- Email: [email protected]
Carbon-supported PdSn nanoparticles as catalysts for formic acid oxidation
Electrochemistry Communications 11 (2009) 1667–1670
Contents lists available at ScienceDirect
Electrochemistry Communications journal homepage: www.elsevier.com/locate/elecom
Carbon-supported PdSn nanoparticles as catalysts for formic acid oxidation Zhaolin Liu *, Xinhui Zhang Institute of Materials Research and Engineering, 3 Research Link, Singapore 117602, Singapore
a r t i c l e
i n f o
Article history: Received 2 June 2009 Received in revised form 22 June 2009 Accepted 23 June 2009 Available online 26 June 2009 Keywords: Carbon-supported PdSn nanoparticles Palladium nanoparticles Catalytic activity Direct formic acid fuel cell
a b s t r a c t Pd and PdSn nanoparticles supported on Vulcan XC-72 carbon are prepared by a microwave-assisted polyol process. The catalysts are characterized by transmission electron microscopy (TEM), X-ray diffraction (XRD), linear sweeping voltammetry, and chronoamperometry. The results show that the Pd and PdSn nanoparticles, which are uniformly dispersed on carbon, are 2–10 nm in diameters. All Pd/C and PdSn/ C catalysts display the characteristic diffraction peaks of a Pd face-centered cubic (fcc) crystal structure. It is found that the addition of Sn to Pd can increase the lattice parameter of Pd (fcc) crystal. The PdSn/C catalysts have higher electrocatalytic activity for formic acid oxidation than a comparative Pd/C catalyst and show great potential as less expensive electrocatalyst for formic acid electrooxidation in DFAFCs. Ó 2009 Elsevier B.V. All rights reserved.
1. Introduction During the past 20 years, direct methanol fuel cells (DMFCs) have been widely studied and considered as possible power sources for the portable electronic devices and electric vehicles. These fuel cells offer a variety of benefits such as high specific energy and the ready availability and portability of methanol. On the other hand, the problem of methanol ‘crossover’ from the anode to the cathode through membrane leads to low system efficiency. Methanol crossover prevents utilization of high concentration of methanol; the limits generally less than 2 M. The feasibility of direct formic acid fuel cells (DFAFCs) based on proton exchange membranes has been demonstrated by Masel and co workers [1,2]. Hsing and co workers [3] have also reported that the rate of fuel crossover can be reduced by a factor of 5 and thus a higher performance can be obtained when formic acid is used in place of methanol under the same conditions. Carbon-supported Pt catalysts for electrooxidation of formic acid are poisoned severely by the adsorbed CO intermediate of the reaction [4,5]. It has been demonstrated [6,7] that PtRu and PtPd alloys can diminish this CO poisoning effect to some extent, but it still limits significantly the catalytic activity for formic acid oxidation. Recently, Masel co workers [8] have disclosed that unsupported Pd and Pd/C catalysts can overcome CO poisoning effect and thereby yield high performances in the DFAFCs. In order to further improve the electrocatalytic performance of the Pd and Pd/ C catalysts, the Pd-based bimetallic catalysts, such as Pd–Ni [9], Pd–Au [10], Pd–Pt [11] and Pd–Ir [12], have been investigated. * Corresponding author. Tel.: +65 68727532; fax: +65 68720785. E-mail addresses: [email protected], [email protected] (Z. Liu). 1388-2481/$ - see front matter Ó 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.elecom.2009.06.023
However, to the best of our knowledge, PdSn bimetallic catalysts for formic acid electrooxidation have not been reported yet. In this paper, PdSn nanoparticles are synthesized by a simple microwave-assisted polyol procedure and deposited on carbon to produce carbon-supported PdSn catalysts, aiming to have a less expensive electrocatalyst in the DFAFC. The physicochemical properties and electrochemical activities of the nanoparticles for formic acid oxidation are investigated. The reasons of oxidation activity enhancement for PdSn catalysts are discussed in detail. 2. Experimental The PdaSnb/carbon black (Cabot Vulcan XC-72, subscript denotes the atomic percentage of the alloying metal) catalysts were prepared by microwave heating of ethylene glycol (EG) solutions of PdCl2 and SnCl22H2O. The Pd content in each sample was 20 wt.%. 1.46, 1.95, 2.93 and 5.85 mL of 0.02 M SnCl22H2O (Aldrich, A.C.S. Reagent) and 5.85 mL of 0.02 M PdCl2 (Aldrich, A.C.S. Reagent) were chosen to yield Pd4Sn1/C, Pd3Sn1/C, Pd2Sn1/C and Pd1Sn1/C, respectively. A typical preparation of Pd2Sn1/C catalyst would consist of the following steps: 5.85 mL of 0.02 M PdCl2 and 2.93 mL of 0.02 M SnCl22H2O was mixed with 30 ml of ethylene glycol (Mallinckrodt, AR). 0.5 mL of 0.8 M NaOH was added dropwise. About 0.05 g of Vulcan XC-72 carbon was added to the mixture and sonicated. The solution was placed in a CEM ‘‘Discover” microwave reactor (CEM Corporation) with the maximum temperature set at 170 °C at atmospheric conditions for 30 s. The resulting suspension was filtered; and the residue was washed with acetone and dried at 100 °C over night in a vacuum oven. For comparison, Pd/C catalyst (20 wt.% Pd loading) was also prepared using the same method.
