ISSN:0003-2654    

DOI:10.1039/AN95681FX035

出版商:RSC    

年代:1956

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN95681BX037

出版商:RSC    

年代:1956

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN95681FP079

出版商:RSC    

年代:1956

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN95681BP085

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

JULY, 1956 THE ANALYST Vol. 81, No. 964 PROCEEDINGS OF THE SOCIETY FOR ANALYTICAL CHEMISTRY XORTH OF ENGLAND SECTION THE Thirty-first Annual General Meeting of the Section was held at 2.15 p.m. on Saturday, January 28th, 1956, at the Engineers’ Club, Albert Square, Manchester. The Chairman of the Section, Mr. J. R. Walmsley, A.M.C.T., F.R.I.C., F.P.S., presided. The following appointments were made for the ensuing year :-Chairman-Mr. J. R. Walmsley. Vice- Chairman-Mr. A. N. Leather. Hon. Secretary and Treasurer-Mr. A. C. Wiggins, J. Lyons & Co. Ltd., 5 Laurel Road, Liverpool, 7. Members of Committee-Messrs. E. G. Brown, -4. C. Bushnell, J. F. Clark, T. W. Lovett, R. Sinar and E. P. Underwood. Hon. Auditors -Messrs. F. Dixon and C. J. House. The Annual General Meeting was followed by an Ordinary Meeting of the Section at which the following paper was presented and discussed : “Applications of Newer Techniques to the Analysis of Pharmaceutical .Products,” by D.C. Garratt, B.Sc., Ph.D., F.R.I.C. NORTH OF ENGLAND SECTION AND MICROCHEMISTRY GROUP A JOINT Meeting of the North of England Section and the Microchemistry Group, together with the Bradford Chemical Society, was held at 6.30 p.m. on Friday, May 25th, 1956, in the Chemistry Lecture Theatre, Bradford Technical College, Bradford 7. The Chair was taken by the Chairman of the Bradford Chemical Society, Dr. R. L. Elliot, F.R.I.C., F.S.D.C. The subject of the meeting was “Microvolumentric Analysis” and the following papers were presented and discussed: “Apparatus and Technique,” by D.W. Wilson, M.Sc., F.R.I.C. ; “Primary Standards,” by R. Belcher, B.Sc., Ph.D., F.Inst.F., F.R.I.C. ; “End-point Location,” by E. Bishop, BSc., A.R.T.C., A.R.I.C. (see summaries below). The meeting was preceeded in the afternoon by a visit, by kind permission of the Coal Tar Research Association, to their laboratories at Gomersal, Leeds. APPARATUS AND TECHNIQUE MR. D. W. WILSOK described the errors relating to volume measurement that become appreciable when titimetric apparatus was reduced in size. Error due to the volume of the emergent burette drop was overcome by adding very small increments, preferably by titrating with the tip of the burette immersed. The entrapment of a variable volume by, and leakage around, the tap were eliminated by substituting other forms of control.Error due to variable drainage was avoided by a radical alteration in the design of the burette. Mr. Wilson described the operation of the horizontal burette, and various types of syringe burette, as well as the “walking-stick” type of syringe pipette. He concluded by saying that the actual measurement of volume on the small scale, using the apparatus described, presented no difficulties. PRIMARY STANDARDS DR. R. BELCHER said that although none of the many recommended standard substances answered exactly the requirements of an analytical primary standard as laid down by F. D. Dodge ( J . Ind. Eng. Chem., 1915, 7, 29), some came near to fulfilling them. The merits of sodium carbonate as a primary standard had recently been questioned. Little would be said about this standard, except to state that it had always given reliable results in the experience of the lecturer.381382 PROCEEDINGS pol. 81 The difficulty of obtaining the pure decahydrate when borax was to be used as an acidimetric standard could be overcome by recrystallising an AnalaR sample and then leaving it for 3 days in a hygrostat. Ilt is not sufficient to place the commercial product in a hygrostat without this prior treatment, Although previous studies on the thermal stability of potassium hydrogen phthalate had shown a safe upper limit of 135" C for drying the material, more recent work had indicated that drying could be carried out at 240" C. The temperature of drying had been given variously in the literature. Even though solutions of sulphamic acid hydrolysed slowly at room temperature, a constant titre should be obtainable became the hydrolytic product is the hydrogen sulphate ion.A sample of commercial potassium bi-iodate gave results from 1-6 per cent. high t o 0.04 per cent. high after three recrystatllisations. Dr. Belcher then outlined a few of the newer primary standards. Of the newer alkalimetric standards, potassium hydrogen bis-(3 : 5-dinitrobenzoate) and 2 : 4 : 6-tri- nitrobenzoic acid both possessed many desirable properties. The latter reagent had the unusual property of serving as its own indicator. Recently recommended standards in acidimetry were tris(hydroxymethy1)amino- methane and 4-aminopyridine. The indicators proposed for use with the former standard were not very satisfactory.4-Aminopyridine appeared on paper to be very attractive, but it remained virtually untried as yet. The concentration of solutions of sulphuric acid (32 to 39 per cent. by weight) could be accurately calculated from specific gravity by means of a simple equation. Such solutions could be kept for at least six :months and were practically non-hygroscopic. The suitability of this as a means of preparing standard solutions of sulphuric acid by direct weighing had been substantiated. In the preparation of standard i0din.e solutions by titration against a primary standard, use of arsenious oxide was usually recommended. Recently, it had been confirmed that barium thiosulphate monohydrate could be used as a direct standard for iodine solutions. A new compound, Ce.en2.(S0,),.2H,SO,.7H,O (en = ethylenediamine), had recently been proposed as a standard substance in cerimetric titrations.The equivalent weight was 774-7. This had been verified over a period of 213 days. This substilnce should have wide applications. END-POINT LOCATION MR. E. BISHOP said that any property of a solution that underwent a change at the equivalence point of a titrimetric reaction might be used to locate that point. The end-point of the titration was the point oE maximum rate of change of the indicating system, but this might not coincide with the equivalence point at which the stoicheio- metric amount of titrant had been added. Aside from certain minor physical methods, end-points were commonly located by visual indicators or by electrical means.The first class included principal-reagent indication, in which the titrant served as its own indicator, e.g., permanganate, cerate, iodine and clear-point titrations with silver ; and ancillary-reagent indication, in which an ancillary reagent which also reacted with the titrant was added to the solution being titrated, e.g., colour, adsorption, luminescent, chemiluminescent , turbidity and surface-active indicators. Such changes might be observed by a variety of visual or photo-electric techniques. The second class involves measurement of electrical quantities, potential, current (polarographic diffusion current), low-frequency resistance (ohmic resistance) and high radio-frequency resistance (reactance plus ohmic resistance). The last technique could be further subdivided into Q-metric or energy absorption, rectified radio-frequency conductance and beat-frequency or oscillometric measurements.A graph of the electrical quantities measured against volume of titrant added showed some form of break at the end-point. The potentiometric method was dependent on the logarithm of the con- centration of the indicating ion, and was often more sensitive than the others, which depended directly on concentration. This class also included many other electrical techniques based upon the behaviour of electrodes in ionic solutions, such as the dead- stop, bimetallic, reference indicator, differential-potentiometric, diff erential-polarographic and di ff erential-elect roly t ic met hods.July, 19561 PROCEEDINGS 383 All these methods might theoretically be used on any scale, but as the scale of operation decreased, physical difficulties of observation and manipulation imposed limitations, and since it was always better to reduce scale rather than reagent concentra- tion, it was found that visual indicators were unsuitable below about O-OZml, radio- frequency methods were at present limited to cell capacities of about 0-5 ml, although they would handle very dilute solutions, and other electrical methods reached a limit at about 0.01 ml owing to difficulties of accommodating electrodes, burette tip, salt bridge and other components in the available solution volume.The advantages of the author’s differential-electrolytic potentiometric method (Mikrochim. Acta, 1956, 619), which has made possible titrations of the order of 0.0005 ml in working volumes of about 0.001 ml, afford an additional useful tool for working on the ultra-micro scale.A brief description of this method was given, and some results obtained in redox reactions were shown. WESTERN SECTION A JOINT Meeting of the Section with the Bristol Sections of the Royal Institute of Chemistry, the Society of Chemical Industry and the Chemical Society and the Oils and Fats Group of the Society of Chemical Industry was held at 7 p.m. on Thursday, January 26th, 1956, at Bristol University, Woodland Road, Bristol, 8. The Chair was taken by the President of The Society for Analytical Chemistry, Dr. K. A. Williams, A.Inst.P., M.Inst.Pet., F.R.I.C., who is also Chairman of the Oils and Fats Group of the Society of Chemical Industry.A lecture on “Industrial Application of Sequestering Agents” was given by R. L. Smith, B.Sc., Ph.D., A.R.I.C., and P. Womersley. PHYSICAL METHODS GROUP THE fifty-fourth Ordinary Meeting of the Group was held jointly with the Photoelectric Spectrometry Group at 2.30 p.m. on Friday, May 25th, 1956, in the Physical Chemistry Laboratory, Oxford. Dr. J. E. Page, F.R.I.C., Chairman of the Physical Methods Group, took the Chair during the afternoon session and Mr. C. G. Cannon, Chairman of the Photo- electric Spectrometry Group, during the evening. The subject of the meeting was “Nuclear and Paramagnetic Resonance’’ and the following papers were presented and discussed : “Analytical Application of Nuclear Resonance Spectro- scopy,” by R. Richards, M.A., D.Phi1. ; “Techniques of Magnetic Resonance Spectroscopy,” by E. E. Schneider, Dr.Phil.Nat. ; “The Detection of Photochemically Formed Radicals by Magnetic Resonance,” by D. J. E. Ingram, M.A., D.Phil. During the tea interval members were invited to visit Dr. R. Richards’s laboratory, where his nuclear magnetic resonance equipment was demonstrated. MICROCHEMISTRY GROUP THE sixth London Discussion Meeting of the Group was held on Wednesday, June 20th, 1956, at 6.30 p.m. in “ The Feathers,” Tudor Street, London, E.C.4. The Chair was taken by the Chairman of the Group, Dr. G. F. Hodsman, A.1nst.P. The subject, “Kjeldahl Determination of Nitrogen,” was introduced by Dr. R. E. Stuckey, F.P.S., F.R.I.C., and Mr. P. R. W. Baker, B.Sc., A.R.I.C., and an informal discussion followed.

