- Email: [email protected]
Preparation and characterization of a graphite electrode containing carbon nanotubes grown in situ by flame synthesis
Electrochimica Acta 56 (2011) 5205–5209
Contents lists available at ScienceDirect
Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta
Preparation and characterization of a graphite electrode containing carbon nanotubes grown in situ by flame synthesis Qi Jiang a,∗ , Rong Yang a , Zhengwen He a , Zhao Liu a , Deyu Xie a , Yong Zhao a,b a Key Laboratory of Advanced Technologies of Materials (Ministry of Education of China) and Superconductivity R&D Centre, Southwest Jiaotong University, Chengdu 610031, PR China b School of Materials Science and Engineering, University of New South Wales, Sydney, 2052 NSW, Australia
a r t i c l e
i n f o
Article history: Received 4 February 2011 Received in revised form 10 March 2011 Accepted 11 March 2011 Available online 21 March 2011 Keywords: Carbon nanotubes Chemically modified electrode Flame synthesis
a b s t r a c t A crystalline flake graphite electrode (GE) was impregnated with nickel particles using direct current electrochemical deposition. The particles were used for in situ growth of carbon nanotubes (CNTs) by flame synthesis with a liquid ethanol flame. The obtained electrode was characterized by X-ray diffraction, and scanning and transmission electron microscopy. The results showed that the deposited Ni catalyst crystal face was mainly (1 1 1). CNTs with a diameter of about 40 nm were uniformly grown on the GE surface. The electrochemical performance of the CNT–GE was characterized by cyclic voltammetry using a [Fe(CN)6]3− /[Fe(CN)6]4− solution, and showed a much greater electrochemical response than that obtained using a material in which CNTs were grown by catalytic chemical vapor deposition. © 2011 Elsevier Ltd. All rights reserved.
1. Introduction Carbon nanotubes (CNTs) have attracted extensive attention in many science and technology fields since they were discovered [1,2]. Due to their unique structure and physical properties, CNTs have been considered as one of the best candidate for fabricating nano-electronic devices [3], composite materials [4,5], gas storage materials [6] and catalyst support materials [7]. Regular methods such as arc-discharge [8], laser ablation [9], and catalytic chemical vapor deposition (CCVD) [10] have been successfully used to synthesize CNTs. Flame synthesis is a new developing method to grow CNTs compared with the methods mentioned above [11–15], and its prominent advantages are energy saving up and simple operation. In flame synthesis, some fuel provides heat while the remainder serves as carbon source for CNT growth. Recently, liquid ethanol had been developed as carbon source for CNT growth by a common ethanol burner [16,17]. In comparison with the regular methods mentioned above, flame synthesis was much simpler and more cost-effective. In electrochemical field, CNT novel characteristics including high aspect ratio, nano-scale dimensions and good electrical conductivity have indicated that CNTs are a type of potential modified materials for chemically modified electrode (CME) [18,19]. According to the literatures, five ways were developed to prepare
∗ Corresponding author at: Superconductivity R&D Centre, Southwest Jiaotong University, 111, No. 1, North, 2 Ring Road, Chengdu 610031, PR China. Tel.: +86 28 87603544; fax: +86 28 87603544. E-mail addresses: [email protected], [email protected] (Q. Jiang). 0013-4686/$ – see front matter © 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.electacta.2011.03.038
CNT–CME [20–24], which are listed as follows: The first is mixing the CNTs, bromoform, mineral oil and liquid paraffin carefully and stuffing the mixture into the glass capillary to prepare electrode. The next is filling the CNTs into the corration platinum electrode to prepare CNT powder microelectrode. The third is embedding the CNTs into the graphite electrode (GE) to prepare embedding modified electrode. The fourth is “abrasive immobilization” (simply rubbing the CNTs onto an electrode surface to prepare electrode). The last is dispersing the CNTs into the H2 SO4 solution, H2 O solvent, dimethyl formamide solvent and coating directly on the surface of the electrode to prepare CNT–CME. Following these ways, CNT–CMEs were done as two steps: firstly, CNTs were prepared by the CNT growth methods mentioned above and then the CNTs were fixed on the electrode surface by some physical ways, for example, using binder. However, the binder could reduce the CNT utilization efficiency by enhancing the resistance or changing the CNT novel nanometer hollow tube structure. In order to avoid these defects, we have introduced a new method to prepare CNT–CME by growing CNTs in situ (named as GSCNT–CME) in our previous work [25]. The GSCNT–CME was obtained by directly growing CNT on a crystalline flake graphite electrode (GE) surface, namely one-step method. The obtained GSCNT–CME can hold the CNT pristine nanometer structure and has showed excellent potential application value. In that report, CNTs were grown by CCVD and CNT catalyst on the electrode was prepared by immersion. However, the operation is relative complicated and inefficient (the operation process is over 4 h and in vacuum). In order to enhance the GSCNT–CME preparation process practicability and maneuverability, flame synthesis and catalyst direct current electrochemical deposition were used to replace CCVD and
5206
Q. Jiang et al. / Electrochimica Acta 56 (2011) 5205–5209
