Synthesis and characterization of p-toluenesulfonate incorporated poly(3,4-ethylenedioxythiophene)

Talanta 72 (2007) 532–538

Synthesis and characterization of p-toluenesulfonate incorporated poly(3,4-ethylenedioxythiophene) Yinghong Xiao a,b,c , Chang Ming Li a,b,∗ , Shucong Yu a , Qin Zhou a,b , Vee. S. Lee d , Shabbir. M. Moochhala d a

School of Chemical and Biomedical Engineering, Nanyang Technological University, Singapore 637457, Singapore b Center for Advanced Bionanosystems, Nanyang Technological University, Singapore 637457, Singapore c School of Chemical Engineering, Nanjing University of Science and Technology, Nanjing 210014, China d Defence Medical & Environmental Research Institute, DSO National Laboratories, Singapore 117510, Singapore Received 30 October 2006; received in revised form 8 November 2006; accepted 8 November 2006 Available online 8 December 2006

Abstract Poly(3,4-ethylenedioxythiophene) (PEDOT), a conducting polymer, was electrochemically synthesized with p-toluenesulfonate (TSNa) as a dopant on gold surface. The electrochemical properties of the polymer were studied by impedance spectroscopy and cyclic voltammetry (CV). It was found that the impedance magnitude of the electrode significantly decreased over a wide range of frequency from 100 to 104 Hz after the polymer deposition. The CV demonstrated enhanced reversibility of the PEDOT film. The surface morphology was investigated by scanning electronic microscope (SEM) and atomic force microscope (AFM). Due to the effect of TSNa structure, nano-fungus was observed. Polymerization time was optimized and 30 min deposition resulted in the highest charge capacity, showing the highest electroactive surface area, possibly due to its porous structured polymer. Moreover, the high specific surface area could be favorable for cell attachment. The stability of PEDOT in glutathione (GSH), a common biologically relevant reducing agent, was studied with polypyrrole (PPy) as a baseline. It showed that the former had much better stability than the latter and it could be an excellent candidate for potential applications of in vivo neural devices. © 2006 Elsevier B.V. All rights reserved. Keywords: PEDOT; Electrochemical properties; Surface morphology; Stability against reducing agents

1. Introduction Since the highly conductive polyacetylene was discovered in the late 1970s [1], conducting polymers have been intensively studied and successfully used in various areas such as sensors and actuators [2–4], batteries [5], antistatic coatings for photographic films [6], processing of electronic circuit boards [7]. Conducting polymers show interesting chemical and physical properties derived from their unique conjugated ␲-electron system [8]. Amongst these polymers, polypyrrole (PPy) is often chosen for biological applications due to its ease of preparation, good conductivity and biocompatibility [9–11]. PPy was applied to modify implantable devices for neural recording and drug delivery [12,13]. However, applications of

∗

Corresponding author. Tel.: +65 67904485; fax: +65 67911761. E-mail address: [email protected] (C.M. Li).

0039-9140/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.talanta.2006.11.017

PPy suffer from its instability in biological environment due to its structural disorder. Oxidized PPy is particularly unstable in biologically relevant reducing agents such as dithiothreitol (DTT) and glutathione (GSH). It is apparently disadvantageous to PPy modified electrodes which are implanted in brain tissues for long term use. Thiophene is difficult to polymerize electrochemically due to its high oxidation potential in aqueous media [14]. Recently a derivative of polythiophene, poly(3,4ethylenedioxythiophene) (PEDOT), which could be easily electrochemically synthesized has aroused great interest of material scientists. PEDOT has been classified as a low band gap conducting polymer. The 3,4-dioxy substitution pattern blocks the possibility of ␣-␤(␤ ) coupling normally presented in PPy (Scheme 1, left), which can result in a more regiochemically defined material (Scheme 1, right), adds electron density to the aromatic heterocycle, and reduces the monomer oxidation potential. Its high thermal stability has been reported [15]. Additionally, the decrease of the polymer reduction potential would

Y. Xiao et al. / Talanta 72 (2007) 532–538

Scheme 1. Chemical structures of EDOT monomer (left) and PEDOT polymer (right).

