Introduction

Automotive catalytic converters are a primary source of emission of platinum, palladium and rhodium (PGE—platinum group elements) into the environment. Due to attrition of the catalyst surface during vehicle operation, PGE-containing particles are emitted with exhaust fumes at rates of several hundred ng km−1 per car. It has been proven that these particles accumulate in airborne and roadside dust, soil and vegetation and are transported to surface waters via road run-off [1–3]. A number of studies have been undertaken to develop reliable methods for the determination of platinum in environmental matrices. For this purpose, electrothermal atomic absorption spectrometry (ETAAS), inductively coupled plasma atomic emission spectrometry (ICP-AES), inductively coupled plasma mass spectrometry (ICP-MS) and adsorptive stripping voltammetry (AdSV) are mostly applied [4].

Only a few examples of chemiluminescence methods of Pt(IV) determination have been reported in the literature [5–7]. These methods are characterized by poor precision (RSD ≤ 10%), because the measurements were performed in a batch mode. A chemiluminescence flow injection (CL-FIA) method of Pt(IV) determination has been developed in our laboratory [8]. It is based on the chemiluminescence reaction of luminol oxidation in aqueous alkaline medium. Luminol is oxidized to yield electronic excited state 3-aminophthalate, which emits the radiation at 425 nm on falling to the ground state [9]. Chlorocomplexes of Pt(IV) acting as a catalyst of this reaction can be determined at very low concentration levels (ng mL−1) [8]. High sensitivity and precision, and inexpensive instrumentation are the main advantages of this method. However, the accuracy of CL determination of platinum in environmental samples is strongly affected by matrix components (other metals and organic compounds). In order to avoid matrix interferences, application of procedure of analyte separation is necessary. It has been demonstrated that biosorbents are very effective materials for the separation and preconcentration of metal ions in solid phase extraction procedures [10]. Biosorbents are prepared from biologically inactive microorganisms, including algae, fungi, yeast and bacteria [e.g. 11, 12]. In order to obtain particles with good mechanical properties, biomass is immobilized in/on natural or synthetic support. The efficiency of metal biosorption by microorganisms mainly depends on the chemical structure of the cell wall. The biomass cell wall is a complex network of various binding sites with different affinities for metal ions [13]. Other important parameters influencing the biosorption process are sample pH, ionic strength and presence of other metal ions competing for functional groups. The mechanism of metal biosorption can be a combination of physical adsorption, ion exchange, complexation and microprecipitation [14]. In our previous studies algae Chlorella vulgaris immobilized on Cellex-T resin support and yeast Saccharomyces cerevisiae immobilized in calcium alginate beads have been applied to the preconcentration and separation of Pt(IV) before its determination by luminol-based CL method [15, 16]. However, the procedures may be used only for the determination of platinum in simple matrices, such as natural waters. Therefore, our further studies focused on searching more selective biosorbent. Filamentous fungi (Penicillium sp., Aspergillus sp., Rhizopus sp.) are ubiquitous in natural environment and important in a variety of industrial processes. The cell wall of the fungi is a thick, rigid structure composed of complex layers of polysaccharides (80–90%), proteins, lipids, and polyphosphates. The most common constituent of the wall is chitin, consisting of N-acetylglucosamine residues [13]. Recently fungi Aspergillus sp. immobilized in calcium alginate and on Cellex-T resin have been applied to the separation and preconcentration of Pt and Pd before their determination in environmental samples by ETAAS [17].

We describe here a method using Aspergillus sp. immobilized on a cellulosic resin as a biosorbent for the separation of Pt(IV) ions prior to their chemiluminescent determination via flow injection analysis. The CL method is based on monitoring the light intensity generated in the reaction of luminol oxidation catalysed by Pt(IV) ions. Our paper also covers the comparison of immobilized fungi with biosorbents studied earlier, namely algae C. vulgaris immobilized on Cellex-T resin [15] and yeast S. cerevisiae immobilized in calcium alginate beads [16] in terms of their biosorption performance and applicability for platinum separation from environmental samples prior to its determination by CL method.

Experimental

Reagents and materials

A standard solution of platinum as hexachloroplatinic(IV) acid (30%) (POCh, Poland, www.poch.com.pl) was used. Stock solution of Pt(IV) (1,000 μg g−1) was prepared in 1 mol L−1 hydrochloric acid (Trace Select, Fluka, France, www.sigmaaldrich.com). Working standard solutions were prepared daily by diluting the stock solution with HCl solution of pH 1.0.

