Electrode passivation caused by polymerization of different phenolic compounds

Electrode passivation caused by polymerization of different phenolic compounds

Electrochimica Acta 52 (2006) 434–442 Electrode passivation caused by polymerization of different phenolic compounds Marystela Ferreira a,1 , Hamilto...

360KB Sizes 0 Downloads 5 Views

Electrochimica Acta 52 (2006) 434–442

Electrode passivation caused by polymerization of different phenolic compounds Marystela Ferreira a,1 , Hamilton Varela a , Roberto M. Torresi b , Germano Tremiliosi-Filho a,∗ a

Instituto de Qu´ımica de S˜ao Carlos, Universidade de S˜ao Paulo CP-780, CEP 13560-970 S˜ao Carlos, SP, Brazil b Instituto de Qu´ımica, Universidade de S˜ ao Paulo CP-26077, CEP 05513-970 S˜ao Paulo, SP, Brazil Received 23 March 2006; received in revised form 17 May 2006; accepted 17 May 2006 Available online 7 July 2006

Abstract An electrochemical quartz crystal microbalance (EQCM) was used to investigate the formation of polymeric products resulting from electrooxidation in aqueous solutions of phenolic compounds on the surface of Au and Pt electrodes. The studied compounds were phenol; m-cresol; 2,5-dimethylphenol and 2,3,5-trimethylphenol. The polymerization process was studied as a function of the methyl substitution in the phenolic structure. Electrochemical quartz crystal microbalance studies show that the polymer formed from substituted phenols is more passivating than that from the non-substituted phenol. In any case, the largest amount of mass was deposited during the first voltammetric cycle and the Pt electrode was more active than the Au electrode for the organic electrooxidation process. FT-IR spectroscopy showed that the films formed upon phenol and m-cresol electrolysis were also oxidized. © 2006 Elsevier Ltd. All rights reserved. Keywords: Phenols oxidation; Polymer film formation; Surface passivation process; EQCM

1. Introduction During the electrooxidation of phenolic compounds a noticeable decrease in the current is observed owing to the formation of a polymeric film on the electrode surface. This film may promote electrode passivation mainly by interfering in the supply of fresh reactants from bulk solution, removing products from the reaction zone or active sites, or decreasing the available overpotential to drive the reaction. On one side, the formation of an adherent film during phenol oxidation delays important electrode processes, such as those involving the treatment of wastewater [1–3] or the synthesis of quinines [4], otherwise, the passive layer is of interest to protect electrodes surface from corrosion [5,6]. The electrochemical oxidation of phenol is a notoriously complex process [7] and the product distribution and the reaction pathway may be affected by the following factors: concentration of phenolic compounds, electrode material, pH, current density, potential, etc. The initial stage in the oxidation of phe∗

Corresponding author.Tel: +55 16 3373 9933; Fax: +55 16 3373-9952. E-mail address: [email protected] (G. Tremiliosi-Filho). 1 Present address: DFQB/UNESP, CP-405, CEP 19090-900 Presidente Prudente, SP, Brazil. 0013-4686/$ – see front matter © 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.electacta.2006.05.025

nol after 4-electron transfer leads to the formation of quinones as showed in Schemes 1 and 2 [7,8]. During phenol oxidation to quinones, phenoxy radicals (3–5) are initially formed as intermediates as illustrated in Scheme 1. These radicals can be further oxidized to quinones (14, 14 , Scheme 1) or can react irreversible (dimerization) to form dimeric products (15–17, Scheme 3 [7,8]). Such dimers can be oxidized again to produce radical (18, Scheme 4 [7,9,10]), which can couple with the phenoxy radical (3, Scheme 1) or with other dimeric radical (18, Scheme 3) to produce the polymer (19), as shown in Scheme 4. These schemes illustrate a possible route for the polymer film formation, through the coupling C O C, though any coupling through dimers shown in Scheme 3 may occur. Electrode passivation during electrolysis of aqueous phenol solutions on a Pt electrode has been studied by Gattrell and Kirk [8,11] using gel permeation chromatography (GPC) and Fourier transform infrared spectroscopy (FT-IR). From FT-IR results they postulated the structures for different films produced by oxidation of p-cresol, o,o’-biphenol and phenol [11]. In a series of papers Jusys and co-workers investigated the electropolimerization of chlorinated phenols on Pt electrode in alkaline media by means of cyclic voltammetry, FT-IR, electrochemical quartz crystal microbalance (EQCM), and gas chromatography-mass

M. Ferreira et al. / Electrochimica Acta 52 (2006) 434–442

435

Scheme 1.

