Characterization of Carbon-Supported AuPt Nanoparticles for

AumPt100-m nanoparticles of 2−3 nm core sizes with different atomic ... ago for the methanol oxidation reaction (MOR) at the anode.6,7 The catalysts...
1 downloads 0 Views 164KB Size
Download PDF
2892

Langmuir 2006, 22, 2892-2898

Characterization of Carbon-Supported AuPt Nanoparticles for Electrocatalytic Methanol Oxidation Reaction Jin Luo, Peter N. Njoki, Yan Lin, Derrick Mott, Lingyan Wang, and Chuan-Jian Zhong* Department of Chemistry, State UniVersity of New York at Binghamton, Binghamton, New York 13902 ReceiVed NoVember 3, 2005. In Final Form: December 30, 2005 In view of the recent finding that the bimetallic AuPt nanoparticles prepared by molecular-capping-based colloidal synthesis and subsequent assembly on carbon black support and thermal activation treatment exhibit alloy properties, which is in sharp contrast to the bimetallic miscibility gap known for the bulk counterparts in a wide composition range, there is a clear need to assess the electrocatalytic properties of the catalysts prepared with different bimetallic composition and different thermal treatment temperatures. This paper reports recent results of such an investigation of the electrocatalytic methanol oxidation reaction (MOR) activities of the carbon-supported AuPt nanoparticle catalysts with different bimetallic composition and thermal treatment temperatures. AumPt100-m nanoparticles of 2-3 nm core sizes with different atomic compositions ranging from 10% to 90% Au (m ) 10∼90) have been synthesized by controlling the feeding of the metal precursors used in the synthesis. The electrocatalytic MOR activities of the carbon-supported AuPt bimetallic catalysts were characterized in alkaline electrolytes. The catalysts with 65% to 85% Au and treated at 500 °C were found to exhibit maximum electrocatalytic activities in the alkaline electrolytes. The findings, together with a comparison with some well-documented catalysts as well as recent experimental and theoretical modeling results, have revealed important insights into the participation of COad and OHad on Au sites in the catalytic reaction of Pt in the AuPt alloys with ∼75% Au. The insights are useful for understanding the correlation of the bifunctional electrocatalytic activity of the bimetallic nanoparticle catalysts with the bimetallic composition and the thermal treatment temperatures.

Introduction Two of the major problems in the development of practical methanol oxidation fuel cells include poor activity of the anode catalysts and ‘‘methanol crossover” to the cathode electrode1,2 which lead to a loss of about one-third of the available energy at the cathode and another one-third at the anode. For the catalysts, Pt-group metals have been extensively studied, a major problem is the poisoning by CO-like intermediate species.3-5 To address the problem, the PtRu catalysts on carbon support were developed decades ago for the methanol oxidation reaction (MOR) at the anode.6,7 The catalysts exhibit a bifunctional catalytic mechanism in which Pt provides the main site for the dehydrogenation of methanol and Ru provides the site for hydroxide (OH) and for oxidizing CO-like species to CO2. In recent years, the unique properties of gold at nanoscale sizes were found to show unprecedented catalytic activities for CO oxidation8-21 and the * To whom correspondence [email protected].

should

be

addressed.

E-mail:

(1) Ren, X. M.; Zelenay, P.; Thomas, S.; Davey, J.; Gottesfeld, S. J. Power Sources 2000, 86, 111. (2) Chu, D.; Jiang, R. Solid State Ionics 2002, 148, 591. (3) Paulus, U. A.; Endruschat, U.; Feldmeyer, G. J.; Schmidt, T. J.; Bonnemann, H.; Behm, R. J. J. Catal. 2000, 195, 383. (4) Antolini, E. Mater. Chem. Phys. 2003, 78, 563. (5) Lu, G. Q.; Wieckowski, A. Curr. Opin. Colloid Interface Sci. 2000, 5, 95. (6) Reddington, E.; Sapienza, A.; Gurau, B.; Viswanathan, R.; Sarangapani, S.; Smotkin, E. S.; Mallouk, T. E. Science 1998, 280, 1735. (7) Liu, R. X.; Smotkin, E. S. J. Electroanal. Chem. 2002, 535, 49. (8) Haruta, M. Catal. Today 1997, 36, 153. (9) Haruta, M.; Date, M. Appl. Catal., A 2001, 222, 427. (10) Bond, G. C.; Thompson, D. T. Gold Bull. 2000, 33, 41. (11) Corti, C. W.; Holliday, R. J.; Thompson, D. T. Gold Bull. 2002, 35, 111. (12) Gold 2003sNew Industrial Applications for Gold, Proceeding Volume; World Gold Council: Vancouver, 2003. (13) Valden, M.; Lai, X.; Goodman, D. W. Science 1998, 281, 1647. (14) Cameron, D.; Holliday, R.; Thompson, D. J. Power Sources 2003, 118, 298. (15) Rolison, D. R. Science 2003, 299, 1698. (16) Jaramillo, T. F.; Baeck, S. H.; Cuenya, B. R.; McFarland, E. W. J. Am. Chem. Soc. 2003, 125, 7148.

