Carbon Nanotubes Supported Pt-Ru-Ni as Methanol Electro-Oxidation Catalyst for Direct Methanol Fuel Cells

Carbon Nanotubes Supported Pt-Ru-Ni as Methanol Electro-Oxidation Catalyst for Direct Methanol Fuel Cells

Journal of Natural Gas Chemistry 16(2007)162–166 Article Carbon Nanotubes Supported Pt-Ru-Ni as Methanol Electro-Oxidation Catalyst for Direct Metha...

429KB Sizes 7 Downloads 115 Views

Journal of Natural Gas Chemistry 16(2007)162–166

Article

Carbon Nanotubes Supported Pt-Ru-Ni as Methanol Electro-Oxidation Catalyst for Direct Methanol Fuel Cells Fei Ye1 ,

Shengzhou Chen2 ,

Xinfa Dong1 ,

Weiming Lin1,2∗

1. School of Chemical and Energy Engineering, South China University of Technology, Guangzhou 510640, Guangdong, China; 2. School of Chemistry and Chemical Engineering, Guangzhou University, Guangzhou 510006, Guangdong, China [ Manuscript received November 27, 2006; revised February 27, 2007 ]

Abstract: Carbon nanotubes (CNTs) supported Pt-Ru and Pt-Ru-Ni catalysts were prepared by chemical reduction of metal precursors with sodium borohydride at room temperature. The crystallographic properties and composition of the catalysts were characterized by X-ray diffraction (XRD) and energy dispersive X-ray (EDX) analysis, and the catalytic activity and stability for methanol electro-oxidation were measured by electrochemical impedance spectroscopy (EIS), linear sweep voltammetries (LSV), and chronoamperometry (CA). The results show that the catalysts exhibit face-centered cubic (fcc) structure. The particle size of Pt-Ru-Ni/CNTs catalyst is about 4.8 nm. The catalytic activity and stability of the Pt-Ru-Ni/CNTs catalyst are higher than those of Pt-Ru/CNTs catalyst. Key words: carbon nanotubes; Pt-Ru-Ni/CNTs; methanol electro-oxidation; direct methanol fuel cells

1. Introduction

Methanol is a fundamental product of C1 chemistry industry. The production processes of methanol are simple, with rich recourses, such as natural gas, coal, heavy oil, and so forth [1]. Therefore methanol has many applications in the chemical industry. In recent years, direct methanol fuel cells (DMFCs) have attracted an increasing attention because of their favorable advantages, such as simple system structure, compatibility with current petroleum distribution network, high energy density, as well as low temperature operation [2−4]. It is suggested that among the various types of fuel cells, DMFCs show the most promising prospect for portable applications (laptops, PDAs, mobile phones, etc.). In operation, DMFCs oxidize methanol with water to form carbon dioxide, protons, and electrons. The carbon dioxide is then ∗

released at the anode [5]. CH3 OH + H2 O → CO2 + 6H+ + 6e− (Ea = 0.046 V) (1) At the cathode, the protons migrating through the electrolyte membrane, together with the electrons passing through the outer circuit combine with oxygen to form water. 3/2O2 + 6H+ + 6e− → 3H2 O(Ec = 1.23 V)

(2)

The overall reaction in a DMFC is the catalytic conversion of methanol with oxygen to carbon dioxide and water; with a maximum thermodynamic voltage of 1.18 V at 25 ℃. CH3 OH + 3/2O2 → CO2 + 2H2 O(Ecell = 1.18 V) (3) Although a great deal of effort has been made to the research and development of DMFCs, several problems remain to be resolved in terms of efficiency

Corresponding author. Tel: 020-87113023; E-mail: [email protected] The project is supported by the National Natural Science Foundation of China (20576023), the Science and Technology Project of Guangzhou City (2005 J1-C0361) and the Key Project of Education Bureau of Guangzhou City (2052).

