Surface properties of self-assembled monolayer films of tetra-substituted cobalt, iron and manganese alkylthio phthalocyanine complexes

Electrochimica Acta 55 (2010) 7085–7093

Contents lists available at ScienceDirect

Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta

Surface properties of self-assembled monolayer films of tetra-substituted cobalt, iron and manganese alkylthio phthalocyanine complexes Isaac Adebayo Akinbulu, Samson Khene, Tebello Nyokong ∗ Department of Chemistry, Rhodes University, Grahamstown 6140, South Africa

a r t i c l e

i n f o

Article history: Received 18 February 2010 Received in revised form 21 June 2010 Accepted 24 June 2010 Available online 3 July 2010 Keywords: Phthalocyanine Self-assembled monolayer Electrochemical impedance spectroscopy Atomic force microscopy Carbofuran

a b s t r a c t Self-assembled monolayer (SAM) films of iron (SAM-1), cobalt (SAM-2) and manganese (SAM-3) phthalocyanine complexes, tetra-substituted with diethylaminoethanethio at the non-peripheral positions, were formed on gold electrode in dimethylformamide (DMF). Electrochemical, impedimentary and surface properties of the SAM films were investigated. Cyclic voltammetry was used to investigate the electrochemical properties of the films. Ability of the films to inhibit common faradaic processes on bare gold surface (gold oxidation, solution redox chemistry of [Fe(H2 O)6 ]3+ /[Fe(H2 O)6 ]2+ and underpotential deposition (UDP) of copper) was investigated. Electrochemical impedance spectroscopy (EIS), using [Fe(CN)6 ]3−/4− redox process as a probe, offered insights into the electrical properties of the films/electrode interfaces. Surface properties of the films were probed using atomic force microscopy (AFM) and scanning electron microscopy (SEM). The films were employed for the electrocatalytic oxidation of the pesticide, carbofuran. Electrocatalysis was evidenced from enhanced current signal and less positive oxidation potential of the pesticide on each film, relative to that observed on the bare gold electrode. Mechanism of electrocatalytic oxidation of the pesticide was studied using rotating disc electrode voltammetry. © 2010 Elsevier Ltd. All rights reserved.

1. Introduction Thin films of metallophthalocyanine (MPc) complexes are frequently employed as electrocatalysts and in the design of electrochemical sensors. Electrocatalytic activities of MPc complexes towards many analytes have been extensively reported [1–6]. Formation of thin films of MPc complexes can be achieved by direct deposition of the monomer on a suitable solid substrate by adsorption [7–10], electrochemical polymerization [11–17], electrodeposition [4,18], Langmuir–Blodgett films [19,20] and self-assembled monolayer [21–25], among others. The method of interest may be informed by the molecular identity of the substituent on the phthalocyanine ligand and the nature of the solid substrate. In the present work, we report the formation of thin films of non-peripherally tetra-substituted iron (SAM-1), cobalt (SAM-2) and manganese (SAM-3) diethylaminoethanethio phthalocyanine complexes (Scheme 1) by self-assembly technique. This technique forms molecular thin films by spontaneous chemisorptions of molecules of interest onto a suitable solid substrate, resulting in a self-organized array of molecules and the formation of reproducible stable insoluble films. Sulfur particularly has high

∗ Corresponding author at: Department of Chemistry, Rhodes University, PO Box 94, Grahamstown 6140, South Africa. Tel.: +27 46 6038260; fax: +27 46 6225109. E-mail address: [email protected] (T. Nyokong). 0013-4686/$ – see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.electacta.2010.06.065

affinity for gold and silver, hence various MPc complexes with substituents containing sulfur group, like those used in this work, have been synthesized for the purpose of forming SAM [26–31]. The synthesis and electrochemical properties of non-peripherally tetra-substituted cobalt (diethylaminoethanethio) phthalocyanine (complex 2 for SAM-2) [32] and manganese(III) acetate (diethylaminoethanethio) phthalocyanine (complex 3 for SAM-3) [33] have been reported elsewhere. The synthesis of the iron analogue (for SAM-1) is reported in this work. Electrochemical properties of the films were determined using cyclic voltammetry, while their impedimentary properties were investigated by electrochemical impedance spectroscopy (EIS). EIS is a powerful technique of characterizing the electrical properties of materials and their interfaces with electronically conducting electrodes [34]. Scanning electron microscopy (SEM) and atomic force microscopy (AFM) were also employed to study the SAM films. Electrocatalytic application of the SAM films was investigated in the electro-oxidation of the pesticide, carbofuran. Carbofuran is a systemic insecticide used to control insects in a wide variety of field crops such as potatoes, corn and soybeans. It is regarded as one of the most toxic carbamate pesticides, with toxicity second to that of parathion and aldicarb, hence, electrocatalytic oxidation of this pesticide, to facilitate its detection in environmental samples, is important. Carbofuran is electrochemically inactive but was converted to the electroactive phenolic derivative by alkaline hydrolysis. Mechanism of electrocatalytic oxidation

7086

I.A. Akinbulu et al. / Electrochimica Acta 55 (2010) 7085–7093

Scheme 1. Synthetic pathway for complex 1 (for SAM-1).

of this pesticide was investigated using rotating disc electrode voltammetry. 2. Experimental 2.1. Materials Potassium carbonate, iron(II) chloride tetrahydrate, 2(diethylaminoethanethiol) hydrochloride, 2-diethylaminoethanol and carbofuran were obtained from Sigma–Aldrich. Anhydrous form of iron(II) chloride was obtained by heating the hydrated form in an oven. Tetrabutylammonium tetrafluoroborate (TBABF4 ) (Aldrich) was used as the electrolyte for electrochemical characterization of complex 1. Aluminum oxide, WN-3: neutral for column chromatography, was purchased from Sigma–Aldrich. Dimethylformamide (DMF) and dichloromethane (DCM) were obtained from Merck. DMF, DCM, methanol and ethanol were distilled before use. Stock solution of carbofuran (2.03 × 10−3 M) was prepared in freshly distilled methanol because of its limited solubility in water. All solutions were prepared with ultrapure water of resistivity 18.2 M cm obtained from a Milli-Q Water system. Electrochemical experiments were carried out in argon-saturated aqueous solutions containing small amounts of methanol from the stock solution of carbofuran. Prior to electrochemical analysis involving carbofuran, a desired volume of the stock solution (1.97 ml used to prepare 10 ml of 200 ␮M solution of carbofuran) was hydrolyzed in 0.5 M solution of NaOH (0.8 ml). Acetic acid (20 ␮L) was added to adjust the pH to 4, since the best voltammetry response of carbofuran was obtained at this pH. The SAM films were also very stable at this pH. 2.2. Electrochemical studies All electrochemical experiments were performed using Autolab potentiostat PGSTAT 302 (Eco Chemie, Utrecht, The Netherlands) driven by the general purpose Electrochemical System data processing software (GPES, software version 4.9) or BioAnalytical Systems (BAS) model 100B/W Electrochemical Workstation [for rotating disc electrode voltammetry (RDE)]. RDE studies were done in the potential range of 0.1–0.45 V at 1600 rpm. A conventional three-electrode system was used. The working electrode was a bare glassy carbon electrode (GCE) (for electrochemical characterization of complex 1) or gold electrode (for SAM formation), Ag|AgCl wire and platinum wire were used as the pseudo-reference and auxil-