electrocatalytic activity for CO and methanol oxidation.21-27 The exploration of nanoscale gold-based bimetallic materials could potentially provide a synergistic catalytic effect by addressing the problem of poisoning of Pt by CO-like species. For example, gold-platinum (AuPt) nanoparticles could provide synergistic catalytic effects which involve the suppression of adsorbed poisonous species and the change in electronic band structure to modify the strength of the surface adsorption. With such a bimetallic system, one of the motivations is to explore the viability of using Pt as the main dehydrogenation site and Au together with Pt to speed up the removal of the poisonous species. The decrease of activation energy for facilitating oxidative desorption and suppressing the adsorption of CO was previously considered to lead to sufficiently high adsorptivity to support catalytic oxidation in alkaline electrolytes.28-31 A recent study32 showed that catalysts prepared by impregnation from Pt and Au precursors (17) Blizanac, B. B.; Arenz, M.; Ross, P. N.; Markovic, N. M. J. Am. Chem. Soc. 2004, 126, 10130. (18) Kandoi, S.; Gokhale, A. A.; Grabow, L. C.; Dumesic, J. A.; Mavrikakis, M. Catal. Lett. 2004, 93, 93. (19) Avgouropoulos, G.; Ioannides, T.; Papadopoulou, C.; Batista, J.; Hocevar, S.; Matralis, H. K. Catal. Today 2002, 75, 157. (20) Haruta, M. Gold Bull. 2004, 37, 27. (21) Meyer, R.; Lemire, C.; Shaikhutdinov, Sh. K.; Freund, H.-J. Gold Bull. 2004, 37, 72. (22) Zhong, C. J.; Maye, M. M.; Luo, J.; Han, L.; Kariuki, N. N. In Nanoparticles: Building Blocks for Nanotechnology; Rotello, V. M., Ed.; Kluwer Academic Publishers: New York, 2004; Chapter 5, pp 113-144. (23) Zhong, C. J.; Luo, J.; Maye, M. M.; Han, L.; Kariuki, N. N. In Nanotechnology in Catalysis; Zhou, B., Hermans, S., Somorjai, G. A., Eds.; Kluwer Academic/Plenum Publishers: New York, 2004; Vol. 1, Chapter 11, pp 222-248. (24) Zhong, C. J.; Maye, M. M. AdV. Mater. 2001, 13, 1507. (25) Borkowska, Z.; Tymosiak-Zielinska, A.; Nowakowski, R. Electrochim. Acta 2004, 49, 2613. (26) Tang, H.; Chen, J. H.; Wang, M. Y.; Nie, L. H.; Kuang, Y. F.; Yao, S. Z. Appl. Catal., A 2004, 275, 43. (27) Umeda, M.; Ojima, H.; Mohamedi, M.; Uchida, I. J. Power Sources 2004, 136, 10. (28) Nishimura, K.; Kunimatsu, K.; Enyo, M. J. Electroanal. Chem. 1989, 260, 167. (29) Morita, M.; Iwanaga, Y.; Matsuda, Y. Electrichim. Acta 1991, 36, 947.

10.1021/la0529557 CCC: $33.50 © 2006 American Chemical Society Published on Web 02/14/2006

Electrocatalytic AuPt Nanoparticles

were similar to those of monometallic Pt catalysts, suggesting that the presence of Au did not affect the catalytic performance of Pt in any significant way. This was attributed to phase segregation of the two metals due to their miscibility gap. As such, only Pt participates in the adsorption of CO and the catalytic reaction. Recent density function theory (DFT) calculations33 have indicated that the d-band shift for Pt in a Pt-Au alloy on Au(111) differs from that for a Pt-Au alloy on Pt(111). A stronger bonding of CO to the first layer of Pt on Au(111) exists in comparison with the binding of CO on clean Pt. The Au substrate surprisingly increases the Pt overlayer reactivity. The modeling suggests that the change in the CO binding energy is proportional to the shift of the d-band center of the metal overlayer. The adsorption of CO showed an increased binding energy in comparison with that of Pt(111), due to the larger lattice constant of Au, leading to an expansion of Pt. The shift of the d-band depends strongly on Au coverage. The CO desorption temperature from a PtAu alloy was seen to depend linearly on Au coverage.34 The understanding of how such a d-band shift applies for AuPt alloy nanoparticles could have important implications to the design of new or improved bimetallic nanoparticle catalysts. We have recently demonstrated the viability of the electrocatalytic MOR activity of AuPt nanoparticles prepared by molecular-capping-based colloidal synthesis and subsequent assembly on carbon black support and thermal activation treatment.35,36 The thermal activation was demonstrated to be very effective in the removal of the organic shell, the control of the particle size, and the alloying of the metals for the monolayercapped nanoparticles supported on some substrates.37,38 Because of the practical application in catalyst membrane electrode assembly (MEA) in fuel cells, this demonstration of the electrocatalytic properties of the carbon-supported bimetallic catalysts is an advance of our earlier work using catalysts prepared by molecularly mediated assembly of similarly synthesized nanoparticles on flat electrode disks and subsequent electrochemical activation.39-43 While the phase or alloy properties of various metals at bulk states have been well characterized, an important question was whether the bimetallic AuPt nanoparticles are single-phase or phase-segregated materials, which is important for understanding the bimetallic catalytic activities. Most recently, our X-ray diffraction study44 has revealed that the diffraction patterns of the bimetallic AuPt nanoparticles of different compositions prepared by molecular-capping-based colloidal (30) Anderson, A. B.; Grantscharova, E.; Seong, S. J. Electrochem. Soc. 1996, 143, 2075. (31) Burke, L. D.; Collins, J. A.; Horgan, M. A.; Hurley, L. M.; O’Mullane, A. P. Electrochim. Acta 2000, 45, 4127. (32) Mihut, C.; Descorme, C.; Duprez, D.; Amiridis, M. D. J. Catal. 2002, 212, 125. (33) Pedersen, M. Ø.; Helveg, S.; Ruban, A.; Stensgaard, I.; Lægsgaard, E.; Nørskov, J. K.; Besenbacher, F. Surf. Sci. 1999, 426, 395. (34) Sachtler, J. W. A.; Somorjai, G. A. J. Catal. 1983, 81, 77. (35) Luo, J.; Maye, M. M.; Kariuki, N. N.; Wang, L.; Njoki, P.; Lin, Y.; Schadt, M.; Naslund, H. R.; Zhong, C. J. Catal. Today 2005, 99, 291. (36) Luo, J.; Jones, V. W.; Maye, M. M.; Han, L.; Kariuki, N. N.; Zhong, C. J. J. Am. Chem. Soc. 2002, 124, 13988. (37) Luo, J.; Maye, M. M.; Han, L.; Kariuki, N.; Jones, V. W.; Lin, Y.; Engelhard, M. H.; Zhong, C. J. Langmuir 2004, 20, 4254. (38) Luo, J.; Jones, V. W.; Han, L.; Maye, M. M.; Kariuki, N. N.; Zhong, C. J. J. Phys. Chem. B 2004, 108, 9669. (39) Lou, Y.; Maye, M. M.; Han, L.; Luo, J.; Zhong, C. J. Chem. Commun. 2001, 473. (40) Luo, J.; Lou, Y. B.; Maye, M. M.; Zhong, C. J.; Hepel, M. Electrochem. Commun. 2001, 3, 172. (41) Luo, J.; Maye, M. M.; Lou, Y.; Han, L.; Hepel, M.; Zhong, C. J. Catal. Today 2002, 77, 127. (42) Maye, M. M.; Luo, J.; Lin, Y.; Engelhard, M. H.; Hepel, M.; Zhong, C. J. Langmuir 2003, 19, 125. (43) Maye, M. M.; Luo, J.; Han, L.; Kariuki, N. N.; Zhong, C. J. Gold Bull. 2003, 36, 75. (44) Luo, J.; Maye, M. M.; Petkov, V.; Kariuki, N. N.; Wang, L.; Njoki, P.; Mott, D.; Lin, Y.; Zhong, C. J. Chem. Mater. 2005, 17, 3086.