1668
Z. Liu, X. Zhang / Electrochemistry Communications 11 (2009) 1667–1670
To determine the actual palladium and tin contents in the PdSn alloys, inductively coupled plasma spectroscopy (ICP) was used to measure the unreacted metal ions remaining in the ethylene glycol mixtures. The PdSn alloy nanoparticles of compositions of Pd4Sn1.05, Pd3Sn1.09, Pd2Sn1.03, and Pd1Sn1.06 were obtained from precursors of Pd4Sn1, Pd3Sn1, Pd2Sn1, and Pd1Sn1. The palladium and tin contents in the PdSn alloy nanoparticles were about the same as that in the initial mixtures. The catalysts were examined by TEM on a JEOL JEM 2010. For microscopic examinations the samples were first ultrasonicated in acetone for 1 h and then deposited on 3 mm Cu grids covered with a continuous carbon film. X-ray diffraction (XRD) patterns were recorded by a Bruker GADDS diffractometer with area detector using a CuKa source (k = 1.54056 Å) operating at 40 kV and
Pd/C Pd4Sn1/C Pd3Sn1/C Pd2Sn1/C
Intensity (a. u.)
Pd(220)
Pd(311)
Pd(200)
b Pd(220)
Intensity (a. u.)
C(002)
Pd(111)
a
40 mA. The samples were prepared by depositing carbon-supported nanoparticles on a glass slide. An AUTOLAB potentiostat/galvanostat and a conventional three-electrode test cell were used for electrochemical measurements. The working electrode was a thin layer of Nafion-impregnated catalyst cast on a vitreous carbon disk held in a Teflon cylinder. The catalyst layer was obtained in the following way: (i) a slurry was first prepared by sonicating for 1 h a mixture of 0.5 ml of deionized water, 13 mg of Pd/C or PdSn/C catalyst, and 0.2 ml of Nafion solution (Aldrich: 5 wt.% Nafion); (ii) 4 ll of the slurry was pipetted and spread on the carbon disk; (iii) the electrode was then dried at 90 °C for 1 h and mounted on a stainless steel support. The surface area of the vitreous carbon disk was 0.25 cm2. Pt gauze and an Ag/AgCl electrode were used as the
Pd/C Pd4Sn1/C Pd3Sn1/C Pd2Sn1/C Pd1Sn1/C
Pd1Sn1/C
10
20
30
40
50
60
70
80
90
63
100
65
67
2 Theta (deg.)
69
71
73
2 Theta (deg.)
c
d
e
f
25
35 30
20 Frequency (%)
Frequency (%)
25 15
10
20 15 10
5 5 0
0 1
2
3
4
5
6
7
Particle size (nm)
8
9
10
1
2
3
4
5
6
7
8
9
10
Particle size (nm)
Fig. 1. (a) XRD patterns of PdSn/C catalysts, (b) an expanded view of the (2 2 0) reflections of the fcc phase, (c) TEM image of Pd4Sn1/C catalyst, (d) TEM image of Pd2Sn1/C catalyst, (e) particle size distribution of Pd4Sn1/C catalyst, and (f) particle size distribution of Pd2Sn1/C catalyst.
1669
Z. Liu, X. Zhang / Electrochemistry Communications 11 (2009) 1667–1670
3. Results and discussion The powder XRD patterns for PdSn/C are shown in Fig. 1a alongside the diffraction patterns of a Vulcan carbon-supported Pd catalyst used as comparison. The diffraction peak at 20–25° observed in all the XRD patterns of the carbon-supported catalysts is due to the (0 0 2) reflection of the hexagonal structure of Vulcan XC-72 carbon. The diffraction peaks at about 40, 47, 68 and 82 are due to the Pd (1 1 1), (2 0 0), (2 2 0), and (3 1 1) reflections, respectively, which represents the typical character of a crystalline Pd face-centered cubic (fcc) phase. There are not other distinct reflection peaks in all spectra than those of the four peaks mentioned above, indicating that these electrocatalysts have prevailed Pd (fcc) crystal structure. The (2 2 0) reflections of Pd are used to calculate the average particle size according to the Scherrer equation [13]:
second peak near 0.6 V can be mainly attributes to formic acid oxidation via the CO pathway [14,15]. The peak current density near 0.2 V increased as the Sn content in the anode catalysts increased.
