ISSN:0003-2654    

DOI:10.1039/AN95479FX017

出版商:RSC    

年代:1954

数据来源: RSC

摘要:

No. 20 April, 1954 THE SOCIETY FOR ANALYTICAL CHEMISTRY BULLETIN FORTHCOMING MEETINGS Ordinary Meeting of the Society, May 5th, 1954 AN Ordinary Meeting of the Society will be held at 7 p.m. on Wednesday, May 5th, 1954, in the Meeting Room of the Chemical Society, Burlington House, London, W.1. The following papers will be presented and discussed- “The Determination of Phosphate in the Presence of Soluble Silicates : Application to Basic Slag,” by H. N. Wilson, F.R.I.C. “The Spectrophotometric Estimation of Total Penicillins by Conversion to Penicillenic Acid and the Importance of Copper in Controlling the Reaction,” by F. G. Stock, M. Pharm., Ph.C., A.R.I.C. “Micro Method for the Determination of Bromide in Presence of Chloride,” by G. Hunter, MA., D.Sc., and A. A. Goldspink. Ordinary Meeting of the Society, July 21st, 1954 AN Ordinary Meeting of the Society will be held on Wednesday, July 21st, 1954, in the Lecture Theatre of the Royal Institution, 21, Albermarle Street, London, W.l.The meeting will deal With “The Use of Perchloric Acid in Analytical Chemistry,” and there will be a preliminary paper by Professor Harold Burton of Queen Elizabeth College, followed by a paper with demonstrations by Professor G. Frederick Smith of the University of Illinois. The demonstrations are believed to be quite striking. Professor Smith is making a special visit to Europe to give this paper, and it is hoped that a very large number of members will come to the meeting. Inaugural Meeting of the Western Section, May 8th, 1954 THE Council of the Society has approved the farmation of a Western Section following an application signed by many members resident in the West.The Inaugural Meeting will be held at 12 noon on Saturday, May Sth, 1954, in the Lecture Theatre of the Technical College, Newport (Mon). The President, Dr. D. W. Kent- Jones, and other members of Council will attend this inaugural meeting, and it is hoped that as many members as pxsible will support them. Following a short morning session, lunch will be taken at the Wedgate Hotel, after which the meeting will continue at 2.30 p.m. at the Technical College to complete the business of the day. After the business session, Dr. D. W. Kent-Jones will address the Section on “Alcohol Determination and its Medico Legal Aspects”.Joint Meeting of the Microchemistry Group with the London Section of the Royal Institute of Chemistry, May 7th, 1954 -4 JOINT Meeting of the Group with the London Section of the Royal Institute of Chemistry will be held at 6.15 p.m.on Friday, May 7th, 1954, in the Zoological Lecture Theatre, The University, Reading. The subject of the meeting will be “Microcliemical Methods in Biochemistry,” and the following papers will be presented and discussed- “The Determination of Esterases,” by \V, N. Aldridge. “The Determination of Sugars,” by G. Harris. “The Measurement of Isotopes of Carbon and Hydrogen,” by R. F. Glascock. During the afternoon visits will be made to the works of Messrs. Huntley and Palmers Ltd., Reading, and to the National Institute for Research in Dairying, Shinfield, nr.Reading. ,4t 5.45 p.m. there will be demonstrations of New Methods in Paper Chromatography and of Continuous Paper Electrophoresis Apparatus in the Agricultural Chemistry Research Laboratory. Meeting of the Physical Methods Group, May 28th, 1954 AN Ordinary Meeting of the Group will be held at the Harris Institute, Preston, Lancs., a t 7.30 p.m. on Friday, May 28th, 1954. The subject of the meeting will be “Fluorimetry,” and the following papers will be presented and discussed- “Quenching Effects in the Fluorimetric Determination of Uranium”, by G. N. Walton. “A Twin Beam Null Point Fluorinieter for Liquid Samples”, by J. P. Dowdall, A.R.C.S., The meeting will be preceded at 2 p.m. by a visit to Siemens Lamp Works. D.I.C., and H. Stretch, A.R.I.C.PAPERS ACCEPTED FOR PUBLICATION IN THE ANALYST THE following papers have been accepted for publication in The Analyst, and are expected to appear in the near future. It is not possible to enter into correspondence about any of them. “The Colorimetric Estimation of Niobium and Tantalum with Pyrogallol,” by E. C. Hunt and R. A. Wells. Colorimetric procedures are described for the determination of niobium and tantalum in mixtures of their oxides. The method is applied to the determination of the two metals in mixtures obtained by their chromato- graphic extraction from minerals and ores. The determinations depend on the formation of a coloured complex between tantalum and pyrogallol in acid solution and between niobium and pyi-ogallol in alkaline solution. Both systems obey Beer’s law, and, with 1-cm cells, the optimum limits of con- centration are 0 to 20 p.p.m.for niobium and 0 to 80 p.p.m. for tantalum. The effect of variation in pH and the interference of a number of cations and anions are recorded. “Inorganic Chromatography on Cellulose. Part XIV. A Shortened Chromatographic Method for the Determination of Niobium and Tantalum in Minerals and Ores,” by R. A. Mercer and R. A. Wells. A chromatographic procedure is described for the extraction of the mixed oxides of tantalum and niobium from minerals and ores. A solution of the sample in hydrofluoric acid containing ammonium fluoride is absorbed on cellulose and transferred to a 3-inch column of cellulose. The niobium and tantalum are completely extracted on passing 400 ml of ethyl methyl ketone containing 15 per cent.v/v of 40 per cent. w/w hydrofluoric acid through the column. The separation is complete from all metals other than tungsten. The two oxides, recovered from this solvent, are subsequently separated by further chromatography or determined without separation by a suitable colorimetric procedure.“‘Inorganic Chromatography on Cellulose. Part XV. A Rapid Chromatographic Method for the Determination of Niobium in Low-grade Samples,” by E. C. Hunt and R. A. Wells. .4 rapid and simple chromatographic method is described for the deter- mination of niobium in a hydrofluoric acid solution of an ore by upward diffusion on a paper strip. The niobium is detected as a yellow band on spraying the strip with aqueous tannic acid.An accurate determination of niobium is permitted by direct visual comparison of the band with standard strips. The chromatographic separation takes 20 minutes and a simple technique is described for carrying out ten separations simultaneously. The accuracy is 10 per cent. on ores containing >0-10 per cent. of niobium pentoside. “The Determination of Titanium by More Precise Absorptiometry,” by W. T. L. Neal. The titanium content of titanium-base alloys and pure titanium metal can be determined absorptiometrically with an accuracy (coefficient of variation) of 0-03 per cent., with a Unicam SP500 spectrophotometer at a wavelength of 4100 A, use being made of the colour of the titanium - hydrogen peroxide compound in solutions with an optical density of 2.5 to 3.0 in 1-cm cells.In this paper an analysis is made of the effect of factors liable to influence the precision and accuracy of the determination, and the techniques required to secure high precision are described in detail. ‘“The Volumetric Determination of Aluminium in Non-ferrous Alloys,” by G. W. (3. Milner and. J. L. Woodhead. The determination of aluminium can readily be accomplished by a volumetric procedure involving the addition of an excess of a standard ethylenediaminetetra-acetic acid solution t o the aluininiiini solution f~llowed by the back-titration of the amount of reagent unused in the forniation of the aluminium - ethylenediaminetetra-acetic acid complex. A standard iron solution is used for the back-titration and salicyclic acid is a suitable indicator for showing the titration end-point.Under these conditions it is possible t o determine up to 60 mg of aluminium with an accuracy of better than & l per cent. This titration has proved advantageous in the rapid analysis of various lion-ferrous materials including copper, zinc and magnesium-base alloys after the preliminary separation of the aluminium as its insoluble benzoate. “‘The Aspecific Detection of Preservatives in Foods by a Simple Fermentation Test, with Special Reference to Cured Meat Products,’’ by D. A. Mossel. The stability of non-sterile canned solid foods, which should not contain preservatives, is occasionally increased by adding low concentrations of highly toxic antimicrobial agents, e.g., derivatives of bromoacetic acid and phenylmercury compounds.Chemical methods for detecting each of the various preservatives are cumbersome and time-consuming. I t was therefore of interest to adopt the Kluyver fermentation test as an aspecific reaction for antimicrobial compounds. For testing the solid food (especially cured meat products), a portion is estractcd with 0-5 per cent. tartaric acid solution (pH 3.0 0.2) and a similar portion is extracted with 0.1 per cent. aqueous sodium hydroxide (pH 8.0 f 0.2). The estracts are pasteurised at pH 3, enriched by adding 0.5 per cent. of Difco yeast extract and 2.5 per cent. of dextrose, and the pH is adjusted t o 4.0. The estracts are then innoculated with sufficient bakers’ yeast to give lo4 cells per ml of solution and they are placed in Einhorn fermentation tubes.2” C, the volume of gas formed is measured. Brominated acetic acid derivatives can be detected when present in meat products at concentrations corresponding to 5 mg of bromine per kg. Sodium chloride, potassium nitrate and sodium nitrite do not interfere at the maximum levels found in meat products, i.c., 5, 0-2 and 0.02 per cent., respectively . After incubation for 24 to 30 hours at 24’“A Volumetric Method for the Rapid Assay of Palladium Jewellery Alloys,” by R. H. Atkinson. A volumetric method suitable for the assay of palladium has been developed. It is based on the precipitation of palladium as palladous iodide under appropriate conditions, with the precipitate as its own indicator. I t has been shown that the presence of 5 per cent. of nickel, iridium, platinum, rhodium, tungsten, molybdenum, copper and tin do not interfere within 5 parts in 1000; interference by gold and silver can be obviated by suitable modifications of procedure.In the analysis of a palladium - ruthenium alloy, 15-mg samples are dissolved in the minimum amount of concentrated aqua regia and the resulting solutions are titrated with 0.01 N potassium iodide, after adding hydrochloric acid and ferrous sulphate, the latter to prevent excess of aqua regia from reacting with the potassium iodide. The titration is continued until there is no cloud when a drop of the iodide solution is added; owing to the slowness with which palladous iodide settles, it is necessary to centrifuge a portion of the alloy solution before adding the test drop. “The Spectrophotometric Determination of Small Amounts of Oxygen in Waters,” by T.C. J- Ovenston and J. H. E. Watson. This paper describes the development of a method for the determination of dissolved oxygen in boiler feed waters applicable to concentrations in the range 0.001 to 0.1 ml of oxygen per litre, with a precision better than 0.001 a t the lower concentrations. The method is based on that of Bairstow, Francis and Wyatt, but the precision is improved by avoiding the use of starch and by determining the iodine release by means of the ultra-violet absorption of the tri-iodide ion. The procedure has been simplified by the elimination of the step involving volumetric dilution prior to photometric measurement. “An Examination of Scottish Heather Honey,” by T.J. Mitchell, E. 11. Donald and J. R. M. Kelso . Ling honey from the nectar of Calluna vulgayis is unique among European honeys in colour, taste, viscosity and the property of thixotropy. The present work sought to find an explanation of this difference by the determination of nitrogen, mineral matter (percentage of ash) and colloid content in samples of honey drawn from widely scattered districts in Scotland and Northumber- land. Forty-two samples were examined, of which 30 were predominantly ling honey. Correlation was found between the pH and the ash content of the honeys. There was some evidence of relationship between the colloid content and the total nitrogen and the thixotropy. The relation between these properties was not exact, probably because of the variation in floral source of the honeys. “Design and Operating Technique of a Vacuum Drying Oven. Part 11. Solids in Cane Molasses,” by S. D. Gardiner and F. J. Farmiloe. With a new carefully-designed vacuum oven, a drying method a t 69” to 70” C was devised for cane molasses. Values for true solids were determined to within 30.1 per cent. An ash correction was derived to permit solids determined by refractometry with the use of sucrose tables to replace the lengthy and difficult determination of solids by drying. Equations were derived relating true solids to refractometer solids, invert sugar and sulphate ash, and true solids to refractometer solids, invert sugar and sulphate ash minus a correction for sodium and potassium sulphates. This correction involves determinations with a flame photometer and gives more accurate results. PRINTED BY W. HEFFER & SONS LTD.. CAMBRIDGE. ENGLAND

ISSN:0003-2654    

DOI:10.1039/AN954790X019

出版商:RSC    

年代:1954

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN95479BX023

出版商:RSC    

年代:1954

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN95479FP045

出版商:RSC    

年代:1954

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN95479BP053

出版商:RSC    

年代:1954

数据来源: RSC

摘要:

APRIL, 1954 THE ANALYST Vol. 79, No. 937 PROCEEDINGS OF THE SOCIETY FOR ANALYTICAL CHEMISTRY ORDINARY MEETING AN Ordinary Meeting of the Society, organised by the Microchemistry Group, was held at 7.15 p.m. on Friday, January 29th, 1954, at the Sir John Cass College, Jewry Street, Aldgate, London, E.C.3. The Chair was taken by the President, Dr. D. W. Kent-Jones, F.R.I.C. A film entitled “Old Masters of Microchemistry” was shown and the following papers were presented and discussed: “Organic Ion Exchange,” by L. Saunders, B.Sc., Ph.D., F.R.I.C. ; “Inorganic Ion Exchange,” by G. H. Osborn, A.M.I.M.M., F.R.I.C. An Exhibition of Microchemical Apparatus was held in the laboratories of the College during the afternoon and evening. NEW MEMBERS ORDINARY MEMBERS Gordon Beswick, B.Sc.(Lond.), A.R.T.C. (Salford) ; Vincent Binns, M.Sc. (Manc.), F.R.I.C. ; Francis William Edward Diggins, Assoc. Inst .Med. Lab.Technology , Dip.Chem. Path. ; Charles Oliver Granger, B.Sc. (Lond.) , A.R.I.C. ; Peter John Jackson, B.Sc. (Lond.), A.R.I.C. ; Francis Raban Johnson, M.B.E., M.Sc. (Lond.), F.R.I.C.; Owen Eldred Mott, B.Sc. (Lond.); Edwin Arthur Robinson, B.Sc. (Lond.), A.R.I.C. ; Peter Haydock Scholes, A.Met. (Sheff.), L.I.M. ; Geoffrey Spencer, M.Sc. (Lond.), A.R.I.C.; Ernest Eric White, F.R.I.C. JUNIOR MEMBERS John Buckett, B.Sc. (Lond.) ; Bernard Collier; James Clinch, B.A. (Cantab.) ; Alan Roy Gunningham, B.Sc. (Lond.) ; Ralph Stem, B.Sc. (Lond.), A.R.I.C. DEATHS WE regret to record the deaths of Harold Grovett Colman Cresacre George Moor. NORTH OF ENGLAND SECTION AN Ordinary Meeting of the Section was held at 2 p.m.on Saturday, November 28th. 1953, at the City Laboratories, Mount Pleasant, Liverpool. The Chairman, Mr. T. W. Lovett, F.R.I.C., presided over an attendance of about 100. At this meeting, Dr. R. L. M. Synge, F.R.I.C., F.R.S., delivered a lecture entitled “Principles of Chromatography” (see summary below). PRINCIPLES OF CHROMATOGRAPHY DR. R. L. M. SYNGE defined chromatography as a method for separating substances by the percolation of fluids through porous and powdered media, and explained the counter-current nature of the chromatographic process. Dr. Synge said that he would consider th-ree types of chromatographic operating procedures : (i) elution development, (id) frontal analysis and (iik) displacement development.186186 PROCEEDINGS [Vol. 79 (i) In elution development, compact bands resulted only with linear isotherms. Under practical conditions isotherms were often curved, which resulted in diffuse bands. With linear isotherms, bands could sometimes be made sharper by the use of slower rates and finer powders. (ii) In frontal analysis a series of steps was obtained, the original solution finally appearing at the base of the column. A practical example was the stripping of a minor component by ion exchange from a large volume of solution. (iii) In displacement development, components were removed in turn from the adsorbent by the use of a substance that was more strongly held by the adsorbent. There were no gaps between the substances as they appeared at the base of the column, and the technique was useful for preparative work.The technique could be modified by the introduction of a substance of intermediate affinity. Examples were the separation of amino-acids with different alcohols, and the adsorption chromatography of sugars and oligosaccharides. Dr. Synge considered four different types of equilibria. These were gas - liquid, gas - solid, liquid - liquid and liquid - solid, Gas - liquid chromatographic techniques were of recent introduction, and an excellent example was furnished by the original work of A. T. James and A. J. P. Martin on the separation of the volatile fatty acids. The relatively non-volatile liquid adsorbent, e.g., high-boiling paraffin, was supported on a solid such as kieselguhr.Special methods of detection, such as those based on thermal conductivity and vapour density, might be required for this type of chromatography. Gas-solid chromatography had so far usually involved a displacement technique, and an example was furnished by S. Claesson’s and C. S. G. Phillips’ work involving separation of paraffin hydrocarbons on carbon. The use of liquid-liquid techniques covered a very wide field, including most substances of biochemical interest. As an illustration of the pharmacological applica- tions of this type of system, an interesting study had been made of the changes that took place in the metabolism of an animal when fluoroacetic acid was administered. There were numerous applications to inorganic separations by partition between an aqueous phase held on paper and an organic solvent.Ion-exchange resins were versatile and were widely used in liquid - solid chromato- graphy. Tswett’s work with columns of calcium carbonate and light petroleum solvent furnished an excellent example of a liquid - solid process. Such systems were applied in about 1930 to the carotenoids and vitamin A. By displacement on ion-exchange resins, separations of amino-acids had been made by Partridge at Cambridge on a fairly large scale. Separations of the rare-earth and trans-uranium elements had also been made on ion exchangers. Fraction cutters were used in conjunction with columns of adsorbent, and there were many different types. Other detection methods involved the use of paper chromatography: radio-autography and bio-autography were important examples. For compounds of molecular weight below 500, chromatography could give a relatively easy method of separation.There was an enormous choice of methods and it was generally only necessary to find the right solvent. There were difficulties with large molecules-it was difficult to find two phases in which distribution was at all even, and the molecular dimensions of the pores in chromatographic working materials often led to interference. Finally, Dr. Synge mentioned his latest work on the separation of large molecules on collodion membranes. Dr. Synge emphasised the use of proper detection methods. THE Twenty-ninth Annual General Meeting of the Section was held at 2.15 p.m. on Saturday, January 30th, 1954, at the Engineers’ Club, Albert Square, Manchester, under the Chairman- ship of Mr.T. W. Lovett, F.R.I.C. The following Officers and Committee Members were elected for the forthcoming year :-Chairman-Mr. T. W. Lovett. Vice-Chairman-Mr. J. R. Walmsley. Hon. Secretary and Treaszcrer-Mr. Arnold Lees, 87, Marshside Road, Southport, Lancs. Elected Committee Members-Messrs. H. Childs, R. Crosbie-Oates, H. Dedicoat, K. A. Hyde, A. 0. Jones and A. N. Leather. Hon. Auditors-A. Alcock and C. J. House.April, 19541 PROCEEDINGS 187 The Annual General Meeting was followed by an Ordinary Meeting of the Section, at which the President, Dr. D. W. Kent-Jones, F.R.I.C., gave an address on “The Society for Analytical Chemistry. ” SCOTTISH SECTION A JOINT Meeting of the Section was held with the Stirlingshire Sections of the Royal Institute of Chemistry and the Society of Chemical Industry at 7.30 p.m.on Wednesday, December 16th, 1953, at the Lea Park Restaurant, Callendar Road, Falkirk. At this meeting Dr. R. L. M. Synge repeated his lecture on “Principles of Chromato- graphy” (see summary above). THE Nineteenth Annual General Meeting of the Section was held in Glasgow on Wednesday, January 20th, 1954, at 12.45 p.m., under the Chairmanship of Mr. R. S. Watson, F.R.I.C. The following office bearers were elected for the forthcoming year:--Chairman-Mr. R. S. Watson. Vice-Chairman-Dr. F. J. Elliott. Hon. Secretary and Treasurer-Mr. J. A. Eggleston, Boot’s Pure Drug Co. Ltd., Motherwell Street, Airdrie, Lanarkshire. Elected Committee Members-Dr. Christina C.Miller and Messrs. J. M. Malcolm, J. K. McLellan, H. C. Moir, R. T. Potter and A. F. Williams. Hon. Auditors-Messrs. J. Andrews and J. W. Gray. MICROCHEMISTRY GROUP THE Tenth Annual General Meeting of the Group was held at the Sir John Cass College, Jewry Street, Aldgate, London, E.C.3, on Friday, January 29th, 1954, at 7 p.m., and the following Officers and Committee Members were elected for the forthcoming year:- Chairman-Dr. A. M. Ward. Hon. Secretary-Mr. A. Bennett, The Brewing Industry Research Foundation, Lyttel Hall, Nutfield, Redhill, Surrey. Hon. Treasurer-Mr. G. Ingram. Elected Committee Members-Messrs. W. Brown, G. S. Crouch, D. F. Phillips, J. T. Stock, C. L. Wilson and D. W. Wilson. Hon. Auditors-Messrs. L. H. N. Cooper and H. Childs. Vice-Chairman-Dr.G. F. Hodsman. PHYSICAL METHODS GROUP THE Ninth Annual General Meeting of the Group was held at 6 p.m. on Tuesday, November 24th, 1953, in the Meeting Room of the Chemical Society, Burlington House, London, W.I. The Chairman of the Group, Dr. J. Haslam, presided. The Group Officers and Elected Members of the Committee for the forthcoming year are as follows:-Chairman-Mr. A. A. Smales. Hon. Secretary and Treasurer-Mr. R. A. C. Isbell, Hilger & Watts Ltd., Hilger Division, 98, St. Pancras Way, London, N.W.l. Members of Committee-Messrs. W. H. Bennett, W. Furness, H. M. N. H. Irving, A. G. Jones, F. C. J. Poulton and R. A. Wells. Hon. Auditors-Messrs. C. A. Bassett and D. C. Garratt. The Annual General Meeting was followed by the Forty-second Ordinary Meeting of the Group, at which the new Chairman, Mr.A. A. Smales, took the Chair. The retiring Chairman, Dr. J. Haslam, gave an address on “Physical Methods in the Analysis of Plastic Materials-Some Observations of an Analytical Chemist’’ (see summary below). Vice-Chairman-Dr. J. E. Page. PHYSICAL METHODS OF ANALYSIS OF PLASTIC MATERIALS-SOME OBSERVATIONS OF AN ANALYTICAL CHEMIST DR. J. HASLAM, in presenting a survey of the methods used in the examination of plastics of various kinds, said that, although some of them were chemical and some combinations of chemical and physical procedures, many were purely physical. Amongst the most important of the purely physical methods was the determination of relative viscosity; in this different solvents were used for different types of plastic: nylon was dissolved in 90 per cent.formic acid, polythene in tetrahydronaphthalene, poly- methyl methacrylate in chloroform, Terylene in o-chlorophenol and polyvinyl chloride in ethylene dichloride. Although the chemist would realise that the only figure determined was the relative viscosity, the results were, except for those found for nylon, usually calculated to arbitrary188 PROCEEDINGS [vol. 79 values by means of empirical factors. Polythene and polymethyl methacrylate were reported as molecular weight, polyvinyl chloride as the so-called K value and Terylene as intrinsic viscosity. In the analysis of nylon, information on the speed and extent of hydrolysis by hydrochloric acid was desirable. This knowledge, difficult to get by chemical means, was easily obtained by potentiometric titration of the hydrolysis products, a physical method.For nylon 66 the end-points were at pH 2.6 for the free hydrochloric acid (hexamethylenediamine dihydrochloride is neutral) and at pH 8 for the adipic acid. Similarly for nylon 610, the first end-point at pH 3.7 to 3.8 marked the titration of the free hydrochloric acid and the second at pH 8.2 corresponded with that of the sebacic acid. For nylon 6 the end-points were at pH 2-6 for the free hydrochloric acid and pH 8.0 for the hydrochloric acid of the eaminocaproic acid hydrochloride. Another physical method, chromatography, has also given valuable results in the examination of nylon; Zahn has described separations on paper that have been useful in dealing with small amounts, and Miss Ayers has shown that nylon polymers can be dealt with by straight chromatography.In nylon analysis, partition methods have also been useful. From the hydrolysis product of copolymers of hexamethylenediamine with adipic and sebacic acids, the two acids can be isolated, equilibrated between ether and water, and so determined by a procedure that has been described in The Analyst. In the analysis of Perlon U, which is not an ordinary nylon polymer, the acid hydrolysis products are hexamethylenediamine hydrochloride and tetrahydrofuran. The furan derivative reacts with an excess of hydrochloric acid to form 1 :4-dichlorobutane, a substance difficult to deal with. The difficulty was overcome by taking advantage of two physical properties of tetrahydrofuran : its low specific gravity and its partial solubility in water.Once separated, the tetrahydrofuran can be identified by conversion to succinic acid. Of the purely physical methods used in nylon analysis, the examination of the infra-red spectrum of the polymer was the most important of all, for it had reduced considerably the amount of chemical work that would otherwise have been necessary to characterise the polymer. Again, for polymethyl methacrylate the method of attack was a combination of physical and chemical methods. The plasticiser was first removed from a solution of the plastic in acetone by precipitating the polymer with light petroleum, and the insoluble polymer was then submitted to vacuum distillation to recover the monomer, which can be identified by its saponification value and refractive index.Copolymers of methyl methacrylate with substances such as styrene, cyclohexyl methacrylate and ethyl acrylate are amenable to the same treatment. Polyvinyl chloride tubing presents a more complicated problem. In addition to the basic polymer or copolymer it could contain a mixed plasticiser, a stabiliser, a pigment, a lubricant and possibly a filler. The plasticiser is removed by extraction with ether; the polymer is then dissolved in tetrahydrofuran, and the fillers and so on are removed by centrifugation. The polymer is recovered by precipitation with ethanol and examined by chemical and physical methods; the polymer itself is examined by infra- red spectroscopy and its chlorine content is determined by fusion with sodium peroxide in a stainless-steel bomb and potentiometric titration of the resulting chloride with standard silver nitrate solution with silver-wire electrodes.The morpholine test is used for polyvinylidene chloride and the methanol - alkali - pyridine test for polyvinyl chloride. The much used principle of solution of a plastic in a solvent, followed by precipita- tion with a non-solvent, is not confined to the recovery of the pure polymer; it is frequently used to release from a preparation substances not amenable to direct solvent extraction. For example, after precipitation of polymethyl methacry late by light petroleum from a solution in acetone, the dibutyl phthalate plasticiser can be recovered from the petroleum solution by evaporation and weighed.Likewise a polythene - polyisobutylene mixture can be separated from a solution in toluene by means of light petroleum and the polyisobutylene recovered from the filtrate for identification by means of its infra-red absorption spectrum. Elementary analysis of hydrocarbons of this kind is useless, for both polythene and olyisobutylene have the same carbon and hydrogen values. Free phenols are li I! erated from phenol - formaldehyde and cresol - formaldehydeApril, 19541 PROCEEDINGS 189 resins by solution in alkali and precipitation of the resin at pH 4.5; they can then be identified and determined in the filtrate. The absorbers of ultra-violet light that are added to Perspex to reduce the amount of yellowing caused by sunlight are also separated by precipitation of polymethyl methacrylate resin with a non-solvent from a solution in acetone, after which they can be identified and determined by methods that have already been published; the results are checked by measuring the optical density of the solution of the original polymer at 308 mp for methyl salicylate, at 312 mp for phenyl salicylate and at 289 and 324 mp for 2 :4-dihydroxybenzophenone, the three substances in common use.Lauryl mercaptan, added to Perspex to stabilise it towards heat, is determined by amperometric titration with silver nitrate, on the principle of Kolthoff and Hams, by means of a rotating platinum-wire indicator electrode and a reference half-cell con- taining mercury in contact with a potassium iodide - mercuric iodide and potassium chloride solution connected to the test solution by a salt bridge.Catechol is released from nylon by hydrolysis with hydrochloric acid in a sealed tube and determined spectroscopically by means of its ultra-violet absorption at 275 mp. In the analysis of plasticisers, physical methods are of great importance and are used in conjunction with chemical analysis. For mixed plasticisers, direct infra-red examination is often unsatisfactory, e.g., for mixed butyl, octyl and nonyl esters. These are hydrolysed with potassium hydroxide in ethylene glycol and the resulting alcohols are determined by their infra-red absorption spectra. Spectroscopic methods are amongst the most important of those used in the plastics industry. They are used in a wide variety of analyses.In the ultra-violet region, acenaphthylene is determined in its copolymers by its absorption at 295 mp in chloroform solution; the purity of dipotassium phthalate from a plasticiser is checked in approxi- mately 0.1 N hydrochloric acid at 276 mp. The ultra-violet spectrum is also used in many other tests for additives or for stabilisers such as diphenylurea and diphenyl- thiourea. The visible region is used for dyestuffs, for example, Durindone Pink FF in Terylene, at 555 mp, after extraction with o-chlorophenol. Inorganic fillers are identified by photographing the spectrum of the sample in juxtaposition with the iron spectrum, which serves as wavelength standard. A recently devised piece of apparatus for showing whether or not Perspex is plasticised with dibutyl phthalate consists of a mercury- vapour lamp, a Beck ultra-violet spectroscope and a fluorescent screen.A sheet of ordinary &-inch Perspex placed in the light beam cuts off the spectrum at 290 mp; with plasticised material the cut-off is at 295 mp. Polarographic methods are used in the analysis of plastics, particularly for zinc, after removal of organic matter, and for the detection of phthalate esters in commercial plasticisers. These sometimes contain castor oil, which causes the usual resorcinol and phenolphthalein tests to fail. In plasticisers of this type, qualitative evidence of the presence of phthalates is given by recording a polarogram on a mixture of the plasticiser with tetramethylammonium iodide in methanol over the range -1.0 to -2.0 volts.If dibutyl phthalate is present, reduction occurs at -1.45 volts versus the mercury pool anode. Phosphates, ricinoleates, citrates, sebacates, Cerechlor and Mesamoll do not interfere. For the determination of coconut-shell flour in a phenol - formaldehyde moulding powder, the lycopodium method of Wallis has been found useful. Many of the methods mentioned above have been described in detail in The Analyst and in The Journal of Applied Chemistry. The microscope is used in the examination of fillers. BIOLOGICAL METHODS GROUP THE Ninth Annual General Meeting of the Group was held in the Anatomy Lecture Theatre, University College, London, W.C.l, at 4.30 p.m. on December llth, 1953. The following Officers and Committee Members were elected for the forthcoming year :-Chairman- Dr. Leslie J. Harris. Hon. Secretary and Treasurer- Mr. s. A. Price, Walton Oaks Experimental Station, Vitamins Ltd., Dorking Road, Tadworth, Surrey. Members of Committee-Messrs. E. M. Bavin, W. A. Broom, H. 0. J. Collier, K. A. Lees, H. Pritchard and G. F. Somers. HOB. Azcditors-Messrs. D. M. Freeland and J. H. Vice-Chairman-Mr. K. L. Smith.190 MACNULTY, REYNOLDS AND TERRY: THE POLAROGRAPHIC Pol. 79 Hamence. Mr. J. W. Lightbown has consented to attend Committee Meetings in the capacity of Hon. Recorder. During the afternoon preceding the Annual General Meeting and the evening following the meeting, an Ordinary Meeting of the Group was held jointly with the Pharmaceutical Society on “The Assay and Detection of Pyrogens.” About 200 members and visitors were present . The Afternoon Session was introduced by the Chairman, Professor J. P. Todd, and the following papers were presented and discussed : “The Occurrence and Importance of Pyrogens,” by T. D. Whittet, B.Sc., Ph.C., D.B.A., A.R.I.C. ; “Routine Pyrogen Testing,” by K. L. Smith, M.P.S.; “The Leucocyte Response in the Rabbit to the Pyrogen from Pyotezcs vulgaris. Mononuclear and Temperature Responses,” by M. Dawson, Ph.C., and J. P. Todd, Ph.D., Ph.C., F.R.I.C. Dr. H. 0. J. Collier took the Chair for the Evening Session, and the following papers were presented and discussed: “Rabbit Responses to Human Threshold Doses of a Bacterial Part I. Pyrogen,” by J. G. Dare, Ph.D., Ph.C.; “Standards of Pyrogenic Activity,’’ by W. L. M. Perry, M.D.