533

SEM (JEOL JSM-6700F FEG, USA) was used to study the surface morphology of the resultant PEDOT/TSNa film with a voltage of 5.0 kV and spot size of 4 (arbitrary units). Surface topography of the PEDOT/TSNa film was investigated with AFM (Dimension 3100 SPM, Veeco, USA) and high resolution surface images were produced. In AFM characterization, the tapping mode with a silicon probe (RTESP, Veeco, USA) over scan sizes of 15 ␮m and the scan rate of 0.20 Hz was used. 2.2. Materials

yield an increased stability in the oxidized conducting form [16]. The superior properties of PEDOT discussed above render it favorable as an electrode material in implantable devices. Implantable electronic systems in the brain tissues particularly require low impedance of the electrode, high charge capacity, high surface area and stability to the reducing agents in the brain tissue environment. There is a great need to provide a stable electrode material that can not only facilitate electron transfer between the neurons and electrode, but can also attach neural cells onto the electrode sites. In our study, PEDOT was electrosynthesized with ptoluenesulfonate (TSNa) as the dopant onto the gold electrode. TSNa was selected since its molecules are large enough to occupy free spaces in the polymer while small enough to dedope from the polymer matrix. The feature provides the possibility to incorporate bulky negatively charged biomolecules by subsequent ion exchange [17] for implantable devices. TSNa could be replaced by the biomolecules in the ion exchange process to eliminate the cytotoxicity of the devices. Thus, the synthesized PEDOT/TSNa films in this work were optimized and characterized for such a potential application. Thomas et al. reported that 3,4-dioxy substituted PPy showed better stability against reducing agents than PPy in DTT solution [16]. For the first time we studied the stability of PEDOT against a typical biological reducing species, GSH, and proposed its stability enhancement mechanisms.

EDOT monomer was purchased from Aldrich, purified via vacuum distillation and kept refrigerated at 4 ◦ C under nitrogen before use. TSNa (95%) from Sigma–Aldrich was of analytical grade and was used as received. l-Glutathione (reduced, 98%) was purchased from Aldrich. Deionized Milli-Q water (18.2 M cm, Millipore Inc.) was used in all experiments. The gold electrodes (2 mm in diameter) were sequentially polished with 1.0, 0.3 and 0.05 ␮m Al2 O3 slurry, then washed with acetone and isopropyl alcohol followed by thorough rinsing with deionized water. All experiments were performed at room temperature, unless otherwise stated. 2.3. Electrosynthesis and stability test Different electrosynthesis methods were discussed in [18]. Preparation of PEDOT in this report was carried out by the potentiostatic method in 0.01 M EDOT + 0.1 M TSNa solution. Before polymerization, the electrochemical cell was ultrasonically cleaned in water and the monomer solution was purged with N2 for 20 min. All electrochemical polymerizations were performed with CHI 760B potentiostat (USA). The polymer stability against the reducing agent was studied by immersing the PEDOT coated electrode in 10 mM GSH aqueous solution. The samples were removed from the solution followed by CV measurements at 1-day intervals.

2. Experimental

3. Results and discussion

2.1. Instrumentation

3.1. Electrodeposition of PEDOT

All controlled-potential electrochemical syntheses were performed with CHI 760B Electrochemical Station (USA). A three-electrode cell was set up for all electrochemical experiments, in which a gold disk electrode, a platinum wire, Ag/AgCl (saturated KCl) were used as working, counter and reference electrode, respectively. Electrochemical impedance spectroscopy and cyclic voltammetry (CV) were performed using the same instrument as for the electrosynthesis and the same cell set-up was employed. 0.01 M PBS (pH 7.4) was used as the electrolyte. During impedance measurements, 5.0 mV ac sinusoid signal was applied as the input perturbation and dc bias potential was set at 0.0 V. The impedance measurements were carried out over 100 –104 Hz. The CV was conducted with a scan rate of 100 mV/s and scan potential range over −0.8 to 0.4 V.

EDOT monomer has poor solubility (0.01 M at room temperature) in water. The solubility can be increased by mixtures of water and some organic solvents [19]. However, water is preferred in our study due to the potential bio-applications of the polymer film. We found that TSNa, as a surfactant, could not cause significant increase of EDOT solubility in water as sodium dodecylsulfate (SDS) or sodium dodecylbenzenesulfonate (SDBS) did [20]. A suitable potential for the potentiostatic synthesis of PEDOT is critical, since the conducting polymers can be over-oxidized at higher potentials, leading to its deterioration and even destruction. The linear sweep voltammetry (LSV) was conducted to determine the suitable polymerization potential. Results in Fig. 1 show two oxidation peaks of PEDOT in a potential range from 0.2 to 1.6 V. As discussed in [21], the

