Electrochemical activation of carbon nanotube/polymer composites

Abstract

Electrochemical activation of carbon nanotube/polysulfone composite electrodes for enhanced heterogeneous electron transfer is studied. The physicochemical insight into the electrochemical activation of carbon nanotube/polymer composites was provided by transmission electron microscopy, Raman spectroscopy, electrochemical impedance spectroscopy, and cyclic voltammetry. Dopamine, ascorbic acid, NADH, and ferricyanide are used as a model redox system for evaluating the performance of activated carbon nanotube/polymer composite electrodes. We demonstrate that polymer wrapping of carbon nanotubes is subject to defects and to partial removal during activation. Such tunable activation of electrodes would enable on-demand activation of electrodes for satisfying the needs of sensing or energy storage devices.

---

1. Introduction

Hybrid carbon nanotube/polymer composites have gained significant attention in recent years due to their enhanced functionality and multifunctional properties.¹ There have been a significant number of reports on the advantages of the use of carbon nanotube/polymer composite electrodes for sensors² and biosensors^3,4 and for energy storage.⁵

There is always a need to enhance the performance of electrodes. One of the ways of increasing the heterogeneous electron transfer rate of carbon-based electrodes is to subject them to electrochemical oxidation (anodization) to introduce defects and oxygen-containing groups on their surface. This process has been known for a long time and has been extensively studied for graphite-based electrodes^6–9 and, more recently, for carbon nanotube-based electrodes.^10–13 Wang et al.¹⁰ electrochemically activated single-walled carbon nanotube (SWCNT) film electrodes but did not provide any insight as to the cause of such activation. Musameh et al.¹¹ electrochemically activated multiwalled carbon nanotube (MWCNT) film electrodes and demonstrated enhanced electrochemical activity towards the redox behavior of several analytes, such as NADH, H₂O₂, hydrazine, and ascorbic acid. Using electrochemical methods Musameh et al. concluded that the enhancement of electrochemistry at the MWCNT film electrodes is due to breaking caps of arc produced nanotubes.¹¹ Recently, we have extensively studied the electrochemical activation of MWCNT film electrodes by cyclic voltammetry, electrochemical impedance spectroscopy (EIS), transmission electron microscopy, X-ray photoelectron spectroscopy, and Raman spectroscopy.¹³ We concluded that the enhancement of the heterogeneous electron transfer rate is due to the introduction of defects in the MWCNT walls. All previous studies^10–13 were carried out for CNT films which were cast on the surface of glassy carbon electrodes. However, for practical applications, the use of CNT/polymer composite electrodes is highly favorable due to their superior mechanical stability. As far as we are concerned, no studies have been conducted to elucidate the reasons for enhanced heterogeneous electron transfer at CNT/polymer composites that underwent electrochemical activation.

Here, we report on the systematic investigation of electrochemical activation of carbon nanotube/polymer composites by transmission electron microscopy (TEM), electrochemistry, electrochemical impedance spectroscopy, and Raman spectroscopy.

2. Experimental

2.1 Apparatus

A JEM 2100F Field Emission Transmission Electron Microscope (JEOL, Japan) operating at 200 kV was used to acquire STEM and HR-TEM images in a scanning TEM mode (0.19 nm point resolution and 0.1 lattice resolution). Raman spectra were obtained by using a T64000 Raman Spectrometer (Horiba Jobin Yvon Ltd., Japan) using 514.5 nm excitation from an Ar ion laser beam employing backscattering geometry. The working electrodes were made by screen-printing using a Dek248 Semiautomatic Screen Printer System (Asflex S.A. Int., Spain). The squeegees used were soft polymer-type and the pressure applied during the printing process was set at 7 kg cm⁻². A double-sweep process was programmed at a speed of 20 mm s⁻¹. All voltammetric and impedimetric experiments were performed using an Autolab PGSTAT30 Electrochemical Analyzer (Eco Chemie, The Netherlands) connected to a personal computer and controlled by General Purpose Electrochemical Systems v 4.9 and Frequency Response Analysis software (Eco Chemie). Electrochemical and impedimetric experiments were performed in a 5 mL voltammetric cell at room temperature (25 °C), using a three-electrode configuration. A platinum electrode served as an auxiliary electrode and Ag/AgCl as a reference electrode. All electrochemical potentials in this paper are stated vs.Ag/AgCl. Cyclic voltammogram experiments were performed at a scan rate of 100 mV s⁻¹ using 50 mM phosphate buffer (pH 7.4).