ISSN:0003-2654    

DOI:10.1039/AN9547900185

出版商:RSC    

年代:1954

数据来源: RSC

摘要:

'190 MACNULTY, REYNOLDS AND TERRY THE POLAROGRAPHIC pox. 79 The Polarographic Determination of Fluoride Part I. Basic Principles of the Method: Application to the Cathode-ray Polarograph BY B. J. MACNULTY, G. F. REYNOLDS AND E. A. TERRY (Presented, together with Part II*, at the meeting of the Physical Methods Group on Tuesday, APriZ 1&h, 1953) Modern interest in traces of fluoride has revealed the need of more sensitive methods of fluoride determination. In this paper a sensitive method for determining fluoride polarographically is described. It is based on the depression by fluoride of the polarographic step given by the reduc- tion of the aluminium - Solochrome Violet R.S. complex. The step depression is shown to be linearly related to the amount of fluoride down to 0.001 p g per ml.The use of this method with the cathode-ray polarograph is described. DISCOVERIES made in the past twenty years have resulted in a greatly increased use of fluoride and a realisation of the importance of trace quantities of this ion. The researches into dental caries have led to the critical consideration of the safe fluoride concentration in drinking water. The effects of fluoride on animals and human beings have resulted in an increased interest in the fluoride content of fuels because of possible contamination of atmosphere and herbage. Industrially fluoride has found use in catalysts, insecticides and as a food preserva- tive, with the resulting interest in the trace amounts appearing in the final products. The quantity of fluoride in strata and fossils has also been used as an aid to archaeological dating.With the increased interest in fluoride much attention has been paid to its detection and numerous methods of determination have been suggested. The following is a list of older methods that have some proved usefulness, together with the more promising of the recent methods- (1) Titration with thorium ~ ~ i t r a t e . ~ s ~ s ~ s * , ~ (2) Methods based on the use of alizarin c o m p l e ~ e s . ~ ~ ~ ~ ~ (3) The use of ferric salicylate or sulphosalicylate.9~10 y 1 l (4) The use of the thorium - thoron complex.12 (5) Fluorescent methods.13 9 1 4 (6) The use of the aluminium - haematoxylin complex.16 (7) The use of the aluminium - Eriochrome cyanine complex.16p17 * To be published shortly in The Analyst.April, 19541 DETERMINATION OF FLUORIDE.PART I 191 Except with samples of potable waters, all methods usually require the prior separation of fluoride by distillation as hydrofluosilicic acid and this usually provides some degree of concentration of the ion. Although the above methods are adequate for various applications in which very low concentrations of fluoride are encountered and the sample size is limited, even the aluminium - haematoxylin method, which requires 25 ml of sample containing 0.1 pg of fluoride per ml as a minimum, is not wholly satisfactory. An example is the fluoride content of hydrochloric acid, a knowledge of which is required whenever this acid is used as a reagent. It is clear that a more sensitive method of fluoride determination is still required in certain circumstances.From previous work13J4~15~10~17 it has been shown that any really sensitive means of determining aluminium would provide a basis for an equally sensitive method for fluoride. The publication by Willard and Deanla of a method for the polarographic determination of aluminium down to a concentration of 0.1 pg per ml made it probable that, with this as a basis, a very sensitive method for fluoride could be developed. Aluminium - dye complex EXPERIMENTAL DEVELOPMENT BASIC PRINCIPLES- The Willard and Dean method is based on the measurement of the polarographic step produced by the reduction of Solochrome Violet R.S. (the sodium salt of 5-sulpho-2-hydroxy- benzeneaz0-2-naphthol), which is known as Pontachrome Violet S.W.in the United States. This compound produces a step with a half-wave potential of about -0.3 volt zlersw the saturated calomel electrode in an acetate buffer solution of pH 4-6. Addition of aluminium to a solution of this dye causes the polarographic step to be reduced in height and a second step to appear at a point about 0.2 volt more negative. Fig. 1 illustrates this effect. The Fig, 1. Curve A: typical step for dye; curve B: typical step for aluminium and dye I I I I 0 2 4 6 8 I 1 Concentration of fluaridc in final solution, ,ug per ml Relation between height of aluminium Fig. 2. step and concentration of fluoride sum of the height of these two steps is equal to the height of the original step, and the size of the second step is proportional to the concentration of aluminium present. In their paper Willard and Dean stated that “fluoride forms a more stable complex with aluminium than does the dye” and their figures for fluoride interference indicated an effect of reasonable magnitude.To decide whether the effect of fluoride was such as to render the development of a method practicable, the following experiment was carried out- A 2-ml portion of a 0.025-mg per ml solution of aluminium in dilute perchloric acid was neutralised to methyl red with sodium hydroxide and treated with 1 ml of 5 N perchloric acid and 5 ml of 2 N ammonium acetate. Up to 15 ml of a standard solution (33.3 pg per ml) of fluoride were then added (giving a range of fluoride concentration in the final solution of 0 to 10 pg per ml) followed by 20 ml of a 0.05 per cent.aqueous192 MACNULTY, REYKOLDS AND TERRY: THE POLAROGRAPHIC [Vol. 79 solution of dye. The solution was made up to 50 ml, heated at 70" C for 5 minutes, cooled, and an aliquot was deoxygenated and examined polarographically . The results, which are plotted in Fig. 2, showed a satisfactory straight-line relationship between the height of the aluminium step and the concentration of fluoride up to 6 pg per ml. At this point the gradient changed and another relationship of smaller slope governed the range 6 to 10 pg per ml of fluoride. The significance of this change will be discussed later. DETERMINATION OF FLUORIDE XN THE RANGE 0.1 TO 0-6pg PER ml- First a method was developed for use in the range 0.1 to 04pg per ml, where other methods of determination could be used to check the results.The following experimental technique was used- To each of a series of 50-mi flasks were added, in solution, 30pg of aluminium, and then 0, 5, 10, 15, 20 and 25 pg of fluoride to successive flasks. To each flask were added 2.5 ml of 2 N ammonium acetate, 0.5 ml of 5 N perchloric acid, 2 ml of 0.05 per cent. dye solution and 1.0 ml of proteose peptone (maximum suppressor). The solutions were made up to 50 ml, heated at 70" C for 5 minutes, cooled, and an aliquot taken for polarography. The pH of the solution at the end of this procedure was 4.6. This technique was the pattern for all succeeding experiments, the concentration of reagents and the pH being altered as desired. The results were encouraging, but it was considered that the differences between replicates were too great, and it was decided to check whether the pH value of 4.6, at which the experi- ments were carried out, was the most satisfactory one.Accordingly the depression of the aluminium - dye step resulting from the addition of a standard amount of fluoride was plotted against pH, and the result (Fig. 3) clearly indicated that a pH value of about 3.9 was the most suitable for fluoride determination; all further work was carried out at this value. Fig. 3. Effect of change of pH on difference in step height brought about by the addition of 0.2 pg of fluoride per ml I I I I I 0.01 0.02 0.03 0.04 005 Fluoride cancentrarion,)ug per rnl Fig. 4. Effect of time. Curve A: samples examined immediately after heating ; curve 13 : samples examined 20 hours after heating The experiments at pH 4-6 were now repeated at a pH value of 3.9, and the results in Table I show that the differences between replicates were satisfactorily small.TABLE L DEPRESSION OF ALUMINIUM - DYE STEP DUE TO FLUORIDE CONCENTRATIONS OF 0 TO 0.6p.g PER Id AT pH 3.9 Step-height, arbitrary units Fluoride, Expt. 1 Expt. 2 Expt. 3 Expt. 4 Expt. 5 Expt.'s pg per 50 ml 0 72 72 69 72 72 72 5 63 63 66 65 66 65 10 54 56 57 56 66 57 15 50 48 48 50 5l 49 20 41 40 39 42 44 42 30 30 29 28 30 32 29April, 19541 DETERMINATION OF FLUORIDE. PART 1 193 The level of fluoride determination reached at this stage (0.3 to 0.6 pg per ml) was only of the same order as that possible by other methods of analysis, and at the same time the limit set by the use of conventional instruments had almost been reached.The cathode- ray polarograph, an instrument of much greater sensitivity, developed in this department by Davis, Reynolds and S e a b o ~ n , ~ ~ , ~ ~ was therefore used. THE USE OF THE CATHODE-RAY POLAROGRAPH- Comparative tests between the cathode-ray polarograph and conventional instruments showed that a similar straight-line relationship between depression of the aluminium - dye step and fluoride concentration (0 to 0 4 p g per ml) existed on both types of instrument. The method was next tested at a lower level of fluoride concentration, 0.01 to 0.05 pg of fluoride per ml, by the same technique with an approximately tenfold decrease in con- centration of reactants, The results obtained by polarography of the solutions immediately showed that, although an effect similar to that expected was produced, the relationship was not strictly linear (Fig.4). If, however, the solutions were set aside overnight for 16 to 20 hours before polarographic measurement, the relationship became linear and had a slope of the expected order (see Fig. 4). No advantage was gained by allowing the solutions to stand for longer and, in fact, after about 30 hours the magnitude of the fluoride effect began to diminish. In Table I1 a number of calibrations on this range of fluoride concentration are shown. As these calibrations were made with different solutions and with material from different sources, and as there were slight variations in the excess of dye present, it is not surprising to find some variation in the recorded step height, although all experiments showed the expected straight-line relationship.Experiments I to 6 of Table 11. were carried out in a water - perchloric acid medium and experiments 7 to 11 in 0.1 M perchloric acid made 0.1 M with respect to hydrochloric acid. TABLE 11 COLLECTED CALIBRATIONS WITH VARIOUS SOLUTIOPU’S OF REAGENTS AND MATERIALS TO SHOW THE EFFECT OF FLUORIDE ON THE STEP HEIGHT OF A STANDARD ALUMINIUM - DYE SOLUTION Measurements on cathode-ray polarograph at sensitivity 20 kR 7 Expt. Fluoride, 1 0 75 0.01 - 0.02 58 0.03 - 0.04 47 0.05 - 0.06 30 0.07 - 0,OS 24 PLg Per ml Expt. 2 65 67 44 36 24 - - - - Step height, arbitrary units -- A Expt. Expt. Expt. Expt. Expt. Expt. 3 4 6 6 7 8 90 96 84 45 68 62 83 a4 74 - 01 - 75 73 62 37 55 52 55 - 66 73 - 59 68 45 31 47 40 51 60 36 - 43 36 - - - 23 - I - - - - - - - Expt.9 86 77 73 64 58 52 - Expt. I 0 83 74 69 61 65 48 - 1 Expt. 11 77 70 64 58 52 48 Experiments 1 to 6 in water - perchloric acid medium; experiments 7 to 11 in 0-1M hydrochloric acid - 0-1M perchloric acid medium. To test the reproducibility, a series of calibrations were carried out by the standardised procedure and with bulked reagents, one or two calibrations being performed daily. The results (Table III), indicate a satisfactory degree of reproducibility. The step heights are measured on a graticule that can be read to half a division, and except a t 0.10 pg per rnl of fluoride, which is near the limit of this range of concentration, differences do not amount to more than one division (04.2 pg of fluoride).SUB-MICROGRAM AMOUNTS OF FLUORIDE- The method was finally extended to deal with sub-microgram mounts of fluoride in the range 0.001 to 0.01 pg per ml. When the technique previously described was used with a further tenfold decrease in reactants, it was found that, ’ although with bulked reagents satisfactory reproducibility could be obtained, the number of “spoilt” results was too high for the method to be reliable.194 MACNULTY, REYNOLDS AND TERRY: THE POLAROGRAPHIC [Vol. 79 It was found that at these low levels of fluoride concentration even AnalaR reagents contained sufficient impurities, such as iron and aluminium, to vitiate the method if they were not removed. A satisfactory method of purifying reagent solutions is by the addition of acetylacetone and extraction of the complexed impurities with chloroform under suitable conditions.TABLE I11 EFFECT OF FLUORIDE ON STEP HEIGHT OF A STANDARD ALUMINIUM-DYE SOLUTION Measurements on cathode-ray polarograph at sensitivity 20 kQ : standardised procedure and bulked reagents used Step height, arbitrary units Fluoride, Expt. I /G Per 0 52 0.02 49 0-04 45 0.06 41 0.08 37 0.10 33 Expt. 2 Expt. 3 Expt. 4 Expt. 5 52 55 51 56 48 51 47 52 44 46 43 43 41 42 39 44 37 38 35 40 34 32 33 35 7 - Expt, 6 Expt. 7 54 53 50 49 46 45 43 42 38 38 34 34 Difficulties arise from vibration of the mercury pool, when this is used as anode, and it is advisable to substitute a suitably conditioned silver wire. The silver wire isconditioned by applying a potential of 1.5 to 2 volts for 5 minutes between it and a dropping-mercury cathode when both are immersed in chloride solution.This treatment results in the deposition of a thin coating of chloride on the silver wire. In water solutions, when silver wire is used, it is also necessary to add a small amount of a salt (0.12 N ammonium chloride) to stabilise the anode potential. The experimental conditions were also re-examined at this low level of fluoride con- centration and it was found that the optimum amounts of aluminium and dye at this level were 6 pg of aluminium and 2 ml of 0.05 per cent. dye solution per 50 ml of stock solution. It was also advisable to keep the amounts of proteose peptone and buffer to a minimum.A series of calibrations under these new conditions are shown in Table IV. The conditions of the experiments were as follows- To a series of 25-ml flasks were added respectively 0, 0.1, 0.2, 0.3, 0.4 and 0.5 pg of fluoride and about 10 ml of water. Then 5 ml of aluminium - dye complex solution were added and the