Luminol (≥98,0%, Fluka, Russia, www.sigmaaldrich.com) stock solution (2.5⋅10−2 mol L−1) was prepared by dissolving an appropriate amount of the compound in 1 mol L−1 NaOH solution and stored in a refrigerator (at 4 °C) for 24 h prior to use, to ensure the stability of the measurements. Working solution (5.0⋅10−3 mol L−1 luminol in 0.2 mol L−1 NaOH) was prepared before use by dilution of the stock solution with Milli-Q water. The analytical grade reagents: sodium hydroxide (POCh, Poland, www.poch.com.pl) and sodium chloride (Sigma-Aldrich, Denmark, www.sigmaaldrich.com) were also used.

A growth medium Czapek Dox Agar (Fluka, Switzerland, www.sigmaaldrich.com) containing sucrose (30 g L−1), agar (15 g L−1), NaNO3 (3 g L−1), K2HPO4 (1 g L−1), KCl (0.5 g L−1), MgSO4·7H2O (0.5 g L−1), FeSO4·7H2O (0.01 g L−1) was used for the cultivation of fungi Aspergillus sp.

Yeast extract (Fluka, Switzerland, www.sigmaaldrich.com)—an extract of autolyzed yeast cells (a mixture of amino acids, peptides, water soluble vitamins and carbohydrates) was used as additive for a culture medium.

Cellex-T resin (Bio-Rad Laboratories, USA, www.bio-rad.com), cellulose anion-exchanger (triethylaminocellulose) was used as a support material for the immobilization of fungi Aspergillus sp.

Reversed-phase C18 bonded silica gel (LiChroprep RP-18, 25–40 μm particle size, Merck, Germany, www.merck-chemicals.com) was used.

The environmental samples analyzed in this study were: water from the Biała river (Białystok, Poland), wastewater collected from municipal sewage treatment plant in Białystok and road run-off taken from the retention reservoir at the ring road of Białystok.

Instrumentation

A CL-FIA manifold used in this work was the same as in our previous study [16], except that the column was filled with fungi Aspergillus sp. immobilized on Cellex-T. Briefly, the flow injection manifold was composed of two peristaltic pumps (Gilson Minipuls, France, www.gilson.com); a rotary injection valve (Model 5021 Rheodyne, USA, www.chromtech.com); laboratory made glassy column (30 mm length, 6 mm i.d.) packed with the biosorbent and installed in the injection valve; flow luminometer (KSP, Poland).

A PU 9100X (Philips Scientific, UK) atomic absorption spectrometer equipped with PU 9390X electrothermal atomizer, FS-90 autosampler, deuterium background correction system and platinum hollow cathode lamp (Photron, Australia, www.photron.com.au) was used. The measuring conditions for Pt determination were described elsewhere [18].

For pH measurements, pHmeter CP-315M (Elmetron, Poland, www.elmetron.com.pl) was used.

A solid phase extraction unit Spe-12G (J.T. Baker, Germany, www.witko.com.pl) equipped with a glassy column filled with LiChroprep RP-18 was used to remove the interfering organic matrix from real samples.

Procedures of cultivation and immobilization of Aspergillus sp

The medium used for the cultivation of fungi Aspergillus sp. was prepared by dissolving 7.5 g of Czapek Dox Agar and 0.58 g of yeast extract in 150 mL of warm Milli-Q water. The growth medium was sterilized by autoclaving at 121 °C and poured onto Petri plates. Aspergillus sp. was inoculated on solidified medium and incubated at 30 °C for 3 days. Then the biomass was separated from the growth medium, washed with 0.1 mol L−1 HCl and Milli-Q water and next was immobilized on Cellex-T support. The biomass was added to the resin (mass ratio 1:3), wetted with Milli-Q water, thoroughly mixed and then dried at 60 °C. The procedure (wetting, mixing and drying) was repeated twice in order to maximize the contact between the biomass and support surface.