ISSN:0003-2654    

DOI:10.1039/AN9568100381

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

384 FRANKS PAPER CHROMATOGRAPHY WITH CONTINUOUS [Vol. 81 Paper Chromatography Change in Solvent Part I. Separation of with Continuous Composition Fatty Acids BY F. FRANKS* (Presented at the meeting of the Society on Wednesday, Aeril 4th, 1956) A method and apparatus described permit paper partition chromatography to be carried out with a solvent whose composition changes continuously in an exponential manner. The method is preferred to more conventional pro- cedures, because closely similar compounds such as members of a homologous series can be separated on a short paper strip, thus avoiding diffusiveness of zo nes. fatty acids obtained from acetone extracts of laundered articles. The method has been applied to the separation of mixtures of long-chain SEVERAL workers have described methods for the chromatographic separation of fatty acids.1,*A4,5~6 Although the lower members of the series can be fairly readily separated, the higher members whose physical properties do not change very much with increase in chain length can only be separated by employing reverse-phase partition chromatography.This involves impregnation of the paper with the stationary phase by immersion, spraying or capillary ascent. It has been found that all these procedures produce concentration gradients on the paper, thus making it difficult to obtain reproducible results. It has also been reported4 that small variations in the mobile phase will lead to considerable fluctuations in the RF values of the fatty acids. Spiteri6 has developed a method that has given fairly satisfactory results.The fatty acids with even-number carbon chains from capric to lignoceric have been resolved on a paper strip, liquid paraffin being used as the stationary solvent and glacial acetic acid as the mobile solvent. Although all the acids were separated, the degree of separation was not very good. Capric, lauric and myristic acids were found to be rather close together and only just beneath the solvent front. As the solvent had travelled a distance of ZOcm, the spots were diffuse and not completely separated from each other. The chief difficulty therefore lies in the choice of the mobile phase. This problem is brought out particularly well by Spiteri and Nunez,l who state that, when a particular mixture of ethanol and water is used, the RF values of lauric and stearic acids are 0.65 and 0.0, respectively.However, when the ethanol concentration is increased, the respective R F values become 0.86 and 0.30. The significance of these results is that for a certain mobile phase concentration only a limited number of acids can be resolved, so that acids with a longer carbon chain will not migrate from the origin, whereas acids with shorter carbon chains will be carried along the paper by the solvent front. The argument that has been developed for the fatty acids applies equally well to other homologous series. In order to achieve better separation of zones, gradient elution analysis has been applied to column chromatography. This technique involves a continuous or stepwise change in the polarity of the elution medium ; it has yielded satisfactory results in the separation of the lower fatty acids' up to C,, and also in the resolution of acids belonging to the citric acid As far as I am aware, no attempt has so far been made to apply gradient elution analysis to the separation of long-chain fatty acids and their derivatives or to adapt the technique to paper chromatography.In view of the difficulties outlined above a method has been studied that involves reverse-phase partition chromatography on paper, the mobile solvent composition being altered by constant-rate dilution or concentration, whichever is necessary. * Present address : South East Essex Technical College, Dagenham, Essex.July, 19561 CHANGE IN SOLVENT COMPOSITION. PART I 385 APPARATUS The apparatus used in shown in Fig.1. Two glass plates, A, of approximate dimensions 50 cm x 20 cm x 1 cm (with circular chromatography the dimensions are 25 cm x 25 cm x 1 cm) hold the filter-paper between them. The lower of the two plates is provided with a hole of diameter 12 mm. The plates rest in a horizontal position on the neck of a conical flask, B, of capacity 100n11, which acts as solvent reservoir. The seal between flask and A -D 111 w - Fig. 1. Diagram of apparatus glass plate is made with plasticine. The filter-paper is cut to the shape shown in Fig. 2, so that two determinations can be carried out simultaneously on the same strip. The flask is fitted with two side-arms: the side-arm C has a €314 socket and D is a length of 5-mm diameter tubing attached to the flask at the 100-ml level. It slopes slightly downwards and is bent to the vertical at a distance of 2 cm from the point of attachment.The vertical section extends to just below the level of the bottom of the flask. The constant-head solvent reservoir, E, is connected with the flask by means of a capillary tap, F, a length of rubber tubing provided with a screw clip and a 0-5-mm capillary tube sealed into a R14 cone; G is a 500-ml separating funnel containing the diluting solvent, The stem of the funnel reaches below the level of the liquid in E, which is determined by the positioii of the overflow tube, H. The magnetic stirrer, J, ensures constant agitation of the solvent in 13.FRANKS : PAPER CHROMATOGRAPHY WITH CONTINUOUS THEORY [Vol. 81 The flask contains a solution whose concentration Solvent is introduced at a constant rate of v ml per second by means of the 386 Let the volume of flask B equal I' ml.is c g per ml. Fig. 2. of fatty acids. Shape and approximate dimensions of filter-paper strips used in the separation Each strip contains two chromatograms capillary in side-arm C. Assuming that the system is homogeneous, i.e., perfect mixing, then- Solution leaves the flask at the same rate through side-arm D. weight of solute leaving flask per second = cv g, fraction of solute leaving flask per second = ~ = -- cl' JT' cv v d ( c V ) = - y ( c V)dt, V where t = time in seconds. Let co be the initial concentration and ct the concentration after time t, then-- ct = co e - 3 NOTE-For a flask having a capacity of 100 ml or less and for rates of dilution up to 30 ml per hour no appreciable error is introduced by assuming perfect mixing. EXPERIMENTAL The paper was impregnated with liquid paraffin by capillary ascent of a 10 per cent.v/v solution of liquid paraffin in benzene. The mixture of fatty acids contained approximately 40 pg each of the following acids: capric, lauric, myristic, palmitic and stearic. Initially water was placed in the solvent reservoir, whereas the separating funnel and the constant-head reservoir contained glacial acetic acid. From observing the movement of the acid zones, it was later decided to substitute 50 per cent. v/v acetic acid for the initial solvent. The papers were placed in position with the wicks dipping into the solvent. The upper glass plate was then placed on top of the papers and the magnetic stirrer was started.The time was taken and tap F was opened so as to allow acetic acid to flow into the flask at a constant rate. The rate was measured by placing a graduated cylinder under tube D and recording the volume after timed intervals. Graphs were constructed on log graph paper of the time in hours against solvent composition for various rates of dilution. It was then possible to determine the concentration of acetic acid in the flask at any time. The chromatogram was developed for periods up to 24 hours. By varying the width of the wick, it was possible to adjust the rate of development. The papers were then removed and dried until opaque; the fatty-acid zones were developed by the silver-salt method.For the separation of fatty acids the method of Spiteri6 was taken as a basis. METHOD Acetic acid, glacial-Analytical-reagent grade. Fatty acids fur re fereme mixture-Lauric, myristic, palmitic, stearic, oleic and linoleic RE-4GENTS- acids were of Kahlbaum reagent grade. Capric acid was of reagent grade, m.p. 30" C.July, 19561 CHANGE I N SOLVENT COMPOSITION. PART I 387 Methanol-Analytical-reagent grade methanol was heated under reflux for 8 hours with silver nitrate and potassium hydroxide ; it was then fractionally distilled. Liquid parafin-Medicinal grade that had been extracted with glacial acetic acid. Benzene-Analytical-reagent grade. Silver nitrate-A 1 per cent. ammoniacal solution prepared from analytical-reagent Sodium suZphide-A very dilute solution prepared from the analytical-reagent grade Ammonia sohtion, sp.gy.0.880-Analytical-reagent grade. Cut out 40-cm x 80-cm strips of Whatman No. 1 filter-paper. grade material. This solution is freshly prepared. solid. PROCEDURE FOR IMPREGNATING THE FILTER-PAPER- Suspend the papers in a chromatographic tank, allowing the lower ends to dip into a 10 per cent. v/v solution of liquid paraffin in benzenea6 When the solution has ascended a distance of 30 cm, remove the papers and allow them to dry at room temperature. This solution is freshly prepared. PROCEDURE FOR APPLYING THE FATTY-ACID MIXTURE- a platinum-wire loop. unknown mixture. the concentrations should be approximately 10 g per litre. paper. PROCEDURE FOR DEVELOPING THE CHROMATOGRAM- Assemble the apparatus as shown in Fig.1. Fill the flask to the level of the overflow tube with 50 per cent. v/v acetic acid solution, and fill the separating funnel and the constant- head solvent reservoir with glacial acetic acid, taking care to remove all air bubbles from the rubber tubing and the capillary tap. Lay the papers on the lower glass plate with their wicks dipping into the solvent in the flask, and place the upper glass plate in position. Start the magnetic stirrer and note the time. Open the capillary tap and adjust its position until the desired rate of flow of solvent is obtained. Measure the volume by means of a graduated cylinder placed under the overflow tube. Allow the chromatogram to run until the solvent front has advanced a distance of 15 to 20 cm, depending on the rate of change of solvent composition.Record the time and measure the total volume in the graduated cylinder. PROCEDURE FOR COLOUR DEVELOPMENT (AFTER SPITERIG)- Place them in a shallow dish and wash them several times with water. Immerse them in the smallest volume of a 1 per cent. solution of ammoniacal silver nitrate, and gently agitate the dish for 10 minutes. Pour away the solution and wash the papers with water until the washings no longer give a positive reaction for silver. Immerse the papers in an extremely dilute solution of sodium sulphide. When dark spots of silver sulphide begin to appear, discard the solution and replace it by a fresh one. When the colour of the spots has developed to its maximum intensity, discard the solution and wash the papers several times with water.Finally dry the papers at room temperature. After cutting the papers to shape, as shown in Fig. 2, apply the mixture of acids with Prepare a chromatogram of a reference mixture alongside the For this purpose prepare standard solutions of fatty acids in methanol; Apply 60 pg of each acid to the Remove the papers, and stop the solvent flaw. Dry the papers at room temperature until they are opaque. RESULTS Fig. 3 shows chromatograms prepared according to the method described above. Fig. 3 (a) illustrates a mixture of oleic and linoleic acids, whereas Fig. 3 (b) shows the resolution of saturated acids from capric to stearic acid. The rate of change of solvent concentration was 30 ml per hour, and the chromatogram was developed for 21 hours, during which period the solvent front moved through a distance of 13 cm.Although RF values determined by this method are subject to the rate of change of solvent composition, nevertheless they do give a good indication of the degree of resolution of the mixture. The RF values obtained by the method outlined above are compared in Table I with those obtained by other workers. Table I1 shows the corresponding values of (R,-z - R,J, where n is the number of carbon atoms. This quantity is a measure of the separation between two adjacent spots on the chromat ogram.388 [Vol. 81 FRANKS : PAPER CHROMATOGRAPHY WITH CONTINUOUS TABLE I RF VALUES O F FATTY ACIDS DERIVED FROM DIFFERENT CHROMATOGRAPHIC METHODS RF found by A I \ Holasek and SDiteri and Kobrle and Number of carbon atoms in acid molecule 10 12 14 16 18 VALUES OF - Number of carbon atoms (n) 12 14 16 18 Winsauer Nunez Zahradnik Spiteri method4 methodl methods method6 Present method 0.44 - 0-80 0.94 0.59 0.56 0.86 0.74 0.90 0.46 0.64 0-77 0-56 0.85 0.33 0.72 0.28 0.39 0-76 0.2 1 0-80 0.24 0.29 0.66 0.13 TABLE I1 &) DERIVED FROM THE RF VALUES SHOWN I N TABLE I 7 Holasek and Winsauer method4 -0.12 - 0.08 - 0.08 - 0.08 (Rn-2 - Rtt) calculated for results by Kobrle and Zahradnik Spiteri A -7 method6 method6 Present method 0.10 0.04 0.13 0-14 0.05 0.13 0.17 0.09 0.12 0.10 0.10 0.08 TABLE I11 ABSOLUTE MOVEMENT OF FATTY ACIDS ON PAPER CHROMATOGRAMS CALCULATED FROM MOVEMENT OF SOLVENT FRONT AND RF VALUES GIVEN IN TABLE I Distance moved in f A > Fatty acid Spiteri method,6 Present method cm cm Capric acid .. .. .. 18.8 Lauric acid . . .. .. 18-0 Myristic acid .. .. 17.0 Palmitic acid .. .. 16.0 Stearic acid . . .. .. 13.2 7.7 6-0 4.3 2.7 1-7 Solvent front .. .. 20 13 Figs. 3 (c), ( d ) and (e) show the application of the method t o the separation of fatty-acid mixtures of unknown composition. Fig. 3 (e) demonstrates the resolution of the free-fatty- acid fraction of the acetone extract of a laundered pillowslip; Fig. 3 (c) is a chromatogram of the component fatty acids of a high-titre soap and Fig. 3 (d) that of low-titre soap, as employed at this laboratory. DISCUSSION OF RESULTS It is generally known that the diffusiveness of a zone on a chromatograni increases with the distance of migration of the zone.It therefore follows that short distances of migration will produce compact spots that are more suitable for quantitative evaluation. On the other hand, a certain minimum distance of migration is necessary for complete separation of the components, so that a compromise must be reached. In most methods of long-chain fatty-acid chromatography the authors specify a minimum distance through which the solvent front should move in order that the acids be completely resolved. For instance, Holasek and Winsauer4 give a minimum distance of 30 cm, Spiteri6 specifies 20 cm and Kobrle and Zahradnik5 employ a distance of 45 cm. In the present method it has been found that a movement of the solvent front of 13 cm has been sufficient for the separation of acids up to stearic acid.In fact, as reference to Table I1 will show, the degree of separation, (R,-z - R,J, is somewhat greater than that obtained by some of the other authors. ,4t the same time, the actual movement of the fatty-acid zones has been considerably less than that described by the other authors. As a comparison, the actual distances through whichJuly, 19561 CHANGE IN SOLVENT COMPOSITION. PART I 389 the zones have moved when Spiteri’s6 method is used and when the present method is used are shown in Table 111. By comparing the corresponding distances it can be seen that the change in solvent composition affects the chromatogram in such a manner that, for a smaller absolute distance of migration, better separation is obtained than has been recorded for the more conventional procedures.At the same time reference to Fig. 3 shows that spots are compact and that no tailing occurs. Although changing the solvent during a chromatographic run has been employed b e f ~ r e , ~ ~ ~ , ~ ~ ~ ~ this has always been confined to column chromatography. Solvent front OL 0 Lin OM 0 0 0 (4 (4 (4 (4 (4 Fig. 3. Chromatographic separation of fatty acids: (a) a mixture of oleic (01) and linoleic (Lin) acids; (b) a mixture of capric (C), lauric (L), myristic (M), palmitic (P) and stearic (St) acids; (c) component fatty acids of a high-titre soap; ( d ) com- ponent fatty acids of a low-titre soap; (e) the free-fatty-acid faction of the acetone extract of a laundered pillowslip As has been observed before, unsaturated acids interfere with the chromatogram, since double bonds reduce the effective chain length of the acid by two carbon atoms per double b ~ n d .~ , ~ Therefore, if a mixture containing saturated as well as unsaturated acids is t o be resolved, a chromatogram of the complete mixture is first developed. The acids are then oxidisedll and another chromatogram is obtained. By comparing the two chromatograms, the presence and nature of any unsaturated acids is determined. Kobrle and Zahradnik12 state that instead of oxidation, it is preferable to prepare dihalogen compounds by the addition of iodine bromide, since their RF values differ more from those of the unsaturated acids than do the RF values of dihydroxy acids. Fig. 3 (a) shows that oleic and linoleic acids, although they contain a chain having 18 carbon atoms, appear in different places on the chromatogram. APPLICATIONS OF THE METHOD The nature of the residual .fatty matter in laundered articles has for scme time been of interest in connection with studies carried out at this laboratory.Fig. 3 (e) shows that three fatty acids are present, one of them in very small quantities. The chromatogram shown was obtained from the extract of a pillowslip. Figs. 3 (c) and 3 (d) show the chromatograms of the constituent fatty acids in soap. The large amounts of stearic and palmitic acids present in the high-titre soap resulted in a rather poor separation. This could be improved either by allowing the spots to move390 FRANKS : PAPER CHROMATOGRAPHY WITH CONTINUOUS [Vol. 81 through a larger distance or by increasing the rate of concentration of the mobile solvent. The low-titre soap shows satisfactory separation into a large spot of oleic acid with smaller amounts of shorter-chain acids. I thank Mr. R. E. Wagg for helpful discussion and the Council of the British Launderers’ Research Association for permission to publish this paper. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. REFERENCES Spiteri, J., and Nunez, G., Compt. Rend., 1952, 2603. Duncan, R. E. B., and Porteous, J. W., Analyst, 1953, 78, 641. Savary, P., Bull. SOC. Chim. Biol., 1954, 36, 927. Holasek, A., and Winsauer, K., Monatsh., 1954, 85, 796. Kobrle, V., and Zahradnik, R., Chem. Listy, 1954, 48, 1189. Spiteri, J., Bull. SOC. Chim. Biol., 1954, 33, 1355. Williams, R. J. P., Analyst, 1952, 77, 905. Bush, H., Hulbert, R. B., and Potter, V. R., J . B i d . Chem., 1952, 196, 717. Donaldson, K. O., Tulane, V. J., and Marshall, L. M., Anal. Chem., 1952, 24, 185. Allen, R. R., and Eggenberger, D. N., Ibid., 1955, 27, 476. Savary, P., and Desnuelles, P., Bull. SOC. Chim. France, 1953, 939. Kobrle, V., and Zahradnik, R., Chem. Listy, 1954, 48, 1703. BRITISH LAUNDERERS’ RESEARCH ASSOCIATION THE LABORATORIES, HILLVIEW GARDENS LONDON, N.W.4 January loth, 1966

