- Email: [email protected]
Determination of Hg2+ on poly(vinylferrocenium) (PVF+)-modified platinum electrode
Talanta 78 (2009) 405–409
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Determination of Hg2+ on poly(vinylferrocenium) (PVF+ )-modified platinum electrode Mutlu Sönmez C¸elebi, Haluk Özyörük, Attila Yıldız, Serdar Abacı ∗ Hacettepe University, Faculty of Science, Department of Chemistry, 06532 Ankara, Turkey
a r t i c l e
i n f o
Article history: Received 16 April 2008 Received in revised form 16 November 2008 Accepted 20 November 2008 Available online 27 November 2008 Keywords: Mercury determination PVF+ Anodic stripping Complexation Interference Aqueous solution
a b s t r a c t A new surface based on poly(vinylferrocenium) (PVF+ )-modified platinum electrode was developed for determination of Hg2+ ions in aqueous solutions. The polymer was electrodeposited on platinum electrode by constant potential electrolysis as PVF+ ClO4 − . Cl− ions were then attached to the polymer matrix by anion exchange and the modified electrode was dipped into Hg2+ solution. Hg2+ was preconcentrated at the polymer matrix by adsorption and also complexation reaction with Cl− . Detection of Hg2+ was carried out by differential pulse anodic stripping voltammetry (DPASV) after reduction of Hg2+ . Mercury ions as low as 5 × 10−10 M could be detected with the prepared electrode and the relative standard deviation was calculated as 6.35% at 1 × 10−6 M concentration (n = 6). Interferences of Ag+ , Pb2+ and Fe3+ ions were also studied at two different concentration ratios with respect to Hg2+ . The developed electrode was applied to the determination of Hg2+ in water samples. © 2008 Elsevier B.V. All rights reserved.
1. Introduction Mercury is one of the most well known toxic metals. The growing awareness of environmental mercury pollution and toxicity makes the determination of very low concentrations very important. Mercury is one of the few metals which strongly bioconcentrates and biomagnifies, has only harmful effects with no useful physiological functions when present in living organisms and easily transformed from a less toxic inorganic form to a more toxic organic form especially in fish [1]. It is usually present at low concentrations in environmental samples therefore, preconcentration of the metal is usually necessary to carry out a successful determination. Several methods can be used for mercury quantification such as cold vapor atomic absorption spectrometry (CV-AAS) [2], cold vapor atomic fluorescence spectrometry (CV-AFS) [3], inductively coupled plasma mass spectrometry (ICP-MS) [4] and, for relatively high concentrations, inductively coupled plasma atomic emission spectrometry (ICP-AES) [5]. All these techniques require expensive instrumentation and complicated sample preparation processes [6]. However, electrochemical methods are very good alternatives and provide easier and cheaper ways. Modification of the working electrode surfaces with an appropriate reagent offers analytical
∗ Corresponding author. Tel.: +90 312 2976080; fax: +90 312 2992163. E-mail address: [email protected] (S. Abacı). 0039-9140/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.talanta.2008.11.028
methods with enhanced selectivity and sensitivity. Modification can be achieved by electrochemical, chemical or physical methods. Stripping electrochemical methods represent an interesting alternative for mercury determination owing to their sensitivity, versatility and low costs [7]. Ju and Leech accumulated mercury ions on a metallothionein-modified gold electrode at open circuit and determined trace Hg2+ by cathodic stripping differential pulse voltammetry [8]. Berchmans et al. preconcentrated Hg2+ chemically using 2-mercaptobenzimidazole-modified gold electrode and determined the mercury ions by stripping voltammetry after reduction of preconcentrated Hg2+ prior to determination by anodic stripping voltammetry [9]. Electrochemical preconcentration of mercury ions has been carried out by Ugo et al. at gold- and polymer-coated electrodes [10]. Preconcentration and voltammetric determination of mercury has also been studied using chemically modified carbon-paste electrodes [11,12], graphite electrodes [13,14], glassy carbon electrodes [15,16] and sol–gel electrodes [17,18]. Poly(vinylferrocene) (PVF), is a redox polymer, which has long been used as a fundamental conducting polymer system, with the advantages of simple electrochemistry (a reversible one-electron process), high stability (allowing multiple measurements to be made over extended time scale), and the ease of deposition of thin films using a variety of methods [19]. The polymer oxidizes from methylene chloride to give the less soluble ferrocenium form of the polymer, which precipitates onto the electrode surface to give a PVF+ -modified electrode.
