Noble metal modifiers for antimony determination by graphite furnace atomic absorption spectrometry in biological samples

Abstract

The use of palladium, iridium or rhodium as a modifier in graphite furnace atomic absorption spectrometry was investigated for the determination of antimony in eluent fractions from high-performance liquid chromatography separation of clinical samples. The separation of albumin and transferrin at the physiological pH was carried out by ion chromatography on a Cosmogel DEAE column. As an eluent, 0.01 mol l⁻¹ Tris-HCl in NaCl 1 mol l⁻¹ gradient was used. Several fractions of 0.5 ml each were collected and the concentration of antimony was determined off-line by GFAAS. Palladium, iridium and rhodium effectively stabilise antimony in an aqueous standard solution. In the presence of proteins, Tris-HCl buffer and NaCl, rhodium loses its stabilising performance. However, palladium and iridium were found to be efficient with respect to stabilisation of antimony up to 1500 °C in matrix-containing solutions.

---

Introduction

Because of the enhanced mobility of antimony in the environment, an increased interest in this element in living organisms has been observed. The natural processes of weathering of rock and soil runoff could be a source of the increased content of antimony in air and waters.¹ However, the main source of antimony in the environment is from anthropogenic activities.^2–4

Compounds of trivalent antimony are generally more toxic than pentavalent forms, because Sb(III) ions are bound irreversibly to thiol-containing enzymes.^5,6 This fact and the affinity of antimony for sulfhydryl groups in red cells lead to problems in the oxidation and binding of oxygen by the red cells of blood.⁷ Antimony compounds are accumulated in the kidneys, lung and liver (the contents vary from 0.7 to 37 µg kg⁻¹) and could be the reason for several diseases.^8,9 The concentrations of antimony in serum vary from 0.07 to 0.76 µg l⁻¹ and in whole blood from 0.08 to 0.88 µg l⁻¹.⁹

One of the common methods used for trace element determination in clinical samples is graphite furnace atomic absorption spectrometry (GFAAS). In the direct determination of antimony, the use of a modifier was found to be essential in order to minimise the interference effects. Palladium, reduced in the graphite tube, was found to be the most universal for many elements.^10–14 Several authors also proposed the use of refractory metals, such as Zr, Rh or Ir.^15–17

Antimony is a volatile element and, therefore, its thermal stabilisation prior to atomisation is very important for avoiding losses of analyte. In this study, the effectiveness of Pd, Ir or Rh used as a modifier in a standard aqueous solution as well as in eluates from HPLC separation of proteins in serum samples was examined.

Experimental

Instrumentation

Graphite furnace atomic absorption measurements were made with a 4100 ZL spectrometer (Perkin Elmer, Germany) with longitudinal Zeeman-effect background correction. Pyrolytically coated THGA graphite tubes with an integrated L'vov platform (Perkin Elmer, Germany) were used. A hollow-cathode lamp for antimony from Beckman (Germany), operated at 12 mA, was used for all measurements. A 0.7 nm spectral bandwidth was selected to isolate the 217.6 nm resonance line. A 20 µl aliquot of the sample was injected into the graphite furnace by means of the AS-70 autosampler (Perkin-Elmer, Germany). The temperature programmes for THGA are given in Table 1.

Table 1 Graphite furnace temperature programme

Step number	Temperature/°C	Ramp time/s	Hold time/s	Read
a 1450 °C – Pd, 1350 °C – Rh, 1450 °C – Ir. b 2150 °C – Pd, 2050 °C – Rh, 2250 °C – Ir.
Programme 1: used for thermal reduction of the modifiers in the graphite furnace
1	140	8	5	−
2	150	7	20	−
3	900	10	10	−
Programme 2: used for the determination of antimony
4	140	8	5	−
5	150	7	20	−
6	Various^a	10	10	−
7	Various^b	0	3	+
8	2500	1	3	−

---

An HPLC pump, L-6210 (Merck-Hitachi), with a variable wavelength UV-VIS diode array detector, L-4500 (Merck-Hitachi), and a Cosmogel DEAE 75 × 7.5 mm id column (Nacalai Tesque, Inc., Japan) were used for sample separation. An injector (Rheodyne, USA) fitted with 5 µl loop was used for sample introduction.

Reagents and solutions

Chromatographic separation. Buffer: 0.01 mol l⁻¹ tris-(hydroxymethyl)aminomethane (Tris) (Sigma, Germany) + HCl (ultrapure) (Merck, Germany) + 1 mol l⁻¹ NaCl (POCh Gliwice, Poland). Solution of serum proteins: 4 g l⁻¹ albumin (Serva, USA) + 0.4 g l⁻¹ transferrin (Sigma, Germany).