ISSN:0003-2654    

DOI:10.1039/AN9568100384

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

390 FRANKS : PAPER CHROMATOGRAPHY WITH CONTINUOUS [Vol. 81 Paper Chromatography with Continuous Change in Solvent Composition Part 11. Separation of Surface-active Agents BY F. FRANKS* (Presented at the meeting of the Society 0% Wednesday, A $ d 4th, 1956) The method and apparatus described in Part I are used to separate mixtures of surface-active agents and commercial detergents. Procedures are described for the resolution of anionic and cationic compositions. It has not been possible to identify the resolved components of commercial deter- gents, since pure reference compounds are not readily available. A REVIEW of the literature shows that chromatography has not yet been applied to the analysis of surface-active compounds, although the determination of traces of detergents has been gradually becoming an important problem in several industries. Further, there does not exist a great deal of information about the exact composition of commercial deter- gents.Blandin and Desalmel have used the migration of dye complexes formed by anionic compounds to locate the presence of certain detergents in mixtures, and Gallo2 has investigated the effect of various solvents on the rate of migration of non-ionic detergents on paper. It is well known that most commercial detergents consist of a mixture of several surface- active species differing mainly in the number and arrangement of carbon atoms in the alkyl group. Holness and Stone3 and the authofl have described chromatographic procedures whereby long-chain alkyl sulphates can be separated.Both methods suffer from a drawback generally encountered with long-chain members of a homologous series, namely that no single solvent will separate all the components. Working with vt-alkyl sulphates with chains having 12 to 18 carbon atoms, I found that, when suitable R F values were obtained for the three lower members, then the octadecyl sulphate did not migrate at all; on the other hand, if the mobile solvent was adjusted so that a reasonable RF value was obtained for octadecyl * Present address : South East Essex Technical College, Dagenham, Essex.July, 19561 CHANGE IN SOLVENT COMPOSITION. PART I1 39 1 sodium sulphate, then the dodecyl compound was washed off the paper by the solvent front. Similar difficulties were encountered in the attempted resolution of commercial detergent compositions.There is, of course, an additional problem in the selection of suitable solvent systems for the chromatography of detergents and similar compounds. These substances are scluble only in water and alcohols. It is therefore necessary t o employ two non-miscible solvent phases composed solely of water and alcohols. In order to achieve this the alcohol constituting the stationary phase must be a long-chain compound the length of whose hydrocarbon chain should be comparable with those of the compounds making up the detergent mixture to be analysed. But the mobile solvent composition must be such that it does not dissolve the stationary phase. As this is impossible to achieve, the mobile phase must always be saturated with the stationary phase, and this in turn requires a fairly constant temperature during the development of the chromatogram.The apparatus described in Part I has been employed for the resolution of detergent mixtures with great success. Not only has it been possible to separate mixtures containing known quantities of pure n-alkyl sodium sulphates, but commercial detergents such as Teepol, Santomerse No. 1 and Fixanol C have yielded chromatograms indicating separation into individual fractions. It has not yet been possible to identify these fractions, as pure reference compounds to be used for comparison are not readily available and are usually difficult to synthesise, especially the sec.-alkyl sulphates. This does not however detract from the usefulness of the method.EXPERIMENTAL The apparatus and theory underlying it have already been discussed in Part I (see p. 384). Initial experiments were made with four straight-chain compounds of known purity, namely lauryl, tetradecyl, cetyl and octadecyl sodium sulphates. The paper was impregnated with the stationary solvent by drawing it quickly through a 2 per cent. w/v solutution of cetyl alcohol in acetone. The initial mobile solvent was a 60 per cent. v/v aqueous solution of ethanol. The chromatogram was allowed to develop for 24 to 30 hours. It was found that the cetyl alcohol coating on the paper slowed down the rate of migration considerably. Great care had to be taken to saturate the initial solvent with cetyl alcohol at room temperature and to ensure that the temperature during the development of the chromatogram remained reasonably constant.As the mobile solvent was gradually diluted with water, stationary solvent was precipitated in the flask. In my experience this never led to any complications, as the precipitated long-chain alcohol was removed through the overflow tube. The main point was that by continuous dilution of the solvent in the flask, it was always saturated with respect to the long-chain alcohol forming the stationary phase. If this measure was neglected, it was generally found that the solvent front did not advance uniformly and that cetyl alcohol was dissolved off the paper as the mobile solvent advanced. After development the paper was dried at room temperature and immersed in an aqueous solution of cupric acetate for 1 minute.After being rinsed with water and placed in an oven at 100" C until completely dry, it was immersed in a solution of rhodamine B and finally washed with water several times. Comparatively large concen- trations of surface-active compounds could be seen as dark red spots on a pink background by daylight and as purple spots under ultra-violet light. Surface-active species that were present in moderate or low concentrations could be detected only under ultra-violet light as purple spots on a strongly fluorescing pinkish background. Depending on the water solubility of the mixture under iavestigation, the initial con- centration of the mobile phase was adjusted. When Teepol was investigated, it was found that the best initial solvent concentration was 50 per cent.v/v aqueous ethanol. Experiments were also carried out to examine the effects of altering the nature of the stationary phase. Decyl, lauryl and tetradecyl alcohols were used, but did not give the same degree of separation as could be obtained with cetyl alcohol. Experiments carried out with all<ylarylsulphonates showed that similar conditions to those appropriate for Teepol could be employed, whereas the only change in the experimental procedure necessary with cationic compounds was in the choice of a siiitable indicator. Bromcresol green was adopted, as it yielded satisfactory results. This was diluted with water at an approximate rate of 15ml per hour. It was then again dried at 100" C.392 FRANKS : PAPER CHROMATOGRAPHY WITH CONTINUOUS METHOD [Vol.81 REAGENTS- Cetyl alcohol-Reagent grade. Acetone-Reagent grade. Ethavtol-Analytical-reagent grade ethanol is heated under reflux for 8 hours with silver nitrate and potassium hydroxide. Cupric acetate-A 0.5 per cent. solution of the analytical-reagent grade material. This solution is prepared as required. Rhodamine B-A 0.1 per cent. solution in water. It is then fractionally distilled. PROCEDURE FOR IMPREGNATING THE PAPER- Cut strips of Whatman No. 1 filter-paper to shape as described in Part I (p. 384). Draw them through a 2 per cent. solution of cetyl alcohol in acetone and let them dry at room temperature. PROCEDURE FOR DEVELOPING THE CHROMATOGRAM- Apply to the paper a solution of the mixture containing up to 200 pg of alkyl sulphates. Assemble the apparatus as described in Part I.Saturate the mobile solvent, in this case 50 per cent. aqueous ethanol, with cetyl alcohol by refluxing for 30 minutes with excess of cetyl alcohol. Allow the mixture to cool to room temperature and set it aside overnight. Finally filter off the excess of cetyl alcohol. Fill the flask with the mobile solvent and the separating funnel and constant-head solvent reservoir with water, taking care to remove air bubbles from the capillary tap and rubber tubing. Place the paper in position, and start the dilution mechanism and the magnetic stirrer. Adjust the dilution rate to approximately 20 ml per hour. Allow the chromatogram to develop until the solvent front has advanced a distance of 15 to 20 cm, or with circular-paper chromatography a distance of 10 to 15 cm.Stop tl-g dilution mechanism; remove the paper and allow it to dry at room temperature. Measure the overflow volume and record the time. PROCEDURE FOR COLOUR DEVELOPMENT- Immerse the paper in a freshly prepared 0.5 per cent. solution of cupric acetate for several minutes. Allow the paper to dry in an oven at 100" C. Subsequently immerse it in a 0-1 per cent. solution of rhodamine B in water for a few minutes. Discard the solution, and wash the paper with several portions of water until the washings possess only a faintly pink colouration. Remove the paper and allow it to dry in an oven at 100" C. When the paper is completely dry, examine it in ultra- violet light and mark with pencil the zones that appear as purple on a strongly fluorescing pink to orange background.RE s u LTS Chromatograms of Teepol developed by the above method show resolution into seven well defined zones having the following RF values: 0.36, 0.45, 0.56, 0.67, 0.76, 0.85 and 0.93. Similar chromatograms, developed by the method of S~hwerdtfeger,~ show partial resolution, but on changing the solvent composition continuously the zones become more clearly defined, and good results have been obtained when the solvent front migration was only 5.5 cm. Similar chromatograms have been prepared with n-alkyl sulphate mixtures, with Santomerse No. 1, an alkylarylsulphonate and with cationic compounds such as cetylpyridinium chloride in which te tradecyl and lauryl compounds could be detected. Discard the solution and wash the paper with water.DISCUSSION OF RESULTS The above results show that a resolution of detergents into pure compounds seems to be feasible. Anyway it is certain that different groups of compounds can be easily separated. For instance, in a mixture of alkylarylsulphonates and n-alkyl sulphates, if water alone is used as mobile solvent, the sulphonates will migrate, while the sulphates remain behind at the origin. Although most commercial detergents contain inorganic salts, sometimes in large con- centrations, they have not interfered with the chromatographic separation as described above.July, 19561 CHANGE IN SOLVENT COMPOSITION. PART I1 393 Several schemes of analysis for commercial detergents have been described,6 3 7 9 8 3 9 #lo all of them laborious and consisting of a very large number of tests that must be performed in order that any conclusions can be reached.A minor difficulty in all the schemes consists in separating the surface-active component by extraction. This separation could easily be carried out by adopting the chromatographic procedure outlined above. All inorganic substances are washed off the paper by the water in the solvent, whereas non-surface-active organic substances would be held at the origin by the stationary solvent. Only the surface- active substances would form the chromatogram and would be distributed on the paper in accordance with their h ydrocarbon-c hain lengths. It is also problable that chromatography of detergents might be usefully employed in quality control during production and might supersede titration and the more laborious methods of analysis.A quantitative evaluation of the coloured spots can probably be developed. Preliminary experiments have shown that on dry-ashing of the spots and subse- quent determination of the copper by an absorptiometric method involving use of dithizone an indication is given of the quantity of detergent present in the spot, When a mixture of surface-active species is involved, it is necessary to know the number of carbon atoms present in each component separable by chromatography before any quantitative evaluation can be made. For this purpose it is necessary to have pure compounds, so that reference mixtures with known components can be prepared and their rates of migration determined. I thank Mr. R. E. Wagg for helpful discussion and the Council of the British Launderers’ Research Association for permission to publish this paper. 1. 2 . 3. 4. 6. 7. 8. 9. 10. I ) . REFERENCES Blandin, S., and Desalme, R., Bull. Mens. ITERG, 1954, 8, 69. Gallo, U., Boll. Chim. Farm., 1953, 92, 332. Holness, H., and Stone, W. R., Nature, 1955, 176, 604. Franks, F., Nature, 1955, 176, 693. Schwerdtfeger, E., Naturwiss., 1953, 40, 201. Gilby, J. A., and Hodgson, H. W., Mfg Chem., 1950, 21, 371 and 423. Wurzschmitt, B., 2. anal. Chem., 1950, 130, 105. Van der Hoeve, J. A., Rec. Trav. Chim. Pay-Bas, 1948, 67, 649. Marconi, &I., Chimica, 1951, 6, 251. Van der Hoeve, J. A., J . SOC. Dyers 6 Col., 1954, 70, 145. BRITISH LAUNDERERS’ KESEARCH ASSOCIATION THE LABORATORIES, HILLVIEW GARDENS LONDON, N.W.4 January loth, 1956

