ISSN:0003-2654    

DOI:10.1039/AN95277FX033

出版商:RSC    

年代:1952

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN95277BX035

出版商:RSC    

年代:1952

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN95277FP101

出版商:RSC    

年代:1952

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN95277BP107

出版商:RSC    

年代:1952

数据来源: RSC

摘要:

SEPTEMBER, I952 THE ANALYST Vol. 77, No. 918 PROCEEDINGS OF THE SOCIETY OF PUBLIC ANALYSTS AND OTHER ANALYTICAL CHEMISTS DEATH Sir Jack Cecil Drummond. WE regret to record the death of The Determination of Small Quantities of Alginates in Rayon Finishes and on Yarn BY E. G. BROWN AND T . J. HAYES (Presented at the meeting of the Society on Wednesday, M a y 7th, 1952) A method is presented for the absorptiometric determination at small concentrations of ammonium di-ethanolamine alginate or ammonium tri- ethanolamine alginate in rayon-finishing liquids and on rayon yarn. The method is also applicable to sodium alginate solutions. It is based on the hydrolysis of alginic acid to furfural and subsequent reaction with Bial’s reagent to give a greenish-blue colour, the absorption of which is measured on a suitable instrument, such as the Spekker absorptiometer. Alginic acid at concentrations of up to 0.8 g per litre can be measured with satisfactory precision.A method for the extraction of alginates from rayon yarn is also described. Ir; the viscose rayon industry a solution of ammonium di-ethanolamine, or tri-ethanolamine, alginate is used in conjunction with a suitable finishing liquid to impart the requisite degree of stiffness to rayon cakes. The finishing agent is usually some type of lubricating oil, of paraffin origin, emulsified with a suitable agent, often a proprietary mixture of highly sulphated oils. The purpose of the paraffin is to provide lubrication for the yarn and simultaneous stiffening is given by the alginate compound. Ammonium di- or tri-ethanola- mine alginate is made by the dissolution of commercial alginic acid in stoicheiometric proportions of di- or tri-ethanolamine and concentrated ammonium hydroxide solution, two molecules’of alginic acid combining with one molecule each of ammonia and di- or tertiary base.The problems arose of (i) rapidly determining ammonium di- or tri-ethanol- amine alginate in finishing liquors for process control purposes where the finishing was part of a continuous process and (ii) determining alginate on rayon yarn after drying and conditioning. It was considered that a report of the alginic acid concentration would be adequate and, consequently, a method was developed by one of us (E.G.B.) involving precipitation of the acid by hydrochloric acid, filtration and alcohol washing of the pre- cipitate, and finally dissolution of the acid in calcium acetate solution and titration of the liberated acetic acid.l This method, although accurate, was tedious and unsuitable for process control work; it took about 3 hours to perform and required constant attention during filtration.It was inapplicable to alginate on yarn because of the small quantities involved, so a search was made for a more speedy method and, preferably, an absorptiometric one. 445446 BROWN AND HAYES: THE DETERMINATION OF SMALL QUANTITIES [VOl. 77 The literature on the detection and determination of alginic acid and alginates can be summarised under three main headings. First, methods depending on the hydrolysis of alginic acid by hydrochloric acid under controlled conditions to give carbon dioxide, which is suitably measured gravimetrically or volume tri~ally.~,~,* Secondly, application of the method of Ludtke5 and Yackel and Kenyon,G entailing the reaction of isolated alginic acid with calcium acetate solution to liberate acetic acid, which is suitably titrated.1,7,8,9 A modification of this procedure utilises the reaction of alginic acid with an excess of sodium hydroxide solution and titration of the excess with standard acid.1 Lastly, the colorimetric method of Ross and Perciva1,lo requiring carbazo1,e and concentrated sulphuric acid, which is based on the preliminary work of Egamill and I)ische.12 PRELIMINARY INVESTIGATIONS- It was originally considered that the carbazole method would be the most suitable for the purpose in hand, in spite of the fact that an examination of the calibration data of Ross and Percival shows that the points are only approximately linear. I t was found, however, that there was an interference effect i n the presence of the finishing liquid and colours either did not develop or had a marked brown tinge.A further disadvantage was that the method required intimate mixing by stirring and cooling of the solution in ice, which makes it not readily usable by unskilled operators. Mitchell, Shaw and Frary13 have detected sodium alginate in ice-cream by isolating silver alginate at a pH value of 4-6 by adding silver nitrate and subsequently hydrolysing the precipitate with hydrochloric acid to give furfuratl, which is qualitatively detected both by aniline and Bial’s reagent.This method presented itself as a possible alternative for quantitative use. Bial’s reagent,l4,l5 or its various modifications, has been extensively used since the original work on the subject, largely for pentoses, uronic acids, nucleosides and nucleotides. The reagent is composed of a solution of orcinol, hydrochloric acid and ferric alum or ferric chloride, the iron acting as a catalyst. When pentoses are heated with Bial’s reagent in a bath of boiling water they yield furfural by hydrolysis. The same reaction probably occurs with heptoses. The furfural then gives a stable bluish-green colour, by complex formation with the orcinol, that can be shaken into amyl alcohol and suitably measured. For example, Scheff16J7 used the method for a spectrophotometric determination of pentoses and glucuronic acid, whilst Fleury and Poirot18 utilised the reaction for the determination of furfural.Recently Nordal and Klevstrandlg adapted a modified reagent to the estimation of sedoheptulose. Many modifications of the composition of the reagent have been proposed, including the use of copper20s21 instead of iron for catalysis, but a critical evaluation of these reagents has been lacking until recently, when Miller, Golder and Miller,= on the basis of a systematic investi- gation, recommended the reagent proposed by DrUryS as the most favourable for the determination of pentoses. EXPERIMENTAL PREPARATION OF STANDARD SOLUTION- It is inconvenient to prepare a standard ammonium di- or tri-ethanolamine alginate solution for several reasons.Commercial di-ethanolamine and tri-ethanolamine are variable mixtures of the mono-, di- and tri- compounds, and it would be necessary to calculate the equivalent weight of the mixture and relate it to commercial alginic acid in this proportion. Secondly, it is difficult to add the calculated amount of concentrated ammonium hydroxide, sp. gr. 0.880, without loss; an excess would almost certainly be required. These effects are minimised under factory conditions by the large volumes used, but they might be undesirable for analytical purposes. Accordingly, it was decided to make up a standard solution of sodium alginate and to test the effect of ammonia and di- and tri-ethanolamine later in the method. Com- mercial alginic acid (70 to 80 per cent.) was analysed for purity by dissolution in standard sodium hydroxide solution and titration of the excess of alkali.A known amount of alginic acid was then dissolved in the calculated amount of standard sodium hydroxide solution to give a standard solution of sodium alginate The commercial acid was not purified because it was felt that as any impurities in it would be present in the finishing liquor, these should be retained in the standard solution to compensate for any effect they might have It was first necessary to prepare a standard alginate solution.Sept., 19521 OF ALGINATES IN RAYON FINISHES AND YARN 447 on colour development. A solution of finish was also made up, of the nominal strength of a factory solution, with the addition of a small amount of sodium bisulphite liquor to keep the pH range between 7 and 8.Suitable aliquots of sodium alginate solution were then placed in a series of standard flasks by means of a pipette and made up to volume with the finishing liquor. This procedure was necessary to simulate factory conditions and to see if the emulsified liquor had any effect on the colour development. COMPOSITION OF BIAL'S REAGENT- The composition of the reagent recommended by Nordallg was used for a few preliminary experiments but was relinquished subsequently in favour of the reagent given by DruryB and quoted by Miller et aZ.,22 with the exception that alcohol was not used for the solution of the orcinol. The reagent was made up in 100-ml portions and stored in dark bottles in a refrigerator, where it apparently remained stable over several months.If kept in the light at room temperature, some reduction from ironU1 to ironII may occur. The reagent gives a small constant blank when the reaction is carried out ; a blank determination should be made on each fresh bottle of reagent and preferably several times on the same bottle. The blank reading for 4 ml of reagent is in the range 0.05 to 0-08 of a drum reading, as measured on a Spekker absorptiometer. PERIOD OF BOILING- The boiling-time for the reaction of Bial's reagent with furfural is criticall6 t17; whilst some investigatorsl6 recommend keeping the test tube containing the reactants stoppered during the reaction, Nordal and Klevstrand19 have shown that it is unnecessary under their conditions to use a stopper if the boiling period does not exceed 20 minutes and we have confirmed this ; a 15-minute boiling period without stoppering gives reproducible results for alginic acid.U S E OF AMYL ALCOHOL AS SOLVENT- The addition of S ml of amyl alcohol to the mixture was standardised, this volume being suitable with the 1-cm cell of a Spekker absorptiometer. Scheff16 recommended swirling the mixture and not shaking, and used a correction factor to take into account the increase in volume of the amyl alcohol by solution of some of the acid layer. Nordal and Klevstrand19 have shown that the increase in volume, and hence the correction factor, can be avoided by adding more water and shaking vigorously; a similar procedure was used in our in- vestigations.MEASUREMENT AND STABILITY OF COLOUR- Preliminary tests showed that with a Spekker absorptiometer (Model H 560), Ilford spectrum red filters No. 608 gave the maximum drum reading in conjunction with the smallest blank reading. Previous investigati~nsl~ have shown that there is a maximum absorption peak for the system at about 620 mp. By the method detailed below, the analysis, as far as taking the extinction reading, can be comfortably completed in 45 minutes; accordingly this was standardised as the reaction time. There was some decrease in drum reading with time; decomposition was at a greater rate than that given by Nordal. There is a fairly rapid drop over the first 2 hours, then a more steady decrease in drum reading (see Table I). TABLE I EFFECT OF TIME ON THE DRUM READING Time, hours 0 0.5 1 1.5 2 4 4.5 5 6 24 Drum reading 0.805 0.798 0-795 0.788 0.786 0.785 0.782 0.782 0.781 0.758 NOTES-1.The solution was kept in the dark in a 1-cm glass cell covered with a microscope slide. 2. The initial reading, 0 hours, is 45 minutes after the start of the analysis. The reading is from both the alginic acid and the blank, and is equivalent to 120 pg of alginic acid. The colour showed little change of drum reading with change in temperature, a decrease of about 0.0013 of a drum reading per O C being obtained between the limits 13" and 24" C. Hence the effect of temperature change within normal working limits is negligible.448 BROWN AND HAYES: THE DETERMINATION OF SMALL QUANTITIES WOl. 77 METHOD OF PREPARING THE CALIBRATION CURVE The calibration curve prepared from a standard sodium alginate solution is shown in Fig, 1.APPARATUS- Use a 1-litre beaker and clamp the Pyrex 2.5 x 15-cm boiling-tube that contains the reactants 2.5 cm from the bottom. Let the initial height of boiling water be 11 cm or alternatively, use the constan t-level water-bat h of Miller.24 REAGENTS- All reagents should be of recognised analytical purity. Amyl alcohol. Bial’s reagent-Transfer 0.714 g of orcinol (3 :5-dihydroxytoluene) and 0.060 g of crushed ferric ammonium sulphate to a 250-ml beaker and dissolve it in about 75ml of I 00 I50 Alginic acid, ,ug/2ml Fig. 1. Calibration curve for alginic acid determination concentrated hydrochloric acid, sp. gr. 1.18. Transfer it to a 100-ml calibrated flask and make up to the mark with hydrochloric acid, sp.gr.1.18. Store in a dark tightly-stoppered bottle in a refrigerator. Sodiam aZginate sohtiort-Transfer about 01.5 g of commercial alginic acid, accurately weighed, to a 250-ml iodine flask and add 40 ml of 0.1 N sodium hydroxide solution. Swirl the flask, stopper it and set it aside for 4 hours. Titrate the excess of alkali with 0.1 N hydrochloric acid , using phenolphthalein as indicator. Calculate the percentage of alginic acid in the sample; 1 ml of 0.1 N sodium hydroxide solution is equivalent to 0.0176 g of alginic acid (C,H,O,) .26 Weigh out the appropriate amount of crude material so that 1 ml of the final solution is equivalent to 10 mg of alginic acid; dissolve it in the calculated volume of 0.1 N sodium hydroxide solution and make up to the requisite volume in a calibrated flask with distilled water.Dilute a suitable volume of this solution ten-fold with distilled water. 1 ml of solution = 1 mg of alginic acid. Transfer volumes of standard sodium alginate solution, covering the range 1 mg to 8 mg, to 100-ml flasks. Dilute each flask to the mark: with a solution containing about 0.4 g per litre of finishing liquid.Sept., 19521 OF ALGINATES IN RAYON FINISHES AND YARN 449 PROCEDURE- Transfer 2 ml of the dilute alginate solution to a boiling-tube and then add from a burette exactly 4 ml of Bial’s reagent. Place the tube in a l-litre beaker containing briskly boiling water maintained at the boil for exactly 15 minutes-timed by a stopwatch.The boiling should be vigorous and even. Remove the tube and cool for exactly 2 minutes in a stream of cold water. Add 8ml of amyl alcohol from a burette, swirl the resultant mixture and pour into a 100-ml separating funnel. Wash the tube with two separate 5-ml portions of distilled water, using a pipette. Add these washings to the separating funnel and shake vigorously for 10 seconds. Allow the two layers to separate, then run off the lower to waste. Dry the inside of the stem of the funnel with a roll of filter-paper and run off the coloured layer into a clean dry l-cm Spekker cell, filtering through a small funnel containing a No. 41 Whatman filter-paper to remove any water present. Stir the solution contained in the cell with a fine glass rod. Exactly 45 minutes after the start of the determination read the absorption of this solution on a Spekker absorptiometer, using amyl alcohol or water in the 1-cm reference cell, No.608 Ilford spectrum red filters and Calorex No. H 503 heat-resisting filters. A tungsten-filament lamp can be used as light source, and the direct method of measurement20 applied. Carry out a blank determination on each fresh batch of Bial’s reagent, or preferably several times on the same batch, by using 2 ml of finishing solution and 4 ml of Bial’s reagent and taking it through the same procedure. Subtract the blank reading from the observed reading and plot the drum reading against concentration to give the calibration curve. EFFECT OF DI-ETHANOLAMINE, TRI-ETHANOLAMINE AND AMMONIA ON THE PROPOSED The validity of the calibration graph prepared from sodium alginate for use with a liquid containing ammonium di- or tri-ethanolamine alginate was tested.For this purpose a factory solution containing di-ethanolamine alginate was analysed for alginic acid content PROCEDURE- TABLE I1 EFFECT OF DI-ETHANOLAMINE, TRI-ETHANOLAMINE AND AMMONIA ON THE PROPOSED METHOD 4 (4 4 (b) Solution series Ammonium di-ethanolamine alginate . . .. Ammonium di-ethanolamine alginate . . .. Ammonium di-ethanolamine alginate . . .. Ammonium tri-ethanolamine alginate . . 