Spatially Directed Functionalization by Co-electropolymerization of Two 3,4-ethylenedioxythiophene Derivatives on Microelectrodes within an Array

All chemicals were reagent grade and used as received unless otherwise specified. The microelectrode arrays were fabricated by standard photolithographic methods on silicon wafers (125 mm diameter, 600–650 μm thick, with a 2 μm layer of thermally grown oxide on the surface) acquired from Silicon Quest International (Santa Clara, CA, USA). A chromium-plated tungsten rod, gold coin (Canadian Maple Leaf, 99.99%), and molybdenum boat (Kurt J. Leskar Company, Clairton, PA, USA) were used for thermal vapor deposition of metals using an Auto 306 (BOC Edwards, West Sussex, UK). Positive photoresist (AZ 4330) and developer (AZ 400K) were procured from EMD Performance Materials (Somerville, NJ, USA). Gold etchant (GE8148) was purchased from Transene (Danvers, MA, USA) and chromium etchant CEP-200 was obtained from Microchrome Technology (San Jose, CA, USA). Cyclotene 4024-40 (benzenecyclobutene, BCB), adhesion promoter (AP3000), and developer (DS2100) were used to insulate the chip-based electrode leads and were obtained from Dow Corning Company (Midland, MI, USA). Photoplot masks for metal and insulating layers were purchased from Fine Line Imaging (Colorado Springs, CO, USA). Electrical connection between the chip and potentiostat was made using an edge connector (solder contact, 20/40 position, 0.5 in pitch) from Sullins Electronics Corporation (San Marcos, CA, USA). The chip was kept in contact with the edge connector using a 0.99 mm thick folded plastic shim. Aqueous solutions were prepared using ACS reagent grade water, ASTM Type I, ASTM Type II (H₂O) from RICCA Chemical Company (Arlington, TX, USA). Succinic anhydride (SA), 4-(dimethylamino)pyridine (DMAP), triethylamine (Et₃N), sodium phosphate dibasic anhydrous, sodium phosphate monobasic anhydrous, and monomer 1 were purchased from Sigma-Aldrich (St. Louis, MO, USA). Sodium dodecyl sulfate (SDS), potassium ferrocyanide trihydrate (K₄Fe(CN)₆ · 3H₂O), and lithium perchlorate (LiClO₄) were obtained from Amresco (Solon, OH, USA), J.T. Baker (Phillipsburg, NJ, USA), and Alfa Aesar (Tewksbury, MA, USA), respectively. Concentrated hydrochloric acid (HCl), dichloromethane (DCM), and alumina (Neutral, Brockman Activity I) were purchased from Fisher Scientific (Pittsburg, PA, USA). Sodium chloride (NaCl), magnesium sulfate (MgSO₄), and acetonitrile (MeCN) were acquired from EMD Chemicals (Billerica, MA, USA). Deuterated chloroform (CDCl₃) was purchased from Cambridge Isotopes Labs (Tewksbury, MA, USA).

Figure 3 shows design schematics and a fabricated example of the electrode array chip. It is 2.54 cm × 2.54 cm in outer dimensions and has 20 individually-addressable electrodes, 16 of which are microband electrodes within an array. The nominal geometric area of each microelectrode is 2 × 10⁻³ cm² (2000 μm × 100 μm), and each one is separated from its neighboring electrode(s) by a 100 μm gap. The length of each microelectrode is defined by the edges of the insulating BCB layer. A detailed description of the microelectrode array fabrication can be found in section 1 of the Supporting Information (available online at stacks.iop.org/JES/167/166511/mmedia).

Zoom In Zoom Out Reset image size

SA was used to derivatize monomer 1 with the desired terminal carboxylic acid group to form monomer 2 based on a modified procedure previously reported. ⁵⁸ DCM was dried over alumina before use. Briefly, 452.7 mg of monomer 1 was dissolved in 25 mL DCM by stirring in a three-neck round-bottom reaction flask under N₂. A mixture of 562.7 mg SA, 28.6 mg DMAP, and 0.527 mL Et₃N was prepared in 10 mL DCM and stirred for 1 h at room temperature to dissolve, then added dropwise to the reaction flask. The reaction was stirred under N₂ overnight. The product was extracted with 10% HCl and rinsed with a saturated NaCl solution until the discarded rinse was the same pH as the saturated NaCl solution. The product was dried over MgSO₄ and remaining solvent evaporated under vacuum to yield monomer 2 (669.9 mg, 93.6% yield). To ensure the reaction was successful, characterization of the product, as well as the starting materials, by proton nuclear magnetic resonance spectrometry (¹H NMR), attenuated total reflectance Fourier transform infrared spectrometry (ATR-FTIR), and liquid chromatography coupled to electrospray ionization and mass spectrometry (LC-ESI-MS) was completed.

¹H NMR (400 MHz, CDCl₃): δ 6.37 (2 H, dd, J = 3.6, 3.6 Hz), 4.06 (1 H, dd, J = 11.7, 6.9 Hz), 4.24 (2 H, dd, J = 11.7, 1.9 Hz), 4.30–4.44 (3 H, m), 2.66–2.78 (4 H, m). ATR-FTIR (cm⁻¹): 1735 (νC=O_ester), 1699 (νC=O_{carboxylic acid}), 1483 (combination of stretching and bending modes of the C=C, C–C, and C–H bonds of thiophene ring), 1174 (νC-O-C_ring, νC=C_thiophene), 913 (νC-S-C_thiophene), 750 (δ C-H_thiophene). LC-ESI-MS (m/z): [M + H]⁺ calcd for C₁₁H₁₂O₆S, 273.043, found 273.1. The ¹H NMR, ATR-FTIR, and LC-ESI-MS results of the starting materials and products can be found in sections 2, 3, and 4 of the Supporting Information (Figs. S1–S7).

A three-electrode electrochemical cell, platinum flag or wire counter electrode, Ag/AgCl (saturated KCl) reference electrode, CHI 650 A with a picoamp booster and faraday cage (CH Instruments, Austin, TX, USA) were used for all electrochemical procedures. Before the chips were used for any electrochemistry (initial characterization and electropolymerization of films) they were rinsed with H₂O, dried with Ar, and cleaned of residual trace organics by oxygen plasma (6.8 W applied to RF coil for 15 min at 60 mtorr, PDC-32G, Harrick Plasma, Ithaca, NY, USA) and stored in H₂O until use.

