ISSN:0003-2654    

DOI:10.1039/AN97500FX037

出版商:RSC    

年代:1975

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN97500BX039

出版商:RSC    

年代:1975

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN97500FP101

出版商:RSC    

年代:1975

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN97500BP107

出版商:RSC    

年代:1975

数据来源: RSC

摘要:

OCTOBER, 1975 The Analyst Vol. 100, No. 11 95 An Improved Procedure for the Determination of Thiamine E. E. Edwin, R. Jackrnan and Nancy Hebert Ministry of Agriculture, Fisheries and Food, Central Vetr rinary Laboratory, Weybridge, Surrey, KTl6 3NB A modified thiochrome procedure for the assay of thiamine is described. It eliminates the use of ion-exchange columns and elutions with hot, saturated potassium chloride solution that are employed in most other methods. The times required for extraction and dephosphorylation have been examined critically and shortened. It was shown by Jansenl that the oxidation of thiamine to thiochrome using akaline potassium hexacyanoferrate( 111) could be used for the determination of thiamine in natural materials and this reaction forms the basis of many assay procedures.In order to remove interfering substances the extracted thiamine was absorbed on an ion-exchange resin, the interfering substances removed by washing and the thiamine eluted with a hot, saturated solution of potassium chloride. In the procedure to be described, thiamine is converted into thiochrome while still on the resin, and the thiochrome then extracted into 2-methylpropan-1-01. The method is simple, reproducible and has been found to be satisfactory in this laboratory. Experimental Reagents All reagents should be of analytical-reagent grade. HydrochZoric acid, 0.1 N. Sodium acetate trihydrate. CZarase. This is available from Miles Laboratories Inc., Elkhart, Indiana, USA, and P.O. Box 37, Stoke Poges, Slough, Berkshire. Decalso F (also called Zerozit S / F ) .This resin is manufactured by Permutit Co. (Zerolit) and marketed by BDH Chemicals Ltd. and Hopkin and Williams Ltd. and is a sodium aluminosilicate cation-exchange resin of mesh size 60-85. It is prepared for use as follows. Spread 1 kg of the resin on a tray and run a strong magnet over it in order to remove any contaminating particles of iron or rust. Mix the resin well with a 25 per cent. solution of sodium chloride in dilute hydrochloric acid (2 ml of concentrated hydrochloric acid per litre). After settling, pour off the supernatant liquid together with much of the “fines.” Repeat this treatment three times. Next, wash the resin repeatedly with acidified water (1 ml of glacial acetic acid per litre of de-ionised water) until the washings are free from chloride.At the end of this treatment only the coarser particles will be left and these should settle to the bottom rapidly. Finally, dry the treated resin in air. NOTE- We have observed batch variations in the activity of Decalso F; although most batches were satis- factory, some gave high blank fluorimeter readings and did not retain thiamine quantitatively (see under Measurement of $uo1*escence). Cyanogen bromide solution. This solution must be freshly prepared before use. It is convenient to store the two reagents needed, viz., saturated bromine water and 10 per cent. aqueous potassium cyanide solution, in a refrigerator at 0.4 “C. The required amount of bromine water is poured into a flask and the 10 per cent. potassium cyanide solution added dropwise until 1 drop causes the yellow colour of free bromine to disappear completely. Any unused cyanogen bromide solution must be poured down the drain with plenty of running water.689690 EDWIN et al. : AN IMPROVED PROCEDURE Analyst, Vol. 100 Mercury(II) chloride solutiovt. Dissolve 1 g of the solid in water and dilute to 100 ml. Sodiwn hydroxide solution. Dissolve 30 g in water and dilute to 100 ml. Potassium hexacyanoferrate(III) solution. Dissolve 1 g in water and dilute to 100 ml. 2-Methylpropan-1-01. Repelcote. This is a 2 per cent. solution of dimethyldichlorosilane in carbon tetrachloride. A fluorescence-free grade must be used. Preparation of Standards As thiamine is deliquescent, it must be dried over phosphorus(V) oxide in a desiccator for several days. , Dissolve an accurately weighed amount (say 100 mg) of the dried material in 100 ml of 0.1 N hydrochloric acid and make further dilutions with the 0.1 N acid in order to obtain standard solutions of 0.1-1 pg ml-l.These solutions must be stored in dark bottles at 0 4 "C. As thiamine tends to become adsorbed on glass surfaces, it is advisable to treat these bottles with Repelcote. Standard solutions must be examined periodically for mould growth. Apparatus Fluorimeter. A Locarte fluorimeter was used. An LF2 filter was used to select the 365-nm excitation wavelength and either an LF14 or LF3 plus LF5, both with a wedge mono- chromator in the fluorescent beam, to select a waveband in the region of 436nm. General Procedure Extraction Dice animal tissues, such as liver, brain or muscle, food concentrates or plant material and weigh about 2 g accurately into 1-02 Universal or Macartney bottles, adding 5-10 ml of 0.1 N hydrochloric acid.Place the loosely capped bottles in a boiling water bath and heat for 15 min. This procedure denatures the protein material and, at the same time, releases any protein-bound thiamine phosphates. Liquid samples. Pipette liquid samples, such as blood (1 ml) or urine (5 ml), into 15 ml of 0.1 N hydrochloric acid in a Macartney bottle and heat as described for solid samples. Defihosphorylation Adjust the pH of the extracts to 4.5 by adding small amounts of solid sodium acetate and using pH indicator paper. Next add about 100 mg of Clarase and mix it in thoroughly. Incubate the mixture for 3 h at 45 "C or 16 h at ambient temperature with occasional shaking and centrifuge at 3000g for 10 min.Transfer the clear supernatant liquid to a calibrated flask (20- or 25-ml capacity) and re-extract the residue twice with 0.1 N hydrochloric acid. Then combine all of the extracts, add 1-2 drops of concentrated hydrochloric acid so that the pH is below 4, and dilute to the mark. Solid samples. Pzcri'catioa Place about 1.5g (about 2nd) of activated Decalso in a stoppered test-tube (capacity about 17 ml) and pipette a portion (normally 1.0 ml) of the extract directly on to the Decalso, avoiding the sides of the tube. Mix the materials by using a mechanical vibrator. With all batches include a set of standards (at least three) and a reagent blank and treat them in the same way as the extracts. Fill the tubes nearly to the top with boiling, de-ionised water and shake them.After allowing the contents to settle pour off the wash water, taking care not to lose any Decalso particles." Repeat the washing procedure three times. Formation of thiochrome Pipette 3 ml of cyanogen bromide or 0.3 ml of mercury(I1) chloride on to the wet Decalso and mix thoroughly. Then add 2 ml of 30 per cent. sodium hydroxide solution and shake the mixture vigorously for at least 30 s by use of a mechanical vibrator. * It has been suggested that the procedure for decantation could be improved by having fine gauzes made to fit over the mouth of the test-tube and that even fine muslin held over the mouth could allow more efficient and safer washing procedures.Alternatively, the supernatant liquid could be removed by suction through a Pasteur pipette attached to a water-pump.October, 1975 FOR THE DETERMINATION OF THIAMINE 691 Extraction into 2-methylpropan-1-01 To the mixtures in the tubes add 5.0 ml of 2-methylpropan-1-01, stopper the tubes and shake them vigorously for 2 min. After centrifugation at 3000 g for 5 min, transfer the clear 2-met hylpropan- 1-01 extract in to clean cuve t t es. Measurement of Jluorescence With all batches include a set of standards (at least three) and a reagent blank. Measure the fluorescence at 436nm, and adjust the instrument so that the highest standard gives a convenient reading (say 100). The readings for the other standards should be linear and values for the samples can then be read off from a standard graph. The blank reading when the highest thiamine standard is 1.0 pg ml-1 must be less than 8 divisions.If higher readings are obtained, it would be advisable to change the batch of Decalso. Critical Assessment of Method Time of Extraction About 2 g of comminuted liver were analysed to determine the thiamine content as des- cribed above. The extraction time was varied, replicate determinations being made at six intervals during the first hour. The extraction was found to be complete within the first 15 min. There was a highly significant rise in mean thiamine content from 0 to 10 min (P<O.Ol) and a corresponding fall from 30 to 60 min (see Table I). TABLE I EFFECT OF EXTRACTION TIME ON THIAMINE CONTENT Time of extraction/min .. .. .. , . 0 5 10 15 30 60 Mean thiamine content in liver (wet mass)/pg g-' 4.04 4-28 4-43 4.45 4.43 4.00 Number of determinations , . . . . . 4 3 4 4 4 4 Standard error of the mean of four determinations = 0.080. Optimum Amount of Clarase The effect of varying the amount of Clarase was studied using bovine liver (about 2 g). It was found (Table 11) that 50 mg of Clarase per gram of liver homogenate was the optimum amount. TABLE I1 EFFECT OF DIFFERENT AMOUNTS OF CLARASE Amount of Clarase per gram of homogenate/mg Amount of thiamine in homogenate (wet mass)/pg g-' 25 1.945 & 0-015 50 2.225 f 0-095 100 2.02 f 0.02 Optimum Amount of Decalso The effect of using different amounts of Decalso in the assay procedure was examined by taking different amounts of the resin and adding 1.0 ml of a solution containing 0-5 pg of thiamine.For each tube of Decalso plus thiamine a blank containing the same mass of Decalso, but no thiamine, was included. The remainder of the assay was as describedunder Purification, Formation of thiochrome, Extraction into 2-met~~y~propaiz-l-oI and Measurement of juorescence. The results (Table 111) show that the blank reading increased as the amount of Decalso was increased. However, this increase does not seriously interfere in the assay provided that the same amount of Decalso is used in all of the tubes.692 Analyst, Vol. 100 Limit of Thiamine Retention on Decalso To ten tubes, each containing 1.5 g of Decalso, different amounts of thiamine (from 0.1 to 1000 pug) were added and thoroughly mixed.The Decalso was washed with boiling water as described in the assay procedure and the amount of thiamine remaining was assayed. EDWIN et al. : AN IMPROVED PROCEDURE TABLE I11 EFFECT ON THE ASSAY OF THIAMINE OF INCREASING THE AMOUNT OF DECALSO Fluorimeter readings Mass of Decalsolg 'Blank With thiamine Corrected 0.75 6 69 63 1-5 7.5 60 62.6 3.0 14.6 70 65.5 For the higher amounts of thiamine the extracts in 2-methylpropan-1-01 were suitably diluted with more 2-methylpropan-1-01. The results (Table IV) show that with 1 mg of thiamine per 1.5g of Decalso the limit of retention had not been reached. TABLE IV LIMIT OF THIAMINE RETENTION ON DECALSO Thiamine added/pg 0 (blank) 0.1 0.26 0.5 0.75 1.0 250 600 760 1000 Fluorimeter reading 6.0 14.5 28.5 51.5 72.0 100.0 56.0 53.0 55.5 52.0 Dilution factor Nil Nil Nil Nil Nil x 500 x 1000 x 1500 x 2000 - Thiamine found/pg 0.1 0.25 0.5 0.73 1.03 - 275 615 810 1010 Precision of the Method The precision of the method was assessed by carrying out repeated assays on an acid extract of a sample of calf liver of low thiamine content.To another portion of the extract thiamine was added at the level of 10 pg g-l and the assay carried out in replicate. The mean values, together with the 95 per cent. confidence limits and standard deviations, are given in Table V. TABLE V PRECISION OF THE METHOD Liver low in thiamine content Extract plus added thiamine Number of assays .. .. 10 12 Mean thiamine contentlpg 6-l . . 0.305 10.22 95 per cent. confidence limits . . 0*161-0*449 10.0 1-1 0.44 Standard deviation .. .. & 0.201 f 0.344 The variation between replicates is greatly reduced with added thiamine. Over-all Recovery of Added Thiamine Approximately 2-g portions of samples of comminuted liver were assayed (in duplicate) with and without added thiamine. From the results the percentage recoveries were calculated (Table VI). Comparative Use of Mercury(I1) Chloride and Cyanogen Bromide for Thiochrome Formation Standard graphs were prepared by using five levels of thiamine ranging from 0-1 to 4 pg. In the assay procedure either 0.3 ml of a 1 per cent. solution of mercury(I1) chloride or 3 ml These recoveries lie within 10 per cent. of expected values.October, 1975 FOR THE DETERMINATION OF THIAMINE TABLE VI RECOVERY OF ADDED THIAMINE Thiaminelpg g-I Thiamine Amount of r added/ pg tissuelg Calculated Found Recovery, per cent.A -l 0 0 5 5 10 10 15 15 2.002 1 2.0042 2.0031 2,0009 2.0004 1.9929 2.0039 1.9973 4.146 4-341 Mean: 4.250 6.74 6.59 98 6-85 6.75 101 9-25 9.9 107 9.27 9-68 104 11.74 12-48 106 11-76 12.87 109 693 of cyanogen bromide solution were used for the oxidation stage. Fig. 1 shows the mean standard graphs evaluated for 14 assays using mercury(I1) chloride and 13 using cyanogen bromide. The regression equations are : Mercury(I1) chloride: y = 95-13 x + 0.53 (standard error of slope, & 1-36) Cyanogen bromide: 1-28) y = 94.63 x - 1.31 (standard error of slope, 100 LL 20 0 Amount of thiamine/pg Fig. 1. Mean standard graph of thio- chrome formation for 14 assays with mercury(I1) chloride (0) and 13 assays with cyanogen bromide (e) as oxidant.where x = thiamine (pg) andy = fluorimeter reading. The slopes did not differ significantly for the two methods. The results are similar with either reagent. Mercury(I1) chloride tends to give lower blanks. Comparative Use of Mercury( 11) Chloride and Cyanogen Bromide for Determination of Thiamine Thiamine was determined in liver, yeast and cattle concentrate nuts, using mercury(I1) chloride or cyanogen bromide as the oxidant. Slightly higher values were obtained with mercury(I1) chloride but Table VII shows that there is little difference between the results obtained by use of these two reagents. Assay of Thiamine in Urine : Comparison of Three Oxidising Agents A sample of bovine urine was acidified by the addition of 2 drops of concentrated hydro- chloric acid and held in a boiling water bath for a few minutes.Three sets of 1-, 2- and 3-ml aliquots were pipetted on to Decalso and washed as described before. The thiamine was then converted into thiochrome using cyanogen bromide or mercury(I1) chloride, but in one694 EDWIN et aE. : AN IMPROVED PROCEDURE Analyst, Vol. 100 TABLE VII ASSAY OF NATURAL MATERIAL USING MERCURY(II) CHLORIDE OR CYANOGEN BROMIDE Thiaminelpg g-1 r -l Material C,yanogen bromide Mercury( 11) chloride Liver . . .. .. 2-35 2.4 Feed concentrate . . 2.75 2-95 Yeast . . .. .. 17.2 18.9 set of tubes the oxidising agent was alkaline potassium hexacyanoferrate(III)2 [3 ml of an oxidising solution made by mixing 1 per cent. potassium hexacyanoferrate( 111) solution (1 ml) and 15 per cent.sodium hydroxide solution (24 ml)]. For each oxidising agent a standard graph was also prepared. The results (Table VIII) show that alkaline potassium hexacyanoferrate(II1) is not a suitable reagent for the measurement of thiamine in urine as the fluorescence of its product does not appear to be linear. TABLE VIII ASSAY OF THIAMINE IN URINE Thiamine found (pg) after oxidation with- - 1 alkaline cyanogen mercury( 11) Volume of urine/ml hexacyanoferrate( 111) bromide chloride 1 0.55 2 1.25 3 2.17 0.42 0.48 0.90 1-04 1.33 1.55 Discussion A major step in the chemical assay of thiamine, its purification by adsorption on the synthetic cation-exchange material Decalso, was introduced by Hennessy and Cere~edo.~ Following the removal of non-adsorbed impurities by washing with water, the thiamine was eluted with a hot, saturated solution of potassium chloride.However, this can be a tedious and time-consuming process, attended by many frustrations, such as the crystallisation of potassium chloride during elution and consequent clogging of the column. I t has also been suggested that incomplete elution may result if the potassium chloride solution is allowed to cool, and on occasions high blanks have been known to result.* These drawbacks have been overcome in the present procedure by eliminating elution with potassium chloride. The thiamine adsorbed on the Decalso is oxidised to thiochrome and extracted directly into 2-methylpropan-1-01, thus avoiding any losses. The formation of thiochrome by the oxidation of thiamine with cyanogen bromide was first demonstrated by F~jiwara,~ and later Fujiwara and Matsui6 applied it to the deter- mination of thiamine and found it to have distinct advantages over the use of alkaline potassium hexacyanoferrate( 111).The conversion into thiochrome was shown to be quanti- tative over a wide range of concentrations of thiamine. Reducing impurities had little or no influence on thiochrome formation. With alkaline potassium hexacyanoferrate( III), how- ever, such impurities can be a serious hindrance, even if the amount of hexacyanoferrate(II1) is carefully adjusted to compensate for the interference. On the other hand, excess of the hexacyanoferrate( 111) reagent can destroy thiochrome (Westenbrink and Goudsmit') and some workers employ a further step to destroy the excess of hexacyanoferrate(II1) with hydro- gen peroxide.8~~ Fujiwara and Matsui6 found that cyanogen bromide can safely be used for the determination of thiamine in urine as it does not produce flucrescence with N-methylnicotin- amide, whereas alkaline hexacyanoferrate( 111) would do so.As this metabolite is likely to occur in urine it could be a source of error. Waring et aZ.,1° who used the latter reagent, had to overcome this interference by separating thiamine and its phosphate esters in urine by thin-layer chromatography before assay. Morita et aZ.11 have used mercury(I1) chloride instead of cyanogen bromide for the con- version of thiamine into thiochrome. This is a useful improvement as it has several advantages; the limit of sensitivity is said to be about 0.02 pg and although both reagents are highlyOctober, 1975 FOR THE DETERMINATION OF THIAMINE 695 selective for thiamine, 1 per cent.mercury(I1) chloride solution is easier to handle than cyanogen bromide, which must be used only in a fume cupboard and is stable for only 3 h at room temperature.* The procedure described in this paper is simple to use, reasonably accurate and considerably reduces the time taken to carry out the assay by conventional rnethods.l2p13 The skilled technical assistance of Messrs. J. Carroll and K. James is acknowledged. 1. 2. 3. 4. 6. 6. 7. 8. 9. 10. 11. 12. 13. References Jansen, B. C . P., Red Trav. Chim. Pays-Bas Belg., 1936, 55, 1046. Strohecker, R., and Henning, H. M., “Vitamin Assay-Tested Methods,’’ Verlag Chemie GmbH, Hennessy, D. J., and Cerecedo, L. R., J . Am. Chem. SOL, 1939, 61, 179. Taku, T. J., and Michinaka, K., Vitamins, Kyoto, 1969, 39, 33. Fujiwara, M., J . J a p . Biochem. Soc., 1949, 21, 200. Fujiwara, M., and Matsui, K., Analyt. Chem., 1953, 25, 810. Westenbrink, H. G. K., and Goudsmit, J., Red Tvav. Chim. Pays-Bas Belg., 1937, 56, 803. Burch, H. B., Meth. Enzym., 1954, 3, 946. Rindi, G., and Perri, V., in Gyorgy, P., and Pearson, W. N., Editors, “The Vitamins: Chemistry, Physiology, Pathology, Methods,” Volume VII, Academic Press, New York, 1968, p. 67. Waring, P. P., Wayne, C. G., and Ziporin, Z. Z., Analyt. Biochem., 1968, 24, 185. Morita, M., Kanaya, T., and Minesita, T., J . Vztam., 1969, 15, 116. Association of Vitamin Chemists, “Methods of Vitamin Assay,” Wiley-Interscience, New York, 1951, p. 111. Horwitz, W., Editor, “Official Methods of Analysis of the Association of Official Agricultural Chemists, Tenth Edition, Association of Official Agricultural Chemists, Washington, D.C., 1965, p. 758. Received January 2 Id, 1976 Accepted June 9th, 1976 WeinheimlBergstrasse, 1966.