Procedure of platinum preconcentration on biosorbent and its CL determination

Before analysis, a column filled with immobilized fungi (0.15 g) was preconditioned by passing 0.25 mol L−1 HCl solution (5 mL) and next Milli-Q water (4 mL) at a flow rate of 2.0 mL min−1. The Pt(IV) standard or sample solution (10 mL) of pH 1.0 was loaded onto the column at a flow rate of 2.0 mL min−1 for the analyte retention and the effluent was directed to wastes. After that, the column was washed with 4 mL of Milli-Q water. The next stage in the procedure involved the elution of the bound platinum from the biosorbent. It was accomplished by passing 5.0 mol L−1 NaCl solution through the column at a flow rate of 2.2 mL min−1. The eluent was merged with the reagent solution (5.0⋅10−3 mol L−1 luminol in 0.2 mol L−1 NaOH) pumped at a flow rate of 2.2 mL min−1 and the resulting solution was directed to the flow cell. Increase in the chemiluminescence intensity relative to the baseline, corresponding to concentration of Pt(IV) ions, was measured with the CL detector. The average lifetime of the column was around 50 analytical cycles.

Pretreatment of environmental samples

Samples of river water and road run-off were spiked with Pt (0.1; 0.25; 1.0 ng mL−1). After 2 h equilibration, samples were filtered through Iso-Disc Filters PTFE (25 mm × 0.45 μm, Supelco, USA, www.sigmaaldrich.com) and adjusted to the required pH with hydrochloric acid.

Samples of wastewater were filtered through cellulose filters with soft and wide pores (Filtrak 388, Germany), acidified to pH 1 with HCl and stored in a refrigerator. Before analysis samples were spiked with Pt(IV) standard (0.25; 1.0 ng mL−1).

The interfering compounds present in environmental samples were retained on reversed phase C18 column packed with 500 mg of LiChroprep RP-18. The sorbent was conditioned by passing 5 mL of methanol and 5 mL of 0.01 mol L−1 hydrochloric acid. The sample solution was passed through the column at a flow rate of 1.0 mL min−1 and the effluent was collected and loaded onto the column filled with the biosorbent.

Results and discussion

Optimization of platinum preconcentration conditions on Aspergillus sp. immobilized on cellex-T

In order to determine the suitability of the biosorbent for the preconcentration of Pt(IV) ions, the following parameters were optimized: the sample pH and the rate of passing the sample solution through the column. The influence of the pH on biosorption efficiency of platinum was studied in the range from 0.25 to 4.0. The Pt(IV) solutions were adjusted to the required pH with hydrochloric acid and loaded onto the column packed with immobilized fungi. The collected effluents were analyzed for the remaining metal by ETAAS. It was observed that efficiency of platinum biosorption was almost constant (97.7–99.2%) in the studied pH range. In further experiments solutions of Pt(IV) were adjusted with HCl to pH 1.0. In such solutions platinum is present in the form of anionic chlorocomplexes (mainly PtCl 2-6 and PtCl5(H2O)-) [19] and can interact with the protonated active sites of the cell wall of investigated microorganism. Literature survey shows that the optimal pH for the retention of other metal ions such as Cu(II), Pb(II), Zn(II), Fe(III), Ni(II), Co(II) and Cd(II) on fungi Aspergillus (A. fumigatus immobilized on Diaion HP-2MG resin [20], A. niger immobilized on silica gel [21], A. niger immobilized on sepiolite [22]) is in the range 6–8. Hence, strongly acidic medium (pH 1.0) should provide a good selectivity of platinum biosorption process. In the case of algae C. vulgaris immobilized on Cellex-T resin platinum was the most efficiently retained in the pH range from 1.5 to 1.8 [23]. For yeast S. cerevisiae immobilized in calcium alginate the highest biosorption efficiency was observed in a wide range of pH from 1.5 to 6.5 [16]. At the optimal pH, immobilized fungi demonstrated higher biosorption efficiency of Pt(IV) chlorocomplexes (99.2%) compared to immobilized algae (95.2%) [23] and yeast (90.3%) [16]. Variations in biosorption efficiency can be attributed to differences in composition and structure of cell walls of tested microorganisms and in consequence to differences in biosorption mechanisms. Also the effect of the type of support used for the immobilization of biomass should not be neglected.

Subsequent studies showed that the flow rate of Pt(IV) solutions (in the range from 0.6 to 2.2 mL min−1) practically does not affect the efficiency of biosorption (98.1–99.2%). In order to shorten the analysis time, the flow rate of 2.0 mL min−1 was chosen for further experiments. The same flow rate was used in the case of algae column [15], but much lower (0.5 mL min−1) in the case of yeast column [16]. This indicates that immobilized fungi and algae are characterized by faster kinetics of biosorption than immobilized yeast.