2.2 Materials

Polysulfone (PSf) was obtained from BASF (BASF Ultrasons S 3010 natur, Frankfurt, Germany). Multi-walled carbon nanotubes (MWCNT) with a length of 0.5–200 μm, external diameter of 30–50 nm, and internal diameter of 5–15 nm were obtained from Aldrich (Steinheim, Germany). The MWCNT/PSf composite electrode consisted of a single working screen-printed electrode (SPE) deposited onto a polycarbonate (PC) substrate. The electrodes were prepared using three inks: conductive silver ink, carbon, and insulating ink were printed and cured consecutively in the furnace at 60 °C for 30 min. The Acheson carbon ink (Electrodag 400 B), conductive silver ink (Electrodag 6037 SS), and insulating ink (Minico M 7000) were obtained from Acheson Colloids Co. (Scheemda, The Netherlands). The area of the working electrode was 20 mm². Ferricyanide, dopamine, ascorbic acid, NADH (nicotinamide adenine dinucleotide, reduced form), N,N-dimethylformamide (DMF), potassium phosphate dibasic, and potassium chloride were purchased from Sigma-Aldrich, Japan.

2.3 Composite electrode preparation

As-received carbon nanotubes were stirred in concentrated nitric acid (6 M) for 24 h at 80 °C for additional purification from residual metallic catalyst nanoparticles.^14,15 The acid/MWCNT mixture was washed with distilled water until neutral pH was reached and subsequently filtered. The purified MWCNT were dispersed in DMF by placing the suspension into an ultrasonic bath for 10 min. The PSf solution was prepared by dissolving the polymer in 7.5% wt DMF. The composite was achieved by mixing the PSf solution with the MWCNT suspension (1 : 1) with a final MWCNT concentration of 5 mg mL⁻¹. The mixture was sonicated for 30 min. The solution was stable for at least 2 months. For casting the working electrode, 20 μL of suspension was pipetted on the working area. The composite was coagulated by phase inversion adding 20 μL of water to it and immersing the electrode in a water bath for 10 s.¹⁶ Thickness of resulting MWCNT/PSf membrane was 120 μm.¹⁷

2.4 Composite characterization

Cyclic voltammetric experiments were performed at a scan rate of 100 mV s⁻¹ using PBS buffer (50 mM phosphate, 100 mM KCl, pH 7.0). Electrochemical impedance spectroscopy was performed in 0.1 M KCl. Fe(CN)₆^3–/4– was used as an electrochemical probe at a concentration of 10 mM. The capacitance values were extracted by fitting to Nyquist plots using Randles equivalent circuit. For the TEM measurements, MWCNT/PSf composites were cut by microtome and then transferred to a TEM grid.

3. Results and discussion

Screen-printed MWCNT/polysulfone composite electrodes were subjected to electrochemical activation at different potentials (0–2 V) and times (0–300 s) and afterwards studied by TEM, cyclic voltammetry, EIS, and Raman spectroscopy.

3.1 Electron microscopy

Fig. 1 shows representative TEM images of MWCNT/PSf composite before (A, B) and after electrochemical activation (C, D). The as-prepared MWCNT/PSf sample contains a thick polymer wrapping (sheath) around a MWCNT core with an average thickness of ~8 nm (A, B). There were no observable defects in the MWCNT structure at the non-activated electrodes. In contrast, TEM observations of the electrochemically activated MWCNT/PSf composites (at 1.75 V for 180 s) show striking differences. The structure of electrochemically activated MWCNT was damaged with observable introduction of wall defects on the MWCNT (C). This observation is consistent with our previous report.¹³ Furthermore, the observed thickness of the amorphous polymer layer covering the MWCNT suffered a sharp decrease after the electrochemical treatment as indicated in Fig. 1(D). This decrease is because electrochemical activation at high potentials leads to the decomposition of water and the generation of oxygen, which oxidizes the polymer wrapping around the MWCNT.

Fig. 1 TEM images of MWCNT/PSf composite before (A, B) and after electrochemical activation at 1.75 V for 180 s (C, D).

