Carbon-Embedded Carbon Nanotubes as Supports of Polymer Electrolyte Membrane Fuel Cell Catalysts

Transcription

1 Copyright 2014 American Scientific Publishers All rights reserved Printed in the United States of America Article Journal of Nanoscience and Nanotechnology Vol. 14, , Carbon-Embedded Carbon Nanotubes as Supports of Polymer Electrolyte Membrane Fuel Cell Catalysts Huaiguang Li, Xiao Zhang, Daping He, Tao Peng, and Shichun Mu State Key Laboratory of Advanced Technology for Materials Synthesis and Processing, Wuhan University of Technology, Wuhan , China Carbon shells embedded carbon nanotubes can facilitate Pt nanoparticles loading and dispersing on the core shell nanostructural support. Carbon shells are prepared by coating polyaniline layers on the core (carbon nanotubes) with in-situ polymerization and subsequent carbonization. The carbon shell embedded carbon nanotube supported Pt catalyst reveals high electrochemical active surface area and mass activity, which are the factors of 1.4 times and 2.2 times higher than that of the pristine carbon nanotube supported Pt, respectively. In addition, the carbon shell embedded carbon nanotube supported Pt catalyst has a higher stability than the carbon nanotube supported Pt catalyst. The improved catalytic activity and stability of our new catalyst can be ascribed to the improved Delivered dispersion by of Publishing Pt nanoparticles Technology on surfaces to: Main of carbon CID is nanotubes and (JPP) the interaction between Pt nanoparticles IP: and carbon shells. On: Tue, 01 Apr :28:08 Copyright: American Scientific Publishers Keywords: Core Shell Structure, Carbon Nanotube, Catalyst, PEM Fuel Cell. 1. INTRODUCTION Polymer electrolyte membrane (PEM) fuel cell as one of the most promising power source candidates are very attractive in various transportation, distributed home power generation, and back-up sources for small portable electronics due to their high efficiency (usually about 50%) and mild operating conditions. 1 3 However, improving the performance and lifetime of noble-metal (such as Pt) catalysts for PEM fuel cells have been the predominant objective of considerable attention because of the big loss and radical agglomeration of noble-metal catalysts on surfaces of supports. 4 5 Several groups reported various methodologies, such as modifying noble metal catalyst with polymer, selecting corrosion resistance support to enhance the noble-metal catalyst lifetime. 6 8 The previous researches showed that the carbon nanotubes (CNTs), with high electrical conductivity, high chemical stability as well as extremely high mechanical strength and modulus properties, could make a major step forward in improving the catalyst lifetime Xin et al. reported that the prepared multiwalled CNT (MCNT) Author to whom correspondence should be addressed. supported Pt cathode catalysts displayed significantly higher performance than the Pt/XC-72 catalyst for direct methanol fuel cells. 13 Yan et al. addressed the issue of durability for CNTs as catalyst supports for PEM fuel cells and showed that the MWNTs, with less surface oxide formation and 30% lower corrosion current under the investigated condition, are electrochemically more stable than the Vulcan XC Simultaneously, the CNT-supported catalyst is available to increase the catalyst utilization and performance of fuel cells compared to the commercial Pt/C catalysts More importantly, the CNT-supported catalyst is more stable than the Pt/C catalyst. 