ISSN:0003-2654    

DOI:10.1039/AN9568100390

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

394 NORRIS AND DARBRE: THE MICROBIOLOGICAL ASSAY OF INOSITOL [Vd. 81 The Microbiological Assay of Inositol with a Strain of Schizosaccharomyces pombe BY F. W. NORRIS AND A. DARBRE The microbiological assay of total and free inositol by a tube method is described. It is accurate and sensitive, and for routine purposes the calculation of results is simple. A modified design permits full statistical computation. Results for a number of food and other materials are given. THE method is essentially that proposed by Northam and N0rris.l As a result of further experiments, the medium has been slightly modified and minor points of procedure have been clarified. For routine purposes it is suggested in this paper that direct-reading methods, with or without logarithmic transformation, with occasional resort to fuller computations, would fulfil normal analytical requirements, METHOD Various methods of computing the results have been examined.The assay is capable of a high order of sensitivity and precision. The organism used is a strain of Schizosaccharomyces pombe, available from these labora- tories.* It is maintained by monthly (or more frequent) subculture for 20 hours at 25” C on malt - agar slopes. The medium used consists of 10 g of dried malt extract (Muntona Ltd.), 0.2 g of dried yeast extract (Norman Evans & Rais Ltd.) and 1.8 g of agar per 100 ml. The basal medium is that described by Northam and Norris,l with the omission of asparagine and a twofold increase in biotin content; its composition is shown in Table I. TABLE I BASAL MEDIUM Figures in brackets indicate concentration per ml of solution used Glucose .. .. .. . . .. 40.0 g Ammonium sulphate . . .. .. 4.0 g Potassium dihydrogen phosphate .. 3.0 g Potassium iodide . . - .. .. .. 0.2 mg (0.2 mg) Calcium D-pantothenate* . . .. I . 1.0 mg (1.0 mg) Nicotinic acid* . . .. . . .. 1.0 mg (1.0 mg) Pyridoxin hydrochloridet . . .. .. 1.0 mg (1.0 mg) Thiamine hydrochloride* . . .. .. 1.0 mg (1.0 mg) Trace-elements solution . . .. . . 2-0 ml Yeast-extract supplement . . .. . . 167.0ml Lactate buffer . . . . .. . . 20.0 ml Water to . . .. . . .. . . 1 0 0 0 ~ 0 d ’ Magnesium sulphate, MgS04.7H,0 . . 1.0 g (0.1 g) Calcium chloride, CaC1,.6H,O . . .. 1-og (0.1 g ) D-Biotin . . . . . . .. .. 0.8 Pg (1.0 P.g) * Solution prepared monthly. t Solution prepared fortnightly, The solutions listed as ingredients of the basal medium in Table I are prepared as follow s- Trace-elements solution-The following are dissolved in water- Boric acid, H,BO, .... .. . . 0-1 g Manganese sulphate, MnS0,.4H20 . . . . 0-04 g Zinc sulphate, ZnS0,.7H20 . . .. . . 0-04 g Copper sulphate, CuSO4.5H20 . . .. . . 0.045 g Ferrous sulphate, FeS0,.7H20 . . .. . . 0.25 g Ammonium molybdate, (NH4),Mo,0,,.4H,0 . . 0.02 g * This organism appears in the list of the Centralbureau voor Schimmel-cultures, Yeast Division, That in the culture collection of the Brewing Industry Delft, Holland, whence it was originally obtained. Research Foundation, Nutfield, Surrey, appears to be identical with our culture.July, 19561 WITH A STRAIN OF Schizosaccharomyces pombe 396 The solution is diluted to 1 litre with water (Emery, McLeod and Robinson2). All ingredients in solution are kept under sulphur-free toluene in glass-stoppered bottles at 0" to 4" C.Yeast-extract supplement-An amount sufficient for 400 assay tubes is prepared by stirring 19.2 g of yeast extract (Norman Evans & Rais Ltd.) for 10 minutes with a mixture of 1520 ml of absolute ethanol and 80 ml of 2 N sodium hydroxide. The mixture is filtered through a Whatman No. 1 filter-paper, neutralised to pH 7.0 (bromo- thymol blue) with pure hydrochloric acid and centrifuged. The clear supernatant liquid is poured off and evaporated almost to dryness under diminished pressure. The deposit is taken up with distilled water, adjusted to pH 4.8 (bromocresol green), filtered through a Whatman No.44 filter-paper and made up to 200ml. After the supplement has been pasteurised for 10 minutes at 60" C, it is stored at 0" to 4" C. Lactate bufer-Concentrated :odium hydroxide solution is added to 250 ml of syrupy lactic acid until on diluting 1 to 100 a pH of 4.8 is reached. The solution is diluted to 500 ml and stored; it keeps indefinitely. PREPARATION OF EXTRACTS- Total inositol-In a study of the distribution of inositol, free and combined, in plant materials, Darbre and Norris3 determine inositol by heating portions of 100 to 200 mg with 2 ml of N hydrochloric acid in sealed Pyrex-glass tubes (150 mm x 16 mm) for 48 hours at 123" C. After being cooled, the tubes are cracked open and the contents filtered through a Whatman No.1 filter-paper into a suitable calibrated flask (usually 250 ml), and the black residue in the funnel is repeatedly washed with distilled water until the flask is about three- quarters full. All the assayable inositol has been washed into the flask when a drop of the washings is neutral to bromocresol green. The contents of the flask are brought to pH 4.8, bromocresol green being used as external indicatur. After being diluted t o volume, the sample is ready for assay. The colour of the solution is usually pale brown, but this has no detectable effect on the Spekker absorptiometer readings. If the solution is not required immediately, it is transferred to a conical flask, which is plugged with cotton-wool, heated in a water bath to about 90" C, allowed t o cool spontaneously and stored at 0" to 4" C.The amount of sodium chloride formed by neutralisation is below the level at which the organism is detectably inhibited. Moreover, the test solution must usually be considerably diluted t o give an inositol concentration of 1 pg per ml for assay purposes. Free inositol-A 1-g portion of the ground material is transferred from a small square of aluminium or nickel foil to a stainless-steel homogeniser (Potter and Elvehjem4; Albaum and UmbreiP). Then 10 ml of 0-04 N hydrochloric acid are added from a pipette to wash down any material adhering to the foil and to the sides of the homogeniser tube. After being homogenised for 2 minutes at about 2000 r.p.m., the mixture is transferred, together with many small washings, to a 50-ml centrifuge tube.This is set aside, with occasional stirring for 10 minutes, or overnight. It is then spun in a centrifuge and the supernatant liquid is transferred to a conical flask. The precipitate is washed, with stirring, four times at the centrifuge with 5-ml portions of 0.04 N hydrochloric acid ; three washings remove all detectable inositol, the fourth is a precaution. The extract and washings are adjusted to pH 4.8, bromocresol green being used as indicator, in the usual way. The flask is plugged with cotton-wool and the solution is autoclaved at 15 lb pressure momentarily. Often the amount of material precipitated on heating makes filtration very slow, and it is advisable to decant the supernantant liquid into a Whatman No. 1 filter-paper to be filtered directly into a calibrated flask (usually 100 ml).The remainder is spun in a centrifuge, and the supernatant liquid is added to the filter. The precipitate is washed four times at the centrifuge as previously described, and the filter-paper is washed again before making up to volume. The solution should be heated to 90" C, as previously indicated, if it is to be stored before assay. PREPARATION OF INOCULUM- The organisms from a 20-hour slope culture are transferred by means of a platinum loop to a mixture of 3 ml of sterile basal medium and 3 ml of distilled water in a plugged sterile tube. The inoculum density is such as to give a reading of 0.15 t o 0.25 in the Spekker absorptiometer against a medium blank. The tube may conveniently be held in the beam of light in the instrument by use of a wooden container such as that described by Xortham396 NORRIS AND DARBRE: THE MICROBIOLOGICAL ASSAY OF INOSITOL [VOl.81 and Norris.6 The density indicated corresponds to approximately 0.2 to 0.4 mg dry weight of yeast per 6ml. ASSAY PROCEDURE- Flat bottomed 4-inch tubes (Samco) are used for the assay, selected so that the internal diameters fall between the limits 1-66 and 1.72 cm. The standard and sample tubes are set up in triplicate, and the respective solutions are added from micro-burettes. The standard solution contains 1.0 pg of inositol per ml, and levels are set up at (0900, 0.75), 1.0, (1-25), 1.5, (1.75), 2.0 and (2.25) ml, the values in brackets being optional. The concentration of inositol in the sample solution should be about the same as that of the standard, and three levels at 1.0, 1.5 and 2.0 ml may be set up.As indicated later, this design can be modified. Basal medium is added at the rate of 3 ml per tube by means of an automatic pipette similar to that proposed by R i d ~ a r d . ~ The tubes, fitted with glass or aluminium (Oxoid) caps, are randomised and then sterilised in the autoclave until 10lb pressure is reached, when they are allowed to cool overnight. (Heating in a steam steriliser for 5 minutes has been found to be equally satis- factory.) Each tube is then inoculated with one drop of the prepared inoculum added by means of a Pasteur pipette. Incubation is carried out at 25" 3- 0.2" C for 72 hours in a water bath iitted with a stirrer. The turbidities are then read in l-cm cells against a medium blank, the contents of the tubes being swirled to ensure evenness of turbidity without inter- ference by air bubbles, as would occur with shaking.It is suggested that time may be saved, without loss of accuracy, if the absorptiometer drum is set a t 1.0 with zero reading on the galvanometer for the medium blank, and the assay tubes are read against this. This avoids two readings for each tube; if the medium blank at 1.0 is checked occasionally when the readings are being taken, accuracy is maintained. Water is then added to each tube to give a total volume of 3 ml. COMPUTATION OF RESULTS For illustrative purposes one example has been used throughout, namely, soya-bean flour that had been germinated for 10 days.BY DIRECT READING FROM CURVES- For routine purposes this method is likely to prove quite satisfactory. It is suggested that curves should be drawn on as large a scale as possible and that mechanical aids are desirable. The full results of an assay are shown in Table 11. TABLE I1 COMPLETE ASSAY RESULTS Standard: 1 ml of solution = 1 pg of inositol Test sample: heated in a sealed tube for 48 hours at 123" C with N hydrochloric acid; 0.1314 g is made up to 800 ml of sample extract Absorptiometer readings* Dose, ml -7 Mean Standard- 0.00 0.75 1.00 1-25 1-50 1.75 2.00 2-25 1.20 1.80 2.40 Test- 999 800 638 524 389 3 60 249 182 655 437 277 994 795 654 525 412 337 264 197 660 424 270 998 790 648 512 447 338 251 189 673 445 270 997 795 647 520 416 345 255 189 663 435 272 * Spekker absorptiometer zero is set a t 1.0, The readings are multiplied by 1000.July, 19561 WITH A STRAIN OF Schizosaccharomyces pornbe 397 The plot of mean absorptiometer readings against dose is shown in Fig.1. By drawing horizontal lines from test curve to standard curve at the three test points, the relative potency of the test may be deduced- Test level, ml . . .. .. , . 