Cyclic voltammetry on recessed nanodisk-array electrodes prepared from track-etched polycarbonate membranes with 10-nm diameter pores

Abstract

This paper described cyclic voltammograms (CVs) on a recessed nanodisk-array electrode (RNE) prepared from a track-etched polycarbonate membrane (TEPCM) with 10-nm diameter pores. A RNE was prepared via sputtering of a gold thin film onto a TEPCM whose pores were coated with polyvinylpyrrolidone. CVs of such a RNE changed from peak-shaped to sigmoidal with decreasing scan rate due to the transition of diffusion modes of redox-active molecules, as with previous reports. However, faradic currents in these CVs were larger than theoretical currents calculated from the membrane thickness, pore density, and pore diameter. The larger faradic currents may reflect the presence of a surface layer on TEPCM nanopores to which the redox species preferentially distribute as compared with the nanopore cavity. The limiting current of uncharged 1,1′-ferrocenedimethanol was not affected by solution pH or supporting electrolyte concentration. In contrast, those of anionic ferricyanide and cationic ferrocenylmethyltrimethylammonium decreased and increased, respectively, with increasing pH from 4 to 9. These pH-dependent changes in limiting current were larger at lower supporting electrolyte concentration. The dependence of the CVs on solution pH and supporting electrolyte concentrations was attributed to electrostatic interaction between the redox species and the charged TEPCM nanopores.

---

Introduction

Due to the nearly cylindrical shapes and narrow size distribution of their pores, track-etched polycarbonate membranes (TEPCMs) have been employed as filtration membranes,¹ as model systems for studying hindered mass transport,² and as templates for synthesizing nanowires and nanoelectrode ensembles.^3,4 TEPCMs incorporated into microfluidic devices could be used as diffusion-controlled separators between two channels,⁵ as potential-controlled gates^6–10 and for the electrokinetic preconcentration of molecules.^6,11,12 In addition, electroless deposition of gold on TEPCM nanopores and their subsequent chemical functionalisation with thiolate self-assembled monolayers provided a means for controlling the selectivity in molecular separation and detection based on size-exclusion and chemical interactions.^13,14 All these applications involve the mass transport of solvated molecules that reflects the surface properties (e.g., charge, hydrophobicity) of TEPCM nanopores.

Previously, the surface charge of unmodified TEPCM nanopores was assessed by measurements of streaming potential,^15–17 electroviscous effects,¹⁷ and flux of fluorescent molecules.^18,19 Results obtained using these measurements suggested the presence of surface charge on the solution-filled TEPCM nanopores that possibly originates from functional groups developed during membrane processing or adsorption of ions from aqueous solution.^16,18,19 Whereas these methods have provided information reflecting the net properties of nanopores, they have not been employed to discuss the heterogeneous nature of the nanopores (e.g., surface vs. cavity). The surface of TEPCM nanopores is often coated with a thin polyvinylpyrrolidone (PVP) layer that may play an essential role in the molecular mass transport behavior.^18,19

In contrast to the above methods, electrochemical methods have been used to discuss the heterogeneous natures of nanoporous media deposited on electrode surfaces.²⁰ A faradic current reflects the diffusion coefficients and concentrations of redox species in the media, the kinetics of the electrode reaction, the time-scale of measurements (e.g., scan rate in voltammetry), solution convection and electrode geometry.²¹ Thus, detailed analysis of faradic currents measured under controlled conditions made it possible to discuss the distinct electron/mass transport modes within nanoporous media. For example, the properties of two bicontinuous phases within anodic aluminium oxide (AAO) nanopores modified with an organic polymer or bilayer assemblies were investigated using voltammetry and chronocoulometry on AAO-based recessed nanodisk-array electrodes (RNEs).^22–25 The mass/electron transport properties and concentration of redox species within a thin organic layer on the nanopore were different from those within the nanopore cavity, resulting in observations of enhanced redox reaction.^23–25