iary electrodes, respectively. Ag|AgCl (3 M KCl) reference electrode was used for electrochemical experiment involving carbofuran. The potential response of the Ag|AgCl pseudo-reference electrode was less than the Ag|AgCl (3 M KCl) by 0.015 ± 0.003 V. Prior to use, the glassy carbon electrode surface was polished with alumina on a Buehler felt pad and rinsed with excess millipore water. The gold electrode surface was also polished in aqueous slurry of alumina on sic-emery paper and subjected to ultrasonic vibration in absolute ethanol to remove residual alumina. The electrode was then etched in hot ‘piranha’ solution (1:3 (v/v) 30% H2 O2 and concentrated H2 SO4 ) for 2 min and rinsed with excess millipore water. It was then scanned in 0.5 M H2 SO4 between −0.5 and 1.5 V vs. Ag|AgCl to obtain a reproducible scan. The electrode was rinsed with freshly distilled DMF and placed in DMF containing the desired complex. Gold-coated glass was used for surface characterization of the SAM films using AFM and SEM. The glass was immersed in solution of the desired complex in DMF to form SAM. Electrochemical impedance spectroscopy measurements were performed with Autolab FRA software between 1.0 mHz and 10 KHz using a 5 mV rms sinusoidal modulation, in 1 mM solution of [Fe(CN)6 ]3− containing 0.1 M KCl as supporting electrolyte, at halfwave potential of [Fe(CN)6 ]3− /[Fe(CN)6 ]4− (0.10 V vs. Ag|AgCl). A non-linear least squares (NNLS) method based on the EQUIVCRT programme [35] was used for automatic fitting of the obtained EIS data. 2.3. Equipment UV/Vis spectra were recorded on Cary 50 UV/Vis/NIR spectrophotometer. IR (KBr discs) was recorded on Bruker Vertex 70-Ram II spectrophotometer. Elemental analysis was performed using Vario Elementar Microcube EL111. 1 H nuclear magnetic resonance (1 H NMR, 400 MHz) spectra were obtained in CDCl3 using Bruker EMX 400 NMR spectrometer. AFM images were obtained in the non-contact mode in air with a CP-11 Scanning Probe Microscope from Veeco Instruments (Carl Zeiss, South Africa) at a scan rate of 1 Hz. SEM images were recorded using scanning electron microscope of Tescan Digital Microscope model. 2.4. Synthesis 2.4.1. Iron tetrakis-(2-diethylaminoethanethio) phthalocyanine (non-peripheral) (complex 1) Complex 1 was synthesized following the procedure reported for the cobalt analogue [32]. A mixture of 3-

I.A. Akinbulu et al. / Electrochimica Acta 55 (2010) 7085–7093

7087

diethylaminoethanethio phthalonitrile (0.40 g, 1.54 mmol) and anhydrous iron(II) chloride (0.048 g, 0.38 mmol) was refluxed in 2-(diethylaminoethanol) (1.2 ml) for 12 h under nitrogen. Thereafter, the mixture was cooled to room temperature and treated with excess MeOH:H2 O (1:1) to precipitate the crude deep green product. The product was filtered and dried in air. Purification was achieved using column chromatography with neutral alumina as column material and DCM/MeOH (10:1) as eluent. Yield: 1.24 g (74%). Calc. for C56 H68 N12 S4 Fe·CH2 Cl2 : C, 57.09; H, 5.78; N, 14.27; S, 10.88%; Found: C, 56.62; H, 5.61; N, 14.12; S, 11.30%. UV–Vis (CH3 OH): max (nm) (log ε): 689 (5.1), 626 (4.6), 334 (5.2); IR (KBr) vmax /cm−1 ; 2966–2803 (Aliph-CH2 ), 1600, 1496, 1457, 1384, 1324, 1196, 1140, 1071, 914, 823, 745.

3. Results and discussion 3.1. Synthesis and characterization of complex 1 Cyclotetramerization of 3-diethylaminoethanethiol phthalonitrile occurred in the presence of FeCl2 to form complex 1. The complex was purified using column chromatography on alumina. Complex 1 was soluble in solvents such as DMF, DCM, CH3 OH and DMSO. It was characterized using IR and UV–Vis spectroscopies as well as elemental analysis. The results obtained were consistent with the predicted structure of the complex. Formation of the complex was confirmed by the disappearance of the sharp C N vibration at 2229 cm−1 of 3-diethylaminoethanethio phthalonitrile. Complex 1 showed Q and B bands at 689 and 334 nm, respectively, in CH3 OH (see Supporting information files, Fig. 1.) Fig. 1A and B shows the square wave and cyclic voltammetry profiles of 1 × 10−3 M of complex 1 in freshly distilled dry DMF containing 0.1 M TBABF4 as supporting electrolyte. Four redox processes were identified and assigned, in comparison with literature for FePc complexes, Table 1. They were also confirmed using spectroelectrochemistry (data not shown). Process I is a quasi-reversible (E = 171 mV vs. Ag|AgCl) ring-based reduction, assigned to the formation of FeI Pc−2 /FeI Pc−3 species (E1/2 = −1.14 V vs. Ag|AgCl) [36]. Process II is also a quasi-reversible (E = 121 mV vs. Ag|AgCl) redox process associated with metal-based reduction and the formation of FeII Pc−2 /FeI Pc−2 species (E1/2 = −0.49 V vs. Ag|AgCl) [37]. Process III is a metal-based oxidation associated with FeIII Pc−2 /FeII Pc−2 species (E1/2 = +0.29 V vs. Ag|AgCl) [36,38]. This process is also quasi-reversible with anodic current more intense than cathodic one. Process IV is an irreversible ring-based oxidation assigned to FeIII Pc−1 /FeIII Pc−2 species (Ep = +0.93 V vs. Ag|AgCl) [38], irreversible oxidation is typical of alkylthio substituted MPc complexes [39]. The first oxidation in FePc complexes is known to occur at the central metal [40]. Spectroelectrochemistry (Fig. 2 of Supporting information files) was used to confirm the nature of the first

