摘要:

ANALYST, MARCH 1986, VOL. 111 305 Extraction = Spectrophotometric Determination of Cadmium Yadvendra K. Agrawal and Tushar A. Desai Analytical Chemistry Laboratory, Department of Pharmacy, Faculty of Technology and Engineering, M.S. University of Baroda, Kalabhavan, Baroda-390 00 I , India A simple and sensitive extraction - spectrophotometric method for the determination of cadmium is described. The binary complex formed between cadmium and N-phenylcinnamohydroxamic acid (PCHA) is extracted with chloroform at pH 9.5, and shows maximum absorbance at 380 nm with a molar absorptivity of 3.6 x lo3 I mol-1 cm-1. The optimum concentration range for measurements is 1.41-1 1.25 p.p.m. The sensitivity of the method was increased by the addition of 4-(2-pyridylazo)resorcinol (PAR) after extracting cadmium with PCHA at pH 9.5.The Cd - PCHA - PAR complex is reddish orange and shows maximum absorbance at 510 nm with a molar absorptivity of 4.8 x lo4 I mol-l cm-I. It obeys Beer's law over the range 0.23-2.25 p.p.m. Various experimental parameters were studied for establishing the optimum conditions for the determination of cadmium. The method had been applied to the determination of cadmium in standard and environmental samples. Keywords: Cadmium determination; N -phen ylcinnamoh ydroxamic acid; 442-p yridylazo)resorcinol; extraction; spectrophotometry The analytical chemistry of cadmium has developed consider- ably in recent years especially owing to its toxicity and ubiquity. High concentrations have been found related to industrial activity in some areas.1 Cd metal itself is not toxic, but most of its compounds have very high toxicity on inhalation.Cadmium and mercury are considered to be two major pollutants for which new and better abatement processes and analytical methods are required. Liquid - liquid extraction is an attractive technique for meeting these challenges.2 Dithizone, Cadion, 4-(2-pyridylazo)resorcinol (PAR) , etc. , have been used for the spectrophotometric determination of cadmium, but the methods are tedious and not selective as many of the common ions interfere."6 Hydroxamic acids have been widely studied and applied as potential analytical reagents for the separation, detection and determination of various metal ions7-9 and have also been employed as possible spectrophotometric reagents for metal ions.10 A few hydroxamic acids have been employed for the extraction and spectrophotometric determination of cad- mium.11-14 Cadmium forms a colourless complex with hydrox- amic acid and spectrophotometric determination is possible utilising some chromogenic reagents. The method presented here provides a simple and precise direct determination of yellow cadmium hydroxamate at 380 nm using N-phenyl- cinnamohydroxamic acid (PCHA) in chloroform. The sensi- tivity of the method was further enhanced by utilising PAR. Moreover, the proposed method also provides a simple and rapid separation of small amounts of cadmium from larger amounts of closely related metal ions. The method has been applied to the determination of cadmium in standard and environmental samples. Experimental Apparatus The spectral measurements were made on VSU 2-P (Carl- Zeiss) spectrophotometer.The pH adjustments were made on Systronics Model 335 digital pH meter, equipped with glass and calomel electrodes. Chemicals and Reagents All the chemicals used were of AnalaR and GR grades from BDH Chemicals and E. Merck, respectively, unless stated otherwise. N- Phenylcinnamohydroxamic acid (PCHA) . Freshly synthesised as described elsewhere15 (m.p. 159 "C; literature16 m.p. , 162 "C). The purity of the reagent was checked by TLC and UV and IR spectroscopy and a 0.10% mlV solution was prepared in chloroform. Standard buffer solutions. Prepared from a mixture of 0.025 M Na2B407 and 0.1 M NaOH solutions (also form a mixture of 0.05 M NaHC03 and 0.1 M NaOH solutions) as described elsewhere.17 Standard cadmium solution, 4.95 x 104 M. Prepared by dissolving 0.3809 g of 3CdS04.8H20 in 1 1 of doubly distilled water. The cadmium content was determined titrimetrically .18 PAR solution 0.1% mlV in 95% ethanol. Extraction Procedure An aliquot of solution containing 5.62 p.p.m. of Cd(I1) (1.0 ml of 4.95 x 10-4 M solution) was transferred into a 60-ml separating funnel and sufficient amounts of distilled water and buffer solution were added to maintain the pH at 9.5 in a total volume of aqueous phase of 10 ml. A 10-ml volume of 0.10% PCHA solution in chloroform was added and the mixture was shaken vigorously for 1&15 min. The phases were allowed to separate and a yellow chloroform layer was collected in a 10-ml calibrated flask after drying over anhydrous sodium sulphate.The absorbance of the yellow chloroform extract was measured at 380 nm against a reagent blank prepared as above but without cadmium. To enhance the sensitivity of the method, PAR solution was added to the Cd - PCHA extract, and the contents were diluted to 25 ml with ethanol. The absorbance was measured at 510 nm against a reagent blank. Results and Discussion Absorption Spectra The yellow Cd - PCHA complex shows maximum absorbance at 380 nm, while the reagent (PCHA) has maximum absor- bance at 290 nm and no absorbance at 380 nm. The Cd - PCHA complex has a molar absorptivity 3.8 X 103 1 mol-1 cm-1. The Cd - PCHA - PAR extract has maximum absorbance at 510 nm with a molar absorptivity of 4.8 x 104 1 mol-l cm-l.The enhanced sensitivity of PCHA as reflected by an increased molar absorptivity is due to the increase in the length of conjugation by introduction of the side-chain double306 ANALYST, MARCH 1986, VOL. 111 bond, -CH=CH- between the R group and the carbon atom of the carbonyl group: R1-N-OH I R&H=CH-C=O where R1 = R2 = phenyl. Effect of pH The optimum pH for the complete extraction of cadmium with PCHA in chloroform was studied over the range 1.0-10.5. It was observed that the extraction commences at pH 6.0 and is maximum at pH 9.0-9.8 (Table 1); above pH 9.8 the extraction decreases. Hence pH 9.5 was chosen for extraction. Effect of PCHA Concentration Extractions were carried out with various amounts of PCHA and it was found that 10 ml of 0.1% PCHA solution in chloroform is adequate for the complete extraction of cadmium.A further excess of PCHA increased the absor- bance of the reagent blank. However, for Cd - PCHA - PAR a large excess of PCHA (15 ml of 0.2% PCHA solution in CHCl3) could be used without any difficulty (Tables 2 and 3). Effect of PAR Concentration Studies with various concentrations of PAR showed that 2.0 ml of a 0.1% solution of PAR in ethanol was adequate for Table 1. Effect of pH on the extraction of the Cd - PCHA complex. Cd(II), 5.62 p.p.m.; PCHA, 10 ml of 0.1% solution in CHCI3; solvent, CHCI,; colour of complex, yellow; A,,,, 380 nm. The extraction ( E ) is given by E(%) = lOOD/(D + Vaq,/Vorg,), and the distribution coefficient (D) is calculated from D = concentration of Cd in organic phase/(total Cd taken - Cd extracted into organic phase) Molar absorptivity/ PH Absorbance 1 mol-l cm-I E , Yo 6.0 7.0 8.0 8.5 9.0 9.5 9.8 10.0 10.5 0.02 0.02 0.02 0.08 0.13 0.18 0.17 0.10 0.06 4.0 x 102 4.0 x lo2 1.6 x 103 2.6 x 103 3.6 x 103 3.4 x 103 2.0 x 103 1.2 x 103 4.0 X lo2 11 11 11 44 72 100 94 55 33 Table 2.Effect of PCHA concentration on the extraction of the Cd - PCHA complex. Cd(II), 5.62 p.p.m.; PCHA, 10 ml of solution in CHCI,; solvent, CHCI,; colour of complex, yellow; A,,,. 380 nm Reagent Molar absorptivity/ concentration, YO Absorbance 1 mol-1 cm-1 E , Yo 0.05 0.15 3.0 x 103 81 0.10 0.18 3.6 x 103 100 0.15 0.13 2.6 x 103 72 0.20 0.10 2.0 x 103 55 Table 3. Effect of PCHA concentration on the extraction of the Cd - PCHA - PAR complex.Cd(II), 1.124 p.p.m.; PCHA, 10 ml of solution in CHCI3; PAR, 2.0 ml of 0.1% solution in EtOH; solvent, CHCI3 - EtOH (1 + 1); colour of complex, reddish orange; h,,,., 510 nm Concentration of Molar absorptivity/ PCHA, YO Absorbance 1 rnol-1 cm-1 E , Yo 0.03 0.30 3.0 x 104 62 0.05 0.38 3.8 x 104 79 0.10 0.44 4.4 x 104 92 0.20 0.48 4.8 x 104 100 0.25 0.39 3.9 x 104 81 complete colour development (Table 4). A further excess of PAR increased the absorbance of the reagent blank. Effect of Shaking Time and Stability The extraction of cadmium complex is rapid under the conditions recommended in the procedure. A shaking time of 10-15 min is sufficient for the complete extraction of cadmium and the binary Cd - PCHA complex is stable for 3-4 h. Effect of Solvents Various solvents were studied for the quantitative extraction of cadmium, viz., chloroform, carbon tetrachloride, toluene, benzene and chlorobenzene. Chloroform was found to be the most suitable (Table 5).Table 4. Effect of PAR concentration on the extraction of the Cd - PCHA - PAR complex. Cd(II), 1.124 p.p.m.; PCHA, 15 ml of 0.2% solution in CHCI,; PAR, 0.1% solution in EtOH; solvent, CHCl3 - EtOH (1 + 1); colour of complex; reddish orange; A,,,,, 510 nm Volume of Molar absorptivity/ PAR solution/ml Absorbance 1 mol-1 cm-1 E , Yo 1 .o 0.25 2.5 x 104 52 1.5 0.37 3.7 x 104 77 2.0 0.48 4.8 x 104 100 2.5 0.34 3.4 x 104 71 3.0 0.24 2.4 x 104 50 ~~ ~ ~ Table 5. Effect of solvents on the extraction of the Cd - PCHA complex. Cd(II), 5.62 p.p.rn.; PCHA, 10 ml of 0.1% solution; colour of complex, yellow; A,,,., 380 nm Molar absorptivity/ Solvent Absorbance 1 mol-1 cm-1 E , Yo Chloroform .. 0.18 3.6 x 103 100 Benzene . . . . 0.13 2.6 x 103 72 Carbon tetra- chloride . . 0.15 3.0 x 103 83 Chlorobenzene 0.10 2.0 x 103 55 Toluene . . . . 0.14 2.8 x 103 78 Table 6. Effect of diverse ions on extraction. Cd - PCHA system (I): Cd(II), 5.62 p.p.m.; PCHA, 10 ml of 0.1% solution in CHC13; solvent, CHCI3; h,,,, 380 nm; absorbance, 0.18. Cd - PCHA - PAR system (11): Cd(II), 1.124 p.p.m.; PCHA, 10 ml of 0.2% solution in CHCI,; PAR, 2.0 ml of 0.1% solution in EtOH; solvent, CHCI3 - EtOH (1 + 1); A, 510 nm; absorbance, 0.48 Absorbance Ion Added as Amount/mg System I System I1 Ba2+ . . BaCI, 40 0.18 0.48 Sr2+ . . SrCI2 50 0.18 0.48 Cd2+ . , CdC12 40 0.18 0.48 50* 0.16 0.47 0.48 Be2+ .. BeSO, 50 0.18 Mg2+ , . MgS04 50 0.18 0.48 0.48 cu2+ . . cuso4 50 0.18 Ni2+ . . NiCI2 50 0.18 0.48 co2+ . . COCI, 50 0.18 0.48 AS^+ . . As203 40 0.18 0.48 Ti4 + . . TiC14 40 0.19 0.48 zI"+ . . Zr(N03)4.5H20 30 0.18 0.48 MO6+ . . (NH4)6M07024 30 0.18 0.48 v5+ . . NH4V03 50 0.18 0.48 Hg2+ . . HgC12 50T 0.17 0.49 Mn2+ . . MnS04 40 * 0.17 0.47 Zn2+ . . ZnSO, 50 0.18 0.48 * Masked with NaF (10 ml of 0.1 M NaF solution). 1- Masked with ascorbic acid (10 ml of 0.1 M ascorbic acid). Pb2+ . . Pb(C2H302)2ANALYST, MARCH 1986, VOL. 111 307 Table 7. Determination of cadmium in standard and environmental samples Cd found Certified Cd by present Standard Sample content, % method, YO deviation, % Copper alloys, BS 2873 C108 . . . .. . . . . . 0.8 Magnesium alloy, BS 2901D6 . . 1.5-2.5 Low-melting solder, BS 219 Grade T (DIN 1707, L-SnPb) . . . . 18 Blood(per100ml) . . . . . . - Urine (pg g-1 creatine) . . . . - Soil* - Plant* - Industrialeffluentt . . . . . . - Tobacco$ - Waste water§ - . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0.79 0.01 2.20 0.02 17.99 p.p.m. 0.02 10.60 - 10.25 - 4.80 - 1.20 - 30.00 - 1.38 24.96 - - * Samples from G.I.D.C. Makarpura area of Baroda, average of ten samples. t Samples from Nandeshari Industrial area of Baroda. $ Samples from Channi area of Baroda. § From Baroda City. Coefficient of Cd found variation, YO by AAS, % 1.25 0.80 0.90 2.20 0.11 18.0 p.p.m. - 10.58 - 10.26 - 4.82 - 1.20 - 30.08 - 1.40 - 24.93 Beer’s Law, Molar Absorptivity and Sensitivity The Cd - PCHA and Cd - PCHA - PAR complexes obey Beer’s law from 1.40 to 11.25 and from 0.23 to 2.25 p.p.m.of cadmium, respectively. The molar absorptivities were found to be 3.6 X 103 and 4.8 x 104 1 mol-1 em-’, respectively. Precision and Accuracy The standard deviation for ten determinations of cadmium (5.62 ppm) was found to be 0.02 p.p.m. and the coefficient of variation was 0.35%. Stoicheiometry of the Complex The stoicheiometric composition of the complex was deter- mined by employing the slope ratio method.19 The stoi- cheiometry of the Cd - PCHA complex was determined by taking a fixed amount of Cd (1.0 ml of 5 x 10-4 M solution) and gradually increasing the amount of 5 x 10-4 M reagent solution (L1), keeping the concentration of PAR constant.The slope of the graph of log D M versus -log[L1] was found to be 2.0, confirming a metal to ligand (PCHA) ratio of 1 : 2 . The stoicheiometery of the Cd - PCHA - PAR complex was determined by taking a fixed volume of metal solution (1 .O ml of 5 X 10-4 M solution) and 10 ml of 5 x 10-4 M PCHA in chloroform and gradually increasing the amount of PAR (L2). The slope of the graph of log DM versus -log[L2] was found to be 2.0, confirming a metal to ligand (PAR) ratio of 1 : 2. It seems that on addition of PAR to the chloroform extract of cadmium, the PCHA is replaced with PAR. Effects of Diverse Ions In order to assess possible analytical applications of the reaction, the effects of diverse ions on the extraction and spectrophotometric determination of Cd(I1) were studied by adding a known amount of the ion in question to a solution containing 5.62 p.p.m.of cadmium and following the recom- mended procedure. Ba2+, Sr2+, Ca2+, Be2+, Mg2+, Cu2+, Ni2+, Co2+, As3+, Ti4+, Zr4+, Mo6+, V5+ and Zn2+ do not interfere in the determination of cadmium even at high concentrations. Hg2+, Pb2+ and Mn2+ interfere; the interference due to Hg2+ can be eliminated with ascorbic acid and that due to Pb2+ and Mn2+ with sodium fluoride (Table 6). Determination of Cadmium in the Environment Depending on the concentration of cadmium in environ- mental samples (soil, plant, waste water, etc.), 20-50 g of material were digested with an excess of perchloric and nitric acids. The mixture was centrifuged and filtered, and the filtrate was evaporated to dryness and the residue dissolved in and diluted to 100 ml with 0.1 M HCl.A 10-ml of sample solution was used for the determination of cadmium. The results are given in the Table 7. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. Forstner, U., and Muller, G., “Schwermetalle in Flussen und Seen als Ausdrucker Umweltverschmutzung,” Springer Ver- lag, Berlin, Heidelberg, New York, 1974, p. 18. Moore, F. L., Environ. Sci. Technol., 1972, 6, 525. Agrawal, Y. K., and Mehd, G. D., Rev. Anal. Chem., 1982,6, 185. Sandell, E. B., and Onishi, H., “Chemical Analysis, Volume 3, Part I, Traces of Metals: General Aspects,” Wiley- Interscience, New York, 1978. Watanabe, H., and Ohonori, H., Talanta, 1979, 26, 959. Anderson, R. q., and Nickless, G., Anal. Chim. Acta, 1967, 39, 469. Agrawal, Y. K., Rev. Anal. Chem., 1980, 5, 3. Agrawal, Y. K . , and Patel, S. A., Rev. Anal. Chem., 1980, 4, 237. Agrawal, Y. K., and Jain, R. K., Rev. Anal. Chem., 1982, 6, 49. Bass, V. C., and Yoe, J. H., Talanta, 1966, 13, 735. Patke, S. K., PhD Thesis, M.S. University of Baroda, 1980. Mehd, G. D., PhD Thesis, M.S. University of Baroda, 1982. Rathi, B. N., PhD Thesis, M.S. University of Baroda, 1982. Brydon, G. A., and Ryan, D. E., Anal. Chim. Acta, 1966,35, 190. Bhura, D. C., and Tandon, S. G., J. Chem. Eng. Data, 1971, 16, 106. Tandon, S. G., and Bhattacharya, S. C., J. Chem. Eng. Data, 1962, 7, 553. Dean, J. A., “Lange’s Handbook of Chemistry,” Twelfth Edition, McGraw-Hill, New York, 1979. Welcher, F. J., “The Analytical Uses of Ethylenediamine- tetraacetic Acid,’’ Van Nostrand, New York, 1965. Branko, T. B., and Jerome, 0. W., Anal. Chem., 1973, 45, 1519. Paper A51292 Received August 8th, 1985 Accepted September 19th, 1985