volume made up to 25 ml with water. The solutions were heated at 75" & 5" C for 30 minutes and examined polarographically after 18 to 20 hours. The complex was made up by taking in a 50-ml flask 6 pg of aluminium (as potassium aluminium sulphate solution) and adding 2 ml of 0.05 per cent. dye solution, 0.25 ml of 0.1 per cent. solution of proteose peptone, 10ml of buffer (as previously described) and 1 ml of 2 N ammonium chloride. Recoveries of fluoride from water solutions by this technique are included in Table IV.METHOD SPECIAL APPARATUS- A +olarogrujd.z-The Cambridge photographic instrument is suitable for larger amounts, but a cathode-ray polarograph is required for amounts of fluoride less than 0.2pg per ml of final solution. REAGENTS- The purest obtainable reagents should be used. It is advisable to use redistilled water throughout. Aluminium sohiion, 10 pg of aluminium per ml-Dissolve 1.758 g of potassium aluminium sulphate in water, add 10 ml of 60 per cent. perchloric acid, sp.gr. 1.54, and dilute to 1 litre. Take 10 ml of this solution, add 1 ml of 60 per cent. perchloric acid, sp.gr. 1.54, and dilute to 100 ml. It is recommended that this last solution should be freshly prepared at fortnightly intervals.Ammonium acetate, 2 N. Aqueous Solochrome Violet R.S. solution, 0.05 per cent.April, 19541 DETERMINATION OF FLUORIDE. PART I 195 Sodium fluoride solution, 10 pg of jluoride per ml-Dissolve 0.5525 g of sodium fluoride* in water and dilute to 250 ml. Dilute 1 ml of this solution to 100 ml for use. This solution appears to keep well, but should be freshly prepared at monthly intervals. Acetylacetone-A 0.1 per cent. v/v solution in water. Proteose peptone, 0.1 per cent. solution-Dissolve 0.1 g of proteose peptone and 0.2 g of phenol in 100ml of water. Bufer solution-To 300 ml of 2 N ammonium acetate solution add sufficient 60 per cent. perchloric acid, sp.gr. 1.54, to adjust the pH to 3.66. Add 1 ml of 1 per cent. acetylacetone and extract five times with 25-ml portions of redistilled chloroform, shaking for 2 minutes each time.Discard the used chloroform. Neat the buffer solution so prepared until all smell of chloroform has been removed. AmmoniHm chloride, 2 N-Dissolve 32.1 g of ammonium chloride in 300 ml of water. Add 1 ml of 1 per cent. acetylacetone solution and extract with chloroform as described for the buffer solution and heat until all smell of chloroform has been removed. PROCEDURE FOR RANGE ONE: 5 TO 40 pg OF FLUORIDE IN 50 ml (0.1 TO 0.8 pg PER m1)- In this range, except for samples such as halogen acids and water, the fluoride is usually separated from the sample by a preliminary distillation as hydrofluosilicic acid by the method of Willard and Winter. To five 50-ml calibrated flasks add 0, 1.0, 2.0, 3.0 and 4-0 ml of the dilute sodium fluoride solution and sufficient water to bring the volume to just less than 40rnl.In other 50-ml calibrated flasks place 40 ml of the solution or &stillate under test, which has been neutralised to methyl red with ammonium hydroxide or acetic acid. In a 100-ml calibrated flask prepare an aluminium-dye solution as follows- To 30 ml of aluminium solution add 25 ml of ammonium acetate solution, 5 ml of proteose peptone solution and 2 rnl of ammonium chloride solution. Adjust the pH t o 3.9 with 5 N perchloric acid against a pH meter, add 20 ml of Solochrome Violet R.S. solution and make up to 100 rnl with water. Add 10 ml of the aluminium - dye solution to each of the 50-ml graduated flasks and make up to the mark with water.Heat the flasks in a bath of water at 70" & 5" C for 5 minutes and then cool them, (NOTE-It is important that the heating conditions should be the same for all flasks; therefore a bath large enough to take all the flasks at once must be used.) Place about 4ml of each solution in polarographic cells, deoxygenate the solutions for 10 minutes and record polarograms at 25" C between -0.2 and -0.9 volt, with the saturated calomel electrode as anode. A large step should be observed {at about E4 = 0.4 volt) followed by a smaller one at approximately 0.1 volt more negative; the first step is due to the reduction of the uncom- plexed dye, and the other, which is the step that must be measured, is due to the aluminium - dye complex. Occasionally a further small step occurs after the one due to the aluminium - dye complex.This is caused by impurities such as iron and should be ignored. Plot a calibration graph of the depression of step height, H, against the fluoride present expressed in micrograms, where H = Hl-H,, HI-H,, etc., and H,, H,, H, . . . are the heights of the aluminium -dye step in presence of 0, 1.0, 2.0 . . . micrograms of fluoride. From this calibration graph, which should be linear, the amount of fluoride in the sample can be calculated. PROCEDURE FOR RANGE TWO: 0-5 TO 2.5 pg OF FLUORIDE I N 25 d (0.02 TO 0.1 pg PER m1)- For this and the next range (range three) of fluoride concentration, distillations are carried out with a semi-micro apparatus. The apparatus used is similar to that used by Horton,12 except that a round-bottomed distillation flask is used and heating is by a small gas flame instead of a heating mantle.It appears that the rate of distillation of fluoride is to some extent governed by the rate at which equilibrium of fluoride compounds between gas and liquid phases in the distillation flask is attained; therefore the largest possible surface area of liquid is desirable. For this reason a round-bottomed flask seems preferable to the inverted-cone type used by Horton. Use of a heating mantle is undesirable, as local hot spots can arise The purest available material must be used; that supplied by Baker and Adamson to A.C.S. specsca- tion is suitable.196 MACNULTY, REYNOLDS AND TERRY: THE POLAROGRAPHIC [Vd. 79 on the sides of the distillation flask above the liquid level and these may result in decom- position of the acid to give products that cause trouble in the subsequent treatment of the distillate.For this reason a small gas burner or an oil-bath is preferred. The use of the semi-micro apparatus permits all the fluoride to be collected in 15 to 20 ml of distillate. This involves some alterations in the total amounts of reagents required and the use of 25-ml in place of 50-ml calibrated flasks. The following modifications of the procedure described for range one are necessary- (a) 25-ml instead of 50-ml calibrated flasks are used. (b) The flasks used for calibration should contain 0.0, 0.5, 1.0, 1-5, 2.0 and 2.5 pg of fluoride. For this, dilute the standard sodium fluoride solution ten times and use 0, 0.5, 1.0, 1.5, 2.0 and 2.5 ml.(c) The aluminium - dye complex solution contains 3 ml of aluminium solution, 2.5 ml of proteose peptone solution, 10 ml of Solochrome Violet R.S. solution, 10 ml of buffer solution (pH 3-66) and 1 ml of 2 N ammonium chloride solution. This is made up to 5Uml with water and 5ml is taken for each flask. (d) The time of heating is made 30 minutes and the solutions must be set aside overnight fur 16 to 20 hours before polarograms are recorded. (e) The cathode-ray polarograph must be used, and a silver chloride anode of the straight-wire type is more satisfactory than the saturated calomel electrode. PROCEDURE FOR RANGE THREE: 0.1 TO 0.5 pg OF FLUORIDE IN 25 ml (0.004 TO 0.02 pg PER Id)- This range is similar to range two, but the following modifications are necessary- (a) The flasks used for calibration should contain 0.0, 0.1, 0.2, 0.3, 0.4 and 0.5p.g of fluoride.(6) The aluminium - dye complex solution contains 0.6 ml of aluminium solution, 0.25 ml of proteose peptone solution, 2 ml of Solochrome Violet R,S. solution, 10 mL of buffer solution and 1 ml of 2 N ammonium chloride solution. This solution is made up to 50 ml and 5 ml is taken for each flask. POINTS OF SPECIAL IMPORTANCE IN RANGE THREE- Impurities-One of the chief difficulties encountered at this low level of fluoride con- centration is the effect of traces of interfering impurities in reagents. The two reagents used in relatively large amounts, the acetate - perchloric acid buffer and the ammonium chloride, are normally purified sufficiently by extracting the metal acetylacetonates by shaking with chloroform .If impurities such as iron or aluminium are found to be present in solutions or distillates under examination, they should be removed as follows. To the solution at pH 2 to 4 add 1 ml of 0.1 per cent, v/v acetylacetone in water, neutralise with ammonium hydroxide and extract by shaking with 10 ml of redistilled chloroform. Discard the chloroform layer. Repeat the extraction three more times. Finally, warm the solution to remove traces of chloroform. Cleaning-The flasks and polarographic cells should be cleaned by rinsing them first with chromic acid, then with tap water, and finally with several changes of distilled water. A better calibration seems to result if, after this initial treatment, the flasks and cells are reserved for use at this range of fluoride concentration only and are cleaned between determinations with hydrochloric acid followed by several rinses with distilled water.Pure acetone can be used for drying purposes or to remove any greasiness. Ripple-Ripple on the cathode-ray trace diminishes the accuracy with which step heights can be measured and should therefore be kept to a minimum. This can be achieved by using only a very small amount of mercury in the cell, together with a silver chloride anode, and at the same time earthing the water-bath to the cathode-ray polarograph. The best point of attachment of the earth Wire varies from day to day and must be determined by trial and error. Oxygm-The least trace of oxygen upsets the method at this concentration and it is usually necessary to deoxygenate the solutions for 15 to 20 minutes before recording the polarograms.April, 19541 DETERMINATION OF FLUORIDE.PART 1 197 RESULTS AND DISCUSSION The method was tested on solutions of fluoride both directly and after separation of the fluoride by distillation as hydrofluosilicic acid. Results are shown in Table IV. TABLE IV RECOVERY OF FLUORIDE After separation of fluoride by Direct determination distillation as H,SiF, r 1 Fluoride Fluoride A A added, Fluoride recovered, added , Fluoride recovered, P€! Pg 5 5.0, 5-5, 5.0 10 10.0, 10.5 15 15.0, 15.5, 15.5 20 19.0, 20.0 0.5 1.0 1.5 2.0 2.5 0.05 0.10 0.15 0.20 0.40 These results are dudicate. as smrious 0.5, 0.5 1.0, 1.0 1.4, 1.5 1.9, 2.0 2.3, 2.3 0.06 0.10 0.13 0.19, 0.19, 0.22 0.39, 0.39, 0.40, 0.38 Pg Crg 2.2 2-3, 2.3, 2-35 6.8 6.7, 6.9, 6.9 9.6 10.5, 10*5, 10.0, 10.2 0.7 1.1 2.2 2-25 0.1 0.2 0.3 0.4 0.7, 0.8 1.0, 1.0, 1.16 2.1, 2-1, 2.3 2-15, 2-25 0.12, 0.12, 0.10, 0-10 0.19, 0.20, 0.16 0.28, 0-25 0.36 satisfactory, but determinations should always be carried out in results occur f airlv f rea uen t lv.The methid has the advantage of perkittiig smaker amounts of fluoride to be determined than hitherto; on the lowest range a concentration of 0.004 pg per ml in the final solution gives a step-height depression of the aluminium - dye step of about 7 units and, as the ripple on the cathode-ray trace is of the order of 1 unit under good working conditions, the limit of detection of the present method is of the order of 0.001 pg per ml in the final solution.Nevertheless, the full potentialities of the method have not as yet been realised. The deter- mination depends at present on the measurement of a small depression of a relatively large combined dye and aluminium - dye complex step. In amplifying the step to obtain maximum sensitivity, the limit is set by the size to which the step can be magnified, whereas it is only the step depression that is of importance in the determination of fluoride. The full sensitivity can only be reached when a differential (the term “subtractive” is to be preferred21) method of polarography is used. Such a method is now being investigated in the authors’ laboratories. The necessity of allowing the solutions of fluoride and aluminium to stand for a long time (18 to 20 hours) on ranges one and two is another instance of the slowness with which dilute solutions of aluminium and fluoride reach equilibrium, particularly when the amount of aluminium is greater than the amount of fluoride.This phenomenon has been noticed previously by ourselves15 and by Brosset,22 and examination of many papers on fluoride determination indicates that this phenomenon is general. Some applications of this method to the determination of fluoride in specific substances will be given in Part I1 of this series of papers. Acknowledgment is made to the Chief Scientist, Ministry of Supply, for permission to publish this paper and to Mr. A. S. Nickelson for his interest and encouragement. REFERENCES 1. 2. 3. 4. 5. Willard, H. H., and Winter, 0. B., Ind. Eng. Chem., Anal. Ed., 1933, 5, 7. Milton, R. F., Liddell, H. F., and Chivers, J. E., Analyst, 1947, 72, 43. Williams, H. A., Ibid., 1946, 71, 175. Rickson, J. B., Ibid., 1950, 75, 84. Wadhwani, F., J . Indian Inst. S G ~ . , 1952, 34, 123.198 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. BREMNER ; ZDENTIFICATION OF KYDROXYLAMINE De Boer, J. H., 2. anorg. Chem., 1926, 126, 213. Sanchis, J. M., Ind. Eng. Chem., Anal. Ed., 1934, 6, 134. Lamar, W. L., and Seegmiller, C . G., Ibid., 1941, 13, 901. Kortum-Seiler, M., Angew. Chem., 1947, A59, 159. Monnier, D., Rusconi, Y,, and Wenger, P., Helv. Chim. Ada, 1946, 29, 521. Lacroix, S., and Labalade, M., Anal. Cham. Acfa, 1950, 4, 68. Horton, A. D., Thompson, P. F., and Miller, F. J., Anal. Chem., 1952, 24, 548. Feigl, F., and Heisig, G. B., Alzal. Chim. Acta, 1949, 3, 361. Willard, H. W., and Horton, C. A., Anal. Chern., 1952, 24, 862. Hunter, G, J., MacNulty, B. J., and Terry, E. A., Anal. Chim. Acta, 1953, 8, 351. Thrun, W. E., Anal. Chem., 1948, 20, 1117, -, Ibid., 1950, 22, 918. Willard, H. H., and Dean, J. A., Ibid., 1850, 22, 1264. Davis, H. M., and Reynolds, G. F., AnaZyst, 1953, 78, 314. Davis, H. M., and Seaborn, Miss J. E., Electronic Eng., 1953, 25, 314. Reynolds, G. F., Chem. & Ind., 1953, 971. Brosset, C., and Orring, J., Suensk Kern. Tidskr., 1943, 55, 101. [VOL 79 CHEMICAL INSPECTORATE STATION APPROACH BUILDINGS KIDBROOKE, LONDON, S.E.3 October 23rd, 1953