1.2 1.8 2.4 Standard equivalent, ml . . . . . . 0.975 1.455 1.950 Inositol content, pg per g . . .. .. 4960 4920 4950 The mean of these results is 4940 pg per g, and the error of the mean is & 15, or 0.3 per cent. LOGARITHMIC TRANSFORMATION- the logarithm of the dose and the Spekker absorptiometer reading. Preliminary trials with the results suggested that there was a linear relation between Computation can then 0 0 .5 1.0 1.5 2 - 0 2-5 Dose, ml Fig. 1. Response curves: A, standard; B, test sample (from results in Table 11) follow lines similar to those suggested by Wood.* The linearity usually holds over the standard dose range of 1 to 2 ml. It is easier to draw straight lines than curved ones, and again for routine purposes the semi-log transformation has much t o recommend it. The lines for the results for soya-bean flour are shown in Fig. 2, and the transformed results in Table 111. TABLE I11 TRANSFORMED RESULTS FROM TABLE 11 {---L- 7 ---------- 7 -4bsorptiometer Absorptiometer Log dose, ml x 10 reading Log dose, ml x 10 reading Standard Test 1-0000 647 1.0792 663 1-0969 520 1.2553 435 1.1761 416 1.3802 272 1.2430 345 1-3010 255398 NORRIS AND DARBRE: THE MICROBIOLOGICAL ASSAY OF INOSITOL [Vol.81 The lines of Fig. 2 were originally drawn by eye in a rectangle of rnillimetre squared paper 40 cm x 25 cm, permitting accurate assessments to be made. The potency of the preparation is derived from the horizontal distance between the lines at any given value of absorptiometer reading, since this is equal to log V , - log V,, where V t and V s represent doses of test and standard, respectively, that give the same reponse. I I ‘ I I 1.1 1.2 1.3 1.4 Fig. 2.’ Response curves: A, standard; B, test sample (from results in Table 111) Log dose By reference to Table IV, the mean horizontal distance for three different values of visual reading is 0.0923. Thus- log ‘Vt - log V , = 0.0923 V 21 antilog 0.0923 = 1.237, V S whence- = 4920 (results of Table 11).800 0.1314 x 1.237 Content of inositol, pg per g = Instead of drawing the lines by inspection, results (Table IV) may more accurately be obtained after calculating the equations for the lines of best fit for the two curves. With assays similar to that shown in Fig. 2, the resulting lines do not differ appreciably from those drawn by eye and the results show close agreement. Normally, lines of best fit would not be calculated, since, if statistical methods are to be used, then a more correct method of dealing with the results is one making simultaneous use of standard and test preparations. TABLE IV RESULTS BY DIRECT READING AND BY CALCULATION Distance between Inositol content Inositol content r-lg Per g Absorptiometer reading lines (visual) by visual method, calculated, PLg Per g 640 0.0910 4940 4920 440 0-0925 4920 4950 280 0.0935 4910 4970 Means 0.0923 4920 4950 STATISTICAL VALIDITY AND FIDUCIAL LIMITS- Neither of the methods described above afford the worker any information as to the validity of the assay or the fiducial limits of the result.Since the dose-response curve is not linear (Fig. l), the method of multiple regression (Burn, Finney and Goodwing) was applied to the assay of Northam and N0rris.l This provides an accurate method of com- putation; but the “bracket” design described will be applicable to routine assays only ifJuly, 19561 WITH A STRAIN OF Schizosaccharomyces #ombe 399 the worker has a fairly precise knowledge of the concentration of inositol in the test solution, so that the test levels may lie within the brackets of the standard.As the relationship between log dose and response is linear, a less involved statistical computation is possible. In fact, advantage may be taken of simplified computations pro- posed by FinneylO and by Wood,ll if the experimental design is altered to a symmetrical (3 + 3) dose design when the doses are, for example, 1.00, 1-41 and 2-00 ml, the log doses then being in arithmetical progression. Such a procedure is strongly recommended when a statistical check on the assay is desired. DISCUSSION CHOICE OF YEAST EXTRACT IN MEDIUM SUPPLEMENT- Norman Evans & Rais yeast extract was normally used in the preparation of the yeast- extract supplement. On several occasions, however, Difco Bacto yeast extract was used.With the latter extract, the assay blanks were not consistent and were always very high. Never- theless, final results obtained with either extract were similar (see Table V). TABLE V RESULTS WITH DIFFERENT YEAST-EXTRACT SUPPLEMENTS Inositol content Inositol content found with Difco Bacto extract, p g per g on dry weight found with Norman Evans & Rais extract, pg per g on dry weight Test sample Barley (total inositol) .. .. 3260 3140 Malt (total inositol) . . .. . . 2990 2990 Oats (free inositol) . . .. . . 95 103 The high assay blanks obtained with media containing Difco yeast-extract supplement are to be attributed to the high free-inositol content of this extract compared with Norman Evans & Rais extract (see Table VII).I t is possible that other yeast extracts might give variable and high assay blanks, which, although theoretically undesirable , may well have little influence on the final result. PREPARATION OF SAMPLES- The choice of hydrolysis method, N hydrochloric acid in sealed tubes for 48 hours a t 123" C, was indicated as a result of many experiments discussed in more detail elsewhere (Darbre and Norris3) and governed by two prime considerations-maximum yield of inositol and minimum amount of acid. For routine purposes the use of the autoclave is obviously the simplest in operation, but experiments showed that short periods of autoclaving will not effect optimum release of inositol. Jones12 determined inositol in yeast by a cup plate method, Saccharomyces carlsbergensis being used as assay organism, and quotes results after autoclaving with hydrochloric acid of various concentrations for 1 and 2 hours.Our results (Table VI) show the superiority of the sealed-tube method, to an extent of about 30 per cent. TABLE VI TOTAL INOSITOL IN DRIED YEAST BY TWO METHODS OF HYDROLYSIS Inositol content, pg per g on dry weight Extraction method N HCl for 1 hour . . .. 2530 3N HC1 for 1 hour . . .. 2780 .. 2980 .. 2690 3N HCl for 2 hours . . .. 3210 5N HC1 for 2 hours . . .. 3200 f water for 72 hours . . .. 2220 Autoclave a t 15 lb pressure ~ N $ ~ ~ ~ ~ 2 1 ~ ~ ~ ; ; .. 4270 .. 4220 4N HC1 for 18 hours . . .. 4290 Sealed tube a t 123' C ~NH~~.o~~~~~~s : : The additional time necessary t o obtain sample extracts for assay with the sealed-tube Although this time may be method is offset by the improvement in the release of inositol.400 NORRIS AND DARBRE [Vol. 81 shortened from 48 hours (with N hydrochloric acid) to 18 hours (4N hydrochloric acidj, the higher concentration of acid is undesirable owing to the possible effect on the organism of the high concentration of sodium chloride consequently introduced into the sample extract.The method has been applied to a wide range of materials, of which a selection is included in Table VII. It gives satisfactory and reproducible results, is capable of considerable precision and is suitable for materials of both high and low potency. TABLE VII INOSITOL CONTENT OF VARIOUS MATERIALS Sample -4ustralian padi .. .. .. Australian padi, germinated 8 days .. Oats, sample A, germinated 2 days . . Oats, sample A, germinated 10 days Oats, sample B, germinated 2 days , . Oats, sample B, germinated 10 days Soya-bean meal, germinated 10 days Malt . . .. .. .. .. Malt extract . . .. .. .. Muntona .. .. .. .. Flour . . .. .. .. .. Soya-bean meal . . .. .. Barley . . .. .. .. * . Beans (Phaseolus vulgaris) . . .. Germ tailings . . . . . . .. Dried yeast . . .. .. .. Yeast extract (Norman Evans & Rais) Yeast extract (Difco) . . * . .. Brewers’ yeast . . .. . . .. Pancreatin . . .. .. .. Papain . . .. .. .. .. Pepsin, sample A . . .. .. Pepsin, sample B . . .. .. Takadiastase . . .. .. . . Free inositol, Total inositol, 50 2080 360 2110 86 2650 200 2400 57 2760 300 249 0 350 5980 420 4920 180 3230 100 2950 930 1890 520 1650 260 4950 140 850 2060 10,750 1970 4250 2470 2870 3880 3980 4470 1410 1450 1570 4350 1800 2150 - 3900 180 220 pg per g on dry weight pg per g on dry weight - We are grateful to Dr. M. Horwood, of Messrs. Norman Evans & Rais Ltd., for samples of dried yeast extract. We also express our gratitude to Dr. E. C. Wood for much helpful advice and constructive criticism on statistical aspects. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. Northam, B. E., and Norris, F. W., J . Gen. Microbiol., 1952, 7, 245. Emery, W. B., McLeod, N., and Robinson, F. A., Biochem. J., 1946, 40, 426. Darbre, A., and Norris, F. W., Ibid., 1956, in the press. Potter, V. R., and Elvehjem, C . A., J . Biol. Chem., 1936, 114, 495. Albaum, G. H,, and Umbreit, W. W., Amer. J . Botany, 1943, 30, 553. Northam, B. E., and Norris, F. W., J . Gen. Microbiol., 1951, 5, 502. Ridyard, H. N., Analyst, 1949, 74, 24. Wood, E. C., Ibid., 1947, 72, 84. Burn, J. H., Finney, D. J., and Goodwin, L. G., “Biological Standardization,” Oxford University Finney, D. J., “Statistical Method in Biological Assay,” Charles Griffin & Co. Ltd., London, 1952. Wood, E. C., Analyst, 1953, 78, 451. Jones, A., Ibid., 1951, 76, 588. Press, 1950. DEPARTMENT OF APPLIED BIOCHEMISTRY DEPARTMENT OF PHYSIOLOGY, BIOCHEMISTRY AND PHARMACOLOGY UNIVERSITY OF BIRMINGHAM UNIVERSITY OF LONDON, KING’S COLLEGE Movembev llth, 1955