3.2 Cyclic voltammetry

The electrochemical performance of the as-produced MWCNT/PSf composite and electrochemically activated composite was evaluated by cyclic voltammetry of ferricyanide, ascorbic acid (AA), NADH, and dopamine (Fig. 2A–D). The responses of all four analytes increased dramatically when the electrodes were electrochemically activated at 1.75 V for 60 s (b) and further improved after activation for 300 s (c).

Fig. 2 Cyclic voltammograms of 5 mM ferricyanide (A), 5 mM ascorbic acid (B), 5 mM NADH (C), and 5 mM dopamine (D) on MWCNT/PSf composite. MWCNT/PSf composite was electrochemically activated at 1.75 V for 0 s (a), 60 s (b), and 300 s (c) in phosphate buffer (50 mM, pH 7.4). Conditions: PBS buffer, pH 7.4; scan rate, 100 mV s⁻¹.

The heterogeneous electron transfer was enhanced when the MWCNT/PSf composite was electrochemically activated, resulting in a lowering of the overpotentials of ascorbic acid, NADH, and dopamine (Fig. 2B, C, D). The overpotential for the oxidation of ascorbic acid was lowered from 0.29 to 0.10 V after electrochemical activation for 300 s at 1.75 V (Fig. 2B, a and c). The overpotential for the oxidation of NADH decreased from 0.56 to 0.41 V (Fig. 2C). For the detection of dopamine, the trend is the same as that observed for NADH and ascorbic acid, achieving a decrease in the overpotential for oxidation from 0.27 to 0.18 V (Fig. 2D). The dopamine also underwent a high decrease of ΔE from 181 mV to 83 mV. In contrast, this effect is not so noticeable for ferricyanide even after 300 s of electrochemical activation (Fig. 2A). The relatively small influence of electrochemical activation on ΔE of ferricyanide is because the heterogeneous electron transfer between the MWCNT/PSf composite and ferricyanide is very rapid and further improvements have little effect on ΔE.

For further demonstrations of the activation effect on the MWCNT/PSf composite electrodes, the influence of different activation potentials were studied on the response to 5 mM of ferricyanide, ascorbic acid, NADH, and dopamine. Fig. 3 shows the dependency of change of ΔE (a, d) and peak potentials (b, c) for the activated electrodes for the different analytes. For ferricyanide (a) and dopamine (d), the peak-to-peak separation ΔE is shown, subtracted from ΔE at the non-activated electrodes. For ascorbic acid (b) and NADH (c), the change of the oxidation potentials (subtracted from the oxidation potentials at the non-activated electrodes) is plotted. It can be seen that for the ascorbic acid, NADH, and dopamine the oxidation potential gradually lowers with an increasing activation potential up to 1.75 V. Additional increasing of the potentials yielded damage on the MWCNT/PSf composite surfaces and difficulties in the detection of some of the electrochemical probes. Data in Fig. 3, A demonstrate that 1.75 V is the optimal potential for the electrochemical activation of MWCNT/PSf composite electrode for the analytes studied in this work.

Fig. 3 Effect of the applied electrochemical activation potential (A) and of electrochemical activation pretreatment time (B) on ΔE (a, d) and oxidation peak potential (b, c) (vs. non-activated composites) for 5 mM of ferricyanide (a), ascorbic acid (b), NADH (c), and dopamine (d). Conditions: PBS buffer, pH 7.4; scan rate, 100 mV s⁻¹; (A) electrochemical activation time, 180 s; (B) electrochemical activation potential, 1.75 V.

We also studied the influence of electrochemical activation time on electrochemical response of MWCNT/PSf composite electrode. The activation potential was fixed at 1.75 V and the time dependence was studied for the same four analytes as plotted in Fig. 3B(a–d). Activation times of 60, 180, 300, and 360 s of electrochemical activation time were used for the study and their effect on the change of oxidation potential or ΔE was observed. It should be noted that ascorbic acid, NADH, and dopamine experienced a sharp lowering of the oxidation potential 60 s after activation. For longer activation times, the detection potentials of the oxidation peaks and ΔE of the electrochemical probes lowered 300 s after activation. A dramatic decrease in these values was observed for longer times, similar to those observed for the high activation potentials in Fig. 3A. Thus, we have demonstrated that applying an electrochemical activation potential of 1.75 V for 300 s is most favorable for analytes such as ferricyanide, AA, NADH, and dopamine in the context of MWCNT/PSf composite electrodes.