catalyst immersion in this paper. And a liquid ethanol was used as the carbon resource. With these improvements, the GSCNT–CME can be prepared in a few minutes and without vacuum condition. Therefore, the operation is much simpler and safer. The CNTs with a diameter of about 40 nm are uniformly grown on the GE surface. All these show that the way used here is of much more practicability and maneuverability. 2. Experimental 2.1. Electrochemical deposition nickel catalyst To prepare Ni catalyst, a galvanostatic – voltage instrument (ZF9, Shanghai, China) was used. Ni slice (99.99%) with 1 cm × 3 cm size was used as the anode. A crystalline flake graphite disk (5.3 mm in diameter, 7.0 mm in length) was used as the electrode. The graphite disk was polished to a mirror surface with the metallographic abrasive paper (W 7-05). Then, it was sealed by fixing into a teflon housing with the polished surface out. The obtained electrode was named as GE. The electrochemical deposition solution was composed of nickel sulfate (NiSO4 ·7H2 O, 300 g L−1 ), nickel chloride (NiCl2 ·6H2 O, 45 g L−1 ), boric acid (H3 BO3 , 40 g L−1 ), saccharin sodium (C6 H4 COSO2 NNa·2H2 O, 5 g L−1 ), sodium dodecyl sulfate (CH3 (CH2 )11 OSO3 Na, 0.1 g L−1 ) and redistilled water. The direct current electrochemical deposition was carried out for 1.5 min using a current of 20 mA. The deposited Ni catalyst GE was named as Ni/GE and used to grow CNT. 2.2. GSCNT–CME preparation A liquid anhydrous ethanol flame was used to grow CNT. The flame height and middle width were about 20 and 60 mm, respectively. The wick height was about 10 mm. In order to grow CNTs, the Ni/GE was placed about 8 mm over the wick for 4 min, where the flame color was blue and the temperature was about 610 ◦ C (tested by an electronic sensor). After that, a layer of black sample appeared on the Ni/GE surface. Then the obtained electrode was treated by concentrated sulfuric acid, supersonic and redistilled water for removing the impurities (including amorphous carbon, catalyst and so on). As this, the GSCNT–CME was prepared. 2.3. Morphology and microstructure characterization The GE and the Ni/GE surfaces were characterized by scanning electron microscopy (SEM, QUANTA 200). The crystal structure of the deposited Ni was characterized by X-ray diffraction (XRD) (XRD, PWI 3040/60). The morphology and microstructure of the CNTs were also characterized by SEM and transmission electron microscopy (TEM, HITACHI H-700H). 2.4. Electrochemical performance testing Cyclic voltammetry (CV) with three-electrode-system was used to test the obtained GSCNT–CME electrochemical performance. A platinum wire was used as the counter electrode, the saturated calomel electrode (SCE) was used as the reference electrode and the obtained GSCNT–CME was used as the working electrode in the testing system. 3. Results and discussion 3.1. SEM, XRD and TEM study Fig. 1 shows SEM images of the GE surface after being polished (a and b) and the Ni/GE surface after the electrochemical deposition
Fig. 1. SEM images of GE before (a and b) and after (c) the catalyst deposition.
(c). It can be seen from Fig. 1a that the GE surface after being polished is very smooth and flat. Fig. 1b shows SEM images with higher amplification (3 times higher) of the GE after being polished. Fig. 1c shows SEM images at the same amplification (30,000) of the Ni/GE surface after electrochemical deposition. Comparing Fig. 1b with c, it is found that there are lots of grains in Fig. 1c, and the average grain diameter is about 35 nm. These grains are the result of direct current electrochemical deposition. In order to characterize the grain component and structure, XRD analysis was used.
Q. Jiang et al. / Electrochimica Acta 56 (2011) 5205–5209
5207
Fig. 2. XRD patterns of the Ni/GE.
Fig. 2 shows XRD patterns of the obtained Ni/GE. It can be seen from this figure that there are three distinct peaks from 35◦ to 60◦ . According to the XRD PDF database with the XRD scanning program, we can have the peak type and crystal face data. There are (1 0 0) crystal face peak of carbon (42.439◦ ), (1 1 1) crystal face peak of Ni (44.574◦ ) and (0 0 4) crystal face peak of carbon (54.571◦ ). The carbon peaks come from GE. Therefore, it can be deduced that Ni nanometer particle introduction is the result of direct current electrochemical deposition. Moreover, the Ni particle is mainly the (1 1 1) crystal face. The average diameter of the deposited Ni is about 34 nm calculated by the SCHERRER equation, which is the same as the SEM result mentioned above. According to our previous reports [26,27], Ni with nanometer size can be used as the catalyst to grow CNTs and Ni (1 1 1) crystal face is even better for the CNT growth. Therefore, the obtained Ni/GE can be directly used to grow CNTs and the method (direct current electrochemical deposition) is suitable for preparing CNT catalyst. Fig. 3 shows SEM images (a and b) and TEM images (c) of the CNTs grown on the Ni/GE by the flame synthesis. From Fig. 3a and b, we can see that the CNTs are dense, clean (without other carbon impurities) and evenly grown on the GE surface. Moreover, the CNT diameter is uniform (about 40 nm) and the length is 1–3 m. In order to study the CNT structure, the CNTs were removed from the electrode surface and characterized by TEM (Fig. 3c). A nanometer hollow structure of the prepared CNTs can be clearly seen from Fig. 3c. The CNT diameter is very uniform and about 40 nm, which is consistent with the Ni catalyst diameter [28]. These phenomena also indicate that the size of the catalyst obtained by the direct current electrochemical deposition is also uniform. 3.2. Electrochemical performance characterization Fig. 3. SEM images (a and b) and TEM images (c) of the CNTs grown on the GE.
