ISSN:0003-2654    

DOI:10.1039/AN95681FX009

出版商:RSC    

年代:1956

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN95681FP011

出版商:RSC    

年代:1956

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN95681BP015

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

FEBRUARY, 1956 THE ANALYST Vol. 81, No. 959 Safety in Numbers IT is now about 30 years since analysts involved in various aspects of food and pharma- ceutical technology took what some of their colleagues appeared to regard as a bold, not to say a reckless, step. The current methods of biological assay, into which they had been more or less pitchforked because there were no other victims handy and-after all-measuring a vitamin in a rat was, although admittedly a p i s aller, only an alternative to weighing it on a balance, soon produced a state of profoiind disquiet in their analytical consciences. Did the results they were getting really mean anything? Could they be made to mean anything? Above all, how could they find out whether they did or could mean anything? Some one-and it would probably be neither rewarding nor helpful to try and find out who it was-noted the incursion of statistical concepts and methods of design in the somewhat remote, at any rate to most analysts, discipline of pharmacology. It occurred to him that the scientific principles applicable to measuring effects on animals might be what he was looking for in his efforts to assess amounts of, say, the antirachitic vitamin, at that time recognised but not isolated. He, or she, began to talk meaningly of variances and correlation coefficients. It was not long before a dozen or more laboratories began to apply to their results the critical probe afforded by the new, or newly realised, technique. Revision of experimental design to meet the stringent requirements of significance tests became at one and the same time the curse and the blessing of those engaged in what was coming to be known as bio-assay.The method of least squares supplanted the ruler, and the mean deviation gave way to the standard error. I t is not necessary to dwell at any length on the importance of this newly developed branch of “analytical chemistry” or, as the first Chairman of our own Biological Methods Group, in his introductory address, urged us rather to call it, “chemical analysis.” The existence of the Group bears witness to the Society’s view that here were scientific procedures worthy of detailed discussion by specialists, in the same way as physical methods and micro- analysis. The many stimulating meetings organised by the Group are a sufficient justification of the Society’s attitude.Yet there is another less obvious benefit to be put to the credit of the statistical developments of bio-assay, in general, and perhaps to some extent also to the biological Methods Group in particular. For the analyst turned bio-assayist, particularly if he occupied an industrial post, did not have his being in even a partial vacuum. He was in daily contact with other chemists, engaged in production, analysis and research. His new approach in due course had its effect on the others’ outlooks. Probability levels and fiducial limits eventually came to mean something to them also. It was inevitable for them, for some of them, anyhow, to ask themselves about their colleague’s new technique whether there might not be, as the saying is, “something in it for us.” Nor, it must be allowed, was his influence operating entirely unaided.A tiny minority of technologists who were involved in such things as the sampling of coal and the quality oE electric-light bulbs had already seen the light of the early statistical dawn. Whatever the relative strengths of these two influences, it was not long beiore the growing demand for knowledge of how to apply statistical concepts and methods to control and development work began to be 17 €4 1-jy lectures, monographs and large books of imposing proportions. Since the war thew Lave been several pblished in English, on both sides of the Atlantic, in which the theory and practice of statistical design and analysis have been expounded for the specific 73 The thing caught on. Some of its rays must also have reached the chemist.74 PROCEEDINGS Pol. 81 assistance of those engaged in the development and control of industrial processes. The non-biological analyst is among the beneficiaries, and few investigations into quantitative analysis, especially the increasing number con'ducted on a collaborative basis, are to-day planned without due consideration of statistical aspects from the outset. It is often said, perhaps a little by way of a trite generalisation, that advances in any one field of chemistry can help those working ir, other fields. In the mid-twentieth century the use of statistical weapons over the whole of applied chemistry, from bio-assay to plant design, is surely an outstanding example of how a benefit to one may become a benefit to all. A. L. BACIIARACH

ISSN:0003-2654    

DOI:10.1039/AN9568100073

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

74 PROCEEDINGS Pol. 81 PROCEEDINGS OF THE SOCIETY FOR ANALYTICAL CHEMISTRY ORDINARY IMEETING , 4 ~ Ordinary Meeting of the Society, organised by the Microchemistry Group, was held at 7.15 p.m. on Friday, January 27th, 1956, in the meeting room of the Chemical Society, Burlington House, London, W.1. The Chair was taken by the President, Dr. K. A. Williams, F.R.I.C., A.Inst .P., M.Inst.Pet. The following papers were presented and discussed : “Microchemical Methods in the Art Gallery and Museum,” by A. E. Werner, MA., M.Sc., D.Phil., A.R.I.C. ; “The Ring-oven Technique and its Application in Archaeology, ” by H. Weisz, Dr.techn.Dipl.-Ing. NEW MEMBERS ORDINARY MEMBERS Frederick Culkin, R.Sc. (Lond.), A.R.I.C.; Albert Arthur Henly, B.Sc., PhD. (Lond.), F.R.I.C.; Douglas William James, B.Sc.(Lond .), A.R.C.S. ; Roger Millward; Derek Ivor Stansell, H.Sc. (Lond.), A.R.I.C.; Catherine Hannah Tinker, B.Sc., Ph.D. (Lond.), A.R.I.C. JUNIOR MEMBERS ,41an William Archer, A.R.I.C. ; Alan Herbert HIelman, A.R.I.C. SCOTTISH SECTION THE Twenty-first Annual General Meeting of the Section was held at 5 p.m. on Friday, January 20th, 1956, at the Rhul Restaurant, 123 Sauchiehall Street, Glasgow. The President, Dr. K. A. Williams, F.R.I.C., A.Inst.P., M.Inst.Pet., the Hon. Assistant Secretary, Dr. R. E. Stuckey, F.R.I.C., F.P.S., and 9 founder members were among the 32 present. The Chairman of the Section, Dr. F. J. Elliott, F.R.I.C., presided. The following office bearers were elected for the forthcoming year :-Chairman-Dr. F. J. Elliot. Vice-Chairman-Dr. Magnus Pyke.Non. Secretary and I’reaswer-Mr. J. A. Eggleston, Boots Pure Drug Co. Ltd., Airdrie Works, Airdrie, Lanarkshire. Members of Committee-Messrs. J. Brooks, J. W. Gray, J. B. Headridge, R. L. Mitchell, R. S. Watson and W. Wilson. Messrs. J. Andrews and J . McL. Malcolm were appointed as Hon. Auditors. On behalf of Council and the whole Society, Dr. Williams congratulated the Section on completing its twenty-first year. The Annual General Meeting was followed at 7.30 p.m. by a film evening, at which the following films were shown : “The Technique of Sampling,” “Gravimetric Analysis” Parts I-IV, “Care of Laboratory Glass-ware,” “Fractional Distillation (Micro)” and “Principles of Chromatography” (colour) . MIDLANDS SECTION AN Ordinary Meeting of the Section was held at ‘7 p.m.on Wednesday, January l l t h , 1956, in the Main Chemistry Theatre, Department of Chemistry, The University, Edgbaston, Birmingham, 15. The Chair was taken by the Chairman of the Section, Mr. J. K. Leech, J.P. The following papers were presented and discussed: “Gas Chromatography,” by J. C. Tatlow, Ph.D., D.Sc., A.R.I.C.; “Ionophoresis” by A. B. Foster, B.Sc., Ph.D. THE First Annual General Meeting of the Section was held at 7 p.m. on Tuesday, Januarjv 24th, 1956, in the Sale Room, Regent House, St. F’hilip’s Place, Birmingham. The ChairmanFeb. 19561 PROCEEDINGS 75 of the Section, Mr. J. R. Leech, J.P., presided. The following appointments were made for the ensuing year :--Chairmm-Mr. J. R. Leech. Vice-Chairman-Dr. R. Belcher.Hun. Secretary-Mr. G. W. Cherry, 48 George Frederick Road, Sutton Coldfield, Warwicks. HOB. Treasuver-Mr. F. C. J. Poulton. Members of Committee-Mr. R. Adkins, Dr. Bella B. Bauminger, Messrs. H. J. G. Challis, F. P. Everett, S. H. Jenkins, C. A. Johnson, H. C. Smith, W. H. Stephenson and T. S. West. Mr. H. J. Alcock and Miss M. E. Tunnicliffe were re-appointed as Hon. Auditors. The Annual General Meeting was followed by an Ordinary Meeting of the Section at which a paper on “Ultra-micro Methods for the Analysis of Organic Compounds” was given by T. S. West, R.Sc., Ph.D., A.R.I.C. AN Ordinary Meeting of the Section was held at 7 p.m. on Tuesday, February 7th, 1956, in the Mason Lecture Theatre, The University, Edmund Street, Birmingham, 3. The Chair was taken by the Chairman of the Section, Mr.J. R. Leech, J.P. The following papers were presented and discussed: “The Analytical Chemistry oi Germanium ” by H. J. Cluley, M.Sc., F.R.I.C. ; ‘‘ The Analytical Chemistry of Gallium,” by G. W. C. Milner, X.Sc., F.R.I.C., A.1nst.P. &IICROCHEMISTRY GROUP THE Twelfth Annual Gei:erd Meeting of the Group was held at 6.45 p.m. on Friday, Januarj- %7th, 1966, in the meeting room of the Chemical Society, Burlington House, Piccadilly. London, ?V.l. In the absence of the Chairman of the Group, the Vice-chairman, Mr. D. I?. Phillips, A.R.I.C., presided. The following Officers and Conunittee Members were elected for the forthcoming year :-Chairman-Dr. G. F. Hodsman. I’ice-C~~ainnan-Mr. D. F. Phillips. Hoiz. Secretary-Mr. D. W. Wilson, Sir John Cass College, Jewry Street, Aldgate, London, E.C.3.Members u f Committee-Messrs, P. R. W. Baker, R. Belcher, R. A. Chalmers, F. Holmes, M. A. Fill and A. M. Ward. Dr. L. H. K. Cooper and Mr. €1. Childs were re-appointed as Hon. Auditors. The Annual General Meetin5 was followed, at 7.15 p.m., by an Ordinary Meeting of the Society, organised by the Microchemistry Group, at which the President, Dr. K. A. Williams, F.R.I.C., A.Inst .P., M.Inst .Pet ., presided. THE following are summaries of the papers presented at the Ordinary Meeting of the Society organised by the Biological Methods Group on Wednesday, November 2nd, 1955, in London. A first report appeared in The L4~za&st, 1955, 80, 781. The subject of the meeting was “The Evaluation of Anti-fungals,” and the following- papers were presented and discussed : “Laboratory Evaluation of Drugs for Clinical Trial Against Dermatomycoses,” by H.0. J. Collier, B.A., Ph.D., M.I.Biol., and G. K. A. Smith, A.I.M.L.T. ; “Some Factors in the Planning of Fungitoxicity Experiments in the Laboratory,” by R. J. W. Byrde, B.Sc., Ph.D., and G. M. Clarke, M.A., DipStat. ; “Cattle Ringworm: Problems in the Evaluation of Treatment,” by K. C. Sellers, Ph.D., B.Sc., D.V.S.M., M.R.C.V.S. Hon. Treasurer-Mr. G. Ingram. SUMMARIES OF PAPERS PRESENTED AT MEEIINGS OF THE SOCIETY LABORATORY EVALUATION OF DRUGS FOR CLINICAL TRIAL AGAINST DERMATOMYCOSES MR. SMITH said that in the evaluation of potential remedies against dermatophytes, skin infections of laboratory animals seldom appeared to have been used.Nevertheless, for this purpose both natural and experimental infections could be considered. Natural fungal infections occurred in laboratory mice, their existence being sometimes detected by the appearance of ringworm among laboratory staff. In more severe epidemics among mice, such as those described by B. H. Booth (Arch. Dermat. Syph., N.Y., 1952, 66, 65), skin lesions or loss of hair were obvious. Such natural infections of mice might be adaptable for chemotherapeutic trial ; but adequate methods of maintenance and control of the infection were first required. Experimental infections of fungi could be obtained in guinea-pigs and rabbits by rubbing fungal cultures into the shaved and scarified skin. E. D. De Lamater and R. W. Benham (1. Iwuest.Dermat., 1938, 1, 451) described the production of such infections and summarised the earlier literature. Within 4 to 6 days of inoculation, virulent strains of Trichophyton caused skin lesions that were active for about a fortnightPROCEBDINGS [Vol. 81 and then healed spontaneously. This type of infection offered opportunities for chemo- therapeutic experiments ; but at present the vast majority of tests reported in the literature were of the in vitro type. Without in vivo therapeutic methods, evaluation of potential remedies rested on a variety of measurements, including those of an vitro antifungal activity and of systemic and skin toxicity. For in vitro tests, plate methods had been used, for example by H. I. Chinn, R. B. Mitchell and A. C. Arnold ( J .Invest. Dermat., 1953, 20, 177); but methods of dilution in liquid media appeared to be more usual and to have certain advantages. They could, for example, be readily adapted to testing antagonism by materials such as hair and to exploring fungicidal activity. Studies of fungicidal action could not be relied upon unless a potent antagonist of the drug under investigation was incorporated in the medium of sub-cultnres. In the bisquaternary ammonium series examined by the present authors (H. 0. J. Collier, M. D. Potter, E. P. Taylor and G. K. A. Smith, Brit. J . Pharmacol., 1955 10, 343), prolongation of exposure to drug increased fungicidal and reduced fungistat ic effects. In addition to studies of systemic toxicity, the reactions of skin to intradermal injections and repeated application of drugs should be investigated.For both these purposes, the shaved skins of rabbits and guinea-pigs could be used. Hairless mice were also being studied as a means of assessing the toxicity of drugs applied repeatedli- to skin. Knowledge of the penetration of antifungal substances into keratin and hair follicles might also contribute to their evaluation ; but at present, except with coloured or fluorescent substances, penetration was difficult to assess experimentally. It was possible that radio- active methods might later be applicable. A. J. E. Barlow and F. W. ChattawaS. ( J . Invest. Dermat., 1955, 24, 65) have shown that treatment of hairs with keratolytic agents rendered them more susceptible to invasion by dermatophytes. While this observation threw doubt on the therapeutic value of keratolytic agents in antifungal preparations, it suggested a method of enhancing virulence of fungi in experimental animals by the application of keratolytic agents before the inoculation of the skin.SOME FACTORS IN THE PLANNING OF FUNGITOXICITY EXPERIMENTS IN THE LABORATORY DR. BYRDE said that an antifungal assay system might be regarded, very simply, as consisting of three components-fungus, medium and toxicant-each of which would contribute to the over-all variability of the results. The fungus inoculum should be obtained by standard methods; methods for spore suspensions had been studied in detail by S. 15. A. McCallan and F. Wilcoxon (Contr. B o p Thompson Inst., 1939, 11, 5). In the medium, controlled hydrogen-ion concentration was important, particularly in tests of weak electrolytes. In general, such compounds were most active in the undissociated form, owing to greater ease of penetration of the lipophilic cell membrane.For valid comparisons of inherent toxicity, they should be tested at a pH level at which they were predominantly undissociated. This could be achieved by buffering the medium 2 pH units below pK (for weak acids) or 2 yH units above pK (for weak bases). If, however, it was desired to apply the results of a test to a specific in vivo system, it was preferable to work at the pH of that system, as did E. W. Simon and G. E. Black- man (“Report of Third Symposium of the Society of Experimental Biology, 1949,” p. 253). In general, the medium should be the simplest that would permit normal growth of the fungus, and, if possible, should consist of specific compounds, e g ., asparagin, rather than biological extracts. The performance of a fungicide may be profoundly modified by its physical form, as described by G. L. McNew and R. P. Burchfield (Contr. Boyce Thompson Insf., 1950, 16, 163). That the presence of a mutual solvent may enhance fungistatic activity wak shown by R. J. W. Byrde and D. Woodcock (Ann. Appl. Biol., 1953, 40, 675). The assay of volatile fungicides presented a special problem, owing to the decreasing concentration in the medium, and results from such experiments should be interpreted with caution.Feb., 19561 PROCEEDINGS 77 MR. CLARKE said that in laboratory comparisons of the effectiveness of fungicides. carried out in incubators, sources of variation in growth of fungal material were common characteristics of the incubators themselves independently of the applied fungicide treatments: vertical trends were often present, growth being better at the top, and hori- zontal trends might also exist.It was also useful to examine whether the replicates of a particular treatment showed any trend due to the order of pouring out of the applied fungicide from its stock preparation. Work along these lines has been described by G. V. Coles, J, T. Martin, and R. J. MT. Byrde (Ann. Rep. Long Ashton Res. S~Q., 1954) and by G. M. Clarke (A$PZ. Statistics, 1955,4,199). Where there was not scope for- eliminating all these effects, a randomised block design might be used, each vertical layer in the incubator forming one complete block.In order to compare the results of several experiments, a standard substance must be included in each run, and would vary slightly in its effect; but the other compounds in each trial could be compared with this standard, and hence indirectly with compounds in other trials. Some fungi had much higher natural variabilities than others, and so if a choice existed, a less variable organism was chosen to make experimental results as precise as possible. Moreover, it was not possible to compare two fungi of different variances by means of a single analysis of variance, nor therefore always directly to compare the effect of a fungicide on two different fungi. CATTLE RINGWORM: PROBLEMS IN THE EVALUATION OF TREATMENT DR. SELLERS said recent work by G.C. Ainsworth and P. K. C. Austwick {Yet. Kec., 1955, 67, 99) and by K. C. Sellers, C. J. La Touche and W. B. V. Sinclair (in preparation) had shown that Tricho$hyton verrucosunt Rodin var. discoides Georg was the most common cause of cattle ringworm in this country. In Yorkshire the infection was found on a high proportion of farms, more commonly in calves in their first year of life and during the winter. Humans contracted the infection, but no accurate estimate of the incidence in this species had been made. As a preliminary to work on the control and treatment of cattle ringworm in the field, experimental observations had been carried out at the Veterinary Investigation Laboratory, Ministry of Agriculture, Leeds.and the Skin Department, Leeds General Infirmary. The results showed that healthy calves could be infected experimentally with material from natural cases of ringworm or with culture suspensions. Histological examinations of experimentally induced and naturally occurring lesion were made at various stages during the course of the disease. These surveys and experimental studies had shown that, although satisfactory trials might be performed in the artificially infected calf, the final and accurate assessments of an antifungal under field conditions presented many problems, The remedy, for use under farm conditions, must be easy t o apply and persistent in action. It must also penetrate through the crust to the depth of the infected hair and follicle. and must not damage the hide nor be toxic to the calf. Other factors such as the age of the lesion and the state of nutrition of the animal may influence an animal’s response to treatment. Consideration should also be given to the use of fungicidal or fungistatic paint or sprays for dressing farm buildings where reservoirs of infection might persist. Nevertheless, cattle ringworm lent itself to well planned clinical trials; not only was it widespread, but ethical and cosmetic considerations did not bulk so largely as in man. Also, it was possible that the results so gained would be applicable to the ring- worms of other species.