1 ml of 1 (a) + 1 ml of standard sodium alginate . . 1 ml of 2 (a) + 1 ml of standard sodium alginate . . 1 ml of 3 (a) + 1 ml of standard sodium alginate . . 1 ml of 4 (a) + 1 ml of standard sodium alginate . . solution containing 20 pg of alginic acid solution containing 60 pg of alginic acid solution containing 40 p g of alginic acid solution containing 40 pg of alginic acid ..Alginic acid & Added, Found, pg per 2 ml - 66 65 - pg per 2 ml 90, 91, 93 (Mean, 92) 91, 93, 93 (Mean, 92) 91, 93, 93 (Mean, 92) 106 107 - 86 83 - 85, 85 (Mean, 85) 83 81 by the proposed method; an initial ten-fold dilution was necessary. Satisfactory precision was attained (see Table 11). One millilitre of this solution was then added to 1 ml of standard sodium alginate solution and the resultant mixture was analysed by the proposed method. Satisfactory recovery of sodium alginate related to alginic acid was obtained (see Table 11). A solution of ammonium tri-ethanolamine alginate was also made up in the laboratory and treated similarly, with satisfactory results (see Table 11).The results show that the presence of tri-et hanolamine hydrochloride, di-et hanolamine hydrochloride and ammonium chloride (formed by the action of concentrated hydrochloric acid on the free bases, which are themselves liberated by hydrolysis) had no effect on the method, and a calibration curve prepared from sodium alginate solution is valid for solutions containing ammonium di-ethanolamine alginate or ammonium tri-ethanolamine alginate.450 BROWN AND HAYES: THE DETERMINATION OF SMALL QUANTITIES [Vd. 77 EFFECT OF DIFFERENT CONCENTRATIONS OF FINISH ON THE PROPOSED METHOD In factory conditions, the concentration of jinish in a finishing liquid is liable to alter within fairly wide limits, and this work was carried out to see whether or not different concentrations would affect the proposed method.During the heating with hydrochloric acid, the white emulsion initially present became coagulated to some extent and some hydrolysis of the sulphated oils present probably occurred. On extraction with amyl alcohol, however, all turbidity disappeared and two clear layers were formed. Three blank determinations were made (a) on pure water, (b) on a dilute finish solution containing about 0.4g per litre and (c) on a strong finish solution containing about 4.0 g per litre. The results, in Table 111, show no significant difference and prove that large variations in the total oily matter content of a finishing solution do not affect the method. TABLE 111 EFFECT OF DIFFERENT CONCENTRATIONS OF FINISH ON THE PROPOSED METHOD Solution used Water An approximately 0.4 g per litre finish An approximately 4.0 g per litre finish Drum reading 0.045 0-048 0.053 METHOD FOR THE DETERMINATION OF AMMONIUM DI- OR TRI-ETHANOLAMINE ALGINATE I N FINISH SOLlJTION The apparatus and reagents used are the same as those used in the calibration procedure (p.448). PROCEDURE- For concentrations of the order of 0.0 to 0.4 g per litre of alginic acid, dilute the finish solution ten-fold with distilled water. Take 2 ml of this solution and analyse according to the method detailed above. Calculate the result in terms of grams per litre of alginic acid. The factor for con- version of alginic acid to ammonium di-ethanolamine alginate (C,H,O,COONH, + C,H,O,COOH.NH(CH,CH,OH) ,) is 1.310 and for alginic acid to ammonium tri-ethanol- amine alginate (C,H,O4COONH, + C,H,O,COOH.N(CH,CH,OH),) is 1-472. RESULTS AND DISCUSSION- The calibration curve is drawn to the best fit as calculated by the method of least squares.On the basis of this calculation the line does not pass through the origin; this portion of the original curve is shown by a dotted line in Fig. 1. The extinction does not obey Beer’s law, but this is usual in this type of reaction owing to furfural not being liberated quantitatively, so the reaction conditions must be rigidly standardised. Even under strict control of experimental procedure, there is still some scattering of points for a given con- centration of alginic acid ; this has been experienced in similar reactions, e.g., by Nordal and Kle~strand.1~ It is considered, however, that the precision of the method on a normal solution of alginic acid of about 0.4 g of litre is about SfI 0.01 g per litre, as the scatter of points for a given concentration does not exceed this value.Table IV shows the precision of typical results as obtained on factory finishing liquors. TABLE IV PRECISION OF RESULTS FOR DETERMINATION O F ALGINIC ACID CONTENT OF FINISH SOLUTIONS Sample A 13 C Alginic acid, g per litre 0.45, 0.46, 0.47 0-46, 0.47, 0.47 0.43, 0.43 The whole analysis can be carried out in 45 minutes, which is a suitable period for process control work and much quicker than the: original gravimetric - volumetric method. Following the recommendations of Ayres2, and Ringbom,28 the percentage absorptionSept., 19521 OF ALGINATES IN RAYON FINISHES AND YARN 451 (calculated on the assumption that drum reading on the Spekker is equivalent to extinction) is plotted against the logarithm of concentration as abscissae (see Fig.2). Ayres2' states that the conclusions drawn in his article can be applied in a general way to any instrument. From a study of Fig. 2, it is seen that the most accurate range of the determination is 20 40 60 100 140 180 Alginic acid concentration, pg/2rnl (log scale) Fig. 2. Range and accuracy between 40 and 1OOpg per 2m1, corresponding to 0.2 to 0 6 g per litre of alginic acid. Ringbom28 has shown that the accuracy of a given procedure for this type of plot is greatest where the curve has its steepest slope.The curve is nearly linear within the limits 40 to 100 pg. A relative analysis error of 3.7 per cent. per 1 per cent. absolute photometric error is established by dividing 230 by the slope of the curve over the range 40 to 1OOpg of alginic acid. In the course of the development of the method, it became evident that with Bial's reagent for the visual detection of alginic acid, the lower limit of identification that allows distinction to be made from the reagent colour was about 10 pg per 2 ml. This corresponds to a concentration limit of 1 in 200,000. APPLICATION OF THE METHOD TO THE DETERMINATION OF THE ALGINATE CONCENTRATION EXPERIMENTAL WORK- The exact state of combination of alginic acid on rayon yarn after it has been dried and conditioned is somewhat doubtful; it was considered, however, that it would be possible to extract the material with boiling water.It would be a difficult matter to prepare a sample of rayon yarn with a known weight of alginate thereon, because of difficulties of drying and decomposition, and so samples with unknown concentrations were used. Preliminary experi- ments with hot-water treatment showed that alginate was extracted but incompletely in a beaker and by decanting off successive extractions, but with a Soxhlet apparatus, extraction of about 1 g of yarn was complete in 2 to 3 hours, as shown by a second 1-hour extraction and subsequent testing with Bial's reagent. A smaller sample can be taken in semi-micro work. During the treatment with boiling water, some finishing material was also extracted, but this was not detrimental to the determination.Extraction of yarn that had had no finishing or alginate treatment and subsequent treatment of an aliquot of the extract with Bial's reagent showed no matter to be extracted that would interfere in the method. ON RAYON YARN452 BROWN AND HAYES: THE DETERMINATION OF SMALL QUANTITIES [VOl. 77 APPARATUS- Otherwise the apparatus is as detailed above. RE AGENTS- As detailed above (p. 448). PROCEDURE- Weigh out accurately 0.1 to 0.15 g of rayon yarn and transfer it to the extraction chamber of the Soxhlet apparatus, using a suitable thimble to retain the yarn. Extract with about 20 ml of distilled water for 2 to 3 hours. Cool the flask and make up to volume in a 25-ml calibrated flask.Take 2 ml of the solution and proceed as in the proposed method. Simultaneously with the initial weighing, take a suitable sample of yam for moisture deter- mination; determine the moisture by drying to constant weight at 100" to 105" C. Calculate the result in terms of percentage of alginic acid on dry cellulose. A semi-micro Soxhlet apparatus of capacity 20 ml with a 30-ml flask is required. RESULTS AND DISCUSSION- finishing solution. are shown in Table V. The precision attained by taking aliquots of a given extract is the same as that on a Some results obtained on different sections of a viscose rayon cake TABLE V DISTRIBUTION OF ALGINATE ON VISCOSE RAYON CAKE Sample % Alginic acid, 1 0.67, 0-57 2 0.72, 0.79 Hence it is evident that the alginate is distributed unevenly over a given rayon cake, but no study has been made as to the possible relationship between concentration and distribution. Whilst it is impossible to verify that extraction is quantitative by using a known amount of alginate, the negative test given by Bial's reagent on a second extraction after initial boiling for 2 to 3 hours with water makes it extremely probable that all alginate is extracted.Thanks are due to Mr. P. Morley for help with the preparation of the calibration curve by the method of least squares, and to Mr. E. Stone, Research and Development Department Manager, British Enka Limited, and to the Directors of British Enka Limited for permission to publish this paper. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 16. 16. 17. 18. 19.20. 21. REFERENCES Cameron, M. C., Ross, A. G., and Percival, E. G. V., J . SOC. Chem. Ivtd., 1948, 67, 161. McCready, R. M., Swenson, H. A., and Maclay, W. D., Ind. Eng. Chem., Anal. Ed., 1946, 18, 290. Burkhart, B., Baur, L., and Link, K. P., J . Bial. Chem., 1934, 104, 171. Taylor, E. W., Fowler, W. F., Jun., McGee, 1'. A., and Kenyon, W. O., J . Amer. Cheun. Soc., Liidtke, M., Biochem. Z . , 1934, 268, 372. Yackel, E. C., and Kenyon, W. O., J . Amer. Chem. SOC., 1942, 64, 121. McGee, P. A., Fowler, W. F., Jun., and Kenyon, W. O., Ibid., 1947, 69, 347. Davidson, G. F., and Nevell, T. P., J . Text. Inst., 1948, 39, T102. Nevell, T. P., Ibid., 1948, 39, T118. Ross, A. G., and Percival, E. G. V., J . SOC. Chem. Ind., 1948, 67, 420. Egami, H., J. Chem. SOC., Japan, 1941, 62, 2'17.Dische, Z., J . Biol:Chem., 1947, 167, 189. Mitchell, D., Shaw, E. H., Jun., and Frary, G. C., J . Ass. Ofl. Agric. Chem., 1945, 28, 97. Bial, M., Dtsch. med. Wochenschr., 1902, 28, 253; Chem. Zentr., 1902, 11, 295. -, Dlsch wed. Wochenschr., 1903, 29, 477; Chem. Zenlr., 1903, 11, 1021. Scheff, G., Biochem. Z . , 1924, 147, 90, 94. -, Ibid., 1927, 183, 341. FIeury, P., and Poirot, G., J . Phavm. Chim., 1922, 26, 87. Nordal, A., and Klevstrand, R., Anal. Chim. rlcta., 1950, 4, 411. Jarrige, P., Bull. SOC. Chim. Biol., Paris, 1951, 32, 1038. Jarrige, P., and Henry, R., Ibid., 1950, 32, 1053. 1947, 69, 342.Sept., 19521 OF ALGINATES IN RAYON FINISHES AND YARN 463 22. 23. 24. 25. 26. 27. 28. Miller, G. L., Golder, R. H., and Miller, E. E., Anal.Chem., 1951, 23, 903. Drury, H. F., Arch. Biochem., 1946, 11, 269. Miller, G. L., J . Lab. Clin. Med., 1950, 36, 325. Alexander, J., “Colloid Chemistry,” Vol. VI, Reinhold Publishing Corporation, New York, 1944, “Instructions for Use of the Spekker Photoelectric Absorptiometer,” Adam Hilger Ltd., London. Ayres, G. H., Anal. Chem., 1949, 21, 652. Ringbom, A., 2. anal. Chem., 1939, 115, 332. p. 675. RESEARCH AND DEVELOPMENT DEPARTMENT BRITISH ENKA LIMITED AINTREE LIVERPOOL, 9 DISCUSSION DR. J. HASLAM congratulated the authors on the excellent way in which they had presented their paper. He drew attention to the fact that the determination of alginates was becoming increasingly important in many industries and for that reason he asked what the authors’ experience of interfering substances had been, zliz., with spin finishes in day-to-day work. He understood that the constituents of the spin finish used by the authors did not interfere in the test, and he would be pleased if the authors could give more information about the general chemical constitution of such finishes.MR. BROWN replied that the finish used in the proposed method consisted of commercial liquid paraffin as a lubricating agent emulsified with a proprietary mixture of highly sulphated oils. As the emulsifying agent was a proprietary brand he was not himself sure as to its exact chemical composition, but he under- stood there were four sulphated oils present. It would be necessary to test each individual finish to see if there were any interference with the method as the authors had not tested the effect of other spin finishes. DR. J. H. HAMENCE said that the determination of alginates present in small quantities in foodstuffs was always a difficult operation, but he imagined that the method described by the author would not be applicable to foodstuffs containing any carbohydrate that was likely to give rise to the liberation of furfural on treatment with acid. In spite of this difficulty, the method obviously would be of great assistance to food chemists in the determination of alginates in those instances where it was possible to separate the other carbohydrate materials. He pointed out that a possible method of removing interfering substances might be by isolating alginic acid with acid or as silver alginate with silver nitrate, according to the method of Mitchell, Shaw and FrarylS for ice-cream. Pectic acid and uronic acids would also interfere with the determination. MR. BROWN agreed. DR. H. AMPHLETT WILLIAMS asked if cellulose esters interfered With the determination. MR. BROWN said he had no specific information on this point, but did not think that cellulose esters would interfere. MR. N. L. ALLPORT asked how the authors ensured that the alginic acid used in the preparation of standards was 100 per cent. pure. MR. BROWN pointed out that before preparation of the standards the crude sample of alginic acid was analysed for purity and the necessary weight of crude material then taken to give the equivalent of 100 per cent. of alginic acid. Cellulose itself did not interfere in the method.

