ISSN:0003-2654    

DOI:10.1039/AN98611FX045

出版商:RSC    

年代:1986

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN98611BX047

出版商:RSC    

年代:1986

数据来源: RSC

ISSN:0003-2654    

DOI:10.1039/AN98611BP049

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST, DECEMBER 1986, VOL. 111 Edit oria I 1353 Publication of Work on Spectrophotometric Methods for the Determination of Common Metals “The Analyst publishes papers on all aspects of the theory and practice of analytical chemistry.” (Instructions to Authors.) However, a large proportion of its contents is devoted to the publication of research papers reporting the results of investigations into new , or developments of well established analytical methods. In general, these investigations may be divided into three broad categories: (a) Project-related analytical problems; (b) original analytical method development; and (c) methods to teach the procedures required for successful research in analytical chemistry. (a) Project-oriented research produces an answer to a specific problem and if it is an area in which other workers are involved it will be of interest to them and will be well worth publishing in The Analyst.(b) Original analytical method development involving new reagents to solve difficult problems or new developments improving the efficiency of an existing procedure or extending a technique will generally be of interest to other workers and The Analyst will be an appropriate journal for publication of the outcome of such work. (c) Projects undertaken in which the prime aim is to teach procedures for successful research in analytical chemistry, such as in the training of a graduate student for a research degree, do not always lead to work that is of interest to other analytical chemists even when it is of high quality.In this respect , particular concern is expressed over the number of papers submitted reporting “new” spectrophotometric methods for easily determined common metals, and for iron and copper in particular. Authors of such papers should be aware that referees are reluctant to recommend publication and lack of interest or a potentially useful application are valid criteria for rejection. A fair proportion of papers submitted to The Analyst and other reputable journals unfortunately fall into this category and it seems desirable for this proportion to be reduced so that valuable space is not wasted on reports of methods that no-one is ever likely to use. It is a fact that the introduction, in large numbers, of atomic absorption spectrometers some twenty years ago, bringing flame atomic absorption (FAA) spectroscopy within the reach of most laboratories, revolutionised trace metal analysis.Fast , inexpensive and reliable determination by FAA was possible for most of the metallic elements that had previously been determined colorimetrically and whereas spectrophotometric methods continue to be of great value in certain areas, their use for trace metal analysis has become minimal, except where a constant matrix enables the process to be automated. The trend away from such methods has been further highlighted by the availability at reasonable cost of inductively coupled plasma atomic emission spectroscopy (ICP-AES) , which has proved to be able to determine most of the refectory elements which are poorly determined by FAA.ICP-AES is inherently even faster than FAA and therefore more economical when a large number of determinations is to be performed. This is not to say that spectrophotometric methods for trace metal analysis are of no further interest, but the main applications now are for methods that can be readily auto- mated or have applications in some of the newer techniques such flow injection analysis (FIA). It is also worth remember- ing that there are a number of elements, particularly some of the metalloids and the non-metals, which are not readily amenable to analysis by FAA or ICP-AES and so there is a need for the continuing development of other analytical methods for these species, and for anions in general, where current spectrophotometric methods are less than ideal, often being time consuming and lacking sensitivity or selectivity.It therefore seems unfortunate that, with so many chal- lenges left to those who wish to work in the field of spectrophotometric trace analysis, so much effort is still being put into developing reagents and methods for the common metals, particularly iron and copper, for which there are already an enormous number of spectrophotometric pro- cedures, and which are readily determined by both FAA and It happens that of all the elements in the Periodic Table, iron and copper may both be determined with high sensitivity and selectivity using reagents based on the ferroin arrange- ment. Such reagents are selective for iron and with suitable substitution can be made essentially specific for copper.The effect of the substitution of various groups near to the basic ferroin entity on both sensitivity and selectivity is well known and although it is a simple matter to produce new reagents on this theme, there is little likelihood of them every being used. There is even less reason to look for new reagents for copper, as apart from the highly selective substituted ferroin-based reagents, there are a number of highly selective procedures based on the high stability of copper - dithocarbamate complexes at very high acidities. However, in spite of the lack of need for reagents for iron and copper, The Analyst still receives far more papers devoted to the determination of iron and copper than any other elements. Much of this work is of high quality and is published, but sadly it is of such little applicability, owing to the plethora of similar already available reagents and more convenient alternative methods, that few analysts will even read it, never mind use the method! Perhaps a reason for this continuing exploration of well trodden paths of reagents for iron and copper is that the recipe for “success” for such a project is simple and the student is assured of a “satisfactory” outcome to the project.However, it would be much more satisfactory to tutor, student and other analysts if the work led to methodology that had a realistic chance of becoming part of the analytical chemist’s arsenal of methods used to investigate the wide variety of problems we are asked to solve, rather than providing a wide variety of methods to solve non-existent problems. Profitable areas for research have previously been men- tioned (reagents for anions, metalloids, non-metals, a few metals such as thorium and cerium, which are very difficult to determine by FAA and cause difficulties in ICP-AES, and systems that have characteristics making them suitable for automatic analysers and FIA). Research projects on spectro- photometric reagents to solve problems in these areas provide just as good training for postgraduate students as continually “reinventing the wheel” with reagents for iron and copper. Such work, however, would have numerous applications and would be of interest to other analytical scientists and would therefore be very welcome in The Analyst. C. A. Watson ICP-AES.

