摘要:

80 BURROWS: A COLORIMETRIC METHOD FOR THE [Vol. 75 A Colorimetric Method for the Determination of Oxalate BY S. BURROWS SYNoPsIs-The routine determination of calcium oxalate in mushroom composts by the precipitation of calcium oxalate from a hydrochloric acid extract of the compost is difficult because of the co-precipitation of colouring matter and organo-metallic complexes which interfere in the final determina- tion of oxalate. A number of published colorimetric methods were investigated, but none was found to be satisfactory. An absorptiometric method based on the fading effect of oxalates on various coloured organic complexes of tervalent iron has been investigated and the green complex of iron and 7-iodo-8-hydroxyquinoline-5-sulphonic acid (ferron) shown to be suitable. By removing the colouring matter and phosphates from a compost by preliminary extraction with citric acid and by close control of the acidity, results reproducible to &4 per cent.have been obtained; but recovery of added oxalate amounts to only 90 per cent. On exposure of iron - ferron solutions containing oxalate to sunlight, further fading, which is related quantitatively to the oxalate present, occurs. THE determination of oxalates in biological materials is usually made by precipitation as calcium oxalate followed by weighing or titration. The method may be expedited and made more sensitive by introducing a centrifuge technique with a cerimetric or iodimetric fini~h.19~9~ Preliminary concentration by acidification and extraction with ether is sometimes req~ired.~J There arose a demand on this laboratory for a routine method of determining oxalate in mushroom composts; Bau's method, as modified by with a pennanganate titration was first used.Considerable difficulty was experienced in removing colouring matter and metal-organic complexes that were precipitated with the calcium oxalate. Extraction of the compost or the contaminated oxalate precipitate with 1 per cent. citric acid, containing calcium chloride and saturated with calcium oxalate, effected improvement, but reprecipitation was often necessary and this added considerably to the work. A specific colorimetric method which would obviate the necessity for purifying the calcium oxalate was therefore sought. The method suggested by Mitchell' in which a red colour is produced by a sodium vanadate - hydrogen peroxide reagent in the presence of oxalate was unsuitable because the mineral acids necessary to dissolve calcium oxalate also produce a red colour.Calkins8 gives details of a method by which oxalic acid is reduced to glycollic acid and treated with 2 : 7-dihydroxynaphthalene in sulphuric acid to produce a violet colour, but the numerous compounds that give colours with concentrated sulphuric acid make the method difficult to work. The fading effect of oxalates on the colours produced by iron and various organic reagents is well known and has been used in a volumetric method for determining oxalate by Erametsa and P a r p ~ l a . ~ They titrated oxalic acid against a standard ferric chloride - sulphosalicylic acid mixture until a predetermined extinction, measured in the Pulfrich photometer, was reached. LanglO also describes a method using the same principle in which the iron thiocyanate colour, partially bleached with oxalate, is matched against standards containing known amounts of oxalate.Preliminary investigations-Despite the possibility of serious interference from phosphates, these fading methods were examined, and it was found that phosphate interference could be reduced by working in acid solutions below pH 2. The iron reagent eventually chosen was 7-iodo-8-hydroxyquinoline-5-sulphonic acid (ferron) introduced by Y oell and used by Fahey12 to determine fluorides by their bleaching effect. This reagent was chosen because it could be used in moderately concentrated acid solutions (pH 1 to 3) and the green colour of the iron complex permitted the use of filters that would minimise interference from the brown colouring matter in composts. According to Mellan13 and Yoe and Haul4 the ferric ferron complex contains one atom of iron to three ferron residues, but in acid solution an equilibrium exists, and uncombined Fe"' ions and ferron are also present, so that Beer's law is not strictly followed.The mechanism of the fading action of oxalate on the colours of ferric complexes is apparentlyFeb., 19501 DETERMINATION OF OXALATE 81 due to the greater stability of the ferri-oxalate complex and the consequent decomposition of the colour complex. Experiments showed that, with ferron, this also is a balanced reaction, so that a high concentration of iron ferron is desirable to ensure sensitivity to oxalate.This necessitates the use of deep coloured solutions and a sensitive photo-electric measuring instrument. The Hilger Spekker Sensitive Model is satisfactory for the purpose, but it was found necessary to use a neutral screen on the left of the light source in order to obtain a balance when setting the drum a t zero with the reagent blank. Interference by phosphate is small at pH 0.9 to 1.0, and the most serious interference is from iron. This difficulty cannot be overcome by using a higher proportion of iron in the reagent since the reagent then becomes relatively insensitive to low amounts of oxalate and an inflection appears in the standard curve. However, when working with composts, in which all the oxalate is present as calcium oxalate, a preliminary extraction with 1 per cent'.citric acid containing calcium chloride and saturated with calcium oxalate removes practically all the iron. This reagent also removes most of the phosphate and colouring matter. The hydrochloric acid extract of the residue can then be used directly for the colorimetric finish. REAGENTS- METHOD Iron ferron-Dissolve 4.00 g. of ferron in about 500 ml. of hot water containing 1.0 g. of A.R. ferric chloride (FeC1,.6H20). Add 300 ml. of 2.0 N hydrochloric acid and 68 g. of sodium acetate (CH3C0,Na.3H,0), cool and make up to 1 litre. Citric acid reagent-Dissolve 20 g. of citric acid and 50 g. of anhydrous calcium chloride in water, make up to 1 litre, heat to boiling and add saturated ammonium oxalate solution drop by drop until a permanent turbidity is formed.Boil, set aside overnight and filter. PROCEDURE- Extract 1.00g. of the dried, ground compost sample in a 100-ml. conical flask with 10 ml. of water and 10 ml. of citric acid reagent, by simmering on a hot-plate for half an hour with a small glass funnel in the neck of the flask for condenser. Filter through a No. 30 Whatman filter paper and wash well with cold water. Drain the flask well and return the filter paper and precipitate to the flask after partially drying for an hour at 100" C. Add, from a pipette, 25 ml. or 50 ml. of 0.4 N hydrochloric acid, using the former quantity if the calcium oxalate expected is below 1 per cent. or the latter if from 1 to 3 per cent.Bring the liquid just to boiling-point, and then keep below boiling-point for about 5 minutes, swirling occasionally. Filter through a small dry paper, reject the first runnings and pipette 4ml. of the filtrate into a dry beaker. Add 5ml. of the iron ferron reagent, protect the solution from direct sunlight and take the drum reading in a 1-cm. cell with the Spekker absorptiometer, using Ilford filters, No. 607, and setting the drum at zero against a reagent blank containing 4 ml. of 0.4 N hydrochloric acid and 5 ml. of iron ferron reagent. It will probably be necessary to place a neutral screen on the left of the light source in order to obtain a balance on the galvanometer. The temperature of the blank and sample solution should be the same to within f0.5" C.Prepare a standard curve using solutions of between 0 and 4 mg. of calcium oxalate (CaC,O,.H,O) in 4 ml. of 0.4 N hydrochloric acid. Discussion-Since the method depends on differences in the light absorption between the sample and a highly coloured reference solution, care must be taken that the reference solution (reagent blank) is accurately made up and that the cells used in the absorption measurement are of uniform dimensions. Cells that were satisfactory for ordinary work have been found to give a slight error with the highly absorbing solutions used in this method because of small deviations from the 1-cm. length. As already mentioned, Beer's law is not followed and a non-linear calibration curve is obtained (curve I). Interfering sabstances-Table I shows the principal interfering substances met with in composts.The effect of fluorides has not been investigated since they have not so far been found in mushroom composts. TABLE I Interfering substance Limits permissible Phosphate . . . . . . . . 100p.p.m. P Hydrochloric acid . . . . . . f 1 per cent. of the amount in the reagent Ferric iron . . .. .. .. 3 p.p.m. Fe"' Citrate . . .. .. .. . . 1000p.p.m. blank82 BURROWS: A COLORIMETRIC METHOD FOR THE [Vol. 75 Close control of acidity is necessary hecause sodium acetate is a poor buffer a t the pH used (0.95). Other buffer systems have not been tried. Dextrose, sucrose and the following acids or their calcium salts do not interfere: formic, acetic, propionic, succinic, malonic, tartaric, glutaric, gluconic, fumaric and maleic.Yellow colorations are tolerated up to a deep straw-yellow. Mg. of calcium oxalate (CaC20,.H20) Fig. 1 Curve I. Calibration curve. Curve 11. Calibration solution after irradiation ResuZts-Table I1 gives the mean of three determinations on separate l-g. sub-samples from three typical composts dried and ground to pass the l-mm. screen. In addition, 2-ml. aliquots of a suspension of calcium oxalate containing 8.10 mg. of CaC,O,.H,O per ml. were added to three sub-samples of sample No. 2; and oxalate was determined in the usual way. Oxalate was also determined by the method of Myers as mentioned in the introduction. TABLE I1 Calcium oxalate found, mg- r 7 Sample Myers’ method Ferron method 1 29.4 28.2 & 0.4 2 12.0 13.8 & 0.6 3 2.7 2.3 f 0.5 2 + 16.2mg.- 28.6 & 0.5 Calcium oxalate recovered = 14.8 mg. = 91 per cent.Feb., 19501 DETERMINATION OF OXALATE 83 The reproducibility obtained (f.4 per cent. in samples containing about 1 per cent. of calcium oxalate) is satisfactory in view of the rapidity of the method. The low recovery of added calcium oxalate, which has been confirmed a number of times, may be due to adsorp- tion of oxalate on the compost during the hydrochloric acid extraction. In addition to this source of error the results are likely to be low because the citric acid reagent, although it is saturated with calcium oxalate in the cold, dissolves another 0.5 mg. per 10 ml. on boiling. A factor to correct for these errors has not been used, as only comparative values have so far been required. The efect of s.unZight-The test solutions, when ready for absorption measurements are stable in the dark for a number of hours, but fading occurs on exposure to sunlight; this is a function of the amount of oxalate added and leads to a drum reading of two to three times the original figure.The reagent blank is unaffected. Curve 11, obtained from these drum readings, shows a slight upward curvature as against the downward curvature of the original calibration curve, I. A number of compounds are known to reduce ferric chloride when irradiated with ultra-violet light, but in the presence of ferron the concentration of ferric ions is much diminished and it appears that complex formation with the iron is necessary before the photo-reaction can take place.Thus, dextrose and all the non-interfering acids already mentioned have no effect on the ferron complex in sunlight; whereas tannic, gallic, pyrogallic and, to a lesser extent, citric and lactic acids behave in a manner similar to oxalic acid. The following equations represent the reactions that are likely to take place with oxalic acid (Q = ferron residue in the colour complex)- .. .. * * (1) FeQ3 + Fe"' + 3Q' . . .. .. .. 2Fe"' + 3C,04" + Fe,(C,O,), . U.V.L. Fe,(C,O,), --+ 2Fe" + 2C20," + 2 0 , .. .. The tendency to form a ferro-oxalate complex is probably slight at the low pH employed, so that the oxalate ions liberated according to equation (3) will combine with more ferric ions from the ferron complex and further oxidation of oxalate will occur until decomposition is complete, so that the concentration of ferron complex will be determined solely by equation (I), and the equilibrium will not be affected by oxalate (equation (2)) as it is in the dark reaction.Thus, at the completion of the light reaction, equation (1) predominates, giving an upward curvature to curve 11, whereas in the dark reaction equation (2) controls the concentration of the ferron complex, causing a downward curvature in the calibration curve I. The approximately three-fold increase in drum reading on irradiation may be explained by comparing equations (2) and (3). In equation (2) (dark reaction), three oxalate ions combine with two ferric ions, whereas in equation (3) (light reaction), only one oxalate radical is required to reduce two ferric atoms.In bright sunlight the maximum drum reading is reached in 20 to 40 minutes, but there- after the drum reading begins to decrease owing to re-oxidation of the iron, so that the conditions of irradiation need to be standardised in order to obtain reproducible drum readings. As an ultra-violet lamp was not available, the possibility of using the light reaction to determine oxalates was not studied further, although there is some likelihood that inter- ference due to phosphates and iron may be reduced by this means. The author wishes to thank the Director of ResearcEi and also the Directors of the Mush- room Research Association for permission to publish this work. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. Koch, K., Biochem. Z., 1936, 283, 422. Smith, G. F., and Getz., C. A., Ind. Eng. Chern., Anal. Ed., 1938, 10, 304. Kirk, P. C., and Tompkins, P. C., Ibid., 1941, 13, 277. Powers, H. H., and Levatin, P., J . Bid. Chem., 1944, 154, 207. Flaschentrager, B., and Muller, P. B., 2. Physiol. Chenz., 1938, 251, 52. Myers, A. T., J. Agric. Res., 1947, 74, 33. Mitchell, G. A., Analyst, 1933, 58, 279. Calkins, V. P., Ind. Eng. Chem., Anal. Ed., 1943, 15, 762.84 RICKSON : AN IMPROVED MICRO-METHOD FOR THE 9. 10. 11. 12. 13. 14. Erametsa, O., and Parpola, T. M., Suomkn. Kern., 1942, 15B, 12. Lange, B., Kolorimetrische Analyse, Edwards, Ann Arbor, U.S.A., 1941. Yoe, J. H., J . Amer. Chem. SOC., 1932, 54, 4139. Fahey, J. J., I n d . Eng. Chern., Anal. Ed., 1939, 11, 362. Mellan, I., Organic Reagents in Inorganic Analysis, Philadelphia, 1941. Yoe, J. H., and Hall, R. T., J . Amer. Chem. SOC., 1937, 59, 872. MUSHROOM RESEARCH ASSOCIATION LIMITED YAXLEY, PETERBOROUGH [Vol. 75 May, 1949

ISSN:0003-2654    

DOI:10.1039/AN9507500080

出版商:RSC    

年代:1950

数据来源: RSC

摘要:

84 An RICKSON : AN IMPROVED MICRO-METHOD FOR THE [Vol. 75 Improved Micro-Method for the Determination of Fluorine Based on an Examination of the Fluoride - Fluosilicate Equilibrium BY J. 13. RICKSON SYNOPSIS-A study of the fluoride - fluosilicate equilibrium system indicates that low results in the micro-determination of fluorine by titrating with thorium nitrate after separation of fluorine from interfering ions by distillation may be caused by some of the fluorine being present as the.fluosilicate ion SiF,”. The fluosilicate ion does not form an un-ionised compound with thorium as does fluoride. This error can be avoided by carrying out the titration in a 50 per cent. alcoholic system buffered at pH 5.3 with gallocyanine as indicator. The method can be used for the determination of 20 to 400 p g .of fluorine. The precision is approximately 1 pg. up to 100 pg. of fluorine, but the experimental error increases with larger quantities of fluorine up to 5 per cent. with 400 pg. The effect of chloride ions and varying pH on the method is discussed. THE determination of small amounts of fluorine is a problem which has received a considerable amount of attention over a long period. The thorium nitrate titration described by Willard and Winter,l with modifications which appear to be based more on personal choice than on any fundamental reasoning; is now widely used for this determination. An essential preliminary when determining fluorine in insoluble fluorides, such as are found in soil, fertilisers, plant ash and similar material, is the separation of the fluorine from ions which would interfere in the subsequent titration.This separation may be carried out by distillation from either sulphuric or perchloric acid, water or steam being added to the distillation flask to maintain the boiling-point of the liquid at a temperature low enough to prevent sulphuric or perchloric acid from appearing in the distillate. One of the most disturbing features of the literature is that no adequate explanation has been offered for the slightly low result which is usually obtained when the distillate is titrated with thorium nitrate. Clifford2 showed that, in general, complete recovery of fluorine is never obtained on distillation from either sulphuric or perchloric acid, although the former acid appears to be the more efficient.The maximum recovery varies from 93 to 97 per cent. If the distillation flasks are freshly boiled with sulphuric acid before distillation, a recovery of 98 per cent. can be attained, but this is not increased by collecting larger quantities of distillate. A possible explanation of this fact is that a non-volatile silicon oxyfluoride may be formed in the presence of hydrated s i l i ~ a . ~ ~ * ~ ~ Although there appears to be no independent confirmation of the existence of such a compound, there is evidence that, in the presence of gelatinous silica (q., as silica gel) the amount of fluorine recovered may be reduced to as little as 10 per cent. of the quantity pre~ent.~ This decrease does not occur if the silica is dehydrated-before being added to the distilling flask.No explanation is offered here for this effect of hydrated silica, but the results reported by Clifford indicate that the slightly low recoveries obtained in the absence of hydrated silica are not wholly due to the distillation technique, some error being inherent in the titration procedure; proposals are made in this paper for altering the conditions of the titration so as to reduce the error involved. In order to understand how this error arises in the thorium nitrate titration now in common use, it is necessary briefly to review the literature since 1933, when Willard and Winter published their paper. According to the original publication the titration system contained equal volumes of water and alcohol, the acidity being adjusted with hydrochloric acid to aFeb., 19501 DETERMINATION OF FLUORINE 85 point which was judged by the colour of the zirconium - alizarin indicator.No measurements of the pH values of the solutions were recorded. Reynolds and Hill6 showed that the addition of zirconium to the alizarin was unnecessary if the solution is buffered, and that a more clearly defined end-point may be obtained by omitting alcohol. The buffer solution used by these workers was half-neutralised monochloroacetic acid, as recommended by Hoskins and Ferris.’ It is to be noted here that the pH of the aqueous alcohol sjrstem used by Hoskins and Ferris was thereby maintained at 3.5, whereas the pH of the aqueous system used by Reynolds and Hill was approximately 2-8. Hammond and MacIntyre8 made a comparison of the thorium nitrate titration in aqueous and alcoholic systems using the chloroacetic acid buffer, and showed that, in the range 0 to 50 pg.of fluorine, a linear relationship, approximating to stoicheiometric requirements, was obtained between fluorine and thorium nitrate in 50 per cent. alcohol ; in aqueous systems the relationship was non-linear. With larger quantities of fluorine, linear relationships were obtained in both aqueous and alcoholic systems. In later papers on this problem more emphasis was given to the importance of the pH at which the determination was carried out. It was showns that below pH 3 the amount of thorium nitrate solution required to cause a colour change of the indicator in a blank titration rapidly increases with decreasing pH, and that above pH 3.5 the end-point becomes indistinct.Consequently it was considered that the optimum pH was between 3.0 and 3.5. Williams,lo in an analysis of the factors influencing the titration, suggested that the ionisation of ThF, depended on the hydrogen ion concentration , and recommended adding hydrochloric acid to control the dissociation at an optimum level. Milton, Liddell and Chiversll investigated the possibility of using some indicator other than alizarin S, and recommended the use of Solochrome Brilliant Blue B.S., which gives a colour change from pink to blue. This indicator was stated to be preferable to alizarin S in that it is more sensitive and the reaction is immediate. These workers also adopted the chloroacetic acid buffer at about pH 3.0 in an aqueous system.The Jluoride -$uosiZicate eq&Zibrium-It has generally been supposed that the pH in the titration must be controlled in order to obtain a sharp end-point with the indicator used and to restrict the ionisation of thorium fluoride. The titration of fluorine with thorium nitrate is generally represented by- In acid solutions the thorium fluoride is only slightly ionised, but as the acidity decreases the amount of fluoride ion in solution increases, presumably owing to the precipitation of basic thorium salts or hydrous thorium oxide. There is, however, yet another factor which may be influenced by the pH. During the separation of fluorine from interfering ions, the silicon tetrafluoride that is distilled off reqcts with steam in the condenser to produce a distillate containing fluosilicic acid with some precipitated silica.When a buffer is added before titration, the fluosilicate ions will be hydrolysed in accordance with the equation- It follows that the proportion of fluorine present as fluoride ion will be greater a t higher pH values. If it is assumed that the fiuosilicate ion does not form a slightly ionised compound with thorium, it becomes necessary to find an optimal pH range to balance two opposing factors ; with increasing pH values the amount of thorium fluoride present would decrease owing to the removal of thorium ions by precipitation, whilst the concentration of fluoride ions would increase through more complete hydrolysis of the fluosilicate. The problem therefore reduces to one of finding a convenient balance between the equilibria of equations (1) and (3), both reactions occurring in the same system. If this reasoning is correct, it would be expected that, when working in solutions of about pH 3, the proportion of fluorine titrated in a distillate would decrease as the absolute amount of fluorine present increased; this would be so because the amount of silicon in the distillate necessarily follows the amount of fluorine, and increasing amounts of silicon will tend to increase the proportion of fluorine in the fluosilicate form under any defined equilibrium conditions.This was confirmed by titrating the fluorine distilled from sodium fluoride and sulphuric acid; it was then found that the percentage of fluorine apparently recovered was lower with .... .. * - (1) 4F + T h +ThF, .. .. .. 3SiF, + 2H,O + 4H’ + BSiF,” + SiO, . . .. .. - - (2) SiF,” + 40H’ + H,SiO, + 6F’ + 2H,O . . .. . . * - (3)86 RICKSON: AN IMPROVED MICRO-METHOD FOR THE [Vol. 75 5mg. fluorine present than with 2mg. (Fig. 1). In both experiments the distillation was continued until titration of aliquots of successive portions of distillate showed that no more fluorine was distilling over. The titrations were carried out in aqueous systems buffered at pH 2.8 with half-neutralised chloroacetic acid, that is, under conditions similar to those recommended by Reynolds and Hill.g Had any fluorine been retained in the distilling flask its amount might have been expected to be independent of the amount of fluorine originally in the flask.On the other hand, any discrepancy which arose in the titration might be expected to vary with the amount of fluorine present, especially if any complex fluorine ions were involved. Further evidence for the adverse effect of the formation of fluosilicate on the thorium titration of fluorine was afforded by an attempt to determine the end-point of the titration by a conductimetric method. As would be expected from equation (l), a marked break in the conductance curve is observed when titrating sodium fluoride with thorium nitrate. 6 6 0 - 8 B 8 4 0 - 1 2 2 0 “1 2-5 ,\v z 8 .H 0 1 1 I I 1 2-0 10 I J l.2 1.3 oe 0 9 5 0 I 0 0 I50 2 0 0 2 5 0 M1. of distillate Fig. 1. Recovery of fluorine from the distillation of sodium fluoride. Titrated with thorium nitrate, using chloroacetate buffer and alizarin S indicator.0 = 2mg. of fluorine, x = 5mg. of fluorine. M1. of 0.025 M Th(NO,), solution Fig. 2. The variation in the thorium nitrate titre of 500 pg. of fluorine with the pH value, using alizarin S indicator. However, with the distillate from an equal amount of fluorine, a curve of completely different shape was obtained, suggesting that there was no close similarity between the ionic com- positions of the two solutions. This dissimilarity was too great to be explained by the fact that one contained sodium ions and the other hydrogen ions in equivalent quantities. Similar results were obtained from conductimetric titrations using silver nitrate instead of thorium nitrate; as silver fluoride is a co-ordinated compound it would be expected to behave in a similar way to thorium fluoride.The apparent deterioration on storage of sodium fluoride solutions used for standardising thorium nitrate solutions, which has been reported by several workers, can also be explained by this view of the fluoride - fluosilicate - thorium system. Ryss and Bakina12 showed that solutions containing sodium fluoride, sodium fluosilicate and silica show a definite change in pH value with time. The direction in which the pH changes depends on the initial con- centration of sodium fluoride, and in all but the highest concentration used by these workers the time taken to reach equilibrium was more than 24 days. These results indicate that the system reaches equilibrium slowly, so that solutions of sodium fluoride stored in unprotected glass vessels will show a slow increase in fluosilicate content and a consequent decrease in thorium nitrate titer.In view of this evidence, information was sought on the conditions of equilibrium of the system represented by equation (3) above. A number of papers have been published on the decomposition of the fluosilicate ion in aqueous systems.12 913914915 Unfortunately the data given in these papers were all obtained with fluosilicic acid solutions partly neutralised with 6NaF + SiO, + 2H,O = Na,SiF, + 4NaOH * . * * (4)Feb., 19501 DETERMINATION OF FLUORINE 87 sodium hydroxide, so that the equilibrium constants calculated from such data are applicable only to systems of a much lower hydrogen ion concentration than those with which we are concerned.However, Malaprade16 has shown, from the neutralisation curve of fluosilicic acid, that the neutralisation takes place in two stages: .. .. * . (5) Na$iF6 + 4NaOH = 6NaF + SiO, + 2H,O .. .. .- (6) H2iiF6 + 2NaOH = N+SiF6 + 2H20 .. The curve also shows that SiF," occurs in appreciable proportions only in solutions with pH below 3. If it is desired to isolate quantitatively either fluorine or silicon from a fluosilicate the pH must be greater than this value. From the evidence given it follows that disturbance of the titration by fluosilicate will be minimised by working at as high a pH as possible, and that the use of hydrochloric acid or of a chloroacetic acid buffer gives pH values that are definitely too low. The necessary conditions for the thorium nitrate titration of jluorine-Before adopting a buffer solution which would give a higher pH value than those previously used, it was necessary to investigate the behaviour of the alizarin S indicator in solutions of varying acidity.This 300 250 20c 8 .r( # JSC bil w w P ma SO 0 M1. of 0.005 M Th(NO,), solution Fig. 3. The fluorine equivalent of 0-005 M thorium nitrate solution. X = a t pH 2.8 in aqueous solution, 0 = at pH 5.3 in aqueous solution, = at pH 5.3 in 50 per cent. alcohol solution. was done by titrating 0.5-mg. quantities of fluorine with thorium nitrate, the pH being adjusted by the addition of either hydrochloric acid or acetic acid. The fluorine was added as a solution of sodium fluoride, the volume of solution being kept constant at 20 ml.The results obtained are plotted in Fig. 2. It will be seen that below pH 2.8 the points are scattered, indicating the difficulty of recognising the end-point. Above pH 3.5 the indicator becomes very sensitive to pH changes, as it behaves as an acid - alkali indicator over the pH range 4 to 5. Hence it appears that the useful pH range of this indicator in the thorium nitrate titration is 2.8 to 3.5 approximately; Matuszak and Brown9 give the optimum pH as 3.0 to 3.5. It was therefore necessary t o find an indicator for the titration reaction which would be suitable at higher pH values. The only other indicator which appears to have88 RICKSON : AN IMPROVED MICRO-METHOD FOR THE [Vol. 75 been used for this titration is Solochrome Brilliant Blue B.S.ll This is not now commercially available, but two other dyestuffs of similar type were obtained, Solochrome Azurine B150 and Solochrome Cyanine R200.Of these the former appeared promising, showing a colour change from pink to blue-violet with an excess of thorium. Unfortunately the coiour change was too slow for this substance to be useful as an indicator. Solochrome Cyanine R200 gave a similar colour change, but only in solutions more alkaline than pH 6; after trying a large number of other substances which might behave as adsorption indicators, it was found that gallocyanine behaves as an adsorption indicator above pH 4, the colour change in 0.1 M acetic acid - sodium acetate buffer at pH 4.12 being only just discernible, whilst in 0.01 M acetate buffer (pH 4.10) the colour change is quite good.It was also observed that the speed of the colour change increased as the pH is increased, so the buffer solution finally adopted for use in the titration was 0.1 M acetate in which the molar ratio of sodium acetate to acetic acid was 2 : 1. In the titration system the buffer is diluted to 0.01 M , giving pH 5.3 in aqueous solution. (The pH values throughout this work were determined with a valve potentiometer, using a glass electrode and a saturated calomel half-cell.) One consequence of titrating at this high pH in aqueous solution is that the degree of dissociation of the thorium fluoride is quite high, so that as the amount of fluorine present increases, the volume of titrant required for a given increment of fluorine decreases.This results in a non-linear standardisation curve, the sensitivity of the titration decreasing as the amount of fluorine increases. The dissociation of thorium fluoride can be decreased by carrying out the titration in a mixture of water and ethyl alcohol; calibration curves for thorium nitrate in aqueous and in 50 per cent. alcoholic systems are shown in Fig. 3. The effect of pH on the dissociation of thorium fluoride is also illustrated here by a comparison of the calibration curves at pH 2.8 and pH 5.3 in aqueous solution. METHOD REAGENTS- Thorium nitrate, 0.005 M-2.76 g. of Th(NOs)+.4H,0 in 1 litre. Bufer - indicator solution-3-4 ml. of acetic acid (1 M ) + 6.6 ml. sodium acetate (1 M ) , diluted to 100 ml. with a 0.01 per cent. aqueous solution of gallocyanine.This gives a solution which is 0.1 M with respect to acetate and approximately 0.01 per cent. with respect to the indicator. The mixed solution deteriorates on keeping, and should be not more than 3 days old when used. EthyZ alcohol-98 per cent. Sodium juoride, for standardisation-To prepare the standard solution, dissolve 0-221 1 g. of sodium fluoride in 500 ml. of water. Ten ml. of this stock solution are diluted to 100 ml. when required; the diluted solution contains 0,020 mg. of fluorine per ml. (It is not possible to obtain “Analytical Reagent” grade sodium fluoride, the best material that is easily obtain- able being “Laboratory Reagent” grade. It is advisable to recrystallise this from water, although little difference in titer has been noticed when this has been done.Through the kindness of the Government Chemist’s Laboratory, a sample of pure sodium fluoride, used as a standard, was obtained, and it was found that the recrystallised material was as good as this sample.) PROCEDURE- Pipette a 5-ml. aliquot of the solution containing fluorine (which may be a distillate from a fluorine separation) into a 25-ml. beaker or specimen tube, and add 5 ml. of alcohol. Add 1 ml. of the buffer - indicator solution. Prepare a blank in a similar way, using 5 ml. of water, 5 ml. of alcohol and 1 ml. of buffer - indicator solution. Using the thorium nitrate solution from a micro-burette (2-ml. capacity graduated in 0.01 ml.), titrate the blank drop- wise until a marked change of colour occurs from the original blue to a reddish-purple shade.This should require only 1 or 2 drops (approx. 0.05 ml.) of thorium nitrate. Titrate the fluorine solution, adding the titrant dropwise until the colour of the solution matches the colour of the titrated blank. Deduct the blank titer from that of the fluorine solution to obtain the volume of thorium nitrate equivalent to the fluorine present. With each batch of determinations, carry out titrations with known amounts of fluorine as sodium fluoride, and prepare a calibration curve, which should be linear. From this curve the amount of fluorine present in the unknown sample is obtained. This frequent standardisation is necessary as the strength of the thorium nitrate may decrease on standing; a 0.005 N solutionFeb., 19501 DETERMINATION O F FLUORINE 89 deteriorated to 95 per cent.of its initial strength in 6 days. In view of earlier opinions on the stability of sodium fluoride solutions, it is of interest that, on titrating with thorium nitrate by the method described above, it was found that the fluorine content of a standard solution of sodium fluoride stored in an unprotected glass bottle, had not changed over a period of 4 months. This is in marked contrast with the apparent deterioration in fluorine content met with when using the older methods of determination. It is not essential to carry out all titrations in 10 ml. of solution, and indeed it may be inconvenient. Provided that (a) the concentration of alcohol is 50 per cent. by volume in the solution before adding the buffer, (b) the amount of indicator - buffer solution is 1 ml.per 10 ml. of the 50 per cent. alcohol solution, and (c) the standardisation and blank titrations are carried out under similar conditions, then the absolute volume is not critical. The effect of titrating in a larger volume of solution is to decrease the thorium nitrate titer for a given amount of fluorine; for example, in 10 ml. of solution, 50 pg. of fluorine are equivalent to 0.68 ml. of titrant (0.005 M ) , and in 20 ml., are equivalent to 0.50 ml. It follows that it is necessary to standardise the thorium nitrate solution under the same conditions as the deter- mination, and that it is advisable to carry out the titration in as small a volume as possible. RESULTS- The amounts of fluorine determined on titrating aliquots of the distillates from known amounts of sodium fluoride are shown in Table I, which also shows the results obtained TABLE I RECOVERY OF FLUORINE FROM Fluorine added Fluorine in to distilling titrated flask aliquot PLg.tLg. 200 10 400 20 600 30 800 40 1000 50 1000 100 2000 200 3000 300 4000 400 SODIUM FLUORIDE Fluorine determined at pH 2.8 P g . 11 21 29 38 48 90 160 280 360 ON DISTILLATION Fluorine determined at pH 5.3 tcg. 10 21 32 40 50 100 205 285 385 on titrating similar aliquots with thorium nitrate using chloroacetic buffer and alizarin indicator. It is seen that the accuracy obtainable by the proposed method is better than that obtained when titrating under the conditions which have been used previously for this titration. The results also indicate that titration at pH 5-3 tends to give low results when more than 200 pg.fluorine are titrated. Under such conditions the end-point of the titration is not sharp owing to the formation of a precipitate, probably thorium fluoride. It is recommended, therefore, that the volume of the aliquot titrated should be such that the amount of fluorine present is less than 200pg., although up to 400pg. may be determined with an error of about 5 per cent. INTERFERING ANIONS- The effect of other anions on the determination has been considered. It is known that PO4"' and SO," react with thorium nitrate and so lead to erroneously high results in the fluorine determination. However, the separation of fluorine by distillation from sulphuric acid is effective in that no phosphate appears in the distillate, and if the apparatus contains an efficient spray trap no trouble arises from contamination of the distillate.by sulphuric acid spray. The presence of organic matter in the parent material can give rise to sulphurous acid in the distillate, which is also undesirable. The results of experiments with soil, carried out at this laboratory, have shown that the effect of organic matter can be eliminated by mixing the soil with an equal weight of calcium oxide or hydroxide and igniting at 550" to 600" C. for 1 hour. MacIntyrel7 has recently published results confirming the value of this treatment; as an alternative he suggests that the fluorine be first distilled off from sulphuric acid, and then the distillate redistilled from perchloric acid.It is considered that this procedure, involving two distillations, is tedious and unnecessary.90 RICKSON : AN IMPROVED MICRO-METHOD FOR THE [Vol. 75 Salts which give rise to volatile acids on heating with sulphuric acid may interfere in the determination; of such substances, soil may contain nitrates, nitrites, sulphides and chlorides. The first two of these are decomposed by the ignition treatment with lime. If sulphides are present they may be oxidised in the distilling flask by adding potassium permanganate.6 Chlorides, however, cannot be removed easily and completely, so the effect of hydrochloric acid in the distillate was determined by titrating a standard fluoride solution with thorium nitrate in the presence of varying amounts of sodium chloride (Table 11).TABLE I1 EFFECT OF CHLORIDE ION ON THE DETERMINATION OF FLUORINE Fluorine Chlorine F1 u orine added added determined 20 20 40 40 40 80 80 80 100 1000 10 100 1000 10 100 1000 20 20 41 40 40 78 78 78 It is seen that there is no adverse effect caused by chloride in any of the concentrations used; the maximum concentration of Cl' was 1 mg. per 10 ml., and the maximum value of the ,Cl'/F' ratio was 50/1. Another point which required investigation was the effect of altering the pH at which the titration was carried out. This was done by altering the sodium acetate - acetic acid ratio in the buffer solution, whilst maintaini,ig the concentration of acetate at 0.01 M in the titrated solution. The results are shown in Table 111, and it is seen that there is no significant difference in titer over the pH range 4.5 to 5.3.These pH values are not absolute, owing to the unreliable behaviour of the glass electrode in alcoholic solutions, but the range of 0.75 pH units is a good indication of the flexibility of this factor. TABLE I11 (50 pg. fluorine present) EFFECT OF VARIATION OF pH ON THORIUM NITRATE TITER OF FLUORINE Th(NO,),, 0.005 M, ml. PH 5-30 0.535 4-90 0-540 4.70 0-540 4.55 0.535 From this information it can be shown that it is not necessary to neutralise the fluosilicic acid with sodium hydroxide before adding the buffer solution, because addition of 0.5 mg. of fluorine as fluosilicic acid lowers the pH of the buffered system by only 0.3 pH unit, which is well inside the pH range given in Table 11.As noted previously this amount of fluorine is too large to be determined with accuracy, owing to the precipitation of thorium fluoride, so that in practice the amount of fluorine present will never cause a fall in pH of as much as 0.3 unit. If hydrochloric acid is present in quantities comparable with the maximum amount of chloride used in Table 11, the pH of the system will be reduced thereby to about 4.1. At this pH the colour of the indicator itself will show that the pH has been lowered appreciably. Conditions of this type might be met when working with saline soils, and for these it would be necessary to remove the bulk of the chloride from the material before distillation. With soil containing 0.02 per cent. of fluorine this would be necessary if the chlorine content were.greater than 1.0 per cent. Precision-At the end-point of the titration the addition of 0.01 ml. of 0.005 M thorium nitrate produces little recognisable change in the colour of the indicator so that this amount must be accepted as the precision of the method (equivalent to approx. 1 pg. of fluorine).Feb., 19501 DETERMINATION OF FLUORINE 91 Thus the accuracy required sets a lower limit to the amount of fluorine which can be deter- mined; for example, if as little as 20 pg. of fluorine is titrated, the error may be 5 per cent. The upper limit is set by the precipitation of thorium fluoride as mentioned earlier, which occurs at concentrations of 400 to 500 pg. of fluorine per 10 ml. of 50 per cent. alcohol. The precision at this end of the range is 0.2 per cent., a figure which is negligible when compared with the accuracy of the method. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. Willard, H. H., and Winter, 0. B., Ind. Eng. Chem., Anal. Ed., 1933, 5 , 7. Clifford, P. A., J . Assoc. Off. Agr. Chem., 1941, 24, 350. Reynolds, D. S., and Jacob, K. D., Ind. Eng. Chem., Anal. Ed., 1931, 3, 366. Reynolds, D. S., J. Assoc. Off. Agr. Chem., 1935, 18, 108. Reynolds, D. S., and Jacob, K. D., Ind. Eng. Chem., Anal. Ed., 1931, 3, 371. Reynolds, D. S., and Hill, W. L., Ibid., 1939, 11, 21. Hoskins, W. M., and Ferris, C . A., Ibid., 1936, 8, 6. Hammond, J. W., and MacIntyre. W. H., J . Assoc. Off. Agr. Chem., 1940, 23, 398. Matuszak, M. P., and Brown, D. R., Ind. Eng. Chem., Anal. Ed., 1945, 17, 100. Williams, H. A., Analyst, 1946, 71, 175. Milton, R. F., Liddell, H. F., and Chivers, J. E., Ibid., 1947, 72, 43. Ryss, J. G., and Bakina, N. P., Comptes Rendues de Z’Acad. Sciences U.R.S.S., 1936, 2, 21. Rees, A. G., and Hudleston, L. J., J . Chem. Soc., 1936, 1334. Kubelka, P., and Pristoupil, V., %. anorg. allgem. Chemie, 1931, 197, 391. Ryss. J, G., and Slutskaya, M. M., J . Phys. Chem. Malaprade, J., Annales de Chimie et de Physique, 1929, 11, 104. MacIntyre, W. H., and Palmer, G., J . Assoc. 08. Agr. Chem., 1948, 31, 419. . U.S.S.R., 1940, 14, 701. ROTHAMSTED EXPERIMENTAL STATION HARPENDEN, HERTS. J d y , 1949