Fig. 1. (A) Square wave and (B) cyclic voltammetry profiles of 1 × 10−3 M of complex 1 in freshly distilled dry DMF containing 0.1 M TBABF4 as supporting electrolyte. Step potential: 5 mV, amplitude: 50 mV vs. Ag|AgCl, frequency: 10 Hz. Scan rate: 100 mV s−1 .

reduction couple II, which could be due to either FeI Pc−2 or FeII Pc−3 . The reduction of FeII Pc to FeI Pc is known to show no (or low intensity) Q band and to give a pink solution due to absorption in the 500 nm region [40]. The formation of a relatively intense peak near 500 nm confirms that couple II is due to reduction of FeII Pc to FeI Pc. The pink colour of the solution has been attributed to disturbance of the  − * spectrum of the phthalocyanine by FeI [40]. 3.2. Formation of SAM SAM films were formed by immersing the bare gold electrode in the desired complex, for 24 h in each case. Formation of SAM films may have been facilitated by coordination of the sulfur group (using its lone pairs of electrons) to gold, with the C–S bond intact on formation of SAMs, as we reported previously [41]. This was proved using electrochemical methods, where MPcs with different substituents, but same central metal, gave different SAM behaviors [41]. Also the existence of C–S bond after SAM formation has been confirmed, using in situ surface enhanced Raman spectroscopy (SERS), for SAMs of anthraquinone derivatives, formed on

Table 1 Summary of peak and half-wave potentials, in E1/2 vs. (Ag|AgCl)/V, of the complex 1 in comparison with that of previously reported thio derivatised FePc complexes. Values were recorded in DMF containing TBABF4 unless otherwise stated. Complex Complex 1 FeODEAETPc(␤) FeTBMPc(␤)a FeTDMPc(␤)a,b FeOBMPc(␤)a,c a

FeI Pc−3 /FeI Pc−4

FeI Pc−2 /FeI Pc−3

FeII Pc−2 /FeI Pc−2

FeIII Pc−2 /FeII Pc−2

FeIII Pc−1 /FeIII Pc−2

References

−1.18

−1.14 ± 0.01 −0.9 −0.78 −0.84 −0.70

−0.49 ± 0.01 −0.35 −0.37 −0.53 −0.26

+0.29 ± 0.01 +0.26 +0.36 +0.62 +0.25

+0.93 ± 0.01 +0.87 +0.70 +1.01 +0.60

TWd [37] [38] [38] [39]

−1.30

TDMPc = tetra dodecylmercapto phthalocyanine, TBMPc = tetra benzylmercapto phthalocyanine, OBMPc = octa benzylmercapto phthalocyanine, ODEAETPc = octa diethylaminoethanethio phthalocyanine. b Values recorded in DCM, using TBABF4 . c Electrolyte = tetrabutylammonium perchlorate. d TW = this work.

7088

I.A. Akinbulu et al. / Electrochimica Acta 55 (2010) 7085–7093

Fig. 2. Cyclic voltammetry profiles of bare (dashed line) and MPc-SAM modified gold electrodes in (A) 0.1 M KOH solution, (B) 1 mM Fe(NH4 )(SO4)2 solution containing 1 mM HClO4 , (C) 1 mM solution of [Fe(CN)6 ]3− containing 0.1 M KCl supporting electrolyte and (D) 1 mM CuSO4 solution containing 0.5 M H2 SO4 . Scan rate: 25 mV s−1 .

repetitive scanning, suggesting the unstable nature of the SAMs at alkaline pH. Fig. 2A was also used to investigate the extent to which the bare gold surface has been isolated from the electrolyte solution (0.1 M KOH) as a result of the presence of the SAMs. This was done by estimating the ion barrier factor ( ibf ) (Eq. (1)) for each SAM film.

gold electrode [42]. Molecular configuration of the anthraquinone derivatives is similar to that of the MPc complexes studied in this work. 3.2.1. Inhibition of faradaic processes Electrochemical properties of the films are closely related to the extent to which they inhibit common faradaic processes associated with bare gold surface. Such processes include gold surface oxidation, solution redox chemistry of [Fe(H2 O)6 ]3+ /[Fe(H2 O)6 ]2+ and underpotential deposition (UDP) of copper. Considerable passivation of these processes was observed after SAM formation. More insights into the electrochemical nature of the films were provided using the fast electron transfer [Fe(CN)6 ]3− /[Fe(CN)6 ]4− redox process, and investigation of the oxidation redox couples of the MPc-SAM modified Au electrodes in pH 4 buffer solution. Fig. 2A shows the oxidation (at +0.37 V vs. Ag|AgCl) and reduction (gold oxide stripping peak, at around 0.02 V vs. Ag|AgCl) of bare gold surface in 0.1 M KOH. There was appreciable passivation of this process in the presence of all the MPc-SAMs. Gold oxidation was considerably inhibited, with drastic decrease in intensity of the gold oxide stripping peak, suggesting the electrolyte was no longer accessible to the bare gold surface, confirming SAM formation. However, there was gradual desorption of the SAMs during

ibf = 1 −

QSAM QBare

(1)

where QSAM and QBare (1.82 × 10−5 C) are the total charges under the reduction peak (gold oxide stripping peak) at the SAM modified and bare gold electrode, respectively (Fig. 2A). Values obtained, with their margin of errors, are indicated in Table 2 (0.87, 0.88 and 0.83 for SAM-1, SAM-2, and SAM-3, respectively). These values are close to unity, indicating a closely packed SAMs, with minimal defects. Fig. 2B shows the cyclic voltammograms obtained for the bare electrode (dashed line) and the SAM modified gold electrodes in 1 mM ferrous ammonium sulfate (Fe(NH4 )(SO4 )2 ) containing 1 mM perchloric acid (HClO4 ). The clearly resolved quasi-reversible redox process ([Fe(H2 O)6 ]3+ /[Fe(H2 O)6 ]2+ ) on the bare electrode (dashed line) was almost completely inhibited on SAM-1 and SAM-2 modified gold electrodes, while the Fe3+ /Fe2+ redox couple shifted to more negative potentials and its intensity decreased

