Probing the Formation Mechanism and Chemical ... - ACS Publications

Mar 16, 2006 - Fadlilatul Taufany , Chun-Jern Pan , John Rick , Hung-Lung Chou , Mon-Che Tsai , Bing-Joe Hwang , Din-Goa Liu , Jyh-Fu Lee , Mau-Tsu Ta...
0 downloads 0 Views 444KB Size
Download PDF
J. Phys. Chem. B 2006, 110, 6475-6482

6475

Probing the Formation Mechanism and Chemical States of Carbon-Supported Pt-Ru Nanoparticles by in Situ X-ray Absorption Spectroscopy Bing Joe Hwang,*,†,‡ Ching-Hsiang Chen,† Loka Subramanyam Sarma,† Jiun-Ming Chen,† Guo-Rung Wang,† Mau-Tsu Tang,‡ Din-Goa Liu,‡ and Jyh-Fu Lee‡ Nanoelectrochemistry Laboratory, Department of Chemical Engineering, National Taiwan UniVersity of Science and Technology, Taipei 106, Taiwan, Republic of China ReceiVed: NoVember 4, 2005; In Final Form: February 5, 2006

The understanding of the formation mechanism of nanoparticles is essential for the successful particle design and scaling-up process. This paper reports findings of an X-ray absorption spectroscopy (XAS) investigation, comprised of X-ray absorption near-edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) regions, to understand the mechanism of the carbon-supported Pt-Ru nanoparticles (NPs) formation process. We have utilized Watanabe’s colloidal reduction method to synthesize Pt-Ru/C NPs. We slightly modified the Watanabe method by introducing a mixing and heat treatment step of Pt and Ru oxidic species at 100 °C for 8 h with a view to enhance the mixing efficiency of the precursor species, thereby one can achieve improved homogeneity and atomic distribution in the resultant Pt-Ru/C NPs. During the reduction process, in situ XAS measurements allowed us to follow the evolution of Pt and Ru environments and their chemical states. The Pt LIII-edge XAS indicates that when H2PtCl6 is treated with NaHSO3, the platinum compound is found to be reduced to a Pt(II) form corresponding to the anionic complex [Pt(SO3)4]6-. Further oxidation of this anionic complex with hydrogen peroxide forms dispersed [Pt(OH)6]2- species. Analysis of Ru K-edge XAS results confirms the reduction of RuIIICl3 to [RuII(OH)4]2- species upon addition of NaHSO3. Addition of hydrogen peroxide to [RuII(OH)4]2- causes dehydrogenation and forms RuOx species. Mixing of [Pt(OH)6]2- and RuOx species and heat treatment at 100 °C for 8 h produced a colloidal sol containing both Pt and Ru metallic as well as ionic contributions. The reduction of this colloidal mixture at 300 °C in hydrogen atmosphere for 2 h forms Pt-Ru nanoparticles as indicated by the presence of Pt and Ru atoms in the first coordination shell. Determination of the alloying extent or atomic distribution of Pt and Ru atoms in the resulting Pt-Ru/C NPs reveals that the alloying extent of Ru (JRu) is greater than that of the alloying extent of Pt (JPt). The XAS results support the Pt-rich core and Ru-rich shell structure with a considerable amount of segregation in the Pt region and with less segregation in the Ru region for the obtained Pt-Ru/C NPs.

1. Introduction Nanometer-sized carbon-supported and unsupported Pt-Ru nanoparticles have received great scientific interest in academic and industrial research due to their superior activity as anode catalysts for methanol electrooxidation and CO tolerance oxidation related to fuel cell applications.1 The current interest in the use of Pt-Ru nanoparticles is largely due to the enhanced activity and selectivity that may be achieved by two metals working synergistically as explained by the so-called bifunctional mechanism. According to this bifunctional mechanism adsorbed CO species are oxidized by OH species generated on Ru surface atoms2 or by electronic effects3 where the presence of ruthenium involves a change in the electronic density of state of platinum leading to the weakening of the CO-Pt bond, or a mixing of both effects as was recently shown by Waszczuk et al.4 A true synergetic effect between the two metals is shown only if they are in strong interaction, e.g., in contact with each other, allowing higher CO mobility from platinum toward ruthenium. It is clear that the catalytic performance is strongly * Address correspondence to this author. Fax: +886-2-27376644. E-mail: [email protected]. † National Taiwan University of Science and Technology. ‡ Current address: National Synchrotron Radiation Research Center, Hsinchu 300, Taiwan.