Our previous study [15] demonstrated that quantitative elution of Pt(IV) from algae C. vulgaris immobilized on Cellex-T resin can be obtained by using sodium chloride solution. Additional advantage of this solution is that it shows a positive influence on analytical signal of Pt(IV) measured by luminol-based CL method. The present study revealed that sodium chloride solution can also be applied for desorbing of platinum from fungi Aspergillus sp. immobilized on Cellex-T. Therefore, NaCl solution was used as an eluent/carrier stream in the CL-FIA manifold equipped with fungi column and the effect of its concentration (in the range 3.0–5.5 mol L−1) on platinum signals was investigated (Fig. 1a). A three-fold increase in CL signals of platinum was observed when the NaCl concentration was raised from 3.0 to 5.0 mol L−1. To ensure maximum sensitivity of measurements, the solution of 5.0 mol L−1 NaCl was selected for further experiments. In our previous studies [15, 16] the highest CL signals of Pt(IV) were obtained for lower concentration of sodium chloride—3.0 mol L−1.

Fig. 1
figure 1

Effect of elution conditions on CL signal of Pt(IV) (1 ng mL−1): a NaCl concentration (CLum = 5.0⋅10−3 mol L−1, CNaOH = 0.2 mol L−1, flow rate = 2.2 mL min−1); b flow rate of eluent and luminol streams (CNaCl = 5.0 mol L−1, CLum = 5.0⋅10−3 mol L−1, CNaOH = 0.2 mol L−1)

Full size image

As can be seen from Fig. 1b, analytical signals of platinum increased with increasing the flow rate of eluent and luminol streams in the range 0.9–2.5 mL min−1. The flow rate equal to 2.2 mL min−1 was chosen for subsequent experiments. At higher values considerable resistance in the flow of eluent through the column was observed. The same flow rate was used in the case of algae [15] and yeast column [16]. The optimal conditions of platinum extraction procedures on tested biosorbents are given in Table 1.

Table 1 Optimal conditions of extraction procedures of Pt(IV) on biosorbents [this work, 15, 16]
Full size table

The recovery of platinum from immobilized fungi obtained at optimal values of studied parameters was 92.0 ± 8.8% (n = 3). It was slightly lower than that obtained for the preconcentration procedure of platinum on algae column (98.5%) [15], but greater than that obtained in the case of yeast column (83.0%) [16]. The complete recoveries of Pt from fungi and algae achieved with concentrated NaCl solution indicate that electrostatic interactions are responsible for binding of Pt(IV) chlorocomplexes to the biosorbents. The lower recovery from immobilized S. cerevisiae suggests that in this case, electrostatic forces are not the only kind of interactions involved in the metal binding. It has been reported by Mack et al. [24] that biosorption of Pt(IV) chlorocomplexes by immobilized yeast from solutions of pH 1.5 involves chemical sorption mechanism such as covalent bonding.

The same concentrations of luminol and NaOH (5.0⋅10−3 mol L−1 luminol in 0.2 mol L−1 NaOH) as in the earlier studies [15, 16] were applied.

Effect of the interfering ions

The interfering effect of other metal ions commonly present in environmental samples on the determination of Pt(IV) by the CL-FIA method was studied. For this purpose, solutions of Pt(IV) ions (1.0 ng mL−1) spiked with different concentrations of matrix ions (as listed in Table 2) were subjected to the separation procedure on Aspergillus sp. immobilized on Cellex-T. For comparison, the influence of metal ions on luminol-based CL determination of platinum after its separation on algae and yeast column was also presented. As can be seen from Table 2, among the tested biosorbents the immobilized fungi showed the greatest ability to the selective retention of Pt(IV) in the presence of other metal ions. The acceptable excesses of foreign ions to analyte were the highest (in the range 50–80 000-fold), except for Cr(VI) and Cu(II). It is worth noting that the studied biosorbent allows the efficient separation of Pt(IV) from other ions belonging to PGE, such as Pd(II), Rh(III) and Pt(II).