Table 2 Electrochemical data of SAMs and parameters for electrocatalytic oxidation of carbofuran for SAMs of complexes 1–3 in pH 4 buffer. MIII /MII (E1/2 vs. (Ag|AgCl)/V)

 MPc-SAM /mol cm−2

Electrochemical data for SAMs of complexes 1–3 SAM-1 0.87 ± 0.02 SAM-2 0.88 ± 0.02 SAM-3 0.83 ± 0.02

+0.11 ± 0.002 +0.13 ± 0.003 −0.21 ± 0.01

7.60 ± 0.17 × 10−11 5.48 ± 0.14 × 10−11 2.69 ± 0.11 × 10−10

Electrode

IP /␮A

Electrode

 ibf

Ep vs. (Ag|AgCl)/V

Parameters for electrocatalytic oxidation of carbofuran SAM-1 0.40 ± 0.01 SAM-2 0.42 ± 0.01 SAM-3 0.48 ± 0.02 Bare Au 0.67 ± 0.01

1.12 1.62 1.88 0.74

± ± ± ±

0.03 0.03 0.06 0.01

Tafel slope/mV

˛

118 ± 4.1 220 ± 13.2 132 ± 3.3

0.50 ± 0.02 0.27 ± 0.02 0.45 ± 0.01

I.A. Akinbulu et al. / Electrochimica Acta 55 (2010) 7085–7093

on SAM-3 modified gold electrode. Inhibition of this process on the SAM modified gold electrodes suggests isolation of the gold surface from the electrolyte, confirming SAM formation. The four pedant amines of the complexes are tertiary (Scheme 1), hence susceptible to protonation in HClO4 , resulting in appreciably positively charged electrode surface. Thus, inhibition of the ([Fe(H2 O)6 ]3+ /[Fe(H2 O)6 ]2+ ) redox process on the SAM modified Au electrodes may also be explained in terms of electrostatic repulsion of the cationic Fe3+ and Fe2+ ions. Fig. 2C shows the cyclic voltammograms of the bare Au electrode (dashed line) and the SAM modified electrodes in 1 mM solution of [Fe(CN)6 ]3− containing 0.1 M KCl as supporting electrolyte. Unlike the [Fe(H2 O)6 ]3+ /[Fe(H2 O)6 ]2+ redox process in Fig. 2B, the [Fe(CN)6 ]3−/4− redox couple was not inhibited on the SAM modified electrodes because of the fast nature of electron transfer for this process. This observation has been reported previously for adsorbed cobalt tetra-amino phthalocyanine films on vitreous carbon electrode [3] and MnPc on gold electrode [43]. Lack of inhibition of this process may also be interpreted in terms of enhanced surface concentration of [Fe(CN)6 ]3− and [Fe(CN)6 ]4− ions on the electrode, due to electrostatic attraction of these anionic species, assuming the amine groups of the adsorbed complexes are positively charged in the unbuffered KCl electrolyte. Observation of this process on the SAM modified Au electrodes can also be evaluated within the context of the electrocatalytic properties of the SAMs. Oxidation potential of the [Fe(CN)6 ]3−/4− redox process (E1/2 = +0.1 V vs. Ag|AgCl) is within the range of metal-based processes of the adsorbed complexes. However, cathodic to anodic peak difference (E) for this process is larger for SAM-1 (80 mV vs. Ag|AgCl) and SAM-2 (72 mV vs. Ag|AgCl) modified gold electrodes, compared to that observed for the bare electrode (57 mV vs. Ag|AgCl); while that observed for SAM-3 modified gold electrode is same as that for the bare electrode. Interestingly, the anodic and cathodic current intensities of the [Fe(CN)6 ]3−/4− redox couple on bare gold electrode are larger than that observed on all the SAM modified electrodes, confirming the formation of the SAM films. The differences in E for the [Fe(CN)6 ]3−/4− redox couple (on SAM-1 and SAM-2 modified electrodes) and current intensities (on all the SAM modified electrodes), relative to that on bare gold electrode, may be associated with the effect of electron tunneling within the films in the SAM modified electrodes. Fig. 2D shows the cyclic voltammograms of the bare Au electrode and the SAM modified electrodes in 1 mM CuSO4 containing 0.5 M H2 SO4 . The UDP of Cu and stripping of the deposited Cu occurred at −0.18 and −0.36 V (vs. Ag|AgCl), respectively, on the bare Au electrode. These processes were significantly inhibited on the SAM modified electrodes, indicating SAM formation. Redox peaks at +0.11, +0.13 and +0.04 V vs. Ag|AgCl, observed for SAM-1, -2 and -3 modified gold electrodes, respectively, are due to metal-based redox processes of the respective MPc complexes. 3.2.2. Characterization of the metal redox processes for MPc complexes (1–3) Metal-based redox processes of the SAM modified gold electrodes were characterized in pH 4 buffer solution. This pH was used since the best voltammetry response of carbofuran was obtained at this value. Also, SAMs of all the complexes were appreciably stable at this pH but showed weakly resolved reversible processes. Poorly resolved nature of metal-based redox processes has been reported before for SAMs of iron phthalocyanine complex [28] and cobalt porphyrin complexes [44,45], limiting their uses in calculation of surface coverages and prediction of orientations of these complexes on electrodes. Enhanced resolution of redox waves can be achieved by repetitive cycling of the SAM modified electrodes in coordinating solvents (e.g. DMF or pyridine) containing electrolyte such as TBABF4 or tetrabutylammonium percholate (TBAP). This was done

7089

Fig. 3. Cyclic voltammetry profiles of (A) SAM-1, (B) SAM-2 and (C) SAM-3 modified gold electrodes in pH 4 buffer (scan rate: 50 mV s−1 ). Insert in (C)=plot of peak current vs. scan rate.

in this work, Fig. 3. The Fe3+ /Fe2+ redox wave appeared at +0.11 V vs. Ag|AgCl, Co3+ /Co2+ at +0.13 V vs. Ag|AgCl and Mn3+ /Mn2+ at −0.21 V vs. Ag|AgCl, Table 2. Linear dependence (Fig. 3C, inset) of peak current (anodic) on scan rate for SAM-3 (representative of the other SAMs), confirmed surface confinement of the complex on gold electrode. For all the SAM modified gold electrodes, the observed metal redox processes have peak separations (E = 73, 98 and 200 mV vs. Ag|AgCl for SAM1, -2 and -3 modified gold electrodes, respectively) larger than that expected for surface confined species. Such observation has been reported for Co3+ /Co2+ couple (E > 150 mV) in cobalt porphyrins in low pH solutions [46], such as the pH used in this work. It was attributed to the slow kinetics of the couple at low pH. Surface coverage for each SAM was estimated from the relevant redox process in Fig. 3A–C using Eq. (2). MPc-SAM =