A 0.2 M sodium phosphate buffer, pH 6.5, was prepared by first dissolving 14.24 g sodium phosphate dibasic anhydrous in a 0.5 L H₂O volumetric flask. Then 12.04 g sodium phosphate monobasic anhydrous was dissolved in 0.5 L H₂O in another volumetric flask. The pH of the sodium phosphate dibasic solution was continuously monitored while enough of the sodium phosphate monobasic solution was added to adjust the pH to 6.5. Electrodes were characterized before (designated as bare gold) and after electropolymerization by CV in two solutions: a 0.2 M sodium phosphate buffer, pH 6.5, and the buffer containing 0.001 M K₄Fe(CN)₆. Area-normalized capacitance, C, of each electrode was determined from the charging current, i_c , measured from the CV response both before and after film modification, using Eq. 1,

$$\begin{eqnarray}&&C\,=\,\displaystyle \frac{{i}_{c}}{vA}\end{eqnarray} \tag{ 1 }$$

where v is the scan rate of the cycle and A is the geometric area of the electrode (metal thickness was assumed to be a negligible contribution to A). The i_c of the CV response in the buffer alone was obtained by dividing the current difference between the forward and reverse sweeps at +0.400 V by two. The i_c for the buffer containing Fe(CN)₆ ⁴⁻ was measured directly from the forward sweep current at 0.000 V to avoid contributions from faradaic current in the return sweep.

To ensure that enough monomer 2 was available for the full set of depositions, a cell (Fig. 4) was designed such that only 0.7 mL was needed and placement of the reference and counter electrodes could be controlled for each deposition. The cell was machined from Delrin® and has two main parts: the bottom bulk of the cell, where mounts for the edge connector and the solution well are located, and the top of the cell, which holds the counter and reference electrodes in place. The top is fixed to the bottom by machine screws. For all electrochemical procedures other than polymer depositions, a small 15 mL glass cell was used.

Zoom In Zoom Out Reset image size

Films were electropolymerized from aqueous solutions containing 0.1 M LiClO₄, 0.05 M SDS and a total monomer concentration of 0.02 M. For electropolymerization by CV, the working electrode was cycled from −0.100 to +1.200 V at a scan rate of 0.1 V s⁻¹ for a total of three cycles, with a 2 s quiet time at −0.100 V between each cycle. Each polymer was generated in triplicate by modifying three microband electrodes within the same array on a given chip; each chip was dedicated to a specific mol% 2 of the two monomers. In cases of inadequate polymer adhesion, the surfaces of the working electrodes were refreshed by swabbing the array first with 1 M sodium hydroxide, followed by rinses of ethanol and H₂O, and finally dried with Ar. Once the depositions were concluded for a given chip, it was carefully removed from the cell and placed in another cell containing the sodium phosphate buffer and allowed to soak for 5 min. Five sequential CV cycles were then performed from −0.100 to +0.700 V at a scan rate of 0.050 V s⁻¹ to equilibrate the film with the characterization electrolyte and reach a reproducible response, one modified electrode at a time. ⁵⁹ The chips were subsequently stored under H₂O overnight before further characterization. The capacitance of each film was measured and characterized with an electroactive model compound in the same manner as described for electrodes before modification.

Films formed from deposition solutions of high mol% 2 (100, 75, 66) exhibited inconsistent polymer adhesion, which was especially regular for the 100 mol% 2 case. Delamination of the films was sometimes observed during the emersion of the chip from the deposition solution and immersion of the chip into characterization solutions. We addressed this complication to achieve the most reliable understanding of these films by repeating electropolymerization as needed, and in the case of 100 and 66 mol% 2, fresh films were produced for the surface analyses (μATR-FTIR, XPS, and profilometry), separate from films used for electrochemical characterizations. For the case of 100 mol% 2, fresh films were also used for both electrochemical characterizations: in buffer solution with ferrocyanide and buffer alone.

Because there are no resolved oxidation peaks during the electrodeposition cycles, a reproducible method was developed to determine E_onset when monomer oxidation becomes visibly significant over the charging current during the electropolymerization cycles. The average background current, between 0.000 and +0.500 V, was determined and subtracted for each electrodeposition cycle. Then the first derivative was approximated using the software accompanying the CHI 650 A. The local minimum of the first derivative plot corresponding to the inflection point was used to extrapolate a tangent line, with proper slope and position, to locate the potential value at zero current of the deposition voltammogram. The process was only performed for the anodic sweep of each deposition cycle and average values with standard deviation are listed in Table I. Further detail can be found in section 5 of the Supporting Information and a schematic of this process is shown in Fig. S-8.

The total charge passed (Q_T ) and total monomer deposited (m) were determined from the electrodeposition voltammograms. Each deposition cycle was separated into forward and reverse sweeps and the average background current, for each sweep, was determined and subtracted as described earlier. The area of the curve to background-subtracted zero current was integrated for both segments of each cycle and added. Because the anodic current is defined as negative in this work, the absolute value was the amount of charge passed for that cycle. The process was repeated for the subsequent cycles and the sum of the charge passed for each of the cycles is Q_T , and given by Eq. 2,

$$\begin{eqnarray}&&{Q}_{T}\,=\,{Q}_{1}+{Q}_{2}+{Q}_{3}\end{eqnarray} \tag{ 2 }$$

where Q₁ , Q₂ , and Q₃ are the average charge passed for the first, second, and third deposition cycles, respectively, across the modified electrodes of a given mol% 2. The values of Q_T , with standard deviation, for each of the seven different films, are listed in Table I. The total number of moles that were electrodeposited, m, was calculated directly from Q_T , using Eq. 3,

$$\begin{eqnarray}&&m\,=\,\displaystyle \frac{{Q}_{T}}{nF}\end{eqnarray} \tag{ 3 }$$

where n is the number of electrons per monomer deposited, assumed to be 2.3, ⁴⁰ and F is the Faraday constant. These values are reported in Table I.

Profilometry of each film was carried out on a Dektak 3030 surface profiling measuring system (Sloan Technology Corporation, Santa Barbra, CA, USA). Each array was scanned with a medium speed and stylus force setting of 20 mg, which correlates to a force of 0.2 mN. To ensure a representative average of the films' profile, each electrode was scanned across its width three times: once near each of the two ends of the microband and once in the middle, about halfway between the two ends. The printout of each measurement was scanned as digital images and digitized with GetData Graph Digitizer software (www.getdata-graph-digitizer.com) to generate tabulated data for analysis. A line was drawn from the base of one side to the base on the other side of each electrode profile, above which the cross-sectional area was determined. To obtain the average cross-sectional area of a given type of film, the average cross-sectional area of the unmodified electrodes on the same chip was subtracted from the average cross-sectional area of the modified electrodes. For the chip with the 0 mol% 2 films, the measurement was completed using six scans, instead of three, because of larger distances between modified electrodes. Also, difficulty obtaining a suitable background resulted in only two 0 mol% 2 modified electrodes were used in the final calculation of the average cross-sectional area. A schematic showing the order and direction of measurements is shown in Fig. S-9 of section 6 of the Supporting Information.