In this study, the surface properties of 10-nm diameter TEPCM nanopores coated with PVP were investigated using cyclic voltammetry on TEPCM-based RNEs. Due to their higher surface-to-volume ratio, TEPCMs with 10-nm diameter pores were expected to offer a faradic current more significantly reflecting the nanopore surface properties as compared to those containing 50-nm diameter pores.²⁶ TEPCM-based RNEs were characterized using several different approaches, including measurements of charging and faradic currents, profilometry to measure the membrane thickness, and scanning electron microscopy (SEM) to determine the pore density and diameter. As compared to AAO membranes^{3,14,22–25,27} and nanoporous membranes derived from block copolymers^28–34 that contain an array of densely-packed cylindrical nanopores, the pore density in the TEPCM was much lower. The low pore density provided a means for characterizing the RNEs from the transition of cyclic voltammogram (CV) shapes measured at different scan rates.^26,35–39 At slower scan rates, sigmoidal CVs reflecting the steady-state, radial diffusion of redox molecules from a solution into the nanopores were observed on these RNEs, which facilitated our discussion on the electrochemical data. The charge within the TEPCM nanopores was assessed from the limiting currents in sigmoidal CVs of charged and uncharged redox species measured at different pH and supporting electrolyte concentrations.

Experimental section

Chemicals and materials

All solutions were prepared with water having a resistivity of 18 MΩ cm or higher (Barnstead Nanopore Systems). Potassium nitrate (Fisher Chemical), potassium dihydrogen phosphate (Fisher Chemical), potassium hydrogen phosphate (Fisher Chemical), potassium hydroxide (Fisher Chemical), potassium ferricyanide (K₃Fe(CN)₆; Arcos Organics), ferrocenylmethyltrimethylammonium iodide (TMAFc⁺I⁻; Strem Chemicals), and 1,1′-ferrocenedimethanol (FcDM; Aldrich) were of reagent grade quality and used without further purification. Ferrocenylmethyltrimethylammonium nitrate (TMAFc⁺NO₃⁻) was prepared by passing a TMAFc⁺I⁻ solution through NO₃⁻–form anion exchange resin (Amberlite IRA-900, Aldrich), and characterized using ion chromatography (column: Hamilton PRP-X100). TEPCMs with pore diameters of 10 nm (25 mm membrane diameter) were obtained from Osmonics. The thickness of a TEPCM was measured using a profilometer (Ambios Technology XP-2 Stylus Profiler). SEM images of a TEPCM were measured using a Zeiss Leo 32 SEM at the Microscopy & Analytical Imaging Laboratory, the University of Kansas, upon deposition of a 5-nm thick Pd/Au alloy film on the sample.

Fabrication of TEPCM-Based RNEs

TEPCM-based RNEs were prepared according to the procedures described previously.²⁶ In short, the rougher face of a TEPCM was coated with a gold layer (≥ 200nm thick) via sputtering (EFFA Coater, Ernest F. Fullam), and then the gold side was mounted on a glass slide (ca. 1 × 1 cm²) coated with a thin film of PDMS (Sylgard 184, Dow Corning). Electrical contact to the Au surface was made using conducting copper tape (Electron Microscopy Sciences) to use the RNE as a working electrode (Fig. 1).

Fig. 1 Schematic illustration of the experimental apparatus used for measuring CVs on a TEPCM-based RNE.

Electrochemical measurements

Electrochemical data was obtained using a CH Instruments Model 618B electrochemical analyzer. Electrochemical experiments were carried out in a three-electrode cell containing a Ag/AgCl (3 M KCl) reference electrode and a Pt counter electrode (Fig. 1).²⁶ The area of the RNE in contact with solution was 0.34 cm². In addition, a 25-μm diameter gold microelectrode was used to measure the diffusion coefficients of Fe(CN)₆³⁻, TMAFc⁺ and FcDM in 0.1 M KNO₃. CV measurements were taken at scan rates ranging from 5 mV s⁻¹ to 1000 mV s⁻¹ at room temperature.

