Enhanced Electron Transfer Rates by AC Voltammetry for Ferrocenes Attached to the End of Embedded Carbon Nanofiber Nanoelectrode Arrays

Abstract

The effect of the interior structure of carbon nanomaterials on their electrochemical properties is not well understood. We report here the electron transfer rate (ETR) of ferrocene (Fc) molecules covalently attached to the exposed end of carbon nanofibers (CNFs) in an embedded nanoelectrode array. The ETR in normal DC voltammetry was found to be limited by the conical graphitic stacking structure interior of CNFs. AC voltammetry, however, can cope with this intrinsic materials property and provide over 100 times higher ETR, likely by a new capacitive pathway. This provides a new method for high-performance electroanalysis using CNF nanoelectrodes.

Keywords: Biosensors; Electron transfer rate; Microstructural electrical network; Nanoelectrode array; Vertically aligned carbon nanofibers.

Fig. 1.

(a) Schematic of the embedded CNFs functionalized with ferrocene molecules at the exposed tip. The bottom brown color represents Si wafer, yellow color represents Cr contact layer under CNFs, blue color represents dielectric SiO₂ deposited using TEOS CVD to encapsulate individual CNFs, dark black circle represents exposed CNF tips and grey circles represent the unexposed CNF tips buried in SiO₂ matrix. Ferrocene moieties are functionalized to the tip and the sidewall of the exposed CNFs. (b) and (c) show top and 45° perspective views of scanning electron microscopic images of embedded CNFs with the scale bars of 300 and 200 nm, respectively.

Fig. 2.

(a) and (b) Cyclic voltammetry measurements of Fcfunctionalized GCE and CNF NEA in 1.0 M KCl, respectively. Each set of measurements was performed at a series of scan rates of 0.020, 0.10, 0.50, and 1.0 V/s. The oxidation and reduction currents were normalized to the geometric surface area defined by the 3-mm i.d. O-ring.

Fig. 3.

(a) The plots of background corrected peak currents (i_p) derived from the CVs in Figure 2 of Fc-functionalized GCE and CNF NEA in 1.0 M KCl. (b) Plot of logarithm of (i_p) vs. logarithm of the scan rate, the solid lines are the linear fitting curves by equation log(i_p)=a+b log(v) with the slope b=0.95 for GCE and b=0.38 for CNF NEA.

Fig. 4.

(a)–(f) AC voltammograms measured at 10, 75, 3500 Hz with a sinusoidal wave of 25 mV in amplitude superimposed on the DC staircase ramp from −0.05 to 0.65 V at a scan rate of 10 mV/sec. (a)-(c) are measurements on a Fc-functionalized GCE electrode, (d)–(f) on a Fc-functionalized CNF NEA. The black curves are the background currents measured with clean GCE and CNF NEA electrodes without Fc attachment in 1.0 M KCl and grey dotted curves are the ACV signals in 1.0 M KCl after functionalization of electrodes with Fc.

Fig. 5.

(a) Background corrected peak currents (i_p,ac) of surface-attached Fc molecules in AC voltammetry vs. the frequency for the Fc-functionalized GCE and CNF NEA, respectively. The maximum i_p,ac is at 75 Hz for the GCE whereas at 3500 Hz for the CNF NEA. (b) The same data plotted in log–log scale.

Fig. 6.

(a) and (b) Nyquist plots of the electrochemical impedance spectra of Fc-functionalized GCE and CNF NEA in 1.0 M KCl. The spectra were recorded at 20 mV voltage amplitude, 0.1 Hz to 100 kHz frequency range, and biased at a DC potential of +0.275 V vs. Ag/AgCl (sat’d KCl). (c) and (d) are the bode plots (phase angle vs. log (frequency)) of the same experiment. The grey solid line is the fitting curve obtained by using the equivalent circuits shown in insets. The fitting parameters are listed in Table 2.