THE AMATEUR PHOTOGRAPHER and PHOTOGRAPHIC NEWS. Edited by F. J. MORTIMER, F.R.P.S. Price TWOPENCE - Every TUESDAY. The Leading Paper for everyone interested in Special Articles on PICTORIAL and MODERN SCIENTIFIC PHOTOGRAPHY. The best and brightest Photographic Weekly for the Beginner or Advanced Worker. Office: 51, LONG ACRE, LONDON, W.C. London: Printed and Published for the Proprietor by EDWIN JOHN DAVEY, at the Office, 16, Newcastle Street, Farringdon Street, E.C. February 12, 1909 Published Weekly. Annual Subscription, free by post, £1. Entered at the New York Post Office as Second Class Mail matter. Vol. 99.-No. 2569. [Copyright Friday, February 19, 1909. Registered as PRICE 44. 95 CHEMICAL SOCIETY .......... a Newspaper. POST FRBB 444, NOTICE to ADVERTISERS. All communications respecting Advertisements in the CHEMICAL NEWS should be addressed to the Sole Contractors, HAZELL, WATSON, and VINEY, Ld., 52, Long Acre, London, W.C. TO CHEMICAL MANUFACturers, enGINEERS, and OTHERS. C. PASS AND SON, LIM., BRISTOL, MESSRS FULLER, HORSEY, SONS, and Are Buyers of LEAD ASHES, SULPHATE OF LEAD, LEAD SLAGS, ANTIMONIAL LEAD, COPPER MATTE, TIN ASHES, &c., ORES, DROSS, or RESIDUES, containing TIN, COPPER, LEAD, and ANTIMONY. Chemist (young) desires Situation in Labora tory. Any capacity, Analytical or Manufactory. Knowledge of general analytical work, Foods and Drugs, and Minerals. Moderate salary required, -Address, F. D., CHEMICAL NEWS Office, 16, Newcastle Street, Farringdon Street, London, E.C. hemist (Suisse-Français), 26 years of age, Ch Doctor of Science, six years' University Course, of which three Wa Wanted by a Young Man just completing a three years' training in a large Technical Laboratory, a Position as Assistant in Metallurgical Laboratory or Works. Well up in electro-chemical methods, and in the analyses of Steels, Ores, Alloys, Lubricating Oils, and Coals. Moderate salary.-Address, CASSELL are instructed to SELL by AUCTION, in Lots, at the WOOD PROCESS WORKS, CARPENTER'S ROAD, STRATFORD, E., on FRIDAY, MARCH 5, at Eleven o'clock precisely, SURPLUS PLANT and MACHINERY, comprising 50 h.p. horizontal Steam Engine by Ruston, Proctor, 7 ft. by 28 ft.; Lancashire Boiler, 120 lbs. w.p.; steel Saturating Cylinder, 6 ft. 6 in. diameter 30 ft. long; two Gwynne's Centrifugal Pumps; new Claritas Filter-press; 12 wrought-iron Tanks, up to 6000 gallons capacity; Avery's Testing Machine; Circular Saw Bench; Drilling Machine; three Blackman Propellers; portable Weighing Machine; quantity Shafting, Pulleys, Belting; twelve steel-framed Bogies; about 1000 ft. Piping; Valves and Pipe Fittings; Engineer's Tools; Chemical Apparatus; quantity useful Timber; and other effects. May be Viewed by orders, and Catalogues had of Messrs. FULLER, HORSEY, SONS, and CASSELL, 11, Billiter Square, E.C, FOREIGN CHEMICAL BOOKS Supplied Rapidly. AMMONIA.-All Qualities and Strengths. Enquiries solicited. BROTHERTON & CO., LTD., CITY CHAMBERS, LEEDS. Works: Birmingham, Glasgow, Leeds, Liverpool, and Wakefield. "Crux," CHEMICAL NEWS Office, 16, Newcastle Street, Farringdon SECOND-HAND BOOKS at Half Prices! Street, London, E.C. For RUSSIA-WANTED as PARTNER. Practical Chemical Manufacturer with some capital, knowing thoroughly the manufacture of Carbonate of Ammonia, eventually also of white raw Carbolic Acid. -Address, J. H., CHEMICAL NEWS Office, 16, Newcastle Street, Farringdon Street, London, E.C. OLD PLATINUM IN ANY FORM PURCHASED FOR CASH. Highest prices allowed by New Books at 25 per cent Discount! Books for Chemical, Pharmaceutical, and ALL other Examinations supplied: sent on approval. State wants. Send for List. BOOKS BOUGHT.W. & G. FOYLE, 135, Charing Cross Rd., W.C. (THE Booksellers). PLATINUM CRUCIBLES, UTENSILS, FOILS WIRE, and SCRAP Bought at Highest Prices. THE LONDON REFINING & METALLURGICAL WORKS, Ltd. ROBERT PRINGLE & SONS, Gold and Silver GRIFFIN & SONS, LO, KINGSWAY, LONDON. Refiners, &., 40 and 42, Clerkenwell Rd., E.C. Photographic Residues reduced and Purchase. ABSOLUTE ALCOHOL, SOLE MAKERS OF REVISED WITH AN ANSWER TO VARIOUS CRITICS By SIR WILLIAM CROOKES, F.R.S. SECOND EDITION. VITH PREFACE AND ADDITIONAL CHAPTER, BRINGING THE With Two Chapters on the Future Wheat Supply of the OPINIONS of the PRESS. .. "Sir William Crookes has propounded a problem which in the next century [written in 1899] is bound to engage the close attention not merely of agricultural experts, but of economists and statesmen."