ISSN:0003-2654    

DOI:10.1039/AN95277FX025

出版商:RSC    

年代:1952

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN95277BX027

出版商:RSC    

年代:1952

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN95277FP073

出版商:RSC    

年代:1952

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN95277BP079

出版商:RSC    

年代:1952

数据来源: RSC

摘要:

JULY, 1952 THE ANALYST Vol. 77, No. 916 PROCEEDINGS OF THE SOCIETY OF PUBLIC ANALYSTS AND OTHER ANALYTICAL CHEMISTS AN Ordinary Meeting of the Society was held at 7 p.m. on Wednesday, May 7th, 1952, in the Meeting Room of the Chemical Society, Burlington House, London, W.l. The Chair was taken by the President, Dr. J. R. Nicholls, C.B.E., F.R.I.C. The following papers were presented and discussed : “A Routine Method for the Analysis of Table Jellies,” by Miss E. M. Chatt, B.Sc., F.R.I.C.; “The Determination of Oxalates in Fresh Plant Material,” by C. J. L. Baker, A.R.I.C. ; “The Determination of Small Quantities of Ammonium Di- or Tri-ethanolamine Alginate in Rayon-Finishing Solutions and on Rayon Yarn,” by E. G. Brown, A.M.C.T., F.R.I.C., and T. J. Hayes. NEW MEMBERS Mary Andross, B.Sc., F.R.I.C.; Arthur Derek Campbell, M.Sc.(N.Z.), A.N.Z.I.C.; George Valentine Francis, B.Sc., P1.D. (Liv.) ; Karl Gunnar Mautitz Gran; Paul Geoffrey Jeffery, B.Sc. (Lond.), A.R.C.S. ; Alfred Henry Moore; Edwin William Stanley Press, R.Sc. (Lond.), F.R.I.C.; John Jeffrey Reid; Neville Stuart Thom, B.A. (Cantab.). SCOTTISH SECTION AN Ordinary Meeting of the Section was held on Friday, May Znd, 1952, a t 7 p.m. in the North British Hotel, Edinburgh. Mr. H. C. Moir presided and forty-two members and friends were present. A lecture entitled “Chemistry and the Law” was given by J. K. McLellan, M.A., B.Sc., A.R.I.C., and was illustrated by lantern slides and exhibits. A discussion followed. MICROCHEMISTRY GROUP A JOINT Meeting of the Group with the Bristol and District Sections of the Chemical Society, the Royal Institute of Chemistry and the Society of Chemical Industry was held in the Lecture Theatre of the Chemical Department of Bristol University on Wednesday, April 23rd, 1952, at 7 p.m.The following paper was presented and discussed: “The Use of Cylinder Oxygen in the Organic Micro-determination of Nitrogen,” by H. Swift and E. S. Morton. This was followed by an open discussion on “Standard Substances for Organic Micro-analysis.” During the afternoon, visits were made to the chocolate and cocoa works of J. S. Fry & Sons Ltd., Somerdale, and to the University of Bristol Agricultural and Horticultural Research Station at Long Ashton. PHYSICAL METHODS GROUP THE Thirty-seventh Ordinary Meeting of the Group was held at 6 p.m.on Friday, May 16th, 1952, a t the University College, Swansea. This was a joint meeting with the South Wales Section of the Royal Institute of Chemistry, and was preceded by a visit to the B.I.S.R.A. Research Laboratories, Swansea. Dr. J. Haslam was in the Chair and fifty-two members and visitors were present. The following papers on Ion Exchange Resins were presented and discussed : “The Theory of Ion Exchange,’’ by Professor C. W. Davies, DSc., F.R.I.C. ; “Some Newer Applica- tions and Techniques of Cation and Anion Exchange Resins in Chemical Analysis,” by 333334 NOTES [Vol. 77 G. H. Osborn, F.R.I.C., A.M.1nst.M.M. ; “The 13etermination of Individual Rare Earths by Radioactivation using Ion Exchange Separation,” by F. W. Cornish, Ph.D., A.R.I.C. ANALYTICAL METHODS COMMITTEE SUB-COMMITTEE ON METHODS OF ANALYSIS OF ICE-CREAM A SUB-COMMITTEE has been appointed to forrnulate methods of analysis of ice-cream in respect of the Ministry of Food Ice-Cream Oraier. The Sub-committee consists of J. H. Hamence, M.Sc., Ph.D., F.R.I.C. (Chairman) ; J. G. Davis, Ph.D., D,Sc., F.R.I.C.; G. E. Forstner, M.Sc., F.R.I.C.; J. King, O.B.E., F.R.I.C.; K. A. Williams, B.Sc., Ph.D., A.Inst.P., F.R.I.C. ; M. G. Read, B.Sc., F.R.I.C. (Honorary Secretary).

ISSN:0003-2654    

DOI:10.1039/AN9527700333

出版商:RSC    

年代:1952

数据来源: RSC

摘要:

334 NOTES [Vol. 77 0 bi tuar y JOHN FRANCIS HUT’CHINS GILBARD JOHN FRANCIS HUTCHINS GILBARD died in London at his home on April 24th, 1952, in his eighty-third year, sixty-two years after his election to membership of the Society. He was born in London and educated at Vermont College, Clapton, and at Finsbury Technical College, under Meldola and Streatfeild. He joined Bernard Dyer in 1888 and was his assistant during the remainder of his working life. From early in this period his chemical knowledge and great analytical skill were of immense assistance t o Dyer in the many original investi- gations that Dyer undertook. It is now nearly sixty years since the first contribution from Dyer and Gilbard appeared in The Analyst in 1893-dealing with the composition of genuine and spent ginger-and this was followed in 1895 by a paper on cinnamon and a review of the examination of over 1000 samples of linseed and other oil-cakes and feeding stuffs with respect to their natural content of free fatty acids.This latter was a detailed account of work of such a specialised character bearing on the question of deteriora- tion of feeding quality due to enzyme and mould activity that it is still commonly consulted when such questions arise in commercial practice. In 1896 a paper on the detection and determination of drawn or exhausted carawayys was published. Finally, in 1911, Gilbard gave a short account of his own work on a colour reaction for the active principle (resins) of caulophyllin, formerly used as an abortifacient. To those of us who knew Gilbard personally he was a strong character with firm and often unpredictable views on all sorts of subjects.So to Bernard Dyer he was occasionally very provocative by his obstinacy in holding views on chemical matters that were more or less diametrically opposed to Dyer’s. This led occasionally to somewhat stormy scenes, but as we, the weaker members of the staff, knew quite well that these would inxitably end happily in reconciliations of a frankly human! nature, we always enjoyed such occasions. But, in fact, he was a pioneer both in food ana!ysis and in bacteriology. In these days of intense activity in the field of antibiotics it is a memory of great interest that Gilbard in his earlier bacteriological work was occasionally bothered by contamination of his plates and then speculated, with what we now recognise as a prophetic foresight, on the odd effects of mould growth.With the outbreak of the second world war, Gilbard’s strength began to give way. Yet he was extraordinarily resilient; for not only was he an active air-raid warden during the war, but in the years that immediately followed he underwent two major operations from which he recovered apparently almost unaffected. To the end he retained his mental faculties, and, on the last occasion I saw him, only a short time ago, he was quietly humorous about his resemblance to Einstein; a likeness that had been noted some years ago and had slowly become more and more marked. Gilbard was elected to the Society in 1890, he became a Fellow of the Chemical Society in 1895, and a Fellow of the Institute of Chemistry-as it then was-in 1899. Sometime about this latter period he was appointed a Gas Examiner to the London County Council and later, additionally, to West Ham. He married twice, first Thirza Hawke in 1!302, and secondly Rita Johnson. GEORGE TAYLOR

ISSN:0003-2654    

DOI:10.1039/AN9527700334

出版商:RSC    

年代:1952

数据来源: RSC

摘要:

July, 19521 CHATT 335 A Routine Method for the Analysis of Table Jellies BY E. M. CHATT (Presented at the meeting of the Society on Wednesday, M a y 7th, 1952) A speedy routine method for the analysis of table jellies is presented, together with analytical data that form the basis of equations used to calculate the percentages of sucrose, invert sugar and glucose from the optical rotation and reducing power of clarified solutions. ALTHOUGH no new principles are involved in the following description of a routine method for the analysis of table jellies, data are presented that form a basis for calculating the per- centages of the individual sugar components. GENERAL PROCEDURE The direct refractometer reading should be determined on several sections taken from the interior of the tablet.The sugar equivalent of such a reading is sometimes about 1.5 per cent. too high, owing to the interference of gelatin, citric acid and non-sucrose sugars. The appropriate corrections to be applied for the calculation of the true soluble solids are presented in Table I. Only the uncorrected reading is required by the Statutory Instrument. TABLE I CORRECTIONS TO BE APPLIED TO APPARENT SOLUBLE SOLIDS BY DIRECT REFRACTOMETER READING WHERE CERTAIN PERCENTAGES OF CONSTITUENTS OTHER THAN SUCROSE ARE PRESENT Constituent Invert sugar .. Glucose dry solids . . Citric acid . . .. Tartaric acid .. Gelatin . . .. .. .. .. .. .. .. Amount of constituent in sample, /O 20 40 50 60 70 75 20 40 1 2 0' .. .. 1 2 .. .. 5 10 Correction when amount of soluble solids is approximately f 60 % + 0-4 + 0-75 + 0-95 + 1.15 - 0.3 - 0.55 + 0.1 + 0.2 +0.1 + 0.25 - 0.85 - 1.7 70% + 0.45 + 0.9 + 1.1 + 1.35 + 1.55 - 0.25 - 0.5 +0.1 + 0-2 +0*1 + 0.25 - 0.8 - 1.6 75 % + 0.5 + 1.00 + 1.25 + 1.5 + 1.75 - 0.25 - 0.5 +0*1 + 0.2 +om1 + 0.25 - 0.75 - 1.5 85 ./o' + 0.6 + 1.2 + 1.5 + 1.8 +2*1 + 2-55 - 0.25 - 0.45 +0.1 + 0.2 f0.1 + 0.25 - 0.7 - 1.4 The sample is prepared for analysis by cutting it into small pieces weighing not more than 0-5g.Gelatin is determined by the Kjeldahl method on about 2-5g of sample and the factor 5.55 is used to convert nitrogen to gelatin protein. For the determination of the percentages of soluble solids by specific gravity measurement and of acid and sweetening components, a 10 per cent. w/v solution is prepared by weighing 25 g of the sample into a beaker and dissolving by warming it to between 40" and 50" C with 75 ml of distilled water in as short a time as possible to prevent inversion of the sucrose.The solution is transferred to a 250-ml flask with cold distilled water, cooled to 20" C and then diluted to the graduation mark with distilled water at 20" C and mixed. If required, the approximate concentration of soluble solids in the 10 per cent. solution can be calculated from the specific gravity at 20" C by subtracting 1 from the specific gravity and dividing by a common solution factor, 0.00386. As this factor is exact for sucrose at 15.5" C only and within a limited range of concentration, the appropriate corrections to be336 CHATT: A ROlJTINE METHOD Wol. 77 applied for temperature and concentration are shown in Table 11.Further corrections for the respective concentrations of acid, gelatin, glucose dry solids and invert sugar in the 10 per cent. solution of the sample are shown in Table 111. TABLE I1 CORRECTIONS TO BE APPLIED TO SOLUBLE SOLIDS WHEN SOLUTIONS ARE MADE .AT 20°C Approximate specific Correction, grams per gravity 20" C/20° C 100 ml of 10% solution 1.023 + 0.01 1.031 + 0.02 1-038 + 0.03 TABLE I11 CORRECTIONS TO B E MADE FOR SUBSTANCES OTHER THAN SUCROSE I N 10 PER CENT. SOLUTIONS OF TABLE JELLIES Approximate amount of constituent Correction, grams per 100 ml Constituent in original sample, of 10% solution Y O Invert sugar . . .. .. 20 0.0 40 - 0.01 60 - 0.01 Glucose dry solids . . .. Citric acid .. .... Tartaric acid . . .. .. Gelatin . . . .. .. .. 20 40 0.5 1.0 0.5 1 4 5 10 - 0.04 - 0.07 0.0 - 0.01 - 0.01 - 0.02 + 0.09 +0-17 The acid content of the sample can be calculated from the volume of 0.1 N sodium hydroxide solution required to neutralise 50 m:l of the 10 per cent. solution, with phenol- phthalein as indicator. As a rule, there is no difficulty in seeing the end-point in coloured solutions. REAGENTS- chloride and of a 10 per cent. solution of phosphotungstic acid. METHOD OF DETERMINING 'THE SUGAR COMPONENTS Phosphotungstic acid reagent-Mix equal volumes of a 10 per cent. solution of sodium Hydrochloric acid, 6-34 N. Fehling's solution-Prepare and accurately standardise as described in Lane and Eynon's meth0d.l The addition of 1 ml of N sul;phuric acid to each litre of the original copper sulphate solution retards the formation of a deposit during storage.Methylene blue-A 1 per cent. aqueous solution. PROCEDURE- Prepare a 7.5 per cent. clarified solution of the sample by placing exactly 150 ml of the 10 per cent. solution into a 200-ml flask. Add 40 ml of the phosphotungstic acid reagent (50 ml may be required if the percentage of gelakin exceeds 10) and dilute the solution with water to 200 ml at 20" C. Shake thoroughly and filter through a fluted 15-cm Whatman No. 4 or No. 1 filter-paper. Filtration should be rapid and the filtrate bright and clear. By measuring the optical rotation, at 20" C, of the filtrate so prepared from a commercial sample of table jelly at intervals over a period of time, it has been established (Table IV) that no appreciable inversion of sucrose occur!j within 3 hours of filtering.Determine the optical rotation of the filtrate a t 20" C and calculate therefrom the specific rotation, D, of the sample. Determine the optical rotation after inversioii as follows. Place a 100-ml flask containing 80 ml of the filtrate and 10 ml of 6.34 N hydrochloric acid in a water-bath at 62" to 63" C. Bring the solution to 60" C in 2 minutes and maintain it at that temperature for 10 minutes.2July, 19521 FOR THE ANALYSIS OF TABLE JELLIES 337 Then cool the solution to 20" C, dilute it to 100 ml and set it aside for 1 hour before taking polarimeter readings at 20" C. If the acid solution becomes opalescent on dilution to 100 ml, it should be cleared by shaking it with 0.1 g of Filtercel and filtered before polarimeter readings are taken.TABLE IV POLARIMETER READINGS AT 20°C ON THE FILTRATE FROM A 7.5 PER CENT. Finally, calculate the specific rotation of the sample. CLARIFIED SOLUTION OF A SAMPLE OF TABLE JELLY Time after filtration, Mean of readings Mean of readings hours in 2-dm tube in 4-dm tube 1 +6*07" + 12.15" 2 + 6-08" + 12-16' 3 + 6.09" + 12.16" 0 + 6.08' - Prepare a 1.5 per cent. solution of the sample, by neutralising 40 ml of the uninverted filtrate with 2 N sodium hydroxide solution and diluting with water to 200 ml, for the purpose of determining the percentage of reducing sugars in terms of invert sugar by the method of Lane and Eyn0n.l As the percentages of sucrose and reducing sugar in table jellies vary over a wide range, it is not always appropriate to use 25 ml of Fehling's solution and in some determinations where 10 ml are used it may be necessary to read from the column headed "1 g of sucrose per 100 ml." Some operators may prefer to use the constant volume titration method3 and not the tables.CALCULATION- The percentages of the sugar components can be calculated from the following equations- D - 1 Sucrose, S, = ~ 0.884 ' where 0.884 is the inversion divisor factor applicable to sodium illumination and angular degrees. The part of D, the specific rotation before inversion, that is contributed by sucrose can be calculated by multiplying S by 0.665. The algebraic sum, A, of the rotation due to invert sugar and glucose is represented by D - 0.665s. If B = the percentage of reducing sugars in the sample, expressed as invert sugar, X = the percentage of glucose dry solids in the sample, Y = the percentage of invert sugar in the sample, G = the specific rotation of glucose dry solids, R = the reducing power of glucose dry solids and it follows that:- -20 = the specific rotation of invert sugar under the above conditions of preparation of solution, .. (1) .. (2) GX - 20Y = lOOA . . .. . . . . R X + Y = B . . . . . . . . . . If the glucose used is available for analysis, the factors for G and R can be substituted in the above equations. Otherwise, the mean values G = +144" and R = 46 per cent., calculated to dry solids, for recent samples delivered over a period of 18 months for analysis at the B.F.M.I.R.A. laboratories (see Table V) can be substituted in equations (1) and (2).These mean values are. in fairly close agreement with the corresponding values, +143.4" and 43-9 per cent., for a range of glucose samples received in 1924. Hence 144X - 2OY = lOOA . . .. . . .. . . (1)' . . . . (2)' 0.46X + Y = B . . . . .. .. Equation (2)' can be expressed as- 144X + 317Y = 317B and by subtracting equation (1)' from it and solving for Y , .. (3) . . .. (4) 337Y = 317B - lOOA . . .. . . X = 2.2(B - Y) . . ..338 CHATT: A ROUTINE METHOD Pol. 77 Equations (3) and (4) can be applied to the calculation of invert sugar and glucose dry solids in the sample. Finally, apply the factor 0.994 for the volume of materid precipitated in clarifying. This factor was derived by comparing the angular rotation of the filtrate from a clarified solution, A, containing initially 6.6 of sucrose and 0.8 g of gelatin per 100 ml, with that of another solution, B, prepared by diluting a 13.2 per cent.solution of sucrose to twice its volume with the filtrate from a 1.6 per cent. solution of gelatin alone, which had Brand 1 2 TABLE V POLARISATION AND REDUCING POWER OF SAMPLES OF CONFECTIONERS' GLUCOSE (DR'I! BASIS) Polarisat ion, + 146.3' 140.6 143.0 143-9 142.0 141.6 141-6 143.2 144.6 140-7 Mean +142.7 + 146.8" 142.2 143.6 144.6 143.4 143-2 140.0 Mean $143.4 Reducing power as percentage of invert sugar 45.6 47.4 46.8 47.2 46.0 47.7 44.7 46-9 46.9 48.0 46-6 47.2 46.3 46.7 46-3 46-5 46.5 47.4 46.4 Polarisa tion, Brand 3 + 144.0' 143.4 147.6 146.2 138.2 143.2 143.1 Mean +143.7 4 + 146.6' 143.0 146-3 149.6 144.1 149.4 144.0 Mean +146-1' 6 + 146*0' Reducing power as percentage of invert sugar 46.6 46.8 42.4 43-7 48.7 47.6 48.3 46.2 45.5 46.2 46.0 42.8 46.8 42-6 44.8 44.5 44.3 Mean of all samples +143-8' 46.9 been clarified with phosphotungstic acid.Hence it was possible to study the change in con- centration under entirely similar conditions. TWO similarly treated 10 per cent. solutions of commercial glucose solutions were compared for the purpose of observing the deviation over a relatively large angle. The readings are shown in Table VI. From the values obtained for B/A, the correction factor, by which all results for the sugar determinations should be multiplied, can be taken as 0.994. TABLE 171 DERIVATION OF CORRECTION FACTOR Readings in 2-dm tube Factor, B/* Clarif --d, led, A €3 Solution Sucrose, 6.6% .. .. .. . . + 8-'78" + 8-73" 0.9943 Glucose, 10% (commercial) . . . , +22.!38' +22*84O 0.9939 RAPID DETERMINATIQN ON A GLUCOSE-FREE SAMPLE- When it is known that a sample contains no glucose, the following rapid routine analysis will provide results that are within approximate1.y 0.6 per cent. of the sum of the sucrose and invert sugar contents. A 1 or 2 per cent. solution of the sample is prepared by diluting a 10 per cent. solution, and the percentage of invert sugar is determined without preliminary clarification. The percentage of invert sugar after inversion is determined on a 0.8 per cent. solution by inverting 40 ml of the 10 per cent. solution with 6; ml of 6-34 N hydrochloric acid solution, in theJuly, 19521 FOR THE ANALYSIS O F TABLE JELLIES 339 manner described above, and diluting it to 50ml with distilled water.Twenty millilitres of the 8 per cent. solution so prepared are neutralised with sodium hydroxide solution and made up to 200 ml. Results obtained by this method are compared with those for clarified solutions in Table VII, from which it will be seen that the error on reducing sugars before inversion is comparatively small, whereas the effect of non-removal of gelatin from the inverted solution gives results which are 0.6 to 0.7 per cent. in excess of the total sugar present and about 0.5 per cent. in excess of the sucrose content of the one sample in which it was determined. TABLE VII COMPARISON OF THE REDUCING POWER OF CLARIFIED AND UNCLARIFIED SOLUTIONS CONTAINING GELATIN Sample Clarified solution Unclariiied solution P 1 Invert Invert sugar, Sucrose, Sucrose , % % % % Table jelly “A,” before inversion .. . . 40.23 28.80 40.36 29.30 Table jelly “A,” after inversion . . . . 70.55 71.20 Table jelly “B,” before inversion . . . . 32.06 32-22 Calves’ feet jelly “C,” after inversion . . 19.09 19.69 Calves’ feet jelly “D,” after inversion . . 11.97 12.66 My thanks are due to Miss J. G. Holliwell and Miss E. M. Johnson for their assistance in the experimental work and to the Council of the British Food Manufacturing Industries Research Association for permission to publish this communication. REFERENCES 1. 2. 3. Lane, J. H., and Eynon, L., J . Soc. Chem. Ind., 1923,42, 32T.Jackson, R. F , and Gillis, C. L., U.S. Bureau of Standards, Scientific P@er No. 375, 1920. Proceedings of the Tenth Session of the International Commission for Uniform Methods of Sugar Analysis 1949, Subject 6; Int. Sug. J., 1950, 52, 184. THE BRITISH FOOD MANUFACTURING INDUSTRIES RESEARCH ASSOCIATION RANDALLS ROAD LEATHERHEAD, SURREY DISCUSSION THE PRESIDENT said that difficulties were frequently encountered with preparations containing liquid glucose together with other sugars. To the confectionery manufacturer, liquid glucose was a sweetening ingredient that could replace sucrose on an equivalent weight basis as far as soluble solids were concerned; but to the chemist it was a sweetening material of which only about one half was sugar. It was a variable product but, as made to-day, the variations were relatively small.The usual methods of analysis were straightforward, but the interpretation of the results in terms of ingredients was difficult. For certain legal purposes a standard had been laid down based on the sum of the sucrose plus the reducing sugars. This figure bore little relation to the sum of the sweetening ingredients used. The paper was likely to be useful to manufacturers and others who required to find the amount of glucose added to table jellies, and it did not seem to necessitate much more analytical work. MR. T. MCLACHLAN asked if the factor 5.55 used for converting nitrogen content to gelatin protein gave the amount of gelatin itself; if not, would the author give the factor she used to convert nitrogen to gelatin.MISS CHATT said that the ash from a table jelly containing about 10 per cent. of gelatin would not be likely to exceed 0.2 per cent. of the sample. The factor 5-55 was therefore sufficiently close to the true figure for dry gelatin content to apply a correction for gelatin to the observed percentage of soluble solids, as determined by refractometer. MR. K. A. SARGENT asked if there was any reason for choosing the hydrolysis conditions quoted by Jackson and Gillis in preference to others such as hydrolysis with N hydrochloric acid for 5 minutes at 155’ F or with approximately 0-5 N hydrochloric acid for 1 minute at 212’ F. MISS CHATT said that the Jackson and Gillis’ method of inversion was the most reliable one for the polarimetric determination of sucrose.Comparable results for reducing sugars had been obtained by other methods of inversion.340 BAKER : THE DETERMINATION OF pol. 77 MR. D. D. MOIR asked whether the author was fully satisfied that the phosphotungstic clearing agent completely precipitated the gelatin. It was not generally realised that gelatin had a negative specific rotation of more than 300°, and that unless the constituent that gave rise to that rotation could be completely precipitated serious error would result. MISS CHATT pointed out that several trials had been made with various concentrations of salt and phosphotungstic acid in the clarifying reagent, but thlz recommended procedure had been found to give the optimum conditions for completely removing gelatin. MR. J.G. MALTBY said, with regard to methods for the inversion of sucrose, that if strong acid were used, care must be taken to avoid a local excess of alkali when neutralising, in order to prevent loss of fructose before estimating total sugars. He made this point although that particular determination was not made in the proposed scheme. When dealing with old dried jelly, he said, there might be migration of some of the constituents either towards or away from the skin (efflorescence, crystallisation and so on). To get an average, it was necessary to dissolve the whole sample. MISS CHATT said that in order to avoid disturbances due to an excess of alkali, the angular rotation after inversion was determined in acid solution. When a solution was being prepared for the determination of reducing sugars, an aliquot of this solution should lie considerably diluted before neutralisation. Special precautions should of course be taken to ensure an average sample when deterioration had occurred. MR. V. H. PARKS said that the equation for the calculation of sugars quoted by Dr. H. E. Cox on page 32 of his book “The Chemical Analysis of Foods” had been found very reliable. MISS CHATT explained that the equation referred to differed mainly in that the optical rotation was expressed in Ventzke units. It appeared from the text of the book that the formula was used to calculate the percentage of glucose syrup and not glucose solids, in which event the reducing power, as invert sugar, was somewhat above the average for present-day syrups. THE PRESIDENT, in thanking the author for presenting the paper, expressed his satisfaction that she had adopted the Jackson and Gillis conditions for inversion. This method had been very carefully worked out as the one most generally suitable for the inversion of sucrose, and it was the standard procedure adopted by the Condensed Milk Committee of the Society many years ago. All who had had experience of the determination of sucrose recognised the advantage of a single standard method of inversion.

ISSN:0003-2654    

DOI:10.1039/AN9527700335

出版商:RSC    

年代:1952

数据来源: RSC

摘要:

340 BAKER : THE DETERMINATION OF m0i. 77 The Determination of Oxalates in Fresh Plant MateriaI BY C. J. 1,. BAKER (Presented at the meeting of the Society 0% Wednesday, May 7th, 1952) A method is described for determining total oxalates in plants, by extraction with hydrochloric acid, precipitation as calcium oxalate from the deproteinised extract and subsequent titration with potassium perman- ganate. Soluble oxalates are determined in a similar manner, in aqueous extract. The later stages of the determination are carried out on a semi-micro scale, so that a number of determinations can be made simultaneously. The method is designed for fresh green plants only, as oxalate is lost on drying the material. OXALIC acid is fouhd in various quantities throughout the plant world, where it occurs mainly as the calcium or potassium salt of the acid.'The amount present, although generally small, is considerable in some plants, viz., docks, sorrels, spinach, rhubarb, certain fungi and the leaves of beet and related succulents. The jpoisonous properties of the free acid and its soluble salts are well known, and Hickenbotham and Bennett,l working in Australia on the poisoning of sheep by Soursob. ( & a h cernzta), suggest that the calcium salt undergoes decomposition and absorption during the later stages of digestion. Talapatra, Ray and Sen2 have studied calcium metabolism in ruminants fed on fodder rich in oxalate. In this country the feeding of sugar beet tops to farm stock is usually practised with caution, scour and other undesirable effects being attributed to the oxalates present.In view of the significance of the oxalate content of plants normally used as foods, it is surprising that so little detailed information about its determination is to be foundJuly, 19521 OXALATES IN FRESH PLANT MATERIAL 341 in the literature, although Talapatra, Ray, Kehar and Sen3 describe a method for paddy straw and similar materials. The increasing use of fodder beet as feed for cattle and pigs in this country, together with the arrival of many new varieties, gave rise to enquiries about the advisability of using the tops as feed, and for that reason the method described here was devised. EXPERIMENTAL The method of Talapatra et aZ. was primarily designed for the analysis of paddy straw and similar coarse fodders. Total oxalates are extracted with sodium carbonate solution, and in this respect the method resembles the one used in toxicological investigations.The method was tried with several samples of beet leaves, but without success. The alkaline extracts were very dark in colour, acidification produced little improvement and the calcium oxalate precipitates were so heavily contaminated that no determinations could be made. It was found that the oxalates could be extracted from finely divided material of this nature with hot N hydrochloric acid solution and that oxalic acid remains stable under these conditions. This confirms the observation of Hoover and Kar~nairatnam~ that oxalic acid in plant materials remains stable during digestion with hot 1.5 N hydrochloric acid for at least 12 hours.The extracts were light brown, and the oxalate precipitates produced by adding a calcium salt at pH 4.5, although much cleaner than those formed in Talapatra's method, were somewhat contaminated by protein material. Contamination was almost completely eliminated by adding a suitable deproteinising agent before pre- cipitating the oxalate. The deproteinising agent must be effective at a pH low enough to retain calcium oxalate in solution and must not form a calcium salt that subsequently co-precipitates with oxalate. A modification of the phosphoric - tungstate reagent of Sendroy and Van Slyke, mentioned by Hawk, Oser and S~mmerson,~ was found to be ideal for the purpose. Oxalate was precipitated from the deproteinised extract with calcium chloride in acetic acid buffer solution of pH 4.5.The precipitate was washed with 5 per cent. acetic acid saturated with calcium oxalate to prevent loss of the salt by solution in the acetic acid. Precipitates were clean and dissolved readily in dilute sulphuric acid, leaving no residue; titration with 0.02 N potassium permanganate gave a sharp and persistent end-point to within one drop. A similar technique was applied to aqueous extracts of the leaves in the determination of water-soluble oxalates; again it was necessary to remove the protein although this was present in comparatively small amounts. By comparing the amount of oxalate found in samples of fresh leaves with that in leaves dried in an oven at 100" C for 24 hours, it was found that drying caused appreciable loss of oxalate, so fresh material was used for routine analysis.Qualitatively, the method is as satisfactory for dry material as for fresh and could probably be adapted to suit the require- ments of dry substances. METHOD APPARATUS- tubes of resistance glass. REAGENTS- A disintegrating machine-a Waring blendor is suitable-and 50-ml tapered centrifuge Diluted hydrochloric acid (1 + 1). Ammonium hydroxide solution-Spgr., 0.880. Phosphoric - tungstate reagent-Dissolve 24 g of sodium tungstate in water, add 40 ml of syrupy phosphoric acid, spgr. 1.75, and dilute to 1 litre. Calcium chZoride bufler-Dissolve 25 g of anhydrous calcium chloride in 500 ml of 50 per cent. v/v glacial acetic acid and add this solution to a solution of 330g of sodium acetate in water, diluted to 500ml.Wash solution-A 5 per cent. v/v solution of acetic acid kept over calcium oxalate at room temperature. Sul#uric acid-A 10 per cent. v/v solution. Potassium permanganate-A 0.02 N solution, prepared as required by diluting a 0.1 N solution. Shake the solution periodically and filter before use.342 BAKER: THE DETERMINATION OF [Vol. 77 PROCEDURE- For total oxalate-Homogenise 60g of chopped green material with about 100ml of water in the blendor and transfer the mixture to a 600-ml beaker with the minimum number of washings. Add 2 volumes of &luted hydrochloric acid (1 + 1) to each 10 volumes of liquid (to give an approximately normal concentratiori) and one or two drops of capryl alcohol and boil for 15 minutes. Allow to cool, transfer to a 500-ml volumetric flash, dilute to the mark and after occasional shaking set it aside overnight.Mix and filter through a dry paper, Transfer, by means of a pipette, 25 ml of filtrate into a tube fitted with a stopper, add 5 ml of phosphoric - tungstate reagent, mix by inverting once or twice and set the mixture aside for 5 hours. Centrifuge for 10 minutes at 3000 r.p.m. and radius 6 inches, transfer exactly 20 ml of the clear solution to a 50-ml centrifuge tube and add ammonium hydroxide dropwise from a burette until the solution is alkaline, as indicated by the formation of a slight precipitate of phosphotungstate. Add 5 ml of the calcium chloride reagent, stir with a fine glass rod and leave the tube overnight in a refrigerator at 5" to 7" C.Centrifuge for 10 minutes, carefully remove the supernatant liquid and wash the precipitate with 20 ml of wash solution, stirring vigorously with a fine rod until the precipitate is broken up and the impurities dissolve. Centrifuge for 10 minutes, carefully remove the washings, dissolve the precipitate in 5 ml of 10 per cent. sulphuric acid, place the tube in a water-bath at 100" C for 2 minutes and titrate the oxalic acid with 0.02 N potassium permanganate. 1 ml of 0.02 N KMnO, = 0.00090 g of (COOH),. Twenty millilitres of deproteinised extract are equivalent to 2.0 g of sample. For water-soluble oxalate-Homogenise another 60 g of sample and transfer it to a beaker with water, as before. Boil for 15 minutes, cool, dilute to 500 ml, mix, set it aside overnight and filter it through a paper capable of retaining calcium oxalate.Transfer 25 ml of the filtered extract to a stoppered tube by means of a pipette, add 2-5 ml of diluted hydrochloric acid (1 + 1) and then 2.5 ml of phosphoric - tungstate reagent ; complete the determination by the procedure described for total oxalate. RESULTS Blank determinations on oxalate-free material such as cabbage and kale indicated the specificity of the method. Quantities of pure sodium oxalate ranging from 0.25 to 1.00 per cent. were added to freshly chopped cabbage, to give concentrations like those found in beet leaves. Recoveries with the acid extractio:n technique are shown in Table I. TABLE I RECOVERIES OF OXALATE ADDED TO FRESHLY CHOPPED CABBAGE Oxalic acid equivalent of sodium oxalate Oxalic acid added, theoretical, found., Recovery, % % % nil nil - 0.168 0.166 98.8 0.336 0.334 99.4 0.504 0.504 100.0 0.672 0.670 99.7 Mean recovery = 99.5 per cent.Recovery tests were also carried out with similar quantities of oxalate added to chopped fodder beet leaves with acid and water extraction techniques, and gave the results shown in Table 11. Care was taken to ensure a reasonably homogeneous sample of chopped leaves, because all errors from sample variation would ;be chargeable to the added oxalate. These results show recoveries to be excellent under the prescribed conditions, amounts of oxalic acid as low as 3mg being recovered in some experiments. Further tests showed rates of precipitation to be related to concentration of oxalate; 10 mg were almost completely recovered after 3 hours, but smaller amounts required more time.Precipitation conditions were finally standardised, a period of 1.6 hours or more in a refrigerator at 5" to 7" C being normally used.July, 19521 OXALATES IN FRESH PLANT MATERIAL TABLE I1 343 RECOVERIES OF OXALATE ADDED TO FRESHLY CHOPPED FODDER BEET LEAVES Oxalic acid equivalent of sodium oxalate added, theoretical, found, added oxalic acid, added oxalic acid, Oxalic acid Recovery of Recovery of % % % % Total oxalate method- - - nil 0.380 0.168 0.549 0.169 100.1 0.336 0.720 0.340 101.0 0.504 0.877 0-497 98-5 0-672 1-044 0.664 99.0 Mean = 99.6 Soluble oxalate method- nil 0.092 0.168 0.263 0.336 0.430 0.504 0.589 0.672 0.765 - 0.171 0.338 0.497 0.673 - 102.0 100-6 98.6 100.2 Mean = 100.4 The efficiency of the extraction method was examined by studying the variation in oxalate content of the extract with length of time in contact with the solid material.The results of eight tests are shown in Table 111. TABLE I11 EQUILIBRIUM OF OXALATE CONTENT OF EXTRACTS REMAINING IN CONTACT WITH INSOLUBLE MATERIAL (BEET LEAVES) Oxalic acid found in the material A r 1 Sample No. After 16 hours, After 48 hours, After 96 hours, % % % 0-437 0.643 0.680 0.527 0.581 0.441 0.549 0.583 0.435 0.661 0.682 0.540 0.580 0.441 0.554 0.589 0.436 0.66 1 0.684 0.533 0.583 0.441 0.554 0.585 It is evident from these results that the oxalate present in the sample soon becomes uniformly distributed through the liquid phase and that determination need not be delayed.The filtered extracts from acid and water treatments were retained and examined at intervals up to 14 days after their preparation. The acid extracts remained clear, but TABLE IV QUADRUPLICATE DETERMINATIONS OF TOTAL OXALIC ACID IN TWO SAMPLES OF BEET LEAVES Mean and its standard Oxalic acid, yo error, yo Variety A . . .. .. 0.624, 0.620, 0-626, 0.632 0.626 f 0.005 Variety B . . . . .. 0-472, 0.464, 0.464, 0.469 0:467 f 0-004 moulds quickly developed in the aqueous solutions. However, the oxalate content remained unchanged throughout, indicating that extracts can be safely kept for several days before canying out determinations. Finally, the reproducibility of the method was tried with two samples of beet leaves, each sample being analysed in quadruplicate. Results are shown in Table IV.344 BAKEF: [Vol.77 As an over-all test of the reproducibility of the method, these results show remarkably close agreement, particularly with material in which sampling errors might well be expected. The author is indebted to Dr. A. Eden for his interest and encouragement during this investigation. REFERENCES 1. 2. 3. 4. 5. Hickenbotham, A. R., and Bennett, W. G., J . Agric. S. Australia, 1931, 34, 1225. Talapatra, S. K., Ray, S. C., and Sen, K. C., J . Agric. Sci., 1948, 38, 163. Talapatra, S. K., Ray, S. C., Kehar, N. D., artd Sen,, K. C., Science and Culture, 1942, 8, 209. Hoover, A. A., and Karunairatnam, M. C., Bioclzem. J . , 1945, 39, 237. Hawk, P. B., Oser, B. L., and Summerson, W. H., “Practical Physiological Chemistry,’’ Twelfth Edition, J.& A. Churchill Ltd., London, 1947, p. 576. MINISTRY OF AGRICULTURE AND FISHERIES ANSTEY HALL, TRUMPINGTON, CAMBRIDGE January, 1952 DISCUSSION THE PRESIDENT thanked the author for presenting a paper that had given evidence of the excellent manipulative technique of agriculturalists working with great accuracy on small amounts of material. He asked if it could be proved that oxalate only was precipitated and not the salt of any other acid. MR. BAKER said that proof of specificity had been mainly by inference, based upon the negative results obtained when the method was applied to oxalate-free plants, but he mentioned that oxalic acid was unique in producing a calcium salt that was insoluble a t pH 4.5. DR. J. HASLAM enquired if it was possible to prove that the calcium salts of other organic acids were not present, by weighing the precipitate as CaC,O,.H,O before dissolving it in acid.MR. BAKER said the oxalate precipitates were too small to permit critical gravimetric checks with the facilities at his disposal. DR. H. AMPHLETT-WILLIAMS said that he had been rather concerned for some years past about the toxicological aspect of the high proportions of oxalic acid that occured in certain vegetable purhes and canned baby foods. He had found 0.24 per cent. of water-soluble oxalate in strained spinach pur6e; this was equivalent to 0.3 g of oxalic acid in a 43-ounce tin, the whole of which was recommended on the label for daily consumption by a baby of 7 months. The minimum fatal adult dose of oxalic acid was generally given as 4 g and, by Gaubins’ 01- Young’s method of calculation, the corresponding dose for a child under 1 year would be one twelfth of this, viz., 0.3 g ; this quantity would be capable of combining with the whole daily requirement of calcium. DR. J. H. HAMENCE said that he was surprised to hear the author placed his solutions in the refrigerator in order to precipitate the oxalate, as in his experience, calcium oxalate came down far more readily if the solution was heated in a water-bath. He asked Mr. Baker if he had tried using acetone to assist precipita- tion. This method had recently been advocated for the determination of calcium in soils and had been used with promising results. It was also used in the wash liquors, particularly when the calcium oxalate was removed by centrifuging. MR. BAKER agreed that the water-bath technique had advantages when precipitating calcium in macro quantities with oxalate, the main effect being to encouirage “granulation.” In his experience, however, small amounts of oxalic acid were not readily precipitated from solutions at high temperatures, and the best results were obtained under the cool conditions indicated in the paper. He thanked Dr. Hamence for drawing his attention to the use of acetone.

ISSN:0003-2654    

DOI:10.1039/AN9527700340

出版商:RSC    

年代:1952

数据来源: RSC

摘要:

July, 19521 FURNESS 345 Changes in Potential of the Dropping-Mercury Electrode during Drop-Formation, and Measurement of Potential in Polarographic Analysis BY W. FURNESS The electromotive force applied across the terminals of a polarographic cell is influenced slightly by the magnitude and direction of the polarographic current. The change in diffusion current due to the growth and fall of mercury drops is, likewise, accompanied by a pulsation in the applied electromotive forcei The amplitude of this pulsation varies partly as the value of the diffusion current at maximum drop size and partly as a function of the position occupied by the sliding contact on the polarograph potentiometer. With a Tinsley polarograph reasonable agreement has been found between calculated and observed effects.Because of the pulsation it is preferable, when plotting polarograms manually, to determine the diffusion current and the potential of the dropping- . mercury electrode when the rate of change of applied electromotive force is least, that is, at the instant of maximum drop size. This potential, with respect to that of a reference electrode, can be measured to the nearest tenth of a millivolt by the Poggendorff compensation method. Polarographic cells incorporating a quiet mercury-pool to conduct the diffusion current are not always satisfactory; in contact with certain oxy-acids of sulphur, for example, such electrodes may acquire insoluble films in which event their potential and electrical resistance become erratic. These com- plications are avoidable by means of cells having a pair of permanently attached calomel electrodes. The use of such cells in conjunction with the Tinsley polarograph has been tested by reference to the well known polarograms of cadmium and thallium.WHILST accuracy in the measurement of diffusion current is the foremost requirement in quantitative polarographic analysis, references to the potential of a polarised electrode are fundamental to the use of polarographic data in qualitative analysis. For many practical purposes it often suffices to record polarograms automatically, in which event the electro- motive force applied to the cell circuit can be found approximately from the scale of the abscissa. For other purposes, which justify manual plotting of a polarogram as described by Miiller,l readings of electromotive force applied by the polarograph potentiometer must be corrected for the iR drop or iR gain throughout the cell circuit in order to find the potential of the dropping-mercury electrode with reference to that of the quiet electrode of the polaro- graphic cell.In the most accurate work, however, the potential of the dropping-mercury electrode at every point on the manual polarogram should be referred to the potential of a standard half-cell that does not conduct the polarographic current , the potential difference being measured potentiometrically. A circuit for this purpose has been described by Lingane and Kolthoff .2 First, it may be noticed that the potential of the dropping-mercury electrode is unsteady at every point upon a polarographic wave except where the diffusion current is very small.Secondly, in investigations of the polarographic behaviour of certain sulphur-containing compounds, the surface of a quiet, internal mercury-pool electrode may become covered by a film of sulphide whose electrical resistance is high enough to prevent the dropping-mercury electrode being maintained, even approximately, at a definite potential. In the present work, therefore, the factors that influence the value of electromotive force applied to the cell circuit have been examined and a technique for measuring the potential of the dropping-mercury electrode has been developed. Throughout this investigation a Tinsley polarograph, model V3211 , was used, but the conclusions will be applicable, although perhaps to a varying degree, t o work with other pen-recording instruments.In the application of this last-mentioned procedure two problems may arise.346 FURNESS: THE MEASUREMENT OF [Vol. 77 INFLUENCE OF THE DIFFUSION CURRENT ON THE ELECTROMOTIVE FORCE APPLIED TO THE CELL CIRCUIT In the basic circuit illustrated in Fig. 1, DB is the uniform potentiometer wire across which the battery maintains a steady current when the switch, S, is closed. The fall in potential per unit length of the potentiometer wire is adjusted to some convenient value by means of the resistance, R,, in series with the battery. The quiet electrode of the polarograph cell is connected through a galvanometer and its shunts, G, to the fixed point, A. The dropping- mercury electrode, whose capillary is immersed in a supporting electrolyte containing, for instance, a reducible ion, is connected to the sliding contact, C, which can occupy any position along DB.Let C move to a point between A and B such that the process of reduction occurs at the dropping-mercury electrode. A current equal to the sum of the diffusion and residual currents then flows from A through G to the quiet electrode, thence through the cell and dropping-mercury electrode back to the potentiometer wire at C. Kolthoff and Lingane3 treated this network as a parallel circuit of ohmic resistances and so arrived a t an expression for the electromotive force, E,, applied to the cell circuit, which can be written as- where EB,, denotes the e.m.f. between A and B, denotes the effective resistance between A and B, denotes the resistance of the potentiometer wire between A and C RbB R, and R,, denotes the resistance of the polarographic cell circuit.But, as the dropping-mercury electrode is subject to concentration polarisation, the current flowing through the cell cannot be proportional to the electromotive force applied externally across its poles, and for this reason the above expression does not at all times strictly define the value of the electromotive force applied across the polarographic cell circuit. The difference in potential between the points A and C is given strictly in accordance with Ohm's Fig. 1. Basic circuit of the polarograph, with polarographic cell having one external half-cell such as the saturated calomel electrode law by the product of the resistance of that portion of the potentiometer wire between A and C a d the current flowing at any instant dong A(:.However, the current flowing from A to C along the potentiometer wire cannot remain constant whilst the current in the polar+ graphic cell circuit is continually changingJhroughout the life of each mercury drop at the capillary, and so it follows that the potential between A and C must also vary throughout the life of each drop. THEORETICAL- Case I-Considering the reduction of a metal ion at the dropping-mercury electrode, let US specify that reduction occurs only when the potential of the dropping-mercury electrode is made negative with respect to that of the quiet electrode of the reference half-cell. As the sliding contact C moves from A in the direction of B no appreciable current is taken from the potentiometer wire until the potential of C with respect to A approaches the polaro- graphic reduction potential of the metal ion.Afterwards, the current taken from theJuly, 19521 POTENTIAL OF THE DROPPING-MERCURY ELECTRODE 347 potentiometer wire follows the current - voltage curve, which is the polarogram of the metal ion. The current flowing in the potentiometer wire from A to C is less than that flowing from D to A or from C to B by a quantity equal, at any given instant, to the current flowing in the polarographic cell circuit. If contact between the dropping-mercury electrode and the point, C, is temporarily broken the current throughout the whole length of the potentiometer wire becomes steady and uniform at all points. Let its value be denoted by I.If the potential difference across the terminals of the battery is V , the resistance per unit length of the potentiometer wire is r and its length I , then- Let contact again be made between the dropping-mercury electrode and the point C, and let the current flowing from A to C in the polarographic cell circuit at any given instant be denoted by il. If the current flowing at the same instant through the battery is denoted by I,, and the distance between A and C is d , then- Hence, from which Let the potential of C on the potentiometer wire against the potential of A be denoted by E whilst the dropping-mercury electrode is temporarily disconnected, and by El when the instantaneous value of the diffusion current is i,, after the connection at C has been re-made.Then, and If it is assumed that the diffusion current is negligibly small at the instant following detach- ment of the previous drop, the change in potential of C with respect to A up to any instant during the life of a single drop is given by- V = I(d + Rl). V = Ilr(Z - d ) + (I1 - i J r d + IlR1. I(YZ + R1) = I1rZ - i1yd + IlR1, I = I , - ilYd/(yZ + R1). E = -Iyd El = -(I1 - i,)Yd. AE = El - E = [I - 11 + il]rd = [il - ilyd/(yZ + Rl)]rd = i l r d [ l - yd/(rZ + Rl)] . . .. .. (1) Therefore, whilst the diffusion current is increasing on account of the growth of the drop, the potential of the dropping-mercury electrode must become more positive with respect to A, and, at any given instant, the extent of the change must be proportional to the instantaneous value of the diffusion current.The extent of the change is dependent also upon the position of the sliding contact C, but is independent of the resistance of the polaro- graphic cell circuit between A and C. Case 11-If polarographic reduction occurs whilst the potential of the dropping-mercury electrode is positive with respect to that of the quiet electrode, that is, whilst contact C lies between D and A, no current is taken from the potentiometer wire. Instead, the current I , flowing through the battery is supplemented between C and A by the diffusion current i,. The potential of C, therefore, which is already more positive than that of A at the beginning of the formation of each drop, becomes still more positive as the diffusion current increases with increasing drop size.By use of the previous system of notation and method of derivation, from which I(Yl + Rl) = I , q - d ) + (1, + i2)yd + IzR,, I = I , + i , ~ d / ( d + Rl). E = Ird Now and therefore, E , = ( I , + QYd, AE = E , - E = [ I ~ - I + ;,pa = [i2 - i,~d/(yZ + R l ) ] ~ d = i , ~ d [ l - rd/(rZ + R J ] .. (2)348 FURNESS: THE MEASUREMENT OF [Vol. 77 Cases 111 and IV-There are two other cases to be considered in which the diffusion current is anodic. If C lies between A and B, the anodic diffusion current supplements the current from the battery in that part of the potentiometer wire between A and C. Finally, when C lies between D and A, current is taken from the potentiometer wire if an anodic current flows in the cell circuit.If the anodic currents flowing from the dropping-mercury electrode towards the quiet electrode within the polarographic cell are given the values i, and i, the changes in the potential of the point C with respect to A during the life of each drop are calculated to be- when C is a point between A and B, and when C is a point between D and A. The general case-If the convention be now adopted that a reduction of the substance responsible for concentration polarisation at the dropping-mercury electrode gives rise to a positive diffusion current, whilst oxidation is accompanied by a negative diffusion current, the symbol i can be substituted in the above equations for i,, i,, 4, and 4,. Further, if the resistance of that part of the potentiometer wire between A and C be denoted by RAc and that of the whole potentiometer wire by RDB, the general expression for AE becomes- .. (3) .. (4) AE = E3 - E = -&7d[1 -- ~ d / ( r Z + R,)] AE = E, - E = -i,rd[l - rd/(rZ + R,)] . . .. . . .. .. .. . . If AE be taken to denote the change in potential of C with respect to that at A during the complete life-period of a single drop, the corresponding value of i to be substituted in equation (5) is the instantaneous value of the diffusion current at maximum drop size. The inferences to be drawn from equation (5) can be stated as follows. (a) When the position of contact C coincides with the point A, AE will be zero irrespective of the value of the diffusion current. (b) For any other given position of C, AE should be directly proportional to i as long as the.resistances RD, and ’R, remain constant. (c) As the position of contact C changes along the potentiometer wire, the ratio AE/i should attain a maximum value equal to a(RDB + R,) when RA, = Q(RDB + R,). (d) The value of the ratio AE/i should be independent of the resistance of the polaro- graphic cell circuit and also independent of the potential gradient along the potentio- meter wire. (e) Since, in the practice of polarography, it is desirable that AE/i should be as s m d as possible, polarographs should be designed so that the resistance (RDB + R,) is as small as is compatible with the maintenance of a very steady potential gradient along the potentiometer wire. \ EXPERIMENTAL- Equation (5) and the inferences (a),< (b), (c) and (d) have been tested by experiment.The potentiometer wire of the Tinsley pola.rograph, type S10600, model V3211, has the high resistance of 118.5 ohms per volt length and this instrument was used in conjunction with a polarographic cell having an external saturated calomel electrode of the type previously d e ~ c r i b e d . ~ ~ ~ Solutions containing Pb”, Fe,”’ S,O,” or S,03” in appropriate supporting electrolytes were selected as depolarisers to provide data appropriate to Cases I , 11, I11 and IV above. Throughout the tests the resistances of the potentiometer wire RDB and of R, remained constant. A Tinsley-vernier potentiometer, type 3126B, standardised frequently against a Weston cell, was connected across the points of the Tinsley polarograph, viz., terminal “E” and terminal “electrode -,” which correspond to points A and C, respectively, of Fig.1. Contact C was adjusted to the desired position on the polarograph potentiometer wire, without having the dropping-mercury electrode connected to it, and the steady potential difference between C and A was measured on the vernier potentiometer. Next, the dropping-mercury electrode was connected to C. The oscillations of the galvanometer in the Poggendorff compensation circuit then coincided with the growth and fall of each drop at the capillary, and the vernier potentiometer was so adjusted that these oscillations showed a balance ofJuly, 19521 POTENTIAL OF THE DROPPING-MERCURY ELECTRODE 349 electromotive force at the instant of maximum drop size.The sum of the diffusion and residual currents recorded at the same instant on the Tinsley pen-recorder was also noted. The two readings on the vernier potentiometer gave, respectively, the values of E and El, 2, 8, or 4, and hence the algebraic difference AE. RESULTS- The experimental data are presented in Tables I and I1 and also in Fig. 2, where com- parisons are made between the observed values of AE/i and values calculated from equation (5). TABLE I DETERMINATION OF THE RATIO AE/i FOR VARIOUS POSITIVE VALUES OF i, WHILST RAc HAS SEVERAL DISCRETE VALUES PolarograDhic reduction of ferric ion in 0.2 N potassium nitrate, with 0.01 per cent. of gelatin Potential of C with respect to A Polarograph potentio- meter, nominal setting, volts SO.1 0 - 0.5 - 1.0 RAC, ohms 11.85 0 59.25 118.5 Normality of ferric ion, x 103 2 4 10 20 2 4 10 20 2 4 10 20 2 4 10 20 , Dropping- mercury electrode connected i.to c, P A mV 6.4 85.76 12.8 85.12 32.3 85.9 64.5 85.3 6.4 - 12.30 12.9 - 11.80 32.6 - 12.43 65.0 - 12.58 6.6 - 520.47 13.0 - 520.93 32.7 -518.5 65.5 -516.1 6.4 -1012*80 12.9 - 101 1.46 31.9 -1010*4 63.5 -1005.8 I Dropping- mercury electrode disconnected from C, mV 85.69 85.98 85.54 84.50 - 12.30 - 11.81 - 12.47 - 12.65 - 520.82 - 521.65 - 520.21 -519.63 - 1013.40 - 1012.65 - 1013.28 - 1011.67 AE, mV 0-07 0.14 0.4 0.8 0.00 0.01 0.04 0.07 0.35 0.72 1-7 3-5 0.60 1-19 2-9 5.9 AE/i - Observed, Cal- mV per culated, 0.011 0.012 0.01 1 0.012 0.012 0~000 0.000 0.001 0*001 0.001 0.054 0.062 0.055 0.052 0.053 0.094 0.089 0-092 0.091 0.093 pA mVperpA When R,, is zero A and C are coincident so that AE should always be zero irrespective of the value of i.On the Tinsley polarograph the precision with which C can be set to coincide with A is not high, nevertheless, Table I shows the values of AE/i to be negligibly small. For any other point on the cathodic wave of ferric ion, where according to convention the diffusion current is positive, the potential of the dropping-mercury electrode always becomes more positive with respect to that of the calomel electrode whenever the diffusion current is increasing, For values of i within the range 6 to 65 pA the ratio AE/i was found to remain constant at each of the specified values of R,. On the other hand, when the diffusion current is negative, as illustrated by the examples of the dithionite (S2O4") and thiosulphate waves reported in Table 11, the change in potential of C with respect to A is in a negative direction as each drop grows at the dropping-mercury electrode.'Here again, the observed values of AE were found to be directly proportional to the corresponding values of i, the ratio varying only according to the chosen value of R,. The values given in Tables I and I1 for i are those recorded directly by the pen-recorder at the instant of maximum drop size. These are known to be lower than the real instantaneous values at maximum drop size by approximately 3 per cent.g For this reason the data presented as observed values of AE/i are subject to correction and, with this allowance, the observed values are in satisfactory agreement with those calculated from equation (6).By the use of one or two accumulators to provide the potential gradient along the potentiometer wire it was possible, with a single solution of lead in a potassium nitrate supporting electrolyte, to observe the variation of the ratio AE/i over a range of values of350 FURNESS: THE MEASUREMENT OF [Vol. 77 RAC from 59.3 to 308.1 ohms. The observed and calculated values of AE/i are not tabulated here but have been plotted, along with other data from Tables I and 11, in Fig. 2. Subject to the previously mentioned correction of 3 per cent., the agreement is regarded as satisfactory. The observed maximum of 0.123 mV per p A (0.119 mV per pA after correction) compares with the calculated value of t(RDB + R,) which is equal to 0-117 mV per PA, and this observed maximum occurs at a point whose abscissa is 235 ohms, a value which is coincident with the calculated value of Q(RDB + R,).In the same experiment, for values of RAc from 106.7 to 154.1 ohms, values of AE/i were obtained both with one and with two accumulators TABLE 11: DETERMINATION OF THE RATIO AE/i FOR VARIOUS NEGATIVE VALUES OF i, WHILST RAc HAS TWO DISCRETE VALUES Polarographic oxidation of dithionite ion (S204”) in 0-5 M di-ammonium hydrogen phosphate, M ammonium hydroxide, with 0.01 per cent. of gelatin Polarographic oxidation of mercury on the wave due to thiosulphate ion in 0.2 N potassium nitrate Potential of C with respect to A r 1 Dropping- Dropping- Polarograph mercury mercury potentiometer, electrode electrode AEli, nominal connected disco:nnected P setting, RAC, i, to c, from C, AE, Observed, Calculated, volts ohms pA mV mV mV mVper pA mVper pA - 0.2 23.70 -88.0 -204.7 - 202.60 -2.1 0.024 0.023 (a) For various concentrations of the dithionite ion- -62.0 -203.9 - 202.48 - 1.4 0.023 - 39.5 - 203.38 - 202.46 - 0.92 0.023 -21.2 -202.94 - 202.46 - 0.48 0.023 ( b ) For various concentrations of the thiosulphate ion- +o-1 11.85 -85.5 85.37 86.37 - 1.00 0.012 - 64.0 85.58 86-30 - 0.72 0.01 1 - 43.0 85.83 86.30 - 0.47 0.01 1 - 22-0 86.05 86.29 - 0.24 0.01 1 0.012 across the polarograph potentiometer, Such a two-fold change in the potential gradient along the potentiometer wire had no effect on the values found for AE/i.Nor did the internal resistance of the polarographic cell have any influence, for similar results were obtained with several types of cells whose internal resistances varied from a few hundred to several thousand ohms.INFLUENCE OF THE DIFFUSION CURRENT ON THE POTENTIAL OF THE DROPPING-MERCURY ELECTRODE Let us suppose a polarographic cell to be constructed so that two identical half-cells, X and Y, are connected through liquid junctions with the supporting electrolyte in which the tip of the dropping-mercury electrode, 2, is immersed (Fig. 3). Also, let the quiet electrode of the half-cell, X, be connected through a galvanometer G to point A, and let the dropping-mercury electrode, 2, be connected to the sliding contact, C. As long as no current flows through the polarographic cell the potentials of the electrodes of the two half-cells, X and Y, must be equal to one another and to that of point A, whilst the potential of the dropping-mercury electrode is equal to that of point C.In this instance, the potential of the dropping-mercury electrode against that of either of the half-cells is given strictly by the position of contact C and is equal to -Ird. This value can be measured with great accuracy by the Poggendorff compensation method, a potentiometer being connected across C and A, or across C and the half-cell, X, or across C and the half-cell, Y. Now let a reducible ion be introduced into the supporting electrolyte so that a diffusion current i, flows from A to C via G, X and 2. The potential difference between C and A now changes to the value -(II - Q r d , with the previous notation, but this is numerically larger than the potential difference between C and the half-cell, X, by the quantity ilRn, whereJuly, 19521 POTENTIAL OF THE DROPPING-MERCURY ELECTRODE 351 Rs denotes the resistance of the direct path taken by the diffusion current between A and X.The potential of C with respect to X is, therefore, given by- As the half-cell, Y, does not conduct any portion of the polarographic current its electrode must be at a more negative potential than that of the half-cell, X. Provided that its liquid junction with the supporting electrolyte does not lie along the path followed by the polaroa graphic current between half-cell X and the tip of the dropping-mercury electrode, 2, half- cell Y can be used as a reference electrode against which the potential of the supporting -(I1 - i 1 ) d + &R=.3 RAG ohms Fig. 2. Relationship between AE/Z and the position of the contact along the potentiometer wire. The plot of AE/i = RAc[l -RAC/(RDB+R~)] is shown by the broken line. Observed values of AE/z are plotted against selected values of RAC, thus -o-o-o-o- electrolyte in immediate contact with the dropping-mercury has a standard value. In these circumstances, the potential of C with reference to the electrode of half-cell Y is given by- where RceU is ‘the resistance of the path within the polarographic cell between the electrode of half-cell X and the supporting electrolyte in immediate contact with the dropping-mercury electrode. Because of the resistance of the thread of mercury, which will be denoted by Rzc, the potential d?.?d.m.e.of the mercury drop with reference to the electrode of half-cell Y is given strictly by the equation- a . .. (6) -(I1 - i,pa + il(RAx + Rcell) * * .. Ed.m.e. = -(I1 - i1)rd + i l ( R n + R e 1 1 + Rzc) .. t 7) If the total change in potential of the dropping-mercury electrode, against the reference electrode, Y, throughout the life of the drop is denoted by A E d . m . e . , and if i, represents the diffusion current at maximum drop size, we can write- AEd.m.e. = -(I1 - i,)rd + il(RAX + Rcell + RZC) - (-Ird) = i,rd[1 - rd/(rJ + R,)I + iI(RAx + R c e l l + Rzc). A little further consideration will show that, if the instantaneous value of a cathodic or anodic diffusion current at maximum drop size be denoted by i, having regard to the convention relating to its sign, the general expression for A E d .m . e . can be written as- AEd.m.e. = iRAC (I - .. (8) MEASUREMENT OF THE POTENTIAL OF THE DROPPING-MERCURY ELECTRODE CHOICE OF METHOD- In all polarographs incorporating the circuit shown in Fig. 3 the potential of the dropping- mercury electrode against a reference half-cell can only be strictly proportional to the distance352 FURNESS : THE MEASUREMENT OF [Vol. 77 AC if the polarographic current is zero. Othenvisle, for any given position of C, the potential of the dropping-mercury electrode pulsates with a frequency equal to the drop rate and with an amplitude proportional to the instantaneous value of the diffusion current at maximum drop size.The maximum value of the first te:rm on the right-hand side of equation (8) is a i ( R D B + R,) but, since the resistance of the potentiometer wire in many polarographs is low, this term can be negligible in comparison with the second. In the Tinsley polarograph, however, the value of a ( R D B + R,) is approximately 117 ohms and, since R,, is very small the value of (R, + Rceu + R,) is chiefly dependlent on the resistance of the polarographic I A vernier I Fig. 3. Basic circuit of the polarograph, but with polaro- graphic cell having two external half-cells and connections to vernier potentiometer. The half-celI on the right is the reference electrode cell, which may be only a few hundred ohms. For this reason, if accuracy in potential measurements to the nearest millivolt or better is desired when plotting manual polarograms with the Tinsley polarograph, the effect of both terms on the right-hand side of equation (8) must be taken into account, especially if the diffusion current approaches or exceeds 10 PA.Some investigators, following Lingane,’ have preferred to work with very dilute solutions of the depolariser in order that the iR correction should be negligibly small, but when half- wave potentials are to be measured this procedure suffers from the disadvantage that the residual current (which is not always accurately known) forms an appreciable part of the total current a t the dropping-mercury electrode. In order to find accurately the potential of the dropping-mercury electrode, Ed.m.e., when the diffusion current is large, it would be possible first to measure the steady potential difference between C and A (Fig. 1) whilst the dropping-mercury electrode is temporarily disconnected from C, secondly to measure the dil’fusion current i when connection at C is re-made, thirdly to calculate A’Ed.m.a from equation (8) and consequently to find arith- metically the value of Ed.m,e.. Such a procedure would be tedious and would involve a knowledge of the internal resistance of the polarographic It is in every respect preferable to measure the potential of point C (Fig. 3) against a reference half-cell such as Y, which does not conduct the polarographic current. The instantaneous value of this potential difference has been defined above, equation (6), and its value should be measured at the instant when its rate of change is least, that is, at the instant of maximum drop size when the rate of change of diffusion current is least.The only correction then needed is for the potential difference iR,, (with due regard to the sign of i) across the capillary of the dropping-mercury electrode and also across any tungsten contacts that may have been incorporated in the dropping-mercury electrode assembly. 9v10 DESIGN OF POLAROGRAPHIC CELL- Polarographic cells with a dropping-mercury edectrode, a quiet “working” mercury-pool electrode and a reference half-cell were used by Lingane and Kolthoff2 and by later workers. Occasionally, the quiet working mercury electrode acquires a surface film whose electrical resistance is so high that the potential of the dropping-mercury electrode becomes erratic.This happens, for example, in the polarographic examination of certain sulphur compounds. The difficulty can be avoided by the use of an external half-cell as the quiet working electrodeJuly, 19521 POTENTIAL OF THE DROPPING-MERCURY ELECTRODE 353 of the polarographic cell, and if a second half-cell is incorporated this can serve as the reference electrode against which the potential of the dropping-mercury electrode may be measured. Thus, the two half-cells would function as X and Y , respectively, of Fig. 3. Such a polaro- graphic cell has been constructed, having two saturated calomel half-cells disposed symmetrically about the central dropping-mercury electrode compartment. In other respects its features and principal dimensions are similar to those of the H-type cell previously de~cribed.~ This cell is intended for use only in the circuit shown diagrammatically in Fig.3 when polarograms are plotted manually; it does not supplant the simpler H-type of polaro- graphic cell when polarograms are being automatically recorded. PROCEDURE FOR MEASUREMENT OF DROPPING-MERCURY ELECTRODE POTENTIAL- During the plotting of a polarogram, the cell is partially immersed in a thermostat a t 25.0" C. The dropping-mercury electrode and one of the saturated calomel electrodes (hereinafter called the working calomel electrode) are connected to the polarising unit of the polarograph, whilst the dropping-mercury electrode and the other saturated calomel electrode (the reference electrode) are connected directly to the terminals of a Tinsley vernier potentio- meter, type 3126B, as indicated in Fig.3. These are the only connections to the polarographic TABLE I11 DATA FOR MANUAL POLAROGRAM OF 1.82 MILLIMOLAR CADMIUM SULPHATE IN 0.1 N POTASSIUM CHLORIDE, WITH 0.01 PER CENT. OF GELATIN Temperature, 25.0" C mW = 1.90 mg% set.-* Values of i recorded at instant of maximum drop size and corrected for residual current EI EI + iRzc EII VS. S.C.E., VS. S.C.E., VS. S.C.E., EI + iRzc - EII, i, (id - i), log,, i / ( i d - i ) volts volts volts volts P A PA -0.41454 - 0.52337 -0.5721 -0.5810 - 0.5894 - 0.5967 - 0.6035 - 0.6094 - 0.6200 -0.6310 - 0.7942 - 0.41454 - 0.52337 - 0.5720 - 0.5809 - 0.5893 - 0.5965 - 0.6033 -0.6091 -0.6197 - 0.6307 -0.7938 - 0.4 1456 - 0.52362 -0.5743 - 0.5843 - 0.5949 - 0.6049 -0.6141 - 0.6224 - 0.6358 - 0.6487 -0.8138 0~00002 0.00025 0.0023 0-0034 0.0056 0.0084 0.0108 0.0133 0.0161 0.0180 0~0200 0 0.16 1.64 2.80 4.40 6.04 7-88 9.20 11-24 12.42 13.76 - - - - 2.07 13.60 12.12 - 1.131 10.96 1-407 9.36 1-672 7-72 1.893 5.88 0.127 4-56 0.305 2-52 0.65 1.34 0.97 - I cell that are required for plotting the polarographic waves manually.In some experiments, however, connections were also taken from the dropping-mercury electrode and working calomel electrode to a second pair of terminals on the vernier potentiometer; these connections are indicated in Fig. 3 by broken lines. For each point to be plotted on the polarographic wave it is only necessary (i) to adjust the polarising unit of the polarograph to apply an appropriate electromotive force between the dropping-mercury electrode and the working calomel electrode, (ii) to observe the value recorded for the diffusion current i at maximum drop size and (iii) to measure EI, the potential of C (Fig.3) against the reference electrode, Y , at the instant of maximum drop size. A little further information, relating only to the internal resistance of the polarographic cell, can be derived by reporting also for each point on the polarographic wave the potential E,, of point C against the working calomel electrode. TREATMENT OF DATA- Values of i plotted against those of E,, would give the curve called by Miillerl the current - voltage curve, but a plot of i against values of E, (corrected for potential difference across the capillary of the dropping-mercury electrode) would give the current - potential curve from which the half-wave potential could be found by inspection. To illustrate the treatment of the data, some results relating to a manual polarogram of a solution of cadmium sulphate (1.82 millimolar in 0.1 N potassium chloride, with 0.01 per cent.of gelatin) are presented in Table 111, This polarogram is not reproduced here, but354 FURNESS: THE MEASUREMENT OF Wol. 77 in Fig. 4 the function log,, i/(ia - i) is plotted against the corresponding values of (E, + iR,) and of En. With values of (E, + iR,) as abscissae the plot is linear and leads directly to the results- and With values of E , as abscissae the plot deviates from the straight line; the displacement measured along the abscissa is proportional to the diffusion current at that point, as shown more clearly inset in Fig.4. The difference between (E, + iR,) and En for any given value of i is equal to the electromotive force required to drive the diffusion and residual currents through the polarographic cell and dropping-mercury electrode. The slope of the 2*303RT/nF = 0.031 volt Eh = -04300 volt against the saturated calomel electrode. Fig. 4. Graphs of the data from. Table I11 relating to the polarogram of 1.82 millimolar ca.dmium sulphate, 0.1 N potassium chloride, with 0.01 per cent. of gelatin. The function log,, i / ( i a - i ) is plotted against (EI + iR,) (linear), and against En (non-linear). The inset shows the linear relationship between (EI $.ZRm - En) and i. The slope, equivalent to 1430 ohms, gives the internal resistance of the polarographic cell and dropping-mercury electrode linear plot in Fig. 4 (inset), therefore, gives the ohmic resistance of the path followed by the diffusion current through the polarographic cell and the capillary of the dropping-mercury electrode. The precision with which values of E, and E,, can be balanced on the vernier potentio- meter at the instant of maximum drop size depends upon (i) the value of AEd.-, (ii) the drop rate of the dropping-mercury electrode and (iiz) the period of the galvanometer in the Poggendorff circuit. Throughout the above work a Tinsley galvanometer, type S.S.2. 45, of period 2.0 seconds was used to detect compensation; with a drop time at the capillary of 3.4 seconds, the point of balance at maximum drop size could be found to the nearest tenth of a millivolt or better so long as AEa.,,,.did not exceed 25mV. This degree ofJuly, 19521 POTENTIAL OF THE DROPPING-MERCURY ELECTRODE 355 precision is adequate, for in plotting polarograms manually the possible errors in measure- ment of diffusion current with the present equipment do not permit half-wave potentials to be reported with an accuracy better than one millivolt. In order to test the procedure outlined above further, the half-wave potentials of the thallous and cadmium ions, at various concentrations, were determined in potassium nitrate and potassium chloride supporting electrolytes. The values so obtained, recorded in Table IV, are in close agreement with other published values.’ TABLE IV HALF-WAVE POTENTIALS, E,, REFERRED TO THE SATURATED CALOMEL ELECTRODE AT 250°C Ion Concentration, Supporting electrolyte vs.S.C.E., E+ millimolar volts Thallous 1.20 0.1 N KNO,, 0.01% gelatin - 0.455 99 * 2.00 99 99 - 0.455 99 4.01 79 99 -0.457, -0.458 99 7.61 n 99 - 0.459 99 10.02 99 99 - 0.460 Cadmium 1.02 0.1 N KNO,, 0.01% gelatin -0.581 99 1-82 99 99 - 0.583 n 1.82 0.1 N KCI, 0.01% gelatin - 0.600 99 4-73 9 ) 99 - 0.600 The same procedure has been applied extensively in plotting manually the polarograms of certain oxy-acids of sulphur. In this field of work the comments of Kolthoffll regarding the unsatisfactory nature of internal mercury-pool electrodes are particularly applicable. In much of this later work it has been advantageous to use relatively high concentrations (up to 10 millimolar) of depolariser. The three-electrode polarographic cell (Fig. 3) has enabled such solutions to be examined without incurring the risks that otherwise would attend the formation of mercurous sulphide films on a mercury-pool anode and without necessitating corrections to potential measurements owing to the internal resistance of the cell. This work will be reported elsewhere. The author is indebted to Dr. W. Cule Davies for many helpful discussions and gratefully acknowledges his friendly advice and encouragement. REFERENCES 1. 2. 3. 4. 6. 6. 7. 8. 9. 10. 11. Muller, 0. H., J . Chem. Educ., 1941, 18, 227. Lingane, J. J., and Kolthoff, I. M., J . Amer. Chem. Soc., 1939, 61, 825. Kolthoff, I. M., and Lingane, J. J., “Polarography,” Interscience Publishers, Inc., New York, Lingane, J. J., and Laitinen, H. A., Ind. Eng. Chm., Anal. Ed., 1939, 11, 604. Furness, W., J . SOC. Dyers & Col., 1950, 66, 270. Furness, W., Analyst, 1952, 77, 246. Lingane, J. J., J . Amer. Chem. Soc., 1939, 61, 2099. IlkoviE, D., Coll. Czech. Chem. Comm., 1932,4, 480. Lingane, J. J., Ind. Eng. Chem., Anal. Ed., 1944, 16, 329. Furness, W., Analyst, 1951, 76, 178. Kolthoff, I. M., Ind. Eng. Chem., Anal. Ed., 1942, 14, 196. 1941, p. 216. BROTHERTON AND COMPANY LIMITED KIRKSTALL LANE LEEDS, 6 CENTRAL RESEARCH DEPARTMENT February, 1962

ISSN:0003-2654    

DOI:10.1039/AN9527700345

出版商:RSC    

年代:1952

数据来源: RSC

摘要:

356 BAGNALL AND STOCK: THE REPRODUCIBILITY OF [Vol. 77 The Reproducibility of G eome trical Correction Procedures in the Spectraphotometric Estimation of Vitamin A BY H. H. BAGNALL AND F. G. STOCK Recent assessments of the precision of geometrical correction procedures for the spectrophotometric estimation of vitamin A are discussed. The spectrophotometric data on the international standard preparation of vitamin A for the instrument used are given. Elxperience with recent modifications of the method, including a detailed table of the results of applying the three correction equation procedure of Cama, Collins and Morton, is described. The intra-laboratory reproducibility is indicated by the application of these three correction equations to each of three weights of oil, so giving nine values for the corrected E:2m at 328 mp for which the average fiducial limits (P = 0.05) of the mean are 2 1.2 per cent.The results of a small-scale inter-laboratory test are given. The use of a factor to allow for the presence of neovitamin A is discussed. THE spectrophotometric determination of vitamin A has been discussed at some length elsewherel and this paper will be concerned solely with the reproducibility of the results of geometrical correction procedures, in the belief that the new method of the B.P. Addendum 1951J2 with some modifications, is worthy of better support than is suggested in the publication of Adamson, Elvidge, Gridgeman, Hopkins, Stuckey and Tay10r.~ Any statement about the precision of the method must surely rest ultimately upon the reproducibility of results within a single laboratory and, although it is impossible to eliminate inter-laboratory variation, the reproducibility within individual laboratories when every possible precaution has been taken must be indicated before a comparison between a number of laboratories can legitimately be made.Cama, Collins and Morton4 have recently derived new standard absorption curves for vitamin-A ester and alcohol using both synthetic and natural vitamin. In the same paper the spectroscopic properties of all-trans vitamin--A alcohol and acetate are dealt with fully and the data given should become accepted as standard for these substances. The following correction equations for eliminating the effect of irrelevant absorption, derived from the absorption curve for all-trans vitamin-A acetate in cydohexane, are given- (a) E (corrected)* = 7 (E a t 327.5 mp - 0.405 E a t 312.5 mp - 0.595 E at 337.7 mp), (b) E (corrected) = 6.58 (E at 328 mp - 0.412 E at 313 mp - 0.588 E at 338.5 mp) and ( c ) E (corrected) = 3.52 (2 E at 328 mp -- E at 316 mp - E at 340 mp). The validity of geometrical correction procedures rests upon the assumption that the irrelevant absorption at the fixation points is linearly related; this assumption has been challenged because it cannot be tested experimentally, but, as data on the reproducibility of the method are to be investigated, comment on the effective linearity of the irrelevant absorption is unnecessary.Seven laboratories assayed each of five vitamin-A oils, readings being made in duplicate with photo- electric instruments.The gross E:2m at 328 m,w values were geometrically corrected for irrelevant absorption and the conclusion, from a statistical analysis of the results, was that the limits of error of a determination of vitamin-A content in duplicate by any one of the seven laboratories were about +15 per cent. for P = 10.05; the corresponding figures for gross E values were &2 per cent. As the authors of that paper state that their concern is with the reproducibility of the method, the apparent lack of precision implied by this conclusion is somewhat disconcerting. If, however, we examine more closely the nature of the data used in the statistical calculations in relation to the deductions made from the latter, we may justifiably entertain some doubts of the value of this inter-laboratory test as a Adamson et aL3 discuss the precision of the three-point correction method.* This equation is used in the method of the B.P. Addendum 1951.July, 19521 GEOMETRICAL CORRECTION PROCEDURES 357 criterion of the possible precision of the method. To take one example, a particular sample of halibut liver oil submitted to each of the seven laboratories was found to have an average irrelevant absorption of 10.7 per cent. ; yet the figure furnished by one of the laboratories was minus 4 per cent.; it is difficult to avoid the conclusion that any results obtained by a laboratory capable of so wide a margin of error would tend to be somewhat unreliable. I n any event, the inclusion of a nonsensical result such as this, from which in practice no con- clusions would be drawn, is surely unjustifiable, and it seems that insufficient attention was paid to the variability of the results of the laboratory in question and from one other.Although it is not possible, in the absence of the original data, to be dogmatic, it seems likely that a significantly greater precision would have been obtained if only the more consistent results submitted by the other five laboratories had been used. Nevertheless, this experiment may be fair enough if considered as a test conducted with a view to deter- mining what might happen if the Morton and Stubbs correction procedure were applied in a number of laboratories to the analytical results from particular oils; but the result gives no indication at all of the ultimate precision of the method, as at this stage laboratories TABLE I THE INTERNATIONAL STANDARD PREPARATION Wavelength, mP 295 300 305 310 311 .312 312.5 313 315 317-5 320 322.5 325 326 32'7 328 330 335 338 338.6 340 345 350 International standard All-trans vitamin-A acetate preparation (Cama, Collins and Morton4), Solvent-cyclohexane, Solvent-cyclohexane Compensator-cotton seed oil Eh/Eh(max.) EA/Eh(max.) 0.448 0.447 0.555 0.555 0.670 0.667 0-806 0.806 0.830 0.829 0.846 0.846 0.857 0.867 0.867 0.894 0.890 0.916 0.935 0.937 0.965 0.985 0.988 0.993 0-995 1.000 0.998 1.000 1,000 0.989 0.991 0.915 0.914 0.853 0.857 0.843 0.81 1 0.814 0.695 0.700 0.556 0.562 vary considerably in experience, expertise and manipulative care.I n fact, the paper of Adamson et aL3 points to the necessity for rigid and meticulous standardisation of tech- nique more than anything else. The Morton and S t u b b ~ ~ ~ ~ ~ ~ correction procedure was used in the above-mentioned test ; by using the revised data of Cama et aL4 an experimental design is made possible that, if care be taken, should greatly increase the precision. It is interesting to compare the experiences of Adamson et aL3 with those of Morgareidge, Blitz, Foy and Aaron8 who, by the use of the procedure of the U.S.P. XIV,9 show the magnitude of the error in a corrected E value determined on the unsaponifiable fraction to be about +S per cent. of the true mean for a single determination (P = 0.05), as deduced from data from a number of laboratories.The authors give their impression that the degree of inter-laboratory variation is gradually decreasing as more experience with the method is gained, and stress the constant attention needed to maintain spectrophotometers in their state of fine adjustment. Wider limits of error are to be expected if the U.S.P. method is used rather than that of the B.P. Addendum 1951, simply on account of the fact that the former involves saponification of the oil and hence more manipulation than the latter, which requires simple solution of the oil. If, therefore, we accept the conclusions of Morgareidge et aZ.,8 the B.P. Addendum method should be358 BAGNALL AND STOCK : THE REPRODUCIBILITY OF [Vol. 77 capable of giving results with limits of error of less than t8 per cent., and experience leads one to believe that this is so.Rigorous attention. to the calibration of the spectrophotometer is a prerequisite to all reliable results, and, when the results of inter-laboratory tests are discussed, special reference from this point of view should be made to the precautions that each laboratory has taken. Greater attention paid to this factor will eventually result in the reduction of errors to a minimum and the attainment of far better agreement between laboratories. EXPERIMENTAL The experimental data that follow, with the exception of those relating to the international standard preparation, are concerned solely with the examination of high-potency vitamin-A ester material, chiefly halibut liver oils with an approximate potency of 30,000 i.u.per g. For the estimation of vitamin A in such preparations, the method of the B.P. Addendum Laboratory and sample number la l b lc Id le 2a 2b 2c 2e 3a 4a INTRA-LABORATORY Gross E::, value - Replicates 16.03 16.00 15-97 16-06 15-92 15-84 16.10 16-18 15-93 15-92 15.94 15.86 15.77 15.72 15-56 15-07 15.00 14-93 15.1 1 15-06 15-00 15.08 14.98 14.97 15.26 14-99 14.95 15.17 15.01 14.94 15.61 15-66 15-55 15.12 15.0 1 15.02 Mean 16.00 15-94 16.07 15.9 1 15-68 15.00 15-06 15-01 18-07 15-04 15.61 16.05 TABLE :[I REPRODUCIBILITY WITH HALIBUT LIVER OILS Corrected E:& value Replicates at three sets of fixation points -- (4 312.5 mp, 327.5 m p and 337.7 mp 13-45 13-41 13.51 13.50 13.55 13.45 13-30 13.69 12.57 12-99 13-33 13.36 12-89 12-77 13.32 13-15 13.29 13-13 13.33 13-16 12.90 12-93 12-66 13.20 13.21 12-97 13.05 13.26 12.98 12.57 12-74 12.69 12.44 13.28 12-76 12-84 (14 313 mp, 328 'mp and 3384 mp 13.59 13.25 13847 13,55 13,,29 13053 13.49 13-68 12483 12-82 13.29 13-49 13.28 13.04 13.02 13.00 13.26 13-03 12.98 12.83 13.03 13.09 12-73 12.98 13.32 12.76 12-80 13.09 12*'91 1242 12-69 12-88 12.437 13.02 12-48 12.t17 (4 .316 mp, 328 mp and 340 mp 13.53 13-67 13.46 13-80 13.42 13.55 13.57 13.51 12-99 12.90 13.48 13.56 13.10 12-98 13.30 12.98 13-40 13.11 12-92 12.90 12-94 12.85 12.47 13.04 13.43 12.77 13.15 13.19 13.14 12.89 12.55 12-50 12.35 13.09 12.52 12.86 Mean of nine values with P = 0.05 fiducial limits 13-48 & 0.08 13.51 f 0.10 13.29 f 0.31 13-25 Ifi 0.22 13-08 & 0.15 13.15 & 0.11 13.00 f 0-12 12.88 f 0.18 13.05 rf: 0.19 12-98 f 0.16 12-61 f 0.13 12-86 f 0.20 Fiducial limits as a percentage of the mean 0.60 0.74 2.33 1.66 * 1.15 0.84 0.92 1.40 1-46 1-23 1-03 1-56GEOMETRICAL CORRECTION PROCEDURES 359 TABLE II-continzced Corrected E:& value A f 1 Replicates a t three sets of fixation points A r > Laboratory and sample number 4b 4c 4a 6a 5b 6C 6a 7a 8a 8b ac 8d & Sf 9a 9b Gross E:Fm value )i Replicates 15-08 15.07 14.97 15.17 15.03 15.05 15.31 15.00 15.04 15.31 15.20 15.31 15.87 15.79 15-73 15.59 15-49 16.47 15.62 15.64 16-44 13.80 13.70 13-77 15-97 16-92 15.84 15.68 16-64 1648 15.68 15.61 15.59 15.76 15.62 15.54 16.48 15.31 16.28 15-81 15-74 15-66 16-71 16.63 15.65 15-42 16.59 16-46 Mean 15-04 15-08 15-12 15-27 15.80 15.52 15.63 13.76 15.91 15.67 15.63 15.64 15.36 16.74 15.66 16.49 (4 312.5 mp, 327.5 m p and 337.7 mp 12.92 13.40 13.07 13-24 13.22 13.46 13.67 12.93 13.0’7 12.37 12.45 12.87 13.29 13.36 13.35 13.15 12.90 12.93 12-74 12.61 12.32 11.03 11-00 11.25 13.41 13.17 13-24 13.22 13.44 13-81 13.10 12.86 12.61 13.86 13.42 13.72 12-96 12.38 12.51 13.07 13-40 13.32 14.46 14.09 14.03 14.03 13-90 13.95 (b) 313 mp, 328 mp and 338.5 mp 12-92 13.12 13.04 13.12 13-08 13.28 13.64 12.90 12-93 11.95 12-52 12.81 13-14 13.37 13.14 13.39 13.05 12s.94 12.77 12.53 12-37 11.00 10.92 11.20 13.36 12.96 13.13 13.31 13.26 13-72 12.79 12.72 12-79 13-63 13.36 13.36 12.84 12-23 12-46 13.36 13.29 12.93 13.92 13.94 13.98 13-79 13.68 13.72 (4 316 mp, 328 m p and 340 mp 13.05 13-02 13.02 13.20 13.10 13.19 13.67 12-91 12.88 12.30 12.44 13-04 13.17 13-13 13-15 13.32 12.99 13-06 12.63 12.35 12.46 11.06 11-02 11-13 13.13 13-04 13.34 13-43 13.47 13.71 13.13 12.85 12.80 13-84 13.66 13.46 12-93 12.11 12.59 13.31 13.20 13.38 14.20 13.61 13.99 13-78 13-93 13.85 P = 0.06 fiducial limits 13.06 f 0.11 13.21 f 0.09 13-18 f 0.28 12.53 & 0.26 13.23 f 0-08 13.08 f 0.13 12.53 & 0.13 11.07 f 0.08 13.20 & 0.12 13*501&:0-16 12.85 f 0.13 13.59 f 0.15 12.56 f 0-20 13.25 f 0.12 14.03 f 0.18 13.85 f 0.09 Mean ( x ) = Fiducial limits as a values with percentage of the Mean of nine mean 0.84 0.68 2.12 2.08 0.60 0.99 1.04 0.72 0.91 1.19 1.01 1.10 1.67 0.9 1 1-28 0.65 1-17360 BAGNALL AND STOCK : THE REPRODUCIBILITY OF Pol.77 1951 was followed with the following modifications, viz. , all three correction equations recom- mended by Cama et aL4 were used with each of three separate weights of oil dissolved in cyclohexane so giving nine “corrected” values for E::, at 327-5 to 328 mp.The fiducial limits (P = 0.05) of the mean of the nine values were then calculated. Experience shows that the wavelength scale of the spectrophotometer can easily get out of adjustment and it is important to check frequently the position of the 4 8 6 1 ~ hydrogen line. Gridgemanlo has drawn attention to the large errors that can be introduced by comparatively small dis- placements of this scale. The instrument used in these determinations was the Unicam photo-electric spectrophotometer. RE s u LTS The international standard preparation, the diluent oil being used as compensator and cyclo- hexane as solvent, would be expected to have an EiFm at 328 mp of 5.21 ( x 1920 = 10,000 i.u.per g). Three determinations were made, separate weighings from different capsules being used. TABLE III EY&, OF HALIBUT LIVER OIL SAMPLES Laboratory A L Oil Wave- , -7 number length, mP 1 312.5 313-0 316.0 327.5 328.0 337.7 338.5 340.0 2 312.5 313.0 316.0 327-5 328-0 337.7 338.5 340.0 3 312.5 313.0 3 16.0 327.5 328.0 337.7 338-5 340.0 E::~ replicates , 16.51 16.69 17-46 19.13 19-16 17.06 16.73 16.26 16.07 16.23 16.99 18-60 18.60 16-43 16-11 15-67 19.84 20.13 2 1.08 23-30 23.30 21.08 20.60 20.13 16-46 16-66 17.37 19.10 ’ 19-14 16.96 16-03 16-14 16.13 16-25 17.02 18-61 18-61 16.50 16.21 15-60 19.86 20.08 21-04 23.20 23.20 20-88 20.60 19.97 16-46 16.63 17.38 19.02 19.02 16.94 16.60 16-13 16.14 16.31 17.04 18-66 18.69 16.45 16.17 15-68 20.04 20.19 21-11 23.35 23.35 20.95 20.68 20.13 Mean 16-47 16.66 17-40 19.08 19.1 1 16.99 16.65 16.18 16.11 16.26 17.02 18-62 18.163 16.16 25.165 19.91 20.13 21.08 23.88 23.28 20.97 20.63 20-08 1646 Laboratory B r A 1 E:?, replicates , 16.42 16.57 17.38 19-04 19-09 16.81 16-62 15.93 16.00 16.14 17-00 18.55 18.57 16.52 16.17 15.60 19.55 19.68 20.60 22.74 22.79 20.52 20.20 .19.58 16.50 16.62 17-33 19-00 19.00 16.91 16-57 16-03 16.08 16.24 16.92 18.59 18.63 16.45 16.12 15.59 19-90 20.05 20.95 23-29 23.34 21.05 20.66 20.05 16-24 16.47 17.26 19.00 18.95 16.84 16.44 15-97 16-17 16-19 17.13 18.80 18-80 16.63 16-39 15.76 19.53 19.70 20.85 23.15 23.18 20.80 20.53 19.88 Mean 16.39 16.55 17-33 19-02 19-02 16-85 16.54 15.98 16-08 16-19 17-02 18.64 18.67 16-34 16.22 15.65 19.66 19.81 20.80 23.06 23-10 20.79 20.46 19.84 Rather than assume that each capsule contained exactly 0.250 g of material, a known amount was weighed from each capsule and compensated with the same concentration of diluent oii.The results were 5.104, 5.099 and 5.112, having a mean of 5.105, which agrees closely with the value of 5.09 on a weighed amount, obtained by (Cams et aL4 The international standard preparation according to these figures is apparently 2 per cent. deficient in activity. There is certainly something abnormal about it, because con tinuation of the absorption curve over the short-wave side of 300 mp shows a very marked departure from the.standard vitamin-A acetate curve.Nevertheless, the standard over the range 300 to 340 mp is very useful as a means of checking the ratio E~/E~~max.) at a particular wavelength, and the values agree very closely with the figures of Cama et aL4 for pure all-tram vitamin-A acetate (Table I), For extreme accuracy it may be necessary to derive particular equations for each instrument with pure all-tram vitamin-A acetate. Table I1 shows the results from twenty-eight samples of a brand of halibut liver oil capsules taken under the Food and Drugs Act from the same number of retail chemists’ shops in Birmingham. They include various numbers of samples originating from nine different halibut liver oils. The P = 0-05 fiducial limits of the mean of nine values for the correctedJuly, 19521 GEOMETRICAL CORRECTION PROCEDURES 361 E:Fm at 328 mp, expressed as a percentage of the mean, vary from & 0.60 to & 2.33, with average value of -+ 1-17.This is a representative set of data from a total of 127 samples, chiefly of halibut liver oils, examined during an analytical survey of these preparations in 1951.11 The intra-laboratory reproducibility indicated by the data shown in Table I1 suggests that by adopting the outlined experimental design an inter-laboratory test should show the precision of the method to be better than has been indicated by previous tests. In order to verify this conjecture, three halibut liver oils were analysed, and samples of the same oils were submitted simultaneously to another laboratory for analysis. 1 The results are recorded in Tables 111, IV and V.The figures under Laboratory A were obtained by us with a Unicam instrument and those under Laboratory B by the other laboratory with a Uvispek instrument. TABLE I V CORRECTED E:Tm AT 327.5 TO 328mp OF HALIBUT LIVER OILS Laboratory A Laboratory B A Correc- r A > f \ Oil tion Corrected Ei$m at 327.5 to Corrected E itm a t 327.5 to number procedure* 328 mp replicates Mean 328 mp replicates Mean 16-03 16-36 15.98 16.12 16-72 15.81 16-84 16.46 17-09 16.45 16.13 16-26 16-78 15.96 16-33 17.14 16.26 16.44 16-61 16.23 15.85 16.38 16.15 15-64 16.02 16.46 16.04 16-07 15.68 16.23 15-93 16-08 16-20 16.40 16.23 (b) (4 15.98 16.20 16.43 16.20 15-98 16.72 16-58 16.43 19-06 19-15 19-47 19.23 18-30 18-95 20-08 19.11 19.04 18-64 18.87 18.82 18.44 19.28 19-69 19.14 19.82 19.56 19.00 19.03 19-26 19.10 19.01 16.36 15.90 16.13 16.40 15.86 I$ 1 (4 2 (4 3 (4 (b) (4 19-86 * Fixation points: (a) 312.5, 327.5 and 337.7 mp; (b) 313, 328 and 338.5 mp; (c) 316, 328 and 340 mp. TABLE V MEAN CORRECTED E:2m AT 327.5 TO 328 mp AND ITS P =0*05 FIDUCIAL LIMITS Oil number Laboratory A Laboratory B 1 16-20 f 0.21 (fl.30%) 16-51 f 0-37 (&2.24%) 2 16.23 f 0.25 ( fl*54y0) 3 19-27 & 0.49 ( &2*54y0) 16-12 & 0.19 ( &l*18y0) 19.05 f 0.20 ( fl*05y0) TABLE VI DATA ON THE INTERNATIONAL STANDARD PREPARATION FROM LABORATORY B Solvent: cyclohexane.Compensator: cotton seed oil (EiTm at 327.5 mp = 5.13) Wavelength, mCc 290.0 300.0 305.0 310.0 312.5 315.0 317.5 320-0 322.5 325.0 326.0 0.327 0.55 1 0.660 0.792 0.849 0.881 0.911 0-935 0.962 0.990 0.994 Wavelength, mp 327.0 327.5 328.0 330.0 335.0 337.0 340.0 345.0 350.0 355.0 360.0 0.998 1.000 0.998 0.994 0.919 0.865 0.811 0.697 0.557 0.424 0.313 DISCUSSION OF RESULTS The reproducibility of a gross E value appears to be better in laboratory A than in laboratory B, the magnitude of the fiducial limit4 of the mean corrected E values for laboratory B being nearly twice those for laboratory A.The agreement between the two laboratories362 BAGNALL AND STOCK : THE IiEPRODUCIBILITY OF [vol. 77 is good, the means of the corrected E values differing by 1 to 2 per cent. The absorption curve from laboratory B on the international stand:ard preparation shows some disagreement with the standard curve for pure all-trans vitamin-A acetate of Cama et aZ.,4 in particular over the critical 310 to 320 mp region, and this may account in some part for the tendency of laboratory B to produce results consistently higher than laboratory A.The results, although inadequate to give a reliable mathematical estimate of the method’s precision, strongly suggest that a larger-scale inter-laboratory test would show it to be satisfactory. The results show that good intra-laboratory reproducibility can be attained by the method described, which is based on the published data of Cama et aL4 The twenty-eight values for gross and corrected E:Fm at 328 mp given in Table I1 yield the following in- formation- Average gross E:m at 328 mp = 15.42 ( x 1600 = 24,670 i.u. of vitamin A per g). Average corrected at 328 mp = 13.07 (>< 1920 = 25,090 i.u.of vitamin A per g). The twenty-eight samples had all been manufactured during the twelve months previous to sampling, with the exception of sample number 7a, whose absorption curve showed signs of oxidation of the vitamin, and it is to be observed that the vitamin-A content from the average gross E;Zm at 328mp x 1600 approximates to that given by the average corrected Elk at 328mp x 1920. This is in agreement with MortonJs12 observation that no better factor (than 1600) for fish liver oils in general could, wen to-day, be chosen for converting gross E:?., values to international units, although oils showing little irrelevant absorption would be somewhat undervalued and oils exhibiting more irrelevant absorption a little over- valued.This statement, however, although valid for fresh and unoxidised oils, is not always applicable to samples submitted to a Public Analyst under. the Food and Drugs Act. Many of these have deteriorated owing to oxidation of the vitamin, so necessitating the use of a correction procedure. This point has been dliscussed in some detail elsewhere.1 The problem of neovitamin A causes some concern, and a factor of 1-04, based upon an assumed average content of 30 to 40 per cent. of neovita.min A, has been used1J3 for calculating the vitamin content in international units from the spectrophotometric data. Yet if, as Harris, Ames and Brinkman14 have recently asserted, the biological potency of vitamin A is significantly higher by 20 to 28 per cent. than that of neovitamin A, and since by a fortunate coincidence the correction procedure in the spectrophotometric estimation discounts neovitamin A to about the same extent as does the rat bio-assay, the use of such a factor will become unnecessary.Experience, as illustrated above, shows that gross E;?.,, at 328 mp x 1600 approximates to corrected E:, at 328 m p x 1920, on average, for a number of unoxidised oils. As the 1600 factor was of biological origin, it would automatically allow for a difference in the biological activity of the cis and trans isomers of vitamin A, and it would seem, therefore, that this approximate relationship between the 1600 and 1920 factors substantiates to some extent the statement of Harris et This point is rather critical, because even though the new international standard is pure crystalline all-trans vitamin- A acetate, the international unit is still the activity of a given weight thereof.lS Hence, if there is a significant difference between the biological activity for all-trans vitamin A and neovitamin A, this must be accounted for in any expression of vitamin-A content in inter- national units.The use of more than one set of fixation points goes some way to meet the criticism that the irrelevant absorption at these points may not be strictly linearly related; for departure from linearity will be reflected in an increase in the fiducial limits of the mean and so the effect will be lessened to some extent. Ab.normally wide fiducial limits should give warning of possible peculiarities in the sample and care should then be taken before expressing an opinion about its vitamin-A content.CONCLUSIONS It is strongly suggested that any future inter-laboratory tests of the precision of geometrical correction procedures in estimating vitamin A would give an estimate significantly different from that given by Adamson et aL3 if aII the relevant factors were taken into consideration, including the examination! of the absorption curves from the inter- national standard preparation obtained by the participating laboratories and the adoption of an experimental design similar to that outlined in this paper.July, 19521 GEOMETRICAL CORRECTION PROCEDURES 363 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. REFERENCES Bagnall, H. H., and Stock, F. G., J . Pharm. Pharmacol., 1952, 4, 81. Addendum 1951 to the British Pharmacopoeia 1948, Appendix XVA, p. 92. Adamson, D. C. M., Elvidge, W. F., Gridgeman, N. T., Hopkins, E. H., Stuckey, R. E., and Cama, H. R., Collins, F. D., and Morton, R. A., Biochem. J . , 1951, 50, 48. Morton, R. A., and Stubbs, A. L., Analyst, 1946, 71, 348. Mofgareidge, K., Blitz, M., Foy, J. R., and Aaron, J. P., Jun., “Current Experience in the Estima- tion of Vitamin A by the U.S.P. XIV Procedure (Spectrophotometric),” Food Research Labora- tories, Inc., and Nopco Chemical Co., Inc., U.S.A. United States Pharmacopoeia XIV Revision, Vitamin-A Assay, Spectrophotometric Method, p. 784. Gridgeman, N. T., Analyst, 1951, 76, 449. Bagnall, H. H., and Stock, F. G., Pharm. J . , 1952, 168, 40. Morton, R. A., J . Pharm. Pharmacol., 1950, 2, 129. Dalvi, P. D., and Morton, R. A., Biochenz. J . , 1951, 50, 43. Harris, P. L., Ames, S. R., and Brinkman, J. H., J . Amer. Chem. Soc., 1951, 73, 1252. World Health Organisation Technical Report Service, 1950, p. 3. Taylor, R. J., Analyst, 1951, 76, 445. 1 , , Biochem. J . , 1947, 41, 625. , Ibid., 1948, 42, 195. -- -- CITY ANALYST’S LABORATORY BIRMINGHAM January, 1952

ISSN:0003-2654    

DOI:10.1039/AN9527700356

出版商:RSC    

年代:1952

数据来源: RSC