WHATMAN Grade FILTER PAPERS Note a few of our later specialities: No. 42-"ASHLESS." Double Acid washed and extremely close of texture. For use with No. 50-Hardened by treatment with Nitric Acid. Very tough, it will resist great pressure, and SOLE MANUFACTURERS : W. & R. BALSTON, LTD., MEDWAY MILL, MAIDSTONE, KENT. FIND OUT ALL ABOUT OUR COMPLETE RANGE FROM OUR REVISED BOOKLET AND FRICE LIST. YOUR LABORATORY FURNISHER CAN SUPPLY COPIES In case of difficulty in obtaining Free Samples, write the Sole Mill RepresentativesH. REEVE ANGEL & CO., 9, BRIDEWELL PLACE, LONDON, E.C. 4. Phosphoric Acid 1500 WATER WHITE Acid Calcium Phosphate MANUFACTURERS: PRESCOTT & CO., HULME, MANCHESTER. Wires: "Corncrake," Manchester. Telephone: City 5470. THE CHEMICAL NEWS. VOLUME CXXI. EDITED BY JAMES H. GARDINER, F.C.S. No. 3142.-JULY 2, 1920. EFFECT OF PRESSURE ON GAS ADSORPTION. WORK was only done at pressures below atmospheric. Six kinds of charcoal in various states of division were studied in a manner similar to that of Macbain (Phil. Mag., 1909, xviii., 816). The charcoals, previously heated to expel air, and in equal volumes (50 cc.), were saturated with dry ammonia at atmospheric pressure and room temperature. They were then placed, in turn, in a dry glass vessel closed with a rubber stopper carrying a manometer and aspirator tube. The apparatus had been previously tested and proved gas-tight. Capacity of vessel, 400 cc. A good water pump was used as aspirator, and as the necessary pressure was reached the aspirator tube was closed, using a screw clip. The pressure was noted at intervals, temperature being kept constant by having the vessel immersed in jar of cold water surrounded by nonconducting material. Experiments were done with each charcoal at two or three different pressures. Occasional blank experiments were done as a control. The pressure in all cases at first increased and ultimately decreased, but this behaviour varied with the charcoal used. The first effect of aspiration would be to remove the gas on the charcoal surface. The increase in pressure would therefore be due to escape of gas from the interior, subsequent decrease being due to renewed surface condensation. In no case was the initial pressure regained, but the results give a good index as to the inner structure of the charcoals used. Thus, where rapid pressure increase is noted, a complex internal structure may be surmised cocoanut, fruit stone, &c. In the case of birch dust its gas-adsorbing capacity depends mainly on surface effect. The palm nut charcoal used was in large pieces, so that most gas it retained would be by absorption. Other Details. I. Apparent density determined by weighing several definite volumes of charcoal and obtaining average mass per cc. Real density determined by specific gravity. Bottle method. In the following results apparent density is given first: Almond Birch chips Fruit stone 2. Carbon content determined by combustion and esti- | STRUCTURE OF MATTER AND THE QUANTUM mation as CO2. The content was generally about 85 per cent. 3. Moisture content of a charcoal was increased by addition of lime, common salt, or potassium carbonate. Unsatisfactory results were obtained as the mass tended to become sticky and unworkable. With phosgene, COCl2, the moisture would be advantageous, as decomposition into CO2 and HCI would be assisted. Further, the water would serve as a reservoir for the HCl. This gas was not studied as it could not be got in sufficient quantities; | also, laboratory class work could not be interfered with. 4. Porosity of a mass of charcoal dust was increased by mixing with solid ammonium bichromate and heating out of air. The efficiency against gases was increased per cc. of actual charcoal, but per cc. of wbole product efficiency was low. 5. Decolorising power was tested, using caramel and indigo solutions, also highly discoloured crude glycerin. Birch charcoal dust was found a remarkably good decoloriser, but subsequent filtration was slow. Coarse charcoals do best at high temperatures, when penetration effect is considerable. Heating immediately before use drives from the charcoal any contained air or hydrocarbons, thus increasing the efficiency. 6. As gas adsorbents the denser charcoals were most efficient. The order of efficiency was found to be-palm nut, cocoanut, fruitstone. The values obtained by Hunter were not even approximately obtained. It is very probable that he used a highly specialised form of charcoal. State of division is an important factor. Fine dusts offer a large resistance to a gas and in large quantities there would be some difficulty in their use. Granules of about 1 mm. diameter were found suitable. 7. Selective adsorption.--Iodine residues were treated with chlorine water and iodine liberated removed by warming the solution with charcoal dust. The charcoal then filtered off and heated in an iron retort. The iodine sublimed into the neck of the retort and removed. The charcoal remaining was of little further use. In one case the charcoal was extracted with hot alcohol, but the process was slow and required a large volume of alcohol. The alcohol was subsequently distilled off from a water-bath, the iodine remaining. Both methods were found to be somewhat tedious in practice but fairly efficient as regards iodine yield. Iodine in iodates could be liberated by using bisulphite solutions after chlorine had liberated iodine from iodides and such iodine bad been removed as above. Later, selective adsorption will be studied as a means of removing alkaloids from alkaline extracts of vegetable tissues. 