Journal of Natural Gas Chemistry Vol. 16 No. 2 2007

and power density. One of the problems is the low activity of methanol electro-oxidation in the anode. It is well known that Pt is the most active metal for methanol electro-oxidation; however, Pt is easy to be poisoned by CO-like species produced during the methanol electro-oxidation and thus loses continuous high catalytic activity [5]. Therefore, Pt-based bimetal, ternary, or quaternary catalysts have been used to improve the CO-tolerance characteristics of Pt through the bifunctional or electronic effect [3]. Recently, Pt-Ru-Ni alloy or Vulcan XC-72 supported Pt-Ru-Ni catalysts have been reported to show higher activity and stability in comparison to state-of-the-art Pt-Ru catalysts [6−10]. Since the discovery of carbon nanotubes (CNTs) in 1991 [11], they have been widely used in many fields. In the fuel cell areas, they could be of interest as catalyst supports [12,13]. CNTs show highly electrochemically accessible surface area and offer a remarkable electronic conductivity compared with the commonly used Vulcan carbon black. Previous studies have showed that CNTs supported catalysts exhibit better performance of methanol electrooxidation as compared to conventional carbon black (XC-72) supported catalysts [14−17]. However, little research has been devoted to CNTs supported ternary catalysts [18]. This study presents the methanol electro-oxidation results on CNTs supported Pt-RuNi catalyst. 2. Experimental 2.1. Oxidative pretreatment of CNTs Well-aligned multi-walled CNTs with purity higher than 95% were purchased from Shenzhen Nanotech Port Co., Ltd., China. The main range of diameter, length, and surface area of the CNTs was 10−20 nm, 5−15 µm, and 40−300 m2 /g, respectively. In order to increase the concentration of grafting sites on the walls of CNTs, an oxidative pretreatment was performed by refluxing with a mixture of concentrated sulfuric and nitric acids (1:1 v/v, 98% and 70%, respectively) at 90 ℃ for 5 h. Following this, the CNTs were filtered and washed using ultra-pure water (18.23 MΩ) until the pH of the filtrate became 7 and consequently dried in an oven at 110 ℃ for 5 h. 2.2. Preparation of catalysts CNTs supported Pt-Ru (with an atomic ratio of

163

1:1) and Pt-Ru-Ni (with an atomic ratio of 6:3:1) catalysts were prepared by chemical reduction of H2 PtCl6 , RuCl3 , and NiCl2 precursors with sodium borohydride at room temperature. The metal loading of the two catalysts was 20% in weight. Appropriate amount of CNTs, metal precursors, and ultra-pure water were ultrasonically mixed for 30 min and then mechanically stirred for 2 h. Excess quantities of 0.2 M sodium borohydride solution were added drop-by-drop to the mixtures and then the bath was stirred for 3 h for the complete reduction of the metals. Finally, the mixtures were filtered, washed, and dried in an oven at 80 ℃ for 2 h. 2.3. Characterizations of catalysts X-ray diffraction (XRD) powder patterns of the CNTs and catalysts were obtained on a XD-3 X-ray diffractometer (Beijing Purkinje General Instrument Co., Ltd., China) using a Cu-Kα source operating at 36 kV and 20 mA. The scanning range and rate are 10o −90o and 8o /min, respectively. Chemical composition analyses of the catalysts were carried out on an energy dispersive X-ray (EDX) analyzer (Oxford INCA300) attached to a scanning electron microscope (LEO 1530 VP, Germany). 2.4. Electrochemical measurements The electrochemical measurements were performed in a solution of 0.5 M H2 SO4 and 1 M CH3 OH at room temperature, using a conventional threeelectrode cell and a Solartron SI 1287 electrochemical interface and SI 1260 impedance/gain-phase analyzer. A Pt mesh and a saturated calomel electrode (SCE, −0.241 V vs. NHE) were used as counterelectrode and reference electrode, respectively. PtRu/CNTs or Pt-Ru-Ni/CNTs modified glassy carbon (GC, Johnson Matthey) electrode was used as the working electrode. The GC electrode was polished by 1.0, 0.3, and 0.05 µm alumina (CHI Inc., USA), respectively, and then washed in ethanol and ultra-pure water ultrasonically. 3 mg catalyst and 1 ml solution (20% isopropanol+73.75% H2 O+6.25% Nafion (5 wt%, Fluka)) were mixed ultrasonically for 30 min. 10 µl slurry was pipetted onto the surface of the polished GC electrode. After the solvent evaporation, the working electrode was obtained. The apparent surface area of the GC electrode was 0.196 cm2 , and the specific loading of the catalyst was about 30 µgmetal /cm2 . The impedance spectra were reg-

