Preconcentration of trace amounts of lead in water samples with cetyltrimethylammonium bromide coated magnetite nanoparticles and its determination by flame atomic absorption spectrometry

2.1 Instrumentation

A Varian (Mulgrave, Victoria, Australia) flame atomic absorption spectrometer model SpectrAA 110, with an air-acetylene flame was used for lead determinations. Absorbance measurements were carried out at 217.0 nm using lead hollow cathode lamp (from Varian) operated at 4.0 mA with deuterium background correction (at 10 mA). The instrumental parameters were used according to the manufacturer’s recommendations. All measurements were based on peak height. The pH of the solutions were measured with a WTW pH meter (Inolab, Germany) which was supplied with a combined pH electrode. TEM images were obtained with an H-800 transmission electron microscope (Hitachi, Japan) and magnetic properties were analyzed using a vibrating sample magnetometer (VSM, LDJ9600).

2.2 Reagents

All chemicals used were of analytical reagent grade. Ferric chloride (FeCl₃·6H₂O), ferrous chloride (FeCl₂·4H₂O), sodium hydroxide, 1-(2-pyridilazo)-2-naphthol (PAN), lead nitrate (Pb(NO₃)₂), CTAB, methanol, acetone, and 1-propanol, hydrochloric acid were all purchased from Merck (Darmstadt, Germany). The stock solution of lead (1000 μg mL⁻¹) was prepared by dissolving the appropriate amounts of Pb(NO₃)₂ in doubly distilled water. A solution of 1 × 10⁻³ mol L⁻¹ PAN was prepared by dissolving the appropriate amount of this reagent in methanol.

2.3 Synthesis of Fe₃O₄ NPs

Fe₃O₄ NPs were prepared by chemical co-precipitation method via a reactor which was designed in our previous work (Faraji et al., 2010b). Briefly, 10.4 g of FeCl₃·6H₂O, 4.0 g of FeCl₂·4H₂O and 1.7 mL of HCl (12 mol L⁻¹) were dissolved in 50 mL of deionized water in order to prepare stock solution of ferrous and ferric chloride in a beaker which was then degassed with nitrogen gas for 20 min. Simultaneously, 500 mL of 1.5 mol L⁻¹ NaOH solution was degassed (for 15 min) and heated to 80 °C in a reactor. Then, the stock solution was added dropwise using the dropping funnel during 30 min under nitrogen gas protection and vigorous stirring (1000 rpm) by a glassware stirrer. During the whole process, the solution temperature was maintained at 80 °C and nitrogen gas was purged to remove the dissolved oxygen. After completion of the reaction, the obtained Fe₃O₄ NPs were separated from the reaction medium by the magnetic field, and then washed with 500 mL doubly distilled water four times. Finally, the obtained Fe₃O₄ NPs were resuspended in 500 mL of degassed deionized water.

2.4 Preconcentration procedure

An aliquot of water sample (500 mL, pH = 9.0) was transferred to 2000 mL glassware beakers. Then, 2.0 mL of 1 × 10⁻³ mol L⁻¹ PAN, 10 mL of the suspension of Fe₃O₄ NPs (10 mg mL⁻¹) and 2.5 mL of the CTAB solution (10 mg mL⁻¹) were sequentially added to the sample solution and then mixed. The mixture was shaken and allowed to complete the extraction process for 2 min. Subsequently, an Nd-Fe-B strong magnet (10 × 5 × 4 cm, 1.4 Tesla) was placed at the bottom of the beaker, and the CTAB-coated Fe₃O₄ NPs were isolated from the solution. After about 10 min, the solutions became limpid and supernatant solutions were decanted. Finally, the preconcentrated lead complex was eluted from the isolated particles with 1.0 mL of 0.5% HCl in methanol and this solution was introduced into FAAS for determination of lead concentration.