ISSN:0003-2654    

DOI:10.1039/AN9750000689

出版商:RSC    

年代:1975

数据来源: RSC

摘要:

696 Analyst, October, 19'15, Vol. 100, @. 696-702 A Simple Method for the Determination of 5,7- Diiodoquinolin-8-01 and 5-Chloro-7-iodoquinolin-8-ol (Clioquinol) Sobhi A. Soliman Department of Pharmaceutical AnalyticaZ Chemistry,CoZlege of Pharmacy, Uniusvsity of Alexandria, Alexandvia, Egypt A simple and accurate method is suggested for the determination of 5,7- diiodoquinolin-8-01 and 5-chloro-7-iodoquinolin-8-01 in a pure state as well as in various pharmaceutical preparations. The liberation of the iodine content of the compounds investigated is based on simple refluxing with 15 per cent. sodium hydroxide solution in the presence of zinc metal powder. Interference in the subsequent steps of the determination, due to the resulting organic products, is overcome by precipitation and filtration of these products prior to titration.The suggested method is applicable to the determination of the compounds under investigation in a convenient concentration range that is suitable for the determination of different dosage forms without numerous dilutions. Any unnecessary dilutions may result in high experimental error. The compounds 5,7-diiodoquinolin-8-01 and 5-chloro-7-iodoquinolin-8-ol (clioquinol) have been widely accepted a s anti-amoebic compounds, both in the treatment of amoebic dysentery and intestinal amoebiasis, and as topical anti-infectives, because of their low toxicity. Official compendia provide different assay procedures for the determination of the two compounds. The methods of assay given in USP XVIIIl and BP 19682 for the determination of 5,7- diiodoquinolin-8-01, both as the pure compound and in tablets, depend on the determination of the halogen content by the oxygen-flask combustion technique.However, while USP XVIII employs an infrared spectrophotometric assay for the determination of 5-chloro-7- iodoquinolin-8-01 and its pharmaceutical preparations, BP 1968 again employs the oxygen- flask combustion technique. The method of BP 1968 was superseded in BP 1973 by a non- aqueous titration for total phenolic substances, coupled with gas - liquid chromatography for 5~chloro-7-iodoquinolin-8-ol. The NF XII13 describes different assay procedures for the determination of 5-chloro-7-iodoquinolin-8-ol in different pharmaceutical preparations. Several methods are available in the literature for the determination of 5,7-diiodoquinolin-8- 01 and 5-chloro-7-iodoquinolin-8-ol as pure substances and in various formulations.Colori- metric methods involve the formation of coloured metal complexes in both instances. A colorimetric m e t h ~ d , ~ based on a previously described reaction,6 has been developed, which depends upon the formation of a coloured complex between 5-chloro-7-iodoquinolin-8-ol and the iron(II1) ion. The method has been used in order to determine 5-chloro-7-iodoquinolin- 8-01 in ointments. The yellow complex formed between copper sulphate and 5,7-diiodoquinolin-8-01 in di- methylformamide has been the basis for a colorimetric determination of this compound in tablets and suspensions.g A method, which is based on the measurement of the orange - yellow colour formed when a solution of 5,7-diiodoquinolin-8-01 or 5-chloro-7-iodoquinolin- 8-01 in glacial acetic acid is treated with sodium nitrite ~olution,~ has been applied to the determination of these compounds in tablets, powders and ointments.Also, an analytical procedure that is based on the formation of a yellow-coloured copper chelate,8 which is readily extractable into chloroform, has been used to determine either compoundin tablets and creams. However, colorimetric methods that are based on chelate formation, in general, suffer from a marked sensitivity to variations in the moisture content of the system.8 The spectrophotometric determination of 5-chloro-7-iodoquinolin-8-ol has been carried out by measurement of the absorbance of a solution of the compound in ethanolic hydro- chloric acid (0.1 N) at 257 or 325 It has also been determined spectrophotometrically at 39On1n.l~SOLIMAN 697 Infrared spectrophotometric methods include the semi-quantitative determination of 5-chloro-7-iodoquinolin-8-ol and its intermediatesll This method has been modified12 to fulfil the conditions of specificity, accuracy and precision that are required for quantitative analysis.The compound has also been determined in various pharmaceutical preparations by use of infrared ~pectrophotometry.~~ Although infrared spectrophotometric methods are specific, sometimes the lengthy extractions required before the final measurement is made render the method impractical for routine analyses in the pharmaceutical industry.Other methods described in the literature include the polarographic determination of 5,7- diiodoquinolin-8-01,~~ the indirect X-ray spectrographic determination of 5-chloro-7-iodo- quinolin-8-0115 and the determination of both compounds by gas - liquid chromatography of their trime th ylsilylethers . l6 Gravimetric and titrimetric methods that have been described include the precipitation of 5,7-diiodoquinolin-8-01 or 5-chloro-7-iodoquinolin-8-ol with cadmium iodide in acetone and then weighing the cadmium complex f0rmed.l' 5-Chloro-7-iodoquinolin-8-ol was also deter- mined by precipitation from solution in ethyl acetate with a 6 per cent. solution of mercury(I1) acetate in acetic acid, in the form of a mercury complex.1° The total halogen content of 5-chloro-7-iodoquinolin-8-ol was determined by liberating the halogens by means of fusion with sodium carbonate, adding an excess of standard silver nitrate solution and determining the excess of the latter by titration against standard potassium thiocyanate solution.In addition, the iodide ion formed was determined by using standard iodate solution.l* This compound has also been determined by precipitation in the form of a copper complex, the copper content of the residue being determined by iodimetric titration.19 Other titrimetric methods that have been described include the fusion of 5,7-diiodoquinolin-8-01 or 5-chloro- 7-iodoquinolin-8-01 with potassium carbonate and determination of the potassium iodide formed by titration against N-bromosuccinimide.20 5,7-Diiodoquinolin-8-01 has also been assayed by non-aqueous titration in dimethylformamide solvent against standard sodium methoxide solution, using thymol blue as the indicator.21 In this paper a simple and more direct analytical method that is applicable to the deter- mination of relatively high concentrations of 5,7-diiodoquinolin-8-01 or 5-chloro-7-iodoquinolin- 8-01 in pharmaceutical preparations without numerous dilutions of the sample is reported.The method has proved successful in determining accurately either compound alone or admixed with many other ingredients that are commonly encountered in dosage forms. Experimental Reagents 5,7-Diiodoquinolin-8-01. 5-Chloro-7-iodoquinolin-8-ol. Zinc metal powder. Chloroform, analytical-reagent grade. Dilute hydrochloric acid, 10 per cent.Concentrated hydrochloric acid, sp. gr. 1.18. Sodium hydroxide, 15 per cent. solution. Dissolve 15 g of sodium hydroxide in distilled water in order to make 100 ml of solution. Standard potassium iodate solution, 0.025 M. Weigh accurately 5.35 g of potassium iodate and dissolve it in distilled water so as to give 1000 ml of solution. Procedure Determination of 5,7-diiodoquinolin-8-01 Weigh, accurately, about 0.3 g of 5,7-diiodoquinolin-8-01, place it in a 250-ml Erlenmeyer flask, add about 0.5 g of zinc powder and then 30 ml of 15 per cent. sodium hydroxide solution. Boil the mixture for 30 min on a hot-plate under a reflux condenser. Disconnect the con- denser, wash down any particles adhering to the walls of the flask and the condenser with about 5 ml of sodium hydroxide solution and a small amount of distilled water, and reflux again for a further 10 min.Cool, filter the contents and wash the flask and the filter with three 10-ml portions of distilled water. Next, neutralise the combined filtrate and washings to litmus by using 10 ml of concentrated hydrochloric acid and about 10-20 ml of the dilute698 SOLIMAN : DETERMINATION OF ~,7-DIIODOQUINOLIN-8-OL Analyst, vol. 100 acid solution. Filter off the precipitate formed and wash it thoroughly by decantation, using four volumes, each of 10 ml, of distilled water and adding the washings to the filtrate. Add about 60 ml of concentrated hydrochloric acid for each 50 ml of flask contents and titrate the solution with 0.025 M potassium iodate solution until the dark brown solution becomes light brown.Then add 10ml of chloroform and continue the titration, shaking well after each addition, until the chloroform layer becomes colourless. Alternatively, separate the precipitate formed on neutralisation by centrifuging for 20 min at 5000 rev min-1, transferring the clear supernatant liquid quantitatively to the titration flask, then wash the residue with distilled water until it is completely suspended, centrifuge it again for a further 10 rnin and combine this supernatant liquid with the solution in the titration flask. Determination of 5-chloro-7-iodoquinolin-8-ol However, after separation of the pre- cipitate formed on neutralisation by either filtration or centrifugation, transfer the filtrate and washings to a suitable separator and extract with two 20-ml portions of chloroform. Reject the chloroform layer, transfer the aqueous phase into the titration flask, run the con- centrated hydrochloric acid through the separator into the flask and titrate as described above.Carry out the determination as described above. Calculate the percentage recovery as follows : A x 9-925 x 100 W Recovery of 5,7-diiodoquinolin-8-01, per cent. = A x 15.275 x 100 W Recovery of 5-chloro-7-iodoquinolin-8-ol, per cent. = where A ml is the volume of 0.025 M potassium iodate solution and W mg is the mass of the Determination of 5,7-diiodoqi~inolin-8-01 and 5-chloro-7-iodoquinolin-8-ol in tablets Transfer into a 250-ml Erlenmeyer flask an accurately weighed amount of the powder that is equivalent to about 300-0 mg of the active ingredient and proceed as described above under Determination of 5,7-diiodoquinolin-8-ol or Deter- mination of 5-chloro-7-iodoquinolin-8-ol. drug.Weigh and powder 20 tablets. Determination of 5,7-diiodoquinoli~~-8-ol and 5-chEoro-7-iodoquinolin-8-ol in suspensions Accurately measure a volume of the suspension equivalent to about 300.0mg of active ingredient and centrifuge it for 30min at 5000 revmin-l. Discard the clear supernatant liquid and wash the residue with a 40-ml portion of distilled water until it is completely suspended, centrifuging for 30 min and rejecting the clear supernatant liquid. Repeat the washing and centrifugation with a further 40-ml portion of water. Transfer the residue quanti- tatively into a 250-ml Erlenmeyer flask by use of 30 ml of 15 per cent.sodium hydroxide solu- tion and a small amount of distilled water and proceed as described above under Determination of 5,7-diiodoqzcinolin-8-oI or Determination of 5-chloro-7-iodoquinolin-8-ol. Results and Discussion The methods that are in general use for liberating the iodine content of organic compounds are the double crucible method, which involves fusion with alkali-metal carbonates, or the oxygen-flask combustion technique. Fusion with alkali-metal carbonates has several dangers and disadvantages, among which are the long time needed to complete the reaction, the risk of loss of halogen by volatilisation and the possibility of significant contamination of the sample by impurities present in the flux required to fuse the sample.On the other hand, the oxygen-flask combustion technique can be performed only on a small amount of sample. Therefore, a method was sought for the assay of 5,7-diiodoquinolin-8-01 and 5-chloro-7-iodoquinolin-8-ol that would circumvent some of these difficulties and, at the same time, be applicable to various pharmaceutical preparations.October, 1975 AND ~-CHLORO-7-IODOQUINOLIN-8-OL (CLIOQUINOL) 699 The analytical method reported in this paper is a modification of the BP 1968 method of assay of sodium