ISSN:0003-2654    

DOI:10.1039/AN9861100305

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST, MARCH 1986, VOL. 111 309 Extraction = Spectrophotometric Determination of Manganese with 3-Phenyl-2-mercaptopropenoic Acid Alvaro Izquierdo, M. Dolors Prat, Nuria Garriga and Jose M. Alegria Department of An a1 ytica I Chemistry, University of Barcelona, Barcelona -28, Spa in An extraction - spectrophotometric procedure for the determination of manganese(l1) based on the formation of a chelate with 3-phenyl-2-mercaptopropenoic acid is described. The 1 : 2 manganese complex is quantitatively extracted into isoamyl alcohol (2-methylbutan-1-01) in the pH range 6.3-9.4. The extract shows maximum absorbance at 625 nm (E = 7.3 x lo3 I mol-1 cm-1) and obeys Beer's law up to 20 pg ml-1 of manganese. The method is selective and has been applied successfully to the determination of manganese in standard samples.Keywords 1 Manganese determination; 3-phen yl-2-mercaptopropenoic acid; spectrophotometry; solvent extraction 3-Aryl-2-mercaptopropenoic acids, characterised by the struc- ture I, are good chelating agents for metal ions. Most of the complexes formed are coloured and readily extractable into oxygen-containing organic solvents, such as alcohols and ketones. In previous papersl-3 we reported some mercapto- propenoic acids for the extraction and determination of several metal ions. In this work the complex formation of manganese(I1) with 3-phenyl-2-mercaptopropenoic acid (PMPA) is described. The green colour (Amax, = 625 nm) of the extracted complex into isoamyl alcohol (2-methylbutan-l- 01) provides the basis of a spectrophotometric method for the determination of manganese.R1 \ C=C-C-OH R2/ I It SH 0 R1 = aryl; R2 = alkyl or H I The sensitivity of the proposed method is about four times that of the most widely used routine procedure based on the oxidation of manganese( 11) to manganese(VII)4.5 and, although it is lower than those of other reagents such as the heterocyclic azo dyes 4-(2-pyridylazo)resorcinol (PAR) or 1-(2-pyridylazo)-2-naphthol (PAN) ,637 the method has the advantage of higher selectivity. The most usual interferents in manganese determinations, such as iron and nickel, have been eliminated using masking agents. The applicability of the method was tested by the determi- nation of manganese in standard samples. Experimental Apparatus A Beckman Acta VII spectrophotometer with 10-mm silica cells, a Perkin-Elmer 4000 atomic absorption spectropho- tometer and a Radiometer PHM 64 pH meter equipped with glass - calomel electrodes were used.Reagents Analytical-reagent grade chemicals were used throughout without further purification, unless stated otherwise. Standard manganese solution. A stock solution of man- ganese(I1) chloride (1 g 1-1 of Mn) was prepared and standardised complexometrically. Working solutions were prepared daily by diluting this stock solution to appropriate concentrations. 3-Phenyl-2-mercaptopropenoic acid (PMPA) solution. A 0.5% PMPA solution in purified isoamyl alcohol was used. The PMPA was synthesised and purified as described previ- ously.8 This solution was prepared fresh daily. Buffer solutions. Acetic acid - sodium hydroxide, hydro- chloric acid - tris(hydroxymethy1)aminomethane and am- monia - ammonium chloride buffer solutions were used.The ionic strength was adjusted by the addition of sodium nitrate. Procedures Determination of manganese Transfer 10 ml of buffered sample solution containing 10-200 pg of manganese into a 50-ml separating funnel and equilib- rate with 10 ml of PMPA reagent in isoamyl alcohol by shaking for 10 min. Collect the organic layer, dry over anhydrous sodium sulphate and measure the absorbance at 625 nm against a reagent blank. Use standard manganese solutions treated in the same way to prepare the calibration graph. The colour is stable for at least 2 h. Analysis of aluminium alloys Weigh up to 0.1 g of sample containing 0.1-2 mg of manganese into a 100-ml beaker.Add 5 ml of 10% sodium hydroxide solution in small portions. Evaporate to a syrupy consistency, remove from the hot-plate and carefully add 5 ml of cold water and cool. Add 5 ml of nitric acid (1 + 1) and heat to dissolve the salts and expel the brown fumes. Cool, transfer into a 100-ml calibrated flask, add 1-3 ml of 0.2 M ascorbic acid, depending on the content of iron, and 5-10 ml of 1 M citric acid to prevent the precipitation of aluminium, and adjust the pH to 8.5-9 with ammonia - ammonium chloride buffer solution. Add 1-2 ml of 2 M potassium cyanide solution and dilute to volume. Use 10 ml of this solution to determine manganese as described. Use standard manganese solutions treated in the same way to prepare the calibration graph.Analysis of stainless steel Weigh up to 0.025 g of sample, Containing 0.1-2 mg of manganese, into a 100-ml beaker. Add 10 rnl of a mixture of hydrochloric and nitric acids (1 + 1) and heat gently until decomposition is complete. Add more acid if necessary. Continue the heating to evaporate the excess of acid, cool and transfer into a 100-ml calibrated flask. Add 4-6 ml of 0.5 M310 ANALYST, MARCH 1986, VOL. 111 ascorbic acid, 2 ml of 1 M citric acid and neutralise. Add 4-6 ml of 2 M potassium cyanide solution, adjust the pH to 8.5-9.0 and dilute to volume. Extract at once 10 ml of this solution by shaking for 10 min with 10 ml of PMPA reagent in isoamyl alcohol and determine manganese as described. Use standard manganese solutions treated in the same way to prepare the calibration graph. Analysis of zinc blende Weigh up to 0.1 g of sample containing 0.1-2 mg of manganese into a 100-ml beaker.Dissolve by heating with hydrochloric acid with the aid of nitric acid. Cool, dilute with distilled water, remove insoluble sulphur residue, if present, by filtration, transfer into a 100-ml calibrated flask and proceed according the method described for steel. The amount of ascorbic acid to be added depends on the content of iron. Results and Discussion Characteristics of the Manganese Complex The green manganese complex formed is soluble in water and in some oxygen-containing organic solvents , such as alcohols and ketones, but is not extractable into non-polar solvents such as benzene or chloroform. The absorption spectrum of the aqueous solution exhibits a maximum at 625 nm, but its absorbance decreases gradually with time.Organic solutions have similar absorption peaks 0.800 0.700 0.600 0.500 o.200w\\ O.lo0 t 500 550 600 650 700 Linm Fig. 1. Absor tion spectra of the manganese(I1) complex extracted into isoam I aiohol at different pH values of the aqueous phase. Mn(II), 6.16 pg ml-1. pH, A, 4.32; B, 5.21; C, 7.18; and D, 10.52 t 0.400 0.300 q I - 0.200 1 0.100 4 5 6 7 8 9 1 0 PH Dependence of complex formation on acidity. Mn(II), 2.9 Fig. 2. pg ml-1 and good stabilities. Under optimum conditions, manganese can be extracted quantitatively from aqueous solutions with PMPA in isoamyl alcohol. The absorption spectra of the manganese complex in isoamyl alcohol are shown in Fig.1. The curves have a maximum at 625 nm and PMPA has negligible absorption at this wavelength. A maximum and constant absorbance of the organic phase at 625 nm is obtained between pH values of the aqueous phase of 6.3 and 9.4, as shown in Fig. 2. In less or more acidic solutions, the absorbance decreases rapidly. The presence of electrolytes slightly increases the maximum recovery of manganese but no appreciable effect is observed when the ionic strength is > 1. A large excess of reagent is necessary for the complete extraction of manganese. The absorbance of the organic phase increases with increasing amount of PMPA up to an 80-fold molar excess. The colour intensity of the complex thus obtained is constant for at least 2 h. Addition of more reagent has no effect on the absorbance but improves the stability of the extracted complex.A shaking time of 2 min is adequate for the quantitative extraction of manganese under the optimum conditions, the recovery being higher than 99.9%. In the presence of high concentrations of foreign species the colour development is retarded in some instances and therefore a shaking period of 10 min is selected in the recommended procedure. The extraction efficiency was calculated by measuring the man- ganese concentration of the aqueous phase by atomic absorp- tion spectrometry. Beer’s Law and Precision The system obeys Beer’s law up to a concentration of 20 pg ml-1 of manganese. The molar absorptivity at 625 nm is 7.3 X 103 1 mol-1 cm-1. The limit of detection, expressed as the concentration that gives an absorbance three times greater than the standard deviation of the blank, is 0.06 pg ml-l of manganese.The optimum concentration range, evaluated by Ringbom’s method, is 1.4-5.6 pg ml-1 of manganese. Ten determinations on standard solutions containing 3.47 pg ml-1 gave a relative standard deviation of 0.28%. Composition of the Complex The stoicheiometry of the extracted species was studied by the equilibrium shift method. Manganese (3.24 pg ml-1) was extracted at constant pH (5.1) using various concentrations of reagent (2.10-8.10 M). The mercapto acid was introduced in ethanolic medium into the aqueous phase because at this pH the equilibrium was attained very slowly when the reagent was dissolved in the organic phase. The ethanol concentration in the aqueous phase before the extraction was kept constant at 10% V/V.The plot of log D vs. PMPA concentration was linear with a slope of 2.1, indicating that the metal to ligand ratio was 1 : 2. Interferences The effects of diverse ions on the determination of manganese were examined under the optimum conditions. The tolerance limit was taken as that concentration which does not cause more than a 2% change in the absorbance. The complexing agents oxalate, tartrate, citrate, phosphate, cyanide and fluoride do not interfere in the determination and were used as masking agents. EDTA inhibits the colour formed by manganese with PMPA. The interference of most of the ions forming coloured species that absorb near 625 nm under similar experimental conditions can be avoided.Thus, addition of cyanide elim- inates the interferences due to Pd(II), Cu(II), Ni(II), Fe(I1) and Fe(II1) (after reduction with ascorbic acid), whereas Ti(1V) can be masked with fluoride. Certain metal ions, suchANALYST, MARCH 1986, VOL. 111 311 Table 1. Results for the determination of manganese in real samples. Each result is the average of five separate determinations Sample Composition, % Mn certified, YO Mn found, % Stainlesssteel(NBS121d) . . . . . . . . . . . . Cr17.4,Ni11.71,Ti0.34,Mo0.16, 1.80 1.81 CuO.12,CoO.1O,SiO.54,CO.67, P 0.019 Fe 0.18, Ni 0.10, Pb 0.05 Pb1.33,As0.14 Duraluminium IV (Hoepfmer Gebr.) . . . . . . . . Si 5.63, Cu 1.68, Mg 0.45, Zn 0.41, 0.25 0.24 Zinc blende I1 (Hoepfmer Gebr.) . , . . . . . . . . Zn 48.50, S 31.65, Fe 9.87, Cu 1.61, 0.61 0.60 as Al(III), Cr(III), Cd(II), Zn(I1) and Pb(II), which precipi- tated at the pH used, cause interference in some instances.This can be overcome by adding citrate or cyanide ions. Addition of ascorbic acid, potassium cyanide and/or citric acid permits the determination of manganese at the 3.5 pg ml-1 level without interference from 500 pg ml-l of copper , nickel, iron, palladium , cadmium, lead, zinc, chro- mium and aluminium. Of the metals tested, only Co(I1) (2 pg ml-1) and V(V) (5 pg ml-1) caused serious interferences. In samples containing large amounts of some foreign species, a precipitate appeared 10-15 min after the test solution was prepared. In order to obtain good results, the extraction must be carried out immediately. Applications To test the validity of the proposed method, manganese was determined in a standard steel sample supplied by the US National Bureau of Standards and in zinc blende and duraluminium samples supplied by Hoepfmer Gebr. (Ham- burg, FRG). The results obtained agreed well with the expected values, as shown in Table 1. 1. 2. 3. 4. 5. 6. 7. 8. References Izquierdo, A., and Calmet, J., Talanta, 1977, 25, 56. Izquierdo, A., GinC, M., and Compaii6, R., J . Znorg. Nucl. Chem., 1981,43, 617. Izquierdo, A., and Carrasco, J., Analyst, 1984, 109, 605. Cooper, M. D., Anal. Chem., 1953, 25,411. Gottschalk, G., Freseniw 2. Anal. Chem., 1965,212, 303. Ahrland, S., and Herman, R. G., Anal. Chem., 1975, 47, 24. Donaldson, E. M., and Inman, W. R., Talanta, 1966, 13,489. Izquierdo, A., and Garriga, N., Talanta, 1985, 32, 669. Paper A51286 Received August 5th, 1985 Accepted October 15th, 1985

ISSN:0003-2654    

DOI:10.1039/AN9861100309

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST, MARCH 1986, VOL. 111 Fe(C10413 0.1% HC104 10 M 313 2.4 2.4 Automated Flow Injection Spectrophotometric Determination of Some Phenothiazines Using Iron Perchlorate: Applications in Drug Assays, Content Uniformity and Dissolution Studies Michael A. Koupparis" and Antonie Barcuchovat Laboratory of Analytical Chemistry, Department of Chemistry, University of Athens, Athens 106 80, Greece An automated flow injection determination of some phenothiazine derivatives, based on their oxidation with iron(ll1) in a strongly acidic medium, is described. Chlorpromazine, promethazine, thioproperazine, promazine, levomepromazine, fluphenazine, trifluoperazine and thioridazine are determined in the range 10-250 pg ml-I. The precision is better than 1% and a measurement rate of 120 per h o u r can be obtained.The method was evaluated by carrying out an interference study with common excipients and other drugs, a recovery study and by the analysis of commercial formulations, the results of which are compared with those obtained by the official method. The method was applied in content uniformity tests and for monitoring the dissolution of solid dosage forms. Keywords: Phenothiazine determination; flow injection analysis; routine assays; content uniformity test; dissolution studies Phenothiazine derivatives are one of the most important groups of medicaments, used as antihistamines, tranquillisers, antiemetics, anti-Parkinson drugs, etc. About 30 phenothia- zine drugs and 100 formulations are commercially available. The importance of these drugs prompted the development of many methods for their determination, reviewed by Blazekl and Fairbrother.2 Apart from the official methods, based on UV spectrophotometry and non-aqueous titrimetry, two-phase, tetraphenylborate, ammonium reineckate, complexometric and oxidative titrations have also been used.A variety of manual spectrophotometric methods are based on coloured complex formation or oxidation reactions yielding intensely coloured radicals. Individual phenothiazines have been determined using molybdophosphoric acid ,3 molybdoarsenic acid,4 chloramine-T,S Van Urk's reagent6 and acid dyes7 as reagents. Other spectrophotometric methods for phenothiazines employ Pd(II),8 Ce(IV),9 Fe(III),lO N-bromosuccinimide ,11 etc. A number of chromatographic and polarographic methods have also been proposed.Despite the large number of manual methods available, automated procedures that can be used for the routine analysis of many samples in governmental and industrial quality control laboratories are limited. A semi-automated method using the Technicon air-segmented continuous flow analyser is based on the dilution and oxidation with NaN02 (if neces- sary), and a double extraction and UV determination.12 Another continuous flow method (unsegmented closed-loop mode) has been proposed by Mottola and Hanna,9 based on the oxidation of phenothiazines with cerium(1V) to produce transient absorbance peaks due to the intermediate red free radicals. In recent years, flow injection analysis (FIA) has found wide application in various fields of routine analysis, including pharma~eutical.13~14 The versatility and simplicity of the FIA technique allow its adaptation at relatively low cost to the different requirements of a variety of analytical problems.The introduction of content uniformity tests (required for all solid formulations with contents equal to or less than 50 mg of drug) increases the workload of quality control laboratories at least ten-fold. This type of analysis, in which a large number of samples of very similar composition have to be analysed, is an attractive field for automated FIA. * To whom correspondence should be addressed. t On leave from the Department of Physical Chemistry, Faculty of Pharmacy, Charles University, Hradec Kr AovC, Czechoslovakia. Recently FIA was successfully interfaced with dissolution apparatus and the automation of dissolution studies of drug formulations was achieved.15 Complete tabulated and graphi- cal multi-point time profiles of dissolutions were obtained.In this paper, the development of an automated FIA method for the determination of some phenothiazines and its application to routine assays, content uniformity tests and dissolution studies are described. The reaction of the pheno- thiazines with iron(II1) in a strongly acidic medium with HC104 has been found to be suitable for adaptation. Experimental Apparatus A laboratory-constructed automated photometric flow injec- tion analyser ,I6 assembled from commercially available com- ponents and instruments, was used. This analyser provides for automatic removal of aliquots of samples from the turntable, or manually processed solutions, injection of an accurate adjustable sample volume into the reagent flow stream, measurement of the absorbance peaks of the reaction mix- tures and manipulation of the analytical data.The analyser is controlled by a versatile and inexpensive microcomputer (Rockwell AIM 65). The analytical manifold designed for the determination of phenothiazines, together with the flow-rates of the reagents, is shown in Fig. 1. The A,,,. for each individual phenothiazine was set on the photometer (Table 1). Because of the use of concentrated HC104, acid-resistant Vidon tubing in the peristaltic pump of the analyser is preferable to the usual silicone-rubber tubing. For the automated dissolution studies on solid dosage forms of phenothiazines, the FIA analyser was coupled with a USP rotating basket apparatus as described previously.15 Flow-rate/ m 1 3 - 1 200 PI Sample Fig. 1. determination of phenothiazines Schematic diagram of the automated FIA system for the314 ANALYST, MARCH 1986, VOL. 111 Reagents All solutions were prepared in de-ionised, distilled water from analytical-reagent grade materials. Pure phenothiazines were obtained directly from the manufacturers and their purity was tested using USP procedures. The formulations analysed and tested were obtained from local commercial sources. Phenothiazine standard solutions. A 500 pg ml-1 stock solution of each phenothiazine was prepared by dissolving 0.1250 g of pure substance in 250 ml of water and stored in amber-coloured bottles in a refrigerator.Working standard solutions, in the range stated in Table 1 for each phenothiaz- ine, were prepared by appropriate dilution of the stock solution. Iron(III) - perchloric acid solution, 0.10% mlV-lO M. Prepared by dissolving 0.500 g of Fe(C1O4I3 in 500 ml of 10 M HC104, prepared by appropriate dilution of concentrated 70% mlV acid. Carrier solution. Water. Dissolution medium. 0.10 M HCl. HC104, 0.10 M. Procedures Sample preparation Assays in dosage forms. (a) Tablets. Not less than 20 tablets were weighed and finely powdered. An accurately weighed portion, equivalent to about 50 mg of the phenothiazine, was transferred into a 100-ml calibrated flask and diluted to volume with water. Using an ultrasonic bath or a mechanical shaker, the powder was completely disintegrated and the solution was filtered.From this sample solution, working solutions were prepared by appropriate dilution, so that their concentration was in the individual phenothiazine working range. (b) Injections and syrups. An accurately measured volume was appropriately diluted so that the phenothiazine concen- tration was in the working range. (c) Creams. An accurately weighed portion of the cream, equivalent to about 100 mg of phenothiazine, was transferred into a separating funnel, dissolved in a mixture of 25 ml of diethyl ether and 10 ml of methanol and extracted with three 25-ml portions 0.10 M HC104. The extractant was diluted to 100 ml with the same medium and further diluted to fall within the working range.Assays for content uniformity tests. Ten individual tablets were selected, disintegrated and diluted with water in separate calibrated flasks to fall within the working range. Measurement procedures The spectrophotometer of the FIA analyser was set at the h,,,, of the individual phenothiazine, the reagents were pumped and the 100% transmittance was set. Data on the number of standards and samples, the runs per standard and sample and the injection and load time (8 and 15 s, 1 .o 0.8 0.6 T 0.4 0.2 0 respectively) were provided to the microcomputer’s routine program. The determination then proceeded automatically. Dissolution studies were performed according to a “dissolution” program loaded into the microcomputer’s memory. A calibration graph was initially obtained using at least three standards of the phenothiazine examined, prepared in the dissolution medium used (0.1 M HC1) and thermostated at 37 “C.Then a dosage form was placed into a 250-mesh screen basket, immersed and rotating at 60 rev min-l, in a double-wall beaker containing 250 or 900 ml of 0.1 M HCl (for formulations containing less or more than 25 mg, respectively). The dissolution medium was thermostated at 37 k 0.5 “C. The filtered dissolution medium was circulated continu- ously through the sample loop (200 pl) and at pre-set time intervals was injected into the carrier stream. At the end of the experiment the entire dissolution profile was presented on the chart recorder as a series of absorbance peaks versus time and also on the computer’s printer as a table consisting of time, absorbance, drug concentration and percentage dissolution.Results and Discussion From the various oxidants, already used for manual spectro- photometric determinations of phenothiazines, iron(II1) was chosen for its mild oxidation power. The formal reduction potential of the Fe(II1) - Fe(I1) couple does not allow further oxidation of the coloured free radicals to the uncoloured sulphoxides.17 Using more powerful oxidants, such as Ce(IV), transient redox absorbing products are produced.9 Optimisation of the Method In order to optimise the proposed FIA method, the effects of the various experimental parameters were studied. Of the various acids tested, perchloric was found to be the most suitable. Its mixed solution with iron(II1) was uncoloured (in contrast to hydrochloric acid) and had no oxidising action on the free radicals produced (in contrast to nitric acid).The perchloric acid concentration was found to have a drastic effect on the absorbance peak heights shown in Fig. 2(a). A 10 M concentration was chosen as the optimum for high sensitivity. This high acid concentration requires the use of acid-resistant pump tubing. Various sources of iron(II1) reagent were tested. NH4Fe(S04)* gave turbid solutions at the high acid concentra- tion used and filtration was required. Fe(N03)3 showed a considerable decreasing effect on the absorbance peaks at concentrations higher than 0.5% m/V because of the oxidising effect of nitrates on the free radicals. Fe(C104)3 was found to be the most suitable and 0.1 Yo m/V (2.75 x 10-3 M) was chosen as the optimum.Fig. 2(b) shows the effect of Fe(N03)3 and Fe(C104)3 concentrations on the absorbance peaks. The reaction of iron(II1) with phenothiazines is very rapid17 and is completed in a few seconds. The effect of the length of the reaction coil ( L ) is shown in Fig. 2(c). An increase in L 0 1 2 3 4 5 50 100 150 200 [ F e ( l l l ) ] / ~ X 10’ Ucm Fig. 2. Effect of experimental parameters on the FIA measurement of 100 pg ml-1 of promethazine.HC1. (a) Effect of HC104; [Fe(III)] = 2.5 .I< M, L = 50 cm. (b) Effect of Fe(II1): A, Fe(C104)3; B, Fe(N03)3; [HC104] = 10 M, L = 50 cm. (c) Effect of reaction coil length; [HC104] = 10 M, [Fe(ClO,),] = 2.75 x 10-3 MANALYST, MARCH 1986, VOL. 111 31.5 decreases the absorbance peaks, because of increasing sample dispersion, and reduces the sample throughput.A 50-cm reaction coil was chosen as the optimum to ensure high reproducibility of mixing of the sample with the reagents, high sensitivity and a high measurement rate. Validation of the Method Typical FIA peaks for the calibration graph of promethazine hydrochloride are shown in Fig. 3, with increasing and decreasing concentrations in order to show the absence of any carry-over effect. Data relevant to the FIA determination of several pheno- thiazines in pure solutions are summarised in Table l. The linearity of the stated calibration graphs was excellent, with correlation coefficients ( r ) varying from 0.999 to 0.99999. The analytical ranges of the determinations are suitable for both assays of formulations (containing 10-100 mg) and dissolution studies over a wide range of concentrations.The precision of the measurements varied from 3.3 to 0.4% relative standard deviation (RSD) (n = 10) for the lowest to the highest concentration of the calibration graphs. The sample throughput is high (120 measurements per hour), assuming a 15-s load time, an 8-s injection time and 7 s for data collection, manipulation and printing. Interference studies In order to use the automated FIA method in assays of commercial formulations, common excipients and other additives and drugs coexisting in formulations were tested for possible interference. Synthetic solutions containing 50.0 pg ml-1 of promethazine hydrochloride and various amounts of foreign substances were measured.The undissolved material, if any, was filtered before measurement. The recovery results are shown in Table 2. As shown, the only possible interferents are carbopol (carboxypolymethylene) , which causes enhancement of the peak height generating coloured aggregates, sodium lauryl sulphate, sodium sulphite, which decreases the peak height because of its reducing action , ascorbic acid, acetylsalicylic acid and codeine phos- phate at high concentrations not found in formulations with phenothiazines. All other common excipients showed recoveries varying from 97.9 to 102.2%. Accuracy The accuracy of the automated FIA method was examined by performing recovery experiments on solutions prepared from chlorpromazine formulations.A mean recovery of 101.4% was found (range 98.4-105.2%) (Table 3). Similar experi- ments for promethazine hydrochloride gave an average recovery 101.6%. The proposed method was also evaluated by analysing commercial formulations of chlorpromazine , promethazine and thioproperazine and comparing the'results obtained with those obtained by the official USP methods. A satisfactory agreement between the results was obtained (Table 4) with a mean relative difference of 1.2% (range 0.2-2.8%). The RSD for FIA determinations of these formulations varied from 0.5 to 2.6% (three samples, three measurements per sample). Content uniformity tests Fig. 4 shows a typical recorded FIA profile of content uniformity test for Phenergan tablets containing 25 mg of promethazine hydrochloride.Table 5 shows the results of content uniformity tests for chlorpromazine, promethazine and thioproperazine formulations. All the samples examined were found to meet the USP requirements (all the samples examined fell within the limits of 85-115% of the average definition in the individual monograph). The standard devia- tions of the contents varied from 0.3 to 0.7 mg Der tablet (I. 6-3.OYo). 120 Fg ml-1 T O.! 0 100 Time - Fig. 3. Typical FIA absorbance peaks used for the calibration graph of promethazine.HC1 with increasing and decreasing concentrations of promethazine.HC1 Table 1. Summary of results pertinent to the determination of phenothiazines by automated FIA (200-pl sample volume) Compound Chlorpromazine hydrochloride Promethazine hydrochloride Promazine hydrochloride .. Methotrimeprazine (levomepromazine) hydrochloride . . . . Thioproperazine bismethanesulphonate . . Fluphenazine dihydrochloride Tri fluoperazine dihydrochloride . . . . Thioridazine hydrochloride Detection Determination ranget/ Sensitivity$/ nm pg ml-1 pgml-l mA pg-' ml L l a x . / limit*/ . . 535 . . 515 . . 513 . . 560 . . 515 . . 500 . . 501 . . 580 0.6 0.6 0.7 0.7 1.3 1.6 1.2 1.6 6-124 6-117 6-128 7-133 12-248 15-312 11-230 16-318 8.07 8.53 7.81 7.50 4.02 3.20 4.35 3.15 * Given as the concentration producing a signal twice the standard deviation of the most dilute standard. t Given as the concentration range producing absorbance peaks of 0.05-1 A. $ Given as the slope of the calibration graph (mA vs. concentration).316 ANALYST, MARCH 1986, VOL.111 Table 2. Analytical recovery of promethazine hydrochloride (50 pg ml- I ) from various additives used as excipients and from some coexisting drugs Type Excipients . . . . Drugs . . . . . , * Carboxypolymethylene. t Polyethylene glycol 4000. Additive Galactose Glucose Lactose Sugar Starch Talc Gelatin Cellulose Cellulose acetate Carbopol* Car bow ax? Sodium lauryl sulphate Magnesium stearate NaCl KH2P04 CaCl, Na2S03 Ethanol Formaldehyde Phenol Na,EDTA hydroxyphthalate Ascorbic acid Acetaminophen Phenacetin Acetylsalicylic acid Caffeine Codeine phosphate Concentration ratio, additive to promethazine 20 20 20 20 20 20 20 20 20 10 20 10 20 20 20 10 25 100 25 100 100 25 25 1 100 50 100 Saturated Saturated 100 50 Recovery, 98.9 100.0 97.9 100.0 100.0 100.0 100.0 100.0 102.1 124.4 100.0 106.4 98.9 100.0 100.0 101.0 80.0 102.2 101.2 102.2 110.2 101.2 53.6 100.0 105.6 100.0 100.0 110.8 100.0 114.7 101.0 Yo ( n = 3) Table 3.Recovery experiments for chlorpromazine hydrochloride added to sample solutions of commercial formulations Clorpromazine. HCI/ pg ml-1 Formulation Initially present Largactil tablets, 25 mg . . . . 54.6 54.6 Largactil tablets, 100mg . . . . 55.2 55.2 Antistresstablets,25mg . . . . 54.6 54.6 55.9 Solidon tablets, 100 mg . . . . 52.2 52.2 Largactil injection, 5 mg ml-1 . . 59.0 59.0 Antistress tablets, 100 mg . . . . 55.9 Added 25.0 50.0 25.0 50.0 25.0 50.0 25.0 50.0 25.0 50.0 25.0 50.0 Recovery, Recovered YO 25.5 49.2 24.9 49.2 25.5 49.2 25.8 50.3 25.8 51.0 26.3 51.9 102.0 98.4 99.6 98.4 102.0 98.4 103.2 100.6 103.2 102.0 105.2 103.8 Mean: 101.4ANALYST, MARCH 1986, VOL.111 317 Table 4. Determination of chlorpromazine, promethazine and thioproperazine in commercial formulations by FIA and the official USP method based on extraction and UV spectrophotometry Drug/mg Type Tablets . . . Formulation (pheno t hiazine) (chlorpromazine) . . . Largactil Antistress Solidon Ancholactil Phenergan Maj e p t il (chlorpromazine) (chlorpromazine) (chlorpromazine) (promethazine) (thioproperazine) Cream . . . . . . . . Phenergan Syrup, injection . . . , . . Phenergan Largactil Claimed 25 100 25 100 100 100 25 10 2% mlm 1 mg ml-1 5 mg ml-1 Found Relative difference (FIA -. Official -official method), FIA* method? Y O 24.5 f 0.4 24.1 f 0.5 1.7 99.0 f 0.5 99.5 _+ 0.7 -0.5 24.5 f 0.2 25.0 f 0.4 -2.0 100.3 t 0.7 100.7 f 1 -0.4 93.7 f 0.8 94.2 t 0.9 -0.5 98.1 f 1 97.9 f 1 0.2 24.3 k 0.3 24.2 f 0.5 0.4 10.6 f 0.1 10.9 f 0.2 -2.8 2.03 f 0.02 2.05 t 0.05 -1.0 1.13 f 0.03 1.15 f 0.04 -1.7 5.29 k 0.07 5.20 f 0.2 1.7 * Average for three samples, each measured three times, f standard deviation.t Average for three samples, each measured once, f standard deviation. Table 5. Content uniformity tests on commercial tablets of some phenothiazines. Ten tablets were analysed; measurements in .;iplicate Range Mean f SD Formulation mg per tablet % of claim mg per tablet YO of claim Phenergan tablet, 25 mg (promethazine. HCI) . . . . . . 24.3-25.6 97.1-102.5 24.9 f 0.4 99.5 k 1.6 100.0 f 2.8 Largactil tablet, 25 mg (chlorpromazine. HCl) .. . . . . 23.7-26.0 94.8-104.0 25.0 f 0.7 Majeptil tablet, 10 mg (thioproperazine bismethanesuiphonate) . . . . . . . . . . . . . . 10.2-1 1 .O 102.0-110.0 10.6 f 0.3 106.0 f 3.0 115 100 85 s +- a 0 0 + 0 1 2 3 4 5 6 7 8 9 1 0 Samples I Fig. 4. Typical FIA profile of the content uniformity test on Phenergan tablets (25 mg) (promethazine. HCl). Statistical treatment of contents (mg per tablet): range, 24.3-25.6; mean, 24.9 f 0.4 (RSD = 1.6%). Only one peak is shown for each individual sample 30 20 10 0 Time/min Fig. 5. Phenergan tablets at pH 1.0 Typical FIA dissolution test for ( a ) Largactil and (b)318 ANALYST, MARCH 1986, VOL. 111 Table 6. Dissolution tests on phenothiazine tablets Dissolution, O h * ~~ ~ Injection No. Time/min Largactilt PhenerganS Majeptil§ 1 2 0 2.9 0 2 4 8.0 9.8 0 3 6 18.4 18.1 14 4 8 28.7 26.5 33.2 5 10 39.0 36.8 42.4 6 12 49.0 46.1 51.6 7 14 57.6 55.9 58.5 8 16 65.8 63.7 63.1 9 18 73.5 72.6 65.4 10 20 79.6 86.8 70.1 11 22 84.4 95.1 74.7 12 24 88.5 100.0 81.5 13 26 92.3 100.0 86.1 14 28 95.7 89.4 15 30 98.2 93.1 16 32 99.3 95.3 17 34 99.8 100.0 18 36 100.0 100.0 * Average of three experiments.t Chlorpromazine.HC1, 100 mg; dissolution medium, 900 ml. $ Promethazine.HC1, 25 mg; dissolution medium, 250 ml. 0 Thioproperazine bismethanesulphonate, 10 mg; dissolution medium, 250 ml. Dissolution tests Fig. 5 shows typical results for dissolution profiles of Largactil and Phenergan tablets at pH 1.0 (0.1 M HCI). Summarised data are shown in Table 6. An average standard deviation of 2.4% dissolution (range 0-5.7%) (n = 3) was obtained in all three instances.Considering the tablet-to-tablet variability, the precision of the proposed technique is excellent. Conclusions The proposed automated FIA method is rapid, sensitive, precise and accurate, with a high sample throughput. The set-up time is very short and reagent consumption is low. Interferences from common excipients are limited and the method can be used for the routine analysis of commercial formulations instead of the tedious official methods where separation of the drug is required. The application of FIA to content uniformity tests should be very useful in meeting the considerably increased workload required to carry out these tests. The automated performance of dissolution studies of formulations, where a “kinetic” dissolution profile is obtained very easily, should be useful in pharmaceutical technology. The data obtained can be treated to calculate dissolution rate constants. The authors thank the Greek Ministry of Education for supporting one of the authors (A.B.). 1. 2. 3. 4. 5. 6. 7. 8, 9. 10. 11. 12. 13. 14. 15. 16. 17. References Blazek, J., Pharmazie, 1967,22, 129. Fairbrother, J. E., Pharm. I . , 1979, 222, 271. Stan, M., Dorneanu, V., and Ghimicescu, G., Talanata, 1977, 24, 140. Ramappa, P. G., Sanke Gowda, H., and Nayak, A. N., Analyst, 1980, 105, 663. Issa, A. S., Beltagy, Y. A., and Mahrous, M. S., Talanta, 1978, 25, 710. Murty, B. S. R., and Baxter, R. M., J. Pharm. Sci., 1970, 59, 1010. Matsui, F., and French, F. M., J. Pharm. Sci., 1971, 60, 287. Mercaldo, D. E., Ann. N.Y. Acad. Sci., 1968, 153, 403. Mottola, H. A., and Hanna, A., Anal. Chim. Acta, 1978, 100, 167. Istvan, F., Floderer, H., and Valeria, H., Acta Pharm. Hung., 1957,27, 152. Taha, A. M., El-Rabbat, N. A., El-Kommos, M. E., and Refat, I. H., Analyst, 1983, 108, 1500. “Drug Autoanalysis Manual,” Second Edition, United States Food and Drug Administration, Washington, DC, 1972, method No. 82. Landis, J. B., in Munson, J. W., Editor, “Pharmaceutical Analysis, Part B,” Marcel Dekker, New York, 1984, pp. Rios, A., De Castro, M. D., and Valcarcel, M., J. Pharm. Biomed. Anal., 1985, 3, 105. Koupparis, M., Macheras, P., and Reppas, C., Znt. J. Pharm., 1984, 20, 325. Koupparis, M., and Anagnostopoulou, P. , J. Autom. Chem ., 1984, 6, 186. Gasco, M. R., and Carlotti, M. E., J . Pharm. Sci., 1978, 67, 168. 217-277. Paper A51255 Received July 15th, 1985 Accepted October 9th, 1985

