Characteristics of Pt-Ru Catalyst Supported on Activated Carbon for ...

3 downloads 0 Views 4MB Size Report
Sep 3, 2003 - Jae-Hoon Junga, Seong-Hwa Honga,. Dong-Hyun Pecka, Dong-Ryul Shina and Eui-sik Kimb. aAdvanced Fuel Cell Research Center, Korea ...
&DUERQ 6FLHQFH

Vol. 4, No. 3 September 2003 pp. 121-125

Characteristics of Pt-Ru Catalyst Supported on Activated Carbon for Direct Methanol Fuel Cell Doo-Hwan Junga, N , Jae-Hoon Junga, Seong-Hwa Honga, Dong-Hyun Pecka, Dong-Ryul Shina and Eui-sik Kimb a

Advanced Fuel Cell Research Center, Korea Institute of Energy Research, Daejeon, Korea b Department of Chemical Engineering, Chungbuk National University, Chungju, Korea N e-mail: [email protected] (Received July 9, 2003; Accepted September 9, 2003)

Abstract The Pt-Ru/Carbon as an anode catalyst supported on the commercial activated carbon (AC) having high surface area and micropore was characterized for application of Direct Methanol Fuel Cell (DMFC). The Pt-Ru/AC anode catalyst used in this experiment showed the performance of 600 mA/cm2 current density at 0.3 V. The borohydride reduction process using NaBH4, denoted as a process A, showed much higher current and power densities than process B prepared by changing the reduction and washing process of process A. The particle sizes are strongly affected by the reduction process than the specific surface area of raw active carbon and the sizes are almost constant when the specific surface area of carbon are over than the 1200 m2/g. Smaller particle size of catalyst and more narrow intercrystalite distance increased the performance of DMFC. Keywords : activated carbon, electrodes, catalyst support, direct methanol fuel cell, surface areas

1. Introduction The DMFC using a Pt/carbon (Pt/C) and Pt-Ru/carbon (PtRu/C) catalyst systems in cathode and anode, respectively, has been considered as a reasonable power source for portable electronic device and electric vehicle [1-3]. Ongoing commercialization of DMFC is creating considerable development activity for improvement of the catalyst layers of methanol oxidation electrode. A primary role of the carbon support is to provide electrical connection between the widely dispersed Pt and Pt-Ru catalyst nano-particles and the porous backing material such as carbon cloth or paper. And it is well known that the performance of DMFC is strongly influenced by the metal catalyst in membrane electrode assembly (MEA) [4-6]. And the dispersion of catalyst, ion conductivity, and electric conductivity are also important to the performance of MEA [7]. In spite of an anticipated future of the DMFC, much higher performance is still expected through the control of the nano-scale structure for MEA, especially its catalyst system. For these reasons, the carbon support materials have been widely studied to improve the performance of electrodes for the DMFC [8-12]. Generally, the carbon black such as Vulcan XC-72 has been used as support materials [13-14]. Carbon is impermeable to gases and water and does not conduct protons, which limits achievable performances. In order to reduce activation polarization thereby obtain maximum utilization of the

catalyst, a carbon material such as a carbon black is usually mixed with the catalyst. In this work, a single cell performance of DMFC with difference anode electrode prepared with the high surface area activated carbons were examined by changing the preparation method of anode catalysts of DMFC. Anode catalysts were prepared with modified Brohydride process suggested at KIER by changing a reduction procedure of the process. The particle size of metal catalyst, specific surface area of carbon before and after the catalyst supporting material, were evaluated using a X-ray diffraction (XRD), BrunauerEmmett-Teller (BET) method, SEM, TEM.

2. Experiments The Pt-Ru/C (atomic ratio of Pt/Ru=1:1) catalysts used in this study were prepared through the different reduction and washing process. In the Pt-Ru/C catalysts, the amount of PtRu metal was loaded 60 wt% on activated carbon. The precess A and B are along the reduction reaction by borohydride as shown in Figure 1. The H2PtCl6·6H2O and RuCl3·xH2O (Alfa Aesar) were used as precursors of catalysts. Some properties of the active carbons (BP15, BP20, BP25, Kuraray Chem. Co.) as support materials are shown in Table 1. The preparation process of catalyst was performed in a glove box of Ar atmosphere and controlled to

122

D.-H. Jung et al. / Carbon Science Vol. 4, No. 3 (2003) 121-125

Fig. 2. Schematic diagram of single cell assembly used in this experiment.