Langmuir, Vol. 22, No. 6, 2006 2893

synthesis and subsequent assembly on carbon black support and thermal activation treatment are characteristic of the fcc-type lattice and show a smooth transition from the pattern of carbonsupported Au to that for carbon-supported Pt with the increase of Pt concentration in the nanoparticles. The lattice parameters were found to scale linearly with the relative Au/Pt content, demonstrating the single-phase character. The alloy properties of the bimetallic AuPt nanoparticles are in sharp contrast to the bimetallic miscibility gap known for the bulk counterparts in a wide composition range (10% to 80% Au).45 In view of this significant finding, there is a clear need to assess the electrocatalytic properties of the catalysts prepared with different bimetallic compositions and different thermal treatment temperatures. We report herein recent results of such an investigation of the electrocatalytic MOR activities of the carbon-supported AuPt nanoparticle catalysts with different bimetallic composition and thermal treatment temperatures, which represents an important advance of our early demonstration of the electrocatalytic MOR activities of carbon-supported AuPt nanoparticle catalysts of limited composition and thermal treatment temperatures.35 Experimental Section Chemicals. Decanethiol (DT, 96%), oleylamine (OAM, 70%), hydrogen tetrachloroaurate (HAuCl4, 99%), tetraoctylammonium bromide (TOABr, 99%), hydrogen hexachloroplatinate (IV) (H2PtCl6‚xH2O, 99.995%), sodium borohydride (NaBH4, 99%), methanol (99.9%), ethanol (99.9%), and Nafion (5 wt %) were purchased from Aldrich and used as received. Other chemicals included hexane (99.9%) and toluene (99.8%) from Fisher. Water was purified with a Millipore Milli-Q water system. Synthesis of Binary Nanoparticles. Gold platinum (AuPt) nanoparticles of 2 nm core size encapsulated with two kinds of organic shells were synthesized by a modified two-phase method.35,44,46 For the preparation of Au65Pt35 nanoparticles, 0.30 g (0.88 mmol) of HAuCl4‚3H2O and 0.48 g (1.17 mmol) of H2PtCl6‚xH2O were dissolved in 20 mL of water, respectively. These Au and Pt precursors were mixed and then followed by phase transfer using 2.0 g (3.66 mmol) of TOABr (tetraoctylammonium bromide) in toluene (80 mL). After separating the water phase, the toluene phase contained AuCl4- and PtCl62-. A volume of 1.0 mL of decanethiol (4.63 mmol) and 1.0 mL of oleylamine (2.13 mmol) were added as capping agents. An amount of 2.00 g (52.34 m mol) of NaBH4 (in 25 mL water) was slowly added as a reducing agent. The reaction was stirred for 4 h, yielding thiol/amine-encapsulated AuPt nanoparticles in the toluene phase. The resulting DT/OAM-encapsulated AuPt nanoparticles in toluene were collected by removing the solvent and cleaned using hexane and ethanol. The particles were then dried and dissolved in hexane. The bimetallic composition of the AuPt nanoparticles was determined by a direct current plasma-atomic emission spectrometer (DCP-AES). Different compositions of alloy nanoparticles were synthesized by controlling the feed ratios of the two metal precursors. Preparation of Binary Nanoparticles on Carbon Support. Carbon black XR-72C obtained from Cabot was used as support materials. The carbon black was first pretreated by suspending it in hexane and sonicating for ∼6 h at room temperature. A controlled amount of AuPt nanoparticles was added into the suspension. The suspension was sonicated for 30 min, followed by stirring overnight. The thus-prepared carbon-supported nanoparticle powders were collected and dried under N2. The loading of AuPt on the carbon support was controlled by monitoring the weight ratio of AuPt nanoparticles versus that of carbon black. The actual loading was determined by thermogravimetric analysis (TGA) and DCP-AES analysis. The carbon-loaded nanoparticles were then treated in a tube furnace under controlled temperature and atmosphere. A typical (45) Ponec, V.; Bond, G. C. Catalysis by Metals and Alloys; Elsevier: Amsterdam, 1995. (46) Hostetler, M. J.; Zhong, C. J.; Yen, B. K. H.; Anderegg, J.; Gross, S. M.; Evans, N. D.; Porter, M. D.; Murray, R. W. J. Am. Chem. Soc. 1998, 120, 9396.