534

Y. Xiao et al. / Talanta 72 (2007) 532–538

Fig. 1. LSV of 0.01 M EDOT in 0.1 M TSNa aqueous solution. Fig. 3. Relationship between magnitude at 1 kHz and deposition time.

first poorly defined peak at about 0.96 V corresponded to the oxidation of the monomer whereas the second one at 1.47 V was attributed to the over-oxidation of the deposited polymer. Apparently, the suitable potential for PEDOT synthesis should be higher than 0.96 V and lower than 1.47 V. In our work, 1.0 V was chosen as the deposition potential. 3.2. Electrochemical behavior of PEDOT/TSNa PEDOT was potentiostatically synthesized with different polymerization times. It was clearly observed that a sky-blue thin film formed on the gold electrode surface within several seconds. Impedance magnitude was measured on both bare and PEDOT coated electrodes. As shown in Fig. 2, the impedance magnitude of all PEDOT-coated electrodes was significantly reduced over 100 –104 Hz in comparison to that of the bare gold. At lower frequency range (1–10 Hz) capacitance was the prime component for the impedance while at frequencies above 30 Hz, including the frequencies of most neural activity (300–1000 Hz), the dominant impedance changed from capacitive to resistive. Since the resistance reduction of the electrode is essential to produce high-quality signals including high S/N and superior sensitivity

Fig. 2. Impedance spectroscopy of bare gold (a) and PEDOT coated electrodes (b–f). Deposition time was 5 min (b), 10 min (c), 20 min (d), 30 min (e) and 60 min (f).

[22] for in vivo measurements of neuron activity, apparently, the PEDOT electrode could be potentially used in miniaturized implantable devices. Since 1 kHz is the frequency often used for measurements of the biologically relevant activity, relationship of the impedance magnitude at 1 kHz versus PEDOT deposited time was studied. The results are shown in Fig. 3. It was observed that the impedance magnitude had a relatively sharp decrease initially with the deposition time, indicating a thin layer formation on the gold surface increased greatly the true electrode surface area due to the porous and very conductive polymer. However, with the further increment of deposition time, the film impedance magnitude became relatively stable even with slight increase. When the deposition time was increased to more than 60 min, the impedance sharply increased even worse than that of the bare gold. CV characteristics of a conducting polymer could show its reversibility, which directly affect its switching ability in electronic measurements. CVs of PEDOT films prepared by using different deposition times were measured in PBS (Fig. 4). Up to 30 min of deposition time, the enclosed area of CV curves increased with the deposition time, and the CV curves showed better reversible, symmetric redox waves (Fig. 4A). The shape of the curve became more symmetric and the redox waves were getting better defined, indicating the doping redox reaction was more reversible for electron or charge transfer in the film. However, when the time was more than 30 min, the area sharply became smaller (Fig. 4B) and the redox waves became less prominent. In order to show the reversibility and stability of the film, multiple CV measurements were conducted on the 30 min deposited film. The results are shown in Fig. 4C. It was observed that the enclosed area was reduced for the first three cycles and then became stabilized up to 200th cycles. The reduction of the enclosed area during the first few cycles of reversible reactions is common for electrodes even inert ones such as Au and Pt due to surface stabilization. However, all CV curves for the PEDOT showed well defined, reversible redox waves, demonstrating good reversibility. The enclosed area of CV is proportional to the charge capacity of

Y. Xiao et al. / Talanta 72 (2007) 532–538

535

Fig. 4. Cyclic voltammograms of PEDOT films, which were prepared with different synthesis times as indicated in the figure (A and B). CV of bare gold was plotted in the figure as a negative control. Electrochemical reversibility of the 30 min deposited film was demonstrated by multiple CV sweeping (C).

the electrode, which was determined by electrochemical doping/dedoping process + double layer capacitance of the PEDOT film. Fig. 5 shows the change of charge capacity versus the deposition time. Clearly, within 30 min, the charge capacity increased because both electrochemically active doping sites and true surface area increased with the film growth. Fig. 3 shows that the resistance did not increase significantly for the 40 and 60 min deposited PEDOT, but their charge capacity sharply dropped and no doping/dedoping redox waves were observed. The results might indicate that further deposited polymer changed the film morphology and porous structure, leading to decrease of real surface area and deactivation of doping sites. The explana-

Fig. 5. Relationship between the film charge capacity and deposition time.