a
Pd/C Pd4Sn1/C
350
Pd3Sn1/C
300
Pd2Sn1/C
250
Pd1Sn1/C Sn/C
200 150 100 50 0 -50 -0.3
0
0.3
0.6
0.9
1.2
E (V vs. Ag/AgCl)
b
70 60
L ¼ 0:9kK a1 =ðB2h cos hB Þ
Pd2Sn1/C
i (A/g Pd)
50
Pd1Sn1/C
Pd3 Sn1 /C
40
Pd4Sn1/C
30 Pd/C
20 10 0
500
1000
1500
2000
2500
Time (s)
c
100 Pd/C Pd4Sn1/C
80
Pd3Sn1/C
i (A/g Pd)
where L is the average particle size, kKa1 is the X-ray wavelength (1.54056 Å for Cu Ka1 radiation), B2h is the peak broadening and hB is the angle corresponding to the peak maximum. An expanded view of the (2 2 0) reflections of the fcc phase is shown in Fig. 1b. The average particle size of Pd (5.2 nm) obtained from XRD patterns was slightly higher than that of Pd4Sn1 (4.8 nm), Pd3Sn1 (4.1 nm), Pd2Sn1 (3.7 nm) and Pd1Sn1 (3.2 nm). The particle size of PdSn alloy nanoparticles obtained by TEM analysis falls in the order Pd (5.5 nm) > Pd4Sn1 (4.8 nm) > Pd3Sn1 (4.1 nm) > Pd2Sn1 (3.7 nm) > Pd1Sn1 (3.2 nm). The values of the average particle size obtained by TEM analysis are almost in good agreement with those calculated from the XRD results, but the former are always higher. The lattice parameter of the (fcc) Pd is estimated to be 3.911 Å. It was found that the addition of Sn to Pd can increase the lattice parameter of Pd (fcc) crystal (in the order of Pd4Sn1 (3.925 Å) < Pd3Sn1 (3.949 Å) < Pd2Sn1 (3.955 Å) < Pd1Sn1 (3.963 Å)) inducing the (2 2 0) reflection peak shift to lower position. The typical TEM images of the Pd4Sn1/C and Pd2Sn1/C catalysts are shown in Fig. 1c and d. It can be seen that a remarkably uniform and high dispersion of metal particles on the carbon surface with an average diameter of 4.8 and 3.7 nm for Pd4Sn1/C and Pd2Sn1/ C, respectively. Evidently, Pd4Sn1 and Pd2Sn1 nanoparticles synthesized by a microwave-assisted polyol process present well-dispersed particles on Vulcan XC-72 and relatively narrow particle size distributions as shown in Fig. 1e and f. Fig. 2a shows the linear sweeping voltammograms of 1 M HCOOH in 0.5 M H2SO4 solution at the different catalysts. There was no significant feature difference between the linear sweeping voltammogramms of room temperature formic acid on carbon supported Pd and carbon supported PdSn catalysts. Obviously, no anodic peak was observed at the Sn/C catalyst, indicating that the Sn/C catalyst has no electrocatalytic activity for the oxidation of formic acid. It can be observed that there are main peaks near 0.2 V and small peaks near 0.6 V. The first anodic peak at near 0.2 V corresponds to formic acid oxidation via the direct pathway, while the
450 400
i (A/g Pd)
counter and reference electrodes, respectively. All potentials in this report are quoted against Ag/AgCl. All electrolyte solutions were deaerated by high-purity argon for 2 h prior to any measurement. For linear sweeping voltammetry and chronoamperometry of formic acid oxidation, the electrolyte solution was 1 M formic acid in 0.5 M H2SO4, which was prepared from high-purity sulfuric acid, high-purity grade formic acid and distilled water. For the electrochemical measurement of the adsorbed CO, CO was bubbled into the solution for 10 min when the electrode potential was fixed at 0 V vs. Ag/AgCl. Then, Ar was bubbled into the solution for 10 min to remove CO in the solution.
Pd2Sn1/C
60
Pd1Sn1/C Sn/C
40
20
0 0
0.2
0.4
0.6
0.8
1
1.2
E (V vs. Ag/AgCl) Fig. 2. (a) Linear sweeping voltammograms of Pd/C and PdSn/C catalysts in 1 M HCOOH, 0.5 M H2SO4 with a scan rate of 50 mV s1 at room temperature. (b) Polarization current vs. time plots for the electrooxidation of formic acid in 1 M HCOOH, 0.5 M H2SO4 electrolyte at 0.3 V (vs. Ag/AgCl) at room temperature. (c) COstripping linear sweeping voltammograms on Pd/C and PdSn/C catalysts in 0.5 M H2SO4 with a scan rate of 50 mV s1 at room temperature.