17 However, one of the major disadvantages of using the CNT supported noble-metal catalyst is the limited loading and inhomogeneous dispersion of metal nanoparticles on surfaces of CNTs because of the smooth and inert nature of CNT surfaces which is adverse to effectively anchor catalyst nanoparticles. The motivation to improve these disadvantages has been triggered by some researchers. For example, Kim et al. reported that the CNTs oxidized by nitric acid in order to introduce functionality of surfaces might facilitate catalysts anchored on CNTs. 18 Wu et al. addressed the synthesis of doped polyaniline in its emeraldine salt form with carboxylic groups containing multi-walled carbon J. Nanosci. Nanotechnol. 2014, Vol. 14, No /2014/14/6929/005 doi: /jnn

2 Carbon-Embedded Carbon Nanotubes as Supports of Polymer Electrolyte Membrane Fuel Cell Catalysts nanotubes via in-situ polymerization 19 and the polyaniline/mwcns and polyaniline/poly materials composites with core shell structures were also prepared by some groups In this work, we proposed a new method to obtain highly dispersed Pt nanoparticles on CNTs, which involves the coating of polyaniline (PANI) layers on CNTs surface analogous to the literature, and then the PANI was carbonized on the CNT surface (C/CNTs) in N 2 protection. Furthermore, Pt nanoparticles were loaded on the C/CNTs (Pt C/CNTs) through glycol reduction. Such a structure originates from the formation of the anchor mechanism in which Pt nanoparticles are stabilized onto the surface of CNTs. The formed C layers outside CNTs can anchor Pt particles due to strong adsorption interaction between them; this effectively prevents Pt nanoparticles from aggregating on such support surfaces. 2. EXPERIMENTAL DETAILS The raw high-purity CNTs were purchased from Shenzhen Nanotech Port Co. Ltd. and used without further treatment. Chemicals used in this work were of analytical grade Preparation of C/CNTs 2.3. Characterization of the Prepared Catalysts Scheme 1 illustrates the approach used to synthesize the C/CNTs. The core shell structural C/CNTs were prepared High-resolution transmission electron microscopy Delivered by Publishing Technology to: Main CID is (JPP) by two steps. 5 ml of aniline IP: and ml of HCl solution at 1 M were added into the CNTsCopyright: solution and American stirred (HRTEM) investigations were carried out in a JEOL On: Tue, Apr :28:08 JEM-2010 transmission electron microscopy (TEM). The Scientific Publishers with ultrasonic treatment for 5 min. During mechanically stirred, 100 ml of water solution of ammonium persulfate (NH 4 S 2 O 8 at different concentrations was added into the above solution dropwise at a rate of about 20 ml/h for polymerization of aniline monomer, where NH 4 S 2 O 8 acted as an initiator. This polymerization reaction was carried out at the temperature of 0 C in an ice-water bath under supersonic agitation. After in-situ polymerization reaction, the core shell polyaniline/carbon nanotubes (PANI/CNTs) Carbon nanotube Carbon shell PANI layer Pt nanoparticles Scheme 1. Fabrication carbon shell on the surface of CNTs. (1) Through in situ polymerization reaction to synthesize the core shell carbon/polyaniline nanotube. (2) The conversion of the core shell carbon/polyaniline nanotube into the carbon shell CNTs. (3) Preparation of Pt C shell CNTs. See experimental section for details. were obtained by filtering, washing with deionized water. On the other hand, the conversion of the PANI/CNTs into the C/CNTs was accomplished in two steps as well, namely, stabilization and carbonization. The stabilization was carried out at 270 C in air for 1 h and followed by the carbonization at 800 CinN 2 atmosphere for 2hatthe heating rate of 5 C /min and N 2 flow rate of 200 ml/min Preparation of Pt C/CNTs and Pt/CNTs Catalysts The Pt C/CNTs catalyst was prepared as follows. 