In order to characterize the GSCNT–CME electrochemical performance, CV testing with three-electrode-system is used. A [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution is also used to characterize the obtained electrode in this article. Fig. 4 shows CV curves of GE, Ni/GE and GSCNT–CME in 1.0 mol L−1 KCl solution (a) and 1.0 × 10−3 mol L−1 K3 Fe(CN)6 + 1.0 × 10−3 mol L−1 K4 Fe(CN)6 in 1.0 mol L−1 KCl solution (b) at a scanning rate of 10 mV s−1 vs. SCE. From Fig. 4a, it can be seen that the GSCNT–CME CV curve is different from those of the GE and Ni/GE under the same testing conditions. The surrounded area of the GSCNT–CME CV curve is much larger than those of the GE and Ni/GE, indicating that the GSCNT–CME has larger specific surface area. There are only CNTs on the GSCNT–CME surface compared with the Ni/GE, so the differ-
ence of the CV curves is the result of the CNTs. That’s because that the CNTs are one-dimensional nanometer materials, which have large specific surface area. The CV curves of the GE and Ni/GE have a little difference, indicating that Ni deposition has a little effect on the electrochemical performance in our testing scope. The reason is that the amount of the deposited Ni is finite and the effect of Ni on the GE can be considered as enhancing the GE electrode conductivity, we consider. From Fig. 4b, it can be seen that the current responses of the GE and Ni/GE to the [Fe(CN)6 ]3− /[Fe(CN)6 ]4− oxidation–reduction pair are very weak with modifying the voltage from 300 mV
5208
Q. Jiang et al. / Electrochimica Acta 56 (2011) 5205–5209
Table 1 Data of CV testing based on the GSCNT–CME with different scanning rates in the [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution vs. SCE. Scanning rate (mV s−1 )
Square root of scanning rate ((mV s−1 )1/2 ) E (mV) Peak current (10−3 mA)
0.83
1.67
5
10
20
50
100
0.91 22.04 212.7
1.29 29.50 390.1
2.23 70.97 922.6
3.16 104.95 1562.5
4.47 151.93 2643.9
7.07 255.00 4956.3
10 380.01 7812.4
to 900 mV, whereas the GSCNT–CME current response to the [Fe(CN)6 ]3− /[Fe(CN)6 ]4− oxidation–reduction pair is very strong. The oxidation and reduction peaks based on GSCNT–CME appear at 709.3 and 601.4 mV, respectively. So the current response to the [Fe(CN)6 ]3− /[Fe(CN)6 ]4− based on GSCNT–CME is much stronger than those of the GE and Ni/GE under the same testing conditions. In order to further research the obtained GSCNT–CME electrochemical performance, CV testing at different scanning rates (0.83, 1.67, 5, 10, 20, 50, 100 mV s−1 ) was operated in a [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution (1 × 10−3 mol L−1 ). The data of the oxidation peak current (Ipk ) and the redox peak voltage difference (E) at the different scanning rates are listed in Table 1. Table 1 shows the data of CV testing based on the GSCNT–CME with different scanning rates in the [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution. It is evident from the table that the E increases with increasing scanning rate, indicating that this is not a reversible system. From the
Fig. 5. Relationships between the scanning rate and the oxidation peak current of the GSCNT–CME obtained by flame synthesis in this paper with 1 × 10−3 mol L−1 [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution (a) and the reported GSCNT–CME obtained by CCVD in Ref. [25] with 1 × 10−2 mol L−1 [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution (b).
relationship between the scanning rate square root and the oxidation peak current of the GSCNT–CME in the 1 × 10−3 mol L−1 [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution, it can be found that the relationship is excellent linear, and the linear regression equation is Ipk = 211.4w1/2 − 153.0 (w is the scanning rate) with 0.996 of the linear correlation coefficient. Therefore, it is evident that the electrochemical reaction process is a typical diffusion control process. Fig. 5 shows the relationships between the scanning rate and the oxidation peak current based on the GSCNT–CME obtained by flame synthesis in this paper with 1 × 10−3 mol L−1 [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution (a) and reported GSCNT–CME obtained by CCVD in Ref. [25] with 1 × 10−2 mol L−1 [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution (b). From Fig. 5, it can be seen that two GSCNT–CME oxidation peak currents to the [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution both increase with increasing the scanning rates. Moreover, the oxidation peak current based on the GSCNT–CME obtained by flame synthesis in this paper enhances much greater than that of the GSCNT–CME obtained by CCVD in Ref. [25] even if the [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution concentration in this article is 1 × 10−3 mol L−1 , lower than that in Ref. [25] (1 × 10−2 mol L−1 ). The results show that the electrochemical response to [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution of the GSCNT–CME obtained in this article is much greater than that of the GSCNT–CME obtained in Ref. [25].
4. Conclusions
Fig. 4. CV curves of the GE, Ni/GE and GSCNT–CME in the substrate solution ((a) 1.0 mol L−1 KCl solution) and the testing solution ((b) 1.0 × 10−3 mol L−1 K3 Fe(CN)6 + 1.0 × 10−3 mol L−1 K4 Fe(CN)6 in 1.0 mol L−1 KCl solution) with the scanning rate of 10 mV s−1 vs. SCE.