Table 2 Acceptable excess of interfering ions to analyte in determination of Pt(IV) by the CL-FIA methods with separation step on biosorbents (relative error of determination ± 5%) [this work, 15, 16]
Full size table

Analytical characteristic

The analytical characteristic of the CL method of Pt(IV) determination after on-line analyte preconcentration on fungi Aspergillus sp. immobilized on Cellex-T is presented in Table 3. The analytical parameters obtained in the flow systems with algae and yeast column are showed for comparison. The method is characterized by the lowest limit of detection (expressed as the concentration of the analyte equal to 3.3-fold standard deviation of the intercept or residual standard deviation divided by the slope of the calibration graph, LOD = 0.02 ng mL−1 Pt) and the best sensitivity (the largest slope of the calibration graph). After the biosorption step on immobilized fungi, the sensitivity of CL measurements increased 2-fold and the limit of detection of Pt(IV) decreased 5-fold compared to the direct measurements (without separation step LOD = 0.1 ng mL−1 Pt). The LOD of the CL method is significantly lower than that obtained by ICP-AES after preconcentration of Pt(IV) on activated carbon modified with ethyl-3-(2-aminoethylamino)-2-chlorobut-2-enoate (9 ng mL−1) [25] and activated carbon modified with 2,6-diaminopyridine (0.29 ng mL−1) [26]. The obtained LOD is comparable with that obtained by ICP-MS after solid-phase extraction of analyte complexes on modified silica gel Separon SGX C18 (0.05 ng mL−1 Pt) [27] and anion-exchange sorbent Dowex AG1-X8 (0.15 ng g−1) [28] and the same as LOD of ETAAS method with biosorption of platinum on Aspergillus fungi [17]. The calibration graph obtained for standard solutions of Pt(IV) was linear in the concentration range 0.05–2 ng mL−1. The precision of the method, expressed as the relative standard deviation (RSD) of 12 subsequent measurements of 0.25 ng mL−1 concentration of Pt(IV), was equal to 2.6%.

Table 3 Analytical characteristic of the CL-FIA methods of determination of Pt(IV) after its preconcentration/separation on biosorbents [this work, 15, 16]
Full size table

Application of biosorbent for the separation of platinum from environmental samples

The samples of river water were spiked with Pt(IV) and analyzed by the CL-FIA method. The approach applied for the removal of organic matrix, which interfere with luminol-based CL determination of platinum, was the same as described previously [16]: samples were passed through a column packed with LiChroprep RP-18 sorbent. The recoveries of Pt determined by the modified procedure, including sample pre-cleaning step on C18 column followed by biosorption of the analyte on fungi column, are shown in Table 4. Selectivity of the procedure ensures effective separation of interfering matrix and accurate CL determination of Pt(IV) in complex samples, such as road run-off waters and municipal wastewater. This confirms the higher selectivity of platinum biosorption process on fungi Aspergillus sp. compared to algae and yeast. The previous studies [15, 16] showed that the extraction procedures using algae and yeast column can be applied only for the separation of Pt(IV) from simple matrices, such as river waters.

Table 4 Recovery of Pt(IV) from real samples after removal of organic matrix on RP-18 column and analyte preconcentration on fungi Aspergillus sp. immobilized on Cellex-T
Full size table

Conclusions

A procedure for the preconcentration and separation of Pt(IV) ions on fungi Aspergillus sp. immobilized on Cellex-T resin prior to their luminol-based CL determination has been developed. Under the optimal conditions of biosorption and elution steps quantitative recovery of platinum was achieved. After the extraction procedure, the sensitivity of CL measurements increased 2-fold and the limit of detection of Pt(IV) decreased 5-fold compared to the direct measurements. For that reason, the method may be applied to the determination of trace amounts of Pt(IV) ions. Due to high selectivity of the used biosorbent, ions of transition metals present in environmental samples, such as river water, road run-off and municipal wastewater, do not interfere with the CL determination of Pt(IV) ions. The dissolved organic matter affecting the chemiluminescence of luminol was removed from the samples by means of C18 column. The biosorption performance of the immobilized fungi was compared with that of the biosorbents studied earlier: algae C. vulgaris immobilized on Cellex-T resin [15] and yeast S. cerevisiae immobilized in calcium alginate beads [16]. Among the three biomaterials tested, the immobilized fungi proved to be the best biosorbent for Pt(IV), as they exhibit the highest selectivity and, unlike the other two biosorbents, can be used to separate platinum from complex matrix of wastewater samples.

Immobilized microorganisms are useful solid sorbents for effective separation/preconcentration of trace metals before their chemiluminescent determination. With proper choice of the microorganism and immobilization technique, biosorbents could satisfactory replace other solid phase materials. The main advantages of using biosorbents are the effectiveness in removing of metal ions at very low concentration levels, the availability of microorganisms, ease of production, and low cost.