Fei Ye et al./ Journal of Natural Gas Chemistry Vol. 16 No. 2 2007

164

istered at frequencies from 100 KHz to 0.05 Hz with an amplitude of 10 mV at a potential of 0.4 V vs. SCE, and ZPlot and ZView software were used to measure and analyze the impedance data. Linear sweep voltammetries (LSV) were plotted within a potential range from −0.241 to 0.50 V vs. SCE with a scanning rate of 5 mV/s, and the chronoamperometry (CA) profiles were obtained at a potential of 0.40 V vs. SCE with a polarization time of 30 min. 3. Results and discussion 3.1. Characterizations of catalysts Figure 1 shows the XRD patterns of CNTs support and Pt-Ru-Ni/CNTs and Pt-Ru/CNTs catalysts. The first peak located at about 25.5o in the three samples is associated with the CNTs support. Four characteristic peaks corresponding to (111), (200), (220), and (311) planes of the fcc crystalline Pt are observed in the catalysts’ patterns, and the corresponding peak angles a little shift to higher 2θ values of 39.76o, 46.24o, 67.45o , and 81.28o for pure Pt fcc, indicating that the alloy catalysts have singlephase disordered structures, and the lattice constants decrease because of Ru or Ni substitution in Pt fcccenter [9]. The average particle size may be roughly calculated from Pt (220) FWHM according to DebyeScherrer Equation [6,9,19]: L=

0.9λCuKα B2θ cos θmax

and Pt-Ru-Ni/CNTs are 5.6 and 4.8 nm, respectively. No peaks associated with either Ru or Ni metal or oxide species are observed, implying that Ru or Ni may enter Pt fcc-center to form Pt-Ru or Pt-Ru-Ni alloys, or partially present as oxide species but not clearly be discerned by X-ray diffraction. EDX measurement was used to analyze the composition of the catalysts. EDX spectra of Pt-RuNi/CNTs and Pt-Ru/CNTs are shown in Figure 2, and the calculated chemical composition of the two catalysts are listed in Table 1. It is evident that Pt, Ru, and Ni are present on the CNTs support, indicating that the H2 PtCl6 , RuCl3 , and NiCl2 precursors can be reduced to their respective metal phases by sodium borohydride at room temperature. According to the composition values in Table 1, the ratio of Pt:Ru for Pt-Ru/CNTs catalyst is 48.87:51.13, and the ratio of Pt:Ru:Ni for Pt-Ru-Ni/CNTs catalyst is 58.03:34.88:7.09, which are quite close to the theoretical values. Wang and co-workers [9] prepared Pt-Ru/C and Pt-Ru-Ni/C by the same method, and the ratio of Pt:Ru (theoretic value 1:1) for Pt-Ru/C catalyst was 56.3:43.7, and the ratio of Pt:Ru:Ni (theoretic value 6:3:1) for Pt-Ru-Ni/C was 63.2:30.1:6.7.

(4)

where, L is the average particle size, λCuKα is the X-ray wave-length (1.5406 ˚ A), B 2θ is the full width at half maximum, and θmax is the angle at peak maximum. The average particle sizes for Pt-Ru/CNTs

Figure 2. EDX spectra of Pt-Ru-Ni/CNTs (a) and Pt-Ru/CNTs (b) catalysts Table 1. The atomic compositions of the catalysts Composition (at.%)

Catalysts Figure 1. XRD patterns of CNTs support (1), and Pt-Ru/CNTs (2) and Pt-Ru-Ni/CNTs (3) catalysts

nominal Pt

Ru

determined by EDX Ni

Pt

Ru

Ni

Pt-Ru/CNTs

50

50



48.87

51.13



Pt-Ru-Ni/CNTs

60

30

10

58.03

34.88

7.09

Journal of Natural Gas Chemistry Vol. 16 No. 2 2007

3.2. Electrochemical measurements of the catalysts Electrochemical impedance spectroscopy (EIS) was used to determine the charge transfer resistance (Rct) during the methanol oxidation process on the Pt-Ru/CNTs and Pt-Ru-Ni/CNTs catalysts. Figure 3 shows the Nyquist plots obtained in a solution of 0.5 M H2 SO4 and 1 M CH3 OH at a potential of 0.40 V vs. SCE. It is interesting that for Pt-Ru-Ni/CNTs catalyst, an inductive loop appears at low frequencies, which could be attributed to the kinetics of the COad oxidation [20−22]. Using the ZView software and the equivalent circuits [22], the impedance data were fitted, and the values of various circuit elements were obtained. The Rct values are 231.3 and 135.3 Ω for Pt-Ru/CNTs and Pt-Ru-Ni/CNTs, respectively. A significant decrease of Rct values for Pt-Ru-Ni/CNTs indicates a smaller reaction resistance and higher catalytic activity for Pt-Ru-Ni/CNTs electrode.