20 ml of hexachloroplatinic acid with a concentration of 1.5 mg Pt/mL and 200 ml glycol were mixed and refluxed at 130 C. ph of the solution was next adjusted to 9 10 with additional NaOH solution. The Pt colloidal nanoparticles were obtained after the solution color changed from yellow to black. Then 0.1 g of the prepared C/CNTs was added to the solution under agitation for 3 4 h. The solid was filtered and washed with deionized water and then dried at 80 C in a vacuum oven, and then the Pt loaded Pt C/CNTs catalyst was prepared. The Pt/CNTs catalyst as the benchmark was made in a similar process. The Pt loading of both the Pt C/CNTs and Pt/CNTs catalysts was 20% in weight. Pt C/CNTs catalyst was characterized by recording their X-ray diffraction (XRD) patterns on a Rigaku X-ray diffractometer with CuK radiation source. The 2 angular regions between 15 and 100 were explored at a scan rate of 2 /min. The electrochemical measurements were conducted in a three-electrode cell setup. A saturated calomel electrode (SCE) and a platinum foil were employed as the reference electrode and counter electrode, respectively. The working electrode was prepared as follows: 25 5 mg of the catalysts and 100 L of 5 wt.% PFSA emulsions were mixed in 1 ml of alcohol solution to form a dispersion of ink. And then 5 L mixed ink was coated on a mirror-polished glassy carbon disk electrode as the working electrode. The measurement was carried out in 0.5 M H 2 SO 4 at the scan rate of 50 mv/s in a potential range of 0 to 1.2 V versus NHE at room temperature. The electrochemical active surface area (ECSA) of the catalysts was calculated on the following formula: ECSA = Q H (1) m q H where Q H is charge for H upd adsorption, m is the loading amount of Pt, and q H (210 C/cm 2 ) is the charge required for monolayer adsorption of hydrogen on Pt surfaces. The oxygen reduction reaction (ORR) activity of the as-prepared catalysts was evaluated by the rotating disk electrode (RDE) technique in 0.5 M H 2 SO 4 electrolyte at 6930 J. Nanosci. Nanotechnol. 14, , 2014

3 Carbon-Embedded Carbon Nanotubes as Supports of Polymer Electrolyte Membrane Fuel Cell Catalysts from 2 to 5 nm with a peak centered at 2.9 nm. In comparison, Figure 2(C) shows the TEM image of Pt/CNTs, which has a limited loading and inhomogeneous dispersion of Pt nanoparticles on surfaces of the CNTs. These results well prove that adding PANI layers on surfaces of CNTs and subsequent carbonization can promote the dispersion of Pt nanoparticles on CNTs. This is because the formation of platinum nanoclusters has two distinct stages: The first stage is the formation of nuclei of the new phase. The nuclei then grow into bigger particles with time. However, the densities of surface functional sites on the smooth and Figure 1. TEM images of the core/shell structure of C/CNTs (A). An inert nature of CNT surfaces are insufficient to provide enlargement of an area in (A) as indicated by the red circle (B). the nuclei for Pt seeds. By contrary, the richness of the 1 functional groups on the carbon shell associated with high a sweep rate of 10 mv s and a speed of 1600 rpm at electrochemically accessible surface area is much more room temperature. suitable to form the nuclei for Pt seeds. As a result, the Electrochemical