Flame atomic absorption spectrometry determination of trace amount of gold after separation and preconcentration onto ion-exchange polyethylenimine coated on Al₂O₃

2.1 Apparatus

A Varian model SpectrAA 220 atomic absorption spectrometer (Victoria, Australia, http://www.varianinc.com) was used for measuring of Au(III) in air-acetylene flame. The instrumental parameters according to manufacturer’s recommendations are as follows; Wavelength, 242.8 nm; slit width, 1.0 nm; current lamp 4.0 mA; air flow 3.5 L min⁻¹ and acetylene flow 1.5 L min⁻¹. Thermo Finnigan Flash EA1112 Microanalyzer was used for the determination of C.H.N percentage. A Metrohm 713 pH meter (http://www.metrohm.de) was used for pH measurement and equipped with a combined glass calomel electrode. Thermogravimetric analysis (TGA) was accomplished on a Perkin–Elmer TGA-USA.

2.2 Reagents

All reagents were of analytical grades. The stock solution (1000.0 ng mL⁻¹) of Au(III) was prepared by dissolving appropriate amount of gold chloride (Bombay–New Dehli, Central Drug House) in HNO₃ 1:1 and dilute with distilled water to total volume of 50 mL. A 5% solution of polyethylenimine (PEI, molecular weight: 600,000–1,000,000) (Fluka) was prepared in distilled water. The sorbent was Al₂O₃ produced by Fluka. A solution 0.5 M thiourea (Merck) was prepared by dissolving thiourea in distilled water. Analytical grade salts of various metal ions (all from Merck) were of the highest purity available. Buffer solution with pH 5.7 was prepared from 0.1 M CH₃COOH and CH₃COONa.

2.3 Preparation of the sorbent

The coating PEI on alumina (giving PEI/Al₂O₃) was carried out as follows:

1.0 g alumina and 0.5 g NaNO₃, with 5 mL PEI 5% were mixed. Then pH was adjusted to 7.2. The obtained suspension was shaken for 24 h. The sorbent that was made, then filtered, rinsed with distilled water several times, and finally dried in desiccators. It was characterized with elemental analysis (C.H.N) and TGA. The thermogravimetric analysis curve of the PEI/alumina sorbent shows a four step mass loss up to 721 °C. The 1.7% mass loss up to 88.6 °C in the first step is due to sorbed water. In the second step mass loss is 3.9% up to 296.4 °C. In the third step mass loss is 3.9%, also, up to 506.5 °C and in the fourth step mass loss is 1.53% up to 723.1 °C. The mass losses in the second, third and fourth steps are corresponding with polyethylenimine. The results of elemental analysis of the dried sorbent showed that on new sorbent N and C elements have existed, thus PEI was coated on alumina (Table 1).

2.4 General procedure

The columns were packed with 50 mg of sorbent (PEI/Al₂O₃) and first was washed with an aliquot of water to wet the surface of the sorbent, then conditioned with a buffer solution at pH 5.7. Then, an aliquot of the sample solution containing of Au(III) (0.5–50.0 μg) was taken in a 50 mL beaker and 3 mL buffer solution was added with pH 5.7. The total volume of the solution was reached to about 30 mL with distilled water. This solution was passed through the column (funnel-tiped glass tube) with flow rate of 2 mL min⁻¹. With this procedure, Au³⁺ ions in water samples were quantitatively adsorbed on the new sorbent and consequently were eluted by 3.0 mL of 0.5 M thiourea and then 2.0 mL of 1.0 M HCl (final volume 5.0 mL) at a flow rate of 4 mL min⁻¹. The final solution was aspirated directly in to the flame AAS against the blank prepared in the same manner without the addition of gold.

3.1 Sorption mechanisms

Polyethylenimine (PEI) is well known for its metal chelation property due to the presence of a large number of amine groups in a molecule, and is often used to modify the adsorbent surface to increase the adsorption capacity (Ghoul et al., 2003; Deng and Ting, 2005; Sabermahani and Taher, 2006). It is known to exist as a linear structure or a branched structure (Fig. 1).

Close

Figure 1

Structure of poly(ethyleneimine) (PEI).

Export to PPT

Commercial branched PEI was employed in this work; it contains primary, secondary and tertiary amino groups in a ratio of approximately 1/4, 1/2 and 1/4, respectively (Amara and Kerdjoudj, 2003). Low-molecular-weight substances can be bound to macromolecules by intermolecular forces, mainly a complex or ionic bond or a combination of both (Molinari et al., 2004). Coating alumina with PEI is a cheap and simple method (one step). By elemental analysis it was observed that prepared sorbent in the neutral media has more capacity than the prepared sorbent in acidic or alkaline media. The experiments showed that the increase in ionic strength causes an increase in the amount of polymer adsorbed on to Al₂O₃ at all measured pH values (Sabermahani and Taher, 2007) NaNO₃ is a salt that causes increasing ionic strength. Various amounts of NaNO₃ were added to alumina and sorption capacity was measured. The optimum value was determined to be 0.5 g NaNO₃ for 1 g alumina.