spectrometry [12–16]. Among the different aspects discussed in these papers, it was clarified that the electrode passivation is likely to be due to the surface blocking by polymeric products formed during oxidation. Moreover, the electropolymerization degree and the surface deactivation were found to depend on the structure and the permeability of the resulting polymers, which in turn is a function of the degree of chlorination and the monomer isomerism, their reactivity and polymerization pathways. EQCM is a powerful tool that has been used to monitor electrochemically induced mass changes at electrode surfaces during electrochemical processes and also to study interfacial phenomena [17–20]. Particularly, in electronically conducting polymers,

many researchers [21–24] have adopted the EQCM technique to study polymer growth and electroneutralization (charge compensation) processes during electrochemical cycling. Concerning the investigation of the processes accompanying the electrooxidation of phenolic compounds, the EQCM has been also used [14,25,26]. It should be emphasized that all these studies are focused mainly on the eletrooxidation of phenol and chlorinated phenols. Wang et al. [25], for instance, used EQCM to investigate polyphenol formation from buffered solutions with pH 7.4 on a gold surface. They observed that the polymerization efficiency depends mainly on the initial phenol concentrations, but the impact of methyl substitution and the substrate were not considered.

Scheme 2.

436

M. Ferreira et al. / Electrochimica Acta 52 (2006) 434–442

2. Experimental 2.1. Chemicals All chemicals used were of analytical grade and the solutions were prepared with ultra-high purity water (MilliPore, MilliQ system). Phenol was purchased from Synth and m-cresol, 2,5-dimethylphenol and 2,3,5-trimethylphenol were supplied by Aldrich. Chloroplatinic acid was obtained from Alfa Aesar. The solution used as supporting electrolyte was H2 SO4 (Merck) 0.5 mol L−1 . All chemicals were used without further purification. 2.2. Apparatus and procedure

Scheme 3.

To our knowledge, the present study is the first to analyze comparatively the film formation during electrooxidation of phenol and methyl-substituted phenolic compounds on different substrates (Au and Pt) using in situ EQCM method with simultaneous voltammetric scans. Additionally, the effect of the methyl substitution in the phenolic structure was also investigated. The compounds employed were phenol (model aromatic organic compound for wastewater treatment), m-cresol, 2,5-dimethylphenol and 2,3,5-trimethylphenol, as illustrated in Scheme 3. The experiments were performed on Au and Pt electrodes at pH 0.5. In addition, ex situ FT-IR was used to analyze samples of passivating films produced on the anode during the electrolysis.

The measurements were performed in a conventional, onecompartment cell with a volume of approximately 30 cm3 . The system was controlled by a Pentium microcomputer through the FAC-QCM software. The working electrode consisted of a 6 MHz AT-cut quartz crystal with piezoelectric area of 0.2 cm2 , both sides of the quartz crystals were coated with gold by vacuum deposition, but only one face was exposed to the electrolyte solution and used as working electrode. Some experiments were carried out with a Pt electrode prepared by electrochemical deposition of Pt on Au by applying a cathodic current of 150 ␮A during 700 s, and the deposited mass was 180 ± 10 ␮g cm−2 , according to EQCM measurements [20]. The solution for platinum electrodeposition was a 2 mM chloroplatinic acid in a 0.5 M H2 SO4 solution. The reference electrode was a Hg/Hg2 Cl2 , KCl(sat.), SCE, and all potential values in this work are referred to it. The auxiliary electrode consisted of a Pt sheet of 1.0 cm2 . The Sauerbrey equation [27], f = −km, was used to relate the frequency response to mass. The proportionality factor k = 5.2 × 107 cm2 Hz g−1 (or 2.6 × 108 Hz g−1 ) was obtained on the basis of the calibration using silver deposition, described previously by Gabrielli et al. [28]. The polymer samples analyzed by FT-IR were produced on a Pt sheet, 5.0 cm2 . After

Scheme 4.

M. Ferreira et al. / Electrochimica Acta 52 (2006) 434–442

437

polymerization, the electrode was rinsed with water to remove electrolyte and excess of monomer and dried under nitrogen. The measurements were carried out using a BOMEM DA8 FT spectrometer. All electrochemical and EQCM studies were carried out in 1 mmol L−1 solutions of the following phenols: phenol, mcresol, 2,5-dimethylphenol and 2,3,5-trimethylphenol dissolved in H2 SO4 aqueous solution (pH 0.5). The cyclic voltammograms were performed from 0.01 to 1.2 V using a sweep rate of 20 mV s−1 . All experiments were performed at room temperature. The polymer film thickness was measured by AFM according Lobo et al. [29] and the average value found was of ca. 70 nm. 3. Results and discussion Fig. 1a and b illustrate the j/E and f/E profiles, respectively, obtained for the first five cycles during phenol oxidation on gold and Fig. 1c shows the corresponding profiles j/E and (df/dt)/E profiles, only for the first cycle. The voltammograms were carried out at 20 mV s−1 . The double layer region of gold in 0.5 mol L−1 H2 SO4 can be identified in Fig. 1a between the potentials of 0.0 and 0.3 V. During the potential scanning in this region only a slight frequency shift of ca. 65 Hz cm−2 is observed (Fig. 1b). This frequency shift is due to changes in the double layer structure, adsorption of submonolayer of hydroxyl species [30] and eventually some (bi)sulfate adsorption [31,32]. When the potential is scanned in the range of 0.3–0.6 V a minor anodic peak is observed around 0.5 V (Fig. 1a). This peak can be associated with the electrosorption of hydroxyl and/or oxygen species on gold and possibly also some phenol adsorption. As a consequence, a more significant frequency shift of ca. 150 Hz cm−2 is observed, Fig. 1b. In any case, the observed variation in frequency occurs due to the adsorbed oxygen [33,34] and some phenol adsorption. The main observed peak that occurs at 0.9 V (Fig. 1a) is accompanied by an abrupt frequency shift of ca. 750 Hz cm−2 (Fig. 1b). This voltammetric peak corresponds formally to participation of two subsequent processes, the continuation of the gold surface oxidation and the phenol electrooxidation. The oxide formation occurs in the potential range from ca. 0.5 to 1.2 V with the growth of a quasithree-dimensional structure of AuO [35]. The subsequent phenol oxidation is a complex process that takes place on the freshly formed gold oxide (AuO) and it may act as an intermediate in the organic oxidation and is called a surface-mediator. The first step of the phenol oxidation involves the formation of an adsorbed layer of phenoxy radicals on the surface-mediator [36] (intermediates 3–5 in Scheme 1). Such radicals remain on the electrode surface and experience further oxidation forming possible many different products such as quinones, aliphatic acids, carbon dioxide [3,8,37,38], dimers [39] and a surface polymeric film through coupling reactions involving adjacent phenoxy and dimeric radicals [6,11,40] (see also Schemes 1, 3 and 4). Above 1.0 V, among the organic and gold surface oxidation, is observed the formation of molecular oxygen that begins and continues until the potential reaches the maximum value in the anodic scan (1.2 V) and the value of ca. 1.0 V during the reverse voltammetric scan (Fig. 1a). The oxygen evolution reaction proceeds only on the exposed gold surface covered with the oxide film. It is possi-

Fig. 1. j/E profiles (a) and f/E curves (b) for phenol. The j/E profile and (df/dt) vs. E curves are shown for the first cycle for phenol (c). Electrode: Au, v = 20 mV s−1 . Solution: 1 mmol L−1 of phenol in H2 SO4 0.5 mol L−1 .

ble that the external surface of the polymeric film does not give any contribution for this process once it exhibits poor conducting properties [41]. Concomitantly to the oxygen evolution, the organic oxidation and the gold oxide formation also occur. The corresponding frequency shift associated to these processes is ca. 250 Hz cm−2 (Fig. 1b). A region of a nearly constant frequency exists between 1.0 and 0.4 V during the negative potential scan (Fig. 1b). The cathodic peak at 0.35 V corresponds to the reduction of the gold oxide [35] (Fig. 1a) and gives a frequency decrease of ca. 450 Hz cm−2 (Fig. 1b). A quantitative analysis of the charge involved in the gold oxide reduction process suggests the formation of a quasi-three-dimensional structure of AuO in a broad potential range (between 0.5 and 1.0 V). It

438

M. Ferreira et al. / Electrochimica Acta 52 (2006) 434–442

should be noted that our results for the gold oxidation/reduction after correction for the roughness factor are consistent with the literature data obtained by EQCM in terms of charge density and mass shift [33]. This series of cyclic events does not restore the initial surface morphology once there is a significant frequency shift between the initial and final states of the order of 750 Hz cm−2 (Fig. 1b). The frequency difference in the whole cycle can thus be mainly attributed to the formation of a stable polymeric film that stays attached to the electrode surface even at the end of the potential cycle. Owing to the intrinsic difficulties due to the existence of parallel processes (phenol oxidation, polymeric film growth, gold surface oxidation, oxygen evolution reaction) it is not be possible to calculate the Faradaic efficiency for the formation of the polymeric film using just voltammetric measurements. Fig. 1a and b also show the current and frequency shift for the other four subsequent cycles during phenol oxidation under the same potential conditions. As can be seen the largest variations were observed during the first cycle, consistent with the fact that the most significant contribution to the electrode passivation process occurs during this cycle. Upon increasing the number of cycles, the frequency variation per cycle becomes smaller. The Sauerbrey equation is used to estimate the mass changes if there is no appreciable change in the viscoelastic properties of the film [20], as it is usually assumed for polyphenol films [14,25]. In the first cycle, the mass increase occurred mainly between 0.4 and 1.2 V, reaching 4.6 ␮g cm−2 . At the cathodic scan the mass remains constant up to 0.4 V, after this potential a decrease of ca. 1.7 ␮g cm−2 is observed. The mass difference in the whole cycle, approximately 2.9 ␮g cm−2 , is attributed to the accumulative formation of the polymeric film. This last value corresponds to the observed frequency variation between the initial and final states in the first voltammetric cycle (750 Hz cm−2 ) allowing us to conclude that during the phenol oxidation mainly the polymer is formed as product. In other words, under the studied conditions the preferred reaction path for the phenol oxidation is the polymerization reaction. Fig. 1c illustrates the (df/dt) versus E (associated to mass flux) as a function of the potential as well as the profile j/E for the first cycle. The shape of both curves is very similar in the anodic and cathodic scans. The anodic branch of the j/E profile is associated with the gold and phenol oxidation and the polymer film formation. On the other hand, the anodic branch for the (df/dt) versus E curve is only due to the mass flux. Thus, the mass flux is associated only with the anodic charge due to the anodic peaks of the cyclic voltammogram. Similarly, the mass flux in the cathodic scan is clearly connected with the cathodic charge of the reduction peak. This behavior observed for the cathodic d(f/dt)/dt variation suggests an independence of the mass flux observed during the gold oxide reduction on the stable polymeric film attached to the surface. From Fig. 1c, it is clear that the details of the mass flux reduction profile correspond to that observed for the voltammetric reduction of a oxidized surface of the bulk-metal (Au). These results therefore imply that the polymeric film is microscopically porous once allows to the EQCM to detect the mass flux during reduction of the gold oxide, process that occurs underneath of the remaining polymeric film,

independently and without any complication due to its presence. Supporting this conjecture, Babai and Gottesfeld [41] observed by ellipsometric studies the same porous nature of the polymeric film grown at constant potential. Additionally they suggest that the electron transport is the rate-determining step during the film formation. M-cresol and phenol show quite similar voltammetric behavior. However, voltammetric differences can be observed during the electrooxidation of the other substituted phenolic compounds, viz. 2,5-dimethylphenol and 2,3,5-trimethylphenol. Fig. 2 shows the electrooxidation behavior for the 2,3,5trimethylphenol. In this case the j/E profile exhibits a large

Fig. 2. j/E profiles (a) and f/E curves (b) for 2,3,5, trimethylphenol. The j/E profile and (df/dt) vs. E curves are shown for the first cycle for 2,3,5trimethylphenol (c). Electrode: Au, v = 20 mV s−1 . Solution: 1 mmol L−1 of 2,3,5-trimethylphenol in H2 SO4 0.5 mol L−1 .

M. Ferreira et al. / Electrochimica Acta 52 (2006) 434–442

anodic peak in 0.35 V during the first cycle, c.f. Fig. 2a. Besides the electrosorption of hydroxyl/oxygen species on gold electrode, this peak can be associated with a significant organic adsorption. These processes are accompanied by an abrupt frequency shift (350 Hz cm−2 ) in the potential range of 0.3 and 0.5 V (Fig. 2b). The remaining features of the voltammograms are similar to those observed for the phenol oxidation. A result of great significance in this work is concerned with the cathodic branch of f/E profiles. In contrast to the case of phenol, there is nearly no frequency shift during the oxide reduction (compare Fig. 1b, 2b and 3) when substituted phenols are oxidized. It seems that such lack of mass variation would be mainly due to the presence of a stable polymeric film that once formed on the gold surface reduces the mass transport through it, but does not affect the tracing out of the gold oxide reduction j/E profile. The same phenomenon is more clearly observable by comparison of Figs. 1c and 2c, where the cathodic branch of the mass flux as a function of the potential does not show the peak at 0.4 V as observed for the non-substituted phenol. This behavior suggests that the polymeric film formed on the electrode surface is almost free of pores, even though the proton can cross the film for charge compensation. Thus, it is very likely that the polymeric films formed during the electrooxidation of the substituted phenols have a considerably different structure from that obtained during the oxidation of the nonsubstituted phenol. In summary, once the electrode is covered with the non-porous polymer, the water electroformed during the reduction of the gold oxide beneath the polymer remains trapped at the interface between the polymeric film and the electrode surface and consequently the mass flux for this process is not observable. During the successive cycles for the oxidation of the substituted phenols (2nd–5th cycles, Fig. 2a) the gold re-oxidation process is not clearly observable in the corresponding anodic branch of the cyclic voltammograms, in the potential range of 0.3 and 0.7 V, as can be seen in Fig. 2a. However, the oxidation of 2,3,5-trimethylphenol is clearly noticed in the oxidation peak at 0.85 V (Fig. 2a). Once the gold electrode surface is covered by the polymeric film formed during the first cycle, the organic oxidation at the successive cycles has to occur at the external part of the remaining stable polymeric film. As for phenol film formation, the largest amount of mass was deposited in the first cycle (Figs. 2 and 3Figs. 2b and 3), and with increasing the number of cycles the deposited mass per cycle decreased. The

439

Fig. 3. f/E curves for cresol. Electrode: Au, v = 20 mV s−1 . Solution: 1 mmol L−1 of m-cresol in H2 SO4 0.5 mol L−1 .

electrooxidation of 2,5-dimethylphenol is quite similar to the behavior observed for the 2,3,5-trimethylphenol. A quantitative analysis of the frequency shift and the anodic charges involved in the different processes associated with the organic oxidation is presented in Table 1. While the frequency shift (mass change) at the first cycle are very similar for all studied compounds, ca. 800 Hz cm−2 , the anodic charge, qox , and the organic oxidation charge, qorg , are higher for phenol and decrease when the number of methyl substituting in the aromatic ring increases. In addition, after five successive potential cycles the same tendency is observed again for the accumulative values of frequency shift and anodic charges. This suggests that almost the same amount of polymer is deposited on the electrode surface during the oxidation process independently of the monomer structure, however, the polymeric films formed show quite different characteristics once they exhibit distinct passivation effects. Further insight into the differences among the film formation when different phenolic compounds are used can be given in terms of the electropolymerization efficiency. The mass of the polymer film electrodeposited at the electrode surface, mp , can be associated to the charge consumed during the organic oxidation, qorg , through the apparent number of electrons transferred per monomer unit, napp , in the following way:

Table 1 Values of qox , qorg and f, for several phenolic compounds (electrode: Au) Compound

First cycle qox

Phenol m-Cresol 2,5-Dimethylphenol 2,3,5-Trimethylphenol

(mC cm−2 )

43.0 20.1 7.7 5.9

Sum of all cycles qorg 24.6 17.6 5.9 4.7

(mC cm−2 )

−f 770 880 625 935

(Hz cm−2 )



qox (mC cm−2 )

134.0 49.3 17.3 14.5



qorg (mC cm−2 )

45.6 37.9 8.1 7.1





f (Hz cm−2 )

1665 1825 1285 1450

qox : anodic charge due to gold and organic oxidations (calculated by subtraction of the charge associated to the oxygen evolution from the total anodic charge); qorg : anodic charge due only to organic oxidation (calculated by subtraction of the reduction charge of the AuO from the qox ); −f: frequency shift (decrease) at the end of the cycle.

440

M. Ferreira et al. / Electrochimica Acta 52 (2006) 434–442

Table 2 Apparent number of moles of electrons per monomer, napp , transferred during electropolymerization of phenolic compounds (electrode: Au) Compound

napp (first cycle)

napp (all cycles)

Phenol m-Cresol 2,5-Dimethylphenol 2,3,5-Trimethylphenol

8.1 5.8 3.1 1.8

6.9 6.0 2.1 1.8

mp = qorg M/napp F, where M is the molar mass of the monomer and F is the Faraday constant. Values of napp when gold electrode was used as substrate were calculated and are displayed in Table 2. In line with the quinol–ether polymerization mechanism discussed above, it would be expected to have napp = 2, i.e., two electrons per monomer unit. As seen in Table 2, this prediction seems to hold only for the di- and trimethyl-substituted phenols. This result is in contrast to that reported for the case of tri- and pentachlorophenols where napp is considerably higher than 2 probably due to the diffusion of low molar mass oligomers from the electrode surface [14]. Hence, it seems to be plausible to extend this difference and concluded that diffusion of side products such as dimers and trimers is less important in the case of methyl-substituted phenols than that of chlorinated phenols. On the other hand, the higher napp values observed for the electropolymerization of phenol and m-cresol are rationalized in terms of the additional charge consumed in the further oxidation to quinone or more saturated structures which occurs at higher potentials. This is in agreement with the FT-IR results described below, as well as the ones discussed by Ezerskis and Jusys [15]. Fig. 4 shows the results for 2,5-dimethylphenol electrooxidation on a Pt electrode. The first five voltammetric cycles are shown in Fig. 4a. All the five successive voltammograms are very well overlapped. Similar behavior is also observed for the others organic compounds. In general, the voltammetric currents show higher values for the phenol oxidation on platinum electrode, pointing to a more significant passivating effect for the substituted phenols. The well-known peaks attributed to the deposition/oxidation of adsorbed hydrogen appear between −0.2 V and +0.1 V and the associated frequency variation is of the order of 100 Hz cm−2 (Fig. 4b). The second observable anodic process starts at ca. +0.6 V and continues until reach the potential value of ca. 0.8 V at the reverse sweep. This process can be related with the electrosorption of hydroxyl and/or oxygen species, phenols adsorption on platinum and phenols oxidation. The corresponding frequency shift for these processes is around 1950 Hz cm−2 (see Fig. 4b). As observed for gold, the frequency shift for the 2nd–5th cycles under the same potential conditions becomes smaller. During the negative scan the platinum oxide reduction peak appears at ca. +0.35 V and the associated frequency variation is of the order of 550 Hz cm−2 (Fig. 4b). In general, all the features of the voltammograms on platinum are similar to those observed for the phenol oxidation on gold as previously discussed. For platinum electrodes a similar behavior as for gold is related with the non-restoration of the initial surface morphology, as indicated by the frequency difference

Fig. 4. j/E profiles (a) and f/E curves (b) for 2,5-dimethylphenol. Electrode: Pt, v = 20 mV s−1 . Solution: 1 mmol L−1 of 2,5-dimethylphenol in H2 SO4 0.5 mol L−1 .

in the whole cycle that is attributed to the formation of a stable polymeric film attached to the platinum surface. Similarly to the observed behavior for phenols oxidation on Au electrode, the largest amount of mass was deposited during the first cycle as can be seen in Fig. 4b. Unlike the case for the oxidation of substituted phenolic compounds on gold, there was clearly a mass decrease in the cathodic scan during the platinum oxide reduction, compare Figs. 3 and 4b. This is an evidence that the microstructure of the polymeric films grown on platinum during the oxidation of substituted or non-substituted phenols is porous in nature once allows to the EQCM to detect the mass flux during reduction of the platinum oxide, process that occurs underneath of the remaining polymeric film independently of its presence. While the gold electrode shows this characteristic only for the non-substituted phenol. Once all the five successive voltammetric cycles overlap very well for all studied compounds, the anodic charges involved in each individual cycle due to platinum and organic oxidations are nearly the same. However, differences between the substituted and the non-substituted phenols (ca. 68 mC cm−2 for phenol and

M. Ferreira et al. / Electrochimica Acta 52 (2006) 434–442

20 mC cm−2 for the substituted phenols) are observed. Similarly, the same tendency is observed for the anodic charges associated only with the organic oxidation (ca. 35 mC cm−2 for phenol and 10 mC cm−2 for the substituted phenols). As for gold, the frequency shift (mass changes) at the first cycle on Pt is almost the same independently of the monomer structure, ca. 1400 Hz cm−2 . This suggests that the same amount of polymer is deposited on the electrode surface during the voltammetric cycle. The mass changes were estimate by the use of the Sauerbrey equation and an increase of about 7.5 ␮g cm−2 in the potential range of 0.2 and 0.8 V (reverse scan in the first cycle) was observed. At the platinum reduction process the mass decrease of ca. 2.3 ␮g cm−2 . The mass difference in the whole cycle, approximately 5.2 ␮g cm−2 , is attributed to the formation of a stable polymeric film on the platinum surface. Although the same amount of polymer is deposited on the electrode surface during the oxidation of the different organic molecules, the structure of the polymers clearly depends on the monomer structure once distinct passivation effects are observed. For both Au and Pt electrodes, higher currents were observed for phenol oxidation, indicating a more significant passivating effect of the substituted compounds. In other words, the polymer formed from substituted phenols is more passivating than the polymer obtained from phenol, which may be associated with effects from the substituting groups in the phenolic structure. Polymer growth occurs through the oxygen atom, as illustrated in Scheme 4. The substitution of H atoms in the position 2 by the methyl group leads to a polymer with more passivating characteristics than the polymer formed when non-substituted phenol is used (compare Figs. 1 and 2). The substitution of H’s for methyl radicals in the positions 3 and/or 5 leads to polymers with similar steric hindrance of that formed by cresol. These substitutions play a significant role in determining the structure of the polymers, suggesting a more linear polymeric chain for substituted phenols and a more ramified and open chain for the non-substituted phenol. This explains why the passivating effect is similar for all substituted compounds (m-cresol, 2,5dimethylphenol and 2,3, 5 trimethylphenol) and more effective than that observed for non-substituted phenol. The charge upon oxidation was consistently larger for the Pt electrode when compared to the Au electrode, especially for the substituted compounds. For Pt electrode, as observed previously for Au, similar charges were measured for the substituted phenolic compounds, which indicates that the activity is independent on the monomeric unit used. The frequency change was higher in the first cycle, for all compounds and for the two electrode materials employed. It was higher in the first cycle of the Pt than on the Au electrodes, thus showing that a thicker polymer film is formed upon phenol oxidation on Pt. The deposited mass was estimated as 4.6 ␮g cm−2 for phenol electrooxidation on Pt, while for Au electrode the mass was ca. 2.9 ␮g cm−2 . Finally, thicker polymeric films should be formed during phenol oxidation because the charges were higher for phenol than for the substituted organic compounds. This is to compensate the differences in the molar masses of the different compounds once the observed mass variations were almost similar for all studied compounds.

441

Fig. 5. Fourier transforms infrared spectra of films formed from phenol (polyphenol) (a) and m-cresol (polycresol) (b). Electrode: Pt. Film growth potential: 1.7 V. Film growth time: 2 h. Solution concentrations: 200 mmol L−1 (phenol) in H2 SO4 0.5 mol L−1 and saturated solution (m-cresol) in H2 SO4 0.5 mol L−1 .

In order to analyze samples of the passivating film produced on the anode during phenolic compound oxidation, infrared spectroscopy was used. In this case, the polymeric films were grown and subsequently analyzed ex situ. Films were potentiostatically grown at 1.7 V, pH 0.5, with an oxidation time of approximately 2 h and using 200 mmol L−1 of phenol and saturated solution of m-cresol. The spectrum obtained for the polymeric film upon phenol oxidation is shown in Fig. 5a. In spite of the differences in the experimental conditions, this spectrum is similar to those obtained by Gattrell and Kirk [11] who studied films obtained during oxidation of phenol, p-cresol and o,o’-biphenol. In their work, the electrolysis was performed with a relatively high concentration (often saturated) solution and the oxidation time was 1 s. The formation of a polymeric film results in the decrease of the ν(O H) vibration at 3300 cm−1 (not observed in our case) and the formation of quinone groups. The main difference between the film grown by Gattrell and Kirk and that in our studies is the absence of the OH band. This is probably due to oxidation of the polymer produced here, which is indicated by the appearance of quinone functional groups with a characteristic ν(C O) vibration around 1640–1660 cm−1 . The spectrum indicates that the polymer does not present a defined structure, probably comprising distinct couplings of dimers, C C or C O C. The oxygen is retained through an ether-like

442

M. Ferreira et al. / Electrochimica Acta 52 (2006) 434–442

structure, as shown by the strong band at 1211 cm−1 . The bands at 1480 cm−1 and 1589 cm−1 refer to the symmetrical stretching C C. The other bands are due to the C–H structures. It is believed that the film formed under the experimental conditions of our work has a larger amount of quinone groups C O than those investigated by Gattrell and Kirk. This also means that the film produced here was oxidized. The oxidation of m-cresol also produces passivating films (polycresol), whose FT-IR spectra is shown in Fig. 5b, which was obtained as described for phenol. Gattrell and Kirk [11] also studied polymer films of p-cresol. Again, the spectra are similar, in spite of the differences in the growth conditions between their work and ours. Analogously to the results for phenol, the ν(O–H) vibration is not present, i.e., the polymer is oxidized, probably due to the experimental conditions, i.e., applied potential and oxidation time, employed here. Polycresol shows an extensive peak at 1644 cm−1 , characteristic of quinone-type structures. The peak at 1481 cm−1 is characteristic of C C in the polymer in the 1,2,4,6 substitution. The ν(C C) vibration at 1439 cm−1 is typical for para or 1,2,4,6 structures. The peaks between 821 cm−1 and 1188 cm−1 refer to ν(C–H) vibrations, with the exception of the peak at 1072 cm−1 that corresponds to ν(CH3 ) vibration. The peak at 1610 cm−1 indicates the presence of C C groups. The film formed under our experimental conditions is probably oxidized, and as such it should present a larger number of quinone (C O)-type structures.

Acknowledgements We are grateful to the Brazilian agencies CNPq, PRONEX/ FINEP and FAPESP for financial support. H. Varela and M. Ferreira acknowledge FAPESP for the scholarships. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10]

[11] [12] [13] [14] [15] [16] [17] [18]

4. Conclusions The in situ EQCM method has been demonstrated as extremely valuable for studying the electropolymerization process of phenolic compounds. Polymer films grown on platinum electrode are not as passivating as those formed on gold electrode, which shows that platinum is more reactive particularly in the electrooxidation of substituted phenolic compounds. The polymeric films formed from all studied compounds on platinum electrode show porous characteristics. Similarly, the same behavior was perceived for the polymer formed on gold electrode from phenol. However, a quite different behavior was observed for films grown from substituted phenolic compounds on gold electrode indicating a non-porous material that block the water flux during the oxide reduction. For both electrodes, it was observed that the largest amount of mass is deposited during the first cycle. Additionally, it was also verified that the same amount of polymer is deposited on the electrode surface during the oxidation process independently of the monomer structure, however the different polymeric films formed show quite different characteristics once they exhibit distinct passivation effects. For both Au and Pt electrodes, higher currents were observed for phenol oxidation, indicating a more significant passivating effect of the substituted compounds. In other words, the polymer formed from substituted phenols is more passivating than the polymer obtained from phenol, which may be associated with effects from the substituting in the phenolic structure. FTIR spectra showed a C O band, indicating that the film formed during phenol and m-cresol oxidation is also oxidized, owing to the high potential and long oxidation time employed.