dependent on the distribution of Pt and Ru sites at the atomic level and believed to be sensitive to the particle’s shape and size and to the nature of the support. Various methods are used for the preparation of Pt-Ru bimetallic NPs, such as (1) the colloidal chemistry methods,5 (2) the reverse micelles method,6 (3) the microwave irradiation method,7 (4) the alcohol-reduction method,8 and (5) the deposition of an organometallic PtRu5C(CO)16 cluster onto a carbon black followed by the decomposition at a high-temperature treatment.9 Of the several methods considered, the preparation of the catalyst via the formation of a sulfite complex, followed by a thermal treatment at 280 °C in a reducing atmosphere gave better results,10 comparable to those obtained with the state-ofthe-art Pt-Ru/C catalysts. However, from the experience of formation mechanism studies on mono- and bimetallic nanoparticles by XAS11 we believe that the change in the ligand environment around metal ions and mixing of metal ion precursors will affect the reducibility of metal ions. Here we attempted to enhance the homogeneity and atomic distribution of metals in the nanoparticle by improving the mixing efficiency of metal ion precursors prior to the reduction step and modified the Watanabe colloidal reduction method accordingly. Even though a wealth of information is available on the synthesis of Pt-Ru bimetallic NPs studies focusing on understanding the

10.1021/jp0563686 CCC: $33.50 © 2006 American Chemical Society Published on Web 03/16/2006

6476 J. Phys. Chem. B, Vol. 110, No. 13, 2006 relationships that exist between composition, size and ultimately the structure adopted are limited. Nuzzo et al. give detailed information about the structural aspects of the Pt-Ru bimetallic NPs and pointed-out that more in-depth studies on the above features are beneficial to gain new insights that could lead to the useful catalytic activities of metallic NPs.12 Recently on the basis of XAS methodology, we have shown that the alloying extent or atomic distribution in nanoparticles is another important factor of concern that has a strong influence on their catalytic activities.13 Thus studies focusing on controlling the size, morphology, and homogeneity of the bimetallic NPs14 are of great importance in current nanoscience research and to achieve this thorough understanding of the metal nanoparticle formation mechanism is necessary.11 Also success in either particle design or scaling-up requires a detailed knowledge of the particle formation mechanism. A full understanding of the particle formation mechanism during the preparation process, especially in terms of the chemical environment around the metal ions to be reduced, and nature of the ligands formed during the course of nanoparticle formation will greatly benefit the development of needed structure-controllable synthetic pathways for metal nanoparticles. For metal nanoparticles X-ray absorption spectroscopy (XAS) has proved to be one of the most suitable methods for investigating structural evolution and in many cases structural properties of metal particles can be probed in situ during the different steps of preparation. Other techniques such as X-ray diffraction (XRD) and transmission electron microscopy (TEM)15 are difficult to employ during in situ conditions because the nanoparticle structure would change during the preparation or transportation of the sample or by the lack of long-range ordering.16 It has been shown in the literature that study at the X-ray absorption near-edge spectroscopy (XANES) region (conventionally from below the edge up to ∼30-50 eV) provides information about the oxidation state, fractional d-electron density, and electronic environment of the absorbing atom. Spectra obtained from the region extending from the XANES region to as high as 2 keV above the edge are known as the extended X-ray absorption fine structure (EXAFS), and are primarily due to the scattering of the photoelectron off nearneighbor atoms. The amplitude of the EXAFS function χ(k), where k is the wave vector, is proportional to the number of nearest neighbors, and the change of phase with the wavelength of the photoelectron depends on the distance between the emitter and the backscattering atom.17 The backscattering strength also depends on the type or atomic number of atoms involved in the backscattering process. Thus, an analysis of EXAFS data yields structural details about the absorbing atom and its local environment. In recent years XAS studies have been well explored on bimetallic nanoparticles.18 The electronic and structural properties of nanometer-scale metallic clusters have been widely studied by utilizing the L-edge XANES.19 Recently Bazin et al. have used a combined full multiple scattering (MS) and principal component analysis (PCA) approach for the K-edge XANES analysis of nanometer-scale transition metals. The combined MS and PCA approach offers a better understanding of the physicochemical processes specific to nanometer scale entities.20 In situ XAS allows investigating the properties of Pt-Ru catalyst particles under working conditions similar to that of the actual fuel cell. Even though XAS has been applied to the study of fuel cell electrocatalysts its application during the formation of Pt-Ru nanoparticles has remained relatively less explored. Note that interesting results can be obtained