tion is discussed in more detail later by the SEM and AFM results. 3.3. Morphologies of PEDOT/TSNa films SEM characterization was conducted on different timedeposited PEDOT films. As shown in Fig. 6, the morphology of the films exhibited a nano-fungus-like structure, which has not been reported up to date. A possible mechanism for the formation of such morphology is that the dopant of TSNa is a surfactant and existed in the form of lamella at the concentration of 0.1 M applied in our experiments [23]. The polymer could grow with the pseudo-molecular template and thus to form the fungus-like shapes. The rough surface shown in Fig. 6 will evidently provide more sites for attachment of neural cells and enhance adhesion between PEDOT and neurons. From image (a) to (f) we clearly observed that the density of the nanofungus was getting higher and higher and finally the adjacent fungi merged together. The surface roughness, i.e. the specific surface area increased until the maximum (d) followed by a decrease with deposition time (from (e) to (f)). The large fungus in (e) and (f) looked like a sintered, melted material and lost its porous structure shown in (d). This might indicate that the deposition for more than 30 min not only grew the film thickness, but also filled the pores. A typical nano-fungus structure can be clearly observed from the lower high magnification SEM image. It is known that the high specific area is mainly contributed by the micropore but not macropore structure. Obviously, the

536

Y. Xiao et al. / Talanta 72 (2007) 532–538

Fig. 6. SEM images of PEDOT film with different deposition time. (a) 5 min, (b) 10 min, (c) 20 min, (d) 30 min, (e) 60 min and (f) 120 min. Lower: high magnification image of (d).

structure in (e) and (f) significantly reduced the specific surface area. This is one of the main reasons to cause the dramatic decrease of the charge capacity in Fig. 5. It demonstrates that the deposition time is very critical to obtain a superior PEDOT film. Three PEDOT samples with different deposition times which showed typical different morphology were characterized by AFM. The results are shown in Fig. 7. It could be seen that the topographies in (b)–(d) had very notable variations from that of the smooth bare gold (a). From (b) to (c), the surface became rougher and size of the clusters was getting more uniform as the film grew thicker. The clusters coalesced as the film continued to

grow (d). The lower 3D images clearly show that the roughness increased with deposition time until its maximum (c) followed by a decrease when some micropores were filled with the material. This might not only decrease the specific surface area, but also could block the doping site. The results are in agreement with the SEM images. 3.4. Stability of PEDOT compared with PPy against biological reducing agent It is well known that lower reduction potential of a conducting oxidized polymer (p-type polymer) can give higher

Y. Xiao et al. / Talanta 72 (2007) 532–538

537

Fig. 7. AFM images of PEDOT coated electrode with deposition time of (a) 0 min, (b) 15 min, (c) 30 min and (d) 60 min. Lower: three-dimension images.

stability against biological reducing agents. Lower reduction potential of the p-type polymer can eliminate reactions of the reducing agents and oxidized form which would result in loss of conductivity of the polymer. The result of PEDOT stability in reducing agent solutions is shown in Fig. 8. As

control, a film of PPy/TSNa was potentiostatically electrosynthesized at 0.8 V and exposed to the same reducing agent. In comparison of the reduction potential for PPy (reduction potential = −0.32 V), PEDOT had a much lower reduction potential (−0.56 V), indicating that the latter had better stability against the reducing agents than the former. The reduction potentials of both PPy (a) and PEDOT (b) positively shifted with immersing in GSH, a common reducing agent in biological system. However, the deteriorate rate of the reduction potential for PEDOT was only about 5 mV/day (about 50 mV shift for 10 days), whereas the deteriorate rate of reduction potential for PPy was about 20 mV/day. This showed that the positive shift rate of the reduction potential for PEDOT was much lower than that for PPy. The results indicate that PEDOT coated electrode is favorable for long-term applications as an implantable device. 4. Conclusion

Fig. 8. Comparison of the stability of PPy (a) and PEDOT (b) against GSH.