ISSN:0003-2654    

DOI:10.1039/AN9547900190

出版商:RSC    

年代:1954

数据来源: RSC

摘要:

198 BREMNER : IDENTIFICATION OF HYDROXYLAMINE [Vol. 79 Identification of Hydroxylamine and Hydrazine by Paper Chromatography BY J. M. BREMNER A method for the separation and identification of microgram quantities of hydroxylamine and hydrazine is described. It involves paper chromato- graphy with acidic solvents and identification by RF values and by the colours produced with picryl chloride and other reagents. ALTHOUGH hydroxylamine has been reported or postulated to be an intermediate in several important biological processes such as the fixation of atmospheric nitrogen by Azotobacter and leguminous bacteria and the oxidation of ammonia to nitrite by Nitrosornonas, little attention has been given to the development of methods for its identification. The method of detection generally used is that of Blom,l which is based on the oxidation of hydroxylamine to nitrite by iodine in acetic acid and detection of the nitrite by means of the colour reaction with sulphanilic acid and 1-naphthylamine.This highly sensitive test has been investigated and improved by Csaky,2 who showed that interference due to nitrous, hyponitrous and nitrohydroxamic acids, which give the colour reaction without pretreatment with iodine, could be obviated by an acid hydrolysis procedure. He found, however, that although only hydroxylamine gave the test after hydrolysis, it was partly or totally destroyed if any of these interfering substances were present in the test solution during hydrolysis. Apart from this, Blom’s test is open to criticism on the grounds that it is an indirect procedure based on the detection of an oxidation product of hydroxylamine and that when used to detect hydroxylamine in complex biological fluids it may be subject to interference from compounds not previously tested. The need for a method of confirming results obtained by Blom’s procedure became apparent during the course of an investigation into the reaction between lignin and nitrous acid and in studies of the oxidation of ammonia t o nitrite by soil micro- organisms.The object of the work reported in this paper was to develop a method by which microgram quantities of hydroxylamine could be identified after separation from interfering substances. Because a method based on paper chromatography appeared to hold the most promise, the behaviour and detection of hydroxylamine on paper chromatograms run in various solvents have been investigated.Paper chromatography of hydrazine, which gives some of the colour and reducing reactions of hydroxylamine, has also been studied.April, 19541 AND HYDRAZINE BY PAPER CHROMATOGRAPHY EXPERIMENTAL APPARATUS AND PROCEDURE- 199 Hydroxylamine was used in the form of a 0.1 M aqueous solution of its hydrochloride and hydrazine as a 0.1 M aqueous solution of its sulphate. Two to four microlitres of the solutions containing 6 to 13 pg of hydroxylamine or hydrazine were used for paper chromato- graphy, which was carried out by the ascending technique. Whatman No. 4 filter-paper was used throughout. Preliminary runs were made in Pyrex boiling tubes with 2-5-cm x 25-cm strips of paper, the technique being essentially that described by Rockland and D ~ n n . ~ Further runs with promising solvents were made in glass filtrate jars with cylinders of paper 33 cm high made from sheets by stapling them so that their edges did not overlap (Bray, Thorpe and White4).The chromatograms were developed until the solvent front had travelled a distance of 20 to 30 cm. Depending on the solvent used, the time required was 3 to 24 hours. After development, the chromatograms were dried at room temperature and sprayed with the detecting reagents described below. RF values reported were determined at 18" & 2" C. DETECTING REAGENTS- Picryl chloride-After being sprayed with a 1-5 per cent. w/v solution of picryl chloride in ethanol, the chromatograms were allowed to dry and then exposed to ammonia vapour.Hydroxylamine was revealed by a bright orange colour and hydrazine by an intense blue colour. Ammortiacal diacetylmonoxime - nickel salt reagent-This was prepared according to Feigl.5 Hydroxylamine on chromatograms sprayed with this reagent was revealed by a red colour; hydrazine gave no colour. ModiJed Csaky2 reagents-For reasons given below the reagents used by Csaky2 for the detection of hydroxylamine in solution were modified as follows for use as sprays. Reagent (a), which was a mixture of 20 ml of a 1.3 per cent. w/v solution of iodine in glacial acetic acid and 20 ml of a 1.0 per cent. w/v solution of sulphanilic acid in 30 per cent. v/v aqueous acetic acid, was prepared immediately before spraying.Reagent (b) was a 0-1 per cent. w/v solution of N-( 1-naphthy1)-ethylenediamine dihydrochloride in water. The two reagents were used successively. Hydroxylamine on chromatograms sprayed with these reagents was revealed by a red colour; hydrazine gave no colour. Both hydroxylamine and hydrazine could be detected as weak yellow spots on spraying with reagent (a), but the yellow spot of hydrazine disappeared and that of hydroxylamine became red on subsequent spraying with reagent (b). Ninhydrin-After being sprayed with a 0.1 per cent. w/v solution of ninhydrin in chloroform, the chromatograms were placed in an oven at 105" C for 5 minutes. Hydrazine was revealed by a yellow colour: hydroxylamine gave no colour. SOLVENTS- In two-phase systems, the non-aqueous phase was used as solvent and the chromatograms were prepared in an atmosphere provided by the aqueous phase.The n-butanol - acetic acid - water mixture (Table I) was of the same composition as that used by Partridge6 for the separation of sugars. Before use this mixture was shaken thoroughly and allowed to stand for 3 days (Bate-Smith and Westall'). Formic acid was used in the form of the 90 per cent. w/w AnalaR grade solution supplied by Hopkin and Williams Ltd. RESULTS AND DISCUSSION The RF values of hydroxylamine and hydrazine in various solvents are recorded in Table I. They were obtained with one batch of filter-paper and represent the means of several deter- minations. The values were found to vary from batch to batch of filter-paper but to be reasonably constant within individual batches.Neither hydroxylamine nor hydrazine could be detected on chromatograms run in basic solvents and their detection on chromatograms run in neutral solvents, such as water- saturated n-butanol, was achieved only when fairly large amounts (20 to 30 pg) of materials were used. Of the acidic solvents tested, only mixtures containing hydrochloric acid were The solvents used were of the purest quality available. The other solvent mixtures were prepared immediately before use.200 BREMNER: IDENTIFICATION OF HYDROXYLAMINE [Vol. 79 found to be satisfactory for paper chromatography of both hydrazine and hydroxylamine, the alcohol - hydrochloric acid series of solvents listed in Table I giving small well-defined spots with either compound.Solvents containing acetic or formic acid proved much less useful. Some, e.g., n-butanol- formic acid - water, or tert.-amyl alcohol - acetic acid - water, gave extensive tailing of both compounds. In others, e.g., benzene - acetic acid - water, or chloroform - formic acid - water, no movement occurred. However, a few gave only slight tailing of hydroxylamine; the RF values are shown in Table I. TABLE I RF VALUES OF HYDROXYLAMINE AND HYDRAZINE ON WHATMAN NO. 4 FILTER-PAPER AT ROOM TEMPERATURE Solvent* Hydroxylamine Hydrazine Methanol (70), 6 N hydrochloric acid (30) .. .. .. Ethanol (70), 6 N hydrochloric acid (30) . . .. .. .. n-Propanol (70), 6 N hydrochloric acid (30) . . .. .. n-Butanol (40), acetic acid (50), water (10) . . .. .. n-Butanol (50), 2 N hydrochloric acid (50) .. . . .. n-Rutanol (50), 6 N hydrochloric acid (50) . . .. .. n-Butanol (70), 6 N hydrochloric acid (30) . . .. .. tert.-Butanol (70), 6 N hydrochloric acid (30) . . .. .. Diethyl ether (50), 90 per cent. w/w formic acid (50) Methanol (80), 90 per cent. w/w formic acid (15), water (5) . . Ethanol (70), 90 per cent. w/w formic acid (ZO), water (10) . . n-Butanol . . .. .. .. .. .. .. .. . . .. 0.53 0.67 0.4 1 0-54 0.36 0.12 0.23 0.18 0.5 1 0.3 1 0.40 0.36 0.34 Streak 0.22 Streak 0.21 0.00 0.07 0.10 0.42 0.19 0.20 Streak * The figures indicate the percentage v/v of each component. Butanol and butanol- acetic acid mixture are unsatisfactory solvents, as it is difficult to detect less than 20pg of hydrazine or hydroxylamine on their chromatograms.The RF values in these solvents are presented in Table I only because they show that the weak yellow spots of identical RF value noted by Bremner and Kentens on chromatograms of hydrazine and hydroxylamine run in butanol (RF 0.02) and butanol- acetic acid (KF 0.12) and sprayed with ninhydrin must have been due to some impurity. This is confirmed by the finding that only hydrazine gives a yellow colour reaction with ninhydrin and that its detection by this reagent on chromatograms run in butanol or butanol- acetic acid is possible only if a large amount of hydrazine is present. It was found that RF values in the alcohol - hydrochloric acid series of solvents could be varied considerably by altering the proportion of alcohol to acid. With mixtures of n-butanol and 6 N hydrochloric acid, for example, the KF value of hydrazine can be increased from 0-15 to 0.42, and that of hydroxylamine from 0.24 to 0.50, by increasing the proportion of 6 N hydrochloric acid from 25 to 50 per cent.With mixtures of n-butanol and 12 N hydrochloric acid, the RF value of hydroxylamine can be increased from 0.08 to 0.76 by increasing the proportion of 12N hydrochloric acid from 10 to 67 per cent., whereas the RF value of hydrazine is unaffected by this change, being less than 0.1 in either mixture. Although complete separation of hydroxylamine and hydrazine can be achieved by the use of any of the alcohol - 6 N hydrochloric acid solvent mixtures described in Table I, the time required varies considerably with the alcohol used. With methanol or ethanol 2 to 4 hours is sufficient, but with tert.-butanol 20 to 24 hours is required.No attempt was made to determine the lower limit of sensitivity for the detection of hydroxylamine or hydrazine by the methods described, but 0-5 ,ug of hydrazine and 3 ,ug of hydroxylamine can be readily detected by picryl chloride on chromatograms run in the solvents containing hydrochloric acid, and 2 pg of hydroxylamine can be detected on these chromatograms by the amrnoniacal diacetylmonoxime - nickel salt reagent or the modified Csaky reagents. According to Kul’berg and Cherke~ov,~ as little as 0.005 pg of hydrazine and 0.2 pg of hydroxylamine can be detected by spot tests on filter-paper by the picryl chloride - ammonia technique. Feig15 has reported that the ammoniacal diacetylmonoxime - nickel salt reagent will detect 1 pg of hydroxylamine in spot tests on filter-paper.This method of detecting hydroxylamine would appear to be highly specific, since it requires condensation between hydroxylamine and diacetylmonoxime with formation of diacetyl- dioxime (dimethylglyoxime) and reaction of this last with ammoniacal nickel salt solution to form red, slightly soluble nickel dimethylglyoxime. The reagents used in Csaky’s methodApril, 19541 AND HYDRAZINE BY PAPER CHROMATOGRAPHY 201 of detecting hydroxylamine,2 which involves oxidation of hydroxylamine to nitrite by iodine in acetic acid solution in the presence of sulphanilic acid followed by destruction of the excess of iodine with sodium arsenite and coupling of the diazotised sulphanilic acid with 1-naphthylamine, were found to be effective in revealing hydroxylamine when sprayed successively on developed chromatograms.It was found, however, that it was unnecessary to remove the excess of iodine with sodium arsenite when sprays of these reagents were used and that a more intense red colour was produced when N-( l-naphthy1)-ethylenediamine was used as coupling agent in place of l-naphthylamine (Shindo). The modified technique described above was therefore adopted. As a reagent for the detection of hydrazine, ninhydrin is considerably less sensitive than picryl chloride. Of the four detecting reagents used, only picryl chloride reveals both hydrazine and hydroxylamine. It has also been used for the detection of tertiary pyridine bases on paper chromatograms.ll Although it has not been possible to supply direct proof of non-interference from all substances that may be present in biological material, the paper chromatographic method of identifying hydroxylamine described above appears to be highly specific, as it is based not only on RF values in various solvents, but also on the colours produced with three different c hromogenic sprays. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. REFERENCES Blom, J., Zbl. Baht., Abt. 2, 1931, 84, 60. Csaky, T. Z., Acta Chem. Scand., 1948, 2, 450. Rockland, L. B., and Dunn, BI. S., Science, 1949, 109, 539. Bray, H. G., Thorpe, W. V., and White, K., Biochem. J.. 1950, 46, 271. Feigl, F. , “Qualitative Analysis by Spot Tests,” Third Edition, Elsevier Publishing Co., New Partridge, S. M., Biochem. J., 1948, 42, 238. Bate-Smith, E. C., and Westall, R. G., Biochim. Biophys. Ada, 1950, 4, 427. Bremner, J. M., and Kenten, R. H., Biochem. J., 1951, 49, 651. Kul’berg, L. M., and Cherkesov, A. I., Zhur. Anal. Khim., 1951, 6, 364. Shim, M. B., Ind. Eng. Chem., Anal. Ed., 1941, 13, 33. Cuthbertson, W. F. J., and Ireland, D. M., Biochem. J., 1952, 52, xxxiv. York and Amsterdam, 1947. ROTHAMSTED EXPERIMENTAL STATION HARPENDEN, HERTS. July 7th, 1953

ISSN:0003-2654    

DOI:10.1039/AN9547900198

出版商:RSC    

年代:1954

数据来源: RSC

摘要:

April, 19541 AND KYDRAZXNE BY PAPER CHROMATOGRAPHY 201 The Determination of Glucosamine BY R. BELCHER, A. J. NUTTEN AND MISS C. M. SAMBROOK The colorimetric method for the determination of glucosamine based on the reactions with acetylacetone and p-dimethylaminobenzaldehyde has been systematically examined. Several improvements have been made in the method, which has been applied to the determination of glucosamine in N-acetyl-a-methylglucosaminide and heparin. The method has been adapted €or use with commercially available instruments. THE colorimetric determination of glucosamine (2-amino-2-deoxyglucose) based on the reactions with acetylacetone and $-dimethylaminobenzaldehyde has given erratic results in the hands of several workers. A systematic examination of the method has been made to trace the sources of error, and several improvements have been introduced.The improved procedure has been applied to the determination of glucosamine in N-acetyl-&-methyl- glucosaminide and heparin and has been adapted for use with commercially available instruments. The determination of glucosamine in heparin is of importance in structural studies; it is one of the few materials available containing glucosamine to which the method described in this paper can be applied. As these materials have to be broken down by acid hydrolysis, a study of the conditions of hydrolysis has been made and the results may be a guide in biochemical work of a similar kind. The most widely used procedure for the determination of glucosamine is that of Elson and M0rgan.l After preliminary treatment of the glucosamine test solution with an alkaline202 BELCHER, NUTTEN AND SAMBROOK: [Vol.79 solution of acetylacetone, addition of a reagent containing 9-dimethylaminobenzaldehyde produces a red colour, which is compared with standards. That the method is not completely satisfactory is evident from the numerous variations of it described in the literature. Nilsson,2 for example, carried out the reaction with acetylacetone at 95" C instead of at 100" C, while Smrensen,3 who followed Nilsson's procedure closely, claimed that the optimum pH for the reaction with acetylacetone was 9.5. Sarrensen also reported that the intensity of the colour remained constant for a "considerable length of time," although Boyer and Furth,4 who used acetic acid as a solvent and heated the solution after the addition of P-dimethylaminobenz- aldehyde, stated that the colour faded within 20 minutes.Blix5 claimed that satisfactory results were attained when the concentrations of both acetylacetone and alkali were increased, and when the reaction was carried out at 96" C in a closed vessel. The capriciousness of the method has been further emphasised by Horowitz, Ikawa and Fling,6 who suggested that the intensity of the colour developed depended not only on the conditions of heating with acetylacetone, but on the size of the vessel in which the reaction was carried out. After addition of the 9-dimethylaminobenzaldehyde, Palmer, Smyth and Meyer7 incubated the solution at 37" C for 30 minutes before measuring the colour. Schloss also favoured incubation, but at 30" C for 24 hours.There is some difference of opinion about the interferences in the method. Elson and Morgan found that certain sugars and amino-acids caused no appreciable interference, but Boyer and Furth reported considerable interference from these substances. Sideris, Young and Kraussg observed that glycine and lysine in the presence of reducing sugars produced a measurable amount of colour, while Bendich and Chargafflo found that glycine in the presence of glucose produced a slight, but not negligible, colour. Blix stated that interference from sugars and amino-acids was appreciable, but claimed to have eliminated the interference by modifying the acetylacetone reagent. Immers and Vasseurll also obtained measurable colours from sugars and amino-acids in admixture when they were treated according to the procedure of Elson and Morgan.In view of the conflicting information about the determination of this biologically important compound, we have carried out a complete systematic examination of the procedure of Elson and Morgan in an attempt to place the method on a sound practical basis. The results of this examination, including the development of a satisfactory method for the determination of glucosamine, are described below. In order that an accurate and quantitative analysis of glucosamine in mucopoly- saccharides can be made, it is essential that the breakdown of the protein - carbohydrate complex with acid be complete. Certain of these mucopolysaccharides, in particular, heparin, are very stable towards acids, and drastic conditions are necessary for the hydrolysis. Little exact information is available about the conditions required for the complete hydrolysis of heparin, and no information is available as to the behaviour of glucosamine itself during a hydrolysis, Accordingly, a lesser, but nevertheless important, consideration during the present investigation was the establishment of conditions that would lead to the complete hydrolysis of heparin without affecting the glucosamine molecule.The compound N-acetyl-a-methylglucosaminide contains linkages similar to those present in certain materials containing glucosamine, e.g., hyaluronic acid, and the conditions necessary for the hydrolysis of this compound were therefore established.EXPERIMENTAL REACTION OF GLUCOSAMINE WITH ACETYLACETONE- (i) Efect of temperature-When the reaction mixture was heated to 100" C, the colours produced were more intense than at lower temperatures, although the temperature coefficient of the reaction was small between 95" and 100" C. A thermostatically controlled heating bath was not necessary to maintain the optimum temperature. The development of the colour was negligible when the reaction mixture was heated to temperatures less than 60" C. (ii) Efect of heating time-A heating time of 10 minutes gave almost linear graphs when optical densities were plotted against concentrations of glucosamine. The development of the colour was incomplete when heating times of less than 10 minutes were used, and non- linear curves were given when heating was continued for more than 10 minutes.203 (iii) Efect of concentration of acetylacetone-The intensities of the colours obtained decreased when the amount of acetylacetone added was increased or decreased, but a small amount of latitude was obviously permissible in the preparation of the reagent (see Fig.1). April, 19541 THE DETERMINATION OF GLUCOSAMINE Concentration of acetylacetone, ml per 50 ml of 0 5 N Na,CO, Effect of concentration of acetylacetone Fig. 1. (iv) Eflect of pH-As the concentration of sodium carbonate in the reagent was increased, the intensities of the colours produced also increased slightly. A t concentrations of sodium carbonate greater than N , however, some difficulty was encountered in redissolving the precipitate initially formed on adding ethanol to the reaction mixture.The concentration coefficient was small, but by no means negligible. The optimum pH for the reaction was 9.8 (that is, with the reagent N with respect to sodium carbonate), but the development of colour was slight in solutions adjusted to this pH with sodium hydroxide. REACTION WITH p-DIMETHYLAMIV I OBENZALDEHYDE- (i) Efect oftemperatuure-When the temperature of the solution was raised after addition of the +-dimethylaminobenzaldehyde reagent to the mixture from the reaction between glucosamine and acetylacetone, a considerable increase in the intensity of the colour occurred. The maximum intensity was attained by heating to 75" C. Above this temperature, serious losses of ethanol occurred by evaporation.When the ethanol was replaced by glacial acetic acid and the solution was heated to above 75" C, development of the colour was markedly decreased, probably because of partial destruction of the chromogen. (ii) Efect of heating time-The heating time was varied between 5 minutes and 3 hours. Approximately linear graphs were obtained after heating for 10 minutes, but the optical densities were not entirely reproducible. Results were somewhat more reproducible after heating for 20 minutes, and after heating for 30 minutes readings were reproducible and somewhat higher, indicating an increase in sensitivity. This increase in sensitivity, although small, continued up to a heating time of 60 minutes, after which measurements became uncertain and not reproducible.Accordingly, a heating time of 30 minutes was taken as the most suitable. (iii) Efect of concentratio% of p-dimethylaminobenxaldehyde-Reagents were prepared in which the concentration of p-dimethylaminobenzaldehyde was (a) half and (b) twice that recommended by Elson and Morgan. When the half-strength reagent was used, the colours204 BELCHER, NUTTEN AND SAMBROOK: [Vol. 79 did not develop to the full intensity, and the graphs plotted were neither linear nor repro- ducible. When the double-strength reagent was used, the development of the colour was considerably enhanced (the reagent itself did not show selective light absorption), but the graphs plotted were again neither linear nor reproducible. Accordingly, the reagent recom- mended by Elson and Morgan was retained.(iu) Efect ojpH-The work of other authors appeared to indicate that close control of pH was not essential (the pH of the final solution in the procedure of Elson and Morgan is approximately 0.3). The preparation of the p-dimethylarninobenzaldehyde reagent is by no means precise, especially with respect to the concentration of hydrochloric acid, for commercial hydrochloric acid may be anywhere between 10 and 12 N . Experiments showed that optimum development of colour occurred when the pH of the final solution was between 0.62 and 0.35 (that is, when the reagent was between 8 and 10.8 N with respect to hydrochloric acid). In more weakly acidic solutions, the intensities of the colours decreased with decrease in pH until, at pH 2.9, no development of colour occurred.Considerable latitude was permissible, therefore, in the preparation of the reagent, and it was found convenient to replace the ethanol in the reagent by concentrated hydrochloric acid. (v) E$ect of ethaaoZ-The amount of ethanol added to the mixture from the reaction between glucosamine and acetylacetone is normally 6ml. In the present work, 5ml of ethanol were added, as it had been found convenient to adjust the volume of the mixture of glucosamine and acetylacetone before reaction to 4 ml (Elson and Morgan specify 3 ml) with distilled water. This ensured that the sides of the reaction tube were washed down thoroughly. When the ethanol was gradually replaced by water, the intensity of the colour developed decreased with increasing amounts of water.The results showed that the amount of ethanol added was not critical, as the graph of optical density against concentration of ethanol reached a maximum after the addition of 44ml of ethanol and remained at this maximum when further amounts of ethanol were added. INTERFERENCES- Under the conditions of the method developed as a result of the studies described above, the following substances, both alone and in combinations of two and three, caused no inter- ference, even when added in twentyfold excess over the amount of glucosamine present: arabinose, galactose, glucose, mannose, glucuronic acid, alanine, arginine, glycine, lysine, phenylalanine, proline, serine, tryptophan and tyrosine. The majority of the solutions of glucosamine encountered in practice are produced by acid hydrolysis of naturally occurring materials.However, when solutions of combinations of the above substances with glucosamine were heated under reflux with 4 N hydrochloric acid for 24 hours, no interference was found. Even a twentyfold excess of these substances could be tolerated. It was concluded, therefore, that the greatly increased sensitivity of the method, and hence the appreciable dilution of the foreign substances likely to be present, was mainly responsible for eliminating the slight interferences reported by other workers. ABSORPTION CWRVE- A 0.05-mg sample of glucosamine was treated according to the method described later) and the optical density of the solution was measured over the wavelength range 400 to 700 mp with a photo-electric quartz spectrophotometer No.SP 600 (Unicam Instruments Ltd., Cambridge). The absorption curve is shown in Fig. 2. The maximum optical density occurs at 512mp, in agreement with the findings of Schloss. Only one maximum occurs in the curve, and this would appear to indicate that other chromogens formed in the reaction between acetylacetone and glucosamine have either been destroyed during the “after heat” or have been converted into the chromogen responsible for the colour. SENSITIVITY AND ACCURACY- A measurable absorption was given by 0-5 pg of glucosamine by the method described later. Colour densities were measured with three instruments, (a) an E.E.L, test-tube colori- meter (Evans Electroseleniurn Ltd.) with No. L 204 tubes and a No.623 green filter, (b) a Spekker photo-electric absorptiometer, model H 760 (Hilger and Watts Ltd.) with a 2-cmApril, 19541 THE DETERMIKATIOX OF GLUCOSAMINE 205 cell and Ilford No. 604 filter, and (c) a photo-electric quartz spectrophotometer, No. SP 500 (Unicam Instruments Ltd.) with a 2-mm cell and a wavelength of 512 mp. A comparison of the results and the errors incurred with these three instruments is included in TabIe I. As the colours produced were stable for at least 12 hours, the coiour .- $cg 0-5--- 6 7 0.4- O" 0.3- .- c, 0.2 - 0.1 - 400 500 600 700 Wavelength, my Fig. 2, Absorption curve for 0.05 mg of glucosamine treated by the procedure described measurements were made successively on the three instruments, and the amounts of glucosamine were read from the standard graphs previously prepared for each instrument.TABLE I COMPARISOT; OF RESULTS WITH DIFFERENT INSTRUMENTS Glucosamine f- present, E.E.L., Pg PLg 10 10.1 20 19.6 30 30.8 40 40-8 50 49.2 60 58-6 70 71.3 80 81.7 90 88.1 100 101.2 Glucosamine found Spekker, Pg 10.0 20.2 30.3 39.7 50.3 5 9.4 69-5 80.7 90-7 99.1 A 7 Unicam, P.lg 10.0 19.9 30.2 40.2 49.8 60.3 69.8 80.4 90.3 99-5 - E.E.L., te +0.1 - 0.4 + 0.8 + 0.8 -0.8 - 1.4 + 1.3 + 1.7 - 1.9 + 1.2 Error Spekker, Pg + 0.0 + 0.2 + 0.3 - 0.3 + 0.3 - 0.6 - 0.5 -- + 0.7 + 0-7 -0.9 - Unicam, Pg + 0.0 -0.1 + 0.2 + 0.2 - 0.2 + 0.3 - 0.2 + 0.4 3-0-3 - 0.5 BLANKS- were negligible, a measurable absorption not being found. Although the sensitivity of the method had been considerably increased, the blanks206 BELCHER, NUTTEN AND SAMBROOK: [Vol.79 APPLICATIONS OF THE METHOD Preliminary experiments were designed to determine the effect of heating glucosamine with hydrochloric acid of various concentrations. No visible decomposition of the molecule occurred when glucosamine solutions were heated for 15 hours with N and 2 N hydrochloric acid. When the solution was neutralised with sodium hydroxide and the amount of gluco- samine was determined, the recoveries were quantitative. When 3 N and 4 N solutions of hydrochloric acid were used, the solutions darkened somewhat; when the solution in 4 N acid was cooled, a brown deposit was formed. Quanti- tative recoveries of glucosamine were, however, obtained at both these concentrations, which indicated that the part of the glucosamine molecule responsible for the colour reaction with p-dimethylaminobenzaldehyde was not affected by the hydrochloric acid, HYDROLYSIS OF N-ACETYL-CX-METHYLGLUCOSAMINIDE- When solutions of N-acetyl-a-methylglucosaminide in 2 N hydrochloric acid were heated under reflux, hydrolysis was complete in 2 hours.The rate of hydrolysis was initially rapid, a 50 per cent. recovery of glucosamine being obtained after 30 minutes. When the hydrolysis was completed, the solution was neutralised with sodium hydroxide and an aliquot was taken for the analysis. HYDROLYSIS OF HEPARIN- When solutions of heparin (sodium salt) in 4 A; hydrochloric acid were heated under reflux for 12 hours, hydrolysis was not complete, although consistent recoveries of glucosamine were obtained.Hydrolysis was assumed to be complete after 15 hours, as no increase in the amounts of glucosamine recovered occurred after further heating. I t is of interest to compare the analytical results for the present sample of heparin with those reported by Wolfrom et aZ.12 on a different sample (Table 11). TABLE I1 COMPARISON OF RESULTS OF PRESENT WORK WITH RESULTS BY WOLFROM et aZ.la Sodium heparinate used by Wolfrom et aZ.12 Sodium heparinate used in present work Mean 1.86 Mean 10.88 1 Nitrogen, per cent. . . 2.6 Dumas: 1-82, 1.86, 1.90, 1.90 Kjeldahl: 1.90, 1.86, 1-85, 1.88 Sulphur, per cent. . . 11-66 Parr bomb: 10.87, 10.78, 10.90 Combustion : 10.87, 10.92, 10.94 Glucosamine, per cent. . . 25 ( i: 1.7) 18.8, 18.9, 18.7, 18.8 (12 hours) 21.1, 21.1, 21-2 (1 5 hours) 21.2, 21.3, 21.2 (20 hours) The ratios of sulphur to nitrogen for the two samples are in close agreement, and the amounts of glucosamine found agree with the values for nitrogen.It is also of interest to note that the accuracy of the method used by Wolfrom and his co-workers for the determina- tion of glucosamine (the method of Elson and Morgan) is considerably less than the accuracy claimed for the method described in this paper. METHOD MATERIALS- at 139" to 140" C and store it in a dark, well-stoppered bottle. three times from ethanol to give pale yellow plates, m.p. 73" C. well-stoppered bottles. Acetylacetone-Twice distil commercial acetylacetone and collect the fraction boiling p-DimethyZaminobenzaZdehyde-Recrystallise AnalaR p-dimethylaminobenzaldehyde Sodium carbonate-Dry AnalaR sodium carbonate at 110" C for 2 hours and store it inApril, 19541 THE DETERMINATION OF GLUCOSAMINE 207 REAGENTS- AcetyZacetone solution-Dissolve 1 ml of acetylacetone with shaking in 50 ml of N sodium carbonate.The reagent must be freshly prepared every 24 hours, and when not in use must be kept in a refrigerator at 0" C. p-Dimethy Zaminobenzaldehyde solution-Dissolve 0.8 g of 9-dimethylaminobenzaldehyde in 60 ml of concentrated hydrochloric acid (AsT grade). PROCEDURE- Measure 1 ml of a neutral glucosamine solution (containing between 10 and 100 pg of ghcosamine hydrochloride) into a 20-ml graduated tube* constricted between the 9 and 10-ml graduation marks, as shown in Fig. 3. Add 1 ml of acetylacetone solution and adjust the volume to 4ml with a fine jet of distilled water, washing down the sides of the tube.Immerse the tube in a bath of water at 100" C, the level of the water in the bath being just higher than the level of the liquid in the tube. The neck of the tube should project a few -14 - 12 -1 I - Fig. 3. Special graduated tube inches above the top of the water-bath. After heating the tube for 10 minutes, remove it from the bath and allow it to cool to room temperature. Adjust the volume of the solution to 91111 with ethanol and place the tube in a thermostatically controlled bath of water at 75" C for 5 minutes. Then add 1 ml of P-dimethylaminobenzaldehyde solution, remove the tube and shake it thoroughly. Replace the tube in the bath so that the water comes just above the level of the liquid in the tube, and heat it at 75" C for 30 minutes with occasional shaking.Remove the tube from the bath and allow it to cool to room temperature. Adjust the volume of the liquid in the tube to 10 ml with ethanol and measure the optical density of the solution. Compare the readings with those for known amounts of glucosamine. Determine the amount of glucosamine from a standard graph prepared for the particular instrument used. Note-When solutions obtained from the hydrolysis of materials containing glucosamine are neutralised, it is recommended that the solution be made just alkaline with a solution of sodium hydroxide, then just acid with hydrochloric acid and finally neutralised exactly with a solution of sodium carbonate to pH 7, a pH meter being used.* Supplied by W. J. George and Beclrer, Birmingham.208 BELCHER, NUTTEN AND SAMBROOK [Vol. 79 SUMMARY A systematic and critical examination has been made of the method of Elson and Morgan for the determination of glucosamine. A satisfactory method has been described for the determination of glucosamine in the presence of certain sugars and amino-acids; it is based on the experimental observations made during the examination. The optimum conditions have been established for the two main reactions involved; in particular, the effects of varying pH, concentration of the reagents, temperature and time of heating on the reactions have been investigated, and information has been obtained about the latitude permissible in the attain- ment of the optimum conditions. Amounts of glucosamine (as hydrochloride) between 10 and 100 pg were determined, but smaller amounts can be determined, if necessary.The behaviour of glucosamine on heating with hydrochloric acid has been examined, and conditions have been established for the hydrolytic breakdown of N-acetyl-or-methyl- glucosaminide and heparin. We wish to thank Professor M. Stacey, F.R.S., and Dr. A. B. Foster, who asked us to One of us (C. M. S.) is grateful to the Colonial Products Research This paper describes work carried out by Miss C. M. Sambrook in partial fulfilment examine this problem. Council for financial assistance. of the requirements of the degree of M.Sc. in the University of Birmingham. REFERENCES 1. 2. 3. 4. 6. 6. 7. 8. 9. 10. 11. 12. Elson, L. A., and Morgan, W. T. J., Biochem. J., 1933, 27, 1824. Nilsson, J., Biochem. Z., 1936, 285, 386. S~rensen, M., Compt. Rend. Lab. Carlsberg, 1938, 22, 487. Boyer, R., and Fiirth, O., Biochem. Z., 1935, 282, 242. Blix, G., Acta Chem. Scand., 1948, 2, 467. Horowitz, H. N., Ikawa, M., and Fling, M., Arch. Biochem., 1950, 25, 227. Palmer, J. W., Smyth, E. M., and Meyer, K., J . Biol. Chem., 1937, 119, 491. Schloss, B., AIzaE. Chem., 1951, 23, 1321. Sideris, C. P., Young, H. Y., and Krauss, B. H., J . Biol. Chem., 1938, 126, 233. Bendich, A., and Chargaff, E., Ibid., 1946, 166, 283. Immers, J., and Vasseur, E., Nature, 1950, 165, 898. Wolfrom, M. L., Weisblet, D. I., Karabinos, J. V., McNeely, W. H., and McLean, J., J . Amer. Chem. SOC., 1943, 65, 2077. DEPARTMENT OF CHEMISTRY THE UNIVERSITY EDGBASTON, BIRMINGHAM, 16 October 20th, 1953