The composition of the modified microelectrodes was analyzed by μATR-FTIR sampling using a Vertex 70 v FTIR spectrometer equipped with a HYPERION 2000 infrared microscope (Bruker Optics, Billerica, MA, USA). The spectrometer bench was maintained under vacuum, and radiation from the interferometer was passed to the microscope through a KBr window. The microscope was kept under a dry N₂ gas purge and secured to the spectrometer bench through a purge tube. All samples were allowed to dry under ambient conditions before analysis. Electrodeposited polymer on the 100 μm wide microelectrodes was analyzed using the diamond ATR probe (32 μm diameter) on the microscope objective (20×, 0.6 NA). The probe sampled within the center region of each electrode. The knife-edge apertures on the microscope were set to isolate the electrode and ensure against sampling regions on the chip beyond the electrode of interest. The probe-electrode force was adjusted as recommended for polymers (∼0.5 N). The microscope was operated with a mid-range liquid nitrogen cooled mercury-cadmium-telluride (MCT) detector. Immediately prior to measurements on the modified electrodes, the ATR crystal was cleaned by gently passing ethanol-soaked lens tissue over the probe surface, and a background spectrum was collected with the probe suspended in air. All interferograms were recorded at 4 cm⁻¹ resolution with 64 scans averaged and processed using the Blackman-Harris three-term apodization function before Fourier transformation. Spectra are displayed in units of pATR,

$$\begin{eqnarray}&&pATR\,=\,-\,\mathrm{log}\,\displaystyle \frac{{I}_{sample}}{{I}_{background}}\end{eqnarray} \tag{ 4 }$$

where I_sample and I_background represent the sample and background single beam spectra, respectively.

XPS measurements were carried out on a Kratos Axis Ultra DLD spectrometer (Kratos Analytical Ltd., Manchester, UK) using a monochromated aluminum source (Al Kα 1486.6 eV). An X-ray gun with an emission current of 0.01 A, anode voltage of 15 kV, and a bias voltage of 125 V was used. The aperture size and iris position were both 110 μm centered on the 100 μm wide modified microelectrodes. High-resolution scans of the C(1s) core electrons were obtained using a pass energy of 40 eV and step size of 0.1 eV. Copper tape was used to short the contact pads of the chip and ground it to the sample stage. A charge neutralizer gun was used to compensate for any charge accumulation during the measurement. All XPS data were analyzed, deconvoluted, and fitted using CasaXPS software (Casa Software, Ltd., Teignmouth, UK). A linear background and line shapes of a Gaussian-Lorentzian sum form were used. Constraints were forced on the peak area, full-width at half maximum (FWHM), and binding energy position of the synthetic components. A more detailed explanation of the deconvolution and fitting process can be found in section 7 of the Supporting Information.

Electrodeposition was performed from aqueous solutions containing 0.05 M SDS, 0.1 M LiClO₄, and monomer 1 and (or) 2 at a total concentration of 0.02 M. The mol% 2 represents the molar ratio of monomer 2 to all monomers in the deposition solution, given by Eq. 5.

$$\begin{eqnarray}&&mol \% \,{\bf{2}}\,=\,\left[\frac{mol{\bf{2}}}{mol{\bf{1}}+mol{\bf{2}}}\right]\,\times \,100 \% \end{eqnarray} \tag{ 5 }$$

Figures 5a through 5g present the electropolymerization voltammograms obtained at 0.1 V s⁻¹ for each mol% 2. Each of the three consecutive cycles (1st, 2nd, 3rd) in the figures is the average current response for that cycle number from three electrodes. Error bars that represent one standard deviation are reported at ten selected potentials in both the forward and reverse sweeps. In all but one case (100 mol% 2), the average voltammograms were from a single set of electrodes (N = 3). The 100 mol% 2 results are from two sets of electrodes (N = 6). Figure 3c shows a micrograph of an array with polymer films on three electrodes, after the three cycles of deposition from the 50 mol% 2 solution. The films change color from dark blue or purple to colorless, as they transition from reduced to oxidized states. Additional color is caused by the diffraction of white light that varies with film thickness. All deposition voltammograms are shown in Fig. S-10, and micrographs for all films are shown in Fig. S-11 in sections 8 and 9 of the Supporting Information, respectively. The general shapes of the CV responses for electrodeposition from pure and comonomer solutions are similar to each other and consistent with those reported previously for EDOT and other derivatives. ^{48,50,58,60,61} They exhibit a characteristic, relatively constant, charging current between −0.100 V and 0.850 V and a significant increase in anodic current at potentials more positive than 0.850 V where polymer formation occurs.

Zoom In Zoom Out Reset image size

A more detailed analysis of each CV cycle is given here to investigate possible differences induced by the two monomers and their mixtures. During the first, forward anodic sweep (red) for all monomer solution compositions, the current remains relatively constant due to double-layer charging at the bare gold surface until the potential nears +0.9 V. Beyond +0.9 V, the anodic current increases dramatically due to the oxidation of the monomers in solution that subsequently form polymer films on the electrode. Monomer oxidation and polymer deposition continue during the reverse sweep past +0.9 V because of the lower oxidation potentials of longer oligomers that were not present in solution during the initial anodic sweep. This hysteresis in the first electrodeposition voltammogram is characteristic in solutions of all monomer compositions studied herein. Figure S-12 in section 10 of the Supporting Information can be used to explain this behavior for monomers 1 and 2. It shows an electropolymerization mechanism modeled from the "oligomer approach" described in the literature for thiophene and pyrrole. ⁶² The mechanism highlights the difference between initial monomer oxidation and deposition of larger chain, insoluble oligomers, on the electrode surface. The first step is the oxidation of two monomers at the electrode surface resulting in monomeric radical cations. These remain in solution and react with each other forming a dimer in the diffusion layer. The first dimerization between two monomers, during the first anodic sweep, occurs at a larger E_onset than subsequent dimerization steps throughout the mechanism. The following steps include successive dimerization reactions at decreasing overpotentials leading to the formation of an octamer. The decreasing solubility of increased oligomer length leads to the deposition of the polymers on the electrode surface through instantaneous nucleation and 3D polymer growth.

