Superior electrochemical platforms based on polymer carbon nanotube composite electrodes

Superior Electrochemical Platforms based on Polymer Carbon Nanotube Composite Electrodes Suriya Ounnunkad1, Andrew I. Minett1, Barry D. Fleming2, Chong-Yong Lee2, Alan M. Bond2, Gordon G. Wallace1 1

ARC Centre of Excellence for Electromaterials Science, Intelligent Polymer Research Institute, University of Wollongong Innovation Campus, AIIM Building, Squires Way, Fairy Meadow, NSW 2519, Australia 2 School of Chemistry, Monash University, Clayton, Victoria 3800, Australia [email protected], [email protected], [email protected], [email protected]

Abstract—Putting insulating polymers into a highly conductive single-walled carbon nanotube paper leads to excellent electrochemical performance; fast redox reactions and high signal to background noise ratio. The ability of such composites to serve as superior electrochemical platforms was investigated by using DC and AC cyclic voltammetry. The electrochemical platforms show benefits in sensing applications with fast signal generation and low limits of detection. Keywords-Buckypaper; intercalation; cyclic voltammetry; AC voltammetry

I.

INTRODUCTION

Over the last two decades, carbon nanotubes (CNTs) have been intensively investigated in various architectures for use as electrochemical sensing platforms [1-4]. These electrochemical platforms can be directly used as electrodes or functionalised with active sensing elements for detection applications with good stability and selectivity [1-6]. Various electrode architectures can be made from CNT soot, such as CNT paste electrodes[9], screen-printed electrodes [10], fibres [11], papers [12], or biogels [13]. CNT structuring and patterning such as aligned CNTs [8] and CNT Nanowebs [14] can be achieved by careful design of chemical vapour deposition processes. CNT papers or Buckypapers (BPs) have a unique character with very high capacitance contributed by their high electro-active surface areas and high porosity, forming large double layer capacitive films on surfaces. For these types of architectures, it is much more difficult to use them as sensing electrodes since they suffer from very large background currents due to this capacitive charging [15]. With low capacitive charging, ultrathin CNT BPs sitting on membranes and their composites also offer enhanced performance as sensing platforms for chemical and biochemical sensors [16] while nafion-CNT composite BPs revealed an ability to oxidize NADH, benefits for glucose biosensors [17]. The high contribution from this capacitive charging current masks the redox process in electrochemical detection techniques; ie: lowering the signal to background ratio. The fabrication of sensing platforms from BPs with good sensitivity and selectivity is a challenging issue in electroanalysis and is the focus of this report. In this present work, an inverse modification process which uses insulating polymers to improve the electrochemical

sensing efficiencies of highly capacitive CNT architectures is reported. This is in contrast to the usual route of embedding CNTs into polymer composites. This process results in a novel CNT BP nanocomposite electrode platform via the intercalation of organic, non-water soluble, non-conducting polymers, which has the added benefit of providing good mechanical properties and stability for practical applications. The structure surprisingly provides an improvement in the signal to noise ratio in terms of Faradaic to background charging current ratio. There have been only three intercalated organic polymers identified so far, which have a very low capacitance. These are; poly(styrene-β-isobutylene-β-styrene) (SIBS), polyisobutylene (PIB) and polystyrene (PS). The polymers effectively decrease the total free volume available by blocking the interstitial, interbundle and/or intertube voids in the architecture (see Fig. 1). They additionally provide an electrode surface that is predominantly more hydrophobic in nature to the raw BP, which reduces electrolyte access into the pore structure reducing the formation of double layer capacitive films. In this study, we use a combination of DC and AC cyclic voltammetry to evaluate the properties of these prepared electrode platforms. Fourier Transformed (FT-AC) voltammetry is a powerful technique that provides information on electron transfer rates and kinetics in a single experiment [18]. II. EXPERIMENTAL DETAILS Free-standing single-walled carbon nanotube (SWCNT) mats were prepared by a vacuum-assisted filtration of welldispersed SWCNT solution [12]. Triton-X100 was used as a dispersant. After SWCNT paper settled on a 0.22-µm hydrophobic PVDF membrane, the paper was washed with Milli-Q water several times to remove non-interacting TritonX100 and followed with ethanol. The SWCNT membrane was peeled off from PVDF membrane. It was left for drying in a room temperature. For polymer-SWCNT composite BP, the three different insulating polymers each was intercalated by soaking the raw BP in a 5%w/v polymer solution at desired periods. After that, intercalated BP was carefully rinsed with the solvent and let dry at room temperature. The composite Buckyelectrodes (BEs) were used as working electrodes in a three-electrode electrochemical cell with a platinum mesh auxiliary electrode and a Ag/AgCl (3M NaCl) reference