ISSN:0003-2654    

DOI:10.1039/AN9547900201

出版商:RSC    

年代:1954

数据来源: RSC

摘要:

April, 19541 JENKINS 209 The Absorptiometric Determination of Traces of Copper in Highly Purified Water BY E. N. JENKINS A method is described for the determination of copper in highly purified water at the low levels significant in aluminium corrosion. Traces of copper down to 0.001 p.p.m. can be measured absorptiometrically as cupric diethyl- dithiocarbamate after a single extraction from a 500-ml sample into 10 ml of chloroform in the presence of a citrate buffer and of disodium ethylenediamine- tetra-acetic acid. No interference results from the presence of 1 p.p.m. of common cations or of the sulphide or cyanide anions. A simple modification is described to obviate the interference of bismuth and antimony. THE biological importance of the presence or absence of traces of copper in natural and treated waters has stimulated the development of sensitive analytical methods for this element.For example, the United States Public Health Service recommends that the copper concentra- tion in drinking waters should not exceed 0-2 parts per million,l and reliable methods have long been established for water purity control at this level, for example, the use of the reagent sodium diethyldithiocarbamate to give an intense yellow colloid2 that can be stabilised by the addition of gum arabic. Such methods reach the limit of their sensitivity at about 0.01 p.p.m. Determination of copper at or below this level becomes necessary in the control of the “pitting” corrosion of aluminium by soft waters, which is greatly influenced by the copper content of the water and is said to persist at copper concentrations as low as 0.02 ~ .p . r n . ~ Analyses at such low levels have normally been preceded by concentration of the sample by evaporation, with the risk of contamination or losses. EXPERIMENTAL It was decided to attempt the solvent extraction of copper as a coloured complex from 500-ml water samples containing 0.01 p.p.m. or less of the metal into the minimum volume of immiscible organic solvent, and to measure the optical density of the entire extract in a cell of maximum path length. The basic problems involved are discussed below. CHOICE OF REAGENT- Reagents currently used for the colorimetric analysis of traces of copper include dithizone, sodium diethyldithiocarbamate, am‘-diquinolyl, 2 : 9-dimethylphenanthroline and biscycb- hexanone oxalyldihydrazone.The extinction coefficients of the appropriate copper complexes (that for dithizone being corrected for the extinction loss of the excess of reagent) are, respectively, 24,700 a t 5080 A,‘ 14,200 at 4320 A,’ 5490 at 5470 A , ~ , ’ 7950 at 4545 A,* and 16,500 at 5950 A’ at a concentration of one gram-atom of combined copper per litre. These reagents are known to form solvent-extractable copper complexes, except the last-named reagent, for which no solvent extraction procedure has been published. Diethyldithio- carbamate was chosen for the preient application because of its availability, high sensitivity and freedom from potential difficulties occasioned by the intense and unstable colour of the excess of the dithizone reagent, which would remain in the organic phase under the acid conditions normally used for the cupric dithizonate extraction.EXTRACTION OF COPPER AT LOW ORGANIC - AQUEOUS PHASE RATIOS- Many organic solvents are reported in the literature as being suitable for the extraction of cupric diethyldithiocarbamate. The volume of solvent used is commonly equal to that of the aqueous sample, which contains not less than 0-1 p.p.m. of copper. Sandelllo recommends that if it is necessary to determine copper in concentrations below 0.1 p.p.m., the sample should be concentrated by evaporation after acidification. Experiments with radioactive copper have shown, however, that an almost complete extrac- tion of copper by chloroform can still be achieved even at very unfavourable concentrations and phase ratios.210 JENKINS : THE ABSORPTIOMETRIC DETERMINATION OF TRACES [Vol.79 A solid dilution of pure cupric oxide in sucrose was irradiated for 12 hours in the Harwell pile (BEPO) at a slow neutron flux of about 10l2 neutrons per sq. cm per second, and dissolved in dilute hydrochloric acid to give a stock solution containing 1 pg of copper per ml “labelled” with radioactive ~ C U (12-9-hour half-life, 0-57-MeV beta-particles, 0.60-MeV positrons, 1.35-MeV gamma-radiation). The beta - gamma activity of the stock solution, and also of “labelled” 500-ml water samples before and after solvent extraction of the copper complex, were determined on 10-ml aliquots with a type M6 (liquid sample, skirt type) Geiger tube. Decay measurements indicated a half-life of about 12 hours, and the copper tracer was considered to be of satisfactory radiochemical purity.Next, 50 ml of 20 per cent. w/v tri-ammonium citrate solution and 0.100 g of sodium diethyldithiocarbamate were added to 500 ml of doubly glass-distilled water, and the mixture was brought to pH 9.3 by dropwise addition of concentrated ammonium hydroxide. Traces of copper present as impurity in the reagent were removed at this stage by an extraction with 25ml of pure chloroform (the first extraction). An addition of 5 pg of “labelled” copper was made ( L e . , the equivalent of a 500-ml sample containing 0.01 p.p.m.), and the aqueous solution was again extracted for 5 minutes with 25ml of chloroform. Radio- chemical assay of the residual aqueous phase demonstrated a satisfactory removal of copper.Further 5-pg additions of copper were made, followed each time by a &minute extraction with a small volume of chloroform. The results presented in Table I show conclusively that 0.01 p.p.m. of copper is extracted efficiently at very low organic - aqueous phase ratios (down to 1 to 100). It also appears that the successive extractions did not seriously deplete the reagent. TABLE I EXTRACTION OF LABELLED COPPER FROM AMMOTU’IUM CITRATE SOLUTION Beta - gamma assay of aqueous phase (corrected for background) Volume of Extraction sequence chloroform, ml Second . . .. .. 25 Third . . .. .. 25 Fourth ,. .. .. 10 Fifth .. .. .. 10 Sixth .. .. .. 5 r 1 before extraction, after extraction, Extraction, counts per minute counts per minute % 320 13 96 339 12 96-5 331 11 96.5 359 13 96 355 21 94 Ry way of interest, a further series of extractions was carried out under acid conditions (0.100 g of reagent in 500 ml of 0.1 N hydrochloric acid).It was again shown that copper was extracted efficiently (to the extent of 98 per cent.) into 10ml of chloroform, provided that, in these conditions, fresh reagent was added before each extraction. The extractions in which labelled copper was used, described above, were not carried out in the presence of ethylenediaminetetra-acetic acid. Subsequently the procedure was modified by the addition of this powerful complexing agent to the sample before the addition of reagent and citrate buffer, in order to mask the interference from cobalt and nickel.Haque, Brown and Bright5 state that the reaction of copper with diethyldithiocarbamate is a little slower in the presence of the ethylenediaminetetra-acetate ion, but that solvent extraction gives complete recovery of the copper within 1 minute. Jewsburyll investigated the time necessary for full colour development from copper with the carbamate reagent, in the aqueous phase, in the presence of the same complexing agent. He reports that the full colour was not developed within 30 minutes at a 25-pg level of copper ; the rate of reaction would be expected to decrease still further as the level of copper falls and as the concentration of complexing agent increases. The concentrations of reagent and complexing agent used by Jewsbury are identical with those present in the aqueous phase when the procedure recommended in this paper is used.The following tests were carried out with 10 pg of copper and five times the normal concentration of complexing agent, Le., in conditions where Jewsbury’s results would lead one to expect a prolonged inhibition of full colour development in the aqueous phase. The results show strikingly that, when solvent extraction is used, the copper colour is formed completely within 5 minutes, and that there is no necessity for a preliminary reaction period. Four pairs of samples containing about 0.02 p.p.m. of copper were analysed; the precise copper level varied from one pair to the next. Analysis of one member of eachApril, 19541 OF COPPER IN HIGHLY PURIFIED WATER 21 I pair was carried out by the method described in the procedure, but with the increased quantity of complexing agent noted above.Before a solvent extraction was made on the other member of the pair, the aqueous phase containing the reagent and complexing agent was allowed to stand at room temperature for 30 to 45 minutes. The results are shown in Table 11. TABLE I1 EXTRACTION OF COPPER IN THE PRESENCE OF COMPLEXING AGENT Optical density I A > Immediate extraction . . .. . . 0.742 0.753 0.639 0.661 Delayed extraction . , .. . . 0.758 0.780 0.630 0.664 It remained to be shown that the colour intensity per microgram of copper was unchanged in the presence of the complexing agent. This was established as follows. One series of very dilute copper solutions, containing 2 to 6pg of copper, was extracted in the presence of the same buffer and reagent concentrations as in the tracer experiments described above, but with only 50ml of aqueous phase (to enhance even further the extraction of copper).The second series, also containing 2 to 6 pg of copper, was extracted from 560 ml of aqueous phase containing 0.50 g of disodium ethylenediaminetetra-acetic acid in addition to the normal citrate buffer and dithiocarbamate reagent. The extractions and subsequent optical density measurements for the first and second series were carried out as detailed in the procedure described below. In the absence of complexing agent, values of 0.065, 0.054, 0.059, 0.062 (mean: 0.060) were obtained for the optical density per microgram of copper. Corresponding values in the presence of complexing agent were 0-0595, 0.061, 0.0605 (mean : It is clear that, under the conditions of reaction and subsequent extraction used in the present work, the ethylenediaminetetra-acetate ion in no way interferes with the extraction of cupric diethyldithiocarbamate.0.060). CHOICE OF ABSORPTION CELL AND FILTER- When it was established that copper could be extracted efficiently from 500-ml samples into, say, 10 ml of chloroform, it was necessary to measure the colour intensity on a Spekker absorptiometer under optimum conditions. The ratios of path length to working volume of the Hilger cells normally supplied with this instrument are unduly low, while the Hilger micro-cell holds only 0.5 ml. The Ogal type CLZ 3 cell, micro modification, supplied by The Tintometer Ltd., has a path length of 4 cm and a working volume of 8 ml, and has proved suitable for our purpose.The internal width of the cell, 6mm, allows the passage of a sufficiently wide beam to give ample sensitivity for detecting the balance point. Adapters were specially constructed to hold two such cells on the carriage of a Spekker model H 503 absorptiometer. The Ilford spectrum filter No. 601 (transmission band around 4300 A) is suitable for optical density measurements on the yellow cupric diethyldithiocarbamate (maximum absorption at 4320 A). MISCELLANEOUS PRELIMINARY TESTS- Before systematic extractions of known amounts of copper were undertaken to set up a calibration graph, two further possible variables were studied. Colow fuding-Sandell10 reports quite a rapid fading of the yellow colour of a solution of copper diethyldithiocarbamate in carbon tetrachloride exposed to diffuse daylight.The coloured complex corresponding to 30 pg of copper was extracted and made up to 50 ml with chloroform. The chloroform solution was allowed to stand in a stoppered volumetric flask in diffuse April daylight on our normal working bench. Samples were withdrawn at intervals and centrifuged, and the optical density was measured. The results show the colour to be stable over a period of an hour under conditions of moderate diffuse daylight, the initial optical density reading of 0.381 being unchanged after this time. Reagent storage-Previous workers have used a 0.1 per cent. aqueous stock solution of the reagent.It was convenient to use a 1.0 per cent. stock solution for the present work, and the efficiency of this reagent after varying periods of storage in the dark in a refrigerator at 0" C was investigated. The colour intensities of extracts containing approximately 10 pg of copper were compared when (a) fresh reagent and ( b ) stored reagent was used. A212 JENKINS : THE ABSORPTIOMETRIC DETERMINATION OF TRACES [Vol. 79 reagent solution stored for 8 days gave a normal colour intensity (an optical density of 0-65 for 10 pg of copper). After storage for 15 days, the full colour did not develop; the optical density was only 0.60. It was concluded that the reagent solution should be freshly prepared at weekly intervals. REAGENTS- 4 Ammoniztrn cityate solution, 20 per cent.w,h-Dissolve 173 g of analytical grade citric acid monohydrate in 800ml of glass-distilled water and slowly add 165 ml of analytical grade ammonium hydroxide, spgr. 04380. Cool and make up the volume to 1OOOrnl in a cylinder. Sodium diet~yZdit2tiocarbamafe solution, 1 per cent. w/v-Dissolve 1-00 g of pure sodium diethyldithiocarbamate, e.g. , the B.D.H. product, in 100 ml of glass-distilled water. Store it in a refrigerator. Dz'sodium ~h~le~~dz'u~z'netatra-acetic acid (solid)-A pure grade, a.g., the R.D.H. product. Chloroform-Analytical reagent grade. PROCEDURE Prepare freshly each week. kPARATUS- Spekker absorptiometer, model H 503. Ilford spectrum filters No. 601. Ogal 4-crn celIs No. CLZF 3 (micro), made by The Tintometer Ltd. Adapter to hold a pair of cells on the Spekker absorptiometer carriage.Clinical centrifuge with 12-rnl centrifuge tubes. PROCEDURE- In a 500-ml water sample dissolve with shaking 0 6 O g of disodium ethylenediamine- tetra-acetic acid and set the solution aside for 30 minutes. The addition of the complexing agent not only prevents interference by cobalt and nickel, but also ensures that the copper is present as a true solution of a stable ionic species. In the meantime, take 50 ml of buffer solution and 10 ml of carbamate reagent in a 500-ml conical separating funnel, add 10ml of chloroform, insert the stopper and shake for 2 minutes. Reject the chloroform layer. Repeat the extraction with a further 101111 of chloroform, again rejecting the lower layer. Extract for a third time with 10 rnl of chloroform, shaking for 5 minutes. Allow the layers to settle, dry the stem of the funnel with filter-paper, and deliver the extract into a 12-1111 centrifuge tube.Centrifuge at maximum speed for 2 minutes. Withdraw the chloroform layer with a 25-ml syringe-type pipette with a drawn-out tip. Avoid the inclusion of any droplets of the aqueous phase. Measure the optical density of this extract without delay, using the Spekker absorptio- meter with the tungsten lamp, No. H 503 heat filters and Ilford spectrum filters No. 601. Set the drum at 1-00 with pure chloroform in the reference cell. The blank measured at this stage (normally an optical density of 0.00 to 0.02) allows for the contribution of traces of copper as impurity in the solvent, together with any possible contribution due to extracted dithio- carbamic acid.Remove any droplets of chloroform from the bottom or stop-cock of the separating funnel. Shake with 10ml of chloroform for 5 minutes. During the extraction, the chloroform layer decreases in volume to about 8 ml. Allow the chloroform layer to settle, dry the stem with filter-paper, and deliver the extract into a dry 10-ml calibrated flask. Wash out the remaining drops of extract by successive additions of 0.5 ml of chloroform, swirling the funnel. Continue until exactly 10.0 ml has been collected. If the copper content of the sample exceeds 15 pg, make the extract volume up to 25 or 50 ml. Transfer the extract to a 12-ml centrifuge tube. Centrifuge, separate, and measure the optical density on the Spekker absorptiometer as described above for the solvent blank.The optical density reading must be corrected for the solvent blank and divided by the calibration factor (an optical density of about 0.065 per microgram of copper) found by experi- ments with standard copper solutions. The quotient is further divided by the sample volume in rnillilitres (normally 600 ml) to find the copper concentration in parts per million. Deliver the solution into a clean dry 4-cm absorptiometer cell. Add the 500-ml sample of water containing the complexing agent.April, 19541 OF COPPER IN HIGHLY PURIFIED WATER 213 CALIBRATION RESULTS A bulk stock of nominally high-purity water, produced by a mixed-bed de-ionisation process, was available. Disodium ethylenediaminetetra-acetic acid was added (1-0 g per litre) and well mixed, and 500-ml aliquots were taken.To each aliquot was added an appro- priate volume of a standard copper solution freshly prepared from analytical-reagent grade copper sulphate pentahydrate. Each sample was analysed for copper as described above. The results of two calibration series with different bulk samples of de-ionised water are presented in Table 111. TABLE I11 CALIBRATION DATA Optical density of I A 1 Optical sample corrected density per Series Copper added solvent blank sample for blanks microgram 2 CLg 2.50 0.020 0-282 0-154 0-062 7.50 0.020 0-629 0.50 1 0.067 12.5 0.01 1 0-972 0.853 0.068 17.5 0.013 1*294* 1.173 0-067 0 0.00 0.121 0.000 2.50 0.00 0.280 0.159 0-064 7.50 - 0.02 0-570 0.471 0.063 12.5 - 0.02 0-93 1 0,830 0.066 17.5 0.00 1-272* 1.151 0.066 Mean .. 