ISSN:0003-2654    

DOI:10.1039/AN9568100394

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

July, 19561 MACDONALD AND REED 401 The Determination of Deuterium by the Mass-spectrometric Method BY A. MAcDONALD AND R. I. REED An attempt has been made to determine the deuterium content of some organic acids. Mass-spectrometric determination of the deuterium - hydrogen ratio has been employed with moderate success. The discrepancy between the results so obtained and those of a reference method is about 5 per cent. of the percentage abundance. IN the course of an investigation, an attempt was made to assay deuterium - hydrogen ratios by mass-spectrometric analysis of the isotope mixture. This method, except in the region of very high and very low deuterium contents, has been but little used. It has, however, the advantage that it can be applied to the analysis of very small quantities of material, 0.1 to 2.0 mg in the present study, particularly when the accuracy required is not high.Reviews of possible procedures, with special reference to precautions and particular methods, have already been given by Barnardl and Roth2; the general method of the latter worker has been adopted in the present work. EXPERIMENTAL The measurements have been made upon certain simple organic acids having deuterium atoms distributed about the alkyl or aryl nucleus. The values so obtained were compared with those from the gradient-tube m e t h ~ d , ~ in which the quantities of material used were about 100 to 150 mg. From the results, shown in Table I, it is seen that there is reasonable agreement between the two sets of values-usually better than 5 per cent.of the measured value. TABLE I COMPARISON OF RESULTS Weight Deuterium found by Deuterium found by Compound of sample, mass-spectrometer, gradient-tube method, mg atom per cent. atom per cent. Adipic acid .. . . 1 1.20 1.21 Benzoic acid ,. .. 2 10.44 11-13 Benzoic acid .. .. 2 9.59 10-02* Mandelic acid . . .. 0.1 8.79 8.39 Silver acetate . . . . 1 8.96 - 9.03 Silver acetate . . .. 1 1.18 1.20 Water . . .. .. 0.5 2.72 2-75? Water . . .. .. 0.2 5.94 594t Water . . .. . . 1 8.40 8-54t * This value was determined by first diluting the benzoic acid and then igniting it. t These samples were subjected to reduction only. The abundance ratio of masses 2 and 3 was determined at a series of different gas pressures by the method of Nier, Stevens and R ~ s t a d , ~ the value for mass 2 being used also as a measure of the total pressure.In the majority of samples examined a plot of the abundance ratio of mass 3 to mass 2 against mass 2 was in the form of a hyperbola (see Fig. l), in which the curvature of the graph was more marked towards lower gas pressures. This is in agreement with the general finding of these workers, even with samples having a very low deuterium content. The asymptote to the curve was constructed by standard method^,^ and the value for the ratio (3 to 2) at which the line intersected this axis was taken to be the true ratio in accordance with the theoretical treatment given by Roth.2 In two or three cases, the experimental results were used to deduce the mathematical equation to the curve, incidentally supporting the hypothesis that this was a hyperbola.The value given by the intersection of asymptote and axis was readily calculated from this equation. The value so obtained was, however, little better than that given by a careful geometrical402 MACDONALD AND REED: THE DETERMINATION OF [Vol. 81 construction of the asymptote. Notwithstanding that the algebraic procedure uses all the experimental values and is free from constructional error, it was considered that the change in value, which is often less than 0.2 per cent., did not warrant the extra labour of the algebraic method. It was assumed that the gases entering the ionisation chamber, which was operated at 200" C in these experiments, were in thermal and chemical equilibrium before ionisation occurred.H, + D, + 2HD was taken as K = 3.62, as determined by Urey and Rittenberg6 The results previously tabulated were obtaived from the corrected abundance ratio of mass 3 to mass 2 value by means of this value of K . Unfortunately the relative inaccuracy of this method prevents any test of this assumption, at least in the range of concentrations examined. In view of the strict precautions necessary, the method is both difficult and tedious. It must, moreover, be regarded as tentative, as in some experiments, particularly when the deuterium content was greater than about 10 per cent., discrepancies of as much as 30 per cent. were found. It does, however, provide a possible method of analysis for small quantities of material. METHOD At this temperature the value for the equilibrium constant for the reaction- 0.10- The oxidations are carried out with analytical-reagent grade potassium dichromate, previously dried over phosphorus pent oxide, in a hard-glass apparatus of the design shown in Fig.2. I I I 1 PROCEDURE- Dry the apparatus by flaming-out, Le., heating it to approximately 400" C and allowing it to cool under vacuum from an oil-pump, with a cold-trap interposed between the apparatus and the pump. When the apparatus is cold, admit air through a guard tube containing phosphorus pentoxide. Carry out the complete operation in two stages, the section of the apparatus from tube A to the breaker seal C being first treated. Enclose a pellet of uranium in a small copper jacket, which is necessary to protect the glass from the hot uranium, and dry this by heating and allowing to cool under high vacuum-the procedure described above for drying the glass apparatus.Introduce the uranium into the part of the apparatus already dried (tube E), as shown in Fig. 2. Finely powder some crystalline potassium dichromate and dry it over phosphorus pentoxide; use this as the oxidising agent. Mix a dry powdered sample of the appropriate acid with about a fivefold excess of the dichromate. Place this mixture over phosphorusJuly, 19561 DEUTERIUM BY THE MASS-SPECTROMETRIC METHOD 403 pentoxide in a desiccator for about 24 hours. Introduce the mixture into tube A of the apparatus, taking care to avoid putting the material in the upper part of the tube. Evacuate this section of the apparatus by means of an oil-pump protected by a cold-trap immersed in a liquid-air bath, and seal the inlet, B, while the apparatus is under high vacuum.Heat the mixture until it is just molten, that is to about 400” C, and maintain it at this temperature for 5 to 10 minutes. Allow the tube to cool to the temperature of the room and further cool it in an acetone - solid carbon dioxide mixture. Break the seal, C, by means of the magnetically operated breaker, G, which is introduced during the construction of the apparatus, and remove the oxygen and other gases by pumping, again using an oil-pump protected by a cold-trap. Seal the tube at D under high vacuum, and transfer the water from tube A to tube E by static-distillation, this procedure being conducted at a temperature of about 35” C during some 30 minutes.The water vapour is reduced by heating tube E at approximately 360” C during 14 hours, and the tube is then allowed to cool. Finally seal the tube at F. L E Fig. 2. Apparatus for oxidation: A, tube containing powdered potassium dichromate and the organic acid; B, D and F, constrictions for sealing the apparatus under vacuum; C, breaker seal; E, tube containing small pieces of uranium in copper jacket; G, metal breaker Immerse the tube in an acetone - solid carbon dioxide trap to freeze any moisture remaining after reduction. Introduce a sample of the hydrogen for analysis into the Metropolitan- Vickers M.S.2 mass spectrometer. Except when all the samples have nearly the same deuterium concentration, it is necessary to use part of the gas sample to flush out the spectrometer.It was found that with all the samples that we have examined there is sufficient gas for this and for subsequent measurement. Allow the initial sample of hydrogen to remain in the machine for at least 1 hour (overnight is preferable) and then remove it by pumping. Introduce the remainder of the hydrogen sample and, in accordance with general practice, measure the abundance ratio of mass 3 to mass 2 at a series of different gas pressures. We express our indebtedness to Mr. J. M. L. Cameron for his advice and for the dilution of certain samples on the micro-scale. REFERENCES 1. 2. 3. 4. 5. 6. Barnard, G. P., “Modern Mass Spectrometry,” Institute of Physics, London, 1953, p. 259. Roth, E., “Conference on Applied Mass Spectrometry,” Institute of Petroleum, London, October, Anfinsen, C., “Preparation and Measurement of Isotopic Tracers,” Edwards Brothers Inc., Ann Nier, A. O.,, Stevens, C. M., and Rustad, B., US. Atomic Energy Commission Report AECD-2767. Mukhopadhyay, A., “Geometry of Conics,” Macmillan & Co. Ltd., London, 1893, p. 160. Urey, H. C., and Rittenberg, D., J . Chew. Phys., 1933, 1, 140. 1953, Paper No. 7. Arbor, Michigan, 1946, p. 61. CHEMISTRY DEPARTMENT THE UNIVERSITY GLASGOW, W.l October 28th, 1955