1670
Z. Liu, X. Zhang / Electrochemistry Communications 11 (2009) 1667–1670
The Pd2Sn1/C catalyst was similar to Pd1Sn1/C catalyst. It is seen from Fig. 2a that the Pd2Sn1/C catalyst exhibits higher current density compared with the Pd1Sn1/C catalyst between 0.15 V and after 0.55 V. This suggests that the presence of tin in the PdSn nanoparticles catalyst markedly enhances the electrocatalytic activity of palladium towards formic acid oxidation. Pd/C and PdSn/C catalysts were biased at 0.3 V vs. Ag/AgCl and the changes in their polarization currents with time were recorded (Fig. 2b). The pattern of current decay was different for each catalyst. For the Pd/C catalyst, the current decayed continuously even after 0.5 h, supposedly because of catalyst poisoning by the chemisorbed carbonaceous species. The Pd2Sn1/C is able to maintain the highest current density for over 0.5 h among all the catalysts, indicating enhanced electrocatalytic activity compared with Pd/C and other PdSn/C catalysts. The comparative tests concluded that Pd2Sn1/C had the best electrocatalytic performance among all carbon supported Pd-based catalysts prepared in this paper. In order to evaluate the poisoning effect of CO, CO-stripping linear sweeping voltammograms of the adsorbed CO at the different catalysts are shown in Fig. 2c. The most notable difference between CO-stripping on PdSn/C catalysts and Pd/C catalyst is the negative shift of the CO oxidation peak in the firmer. At the Pd/C catalyst, a strong anodic peak of adsorbed CO is located at 0.76 V. However, at the PdSn/C catalysts, the anodic peaks of the adsorbed CO are located between 0.67 and 0.72 V, which are more negative than that at the Pd/C catalyst. This is an indication that addition of Sn is helpful to weakening the adsorption strength of CO on Pd through the interaction between Pd and Sn. It was observed that no anodic peak appears at the Sn/C catalyst, indicating that CO cannot be adsorbed on the Sn surface. Addition of Sn exhibits the ability of weakening the adsorptive bond of the reactive intermediates, such as COads and COOHads. The effect of Sn could have prevented the accumulation of poisoning-intermediates and more Pd sites were available for the direct HCOOH decomposition via the direct pathway to CO2. Thus, Sn addition results in a remarkable enhancement of the oxidation activity for formic acid. When the content of Sn in the PdSn/C catalyst is increased, the potential of the oxidation peak of the adsorbed CO is shifted more negatively and the current density of the anodic peak is decreased, suggesting that the increase in the content of Sn in the PdSn/C catalyst changes the adsorption
strength and amount of CO. However, the electrocatalytic activity of the Pd1Sn1/C catalyst would be decreased when the content of Sn in the PdSn/C catalysts is high. This may be due to the lack of electrocatalytic activity on Sn for the oxidation of formic acid. Only when the atomic ratio of Pd and Sn is suitable, such as Pd2Sn1/C catalyst, its electrocatalytic activity can be better than that of the other Pd/Sn mole ratio catalysts. 4. Conclusion Pd and PdSn nanoparticles supported on Vulcan XC-72 carbon have been prepared by a microwave-assisted polyol process. The Pd and PdSn particles are nanosized and have relatively narrow particle size distributions. X-ray diffraction patterns show that the Pd/C and PdSn/C catalysts display the characteristic diffraction peaks of a face-centered cubic Pd structure. It is found that the addition of Sn to Pd can increase the lattice parameter of Pd (fcc) crystal. The results of linear sweeping voltammetry and chronoamperometry show that the Pd2Sn1/C catalyst has better electrocatalytic activity than Pd/C and other Pd/Sn mole ratio catalysts. The Pd2Sn1/C catalyst shows great potential as less expensive electrocatalyst for formic acid electrooxidation in DFAFCs. References [1] C. Rice, S. Ha, R.I. Masel, P. Waszczuk, A. Wieckowski, T. Barnard, J. Power Sources 111 (2002) 83. [2] Y. Zhu, S. Ha, R.I. Masel, J. Power Sources 130 (2004) 8. [3] X. Wang, J.M. Hu, I.M. Hsing, J. Electroanal. Chem. 562 (2004) 73. [4] J. Jiang, A. Kucernak, J. Electroanal. Chem. 520 (2002) 64. [5] S. Park, Y. Xie, M.J. Weaver, Langmuir 18 (2002) 5792. [6] D. Capon, R. Parsons, J. Electroanal. Chem. 65 (1975) 285. [7] M. Arenz, V. Stamenkovic, T.J. Schmidt, K. Wandelt, P.N. Ross, N.M. Markovic, Phys. Chem. Chem. Phys. 5 (2003) 4242. [8] S. Ha, R. Larsen, Y. Zhu, R.I. Masel, Fuel Cells 4 (2004) 337. [9] T. Shobha, C.L. Aravinda, P. Bera, L.G. Devi, S.M. Mayanna, Mater. Chem. Phys. 80 (2003) 656. [10] L.A. Kibler, A.M. El-Aziz, D.M. Kolb, J. Mol. Catal. A: Chem. 199 (2003) 57. [11] R.S. Jayashree, J.S. Spendelow, J. Yeom, C. Rastogi, M.A. Shannon, P.J.A. Kenis, Electrochim. Acta 50 (2005) 4674. [12] X. Wang, Y. Tang, Y. Gao, T. Lu, J. Power Sources 175 (2008) 784. [13] B.E. Warren, X-ray Diffraction, Addison-Wesley, Reading, MA, 1996. [14] W.J. Zhou, J.Y. Lee, Electrochem. Commun. 9 (2007) 1725. [15] Z. Liu, L. Hong, M.P. Tham, T.H. Lim, H. Jiang, J. Power Sources 161 (2006) 831.