ISSN:0003-2654    

DOI:10.1039/AN9507500084

出版商:RSC    

年代:1950

数据来源: RSC

摘要:

Feb., 19501 DETERMINATION OF FLUORINE 91 An Absorptiometric Method for the Determination of Magnesium BY J. G. HUNTER SYNoPsIs-The method is based on the formation of coloured complexes of magnesium hydroxide with certain dyes of the Thiazol Yellow class. A constant quantity of the dye, more than is necessary for the formation of the complex, is taken, and the quantity of magnesium present is ascertained from the excess of unused dye, which is extracted from the reaction mixture by means of n-butyl alcohol and read as a solution in that solvent, on a Spekker absorptiometer. A calibration curve relates quantities of magnesium directly with drum readings of the unused dye solution. The method was developed for the determination of magnesium in plant tissues and soil extracts, and can be adapted also to trichloroacetic acid filtrates of milk and blood serum.It determines from 0.02 to 0-20mg. of magnesium with an accuracy satisfactory for this type of determination. Permissible concentration limits of certain interfering ions and means for controlling their interference are given where necessary. Data on the concentrations of magnesium, iron, calcium, manganese and aluminium in extracts of a number of soils, prepared with (a) N ammonium acetate solution, (b) Morgan’s solution and (c) 0.5 N acetic acid, are tabulated. INTRODUCTION-Methods previously published for the determination of magnesium on a semi-micro scale by means of the red-coloured co-ordination complex of magnesium hydroxide and a dye such as Thiazol Yellow, have been based on direct absorptiometric measurement of the complex itself.In the method now proposed a constant quantity of dye is used and the excess over and above that required for forming the complex is separated and determined absorptiometrically in true solution. The method described can be used for the determination of magnesium in plant tissue, soil, milk and blood. The use of the magnesium hydroxide - dye complex for the determination of magnesium is subject to certain fundamental difficulties. When the pH of a solution containing magnesium ions and an appropriate dye is adjusted to about 12.5, the complex is formed very rapidly and is not consistent in physical type and colour intensity. The complex in the solution is not constant in degree of dispersion and may be anything from a suspensoid92 HUNTER: AN ABSORPTIOMETRIC METHOD FOR THE [Vol.75 to a sol. Errors may therefore arise from inconsistency in colour intensity and from difficulty in measuring it accurately. Errors may also be introduced by the presence of other substances (for example, calcium, iron and manganese ions) which may affect the complex or form precipitates in the alkaline medium. In numerous published methods1~2~3~4~5~6~7~8~9~10~11 attempts to overcome these difficulties are usually made by incorporating reagents to stabilise the colloids or to prevent or standardise interferences. Peech and Englishg. in a paper on soil tests, discuss these measures and describe a method that they consider to be a significant advance, although not so satisfactory as the methods they use for the determination of other ions.In the method now to be described the sources of error are substantially controlled. The complex is formed slowly and is more consistent physically, a non-colloidal solution is used for the colour intensity measurement, and interference from other ions is materially reduced. The slow rate of complex formation is obtained by means of a tartrate. The true solution for colour intensity measurement is obtained by producing an alcoholic solution of that part of the dye that has not reacted with the magnesium hydroxide; it is the upper layer of the two-phase system which forms after shaking an organic solvent with the mixture containing the complex and the excess of dye. Prevention of interference by other ions is attained by introducing certain reagents, which are discussed more fully in the section on interference. The dyes that can be used in this method are fundamentally diazoamino compounds of derivatives of dehydrothio-9-toluidine, and among the names used rather indeterminately for them are Titan Yellow, Clayton Yellow, Thiazol Yellow, Tourmarine and Mimosa.There are considerable differences in sensitivity among the dyes and although samples may have the same name and even be from the same source, their sensitivities may not be identical. This difference in sensitivity has been emphasised by Mikkelsen and Toth,12 who show that the most satisfactory dye is Thiazol Yellow (General Aniline Works, Rensselaer, New York). In the present investigation a sample of this dye was compared in equivalent quantities with samples of Titan Yellow (British Drug Houses) and Chlorazol Yellow (Imperial Chemical Industries) and its superiority for the present purpose confirmed.The method is suitable for determining between 0.02 and 0.20mg. of magnesium, and over this range the degree of accuracy is satisfactory. The method is relatively rapid and simple, although not in the “quick test” category. A hundred determinations, including the measurement of aliquots, can be completed in 18 days. METHOD PREPARATION OF THE SAMPLE- The aim of any preliminary treatment of a sample to be used for the determination of magnesium by this method is to produce a solution which does not contain excessive amounts of organic matter or interfering ions and in which the magnesium concentration is such that 0.02 to 0.20 mg.of magnesium can be obtained in easily handled aliquots. In addition, the magnesium extractant should be easily removable so that it can be replaced by Morgan’s reagent4; if Morgan’s reagent is itself the extractant, the required aliquot should be not more than 15hl. Some satisfactory methods of preparing solutions for analysis are outlined below. Plants-The usual technique of ashing followed by solution in dilute hydrochloric acid is suitable for plant tissues; thereafter, a convenient aliquot is transferred to the reaction flask, evaporated to dryness and the residue dissolved in exactly 15 ml. of Morgan’s reagent. Plant tissues extracted with Morgan’s reagent are usually used without treatment other than removal of excessive amounts of organic matter by activated carbon ( H e ~ t e r ~ ~ ) .It is essential that the carbon used shall contain no extractable magnesium; ordinary grades can be purified by acid washing. Soils-Solutions prepared by extracting the soil with Morgan’s reagent as described by Peech and English9 are used as indicated above. If the extract has been prepared by using neutral N ammonium acetate (PipeP) or 0.5 N acetic acid (Williams19 and Williams and Stewart26) it is evaporated to dryness and the residue is dissolved in a convenient volume of dilute hydrochloric acid, to give a solution from which the aliquot is taken. Alternatively, a convenient aliquot of the original extract is evaporated to dryness in the reaction flask and the residue dissolved in 15 ml.of Morgan’sFeb., 19501 DETERMINATION OF MAGNESIUM 93 reagent. In either technique, excessive amounts of organic matter can be removed by repeated addition of 20-volume hydrogen peroxide as the evaporation nears completion. Milk and blood-Suitable solutions can be made from these by an ashing procedure similar to that for plants. Magnesium is then determined as described in the section on procedure. These solutions can be prepared more easily by means of trichloroacetic acid (for example, Sanders26). In analysing milk, a convenient aliquot (usually 10 ml.) is diluted to 100 ml. with tri- chloroacetic acid reagent (200 g. of trichloroacetic acid dissolved in water and diluted to 1 litre). The mixture is shaken occasionally during 30 minutes and then filtered through a dry No.30 Whatman filter-paper. Ten ml. of the filtrate are pipetted into the reaction flask, 5 ml. of Morgan’s reagent are added and the magnesium is determined as in the section on procedure-with a slight alteration in the method of calibration, which is also described. A similar method can be used for blood. For this, however, the sample should be diluted with an equal volume of the trichloroacetic acid solution, and the resulting precipitate removed by centrifuging rather than by filtering. REAGENTS- Morgan’s reagen+lOOg. of hydrated sodium acetate and 30ml. of glacial acetic acid are dissolved in water and diluted to 1 litre. Oxalic acid reagent-15 g. of hydrated oxalic acid are dissolved in water and diluted to 1 litre.Tartrate reagent-log. of sodium hydrogen tartrate, log. of mannitol and 2.5g. of hydrazine sulphate are dissolved separately in water, and the mixed solutions diluted to 1 litre. Sodium hydroxide reagent-150 g. of sodium hydroxide are dissolved in water and diluted to 1 litre. Dye reagent-The concentration of dye required will depend on the particular specimen of dye used and should be chosen to give a convenient range of Spekker readings. The following may prove satisfactory: 0.09g. of Thiazol Yellow (or 0.15g. of Titan Yellow or 0.25 g. of Chlorazol Yellow) dissolved in water, diluted to 1 litre, mixed with 2 drops of sodium hydroxide reagent, and filtered after 24 hours. Dye solvent-600 ml. of butan-1-01 (pure technical n-butyl alcohol) mixed with 400 ml.of ethanol (rectified spirits). Magnesium stock reagent (400 p.p.m. magnesium)--4.054 g. of A.R. magnesium sulphate (MgS0,.7H20) are dissolved in Morgan’s reagent and diluted accurately to 1 litre with Morgan’s reagent. Magnesium standard solution (20 p.p.m. magnesium)--50 ml. of magnesium stock reagent are accurately diluted to 1 litre with Morgan’s reagent. Calcium comfiensating reagent (200 p.p.m. calcium)--0.50 g. of A.R. calcium carbonate is dissolved in Morgan’s reagent and diluted to 1 litre with Morgan’s reagent. PROCEDURE- An aliquot containing from 0.02 to 0.20mg. of magnesium is put in the reaction flask (100-ml. conical) and treated as already described in the section on Preparation of the Sample, so that ultimately the magnesium is present dissolved in 15 ml.of. Morgan’s reagent-except where a trichloroacetic acid filtrate from milk or blood is concerned. Five ml. of the oxalic acid reagent are added to the solution, which is mixed and left for at least 1 hour at room temperature. Five ml. of the tartrate reagent are then added followed in succession by exactly 5 ml. of the dye reagent and 20 ml. of the sodium hydroxide reagent. The contents of the flask are mixed after each addition and the final mixture is left overnight. Fifty ml. of the dye solvent are then added and shaken with the mixture for about 30 seconds. Within 30 minutes as much of the upper layer as can be conveniently decanted off is transferred to a small conical flask containing 0-5 ml. of acetone. The colour intensity of this solution is determined within 30 minutes; a Spekker photo- electric absorptiometer with 4-cm.cells and glass violet filters can be used. The graph required for the conversion of the Spekker readings is constructed by diluting 0, 1, 2, . . . 10ml. of magnesium standard solution to 10ml. with Morgan’s reagent and The reagent is stable for several months. After standing for 1 minute two phases will have separated. The contents of this flask are then mixed.94 HUNTER : AN ABSORPTIOMETRIC METHOD FOR THE [Vol. 75 adding 5 ml. of the calcium compensating reagent. and the determination completed as above. Typical calibration readings are given below. Yellow, on a Spekker absorptiometer, with 4-cm. cells and water setting of 1.00. The oxalic acid reagent is then added They were obtained at 16" C.with Thiazol Magnesium, mg. nil 0.02 0.04 0.06 0.08 0.10 0.12 0.14 0-16 0-18 0.20 Spekker reading 0-070 0.150 0.220 0.295 0.380 0.465 0655 0.635 0.705 0.770 0.810 NOTES ON THE METHOD- If it is desired to obtain results more quickly, the procedure can be accelerated by heating the solution to boiling after the addition of the oxalic acid reagent, then cooling, and allowing the colour to develop for 1 hour; any such changes must, of course, be incorporated also in the calibration procedure. It should be noted that this modified method effects no saving in the time required for the manipulations, although the results are obtained sooner; moreover, results obtained in this way tend to be less accurate. The sodium hydroxide reagent must be added immediately after the dye reagent and in such a way that it washes all the dye from the sides of the flask into the solution.It is important that the 5ml. of dye reagent should be measured accurately. The contents of all the reaction flasks during the colour development period (especially in the initial stage) and the dye extraction period, should be maintained at the same tempera- ture, which must lie between 10" and 35" C. The temperature of the sodium hydroxide reagent should also be within that range. A constant temperature can be maintained thermostatically, but it is simpler to work at room temperature (provided it is within the required range) and to prepare a calibration graph with each set. While the mixture is being shaken with the dye solvent, the flask should be closed by a glass or rubber stopper.The phases which separate consist of a clear orange-coloured alcoholic upper layer containing the excess of dye, and a turbid aqueous lower layer containing the magnesium hydroxide - dye complex and other insoluble substances. The upper layer has a volume of about 50 ml. and it is an easy matter to pour off approximately the same amount each time; although slight differences in the decanted volume do occur, the dilution error introduced in adding the 0.5ml. of acetone is small and in any case is within the limits of accuracy. Acetone is added to prevent the alcoholic solution becoming turbid; it is ineffective if incor- porated with the mixture before separation. Once the solution is poured into the Spekker cell the reading should be made without delay. Any adjustment necessary to keep the process within time limits may suitably be made by dividing sets into groups of not more than 15 immediately before extracting the excess of dye, and completing the determinations on one group before proceeding with the next.Although oxalic acid reagent effects control of calcium interference within satisfactory limits, there is a tendency for solutions that contain no calcium to give rather high results; for this reason some calcium is incorporated in the standards used to construct the calibration curves. When trichloroacetic acid is present in the solution, approximately the same amount should be included in the calibration solutions; this can be easily arranged by incorporating the appropriate amount in the calcium compensating reagent.INTERFERENCES- The effect of certain ions on the method was investigated and it was found that some of these influenced the results when present in amounts outside the tolerance limits given in TABLE I Permissible amounts (milligrams) of ions in the aliquot taken for analysis ION TOLERANCE LIMITS Magnesium . . .. Calcium .. .. . . 0-4.0 Iron . . .. . . . . 0-0.6 Manganese . . . . . . 0-0.6 Aluminium . . . . . . 0-0.6 Titanium . . . . . . 0-2.0 Ammonium . . . . . . 0-700* Potassium . . . . . . 0-20*0* . . 0-02-0.20 Sodium . . .. .. . . 0-20.0" SiIicatet .. . . . . 0-2.0 Carbonate . . . . . . 0-10.0* Chloride . . . . . . 0-1300" Nitrate . . . . . . . . 0-10*0* . . 0-10.0* Phosphate . . . . . . 0-10*0* Sulphate .. . . * The effect of greater amounts was not investigated. t Expressed as silica.Feb., 19501 DETERMINATION OF MAGNESIUM 95 Table I. From Table I1 it is seen that variable magnesium values result from excessive amounts of calcium or iron in the aliquot and high values from excessive manganese. Low values were obtained when the amounts of aluminium, titanium and silicate exceeded the limits. TABLE I1 ION INTERFERENCE Effect of calcium, iron and manganese on the determination of magnesium (Amounts as milligrams per determination) Magnesium present 0.10 0.10 0.10 0.10 0.18 0.18 0.18 0.10 0.10 0.10 0.10 0.10 0.10 Calcium added 2.00 4-00 6.00 4.00 6.00 10.00 1.00 1.00 1.00 1.00 1.00 1.00 - Iron Manganese added added - - 0.60 - 1.00 - 2.00 - - 0-60 - 1.00 - 2.00 Magnesium Percentage found error 0.10 0 0.10 0 0.099 - 1 0-08 - 20 0- 184 + 2 0.206 + 14 0.10 - 44 0-10 0 0.094 - 6 0.08 - 20 0.10 0 0.113 + 13 0.133 + 33 Control within the established limits is obtained for calcium by oxalic acid, for manganese by a tartrate, and for iron and aluminium by a reagent containing a tartrate, hydrazine sulphate and mannitol.The tolerance limits are not often exceeded in the solutions as usually prepared for analysis, but excessive amounts of calcium, iron, manganese and aluminium do sometimes occur. The possibility of interference is discussed below. Few of the several thousand plant analyses enumerated by Beeson13 show excessive calcium and even fewer show excessive iron or manganese. The aluminium values given for ordinary plants by Winter and Bird1* and Hutchinson15 indicate that excessive amounts are not present.However, plants grown on some very acid soils have very high manganese contents (Hale and Heintzel* and Hunter and McGregorl’), and a few plant species are recognised aluminium accumulators (H~tchinsonl~) ; interference may be found in these. In general then, calcium interference is seldom encountered in analysing plant tissue and interference from iron, manganese or aluminium is exceptional. In soil analysis, the concentration of interfering ions in an extract will depend not only on their content and form in the sample, but also on the method used to prepare the extract. To obtain some idea of the possible occurrence of interfering substances in extracts, twelve soils of various types, mostly very acid, were extracted with (a) neutral N ammonium acetate solution (Piperls), (b) Morgan’s reagent (see section on reagents, p.93) pH 4-8 (as described by Peech and Englishg but extraction period increased t o 2 hours), and (c) 0-5 N acetic acid, pH 2.5 (Williamslg). The concentrations o€ calcium, iron, manganese and aluminium in the extracts were determined as described by Peech and Engli~h,~ and magnesium by the method described in the section on procedure, interiering ions being removed where necessary. The results are given in Table 111. It will be seen that calcium was extracted in excessive amounts from two of the soils by ammonium acetate and Morgan’s reagent, and from three by acetic acid; this extraction of excessive calcium from some soils by all three solutions was not unexpected and if more calcium-rich soils had been examined a higher proportion of extracts containing excessive calcium would have been obtained.Ammonium acetate extracted little iron, manganese and aluminium, and although Morgan’s reagent removed more, the extracts did not contain excessive amounts; it is possible, however, that large amounts of iron, manganese and aluminium, cspecially the last, will be extracted from some soils, and if the magnesium content of these soils is low and the maximum permissible aliquot for Morgan’s reagent (15 ml.) is taken, interference may occur. The acetic acid extracts contained more iron, manganese and aluminium than the other extracts, interfering amounts of iron being present in two, manganese in one, and aluminium in seven; this is characteristic of the more acid extractants.Nature of soil A r - Parent material Raised beach Fluvio-glacial Till (quartzose Till (acid igneous) sand and gravel schist) Till (Old Red Sandstone) Till (basic igneous) Till (quartzose schist) Till (serpentine) Till (basic igneous) Till (Old Red Sandstone) Basaltic Raised beach Texture Sand Loamy sand Loamy sand Coarse sandy Loam loam Loam Organic Organic Organic Clay loamy sand loamy sand sandy loam Organic Loam sandy loam TABLE I11 ION INTERFERENCE Magnesium, calcium, iron, manganese and aluminium in soil extracts (Concentrations expressed as parts per million of air-dry soil) Per- centage loss on pH ignition 4.9 4 4.7 7 4.8 9 4-6 8 4.3 8 4.9 7 4 4 18 4.6 54 4.8 21 5.5 9 5.3 26 7.0 9 Magnesium - (4 (b) (4 21 19 22 30 39 41 44 44 49 41 43 45 13 12 14 69 63 66 26 24 23 281 225 306 125 125 115 340 310 335 220 200 229 105 140 150 Calcium - 7 (4 (b) (4 (4 160 153 180 5 230 187 300 5 45 33 130 15 600 550 800 5 800* 900* 1400* 5 1000 750 1500* 5 130 75 185 5 1500 1500 2500 175 650 650 900 25 1850 1800 1850 5 400 400 450 5 3400* 5000* 7500* 5 (a) Extracted by N ammonium acetate.(b) Extracted by Morgan's reagent. (c) Extracted by 0-5 N acetic acid. * Likely to interfere with determination of magnesium. Iron + (4 (4 30 30 30 30 45 150* 15 35 10 10 5 25 30 45 300 400 150 500* 15 45 5 90 5 25 Manganese A (4 (b) (4 0 0 5 12 25 50 0 0 0 5 25 30 0 5 25 10 20 45 25 35 120* 75 250 500 13 25 45 0 5 25 15 15 80 7 10 25 Aluminium A x C (4 (b) (4 Z 0 50 ZOO* 3 10 75 500* * ?? 0 60 75 z * b v 0 25 75" 4 E; Ei 5 150 750" z 5 150 750* e E +I x 0 U 10 75 750 5 75 1000* w 0 w 0 20 25 2 M 40 250 1500* 5 10 100 n c sFeb., 19501 DETERMINATION OF MAGNESIUM 97 In general then, it can be taken that with ammonium acetate and Morgan’s reagent interference may frequently arise from excessive extraction of calcium, but seldom of other ions.On the other hand, excessive calcium, iron, manganese and aluminium occur commonly in acetic acid extracts (and in solutions from alkaline fusion and from relatively concentrated acid extraction). The composition of normal cow’s milk (Bushill and Rooke20) is such that interference from calcium , iron, manganese and aluminium does not arise. However, calcium interference may occur in the analysis of milk deficient in magnesium.The calcium in normal human whole blood and serum, and cattle serum (Evans21 and Sjollema22), does not interfere in the determination of magnesium, but when the magnesium content of blood is low, as it is in cattle affected by grass tetany, calcium may be in excess. The iron content of normal human whole blood, plasma and corpuscles is given by Evans21; in plasma (or serum) the content is small and does not interfere, but in whole blood or corpuscles the content is very large and interference occurs. The amounts of manganese and aluminium in blood serum (Kehoe, Cholak and Story23) are small and unlikely to cause interference. When an excessive amount of calcium, iron, manganese or aluminium does occur in a solution in which magnesium is to be determined, a removal procedure such as described TABLE IV DEGREE OF ACCURACY OF THE METHOD Statistical analysis of replicate determinations Milligrams of magnesium per Number of Standard Standard Percentage determination determinations deviation error error 0.0200 10 0.00292 0.000923 15.0 0-0200 10 0.00266 0.000841 13.7 0*0300 10 0.00220 0.000696 7-54 0.0300 10 0.00162 0*0005 12 5.55 0.0406 10 0*00196 0.000620 4.96 0-0404 10 0.00 148 0.000468 3-76 0.100 10 0.00238 0.000752 2.45 0.100 10 0-00196 0.000620 3.00 0.160 10 0.00262 0.000820 1.67 0.160 10 0.00288 0*000911 1-85 0.200 10 0.00322 0.00 102 1.66 0,200 10 0.00220 0.000696 1.13 * Percentage ewor A .Percentage error B. A* This will not be exceeded 99 times out of 100.This will not be exceeded 19 times out of 20. Percentage error B* 10-4 9.52 5-24 3-86 3.45 2.62 1-70 1-39 1-16 1.29 1-15 0.78 immediately below must be introduced. The presence of excessive calcium is readily detected in the injtial stage of a magnesium determination by the excessively large precipitate produced on adding the oxalic acid reagent; excessive iron is detected by the yellow-brown colour of the original solution, and excessive manganese by the production of an orange-red instead of a red complex. Removal of excessive calcium, iron , manganese and aluminium-When excessive amounts of calcium are indicated by the size of the precipitate on adding oxalic acid, a further 15 ml. of Morgan’s reagent and 5 ml. of the oxalic acid reagent are added and after about an hour the mixture is filtered through a dry No.40 Whatman filter-paper. Twenty ml. of the filtrate, which will be slightly turbid, are then transferred to a reaction flask and the colour is developed in the usual way. The magnesium value found must be adjusted by the dilution factor of 2. Excessive iron, manganese and aluminium can be simply removed from solution in Morgan’s reagent or dilute hydrochloric acid as described below, but when in other extractants, such as ammonium acetate solution or acetic acid, it is usually better to evaporate the solution to dryness and dissolve the residue in dilute hydrochloric acid. When the extractant is Morgan’s reagent, the iron, manganese and aluminium are precipitated in presence of about 10 ml. of 20-volume hydrogen peroxide per 100 ml.of solution by adding sodium hydroxide solution dropwise until the liquid is alkaline (litmus paper). This is boiled for 5 minutes, cooled and filtered. The precipitate, etc., is washed with water and the filtrate and washings are evaporated until the volume is less than the original. Five ml. of glacial acetic acid are added per 100 ml. of the original solution and the whole is diluted with water to the original volume. When the solvent is dilute hydrochloric acid, the procedure is similar except that98 HUNTER: AN ABSORPTIOMETRIC METHOD FOR THE [Vol. 75 the filtrate and washings are made up to any convenient volume with water; a suitable aliquot is then transferred to a reaction flask and evaporated to dryness, the residue being dissolved in 16 ml.of Morgan’s reagent. It is unlikely that the residue will contain ammonium salts in interfering amounts, but where necessary these may be removed by gentle ignition. ACCURACY- To determine the degree of accuracy of the method, solutions containing known amounts of magnesium were prepared and their magnesium contents determined. The statistical analysis of these determinations is given in Table IV. It will be seen that the standard deviation and standard error for each set of determina- tions is relatively constant. In order to estimate on a percentage basis the degree of accuracy at different levels of magnesium concentration, the product of each standard error and the appropriate t value is expressed as a percentage of the mean determination. In this way the percentage accuracy that will be obtained 99 times out of 100, and the percentage accuracy TABLE V DETERMINATION OF MAGNESIUM IN PLANTS Material Tomato leaf-laminae Oat leaves Swede leaf -petioles Tomato stems Oat leaves Milligrams magnesium per 100 g.dry matter Detem+ned Determined A I v Method of preparing gravi- absorptio- Percentage solution for analysis metrically metrically difference Dry matter ashed 260 267 + 2.7 and extracted by 612 600 - 2.0 hydrochloric acid 212 216 + 1.9 Dry matter extracted 162 156 - 3.7 { by Morgan’s reagent } 426 424 - 0.5 obtained 19 times out of 20 were estimated and are recorded in Table IV. These show that for 0.04 to 0-20 mg. of magnesium the result will be within 5 per cent. of the true value 99 times out of 100, and for 0.03 to 0.04 mg.of magnesium, 19 times out of 20. The errors are relatively greater for smaller amounts of magnesium and the same level of accuracy is not reached. Thus, for 0.02 mg. of magnesium, 10 determinations will give a mean value within 10 per cent. of the true value in 19 cases out of 20. These results indicate that the degree of accuracy of the method can be considered satisfactory by the standards usually recognised for this type of determination. The magnesium contents of numerous samples of plant tissue, soil, milk and blood were determined by the method described. For some of the results the degree of accuracy was estimated by comparison with results obtained by established methods. The remainder were examined by repeating their analysis, using aliquots of different volume as the replicates; in such cases, satisfactory agreement between replicates would not be obtained if interfering ions were present (since no other ion examined affects the dye as does magnesium).Both methods of examination gave satisfactory results, some of which are shown in Tables V, VI and VII. TABLE VI DETERMINATION OF MAGNESIUM IN SOILS ( N ammonium acetate extraction) PH 8 6.7 6.1 5-7 4-1 Percentage loss on ignition 6-6 5.5 11.4 3-6 16.8 Milligrams of exchangeable magnesium per 100 g. soil Determined Determined absorptio- Percentage gravimetrically metrically difference 206 197 - 4.4 40.6 42.6 + 4.9 6-80 6.56 - 3.6 12.5 12.6 + 0.8 10.3 10-6 + 2.9 A I \Feb., 19501 DETERMINATION OF MAGNESIUM 99 TABLE VII DETERMINATION OF MAGNESIUM IN MILK AND BLOOD SERUM Solution prepared for analysis by trichloroacetic acid method Milligrams magnesium per 100 ml.r 3 Determined absorptiometrically Determined f L \ A Material gravimetricall y 5-ml. aliquot 10-ml. aliquot Milk 12.4 Milk - Milk - Blood (serum) 2.70 Blood (serum) - 12.5 10.9 10.8 2.72 2.12 12.4 10.8 11.2 2.76 2.08 The preliminary work on this method was carried out in the Chemistry Department of the West of Scotland Agricultural College. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. Kolthoff, I. M., Biochem. Z., 1927, 185, 344. Hirschfelder, A. D., and Serles, E. R., J . Biol. Cham., 1934, 104, 635. Spurway, C. H., Mich. Agr. Exfit. Sta. Tech, Bull., 1936, 132. Morgan, M. F., Conn. Agr. Expt. Sta. Bull., 1937, 392, 126. Peech, M., PYOC. Soil Sci. SOC. Amer., 1939, 4, 189. Gillam, W. S., Ind. Eng. Chem., Anal. Ed., 1941, 13, 499. Garman, W. H., and Merkle, F. G., Soil Sci., 1942, 53, 15. Stross, W., Analyst, 1942, 67, 317. Peech, M., and English, L., Soil Sci., 1944, 57, 167. Garner, R. J., Biochem. J., 1946, 40, 828. Pieters, H. A. J., Hanssen, W. J., and Geurts, J. J., Anal. Chim. Actu, 1948, 2, 241, Mikkelsen, D. S., and Toth, S. J., J . Amer. Soc. Agron., 1947, 39, 166. Beeson, K. C., U S . Dept. Agr. Misc. Publ., 1941, 369. Winter, 0. B., and Bird, 0. D., J . Amer. Chem. SOC., 1929, 51, 2964. Hutchinson, G. E., Soil Sci., 1945, 60, 29. Hale, J. B., and Heintze, S. G., Nature, 1946, 157, 554. Hunter, J. G., and McGregor, A. J., Scot. Agric., 1946, 26, 30. Piper, C. S., Soil and Plant Analysis, The University, Adelaide, 1944, p. 171. Williams, R., J , Agric. Sci., 1928, 18, 439. Bushill, J . H., and Rooke, H. S., Thorpe’s Dictionary of Applied Chemistry, 1947, 8, 98. Evans, C. L., Principles of Human Physiology, Churchill, London, 1946, p. 481. Sjollema, B., Nut. Abstr. & Rev., 1932, 1, 621. Kehoe, R. A,, Cholak, J.. and Story, R. V., J. Nut., 1940, 20, 85. Hester, J. B., J . Amer. Soc. Agron., 1940, 32, 549. Williams, E. G., and Stewart, A. B., J . SOC. Chem. Ind., 1941, 60, 291. Sanders, G. P., J. Biol. Chem., 1931, 90, 747. THE MACAULAY INSTITUTE FOR SOIL RESEARCH CRAIGIEBUCKLER, ABERDEEN First submitted, Mavch, 1949 Amended, Decembev, 1949

ISSN:0003-2654    

DOI:10.1039/AN9507500091

出版商:RSC    

年代:1950

数据来源: RSC

摘要:

100 SCOTT: THE EFFECT OF BICARBONATE ON THE [Vol. 75 The Effect of Bicarbonate on the Colorimetric Determination of Uranium BY T. R. SCOTT SYNOPSIS-when peruranate colours are to be developed in sodium carbonate solutions in the hydrogen peroxide method of determining uranium, the presence of bicarbonate in the solution causes errors. Their magnitude and means of avoiding them are considered. They may be eliminated by con- ducting the preliminary neutralisation of the solution with sodium hydroxide and measuring the peruranate colour within a few hours of making the solution alkaline with sodium carbonate. If readings are made at 445 mp., no error occurs even in presence of bicarbonate, but the sensitivity is much less than at 360 mp. The decrease in colour at this lower wavelength is attributed to the conversion of higher to lower peruranates, the latter increasing in amount as the pH of the solutions is decreased.THE intense yellow colour developed on addition of hydrogen peroxide to alkaline solutions containing uranium has been used for many years in the colorimetric estimation of that element. Precautions must be taken, however, when the colour is developed in the customary fashion1 in sodium carbonate solution, since considerable errors may be introduced if sodium bicarbonate is present. The magnitude and variability of the error is demonstrated below and has not been previously reported except for a preliminary note2 by the writer. The purpose of this paper is to amplify and supplement the original short note. EXPERIMENTAL FORMATION OF BICARBONATE IN THE ANALYTICAL PROCEDURE- Uranium is conveniently separated from most other elements by ether e~traction,~ the ether extract being allowed to evaporate in presence of water, leaving an aqueous solution of uranyl nitrate and nitric acid.When this solution is made alkaline with sodium carbonate, considerable quantities of sodium bicarbonate are formed, which cause a marked reduction in the intensity of the peruranate colour subsequently formed with hydrogen peroxide. The results of experiments with synthetic ether extracts (containing 30ml. of ethyl ether, 4mg. of U,O, and 2ml. of concentrated nitric acid) are given in Table I. These extracts were poured into different volumes of sodium carbonate solution, the ether was evaporated and the resulting aqueous solutions were made up to 50ml.The peruranate colours were developed by addition of 0.1 ml. of 30 per cent. hydrogen peroxide per 10 ml. of solution and the extinction values determined at 360mp. with a Coleman spectrophoto- meter. In an attempt to eliminate bicarbonate, the solutions were boiled vigorously for 45 minutes, and the rise in pH shown in Table I is evidence that considerable amounts of carbon dioxide were lost, but the extinction values, although increased by this procedure, were still considerably lower than the value for solutions free from bicarbonate. TABLE I FORMATION OF BICARBONATE FROM ETHER EXTRACT Volume of Before boiling After boiling 10 per cent. Na,CO, r A > I A > solution used, PH E* PH E ml. 7.46 0.10 9.65 0.46 8-23 0.28 9.93 0.45 1st 20 30 9-09 0-35 9.97 0.41 * E = 0.554, in absence of bicarbonate.t 15.2 ml. required to neutralise the acid present, according to the equation 2HN0, + Na,CO, = ZNaNO, + H,O + CO,. Carbon dioxide is also absorbed from the air by solutions of sodium carbonate, so that solutions initially free from bicarbonate may give erroneous extinction values if several days are allowed to elapse (as over a week-end) before the colours are developed. In Table I1 details are given of the decrease in pH (measured with an “Alki” glass electrode) in solutionsFeb., 19501 COLORIMETRIC DETERMINATION OF URANIUM 101 of sodium carbonate, kept in beakers covered with watch glasses at room temperature. Each solution contained in addition 80 mg. of U,O, per litre, and the extinction values given in the last column show the effect of bicarbonate formation after 2 and 6 days.TABLE I1 EFFECT OF ABSORPTION OF CARBON DIOXIDE BY SODIUM CARBONATE SOLUTIONS PH *Extinction (E) 7 \ * After After After After After After Solution Initial 1 day 2days 3days 6days 2days 6days 9 percent. Na&O, . . 11450 11-38 11.32 11.20 10.93 0.652 0.553 2.5 99 97 . . 11.39 11.25 11.16 11.04 10.74 0.541 0-530 0.5 39 9 ) . . 11.24 10.89 10.67 10.48 10-08 0.503 0.458 * In absence of bicarbonate, E = 0.554. It is evident that no detectable error occurs with concentrated sodium carbonate solutions, even after 6 days, but that errors appear with relatively dilute solutions within 2 days. When the beakers are uncovered, uptake of carbon dioxide is much more rapid and, for 2-5 per cent.solutions, the extinction value is in error by 3 per cent. after only 7 hours’ exposure. It is thus advisable to eliminate this source of error either by determining extinction values within a few hours of making the solutions alkaline, or by keeping the solutions in stoppered flasks. MAGNITUDE OF THE BICARBONATE ERROR- To solutions containing 6mg. of U,O, per 100ml. and known amounts of sodium carbonate and bicarbonate, 1 ml. of 30 per cent. hydrogen peroxide was added per 100 ml. I I I I I t 2-0$ Na, \ 0.5 1.0 1-51 2.0 2.5 Per cent NaHCO, Fig. 1. Effect of bicarbonate on extinction values and the extinction values were determined at 360mp. Fig. 1 shows that, with a fixed amount of bicarbonate, the effect is more pronounced as the concentration of sodium carbonate becomes less.On the other hand, the decrease in extinction becomes larger for a given carbonatefiicarbonate ratio as the carbonate concentration increases. This non-linearity of the relationship between extinction and bicarbonate concentration leads to considerable difficulty in correcting results once bicarbonate has been introduced into the solutions. Most of the extinction values shown in Fig. 1 fall approximately on a straight line when plotted against pH, except at pH values below 9.8, where the decrease in extinction only varies slightly with increased bicarbonate concentration. It must be stressed, however,102 SCOTT: THE EFFECT OF BICARBONATE ON THE [Vol. 75 that points do not necessarily fall on or near this line when colours are developed in media other than the carbonate - bicarbonate solutions under consideration.ELIMINATION OF THE BICARBONATE ERROR- According to Stumper"l the decomposition of sodium bicarbonate in aqueous solution is a first-order reaction and prolonged boiling is therefore needed to destroy all the bicarbonate. In presence of excess of sodium carbonate the decomposition is further retarded. This evidence, taken in conjunction with the findings shown in Table I, suggests that errors may still persist despite prolonged boiling of carbonate - bicarbonate solutions, and indicates that alternative methods of avoiding the error are to be preferred. The addition of sodium hydroxide to convert bicarbonate to carbonate is not recom- mended, for excess of the reagent will augment the colour intensity beyond that produced in pure Fig.2. Transmittance curves for sodium peruranate solutions (using Beckman spectrophotometer) As shown in Fig. 2, the bicarbonate effect becomes less marked a t longer wavelengths, and is non-existent a t 445 mp. No error will be caused if readings are made at this wave- length, but the extinction for a given concentration of uranium is barely one-third of that at 360mp. As the ratio of bicarbonate to carbonate decreases, the wavelength at which the two curves cross becomes slightly less, but the error introduced by using 445 mp. for all ratios should never exceed 2 per cent. The simplest method for completely eliminating the source of error is to neutralise solutions with sodium hydroxide, with addition of sodium carbonate at a later stage. In this way no bicarbonate is formed.Although more intense colours are produced when sodium hydroxide is used instead of sodium carbonate to make the solutions alkaline, the method has several disadvantages. Care must be taken to ensure that uptake of carbon dioxide from the air, or the presence of sodium carbonate as an impurity, does not lead to errors similar to those occurring with solutions made alkaline with sodium carbonate. Moreover, when alkaline solutions are coloured with compounds other than uranium salts, suitable blanks cannot be produced with sodium hydroxide, since sodium uranate is precipitated: with sodium carbonate, on the other hand, the uranium in the blank remains in solution as the double carbonate.DISCUSSION At first sight it may appear that the decrease in colour in the presence of sodium bicarbonate can be explained by the reversible reaction- 2Na&JO, + 8NaHC0, = 2Na4(UO2)(C0& + 2Na&O, + 4H202 . . - . (1)Feb., 19501 COLORIMETRIC DETERMINATION OF URANIUM 103 In the absence of sodium bicarbonate, the equilibrium in carbonate solutions with excess of hydrogen peroxide would be displaced to the left with formation of a peruranate (Na,UO, is only one of several possibilities) and the minor amounts of bicarbonate formed by the reaction with milligram quantities of uranium would have no significant effect. With gross amounts of bicarbonate, however, the equilibrium would be displaced to the right and the intense peruranate colour would be partly replaced by the relatively weak colour of sodium uranyl tricarbonate.This simple explanation can be, at best, only partly true, because inspection of Fig. 2 shows that the transmittance curve for bicarbonate solutions is quite different from that for carbonate solutions, whereas it should be similar (but with decreased absorption) if a diminu- tion in peruranate concentration has taken place. The change in the shape of the curve cannot be ascribed to sodium uranyl tricarbonate, which shows comparatively little absorption above 350 mp. at the concentrations involved. Hence it must be concluded that the colour- producing compound has changed in whole or part because of the presence of sodium bicarbonate. In the three peruranates, Na2U20,,, Na2U0, and Na,UO,, described by Rosenheim and Daehr,5 the ratio of active oxygen to uranium increases in the order given.Relationships between these compounds are shown in equations (2) and (3) and indicate that the formation of higher peruranates (it?., those containing a greater proportion of active oxygen) is favoured by increase in pH and hydrogen peroxide concentration. 2Na4U0, + 4H20 = 2Na2U0, +. 4NaOH + 2H202 . . .. .. * (2) 2NaJJ0, + 2H,O = Na&J20,, + 2NaOH + H,02 . . .. .. Reduction of pH by addition of sodium bicarbonate can thus be expected to favour of the formation of lower peruranates, as shown in equation (3), which may be rewritten, to compare with equation (l), in the form- 2Na2U0, + 2NaHC0, = Na2U20,0 + 2Na2C0, + H202 . . .. ' (4) In view of the high concentrations of hydrogen peroxide and sodium carbonate in the solutions studied, it is understandable that reduction in pH would not cause complete dis- placement of equilibrium to the right-hand side of equation (4) and the transmittance curves shown in Fig.2 more probably represent the effect of mixtures of two (or more) peruranates, with the lower peruranate preponderating in the curve for the solution containing sodium bicarbonate. If the lower peruranate has a smaller extinction coefficient than the peruranate that predominates in the absence of bicarbonate, the decrease in the extinction values in the presence of bicarbonate can be readily understood. An alterndive that cannot be over- looked is that the lower peruranate has a higher extinction coefficient, in which instance a low extinction value in the presence of bicarbonate must be ascribed to the formation of sodium uranyl carbonate (with a very small extinction coefficient at 360mp.) as well as a lower peruranate.Insufficient data are available to make it possible to choose between these alternatives. It should be emphasised that the equations given above are merely illustrative and the actual peruranates responsible for the colours may be different from those mentioned. In the course of work in hand, however, it has been found that the transmittance curve obtained in presence of bicarbonate can be duplicated in absence of bicarbonate by appropriate reduction of the concentration of hydrogen peroxide. This fact might have been predicted from a study of equation (4) andeonfirms the belief that a lower peruranate is formed under these conditions. CONCLUSIONS When peruranate colours are to be developed in sodium carbonate solutions, errors due to bicarbonate may be eliminated by conducting the preliminary neutralisation of the solution with sodium hydroxide and measuring the intensity of the peruranate colour within a few hours of making the solution alkaline. If readings are made at 446 mp., no error occurs even in presence of bicarbonate, but the sensitivity is much less than at 360mp.The decrease in colour at this lower wavelength is attributed to the conversion of higher to lower peruranates, the latter increasing in amount as the pH of the solutions is decreased.104 LINDSEY : THE QUANTITATIVE ELECTRO-DEPOSITION OF TIN [Vol. 75 The work described in this paper was carried out as part of the research programme of the Division of Industrial Chemistry, Council for Scientific and Industrial Research, Melbourne, Australia. REFERENCES 1. Sandell, E. B., Colorimetric Determination of Traces of Metals, Interscience Publishers Inc., New 2. Scott, T. R., Nature, 1949, 163, 768. 3. - , Analjst, 1949, 74, 486. 4. Stumper, R., Z. artovg. Chem., 1933, 210, 264. 6 . Rosenheim, A., and Daehr, H., Ibid., 1932, 208, 81. York, 1944. COUNCIL FOR SCIENTIFIC AND INDUSTRIAL RESEARCH MELBOURNE, AUSTRALIA March, 1949

ISSN:0003-2654    

DOI:10.1039/AN9507500100

出版商:RSC    

年代:1950

数据来源: RSC

摘要:

104 LINDSEY : THE QUANTITATIVE ELECTRO-DEPOSITION OF TIN [Vol. 75 The Quantitative Electro-deposition of Tin from Chloride Solutions Part 11. Stannous Tin Solutions BY A. J. LINDSEY SYNOPSIS-The electro-deposition of tin. from hydrochloric acid solutions has been described by several authors, but until Part I of this work was published, Analyst, 1940, 65, 498, no explanation had been given of occasional failure to recover all the tin from solution. I t was shown in Part I that losses previously noted by various authors were probably due to re-solution of the tin deposit during the processes of washing and disconnection. In the present part, the electrolysis of stannous and mixtures of stannous and stannic tin solutions are studied. Strict adherence to the technique of Sand is shown to be essential and by observing the precautions described, satisfactory results are obtained.In the range 0.1 to 0.3 g. the recoveries show an average error of f0.05 mg. THE advantages of using hydrochloric acid solutions for the rapid electrolytic analysis of metals have been set forth in previous communications from this laboratory1J,*s4 and in Part I of the present researche~.~ Prior to 1940 no adequate explanation had been given of the difficulties which various analysts had found in recovering residual quantities of tin in the electrolysis of chloride solutions1 p6 p7 s8 and the present investigation is a completion of work that was interrupted by the war. In Part I it was established that losses caused by volatilisation of stannic chloride, cathodic formation of stannane or mechanical loss of crystals of tin were unlikely to have accounted for the deficiences which had been noted.When using stannic solutions it was shown that during washing there is a considerable risk of losing a portion of a tin deposit by electrolytic action between the deposit and the platinum electrode. Strict adherence to the details of the method reduced the average error to 0.2 mg. with deposits weighing from 0.13 to 0-25 g. The deposition of tin from stannic solutions causes very little reduction to the stannous state, but if other metals are deposited first, partial reduction to the stannous state occurs. It is therefore important to test the method for stannous tin solutions and for mixtures of both valency states in solution.As in Part I of this work, depositions were made on copper-coated cathodes and on plain platinum cathodes. The solutions for analysis were prepared by dissolving “Chempur” tin in 10 to 12 ml. of hydrochloric acid with addition of 5 ml. of water. As depolariser, 2 g . of either hydrazine sulphate or hydroxylamine hydrochloride were then added and the solutions electrolysed at 40” to 45” C. and with the cathode at -0.6 to -0.75 volt with respect to the saturated calomel auxiliary electrode of Sand.@ The apparatus and electrodes of SandDJo were used in some of the experiments, and those of Fischerll in a commercially made electrolysis stand were used for others. The results are given in Table I. Further experiments were carried out as above, but immediately after dissolving the tin, measured quantities of a standard solution of potassium chlorate were added and theFeb., 19501 PART 11.STANKOUS TIN SOLUTIOXS 105 solution warmed to oxidise some of the tin to the stannic condition. The electrolyses were completed in the manner described above. The results are given in Table 11. SUMMARY- Good analytical results are obtained in the quantitative deposition of tin from either stannous or stannic chloride solutions or from mixtures containing the two valency states when potential control is exercised. Ammonium salts present in considerable amount (up to 2 g.) cause errors due to partial re-solution of deposited tin, but these errors are reduced TABLE I DETERMINATION OF STANNOUS TIN Tin taken, g . 0.1077 0.1115 0-1274 0.1441 0.1486 0.1613 0.1958 0.2212 Tin found, g.0.1076 0.1 114 0.1276 0.1444 0.1489 0.1618 0.1954 0.2209 Error, g* - 0.0001 - 0~0001 + 0*0002 + 0*0003 + 0-0003 + 0*0005 - 0*0004 - 0.0003 to a minimum if the electrolyte is neutralised at the end of a determination, and if the platinum electrode is previously coated with copper. The considerable losses reported by earlier workers were almost certainly due to re-solution of the deposit during washing. Rapid washing and drying and strict adherence to the technique of Sand are recommended. The increased reliability which is now attainable in the deposition of tin from hydrochloric acid solutions improves the methods previously described for the electro-analysis of bearing TABLE I1 DETERMINATION OF STANNOUS AND STANNIC TIN Tin taken, g* 0.0930 0.1419 0.1491 0-2 106 0.2114 0.3079 S tannic, Tin found, Error, g.g - g- 0.050 0-0933 + 0.0003 0.050 0.1413 - 0.0006 0.025 0.1489 - 0*0002 0.100 0.2 102 - 0.0004 0.1 00 0-2110 - 0.0004 0*100 0.3080 + 0~0001 metals, solders and bronzes. These two papers make it clear that equally accurate results can be obtained in the electro-deposition of tin from hydrochloric acid solutions by the method described in Part I, whether the tin is present in the stannic or stannous condition. The author wishes to acknowledge with thanks the loan of apparatus by Imperial Chemical Industries Ltd. He also wishes to thank his forrner co-workers, Messrs. F. G. Kny-Jones and A. C. Penney, for their co-operation and for agreeing to his wish to proceed with work which was brought to a halt in the autumn of 1939. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. REFERENCES Lindsey, A. J., and Sand, H. J . S., Analyst, 1934, 59, 335. Torrance, S., I b i d . , 1937, 62, 719. Lindsey, A. J., I b i d . , 1938, 63, 159. Kny-Jones, F. G., Ibid., 1939, 64, 172. Kny-Jones, F. G., Lindsey, A. J., and Penny, A. C., Ibid., 1940, 65, 498. Schoch, E. P., and Brown D. J., J . Amev. Chem. Soc., 1916, 38, 1660. Furman, N. H., I n d . Eng. Chem., Anal. Ed., 1923, 15, 1071. Lassieur, A., Compt. rend., 1924, 179, 827. Lindsey, A. J., and Sand, H. J. S., Analyst, 1934, 59, 328. Sand, H. J . S., I b i d . , 1029, 54, 275. Fischer, A., Z. Elektrochem., 1907, 13, 469. SIR JOHN CASS TECHNICAL INSTITUTE JEWRY STREET LONDON, E.C.3 May, 1949

ISSN:0003-2654    

DOI:10.1039/AN9507500104

出版商:RSC    

年代:1950

数据来源: RSC

摘要:

106 NOTES [Vol. 75 Notes A CONFIRMATORY TEST FOR THE PRESENCE OF DIPHENYLAMINE I N COMMERCIAL PETROL FOR some time we have sought a satisfactory confirmatory test for the presence of diphenylamine in commercial petrol. The following chromatographic test has been found helpful in distinguishing diphenylamine from other compounds giving a positive reaction in the official test. The method is based on the circular filter-paper technique described by l3utter.l PROCEUURS Carry out the official test as described in the Ministry of Fuel and Power Memorandum KP 1/8/7/3 to the final stage at which the diphenylamine is extracted with light petroleum spirit. Drain off as much of the aqueous phase as possible and evaporate the petroleum spirit (avoiding drops of water) in a 50-ml. flask to about 5ml.Divide the petroleum spirit extract that remains approximately equally between two small 7 x 1 cm. test tubes. To one tube add 0.1 ml. of 0.05 per cent. solution of diphenylamine in light petroleum. Evaporate the petroleum spirit in the tubes by placing them in a beaker of hot water until only 0.1 ml. of residue remains in each tube. Transfer the residues with the aid of micro-pipettes to the centres of two ll-cm. Whatman No. 2 filter-papers, controlling the rate of flow from the pipettes so that the central spots do not exceed 1.6cm. in diameter. In each of the filter-papers, make two parallel cuts 3 mm. apart, starting from the outer edge and extending to the centre of the petroleum deposit, thus forming a ‘‘tail” or tongue, which is then folded downwards at right angles to the paper.Place each of the filter papers between two small sheets of glass, one of which has a hole 4 mm. in diameter drilled in the centre, through which the filter paper “tail,” cut as described above, is placed. The glass sheets are then placed on top of a dish containing light petroleum, boiling range 40” to 60” C., into which the filter paper “tail” dips. Allow the chromatograph to develop until the circular solvent front has a diameter of about 8 cm. Remove the chromatograph from between the glass plates and cut out a suitable segment for testing. Dab the segment with a piece of cotton wool wetted with the official ammonium vanadate reagent in hydrochloric acid. When diphenylamine is present a bright blue ring is developed near the edge of the “solvent front” and is usually adjacent to a red colour ring formed by traces of dye remaining in the extracting solvent.With pure diphenylamine, no other rings or discs are obtained near the original “central spot’’ on the filter-paper. The chromatograph obtained with the added diphenylamine is the more important test. In this test only one blue band is obtained if the blue colour in the official test is due to diphenylamine. Other nuclear substituted compounds such as diphenyl benzidine and the amino compounds of diphenylamine yield blue rings of various shades of blue near the original central spot. If suitable sheets of glass are not available, the test may be conducted without them provided precautions are taken to avoid evaporation of the petroleum spirit during development of the chromatograph.The small central hole in the glass plate is easily made by the aid of a drill and emery powder. 1. Rutter, L., Analyst, 1950, 75, 37. REFERENCE * DR. B. DYER AND PARTNERS LONDON, E.C.3 J. HUBERT HAMENCE December, 1949 THE DETERMINATION OF PENICILLINASE ACTIVITY THE action of the enzyme penicillinase on penicillin is well known.1 I t converts it to a compound, possibly penicilloic acid, which has no antibiotic activity and has one extra acidic group for each molecule of penicillin originally present. Penicillinase activity is measured by the rate of decrease in penicillin consentration or else by the rate of production of the acidic group. McQuarrie and Liebmann2 used the former method, measuring the penicillin concentration by means of a biological assay.Henry and Housewright3 followed the production of the acidic group from penicillin by measuring the volume of carbon dioxide evolved from a mixture of penicillin, penicillinase and bicarbonate buffer in a Warburg apparatus. A simpler method of estimating penicillinase has been developed, based on the observation by Murtaugh and Levy“ that the total acid produced from the penicillin by the action of penicillinase can be titrated with The chief objection to these methods is that they are rather tedious to use.Feb., 19501 KOTES 107 sodium hydroxide. The rate of production of the acidic group is measured by titrating with sodium hydroxide at a constant pH. It has been found that there is an optimum pH value for the maximum rate of reaction; the titration is therefore carried out electrometrically so as to maintain the pH continuously at this value.The penicillinase activity is then measured by the rate of addition of alkali at this pH and at a definite temperature. Procedure-Use a preparation of penicillinase, diluted if necessary, such that the reaction takes at least 3 minutes. Place about 50ml. of a penicillin solution containing 200 to 300 units per ml. in a beaker kept in a thermostat at 25" k 0.2" C. Insert the electrodes of a pH meter into the contents of the beaker, above which is placed a burette to deliver 0.01 N sodium hydroxide solution. The contents of the beaker should be stirred; a stream of carbon dioxide-free air is convenient for this purpose.Adjust the pH of the penicillinase solution to 7-8 just before use. Time in minutes Fig. 1. Burette readings against time at two enzyme concentrations Adjust the pH of the penicillin solution to 7.8 and then add a known volume of the penicillinase solution, In order to prevent a lowering of pH, add sodium hydroxide continuously and at such a rate that the pH is constant at 7.8. Note a number of burette readings together with the corresponding times. The slope of a graph of burette readings plotted against time, in ml. of 0.01 N sodium hydroxide per min., is a measure of penicillinase activity. Since the rate of reaction is proportional to the total quantity of enzyme present in the solution, and independent of the penicillin concentration, it remains constant throughout, despite dilution by the titration.We have thus taken as a convenient unit of penicillinase activity the volume in ml. of 0-01 N sodium hydroxide per min. per 1 ml. of the original enzyme preparation. All values are taken at the optimum pH of the enzyme, in this case 7-8, and at a temperature of 25" C. Resulfs-Typical experimental results are given in Fig. 1, which shows clearly that the rate of reaction is constant with respect to time, i.e., independent of the penicillin concentration. Thus the quantity of penicillin used in the estimation is not critical. The dependence of rate of reaction on enzyme concentration is shown in Fig. 2. It will be seen that the rate of reaction is proportional to the enzyme concentration over a wide range.108 XOTES [Vol.75 The rate of reaction was found to depend on the purity of the penicillin used, very impure samples giving lower rates. In general, it was found that commercial penicillins, of potency greater than about 1200 units per mg. were satisfactory, especially when comparative results were needed, for example, in studying the up-grading of a penicillinase-containing broth. The purity of the penicillinase used did not affect the determination unless it was so highly buffered that the change in pH on formation of the acid group from penicillin was too small to be measured easily. In these circumstances, the penicillinase preparation was dialysed before the activity was determined. Concentration of enzyme, arbitrary units Fig. 2. Rate of reaction against enzyme concentration In general, the determined kinetics of the reaction confirm the results of Henry and House- wright.The optimal pH, however, was found to be 7-8, which differs somewhat from the American value of 7.2 (which, however, was for a temperature of 36" C.). These investigators found that at high enzyme concentratioiis the rate of reaction at 36" C. was independent of the enzyme concentra- tion, the limiting rate being 0-08 ml. of 0.01 N sodium hydroxide per min. Fig 2 shows that the rate of reaction varies directly with enzyme concentration up to a rate of a t least 8 ml. per min. at 23" C. The explanation may be that the enzymes have different characteristics, that the difference in temperature was large enough to affect the kinetics in this way, or that the Warburg apparatus gave anomalous results at fast sates of carbon dioxide evolution.Unfortunately, it has not been possible to investigate this point further. We are indebted to the Directors of the Distillers Company Limited for permission to publish this note. 1. 2. 3. 4. RESEARCH AND DEVELOPMENT DEPARTMENT G. H. TIYIGG March, 1949 REFERENCES Penicillin ; its Properties, Uses and Preparations, Pharmaceutical Society, London, 1041. McQuarrie, E. B., and Liebmann, A. J., Arch. Biochem., 1944, 5, 307. Henry, R. J., and Housewright, R. D., J . Biol. Chenz., 1947, 167, 559. Murtaugh, J . J., and Levy, G. B., J . Anier. Chein. SOC., 1945, 67, 1042. THE DISTILLERS COMPANY LIMITED w. s. WISE GREAT BURGH, EPSOM, SURREY DETECTION OF PHENYL BARBITURATES PESEZ~ describes a method for the detection of phenyl barbiturates based on the nitration of the benzene nucleus and the subsequent identification of the nitro-compound by its colour reaction with sodium hydroxide in acetone solution.He uses about 50mg. of the barbiturate for the test. The three phenyl barbiturates examined were- The method described below is sensitive to 0.5 mg. Luminal : 5-ethyl-5-phenyl barbituric acid. Rutonal : 5-methyl-5-phenyl barbituric acid. Prominal : l-methyl-5-phenyl-5-ethyl barbituric acid. Procedure-Heat about 1.0 mg. of the barbiturate in a test tube with 200 mg. of potassium Cool Transfer to a separating funnel and extract, first with nitrate and 20 drops of concentrated sulphuric acid in boiling water for about 20 minutes. and dilute with water to about 30ml.Feb., 19501 NOTES 109 30 ml.of ckloroform, and then with 30 ml. of ethyl ether. Wash each extract separately with a little water. Evaporate the solvents, dissolve each residue in 2 ml. of acetone and transfer to two test tubes. Add 1 drop of a 50 per cent. solution of sodium hydroxide to each and shake vigorously. Allow to stand and observe the colour of the acetone layers. The following colour changes were observed with 0.5 mg. of barbiturate- Chloroform extract Ether extract Luminal . . . . .. ... Purple light purple Rutonal . . .. .. .. purple - Prominal. . . . .. . . blue changing slowly - to a dark purple With luminal and rutonal the colour fades quickly. Ethyl ether should be washed with sodium hydroxide solution before use, as some samples of ether give a pinkish colour with aqueous sodium hydroxide.REFERENCE 1. Pesez. M., Analyst, 1938, 63, 353. GOVERNMENT ANALYST’S LABORATORY COLOMBO, CEYLON E. RATHENASINKAM May, 1949 APPARATUS FOR PRESSURE AND SUCTION REGULATION IT\; CHROMATOGRAPHIC COLUMNS IN the course of chromatographic adsorption analysis trouble is often experienced with columns having a low percolation rate. An extreme example is that of a long column of very fine powder, which may take some hours for complete development. Filter aids may be mixed with dense powders, but the extent to which such dilution can be tolerated is limited. By applying suction to the bottom of the column percolation may be hastened, but the maximum effective pressure available for driving the liquid through the column cannot exceed 1 atmosphere and is usually much less.Pressure can be applied to the liquid on the top of the column, but most types of apparatus so far described for achieving this are complicated. The simplest method is to lead a gas under pressure through a bung in the top of the chromatographic tube. This method is illustrated by Williams,’ by Gridgeman, Gibson and Savage2 and by others. The pressure that can be applied by this means is limited by the tenacity with which the bung can be held in the tube. Williams, Zoc. cit., describes an apparatus for overcoming this disadvantage, but each apparatus can be used only on tubes of one diameter and it takes several minutes every time to replenish the liquid. The chromatofuge of Hopf3 is at its best when used on a large scale.An apparatus for applying pressure has therefore been made and this can be brought into action in a few seconds, can be applied to ordinary adsorption tubes of different diameters and length, and is instantly disconnectable for replenishing the solvent. The apparatus is shown in use in the figure. The part of it above the laboratory bench is concerned with the application of gas pressure above the column. The part below the bench is concerned with the control of suction. For the pressure apparatus, a board A (shown both in elevation and in plan in the figure), 50cm. long by 28cm. thick by 8 cm. wide, is hinged at one end to a shelf, S , 65cm. above the bench. A bar of wood, B, of 24 cm. square section by 20 cm. long, is pivoted in a slot at the end of the board farthest from the hinge.This bar is constrained to remain vertical by a spring at its top end and a cord at its foot. The other end of the cord is tied to an inverted shelf bracket. The board, the bracket, the cord and the vertical bar form a rough parallelogram. When not in use the apparatus is allowed to swing down under the shelf; the bar is drawn inwards by the spring until it stops against the weight, W, and the cord slackens. The weight is a block of lead 15 cm. in diameter weighing 5 kg. The weight is held by a slanting nail and its force can be varied by moving it to a different nail. A rubber bung, R, 4 em. in diameter, is pushed up into a brass tube, which is so cut away that it can be screwed to the vertical bar. Through the bung passes a tube, T, which can be connected to a source of compressed gas.A pressure reducing valve attached to a cylinder of nitrogen is convenient but not essential. To bring the pressure apparatus into use it is swung up by grasping the handle H with one hand; the chromatographic tube is centred under the rubber bung and the nitrogen is turned on. The gas pressure that is applied may be regulated at the source. The maximum pressure on the liquid in the column is limited by the downward thrust of the apparatus. This is regulated by the mass and position of the lead weight. I f the The spring also adds a little to the thrust.110 NOTES [Vol. 75 El \ \ \ \ I PLAN - -- \ \ \ \ \ \ --- l r ' to suction pump \ H ! ! 1 : * I : i I . I ! * * i ! . . bench Apparatus for applying pressure over, and suction below, a chromatographic column.The suction apparatus, which is shown in the idle position, is independent of the pressure apparatus, which is shown in use (firm lines) and out of use (broken lines). The inset is a plan of the bung-holder, etc.Feb., 19501 OFFICIAL APPOINTMENTS 111 gas pressure exceeds the thrust pressure some gas escapes between the rubber bung and the top of the chromatographic tube. In this way the junction between the bung and the tube acts as a safety valve. Moreover, the pressure can be released at once by simply lifting the apparatus. The principal parts of the suction control apparatus below the bench have already been briefly mentioned (Booth4), but without a diagram. A T-join, J , is inserted in the vacuum pump line. The T opens to the air through a pressure rubber tube which passes under a lever, L, fitted beneath the bench. The lever is joined by a cord to a pedal, P, depression of which clamps the tube. Release of the pedal admits air and suction ceases at once. This arrangement permits immediate control of the suction and leaves both hands free. This enables the pedal to be drawn back and held under the bracket for prolonged suction, The pedal is released from the bracket by a kick. The pedal-controlled suction apparatus is used first and the pressure apparatus is brought into use if the percolation rate with suction is still too slow. REFERENCES Suction is thereby applied to the flask. The pivot securing the idle end of the pedal fits loosely. Either device may be used independently of the other. 1. 2. 3. 4. Williams, T. I., An Introduction to Chronzatogrufiky, Blackie, London, 1946. Gridgeman, N. T., Gibson, G. P., and Savage, J. P., Analyst, 1948, 73, 662. Hopf, P. P., Ind. Eng. Ckem., 1947, 39, 938. Booth, V. H., J . SOC. C h e w Ind., 1945, 64, 1621.. DUNN NUTRITIONAL LABORATORY UNIVERSITY OF CAMBRIDGE AND MEDICAL RESE.4RCH COUNCIL V. H. BOOTH May, 1949

ISSN:0003-2654    

DOI:10.1039/AN9507500106

出版商:RSC    

年代:1950

数据来源: RSC

摘要:

Feb., 19501 OFFICIAL APPOINTMENTS 111 Official Appoint men ts PUBLIC ANALYST APPOINTMENTS NOTIFICATION of the following appointments has been received from the Ministry of Food since the last record in The Analyst (1950, 75, 57). Public Analyst Appointments ALLEN, David George (Deputy) . . ALLEN, David George (Deputy) . . CHALMERS, Frederick Grant Ducan CREGEEN, William Allan (Deputy) . . DALLEY, Richard Arthur (Deputy) . . (Deputy) . . . . .. . . JAMES, George Vaughton . . . . JENKINS, Daniel Ceiriog . . .. JENKINS, Daniel Ceiriog . . .. SMITH, Arthur (Deputy) . . .. MINOR, Roland Gordon (Deputy) . . WATSON, Robert Sinclair (Additional) Metropolitan Borough of Bermondsey. County Borough of Southampton. County Borough of Wslsall. County of Somerset. County Borough of Leeds. County Borough of Bournemouth. County of Warwickshire. County Borough of Walsall. Metropolitan Borough of Islington. East Riding of Yorkshire. City of Glasgow. OFFICIAL AGRICULTURAL ANALYST APPOINTMENTS NOTIFICATION of the following appointments has been received from the Ministry of Agriculture and Fisheries since the last record in The Analyst (1949, 74, 604). Agricultural Analyst Appointments County Borough of Leeds. DALLEY, Richard Arthur (Deputy) . . JAMES, George Vaughton . . . . County Borough of Bournemouth. JONES, Daniel Evans . . .. . . County Borough of Merthyr Tydfil.

ISSN:0003-2654    

DOI:10.1039/AN9507500111

出版商:RSC    

年代:1950

数据来源: RSC

摘要:

112 MINISTRY OF FOOD [Vol. 75 Ministry of Food STATUTORY INSTRUMENTS* 1949.-No. 1816. The Food Standards (Curry Powder) Order, 1949. Price Id. This Order, which comes into force (a) as respects sales by the manufacturer of curry powder, on the 1st day of April, 1950, (b) as respects saEes by wholesale, on the 1st day of July, 1960, (c) as respects sales by retail on the 1st day of October, 1950, i s to be read with the Food Standards (General Provisions) Order (as amended). It prescribes the following standards for curry powder: (a) Curry powder shall contain not less than 85 per cent. spices, aromatic seeds and aromatic herbs. (b) No curry powder shall contain lead in excess of 10 parts of lead per million parts of curry powder. - No. 1817. The Food Standards (Tomato Ketchup) Order, 1949.Price Id. This Order, which comes into force (a) as respects sales by the manufacturer of tomato ketchup, catsup, sauce or relish, on the 1st day of April, 1950, (b) as respects sales by wholesale on the 1st day of July, 1950, (c) as respects sales by retail on the 1st day of October, 1950, and which should be read with the Food Standards (General Provisions) Order, 1944 (as amended), prescribes a standard for tomato ketchup, catsup, sauce and relish and Prescribes a maximum limit for traces of copper therein as foliows- (a) Tomato ketchup, catsup, sauce and relish shall contain not less than six per cent. by weight of tomato solids derived from clean and wholesome tomatoes or from tomato puree, or its equivalent, made from clean and wholesome tomatoes. The tomatoes, tomato puree or its equivalent or the tomato ketchup, catsup, sauce or relish shall be so strained, with or without heating, as to exclude seeds or other coarse or hard substances.Tomato ketchup, catsup, sauce and relish shall contain no fruit or vegetables other than tomatoes except onions, garlic and spices added for flavouring purposes. No tomato ketchup, catsup, sauce or relish shall contain copper in excess of 50 parts of Proceedings in England and Northern Ireland for a n infringement of Article 1 of the Food Standards (General Provisions) Order, 1944, as atnended by S.R. & O., 1944 (Nos. 42 and 654), II, p p . 505 and 508, in respect of tomato ketchup, catsup, sauce or relish ma-y be brought by a Food and Drugs Authority without the consent of the Minister.(b) (c) copper per million parts of the dried total solids. No. 1893. The Food Standards (Preserves) (Amendment) Order, 1949. Price Id. This Order, which came into force October 16th, 1949, amepids the Food Standards (Preserves) Order, 1944, as amended by S.R. & O., 1944 (No. 842). It prescribes standards for f r u i t curd and mincemeat that were previously controlled administratively under the Preserves Order, 1947. This i s eflected by substituting for paragraph 2 of Part I of tke Schedule to the Order of 1944 (as amended) the following paragraphs- “2. The standard for fruit curd (including fruit flavour curd) shall be as follows, t h a t is to say:- (i) Each 100 parts of fruit curd shall contain not less than- (a) 4 parts of fat, (b) (c) ( d ) 0.33 part of citric acid, 0.125 part of oil of lemon (or 0.25 part of oil of orange), 1 part of dried whole egg or 14 parts of sugar dried whole egg or 3% parts of liquid or frozen whole egg or 4+ parts of shell egg; (ii) The percentage of soluble solids contained in fruit curd shall be not less than 65 per cent. (i) Each 100 parts of mincemeat shall contain- (a) not more than 0.5 part of acetic acid (80 per cent. or glacial), (b) not less than 30 parts of added sugar, (c) not less than 30 parts of dried fruit and peel, (d) not less than 2.5 parts of suet or equivalent fat; (ii) The percentage of soluble solids contained in mincemeat shall be not less than 66 per cent.” 2 ~ . The standard for mincemeat shall be as follows, that is to say:- * Obtainable from H.M. Stationery Office. Italics indicate changed wording.

ISSN:0003-2654    

DOI:10.1039/AN9507500112

出版商:RSC    

年代:1950

数据来源: RSC

摘要:

Feb., 19501 REVIEWS 113 Reviews ISOTOPIC CARBON : TECHNIQUES IN ITS MEASUREMENT AND CHEMICAL MANIPULATION. By M. CALVIN, C. HEIDELBERGER, J. C . REID, B. M. TOLBERT and P. F. YANKWICH. Pp. xiii + 376. 1949. Price $6.60; 44s. The significance of carbon isotopes for tracer studies in chemistry and biology cannot be over- estimated. The usual definition of organic chemistry as the chemistry of carbon compounds indicates the unique importance of these isotopes. Ordinary stable carbon is a mixture of 98.9 per cent. of carbon-12 and 1.1 per cent. of carbon-13. The latter can be separated from carbon-12 by either chemical exchange or thermal diffusion and is available for experiments requiring a stable carbon isotope. The three other isotopes are radioactive. Carbon-I0 and -11 emit energetic positrons and have half-lives of 8-8 seconds and 20.5 minutes respectively.Carbon-14 is produced in the nuclear pile: it emits slow electrons and has a half- life of about 5000 years. Before carbon-14 was discovered, cyclotron-produced carbon-1 1 was used in tracer experiments, but it has now been largely superseded by the heavier isotope. The decay of carbon-10 is too rapid for tracer work. Practically all the early tracer work with carbon isotopes was carried out in the United States, but recently, thanks to the activities of the Atomic Energy Research Establishment at Harwell, supplies of carbon-14 have become available to British research institutes, and the subject of this excellent monograph by Professor Calvin and his colleagues has thus become of more than academic interest to British analysts.In this volume, both the methods available for the determination of carbon-1 1, -13 and -14 and the chemical maniplations needed for preparing carbon-labelled compounds, by direct synthesis or biosynthesis, are described in some detail, The volume is essential to all chemists using carbon-14 and is also of considerable value to chemists working with other isotopes emitting electrons of low energy. Many of the experimental procedures were developed in the authors’ laboratories at the University of California and are described in greater detail than in ordinary publications. Since carbon-14 is a very weak electron- emitter, particular care must be taken in preparing samples for radioactivity measurement and special detection devices must be employed.These techniques are equally suitable for other isotopes emitting electrons of low energy, such as sulphur-35. However, although synthese involving stable carbon- 13 are presented along with those involving radioactive carbon- 11 and -14, the measurement of carbon-13 is not described so fully as that of carbon-14, and anyone proposing to use the stable isotope would have to go elsewhere for detailed information. Some of the appendixes, such as those dealing with isotope dilution methods, statistical treatment of counting data and the determination of counter efficiency, are of particular interest to the analyst. Carbon- 14 is normally supplied as barium carbonate and consequently the greatest ingenuity has had to be exercised in using the methods of classical organic chemistry for introducing labelled carbon atoms into organic molecules ranging in complexity from acetone to testosterone.Labelled glucose and sucrose can be prepared by use of biosynthetic methods. All these operations must be carried out on a micro-scale. The authors believe that it will soon be possible to label almost any carbon compound. In this volume every isotopic carbon synthesis reported in the literature up to April, 1948, has been described in sufficient detail for the procedure to be used in the laboratory. The style is generally clear and easy to follow. The book is well-printed and bound but, as is now inevitable with American books, the price is high. This is a serious matter for monographs that discuss rapidly developing subjects and may therefore be nearly out-of-date before they are printed, J.E. PAGE New York: John Wiley & Sons Inc. London: Chapman & Hall Ltd. Five carbon isotopes with atomic weights ranging from 10 to 14 are known.114 REVIEWS [Vol. 75 CHEMISCHE SPEKTRALANALYSE. EINE ANLEITUNG ZUR ERLEKNUNG UND AUSFUHRUNG VON By W. SEITH and K. RUTHARDT. Seith and Ruthardt have produced a new edition of their earlier handbook of spectrochemical analysis, an essentially practical textbook for those working in laboratories of industrial and teaching organisations. As a work on s-pectrochemical analysis in Germany the volume is comprehensive, but the authors have admittedly confined the scope almost exclusively to the apparatus and technique employed there.They have avoided all but passing reference to developments elsewhere, both in instrumentation and applications ; the volume thereby loses almost entirely any wider appeal. The book is liberally illustrated with diagrams and photographs of apparatus, some of which are inadequately described in the text. I t seems pointless to devote valuable space to the description of apparatus that is obsolete in comparison with present-day standards. An early micro-photometer is described in some detail, followed by an account of an excellent up-to-date model. An adequate theoretical treatment of the subject is given together with a wealth of practical detail covering all branches of emission analysis. Some details of absorption spectra in the visible and ultra-violet regions of the spectrum are also included.This volume, though a useful handbook on the subject in its country of origin, could have been of much greater value in Britain had the authors not restricted their field to German products and methods. SPEKTRALANALYSEN IM CHEMISCHEN LABORATORIUM. Fourth Edition. Pp. vii + 173. Berlin: Springer-Verlag. 1949. Price DM.16.50. J. A. C. MCCLELLAND CHEMICAL ANALYSIS. v01. v, AQUAMETRY. APPLICATION OF THE KARL FISCHER REAGENT TO QUANTITATIVE ANALYSES INVOLVING WATER. By J. MITCHELL jun. and DONALD M. SMITH. Pp. xi + 444. New York and London: Interscience Publishers. 1948. Price $8.00; 48s. The authors apparently intend their newly coined word aqzmrnetry to signify the determination of water in any substance by any method.The word is no doubt etymologically correct, but is not particularly euphonious, and the reviewer, for one, is doubtful whether it will find wide acceptance. Having created a perfectly general term and used it as the title of their work, the authors proceed, after an introductory chapter on miscellaneous methods of water determination, to describe one particular method in minute detail. In other words, the sub-title more correctly describes the contents of the book than does the main title. It is now widely known that the Fischer technique involves the use of a reagent containing iodine and sulphur dioxide in pyridine solution. It is only some fifteen years since the method was introduced, yet it has been the subject of widespread study, and its popularity and outstanding value are evident from the large number of literature references quoted in the book.The authors (members of the duPont organisation) are well equipped for their task, for they themselves have contributed much to the application of the Fischer method in various directions. The book is in two main parts, the first dealing with the determination of water present as such in both organic and inorganic substances, and the second with an extension of the method to the determination of organic functional groups, in which the water is produced (or consumed) by a reaction between the sample and an appropriate reagent. Part I describes, with wealth of detail, the stoicheiometry of the reaction, the nature of the parasitic side reactions, the specifications of the components of the Fischer reagent, and the prepara- tion and standardisation of the reagent and of the standard water-in-methanol solution. There are next given detailed instructions for carrying out the titration, by both visual and by electrical (potentiometric and “dead-stop end-point’ ’) methods, followed by instructions for determining moisture in a very wide variety of substances, the organic ones, of course, predominating.It is clearly indicated that, in general, attention must be paid to (1) methods for getting the water in the sample into a form suitable for titrating, e.g., by dissolving in, or extracting with, anhydrous methanol, (2) avoiding undesired reactions in which iodine is absorbed by reacting with the substance under analysis. In this connection many ingenious devices have been used, e g ., the -CHO group may be rendered inert to the unmodified Fischer reagent by prior reaction with hydrocyanic acid to form the cyanohydrin. By such devices and, in some cases, modification of the composition of the Fischer reagent, water may be determined in a very large variety of organic substances, such as acids, alcohols, esters, acetals, acyl halides, amines, amides, carbonyl compounds, hydrazine derivatives, hydrocarbons, mercaptans and peroxides.Feb., 1950j REV IE W S 115 Part I1 of the book deals with the application of the Fischer method to the determination of functional groups in organic substances, i.e., to such determinations as aliphatii: alcohols (alone or in presence of phenols) , amines, carboxylic acids, acid anhydrides, carbonyl compounds, amides, nitriles, peroxides. As always with a new technique, there are doubtless many cases where its application may not possess any advantage over older alternative methods; there must be many others, however, where it will supply information which would not otherwise be readily obtainable.In most of the applications of the method the authors quote detailed results which have been obtained in establishing the method; these give a very impressive idea of both the accuracy and the extreme sensitivity of the method. The book has clearly been compiled with much care, and it will undoubtedly be found of great value in organic research and analytical laboratories. N. STRAFFORD LABORATORY AND WORKSHOP NOTES. Compiled and edited by RUTH LXNG, Ph.D., A.1nst.P.During the past 25 years the Laboratory and Workshop Notes published in the Journal of Scienti$c Instruments have provided many beautiful examples of the simple and ingenious solution of experimental problems. Feeling that the contents of these Notes should be more widely known, the Board of the Institute of Physics invited Dr. Lang to make a selection from those published in the Journal. This book, the royalties from the sale of which are to be placed to the credit of the Institute’s Benevolent Fund, is the outcome. In an undertaking of this kind, omission of subject-matter rather than its inclusion is the difficulty. In choosing the 181 Notes for this book (less than half of those selected for publication in the Journal), Dr. Lang must have had many moments of sore trial.The highly-assorted nature of the subjects included renders classification no easy matter and the grouping is necessarily somewhat arbitrary. The arrangement adopted is the division of the Notes into eight sections : laboratory and workshop tools, processes and devices ; clamps, supports and agitators ; soldering, brazing and welding ; technique of glass manipulation and silvering; vacuum and pressure techniques ; electrical and magnetic devices and techniques ; optical devices and techniques and devices for liquids and gases. A useful index supplements the full list of titles. There are numerous examples of the construction of special tools and of the solution of really difficult problems. The reviewer, who has never ceased to marvel at the never-ending uses of an ordinary T-piece, was particularly gratified to find, in addition, a host of items having simplicity as the keynote. Apart from its principal function as a source of ideas and information, the whole work makes fascinating reading. The galaxy of novel ideas is neatly prevented from causing mental indigestion by the inclusion of several longer Notes on more general topics such as cements, soldering, glass- working, etc. In dealing with such subjects, the older well-tried techniques are naturally introduced along with much that is new, thereby affording the reader just the right degree of relief. It appears from the foreword that a further volume of these Notes may be published. If the standard of the first volume can be maintained, this cannot take place too soon. Pp. xii + 272. London: Edward Arnold & Co. 1949. Price 21s. An enormous amount of information is packed between the covers of this book. J. T. STOCK

ISSN:0003-2654    

DOI:10.1039/AN950750113b

出版商:RSC    

年代:1950

数据来源: RSC

摘要:

REPORTS OF THE ANALYTICAL METHODS COMMITTEE OBTAINABLE THROUGH THE EDITOR The Reports of the Analytical Methods Committee listed below may be obtained direct from the Editor of THE ANALYST, 7-8, Idol Lane, London, E.C.3 (not through Trade Agents), a t the price of 1s. 6d. to Members of the Society, and 2s. Od. to non-Members. Remittances must accompany orders and be made payable to “Society of Public Analysts.” The Reichert-Polenske-Klrschner Process. (Test for Butter Fat.) To be reprinted. Milk Products Sub-committee : Reports Nos. 1 and 2. Report No. 3. Report No. 4. Analysis of Condensed Milks. Analysis of Sweetened Condensed Milk in which the Sucrose has altered during Storage. To be reprinted. Determination of Water, of Total Solids and of Fat in Dried Milk. Sub-committee on Dirt in Milk.Report on the Determination of Total Solids in Fresh Liquid Milk. Essential Oil Sub-committee : Report. Determination of Dirt in Milk. Report No. 1. Estimation of Cineole in Essential Oils. Report No. 2. Physical Constants (1). Report No. 3. Physical Constants (2). Report No. 4. Interim Report on the Determination of Acetylisable Constituents in Essential Oils. Report No. 5. Determination of Phenols in Essential Oils. Report No. 6. Determination of Citral in Lemon Oil. Report No. 7. Determination of Solubilities. Report No. 8. Determination of Cineole in Essential Oils. (2) Camphor Oil. (3) Other Oils. Report No. 9. Determination of Carvone and Menthone. Report No. 10. Determination of Citronellal. Report No. 11. Determination of Aldehydes other than Citronella].Report No. 12. Determination of Ascaridole. Report No. 13. Determination of Esters. Report No. 14. Solubility Test for Ceylon Citronella Oil. (Gratis.) (1) Cajaput and Eucalyptus Oils. Metallic Impurities in Foodstuffs Sub-Committee (formerly Sub-committee on the Determination of Arsenic, Lead, etc. in Food Colouring Materials): Report No. 1. Determination of Arsenic. Out of print. Report No. 2. Determination of Lead. Report No. 3. Determination of Copper. Report No. 4, Determination of Zinc. Sub-committee on the Determination of Unsaponifiable Matter in Oils and Fats and of Un- saponified Fat in Soaps : Report No. 1. Report No. 2. Report No. 3. Report No. 4. Report No. 6. Report No. 6. Determination of Unsaponifiable matter in Oils and Fats.Determination of Unsaponified Fat in Soap. Determination of Free Alkali in Soaps. Determination of Free Alkali and Silica in Silicated Soaps. Determination of Rosin in Soaps. Determination of Phenols in Soaps. Poisons Sub-committee appointed to investigate Methods of Assay for Various Substances appearing in the Poisons Schedules of the Poisons Regulations, 1935 : Report No, 1. Report No. 2. Report No. 3. Report No. 4. Report No. 5. Report No. 6. Assay of Lobelia (Lobelia Inflata). Assay of Gelsemium Assay of Aconite. Assay of Yohimba. Assay of Jaborandi. Assay of Ephedra and of Ephidrine in Nasal Sprays. Fluorine in Foods Sub-committee : Report on the Determination of Fluorine in Foods. Addendum to above Report. (Gratis.) Sub- Committee on Vitamin Estimations. Microbiological Panel : Report on the Microbiological Assay of Riboflavine and Nicotinic Acid. Tragacanth Sub-committee : Report No. 1. Evaluation of Powdered Tragacanth. Report No. 2. Evaluation of Flake Tragacanth.

ISSN:0003-2654    

DOI:10.1039/AN9507500116

出版商:RSC    

年代:1950

数据来源: RSC