ISSN:0003-2654    

DOI:10.1039/AN9861101353

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST, DECEMBER 1986, VOL. 111 1355 Simultaneous Determination of Zinc, Cadmium and Lead in Manganese Sulphate Electrolyte by Differential-pulse Anodic-stripping Voltammetry Samuel B. Adeloju Trace Analysis Unit, Division of Chemical and Physical Sciences, Deakin University, Victoria 32 17, Australia and Tam Tran BHP Central Research Laboratories, PO Box 188, Wallsend 2287, Australia Conditions are described for the direct simultaneous determination of zinc, cadmium and lead in process manganese sulphate electrolyte by differential-pulse anodic-stripping voltammetry. The suitability of the technique in this medium is influenced by the solution pH, scan rate and pulse height. Reliable determinations of the elements are accomplished in the concentrated electrolyte at pH 4.1 with a scan rate of 8 mV s-1 and a pulse height of 50 mV.Under these conditions, the limits of detection with a deposition time of 30 min are 0.05, 0.03 and 0.10 pg I-’ for the three elements, respectively. The precision of the method is also satisfactory with relative standard deviations of 0.8, 1.5 and 1.2% for zinc, cadmium and lead, respectively, with 10 samples at the 1 pg 1-1 level. However, the high manganese sulphate background is inadequate for copper determination at concentrations less than 50 pg 1-1. Keywords; Anodic-stripping voltammetry; cadmium determination; lead determination; zinc determination; manganese sulphate electrolyte Electrolytic manganese dioxide (EMD) is widely used as a cathode active material in dry cells and is usually deposited from manganese sulphate solution.Unfortunately, the heavy metal impurities in such solutions may be co-deposited or adsorbed on to the EMD during its production and may, consequently, be leached out into the battery electrolyte during storage. These impurities will, in turn, corrode the anode and eventually decrease the battery performance. For these reasons, stringent specifications have now been imposed on the EMD manufacturers to produce high-quality cathode active materials that contain relatively low concentrations of most heavy metal impurities. 1 This requires extensive purifi- cation of the process manganese sulphate electrolyte prior to the electrodeposition of the EMD .* Successful purification will rely on the availability of sensitive analytical techniques that can reliably determine ultra-trace amounts of impurities in the electrolyte. The reported levels of some of the heavy metal impurities in manganese sulphate electrolyte and EMD products are often less than the detection limits (1 mg 1-1 or 1 pg g-1) of the techniques currently used for analysis.3 Evidently, there is a need for the development of sensitive and selective analytical techniques that can be used in conjunction with the purifica- tion process in order to lower the ultimate concentrations of the impurities in the process electrolyte, prior to the EMD production.Ideally, such techniques should be amenable to both plant monitoring and the reliable determination of the heavy metal impurities in the EMD products. The only restriction in this regard is the high manganese sulphate background, which may cause severe interference problems with some of the available sensitive analytical techniques.The possible exception to this type of interference is voltammetry, which may be able to utilise the high background as a suitable “supporting electrolyte” for such determinations. However, no previous application of voltammetric techniques to the determination of heavy metal impurities in manganese sul- phate electrolyte has been reported. In the work reported in this paper, the suitability of differential-pulse anodic-stripping volt ammetry (DPASV) for the simultaneous determination of zinc, cadmium and lead in process manganese sulphate electrolyte was investigated. In particular, the critical dependence of the reliability of this method on pH and some instrumental parameters was carefully examined.Experimental Reagents and Standard Solutions The acids and ammonia used were of Aristar purity (BDH Chemicals) and the other reagents were of analytical-reagent grade. Distilled, de-ionised water, prepared as previously described,4 was used for all sample and solution preparations. Ammonia buffer (1 M, pH 9.5) was prepared by mixing equal volumes of 4 M ammonia and 2 M acetic acid solutions. Stock solutions (1 g 1-1) of zinc, cadmium, lead and copper were prepared by dissolving appropriate amounts of the chloride and nitrate salts in 1 M hydrochloric acid. The required daily standard (1 mg 1-1) was prepared weekly by dilution of the stock with 0.1 M hydrochloric acid.Instrumentation An EG & G Princeton Applied Research microprocessor- based polarographic analyser (PAR Model 384), equipped with a PAR Model 303 static mercury drop electrode and a PAR Model 305 stirrer, was used to record all stripping voltammograms. The electrode compartment consisted of a hanging mercury drop electrode (HMDE), a silver - silver chloride electrode (saturated KCI) and a platinum wire electrode as its working, reference and auxiliary electrodes, respectively. Solution pH measurements were made on an Activon Scientific (Sydney, Australia) portable pH/mV meter. Standard solutions of zinc, cadmium and lead were added to the polarographic cell with fixed volume Soccorex micropipettes with disposable tips. Glassware All glassware and polyethylene bottles were soaked in 2 M nitric acid for at least 7 days and rinsed several times with distilled, de-ionised water prior to use.Between experiments, the used glassware and bottles were soaked in 2 M nitric acid for at least 12 h and again rinsed several times with distilled, de-ionised water before use. Plant Process Electrolyte For preliminary investigations, a solution of 1 M manganese sulphate was prepared by dissolving appropriate amounts of1356 ANALYST, DECEMBER 1986, VOL. 111 the analytical-reagent grade salt in distilled, de-ionised water. This solution (1 1) was then purified by adding 0.5 g of calcium sulphide, stirring with a glass rod to mix thoroughly, and was left to stand for 1 h before finally being filtered through a Whatman No. 541 hardened ashless paper.The resulting pink solution was left overnight, filtered again the next day and 100 yl of concentrated HN03 were added before aliquots for voltammetric measurement were taken. The ammonia buffer was used to adjust the solution pH to the desired value for the ASV determination. The process manganese sulphate electrolyte used in the production of EMD was provided by Broken Hill Proprietary Ltd. (Wallsend, NSW, Australia). The electrolyte contained about 1.2 M manganese sulphate and had a pH of 3.0. Stripping Voltammetric Determinations An aliquot (5 ml) of the electrolyte was transferred into the polarographic cell, de-oxygenated for 5 min and maintained under a flow of nitrogen during the experiment. The three elements were then determined by ASV at the HMDE using the following conditions: operating mode, DPASV; deposi- tion potential, -1.15 V vs.Ag - AgCl (saturated KC1); final potential, -0.25 V; scan rate, 8 mV s-1; duration between pulses, 0.5 s; modulation amplitude, 50 mV; deposition time, 285 s (stirred); and equilibration period, 15 s (unstirred). The deposition of the three elements on to the mercury electrode was achieved by using a fast stirring rate and a medium-sized drop with a surface area of 0.015 cm2. The concentrations of the three elements in the electrolyte were determined by the standard additions method using the peaks that appear at about -1.0, -0.7and -0.4Vvs. Ag-AgClatpH4.1forzinc, cadmium and lead, respectively. The solution was de-oxygen- ated after each addition for 30 s prior to the ASV measure- ment.Working Area All reagent preparations and sample manipulations were carried out in a Class 100 clean room, controlled at a temperature of 22.5 k 0.5 "C. All stripping voltammetric determinations were carried out in a Class 1000 clean room under similar temperature control. Both of these laboratories form part of the Deakin University Trace Analysis Unit. Results and Discussion pH Dependence The reliable determination of zinc, cadmium and lead in the manganese sulphate electrolyte by ASV is dependent on the pH of the solution. The results obtained in this study indicate that the sensitivity and resolution of the stripping peaks for the three elements varied considerably with increasing solution pH. Generally, as can be seen from the data in Table 1, there is no significant difference in the sensitivity of the cadmium peak at pH 22.9, whereas the sensitivities for the lead and zinc peaks varied considerably with increasing pH.Careful exam- ination of these data reveals that the optimum sensitivities for lead and zinc were obtained at pH 4.1 and 7.7, respectively. The data also indicate that the lowest sensitivity for lead was obtained at pH 7.7, whereas the zinc peak was reasonably sensitive at pH 4.1. From these observations, it was concluded that pH 4.1 is adequate for the reliable simultaneous determination of the three elements in the manganese sulphate electrolyte by ASV. The data in Table 1 also indicate that the chosen deposition potential has some influence on the sensitivities of the zinc and lead peaks between pH 1.8 and 5.1.The more negative deposition potential (- 1.15 V) gave a better sensitivity for the zinc peak, whereas the lead peak seems to be influenced by both the solution pH and the chosen deposition potential. Evidently the use of a deposition potential of - 1.15 V is useful in maintaining an adequate sensitivity for the three elements in solution between pH 1.8 and 5.1. However, beyond pH 7.0 the use of a deposition potential of -1.2 V is necessary to obtain an adequate stripping peak for zinc, owing to the negative shift in its peak potential with increasing solution pH. The observed reduction in the sensitivity of the lead peak beyond pH 4.1 may be associated with the increasing tendency to precipitate manganese hydroxide from the solution.This view is supported, to some extent, by the sudden change in the solution colour from pink to orange at pH > 5.0. Eventually, at pH > 7.7, a brownish orange precipitate is formed on the addition of more buffer solution, or when left overnight. It is also likely that the increasing tendency to form manganese hydroxide in the electrolyte at the higher pH aids the precipitation of some of the lead and consequently results in the decreasing peak current measurements for this element. Another interesting observation in the ASV determination of the three elements in the manganese sulphate electrolyte was the absence of a voltammetric response for copper. The only response for the element was noted at pH 2.9 (Fig. l ) , but no increase in the peak current was observed on the addition of 4 pg 1-1 of the analyte.Under normal conditions (in the absence of such a high background), less than 4 pg 1-1 of the element can be determined in a supporting electrolyte such as 0.1 M hydrochloric acid or acetate buffer solution by ASV. It is conclusive, therefore, that the high manganese sulphate background interferes seriously with the ASV determination of copper, possibly via intermetallic compound formation. Copper is known to form intermetallic compounds with most elements.5 However, a definite and quantitative r zsponse was obtained for the element, as shown in Fig. 1, at concentrations 250 yg 1-1. This level is considerably higher than the usual detection limit (0.1 yg 1-1) obtained in other supporting electrolytes.6 Influence of Scan Rate and Pulse Height The sensitivity and resolution of the stripping peaks obtained for zinc, cadmium and lead in the manganese sulphate electrolyte were also influenced by a number of instrumental parameters.In particular, the chosen scan rate and pulse height had a considerable influence on the sensitivity and resolution of the stripping peaks. Fig. 2 shows that the resolution of the peaks decreased with increasing scan rate. However, no significant difference was observed in either the resolution or sensitivity of the stripping peaks with the slower scan rates (S4 mV s-1). Even with the use of a scan rate of 8 mV s-1, only slight decreases in peak currents were observed for cadmium and lead, but resolution was still maintained. In contrast, the use of a faster scan rate (310 mV s-1) resulted in a considerable reduction in the peak currents and affected the resolution for both elements.From these observations, it was concluded that the resolution and sensitivity obtained with a Table 1. Influence of the pH of manganese sulphate electrolyte on the sensitivity of the stripping peaks for zinc, cadmium and lead SensitivityhA (pg l - I ) - l * PH Ag - AgCl Zn Cd Pb EdJV V S . 0.9 1.8 2.9 3.8 4.1 4.7 5.1 6.3 7.7 -1.10 -1.10 -1.15 -1.10 -1.15 -1.10 -1.10 -1.15 -1.10 -1.15 -1.15 -1.20 - 3.0 5.1 9.8 9.8 10.3 13.3 10.0 15.0 15.3 17.0 21.5 22.8 23.0 28.5 28.5 29.8 29.8 29.3 29.5 30.8 32.3 29.8 4.0 4.5 4.5 5.0 8.3 8.3 8.8 9.0 7.3 6.5 4.9 1.9 * Based on addition of 4 pg I-' of the analytes, t, = 300 s, 8 mV s-I.ANALYST, DECEMBER 1986, VOL.111 I 1357 I scan rate of 8 mV s-1 are adequate for the reliable and rapid determination of the three elements. Consequently, this scan rate was used for all determinations in this work. Similarly, the applied pulse height (or modulation ampli- tude) also influenced the sensitivity and resolution of the stripping peaks for zinc, cadmium and lead in the manganese sulphate electrolyte. The results in Fig. 3 show that the peak currents for the three elements increased with increasing pulse height, except at values >50 mV where the sensitivity of the lead peak was reduced. In addition, the resolution of the three stripping peaks was affected by the use of high pulse heights. On this basis, a pulse height of 50 mV was chosen as the optimum with respect to the sensitivity and resolution of the three stripping peaks.Cd Zn I 1.15 0.35 - Etv Fig. 1. Simultaneous determination of zinc, cadmium, lead and copper in manganese sulphate electrolyte by DPASV. Electrolysis time ( t e ) , 120 s; pH, 4.1; concentration of analytes added, 205 pg 1--1; other conditions as described under Experimental Application to Process Electrolyte The ultimate adequacy of the established ASV conditions for the reliable determination of zinc, cadmium and lead in the process plant electrolyte is dependent on the linear concentra- tion ranges for the three elements. It can be expected that the concentrations of the elements in such samples will vary bepending on the effectiveness of the chemical purification step.It is therefore desirable to incorporate some flexibility in the method for handling the possible variations in the concentrations at various stages of the purification process. Fig. 4 shows that the use of a deposition time of 120 s with the established ASV conditions gave linear calibration graphs for zinc and cadmium up to 200 pg 1-1, and up to 100 pg 1-1 for lead. The observed increase in the lead peak currents at concentrations >lo0 pg 1-1 (Fig. 4) suggest that the matrix or intermetallic effects on the ASV determinations of the element were progressively reduced at the higher concentra- tions. Nevertheless, the linear concentration range for lead can also be extended to 200 pg 1-1 by use of a 60 s deposition time. However, as indicated by the data in Table 2, the detection limit for lead was also affected by the chosen deposition time.In general, the use of a 60 s deposition time is adequate for the reliable determination of ultra-trace amounts of the three elements in the concentrated manganese sulphate electrolyte. Providing that the process plant electrolyte can be adequately purified, as little as 0.05 pg 1-1 of Zn, 0.03 pg 1-1 of Cd and 0.10 pg 1-1 of Pb can be reliably determined by the ASV method. These estimated limits of detection are compar- able to those obtained in various supporting electrolyte^^^^ and, hence, indicate that the high manganese sulphate background did not interfere to any great extent with the simultaneous determination of the three elements by ASV. At concentrations greater than 200 pg 1-1, the three elements were rapidly determined in the manganese sulphate elec- trolyte by differential-pulse polarography (DPP).This pro- l 1.15 0.35/1.15 0.35 - E N Fig. 3. Influence of pulse height on the resolution and sensitivity of zinc, cadmium and lead stri ping peaks. Pulse heights: ( a ) 10 mV; ( b ) 25 mV; ( c ) 50 mV; and (8 100 mV; scan rate, 8 mV s-1; analyte concentration and pH as in Fig. 21358 ANALYST, DECEMBER 1986, VOL. 111 2 1 B A 1.0 0 50 100 150 200 Concent ra t i on/pg I-’ Fi . 4. Calibration graphs obtained for (A) zinc, (B) cadmium and (C! lead in manganese sulphate electrolyte by DPASV. Conditions as in Fig. 1, except for analyte concentrations 1.2 0.8 0.9 -EN 0.3 Fig. 5. Sequential simultaneous determination of zinc, cadmium and lead in process plant electrolyte by ( a ) DPP and (b) DPASV. Conditions for ( a ) : pulse height, 100 mV; 4 mV s-l; drop time, 1 s .Standard additions: (A) 0; (B) 1; (C) 2; and (D) 3 mg 1-1 of Zn. Conditions for ( b ) : te = 120 s. Standard additions: (A) 0; (B) 1; (C) 3; and (D) 5 pg 1-l (Cd, Pb). pH 3.0; other conditions as described under Experimental Table 2. Estimated limits of detection* for zinc, cadmium and lead in manganese sulphate electrolyte with different deposition times Deposition time/s Zn/pg 1- Cd/pg I-* Pb/pg 1- 1 60 2.0 1 .o 3.0 120 1 .o 0.5 2.0 300 0.4 0.2 0.6 600 0.2 0.1 0.3 1800 0.05 0.03 0.1 * Estimated from the data in Table 1 and Fig. 5 , pH 4.1. vides suitable flexibility in handling possible variations in the analyte concentrations during the chemical purification of the process electrolyte.The application of the ASV method to the determination of the three elements in a process plant electrolyte supplied by Broken Hill Proprietary Ltd. revealed that zinc is present at excessively higher concentrations than cadmium and lead. Based on the peak current measurements it was determined that the zinc concentration in the electrolyte is >500 pg 1-1. The use of ASV for such high concentrations is not recom- mended as it may result in serious contamination of the electrode.8 As a result, zinc determination in the electro- lyte was accomplished by DPP, whereas cadmium and lead were determined by ASV. Fig. 5 shows the utilisation of both techniques for the reliable determination of the three elements in a single electrolyte sample.The concentrations of the three elements in the plant electrolyte determined by this approach with standard additions were 625 2 5 pg 1-1 of Zn, 1.35 k 0.02 pg 1-1 of Cd and 4.63 k 0.06 pg 1-1 of Pb. Alternatively, the zinc in the sample may be determined by ASV after dilution, but such an approach is more time consuming and requires the use of a separate sample aliquot for the direct determination of cadmium and lead. Fig. 5 also shows that it was not necessary to adjust the pH of the process electrolyte to the recommended value (pH 4.1) for the cadmium and lead determinations as the data in Table 1 indicate that there is no significant difference in the sensitivi- ties obtained for both elements between pH 2.9 and 4.1. Evidently, the ASV method is adequate for the reliable determination of ultra-trace amounts of the three elements in manganese sulphate solution and can readily be combined with other less sensitive voltammetric techniques such as DPP to handle possible variations in analyte concentrations during the chemical purification of the electrolyte.Conclusion The high manganese sulphate background in the process plant solution was suitable as a “supporting electrolyte” for the simultaneous determination of zinc, cadmium and lead by ASV. Under the established optimum conditions, as little as 0.05, 0.03 and 0.10 pg 1-1, respectively, can be reliably determined. The precision of the method was also satisfactory with a relative standard deviation between 1 and 2% for the three elements. However, the determination of copper in the electrolyte was considerably affected, possibly as a conse- quence of an intermetallic interference, but its determination was still possible at concentrations >50 pg 1-1. More work is being undertaken to improve the sensitivity and detection limit of the ASV method for this element. The authors are grateful to Broken Hill Proprietary Ltd., NSW (Australia), for providing the funds for this research and giving permission for publication. 1. 2. 3. 4. 5. 6. 7. 8. References Deane, M. E., Inst. Min. Metall. Trans., Sect. A, 1985, 94, A169. Kozawa, A., in Kordesch, K. V., Editor, “Batteries, Volume One, Manganese Dioxide,” Marcel Dekker, New York, 1974, Chapter 3, p. 385. Toyo Soda Manufacturing Co. (Japan), in Kozawa, A., and Nagayome, M., Editors, “Proceedings of the IBA Symposium, Brussels, 1983,” International Battery Material Association, Cleveland, OH, 1984, pp. 275-279. Adeloju, S . B., Bond, A. M., Briggs, M. H., and Hughes, H. C., Anal. Chem., 1983, 55, 2076. Wang, J., “Stripping Analysis,” VCH, Deerfield Beach, FL, Adeloju, S. B., Bond, A. M., and Hughes, H. C., Anal. Chim. Acta, 1983, 148, 59. Adeloju, S. B., and Bond, A. M., Anal. Chem., 1985,57,1728. Adeloju, S. B., Bond, A. M., and Briggs, M. H., Anal. Chem., 1985, 57, 1386. Paper A61155 Received May 21st, 1986 Accepted July 3rd, 1986 1985, pp. 93-98.