ISSN:0003-2654    

DOI:10.1039/AN9861100313

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST, MARCH 1986, VOL. 111 319 Spectrophotometric Determination of Acetaminophen, Oxyphenbutazone and Salicylamide by Nitration and Subsequent Complexation Reactions Afaf A. El Kheir Pharmaceutical Chemistry Department, Faculty of Pharmacy, Zagazig University, Zagazig, Egypt Saied Belal* Faculty of Pharmacy, Alexandria University, Alexandria, Egypt and Mohammad El Sadek and Abdullah El Shanwani Pharmaceutical Chemistry Department, Faculty of Pharmacy, Zagazig University, Zagazig, Egypt A simple and sensitive spectrophotometric method for the assay of three antipyretic drugs through their nitration and subsequent complexation with an nucleophilic reagent is proposed. The experimental conditions leading to optimum chromagen stability and intensity were studied. The results of the application of the method to the assay of the test compounds in unit doses were compared statistically with those obtained by official methods, and demonstrated good accuracy and precision.Keywords: Antipyretics analysis; phenolic drugs analysis; nitro derivatives; spectrophotometry Acetaminophen, oxyphenbutazone and salicylamide tablets are important and widely used analgesic - antipyretic and antirheumatic drugs. Their spectrophotometric determination in dosage forms and biological fluids offers advantages of both sensitivity and simplicity. Most spectrophotometric methods for these compounds were based on the utility of their phenolic hydroxy groups to yield spectrophotometrically useful chromagens through diazo couplingl-5 or nitrosation reactions.6-11 The use of nitration has also been reported12J3 for the determination of acetaminophen; the other two compounds concerned here were not involved.This work was concerned with the study of the preparation and use of the polynitro derivatives of the investigated drugs as intermediates in their spectrophotometric determination, through interaction with alkaline ketone reagents to form Meisenheimer-type complexes. The aim was to develop a simple and sensitive assay procedure for these drugs. Polynitro aromatic compounds are known to form various intensely coloured complexes with nucleophilesl4 and other charge-transfer complexes and adducts with electron-rich species.15 One of these interactions occurs between dinitro or trinitro aromatic compounds with bases, including anions generated from a base and a ketonic compound, forming anionic sigma complexesl6-18 in which the reacting species are linked with a sigma bond.16 These complexes are more intensely coloured than the original nitro derivatives.This type of interaction has been widely used to determine ketosteroids with dinitrobenzene - acetone - potassium hydroxide,19 to determine some amines20 and to detect polynitro aromatic compounds21>22 and compounds containing an active methylene or a hydrogen a to a carbonyl gro~p.23-~~ It was claimed that test compounds that easily form polynitro derivatives may form an anionic sigma complex. This sugges- tion was investigated to establish whether the intensely coloured products may be spectrophotometrically useful in determining the parent compounds.The reaction sequences were developed into a spectrophotometric assay of the test drugs. Experimental Apparatus A Pye Unicam SP-400 UV - visible spectrophotometer with 1-cm quartz cells was used. * To whom correspondence should be addressed. Present address: College of Medicine and Allied Sciences, King Abdulaziz University, Jeddah, Saudi Arabia. Materials and Reagents Acetaminophen, oxyphenbutazone and salicylamide powders. Obtained from Alexandria, El Nile and CID Pharmaceutical Companies, respectively. Concentrated nitric acid and concentrated sulphuric acid. Prolabo. Acetone. BDH Chemicals, laboratory-reagent grade. Potassium hydroxide solution, 10% mlV. Standard ethanolic or methanolic solutions of the drugs. Procedures Pure drugs An accurately weighed amount of the drug (25-100 mg) was transferred into a 100-ml calibrated flask, treated with 2 ml of nitric acid and 2 ml of sulphuric acid (care must be taken during handling of the nitration mixture), left to stand for 10 min, then cooled and diluted to volume with distilled water.A 20-ml aliquot of the solution was diluted with distilled water to 100 ml in a calibrated flask. Aliquots of 0.5-3 ml of the diluted solution were transferred into 25-ml calibrated flasks, treated with 2-3 ml of acetone and 5 ml of potassium hydroxide solution and diluted to volume with distilled water. The absorbance of the resulting colour was measured against a reagent blank at 355 nm (acetaminophen) or 390 nm (oxy- phenbutazone or salicylamide). Construction of calibration graphs Aliquots of standard alcoholic solutions of the drugs (~10-20 mg of acetaminophen and 5-10 mg of salicylamide or oxyphenbutazone) were evaporated to dryness in small beakers at about 60 "C.The residues were treated with 2 ml of nitric acid and 2 ml of sulphuric acid, left to stand for 10 min, cooled, transferred quantitatively into 100-ml calibrated flasks and diluted to volume with distilled water. A 1-ml volume of each solution was transferred into a 25-ml calibrated flask and the assay was completed as above. Application to tablets Twenty tablets were weighed, powdered, mixed and an amount of the powder (= 60 mg of acetaminophen and 20 mg of oxyphenbutazone or salicylamide) was transferred into a small beaker. The powder was subjected to nitration as above and the reaction mixture was transferred into a 100-ml calibrated flask, diluted to volume with distilled water and filtered.A 1-ml volume of the filtrate was transferred into a 25-ml calibrated flask and the assay was completed as above.320 ANALYST, MARCH 1986, VOL. 111 The concentration of drug was calculated using a regression equation obtained by applying the procedure to serial standard concentrations of the drug. The equation was checked frequently. Results and Discussion Nitration of acetaminophen and salicylamide (Schemes 1 and 3) probably results in a dinitro derivative, in spite of the excess of nitric acid used. This was proved by treating the nitro derivative with trimethylamine in a non-aqueous solvent; failure to develop a colour precluded the possibility of formation of a trinitro derivative.This was expected from steric hindrance considerations. On the other hand, oxyphen- butazone, owing to its additional benzene ring (in addition to the p-hydroxyphenyl moiety), is able to form a polynitro derivative (probably the tetranitro derivative). This is demon- strated from the more intense colour produced, the red shift NHCOCH3 Scheme 1 6 o>N,(y(o 0 0 (with respect to acetaminophen) of A,,,, and the higher absorptivity of the final complex. Furthermore , treatment of the chloroform extract of oxyphenbutazone nitro derivative with trimethylamine resulted in an intense red colour, which suggests that more than two nitro groups are present in the derivative molecule in acetaminophen and the p-hydroxy- phenyl moiety (Schemes 1 and 2).In oxyphenbutazone, it is assumed that the two nitro groups enter the two ortho positions (with respect to the OH), whereas in salicylamide (Scheme 3) one nitro group should be directed to the vacant para position. In the second benzene ring of the nitro oxyphenbutazone structure, the two entering nitro groups would be metu to each other. The optimum volume of nitric acid leading to the maximum final colour intensity was 2 ml. Such a large volume ensured the introduction of a minimum of two groups into the molecule, essential for the electron-acceptor property. Treatment of the electron-deficient polynitro derivatives with ketones such as acetone in alkaline medium resulted in intensely coloured complexes by interaction with anions such as CH3COCH2- (Schemes 1-3).These complexes are of the Meisenheimer type15-17 and differ in colour from those obtained by adding only alkalis to the nitro derivatives. The latter colours originate from the formation of alkali metal salts and the generation of quinoid structures.19 The optimum volume of acetone was 2-3 ml and the optimum volume of alkali was 5 ml of a 10% solution of potassium hydroxide. The absorption curves of the complexes are shown in Fig. 1. The low colour intensities and shorter A,,,. for the acetaminophen colour may be attributed to the auxochrome (-NHCOCH3) in the para position to the OH group, in contrast to the chromophoric (NO,) group for the salicylamide colour. The intense colour and higher absorptivity in oxyphenbutazone colours may be attributed to complexation in both benzene rings.The coloured complexes are stable for at least 1 h, and give clear aqueous solutions in spite of the poor solubility of the nitro derivatives in cold water. The calibration graphs obtained obeyed Beer's law in the range 0.2-1,0.01-0.05 and 0.05-0.30 mg ml-1 for acetamino- phen, oxyphenbutazone and salicylamide, respectively. Using the least-squares method,28 the regression equation describing the calibration graphs ( A = a + bC, where A = absorbance of a 1-cm layer, a = intercept, b = slope and C = concentration in mg-% of the final measured solution) were A = 0.001 + 0.314C, A = 0.001 + 9.60C and A = 0.002 + 0.69C for acetaminophen, oxyphenbutazone and salicylamide , respect- ively.OH OH -0 0- Scheme 2 Scheme 3ANALYST, MARCH 1986, VOL. 111 321 Table 1. Results for the determination of the investigated drugs in unit doses using the proposed and official meth0ds*~.3~ Results, % (mean k S.D.) Compound/preparation Proposed Official method method t-test F-test Acetaminophen powder . . . . . . . . . . . . . . 100.01 _+ 0.60 Acetaminophen tablets (paracetamol tablets) . . . . . . 99.94 k 1.49 99.46 k 1.76 0.46 1.666 (2.31) (6.39) Salicylamide powder . . . . . . . . . . . . . . 99.95 f 0.48 Salicylamide tablets (Cidal forte tablets) . . . . . . . . 101.1 k 1.07 100.90 k 0.93 0.32 1.324 (2.31) (6.39) Oxyphenbutazone powder . . . . . . . . . . . . 99.97 f 0.43 Oxyphenbutazone tablets (Tandril tablets) . . . . . . 99.69 5 0.86 100.1 * 1.22 0.82 1.33 (2.31) (6.39) * Means of five determinations; results are percentages found with respect to the label claim.Figures in parentheses are tabulated values o f t and F. 1 1 290 330 370 410 450 490 Wavelengthinm Fig. 1. Absorption graphs for the reaction products obtained from: A, acetaminophen; B, salicylamide; and C, oxyphenbutazone The validity of the regression equation was tested by analysing commercial tablets using the proposed and official methods29930 to analyse unit doses of the test drugs (Table 1). Statistical analysis31 of the results revealed that a the 95% confidence level, the calculated t and Fvalues indicate that the proposed method is sufficiently accurate and precise. The proposed method is fairly sensitive, selective and of good reliability owing to the stability of the complex.The results were reproducible. The method could be considered as a general method for determining drug substances having phenolic OH groups. These merits, in addition to the use of simple reagents, suggest its use in drug control laboratories. References 1. 2. 3. 4. 5. 6. Belal, S., El Sayed, M. A. H., El Walily, A., and Abdine, H., J. Pharm. Sci., 1979, 86, 750. Sane, R. T., and Amber Dekar, A. B., Indian Drugs, 1981, 19, 115. Hassan, S. M., Walash, M. I . , El Sayed, S. M., and Abou Ouf, A. M., J. Assoc. Off. Anal. Chem., 1981,64, 1442. Svatek, E., and Hardkava, A,, Cesk. Farm., 1966, 15, 76. Belal, S., El Kheir, A. A., El Shanwani, A. A., Analyst, 1985, 110, 205. Amer, S. M., Ellaithy, M. M., and El Tarrasse, M. F., Pharmazie, 1982, 37, 182.7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22 I 23. 24, 25. 26. 27. 28. 29. 30. 31. Inamdar, M. C., and Kaji, N. N., Indian J . Pharm., 1969,31, 79. Belal, S. F., El Sayed, M. A. H., El Walily, A. M., and Abdine, H., Analyst, 1979, 104,919. Abou Ouf, A., Walash, M. I., Hassan, S. M., and El-Sayed, S. M., Analyst, 1980, 105, 169. Sanghaiv, M. M., Jivani, N. G., and Mluye, P. D., Zndia J . Pharm. Sci., 1977,39, 87. Amer, S. M., Ellaithy, M. M., and El Tarrasse, M. F., Anal. Lett., 1980, 13, 1625. Hanegraff, C., Chastaigner, N., and DeMonrely, E., Ann. Pharm. Fr., 1969,27, 663. Le Pedriel, F., Hangroaff, C., Chastaigner, N., and DeMon- rely, E., Ann. Farm. Fr., 1968,28,227. Gold, V., “Advances in Physical Organic Chemistry,” Volume 7, Academic Press, New York, 1969, p.211. Strauss, M. J., Chem. Rev., 1970, 70, 667. Miller, R. E., and Wyne-Jones, W. F. K., J . Chem. SOC., 1959, 2375. Foster, R., and Mackie, R. K., J. Chem. Soc., 1959, 3508. Foster, R., and Mackie, R. K., Tetrahedron, 1961, 6 , 119. Feur, H., “The Chemistry of Nitro and Nitroso Compounds, Part 2,” Interscience, New York, 1970, p. 329. Glover, D. J., and Kyser, E. G., Anal. Chem., 1968,40,2055. Anas, S . A. H., and Yallop, H. J., Analyst, 1966, 91, 336. English, F. L., Anal. Chem., 1948, 20,745. Zimmermann, W., Hopper-Seylers Z . Physiol. Chem., 1937, 47,245. Kimura, M., Obi, N., and Kawazoi, M., Chem. Pharm. Bull., 1972,20, 452. Burns, L. B., Stedman, R. J., and Tuckerman, M. M., J . Pharm. Sci., 1977,66, 753. Canbeck, T., Svensk. Farm Tidskr., 1950, 54, 225. Kohashi, K . , Tsuruta, Y., Yamaguchi, M., and Ohkura, Y., Chem. Pharm Bull, 1979, 27, 2122. Spiegel, M. R., “Theory and Problems of Probability and Statistics,” McGraw-Hill, New York, 1975, pp. 215 and “The Pharmacopoeia of Japan,” Eighth Edition, Society of Japanese Pharmacopoeia, Tokyo, 1971, p. 674. Davies, 0. L., and Goldsmith, P., “Statistical Methods in Research and Production,” Fourth Edition, Oliver & Boyd, Edinburgh, 1972, p. 178. “British Pharmacopoeia 1980,” Volume 1, HM Stationery Office, London, 1980, pp. 322 and 340. Paper A51204 Received June 7th, I985 Accepted September 25th, I985 259-270.

