Preconcentration of Trace Elements by Using 1-( 2-Pyridylazo )-2-naphthol Functionalized Amberlite XAD-1180 Resin and their Determination by FAAS Ş erife

Neste estudo foi sintetizada a resina Amberlite XAD funcionalizada por 1-(2-piridilazo)-2naftol (PAN) e a resina resultante foi usada na pré-concentração dos íons Cr(III), Mn(II), Fe(III), Ni(II), Cu(II), Cd(II) e Pb(II) presentes em baixas concentrações em diferentes amostras. Para isso, o copolímero foi nitrado, reduzido à correspondente amina, convertido ao sal diazonato com nitrito e em reação com PAN produziu a resina complexante XAD-1180-PAN. Para o método descrito, parâmetros analíticos para recuperação dos elementos traço foram avaliados tais como, pH, volume da amostra, quantidade de resina, vazão de retenção e remoção, volume e tipo de eluente. Os metais retidos na resina XAD-1180-PAN foram eluídos com HNO 3 2 mol L. A influência dos íons presentes na matriz, ou seja, Na(I), K(I), Ca(II) e Mg(II), sobre a recuperação dos elementos traço também foi avaliada, quando presentes isoladamente ou em conjunto. A repetibilidade do método foi avaliada nas condições experimentais ótimas estabelecidas. Os valores de recuperação, para todos os elementos, exceto o Cr(III), foram ≥92% e os desvios padrão relativos ≤8.5%. Os limites de detecção (3 s branco / coeficiente angular) para Cr(III), Mn(II), Fe(III), Ni(II), Cu(II), Cd(II) e Pb(II) foram 4,1; 0,13; 2,7; 1,2; 0,19; 0,06 e 0,13 μg L, respectivamente. O método desenvolvido foi utilizado para a pré-concentração e determinação de Mn(II), Fe(III), Ni(II), Cu(II), Cd(II) e Pb(II) em águas de abastecimento urbano, pluvial e interior, e de Ni(II), Cd(II) e Pb(II) em materiais de referência certificados (RM 8704 sedimento do rio Buffalo) por espectrometria de absorção atômica com chama (FAAS).


Introduction
In the analysis of trace metal ions present in various samples such as natural and waste water, biological and alloy samples, the direct determination with various instrumental methods is not possible owing to matrix effect and low concentration of metal ions in a sample.In trace analysis, therefore, a preconcentration and/or separation of trace elements from the matrix is frequently necessary to improve the detection limit and selectivity. 1,2arious methods used for this purpose are coprecipitation, 3 liquid-liquid extraction, 4 ion exchange, 5 and solid phase extraction. 6,7The solid phase extraction technique is now routinely used in various research and application areas due to its characteristics and advantages over the classical solvent extraction or other preconcentration techniques.Chelating resin sorption method is one of the most effective multielement preconcentration methods.The chelating resins include good selectivity, preconcentration factor, binding energy and mechanical stability, easy regeneration for multiple sorption-desorption cycles and good reproducibility in the sorption characteristics.[10] A number of supports and chelating ligands have been widely used in the design of chelating resins for the preconcentration and separation of trace metal ions from various matrices.The most prominent among the supports used are Amberlite XAD resins, 11,12 silica gel, 13 activated carbon, 14 activated alumina, 15 sepiolite, 16 and related polymeric supports. 17,18Some important and recently reported chelating matrices described using such supports are: silica gel modified with purpurogallin, 19 1-(2pyridylazo)-2-naphthol, 20 3-aminopropyltriethoxysilane, 21 1,8-dihydroxyanthraquinone, 22 2-aminothiazole, 23 N-propylsalicylaldimine, 24 surfactant coated alumina modified with 1-(2-pyridylazo)-2-naphthol, 15 Amberlite XAD-2 modified with chromotropic acid, 25 alizarin red-S, 26 Nitroso R salt, 27 Amberlite XAD-4 modified with o-aminobenzoic acid, 28 N-hydroxyethylenediamine, 29 and Amberlite XAD-16 modified with 1-(2-thiazolylazo)-2naphthol. 30mberlite XAD 1180 resin (Rohm and Haas Co., Philadelphia, USA) is a non-ionic, hydrophobic, crosslinked polystyrene-divinylbenzene copolymer having a surface area of 650 m 2 g -1 and a particle size of 0.3-0.9mm.It has physical, chemical and thermal stability. 31mberlite XAD-1180 resin has been used for the preconcentration and separation of trace metals in various samples in our previous works. 31,32-(2-pyridylazo)-2 naphthol forms stable complexes with many transition metals of interest; it is one of the most extensively used complexing agents for trace element analysis.Essentially it is insoluble in water and in alkaline solutions in which [OH -] is < 0.01 mol L -1 .It is soluble in alcohols, CHCl 3 , CCl 4 (≈ 0.01 mol L -1 ) etc. Its solutions are stable.In acidic solutions the pyridine nitrogen is protonated and in basic solutions the H of the OH group is ionized.The metal chelates of PAN have the metal atom bonded to O of the OH group, to pyridine N and to azo N. 33,34 Up to now, any information in literature could not found relevant to the usage of Amberlite XAD-1180-PAN chelating resin with the purpose of preconcentrating of trace metal ions from different samples.
In the present study, Amberlite XAD-1180-PAN chelating resin was prepared and characterized by using FTIR.The adsorption and preconcentration capabilities of the resin for Cr(III), Mn(II), Fe(III), Ni(II), Cu(II), Cd(II), and Pb(II) were ascertained.