acetrizoate injection.2 It has been found that the direct application of this method, without any modification, to the determination of 5,7-diiodoquinolin-8-01 and 5- chloro-7-iodoquinolin-8-01 is not possible in view of the fact that the organic products, resulting from the hydrolysis process, interfere in the subsequent steps of the analysis. Preliminary investigations on pure 5,7-diiodoquinolin-8-01 and 5-chloro-7-iodoquinolin-8-01 have indi- cated that the presence of such organic compounds, which are believed to be quinolin-8-01 and/or decomposition products thereof, positively interfere with the titration against stan- dard iodate solution, thus leading to higher recoveries.This interference can be explained as being due to partial iodination of these products under the conditions of the experiment, and/or consumption of iodate in an oxidation reaction. Such interference has been overcome by a suitable modification of the BP method in order to effect accurate determination of 5,7-diiodoquinolin-8-01 or 5-chloro-7-iodoquinolin-8-ol in the pure form as well as in various dosage forms.Further investigation has indicated that the interfering organic products can be precipitated quantitatively by neutralising the reaction mixture to litmus and filtering before the titration against the standard iodate solution is carried out. The solubility of the resulting precipitate increases in acidic or alkaline medium and therefore caution should be exercised not to overshoot the neutral point. In the determination of 5,7-diiodoquinolin-8-01 or 5-chloro-7-iodoquinolin-8-ol in sus- pensions in which these drugs are not the only therapeutic agents, advantage is taken of the complete insolubility of the two drugs in water to separate them, by centrifugation, from soluble ingredients prior to determination. This step also eliminates any interference in the deter- mination due to the presence of reducing sugars in suspensions.In the determination of 5-chloro-7-iodoquinolin-8-ol as the pure substance, as well as in the dosage forms, it is necessary to shake the solution gently with chloroform just before the titration is carried out, discarding the organic solvent. This extraction will eliminate any colour, which, if present in solution, will interfere with the detection of the end-point. Table I summarises the analytical results obtained in the determination of 5,7-diiodoquino- lin-8-01 and 5-chloro-7-iodoquinolin-8-ol by means of the developed method and the non- aqueous titration method. The non-aqueous titration21 of 5,7-diiodoquinolin-8-01 was carried out with dimethylformamide as the solvent and 0.1 N sodium methoxide solution as the titrant. The end-point was detected visually, by using thymol blue as indicator. 5-Chloro-7-iodoquinolin-8-01 was determined in the same manner but using pyridine as the solvent.The mean percentage recoveries (P‘ = 0.05) obtained by applying the proposed method to the determination of 5,7-diiodoquinolin-8-01 and 5-chloro-7-iodoquinolin-8-oP TABLE I DETERMINATION OF ~,7-DIIODOQUINOLIN-8-OL AND ~-CHLORO-7-IODOQUINOLIN-8-OL BY THE PROPOSED METHOD AND THE NON-AQUEOUS TITRATION PROCEDURE Proposed method --------7 Mass Mass Recovery, takenlmg found/mg per cent. 302.8 303.7 100.3 300-1 301.7 100-5 302.0 294-3 97.4 301-8 300.7 99.6 301.0 296-7 98.6 3 19.6 320.5 100.3 298.0 290.3 97.4 Mean (P’ = 0.05) = 99.2 f 1.26 5,7-Diiodoquinolin-8-ol- Non-aqueous titration procedure Mass Mass Recovery, takenlmg found/mg per cent.I L > 399.5 404.5 101-3 402-1 404.5 100.6 400.4 404.5 101.0 388.4 398.9 102.7 401.5 416.8 103.8 422-1 441.4 104.6 Mean (P’ = 0.05) = 102.3 f 1.71 5-Chloro-7-iodoqu~taoli~-8-ol- 105.6 109.9 104.1 499- 7 523-9 104-8 108- 8 106.9 98.2 500-5 619.4 103.8 133.0 13 1-4 98.8 499.9 512-9 102.6 210.8 216.9 102.9 304.4 294.8 96-9 Mean (I” = 0.05) = 100.2 3-89 Mean (P’ = 0.05) = 103-7 f 2.74700 SOLIMAN : DETERMINATION OF 5,7-DIIODOQUINOLIN-8-OL Analyst, VOZ. IOO were 99.2 & 1.26 and 100.2 3 3.89, respectively, compared with 102.3 & 1-71 and 103.7 & 2.74, respectively, when the non-aqueous titration method was applied.The high results obtained by using the latter method may be attributable to the presence of such impurities as halogenated isomers in commercial samples of 5-chloro-7-iodoquinolin-8-ol and other halogenated isomers in 5,7-diiodoquinolin-8-01. The results obtained from the determination of 5,7-diiodoquinolin-8-01 and 5-chloro-7- iodoquinolin-8-01 in tablets are given in Table 11. 5,7-Diiodoquinolin-8-01 was determined in tablets especially prepared so as to contain 300mg of the drug in each. The mean re- covery (P’ = 0.05) was 100-3 & 1.99 per cent. Tablet excipients, such as starch, gelatin, talc, soluble saccharin and fumed silica, did not interfere in the determination. The recovery of 5-chloro-7-iodoquinolin-8-ol from different batches of tablets that are marketed in Egypt and labelled to contain 250 mg of the drug was also investigated.The mean recovery (P’ = 0.05) was 101.6 & 4-55 per cent. TABLE I1 DETERMINATION OF 5,7-DIIODOQUINOLIN-8-OL AND ~-CHLORO-7-IODOQUINOLIN-8-OL I N SYNTHETIC AND COMMERCIAL TABLETS Mass of Average powdered mass of tablets Content of Mass Recovery, Brand tablet/mg taken/mg druglmg found/mg per cent. A 500 334.0 200-4 202.9 101.3 A 500 498.4 299.0 298-5 99.8 A 500 500.3 300.2 3044 101.4 A 500 501-9 301.1 297.5 98.8 Synthetic 5,7-diiodoquinolin-8-01 tablets (300 mg)- Mean (P’ = 0.05) = 100.3 -+ 1.99 Commercial 5-chloro-7-iodoquinolin-8-ol tablets (250 mg)- I3 399.2 479.7 300-2 307.0 102.3 B 399.2 328.1 205-5 209-3 101.8 €3 399-2 326.7 204.6 200.1 97.8 B 399.2 163-0 102.1 106.9 104-7 Mean (P’ = 0.05) = 101.6 f 4-55 C 400-4 236.5 147.6 140.5 95-2 D 398.5 507.1 318.1 307.0 96.5 E 396.9 160.1 100.8 93.2 92.5 E 396.9 480.8 302-8 279.5 92.3 The proposed method has been applied to the determination of 5,7-&iodoquinolin-8-01 in suspensions.The results of these analyses can be seen in Table 111. The mean recoveries (P’ = 0.05) of the drug in specially prepared suspensions containing 2.5 per cent. m/V, and in commercial suspensions labelled to contain the same percentage, were 10043 & 3.26 and 101.5 & 4.18 per cent., respectively. Substances that are commonly encountered in suspensions of 5,7-diiodoquinolin-8-01, such as sulphaguanidine (8 per cent .) , bismuth sub- carbonate (5 per cent.), light kaolin (8 per cent.), sodium citrate (5 per cent.), sucrose (40 per cent .), carboxymethylcellulose (1 per cent .) , Tween 80 (1 per cent .), methyl hydroxybenzoate (0.15 per cent.), propyl hydroxybenzoate (0.05 per cent.), liquid extract of belladonna (0.1 per cent.) and vanillin (0.1 per cent.), do not interfere in the determination. The calculated mean recovery (P’ = 0.05) for the determination of 5-chloro-7-iodoquinolin-8-ol in commercial suspensions, as shown in Table 111, was 97-9 4.84 per cent.Up to 7 per cent. m/V of formosulphathiazole, the condensation product of sulphathiazole and formaldehyde, does not interfere in its determination in suspensions. The recovery of 5,7-diiodoquinolin-8-01, when added to commercial suspensions already containing it, was determined by adding a known amount of the pure drug to suspensions, the 5,7-diiodoquinolin-8-01 content of which had previously been determined.The results given in Table IVY obtained with some of these suspensions, are representative. These control experiments were conducted in order to differentiate between experimental errors andOctober, 1975 AND ~-CHLORO-7-IODOQUINOLIN-8-OL (CLIOQUINOL) TABLE I11 DETERMINATION OF 5,7-DIIODOQUINOLIN-8-OL AND ~-CHLORO-7-IODOQUINOLIN-8-OL I N SYNTHETIC AND COMMERCIAL SUSPENSIONS Amount Volume Content of of drug Recovery, taken/ml drug/mg found/mg per cent. 12-5 312.5 307.7 98-4 12.5 312.5 322.5 103.2 5.0 125.0 127.0 101.6 5.0 125.0 125.1 100.1 Synthetic suspensio.lzs containing 2.5 per cent. m/V of 5,7-diiodoquinolin-8-ol- Mean (I" = 0.05) = 100.8 & 3.26 Commercial sus+ensions labelled to contain 2.5 per cent.mlV of 5,7-diiodoquinolin-8-01- 10.0 250.0 252.1 100.8 10.0 250.0 252-1 100.8 10.0 250.0 242-2 96.9 10.0 250-0 258.0 103.2 10.0 250.0 264.9 106.0 Mean (I" = 0.05) = 101.5 f 4.18 Commercial szsspensions labelled to contain 3.0 per cent. mlV of 5-chloro-7-iodoqui~olin-8-ol- 10.0 300.0 297.9 99.3 10.0 300-0 296.3 98.8 10.0 300.0 287.2 95.7 Mean (P' = 0.05) = 97.9 f 4.84 701 errors due to the interaction of other constituents encountered in the system or caused by bulk production. The method developed in this investigation has the advantages of being simple and appli- cable, over a convenient range of concentrations, to the determination of 5,7-diiodoquinolin- 8-01 and 5-chloro-7-iodoquinolin-8-ol in the pure state or in different pharmaceutical for- mulations, without numerous dilutions of the sample.The high concentrations of the drugs TABLE IV RECOVERY OF 5,7-DIIODOQUINOLIN-8-OL FROM COMMERCIAL SUSPENSIONS LABELLED TO CONTAIN 2.5 PER CENT. WITH ADDED ~,7-DIIODOQUINOLIN-8-OL Number Volume of Labelled Added Total suspension amount of amount of amount of Recovery/ takenlml druglmg druglmg druglmg mg 1 5.0 125.0 0 125.0 125.1 5.0 125.0 98.8 223.8 225.3 Recovery of added amount of drug = 100.2 mg (101.4 per cent.) 2 5.0 125-0 0 125.0 124.1 5.0 125-0 204.0 929.0 326.5 Recovery of added amount of drug = 202.5 mg (99.3 per cent.) 3 5.0 125-0 98.6 223.6 215.9 4 10.0 250.0 100-6 350-6 362.2 Recovery, per cent. 100.1 100.7 99.3 99.2 96.6 103.3 usually encountered in their formulations cause no problems and render the great dilutions and low-level determinations that are inherent in instrumental methods unnecessary.Such dilutions can result in high experimental error. In addition, the method requires no special skills and does not necessitate the use of expensive instruments, which may not be readily available in developing and underdeveloped countries where preparations containing 5,7-di- iodoquinolin-8-01 and 5-chloro-7-iodoquinolin-8-01 are most needed. The author thanks Dr. Ibrahim H. Abdallah, President of the Alexandria Company for Pharmaceuticals and former Professor of Pharmaceutical Chemistry, for generously supplying702 SOLIMAN chemicals and providing laboratory facilities, and Miss Sdwa A.Moustafa for technical assistance. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. References “United States Pharmacopeia,” XVIIIth Revision, Mack Publishing Co., Easton, Pa., 1970, pp. 201 “British Pharmacopo:@ 1968,’’ The Pharmaceutical Press, London, 1968, pp. 226, 338 and 858. “National Formulary, Cohen, J., and Kluchesky, E., J. Pharm. Sci., 1963, 52, 693. Haskins, W. T., and Luttermoser, G. W., Analyt. Chem., 1951, 23, 456. Ozsoz, B., Turk. Ij. Tecr. Biyol. Derg., 1963, 23, 113; Analyt. Abstr., 1963, 10, 4888. Vadodaria, D. J., Desai, B. R., and Mukherji, S. P., Indian J. Pharm., 1965, 27, 257; Analyt. Abstr. Windheuser, J. J., and Chu, D. Y., J. Pharm. Sci., 1967, 56, 519. Melysz, M., and Nowakowska, Z., Acta Pol. Pharm., 1966, 23, 39; Analyt. Abstr., 1967, 14, 3538. Moheric, J., Farmaceutski Vest., 1969, 20, 1; Analyt. Abstr., 1970, 19, 670. Bigeard, F., Clerque, H., Kern, P., and Vaupre, R., Annls Pharm. Fr., 1964, 22, 667. Urbanyi, T., Sloniewsky, D., and Tishler, F., J. Pharm. Sci., 1966, 55, 730. Urbanyi, T., and Stober, H., J. Pharm. Sci., 1969, 58, 232. Brown, L. W., and Krupski, E., J. Pharm. Sci., 1961, 50, 49. Papariello, G. J., Letterman, H., and Mader, W. J., Analyt. Chem., 1962, 34, 1261. Gruber, M. P., Klein, R. W., Fox, M. E., and Campisi, J., J. Pharm. Sci., 1972, 61, 1147. Anantanrayanan, K. G., Kudalkar, V. G., Madiwale, M. S., Desai, H. H., and Walawalker, &I. B., Mukerje, S. L., and Dey, A. P., Drug Stand., 1959, 27, 18; Analyt. Abstr., 1959, 6 , 4168. Elmer, V., Ferenc, K., and Marianna, G., Acta Pharm. Hung., 1972, 42, 141; Analyt. Abstv., 1972, Barakat, M. Z., Bassioni, M., and El-Wakil, M., Analyst, 1972, 97, 466. Kavarana, H. H., Am. J . Pharm., 1959, 131, 1959. and 338. 13th Edition, Mack Publishing Co., Easton, Pa., 1970, p. 371. 1967, 14, 1005. Indian J. Pharm., 1959, 21, 263; Analyt. Abstr., 1960, 7, 2947. 23, 4910. Received September 23rd, 1974 Amended December 2nd, 1974 Accepted March 26th, 1975