Results and discussion

Electrochemical characterization of TEPCM-based RNEs with 10-nm diameter pores

First, TEPCM-based RNEs were characterized by measuring (a) capacitive currents and (b) CVs of uncharged FcDM at different scan rates. The former provides a means for determining the total active area of a RNE that makes it possible to directly assess the seal between the gold layer and TEPCM.⁴⁰ The latter offers the scan rate that gives the transition between sigmoidal and peak-shaped CVs, permitting us to discuss the diffusion behavior of FcDM within the TEPCM nanopores.^26,35,41

(a) Capacitive current measurements. Capacitive current on a TEPCM-based RNE was obtained by measuring CVs in 0.1 M KNO₃ (data not shown). The current at 0.1 V vs. Ag/AgCl was used to calculate the active electrode area of a RNE using the double layer capacitance of gold surface that we measured on a gold disk electrode (5.7 × 10⁻⁵ F cm⁻²). If the seal between the TEPCM and gold film was perfect, the active electrode area of a RNE would be close to the total cross-sectional area of the open pores (5.1 × 10⁻⁴ cm²), which was calculated from the pore radius (7.1 nm) and pore density (9.5 pores μm⁻²) determined from SEM images (see ESI†). If the gold film is completely detached from the TEPCM, the active area would be close to the RNE area in contact to the solution (0.34 cm²). The experimental active area was (1.1 ± 0.7) × 10⁻² cm² (average ± standard deviation from 4 separate RNEs), suggesting that the solution crept into gaps between the TEPCM and the gold film formed due to their imperfect seal. The charging current of a RNE was unchanged for 5–7 days, indicating the strong adhesion of the sputter-coated gold film onto the TEPCM in contrast to a vapor-deposited gold film.²⁶

(b) CV measurements at different scan rates. Subsequently, CVs of 3.0 mM FcDM on TEPCM-based RNEs with 10-nm diameter pores were measured at different scan rates (v). As shown in Fig. 2a, the transition from sigmoidal to peak-shaped CVs was observed with increasing scan rate. The sigmoidal CVs reflected the radial diffusion of redox molecules from the outside of the pores, whereas the peak-shaped CV originated from their linear diffusion within the nanopores.^26,35,36 The observation of sigmoidal CVs at slower scan rates suggested that the spacing between pores was so large that the diffusion layers from individual pores did not overlap.⁴²

Fig. 2 (a) Cyclic voltammograms for a TEPCM-based RNE (10 nm in pore diameter) in a solution of 3.0 mM FcDM containing 0.01 M KNO₃ and 0.01 M KH₂PO₄–K₂HPO₄ buffer (pH 6.3) at five different scan rates (0.01, 0.05, 0.1, 0.4 and 1 V s⁻¹). (b) Relationship between the maximum oxidation current of 3.0 mM FcDM and the square root of scan rate. Measured in 0.01 M KNO₃ and 0.01 M KH₂PO₄–K₂HPO₄ buffer (pH 6.3) on the TEPCM-based RNE that gave CVs shown in Fig 2a.

Fig. 2b summarizes the relationship between the maximum oxidation current (i.e., a limiting current (i_lim) in a sigmoidal CV; a peak current (i_p) in a peak-shaped CV) and scan rate, i.e., v^1/2. The maximum current was similar at lower v, and then linearly increased for v greater than 0.05 V s⁻¹. Similar results were obtained on four separate TEPCM-based RNEs with 10-nm diameter pores. These observations are qualitatively consistent with the following theoretical equations regarding i_lim in a sigmoidal CV and i_p in a peak-shaped CV on a RNE:^35,43

where n is the number of electrons (n = 1 for the redox-active molecules used in this study), F is Faraday's constant (= 96485 C/mol), a is the pore diameter, R is the gas constant (= 8.31 J K⁻¹ mol), and T is temperature (= 298 K for 25 °C). N is the total number of nanopores participating (vide supra), and L is the length of the nanopore, which would be close to the thickness of the TEPCM measured using profilometry (6.4 μm). These equations were derived under the assumption that the apparent diffusion coefficients and concentrations of redox species within nanopores (D_p, C_p) are different from those in a solution (D_s, C_s). Eqn (1) indicates i_lim is similar regardless of v, and eqn (2) indicates i_p is proportional to v^1/2, as shown in Fig. 2b.

