solution interface of ex situ prepared bismuth film electrodes: A scanning electrochemical microscopy (SECM) and atomic force microscopy (AFM) investigation

Electrochimica Acta 53 (2007) 555–560

Reactivity at the film/solution interface of ex situ prepared bismuth film electrodes: A scanning electrochemical microscopy (SECM) and atomic force microscopy (AFM) investigation Samo B. Hoˇcevar a , Salvatore Daniele b,∗ , Carlo Bragato b , Boˇzidar Ogorevc a a

Analytical Chemistry Laboratory, National Institute of Chemistry, P.O. Box 660, SI-1001 Ljubljana, Slovenia b Department of Physical Chemistry, University of Venice, Calle Larga, S. Marta, 2137, 30123 Venice, Italy Received 29 May 2007; received in revised form 4 July 2007; accepted 6 July 2007 Available online 22 July 2007

Abstract Bismuth film electrodes (BiFEs) prepared ex situ with and without complexing bromide ions in the modification solution were investigated using scanning electrochemical microscopy (SECM) and atomic force microscopy (AFM). A feedback mode of the SECM was employed to examine the conductivity and reactivity of a series of thin bismuth films deposited onto disk glassy carbon substrate electrodes (GCEs) of 3 mm in diameter. A platinum micro-electrode (φ = 25 ␮m) was used as the SECM tip, and current against tip/substrate distance was recorded in solutions containing either Ru(NH3 )6 3+ or Fe(CN)6 4− species as redox mediators. With both redox mediators positive feedback approach curves were recorded, which indicated that the bismuth film deposition protocol associated with the addition of bromide ions in the modification solution did not compromise the conductivity of the bismuth film in comparison with that prepared without bromide. However, at the former Bi film a slight kinetic hindering was observed in recycling Ru(NH3 )6 3+ , suggesting a different surface potential. On the other hand, the approach curves recorded by using Fe(CN)6 4− showed that both types of the aforementioned bismuth films exhibited local reactivity with the oxidised form of the redox mediator, and that bismuth film obtained with bromide ions exhibited slightly lower reactivity. The use of SECM in the scanning operation mode allowed us to ascertain that the bismuth deposits were uniformly distributed across the whole surface of the glassy carbon substrate electrode. Comparative AFM measurements corroborated the above findings and additionally revealed a denser growth of smaller bismuth crystals over the surface of the substrate electrode in the presence of bromide ions, while the crystals were bigger but sparser in the absence of bromide ions in the modification solution. © 2007 Elsevier Ltd. All rights reserved. Keywords: Bismuth film electrode; Reactivity; Ex situ; SECM; AFM; NaBr

1. Introduction Since its introduction in stripping analysis [1], the bismuth film electrode (BiFE) has been accepted in many electroanalytical laboratories worldwide [2]. The electroanalytical performance of the BiFE, which approaches, or in some cases even surpasses that of mercury analogues, has encouraged many scientists to further investigate and expand its scope and applicability for measurement of numerous heavy metals and some selected organic compounds [3–5]. Different kinds of bismuth electrodes have been proposed, including in situ and ex situ pre-

∗

Corresponding author. Tel.: +39 0412348630; fax: +39 0412348594. E-mail address: [email protected] (S. Daniele).

0013-4686/$ – see front matter © 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.electacta.2007.07.035

pared BiFE [1,6], bismuth bulk electrode (BiBE) [7], and several modifications of bismuth-based paste electrodes [8,9]. Among various types of bismuth film electrode, the ex situ prepared bismuth films require adequate physical and chemical stability, as they have to be transferred from the preparation/modification solution to the measuring cell device, and usually need to exhibit enhanced stability for multiple measurements. Recently, it was observed that the addition of complexing bromide ions into the modification solution is beneficial for the ex situ preparation of a bismuth film in connection with its electrochemical stripping operation [10–12]. This deposition route, reported for the first time by Krolicka et al. [10], has provided BiFEs of higher (physical) stability and enhanced reproducibility of measurements associated with the electrochemical stripping analysis. However, whether or not the latter improvements could also be

556

S.B. Hoˇcevar et al. / Electrochimica Acta 53 (2007) 555–560

ascribed to an enhanced chemical stability of the ex situ prepared bismuth deposits in contact with the electrolyte solution has not been investigated yet. Scanning electrochemical microscopy (SECM) [13] is a technique that has been exploited to obtain valuable information concerning spatial localization of chemical reactions of a variety of substrate surfaces and thin solid films [14–16]. This can be done by using SECM operating in the feedback mode [17], where a micro-electrode tip-current is perturbed from its value in the bulk solution by the presence of substrate in a close proximity. In particular, when the micro-electrode tip approaches an insulator, the current decreases with the tip/substrate distance due to blockage of diffusion of the electroactive species toward the tip (negative feedback). When the tip approaches a conductor, the current increases with decreasing the distance due to regeneration of the redox mediator at the substrate/solution interface (positive feedback). This regeneration may be due to lateral charge-transfer process or it may proceed by reaction between the substrate and the tip-generated species at the substrate/solution interface [18]. Therefore, the plots of the current against the tip/substrate distance (approaching curves) along with their analysis may provide useful information about the nature and reactivity of the investigated substrate [13]. In this paper we utilized SECM operating in both feedback and scanning mode to probe the local properties of bismuth films, which were deposited ex situ on a glassy carbon substrate electrode (GCE) either in the presence or absence of complexing bromide ions in the modification solution. The GCE has been chosen, as it is typically employed as a substrate for deposition of bismuth films [3]. The SECM investigation is performed with the substrate being unbiased, thus allowing the measurements to be carried out on the bismuth films as grown in pristine conditions, without any direct electrochemical perturbation. Two different redox mediators hexaaminoruthenium(III) trichloride and potassium hexacyanoferrate(II) having different formal potentials were employed, thus allowing us to investigate the effect of redox couple upon the SECM feedback characteristics of the ex situ prepared bismuth films. In addition, AFM was used to illustrate the influence of the modification solution conditions upon the bismuth film growth and its morphology. 2. Experimental 2.1. Apparatus Unless otherwise stated, voltammetric measurements were performed in a three-electrode electrochemical cell. A glassy carbon electrode (GCE) of 3 mm in diameter was employed as the working/substrate electrode for deposition of bismuth film, an Ag/AgCl (sat. KCl) served as the reference electrode and a platinum spiral wire was used as the counter electrode. The surface of the glassy carbon electrode was polished to a mirrorlike appearance with alumina powder (0.3 and 0.05 ␮m) on a polishing pad. For SECM measurements a platinum micro-disk of 25 ␮m in diameter was employed as the tip-electrode. The platinum micro-disk was prepared by sealing a platinum wire of 25 ␮m in diameter into a glass capillary, which was afterward

tapered to a conical shape, such that the overall tip to electrode radius ratio (RG) was equal to 10. The micro-disk was polished with graded alumina powder of different sizes (1, 0.3 and 0.05 ␮m) on a polishing pad. The SECM micro-positioning device consisted of a set of three stepper motor stages with a 0.1 ␮m resolution (MICOS) with optical encoder (ZEISS), and the motion was controlled by a closed loop motion controller board PCI-7324 (National Instruments). The data acquisition was performed by a PCI-6035E Multifunction I/O board controlled with Lab View (National Instruments). A CH700B workstation (CH Instruments) was employed for SECM measurements and for other electrochemical experiments. All measurements were carried out in a three-electrode electrochemical cell placed in a Faraday cage. AFM measurements were performed in a Veeco electrochemical cell with MultiMode V in combination with NanoScope V controller and Veeco Univecpot Bipotentiostat. A glassy carbon pellet (φ = 1 cm) was used as the substrate electrode for electrochemical deposition of bismuth films, an Ag/AgCl wire was employed as the quasi-reference electrode, and a platinum wire as the counter electrode. 2.2. Reagents and solutions All chemicals employed were of analytical-reagent grade. K4 Fe(CN)6 , KCl, CH3 COOH, CH3 COONa and NaBr were obtained from Aldrich. Hexaamminoruthenium(III) trichloride was purchased from J. Mattey and used as received. All solutions were prepared by using water, which was purified via a Milli-Q system (Gradient, Millipore, Bedford, USA). Acetate buffer solution (0.1 M, pH 4.5) was prepared by mixing together appropriate amounts of CH3 COOH and CH3 COONa. All SECM measurements were performed in aerated aqueous solutions containing 0.1 M KCl as a supporting electrolyte. All measurements requiring no oxygen were carried out in solutions that had been purged with pure nitrogen (99.99%) provided by Siad, Italy. 2.3. Ex situ preparation of BiFEs The ex situ BiFE was prepared by electrodeposition of metallic bismuth onto a glassy carbon substrate electrode from a 0.1 M acetate buffer solution (pH 4.5) containing 100 mg/L bismuth(III) by applying a potential of −1.0 V for 5 min while the solution was stirred. BiFE in the presence of bromide ions was prepared similarly from a 0.1 M acetate buffer solution (pH 4.5) containing 50 mg/L bismuth(III) and 50 mg/L NaBr by applying a potential of −0.3 V for 60 s. The above plating conditions were adopted from previous investigations, since they provided longterm electrochemical and mechanical stability of the bismuth film as well as its good electroanalytical stripping performance [12]. Other details regarding the ex situ BiFE preparation can be found also in Ref. [12]. Following its preparation, BiFE was transferred directly into the cell of SECM device. In the case of AFM measurements, the bismuth film was grown directly in an AFM electrochemical cell onto the surface of a glassy carbon substrate electrode as described above.

S.B. Hoˇcevar et al. / Electrochimica Acta 53 (2007) 555–560

557

Scheme 1. Lateral electron-transfer and positive feedback at an unbiased substrate for O/R redox mediator. Fig. 1. Approach curves recorded at different substrates with Ru(NH3 )6 3+ as redox mediator; bare GCE (—); BiFE prepared without (· · ·) and with (- - -) bromide ions.

3. Results and discussion Preliminary, SECM experiments were performed using a bare unbiased substrate GCE with the tip-electrode held at −0.4 V and +0.5 V for Ru(NH3 )6 3+ and Fe(CN)6 4− , respectively, which were used as redox mediators, and typical normalized approach curves, i.e., normalized current I (I = i/il∞ , i is the current at given tip/substrate distance d and il∞ is the current in the bulk solution) against normalized distance L (L = d/a, a is the radius of micro-electrode) obtained are shown in Figs. 1 and 2 (full lines). Under these conditions, the positive feedback effect was observed in all cases in accordance with regeneration mechanism characteristics of a conductive substrate [13–18] presented in Scheme 1 for the case of generic O(oxidised)/R(reduced) couple. The positive feedback arises since a significant area of the substrate away from the tip is bathed in a solution whose concentration has not been perturbed by the tip-reaction [13–18]. Thus, in the latter zone, the substrate is poised at potentials established by the species in the bulk solution (i.e., the oxidant for Ru(NH3 )6 3+ or the reductant for Fe(CN)6 4− ), thus at open circuit its potential is positive or negative with respect to the

Fig. 2. Approach curves recorded at different substrates with Fe(CN)6 4− as redox mediator; bare GCE (—); BiFE prepared without (· · ·) and with (- - -) bromide ions.

halfwave potential (E1/2 ) of the corresponding redox couple. When the tip approaches the substrate, a portion of the substrate directly under the tip is exposed to the electrode reaction product, which then instantaneously undergoes the inverse redox reaction. It must be noted that the potential of an unbiased substrate is also perturbed by current flow, which is required for recycling the mediator [18]. Thus, the feedback response can also be perturbed by finite heterogeneous electron-transfer kinetics [18]. The lateral electron transport of Scheme 1 is operative in conductive substrate whose surface area is larger than that of the micro-tip counterpart, the condition, which is also fulfilled in the SECM experiments considered in this paper. To further confirm that the positive feedback effect was due to aforesaid properties, additional approaching curves were recorded in the same redox solutions above the Teflon insulator, which surrounds the GCE surface. In this case, as expected, typical negative feedback approach curves were obtained (not shown). In addition, considering again the approaching curves in Figs. 1 and 2, it is evident that the Ru(NH3 )6 3+/2+ redox couple yields larger currents. This phenomenon can be attributed to faster heterogeneous electron-transfer rate that characterizes the latter redox couple with respect to the Fe(CN)6 4−/3− couple [19–21]. It was also verified that the experimental approach curve recorded in the solution of Ru(NH3 )6 3+ fit the theoretical diffusion controlled one, whereas some discrepancies arose with that recorded with the Fe(CN)6 4− . This feature further confirms the above view on the shape of the approach curves recorded with two different redox mediators. Measurements similar to those described above were performed on bismuth films, which were prepared ex situ on GCE in the presence and in the absence of NaBr. Typical currentdistance curves are included in Figs. 1 and 2 for Ru(NH3 )6 3+/2+ and Fe(CN)6 4−/3− redox couples, respectively. Also in this case, positive feedback responses were obtained, corroborating the conductive nature of investigated bismuth films. However, the approach curves displayed significantly higher currents at low tip/substrate distances, when the Fe(CN)6 4− species was employed as the redox mediator (compare curves with a dashed line in Figs. 1 and 2). These results can be explained by considering redox potentials of all species, which were involved in these measurements (see Table 1 for standard redox potentials). Since the standard redox potential of bismuth is between those

558

S.B. Hoˇcevar et al. / Electrochimica Acta 53 (2007) 555–560

Table 1 Redox potentials of the species involved in SECM measurements Redox couple 3−/4−

Fe(CN)6 Ru(NH3 )6 3+/2+ Bi3+/0 BiCl4 − /Bi0 + 4Cl−

E◦ (V) (vs. NHE)

Ref.

0.36 0.05 0.294 0.160

[5] [6] [8] [8]

of Ru(NH3 )6 3+/2+ and Fe(CN)6 4−/3− redox systems, a chemical redox reaction may take place between metallic bismuth film and the Fe(CN)6 3− which is locally/electrochemically formed directly under the electrode tip: Tip reaction :

Fe(CN)6 4− ⇔ Fe(CN)6 3− + e

(1)

Substrate reaction : 3Fe(CN)6 3− + Bi0 ⇔ 3Fe(CN)6 4− + Bi3+

(2)

The reaction pathway (1) and (2) ultimately provides an electrocatalytic cycle, which is responsible for enhanced feedback current. No such cycle was instead operative when Ru(NH3 )6 3+ was employed as a redox mediator. It must be considered that the enhanced current observed with the Fe(CN)6 4− redox system cannot be attributed to the reduction of Bi3+ ions at the tip-electrode, which are produced in the solution within the tip/substrate gap, as its reduction process occurs at more negative potentials than those applied at the SECM micro-tip [22]. Considering again the plots in Figs. 1 and 2, it can be seen, that with both redox mediators a small difference exists between the approach curves, recorded at the bismuth films prepared with (dashed line) and without (dotted line) bromide ions. The approach curves obtained at the surface of bismuth films, which were formed in the presence of bromide ions, exhibit, at the same distance L, slightly lower currents in comparison with those prepared without bromide ions. This can be attributed either to a different surface potential (for the Ru(NH3 )6 3+ species) [18], or to a slightly higher chemical stability of bismuth film (for the Fe(CN)6 4− species), which was obtained in the presence of bromide, consistently with previous findings [12] concerning the stability and performance in ASV analysis. 3.1. SECM and AFM images A series of SECM and AFM images of BiFEs were also recorded in order to ascertain whether the adopted plating conditions provided the films with uniform chemical reactivity and/or physical characteristics. Fig. 3 shows typical SECM images, which were recorded at the GCE either before or after its modification with Bi films with Ru(NH3 )6 3+ as redox mediator. They were obtained by positioning the platinum micro-tip at about 10 ␮m above the surface of the substrate and then scanned across the x–y plane parallel with the surface. Similar images were recorded also by using Fe(CN)6 4− (not shown). The SECM images are essentially featureless, showing in all cases positive feedback responses throughout the surface. However, statistical comparison of normalized currents provided some useful infor-

Fig. 3. SECM images obtained at different substrates with Ru(NH3 )6 3+ as redox mediator; bare GCE (a), BiFE prepared without (b) and with (c) bromide ions.

mation. The root mean square current fluctuation (Rq ) obtained within the area analysed of the various substrates was calculated by the following expression:   (Zi − Zave )2 Rq = N where Zave is the average of Z values within the given area, Zi the normalized current Z value, and N is the number of points within the area. Results are listed in Table 2, which also include the average normalized current for each sample. These data indicate a flattening of the surface with Bi film obtained by using the plating protocol with bromide ions, which results in more uni-

S.B. Hoˇcevar et al. / Electrochimica Acta 53 (2007) 555–560 Table 2 Average normalized current and root mean square current fluctuation (Rq ) obtained within the area 200 ␮m × 200 ␮m of the indicated substrates Substrate

Zave a

Rq a

Zave b

Rq b

Bare GCE BiFE BiFE + NaBr

1.22 2.23 1.87

0.05 0.11 0.03

1.32 1.43 1.28

0.03 0.09 0.02

a b

Redox mediator: Fe(CN)6 4− . Redox mediator: Ru(NH3 )6 3+ .

form reactivity of the surface. Higher current fluctuations and higher average normalized current values were instead observed in the images recorded at the Bi film substrate produced by using the plating protocol without bromide ions. These results agree with the above findings obtained through the approach curves and once again suggest an enhanced chemical stability of the Bi film obtained in the presence of NaBr. To confirm that the plating protocol also provided physically more uniform films, a series of AFM measurements were carried out on Bi films prepared similarly as for the SECM measurements. Fig. 4 shows the AFM images of bare GCE (a), BiFE prepared without bromide ions (b), and BiFE prepared in the presence of bromide ions in the modification solution (c). All images were recorded in 0.1 M acetate buffer solutions containing either 100 mg/L of Bi3+ ions (b) or 50 mg/L of Bi3+ ions together with 50 mg/L of bromide ions (c) from which both bismuth films were directly in situ electrodeposited onto the surface of substrate GCE. It is evident that the bismuth film, which was grown without bromide ions in the modification solution (b), exhibits relatively large and sparse bismuth crystals, as indicated by the higher surface roughness (i.e., darker (blue) nuances) of the Bi film which are deposited

559

uniformly over the substrate surface. In contrast, the deposition protocol in the presence of bromide ions yielded a denser growth of smaller bismuth crystals (i.e., lighter nuances), which were also distributed uniformly on the surface of the substrate GCE. This phenomenon can be ascribed, and is also consistent with earlier observation [23], to different stabilities of bismuth hydroxo- and halogen-complexes, which affect the deposition of bismuth film, and/or to the formation of bismuth bromide moieties, which stabilize the bismuth film, and consequently preclude a further growth of bigger crystals [23]. In addition, this assumption is also consistent with the aforementioned SECM experiments, which revealed slightly lower chemical activity of bismuth film prepared in the presence of bromide ions. 4. Conclusions The results presented in this work suggest that the introduction of bromide ions during ex situ preparation of bismuth films enhances the stability of ex situ BiFEs, without adversely affecting their conductivity and electrochemical characteristics. These findings are consistent with earlier results, which showed a significantly enhanced stability of an ex situ prepared BiFE under its electrochemical stripping operation attributed to a more uniform bismuth film formation in the presence of bromide ions. This was confirmed in the present investigation by SECM experiments by recording approach curves and employing scanning operation mode and in addition by AFM experiments, which were performed on bismuth films produced under identical experimental conditions as those for SECM measurements. Acknowledgements Financial support from the Slovenian Research Agency (P1-0034) and MUR (Italian Ministry of the University and Research) is gratefully acknowledged. References