Although the shapes of the deposition voltammograms in Fig. 5 are similar, a comparison of the anodic sweeps for each mol% 2 reveal different E_onset values of monomer oxidation and are listed in Table I. For all three deposition cycles, the 100 mol% 2 E_onset values are greater than those for the 0 mol% 2. The values for the first cycle differ by 0.08 ± 0.02 V. The lowest E_onset value for the 100 mol% 2 (Cycle 3, 0.969 ± 0.004 V) is greater than the highest E_onset value for the 0 mol% 2 (Cycle 1, 0.946 ± 0.003 V). These results suggest that monomer 1 is easier to oxidize than monomer 2 at the electrode. Because the anodic sweep of the first cycle is primarily the oxidation of the monomers to their radical cations at a bare gold electrode, the difference in oxidation potentials is most likely due to steric effects. The side chain of monomer 2 introduces a conformational barrier to the electrode surface for oxidation. A similar effect was observed between monomers differing in structure by even just a single hydroxymethyl group, as is the case for EDOT and monomer 1 reported in the literature, albeit at a much smaller difference. ⁵⁵ The E_onset values for the first deposition cycle of the different mixed monomer solutions are different from each other. Still, they do not follow a consistent trend with the relative monomer composition in solution. However, the E_onset value does decrease consistently for a given solution composition from cycle 1 to cycle 3.

Evidence for growth of the polymer, layer by layer, from one cycle to the next can be seen in the increasing charging current with each cycle for all monomer solutions. The anodic current due to oxidative electropolymerization of the monomer also increases by a similar amount due to this. This behavior has been observed previously with EDOT and EDOT derivatives. ^48,58,60 However, for a larger mol% 2 in the deposition solutions, the magnitude of both the maximum anodic and background currents are smaller, suggesting less electropolymerization occurs when monomer 2 is present.

To quantify this observation, integration of the background-subtracted current with respect to time during the oxidative deposition portion of the voltammograms was performed. This yields a charge, Q, that is proportional to the amount of monomer deposited. Figure 6a shows how the dependence of the total charge, Q_T , added over the three cycles, and averaged from multiple electrodes, varies with the composition of the solution. Figures 6b–6c illustrate how and where the integration of each voltammogram was performed. As seen in Fig. 6a Q_T decreases with increasing mol% 2 in the deposition solution. These values can be directly converted to the amount of monomer deposited, m, assuming 2.3 moles of electrons per mole of monomer deposited ⁴⁰ using Eq. 2. Each deposition solution resulted in films with unique values of m. Although the trend exhibits a relatively weak fit to a linear model (R² = 0.9889), there is a distinctive dependence of m on the mol% 2. This dependence is most likely a consequence of the decreasing pH of the deposition solution caused by the increasing concentration of monomer 2. The pH of the deposition solution affects proton elimination, which is the overall rate limiting step of the electropolymerization mechanism, as the nucleation and growth happen simultaneously (Fig. S-12). Therefore, the rate of proton elimination and thus the amount of oligomer formation from solutions comprised mostly of the acid, monomer 2, are expected to be lower than in solutions comprised mostly of monomer 1.

Zoom In Zoom Out Reset image size

Table I shows the measured pH values of each deposition solution, along with a monomer-less solution containing only 0.1 M LiClO₄ and 0.05 M SDS. The pH of the solutions decreases from 5.42 to 2.96 as the mol% 2 increases from 0 to 100 mol% 2. Although a pK_a value for monomer 2 could not be found in the literature, it is likely to be similar to the pK_a value of succinic acid, 4.16. ⁶³ This is because succinic acid is nearly identical to the side chain of monomer 2. The pH of all deposition solutions containing monomer 2 are below this estimated pK_a value. Applying the Henderson-Hasselbach equation and the known total concentrations of monomer 2 estimates the amount of protonated carboxylic acid moieties to range from 94% for the 100 mol% 2 deposition solution (pH 2.96) to 88% for the 25 mol% 2 deposition solution (pH 3.29). The higher pH value of the 0 mol% 2 solution (5.42) compared to that of the monomer-less solution (5.31) suggests that the presence of monomer 1 also affects the pH. The presence of SDS does not affect the pH of the solutions. Although the pH of the deposition solution can predict the extent of the electropolymerization, and thus m, it does not provide information regarding the relative amounts of monomer incorporated into the film.

In addition to pH, there is another possible explanation for the trend observed for mixed monomer deposition solutions in Fig. 6a. Previously reported co-electropolymerizations of EDOT and monomer 1 in aqueous solutions, for example, show a preference for the deposition of EDOT, over monomer 1, up to equimolar ratios of the monomers in the deposition solution. ⁵⁵ Thus, one could argue that in our case, the decreasing m with increasing mol% 2 suggests that monomer 1 is the one that is predominately deposited. Note, however, that our solution conditions are substantially different (an aqueous solution of 0.1 M LiClO₄ and use of a solubilizer) than in the other report (methanol, diluted with aqueous 0.1 M LiClO₄). Consequently, the ratio of monomers in our deposited films could also be different. Therefore, an analysis of the ratio of monomers incorporated into the films was performed using μATR-FTIR and XPS, and is described in a subsequent section below.

The packing of oligomers on the electrode surface could affect the electrochemical behavior of the films, including access to counter ions migrating in and out of the polymer network, as well as the path that the electrons travel from one oligomer to the next. Given the proposed mechanism of Fig. S-12 of the Supporting Information, we do not expect the side chain of monomer 2 to play a major role in the polymerization mechanism. Like EDOT, polymerization of the monomers 1 and 2 is linear through α-α' coupling of the thiophene rings, and the ethylenedioxy ring blocks coupling at the 3 and 4 positions. However, due to steric effects, monomers with a bulkier side chain, as is the case for monomer 2, should affect the packing of the deposited oligomers and produce a less dense polymer film.