For gold, by contrast, +3 is the element’s best known oxidation state. Au (III) is trend to the oxygen, especially to sulfur and nitrogen. Between various complexes of gold, the square-planar ion [AuCl₄]⁻ is well-known. Other square-planar ions of the type [AuX₄]⁻ can be derived in which X = F, Br, I, CN, SCN and NO₃. Numerous cationic complexes have been prepared with amines, both unidentate (e.g. pyridine, quinoline, as well as NH₃) and chelating (e.g. ethylenediamine, bipyridine, phenanthroline) (Greenwood and Earnshaw, 1997). A yellow colored complex has been formed between Au³⁺ and PEI. The eluent also has sulfur and nitrogen groups and Au is complexed with Cl⁻ in HCl and formed [AuCl₄]⁻. In the presence of PEI on alumina the recovery of gold was almost 100%, but with alumina alone, the recovery of Au(III) was 29%.

3.2 Choice of eluent

In order to achieve the maximum recovery for the retained Au(III) ions, by PEI/alumina, various eluents were tested (Table 2). As can be seen from the tests, the best eluent is 3.0 mL of 0.5 M thiourea and then 2.0 mL of 1.0 M HCl.

3.3 Effect of pH

The influence of the pH on the strength of the complexes between Au³⁺ and PEI on alumina was investigated. In order to optimize this parameter, the effect of pH on the recovery of gold was studied in the range 2.0–12 (Fig. 2). Gold was quantitatively adsorbed in the range of 5.0–8.5. That is almost a wide range. Under pH 5, a majority of amine sites are protonated and can not take part in the chelation process. However, decrease in Au³⁺ recovery in solution of pH >8.5 is due to the hydrolysis of Au(III). But in the mentioned range a proton of amino groups is in a neutral form, leading to a donor–acceptor interaction with gold ions.

Close

Figure 2

Effect of pH on recovery of gold. Conditions: Au, 20.0 μg; sample flow rate, 2 mL min⁻¹; eluent flow rate, 4 mL min⁻¹; sorbent, 50 mg; final solution, 3.0 mL thiourea 0.5 M then 2.0 mL HCl 1.0 M, instrument settings were as follows; wavelength, 242.8 nm; slit width, 1.0 nm; current lamp 4.0 mA; air flow 3.58 L min⁻¹ and acetylene flow 1.5 L min⁻¹.

Export to PPT

3.4 Effect of flow rates and sample volume

The effect of flow rates of sample and eluent solutions on recovery of Au³⁺ ion was studied. The flow rate of sample solution was examined in the range 0.15–5.0 mL min⁻¹ and for the eluent flow rate, the range 0.15-8.0 mL min⁻¹ was tested. In the range of 0.15-2.0 mL min⁻¹ for the sample solution and 0.15–4.0 mL min⁻¹ for the eluent, the flow rate has no significant effect on the recovery of gold. Consequently, all subsequent experiments were performed at 2 mL min⁻¹ and 4 mL min⁻¹ for flow rates of the sample and eluent, respectively.

To determine the maximum applicable sample volume, solutions containing a fixed amount of Au³⁺ ions (20.0 μg), were varied in the range of 100–3000 mL under the optimum conditions and was passed through a column (Fig. 3). It was observed that the recovery of the analyte was constant up to 2300 mL (break through volume). By using 5.0 mL of the eluent solution, the preconcentration factor was obtained at 460 and it was found to be an excellent factor in contrast to other works for gold.

Close

Figure 3

Effect of sample volume on the recovery of gold. Conditions and instrument settings were the same as Fig. 2.

Export to PPT

The maximum capacity of the PEI/alumina sorbent was found 6.0 mg gold per sorbent g. The sorption capacity was determined by the batch process, in the following way: shaking the mixture of 1 g sorbent with various amount of Au³⁺ (1–8 mg), for about 2 h at room temperature.

3.5 Effect of diverse ions

Various salts and metal ions were added to the solution containing 20.0 μg of Au(III) individually. For most of them the tolerance limit was high. An advantage of this sorbent for gold was defined as freedom from high amount of diverse ions (Table 3). It is probably because of this fact that gold is almost a soft cation and the competition between hard cations (included alkaline and alkalin-earth metal) is weak.

3.6 Calibration and sensitivity

Calibration curves for the determination of Au(III) according to optimized conditions were obtained in the linear range of 0.1–10.0 μg mL⁻¹. Calibration curve equation A = 0.0357c (ppm) + 0.0024 with correlation factor of 0.995 was found. The limit of detection (LOD) was calculated as three times the standard deviation (3σ) of 10 replicate measurements of blank sample with the preconcentration step. The detection limit was calculated by dividing the instrumental detection limit by the preconcentration factor. Therefore, the detection limit was obtained 26.2 ng L⁻¹ in original solution. Sensitivity for 1% absorbance of gold(III) was obtained at 56.0 μg L⁻¹.

The repeatability of gold ion with PEI/alumina, was also investigated at optimum condition for 10 measurements of 4.0 μg mL⁻¹ of gold. Relative standard deviation was obtained 1.46% (n = 10).

3.7 Analysis of gold in water samples and standard alloys

The applicability of the proposed method was investigated in real samples with different matrices. The result, as the average of four separate determinations is shown in Table 4. Furthermore, standard solution containing Au³⁺ (10.0 μg) was added to the alloy (NKK, No. 1021); the results are also summarized in Table 4. As the table shows, in all cases, the gold recovery is almost quantitative.

3.8 Comparison to other methods

A comparison of the proposed system with other preconcentration procedures using several sorbents is given in Table 5. Preconcentration factor obtained is 460. Thus, it is higher than the presented results reported in the literature. Sorption capacity of the PEI/Al₂O₃ sorbent is 6 mg per sorbent g. The results obtained shows that the proposed method can be applicable for the separation of trace gold ions in a variety of water samples.