3.3 Electrochemical impedance spectroscopy

The electrochemical resistance of the electrode and its capacitance was studied using electrochemical impedance spectroscopy (EIS). Fig. 4 shows the impedance spectra of a MWCNT/PSf composite electrode for 10 mM ferricyanide in 0.1 mM KCl solution. Our results suggest that a compromise between high capacitance and low resistance to heterogeneous electron transfer should be taken. These results agree with the electrochemical experiments (section 3.2), demonstrating that the best activation condition for the MWCNT/PSf electrode is 1.75 V and a long activation time.

Fig. 4 Impedance spectra for MWCNT/PSf composite for electrochemical activation for 0 s (a), 60 s (b), and 300 (c) s at potential of 1.9 V.

Table S1 in the ESI† summarizes the EIS data obtained after the application of different activation potentials and different activation times for the MWCNT/PSf composite electrodes. The initial capacitance (C_pe) of the electrodes before activation was between 2.1 × 10⁻⁴ and 4.5 × 10⁻⁴ F. These high values indicate that a large surface area is present within the MWCNT/PSf composite. The heterogeneous electron transfer resistance was between 27 and 37 Ω. The C_pe values do not increase significantly after the application of 1.0, 1.5, and 1.6 V, as shown in Table S1,† even after 300 s of activation. However, higher capacitances can be achieved by activating the electrodes using potentials above 1.75 V. The lowest electrode resistance (3.5 Ω) was found for 300 s activation at 1.75 V. It is important to mention that at the electrochemical activation potentials <1.5 V there is negligible effect on activation of MWCNT.^13,18

3.4 Raman spectroscopy

The Raman spectra of the MWCNT/PSf composite were examined before and after electrochemical activation. Polysulfone polymer itself was also studied by Raman spectroscopy. Selected representative Raman spectra are shown in Fig. 5(A–C). In the PSf polymer spectra (A), different sulfur states appeared in the region from 700–1200 cm⁻¹. A peak at 795 cm⁻¹ was found and is attributed to Ar–S–Ar stretch (not shown). The peak at 1152 cm⁻¹ is attributed to the symmetric—SO₃^– stretch of the sulfonate group. Fig. 5B and C compare the Raman spectra of the as-prepared MWCNT/PSf composite (B) and the electrochemically activated composite at 1.75 V for 360 s (C). In both composites, the peaks from the sulfonate groups are shown and high intensity bands are present at 1350 and 1580 cm⁻¹. These bands are commonly attributed to the D band (disorder sp³) and G band (stretching mode in graphene plane) of graphene sheets of MWCNT.¹⁹

Fig. 5 Raman spectra of PSf (A), MWCNT/PSf composite before electrochemical activation (B), and after 360 s at 1.75 V (C). (D) D/G ratio for different electrochemical activation potentials and times; n = 3.

A sharp increase in the D-band is observed when electrochemical activation is applied to the composite. This increase is due to the introduction of “edge-plane-like” defects to the outer graphene sheet of MWCNT.¹³ The D/G intensity ratio increased from 0.7 for the as-prepared composite to 1.1 for the electrochemically activated composite after 360 s, as shown in Fig. 5D. There was no significant increase in the D/G ratio when the MWCNT/PSf composite electrodes were exposed to electrochemical activation for 60 s showing a very small increase from 0.70 to 0.75 for 1.75 V, suggesting that at such a short electrochemical activation time the damage to the sp² lattice of MWCNT is limited. It is possible to calculate crystallite size (L_a) of sp² lattice using equation:²⁰

where λ_laser is the wavelength of laser in nm units and G and D intensities of G- and D-bands. Based on eqn (1), we calculated L_a for MWCNT for various electrochemical activation potentials and times. Crystallite sizes are of 24.0, 24.0, 22.4, 22.4 23.0 and 23.0 nm for electrochemical activation potentials of 0, 1.5, 1.6, 1.75, 1.9 and 2 V for electrochemical activation time of 60 s. L_a are of 24.0, 20.8, 19.8, 16.0 nm for electrochemical activation potentials of 1.0, 1.5, 1.6 and 1.75 V for electrochemical activation time of 300 s. These data are in agreement with the electrochemical and impedance results as with the TEM observations of the MWCNT/PSf composite.