Contents lists available at ScienceDirect
Electrochemistry Communications journal homepage: www.elsevier.com/locate/elecom
Carbon-supported PdSn nanoparticles as catalysts for formic acid oxidation Zhaolin Liu *, Xinhui Zhang Institute of Materials Research and Engineering, 3 Research Link, Singapore 117602, Singapore
a r t i c l e
i n f o
Article history: Received 2 June 2009 Received in revised form 22 June 2009 Accepted 23 June 2009 Available online 26 June 2009 Keywords: Carbon-supported PdSn nanoparticles Palladium nanoparticles Catalytic activity Direct formic acid fuel cell
a b s t r a c t Pd and PdSn nanoparticles supported on Vulcan XC-72 carbon are prepared by a microwave-assisted polyol process. The catalysts are characterized by transmission electron microscopy (TEM), X-ray diffraction (XRD), linear sweeping voltammetry, and chronoamperometry. The results show that the Pd and PdSn nanoparticles, which are uniformly dispersed on carbon, are 2–10 nm in diameters. All Pd/C and PdSn/ C catalysts display the characteristic diffraction peaks of a Pd face-centered cubic (fcc) crystal structure. It is found that the addition of Sn to Pd can increase the lattice parameter of Pd (fcc) crystal. The PdSn/C catalysts have higher electrocatalytic activity for formic acid oxidation than a comparative Pd/C catalyst and show great potential as less expensive electrocatalyst for formic acid electrooxidation in DFAFCs. Ó 2009 Elsevier B.V. All rights reserved.
1. Introduction During the past 20 years, direct methanol fuel cells (DMFCs) have been widely studied and considered as possible power sources for the portable electronic devices and electric vehicles. These fuel cells offer a variety of benefits such as high specific energy and the ready availability and portability of methanol. On the other hand, the problem of methanol ‘crossover’ from the anode to the cathode through membrane leads to low system efficiency. Methanol crossover prevents utilization of high concentration of methanol; the limits generally less than 2 M. The feasibility of direct formic acid fuel cells (DFAFCs) based on proton exchange membranes has been demonstrated by Masel and co workers [1,2]. Hsing and co workers [3] have also reported that the rate of fuel crossover can be reduced by a factor of 5 and thus a higher performance can be obtained when formic acid is used in place of methanol under the same conditions. Carbon-supported Pt catalysts for electrooxidation of formic acid are poisoned severely by the adsorbed CO intermediate of the reaction [4,5]. It has been demonstrated [6,7] that PtRu and PtPd alloys can diminish this CO poisoning effect to some extent, but it still limits significantly the catalytic activity for formic acid oxidation. Recently, Masel co workers [8] have disclosed that unsupported Pd and Pd/C catalysts can overcome CO poisoning effect and thereby yield high performances in the DFAFCs. In order to further improve the electrocatalytic performance of the Pd and Pd/ C catalysts, the Pd-based bimetallic catalysts, such as Pd–Ni [9], Pd–Au [10], Pd–Pt [11] and Pd–Ir [12], have been investigated. * Corresponding author. Tel.: +65 68727532; fax: +65 68720785. E-mail addresses: [email protected], [email protected] (Z. Liu). 1388-2481/$ - see front matter Ó 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.elecom.2009.06.023
However, to the best of our knowledge, PdSn bimetallic catalysts for formic acid electrooxidation have not been reported yet. In this paper, PdSn nanoparticles are synthesized by a simple microwave-assisted polyol procedure and deposited on carbon to produce carbon-supported PdSn catalysts, aiming to have a less expensive electrocatalyst in the DFAFC. The physicochemical properties and electrochemical activities of the nanoparticles for formic acid oxidation are investigated. The reasons of oxidation activity enhancement for PdSn catalysts are discussed in detail. 2. Experimental The PdaSnb/carbon black (Cabot Vulcan XC-72, subscript denotes the atomic percentage of the alloying metal) catalysts were prepared by microwave heating of ethylene glycol (EG) solutions of PdCl2 and SnCl22H2O. The Pd content in each sample was 20 wt.%. 1.46, 1.95, 2.93 and 5.85 mL of 0.02 M SnCl22H2O (Aldrich, A.C.S. Reagent) and 5.85 mL of 0.02 M PdCl2 (Aldrich, A.C.S. Reagent) were chosen to yield Pd4Sn1/C, Pd3Sn1/C, Pd2Sn1/C and Pd1Sn1/C, respectively. A typical preparation of Pd2Sn1/C catalyst would consist of the following steps: 5.85 mL of 0.02 M PdCl2 and 2.93 mL of 0.02 M SnCl22H2O was mixed with 30 ml of ethylene glycol (Mallinckrodt, AR). 0.5 mL of 0.8 M NaOH was added dropwise. About 0.05 g of Vulcan XC-72 carbon was added to the mixture and sonicated. The solution was placed in a CEM ‘‘Discover” microwave reactor (CEM Corporation) with the maximum temperature set at 170 °C at atmospheric conditions for 30 s. The resulting suspension was filtered; and the residue was washed with acetone and dried at 100 °C over night in a vacuum oven. For comparison, Pd/C catalyst (20 wt.% Pd loading) was also prepared using the same method.