165

the Pt-Ru-Ni/CNTs catalyst shows a higher current density at all potential than Pt-Ru/CNTs catalyst, indicating superior catalytic activity, which may be attributed to the promoting effect of Ni in Pt-RuNi/CNTs catalyst for methanol electro-oxidation.

Figure 4. Linear sweep voltammetries of methanol electro-oxidation on Pt-Ru-Ni/CNTs (1) and Pt-Ru/CNTs (2) catalysts Test conditions: solution 0.5 M H2 SO4 and 1 M CH3 OH, scan rate 5 mV/s, room temperature

Figure 3. Nyquist plots for Pt-Ru/CNTs (1) and Pt-Ru-Ni/CNTs (2) catalysts Test conditions: solution 0.5 M H2 SO4 and 1 M CH3 OH, potential 0.40 V vs. SCE, room temperature

The catalytic activities of Pt-Ru/CNTs and PtRu-Ni/CNTs were also analyzed by linear sweep voltammetries (LSV) with scanning from −0.241 to 0.50 V vs. SCE and a scan rate of 5 mV/s (Figure 4). Theoretically, methanol electro-oxidation may proceed at 0.04 V vs. NHE, but the potential for methanol electro-oxidation on Pt-based catalysts is much larger than the theoretical value due to the poison of Pt sites by the reaction intermediates, such as formaldehyde, formalic acid, carbon monoxide, and so on. According to Figure 4, the onset potential of methanol electro-oxidation on the Pt-RuNi/CNTs catalyst is approximately similar to that of Pt-Ru/CNTs, that is, 0.08 V vs. SCE. However,

The chronoamperometry (CA) profiles of methanol electro-oxidation on Pt-Ru/CNTs and PtRu-Ni/CNTs catalysts at a potential of 0.4 V vs. SCE are shown in Figure 5. At a potential of 0.4 V vs. SCE, methanol can be continuously electro-oxidized on the surfaces of Pt-based catalysts. However, the accumulation of reaction intermediates would appear if the removal reaction cannot be in line with that of methanol electro-oxidation. Therefore, the decrease in electro-oxidation current density will occur, and the decrease with less extent is indicative of better CO-resistance. For the two catalysts, the onset current densities, as well as, the decrease in electro-oxidation current density are different. The onset current densities at Pt-Ru/CNTs and Pt-RuNi/CNTs electrodes are about 0.00030 A/cm2 and 0.00045 A/cm2 , respectively. And Pt-Ru-Ni/CNTs have higher current densities throughout the test time, that is, the current density at Pt-Ru-Ni/CNTs electrode at 1800 s is almost twice larger than that at Pt-Ru/CNTs electrode. These results further demonstrate the improved catalytic activity and stability of Pt-Ru-Ni/CNTs catalyst in comparison to PtRu/CNTs catalyst. The improved catalytic activity and stability of Pt-Ru-Ni/CNTs for methanol electro-oxidation can be explained by the electronic effect and the presence of Ni(OH)2 and NiOOH species on the Pt-Ru-

Fei Ye et al./ Journal of Natural Gas Chemistry Vol. 16 No. 2 2007

166

Ni/CNTs catalyst [6,9]. The electronic transfer from Ni to Pt may contribute to the decrease of Pt-CO binding energy and the releasing of more Pt sites for continuous methanol electro-oxidation. The Ni hydroxide layer that is present on the Pt-Ru-Ni/CNTs catalyst surface may have favorable properties, such as proton and electronic conductivities, anticorrosion under methanol electro-oxidation conditions, and the facilitating oxidation of the poisoning CO-like species on Pt to CO2 , thus improving the catalytic activity and stability with respect to methanol electrooxidation, with the following reaction scheme for PtNi nanoparticles [6]: NiOOH + xPt-H −→ Ni(OH)2 + xPt Ni(OH)2 −→ NiOOH + H+ + e−