To investigate the extent of packing, profilometry was used to measure the average cross-sectional areas of the polymer films. Representative height profiles from which the cross-sectional areas were obtained are shown in Fig. S-9 in section 6 of the Supporting Information. The profiles give information about how the polymers were deposited on the electrodes. As expected, there is a build-up of polymer along the edges of the electrodes, for all films, due to enhanced mass transport there from radial diffusion during the electrodeposition. Figure S-13 in section 11 of the Supporting Information shows that the average cross-sectional areas for the polymer films from the mixed monomer solutions do not exhibit a consistent trend. The two extreme cases of the pristine polymers, 0 and 100 mol% 2, yielded areas of 60 ± 19 μm² and 30 ± 13 μm², respectively. Assuming a uniform distribution of polymer along the entire 2000 μm length of the microbands, the molar concentrations of monomer units within the space occupied by the polymer are 12 ± 4 mol L⁻¹ and 9 ± 4 mol L⁻¹, respectively. Figure S-14 in section 11 of the Supporting Information compares these molar concentrations for all of the films as a function of mol% 2. There does not appear to be a definitve trend in this quantity, because of the substantial amount of error (as much as ∼36% orginating from the cross-sectional area measurements). Although there is some inherent error in the use of the profilometry instrument (±0.05 μm) and in digitizing the results, a greater error is likely due to inaccurate profiling of the polymer edges given the speed and size of the profilometer probe. Consistent with this hypothesis is that the average cross-sectional area of the bare gold electrodes (29 ± 5 μm²), which are flat, is within the error of the expected value of 26 um² based on the amount of gold and chromium deposited during thermal evaporation.

Another way to evaluate the profilometry data is to avoid the edges and simply compare the thicknesses for the different films based on the height in center portion of the profiles. These results are also reported in Fig. S-13. The heights also vary, where films deposited from solutions of 0 and 100 mol% 2 measured 0.27 ± 0.05 and 0.15 ± 0.13 μm, respectively. There is no discernable trend among the films formed from the mixed monomer solutions.

Determining the relative amounts of the two monomers incorporated into the polymer films from the mixed solutions addresses the questions regarding factors affecting the deposition and provides guidance about conditions for preparing films with desired properties. Quantitative analysis was performed using μATR-FTIR sampling, which can measure spectra from within the 100-μm wide microbands without overlapping with the gaps or other adjacent structures on the chip.

Figure 7a shows μATR-FTIR spectra plotted over the 1900–750 cm⁻¹ range for polymer films electrodeposited from solutions of 100, 66, 50, 25, and 0 mol% 2. These films had been electrochemically characterized after film formation, thoroughly rinsed and soaked in water and allowed to dry in the air under ambient conditions. Peaks for the films are much broader and some are slightly shifted compared to those for the individual monomers (Figs. S-3a and S-3b in the Supporting Information). The polymer spectra are also similar to those previously reported for poly(2) and related polymers and monomers. Table III summarizes the proposed assignments of vibrational modes of the main peaks highlighted in Fig. 7a based on spectra reported in the literature. Absorbance bands at 836 and 976 cm⁻¹ have been assigned to the stretching of the C–S bond of the thiophene ring. ^49,64–69 Stretching vibrations within the ethyelenedioxy ring absorb at 1062, 1104, 1220 cm⁻¹. ^64,66–74 Several IR peaks can be traced to the stretching of the C=C and C–C bonds within the thiophene ring at 1169, 1370, 1477, and 1525 cm⁻¹. ^{49,67,68,70–73,75} The bending vibration of H₂O molecules retained within the film and the stretching of the C=O bond within the ester of monomer 2 absorb at 1642 and 1739 cm⁻¹, respectively. ^50,76

Zoom In Zoom Out Reset image size

Carbonyl group stretching modes within esters exhibit some of the more distinct absorbances in IR spectroscopy, and in the present studies, provide a unique marker for monomer 2. Figure 7b shows an overlay of spectra in the expanded region of 1800–1580 cm⁻¹ to enable comparison of changes in the ester carbonyl band for polymers formed from varying compositions of deposition solutions and in the bending vibration for the retained H₂O. Although two carbonyl groups are present on the side chain of monomer 2, each with distinct vibrational bands in IR spectroscopy, only the C=O stretching vibration of the ester group is evident in spectra of the electropolymerized films. Because the carboxylic acid groups neutralize during the electrochemical characterization of the films in the sodium phosphate buffer, the characteristic band for the carboxylic acid stretch is replaced by lower wavenumber carboxylate group vibrations near 1400 cm⁻¹. Figure 7b shows the expected increase in pATR₁₇₃₉ for films formed from deposition solutions with increasing mol% 2. Interestingly, pATR₁₆₄₉ displays the opposite trend, decreasing with increasing deposition mol% 2. The response indicates the electropolymerized films become more resistant to wetting with increasing monomer 2 content, possibly a result of the side-chain hydrophobicity.

A plot of the pATR₁₇₃₉ vs the mol% 2 of the deposition solution is shown in Fig. 7c. The background pATR₁₇₃₉ derived from the 0 mol% 2 sample has been removed by subtraction. The near linear calibration relationship (R² = 0.9898) demonstrates the content of monomer 2 in the electropolymerized films can be closely estimated from the deposition solution composition. These data also support the notion that the decreased quantity of electropolymerized monomer that occurs with increasing mol% 2 in solution is due to a change in pH and not the idea that the deposition is dominated by monomer 1. The fact that the ratio of the two monomers in the polymers closely tracks their ratio in the deposition solutions, regardless of the total quantity of polymer electrodeposited, suggests that our conditions involve a factor that normalizes the monomer deposition behavior. This factor is SDS, and its effect is described in more detail in the next section.

Further confirmation of the chemical composition of the films was obtained by XPS. Because the spot size (110 μm) was slightly larger than the microelectrode width (100 μm), some of the survey spectra (not shown) measured peaks for silicon from the SiO₂-coated silicon substrate. Thus, the analysis of the XPS data here focuses on the C(1s) peaks and does not include the O(1s) peaks. Figure 8 shows C(1s) narrow scans of the 0, 25, 34, 50, 66, and 100 mol% 2 films. Deconvolution and fitting of the spectra were performed to confirm contributions of the different carbons and, therefore, the relative amounts of the two monomers toward the spectra. A careful balance between mathematical overfitting and reasonable assessment of the spectra considering the expected chemistry was the overall goal. Because the samples were prepared and stored under ambient conditions, there is adventitious carbon on the surfaces. Adventitious carbon is often used as a calibration reference for binding energy values (284.6–285 eV). For our measurements, however, we are interested in the different carbonaceous species within the films. Thus, we focused on the intensity of the XPS signals of carbonyl carbons at higher binding energies than the adventitious carbon, as this is an indication of the relative amount of monomer 2 present in the film.