2894 Langmuir, Vol. 22, No. 6, 2006 thermal treatment protocol included shell removal by annealing at 300 °C under 20% O2/N2 for 1 h and treatment at different temperatures under 15% H2/N2 for 2 h. The carbon-loaded AuPt nanoparticles are denoted as AuPt/C. It is important to note that while this treatment condition was found to produce very active catalysts, poor catalytic activities were occasionally observed for some catalysts, which we believe was likely due to burning of the carbon support materials catalyzed by the highly active bimetallic nanoparticle catalysts. The origin of the “burning effect” is not completely clear at this point. Preparation of Catalysts on Electrodes. Glassy carbon (GC) disks (geometric area, 0.07 cm2) were polished with 0.03 µm Al2O3 powders. The geometric area of the substrate electrode (glassy carbon), not the surface area of the catalyst itself, provides a measure of the loading of catalyst on the electrode surface. A typical suspension of the catalysts was prepared by suspending 1 mg of catalysts (AuPt/ C) in 1 mL of 0.25% Nafion solution and sonicating for 15 min. The suspension was stable for days. The suspension was then quantitatively transferred to the surface of the polished GC disk. The electrodes were dried overnight at room temperature. Instrumentation and Measurements. Direct Current PlasmaAtomic Emission Spectroscopy (DCP-AES). DCP-AES was used to analyze the composition of the bimetallic nanoparticles using an ARL Fisons SS-7 DCP-AES instrument; the measurements were made on emission peaks at 267.59 and 265.95 nm, for Au and Pt, respectively. The nanoparticle samples were dissolved in concentrated aqua regia and then diluted to concentrations in the range of 1-50 ppm for analysis. Calibration curves were made from dissolved standards with concentrations from 0 to 50 ppm in the same acid matrix as the unknowns. Standards and unknowns were analyzed 10 times each for 3 s counts. Instrument reproducibility, for concentrations greater than 100 times the detection limit, results in < (2% error. For example, the DCP analysis of the bimetallic nanoparticles synthesized under the above condition yielded an atomic composition of 65% Au and 35% Pt. The nanoparticles of this composition are denoted as Au65Pt35. Transmission Electron Microscopy (TEM). TEM was performed on a Hitachi H-7000 electron microscope (100 kV). For TEM measurements, AuPt or AuPt/C samples were suspended in hexane solution and were drop cast onto a carbon-coated copper grid followed by solvent evaporation in air at room temperature. ThermograVimetric Analysis (TGA). TGA was performed on a Perkin-Elmer Pyris 1-TGA for determining the metal loading on carbon black. Typical samples weighed ∼4 mg and were heated in a platinum pan. Samples were heated in 20% O2 at a rate of 10 °C/min. X-ray Powder Diffraction (XRD). XRD data were collected on a Philips X’Pert diffractometer using Cu KR radiation (λ ) 1.5418 Å). The measurements were done in reflection geometry, and the diffraction (Bragg) angles 2θ were scanned at a step of 0.025°. Each data point was measured for at least 20 s, and several scans were taken on each sample. Measurement of Electrocatalytic ActiVity. The cyclic voltammetry (CV) measurements were performed using a microcomputercontrolled electrochemical analyzer (CHI600a, CH Instruments). The experiments were performed in three-electrode electrochemical cells at room temperature. All electrolytic solutions were deaerated with high-purity argon or nitrogen before the measurement. The potentials are given with respect to the reference electrode of Ag/ AgCl saturated KCl.

Results and Discussion The discussion of the experimental results is divided into two sections. In the first section, the morphological, compositional, and structural properties of the bimetallic alloy nanoparticle catalysts are described. The results establish the correlation between the bimetallic composition in the nanoparticles and the feeding of the metal precursors in the synthesis. In the next section, the results from the characterization of the electrocatalytic

Luo et al.

Figure 1. TEM micrograph of Au82Pt18/C after treatment at 500 °C. Bottom inset: high-resolution TEM micrograph for the indicated nanoparticle. Top inset: TEM micrograph for the Au82Pt18 nanoparticles before assembly.

Figure 2. Composition data for AuPt nanoparticles derived from DCP-AES analysis. The plot shows the relationship between the Au % from the synthetic feeding and Au % in the nanoparticles.