Hwang et al. through wide-angle X-ray scattering (WAXS) on Pt-Ru nanoparticles.21 In this work, we demonstrate a simple XAS methodology for following the formation of Pt-Ru nanoparticles by analyzing the XAS spectra recorded at the Pt LIII-edge (11564 eV) and Ru K-edge (22117 eV). In addition, we discuss the atomic distribution or alloying extent of Pt and Ru atoms inside the Pt-Ru nanoparticles based on the XAS structural parameters. The methodology proposed in this contribution to study the formation mechanism would be beneficial to a rational design and size and shape control of metal nanoparticle fabrication which is highly required for their superior catalytic activities. 2. Experimental Section Synthesis of Pt-Ru Bimetallic NPs. Carbon-supported PtRu bimetallic NPs were prepared by using a slight modification of the colloidal reduction method developed by Watanabe et al.10 In brief, the pH of equimolar aqueous H2PtCl6 and RuCl3 solutions was adjusted to 7 and 4, respectively, with 0.6 M Na2CO3 and reduced by using NaHSO3 to their corresponding intermediate compounds. To each compound hydrogen peroxide was added and again the pH was maintained at 5 by using 1 M NaOH. These two solutions were then mixed and the pH was maintained at 5. Later, an appropriate amount of Vulcan XC72R carbon was added, and the mixture was mixed and heated at 100 °C for 8 h. The resulting colloidal product was then washed with ultrapure water and dried. Hydrogen reduction was performed on the colloidal product at 300 °C for 2 h to achieve carbon-supported Pt-Ru bimetallic NPs. XRD, EDX, and TEM Measurements. Powder X-ray diffraction (XRD) patterns for the final Pt-Ru/C sample were obtained on a diffractometer (Rigaku Dmax-B, Japan) using a Cu KR source that was operated at 40 kV and 100 mA. The X-ray diffractogram was obtained at a scan rate of 0.05 deg s-1 for 2θ values between 20° and 90°. The EDX measurements were performed with a JSM 6500 EDX analyzer. Transmission electron microscopy (TEM) examination was performed on a JEOL JEM-1010 microscope that was operated at an accelerating voltage of 200 kV. Specimens were prepared by ultrasonically suspending the catalyst powders in ethanol, applying the specimen to a copper grid and drying in air. XAS Measurements. The X-ray absorption spectra were recorded at the Taiwan Beam Line of BL12B2 at the Spring-8, Hyogo, Japan. The electron storage ring was operated at 8 GeV. A double Si(111) crystal monochromator was employed for energy selection with a resolution ∆E/E better than 1 × 10-4 at both the Pt LIII-edge (11 564 eV) and the Ru K-edge (22 117 eV). All the reaction sequences during the formation of PtRu/C bimetallic nanoparticles were conducted in a homemade cell made with PTFE for XAS study. Two holes were made, one on top of the cell and the other on one side. After placing the liquid samples, the top hole was closed with a Teflon rod to avoid the exposure of the sample to the outer atmosphere. A hollow Teflon rod with a Kapton film cap at one end was inserted into the other end in the XAS cell. The position of the Teflon rod was adjusted to reach the optimum absorption thickness (∆µx ≈ 1.0, ∆µ is the absorption edge, x is the thickness of the liquid layer) so that the proper edge jump step could be achieved during the measurements. All of the spectra were recorded at room temperature in a transmission mode. Higher harmonics were eliminated by detuning the double crystal Si(111) monochromator. Three gas-filled ionization chambers were used in series to measure the intensities of the incident beam (I0), the beam transmitted by the sample (It), and

Formation of C-supported Pt-Ru Bimetallic Nanoparticles the beam subsequently transmitted by the reference foil (Ir). The third ion chamber was used in conjunction with the reference sample, which was a Pt foil for Pt LIII-edge measurements and Ru powder for Ru K-edge measurements. The control of parameters for EXAFS measurements, data collection modes, and calculation of errors were all done as per the guidelines set by the International XAFS Society Standards and Criteria Committee.22 EXAFS Data Analysis. The XAS experimental data were treated by utilizing the standard procedures. The EXAFS function, χ, was obtained by subtracting the postedge background from the overall absorption and then normalized with respect to the edge jump step. The normalized χ(E) was transformed from energy space to k-space, where “k” is the photoelectron wave vector. The χ(k) data were multiplied by k2 to compensate for the damping of EXAFS oscillations in the high k-region. Subsequently, k2-weighted χ(k) data in k-space ranging from 3.6 to 12.6 Å-1 for the Pt LIII-edge and from 3.6 to 11.6 Å-1 for the Ru K-edge were Fourier transformed (FT) to r-space to separate the EXAFS contributions from the different coordination shells. A nonlinear least-squares algorithm was applied to the curve fitting of an EXAFS in the r-space between 0.7 and 3.3 Å for both Pt and Ru depending on the bond to be fitted. The Pt-Ru reference file was determined by a theoretical calculation. Reference phase and amplitude for the Pt-Pt, Pt-O, and Pt-S absorber-scatterer pairs were obtained from a Pt foil, Na2Pt(OH)6, and H2PtCl6, respectively. For RuRu and Ru-O absorber-scatterer pairs the phase and amplitude were obtained from reference Ru powder and RuO2, respectively. All the computer programs were implemented in the UWXAFS 3.0 package23 with the backscattering amplitude and the phase shift for the specific atom pairs being theoretically calculated by using the FEFF7 code.24 From these analyses, structural parameters such as coordination numbers (N), bond distance (R) and the Debye-Waller factor (∆σj2) and inner potential shift (∆E0) have been calculated. The amplitude reduction factor, S02, values for Pt and Ru were obtained by analyzing the Pt foil and Ru powder reference samples, respectively, and by fixing the coordination number in the FEFFIT input file. The S02 values were found to be 0.95 and 0.88 for Pt and Ru, respectively. 3. Results and Discussion The Pt LIII-edge XANES data were obtained for all the reaction steps during the preparation of Pt-Ru nanoparticles and for Pt foil, PtO2, and H2Pt(OH)6 reference compounds and are compared in Figure 1. In the first step the pH of the beginning compound, i.e., H2PtCl6, is adjusted to 7 and XANES spectra are recorded; the first peak on the rising edge at 11566 eV results from the electronic transition from a 2p3/2 to the unoccupied “d” states near or above the Fermi level. This sharp peak is generally called as white line and its intensity is sensitive to the degree of electron occupancy in the valence orbits of the absorber.25 Generally speaking, the lower the white line intensity, the higher the electron density and the lower the oxidation state of Pt. Hence, changes in the white line intensity may be regarded as an indication of the change in the oxidation state of Pt. Another feature appearing at, postedge, 11 580 eV in the spectrum of H2PtCl6 is very much similar to the one found by Ankudinov et al. in the XANES spectra of Pt(IV) chlorides.26 This peak is assigned as a hybridization peak by these authors and it arises due to the hybridization of the Pt d photoelectron state with the unoccupied atomic Cl 3d states, mediated by multiple scattering. In the second step the addition of reducing