8. Adsorption of iodine from its solution in benzene and of benzoic acid from benzene was studied. The iodine was removed to a small extent comparatively, 17 per cent. Since iodine in KI solution exists as the ion 13, and is removed to the extent of 80 per cent under the same conditions of temperature, &c., it is probable that iodine in benzene exists in a simpler form. Benzoic acid was removed to the extent of 80 per cent; this points to benzoic acid molecules being associated in benzene. It was noticed throughout the preceding work that the more complex in structure a substance may be the more readily is it adsorbed by charcoal. This fact should be useful in the case of alkaloid extraction. PHYSICAL APPARATUS FOR CANADIAN UNIVERSITY.A report has been received in the Department of Overseas Trade from His Majesty's Trade Commissioner at Toronto (Mr. F. W. Field) to the effect that a professor of a Canadian university desires to receive catalogues from United Kingdom manufacturers of physical apparatus for university laboratories for his guidance in placing orders. The name and address of the professor referred to will be furnished by the Department to any United Kingdom manufacturers interested, on application. THEORY. SUPPLEMENTARY NOTE TO PART IV. By F. H. LORING. UPON further study I find that the scheme (CHEMICAL NEWS, cxx., p. 183) can be made quite regular up to and including nickel. The following table gives the additional values :— CI 32:06 32'07 35'50 35:46 39.88 39'9 39 07 39°10 40:06 40'09 43'50? 44'I Ti 48.22 48.1 The 1920 atomic-weights are given alongside the ones deducible by means of the scheme. The significance of Note 1 of Part IV. is now made more apparent. The order of the elements in the above table is necessitated by the scheme, and this agrees with that given by the atomic numbers. Argon resembles beryllium in its composition, having apparently three outriders Cobalt and nickel have isotopes of mass 59, the others being 57 and 58 respectively. It looks as if the iron isotope of mass 56 is responsible for the high magnetisability of iron. The regularity of odd and even valencies associated with odd- and even-number isotopes seems to break down at nickel and cobalt. The calculated atomic-weights of nickel and iron when averaged become equal to the average of the experimental values, suggesting thereby a certain amount of perhaps unavoidable contamination in the experimental determinations. From a study of the values obtained by the method of positive rays and these relations, I have come to the conclusion that further experimental work on isotopes may reveal the fact that, under certain conditions of extreme dilution, it will be possible to get "mass spectra " of many metallic elements. Moreover, that some of the atoms may be broken up so that the values recorded are not always representative of those functioning in ordinary chemical actions, or in ordinary physical processes. There is also the possibility of a small percentage of contamination due to what might be termed broken or disintegrated atoms-not necessarily due to radio-activity. This may account for the masses obtained by chemical methods deviating from the theoretical values; but, on the whole, the values seem very concordant: probably they are, for the most part, within the range of experimental error. There is a regularity in the extended table which points to the possible modification of the nitrogen outrider, in that it may consist of two a-parts taken together instead of one more 66 compact" b-part; but as Sir E. Rutherford lays stress on the disrupted member of nitrogen having possibly a mass of 2 (see CHEMICAL NEWS, 1919, cxviii., p. 311), such evidence would appear possibly to contradict experiment. I am, however, aware that further experiments have been made which may clear up the difficulty. In the absence of full details of such experiments, this point cannot be discussed from the experimental side. An attempt to explain the non-isotopic characteristic of C, N, O, and F has already been made (see CHEMICAL NEWS, 1915, cxi., p. 157), but there may be a further peculiarity as a contributory cause. These particula elements may be made up of sub-atoms of masses indicated, minutes, at the end of which time it is carefully removed by n in the table on p. 183, and the number (") may be 4 and the pressed cake weighed, any camphor adhering to in each case. the lint being carefully brushed off and added to the cake. From the loss in weight the amount of water plus oil expressed is found. EQUATION OF STATE. By FRED. G. EDWARDS. THE equation of molecular heat, namely, k = mx log T, may be converted into the equation of state b(k + a log 0) = mx log T, where @ is the number of degrees below the critical temperature, a is an infinitesimal coefficient producing the relative values of the intrinsic pressures at the critical and observation temperatures respectively, and b is a constant making the sum in brackets equal to the intrinsic pressure, the surface tension, or the vo ume density, with any given units. The values of y obtained from m=5'95/y and m = 4 - (2y/2·95) will give the specific heats at variable volume, while m will be integral at the natural temperature of change of phase and recalescence. The specific density ratio D- Amx log T, and the coefficient of linear expansion d.V/3d.Tid.1/Am log T = T/3Amx. = The atomic volume ratio V = 1/m log T is a more general form of the equation of Lothar Meyer's curve V=A/D. The tangent to the curve d.V/d.TT/m, and the minimum values d.2V/d.T2 = m−x = 0. Also, d.D/d.T