Several conclusions can be drawn from the experimental results and discussion mentioned above, including: Firstly, the nickel catalyst with grain size from 30 nm to 40 nm can be deposited on the GE surface by the direct current electrochemical deposition in 1.5 min. The deposited nickel can be directly used as the catalyst to grow CNTs.
Q. Jiang et al. / Electrochimica Acta 56 (2011) 5205–5209
Secondly, CNTs with the diameter of about 40 nm can be grown on the GE surface in 4 min without vacuum condition by flame synthesis. GSCNT–CME can be obtained by this simple, economic and safe method, namely depositing nickel catalyst on a GE surface by direct current electrochemical deposition and then growing CNTs by flame synthesis with the liquid ethanol burner flame. Thirdly, the obtained GSCNT–CME electrochemical performance testing has been carried out. The researching results show that the obtained GSCNT–CME has better electrochemical performance than that of the GSCNT–CME obtained by CCVD in Ref. [25], indicating that the flame synthesis is a more potential and suitable for preparing GSCNT–CME. Acknowledgements The project was supported by the National Natural Science Foundation of China (50907056, 50872116, 50588201), the Changjiang Scholars and Innovative Team Program of China (IRT0751), the National Hi-Tech Project of China (2007AA03Z203), the Fundamental Research Funds for the Central Universities (SWJTU09CX053), the Science and Technology Research Funds of Sichuan Province (2006Z02-006-1), the Applied Basic Research Funds of Sichuan Province (2008JY0061), and the Fundamental Science Funds of Southwest Jiaotong University (2007B20). References [1] S. Iijima, Nature 354 (1991) 56. [2] M. Monthioux, V.L. Kuznetsov, Carbon 44 (2006) 1621.
5209
[3] K. Tsukagoshi, N. Yoneya, S. Uryu, Y. Aoyagi, A. Kanada, Y. Ootuka, et al., Physica B 323 (2002) 107. [4] A. Peigney, C. Laurent, E. Flahaut, A. Rousset, Ceram. Int. 26 (2000) 677. [5] K.T. Lau, M. Chipara, H.Y. Ling, D. Hui, Composites Part B 35 (2004) 95. [6] E. Poirier, R. Chanhine, P. Bénard, D. Cossement, L. Lafi, E. Mélancon, et al., Appl. Phys. A 78 (2004) 961. [7] W.Z. Li, C.H. Liang, J.S. Qiu, W.J. Zhou, H.M. Han, Z.B. Wei, Carbon 40 (2002) 791. [8] X. Zhao, M. Ohkohchi, M. Wang, S. Iijma, T. Ichihashi, Y. Ando, Carbon 35 (1997) 775. [9] F. Kokai, K. Takahashi, D. Kasuya, T. Ichihashi, M. Yudasaka, S. Iijima, Chem. Phys. Lett. 332 (2000) 449. [10] C. Singh, M.S.P. Shaffer, A.H. Windle, Carbon 41 (2003) 359. [11] C. Pan, X. Xu, J. Mater. Sci. Lett. 21 (2002) 1207. [12] L. Wang, C. Li, Q. Zhou, F. Gu, L. Zhang, Phys. B: Condens. Matter 398 (2007) 18. [13] Q. Zhou, C. Li, F. Gu, H.L. Du, J. Alloys Compd. 463 (2008) 317. [14] L. Yuan, K. Saito, W. Hu, Z. Chen, Chem. Phys. Lett. 346 (2001) 23. [15] W. Merchant-Merchant, A. Saveliev, L.A. Kennedy, A. Fridman, Chem. Phys. Lett. 354 (2002) 20. [16] Y. Liu, Q. Fu, C. Pan, Carbon 43 (2005) 2264. [17] Y.X. Liu, J.H. Liu, C.C. Zhu, Appl. Surf. Sci. 255 (2009) 7985. [18] R. Saito, M. Fujita, G. Dresselhaus, M.S. Dresselhaus, Appl. Phys. Lett. 60 (1992) 2204. [19] J. Wang, Electroanalysis 17 (2005) 7. [20] C.E. Banks, T.J. Davies, G.G. Wildgoose, R.G. Compton, Chem. Commun. 7 (2005) 829. [21] P.J. Britto, K.S.V. Santhanam, P.M. Ajayan, Bioelectrochem. Bioenerg. 41 (1996) 121. [22] Y.D. Zhao, W.D. Zhang, H. Chen, Q.M. Luo, S.F.Y. Li, Sens. Actuators B: Chem. 87 (2002) 168. [23] Z.H. Wang, J. Liu, Q.L. Liang, Y.M. Wang, G.A. Luo, Analyst 127 (2002) 653. [24] M. Musameh, J. Wang, A. Merkoci, Y. Lin, Electrochem. Commun. 4 (2002) 743. [25] Q. Jiang, L.J. Song, H. Yang, Z.W. He, Y. Zhao, Electrochem. Commun. 10 (2008) 424. [26] L.J. Song, Q. Jiang, J. Yi, X.T. Zhu, Y. Zhao, Chin. J. Inorg. Chem. 21 (2006) 1345. [27] Q. Jiang, Z.W. He, B. Du, X.F. Zhao, Y. Zhao, J. Phys.: Conf. Ser. 152 (2009) 012019. [28] M. Chhowalla, K.B.K. Teo, C. Ducati, N.L. Rupesinghe, G.A.J. Amaratunga, A.C. Ferrari, et al., J. Appl. Phys. 90 (2001) 5308.