J. Phys. Chem. B, Vol. 110, No. 13, 2006 6477

Figure 1. In situ XANES spectra at Pt LIII-edge for various reaction steps during the formation of Pt-Ru bimetallic NPs. The XANES patterns of reference compounds Pt foil, PtO2, and H2Pt(OH)6 were also shown.

agent NaHSO3 to the H2PtCl6 witnessed a sharp decrease in the white line intensity and its XANES spectra indicate the decrease in Pt oxidation state (from +4 to +2). However, chemical speciation of the compound formed in this step is rather complicated with only XANES results and we will confirm it later during the discussion of EXAFS results. In the third step we have added hydrogen peroxide to the intermediate complex species and in the corresponding XANES spectra there is a sharp increase in white line intensity. Comparison of the spectrum with those of reference H2Pt(OH)6 and PtO2 compounds reasonably suggests the species formed are similar to H2Pt(OH)6, which is further supported by EXAFS analysis, indicating that the Pt oxidation state is changed from +2 to +4. After mixing the H2Pt(OH)6 species with RuO2, heat treatment at 100 °C for about 8 h decreases the white line intensity probably due to the presence of both metallic and ionic contributions of Pt and Ru. We relate this observation to the well-mixed state of Pt and Ru oxidic species and initiation of the reduction reaction. It is interesting that hydrogen peroxide, which acts as an oxidizing agent in the previous step, i.e., during the conversion of Pt(SO3)42- complex ion to Pt(OH)62-, will play a different role when RuO2 is added to H2Pt(OH)6. The presence of RuO2 species probably catalyzes the decomposition of hydrogen peroxide present in excess in the reaction medium and facilitates the in situ reduction of Pt and Ru oxides to a certain extent and results in bimetallic Pt-Ru and ionic contributions in the species. This phenomena observed in the present study is consistent with the findings of Watanabe et al., who observed the formation of bimetallic Pt-Ru clusters when RuO2 species were added to platinum oxidic species as was evidenced by XPS studies.10 Even though our XAS results reasonably suggest the presence of both Pt and Ru metallic and ionic contributions in the colloidal product, obtained when H2Pt(OH)6 species are mixed with RuO2 followed by heat treatment at 100 °C for about

6478 J. Phys. Chem. B, Vol. 110, No. 13, 2006

Hwang et al.

Figure 3. FT-EXAFS spectra at the Pt LIII-edge of various reaction steps during the formation of Pt-Ru bimetallic NPs; reference compounds are Pt foil, PtO2, and H2Pt(OH)6. Figure 2. Ru K-edge in situ XANES spectra for various reaction steps during the formation of Pt-Ru bimetallic NPs. The XANES patterns of reference compounds Ru powder and RuO2 were also shown.

8 h, the exact mechanism of this step needs further study and we will address this in our future contributions. The XANES spectra recorded after performing the H2 reduction at 300 °C for 2 h on the reaction mixture are comparable with that of Pt foil spectra indicating the complete reduction of Pt ions to metallic Pt and the oxidation state of Pt is decreased from +4 to 0. XANES analysis at the Ru K-edge of all the reaction steps and reference compounds Ru powder and RuO2 is shown in Figure 2. Upon NaHSO3 addition, the spectrum of RuCl3 changed significantly. The edge energy (the first inflection point on the rapidly rising portion of the absorption edge) has shifted about 2 eV to a lower energy, indicating that Ru has been reduced from +3 to +2. We believe that the reduced species are in the form of [Ru(OH)4]2-, which will be confirmed later with the EXAFS results. After the addition of hydrogen peroxide to the intermediate [Ru(OH)4]2- species the edge position is comparable to the near-edge structure of the RuO2 reference compound indicating that Ru has been oxidized from +2 to +4. From these changes and comparison with the reference compound, we infer that the local structure of the formed compound is similar to that of RuO2. However, when we mix the RuO2 species with H2Pt(OH)6 followed by heat treatment at 100 °C for about 8 h the Ru edge energy shifts to lower values indicating that formation of a mixed colloid of Pt and Ru oxidic species further supports the enhanced mixing of these species and initiation of the reduction reaction. The XANES spectrum recorded after performing the H2 reduction at 300 °C for 2 h on the reaction mixture closely resembles the Ru reference spectrum indicating the complete reduction of Ru ions to metallic Ru and the oxidation state of Ru is decreased from +4 to 0. The magnitude of the k2-weighted Pt LIII-edge FT EXAFS spectra for all the reaction steps and reference compounds PtO2 and H2Pt(OH)6 is shown in Figure 3. The transform for the