MOR activity are described. Insights gained from the findings into the mechanistic aspects are also discussed. 1. Morphological, Compositional, and Structural Properties. Figure 1 shows a typical set of TEM micrographs for the as-synthesized bimetallic AuPt nanoparticles (upper-left inset) and the carbon-supported AuPt nanoparticles after thermal treatment as well as the HRTEM in the lower-right inset (example: Au82Pt18). Similar to DT-capped AuPt nanoparticles,35 the as-synthesized DT/OAM-encapsulated AuPt nanoparticles (Figure 1, upper-left inset) were found to display an average size of 1.7-2.4 nm with a relatively high monodispersity (e(0.6 nm) based on a count of ∼2000 particles. As evidenced by the uniform interparticle spacing, the particles seem to be well separated from each other due to the presence of the capping organic monolayer. The presence of such a capping layer is supported by FTIR characterization. AumPt100-m nanoparticles with different atomic compositions ranging from 10% to 90% Au (m ) 10∼90) have been synthesized by the method described in the Experimental Section. The bimetallic composition in the nanoparticles was controlled by varying the feeding of the metal precursors used in the synthesis. Figure 2 shows a representative set of results comparing the bimetallic composition in the nanoparticles and the feeding of the metal precursors in the synthesis. The relationship between the molar feeding of the two metal precursors in the synthetic solution and the resulting composition (Au %) in the nanoparticles exhibits a clear proportional trend, though not in a well-defined linearity as observed in our previous work in synthesizing watersoluble AuPt nanoparticles.47 The significant deviation appears around ∼50% Au. The observation that the percent of Au in the nanoparticles is larger than the percent of Au in the synthetic

Electrocatalytic AuPt Nanoparticles

Figure 3. XRD spectra of Au82Pt18/C catalysts treated at 400 °C (solid line) and 500 °C (dashed line).

feed solution by a factor of 1.1 to 1.5 suggests that there is some loss of Pt from the nanoparticles. On the basis of an analysis of the solution left after collecting the nanoparticles, Pt (likely in the form of PtCl62-) was detected. The loss of Pt reflects the instability due to relatively poor encapsulation by thiols or amines on Pt in comparison with encapsulation of Au by thiols. As stated in the Experimental Section, the thermal treatment involved heating the catalyst at 300 °C under 20% O2 followed by treatment at 400-600 °C under 15% H2. It has been found that the temperature range of 400-500 °C is the most appropriate thermal treatment condition for both the electrocatalytic (shown in next section) and the alloying properties.44 After the thermal treatment of AuPt/C, the particle dispersion remains relatively high, but the average size is found to increase by a certain degree that is dependent on the treatment temperature (see the example in Figure 1). The high-resolution TEM image (Figure 1, bottom inset) indicates that the thermally treated nanoparticles exhibit highly crystalline morphology. As demonstrated in our earlier report,37 there are two important pieces of evidence supporting the effective removal of the capping monolayers from the nanoparticles by the thermal treatment condition. First, the vibrational bands characteristic of the capping molecules in the C-H stretching region (νa(CH3), 2955 cm-1; νs(CH3), 2872 cm-1; νa(CH2), 2917 cm-1; νs(CH2), 2848 cm-1) were not detected by FTIR. Second, XPS analysis showed that the bands associated with sulfur species (S(2p1/2), 163.8 eV; S(2p3/2), 162.5 eV) were absent after the thermal treatment. The thermal treatment of the catalysts is also believed to influence the core and the surface crystallinity properties of the nanocrystals. As stated in the Introduction on our recent XRD study,44 the carbon-supported AuPt nanoparticles treated at 500 °C exhibit single-phase character as demonstrated by the linear relationship between the lattice parameters and the composition. For the change of the surface properties, we are currently performing an in-depth FTIR probing of CO adsorption on the catalysts. While a detailed account of the results will be described in another report, the results also showed that the surface bimetallic properties are consistent with the core bimetallic properties. To assess the temperature effect on thermally treated bimetallic nanoparticles, Figure 3 shows a set of XRD data for the Au82Pt18/C catalyst treated at 400 and 500 °C. While the diffraction patterns are characteristic of the fcc-type lattice,44 there are subtle differences in peak shape, width, and position. The lattice parameters were determined by carefully determining the positions of the Bragg peaks in the diffraction patterns. The Bragg peaks (47) Njoki, P. N.; Luo, J.; Wang, L.; Maye, M. M.; Quaizar, H.; Zhong, C. J. Langmuir 2005, 21, 1623.

Langmuir, Vol. 22, No. 6, 2006 2895

Figure 4. Cyclic voltammetric curves for Au82Pt18/C (solid line) and Au/C (dashed line) catalysts (treated at 400 °C) on a GC electrode (0.07 cm2) in 0.5 M H2SO4 electrolyte saturated with Ar. Scan rate: 50 mV/s.