Contents lists available at ScienceDirect
Electrochimica Acta journal homepage: www.elsevier.com/locate/electacta
Preparation and characterization of a graphite electrode containing carbon nanotubes grown in situ by flame synthesis Qi Jiang a,∗ , Rong Yang a , Zhengwen He a , Zhao Liu a , Deyu Xie a , Yong Zhao a,b a Key Laboratory of Advanced Technologies of Materials (Ministry of Education of China) and Superconductivity R&D Centre, Southwest Jiaotong University, Chengdu 610031, PR China b School of Materials Science and Engineering, University of New South Wales, Sydney, 2052 NSW, Australia
a r t i c l e
i n f o
Article history: Received 4 February 2011 Received in revised form 10 March 2011 Accepted 11 March 2011 Available online 21 March 2011 Keywords: Carbon nanotubes Chemically modified electrode Flame synthesis
a b s t r a c t A crystalline flake graphite electrode (GE) was impregnated with nickel particles using direct current electrochemical deposition. The particles were used for in situ growth of carbon nanotubes (CNTs) by flame synthesis with a liquid ethanol flame. The obtained electrode was characterized by X-ray diffraction, and scanning and transmission electron microscopy. The results showed that the deposited Ni catalyst crystal face was mainly (1 1 1). CNTs with a diameter of about 40 nm were uniformly grown on the GE surface. The electrochemical performance of the CNT–GE was characterized by cyclic voltammetry using a [Fe(CN)6]3− /[Fe(CN)6]4− solution, and showed a much greater electrochemical response than that obtained using a material in which CNTs were grown by catalytic chemical vapor deposition. © 2011 Elsevier Ltd. All rights reserved.
1. Introduction Carbon nanotubes (CNTs) have attracted extensive attention in many science and technology fields since they were discovered [1,2]. Due to their unique structure and physical properties, CNTs have been considered as one of the best candidate for fabricating nano-electronic devices [3], composite materials [4,5], gas storage materials [6] and catalyst support materials [7]. Regular methods such as arc-discharge [8], laser ablation [9], and catalytic chemical vapor deposition (CCVD) [10] have been successfully used to synthesize CNTs. Flame synthesis is a new developing method to grow CNTs compared with the methods mentioned above [11–15], and its prominent advantages are energy saving up and simple operation. In flame synthesis, some fuel provides heat while the remainder serves as carbon source for CNT growth. Recently, liquid ethanol had been developed as carbon source for CNT growth by a common ethanol burner [16,17]. In comparison with the regular methods mentioned above, flame synthesis was much simpler and more cost-effective. In electrochemical field, CNT novel characteristics including high aspect ratio, nano-scale dimensions and good electrical conductivity have indicated that CNTs are a type of potential modified materials for chemically modified electrode (CME) [18,19]. According to the literatures, five ways were developed to prepare
∗ Corresponding author at: Superconductivity R&D Centre, Southwest Jiaotong University, 111, No. 1, North, 2 Ring Road, Chengdu 610031, PR China. Tel.: +86 28 87603544; fax: +86 28 87603544. E-mail addresses: [email protected], [email protected] (Q. Jiang). 0013-4686/$ – see front matter © 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.electacta.2011.03.038
CNT–CME [20–24], which are listed as follows: The first is mixing the CNTs, bromoform, mineral oil and liquid paraffin carefully and stuffing the mixture into the glass capillary to prepare electrode. The next is filling the CNTs into the corration platinum electrode to prepare CNT powder microelectrode. The third is embedding the CNTs into the graphite electrode (GE) to prepare embedding modified electrode. The fourth is “abrasive immobilization” (simply rubbing the CNTs onto an electrode surface to prepare electrode). The last is dispersing the CNTs into the H2 SO4 solution, H2 O solvent, dimethyl formamide solvent and coating directly on the surface of the electrode to prepare CNT–CME. Following these ways, CNT–CMEs were done as two steps: firstly, CNTs were prepared by the CNT growth methods mentioned above and then the CNTs were fixed on the electrode surface by some physical ways, for example, using binder. However, the binder could reduce the CNT utilization efficiency by enhancing the resistance or changing the CNT novel nanometer hollow tube structure. In order to avoid these defects, we have introduced a new method to prepare CNT–CME by growing CNTs in situ (named as GSCNT–CME) in our previous work [25]. The GSCNT–CME was obtained by directly growing CNT on a crystalline flake graphite electrode (GE) surface, namely one-step method. The obtained GSCNT–CME can hold the CNT pristine nanometer structure and has showed excellent potential application value. In that report, CNTs were grown by CCVD and CNT catalyst on the electrode was prepared by immersion. However, the operation is relative complicated and inefficient (the operation process is over 4 h and in vacuum). In order to enhance the GSCNT–CME preparation process practicability and maneuverability, flame synthesis and catalyst direct current electrochemical deposition were used to replace CCVD and
5206
Q. Jiang et al. / Electrochimica Acta 56 (2011) 5205–5209