starting compound H2PtCl6 exhibits a strong peak at 2.0 Å characteristic of the presence of a Pt-Cl bond. The position and nature of the peak and the number of chlorine atoms are found to be comparable with the constants given in the literature.27 The magnitude of the peak corresponding to the Pt-Cl bond decreases upon the addition of NaHSO3 to H2PtCl6 indicating the progressive reduction of Pt4+ ions. Analysis of the EXAFS spectra of this stage revealed that the framework of the [PtCl6]2- is destroyed upon the addition of NaHSO3. The characteristic Pt-Cl bond of the precursor is not observed and the Fourier transform exhibits feature characteristics of only Pt-S (2.31 Å) and Pt-O (2.91 Å) bonds. The coordination of sulfur and oxygen around platinum is found as 3.9 and 2.9, respectively. At this stage it is reasonable to expect that the Pt ions are surrounded by 4 SO32- groups. However, the bond distance of Pt-O (2.91 Å) observed in this step is quite large when compared to the bond distance of Pt-O (2.07 Å) in PtO2 reference compound.28 Hence, we believe that the contribution of oxygen around Pt comes from the attached SO32- groups as a result of multiple scattering. Both the XANES and EXAFS results suggest that the precipitated species formed in this step are like [PtII(SO3)4]6-. Petrow and Allen found six sodium and four moles of SO32- per atom of platinum when NaHSO3 is added to the H2PtCl6 solution with a pH controlled at 7 and the present results are consistent with their findings.29 They have prepared a corresponding platinum sulfite complex of the form H3Pt(SO3)2OH from Na6[Pt(SO3)4] by replacing sodium atoms via treatment with strong acid resin. They found that the addition of H2O2 to the platinum sulfite complex formed produced platinum oxide species as shown in eq i.27

H3Pt(SO3)2OH + 3H2O2 f PtO2 + 3H2O + 2H2SO4 (i) Watanabe et al.10 have suggested that the white precipitate obtained after the addition of H2O2 to the platinum sulfite complex acid consists of not only PtO2 but also mixtures of oxides at least in two oxidation states. They reached this conclusions based on the comparison of XPS of the white precipitate with the XPS on pure platinum.

Formation of C-supported Pt-Ru Bimetallic Nanoparticles

J. Phys. Chem. B, Vol. 110, No. 13, 2006 6479

TABLE 1: EXAFS Fitting Parameters at the Pt LIII-Edge for Various Reaction Steps during the Formation of Pt-Ru Nanoparticles calcd parameters reaction steps H2PtCl6 H2PtCl6 + NaHSO3 H2PtCl6 + NaHSO3 + H2O2 after mixing Pt(IV) + Ru(IV), heat treatment at 100 °C after H2 reduction at 300 °C, 2 h on colloidal product

shell

N

R (Å)

∆σj2 (Å2)

∆E0 (eV)

Pt-Cl Pt-S Pt-O Pt-O Pt-O Pt-Pt Pt-Ru Pt-Ru Pt-Pt

6.0((0.5) 3.9((0.1) 2.9((0.4) 5.9((1.2) 2.8((0.9) 0.8((0.2) 1.3((0.7) 1.9((0.6) 6.9((0.7)

2.32((0.00) 2.31((0.00) 2.91((0.01) 2.00((0.05) 2.05((0.01) 2.76((0.05) 2.73((0.04) 2.70((0.02) 2.74((0.02)

0.0026 0.0039 0.0046 0.0023 0.0074 0.0000 0.0073 0.0022 0.0055

8.34 6.48 28.8 5.87 6.50 -5.14 4.30 5.20

However, careful examination of XAS results of the present study indicates that when the anionic complex of [PtII(SO3)4]6is oxidized with hydrogen peroxide the magnitude of the peak corresponding to the Pt-S bond is decreased and a new peak at 1.62 Å corresponding to the presence of the Pt-O bond (NPt-O, 5.9) is observed. The corresponding FTs closely match with that of reference H2Pt(OH)6 rather than PtO2. This observation reveals that the species formed are more like H2Pt(OH)6. The following possible reactions are deduced from the XAS results of the present study:

H2PtCl6 + Na2CO3 f Na2PtCl6 + H2O + CO2

(ii)