Amx/T. The datum value of k will vary for each change of phase with the intrinsic pressure. The values of x are o·0698A! generally, for certain gaseous substances 0.76384A, and for hydrogen, unity. It will be found that Groups V. and VI. of the Periodic Table form one homogeneous group as required for symmetrical atomic shape built up from a primordial mass unit, which may be nebulium. The coefficients of these equations can only be determined by a very extended analysis of the physical state of liquid and solid substances; but inspection shows that the terms are of the right order of magnitude, and, moreover, they are derived from an equation showing exact agreement with the specific heat at constant volume, and there is a distinct parallelism between the coefficient of expansion and the specific heat at variable volume. THE EXAMINATION OF CHINESE CRUDE CAMPHOR. IN the examination of crude camphor, the estimations usually required are those of non-volatile matter (or dirt), moisture, and oil, and the sum of these impurities sub. tracted from 100 per cent is supposed to represent the camphor present. The moisture may be conveniently estimated by the calcium carbide method, allowing three hours for the evolution of gas, and the dirt by the residue left after volatilising a weighed portion of camphor. So far as is known to the writer, no reliable method for the estimation of camphor oil in crude camphor has been published, other than the melting-point method given in Allen's "Commercial Organic Analysis," vol. iv., p. 197. | The moisture is then estimated on the pressed cake, and from the difference between the result and the original moisture the amount of water expressed is found, and, by difference, the amount of oil in the expressed liquid is found. It is then assumed that the water still remaining in the pressed cake is associated with as much oil as that in the expressed liquid, and the total oil calculated on that basis. The accuracy of this assumption may be open to question, but, with a good press, very little moisture remains in the cake, while the m. pt. of the pressed camphor usually indicates a fairly high degree of purity.-The Analyst, June, 1920. A NUMBER of methods for the detection of petroleum spirit in vegetable oils have been proposed. A careful study of these has shown that they are not satisfactory, but it has been found that Nastjukoff's formolite reaction may be applied successfully for the purpose as follows: The ol (50-100 grms.) is saponified by means of potassium hydroxide solution. Distilled water and pure calcium chloride solution are added, the liquid distilled by means of steam, and the distillate treated with 40 per cent formaldehyde solution and a few drops of concentrated sulphuric acid. A reddish brown film coloration on the surface of the liquid, gradually changing to deep yellow, indicates presence of petroleum spirit. If a few drops of the distillate are added to water, a brilliant interference ring of optical waves is produced on the surface of the water; this ring becomes almost invisible after standing for some time and disappears completely on beating. With soya bean oil the ring does not change, even on beating. The above process is capable of detecting traces of petroleum spirit in vegetable oils, and may be made the basis of a quantitative method, the formolite precipitate being weighed after drying at 110-115° C.Journal Society Chemical Industry, June 15, 1920. A REVISION OF THE ATOMIC WEIGHT OF TIN.* We have determined in a new and more careful series of trials the ratio SnBr4: Ag by analysis of newly prepared tetrabromide of tin. The materials for this purpose were prepared with great care and cleanliness, and chief attention was paid to removing all traces of moisture from the bromine used for the synthesis of the tetrabromide, and to keeping the latter in an absolutely dry state. Individual analyses were carried out in the same way as before (B. Brauner and H. Krepelka (Fourn. Am. Chem. Soc., xlii., No. 5), in the earlier work. The details follow. although one profited, of course, by the experience gained Preparation of New Materials. As this laboratory was called upon to examine a considerable number of Chinese crude camphors, the follow ing method worked out here may be of interest :-The moisture is first estimated on the well-mixed sample, then 100 grms., weighed to the nearest o'i grm., are trans(The remaining reagents not mentioned here were preferred to a press and pressed between two layers of lint. The press designed for this work has a steel cylinder pared in the same way as stated in the preliminary work). Bromine.-Ordinary bromine was shaken in portions in 2 inches in diameter and 6 inches deep, and is furnished a separatory funnel containing distilled water, every with a movable perforated bottom plate. The piston is operated by a strong screw thread. The sample is Presented to the Bohemian Academy of Sciences, Prague, 1919. allowed to remain in the press under pressure for fifteen From the Journal of the American Chemical Society, xlii., No. 5. portion being thus treated three times. This bromine was then redistilled from a saturated solution of potassium bromide. The distillate was allowed to drip into a solution of potassium oxalate prepared by neutralising pure oxalic acid with pure potassium carbonate. The potassium bromide thus obtained was recrystallised three times, and its solution was evaporated with a small quantity of potassium dichromate and twice distilled sulphuric acid. The bromine thus set free removed all iodine that may have