catalyst immersion in this paper. And a liquid ethanol was used as the carbon resource. With these improvements, the GSCNT–CME can be prepared in a few minutes and without vacuum condition. Therefore, the operation is much simpler and safer. The CNTs with a diameter of about 40 nm are uniformly grown on the GE surface. All these show that the way used here is of much more practicability and maneuverability. 2. Experimental 2.1. Electrochemical deposition nickel catalyst To prepare Ni catalyst, a galvanostatic – voltage instrument (ZF9, Shanghai, China) was used. Ni slice (99.99%) with 1 cm × 3 cm size was used as the anode. A crystalline flake graphite disk (5.3 mm in diameter, 7.0 mm in length) was used as the electrode. The graphite disk was polished to a mirror surface with the metallographic abrasive paper (W 7-05). Then, it was sealed by fixing into a teflon housing with the polished surface out. The obtained electrode was named as GE. The electrochemical deposition solution was composed of nickel sulfate (NiSO4 ·7H2 O, 300 g L−1 ), nickel chloride (NiCl2 ·6H2 O, 45 g L−1 ), boric acid (H3 BO3 , 40 g L−1 ), saccharin sodium (C6 H4 COSO2 NNa·2H2 O, 5 g L−1 ), sodium dodecyl sulfate (CH3 (CH2 )11 OSO3 Na, 0.1 g L−1 ) and redistilled water. The direct current electrochemical deposition was carried out for 1.5 min using a current of 20 mA. The deposited Ni catalyst GE was named as Ni/GE and used to grow CNT. 2.2. GSCNT–CME preparation A liquid anhydrous ethanol flame was used to grow CNT. The flame height and middle width were about 20 and 60 mm, respectively. The wick height was about 10 mm. In order to grow CNTs, the Ni/GE was placed about 8 mm over the wick for 4 min, where the flame color was blue and the temperature was about 610 ◦ C (tested by an electronic sensor). After that, a layer of black sample appeared on the Ni/GE surface. Then the obtained electrode was treated by concentrated sulfuric acid, supersonic and redistilled water for removing the impurities (including amorphous carbon, catalyst and so on). As this, the GSCNT–CME was prepared. 2.3. Morphology and microstructure characterization The GE and the Ni/GE surfaces were characterized by scanning electron microscopy (SEM, QUANTA 200). The crystal structure of the deposited Ni was characterized by X-ray diffraction (XRD) (XRD, PWI 3040/60). The morphology and microstructure of the CNTs were also characterized by SEM and transmission electron microscopy (TEM, HITACHI H-700H). 2.4. Electrochemical performance testing Cyclic voltammetry (CV) with three-electrode-system was used to test the obtained GSCNT–CME electrochemical performance. A platinum wire was used as the counter electrode, the saturated calomel electrode (SCE) was used as the reference electrode and the obtained GSCNT–CME was used as the working electrode in the testing system. 3. Results and discussion 3.1. SEM, XRD and TEM study Fig. 1 shows SEM images of the GE surface after being polished (a and b) and the Ni/GE surface after the electrochemical deposition
Fig. 1. SEM images of GE before (a and b) and after (c) the catalyst deposition.
(c). It can be seen from Fig. 1a that the GE surface after being polished is very smooth and flat. Fig. 1b shows SEM images with higher amplification (3 times higher) of the GE after being polished. Fig. 1c shows SEM images at the same amplification (30,000) of the Ni/GE surface after electrochemical deposition. Comparing Fig. 1b with c, it is found that there are lots of grains in Fig. 1c, and the average grain diameter is about 35 nm. These grains are the result of direct current electrochemical deposition. In order to characterize the grain component and structure, XRD analysis was used.
Q. Jiang et al. / Electrochimica Acta 56 (2011) 5205–5209
5207
Fig. 2. XRD patterns of the Ni/GE.
Fig. 2 shows XRD patterns of the obtained Ni/GE. It can be seen from this figure that there are three distinct peaks from 35◦ to 60◦ . According to the XRD PDF database with the XRD scanning program, we can have the peak type and crystal face data. There are (1 0 0) crystal face peak of carbon (42.439◦ ), (1 1 1) crystal face peak of Ni (44.574◦ ) and (0 0 4) crystal face peak of carbon (54.571◦ ). The carbon peaks come from GE. Therefore, it can be deduced that Ni nanometer particle introduction is the result of direct current electrochemical deposition. Moreover, the Ni particle is mainly the (1 1 1) crystal face. The average diameter of the deposited Ni is about 34 nm calculated by the SCHERRER equation, which is the same as the SEM result mentioned above. According to our previous reports [26,27], Ni with nanometer size can be used as the catalyst to grow CNTs and Ni (1 1 1) crystal face is even better for the CNT growth. Therefore, the obtained Ni/GE can be directly used to grow CNTs and the method (direct current electrochemical deposition) is suitable for preparing CNT catalyst. Fig. 3 shows SEM images (a and b) and TEM images (c) of the CNTs grown on the Ni/GE by the flame synthesis. From Fig. 3a and b, we can see that the CNTs are dense, clean (without other carbon impurities) and evenly grown on the GE surface. Moreover, the CNT diameter is uniform (about 40 nm) and the length is 1–3 m. In order to study the CNT structure, the CNTs were removed from the electrode surface and characterized by TEM (Fig. 3c). A nanometer hollow structure of the prepared CNTs can be clearly seen from Fig. 3c. The CNT diameter is very uniform and about 40 nm, which is consistent with the Ni catalyst diameter [28]. These phenomena also indicate that the size of the catalyst obtained by the direct current electrochemical deposition is also uniform. 3.2. Electrochemical performance characterization Fig. 3. SEM images (a and b) and TEM images (c) of the CNTs grown on the GE.