ISSN:0003-2654    

DOI:10.1039/AN9861100319

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST, MARCH 1986, VOL. 111 323 Spectrophotometric and Fluorimetric Methods for the Determination of lndomethacin C. S. P. Sastry, D. S. Mangala and K. Ekambareswara Rao Foods and Drugs Laboratories, School of Chemistry, Andhra University, Waltair 530 003, A. P., India Three spectrophotometric methods and a fluorimetric method are described for the determination of indomethacin in bulk samples and pharmaceutical preparations based on the formation of coloured species with m-aminophenol - chloramine-T, resorcinol - sodium hypochlorite or phloroglucinol - hypochlorite reagent and a fluorescent species with m-aminophenol - chloramine-T reagent, respectively, under specified experimental conditions. All the methods are simple, sensitive and reproducible to within + I . ~ O / O .Keywords: lndomethacin determination; spectrophotometry; fluorimetry Indomet hacin , [ 1 - (4-chloro benzoyl)-5-methoxy-2-methyl- indol-3-yllacetic acid, is well known for its antipyretic anal- gesic action and is extensively used in rheumatoid arthritis. The methods reported for the determination of indomethacin include spectrophotometric~-3 and fluorimetric methods4 and the official BP procedure.5 The existing spectrophotometric methods are time consuming, tedious and require some preliminary treatment. This paper describes three simple spectrophotometric (visible region) methods and one fluor- imetric method using pairs of reagents, rn-aminophenol - chloramine-T, resorcinol - hypochlorite, phloroglucinol - hypochlorite or rn-aminophenol - chloramine-T, under speci- fied conditions.Experimental Apparatus A Systronics Model 105 (Mk. 1) spectrophotometer with 1-cm path length cuvettes, a Perkin-Elmer Model 203 fluorescence spectrophotometer with the sensitivity setting adjusted to 10 and a Systronics Model 305 pH meter were used for absorbance, fluorescence and pH measurements, respec- tively. Reagents All the reagents were of analytical-reagent grade and all solutions were prepared in doubly distilled water. All the pharmaceutical preparations used were available commer- cially. Aqueous solutions of m-aminophenol (mAP, 0.1% in 0.025 M HCl), phloroglucinol (0.2%), resorcinol (0.270)~ chlor- amine-T (CAT, 0.03 M), sodium hypochlorite (OC1-, 0.05 M), HC1 (2 M) and potassium acid phthalate (0.05 M) were prepared. Standard drug solution.A 50-mg amount of BP-grade indomethacin (supplied by Themis Chemicals Ltd. , Bombay, India) was initially dissolved in 10 ml of 1.25 M NaOH solution and then diluted to 250 ml in a calibrated flask. Working solutions for methods A and D were prepared by suitable dilution of the stock standard solution. The solutions were stable for 2 h. Sample solution. Twenty capsules were emptied and pul- verised and an amount equivalent to 50 mg of indomethacin was taken, dissolved as described above and filtered. The solution was stable for 2 h. Spectrophotometric Procedures A. With mAP - CAT reagent A 15-ml volume of potassium hydrogen phthalate, 3 ml of CAT and 3 ml of mAP solution were placed in a 25-ml calibrated flask. A 1.04.0-ml portion of indomethacin (100 pg ml-1) solution and the requisite volume of distilled water were added to make the total volume 25 ml.The pH of the resulting solution was between 4 and 5.0. The absorbance of the coloured species was measured at 490 nm within 3 rnin against a reagent blank prepared in a similar manner. The indomethacin concentration was calculated from a calibration graph prepared with a standard solution under identical conditions. B or C. With resorcinol or phloroglucinol - OCI- reagents Aliquots of 1.0-4.0 ml of indomethacin solution (200 pg ml-1) were placed in a 25-ml calibrated flask containing 1 ml of 2 M HC1, mixed well (for 1 min), 1 ml of OC1- was added and the mixture was allowed to stand for 3 min for resorcinol or 2 rnin for phloroglucinol. A 3-ml volume of resorcinol or phloroglu- cinol was added after this period and the absorbance of the coloured species was measured at 460 nm after 10 rnin (the stability periods were as follows: B, resorcinol - OC1-, 5-20 min; C, phloroglucinol - OC1-, 10-35 min) against the corresponding reagent blanks prepared in a similar manner.The indomethacin concentration was obtained from calibra- tion graphs obtained under identical conditions. Fluorimetric Procedure D. With mAP - CAT reagent A 0.2-3.0-ml portion of indomethacin (10 pg ml-1) was placed in a 10-ml calibrated flask containing 0.5 ml of 2 M HCl, 2 ml of CAT and 2 ml of mAP solution and diluted to the mark with distilled water. The fluorescence of the solution was measured between 10 and 120 min at excitation and emission wave- lengths of 465 and 490 nm, respectively, against the reagent blank prepared in a similar manner.The concentration of indomethacin was calculated from a calibration graph. Results and Discussion Of the various combinations of phenols (phenol, catechol, resorcinol, pyrogallol and phloroglucinol) , aminophenols (0-, m-, p- and p-N-CH3) or phenylenediamines [o-, rn-, p - , and p-N7N-(CH3)2] and oxidising agents [CAT, OC1-, 104- , 103-, Cr(VI), Fe(III), S208*- , N-bromosuccinimide and324 ANALYST, MARCH 1986, VOL. 111 Table 1. Optical characteristics, precision and accuracy Method Parameter Beer’slawlimits/pgml-1 . . . . . . Molar absorptivity/l mol-1 cm-1 . . , . Sandell’s sensitivity/pg cm-2 per 0.001 absorbance unit . . . . Intercept . . . . . .. . . . . . Correlationcoefficient . . . . . . Relative standard deviation, % . . . . Range of error, % (95% confidence limits) . . . . . . Slope . . . . . . . . . . . . A 4.0 - 16 5.72 x 103 0.063 6.32 x 10-4 -3.49 x 10-3 0.999 1.39 k 1.46 B 8.0 - 24 2.27 x 103 0.16 2.41 x 10-4 3.00 x 10-3 0.998 1.65 k1.73 C D 5.5 - 24 3.18 x 103 0.2 - 3 - 0.11 - 3.42 x 10-4 0.3186 0.1698 2.03 x 10-3 0.999 0.999 1.55 1.32 k 1.62 k 1.39 Table 2. Assay and recovery of indomethacin in dosage forms Amount found/mg Labelled Proposed method Recovery, Yo amount/ Reported Capsules mg A BorC D method A B or C* D Hyderabad) . . . . 25 23.44 23.50 (B) 23.52 23.45 97.8 98.4 (B) 98.5 Idicin(IDPL, . . . . Microcid Inobid (Thomis, (Microlabs, Madras) 25 24.23 24.18 (C) 24.30 24.21 98.2 98.2 (C) 98.7 Bombay) .. . . 75 73.32 73.30 (B) 73.36 73.29 97.6 98.6 (B) 99.1 H2021 tried for developing the colour or fluorescence under acidic or alkaline conditions at laboratory temperature, mAP - CAT was found to be superior. Of the other combinations, resorcinol or phloroglucinol and OC1- were found to be suitable as chromogenic reagents. The optimum conditions for each method (A-D) were established after a thorough systematic study of the para- meters such as acid strength (pH), reagent concentration and order of addition of the reagents. Potassium hydrogen phthalate solution was found to be necessary only in method A for maintaining the resulting pH between 4 and 5. Even trace amounts of miscible alcohols such as methanol and ethanol were found to quench the fluores- cence intensity.The chromophore or fluorophore in methods A-D was partially extractable into butan-1-01 and not extractable into chloroform. The excitation spectra of the fluorescent species showed three maxima, at 245,305 and 465 nm. The peaks in the UV region were not considered as the blank interferes. The optical characteristics such as Beer’s law limits, molar absorptivity and Sandell’s sensitivity for each method are given in Table 1. The slopes, intercepts and correlation coefficients obtained by linear least-squares treatment6 of the results for the systems involving indomethacin with the mentioned reagents are also presented in Table 1. The reproducibility of the methods was found by measuring the absorbances (in methods A, B and C) or fluorescence intensity (in method D) of six replicate samples containing a known amount of drug (400 pg per 25 ml in A, B and C and 20 pg per 10 ml in D) and the results obtained are given in Table 1.The accuracy of the methods was further confirmed by adding known amounts of indomethacin to previously analysed H I OH- CH&OOH H3C0 CHZCOOH H3C0 I R‘ II I CHzCOOH 0 CHZCOOH IV I l l rn-Aminophenol: R = NH; R’ = H Phloroglucinol : R = 0; R’ = OH Resorcinol : R = 0; R’ = H Scheme 1. Reaction mechanismANALYST, MARCH 1986, VOL. 111 samples and the recoveries obtained are given in Table 2. The results obtained by the proposed and reported1 methods for indomethacin in dosage forms are also included in Table 2. The ingredients usually present in pharmaceutical prepara- tions of indomethacin, such as glucose, lactose, sodium metabisulphite, sodium chloride, magnesium stearate, starch, talc, sodium citrate and other analgesics such as paracetamol, phenacetin and analgin, did not interfere in the proposed methods.Mechanism The species reacting with the proposed reagents (Scheme 1) appears to be 5-methoxy-2-methylindol-3-ylacetic acid (11), the hydrolysis product of indomethacin (I) ,’ as the indo- methacin sample dissolved only in alkali and not in ethanol produces colour or fluorescence. The failure of tryptophan and indole-3-acetic acid to develop colours with the proposed reagents indicates the necessity to have a 5-OMe group in the indole moiety. Compound I1 reacts initially with the oxidant (CAT or OC1-) to produce the highly reactive and less stable p-benzoquinone monoimine derivative (111), as with p-phene- tidin (p-ethoxyaniline) .8 Compound I11 may react further with coupler (m-aminophenol, resorcinol or phloroglucinol) to give 7-substituted-I11 such as p-N-acetylbenzoquinone monoimine and cysteine.9 In conclusion, the proposed methods are simple, sensitive, selective and can be used for the routine determination of indomethacin in pharmaceutical preparations.325 The authors are grateful to the authorities of Andhra University and also to the Council of Scientific and Industrial Research, New Delhi, for awarding a fellowship to K. E. R. References 1. 2. 3. 4. 5. 6. 7. 8. 9. Baggi, T. R., Mahajan, S. N., and Rao, G. R., Indian 1. Pharm. Sci., 1976,38, 101. Sanghavi, N. M., and Kamala, S . , Indian J. Pharm. Sci., 1978, 40, 71. Peterkova, M., Kakac, B., and Matousova, O., Cesk. Farm., 1980, 29, 73. Garcia, C. R., Lopez, A. A., and Benet, L. Z . , Rev. SOC. Quim., 1980, 24, 68. “British Pharmacopoeia 1980,” Pharmaceutical Press, London, 1980, p. 239. Pattergill, M. D., and Sands, D. E., J. Chem. Educ., 1979,58, 244. Hajratwala, B. R., and Dawson, J. E., J. Pharm. Sci., 1977,66, 27. Davis, D. R., Fogg, A. G., Thorburn Burns, D., and Wragg, J . S . , Analyst, 1974, 99, 12. Blair, 1. A., Boobis, A. R., Davis, D. S., and Cresh, M., Tetrahedron Lett., 1980, 4947. Paper A51254 Received July 15th, 1985 Accepted September 4th, 1985