406
M. Sönmez C¸elebi et al. / Talanta 78 (2009) 405–409
In the present study, the use of PVF+ -modified Pt electrode for determination of Hg2+ ions by differential pulse anodic stripping voltammetry was investigated in terms of the parameters that can affect the analytical response. Possible interferences and the repeatability of the method were determined. 2. Experimental PVF was prepared by chemical polymerization of vinylferrocene (Alfa Product) at 70 ◦ C for 24 h using 2,2 -aso-bis(2-methylpropionitrile) (AIBN) (Alfa Product) as the initiator [20]. Methylene chloride (Aldrich), which was used for preparing polymer solutions, was washed with concentrated H2 SO4 (Merck), triple distilled water, Na2 CO3 (Merck) (5%), and triple distilled water. Then it was distilled over P2 O5 (Merck) [21]. 1 × 10−3 M stock solution of Hg2+ was prepared from HgCl2 (BDH, Analar) using triple distilled water. NaCl solutions were prepared using triple distilled water with NaCl (Carlo Erba). Ag+ , Fe3+ and Pb2+ solutions were prepared from AgNO3 (BDH, 99.8% pure), FeNO3 (Merck) and Pb(NO3 )2 (BDH, Analar) respectively Tetra-n-butyl ammonium perchlorate (TBAP) was used as the supporting electrolyte in the polymer solution. TBAP was obtained by the reaction of tetra-n-butyl ammonium hydroxide (40% aqueous solution, Merck) with perchloric acid (BDH) and recrystallised from the 1:9 mixture of water and ethyl alcohol by volume several times. It was then dried at 120 ◦ C under vacuum for 12 h. This salt was always kept under nitrogen atmosphere. NaCl (Carlo Erba) was used as the supporting electrolyte in the electrochemical experiments. The polymer solution and the NaCl solution that was used as the supporting electrolyte were deoxygenated by bubbling pure nitrogen gas (BOS). In electrochemical studies, a platinum (Pt) disc electrode (r = 0.05 cm) or Pt foil electrode (1.2 cm × 0.8 cm) was used as the working electrode. Before each experiment, the working electrode was polished with slurry of Cr2 O3 with water, then rinsed with triple distilled water, cleaned in ultrasonic bath and dried. Finally, the electrode was washed with the solvent that was used for the experiment. In the methylene chloride medium, a Ag/AgCl electrode was used as the reference electrode. The electrode was immersed in a separate compartment containing methylene chloride/0.1 M TBAP solution with a saturated amount of AgCl. In the electrochemical experiments that were carried out in methylene chloride medium, a Pt wire in separate compartment containing methylene chloride/0.1 M TBAP solution was used as the counter electrode. In aqueous medium, saturated calomel electrode (SCE) was used as the reference electrode and a Pt wire electrode with a surface area of 2 cm2 in spiral form was used as the counter electrode. The cyclic voltammetric and potential-controlled coulometric studies were carried out with PAR system, which consists of Model 175 Universal Programmer, Model 173 Potentiostat and Model 179 Digital Coulometer. The differential pulse anodic stripping voltammetric studies were performed with PAR Model 174A Polarographic Analyzer. Cyclic voltammograms and differential pulse anodic stripping voltammograms were recorded with EG&G PAR Model RE0150 X-Y recorder. The PVF+ ClO4 − film was electrodeposited on the electrode surface by the electrooxidation of 1.0 mg/mL PVF solution in methylene chloride containing 0.1 M TBAP at +0.7 V vs. Ag/AgCl. The uptake of ClO4 − as the counter anion to the polymer film was shown in a previous study by infrared spectroscopy [22]. The thicknesses of PVF+ ClO4 − films were controlled by the charge passed during the electroprecipitation. This charge was considered as an indication of polymer film thickness. A charge of 1 × 10−3 C corresponded to 1.32 × 10−6 moles of the oxidized
PVF/cm2 (dry thickness of ∼300 m [23], which corresponds to about 3 × 105 layers). 2.1. Procedure The procedure of preconcentration and voltammetric determination of Hg2+ ion on PVF+ Cl− -modified electrode includes the following steps: (1) coating the Pt electrode with PVF+ ClO4 − film; (2) immersion of the film in aqueous solution of Cl− ; (3) immersion of PVF+ Cl− film in aqueous solution of Hg2+ at open circuit; (4) electrochemical reduction of Hg2+ ions prior to determination; (5) determination of Hg2+ ions by DPASV. 3. Results and discussion The voltammetric behavior of PVF+ ClO4 − -coated Pt electrode is shown in Fig. 1. No appreciable current and potential change for oxidative and reductive behaviors was recorded in this voltammogram with time and surface was determined as stable. PVF+ ClO4 − -coated Pt electrode was dipped into a solution which contained 1 × 10−4 M Hg2+ for 15 min, then, removed, washed and immersed into a blank solution (a solution only contained 0.1 M NaCl). A potential of −0.2 V vs. SCE electrode, which is beyond reduction potential of Hg2+ to metallic Hg, was applied for 5 min for reducing presumably immobilized Hg2+ . Then, differential pulse anodic stripping voltammogram (DPASV) in blank solution was recorded (Fig. 2b). As can be seen, immobilized mercury stripped around at −0.05 V vs. SCE. This result indicated that some of the Hg2+ from the solution was adsorbed on PVF+ ClO4 − matrix possibly due to electrostatic attraction. However, when PVF+ ClO4 − -coated Pt electrode was initially immersed into 5 × 10−2 M Cl− ion for 15 min before other processes described above was applied, stripping peak current considerably increased (Fig. 2c). Voltammogram of the PVF+ ClO4 − -coated electrode was also recorded (Fig. 2a). This result was explained as follows. PVF+ ClO4 − matrix is sensitive to the anions present in the solution. Since the Cl− ions are negatively charged. The electrostatic immobilization of Cl− ions on PVF+ ClO4 − -coated Pt electrode was accomplished via anion exchange. Complexation reaction occurred between Hg2+ and Cl− ions, thus, deposited amount of Hg2+ on electrode surface, therefore, stripping peak of mercury enhanced. The immobilization and detection mechanism can be proposed as following: PVF+ ClO4 − + Cl− → PVF+ Cl− + ClO4 −
(anion exchange)
(1)
Fig. 1. Cyclic voltammetric behavior of PVF+ ClO4 − -coated Pt disc electrode in 0.1 M NaCl solution. v = 100 mV/s; A = 7.85 × 10−3 cm2 .
M. Sönmez C¸elebi et al. / Talanta 78 (2009) 405–409
407
Fig. 3. The effect of Cl− ion concentration on the oxidation peak current for 1 × 10−4 M Hg2+ (1.0 mC film thickness, 15 min immersion time in Cl− ion solution, 15 min preconcentration time, −0.2 V electrolysis potential, 5 min electrolysis time).