However, in contrast to previously reported results obtained on TEPCM-based RNEs containing 50-nm diameter pores,²⁶ the currents observed on the RNEs with 10-nm diameter pores were larger than those calculated using the theoretical equations. For example, the average and standard deviation of i_lim measured on four separate RNEs was 2.3 ± 0.7 μA, and was larger than 1.4 μA that was obtained from eqn (1) by assuming D_p = D_s = 6.1 × 10⁻⁶ cm² s⁻¹ for FcDM. In addition, i_p was much larger than the theoretical one calculated using eqn (2) at C_p = C_s = 3.0 mM and D_p = D_s: For example, i_p obtained at 0.4 V s⁻¹ from the four RNEs was 4.5 ± 0.4 μA, whereas the theoretical value was 0.8 μA. The large currents suggest the involvement of enhanced electron/mass transport (i.e., larger apparent D_p) within the nanopores such as electron hopping through a nanopore surface layer that contains the redox species at higher concentration.²⁰ The surface layer may originate from a PVP layer immobilized on the nanopore surface, considering its hydrophobic nature as compared with aqueous solution within the nanopore cavity. An enhanced faradic current was previously reported on AAO-based RNEs whose pore surface was covered with such an organic polymer layer.^22–25

On the other hand, the scan rate that gave the transition (v_t) was ca. 0.05 V s⁻¹ (Fig. 2b), which was much slower than the scan rate (0.8 V s⁻¹) estimated using eqn (3) under the assumption of L = 6.4 μm and D_p ∼ D_s.^26,27

Smaller D_p than D_s due to strong interactions with nanopores and/or larger L due to the tilted track angle may explain the slower v_t, but are not consistent with the larger i_lim and i_p (vide supra). Currently, we do not have a comprehensive model that can explain both the larger faradic current and smaller v_t.

CV measurements of uncharged and charged redox species on TEPCM-based RNEs at different pH and supporting electrolyte concentrations

Subsequently, CVs of three different redox species were measured at different pH (pH 4–9) and supporting electrolyte concentrations ([KNO₃] = 0.1 and 0.01 M in addition to 0.01 M KH₂PO₄–K₂HPO₄). Here, we discuss a series of sigmoidal CVs measured at 0.01 V s⁻¹ on an identical TEPCM-based RNE: Limiting currents in such CVs may be affected by the gaps suggested by the capacitive current measurements (vide supra) less significantly than peak currents in peak-shaped CVs, because they reflect the diffusion of redox species from the outside of the nanopores. Eqn (1) suggests that difference in i_lim can be attributed to changes in D_p. D_p reflects the partition of the redox species into the nanopores in addition to the mass/electron transport properties of the nanopores.^20,44 Thus, comparison of the sigmoidal CVs of charged and uncharged redox species at different pH will permit us to discuss interactions between the redox species and the 10-nm diameter nanopores.

Fig. 3 shows CVs of 1.0 mM Fe(CN)₆³⁻ measured on a RNE at 0.01 V s⁻¹ in solutions of three different pH conditions containing (a) 0.1 M and (b) 0.01 M KNO₃. In 0.1 M KNO₃, i_lim at pH 4.4 was slightly larger than those at pH 6.1 and 8.0 (Fig. 3a). In 0.01 M KNO₃, the decrease in i_lim at higher pH was larger (Fig. 3b) as compared with that in 0.1 M KNO₃. The shapes of the CVs were very similar, suggesting that the decrease in i_lim originated from changes in nanopore properties rather than changes in electrode surface properties. These observations suggest that electrostatic interactions between Fe(CN)₆³⁻ and the nanopores play an essential role in the changes of i_lim.^31,32,45

Fig. 3 Cyclic voltammograms (scan rate: 0.01 V s⁻¹) of 1.0 mM K₃Fe(CN)₆ solutions containing (a) 0.1 M KNO₃ + 0.01 M KH₂PO₄–K₂HPO₄ and (b) 0.01 M KNO₃ + 0.01 M KH₂PO₄–K₂HPO₄ at three different pH on a TEPCM-based RNE (10 nm in pore diameter). These data were obtained using a RNE different from that gave the data shown in Fig. 2.