ISSN:0003-2654    

DOI:10.1039/AN9861100323

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST, MARCH 1986, VOL. 111 327 Spectrofluorimetric Determination of Zinc with Pyrocatechol- 1 =aldehyde 2-Pyridylhydrazone Ana M. Afonso, Jose J. Santana and Francisco Garcia Montelongo Department of Analytical Chemistry, University of La Laguna, La Laguna, Tenerife, Spain Pyrocatechol-1 -aldehyde 2-pyridyl hydrazone was synthesised and its ionisation constants spectrophoto- metrically determined. A procedure wasdevelopedforthe spectrofluorimetricdetermination of 12-250 ng ml-1 of zinc in 50% V/Vethanol -water medium, acetate-buffered to apparent pH 5.5 (kex. = 398 nm, kern. = 521 nm), using the above reagent. Interferences were evaluated and the procedure was applied satisfactorily to the determination of zinc in potable tap water. Keywords: Zinc determination; spectrofluorimetry; pyrocatechol- I-aldeh yde 2-pyridylh ydrazone Hydrazones have been widely used in the spectrophotometric determination of metal ions1 but only in recent years have they found application as fluorogenic reagents for metals.2 Haddad et al.3 determined cobalt fluorimetrically after extraction of its ternary complex with 2-pyridylaldehyde 2-pyridylhydrazone and eosin, and 2-pyridylhydrazones derived from 3-hydroxy- 2-pyridylaldehyde, benzyl 2-pyridyl ketone, 2-dipyridyl ketone, salicylaldehyde , P-resorcylaldehyde and 2-hydroxy-l- naphthaldehyde have been used in the fluorimetric determina- tion of Al(III), Ga(III), In(III), Sc(II1) and Zn(II).4-8 Other 2-pyridylhydrazones have been used as substrates for the kinetic and catalytic fluorimetric determination of several metal ions.9-11 In this paper the characteristics and analytical properties of pyrocatechol-1-aldehyde 2-pyridylhydrazone (PCAPH) are described, and a rapid and simple method for the spectrofluo- rimetric determination of 12-250 ng ml-1 of zinc based on its fluorescent complex with PCAPH is reported. Experimental Apparatus All fluorescence measurements were made with a Perkin- Elmer MPF-44A recording spectrofluorimeter equipped with a 150-W Osram XBO xenon arc lamp, a DSCU-1 corrected spectra unit (0.5% Rodamine B in ethylene glycol as the reference), a UDR-3 digital read-out, a Selecta Frigitherm ultrathermostat and 1-cm quartz cells.The emission intensity measuring system of the spectrofluorimeter was calibrated daily by using the Perkin-Elmer set of fluorescent polymer blocks.A Radiometer PHM84 digital pH meter with glass and calomel electrodes was also used. pH values in 50% V/V ethanol - water were not corrected and are referred to as pH*. Reagents Analytical-reagent grade chemicals and de-ionised water were used throughout without further purification. Pyrocatechol-1-aldehyde 2-pyridylhydrazone. Synthesised by condensation of pyrocatechol-1-aldehyde and 2-pyridylhy- drazine.13 The crude product was washed with diethyl ether until only one spot was revealed by thin-layer chromatography [silica gel G (Merck); benzene - ethanol (9 + 1); iodine vapour as detection agent]. The crystals obtained (yield 85%) melted at 192-194 "C. High-resolution mass spectrometry showed a parent molecular ion ( M + ) at mlz 229.083 and C12H1102N3 as the most probable composition.A 1.0 x 10-2 M solution of the reagent in absolute ethanol was prepared and diluted as required. Zinc perchlorate standard solution, 0.1 M. Prepared from zinc oxide by perchloric acid treatment and standardised complexometrically. The ionic strength was controlled by adding suitable amounts of 2.5 M sodium perchlorate solution. A pH 4.5 acetic acid - sodium acetate (0.1 M) buffer solution was used as indicated. Procedure Determination of zinc To a solution of zinc (up to 10 ml) containing 0.3-25.5 pg of zinc in a 25-ml calibrated flask add 3 ml of the acetic acid - sodium acetate buffer solution, 2 ml of 2.5 M sodium perchlorate solution, 5 ml of a 4.15 X 10-5 or 1.7 X 10-4 M ethanolic solution of PCAPH (according to the expected zinc concentration, see later) and 7.5 ml of absolute ethanol and dilute to volume with de-ionised water.Measure the fluores- cence at 521 nm using excitation at 398 nm, against a reagent blank. Determine the amount of zinc present in the sample from calibration graphs prepared under the same experimen- tal conditions. If copper is present, add 2 ml of 10-2 M sodium thiosulphate solution and a few crystals of ascorbic acid. If aluminium or iron(II1) is present, add 0.25 ml of 0.25 M ammonium fluoride solution. Determination of zinc in potable tap water Analyse suitable aliquots according to the above method. Results and Discussion Characteristics of the Reagent The infrared spectrum of PCAPH (KBr pellet) was obtained and bands were assigned as follows: phenolic OH (3400 cm-l), -NH- (3250 cm-I), pyridinic -C=N- (1620 cm-l), >C=N- (1286 cm-1) and benzylic ; CH-.(750 cm-1).The NMR spectrum (dimethyl sulphoxide-d6, tetramethyl- silane) at 90 MHz was as follows: 6 (benzenic moiety) = Hz, H5); 6 (pyridinic moiety) = 6.95 (lH, s, H3), 7.08 (lH, s, H5), 7.44 (lH, s, H4), 8.09 (lH, s, H6); 6 (benzylic =CH-) = 8.99 (lH, s). PCAPH is very soluble in ethanol, dimethyl sulphoxide, dimethylformamide and concentrated alkalis, soluble in water and slightly soluble in benzene and diethyl ether. Aqueous ethanolic solutions of the reagent slowly hydrolyse when the pH is less than 1 or higher than 9. PCAPH behaves as a tribasic substance with protonation of the heterocyclic nitrogen atom and deprotonation of the meta- and ortho-hydroxy groups.The corresponding ionisation constants were calculated from the variation of the absorbance at different wavelengths with pH by application of the method 6.80(1H,d,J=7H~,Hg),7.26(1H,~,H4),7.64(1H,d,J=7ANALYST, MARCH 1986, VOL. 111 I 1 I I 260 340 420 500 580 6 Wavelengthinrn Fig. 1. Corrected excitation (1-3) and emission (1'-3') spectra of PCAPH in different media. 1,l' = Neutral medium; 2,2' = acidic medium; 3,3' = basic medium. C, = 6.0 x 10-5 M; solvent = 50% V/V ethanol - water Table 1. Spectral characteristics of the fluorescent complexes of PCAPH pH* = lt pH* = 5$ Complex of he,,/nm h,,./nm Zn(I1) . . . . - - Ca(I1) . . . . 385 440 Mg(I1) , . . . 384 438 Cd(I1) . . . . 388 440 Al(II1) . . . . 405 492 Ga(II1) . .. . 400 500 La(II1) . . . . 382 452 In(II1) . . . . - - hex./nm 400 3 87 386 396 390 400 376 - t Perchloric acid. Acetic acid - sodium acetate buffer solution. hern./nm 494 482 489 490 486 510 424 - of HniliEova and Sommer.14 The mean values found were pKal = 5.42 k 0.06 and pKa2 = 8.42 k 0.04; pKa3 could not be calculated because of the slow hydrolysis of the reagent at higher pH. The fluorescence maxima of PCAPH at pH* 7.0 (hex. = 299 nm, hem, = 366 nm), Fig. 1, show a bathochromic change (Aex. = 311 nm, hem, = 394 nm) in acidic media, the fluorescence emission intensity increasing as the pH decreases. This may be due to protonation of the ring nitrogen atom, which stabilises its free electron pair and therefore stabilises the corresponding excited states. In ammoniacal media PCAPH also shows a bathochromic change (Aex, = 380 nm, kern, = 490 nm).The reactions of PCAPH with 50 metal ions at pH* 1.0 and 5.0 were investigated. Only Cd(II), Mg(II), Ca(II), Ga(III), In(III), Al(II1) and La(II1) showed fluorescence under UV light. The spectrofluorimetric characteristics of these com- plexes are given in Table 1. Reaction with Zinc With Zn(I1) ions PCAPH forms a yellow - green complex with strong yellow fluorescence under UV irradiation. The correc- ted excitation and emission spectra are shown in Fig. 2; its highest fluorescence emission intensity is at pH* 5-6 (Fig. 80 8 >r 4- .- v) C 60 .- a C (u v) $ 40 3 a m 0) - Lc .- 4- - a 20 1 i Wavelengthinm Fig. 2. Corrected (1) excitation and (2) emission spectra of the ZntII) - PCAPH complex in a 50% V/V ethanol - water medium at pH 5.5 (acetic acid - acetate buffer).C,, = 5.3 x 1 0 - 6 ~ ; CR = 1.0 x 10-4 M L 4 6 8 10 PH Ethanol,% V/V 1 0 I J 20 40 60 80 Fig. 3. Effect of (A) pH* and (B) amount of ethanol on the fluorescence intensity of the Zn(I1) - PCAPH complex. (hex, = 398 nm, hem, = 521 nm) 3A). A pH 4.5 acetic acid - sodium acetate (0.1 M) buffer solution, which gives pH* 5.5 in the 50% V/V ethanol - water medium used, was employed in all subsequent work. All fluorescence measurements were made at 25 & 0.1 "C, at which temperature the fluorescence emission remains stable for at least 1 h. The fluorescence intensity decreases linearly at ca. 0.38% "C-1 as the temperature increases between 35 and 55 "C. This effect might be due to the easier non-radiative deactivation of the excited singlet state as the temperature increases.The fluorescence intensity is critically dependent on the concentration of ethanol, increasing steadily with the percen- tage of ethanol in the medium (Fig. 3B). A 50% V/V ethanol - water medium was chosen as the best balance between fluorescence emission and volumes of aqueous solution of sample and reagents to be added. Several sodium and potassium salts were tested at different concentrations in order to study the influence of the ionic strength on the fluorescence emission. The results show that variations in the ionic strength and the concentration of the electrolyte used to control it do not influence the fluorescence emission. In all subsequent work a 0.2 M ionic strength, adjusted with sodium perchlorate, was used.ANALYST, MARCH 1986, VOL.111 329 Table 2. Interference levels of foreign ions on the spectrofluorimetric determination of 60 ng ml-1 of Zn(I1) with PCAPH Tolerance ratio, mlm 6500 6300 3300 2800 1000 800 500 200 50 15 10 5 3 2 1 0.5 Strong interference Ion added C1- S042- Mg2+ Ca2+ NO3-, F-, ascorbic acid S2032- C032- I-, HP042 - , B40,2- Br-, SCN- Mn2+ Fe3+ As(III), U(V1) Hg(I), Pb(II), TW) Cd(II), Sn(II), La(III), Ag(I), Te(IV), tartrate Mn(VII), Cr(VI), Ce(IV), Zr(IV), V(IV), Se(IV), Au(III), Ru(III), Pd(II), C2042- V(V), Th(IV), In(III), Sb(III), Fe(II), Pt(II), Al(III),EDTA,EGTA,DCTA Mo(VI), Ti(IV), Ga(III), Bi(III), Co(I1) Variation of the order of addition of sample and reagents does not have a marked influence on the fluorescence emission.The stoicheiometry of the complex was studied under the established experimental conditions by the continuous varia- tions and molar ratio methods. A metal to ligand ratio of 1 : 1 was found. The effect of the reagent concentration on the fluorescence intensity of solutions containing 35 ng ml-1 of zinc(I1) was studied under conditions similar to those recommended under Experimental. The fluorescence intensity increases with increasing the reagent concentration up to 5.0 x 10-5 M, and remains constant between 5.0 x 10-5 and 2.5 x 10-4 M. At higher reagent concentrations, the fluorescence decreases markedly, mainly owing to autoabsorption phenomena. Spectrofluorimetric Determination of Zinc Under the experimental conditions outlined in the recommen- ded procedure, there is a linear relationship between emitted fluorescence intensity and Zn(I1) concentration in the ranges 12-80 and 80-250 ng ml-1, using 5 ml of 4.15 X 10-5 or 1.7 X 10-4 M solutions of the reagent, respectively.The detection limit, as defined by IUPAC,15 was determined to be 5.5 ng ml-1. When the developed method was applied to two series of eleven samples containing 35 and 170 ng mi-1 of zinc, relative errors of 2.80 and 1.14% (95% confidence limits), respec- tively, were obtained. A study of the effect of several ions on the determination of 60 ng ml-1 of zinc was carried out by first applying the recommended method to solutions containing a 10 000-fold (mlm) ratio of interferent to zinc and, if interference occurred, reducing this ratio until interference ceased.Higher ratios were not tested. The criterion for interference was a variation in the concentration found for zinc of more than ?4%0 from the value taken. The results are shown in Table 2. Copper can be tolerated up to a Cu to Zn ratio of 7 when it is masked with thiosulphate and ascorbic acid, Al(II1) up to an A1 to Zn ratio of 1 and Fe(II1) up to an Fe to Zn ratio of 40 in the presence of fluoride as the masking reagent. The interferences in the proposed method come mainly from ions such as Mg(II), Ca(I1) and Al(III), which also form Table 3. Spectrofluorimetric analysis of potable tap water for zinc Zn foundhg ml-1* Sample No. PCAPH (range) AAS 1 104( 102-105) 104 2 277( 266-284) 277 3 1295(1258-1319) 1294 4 147( 145-1 49) 147 5 160( 158-162) 160 6 271(264-277) 273 * Mean of three determinations. fluorescent complexes with PCAPH as previously shown, and from those giving coloured complexes with the reagent such as Fe(II), Fe(II1) and Ni(I1).Applications The method developed was applied to the determination of zinc in samples of tap water from the distribution systems in several cities of the Canary Islands, collected and preserved as recommended.16 The results are shown in Table 3, where they are compared with those obtained by the standard AAS method.17 The zinc concentrations found are well under the tolerance limit set by the Spanish Food Directorate ( S 1.5 mg 1-1). The authors acknowledge financial support of this work by CAICYT (Spain), grant No.4122/79. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. References Katyal, M., and Dutt, Y., Talanta, 1975, 22, 151. Singh, R. B., Jain, P., and Singh, R. P., Talanta, 1982,29,77. Haddad, P. R., Alexander, P. W., and Smythe, L. E., Talanta, 1976, 23, 275. Laserna, J. J., Navas, A., and Garcia SBnchez, F., Anal. Lett., 1982, 14A, 833. Laserna, J. J., Navas, A., and Garcia SBnches, F., Anal. Chim. Acta, 1980, 121, 295. Laserna, J. J., Navas, A., and Garcia SBnchez, F., Microchem. J., 1982, 27, 312. Cano, J. M., Trujillo, M. L., and Garcia de Torres, A., Anal. Chim. Acta, 1980, 117, 319. Sommer, L., Maung-Gyee, W. P., and Ryan, D. E., Scr. Fac. Sci. Nut. Univ. Purkynianae Brun. 2, 1972, 2, 115; Chem. Abstr., 1973, 79, 121493~. Garcia SBnchez, F., Navas, A., and Laserna, J. J., Anal. Chem., 1983, 55, 253. Rubio, S., G6mez-Hens, A., and Valcarcel, M., Anal. Lett., 1984, 17A, 651. Rubio, S., Gomez-Hens, A., and ValcBrcel, M., Analyst, 1984, 109, 717. Afonso, A. M., Santana, J. J., Gonzalez, M. P., and Garcia Montelongo, F., Mikrochim. Acta, 1984,II, 53. Odashima, T., Anzai, P., and Ishii, H., Anal. Chim. Acta, 1976, 86,231. HniliCkovB, M., and Sommer, L., Talanta, 1966, 13, 667. Irving, H. M. H. N., Freiser, H., and West, T. S . , Editors, “IUPAC Compendium of Analytical Nomenclature, Definitive Rules, 1977,” Pergamon Press, Oxford, 1978. American Public Health Association, American Water Works Association and Water Pollution Control Federation, “Stan- dard Methods for the Examination of Water and Wastewater,” Fourteenth Edition, American Public Health Association, Washington, DC, 1976, p. 38. Reference 16, p. 143. Paper A51220 Received June 20th, I985 Accepted September 20th, I985

ISSN:0003-2654    

DOI:10.1039/AN9861100327

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST, MARCH 1986, VOL. 111 331 Determination of Theaflavins in Tea Solution Using the Flavognost Complexation Method Michael Spiro* and William E. Price Department of Chemistry, Imperial College of Science and Technology, London SW7 ZAY, UK Several aspects of the flavognost method for determining theaflavins (TF) in tea solutions have been quantitatively investigated. Double extraction experiments showed that the partition coefficient of TF between water and isobutyl methyl ketone (IBMK) is 4.15 and approximately independent of temperature. The efficiency of the extraction step using 2 volumes of IBMK to 1 volume of tea infusion is then equal to 89%. A structure has been proposed for the green complex formed between the extracted TF and flavognost in a 25% VNIBMK - 75% WVethanol mixture, and the equilibrium constant for its formation at 25 "C was found to be 0.073.The normal analytical procedure thus leaves 45% of TF uncomplexed. As only 85% of the TF in a sample of tea solution eventually appears in the form of the green flavognost complex, it is important not to vary the standard analytical procedure. Keywords: Theaflavins determination; flavognost; Hilton method; borate complexation; tea solutions Theaflavin is an important constituent of black tea. Although present at only 0.5-2% m/m in the leaf,l it is responsible for much of the attractive red colour of tea infusions and a correlation exists between the market price of a given black tea and its theaflavins (TF) content.2.3 The latter comprises both theaflavin itself and its mono- and digallates.The earlier method for total TF determination, the solvent-extraction procedure of Roberts and Smith,1>4 has been criticised on several counts.5 In many laboratories the determination of TF is now carried out by the simple flavognost method,6 which was further developed by Hilton.778 Here the TF is extracted by an organic solvent such as isobutyl methyl ketone (IBMK) and is then complexed with 2-aminoethyl diphenylborate (flavognost). The absorbance of the green complex formed is measured spectrophotometrically . It was the purpose of this study to examine several aspects of this method in more detail. It should be mentioned that although individual theaflavins can now be determined by HPLC79J0 this method requires prior work-up of samples as well as more expensive equip- ment, and is less suitable for small or dilute samples.ll Experimental The general procedure involved the infusion of black Kap- chorua tea (x g) in 200 cm3 of distilled water at 80 "C for 30 min to reach partition equilibrium.The pH of the infusion was typically 4.8. Then an aliquot (V cm3) of the tea solution was removed so as to exclude any leaf, and shaken vigorously for 1 min in a stoppered tube with n times its volume of IBMK (BDH Chemicals, AnalaR grade) (Table 1). The layers were allowed to separate; if the shaking had produced a suspension, separation was achieved by centrifuging for 5 min in a Gallenkamp Junior centrifuge (setting 1). With a clean syringe, 2 cm3 of the IBMK layer were removed and added to v cm3 (usually v = 2) of an ethanolic 2% mlV solution of flavognost (Koch-Light) and (6 - v) cm3 of absolute ethanol (J.Burrough). The green TF - flavognost complex was allowed 15 min to develop completely before its absorbance (A) was measured in a 1 cm cuvette at 625 nm with a Unicam SP 1800 spectrophotometer. The blank cuvette held a mixture of IBMK - flavognost solution - ethanol [2 : v : (6 - v)]. The TF content of the aqueous solution was calculated from Beer's law by the equation [TF](mmol dm-3) = fnA X 9/375 = 0.024fnA (1) where n is the dilution factor in the extraction andfis a molar * To whom correspondence should be addressed. Table 1. Summary of principal symbols Symbol Unit Significance X g Mass of tea leaf infused in 200 cm3 of water V cm3 Volume of aliquot of tea infusion removed n - Volume of IBMK/ volume of aqueous aliquot V cm3 Volume of flavognost solution added to 2 cm3 IBMK layer conversion factor taken as 47.9 for n = 1.8 A modified value applies for other values of n, as explained below.The figures 9 and 375 take account of the original conditions7 in which 9 g of tea leaf were infused in 375 cm3 of water. The tea employed was Kapchorua (Pekoe Fannings and Dust) whose various size fractions have all been shown to contain 0.022 k 0.001 mol kg-1 of TF.11 Results and Discussion Effect of Sample and Solvent Temperature Three sets of experiments were performed to test whether the analytical result depended upon the temperature of either phase in the solvent extraction step. Unsieved Kapchorua Pekoe Fannings tea was employed, and x = 10 g, V = 5 cm3, n = 2 and v = 2 cm3.In experiments of type A, the hot (80 "C) 5-cm3 sample of tea solution was added to 10 cm3 of IBMK that had been pre-heated to 80 "C in the thermostated bath. The mixture was shaken vigorously and returned to the 80 "C bath. The layers separated easily under these conditions and no centrifuging was necessary. In type B experiments, the hot tea sample was shaken with cold (room temperature) IBMK, with all subsequent steps being carried out at room tempera- ture. Analyses of type C were carried out completely at room temperature, the filtered tea liquor having been cooled to 24 "C before a sample was removed. A small amount of tea cream appeared at the interface only. The resulting absorbances of the TF - flavognost complex in duplicate experiments were identical within experimental error: 0.294, 0.293 (type A); 0.294,0.294 (type B); 0.295,0.294 (type C).The temperatures of the two phases are therefore not of importance in the procedure. For convenience, in subsequent work the hot tea sample was added to room-temperature IBMK. Efficiency of the Solvent Extraction Step Although Hilton has stated7 that a single IBMK extraction does not exhaustively remove the theaflavins, this point is332 ANALYST, MARCH 1986, VOL. 111 Fl Initially Claq First extraction c20rg c2aq Second extraction Fig. 1. Schematic illustration of successive extraction steps normally ignored and has never been quantified. For this reason the partition coefficient (Kp) of TF between water and IBMK has now been determined.This was performed by carrying out two successive extractions as depicted schematic- ally in Fig. 1. The TF concentration in the original tea sample, coaq, is reduced to claq after equilibration with n times its own volume of IBMK and to c2aq after further equilibration with n times its volume of fresh IBMK. The concentration of TF in the two organic layers, cprg and c p g , can be determined by complexation with flavognost. Any uncertainty about the exact value of the molar absorptivity of the complex is removed by considering only the ratio cl~~g/c2~rg = r. From the conservation of mass: coaq = claq + nclorg . . . . . . (2) claq = c2aq + nc20rg . . . . . . (3) If the partition coefficient is the same in both extractions: It follows that Hence from equations (3), (4) and ( 5 ) : r = 1 + nKP .. . . . . . . (6) whence Kp = (r- l)/n. The theory shows how values of Kp can be evaluated by performing two successive extractions with IBMK. The results of various experiments using unsieved Kapchorua Pekoe Fannings tea and v = 2 cm3 are listed in Table 2. Because the absorbances after the second extractions are small, most Kp values have an attached uncertainty of about k12%. Within these limits Kp is reproducible and shows no significant variations with the TF concentrations of the infusion (x), the ratio of IBMK to water in the solvent extraction step ( n ) , or the pH of the tea infusion over the range 4.8 (its normal value) down to 2.9 (lemon tea). The mean Kp value over all these experiments is 4.15 _+ 0.l5.This result has an important implication for the normal flavognost method based on a single extraction. From equations (2) and (4) it follows that coaq/clorg = n + (I/Kp) . . . . . . (7) Were Kp infinitely large, the extraction would be complete and coaq/(clorg)compl, would equal n . As Kp is finite, Kp = Clorg/Claq = c20rg/c2aq . . . . (4) r = cl0~g/c2org = claq/c2aq . . . . ( 5 ) nKP * * (8) - n - clo'g (Clorg)compl. n + (1lKp) nKp + 1 which equals 80.6 (k O.6)% when n = 1 and 89.2 (k 0.4)% when n = 2. The extraction efficiencies are therefore functions of n. The conversion factor in equation (1) must vary accordingly, and Hilton's value forfof 47.9 obtained with n = 1 has to be changed to 47.9 x (80.6h9.2) = 43.3 with n = 2. This gives [TF](mmol dm-3) = 0.024 x 43.3 x 2A = 2.08A .. (9) Although the extraction is not exhaustive, the partition Table 2. Partition coefficients for theaflavins between aqueous tea infusions and IBMK X k 10 10 10 10 20 20 20 10 10 10 10 n 2 2 2 2* 1 2 3 2-t 2t 21 29 A1 1st extn. 0.288 0.318 0.300 0.294 0.83 0.462 0.339 0.387 0.380 0.373 0.368 A2 2nd extn. 0.026 0.034 0.030 0.031 0.145 0.057 0.029 0.046 0.042 0.041 0.041 KP 5.0 4.2 4.5 4.2 4.7 3.6 3.6 3.7 4.0 4.05 4.0 * Extracted with hot (80 "C) IBMK. t Tea infusion containing a citric acid (0.32 M) + NaOH (0.11 M) $ Tea infusion containing the same citrate buffer as above, with the 0 Tea infusion containing 0.11 M citric acid (pH 2 at the start of buffer of ionic strength 0.11 M and pH 3 at 80 "C. sample neutralised with NaOH to pH 4.8 before analysis.infusion and pH 2.9 at the end). coefficient has been shown to be independent of various experimental parameters. The flavognost method therefore retains its usefulness for comparing data from different tea infusions provided the same value of n is employed or else the appropriate molar conversion factor f. Calibration of the method must be effected by dissolving a known amount of TF in the aqueous phase and not directly in IBMK. Stability Constant of the TF - Flavognost Complex Although the flavognost method depends on the formation of a stable green complex in 25% V/V IBMK - 75% V/V ethanol, no information has been available on its formula or stability constant. It is well known that boron acids and borates form complexes with organic substances containing cis-l,2- dihydroxy groups; with various polyols both 1: 1 and 1 : 2 complexes are formed12 while only 1: 1 complexes are produced with catechols and their derivatives.13 This latter situation would apply to theaflavin whose only pair of cis-l,2- dihydroxy groups is attached to a benzene ring.With a diphenyl-substituted boron compound, such as flavognost , only a 1 : 1 complex is possible, and so the complexation reaction may be written + HOCH2CHZNH3+.. .. .. (10) The suggested structure of the boron complex follows similar structures proposed by workers in the field712J3 and one such structure has received confirmation in a Raman study.14 Theaflavin gallates, however, also contain 3,4,5- trihydroxybenzoyl groups that could themselves complex with flavognost.A more general reaction scheme should therefore be considered: TF + mB complex + HOCH2CH2NH3+ . . (11) co-cx bo-mc, c, cx where B represents flavognost and bo is its initial concentra- tion in the IBMK - ethanol mixture; co is the initial concentration of TF in the same mixture and c, the equilib-ANALYST, MARCH 1986, VOL. 111 333 rium concentrtion of the green complex. Appreciable ion pairing of the two products in equation (10) is unlikely for such dilute solutions in a mainly ethanolic solvent.15 Provided only one type of complex is formed, its stability constant is thus given by . . . . . . (12) cx2 (co- CX) (bo - mcxP K = It may be noted that bo >> cx. Experiments were carried out with Kapchorua Pekoe Dust (250-710 pm) and x = 4 g, V = 20 cm3, n = 1, and six different values of v ranging from 0.1 to 1.5 cm3 of an ethanolic 1 % m/V (= 0.0445 M) solution of flavognost.A further 2-cm3 sample of the IBMK layer was analysed using v = 2 cm3 of an ethanolic 2% m/V solution of flavognost to provide a mixture with a sufficient excess of flavognost to convert almost all the TF into the complex. For these experiments the flavognost had been dried in vacuo at 100 “C and the ethanol was of AnalaR grade. The mixtures were placed in a thermostated bath at 25 “C for 15 min before the absorbances of the complex were read at 625 nm. The results are listed in Table 3. The experiment with 2% mlVflavognost yielded 89.3 p~ as a first approximation for co. When this value is inserted into the equilibrium equation (12) with m = 1, the K values become constant only at the higher values of bo.This trend largely disappears when co is increased by several per cent., the best fit being obtained near co = 94 VM. All the K values, except that at the lowest concentration, are now in agreement when one makes due allowance for the sensitivity of K to small experimental errors in cx. The mean K value is 0.073 with a standard deviation of the mean of 0.003. It can be seen that similar calculations with m = 2 and rn = 1/2 produce K values with pronounced monotonic trends (falling with increasing c, for m = 2, and rising for m = 1/2), and these cannot be removed by any reasonable adjustment of co. The following conclusions may now be drawn. (1) The fact that m = 1 shows that flavognost complexes only with the cis-1,2-dihydroxy group of theaflavin itself but not with the gallate side groups.This point was confirmed by some further complexation experiments carried out with small samples of ungallated theaflavin kindly donated by Mr. A. N. Smith (Unilever Research Laboratories, Colworth). (2) The equilib- rium constant is ca. l o 3 times greater than that found for PhB( OH), complexation with aqueous catechol. 13 Neverthe- less, the K value is not sufficiently large for all the extracted TF to be complexed under the normal analytical conditions. If v = 2, so that bo = 22.2 mM in the IBMK - ethanol mixture, Table 3. Equilibrium measurements on the TF - flavognost complex in 25% V/V IBMK - 75% V/V ethanol at 25 “C K* K? 1 0 5 ~ * / p ~ - l K * / p ~ l / ~ vIcm3 bdmM C J ~ M (m= 1) (m= 1) (m= 2) ( m = 1/2) 0.1 0.556 35.9 0.046 0.043 10.3 1.04 0.2 1.112 53.2 0.074 0.065 7.74 2.38 0.4 2.225 66.4 0.089 0.074 4.40 4.12 0.6 3.338 74.2 0.112 0.085 3.58 6.34 0.9 5.006 77.3 0.101 0.073 2.11 7.06 1.5 8.344 82.3 0.117 0.070 1.45 10.6 * co = 89.3 p ~ .t c, = 94.0 VM. and with a typical TF concentration in the mixture of 80 p ~ , it follows from the above equilibrium constant that c,/co = 0.96. A very similar result emerges from the amount by which co had to be increased to yield a constant value of K, as 89.3/94 = 0.95. Incomplete complexation therefore leads to an under- estimate by 4 5 % in the TF content of the organic phase. To check this conclusion, a set of analyses was carried out on a similar Kapchorua tea solution with n = 2, v = 2 and using both a 2% m/V and a 4% m/V solution of flavognost.As expected, the absorbances at 625 nm were consistently 2% greater with the more concentrated flavognost reagent. Conclusions The solvent extraction step in the Hilton flavognost method has been shown to be 89% efficient if a sample of tea solution is extracted by twice its volume of IBMK. The subsequent complexation of the extracted TF with flavognost is ca. 95% complete when 2 cm3 of the IBMK extract is mixed with 2 cm3 of 2% mlvethanolic flavognost and 4 cm3 of ethanol. Overall, only 85% of the TF in a sample of tea solution ultimately appears in the form of the green flavognost complex whose absorbance is measured. The use of an appropriate propor- tionality constant in the Beer’s law equation can overcome this problem provided that a standard analytical procedure is always employed (preferably n = 2, v = 2 cm3).The method must be standardised by dissolving a known amount of TF in an aqueous solution and not by dissolving it directly in IBMK. The authors thank the SERC for the award of a CASE Studentship to W.E.P. and Unilever plc and particularly Dr. D. R. Haisman and Mr. A. N. Smith for their help and support and for supplying the tea leaf and a sample of ungallated theaflavin. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. References Roberts, E. A. H., and Smith, R. F . , J. Sci. FoodAgric., 1963, 14, 689. Hilton, P. J., and Ellis, R. T., J. Sci. Food Agric., 1972, 23, 227. Cloughley, J. B., J. Sci. Food Agric., 1980, 31, 911. Roberts, E. A. H., and Smith, R. F., Analyst, 1961, 86, 94. Collier, P. D., and Mallows, R., J. Chromatogr., 1971, 57, 19. Nestle’s Products Ltd., Br. Pat., 1 034 670, 1966. Hilton, P. J., in Snell, F. D., and Ettre, L. S . , Editors, “Encyclopedia of Industrial Chemical Analysis,” Volume 18, Wiley, New York, 1973, p. 455. Hilton, P. J. R., “Tea Research Foundation of Central Africa Annual Report,” Mulanje, Malawi, 1972/73, Section 4, p. 80. Hoefler, A. C., and Coggon, P., J. Chromatogr., 1976, 129, 460. Wellum, D. A., and Kirby, W., J. Chromatogr., 1981,206,400. Price, W. E., and Spiro, M., J. Sci. FoodAgric., 1985,36,1303. Conner, J. M., and Bulgin, V. C., J. Znorg. Nucl. Chem., 1967, 29, 1953. Pizer, R., and Babcock, L., Znorg. Chem., 1977, 16, 1677. Oertal, R. P., Inorg. Chem., 1972, 11, 544. Fernhndez-Prini, R., in Covington, A. K . , and Dickinson, T., Editors, “Physical Chemistry of Organic Solvent Systems,” Plenum Press, London, 1973, Appendix 5.1. Paper A5/230 Received June 26th, 1985 Accepted September 26th, 1985