ISSN:0003-2654    

DOI:10.1039/AN9527700445

出版商:RSC    

年代:1952

数据来源: RSC

摘要:

454 FOG TURBIDITY I N PHOTOMETRY [Vol. 77 Turbidity in Photometry* Correction for Turbidity in Photometric Methods BY JORGEN FOGt The interfering effects of turbidity in photometric studies are overcome by a technique suitable for use in various photometric determinations, The spectral absorption curves of various polydispersed sols and sus- pensions show that extinctions are closely related logarithmically to the wavelengths. Extinctions read at two different wavelengths will then suffice to allow the extinction at a third wavelength to be determined. If the absorption curve of a clear coloured solution, which shows two troughs (El and E,) and one peak (E,) at the wavelengths A,, A3 and A,, respectively, is plotted logarithmically, and a straight line through El and E, crosses A, at AE2, then the difference (E, - AE,) is linearly related to concentration, provided that Beer’s law is valid.When a turbid solution is added, the difference is still closely related to concentration. The error is either negligible or calculable if (i) the difference approaches E,, (ii) E, > El when A1 > A3, and (iii) E, and E, are chosen at the minima of the absorption curves with foreign coloured substances. Spectral absorption curves and diagrams demonstrate how the effect of turbidity is eliminated when an artificial turbidity is produced in coloured solutions, even if the degree of turbidity and the amount of the coloured sub- stances are varied within wide limits. The equation fits well for protein precipitates and dispersed fat, which are the most common causes of turbidity in biological samples.THE accuracy attained in photometry with turbid samples is strictly limited, especially at the shorter wavelengths, and even if the turbidity is almost invisible to the eye. Morton and Stubbsl propose a device for correcting absorption spectra of vitamin A in fish-liver oils, and Lowry and Hastings2 propose one for determining the quantity of blood retained in animal organs. They assume that within a rather narrow wavelength band the extinction (optical density) is linearly related to the wavelength. This assumption limits the general applicability of their methods, however, because with turbid solutions extinctions and wavelengths are related logarithmically. Zerban, Sattler and Lorge3 in extensive studies on turbid systems containing coloured matter describe a photometric method for characterising both the absorbing and scattering properties of sugar solutions, but their method is confined to that one particular system only.Snell and Snell4 consider the use of an integrating-sphere photometefl to be the only satisfactory technique for turbid solutions. Fogs shows how the error caused by turbidity can be approximately eliminated in photometric determinations of the icterus index, i.e., the yellow colour in blood serum. This report describes how the effect of turbidity is adequately overcome by a technique well fitted to modification to suit various photometric determinations. THEORETICAL CONSIDERATIONS Although the fraction of light that is absorbed in travelling through a colourless mono- dispersed sol or suspension can be calculated for an ideal turbid medium,’ $8 $9 the theoretical treatment only applies when the particles are spherical and of uniform size and when the wave- length of the light is confined to narrow limits.These conditions do not apply to the turbid solutions likely to be met with in problems of practical absorptiometry. A number of experiments with non-coloured turbid samples shows the relationship between extinction and wavelength to be, as shown in Fig. 1, a law of the form E = kM, where k and y are constants. Extinctions at two different wavelengths then enable the constants to be eliminated and the extinction to be determined at a third wavelength, e.g., the value can be found graphically by plotting log E and log X (see Fig.2) assuming that El’, E,’ and E,’ satisfy the equation E = kM. * Read before the Seventh Scandinavian Physiological Congress in Arhus, Denmark, 1961. t Present Address: Oslo City Hospitals, Ullev&l sykehus, Department VII, Oslo.Sept., 19521 ' 002- FOG : TURBIDITY IN PHOTOMETRY Iw---=- - - - 0.02 I-* : = z = = . I I I I . . , I . . I 1 4 $ 6 7 8 9 1 . 0 Filters 1 . 2, I I . :1,00 -0.80 - 0.60 . m - 0.40 $ n . 0 3 -0.20g x 0.' -0.102 - 0.08 - 0.06 -0.04 1-00 0 8 0 C - Y .- 2 0.60 U UJ 0.40 0.20 wavelength Filters 4 5 6 7 8 9 10 I 455 Fig. 2. Cobaltous nitrate. Extinctions read with a Zeiss Pulfrich The filters Nos. 4, 5, 6, 7, 8, 9 and 10 correspond to (On the left are photometer. s61, s57, s55, s53, s50, s47 and s43, respectively.linear and on the right logarithmic co-ordinates)456 FOG TURBIDITY IN PHOTOMETRY [Vol. 77 In Fig. 2 the lower curve with El, E, and E, at A1, A, and A3, respectively, represents the absorption free from turbidity and the values are therefore identical with those of the pure substance at that concentration. If El, E, and E, are plotted against X on logarithmic scales the straight line joining the two points El and E, gives a value of AE, at A,. E2- AE, is proportional to concentration if Beer's law is valid; if not, a calibration curve can be used (see Figs. 3 and 4). If the solution is turbid, the observed curve E" will be the sum of the E and the turbidity E'. Wavelength, mp (log scale) Dilution ' Fig. 3. Alkaline glycerol - copper sulphate.Extinctions read at the wavelength 625 mp and corrected by means of extinctions read a t 900 and 440 mp E corrected with clear solutions 0 0 a 0 E corrected with turbid solutions o o o o E at 625 m p with turbid solutions If AE," is calculated (assuming that El", AE," and E," satisfy the equation E = khy), then Egl' - AE;' + x = E, - AE, because E;' = E, + E,' and AE," = AE2 + E,' + x. Here x represents an over-correction because the sum of values obtained from the two expressions E = kAy and E' = B'AY' do not exactly follow the expression E" = k"Ay". The quantity x will be small (i) if AE, is < E,, (ii) if E, > El where A, > A3 and (iii) if A1 and A, are at absorption minima on the curve for any selectively absorbing impurity. If x should exceed the experimental error of the measurement of E it can be calculated by successive approximations.Thus- El = (E2" - AE;') (E," 2;E2O) E3O and E, = (E," - AE,") (El0 -A,,.> to a first approximation. The quantities designated E" refer to the absorption curve measured with a non-turbid solution of the pure substances. Approximate values for E,' and E i are given by- El' = El,',' - El (approximately). E,' = E, - E, (approximately). Approximate values of AE2 and E,' can then be obtained graphically, so that z = AE," - AE, - E,' as a first approximation. With this value of x a second approximation can be reached and so on until x is less than the experimental error of E. EXPERIMENTAL The experiments were performed with a Beckman spectrophotometer model B and a In the series with coloured substances, certain Zeiss Pulfrich visual filter photometer.Sept., 19521 FOG : TURBIDITY IN PHOTOMETRY 457 amounts of stock solutions were diluted to the mark in a graduated 10-ml cylinder. The same pipettes and cylinders were used for each member of the series. Fig. 1 shows how logarithmically expressed extinctions and wavelengths are nearly linearly related in oil-dispersed-in-water emulsions , mastix-sols, precipitated proteins and 0 5 '3 0.10- .- 0.08: 2 0.06- 0.04 - W - ? 0.20 v 0'401 Wavelength, mp (log scale) 1 1 0 . 2 0 4 5 6 7 . 