been present. This evaporation was repeated three times. In order to remove organic matter the dry bromide was melted in small quantities in a platinum crucible. From this remelted bromide bromine was set free by means of an amount of potassium dichromate and sulphuric acid, such as to leave undecomposed bromide in the distilling flask., The bromine was then redistilled, and from a portion of this distillate was prepared the calcium bromide used to dry the remaining part of the bromine, which was then further dried by shaking it with phosphorus pentoxide (twice sublimed in a stream of oxygen), and was afterwards distilled directly into the apparatus in which the synthesis of tetrabromide took place. Tetrabromide of tin was prepared in the modified Lorenz apparatus. During the preparation the communication with the outside air was effected by a drying system (described in a previous communication-Journ. Am. Chem. Soc., xlii., No. 5, p. 917) to which were added tubes containing sublimed phosphorus pentoxide. The course of the reaction showed that both the bromine used and atmosphere of the reaction bottle were absolutely free from moisture, since the first drop of bromine did not at once react with the tin-foil-only after five minutes did the reaction begin slowly to take place. As soon as the first foil was covered with tetrabromide the reaction became violent and accompanied by such heat that the surrounding pieces of foil were melted into a ball, and the whole reaction bottle had to be quickly cooled. The tetrabromide obtained was introduced into glass bulbs provided with cone-shaped necks and there sealed. This arrangement removed the difficulties encountered at the same operation during the preliminary work, when the bulbs used had straight necks. Silver.-750 grms. of pure ordinary silver was dissolved in portions in distilled nitric acid (1:2) in such a way as to leave a small part of the silver undissolved. The solution of silver nitrate thus obtained was heated to the boiling-point, and, after the expulsion of nitrous gases, was filtered. The clear solution, coloured pale blue by copper, was allowed to crystallise by evaporation. Silver nitrate thus obtained was fused in a porcelain dish until the melted mass became black. This black substance was dissolved in water, and the solution was filtered and allowed to crystallise by evaporation. The crystallisation was repeated three times. A spectroscopic examination of the last crystals showed only the lead line 3683 62 (intensity 1000), proving thus that all other metals had been removed. The silver nitrate obtained in this manner was reduced by ammonium formate (prepared from pure formic acid and freshly distilled ammonia) and the reduced silver was washed with distilled water until the Nessler reagent gave no test for ammonia, dried in an electric drying oven at 150°, and then melted in a current of pure hydrogen according to the method elaborated by T. W. Richards (Journ. Am. Chem. Soc., 1955, xxvii., 472). Pure hydrogen for this purpose was obtained by the electrolysis of a solution of pure sodium hydroxide prepared from pure metallic sodium. The apparatus used consisted of a long U-shaped tube, in one arm of which was evolved oxygen and in the other hydrogen. The arms were long in order to prevent the mixing of the two gases. The hydrogen was conducted into a purifying system composed of two Richards' washing flasks filled with a saturated solution of silver sulphate, a U-tube containing fused sodium hydroxide, a tube filled with red-hot pumice stone covered with platinum and of another Utube containing fused sodium hydroxide. The individual parts of the apparatus were connected by means of airtight ground joints. Grains of remelted silver were etched with dilute nitric acid which had been redistilled, then washed successively in distilled water, ammonia water, and again in distilled water. The final delicate operation, namely, the preparation of small pieces of pure silver for weighing, was carried out in the following manner:-The greater portion of the silver buttons was cut up into small pieces on a slab of pure silver with a sharp steel chisel-the pieces were then etched and washed as above. The remaining grains were rolled between clean steel rollers in such a way that after each passing through the roller the silver was etched and washed in order to remove any trace of iron which it might have acquired. The silver foil was then cut by scissors into small pieces, which were then etched and washed as before. Weighing was done on the same balance as in the preliminary work, but not until after the weights bad been tested and corrected. Sx analyses were made in this series, and the results are given in the accompanying table. Discussion.-The mean value from six determinations is 118 699 (0.016). The maximum figure, 118.727, found only once, was arrived at by the analysis of Bulb No. 18, which was filled as far as to the neck. The minimum was 118 674. and this was the result of the analysis of Bulb No. 8, 1.kewise filled up to the neck. The maximum difference between the highest and lowest values is thus o 053. The mean value of all six ratios, SnBr1: 4Ag, is 101586. The only probable source of error of this series of analyses appears to have been the space in the necks of the bulbs not filled with tetrabromide. The agreement of the resulting mean value of the atomic weight of this series of analyses with the mean value of the preliminary determination increases the probability of this figure, and supports the value of the atomic weight of tin, |