(2θ) are found at 38.6° or 38.3° for 400 or 500 °C, respectively. It seems that the changes in peak width and symmetry are more significant than the peak position, which likely suggests better crystallinity for the catalysts treated at 500 °C in comparison with that at 400 °C. There was also a slight increase in particle size, and the estimate from the XRD peak widths yielded 4.2 (( 0.5) nm and 5.0 (( 0.5) nm after the treatment at 400 and 500 °C, respectively, which are close to those determined from TEM data.44 Upon further increasing the treatment temperature to higher than ∼650 °C, experiments revealed indications of phase segregation and larger particle sizes. 2. Electrocatalytic MOR Activity. As an initial step to determine the electrocatalytic activity, the presence of Pt on the surface of AuPt nanoparticle catalysts was first assessed by CV measurements of the catalysts loaded on a glassy carbon electrode in 0.5 M H2SO4 solution. The CV data for Au/C and AuPt/C catalysts were compared, as shown by a representative set of CV curves in Figure 4. In contrast to the largely silent feature of the voltammetric current at E < 0.0 V for the Au/C catalyst, the CV curve for the AuPt catalyst reveals hydrogen evolution/adsorption-desorption waves indicative of the presence of a Pt component on the surface of the bimetallic catalyst. The fact that the general feature for the hydrogen evolution/adsorptiondesorption waves does not completely resemble those fine features observed for a pure Pt catalyst is believed to reflect the Pt-Au alloy character on the Au82Pt18/C catalyst surface. Depending on the relative Pt concentration in the bimetallic nanoparticles, features characteristic of hydrogen adsorption waves and hydrogen evolution current characteristic of Pt in the -0.2 V to +0.1 V potential range were found to be modified by the presence of Au atoms on the surface. It is thus interesting to examine how such a modification affects the operation of the bimetallic composition of the AuPt nanoparticles in the electrocatalytic reaction. The electrocatalytic MOR activity of the AuPt/C catalysts loaded on the glassy carbon electrode was examined in both acidic and alkaline electrolytes. While the results from the acidic electrolyte showed a low activity, the electrocatalytic activity was found to be high in the alkaline electrolyte, which is the focus of the present paper. The methanol oxidation activity was found to be dependent on both composition and treatment temperature. Figure 5 shows a typical set of CV curves obtained for MOR in alkaline electrolyte on Au82Pt18/C catalysts (20% metal loading) treated at two different treatment temperatures, 400 and 500 °C. In the absence of methanol, both catalysts exhibit redox waves of Au/Au-oxide on the surface. The gold oxidization wave was found at 0.3 V, whereas the reduction wave was around 0.06 to 0.07 V for both

2896 Langmuir, Vol. 22, No. 6, 2006

Luo et al.

Figure 5. Cyclic voltammetric curves for Au82Pt18/C (metal loading, 20%) catalysts treated at different treatment temperatures (A, 400 °C; B, 500 °C, on a GC electrode, 0.07 cm2) in 0.5 M KOH electrolyte with (solid curves) and without (dashed curves) 0.5 M methanol. Scan rate: 50 mV/s.

catalysts. There seems to be a subtle difference in the redox current, which is perhaps suggestive of the surface composition difference. The Au82Pt18/C catalysts treated at 500 °C showed larger redox current for gold than that treated at 400 °C. In comparison with the data from the above control experiment (dashed lines), there is a large anodic wave at -0.16 V or at ∼ -0.17 V for catalysts treated at 400 and 500 °C, respectively, in the presence of methanol. This anodic wave corresponds to the electrocatalytic oxidation of methanol. The peak potentials are much more negative than that for a monometallic Au catalyst (Au/C).35 The magnitude of the anodic current for the AuPt/C treated at 500 °C is greater than that treated at 400 °C. This enhanced electrocatalytic activity is likely associated with the better alloy properties of the catalyst treated at 500 °C than that treated at 400 °C, as evidenced by the XRD data (Figure 3), leading thus to more continuous Au atoms on the surface of the Au82Pt18 catalyst. Furthermore, a smaller anodic wave is observed at ∼ -20 mV on the reverse sweep for these AuPt/C catalysts, which is attributed to oxidation of methanol on reactivated catalyst surface.48 The electrochemical data are further compared with those from well-documented Pt and PtRu catalysts (e.g., E-tek’s Pt/C and PtRu/C). For the purpose of comparison, we used the catalysts with same catalyst loading, i.e., 20% Au82Pt18/C, 20% Au72Pt28/C, 20% Pt/C, and 20% PtRu/C (wt % of metals in the catalyst). The catalyst thin films on the GC electrode were prepared under the same condition. Figure 6 shows a representative set of cyclic voltammetric curves comparing the electrocatalytic MOR characteristics in alkaline solution between different catalysts, (48) Page, T.; Johnson, R.; Hormes, J.; Noding, S.; Rambabu, B. J. Electroanal. Chem. 2000, 485, 34.

Figure 6. Cyclic voltammetric curves for several catalysts (on a GC electrode, 0.07 cm2) in 0.5 M KOH with (solid curves) and without (dashed curves) 0.5 M methanol: Pt/C (A), PtRu/C (B), Au82Pt18/C (C), Au72Pt28/C (D), and Au/C (E). All data were obtained with 20% metal loading except Au/C with 17% metal loading. Scan rate: 50 mV/s.

i.e., Pt/C, PtRu/C, Au82Pt18/C, and Au72Pt28/C catalysts. The data for Au/C catalysts35 are also included in Figure 6 for comparison. To ensure the quantitative comparison of the data under a relatively comparable loading condition, we also measured the peak current versus the amount of catalysts loaded onto the electrode surface. The observed linear relationship for our catalyst loading range35 indicates comparability of data because the catalysts loading on the electrode surface falls in a similar range. It is evident that the general electrocatalytic characteristic for AuPt/C catalysts is quite similar to those observed for the Pt/C and PtRu/C catalysts. A close examination of the peak potentials and currents reveals subtle differences. The peak potential in the forward scan for Au82Pt18/C catalysts (∼ -170 mV) is more positive by +10 mV than that for the Pt/C (∼ -180 mV) and by +80 mV than that for PtRu/C (∼ -250 mV) catalysts. The peak current density for the Au82Pt18/C catalyst (∼8500 mA‚cm-2‚(mg AuPt)-1), after being normalized to the total metal loading, is larger than that for the PtRu/C catalyst (∼7600