ISSN:0003-2654    

DOI:10.1039/AN9750000696

出版商:RSC    

年代:1975

数据来源: RSC

摘要:

Analyst, October, 1975, Vol. 100, pp. 703-707 703 Differential Spectrophotometric Method for the Determination of Vitamin A (Retinol) by Using Trifluoroacetic Acid, and Its Application to Related Compounds Samir A. Gharbo and Leo A. Gosser Wawen-Teed Research Center, Warren-Teed Pharmace.uticaEs Inzc., 582 West GoodaZe Street, Colzsmbzcs, Ohio 43215, U.S.A . A differential spectrophotometric procedure for the determination of vitamin A (retinol) based on the formation of a pink colour in a trifluoroacetic acid - perchloric acid medium has been developed, Maximum absorbance of the pink colour with Amax. at 502nm was attained within 7-12min and remained stable for at least 30 s ; the intensity of this colour was equivalent to 45 per cent. of the intensity of the blue colour produced by the well known antimony(II1) chloride procedure.The pink colour could be destroyed within 2-4 min by the addition of pentane-2,4-dione followed by hydrogen peroxide. A rectilinear graph of absorbance versus amount of vitamin A was obtained for the 0-20-pg range, optimum results being achieved in the 6-10-pg range. The relative standard deviation of the method for 8-3 pg was f0-7 per cent. The reactions of vitamin A acetate, retinal, retinoic acid, 8-carotene, ergocalciferol, cholecalciferol, cholesterol, ergosterol, phytonadione and a-tocopherol were also investigated employing the same reagent additions. Several methods are available for the determination of vitamin A (retino1)l in foods,2 phar- maceutical preparations3 and blood, but of these the colorimetric method with antimony(II1) chloride4 is the most widely used.The blue colour (Amx. 616-620 nm) that develops in the reaction with antimony( 111) chloride, trichloroacetic acid5 and other strong Lewis acids6 is very sensitive but is subject to rapid fading within several seconds. Vitamin A gives the same colour reaction with trifluoroacetic acid (TFA)6s7 with much less sensitivity to moisture8 and this reagent is preferred, especially in micro- determination^.^ The blue colour produced with TFA decays to a more stable secondary pink colour (Amx. 520-550 nm) within 1.5-2 h, but the intensity of this pink colour is less than one third of that of the blue colour.6 Thus, the difficulties with the colorimetric determination of vitamin A are that the blue colour, although very sensitive, is unstable while the pink colour, although more stable, is less sensitive and develops slowly.During the investigation of chemical methods for the determination of vitamin A in vitamin mixtures, it was noted that temperature, solvents and oxidising agents all have an effect on the speed of development and the sensitivity of the pink colour. It was found that a mixture of trifluoroacetic acid and 0.1 N perchloric acid (5 + 1) added to a dichloromethane solution of vitamin A gave the best sensitivity within 7-12 min and the absorbance at 502 nm was stable for at least 30 s. This absorbance could be destroyed rapidly within 2 4 min by the addition of pentane-2,4-dione followed by hydrogen peroxide. In this paper, the optimum conditions for a differential spectrophotometric procedure for determining vitamin A are described.In addition, the reactions of related compounds that may be present with vitamin A in pharmaceutical preparations or in blood are reported. Experimental Reagents acid and should be handled with care.) Tri$uoroacetic acid. “OR” grade. (CAUTION-Trifluoroacetic acid is an extremely corrosive Perchloric acid, 0.1 N solution in glacial acetic acid. Dichloromethane. Spectroscopic quality. Pentane-2,4-dione (acetylacetone) .704 GHARBO AND GOSSER : DIFFERENTIAL SPECTROPHOTOMETRIC Analyst, Vol. 100 Hydrogen peroxide solution , 30 per cent. Analytical-reagent grade. All-trans-vitamin A. Vitamin A acetate. USP reference standard in oil. All-trans-retinal. A ll-t rans-retinoic acid.p-Carotene, ergosterol, cholesterol and cholecalciferol, 100 per cent. pure, crystalline. Ergocalciferol. USP reference standard. Phytonadione. USP reference standard. m-a-Tocopherol. Apparatus Perkin-Elmer, Coleman Model 124, spectrophotometer. Rotary evaporator. Fast-delivery pipette. Capacity, 2.0 ml. Procedure Transfer an aliquot of a solution of the sample, preferably dry, in a volatile organic solvent, containing 5-10 pg of vitamin A, into a 10- or 25-ml flask and evaporate it to dryness under vacuum at 3540 "C. Add 1 ml of dichloromethane to the flask and again evaporate to dryness. To the residue in the flask, add 0.20 ml of dichloromethane, mix well to effect dissolution, and quickly add 2.0 ml of the freshly mixed solvent trifluoroacetic acid - 0.1 N perchloric acid (5 + 1).Gently swirl the flask in order to mix the contents and immediately transfer the mixture into a 1-cm spectrophotometer cuvette. Monitor the absorbance at 502 I f 2 nm and record the maximum value attained (usually within 7-12 min of mixing), zeroing theinstrument with the mixed solvent as blank. After the maximum absorbance has been recorded, add 1 drop of pentane-2,4-dione, with mixing, followed by 1 drop of hydrogen peroxide solution, with mixing, and measure the absorbance at the same wavelength 3 min -J= 30 s after the addition of the hydrogen peroxide. Subtract the second absorbance reading from the maximum absorb- ance reading and calculate the amount of vitamin A by comparison with a similarly treated standard solution. Results and Discussion Efforts to utilise the pink colour developed in the reaction of vitamin A with trifluoroacetic acid, led to an investigation of this reaction and the parameters that affect it.Simply allowing vitamin A to react with TFA at room temperature gives an initial blue colour that decays to a pink colour. It was found that a fairly stable isosbestic point was reached within 5- 10 min at 532 I f 2 nm, which was unsatisfactory as the basis of an analytical procedure, as the absorbance value obtained was relatively weak and easily affected by solvent and pH variations. Heating vitamin A in various mixtures of chloroform and TFA at 50 "C for time intervals of 1 4 m i n demonstrated that the pink colour could be developed more rapidly and completely, but again it was easily affected by slight variations in experimental con- ditions.Encouraging results were obtained initially when it was found that the addition of hydrogen peroxide as an oxidant, in conjunction with TFA, gave increased sensitivity to the pink colour as the maximum absorbance at 502 nm developed within 10 min and remained stable for about 1 min. The effects of nitric acid, perchloric acid and mixtures of these acids, in lieu of hydrogen peroxide, were then investigated. The most satisfactory combination was found to be a 5 + 1 mixture of TFA with 0.1 N perchloric acid in glacial acetic acid. The presence of dichloromethane before the addition of TFA was found to be advantageous, but the sensitivity decreased somewhat with the use of 1,2-dichloroethane, chloroform, carbon tetrachloride and benzene, in that order, with absolute ethanol and acetone being definitely inferior.During the initial work with hydrogen peroxide as an aid to TFA for the rapid development of the pink colour, it was found that the addition of a small amount of acetone could destroy the colour within 15 min. Then, when developing the colour with a trifluoroacetic acid - 0.1 N perchloric acid (5 + 1) mixture, it was found that the colour was only slightly affected by the addition of a small amount of either hydrogen peroxide or acetone alone; however, ifOctober, 1975 DETERMINATION OF VITAMIN A WITH TRIFLUOROACETIC ACID 705 hydrogen peroxide and acetone were added successively or in a 1 + 1 mixture, the colour was removed within 15-20 min.On the assumption that the keto group of the acetone had a role in this reaction, pentane-2,4-dione was substituted for acetone with the result that the absorbance of the colour was reduced to less than 0.02 within 2-4 min. Based on these observations, the conditions described in the procedure were deemed to be the optimum for the determination of vitamin A. The visible spectrum obtained in the reaction of trifluoroacetic and perchloric acid with vitamin A at maximum intensity (Fig. 1A) illustrates the absorbance band at 502 nm with only a slight shoulder at 528 nm. The absorbance at 502 nm increases rapidly in the first 5 min, attains maximum intensity within 12 min and remains stable for a t least 30 s, before decreasing slowly. The primary blue colour, which appears at 595 nm and which decays rapidly within a few seconds of the addition of the reagent, has about twice the intensity of the secondary pink colour.On the addition of pentane-2,4-dione, with mixing, the absorbance falls only slightly, but on subsequent addition of hydrogen peroxide, with mixing, the absorbance falls rapidly in the first 1 min and then attains a stable intensity (less than 0.02 absorbance unit) within 2 4 min (Fig. 1B). Addition of the last two reagents in the reverse order gives essentially the same result. The spectrum shown in Fig. 1B is also given by a blank reaction without vitamin A. It is important that the sample and standard solution are treated in a similar way as the concen- tration of TFA may vary from batch to batch and degree of use.The perchloric acid should be mixed with the TFA shortly before use (preferably less than 1 h) as the mixture tends to lose potency with time. A" 0.6 0.4 0 m e 8 n a 0.2 Wavelengthhm Fig. 1. A typical spectrum of colour developed in the reaction of vitamin A with trifluoroacetic acid - perchloric acid mixture before (A) and after (B) the addition of pentane-2,4-dione, then hydrogen peroxide. Various dilutions of vitamin A [Fig. 2(a)] and of the USP reference standard vitamin A acetate [Fig. 2 ( b ) ] in chloroform were used to establish thelinearity of response. The graphs of maximum absorbance at 502 nm versus the amount of vitamin A and that of the acetate was rectilinear in the range 0-20 pg, with optimum results being obtained in the 5-10-pg range. The relative standard deviations of the absorbance for six aliquots of vitamin A containing 8.3 pg and of the acetate containing 8.4 pg were found to be 0.66 and 1-07 per cent., respect- ively.706 GHARBO AND GOSSER : DIFFERENTIAL SPECTROPHOTOMETRIC Analyst, VoZ.100 2.0 I 1 Amount of vitamin A/pg Amount of vitamin A acetate/pg Fig. 2. Linear relationship of different samples and their dilutions of vitamin A (a) and reference standard vitamin A acetate (two samples) ( b ) to the colour development in the trifluoroacetic acid - perchloric acid procedure. The Ei& values calculated from the average of several determinations of vitamin A and its acetate (Table I) indicate that the value of the acetate is 2 per cent. higher than that of the alcohol, which is in contrast to a value expected to be 15 per cent.lower based on the difference in relative molecular masses and according to the comparable literature values for the blue colour with TFA.6p9 Hydrolysis of vitamin A acetate by the USP XVIII10 pro- cedure and determination of the resulting alcohol by both the ultravioletlo and the colori- metric procedures gave 97.9 & 1.8 and 96.9 & 2.4 per cent. recoveries, respectively. Vitamin A acetate was adsorbed on deactivated neutral silica, chromatographed on neutral alumina, eluted with acetone and the acetate determined in the eluate by both the ultraviolet and the colorimetric procedures, which gave 99.9 & 0.7 and 97.5 & 1.5 per cent. recoveries, respec- tively. Apparently, the acetate does indeed produce a higher colour yield than does the alcohol on an equimolar basis.The colorimetric procedure was applied to some related compounds that may be present with vitamin A in foods, pharmaceutical preparations and in blood, and the results are shown in Table I. The spectra of all of the compounds in the table showed single peaks except for /3-carotene, which showed a secondary peak at 475 nm, and a-tocopherol, which showed an additional shoulder at 430nm. All the peaks disappeared on the addition of pentane-2,4- dione and hydrogen peroxide. However, it is evident that differential spectrophotometry could not be used for compounds that have their maxima at wavelengths below 460 nm as the blank solvents begin to make a significant contribution to the absorbance (Fig.1B). TABLE 1 COMPARATIVE RESULTS OF THE APPLICATION OF THE DIFFERENTIAL SPECTROPHOTOMETRIC PROCEDURE FOR THE DETERMINATION OF VITAMIN A TO RELATED COMPOUNDS Time taken to reach Wavelength maximum (Amax. f 2)/ Et:k absorbance/ Compound nm at Amax. min Vitamin A . . . . . . . . 502 1670 10 f 2 Vitamin A acetate . . .. . . 502 1704 10 f 2 Retinal . . .. . . . . .. 408 2138 8 f 2 Retinoic acid . . .. . . . . 450 1870 13 f 3 Minimum time of stability of Ei& 30 1670 30 1704 180 142 60 122 absorbancels a t 502 nm p-Carotene . . .. .. Ergocalciferol . . .. .. .. 485 Cholecalciferol . . .. .. .. 485 Cholesterol . . .. * . .. 412 Ergosterol . . ,. . . .. 383 Phytonadione . . .. .. . . 400 DL-a-Tocopherol . . . . .. 458 * At 90 min. 332 108 1161 1096 163 213 125 28" 7 f l 30 17 f 2 60 3 z t 1 60 3 f l 30 2 f 3 10 2.5 f 1 20 60 f 6 120 > 90 - 92 102 641 517 2 34 0.8 2*5* Several advantages over other colorimetric methods that are used for the determination of vitamin A could be attributed to the method described in this paper. The developedOctober, 1975 DETERMINATION OF VITAMIN A WITH TRIFLUOROACETIC ACID 707 colour, in addition to giving an absorbance graph that is rectilinear with concentration, is relatively stable, sensitive and rapidly formed. The procedure is simple and provides a distinct advantage in the use of differential spectrophotometry in the presence of spectral inter- ferences of a general background nature.There were no significant day-to-day variations in results noted under similar conditions. 1. 2. 3. 4. 8. 6. 7. 8. 9. 10. References Hashmi, M., “Assay of Vitamins in Pharmaceutical Preparations,” John Wiley and Sons Ltd., New Parrish, D. B., J . A s s . Off. Analyt. Chem., 1974, 57 (4), 897 and 903. Chesterfield, N. J., Aust. J . Pharm., 1969, 50, 597 (Supplement 82). Carr, F. H., and Price, E. A., Bzochem. J.. 1926, 20, 497. Bayfield, R. E., Analyt. Biochem., 1971, 39, 282. Dugan, R. E., Frigerio, N. A., and Siebert, J. M., Analyt. Chem., 1964, 36, 114. Neeld, J. B., and Pearson, W. N., J . Nutr., 1963, 79, 464. Stross, P. S., and Brealey, L., J . Pharm. Pharmac., 1955, 7, 739. Bradley, D. W., and Hornbeck, C. L., Biochem. Med., 1973, 7, 78. “United States Pharmacopeia,” XVIIIth Revision, Mack Publishing Co., Easton, Pa., 1970, p. 914. York, 1973, p. 19. Received January 27th, 1975 Accepted May 6th. 1975