3.2. The effect of immersion time in Cl− solution The immersion time in Cl− ion solution was considered as an important parameter and its effect was studied. Fig. 4 shows the change of stripping peak currents with the immersion time of the modified electrode in Cl− solution. As can be seen, stripping peak currents showed an increase up to 15 min immersion time and decreased after this value. The result supported our claim about the damaging effect of excess Cl− ions deposited on the surface. Therefore, 15 min immersion time seemed the optimum immersion time. 3.3. The effect of polymeric film thickness
Hg2+ + PVF+ Cl− → PVF+ Cl− (Hg2+ ) (complexation)
(2)
PVF+ Cl− (Hg2+ ) → PVF+ Cl− (Hg)
(reduction)
(3)
The effect of polymeric film thickness on the stripping peak current was investigated by varying the charge passed during the electrooxidation of PVF to PVF+ and keeping the other variables constant (Fig. 5). When the thickness of PVF+ ClO4 − or the amount of PVF+ increased, accumulated amount of Cl− also increased and this caused stripping peak current to enhance due to higher amount of Hg2+ –Cl− complex formation on the surface. However, after 0.8 mC film thicknesses, stripping peak current started to decrease with increase in film thickness due to excess Cl− accumulation. This is also another indication of damaging effect of excess Cl− which was described above. The decrease in the stripping peak current is also attributed to the change in the porosity of the polymer matrix at elevated film thicknesses.
(stripping)
(4)
3.4. The influence of preconcentration time
Fig. 2. Differential pulse anodic stripping voltammogram of (a) PVF+ ClO4 − film corresponding to 0.8 mC film thickness, (b) PVF+ ClO4 − film immersed in 1 × 10−4 M Hg2+ solution for 15 min, (c) PVF+ ClO4 − film immersed in 50 mM Cl− solution for 15 min, then in 1 × 10−4 M Hg2+ solution for 15 min, after 5 min cathodic electrolysis at −0.2 V vs. SCE. v = 5 mV/s; mod. amp. = 100 mV; A = 7.85 × 10−3 cm2 .
+
−
+
−
PVF Cl (Hg) → PVF Cl (Hg
2+
)
PVF+ ClO4 − -coated
These results and mechanism indicated that Pt electrode can be used for Hg2+ detection in aqueous solutions. However, several parameters had to be evaluated to get the optimum performance from this surface.
As can be determined from Fig. 6, the stripping peak current shows a steady increase up to 10 min preconcentration time and
3.1. The effect of Cl− ion concentration Fig. 3 shows the effect of Cl− ion concentration on the preconcentration of Hg2+ ions in the polymeric matrix. As can be seen, stripping peak current increased with the increase of Cl− ion concentration. This was an expected result according to proposed mechanism because increased concentration of Cl− leads to higher amounts of Hg2+ –Cl− complex on the surface. However, when Cl− ion concentration exceeded 40 mM, stripping peak current started to decrease and the rupture of polymer films were visually observed when the concentration of Cl− ion was 80 mM. This was a clear indication of damaging effect of excess Cl− on the surface since it initiated pitting and that caused the rupture of the film.
Fig. 4. The effect of immersion time in Cl− ion solution on the stripping peak current for 1 × 10−4 M Hg2+ (1.0 mC film thickness, 40 mM Cl− ion concentration, 15 min preconcentration time, −0.2 V electrolysis potential, 5 min electrolysis time).
408
M. Sönmez C¸elebi et al. / Talanta 78 (2009) 405–409 Table 1 Equation of calibration curves and R2 values for three concentration intervals. Concentration range (M) −3
−5
1 × 10 to 5 × 10 1 × 10−5 to 1 × 10−6 1 × 10−7 to 5 × 10−10
Equation of the calibration curve
R2
y = 1884.1x + 8147 y = 3.7627x + 23.16 y = 0.1482x + 1.3908
0.9989 0.9900 0.9832
Table 2 Interferences of some ions to the response of the electrode in 1:1 and 1:10 concentration ratios with respect to Hg2+ . Interfering ion Fig. 5. The effect of polymeric film thickness on the stripping peak current for 1 × 10−4 M Hg2+ (50 mM Cl− ion concentration, 15 min immersion time in Cl− ion solution, 15 min preconcentration time, −0.2 V electrolysis potential, 5 min electrolysis time).
remains constant after this time because of diffusion limitation. During evaluation of performance of the electrode, 5 min preconcentration time was used in order to assure that the electrode was not saturated with metal ions.
% Interference
Ag+ Fe3+ Pb2+
1:1
1:10
91.8 52.2 0.472
92.1 92.3 80.1
4 min, formed neutral mercury metal might have been released from the polymer matrix, consequently, stripping peak currents decreased. Thus, optimum electrolysis time was considered to be 4 min.
3.5. The influence of reduction potential 3.7. Repeatability, linearity and interferences Various potentials were applied to for reduction of Hg2+ to Hg. The peak current showed a maximum at −0.3 V vs. SCE and decreased after this potential. Two processes occur during the electrolysis of the modified electrode: (a) reduction of the metal ion in the polymer matrix and (b) reduction of PVF+ to PVF. The latter is an undesirable process because, when the amount of PVF+ in the polymer matrix decreases, the amount of the negatively charged complex species also decreases. It was thought that, up to −0.3 V, as the reduction of the metal ion species is dominant, the stripping peak current increased. After this potential, because of the decrease at the stripping peak current, the second process was considered to be dominant. Therefore, −0.3 V seemed to be the optimum electrolysis potential. 3.6. The influence of electrolysis time The change in stripping peak current with electrolysis time was studied to obtain the optimum time needed for reduction of Hg2+ ions. −0.3 V is the optimum electrolysis potential for the reduction of Hg2+ ions to Hg metal. Therefore, electrolytic reduction of the Hg2+ is the dominant process at this potential. Stripping peak current increased with increasing electrolysis time up to 4 min. However, when the electrolysis time exceeded
The performance of the method was evaluated with Pt foil electrode (1.2 cm × 0.8 cm) using the following conditions: 70 mC polymeric film thicknesses, 40 mM Cl− ion concentration, 15 min immersion time in Cl− solution, 5 min preconcentration time, 4 min electrolysis at −0.3 V vs. SCE. The repeatability was evaluated with 1 × 10−6 M Hg2+ solution for six observations. The relative standard deviation value was calculated as 6.35%. Linearity of the electrode was investigated by starting with 1 × 10−3 M Hg2+ concentration in immobilization solution and gradually lowering the Hg2+ concentration and taking the stripping peak current of Hg as basis. Concentration of Hg2+ solutions was calculated by calibration curve method. There was linearity up to a concentration value of 5 × 10−5 M. After this value, slope of the curve changed and linearity was observed between 1 × 10−5 M and 1 × 10−6 M. After 1 × 10−6 M Hg2+ concentration, a greater sized platinum electrode was used to better detect the low amounts of mercury and linearity was obtained concentration interval 1 × 10−7 M and 5 × 10−10 . Table 1 summarizes the equation of the calibration curve and R2 for three concentration intervals. No appreciable change in peak currents was observed for Hg2+ concentrations lower than 5 × 10−10 M. Influences of interfering metal ions which are capable of forming chloride complexes on the response of the electrode were examined. The voltammogram of 3 × 10−3 M Hg2+ solution was recorded in the presence of Ag+ , Fe3+ and Pb2+ ions in 1:1 and 1:10 con-
Table 3 Determination of Hg2+ ions in water samples.
Fig. 6. The effect of preconcentration time on the stripping peak current for 1 × 10−4 M Hg2+ (0.8 mC film thickness, 40 mM Cl− ion concentration, 15 min immersion time in Cl− ion solution, −0.2 V electrolysis potential, 5 min electrolysis time).
Sample
Hg2+ added (ng mL−1 )
Hg2+ found (ng mL−1 )
% Recovery
Tap water
– 40
n.d.* 43.71 ± 0.15**
– 109.3
Natural spring water
– 40
n.d. 40.60 ± 0.18
– 101.5
* **
Not detected. Mean value ± standard deviation (n = 3).
M. Sönmez C¸elebi et al. / Talanta 78 (2009) 405–409
centration ratios with respect to Hg2+ . Interferences are given in Table 2. 3.8. Determination of mercury in water samples The validity of the proposed method was tested in water samples by spiking known concentrations of mercury to water samples [24]. The results are summarized in Table 3. 4. Conclusions PVF+ -coated Pt surface has been shown to be a good choice for determining Hg2+ from aqueous solutions. The initial immersion of surface into Cl− solution increased the accumulation of Hg2+ on the surface due to formation of Hg2+ –Cl− complex. However, excess amounts of Cl− on the surface caused pitting and rupture of the surface film. Detection mechanism was proposed. The reproducibility and sensitivity was satisfactory. Interferences from ions that might also form complexes with Cl− were examined and reported. The validity of method was tested in water samples. It can be claimed that these surfaces are very good alternative to expensive analytical techniques such as atomic absorption spectrometry and ICP-MS. Another advantage of the method between electrochemical techniques is that, preparation of the surface and application of the method is simple and also cheaper than more costly materials such as gold.
409
References [1] R.J. Irwin, M. VanMouwerik, L. Stevens, M.D. Seese, W. Basham, Environmental Contaminants Encyclopedia, National Park Service, Water Resources Division, Fort Collins, Colorado, 1997, p. 7. [2] S.C. Hight, J. Cheng, Food Chem. 91 (2005) 557. [3] M. Roulet, M. Lucotte, J.R.D. Guimarães, I. Rheault, Sci. Total Environ. 261 (2000) 43. [4] P. Ugo, S. Zamperi, L.M. Moretto, D. Paolucci, Anal. Chim. Acta 434 (2001) 191. [5] A.C. Barbarosa, G.A. East, Ecol. Trace Element Res. 60 (1997) 153. [6] C.C. Huang, H.T. Chan, Anal. Chem. 78 (2006) 8332. [7] A. Giacomino, O. Abollino, M. Malandrino, E. Mentasti, Talanta 75 (2008) 266. [8] H. Ju, D.J. Leech, Electroanal. Chem. 484 (2000) 150. [9] S. Berchmans, S. Arivukkodi, V. Yegnaraman, Electrochem. Commun. 2 (2000) 226. [10] P. Ugo, S. Zampieri, L.M. Moretto, D. Paolucci, Anal. Chim. Acta 434 (2001) 291. [11] R. Agraz, M.T. Sevilla, L. Hernandez, J. Electroanal. Chem. 390 (1995) 47. [12] J. Wang, M. Bonakdar, Talanta 35 (1988) 277. [13] C. Faller, N.Y. Stojko, G. Henze, K.Z. Brainina, Anal. Chim. Acta 396 (1999) 195. [14] J. Labuda, V. Plaskon, Anal. Chim. Acta 228 (1990) 259. [15] P. Ugo, L.M. Moretto, G.A. Mazzocchin, Anal. Chim. Acta 305 (1995) 74. [16] J. Lu, X. He, X. Zeng, Q. Wan, Z. Zhang, Talanta 59 (2003) 553. [17] Y. Guo, A.R. Guadalupe, J. Pharmaceut. Biomed. 19 (1999) 175. [18] G. Cabello-Carramolino, M.D. Petit-Dominguez, Anal. Chim. Acta 614 (2008) 103. [19] L. Yu, M. Sathe, X. Zeng, J. Electrochem. Soc. 152 (2005) E357. [20] C. Aso, T. Kunitake, T. Nakashima, Macromol. Chem. 124 (1969) 232. [21] D.D. Perrin, W.L.F. Armarego, D.R. Perrin, Purification of Laboratory Chemicals, Pergamon Press, 1980. [22] H. Gülce, H. Özyörük, A. Yıldız, Electroanalysis 7 (1993) 178. [23] P.J. Peerce, A.J. Bard, J. Electroanal. Chem. 112 (1980) 97. [24] N. Rajesh, M.S. Hari, Spectrochim. Acta Part A 70 (2008) 1104.
Contents lists available at ScienceDirect
Talanta journal homepage: www.elsevier.com/locate/talanta
Determination of Hg2+ on poly(vinylferrocenium) (PVF+ )-modified platinum electrode Mutlu Sönmez C¸elebi, Haluk Özyörük, Attila Yıldız, Serdar Abacı ∗ Hacettepe University, Faculty of Science, Department of Chemistry, 06532 Ankara, Turkey
a r t i c l e
i n f o
Article history: Received 16 April 2008 Received in revised form 16 November 2008 Accepted 20 November 2008 Available online 27 November 2008 Keywords: Mercury determination PVF+ Anodic stripping Complexation Interference Aqueous solution
a b s t r a c t A new surface based on poly(vinylferrocenium) (PVF+ )-modified platinum electrode was developed for determination of Hg2+ ions in aqueous solutions. The polymer was electrodeposited on platinum electrode by constant potential electrolysis as PVF+ ClO4 − . Cl− ions were then attached to the polymer matrix by anion exchange and the modified electrode was dipped into Hg2+ solution. Hg2+ was preconcentrated at the polymer matrix by adsorption and also complexation reaction with Cl− . Detection of Hg2+ was carried out by differential pulse anodic stripping voltammetry (DPASV) after reduction of Hg2+ . Mercury ions as low as 5 × 10−10 M could be detected with the prepared electrode and the relative standard deviation was calculated as 6.35% at 1 × 10−6 M concentration (n = 6). Interferences of Ag+ , Pb2+ and Fe3+ ions were also studied at two different concentration ratios with respect to Hg2+ . The developed electrode was applied to the determination of Hg2+ in water samples. © 2008 Elsevier B.V. All rights reserved.
1. Introduction Mercury is one of the most well known toxic metals. The growing awareness of environmental mercury pollution and toxicity makes the determination of very low concentrations very important. Mercury is one of the few metals which strongly bioconcentrates and biomagnifies, has only harmful effects with no useful physiological functions when present in living organisms and easily transformed from a less toxic inorganic form to a more toxic organic form especially in fish [1]. It is usually present at low concentrations in environmental samples therefore, preconcentration of the metal is usually necessary to carry out a successful determination. Several methods can be used for mercury quantification such as cold vapor atomic absorption spectrometry (CV-AAS) [2], cold vapor atomic fluorescence spectrometry (CV-AFS) [3], inductively coupled plasma mass spectrometry (ICP-MS) [4] and, for relatively high concentrations, inductively coupled plasma atomic emission spectrometry (ICP-AES) [5]. All these techniques require expensive instrumentation and complicated sample preparation processes [6]. However, electrochemical methods are very good alternatives and provide easier and cheaper ways. Modification of the working electrode surfaces with an appropriate reagent offers analytical
∗ Corresponding author. Tel.: +90 312 2976080; fax: +90 312 2992163. E-mail address: [email protected] (S. Abacı). 0039-9140/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.talanta.2008.11.028
methods with enhanced selectivity and sensitivity. Modification can be achieved by electrochemical, chemical or physical methods. Stripping electrochemical methods represent an interesting alternative for mercury determination owing to their sensitivity, versatility and low costs [7]. Ju and Leech accumulated mercury ions on a metallothionein-modified gold electrode at open circuit and determined trace Hg2+ by cathodic stripping differential pulse voltammetry [8]. Berchmans et al. preconcentrated Hg2+ chemically using 2-mercaptobenzimidazole-modified gold electrode and determined the mercury ions by stripping voltammetry after reduction of preconcentrated Hg2+ prior to determination by anodic stripping voltammetry [9]. Electrochemical preconcentration of mercury ions has been carried out by Ugo et al. at gold- and polymer-coated electrodes [10]. Preconcentration and voltammetric determination of mercury has also been studied using chemically modified carbon-paste electrodes [11,12], graphite electrodes [13,14], glassy carbon electrodes [15,16] and sol–gel electrodes [17,18]. Poly(vinylferrocene) (PVF), is a redox polymer, which has long been used as a fundamental conducting polymer system, with the advantages of simple electrochemistry (a reversible one-electron process), high stability (allowing multiple measurements to be made over extended time scale), and the ease of deposition of thin films using a variety of methods [19]. The polymer oxidizes from methylene chloride to give the less soluble ferrocenium form of the polymer, which precipitates onto the electrode surface to give a PVF+ -modified electrode.
406
M. Sönmez C¸elebi et al. / Talanta 78 (2009) 405–409
In the present study, the use of PVF+ -modified Pt electrode for determination of Hg2+ ions by differential pulse anodic stripping voltammetry was investigated in terms of the parameters that can affect the analytical response. Possible interferences and the repeatability of the method were determined. 2. Experimental PVF was prepared by chemical polymerization of vinylferrocene (Alfa Product) at 70 ◦ C for 24 h using 2,2 -aso-bis(2-methylpropionitrile) (AIBN) (Alfa Product) as the initiator [20]. Methylene chloride (Aldrich), which was used for preparing polymer solutions, was washed with concentrated H2 SO4 (Merck), triple distilled water, Na2 CO3 (Merck) (5%), and triple distilled water. Then it was distilled over P2 O5 (Merck) [21]. 1 × 10−3 M stock solution of Hg2+ was prepared from HgCl2 (BDH, Analar) using triple distilled water. NaCl solutions were prepared using triple distilled water with NaCl (Carlo Erba). Ag+ , Fe3+ and Pb2+ solutions were prepared from AgNO3 (BDH, 99.8% pure), FeNO3 (Merck) and Pb(NO3 )2 (BDH, Analar) respectively Tetra-n-butyl ammonium perchlorate (TBAP) was used as the supporting electrolyte in the polymer solution. TBAP was obtained by the reaction of tetra-n-butyl ammonium hydroxide (40% aqueous solution, Merck) with perchloric acid (BDH) and recrystallised from the 1:9 mixture of water and ethyl alcohol by volume several times. It was then dried at 120 ◦ C under vacuum for 12 h. This salt was always kept under nitrogen atmosphere. NaCl (Carlo Erba) was used as the supporting electrolyte in the electrochemical experiments. The polymer solution and the NaCl solution that was used as the supporting electrolyte were deoxygenated by bubbling pure nitrogen gas (BOS). In electrochemical studies, a platinum (Pt) disc electrode (r = 0.05 cm) or Pt foil electrode (1.2 cm × 0.8 cm) was used as the working electrode. Before each experiment, the working electrode was polished with slurry of Cr2 O3 with water, then rinsed with triple distilled water, cleaned in ultrasonic bath and dried. Finally, the electrode was washed with the solvent that was used for the experiment. In the methylene chloride medium, a Ag/AgCl electrode was used as the reference electrode. The electrode was immersed in a separate compartment containing methylene chloride/0.1 M TBAP solution with a saturated amount of AgCl. In the electrochemical experiments that were carried out in methylene chloride medium, a Pt wire in separate compartment containing methylene chloride/0.1 M TBAP solution was used as the counter electrode. In aqueous medium, saturated calomel electrode (SCE) was used as the reference electrode and a Pt wire electrode with a surface area of 2 cm2 in spiral form was used as the counter electrode. The cyclic voltammetric and potential-controlled coulometric studies were carried out with PAR system, which consists of Model 175 Universal Programmer, Model 173 Potentiostat and Model 179 Digital Coulometer. The differential pulse anodic stripping voltammetric studies were performed with PAR Model 174A Polarographic Analyzer. Cyclic voltammograms and differential pulse anodic stripping voltammograms were recorded with EG&G PAR Model RE0150 X-Y recorder. The PVF+ ClO4 − film was electrodeposited on the electrode surface by the electrooxidation of 1.0 mg/mL PVF solution in methylene chloride containing 0.1 M TBAP at +0.7 V vs. Ag/AgCl. The uptake of ClO4 − as the counter anion to the polymer film was shown in a previous study by infrared spectroscopy [22]. The thicknesses of PVF+ ClO4 − films were controlled by the charge passed during the electroprecipitation. This charge was considered as an indication of polymer film thickness. A charge of 1 × 10−3 C corresponded to 1.32 × 10−6 moles of the oxidized
PVF/cm2 (dry thickness of ∼300 m [23], which corresponds to about 3 × 105 layers). 2.1. Procedure The procedure of preconcentration and voltammetric determination of Hg2+ ion on PVF+ Cl− -modified electrode includes the following steps: (1) coating the Pt electrode with PVF+ ClO4 − film; (2) immersion of the film in aqueous solution of Cl− ; (3) immersion of PVF+ Cl− film in aqueous solution of Hg2+ at open circuit; (4) electrochemical reduction of Hg2+ ions prior to determination; (5) determination of Hg2+ ions by DPASV. 3. Results and discussion The voltammetric behavior of PVF+ ClO4 − -coated Pt electrode is shown in Fig. 1. No appreciable current and potential change for oxidative and reductive behaviors was recorded in this voltammogram with time and surface was determined as stable. PVF+ ClO4 − -coated Pt electrode was dipped into a solution which contained 1 × 10−4 M Hg2+ for 15 min, then, removed, washed and immersed into a blank solution (a solution only contained 0.1 M NaCl). A potential of −0.2 V vs. SCE electrode, which is beyond reduction potential of Hg2+ to metallic Hg, was applied for 5 min for reducing presumably immobilized Hg2+ . Then, differential pulse anodic stripping voltammogram (DPASV) in blank solution was recorded (Fig. 2b). As can be seen, immobilized mercury stripped around at −0.05 V vs. SCE. This result indicated that some of the Hg2+ from the solution was adsorbed on PVF+ ClO4 − matrix possibly due to electrostatic attraction. However, when PVF+ ClO4 − -coated Pt electrode was initially immersed into 5 × 10−2 M Cl− ion for 15 min before other processes described above was applied, stripping peak current considerably increased (Fig. 2c). Voltammogram of the PVF+ ClO4 − -coated electrode was also recorded (Fig. 2a). This result was explained as follows. PVF+ ClO4 − matrix is sensitive to the anions present in the solution. Since the Cl− ions are negatively charged. The electrostatic immobilization of Cl− ions on PVF+ ClO4 − -coated Pt electrode was accomplished via anion exchange. Complexation reaction occurred between Hg2+ and Cl− ions, thus, deposited amount of Hg2+ on electrode surface, therefore, stripping peak of mercury enhanced. The immobilization and detection mechanism can be proposed as following: PVF+ ClO4 − + Cl− → PVF+ Cl− + ClO4 −
(anion exchange)
(1)
Fig. 1. Cyclic voltammetric behavior of PVF+ ClO4 − -coated Pt disc electrode in 0.1 M NaCl solution. v = 100 mV/s; A = 7.85 × 10−3 cm2 .
M. Sönmez C¸elebi et al. / Talanta 78 (2009) 405–409
407
Fig. 3. The effect of Cl− ion concentration on the oxidation peak current for 1 × 10−4 M Hg2+ (1.0 mC film thickness, 15 min immersion time in Cl− ion solution, 15 min preconcentration time, −0.2 V electrolysis potential, 5 min electrolysis time).