Fig. 1. Preparation processes of Pt-Ru/AC anode catalysts. Table 1. Properties of as-received active carbons used in this study BP15 BP20 BP25 Amount of iodine adsorption (mg/g) Density (g/ml) Average particle size (µm) Specific surface area (measured) (m2/g)

1700 0.54 5.31 1500

2050 0.45 5.25 2300

2440 0.30 1.97 2700

be Pt and Ru atomic ratio of 1:1. In the process A, RuCl3·xH2O was dissolved in the distilled water after dispersion of the active carbon in the water with controlling of pH within 2-4. The H2PtCl6·6H2O was added subsequently into the solution. The reduction of the mixed solution in process A was conducted by dropping slowly the solution in the 0.5 M NaBH4 for 1 hr. After the reduction process, the mixed solution was washed using soxhlet by distilled water for 2 hours, and subsequently dried at 80oC for 24 hours, as

shown in Figure 1. In process B, the mixed solution was reduced by dropping slowly 0.5 M NaBH4 into the mixed solution for 1 hr as shown in Figure 1. 5 w/o of Teflon emulsion and 30 w/o of Nafion solution were add subsequently to the catalyst for preparing the anode. The solution was stirred for 24 hr for preparing catalyst slurry. The slurry was coated using brush on the pretreated carbon paper and dried at the room temperature. The amount of metal catalyst was controlled to 3 mg/cm2. An Pt-black from Johnson Matthey and the 7 wt% Nafion solution was used in the cathode catalyst The cathode was also prepared through the stirring, brushing on the carbon paper and drying. The amount of cathode catalyst was controlled to 5-7 mg/cm2. Figure 2 shows a schematic diagram of single cell assembly. The gaskets, bipolar plates, current collectors, and stainless steel end plates are fixed subsequently on both sides of MEA. The MEA was fabricated by using early-denoted Jung [15]. Figure 3 shows a schematic diagram of cell test apparatus. 2 M methanol was fed into the node. Pure oxygen was used in the cathode. The Pt-Ru/active carbon (Pt-Ru/AC) was characterized using XRD (Rigaku, RINT 2000) and TEM (Carl Zeiss, EM912 Omega). The crystal size of metal catalyst is calculated from Scherrer equation, Dn(nm)=0.9γ /(Bcosθ) (here, γ is wave length of X-ray beam, B is the angular radian width of the diffraction peak at half-maximum intensity, and θ is the Bragg angle appeared around 2θ=68o. The surface properties of activated carbon and prepared catalysts were investigated through the BET method using a surface analyzer (Micromeritics, ASAP 2010).

Characteristics of Pt-Ru Catalyst Supported on Activated Carbon for Direct Methanol Fuel Cell

123

Fig. 4. XRD patterns of 60 wt% Pt-Ru/AC catalysts prepared through the process A. Fig. 3. Schematic diagram of cell test apparatus.

3. Results and Discussions Fig. 4 and 5 show XRD patterns of 60 w/o Pt-Ru/AC catalysts prepared through the process A and B, respectively. Generally, crystal size of Pt(220) particle is determined by the fcc (face-centered cubic lattice) peak appeared around 2θ=68o, because this peak do not overlap with the peak of Ruthenium. The peaks around 2θ=68o of samples prepared through the process A showed broad patterns than those of the process B. Table 2 shows the catalyst sizes and specific surface areas examined by BET method after preparation of Pt-Ru/C anode catalysts through the preparation process A and B, respectively. The surface areas of as-received activated carbons subscribed as BP15, BP20, and BP25 showed 1506, 2300, and 2700 m2/g, respectively. The surface areas of both supported catalysts were decreased drastically after the reduction process. The specific surface areas of catalysts such as BP-15-A, BP-20-A and BP-25-A showed drastic decrease of 65, 76, and 80%, respectively, after preparation