Fig. 4. AFM recorded at different substrates; bare GCE (a), BiFE prepared without (b) and with (c) bromide ions in the modification solution.

[1] J. Wang, J. Lu, S.B. Hocevar, P.A.M. Farias, B. Ogorevc, Anal. Chem. 72 (2000) 3218. [2] J. Wang, Electroanalysis 17 (2005) 1341. [3] A. Economou, Trends Anal. Chem. 24 (2005) 334. [4] E.A. Hutton, B. Ogorevc, S.B. Hocevar, F. Weldon, M.R. Smyth, J. Wang, Electrochem. Commun. 3 (2001) 707. [5] M.A. Baldo, S. Daniele, Anal. Lett. 37 (2004) 995. [6] E.A. Hutton, S.B. Hocevar, B. Ogorevc, M.R. Smyth, Electrochem. Commun. 5 (2003) 765. [7] R. Pauliukaite, S.B. Hocevar, B. Ogorevc, J. Wang, Electroanalysis 16 (2004) 719. [8] S.B. Hocevar, I. Svancara, K. Vytras, B. Ogorevc, Electrochim. Acta 51 (2005) 706. [9] K. Vytras, I. Svancara, R. Metelka, Electroanalysis 14 (2002) 19. [10] A. Krolicka, A. Bobrowski, K. Kalcher, J. Mocak, I. Svancara, K. Vytras, Electroanalysis 15 (2003) 1859. [11] A. Kr´olicka, A. Bobrowski, Electrochem. Commun. 6 (2004) 99. [12] E.A. Hutton, S.B. Hocevar, B. Ogorevc, Anal. Chim. Acta 537 (2005) 285. [13] A.J. Bard, M.V. Mirkin (Eds.), Scanning Electrochemical Microscopy, Marcel Dekker Inc., New York, Basel, 2001.

560

S.B. Hoˇcevar et al. / Electrochimica Acta 53 (2007) 555–560

[14] M.V. Mirkin, M. Arca, A.J. Bard, J. Phys. Chem. 97 (1993) 10790. [15] P. Liljeroth, D. Vanmaekelbergh, V. Ruiz, K. Konturri, H. Jiang, E. Kauppinen, B.M. Quinn, J. Am. Chem. Soc. 126 (2004) 7126. [16] A.L. Whitworth, D. Mandler, P.R. Unwin, Phys. Chem. Chem. Phys. 7 (2005) 356. [17] J. Kwak, A.J. Bard, Anal. Chem. 61 (1989) 1221. [18] D.O. Wipf, A.J. Bard, J. Electrochem. Soc. 138 (1991) 138. [19] E. Sabatani, I. Rubinstein, J. Phys. Chem. 91 (1987) 6663.

[20] S. Daniele, P. Ugo, G.A. Mazzocchin, D. Rudello, in: R.A. Mackay, J. Texter (Eds.), Electrochemistry in Colloids and Dispersions, VCH, New York, 1992, p. 55. [21] K.K. Cline, M.T. McDermott, R.L. McCreery, J. Phys. Chem. 98 (1994) 5314. [22] M.A. Baldo, S. Daniele, C. Bragato, J. Phys. IV France 107 (2003) 103. [23] I. Svancara, L. Baldrianova, M. Vlcek, R. Metelka, K. Vytras, Electroanalysis 17 (2005) 120.