Q nFA

(2)

7090

I.A. Akinbulu et al. / Electrochimica Acta 55 (2010) 7085–7093

where Q is the background-corrected charge of the anodic peaks in Fig. 3A–C, A is the real surface area of the gold electrode, n is the number of electron transferred (approximately 1) and F = Faraday’s constant. The real surface area of the electrode was estimated using the well-established method [47] applying Eq. (3) (Randles–Sevcik equation). [Fe(CN)6 ]3− was used as the redox active species because it has a known diffusion coefficient. Ipa = (2.69 × 105 )n3/2 D1/2 v1/2 AC

(3)

where n is the number of electron transferred (n = 1), D is the diffusion coefficient of the redox active species (7.6 × 10−6 cm2 s−1 ) [47], A is the geometric surface area (0.020 cm2 ), v is the scan rate (0.05 V s−1 ) and C is the bulk concentration of [Fe(CN)6 ]3− (0.01 M). Roughness factor of the electrode was calculated to be 1.05 (ratio of Ipa (exptal)/Ipa (theor)), where Ipa (exptal) and Ipa (theor) are the experimental current and current calculated using Eq. (3), respectively. The geometrical area was used in the estimation of Ipa (theor). The product of roughness factor and theoretical surface area gives the real surface area (0.021 cm2 ). The values of surface coverage obtained are 7.60 × 10−11 , 5.48 × 10−11 and 2.69 × 10−10 mol cm2 for SAM-1, SAM-2 and SAM-3, respectively (Table 2). Generally, surface coverage of approximately 1 × 10−10 mol cm2 has been reported for monolayer of metalloporphyrin and metallophthalocyanine complexes lying flat on a substrate [48,49]. Values obtained for SAM-1 and SAM-2 are very close to that reported for cobalt meso-tetrakis (4-pyridyl) porphyrin (7.0 × 10−11 mol cm2 ) [50] lying flat on gold electrode. For SAM-3, the value of surface coverage suggests a different orientation of the SAM on gold, compared to that observed for SAM-1 and SAM-2. It has been reported that a perpendicular orientation is expected for cobalt meso-tetrakis (4-pyridyl) porphyrin adsorbed on gold surface, with a surface coverage of 3.3 × 10−10 mol cm2 [50]. Hence, high value of surface coverage obtained for SAM-3 may indicate perpendicular orientation of the SAM on gold surface. This may have been facilitated by the presence of acetate axial ligand, which distorts flat orientation of the SAM on the substrate. All the SAMs were substantially stable in buffer of pH range 1–7, but desorbed at alkaline pH. Metal-based processes were not observed at alkaline pH, due to desorption of the SAMs at this pH. This is consistent with previous report for octabutylthio substituted FePc complex [28]. Stability of these SAMs may be interpreted in terms of their densely packed crystalline nature, derived from the bulky diethylaminoethanethio substituent. This has been reported for SAMs of n-alkyl thio complexes formed on gold electrode [51]. Density and degree of crystallinity of the SAMs were directly proportional to the length of alkyl chain. 3.3. Impedimentary properties of the films Impedimentary properties of the films were studied using the [Fe(CN)6 ]3−/4− redox process as probe. These properties are closely related to the electrical nature of the films/electrode interfaces. Fig. 4A shows the impedance spectra (Nyquist plots) of (a) bare gold electrode and (b–d) the SAM modified gold electrodes in 1 mM solution of [Fe(CN)6 ]3− containing 0.1 M KCl as supporting electrolyte. The impedance spectrum of the bare gold electrode shows a semi-circle, characteristic of charge transfer-limited impedance, in the high-frequency region, and a straight line, associated with a purely diffusion-limited reaction, in the low-frequency limit. The suitable equivalent circuit representative of this behavior is shown in Fig. 4B(a), where RS is the resistance of the electrolyte solution between the reference and the working electrodes, RCT is the charge transfer resistance, ZW is mass-transfer or Warburg impedance and Cdl is the double-layer capacitance, representative of the capacitance of the electrochemical double layer of the cell. The values of these quantities and their percentage errors are shown in Table 3.

Fig. 4. (A) Nyquist plots obtained for (a) bare, (b) SAM-1, (c) SAM-2 and (d) SAM-3 modified gold electrodes in 1 mM solution of [Fe(CN)6 ]3− containing 0.1 M KCl as supporting electrolyte. Applied potential = 0.10 V vs. Ag|AgCl. (B) Suggested equivalent circuits for the impedance spectral of (a) bare gold electrode, (b) SAM-1 and (c) SAM-2 and SAM-3 modified gold electrodes.

Fig. 4A(b) is the impedance spectrum of SAM-1 modified Au electrode. The spectrum has a feature of charge transfer-limited behavior and negligible diffusion-controlled nature. However, a huge error resulted when the experimental impedance data was fitted using the circuit in Fig. 4A(a). A better fit was obtained when the data was interpreted using the circuit in Fig. 4(b), eliminating contribution from diffusion-limited impedance, where RS , RCT are as defined previously. The third component is the so-called constant phase element (CPE). The impedance spectrum in Fig. 4B(b) is not semicircular; hence inclusion of Cdl as a circuit element resulted in non-linearity in the equivalent circuit proposed. Wu and co-workers [52,53] described this model-depended non-linearity as intrinsic. As observed in this work, they suggested the substitution of Cdl with a distributed circuit element like CPE will give a better fit and reduce errors associated with non-linearity of the proposed circuit. CPE is an empirical representation of deviation of the double-layer capacitance from ideal behavior. It is characteristic of impedance behavior of the double-layer capacitance at solid electrodes [54]. This behavior depends mainly on the state of the electrode’s surface, like its roughness and degree of polycrystallinity, and most importantly, on anion adsorption [54]. In the present work, the non-ideal capacitive behavior of SAM1 film/electrode interface may be associated with adsorption of

I.A. Akinbulu et al. / Electrochimica Acta 55 (2010) 7085–7093

7091

Table 3 Summary of impedance data obtained for the bare and SAM modified gold electrodes at applied potential of 0.1 V vs. Ag|AgCl. Electrode Brae Au SAM-1 SAM-2 SAM-3

RS / 454 695 553 444

± ± ± ±

RCT / 12 14 4 2

5920 ± 148 6080 ± 206 None None

[Fe(CN6 )]3− anion, with the physical nature of the film influencing the kinetic of adsorption. This position is supported by reports that almost ideal capacitive behavior was observed at a deliberately roughened Pt surface [54]. The conclusion was that anion adsorption, rather than the nature of the electrode surface is central to non-ideal capacitive behavior; as a result of the frequencydependent adsorption pseudocapacitive nature associated with anion chemisorptions and associated kinetics of the process. Impedance spectra of SAM-2 and -3 modified gold electrodes (Fig. 4A(c) and (d), respectively) completely exhibited a diffusion-limited behavior, eliminating contribution from charge transfer-mediated impedance. The most suitable equivalent circuit that mimic the physical electrochemistry of these cells is shown in Fig. 4B(c), where RS , Cdl and ZW have the same meanings as defined previously. Table 3 shows the fitted experimental impedance data, inclusive of their percentage errors. Differences in values of impedance parameters suggest the conducting pathways of the films are closely associated with the central metal of the phthalocyanine complexes constituting the films. SAM-2 and -3 modified gold electrodes offered no resistance to electron transfer (RCT ) while SAM-1 modified gold electrode, with the highest RCT , significantly inhibited electron transfer in the [Fe(CN)6 ]3−/4− redox couple. This is consistent with the highest value of E (80 mV vs. Ag|AgCl), obtained for the [Fe(CN)6 ]3−/4− redox process on SAM-1 modified Au electrode. Expectedly, the resistance of the electrolyte solution (RS ) does not show any significant variation (444–695 , Table 3) for the bare and SAM modified gold electrodes. Bode plots (Fig. 5) offered more insights into the electrical properties of the bare and SAM modified gold electrodes. Phase angle and frequency of the peak due to the bare gold electrode (Fig. 5A) are 55.7◦ and 500 Hz, respectively. In the presence of the SAM films, the phase peaks shifted to lower frequencies (between 2 and 16 Hz with phase angles between 36.7 and 49.3◦ , Table 3), Fig. 5B–D. The shift of the phase peak to lower frequency shows that electron transfer rate decreases as a result of electrode modification. The shifted phase peak was observed more clearly for SAM-1 in Fig. 5b and the phase peaks were less clearly resolved for SAM-2 and SAM-3, Fig. 5C and D, this is con-

Fig. 5. Bode plots (phase angle vs. log f) obtained for (a) bare, (b) SAM-1, (c) SAM-2 and (d) SAM-3 modified gold electrodes in 1 mM solution of [Fe(CN)6 ]3− containing 0.1 M KCl as supporting electrolyte. Applied potential = 0.10 V vs. Ag|AgCl.

ZW / s−1/2

Phase angle/◦

f/Hz −5

4.80 ± 0.14 × 10 None 6.02 ± 0.03 × 10−5 5.54 ± 0.01 × 10−5

500 16 13 2

± ± ± ±

2.5 0.05 0.03 0.01

55.7 49.3 39.1 36.7

± ± ± ±

0.28 0.16 0.10 0.08

sistent with the diffusion-controlled nature of their impedance behavior. 3.4. Microscopy studies of the SAMs Surface natures of the SAM films were probed using SEM and AFM (in the non-contact mode). Gold-coated glass was employed, in place of gold electrode, for probing surface properties of the SAM films. SEM and AFM are used to give information beyond the surface blocking characteristics, such as whether the MPcs are aggregated or form a uniform layer on the gold surface. SEM images (see Supporting information files, Fig. 3) of gold-coated glass before and after formation of SAM-3 (representative of the other SAMs) show some differences. The bare gold-coated glass depicts a very rough surface while the SAM modified gold-coated glass shows a relatively smooth surface. The difference in topographies of the surfaces may be ascribed to charging effects resulting from SAM formation. AFM images (see Supporting information files, Fig. 4) of the gold-coated glass before and after formation of SAM-2 (representative of the other SAMs) also indicate significant differences in topographies of the surfaces. The mean roughness and thickness of the gold-coated glass before SAM formation are 0.44 nm and 0.004 ␮m, respectively, while the mean roughness and thickness after SAM formation are 2.20 nm and 0.045 ␮m, respectively, suggesting reconstruction of the gold-coated surface in the presence of the SAM film. 3.5. Electrocatalytic oxidation of carbofuran and mechanism of electrocatalysis Carbofuran is not electrochemical active, but the electroactive phenolic analogue was formed by hydrolysis as explained in Section 2. The best voltammetry signal of carbofuran was observed at pH 4, the SAM films were also very stable at this pH, hence the use of this pH value for voltammetry studies. Fig. 6 shows the voltammetry responses for irreversible oxidation of 200 ␮M hydrolyzed solution of carbofuran on bare (dashed line) and SAM modified gold electrodes. Oxidation potentials and current responses of carbofuran on the SAM modified gold electrodes (SAM-1 = 0.40 V, 1.12 ␮A, SAM-2 = 0.42 V, 1.62 ␮A and SAM-3 = 0.48 V, 1.88 ␮A),

Fig. 6. Cyclic voltammetry profiles of bare (dashed line) and MPc-SAM modified gold electrodes in 200 ␮M solution of carbofuran (pH 4). Scan rate: 100 mV s−1 .

7092

I.A. Akinbulu et al. / Electrochimica Acta 55 (2010) 7085–7093

by Eq. (5) [55]. b = 2.3RT/˛na F

(5)

where na is the number of electrons involved in the ratedetermining step and ˛ is the electron transfer coefficient. Tafel slopes of 118, 132 and 220 mV, Table 2 (inclusive of their margin of errors) were obtained for SAM-1, SAM-3 and SAM-2, modified gold electrodes, respectively. The corresponding values of ˛ are 0.5, 0.45 and 0.27 (Table 2) for SAM-1, SAM-3 and SAM-2 modified gold electrodes, respectively. This suggests low probability of forming product during oxidation of carbofuran, especially on SAM-2 and SAM-3 modified gold electrodes. The maximum value of ˛ possible is 1, the closer to unity the value of ˛, the higher the probability of product formation in a particular electrochemical reaction. Tafel slope for SAM-1 modified gold electrode is within the normal value of 30–120 mV. Tafel slope within this range is indicative of one-electron transfer process, as the rate-limiting step, and the absence of substrate–catalyst interaction. Outer-sphere mechanism is thus proposed for the catalysis of electro-oxidation of carbofuran on SAM-1 modified gold electrode (Eqs. (6) and (7)). The Fe3+ /Fe2+ species, implicated in catalysis, was observed on SAM-1 modified gold electrode in pH 4 buffer solution (E1/2 = +0.11 V vs. Ag|AgCl).