ISSN:0003-2654    

DOI:10.1039/AN9568100074

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

OBITUARY Fd. 81 0 bituar y OSMAN JONES OSMAN JONES died in hospital at Oxford on June 22nd, 1955, after a very brief illness. It cxme as a great shock to his friends and col1e;tgues as, feeling the strain of overwork he had entered the hospital only a few days previously for observation. Born in 1886, Osman Jones was articled to J. Kear-Colwell, Public Analyst for St. Pancras, Holborn, Bedford and Luton in 1903. During this period he attended the City and Guilds of London Institute when Sylvanus Thompson was the Principal and Meldola the Professor in the Department of Technical Chemistry, and gained the Diploma in this Department in 1908. In 1912 he was appointed Chief Chemist to the English Margarine Works, Broad Green, Liverpool, where he remained until 1918. He was then appointed Chief Chemist to Messrs.C. & T. Harris (Calne) Ltd. and began his long association with the meat industry, which was in effect his life’s ~70rk. During this period he was responsible for inauguration of the By-products Factory and also the Water Softening Plant, giving to both projects his usual unbounding enthusiasm and energy. He remained at Calne until 1946, when he was appointed Scientific Advisor to Messrs. Lovell and Christmas, and was responsible for organising the St. John’s Laboratory, of which he was in control until his death. During his long career Osman Jones was actively connected with many professional and learned societies. Elected a fellow of the Royal Institute of Chemistry in 1922, he served on the Council from 1943 to 1946 and was Streatfeild Memorial Lecturer in 1944, choosing as his subject “Modern Methods of Food Preservation.” It will be remembered that Streatfeild was a Lecturer at the Finsbury Technical College during his period as a student.He was a member also of the other two chartered chemical societies, as well as the Society for Analytical Chemistry, the Society for General Microbiology, the Society for Applied Bacteriology, the American Chemical Society, the Institute of Refrigeration and the Institute of Meat. Active membership of these many bodies does not complete the story of Osman Jones’s work for the advancement of Food Science nor his interest in the social aspects of his life’s work. A staunch member of the Chemical Club, he was for many years a member of its Executive Committee. I n 1937 he was appointed Chairman of the Neat and Fish Products Panel of the British Food Manufacturers Research Association.He was an active member of the Meat Extract Sub-committee of the Analytical Methods Committee and had accepted membership of the newly formed Meat Products Sub-committee. Osman Jones will long be remembered as a leading British expert on the technology of meat and bacon, a reputation that he also held in Continental countries, especially Holland, which he visited frequently in the course of his duties. His book “Canning Practice and Control,” first published in 1937 and since issued in two revised editions, is as well known to food technologists as are the many articles that he wrote for the technical press, often on very controversial subjects on which he held strong and, not infrequently, individualistic views. Little need be written on his skill as a food chemist nor on the broadness and originality of his outlook, as these are only too well known by his colleagues and friends. Equally well known was his readiness and eagerness to be of assistance even when pressure of a very full life made this seem well nigh impossible. Osman Jones’s home life was of the happiest. His chief hobby was gardening, especially rose growing; he had a wonderful collection. He is survived by his wife and three married daughters. to whom and to his grandchildren he was devoted. H. G. REES