In order to characterize the GSCNT–CME electrochemical performance, CV testing with three-electrode-system is used. A [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution is also used to characterize the obtained electrode in this article. Fig. 4 shows CV curves of GE, Ni/GE and GSCNT–CME in 1.0 mol L−1 KCl solution (a) and 1.0 × 10−3 mol L−1 K3 Fe(CN)6 + 1.0 × 10−3 mol L−1 K4 Fe(CN)6 in 1.0 mol L−1 KCl solution (b) at a scanning rate of 10 mV s−1 vs. SCE. From Fig. 4a, it can be seen that the GSCNT–CME CV curve is different from those of the GE and Ni/GE under the same testing conditions. The surrounded area of the GSCNT–CME CV curve is much larger than those of the GE and Ni/GE, indicating that the GSCNT–CME has larger specific surface area. There are only CNTs on the GSCNT–CME surface compared with the Ni/GE, so the differ-
ence of the CV curves is the result of the CNTs. That’s because that the CNTs are one-dimensional nanometer materials, which have large specific surface area. The CV curves of the GE and Ni/GE have a little difference, indicating that Ni deposition has a little effect on the electrochemical performance in our testing scope. The reason is that the amount of the deposited Ni is finite and the effect of Ni on the GE can be considered as enhancing the GE electrode conductivity, we consider. From Fig. 4b, it can be seen that the current responses of the GE and Ni/GE to the [Fe(CN)6 ]3− /[Fe(CN)6 ]4− oxidation–reduction pair are very weak with modifying the voltage from 300 mV
5208
Q. Jiang et al. / Electrochimica Acta 56 (2011) 5205–5209
Table 1 Data of CV testing based on the GSCNT–CME with different scanning rates in the [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution vs. SCE. Scanning rate (mV s−1 )
Square root of scanning rate ((mV s−1 )1/2 ) E (mV) Peak current (10−3 mA)
0.83
1.67
5
10
20
50
100
0.91 22.04 212.7
1.29 29.50 390.1
2.23 70.97 922.6
3.16 104.95 1562.5
4.47 151.93 2643.9
7.07 255.00 4956.3
10 380.01 7812.4
to 900 mV, whereas the GSCNT–CME current response to the [Fe(CN)6 ]3− /[Fe(CN)6 ]4− oxidation–reduction pair is very strong. The oxidation and reduction peaks based on GSCNT–CME appear at 709.3 and 601.4 mV, respectively. So the current response to the [Fe(CN)6 ]3− /[Fe(CN)6 ]4− based on GSCNT–CME is much stronger than those of the GE and Ni/GE under the same testing conditions. In order to further research the obtained GSCNT–CME electrochemical performance, CV testing at different scanning rates (0.83, 1.67, 5, 10, 20, 50, 100 mV s−1 ) was operated in a [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution (1 × 10−3 mol L−1 ). The data of the oxidation peak current (Ipk ) and the redox peak voltage difference (E) at the different scanning rates are listed in Table 1. Table 1 shows the data of CV testing based on the GSCNT–CME with different scanning rates in the [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution. It is evident from the table that the E increases with increasing scanning rate, indicating that this is not a reversible system. From the
Fig. 5. Relationships between the scanning rate and the oxidation peak current of the GSCNT–CME obtained by flame synthesis in this paper with 1 × 10−3 mol L−1 [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution (a) and the reported GSCNT–CME obtained by CCVD in Ref. [25] with 1 × 10−2 mol L−1 [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution (b).
relationship between the scanning rate square root and the oxidation peak current of the GSCNT–CME in the 1 × 10−3 mol L−1 [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution, it can be found that the relationship is excellent linear, and the linear regression equation is Ipk = 211.4w1/2 − 153.0 (w is the scanning rate) with 0.996 of the linear correlation coefficient. Therefore, it is evident that the electrochemical reaction process is a typical diffusion control process. Fig. 5 shows the relationships between the scanning rate and the oxidation peak current based on the GSCNT–CME obtained by flame synthesis in this paper with 1 × 10−3 mol L−1 [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution (a) and reported GSCNT–CME obtained by CCVD in Ref. [25] with 1 × 10−2 mol L−1 [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution (b). From Fig. 5, it can be seen that two GSCNT–CME oxidation peak currents to the [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution both increase with increasing the scanning rates. Moreover, the oxidation peak current based on the GSCNT–CME obtained by flame synthesis in this paper enhances much greater than that of the GSCNT–CME obtained by CCVD in Ref. [25] even if the [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution concentration in this article is 1 × 10−3 mol L−1 , lower than that in Ref. [25] (1 × 10−2 mol L−1 ). The results show that the electrochemical response to [Fe(CN)6 ]3− /[Fe(CN)6 ]4− solution of the GSCNT–CME obtained in this article is much greater than that of the GSCNT–CME obtained in Ref. [25].
4. Conclusions
Fig. 4. CV curves of the GE, Ni/GE and GSCNT–CME in the substrate solution ((a) 1.0 mol L−1 KCl solution) and the testing solution ((b) 1.0 × 10−3 mol L−1 K3 Fe(CN)6 + 1.0 × 10−3 mol L−1 K4 Fe(CN)6 in 1.0 mol L−1 KCl solution) with the scanning rate of 10 mV s−1 vs. SCE.
