- Email: [email protected]
Manual and fia methods for the determination of cadmium with malachite green and iodide
Taalanra, Vol. 35, No. I I, pp. 885-889, 1988 Printedin Great Britain.All rightsreserved
0039-9 140/88 $3.00 + 0.00 Copyright Q 1988 Pergamon Press plc
MANUAL AND FIA METHODS FOR THE DETERMINATION OF CADMIUM WITH MALACHITE GREEN AND IODIDE I. LOPEZ GARCIA, P. NAVARRO and M. HERNANDEZ CORD~BA Department of Analytical Chemistry, Faculty of Sciences, University of Murcia, 30001 Murcia, Spain (Received 27 November 1987. Revised 4 April 1988. Accepted 9 July 1988) Summary-A sensitive and rapid spectrophotometric method for the determination of cadmium is described, based on the formation of a blue complex at pH 4 between the anionic iodide complex of cadmium(I1) and Malachite Green; the colour is stabilized with poly(viny1 alcohol). The calibration graph for measurement at 685 nm is linear over the range l-50 fig of cadmium per 25 ml of final solution, with a relative standard deviation of + 1.7% for 1 j~g/ml cadmium. The molar absorptivity is 6.1 x IO4 I. mole-’ cm-‘. The method can be successfully adapted for FIA, the peak height being proportional to the cadmium concentration over the range 0.1-3 pg/ml; a two-channel manifold is used and an improvement in selectivity is obtained. The use of a gradient tube is demonstrated to give a good calibration for Cd(R) over the range 2 x lo-‘-2 x 10m6M.
The determination of traces of cadmium has attracted considerable attention owing to the toxicity of cadmium and its compounds. In addition to the AAS
working range, cadmium can be determined by FIA in the range 2 x 10e6-2 x 10m2M. The aim of our work was to develop a rapid method for the routine determination of cadmium, for industrial purposes. We report here our findings on the interaction of the anionic iodo-complex of cadmium with Malachite Green, as the basis for a spectrophotometric method involving no extraction step. To decrease the analysis time the batch procedure has been adapted to FIA systems. The use of simple FIA manifolds allows a high sampling rate with good reproducibility and moderate selectivity.
and ICP methods, the use of ternary complex formation provides an attractive alternative to the conventional extraction with dithizone.’ The procedures so far described are sensitive, but as a general rule, separation of the complex into an organic solvent is needed.24 However, certain dyes show spectral shifts on formation of ion-association systems involving cadmium.>’ If this occurs, a solvent extraction step is unnecessary. With this in mind, we have studied the interaction of Malachite Green with cadmium in the presence of large amounts of iodide, and have EXPERIMENTAL developed a new method for the spectrophotometric Apparatus determination of cadmium without extraction. A Pye-Unicam SP8-200 spectrophotometer with IO-mm The characteristics of the proposed method make path-length glass cells was used for recording spectra and it suitable for use in flow-injection analysis (FIA). making absorbance measurements, and a Radiometer Although several FIA methods have been proposed PHM63 pH-meter was used for pH measurements. for cadmium, with electrometric,8-‘3 atomicThe flow injection system (see Fig. 1) consisted of a absorption’4.‘5 and fluorimetric*’ detection, very few Gilson HP8 peristaltic pump, an Chnnifit injection valve and a Perkin-Elmer 550 SE spectrophotometer as detector. spectrophotometric procedures have been developed Connecting tubing was 0.5 mm bore P’TFE tubing and either with2’~23or withoutZ4 extraction. In this study various end fittings and connectors (Omnifit) were used. the cadmium-iodide-Malachite Green system has The output from-the detector was monitored either by a Perkin-Elmer 561 chart recorder or a Perkin-Elmer Siama been used as the basis for a FIA-spectrophotometric 15 chromatography data station. method for the routine determination of cadmium. A A gradient tube consisting of a straight “Perspex” tube very simple manifold is proposed: good selectivity is approximately 50 mm long and 2 mm in bore was used. obtained from peak height measurements. Reagents The use of peak width as a quantitative parameter All inorganic chemicals used were of analytical-reagent in FIA was first demonstrated by RdiiEka et al.25.26 grade and were used without further purification. Doubly and since then, the method has been studied and distilled water was used throughout. Malachite Green soluapplied by various workers.27m29Recently an accurate tion (2 x lo-‘M) was prepared from a commercial product. Potassium iodide-ascorbic acid solution was prepared by equation relating peak width to analyte concendissolving 16.6 g of potassium iodide and 2 g of ascorbic tration, when FIA manifolds with mixing chambers acid in water and diluting to 100 ml; a fresh solution was or other gradient-forming devices are used, has been prepared every week. Poly(vinyl alcohol) (PVA) solution, studied in depth by Tyson.“.” When advantage is I %. was prepared from commercially available PVA. Stantaken of the peak-width method to extend the dard cadmium solution was prepared by dissolving 3.00 g of 885