4. Conclusions

We have demonstrated for the first time the effect of electrochemical activation of carbon nanotube/polymer composite electrodes. We showed that the electrochemical activity of MWCNT/PSf electrodes could be tuned between not activated and fully activated. We proved that lowering the resistance to heterogeneous electron transfer is primarily caused by partial removal of the polymer wrap around the MWCNT during harsh electrochemical activation. We also demonstrated that, in part, this is due to the introduction of defects to the MWCNT walls. Our findings will have a broad impact on the fabrication of CNT/polymer composite electrodes for sensing and energy storage applications. Such tunable activation of electrodes would allow on-demand activation of sensing and biosensing devices. In addition, tunable activation of fuel cells would enable on-demand power generation, satisfying the specific needs of power consuming devices.

Acknowledgements

This research was supported by NIMS start-up fund and by Japanese Ministry for Education, Culture, Sports, Science and Technology (MEXT) through MANA program (M.P.), by the research program CTQ2006-15681-C0 from the Spanish Ministry of Education and Science (S.S. and E.F.) and by a BE-2007 grant from the Generalitat de Catalunya, Spain (S.S.).

References

1. P. M. Ajayan and J. M. Tour, Nature, 2007, 447, 1066.
2. A. Merkoçi, M. Pumera, X. Llopis, B. Perez, M. del Valle and S. Alegret, Trends Anal. Chem., 2005, 24, 826.
3. G. A. Rivas, M. D. Rubianes, M. C. Rodríguez, N. F. Ferreyra, G. L. Luque, M. L. Pedano, S. A. Miscoria and C. Parrado, Talanta, 2007, 74, 275.
4. M. Pumera, S. Sánchez, I. Ichinose and J. Tang, Sens. Actuators, B, 2007, 123, 1195.
5. P. G. Bruce, B. Scrosati and J.-M. Tarascon, Angew. Chem., Int. Ed., 2008, 47, 2.
6. R. J. Bowling, R. T. Packard and R. L. McCreery, J. Am. Chem. Soc., 1989, 111, 1217.
7. C. A. Goss, J. C. Brumfield, E. A. Irene and R. W. Murray, Anal. Chem., 1993, 65, 1378.
8. A. A. Gewirth and A. J. Bard, J. Phys. Chem., 1988, 92, 5563.
9. K. S. Prasad, G. Muthuraman and J.-N. Zen, Electrochem. Commun., 2008, 10, 559.
10. J. Wang, M. Li, Z. Shi, N. Li and Z. Gu, Anal. Chem., 2002, 74, 1993.
11. M. Musameh, N. S. Lawrence and J. Wang, Electrochem. Commun., 2005, 7, 14.
12. C.-M. Liu, H.-B. Cao, Y. P. Li, H. B. Xu and Y. Zhang, Carbon, 2006, 44, 2919.
13. M. Pumera, T. Sasaki and H. Iwai, Chem. Asian J., 2008 DOI:10.1002/asia.200800218.
14. M. Pumera, Langmuir, 2007, 23, 6453.
15. T. Kolodiazhnyi and M. Pumera, Small, 2008, 4, 1476.
16. S. Sanchez, M. Pumera, E. Cabruja and E. Fabregas, Analyst, 2007, 132, 142.
17. S. Sanchez, M. Roldan, S. Perez and E. Fabregas, Anal. Chem., 2008, 80, 6508.
18. T. Ito, L. Sun and R. M. Crooks, Electrochem. Solid-State Lett., 2003, 6, C4.
19. M. A. Pimenta, G. Dresselhaus, M. S. Dresselhaus, L. G. Cançado, A. Jorio and R. Saito, Phys. Chem. Chem. Phys., 2007, 9, 1276.
20. L. Cancado, K. Takai, T. Enoki, M. Endo, Y. A. Kim, H. Mizusaki, A. Jorio, L. N. Coelho, R. Magalhaes-Paniago and M. A. Pimenta, Appl. Phys. Lett., 2006, 88, 163106.

---

Footnote

† Electronic supplementary information (ESI) available: Summarized EIS data obtained after the application of different activation potentials and different activation times for the MWCNT/PSf composite electrodes. See DOI: 10.1039/b814599g

---