Trace measurements of rhodium by adsorptive stripping voltammetry

0039-9140/91 $3.00 + 0.00 Pergamon Press plc

Talanra, Vol. 38, No. 5, pp. 489492, 1991 Printed in Great Britain. All rights reserved.

TRACE MEASUREMENTS OF RHODIUM BY ADSORPTIVE STRIPPING VOLTAMMETRY JOSEPHWANG and ZIAD TAHA Department of Chemistry, New Mexico State University, Las Cruces, NM 88003, U.S.A. (Received 14 August 1990. Revised 12 September 1990. Accepted 21 September 1990) Summary-A sensitive stripping voltammetric procedure for quantifying rhodium is described. The complex of rhodium with chloride ions is adsorbed on the hanging mercury drop electrode, and the reduction current of the accumulated complex is measured during a negative-going scan. Cyclic voltammetry is used to characterize the interfacial and redox behaviors. The effect of pH, chloride concentration, accumulation potential and other variables is discussed. The detection limit is 1 x lO_sM (- 1 ng/ml) with 5-min accumulation. A linear current-concentration relationship is observed up to 7 x lo-‘M and the relative standard deviation (at the 2 x 10-‘&f level) is 3.0%. Possible interferences by co-existing metals are investigated.

Because of the importance of rhodium, a sensitive method is required for its reliable measurement. In particular, the quantification of rhodium at trace levels is desired for geological surveys, catalytic applications and materials science. Spectroscopic procedures have been proposed for measuring low levels of rhodium, after solvent extraction,* liquid chromatography2 or preferential complexation3 Such schemes, however, are time-consuming and costly. The present paper describes a sensitive stripping voltammetric procedure for trace measurement of rhodium. The polarographic behavior of rhodium has been explored by different groups.67 These studies elucidated the redox mechanism of rhodium in the presence of various complexing agents. The quantification of rhodium by coulometry6 or catalytic polarography’ has also been reported. However, the utility of stripping analysis (the most sensitive electrochemical techniques) for rhodium has not been reported. The combination of the voltammetric activity of rhodium chlorocomplexes with their surface-active properties results in an effective adsorptive stripping procedure, the characteristics of which are described in this paper. EXPERIMENTAL

Apparatus

A PAR Model 264 voltammetric analyzer was used with a PAR Model 303 static mercury drop electrode.

Reagents

All solutions were prepared with doubly distilled water. A 1000 mg/l. rhodium stock solution (atomic absorption standard; Aldrich) was diluted as required for standard additions. Potassium chloride solutions were prepared by dissolving the reagent in water and adjusting the pH with hydrochloric acid or sodium hydroxide solution. Procedure

Potassium chloride solution (10 ml, pH 1.3) was pipetted into the polarographic cell and purged with nitrogen for 4 min. The preconcentration potential (- 0.20 V) was applied to a fresh mercury drop while the solution was stirred. After the preconcentration, the stirring was stopped and 15 set later the background voltamperogram obtained by applying a negative-going scan terminating at -0.60 V. A known volume of rhodium standard was then added and the adsorptive and stripping cycle was repeated with a new mercury drop. RESULTS AND DISCUSSION

Repetitive cyclic voltamperograms for 50 ng/ml rhodium in 2.5iU potassium chloride (pH 1.3) were used to evaluate the interfacial and redox behavior (Fig. 1). In the absence of prior accumulation only a small peak from the reduction of the rhodium chloro-complex was observed, at -0.52 V (Fig. 1, B). When the 489

JOSEPH WANG and ZLUJ TAIU

490

experiment was repeated with a 120-set stirring period (at -0.20 V) a large peak was observed at -0.52 V in the first negative-going scan. Subsequent scans gave a much smaller peak (of approximately constant height), indicating desorption of the product from the surface (Fig. 1, A). According to the work of Van Loon and Page,6 the complex adsorbed under these experimental conditions should be mainly RhCe-. Figure 2 shows that for a 20 ng/ml rhodium solution, the longer the preconcentration time, the more metal complex is adsorbed on the surface. and the larger is the peak current. For example, with 120-set preconcentration (Fig. 2, e) there is a fivefold enhancement of the peak current relative to that obtained without preconcentration (Fig. 2, a). As a result, excellent signal-to-background characteristics are obtained which permit measurements at nanomolar concentration levels. Also shown in Fig. 2 (inset) are plots of peak current us. preconcentration time at two levels of rhodium [(A) 20 and (B) 40 ng/ml]. In both cases the current increases linearly with time. The slopes are 9.6 nA/sec (A) and 26.2 nA/sec (B) (correlation coefficients, 0.999 and 0.996, respectively).

I

400 nA

-0.2

-0.4 Potential

I

12

0

r

0

60

nA

b

120

180

240

Fig. 1. Repetitive cyclic voltamperograms for 50 ng/ml rhodium, in 2SM KC1 @H 1.3) solution, (A) with and (B) without 120-set preconcentration (at -0.20 V), with 406 _~ rpm stirring; scan-rate, 50 mV/sec.

360

E

2 f 0

200 t

I

-0.w.s Potential

(V)

Fig. 2. Differential pulse voltamperograms for 20 ng/ml rhodium after different preconcentration times: (a) 0, (b) 30, (c) 60, (d) 90, (e) 120 set, and 400 rpm stirring. Inset are current vs. pteconcentration time plots for (A) 20 ng/ml and (B) 40 ng/ml Rh. Conditions: 2.5M KC1 @H 1.3); accumulation at -0.20 V; scan-rate 10 mV/sec; pulse amplitude, 25 mV.