ISSN:0003-2654    

DOI:10.1039/AN9861100331

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST, MARCH 1986, VOL. 111 335 Micro-determination and Separation of Manganese Using a Liquid Ion Exchanger Sobhana K. Menon and Yadvendra K. Agrawal Analytical Chemistry Laboratory, Pharmacy Department, Faculty of Technology and Engineering, M.S. University of Baroda, Kalabhavan, Baroda-390 00 1, India A sensitive and selective method for the micro-determination of manganese has been developed involving the extraction of the wine-red manganese - benzohydroxamic acid (BHA) complex with Aliquat 336 liquid ion exchanger, and the optimum conditions have been established. Kinetic and stability studies on the complex were carried out. The extracted metal can be quantitatively eluted with 0.25 M hydrochloric acid, thus rendering the method applicable for the concentration, separation and determination of manganese in samples containing very low levels of the metal.Keywords: Manganese determination; spectrophotometry; liquid ion exchanger Liquid ion exchangers are extensively used for the pre- concentration, separation and recovery of several metal i ~ n s l - ~ and are becoming popular owing to their potential for industrial chemical separations and for the recovery of costly chemicals.4 The superior extraction ability of liquid ion exchangers can also be utilised for the effective removal of toxic metal ions.5 The extraction of a coloured species of the metal ion by a liquid ion exchanger permits its direct detection together with its pre-concentration, separation and determi- nation. Interferences can be considerably reduced either by masking before extraction or by removing the interfering ions with a suitable eluant from the liquid ion exchanger phase before the elution of the metal. This method has been adopted for the determination of manganese as reported for the determination of titanium.6 Hydroxamic acids have already proved to be very sensitive and selective reagents for the trace determination of metals.7.8 A few hydroxamic acids have been reported for the determi- nation of manganese9JO in basic media.The wine-red com- plex of manganese with benzohydroxamic acidlo was found to be extracted instantaneously into Aliquat 336, with a consider- able enhancement of the sensitivity; this can be applied to the determination of microgram amounts of manganese and to its separation from closely associated metals by selective elution.Experimental Chemicals and Reagents All chemicals were of analytical-reagent or general-reagent grade from BDH Chemicals and E. Merck, respectively, unless specified otherwise. Benzohydroxamic acid (BHA). This was prepared as described elsewhere11 and was further purified by our modified method.6 A 0.1 M solution in doubly distilled water was prepared. Manganese standard solution. A stock solution of man- ganese was prepared by dissolving the requisite amount of MnS04.4H20 in doubly distilled water. The solution was standardised titrimetrically with EDTA12 and the metal content was found to be 0.703 mg ml-1. The solution was diluted as required. Liquid Anion Exchangers Amberlite LA-1 [N-dodecyl(trialkylmethyl)amine] (Rohm & Haas, Philadelphia, PA, USA), Aliquat 336 (tricapryl- methylammonium chloride) (Fluka, Buchs, Switzerland) and tnoctylamine (Fluka) dissolved in suitable diluents in varying proportions, were used.Apparatus A VSU2-P spectrophotometer (Carl Zeiss, Jena, GDR) with matched quartz cells was used for spectral measurements. pH measurements were made on an Elico digital pH meter equipped with calomel and glass electrodes. Procedure A sample solution containing 12-200 pg of the metal was placed in a 60-ml separating funnel and ammonia solution was added so that the basicity of a total volume of 15 ml of aqueous phase was between 0.2 and 0.3 M of ammonia. A 2-ml volufhe of a 0.1 M solution of BHA was added immediately to the above solution, which was then mixed well and kept for 5 min. The mixture was shaken gently with 15 ml of a 4% solution of Aliquat 336 in xylene for about 1 min after the addition of 5 ml of 2 M ammonium chloride solution. The phases were allowed to separate and the organic extract was dried over anhydrous sodium sulphate and transferred into a 25-ml calibrated flask. To ensure the complete recovery of manganese, the extraction was repeated with 5 ml of the extraction solvent, followed by drying with sodium sulphate and finally dilution of the combined extract to the mark with the solvent.The absor- bance was measured at 480 nm against a reagent blank. To calculate the distribution ratio, D, and the percentage extraction, E, the manganese concentration in the aqueous phase was determined with 4- (2-p yrid ylazo) resorcinol (PAR).13 The metal content in the organic phase was determined after elution of the metal from the organic phase as described below.For the elution of manganese, the organic layer was shaken with 15 ml of 0.25 M hydrochloric acid for 2 min. The two phases were allowed to settle and the aqueous layer was withdrawn carefully. The metal content was determined using PAR. The pH of the solution was adjusted to 7-8 and 5 ml of borate buffer (pH 10) were added together with 5 ml of 0.1 M PAR solution. The solution was diluted to 50 ml and measurements were taken at 500 nm after keeping the mixture for 15 min. The amount of manganese was calculated from a calibration graph. Results and Discussion The manganese complex is wine-red and has a fairly wide absorption band at 480 nm. The complex formation is accompanied by a change in the oxidation state of the manganese from +2 to +3, which was accomplished by the oxygen dissolved in the solution.The colour formation will not take place if the mixture of the reagent and manganese ions is336 ANALYST, MARCH 1986, VOL. 111 boiled to eliminate the dissolved oxygen. However, on exposure to air or on addition of an oxidising agent such as dilute hydrogen peroxide the wine-red colour appears. The presence of small amounts of hydroxylammonium chloride , which prevents the oxidation of Mn(II), will also inhibit the appearance of the characteristic colour, thereby ensuring the tervalent state of manganese in the complex. The instan- taneous extraction of the complex into the liquid anion exchanger shows the anionic nature of the complex.Effect of Variables on the Extraction Basicity The complex formation is most favoured in ammonia solution. Even though the complex will form in sodium or potassium hydroxide solution, the rate of formation is much slower than in ammonia solution. Also, the former bases will cause interferences due to the formation of precipitates of hydrox- ides with certain metallic ions, whereas precipitates that are soluble in an excess of ammonia are formed in most instances in an ammoniacal medium. The extraction is quantitative about pH 10 and remains constant up to 0.4 M of ammonia. The sensitivity decreases very slowly above 0.5 M of ammonia. Therefore, extractions were carried out at ammonia concen- trations around 0.3 M.Reagent concentration The absorbance of the manganese complex was constant with the use of excess of the reagent. Extractions with various concentrations of the reagent showed that 1-2 ml of 0.1 M BHA solution was adequate for quantitative extraction of the manganese. The reagent is added immediately after the addition of ammonia solution and mixed well so as to prevent the precipitation of manganese hydroxide , which will delay the colour formation. Electrolytes The extraction was carried out in the presence of various concentrations of electrolytes such as NH4C1, NaCl, Na2S04, KCl and K2SO4 as the separation of the organic and aqueous layers was not clear in the absence of an electrolyte. A 2 4 M solution of NH4C1 gave satisfactory results.For each extrac- tion, 5 ml of 2 M NH4Cl solution were used. Aliquat 336 concentration The optimum concentration of Aliquat 336 was studied by varying the concentration from 1 to 10% in xylene. The extraction was quantitative from 3% and remained constant up to 8%. A tendency to form emulsions was observed at higher concentrations of the ion exchanger. A 4% solution of the exchanger was used for extraction. Diluents Manganese was extracted with 4 and 6% Aliquat 336 solution in various diluents. Equilibration was effected by maintaining the ratio of organic to aqueous phase at 1: 1 and the Table 1. Effect of various diluents on the extraction (YO) of manganese with Aliquat 336 Mn extracted, YO ~~ ~ Diluent 4% Aliquat 336 6% Aliquat 336 Benzene . .. . . . . . 99.9 99.9 Toluene . . . . . . . . 83.5 83.0 Hexane . . . . . . . . 55.8 55.3 Chloroform . . . . . . 45.0 45.5 Carbon tetrachloride . . 51.2 50.8 Xylene . . . . . . . . 99.5 99.0 percentage extraction was calculated in each instance (Table 1). The extraction was complete and quantitative with benzene and xylene and a clear separation was obtained. As benzene is highly toxic, xylene was used as the diluent in subsequent work. Type of liquid anion exchanger Manganese was extracted with three extractants in various diluents (Table 2). Aliquat 336 in benzene or xylene was found to be the best extractant. Equilibration time and stability The extraction was very rapid and required only a few seconds for quantitative and complete extraction. The time of shaking was varied from 30 s to 5 min.The extraction was quantitative within 30 s. The complex extracted under optimum conditions was stable for several days. Optical Properties The colour system obeyed Beer's law from 0.35 to 9.0 p.p.m. of manganese at 480 nm and the optimum range (Ringbom plot)l4 was 0.5-10 p.p.m. The molar absorptivity was 7.5 x 103 1 mol-1 cm-1. Kinetic Study of Colour Formation and Extraction As the colour formation of the complex and consequently the sensitivity of the extraction depend on time, a study was carried out to follow the rate of the reaction and to evaluate the rate constant. Extractions were carried out at room temperature (30 "C) at regular intervals of 1 min starting from the first to the tenth minute, the time referring to the interval between the time of mixing of the reagent and completion of extraction.The absorbance increased steadily and became constant after the fifth minute, indicating completion of the colour reaction. Therefore, extractions were carried out 5 min after mixing of the reactants. The rate constant was calculated (Table 3) by substituting the values of absorbance and time in the equations for first-and second-order reactions. The rate constant was found to be 0.65 min-1 following the first-order reaction. Table 2. Effect of different liquid anion exchangers on the extraction of manganese Liquid anion exchanger Diluent Extraction, YO Aliquat336(4%) . . . . Benzene Amberlite LA-1 (4%) . . . . Benzene Trioctylamine(4%) . . . . Benzene Xylene Chloroform Xylene Chloroform Xylene Chloroform 99.9 99.5 45.0 81.5 70.8 41.5 72.0 48.3 32.5 Table 3.Kinetic study of complexation Timelmin 1 2 3 4 5 6 7 Extraction, YO 62.2 70.8 83.3 92.2 99.0 99.9 99.9 * Average rate constant, K = 0.65. Rate constant, K*/min - * 0.69 0.67 0.63 0.63 0.61 M MANALYST, MARCH 1986, VOL. 111 337 Table 4. Effect of diverse ions. Mn taken = 70 pg per 25 ml Foreign ion Ag+ . . . . Be2+ . . . . Mg2+ . . Ca2+ . . . . sr2+ . . . . Ba2+ . . . . Sn2+ . . . . Pb2+ . * . . a * + . . , . AS3+ . . . . c o 2 + . . . . c u 2 + . . , . Hg2+ . . Ni2+ . . . . Zn2+. . . . Ti4+ . . . . cr3+ . . . . A13+ . . . . v5+ . . . . M070246- . . zr4+ . . . . U6+ . . . . W6+ . . . . Sb3+ . . . . ce4+ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tolerance limithg 15* 25 30 30 30 29 25 107 15t 30 10* 20* 10t 30 * 2 20 3 1.5 2% 1% 20 107 20 20 10 * Masked with 0.1% sodium cyanide solution. t Masked with 0.5% potassium iodide solution. $ Eluted with 0.2 M sodium acetate solution. § Stripped with 0.2 M sodium chloride solution. 7 Stripped with 0.1 M ammonia solution. Table 5. Determination of manganese in NBS standard samples Certified Mn Standard Sample concentration, % Mn found,* YO deviation, YO Mgalloy . . . . 0.44-0.46 0.455 0.002 Steel14C . . 0.455-0.470 0.450 0.003 Mn bronze 164 . . 4.65-4.72 4.65 0.01 Sibronze . . . . 1.30-1.33 1.31 0.01 * Average of six determinations. Composition and Stability Constant The composition of the manganese complex was studied by plotting a graph of the logarithm of the distribution coefficient of the metal [log D(M)] against the logarithm of the ligand concentration (log[ligand]).15 The extraction was carried out by taking a fixed amount of manganese in the presence of (a) a constant amount of Aliquat 336 and varying the concentration of BHA and (b) a constant amount of BHA and varying the concentration of Aliquat 336. In both instances the plot of log D ( M ) against log[ligand] gave straight lines with slopes of 3 and 1.7, respectively, which indicates that the composition of the complex is manganese : BHA = 1 : 3 and manganese : BHA : Aliquat 336 = 1 : 3 : 2. It has already been seen that the complex formation is accompanied by a change in the oxidation state of the metal from +2 to +3.The anionic nature of the manganese complex necessitates the reaction with the enol form of the reagent. It has also been establishedlOJ6 that in basic solution a consider- able portion of the benzohydroxamic acid exists in the enol form of the two possible structures of the reagent molecule: __ 0 0 -H C6H5-C-N-OH S C6H5-C=N-O-H II I I I H (keto) (enol) Therefore, the possible structure of the complex is where R4N+ represents the cationic part of the liquid ion exchanger. The stoicheiometric proportion of 2 parts of the liquid ion exchanger in the complex shows that there are only two negative charges on the 1 : 3 manganese - BHA complex. This can be explained by assuming that of the three molecules of the reagent reacting with the metal ion, two are in the enol form and one is in the keto form, leaving behind two negative charges on the metal complex, which will form a neutral ion pair by taking up two cationic parts of the liquid ion exchanger. The stability constant of the complex determined by the spectrophotometric method17 was 1.175 x 108.Stripping After extraction of the manganese into the organic phase, it was stripped with 15 ml of varying concentrations (0.05-5 M) of sulphuric acid, hydrochloric acid, nitric acid, sodium sulphate, sodium chloride , sodium carbonate and sodium hydroxide solutions. The stripping was complete with 0.25 M hydrochloric acid. The metal constant was determined pho- tometrically using PAR.13 Effects of Diverse Ions Manganese was extracted and separated in the presence of a large number of different ions (Table 4).Interference studies were made by measuring the absorbances of the liquid ion exchanger phase and conditions were established for the removal of interfering ions from the organic phase by eluting with suitable solvents. The tolerance limit was set as the amount of foreign ion causing a change in absorbance of k0.02 unit or f 2 % error in the recovery of manganese. Moderate amounts of various metal ions commonly associated with manganese were tolerated, and also most anions. Common anions such as nitrate, chloride, sulphate, acetate, carbonate, thiocyanate , iodide and thiosulphate will not interfere even at very high concentrations, whereas fluoride and cyanide show some interference at higher concentrations (> 5 mg).Up to 20-fold concentrations of zinc and aluminium can be tolerated as their metal hydroxides are soluble in an excess of ammonia. The interference caused by Cu2+, Ni2+, Co2+ and Ag+ can be masked with 0.1% sodium cyanide solution. A 0.5% solution of potassium iodide can be used to mask Cd2+, Hg2+ and Pb2+. Fe3+ interferes owing to the precipitation of hydrous iron(II1) oxide. This can be overcome by a single selective extraction of Fe3+ from 6 M hydrochloric acid with an equal volume of diethyl ether saturated with water prior to the determination of manganese. The interfer- ing ions Fe3+, Co2+ and Cu2+ can be separated simultaneously from Mn2+ by an ion-exchange technique. The separation is effected by the absorption of these ions on Dowex 50W-X8 from 6 M hydrochloric acid, which retains iron, cobalt and copper on the resin column.The effluent contains only manganese and is evaporated to dryness and made alkaline with ammonia solution for further analysis. This method of separation will be extremely useful for the analysis of manganese in standard samples of steel and bronze in which the major constituents are a combination of the above metals in varying proportions.338 In certain instances interfering ions can be removed by selective stripping before the recovery of manganese. Thus vanadium, extracted together with manganese, can be stripped first by shaking the organic phase with 10 ml of 0.2 M sodium acetate solution for 5 min. Manganese is subsequently stripped with 0.25 M hydrochloric acid.The extracted molyb- denum can be removed by washing the organic phase with 5 ml of 0.2 M sodium chloride solution before the separation of manganese. Uranium can also be separated from manganese by stripping with 5 ml of 0.1 M ammonia solution and manganese can be recovered with 10 ml of 0.25 M hydrochloric acid. The method is sensitive and reasonably selective. All the interfering ions can either be masked or eluted selectively. The serious interference from iron can also be easily over- come. The interference from iron, cobalt and copper can be easily removed by a simple ion-exchange technique, which makes the method suitable for the analysis of steel and bronze. Determination of Manganese in Standard Samples Manganese alloy and ore samples were digested with a mixture of concentrated HN03 + HC104 (1 + 3) and evaporated.The semi-solid mass was heated with concen- trated HC1, centrifuged and finally diluted to 250 ml with 0.1 M HC1. An aliquot of this solution was taken and analysed for manganese by the proposed method, and the results are given in Table 5. ANALYST, MARCH 1986, VOL. 111 We are indebted to UGC, New Delhi, for awarding a Research Associateship to one of us (S. K. M.). 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. References Green, H., Talanta, 1964, 11, 1561. Asha, R. P., and Khopkar, S. M., J. Sci. Znd. Res., 1971, 30, 16. Green, H., Talanta, 1973, 20, 139. Moore, F. L., Environ. Sci. Technol., 1972,6, 525. McDonald, C. W., and Moore, F. L., Anal. Chem., 1973,45, 983. Menon, S. K., and Agrawal, Y. K., Analyst, 1984, 109,27. Agrawal, Y. K., and Patel, S. A., Rev. Anal. Chem., 1980, 4, 237. Agrawal, Y. K., Rev. Anal. Chem., 1980, 5 , 3 . Dutta, R. L., J. Indian Chem. SOC., 1957, 34, 311. Dwight, 0. M., and John, H. Y., Talanta, 1960, 7, 107. Hauser, E. R., and Renfrow, W. B., Jr., “Organic Synthesis,” Wiley, New York, 1944, p. 7. Welcher, F. J., “The Analytical Uses of Ethylenediamine- tetraacetic Acid,” Van Nostrand, Princeton, NJ, 1958-, p. 220. Ahrland, S., and Herman, R. G., Anal. Chem., 1957,47,2422. Ringbom, A., Fresenius 2. Anal. Chem., 1949, 21, 332. Yoe, J. H., and Jones, A. L., Ind. Eng. Chem., Anal. Ed., 1944, 16, 111. Plapinger, R. E., J. Org. Chem., 1959, 24, 802. Harvey, A. E., and Manning, D. L., J. Am. Chem. SOC., 1950, 72, 4488. Paper A41386 Received November 7th, 1984 AcceDted SeDtember 25th. 1985