886 C2 P
R, -
R2 -_
wcz+?-@ Cl
T
W
Fig. I. Two-line FIA manifold where the carrier solution (see text for details) is a mixture of two solutions each propelled at 0.7 ml/min by the peristaltic pump P; S = sample injector (135 gl loop size), C, and C, = coils (0.5 mm bore, lengths 1 m and 0.25 m, respectively), D = detector operating at 685 nm; R = recorder; T=gradient tube for peak-width measurements (2 mm bore, 50 mm length), R, and Rz =reagent solutions; W = waste. high-purity cadmium in 10 ml of concentrated hydrochloric acid and accurately diluting to 1 litre with water. Acetic acid/sodium acetate buffer solution, 2M, with pH 4.0, was made. Manual batch procedure
Transfer up to 20 ml of sample containing no more than 50 pg of cadmium into a 25-ml standard flask and dilute to 20 ml with water if necessary. Add 1 ml of iodide/ascorbic acid solution, 1 ml of buffer solution, 2 ml of PVA solution, mix and add 1 ml of Malachite Green solution. Mix thoroughly and measure the absorbance at 685 nm after 1.5 min, against a reagent blank. Beer’s law is obeyed over the concentration range I-50 .ug of cadmium in 25 ml of solution. FIA procedure
Use the two-channel manifold shown in Fig. 1. The peak height at 685 nm is proportional to cadmium con~ntration between 0.1 and 3 pig/ml (f35911 sample loop). If the peak-width method is preferred, use the gradient tube instead of the coil. Prepare cadmium solutions covering the range 2 x IO-‘-2 x 10e2M by dilution of the stock solution and inject 1500 ~1 of each into the manifold. Plot a graph of the time interval between the doublet peaks (1,) us. the logarithm of the initial cadmium concentration (log C,). RESULTS AND
DISCUSSlON
Batch method
Preliminary studies indicated that a major problem was the gradual precipitation of the complex on standing, which made absorbance measurements difficult. The complex was colloidal in nature and
could be stabilized by addition of the protective colloid PVA, which successfully retarded precipitation of the complex, even on standing overnight. As a matter of routine, PVA was therefore added before the other reagents, for stabilization purposes. Figure 2 shows the abso~tion spectra of MaIachite Green with different amounts of cadmium in the presence of an excess of potassium iodide at pH 4. It is evident that the interaction between the cadmium iodide complex and the Malachite Green results in a considerable red shift and that the complex shows maximum absorption at 685 nm, compared with 615 nm for the reagent. The optimum pH was found by using universal buffer solutions over the range 2-8. Measurements of the absorbances of the complex and the reagent blank
700
600
650
Wavelength (nm 1 Fig. 2. Absorption spectra: (A) reagent blank; (B)--(E) as for (A) with the addition of 14, 18, 42 and 56 pg of
showed that the colour system is unaffected by pH over the range 2-5, but iodine is liberated at below pH 2. Although the reaction proceeds over a wide pH range, the pH of the sample and the blank should not be widely different, as the blank absorbance varies slightly with pH (Fig. 3). The pH was maintained at 4.0 in all subsequent investigations. To avoid the use of the complex universal buffer solution, simpler buffers for pH 4.0 were tested. Either potassium hydrogen phthalate or an acetate buffer could be used without affecting the colour system; varying the buffer concentration in the range 0.04-o. 12M did not affect the absorbance. All subsequent investigations were done with the acetate buffer. Two series of experiments were performed to investigate the influence of the concentration of the reagents on the development of the colour. In one, various amounts of 0.2M potassium iodide were added to a mixture of 1 ml of 20 pgg/ml cadmium solution and 5 ml of 5 x 10e4M Malachite Green. The optimum amount was found to be 5 ml (Fig. 4). In the second series, various amounts of Malachite 0.6
0.4 s 6 0
/of
C
$ 2
0.2
tw”.k
h
Fig. 3. Effect of pH on absorbance at 685 nm: (A) reagent blank (reference water); (B) and (C) with 15 pg of cadmium [(B) reference water and (C) reference reagent blank].