Na2PtCl6 + 6NaHSO3 + 2H2O f Na6[Pt(SO3)4] + 2H2SO4 + 2NaCl + 4HCl (iii) Na6[Pt(SO3)4] + 5H2O2 f H2Pt(OH)6 + H2SO4 + 3Na2SO4 (iv) Mixing of H2Pt(OH)6 with RuO2 species followed by heat treatment at 100 °C for 8 h produces Pt-Pt and Pt-Ru bonds apart from the Pt-O bonds. However, hydrogen reduction at 300 °C for 2 h on this colloidal mixture led to the appearance of Pt-Ru and Pt-Pt bonds only with bond lengths 2.70 and 2.74 Å, respectively. The best fit EXAFS parameters (N, coordination number; R, bond distance; σ2, Debye-Waller factor; ∆E0, inner potential shift) are summarized in Table 1. The FT EXAFS spectra collected at the Ru K-edge for different reaction steps and reference compounds Ru powder and RuO2 are shown in Figure 4. The different coordination numbers and interatomic distances of the first shell, obtained by fitting the data, are given in Table 2. The RuCl3 FT-EXAFS spectrum shows a maximum between 1.4 and 2.3 Å corresponding to the nearest chlorine neighbors of Ru. The NRu-Cl is found to be 6.0 with a Ru-Cl bond length of 2.34 Å. After the addition of NaHSO3 the magnitude of the FT peak of Ru-Cl is decreased and the position of the peak is slightly shifted to lower R values indicating change in the ligand environment from chloride to oxygen. We found oxygen coordination around Ru (NRu-O, 3.9) with a bond length of 2.12 Å. The Ru-O bond length value is similar to Ru-O in Ru(OH)x compounds.30 This observation indicates that the species formed during this step are in the form of [Ru(OH)4]2-. Upon addition of H2O2 the FT-EXAFS spectrum exhibits a peak at 1.2 Å corresponding to the nearest neighbors of Ru: six O atoms at 2.05 Å; and it closely resembles that of the reference RuO2 compound peak indicating the local structure of the obtained species is similar to that of RuO2.

Figure 4. FT-EXAFS spectra at the Ru K-edge of various reaction steps during the formation of Pt-Ru bimetallic NPs; reference compounds are Ru powder and RuO2.

The corresponding chemical reactions can be written as follows:

2RuCl3 + 6NaHSO3 + 4H2O f 2Na2[Ru(OH)4] + Na2SO4 + 6HCl + 5SO2 (v) Na2[Ru(OH)4]+ H2O2 f RuO2 + 2NaOH + 2H2O

(vi)

Later when we mixed RuO2 with the H2Pt(OH)6 species and followed by heat treatment at 100 °C for 8 h, we observed the Ru-O, Ru-Ru, and Ru-Pt coordination in the FT-EXAFS spectra and their coordination numbers are found to be 4.1 (NRu-O), 1.6 (NRu-Ru), and 1.2 (NRu-Pt). The observation of Ru and Pt coordination around Ru in this step is believed to be due to the formation of a mixed colloidal sol of unknown composition containing both Pt and Ru ionic as well as metallic contributions. After H2 reduction Pt coordination around Ru is observed (NRu-Pt, 1.9) with a bond length of 2.70 Å, revealing the formation of Pt-Ru species. It is also evidenced from the XRD pattern of Pt-Ru/C particles as shown in Figure 5 that the characteristic platinum fcc peaks (111), (200), (220), and (311) are shifted slightly to higher 2θ values. If the homogeneous solid-solution of Pt-Ru is not formed, then the XRD spectra of pure Ru in an hcp structure would be observed in the scan. However, in the XRD scan no observable peaks correspond to those of tetragonal RuO2 and of the hcp structure of pure ruthenium. The increase in 2θ values corresponds to a

6480 J. Phys. Chem. B, Vol. 110, No. 13, 2006

Hwang et al.

TABLE 2: EXAFS Fitting Parameters at the Ru K-Edge for Various Reaction Steps during the Formation of Pt-Ru Nanoparticles calcd parameters reaction steps

shell

N

R (Å)

∆σj2 (Å2)

∆E0 (eV)

RuCl3 RuCl3 + NaHSO3 RuCl3 + NaHSO3 + H2O2 after mixing Ru(IV) + Pt(IV), heat treatment at 100 °C

Ru-Cl Ru-O Ru-O Ru-O Ru-Ru Ru-Pt Ru-Ru Ru-Pt

6.0((0.3) 3.9((0.7) 6.0((0.5) 4.1((0.6) 1.6((0.5) 1.2((0.7) 4.7((0.9) 1.9((0.6)

2.34((0.01) 2.12((0.01) 2.05((0.05) 2.00((0.04) 2.68((0.01) 2.73((0.04) 2.67((0.02) 2.70((0.04)

0.0096 0.0012 0.0124 0.0086 0.0101 0.0056 0.0059 0.0011

-2.21 9.30 1.92 -5.28 -6.59 0.92 3.95 0.92

after H2 reduction at 300 °C, 2 h on colloidal product

Figure 5. XRD patterns of Pt-Ru/C nanoparticles obtained by a modified-Watanabe process.