-Speaker. "The appearance of the papers in this convenient form will be welcome to everyone who appreciates the importance of the problem."-Scotsman. "If these somewhat gloomy prognostications result in drawing the attention of chemists more seriously to what has hitherto been only an interesting laboratory problem, Sir William Crookes will have conferred an incalculable benefit on the race."-Western Morning News. "The student of economic science and sociology will find this volume full of interesting material. The entire subject is of the profoundest interest, and an excellent pur pose has been served by the publication of these papers in a single volume."-The Eagle (Brooklyn, N.Y.). CHEMICAL NEWS OFFICE. 16 NEWCASTLE ST., FARRINGDON ST., E.C. SULPHUROUS ACID AND SULPHITES. Liquid SO, in Syphons, for Lectures, &c. PHOSPHORIC ACID and PHOSPHATES. CARAMELS and COLORINGS for all purposes. A. BOAKE, ROBERTS, & CO. (LIMITED), STRATFORD, LONDON, E. THE CHEMICAL NEWS VOL XCIX., No. 2569. SCANDIUM.* PART I. THE SEPARATION OF SCANDIUM FROM By R. J. MEYER. Historical. 85 money required for the preparation even of small quantities, In the year 1879 Nilson (Ber., xii., 550, 554, March 22, 1879), who was preparing pure ytterbium, discovered by Marignac, by the fractional melting of the nitrates of the earths from gadolinite and euxenite, discovered a new earth which was present in small quantities in the ytterbium fractions. It was distinguished from the ytterbium earths by the readiness with which its nitrate decomposed on being heated-it was thus more electro-negative than they -and by its very low molecular weight. Thalén's examination of the spark spectrum in the optical region showed that a new element was present. Nilson named it scandium. However, he only succeeded in obtaining about 0'3 grm. of a product in which ytterbium still preponderated; its atomic weight was about go. Some weeks Quite recently, G. Eberhard published a much more later, Cleve (Bull. Soc. Chim., [2], xxxi., 486, May 13, comprehensive account of the occurrence of scandium 1879) announced that he also had succeeded in isolating (Sitzungsber. Kgl. Preuss. Akad. Wissensch., 1908, xxxviii., scandium, both from gadolinite and from yttrotitanite 851). Starting from the fact that the strongest lines of (keilhauite). The yields first given by him of 0.02 to 0.04 the scandium spectrum are observed in the spectra of stars per cent relate, however, to an oxide, which Nilson afterwards calculated to contain only about 30 per cent of in the most different stages of development, he concluded scandium earth, while the actual amount present in the that scandium must be universally distributed on the earth minerals worked up by Cleve was estimated to be about (the same reasoning foretells the occurrence of helium on 0.002 to 0.003 per cent. Subsequently, Cleve, working minerals and rocks he obtained the remarkable result that the earth), and by investigating the arc spectrum of 366 with richer raw material, succeeded in obtaining a rather scandium in small quantities is actually one of the most larger amount of the earth in an approximately pure state; widely distributed of the elements on the earth, and that he determined its atomic weight with some degree of accuracy, and by preparing some of its compounds he by spectral methods it may be detected in almost all the rocks which form the earth's crust. He further showed was able to give some description, if incomplete, of the that it is most frequently met with in zirconium minerals, properties of the element. In two determinations he obtained for the atomic weight Sc=44'91 ard 45:12. This beryls, titanates, niobates, and titan-niobates of the rare discovery was regarded as the more striking because earths, in micas, and finally-and this is the most important result from a chemical point of view-that wolframite and Mendeleeff (Lieb. Ann., Suppl., viii., 133) had foretold the tinstone from some places in the mountain ranges of existence and the properties of a new element of atomic weight 44, ekaboron, basing the prophecy on his law of Saxony and Bohemia contain scandium in such amounts that the extraction of the earth from these minerals would periodicity, and by the fulfilment of the prediction by the be profitable. Undoubtedly the disadvantage of the experimental discovery the position of