Determination
of cadmium 0.6
0.6
I
I
I
s
10
1.5
887
r
Fig. 4. Effect of iodide concentration on absorbance at 685
nm: (A) reagent blank (reference water); (B) and (C) with 15 pg of cadmium [(B) reference water and (C) reference reagent blank]. Green were added to 1 ml of 1M potassium iodide. The colour development reached a maximum at 5 x 10m5M Malachite Green and remained the same for Malachite Green concentrations up to 10e4M (Fig. 5). Poly(viny1 alcohol), PVA, plays an important role as stabilizing agent and its position in the order of addition of the reagents significantly affects the development of the colour. In the absence of PVA the colour faded gradually, and if PVA was added after the dye the absorbances were poorly reproducible. The best results were obtained when the PVA was added before the dye, as in the procedure. Under these conditions a standing time of 15 min was necessary in order to obtain constant absorbance measurements. The PVA concentration could be varied over the range 0.06412% without effect. The ratio of cadmium to Malachite Green in the complex in the presence of an excess of iodide was established as 1:3 by the conventional Job method. The ratio of cadmium to iodide in the complex was investigated in the presence of an excess of Malachite Green, and evidence for formation of the pentaiodocadmate(I1) anion was found. Since the method uses a large excess of both iodide and dye, it is concluded that the complex formed is [CdI,]‘-[R+], where R+ is the Malachite Green cation. Beer’s law is obeyed over the range l-50 pg of
6
4
12
Malachite Green (IO-‘&f
ml of 0.2M KI
cadmium in a final volume of 25 ml. Under the conditions described, the molar absorptivity is 6.1 x 10m4 1.mole-’ . cm-‘. A series of 10 standard solutions each containing 25 pg of cadmium was analysed; the standard deviation and relative standard deviation (RSD) of the absorbances were 0.008 and 1.7%, respectively. Numerous cations and anions were examined as potential interferents. The results are given in Table 1, column A. The tolerance limit for the foreign ion was taken as the concentration causing an error of f3% in the absorbance. The selectivity can be enhanced by using the FIA system, where kinetic effects play an important role (Table 1, column B). Flow-injection
method
Since the absorbance for the blank is low and the reaction is fast, the method described here is suitable for the routine determination of cadmium. A simple manifold using a mixture of Malachite Green and iodide as carrier is not suitable, because the Malachite Green/iodide ion-pair precipitates in the reagent reservoir. When a two-channel FIA manifold (Fig. 1) is used, there are no problems of precipitation in the coil or in the flow-cell, owing to the very short time of contact between the reagents. As a consequence, it is not necessary to add PVA, and the development of the colour is then faster.
Table 1. EtTect of diverse ions on the determination
of 0.7 pg/ml cadmium Tolerance ratio
IIonl/lCd(II)l Suecies added Fluoride, nitrate, citrate Sulphate Mg, Ca, Ba, Co(II), Ni, Zn, Al, Mn(I1) Bromide, U(VI), Fe(III), Cr(II1) NTA, Cr(VI), V(V) As(V), Tl(III), MO(W), Pb(I1) EDTA, perchlorate, Pd(II), Pt(IV), Cu. Hg(II), Ag, Au(III) *Maximum molar ratio tested.
TAL35r,,--E
1
Fig. 5. Effect of Malachite Green concentration on absorbance at 685 nm: (A) reagent blank (reference water); (B) and (C) with 15 pg of cadmium [(B) reference water and (C) reference reagent blank].
A,
B,
Spectrophotometry
FIA
4000. 2000 2000 1000 200 10 1
40002 5000 3000 3000 500 100 5
I.
888
i.OPEZ GARCIA
et al.
on the ordinate. The RSD (P = 0.05) for the determination of 1.5 pg/ml cadmium was 1.8%, at a sampling frequency of 120 samples per hour, Peak -width FIA ~e?hod
The mathematical model usedB~loassumes a molar reagent concentration C,R in the carrier flow-rate Q ($l/sec), an injected sample volume of V, ~1, with molar concentration Ci, and a mixing chamber of volume V gl. The time interval ces (XC) between the doublet peaks is given by:
4LI0
I
I
200
Loop size I 0
400 (pl I l
I
I
I 600
In order to strike a compromise between linearity of calibration and the sampling frequency a gradient tube was included in the manifold. Solutions containing cadmium in the range 2 x 10s6 - 2 x IO-‘&t were prepared and a 1500~P1 volume of each was injected into the carrier. A least-squares regression analysis of the graph of t,, vs. In C,S gave a slope of 14.5 f: 0.S and a correlation coefficient of 0.9976.
I 3
Pumpi& rat6 ( mGmin 1 A I 0
I 3
I 2
I 3
C, coil length t m 1 o Fig. 6. Effect of loop size (0). pumping rate (A) and C, coil length (0) on peak height at 664 nm. Sample injected 1 fig/ml Cd(H). The arrows mark the selected values for these parameters.