electrode. AC voltammetry with a sine-wave perturbation was undertaken  using  a  Fourier  Transformed  alternating  current  (FT‐AC)  instrument  build  by  Prof.  Alan  Bond’s  research  group  at  Monash  University  [19].  Direct  current  (DC)  voltammograms  were  recorded  using  the  FT‐AC  instrument  with  a  zero  amplitude  for  the  sine  wave.  The  simulation  for  DC  data  was  performed  by  using  a  commercial package program (DigiSim) [20].  III.

(a)

RESULTS AND DISCUSSION

Fig. 1 shows AFM images scanned on surfaces of raw BP (a) and SIBS-intercalated BP (b) showing a random oriented structure of SWCNT networks in which a porous native is evident. Pores or “pocket” refer to the intertube/interbundle spacing. After intercalation of the SIBS polymer, the resultant structure shows less porosity, which suggesting pore filling by the polymers. This suggests that the composite BPs allow less electrolyte entering into the internal CNT network structure to form high double-layer capacitive charging films along the electrode surface. The resultant composite structure is a randomly arranged CNT nano-/micro-electrode array, as shown in Fig. 1c. The microarray electrodes usually have unique electrochemical properties such as small capacitivecharging currents, reduced iR drop, and steady-state diffusion currents [21]. In addition, no capacitance was contributed from insulating polymers; SIBS, PIB and PS. These in turn govern the dramatically reduced background charging current observed in Fig.2. (a)

significantly improved Faradaic responses are also observed at the BP polymer composite platforms, indicating fast electron transfer rates whilst maintaining significantly more surface reactivity than at the bare BP electrode.

(b)

(b)

(c)

Figure 1. Platform structure: AFM images for (a) bare and (b) SIBSintercalated Buckyelectrodes. (c) Cartoon represents a CNT microelectrode domain array.

For example, Fig. 2a-b shows the comparison of DC cyclic voltammograms and FT-AC Fundamental components respectively for the [Ru(NH3)6]3+/2+ redox process at bare and PIB-intercalated BPs, which are also representative of the results from the intercalation of PS and PIB. The bare BP shows very high capacitive charging currents in both the AC and DC voltammograms whilst a ten-fold decrease in capacitive current is observed after intercalation of PIB into the CNT architecture. The small peak-to-peak separation (∆Ep) and

Figure 2. Comparison of Faradaic to Background charging current ratios (a) DC cyclic voltammograms and (b) in fundamental components derived from FT-AC cyclic voltammograms, respectively obtained from reduction of 1.00mM [Ru(NH3)6]3+ in PBS solution (pH7.4, 0.1M KCl) at bare (blue line) and PIB-intercalated Buckypapers (brown line, with an intercalation period of 30min) at ν = 59.60mV.s-1 and room temperature. AC conditions employed: a single sine wave, f = 34.98Hz, ∆E = 80mV, Estart = 200mV, Eswitch = -600mV. TABLE 1. MIDPOINT POTENTIAL (EM), PEAK-TO-PEAK SEPARATION (∆EP) AND FARADAIC TO BACKGROUND CHARGING CURRENT RATIO (IF/ICDL ) FOR THE [RU(NH3)6]3+/2+ REDOX COUPLE AT THE ELECTRODE SURFACES