scandium in the always small absolute amount of rare earths which these Periodic Table was settled once for all. In 1880 Nilson minerals contain, and which has hitherto quite escaped published what were for the time being the final statements on the subject (Ber., xiii., 1439, June 12, 1880; Ofvers af that scandium forms the predominant constituent of the detection by chemical analysis, is outweighed by the fact K. Svenska Vet. Akad. Förhandl., 1880, No. VI.). He had succeeded in isolating from 10 kilogrms. of euxenite worked up by Nilson and Cleve contained only traces of rare earths generally present in them, while the minerals some grms. of scandium oxide of a high degree of purity; scandium distributed throughout an enormous excess of he had also acquired further knowledge of its chemical properties, and had finally established its atomic weight by four determinations which gave the values 43'99, 44'07, 44'05, and 44'02. (By studying the arc spectrum Exner and Haschek showed that Nilson's preparations contained small amounts of ytterbium). The spark spectrum of this pure material was measured by Thalén (Comptes Rendus, xci., 45). The investigation of scandium was then compulsorily abandoned for a time, and hardly any mention is made of it in literature during the period from 1880 to 1907. Two which had been shown by its discoverers to be highly explanations may be given for this neglect of a substance interesting; owing to its extreme rarity investigators could hardly hope to be rewarded by appreciably increased yields for the considerable outlay of time, trouble, and • Zeitschrift für Anorganische Chemie, Ix., 134, November 17, 1908. other earths. theories we arrive at a practical result, it may also be fore- Feb. 19, 1909 The ready solubility of the hydroxide and the salts in soda, | examination for fluorine gave a negative result. examination for fluorine was performed by K. Daniel's in water. (NOTE.-Brauner appropriately calls scandium "an extrapolation" in the series of the elements of the rare earths (Abegg, "Handb. d. Anorg. Chem.," III., i., p. 340). If it is included amongst the cerium earths because of the difficulty with which its potassium double sulphate dissolves, then it occupies the same isolated position in this sub-group as yttrium in the group of the yttrium earths). The results obtained by Eberhard do not enable us to detect an evident connection between the chemical nature of scandium and its occurrence on account of its very wide distribution, but it is clear that the occurrence of the element is not restricted to the typical minerals of the rare earths. As regards the occurrence in wolframite and tinstone from the Ore Mountains, this localisation in a small region shows that specific geological causes are responsible for the concentration of the element in this place. For the investigation now to be described of the quantitative isolation of scandium from wolframite, it seemed desirable to acquire a knowledge of the most important reactions of the element from personal observation. I owe hearty thanks to Prof. Dr. W. Nernst, who enabled me to do so by sending me one of Cleve's original preparations. I should not have been able to carry through this task with any hope of success if the best procedure had not been pointed out by the spectrographic examination of almost all the precipitates and filtrates which were obtained in the numerous attempts to find the most advantageous method. These tedious examinations were very kindly performed by Prof. Dr. G. Eberhard, of the Astrophysical Observatory, in Potsdam. He used for the purpose a grating spectrograph belonging to the observatory (cf. G. Eberhard, loc. cit.). I wish to express to him also my hearty thanks for this valuable aid, which was indispensable for the success of the work. Preliminary Experiments. For the preliminary experiments I obtained from the Department for the sale of minerals of the Kgl. Sächs. Bergakademie, in Freiberg, wolframite from Zinnwald, near Attenberg, in the Saxony Ore Mountains; it consisted of large pure pieces containing only very little matrix. The quantitative analysis of the mineral will be published in a later communication. It was first necessary to ascertain into which part the scandium would go when the mineral was opened up with soda. A priori it was to be expected that when the melt was extracted with water the scandium