Electrosynthesis and characterization of PEDOT on gold was conducted. The suitable potential for the polymerization was experimentally determined to be 1.0 V to prevent

538

Y. Xiao et al. / Talanta 72 (2007) 532–538

the polymer from over-oxidation. The PEDOT film greatly decreased the impedance of the electrode, which is favorable for obtaining high-quality signals in an electronic device in a biological system. At the characteristic frequencies of neural biological activity (300–1000 Hz), the dominant factor of the PEDOT impedance changed from capacitive to more resistive. The PEDOT electrode was most reversible and had highest charge capacity when it was formed with 30 min deposition time. The electrochemical properties correlated well with the surface morphology variation during the polymerization process observed by SEM and AFM. The change of the surface area and morphology plays a critical role in the electrochemical and electronic properties of the film. The results demonstrated that there was an optimized deposition time to prepare the superior PEDOT electrode. With the potentiostatic deposition method, our experimental results demonstrated that the 30 min deposited polymer had the best electrochemical reversibility, highest charge capacity and lowest resistance. This work also demonstrated that the PEDOT stability against reducing agents was significantly better than that of PPy. Thus, PEDOT could be a good candidate to apply in the implantable neural devices for long-term operation. The cell attachment and the biocompatibility of PEDOT/TSNa electrode are under investigation in our lab for potential applications. Acknowledgments The work is financially supported by Singapore DSO under contract grant number of DSOCL05047 and Center of Advanced Bionanosystems, Nanyang Technological University. Dr. Yinghong Xiao would like to thank the support from National Science Foundation of China (Contract no. 50373019).

References [1] H. Shirakawa, E.J. Lewis, A.G. MacDiarmid, C.K. Chiang, A.J. Heeger, J. Chem. Soc. Chem. Commun. (1977) 578. [2] C.M. Li, W. Chen, X. Yang, C.Q. Sun, C. Gao, Z.X. Zheng, J. Sawyer, Front. Biosci. 10 (2005) 2518. [3] J. Jang, M. Chang, H. Yoon, Adv. Mater. 17 (2005) 1616. [4] Y. Berdichevsky, Y.-H. Lo, Adv. Mater. 18 (2006) 122. [5] J.H. Burroughs, D.D.C. Bradley, A.R. Brown, R.N. Marks, K. Mackay, R.H. Friend, P.L. Burns, A.B. Holmes, Nature 347 (1990) 539. [6] L. Groenendaal, F. Jonas, D. Freitag, H. Pielartzik, J.R. Reynolds, Adv. Mater. 12 (2000) 481. [7] Z.N. Bao, Adv. Mater. 12 (2000) 227. [8] A.J. Heeger, Angew. Chem. Int. Ed. 40 (2001) 2591. [9] G. Shi, M. Rouabhia, Z. Wang, L.H. Dao, Z. Zhang, Biomaterials 25 (2004) 2477. [10] C.E. Schimidt, V.R. Shastri, J.P. Vacanti, R. Langer, Proc. Natl. Acad. Sci. U.S.A. 94 (1997) 8948. [11] N.A. Peppas, R. Langer, Science 263 (1994) 1715. [12] P.M. George, D.A. LaVan, J.A. Burdick, C.-Y. Chen, E. Liang, R. Langer, Adv. Mater. 18 (2006) 577. [13] D.-H. Kim, M. Abidian, D.C. Martin, J. Biomed. Mater. Res. 71A (2004) 577. [14] L.B. Groenendaal, G. Zotti, P.-H. Aubert, S.M. Waybright, J.R. Reynolds, Adv. Mater. 15 (2003) 855. [15] H. Yamato, M. Ohwa, W. Wernet, J. Electroanal. Chem. 397 (1995) 163. [16] C.A. Thomas, K. Zong, P. Schottland, J.R. Reynolds, Adv. Mater. 12 (2000) 222. [17] Y.H. Xiao, J.F. Che, C.M. Li, C.Q. Sun, Y.T. Chua, V.S. Lee, J.H.T. Luong, J. Biomed. Mater. Res. A 80A (2006) 925. [18] C.M. Li, C.Q. Sun, W. Chen, L. Pan, Surf. Coat. Tech. 198 (2005) 474. [19] M. Ilic, F.-H. Haegel, S. Zlatanovic, N. Potkonjak, B. Simonovic, Environ. Chem. Lett. 3 (2005) 82. [20] A. Lima, P. Schottland, S. Sadki, C. Chevrot, Synthetic Met. 93 (1998) 33. [21] X. Du, Z. Wang, Electrochim. Acta 48 (2003) 1713. [22] D.R. Kipke, R.J. Vetter, J.C. Willianms, J.F. Hetke, IEEE T Neur. Sys. Reh. 11 (2003) 151. [23] M.J. Rosen, Micelle Formation by Surfactants, third ed., John Wiley & Sons, Inc., New York, 2004 (Chapter 3).