ISSN:0003-2654    

DOI:10.1039/AN9568100078

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

Feb. 19SSj HUNTER AYD MILLEK 79 The Separation of Zinc from Some Other Elements by Means of Anion Exchange and Solvent Extraction, and its Titrimetric Determination with Disodium E thylenediaminetetra-ace tate BY JOHN A. HUNTER* AND CHRISTINA C. NILLEK (Presented at the meeting of the Scottish Section on Thursday, Ajb~il 2Sth, 1955) Previous work on the separation of zinc from other elements has been extended to include cadmium, gallium and indium. Zinc is adsorbed from 2 N hydrochloric acid solution by the anion-exchange resin, Amberlite IRA-400(Cl). Cadmium is then fixed on the resin by means of hydriodic acid and zinc is eluted with water and 0-25 N nitric acid. Much tin, some molybdeniim, a little indium, lead, platinum, titanium and tungsten (as phosphotungstate) , and a very little antimony, copper, gallium, ironlI1, manganese, thallium, thorium, uranium and zirconium may accompany zinc.Pre-treatment of the resin with methylarsonic acid, and addition of some to the sample solution before transfer to the resin, greatly reduce the amount of tin accompanying zinc. Quantities of zinc ranging from 0.1 to 5 mg are titrated with disodium ethylenediaminetetra-acetate solution, Solochrome black being used as indicator in a spectrophotornetric determination of the end-point. The small amounts of all the above-named elements accompanying zinc, except gallium, indium, titanium and uranium, can be tolerated. These four elements, together with lead and tin, can be eliminated by chloroform extraction of zinc as zinc pyridine thiocyanate from the eluates collected from the resin.The zinc, together with copper, manganese and ironII, is withdrawn from the chloroform into an ammoniacal solution and titrated as usual. For the determination of 5-mg and 0.5-mg amounts of zinc in mixtures in which the total cation content may be 100 mg, the results show no sig- nificant systematic error. The standard deviations are 15 and 8 pg, respec- tively. Results are given for a few alloys and a glass. RECENTLY,' we showed that the adsorption of zinc by a strong-base anion-exchange resin, from a solution 2 N in hydrochloric acid,2 could be utilised for separating it from aluminium, beryllium, calcium, chromium, cobalt, copper, iron, magnesium, manganese, nickel, thorium, titanium, uranium and zirconium, which show no, or only slight, affinity for the resin under the specified conditions.Antimony, bismuth, cadmium, tin and a little indium and lead accompanied zinc on the resin and, when the zinc was eluted by means of water and dilute nitric acid, all the cadmium and indium, a considerable proportion of the tin, some antimony, a little bismuth and lead, and traces of copper and iron, and perhaps of certain other elements, were simultaneously eluted. In experiments with 5 to 50 mg of zinc in synthetic solutions, or in alloys containing a maximum of 100 mg of the above-mentioned elements, excluding cadmium and indium, zinc, after separation by anion exchange, was determined gravi- metrically by means of 8-hydroxyquinoline, under conditions in which all potential inter- ference, save from a trace of copper, was suppressed.In a few experiments involving micro- techniques, the quantitative separation of 0.5 mg of zinc from 100mg of aluminium or magnesium was achieved. Recently, Rush and Yoe3 have used a similar anion-exchange process for separating 0.02 to 1 per cent. of zinc in a few materialsbefore its colorimetric determination, The general procedure of our earlier experiments has now been extended to cover the quantitative separation of zinc and cadmium, and also the separation and determination of 0.6-mg amounts of zinc in association with many other elements besides aluminium and magnesium. For the determination of 5 to 50 mg of zinc, initially associated with cadmium Present address : United Kingdom -4tomic Energy Authority, Atomic Weapons Research Estab- lishment, Aldermaston, Rerlrs.80 HUNTER AND MILLER: THE SEPARATION OF [Vol.81 and with other elements whose behaviour might be influenced by the introduction of an additional step in the anion-exchange separation, the gravimetric method with 8-hydroxy- quinoline has, for convenience, been retained. The use of the disodium salt of ethylene- diaminetetra-acetic acid for the titrimetric determination of 0-5-mg to 5-mg quantities of zinc has been comprehensively studied. Adequaie accuracy and precision have been obtained by determining end-points spectrophotometrically, with Solochrome black as indicator. EXPERIMENTAL SEPARATION OF 5 TO 50mg OF ZINC FROM CADMIUM BY MEANS OF ANION EXCHANGE- The previously described method for the adsorption of zinc by the chloride form of the strong-base anion-exchange resin, Amberlite IRA-400, was equally applicable to cadmium.The simplicity and effectiveness of the separation of zinc from many important elements was such that a means was sought of treating the resin on which zinc and cadmium were held, so as to prevent their simultaneous elution, and thus permit the ready determination of zinc. Concentrated hydrochloric acid, fluorides, iodides and other reagents were tested for this purpose and hydriodic acid was found to be the most suitable. Recently, Raggott and Willcocks* described a procedure for separa-ting a small amount of zinc from cadmium metal, which depended on the formation of an anionic complex of iodide and cadmium, but not of zinc, in a solution containing sulphates, which was then passed through an appropriately prepared column of the strong-base anion-exchange resin, De-Acidite FF.The cadmium was adsorbed and the zinc passed through. Their method would be unsuitable for solutions containing the large assortment of elements under consideration in this paper. After the 2 N hydrochloric acid solution containing the various elements was transferred to the column and washing with 2 N hydrochloric acid was completed, 5 ml of 2 N hydriodic acid were added to the column to give complete retention of 100 mg of cadmium. If, instead, 2 ml were used, 0.5 mg of cadmium was subsequently found with the zinc. The effluent was collected and the column was further treated with 15 ml of water and 40 or 60 ml of 0.25 K nitric acid; 60 ml led to the recovery of slightly more zinc.In the combined effluents zinc was determined gravimetrically by means of 8-hydroxyquinoline. In order to avoid possible interaction of liberated iodine with the latter, and resultant interference with the determina- tion of zinc, the effluents were treated with a small excess of nitric acid, and cautiously evaporated to break down iodides and expel iodine, before the precipitation of zinc 8-hydroxy- quinolinate was effected as previously described. Of the other elements that were hitherto found to be mainly or completely adsorbed by the resin from 2 N hydrochloric acid solution and partly eluted by water and nitric acid, 100mg of bismuth and 50 of antimonyIII were now found to yield none to the eluate and 50mg of antimonyV yielded 500pg.The amount of tin escaping from 100mg increased from 20 to 30 mg and 1 mg of lead reached the eluate. The behaviour of some other elements, believed not to be significantly adsorbed by the resin, was more closely examined. It was found that 100 mg of chromiumlIT, manganeseII, thorium, uraniumv1 and zirconium yielded none, 20, 10, 100 and 100 pg, respectively, to the hlydriodic acid - nitric acid solution. Indium and titaniumIv, which were slightly adsorbed, yielded 3 and 2 mg, respectively. Of other elements whose behaviour had not hitherto been examined, and of which 50-mg amounts were now tested, gold, mercuryII, platinum’” and thalliumIII were completely held initially on the resin, but 2 mg of platinum and 50 pg of thallium as T1I were removed in the hydriodic acid - water - nitric acid treatment.were not significantly adsorbed. The bulk of the gallium and much of the molybdenumV1, telluriumIV and tungsten as phosphotungstate were found in the hydrochloric acid effluent. The amounts in the hydriodic acid - nitric acid effluents were approximately 0.1, 8, 5 and 2 mg, respectively. Results for the separation of zinc from various metals, in experiments in which the iodide treatment of the resin was used, are shown in Table I. The zinc 8-hydroxyquinolinate, from the experiments in which 5 mg of zinc were separated from 100 mg of cadmium, was tested for cadmium with 4-nitronaphtha1enediar:oaminoazobenzene (cadion 2B), and none (less than 5 pg) was detected. SEPARATION OF ZINC FROM AN EXCESS OF TIN- The introduction of more tin into the zinc-containing eluate from the resin column was a disadvantage of the iodide treatment. Although some tin could be eliminated as stannic halides in preparing the solution of the sample, it was preferred to try to form a SeleniumIv and vanadiumIV andFeb., 19561 ZINC FROM SOME OTHER ELEMENTS 81 complex that would be washed out of the resin column with the hydrochloric acid.The following simple procedure involving the use of methylarsonic acid , with which considerable complexing apparently occurs in 2 N hydrochloric acid solution, even though no precipita- tion takes place,5 was eventually adopted. The resin was treated with 5 ml of a 5 per cent. solution of anhydrous disodium methylarsonate, 2 N in free hydrochloric acid, and 400 mg of the methylarsonate were also added to the solution containing tin, in which the concen- tration of free hydrochloric acid was similarly made 2 N .After transfer of the solution to the column and washing with 2 N hydrochloric acid as usual, only a small amount of tin remained from 50 mg initially taken. With 100 mg, the subsequent treatment with hydriodic acid caused darkening of the resin, indicative of the presence of some stannic iodide, and 3 mg of tin were eluted. In experiments in which 50 and 5mg of zinc were separated from 50 and 100mg of tin, respectively, and determined in acid solution with 8-hydroxyquinoline, the errors in duplicate determinations of the larger amount of zinc were -0.27 and -0.24 mg, as compared TABLE I DETERMINATION OF ZINC ALONE AND IN MIXTURES AFTER ADSORPTION ON, AND ELUTION FROM, THE ANION-EXCHANGE RESIN AMBERLITE IRA-400(Cl) Approxi- mate weight of zinc taken, 50 20 5 10-8 g 5.0 Approxi- mate weight of Other other metals metals taken, 10-8 g Error, 10-6 g -9,* -11,* 0, $2 -6,* -4* +51* +3* -27,* -13 -2,* f 3 +2.* + l o + 19 + 13§ - 253 - 145 Approxi- mate weight of zinc taken, 10-8 g 506} 5 : } ?} 5 Approxi- mate weight of Other other metals metals taken, 10-8 g Bi { !!: cu {I:: Error, 10-6 g - 1 3 t j + 14t + -115 -19 .24,$ -108 * Only 40 ml of 0.25 N nitric acid were used.t Corrected for contaminating iron. fj Precipitation with 8-hydroxyquinoline in alkaline tartrate solution. Corrected for contaminating copper. with an error of -064mg in earlier experiments without methylarsonic acid, and with precipitation of zinc 8-hydroxyquinolinate in alkaline tartrate medium.The precipitates from the smaller amounts of zinc were difficult to filter off, because of contamination by hydrated stannic oxide, but virtually correct results for zinc were obtained by dissolving the precipitates and weighing the contaminant. The presence of stannic oxide was attributed to the conversion of a-stannic compounds into p-compounds, a process induced by nitric acid, long standing and heating. When a-stannic chloride was put on a resin column and 200 mg of tartaric acid were placed in the receiver for the usual eluate, the conversion was greatly delayed. Should the eluate contain iodides, however, the oxidation to eliminate it would counteract the effect of tartrate, and if iodide were absent, the presence of tartaric acid might necessitate the application of the less desirable precipitation of zinc with 8-hydroxyquinoline in alkaline tartrate solution.DETERMINATION OF SMALL AMOUNTS OF ZINC BY MEANS OF DISODIUM ETHYLENEDIAMINETETRA- Before studying further the separation of small amounts of zinc from 100 mg of other elements, we sought a simpler means of determining quantities ranging from 5 to 0-5 mg or less, in the eluates obtained from resin columns. The versatility of disodium ethylene- diaminetetra-acetate as a reagent and its value in the determination of low concentrations of metal ions6 led to its early consideration as a substitute for 8-hydroxyquinoline. When hydriodic acid was used, the eluate (80 ml) from resin columns was 0.12, 0.12 and 0.19 N , respectively, in hydrochloric, hydriodic and nitric acids, and might contain in addition to zinc the maximum quantities of various elements mentioned on p.80. The amount of tin would be reduced to 3 mg if methylarsonic acid were used. In a solution containing ammonia ACETATE-82 HUNTER AND MILLER: 'THE SEPARATION OF [Vol. 81 and ammonium chloride, buffered to pH 10, zinc can be very satisfactorily determined directly with ethylenediaminetetra-acetate in the presence of Solochrome black as an indicator.' By this means, Debneys determined 6 pg of zinc in 2 ml of solution, and Blaedel and Knight,g using a high-frequency technique, determined 0.65 to 13 mg titrimetrically with a high degree of accuracy, in 80 to 100ml of solution.In titrations in which the colour change of Solochrome black was observed visually, it was found that, provided the rate of addition of the ethylenediaminetetra-acetate solution was adequately controlled, its zinc equivalent remained essentially constant over a range of 5 to 0.2 mg of zinc, whether the volume for tiltration was 100 ml or 25. End-points were more distinct in the larger volume, which was preferred. A small blank correction was made for titratable impurity in the reagents. Iodide and nitrate did not influence the results. This simple direct titration procedure would not be applicable in the analysis of complex materials, where small amounts of other elements would accompany zinc. 300 . 400 500 600 700 800 Wavelength, rnv Fig.1. Absorption spectra: curve A, zinc - Solochrome black An important means of improving the selectivity of ethylenediaminetetra-acetate as a reagent for zinc is to convert certain metals, e.g., cobalt, copper, mercury and nickel, and zinc into complex cyanides, which do not react with ethylenediaminetetra-acetate and Solochrome black. Various uncomplexed elements, including indium , lead and manganese, may then be titrated under appropriate conditions with the reagent. If formaldehydelO911,12 or chloral hydrate,13 which is easily obtained pure, be next added, the complex cyanide of zinc, but not those of the other elements cited, is broken down and excess of potassium cyanide destroyed, and the freed zinc can be titrated in the presence of Solochrome black as an indicator.IronIII, which forms at pH 9 to 10 a more stable complex with Solochrome black than it does with ethylenediaminetetra-acetate, binds the indicator and must be eliminated. It may be reduced in nitrate-containing solutions by ascorbic acid14 and effectively masked as ferr0~yanide.l~ Experiments were made to find the minimum amounts of potassium cyanide and chloral hydrate required to effect the desired reactions for 10 mg of zinc in about 100 ml of solution, such as might be obtained from resin columns. Acids were neutralised with sodium hydroxide. Various amounts of zinc were then determined tty the method evolved, a correction being made for the titre found in a blank run. The ethylenediaminetetra-acetate was standardised with reference to zinc in a solution containing only ammonia and ammonium chloride.The end-points obtained in solutions to which ascorbic acid, cyanide and chloral hydrate had been added were much less sharp than those obtained in simple buffered solutions, and, in the blank experiments, it was often difficult to decide whether or not titration was needed. Errors throughout were positive, being about 20 p g for 2 to 5 mg of zinc, but ranging from 4 to 77 pg for 0.5 mg. The results for blank runs varied erratically from 0 to 25 pg of zinc and under-estimation here probably accounted in part for positive errors elsewhere. It was decided to examine the end-point behaviour with the aid of a spectrophotometer. complex; curve B, Solochrome blackFeb., 19561 ZINC FROM SOME OTHER ELEMENTS 83 Spectrop kotometric evaluation of end- points-Sweetser and BrickeP recently determined zinc in an ammonia - ammonium chloride solution by titrating with ethylenediaminetetra- acetate and noting the change in optical density that occurred at a wavelength of 228 mp, where the reagent absorbed strongly as compared with zinc and its complex with the titrant.In this region of the spectrum, potassium cyanide and chloral hydrate rendered the solution opaque. The absorption spectra of Solochrome black and its zinc complex were therefore next measured in ammonia - ammonium chloride solutions buffered at pH 9.5, in order to find if titration in the presence of an indicator would be possible. Recently, Karsten and inches 3.9 inches .L c Fig. 2. Perspex titration cell and attachments his co-workers1' showed that magnesium could be titrated under comparable conditions a t a wavelength of either 530 or 630 mp.From the curves shown in Fig. 1 it is evident that there is sufficient difference in the optical densities at a wavelength of about 665 m p to permit end-point determination in the titration of zinc with ethylenediaminetetra-acetate. During the titration the optical density will change only slightly until the end-point is approached, when it will increase sharply, as a result of the liberation of the indicator from its zinc complex, and reach a limiting value when liberation is complete. End-points were obtainable after the application of potassium cyanide and chloral hydrate. As the volume for titration was expected t o be about 100m1, a Perspex cell (cf.Chalmersls), illustrated in Fig. 2, was made to fill the cell compartment of a Unicam SP500 spectrophotometer, and the compartment was fitted with a light-tight cover provided with two holes, one to accommodate a small electrically driven stirrer, the other for the insertion of the long fine tip of a microburette. Precautions were taken to exclude light at the apertures. When various amounts of zinc, contained in ammonia - ammonium chloride solutions, were titrated with ethylenediaminetetra-acetate and correction was made for the reagent blank (3 pg of zinc), the zinc-equivalents of 1 ml of the titrant were- Weight of zinc, pg . . .. .. 128 554 924 9961 Zinc-equivalent of titrant, pg per ml . . 166 167.9 168.0 168.084 HUNTER AND MILLER: THE SEPARATION OF 0.10 0-08 0.06 0.04 0.02 0 0.10 x Y .- 0.08 U V c .- Z.0.06 .