: acid, and at the time when this investigation was planned | o-, m-, and p-toluidine and B-naphthylamine compounds existence of these substances in certain reactions has been frequently postulated with varying degrees of certainty, and opportunity has now been taken to test the accuracy of certain assumptions concerning their reactivities. Fromm and his collaborators (Ber., 1908, xli., 3397: 1909, xlii., 3816) have advanced the view that in alkaline solution the sulphoxylic acids are unstable, and tend to undergo simultaneous reduction and oxidation, mercaptans and various products of oxidation being formed. This is fully confirmed by the behaviour of thiobenzanilide oxide in alkaline solution. The sodium salt may be obtained by rapidly cooling a warm solution of the substance in 10 per cent aqueous sodium hydroxide; it forms very pale yellow needles, which are soluble in cold water, but sparingly so in the cold alkaline medium. The salt is not stable, and if the aqueous solution is warmed it is completely decomposed, yielding benzanilide and thiobenzanlide. It is clear that the latter is formed by reduction of the oxide, and independent experiment shows that the former product is obtained when the oxide is treated with alkaline oxidising agents. In explaining the reactions of disulphides with concentrated sulphuric acid, it has also been postulated (Prescott and Smiles, Trans., 1911, xcix., 642) that the sulphoxylic acids are capable of condensing in that medium with an aromatic nucleus yielding sulphides, for example: R.SOH+HAr= R·S·Ar + H2O. Support is given to this assumption by the fact that thiobenzanilide oxide is almost quantitatively transformed by warm sulphuric acid into the phenylbenzothiazole, which has been previously obtained (for example, Ber., 1886, xix., 1068) in other ways : S C6H6 HOC C6H1 = C6H1NC·C6H5+ H2O. Among other reactions of this oxide mention may be made of that which takes place with dry hydrogen chloride in presence of a dehydrating agent; sulphur is eliminated, and the iminochloride of benzanilide is formed as expressed by the following equation :— C6H5*N:C⭑C6H5 LOH + HCl = C6H5⋅N:CCl·C6H5+S+H2O. The further investigation has been abandoned owing to the recent discovery (Ber., 1912, xlv., 2965) of other sulphoxylic acids of less intricate structure. 333. "The Condensation of a Keto-B anilino a phenyl. ethane and its Homologues with Carbonyl Chloride, Phenylcarbimide, and Phenylthiocarbimide." By HAMIL TON MCCOMBIE and HAROLD ARCHIBALD SCARBOROUGH. McCombie and Parkes (Trans., 1912, ci., 1991) have shown that carbonyl chloride and a-keto-B-anilino-aß diphenylethane can be condensed to give 3:4:5-triphenyl-2: 3-dihydro-2-oxazolone. This synthesis has now been extended to a-keto-3-anilino-a-phenylethane, which with carbonyl chloride gives 3: 5-diphenyl-2 3-dihydro-2oxazolone (I.). This reaction has been applied also to compounds in which aniline has been replaced by o-, m., and p-toluidine and 3-naphthylamine. : These oxazolones are stable compounds, and their basicity is so slight that no pierates or hydrochlorides could be isolated. CPh−0, CO I. CPh•NPh CPh•NPh CH•NPh The glyoxalones yield salts with picric acid, but not with hydrochloric acid. These picrates consist of one molecule of the base combined with one molecule of the acid. Phenylcarbimide has been replaced by phenylthiocarbimide yielding 1:3: 4-triphenyl-2 : 3 dihydro-2-glyoxalthione (III.). 334. "Note on the Nitration of p-Hydroxyacetophenone." By FRANK GEO. POPE. Gattermann (Ber., 1892, xxv., 3523) obtained 3-nitro-phydroxyacetophenone in small yield (11 per cent) by the condensation of o-nitroanisole with acetyl chloride in the needles melting at 130.5°. The position of the nitropresence of aluminium chloride. It crystallises in yellow group was determined by oxidising the methyl ether with dilute nitric acid to 3 nitroanisic acid, which had already been obtained by Salkowski. Since the hydroxy-ketone is merely a p-substituted phenol, there appeared to be no reason why nitration should not be directly attempted, and the following experiment was consequently carried out. 35 grms. of the finely-powdered hydroxy-ketone were added in small quantities at a time to a well-cooled mixture of 20 cc. of nitric acid (D 1'42) and 20 cc. of concentrated sulphuric acid, with constant stirring. After half an hour the mixture was poured on to crushed ice, and the faintly yellow precipitate was collected, well washed, and dried. The amount of solid matter so obtained was 3'5 grms., and in the crude state melted at 128-130°. represents a 75 per cent yield of nitrohydroxyaceto. phenone. This The nitro