Zoom In Zoom Out Reset image size

Figure 8a is the deconvolution and fit of the 0 mol% 2 film to four synthetic components, one for each type of carbon and one for the contribution from π-π* shake-up within the thiophene; for Figs. 8b–8f there are eight total synthetic components. See the color-coded carbons in the structure shown in Fig. 8g as a guide. The first and second components (cyan and black, respectively) are from the carbons that are in-between the ester and carboxylic acid and present only in monomer 2 (284.1 ± 0.4 eV) and the carbons adjacent to the sulfur in the thiophene ring that are present in both monomers (284.62 ± 0.11 eV). The third component (red) is attributed to the other two carbons in the thiophene ring at 285.44 ± 0.08 eV. One carbon in the ethylenedioxy bridge and the methylene carbon have been designated as equivalent and contribute to the fourth component (orange) at 286.3 ± 0.3 eV. The other carbon in the ethylenedioxy bridge is accounted for in the fifth component (blue) at 287.4 ± 0.3 eV. The sixth and seventh components are only present in films containing monomer 2 and are due to the carbons of the ester (green) and carboxylic acid (purple). They give rise to peaks at 288.6 ± 0.3 and 289.0 ± 0.4 eV, respectively. Despite the synthetic components and their specific position, there is an increase in signal corresponding to carbons with a higher binding energy as the films were produced from deposition solutions with increasing mol% 2. Therefore, the XPS C(1s) spectra support the μATR-FTIR data and show an increase in the ester content of the films as the mol% 2 in the deposition solution increases. Because XPS is a surface analysis technique (depth of ∼10 nm), these results also suggest there is not a major difference between the surface and the bulk composition of the films.

Models have been developed to describe how SDS enhances the solubility of other thiophenes and pyrrole in aqueous solutions and has a catalytic effect on their electropolymerization. ^32,77,78 These can be used to explain how copolymer film compositions closely reflect those of solution mixtures of monomers 1 and 2. It has been proposed that the hydrophobic tails of SDS molecules associate with pyrrole monomers to solubilize them in aqueous solutions and that a sufficient amount of SDS is needed to efficiently supply monomer to the electrode surface, which is achieved when the SDS is in micellar form. ⁷⁷ This is the case for our experiments, where the concentration of SDS (0.05 M) is above the critical micelle concentration (CMC), 0.0082 M at 25 °C. ⁷⁹ Although monomer 1 is water-soluble, it is likely to interact with SDS in a similar way as monomer 2 due to the hydrophobic thiophene region. It is also possible that more than one monomer is associated with each micelle. ⁷⁷ The resulting, multi-molecular assemblies could then experience mass-transport to the electrode surface that does not depend on the individual monomer shape, size and hydration sphere. SDS micelles also form a bilayer at the electrode, which is stabilized with increased SDS concentration. The bilayer could further facilitate presentation of the micelle/monomer assemblies to the surface and host monomers within it. ⁷⁷

It has been shown that the oxidation potential of EDOT for electropolymerization decreases with increasing concentration of SDS until an effective CMC for the electrochemical system is reached, at which point the oxidation potential plateaus. ⁷⁸ In a 0.1 M LiClO₄ aqueous solution, that concentration is ∼0.005 M. This behavior is ascribed to the formation of a 1:1 pseudocomplex between EDOT^+• and the anionic head group DS⁻ that has a relatively large equilibrium constant of formation, 5.4 × 10³ M⁻¹, and which attains a steady concentration once the CMC is achieved and all the EDOT is incorporated in the micellar phase. In our studies with aqueous 0.1 M LiClO₄ and 0.05 M SDS, well above the effective CMC, E_onset for monomer 1 is indeed more negative (by 73 mV) than that reported by Doherty et al. ⁵⁵ for this monomer in aqueous (10% v/v methanol) 0.1 M LiClO₄ solutions without SDS.

The role of SDS can be revealed by comparing the behavior of E_onset for increasing amounts of one monomer in the presence of another. We found no particular trend in E_onset for deposition mixtures containing all combinations of monomers 1 and 2, even though the difference in E_onset between 0 mol% 2 and 100 mol% 2 is 85 mV. The E_onset for mixtures remained within 37 mV of that of monomer 1 alone (0 mol% 2). However, Doherty et al. report a more consistent trend in E_onset for mixtures without SDS and with a narrower 30 mV difference of E_onset for monomer 1 alone to E_onset for EDOT alone.

The effect of SDS micelles on normalizing the electrodeposition of 1 and 2 is especially dramatic when evaluating Fig. 7(c) in light of the copolymer composition curves of Doherty et al. ⁵⁵ For a better comparison, the figure was rescaled from 0 to 1 on the ordinate and abscissa and is shown in Fig. S-15 in section 12 of the Supporting Information. At a depletion of only 0.007% of the monomers during electropolymerization from an equimolar solution, the mole fraction of monomer 2 (higher E_onset ) in the copolymer film is 0.41. In contrast, a depletion of 50% of the monomers from an equimolar ratio of 1 and EDOT in an aqueous deposition solution without SDS was required to achieve a similar mole fraction in the film for the monomer with the higher E_onset,. ⁵⁵

Another way to quantify the reactivity of the two monomers in this SDS solution is to determine the reactivity ratios, r, for a two-monomer system, ^55,80

$$\begin{eqnarray}&&{r}_{1}\,=\,\displaystyle \frac{{k}_{11}}{{k}_{12}}\,{\rm{and}}\,{r}_{2}\,=\,\displaystyle \frac{{k}_{22}}{{k}_{21}}\end{eqnarray} \tag{ 6 }$$

where k₁₁ and k₂₂ are the homogeneous rate constants of the reaction of the cation radical of a monomer with another monomer of the same kind for monomer 1 and monomer 2, respectively. The k₁₂ and k₂₁ are the rate constants for the cation radical of monomer 1 reacting with monomer 2 and for cation radical of monomer 2 reacting with monomer 1, respectively. The relative mole ratio of the monomers in the copolymer, m₁ /m₂ , can be predicted with Eq. 7 and using these reactivity ratios and the initial concentrations of the two monomers, [M₁] and [M₂], under the assumption of steady-state conditions (low depletion of monomers from solution, which is the case here).

$$\begin{eqnarray}&&\displaystyle \frac{d{M}_{1}}{d{M}_{2}}\,=\,\displaystyle \frac{\left[{M}_{1}\right]}{\left[{M}_{2}\right]}\displaystyle \frac{{r}_{1}\left[{M}_{1}\right]+\left[{M}_{2}\right]}{{r}_{2}\left[{M}_{2}\right]+\left[{M}_{1}\right]}\approx \displaystyle \frac{{m}_{1}}{{m}_{2}}\end{eqnarray} \tag{ 7 }$$