ISSN:0003-2654    

DOI:10.1039/AN9861101355

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST, DECEMBER 1986, VOL. 111 1359 Coated-wire Ion-selective Electrode for the Determination of Thal I iu m( 111) Concepcion Sanchez-Pedrefio," Joaquin A. Ortufio and Maria C. Torrecillas Department of Analytical Chemistry, Faculty of Sciences, University of Murcia, 3007?-Murcia, Spain The construction, performance characteristics and applications of a coated-wire thallium(lll) ion-selective electrode, based on the ion pair between TIC14- and the 1,2,4,6-tetraphenylpyridinium cation in a poly(viny1 chloride) matrix, are described. The influence of membrane composition, hydrochloric acid concentration and foreign ions was investigated. The electrode shows a near-Nernstian response over the thallium(lll) concentration range IO-5-lO-* M, with good selectivity and precision. Applications to the potentiometric titration of 0.255.0 mg of thallium(lll) and to the direct potentiometric determination of thallium in sphalerites and zinc concentrates are reported.Keywords: Thallium ion-selective electrode; coated-wire electrode; 1,2,4,6-tetraphenylpyridinium tetrachlorothallate(lIl); potentiometric titration; thallium(111) determination An exciting advance was made in ion-selective electrodes by Catrall and Freiserl when they developed coated-wire ion- selective electrodes (CWE). CWEs have become popular owing mainly to their simple and cheap construction. Coated-wire electrodes, with ion-association compounds formed between negatively charged halide complexes of the metal ions to be determined and strong hydrophobic cations as the electroactive materials, have been reported previously.2-* Iron,2 mercury,3 copper,4 zinc,5 bismuth6 and gold7 have been determined as their chloro complexes and cobalt as its tetrathiocyanato cobaltate(I1) complex.8 Aliquat 336s has been frequently used as a counter ion.2-6 The formation and extraction of an ion-association com- pound of tetrachlorothallate(II1) and the 1,2,4,6-tetra- phenylpyridinium cation have been reported.9 The high degree of extraction and good selectivity of the ion pair of thallium(II1) led us to investigate this compound as the electroactive material in a thallium(II1) CWE.The behaviour of electrodes based on organic ion exchangers depends on the extraction characteristics of the compounds involved. 10 In this paper we report the construction and applications of an electrode for the selective determination of thallium.Experimental Apparatus Potentials were measured using a Philips PW9415 ion-selec- tive meter and an R44/2-SDl calomel double-junction refer- ence electrode, containing 1 M KCl solution in the outer compartment. All solutions were stirred continuously with a magnetic stirrer in a thermostatically controlled 100-ml beaker (25 k 0.1 "C). Automatic titrations were performed with a Radiometer ABU12b autoburette and an OmniScribe D5000 chart recorder. Reagents All inorganic chemicals used were of analytical-reagent grade. Doubly distilled water was used throughout. Tetrahydrofuran (THF) and dibutyl phthalate (DBP) were supplied by Merck. The poly(viny1 chloride) (PVC) of high relative molecular mass used for the ion-selective electrode was from Fluka.1,2,4,6- Tetraphenylpyridinium acetate (TPPA) solution, 0.1 M. Prepared by the method of Chadwickll and standardised * To whom correspondence should be addressed. gravimetrically with perchlorate. Working solutions were prepared by dilution with doubly distilled water. Thallium(III) standard solution, 0.01 M. Prepared by dissolving thallium(II1) chloride in 0.1 M hydrochloric acid and standardised by titration with EDTA. l2 Electroactive Material This was prepared by adding slowly a slight excess of 1,2,4,6-tetraphenylpyridinium acetate (10.5 ml of 0.02 M solution) to 20 ml of 0.01 M thallium(II1) solution in 1 M hydrochloric acid. The mixture was stirred for 30 min and the resulting white precipitate was filtered on a sintered-glass crucible (porosity 4), washed with doubly distilled water and dried at 100 "C.Construction of Electrodes The coated-wire electrodes were constructed as described elsewhere? Powdered PVC, dibutyl phthalate (plasticiser) and the electroactive material were dissolved in tetrahydro- furan; coating solutions and membrane compositions are shown in Table 1. A platinum wire, about 2 cm long and 1.0 mm in diameter, sealed into the end of a glass tube and soldered on to a shielded cable, was dipped into this solution 20 times and the solvent was evaporated with an air gun each time. A membrane was formed on the platinum surface and the electrode was allowed to set overnight. Conditioning and Direct Potentiometric Measurements These electrodes were conditioned by soaking with constant stirring in a 10-3 M thallium(II1) solution with the same hydrochloric acid concentration as the standards, until the electrodes gave a constant potential.The same procedure was then followed with a solution containing only hydrochloric acid of the same concentration. The standard thallium(II1) solutions were then determined in ascending order of concen- tration. The electrodes were stored dry and conditioned as above before a series of determinations. The first conditioning time was about 90 min, and then only 30-40 min for successive uses. Potentiometric Titrations An aliquot of the sample solution containing 0.25-5.0 mg of thallium(II1) was pipetted into the 100-ml beaker, 5 ml of 5 M HCl were added and the solution was diluted to 25 ml with doubly distilled water.This was titrated automatically with1360 ANALYST, DECEMBER 1986, VOL. 111 Table 1. Preparation of the coating solutions and compositions of the membranes Coating solution* Membrane composition, % m/m Electroactive Electroactive Membrane PVC/mg DBP/mg material/mg PVC DBP material A . . . . . . 69.4 136.9 9.8 32.1 63.3 4.5 B . . . . . . 103.1 107.4 10.0 46.8 48.7 4.5 c . . . . . . 134.4 67.4 9.4 63.6 31.9 4.5 * Dissolved in 3 ml of tetrahydrofuran. 1 X 10-3-2 x 10-2 M TPPA solution. Titration rates were held constant at 0.3 ml min-1 for samples containing 0.25-1.25 mg of thallium and at 0.15 ml min-1 for larger amounts. Potentials were then monitored with the thallium(II1) CWE. Determination of Thallium in Sphalerites and Zinc Concen- trates The samples were dissolved in aqua regia and boiled nearly to dryness three times with doubly distilled water, to reduce the acidity.The sample solution was transferred into a calibrated flask, diluted and the hydrochloric acid concentration adjus- ted to 1 M. Thallium(II1) was determined by direct poten- tiometry, measuring the potentials of both the standards and the sample. Results and Discussion Composition of the Membrane Fiedler and RfiiiCka13 have suggested several suitable plasti- ciser - polymer combinations. Three membrane compositions were investigated by varying the ratio DBP to PVC (Table 1). The responses for several thallium(II1) concentrations in 1 M hydrochloric acid solution are shown in Fig. 1. As can be seen, only membrane B, corresponding to a 1 : 1 m/m DBP to PVC ratio, shows a fast and stable response.Membrane A (2: 1 ratio) shows a severe drift, as has been observed in some CWE by other workers at high plasticiser contents. 14 Membrane C (1 : 2 ratio) shows unstable responses, and therefore all subsequent investigations were carried out with membrane B. Time -+ Fig. 1. Responses of membranes A, B and C to different thall- ium(II1) concentrations in 1 M hydrochloric acid. The numbers above the lines indicate pT1I11 Table 2. Potentiometric selectivity coefficients, KE,:, for the coated-wire thallium(II1)-selective electrode Ion @?,; * Cu", Ni", CdII, Mn", Zn", Pb", FeIII, CrIII, InIII, BPI, SnIV, AsV, AgI, nitrate, perchlorate, sulphate . . . . . . . . . .. . . . . . <2 x lO-4a mrr . . <5 x 10-4a PdI', GaIII, PtIV . . . . . . . . . . . . . . <2 x 10-3b AU"I 1.oc Sb\' 15.8= * Concentrations of interferent: a 10-2 M; b 10-3 M; c 10-4 M . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Response Range and Effect of Hydrochloric Acid Concentra- tion The response of the electrode in the thallium(II1) concentra- tion range 10-6-10-2 M was studied at various hydrochloric acid concentrations ranging from 0.01 to 1 M. The electrode gave a near-Nernstian response in the range 10-5-10-2 M thallium(II1) at 0.1-1 M hydrochloric acid concentrations, corresponding to the response to the T1C14- anion. The slopes (S) and correlation coefficients ( r ) obtained were as follows: 1 M HC1, S = -57.4, r = 0.9999; 0.5 M HC1, S = -55.9, r = 0.9999; 0.1 M HC1, S = -55.8, r = 0.9997; 0.01 M HC1, S = -48.9, r = 0.9987.The highest slope and the best correlation coefficient were obtained for 1 M HCl. This strongly acidic medium is also more convenient for the direct analysis of thallium in real samples. Taking all these results into account, a 1 M hydrochloric acid medium was selected for further work. Response Time and Detection Limit Response times of the electrode were tssyO within 10 s for 10-2-10-4 M thallium(II1) and within 1 min for 10-'-10-6 M and tgsyO within 1.5 min for 10-2-10-4 M and within 3-4 min for lo-5-10-6 M thallium(II1). The detection limit of the electrode, considered as the thallium(II1) concentration at which the potential deviates by 18 mV from the extrapolation of the linear portion, is 2 x 10-6M.Reproducibility and Stability The average change in potential for consecutive measure- ments of the 10-4 M standard thallium(II1) solution (five determinations) is kO.1 mV. The reproducibility and stability of the electrode were evaluated by determining replicate calibration graphs (n = 10) over a period of 2 weeks. The electrode was stored dry and conditioned each time. Although the absolute potential of the electrode changed, the calibration slope remained constant over this period (mean k standard deviation, 57.28 k 0.18). Selectivity The interference of various ions was studied by the mixed solution method.15 The concentration of the interfering ion was generally fixed at 10-2 or 10-3 or 10-3 M while the concentration of thallium(II1) was varied between 10-2 and M solution of the interfering ion was used.The selectivity coefficients, Kr;;.), presented in Table 2 show very good selectivity with respect to many common ions. Gold(II1) and antimony(V) cause the largest interferences. M. In examples where the interference was large, aANALYST, DECEMBER 1986, VOL. 111 1361 Table 3. Determination of trace amounts of thallium in materials Thallium obtained*/mg g- Sample Potentiometry Atomic absorption Sphalerite . . . . . . 0.0743 0.0749 Zincconcentrate . . . . 0.334 0.339 * Average of three determinations. Potentiometric Titrations The use of the coated-wire thallium(II1) electrode in the potentiometric titration of thallium(II1) solutions containing 1 M hydrochloric acid with 1,2,4,6-tetraphenylpyridinium acet- ate solutions was investigated.The method relies on the decrease of the concentration of the tetrachlorothallate(II1) anion by precipitation with the 1,2,4,6-tetraphenyIpyridinium cation. In the titrations of 0.5 and 5.0 mg of thallium(II1) with 2 x 10-3 and 2 x M TPPA solutions, respectively, following the recommended procedure, typical potentiometric titration curves with steepness near the end-point (12 and 36 mV per 0.05 ml, respectively) were obtained. The coefficients of variation for the titration of different amounts of thallium in the recommended range (five titrations each) vary between 1.1 and 3.0% and the relative errors are between +0.7 and -2.5%. Applications The method has been applied satisfactorily to the determina- tion of thallium in sphalerites and zinc concentrates. The results of the direct potentiometric analysis shown in Table 3 are compared with those obtained by atomic absorption spectrometry.Good agreement was found between the two methods. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. References Catrall, R. W., and Freiser, H., Anal. Chem., 1971, 43, 1905. Catrall, R. W., and Chin-Poh Pui, Anal. Chem., 1975, 47, 93. Catrall, R. W., and Chin-Poh Pui, Anal. Chem., 1976,48,552. Catrall, R. W., and Chin-Poh Pui, Anal. Chim. Acta, 1976,83, 355. Catrall, R. W., and Chin-Poh Pui, Anal. Chim. Acta, 1976,87, 419. Alexander, P. W., and Joseph, J. P., Talanta, 1981, 28, 931. Ortuno, J. A., PCrez Ruiz, T., and Sanchez-Pedrefio, C., Anal. Chim. Acta, in the press. Burger, K., and Petho, G., Anal. Chim. Acta, 1979, 107, 113. PCrez Ruiz, T., Sanchez-Pedreiio, C., and Ortufio, J. A., Analyst, 1982, 107, 185. Koryta, J., and Stulik, K., “Ion-selective Electrodes,” Second Edition, Cambridge University Press, Cambridge, 1983, p. 30. Chadwick, T. C., Anal. Chem., 1976, 48, 1201. Kinnunen, J., and Wennerstrand, B., Chemist Analyst, 1957, 46, 92. Fiedler, U., and Rfiiitka, J., Anal. Chim. Acta, 1973, 67, 179. Catrall, R. W., Drew, D. M., and Hamilton, I. C., Anal. Chim. Acta, 1975, 76, 269. Bailey, P. L., “Analysis with Ion-Selective Electrodes,” Heyden, London, 1976, p. 48.