0-065 1 0 0-022 0.130 0.000 * Calculated for an extract volume of 10 ml from optical density measurements made on 25 ml. The results, corrected for the solvent blanks and for the water blanks, indicate a linear relation between copper content and optical density up to 17.5 pg of copper. Preliminary experiments indicated that linearity persisted up to at least 80 pg. The results also give figures of 0-003 p.p,m. and 0.004 p.p.rn. for the copper contents of the two bulk samples of de-ionised water. Doubly glass-distilled water normally had no detectable copper content ; this served also to demonstrate the absence of any significant copper impurity in the B.D.H. disodium ethylenediaminetetra-acetic acid used.The limit of detection of the method is set by the small solvent blank of optical density 0.00 to 0.02, which recurs even on repeated extraction and is variable from day to day. A 500-rnl sample containing 0.001 p.p.m. of copper would give a final optical density of 0-033 higher than the blank. INTERFERENCES Factors likely to interfere with the quantitative extraction of copper as the cupric diethyldithiocarbamate salt include the following. Metals reacting with the reagent to decrease its effective concentration and in some instances to form extractable salts with appreciable absorptions a t 4320 A. Anions that mask the copper and inhibit the colour reaction. Oxidking or reducing agents, which oxidise the reagent, e.g., to the disulphide, or which reduce the copper to the cuprous state, which forms a salt of relatively weak colour intensity. These factors are considered below.OTHER METALS- Elements reported to form coloured or white insoluble salts with diethyldithiocarbamic acid in alkaline solution12 include copper (cuprous or cupric) , antimony, bismuth, cadmium, cobalt, chromium (as dichromate), gold, indium, iron (ferrous or ferric), lead, manganese, mercury, nickel, platinum metals, silver, tellurium, thallium (thallous or thallic) and zinc. These salts will usually be extractable into chloroform; often the chloroform solutions are coloured and have a certain degree of absorption within the range of the Ilford 601 filter (3800 to 4 8 0 0 ~ ) . Iron, cobalt, nickel and bismuth are usually considered to present the most serious colour interferences to the determination of copper in alkaline solution. The214 JENKINS: THE ABSORPTIOMETRXC DETERMINATION OF TRACES [Vol.79 first element in the ferric state can be masked by the citrate ion. Cobalt and nickel can be formed into complexes with dimethylglyoximelO or with ethylenediaminetetra-acetic a ~ i d . ~ , l ~ The present work included tests on a series of 500-ml water samples each containing 5 mg of a potentially interfering eIernent. Each sample was analysed for copper by the method given above, and the results were corrected for the blank obtained on the pure water sample. No significant interfering colour (ie., the equivalent of less than 1.0 pg of copper) was found in the tests with cadmium, cobalt, chromium (as dichrornate), chromium (as the chromic cation), ferric and ferrous iron, lead, manganese (as manganous), manganese (as perrnanganate), nicke1, stannous tin, thallium (as thallous ion) and zinc.I t should be noted that the inhibition of cobalt and nickel interferences depends on the addition of cornplexing agent before the reagent. Preliminary experiments in which the nickel or cobalt solutions were added to a mixture of citrate buffer, reagent and complexing agent showed only a partial reduction in the intensity of the extracted colour. Kinetic effects are probably operative, i.e., a cobalt or nickel diethyldithiocarbamate salt, possibly in colloidal suspension, may react only slowly with the complexing agent. In a similar manner, cobalt and nickel sulphides and dithizonates are resistant to decomposition by dilute acid.14 Very minor interferences, equivalent to less than 1Opg of copper, were detected with 5 nig of silver or of mercuryIT. For each metal these small interferences were confirmed by duplicate experiments. For all metals studied and listed above, complete recoveries of 10 pg of added copper were achieved from the residual aqueous phases in a further extraction. Bismuth and antimony give extractable yellow compounds with the reagent in the procedure described above.The elimination of these interferences is discussed later. MASKING BY ANIONS- We have already established that the strong cornplexing of traces of copper by 0.003 M disodium ethylenediaminetetra-acetic acid at pH 9 does not in any way inhibit the subsequent extraction of this metal as the diethyldithiocarbamate.In view of this fact, it seems extremely unlikely that traces of anions in general will mask the coIour reaction, with the possible exceptions of cyanide and sulphide, which form a very stable complex anion and a very insoluble salt, respectively. Amounts of cyanide or sulphide from 1 to 10 p.p.m. were added to a series of 500-ml water samples containing 0.020 p.p.m. of copper. The final mixtures were weakly alkaline, pH 9.5 to 11. Some of the mixtures were allowed time to age before the normal addition of 500 mg of disodiurn ethylenediaminetetra-acetic acid, followed by addition of the mixture to the copper-free buffer - reagent mixture and final extraction with 10ml of chloroform. The oDtical densitv measured on the centrifuged chloroform extracts (made up t o 10.0 ml) are iresented in 'Tables IV and V.v TABLE IV INTERFERENCE BY CYANIDE IK COPPER EXTRACTIOX Cyanide, p.p.rn. . . .. . . 0-0 1.0 5-0 10~0 1.0 Ageing . . * . . . . . None None None None Z day Optical density . . 1 . . . 0.637 0.634 0.302 0.306 0.633 0-622 0.145 TABLE V INTERFERENCE BY SWLPKIDE I N COPPER EXTRACTION Sulphide, p.p.m. . . .. . . 0.0 1.0 10.0 5.0 Ageing . . .. .. . . None None None 1 day OpticaI density . . .. . . 0.637 0.630 0.631. 0.626 5.0 1 day 0.256 10.0 1 day 0-59 1 It is clear that up to 5p.p.m. of sulphide and 1 p.p.rn. of cyanide do not inhibit the full extraction of 0.01 p.p.m. of copper. Larger amounts of cyanide produce a marked inhibition. OXIDISING AND REDUCING AGENTS- Operations in the atomic energy industry may expose water, e g ., cooling water, to heavy ionising radiations, when traces of hydrogen peroxide will be formed. Excessive amountsApril, 19541 OF COPPER IN HIGHLY PURIFIED WATER 215 of peroxide may be expected to oxidise the diethyldithiocarbamate reagent ; interference with the copper determination could then arise either through depletion of the reagent or possibly through the oxidation products. In a series of five experiments, 5 to 26 p.p.m. of hydrogen peroxide were added to 500 ml of pure water containing 10 pg of copper and 0-50 g of com- plexing agent. After being mixed and set aside for at least 1 hour, the samples were analysed for copper in the standard manner. The results, shown in Table VI, establish that concentra- tions of peroxide up to 26 p.p.m.do not seriously interfere. TABLE VI INFLUENCE OF HYDROGEN PEROXIDE Hydrogen peroxide, p.p.m. . . 0 5.2 5.2 26 26 Corrected optical density . . 0-636 0.646 0.663 0.650 0.663 The effect of traces of hydroxylamine hydrochloride, which might reduce copper t o the cuprous state to give a less intensely coloured diethyldithiocarbamate, was examined in a similar fashion. The results presented in Table VII show that 1 p.p.m. of hydroxylamine hydrochloride does not seriously interfere. TABLE VII INFLUEXCE OF HYDROXYLAMINE Hydroxylamine hydrochloride, p.p.m. - . 0.0 1.0 1-0 1.0 Corrected optical density . . . . . . 0-636 0.640 0,630 0.610 REMOVAL OF BISMUTH INTERFERENCE Bismuth and antimony interference persists even in the presence of the complexing agent.Experiments in which bismuth nitrate solution was added to 560 ml of a copper-free (pre-extracted) solution of reagent, buffer and complexing agent, after which the usual extraction was carried out, gave a consistent ratio of about 20 parts of bismuth equivalent to 1 part of copper. Preliminary experiments with antimony gave a ratio of 120 to 1. Attempts at masking the bismuth with 8-hydroxyquinoline-8-sulphonic acid, or its 7-iOdO derivative, were not successful. Though the antimony and bismuth are still extracted into the chloroform, it may yet be possible to separate them from the copper by a suitable back- extraction. Gleu and Schwab12 infer that macro amounts of antimony and bismuth do not give a precipitate with sodium diethyldithiocarbamate from 1.0 N sodium hydroxide solutions. This suggested an attempt to remove these elements from the chloroform extract by back- extraction with an equal volume of 1.0 N sodium hydroxide solution. For these experiments, 500pg of antimony or bismuth were added to a copper-free solution of the reagent, buffer and ethylenediaminetetra-acetic acid, and extracted into 10 ml of chloroform.The extract was separated and made up to 50.0m1, 10-ml aliquots from which were then subjected to various back-extraction treatments in a 100-ml separating funnel, and the optical densities of the residual chloroform layers were compared (Table VIII). TABLE VIII BACK-EXTRACTION OF BISMUTH FROM CHLOROFORM SOLUTION Aqueous solution used None 10 ml of for back-ex tract ion carbonate- free 1.0 N sodium hydroxide Optical density of 0.308 0.057 residual chloroform (mean) 0.059 layer, in replicate 0-046 experiments 0-049 Two extrac- tions with 25ml of 10ml of carbonate- carbonate- free free 3.0 N 1.0 N sodium sodium hydroxide hydroxide 0.039 0.009 10ml of 10ml of sodium free hydroxide 0.1 N (analytical sodium 1.0 N carbonate- grade) hydroxide 0.074 0.242 0.138 0.151 0.055 10ml of 0.5 N sodium hydroxide + 0-5 N sodium carbonate 0.125 A single back-extraction with an equal volume of carbonate-free 1.0 N sodium hydroxide removes 80 to 85 per cent.of the bismuth colour, a second extraction bringing the over-all216 J EN KJNS [Vol. 79 figure up to 97 per cent. A single preliminary experiment with antimony showed a 90 per cent.removal of the colour. The importance of maintaining a very high pH value during the back-extraction is obvious from the variable results obtained with freshly prepared solution of analytical grade sodium hydroxide and from the poor results with 0.1 N sodium hydroxide (even in the absence of carbonate) and with a mixture of 0.5 N sodium hydroxide and 0.5 IV sodium carbonate. The time of shaking during extraction did not appear to be critical; the yellow colour of the chloroform layer disappeared almost immediately and excellent results were achieved after shaking for only 1+ minutes. It may beestimated that 0.5 mg of added bismuth will show only the equivalent of 0.7 pg of copper after a double back-extraction. The copper diethyldithiocarbamate in chloroform solution completely resists the back- extraction with 1.0 N sodium hydroxide, as shown by the following results: 20 pg of copper in an extract made up to 50ml gave an optical density of 0.258.After back-extraction, the optical density was 0-272 and 0.272 in duplicate experiments. MODIFIED PROCEDURE IN THE PRESENCE OF ANTIMONY OR BISMUTH Follow the normal procedure, described above, to the point where the chloroform extract has been made up to exactly 10ml in a calibrated flask. Shake the extract for 3 minutes in a 100-ml separating funnel with an equal volume of carbonate-free 1.0 N sodium hydroxide solution (conveniently prepared by passing a 4 per cent. w/v solution of analytical-reagent grade pellets through a short column of a strong-base anion-exchange resin, such as De-acidite FF, and rejecting the first fraction). The efficiency of bismuth (and probably also of antimony) removal can be increased from 80 to 85 per cent. to 97 per cent. by a second treatment of the chloroform solution with an equal volume of alkali. Centrifuge the residual chloroform extract and measure the optical density as previously described. Separate the chloroform layer. The author wishes to acknowledge the technical assistance of Mr. P. Tumber and Mr. C. Scott-Taggart, and to thank the Director of the Atomic Energy Research Establishment for permission to publish this paper. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13, 14. Monier-Williams, G. W., “Trace Elements in Food,” Chapman and Hall Ltd., London, 1949, p. 37. Institution of Water Engineers, “Approved Methods for the Physical and Chemical Examination Porter, F. C., lecture summary, Chem. G. Ind., 1953, 561. Liebhafsky, H. A., and Winslow, E. H.. J . Amer. Chem. SOC., 1937, 59, 1966. Haque, T. L., Brown, E. D., and Bright, H. A., J . Res. Nut. Bur. Stand., 1951, 47, 380. Hoste, J., Anal, Chim. Acta, 1950, 4, 23; Research, 1948, 1, 713. Guest, R. J., Unclassified report TR-105/52, Department of Mines, Ottawa, 1952. McCurdy, W. H., and Smith, G. F., Analyst, 1952, 77, 846. Wetlesen, C.-U., and Gran, G., Svensk Pafiperstidning, 1952, 55, 212. Sandell, E. B., “Colorimetric Determination of Traces of Metals,” Interscience Publishers Inc., Jewsbury, A., Analyst, 1953, 78, 363. Gleu, K., and Schwab, R., Angew. Chem., 1950, 60, 320. Sedivec, V., and Vasak, V., Coll. Czech. Chem. Comm., 1950, 15, 260. Sandell, E. B., 09. czt., p. 91. ATOMIC ENERGY RESEARCH ESTABLISHMENT of Water,” London, 1949. New York, 1950, p. 317. HARWELL, NR. DIDCOT, BERKS. October 9th. 1953

ISSN:0003-2654    

DOI:10.1039/AN9547900209

出版商:RSC    

年代:1954

数据来源: RSC