ISSN:0003-2654    

DOI:10.1039/AN9750000703

出版商:RSC    

年代:1975

数据来源: RSC

摘要:

708 Analyst, October, 1975, Vole 100, pp. 708-715 Spectrophotometric Determination of Trace Amounts of Vanadium by Formation of the Vanadium - 4-(2- Pyridy1azo)resorcinol (PAR) = Crystal Violet Complex: Application to the Analysis of Plant Materials J Minczewski, J. Chwastowska and Pham thi Hong Mai Laboratory of Analytical Chemistry, Polytechnic Institute of Wavsaw, 3 Noakowski Street, 00-664 Warsaw, Poland The anionic complex of vanadium with 4-(2-pyridylazo)resorcinol (PAR), obtained a t pH 4-6-51, forms an ion pair with Crystal Violet. This ion pair can be extracted with benzene - isobutyl methyl ketone (3 f 2). In the extracted ion pair the proportions of vanadium to PAR to Crystal Violet are 1 : 1 : 1. The coloured extract gives an absorption maximum at 585 nm, and Beer's law is obeyed in the concentration range 0-05-0.5 p g ml-l of vanadium. The molar absorptivity is equal to 1.1 x lo5 1 mol-I cm-l.The method was applied to the determination of trace amounts of vanadium in dried plant materials after wet ashing of the sample and extraction of the vanadium as the N-benzoyl-N-phenylhydroxylamine complex. The results obtained for the determination of vanadium in serradella, rye straw and grain and trefoil showed a relative standard deviation of 8-15 per cent. a t the 0-1-0.25 p.p.m. level of vanadium and of about 25 per cent. a t the 0.05 p.p.m. level. Ion pairs composed of an anionic complex of the metal ion to be determined and a cationic dye molecule are known to have high molar abs0rptivities.l It was assumed that a sensitive spectrophotometric method for the determination of vanadium in plant materials would make it possible to determine vanadium in a sample of not more than 5-10 g of dry material.The concentration of vanadium in plant material varies in the range 1-0-01 p.p.m. and there- fore the 5-g (10-g) sample would contain 5-4-05 pg (10-0.1 pg) of vanadium. Assuming that this amount of vanadium is finally contained in 10 ml of solution, then the range of concen- trations for the solution of the coloured vanadium complex corresponds to 14-01 pg ml-l of vanadium. To obtain reasonable values for the absorbance, the molar absorptivity of the coloured compound must be of the order of about lo5 1 mol-l cm-1. The ion pairs mentioned above have this property but sufficiently sensitive methods for the determination of vanadium are not known.The purpose of this work was to devise such a method, using the anionic complex of vanadium with 4-(2-pyridylazo)resorcinol( PAR)2-5 as the anionic component of the ion pair. Crystal Violet has been chosen as the cationic component. This paper presents the characterisation of this complex and its use in the determination of vanadium in plant materials. Experimental Reagents M. Vanadium(V) oxide (0.1785 g and 0.4547 g, respectively), freshly heated to 500 "C, was dissolved in 100 ml of 1 M sodium hydroxide solution. Lower concentrations were obtained by diluting the stock solutions with water . M. Crystal Violet (0.5701 g) was dissolved in water and the solution made up to 100ml with water. PAR (0.0064g) was dissolved in two drops of 1 M sodium hydroxide solution and the solution made up to 100 ml with water, Vanadium stock solutions, 100 pg ml-l and 5 x Crystal Violet solution, PAR solution, 5 x Cerium(IV) sulphate, 1 per cent.m/V solution in 1 N sulphuric acid. Bufer solution, pH 4-8. This solution was obtained by mixing the appropriate volumes M.MINCZEWSKI, CHWASTOWSKA AND MA1 709 of 1 M potassium dihydrogen orthophosphate solution and 1 M disodium hydrogen ortho- phosphate solution. Extraction solvent, benzene - isobutyl methyl ketone (3 + 2). Benzene was previously washed three times with water. Isobutyl methyl ketone was washed with 1 M sodium hydroxide solution and then with water until it gave a neutral reaction. Apparatus Wet-ashing apiharatus.The apparatus described by Bethge6 was used. Spectrophotometers. Perkin-Elmer 450 and Spekol, Carl Zeiss, Jena. p H meter. Elpho N-512. Choice of Dye and Extraction Solvent The type of dye used has an important influence on the sensitivity of the method and on the absorbance of the blank. The use of Crystal Violet, Methyl Violet, Diamond Green CB, Malachite Green, Rhodamine B and Rhodamine 6G has been investigated. The extraction solvent used should extract completely the vanadium - PAR - dye complex and should not extract the dye itself or its complex with PAR. Use of benzene and its mixtures with acetone, isobutyl methyl ketone and diisopropyl ether, and also of n-pentyl acetate, has been studied and the results are shown in Fig. 1. It can be seen that the best results were obtained when using the solvent benzene - isobutyl methyl ketone (3 + 2).0.700 0-600 0.500 a C m -f) 0.400 9 a a 0.300 0.200 0.1 00 Benzene-acetone A I \ 9 + 1 3+2+15% of ethanol Benzene -isobutyl methyl ketone - 9 + 1 3+2+15 of ethanc Benzene-diisopropyl ether A I 9 + 1 5 + 1 - n - Pe n tyl acetate +15% of ethanol n Fig. 1. Absorbance and blank values for the vanadium - PAR - dye complexYn various solvents: un- shaded regions, absorbances of complex; shaded regions, blank values. [Vanadium] = 5 x M ; [PAR] = 3 x M ; [dyeJ = 1 x 10-4 M. Path length, 1 cm. Wavelengths (nm) are given a t the bottom of each column. Dyes: 1, Crystal Violet; 2, Methyl Violet; 3, Diamond Green CB; 4, Malachite Green; 5, Rhodamine B; and 6, Rhodamine 6G. After extraction the organic phase was often turbid owing to the presence of water.The addition of 15 per cent. V/V of ethanol made this phase fully transparent. The addition of ethanol also caused a shift in the absorption maximum and an increase in the absorbance (Table I).710 MINCZEWSKI st al. : SPECTROPHOTOMETRIC DETERMINATION Analyst, VoZ. 100 The highest values for the absorbance were obtained when using Crystal Violet and Rhodamine B, the blank absorbance with the latter, however, being twice as high as that with Crystal Violet. In the system finally chosen, Crystal Violet was used as the dye in the vanadium - PAR - dye complex and benzene - isobutyl methyl ketone (3 + 2) as extraction solvent. After extraction 15 per cent. V/V of ethanol was added. TABLE I EFFECT OF THE ADDITION OF ETHANOL TO THE EXTRACTION SOLVENT ON THE ABSORBANCE OF THE VANADIUM - PAR - CRYSTAL VIOLET COMPLEX Solvent Amax./nm Absorbance Benzene - isobutyl methyl ketone (2 + 1) . . 545 0.440 + 16 per cent. V / V of ethanol . . . . 685 0-510 Benzene - isobutyl methyl ketone (3 + 2) . . 547 0.430 + 15 per cent. VlV of ethanol . . . . 585 0.550 Characteristics of Absorption Spectra In Fig. 2 the absorption spectra of Crystal Violet (l), the vanadium - PAR complex in water (3), the vanadium - PAR - Crystal Violet complex (2) and the reagent blank (4), all except (3) in solution in benzene - isobutyl methyl ketone (3 + 2) plus 15 per cent. VlV of ethanol, are given. The vanadium complex gives an absorption maximum at a wavelength of 580- 585 nm. Curve 2, for the vanadium complex, is very similar to that for the Crystal Violet alone (I) and shows a small deformation at 540-570 nm, caused by absorption due to the vanadium - PAR complex.1 -0 0) C m .t! % n a 0.5 0 I I I 400 450 500 550 600 Wavelengthhm Fig. 2. Absorption spectra of: 1, Crystal Violet solution in the solvent benzene-isobutyl methyl ketone (3 + 2) + 15 per cent. of ethanol ([Crystal Violet] = 1 x M) ; 2, vanadium - PAR - Crystal Violet complex in the same solvent ([vanadium] = 0.96 x 1 0 - 5 ~ , [PAR] = 6 x 1 0 - 5 ~ , [Crystal Violet] = 1 x 10-5 M) ; 3, vanadium - PAR complex in water ([vanadium] = 0-96 x 10-5 M ; [PAR] = 5 x 10-5 M); and 4, reagent blank. Path length, 1 cm. Formation and Extraction of the Vanadium - PAR - Crystal Violet Complex The effect of the pH of the aqueous phase and of the initial concentrations of PAR and Crystal Violet on the absorbance of the extract has been investigated.The effect of pH inOctober, 1975 OF TRACE AMOUNTS OF VANADIUM I N PLANT MATERIALS 711 the range 3-9 has been examined and it can be seen from Fig. 3 that maximum absorbance is obtained in the pH range 4.6-5-1. At pH values above 5-1 the blank absorbance increases considerably owing to the fact that at pH values greater than 6 the PAR is completely in the form of an anion and the possibility of PAR - Crystal Violet complex formation is much higher. 0.7 0.6 0.5 8 0.4 c m e d 0.3 n 0.2 0.1 01 I I I I I 1 5 3 4 5 6 7 8 9 PH Fig. 3. Variation of absorbance with pH for the vanadium - PAR - Crystal Violet complex in aqueous solution (curve 1).Curve 2, reagent blank. Curve 3, the difference between curves 1 and 2. [Vanadium] = 5 x M. Wavelength, 585 nm. Path length, 1 cm. M; [PAR] = 3 x M; [Crystal Violet] = 1 x To maintain the pH in the range 44-5.1, phosphate buffer solution was used. It was shown that 2 ml of buffer solution were sufficient to keep the pH constant and did not influence the formation of the vanadium - PAR - Crystal Violet complex. The dependences of the absorbance of the extract and of the blank value on the initial concentration of PAR and on the initial concentration of Crystal Violet are shown in Figs. 4 and 5, respectively. For a concentration of vanadium of 5 x M the most efficient extraction of the complex occurs with an initial concentration of PAR above 2.5 x M and with an initial concen- tration of Crystal Violet above 1 x 10" M.Therefore the proportions of vanadium to PAR to Crystal Violet are 1 : 5 : 20. After extraction, the complex is stable for 5 min, after which the absorbance continuously decreases. The extraction proceeds rapidly and it is sufficient to shake the mixture for 2 min in order to achieve 100 per cent. extraction of the complex. Beer's law is obeyed by the coloured solution of the complex for concentrations of vanadium in the range 0.05-0.5 pg ml-1 and the molar absorptivity is equal to 1.1 x lo5 1 mol cm-l. Composition of the Complex Job's method and that of Asmus7 have been used for investigation of the ratio of vanadium to PAR and of vanadium- PAR to Crystal Violet. The results obtained show that the complex formed in a weakly acidic medium coiisists of vanadium, PAR and Crystal Violet in equimolar proportions.These results confirm the literature data,2$4 which show that vanadium forms with PAR a 1 : 1 anionic complex that has a single negative charge. This complex associates with one molecule of Crystal Violet. In solution at a pH of approxi- mately 4.5 vanadium is mainly in the form of V03-.899 Thus, the following mechanism for the formation of the final complex can be formulated: HVO, + HR- + V0,R- + H,O