Apparatus
The concentrations of metal ions were determined by a Perkin Elmer 3110 model atomic absorption spectrometer without background correction, equipped with an air-acetylene burner and hollow cathode lamps of the analytes.The operating conditions such as wavelength, bandwidth, and acetylene/air flow rates were adjusted according to the recommendations of manufacturer's.All pH measurements were made with a Jenko 672 model digital pH meter equipped with a combined pH electrode.Functional groups of the chelating resin synthesized were identified with a Jasco 460 model FTIR spectrometer.

Resin and reagents
Analytical reagent-grade chemicals were used without further purification.Stock solutions containing 1000 mg L -1 of Cr(III), Mn(II), Fe(III), Ni(II), Cu(II), Cd(II), and Pb(II) were prepared by dissolving appropriate amounts of their nitrate salts (purchased from Merck, Darmstadt, Germany) and diluting to 1000 mL with 1% (m/m) HNO 3 .Appropriate standard and model solutions of metal ions were prepared by diluting the stock solution prior to use.The buffer solutions used for adjusting pH value of solutions were CH 3 COOH/CH 3 COO -buffer for pH 4-6 and NH 3 /NH 4 Cl buffer for pH 8-10.

Synthesis of XAD-1180 -PAN chelating resin
The XAD-1180-PAN chelating resin was prepared as described in literature. 305 g of Amberlite XAD-1180 resin was tranferred to a 250 mL of flask containing 40 mL of concentrated HNO 3 and 60 mL of concentrated H 2 SO 4 .The mixture was heated to 60 °C and the heating procedure maintained for 1 h.Then the reaction mixture containing the nitrated resin was poured into a beaker filled with ice water.The resin was filtered and washed with distilled water until the pH 4-5.The nitrated resin (XAD-1180-NO 2 ) was reduced by refluxing at 90 °C for 24 h in a mixture of conc.HCl and ethanol (45:50, v/v) containing 50 g of stannous chloride.After cooling to ambient temperature, the product reduced was filtered and washed with a mixture of concentrated HCl and ethanol (45:50, v/v), 2 mol L -1 NaOH and distilled water, successively.The aminated polymer was suspended in 200 ml of ice-cold water and then diazotized with ~75 mL of 1 mol L -1 NaNO 2 and 100 mL of 1 mol L -1 HCl at 0-3 °C until the reaction mixture started to change the colour of iodide paper to violet.The diazotized resin was filtered, washed with ice-cold water and reacted with 2.5 g of PAN which was dissolved in a mixture including 100 mL of 10% (m/v) NaOH and 50 mL of methanol and afterwards the mixture was maintained at 0-3 °C for 24 h.After stirring for 30 min, the resulting product was filtered and washed with distilled water and 1 mol L -1 HCl until it was free from alkali.Afterwards it was dried at 60 °C over P 2 O 5 in vacuum.

Optimized separation and preconcentration procedure
The method was tested using the model solutions prior to its application to the real samples.For this purpose, the pH of the model solution including analyte ions, 10 μg (0.2 μg mL -1 ) each of Ni(II), Fe(III) and Cu(II), 20 μg (0.4 μg mL -1 ) each of Cr(III) and Pb(II), 2.5 μg (0.05 μg mL -1 ) of Cd(II), and 5 μg (0.1 μg mL -1 ) of Mn(II) was adjusted to the desired value (pH 9.0) by using 1 mol L -1 of NH 3 /NH 4 Cl buffer solution, and then the volume of the solution was completed to 50 mL with distilled water.The resulting solution was passed through the column filled with XAD-1180-PAN resin at a flow rate of 2 mL min -1 .The column was washed with 5-10 mL of distilled water.The metal ions retained on the resin were eluted by using 20 mL of 2 mol L -1 HNO 3 .The eluate was evaporated near to dryness on a hot plate and then the moist residue was diluted to 5 mL with 1 mol L -1 HNO 3 .The Cr(III), Mn(II), Fe(III), Ni(II), Cu(II), Cd(II), and Pb(II) contents were determined by flame AAS.The chelating resin was used repeatedly after washing with 2 mol L -1 HNO 3 and water.The blank analyses were made in the same way.

Analysis of water samples
A tap water sample taken from the laboratory of the University was analysed without pre-treatment.The rain water and stream water samples were collected in prewashed polyethylene bottles.These samples were filtered through a Millipore cellulose membrane filter of 0.45 μm pore size and then acidified with 5 mL of concentrated nitric acid per liter of sample.The samples were analyzed by the separation/preconcentration procedure mentioned above.For the analysis, the aliquots of 400 mL of tap water and 250 mL of rain water and stream water were used.

River Sediment Digestion (SRM)
A 50 mg of the reference material (RM 8704 Buffalo river sediment, dried at 110 °C for 2 h) was digested in a beaker by adding 10 mL of aqua regia.The mixture was heated on a hot plate until to almost dryness.Then, 10 mL of aqua regia was added again to the residue and the mixture was evaporated to dryness.The insoluble part was filtered through a filter paper with blue band, and the filter paper and beaker were rinsed with hot 1 mol L -1 HNO 3 .5 mL of 1 mol L -1 NaF was added to the filtrate in order to prevent the precipitation of Fe(III) found in the sample in excess amounts in hydroxy forms.So, Fe(III) ions have been masked by F -in the solution as (FeF 6 ) 3 -anionic complex.The pH of this solution was adjusted to pH 9 by adding NH 3 / NH 4 Cl buffer.The total volume was made up to 50 mL with water.The column procedure described above was applied to separation and preconcentration of the metal ions from the sample solution.After the elution, the eluate was evaporated near to dryness and completed to 5 mL with 1 mol L -1 HNO 3 .The determination of the studied metals in the final solution, except for Fe(III), was performed by using FAAS.

Characteristics of XAD-1180-PAN chelating resin
Analytical data of the intermediates and synthesized resin were illustrated in Table 1.XAD-1180-NO 2 shows two strong bands at 1526 (asymmetrical stretching) and 1348 cm -1 (symmetrical stretching), which describe characteristic nitro groups.In the IR spectrum of XAD-1180-NH 2 , a band was observed at 1625 cm -1 owing to amine groups (N-H bending).The IR spectrum of PAN functionalized Amberlite XAD-1180 was compared with that of pure Amberlite XAD-1180.As it can be seen in Table 1, there are additional bands at 3433 cm -1 (O-H stretching) and 1630 cm -1 (-N=N-stretching), which appear to originate due to modification of the resin by the ligand and are characteristics of O-H and -N=Nvibrations, respectively.The results obtained are in good agreement with informations given in literature. 2,8,30