Fig. 5. XRD patterns of 60 wt% Pt-Ru/AC catalysts prepared through the process B.

of catalyst. The particle sizes of Pt (220) prepared through the process A and B were about 2 nm and 5 nm, respectively, and increased slightly with increasing the specific surface

Table 2. Catalyst particle size and specific surface area examined by BET method after preparation of Pt-Ru/C anode catalysts through the preparation process A and B Material BP15-A BP20-A BP25-A BP15-B BP20-B BP25-B

BET Surface area (raw) 2

2

526 m /g (1506 m /g) 484 m2/g (2300 m2/g) 493 m2/g (2700 m2/g) 501 m2/g (1506 m2/g) 530 m2/g (2300 m2/g) 778 m2/g (2700 m2/g)

Average pore diameter

Particle size of Pt

Intercrystallite distance

2.5 nm 2.6 nm 2.2 nm 1.9 nm 1.8 nm 2.1 nm

1.9 nm 2.2 nm 2.4 nm 4.9 nm 5.1 nm 5.1 nm

11.06 nm 12.31 nm 12.81 nm 21.98 nm 24.47 nm 25.47 nm

124

D.-H. Jung et al. / Carbon Science Vol. 4, No. 3 (2003) 121-125

area of raw AC. The particle sizes of catalysts prepared through the process A were shown to be 1.9, 2.2, and 2.4 nm of the BP15, BP20, and BP25, respectively. The particle sizes of catalysts prepared through the process B were 4.9, 5.1, and 5.12 nm. As shown in the table, the difference of specific surface area of raw AC was about 1000 m2/g, the difference of particle size of Pt on active carbon was 0.5 nm in process A, and 0.2 nm in process B. The particle sizes are strongly affected by the reduction process than the specific surface area of raw active carbon. By the research results of Uchida [16], the specific surface area of AC was deeply affected on the particle size of Pt that was prepared with colloidal process. About 3.7 nm of Pt particle size was obtained when the specific surface area of AC was about 58 m2/g. But average about 1.0 nm of Pt particle sizes was obtained when they used AC at the range of 800-1500 m2/g. When the raw active carbon was produced different process, the size of Pt particle changed from 1.0 nm (KBEC, 800 m2/g) to 1.6 nm (AB18, 835 m2/g). But the surface of 1270 m2/g (KB600JD), the surface area of 1500 m2/g (No. 3950) and 1480 m2/g (Black pear2000) showed same Pt particles size with about 1.0 nm. It means that the particle size of Pt on active carbon is controlled by the combined effective of raw material and specific surface area of active carbon. The results of Table 2 and Uchida [16], reveals that the catalyst particle size are controlled by preparation process and the sizes are almost constant when the specific surface area of carbon are over than the 1200 m2/g. Figure 6 shows a TEM photograph of Pt-Ru/AC catalyst prepared through process A, BP25-A (60 wt%Pt-Ru/C).

Fig. 7. Performances of DMFC single cells examined at 90oC, 2 M methanol of anode fuel (2 cc/min), and pure oxygen of cathode feed (300 cc/min).

Although partially agglomerated catalyst particles could be find owing to the some high loading of Pt-Ru on active carbon (60 wt%Pt-Ru/C), good dispersion of the catalyst having the size of under 2 nm was observed on the support material. Figure 7 shows the I-V characteristics of DMFC single cell examined with 2 M methanol and pure of cathode with anode and cathode fuel at 90oC, 2 atm. As shown in the figure, the single cell performance by using the catalyst prepared through the process A showed higher value about 2-3 times than that of the process B. It means that the catalyst prepared by process A shows higher utilization than the process B. Some aspect, utilization of catalyst on carbon can be expresses as the intercrystallite distance from particle to particle proposed by Uchida [16]. The mean intercrystallite distance of PtRu was calculated as following. 1 x = --3

Fig. 6. TEM photograph of the Pt-Ru/AC catalyst (BP25-A).