accelerated durability tests (ADT) were core/shell structure of the C/CNTs can be available to synconducted to characterized the long-term performance of thesize high-loading, high-dispersion Pt catalysts. Pt C/CNTs catalysts, which were tested in the current XRD measurements were performed for further invesstudy with CV curves between 0.6 and 1.20 V in 0.5 M tigation of the Pt nanoparticle size. The powder XRD H2 SO4. patterns for the Pt C/CNTs and Pt/CNTs are shown in Figure 3. Both the catalysts display the characteristic pat3. RESULTS AND DISCUSSION terns of Pt face centered cubic (fcc) structure. The peaks at The characterization of TEM was conducted to examaround 2 = 39 7, 46.4, 67.7, and 81.4 are assigned to ine the core/shell structure of C/CNTs. A typical TEM the (111), (200), (220), and (311) planes of Pt respectively. image which reveals the core/shell structure is shown in Since the Pt (220) peak is isolated from the carbon-support Figure 1(A). Figure 1(B) is an enlargement of one area seen peaks, the average crystalline size of Delivered by Publishing Technology graphite to: Maindiffraction CID is (JPP) in Figure 1(A) as indicated by theip: red circle. There is a On: dis- Tue, Pt01 nanopaticles be calculated from this peak according Apr 2014 can 09:28:08 tinct boundary between the shell and Copyright: core, and the lattice Scientific American Publishers to Scherrer s formula:13 24 fringes of CNTs as the core can be seen very clearly, which suggests the existence of carbon shells that cover the CNT L = 0 9 /B cos (2) core. The thickness of the carbon shell is about 2 nm. where L is the average diameter of the crystal pellets; is As shown in Figures 2(A) and (B), the TEM images disthe X-ray wavelength (Cu KR ìkr1) Å; is the play high and homogeneous dispersion of the Pt nanoparbragg angle corresponding to the peak maximum; and B is ticles on the wall of C/CNTs. Further characterization of the half-peak width for Pt (220) in radians. The calculated HRTEM reveals that the Pt nanoparticles, with a uniform average crystalline sizes of Pt for Pt C/CNTs and Pt/CNTs size, are highly dispersed on the surface of C/CNTs. catalysts are about 3.0 nm and 3.1 nm, respectively. This The analysis of Pt particle-size diameters for Pt C/CNTs catalysts indicates a dispersion of the Pt particles ranging is in a good agreement with TEM results. Figure 2. TEM images of Pt C/CNTs catalysts (A) and HRTEM image of Pt C/CNTs catalysts (B). HRTEM image of Pt /CNTs catalysts (C). J. Nanosci. Nanotechnol. 14, ,

4 Carbon-Embedded Carbon Nanotubes as Supports of Polymer Electrolyte Membrane Fuel Cell Catalysts Figure 3. XRD patterns of Pt C/CNTs (A) and Pt /CNTs (B) catalysts. kinetic diffusion controlled region compared to that of the To investigate the electrochemical performance of the Pt/CNTs catalyst. The kinetic current was calculated from Pt C/CNTs catalyst, the steady-state CVs are presented in the ORR polarization curve according to the Koutecky- Figure 4. Typical adsorption and desorption behaviors of Levich Equations. hydrogen and oxygen can be clearly detected. It is noted that the current peak value of the Pt C/CNTs is obviously higher than that of Pt/CNTs catalyst. The ECSA was i = (3) 1 i k i d calculated by Eq. (1) shows that the Pt C/CNTs catalyst where i is the experimentally measured current, i provides a higher ECSA (58.5 m 2 /g) which is almost 1.4 d is the diffusion-limiting current, and i times of that of the Pt/CNTs Delivered (44.13by mpublishing 2 /g). It well Technology known k is the kinetic current. Therefore, to: Main CID theis kinetic current (JPP) can be obtained by Eq. (3) that ECSA is mainly controlledip: by size of Pt nanoparticles dispersed on the support and PtCopyright: utilization American which is Scientific On: Tue, which 01 Apr suggests :28:08 that the mass activity of the Pt C/CNTs Publishers relevant to Pt nanoparticles in contact with ionomer. 