ISSN:0003-2654    

DOI:10.1039/AN9861101359

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST, DECEMBER 1986, VOL. 111 1363 Ion-selective Electrode for the Determination of Metoclopramide S. S. Badawy, A. F. Shoukry and Y. M. lssa Department of Chemistry, Faculty of Science, Cairo University, Giza, Egypt A metoclopramide (MCP) ion-selective PVC membrane electrode based on the ion-pair complex of MCP with sodium tetraphenylborate was prepared with dioctyl phthalate as a plasticiser. The electrode exhibits a linear response with an approximate Nernstian slope (50 mV decade-’ at 20 “C) within the concentration range 10-1.7-10-5.6 M MCP. The effects of ionic strength and pH of the test solution on the electrode performance were studied. The electrode exhibited very good selectivity for MCP with respect to a large number of inorganic and organic cations of biological importance.The standard additions method and potentiometric titration were used to determine the MCP in pure solutions and in pharmaceutical preparations with satisfactory results. Keywords: Metoclopramide ion-selective electrode; poly(viny1 chloride) membrane; potentiometric titration; drug analysis Metoclopramide [monohydrate of 4-amino-5-chloro-N- (2-diethylaminoethyl)-2-methoxybenzamide hydrochloride, MCP] is the active ingredient of many pharmaceutical preparations concerned with the modification of digestive behaviour. The interest in MCP is related to the elective character of its action in various digestive manifestations commonly seen in medical practice, e . g . , nausea, meteorism, epigastric discomfort, hiccups and sensations of heaviness associated with digestive disorders.However, extra-pyrami- dal symptoms (localised or generalised muscle spasms) may appear on overdosage of MCP. The same phenomenon may exceptionally occur with therapeutic doses in patients who are particularly sensitive or who are simultaneously treated with neuroleptics. Although many methods, other than potentiometry, have been reported for the determination of MCP,1-7 most of them are complicated and need sophisticated instrumentation such as HPLC or GLC. The low cost and ease of operation of potentiometric instrumentation make the potentiometric determination of MCP+ a highly desirable alternative. For this reason, and as a result of the medical importance of MCP, it was of interest to investigate the performance characteristics of an MCP-selective membrane electrode based on the MCP ion pair with tetraphenylboron (TPB) incorporated in a poly(viny1 chloride) (PVC) matrix and to use this electrode for the analysis of Primperan syrup and Plasil tablets.Experimental Metoclopramide was provided by AMSA (Milan, Italy), sodium tetraphenylborate(II1) and NaTPB by Aldrich, dioctyl phthalate (DOP) and the PVC for ISEs by Fluka and the pharmaceutical formulations were from Laboratoires Delag- range, Paris (Primperan syrup) and Lepetit, Milan (Plasil tablets). The precipitate of the MCP - TPB ion pair was prepared by mixing 50 ml of an aqueous solution containing 2.5 X 10-3 mol MCP with a solution containing an equimolar amount of NaTPB. The precipitate was filtered, washed throughly with distilled water and dried at room temperature.Electrode Preparation A mixture of 320 mg of PVC, 320 mg of DOP and 50 mg of the ion pair was dissolved in tetrahydrofuran (THF). The resulting mixture was poured into a Petri dish with a diameter of 9.5 cm and the THF was left to evaporate at room temperature. A transparent membrane of about 0.3 mm thickness was obtained, from which a disc of about 12 mm diameter was cut out and glued to the polished end of a PVC tube by means of a PVC - THF solution. The electrode was then filled with a solution of 10-1 M NaCl and 10-3 M MCP as the internal solution. Potential Measurements The electrochemical system was as follows: Ag,AgClI internal filling solution I membrane 1 test solution I I KC1-saturated salt bridge I I saturated calomel electrode.The potential was measured with a Chemtrix Type 62 digital pH/mV meter in a constantly stirred solution. The electrode was soaked in 10-3 M MCP solution for 2 h prior to daily measurements and stored dry and in a closed vessel in a refrigerator. Construction of the Calibration Graph Suitable increments of standard MCP solutions were added to 100 ml of 10-7 M MCP solution to cover the concentration range 10-7-10-1 M MCP. The MCP and reference electrodes were immersed in this solution and the e.m.f. values were recorded after each addition and plotted versus pMCP values. The electrode was repeatedly calibrated over a period of two months. Selectivity of the Electrode The selectivity coefficients @;&+ ,Jz+ were evaluated by the separate solution method8 in which the following equation was applied: wht;ie El is the electrode potential in 10-3 M MCP solution and E2 is the potential of the electrode in a 10-3 M solution of the interferent J z + .1364 ANALYST, DECEMBER 1986, VOL.111 Potentiometric Determination of MCP The standard additions method, in which small increments of a standard solution (10-1 M) of MCP were added to 100 ml samples of various concentrations, was used. The change in the mV reading was recorded after each addition and used to calculate the concentration of the MCP sample solution. For the analysis of MCP formulations, aliquots of 2-5 ml of syrup or 100-150 mg of tablet powder were quantitatively transferred into 150-ml beakers, each containing 100 ml of distilled water, and the standard additions technique was applied as described above.Potentiometric Titration of MCP An aliquot of the MCP solution containing 0.71-7.1 mg of MCP was pipetted into a 150-ml beaker. A 10 ml aliquot of 0.1 M NaCl was added and the solution diluted to 100 ml with distilled water. The resulting solution was titrated with 0.01 M standard NaTPB solution using the MCP membrane electrode as the sensor. For MCP-containing preparations, 1-10 ml aliquots of the syrup or 100-150 mg of the powdered tablets were transferred into 150-ml beakers, each containing 100 ml of water, and titrated as above. Results and Discussion Electrode Response The e.m.f. was measured in solutions containing lo-7-10-1 M MCP at 20 "C. The total ionic strength of each solution was +200 - +160 - +120 - > E +80 - G +40 - 0.0 -40 - - , 8 6 4 2 0 pMCP Fig.1. Calibration graph for the MCP membrane electrode > E t a t + + 0 , 2 1 4 1 6 1 I D 0 2 4 6 E 0 L 2 , 4 , 6 , I~ 0 2 4 6 G pMCP 0 2 4 6 Fig. 2. Effect of ionic strength on performance characteristics of the MCP electrode. u = A, 0; B, 0.05; C, 0.10; D, 0.20; E, 0.40; F, 0.60; and G, 1.00 0.01 (except at the highest concentration) and NaCl was the supporting electrolyte. The MCP+ ion caused a regular increase in the e.m.f. value with a slope of 50 k 1 mV per concentration decade of MCP within the concentration range 10-5.6-10-1.7 M. The response was rapid and reversible and the equilibrium response time was 5-10 s after the membrane electrode was placed in solution. The reproducibility of repeated measurements on the same solutions was +1 mV. The calibration graph of the investigated electrode is shown in Fig.1. The intercept of the linear part of this graph at pMCP = 0 is +269 mV. Table 1. Effect of ionic strength on the performance characteristics of the MCP electrode Ionic Intercept at strength* Slope pMCP = 0.0 0.00 49 +225 0.05 49 +218 0.10 48 +212 0.20 42 +199 0.40 37 +194 0.60 36 +188 1.00 35 +182 * NaCl added. Usable pMCP range 4.8-2:2 4.6-2.1 4.5-2.0 4.2-2.0 3.9-2.1 4.0-2.0 3.9-2.1 Response time/s <lo 910 G 10 G 10 G 15 S15 G20 +601 + 40 , , I I \I 0 2 4 6 8 1 0 PH Fig. 3. Effect of pH on the potential of the MCP membrane electrode. [MCP] = 5 x 10-3 M TPB ion-pair -r 0 2 4 6 NaTPB, 5 x 10-3 dml Fig. 4. Potentiometric titration of pure MCP and Primperan drug solutions using the MCP membrane electrode as the sensor.A, B and C, 2.5 ml of 2 X 10-3 M MCP solutions and D, E and F, 5.0 ml of 2.82 x 10-3 M drug solution Table 2. Selectivity coefficients of the MCP electrode Interferent Log KG;~,~z+ Interferent Na+ . . , . . . -3.12 Sucrose . . K+ . . . . . . -2.87 Glycine . . NH4+.. . . . . -2.78 Alanine . . Mg2+ . . . . . . -4.08 Phenylalanine Ca2+ . . . . -4.02 Me2NH2+ . . D(+)-GiUCOse . . -2.62 Et2NH2+ . . Lactose . . . . -2.71 Et,NH+ . . Maltose . . . . -2.82 Me,N+ . . Log KG,Lp,J~+ . . -2.88 . . -3.26 . . -3.29 . . -3.10 . . -3.04 . . -2.62 . . -3.19 . . -2.71ANALYST, DECEMBER 1986, VOL. 111 1365 Table 3. Potentiometric determination of MCP Standard additions method Potentiometric titration Amount Standard Amount Standard Solution takedmg Recovery, 70 deviation, YO taken/mg Recovery, YO deviation, YO PureMCP .. . . 3.54-7.08 98.5 0.81 0.71-7.1 100.8 1.05 Plasil-t . . . . 10.0-15.0 102.0 0.95 10.0-15.0 102.2 2.30 Primperan* . . 2.0-5.0 97.0 1.60 1.0-10.0 99.8 1.02 * Laboratoires Delagrange, Paris. t Lepetit, Milan. Effect of Ionic Strength The study of the effect of ionic strength (u) on the perfor- mance characteristics of the electrode aimed to determine the pMCP range in which the electrode could be used in different media. For this purpose, calibration graphs were plotted using MCP solutions of various ionic strengths obtained by adding KC1 as the supporting electrolyte. The results are given in Fig. 2 and Table 1. It is evident that an increase in ionic strength generally has a negative effect on the slope and the usable range of the calibration graph.The intercept of the linear parts of the graphs with the e.m.f. axis shows a gradual shift to less positive values as the ionic strength increases. It is also noted that up to an ionic strength of 0.2, the equilibrium response time is less than 10 s, but it gradually increases to 15 s for u = 0.4-0.6 and to 20 s for u = 1.0. Effect of pH The effect of the pH of the test solution (10-3 M MCP, 5 x 10-2 M NaC1) on the electrode potential was investigated by following the variation of potential with change in pH by the addition of very small volumes of HC1 and/or NaOH (0.1-1 M of each) (Fig. 3). From the graph obtained it is clear that pH has a negligible effeet within the range 2.5-7.3 and in this range the electrode can safely be used for MCP determination.Fig. 3 also shows that’at pH values lower than 2.5, the MCP electrode becomes progressively sensitive to the diprotonated MCP species and the e.m.f. readings decrease with decreasing pH. At pH values higher than 7.3 the MCP base precipitates and consequently the concentration of the protonated species decreases. As a result lower e.m.f. readings are recorded. Selectivity of the Electrode The influence of some inorganic cations, sugars, amino acids and organic amines on the MCP electrode was investigated, the selectivity coefficients being determined by the separate solution method (Table 2). None of the investigated species was found to interfere, as shown by the very small values of KE&+,Jz+.This reflects a very high selectivity of the investigated electrode towards MCP. The inorganic cations do not interfere owing to the differences in ionic size, and consequently their mobilities and permeabilities, as compared with those of MCPf. The high selectivity of sugars, amino acids and amines is mainly attributed to the differences in polarity and lipophilic nature of their molecules relative to those of MCP. Analytical Applications The electrode proved to be useful in the potentiometric determination of MCP in pure solutions and in pharmaceutical preparations by direct potentiometry using the standard additions and potentiometric titration methods described (Table 3). Representative titration curves for MCP both in a pure solution of ionic strength 0.01 (NaC1) and in Primperan drug are given in Fig.4. The values of standard deviation and recovery given in Table 3 prove that the electrode is very successful for the microdetermination of MCP either in pure solutions or in the pharmaceutical preparations Primperan (syrup) and Plasil (tablets). These values are comparable to those obtained by applying a spectrophotometric method9 to the determination of MCP in pharmaceutical preparations (recovery range 97.61-99.60% and relative standard deviation The use of the electrode in the analysis of Primperan suppositories was unsuccessful, possibly owing to the presence of greasy materials poisoning the membrane surface. 1.4 Yo). 1. 2. 3. 4. 5. 6 . 7. 8. 9. References Baeyens, W., and De Moerloose, P., Analyst, 1978, 103, 359. Park, M. K . , Lim, B., Yu, K., and Yong, K. H., Yakhak Hoeji, 1978, 22, 27. Tam, Y. K., Axelson, J. E., and Ongley, R., J. Pharm. Sci., 1979,68, 1254. Riley, C . M., J. Pharm. Biomed. Anal., 1984, 2, 81. Tam, Y. K., and Axelson, J. E., J. Pharm. Sci., 1978,67,1073. Shingbal, D. M., and Naik, S. D., Zndian Drugs, 1981,18,441. Groszkowski, S . , Krzemieniewska, G., and Ochocki, Z . , Farm. Pol. , 1984, 40, 419. CoSofrej, V. V., and Buck, R. P., Analyst, 1984, 109, 1321. Kamalapur Kar, 0. S . , and Chudasama, J. J., Zndian Drugs, 1983, 20, 298. Paper A61136 Received May 9th, 1986 Accepted July 30th, 1986