ISSN:0003-2654    

DOI:10.1039/AN9861100335

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST, MARCH 1986, VOL. 111 339 Synthetic Inorganic Ion-exchange Materials Part XLII.* Ion-exchange Selectivity of Divalent Transition Metals and Lead on Titanium Antimonate and Some Chromatographic Separations R. Chitrakar and M. Abet Department of Chemistry, Facultry of Science, Tokyo Institute of Technology, 2- 12- 1, Ooka yama, Meguro-Ku, Tokyo 152, Japan The equilibrium distribution coefficients of divalent transition metals and lead have been determined on a titanium antimonate (TiSbA) cation exchanger. The selectivity sequence Mn(ll) < Ni(ll) < Cd(tl) < Zn(ll) < Co(ll) < Cu(ll) < Fe(ll) < Pb(ll) was established for divalent metal ions (10-4 M) on TiSbA with different Ti to Sb molar ratios in nitric acid media. Useful separations of lead from transition metals and alkaline earth metals using relatively small columns of TiSbA are discussed.Keywords: Titanium antimonate; ion exchanger; chromatographic separation; transition metals; lead Many useful inorganic ion-exchange materials have been synthesised during the last two decades that have found many applications in areas of analytical chemistry, radiochemistry , environmental chemistry and biochemistry. 1 Inorganic ion exchangers possessing high selectivities for certain ions or groups of ions2-5 can be utilised for chromatographic separations of elements. Most investigations with insoluble acid salts of quadrivalent metals have been carried out on zirconium and titanium phoshates of vari6us types.1 Quadrivalent metal antimonate ion exchangers were studied first by Abe and Ito.6 Tin(IV) and titanium(1V) antimonates (SnSbA and TiSbA) behaved as cation exchangers with relatively high capacity, and showed unusual selectivities for alkali metal ions in the order Na < K < Rb < Cs << Li on SnSbA,7 and Na < K < Rb < Li < Cs on TiSbA,s while the usual selectivity for alkaline earth metal ions was observed on both exchangers. SnSbA, having a specific selectivity for the lithium ion, has proved a promising ion-exchange material for the selective separation of lithium from sea water.9 This paper describes adsorption and chromatographic separations of some divalent transition metals and lead on TiSbA with different Ti to Sb molar ratios.The chromato- graphic separations of alkaline earth metals from some transition metals and lead are also discussed.Experimental Reagents Antimony pentachloride (Yotsuhata Chemical Co., Japan) and titanium tetrachloride (Wako Chemical Co., Japan) of high purity (> 99% as metal) were used without further purification. Standard solutions of transition metals and lead were prepared by dissolving metals of high purity (> 99.9%) in a minimum amount of nitric acid. For Fe(I1) adsorption experiments, all reagents were prepared with nitrogen gas bubbling. Iron(I1) nitrate solution was obtained by passing iron(I1) sulphate solution through a column of Dowex-1 anion exchanger in the nitrate form in a nitrogen gas atmosphere. Preparation of TiSbA Titanium antimonate was synthesised as described previ- ously.* A 4 M antimony pentachloride solution was mixed with 4 M titanium tetrachloride solution at different Ti to Sb molar * For Part XLI of this series, see Abe, M., and Furuki, N., Solvent t To whom correspondence should be addressed.E m . Ion Exchange, in the press. ratios of 60 "C. The mixed solutions (40 ml) were imme- diately hydrolysed in 960 ml of de-mineralised water at the same temperature. The precipitate obtained was kept in the mother liquor overnight, filtered and washed with cold de-mineralised water using a centrifuge operated at 1000 rev min-11 until the pH of the supernatant solution was higher than 1.5. The product obtained was dried at 60 "C for 4 d, ground and sieved to 100-200 mesh and the samples were washed with cold de-mineralised water in order to remove fine adherent particles and conditioned with 1 M HN03 until free from C1- and Na+ ions.Finally, the samples were re-washed with de-mineralised water and air dried. Characterisation of TiSbA The determination of antimony and titanium by X-ray diffraction and thermal analysis were carried out as described previously .8 Distribution Coefficients The values of the distribution coefficients (Kd) of several metals ions were determined as follows: 0.10 g of TiSbA was equilibrated with 10.0 ml of solution containing 10-4 M metal ions in different concentrations of nitric acid with intermittent shaking at 30 "C. The concentrations of metal ions in the solid 100 - 80 . 8 9 r' 60 0 . .- c 4 0 - 3 20 - -s------ I 1 I 5 10 15 20 25 30 Ti me/d Fig. 1. Time dependence of the adso tion of transition metals and lead on TiSbA.TiSbA (Ti : Sb = 1.6),?25 g; initial concentration of metal ions, M; total volume, 25.0 ml; temperature, 30 k 0.5 "C; concentration of HN03, 0.07 M for Mn(II), Ni 11) and Co(II), 0.1 M for Cu(II), Zn(I1) and Cd(I1) and 1 M for Pb(& and Fe(I1)340 ANALYST, MARCH 1986, VOL. 111 and the liquid phases were deduced from the concentration relative to the initial concentration in the solution. The Kd values were calculated at equilibrium using the following equation: Kd = X Amount of metal ions in exchanger Amount of metal ions in solution Volume of solution (ml) Mass of exchanger (g) The concentrations of metal ions were determined by using a Varian Techtron 1100 atomic absorption spectrometer. Results and Discussion Characterisation of TiSbA Among various preparations, titanium antimonates with Ti to Sb molar ratios of 1.1, 1.6, 2.1 and 2.8 were synthesised.The X-ray diffraction and the thermal analysis of all the samples were examined, the results obtained being in good agreement with previous work.8 103 r I 0) - E . 9 1 02 10' ' I I \ I 10-2 lo-' 100 10' [HNO~]/M Fig. 2. Distribution coefficients (&) of transition metals and lead on TiSbA as a function of concentration of HN03. TiSbA (Ti : Sb = 1 .6), 0.10 g; initial concentration of metal ions, 10-4 M; total volume, 10.0 ml; temperature, 30 k 0.5 "C 103 .- I 0 - E , 9 102 10' cu2+ Zn2+ Cd2+ Ni2+ Mn2+ 1 .o 2.0 3.0 Molar ratio, Ti : Sb Fig. 3. Kd values of transition metal ions on TiSbA with different molar ratios of Ti to Sb.Conditions for & determination as in Fig. 2 Ion-exchange Selectivity The time dependence of the adsorption of divalent metal ions on TiSbA in nitric acid was measured qualitatively in order to determine the equilibrium distribution coefficients. The time required to attain equilibrium varied among the metal ions studied (Fig. 1). For lithium and caesium ions studied on TiSbA,8 slow rates of adsorption were also reported. Slow rates of adsorption for divalent transition metal ions were also observed on SnSbA.10 When reagents prepared without bubbling nitrogen gas through were used in the Fe(I1) adsorption experiment , TiSbA (white granules) changed to dark grey; this characteristic dark grey of the Sn02 exchanger was also noted in Fe(I1) sorption experiments.llJ2 However, the dark grey colour of TiSbA was not observed when reagents prepared with nitrogen gas bubbling were used. After equilibration, log & values of metal ions were plotted against log[HN03]. A linear relationship with a slope of -2 was obtained for all metal ions studied at different Ti to Sb molar ratios, the result using a Ti to Sb ratio of 1.6 being shown in Fig.2. This indicated that the adsorption of metal ions on TiSbA proceeded via an ideal ion-exchange mechan- ism. The selectivity series Mn < Ni < Cd < Co < Zn < Cu < Fe < Pb was established for 10-4 M concentrations of divalent metal ions on TiSbA at different Ti to Sb molar ratios in nitric acid. The & values of transition metal ions in 0.1 M HN03 5 4 3 9 0, -I 2 1 7 9 1 1 -Log K11 Fig.4. Plot of log & against first hydrolysis constant of cationslg 104 103 7 I 0) - E 9 1 102 10' 10-5 10-4 10-3 10-2 Initial concentration of metal ionsh Fig. 5. Kd values of transition metal ions at different concentrations of metal ions. TiSbA (Ti: Sb = 1.6), 0.10 g; total volume, 10.0 ml; HN03 concentration, 0.1 M; temperature, 30 k 0.5 "CANALYST, MARCH 1986, VOL. 111 341 - - Table 1. Distribution coefficients, Kd, and separation factors, a,* for divalent transition metals and lead on TiSbA and Bio-Rad AG50W-X819 I (0 I - 3 : C 0 .- 4- - 2 ; 1 +8 al C N 0 0 TiSbA ~~ ~ HN03/ Ti: Sb M ratio Parameter Mn Ni Cd Zn c o c u Fe t Pbf. 0.1 1.17 Kd 89.1 118.2 396.5 781.1 1392.5 2683.2 117 >lo4 1.63 Kd 33.3 53.1 100.0 355.8 699.5 1472.8 93 >lo4 a 1.3 3.3 1.9 1.7 1.9 a 1.6 1.8 3.5 1.9 2.1 a 2.0 1.0 4.1 1.2 4.2 a 1.4 1 .o 7.7 1.2 4.8 2.17 Kd 30.1 60.6 66.2 275.8 346.6 1470.5 131 280 2.79 Kd 10.1 14.3 14.7 114.3 140.6 681.3 75 140 Bio-Rad AGSOW-X8 H N 0 3 / ~ Parameter Zn c u Ni Mn c o Cd Pbf.0.1 Kd 1020 1080 1140 1240 1260 1500 35.7 a 1.05 1.05 1.08 1.01 1.19 * a is defined by f. Kd value at 1 M HN03. I m I 1.6 . S 0 .- - 1.2 F s 4- 0.8 0.4 .- Z 0 E I u ate vo I u meim 1 Eluate volume/ml Fig. 6. Separation of (a) Ni(I1) from Cu(I1) and (b) Ni(I1) from Co(I1) with TiSbA usin nitric acid. (a) Loading, 20 pmol for Ni(I1) and 0.2 pmol for Cu(I8; flow-rate, 0.20 ml min-1. ( b ) Loading, 2 pmol for Ni(I1) and 0.02 pmol for Co(1I); flow-rate, 0.15 ml min-' HN03 2.0 I Ni2+ II 1.6 1.2 I 7 0.8 2 .- 0.4 f! .c. 4- c 3 0 20 40 60 L 80 100 L. 0 Eluate volume/ml 5 M HN03 H were plotted against the Ti to Sb molar ratios (Fig. 3), the Kd values of Fe(I1) and Pb(I1) not being included because these metal ions were completely adsorbed in 0.1 M HN03. The metal ions showed high Kd values at a Ti to Sb ratio of 1.1 and the values decreased with an increase in the Ti to Sb molar ratio. The exchangeable protons in TiSbA decreased with an increase in the Ti to Sb ratio, causing decreased Kd values of metal ions. Some inorganic ion exchangers such as C-SbA1 or Mn0213 show size preference selectivity for ions having specific ionic radii, but no such correlation was found between the Kd values and effective ionic radii14 of metal ions studied on TiSbA. It is known that most ion exchangers exhibit a general selectivity sequence between members of four groups of ions, i.e., polyvalent metals > divalent transition metals > alkaline earth metals > alkali metals.15 This behaviour is not shown by E I uate vol u me/m I Fig.7. Separation of (a) Ni(II), Co(I1) and Cd(I1) from Pb(I1) and (b) Zn(I1) and Cu(I1) from Pb(1I) with a TiSbA column. Pre- treatment of the column: (a) 0.1 M HN03 and ( b ) 0.2 M HN03; loading, 1 pmol of each of metal ion; flow-rate, 0.2 ml min-1 the insoluble salts of inorganic acids, e.g., zirconium and tin phosphate gels, which exhibit considerable overlap in the selectivity between three groups of the ions.15 The selectivities of the a- and 8-zirconium phosphatesl6J7 for transition metal ions have been reported to be in the order Ni < Co < Mn < Zn < Cu.A slightly different selectivity order was observed on TiSbA at any Ti to Sb molar ratio. The log Kd values of transition metal and lead ions on TiSbA were plotted against the first hydrolysis constants of cations (Fig. 4).18 The results showed a good correlation except for Fe(I1) and Cu(I1).342 ANALYST, MARCH 1986, VOL. 111 5 m I : 4 2 3 8 0 .- 4- a, C ts 2 N m u U c 1 m t N r" 5 1 2.4 d z . 8 0 m .- - c 1.6 .c C 0 .- 4- F E 0.8 0 8 0 0 ( a ) I-* 0.05 M HN03 P ~ M H N O J 0 20 - 40 60 80 E I uate volu meim I 100 120 140 ~ b) 0.2 M HN03 -6 M HN03 h'"'+ - Mg2+, Ca2+ -Sr2+ 5 m I 0 8 0 F . .- c 3 ; 2 +s Q) 8 N 8 N I F t 3 N 0 - 0 20 40 60 80 100 E I uate vol u meim I Fig. 8. Separation of (a) Mg(I1) and Ca(I1) from Zn(I1) and Cu(I1) and (b) Mg(II), Ca(I1) and Sr(I1) from Pb(I1) with a TiSbA column.(a) Loading, 20 pmol each of Mg(I1) and Ca(II), 1 vmol of each of Zn(I1) and Cu(I1); (b) loading, 1 pmol of each metal ion; flow-rate, 0.20 ml min-1; column pre-treated with 0.2 M HN03 The distribution coefficients, Kd, and separation factors, a, of metal ions studied on TiSbA at different Ti to Sb ratios are given in Table 1, the values obtained on Bio-Rad AG50W-X819 being included for comparison. The separation factors between neighbouring pairs of metal ions on TiSbA are larger than those observed on the organic resins, which show very similar Kd values in nitric or perchloric acid. The separation factor of Pb(I1) and the first transition metal ions on resin shows a large value (ca.5 ) for a low concentration of acid and decreases with increasing concentration (ca. 2 in 4 M acid). An improved separation factor was observed between lead and transition metals on Dowex 50W-X8 cation-exchange resin by complex formation with HC1 or HBr.20321 The separation of Pb(I1) from other elements was achieved with 0.6 M HBr on a heated column, which prevented the precipitation of lead bromide.20.21 The distribution coefficient, Kd, of Pb(I1) on Bio-Rad AG50W-X8 is low in hydrochloric acid - acetone, which makes the method less suitable for the separation of small amounts of lead from large amounts of other elements.22 An improved separation factor was also obtained on the resin in perchloric acid media and lead could be separated from other metals in 2 M HC104.23 Further, a good separation is possible by eluting Pb(I1) with 0.5 M HBr in 40% acetone while Zn(II), Cu(I1) and Ni(I1) are strongly retained on the resin .24 The anion-exchange behaviour of bromide complex-forming elements in nitric acid - hydrobromic acid showed that lead could be separated from other elements by preferential elution with HN03 - HBr eluent.25-27 Some inorganic ion exchangers, e.g., thorium phosphate28 and cerium phosphate,29 showed high selectivity for lead.Although lead could be separated from other elements, a long tailing effect could be observed on elution.28 On organic ion exchangers, the Kd values of metal ions are usually constant at concentrations below 1 mM at a constant acid concentration. As TiSbA with a Ti to Sb ratio of 1.6 showed a slightly high separation factor between some pairs of metal ions, the concentration dependence of Kd values on the concentration of metal ion on TiSbA was studied at a Ti to Sb ratio of 1.6 in 0.1 M HN03.The results are shown in Fig. 5. It was apparent that Kd values of cu(II), Co(I1) and Zn(I1) increased considerably compared with those of Mn(II), Ni(I1) and Cd(I1) at low metal ion loadings. Such a concentration dependence of the Kd values of divalent transition metal ions has also been reported on An-HTD0.30 Chromatographic Separation It is evident from the Kd values that the separation of microamounts of Co(II), Zn(I1) and Cu(I1) from macro- amounts of Mn(I1) and Ni(I1) can be achieved on a TiSbA column without the use of a complexing agent.Relatively small columns (2 cm x 0.5 cm i.d.) containing TiSbA (100-200 mesh) with a Ti to Sb ratio of 1.6 were used throughout. A 2-pmol amount of each of Mn(II), Ni(II), Cd(II), Zn(II), Co(I1) and Cu(I1) were loaded on top of the column and the elution was carried out immediately using nitric acid of different concentrations. The order of elution of metal ions did not coincide with the equilibrium distribution coefficients because Co(I1) and Zn(I1) could be detected before Cd(I1). Such a difference was considered to be due to the slower adsorption process of Co(I1) and Zn(I1) than Cd(I1) on TiSbA. The elution peaks of Mn(I1) and Ni(I1) were sharp; however, long tails were observed for Cd(II), Zn(II), Co(II), Cd(I1) and Zn(I1) and the yields were quantitative with incomplete separation. An attempt was made to separate Cu(I1) from Mn(II), Ni(II), Co(II), Cd(I1) and Zn(I1) at lower column loadings and by decreasing the concentration of nitric acid.Although Cu(I1) was separated from other metal ions, tailing during elution and low elution yields were encountered. Such behaviour was also reported on an Sn02 column during an elution experiment with transition metal ions.12 Low yields were considered to be due to the strong retention of cations to the exchanger particle during the time between feeding and elution periods. The respective yields were 88, 87, 36, 50, 49 and 58% for Mn(II), Ni(II), Cd(II), Zn(II), Co(I1) and Cu(I1). Coloration of the column during the loading of the metal ions and the elution periods was not observed.From the elution curve, it seems possible to separate micro-amounts of Co(I1) and Cu(I1) from macro-amounts of Mn(I1) and Ni(I1). A 20-pmol amount of Ni(I1) and a 0.2-pmol amount of Cu(I1) were loaded on to the column and the elutions were carried out with nitric acid. Ni(I1) was separated from Cu(II), the respective yields being 99 and 70% [Fig. 6(a)]. A similar elution behaviour was observed for the Ni(I1) - Co(I1) separation with 99% recovery for both metal ions [Fig. 6 ( b ) ] . The separation of Ni(I1) - Cu(I1) and Ni(I1) - Co(I1) can be utilised in Ni - Cu base alloys and steel, where trace amounts of cobalt and copper should be removed. As the TiSbA showed extremely high selectivity for Pb(II), the separation of Pb(I1) from divalent transition metal ions was carried out.The tailing observed for some transition metals in the column experiments could be reduced to some extent if the TiSbA column was pre-treated with dilute nitric acid. Improved recovery for these metal ions was observed with a yield higher than 99%. For separation of Pb(I1) from Ni(II), Cd(I1) and Co(II), the column was pre-treated with 0.1 M HN03 and the metal ions were then eluted with nitric acid of different concentrations [Fig. 7(a)]. For separation of Pb(I1) from Zn(I1) and Cu(II), the exchanger was pre-treated with 0.2 M HN03 and the elution was carried out [Fig. 7(b)], the yields being 99% for all metal ions. It could be concluded that better yields for transition metal ions could be achieved by pre-treatment of the TiSbA column with dilute nitric acid.ANALYST, MARCH 1986, VOL.111 343 The TiSbA exchanger showed low affinity for alkaline earth metals8 compared with Zn(II), Co(I1) and Cu(I1). The separation of micro-amounts of Zn(I1) and Cu(I1) from macro-amounts of Mg(II), Ca(I1) and Sr(I1) on a pre-treated column was performed [Fig. 8(a)]. Sharp elution peaks were observed with 99% recovery for all metal ions. Separation of Pb(I1) from Mg(II), Ca(I1) and Sr(I1) was achieved with a 1-pmol loading of each metal ion [Fig. 8(b)] and recoveries were 99% for all the metal ions. In summary, TiSbA can be utilised for the separation of trace amounts of transition metals from salt solutions of alkaline earth metals. The chromatographic separation can be utilised for the determination of lead in a wide variety of minerals and materials.1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. References Clearfield, A., Editor, “Inorganic Ion Exchange Materials,” CRC Press, Boca Raton, FL, 1982. Veselg, V., and Pakarek, V., Talanta, 1972, 19, 219. Abe, M., and Ito, T., Bull. Chem. SOC. Jpn., 1967, 40, 1013. Abe, M., and Uno, K., Sep. Sci. Technol., 1979, 14, 355. Abe, M., and Kasai, K., Sep. Sci. Technol., 1979, 14, 895. Abe, M., and Ito, T., Kagaku Zasshi, 1967, 70, 440. Abe, M., and Hayashi, K., Solvent Extr. Zon Exchange, 1983, 1, 97. Abe, M., Chitrakar, R., Tsuji, M., and Fukumoto, K., Solvent Extr. Ion Exchange, 1985, 3, 149. Abe, M., and Hayashi, K., Hydrometallurgy, 1984, 12, 83. Abe, M., and Furuki, N., Bull. Chem. SOC. Jpn., 1985, 58, 1812. Donaldson, J. D., and Fuller, M. J., J. Znorg. Nucl. Chem., 1968. 30, 1083. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23, 24. 25. 26. 27. 28. 29. 30. Donaldson, J. D., Fuller, M. J., and Price, J. M., J. Znorg. Nucl. Chem., 1968, 30, 2841. Tsuji, M., and Abe, M., Solvent Extr. Ion Exchange, 1984, 2, 253. Shannon, R. D., and Prewitt, C. T., Acta Crystallogr., Sect. B , 1969, 25, 925. Fuller, M. J., Chromatogr. Rev., 1971, 14, 45. Clearfield, A., and Kalnins, J. M., J . Znorg. Nucl. Chem., 1976, 38, 849. Allulli, S . , Ferragina, C., La Ginestra, A., Massicci, M. A., and Tomassini N., J. Chem. SOC., Dalton Trans., 1977, 1880. Baes, C. F., Jr., and Mesmer, R. E., “The Hydrolysis of Cations,” Wiley, New York, 1971. Strelow, F. W. E., Rethemeyer, R., and Bothma, C. J. C., Anal. Chem., 1965, 37, 106. Fritz, J. S . , and Garralda, B. B., Anal. Chem., 1962, 34, 102. Fritz, J. S., and Greene, R. G., Anal. Chem., 1963, 35, 811. Strelow, F. W. E., Victor, A. H., Van Zyl, C. R., and Eloff, C., Anal. Chem., 1971, 43, 870. Strelow, F. W. E., and Sondorp, H., Talanta, 1972, 19, 1113. Strelow, F. W. E., Hanekom, M. D., Victor, A. H., and Eloff, C., Anal. Chim. Acta, 1975, 76, 377. Strelow, F. W. E., and Toerien, F. Von S . , Anal. Chem., 1966, 38, 545. Strelow, F. W. E., Anal. Chem., 1978, 50, 1359. Strelow, F. W. E., Anal. Chim. Actu, 1978, 100, 577. De, A. K., and Chowdhury, K., 1. Chromatogr., 1974,101,73. Alberti, G., Casciola, M., Costantino, U., and Luciani, M. L., J. Chromatogr., 1976, 128, 289. Abe, M., Tsuji, M., Qureshi, S. P., and Uchikoshi, H., Chromatographia, 1980, 13, 628. Paper A51253 Received July 15th, 1985 Accepted November 1 Oth, 1985

