide and carbonic anhydride where the temperature coefficient is small, and argues that the usual formulæ for these gases only expresses the con

stitution of the large majority of the molecules, whilst others of greater and less complexity also exist. With rise of temperature, dissociation proceeds, and an increased absorption of heat takes place. H. K. T.

Influence of Change of Condition from the Liquid to the Solid State on Vapour-pressure. By W. RAMSAY and S. YOUNG (Phil. Mag., 23, 61-68).-Fischer (Ann. Chim. Phys. [2], 28, 400) has stated that although he found the pressure of vapour in presence of water and in presence of ice identical at the melting-point of the latter, yet this was not the case with benzene. The authors have found that Fischer's statement, which is opposed to the second law of thermodynamics, is based on a wrong interpretation of his own results; for Fischer applied a formula of the form p = a + bt +ct, which is not adapted to express the relations of the temperatures and pressures of saturated vapours, instead of the formula devised by Biot and employed by Regnault, p = a + bat + cẞ', and, moreover, did not make use of his own results at low temperatures. On recalculating from Fischer's results by means of the latter formula the vapour-pressure of liquid and solid benzene at the melting point of the latter, it is evident that the values are identical. The authors have also redetermined these constants for benzene, and confirm Fischer's results as regards the liquid; but their results do not quite agree with Fischer's as regards the solid. By making use of Regnault and Schiff's determinations of the heat of volatilisation of benzene, Petersen and Widmann's and Fischer's number for the heat of fusion of solid benzene, Schiff's formula to express the specific heat of liquid benzene, and Fischer's determination of the specific heat of solid benzene, it is possible to bring a check to bear on the value of Fischer's and the authors' experiments. The result proves that the balance of evidence is in favour of the authors' determinations of the pressures of benzene vapour in contact with the solid.

W. R.

Nature of Liquids, as shown by a Study of the Thermal Properties of Stable and Dissociable Substances. By W. RAMSAY and S. YOUNG (Phil. Mag., 23, 129-138).-The authors refer to previous memoirs (Phil. Trans., i, 1884 and 1886, and Chem. Soc. Trans., 1886, 790), in which they have shown that whilst the density of the saturated vapours of stable substances, such as ethyl alcohol and ethyl ether, becomes normal at low temperatures and correspondingly low pressures, those of acetic acid, and, as is shown by results lately published by the Messrs. Natanson, also those of nitric peroxide, increase with fall of temperature. It is held by many chemists that gaseous molecules, in changing to liquid, form molecular groups of definite complexity, exercising cohesive attraction on each other; on the other hand, it is conceivable that the liquid condition is a purely physical one, and that a liquid consists of molecules similar in every respect to those of a gas, but, owing to their closer proximity, exhibiting only that form of attraction known as cohesion. The arguments which have led the authors to adopt the latter view as correct are, that it is difficult to conceive that the rise of density of the saturated vapour of acetic acid, both at high and at low temperatures, can be

due to the same cause, for at high temperatures the conditions are unfavourable to chemical combination, but owing to the necessarily high pressure, the molecules are in close proximity, and the substance exhibits a high vapour-density; whereas at low temperatures the conditions are favourable to chemical combination, whilst the molecules, owing to the corresponding low pressures, are very far apart; so that cohesive attraction is out of the question. With alcohol and ether, a rise of density does not accompany fall of temperature; but at high temperatures they exhibit that rise of density common to all substances, stable or dissociable. Hence it is argued that there is no combination of the nature termed "chemical," that is elective, in the case of stable liquids; each molecule exerts cohesive attraction on all within its influence, but does not single out any small number of other molecules with which to combine. The authors also call attention to the fact observed by them that while the vapour and liquid of a stable substance exist in a closed space in presence of each other, pressure remains absolutely constant through the widest possible changes of volume; but with dissociable substances the pressure rises somewhat as volume is decreased, thereby revealing the partial formation of more complex molecules during the act of condensation. It is shown that the thermodynamical formula suggested by Willard Gibbs to represent the dependence of dissociation on pressure and temperature, does not represent facts at high temperatures in the case of acetic acid. The conclusion at which the authors arrive is that the difference between liquids and gases is one of degree, not of kind; is quantitative, and not qualitative. W. R.