compound was then crystallised from boiling water, separating in slender, yellow needles, melting at 135°. (Found, N=7'92. C8H704N requires N = 7.73 per cent.) Hence the nitro compound is 3-nitro-p-hydroxyacetophenone, and further proof is given by the fact that on methylation and subsequent oxidation with dilute nitric acid, 3 nitroanisic acid is obtained. The acid prepared in this manner melts sharply at 190—191o, whereas Salkowski gives 186-187°. The methyl ether which was also obtained simultaneously by Gattermann is readily prepared by methylating sulphate, and then nitrating the methyl ether in the manner phydroxyacetophenone in alkaline solution with methyl described above. On oxidation with dilute nitric acid it specimen of the acid as obtained by the two different also yields 3-nitroanisic acid (m. p. 190-191°), a mixed methods also giving the same melting point. 335. Bromoxylenols obtained from Dimethyldihydroresorcin." (Preliminary Note). By ARTHUR WILLIAM CROSSLEY and SYDNEY SMITH. In 1903 Crossley and Le Sueur (Trans., 1903, lxxxiii., 110) described experiments on the action of phosphorus haloids on dimethyldihydroresorcin, and showed that both with phosphorus pentachloride and especially phosphorus pentabromide, the resulting substances readily undergo transformation into aromatic derivatives, giving in the latter case bromoxylenols. There were described a mono bromoxylenol melting at 83.5-84° and a dibromoxylenol melting at 96.5°. Both these substances gave on treat. ment with bromime a tribromoxylenol meiting at 183°, which, it was suggested, might be identical with tribromo0 3-xylenol melting at 184°. A second tribromoxylenol was encountered, melting at 177°. Brazier and McCombie (Trans., 1912, ci., 2352) acted on a-keto-ß-anilino-ad-diphenylethane with phenylcarbimide, and obtained 1:3:4: 5-tetraphenyl 2: 3-dihydro2-glyoxalone. This reaction has now been extended to a-keto-3-anilino-a-phenylethane, which gives with phenylcarbimide, 13: 4 triphenyl- 2:3 - dihydro- 2 glyoxalone (II.). The reaction occurs in two stages; the two compounds are heated together on a water-bath until Further work has shown that the reactions are very a white mass is obtained, which on treatment with alcoholic complicated and susceptible to the slightest variation in hydrogen chloride yields the glyoxalone; the corresponding | experimental conditions. Although by no means com 1 plete, it is considered desirable to place on record a brief | after prolonged drying in a vacuum over potassium hysummary of the definite results so far obtained, as the droxide corresponds with the formula 2C11HION,CI, HCI. are unable to continue present authors the work The platinichloride, (C11H11ON4)2PtCl6, the aurichloride, conjointly. (C11HION4) AuC14, and the very explosive dichromate, (1) The monobromoxylenol, m. p. 84°, and the dibro (11H11ON4)2Cr2O7, have the normal composition. moxylenol, m. p. 96.5, have been proved to be derivatives of 0-3 xylenol; and although the position of the bromine atom in the former has not yet been decided, the latter has been proved by synthesis to be 4: 5-dibromo 0 3-xylenol. The synthetic formation is indicated by the following formulæ : (3) There has been isolated from the action of phosphorus pentabromide on dimethyldihydroresorcin a monobromoxylenol melting at 103°. Although the constitution of this substance has not been proved by synthesis, it would appear to be a derivative of 0-4-xylenol, because on further treatment with bromine it gives a tribromoxylenol. and from this an acetyl derivative, the melting points of which are not lowered on admixture respectively with tribromo 0-4-xylenol and its acetyl derivative. In this rearrangement, therefore, a methyl group has again wandered into an ortho-position; but from carbon atom I to carbon atom 6, instead of carbon atom 2:— The work is being continued, and it is hoped shortly to communicate more detailed accounts of these transforma tions and of the nature of the resulting products, for which purpose the synthesis of a number of bromoxylenols is being carried out. 336. "Non-aromatic Diazonium Salts." (Preliminary Note). By GILBERT T. MORGAN and JOSEPH REILLY. The authors have investigated the degree of diazotis ability of the following partly aromatic and non-aromatic amines: 4 aminoantipyrine, 6 amino-2 : 4-dimethylpyrimidine, 4 amino-3: 5-dimethylpyrazole, 4 amino 1: 3:5 trimethylpyrazole, 2 amino 6 oxypurine (guanine), and aminomethyltriazole. When 4-aminoantipyrine hydrochloride is diazotised with ethyl nitrite in alcoholic solution without excess of mineral acid an uncrystallisable, resinous product is obtained on concentrating the solution. With excess of acid a crystalline diazonium hydrochloride is produced, which Although possessing a chemical constitution closely allied to the aromatic amines, 6-amino-2: 4-dimethylpyrimidine is not diazotisable under the ordinary experi mental conditions, and prolonged treatment with nitrous acid leads to the formation of the corresponding 6-hydroxy2: 4-dimethylpyrimidine. 