ISSN:0003-2654    

DOI:10.1039/AN9568100401

出版商:RSC    

年代:1956

数据来源: RSC

摘要:

404 WEATHERLEY : CHROMATOGRAPHIC SEPARATION, [Vol. 81 Chromatographic Separation, Detection and Determination of Selenium BY E. G. WEATHERLEY The chromatographic behaviour of selenium in association with twenty other elements has been investigated. Procedures are described for the separation and determination by chromatography of small amounts of selenium in silicate, electrolytic-slime and flue-dust materials. The method has an accuracy comparable with that obtainable by conventional methods at concentrations between 0.1 and 1.0 per cent. of selenium and is more rapid. PUBLISHED methods1y2y3,4,5,6 for the detection7 p8 ,9 *lo and determination of selenium by ~hromatographic2~ and cation-e~change~~,~~ procedures are generally of limited application or are time-consuming. In the present world shortage of the metal owing to its increased use in the electrical, plastics and glass industries, a quicker method for its detection and determination in some processed and natural materials is becoming more necessary. The chromatographic behaviour of selenium in association with a number of elements has been investigated therefore, to ascertain whether a more rapid method could be evolved.Dietzel and Hirsch32 found that no loss of selenium, whether it was in the form of metal or salt, occurred during fusion with sodium carbonate, whereas CalcagnP and others found that metallic selenium and its salts were soluble in concentrated solutions of alkali hydroxides. In the investigations described in this paper, synthetic mixtures containing several combinations and different percentage compositions of elements present in silicate minerals, electrolytic slimes and flue dusts were used.The mixtures contained Al, Ag, As, Cd, Cu, Co, Cr, Ca, Fe, Hg, Mo, Mg, Na, Ni, Pb, Sb, Se, Si, Te, V and Zn, in the form of silicate, oxide, carbonate, nitrate, chloride, sulphate, sodium salt, selenide, selenite or selenate. Solu- tion was effected either by treating with a known volume of dilute mineral acid, with sodium carbonate or with sodium hydroxide solutions, with or without prior sintering with sodium carbonate. Aliquots from these solutions were spotted on -prepared paper strips, air-dried and chromatographed with various solvent mixtures ; upward diffusion of the solvent was employed. The method of preparing the paper to effect a number of simultaneous separations was essentially that of Hunt and Wells,= differing only in the size of paper used, which was 10 inches long by 6 inches deep.After solvent diffusion was complete, the strips were air-dried and then sprayed with various reagents to detect the positions of the elements present. The separation, detection and determination of selenium was found to be possible with solvent mixtures consisting of (a) an acetone - hydrobromic acid - ethanol system, (b) an acetone - hydrochloric acid - ethyl methyl ketone system, (c) an ethyl methyl ketone - hydro- bromic acid system, and (d) an ethyl methyl ketone - hydrofluoric acid system. gravimetric ,11,12,13,1* ,1546 titrimetric,l7 $18 919 ,20,21922,23,24 colorimetric,25 326 spectrophotometric,27,28 The lower limit of detection on a strip $ inch wide is 5 pg of selenium. EXPERIMENTAL The artificial mixtures used are given in Table I.PREPARATION OF SAMPLE SOLUTIONS- A &-g portion of silicate material was sintered in a small covered platinum dish with 1 g of sodium carbonate, the sintering or fusing being conducted rapidly. Exactly 5 ml of diluted hydrochloric acid (5 + 3) or diluted nitric acid (5 + 3) were added to each melt, disintegration being assisted by crushing. For flue dust and electrolytic slime four procedures were adopted, 0-5-g amounts, contained in small covered resistance-glass dishes, being treated with: (i) 5 ml of diluted hydrochloric acdi (5 + 3), (ii) 5 ml of diluted nitric acid (5 + 3), (iii) 5 ml of water and 1 g of sodium car- bonate, and (iv) 5ml of 40 per cent.sodium hydroxide solution. Solution of soluble material in the mixtures A, B and C was then effected by digestion of the covered dishes at 100" C, for 30 minutes.July, 19561 A1203 AgzO CdO CUO c o o CaO cr203 PbO SeO, Si02 TeO, ZnO Sb204 VZO, .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . . .. . . .. .. .. .. .. .. .. .. . . DETECTION AND DETERMINATION OF SELENIUM TABLE I COMPOSITION OF MIXTURES USED Oxide taken for silicate material (A), % 3.00 2-00 2-00 1.00 - - - 4.00 2-00 - - 4-00 9.43 1.00 0.57 69.00 1-00 1-00 - - Oxide taken for flue dust (B), / O 3-00 3.00 2.00 0 ' - - - I 8643 1 -00 - - 1.00 0.5.7 - - 1-00 1.00 1.00 405 Oxide taken for electrolytic slime (C), O f /O - 1.00 1.00 66.43 2.00 1.00 20.00 1.00 - - - - 1 -00 1-00 0.57 - 5.00 CHROMATOGRAPHIC SEPARATION- After digestion the prepared samples became covered by a thin layer of clear super- natant liquid.An aliquot of each sample (0.05 ml) was spotted on a datum line at one end of each strip by means of a calibrated glass capillary pipette, the pipette being drawn across the strip to produce a narrow rectangular wetted patch. The spotted papers were dried in an atmosphere of controlled humidity (44 per cent. relative ; over potassium carbonate slurry) to obviate fluctuations of paper moisture content influencing comparison of separations. Each sheet was spotted and dried just before use. TABLE I1 ORDER OF SPRAYING REAGENTS 1st Rubeanic acid-A 0.25 per cent.solution of the solid in ethanol. 2nd Diphenylcarbaxide-A 1 per cent. w/v solution in ethanol. 3rd m-DigaZZic acid-A 10 per cent. w/v aqueous solution. 4th Thiourea-A 3 per cent. w/v solution in N hydrochloric acid. 5th m-DigaZZic acid-A 10 per cent. w/v aqueous solution. 6th Aluminon-A 0-1 per cent. aluminon - 1 per cent. ammonium acetate solution in water. 7th Violuric acid-A 0.1 per cent. aqueous solution (for preparation, see Pollard and McOmie36). 8th Sodium hypophosphite-A 5 per cent. w/v in 2 N hydrochloric acid. 9th Stannous chloride-A 10 per cent. solution in diluted hydrochloric acid (4 + 1). The sheets, bent to form cylinders clipped a t the top by paper clips, were placed with the sample spots to the bottom in 25ml of solvent mixture contained in a plastic vessel fitted with an air-tight lid. The internal diameter and height36 of the tank were just sufficient to give the paper cylinder an all round clearance of 1 inch.The atmosphere of the system was always saturated with the most volatile organic constituent of the solvent mixture in use; 5ml of the particular constituent were used for this purpose just before running the chromatogram. All solvent mixtures were made up immediately before use, except when immiscible liquids requiring phase-saturation were encountered. For this saturation, a standard pro- cedure was adopted of shaking violently for 5 minutes, and then setting aside for 1 hour before separation of the mobile phase. Upward diffusion of the solvent for 33 inches from the datum line at a constant ambient temperature of 23" C was used for all separations.After resolution of the chromatogram and air-drying, the position of each element band was detected by spraying with an atomiser-type spray.406 WEATHERLEY : CHROMATOGRAPHIC SEPARATION, [Vol. 81 SPRAY REAGENTS AND IDENTIFICATION COLOURS- A multiple spray technique was adopted for the identification and confirmation of element positions on the chromatograms. The reagents used, order of spraying and colour reactions obtained are given in Tables I1 and 111. TABLE I11 COLOUR REACTIONS Reagents -41 Ag As Cu Co Cd Cr Ca Fe Hg Mg Mo Na Ni Pb Se Sb Te V Zn 1st + NH, - F - DGYBPYG- - - BB - - - B L Y - - - G G - 2nd - F - DGYBBM V - - - R C P B L S P - - - M H - - B L - 3rd+HAc - - - - - - - - BL - - RD - - - I O Y Y - - 4th - BL - 6th R D - - - - - - - - - - - - - - - - - - - 7 th - - - Y B - - G B - B L - - - v - - - - - - PR - - B B - I - - - - - I - - - - - - - - - O - B - - - - - B B B - - - - - - B - - - - - - _ _ _ _ _ _ _ - - - - - _ _ Sth+NH, - - - - - - - BG - - R - - - - - - 8th 9th F = fawn DG = dark green YB = yellow-brown PYG = pale yellow green BB = brown-black BL = blue Y = yellow GG = grey-green BM = bright mauve V = violet R = rose C = cerise P = pink SP = salmon pink M = mauve H = heliotrope RD = red 0 = orange BG = blue-green BY = brown-yellow BG = green-black PR = pale red B = black DISCUSSION OF SEPARATIONS ACHIEVED- Many conventional and unusual solvent mixtures were examined.Separation of the selenium at the solvent front as a tight compact band free from any interfering element was achieved with a number of solvent mixtures.The best separations were with the following solvent systems: (a) 80 ml of acetone - 10 ml of hydrobromic acid - 10 ml of ethanol, (b) 80 ml of acetone - 10 ml of hydrochloric acid - 10 ml of ethyl methyl ketone, (c) 90 ml of ethyl methyl ketone - 10 ml of hydrobromic acid, and (d) 80 ml of ethyl methyl ketone - 20ml of 40 per cent. w/v hydrofluoric acid. Fig. 1 illustrates the type of separation achieved. Solvent (d) gave excellent results for silicate material A after fusion with sodium car- bonate and for materials B and C after extraction with dilute hydrochloric acid. Solvents (a), (b), (c) and (d) gave excellent results with materials B and C after extraction with sodium carbonate or sodium hydroxide solutions.No really satisfactory clean compact separation of selenium could be achieved with silicate material A in nitric acid after fusion with sodium carbonate, or with materials B and C after extraction with nitric acid. The optimum condition for tight-banding of the selenium would appear to be for it t o travel on the chromatogram as a selenious acid complex. In acid solutions, this condition in the presence of some cations can only be partly achieved and diffuse banding results. By comparison of band width with pure selenious, selenic and selenious - selenic acid solutions on chromatograms resolved with solvents (a), (b), (c) and (d), the selenium complex was found predominantly in the selenious form with (b) and (d) and partly selenic with (a) and (c).By utilising the sodium carbonate extraction procedure, the majority of those elements preventing the formation of selenious - selenic complexes on the chromatograms are eliminated. With the sodium hydroxide solution method, which must be used if metallic selenium is thought to be present, solution of Al, As, Bi, Sn, Te and Zn will also occur; none of these prevents the formation of the desired complex. The procedures .adopted to effect solution of the artificial mixtures, together with the satisfactory separation and band formation attained with the four solvent systems given, were considered worthy of further investigation to see whether a quantitative determination by means of the now well known chromatogram comparison method were possible. APPLICATION OF METHOD TO EXPERIMENTAL MATERIALS For silicate material, amounts of selenate, selenite, selenide and selenium metal were added to a mixture of combined cobalt - nickel; chromium - iron and molybdenum - cadmium glasses to give a selenium content of 0-35 per cent.406 WEATHERLEY : CHROMATOGRAPHIC SEPARATION, [Vol.81 SPRAY REAGENTS AND IDENTIFICATION COLOURS- A multiple spray technique was adopted for the identification and confirmation of element positions on the chromatograms. The reagents used, order of spraying and colour reactions obtained are given in Tables I1 and 111. TABLE I11 COLOUR REACTIONS Reagents -41 Ag As Cu Co Cd Cr Ca Fe Hg Mg Mo Na Ni Pb Se Sb Te V Zn 1st + NH, - F - DGYBPYG- - - BB - - - B L Y - - - G G - 2nd - F - DGYBBM V - - - R C P B L S P - - - M H - - B L - 3rd+HAc - - - - - - - - BL - - RD - - - I O Y Y - - 4th - BL - 6th R D - - - - - - - - - - - - - - - - - - - 7 th - - - Y B - - G B - B L - - - v - - - - - - PR - - B B - I - - - - - I - - - - - - - - - O - B - - - - - B B B - - - - - - B - - - - - - _ _ _ _ _ _ _ - - - - - _ _ Sth+NH, - - - - - - - BG - - R - - - - - - 8th 9th F = fawn DG = dark green YB = yellow-brown PYG = pale yellow green BB = brown-black BL = blue Y = yellow GG = grey-green BM = bright mauve V = violet R = rose C = cerise P = pink SP = salmon pink M = mauve H = heliotrope RD = red 0 = orange BG = blue-green BY = brown-yellow BG = green-black PR = pale red B = black DISCUSSION OF SEPARATIONS ACHIEVED- Many conventional and unusual solvent mixtures were examined. Separation of the selenium at the solvent front as a tight compact band free from any interfering element was achieved with a number of solvent mixtures.The best separations were with the following solvent systems: (a) 80 ml of acetone - 10 ml of hydrobromic acid - 10 ml of ethanol, (b) 80 ml of acetone - 10 ml of hydrochloric acid - 10 ml of ethyl methyl ketone, (c) 90 ml of ethyl methyl ketone - 10 ml of hydrobromic acid, and (d) 80 ml of ethyl methyl ketone - 20ml of 40 per cent. w/v hydrofluoric acid. Fig. 1 illustrates the type of separation achieved. Solvent (d) gave excellent results for silicate material A after fusion with sodium car- bonate and for materials B and C after extraction with dilute hydrochloric acid. Solvents (a), (b), (c) and (d) gave excellent results with materials B and C after extraction with sodium carbonate or sodium hydroxide solutions.No really satisfactory clean compact separation of selenium could be achieved with silicate material A in nitric acid after fusion with sodium carbonate, or with materials B and C after extraction with nitric acid. The optimum condition for tight-banding of the selenium would appear to be for it t o travel on the chromatogram as a selenious acid complex. In acid solutions, this condition in the presence of some cations can only be partly achieved and diffuse banding results. By comparison of band width with pure selenious, selenic and selenious - selenic acid solutions on chromatograms resolved with solvents (a), (b), (c) and (d), the selenium complex was found predominantly in the selenious form with (b) and (d) and partly selenic with (a) and (c). By utilising the sodium carbonate extraction procedure, the majority of those elements preventing the formation of selenious - selenic complexes on the chromatograms are eliminated.With the sodium hydroxide solution method, which must be used if metallic selenium is thought to be present, solution of Al, As, Bi, Sn, Te and Zn will also occur; none of these prevents the formation of the desired complex. The procedures .adopted to effect solution of the artificial mixtures, together with the satisfactory separation and band formation attained with the four solvent systems given, were considered worthy of further investigation to see whether a quantitative determination by means of the now well known chromatogram comparison method were possible.APPLICATION OF METHOD TO EXPERIMENTAL MATERIALS For silicate material, amounts of selenate, selenite, selenide and selenium metal were added to a mixture of combined cobalt - nickel; chromium - iron and molybdenum - cadmium glasses to give a selenium content of 0-35 per cent.July, 19561 DETECTION AND DETERMINATION OF SELENIUM 407 For the investigation of electrolytic slime, selenite, selenide and selenate were added to an electrolytic anode slime containing Au, Ag, Co, Cu, Cr, Fe, Mn, Ni, Te and Zn to give a selenium content of 0.35 per cent. For flue dusts, metallic selenium, selenate, selenite and selenide to give a selenium content of 0.35 per cent. were added to a pyrites flue dust containing Fe, As, S, Zn, Te and V.The silicate material was fused with sodium carbonate, and soluble material was extracted with dilute hydrochloric acid ; an aliquot was spotted on a prepared paper strip and chromato- graphed, and the colour of the selenium band was compared with those of a series of chromato- grams containing standard amounts of selenium. For the electrolytic slime, extraction with sodium carbonate solution was used, a sodium hydroxide extraction being the procedure with the flue dust. Aliquots from both these materials were chromatographed and compared with standard selenium chromatograms in the same manner as for the silicate material. The results obtained are given in Table IV. TABLE IV RESULTS FOR ARTIFICIAL MATERIALS Mean, Selenium added, Selenium .found, % % % Silicate material .. . . 0.35 0.37, 0.37, 0.39 0-38 Electrolytic slime . . .. 0.35 0-32, 0-35, 0.32 0-33 Flue dust . . .. .. 0-35 0.34, 0.37, 0.37 0.36 The accuracy and reproducibility axe considered as good as those by conventional gravi- metric procedures, but the time involved is considerably less, being about 3 hours, only a small portion of which is required for actual manipulation. METHOD Analytical-reagent grade reagents and Whatman No. 1 filter-paper should be used in all procedures. PROCEDURE FOR PREPARING SELENIUM STANDARDS- Dissolve 0-2190 g of pure anhydrous sodium selenite in 10ml of distilled water, add 10 ml of concentrated hydrochloric acid and make up to 100-ml in a calibrated flask. Spot suitable aliquots to give amounts of selenium from 5pg to 45 pg on a prepared paper sheet, using Whatman No.1 filter-paper. Dry, and resolve the chromatogram by using a solvent consisting by volume of 80 parts of ethyl methyl ketone and 20 parts of 40 per cent. w/v hydrofluoric acid. Saturate the tank atmosphere with 5 ml of ethyl methyl ketone just before running the chromatogram. Dry the paper, spray it with thiourea solntion and store it between glass sheets in the dark. The standards (see Fig. 2) keep for at least 3 months before discolouring; a yellowing and darkening of the colour takes place after this time. PROCEDURE FOR SILICA MATERIAL- Fuse 0.5 g of the material with 1 g of anhydrous sodium carbonate in a small covered platinum dish. Add 5 ml of diluted hydrochloric acid (5 + 3) to disintegrate the melt, and digest on the water bath for 30 minutes to effect solution of the soluble material.Spot several 0.05-ml aliquots on a prepared paper sheet, dry it and resolve the chromatogram by using a solvent mixture consisting by volume of 80 parts of ethyl methyl ketone and 20 parts of 40 per cent, w/v hydrofluoric acid. Dry the paper, spray it with thiourea, dry it once more and compare the colour of the selenium band at the solvent front with those on the standard chromatogram, taking the mean value obtained from the several aliquots. PROCEDURE FOR ELECTROLYTIC SLIMES- Digest 0.5 g of the material with 5 ml of 20 per cent. aqueous sodium carbonate solution for 30 minutes on a water bath. Spot several 0.05-ml aliquots of the clear supernatant liquid obtained after digestion on a prepared paper sheet, dry, and resolve the chromatogram in a solvent mixture consisting by volume of 80 parts of ethyl methyl ketone and 20 parts of 40 per cent.w/v hydrofluoric acid. After drying the paper and spraying it, compare the colour of the selenium band at the solvent front with those on the standard chromatogram.July, 19561 DETECTION AND DETERMINATION OF SELENIUM 407 For the investigation of electrolytic slime, selenite, selenide and selenate were added to an electrolytic anode slime containing Au, Ag, Co, Cu, Cr, Fe, Mn, Ni, Te and Zn to give a selenium content of 0.35 per cent. For flue dusts, metallic selenium, selenate, selenite and selenide to give a selenium content of 0.35 per cent.were added to a pyrites flue dust containing Fe, As, S, Zn, Te and V. The silicate material was fused with sodium carbonate, and soluble material was extracted with dilute hydrochloric acid ; an aliquot was spotted on a prepared paper strip and chromato- graphed, and the colour of the selenium band was compared with those of a series of chromato- grams containing standard amounts of selenium. For the electrolytic slime, extraction with sodium carbonate solution was used, a sodium hydroxide extraction being the procedure with the flue dust. Aliquots from both these materials were chromatographed and compared with standard selenium chromatograms in the same manner as for the silicate material. The results obtained are given in Table IV. TABLE IV RESULTS FOR ARTIFICIAL MATERIALS Mean, Selenium added, Selenium .found, % % % Silicate material .. . . 0.35 0.37, 0.37, 0.39 0-38 Electrolytic slime . . .. 0.35 0-32, 0-35, 0.32 0-33 Flue dust . . .. .. 0-35 0.34, 0.37, 0.37 0.36 The accuracy and reproducibility axe considered as good as those by conventional gravi- metric procedures, but the time involved is considerably less, being about 3 hours, only a small portion of which is required for actual manipulation. METHOD Analytical-reagent grade reagents and Whatman No. 1 filter-paper should be used in all procedures. PROCEDURE FOR PREPARING SELENIUM STANDARDS- Dissolve 0-2190 g of pure anhydrous sodium selenite in 10ml of distilled water, add 10 ml of concentrated hydrochloric acid and make up to 100-ml in a calibrated flask.Spot suitable aliquots to give amounts of selenium from 5pg to 45 pg on a prepared paper sheet, using Whatman No. 1 filter-paper. Dry, and resolve the chromatogram by using a solvent consisting by volume of 80 parts of ethyl methyl ketone and 20 parts of 40 per cent. w/v hydrofluoric acid. Saturate the tank atmosphere with 5 ml of ethyl methyl ketone just before running the chromatogram. Dry the paper, spray it with thiourea solntion and store it between glass sheets in the dark. The standards (see Fig. 2) keep for at least 3 months before discolouring; a yellowing and darkening of the colour takes place after this time. PROCEDURE FOR SILICA MATERIAL- Fuse 0.5 g of the material with 1 g of anhydrous sodium carbonate in a small covered platinum dish. Add 5 ml of diluted hydrochloric acid (5 + 3) to disintegrate the melt, and digest on the water bath for 30 minutes to effect solution of the soluble material. Spot several 0.05-ml aliquots on a prepared paper sheet, dry it and resolve the chromatogram by using a solvent mixture consisting by volume of 80 parts of ethyl methyl ketone and 20 parts of 40 per cent, w/v hydrofluoric acid. Dry the paper, spray it with thiourea, dry it once more and compare the colour of the selenium band at the solvent front with those on the standard chromatogram, taking the mean value obtained from the several aliquots. PROCEDURE FOR ELECTROLYTIC SLIMES- Digest 0.5 g of the material with 5 ml of 20 per cent. aqueous sodium carbonate solution for 30 minutes on a water bath. Spot several 0.05-ml aliquots of the clear supernatant liquid obtained after digestion on a prepared paper sheet, dry, and resolve the chromatogram in a solvent mixture consisting by volume of 80 parts of ethyl methyl ketone and 20 parts of 40 per cent. w/v hydrofluoric acid. After drying the paper and spraying it, compare the colour of the selenium band at the solvent front with those on the standard chromatogram.

ISSN:0003-2654    

DOI:10.1039/AN9568100404

出版商:RSC    

年代:1956

数据来源: RSC