- Y 2 0-04 d 0.02 0 Wol. 81 0.5 Volume of 0. Fig. 3. Titration curves for corresponding blanks (X) 1.0 1-5 0025 M EDTA, ml about 100 pg of zinc (Y) and for theFeb., 19561 ZINC FROM SOME OTHER ELEMENTS 86 The sharpness of the end-points in the determination of a small amount of zinc and in a blank experiment are shown in Fig. 3a. The above figures are to be compared with the value 167.5, averaged from four similar titrations in which 0-5-mg amounts of zinc were used and end-points were determined visually. When, however, solutions containing zinc and quantities of acids such as are obtained in eluates from resin columns were neutralised with sodium hydroxide, treated with ascorbic acid, cyanide and chloral hydrate and titrated spectrophotometrically, the errors found, after applying a blank correction of 114 pg of zinc, were -20, -62 and -58 pg on about 0-1, 1 and 10 mg of zinc, respectively. Typical end-point curves are shown in Fig.3b. Further tests proved that sodium hydroxide was responsible for the errors, probably because it contained a trace of impurity that was un- masked by cyanide and reacted with Solochrome black, adversely affecting the end-point behaviour for zinc. When acids were neutralised instead with ammonia, no difficulty was experienced in interpreting end-point behaviour, and the over-all error in determining 0.1 to 10 mg of zinc in a series of six experiments where the blank value was 55 pg of zinc was +_ 4pg.In such solutions, the optical density begins to change perceptibly much farther from the end-point than in a simple ammonia - ammonium chloride solution. Never- theless, the end-point break is quite sharp (Fig. 3c). In all subsequent work ammonia was used for neutralising acids before titrimetric determinations with ethylenediaminetetra- acetate. It was desirable to find to what extent visual determinations of end-points would be improved by the omission of sodium hydroxide in neutralisations. Series of experiments were performed in which acids were neutralised with ammonia, and treatment with ascorbic acid, cyanide and chloral hydrate was given. The results shown in Table I1 were obtained. A set with spectrophotometric determination of end-points is included for comparison.TABLE I1 TITRATION OF ZINC WITH ETHYLENEDIAMINETETRA-ACETATE, WITH SODIUM HYDROXIDE EXCLUDED AND THE END-POINTS (EXCEPT IN (v)) DETERMINED VISUALLY Approximate weight of Remarks zinc taken, mg 5 ml of 2 N hydrochloric acid present initially . . .. . . .. .. 1.5 Hydrochloric, nitric and hydriodic acids as from resin columns . . .. . . 1 to 1.3 As (ii) but all titrations done indepen- dently of each other . . .. .. 0.25 0.5 2 As (ii) but purer reagents , . .. 0.35 As (iv) but spectrophotometric titration 0.35 Blank correction calculated as zinc, Pg 5 21 f 5 24 4 & 4 31 f 0 Error, Pg $33Z $31J +27 -3, -15, -1, 3-1, $11, -1 -16, -6 -31, +11 4-5 +24, +38, +22, 4-24, 3-18 4-4, +71 d-1 It is evident that, in visual work, errors of the order of 30 pg are possible, although precision may be good when experiments by the titration procedure are compared with one another.Spectrophotometric determination of end-points gives superior results. INFLUENCE OF OTHER ELEMENTS ON THE DETERMINATION OF SMALL AMOUNTS OF ZINC BY Details concerning the amounts of various elements that could be present with zinc have been given on p. 80. In addition, 0.35-mg amounts of zinc in hydrochloric - hydriodic - nitric acid solutions were determined titrimetrically with ethylenediaminetetra-acetate by the full ascorbic acid - cyanide - chloral hydrate procedure in the presence of small amounts of some other elements whose effects were not known with certainty. Tungsten was present as phosphotungstate and tartrate was added to prevent the deposition of hydroxides of thorium and so on.The following results, corrected for the blank, are all within the limit of accuracy of the method- MEANS OF ETHYLENEDIAMINETETRA-ACETATE- Other element . . Mom WVI P t l V Th Zr TI1 Amount, mg .. 10 2-5 2.5 0-5 0.8 0.1 Error on zinc, pg . . - 5 -4 +l +2 -2 -8 91atinumIv was masked by cyanide.86 HUNTER AND MILLER: THE SEPARATION OF [Vol. 81 TelluriumIV was reduced to the element by ascorbic acid and would have to be removed. Tin, lead, manganese, gallium, indium, titanium and uranium required further consideration. None is inactivated by cyanide. FlaschkalS showed that lead could be directly titrated with ethylenediaminetetra-acetate at a pH of 8 to 10, if tartrate were present to prevent deposition of lead hydroxide.Elements such as zinc and copper could be masked with cyanide.20 Similar titrations of manganese21 and indium22 were done in hot solutions. I t was essential to prevent atmospheric oxidation of manganous ions by having ascorbic acid present. Kinnunen and M e r i k a n t ~ ~ ~ titrated manganes'e at ordinary temperature. Flaschka and Abdine24 were unable to titrate gallium under the same conditions; its Solochrome black complex was so weak that it was necessary to add an excess of ethylenediaminetetra-acetate and titrate back with another solution. This had to be done at a pH of 8 to 9, with a lead or a manganous solution, if there was masking of zinc, etc., by cyanide. Flaschka stated that tinIV did not form a complex with ethylenediaminetetra-acetate, but that it affected the indicator behaviour.Under the conditions of our experi- ments, the titration of lead was extremely slow, and heating of the solution was impracticable in the presence of tin and with a spectrophotometric determination of the end-point. Titanium and uranium blocked the indicator. If titrations were done quickly, immediately after the addition of the indicator, 0-5 rng of titanium and the amount of uranium left after 5 mg had been submitted to the ion-exchange procedure could be tolerated, but the maximum possible amounts could not. As tin, lead and manganese more frequently accompany zinc than do gallium, indium, titanium and uranium, an appropriate means of allowing for the first three was first sought. DETERMINATION OF SMALL AMOUNTS OF ZINC IN ASSOCIATION WITH TIN, LEAD AND The presence of 3 mg of tin ( c j .p. 81) corisiderably reduced the sharpness of the end- point in the determination of zinc, with Solochrome black as indicator. The colour-change was from lilac-grey to blue, making accurate visual detection of the end-point impossible. Spectrophotometrically, the change in optical density was one-third of the usual and spread over a larger volume of the titrant, but end-point determination was satisfactory. The presence of tartaric acid was essential to prevent deposition of stannic oxide, which caused loss of zinc. The temperature of conversion of iron into ferrocyanide had to be lowered from 70" to 40" C, and 20 minutes allowed, during which time no deposition of stannic oxide occurred.When solutions contain tin, there must be no delay in completing the experiments. No difficulty arose in titrating manganese directly in solutions containing tartrate, cyanide and ascorbic acid. With lead, the coloiir change at the end-point with Solochrome black was from purple to blue. The change in optical density at the wavelength of 665 mp was smaller than for zinc, but the end-point was sharp. Each small addition of titrant required, however, 5 minutes' waiting before a steady reading of optical density could be obtained, and a titration required 45 minutes. Titration was expedited by the addition of an excess of ethylenediaminetetra-acetate and ti tration of the unused reagent with a dilute standard solution of manganous chloride.Because of slight sluggishness in the latter reaction, it was desirable to have only a slight excess of ethylenediaminetetra-acetate, control of the addition of which was facilitated by adding to the solution beforehand about 100 pg of manganese", whose more distinct colour change with Solochrome black permitted the more rapid addition of ethylenediaminetetra-acetate in small excess by visual observation. Ascorbic acid had to be present in the solution to which manganous chloride was added and sufficient time allowed for reduction of any manganeseIII, which would oxidise the indicator. In the final titration with manganous chloride, spectrophotometrically at the usual wavelength, slight blocking of the indicator occurred, without affecting end-point determination, and it could no doubt be obviated by having a little ascorbic acid in the manganous chloride solution.After the titration, the slight excess of manganous chloride necessarily added could be determined graphically. Zinc was liberated from its complex cyanide and titrated together with the excess of manganous chloride. This triple titration procedure required 30 to 40 minutes. No adverse influence of stannic oxide was noted in experiments in which a little tin was present. The same result was obtained for 1OOpg of zinc in a solution containing lead and manganese, whether a slow direct titration of these was done or the more rapid back-titration procedure was used. When the latter method was applied to the determination of zinc Antimony caused no trouble. The addition of fluoride did not prevent their interference.MANGANESE-Feb., 19561 ZINC FROM SOME OTHER ELEMENTS 87 accompanied by 2 mg of lead in hydrochloric - hydriodic - nitric acid solutions such as would be obtained from resin columns, and the appropriate correction was applied for the control run, the following results were obtained- Weight of zinc taken, pg .. 464 533 5030 5144 Error, pg . . .. . . .. +l4 - 7 + 33 +6 If desired, the titration of manganese present in eluates from resin columns may also be done in this way. In general, when it is seen that titration is required and it is not known what element necessitates it, it is best to adopt the triple titration procedure. DETERMINATION OF SMALL AMOUNTS OF ZINC I N ASSOCIATION WITH GALLIUM, INDIUM, TITANIUM, URANIUM AND TIN- For the determination of a small amount of zinc originally present with a large amount of gallium, indium, titanium and uranium, a simple means of separating it from the small amounts of these elements that accompany it from the resin column was sought, in order that it might be determined finally with ethylenediaminetetra-acetate.Numerous solvent extraction procedures based on the use of thiocyanate, which have been used in recent years for effecting separations of zinc from various elements, were inadequate for the present purpose. The most promising results were obtained with pyridine and thiocyanate in conjunction with chloroform as solvent. In neutral solutions a number of bivalent metals, e.g., cadmium, cobalt, copper, nickel and zinc, form complex metal pyridine thiocyanates with pyridine and ammonium or potas- sium thiocyanate. Zinc pyridine thiocyanate, Zn(C,H,N),.(SCN),, is soluble in chloroform. Lead may form Pb(OH)SCN, which was found to be ins0luble.~5 Since the eluates from resin columns are liable to contain iodide, there is also the possibility of formation of complex pyridine iodides of, for example, zinc and antimony. Zinc pyridine iodide dissolved readily in chloroform. Antimony pyridine iodide, if extracted, would not upset the subsequent titration. The zinc from zinc pyridine thiocyanate dissolved in chloroform could be readily extracted into acid or alkaline solutions. Since acid combined with excess of pyridine dissolved in the chloroform, and pyridine salts might later influence the buffering of the solution for titration, an alkaline extractant was preferred.In acid solutions (80ml), neutralised to methyl red, to which ammonium thiocyanate and pyridine were added, the following results were obtained for single extractions with 10 ml of chloroform at 15" C; it was immaterial whether the original solution contained only hydrochloric acid or hydriodic and nitric acids as well- Volume of 2 M ammonium thiocyanate, ml 0.25 0.25 0-25 0-1 1 2 5 Volume of pyridine, ml . . .. . . 0.5 1 1.5 1 1 1 1 Percentage of 2 mg of zinc extracted . . 20 46 45 55 80 95 95 In the first experiment, the pH of the residual aqueous layer was 6-7, whereas in all other experiments it was at least 7 . Further experiments with 5-mg and 0.5-mg amounts of zinc, in which 2 ml of thiocyanate solution and 1 ml of pyridine were used, and two extractions with 10 ml of chloroform were made, with addition of an extra 0.5 ml of pyridine after the first extraction in order to ensure a sufficiently high pH value, gave nearly complete recovery of zinc.The efficiency of extrac- tion was reduced by tartrate, whose adverse influence was, however, counteracted by increasing the concentration of thiocyanate and making a third extraction. An extraction apparatus (Fig. 4a) designed for maximum efficiency was made. The chloroform layers were run into the receiver (Fig. 4b), in which they were shaken up with the alkaline extractant, which was afterwards withdrawn by suction through a capillary tube into a conical flask. Acid solutions to be submitted to solvent extraction were neutralised with ammonium hydroxide, just sufficient being added to change methyl red to yellow.An appreciable excess prevented the extraction of zinc pyridine thiocyanate. At this time all determinations of zinc were being made with visual observation of end- points. When 0-5-mg and 5-mg amounts of zinc in 80ml of solution were submitted to treatment with pyridine and thiocyanate and solvent extraction, and the zinc was recovered from the chloroform and determined, the errors noted, after the application of the appro- priate blank corrections, averaged +20 pg and -20 pg, respectively. When similar quanti- ties of zinc were carried through the full ion-exchange procedure with the use of hydriodic acid and methylarsonate, followed by chloroform extraction of zinc pyridine thiocyanate and so on, the corresponding errors were -13 and -60 pg.The results were as good as The amount of pyridine was presumably too small.88 HUNTER AND MILLER: THE SEPARATION OF [Vol. 81 could be expected with visual determination of end-points in titrations. With spectrophoto- metric titrations, the following results, corrected for the corresponding blank, were obtained for zinc that had been submitted to chlorofonn extraction of its pyridine thiocyanate complex- Weight of zinc taken, pg .. 372 344 1329 1347 4925 4937 *- - w Blank, pg . . .. * . .. 25 41 35 Error, pg . . .. .. .. +6 +8 +4 +4 -17 -21 The behaviour of other metals-When put through the solvent-extraction procedure, 5 mg of gallium, 2.5 mg of indium, titaniumIv, bismuth or antimony, and 1 mg of uraniumm imparted nothing to the chloroform.From 3mg of tinIV, the maximum amount to be expected if methylarsonic acid were used, no extraction was noted; 10 mg gave about 0-5 mg. But 2mg of lead imparted to the chloroform a. very small amount of some material that was readily titrated with ethylenediaminetetra-acetate in cyanide solution and therefore (a) (6) Fig. 4. Extraction. apparatus was not lead. With ironIII, some reduction to ironlI occurred, even when no ascorbic acid was used, and this, together with any manganese or copper present, would be extracted, but these elements are easily dealt with in the ensuing titration procedure. A few mixtures containing zinc were also put through the extraction procedure and zinc was determined with the following results, control experiments being made with zinc omitted and the appropriate corrections applied- Other metals, mg Sn(5), Pb(2) In(5), U(l), 'Ti(26) Sn(5), In(5), Ti(2-5), Pb(2), U(2-5) L--y--J v- \ V J Zinc taken, pg .. 432 444 342 355 Error, pg . . -3 -11 -11 -11 4975 1930 -32 -20 FINAL METHODS FOR THE DETERMINATION OF 0.5 TO 5mg OF ZINC IN MIXTURES APPARATUS- Use polythene bottles for ammonia, potassium cyanide and ethylenediaminetetra-acetate solutions and zinc-free glassware for other solutions and general operations, Erratic results may be encountered if the ethylenediaminetetra-acetate solutions are stored in glass. Calibrate volumetric ware. Columns for the resin have already been described,l and the solvent-extraction apparatus, which must be thoroughly cleaned before use, is shown in Fig.4.Feb., 19561 ZINC FROM SOME OTHER ELEMENTS 89 In this work, a Unicam SP500 spectrophotometer was used for the determination of end-points. An appropriate Perspex cell for titrations is illustrated in Fig. 2 and other necessary arrangements are described on p. 83. REAGENTS- Distilled water should not be from a copper still. Purify and distil hydriodic acid (store under carbon dioxide) and other acids, if necessary, and prepare ammonium hydroxide solutions from cylinder ammonia. Purify and distil pyridine and chloroform. Prepare disodium methylarsonateZs and heat it at 160" C in order to form the anhydrous salt. Ammonium thiocyanate solution, 2 M. Ammonium chloride solution-A 10 per cent.w/v aqueous solution. Ascorbic acid solution-A 1 per cent. w/v aqueous solution, freshly prepared. Potassium cyanide solution-A 5 per cent. w/v aqueous solution. Chloral hydrate solation-A 50 per cent. w/v aqueous solution. Disodium ethylenediaminetetra-acetate solutions, 0.008 and 0.0025 M. Hydrochloric acid, 2.0 N. Zinc chloride solzttion-Prepare a standard solution containing a slight excess of hydro- chloric acid, either from pure metallic zinc or from freshly ignited zinc oxide. Arrange the concentration so that aliquots containing 5 or 1.5 mg of zinc can be readily taken. Manganous chloride solution, 0.001 M. Solochrome black indicator solution-Extract 0.2 g of Solochrome black (British Drug Houses) thoroughly with small portions of warm pyridine, filter and make up to 100 ml with pyridine.The final solution is about 0.1 per cent. and, if kept in a well stoppered glass bottle, is usable for 3 months. Use reagents of recognised analytical grade. PROCEDURE- For the preparatiofi and use of columns of resin, consult our earlier paper.l The anion-exchange separation-If less than 10 mg of tin are present, apply to the resin 5 ml of a 2 N hydrochloric acid solution containing up to 5 mg of zinc and a maximum of 100 mg of metallic elements, and continue to add 2.0 N hydrochloric acid until 50 ml of solution are collected. Should the amount of tin exceed 10 mg, first treat the resin column and the sample with methylarsonate as described on p. 81, and then transfer the sample to the column and wash as usual with 2.0 N hydrochloric acid.In order to fix cadmium and bismuth on the resin, next apply 5 ml of 2 N hydriodic acid and collect the eluate in a vessel containing 200 mg of tartaric acid dissolved in a little water. Continue the elution with 15 rnl of water and 60 ml of 0-25 N nitric acid. If the solvent-extraction procedure is required, the receiver should be the extraction funnel (Fig. 4a). If the solution contains tin, proceed at once with the next operation. The formation and extraction of zinc pyridine thiocyanate, and separation from small amounts of