Figure 6 shows the effect of pumping rate, loop size and length of coil C,. From these results it seems clear that a coil length of 1 m is best if a decrease in
sensitivity is to be avoided. On the other hand, the tube C, has to be as short as possible, to avoid dispersion effects. The optimum values for the FIA variables are as follows: flow-rate (q) = 1.4 ml/min; injected volume (u,) = 135 ~1; reactor lengths, C, = 1 m and C, = 25 cm; inner diameter of the reactors (#J)= 0.5 mm. These values allow the cadmium-iodide-Malachite Green complex to be formed by the time the sample plug passes through the detector, where it is monitored at 685 nm. The optimum concentrations for the carrier solutions were found to be 0.08M for the potassium iodide (R,) and 2.5 x 10m4M Malachite Green (R,), both in 0.1 M sodium acetate-acetic acid buffer (PI-I 4). With the manifold described, a plot of peak absorbance (corrected for the blank) vs. concentration of cadmium in the sample injected was linear over the range 0.1-3 pg/mI, with a small negative intercept Table 2. Determination
The procedure described provides a simple means for the determination of trace amounts of cadmium. The sensitivity, similar to that of the standard dithizone method, is adequate for routine purposes but the procedure shows limited sensitivity. To validate the method, four samples of liquids from an electrolysis plant manufa~tu~ng zinc and cadmium were analysed by the batch procedure without any pretreatment. The results, shown in Table 2, were compared with those obtained by AAS. Recovery tests for cadmium were also made. ~c~~~~~e~~e~e~f-~e
REFERENCES 1. Z. Marczenko, ~~c~~o~ho!o~et~ic ~ezer~~nation of Elements, pp. 177, 322. Horwood, Chichester, 1976. 2. P. P. Kish and I. S. Balog, Zh. Analit. Khim., 1977, 32, 482. 3. S. Hoshi, T. Hirai, S. Inoue and M. Matsubara, Eunseki Kagaku, 1984, 33, 565.
of cadmium in liquid samples from zinc production Recovery test*
Sampte Sample Sample Sample
I II II1 IV
Cadmium found, rcgW 825 640 I.6 180
authors are grateful to the CAI-
CYT (project no 84-0374) for financial support. Thanks are also extended to the Espatiola de1 Zinc Company for supplying the liquid samples.
Taken, /@Y 16.5 12.8 16.0 9.0
*IOpg of cadmium added.
Found, w 26.4 23.0 25.9 19.1
Recovery, % 99 102 99 101
Cadmium found by AAS, &ml 827 637 1.52 182
Determination 4. J. Courtot-Coupez and P. Geurder, Bull. Sot. Chim. France, 1961, 1942. 5. Liu Shaopu, Liu Yi and Liu Zhonfan, Mikrochim. Acta, 1983 IQ-355 6. T. R. Prasada and T. V. Ramakrishna. Analyst, 1982, 107,704.
I. B. Huang, S. Yu and B. Pu, Fenxi Huaxue, 1986, 14, 279. 8. W. Frenzel and P. Braetter, Anal. Chim. Acra, 1986, 179, 389. 9. A. Hu, R. E. Dessy and A. Granelli, Anal. Chem., 1983, 55, 320. 10. J. Wang, H. D. Dewald and B. Greene, Anal. Chim. Acra, 1983, 146, 45. 11. P. W. Alexander and U. Akapongkul, ibid., 1983,148, 103. 12. J. Wang and H. D. Dewald, ibid., 1983,153,325; 1984, 162, 189;Anal. Chem., 1984,56, 156; Tafanra, 1984,31, 387. 13. G. Schulze, M. Husch and W. Frenzel, Mikrochim. Acra, 1984 1, 191. 14. Z. Fang, S. Xu and S. Zhang, Anal. Chim. Acfa, 1984, 164,41.
15. S. Olssen, L. C. Hansen, Analyst, 16 K. E. Lawrence, Chem., 1984, 56,
R. Pessenda, J. RdiiEka and E. H. 1983, 108, 905. G. W. Rice and V. A. Fassel, Anal. 289.
of cadmium
889
17. K. BiickstrBm, L. G. Danielsson and L. Nord, Analyst, 1984, 109, 323. 18. Z. Fang, J. RtiEka and E. H. Hansen, Anal. Chim. Acra, 1984, 164, 23. 19. S. D. Hartenstein, J. Rbiieka and G. D. Christian, Anal. Chem., 1985, 57, 21. 20. J. L. Burguera, M. Burguera and A. Townshend, Anal. Chim. Acra, 1981, 127, 199. 21. 0. Klinghoffer, J. RdZiEka and E. H. Hansen, Talanta, 1980, 27, 169. 22. J. L. Burguera and M. Burguera, Anal. Chim. Acta, 1983, 153, 207. 23. E., A. Jones, Tech. Rep. Minfek, 1983, Mill, 32. 24. E. Bylund, R. Andersson and J. A. Carlsson, Pharmacia, Uppsala, Sweden, 1978. 25. J. RbIiEkaand E. H. Hansen, Flow Injection Analysis, 1st Ed., pp. 137, 138 and Fig. 6.6. Wiley, New York, 1981. 26. A. U. Ramsing, J. RdiiEka and E. H. Hansen, Anal. Chim. Acfa, 1981, 129, 1. 27. K. K. Stewart and A. G. Rosenfeld, Anal. Chem., 1982, 54, 2368. 28. J. F. Tyson, Analyst, 1984, 109, 319. 29. H. ._. L._^Pardue and B. Fields, Anal. Chim. Acra, 1981, 124, 39. 30. J. F. Tyson, ibid., 1986, 179, 131. 31. Idem, Analyst, 1987, 112, 523.
