Characterization of an EDTA Bonded Conducting Polymer Modified

Jan 31, 2003 - Calibration plots for the EDTA-CPME with square wave voltammetry were obtained in the concentration range between 5.0 × 10-10 and 1.0 ...
0 downloads 0 Views 383KB Size
Download PDF
Anal. Chem. 2003, 75, 1123-1129

Characterization of an EDTA Bonded Conducting Polymer Modified Electrode: Its Application for the Simultaneous Determination of Heavy Metal Ions Md. Aminur Rahman,† Mi-Sook Won,‡ and Yoon-Bo Shim*,†

Department of Chemistry and Korea Basic Science Institute, Pusan National University, Pusan 609-735, South Korea

An EDTA bonded conducting polymer modified electrode (EDTA-CPME) was fabricated by polymerization of 3’,4’diamino-2,2’;5’,2’’-terthiophene monomer on a GCE, followed by the reaction with EDTA in the presence of catalyst. The surface of the resulting modified electrode was characterized with EQCM, ESCA, SEM, Auger electron spectroscopy, scanning Auger microscopy, and electrochemical methods. The amounts of polymer and EDTA attached on the polymer film were determined. Simple immersing of the EDTA-CPME into a sample solution led to the chemical deposition through the complexation with Pb2+, Cu2+, and Hg2+ ions, simultaniously. Various experimental parameters that affect the simultaneous analysis of the metal ions, e.g., EDTA amount, pH, deposition time, and deposition temperature, were optimized. Calibration plots for the EDTA-CPME with square wave voltammetry were obtained in the concentration range between 5.0 × 10-10 and 1.0 × 10-7 M for Cu(II) and between 7.5 × 10-10 and 1.0 × 10-7 M for Pb(II) and Hg(II). The detection limits for Pb(II), Cu(II), and Hg(II) ions were determined to be about 6.0 × 10-10, 2.0 × 10-10, and 5.0 × 10-10 M, respectively. Interference effects from other metal ions were studied at various pHs and it was found that there was little or no effect on the simultaneous determination. The stability of the EDTACPME was remarkably improved by coating the surface with the Nafion film, and the electrode can be used for more than one month. Analytical availability of the EDTACPME was demonstrated by the application for the certified standard urine reference material and tap water. Stripping voltammetry is the most sensitive electrochemical method for the determination of trace metals. The method is based on the preconcentration and stripping of an analyte in the sample solution.1,2 Stripping analysis using a chemically modified electrode (CME) has higher selectivity due to the selective reaction to * Corresponding author. Phone: (+82) 51 510 2244. Fax: (+82) 51 514 2430. E-mail: [email protected]. † Department of Chemistry. ‡ Korea Basic Science Institute. (1) Vydra, F.; Stulik, K.; Julakova, E. Electrochemical Stripping Analysis; Horwood: Chichester, U.K., 1976. (2) Wang, J. Stripping Analysis: Instrumentation and Application; Verlag Chemie: Deerfield Beach, FL, 1985. 10.1021/ac0262917 CCC: $25.00 Published on Web 01/31/2003

© 2003 American Chemical Society

specific metal ions.3-5 One major advantage of using the CME is that it can detain a wider adjustable preconcentration without interferences compared with the conventional voltammetric techniques. When the experimental condition is optimized, the method with a CME has the least interference by other species in the sample. Electroanalytical approaches of a conductive polymer based on CMEs (CPME) toward trace metal ion detection have received considerable attention.6 Polypyrrole and polypyrrole-N-carbodithioate modified electrodes had been used in the anodic stripping voltammetric determination of silver and mercury.7 The preconcentration and determination of silver ions with conductive polymers having amine groups8,9 were also reported. In addition, functionalization of conducting polymers adsorbed on the electrode surface should increase the scope of voltammetric analysis for metal ion determination. The functionalization of conducting polymers can be carried out in three ways: before, during, and after the polymerization process. The first way involves covalently linking a specific group to the starting monomer and subsequently preparing the functionalized polymer.10 This method can be adopted only if a specific group is stable during the polymerization. For the second one, some of the specific anions are electrostatically incorporated simultaneously during the electropolymerization. In this way, the functionalization is obtained if the doping anion is irreversibly captured in the polymer matrix. The incorporation of anionic complexing ligands to the conducting polymers comes under this category.11 2,6-Pyridinedicarboxylic acid (PDCA) and ethylenediaminetetraacetic acid (EDTA) simply doped in polypyrrole film were employed for the determination of silver ions.12 In this case, the formation of the polypyrrole/ EDTA film was not reproducible due to simple doping of EDTA in the polypyrrole film. The third way is the after polymerization method. In this case, the functionalization is performed after the (3) Murray, R. W. In Electroanalytical Chemistry; Bard, A. J., Ed.; Marcel Dekker: New York, 1983; Vol. 13, pp 191-368. (4) Murray, R. W. Acc. Chem. Res. 1980, 13, 135-141. (5) Snell, K. D.; Keenan, A. G. Chem. Soc. Rev. 1979, 8, 259-282. (6) Imisides, M. D.; John, R.; Riley, P. J.; Wallace, G. G. Electroanalysis 1991, 3, 879-889. (7) O’Riordan, D. M. T.; Wallace, G. G. Anal. Chem. 1986, 58, 128-131. (8) Lee, J.-W.; Park, D.-S.; Shim, Y.-B.; Park, S.-M. J. Electrochem. Soc. 1992, 139, 3507-3514. (9) Park, D.-S.; Shim, Y.-B.; Park, S.-M. Electroanalysis 1996, 8, 44-48. (10) Cai, Q.; Khoo, S. B. Anal. Chem. 1994, 66, 4543-4550. (11) Shiu, K. K.; Chan, O. Y.; Pang, S. K. Anal. Chem. 1995, 67, 2828-2834. (12) Wallace, G. G.; Lin, Y. P. J. Electroanal. Chem. 1988, 247, 145-156.

Analytical Chemistry, Vol. 75, No. 5, March 1, 2003 1123

polymerization if one of an active group in a polymer is allowed to covalently bind to another active group of a specific molecule.13 Recently, we studied the EDTA functionalized CPME. Our preliminary results show that this modified electrode can be used to detect various metal ions at different pHs. While there are some reports on the metal ion detection by conducting polymers, to our knowledge, there is no report on simultaneous determination of metal ions with a modified surface of conducting polymer obtained through covalent bonding between a conducting polymer and a modifier. Therefore, we characterized and used the EDTA-CPME in the simultaneous determination of metal ions. The polymer film formation on a glassy carbon electrode (GCE) and functionalization of the EDTA on the CPME has been confirmed by EQCM, SEM, Auger electron spectroscopy (AES), scanning Auger microscopy (SAM), ESCA, and electrochemical methods. The characteristics of the EDTA-CPME were investigated for the chemical preconcentration/stripping of metal ions. Various experimental parameters that affect the sensitivity of EDTA-CPME, such as EDTA amount, deposition time, deposition temperature, pH, and interferences were optimized. Finally, this method was applied for the simultaneous analysis of Pb, Cu, and Hg with linear sweep and Osteryoung square wave voltammetric methods.14 EXPERIMENTAL SECTION Reagents. Standard metal ion solutions of 1000 ppm were obtained from Kanto Chemical Co. Inc. and diluted step by step (1.0 × 10-3-1.0 × 10-10 M) to an adequate concentration immediately prior to use. An acetate buffer solution was prepared by adjusting 0.2 M sodium acetate (Aldrich) to the desired pH with the addition of 0.2 M glacial acetic acid. A phosphate buffer solution was prepared by adjusting 0.1 M disodium hydrogen phosphate (Sigma) with 0.1 M sodium dihydrogen phosphate (Aldrich). 1-Ethyl-3 (3-(dimethylamino)propyl) carbodiimide (EDC) and dichloromethane (99.8%, anhydrous, sealed under N2 gas) were received from Sigma Co. and were used as received. 3’,4’Diamino-2,2’; 5’,2’’-terthiophene (DATT) was synthesized by a method in the literature.15 Tetrabutylammonium perchlorate (TBAP, electrochemical grade) was received from Fluka Co., purified, and then dried under vacuum at 10-5 Torr. Other chemicals were of analytical reagent grade. Distilled water was obtained from a Millipore Milli-Q water purification system (18 MΩ cm). Apparatus. An EDTA-CPME on a GCE (area, 0.03 cm2), an Ag/AgCl (in saturated KCl) electrode, and a Pt wire were used as working, reference, and counter electrodes, respectively. Cyclic and linear sweep voltammograms were recorded using a potentiostat/galvanostat (Kosentech, Korea, model KST-CV 104A). Square wave voltammograms were recorded using a BAS-CV 50-W voltammetric analyzer (Bioanalytical Systems inc., West Lafayette, IN) using a pulse height of 5 mV, square wave amplitude of 25 mV, and square wave frequency of 60 Hz. Solutions were degassed 10 min before every measurement by purging with N2 gas, which was also flowed over the solution during the experiment. The electrochemical quartz crystal microbalance (EQCM) experiment (13) Wierl, L. M.; Guadalupe, A. R.; Abruna, H. D. Anal. Chem. 1985, 57, 20092011. (14) Osteryoung, J, G.; Osteryoung, R. A. Anal. Chem. 1985, 57, 101A. (15) Kitamura, C.; Tanaka, S.; Yamashita, Y. Chem. Mater. 1996, 8, 570-578.

1124

Analytical Chemistry, Vol. 75, No. 5, March 1, 2003

Scheme 1

was performed using a Seiko EG & G model QCA 917 and PAR model 263A potentiosatat/galvanostat. A gold working electrode (area, 0.196 cm2; 9 MHz; AT-cut quartz crystal) was used for EQCM experiments. SEM images were obtained using a Cambridge Stereoscan 240. ESCA was performed using a VG Scientific Escalab 250 XPS spectrometer with monochromated Al KR source with charge compensation at KBSI (Busan). AES was performed using a VG Scientific Microlab 350, UK at KBSI (Busan). Primary electron beam energy of 5.0 keV and primary beam current of ∼200 nA were used in all AES analyses. Electrochemical Polymerization and Modification. Prior to electropolymerization, a bare GCE was polished with 0.5-µm alumina/water slurry on a polishing cloth to mirror finish, followed by sonicating and rinsing with distilled water. A 1 mM DATT monomer solution was prepared in a 0.1 M TBAP/CH2Cl2 solution. The polymerization onto a GCE was carried out by cycling the potential between 0.0 and + 1.6 V. After electropolymerization, the CPME was washed with dichloromethane to remove excess monomers from the electrode surface. The CPME was then immersed in a phosphate buffer solution of pH 7.0 containing 0.1 M EDC (catalyst) and 0.01 M EDTA with stirring in order to covalently bond with the COOH groups in EDTA and NH2 groups in poly-DATT as shown in Scheme 1. The resulting EDTA-CPME was rinsed carefully with distilled water and dried before each subsequent experiment. RESULTS AND DISCUSSION Characterization of the EDTA Functionalized Conducting Polymer Film. Figure 1a shows a series of cyclic voltammograms (CVs) recorded for a 1.0 × 10-3 M DATT monomer in a 0.1 M TBAP/CH2Cl2 solution, while the potential was cycled between 0.0 and +1.6 V versus Ag/AgCl. The CV exhibited two oxidation peaks at +0.75 and +1.45 V (vs Ag/AgCl) during the first anodic scan, which are due to the oxidation of amine group to imine group and the oxidation of monomer to form the polymer, respectively. The currents of these peaks decreased as the cycle numbers increased. This clearly demonstrates that polymer film

Figure 1. (a) Consecutive CVs (10 cycles) recorded for the oxidation of 1.0 × 10-3 M DATT monomer in 0.1 M TBAP/CH2Cl2. (b) CVs recorded for the oxidation of polymer film in pH 4.25 acetate buffer of (solid line) and in 0.1 M TBAP/ CH2Cl2. The scan rate was 100 mV/s.

immediately forms after the oxidation of the DATT monomer at +1.45 V. The thickness of the polymer film increased as the cycle number increased. The small and broad reduction peak of the polymer was observed at +0.48 V (vs Ag/AgCl) at the cathodic scan. The thickness of the polymer film formed after 10 cycles was estimated as ∼500 nm from the SEM image. Figure 1b shows CVs recorded for the oxidation of the polymer in a 0.2 M acetate buffer (dash line) and in a dichloromethane solution containing 0.1 M TBAP (solid line). The redox peaks of the polymer film in an acetate buffer solution and in dichloromethane were observed at 0.53/1.0 and 0.40/0.94 V (vs Ag/AgCl), indicating that the polymer film is electrochemically active. The peak currents of these redox peaks were proportional to the scan rate, indicating that the current flows due to the redox behavior of the adsorbed polymer film. The peak separations between the peaks increase as the scan rate increased, indicating that the electron-transfer process from the electrode to the polymer film is quasi-reversible. EQCM studies were carried out concurrently with recording the CV to determine the amount of the polymer formed on the electrode surface. Figure 2a shows the frequency and mass changes during polymerization of a 1.0 × 10-3 M DATT monomer in a 0.1 M TBAP/CH2Cl2 solution. The frequency decreased and the mass adsorbed on the electrode increased as the time passed, which indicated the growth of the polymer film on the electrode surface. After 10 cycles, the frequency change, ∆f was found to be 3.845 kHz. Mass change, ∆m, was calculated using the following equation:16

∆m ) ∆f × 5.608 ( ng/cm2)/Hz × 0.196 cm2 where, ∆f was the change in frequency, 5.608 (ng/cm2)/Hz was

Figure 2. Frequencies and masses changed (a) during polymerization of a 1.0 × 10-3 M DATT monomer in a 0.1 M TBAP/CH2Cl2 solution and (b) during EDTA attachment with poly-DATT in CPME.

the sensitivity factor calculated from the physical constant for quartz, and 0.196 cm2 was the electrode area. The amount of the polymer film formed at the GCE surface after 10 cycles was calculated to be ∼4.2 ( 0.2 µg. The thoroughly washed CPME was then used to attach EDTA through the covalent bond. Figure 2b shows the frequency and mass changes during EDTA immobilization with poly-DATT in CPME. The frequency gradually decreased and reached a steady state after ∼1.5 h. After 2-h immobilization, ∆f was found to be 0.505 kHz. The mass change, ∆m of EDTA on the CPME was calculated using the following equation:

∆m ) ∆f × 5.608 (ng/cm2)/Hz × 0.196 cm2 The amount of EDTA attached on the polymer film was calculated to be ∼0.55 ( 0.1 µg. Figure 3 shows ESCA spectra for the CPME and the EDTACPME. The C 1s spectrum for poly-DATT in the CPME exhibited two peaks at about 286 and 284.5 eV, which corresponded to CsN and CsH, CsS, or CsC bonds, respectively.17 In the case of EDTA bonded poly-DATT, an additional peak for the C 1s spectrum was observed at 288.2 eV, which is due to NsCdO along with CsN and CsC bonds as observed in CPME.18 In the C 1s spectrum, CsN peaks slightly shifted to the higher energy in the case of EDTA bonded poly-DATT, indicating an interaction between CsN and NsCdO groups. As shown in Figure 3b, two N 1s peaks in the spectrum of poly-DATT were observed at about 398.7 and 396.6 eV, due to the sNH2 and dNH groups, respectively.19-21 A sNH2 peak at ∼398.7 eV disappeared in the (16) Lee, T.-Y.; Shim, Y.-B. Anal. Chem. 2001, 73, 5629-5632. (17) Pantano, P.; Khur, W. G. Anal. Chem. 1993, 65, 623-630. (18) Chan, H. S. O.; Ng, S. C.; Sim, W. S.; Tan, K. L.; Tan, B. T. G. Macromolecules 1992, 25, 6029-6034. (19) Tan, K. L.; Tan, B. T. G.; Kang, E. T.; Neoh, K. G. Phys. Rev. B 1989, 39, 8070.

Analytical Chemistry, Vol. 75, No. 5, March 1, 2003

1125

Figure 3. ESCA spectra of (a) C 1s, (b) N 1s, (c) S 2p, and (d) O 1s for (1) poly-DATT in CPME (dash line) and EDTA bonded polyDATT in EDTA-CPME (solid line).

Figure 4. Auger electron spectra of (a) bare GCE, (b) poly-DATT in CPME, and (c) EDTA bonded poly-DATT in EDTA-CPME.

spectrum of EDTA bonded poly-DATT. This indicates that the formation of amide bond results in disappearance of the sNH2 group. One peak at 163.5 eV (Figure 3c) in the S 2p spectra was observed in the both poly-DATT and EDTA bonded poly-DATT, which is due to SsC.22 The O 1s spectra in Figure 3d was also obtained for poly-DATT and EDTA bonded poly-DATT. No peak in the O 1s spectrum was observed in the poly-DATT, while EDTA bonded poly-DATT spectrum shows a strong O 1s peak at about 531.5 eV corresponded to CdO.23 The presence of the CdO peak in the O 1s spectrum obtained for the EDTA bonded poly-DATT confirmed that the EDTA molecules covalently attached on the poly-DATT film. Figure 4 shows the AES spectra for (a) bare GCE, (b) polyDATT, and (c) EDTA bonded poly-DATT. CKLL Auger peak was observed at 270 eV in the spectrum of bare GCE. Auger peaks of SLMM, CKLL, and NKLL were observed at 150, 270, and 380 eV, respectively, in poly-DATT, indicating that the formation of poly(20) Menardo, C.; Nechtschein, M.; Rousseau, A.; Travers, J. P.; Hany, P. Synth. Met. 1988, 25, 311-314. (21) Chan, H. S. O.; Ang, S. G.; Ho, P. K. H.; Jhonson, S. Synth. Met. 1990, 36, 103-107. (22) Bernede, J. C.; Tregouet, Y.; Gourmelon, E.; Martinez, F.; Neculqueo, G. Polym. Degrad. Stab. 1997, 55, 55-64. (23) Ng, S. C.; Chan, H. S. O.; Wong, P. M. L.; Tan, K. L.; Tan, B. T. G. Polymer 1998, 39, 4963-4968.

1126 Analytical Chemistry, Vol. 75, No. 5, March 1, 2003

Figure 5. SAM images (T, 5 µm) of (i) C, (ii) S, (iii) N, and (iv) O in EDTA bonded poly-DATT. (v) SEM images (T, 50 µm) of polyDATT and (vi) EDTA bonded poly-DATT.

DATT film on the GCE surface. The presence of OKLL peak at 508 eV in the EDTA bonded poly-DATT clearly indicated that the EDTA molecules adsorbed on poly-DATT. The morphologies of the bare GCE, poly-DATT, and EDTA bonded poly-DATT were characterized by SAM and SEM. Figure 5 shows SAM and SEM images obtained for poly-DATT and EDTA bonded poly-DATT. The SAM images show that C and S elements were distributed similarly in the poly-DATT and EDTA bonded poly-DATT. While the image from the N elements were observed as mountain-like and valley-like shapes in poly-DATT and EDTA bonded poly-DATT. The image of O elements was observed as the island-like shape on the EDTA bonded poly-DATT film and the O elements did not distribute uniformly. Darker and lighter parts in the images correspond to lower (valleys) and higher parts (mountains) of the surface materials. From the SEM image of poly-DATT, it can be seen that poly-DATT film formed uniformly on the GCE surface. A SEM image of EDTA bonded poly-DATT clearly shows some white small-sized aggregates on the surface of the EDTA-CPME. These aggregates on the surface were not seen in the SEM image for poly-DATT. This means that EDTA molecules covered the poly-DATT film successfully. However, EDTA did not cover the whole surface completely and densely. The random island-like distribution of EDTA molecules can be seen in the SEM image for EDTA-CPME. On the basis of the SAM and ESCA data obtained before and after attaching EDTA moieties on the polymer-coated surface, we can conclude that small white aggregates were due to the EDTA molecules. Electrochemical Response of Modified Electrodes for Metal Ions. Immersing of the EDTA-CPME into the measuring solution containing heavy metal ions led to the chemical deposition of the ions on the electrode surface through complexation between the ions and the EDTA-CPME surface. Complexation of metal ions with EDTA-CPME was carried out in a sodium acetate/acetic acid buffer solution containing 1.0 × 10-5 M metal ions such as Pb(II), Cu(II), Hg(II), Cr(II), Fe(II), Co(II), Ni(II), Zn(II), Cd(II), and Ag (I) by immersing the electrode for 10 min with stirring. The accumulation of the metal ions on the EDTA-CPME were carried out chemically without applying a reduction potential to avoid the affect of other interfering species reducible by

Figure 6. Effects of (a) EDTA concentration, (b) deposition temperature, (c) deposition time, and (d) pH on the LSV peak heights of 1.0 × 10-5 M Pb(II) (dot line), Cu(II) (solid line), and Hg(II) (dash line) ions in sodium acetate/acetate buffer solution.

applying the potential in the preconcentration solution. After accumulation of the metal ions in the preconcentration solution, EDTA-CPME was taken off and washed with distilled water and then transferred to a cell containing a blank solution of sodium acetate/acetic acid buffer. Before recording the stripping currents, metal ions complexed on the EDTA electrode surface were reduced at -0.9 V for 10 s in a measuring blank solution. Linear sweep voltammograms (LSVs) for the EDTA-CPME after this step were recorded from -0.9 to +0.75 V in a separate blank electrolyte solution at the scan rate of 50 mV/s. LSV exhibited only three anodic peaks at about -0.30, 0, and +0.40 V versus Ag/AgCl, corresponding to the oxidation of reduced Pb(0), Cu(0), and Hg(0) to Pb(II), Cu(II), and Hg(II), respectively, at the EDTACPME surface. Other metal ions did not show any response at this experimental condition, although individual metal ion responsed with the EDTA-CPME. This means that other metal ions including that in the sample solution did not compete with the complexation of Pb(II), Cu(II), and Hg(II) with EDTA-CPME at this pH range. EDTA forms a complex with most of metal ions, and the formation constants of EDTA complexes with the metal ions increased as follows: Hg(II) > Cu(II) > Ni(II) > Pb(II) > Zn(II) > Cd(II) > Co(II) > Fe(II) > Ag(I).24 In the present work, complexation reactions of EDTA with metal ions were studied in a moderately acidic medium (pH 2-6). The complexation of metal ions with EDTA may be affected by the poly-DATT. Optimization of Analytical Conditions. CPMEs modified with various amounts of EDTA in molar concentrations were studied in order to find the optimum EDTA quantity for the analysis of metal ions. Figure 6a shows the effect of EDTA amount on the LSV peak currents of 1.0 × 10-5 M Pb(II), Cu(II), and Hg(II) ions, which were deposited for 10 min on the EDTA-CPME. The peak currents were found to be increased as the EDTA concentration increased from 0.1 to 10 mM. An increase of EDTA (24) Skoog, A. D.; West, D. M.; Holler, F. J. Fundamentals of Analytical Chemistry, 5th ed.; Saunders College Publishing: Philadelphia, 1988.

concentration would improve the efficiency of the complexation with metal ions. The 100-fold increase in the concentration of EDTA compared with 0.001 M seemed to produce an increase in the quantity of active ligand sites being complexed with metal ions. This would result in an increase of peak currents. The further 10-fold increase of EDTA concentration from 0.01 to 0.1 M did not give a noticeable change in the peak currents for the oxidation of Pb, Cu, and Hg. Therefore, the concentration of EDTA to be bound on the CPME was determined as 0.01 M. The effect of the deposition temperature on the stripping currents was studied in the range from 20 to 50 °C under the previous condition as shown in Figure 6b. As the temperature of the preconcentration solution increased gradually from 20 to 30 °C, the peak currents increased. Then, the current began to decrease above 30 °C. The decrease in the peak currents from 35 to 50 °C might be due to the instability of the modified electrode at these temperatures. Thus, the deposition temperature of 30 °C was used in all subsequent experiments. The effect of deposition time for the metal ions on the EDTACPME was studied between 1 and 15 min and is shown in Figure 6c. The EDTA-CPME was immersed in 1.0 × 10-5 M metal ions containing sodium acetate/acetic acid buffer solution at each deposition time at 30 °C. The peak currents increased linearly as the deposition time increased from 1 to 10 min for Pb, Cu, and Hg, indicating an enhancement of uptake of metal ions on the electrode surface at a longer deposition time. The increase in the peak current continued until a constant-current level was attained due to the saturation of active sites of EDTA molecules by metal ions. The peak currents were found to be almost constant at 15min deposition time. Therefore, the optimum deposition time was chosen as 10 min. Figure 6d shows pH dependency of the stripping current of metal ions at the range of pH 1.5-6.5 in 0.2 M sodium acetate/ acetic acid buffer solutions. In this case, the chemical deposition was done under the same condition (1.0 × 10-5 M metal ion solution for 10 min at 30 °C). The current responses were strongly influenced by the pH of the deposit solution. From pH 1.7 to pH 4.25, the stripping current gradually increased and the maximum current was observed at pH 4.25. Above pH 4.25, the current decreased and the current responses became very noisy over pH 6.5. The noisy current response might be caused by the formation of hydroxide precipitates of some metal ions. Since the composition of EDTA in solution was strongly influenced by pH, the pKa value of EDTA might change the completeness of the complexation reaction. The possible dissolution and release of EDTA from the EDTA-CPME surface into the solution at a pH over 6.5 may also make the EDTA-CPME surface rough. It was observed for all metal ions, and the pH of the preconcentration solution was selected as 4.3 for all subsequent measurements. Also, it is noted that a moderately acidic environment is satisfactory for the divalent heavy metal cations including Pb2+, Cu2+, and Hg2+ in EDTA titration.25 Interference Effects. The selectivity of this method for the determination of heavy metal ions was also evaluated for various heavy metal ions including Pb(II), Cu(II), and Hg(II) solutions at the preconcentration step. Other metal ions can interfere if they compete for complexation with EDTA attached on the polymer (25) Reilley, C. N.; Schmid, R. W. Anal. Chem. 1958, 30, 947.

Analytical Chemistry, Vol. 75, No. 5, March 1, 2003

1127

Figure 7. Effect of storage time of the Nafion-coated EDTA-CPME on the response of 1.0 × 10-7 M Pb(II) (cube), Cu(II) (circle), and Hg(II) ions (triangle) in sodium acetate/acetate buffer solution at pH 4.3 (deposition time 10 min at 30 °C).

film. EDTA is a kind of multidentate ligand that can form complexes with numbers of metal ions at different pHs. In the present work, interferences of other metal ions such as Cr(II), Mn(II), Fe(II), Co(II), Ni(II), Zn(II), Cd(II), and Ag(I) were studied at different pHs. The current responses of Pb(II), Cu(II), and Hg(II) ions were reduced ∼8% by the presence of 100-fold excess of other metal ions at all the pH ranges. However, equimolar concentration of other metal ions did not affect the simultaneous determination of Pb(II), Cu(II), and Hg(II) ions. Stability of the EDTA-CPME. The availability of the EDTACPME as a sensor was examined. After an analytical measurement, the EDTA-CPME was regenerated for its next use by potential cycling in the blank solution until no peaks were observed after spiking the electrode into a 0.1 M HNO3 solution for several seconds. However, by regenerating the surface of the modified electrode, the EDTA-CPME can be used only for a few days. These may be because of the loss of polymer and EDTA from the electrode surface to solutions. A change in film properties such as a decrease in conductivity may also affect the electroanalytical results.26 To improve the stability of the EDTA-CPME, 5 µL of 5% Nafion was used to coat the surface of the EDTA-CPME. Nafion coating over the EDTA-CPME slowed the dissolution or degradation of the poly-DATT and EDTA. In this case, although the sensitivity decreased, the stability of the electrode remarkably improved and thus the EDTA-CPME can be used for more than one month. Figure 7 shows the effect of storage time on the responses of the Nafion-coated EDTA-CPME. Up to six weeks, the sensitivity of the Nafion-coated EDTA-CPME retained 98% of initial sensitivity. The Nafion-coated EDTA-CPME showed an almost steady sensitivity during this time. After use for six weeks, the sensitivity gradually decreased and the response became insignificant, which is possibly due to the loss of EDTA from the electrode on exposure to solutions. Calibration Plot. Osteryoung square wave voltammetry was employed for the simultaneous analysis of trace Pb, Cu, and Hg. Figure 8a shows the square wave voltammograms recorded for various concentrations of Pb(II), Cu(II), and Hg(II) ions ranging from 5.0 × 10-10 to 1.0 × 10-8 M in a sodium acetate/acetic acid buffer solution of pH 4.3. To avoid the memory effect of the EDTACPME, the electrode was spiked in 0.1 M HNO3 for a few seconds (26) Karagozlar, A. E.; Ataman, O. Y.; Galal, A.; Xue, Z. L.; Zimmer, H.; Mark, H. B., Jr. Anal. Chim. Acta 1991, 248, 163-169.

1128 Analytical Chemistry, Vol. 75, No. 5, March 1, 2003

Figure 8. (a) Square wave voltammograms recorded using EDTACPME for (i) blank and (ii) 5.0 × 10-10, (iii) 7.5 × 10-10, (iv) 1.0 × 10-9, (v) 2.5 × 10-9, (vi) 5.0 × 10-9, and (vii) 1.0 × 10-8 M Pb(II), Cu(II), and Hg(II) ions in sodium acetate buffer solution of pH 4.3 (deposition time 10 min at 30 °C). (b) Calibration plots constructed for Cu (solid line), Hg (dash line), and Pb (dot line) from square wave voltammetric studies.

after each measurement and then potential cycling in the blank solution until no peaks were observed. Linear calibration plots (Figure 8b) were obtained in the concentration range from 5.0 × 10-10 to 1.0 × 10-7 M for Cu(II) and 7.5 × 10-10 to 1.0 × 10-7 M for Hg(II) and Pb(II) with 10-min preconcentration time. The relative standard deviations at 1.0 × 10-9 M Pb(II), Cu(II), and Hg(II) ion solution were found to be 5.0%, 4.5%, and 5.3%, respectively. These linear dependences gave the regression equations of ip (µA) ) -0.001 + (-0.003) [Pb(II)], ip (µA) ) -0.02 + (-0.005) [Cu(II)], and ip (1µA) ) -0.008 + (-0.003) [Hg(II)] with correlation coefficients of 0.999, 0.998, 0,999 for Pb, Cu, and Hg, respectively. The detection limits (DLs) were determined based on three times the standaed deviation of the blank noise (95% confidence level, k ) 3, n ) 5). The DL values for Pb(II), Cu(II), and Hg(II) ions were determined to be 6.0 × 10-10, 2.0 × 10-10, and 5.0 × 10-10 M, at 10-min deposition time. When 20-min deposition time was used, the detection limits of Pb, Cu, and Hg, were obtained as 4.0 × 10-10, 1.0 × 10-10, and 3.0 × 10-10 M, respectively. Linear calibration plots were also obtained for the Nafion-coated EDTA-CPME in the concentration range between

5.0 × 10-9 and 5.0 × 10-7 M with 10-min preconcentration under the same experimental condition. These linear dependences gave the regression equations of ip (µA) ) -0.0003 + (-0.07) [Pb(II)], ip (µA) ) -0.19 + (-0.24) [Cu(II)], and ip (µA) ) -0.025 + (-0.12) [Hg(II)] with correlation coefficients of 0.999, 0.999, and 0.999 for Pb, Cu, and Hg, respectively. The relative standard deviations at 1.0 × 10-8 M were of 6.3%, 5.2%, and 5.7% for Pb(II), Cu(II), and Hg(II) metal ions, respectively. The DLs for Pb(II), Cu(II), and Hg(II) ions were determined to be 4.0 × 10-9, 1.0 × 10-9, and 3.0 × 10-9 M for 10-min deposition time, which was ∼1 order of magnitude lower than the detection limits evaluated without Nafion coating. We previously studied27 the simultaneous determination of lead, copper, and mercury at a modified carbon paste electrode containing humic acid. The detection limits of lead, copper, and mercury were determined as 5.0 × 10-9, 8.0 × 10-8, and 8.0 × 10-9 M, with 20-min deposition, which was higher than the present work. Moreover, the deposition time used was 2 times higher than that of the present work. The low detection limits and stable surface of the EDTA-CPME in the present work clearly demonstrated that this method is very suitable for the analysis of Pb, Cu, and Hg in real samples. Standard Reference Material and Real Sample Analyses. To test the availability of the EDTA-CPME, an experiment for the simultaneous determination of Pb(II), Cu(II), and Hg(II) in the urine standard reference material (SRM 2670 freeze-dried urine) was carried out. The certified metal contents in this material (elevated level) were 109, 370, and 105 ppb for lead, copper, and mercury, respectively. Other metals present in the 2-300 ppb range included Al, As, Be, Cd, Cl, Ca, Au, Mg, Mn, Ni, Pt, K, Se, Na, and V. To avoid adsorption of organic species arising from the urine matrix, reference urine sample was boiled in concentrated nitric acid to destroy the organic species in the sample. After appropriate dilution, the pH of the solution was adjusted to the 4.3 with acetate buffer. The preconcentration step was the same as described before. Square wave voltammograms were recorded under the same experimental conditions and calibration plots were attained. The concentrations of Pb(II), Cu(II), and Hg(II) ions in the standard urine reference material were observed as 105 ( 6.5, 372 ( 7.7, and 102 ( 4.5 ppb, respectively. The obtained values showed good agreement with the certified values with good precision. The EDTA-CPME was also applied successfully for the simultaneous determination of Pb(II), Cu(II), and Hg(II) ions in tap water. Only Pb(II) and Cu(II) ions were detected as 0.48 ( 0.03 and 6.9 ( 0.4 ppb. To compare the results, the tap water was also analyzed using ICPMS. The results show that 0.5 and 7.5 ppb Pb(II) and Cu(II) were present in the tap water. (27) Jeong, E.-D.; Won, M.-S.; Shim, Y.-B. Electroanalysis 1994, 6, 887-893. (28) Lead and Copper Rule Minor Revision; EPA 815-F-899-010; U.S. Environmental Protection Agency, 1999.

Hg(II) ion was not detected by ICPMS. The results obtained by EDTA-CPME showed a good agreement with the results obtained by ICPMS. The Pb2+ content in the drinking water should be at least 10 times lower than the 15 ppb action level recommended by the U.S. Environmental Protection Agency.28 The EDTA-CPME developed in this work can detect Pb2+ content in tap water, which is ∼10 times lower than the EPA action level. CONCLUSION A new conducting polymer modified electrode based on EDTA functionalized via covalent bonding (EDTA-CPME) was constructed. The surface of the modified electrode was characterized by EQCM, ESCA, AES, SAM, and SEM, which shows that the EDTA is well bound on the poly-DATT film. The fuctionalized electrode was applied for the simultaneous determination of Pb(II), Cu(II), and Hg(II) ions in the certified and real samples. The dynamic ranges for the simultaneous determination were found as 7.5 × 10-10-1.0 × 10-7 and 5.0 × 10-9-1.0 × 10-7 M with the EDTA-CPME and the Nafion-coated EDTA-CPME, respectively. The detection limits for Pb(II), Cu(II), and Hg(II) ions were determined to be about 6.0 × 10-10, 2.0 × 10-10, and 5.0 × 10-10 M, respectively for 10-min deposition time without coating the EDTA-CPME surface with Nafion. With the Nafion-coated EDTACPME, the detection limits for Pb(II), Cu(II), and Hg(II) ions were determined to be about 4 × 10-9, 1.0 × 10-9, and 3 × 10-9 M for 10-min deposition time, which was ∼1 order of magnitude higher than the detection limits estimated without Nafion coating. The Nafion-coated EDTA-CPME exhibited higher stability than the uncoated EDTA-CPME. Rapid and convenient renewal of the electrode surface by oxidative removal of test ions allows the use of a single modified electrode surface in multiple analytical determinations for more than a month. The EDTA-CPME was applied to the simultaneous determination of Pb(II), Cu(II), and Hg(II) ions in SRM and in a real sample. Pb(II) and Cu(II) ions were detected as 0.48 ( 0.03 and 6.9 ( 0.4 ppb with the relative standard deviations of 6.2% and 5.8%, respectively, in the tap water and showed a good agreement with the results obtained with ICPMS. ACKNOWLEDGMENT This study was supported by “Center for Integrated Molecule Systems” through the Korean Science and Engineering Foundation (KOSEF). We are grateful to Prof. Sung-Chul Shin of the Chemistry Department of Gyeongsang National University for providing us the DATT monomer. M.A.R. gratefully acknowledges a postdoctoral fellowship from the center. Received for review November 7, 2002. Accepted January 9, 2003. AC0262917

Analytical Chemistry, Vol. 75, No. 5, March 1, 2003

1129