decrease in the lattice constants due to the incorporation of Ru atoms. Such incorporation in the fcc structure of platinum indicates the formation of Pt-Ru alloy particles.31 The average particle size of the Pt-Ru/C particle was determined by using the peak associated with the (220) plane of fcc Pt by using Scherrer’s equation and found to be 2.9 nm.32 Formation Mechanism of Bimetallic Pt-Ru/C NPs. On the basis of the XAS results we have attempted to discuss the formation mechanism of bimetallic Pt-Ru/C NPs. Comparing the FT-EXAFS spectra and fitting results of both Pt LIII-edge and Ru K-edge a model is proposed for the mechanism of PtRu/C NPs formation and is shown in Scheme 1. From Pt LIIIedge XAS, for the beginning compound H2PtCl6 we found NPt-Cl ) 6.0, and it shows that the Pt4+ ion is surrounded by six chloride ions. Upon addition of NaHSO3 we observed the contribution of S around Pt (NPt-S, 3.9). The EXAFS results suggest that the species formed at this stage is in the form of [Pt(SO3)4]6- in which the Pt2+ ion is surrounded by four SO32ionic groups. Addition of H2O2 to these species increases NPt-O to 5.9 with a bond length of 2.00 Å, which is comparable with that of the Pt-OH bond length indicating that Pt2+ ions are oxidized to Pt4+ and surrounded by six hydroxide ions (all these reaction steps are schematically shown in Scheme 1A). Ru K-edge XAS analysis reveals NRu-Cl ) 6.0 for RuCl3 and after the addition of NaHSO3 the contribution from Ru-O coordination with NRu-O ) 3.9 similar to the [Ru(OH)4]2- species appeared. Upon addition of H2O2, the Ru-O coordination with NRu-O ) 6.0 similar to that of the RuO2 species is obtained (see Scheme 1B). Later mixing the RuO2 species with the H2Pt(OH)6 and heating this mixture at 100 °C for about 8 h produces Ru and Pt coordination around Ru of 1.6 and 1.2, respectively. Similarly, Pt and Ru coordination around Pt is

Figure 6. TEM image of Pt-Ru/C nanoparticles obtained by a modified-Watanabe process.

found to be 0.8 and 1.3, respectively. The oxygen contribution around Pt and Ru is found to be 2.8 and 4.1, respectively. When hydrogen reduction is performed on this mixture NRu-Pt is increased to 1.9 and NPt-Ru increased to 1.9 revealing the formation of Pt-Ru bimetallic NPs (see Scheme 1C). In our previous results of Pt nanoparticles formation in AOT reverse micelles we observed that the reduction of Pt ions which are surrounded by OH- ligands produces well-dispersed Pt nanoparticles with a particle size down to 2.0 nm.11a As the reduction pathway of the present investigation also involves the OHligand environment around Pt ions we expect that the particles formed are well-dispersed. The representative TEM image of the Pt-Ru/C particle as shown in Figure 6 reveals that metal particles of high contrast are well-dispersed over the surface of the carbon. The particle diameter of 2-3 nm was obtained for Pt-Ru/C from TEM measurements. The coordination number derived from XAS is a strong and nonlinear function of the particle diameter up to 3-5 nm. This property has been widely used in EXAFS analysis to determine the size of the nanoparticle.12 In the resulting Pt-Ru clusters the total coordination number of Pt and Ru around absorbing “Pt” atoms (NPt-Pt + NPt-Ru ) 8.8) is similar to the total coordination number of Ru and Pt around absorbing “Ru” atoms (NRu-Ru + NRu-Pt ) 6.6) indicating that the size of the Pt-Ru clusters is between 2 and 2.5 nm. The particle size obtained from XRD is in good agreement with the TEM and XRD measurements. Composition and Atomic Distribution of Bimetallic PtRu/C NPs. The EDX measurements on the final Pt-Ru/C

Formation of C-supported Pt-Ru Bimetallic Nanoparticles

J. Phys. Chem. B, Vol. 110, No. 13, 2006 6481

SCHEME 1: Schematic Presentation of All the Reaction Steps during the Formation of Pt-Ru Bimetallic NPs

product give a Pt/Ru atomic ratio of 57:43 (spectra not shown here). The composition of Pt-Ru/C is also calculated from XAS by measuring the edge jump at Pt LIII-edge and Ru K-edge and it is found that the atomic ratio of Pt:Ru is 1:0.94 (Pt0.52Ru0.48). We note that XAS data presented here sample many nanoparticles and are not an on-particle method of characterization. However, as the particles obtained are uniformly distributed we believe that the compositional variation is insignificant in the present study. Recently we have explored an XAS based methodology to determine the atomic distribution or alloying extent of bimetallic nanoparticles.13 We have applied this methodology for the Pt-Ru/C system investigated here. In the case of Pt-Ru bimetallic clusters after H2 reduction the coordination numbers of Pt and Ru atoms around the Pt atom are found to be 6.9 and 1.9, respectively, and the total coordination number ∑NPt-i is 8.8. The coordination numbers of Ru and Pt atoms around the Ru atom are determined as 4.7 and 1.9, respectively, and the total coordination number ∑NRu-i is calculated as 6.6. From these values the structural parameter Pobserved (NPt-Ru/∑NPt-i) and Robserved (NRu-Pt/∑NRu-i) was calculated as 0.22 and 0.29, respectively. Then the alloying extent of Pt (JPt) and Ru (JRu) values is calculated by using eqs 1 and 2, respectively.