ISSN:0003-2654    

DOI:10.1039/AN9861101363

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST, DECEMBER 1986, VOL. 111 1367 Liquid Membrane Electrode for the Direct Determination of Ephedrine in Pharmaceutical Preparations* Saad S. M. Hassant and M. M. Saoudi Department of Chemistry, Faculty of Science, Ain Shams University, Cairo, Egypt The construction and performance characteristics of a liquid membrane electrode responsive to the ephedrine cation are described. The electrode is based on the use of the ephedrine-5-nitrobarbiturate ion-pair complex in nitrobenzene a s an ion-exchange site. The electrode shows a stable, near-Nernstian response for 10-2-10-5 M ephedrine over the pH range 4-7. The lower limit of detection is 4.5 x 10-6 M and the response time 20-90 s, and the selectivity coefficients for ephedrine relative to a number of interfering substances were investigated.Many organic and inorganic cations and pharmaceutical excipients and diluents commonly used in drug formulations do not interfere. The determination of 0.1-2000 pg ml-1 of ephedrine in aqueous solutions shows an average recovery of 99.2% and a mean relative standard deviation of 1.5%. The direct determination of ephedrine in some pharmaceutical preparations gives results that compare favourably with those obtained by the British Pharmaceutical Codex method. Keywords: Ephedrine electrode; liquid membrane; potentiometry; pharmaceutical analysis; 5-n itro ba rbitu ra te ion -pa ir complex Ephedrine is a sympathomimetic drug that stimulates both a- and P-adrenergic receptors. It is used in therapeutic doses at the level of 15-60 mg to produce peripheral vasoconstriction, to raise blood pressure, to prevent hypotension and to treat allergic states, catalepsy and myasthenia gravis.It is also utilised as an antidote for poisoning by central nervous system depressants. Official methods used for the determination of ephedrine in various pharmaceutical preparations are usually based on its extraction as a free base, followed by spectro- photometric determination at 241 nm.l Many organic com- pounds, drug excipients and various organic bases, however, absorb at the same wavelength and hence significantly interfere. The determination of ephedrine by spectrophotometric methods has been suggested based on oxidation with perio- date,2 hypohalite,3 chromate4 and hydrogen peroxide5 to give products that can be determined either directly in the ultraviolet region223 or in the visible region after condensation with semicarbazide or aminoantipyrine.2.4>5 Chromogenic reactions using l-fluoro-2,4-dinitrobenzene,6 picryl chloride ,7 ninhydrin,s bromothymol blue,9 tetrabromophenolphthalein ethyl esterlo and copper salts11 have also been described.These methods involve a time-consuming extraction step, require strictly controlled reaction conditions and suffer from severe interference from amines, amides and amino acids. Ephedrine in pharmaceutical preparations has been deter- mined by polarographic and potentiometric techniques. The polarographic methods are based on a prior conversion of ephedrine into polarographically active derivatives through brominationl2 or nitrosation13 reactions, followed by measur- ing the redox wave.Most of the potentiometric methods are based on the extraction of the free base, followed by titration with standard acids in non-aqueous media.14 Membrane electrodes incorporating ephedrine tetraphenylborate dis- persed in organic solvents or poly(viny1 chloride) have recently been used for direct potentiometric monitoring and potentiometric titration of ephedrine. 15-19 The presence of tertiary amines, various classes of alkaloids, arenediazonium salts, some inorganic cations and many of the drug excipients seriously interfere. * Presented at the 30th IUPAC Congress, Manchester, UK, 9-13 t To whom correspondence should be addressed. Current address: September 1985. Department of Chemistry, Quatar University, Doha, Quatar.In earlier papers, we have described liquid membrane electrode systems for the determination of strychnine,20 atropine ,21 caffeine ,22 lidocaine23 and some biogenic amines.24 These electrodes incorporate the picrolonate, reineckate, picrylsulphonate and flavianate ion-association complexes of these compounds as electroactive materials. In this study, a liquid membrane electrode with significantly improved characteristics for ephedrine was developed. It is based on the use of the ephedrine-5-nitrobarbiturate ion-pair complex in nitrobenzene as an electroactive material. The electrode has been successfully used for the determination of ephedrine in simple and complex matrices. Experimental Reagents All solutions were prepared with de-mineralised, doubly distilled water and all chemicals were of analytical-reagent grade unless stated otherwise.Standard 10-2-10-6 M aqueous ephedrine solutions were freshly prepared by accurate dilutions of a 10-1 M ephedrine hydrochloride stock solution (20.17 g 1-1). An aqueous 10-1 M solution of 5-nitrobarbituric acid was prepared by dissolving 0.207 g of 5-nitrobarbituric acid trihydrate in 100 ml of de-mineralised, doubly distilled water and filtering. Pharmaceutical preparations containing ephedrine were obtained from local drug stores. Equipment The potentiometric measurements were made with an Orion microprocessor Ionalyzer (Model 901) using the ephedrine-5- nitrobarbiturate liquid membrane electrode in conjunction with an Ag - AgCl double-junction reference electrode (Orion Model 90-02) containing a 10% mlV KN03 solution in the outer compartment.The cell potentials were measured for stirred solutions at 25 k 1 "C at pH 4-8 using the following electrochemical cell: Ag - AgClI 10-2 M ephedrine hydro- chloride - 10-2 M KC11 10-2 M ephedrine-5-nitrobarbiturate in nitrobenzenel !porous membrane1 Isample test solution1 Ag - AgCl double-junction reference electrode. An Orion glass - calomel combination electrode (Model 91-02) was used for pH adjustment.1368 ANALYST, DECEMBER 1986, VOL. 111 Procedure Membrane preparation The ephedrine-5-nitrobarbiturate ion-pair complex was pre- pared by mixing 15 ml of aqueous 10-2 M ephedrine hydro- chloride with 20 ml of aqueous 10-2 M 5-nitrobarbituric acid. The mixture was cooled in an ice - water mixture, the precipitate was filtered off by suction, washed with de- mineralised, doubly distilled water, dried at ca.70 "C for 15 min and then ground. Elemental analysis data of the product agreed with the composition CI4Hl8N4O6. The infrared spectrum of the product displays almost all the absorption bands that appear in the spectra of both reactants and also a stretching vibration band at 2460 cm-1 assigned to the imino group. The liquid ion-exchange membrane was prepared by making a 10-2 M solution of ephedrine-5-nitro- barbiturate in nitrobenzene. u OH Electrode preparation The body of an Orion Series 92 electrode, equipped with a microporous membrane (Orion 92-05-04), was used. The electrode was assembled and the internal reference and liquid ion-exchange solutions were injected into the appropriate ports.The internal reference solution was a mixture of an equal volume of KC1 and ephedrine hydrochloride solutions, 2 x 10-2 M each. The electrode was conditioned by soaking in a 10-3 M aqueous ephedrine hydrochloride solution for 2 d before use. When not in use, the electrode was immersed in the same solution. The selectivity coefficients were measured using the mixed solution method.25 The performance characteristics of the electrode were evaluated as previously described.2@24 Determination of ephedrine in pharmaceutical preparations The contents of five ephedrine vials were mixed and a volume equivalent to one vial was transferred into a 50-ml beaker, followed by dilution with 30 ml of de-mineralised, doubly distilled water.The solution was acidified with 1 ml of 0.1 M HCl, heated at ca. 70 "C for 5 rnin and cooled to room temperature. The pH of the solution was adjusted to 4-7 with 0.05 M NaOH. The solution was then transferred into a 50-ml calibrated flask, diluted to the mark with de-mineralised, doubly distilled water, shaken and transferred into a 100-ml beaker. The ephedrine-5-nitrobarbiturate liquid membrane electrode in conjunction with the reference electrode was immersed in the test solution. The potential reading was recorded when stable and compared with a calibration graph prepared from pure ephedrine hydrochloride solutions under identical conditions. For the determination of ephedrine in pharmaceutical tablets, ten tablets were pulverised and a weighed portion equivalent to one tablet was transferred into a 50-ml beaker and dissolved in 30 ml of demineralised, doubly distilled water.The procedure used for the determination of ephedrine in vials was then followed. OH CH3 CH3 OH Fig. 1. Ephedrine-5-nitrobarbiturate ion-pair complex I I I I L 6 5 4 3 2 -Log ([ephedrineli~) Fig. 2. ephedrine-5-nitrobarbiturate liquid membrane electrode Calibration graph for ephedrine in the pH range 4-7 using the 70 30 > E Gi -10 -50 Results and Discussion Membrane Material and Characteristics Some organic bases, including ephedrine, can be identified by examining their photomicrographs or by measuring the melting-points of their 5-nitrobarbiturate derivatives.26 In this study, the ephedrine-5-nitrobarbiturate ion-pair complex was prepared (Fig.l ) , characterised and tested as a novel ion-exchange site in a liquid membrane electrode responsive to ephedrine. Solutions of the ephedrine-5-nitrobarbiturate ion-pair com- plex (10-2-10-3 M) were prepared in lipophilic solvents (such as nitrobenzene, octan-1-01 and decan-1-01) and were tested as 3 6 9 Effect of pH on the potential of the ephedrine-5-nitrobarbitu- PH Fig. 3. rate liquid membrane electrode liquid ion-exchange membranes. Potentiometric measure- ments at 25 k 1 "C with the ephedrine-5nitrobarbiturate liquid membrane electrode incorporating nitrobenzene as a membrane solvent give a near-Nernstian response for the ephedrine cation over the concentration range 10-2-10-5 M (Fig. 2). The initial slope of the calibration graph is 55 mV per concentration decade.The least-squares equation obtained from the calibration data is E(mV) = (55 k 0.5) log C + (160 k 0.7).ANALYST, DECEMBER 1986, VOL. 111 1369 The standard deviation is 1.1 mV and the detection limit calculated according to the IUPAC recommendation27 is 4.5 X 10-6 M. The slope and limit of detection offered by this electrode system are better than those obtained using either octan-1-01 or decan-1-01 as membrane solvents; this is probably due to the ease of dissolution and dissociation of the complex in nitrobenzene. Response Time and Stability of the Membrane The average time required for the ephedrine-5-nitrobarbitu- rate liquid membrane electrode to reach a potential within k 1 mV of the final equilibrium value after successive immer- sion in a series of ephedrine solutions, each having a 10-fold difference in concentration, was measured.Stable responses were achieved almost instantaneously for concentrations 210-3 M, and within 60-90 s for concentrations < l o - 4 M. The electrode exhibits a day-to-day reproducibility of about 2 2 mV for the same solutions for 6 weeks after preparation, provided that the electrode is not used in the presence of high concentrations of strong interfering compounds or in highly alkaline or acidic solutions. The influence of pH on the electrode response to different ephedrine concentrations is shown in Fig. 3. The electrode potential is independent of pH in the range 4-7. Over this range the potential does not vary by more than -2.0 mV for any ephedrine concentration in the range 10-2-10-4 M.Prolonged immersion of the electrode in alkaline solutions (pH > 9) causes a deterioration in its response, probably because of the back-extraction of the 5-nitrobarbiturate anion from the organic membrane phase to the aqueous test solution and/or gradual precipitation of the ephedrine base in the test solution. The decrease of the potential below pH 4 may be due to interference from H+. Table 1. Selectivity coefficients for ephedrine-5-nitrobarbiturate liquid membrane electrode Selectivity coefficient, Interfering species (B) e;;, B Me thylurea . . . . . . . . Ace tamide . . . . . . . . Aminobenzoicacid . . . . Piperidine . . . . . . . . Diethylamine . . . . . . Triethanolamine . . . . . . Tetramethylammonium chloride G1 ycine .. . . . . . . . . Alanine . . . . . . . . Nicotine . . . . . . . . Nicotinicacid . . . . . . Strychnine . . . . . . . . Caffeine . . . . . . . . Quinine . . . . . . . . Li+ . . . . . . . . . . Ba2+ . . . . . . . . . . NH4+ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . * . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1 x 10-3 3.6 x 10-3 2.4 x 10-3 2.1 x 10-2 1.9 X 10-2 2.1 x 10-2 6.7 x 10-2 6.3 X 10-3 8.1 x 10-1 6.6 x 10-2 2.8 1.7 1.4 3.1 x 10-3 6.2 x 10-3 8.2 x 10-3 6.1 x 10-3 7.3 x 10-2 5.4 x 10-2 2.5 x 10-3 Table 2. Direct potentiometric determination of ephedrine using ephedrine-5-nitrobarbiturate liquid membrane electrode and the calibration graph method Ephedrine hydrochloride Standard deviation, Addedtyg ml-1 Recovery,* % Y O 2000.0 98.5 0.4 1500.0 98.4 0.4 1000.0 99.1 1.5 500.0 98.1 1.6 100.0 98.0 1.7 50.0 99.0 1.4 10.0 98.0 1.4 5.0 99.5 1.8 1 .o 101.5 2.0 0.10 101.8 2.8 * Average of three measurements.Selectivity Coefficients of the Membrane The potentiometric selectivity coefficients (K$&B) of the ephedrine-5-barbiturate liquid membrane electrode were experimentally determined by the mixed solution method recommended by IUPAC and were calculated from the modified Nernst equation .25327 The ephedrine concentrations were varied from 10-3 to 10-5 M while the concentration of the competing cation was kept constant at 10-3 M. The data in Table 1 show the selectivity of the ephedrine electrode over a series of potentially interfering organic and inorganic cations.Soluble drug excipients and diluents such as maltose, glucose, lactose, starch and gelatin binder that are present in some tablets do not interfere with the response of the electrode. Further, the electrode exhibits negligible interference from many nitrogenous compounds such as amines, amides and amino acids. The electrode is, however, not really selective for ephedrine over some other alkaloids such as strychnine, quinine and caffeine. Determination of Ephedrine Ephedrine solutions in the concentration range 0.1- 2000 pg ml-1 were prepared from pharmaceutical grade reagents and determined by the direct potentiometric method using the ephedrine-5-nitrobarbiturate liquid membrane elec- trode. The potentials recorded in these solutions were compared with a calibration graph. The results obtained (Table 2) for ten samples, each determined in triplicate, showed an average recovery of 99.2% and a mean standard deviation of 1.5%. Ephedrine was also determined in some pharmaceutical preparations.Injections and tablets were homogenised, treated with 0.1 M HC1, heated to effect complete solubilisa- tion and diluted with de-mineralised, doubly distilled water. The potential was then measured after the adjustment of pH to 4-7 and compared with a calibration graph. The results obtained (Table 3) show an average recovery of 99.1% of the nominal values and a mean standard deviation of 1.7%. No interference was caused by active or inactive ingredients and diluents commonly used in drug formulations.The British Table 3. Determination of ephedrine in some pharamaceutical preparations using ephedrine-5-nitrobarbiturate liquid membrane electrode Electrode method BPC method1 Preparation Labelled ephedrine Recovery,*% S.D.,% Recovery,*% S.D.,% Ephedrine sulphate (injection) . . . . 25 mg ml- 100.0 1.7 98.8 2.0 Ephedrine sulphate (injection) . . . . 50 mg ml- 99.4 1.5 98.7 1.8 Ephedrine hydrochloride (tablet) . . . . 25 mg per tablet 98.3 1.6 102.2 2.3 Ephedrine hydrochloride (tablet) . . . . 50 mg per tablet 98.6 1.8 101.4 2.0 * Average of five measurements.1370 ANALYST, DECEMBER 1986, VOL. 111 Pharmaceutical Codex method,l involving a prior extraction of the free base with cyclohexane followed by spectropho- tometric determination at 241 nm, was also used for compari- son.The results obtained by both methods are in good agreement (Table 3). The electrode method, however, offers several advantages in term of simplicity, selectivity and precision. Further, it eliminates the time-consuming extrac- tion step and is directly applicable to coloured and turbid solutions. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. References “British Pharmaceutical Codex 1973,” Pharmaceutical Press, London, 1973, p. 671. Wallace, J. E., J. Pharm. Sci., 1969, 58, 1489. Chafetz, L., Gosser, L. A . , Schriftman, H., and Daly, R. E . , Anal. Chim. Acta, 1970, 52, 374. Wallace, J. E., Anal. Chem., 1967, 39, 531. Jaakko, H., and Arja, K., Farm. Aikak, 1976, 85, 125. Tammilehto, S . , Farm. Aikak, 1975, 84, 53. SiirSunovi, M., and Chi, N. T. K., Cesk.Farm., 1966,15,474. Zobin, A., and Gracza, M., Acta Pharm. Hung., 1975,45,101. Kafedzhieva, P., and Vuleva, E., Farmatsiya (Sofia), 1977,27, 5. Sakai, T., Hara, I., and Tsubouchi, M., Chem. Pharm. Bull., 1976, 24, 1254. Morvay, J., and Sti’er, G., Acta Pharm. Hung., 1965,35, 199. DuSinsky, G . , and Aavabik, T., Cesk. Farm., 1959, 8, 205. DeMarco, A., and Mecarelli, E . , Farmaco, Ed. Prat., 1967,22, 795. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. Sell, E., and Rajzer, D., Chem. Anal. (Wursaw), 1976,21,933. Selinger, K., and Staroscik, R., Pharmazie, 1978, 33, 208. Fukamachi, K., and Nakagawa, R., Morimoto, M., and Ishibashi, N., Bunseki Kagaku, 1975, 24, 428. Goina, T., Hobai, S., and Rosenberg, L., Farmacia (Bucharest), 1978, 26, 141. Zeng, J., Yaoxue Xuebao, 1982, 17, 841. Selinger, K., and Staroscik, R., Chem. Anal. (Warsaw), 1982, 27, 223. Hassan, S. S. M., and Elsayes, M. B., Anal. Chem., 1979, 51, 1651. Hassan, S. S. M., and Tadros, F. S., Anal. Chem., 1984, 56, 542. Hassan, S. S. M., Ahmed, M. A., and Saoudi, M. M., Anal. Chem., 1985,57, 1126. Hassan, S. S. M., and Ahmed, M. A., J. Assoc. Off. Anal. Chem., 1986, 69, 618. Hassan, S. S. M., and Rechnitz, G. A . , Anal. Chem., 1986,58, 1052. Ma, T. S . , and Hassan, S. S. M., “Organic Analysis Using Ion Selective Electrodes,” Volumes 1 and 2, Academic Press, London, 1982. Chatten, L. G., and Barry, P. J., Can. J. Pharm. Sci., 1968,3, 40. IUPAC Analytical Chemistry Division, Commission on Ana- lytical Nomenclature, “Recommendations for Nomenclature of Ion Selective Electrodes,” Pure Appl. Chem., 1976, 48, 127. Paper A61126 Received April 25th, 1986 Accepted July 3rd, 1986