gallium, indium, titanium, uranium, bismuth, tin and lead-If the original material contained gallium or indium, much titanium or uranium, or bismuth that was not fixed on the resin because the use of hydriodic acid was omitted, or it is desired to do a visual titration and the original material contained much tin, treat the eluate from the anion-exchange separation, which should be in the appropriate extraction funnel, as follows.If necessary, add sufficient ascorbic acid to remove free iodine, and make just alkaline to methyl red by means of ammonia. Add 5 ml of ammonium thiocyanate solution, mix and then add 1 ml of pyridine. Mix again by shaking, add 10 ml of chloroform and shake the mixture briskly for 1 minute. Allow the phases to settle and run off the chloroform layer into the back-titration tube (Fig. 4b) contain- ing 2 ml of 5 N ammonia, 0.05 ml of ascorbic acid solution, 2 ml of potassium cyanide solution and water to make 10ml. Repeat the extraction twice, adding to the aqueous layer in the funnel 0.5 ml of pyridine and 10 ml of solvent on each occasion. In order to recover the zinc from the organic solvent, shake the back-extraction tube briskly for 1 minute and, after the phases have separated, withdraw the aqueous layer as completely as possible, via a bent capillary with upward-pointing tip, into a conical flask containing 2 ml of ammonium chloride solution. Wash the residual chloroform thrice with 5-ml portions of water, shake, and collect the washings as before.Heat the aqueous solution to 75" C for 10 minutes in order to convert any ironU that may be present into ferrocyanide, and to remove a little entrained chloroform, which would attack the Perspex cell. Proceed as described on p. 91.90 HUNTER AND MILLER: THE SEPARATION OF [Vol.81 Standardisation of a9firoximately 0.0025 M disodiuln ethylenediaminetetra-acetate, with spectrophotometric determination of the end-$oinL-If a 10-ml microburette is to be used, measure into the titration vessel an aliquot of the standard zinc chloride solution containing about 1.5 mg of zinc and raise the hydrochloric acid content to 5 ml of 2 N . Add ammonia to increase the pH to between 9 and 10, dilute with water to 100 ml and add 0.1 ml of indicator solution. Place the vessel in the cell compartment of the spectrophotometer, which should have been switched on 15 minutes earlier to warm up, and arrange the cover, stirrer and burette in position. Adjust the speed of the stirrer so that it does not affect readings. With a wave- length of 665mp and the switch at “check,” balance the photometer bridge by adjusting the slit width, as is normally done with the solvent cell in position.Switch to “1.” Add the ethylenediaminetetra-acetate solution until about 0.5 ml short of the expected end-point and, when equilibrium has been reached, measure the optical density. Continue the addition of the reagent in small measured increments, at first 0-1 11-11 and then, when the optical density changes rapidly, 0.05 ml. After each, record the optical density. Proceed until the latter becomes steady and finally add 06ml more of the reagent, in order to check that the maximum reading has been attained. Plot optical-density measurements against burette readings and, from the edge of the plateau, determine the burette reading at the end-point. Titrate a blank containing all the reagents, but add the titrant in small increments from the start.It is not worth while to rebalance the spectrophotometer with the untitrated solution, and the titration may therefore be started at an optical density reading of, say, 0.075. During titrations, any drift in the dark current should be counterbalanced in the usual way. For the standardisation of the 0.008 M ethylenediaminetetra-acetate solution, take an aliquot of standard zinc chloride solution containing about 5mg of zinc. Wheh the bulk of the zinc has been titrated, replace the 0.008 111 ethylenediaminetetra-acetate solution by the standardised 0.0025 M solution and complete the titration as already described. Titration of the eluates from resi.12 columns with ethylenediaminetetra-acetate (gallium, idurn, uranium and more than 0.5 mg of titanium absent)-Add to the solution 0.1 ml (0.3 ml if lead or manganese may be present) more of ascorbic acid solution than is necessary to decolorise any iodine present.If lead may be present, add about 2 ml of manganous chloride solution. Next add sufficient 5 N ammonia to raise the pH to between 9 and 10, followed by 1 ml of potassium cyanide solution. Heat at 40” C for 20 minutes (or at 70” C for 5 minutes if tin is absent) in order to form ferrocyanide, cool quickly and add 0.1 ml of indicator solution. Proceed as described under (a) or (b). (a) The colozcr of the solution is bhe (lead and manganese absent)-Add 1 ml of chloral hydrate solution, transfer the solution to the titration cell and arrange for titration spectrophotometrically.Balance the instrument with the switch in the “check” position, then switch to “1,” set the optical-density scale at 0405 or 0.01 and, with the dark-current shutter open, add the ethylenediaminetetra-acetate solution slowly, with stirring, until the galvanometer balances. Close the dark-current shutter, re-adjust the dark current if necessary and, after re-opening the dark-current shutter, continue the addition of the titrant in sma.11 increments, as already described, until the end-point is reached and then exceeded. Carry out a similar titration on a blank which has been subjected to the full procedure. Determine the end-points in the usual way. If the amount of zinc is known to be large, use the more concentrated solution of ethylenediaminetetra-acetate for much of the titration, but obtain the end-point by titration with the less concentrated solution.If tin is present, titrate very cautiously, as the end-point reaction is slow and the change in optical density is only one-third of the normal. Up to 0.5 mg of titanium may be tolerated only if titration is carried out rtapidly after the addition of the indicator. (b) The colour of the solution is pink (lead or manganese or other titratable element $resent)-Check that sufficient potassium cyanide has been added to mask zinc. If lead is known to be present and manganous chloride has been added, titrate very slowly with 0.0025 M ethylenediaminetetra-acetate until the colour is seen to be permanently blue and an excess of about 0-2 ml has been added.The excess need not be accurately known. Transfer the solution to the titration cell, assemble in the spectrophotometer as before and arrange for back-titration, at the usual wavelength, with the standardised manganous chloride solution. Switch to “1,” set to an optical density of 0-1 and balance In a rough visual experiment find the approximate position of the end-point.Feb., 19561 ZINC FROM SOME OTHER ELEMENTS 91 the instrument by means of the slit and sensitivity controls. Add the titrant slowly in small increments until the optical density starts to decrease, after which reduce the size of the increments and wait longer before measuring optical density. Do not go further than necessary to locate the end-point.From the graph, determine the end-point and the excess of manganous chloride added. For the determination of zinc add 1 ml of chloral hydrate solution and proceed as already prescribed. Correct for the excess of manganous chloride and for the relevant blank experiment. Standardise the manganese chloride solution by adding a known excess of ethylenediaminetetra-acetate solution to a titrated blank solution and titrating with the manganese solution. In the absence of lead, a pink solution at the cyanide stage may, alternatively, be titrated directly with ethylenediaminetetra-acetate by the spectrophotometric technique. Titration of zinc with ethylenediaminetetra-acetate after the solvent-extraction procedure- Transfer the cooled, chloroform-free solution (p.89) to the titration cell and add 0.1 ml of Solochrome black solution. If the colour is not blue, titrate directly with the reagent by the spectrophotometric method until it is. Since lead has been removed, slowness in titration is not expected. If the colour is blue or has been brought to blue by titration, add 2 ml of chloral hydrate solution and titrate the liberated zinc in the usual manner. Approach the end-point cautiously, as it is sharper than for a normal titration where cyanide and chloral hydrate have been used. Correct for the full extraction blank. The time required for solvent extraction, back-extraction and titration is 30 to 35 minutes. Preparation of alloy solutions-Treat 100mg of alloy with hydrochloric acid and the minimum of 100-volume hydrogen peroxide or of nitric acid, excess of which should be removed by evaporation to small volume in the presence of an excess of hydrochloric acid.Dissolve the residue in a minimum of hot 2 N hydrochloric acid; then, if lead chloride is present, cool to 0" C and separate the latter, together with any insoluble matter, by centrifuging. If more than 10 per cent. of tin may be present, add 400mg of disodium methylarsonate and make the solution 2 N in free hydrochloric acid. Transfer the solution and the appro- priate washings with 2 N hydrochloric acid to a suitably prepared resin column. Proceed as required. Preparation of a solution from borosilicate glass-To a 100-mg sample add a few drops of water and 0.5 ml of 40 per cent. w/v hydrofluoric acid and evaporate as far as possible to dryness on a hot-plate with a steady surface temperature of 100" C , stirring occasionally with a platinum wire.Add to the residue 0.1 ml of water, 0.25 ml of hydrofluoric acid and 0.1 ml of 60 per cent. w/w perchloric acid, and re-evaporate, with final gradual increase in the temperature, to expel perchloric acid. Reheat with 0.2 ml of perchloric acid until this acid is removed. Dissolve the residue in 5 ml of 2 N hydrochloric acid and proceed to the anion-exchange separation, using only hydrochloric and nitric acids, and the titration of zinc (spectrophotometric determination of end-point) via the ascorbic acid - potassium cyanide - chloral hydrate method. RESULTS The methods prescribed were applied to the determination of zinc in synthetic mixtures and in various materials, with the results shown in Tables 111 and TV, respectively.All the results for zinc in mixtures refer to weight aliquots of standard solutions of zinc chloride prepared from Hilger's "H.H.P." zinc (99.99 per cent.). Many of the additional elements were derived from Specpure solutions or from similar high-quali ty solids. Control experi- ments with the other elements were made without addition of zinc chloride in order to allow for traces of titratable impurity. The results for these seldom differed significantly from those for the associated blank experiments. The blanks were commonly of the order of 50% of zinc. The effects of various acid radicals were examined, as the corresponding acids might be required or encountered in preparing solutions of materials, and it would be an advantage if small amounts could be tolerated in the anion-exchange process.In the experiment with 5 mg of uranium an end-point was obtained on titrating rapidly but the titrated solution soon became pink again as a result of blocking of the indicator. A duplicate of the experiment with cobalt, nickel and manganese gave no end-point, supposedly because of accidental oxidation of the indicator by manganeseIL1. The mean error for the 48 experiments with about 0.5 mg of zinc is -1 pg and for the 21 with 5 mg of zinc +4 pg. The corresponding standard deviations are 8 and 15 pg,92 HUNTER AND MILLER: THE SEPARATION OF TABLE 111 [Vol. 81 TITRIMETRIC DETERMINATION OF ZINC IN SOLUTIONS AFTER SEPARATION BY ANION EXCHANGE Approximate Approximate weight of Approximate Approximate weight of other elements weight of weight of zinc taken, or radicals, Error, zinc taken, other elements, Error, 10-8 g 10-8 g 10-8 g 10-8 g 10-3 g 10-6 g Elution with water and nitric acid only- 5 0.6 - 0, - 5 0-5 Be, 5 3-6 0.4 SO, , 250 1-21 0.5 Cr.5 +3 5 PO4”’, 50 +S 0.6 Cr, 100 +7 0.4 P O P , 50 -6, 0 0.5 Th, 5 +3 5 AsO;”, 50 +3 0-55 Th, 100 +2 0.5 As04/”, 50 -6 0.5 Zr, 5 - 10 5 AsOs”’, 50 1-1 0.5 Ti, 5 -9 0.55 AsO/’, 50 - 13 0-45 Zr, 50; Ti, 50 4-7, +5 5 NO,’, 50; ClO,’, 50 -3 0.5 u, 5 +7 0.5 NOs’, 50; ClO,‘, 50 $6 085 Co, 20; Nil 40; Mn, 40 +1 (a) +17, -24 0.5 cu, 100 -k5 - 5 SO4::. 250 + 20 0.55 Be, 100 3-3 (b) Elution with hydriodic acid, water and nitric acid- 5 Cd, 100 +6, + l o 5 Bi, 50 -2, +4 0.5 Cd, 100 -5, +7 0.4 Bi, 50 -7, - 8 5 Pb, 100 1-26 +36 0.4 Al, 100 -81 +3 0.4 Pb, 100 -3, 0 0.4 SbIII, 50 -7, -9 0-4 SbV, 50 -9, +5 Hg*I, 25, TlIII, 25 5 0.5 5 SbIII, 50 f 1 3 1 + 8 0-4 FeIII, 100 +131 +8 5 Sbv, 50 +11, +6 0*5 Au, 25, Pt, 25, -24, - 9 (c) Methylarsonate treatment and elution as in (b)- - -12, + 2 5 Sn, 100 -31, -7 - + I , +lo 0.4 Sn, 100 -11, -3 (d) Elution as in (a) , followed by the solvent-extraction proccdure- 0.5 Ga, 100 1-11 +3 0.5 u, 100 -7, -6 0.5 In, 100 -4, -8 TABLE IV DETERMINATION OF ZINC IN ALLOYS AND IN A GLASS Approximate composition of material c A v Alloy All Cu, Sn, Pb, Sb, Fe, Ni, Zn, Zinc found, % % % % % % % % Yo <1 2 2.16 to 2.54* 2.41, 2-42 Aluminium alloy . ... 89 2 - - <1 - 0.80t 0.94, 0.94 1.72 to 2.00*++ 1.88, 1.89 Aluminium alloy No.181 .. Bronze No. 183 . . .. - 83 10 2 <1 <1 - <1 5 83 12 <<I. - 0.02 to 0.16* 0.03, 0.03 White metal No. 177 . . - 4 8 4 4 8 < 1 - 0.31 to 0.47*$ 0.44, 0.44 White metal No. 178 . . - Al,Os BaO Na,O + + + 5 - ? 3 2 - 87 SiO,, Fe,O,, CaO, K O , B20s, ZnO, ZnO , % % % % % % % Borosilieate glass . . . . 65 5 4 9 15 1*86§ 1.89, 1-87 * British Chemical Standard, range in certificate values. t Spectrographic determination by the North British Rubber Company. $ The methylarsonate treatment was used. 5 Value supplied by Research Laboratories of The General Electric Company. respectively. It is evident that 50-mg amounts of zinc would be determined with adequate accuracy and precision if, after ion exchange, an aliquot of the eluate from the resin, containing 6 mg of zinc, were titrated spectrophotometrically.It is highly probable, however, that visual titration of larger amounts of zinc with a more concentrated solution of ethylene- diaminetetra-acetate will suffice. In the analyses of the British Chemical Standa.rd samples, all the results fall within the range of the certificate values and precision is excellent. It is obvious that the prescribed methods have a wide range of applicability.Feb., 19561 ZINC FROM SOME OTHER ELEMENTS 93 One of us (J. A. H.) is grateful for the award of an Edinburgh University Studentship. We are indebted to Mr. R. C. Chirnside, of the Research Laboratories of The General Electric Company, Wembley, and to the North British Rubber Company, Edinburgh, for analysed samples, and also to the Geigy Company, Manchester, and Versenes Incorporated, Framing- ham, Mass., U.S.A., for samples of chelating reagents.We thank Dr. H. Flaschka of Graz, Austria, who gave us valuable pre-publication information about the use of ascorbic acid for reducing ironx1’ 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. REFERENCES Miller, C, C., and Hunter, J. A., Analyst, 1954, 79, 483. Kraus, K. A., and Moore, G. E., J . Amer. Chem. SOG., 1953, 75, 1460. Rush, R. M.., and Yoe, J. H., Anal. Chem., 1954, 26, 1345. Baggott, E. R., and Willcocks, R. G. W., Analyst, 1955, 80, 53. Welcher, F. J., “Organic Analytical Reagents,” D. Van Nostrand Company, Inc., New York, Flaschka, H., Mikrochem.Mikrochim. Acta, 1952, 39, 38. Biedermann, W., and Schwarzenbach, G., Chimia, 1948, 2, 56. Debney, E. W., Nature, 1952, 169, 1104. Blaedel, W. J., and Knight, H. T., Anal. Chem., 1954, 26, 743. Kinnunen, J., and Merikanto, B., Chemist-Analyst, 1952, 41, 76. Flaschka, H., 2. anal. Chem., 1953, 138, 332. Brown, E. G., and Hayes, T. J., Anal. Chim. A d a , 1953, 9, 408. Pi-ibil, R., Coll. Czech. Chem. Comm., 1953, 18, 783. Erdey, L., and Bodor, E., Anal. Chem., 1952, 24, 418. Flaschka, H., and Puschel, R., 2. anal. Chem., 1954, 143, 330. Sweetser, P. B., and Bricker, C. E., Anal. Chem., 1954, 26, 195. Karsten, P., Kies, H. L., van Engelen, H. Th. J., and de Hoog, P., Anal. Chim. A d a , 1955, 12, 04. Chalmers, R. A., Analyst, 1954, 79, 519. Flaschka, H., Mikrochem. Mikrochim. A d a , 1952, 39, 315. Flaschka, H., and Huditz, F., 2. anal. Chem., 1952, 137, 172. Flaschka, H., Chemist-Analyst, 1953, 42, 56. Flaschka, H., and Amin, A. M., 2. anal. Chem., 1953, 140, 6. Kinnunen, J., and Merikanto, B., Chemist-A nalyst, 1954, 43, 93. Flaschka, H., and Abdine, H., Mikrochim. Acta, 1954, 657. Welcher, F. J., “Organic Analytical Reagents,’’ D. Van Nostrand Company, Inc., New York, 1948, Volume 111, pp. 1 t o 27. Klinger, H., and Kreutz, A., Annalen., 1888, 249, 147. 1948, Volume IV, p. 69. CHEMISTRY DEPARTMENT THE UNIVERSITY, EDINBURGH, 9 June 27th, 1955 DISCUSSION DR. R. J. MAGEE asked how germanium behaved in the anion-exchange procedures and whether i t interfered in the titrations. MR. HUNTER replied that some germanium was adsorbed and eluted along with the zinc. but since i t could be removed by volatilisation as germanium tetrachloride, i t was not examined in titration. DR. F. J. ELLIOTT asked whether the method was likely to be applicable to the determination of zinc in biological materials. MR. HUNTER said that, after wet-ashing with, say, perchloric acid, the method could be directly applicable. Perchlorate, sulphate, nitrate and phosphate did not interfere with the adsorption of zinc. DR. MAGEE asked how arsenic, antimony and tin behaved during anion exchange and whether they had been eluted and separated. MR. HUNTER replied that arsenic was not held a t all, all the antimony111 was held, 75 per cent. of the antimonyV and all the tinIv, from 2 N hydrochloric acid solutions. Methylarsonic acid prevented adsorption of much tinIV. No method for complete elution of antimony or of tin had been found, but change in the concentration of hydrochloric acid might help. DR. A. MAGNUS PYKE asked if polarography had been considered for determining the small amounts of zinc. MR. HUNTER said i t had, but for about 5 mg, an accuracy of & 10 pg, being 1 in 500, could hardly be expected from polarography. Further, the small amounts of various other elements liable to accompany zinc would be difficult to eliminate in a polarographic procedure. MR. J. A. EGGLESTON asked if the method was likely to be usable for amounts of zinc less than 50 pg, and whether, with smaller amounts of zinc, the blank on the reagents would not be appreciable. MR. HUNTER replied that if the blank could be lowered, the limit of determination of zinc could be taken somewhat below 50 pg, and the lower limit would then be governed by the over-all accuracy of f 10 pg.