JPt )

Pobserved × 100 PRandom

(1)

JRu )

RObserved × 100 RRandom

(2)

where Prandom and Rrandom can be taken as 0.5 for perfect alloyed bimetallic NPs if the atomic ratio of “Pt” and “Ru” is 1:1. This

value can be achieved by assuming NPt-Pt ) NPt-Ru and NRu-Ru ) NRu-Pt, which is generally true for perfect alloyed bimetallic NPs. By using eqs 1 and 2 the alloying extent of Pt (JPt) and Ru (JRu) values is calculated as 44% and 58%, respectively. The higher value of Robserved, 0.29, and JRu, 58%, indicates the higher extent of atomic dispersion or alloying extent of “Ru” atoms when compared to Pt. The observed parameter relationship ∑NPt-i > ∑NRu-i and JRu > JPt indicates that the obtained Pt-Ru/C NPs adopt a Pt rich in core and Ru rich in shell structure and the schematic representation of the structure is given in Figure 7. The observed parameter relationship, i.e., ∑NPt-i > ∑NRu-i for the Pt-Ru/C nanoparticles investigated here, is consistent with the relationship NAA + NAB > NBA + NBB for a homogeneous system of A-B bimetallic NPs for which the core of the cluster is composed of N atoms of A (NA) and the surface is made of N atoms of B (NB), the total coordination number (NAA + NAB) for the “A” atom and greater than the total coordination for the “B” atoms (NBA + NBB).33

Figure 7. Structural model deduced for the obtained Pt-Ru bimetallic NPs based on XAS structural parameters.

6482 J. Phys. Chem. B, Vol. 110, No. 13, 2006 From the quantitative extent of alloying values we can see that in the catalyst nanoparticles a considerable amount of Ru is dominated in the shell region with a lesser extent of segregation. The larger JRu value indicated that most of the Ru is involved in alloying and hence less segregation of Ru in the shell region. The lower JPt value reveals that in the core homometallic Pt-Pt bonds are preferred rather than heterometallic Pt-Ru bonds. Thus the results obtained from XAS support the Pt-rich core and Ru-rich shell structure for the present carbon-supported Pt-Ru nanoparticles (see Figure 7). 4. Summary and Conclusions The studies described in this paper indicate that XAS-based methodology is promising to elucidate the formation mechanism of bimetallic NPs. The XAS results demonstrate that the reduction of platinum hydroxide and ruthenium oxide species under flowing hydrogen resulted in metallic Pt-Ru nanoparticles. XAS results of the present study reveal that the mixing of Pt4+ ions with a ligand environment of OH- groups and Ru4+ ions surrounded by oxygen groups at 100 °C for 8 h prior to H2 reduction initiates the reduction reaction as evidenced by the presence of Pt and Ru bimetallic and ionic contribution and is beneficial to enhance the atomic distribution and is suitable for the formation of well-dispersed Pt-Ru/C nanoparticles. Even though our XAS results reasonably suggest the presence of both Pt and Ru metallic and ionic contributions in the colloidal product the exact mechanism of this step needs further study and we will address this in our future contributions. The carbonsupported Pt-Ru nanoparticles thus obtained have a structure similar to that of Pt-rich core and Ru-rich shell. On the basis of the XAS structural parameters we found that atomic-scale distribution of Ru is much better than that of Pt. The proposed methodology is quite general and easy to extend to study the formation mechanism of other metallic clusters. Acknowledgment. The financial support from the National Science Council (under contract numbers NSC93-2811-E-011008, NSC94-2214-E-011-010, and NSC94-2120-M-011-002), facilities from the National Synchrotron Radiation Research Center (NSRRC), and the National Taiwan University of Science and Technology, Taiwan, R.O.C., is gratefully acknowledged. References and Notes (1) Hogarth, M. P.; Hards, G. A. Platinum Met. ReV. 1996, 40, 150. (2) Watanabe, M.; Motto, S. J. Electroanal. Chem. 1975, 60, 267. (3) (a) Lu, C.; Rice, C.; Masel, M. I.; Babu, P. K.; Waszczuk, P.; Kim, H. S.; Oldfield, E.; Wieckowski, A. J. Phys. Chem. B 2002, 106, 9581. (b) Waszczuk, P.; Lu, G. U.; Wieckowski, A.; Lu, C.; Rice, C.; Masel, M. I. Electrochim. Acta 2002, 47, 36. (4) Waszczuk, P.; Wieckowski, A.; Zelenay, P.; Gottesfeld, S.; