One-step preparation of bifunctionalized surfaces by bipolar electrografting

Abstract

Bipolar electrochemistry (BPE) is widely used to trigger electrochemical reactions on conducting objects without direct electrical wiring. In this study a novel methodology is reported, which for the first time allows simultaneous deposition of two different organic films at each end of a glassy carbon substrate (1 × 1 cm²). The approach is based on the use of an organic bifunctional molecule, which may be oxidatively and reductively electrografted at the same time. The reduction process goes through the diazonium group, while the oxidation proceeds via the primary amine. The double functionalized plates are investigated by ellipsometry, cyclic voltammetry, condensation imaging, X-ray photoelectron spectroscopy, and time-of-flight secondary ion mass spectrometry. Post-modification of one of the anchoring layers illustrates the versatility of the system, pointing to its potential use in fields going from molecular electronics to targeted drug delivery.

---

Introduction

Bipolar electrochemistry is a well-established technique,^1,2 which has recently attracted increasing interest in several fields, including fabrication of wireless microelectrodes,³ motion of objects,^4,5 doping of electroactive polymers,⁶ and corrosion screening.⁷ The versatility of this technique allows several kinds of surface modifications to be carried out, e.g. electrodeposition and/or electrodissolution of metals,^8,9 fabrication of patterned polymer brushes,¹⁰ and electrooxidation of monomers to generate polymer films.¹¹

Most electrodeposition applications of BPE utilize a reduction or oxidation process at one side of the electrode to achieve surface modification, along with a non-grafting counter process at the opposite side. However, a few examples exist of asymmetric Janus-type objects, where both ends of the bipolar electrode are modified.^12,13 To achieve this, either polarity reversal of the electric field is used, or two different kinds of molecules are added to the solution, one for reductive grafting and the other for oxidative grafting.

In this study a novel bipolar electrografting methodology is reported, which accomplishes, by using only a single grafting agent, concomitant deposition of two chemically different organic layers at opposite ends of a glassy carbon (GC) substrate. This result is achieved in a single step based on a simultaneous reductive and oxidative grafting of an appropriately designed bifunctional organic molecule. Using one molecule instead of two is advantageous, because undesired spontaneous reactions, when the two molecules are mixed together in the same solution, may be avoided. The asymmetric Janus objects created in this straightforward way are of potential interest for applications in areas such as sensing, new electronic devices, and catalysis.¹⁴

To carry out a BPE experiment, a conducting object (the bipolar electrode, BE) is placed in an electric field between two feeder electrodes.¹⁵ The resulting polarization of the object with respect to the solution is proportional to the electric field and the dimensions of the BE. The electrode is a conductor, and therefore its potential is (nearly) identical everywhere on its surface. However, the interfacial potential difference between the BE and the solution varies along its length due to the presence of the electric field in solution. These polarization potentials drive two opposite electrochemical reactions at the poles of the BE.¹⁶ The potential difference, ΔV, between the two ends of the BE under the influence of an external electric field is given by eqn (1).

In this expression ΔE and L denote, respectively, the potential difference and the distance between the feeder electrodes, and l is the length of the BE, as illustrated in Fig. 1a. To carry out the two reactions at the opposite ends of the BE, ΔV has to be at least equal to the difference between the thermodynamic potentials of the two electrode processes, but usually higher values are needed to overcome slow kinetics.

Fig. 1 Drawings illustrating (a) the principle of BPE and (b) the interfacial potential polarization between the solution and the BE that promotes bipolar reactions at the two opposite sites. Furthermore, the decrease of the driving force along the BE from the edge along the substrate is shown assuming a linear potential profile.

4-(2-Aminoethyl)benzenediazonium was selected as the bifunctional electrografting agent. This molecule was chosen, because it combines in the same molecule two different functionalities that can be used to electrograft organic layers. Radicals that allow the grafting can be generated following two different pathways; by reduction of the diazonium moiety,¹⁷ and by oxidation of the amine.¹⁸ The well-known reduction of the diazonium group¹⁹ may take place at one end of the BE, while the primary amine, another classical grafting agent, will be oxidized at the opposite end (see Scheme 1). The versatility of the system is further investigated by post-modification of one of the anchoring layers. Additionally, it is shown that this technique is not limited to 4-(2-aminoethyl)benzenediazonium, but that other molecules can be designed and employed as well.

Scheme 1 Simultaneous grafting of two different organic films on glassy carbon using bipolar electrochemistry with 4-(2-aminoethyl)benzenediazonium.

Experimental section

Materials

Acetonitrile (HPLC grade, ≥99.9%), dichloromethane (HPLC grade, ≥99.8%), 4-(2-aminoethyl)aniline (97%), 4-aminophenylacetic acid (98%), trichloroacetyl chloride (99%), and triethylamine (99%) were purchased from Sigma-Aldrich. The diazonium salts were synthesized according to procedures published elsewhere.^20,21

Electrodes

Glassy carbon (GC) plates (Sigradur G, HTW, 10 mm × 10 mm × 1 mm) were cleaned by sonication in Milli-Q water, HPLC grade acetone, and pentane (10 min in each solvent) prior to ellipsometry measurements and bipolar grafting. Graphite feeder electrodes (Alfa Aesar, 1 mm thick, 97% metals basis) were cut into pieces with dimensions of 5 × 1 cm². In between experiments they were polished using a 1000 grit sandpaper.

Ellipsometry

For thickness measurements a Nanofilm EP3 nulling ellipsometer equipped with a high-pressure arc lamp and an interference filter wheel covering the wavelength range from 370–1000 nm was employed. The incident angle was set to 65°. Because the measurements was carried out on a dried and, therefore, collapsed film, the refractive index of the organic film is fixed at a constant value of 1.55, independent of the thickness. The clean plates are measured prior to modification at x = 0, x = 0.5, and x = 1 as the zero values. After bipolar grafting ellipsometry is performed again on the same three spots.

Optical microscopy and condensation imaging

A Macroscope Z16 APO from Leica equipped with a 0.5× objective and a DFC295 camera was used for optical microscopy. Imaging of water condensation on the modified substrate was carried by breathing on the samples and acquiring an image after a few seconds, when the equilibrium was reached.

X-ray photoelectron spectroscopy (XPS)

XPS analyses were conducted using a Kratos Axis Ultra-DLD spectrometer (Kratos Analytical Ltd, Manchester, UK) with a monochromatic Al Kα X-ray source at a power of 150 W with an analysis area of 300 × 700 μm². Survey spectra were acquired by accumulating two sweeps in the 0–1350 eV range at a pass energy of 160 eV. High-resolution scans were acquired by accumulating four sweeps at a pass energy of 20 eV. Spectral processing was carried out using CasaXPS software (Casa Software Ltd, Teignmouth, UK). Atomic surface concentrations were determined from the survey spectra following a Shirley background subtraction using the manufacturer’s sensitivity factors. Binding energies of the components in the spectra were determined by calibrating against the C=C, sp² peak for C 1s at 284.7 eV. The systematic error is estimated to be in the order of 5–10%.

Time-of-flight secondary ion mass spectrometry (ToF-SIMS)

ToF-SIMS analysis was achieved using a TOF.SIMS 5 (ION-TOF GmbH, Münster, Germany) instrument. Static SIMS condition with a total ion dose <10¹³ ions cm⁻² per analysis was employed using a 25 keV Bi₃⁺ primary ion beam with a 0.35 pA ion current and cycle time of 100 μs. The ion beam operated in the high current bunched mode for high spectral resolution >10⁴ at low mass (m/z = 29 u). In total 256 pixels mm⁻¹ were collected for all images with a field of view of 10.5 × 10.5 mm² using macro-rastering mode. All spectra and images were normalized by the total ion count and are presented based on corrected intensities. Fragments of known composition such as CH₃⁺, C₂H₅⁺, C₇H₇⁺, OH⁻, CH⁻, and C₄H⁻ were used for mass calibration.

Bipolar grafting of 4-(2-aminoethyl)benzenediazonium

4-(2-Aminoethyl)benzenediazonium tetrafluoroborate (5.9 mg, 0.025 mmol) was dissolved in 5 mL acetonitrile to obtain a 5 mM solution. A homemade bipolar electrochemistry cell made from polydimethylsiloxane (PDMS) (see Fig. S1†) was used for the grafting experiments. In this setup the GC bipolar electrode is positioned in the center, while the right and left compartments each accommodate a graphite feeder electrode. After filling the cell with the diazonium solution, the desired ΔE (15 V, unless otherwise stated) was applied for 5 min. Finally, the grafted plate was rinsed thoroughly with acetonitrile.

Bipolar grafting of 4-(carboxymethyl)benzenediazonium

4-(Carboxymethyl)benzenediazonium tetrafluoroborate (2.5 mg, 0.1 mmol) was dissolved in 10 mL acetonitrile to obtain a 10 mM solution. The same PDMS cell, as above described, was used for the experiments. After filling the cell with the diazonium solution, ΔE = 10 V was applied for 5 min. Finally, the grafted plate was rinsed thoroughly with acetonitrile.

Post-modification of bipolar grafted substrate

The bifunctionalized substrate obtained from the bipolar electrografting of 4-(2-aminoethyl)benzenediazonium was immersed in a solution of 0.5 M trichloroacetyl chloride and 0.05 M triethylamine in dichloromethane at room temperature for 2 h. The substrate was rinsed with HPLC grade acetone and sonicated in HPLC grade acetone and in water with 0.1 M NaBF₄ (10 min in each).

Results and discussion

In the case of 4-(2-aminoethyl)benzenediazonium, the peak potential of the oxidation process, E_p,ox = 1.51 V vs. SCE, and the reduction process, E_p,red = −0.56 V vs. SCE, was found by cyclic voltammetry (see Fig. 2a and b and Fig. S2, ESI†). Consequently, the polarization has to generate a minimum potential difference of ΔV_min = |E_p,red − E_p,ox| = 2.07 V in order to drive this particular dual electrografting. According to eqn (1), for a BPE cell with L = 4 cm, the bipolar grafting at the two opposite ends of a 1 × 1 cm² GC plate (l = 1 cm) thus requires at least ΔE = 8 V to trigger the reaction. This value should be considered as a lower limit for ΔE, since additional potential drops occur at both feeder electrode/solution interfaces. Furthermore, the absence of supporting electrolyte in these experiments can lead to additional potential drops and perturbations of the electrical field.

Fig. 2 Cyclic voltammograms of a 5 mM solution of 4-(2-aminoethyl)benzenediazonium at a sweep rate of 0.1 V s⁻¹ in 0.1 M Bu₄NBF₄/CH₃CN showing (a) the oxidative grafting of the amine functionality, and (b) the reductive grafting of the diazonium moiety. Optical images of BE (c) before and after water condensation for ΔE equal to (d) 11, (e) 13, and (f) 15 V. The left side of the BE corresponds in all cases to x = 0 and the right side to x = 1.

To investigate the potential dependence, the bipolar grafting of 4-(2-aminoethyl)benzenediazonium in acetonitrile was carried out using three different values of ΔE, i.e. 11, 13, and 15 V. Although it is not straightforward to reveal the grafted areas with optical microscopy (see Fig. 2c), the first evidence of a successful grafting was provided by optical images recorded of the substrates after water condensation. As shown in Fig. 2d–f the difference in wettability of grafted (hydrophilic) and ungrafted (hydrophobic) regions makes the grafting features quite visible. It also can be seen from this series of figures that for higher potential differences the grafted areas are more hydrophilic (Fig. 2f). This can be explained by an increase of the density of molecules immobilized.

At the same time the width of the grafted areas also increases as a function of ΔE, with a tendency of having the greatest impact on the oxidatively grafted region. This shows that the presence of the two electroactive functionalities of 4-(2-aminoethyl)benzenediazonium leads to a composition gradient along the BE. In addition, the dry state thickness, d, of the films was measured using ellipsometry at the oxidation site (x = 0), in the middle of the BE (x = 0.5), and at the reduction site (x = 1). As can be seen in Table 1, d increases at both ends of the BE as a function of ΔE, while only small changes are observed in the middle. This is in good agreement with what is expected, considering the variation of the overpotential, η, along the length of the BE (see illustration in Fig. 1b).²² If the kinetics of the oxidation and reduction reactions are the same, the anodic and cathodic reactions zones should be equal in size and at the center of the BE, x₀ = 0.5, no reaction should take place.

Table 1 Dry film thicknesses (d/nm) at various positions, x, on the substrate as a function of applied potential, ΔE; x = 0 corresponds to the oxidation site, x = 0.5 to the middle of the substrate, and x = 1 to the reduction site

ΔE	x = 0 d/nm	x = 0.5 d/nm	x = 1 d/nm
11 V	1.8 ± 0.0	0.5 ± 0.0	1.8 ± 0.1
13 V	2.8 ± 0.1	0.1 ± 0.0	3.0 ± 0.0
15 V	3.4 ± 0.1	0.0 ± 0.2	3.7 ± 0.3

---

The exact position of x₀ may shift during an experiment in order to keep the total rates of the oxidation and reduction processes equal at all time. Considering that the electrochemical rate of the two grafting processes most likely decreases differently as the grafting proceeds, x₀ will move to keep the overall charge balance. For the case that one end gets totally blocked, the active length, l, shrinks, and the grafting will come to a halt, once ΔV becomes too small to drive the two opposite redox reactions. Hence, both the grafting efficiency and electron transfer blocking properties of the grafted layers play central roles in BPE.¹

X-ray photoelectron spectroscopy (XPS) was employed to determine the surface concentration, Γ, at x​ = 1, according to eqn (2).²³

In this expression denotes the ratio of the intensities of the N 1s peak and the C 1s peak, while the factor 7.3 × 10⁻⁹ mol cm⁻² is the carbon atom surface density based on the assumption that it may be set equal to that of basal plane graphite.²³ It then follows that in this case, Γ = 2.4 × 10⁻¹⁰ mol cm⁻², which is similar to previously reported results for electrografting of 4-(2-aminoethyl)benzenediazonium.²⁴

Imaging, using time-of-flight secondary ion mass spectrometry (ToF-SIMS), was performed on a sample obtained with ΔE = 15 V to demonstrate and identify the immobilization of the two different functionalities, i.e. the diazonium and amine functionalities. First, Fig. 3a shows the sum of the fragments CH₂NH₂⁺ (m/z = 30 u) and C₂H₄NH₂⁺ (m/z = 44 u) originating from 4-(2-aminoethyl)benzene, the presence of which would be expected only in the case of a successful grafting of the diazonium salt. Although these fragments are observed at both ends, the region on the right side (x = 1) is broader and much more intense. The alkylamine observed on the left side may well be due to the initial molecule trapped in the immobilized organic layer. The same phenomenon has been observed during the simultaneous deposition of polypyrrole and copper, where copper was detected inside the polypyrrole film.¹² Fig. 3b shows that the fragment CN₂⁻ (m/z = 40 u) is detected in a rather narrow region at x = 0, which is consistent with the presence of diazonium groups after the oxidative grafting of the primary amine. Hence, it can be concluded that a dual functionalization of the GC substrates has been successfully accomplished. At the same time it may be noted that the oxidatively grafted area (via the amine group) to the left is much narrower than the reductively grafted area (via the aryl diazonium salt) to the right, as observed in Fig. 2d–f. This difference can be explained by the different kinetics of the oxidation and the reduction reaction. The diazonium reduction is more efficient and, therefore, occupies a larger surface area.

Fig. 3 Normalized ToF-SIMS images for a sample obtained with ΔE = 15 V, showing (a) the sum of CH₂NH₂⁺ and C₂H₄NH₂⁺ and (b) CN₂⁻ for the electrografted BE and (c) CCl₃⁻ for the post-modified sample. Field of view is 10.5 × 10.5 mm² with x = 0 being at the left side and x = 1 at the right side of each image.

To further demonstrate the versatility of this system a post-modification of one of the anchoring layers was carried out. It is well-known that surfaces modified with primary amines or diazonium functionalities can be further modified, i.e. the diazonium moieties could be used for further coupling to other surfaces in a subsequent step.^20,25,26 In this study an acyl substitution was carried out by reacting the surface amine groups with trichloroacetyl chloride in dichloromethane. Line-scan XPS was used to analyze the post-modified BE every 0.5 mm in the range from x = 0–1 (i.e. 1 cm in total). Fig. 4 shows the profiles of the atom% of Cl and N. At x = 1 the amine has been converted into the corresponding amide with a CCl₃ functionality. In comparison, less than 0.3% and 0.9% of Cl was observed at x = 0.5 and 0, respectively, showing that no or, in the latter case, very little post-modification had occurred in these regions. In fact, it is plausible that the secondary amines obtained from grafting of the primary amine group at x = 0 are able to react to a small extent with trichloroacetyl chloride. In a control experiment with a blank sample the Cl content was found to be <0.3%. Furthermore, the intensity map of the fragment CCl₃⁻ (m/z = 117 u) in ToF-SIMS imaging of the same post-modified sample confirmed these results (Fig. 3c).

Fig. 4 XPS profile showing the atomic percentage of Cl (■) and N (□), recorded every 0.5 mm across a post-modified BE (i.e. a grafted BE reacted with trichloroacetyl chloride).

The profile of the atom% of N shows a characteristic center region at 0.25 < x < 0.55 (denoted region II) with low values, equal to the level for a blank GC plate. At 0 < x < 0.25 the oxidatively grafted region (denoted I) can be revealed with the atom% of N rising sharply to almost 4%. The reductively grafted region III at 0.55 < x < 1 occupies a broader section with a maximum of ∼4%. Hence, the experimental ratio of atom% of N at x = 0 and 1 is 1 : 1, while the theoretically expected one should be 3 : 1; 3 N atoms would be expected at x = 0; 1 from the secondary amine and 2 from the diazonium group, whereas at x = 1 only the N from the primary amine is expected. The experimental ratio of 1 : 1 may be explained by the instability of the diazonium group both during and after the grafting. Deconvolution of the XPS N 1s was attempted, but did not reveal any particular trend in the binding mode of N in the three regions.

In summary, the condensation experiments, the ToF-SIMS images, and the XPS profiles lead to a consistent overall picture, where the ungrafted region II with x₀ ≈ 0.4 is somewhat off-centered from the midpoint value of 0.5. Also, the width of the oxidatively grafted region I is roughly half the width of the reductively grafted region III. However, as discussed elsewhere the balancing of the two processes is more complicated and should also include faradaic depolarization effects.^27–30 This results in nonlinear electric fields close to the BE due to the low ion (i.e. diazonium salt) concentration and its gradual removal by grafting.

It is expected that the dual grafting procedure presented in this work can easily be extended to include other molecules than 4-(2-aminoethyl)benzenediazonium. Previously, it has been shown that 4-(carboxymethyl)benzenediazonium can be grafted via reduction of the diazonium group or oxidation of the carboxylate group, respectively.²¹ To investigate the suitability of this particular bifunctional molecule in the BPE setup, a few experiments were carried out in the same manner as described for the 4-(2-aminoethyl)benzenediazonium salt (see Fig. S3 and Scheme S1, ESI†). XPS analysis shows that the atom% of O of the modified GC surface at x = 0.5 equals 2, which is the same value as found for a bare GC plate.³¹ At the oxidation site (i.e. x = 0) XPS reveals a high oxygen content of 5.0 ± 0.2 atom%, which is surprising, considering that the oxidation of the carboxylate group should result in grafting of the benzyl part with a concomitant expulsion of carbon dioxide.²¹ We believe that this may simply be ascribed to the fact that oxygen is present during our grafting process, in that we have chosen to use the simplest cell set-up possible with free access to air. This is confirmed by the observation that we also observe an increase in the atom% of oxygen at x = 0 (not included here) when 4-(2-aminoethyl)benzenediazonium is the precursor. Most importantly, at the reduction site (i.e. x = 1), where the diazonium salt group is reduced, the oxygen content becomes as high as 13.6 ± 0.3 atom% due to the presence of the carboxylic acid groups in the grafting agent.

Fig. 5 shows the deconvolution of the C 1s core XPS spectrum recorded at x = 1 (see Fig. S4, ESI† for similar spectra at x = 0 and x = 0.5). At x = 0.5 the main contribution originates from sp²-hybridized carbon, intrinsically present on the GC substrate. Similarly, at the oxidation site (i.e. x = 0) sp²-hybridized carbon is the main contributor, but as expected sp³-hybridized carbon is also observed, because of the introduction of the benzyldiazonium group during the oxidative grafting. For the reduction site (i.e. x = 1) the C 1s peak is broad due to contributions from C–O, O–C=O, and C=O bonds. The contribution from O–C=O confirms the presence of the COOH group. In particular, the introduction of this group opens up numerous possibilities for carrying out post-modifications.

Fig. 5 C 1s core level XPS spectrum for a BE grafted using 4-(carboxymethyl)benzenediazonium at the reduction site (i.e. x = 1). The spectrum is fitted using an asymmetric peak shape for sp²-hybridized C and symmetric peak shapes for the remaining contributions.

Conclusions

In conclusion, we have succeeded in performing simultaneous covalent grafting of two different organic films on glassy carbon. This is possible by the appropriate design of a grafting agent with two functional groups prone to be grafted via an oxidative or reductive route, respectively, using a bipolar setup. Specifically, in this study it was shown that immobilization of diazonium salts on one side and amines or carboxylic acids at the other side of a conducting object can be accomplished. Selective post-modification of the covalently grafted functional groups opens up further possibility for developing surfaces with a more complex chemical composition fulfilling specific requirements of given applications in various fields such as analysis or catalysis.

Acknowledgements

The Danish Council for Strategic Research is thankfully acknowledged for financial support (DA-GATE DSF 12-131827). Also, we are deeply appreciative of the generous financial support from the Danish National Research Foundation (grant no. DNRF118). Dr Dario Bassani's help with ellipsometry measurements is highly appreciated.