Electrocatalytic AuPt Nanoparticles

Langmuir, Vol. 22, No. 6, 2006 2897

Table 1. Electrocatalytic MOR Activities for AuPt/C Catalysts in 0.5 M KOH + 0.5 M Methanola treated at 400 °C catalyst on carbon

Tafel slope (mV/dec)

mass activity (mA/cm2/mg Mt)

treated at 500 °C Tafel slope (mV/dec)

mass activity (mA/cm2/mg Mt)

Au Pt Au97Pt3 Au82Pt18 Au72Pt28 Au65Pt35 Au60Pt40 Au56Pt44 Au35Pt65 a

treated at 600 °C Tafel slope (mV/dec)

mass activity (mA/cm2/mg Mt)

120

4078

349 8092 127 155

6518 2821

154

855

106 149 148 106 165 101

1869 8536 5291 7875 4482 4432 4984

Note: there is (5-6% error bar for the mass activity (Mt is the total amount of metals).

Figure 7. Tafel plots for Au82Pt18/C (a) (treated at 500 °C) and PtRu/C (b) catalysts (on GC electrode, 0.07 cm2) in 0.5 M KOH with 0.5 M methanol.

mA‚cm-2‚(mg PtRu)-1) and slightly smaller than the Pt/C catalyst (∼8900 mA‚cm-2‚(mg Pt)-1). This observation indicates that there is a major improvement in comparison with that of the Au/C catalysts in terms of the peak potential (by ∼ -600 mV) and the peak current (by ∼25×). The presence of a small fraction of Pt in the Au-based bimetallic nanoparticles significantly modified the catalytic properties. The catalytic modification of the bimetallic composition is in fact further reflected by the remarkable difference of the voltammetric characteristic observed in the reverse scan, especially in the alkaline electrolyte. For Pt/C and PtRu/C, the reverse wave for alkaline electrolyte occurs at a potential less positive than the forward wave by ∼200 mV. In contrast, the reverse wave for Au82Pt18/C occurs at a potential which differs from the potential for the wave in the forward sweep by only ∼20 mV. The relative peak current of the reverse/forward wave is also found to be dependent on Au % in the bimetallic nanoparticle. The oxides formed on the catalyst surface at the potential beyond the anodic peak potential in the positive sweep are reduced in the reverse sweep.48 Poisonous CO species formed on the Pt surface can also be removed in the reversed sweep. The observation of the more positive potential for the reverse wave likely reflects the bimetallic effect on the reactivation of the catalyst surface after the anodic sweep, a scenario that is under further investigation using FTIR spectroscopic techniques. The reactivation of the surface catalytic sites after the anodic sweep is likely modified by the presence of Au in the catalyst, which leads to the shift of the peak potential of the reverse wave to a more positive potential (by ∼200 mV) for Au82Pt18/C than for Pt/C. We further compare in Figure 7 the Tafel plots for our Au82Pt18/C catalyst (treated at 500 °C) and the PtRu/C catalyst in 0.5

M KOH with 0.5 M methanol. The Tafel slope for our Au82Pt18/C catalyst is 148 mV/dec, which is slightly larger than that for the PtRu/C catalyst. It is important to note that the Tafel slope found for our AuPt catalysts are somewhat larger than those reported for many known catalysts such as PtRu/C and Pt/C (90 to 140 mV/dec).49-51 The transfer coefficient (Rn) calculated from the slopes yielded a value of 0.4. On the basis of the experimental data and the relevant studies,49,52 the rate-determining step on the Au82Pt18/C catalysts could be the first electrontransfer step (first C-H bond breakage of methanol or the activation of water). The correlation between the composition and the electrocatalytic activity for the above catalysts studied is assessed by analyzing the mass activity (i.e., the peak current density per unit total mass of metals) and the kinetic parameters (Tafel slope). Table 1 summarizes a set of the results for the electrocatalytic MOR in 0.5 M KOH. In general, the results from electrocatalytic MOR activities demonstrate that the bimetallic composition can significantly modify the electrocatalytic properties of both Au and Pt, and the electrocatalytic activity of the carbon-supported AuPt nanoparticle catalysts depends on the composition, the temperature used for the thermal treatment, and the nature of the electrolytes. The mass activity for the AuPt/C catalyst treated at 600 °C or higher temperatures did not show further improvement of the mass activity. One of the most significant findings is that the mass activity appears to exhibit a maximum around the composition of 65% to 85% Au, whereas the variation of the Tafel slope appears to be slightly higher (∼160 mV/dec) in this range than those outside of this range (∼100 mV/dec). This maximum mass activity for the AuPt/C catalysts (5500 to 8500 mA/cm2/mg Mt), which contain only 15% to 35% Pt, appears to be quite comparable to that corresponding to the Pt/C catalyst (∼8000 mA/cm2/mg Mt) and much higher than that corresponding to the Au/C catalyst (∼350 mA/cm2/mg Mt). This finding is intriguing and raises the question about the origin of the high catalytic activity for the AuPt catalysts with 65% to 85% Au. While we do not have a definite explanation for the origin at this time, the most recent modeling results based on the DFT calculation for CO adsorption on small clusters of AuPt33,53 seem to shine some interesting insights into this finding. The modeling results showed that the CO adsorption energy increases with Pt % which maximizes at ∼30% Pt for CO adsorption on Au atoms and increases with Pt % for CO adsorption on Pt atoms for all compositions. This finding is in fact qualitatively consistent with (49) Biswas, P. C.; Nodasaka, Y.; Enyo, M. J. Appl. Electrochem. 1996, 26, 30. (50) Kucernak, A.; Jiang, J. H. Chem. Eng. J. 2003, 93, 81. (51) Gojkovic, S. L. J. Electroanal. Chem. 2004, 573, 271. (52) Tapan, N. A.; Mustain, W. E.; Gurau, B.; Sandi, G.; Prakash, J. J. New Mater. Electrochem. Syst. 2004, 7, 281. (53) Song, C.; Ge, Q.; Wang, L. J. Phys. Chem. B 2005, 109, 22341.