In contrast to the decrease in i_lim of Fe(CN)₆³⁻, TMAFc⁺ exhibited an increase in i_lim with increasing pH, as shown in Fig. 4. The change in i_lim in 0.01 M KNO₃ (Fig. 4b) was larger than that in 0.1 M KNO₃ (Fig. 4a). The reversibility of the CVs was very similar regardless of the pH and supporting electrolyte concentrations. These results support the involvement of electrostatic interactions in the pH dependence of i_lim of the cationic TMAFc⁺.

Fig. 4 Cyclic voltammograms (scan rate: 0.01 V s⁻¹) of 3.0 mM TMAFc⁺ solutions containing (a) 0.1 M KNO₃ + 0.01 M KH₂PO₄–K₂HPO₄ and (b) 0.01 M KNO₃ + 0.01 M KH₂PO₄–K₂HPO₄ at three different pH on a TEPCM-based RNE (10 nm in pore diameter). The data shown in Figure 4 were obtained using a RNE that gave the data shown in Fig. 2.

CVs of uncharged FcDM were similar regardless of pH and supporting electrolyte concentrations examined (Fig. 5). These observations indicated that the diffusion of uncharged FcDM was not affected by electrostatic interactions in contrast to that of charged Fe(CN)₆³⁻ and TMAFc⁺ (Fig. 3 and 4).

Fig. 5 Cyclic voltammograms (scan rate: 0.01 V s⁻¹) of 3.0 mM FcDM solutions containing (a) 0.1 M KNO₃ + 0.01 M KH₂PO₄–K₂HPO₄ and (b) 0.01 M KNO₃ + 0.01 M KH₂PO₄–K₂HPO₄ at three different pH on a TEPCM-based RNE (10 nm in pore diameter). The data shown in Fig. 5 were obtained using a RNE that gave the data shown in Fig. 3.

Fig. 6 summarizes the pH dependence of the ratio of i_lim/C_s between charged redox species X (where X is Fe(CN)₆³⁻ or TMAFc⁺) and FcDM measured on an identical RNE. The ratio of i_lim/C_s can be directly correlated to the ratio of apparent diffusion coefficients of the two species (D_p,X/D_p,FcDM) as shown by the following equation derived from eqn (1) under the assumption that a, L and N are unchanged during the measurements and L(D_s/D_p) ≫ πa:

where i_lim,X and i_lim,FcDM are the limiting currents of charged redox species X and FcDM, and C_s,X and C_s,FcDM are the concentrations of X and FcDM in the sample solutions. D_s,X and D_s,FcDM are the diffusion coefficients of X and FcDM in solution, i.e., 6.1 × 10⁻⁶, 6.7 × 10⁻⁶ and 7.3 × 10⁻⁶ cm² s⁻¹ for FcDM, Fe(CN)₆³⁻ and TMAFc⁺, respectively.

Fig. 6 pH dependence of the ratio of i_lim/C between charged redox species (X = Fe(CN)₆³⁻ (triangles) or TMAFc⁺ (circles)) and FcDM. The i_lim values were measured at 0.01 V s⁻¹ on TEPCM-based RNEs (10 nm in pore diameter) in solutions containing 0.1 M (filled symbols) or 0.01 M (open symbols) KNO₃ in addition to 0.01 M KH₂PO₄–K₂HPO₄. The ratios of the diffusion coefficients in 0.1 M KNO₃ + 0.01 M KH₂PO₄–K₂HPO₄ (pH 6.3), which were measured using a gold microelectrode (25 μm in diameter), were shown in dashed lines.