The stripping current depends strongly on the solution pH (Fig. 3A). Increasing the pH from 3 to 7 results in a sharp decrease in peak height. This could be explained by the affinity of Rh3+ for hydroxide ions at high PH.~ A similar pH-current profile was obtained for polarographic work. A sharp increase in the peak width was observed as the pH was increased from 4 to 7 (Fig. 3B). The effect of convection mass transport was also evaluated.

-0.6

(V)

300

0

4

8

0

PH Fig. 3. Effect of pH on the adsorptive stripping current (A) and the peak half-width (B): 50 ng/ml Rh; preconcentration time, 60 set; other conditions as for Fig. 2.

491

Trace measurements of rhodium

Accumulation of rhodium chloro-complexes from a stirred solution gave rise to a peak current 2.2 times as large as that obtained with the quiescent solutions. The chloride ion concentration has a pronounced effect on the stripping current (Fig. 4A). The stripping peak for 50 ng/ml rhodium increases linearly with increasing potassium chloride concentration between 1 and 2SM, and then decreases. As the chloride concentration increases, the mole fraction of RhCli- increases, giving rise to a higher signal. According to the work of Shlenskaya et al.: this increase is almost linear between 1.5 and 2.5M potassium chloride, in agreement with the signal trend shown in Fig. 4A. At low chloride concentration aquo-chloro mixed-ligand complexes are formed.4g6 The dependence of the stripping current on the preconcentration potential over the range from + 0.10 to - 0.40 V was examined (Fig. 4B). The response increased sharply between 0.00 and -0.20 V and then decreased at more negative potentials. The optimal conditions for rhodium measurements are there-

3i--(4

I 0

i

6

KCI concentration

0

.-

(M)

e-0

/ 0

Accumulation

-0.2

potential

-0.4

(V)

Fig. 4. Dependence of the adsorptive stripping current on (A) the KC1 concentration and (B) the accumulation potential: 50 ng/ml Rh; other conditions as for Fig. 2.

‘r t

60

nA

Concentration

\ -0.2

Potential

-0.4

-0.6

(V)

Fig. 5. Stripping voltamperograms obtained for solutions of increasing rhodium concentration over the range l&30 ng/ml (a-c). Preconcentration for 45 set; other conditions as for Fig. 2. Inset are calibration plots for different accumulation times: (A) 45 and (B) 90 sec.

fore 2.5M potassium chloride as supporting electrolyte (pH 1.3) and a preconcentration potential of -0.20 V. Figure 5 shows stripping voltamperograms obtained for IO-30 ng/ml rhodium solutions after a 45-set preconcentration time. Calibration plots over a wider concentration range (10-60 ng/ml), obtained by using different preconcentration times, are also shown in Fig. 5 (inset). The response is linear over this range [slopes, (A) 24.7 and (B) 54.8 nA.ml.ng-‘, both with correlation coefficients of 0.9971. The detection limit was estimated to be 1OnM (1 ng/ml) with a 5-min accumulation (S/N = 3). The precision was estimated from eight successive measurements of 50 ng/ml rhodium solution (90~set preconcentration); the mean peak current was 1426 nA (range 1360-1470 nA and relative standard deviation 3.0%). The following metal ions (20 ng/ml) were tested and found not to affect the 20 ng/ml rhodium peak: Ag(I), Ni(II), Zn(II), Co(II), Pd(II), Au(III), Ti(IV), Fe(II), Fe(III), Pb(I1) and Cu(I1). A decrease in the rhodium stripping peak was observed in the presence of Mn(I1) and Sn(I1). Hence, a separation step would be required in the presence of these metals. The effects of several anions were also examined. No interference was observed in the presence of 2mM nitrate or bromide. An additional peak was observed in the presence of 2mM iodide, at ca. 140 mV negative to the peak of interest, but did not affect the determination of rhodium.

492

JOSEPHWANGand Zulu, Tw

In conclusion, the method described provides a simple approach for the determination of trace levels of rhodium. The interfacial accumulation results in a substantial enhancement of the voltammetric response, permitting convenient quantification at the ng/ml level. Depending on the sample matrix, a separation step may be required to isolate rhodium from interfering metals. REFERENCES 1. Y. Shijo, K. Nakaji and T. Shim& 519.

Analyst, 1988, 113,

2. B. J. Mueller and R. J. Lovett, Anal. Chem., 1985, 57, 2693. 3. A. L. J. Rao, U. Gupta and B. K. Pm-i, Analyst, 1986, 111, 1401. 4. D. Cozzi and F. Pantani, J. Inorg. Nucl. Chem., 1958, 8, 385. 5. J. B. Willis, J. Am. Chem. Sot., 1944, 66, 1067. 6. G. Van Loon and J. A. Page, Talanta, 1965, 12, 227. 7. P. W. Alexander and G. L. Orth, J. Electroanal. Gem., 1971, 31, App. 3. 8. J. Wang, Stripping Analysis: Principles, Instrumentation and Applications, VCH, Deerfield Beach, 1985. 9. V. I. Shlenskaya, 0. A. Efremenko, S. V. Oleinikova and I. P. Alimarin, Bull. Acad. Sci. USSR, Chem. Sci., 1969, 18, 1525.