2898 Langmuir, Vol. 22, No. 6, 2006

Luo et al.

our recent FTIR study of CO adsorption of AuPt nanoparticles54 which reveals that the Au-atop CO band decreases but remains significant, whereas the Au-bridge CO band shifts toward higher frequency but remains mostly Au-like features for < ∼50% Pt. This is important because it implies that gold atoms surrounding the Pt atom in the AuPt alloy with the relative high percentage of Au may have played an important role in either removing the intermediate CO-like species or providing oxygenated species in the methanol oxidation process. This assessment is consistent with the known facts that the nanoscale Au is catalytically highly active for CO oxidation,8,9,55 and Au is known to be capable of producing surface-oxygenated species in basic electrolytes.56,57 On the basis of the above results and discussion, we believe that bifunctional electrocatalytic properties may be operative for MOR on the AuPt nanoparticle catalysts in the alkaline electrolyte. The similarity of electrocatalytic current density for the bimetallic catalyst with 65% to 85% Au to those for the pure Pt catalyst is suggestive of the participation of Au in the catalytic reaction of Pt. In view of the alloy character for the AuPt/C catalysts as revealed in our recent work,44 the Pt atom for a AuPt alloy with ∼75% Au would be practically surrounded by Au atoms. Mechanistically, the bifunctional properties could then involve a possible combination of the following reactions:

MeOH + Pt w Pt-COad + CO3-

(1)

Pt-COad + Au w Au-COad + Pt

(2)

Au + OH- w Au-OHad + e-

(3)

Pt-COad + Au-OHad w CO3- + Pt + Au + e-

(4)

Au-COad + Au-OHad w CO3- + Au + e-

(5)

While a detailed delineation of the proposed electrocatalytic mechanism is part of our ongoing work, several important insights are supportive of the mechanistic view. The formation of intermediate COad species on Pt (eq 1) is a well-known fact. The transfer of the intermediate COad species from Pt-atop site to a neighboring Au-atop site (eq 2) is possible in view of the favorable (54) Mott, D.; Smith, A.; Luo, J.; Zhong, C. J. Presented at the Materials Research Society 2005 Fall Meeting, Boston, MA, Nov 28-Dec 2, 2005; O1.6. (55) Haruta, M. Nature 2005, 437, 1098. (56) Burke, L. D. Gold Bull. 2004, 37, 125. (57) Luo, J.; Maye, M. M.; Han, L.; Zhong, C. J.; Hepel, M. J. New Mater. Electrochem. Syst. 2002, 5, 237.

adsorption of CO on Au nanoparticles known from both experimental measurements55 and theoretical calculations.33,53 The formation of Au-OHad or surface oxides on gold in alkaline electrolyte (eq 3) was in fact proposed to explain some of the electrocatalytic properties observed for the gold electrode (e.g., the incipient hydrous oxide/adatom mediator model56). Our previous in situ measurement of the interfacial mass change accompanying the electrocatalytic oxidation of methanol also indicated the formation of Au oxides (Au2O3, AuOH, or Au(OH)3) on gold nanoparticle surfaces.57 The synergistic cooperation of these reactions could have played important roles in the electrocatalytic reactions (eqs 4 and 5) toward the final product. We also note that the enhanced electrocatalytic activity associated with the better alloy properties of the catalyst treated at 500 °C than that treated at 400 °C as revealed in Figure 5 is in fact supportive of the above mechanistic view.

Conclusion In conclusion, our studies of the electrocatalytic properties of the catalysts prepared with different bimetallic compositions and different thermal treatment temperatures have shined important insights into the mechanistic correlations between the catalytic activity and the composition. The relationship between the feeding of the Au and Pt metal precursors in the synthetic solution and the resulting composition in the nanoparticles is established for the modified two-phase synthesis protocol. The characterization results showed important details for the dependence of the morphology and the surface binding properties of the bimetallic nanoparticles on the treatment temperature. The carbon-supported AuPt nanoparticle catalysts with 65% to 85% Au and treated at 500 °C are shown to exhibit maximum electrocatalytic activities in the alkaline electrolytes. The findings, together with a comparison of the electrocatalytic activities with some welldocumented catalysts as well as recent experimental and theoretical modeling results, have revealed important insights into the participation of COad and OHad on Au sites in the catalytic reaction of Pt in the AuPt alloys with ∼75% Au. A detailed delineation of the proposed electrocatalytic mechanism is part of our ongoing work. Acknowledgment. This work was supported in part by the National Science Foundation (CHE 0316322) and the Petroleum Research Fund administered by the American Chemical Society. We thank Dr. V. Petkov of Central Michigan University for XRD measurements and Dr. H. R. Naslund of State University of New York at Binghamton for DCP-AES analysis. LA0529557