0039-9 140/88 $3.00 + 0.00 Copyright Q 1988 Pergamon Press plc
MANUAL AND FIA METHODS FOR THE DETERMINATION OF CADMIUM WITH MALACHITE GREEN AND IODIDE I. LOPEZ GARCIA, P. NAVARRO and M. HERNANDEZ CORD~BA Department of Analytical Chemistry, Faculty of Sciences, University of Murcia, 30001 Murcia, Spain (Received 27 November 1987. Revised 4 April 1988. Accepted 9 July 1988) Summary-A sensitive and rapid spectrophotometric method for the determination of cadmium is described, based on the formation of a blue complex at pH 4 between the anionic iodide complex of cadmium(I1) and Malachite Green; the colour is stabilized with poly(viny1 alcohol). The calibration graph for measurement at 685 nm is linear over the range l-50 fig of cadmium per 25 ml of final solution, with a relative standard deviation of + 1.7% for 1 j~g/ml cadmium. The molar absorptivity is 6.1 x IO4 I. mole-’ cm-‘. The method can be successfully adapted for FIA, the peak height being proportional to the cadmium concentration over the range 0.1-3 pg/ml; a two-channel manifold is used and an improvement in selectivity is obtained. The use of a gradient tube is demonstrated to give a good calibration for Cd(R) over the range 2 x lo-‘-2 x 10m6M.
The determination of traces of cadmium has attracted considerable attention owing to the toxicity of cadmium and its compounds. In addition to the AAS
working range, cadmium can be determined by FIA in the range 2 x 10e6-2 x 10m2M. The aim of our work was to develop a rapid method for the routine determination of cadmium, for industrial purposes. We report here our findings on the interaction of the anionic iodo-complex of cadmium with Malachite Green, as the basis for a spectrophotometric method involving no extraction step. To decrease the analysis time the batch procedure has been adapted to FIA systems. The use of simple FIA manifolds allows a high sampling rate with good reproducibility and moderate selectivity.
and ICP methods, the use of ternary complex formation provides an attractive alternative to the conventional extraction with dithizone.’ The procedures so far described are sensitive, but as a general rule, separation of the complex into an organic solvent is needed.24 However, certain dyes show spectral shifts on formation of ion-association systems involving cadmium.>’ If this occurs, a solvent extraction step is unnecessary. With this in mind, we have studied the interaction of Malachite Green with cadmium in the presence of large amounts of iodide, and have EXPERIMENTAL developed a new method for the spectrophotometric Apparatus determination of cadmium without extraction. A Pye-Unicam SP8-200 spectrophotometer with IO-mm The characteristics of the proposed method make path-length glass cells was used for recording spectra and it suitable for use in flow-injection analysis (FIA). making absorbance measurements, and a Radiometer Although several FIA methods have been proposed PHM63 pH-meter was used for pH measurements. for cadmium, with electrometric,8-‘3 atomicThe flow injection system (see Fig. 1) consisted of a absorption’4.‘5 and fluorimetric*’ detection, very few Gilson HP8 peristaltic pump, an Chnnifit injection valve and a Perkin-Elmer 550 SE spectrophotometer as detector. spectrophotometric procedures have been developed Connecting tubing was 0.5 mm bore P’TFE tubing and either with2’~23or withoutZ4 extraction. In this study various end fittings and connectors (Omnifit) were used. the cadmium-iodide-Malachite Green system has The output from-the detector was monitored either by a Perkin-Elmer 561 chart recorder or a Perkin-Elmer Siama been used as the basis for a FIA-spectrophotometric 15 chromatography data station. method for the routine determination of cadmium. A A gradient tube consisting of a straight “Perspex” tube very simple manifold is proposed: good selectivity is approximately 50 mm long and 2 mm in bore was used. obtained from peak height measurements. Reagents The use of peak width as a quantitative parameter All inorganic chemicals used were of analytical-reagent in FIA was first demonstrated by RdiiEka et al.25.26 grade and were used without further purification. Doubly and since then, the method has been studied and distilled water was used throughout. Malachite Green soluapplied by various workers.27m29Recently an accurate tion (2 x lo-‘M) was prepared from a commercial product. Potassium iodide-ascorbic acid solution was prepared by equation relating peak width to analyte concendissolving 16.6 g of potassium iodide and 2 g of ascorbic tration, when FIA manifolds with mixing chambers acid in water and diluting to 100 ml; a fresh solution was or other gradient-forming devices are used, has been prepared every week. Poly(vinyl alcohol) (PVA) solution, studied in depth by Tyson.“.” When advantage is I %. was prepared from commercially available PVA. Stantaken of the peak-width method to extend the dard cadmium solution was prepared by dissolving 3.00 g of 885