ISSN:0003-2654    

DOI:10.1039/AN9861100339

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST, MARCH 1986, VOL. 111 345 Determination of Inorganic and Organomercury Compounds by High-performance Liquid Chromatography - Inductively Coupled Plasma Emission Spectrometry with Cold Vapour Generation Ira S. Krull* and D. S. Bushee Barnett Institute of Chemical Anal ysis and Department of Chemistry, Northeastern University, 360 Huntington Avenue, Boston, MA 021 15, USA and R. G. Schleicher and S. 6. Smith, Jr. Allied Analytical Systems, 590 Lincoln Street, Waltham, MA 02254, USA An inductively coupled plasma (ICP) emission spectrometer was used as a high-performance liquid chromatographic detector for the determination of mercury compounds. Post-column cold vapour generation was used to obtain improved detection limits. The replacement of the conventional polypropylene spray chamber of the ICP by an all glass chamber is described.A comparison of band broadening indicates that the glass chamber is useful when a severe memory effect is observed with the polypropylene spray chamber. Detection limits ranged from 32 to 62 p.p.b. of mercury for four mercury compounds, based on a signal to noise ratio of 2 : 1. This represents a three to four order of magnitude enhancement over detection limits obtained without cold vapour generation. The approach is linear over three orders of magnitude. A blind, spiked distilled water study illustrates the reproducibility and accuracy of the method. Keywords: Mercury determination; high-performance liquid chromatography; inductively coupled plasma; cold vapour generation The most common method of mercury speciation in use today is gas chromatography followed by electron-capture detection (GC-ECD).The poor selectivity of the ECD to organomer- cury species and the time-consuming and elaborate clean-up procedures necessary prior to injection of many samples indicates the need for a simple, more selective approach for the analysis of organomercury and inorganic mercury species. Selectivity in gas-chromatographic analysis has been improved by using atomic-absorption and microwave plasma emission detectors.3-6 The use of high-performance liquid chromatography (HPLC) for mercury speciation has the advantage of simpli- fied sample preparation. In GC-ECD analysis it is essential to form strong, thermally stable derivatives, whereas it is not necessary to derivatise samples prior to analysis by HPLC.7 The combination of HPLC with spectroscopic detection provides a simple and selective method of metal specia- tion.8-15 In order to obtain practical detection limits with these detectors, it was often necessary to generate a more volatile form of the metal of interest.For mercury, this was carried out by forming the cold vapour derivative of the compounds as they eluted from the HPLC column. Previous reports des- cribed relatively complex apparatus for the interface between the HPLC column and the atomic-absorption detector.8J0 In this paper, a simple interface between an HPLC system and an inductively coupled plasma emission spectroscopic detector is described for the speciation of inorganic and organometallic mercury compounds.Experimental Apparatus and Operating Conditions The chromatographic system consisted of two Laboratory Data Control (LDC) (Riviera Beach, FL, USA) Constametric I11 pumps with a gradient controller and a Rheodyne Model 7125 injection valve (Rheodyne Corp., Cotati, CA, USA) fitted with a 2 0 0 4 loop. The separation7 was performed on two Waters Resolve columns (Waters Assoc., Milford, MA, USA), 5 ym C18 stationary phase, 15 cm x 3.9 mm i.d. placed ~~~~ ~ ~ * To whom correspondence should be addressed. in series with a mobile phase consisting of 0.06 M ammonium acetate and 0.005% V/V 2-mercaptoethanol with a gradient from 15 to 75% of acetonitrile. A flow-rate of 1.0 ml min-1 was used for all analyses. The post-column reaction system has been described previously12 and is shown schematically in Fig.1. An aqueous solution of 0.5% m/V sodium tetrahydro- borate(II1) in 0.25 M sodium hydroxide solutio? and a 1.2 M solution of hydrochloric acid served as the two reagents. The reagent flow-rates were 0.5 ml min-l. An Instrumentation Laboratory Model 200 plasma (Allied Analytical Systems, Waltham, MA, USA), modified for autotune operation, was used to monitor the HPLC effluent at a wavelength of 253.7 nm. The polypropylene nebuliser - spray chamber of the ICP showed a significant memory effect for mercury.16 An all-glass chamber (Fig. 2) was developed, which replaced the polypropylene nebuliser - spray chamber completely. A short length of 1/16-in Teflon tubing was inserted into each arm. Inlet A of the chamber was used to introduce an argon purge gas at a flow-rate of 0.4 1 min-1.For optimum performance, the end of the Teflon tubing was turned upwards to direct the gas towards the plasma torch. The tubing was held in place by the walls of the chamber. The inlet B of the chamber served to introduce the HPLC effluent. The Teflon tube in this arm extended about 1 cm below the n Injector n I ICP I 1 Columns 1 I Mixing Glass Fig. 1. Schematic diagram of the HPLC - cold vapour ICP instrumentation346 ANALYST, MARCH 1986, VOL. 111 glass tube, to prevent the gaseous effluent from travelling back up the glass arm. Just prior to the glass chamber, the effluent had passed through the post-column reactor and was in the form of a gaseous mixture. The mercury compounds, now in their cold vapour form, were swept up into the plasma with the purge gas, by way of outlet C.The chamber was connected to the plasma by means of a short length of Tygon tubing. The aqueous mobile phase flowed to waste through the bottom of the chamber, outlet D. A drain trap was positioned directly below the glass chamber. The ICP peak-height response was monitored on a Honeywell Corp. (Minneapolis, MN) strip- chart recorder, and peak-area data were collected on a Radio Shack TRS-80 Model I1 computer (Tandy Corp., Fort Worth, TX, USA). Determination of Cold Vapour Transfer Efficiency Three aqueous solutions of methylmercury chloride, ethyl- mercury chloride and mercury(I1) chloride (0.97, 0.87 and 1.88 p.p.m. of mercury, respectively) were pumped through the cold vapour generation interface into the ICP.Aliquots of 15 ml of waste were collected from the glass chamber after the signal had stabilised. The waste was collected in 30-ml graduated, plastic cups containing distilled water made basic by the addition of about 90 mg of sodium hydroxide. Three aliquots of waste were collected for each mercury solution, and blank samples were obtained by pumping mobile phase through the system between each mercury solution. 152 mrn 36 mm 0.d. 3 mrn i.d. (C) To plasma 3 mm (A) f (€3) From Argon HPLC purge I mm i.d. d i.d. 70 mm 3 mm i.d. Fig. 2. Schematic diagram of glass sample introduction device Each waste sample was transferred into a 50-ml plastic bottle and digested overnight with 0.5 ml of 5% potassium dichromate solution and 5 ml of water - hydrochloric acid - nitric acid (4 + 3 + 1).A standard solution of each mercury compound was prepared at the 100-200 p.p.b. of mercury level. The standard solutions of each mercury compound were treated in the same manner as the samples. The digested waste samples were analysed by direct cold vapour ICP spectrometry using the conventional nebuliser spray chamber assembly. A reducing solution was prepared by adding 50 ml of concentrated sulphuric acid to about 300 ml of distilled water and cooling to room temperature. Sodium chloride (15 g), tin(I1) chloride (25 g) and hydroxylammonium chloride (15 g) were then added, and the solution volume was brought to 500 ml with distilled water.17 Both the sample and reducing solutions were introduced at a flow-rate of 1 ml min-1.Reagents All reagents were of analytical-reagent grade and used without further purification. The mercury compounds and sodium tetrahydroborate(II1) (98%) were obtained from Alfa Products (Danvers, MA, USA). Mercury standard solutions were prepared fresh daily and were stored in a dark, cool place when not in use. Mobile phase and sodium tetrahydrobor- ate(II1) solutions were filtered through 0.45-ym filters (Milli- pore Corp., Bedford, MA, USA). Results and Discussion Evaluation of Glass Chamber A comparison of band broadening was made between the three sample introduction devices. The band broadening was calculated by the method of Foley and Dorsey.l* It was not possible to detect mercury easily by flow injection analysis - ICP spectrometry with the conventional nebuliser in the absence of cold vapour generation. In order to include a value for the band broadening for all three devices, it was necessary to use arsenic as the model compound.Arsenic exhibited no substantial absorbance on to either the polypropylene or glass chambers. The conditions for the formation of arsine post- column were similar to those for the formation of the mercury cold vapour and have been described earlier.12 The results of the flow injection analysis of a solution of 139 p.p.m. of arsenic as arsenite using each introduction device are sum- marised in Table 1. The conventional nebuliser with hydride generation produced a more symmetrical peak than either of the other configurations. The formation of a sharper peak with hydride generation compared with the response without post-column hydride generation can be explained by consider- ing all open tubing after the hydride generation device as a gas-segmented, open-tubular reactor.The concept of segmen- tation in flowing streams with air or gas bubbles has been established as effectively suppressing dispersion of sample zones.19 Of the two interfaces in the hydride formation mode, Table 1. Band broadening and asymmetry in flow injection analysis (FIA) - ICP. Hydride generation conditions for arsenite: 4% sodium tetrahydroborate(II1) solution and concentrated hydrochloric acid introduced at 0.5 ml min - 1 Interface FIA - ICP with polypropylene FIA - hydride generation - ICP with FIA - hydride generation - ICP with ne buliser . . .. . . . . . . polypropylenenebuliser . . . . glass chamber . . . . . . . . Band broadening* Relative band (Pl? broadening 17768 k 434 8160 k 347 12767 k 482 2.2 1 .o 1.6 Asymmetry factor" 1.90 f 0.08 1.69 k 0.06 1.8 k 0.1 * Calculated by the method of Foley and Dorsey.18 Numbers represent the average k standard deviation for at least four injections with each interface.ANALYST, MARCH 1986, VOL. 111 347 I I I I I I 1 I I I I 0 2 4 6 8 10 12 14 16 18 20 22 Timelmin Fig. 3. HPLC - cold vapour ICP chromatogram of mercur species. HPLC on two Waters Resolve columns in series. Gradient elution using (A) 0.06 M ammonium acetate, 0.005% 2-mercaptoethanol; (€37 0.06 M ammonium acetate, 0.005% 2-mercaptoethanol and 75% acetonitrile. Initial [B] = 20%, final [B] = 100%; flow-rate, 1 ml min-1; slope = 5 ; and injection volume, 200 p1 Table 2.Comparison of HPLC - cold vapour - ICP detection limits (p.p.b. of Hg) for sample introduction devices. Conditions as in Fig. 3 Conventional nebuliser interface Glass chamber interface Compound No cold vapour With cold vapour* with cold vapourt CH3HgCl . . . . . . >232 x 103 47 f 4 37 f 2 CH3CH2HgCl . . . . >302 x 103 74 k 10 62 k 10 HgCI2 . . . . . . . . >409 x 103 202 k 30 35 k 15 (CH3)ZHg . . . . . . 267 k 30 p.p. b. Hg 267 * 21 62 ?I 37 * Numbers represent the average f standard deviation for three injections. t Numbers represent the average k standard deviation for at least nine injections. Detection limit values were normalised based on a signal to noise ratio of 2 : 1. the cross-flow nebuliser produced the least peak band broadening.This indicates that the conventional nebuliser would be satisfactory for most applications involving a post-column volatilisation reaction. The glass chamber inter- face need only be used when the possibility of strong absorption of analyte on to the walls of the polypropylene spray chamber exists, as with mercury. Separation of Mercury Compounds Fig. 3 illustrates the chromatographic separation of the four mercury compounds. The separation was complete in less than 20 min. If only dimethylmercury was of interest in a given sample, it would be possible to elute the dimethylmercury in a much shorter time by using an isocratic system consisting of 75% acetonitrile in the mobile phase. Under these conditions, dimethylmercury eluted in 4.5 min, while the other three mercury compounds eluted as a single peak in the solvent front (Fig.4). Detection Limits and Linear Range The detection limits for the four mercury species were determined using the conventional cross-flow nebuliser with- out and with cold vapour generation, and also with the glass chamber with cold vapour generation. Detection limits were defined as the minimum concentration of mercury that gave a detector signal twice the peak-to-peak noise on the strip-chart recorder. These results are compared in Table 2. Using the cross-flow nebuliser without post-column cold vapour genera- tion, the methylmercury chloride, ethylmercury chloride and mercury(I1) chloride were not detected at or below the indicated levels.More concentrated solutions were not tested because of solubility problems and the extreme toxicity of these compounds at high levels. A three to four order of magnitude enhancement was observed in each instance. 0 2 4 Ti me/m in I Fig. 4. HPLC - cold vapour ICP chromatogram of dimethylmercury with mercury(I1) chloride, methylmercury chloride and ethylmercury chloride eluting in the solvent front. Isocratic mobile phase of 0.06 M ammonium acetate, 0.005% 2-mercaptoethanol and 75% acetonitrile. Other conditions as in Fig. 3 The detectability for dimethylmercury was increased by a factor of four by using post-column cold vapour generation with the glass chamber. The low detection limit observed with the conventional nebuliser was probably due to the inherent volatility of the dimethylmercury before derivatisation.Calibration graphs for each of the mercury species were linear over three to four orders of magnitude, ranging from the detection limit to the mid-parts per million region. The correlation coefficients for these calibration graphs ranged from 0.9801 to 0.9995.348 ANALYST, MARCH 1986, VOL. 111 Stability in Aqueous Solution The stability of the four mercury compounds in water was studied. Different combinations of the four compounds were dissolved in distilled water at known levels and the responses of these mixtures were compared with those of standards prepared separately. The results in Table 3 indicate that there was no significant interconversion between mercury(I1) chloride, methylmercury chloride or ethylmercury chloride.Mixtures 4 and 5 were prepared with dimethylmercury in solution with one of the other mercury compounds. In each instance, the dimethylmercury was not completely recovered. Interconversion between the mercury compounds was clearly evident in mixture 6. The complete disappearance of the two mercury forms originally spiked in the mixture was observed, with the corresponding appearance of a peak for methylmer- cury chloride. The instability of the dimethylmercury, and its ready conversion into methylmercury chloride when in the presence of other mercury compounds in an aqueous system, led to its exclusion from the blind spiked distilled water study. Spiked Water Analysis A series of blind spiked, distilled water samples, with known concentrations of the mercury compounds, were studied by Table 3.Mercury interconversion in aqueous solutions Recovery, YO t Mixture* HgC12 CH3HgCl C2HSHgCl (CH3)2Hg 1 96k 11 103f 4 99 f 5 9 2 106 k 11 112 f 11 9 § 3 § 116f 6 1 0 7 f 6 9 4 § 103f 3 § 54 f 7 5 § + 77 f 8 46 f 6 6 0 + § 0 * Mixtures contained between 375 and 675 p.p.b. of mercury and were compared against single standards made up at the same concentration. t Recovery was determined by HPLC - cold vapour ICP spectro- metric analysis of the mixtures immediately following preparation. HPLC - cold vapour ICP conditions as in Fig. 3. Numbers are the average f standard deviation for at least three injections of each mixture. 5 Indicates the formation of methylmercury not initially present in the solution. § Indicates no addition of this mercury species to the mixture. HPLC - cold vapour ICP spectrometry (Table 4).These results indicate a general agreement between the levels of mercury species spiked and the values determined. A linear regression analysis of these results and the actual spiked values gave correlation coefficients of 0.9440, 0.9971 and 0.9869 for mercury(I1) chloride, methylmercury chloride and ethylmer- cury chloride, respectively. The correlation for mercury(I1) chloride was not as good as that for the two organomercury species. This was due to the lack of base-line resolution between the mercury(I1) chloride and the methylmercury chloride, and could be improved by further optimisation of the chromatographic conditions. Conversion/Transfer Efficiency The actual amount of derivatised analyte reaching the plasma was studied.All mercury in the waste samples was converted into the inorganic form and then reduced with the tin(I1) chloride reagent solution. The concentration of mercury in the waste was determined by direct cold vapour ICP spectrometry as described under Experimental. The amount of mercury found in the waste consists of all the mercury not converted into the cold vapour form or swept into the ICP. All other mercury is assumed to have reached the plasma. The concentrations of mercury(I1) chloride, methylmercury chloride and ethylmercury chloride reaching the plasma were 82 k 2, 77.1' rt 0.5 and 79.3 k 0.3, respectively. These values represent the mercury that was converted into the cold vapour form and that was transferred successfully to the plasma during HPLC - cold vapour ICP spectrometry, a combination of conversion efficiency and transfer efficiency.Conclusion A sample introduction method has been developed that provides for the rapid and direct speciation of low levels of mercury compounds in aqueous systems by HPLC - cold vapour ICP spectrometry. This approach could be used in a number of environmental, biological, industrial and toxicolog- ical applications. We acknowledge the assistance of W. Lacourse, S . Colgan and M. Swartz of Northeastern University in the preparation of various spiked water samples. We appreciate the interest and encouragement of K. Panaro of the Boston District Food and Drug Administration. This work was supported by a research and development contract from Allied Analytical _ _ _ _ _ _ _ ~ ~ ~ ~ ~ ~ Table 4.HPLC - cold vapour ICP of blind spiked distilled water. Conditions as in Fig. 3 Mixture Hg spiked, Hg measured, No. Hg species p.p.b. p.p.b.* Recovery, % * 1 HgC12 CH3HgCl CH3CH2HgCl CH3HgCl 2 HgC12 CH3CH2HgCl 3 HgC12 CH3HgCl CH3CHZHgCl 4 HgClz CH3HgCl CH3CHZHgCl 5 HgC12 CH3HgC1 CH3CH2HgCl 6 HgClz CH3HgCl CH3CHzHgCl 0 290 179 46 1 0 303 0 572 242 248 1090 446 419 537 0 0 767 575 NDt 310 f 20 190 + 20 530 f 60 ND 290 k 30 ND 610 f 40 270 f 20 220 f 40 1040 f 50 510 f 30 380 f 70 570 f 20 0 ND 780 f 20 580 f 20 -$ 107 f 8 110 f 10 120 f 10 96 f 9 107 k 7 111 k 8 90 f 20 95 f 5 114 f 4 90 f 20 106 f 3 - - - - 102 f 2 101 f 3 * Numbers represent the average k the standard deviation of a minimum of three injections of each mixture.t ND = not detected. 5 Indicates no addition of this mercury species to the mixture.ANALYST, MARCH 1986, VOL. 111 349 Systems to Northeastern University that allowed us to undertake and complete the work described. This is contribu- tion number 262 from the Barnett Institute of Chemical Analysis and Materials Science at Northeastern University. 1. 2. 3. 4. 5. 6. 7. 8. 9. References Rodriquez-Vazques, J. A., Talanta, 1978, 25, 299. Hight, S. C., and Capar, S. G., J. Assoc. Off. Anal Chem., 1983, 66, 1121. Bye, R., and Paus, P. E., Anal. Chim. Acta, 1979, 107, 169. Talmi, Y . , Anal. Chim. Acta, 1975, 74, 107. Bzezinska, A., Van Loon, J., Williams, D. Oguma, K., Fuwa, K., and Haraguchi, I. H., Spectrochim. Acta, Part B, 1983,38, 1339. Sklarew, D. S., Olsen, K. B., and Evans, J. C., Paper presented at the 189th ACS National Meeting, Miami Beach, FL, April 28-May 3, 1965, ANAL. 61. MacCrehan, W. A., and Durst, R. A., Anal. Chem., 1978,50, 2108. Holak, W., Analyst, 1982, 107, 1457. Holak, W., J. Liq. Chromatogr., 1985, 8, 563. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. Fujita, M., and Takabatake, E., Anal. Chem., 1983, 55, 454. Holak, W., “Innovative Techniques for the Analysis of Iodine and Methylmercury,” 10th Annual Spring Training Workshop, AOAC, Dallas, TX, April 10, 1985. Bushee, D., Krull, I. S . , Demko, P., and Smith, S. B., Jr., J. Liq. Chromatogr., 1984, 7 , 861. Krull, I. S . , in Lawrence, J. F., Editor, “Liquid Chromato- graphy in Environmental Analysis,” Humana Press, Clifton, NJ, 1984, Chapter 5. Krull, I. S., Panaro, K. W., and Gershman, L. L. J . Chromatogr. Sci., 1983, 21, 460. Krull, I. S., Bushee, D., Savage, R. N., Schleicher, R. G., and Smith, S. B., Jr., Anal. Lett., 1982, 15A, 267. Kuldvere, A., Analyst, 1982, 107, 179. Holak, W., J. Assoc. Off. Anal. Chem., 1983 66, 1203. Foley, J. P., and Dorsey, J. G., Anal. Chem., 1983, 55, 730. Frei, R. W., in Frei, R. W., and Lawrence, J. F., Editors, “Chemical Derivatization in Analytical Chemistry,” Volume 1, Plenum Press, New York, 1981, Chapter 4. Paper A51226 Received June 24th, 1985 Accepted August 2 7th, 1985

ISSN:0003-2654    

DOI:10.1039/AN9861100345

出版商:RSC    

年代:1986

数据来源: RSC