This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright
Transcript
This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies areencouraged to visit:
Glassy carbon electrode modified by conductive polyaniline coating fordetermination of trace lead and cadmium ions in acetate buffer solution
Zhaomeng Wang, Erjia Liu ⁎, Xing ZhaoSchool of Mechanical and Aerospace Engineering, Nanyang Technological University, 50 Nanyang Avenue, Singapore 639798, Singapore
a b s t r a c ta r t i c l e i n f o
Available online 20 January 2011
Keywords:Polyaniline modified glassy carbon electrodeSurface morphologySquare wave voltammetryTrace lead and cadmium
Polyaniline (PANI) coatings were electrodeposited on the surfaces of glassy carbon electrodes (GCEs) to formnew electrodes, i.e. PANI/GCEs. It was found that with increased deposition time, the PANI coatings becamemore compact while the charge transfer resistance of the coatings became higher. The PANI/GCEs were usedto detect Cd2+ and Pb2+ ions contained in 0.1 M acetate buffer solutions using square wave anodic strippingvoltammetry (SWASV). It was found that the PANI/GCE had a highest anodic stripping peak current in asolution of pH 5.3. The study of the cleaning performance of the PANI/GCEs indicated that there were lessremaining metals on the surfaces of the PANI/GCEs compared to the bare GCEs after cleaning at a potential of0.4 V, which was probably due to that the PANI coatings could effectively prevent the deposition of the metalsinto the surface defects of the GCEs. The PANI coatings could also reduce the passivation effect of the GCEs,thus improving the repeatability of the electrodes.
The presence of trace heavy metals such as mercury (Hg), lead(Pb), cadmium (Cd) and copper (Cu), etc. in an aquatic ecosystemdirectly or indirectly impacts biota and human being, which hasresulted in an ever-increasing demand for the detection of heavymetal contaminants [1,2]. Square wave anodic stripping voltammetry(SWASV) is a widely used analytical technique for detection anddetermination of trace heavy metals at low cost [3,4].
Recently, modifications of electrodes for detection of trace heavymetals by means of conductive polymers have received considerableattention due to their superior electrical conductivities, good adhesionproperties and suitable structural characteristics [5,6]. Due to its facilepreparation, high conductivity and good environmental stability [7],conductive polyaniline (PANI) can be electrochemically coated on thesurfaces of glassy carbon electrodes (GCEs) and forms a porouscoating [8,9]. PANI coatings are stable and can remain intact for a longtime as long as they are not mechanically damaged [10].
Themicrostructure of PANI coatings can be controlled by fabricationmethods and conditions such as temperature, monomer concentration,deposition potential and time, which then greatly influences theirelectrical conductivities [11]. For example, PANI coatings prepared byconventional potentiodynamic and potentiostatic methods wereusually thick and compact [12]. A compact coating has a relativelysmall specific surface area and a poor electrical conductivity, which are
unfavorable for the construction of an electrochemical sensor. Thus,PANI coatings formed by electrodeposition on GCEs have usually beenvia a method of cyclic voltammetry (CV).
PANI coatings can be used as substrates of Bi, Hg, Au or carbonnanotube (CNT) for detection of trace heavy metals by anodicstripping voltammetry [10,13–17]. PANI can also be mixed withcarbon such as template carbon or CNT [12,17]. When a PANI coatingmodified glassy carbon electrode is immersed in a solution containingtrace heavy metal ions (e.g. Hg2+), interactions occur betweenelectron-rich sites (functional groups) and positively charged metalspecies, and thus the metal ions are adsorbed onto the electrodesurface and further separated from the bulk solution [18–20]. It wasreported that metal ions could be accumulated (during pre-concen-tration process) via a two-step process with PANI coated electrodes,i.e., selective pre-concentration of metal ions (with the electrodesimmersed in an original solution) followed by first cleaning and thenelectrodeposition (in a fresh solution), instead of direct electrodepo-sition in the original solution [10,20].
Conductive polymers deposited on electrode surfaces couldenhance the stripping responses in the detection of metal ions, buttheir superiority was reported to be not obvious [15,16]. It wasreported that when PANI coatings were used for anodic strippingvoltammetric determination of trace heavy metals, low pH valueswere not preferred because high PANI oxidation peaks induced at lowpH values could overlap with the stripping peaks of target metals [6].In a report, conductive PANI was deposited on a GCE surface tosupport bismuth nanoparticles to form a composite electrode (Bi/PANI/GCE) for heavy metal stripping analysis [6,16]. However, therehave been no detailed reports on heavy metal stripping using only
j ourna l homepage: www.e lsev ie r.com/ locate / ts f
Author's personal copy
PANI layer modified GCEs (PANI/GCEs). In this paper, PANI/GCEs wereused for the anodic stripping determination of Pb2+ and Cd2+ ions,and their stripping performances were studied with respect to thoseof bare GCEs. The cleaning performance of the PANI/GCE surfaces aftereach stripping test cycle was studied for the first time. A CV methodwas used to polymerize aniline monomers in a solution to form PANIcoatings, and the metals were deposited on the electrode surfaces bymeans of direct electrodeposition.
2. Experimental details
Aniline monomers (Fluka) were freshly distilled under a reducedpressure and stored at a low temperature (−5 °C) in a nitrogenatmosphere. CH3COOH (Fluka) and CH3COONa (Fluka) were used forthe preparation of a 0.1 M acetate buffer solution. Stock solutions of Pb(NO3)2 and Cd(NO3)2 (Sigma-Aldrich) were diluted using doublydistilled water to a concentration of 1 mM each and stored at roomtemperature. All the chemicals used were of analytical reagent grade.
All electrochemical experiments such as polymerization of PANI andvoltammetric measurements were performed using an electrochemicalworkstation (CHI 660C), having a conventional three-electrode cellconfiguration with a GCE or PANI/GCE of a diameter of 3 mm as theworking electrode, a platinum mesh as the counter electrode and anAg/AgCl (saturated KCl) reference electrode. A magnetic stirrer(Heidolph MR3001K) was used to stir the testing solutions duringcleaning and preconcentration.
Thoroughly polished GCE surfaces using a slurry containing 0.3 μmα-Al2O3 powders on a soft cloth were sonicated in first ethanol andthen doubly distilled water for 3 min each to remove possiblecontaminants. The PANI coatings were formed on the GCE surfacesby dipping the polished GCEs in a 0.25 M H2SO4 electrolyte containing7.3 mM aniline monomers via a CV process from −0.2 to 0.9 V at ascan rate of 50 mV/s for 30 cycles under a nitrogen environment. Afterthe polymerization of PANI, the fabricated PANI/GCEs were dippedinto doubly distilled water for 3 min to remove unpolymerized anilinemonomers remaining in the PANI coatings if any.
For the study of sensing trace heavy metals in aqueous solutionsusing the electrodes developed in this study, the electrodes weredipped in a 0.1 M acetate buffer solution (pH 5.3) containingpredetermined concentrations of Pb2+ and Cd2+ ions that were thetarget metals to be investigated. Then square wave anodic strippingvoltammetric (SWASV) measurements were performed. Firstly, apreconcentration potential of −1 V was applied to the workingelectrode for 120 s with continuous stirring. Next, a quiet time of 30 swas taken to stabilize the solution. Finally, the anodic stripping wasscanned from −1 to −0.2 V with a frequency of 50 Hz, increment of5 mV/s and amplitude of 50 mV, with the voltammograms recordedfor analysis. For repetitive measurements, the electrode surfaces werecleaned at 0.3 V for 120 s with continuousmagnetic stirring to removethe residual metals after each experiment. All the experiments werecarried out at room temperature under a nitrogen environment.
3. Results and discussion
Usually there are two ways to increase the amount of PANI coatingduring polymerization: a higher aniline monomer concentration in asolution and a longer polymerization time. In this paper, polymerizationtimewas increased by increasing the number of scan cycles during the CVdeposition. The voltammograms measured during the synthesis of thePANI coatings on the GCE surfaces in the electrolyte of 0.25 M H2SO4
containing 7.3 mM aniline monomers using CV method are shown inFig. 1. The oxidation peaks at about 0.235 V are related to thetransformation of the deposited PANI coatings from leucoemeraldineform (fully reduced state) to emeraldine salt (neutral state). The smalloxidation peaks at about 0.4 V, which are not obvious but can beidentifiedwith reference to the corresponding reduction peaks at around
0.37 V, are due to the branched structure of the PANI layers. The oxidationpeaks at about 0.513 V refer to the state transformation from emeraldineto pernigraniline (fully oxidized state). The oxidation peaks at about0.706 V are related to the polymerization reactions of aniline. The peakcurrents of the two main oxidation peaks at about 0.235 and 0.513 V arerelated to the amounts of PANI deposited on the GCE surfaces. As thenumber of scan cycles increases, the two main peaks increase, whichindicates that thicker PANI coatings have been formed. Because the PANIoxidation peaks are higher than 0.1 V as shown in Fig. 1, the PANIcoatings are ‘electro-inactive’ within the potential range from −1.4 to0.1 V (vs. Ag/AgCl), and thus they are neither oxidizable nor reducibleand hence have no interferences with the redox reactions of the metalions in the solution. Thus, the PANI coatings can be used to modify theGCEs for the application of anodic stripping voltammetric determinationof trace heavy metals.
The PANI coatings have a porous and branched structure that canincrease the specific surface area. From the SEM micrographs shownin Fig. 2, the PANI coatings appear to be denser (reduced specificsurface area) with the increased number of deposition cycles, andmaylead to a larger film charge transfer resistance that is not preferred.The insets of Fig. 2 show the photos of the PANI coated electrodesurfaces with different CV deposition cycle numbers, where thecoating deposited with 25 cycles is thin and nonuniform due torelatively short deposition time. As the number of scan cyclesincreases up to 35, the coatings get thicker and more uniform, andat the same time, the colour of the coatings darkens from light greento dark jade green that remains unchanged even after a few days ofexposure to air.
At the film-solution interfaces the charge transfer resistance of thePANI coatings is affected by the thickness, specific surface area,conductivity, doping level, and oxidation state of the coatings. Asdiscussed, a longer deposition time can lead to a thicker PANI coatingwith a reduced density of porosities, which results in a higher chargetransfer resistance as confirmed by the Nyquist plots shown in Fig. 3.The Randles circuit as shown in the lower inset of Fig. 3 indicatesmixed kinetic and charge transfer control, where Rs refers to theelectrolyte resistance, Cdl is the double layer capacitance, Rct is thecharge transfer resistance that equals the diameter of the semicircle ina Nyquist plot, and W is the Warburg element referring to anelectrochemical diffusion.
As shown in the upper inset of Fig. 3, the charge transfer resistanceof the PANI coating increases with the increase of cycles, which is inagreement with that previously discussed. Thus, the optimumnumber of CV deposition cycles to achieve both uniform and highlyconductive PANI coatings is about 30.
Fig. 1. In-situ voltammograms of a PANI coating measured during its deposition up to40 cycles with a scan rate of 50 mV/s from −0.2 to 0.9 V.
5286 Z. Wang et al. / Thin Solid Films 519 (2011) 5285–5289
Author's personal copy
The simultaneous anodic stripping voltammetric determination ofPb2+ and Cd2+ ions using the PANI/GCEs in the 0.1 M acetate buffersolutions containing Pb2+ and Cd2+ of 3 μMeachwith respect to the pHvalues of the solutions is illustrated in Fig. 4. The stripping peak currentsof the bothmetals increase as the pH increases from3.5 to 5.3. However,when the pH is higher than 5.3, the both peak currents drop. When thepH reduces from 5.3 to 3.5, the PANI coatings have increasedconductivities due to higher concentrated H+ doping at lower pHvalues, which may tend to increase the peak currents. However, thePANI coatings more positively charged can repulse the metal ions thatare also positively charged due to higher concentrated H+ doping atlower pH values. In addition, the hydrogen evolution at a lower pH canbe much easier, which can reduce the electrode surface activities.Because of these two effects, the peak currents of the PANI/GCEs dropwith the decrease of pH from 5.3 to 3.5. The following discussionwill bebased on the results measured from the solutions of pH 5.3 that favorshigher stripping responses.
Usually, it is preferred that one electrode can be used for many tests,which requires good stability and repeatability of the electrode. However,it is observed that the previously deposited metals on the electrodesurface cannot be completely stripped off the electrode surface after eachmeasurement, and the residual metals can influence the results from thesubsequent measurements. Therefore, the cleaning performance of theelectrodes, which is defined as the fraction of remaining metals on anelectrode surface after cleaning at a certainpotential, is studied for thefirsttime with the procedures as demonstrated below. A freshly polished orPANI coated electrode is scanned within the potential range for strippingof Pb2+ ions in an acetate buffer solution (pH 5.3) containing 3 μM Pb2+
ions toget a stripping current Is. After stripping theelectrode isdipped intodoubly distilledwater for 3 min to remove the adsorbedmetal ions. Then,a cleanliness test is performedwith the electrode being scanned in a freshacetate buffer solution (pH 5.3)without Pb2+ ions to get a peak current Ir.Finally, the fraction of the remaining Pb on the electrode tested iscalculated using the formula:
Iremaining = Istripping + Iremaining
� �= Ir = Is + Irð Þ:
Fig. 2. SEM micrographs of PANI coatings deposited from a 0.25 M H2SO4 solutioncontaining 7.3 mM aniline using CV method for (a) 25, (b) 30 and (c) 35 cycles. Theinsets are the photos of the respective PANI coatings.
Fig. 3. Nyquist plots of PANI coatings deposited for 25, 30 and 35 cycles. Lower insetshows the Randles circuit model for EIS measurements, while upper inset shows thecharge transfer resistances.
Fig. 4. Voltammograms measured with PANI/GCE in solutions containing 3 μM Pb2+
and 3 μM Cd2+ at different pH values. The inset shows the effect of pH value onstripping peak current.
5287Z. Wang et al. / Thin Solid Films 519 (2011) 5285–5289
Author's personal copy
The average percentage of the remaining lead on the PANI/GCEs isabout 7.85%, that is smaller than that of the bare GCEs (about 10.91%),indicating that the PANI layers can effectively reduce the amount ofremaining Pb on the PANI/GCEs. A possible explanation is schematicallyillustrated in Fig. 5 in which the reduced heavymetals can be depositedon the GCE surface and trapped at the internal defects near the surfaceduring the preconcentration. During the stripping, the metal particlestrapped at the defects can also be oxidized and thus contribute to thestripping currents alongwith themetal ions stripped from the electrodesurface. However, the oxidized ions from the trapped metal deposits atthe defects cannot easily or completely diffuse out of the electrodesurface. The conductive PANI layers coated on the electrodes can shieldthe passages to the internal defects of the GCE surfaces, which haveeffectively prevented the targetmetals frombeing trappedat thedefectsduring preconcentration. Another possible explanation is that with alayer of PANI coating, most of the metal particles are deposited on theconductive polymer surface rather than on the bare GCE surfaceexposed through the porosities in the PANI coating, which has beenconfirmed with SEM (micrographs not shown). Since the cleaningperformance is one of the key factors affecting the repeatability of theelectrodes, a lower amount of remaining Pb on the PANI/GCEs can serveto improve the repeatability of the electrodes.
Natural environmental samples in which trace heavy metals need tobe analyzed usually contain some kinds of surface-active substances(surfactants), due to the adsorption from the surrounding environment.The attachments of these surfactants onto electrode surfaces may lead toweaker or broader stripping current peaks as well as peak shifts inpotentials. The passivation of the electrodes can be limited by theelectrode modification using a thin layer of PANI. A series of tests isperformed using the both GCE and PANI/GCE in the solutions containingvarying Pb2+ concentrations from 0 to 3 μM as shown in Fig. 6. FromFig. 6 it can be seen that the GCE shows a nearly second order curve withreducing slopes due to the passivation of the electrode, while the PANI/GCE depicts a near linear relationship with respect to Pb2+ concentrationpossibly due to the branch structured PANI layer that can block thesurface-active substances from reaching the GCE surface. Since the samePANI/GCE is used for all the measurements shown in Fig. 6, sometimesthe PANI layer can be partially damaged during themeasurements, whichleads to a step reduction of the peak current and thus slightly affects thestability of the electrode.