4-Amino-3: 5 dimethylpyrazole gives rise to diazonium salts as stable as those of 4-aminoantipyrine, and these substances are being investigated, together with those of 4-amino-13: 5-pyrazole. 2-Amino-6 oxypurine does not diazotise under conditions which readily give rise to diazonium salts in the case of aminomethyltriazole. SOCIETY OF CHEMICAL INDUSTRY. (LONDON SECTION.) Ordinary Meeting, January 6th, 1913. Dr. W. R. HODGKINSON in the Chair. The following papers were read and discussed :"Estimation of Glyceryl Acetate in Essential Oils." By S. GODFREY HALL and A. J. HARVEY. The authors discuss two methods which have been proposed for the estimation of glyceryl acetate in essential oils. They suggest a process dependent upon the separation and estimation of the glyceryl as such. The glycerol is liberated by a process of saponification and separation of oily matter, and its amount is determined by the triacetin method. The details of the process and some results obtained with mixtures of known composition are given in the paper. "Estimation of Moisture." By F. H. CAMPbell. calcium carbide as a reagent for the estimation of moisture, He gives comparative results obtained with this and other processes for various substances, such as coal, tea, coffee, butter, &c., and shows that the carbide and vacuum processes give the best agreement. "Determination of Moisture in Foods and other Organic Substances." By W. P. SKERTCHLY. In referring to the drying of substances by exposure to sulphuric acid or other dehydrating agent under reduced pressure, the author points out that this method of determining moisture is particularly suitable in the cases of substances which undergo changes when heated, as in the usual method of carrying out the determination, and that, in the case of liquids, it is advisable to pour them upon an absorbent material such as sand, asbestos, &c., to expose the maximum surface. He gives in tabular form the comparative results of drying such substances as flour, biscuit, maize meal, arrowroot, ground rice, soap, cotton wool, mustard, and manure from hides, to a constant weight in vacuo for about twenty-four hours at the ordinary temperature and heating for about two hours (fifteen hours in the case of soap) in a water oven at 100° C. It is shown that, in all cases, more moisture is yielded when the substances are dried in vacuo, and that, in the case of some of the farinaceous materials, the difference amounted to nearly 2 per cent. It was also observed that when the heating at 100° C, was prolonged for more than two hours, the substances began to increase in weight to the extent of a few mm. for each period of one hour. "Determination of Water." By G. N. HUNTLY and J. H. COSTE. The authors point to the frequent importance of an accurate determination of water in commercial products, and discuss the following methods for the determination | of water in various substances, devoting particular attention to VIII, and IX. A. Direct Methods. I. Water driven off by ignition, condensed in part of ignition tube, and weighed directly. (Penfold.) II. Substance heated in a current of dry gas, or in a vacuum, water vapour collected in calcium chloride or sulphuric acid, and weighed. III. Substance mixed with an excess of a volatile nonmiscible liquid such as xylene, distilled, and the water measured under the hydrocarbon layer, with or without corrections for the mutual solubilities of the water and liquid used for its removal. IV. The material directly heated to 130° C. by a vapour jacket (high pressure steam), the steam given off condensed and measured. THE new feature in the above well-known volume consists in an account of the latest rules issued by the Lord Chamberlain as to the wearing of orders, medals, &c., at public entertainments. This addition is likely to be of value to many of those whose names appear in the book. The "new honours" have increased the number of pages by seventeen over last issue. One addition is recorded in the "Order of Merit," Admiral of the Fleet Sir A. Knyvett Wilson, G.C.B., G.C.V.O., V.C. The International Whitaker, 1913. London: Joseph Whitaker. THIS is an entirely new work, and will certainly be welcomed by English-speaking people abroad. The book is divided into four parts. The first deals with the relative functions of the components of the Universe; the second with a description of the land surface of the Earth, physical geography, and ethnographical divisions of Mankind; the third with each Nation, giving its Area, Population, Production, Finances, and other important particulars; the fourth part gives a list of British and American Diplomatic and Consular Representatives in Foreign Countries. There is a very complete index, occupying close upon forty pages, which includes every reference that can be reasonably demanded. MISCELLANEOUS. Royal College of Science, London.