3.2. The effect of immersion time in Cl− solution The immersion time in Cl− ion solution was considered as an important parameter and its effect was studied. Fig. 4 shows the change of stripping peak currents with the immersion time of the modified electrode in Cl− solution. As can be seen, stripping peak currents showed an increase up to 15 min immersion time and decreased after this value. The result supported our claim about the damaging effect of excess Cl− ions deposited on the surface. Therefore, 15 min immersion time seemed the optimum immersion time. 3.3. The effect of polymeric film thickness
Hg2+ + PVF+ Cl− → PVF+ Cl− (Hg2+ ) (complexation)
(2)
PVF+ Cl− (Hg2+ ) → PVF+ Cl− (Hg)
(reduction)
(3)
The effect of polymeric film thickness on the stripping peak current was investigated by varying the charge passed during the electrooxidation of PVF to PVF+ and keeping the other variables constant (Fig. 5). When the thickness of PVF+ ClO4 − or the amount of PVF+ increased, accumulated amount of Cl− also increased and this caused stripping peak current to enhance due to higher amount of Hg2+ –Cl− complex formation on the surface. However, after 0.8 mC film thicknesses, stripping peak current started to decrease with increase in film thickness due to excess Cl− accumulation. This is also another indication of damaging effect of excess Cl− which was described above. The decrease in the stripping peak current is also attributed to the change in the porosity of the polymer matrix at elevated film thicknesses.
(stripping)
(4)
3.4. The influence of preconcentration time
Fig. 2. Differential pulse anodic stripping voltammogram of (a) PVF+ ClO4 − film corresponding to 0.8 mC film thickness, (b) PVF+ ClO4 − film immersed in 1 × 10−4 M Hg2+ solution for 15 min, (c) PVF+ ClO4 − film immersed in 50 mM Cl− solution for 15 min, then in 1 × 10−4 M Hg2+ solution for 15 min, after 5 min cathodic electrolysis at −0.2 V vs. SCE. v = 5 mV/s; mod. amp. = 100 mV; A = 7.85 × 10−3 cm2 .
+
−
+
−
PVF Cl (Hg) → PVF Cl (Hg
2+
)
PVF+ ClO4 − -coated
These results and mechanism indicated that Pt electrode can be used for Hg2+ detection in aqueous solutions. However, several parameters had to be evaluated to get the optimum performance from this surface.
As can be determined from Fig. 6, the stripping peak current shows a steady increase up to 10 min preconcentration time and
3.1. The effect of Cl− ion concentration Fig. 3 shows the effect of Cl− ion concentration on the preconcentration of Hg2+ ions in the polymeric matrix. As can be seen, stripping peak current increased with the increase of Cl− ion concentration. This was an expected result according to proposed mechanism because increased concentration of Cl− leads to higher amounts of Hg2+ –Cl− complex on the surface. However, when Cl− ion concentration exceeded 40 mM, stripping peak current started to decrease and the rupture of polymer films were visually observed when the concentration of Cl− ion was 80 mM. This was a clear indication of damaging effect of excess Cl− on the surface since it initiated pitting and that caused the rupture of the film.
Fig. 4. The effect of immersion time in Cl− ion solution on the stripping peak current for 1 × 10−4 M Hg2+ (1.0 mC film thickness, 40 mM Cl− ion concentration, 15 min preconcentration time, −0.2 V electrolysis potential, 5 min electrolysis time).
408
M. Sönmez C¸elebi et al. / Talanta 78 (2009) 405–409 Table 1 Equation of calibration curves and R2 values for three concentration intervals. Concentration range (M) −3
−5
1 × 10 to 5 × 10 1 × 10−5 to 1 × 10−6 1 × 10−7 to 5 × 10−10
Equation of the calibration curve
R2
y = 1884.1x + 8147 y = 3.7627x + 23.16 y = 0.1482x + 1.3908
0.9989 0.9900 0.9832
Table 2 Interferences of some ions to the response of the electrode in 1:1 and 1:10 concentration ratios with respect to Hg2+ . Interfering ion Fig. 5. The effect of polymeric film thickness on the stripping peak current for 1 × 10−4 M Hg2+ (50 mM Cl− ion concentration, 15 min immersion time in Cl− ion solution, 15 min preconcentration time, −0.2 V electrolysis potential, 5 min electrolysis time).
remains constant after this time because of diffusion limitation. During evaluation of performance of the electrode, 5 min preconcentration time was used in order to assure that the electrode was not saturated with metal ions.
% Interference
Ag+ Fe3+ Pb2+
1:1
1:10
91.8 52.2 0.472
92.1 92.3 80.1
4 min, formed neutral mercury metal might have been released from the polymer matrix, consequently, stripping peak currents decreased. Thus, optimum electrolysis time was considered to be 4 min.
3.5. The influence of reduction potential 3.7. Repeatability, linearity and interferences Various potentials were applied to for reduction of Hg2+ to Hg. The peak current showed a maximum at −0.3 V vs. SCE and decreased after this potential. Two processes occur during the electrolysis of the modified electrode: (a) reduction of the metal ion in the polymer matrix and (b) reduction of PVF+ to PVF. The latter is an undesirable process because, when the amount of PVF+ in the polymer matrix decreases, the amount of the negatively charged complex species also decreases. It was thought that, up to −0.3 V, as the reduction of the metal ion species is dominant, the stripping peak current increased. After this potential, because of the decrease at the stripping peak current, the second process was considered to be dominant. Therefore, −0.3 V seemed to be the optimum electrolysis potential. 3.6. The influence of electrolysis time The change in stripping peak current with electrolysis time was studied to obtain the optimum time needed for reduction of Hg2+ ions. −0.3 V is the optimum electrolysis potential for the reduction of Hg2+ ions to Hg metal. Therefore, electrolytic reduction of the Hg2+ is the dominant process at this potential. Stripping peak current increased with increasing electrolysis time up to 4 min. However, when the electrolysis time exceeded
The performance of the method was evaluated with Pt foil electrode (1.2 cm × 0.8 cm) using the following conditions: 70 mC polymeric film thicknesses, 40 mM Cl− ion concentration, 15 min immersion time in Cl− solution, 5 min preconcentration time, 4 min electrolysis at −0.3 V vs. SCE. The repeatability was evaluated with 1 × 10−6 M Hg2+ solution for six observations. The relative standard deviation value was calculated as 6.35%. Linearity of the electrode was investigated by starting with 1 × 10−3 M Hg2+ concentration in immobilization solution and gradually lowering the Hg2+ concentration and taking the stripping peak current of Hg as basis. Concentration of Hg2+ solutions was calculated by calibration curve method. There was linearity up to a concentration value of 5 × 10−5 M. After this value, slope of the curve changed and linearity was observed between 1 × 10−5 M and 1 × 10−6 M. After 1 × 10−6 M Hg2+ concentration, a greater sized platinum electrode was used to better detect the low amounts of mercury and linearity was obtained concentration interval 1 × 10−7 M and 5 × 10−10 . Table 1 summarizes the equation of the calibration curve and R2 for three concentration intervals. No appreciable change in peak currents was observed for Hg2+ concentrations lower than 5 × 10−10 M. Influences of interfering metal ions which are capable of forming chloride complexes on the response of the electrode were examined. The voltammogram of 3 × 10−3 M Hg2+ solution was recorded in the presence of Ag+ , Fe3+ and Pb2+ ions in 1:1 and 1:10 con-
Table 3 Determination of Hg2+ ions in water samples.
