With the object of testing experimentally the third law of thermodynamics as applied to crystalline solutions, the chief thermodynamic properties of such a solution of silver bromide and silver chloride have been studied. Measurements described and recorded include: the free energy of formation, obtained from measurements of the e.m.f. of appropriate silver-silver halide electrodes against hydrogen; the heat of formation, from measurements of heat absorbed in dissolving the various phases in a thiosulfate solution; the specific heats between 15°K and 298°K of the pure constituents and the solid solution. The entropy of silver bromide and of silver chloride is calculated from the specific heat curves. The entropy change in the formation of the solid solution at 298°K is calculated from the experimental data and is found to agree with the result calculated statistically. The latter is identical in form and magnitude with the entropy as calculated thermodynamically for an ideal solution, though the heat content and free energy show that the solution is far from perfect. From the specific heat curves it appears that the entropy of mixing persists essentially unchanged in magnitude to the lowest temperature of the measurements, and presumably to the absolute zero. The existence of finite entropies, or entropy differences, at the absolute zero is regarded as established thermodynamically, and the requisite interpretation of statistical ideas to secure concordance with this result is discussed. Finally, although the existence of a valid but limited third law is accepted, and a concise general statement of it suggested, it is pointed out that the ``principle of the unattainability of the absolute zero'' must be regarded as extra-thermodynamic in character.

No abstract

The heat capacity of primary normal amyl ammonium chloride was measu red between 20° and 280°K. If the substance is quenched to 90°K a metastable form is obtained which upon heat ing above 165°K slowly changes toward an apparently st able form. Two regions of gradual transition were found. X-ray powder diffraction photographs taken above and below the transition regions were remarkably similar. The experimental results are interpreted on the basis of rotation of the NH3C5H11+ groups in the crystals.

No abstract

Huffman, Parks and Daniels between 90 and 300°K. well within the limits of accuracy claimed by them (1%). Furthermore, there is no tendency for one curve to be consistently higher or lower than the other up to 250°K. It should be pointed out while Huffman, Parks and Daniels used a "vacuum" calorimeter, their method of operation was entirely different and it seems probable that their corrections for radiation lead to the divergence at the higher temperatures.It is interesting to note that the value Huffman, Parks and Daniels obtained by extrapolation from 90 to 0°K. for the entropy at 90°K. was 12.69 e. u. while the value found here is 12.52'e. u. with an extrapolation of only 0.43 e. u. from 15 to 0°K.Acknowledgment.-The authors are indebted to Dr. R. T. Milner for his helpful suggestions in the construction of the apparatus and for his assistance in measurements on the empty calorimeter. They are indebted to Mr. J. W. Cook for liquefying the hydrogen and to Mr. R. A. Nelson for assisting in the measurements and calculations on naphthalene. Summary 1. An improved adiabatic calorimeter for the measurement of specific heats of liquids and solids between 14 and 300°K. has been constructed. It can be used with a precision of about 0.1% and is particularly suited to the study of slow transitions and thermal changes.2. The molal heat capacity of naphthalene was determined between 14 and 300°K. and the entropy and free energy of formation at 298.16°K. calculated to be 39.89 =*= 0.12 e. u. and +48.5 kg. cal., respectively.