ISSN:0003-2654    

DOI:10.1039/AN9568100079

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

94 SHRIMPTON : THE ESTIMATION OF [Vol. 81 The Estimation of Vi tamin-Bl2 Activity in Feeding Stuffs with Lacto bacillus leichmannii and Ochromonas malhamensis BY D. H. SHRIMPTON* (Presented at the meeting of the Biological Methods Grozt$ on Friday, May 13th, 1955) The vitamin-B,, activity of some animal by-products used as feeding stuffs has been determined with Lactobalsillus leichmannii and Ochromonas malhamensis. A description is given of the microbiological methods ; these have been adapted to determine the vitamin-B,, activity of feeding stuffs with as little manipulation as possible. .A semi-automatic syringe used for dispensing solutions is described, and also a shaker of simple design, in which cultures of 0. malhamensis were grown. All but one of the feeding stuffs had different vitamin-B,, activities for each of the test micro-organisms. The activities for 0.malhamensis were less than or similar to the activities for L. leichmannii, and i t was concluded that at best 0. malhamensis was no better than L. leichmannii in predicting the vitamin-B,, activity of feeding stuffs for chicks. IT is known that many biological materials contain substances related to vitamin B,,, which, although they cannot be converted to cyanocobalamin by treatment with cyanide, show a vitamin-B,,-like activity for many micro-organisms and animals. In many cases, however, the relative activity of these substances differs according to the micro-organism or animal used for the assay of vitamin B,,.192 In mixtures of these many growth factors only cyano- cobalamin, at present, may possibly be determined specifica1ly.l In the absence of further evidence it is convenient to attribute vitamin-B,,-like activity for Ochmmonas malhamensis to cyanocobalamin and activities for other micro-organisms and test animals to vitamin- B,,-like substances, which may or may not be cyanocobalamin.A question that must therefore be posed is: can the vitamin-B,, activity of a feeding stuff for animals be predicted from its content of cyanocobalamin? As a contribution to the solution of this problem, the cyanocobalamin content of several feeding stuffs has been determined and also the vitamin-B,, activity for LactubaciZZw leichmannii of the same samples. A detailed description of the methods used to obtain these values is reported in this paper and the results are discussed with particular reference to the vitamin-B,, activity of feeding stuffs for chicks.METHODS Vitamin-B,, activity has been determined microbiologically. Although many workers have developed techniques of assay that are suited to their particular problems, there is no published account of the techniques used in assaying feeding stuffs. A description of the methods used in this work therefore follows. CULTURE OF THE TEST MICRO-ORGANISMS- L. leichmannii ATCC4797 was cultured in a single-strength basal medium,S modified by the addition of 40 mg of asparagine4 to every. 100 ml and by replacing L-cystine with ~-cysteine.4 A commercial casein hydrolysate (Allen & Hanbury Ltd.) was used. The organism was maintained in stab cultures of this medium containing 3 per cent.w/v of agar and supplemented with 0-5 pg of standard vitamin B,, (Cytamen, Glaxo Laboratories Ltd.) to every 100rnl. The organism was subcultured once a fortnight, first into a liquid culture of similar composition but containing no agar, arid then into a fresh stab culture, which was incubated at 37" C for 16 hours and then stored at 1" C. An inoculum for use in an assay was prepared in two stages. Ten millilitres of the modified single-strength basal medium were supplemented with 50mpg of standard vitamin B1,, inoculated from the stock stab culture and incubated in a 50-ml centrifuge tube at 37" C for 48 hours. The culture was centrifuged and resuspended in 10ml of physiological saline. A second similar tube of Present address: Low Temperature R.esearch Station, Cambridge.Feb., 19561 VITAMIN-Biz ACTIVITY IN FEEDING STUFFS 96 medium, but supplemented with only 0.5 mpg of standard vitamin B,,, was then inoculated with one drop of the suspension and incubated at 37°C for 16 hours.The culture was centrifuged and resuspended in physiological saline, one drop being used to inoculate each tube prepared for assay. An inoculum of a “depleted” culture of this type gave a low blank, but nevertheless grew vigorously in the presence of vitamin-B,,-like growth factors. 0. malhamensis was grown in the medium described by F0rd.l One drop of this dark green-brown culture was used to inoculate each tube prepared for assay. STANDARD VITAMIN B12- A 1-ml ampoule of Cytamen 50 (Glaxo Laboratories Ltd.) containing approximately 50 pg of vitamin B,, was diluted to 10 ml and stored at +lo C for up to 1 month.The concentration of vitamin B,, in this dilute solution was determined by measuring EgEttm at 361 mp and comparing this with a reference value5 for EgTcEm of 20.4. The concentration of vitamin B,, was found to be 5.30pg of the vitamin per ml of solution. PREPARATION OF SAMPLES FOR ASSAS- The procedure of Scheid and Schweigert,6 in which 1 g of the finely ground sample was heated in an autoclave at 15 lb per sq. in. for 30 minutes with 20ml of a 1 per cent. w/v solution of sodium acetate at pH 4.5, was modified by the inclusion of potassium cyanide in the acetate solution at 0.01 per cent. w/v. Two extracts were prepared for each sample. The extracts were centrifuged and the supernatant liquid was diluted so that it contained a vitamin-B,, activity approximately equivalent to 0.1 mpg of the vitamin per ml of extract.When the vitamin-B,, activity was determined with 0. malhamensis, a similar extraction procedure was followed, modified only by the exclusion of sodium acetate. Fordl reported that the presence of more than 0.1 per cent. of sodium acetate in the growth medium markedly inhibited the growth of this organism. With some materials, but not with those described below, it may be necessary to modify this extraction procedure. Thus, for example, when sows’ milk is assayed, digestion with papain is necessary,7 and with some balanced rations the temperature of extraction must be kept below 100” C to prevent turbidity in the extract.PREPARATION OF ASSAY TUBES- Wire baskets, each holding 70 rimless tubes (19 mm x 120 mm), were boiled for 10 minutes in a 2 per cent. w/v solution of a proprietary glass cleanser (Brylyanz, obtainable from G. T. Gum Ltd., London, S.W.6). Each tube, while remaining in the basket, was rinsed three times each with tap water and distilled water and then placed in a drying oven. This method was adopted in place of two other methods that had previously been used. When the tubes were washed with chromic acid, occasional erratic results were obtained. The impression given was that of a cumulative effect, and it is possible that in older scratched tubes the chromic acid was not always removed by subsequent rinsing. Soda washing followed by an acid rinse was satisfactory but laborious.ASSAY PROCEDURE- It has frequently been observed in this laboratory that results calculated from replicate sets of assay tubes of a single extract were in better agreement than those calculated from replicate sets of assay tubes from two similar extracts of the same material. Also, results calculated from replicates assayed at the same time were in better agreement than when they were calculated from replicates assayed on different occasions. The standard procedure adopted has therefore been to prepare two extracts of each material to be assayed and to set up duplicate sets of assay tubes for each extract. The entire procedure was then repeated with fresh extracts. If the four results obtained differed from their own mean by more than 10 per cent., the determination was repeated. The vitamin-B,, activities of feeding stuffs reported below are each the mean of four results that differ from their respective means by less than 10 per cent.Duplicate sets of assay tubes were filled, as shown in Table I, for each of the two extracts prepared from each sample. A triplicate set of assay tubes containing graded levels, up to 0.5 mpg per tube, of standard vitamin B,, was also prepared. The total volume contained in each tube was always 10 ml when L. Zeichmannii was the test organism and 5 ml when96 SHRIMPTON : THE ESTIMATION OF [Vol. 81 0. malhamensis was the test organism. The extract and water and, when 0. malhamensis was the test organism, the basal medium were dispensed with a Hauptner 30-ml capacity syringe, illustrated in Fig.1. This instrument delivered 1, 2, 3, 4 or 5 ml at a time with an error of less than 1 per cent. and was supplied by Messrs. Willows Francis Ltd. (73,75 and 89a, Shacklewell Lane, London, E.8). When L. Zeichmannii was the test organism, the basal medium was dispensed with an automatic syringe of the type used for preparing bacterial plates. This instrument delivered 5 ml at a time with an error of less than 2 per cent. The 20 tubes necessary for the assay of two extracts of one sample were arranged in one rack numbered in one comer so that marking of individual tubes was unnecessary. TABLE I QUANTITIES OF HYDROLYSATE, STANDARD , WATER AND BASAL MEDIUM DISPENSED INTO ASSAY TUBES With L. leichmannii as test organism With 0.malhamensis as test organism A A r \ I \ Tube A B C D E A B C D E - 1 2 Hydrolysate, ml . . 1 2 3 1 1 1 2 3 1 1 Standard, ml .. - - 1 2 Water, ml . . .. 4 3 2 3 2 3 2 1 2 1 Basalmedium, ml . . 5 5 5 5 5 1 1 1 1 1 Total volume, ml .. 10 10 10 10 10 5 5 5 5 5 - - - - - - - - - - - - - When L. leichmannii was the test organism, the tubes were covered with metal caps (0x0 Ltd.) and steamed for 30 minutes. After cooling and inoculation the tubes were incubated in a constant-temperature room at 37°C for 48 hours. When 0. malhamensis was the test organism, the tubes were covered with glass specimen tubes and heated in an autoclave at 10 lb per sq. in. for 10 minutes. After cooling and inoculation the tubes were placed in a specially constructed shaker of simple design, shown in Fig.2, and incubated in a constant-temperature room at 30" C for 96 hours. The shaker consisted of a box, constructed of $-inch thick wood, which was suspended on four spring-steel strips from a steel gantry. A steel roller on one side of the box was struck by a steel striker approximately once every two seconds. The striker was turned by a NECO single-phase 4-h.p. induction motor, type R.I. (Normand Electrical Co. Ltd., Neco Works, London, S.W.4), which was cooled by a fan and had a built-in gear box to give a drive of 35 r.p.m. At the end of incubation the tubes were steamed for 30 minutes, cooled and shaken. The suspensions were transferred to special tubes before their turbidities were determined in an EEL nephelometer (Evans Electroselenium Ltd.) .COMPUTATION OF RESULTS- The turbidities were plotted against the corresponding volumes of standard vitamin B,, or of extract, and the mean slope ratios were calculated according to the principle of Kodicek and Pepper,s by using their internal-standard method. The results were always plotted and-the curves inspected. If the response curves were linear, their mean gradients were calculated as follows. Where A , B, C and D were the mean turbidities of solutions containing 1, 2, 3 and 4ml of standard vitamin B1,, respectively, the mean slope of the standard response curve was- " $- 3c + 2D - ' l A ; let this be equal to z. 18 Also, where a, b and c were the mean turbidities corresponding to solutions containing 1 , 2 and 3 ml of extract, respectively, and where d and e were the mean turbidities correspond- ing to 1 ml of extract with the addition of 1 or 2 ml of standard vitamin B,,, respectively, the slopes of the extract curve and of the internal-standard curve were- 2b $- ' - 3a; let this be equal to x , and 4 2d 4- - 3a; let this be equal toy.4Fig. 1. Hauptner 30-nil capacity syringe Fig. 2. Specially constructed shaker used for determining vitamin-B,, activity with Ochromonas rnalhamensis [To face p. 96Feb., 19561 VITAMIN-BIZ ACTIVITY I N FEEDING STUFFS The vitamin-B,, activity of the extract in pg per g was therefore given by- 97 x [standard vitamin B,,] r X [extract] The ratio of the slopes of the internal standard and standard curves was given by y/z, and when this was between 0.9 and 1.1 the assay was considered valid.When the response curves were not linear over the whole of the range, the calculation was modified so that only these points within the linear portion of the response were used. A value so obtained was checked with a further assay whenever possible. Since the calculation of potency was made only from the results of the internal standard and the extract, the assay was basically of Wood's "common-zero five-point" de~ign,~ the zero in this case being the growth response resulting from 1 ml of extract. April I952 I953 April - September March */" February. :* w 0 I I 1 I t 1 1 1 I I t I I ! 1 , 0:. I .*. I Standard vitamin-B,, solution, ml per tube Fig. 3. vitamin B,, The growth response of Lactobacillus Zeichmannii after 48 hours at 37" C to Cytamen FEEDING STUFFS ASSAYED- The vitamin-B,, activities have been determined of the following feeding stuffs : herring meal, white-fish meal, whale-meat meal, meat and bone meal and dried skim milk.They are used in Great Britain as protein supplements, and whenever possible the sample assayed was from a batch for which the processing history was known. RESULTS AND DISCUSSION GROWTH OF MICRO-ORGANISMS- The standard curves of both L. Zeichmannii and 0. maZhamensis were linear over part of their ranges. A series of such standard curves obtained over a period of several months is shown for L. leichmannii in Fig. 3 and for 0. malhamensis in Fig. 4. Three tubes were used at each level, and when two or more points are coincident they are placed side by side.Since a straight-line response was obtained consistently, no transformations of the type recommended by Clarke,lo log of dose against response, or Wood,ll log of dose against log of response, were made. When L. Zeichzmannii has been the test organism, a series of successful assays has some- times been followed by a period in which growth was either slight or excessive at all levels98 SHRIMPTON : THE ESTIMATION OF [vol. 81 of standard. It has always been possible to correct failures of this type by growing a depleted inoculum from a freeze-dried culture obtained from the National Collection of Industrial Bacteria and repeating the assays. When the growth has been erratic, several steps in the assay procedure have, at different times, been suspect, but it has not been proved that any one of these has been responsible.With 0. malhamensis as the test organism, it has been found that growth was some- times unsatisfactory when the assay tubes were covered with metal caps while they were shaken. It seemed probable that metal filings were falling into some of the tubes, resulting in a blackening of the cultures. There has been no difficulty with glass covers. VITAMIN-B,, ACTIVITY OF FEEDING STUFFS- The vitamin B,, activities, expressed as pg of standard vitamin B12.per g of dry feeding stuff, for 0. malhamensis and for L. leichmannii are given in Table 11, in which the feeding ~~~~ ~~ 1954 I955 November April Standard vitamin-B, solution, ml per tube Fig. 4. The growth response of Ochromonas malhamensis after 96 hours stuffs are arranged in descending order of activity for 0.malhamensis. In only herring meal was the activity similar for both micro-organisms, when the activities differed from their mean by less than 10 per cent. Dried skim milk had a greater activity for 0. malhamensis, which was an unexpected result in view of the believed specificity of this organism for cyanocobalamin. However, the difference is not large and it is possible that the activity for L. leichmannii might have been suppressed. The remaining three meals each had an appreciably greater activity for L. leichmannii, white-fish meal having more than twice the activity for L. leichmannii that it had for 0. malhamensis. The similarity of the vitamin-B,, activities of herring meal for the two micro-organisms is surprising in view of the differences found in the other feeding stuffs and most particularly in white-fish meal.Fordl also has reported that the vitamin-B,, activities of fish solubles, presumably prepared from herring, for 0. malhamensis and for Escherichia coli were identical. I t has been frequently stated that animal by-products, such as those evaluated in this work, contain unidentified growth factors. Possibly owing to the absence of any method of assaying such factors, it has been customary to refer to growth effects of a vitamin-like nature for chicks as those due to the “vitamin-B,,” content of the feeding stuffs. Unfor- tunately, there are many difficulties in biological assays of the vitamin-B,, activities of feeding stuffs for chicks,2 and they are not suitable for routine work.At present, therefore, the analyst of feeding stuffs designed for animals must make some prediction of the vitamin-B,,Feb., 19563 VITAMIN-€312 ACTIVITY IN FEEDING STUFFS 99 activity of that feeding stuff for an animal, even though it is possible that only cyanocobalamin can be determined specifically. Denton, Lillie and Sizemore12 have studied the correlation between vitamin-B,, activity of animal by-products for L. Zeichmannii and for the chick. They reported agreement between the methods for a number of feeding stuffs including herring meal, steam fish meal, whale solubles and dried skim milk. When the methods gave results that were not in agree- ment, the activity for the chick was greater than for L. Zeichmannii (excepting those from sewage sludge preparations, for which the activity for L.Zeichmannii was the higher). Thus, a closer determination of the vitamin-B,, activity of a feeding stuff for chicks would only TABLE I1 THE VITAMIN-BIZ ACTIVITIES OF FEEDING STUFFS FOR 0. mudhamensis AND FOR L. leichmannii Standard vitamin-B,, activity per g of dry feeding stuff for Feeding stuff 0. malhamensis, L. leichmannii, I.Lg Pg Herring meal . * * . .. 0.16 0.18 Whale-meat meal . . .. .. 0.053 0.092 0.06 1 Meat and bone meal .. a . 0.048 Dried skim milk . . .. .. 0-036 0.028 White-fish meal . . .. .. 0.028 0.061 be expected from a microbiological method for which the calculated activities were similar to or greater than those calculated from L. Zeichmannii. 0. malhamensis has not fulfilled this condition for the feeding stuffs studied in this work.Although a knowledge of the cyanocobalamin content of feeding stuffs as determined by 0. maZhamensis is of interest, the micro-organism is at best no better, in the feeding stuffs studied, than L. Zeichmannii in predicting their vitamin-B,, activity for chicks. The reported correlations between assays of vitamin-B,, activity with chicks and with L. Zeichmannii may indeed be fortuitous, but at the present state of development it is nevertheless valuable. I am indebted to Dr. J. E. Ford for a culture of 0. malhamensis. It is a pleasure to acknowledge the technical assistance of Miss K. Ramsay and to thank Dr. K. J. Carpenter for advice and useful criticism. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. Ford, J. E., Brit. J . Nutrit., 1953, 7, 299. Coates, M. E., Harrison, G. F., and Kon, S. K., AnaZyst, 1951, 76, 146. Skeggs, €3. R., Nepple, H. M., Valentik, K. A., Huff, J. W., and Wright, 1,. D., J . Bid. Chem., Price, S. A., private communication, 1951. Folkers, K., and Wolf, D. E., Vit. 6 Horm., 1954, 12, 1. Scheid, H. E., and Schweigert, B. S., J . Biol. Chem., 1951, 193, 299. Gregory, M. E., Brit. J . Nutrit., 1954, 8, 340. Kodicek, E., and Pepper, C. R., J . Gen. Microbiol., 1948, 2, 306. Wood, E. C., Analyst, 1946, 71, 1. Clarke, M. F., Anal. Chem., 1953, 25, 1247. Wood, E. C., Analyst. 1953, 78, 451. Denton, C. A., Lillie, R. J., and Sizemore, J. R., Proc. 10th World's PouEtry Congress (Edinburgh), 1950, 184, 211. 1954, 1, 152. ROWETT RESEARCH INSTITUTE BUCKSBURN ABERDEENSHIRE July 27th, 1965