1668
Z. Liu, X. Zhang / Electrochemistry Communications 11 (2009) 1667–1670
To determine the actual palladium and tin contents in the PdSn alloys, inductively coupled plasma spectroscopy (ICP) was used to measure the unreacted metal ions remaining in the ethylene glycol mixtures. The PdSn alloy nanoparticles of compositions of Pd4Sn1.05, Pd3Sn1.09, Pd2Sn1.03, and Pd1Sn1.06 were obtained from precursors of Pd4Sn1, Pd3Sn1, Pd2Sn1, and Pd1Sn1. The palladium and tin contents in the PdSn alloy nanoparticles were about the same as that in the initial mixtures. The catalysts were examined by TEM on a JEOL JEM 2010. For microscopic examinations the samples were first ultrasonicated in acetone for 1 h and then deposited on 3 mm Cu grids covered with a continuous carbon film. X-ray diffraction (XRD) patterns were recorded by a Bruker GADDS diffractometer with area detector using a CuKa source (k = 1.54056 Å) operating at 40 kV and
Pd/C Pd4Sn1/C Pd3Sn1/C Pd2Sn1/C
Intensity (a. u.)
Pd(220)
Pd(311)
Pd(200)
b Pd(220)
Intensity (a. u.)
C(002)
Pd(111)
a
40 mA. The samples were prepared by depositing carbon-supported nanoparticles on a glass slide. An AUTOLAB potentiostat/galvanostat and a conventional three-electrode test cell were used for electrochemical measurements. The working electrode was a thin layer of Nafion-impregnated catalyst cast on a vitreous carbon disk held in a Teflon cylinder. The catalyst layer was obtained in the following way: (i) a slurry was first prepared by sonicating for 1 h a mixture of 0.5 ml of deionized water, 13 mg of Pd/C or PdSn/C catalyst, and 0.2 ml of Nafion solution (Aldrich: 5 wt.% Nafion); (ii) 4 ll of the slurry was pipetted and spread on the carbon disk; (iii) the electrode was then dried at 90 °C for 1 h and mounted on a stainless steel support. The surface area of the vitreous carbon disk was 0.25 cm2. Pt gauze and an Ag/AgCl electrode were used as the
Pd/C Pd4Sn1/C Pd3Sn1/C Pd2Sn1/C Pd1Sn1/C
Pd1Sn1/C
10
20
30
40
50
60
70
80
90
63
100
65
67
2 Theta (deg.)
69
71
73
2 Theta (deg.)
c
d
e
f
25
35 30
20 Frequency (%)
Frequency (%)
25 15
10
20 15 10
5 5 0
0 1
2
3
4
5
6
7
Particle size (nm)
8
9
10
1
2
3
4
5
6
7
8
9
10
Particle size (nm)
Fig. 1. (a) XRD patterns of PdSn/C catalysts, (b) an expanded view of the (2 2 0) reflections of the fcc phase, (c) TEM image of Pd4Sn1/C catalyst, (d) TEM image of Pd2Sn1/C catalyst, (e) particle size distribution of Pd4Sn1/C catalyst, and (f) particle size distribution of Pd2Sn1/C catalyst.
1669
Z. Liu, X. Zhang / Electrochemistry Communications 11 (2009) 1667–1670
3. Results and discussion The powder XRD patterns for PdSn/C are shown in Fig. 1a alongside the diffraction patterns of a Vulcan carbon-supported Pd catalyst used as comparison. The diffraction peak at 20–25° observed in all the XRD patterns of the carbon-supported catalysts is due to the (0 0 2) reflection of the hexagonal structure of Vulcan XC-72 carbon. The diffraction peaks at about 40, 47, 68 and 82 are due to the Pd (1 1 1), (2 0 0), (2 2 0), and (3 1 1) reflections, respectively, which represents the typical character of a crystalline Pd face-centered cubic (fcc) phase. There are not other distinct reflection peaks in all spectra than those of the four peaks mentioned above, indicating that these electrocatalysts have prevailed Pd (fcc) crystal structure. The (2 2 0) reflections of Pd are used to calculate the average particle size according to the Scherrer equation [13]:
second peak near 0.6 V can be mainly attributes to formic acid oxidation via the CO pathway [14,15]. The peak current density near 0.2 V increased as the Sn content in the anode catalysts increased.