Effect of pH on recoveries
The sorption of metal ions on chelating resin is dependent on the pH of a sample solution due to the competitive reaction between chelate forming groups and hydrogen ions in the solutions. 3010 μg of Ni(II), Fe(III) and Cu(II), 20 μg of Cr(III) and Pb(II), 2.5 μg of Cd(II), and 5 μg of Mn(II) were spiked to a 50 mL of the model solution.The pH of this solution was adjusted in the range of 4 to 10 using different buffer solutions and then the separation/preconcentration procedure described was applied.As it can be seen in Figure 1, quantitative recoveries (≥ 95%) were obtained for all the elements at pH 9.0.Therefore, the working pH was chosen as 9.0 for the following experiments.In addition, in the column study performed with only the original Amberlite XAD-1180 resin without using PAN reagent, the recovery values for the elements at the pH 9.0, except for Cu which presented a recovery around 95%, were not quantitative.

Effect of resin amount
The resin amount is one of the other important parameters for obtaining quantitative recovery.After adjusting the pH of model solution including trace metal ions to 9.0 by using NH 3 /NH 4 Cl buffer solution, the model solutions were passed through the column filled with 100-500 mg of the XAD-1180-PAN resin.The metal ions adsorbed were eluted by 20 mL of 2 mol L -1 HNO 3 .The recovery (%) values for all the elements were found as quantitative for 0.4 g of the resin (Figure 2).

Effect of type, concentration and volume of eluent
The elution of the retained metals from the XAD-1180-PAN resin was studied by using various eluting agents and different eluent volumes.As seen in Table 2, 20 mL of 2 mol L -1 HNO 3 was found to be the most adequate eluent for quantitative recoveries of all the elements examined.

Effect of sample and eluent flow rate
The effect of flow rates of the sample and eluting solutions on metal sorption and desorption, respectively, was investigated for the flow rates of 1.0, 2.0, 2.5 and 3.5 mL min -1 under the optimum conditions.It was found that the retention of all metal ions, except for Fe(III) and Cr(III), was not affected from the flow rate up to 3.5 mL min -1 .The effect of eluent flow rate on the recovery was examined for the flow rates of 1.0, 2.0 and 3.5 mL min -1 .The optimum flow rate for the elution of all the elements was found as 2.0 mL min -1 .

2.
Effect of the amount of the chelating resin on the recovery of the elements.J. Braz.Chem.Soc.

Effect of sample volume
In the analysis of real samples using a preconcentration procedure, the sample volume is one of the important parameters for obtaining high preconcentration factors.For that reason, the volumes of model solutions containing 10 μg each of Ni(II), Fe(III) and Cu(II), 20 μg each of Cr(III) and Pb(II), 2.5 μg of Cd(II), and 5 μg of Mn(II) were varied from 50 to 750 mL to determine the maximum applicable volume of sample solution for natural water samples.The metals were preconcentrated on the XAD-1180-PAN resin by applying the proposed procedure.The effect of the sample volume on the recovery was illustrated in Figure 3.The recoveries for Cu(II), Ni(II) and Fe(III) were quantitative up to the sample volume of 750 mL.Breakthrough volume for Cr(III) was 400 mL and almost 750 mL for Mn (94%) and Cd (92%), and also 500 mL for Pb (91%).If the volume of final solution is down to 5 mL, thus an enrichment factor of 100 for Cu(II), Ni(II), Fe(III) and Pb(II), of 80 for Cr(III), of 150 for Cd(II), and Mn(II) can be acquired.

Effect of matrix elements with and without using the preconcentration procedure
The effect of Na(I), K(I), Ca(II), Mg(II), Cl -and NO 3 -ions on the signals of the metals was examined both with the proposed method and without it.When the influence of matrix components without using the enrichment procedure proposed was investigated, it was seen that especially the concentrations of both 500 mg L -1 and 1000 mg L -1 of Ca(II), and Mg(II) ions had a decreasing effect on Fe(III), Cd(II), Cr(III) and Mn(II) absorbance signals.When the concentrations of Na(I) and K(I) ions were 500 mg L -1 , any interference on the elements studied was observed, except for Cd(II), at the concentration of 500 mg L -1 K(I).While the model solution containing 500 mg L -1 of each of Na(I), K(I), Ca(II) and Mg(II) ions shows a decreasing effect on Ni(II), Fe(III), Cr(III) and Pb(II) signals, the solution containing 1000 mg L -1 of matrix ions has an important decreasing effect on the signals of all the elements.These results show that a separation procedure is requisite for the determination of analytes in a complex matrix medium prior to their measurements by FAAS.
Because of the decreasing influence of the matrix ions on the metal signals, the separation procedure is necessary.For this purpose, the metal ions at appropriate amounts were added to the model solutions of 50 mL containing the matrix ions [Ca(II), Mg(II), K(I) and Na(I)] at increasing concentrations.The pH of the solutions was adjusted to 9.0.The results showed that the recovery of the metal ions studied was not affected by the solution containing the matrix ions at high concentrations (see Table 3).

Repeatability and detection limit of the method
The repeatability of the method was examined under the optimum experimental conditions described above by using the model solutions.For this purpose, the analyses were performed by applying the proposed method at three consecutive days (totally eleven replicates).The first three runs were performed at the first day while the others (four runs for each day) were performed at the second and the third days.The contents of the elements in the elution solution were determined by FAAS.The repeatability of the described method was evaluated by monitoring the changes in the recoveries of the analytes (R±s%), 103±3 for Ni(II), 98±5 for Fe(III), 97±2 for Cu(II), 94±5 for Cr(III), 97±5 for Pb(II), 106±8 for Cd(II), and 94±8 for Mn(II).
For the determination of detection limit of the method, 50 mL of blank solution was passed through the column (n=20).The detection limit was calculated as the ratio of the three standard deviations of the blank to the slope of the calibration curve.When an enrichment factor of 100 for Cu(II), Ni(II), Fe(III) and Pb(II); 80 for Cr(III); and 150 for Cd(II) and Mn(II) is considered, the following detection limits: 4.1, 0.13,

The accuracy of the method
As no standard reference material for water sample was available in our laboratory, the accuracy of the described separation/preconcentration method was tested in the recovery studies by adding known amounts of the elements to the water samples of 50-100 mL and also by analyzing of a standard reference material (RM 8704 Buffalo river sediment).The results obtained from the certified reference material and the water samples are depicted in Table 4 and 5, respectively.The recovery values obtained from the water samples were satisfactory, except for Cr(III).The results obtained from the described method were in good agreement with the analysis of the certified reference material for Ni(II), Pb(II) and Cd(II), except for Mn(II) and Cr(III).The reason of the exception for Mn(II) and Cr(III) may be interferences of some matrix components in the sediment sample, such as silicate and aluminum.

Table 1 .
Analytical data of the intermediates and synthesized resin Figure 1.Effect of pH on the recovery of metal ions on XAD-1180-PAN resin.

Table 2 .
Effect of type, concentration and volume of eluents on the recovery (n=3) Figure 3.Effect of sample volume on the recovery of the elements.

Table 3 .
Influence of matrix ions on the retention of metal ions on the Amberlite XAD-1180-PAN chelating resin (n=3)

Table 5 .
The recovery of the elements added to the tap water and rain water samples after preconcentration on the Amberlite XAD-1180-PAN chelating resin (in triplicates for each concentration)

Table 6 .
Determination of Ni(II), Fe(III), Cu(II), Pb(II), Cd(II) and Mn(II) in the water samples by FAAS after preconcentration on the Amberlite XAD-1180-PAN chelating resin Number of replicates, b Limit of detection. a

Table 7 .
Comparison of detection limits (LOD, µg L -1 ) and enrichment factors (EF) of some metal ions using the chelating resins prepared with different ligands and supports