886 C2 P
R, -
R2 -_
wcz+?-@ Cl
T
W
Fig. I. Two-line FIA manifold where the carrier solution (see text for details) is a mixture of two solutions each propelled at 0.7 ml/min by the peristaltic pump P; S = sample injector (135 gl loop size), C, and C, = coils (0.5 mm bore, lengths 1 m and 0.25 m, respectively), D = detector operating at 685 nm; R = recorder; T=gradient tube for peak-width measurements (2 mm bore, 50 mm length), R, and Rz =reagent solutions; W = waste. high-purity cadmium in 10 ml of concentrated hydrochloric acid and accurately diluting to 1 litre with water. Acetic acid/sodium acetate buffer solution, 2M, with pH 4.0, was made. Manual batch procedure
Transfer up to 20 ml of sample containing no more than 50 pg of cadmium into a 25-ml standard flask and dilute to 20 ml with water if necessary. Add 1 ml of iodide/ascorbic acid solution, 1 ml of buffer solution, 2 ml of PVA solution, mix and add 1 ml of Malachite Green solution. Mix thoroughly and measure the absorbance at 685 nm after 1.5 min, against a reagent blank. Beer’s law is obeyed over the concentration range I-50 .ug of cadmium in 25 ml of solution. FIA procedure
Use the two-channel manifold shown in Fig. 1. The peak height at 685 nm is proportional to cadmium con~ntration between 0.1 and 3 pig/ml (f35911 sample loop). If the peak-width method is preferred, use the gradient tube instead of the coil. Prepare cadmium solutions covering the range 2 x IO-‘-2 x 10e2M by dilution of the stock solution and inject 1500 ~1 of each into the manifold. Plot a graph of the time interval between the doublet peaks (1,) us. the logarithm of the initial cadmium concentration (log C,). RESULTS AND
DISCUSSlON
Batch method
Preliminary studies indicated that a major problem was the gradual precipitation of the complex on standing, which made absorbance measurements difficult. The complex was colloidal in nature and
could be stabilized by addition of the protective colloid PVA, which successfully retarded precipitation of the complex, even on standing overnight. As a matter of routine, PVA was therefore added before the other reagents, for stabilization purposes. Figure 2 shows the abso~tion spectra of MaIachite Green with different amounts of cadmium in the presence of an excess of potassium iodide at pH 4. It is evident that the interaction between the cadmium iodide complex and the Malachite Green results in a considerable red shift and that the complex shows maximum absorption at 685 nm, compared with 615 nm for the reagent. The optimum pH was found by using universal buffer solutions over the range 2-8. Measurements of the absorbances of the complex and the reagent blank
700
600
650
Wavelength (nm 1 Fig. 2. Absorption spectra: (A) reagent blank; (B)--(E) as for (A) with the addition of 14, 18, 42 and 56 pg of
showed that the colour system is unaffected by pH over the range 2-5, but iodine is liberated at below pH 2. Although the reaction proceeds over a wide pH range, the pH of the sample and the blank should not be widely different, as the blank absorbance varies slightly with pH (Fig. 3). The pH was maintained at 4.0 in all subsequent investigations. To avoid the use of the complex universal buffer solution, simpler buffers for pH 4.0 were tested. Either potassium hydrogen phthalate or an acetate buffer could be used without affecting the colour system; varying the buffer concentration in the range 0.04-o. 12M did not affect the absorbance. All subsequent investigations were done with the acetate buffer. Two series of experiments were performed to investigate the influence of the concentration of the reagents on the development of the colour. In one, various amounts of 0.2M potassium iodide were added to a mixture of 1 ml of 20 pgg/ml cadmium solution and 5 ml of 5 x 10e4M Malachite Green. The optimum amount was found to be 5 ml (Fig. 4). In the second series, various amounts of Malachite 0.6
0.4 s 6 0
/of
C
$ 2
0.2
tw”.k
h
Fig. 3. Effect of pH on absorbance at 685 nm: (A) reagent blank (reference water); (B) and (C) with 15 pg of cadmium [(B) reference water and (C) reference reagent blank].