8 9 Concentration, mg per 100 mI 1.10 t 1.00 -0.90 .0.80 -0.70 -0.60 -0.50 -0.40 -0.30 Fig. 4. Sodium salicylate ( + ferric chloride). Extinctions read with the wavelength 525 mp corrected by means of extinctions read a t 1000 and 390 mp E corrected with clear solutions 0 0 0 0 E corrected with turbid solutions o o o o E a t 525mp with turbid solutions suspensions of barium sulphate and bacteria.The figures plotted in Fig. 1 are mean values. The suspensions were shaken, the extinctions read from the red to the blue wavelengths, re-shaken, and then read in the opposite direction. With heavy turbidities and relatively large particles, the slope of the curves always declines with the shorter wavelengths. Scattered light is probably reflected by the particles and re-scattered to proceed in the direction of the eye or the photo-cell. In the scattered light, the blue fraction predominates and extinctions read at the shorter wavelengths are consequently diminished (Fig.1, curve number 7). This effect may increase if the bottom 0 -I .20 - I a00 460 I .oo- 0 -0.20 Wavelength, mp (log scale) Serum, Fig. 5. Bilirubin in human serum. Extinctions read with the wavelength 460 mp corrected by means of extinctions read a t 620 and 610 mp. The human serum was diluted with distilled water A, human serum with slight haemolysis B, human serum with heavier haemolysis C, oxyhaemoglobin D, bilirubin in human serum*468 FOG : TURBIDITY IN PHOTOMETRY Pol. 77 and the sides of the cells are frosted, but usually, with moderate turbidities, only a decrease in the slope of the whole curve occurs. Figs. 2, 3 and 4 show the spectral absorption curves for a solution of cobaltous nitrate, an alkaline solution of glycerol - copper sulphate (Haines reagent) read against water, and sodium salicylate dissolved in a dilute acid solution of ferric chloride read against the ferric chloride solution, respectively. A suspension of barium sulphate was added to solutions of the coloured substances a t different concentrations.The filters s61, s50 and s43 were used with the cobaltous nitrate, and the wavelengths 900, 625 and 440mp and 1000, 525 and 390 mp with the copper sulphate and the salicylate, respectively. Non-turbid solutions were prepared in triplicate for comparison. The corrected extinctions agree to a great extent independently of turbidity, as shown in Figs. 3 and 4 and Table 111. In Table I are shown the readings with different-amounts of the turbid solution added to solutions of the coloured substances at the same concentrations.Table I1 shows corresponding values obtained with cobaltous nitrate solutions. Blood serum is freed from turbidity and most haemoglobin only with great skill and luck. Spectral absorption curves of undiluted serum plotted in Fig. 5 (A and B) show peaks at 577 and 414mp caused by oxyhaemoglobin. The pointed curves show the absorptive properties of pure oxyhaemoglobin (C) and bilirubin (D). Bilirubin in human serum has its maximum absorption at 460mp.lO When serum is added to distilled water, proteins are precipitated to give a turbidity that is dependent on the degree of dilution.11 In Fig. 5 the wavelengths 620 and 510 mp are used to correct the extinctions read at 460 mp (bilirubin). Corrected extinctions for bilirubin are nearly linearly related to the dilution of serum in spite of the variation in turbidity (see Fig.5). DISCUSSION OF RESULTS The extinctions of polydispersed sols and suspensions can usually be expressed by the formula E = kAy, and it is especially to be noted how well the equation fits for protein TABLE I EFFECT OF DIFFERENT QUANTITIES OF A TURBID BARIUM SULPHATE SUSPENSION ADDED TO A CONSTANT AMOUNT OF THE COLOURED SUBSTANCES Alkaline glycerol - copper sulphate (900-625-440 mp)* Sodium salicylate - ferric chloride (1000-525-390 mp) t A A f \ r 7 Difference of Difference of ECOIT. from ECOIT. from E a t the mean E at the mean 625 mp Ewrr. of Ewrr. 525 mp Ecorr. of Ecorr. 0.670 0.392 0.005 0.793 0.680 0.010 0-632 0.397 0 0.765 0.664 0.006 0.596 0.397 0 0.764 0.670 0 0.569 0-399 0.002 0.750 0.667 0.003 0.527 0.399 0.002 0.726 0.658 0.012 0.488 0.398 0.001 0.694 0.678 0.008 0.397 0.670 (mean) (mean) * Read against water.t Read against ferric chloride solution. precipitates, which , together with finely dispersed fat (oil-dispersed-in-water emulsions) , are the common causes of turbidity when photometry is used with biological samples. Only if E, > El (A, > h3) is z negligible, and (E2” - A E2”) should be as great as possible (approaching E,“) to minimise the influence of errors in the correcting factors, especially as only two significant figures are usually read from the nomogram with a sufficient degree of accuracy. With complicating foreign coloured substances, the wavelengths used to correct for turbidity should be chosen at the minima of their absorption curves (Fig.5 ) . The salicylate solution (Fig. 4) did not absorb at 1000 mp, and a relatively small number (0.001) was therefore added with the clear solutions to simulate a turbidity. In Table I, for alkaline copper sulphate, the maximum differences between the values obtained for E at 625mp and ECom. are 0.182 and 0.007, respectively, and the maximum deviation of E,,, from the mean is only 1.3 per cent. With the salicylate solutions the459 Sept., 19521 FOG : TURBIDITY I N PHOTOMETRY TABLE I1 EFFECT OF DIFFERENT QUANTITIES OF TURBID BARIUM SULPHATE SUSPENSION ADDED TO A Extinctions read in a Zeiss Pulfrich photometer against water. Filters s61 and s43 The values are means of CONSTANT AMOUNT OF COBALTOUS NITRATE SOLUTION were used to correct the extinctions read with the filter s50.extinctions read 5 times Clear solutions* Turbid solutions* f h > I A \ Difference of Difference of E (s50 filter) Ecorr. 0-443 0.367 0-444 0.370 0.430 0.358 0-436 0.359 0.45 1 0.374 0.43 1 0.357 0-436 0.364 0.444 0.364 0.434 0.356 0-449 0.37 7 0.365 (mean) * The clear and turbid EXTINCTIONS ECOZ. from the mean of Ecorr. 0.002 0-005 0-007 0.006 0.009 0.008 0-001 0-001 0.009 0-012 E (s50 filter) 0.450 0-459 0.506 0.528 0.574 0.579 0.644 0-665 0.727 0.737 Ecorr. 0.36 1 0.35 1 0.364 0.357 0.353 0,348 0.352 0.355 0.358 0.359 0.356 (mean) EWr. from the mean of Ecorr. 0-005 0.005 0.008 0.00 1 0.003 0.008 0.006 0.00 1 0.002 0.003 solutions were prepared from different stock solutions. TABLE 111 OF ALKALINE GLYCEROL - COPPER SULPHATE SOLUTIONS Extinctions read at a wavelength of 625mp and corrected for turbidity by means of the wavelengths 900 and 440 mp (Eiorr.) and 800 and 440 mp (EFo,.) in a Beckman model B swctroDhotometer I A Dilution 6/10 5/10 4/10 3/10 2/10 1/10 Type of solution Clear T&bid Clear Tiibid Clear T;;bid Clear T&bid Clear T;;bid Clear T&bid *, ,, ,, I* E at 625 mp 0.809 0.804 0.806 0.943 0.677 0.681 0.68 1 0.842 0.561 0.555 0.552 0.702 0.427 0.427 0-424 0,569 0-289 0.294 0.289 0.420 0.151 8.149 0.151 0.293 ELT.0-781 0.775 0.781 0.778 0-654 0.656 0.656 0.660 0.541 0.534 0.533 0.538 0.41 1 0.412 0.4 10 0.408 0.277 0-282 0-279 0-278 0.146 0.144 0-146 0.145 Difference from the mean* 0.002 0.004 0.002 0.001 0.001 0.001 0.00 1 0.005 0.005 0.002 0-003 0.Q02 0 0.00 1 0.00 1 0-003 0.002 0.003 0 0.001 0.001 0.001 0.001 0 E20rr.l 0-732 0.726 0.727 0-713 0.612 0.614 0.613 0.602 0.506 0.499 0-496 0.490 0-381 0.383 0.381 0.370 0-259 0-262 0.260 0-252 0,135 0.134 0-135 0.131 Difference from the mean* E20n.a 0.004 0.002 0.001 0.0 15 0.730 0.001 0.001 0 0.01 1 0.619 0-006 0.00 1 0.004 0.010 0.503 0.00 1 0.00 1 0.00 1 0.0 12 0.382 0.001 0.002 0 0.008 0.261 0 0*001 0 0.004 0.137 Difference from the mean* 0.002 0-006 0.003 0 0.001 0.002 * Difference from the mean values of extinctions read with the clear solutions.460 ARMSTRONG : THE DETERMINATION OF [Vol.77 maximum deviation from the mean is 1.8 per cent. in spite of the uncertainty attaching to transference of salicylate solutions by means of a pipette (droplets adhere to the glass).Table 11, for cobaltous nitrate, shows that the maximum deviation of E,,,. from the mean was 3.3 per cent. with the clear and 2.2 per cent. with the turbid solutions, With the clear solutions, the numerical values of the extinctions read with the filters s61 and s43 are rather small and therefore more liable to error than with the turbid solutions. Table I11 demonstrates again how excellently the method works with the alkaline copper sulphate solutions even if E, is greater than E,, but it is then essential that x be calculated and added to the first value of ECorr.. This investigation has been supported by grants from the Norwegian Research Council. The author wishes to thank J. H. Vogt, M.D., for his encouragement in the development of the method. REFERENCES 1. 2. 3. 4. 5 . 6. 7. 8. 9. 10. 11. Morton, R. A., and Stubbs, A. L., Analyst, 1946, 71, 348. Lowry, 0. H. and Hastings, A. B., J . Biol. Chem., 1942, 143, 257. Zerban, F. W., Sattler, L., and Lorge, T., Ind. Eng. Chem., Anal. Ed., 1931, 3, 326; 1934, 6, 178; Snell, F. D., and Snell, C. T., “Colorimetric Methods of Analysis,” Van Nostrand Co. Inc., n’ew Dognon, A., Compt. Rend. SOC. Biol., 1941, 135, 113. Fog, J., Scand. J . Clin. Lab. Invest., 1949, 1, 255. “Tables of Scattering Functions for Spherical Particles,” National Bureau of Standards, Applied Barnes, M. D., and La Mer, V., J . Colloid Sci., 1946, 1, 79. La Mer, V., J . Phys. Colloid Chem., 1948, 52, 65. Heilmeyer, L., “Medizinische Spektrophotometrie,” G. Fisher, Jena, 1933, p. 152. NikkilB, E. A., Scand. J , Clin. Lab. Invest., 1950, 2, 322. 1935, 7, 157; 1936, 8, 168; 1937, 9, 229. York, 1948, p. 145. Mathematics Series 4, Washington, 1949. DEPARTMENT OF MEDICINE NAMDAL HOSPITAL NAMSOS, NORWAY Februavy, 1952

ISSN:0003-2654    

DOI:10.1039/AN9527700454

出版商:RSC    

年代:1952

数据来源: RSC

摘要:

460 ARMSTRONG : THE DETERMINATION OF [Vol. 77 The Determination of Potassium Bromate in Flour BY A. W. ARMSTRONG A titrimetric method for the determination of potassium bromate in flour, is described. A suspension of flour in zinc sulphate solution is clarified by filtration of precipitated zinc hydroxide. Bromate is estimated iodi- metrically in an aliquot of the filtrate by titration of an excess of standard thiosulphate with standard iodate. Results can be corrected by applying a “recovery factor.’’ OF the methods so far published for the estimation of potassium bromate in flour, those of Geddes and Lehbergl and Geddes2 are too time-consuming for routine use. The colorimetric methods of Hoffer and Alcock3 and Johnson and Alcock4 are complicated and require a spectro- photometer.Howe5 has described the estimation of bromate in zinc filtrates of flour suspensions, but we failed to obtain satisfactory results on applying her method to National flours and have had to modify it. On treating 40-g samples of National flours with zinc sulphate and sodium hydroxide according to Howe’s procedure, filtrates were often obtained that, on acidification and after addition of potassium iodide, gave a precipitate that completely obscured the thiosulphate end-point. Even when clear solutions were obtained the end-point was indefinite, the starch iodide colour changing from purplish blue, through reddish purple and orange, to a yellow that faded slowly. Howe’s method of clarification is based on that of Auerbach, Eckert and Angell,6 which is, itself, an adaptation of Sornogyi’s’ pa method of preparing protein-free filtrates from blood and plasma.Auerbach et al. found that “with a very few samples, more than the recommended amount of zinc hydroxide was required to bring about a clear zinc filtrate.” They prescribe 0.3 N zinc sulphate solution when, presumably, 0-3 M isSept., 19521 POTASSIUM BROMATE IN FLOUR 461 intended, since the “very faintly acid” mixture which they describe can be obtained only with the latter. Howe prescribes 0.18 N zinc sulphate solution, containing 51.7 g per litre. With the heptahydrate, 51-7 g of zinc sulphate per litre gives a 0.18 M solution. If 0.09 M , i.e., 0.18 N , zinc sulphate solution is used, the pH value of the filtrate is above 9. When 0.18 M zinc sulphate solution is used the filtrate is nearly neutral.By increasing the zinc concentration in the clarifying medium, we have been able to obtain from 5 0 g of flour more than 100ml of filtrate that remains clear throughout the estimation. A 50-ml aliquot of this filtrate is equivalent to 10 g of flour, and can be used for the estimation of additions of bromate of between 1 and 100 parts per million; for the estimation of larger additions 25-ml or smaller aliquots must be used. The difficulty of judging the thiosulphate end-point is obviated by adding an excess of thiosulphate and titrating the excess with potassium iodate. Thiosulphate in acid solution in presence of an excess of iodide can be titrated accurately with potassium iodate. Standard potassium iodate, which is stable in solution, is therefore used for standardising the thiosulphate and also for titrating the excess of it.A trace of ammonium molybdate is used to catalyse the liberation of iodine from iodide by bromate, so that an excess of thiosulphate can be added immediately. In the titration of an aliquot of filtrate representing 10 g of flour, 1 ml of 0.00359 M thiosulphate solution is equivalent to 10 parts of bromate per million in the flour. In the titration with iodate of the same strength, discrimination at the end-point, in the procedure to be described, is better than 0.05 ml; this corresponds to 0-5 parts of bromate per million in the flour. The precision of the titration is much greater than is necessary for routine work since, in the analysis of cor imercial flours containing bromate, which may be unevenly distributed and the particles of which may vary greatly in size, the sampling error is often large.Only about 95 per cent. of the bromate added in solution to flour suspensions can be recovered. Absorption of brom tte ion by zinc hydroxide, per se, is excluded, since bromate added to solutions of zinc sulphate can be recovered quantitatively from the filtrates after precipitation of the zinc as hydroxide. Reduction of bromate by soluble reducing substances in flour is also excluded; bromate added to flour filtrates can be recovered quantitatively. The low recoveries of bromate added to flour suspensions can be partly accounted for by postulating that dilution of the bromate takes place during the clarifying process.The liquid phase, being hypertonic with respect to wheat endosperm, will extract some moisture from the flour. Further dehydration is almost certainly associated with the actual clarifica- tion, which includes at least partial denaturation of the flour proteins. A total extraction of 7.5 g of water from 50 g of flour would, in the procedure to be described, account for an apparent loss of 3 per cent. of the added bromate. Apparent losses of this order are of no importance in ordinary control analyses, but, for more precise work, a “recovery factor” can be determined by recovery of bromate added, in solution, to flour suspensions (Table I). All the flours examined, including strong and weak, agenised and non-agenised, short extraction and long extraction, have yielded nearly the same recovery factor and, for a given kind of flour, the factor varies around 1 per cent.in replicate determinations at various bromate concentrations. Greater precision is precluded by the limited volumetric accuracy of the procedure. Further, since the discrimination at the end-point of the final iodate titration is about 0.03 ml, recovery factors should not, normally, be calculated from, nor applied to, results obtained when the flour bromate is less than 30p.p.m. Howe emphasises that her method is not highly specific and that, if iodate or persulphate are present, they will interfere. She adds, “From a practical standpoint, however, benzoyl peroxide, chlorine, and chlorine dioxide (the bleaching and maturing agents commonly used on flour) do not interfere and, since iodate and persulphate are not permitted in the Standards of Identity of Wheat Flour, bromate normally is the only compound present that would be measurable by the method outlined.” I t is, at least, improbable that iodate would be encountered in British flours; if it were, it would react quantitatively with iodide in this modification of Howe’s procedure, and the method is, to that extent, non-specific. Persulphate is commonly added to British flours.There is, however, a priori, no reason for expecting that it will interfere in the method described here for the determination of bromate, since (i) in acid solution the reaction between persulphate and iodide is extremely slow at low concentrations of the reactants and (ii) Auerbach et aL6 find that on adding462 ARMSTRONG: THE DETERMINATION OF [vol.77 water to flour containing persulphate, the persulphate disappears rapidly. At 27" C, a dough prepared by adding 60 parts of water to 100 parts of flour, which contained 200 parts of ammonium persulphate per million, had lost 72 per cent. of the persulphate 8 minutes after addition of the water. Experimentally, it was found that the addition to a suspension of flour (containing bromate) of a freshly prepared persulphate solution, equivalent to 160 parts of ammonium persulphate per million in the flour under test, was without effect upon the recovery of the bromate present. The addition of 2000 parts of benzoyl peroxide per million to flour did not affect the recovery of added bromate.Neither agene nor' chlorine treatments of flour affected the recovery of added bromate. The method has not been applied to flours treated with chlorine dioxide. REAGENTS- Zinc sulphate solution-Dissolve 20 g of zinc sulphate (ZnS0,.7H20) in 800 ml of water and dilute to 1 litre. Sodium hydroxide, 0.4 N-Dissolve 17 g of sodium hydroxide in 1 litre of water. Titrate the solution against standard acid and adjust the strength to 0.4 (-+ 0.01) N . Sodium hydroxide, 0.5 N-Dissolve 21 g of sodium hydroxide in 1 litre of water. Titrate the solution against standard acid and adjust the strength to 0.5 (5 0.01) N . Sulphuric acid-An approximately 4 N solution. Add 112 ml of concentrated sulphuric acid to 800 ml of water. Potassium iodide solution-Dissolve 25 g of potassium iodide in 30 ml of water and dilute to 50 ml.Discard any solution that shows a yellow tinge of free iodine. Ammonium molybdate-Dissolve 3 g of ammonium molybdate ( (NH4),Mo,024.4H,0) in 80 ml of water and dilute to 100 ml. Starch solution-Pour a suspension of 1 g of soluble starch in 5 ml of water into 100 ml of briskly boiling water. Prepare a fresh solution daily. Standard potassium bromate solution-Prepare a stock solution by dissolving 5.000 g of potassium bromate (dried for 1 hour at 110" C) in 800 ml of water, and dilute to 1 litre. Prepare a standard potassium bromate solution by diluting 25 ml of the stock solution to 250 ml. Potassium iodate, 0.0898 N-Dissolve 3.204 g of potassium iodate (dried for 1 hour at 110" C) in 800 ml of water and dilute to 1 litre.Potassium iodate, 0.00359 N-Dilute 10 ml of the 0-0898 N potassium iodate solution to 250 ml. Stock sohtion of sodium thiosul~hate-Dissolve 22.5 g of sodium thiosulphate (Na,S20,.5H,0) and 0.06 g of anhydrous sodium carbonate in 800 ml water, and dilute to 1 litre. Dilute 10 ml to 250 ml. Transfer 5 ml of this diluted solution to a 250-ml conical flask. Add 100 ml of water, 10 ml of the approximately 4 N sulphuric acid and 1 ml of potassium iodide solution. Add 5 ml of starch solution and titrate with 0.00359 N potassium iodate from a 5-ml burette graduated to 0.01 ml. Adjust the titre of stock sodium thiosulphate so that a 10 to 250 dilution is 0.00359 N . Store the stock solution in an amber-coloured bottle in a cool place. Sodium thiosulphate solution, 040359 N-Dilute 10 ml of the stock sodium thiosulphate solution to 250ml.Prepare a fresh diluted solution daily. Check the titre of the diluted solution as above at least monthly. PROCEDURE- Transfer 50 g of flour to a 500-ml conical flask. Add 200 ml of zinc sulphate solution. Close the flask with a rubber bung and shake vigorously at intervals for 10 minutes. While vigorously swirling the contents of the flask, add, from a burette, 50 ml of 0.4 N sodium hydroxide. Filter through a 24-cm Whatman No. 5 (or No. 40) folded filter-paper. Return the first few millilitres of filtrate to the paper. Alternatively, centrifuge the mixture and, if necessary, clarify the supernatant liquid by filtration. Transfer 50ml of the filtrate to a 250-ml conical flask. If a smaller aliquot is taken, make the volume up to 50 ml with water.Add 10 ml of the approximately 4 N sulphuric acid, 1 ml of potassium iodide solution and 1 drop of ammonium molybdate solution. Dilute the mixture with METHOD Cool the solution and dilute it to 1 litre. Store in an amber-coloured bottle in a cool place. Boil for 2 minutes and cool rapidly. Prepare a fresh dilution daily. Shake vigorously and allow the mixture to settle for 5 minutes. Collect the clear filtrate in a dry vessel.Sept., 19521 POTASSIUM BROMATE IN FLOUR 463 50 ml of water. With steady mixing, add, from a pipette, an excess of 0.00359 N sodium thiosulphate (5 or 10 ml). Add 5 ml of starch solution and titrate the excess of thiosulphate with 0.00359 N potassium iodate to the first permanent faint purple tinge of the liquid.Parts of potassium bromate per million in the flour = x (y - titration)/5, where x = volume of the aliquot andy = volume of thiosulphate used. DETERMINATION OF RECOVERY FACTOR- Dilute a known volume (% ml), greater than 3 ml but less than 10 ml, of standard potassium bromate solution to 250 ml. To 50 ml of this solution add 10 ml of the approxi- mately 4 N sulphuric acid, 1 ml of potassium iodide solution and 1 drop of ammonium molybdate solution. Dilute the mixture with 50 ml of water and, with steady mixing, add from a pipette an excess (5 or 10 ml) of 0.00359 N sodium thiosulphate solution. Add 5 ml of starch solution and titrate with 0.00359 N potassium iodate solution. Suspend two 50-g portions of flour in two 200-ml portions of zinc sulphate solution. To one (blank) suspension, add 10 ml water.To the other (recovery) suspension, add n ml of standard potassium bromate solution and (10 - n) ml of water. Shake the mixtures vigorously at intervals for 10 minutes. Vigorously swirling the contents, add to each flask 40 ml of 0.5 N sodium hydroxide from a burette. Thereafter proceed as for the estimation of bromate in flour as described above, using 5 rnl of thiosulphate for the blank and 10 ml for the recovery estimation. Deduct the blank value, if any, from the value of the bromate found in the recovery estimation to give “Recovered bromate.” Added bromate, pap.m. = 10 (y - titration), wherey = volume of thiosulphate used. Added bromate Recovered bromate ’ Recovery factor = TABLE I REXOVERIES O F POTASSIUM BROMATE ADDED I N SOLUTION TO FLOUR SUSPE NSIONS Recovery Factor = 1.067 Bromate added, p.p.m. 1 2 5 10 15 20 30 40 60 80 Bromate recovered, p.p.m. 1.0 2.0 5.0 9.2 14.4 18.3 28.1 37.2 56.4 75.1 Recovery x 1.067, p.p.m. 9.8 15.4 19.5 30.0 39.7 60.2 80.1 Recovery x 1.067, % 98.0 102.7 97.5 100.0 99.3 100.3 100.1 REFERENCES 1. Geddes, W. F., and Lehberg, F. H., Cereal Chern., 1938, 15, 49. 2. Geddes, W. F., J - Ass. 08. A&. Chew., 1948, 31, 260. 3. Hoffer, A., and Alcock, A. W., Cereal Chem., 1946, 23, 66, 4. Johnson, R., and Alcock, A. W., Ibid., 1948, 25, 266. 5. Howe, M., Ibid., 1951, 28, 132. 6. Auerbach, M. E., Eckert, H. W., and Angell, E., Ibid., 1949,26, 490. 7. Somogyi, M.. Proc. SOC. Exp. Biol. Med., 1929, 26, 353. 8. - , J . Biol. Chem., 1930, 86, 655. SCOTTISH CO-OPERATIVE WHOLESALE SOCIETY, LIMITED CEREAL LABORATORY REGENT MILLS GLASGOW, C.3 April, 1952

ISSN:0003-2654    

DOI:10.1039/AN9527700460

出版商:RSC    

年代:1952

数据来源: RSC

摘要:

464 HASLAM, RUSSELL AND WILKINSON : THE DETERMINATION OF [Vol. 77 The Determination of Nickel and Manganese in Uranium* BY J. HASLAM,? F. R. RUSSELL AND N. T. WILKINSON Methods have been developed for the determination of 0 to 0.02 per cent. of nickel, and 0 to 1.0 per cent. of manganese in uranium metal, The method for manganese depends on solution of the sample in nitric acid, and separation of silica by dehydration of the solution in the presence of hydrochloric acid. In the determination of nickel the sample is dissolved in nitric acid, which is subsequently removed by evaporation in the presence of sulphuric acid. In the nickel determination metals that form insoluble sulphides are separated directly in the diluted sulphuric acid solution; and in the manganese determination, after the separation of silica.The filtrate from the sulphide precipitation is boiled until free from hydrogen sulphide, and the nickel and manganese precipitated in the presence of iron by sodium carbonate. Precautions are taken to ensure complete precipitation of the nickel and manganese by decomposition of any bicarbonate by boiling. The mixed precipitate is filtered and washed with dilute sodium carbonate solution. The precipitate is then dissolved in acid and the nickel determined colorimetrically by the hypobromite - dimethylglyoxime method. Manganese is determined volumetrically with standard ferrous ammonium sulphate on a solution of the precipitate obtained in the same manner as for nickel after oxidation to permanganate. OUR first experiments were directed to the investigation of the precipitation of nickel in the presence of iron by means of sodium carbonate.Solutions containing appropriate amounts of iron and nickel were precipitated with sodium carbonate, the precipitate dissolved in nitric acid, the solution evaporated to dryness and the residue dissolved in a little hydrochloric acid. The iron in this solution was then precipitated with ammonium hydroxide and the nickel in the filtrate determined by the application of the colorimetric dimethylglyoxime procedure. As was not entirely unexpected, the recoveries of nickel were low, but our experiments suggested that it might be possible to apply the glyoxime reaction directly to the solution containing nickel and iron, interference of iron being prevented by the addition of tartrate ion.It was found that the amount of iron present in the test solution did not interfere when the colorimetric dimethylglyoxime reaction was applied in the presence of tartrate ion. Experiments on iron - nickel solutions gave satisfactory recoveries of nickel with this direct met hod. When carrying out the determination, by preliminary precipitation of iron and nickel with sodium carbonate from an acid solution containing uranium, the recovery of nickel was low, and in general the results were erratic. Modifications of the concentration of the various substances present did not effect much improvement. It appeared from the experiments that complete precipitation of iron and nickel was not being effected by the sodium carbonate because the application of the test to samples of uranium involved solution of the sample in an excess of acid.On the addition of sodium carbonate to the solution containing the excess of acid, appreciable amounts of bicarbonate would be produced, so that in effect the precipitation of iron and nickel was being carried out with sodium carbonate in the presence of sodium bicarbonate. It appeared, therefore, that for complete precipitation of iron and nickel any sodium bicarbonate that was produced must be decomposed. * This work was carried out originally in 1946 and 1946, and forms the subject of two reports, BR 623 and BR 687, which were declassified by the Department of Atomic Energy, Ministry of Supply, on February 13th, 1962. Present address : Imperial Chemical Industries Limited, Plastics Division, Welwyn Garden City, Herts.Sept., 19521 NICKEL AND MANGANESE IN URANIUM 466 Efforts to overcome the trouble were made as follows- (1) By adding an excess of sodium carbonate solution to the cold acid solution of the sample, followed by boiling to effect decomposition of the bicarbonate.(2) By adding sodium carbonate solution to the boiling solution of the sample. This gave a yellow precipitate. (3) By adding an excess of sodium carbonate solution to the cold acid solution of the sample, then adding ammonium hydroxide and boiling. (4) By adding an excess of sodium carbonate solution to the cold solution of the sample, followed by boiling to decompose the bicarbonate. After this operation the solution was diluted and re-boiled.Of these four procedures, (4) proved fo be the most satisfactory, and the method of determination of nickel in samples of uranium is based on this procedure. The same procedure for the preliminary separation of manganese was applied to the determination of manganese in uranium. The results were satisfactory. It is necessary to remove silica and metals precipitated by hydrogen sulphide in acid solution before proceeding to the precipitation of the iron, nickel and manganese. For the precipitation of the latter metals a specially purified solution of sodium carbonate is used. METHOD FOR THE DETERMINATION OF NICKEL SPECIAL REAGENT- Puri$ed sodium carbonate sola&ort, 14 j5er cent. w/v-Dissolve 70 g of sodium carbonate in 300 ml of water and add 100 ml of standard ferric chloride solution (1 ml = 0.0001 g of Fe,O,).Add a small amount of filter-paper pulp and digest the mixture on a steam-bath for 30 minutes. Cool the solution, filter through a Whatman No. 40 filter-paper and wash the filter with a little water. Dilute the filtrate to 500 ml. PROCEDURE- Dissolve about 5 g of the sample (accurately weighed) in 10 ml of concentrated nitric acid by heating the mixture, in a 250-ml beaker, on a water-bath. Add 9 ml of concentrated sulphuric acid and heat the solution to fuming on a sand-bath. Hold the beaker in a clamp and swirl the mixture continuously during this evaporation. When white fumes of sulphur trioxide appear, cool the mixture and dilute with water. Usually the amount of insoluble matter is small and filtration is unnecessary.Dilute the solution to about 150m1, then saturate it with hydrogen sulphide by passing the gas into the solution for about 15 minutes. Heat the solution on the water-bath for about 30 minutes in order to coagulate the precipitate, filter and wash the insoluble precipitate with sulphuric acid solution (2 per cent. v/v) saturated with hydrogen sulphide. Boil the filtrate and washings until free from hydrogen sulphide, add a few millilitres of bromine water and boil the solution to free it of the excess of bromine. Cool the solution and dilute it to 200ml with water in a graduated flask. Take 50ml of the solution and adjust* the iron content of the solution to 0.001Og by adding standard iron solution (1 ml = 0.0001 g of Fe,O,). Dilute the solution to 100 ml and add 50 ml of purified sodium carbonate solution to the cold solution.Evaporate the solution to a volume of approximately 80 ml, add 70 ml of hot water and again evaporate to 80ml. Add a small amount of filter-paper pulp, and filter the solution through a 7-cm Whatman No. 42 filter-paper; wash the insoluble matter with sodium carbonate solution (2 per cent. w/v) until free from uranium. Return the filter-paper pulp and precipitate to the beaker used for the original pre- cipitation by means of a jet of water (about lOml), add 2ml of concentrated nitric acid and heat the mixture until the precipitate has dissolved; pour this solution through the 7-cm filter-paper and collect the filtrate in a clean 100-ml beaker. Re-treat the filter-paper pulp with a mixture of 10 ml of water and 1 ml of concentrated nitric acid and add the extract through the 7-cm filter-paper to the original extract. Wash the filter-paper pulp well with water.aliquot of the test solution. Allow this solution to stand overnight. Reserve the filtrate and washings. * When the iron content of the uranium sample is excessively high, it may be desirable to take a smaller466 HASLAM, RUSSELL AND WILKINSON : THE DETERMINATION OF Evaporate the combined filtrate and washings (in the 100-ml beaker) to dryness by immersing the beaker in a bath of boiling water and bring the residue into solution with 0.5 ml of N hydrochloric acid solution by gentle warming. Wash the solution into a measuring flask of 100-ml capacity, add 5 ml of Rochelle salt solution (30 g of Rochelle salt dissolved in water and diluted to 100 ml and filtered), and then add 2 ml of saturated bromine water and 1 ml of ammonium hydroxide solution (1 volume of ammonium hydroxide, sp.gr.0.880, to 2 volumes of water). After setting the solution aside for 5 minutes, add 2 ml of dimethylglyoxime solution (1 g of dimethylglyoxime dissolved in 100 ml of alcohol) and examine the solution in a Spekker photo-electric absorptiometer using, according to the depth of colour, either a 4-cm or a I-cm cell and No. 5 green filters. Deduce the amount of nickel present in the 100ml of solution from previously prepared calibration curves that relate the amount of nickel in grams with the Spekker indicator drum readings. These curves are smooth and pass through the following points- [vol.77 With 4-cm cells and No. 5 green filters Nickel per 100 ml g A r z Indicator drum of solution, reading 0.000005 0.0 17 0~000010 0.037 0*000020 0.083 0~000040 0.162 0.000060 0.255 0~000120 0.490 With l-cm cells and No. 5 green filters Nickel per 100 ml !of solution, reading g r A \ Indicator drum 0*00002 0,024 0.00006 0.07 1 0~00010 0.1 17 0.00018 0.2 12 0.00024 0-280 Where the nickel content of the uranium sample is comparatively high, viz., of the order of 0.003 per cent., a further 0.0010 g of iron (as ferric chloride solution) is added to the previously reserved filtrate and the nickel determined in this filtrate by carrying out the procedure described on p. 465. The above method has been applied to known uranium - nickel solutions with the results shown in Table I.Each solution used for the precipitation with sodium carbonate contained the equivalent of 1.25 g of uranium. The nickel contents are calculated as percentages of the uranium. All the results are corrected for a control test on the reagents. TABLE I DETERMINATION OF NICKEL IN URANIUM - NICKEL SOLUTIONS Nickel found Nickel f o n d Nickel added, (one precipitation), (two precipitations), % % % nil 0.00040 0.00080 0.00160 0.00320 0*00480 0.00960 0.00960 0.0 1920 0.01920 nil 0.00032 0.00072 0.00 152 0.00304 0.00456 0-00896 0.00872 0.0189 0.0186 nil 0.00032 0.00072 0.00160 0.00320 0.00488 0.00936 0.00920 0.0191 0.0189 METHOD FOR THE DETERMINATION OF MANGANESE SPECIAL REAGENT- Purified sodium carbonate solution, 14 p e r cent.w/v-Prepare as described on p. 465. PROCEDURE- Dissolve about 5 g of the sample (accurately weighed) in 20ml of concentrated nitric acid by heating on a water-bath; evaporate the acid solution to dryness on a sand-bath. After the removal of the nitric acid, evaporate twice further with hydrochloric acid usingSept., 19521 NICKEL AND MANGANESE IN URANIUM 467 10 ml of the concentrated acid on each occasion. Heat the residue in an oven at 120" C overnight. Treat the baked residue with 5 ml of concentrated hydrochloric acid and 20 ml of water and heat the mixture on the water-bath to effect complete solution of the uranium salts. Filter off the insoluble silica on a 9-cm Whatman No. 40 filter-paper and wash with 5 per cent. v/v hydrochloric acid solution and with hot water to remove all uranium salts.Ignite the insoluble siliceous residue, weigh it in a platinum crucible and treat it with dilute sulphuric acid and hydrofluoric acid in order to remove the silica. Fuse the residue with a little potassium acid sulphate and incorporate the aqueous extract of the melt in the main uranium solution. Saturate this solution, of volume approximately 200 ml, with hydrogen sulphide. Filter off the precipitated hydrogen sulphide metals, wash with acidulated hydrogen sulphide water and boil the filtrate to free it from hydrogen sulphide. Oxidise the solution by adding bromine water, free it from bromine by boiling, cool and dilute to 250ml in a measuring flask. Take an aliquot of this solution equivalent to about 1 g of the original uranium metal for the manganese determination.If from previous iron determinations it is known that the aliquot taken daes not contain as much as 0.002 g of iron, then add standard iron solution (1 ml = 0.0001 g of Fe20,) in amount sufficient to give a total of 0.002 g of iron; if the aliquot contains more than 0.002 g of iron, then add no further standard iron solution. Add 50 ml of 14 per cent. w/v sodium carbonate solution and dilute the solution to 150 ml. Boil down the solution to a volume of 80 ml, dilute to 150 ml and again boil down to 80 ml. Add a small amount of paper pulp to the mixture, which is allowed to coagulate on the water-bath for about 10 minutes. After cooling the precipitate filter it off on a 9-cm Whatman No. 40 filter-paper and wash the residue three times with 2 per cent.w/v sodium carbonate solution. Prepare this sodium carbonate solution by diluting the 14 per cent. w/v sodium carbonate solution. Place the paper and precipitate in the beaker originally used for the precipitation and heat on the water-bath with a mixture of 5 ml of concentrated nitric acid and 10 ml of water until all the precipitate dissolves and the filter-paper is quite white. Filter the mixture through an ll-cm Whatman No. 40 filter-paper and wash it first with 10 ml of hot 30 per cent. v/v nitric acid solution and then thoroughly with hot water. Boil the filtrate (volume approximately 150ml) down to about 50m1, cool, add 0.5g of sodium bismuthate and then heat to boiling. Decolorise the clear pink solution with a few drops of sulphur dioxide water and boil off the excess of sulphur dioxide.Again cool the solution, add 10 ml of concentrated nitric acid, followed by 0.5 g of sodium bismuthate and then set it aside in the dark for 1 hour. Filter the mixture with the aid of suction through a sintered-glass funnel* of porosity 4, and wash the insoluble residue with the minimum amount of 3 per cent. v/v nitric acid solution. Add standard 0.02 N ferrous ammonium sulphate to the filtrate (approximate volume 80 ml) until the permanganate colour is destroyed, after which add an excess of 2 to 3 ml. Add 0.02 N potassium permanganate fairly quickly at first, then 2 drops at a time, until the solution is just pink. At this stage add 1-00 ml of 0.02 N potassium permanganate, then 1.OOml of 0-02 N ferrous ammonium sulphate and finally 0.02 N potassium permanganate solution, 2 drops at a time, until the permanent pink end-point appears. Calculate the proportion of manganese in the sample from the amount of ferrous solution consumed by the permanganate produced in the bismuthate oxidation. The principle of the method has been tested by applying the procedure to known mixtures of thrice crystallised uranium nitrate (equivalent to 1 g of uranium) and reduced standard potassium permanganate solution. To all the mixtures, the equivalent of 0.002 g of iron was added. The following results were obtained, the manganese added being calculated as a per- centage of the uranium present. Manganese added, yo . . 0.11 0.11 0.22 0.22 0.44 0-55 0.55 1-10 1-10 1.10 Manganese found, yo . . 0.088 0.090 0.216 0.206 0.444 0-533 0.535 1.095 1.100 1.060 IMPERIAL CHEMICAL INDUSTRIES LIMITED ALKALI DIVISION NORTHWICH, CHESHIRE April, 1962 * Before using the sintered-glass funnel for the final filtering operation, wash it first with potassium permanganate solution, then with 20 per cent. nitric acid and finally with water.

ISSN:0003-2654    

DOI:10.1039/AN9527700464

出版商:RSC    

年代:1952

数据来源: RSC

摘要:

468 FROST- JONES AND YARDLEY : THE COLOUR REACTIONS [Vol. 77 The Colour Reactions of Chloranilic Acid with Particular Reference to the Estimation of Calcium and Zirconium BY R. E. U. FROST-JONES AND J. T. YARDLEY The colour reactions of chloranilic acid are studied in some detail and with particular reference to interference in the colorimetric estimation of calcium and zirconium. A method is described for cancelling interference by sodium, and a detailed procedure is described for estimating zirconium in the presence of a number of other ions. The influence of the perchloric acid concentration on the determination is discussed. IN recent years some intcrest seems to have been focused on the use of chloranilic acid as a colorimetric reagent for c dcium. The demand for analytical grades of the material is fairly substantial and, as the literature is not extensive, it seemed desirable to explore the poten- tialities of the reagent with a view to providing information beyond that already available.THE REACTION WITH CALCIUM Chloranilic acid (2 :5-dichloro-3 :6-dihydroxy-1 :Pbenzoquinone) is a bright red crystalline powder, which is sparingly soluble in water and forms solutions that are also intensely red. The calcium salt of the acid is almost insoluble in cold water and the colour diminution that accompanies its precipitation has been used as the basis of a colorimetric method for the estimation of this metal. Barrettol described both gravimetric and colorimetric methods, but the gravimetric determination shows no advantages over the established oxalate procedure and has received little further attention.The colorimetric method has, however, been the subject of several paperszS3t4 that are concerned with the estimation of calcium in plant materials and soil extracts. Chloranilic acid is not a highly selective reagent under the conditions for the determination of calcium, and interference is extensive. Tyner,2 Gammon and Forbes3 and Le Peintre4 have investigated the influence of the ions that are of common occurrence in plant ash and soil extracts, and we have extended the investigation with a view to broader applications. The classification of types of interference given by Tyner is of some general interest. Ions such as Fe"' and Al"' form soluble complexes that hinder or prevent precipitation of the calcium salt; and barium, strontium, copper and manganese interfere by producing insoluble compounds.Another group of metals, including sodium, potassium and magnesium, were said to interfere because of the occlusion of the metal-chloranilate complexes by the calcium salt, Later work, however, does not support the occlusion mechanism. Gammon and Forbes3 failed to detect significant amounts of magnesium in calcium precipitates by a spectro- graphic method, and the present authors observed colour diminution with all three metals in solutions containing no calcium. The results of our experiments on general interference are summarised in Table I. Where toleration limits are given these are generally in substantial agreement with those previously published.,2 but the magnitude of the interference by sodium proved to be much greater than was expected. Table I shows the effect of foreign ions present at a concentration of 10 mg per 25 ml of solution containing 5 ml of 0.1 per cent.aqueous.chloranilic acid and compared visually with a blank containing chloranilic acid only. The items marked with an asterisk are discussed more fully in the following section. It will be seen that some kind of direct interference was observed with almost every ion examined, but the amounts concerned were equivalent to about ten times the average amount of calcium that would be present in the same volumes in an analysis. Certain ions are discussed in more detail below.Sept., 19521 OF CHLORANILIC ACID 469 THE QUANTITATIVE EFFECT OF CERTAIN COMMON IONS- AZzcmiizizcm-Although aluminium does not noticeably affect the colour of chloranilic acid solutions, the presence of 10 mg almost completely inhibits the precipitation of calcium at levels of about 0.5 mg.Amounts of aluminium somewhat smaller than that of the calcium present can be tolerated without very serious effects. TABLE I INTERFERENCE EFFECT OF FOREIGN IONS Foreign ion Aluminium* .. Ammonium* .. Antimony (Sb'") . . Arsenic (As04''') . . Barium .. .. Beryllium . . .. Bismuth . . .. Cadmium . . .. Chromium (CrO,") Cobalt .. Copper* (:!*') . . Iron* (Fe ) . . Iron (Fe") . , 1. Lead .. .. Magnesium* Manganese* (Mn**f * Mercury (Hg') . . Molybdate Nickel . . . Palladium . . .. Potassium*. . .. Silver . . .. Sodium" . . .. Strontium .. .. Thallium (TI') . . Thorium .. .. Tungsten (WO,") . . Uranium (UO,") . . Zinc .. .. Zirconium .* . .. .. .. Acid radicles- Oxalate . . .. Tartrate . . .. Citrate . . .. Cyanide .. .. Sulphate and chloride .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. * I .. .. .. .. .. .. .. .. .. .. Effect No detectable difference from the blank Little or no effect Very slight colour change Considerable diminution of intensity Precipitate Deeper red colour Green colour and precipitate Precipitate Yellow colour Precipitate Some colour diminution and production of violet tint Very deep purple - mauve colour Similar to (ironII1) but less intense (possibly traces of ferric iron) Precipitate Considerable colour diminution Precipitate Precipitate Considerable colour diminution Violet colour Deep straw colour Slight colour diminution Precipitate Slight colou r diminution Precipitate Slight colour diminution Brilliant violet colour Considerable colour diminution Brown colour Pronounced colour diminution with change of tint Flocculent precipitate Considerable colour diminution Slight colour diminution Some colour diminution Some colour diminution No interference apparent * These are discussed more fully in the text.Magnesizcm-Amounts approaching that of the calcium to be determined can be tolerated. The presence of 10 mg of magnesium ion largely decolorises the solution used in the general method for the estimation of calcium (p. 470). Le Peintred minimised interference by means similar to those adopted by us for cancelling the effect of sodium (see 470).Iron (Fee")-Up to about 0.01 mg of ferric iron is tolerable in the estimation of amounts of calcium of the order of 1 mg. Amounts of iron comparable with the calcium present vitiate the method chiefly by intensifying the colours considerably. The shade of the colour is also modified. Gammon and Forbes3 drew attention to the mutually compensating effect observed when both iron and magnesium are present in soil extracts or plant materials. They recom- mend the prior determination of these constituents with o-phenanthroline and thiazole yellow, respectively, so that additions can be made to the standards. The successful applica- tion of this procedure would seem to be limited to cases where the iron level is not high. Attempts were made to suppress the iron colour by complexing with various reagents, namely citrate, tartrate, cyanide, pyrophosphate and thioglycollic acid.Of these, tartrate, had little effect, citrate and cyanide materially reduced the colour intensity of chloranilic470 FROST- JONES AND YARDLEY: THE COLOUR REACTIONS [Vol. 77 acid solutions ; pyrophosphate suppressed the iron colour but, a s anticipated, it also inhibited precipitation of the calcium salt. Thioglycollic acid, in acid solution, was only partially effective in reducing the iron colour unless present in quantities that gave rise to extensive fading of the reagent, presumably through reduction of the chloranilic acid. Co@$er-Amounts up to about one-quarter of the calcium present are tolerated, but larger amounts, of the order af ten times the amount of calcium present, cause considerable colour diminution together with change of tint. Manganese-Amounts up to 0.025mg have no effect on the determination of calcium at a level of about 0.5 mg.Sodium-The colour intensity is continuously and rapidly decreased by increasing amounts of sodium up to 10 mg, but beyond this level the rate of decrease falls off rapidly. Thus, the effect of 500 mg of sodium is barely double that of 10 mg. Moreover, in the range of 0 to 10 mg of sodium, the effect on the chloraniiic acid coluur is independent of the presence of calcium, so the interference is not due to occlusion. These findings led to the supposition that the influence of moderate amounts of sodium could be effectively cancelled by adding a large excess of sodium ions (as chloride) to both test solutions and standards alike.This modification proved satisfactory; some typical results are shown in Table 11. TABLE I1 THE DETERMINATION OF CALCIUM AFTER THIS ADDITION OF 400mg OF SODIUM ION TO THE TEST SAMPLE Present in the test sample Calcium, Sodium, mg mg 0-60 0.5 0.70 10.0 0.80 1.0 0.90 5.0 1.00 20.0 1.20 10.0 1.40 1.0 f A .I, AND STANDARDS Calcium found, mg 0-57 0.70 0.81 0-92 1-03 1-20 1.38 The average effect of 20mg of sodium on the recovery of 0.5mg of calcium, by the general method (see below), was to introduce a positive error of about 25 per cent. Potassium-In the presence of 100 mg of potassium the colour intensity curve was almost identical with that given by 100 mg of sodium in the same range of calcium concentrations.Although no further quantitative experiments were made, there is every likelihood that the effect of potassium is altogether similar to that of sodium and might be minimised in a similar way. Ammonium-Large amounts of ammonium i o n led to colour diminution, but no quantita- tive experiments were made, as this ion could normally be removed by ignition or be replaced by sodium. GENERAL METHOD FOR THE ESTIMATION OF CALCIUM The following procedure, based on Tyner’s method,2 has proved convenient in practice, and is suggested as a basis for the more general application of the reagent to calcium estimations. To 10 ml of test solution, made neutral or faintly acid with acetic acid, containing 0.2 to 1-5 mg of calcium, in a 25-ml standard flask, add 10 ml of a 0-1 per cent.aqueous solution of chloranilic acid (filtered if necessary) from a pipette. Shake well and set aside overnight (or for 3 hours), preferably in an ice-chest. Let the solution attain room temperature, dilute to the mark with distilled water and filter through a dry Whatman No. 41 filter-paper. As an alternative to filtration, an aliquot of the solution (after dilution to the mark) may be centri- fuged at a relative centrifugal force of about 5000 for 5 minutes. A suitable portion of the supernatant liquor can then be transferred to the absorptiometer cell for measurement of its optical transmittance. Finally, measure the colour intensity of the filtrate at a wavelength of 550mp. If a Spekker absorptiometer is used, either the green or yellow-green filters should be used with a l-cm cell and a water setting of one on the drum.Two standards should be included in each batch of test samples (p. 471).Sept., 19521 OF CHLORANILIC ACID 471 The transmittance of chloranilic acid solution is said to vary slowly but continuously.2 Temperature, pH value and other factors also have some effect on the colour intensity. The possibility of minimising calibration variations by rigid control of the conditions of experiment was considered, but as two suitably chosen standards (say 0.3 and 0.9mg for the range 0 to 1 mg of calcium) are sufficient to define a calibration curve, this further com- plication of the method appeared to be unnecessary. All the calibration curves prepared were linear and, apart from the waiting period, the method is rapid.A large number of test samples can be handled at a time, so the inclusion of an additional pair of standards is probably less troublesome than the introduction of refinements in general working technique. The solid reagent is stable over a period of years. Le Peintre4 recommended working at a pH value between 4 and 5, both in the presence and absence of magnesium, but we did not find it necessary to take any steps beyond ensuring approximate neutrality in the unknown solution. Although no work was done outside the range of 0.1 to 2 mg of calcium, the use of larger amounts of chloranilic acid would enable larger amounts of calcium to be estimated, but the increased amount of reagent would not be suitable for the lower calcium ranges because of the intense residual colours that would result.The use of smaller amounts of chloranilic acid is advantageous when working towards the lower end of the recommended range, but below this range the usefulness of the reagent is limited by the appreciable solubility of the calcium salt. The effect of variables such as temperature and pH value also becomes more marked at low concentrations of calcium and of the reagent. MODIFICATION OF "GENERAL METHOD" IN THE PRESENCE OF SODIUM- When sodium is present in the test solutions, the general method may be followed, except that 400 mg of sodium should be added both to the test solutions and to the standards. This addition may conveniently be made by minimising the volume of the aliquots initially taken and adding 10 ml of a 10 per cent.aqueous solution of sodium chloride before adding the chloranilic acid. Typical results obtained by this modified procedure have been given in Table 11. Le Peintre4 minimised interference from magnesium in a similar manner by adding a large excess of magnesium sulphate. THE REACTION WITH ZIRCONIUM In contrast to its behaviour in the calcium reaction, chloranilic acid yields a relatively selective reaction for zirconium in the presence of perchloric acid. The magenta colour produced was studied by Thamer and Voigt5 who, in a recent paper devoted largely to physico- chemical aspects of the reaction, outlined a method for estimating zirconium at concentrations between 2 x 10-6 and 5 x Intensities were measured at 330 mp and, in 2 M perchloric acid, interference from Hf"", U"", Th"'., Sn"", Ti'"' and Fe"' was mentioned. No visible reaction was reported with Fe", Cr'**, AY**, Cu", Co", Mn, Ba or K.The following summary of our experiments, carried out in the visible waveband, confirms a good degree of selectivity and moderate sensitivity under these conditions. molar. PROCEDURE- Transfer a slightly acid aliquot (not more than 25ml) of the test solution, containing not more than 1.5 mg of -zirconium, to a 50-ml calibrated flask, and add precisely 20 ml of 4 M perchloric acid. Mix, dilute with water to about 45 ml and add 4 ml of a 0.1 per cent. aqueous chloranilic acid solution. Transfer the flask to a constant temperature bath at 20" C for at least 20 minutes, dilute to the mark and measure the colour intensity at a wavelength of 525mp.If the Spekker absorptiometer is used, a 4-cm cell, tungsten lamp and Ilford No. 604 green filter and a water setting of 1 are convenient. When a large number of readings are to be taken, the above method may be somewhat simplified by preparing a composite reagent from the 4 M perchloric acid by diluting 500 ml with 100 ml of 0.1 per cent. chloranilic acid solution and 25 ml of water. Twenty-five- millilitre portions of this solution can then be added instead of the two separate additions, but the solution does not keep well. A calibration curve should be constructed with known amounts of zirconium. For this purpose "pure" zirconium nitrate can be used. This salt is usually basic (more basic than472 FROST- JONES ANT) YARDLEY [vol.77 would correspond to the zirconyl salt) but the zirconium content can be readiLy estimated by ignition to ZrO,. It is possible to estimate rather larger amounts of zirconium; but beyond the limit given above, calibration curves were not always satisfactory. Precipitation never occurred within a reasonable time, but the coloured solutions were not always clear and intensities were often irregular within the range 1.5 to 2.0nig of zirconium. INTERFERENCES- The colour intensities of a number of solutions containing 1 mg of zirconium plus 1 mg of foreign ions were measured and compared with the intensities of solutions containing 1 mg of zirconium alone. With water settings of 1 drum unit, solutions containing the following ions gave Spekker readings within 0.01 unit of those given by the pure zirconium solutions-NH,, Na, K, Mg, Ca, Sr, Ba, Cu", Pb, Zn, T1*, As"', Hg', Al, Co, Cr"' and Mn".Slightly greater differences were given by Bi, Ni, Pd, Th, UO," and tartrate, whilst Sb"', W04", Moo4", Fe"', Sn", Sn"", Cd and Ag gave rise to serious interference of varying degrees. The effect of chromate, oxalate and quadrivalent titanium was considerable. The concentration of perchloric acid has a marked effect on the colour intensity of zirconium chloranilate solutions ; this intensity decreases quite sharply over the concentration range 0 to about 3.5 molar perchloric acid (higher acid concentrations were not examined). Unfortunately, interference effects are much more pronounced at low perchloric acid con- centrations. Iron, for example, when present in quantity equal to that of the zirconium to be determined, has no effect at perchloric acid concentrations of 3.5 molar, but its effect soon becomes apparent as the perchloric acid concentration is diminished. The iron colour completely obscures the reaction its the perchlorate concentration approaches zero. Con- ditions then become virtually identical. with those prevailing in- the calcium determination and interference generally becomes widespread. The acidity level finally chosen is merely a working compromise and individual circum- stances, such as those considered above, might demand higher acid concentrations. Lower acid concentrations, with consequent gain in sensitivity, might equally be desirable in the absence of interfering ions. Thanks are due to the Directors of Hopkin &: Williams Ltd. for permission to publish this paper. REFERENCES 1. Barretto, A., Bol. SOC. Brad Agron., 1945, 8, 351. 2. Tyner, E. H., Anal. Chem., 1948, 20, 76. 3. Gammon, N., and Forbes, R. B., Ibid., 1949, 21, 1391. 4. Le Peintre, M., Compt. Rend., 1950, 231, 968. 5. Thamer, B. J., and Voigt, A. F., J . Amer. Chem. Soc., 1951, 73, 3197. HOPKIN AND WILLIAMS LTD. ANALYTICAL LABORATORY CHADWELL HEATH March, 1962

ISSN:0003-2654    

DOI:10.1039/AN9527700468

出版商:RSC    

年代:1952

数据来源: RSC

摘要:

Sept., 19521 GLOVER AND HODGSON 473 The Polarography of cyclooctatetra-ene and Vinyl cyclooctatetra-ene BY J. H. GLOVER AND H. W. HODGSON The polarography of cyclooctatetra-ene and of a new derivative, vinyl cyclooctatetra-ene, has beeh examined. Previously published work by Elofsen on cyclooctatetra-ene has been confirmed, and it has been shown that the recently isolated vinyl derivative is almost identical in polarographic behaviour with the parent compound. Both compounds give waves at the same half-wave potential, which is independent of the pH of the solution. It has also been found that cis-1-phenyl-1 : 3-butadiene is polarographically inert. CYCZOOCTATETRA-ENE is a highly reactive hydrocarbon that has potential uses in organic synthesis. Its reduction at the dropping-mercury electrode has been examined by Elofsen: who showed that polarographic waves, suitable for analytical use, are obtained in base solutions containing tetramethylammonium ions.The half-wave potential was - 1.51 volt, measured against the saturated calomel anode, and it was found to be independent of the pH of the base solution. On the basis of an analysis of the cyclooctatetra-ene wave, Elofsen concluded that two electrons are involved in the electrode reaction. The formation of negatively charged cyclooctatetra-ene ions was suggested as the primary electrode process, these ions being stabilised by the tetramethylammonium ion, which appears to be essential for the reduction. During experimental work involving the use of Elofsen’s method, it was found that a polarographic wave was obtained from an unidentified fraction containing CloHlo homologues of cyclooctatetra-ene.The wave was in all respects identical to the cyclooctatetra-ene wave, but evidence from other sources indicated that there was no cyclooctatetra-ene present. Recently, Craig and Larrabie2 and Withey3 have identified the constituents of this CloH,, fraction as vinyl cyclooctatetra-ene and cis-l-phenyl-l:3-butadiene. The polarographic behaviour of these two compounds has been examined, and it has been shown that vinyl cyclooctatetra-ene is polarographically reducible, and would therefore constitute a source of interference in Elofsen’s method. The calibration was carried out with purified cyclooctatetra-ene and vinyl cyclooctatetra- ene, and the accuracy is consistent with that of normal direct polarographic methods.CYCZOOCTATETRA-ENE Efect of pH-The polarographic behaviour of cyclooctatetra-ene was examined in a series of buffer solutions covering the pH range 5 to 13. The buffer solutions were prepared by adding a dilute solution of citric or boric acid in 50 per cent. alcohol to 0.2 M tetramethyl- ammonium hydroxide solution. An examination of the graphs of pH value plotted against neutralisation of tetramethyl- ammonium hydroxide shows that neutralisation with citric acid gives rise to a well-buffered solution over the range pH 4.0 to pH 7-5, and neutralisation with boric acid covers the range 9.5 to 11.0. Tetramethylammonium hydroxide is buffered at pH 13.0. In practice, the buffer solutions were prepared by titrating tetramethylammonium hydroxide with the acid until the desired pH was reached, the glass-electrode system being used as a continuous pH indicator.The pH of each solution was checked after mixing with the cyclooctatetra-ene solution. The effect of pH on the half-wave potential of the cyclooctatetra-ene wave confirms Elofsen’s findings, in that no variation is obtained from pH 7.5 to 13.0. The actual values are shown in Table I. No well-defined diffusion current was obtained at pH 5.3 with either compound; it is probable that the hydrogen wave interferes at this pH. The slope of the limiting current increases with decreasing pH, and definition of the wave was best at a pH of 13-0. Calibration experiments showed that the diffusion current bore a linear relationship to the concentration of cyclooctatetra-ene, and details of the calibration constants are shown in Table 11.EXPERIMENTAL474 GLOVER AND HODGSON THE POLAROGRAPHY OF [Vol. 77 VINYL CYC~OOCTATETRA-ENE- This compound was examined in a similar series of base solutions to that used for cyclo- octatetra-ene, in an attempt to separate the waves due to the two compounds. It was found, however, that no separation was possible, since the half-wave potential of vinyl cyclooctatetra- ene was also independent of pH and was within 0.05 volt of the cyclooctatetra-ene wave over the pH range 7.5 to 13.0. At pH 5.3, no well defined diffusion zone was obtained, and the same variation in the slope of the limiting current was given as with cyclooctatetra-ene.Half-wave potentials of these two substances at different pH values are presented in Table I. A linear relationship between diffusion current and concentration for vinyl cyclooctatetra-ene TABLE I[ EFFECT OF pH ON HALF-WAVE POTENTIALS Half-wave potential us. S.C.E. f - pH cyclooctatetra-ene, Vinyl cyclooctatetra-ene, 5.3 7.5 - 1.53 - 1.49 10.0 - 1.53 - 1.49 11.0 - 1.52 - 1.49 13.0 - 1.50 - 1.49 volts volts - - was found in calibration experiments; details of tlhis calibration are compared with those of cyclooctatetra-ene in Table 11. An attempt was made to estimate vinyl cyclooctatetra-ene in the presence of cyclo- octatetra-ene by acid hydrolysis of the vinyl group to acetaldehyde. Previous experiments on mixtures of vinyl cyclooctatetra-ene and acetaldehyde showed that it is possible to estimate the two compounds polarographically in tetramethylammonium hydroxide base solutions, at a pH value of 13.Well-defined acetaldehyde waves were obtained (Fig. 1). Hydrolysis TABLE 11: CALIBRATION DATA FOR CYC~OOCTATETRA-ENE AND VINYL CYC~OOCTATETRA-ENE cycZoOctatetra-ene m&9, 2-01 mg% sec.-& at 20' C A I \ Concentration, Diffusion current, 2.0 0.80 4.0 1.56 6.0 2-36 8.0 3.14 10.0 3-92 mg per 100 ml Pa Diffusion current constant4-2-64 Vinyl cyclooctatetra-ene nztt6, 2.35 mg% sec.-t at 20' C &centration, Diffusion curre&, 2.0 1-00 4.0 2.04 6.0 3.16 8.0 4.12 10.0 5.16 A mg per 100 ml Pa 2-28 with hydrochloric acid, for times varying from five minutes to two hours, failed to produce any acetaldehyde; there was a variable reduction of the vinyl cyclooctatetra-ene wave after this treatment owing to polymerisation to the dinier.No basis for a separation of the two compounds appears to be possible on these lines. Cis-1-PHENYL-1: 3-BUTADIENE- This compound, which occurs in the C1,H,, fraction from cyclooctatetra-ene, was found to be non-reducible at the dropping-mercury electrlode over the whole of the pH ranges tried. Tetrahydrofurfuryl alcohol, diisoamyl acetal, and diethylcarbitol, which are solvents for cyclooctatetra-ene, were also examined and found to be non-reducible. The criteria of purity of the compounds examined are- cyclo0ctatetra-ene-Redistilled; nE5 = 1.5347. (E1ofsen;ln;' = 16342 ; Benson and Cairns,6 Vinyl cyclooctatetra-ene-Freshly distilled material was used for all experiments, as the pure material polymerises slowly on standing, to form an orange glass; nE6= 1-5695.B.p. 83" C at 21 mm of mercury (Craig and Larrabee;2 nz5 = 1.5682, b.p. 83.3" C at 20 mm of mercury). 72:' = 1.5350.)Sept., 19521 CYC~OOCTATETRA-ENE AND VINYL CYC~OOCTATETRA-ENE 475 METHOD The following method is proposed for the estimation of small concentrations of cyclo- oct at etra-ene or vinyl cyclooct atetra-ene. REAGENT- Base solution-Dilute 9.1 g of tetramethylammonium hydroxide solution (10 per cent. in water as obtained from .the suppliers) to 100 ml with 50 per cent. v/v of aldehyde-free ethanol. + \ - c.c I 0.8 x I ( -I -7 Voltage Fig. 1. Polarogram of vinyl cyclooctatetra-ene and acetaldehyde in tetramethylammonium hydroxide base solution PROCEDURE- Prepare a solution of the sample in aldehyde-free ethanol so that 5ml contains up to 2 mg of cyclooctatetra-ene.Place exactly 5 ml of the base solution into a dry container, by means of a pipette, and then 5 ml of the sample solution. Mix and transfer to the polarographic cell and remove oxygen by passing a stream of nitrogen for 5 minutes. No significant loss of cyclooctatetra-ene occurs during this period. Polarograph over the range - 1.2 to - 1.9 volts, measured against the saturated calomel electrode, and calculate the cyclooctatetra-ene content by reference to the calibration curve. This method will give a measure of the combined cyclooctatetra-ene and vinyl cyclooctatetra-ene contents of the samples, if both compounds are present. DISCUSSION Vinyl cyclooctatetra-ene and the parent compound cyclooctatetra-ene showed almost identical behaviour at the dropping-mercury electrode.The electrode reaction appears to be reversible with both compounds, and a linear relation exists between cathode potential and log;/(& - i). By applying values obtained from this relation, the number of electrons476 MAYER AND BRADSHAW : THE ABSOFLPTIOMETRIC DETERMINATION vd. 77 involved in the reduction has been shown to be two) with both compounds; the relevant values for n fall between 1.9 and 2.1 in all the experiments. Elofsen’s findings with respect to the independence of half-wave potential and pH have been confirmed for cyclooctatetra-ene, and support has been gained for his theory of the electrode reaction. All the compounds examined were prepared in the Research Laboratory. The authors are indebted to the Directors of the British 0x:ygen Company for permission to publish their findings. REFERENCES 1. 2. 3. 4. 5 . Elofsen, R. M., Anal. Chem., 1949, 21, 917. Craig, L. E., and Larrabie, C. E., J . Amer. Chem. Soc., 1951, 73, 1191. Withey, D. S., J . Chem. Soc., 1952, 1930. Lingane, J. J., Ind. Eng. Chem., Anal. Ed., 1943, 15, 583. Benson, R. E., and Cairns, T. L., J . Amer. Chenz. Soc., 1950, 72, 5355. ANALYTICAL LABORATORY RESEARCH AND DEVELOPMENT DEPARTMENT BRITISH OXYGEN COMPANY, LIMITED LONDON, S.W.19 April, 1962

ISSN:0003-2654    

DOI:10.1039/AN9527700473

出版商:RSC    

年代:1952

数据来源: RSC