Fig. 6 clarifies the pH-dependent changes in i_lim of the charged redox species in addition to the influence of KNO₃ concentration on i_lim, as described above. In addition, for the pH range examined, the ratios for Fe(CN)₆³⁻ were smaller than its D_s,X/D_s,FcDM, whereas those of TMAFc⁺ were larger than its D_s,X/D_s,FcDM. These results suggest that the 10-nm diameter nanopores were negatively charged over the pH range examined, and the charge increased at the higher pH. The increase in the negative charge at higher pH enhanced electrostatic attraction and repulsion with TMAFc⁺ and Fe(CN)₆³⁻, resulting in the increase and decrease in the i_lim/C_s ratio, respectively. The larger and smaller redox currents of Fe(CN)₆³⁻ and TMAFc⁺ at the lower KNO₃ concentration can be explained by the weaker screening of the negative charge of the nanopore surface that enhanced the electrostatic interactions.^16–18 The pH range that gave the change in i_lim/C_s ratios was pH 4–6 for both charged redox molecules, which is similar to the pH range that gave an increase in the negative streaming potential.¹⁶ The pH range suggests that the negative surface charge may originate from functional groups such as –COOH groups (pK_a ∼ 4.5) that were developed during membrane processing.¹⁶ The PVP layer on the nanopore surface would not directly involve in the negative surface charge because of the high pK_a value of its amide group.⁴⁶ Importantly, the deviation of D_p,X/D_p,FcDM from D_s,X/D_s,FcDM for Fe(CN)₆³⁻ was negligible on TEPCM-based RNEs with 50-nm diameter pores as suggested by the excellent correlation between experimental and theoretical currents.²⁶ probably due to the smaller contributions of the surface properties to the mass/electron transport through the 50-nm diameter nanopores.

Conclusions

This paper reported the CVs of uncharged and charged redox species on TEPCM-based RNEs with 10-nm diameter pores. The effects of pH and supporting electrolyte concentrations on the CVs suggested the presence of negative charge within the nanopores. The faradic currents measured in this study were larger than those estimated using theoretical equations, suggesting the enhanced mass and/or electron transport through the nanopores. These results may reflect the presence of a distinct surface layer on the 10-nm diameter nanopores to which the redox species partition preferentially. The CV data reported here offered information on the charge properties within the 10-nm diameter nanopores, which may make it possible to tune the properties of TEPCMs as templates for nanomaterials synthesis and separation membranes.

Acknowledgements

The authors thank Dr Daniel Higgins (Department of Chemistry, Kansas State University) for his suggestions, Dr Christopher Culbertson (Department of Chemistry, Kansas State University) for access to the surface profiler and Dr Bruce Cutler (Microscopy & Analytical Imaging Laboratory, University of Kansas) for his assistance with SEM measurements. The authors gratefully acknowledge the ACS-PRF and Kansas State University for financial support of this work.