ISSN:0003-2654    

DOI:10.1039/AN9861101367

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST DECEMBER 1986 VOL. 111 1371 Decomposition and Stability Studies of Methylmercury in Water Using Cold Vapour Atomic Absorption Spectrometry Riaz Ahmed" and Markus Stoepplert Institute of Applied Physical Chemistry Chemistry Department Nuclear Research Centre Julich (KFA), P.O. Box 1913 0-5170 Julich FRG The long-term stability of Hg*+ and methylmercury chloride (MeHgCI) and the decomposition of MeHgCl under various conditions were investigated. In the absence of light MeHgCl does not decompose to Hg2+, even in the presence of 25% acids during a period up to 3 d. MeHgCl is not easily decomposed by heating in the presence of 50% acid concentrations; however with UV light the decomposition of MeHgCl to Hg2+ takes place immediately depending on the intensity of UV light.In the presence of SH groups (L-cysteine) partial stability and complexation of mercury was observed. For the long-term storage of Hg and MeHgCl at low concentrations in addition to certain reagents the material of the container is important. Keywords Mercury; methylmercury; decomposition and stability studies; water; cold vapour atomic absorption spectrometry Owing to their toxicological significance much work has been carried out on the determination of mercury1 and methylmer-cury.2 Methylmercury (MeHgCl) has most often been deter-mined in fish as this is the predominant form in these organisms.3 Studies have been carried out on the decomposi-tion of MeHgCl in fish samples using high concentrations of acids,4 and there is now increasing interest in the determina-tion of total and inorganic mercury and MeHgCl in water.5 For the decomposition and determination of MeHgCl in aqueous samples acid concentrations as high as those employed for fish samples cannot be used owing to the extremely low concentrations of mercury in water.Therefore, decomposition studies of MeHgCl in water are required using lower more practical acid concentrations. It has been reported that mercury can be taken up into the cysteine disulphide bridges of proteins. 6 Sulphydryl com-pounds (L-cysteine) also act as effective sensitisers for the photochemical methylation of inorganic mercury,7 and there is a possibility of mercury entering the biosphere as an Hg -cysteine complex.8 As sulphydryl compounds can be present in several different environmental systems particularly sea water and rain water it seemed appropriate to investigate the behaviour of MeHgCl with respect to its decomposition in the presence of cysteine a typical sulphydryl compound.Numerous papers have been published concerning the stability of ionic mercury in solutions9 but very little is known about the stability of MeHgCl in aqueous systems.10There are few references available that discuss the stability of MeHgCl in aqueous systems close to natural levels of mercury and MeHgCl. This paper describes the behaviour of MeHgCl in water when exposed to light and heat in the presence of different reagents acids sulphydryl compounds (L-cysteine) and also long-term stability studies of MeHgCl in the presence of different reagents and different container materials.Experimental Chemicals All the acids used (HCl HN03 HC104 and H2S04) were of Suprapur grade from E. Merck FRG. All other chemicals (NaC1 L-cysteine SnClz - H2S04 methylmercury HgO, H202 Na2SOq NaOH sodium acetate and toluene) were of * Present address NCD-PINSTECH P.O. Nilore Islamabad, 1- To whom correspondence should be addressed. Pakistan. analytical-reagent grade from E. Merck. Labelled methyl-mercury (1 mCi = 3.7 X 107 Bq) was obtained from Amersham UK. Apparatus The UV lamp for decomposition studies was a 150 W mercury vapour lamp and the samples were irradiated from a distance. An atomic absorption spectrometer (Bodenseewerk Perkin-Elmer FRG Model 400) with a mercury vapour lamp (0.2 A/15 V) as a light source was used at a wavelength of 253.7 nm and a 2.0 nm slit width.The mercury reduction vessel (1 1) was of Pyrex glass. Additional equipment included a recorder (SE 120 Goertz Metrawatt FRG). The automatic heating and gas flow control system used was constructed in the workshop of Dr. Beerwald Bochum FRG. A gas - liquid chromatograph Model 5710 equipped with an electron-capture detector (Hewlett-Packard USA) was used for the MeHgCl determination. 11 A multichannel Series 8 Analyzer from Canberra USA with Canberra Spectron F version V2 D1 software and a PDP 11/03 computer with a 64K memory was used for the activity measurements. Procedure Approximately 50 ml of a 10% SnC12 + 20% H2S04 solution was taken in the reduction vessel. A known volume of sample, usually 1-50 ml was taken.Nitrogen was passed through this solution at a rate of 2.0-2.5 1 min-1 and Hg was pre-concentrated on gold wool. After this step the nitrogen flow-rate was decreased to 50-100 ml min-1 and the gold wool heated to 700-800 "C to volatilise the Hg which was subsequently determined in the cuvette of the atomic absorp-tion spectrometer. After the determination of the sample a known amount of the standard solution was added to the same solution in the reduction vessel and determined at the same matrix conditions as for the sample. The apparatus and procedure have been described previously12 and will be given together with all recent modifications in detail elsewhere.13 Results and Discussion Precision and Linearity The cold vapour AAS (CVAAS) procedure has a detection limit of approximately 0.1 ng 1-1 for 100-ml water samples.The relative standard deviation for the analysis of mercury at ca. 1 ng levels (absolute) is <5"h. The procedure has a wide linearity range from 0.0 to 10 ng of mercury. The mercury concentrations used for the experiments were 0.005-1 . 1372 ANALYST DECEMBER 1986 VOL. 111 pg 1-1. The reduction mixture used for ionic mercury was 10% SnCl2 + 20% H2S04 and it was found that this reduction mixture did not decompose MeHgCl during the reduction aeration period. Decomposition of MeHgCl in the Presence of NaCl and HCI Sodium chloride and HC1 are often added to environmental samples for the extraction of MeHgCl by organic solvents r I 0.6 -c 0 1 2 3 Timeidays Fig.1. Decomposition of MeHgCl (1.03 pg 1-l) in Pyrex flasks covered with A1 foil in the presence of A 15% NaCl + 0.2 M HCl; B 15% NaCl + 2.2 M HCI; C 15% NaCl + 5.0 M HCI; D 5% NaCl + 0.2 M HCI; and E 5% NaCl + 2.2 M HCl 1.0 -0.8 -c I - . 0 1 2 3 4 Timelh Fig. 2. Decomposition of MeHgCl(l.03 pg 1-l) in quartz glass flasks in the resence of A 15% NaCl + 0.2 M HCI; B 15% NaCl + 2.2 M HCI; 8 15% NaCl + 5.0 M HCI; D 15% NaCl; and E 5.0 M HCl (toluene benzene) and the subsequent determination of MeHgCl by gas chromatography. Decomposition studies of MeHgCl in the presence of these reagents in Pyrex flasks covered with A1 foil were performed (Fig. 1). High concentra-tions of HCl or of NaCl alone did not decompose MeHgCl as quickly but when 15% NaCl + 2.2 M HCl were added up to 40% of the MeHgCl was decomposed within one day.This combination has been extensively used for the extraction -separation of MeHgC1.14 It is apparent from Fig. 1 that in the presence of these reagents losses of MeHgCl may take place if B 6 0 C e D ~ 0.010 0 I 0.01 5 -0 0.5 1 .o 1.5 2.0 AcidslM Fig 3. Decomposition studies of MeHgCl (0.0206 pg 1-l: presence of H202 (1.0%) and different concentrations of ac HCI; B HN03; C H2S04; and D HClO4 0.020 0.01 5 r I 0 0 I -2 0.010 0.005 in the Is A, 0 1 2 3 4 H202 Yo Fig. 4. Decomposition studies of MeHgCl (0.0206 pg I-l) in the presence of acids and different concentrations of H202. A 1.0 M HCI; B 1.0 M HNO,; C 1.0 M H2S04 and D 1.0 M HC104 ~ ~ Table 1.Decomposition studies of MeHgCl in the presence of different reagents with and without UV irradiation. 1.03 pg I-' of MeHgCl was added to each sample Concentration of Hg found with UV irradiation Sample (20 min)/ number Reagents added After0.0d After 1.0d After2.0d After3.0d After4.0d ygl-1 Concentration of Hg found without UV irradiatiodyg 1-l 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 10% HC1+ 2.5% HN03 20.0% HCI + 5.0% HN03 5.0% HCl + 20.0% HN03 2.0% HC104 + 10.0 HN03 1 .O% HN03 5.0% HN03 50.0% HCl 4.0% HC104 + 20.0% HNO, 15.0% NaCl 15.0% NaCl + 1.0% HCl 1 .O% H2S04 5.0% H2S04 1.0% HC104 10.0% HC104 H20 0.02 0.03 0.03 0.02 0.02 0.05 0.05 0.003 0.006 0.02 0.04 0.04 0.06 0.06 0.04 0.02 0.03 0.03 0.01 0.01 0.05 0.05 -0.07 0.11 0.08 0.06 0.08 0.09 0.14 0.09 0.08 0.09 0.05 -0.05 -0.05 -0.02 -0.02 -- 0.04 - 0.04 - -1.04 1.02 0.74 0.61 0.58 1 .oo 0.75 1.03 1.04 1.05 0.98 1.01 1.02 1.03 1.0 ANALYST DECEMBER 1986 VOL.111 1373 more time is taken for the extraction. In the presence of 15% NaCl + 5 M HC1 the rate of decomposition of MeHgCl is less than that seen with 15% NaCl + 2.2 M HCl; the reason for this may be the low solubility of NaCl in the presence of 5 M HCl. In order to determine the influence of light on the decomposition of MeHgCl studies were carried out in the presence of NaCl and HC1 in quartz glass flasks kept in the laboratory. There was no direct sunlight in the laboratory, only diffused daylight.It can be seen from Fig. 2 that decomposition of MeHgCl is very rapid and that in the presence of 15% NaCl + 2.2 M HCl MeHgCl is decomposed completely in less than 4 h. Hence it is clear that if extractions of MeHgCl are carried out in quartz vessels without covers, the concentrations of MeHgCl found will be unreliable. Effect of Ultraviolet Light on the Decomposition of MeHgCl In order to study the effect of UV light on the decompo-sition of MeHgCl various solutions were prepared in quartz glass flasks and exposed to UV light (Table 1). In the presence of NaCl HC1 HC104 H2S04 1 YO HN03 and H20 MeHgCl decomposes completely after less than 20 min of UV irradiation but in the presence of higher concentrations of HN03 the decomposition of MeHgCl is complete.It is also apparent from Table 1 that MeHgCl is stable in the presence of up to 25% acids without UV light. Detailed studies have shown15 that the decomposition of MeHgCl is difficult in the presence of more than 20% HN03 even with long UV irradiation times. The reason could be the negligible optical opacity in the UV region of strong HN03 solutions. MeHgCl was also not completely decomposed by UV light in the presence of NaOH and decomposition was more difficult at lower concentrations of NaOH (0.1 M) than at 0.75 M. Hence NaOH has the opposite effect to that of HN03 on the decomposition of MeHgCl with UV irradiation. Decomposition of MeHgCl in the Presence of Acids and H202 In the presence of 1% H202 and various concentrations of acids (0.25-2.0 M HC1 HN03 H2SO4 and HC104) the percentage of MeHgCl decomposed remains the same (Fig.3). Increasing acid concentrations do not increase the decom-position of MeHgCl and the effect on the decomposition of MeHgCl decreases in the order HN03 > H2S04 3 HC104 > HCl. If the acid concentration is kept constant and the H202 concentration increased the decomposition of MeHgCl increases with increasing H202 concentration (Fig. 4). The rates of increase in MeHgCl decomposition in the presence of HC1 and H2S04 are comparable and the rates of increase in MeHgCl decomposition in the presence of HN03 and HClO4 are also similar (Fig. 4). In the presence of acids and H202 the decomposition of MeHgCl takes place immediately on the addition of H202, and no further decomposition takes place when the solutions are allowed to stand for 3 d.It is possible that H202 reacts immediately in the presence of acids and is decomposed. 