-The Fifth Annual Dinner of Old Students of the Royal College of Science, London, will be held at the Cafe Monico, Shaftesbury Avenue, on Saturday, January 25th, 1913, at 7 for 7.30 p.m. The President of the Old Students Association (Sir William Crookes, O.M., D.Sc., F.R.S.) will preside, and the guests will include Sir Alfred Keogh, K.C.B., Sir Henry Miers, F.R.S., Sir Robert Morant, K.C.S., Lt. Col. Sir David Prain, C.I.E., Sir Amherst Selby-Bigge, K.C.B., Dr. R. T. Glazebrook, F.R.S., and Dr. H. Frank Heath, C.B. With the object of giving the dinner as much as possible the character of a social reunion for old students, the Dinner will be arranged in tables of eight, and tables will be specially arranged for parties as desired. Friends of old students, including will be shorter than in previous years, and a programme ladies, are invited to attend the Dinner. The toast list of music provided. Tickets may be obtained on applica tion to the Secretary of the Old Students Association, 3, Selwood Place, S.W. MEETINGS FOR THE WEEK. MONDAY, 20th.-Royal Society of Arts, 8. (Cantor Lecture). "Liquid Royal Society. "Relation of the Islets of Langerhans Influ Quinone-ammonium Derivatives-Part II., Nitro-haloid, Di-haloid, and Azo-compounds," by R. Meldola and W. F. Hollely. "Chemical Nature of some Radio-active Disintegration Products," by A. Fleck. "Action of Ammonia and Alkyl Amines on Reducing Sugars," by J. C. Irvine, R. F. Thomson, and C. S. Garrett. "Mode of Combustion of Carbon," by T. F. E. Rhead and R. V. Wheeler. "Chlorination of Iodophenols. Part II. Action of Chlorine on 2:46-Tri-iodo-, 2: 6-Dibromo-4-iodo-, 2:4-Diiodo- and 2-Bromo-4-iodo-phenol," by G. King and H. McCombie. Quercetagetin," by A. G. Perkin. "Oxyhydroquinone Phthalein Andydride and Oxyhydroquinone Benzein," by K. N. Ghosh and E. R. Watson. "2: 2'-Ditolyl-5: 5'-dicarboxylic Acid," by J. Kenner and E. Witham. FRIDAY, 24th.-Royal Institution, 9. "Recent Advances in Scientific Steel Metallurgy," by Prof. J. O. Arnold. Physical, 5. "Electrical Conductivity and Fluidity of Strong Solutions," by W. S. Tucker. "Resistance of Electrolytes," by S. W. J. Smith and H. Moss. "Reca escence of Iron Carbide," by S. W. J. "Aspects of Harmony," by Smith and 1. Guild SATURDAY, 25th.-Royal Institution, 3. H. Walford Davies. THE CHEMICAL NEWS. when dry they are rapidly decomposed by steam, hydro VOL. CVII., No. 2774. THE PREPARATION OF and showed that although very stable at high temperatures Aluoric acid being liberated and a metallic oxide left. He found, for example, that on heating mercuric fluoride it volatilised without decomposition when dry, but when damp decomposed giving hydrofluoric acid, mercury, and oxygen. Again, silver fluoride decomposed similarly in presence of moisture, but could not be decomposed by heat when anhydrous. He investigated the action of chlorine ANHYDROUS HYDROFLUORIC ACID AND THE and of oxygen on calcium fluoride heated to a high temISOLATION OF FLUorine. By F. D. CHATTAWAY. THE failure of so many distinguished chemists in their attempts to isolate, fluorine induced Fiémy, Professor of Chemistry at the Ecole Polytechnique in Paris, to examine with great care the method of preparing hydrofluoric acid, and to study anew its properties, in order to ascertain if possible whether the element could be isolated or not. He recognised the magnitude of his task, for he begins*: "In such a work, the difficulties of which are known to all chemists, and coming after such men as Davy, GayLussac, Berzelius, and Thenard, I recognised that I ought not to count upon any such scientific good fortune as would lead me immediately to the discovery of fluorine, but I knew that a general study of the fluorides would be to the advantage of science, since it would complete the history of a series of compounds still little investigated, would probably destroy some accredited errors, and would indicate the path to follow when attempting the isolation of the radical. It is this thought that has constantly sus tained