GFAAS. Solutions of: PdCl₂ 1 g l⁻¹ (Merck, Germany); IrCl₃ 1 g l⁻¹ (Merck, Germany); RhCl₃·3H₂O (Koch-Light Laboratories Ltd., England).

A standard solution of Sb(III) at 100 µg l⁻¹ was prepared from SbCl₃ (Merck, Germany).

Procedure

HPLC separation of proteins. A model solution of serum proteins contained 4 g l⁻¹ of albumin and 0.4 g l⁻¹ of transferrin. Before separation, antimony was added to the solution of serum proteins so that the concentration of Sb(III) was 100 mg l⁻¹. The solution was injected onto the Cosmogel DEAE chromatographic column. Signals of albumin and transferrin were registered by a UV/VIS detector at 220 nm. The eluate was then collected as a set of fractions of 0.5 ml volume. In each fraction the antimony concentration was determined by GFAAS.

GFAAS determination of antimony. The modifier was applied as follows: the modifier solution was first injected, dried and pyrolysed (Table 1, Programme 1) to obtain the respective metal in its reduced form, then the sample was injected and analysed (Table 1, Programme 2).

The repetitive determination of antimony was carried out (Table 1, Programme 2) for the investigation of the stability of the modifier performance after a single injection and thermal reduction of 10 µg of Pd, Ir or Rh (Table 1, Programme 1).

Results and discussion

Determination of antimony in a standard solution

In order to compare the effectiveness of different modifiers, pyrolysis studies were performed for the determination of antimony in an aqueous standard solution. The pyrolysis temperature varied from 1200 to 1700 °C. When an aliquot containing 10 µg of Pd, Ir or Rh was introduced before each sample injection, antimony was stabilised up to 1350 °C in the presence of Pd, and up to 1500 °C in the presence of Rh or Ir (Fig. 1A). The absorbance value for 2 ng Sb(III) in the presence of Pd (pyrolysis temperature 1350 °C) was comparable with that when no modifier is used for pyrolysis at 200 °C. Although iridium and rhodium stabilise antimony to a higher temperature when compared with palladium, a significant decrease in absorbance value of 23% and 30%, respectively, was observed.

Fig. 1 Influence of the pyrolysis temperature (step 6 in Programme 2, Table 1) on the integrated absorbance of 2 ng of antimony in the presence of thermally reduced 10 µg of Pd, 10 µg of Ir or 10 µg of Rh: A, standard solution; and B, solution containing 4 g l⁻¹ albumin, 0.4 g l⁻¹ transferrin and 1 mol l⁻¹ NaCl.

Determination of antimony in eluent fractions

During HPLC separation, investigated proteins were eluted using a solution containing 0.01 mol l⁻¹ Tris-HCl and 1 mol l⁻¹ NaCl. A linear 4 min gradient of NaCl was used. It was important to evaluate whether the conditions described previously allow the interference-free determination of antimony in eluent fractions. The influence of the pyrolysis temperature on the integrated absorbance of 100 µg l⁻¹ antimony in the solutions of both proteins and the eluent mixture was investigated. From the results shown in Fig. 1B it could be concluded that reduced palladium stabilises antimony to a higher temperature in the presence of proteins compared with a standard solution. Both palladium and iridium stabilise antimony up to 1500 °C, which is, in the case of iridium, comparable to the temperature for a matrix-free standard solution (Fig. 1A). It should be pointed out that the absorbance for antimony in eluate solutions in the presence of iridium is higher by 33% when compared with a standard solution. This phenomenon could be explained by the formation, in the presence of a chloride containing matrix, of intercalation compounds of iridium with graphite.¹⁸

The presence of the organic matrix is known to be responsible for a high background absorbance. Indeed, when the pyrolysis temperature was below 1300 °C, background absorbance was above 1.5 absorbance units, which results in overcorrection effects (Fig. 2). Moreover, above 1400 °C, the background absorbance was found to be less then 0.01 integrated absorbance units.

Fig. 2 Effect of pyrolysis temperature on the atomic absorption of antimony and background signals in the presence of 10 µg of iridium.

Although the absorbance of antimony in the presence of rhodium in investigated standard solutions as well as eluent solutions is comparable when 1350 °C is used at the pyrolysis step, above this temperature the signal decreased rapidly. This means that in the presence of matrix components, rhodium is no longer efficient in antimony stabilisation.