Apparatus for Determining Vapour-densities. By G. DYSON (Chem. News, 55, 88).-The apparatus described is a modification of Victor Meyer's apparatus, and is arranged, by means of a manometer attached to it, to read the pressure produced by the volatilisation of a known weight of substance in a space of known capacity.

D. A. L.

Thermodynamics and Chemistry. By H. LE CHATELIER (Bull. Soc. Chim., 46, 737–746).—The writer considers the conditions of equilibrium of a gaseous mixture, such as a mixture of hydrogen, oxygen, and water at temperatures and pressures such that the water is partially dissociated. If we denote by N the number of molecular weights of the various substances existing in the mixture, this number will be a measure of the degree of dissociation, and will depend on two other quantities only, namely, the pressure P and the temperature T. Between these three quantities, there must exist a relation of the form F(N,P,T) = 0, which is the law of equilibrium of the system under consideration. The form of this function cannot be directly determined, but if we suppose a surface to be described representing the nature of the function for any particular system, then the sections of this surface made by three planes perpendicular to each of the coordinate axes respectively, will give us three distinct curves. The one in which the temperature remains constant is an isothermal, that in which the pressure is constant is an isobar, and the third is a curve of equal dissociation. By applying the second law of thermodynamics

and assuming the correctness of Boyle's and Gay-Lussac's laws, the author obtains the differential equation to the curves of equal dissociation, and points out that the equation contains no coefficient depending on the proportions of the substances initially present in the


It also appears that the sign of the change of pressure consequent on elevation of temperature depends on the sign of the latent heat of reaction of the system.

In general, it is easy to calculate or study experimentally the equations to the isothermals and isobars; and when these are known, the form of the function can be determined. The author applies this method to several cases, including saturated vapours, liquids in presence of soluble gases, and the dissociation of gaseous mixtures. In many cases (for instance, the vapour-tension of mixed liquids), the forms of the isothermals are not known with sufficient accuracy to admit of calculation. The author criticises severely the manner in which the principles of thermodynamics have been applied by some writers to the study of chemical phenomena, and points out the uncertainty of the assumption made in W. Gibbs's mathematical theory, namely, that the entropy of a mixture of gases is equal to the sum of the entropies of its constituents. L. T. T.

Representation of the Connection between the Gaseous and Liquid States of Matter by Isopyknics. By S. v. WROBLEWSKI (Ann. Phys. Chem. [2], 29, 428-451).-All former researches on this subject were based on investigations of the isothermal, a curve which gives the relation between pressure and volume at different temperatures. Jamin (Abstr., 1884, 5) has recently substituted for volume its reciprocal density. The author proposes the following:-Suppose a mass of gas or liquid of any definite density; when the temperature is changed, the pressure must also be changed in order to keep the density constant. The curves which represent the relation between pressure and temperature for different densities are named isopylnics (from ἶσος πυκνός). If a system, of such curves be drawn for any homogeneous isotropic substance, these curves can nowhere intersect each other.

Such a system has been constructed by the author for carbonic acid by means of Sarrau's formula (Abstr., 1886, 203), which is taken as most nearly representing the experimental results. The curve corresponding to density d, of which the values of T (absolute temperature) are abscissæ, and of p (pressure in atmospheres) ordinates, is called for brevity the isopyknic d. A diagram is given showing the course of the isopyknics from d = 0.025 to d = 1.2 for temperatures up to T = 370° and p = 400 atmospheres; and on the same diagram is drawn the curve of minimum values of pv. For temperatures below 30°, this is of course identical with the curve of liquefaction. This portion has been calculated from the data of Regnault and of Pictet (Ann. Chim. Phys. [5], 13, 213); it is slightly conver towards the axis of temperature. The upper part of the curve, which is slightly concave towards the axis of T, has been calculated from the experiments of Amagat (ibid. [5], 22, 374) from 35° to 100°.

« PreviousContinue »