Figure 5. Chronoamperometry prof iles of methanol electro-oxidation on Pt-Ru-Ni/CNTs (1) and Pt-Ru/CNTs (2) Test conditions: solution 0.5 M H2 SO4 and 1 M CH3 OH, potential 0.40 V vs. SCE, room temperature

4. Conclusions CNTs supported Pt-Ru and Pt-Ru-Ni catalysts were prepared by chemical reduction of metal precursors with sodium borohydride at room temperature. The catalytic activities and stabilities of methanol electro-oxidation on the prepared catalysts were investigated by electrochemical impedance spectroscopy (EIS), linear sweep voltammetries (LSV), and chronoamperometry (CA). The results in this study indicate that the addition of Ni in Pt-Ru/CNTs greatly enhances the catalytic activity and stability of Pt-Ru-Ni/CNTs. The current density at the Pt-RuNi/CNTs electrode in chronoamperometry measurement at 1800 s is almost twice larger than that at Pt-Ru/CNTs electrode. The promoting effect of Ni could be ascribed to the electronic effect and the fa-

vorable properties of Ni hydroxide species on the PtRu-Ni/CNTs catalyst. Therefore, Pt-Ru/CNTs catalyst can be used as an efficient anode catalyst for direct methanol fuel cells. References [1] Tijm P J A, Waller F J, Brown D M. Appl Catal A: General, 2001, 221(1-2): 275 [2] Wasmus S, Kuver A. J Electroanal Chem, 1999, 461(1,2): 14 [3] Arico A S, Srinivasan S, Antonucci V. Fuel Cells, 2001, 1(2): 133 [4] Liu H, Song C, Zhang L, Zhang J, Wang H, Wilkinson D P. J Power Sources, 2006, 155(2): 95 [5] Hogarth M P, Ralph T R. Platinum Met Rev, 2002, 46(4): 146 [6] Park K W, Choi J H, Kwon B K, Lee S A, Sung Y E, Ha H Y, Hong S A, Kim H, Wieckowski A. J Phys Chem B, 2002, 106(8): 1869 [7] Choi J H, Park K W, Kwon B K, Sung Y E. J Electrochem Soc, 2003, 150(7): A973 [8] Wang Z B, Yin G P, Shi P F, Sun Y C. Electrochem Solid -State Lett, 2006, 9(1): A13 [9] Wang Z B, Yin G P, Zhang J, Sun Y C, Shi P F. Electrochim Acta, 2006, 51(26): 5691 [10] Martinez-Huerta M V, Rojas S, Gomez de la Fuente J L, Terreros P, Pena M A, Fierro J L G. Appl Catal B: Environ, 2006, 69(1-2): 75 [11] Iijima S, Yudasaka M, Yamada R, Bandow S, Suenaga K, Kokai F, Takahashi K. Chem Phys Lett, 1999, 309(3,4): 165 [12] Wang S, Sun G, Wang G, Zhou Z, Zhao X, Sun H, Fan X, Yi B, Xin Q. Electrochem Commum, 2005, 7(10): 1007 [13] Tian Z Q, Jiang S P, Liang Y M, Shen P K. J Phys Chem B, 2006, 110(11): 5343 [14] Rajesh B, Thampi K R, Bonard J M, Viswanathan B. J Mater Chem, 2000, 10(8): 1757 [15] Han K I, Lee J S, Park S O, Lee S W, Park Y W, Kim H. Electrochim Acta, 2004, 50(2-3): 791 [16] Lin Y H, Cui X L, Yen C H, Wai C M. Langmuir, 2005, 21(24): 11474 [17] Prabhuram J, Zhao T S, Tang Z K, Chen R, Liang Z X. J Phys Chem B, 2006, 110(11): 5245 [18] Liao S J, Holmes K A, Tsaprailis H, Birss V I. J Am Chem Soc, 2006, 128(11): 3504 [19] Radmilovic V, Gasteiger H A, Ross P N. J Catal, 1995, 154(1): 98 [20] Muller J T, Urban P M, Holderich W F. J Power Sources, 1999, 84(2): 157 [21] Sugimoto W, Aoyama K, Kawaguchi T, Murakami Y, Takasu Y. J Electroanal Chem, 2005, 576(2): 215 [22] Ocampo A L, Miranda-Hernandez M, Morgado J, Montoya J A, Sebastian P J. J Power Sources, 2006, 160(2): 915