Once more, the very small values appear, and, as we might expect, the constant of the tertiary or allo-carboxylic acid is less than that of the secondary or ortho-carboxylic acid. The constant s of the isocamphoric acids are still less than those of the camphoric acids proper, in accordance with the fact that the former have the cis-configuration, and the latter the cistrans-configuration. Thus, for l-isocamphoric acid we have the constant 0.00160, and for allo-hydrogen ortho-ethyl camphorate the constant 0.00065. We conclude then from these measurements that the low value of the constant for camphoric acid need not be due to the carboxyl groups being far apart in the molecule, but to the fact that all but one of the hydrogen atoms of the succinic acid, from which, according to the Perkin-Bouveault formula, camphoric acid is derived, are replaced by hydrocarbon groups, which exert their primary effect in lowering the dissociation constant. The author desires to express his indebtedness for valuable criticism and suggestions to Professor W. H. Perkin, who kindly revised this paper in manuscript. UNIVERSITY COLLEGE, DUNDEE. XXXIX. The Maximum Pressure of Naphthalene Vapour. By RICHARD WILLIAM ALLEN, M.A., University College, THE formation of naphthalene during the carbonisation of coal in The absence of decimals (other than the 5) suggests that they are only approximations, and they were, in fact, determined in the ordinary barometric vacuum without any special precautions; the tube was jacketed for a few centimetres only below the top of the mercury column, and no method of direct comparison with a standard barometer was employed. Moreover, the value 2 mm. for the vapour pressure at 15° is improbable, for any substance having that vapour pressure at 15° would soon disappear if freely exposed to the air, whereas a small piece of naphthalene so treated shows little diminution in size, even after many weeks. That Naumann was satisfied with an approximation is not to be wondered at, since, except from the point of view of the coal gas industry, a knowledge of the maximum pressure of naphthalene vapour does not appear to be of much importance. Ramsay and Young (Phil. Trans., 1884, Part I, xxxvii, 461) applied the apparatus they devised for measuring the influence of pressure on the temperature of volatilisation of solids to the measurement of the vapour pressures of benzene, water, acetic acid, and camphor, but the preliminary observations they made with naphthalene led to no result, and were abandoned owing to the low vapour pressure of the hydrocarbon (9 mm.) at its melting point (97.6°), and to the want of data regarding its vapour pressure at low temperatures; it may therefore be assumed that the method is inapplicable to naphthalene at the ordinary temperature. The vapour pressure of naphthalene at the ordinary temperature being very small, the ordinary method of measurement in the barometric vacuum was at first deemed unsuitable. Among other methods, that of evaporation, which has been occasionally employed in the case of liquids, seemed to offer decided advantages. In it, as is well known, a given volume of an indifferent gas at a definite temperature is passed over a weighed quantity of the substance whose vapour pressure is to be measured, and the loss in weight of the latter estimated; this loss represents the weight, and hence the volume, of the vapour required to saturate the given volume of the gas at the given temperature and pressure, and a simple calculation yields the value of the vapour pressure at that temperature. The method is based on the assumption which was proved to be legitimate by the experiments of Regnault and Magnus, that the vapour pressure of a volatile liquid in a space filled with an indifferent gas is very nearly equal to the vapour pressure in a vacuum, and the more nearly is it so the smaller the pressure of the vapour. J. Walker (Zeit. physikal. Chem., 1888, 2, 602) measured the relative vapour pressures of aqueous solutions of various inorganic salts, and Wall and Bredig (Ber., 1889, 22, 1084) those of various alcoholic solutions by this method and obtained very good results. The process being unlimited as regards the length of time for which the gas is passed, the slowness of the operation, and the volume of the gas employed, it seemed capable of great delicacy and exactness, and hence especially well adapted for measuring the low vapour pressures of such solids as naphthalene. Some pure naphthalene was accordingly first prepared by digesting a commercial sample alternately with sulphuric acid and caustic potash several times and washing repeatedly with water between each digestion; in this way, most of the phenols and hydrocarbons were removed. The purified sample was then three times distilled with steam; finally, it was drained free from water, melted, powdered, and set to dry for several weeks in a large bell jar over concentrated sulphuric acid. It then melted at 79-6° and gave only a slight pink coloration on boiling with strong sulphuric acid; hence it might be considered pure. This dry, powdered naphthalene was then filled into light U-tubes, above the powder a little cotton wool was placed to prevent any of it being carried over mechanically, and dry corks, provided with glass outlet tubes, were fitted into the tubes and rendered air FIG. 1. gas To Aspirator G tight with a coating of cement. In order that the temperature of the experiment might be kept perfectly under control, a large zinc water-bath three feet long, two feet six inches high, and two feet three inches wide was constructed. It had six air compartments running lengthwise through it, arranged in two tiers of three each, each compartment being six inches high and four inches wide. The water could circulate freely round each compartment, and was kept well stirred by a large iron stirrer working up and down and driven by a water engine at any desired speed. The temperature of the water was controlled by means of a mercury thermostat, to which was attached a non-atmospheric burner capable of very delicate adjustment. The bath was fitted with a close lid which had three small holes, one in the centre for the rod of the stirrer, and two close together at one end and in the middle line, one being for the thermostat, the other for the thermometers, which were all divided to tenths of a degree. By these means, the temperature could be kept constant throughout the bath to 0.1° for 24 hours. Fig. 1 shows the arrangement for one compartment: the gas, entering through the wash bottle into the safety tube, A, inserted to measure the pressure in the apparatus and prevent it exceeding that of the atmosphere, was partially dried by passage up a cylinder, B, filled with beads moistened with concentrated sulphuric acid. After being further dried and raised to the temperature of the bath by passage through the sulphuric acid bulb, C, and the calcium chloride tube, D, the gas was passed through the weighed tube of naphthalene, E. In the succeeding sulphuric acid tube, F, most of the naphthalene carried over was re-absorbed and thus prevented from condensing in the tubes outside the bath and blocking them up. The tube G is connected with an aspirator by means of which the gas is drawn through the system of tubes. The several aspirators had slightly different capacities, but as each experiment was concluded the pressure and temperature of the atmosphere were observed, and due correction being also applied for the water vapour saturating the gas in the aspirator, the weight of naphthalene carried over in the given time by 36 litres of dry gas at 15° and 760 mm. pressure was calculated from the observed values. Thus the figures given in the third column of Tables I, II, III, and IV are strictly comparable. The ends of each air compartment were carefully closed, as shown in Fig. 1, by stoppers made of triple layers of wood, each layer being covered by flannel and separated with cork from the adjoining one. In this way, air currents were prevented, and the temperature in a compartment ensured as being that of the water of the bath. The temperature of the bath having been raised in the first place exactly to 30°, and the various tubes arranged as described, the gas was passed at various speeds through the weighed tubes of naphthalene, and the loss in weight of the tubes observed; from this the weight of naphthalene required to saturate 36 litres of dry gas measured at 15° and 760 mm. was calculated. It remains to be mentioned that, in addition to air, hydrogen and coal gas were employed in these experiments, as it was deemed better not to rely on results obtained by one gas only when these gases were employed, the U-tubes of naphthalene were always swept clear of them before weighing by blowing a small quantity (about 50 c.c.) of dry air through them; trial showed that the extra loss of naphthalene due to this precaution was quite negligible. In Table I (p. 403) are given the observed results. It is noticeable that the weight of naphthalene carried over increases slightly as the rate at which the gas is passed over diminishes. The nature, however, of the gas into which the naphthalene diffuses seems to exercise no influence on the rapidity of the diffusion; the values in Tables II and III confirm this. From the above figures, it would appear that practically complete saturation was attained in the slowest experiments; owing, however, to the obvious difficulty of keeping a large bulk of water at a constant temperature much higher than that of the surrounding air for days at a time, and inasmuch as 97 per cent. saturation was obtained in the 24 and 30 hours experiments, it was decided to limit future experiments to about 24 hours, and allow an extra 3 per cent. on the observed values in order to obtain the true saturation or "limit" value. Tables II, III, and IV contain the results of experiments conducted at other temperatures. As previously, due corrections have been made in each experiment for temperature and pressure, and for the hygrometric state of the gas in the aspirators, the values given being those for 36 litres of dry air at 15° and 760 mm. pressure. |