Fig. 7. (A) RDE voltammograms for the oxidation of 200 ␮M solution of carbofuran, pH 4, at 1600 rpm, on (a) SAM-2, (b) SAM-3 and (c) SAM-1 modified gold electrodes. Scan rate: 20 mV s−1 . (B) Tafel plots for the oxidation of 200 ␮M solution of carbofuran, pH 4, on (a) SAM-2, (b) SAM-3 and (c) SAM-1 modified gold electrodes, Ik /␮A values were obtained from the RDE voltammograms in (A).

Table 2, were individually better than that on bare gold electrode (0.67 V, 0.74 ␮A), underlining the electrocatalytic properties of the SAM films. Metal-based processes of the adsorbed complexes, in pH 4 buffer alone, were observed at E1/2 = +0.11, +0.13 and −0.21 V for SAM-1, -2 and -3 modified gold electrodes, respectively (Table 2), suggesting involvement of these processes in electrocatalysis of carbofuran. This is discussed later in this work. The current signals indicated above were background-corrected. Importantly, there was gradual decrease in these signals, on the bare and SAM modified gold electrodes, during repeated cyclic voltammetry scans. This can be attributed to passivation of the electrode surface, resulting from the deposition of polyphenylene oxide, obtained from polymerization of the oxidation product of phenol. Interestingly, severity of this occurrence was curtailed on the SAM modified gold electrodes, relative to that on bare gold electrode. Mechanism of electrocatalytic oxidation of carbofuran was investigated by rotating disc electrode voltammetry. Fig. 7A shows the rotating disc electrode voltammograms of the SAM modified gold electrodes ((a) SAM-2, (b) SAM-3 and (c) SAM-1) in 200 ␮M hydrolyzed solution of carbofuran (pH 4) at 1600 rotation per minute (rpm). Mechanistic information, such as Tafel slopes and electron transfer coefficient, were obtained from the Tafel plots (plots of  (EP ) vs. log Ik ) (Fig. 7B), where Ik is the kinetic current corrected for mass transport (Ik = (I × IL )/(IL − I)) and  is overpotential, IL and I are the limiting current (current obtained at 0.42 V vs. Ag|AgCl) and the current at the foot of the wave (Fig. 7A), respectively. Ik is linearly related to overpotential by the Tafel equation (Eq. (4)).  = a + b log Ik

(4)

where  is overpotential, a is exchange current density (Io ) and b is Tafel slope. For an irreversible process, like the oxidation of carbofuran on the SAM films, Tafle slope (b) is expressed

FeII Pc → [FeIII Pc]+ + e−

(6)

[FeIII Pc]+ + RO− → FeII Pc + RO•

(7)

RO represents carbofuran. High values of Tafel slope, obtained for SAM-2 and SAM-3 modified gold electrodes, have been reported for the electrocatalytic oxidation and trace detection of amitrole using a Nafion/lead-ruthenium oxide pyrochlore chemically modified electrode (239 mV) [56]. Also, oxidation of glucose, in alkaline solution, on a RuO2 -carbon paste composite electrode [57] and oxidation of ␤-cyanoethyl ether on a platinum electrode in solution of sulfuric acid [58] have been reported to have high value of Tafel slope (∼240 mV). This observation was interpreted in terms of strong substrate–catalyst interaction, resulting in the formation of substrate-catalyst-intermediate. A strong substrate–catalyst interaction is thus envisaged on SAM-2 modified gold electrode, but weak interaction on SAM-3 modified gold electrode. This suggests likely existence of substrate–catalyst intermediate step in the oxidation of carbofuran on SAM-2 and SAM-3 modified gold electrodes. Inner sphere mechanism is thus proposed (Eqs. (8)–(10)) for the catalysis of electro-oxidation of carbofuran on SAM-2 modified electrode (representative of that on SAM-3). Co3+ /Co2+ species, observed on SAM-2 modified gold electrode, in pH 4 buffer solution (E1/2 = +0.13 V vs. Ag|AgCl), is implicated in catalysis. CoII Pc → [CoIII Pc]+ + e−

(8)

[CoIII Pc]+ + RO− → CoIII Pc-RO−

(9)

CoIII Pc-RO− → CoII Pc + RO•

(10)

4. Conclusions Electrochemical, impedimentary and surface properties of self-assembled monolayer films of non-peripherally tetra-substituted cobalt, iron and manganese(III) acetate-(2diethylaminoethanethio) phthalocyanine complexes are reported. Formation of the SAM films was confirmed by inhibition of common faradaic processes usually associated with bare gold surface, and differences in the topographies of the surfaces of the gold-coated glass before and after SAM formation. EIS suggested the electrical natures of the films were closely related to the central metal in the phthalocyanine macrocycles constituting the