ISSN:0003-2654    

DOI:10.1039/AN9568100094

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

100 LINCOLN AND CHINNICK : DETERMINATION OF QUATERNARY [Vol. 81 Determination of Quaternary Ammonium Compounds as Phosphotungstates BY P. A. LINCOLN AND C. C. T. CHINNICK Surface-active quaternary ammonium compounds are shown to be quantitatively precipitated by phosphotungstic acid, whereas alternative gravimetric methods give inaccurate results with a number of such compounds. By weighing the quaternary phosphotungstate after drying and again after ignition, it is possible to calculate the amount of quaternary cation in an unknown sample and also the ionic weight of the quaternary salt. All alternative methods of analysis hitherto proposed require the molecular weight of the quaternary compound to be known before the amount present in a sample can be determined, leading to a vicious circle of approximations, particularly with unknown materials. Interferences are discussed and general techniques by which these may be overcome are suggested.FOR the determination of quaternary ammonium compounds several methods that make use of precipitation reactions with complex anions have been proposed. Among the anions that have been used may be mentioned ferrocyanide,l ferricyanideJ2 dichromate3 and r e i n e ~ k a t e . ~ ~ ~ These methods are limited in application, since the reactions are not always stoicheiometric, particularly with cations of low equivalent weight .6 Further, when calcu- lating the results of such determinations the molecular weight of the quaternary must be known, whereas in general the molecular weight of the compound is unknown or at best assumed.Lematte, Boinot, Kahane and Kahane' used phosphotungstic and silicotungstic acids to assay low molecular weight quaternary ammonium compounds and, although it has been shown that surface-active quaternary ammonium compounds are precipitated by these reagent~,89~ no use has been made of the reaction for their quantitative determination. The present investigation shows that they are quantitatively precipitated as phosphotungstates. (A summary of some of the results has been reported previously.6) Ignition of the quaternary ammonium phosphotungstate at dull-red heat results in an inorganic residue free of organic matter'- This residue may be dehydrated at higher temperatures as follows- Such dehydration corresponds to a loss in weight of only three parts per thousand and is therefore unlikely to introduce a serious error. EXPERIMENTAL Standard quaternary ammonium compound solutions, 0.01 N, were prepared by dis- solving the required amounts of chemically pure materials (which are available, some in limited amounts, from Milton Industrial Chemicals (London) Ltd.and by arrangement through most chemical supply houses) in water and standardising with 0.01 N silver nitrate by the method of Dubois.lo A 20-ml sample of the 0.01 N quaternary ammonium compound solution was diluted to approximately 200 ml in a 400-ml beaker. Then 2 to 3 g of sodium chloride and 10 ml of 10 per cent. hydrochloric acid were added and the solution was then raised to the boil. An excess (10 ml) of 10 per cent. phosphotungstic acid solution was then added, and the solution was boiled gently until the precipitate coagulated. This usually occurred within 5 minutes.The solution was cooled to room temperature and the precipitate was collected on a weighed No. 3 sintered-porcelain crucible. The precipitate was washed with about 200 ml of water and dried at 105" C to constant weight. A series of results obtained by this method is shown in Table I. . . * - (-4 Q,P0,.12W03 --+ HP03.12W0, . . * . . . 2HP0,.12W03 -+ P@5.24w03 $- H20AMMONIUM COMPOUNDS AS PHOSPHOTUNGSTATES TABLE I PRECIPITATION OF QUATERNARY SALTS AS PHOSPHOTUNGSTATES Quaternary ammonium compound Cetyltrimethylammonium bromide . . Myristyltrimethylanimonium bromide . . Lauryltrimethylammonium bromide . . Decyltrimethylammonium bromide .. Cetylpyridiniuni chloride . . . . . . Myristylpyridinium chloride . . . . 1,aurylpyridinium chloride .. . . Decylpyridinium bromide . . .. . . Cetylbenzylmorpholinium chloride . . Cetylniethyldiethanolammonium chloride p-tert.-Octylphenoxyethoxyethyldimethyl- benzylammonium chloride . . .. Weight of precipitate, mg . . 246.8 248.0 , . 242.6 243.3 . . 235.5 236-7 . . 228.7 230.0 . . 251.8 252.3 . . 246.6 247.2 . . 242.0 243.1 .. 233.8 236.4 .. 270.4 271.3 . . 260-1 261.7 . . 275.0 275-5 Weight of precipitate calculated* [Q&'O,.12wOJ t mg 248.8 243-1 237.5 231.9 262.8 247.1 241.5 235.9 2724 260.8 274-4 101 Average deviation, % 0.56 0.16 0.59 1.08 0.32 0.12 0.45 0.65 0.59 0.3 1 0.33 * Assuming solutions employed were exactly 0.01 N . TABLE I1 LOSS OF WEIGHT ON IGNITION OF QUATERNARY PHOSPHOTUNGSTATES 1 2 3 4 5 6 Weight of precipitate - found Weight of residue after -7 heating found at calcu- Devia- after calcu- Devia- Quaternary ammonium compound 105" C, lated,* tion,? ignition, lated, 1 tion,$ mg mg % mg mg % Myristyltrimethylammonium chloride .. 485.5 486.2 0.14 382.2 381.1 0.26 484.4 0.37 380.6 380.4 0-05 Decyltrimethylammonium bromide . . 461-2 463.8 0.56 380.4 379-7 0.18 462.4 0.30 381.2 380.6 0.16 Cetylpyridinium chloride . . .. . . 507.3 505.6 0.34 381-2 384.0 0.74 504.6 0.20 379.7 381.0 0.34 Cetylbenzylmorpholinium chloride . . 543-0 544.8 0.33 380.8 380-5 0.08 647.1 0.42 381.4 383.4 0.53 fi-tert.-Octylphenoxyethoxyethyldimethyl- benzylammonium chloride . . . . 546.8 548.5 0-49 378.4 380.0 0.42 548.0 0.09 379.5 381.4 0.50 * Assuming solutions 0.01 N exactly.t Deviation calculated from columns 1 and 2. $ Weight of residue calculated from column 1, the reaction Q,P0,.12\Y03 -+ HP03.12W0, being 5 Deviation calculated from columns 4 and 5. assumed. Experiments were then conducted to determine the loss on ignition of a selection of quaternary phosphotungstates. The various phosphotungstates were precipitated as before, but 40-ml samples of 0.01 N quaternary salt solutions were taken and the precipitates were collected on No. 3 sintered-porcelain crucibles. After being washed and dried at 105" C, the precipitates were ignited at a dull-red heat. The results obtained are shown in Table 11.102 LINCOLN AND CHINNlCK : DETERMINATION OF QUATERNARY TABLE I11 MOLECULAR WEIGHTS CALCULATED FROM TABLE II [Val.81 Quaternary ammonium compound Molecular weight determined Myristyltrimethylammonium chloride . . . . . . . . 289.6 Decyltrimethylammonium bromide . . . . . . . . 277.8 Cetylpyridinium chloride . . . . . . . . . . 345.0 Cetylbenzylmorpholinium chloride . . .. .. . . 440.9 p-tert. -0ctylphenoxyethoxyethyldimethylbenzylam~noniuin chloride .. . . .. .. .. .. . . 453.5 Molecular weight calculated 291.9 280.2 340.0 438-1 448.1 DISCUSSION The results given in Table I for a variety of surface-active quaternary ammonium com- pounds and the results obtained by Lematte et al. for some low molecular weight quaternary ammonium compounds show that precipitation as phosphotungstates is an accurate method of general application to quaternary ammonium bases.The decomposition reaction A (p. 100) is confirmed by the results shown in Table 11. This reaction is of great importance, as it permits the equivalent weight of the quaternary cation to be calculated. The anion present can be determined by conventional methods and hence the mean molecular weight of the quaternary ammonium compound can be determined. The results given in Table I1 have been employed to calculate the molecular weights of the respective compounds in Table 111. In all other procedures that have been advocated for the determination of quaternary ammonium compounds a molecular weight must be assumed before the results can be calculated. By weighing the phosphotungstate after it has been dried at 105" C and again after ignition, the amount of quaternary ammonium compound present in the sample taken can be determined without knowledge of the molecular weight, which may be calculated from the same results if required.This is invaluable in the analysis of unknown samples, when even the most comprehensive qualitative analysis cannot establish the molecular weight of the quaternary ammonium compound present, and in addition most commercially available surface-active quaternary ammonium compounds are mixtures containing smaller or greater amounts of homologues of the main constituent. The main disadvantage of this method is that ammonia and amines, e.g., pyridine and trimethylamine, form insoluble phosphotungstates and these materials must therefore either be removed before the quaternary ammonium phosphotungstate is precipitated, or deter- mined by some other means such as potentiometric titration and a correction for ammonium or amine phosphotungstate applied.If the amine is volatile and the quaternary base is not decomposed on boiling, the amine may be removed by making the aqueous solution of the sarhple alkaline and boiling. The solution is then acidified with hydrochloric acid and the quaternary ammonium phospho- tungstate precipitated in the usual manner. A preferred method, which has been applied to a number of surface-active quaternary ammonium compounds, depends upon selective extraction. It has been found that the distribution coefficient of a surface-active quaternary ammonium compound between isoamyl alcohol and an aqueous solution of an inorganic acid is usually over 2000.(If the isoamyl alcohol - water - quaternary ammonium compound system is free from inorganic solute, separation into phases often does not take place.) Thus a solution of quaternary ammonium compound in isoamyl alcohol (0-1 to 0 . 5 N ) may be freed from amine by washing with N hydrochloric acid. If the isoamyl alcohol solution is washed three times with an equal volume of aqueous acid, the loss of quaternary ammonium compound by transference to the aqueous phase is usually less than 0.2 per cent. The actual amount so transferred can be determined by combining the wash liquors and assaying by the bromophenol blue method of Barr, Oliver and Stubbings.1l The isoamyl alcohol solution is then diluted with water, a little isopropanol or denaturant-free ethanol being added if necessary to give a clear solution, which is then treated with phosphotungstic acid in the usual way. Deviation, 0.82 0.86 1.47 0.64 1-24 %Feb., 19561 AMMONIUM COMPOUNDS AS PHOSPHOTUNGSTATES 103 METHOD I t is assumed that the sample is free of interfering amines, i.e., that if the initial sample contained amines these have been removed by one or other of the procedures discussed above.RE AGENTS- Phosphotungstic acid-A 10 per cent. solution of P,0,.24W0,.xH20. Hydrochloric acid, 10 per cent. Sodium chloride. PROCEDURE- Place a sample, W g, containing quaternary ammonium compound equivalent to about 0.0002 to 0.0006 gram-molecules in a 400-ml beaker and make up to about 250 ml with water. Make the solution slightly acid with hydrochloric acid, and then add an excess (10 ml) of 10 per cent. hydrochloric acid.Add about 2 g of sodium chloride and raise the solution to the boil. Add 10ml of phosphotungstic acid solution and boil the mixture for 2 to 3 minutes, i.e., until the precipitate has coagulated. Test the supernatant solution for com- pleteness of precipitation by addition of a few drops of phosphotungstic acid. Cool the mixture to room temperature, collect the precipitate on a weighed No. 3 sintered-porcelain crucible and wash it with 100 to 200 ml of water. Dry the crucible at 105" C to constant weight and then ignite it at dull-red heat and reweigh it. If the molecular weight only of the quaternary ammonium compound is to be determined, the phosphotungstate may be precipitated from a much larger (unweighed) sample, and this precipitate collected and washed on a Buchner funnel.After being dried at 105" C, a 1 to 2-g sample of the quaternary phosphotungstate should be weighed in an ordinary porcelain or silica crucible, and reweighed after being ignited at a dull-red heat. The accuracy of the molecular-weight determination is thus increased. CALCULATION- The reactions that have taken place are- 3QX + H3P0,.12W03 -+ Q,P0,.12W03 (precipitate dried at 105" C) 4 HP0,.12W03 (residue left on ignition) (i) If the molecular weight of the quaternary ammonium compound is known, then- - quaternary ammonium compound, per cent. . . . . 100 W,M W(M-X + 959.3)- where M = molecular weight of quaternary ammonium compound, X = ionic weight of anion, W = weight of sample, and W , = weight of precipitate dried at 105" C.(ii) If the molecular weight of the quaternary ammonium compound is unknown, then- . . * - (2) . . . . . . . . . . M - 954.3 (Wl - W2) - 5 + x w2 where M , X and Wl are as in (i) and W2 = weight of residue left on ignition. (iii) If the molecular weight of the quaternary ammonium compound is unknown, and it is required to calculate the percentage present in the sample without intermediate determination or calculation of the molecular weight, then by combining equations (1) and (2) the following equation is obtained- = quaternary ammonium compound, per cent.As only a few anions are normally present in quaternary ammonium compounds, the last term in the above equation can be reduced to a constant by substituting the appropriate value of X to give- ’ E [ W , - KW,] = quaternary ammonium compound, per cent. W where K = 0.9681 for quaternary ammonium chlorides, K = 0.9215 for quaternary ammonium bromides, and K = 0.8722 for quaternary ammonium iodides. The authors thank the Directors of Milton Industrial Chemicals (London) Ltd, for permission to publish this paper. REFERENCE s 1. 2. 3. Flotow, E., Pharrn. Zentralh., 1942, 83, 181. 4 . 5. - , Ibid., 1954, 37, 379. 6. 7. 8. 9. 10. 11. Lottermoser, A., and Steudel, R., Kolloid Ztg.. 1938, 83, 37. Wilson, J. B., J . Ass. Ofl. Agric. Chem., 1946, 29, 311. Wilson, J. B., J . Ass. Off. Agric. Chem., 1952, 35, 455. Chinnick, C. C. T., and Lincoln, P. A., “Proceedings, 1st World Congress on Surface-active Agents, Lematte, L., Boinot, G., Kahane, E., and Kahane, Mme E., Compt. Rend., 1930, 191, 1130. Renard, T., J . Pharm. Belg., 1952, 7 , 403. Benk, E., Seifen-ole, 1953, 79, 289. Dubois, A. S., Ind. Eng. Chem., Anal. Ed., 1945, 17, 744. Barr, T., Oliver, J., and Stubbings, W. V., J . SOC. Chem. Ind., 1948, 67, 45. Paris 1954,” Chambre Syndicale Tramagras, Paris, 1955, Section 2, p. 41. MILTON INDUSTRIAL CHEMICALS (LONDON) LTD. 42-46 WEYMOUTH STREET LONDON, W.l October 7th, 1965