Palladium, rhodium or iridium as a permanent modifier

Based on previous experience¹⁹ it was interesting to evaluate the permanent performance of the investigated modifiers. For this purpose, antimony was determined after a single injection of matrix solutions and reduction of palladium, rhodium or iridium. It was found that palladium and rhodium must always be introduced before each atomisation cycle as, for both modifiers, the absorbance value for the next firing decreased by about 90%. In the case of iridium, after a preliminary five firings, the modifier kept its performance even in the presence of a chloride-containing matrix for at least 20 atomisation cycles with an RSD = 6%. This was found to be sufficient for the determination of antimony in all fractions from one chromatographic cycle. An example of a chromatographic profile, used for the investigation of antimony speciation in serum samples, is shown in Fig. 3.

Fig. 3 The chromatogram of A, transferrin and B, albumin with a superimposed histogram of absorbance value for antimony.

Conclusion

The aim of this work was to find the best conditions for the determination of antimony in collected fractions of eluate after chromatographic separation of albumin and transferrin for speciation study.

When antimony was determined without any stabilising agents (modifiers) the maximum pyrolysis temperature cannot exceed 200 °C. This is unacceptable for the determination of antimony in eluent fractions containing proteins and sodium chloride, as a high background could be detected up to 1300 °C. All the investigated modifiers, Pd, Rh, and Ir (thermally reduced in the graphite furnace before sample injection), offer good stabilisation performance for antimony in an aqueous standard solution. In the presence of matrix components (NaCl + Tris-HCl buffer and proteins) rhodium loses its stabilising performance. Palladium and iridium proved to be good stabilising agents in matrix-containing solutions. Iridium offered the best performance with respect to stabilisation up to 1500 °C and the possibility of use for several firings.

acknowledgement

The financial support from BST 662/2/2000 grant is gratefully acknowledged.

References

1. M. B. de la Calle-Guntiñas, Y. Madrid and C. Cámara, Quality Assurance for Environmental Analysis, Elsevier, Amsterdam, 1995..
2. Y. A. Gawargious, N. B. Tadros, A. Besada and L. F. Ibrahim, Analyst, 1987, 112, 545.
3. M. O. Andreae, J-F. Asmodé, P. Foster and L. Van't dack, Anal. Chem., 1981, 53, 1766.
4. Y. Xiu-Ping, W. Van Mol and F. Adams, Analyst, 1996, 121, 1061.
5. R. Iffland, Handbook on Toxicity of Inorganic Compounds, ed. H. G. Seiler and H. Sigel, Marcel Dekker, New York, 1988..
6. K. L. Stemmer, Pharm. Ther. A, 1976, 1, 157.
7. E. Berman, Toxic Metals and Their Analysis, Heyden, London, 1980..
8. K. Kanematsu, M. Hara and T. Kada, Mutat. Res., 1980, 77, 109.
9. H. T. Delves, C. E. Sieniawska, G. S. Fell, T. D. B. Lyon, C. Dezateux, A. Cullen, S. Variend, J. R. Bonham and S. M. Chantler, Analyst, 1997, 122, 1323.
10. G. Schlemmer and B. Welz, Spectrochim. Acta, Part B, 1986, 41B, 1157.
11. B. Welz, G. Schlemmer and J. R. Mudakavi, J. Anal. At. Spectrom., 1992, 7, 499.
12. E. Bulska and K. Pyrzyńska, Spectrochim. Acta, Part B, 1997, 52, 1283.
13. H. Emteborg, E. Bulska, W. Frech and D. C. Baxter, J. Anal. At. Spectrom., 1992, 7, 405.
14. D. Styris, L. J. Prell, D. A. Redfield, J. A. Holcombe, D. A. Bass and V. Majidi, Anal. Chem., 1991, 63, 508.
15. E. Bulska, Z. Grobenski and G. Schlemmer, J. Anal. At. Spectrom., 1990, 5, 203.
16. S. Garboś, M. Wałcerz, E. Bulska and A. Hulanicki, Spectrochim. Acta, Part B, 1995, 50, 1669.
17. C. J. Rademeyer, B. Radziuk, N. Romanova, N. P. Skaugset, A. Skogstad and Y. Thomassen, J. Anal. At. Spectrom., 1995, 10, 49.
18. E. Bulska and H. M. Ortner, Spectrochim. Acta, Part B, 2000, 55, 491.
19. E. Bulska, M. Piaścik and K. Pyrzyńska, Anal. Lett., 2000, 33(7), 1399.

---