Notes and references

1. S. E. Fosdick, K. N. Knust, K. Scida and R. M. Crooks, Angew. Chem., Int. Ed., 2013, 52, 10438–10456.
2. G. Loget, D. Zigah, L. Bouffier, N. Sojic and A. Kuhn, Acc. Chem. Res., 2013, 46, 2513–2523.
3. K.-F. Chow, F. Mavré, J. A. Crooks, B.-Y. Chang and R. M. Crooks, J. Am. Chem. Soc., 2009, 131, 8364–8365.
4. G. Loget and A. Kuhn, Nat. Commun., 2011, 2, 535.
5. M. Sentic, G. Loget, D. Manojlovic, A. Kuhn and N. Sojic, Angew. Chem., Int. Ed., 2012, 51, 11284–11288.
6. Y. Ishiguro, S. Inagi and T. Fuchigami, J. Am. Chem. Soc., 2012, 134, 4034–4036.
7. S. Munktell, L. Nyholm and F. Björefors, J. Electroanal. Chem., 2015, 747, 77–82.
8. J.-C. Bradley and Z. Ma, Angew. Chem., Int. Ed., 1999, 38, 1663–1666.
9. J.-C. Bradley, H.-M. Chen, J. Crawford, J. Eckert, K. Ernazarova, T. Kurzeja, M. Lin, M. McGee, W. Nadler and S. G. Stephens, Nature, 1997, 389, 268–271.
10. N. Shida, Y. Koizumi, H. Nishiyama, I. Tomita and S. Inagi, Angew. Chem., Int. Ed., 2015, 54, 3922–3926.
11. S. Kong, O. Fontaine, J. Roche, L. Bouffier, A. Kuhn and D. Zigah, Langmuir, 2014, 30, 2973–2976.
12. G. Loget, V. Lapeyre, P. Garrigue, C. Warakulwit, J. Limtrakul, M.-H. Delville and A. Kuhn, Chem. Mater., 2011, 23, 2595–2599.
13. Z. Fattah, P. Garrigue, V. Lapeyre, A. Kuhn and L. Bouffier, J. Phys. Chem. C, 2012, 116, 22021–22027.
14. A. Walther and A. H. E. Müller, Soft Matter, 2008, 4, 663–668.
15. M. Fleischmann, J. Ghoroghchian, D. Rolison and S. Pons, J. Phys. Chem., 1986, 90, 6392–6400.
16. F. Mavré, K.-F. Chow, E. Sheridan, B.-Y. Chang, J. A. Crooks and R. M. Crooks, Anal. Chem., 2009, 81, 6218–6225.
17. M. Delamar, R. Hitmi, J. Pinson and J. M. Savéant, J. Am. Chem. Soc., 1992, 114, 5883–5884.
18. R. S. Deinhammer, M. Ho, J. W. Anderegg and M. D. Porter, Langmuir, 1994, 10, 1306–1313.
19. D. Bélanger and J. Pinson, Chem. Soc. Rev., 2011, 40, 3995–4048.
20. M. Ceccato, L. T. Nielsen, J. Iruthayaraj, M. Hinge, S. U. Pedersen and K. Daasbjerg, Langmuir, 2010, 26, 10812–10821.
21. H. Hazimeh, S. Piogé, N. Pantoustier, C. Combellas, F. I. Podvorica and F. Kanoufi, Chem. Mater., 2013, 25, 605–612.
22. J. Duval, J. M. Kleijn and H. P. van Leeuwen, J. Electroanal. Chem., 2001, 505, 1–11.
23. Y.-C. Liu and R. L. McCreery, J. Am. Chem. Soc., 1995, 117, 11254–11259.
24. T. Breton and D. Bélanger, Langmuir, 2008, 24, 8711–8718.
25. P. Viel, X. T. Le, V. Huc, J. Bar, A. Benedetto, A. Le Goff, A. Filoramo, D. Alamarguy, S. Noël, L. Baraton and S. Palacin, J. Mater. Chem., 2008, 18, 5913–5920.
26. B. D. Assresahegn, T. Brousse and D. Bélanger, Carbon, 2015, 92, 362–381.
27. J. F. L. Duval, G. K. Huijs, W. F. Threels, J. Lyklema and H. P. van Leeuwen, J. Colloid Interface Sci., 2003, 260, 95–106.
28. J. F. L. Duval, M. Minor, J. Cecilia and H. P. van Leeuwen, J. Phys. Chem. B, 2003, 107, 4143–4155.
29. J. F. L. Duval, H. P. van Leeuwen, J. Cecilia and J. Galceran, J. Phys. Chem. B, 2003, 107, 6782–6800.
30. J. F. L. Duval, J. Colloid Interface Sci., 2004, 269, 211–223.
31. M. Lillethorup, K. Torbensen, M. Ceccato, S. U. Pedersen and K. Daasbjerg, Langmuir, 2013, 29, 13595–13604.

---

Footnote

† Electronic supplementary information (ESI) available: Cell design, cyclic voltammetry, and X-ray photoelectron spectroscopy. See DOI: 10.1039/c5ra20156j

---