me during the prolonged investigation of which I now present the chief results." Fremy then describes how to prepare pure hydrofluoric acid, it never having been obtained previously absolutely free from other compounds or from water. He states: "The acid which is yielded by the action of sulphuric acid on fluor-spar always contains water, sulphuric acid, sulphurous acid, hydrofluosilicic acid, and other impurities. After many fruitless attempts I have given up attempting to purify and dehydrate the crude material obtained thus, and have sought another method of preparing anhydrous hydrofluoric acid." "The following procedure has given me a very satisfactory result. The substance which I use for preparing the acid is the acid fluoride of potassium. I obtain the salt very easily by pouring a strong solution of hydrofluoric acid into a similar neutral solution of potassium fluoride which has been produced by saturating hydrofluoric acid with carbonate of potash; the acid fluoride which is only slightly soluble crystallises immediately. As thrown down at first it is not pure, but contains hydrofluosilicate of potash, which can, however, be separated by several crystallisations. The acid fluoride of potassium is dried first between folds of unsized paper, next under the receiver of an air-pump, and finally in a drying oven; it can then serve for the preparation of anhydrous hydrofluoric acid. I place the well dried salt in a little platinum retort attached to a platinum receiver, plunged in a freezing mixture. Having connected the neck of the retort to the receiver I carefully heat the salt in the retort strongly so as to drive off the last traces of water, and I only collect the hydrofluoric acid when a part of the salt has been decomposed. The anhydrous hydrofluoric acid obtained by this method is gaseous at the ordinary temperature, but is condensed to a limpid liquid by a mixture of ice and salt. .. I have also obtained anhydrous hydrofluoric acid by decomposing fluoride of lead by pure and dry hydrogen in a platinum tube. The acid obtained in this way displays all the characteristics of that obtained by calcining the acid fluoride of potassium." Frémy studied with care a number of metallic fluorides, perature in a platinum tube, and found it to be slight. He further attempted to obtain fluorine from various metallic fluorides, including fluor-spar and potassium fluoride, by electrolysing the fused salts in a platinum crucible, using a platinum rod as a positive electrode. He found that such fused fluorides were decomposed, the metal going to the negative pole, while the platinum rod forming the positive pole was rapidly attacked. A gas was also liberated at the positive pole, which decomposed water, producing hydrofluoric acid, and displaced icdine from iodides. This gas Frémy believed to be fluorine, but he was not able to study it with any care, as from various causes, such as the wearing away of the positive pole or the perforation of the crucible, such electrolysis could only be carried on for a very short time. Frémy's attempts having proved unsuccessful, in spite of his skill and perseverance, little further was done on the subject until Gore, in 1869, made a fresh systematic study of hydrofluoric acid. Having prepared the pure anhydrous acid by Frémy's method, he determined its boiling point and vapour pressure at different temperatures, and investigated most of its chief properties in a manner much more exact than had been done previously. Faraday* apparently had discovered that absolutely anhydrous hydrofluoric acid would not conduct an electric current. This observation Gore verified; he showed that when a little water was present the current passed, but only as long as this was being decomposed; when the water disappeared the current stopped." After a few other unimportant attempts to isolate the element had been made, Moissan, who had been a pupil of Frémy, about the year 1883 took up the subject in a much more careful and systematic manner than had hitherto been possible. After endeavouring unsuccessfully to obtain the element from one of its compounds with silicon, phosphorus, or arsenic, he came to the conclusion that no reaction carried out at a high temperature was likely to lead to any result, and decided to try again the electrolysis of hydrofluoric acid. The hydrofluoric acid used was prepared by Frémy's method, and great precautions were taken to obtain it anhydrous. Moissan thus describes its preparation:"A known volume of carefully made commercial hydrofluoric acid was taken and one quarter of it neutralised by an alcoholic solution of potassium hydroxide, or