Fig. 6. The effect of preconcentration time on the stripping peak current for 1 × 10−4 M Hg2+ (0.8 mC film thickness, 40 mM Cl− ion concentration, 15 min immersion time in Cl− ion solution, −0.2 V electrolysis potential, 5 min electrolysis time).
Sample
Hg2+ added (ng mL−1 )
Hg2+ found (ng mL−1 )
% Recovery
Tap water
– 40
n.d.* 43.71 ± 0.15**
– 109.3
Natural spring water
– 40
n.d. 40.60 ± 0.18
– 101.5
* **
Not detected. Mean value ± standard deviation (n = 3).
M. Sönmez C¸elebi et al. / Talanta 78 (2009) 405–409
centration ratios with respect to Hg2+ . Interferences are given in Table 2. 3.8. Determination of mercury in water samples The validity of the proposed method was tested in water samples by spiking known concentrations of mercury to water samples [24]. The results are summarized in Table 3. 4. Conclusions PVF+ -coated Pt surface has been shown to be a good choice for determining Hg2+ from aqueous solutions. The initial immersion of surface into Cl− solution increased the accumulation of Hg2+ on the surface due to formation of Hg2+ –Cl− complex. However, excess amounts of Cl− on the surface caused pitting and rupture of the surface film. Detection mechanism was proposed. The reproducibility and sensitivity was satisfactory. Interferences from ions that might also form complexes with Cl− were examined and reported. The validity of method was tested in water samples. It can be claimed that these surfaces are very good alternative to expensive analytical techniques such as atomic absorption spectrometry and ICP-MS. Another advantage of the method between electrochemical techniques is that, preparation of the surface and application of the method is simple and also cheaper than more costly materials such as gold.
409
References [1] R.J. Irwin, M. VanMouwerik, L. Stevens, M.D. Seese, W. Basham, Environmental Contaminants Encyclopedia, National Park Service, Water Resources Division, Fort Collins, Colorado, 1997, p. 7. [2] S.C. Hight, J. Cheng, Food Chem. 91 (2005) 557. [3] M. Roulet, M. Lucotte, J.R.D. Guimarães, I. Rheault, Sci. Total Environ. 261 (2000) 43. [4] P. Ugo, S. Zamperi, L.M. Moretto, D. Paolucci, Anal. Chim. Acta 434 (2001) 191. [5] A.C. Barbarosa, G.A. East, Ecol. Trace Element Res. 60 (1997) 153. [6] C.C. Huang, H.T. Chan, Anal. Chem. 78 (2006) 8332. [7] A. Giacomino, O. Abollino, M. Malandrino, E. Mentasti, Talanta 75 (2008) 266. [8] H. Ju, D.J. Leech, Electroanal. Chem. 484 (2000) 150. [9] S. Berchmans, S. Arivukkodi, V. Yegnaraman, Electrochem. Commun. 2 (2000) 226. [10] P. Ugo, S. Zampieri, L.M. Moretto, D. Paolucci, Anal. Chim. Acta 434 (2001) 291. [11] R. Agraz, M.T. Sevilla, L. Hernandez, J. Electroanal. Chem. 390 (1995) 47. [12] J. Wang, M. Bonakdar, Talanta 35 (1988) 277. [13] C. Faller, N.Y. Stojko, G. Henze, K.Z. Brainina, Anal. Chim. Acta 396 (1999) 195. [14] J. Labuda, V. Plaskon, Anal. Chim. Acta 228 (1990) 259. [15] P. Ugo, L.M. Moretto, G.A. Mazzocchin, Anal. Chim. Acta 305 (1995) 74. [16] J. Lu, X. He, X. Zeng, Q. Wan, Z. Zhang, Talanta 59 (2003) 553. [17] Y. Guo, A.R. Guadalupe, J. Pharmaceut. Biomed. 19 (1999) 175. [18] G. Cabello-Carramolino, M.D. Petit-Dominguez, Anal. Chim. Acta 614 (2008) 103. [19] L. Yu, M. Sathe, X. Zeng, J. Electrochem. Soc. 152 (2005) E357. [20] C. Aso, T. Kunitake, T. Nakashima, Macromol. Chem. 124 (1969) 232. [21] D.D. Perrin, W.L.F. Armarego, D.R. Perrin, Purification of Laboratory Chemicals, Pergamon Press, 1980. [22] H. Gülce, H. Özyörük, A. Yıldız, Electroanalysis 7 (1993) 178. [23] P.J. Peerce, A.J. Bard, J. Electroanal. Chem. 112 (1980) 97. [24] N. Rajesh, M.S. Hari, Spectrochim. Acta Part A 70 (2008) 1104.