Several conclusions can be drawn from the experimental results and discussion mentioned above, including: Firstly, the nickel catalyst with grain size from 30 nm to 40 nm can be deposited on the GE surface by the direct current electrochemical deposition in 1.5 min. The deposited nickel can be directly used as the catalyst to grow CNTs.
Q. Jiang et al. / Electrochimica Acta 56 (2011) 5205–5209
Secondly, CNTs with the diameter of about 40 nm can be grown on the GE surface in 4 min without vacuum condition by flame synthesis. GSCNT–CME can be obtained by this simple, economic and safe method, namely depositing nickel catalyst on a GE surface by direct current electrochemical deposition and then growing CNTs by flame synthesis with the liquid ethanol burner flame. Thirdly, the obtained GSCNT–CME electrochemical performance testing has been carried out. The researching results show that the obtained GSCNT–CME has better electrochemical performance than that of the GSCNT–CME obtained by CCVD in Ref. [25], indicating that the flame synthesis is a more potential and suitable for preparing GSCNT–CME. Acknowledgements The project was supported by the National Natural Science Foundation of China (50907056, 50872116, 50588201), the Changjiang Scholars and Innovative Team Program of China (IRT0751), the National Hi-Tech Project of China (2007AA03Z203), the Fundamental Research Funds for the Central Universities (SWJTU09CX053), the Science and Technology Research Funds of Sichuan Province (2006Z02-006-1), the Applied Basic Research Funds of Sichuan Province (2008JY0061), and the Fundamental Science Funds of Southwest Jiaotong University (2007B20). References [1] S. Iijima, Nature 354 (1991) 56. [2] M. Monthioux, V.L. Kuznetsov, Carbon 44 (2006) 1621.
5209
[3] K. Tsukagoshi, N. Yoneya, S. Uryu, Y. Aoyagi, A. Kanada, Y. Ootuka, et al., Physica B 323 (2002) 107. [4] A. Peigney, C. Laurent, E. Flahaut, A. Rousset, Ceram. Int. 26 (2000) 677. [5] K.T. Lau, M. Chipara, H.Y. Ling, D. Hui, Composites Part B 35 (2004) 95. [6] E. Poirier, R. Chanhine, P. Bénard, D. Cossement, L. Lafi, E. Mélancon, et al., Appl. Phys. A 78 (2004) 961. [7] W.Z. Li, C.H. Liang, J.S. Qiu, W.J. Zhou, H.M. Han, Z.B. Wei, Carbon 40 (2002) 791. [8] X. Zhao, M. Ohkohchi, M. Wang, S. Iijma, T. Ichihashi, Y. Ando, Carbon 35 (1997) 775. [9] F. Kokai, K. Takahashi, D. Kasuya, T. Ichihashi, M. Yudasaka, S. Iijima, Chem. Phys. Lett. 332 (2000) 449. [10] C. Singh, M.S.P. Shaffer, A.H. Windle, Carbon 41 (2003) 359. [11] C. Pan, X. Xu, J. Mater. Sci. Lett. 21 (2002) 1207. [12] L. Wang, C. Li, Q. Zhou, F. Gu, L. Zhang, Phys. B: Condens. Matter 398 (2007) 18. [13] Q. Zhou, C. Li, F. Gu, H.L. Du, J. Alloys Compd. 463 (2008) 317. [14] L. Yuan, K. Saito, W. Hu, Z. Chen, Chem. Phys. Lett. 346 (2001) 23. [15] W. Merchant-Merchant, A. Saveliev, L.A. Kennedy, A. Fridman, Chem. Phys. Lett. 354 (2002) 20. [16] Y. Liu, Q. Fu, C. Pan, Carbon 43 (2005) 2264. [17] Y.X. Liu, J.H. Liu, C.C. Zhu, Appl. Surf. Sci. 255 (2009) 7985. [18] R. Saito, M. Fujita, G. Dresselhaus, M.S. Dresselhaus, Appl. Phys. Lett. 60 (1992) 2204. [19] J. Wang, Electroanalysis 17 (2005) 7. [20] C.E. Banks, T.J. Davies, G.G. Wildgoose, R.G. Compton, Chem. Commun. 7 (2005) 829. [21] P.J. Britto, K.S.V. Santhanam, P.M. Ajayan, Bioelectrochem. Bioenerg. 41 (1996) 121. [22] Y.D. Zhao, W.D. Zhang, H. Chen, Q.M. Luo, S.F.Y. Li, Sens. Actuators B: Chem. 87 (2002) 168. [23] Z.H. Wang, J. Liu, Q.L. Liang, Y.M. Wang, G.A. Luo, Analyst 127 (2002) 653. [24] M. Musameh, J. Wang, A. Merkoci, Y. Lin, Electrochem. Commun. 4 (2002) 743. [25] Q. Jiang, L.J. Song, H. Yang, Z.W. He, Y. Zhao, Electrochem. Commun. 10 (2008) 424. [26] L.J. Song, Q. Jiang, J. Yi, X.T. Zhu, Y. Zhao, Chin. J. Inorg. Chem. 21 (2006) 1345. [27] Q. Jiang, Z.W. He, B. Du, X.F. Zhao, Y. Zhao, J. Phys.: Conf. Ser. 152 (2009) 012019. [28] M. Chhowalla, K.B.K. Teo, C. Ducati, N.L. Rupesinghe, G.A.J. Amaratunga, A.C. Ferrari, et al., J. Appl. Phys. 90 (2001) 5308.