28 For the Pt C/CNTs catalyst, the increase of ECSA might be attributed to the well-dispersed, small-sized Pt particles on the support surfaces and to higher Pt utilization. 29 The ORR activity of the Pt C/CNTs and Pt/CNTs catalysts is shown in Figure 5. It exhibits the Pt C/CNTs has a higher reduction current in comparison with the Pt/CNTs catalyst. The half-wave potential of the Pt C/CNTs presents an anodic shift of about 30 mv in the mixed Figure 4. Cyclic voltammetry curves and electrochemically active surface area of catalysts in 0.5 mol/l H 2 SO 4 at a scan rate of 50 mv/s at room temperature. Figure 5. Current-potential curves for ORR in O 2 -satuated 0.5 mol/l H 2 SO 4 and comparisons of ORR mass activities and specific activity of catalysts. (4.8 ma/mg) is as high as 2.2 times of that of the Pt/CNTs (2.3 ma/mg), this indicates that the Pt C/CNTs has a higher ORR activity compared to the Pt/CNTs. Xin et al. addressed that a high and homogeneous platinum distribution in the Pt/MWNT catalyst could result in an enhanced interaction between Pt and MWNTs and thus lead to a higher mass activity. 13 It is evidence that the core shell structural C/CNTs support is beneficial to load and disperse the Pt nanoparticles compared to the pure CNTs. To further substantiate the difference in degradation between the Pt C/CNTs and Pt/CNTs, both the catalysts were investigated by ADT. The retained ECSA, normalized with the initial one, is plotted as a function of cycle numbers in Figure 6. It clearly demonstrates that the ECSA loss of these two catalysts decreases with increasing the number of cycles under the same test conditions. Although some loss of ECSA on the Pt C/CNTs is observed, it shows higher retention of ECSA compared with the Pt/CNTs catalyst. After 2000 cycles, 46% of the initial ECSA for the Pt C/CNTs remains, which is higher than that of the Pt/CNTs (40%). It indicates that Pt on the C/CNTs support is more stable than that on CNTs under the test conditions. The improvement in lifetime of the Pt C/CNTs catalyst can be attributed to the presence of carbon shell. This is because the outer carbon shell is much less graphitized than the CNTs and thus is propitious to the Pt nanoparticles loaded onto CNTs. Thus the Pt 6932 J. Nanosci. Nanotechnol. 14, , 2014

5 Carbon-Embedded Carbon Nanotubes as Supports of Polymer Electrolyte Membrane Fuel Cell Catalysts Pt relative surface area (vs initial) Pt/CNTs Pt/C-CNTs Number of potential circles Figure 6. Loss of ECSA of catalysts with the increased potential cycles for Pt/CNT and Pt C/CNTs catalysts. nanoparticles can be prevented from aggregating with each other, and tightly anchored on the surface of supports by the adsorption interaction between Pt and carbon shells. 