Electrodes Edge-plane graphite Raw BP SIBS-intercalated BP PIB-intercalated BP PS-intercalated BP

Electrochemical parameters IF/ICdl Em (mV) ∆Ep (mV) -169 59 4.22c -258 277 0.004c a a -168 69 4.76c b b -177 73 8.85c b b -151 77 4.16c a. Intercalation period 72h. b. Intercalation period 30min. c. Measured from FT-AC Fundamental component

The electrochemical parameters from DC voltammetry for novel BP platforms intercalated with the three polymers (SIBS, PIB, and PS) are illustrated in Table 1, in comparison with a well known electrochemical electrode, edge-plane graphite and the reference bare BP. The electrochemical performance of the polymer-intercalated BP platforms behave similarly to an edgeplane like electrode with high Faradaic to background charging current ratio and small ∆Ep values. Within experimental error, the reversible potentials, Em for novel SIBS and PIB composites are also close to that of edge-plane graphite electrodes, indicating the same surface energy and electrochemical reactivity. In comparison, the bare BP electrode exhibits a three orders of magnitude lower Faradaic to background charging current ratio. Furthermore, FT-AC fundamental harmonic voltammograms show that the redox reaction at the electrode platforms becomes more reversible after BPs were intercalated with polymer for intercalation periods above 1h (see Figure 3). Superimposing the oxidation and reduction peak currents or scan forward and backward directions for composite electrodes with different intercalation periods are shown. For a reversible electron transfer process, the peak current positions of all harmonics should coincide at the reversible or midpoint potential [22].  The positions of the reductive and oxidative peaks for FMCA0/+ process, show improvement after the intercalation period of 30min and completely coincident at 6h for the PIB-intercalated BP electrode system, possessing high reversibility (fast kinetics). As proposed in Fig.1c, the structure of the electrode surface contributes to the characteristics of the CVs. It is believed that the inner-sphere surface-sensitive ferricyanide ([Fe(CN)6]3-) redox couple are dependent on the electrochemical reactivity and surface morphology of the electrode [23]. Therefore, this present study uses this redox probe to elucidate the relationship between electrode surface and electrochemistry. It was found that the peak positions, but not the voltammetric shape match the experimental data by assuming the entire electrode area is electroactive. Clearly, the shape (sigmoidal component) has some steady-state characteristic associated with microelectrode behaviour [24], and simulation based on this component was introduced by considering each array component as an independent electroactive centre [20]. Fig. 4 demonstrates the simulation of reduction of ferricyanide with overlaid different simulation data. The simulation of experimental data was optimised by varying the size and number of array microelectrodes on the composite BP electrodes to characterise the steady-state shape behaviour. In order to achieve the right current densities, the current of each simulated CV for a microelectrode was multiplied by the required electroactive microelectrode component number as explained if the randomly dispersed CNT networks are embedded in the insulating SIBS polymer. Therefore, a microarray behaviour model with hemispherical diffusion of each array microelectrode was used for the simulation of the DC CV for the reduction of ferricyanide at a SIBS-intercalated BP electrode. It is plausible that the intercalated SWCNT network mat acts as an individual SWCNT/SWCNT bundle nano-/micro-electrode array where the SWCNTs and/or bundles are surrounded by insulating polymer domains.

Figure 3. Comparison for current amplitude of fundamental components derived from FT-AC cyclic voltammograms for oxidation of 1.00mM ferrocenemonocarboxylic acid (FMCA) in PBS solution (pH7.4, 0.1M KCl) at PIB-intercalated BPs with different intercalation periods. The condition employed: a single sine wave, f = 34.98Hz, ∆E = 80mV, Estart = 0mV, Eswitch = 800mV, ν = 59.60mV.s-1 and room temperature.