better by pure potassium carbonate prepared from the bicarbonate. The two portions were then mixed, and the whole distilled in a lead retort heated in an oil bath to a temperature of 120°. At this temperature silico fluoride of potassium is not decomposed, and an acid was collected free from silica, which commercial hydrofluoric acid always contains in considerable quantity. This distilled acid was then divided into two equal parts, and one was exactly neutralised by pure potassium carbonate. The solution of neutral potas Gore (Phil. Trans., 1869, clix., 189) states that Faraday first made this observation, and refers to the "Handbook of Chemistry" by Leopold Gmelin, translated by Watts, London, 1858, vol. i., p. 455), where, speaking of hydrofluoric acid, the following statement occurs: "The hydrated acid is not decomposed only by the water with which it is united suffering decomposition (Faraday)." The only actual statement by Faraday on the subject that I have been able to find is the following which occurs in the "Experimental Researches in Electricity," vol. i., 2nd edition, London, 1849, Section 770, p. 227"Solutions of hydrofluoric acid did not appear to be decomposed under the influence of the electric current; it was the water which gave way apparently." sium fluoride thus obtained was mixed with the second half of the acid, and thus converted into the acid fluoride. The salt itself was obtained by evaporating the solution on a water-bath at 100° in a platinum basin. It was finally completely dried by exposing it in a vacuum over concentrated sulphuric acid. A silver crucible containing two or three sticks of fused caustic potash was also exposed in the vacuous space alongside it. The sulphuric acid and potash were renewed every twenty-four hours for about fifteen days, and the pressure in the bell-jar was maintained at about 20 mm. of mercury. Care must be taken during the drying to powder the salt from time to time in an iron mortar so as to expose fresh surfaces; when completely free from water the acid potassium fluoride falls to powder, and can then be used to prepare the anhydrous acid. It should be noted that properly prepared acid potassium floride is not deliquescent like the neutral salt. To prepare the acid the dry acid salt was placed as quickly as possible in a platinum retort previously heated to redness so as to free it completely from moisture. The salt was first heated for about an hour to a moderate temperature so that slight decomposition only occurred, the small quantity of acid distilling over during this preliminary heating, which contained any traces of water absorbed during the manipulation of the salt, was allowed to escape. A platinum receiver was next fitted to the retort, which was then more strongly heated, but only to a temperature sufficiently high to decompose the salt slowly; the receiver was cooled by a mixture of ice and salt. From the moment of cooling the receiver all the hydrofluoric acid was condensed. It condenses, as is well known, to a limpid liquid, boiling at 19.5° C., which is very hygr scopic and fumes strongly in presence of damp air. Hydrofluoric acid thus obtained sometimes contains a small quantity of alkaline fluoride carried over by the acid vapours during the decomposition of the salt. It is not necessary for the end in view to avoid this as it helps to render the acid a conductor. If it is desired to obtain pure hydrofluoric acid a much larger platinum retort must be used, and the acid vapours must be led to the condenser through a long platinum tube sloping upwards from the retort, and kept above the temperature at which the acid boils." Hydrofluoric acid is so hygroscopic that it can be kept in the anhydrous state only with the greatest difficulty, and it was found that the one practicable procedure was to prepare it immediately before use. With acid prepared thus, Moissan confirmed Gore's obser vation that anhydrous hydrofluoric acid does not conduct. He states that if the acid contains a small quantity of water, either added intentionally or present through the acid having been prepared carelessly, this alone is decomposed, ozone being set free at the positive pole. As the water is broken up the conductivity of the acid diminishes, and when the whole disappears and the acid becomes anhydrous the current no longer passes. Moissan states:-"In several experiments an acid so free from water was obtained that a current of 35 ampères furnished by fifty Bunsen cells was totally stopped." Guided by the experience gained in his previous attempts to electrolyse arsenic fluoride, Moissan then tried the effect of adding to the