Determination
of cadmium 0.6
0.6
I
I
I
s
10
1.5
887
r
Fig. 4. Effect of iodide concentration on absorbance at 685
nm: (A) reagent blank (reference water); (B) and (C) with 15 pg of cadmium [(B) reference water and (C) reference reagent blank]. Green were added to 1 ml of 1M potassium iodide. The colour development reached a maximum at 5 x 10m5M Malachite Green and remained the same for Malachite Green concentrations up to 10e4M (Fig. 5). Poly(viny1 alcohol), PVA, plays an important role as stabilizing agent and its position in the order of addition of the reagents significantly affects the development of the colour. In the absence of PVA the colour faded gradually, and if PVA was added after the dye the absorbances were poorly reproducible. The best results were obtained when the PVA was added before the dye, as in the procedure. Under these conditions a standing time of 15 min was necessary in order to obtain constant absorbance measurements. The PVA concentration could be varied over the range 0.06412% without effect. The ratio of cadmium to Malachite Green in the complex in the presence of an excess of iodide was established as 1:3 by the conventional Job method. The ratio of cadmium to iodide in the complex was investigated in the presence of an excess of Malachite Green, and evidence for formation of the pentaiodocadmate(I1) anion was found. Since the method uses a large excess of both iodide and dye, it is concluded that the complex formed is [CdI,]‘-[R+], where R+ is the Malachite Green cation. Beer’s law is obeyed over the range l-50 pg of
6
4
12
Malachite Green (IO-‘&f
ml of 0.2M KI
cadmium in a final volume of 25 ml. Under the conditions described, the molar absorptivity is 6.1 x 10m4 1.mole-’ . cm-‘. A series of 10 standard solutions each containing 25 pg of cadmium was analysed; the standard deviation and relative standard deviation (RSD) of the absorbances were 0.008 and 1.7%, respectively. Numerous cations and anions were examined as potential interferents. The results are given in Table 1, column A. The tolerance limit for the foreign ion was taken as the concentration causing an error of f3% in the absorbance. The selectivity can be enhanced by using the FIA system, where kinetic effects play an important role (Table 1, column B). Flow-injection
method
Since the absorbance for the blank is low and the reaction is fast, the method described here is suitable for the routine determination of cadmium. A simple manifold using a mixture of Malachite Green and iodide as carrier is not suitable, because the Malachite Green/iodide ion-pair precipitates in the reagent reservoir. When a two-channel FIA manifold (Fig. 1) is used, there are no problems of precipitation in the coil or in the flow-cell, owing to the very short time of contact between the reagents. As a consequence, it is not necessary to add PVA, and the development of the colour is then faster.
Table 1. EtTect of diverse ions on the determination
of 0.7 pg/ml cadmium Tolerance ratio
IIonl/lCd(II)l Suecies added Fluoride, nitrate, citrate Sulphate Mg, Ca, Ba, Co(II), Ni, Zn, Al, Mn(I1) Bromide, U(VI), Fe(III), Cr(II1) NTA, Cr(VI), V(V) As(V), Tl(III), MO(W), Pb(I1) EDTA, perchlorate, Pd(II), Pt(IV), Cu. Hg(II), Ag, Au(III) *Maximum molar ratio tested.
TAL35r,,--E
1
Fig. 5. Effect of Malachite Green concentration on absorbance at 685 nm: (A) reagent blank (reference water); (B) and (C) with 15 pg of cadmium [(B) reference water and (C) reference reagent blank].
A,
B,
Spectrophotometry
FIA
4000. 2000 2000 1000 200 10 1
40002 5000 3000 3000 500 100 5
I.
888
i.OPEZ GARCIA
et al.
on the ordinate. The RSD (P = 0.05) for the determination of 1.5 pg/ml cadmium was 1.8%, at a sampling frequency of 120 samples per hour, Peak -width FIA ~e?hod
The mathematical model usedB~loassumes a molar reagent concentration C,R in the carrier flow-rate Q ($l/sec), an injected sample volume of V, ~1, with molar concentration Ci, and a mixing chamber of volume V gl. The time interval ces (XC) between the doublet peaks is given by:
4LI0
I
I
200
Loop size I 0
400 (pl I l
I
I
I 600
In order to strike a compromise between linearity of calibration and the sampling frequency a gradient tube was included in the manifold. Solutions containing cadmium in the range 2 x 10s6 - 2 x IO-‘&t were prepared and a 1500~P1 volume of each was injected into the carrier. A least-squares regression analysis of the graph of t,, vs. In C,S gave a slope of 14.5 f: 0.S and a correlation coefficient of 0.9976.
I 3
Pumpi& rat6 ( mGmin 1 A I 0
I 3
I 2
I 3
C, coil length t m 1 o Fig. 6. Effect of loop size (0). pumping rate (A) and C, coil length (0) on peak height at 664 nm. Sample injected 1 fig/ml Cd(H). The arrows mark the selected values for these parameters.