From the calibration curves (Fig. 7) of the PANI/GCE electrodes withrespect to the concentrations of Pb2+andCd2+ ions (both in the rangeof
Fig. 5. Schematic diagrams showing the effect of PANI coating on cleaning performance of electrode surface.
Fig. 6. Stripping peak currents measuredwith (a) GCE and (b) PANI/GCEwith respect toPb2+ concentrations in solutions.
5288 Z. Wang et al. / Thin Solid Films 519 (2011) 5285–5289
Author's personal copy
0 to 2 μM), the relationships between stripping current and Pb2+ andCd2+ concentrations are:
I = −3:616 + 12:115 C½ �; with R = 0:989 and DL = 0:1 μM for Pb2+
and
I = −0:688 + 2:368 C½ �; with R = 0:974 and DL = 0:13 μM for Cd2+
where I is the stripping current in μA, [C] is the concentration of heavymetal ions in μM, R is the correlation coefficient, and DL is thedetection limit.
The high R values indicate the excellent performance of the PANI/GCEs in the detection of trace lead and cadmium.
4. Conclusions
PANI coatings were electrodeposited on glassy carbon electrodesto form PANI/GCE electrodes. With increased deposition time the
PANI coatings became thicker and denser, leading to higher chargetransfer resistances. The amount of remaining Pb on the electrodeswith or without PANI coatings after stripping was examined, whichshowed that the electrodes coated with PANI had a better cleaningperformance. It was found that the PANI coatings reduced thepassivation of the electrodes, which was attributed to their branchedstructures that could block the surface-active molecules from reach-ing the electrode surfaces.
Acknowledgements
Z.M.Wang and X. Zhaowere grateful for the PhD scholarships fromthe Nanyang Technological University (NTU), Singapore.
References
[1] T.U. Aualiitia, W.F. Pickering, Water Res. 20 (11) (1986) 1397.[2] M. Heitzmann, L. Basaez, F. Brovelli, C. Bucher, D. Limosin, E. Pereira, B.L. Rivas, G.
Royal, E. Saint-Aman, J.C. Moutet, Electroanalysis 17 (21) (2005) 1970.[3] E.P. Achterberg, C. Braungardt, Anal. Chim. Acta 400 (1999) 381.[4] J.B. Jia, L.Y. Cao, Z.H. Wang, T.X. Wang, Electroanalysis 20 (5) (2008) 542.[5] M.D. Imisides, R. John, P.J. Riley, G.G. Wallace, Electroanalysis 3 (9) (1991) 879.[6] W.W. Zhu, N.B. Li, H.Q. Luo, Anal. Lett. 39 (11) (2006) 2273.[7] Z.X. Wei, M.X. Wan, T. Lin, L.M. Dai, Adv. Mater. 15 (2) (2003) 136.[8] S. das Neves, M.A. De Paoli, Synth. Met. 96/1 (1998) 49.[9] H.N. Dinh, J.F. Ding, S.J. Xia, V.I. Birss, J. Electroanal. Chem. 459 (1) (1998) 45.
[10] J.H. Santos, M.R. Smyth, R. Blanc, Anal. Commun. 35 (10) (1998) 345.[11] J.M. Ribo, M.C. Anglada, J.M. Hernandez, X.D. Zhang, N. Ferrer-Anglada, A. Chaibi,
B. Movaghar, Synth. Met. 97 (3) (1998) 229.[12] X.H. Gao, W.Z. Wei, L. Yang, M.L. Guo, Electroanalysis 18 (5) (2006) 485.[13] A.N. Chowdhury, S. Ferdousi, M.M. Islam, T. Okajima, T. Ohsaka, J. Appl. Polym. Sci.