3 πρ d Sc ( 100−y )/y 3

Where x is mean intercrystallite distance (nm), ρ is the density of PtRu (16.09 × 10−21 g/nm3), d is the mean particle diameter of Pt (nm), Sc is the specific surface area of carbon (nm2/g), and y is the Pt-Ru content (w/o). By the above equation, the intercrystallite distance (nm) was about 11.06 nm with process A and 21.98 nm with process B, respectively. In the process A, the intercrystallite distance (nm) was 11.06 when BP15 was used as catalyst supporting material. BP20 and BP25 showed intercrystallite distance (nm) of 12.31 nm and 12.81 nm, respectively. From the results of Figure 6, decreasing of intercrystallite distance increased the performance of DMFC single cell. The same result was obtained in the process B. Figure 8 shows the relationship between the particle size of Pt-Ru/AC catalyst and power density of DMFC single cell at 0.3 V. The highest performance was shown to be 194 mW/

Characteristics of Pt-Ru Catalyst Supported on Activated Carbon for Direct Methanol Fuel Cell

125

surface area of carbon is over than the 1200 m2/g. Smaller particle size and more narrow intercrystallite distance of catalyst increased the performance of DMFC.

References

Fig. 8. Relationship between the particle size of Pt-Ru/AC catalyst and power density of DMFC single cell at 0.3 V.

cm2 at the BP 15-A. The power density depended on the catalyst particle size in each processes. However, The catalyst of BP-15-B showed the second performance to be 160 mW/cm2 in this experiment. Therefore, from the results of Figure 7 and Figure 8, it could be say that much smaller particle size of catalyst and more narrow intercrystalite distance would be increased the performance of DMFC.

4. Conclusion The Pt-Ru/AC catalyst for direct methanol fuel cell using modified sodium borohydride method was developed. The particle sizes of catalyst are strongly affected by the reduction process than the specific surface area of raw active carbon. The particle sizes are almost constant when the specific

[1] Wasmus, S.; Kuver, A. J. Electroanal. Chem. 1999, 461, 14. [2] Dohle, H.; Schmitz, H.; Bewer, T.; Mergel, J.; Stolten, D. J. Power Sources 2002, 106, 313. [3] Antoloni, E. Mater. Chem. Phys. 2003, 78, 563. [4] Lizcano-Valbuena, W. H.; Paganin, V. A.; Gonzalez, E. R. Electrochim. Acta 2002, 47, 3715. [5] Arico, A. S.; Antonucci, P. L.; Modica, E.; Baglio, V.; Kim, H.; Antonucci, V. Electrochim Acta 2002, 47, 3723. [6] Lim, C.; Wang, C. Y. J. Power Sources 2003, 113, 145. [7] Muller, J. T.; Urban, P. M. J. Power Sources 1999, 84, 157. [8] Li, W.; Liang, C.; Qiu, J.; Zhou, W.; Han, H.; Wei, Z.; Sun, G.; Xin, Q. Carbon 2002, 40, 91. [9] Liu, Y. C.; Qiu, X. P.; Huang, Y. Q.; Zhu, W. T. J. Power Source 2002, 111, 160. [10] Fuente, A. M.; Pulgar, G.; Gonzalez, F.; Pesquera, C.; Blanco, C. Appl. Catal. 2001, 208, 35. [11] Aksoylu, A. E.; Madalena, M.; Freitas, A.; Fernando, M.; Pereira, R.; Figueiredo, J. L. Carbon 2001, 39, 175. [12] Ramesh, K. V.; Sarode, P. R.; Vasudevan, S.; Shukla, A. K. J. Electroanal. Chem. 1987, 223, 91. [13] Goodenough, J. B.; Hamnett, A.; Kennedy, B. J.; Manoharan, R.; Weeks, S. A. J. Electroanal. Chem. 1988, 240, 33. [14] Watanabe, M.; Uchida, M.; Motoo, S. J. Electroanal. Chem. 1987, 229, 395. [15] Jung, D. H.; Lee, C. H.; Kim, C. S.; Shin, D. R. J. of Power Sources 1998, 71, 169. [16] Uchida, M.; Fukuoka, Y.; Sugawara, Y.; Eda, N.; Ohta, A. J. Electrochem. Soc. 1996, 43, 2245.