a
Pd/C Pd4Sn1/C
350
Pd3Sn1/C
300
Pd2Sn1/C
250
Pd1Sn1/C Sn/C
200 150 100 50 0 -50 -0.3
0
0.3
0.6
0.9
1.2
E (V vs. Ag/AgCl)
b
70 60
L ¼ 0:9kK a1 =ðB2h cos hB Þ
Pd2Sn1/C
i (A/g Pd)
50
Pd1Sn1/C
Pd3 Sn1 /C
40
Pd4Sn1/C
30 Pd/C
20 10 0
500
1000
1500
2000
2500
Time (s)
c
100 Pd/C Pd4Sn1/C
80
Pd3Sn1/C
i (A/g Pd)
where L is the average particle size, kKa1 is the X-ray wavelength (1.54056 Å for Cu Ka1 radiation), B2h is the peak broadening and hB is the angle corresponding to the peak maximum. An expanded view of the (2 2 0) reflections of the fcc phase is shown in Fig. 1b. The average particle size of Pd (5.2 nm) obtained from XRD patterns was slightly higher than that of Pd4Sn1 (4.8 nm), Pd3Sn1 (4.1 nm), Pd2Sn1 (3.7 nm) and Pd1Sn1 (3.2 nm). The particle size of PdSn alloy nanoparticles obtained by TEM analysis falls in the order Pd (5.5 nm) > Pd4Sn1 (4.8 nm) > Pd3Sn1 (4.1 nm) > Pd2Sn1 (3.7 nm) > Pd1Sn1 (3.2 nm). The values of the average particle size obtained by TEM analysis are almost in good agreement with those calculated from the XRD results, but the former are always higher. The lattice parameter of the (fcc) Pd is estimated to be 3.911 Å. It was found that the addition of Sn to Pd can increase the lattice parameter of Pd (fcc) crystal (in the order of Pd4Sn1 (3.925 Å) < Pd3Sn1 (3.949 Å) < Pd2Sn1 (3.955 Å) < Pd1Sn1 (3.963 Å)) inducing the (2 2 0) reflection peak shift to lower position. The typical TEM images of the Pd4Sn1/C and Pd2Sn1/C catalysts are shown in Fig. 1c and d. It can be seen that a remarkably uniform and high dispersion of metal particles on the carbon surface with an average diameter of 4.8 and 3.7 nm for Pd4Sn1/C and Pd2Sn1/ C, respectively. Evidently, Pd4Sn1 and Pd2Sn1 nanoparticles synthesized by a microwave-assisted polyol process present well-dispersed particles on Vulcan XC-72 and relatively narrow particle size distributions as shown in Fig. 1e and f. Fig. 2a shows the linear sweeping voltammograms of 1 M HCOOH in 0.5 M H2SO4 solution at the different catalysts. There was no significant feature difference between the linear sweeping voltammogramms of room temperature formic acid on carbon supported Pd and carbon supported PdSn catalysts. Obviously, no anodic peak was observed at the Sn/C catalyst, indicating that the Sn/C catalyst has no electrocatalytic activity for the oxidation of formic acid. It can be observed that there are main peaks near 0.2 V and small peaks near 0.6 V. The first anodic peak at near 0.2 V corresponds to formic acid oxidation via the direct pathway, while the
450 400
i (A/g Pd)
counter and reference electrodes, respectively. All potentials in this report are quoted against Ag/AgCl. All electrolyte solutions were deaerated by high-purity argon for 2 h prior to any measurement. For linear sweeping voltammetry and chronoamperometry of formic acid oxidation, the electrolyte solution was 1 M formic acid in 0.5 M H2SO4, which was prepared from high-purity sulfuric acid, high-purity grade formic acid and distilled water. For the electrochemical measurement of the adsorbed CO, CO was bubbled into the solution for 10 min when the electrode potential was fixed at 0 V vs. Ag/AgCl. Then, Ar was bubbled into the solution for 10 min to remove CO in the solution.
Pd2Sn1/C
60
Pd1Sn1/C Sn/C
40
20
0 0
0.2
0.4
0.6
0.8
1
1.2
E (V vs. Ag/AgCl) Fig. 2. (a) Linear sweeping voltammograms of Pd/C and PdSn/C catalysts in 1 M HCOOH, 0.5 M H2SO4 with a scan rate of 50 mV s1 at room temperature. (b) Polarization current vs. time plots for the electrooxidation of formic acid in 1 M HCOOH, 0.5 M H2SO4 electrolyte at 0.3 V (vs. Ag/AgCl) at room temperature. (c) COstripping linear sweeping voltammograms on Pd/C and PdSn/C catalysts in 0.5 M H2SO4 with a scan rate of 50 mV s1 at room temperature.