3.1 Characterization of Fe₃O₄ nanoparticles

Characterization of synthesized Fe₃O₄ NPs was done using IR, VSM and TEM instruments (Fig. 1). IR spectra of pure Fe₃O₄ NPs (i) and CTAB-coated Fe₃O₄ NPs (ii) are shown in Fig. 1a. The peak at ∼3455 cm⁻¹ is attributed to the stretching vibrations of –OH, which is assigned to surface OH groups of Fe₃O₄ NPs. The peak at ∼571 cm⁻¹ is attributed to the Fe–O band vibration of Fe₃O₄. Also, the peak at ∼1380 cm⁻¹ is attributed to C–N band and the peaks at 2885 and 2925 cm⁻¹ are attributed to two different C–H bands vibration of CTAB. The IR spectra show that the surface of Fe₃O₄ NPs was well modified by CTAB. Also, the TEM image of the prepared NPs is shown in Fig. 1b. Based on the TEM image, the Fe₃O₄ surface morphology analysis demonstrated the agglomeration of many ultrafine particles with a diameter of about 20 nm. On the other hand, it is important that Fe₃O₄ NPs should possess sufficient magnetic and superparamagnetic property for magnetic carrier technology (MCT) practical application. Fig. 1c shows VSM magnetization curves of Fe₃O₄ NPs at room temperature. Fe₃O₄ NPs exhibited typical superparamagnetic behavior due to not exhibiting hysteresis, remanence and coercivity. The large saturation magnetization of Fe₃O₄ NPs was 76 emu/g, which is sufficient for magnetic separation with a conventional magnet.

Close

Figure 1

Characterization of the CTAB-coated Fe₃O₄ NPs. (a) The IR spectra before (i) and after the modification with CTAB (ii). (b) The TEM image of the Fe₃O₄ NPs. (c) VSM magnetization curves of Fe₃O₄ NPs.

Export to PPT

3.2 Effect of solution’s pH

The charge density of mineral oxide surface is a main factor affecting the adsorption of analytes and it varies strongly with pH. Thus, pH is a very important parameter for the adsorption of target compounds. On the other hand, pH plays an important role in the complexation of lead with PAN. To examine the effect of pH on extraction efficiency of lead ions, pH of the solution varied in the range of 6.0–12.0 (Fig. 2). As can be seen from Fig. 2, the Fe₃O₄ NPs do not exhibit obvious adsorption of lead as Pb–PAN complex when the pH value was lower than 8.0. With increasing pH, the adsorption efficiency increased dramatically and reached maximum when the pH was between 9.0 and 10.0. This is mainly due to the fact that negative charges were favorable for the adsorption of cationic surfactants. When pH value was around NPs isoelectric point (about 7.0) the charge density of Fe₃O₄ NPs surface was very low (Zhao et al., 2008b). When the pH value was above their isoelectric point, the Fe₃O₄ NPs surface became negatively charged which gave rise to the strong electrostatic attraction between the CTAB molecules and the charged Fe₃O₄ NPs surface. Under this circumstance, hydrophobic complexes of Pb–PAN can be interacted with CTAB-coated Fe₃O₄ NPs surface. But, at pH higher than 10.0, lead can precipitate as hydroxide form. On the other hand, pH between 9.0 and 10.0 is favorable for complexation of lead and PAN (Marczenko, 1986). Therefore, pH of 9.0 was selected for further studies.

Close

Figure 2

Effect of sample pH on extraction efficiency. Conditions: 500 mL of 50 ng mL⁻¹ of lead; 1.0 mL of 0.001 mol L⁻¹ 351 PAN solution (L/M = 8); 100 mg Fe₃O₄ NPs; 50 mg CTAB; 5 min stirring, elution with 1.0 mL methanol.

Export to PPT

3.3 Effect of PAN concentration

PAN acts as a tridentate ligand and can form very stable complexes with metal ions (as ML or ML₂) through hydroxyl oxygen atom, nitrogen atom of pyridine and one of the azo group nitrogen atoms (Marczenko, 1986). PAN complexes of lead can be easily interacted with CTAB-coated Fe₃O₄ NPs, which increases extraction efficiency of lead. At 50 ng mL⁻¹ of lead, the effect of PAN concentration on extraction efficiency was studied by changing the molar ratio of PAN to lead in the range of 0–20 (Fig. 3). Results showed that by increasing the PAN to lead ratio up to 16 will cause an increase in the extraction recovery; after that, it remained constant. Therefore, PAN to lead ratio of 16 was used for the subsequent studies.

Close

Figure 3

Effect of PAN concentration on extraction efficiency. Conditions: 500 mL of 50 ng mL⁻¹ 355 of lead; sample’s pH = 9.0, 100 mg Fe₃O₄ NPs; 50 mg CTAB; 5 min stirring, elution with 1.0 mL methanol.

Export to PPT

3.4 Effect of the amount of surfactant

The adsorption of surfactant on the surface of mineral oxides can be divided into three regions (hemimicelles, mixed hemimicelles, and admicelles). In the first region, the zeta potential of mineral oxides changed from negative to zero due to the fact that hemimicelles were gradually formed on the mineral oxide surfaces. Upon increasing the cationic surfactant concentration, admicelles were formed dominatingly on the surface and the zeta-potential increases from zero to positive (Zhao et al., 2008a). Therefore, the outer surface of hemimicelles is hydrophobic whereas that of admicelles is ionic, which provides different mechanisms for retention of compounds and are both suitable for the SPE method. In the mixed hemimicelles phase, both hemimicelles and admicelles are formed in the surface of mineral oxides and the adsorption is driven by both hydrophobic interaction and electrostatic attraction. The effect of CTAB amounts on the extraction efficiency was considered in the ranges of 0–100 mg. Based on the obtained results (Fig. 4), increasing the amount of CTAB up to 25 mg resulted an increase in the extraction recovery; after that, it decreased due to unsuitable interactions. So, 25 mg of CTAB was chosen as optimum amount in order to achieve the highest possible extraction efficiency.

Close

Figure 4

Effect of CTAB amount on extraction efficiency. Conditions: 500 mL of 50 ng mL⁻¹ of lead; sample’s pH = 9.0, 2.0 mL of 0.001 mol L⁻¹ 359 PAN solution (L/M = 16); 100 mg Fe₃O₄ NPs; 50 mg CTAB; 5 min stirring, elution with 1.0 mL methanol.

Export to PPT

3.5 Effect of the adsorbent amount and extraction time

In comparison with the traditional sorbents (micro size sorbents), NPs offer a significantly higher surface area-to-volume ratio. Therefore, satisfactory results can be achieved with fewer amounts of NPs. In order to study the effect of the adsorbent, 10–150 mg of the Fe₃O₄ NPs was added to 500 mL of the sample solution. The obtained results showed that by increasing the adsorbent amounts from 25 up to 100 mg; extraction efficiency was increased slowly due to the increasing of accessible sites, and remained constant. So, a 100 mg of the Fe₃O₄ NPs was selected for all subsequent experiments.

Also, the effect of the extraction time on the extraction recovery of lead was investigated. Due to the shorter diffusion route for NPs plus the magnetically assisted separation of the magnetic NPs from the sample solutions, extraction of lead can be achieved in less time even for larger volumes of samples. The effect of the extraction time on the extraction recovery was investigated in the range of 1–5 min. Based on the obtained results, extraction recovery increased up to 2 min and then remained constant. Therefore, 2 min was chosen as optimum value in order to obtain a shorter analysis time.

3.6 Desorption conditions

Organic solvents are known to disrupt surfactant aggregates and were used to elute lead complex from the CTAB-coated Fe₃O₄ NPs surface (Zhao et al., 2008a,b; Sun et al., 2009). So, acetone, ethanol and methanol were selected as eluent. Among these eluents, methanol showed better desorption ability. In continue, effect of methanol’s acidity on desorption efficiency was considered. Results showed that desorption ability of methanol containing 0.5% HCl was found to be superior to that without adding HCl. Lead complex was quantitatively eluted from CTAB aggregates by using 1.0 mL of 0.5% HCl in methanol. This can be attributed to the fact that when the pH level was below or around its isoelectric point, charge density in Fe₃O₄ NPs surface was weak and favorable for the disruption of CTAB aggregates. Thus, 1.0 mL of 0.5% HCl in methanol was selected for desorption. On the other hand, in order to obtain maximum desorption efficiency, effect of the desorption time was investigated in the range of 1–5 min. Based on the obtained results, a time of 2 min was chosen for further studies. Moreover, further studies showed that the adsorbent can be used for at least five successive analyses without a considerable change.