4. CONCLUSIONS This study provided an effective method to synthesize core shell structural C/CNTs as the support of metal nanoparticles. This novel C/CNTs support can improve the distribution of the Pt nanoparticles, Delivered by and Publishing increasetechnology the catalytic activity, Pt loading as well IP: as the stability ofon: the Tue, 01 Apr :28:08 to: Electrochim. Main CID Acta is , 791 (2004). (JPP) catalyst. Both Cyclic voltammetry and Copyright: ADT tests American exhib- Scientific Publishers ited our novel Pt C/CNTs catalyst was electroactive and relatively stable. We believe that the presence of carbon shells makes Pt nanoparticles well dispersed on surfaces of the supports with a narrow size, and effectively prevent Pt nanoparticles from aggregating. These promise potential applications in fuel cells and other industrial field. Acknowledgments: This work was supported by the National Natural Science Foundation of China (NSFC) (No ), the Major National Basic Research Development Program of China (973 Program) (No. 2012CB215504), and the NSFC (No ). References and Notes 1. Z. W. Chen, D. Higgins, A. P. Yu, L. Zhang, and J. J. Zhang, Energy Environ. Sci. 4, 3167 (2011). 2. J. J.-Hae, C. M.-Soon, and K. J.-Bom, J. Nanosci. Nanotechnol. 12, 5412 (2012). 3. J. Prabhuram, T. S. Zhao, Z. X. Liang, and R. Chen, Electrochim. Acta 52, 2649 (2007). 4. P. D.-Hwan, J. Yukwon, O. Jinhee, P. Jooil, Y. S.-Ho, C. J.-Ho, and S. Y.-Gun, J. Nanosci. Nanotechnol. 12, 5669 (2012). 5. Y. Y. Shao, J. Liu, Y. Wang, and Y. H. Lin, J. Mater. Chem. 19, 46 (2009). 6. S. B. Yin, S. C. Mu, H. F. Lv, N. C. Cheng, M. Pan, and Z. Y. Fu, Appl. Catal. B: Environ. 93, 233 (2010). 7. R. Vinodh and D. Sangeetha, J. Nanosci. Nanotechnol. 13, 5522 (2013). 8. H. F. Lv, N. C. Cheng, T. Peng, M. Pan, and S. C. Mu, J. Mater. Chem. 22, 1135 (2012). 9. C. H. Yen, X. L. Cui, H. B. Pan, S. F. Wang, Y. H. Lin, and C. M. Wai, J. Nanosci. Nanotechnol. 5, 1852 (2005). 10. D. P. He, S. C. Mu, and M. Pan, Carbon 49, 82 (2011). 11. K. Y. Niu, J. Sun, J. Yang, and X. W. Du, Sci. Adv. Mater. 4, 463 (2012). 12. H. Y. Li, H. Wang, Y. L. Wu, X. Zhang, and J. P. Zheng, Sci. Adv. Mater. 5, 453 (2013). 13. W. Z. Li, C. H. Liang, W. J. Zhou, J. S. Qiu, Z. H. Zhou, G. Q. Sun, and Q. Xin, J. Phys. Chem. B 107, 6292 (2003). 14. X. Wang, W. Z. Li, Z. W. Chen, M. Waje, and Y. S. Yan, J. Power Sources 158, 154 (2006). 15. G. Wu, Y. S. Chen, and B. Q. Xu, Electrochem. Commun. 7, 1237 (2005). 16. D. P. He, S. C. Mu, and M. Pan, Int. J. Hydrogen Energy 37, 4699 (2012). 17. Y. Y. Shao, G. P. Yin, Y. Z. Gao, and P. F. Shi, J. Electrochem. Soc. 153, A1093 (2006). 18. K. I. Han, J. S. Lee, S. O. Park, S. W. Lee, Y. W. Park, and H. Kim, 19. T. M. Wu, Y. W. Lin, and C. S. Liao, Carbon 43, 734 (2005). 20. Y. Zhou, Z. Y. Qin, L. Li, Y. Zhang, Y. L. Wei, L. F. Wang, and M. F. Zhu, Electrochim. Acta 55, 3904 (2010). 21. S. Kalluri, A. A. Madhavan, P. A. Bhupathi, R. Vani, A. Paravannoor, A. S. Nair, S. Nagarajan, K. R. V. Subramanian, S. V. Nair, and A. Balakrishnan, Sci. Adv. Mater. 4, 1220 (2012). 22. M. C. Liu, X. L. Wu, C. L. Chen, Q. Wang, T. Wen, and X. K. Wang, Sci. Adv. Mater. 5, 1686 (2013). 23. M. D. Ellison and P. J. Gasda, J. Phys. Chem. 112, 738 (2008). 24. D. P. He, C. Zeng, C. Xu, N. C. Cheng, H. G. Li, S. C. Mu, and M. Pan, Langmuir 27, 5582 (2011). 25. Y. Y. Shao, G. P. Yin, J. Zhang, and Y. Z. Gao, Electrochim. Acta 51, 5853 (2006). 26. T. J. Schmidt, H. A. Gasteige, G. D. Stäb, P. M. Urban, D. M. Kolb, and R. J. Behm, J. Electrochem. Soc. 145, 2354 (1998). 27. L. C. Ciacchi, W. Pompe, and A. D. Vita, J. Phys. Chem. B 107, 1755 (2003). 28. S. Park, Y. Y. Shao, H. Y. Wan, P. C. Rieke, V. V. Viswanathan, S. A. Towne, L. V. Saraf, J. Liu, Y. Lin, and Y. Wang, Electrochem. Commun. 13, 258 (2011). 29. D. P. He, K. Cheng, H. G. Li, T. Peng, F. Xu, S. C. Mu, and M. Pan, Langmuir 28, 3979 (2012). Received: 30 November Accepted: 14 January J. Nanosci. Nanotechnol. 14, ,