From m₁ /m₂ ratios obtained from pATR of the copolymer films for the different combination of points at 25, 50 and 66 mol% 2, r₁ = 1.4 ± 0.2 and r₂ = 0.5 ± 0.1. This suggests that the propagating cation radical of monomer 1 reacts preferentially with itself (r > 1) and the propagating cation radical of monomer 2 also reacts preferentially with monomer 1 (r < 1). It is consistent with a slightly higher dominance of monomer 1 in the film. For comparison, Doherty et al. report reactivity ratios that deviate even more from unity for a two-monomer system without SDS at low depletion (monomer 1 (r = 0.15) and EDOT (r = 1.65)), but closer to unity for high depletion (monomer 1 (r = 0.71) and EDOT (r = 1.58)). This further confirms the role of SDS micelles at low depletion to produce a copolymer film that is more representative of monomer concentrations in solution. Interestingly, the product, r₁ r₂ , can provide information about the relative arrangement of monomer units along the copolymer backbone. When r₁ r₂ = 0, an alternating sequence of monomer units is predicted; when r₁ r₂ = 1, a random arrangement is expected, and when r₁ r₂ > 1, a block copolymer is formed. ^55,80 For the conditions reported here, r₁ r₂ = 0.7 ± 0.3, so that the arrangement of monomer units is more random than it is alternating.

Figure S-16 in section 13 of the Supporting Information shows CV responses of polymer films obtained in two different solutions: 0.2 M sodium phosphate buffer and 0.2 M of the buffer containing 0.001 M K₄Fe(CN)₆. The shapes of the voltammograms in buffer alone are similar to those observed for PEDOT in aqueous supporting electrolytes, where the current is relatively flat across the entire voltage range of −0.100 to +0.700 V. ⁶⁰ This behavior is similar to the charging of a capacitor. Such behavior is also observed for CV in buffer containing ferrocyanide during the forward sweep at potentials negative of the E° value and before the faradaic current becomes pronounced. The capacitances (C) obtained from CV in both of these characterization solutions are shown in Fig. 9 and listed in Table II for the polymer films as a function of the mol% 2 of their deposition solutions. The C values obtained from the redox solution are within error of or slightly larger than those obtained in solutions of buffer alone. The small increase in C could be due to the increased ionic strength from the additional K₄Fe(CN)₆.

Zoom In Zoom Out Reset image size

Table II. Electrochemical and infrared spectroscopy characterization of polymer films.

Mol% 2	Electrochemical Characterization						μATR-FTIR
	C (mF cm−2)		Faradaic Response				pATR
	Sodium Phosphate a)	Ferrocyanide a)	ΔEp (V) a)	ipa (μA) a)	ipc (μA) a)	ipc /ipa a)	$\bar{v}$ = 1739 cm−1 (au)
Bare Au	0.0650 ± 0.0004	d)	0.14 ± 0.03	−0.44 ± 0.05	0.37 ± 0.06	1.0 ± 0.2	e)
0	3.58 ± 0.02	3.71 ± 0.05	0.064 ± 0.007	−0.63 ± 0.02	0.83 ± 0.01	1.32 ± 0.03	0.0
25	3.20 ± 0.01	3.5 ± 0.2	0.075 ± 0.003	−0.567 ± 0.009	0.615 ± 0.007	1.08 ± 0.02	0.0181
34	2.79 ± 0.02	2.80 ± 0.06	0.074 ± 0.003	−0.57 ± 0.01	0.571 ± 0.006	1.01 ± 0.01	e)
50	2.70 ± 0.02	2.97 ± 0.06	0.075 ± 0.003	−0.57 ± 0.02	0.628 ± 0.006	1.10 ± 0.01	0.0358
66	c)	2.5 ± 0.1	0.073 ± 0.003	−0.84 ± 0.04	0.83 ± 0.03	1.00 ± 0.05	0.0511
75	2.17 ± 0.01	2.1 ± 0.1	0.073 ± 0.003	−0.622 ± 0.008	0.58 ± 0.01	0.93 ± 0.03	f)
100	0.065 ± 0.004 b)	d)	0.17 ± 0.03 b)	−0.53 ± 0.09 b)	0.47 ± 0.06 b)	1.0 ± 0.2 b)	0.0922

^a)Averages ± standard deviation obtained from the different electrodes on the same chip, except for the 100 mol% 2 case. ^b)Averages ± standard deviation obtained from six different electrodes on the same chip for the 100 mol% 2. ^c)Inconsistencies in CV responses resolved in later experiments but on older films and therefore not reported here but shown in Fig. S-16 of the Supporting Information. ^d)Offset made measurement of charging current not possible. ^e)Spectra not obtained. ^f)Inconsistencies in spectra due to lack of film homogeneity.

The average C of the bare gold microelectrodes (0.0650 ± 0.0004 mF cm⁻², N = 17) before modification has the expected magnitude for these conditions. Interestingly, it is indistinguishable from that for the films resulting from the 100 mol% 2 deposition solution (0.065 ± 0.004 mF cm⁻², N = 3). Yet, a purple color, a substantial amount of charge passed during electrodeposition, and the micrographs of the arrays after electrodeposition (Fig. S-11) confirm the presence of a 100 mol% 2 film. Although the electropolymerization in the 100 mol% 2 solution deposited as much as half of the number of monomers that were deposited from the 0 mol% 2 solution (Fig. 6a), it is as if those sites are inaccessible in the polymer for oxidation and re-reduction and therefore the only contribution to the background current is the charging of the double layer at the polymer/solution interface. It is possible that the 100 mol% 2 films have lost some electrical connectivity with the electrode during the electrochemical characterizations, and thus we are careful not to overinterpret the low C value.