ISSN:0003-2654    

DOI:10.1039/AN9568100100

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

104 JEFFERY: THE SIMULTANEOUS PHOTOMETRIC DETERMINATION LVol. 81 The Simultaneous Photometric Determination of Molybdenum and Tungsten in Silicate Rocks BY P. G. JEFFERY A procedure is described for the simultaneous determination of molyb- denum and tungsten in silicate rocks. Fusion with sodium hydroxide under oxidising conditions is used to extract these elements as molybdate and tungstate, respectively. Silica is removed by evaporation of the alkaline solution with hydrochloric acid. The complexes of molybdenum and tungsten with u-benzoinoxime are then extracted from the acid solution. When these elements are present in similar proportions, the complexes with toluene- 3 :4-dithiol are formed, and are extracted together, the molybdenum and tungsten being determined simultaneously by photometric measurement at wavelengths of 630 mp and 680 mp. If the proportion of either of these elements greatly exceeds that of the other, then the Allen and Hamilton procedure is applied for the separation of molybdenum from tungsten.THE conditions for the quantitative formation and extraction of the complexes of molybdenum and tungsten with toluene-3 : 4-dithiol were established by Allen and Hamilton,' who des- cribed a method for the determination of these elements in biological materials. These workers recorded the existence of absorption maxima at wavelengths of 630mp for the molybdenum complex with toluene-3 : 4-dithiol, and at 680 mp for the tungsten complex. Provided that Beer's law is valid for the solutions used, and that there is no interaction between the two complexes when present together, then the optical density of a solution containing both metal complexes is given by- where t is the length of the light path: El and E , are the specific optical densities; and C, and C, are the concentrations of the component species.Measurement of the optical Log ],/I = t (E,G + EZC,),Feb., 19561 OF MOLYBDENUM AND TUNGSTEN IN SILICATE ROCKS 105 densities of solutions containing both molybdenum and tungsten complexes with toluene- 3: 4-dithiol at 630 mp and at 680 mp permit the concentrations C, and C, to be calculated from the equations- The values of the constants a and b at 630 mp and at 680 mp are determined experimentally, standard molybdenum and tungsten solutions being used. The complexes of molybdenum and tungsten with toluene-3 : 4-dithiol are not, however, formed under the same conditions, a sulphuric acid concentration of 6 to 14 N being required for the quantitative formation of the molybdenum complex.At this acid concentration the corresponding tungsten complex is not formed. For the quantitative formation of the tung- sten complex with toluene-3 : 4-dithiol, a pH of 0.5 to 2 is required. Between these pH values the formation of the molybdenum complex is not quite complete (Allen and Hamilton report "greater than 90 per cent."). In order to effect complete extraction of both molybdenum and tungsten, the following procedure is adopted. First, the complex of tungsten with toluene-3:4-dithiol is developed at a pH of 2 and a temperature of 97" C, together with the major portion of the corresponding molybdenum complex.The remaining portion of the molybdenum is then converted into the complex with toluene-3:4-dithiol by adding diluted sulphuric acid (1 + 1 v/v) to bring the acid concentration to approximately 10 N , and then a further quantity of the organic reagent solution. Complete formation of both complexes is thus obtained in the same solution, from which extraction with light petroleum will quantitatively remove both the molybdenum and the tungsten. In order to apply this procedure to the determination of molybdenum and tungsten, these elements are separated from the solution of the silicate rock, after removal of silica, by extraction with an organic solvent. The procedure of Allen and Hamiltonf for the extraction of these elements as complexes with cupferron into isoamyl alcohol may be used for this purpose.An alternative procedure, which may be employed for the extraction of molybdenum and tungsten, and which also effects a separation from all interfering ions, is extraction as the complex with or-benzoinoxime from an acid solution with chloroform. None of the remaining elements accompanying molybdenum and tungsten in the filtrate from an alkaline fusion interferes with the determination, with the exception of silica, which, on precipitation, removes a portion of both the molybdenum and the tungsten from the solution. This portion is recovered by dehydration of the silica with hydrochloric acid, evaporation with hydrofluoric acid, and fusion of the residue remaining after this treatment with a few milligrams of alkali carbonate.Chromium and vanadium, which, according to Knowles,2 also form insoluble precipitates with a-benzoinoxime, are not extracted from acid solutions under the conditions employed, nor does their presence in the solution interfere with the extraction of molybdenum and tungsten. Aluminium, iron, titanium, fluorine and phosphate, all of which may be present in the alkaline solution, are not extracted by the procedure described, nor do they interfere with the extraction of molybdenum and tungsten. Fusion of the finely ground silicate rock with sodium or potassium hydroxide and nitrate in a nickel crucible, in the manner described by Harwood3 for the determination of chromium and vanadium, is used to decompose the sample, and at the same time to effect a separation of molybdenum and tungsten from iron, manganese and other interfering metals.This method of decomposition is more rapid than fusion with alkali carbonate of a large (5-g) sample, although the method of Bennett and P i c k ~ p , ~ in which a platinum dish in an electric furnace at 1000" C is used, may also be used for this purpose. Fusion with alkali carbonate has no advantage over fusion with alkali hydroxide, as platinum dishes are required, and the time spent in extracting the solidified melt with water is, in general, much greater. EXPERIMENTAL The results obtained from the analyses of mixtures of standard molybdenum and tungsten solutions by the simultaneous photometric method described below are given in Table I.These results indicate that, at the wncentrations employed, Beer's law is valid, and that there is no interaction between the two complexes when formed and extracted from the same solution.106 JEFFERY : THE SIMULTANEOUS PHOTOMETRIC DETERMINATION TABLE I wol. 81 THE RECOVERY OF MOLYBDENUM AND TUNGSTEN FROM MIXED STANDARD MOLYBDATE AND TUNGSTATE SOLUTIONS Molybdenum trioxide added, Pg 10 10 20 10 20 30 Tungsten trioxide added, P8 10 20 20 30 30 30 Molybdenum trioxide recovered, Pg 9.8 9.8 19.9 10.0 20.1 30.2 Tungsten trioxide recovered, clg 10.1 20.2 20-1 30.0 29-6 29.3 The acid concentration for extraction of the complexes with a-benzoinoxime is not critical, both molybdenum and tungsten being completely recovered by four extractions from 0 to 1.8 N hydrochloric acid solutions, as shown in Fig.1. Molybdenum is extracted Hydrochloric acid concentration, N Fig. 1. Relation between percentage extraction and acid concentra- tion : curve A, 20 pg of molybdenum trioxide extracted with a-benzoinoxime and chloroform; curve B, 20 pg of tungsten trioxide extracted with a- benzoinoxime and chloroform more readily than tungsten; three extractions were found to be sufficient to transfer 20 pg of molybdenum to the organic solvent. As shown in Fig. 2, four extractions were found to be necessary to transfer a similar quantity of tungsten to the organic phase. Four extrac- tions at a hydrochloric acid concentration of 1.5 N were used subsequently for all extractions of molybdenum and tungsten from silicate rocks.TABLE I1 LOSS OF MOLYBDENUM AND TUNGSTEN OWING TO CO-PRECIPITATION WITH SILICA Molybdenum and tungsten not recovered from precipitated silica I 1 Sample MOO,, wo,, Pg l g o f granite .. .. .. .. 0.5 2.5 1 g of granite + 10 pg of molybdenum and 10 pg of tungsten .. .. .. 9.7 11.3 1 g of granite + 15 pg of molybdenum and 20 pg of tungsten .. .. .. 14.2 18.8 Molybdenum and tungsten recovered from precipitated silica c MOO,, wo,, Pg 1.1 3-2 11.1 13-2 16.9 23.6Feb., 19561 OF MOLYBDENUM AND TUNGSTEN I N SILICATE ROCKS 107 The loss of molybdenum and tungsten by co-precipitation with silica is shown by the results in Table 11. These were obtained by adding aliquots of standard molybdate and tungstate solutions to the rock solution after fusion, and before filtration to remove the iron oxides, and so on.METHOD REAGENTS- Water-The formation of a black toluene-3 : 4-dithiol complex, insoluble in light petroleum, during the course of the analysis was found to be due to the presence of copper in the water, derived from a copper still used for the distillation of water. Only water that had percolated through an ion-exchange demineraliser was subsequently used. Biodeminrolit mixed-bed resin was used for this purpose.6 Toluene-3: 4-dithiol-Dissolve 1 g of the reagent in 300 ml of a 1 per cent. aqueous sodium hydroxide solution. When solution is complete, add 5 ml of thioglycollic acid. Store this solution in a refrigerator. 1 2 3 4 Fig. 2, Relation between percentage extraction and number of extractions: curve A, 20 pg of molybdenum trioxide extracted with chloroform and a-benzoinoxime from N hydrochloric acid; curve B, 20 pg of tungsten trioxide extracted with chloroform and a-benzoinoxime from N hydrochloric acid Number of extractions Iron solution-Dissolve 1 g of pure iron wire in 20 ml of diluted sulphuric acid (1 + 1 v/v), and dilute to 1 litre.Light petroleum-The product X3B, supplied by the Shell Company of East Africa, with a nominal boiling range of 100" to 132" C, was used. Purify thissolvent before use by shaking with concentrated sulphuric acid, neutralising, and washing with water, Sulphzcric acid, diluted (1 + 1 v/v)-Mix equal volumes of water and concentrated sulphuric acid. Szdphuric acid, diluted (1 + 3 v/v)-Mix equal volumes of water and diluted sulphuric acid (1 + 1 v/v).a-Bemoilzoxime-Dissolve 2 g of reagent in 100 ml of ethanol. Standard molybdenum and tufigsten solutions-Dissolve 0.25 g of pure molybdenum oxide in a small quantity of ammonium hydroxide, and dilute to 250 ml with water. From this solution, prepare a working solution containing 10 pug per ml by dilution with water when required. Prepare the standard tungsten solution similarly, using 0.25 g of tungstic oxide. PROCEDURE- Fuse approximately 25 g of sodium hydroxide with 1 g of sodium nitrate in a nickel crucible until all water is expelled. When cold, transfer approximately 5 g of the silicate rock to the surface of the melt, re-melt the contents of the crucible, and maintain in a molten condition for 1 hour. Extract the cold melt with hot water containing a few drops of108 JEFFERY THE SIMULTANEOUS PHOTOMETRIC DETERMINATION [Vol. 81 ethanol.Collect the residue on a large hardened filter-paper, and wash the precipitate well with a hot 2 per cent. sodium carbonate solution. Discard this residue. Evaporate the filtrate to dryness with concentrated hydrochloric acid in a platinum dish to precipitate and dehydrate silica. Add 20 ml of concentrated hydrochloric acid and approximately 200 ml of water. When all soluble salts have passed into solution, collect the precipitated silica on an open-textured filter-paper, and wash the precipitate well with hot water. Reserve the filtrate. Ignite the precipitate in a platinum crucible, and remove the silica by evapora- tion with hydrofluoric acid in the usual way; avoid overheating the residue.Recover the traces of molybdenum and tungsten by fusion of this residue with a few milligrams of fusion mixture, and extraction with water. Add this solution to the filtrate from the silica precipitation. Transfer this solution quantitatively to a large separating funnel, and dilute to 250 ml. Add 2 ml of cc-benzoinoxime solution, and shake well. Add 10 ml of chloroform, and shake for a period of l g minutes to extract the molybdenum and tungsten into the organic phase. Allow the two phases to separate, and carefully withdraw the lower organic layer. Add 5 ml of chloroform to the aqueous phase in the separating funnel, and repeat the extraction for a further period of 14 minutes. Repeat this extraction with two further 5-ml portions of chloroform.Collect the four chloroform extracts in a 50-ml Kjeldahl flask, and distil off the organic solvent by placing the flask in a bath of boiling water. Add 1 ml of diluted sulphuric acid (1 + 3 v/v), and destroy the organic matter present by evaporation to fumes, first with nitric acid and then with perchloric acid. Add 20 ml of water, 0.5 ml of iron solution containing 1 g of iron per litre and 5 ml of toluene-3:4-dithiol solution. After a period of 1 hour, remove the flask, and add 12 ml of diluted sulphuiic acid (1 + 1 v/v) and 3 ml of toluene-3:4-dithiol solution, and then set aside for a further period of 1 hour. Cool, and by pipette put exactly 10ml of light petroleum in the flask, cork it well, and extract the molybdenum and tungsten into the organic phase by shaking for three separate periods of 1& minutes each.Separate the organic phase, and measure the optical density a t wavelengths of 630 mp and 680 mp, using a band width of 0.3 mp. For the instrument that I used (Uvispek, Hilger & Watts Ltd.) this corresponded to a slit width of 0.06 mm a t the wavelengths employed. When one element predominates, the following procedure, essentially the same as that described by Piper and Beckwith6 and Allen and Hamilton,l may be employed to separate molybdenum from tungsten. Proceed as described above to obtain the solution of molybdenum and tungsten in sulphuric acid, all the organic matter having been removed by evaporation to fumes of sulphuric acid with nitric and perchloric acids. Add 10 ml of diluted sulphuric acid (1 + 1 v/v), 7 ml of water and 1 ml of toluene-3:4-dithiol solution.Set this solution aside at room temperature for 1 hour. Add 10 ml of light petroleum, and extract the molybdenum - toluene-3 : 4-dithiol complex by shaking for three separate periods of 16 minutes as described above. Transfer the aqueous phase remaining after the extraction of molybdenum, or an aliquot of it containing not more than 40g of WO,, to a 50-ml Kjeldahl flask, and evaporate the solution to approximately 0.5 ml. Neutralise the remaining sulphuric acid with dilute ammonium hydroxide (2 N ) , and boil off the excess of ammonia. Add 1 ml of diluted sulphuric acid (1 + 3 v/v) and 20 ml of water. Add 3 ml of the toluene-3 : 4-dithiol solution, and place the flask on a boiling-water bath for a period of 1 hour.Cool the flask and its contents to room temperature, and add 10ml of light petroleum. Extract the tungsten complex into the organic phase by shaking for three separate periods of 13 minutes each. Remove the organic layer, and measure the optical density at 680mp. A determination of molybdenum and tungsten in the reagents should be made, and the blank thus obtained subtracted from the results obtained with the rock samples. Allow to cool. Place the flask in a bath of boiling water. Measure the optical density of the organic phase at 630mp.. DISCUSSION Molybdenum and tungsten may be determined simultaneously in silicate rocks provided that the ratio of the contents of these metals does not exceed about 1 to 12. No difficulty is experienced in determining molybdenum and tungsten when they are not present below this ratio, although the separation of molybdenum is then necessary.Feb., 19561 OF MOLYBDENUM AND TUNGSTEN IN SILICATE ROCKS 109 Results obtained on various samples by the method described in this paper and results obtained after adding known amounts of molybdenum and tungsten to them are shown in Table 111.TABLE I11 RECOVERY OF MOLYBDENUM AND TUNGSTEN ADDED TO SILICATE ROCKS Molyb- denum trioxide Sample found, Pg Granite, Singo . . .. . . 5.6 Mudstone, Karroo Beds, Entebbe.. 13.8 Granite, Mubende . . .. . . 5.5 Molyb- denum trioxide added, P8 10 5 10 Total molyb- denum trioxide found, PLg 16.0 19.0 16.2 Total Tungsten Tungsten tungsten trioxide trioxide trioxide found, added, found, Pg Pg CLg 58.0 10 66-0 10.5 5 15.5 18-3 5 23.0 Some further results for silicate rocks found in the Uganda Protectorate are shown in Table IV, together with results for the same rocks by the methods of Kuroda and Sandell? for molybdenum and of Sandella for tungsten.TABLE IV COMPARISON OF RESULTS Molybdenum trioxide found by present p.p.m. Sample method, Granite, Mubende . . .. .. 1.1 Granite, Entebbe . . .. .. 1.0 Mudstone, Karroo Beds, Entebbe . . Phonolite, Toror Hills, Karamoja . . 2.8 6.2 Molybdenum trioxide found by method of Kuroda and Sandell,‘ p.p.m. 1.2 1.3 3.1 7.9 Tungsten trioxide found by present method, p.p.m. 3.7 8.9 2.1 9.4 Tungsten trioxide found by method of Sandell,* p.p.m. 3.5 8.8 2.2 8-5 From these results it can be seen that remarkable agreement has been obtained, although there is a tendency for the results by Kuroda and Sandell’s method for molybdenum to be higher than those by the present method. This can probably be ascribed to the intrinsic error of the thiocyanate method, in which the thiocyanate complexes of molybdenum and tungsten are extracted together and reported as molybdenum. The author is grateful to the Acting Director, Geological Survey of Uganda, for permission to publish this paper. REFERENCES 1. 2. 3. 4. 6. 6. 7. 8. Allen, S. H., and Hamilton, M. B., AnaZ. Chim. Acta, 1953, 7, 483. Knowles, H. B., J . Res. Nut. Bur. Stand., 1932, 9, 1. Harwood, H. F., “Practical Rock Analysis for Geologists,” Imperial College of Science and Bennett, W. H., and Pickup, R., Colon. Geol. Min. Res., 1952, 3, 171. The Permutit Co. Ltd., “Deminrolit,” Monograph No. 7, 1953. Piper, C. S., and Beckwith, R. S., J. SOL. Chem. Id., 1948, 67, 374. Kuroda, P. K., and Sandell, E. B., Geochim. Cosmochim. Acta, 1954, 6, No. 1, 35. Sandell, E. B., Ind. Eng. Chem., Anal. Ed., 1946, 18, 163. Technology, London, 1933. THE GEOLOGICAL SURVEY OF UGANDA POST OFFICE Box No. 9 ENTEBBE, UGANDA First submitted, March 3 4 1955 Amended, September 6th, 1955

ISSN:0003-2654    

DOI:10.1039/AN9568100104

出版商:RSC    

年代:1956

数据来源: RSC