Figure 4. Simulation profiles of DC cyclic voltammograms for reduction of 1.00mM [Fe(CN)6]3+ in PBS buffer solution (pH7.4, 0.1M KCl) at SIBSintercalated BPs (soaking time of 72h) at scan rate = 59.60 mV.s-1 and room temperature. Parameters used in simulation are rate constant = 0.0045 cm.s-1, electron transfer coefficient = 0.5, reversible potential = 0.232 V, diffusion coefficients of oxidized and reduced forms = 6.3 × 10-6 cm2.s-1 and 7.6 × 10-6 cm2.s-1 respectively, concentration of [Fe(CN)6]3- = 1.00 mM, scan rate = 59.60 mV.s-1, temperature = 298.2K. Simulated data based on microarray model with hemispherical diffusion which shows variable size and number of array microelectrodes and optimised parameters; area of each designated microelectrode = 1.846 × 10-4 cm2, consisting of 650 individually array CNT domain microelectrodes, double-layer capacitance = 40 μF.cm-2.

Figure 5. FT-AC fourth harmonic for oxidation reaction of 1mM uric acid at a novel SIBS-intercalated SWCNT platform in PBS (pH 7.4). The condition employed: a sine wave, f = 35 Hz, ∆E = 80 mV, v =74.51 mVs-1, Estart = -200 mV, and Eswitch = 800 mV.

By eyesight the optimisation of size and number of array microelectrode on the composite BP electrode surface placed in the electrolyte and the optimal simulated DC CV with an enough quality is provided (see Fig. 4). The simulation data are based on the hypothesis that the electrode consists of a random assembly of non-intercalating microelectrodes with hemispherical diffusion by variable size and number. The simulation parameters are shown in the caption of Figure 4. Fig. 5 emphasizes FT-AC fourth harmonic for electrocatalytic oxidation of uric acid at a novel SIBSintercalated SWCNT BP via a heterogeneous two-electron transfer, irreversible process [25], indicating fast reaction processes. This would suggest fast signal generation in electrochemical sensing devices made from these novel electrodes. IV.

CONCLUSIONS

In summary, BP polymer composites have high capability for use as high performance electrochemical platforms. The morphology and voltammetric data of the BP composite has good agreement with the assumption of a microelectrode array model. The addition of an intercalated insulating polymer leads to significant improvements in the Faradaic to background capacitive charging current ratio and electron transfer kinetics of modified carbon nanotube electrodes. The novel sensing platforms also offer high sensitivity and fast signal generation, compared to a bare BP electrode which will present significant benefits with respect to electro-analytical applications. ACKNOWLEDGMENT The financial support of the Australian Research Council (ARC) is gratefully acknowledged. S.O. also acknowledges award of a PhD Scholarship in Nanoscience and Nanotechnology (Nanodevices) by the Higher Educational Strategic Scholarships for Frontier Research Network from the Commission on Higher Education, Ministry of Education, Royal Thai Government. A.I.M. acknowledges a QEII Research Fellowship from the ARC. A.M.B and G.G.W gratefully acknowledge the support of the ARC Federation Fellowship scheme. REFERENCES [1] [2]

[3] [4] [5]

[6]

J. Wang, “Carbon-nanotube based electrochemical biosensors: A review, Electroanalysis,” 17, 7-14 (2004). L. Agüí, P. Yáñez-Sedeño, J.M. Pingarrón, “Role of carbon nanotubes in electroanalytical chemistry: A review,” Analytica Chimica Acta 622, 1147 (2008). G.A. Rivas, et al. “Carbon nanotubes for electrochemical biosensing,” Talanta 74, 291-307 (2007). M. Trojanowicz, “Analytical applications of carbon nanotubes: a review,” TrAC Trends in Analytical Chemistry 25, 480-489 (2006). D. Vairavapandian, P. Vichchulada, M.D. Lay, “Preparation and modification of carbon nanotubes: Review of recent advances and applications in catalysis and sensing,” Analytica Chimica Acta 626, 119129 (2008). C. Lynam, N. Gilmartin, A.I. Minett, R. O'Kennedy, and G. Wallace, “Carbon nanotube-based transducers for immunoassays,” Carbon 47, 2337-2343 (2009).