0 1 2 3 4 Time/h Fig. 5. Decomposition studies of MeHgCl (1.134 pg I-') in the presence of acids and heating at 200°C under pressure. A 50% HNO,; B 40% HN03 + 10% HC104 Effect of Heating on the Decomposition of MeHgCl MeHgCl decomposes immediately in the presence of UV light but it is very stable during heating. MeHgCl did not decompose when heated for 2.5 h at 80°C in the presence of 1-5% HCl or HN03 or 5% NaCl or in the absence of any addition. MeHgCl only decomposes in the presence of 100% HN03 or 90% HN03 + 10% HC104 when heated at 200°C under slight pressure for 1 h.16 However for the decompo-sition of MeHgCl in water samples it is impossible to add 100% acid because of the low concentrations of mercury and MeHgCl in water samples.Only up to ca. 50% acids can be added and this creates large blank values and contamination problems. Studies were carried out to determine the decom-position of MeHgCl in the presence of 50% HN03 and 40% HN03 + 10% HC104 and with heating at 200 "C under slight overpressure for longer times. It was found that even on heating for 3 h in the presence of 50% HN03 only 25% of the MeHgCl was decomposed and on heating for 4 h this decreased to 20% (Fig. 5 ) . When heated for 3 h in the presence of 40% HN03 + 10% HC104 only 34% of the MeHgCl was decomposed and on heating for 4 h this decreased to 29%.This decrease in Hg when heated for more than 3 h could be due to losses from prolonged heating. In conclusion it is difficult to decompose more than one third of the total MeHgCl using 50% acids and heating under pressure at 200°C. Disappearance of Detectable Mercury in the Presence of Cysteine Humic substances and sulphydryl compounds are present in sediments sea water and other systems of the environment. Cysteine was selected for a study of the decomposition and analysis of MeHgCl as mercury in aqueous samples in the presence of this compound. Detailed studies showed15 that if water samples were exposed to light in the presence of MeHgCl and cysteine an appreciable amount of the mercury could not be detected by CVAAS.When the samples containing MeHgCl and cysteine were exposed to UV irradiation in the presence of acids NaCl or even without any addition the major fraction of the decomposed MeHgCl could be initially determined but when the time of UV irradiation was increased to more than 1 h or even if the samples were allowed to stand for this length of time the major part of the mercury that was initially detectable changed into an undetectable form. In order to investigate the influence of chlorides and cysteine on the determination of MeHgCl after its decomposi-tion using CVAAS various experiments were carried out. Chlorides were selected as sea water and other water samples contain large amounts of chlorides. The detectable concentra-tion of mercury in the presence of 0.05% cysteine and various concentrations of HC1 after UV irradiation is plotted in Fig.6. The ratio of cysteine to HC1 is 0.0025 at the minimum 0.6 I I I I J 0 1 2 3 4 5 H C h Fig. 6 . Decomposition of MeHgCl (1.03 pg I-') in the presence of different concentrations of HCI 0.05% cysteine and after UV irradiation for 90 mi 1374 ANALYST DECEMBER 1986 VOL. 111 Table 2. Percentage recovery of MeHgCl after storage in Pyrex polyethylene and PTFE containers Pyrex Polyethylene PTFE Sample number Reagent added 11 d 35 d 5d 26 d 6 d 26 d 1 1.0% HCl 91.5 81.0 53.8 1.5 97.8 83.6 2 1 .0% HN03 96.8 87.5 94.1 72.2 88.4 68.1 3 5.0% NaCl 95.6 89.5 100.0 86.1 92.0 82.6 4 H20 18.8 3.13 26.2 1.4 21.8 4.4 1.2 , I / Cysteine O h Fig. 7. Decomposition of MeHgCl (1.07 pg 1-I) in the presence of 4.0% NaCl and different concentrations of cysteine and after UV irradiation tor 3.5 h detectable concentration of mercury.For different concentra-tions of cysteine at 0.2 M HC1 the minimum detectable concentration of mercury was at 0.005% cysteine; here also the ratio of cysteine to HC1 is 0.0025.15 In the presence of 4.0% NaCl and increasing concentrations of cysteine the minimum detectable mercury concentration is at 0.01% cysteine and again the ratio of cysteine to NaCl at the minimum is 0.0025 although here the minimum is present from ratios of 0.0025 to 0.0125 (Fig. 7). It is important to point out that for the determination of mecury in sea water the water samples are usually slightly acidified and then UV irradiated and measured for total mercury.Sea waters contain ca. 4.0% NaCl. If these sea waters contain small amounts of sulphydryl compounds as is usually so then the determination of total mercury will be unreliable. However as has been seen 50% acids cannot decompose MeHgCl in water and with UV irradiation in the presence of sulphydryl compounds much of the mercury goes into an undetectable form. Also many sea waters contain particulate matter which acts as a scavenger for mercury and MeHgCl. Thus the determination of mercury and MeHgCl in sea water samples is not easy although some approaches have been reasonably successful.~7~~~ In order to find out whether the undetectable mercury was converted back to MeHgCl, experiments were carried out using MeHgCl labelled with 203Hg measuring y-radiation with the multichannel Analyzer and also by gas - liquid chromatography with an electron-capture detector.15 It was found that this undetectable mercury remained in solution and was not MeHgCl although the possibility of the formation of complexes or sulphides should not be ruled out.Long-term Stability Studies of Mercury and MeHgCl The use of 0.1 M HC1 0.1 M NaOH 0.1 M HN03 and 5.0% NaCl for the preservation and storage of ionic mercury was compared for a period of 59 d as these reagents are either present in environmental water samples or are usually added before analysis. Only HN03 and NaCl can be used for short-term preservation purposes. High concentrations of HN03 cannot be used as they prevent the decomposition of MeHgCl with UV light.Stability studies of MeHgCl in glass polyethylene and PTFE bottles are shown in Table 2. In all three types of containers MeHgCl decomposes very rapidly in the absence of any additions of other reagents although the loss of ionic mercury is faster than the decomposition of MeHgCl in polyethylene bottles whereas in glass and PTFE containers the loss of ionic mercury is less than the rate of decomposition of MeHgCl. Thus without any additions MeHgCl decomposes very quickly at all concentrations (the mercury concentrations examined were 0.412 pg 1-1 in glass 0.0412 yg 1-1 in polyethylene and 0.0206 pg 1-1 in PTFE bottles). Various concentrations were taken to see if the stability increases with concentration but the concentration of MeHgCl had no effect on its stability.The stabilisation effects of 1% HCl 1% HN03 and 5% NaCl on the storage of MeHgCl in different types of containers are compared in Table 2. As can be seen the best preservative for MeHgCl is 5% NaC1 followed by 1% HN03 and HC1 and the best container material is glass then PTFE and polyethylene. During storage of MeHgCl solutions, maximum care must be taken to avoid any exposure to light. The valuable technical assistance of Mr. K. May is gratefully acknowledged. 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. References Greenwood M. R. and Von Burg R. in Merian E. Editor, “Metalle in der Umwelt,” Verlag Chemie Weinheim 1984, p. 511. Schreiber W. Sci. Total Environ.1983 31 283. Egan H. Proc. Anal. Div. Chem. SOC. 1978 15 117. Harms U. Z. Lebensm. Unters. Forsch. 1976 162 365. May K. Ahmed R. Reisinger K. Torres B . and Stoeppler, M. in Lekkas T. D. Editor “Proceedings of the 5th International Conference on Heavy Metals in the Environ-ment Athens September 1985,” Volume 2 CEP Consultants, Edinburgh 1985 p. 513. Marston A. W. and Wright H. T. J . Biochem. Biophys. Methods 1984 9 307. Zuo Y. and Pang S. Huanjing Kexue Xuebao 1985,5,239. Ponnamperuma C. “Investigation of Mercury - Amino Acid Complexes in the Aqueous Environment,” NTIS Govt. Rep. Announc. Index (U.S.) 1984 84 38. Christman D. R. and Ingle J. D. Jr. Anal. Chim. Acta, 1976 86 53. Stoeppler M. and Matthes W. Anal. Chim. Acta 1978 98, 389. Torres B. Reisinger K. Stoeppler M. and Niirnberg, H. W. in Miiller G. Editor “Proceedings of the 4th International Conference on Heavy Metals in the Environ-ment Heidelburg September 1983,” Volume 2 CEP Consul-tants Edinburgh 1983 p. 838. Stoeppler M. Spectrochim. Acta Part B 1983 38 1559. May K. and Stoeppler M. in preparation. Rodriguez-Vasquez J. A. Talanta 1978 25 299. Ahmed R. and Stoeppler H. in Stoeppler H. and Ourlech, H. W. Editors “Decomposition and Stability Studies of Methylmercury in Water,” Jii1.-Spez. No. 349 Kernforschungsanlage Jiilich Jiilich 1986 52 pp. May K. and Stoeppler M. Fresenius 2. Anal. Chem. 1984, 317 248. Farey B. J. Nelson L. A, and Rolph M. G. Analyst 1978, 103 656. Ahmed R. May K . and Stoeppler M. Sci. Total Environ., in the press. Paper A611 92 Received June I6th 1986 Accepted July 30th 198

ISSN:0003-2654    

DOI:10.1039/AN9861101371

出版商:RSC    

年代:1986

数据来源: RSC

摘要:

ANALYST, DECEMBER 1986, VOL. 111 1375 Rapid Atomic Spectrometric Determination of Sodium, Potassium, Calcium and Magnesium in Powdered Milk by Direct Dispersion Miguel de la Guardia Departamento de Quimica Analitica, Facultad de Quimicas, Burjasot, Valencia, Spain Amparo Salvador Departamento de Didactica de las Ciencias Experimentales, Escuela Universitaria del Profesorado de Educacidn General Basica, Valencia, Spain and Pilar Bayarri and Rosaura Farre Departamento de Bromatologia, Toxicologia y Quimica Analitica, Facultad de Farmacia, Valencia, Spain A method for the determination of Na, K, Ca and Mg in powdered baby milk formulas is proposed. The procedure consists in weighing 0.1 g of milk, adding 10 ml of a solution of lanthanum (4% mlV) and dispersing the mixture in water to give a volume of 100 ml.The samples are then directly introduced into an air - acetylene flame and Na, Ca and Mg are determined by atomic absorption spectrometry using multi-element standards containing La. K is determined by atomic emission spectrometry. Keywords: Direct dispersion atomic spectrometry; powdered baby milks; sodium, potassium, calcium and magnesium determination The concentrations of alkali and alkaline earth elements present in powdered baby milk formulas must be kept within narrow limits in order to satisfy infants' nutritional require- ments and to avoid physical disorders. Several international organisations , including ESPGAN' and the Codex Alimen- tarius Mundi,2 recommend minimum and maximum limits for sodium, potassium, calcium and magnesium in powdered milk, and these are used to set legal limits on element concentrations. It is common practice, both in industry and in quality control laboratories, to determine these four elements when analysing dairy formulas. The determinations are often carried out in liquid and powdered baby milk formulas by the destruction of organic material, either by wet digestion3-5 or by dry ashing.6 Other methods that do not require this destruction process have been proposed for liquid milk, e.g., protein precipitation and separation ,3,7-9 or dilution followed by direct analysis4J0,11.No such alternatives have been suggested for powdered milks, probably because they are not totally soluble in water. However, methods that utilise either flame emission or atomic absorption spectrometry have been proposed for the direct determination of elements in solid samples in suspension other than powdered milk,l2-15 although L'vov16 has pointed out that the direct determination of solids by atomic absorption spectrometry is impossible because of the characteristics of conventional nebulisers. The aim of this study was to compare the results obtained when sodium, potassium, calcium and magnesium in pow- dered milk are determined using atomic absorption and flame emission spectrometry, after ashing or dispersion in water. A series of samples was analysed by two different analysts in different laboratories.Experimental Apparatus A Pye Unicam SP 1900 atomic absorption spectrometer equipped with Na, Ca and Mg hollow-cathode lamps was used for the test method and a Perkin-Elmer 2380 atomic absorp- tion spectrometer with Na, K, Ca and Mg hollow-cathode lamps was used for the reference method.Reagents Test method Sodium, potassium, calcium and magnesium standard solu- tions, 100 p.p.m. Prepared from NaCl (Probus), KCl (Merck), CaC03 (Probus) and Mg metal (Merck). Concentrated nitric acid. From Probus, d = 1.38. Lanthanum solution. Prepared from La203 (Phaxe), 4% NaCl, 2% mlV aqueous solution. Multi-element standards containing sodium (0.5-4 p.p.m.), potassium (0.5-10 p.p.m.), calcium (0.5-10 p.p.m.) and magnesium (0.02-1.5 p.p.m.) with 0.4% mlvlanthanum were prepared from these solutions. Genapol PF 10 surfactant. Condensed from ethylene oxide and propylene oxide (Hoechst). A-11 milk sample (powder).Certified by the International Atomic Energy Agency. Commercial samples of powdered milk. mlV, and from La203 (Merck), 5% mlV. Reference method Sodium, potassium, calcium and magnesium standard solu- tions, 100 p.p.m. Prepared from NaCl (Merck), KCI (Merck), CaC03 (UCB) and Mg metal (Panreac). Lanthanum solution, 5% mlV. Prepared from La203 (Merck). NaCl, 2% mlV aqueous solution. Concentrated nitric acid. From Probus, d = 1.38. The following standards were prepared from the stock solutions: sodium (0.5-4 p.p.m.), potassium (0.5-2 p.p.m., containing 2000 p.p.m. of sodium in the form of NaCl), calcium (1-10 p.p.m.), magnesium (0.2-1 p.p.m., containing 0.1% lanthanum). A-11 milk sample (powder). Certified by the International Atomic Energy Agency. Commercial samples of powdered milk.General Procedure Test method Tests were carried out to determine whether commercial surfactants encouraged the dispersion of the samples in water1376 ANALYST, DECEMBER 1986, VOL. 111 Table 1. Instrumental parameters Intensity of Angle of Flow-rate Flow-rate Wavelength/ lamp current/ Slit width/ burner/ Burner height/ of C2H2/ of air/ Element nm mA mm degrees cm min min Sodium . . . . . . 589.6 6 0.15 30 0.8 1 5 Potassium . . . . . . 766.5 - 0.15 90 0.8 1 5 Calcium . . . . . . 422.7 6 0.11 0 1 0.8 5 Magnesium . . . . 285.2 3 0.15 10 0.6 0.9 5 Table 2. Effect of the addition of surfactants on the determination of sodium, potassium, calcium and magnesium in powdered milks With surfactant Without surfactant Element X S t x* SS Sodium .. . . 0.26 0.006 0.27 0.006 Potassium . . 0.73 0.006 0.77 0.006 Calcium . . . . 0.48 - 0.49 0.006 Magnesium . . 0.45 0.006 0.46 0.006 * k = Mean value of the measured concentration in three independent analyses expressed as grams of element per 100 g of milk powder. t s = Standard deviation for three independent assays. between the slopes of the calibration graphs obtained with multi-element and with single-element standards of +8.6% for sodium, +0.9’/0 for potassium, +1.1% for calcium and -4.3% for magnesium. These differences can be reduced, especially for sodium and calcium, by reaching a compromise between the best sensitivity and the best agreement between absorbance values for each element in single and multi- element standard solutions. The selection of the optimum experimental conditions is based on this criterion.The burner was placed at an appropriate angle in order to reduce the sensitivity of the readings and increase the dynamic range, thereby avoiding excessive dilution of the samples and making it possible to measure the four elements in the same sample aliquot. Table 1 shows the instrumental conditions selected. Table 3. Results obtained for the analysis of a certified sample of the International Atomic Energy Agency’s A-1 1 powdered milk reference sample Certified Reference Element value, YO Test method, YO method, YO Sodium . . 0.442 k 0.033 0.42 f 0.03 0.44 Potassium . . 1.72 f 0.10 1.70 f 0.04 1.75 Calcium . . 1.29 rt 0.08 1.27 k 0.04 1.31 Magnesium 0.110 rt 0.008 0.110 f 0.004 0.10 and the conditions for using the multi-element standards were also studied.In view of the results obtained, the following procedure was used in further work. After weighing 0.1 g of milk powder, 10 ml of a 4% m/V solution of lanthanum were added and the mixture was dispersed in water to a total volume of 100 ml. The samples were then introduced directly into the air - C2H2 flame, and the absorption of sodium, calcium and magnesium and the emission of potassium under instrumental conditions estab- lished previously were measured using the multi-element standards containing sodium, potassium, calcium and mag- nesium in appropriate proportions and 0.4% m/V of lan- thanum. Reference method Weigh 1-2 g of a milk powder in a porcelain crucible, heat with a Bunsen burner flame until ignition and ash in a muffle furnace at 500°C for 24 h. Dissolve the residue using the minimum volume of concentrated nitric acid and dilute to 100 ml.Take 5 mi from this solution and dilute it to 100 ml, adding NaCl for the K determination, or La for the Mg determina- tion. Results and Discussion Instrumental Parameters Single-element standards of each of the elements under study were prepared in the presence of 0.4% lanthanum as an ionisation buffer and were compared with standards at the same concentration that contained all the elements con- sidered. Under the instrumental conditions that give the maximum sensitivity for each element, we found relative differences Use of Surfactant In order to achieve a better dispersion of the milk samples in water, and to guarantee stability during the assay, we added surfactants to the mixture.A non-ionic surfactant, Genapol PF 10, was chosen because it does not contain any of the elements under consideration. The sodium, potassium, calcium and magnesium content in a sample of powdered milk was determined using the method described under General Procedure, and the results were compared with those obtained by adding 0.75% of the surfactant to samples and standards. As can be seen from Table 2, the values obtained with the two procedures are similar, and it can therefore be concluded that it is not necessary to add surfactants in order to guarantee thorough dispersion of the samples. Analysis of Certified Sample In order to evaluate the precision of the two procedures used in this work, a determination of the concentration of the four elements in a certified sample of the International Atomic Energy Agency’s A-11 powdered milk was carried out.The results obtained are summarised in Table 3, from which it can be seen that the two procedures give analogous results and are in agreement with the reference values if the precision of the determination is taken into account. Comparison of Test Method with Reference Method The content of sodium, potassium, calcium and magnesium in powdered maternal milk and in other powdered milks was determined. The data given in Table 4 indicate that the results obtained from the two methods are similar, neither method yielding systematically lower values than the other. Table 5 shows the results obtained expressed in mg kcal-1.When these data are compared with the maximum and minimum concentrations established by legislation in several different countries (Table 6) and with the limits recommended by ESPGAN and the Codex Alimentarius Mundi (Table 7), it is observed that the values obtained by either procedure in most of the samples are within the legally permitted limits.ANALYST, DECEMBER 1986, VOL. 111 1377 Table 4. Comparison of determinations of sodium, potassium, calcium and magnesium by direct analysis (test method) and by analysis following incineration of samples (reference method). Results are expressed as the means of three determinations k standard deviation. Samples 5 and 6 correspond to dietetic milks that are not subject to the same laws as maternal milk. These data are given to show that the procedures are applicable to different milk samples Sodium Potassium Calcium Magnesium Reference Test Reference Test Reference Test Reference 1 2.70 f 0.06 2.60 k 0.04 7.7 k 0.1 7.40 k 0.08 4.90 IL 0.06 5.00 k 0.08 0.460 f 0.006 0.420 f 0.007 2 2.90 3.0 k 0.2 8.40 f 0.06 8.00 k 0.07 6.1k0.2 6.10+0.05 0.560k0.006 0.53k0.02 3 2.50 k 0.06 2.6 f 0.3 6.60 k 0.06 6.2 t 0.3 3.8 3.9 k 0.2 0.450 0.41 t 0.02 4 2.60k0.06 2.60k0.01 8.60k0.06 8.0 5.80 k 0.06 6.13 0.530 k 0.006 0.47 t 0.007 5 9.7 k 0.3 11.76 t 0.04 22 * 2 20.1 k 0.6 8 .2 f 0.2 9.5 + 0.08 2.96 k 0.03 3.27 i- 0.06 6 8.8 k 0.1 11.3 t 0.1 20.1 +- 0.5 18.1 + 0.3 8.10 _t 0.09 9.4 k 0.08 2.854 0.004 3.20 i- 0.01 Sample Test Table 5. Comparison of test and reference methods (results expressed in mg per 100 kcal) Sodium Potassium Calcium Magnesium Sample Test Reference Test Reference Test Reference Test Reference 1 54 52 154.0 148.0 98 100 9.2 8.4 2 60.4 62.5 175.0 166.6 127.1 127.1 11.6 11.0 3 48.1 50.0 126.9 119.2 73.1 75.0 8.6 7.9 4 57.5 57.5 190.3 177.0 128.3 135.6 11.7 10.4 ~ _____ _ _ _ _ ~ Table 6.Legislation: permitted sodium, potassium, calcium and magnesium concentrations in milk formulas for infants expressed in mg per 100 kcal USA Canada Net herlands Thailand Yugoslavia Spain Min. Max. Min. Max. Min. Max. Min. Max. Min. Max. 60 20.0 60.0 20.0 60.0 20.0 60.0 20.0 60.0 20.0 60.0 - 80.0 200.0 80.0 200.0 75.0 150.0 80.0 200.0 80.0 200.0 Element Min. Max. Sodium . . - Potassium . . - Calcium . . - - 50.0 - 50.0 - 50.0 140.0 50.0 - 50.0 - 6.0 - 6.0 - 6.0 12.0 6.0 - 6.0 - Magnesium .. - - Table 7. Recommendations for sodium, potassium, calcium and magnesium content (mg per 100 kcal) in an adapted baby formula. N.s. -= not specified Codex Alimentarius Element ESPGAN Min . Max. Sodium . . . . Max.40-48 20 60 Calcium . . . . Min. 60 50 N.s. Magnesium . . . . Min.6 6 N.s. Conclusions The determination of sodium, potassium, calcium and mag- nesium in powdered milk by atomic spectrometric analysis of dispersed samples gives results that are analogous to those obtained when the samples are first incinerated. The method has the advantage of requiring less manipulation of the samples and is less time consuming than the standard methods also studied. We therefore propose that this method could be used routinely to control the alkali and alkaline earth metal content of powdered milk.2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. Codex Alimentarius Commission, “Recommended Interna- tional Standards for Foods for Infants and Children,” Joint FAOIWHO Food Standards Programme, FAOIWHO, Rome, 1976, CACIRS 72/74. Maurer, J., 2. Lebensm. Unters. Forsch., 1977, 165, 1. Razifard, R., Lait, 1972, 52, 567. Tanner, J. T., J. Assoc. Off. Anal. Chem., 1982, 65, 1488. Murthy, G. K., J. Dairy Sci., 1967, 50, 313. Juarez, M., Martinez-Castro, I., Ramos, M., and Martin- Alvarez, P. J., Milchwissenschaft, 1979, 34, 149. Brooks, I. B., Luster, G . A. and Easterly, D. G., At. Absorpt. Newsl., 1970, 9, 93. Juarez, M., and Martinez-Castro, I., Rev. Agroquim. Tecnol. Aliment. 1979, 19, 45. Rebmann, V. H . , and Hoth, H. J., Milchwissenschaft, 1971,26, 411. Arpadjan, S., and Stojanova, D., Fresenius Z. Anal. Chem., 1980, 302, 206. Willis, J. B., Anal. Chem., 1975, 47, 1752. O’Reilly, J. E., and Hicks, D. G., Anal. Chem., 1979,51, 1905. Van Loon, J . C., Anal. Chem., 1980,512, 955A. Fietkav, R . , Wichman, M. D., and Fry, R . C . , Appl. Spectrosc., 1984, 38, 118. L’vov, B. V., Talanta, 1976, 23, 109. References 1. ESPGAN, Committee on Nutrition, Actu Pediutr. Scand., 1977, Suppl. No. 262. Paper A611 79 Received June 3rd, 1986 Accepted July 14th, 1986

ISSN:0003-2654    

DOI:10.1039/AN9861101375

出版商:RSC    

年代:1986

数据来源: RSC