[19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41]

V.S. De Sucre, A.P. Watkinson, Can J. Chem. Eng. 59 (1981) 52. M. Gattrell, D.W. Kirk, Can J. Chem. Eng. 68 (1990) 997. M. Gattrell, D.W. Kirk, J. Electrochem. Soc. 140 (1993) 1534. C. Oloman, AIChE Symp. Ser. 77 (1981) 264. G. Mengoli, M.M. Musiani, J. Electrochem. Soc. 134 (1987) 643C. G. Mengoli, M.M. Musiani, Prog. Org. Coat. 24 (1994) 237. M. Ferreira, Ph.D. Thesis, Universidade de S˜ao Paulo, S˜ao Carlos, SP, Brazil (2000). M. Gattrell, D.W. Kirk, J. Electrochem. Soc. 140 (1993) 903. A. Nilsson, A. Ronl´an, V.D. Parker, J. Chem. Soc. I. (1973) 2337. A. Rol´an, in: A.J. Bard, H. Lund (Eds.), Encyclopedia of Electrochemistry of Elements, vol. XI, Marcel Dekker, Inc., New York, 1978, p. 242. M. Gattrell, D.W. Kirk, J. Electrochem. Soc. 139 (1992) 2736. Z. Ezerskis, Z. Jusys, Pure Appl. Chem. 73 (12) (2001) 1940. Z. Ezerskis, Z. Jusys, J. Appl. Electrochem. 31 (10) (2001) 1117. Z. Ezerskis, G. Stalnionis, Z. Jusys, J. Appl. Electrochem. 32 (1) (2002) 49. Z. Ezerskis, Z. Jusys, J. Appl. Electrochem. 32 (7) (2002) 755. Z. Ezerskis, Z. Jusys, J. Appl. Electrochem. 32 (5) (2002) 543. D.A. Buttry, in: A.J. Bard (Ed.), Electroanalytical Chemistry. A Series of Advances, vol.17, Marcel Dekker Inc., New York, 1991, p. 1. M. Hepel, in: A. Wieckowski (Ed.), Interfacial Electrochemistry. Theory, Experiment, and Applications, Marcel Dekker Inc., New York, 1999, p. 599. J. Lee, H. Varela, S. Uhm, Y. Tak, Electrochem. Commun. 2 (2000) 646. H. Varela, M. Malta, R.M. Torresi, Quim. Nova 23 (2000) 664. H. Varela, R.M. Torresi, J. Electrochem. Soc. 147 (2000) 665. D. Orata, D.A. Buttry, J. Am. Chem. Soc. 109 (1987) 3574. G. Inzelt, Electroanalysis 7 (1995) 895. H. Varela, R.M. Torresi, D.A. Buttry, J. Braz. Chem. Soc. 11 (2000) 32. J. Wang, M. Jiang, F. Lu, J. Electroanal. Chem. 444 (1998) 127. M.S. Ureta-Zanartu, P. Bustos, M.C. Diez, M.L. Mora, C. Gutierrez, Electrochim. Acta 46 (16) (2001) 2545. G. Sauerbrey, Z. Phys. 178 (1964) 457. C. Gabrielli, M. Keddam, R. Torresi, J. Electrochem. Soc. 138 (1991) 2657. R.F.M. Lobo, M.A. Pereira-da-Silva, M. Raposos, R.M. Faria, O.N. Oliveira Jr., Nanotechnology 10 (1999) 389. W. Kautek, M. Sahre, D.M. Soares, Ber. Bunsenges. Phys. Chem. 99 (1995) 667. P. Zelenay, L.M. Ricejackson, A. Wieckowski, J. Electroanal. Chem. 283 (1990) 389. M. Weber, F.C. Nart, Langmuir 12 (1996) 1895. Z. Jusys, S. Bruckenstein, Electrochem. Solid-State Lett. 1 (1998) 74. J.S. Gordon, D.C. Johnson, J. Electroanal. Chem. 365 (1994) 267. G. Tremiliosi-Filho, L.H. Dall’Antonia, G. Jerkiewicz, J. Electroanal. Chem. 422 (1997) 149. P.I. Iotov, S.V. Kalcheva, J. Electroanal. Chem. 442 (1998) 19. Ch. Comninellis, C. Pulgarin, J. Appl. Electrochem. 21 (1991) 703. H. Sharifian, D.W. Kirk, J. Electrochem. Soc. 133 (1986) 921. F.P. Bub, K. Wisser, W.J. Lorenz, W. Heimann, Ber. Bunsenges. Phys. Chem. 77 (1973) 823. F. Beck, Electrochim. Acta 33 (1988) 839. M. Babai, S. Gottesfeld, Surf. Sci. 96 (1980) 461.