[7]

[8]

[9] [10]

[11]

[12]

[13] [14] [15] [16]

[17]

[18]

[19] [20]

[21]

[22] [23]

[24]

[25]

G.Z. Fang, J.X. He, S. Wang, “Multiwalled carbon nanotubes as sorbent for on-line coupling of solid-phase extraction to high-performance liquid chromatography for simultaneous determination of 10 sulfonamides in eggs and pork,” Journal of Chromatography A, 1127, 12-17 (2006). S. Goswami, et al. “Aligned carbon nanotube stationary phases for electrochromatographic chip separations." Chromatographia 69, 473-480 (2009). M.D. Rubianes, G.A. Rivas, “Carbon nanotubes paste electrode,” Electrochemistry Communications 5, 689-694 (2003). A. Bonanni, M.J. Esplandiu, M. del Valle, “Impedimetric genosensors employing COOH-modified carbon nanotube screen-printed electrodes,” Biosensors and Bioelectronics 24, 2885-2891 (2009). C. Dechakiatkrai, et al. “Single-Walled Carbon Nanotube/Trititanate Nanotube Composite Fibers,” Advanced Engineering Materials 11, B55B60 (2009). C.J. Frizzell, et al. “Reinforcement of macroscopic carbon nanotube structures by polymer intercalation: The role of polymer molecular weight and chain conformation,” Physical Review B 72, 245420 (2005). B.C. Thompson, et al. “Carbon nanotube biogels,” Carbon 47, 12821291 (2009). J. Chen, et al. “Direct growth of flexible carbon nanotube electrodes,” Advanced Materials 20, 566-569 (2008). C. Zhou, PhD Thesis, “Carbon nanotube based electrochemical supercapacitors,” Georgia Institute of Technology, 2006. C. Hu, Y. Ding, Y. Ji, J. Xu, and S. Hu, “Fabrication of thin-film electrochemical sensors from single-walled carbon nanotubes by vacuum filtration” Carbon In press (2010). C. Luo, et al. “Flexible carbon nanotube-polymer composite films with high conductivity and superhydrophobicity made by solution process,” Nano Letters 8, 4454-4458 (2008). C.Y. Lee and A.M. Bond, “Evaluation of levels of defect sites present in highly ordered pyrolytic graphite electrodes using capacitive and faradaic current components derived simultaneously from largeamplitude Fourier Transformed ac voltammetric experiments,” Analytical Chemistry 81, 584-594 (2009). A.M.  Bond,  N.W.  Duffy,  S.‐X.  Guo,  J.  Zhang,  D.  Elton,  “Changing  the  look of voltammetry,” Analytical Chemistry 77, 186A‐195A (2005). T.J. Davies, C.E. Banks, and R.G. Compton, “Voltammetry at spatially  heterogeneous electrodes,” Journal of Solid State Electrochemistry 9,  797‐808 (2005). X.-J. Huang, A.M. O’Mahony, and R.G. Compton, “Microelectrode arrays for electrochemistry: Approaches to fabrication,” Small 5, 776788 (2009) A.J. Bard and L.R. Faulkner, Electrochemical Methods: Fundametals and Applications, 2nd ed., John Wiley & Sons, 2001, pp.388-401. X. Ji, C.E. Banks,A. Crossley, and R.G. Compton, “Oxygenated edge plane sites slow the electron transfer of the ferro-/ferricyanide redox couple at graphite electrodes,” ChemPhysChem 7, 1337-1344 (2006) J. Guo and E. Lindner, “Cyclic voltammograms at coplanar and shallow recessed microdisk electrode arrays: Guidelines for design and experiment,” Analytical Chemistry 81, 130-138 (2009). A.P. O’Mullane, J. Zhang, A. Brajter-Toth, A.M. Bond, “Higher harmonic large-amplitude Fourier Transformed alternating current voltammetry: Analytical attributes derived from studies of the oxidation of ferrocenemethanol and uric acid at a glassy carbon electrode,” Analytical Chemistry 80, 4614-4626 (2008).