3.7 Effect of sample volume

In order to obtain a higher enrichment factor, a larger volume of sample solution is required. Fortunately, due to the magnetically assisted separation of the adsorbent (Fe₃O₄ NPs), it is possible to collect the adsorbent from larger volumes of the sample solution. Thus the extraction of 25 μg of lead from different volumes of the water samples ranging from 500 to 1500 mL was investigated (Fig. 5). It was found that the best quantitative recovery (>90%) was obtained when the sample volumes were less than 1000 mL. However, the extraction efficiency would slightly decrease when the sample volumes were more than 1000 mL. Hence, a sample volume of 1000 mL was selected as the ideal volume for trace analysis of lead in water samples.

Close

Figure 5

Effect of sample volume on extraction efficiency. Conditions: 500 mL of 50 ng mL⁻¹ of lead; sample’s pH = 9.0, 2.0 mL of 0.001 mol L⁻¹ 364 PAN solution (L/M = 16); 100 mg Fe₃O₄ NPs; 25 mg CTAB; 2 min stirring, elution with 1.0 mL of 0.5% HCl in methanol.

Export to PPT

3.8 Interference studies

In view of the high selectivity provided by FAAS, the only interference may be attributed to the preconcentration step, in which the cations may react with PAN which may lead to a decrease in extraction efficiency. So, interference of co-existing ions on the preconcentration of lead was investigated. In order to investigate the effect of foreign ions on the extraction efficiency, solutions (500 mL) containing 50 ng mL⁻¹ of lead ions and various amounts of foreign ions were prepared and were extracted according to the procedure described in the experimental section. The deviation of ±10% from the recovery of the standard solution was taken as interference. The results are shown in the Table 1. It can be seen that a very good selectivity was achieved.

3.9 Analytical features

The analytical features of the proposed method such as precision, linear range of calibration curve and limit of detection were also examined. The results are summarized in Table 2. Precision of the method was estimated by applying successive retention and elution cycles with 1000 mL of a sample solution containing 10 μg of lead under the optimum conditions. The recovery of lead was quantitative (>95%) and the precision of the method was very good (RSD = 2.1%). The linear calibration range for measurements under the optimum conditions was 0.05–100 ng mL⁻¹ (R² = 0.9996) for 1000 mL of solution. The detection limit based on three-times of the standard deviation of the blank solution (3Sb/m) was found to be 0.005 ng mL⁻¹. Also, an enhancement factor of 1050 was calculated as the ratio between the slopes of the calibration curve submitted by the SPE procedure (for V = 1000 mL) and the direct calibration curve (without preconcentration).

3.10 Analysis of environmental water samples

In order to check the applicability of the proposed method for various water samples, lead ions were determined in different water samples including Tehran tap water, well water (Jaber Ebne Hayyan Pharmaceutical Company), mineral water (Damavand Brand) and wastewater sample collected around Tehran Refinery. An appropriate volume of sample solutions was adjusted to the optimum pH and subjected to the recommended procedure with external calibration method. Low concentrations of lead ions were found in the well water and wastewater samples. Reliability was checked by spiking samples. 1000 mL of the water samples spiked with 2.0 and 10.0 ng mL⁻¹ of lead ions was extracted by the proposed SPE procedure. Table 3 lists the recoveries and concentrations found for lead, expressed as the mean value (n = 3). Recoveries of lead ranged between 91% and 105% for the all four water samples. The results demonstrate the applicability of the proposed method to water analysis.

3.11 Comparison of the proposed method with other methods

The proposed method was compared with a variety of methods reported recently in the literature for the preconcentration and determination of lead. The distinct features of the proposed method are summarized in Table 4. As can be seen from Table 4, it is evident that the enhancement factor obtained with the CTAB-coated Fe₃O₄ NPs is very high in comparison with other methods. Large dynamic linear range, also, very low LOD and RSD are the other significant features of the method which are comparable to or even better than off-line cloud point extraction (CPE) followed by the very sensitive detection method (ET-AAS) (Chen et al., 2005).