References

1. J. Han, J. Fu and R. B. Schoch, Lab Chip, 2008, 8, 23–33.
2. W. M. Deen, AIChE J., 1987, 33, 1409–1425.
3. C. R. Martin, Science, 1994, 266, 1961–1966.
4. J. C. Hulteen and C. R. Martin, J. Mater. Chem., 1997, 7, 1075–1087.
5. R. F. Ismagilov, J. M. K. Ng, P. J. A. Kenis and G. M. Whitesides, Anal. Chem., 2001, 73, 5207–5213.
6. T.-C. Kuo, D. M. Cannon Jr., M. A. Shannon, P. W. Bohn and J. V. Sweedler, Sens. Actuators, A, 2003, 102, 223–233.
7. T.-C. Kuo, D. M. Cannon Jr., Y. Chen, J. J. Tulock, M. A. Shannon, J. V. Sweedler and P. W. Bohn, Anal. Chem., 2003, 75, 1861–1867.
8. D. M. CannonJr., T.-C. Kuo, P. W. Bohn and J. V. Sweedler, Anal. Chem., 2003, 75, 2224–2230.
9. K. Fa, J. J. Tulock, J. V. Sweedler and P. W. Bohn, J. Am. Chem. Soc., 2005, 127, 13928–13933.
10. E. N. Gatimu, J. V. Sweedler and P. W. Bohn, Analyst, 2006, 131, 705–709.
11. J. Dai, T. Ito, L. Sun and R. M. Crooks, J. Am. Chem. Soc., 2003, 125, 13026–13027.
12. Y. Zhang and A. T. Timperman, Analyst, 2003, 128, 537–542.
13. C. R. Martin, M. Nishizawa, K. Jirage, M. Kang and S. B. Lee, Adv. Mater., 2001, 13, 1351–1362.
14. L. A. Baker, P. Jin and C. R. Martin, Crit. Rev. Solid State Mat. Sci., 2005, 30, 183–205.
15. K. J. Kim, A. G. Fane, M. Nystrom and A. Pihlajamaki, J. Membr. Sci., 1997, 134, 199–208.
16. C. Lettmann, D. Mockel and E. Staude, J. Membr. Sci., 1999, 159, 243–251.
17. I. H. Huisman, P. Pradanos, J. I. Calvo and A. Hernandez, J. Membr. Sci., 2000, 178, 79–92.
18. P. J. Kemery, J. K. Steehler and P. W. Bohn, Langmuir, 1998, 14, 2884–2889.
19. T.-C. Kuo, L. A. Sloan, J. V. Sweedler and P. W. Bohn, Langmuir, 2001, 17, 6298–6303.
20. R. W. Murray, Ed., Molecular Design of Electrode Surfaces, John Wiley & Sons, New York, 1992.
21. A. J. Bard and L. R. Faulkner, Electrochemical Methods, Fundamentals and Applications, 2nd Ed., Wiley, New York, 2001.
22. C. J. Miller and M. Majda, J. Am. Chem. Soc., 1985, 107, 1419–1420.
23. C. J. Miller and M. Majda, J. Electroanal. Chem., 1986, 207, 49–72.
24. C. J. Miller and M. Majda, J. Am. Chem. Soc., 1986, 108, 3118–3120.
25. C. J. Miller, C. A. Widrig, D. H. Charych and M. Majda, J. Phys. Chem., 1988, 92, 1928–1936.
26. T. Ito, A. A. Audi and G. P. Dible, Anal. Chem., 2006, 78, 7048–7053.
27. C. J. Brumlik, C. R. Martin and K. Tokuda, Anal. Chem., 1992, 64, 1201–1203.
28. M. A. Hillmyer, Adv. Polym. Sci., 2005, 190, 137–181.
29. D. A. Olson, L. Chen and M. A. Hillmyer, Chem. Mater., 2008, 20, 869–890.
30. E. Jeoung, T. H. Galow, J. Schotter, M. Bal, A. Ursache, M. T. Tuominen, C. M. Stafford, T. P. Russell and V. M. Rotello, Langmuir, 2001, 17, 6396–6398.
31. Y. Li, H. C. Maire and T. Ito, Langmuir, 2007, 23, 12771–12776.
32. Y. Li and T. Ito, Langmuir, 2008, 24, 8959–8963.
33. Y. Li and T. Ito, Anal. Chem., 2009, 81, 851–855.
34. H. C. Maire, S. Ibrahim, Y. Li and T. Ito, Polymer, 2009, 50, 2273–2280.
35. K. Tokuda, K. Morita and Y. Shimizu, Anal. Chem., 1989, 61, 1763–1768.
36. C. S. Henry and I. Fritsch, Anal. Chem., 1999, 71, 550–556.
37. B. Kralj and R. A. W. Dryfe, Phys. Chem. Chem. Phys., 2001, 3, 3156–3164.
38. A. I. Bhatt and R. A. W. Dryfe, J. Electroanal. Chem., 2005, 584, 131–140.
39. C. G. Zoski, N. Yang, P. He, L. Berdondini and M. Koudelka-Hep, Anal. Chem., 2007, 79, 1474–1484.
40. R. M. Penner and C. R. Martin, Anal. Chem., 1987, 59, 2625–2630.
41. C. Amatore, J.-M. Saveant and D. Tessier, J. Electroanal. Chem., 1983, 147, 39–51.
42. C. Beriet, R. Ferrigno and H. H. Girault, J. Electroanal. Chem., 2000, 486, 56–64.
43. A. M. Bond, D. Luscombe, K. B. Oldham and C. G. Zoski, J. Electroanal. Chem., 1988, 249, 1–14.
44. J.-M. Saveant, J. Electroanal. Chem., 1991, 302, 91–101.
45. G. Wang, B. Zhang, J. R. Wayment, J. M. Harris and H. S. White, J. Am. Chem. Soc., 2006, 128, 7679–7686.
46. G. Lessene, M. Bordeau, C. Biran, D. de Montauzon and J. Gerval, J. Electroanal. Chem., 2000, 490, 79–84.

---

Footnote

† Electronic supplementary information (ESI) available: FE-SEM image and distribution of pore diameters. See DOI: 10.1039/b917517b

---