would go over into the alkaline solution with the tungstic acid, since it forms soluble double carbonates. However, when the oxides which remained undissolved on extraction with water were subjected to spectroscopic examination, it was found that they contained all the scandium, while the alkaline filtrate, from which the tungstic acid was separated in the usual way, contained none at all. This behaviour was explained by an observation made later, that solutions of scandium salts in soda, if they are diluted with water and boiled, undergo a complete hydrolytic dissociation with separation of the carbonate. The assumption which was first made, that wolframite possibly contains scandium in the form of a fluoride insoluble in alkali, was not confirmed, for the In the first experiments the methods of isolating the scandium quantitatively were worked out. For this purpose the above mentioned very pure wolframite was used; it was finely powdered, and portions of 100 grms. were heated in Chamotte crucibles in a Perrot's furnace with 220 to 230 grms. of free soda, some potassium nitrate being added. The melt was washed with boiling water containing some alcohol, and the brown mixture of oxides left was dissolved in concentrated hydrochloric acid, after having been washed. The silica which separated was removed, the solution was evaporated to dryness on the water-bath, and the residue taken up with water. More recently, after the completion of the preliminary experiments, a manufactory which is engaged in the preparation of metallic tungsten placed at my disposal large quantities of oxide residues from Zinnwald wolframite. These residues from the manufacture of tungsten provide an excellent and cheap raw material for the preparation of scandium, but they can only be used for this purpose if they have been obtained from good ore ("Stuferz "), while the so-called "Setz" or "Schlicherz" (small ore or slick), which contains large quantities of gangue, gives a less productive oxide residue. Even the best ore often contains some matrix, and the residues yield large quantities of silica when they are dissolved in hydrochloric acid. It was usually found that 20 per cent by weight of this "technical wolframite oxides" is insoluble, a circumstance which must be taken into consideration in calculating the from some places in the Saxon and Bohemian Ore yield. I must again point out that only the wolframite Mountains is worth working up for scandium. Eberhard has proved spectrographically that specimens from other places only contain traces of it. result by working up the residues of an Australian wolframite, which gave a yield of only o'015 per cent of rare earths, while the normal yield of residues from Zinnwald wolframite amounts to o'30 to 0.33 per cent. I obtained the same METHODS OF SEPARATION. I will now describe the methods which lead to the separation and preparation in the pure state of scandium from the solution of wolframite oxides. The chief aim of these experiments was to obtain a "quantitative " isolation, which could only be attained if each precipitation was controlled by means of spectrographic examination. A.-Concentration by Precipitation with Oxalic Acid. It is well known that the group of the rare earths is usually totally separated from other substances, such as iron, manganese, calcium, &c., by precipitation with oxalic acid from acid solution. But it must be remembered that scandium oxalate, unlike the oxalates of the other rare earths, is very readily soluble in acids, and that it is impossible to precipitate with alkali oxalates owing to the formation of soluble oxalate complexes. Moreover, the large quantities of iron which the solution contains have a disturbing effect on the separation. When oxalic acid is added to concentrated ferric solutions, the ferric ion, Fe..., yields a complex ferric-oxalate ion, as shown by the colour change to green. Thus the solution uses up oxalic acid until all the iron is bound in the complex. At the same time the concentration of the H-ions is continually raised. When to the neutral or feebly acid solution of the wolframite oxides a certain quantity of oxalic acid is added, at first no precipitate is formed, but when the solution is warmed or it has stood for a long time, a precipitate appears owing to the formation of ferrous ions by autoreduction, the latter being gradually separated as difficultly |