Figure 6 shows the effect of pumping rate, loop size and length of coil C,. From these results it seems clear that a coil length of 1 m is best if a decrease in
sensitivity is to be avoided. On the other hand, the tube C, has to be as short as possible, to avoid dispersion effects. The optimum values for the FIA variables are as follows: flow-rate (q) = 1.4 ml/min; injected volume (u,) = 135 ~1; reactor lengths, C, = 1 m and C, = 25 cm; inner diameter of the reactors (#J)= 0.5 mm. These values allow the cadmium-iodide-Malachite Green complex to be formed by the time the sample plug passes through the detector, where it is monitored at 685 nm. The optimum concentrations for the carrier solutions were found to be 0.08M for the potassium iodide (R,) and 2.5 x 10m4M Malachite Green (R,), both in 0.1 M sodium acetate-acetic acid buffer (PI-I 4). With the manifold described, a plot of peak absorbance (corrected for the blank) vs. concentration of cadmium in the sample injected was linear over the range 0.1-3 pg/mI, with a small negative intercept Table 2. Determination
The procedure described provides a simple means for the determination of trace amounts of cadmium. The sensitivity, similar to that of the standard dithizone method, is adequate for routine purposes but the procedure shows limited sensitivity. To validate the method, four samples of liquids from an electrolysis plant manufa~tu~ng zinc and cadmium were analysed by the batch procedure without any pretreatment. The results, shown in Table 2, were compared with those obtained by AAS. Recovery tests for cadmium were also made. ~c~~~~~e~~e~e~f-~e
REFERENCES 1. Z. Marczenko, ~~c~~o~ho!o~et~ic ~ezer~~nation of Elements, pp. 177, 322. Horwood, Chichester, 1976. 2. P. P. Kish and I. S. Balog, Zh. Analit. Khim., 1977, 32, 482. 3. S. Hoshi, T. Hirai, S. Inoue and M. Matsubara, Eunseki Kagaku, 1984, 33, 565.
of cadmium in liquid samples from zinc production Recovery test*
Sampte Sample Sample Sample
I II II1 IV
Cadmium found, rcgW 825 640 I.6 180
authors are grateful to the CAI-
CYT (project no 84-0374) for financial support. Thanks are also extended to the Espatiola de1 Zinc Company for supplying the liquid samples.
Taken, /@Y 16.5 12.8 16.0 9.0
*IOpg of cadmium added.
Found, w 26.4 23.0 25.9 19.1
Recovery, % 99 102 99 101
Cadmium found by AAS, &ml 827 637 1.52 182
Determination 4. J. Courtot-Coupez and P. Geurder, Bull. Sot. Chim. France, 1961, 1942. 5. Liu Shaopu, Liu Yi and Liu Zhonfan, Mikrochim. Acta, 1983 IQ-355 6. T. R. Prasada and T. V. Ramakrishna. Analyst, 1982, 107,704.
I. B. Huang, S. Yu and B. Pu, Fenxi Huaxue, 1986, 14, 279. 8. W. Frenzel and P. Braetter, Anal. Chim. Acra, 1986, 179, 389. 9. A. Hu, R. E. Dessy and A. Granelli, Anal. Chem., 1983, 55, 320. 10. J. Wang, H. D. Dewald and B. Greene, Anal. Chim. Acra, 1983, 146, 45. 11. P. W. Alexander and U. Akapongkul, ibid., 1983,148, 103. 12. J. Wang and H. D. Dewald, ibid., 1983,153,325; 1984, 162, 189;Anal. Chem., 1984,56, 156; Tafanra, 1984,31, 387. 13. G. Schulze, M. Husch and W. Frenzel, Mikrochim. Acra, 1984 1, 191. 14. Z. Fang, S. Xu and S. Zhang, Anal. Chim. Acfa, 1984, 164,41.
15. S. Olssen, L. C. Hansen, Analyst, 16 K. E. Lawrence, Chem., 1984, 56,
R. Pessenda, J. RdiiEka and E. H. 1983, 108, 905. G. W. Rice and V. A. Fassel, Anal. 289.
of cadmium
889
17. K. BiickstrBm, L. G. Danielsson and L. Nord, Analyst, 1984, 109, 323. 18. Z. Fang, J. RtiEka and E. H. Hansen, Anal. Chim. Acra, 1984, 164, 23. 19. S. D. Hartenstein, J. Rbiieka and G. D. Christian, Anal. Chem., 1985, 57, 21. 20. J. L. Burguera, M. Burguera and A. Townshend, Anal. Chim. Acra, 1981, 127, 199. 21. 0. Klinghoffer, J. RdZiEka and E. H. Hansen, Talanta, 1980, 27, 169. 22. J. L. Burguera and M. Burguera, Anal. Chim. Acta, 1983, 153, 207. 23. E., A. Jones, Tech. Rep. Minfek, 1983, Mill, 32. 24. E. Bylund, R. Andersson and J. A. Carlsson, Pharmacia, Uppsala, Sweden, 1978. 25. J. RbIiEkaand E. H. Hansen, Flow Injection Analysis, 1st Ed., pp. 137, 138 and Fig. 6.6. Wiley, New York, 1981. 26. A. U. Ramsing, J. RdiiEka and E. H. Hansen, Anal. Chim. Acfa, 1981, 129, 1. 27. K. K. Stewart and A. G. Rosenfeld, Anal. Chem., 1982, 54, 2368. 28. J. F. Tyson, Analyst, 1984, 109, 319. 29. H. ._. L._^Pardue and B. Fields, Anal. Chim. Acra, 1981, 124, 39. 30. J. F. Tyson, ibid., 1986, 179, 131. 31. Idem, Analyst, 1987, 112, 523.