1670
Z. Liu, X. Zhang / Electrochemistry Communications 11 (2009) 1667–1670
The Pd2Sn1/C catalyst was similar to Pd1Sn1/C catalyst. It is seen from Fig. 2a that the Pd2Sn1/C catalyst exhibits higher current density compared with the Pd1Sn1/C catalyst between 0.15 V and after 0.55 V. This suggests that the presence of tin in the PdSn nanoparticles catalyst markedly enhances the electrocatalytic activity of palladium towards formic acid oxidation. Pd/C and PdSn/C catalysts were biased at 0.3 V vs. Ag/AgCl and the changes in their polarization currents with time were recorded (Fig. 2b). The pattern of current decay was different for each catalyst. For the Pd/C catalyst, the current decayed continuously even after 0.5 h, supposedly because of catalyst poisoning by the chemisorbed carbonaceous species. The Pd2Sn1/C is able to maintain the highest current density for over 0.5 h among all the catalysts, indicating enhanced electrocatalytic activity compared with Pd/C and other PdSn/C catalysts. The comparative tests concluded that Pd2Sn1/C had the best electrocatalytic performance among all carbon supported Pd-based catalysts prepared in this paper. In order to evaluate the poisoning effect of CO, CO-stripping linear sweeping voltammograms of the adsorbed CO at the different catalysts are shown in Fig. 2c. The most notable difference between CO-stripping on PdSn/C catalysts and Pd/C catalyst is the negative shift of the CO oxidation peak in the firmer. At the Pd/C catalyst, a strong anodic peak of adsorbed CO is located at 0.76 V. However, at the PdSn/C catalysts, the anodic peaks of the adsorbed CO are located between 0.67 and 0.72 V, which are more negative than that at the Pd/C catalyst. This is an indication that addition of Sn is helpful to weakening the adsorption strength of CO on Pd through the interaction between Pd and Sn. It was observed that no anodic peak appears at the Sn/C catalyst, indicating that CO cannot be adsorbed on the Sn surface. Addition of Sn exhibits the ability of weakening the adsorptive bond of the reactive intermediates, such as COads and COOHads. The effect of Sn could have prevented the accumulation of poisoning-intermediates and more Pd sites were available for the direct HCOOH decomposition via the direct pathway to CO2. Thus, Sn addition results in a remarkable enhancement of the oxidation activity for formic acid. When the content of Sn in the PdSn/C catalyst is increased, the potential of the oxidation peak of the adsorbed CO is shifted more negatively and the current density of the anodic peak is decreased, suggesting that the increase in the content of Sn in the PdSn/C catalyst changes the adsorption
strength and amount of CO. However, the electrocatalytic activity of the Pd1Sn1/C catalyst would be decreased when the content of Sn in the PdSn/C catalysts is high. This may be due to the lack of electrocatalytic activity on Sn for the oxidation of formic acid. Only when the atomic ratio of Pd and Sn is suitable, such as Pd2Sn1/C catalyst, its electrocatalytic activity can be better than that of the other Pd/Sn mole ratio catalysts. 4. Conclusion Pd and PdSn nanoparticles supported on Vulcan XC-72 carbon have been prepared by a microwave-assisted polyol process. The Pd and PdSn particles are nanosized and have relatively narrow particle size distributions. X-ray diffraction patterns show that the Pd/C and PdSn/C catalysts display the characteristic diffraction peaks of a face-centered cubic Pd structure. It is found that the addition of Sn to Pd can increase the lattice parameter of Pd (fcc) crystal. The results of linear sweeping voltammetry and chronoamperometry show that the Pd2Sn1/C catalyst has better electrocatalytic activity than Pd/C and other Pd/Sn mole ratio catalysts. The Pd2Sn1/C catalyst shows great potential as less expensive electrocatalyst for formic acid electrooxidation in DFAFCs. References [1] C. Rice, S. Ha, R.I. Masel, P. Waszczuk, A. Wieckowski, T. Barnard, J. Power Sources 111 (2002) 83. [2] Y. Zhu, S. Ha, R.I. Masel, J. Power Sources 130 (2004) 8. [3] X. Wang, J.M. Hu, I.M. Hsing, J. Electroanal. Chem. 562 (2004) 73. [4] J. Jiang, A. Kucernak, J. Electroanal. Chem. 520 (2002) 64. [5] S. Park, Y. Xie, M.J. Weaver, Langmuir 18 (2002) 5792. [6] D. Capon, R. Parsons, J. Electroanal. Chem. 65 (1975) 285. [7] M. Arenz, V. Stamenkovic, T.J. Schmidt, K. Wandelt, P.N. Ross, N.M. Markovic, Phys. Chem. Chem. Phys. 5 (2003) 4242. [8] S. Ha, R. Larsen, Y. Zhu, R.I. Masel, Fuel Cells 4 (2004) 337. [9] T. Shobha, C.L. Aravinda, P. Bera, L.G. Devi, S.M. Mayanna, Mater. Chem. Phys. 80 (2003) 656. [10] L.A. Kibler, A.M. El-Aziz, D.M. Kolb, J. Mol. Catal. A: Chem. 199 (2003) 57. [11] R.S. Jayashree, J.S. Spendelow, J. Yeom, C. Rastogi, M.A. Shannon, P.J.A. Kenis, Electrochim. Acta 50 (2005) 4674. [12] X. Wang, Y. Tang, Y. Gao, T. Lu, J. Power Sources 175 (2008) 784. [13] B.E. Warren, X-ray Diffraction, Addison-Wesley, Reading, MA, 1996. [14] W.J. Zhou, J.Y. Lee, Electrochem. Commun. 9 (2007) 1725. [15] Z. Liu, L. Hong, M.P. Tham, T.H. Lim, H. Jiang, J. Power Sources 161 (2006) 831.