I.A. Akinbulu et al. / Electrochimica Acta 55 (2010) 7085–7093

films. The films exhibited interesting electrocatalytic properties towards the electro-oxidation of the pesticide, carbofuran. Current responses of carbofuran were a function of the nature of the central metal in the phthalocyanine macrocycles constituting the SAM films. Mechanistic investigation suggested substrate–catalyst interaction in the electro-oxidation of carbofuran on SAM-2 and SAM-3 modified gold electrodes. Acknowledgements This work was supported by the Department of Science and Technology (DST) and National Research Foundation (NRF) of South Africa through DST/NRF South African Research Chairs Initiative for Professor of Medicinal Chemistry and Nanotechnology and Rhodes University. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.electacta.2010.06.065. References [1] Y. Tse, P. Janda, H. Lam, J. Zhang, W.J. Pietro, A.B.P. Lever, J. Porphyr. Phthalocya. 1 (1997) 3. [2] J.H. Zagal, M. Gulppi, M. Isaacs, G. Cardenas-Jiron, M.J. Aguirre, Electrochim. Acta 44 (1998) 1349. [3] S. Griveau, J. Pavez, J.H. Zagal, F. Bedioui, J. Electroanal. Chem. 497 (2001) 75. [4] P. Janda, J. Weber, L. Dunsch, A.B.P. Lever, Anal. Chem. 68 (1996) 960. [5] J. Zagal, F. Bedioui, J.P. Dodelet (Eds.), N4-Macrocyclic Metal Complexes, Springer, New York, 2006. [6] S. Griveau, F. Bedioui, Electroanalysis 13 (2001) 253. [7] M.J. Cook, Pure Appl. Chem. 71 (1999) 2145. [8] J. Zagal, J. Electroanal. Chem. 109 (1980) 389. [9] J.H. Zagal, P. Herrera, Electrochim. Acta 30 (1985) 449. [10] M.J. Aquirre, M. Isaacs, F. Armijo, L. Basaez, J.H. Zagal, J. Electroanal. 14 (2002) 356. [11] A. Goux, F. Bedioui, L. Robbiola, M. Pontie, Electroanalysis 15 (2003) 969. [12] J. Obirai, T. Nyokong, Electrochim. Acta 49 (2004) 1417. [13] J. Wang, Anal. Lett. 29 (1996) 1575. [14] K.I. Brown, H.A. Mottola, Langmuir 14 (1998) 3411. [15] Y. Tse, P. Jadan, H. Lam, A.B.P. Lever, Anal. Chem. 67 (1995) 981. [16] I.A. Akinbulu, T. Nyokong, Electrochim. Acta 55 (2009) 37. [17] J. Obirai, T. Nyokong, J. Electroanal. Chem. 573 (2004) 77.

[18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58]

7093

Y.H. Tse, A.B.P. Lever, Anal. Chem. 66 (1994) 384. D. Xie, Y. Jiang, J. Jiang, Z. Wu, Y. Li, Sens. Actuators B: Chem. 77 (2001) 260. M.A. Meineke, J.D. Gezelter, J. Phys. Chem. B 105 (2001) 6515. K. Ozoemena, P. Westbroek, T. Nyokong, J. Porphyr. Phthalocya. 6 (2002) 98. M.P. Somashekarappa, S. Sampath, Chem. Commun. (2002) 1262. D.J. Revell, I. Chambrier, M.J. Cook, D.A. Roussell, J. Mater. Chem. 10 (2000) 31. M.P. Somashekarappa, J. Keshavayya, S. Sampath, Pure Appl. Chem. 74 (2002) 1609. S. Nitahara, N. Terasaki, T. Akiyama, S. Yamada, Thin Solid Films 499 (2002) 354. E. Solomon, T. Angot, N. Papageorgioi, J.M. Layet, Surface Sci. 596 (2005) 74. K. Ozoemena, P. Westbroek, T. Nyokong, Electrochem. Commun. 3 (2001) 529. K. Ozoemena, T. Nyokong, Electrochim. Acta 47 (2002) 4035. Z. Li, M. Lieberman, Supramol. Sci. 5 (1998) 485. T. Akiyama, S. Nitahara, S. Inoue, S. Yamada, Photochem. Photobiol. Sci. 3 (2004) 26. K. Ozoemena, T. Nyokong, Michrochem. J. 75 (2003) 241. I.A. Akinbulu, T. Nyokong, Polyhedron 29 (2010) 1257. I.A. Akinbulu, T. Nyokong, Inorg. Chim. Acta., in press. J. Barsoukov, R. Macdonald, Impedance Spectroscopy: Theory, Experiment, and Applications, 2nd ed., John Wiley and Sons, Inc., Hoboken, NJ, 2005 (Chapter 1). B.A. Boukamp, Solid State Ionics 20 (1986) 31. I.A. Akinbulu, T. Nyokong, Polyhedron 28 (2009) 2831. B.O. Agboola, K.I. Ozoemena, T. Nyokong, Electrochim. Acta 51 (2006) 4379. F. Matemadombo, M.D. Maree, K.I. Ozoemena, P. Westbroek, T. Nyokong, J. Porphyr. Phthalocya. 9 (2005) 484. A.R. Ozkaya, A.G. Gurek, A. Gul, A.O. Bekaroglu, Polyhedron 16 (1997) 1877. M.J. Stillman, T. Nyokong, in: C.C. Leznoff, A.B.P. Lever (Eds.), Phthalocyanines: Properties and Applications, vol. 1, VCH, New York, 1989. K. Ozoemena, T. Nyokong, P. Westbroek, Electroanalysis 15 (2003) 1762. K. Nishiyama, S. Tahara, Y. Uchida, S. Tanoue, I. Taniguchi, J. Electroanal. Chem. 478 (1999) 83. B. Agboola, K.I. Ozoemena, P. Westbroek, T. Nyokong, Electrochem. Commun. 9 (2007) 310. J.E. Hutchison, T.A. Postlethwaite, R.W. Murray, Langmuir 9 (1993) 3277. J. Zak, H. Yuan, M. Ho, L.K. Woo, M.D. Porter, Langmuir 9 (1993) 2772. R.R. Durand, F.C. Anson, J. Electroanal. Chem. 134 (1982) 273. H.O. Finklea, in: A.J. Bard, I. Rubinstein (Eds.), Electroanalytic Chemistry, vol. 19, Marcel Dekker, New York, 1996, pp. 109–335. J.H. Zagal, M.A. Guilppi, C. Depretz, D. Lelievre, J. Porphyr. Phthalocya. 3 (1999) 355. N. Kobayashi, P. Janda, A.B.P. Lever, Inorg. Chem. 31 (1992) 5172. M. Ebadi, Electrochim. Acta 48 (2003) 4233. M.D. Porter, T.B. Bright, D.L. Allara, C.E.D. Chidsey, J. Am. Chem. Soc. 109 (1987) 3568. X. Wu, W. Zang, J. Electroanal. Chem. 383 (1995) 1. X. Wu, W. Zang, H. Yu, J. Electroanal. Chem. 398 (1995) 1. J. Barsoukov, R. Macdonald, Impedance Spectroscopy: Theory, Experiment, and Applications, 2nd ed., John Wiley and Sons, Inc., Hoboken, NJ, 2005 (Chapter 4). M.E.G. Lyons, C.A. Fitzerald, M.R. Smyth, Analyst 119 (1994) 855. J.-M. Zen, A.S. Kumar, M.-R. Chang, Electrochim. Acta 45 (2000) 1691. D.A. Van Galen, M. Majda, Anal. Chem. 60 (1988) 1549. B. Wermeckers, F. Beck, Electrochim. Acta 30 (1985) 1491.