The incorporation of increasing amounts of monomer 2, however, does not adversely affect the measured capacitances for the mixed polymers. For the range of 0 to 75 mol% 2, the decrease in capacitance (Fig. 9, slope = −5.1 × 10⁻³ ± 6 × 10⁻⁴ mol% 2⁻¹, normalized to the 0 mol% 2 value) is the same as the decrease in the quantity of monomers deposited (Fig. 6, slope = −5.2 × 10⁻³ ± 3 × 10⁻⁴ mol% 2⁻¹, normalized to the 0 mol% 2 value). Another way to illustrate this point is by plotting C_normalized vs. m_normalized as shown in Fig. S-17 in section 13 of the Supporting Information. This comparison assesses whether changes in capacitance are simply proportional to the quantity of polymer (expecting a linear relationship and slope of unity) or influenced more by the relative composition of 2:1 in the polymer (expecting a nonlinear relationship and a slope not equal to one). Least squares analysis for points 0 to 75 mol% 2 in Fig. S-17 yields: y_Cnormalized = (1.01 ± 0.09 C_normalized (m_normalized )⁻¹) ${x}_{{m}_{normalized}}$ – 0.01 ± 0.07 C_normalized (R² = 0.9672). The R² value suggests that a linear relationship is not strong. However, the slope is indeed unity, confirming that it is the quantity of monomer deposited that determines the capacitance and not the relative composition of monomers 2 and 1 in the polymer films. The y-intercept of the linear regression ideally represents the double layer capacitance when there is no film (a bare electrode). Its error brackets the measured (and normalized) double layer capacitance of the bare electrodes (0.0650 ± 0.0004 mF cm⁻²/3.65 ± 0.06 mF cm⁻² = 0.0178 ± 0.0003).

This retained electrochemical activity suggests that the bulkier side group of monomer 2 does not disrupt the conductivity between neighboring oligomer groups within the polymer network. In addition, it appears that the negative charge of monomer 2 due to deprotonation of the carboxylic acid groups, as is supported by the μATR-FTIR data, does not impede counter ion compensation that is needed for the polymer to oxidize and reduce. The presence of carboxylate anions within the polymer likely contribute toward ion compensation themselves, requiring cations to enter the films because the carboxylates are tethered, transferring the role of charge compensation from anions to cations, as is known to occur in polymers of PEDOT formed with polystyrenesulfonate (PSS). ⁸¹ It should also be noted that the rinsing, electrochemical cycling in the buffer, and storage in water after the electropolymerization step remove residual SDS. This is supported by the lack of absorption bands specific to SDS in the μATR-FTIR data. Thus, the normalizing effect that SDS has on the different side chains of the monomers during electropolymerization no longer applies to the monomer units within the polymer after the films are formed.

Investigation of the faradaic electrochemistry of [Fe(CN)₆]⁴⁻ in buffer using CV at electrodes modified with the conducting polymer films was intended to establish the facility to electron transfer across the polymer/solution interface. The shapes of the resulting CV responses (see Fig. S-16 in section 13 Supporting Information) for all films at 0.100 V s⁻¹ are consistent with mass transport that is dominated by planar, diffusion-limited behavior. The diffusion length is estimated to be approximately 80 μm by the Einstein equation, which is on the same order of magnitude as the width of the electrodes and suggests that there may be some minor contribution from hemicylindrical diffusion as well. The background-subtracted forward peak currents, i_pa , reverse peak currents, i_pc , their ratio, i_pc /i_pa , and peak separations, ∆E_p are listed in Table II.

The faradaic peak currents (background-subtracted) from the CV responses at the different polymer films can be compared. The Randles-Sevcik equation predicts the forward peak current to be 0.43 μA at a planar electrode with dimensions of the microband electrodes (100 μm × 2000 μm, assuming the height 0.025 μm provides a negligible contribution), at 23 °C, and for [Fe(CN)₆]⁴⁻ (n = 1 mole of electrons/mole of molecules) using a diffusion coefficient of 0.65 × 10⁻⁶ cm² s⁻¹. ⁸² This predicted value is within the error of the i_pa obtained for the bare gold electrodes. The i_pa values for the polymer films range from 22 to 93% larger than this, with a median of 31%, and do not appear to follow any particular trend with the mol% 2. The higher current could result from the higher surface area of the polymer films. The variation in i_pa could derive from polymer growth that is affected by a combination of planar and hemicylindrical diffusion during each deposition sweep that builds on the morphology of the previous deposition, producing nonuniform films. This is consistent with cross sectional areas and nonuniform profiles measured by profilometry (Fig. S-9 in section 6 of the Supporting Information).

The ratio of the reverse peak current to that of the forward peak (i_pc /i_pa ) is expected to be unity for a reversible electron transfer with a stable anodic product. This ratio also allows comparisons of electrochemical behavior between different films by normalizing out the varying i_pa values. The ratio appears to follow a trend from the largest of 1.32 ± 0.04 for the 0 mol% 2 film to the smallest of 0.89 ± 0.02 for the 100 mol% 2 film. A ratio greater than unity is consistent with films that exhibit an expansion during oxidative excursions to accommodate the incorporation of anions to compensate charge. The film is most expanded in its anodic state, as has been observed for thick films of PEDOT as described previously. ⁶⁰ A growing film can oxidize increasingly more ferrocyanide in the forward sweep past the anodic peak and possibly push through the diffusion layer to access fresh solution. As the potential passes E° in the return sweep, a larger peak occurs because of a higher ferricyanide concentration, than would have been produced had the film not increased size. Ratios less than unity for films with high mol% 2 suggest that the least expanded film is in its anodic state, when tethered carboxylate anions internally compensate for the positive charge along the backbone. The oxidation peak of ferrocyanide appears with diminishing oxidation in the forward sweep as the film contracts. It is during the reductive excursions that the film expands, when cations must enter the polymer to compensate for the immobilized negative charges of the carboxylates, but there is less ferricyanide available to reduce than if the film had not changed size. A smaller faradaic peak in the return sweep of films of high mol% 2 could also be partially explained by more repulsion of the anionic redox species, whereas in the forward anodic sweep, [Fe(CN)₆]⁴⁻ can also serve as a counter ion for the polymer oxidation. However, the high concentration of supporting electrolyte should shield most of the migration of the redox species in solution.

The separation of the forward and reverse peaks, ΔE_p, is near the ideal value of +0.059 V for n = 1, 25 °C, and fast heterogeneous electron transfer kinetics on the time scale of the experiment, ⁸³ for all films containing monomer 1 (0–75 mol% 2). However, the ∆E_p for films from the 100 mol% 2 solution. is + 0.17 ± 0.03 V. This valueis within a standard deviation of that for the bare gold electrodes, ΔE_p = + 0.14 ± 0.03 V. The larger ∆E_p could be due to two reasons. One is that the electron transfer kinetics are slower than at films containing monomer 1, because of different electrode surface composition to which ferrocyanide (an inner-sphere redox species) is known to be sensitive. ⁸⁴ The other is that there is substantial uncompensated resistance within the electrode or polymer. Because the 100 mol% 2 films behave like the bare gold electrodes in buffer alone for the proposed reasons described above, the large ΔE_p is most likely due to slower heterogeneous kinetics.