where HR- represents PAR, and712 MINCZEWSKI et aZ. : SPECTROPHOTOMETRIC DETERMINATION Analyst, VoZ. 100 V0,R- + CVf + V0,R-.CV+ where CV represents Crystal Violet. The following structural formula for the ion-association complex can be assumed : L Interference by Anions The proposed method for the determination of vanadium by formation of a complex with PAR and Crystal Violet is not selective. The PAR anion forms coloured, extractable com- plexes with some cations and the Crystal Violet cation can form extractable ion-association complexes with simple anions.Thus, if the concentration of interfering ions considerably exceeds the concentration of vanadium, erroneous results for the absorbance of the vana- dium - PAR - Crystal Violet complex will be obtained. Table I1 shows the absorbances of the vanadium - PAR - Crystal Violet complex and of the blank solution in the presence of some common anions. It can be concluded that only sulphate and acetate ions do not interfere in the determination.Chlorides, fluorides, oxalates, citrates and tartrates do not interfere at concentrations lower than 100 times the vanadium concentration. The anions that interfere most are perchlorate, thiocyanate, nitrate and iodide ions. The EDTA anion does not influence the blank determination, but interferes with the formation of the coloured complex, acting as a masking agent on the vanadium ion. 01 I I I I I 1 2 3 4 5 6 Concentration of PAR x 1 0 5 / ~ Fig. 4. Variation of absorbance with the initial concentration of PAR for the vanadium - PAR - Crystal Violet complex (curve 1). Curve 2, reagent blank. Curve 3, the difference between curves 1 and 2. [Vanadium] = 5 x M ; [Crystal Violet] = 1 x M. pH, 4.8. Wavelength, 585 nm. Path length, 1 cm. :::I 1 - y I 0.5 1 1.5 2 2.5 Concentration of Crystal Violet X i 0 4 / ~ Fig.5. Variation of absorbance with the initial concentration of Crystal Violet for the vanadium - PAR - Crystal Violet complex (curve 1 ) . Curve 2, reagent blank. Curve 3, the difference between curves 1 and 2. [Vanadium] = 5 x 1 0 - * ~ ; [PAR] = 3 x 1 0 - 5 ~ . pH, 4.8. Wavelength, 585 nm. Path length, 1 cm.October, 1975 OF TRACE AMOUNTS OF VANADIUM IN PLANT MATERIALS TABLE II EFFECT OF SOME COMMON ANIONS ON THE ABSORBANCE OF THE VANADIUM - PAR - CRYSTAL VIOLET COMPLEX 713 Anion added - F- c1- Br- I- Br0,- NO,- soda- Molar ratio of anion to vanadium 5 50 100 5 100 2000 5 100 5 5 5 50 5 100 2000 - Absorbance a t 585 nm +- Blank 0.100 0.100 0.125 0.145 0*105 0.170 0.225 0.115 0.170 0.770 0.120 0.270 0.860 0.100 0.120 0,120 Vanad;um - PAR - Cry- stal Violet 0.550 0.550 0.555 0.545 0.545 0.510 0.485 0.565 0.630 0.430 0.530 0.450 0- 150 0.540 0.540 0.530 Anion added c10,- SCN- CHaCO Ca04a- C,H,0,8- C4H40eB- EDTA DCTA* Molar ratio of anion to vanadium 2 5 5 100 2000 6 50 100 5 50 100 5 60 100 5 100 5 100 Absorbance at 585 nm r-- Blank 1.400 1.900 0.100 0.110 0.170 0.100 0.145 0.150 0.126 0.145 0-170 0.096 0.1 10 0.140 0.100 0.100 0-105 0.160 Vanadium - PAR - Cry- stal Violet - - 0-540 0.560 0.500 0.660 0.495 0.490 0.535 0.525 0.510 0.566 0.560 0.545 0.420 0.060 0.516 0.520 * 1,2-Diaminocyclohexane-NNN”’-tetraacetic acid.Determination of Vanadium in Plant Material Extraction of Vanadium from the Plant Material After wet ashing of the plant material it is necessary to separate the trace amounts of vanadium from micro-amounts of other elements that are present in the solution.The ex- traction of vanadium, in the form of its complex with N-benzoyl-N-phenylhydroxylamine (BPHA), from a solution that was 4 N in hydrochloric acidlo was used for this purpose. This method, devised for vanadium in the concentration range 20-500 pg m1-l, was not suitable for amounts of vanadium of a few micrograms. In the presence of iron, which is always present in plant material, it was impossible to extract such small amounts of vanadium quantitatively. It was found necessary to carry out the extraction from a highly oxidising solution [containing cerium(1V) sulphate], immediately after adjustment of the concentration of hydrochloric acid to 4 N (using 6 N hydrochloric acid).From a solution that is 4 N in hydrochloric acid, BPHA can completely extract niobium, hafnium, tungsten and molybdenum and can partially extract zirconium, tantalum and titanium. Zirconium, molybdenum and titanium may be present in plant materials in concentrations comparable with that of vanadium. As can be seen from the literature, and from our own experiments, these elements do not influence the determination of vanadium using the method devised. In order to return the vanadium into aqueous solution three procedures were examined : a re-extraction with nitric acid, sodium hydroxide and hydrogen peroxide solutions ; wet ashing with nitric acid - perchloric acid; and dry ashing after evaporation of the solvent. The last method gave the best results and proceeded rapidly, without additional reagents.The vanadium(V) oxide obtained could be dissolved easily in a small volume of sodium hydroxide solution. Procedure Powder the sample of dried plant material and weigh an amount containing 1-6pg of vanadium. Apply the wet-ashing procedure of Minczewski et aE. ,11 using Bethge’s apparatus, with 4 ml of concentrated nitric acid and 2 ml of 60 per cent. perchloric acid per 1 g of sample. When the wet ashing is complete, boil off the acids until the residue in the flasl, is nearly dry. Dissolve the residue in a small amount of water and filter off any undissolved material. Add 0.2 ml of 0.1 per cent. cerium(1V) sulphate solution to the sample solution and leave714 MINCZEWSKI et al. : SPECTROPHOTOMETRIC DETERMINATION Analyst, VOZ.100 for 5 min. Then add sufficient 6 N hydrochloric acid to adjust the concentration of hydro- chloric acid in the solution to 4 N. Immediately add to the solution two 5-ml portions of a 0.1 per cent. solution of N-benzoyl-N-phenylhydroxylamine in chloroform, each time shaking for 1 min. Combine the extracts and wash them with 5 ml of 6 N hydrochloric acid. Transfer the solution into a quartz crucible, evaporate the solvent on a water-bath and then gently heat the crucible with a burner in order to destroy the complex, finally heating vigorously to ash the residue. After cooling the crucible, add 0.5 ml of 0.1 N sodium hydroxide solution, then 6 ml of water, and heat the solution gently. After cooling again, add 1 ml of PAR solution, adjust to pH 5 with 0.1 N sulphuric acid and transfer the solution quantitatively to the separator.Add 2 ml of phosphate buffer solution, 1 ml of the Crystal Violet solution and 8 ml of benzene - isobutyl methyl ketone (3 + 2). Shake the mixture for 2 min, separate the organic phase quantitatively and transfer it to a 10-ml calibrated flask containing 1.5 ml of ethanol. Make up to the mark with the extraction solvent, mix well and immediately measure the absorbance at wavelength 585nm, using the extraction solvent as a blank. Carry out the blank experiment in parallel. Prepare a calibration graph for the concentration range 0.1-0-5 pg ml-l of vanadium, including vanadium taken through the procedure. The solutions used for calibration should contain a concentration of iron similar to that in the sample of plant material.Results and Conclusion By using the procedure described above vanadium was determined in four types of dry plant material, viz., in serradella (Or.nitho$zls), rye (Secale) straw, rye grain and trefoil (Trifolizcm). Five-gram samples were used. For serradella and rye straw, the full procedure was checked by addition of vanadium standard solution containing 0 4 p g of vanadium. The results are given in Tables I11 and IV. As can be seen from Table 111, the recovery of vanadium is good (94-96 per cent.) in spite of the rather complicated procedure. TABLE I11 RECOVERY OF VANADIUM FROM PLANT MATERIAL The results given are the means of six determinations. Vanadium Vanadium Recovery, Plant material added/pg found/pg per cent.Serradella . . .. .. 0-50 0.48 96.0 Ryestraw .. .. .. 0.50 0-47 94.0 For vanadium concentrations in the plant material in the range 0.13-0.25 p.p.m. the relative standard deviation is equal to 8-15 per cent. For concentrations below 0.1 p.p.m. the relative standard deviation is higher, i.e., 24 per cent. (Table IV). Taking into account the rather complicated procedure the results can be considered to be acceptable. TABLE IV STATISTICAL EVALUATION OF THE RESULTS Concentration Standard Relative Plant material determinations p.p.m. p.p.m. per cent. Number of of vanadium, deviation, standard deviation, Serradella .. .. 6 Rye grain . . .. 6 Trefoil . . .. .. 5 Ryestraw .. .. 6 0.15 0.016 10.0 0-13 0.019 14.6 0-05 0.012 24.0 0.25 0.022 8.8 The method described has been used to determine vanadium in the concentration range 0-05-0-5 p.p.m.Sensitive methods such as this are not usually specific and the pro- cedure for the determination is rather complicated. It is necessary to observe the specified conditions, the reagent concentrations and the composition of the organic solvent. It is especially important in this method because of the blank absorbance, which is constant only under specified conditions. In spite of the rather high value of the blank absorbance, which is about 0.1, the level of absorbance of the vanadium complex is considerably higher than thatOctober, 1975 OF TRACE AMOUNTS OF VANADIUM IN PLANT MATERIALS 715 of the blank (molar absorptivity of the complex is equal to 1.1 x lo5 1 mol-1 cm-1). From Table IV it can be seen that the error of the method is typical for spectrophotometric methods for the determination of trace elements. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. References Pilipenko, A. T., and Tananaiko, M. M., Zh. Analit. Khim., 1973, 28, 746. Siroki, M., and Diordjevic, C., Analytica Chim. Acta, 1971, 57, 301. Karpova, 0. I., Lukachina, W. W., and Pilipenko, A. T., UKr. Khim. Zh., 1973, 39, 194. Babko, A, K., Volkova, A. I., and Getman, T. E., Zh. Neovg. Khim., 1966, 11, 374. Babenko, N. L., Busev, A. I., and Simakova, L. K., Zh. Analit. Khim., 1970, 25, 1639. Bethge, P. O., Analytica Chim. Acta, 1954, 10, 317. Asmus, E., 2. Analyt. Chem., 1960, 178, 104. Schiller, K. S., and Thilo, E., 2. Anorg. Allg. Chem., 1961, 310, 261. Dyrssen, D., and Sekine, T., J . Inovg. Nucl. Chern., 1964, 26, 981. Vita, 0. A., Levier, W. A., and Litteral, E., Analytica Chim. Acta, 1968, 42, 87. Minczewski, J., Chwastowska, J., and Pham thi Hong Mai, Chemia Analit., 1973. 18, 1189. Received January 29th, 1976 Accepted June llfh, 1976

ISSN:0003-2654    

DOI:10.1039/AN9750000708

出版商:RSC    

年代:1975

数据来源: RSC

摘要:

716 Analyst, October, 1975, Vol. 100, pp. 716-720 Elimination of Interference from Aluminium in the Determination of Total Iron in Soils and Plant Materials Using 1,lO-Phenanthroline Reagent T. C. Z. Jayman, S. Sivasubramaniam and M. A. Wijedasa Tea Reseurch Institwte, Agricultural Chemistry Department, St. Coombs, TaZawakeEEe, S7i Lalaka Aluminium, in solution, enhances the iron - 1,lO-phenanthroline colour, leading to high results in the determination of iron. Both the iron and aluminium complexes of phenanthroline exhibit identical absorption charac- teristics. Attempts to mask the aluminium in solution with sodium fluoride have been unsuccessful as the fluoride ions suppress the colour formed with iron and the reagent. The determination of iron after the separation of aluminium and phosphates is simple and rapid.The method presented is reliable and recoveries are quantitative. There are several methods for the determination of iron in soils and plant materials, The method commonly adopted is that involving the use of 1,lO-phenanthroline. However, there appears to be considerable uncertainty when this method is used to determine iron in biological materials. Sandelll lists a number of ions that interfere in this method and indicates that pyrophosphates could cause serious errors. However, he also indicates that one of the main advantages of the method is that it can be used in a weakly acidic medium, so that the hydroxides and phosphates of many metals are not precipitated. He also discusses the degree to which interfering ions can be tolerated and suggests means of overcoming such interferences.Many workers have drawn attention to low recoveries of iron as a result of the interference caused by aluminium and phosphates. Cowling and Benne2 have shown that if a relatively large amount of aluminium is present together with phosphates, low results are obtained for iron because iron is co-precipitated with aluminium phosphate, which precipitates in a weakly acidic medium. They overcame this interference by adding citrate before the adjustment of pH with sodium acetate. However, there has been no reference in the literature to obtaining high results for iron as a result of a positive interference from the presence of aluminium in the solution. That this last effect occurs was established during the course of an investigation in which tea soils, originally known to be high in aluminium, had to be analysed to determine total iron and phosphates.Yamamura and Sikes3 have presented a method with which they claim that in situ masking using a double masking reagent is effective in determining iron in the presence of a 1000-fold increase in the concentration of aluminium. The authors decided to examine the method of Yamamura and Sikes3 and it was found, for reasons that will be discussed below, that their masking agent was ineffective in preventing the interference caused by aluminium in solution. It was therefore decided to examine the original method and effect the necessary modifications. The results of the investigation are reported and a modified method is presented. Reagents Dissolve 20.75 g of analytical-reagent grade sodium acetate in about 50 ml of glass-distilled water in a beaker, add 30 ml of glacial acetic acid, mix well and transfer the mixture quantitatively into a 250-ml calibrated flask, washing the beaker with several small volumes of distilled water.Then stopper the flask and mix the contents well. The pH of this solution should be 4.0. Hydyoquinone (quinol) solution, 1 per cent. m/V. Dissolve 1 g of hydroquinone in a little water in a beaker, heating gently to dissolve the solid, transfer the solution quantitatively into a 100-ml calibrated flask and dilute to the mark with distilled water, Stopper the flask and mix the contents well. The reagent should be prepared freshly and discarded if there is any slight discoloration. Experimental Sodium acetate bufer solution.JAYMAN, SIVASUBRAMANIAM AND WI JEDASA 717 1 ,lO-Pkenanthroline reagent, 0.50per cent.m/V. Dissolve 0-50 g of the reagent in about 40 ml of distilled water in a beaker, warming gently to effect dissolution, and transfer the solution quantitatively into a 100-ml calibrated flask. Dilute to the mark with distilled water, stopper the flask and mix well. If the solid tends to recrystallise on standing, warm the flask in order to redissolve the reagent before use. Ammonia solution, sj5. gr. 0.92. Analytical-reagent grade. Ammunia solution, 4 N. Bromocresol purple indicator soldon. Standard iron sohution. Dilute the analytical-reagent grade ammonia solution. Dissolve 0.1000 g of pure iron wire in 15 ml of 0.6 N hydrochloric acid plus 2 ml of concentrated nitric acid and when it has completely dissolved transfer the solution quantitatively into a 1-1 calibrated flask, adjusting the volume to the mark.1 ml of solution = 100 pg of iron. Methods Calibration of Standard Graph Transfer, by pipette, into 25-ml calibrated flasks aliquots of solutions containing from 0 to 9Opg of iron; adjust the volumes to approximately 15ml with distilled water. Add, in order, 5 ml of the sodium acetate buffer solution, 2 ml of the hydroquinone solution and, finally, 1 ml of the phenanthroline reagent. Mix the solutions and dilute to the mark with distilled water, stopper the flasks and mix again. A reference blank is prepared in the same manner except that the iron is omitted.By use of a colorimeter read the values of the colours produced against the reference blank in 2-cm cells at a wavelength of 490 nm. The resulting graph of instrument reading versus concentration of iron should be linear, passing through the origin. Procedure for Separation of Iron from Interfering Ions To centrifuge tubes marked with a 50-ml calibration, transfer, by pipette, suitable portions (say 10-20 ml) of solutions from plant materials or soils. Add 10 drops of bromocresol purple indicator to each and place the tubes in a boiling water bath for 3 min. Next add ammonia solution, sp. gr. 0.92, dropwise until precipitation just begins. Thereafter, add 4 N ammonia solution dropwise until the indicator turns purple; then add a further 2 ml of the 4 N ammonia solution.Stir the solution well with an air jet and place the tubes in a boiling water bath for 5 min in order to complete the precipitation. When the solution has cooled, adjust the volume to 50 ml, stir it well with the air jet and centrifuge at 200 rev min-1 for 5 min. To the residue in the tube add 3 ml of hot, 6 N hydrochloric acid and place the tubes in a boiling water bath for 2 min. Thereafter add 10 ml of hot, 25 per cent. m/V sodium hydroxide solution and again place the tubes in a boiling water bath for 5 min. After cooling, adjust the volume of solution to 50 ml with distilled water, stir well with the air jet and centrifuge at 2000rev min-l for 5 min. Then discard the supernatant solution. To the residues in the centrifuge tubes add 3 ml of hot, 6 N hydrochloric acid and warm them in a boiling water bath for 2 min in order to effect complete dissolution.Transfer the contents quantitatively into 50-ml calibrated flasks, washing the tubes with several small por- tions of water. Next, make the contents of the flasks up to the mark with distilled water, stopper and mix well. The solutions in the 50-ml calibrated flasks contain iron free from interfering ions. Transfer, by pipette, portions containing 0-90 pg of iron from the separated solutions into 25-ml calibrated flasks, develop the iron colour and proceed as described under Calibration of Standard Graph. Any suitable colorimeter can be used. Discard the supernatant liquid. Sample Preparation Ashing of plant samples Weigh 0.20 g of finely ground sample into a number of glass ignition tubes held in place with a stainless-steel tray and leave them overnight in a muffle furnace at 450 "C.Cool, add a few drops of distilled water to the tubes in order to moisten the ash, then add by pipette 2 ml of the digestion mixture, nitric acid - hydrochloric acid - water (25 + 25 + 50 V/V), and evaporate the mixture to dryness on a hot-plate. Add exactly 10 ml of 0.05 N hydro- chloric acid, warm the tubes so as to dissolve the contents, stopper and shake them to mix the718 solution well. suitable portions for the determination of iron. Acid digestion of soil samples To 2-5 g of a soil sample placed in the digestion tubes, 10 ml of 10: 1 digestion mixture (100 ml of 60 per cent. perchloric acid mixed with 10 ml of concentrated sulphuric acid) are added.Several such tubes are slowly heated in an orthophosphoric acid bath until the perchloric acid has boiled off and only about 1-2 ml of solution remain. If the samples are not digested to a clear white residue, a few inore millilitres of the digestion mixture are added and the digestion is continued. When the tubes have cooled they are removed from the bath and the digested samples are diluted to about 15 ml with distilled water. The solutions are then filtered into 250-ml calibrated flasks using filter-paper No. 542. The residues in the digestion tubes are washed repeatedly with hot distilled water and the washings passed through the filter-papers into the flasks. Finally, the soil extracts in the flasks are diluted to the mark with distilled water, mixed well and suitable aliquots used for the determination of iron.The method is that described by Jack~on.~ Interference from Aluminium The calibrated standards obeyed Beer’s law over the range 0-90pg of iron per 25ml. When aluminium was added, however, there was a definite enhancement in colour that resulted in positive error. The results are shown in Table I. JAYMAN et al. : ELIMINATION OF INTERFERENCE FROM ALUMINIUM Analyst, vol. 100 Leave the tubes aside for 1 h so as to allow any silica to settle, then use Results and Discussion TABLE I INTERFERENCE FROM ALUMINIUM WHEN ADDED TO SAMPLES BEING ANALYSED TO DETERMINE IRON Iron present/ Aluminium added/ Iron found/ pg per 25 ml pg per 25 ml pg per 25 ml 50.0 Nil 50.0 50.0 500 52.0 50.0 1000 57.0 50.0 2000 63.0 60.0 4000 7 6.0 In order to find out whether aluminium in solution also formed a complex with phenanthro- line reagent, aluminium standards containing from 200 to 8000 pg per 25 ml were developed alone in the same manner as the iron standards and the colours were measured against a blank solution at a wavelength of 490 nm in 2-cm cells.The results obtained are shown in Table 11. TABLE I1 &LOUR DEVELOPMENT OF ALUMINIUM STANDARDS WITH PHENANTHROLINE Aluminium Apparent iron Aluminium Apparent iron present/ concentration/ present/ concentration/ pg per 25 ml pg per 25 ml pg per 25 ml pg per 25 ml 200 3.0 2000 13.0 400 4.0 4000 24.0 800 5.0 8000 63-0 1000 7.0 Colours developed with phenanthroline by iron and aluminium standards were read against a blank at various wavelengths in order to study their absorption characteristics. The absorption spectra are shown in Fig.1. It can be seen from Fig. 1 that both the iron and aluminium complexes of phenanthroline exhibited identical absorption characteristics under the conditions of colour development, the only difference being that the iron-phenanthroline complex was more sensitive than the aluminium - phenanthroline complex. That being so, it is evident that the presence of aluminium in the system will cause a positive error in the determination of iron. Efforts to suppress the interference from aluminium using sodium fluoride were unsuccessful because the fluoride ions markedly suppressed the colour formed by iron with phenanthroline. The results are shown in Table 111.October, 1975 IN THE DETERMINATION OF TOTAL IRON IN SOILS AND PLANTS 719 425 445 475 490 505 525 555 Wavelengthhm Fig.1. Absorption spectra of iron and aluminium complexes of phenanthroline a t pH = 4. A, iron(I1) - phenanthroline complex ; B, aluminium - phenanthro- line complex. TABLE I11 EFFECT OF DIFFERENT VOLUMES OF SODIUM FLUORIDE SOLUTION (5 per cent. mfY) ON IRON - PHENANTHROLINE COLOURS Volume of Iron present/ sodium fluoride Colorirneter p g per 25 ml solution/ml reading 50.0 Nil 48.0 50.0 0.50 44.0 50.0 1.00 33.0 60.0 2.00 22.0 50.0 4.00 7.0 Apparent iron concentration/ pg per 25 ml 45.0 41.5 31.0 21.0 6-6 Suppression of Aluminium Xnterference by the Method of Yamamura and Sikes3 The colours from pure aluminium standards were developed by the method described by Yamamura and Sikes3 and the absorbances of the solutions were measured in 2-cm cells at 507 nm.TABLE IV INEFFECTIVENESS OF THE DOUBLE MASKING REAGENT IN PREVENTING INTERFERENCE FROM The results obtained are given in Table IV. ALUMINIUM IN SOLUTION Colorimeter Apparent iron A1 presentlpg reading in 2-cm concentration/ cells a t 507 nm pg per 25 ml 1000 15-0 14-0 2000 21.0 20.0 3000 26.0 25.0 4000 32-0 31.0 5000 39-0 38.0 It can be seen from the results in Table IT' that despite the presence of the double masking reagent, the aluminium formed a complex with phenanthroline and the absorbance readings of the colour given by the aluminium - phenanthroline complex increased linearly with the increase in aluminium concentration.This proportionality indicated that the aluminium - phenanthroline complex was much stronger and more stable than either the aluminium - citrate or aluminium - EDTA complexes under the conditions of colour development. Because the masking agents that were tried failed to achieve the desired results, it was decided to separate iron from all other interfering ions by use of the proposed method. The results obtained for iron in soil and tea plant foliage by using the proposed procedure are presented in Tables V and VI. per 25 ml720 JAYMAN, SIVASUBRAMANIAM AND WI JEDASA TABLE V TOTAL IRON IN SOIL WITH AND WITHOUT SEPARATION OF ALUMINIUM Soil after separation of A1 and P, p.p.m. p.p.m. Fe + added A1 Fe A1 P Fe in original soil Original soil (without separation), A 1 A I \ equivalent to A1 100000 50000 10000 81 250 100 000 TABLE VI IRON IN TEA LEAVES BY THE EXISTING AND PROPOSED METHODS Iron by atomic- Aluminium Iron b y existing Iron by proposed absorption present, p.p.m.method, p.p.m. method, p.p.m. method, p.p.m. 600 200.0 160.0 155.0 1700 320.0 200.0 205-0 2400 350.0 290.0 290.0 4200 485.0 365.0 362.0 The results given in Tables V and VI clearly illustrate the enhancing effect of aluminium, the amount of which was determined by the method of Jayman and Sivasubramaniam,s on the iron values. Once the aluminium in solution was eliminated, the iron values obtained agreed well with the results obtained by atomic-absorption spectrophotometry. In the soil samples (Table V), when the calculated amount of the aluminium contained in the original soil was added to an aliquot of the extract from which aluminium had been separated, the value obtained for iron agreed with the apparent iron content of the unseparated soil extract, illustrating clearly the interference caused by aluminium. The modified method is simple and, with a little practice, the separations can be carried out rapidly and quantitatively, enabling large numbers of samples to be analysed in one working day. The authors thank Mr. P. W. Uduwawala for typing the manuscript and the Director of the Tea Research Institute for permission to publish this paper. References 1. 2. 3. 4. 6. Sandell, E. B., “Chemical Analysis,” Volume 3, “Colorimetric Determination of Traces of Metals,” Cowling, H., and Benne, E. J., J. Ass. Off. Agric. Chem., 1942, 25, 555. Yamamura, S. S., and Sikes, J. H., Analyt. Chem., 1966, 38, 793. Jackson, M. L., “Soil Chemical Analysis,” Constable & Go. Ltd., London, 1962. fayman, T. C. Z., and Sivasubramaniam, S., Analyst, 1974, 99, 296. Second Edition, Interscience Publishers Tnc., New York and London, 1950. Received February 18th, 1976 Accepted April 16th, 1976

ISSN:0003-2654    

DOI:10.1039/AN9750000716

出版商:RSC    

年代:1975

数据来源: RSC

摘要:

AIzalyst, October, 1975, Vol. 100, pp. 721-725 721 Atomic-absorption Determination of Strontium in Silicate Rocks: A Study of Major Element Interferences in the Nitrous Oxide - Acetylene Flame D. Carter, J. G. T. Regan and J. Warren Departmefit of Industry, Laboratory of the Government Chemist, Cornwall House, Stamford Street, London, SEI 9NQ A study has been made of the major element interferences associated with the atomic-absorption determination of trace amounts of strontium in silicate rocks by using a nitrous oxide - acetylene flame. Aluminium causes sup- pression of the strontium signal, while calcium and magnesium act as partial releasing agents, thus reducing the effect of the aluminium. Exact matching of samples and standards can be avoided by the use of lanthanum, which has been shown to be an effective releasing agent in the nitrous oxide - acety- lene flame.Strontium levels in the United States Geological Survey rocks AGVl, BCRl and GSPI, in the range 244-670 pg g-l, have been determined with a precision of better than 2 per cent. A spiking and recovery experiment has also been carried out. Strontium can be measured in silicate rocks with a limit of determination of 0.6 pg 8-l. The determination of strontium by means of atomic-absorption spectrophotometry can be accomplished by using an air - acetylene flame or, with greater sensitivity, by using a nitrous oxide - acetylene flame. The determination with the air - acetylene flame is subject to a variety of serious interferences, which can be controlled by using lanthanum as a releasing agent.1-3 The higher temperature and greater reducing capacity of the nitrous oxide - acety- lene flame generally serves to reduce a number of inter-element interferences.Magill and Svehla3 investigated the effect of determining strontium in the presence of a number of inter- ferents (using a strontium - interferent ratio of 1 : 6) and found that use of the nitrous oxide - acetylene flame successfully eliminated them. However, Bowman and Willis4 found, when determining strontium in rock solutions , that appreciable interference did occur, even in the nitrous oxide - acetylene flame, and recommended that the method of standard additions be used in order to compensate for variations in the levels of major elements found in different rocks. During recent work5 on the determination of trace elements in silicate rocks we also found that the nitrous oxide - acetylene flame did not completely eliminate the interferences with strontium, even when the standards used contained a background matrix that was similar to, but not identical with, that of the rock sample.As the standard additions method suggested by Bowman and Willis is time consuming it was decided to study the effects that the varying amounts of major components present in silicate rock (aluminium, calcium, iron and magnesium) had on the strontium signal, and to determine whether lanthanum would prove to be effective in overcoming the interferences. Experimental Apparatus A Perkin-Elmer 403 atomic-absorption spectrophotometer, fitted with a single-slot nitrous oxide - acetylene burner head, was used with a Telsec chart recorder of 10-mV full-scale deflection.A low-pressure PTFE bomb from Perkin-Elmer Ltd. was also used. Reagents The following Specpure materials (Johnson Matthey Chemicals Ltd.) were used : aluminium rod, iron sponge, magnesium crystals, calcium carbonate, strontium nitrate, potassium chloride Crown Copyright.722 Analyst, VoZ. 100 and sodium chloride. The other reagents were of analytical-reagent grade : boric acid, hydro- chloric acid (sp. gr. 1-16), hydrofluoric acid (40 per cent.), nitric acid (sp. gr. 1.42), perchloric acid (60 per cent.) and lanthanum chloride solution (10 per cent. m/V lanthanum content} for atomic-absorption purposes. CARTER et al. : INTERFERENCES IN THE ATOMIC-ABSORPTION Dissolution of Rock The dissolution procedure used in this study has already been reported5 in connectionwith an atomic-absorption method for the determination of trace amounts of copper, vanadium, chromium, nickel, cobalt and barium in silicate rocks.Briefly, it consists in digesting the rock with a hydrofluoric - nitric - perchloric acid mixture at 100 "C in a low-pressure PTFE bomb, followed by evaporation until perchloric acid fumes appear and final dissolution in a mixture of hydrochloric and boric acids. A suitable amount of potassium chloride (ionisation buffer) is then added to the mixture, which is diluted with distilled water to give a solution containing 1 per cent. of rock (original mass). Study of Interferences Potential interferences were monitored by observing whether the signal produced from a standard strontium solution was influenced by the presence of the major element under investigation.The concentration of strontium used throughout this study was 8 pg ml-1, which is equivalent to 800 pg g-l in the original rock. The concentration ranges of the major elements studied were chosen to be similar to those normally found in silicate rocks. All of the interference studies were carried out on solutions that were also 0.1 per cent. in potassium, 0-8 per cent. in boric acid, 2 per cent. in perchloric acid and 5 per cent. in hydrochloric acid, so that the results obtained in this study would be directly applicable to our previously reported5 atomic-absorption procedure for rock analysis. Efect of aluminium, calcium, magnesium and iron on the strontium signal The effects that aluminium, calcium, magnesium and iron have separately on strontium were determined as follows.Aluminium on strontium. Six solutions that contained 8 pg ml-l of strontium were pre- pared so as to contain 0, 159, 397, 794, 1191 and 1588 pg ml-l of aluminium. These levels correspond to 0, 3.0, 7-5, 15.0, 22.5 and 30.0 per cent. of aluminium oxide in the original rock. Calcium on strontium. Six solutions that contained 8 pg ml-l of strontium were prepared so as to contain 0, 179, 357, 536, 714 and 1071 pg ml-l of calcium. These levels correspond to 0, 2-5, 5.0, 7.5, 10.0 and 15-0 per cent. of calcium oxide in the original rock. Magnesium on strontium. Five solutions that contained 8 pg ml-l of strontium were pre- pared so as to contain 0, 121, 241, 482 and 723 pg ml-1 of magnesium.These levels corres- pond to 0, 2.0,4*0, 8.0 and 12.0 per cent. of magnesium oxide in the original rock. Five solutions that contained 8 pg ml-l of strontium were prepared so as to contain 0, 280, 560, 840 and 1400pgml-l of iron. These levels correspond to 0, 4.0, 8-0, 12.0 and 20.0 per cent. of iron oxide (total iron expressed as Fe,O,) in the original rock. Iron on strontium. Efects of calcium, magnesium and iron on strontium in the presence of aluminium The effects that calcium, magnesium and iron have separately on strontium in the presence of aluminium were determined on solutions containing 8 pg ml-l of strontium and 794pg ml-1 of aluminium. These levels correspond to 800 pg g1 of strontium and 15 per cent.of alu- minium oxide in the original rock. The concentrations of calcium, magnesium and iron used in this experiment were the same as in the investigations of the effects of aluminium, calcium, magnesium and iron alone. Eflect of lanthanum The above two interference experiments were repeated with duplicate solutions containing 1 per cent. m/V of lanthanum, so that the effectiveness of lanthanum as a releasing agent in the nitrous oxide - acetylene flame could be observed.October, 1975 DETERMINATION OF STRONTIUM I N SILICATE ROCKS Results and Discussion Efects of aluminizcm, calcium, magnesium and iron on the strontium signal 723 Aluminium on strontium. The influence that the aluminium concentration has on the strontium signal is shown in Fig.1. It can be seen from line A that while a solution of 8-0 pg ml-1 of strontium (800 pg g1 in the rock) gave an absorbance of 0.420, the addition of aluminium decreased the signal to give an absorbance of 0.180 at a level of 1588 pg ml-l of aluminium (30 per cent. of aluminium oxide in rock). The effect of repeating the atomic- absorption measurements on duplicate solutions containing 1 per cent. of lanthanum (as described under Efect of lanthanum) is shown in Fig. 1, line B, and illustrates the effectiveness of lanthanum as a releasing agent. A slight decrease in absorbance (from 0.420 to 0.410) was observed in the range corresponding to 22.5-30-0 per cent. of aluminium oxide in the original rock. However, this decrease was completely eliminated when solutions containing 2 per cent.of lanthanum were used. 0 5 10 15 20 25 30 Aluminium oxide in rock, per cent. Fig. 1. The effect of aluminium on the absorbance of strontium (800 pg g-l in rock): A, aluminium; and B, aluminium with 1 per cent. of lanthanum. Efects of calcium, magnesium and iron on strontium. The presence of these elements had no effect on the strontium signal over the ranges investigated in this work. The addition of 1 per cent. of lanthanum to the solutions also had no effect, and the measured absorbance of 0-420 indicated that no interference occurred. Efects of calcium, magnesium and iron on strontium in the presence of aluminium Because the experiment to determine the effects of aluminium, calcium, magnesium and iron separately indicated that of the elements investigated only aluminium interfered directly with strontium, it was decided to study separately the effect that calcium, magnesium and iron had on the aluminium plus strontium system so as to determine if thcy had any indirect effect on the interference mechanism. The influence of 1 per cent.of lanthanum on this system was also studied, as described under Efect of lanthanum. The results of the experiment to determine the effect of calcium, magnesium and iron in the presence of aluminium are shown graphically in Fig. 2, and from line A it can be seen that iron does not influence the interference of aluminium in the determination of strontium. Calcium and magnesium do, however, cause a reduction in the interference effect by acting as releasing agents, and lines B and C indicate that the releasing action is greater for calcium than magnesium.The addition of 1 per cent. of lanthanum to the iron, calcium and mag- nesium systems completely eliminates the interference, as is shown by line D, which represents an absorbance of 0420. Possible interference mechanism It is possible that in the flame, aluminium salts are converted into aluminate, which combines with the strontium2 in a manner similar to that of other oxyanions, such as phosphate and silicate, perhaps giving rise to mixed strontium - aluminium oxides, which would inhibit the formation of ground-state strontium atoms.724 Analyst, VoZ. 100 Magnesium, calcium and lanthanum would seem to combine with aluminium in a manner similar to that of strontium, thus acting as releasing agents for strontium.The order of effective- ness as releasing agents was found to be lanthanum > calcium > magnesium. This observed effectiveness is probably due to the relative affinity of the elements for aluminium. Increasing the concentration of the releasing elements also increases their effectiveness, as would be expected from the Law of Mass Action. However, it appears, from Fig. 2, that neither calcium nor magnesium at acceptable levels would effect a complete release of strontium. CARTER et aE. : INTERFERENCES IN THE ATOMIC-ABSORPTION 0.1 I 1 I I 0 5 10 15 20 Metal oxide in rock, per cent. Fig. 2. The effect of iron, magnesium and calcium on the absorb- ance of strontium (800 pg g-l in rock) in the presence of aluminium (16 per cent.of aluminium oxide in rock): A, iron; B, magnesium; C, calcium; and D, iron, magnesium and calcium with 1 per cent. of lanthanum. DetermiNation of strontium in silicate rocks The previous experiments indicate that the effect on strontium of the major elements normally present in silicate rocks can be overcome by the addition of 1 per cent. of lanthanum to the analytical spray solution. A high aluminium content rock containing aluminium oxide at a level of 22-5-30 per cent. should be analysed in the presence of 2 per cent. of lanthanum. Therefore, it was decided to determine the strontium content of some silicate rocks by using a previously reported5 atomic-absorption method with the modification of adding lanthanum chloride to the rock solutions and standards to give a final concentration of 1 per cent.m/V of lanthanum. It was found that these solutions had a shelf-life of approxi- mately 3 d before precipitation of lanthanum fluoride occurred. Because of this the lanthanum was added just prior to the atomic-absorption determination. An attempt was made to assess the accuracy and precision of this modified atomic-absorp- tion method by analysing the United States Geological Survey rocks AGVI, BCRI and GSPI. From the results shown in Table I it can be seen that the precision is approximately 1-2 per cent. and that there is reasonable agreement between the atomic-absorption values and the reported literature v a l ~ e s . ~ , ~ However, it can also be seen that the individual litera- ture values cover a wide range, and this must influence the credibility of the literature averages. Because of this drawback it was decided to obtain a further indication of the accuracy by carrying out some recovery experiments on a rock that had been spiked with 100 pg g-1 of strontium.The rock used was a sample of olivine-dolerite that contained 17.3 per cent. TABLE I ATOMIC-ABSORPTION DETERMINATION OF STRONTIUM IN UNITED STATES GEOLOGICAL SURVEY ROCKS Coefficient of Mean of variation of Literature Literature atomic-absorption atomic-absorption Number of Rock average/pg g-l rangelpg g-l resultslpg g-' results, per cent. determinations AGV-1 657 348-1050 670 0.75 9 BCR-1 330 244-525 344 1.6 10 GSP-1 233 148-400 244 1.6 15October, 1975 DETERMINATION OF STRONTIUM I N SILICATE ROCKS 725 of aluminium oxide, 10.8 per cent, of iron oxide (total iron as Fe,O,), 9.0 per cent.of mag- nesium oxide and 10.7 per cent. of calcium oxide. The spiking was carried out in a PTFE bomb by adding 20ml of 10pgml-1 strontium solution to 2 g of rock. The contents of the bomb were evaporated to dryness before the dissolution procedure was commenced. Five replicate portions of spiked and unspiked rock were analysed and from the results shown in Table I1 it can be seen that the recoveries determined in this way, and at these levels, varied from 98.2 to 101.6 per cent. TABLE I1 SPIKING EXPERIMENT WITH 100 pg g-l OF STRONTIUM Strontium concentration in unspiked rock/ 191.6 190.5 192.8 190.5 189-3 Ctg g-' Strontium concentration in spiked rock/ Recovery, Ctg g-l per cent. 291.0 99.4 291.0 100.5 291.0 98.2 292.1 101.6 289.5 100.2 The limit of determination of strontium in silicate rocks by use of this atomic-absorption The authors thank the Government Chemist for permission to publish this paper. method is 0.6 pg gf. References 1. 2. 3. 4. 6. 6. 7. Belcher. C. B., and Brooks, K. A., Analytica Chim. Acta, 1963, 29, 202. Trent, D., and Slavin, W., Atom. Absorption Newsl., 1964, 22, 63. Magill, W. A., and Svehla, G., 2. Analyt. Ckem., 1974, 270, 177. Bowman. J. A., and Willis, J. B., Analyt. Chem., 1967, 39, 1210. Warren, J., and Carter, D., Calz. J . Spectrosc., 1976, 20, 1. Flanagan, F. J., Geochim. Cosmochirn. Acta, 1969, 33, 81. Flanagan, F. J., Geochim. Cosmochim. Acta, 1973, 37, 1189. Received February 19th. 1976 Accepted May 19tk, 1976

ISSN:0003-2654    

DOI:10.1039/AN9750000721

出版商:RSC    

年代:1975

数据来源: RSC