non-conducting hydrofluoric acid a small quantity of very dry potassium hydrogen fluoride, KF.HF. This was found to dissolve readily in the anhydrous acid. Moissan states:-"If fragments of potassium hydrogen fluoride are added to anhydrous hydrofluoric acid contained in a platinum crucible they are seen to disappear rapidly, potassium hydrogen fluoride being very soluble in the anhydrous acid." Submitting such a solution to electrolysis, he found that the current passed and that a gaseous product was liberated at each electrode. To enable the products to be collected separately the well known platinum U-tube apparatus was devised. * Compare Gore (loc. cit.). The electrolysis was carried out at a very low temperature, to prevent any gaseous product formed being largely diluted with the vapour of hydrofluoric acid and to diminish the risk of the apparatus being attacked destructively by any fluorine liberated. For this purpose the U-tube was immersed in a bath of methyl chloride, the temperature of which, although it boils under the ordinary pressure at 23°, can easily be lowered to 50°, when its evaporation is increased by aspirating through it a current of dry air. Moissan subsequently found that a product more free from hydrogen fluoride could be obtained by conducting the electrolysis at a temperature lower than this, and in his later experiments used a bath of acetone holding solid carbon dioxide in suspension, with which a temperature of -80° could be obtained. Moissan thus describes an experiment :-"While the hydrofluoric acid was being prepared the platinum U-tube and the electrodes were dried in an air bath at 120°. About 6 to 7 grms. of thoroughly dried acid potassium fluoride were introduced into the apparatus, the stoppers were very carefully screwed in and covered outside with a coating of melted shellac. The U-tube was then fixed into a glass cylinder by a cork, the entrance of moisture being prevented by attaching to the outside tubes drying tubes containing recently fused potash. Methyl chloride was then introduced into the glass cylinder, and the temperature in this way reduced to 23°. The hydrofluoric acid was then drawn into the apparatus, from the receiver in which it had been condensed, through one of the side tubes by means of a mercury aspirator. In each experiment from 15 to 16 grms. of acid were used. As soon as the current was passed gas was regularly evolved from each pole. At the negative pole hydrogen was obtained which burnt with an almost invisible flame, yielding water vapour, the properties of which could easily be verified. At the positive pole a gas was disengaged, apparently colourless and endowed with very great chemical activity... Fluorine was thus isolated for the first time by Moissan on June 26th, 1886. The apparatus used at first was made wholly of platinum, but it was subsequently found that a copper U-tube could be employed. In early experiments the electrodes were made of platinum containing 10 per cent of iridium, this alloy being somewhat less readily attacked than pure platinum; later electrodes of pure platinum were used, made club shaped at the end so that they would resist the attack of the fluorine for a longer time. The positive electrode was often completely corroded away during an experiment, but the U-tube was never found to have lost any weighable amount of platinum. In order to study the properties of fluorine it was absolutely necessary to obtain it free from the vapours of hydrofluoric acid derived from the acid undergoing electrolysis. To do this the fluorine was passed through a little platinum worm serving as a condenser, and cooled by a bath of methyl chloride to about - 50°. The gas issuing from this only carries away the small quantity of hydrogen fluoride corresponding to its vapour pressure at this temperature, which is about 70° below its boiling-point. It was freed from the last remaining traces of hydrofluoric acid by passing it through two small platinum tubes filled with fragments of fused sodium fluoride, which does not attract moisture as does potassium fluoride, and which combines readily at the ordinary temperature with hydrogen fluoride. Using 26 to 28 Bunsen cells mounted in series, and employing an apparatus containing from 90 to 100 grms. of anhydrous hydrofluoric acid in which 20 to 25 grms. of potassium hydrogen fluoride had been dissolved, Moissan obtained a yield of 2 to 3 litres of fluorine per hour. * Moissan again notes that a higher temperature than this, say - 10°, is insufficient to prevent the gas escaping at each pole from being mixed with a large excess of the vapour of hydrogen fluoride. |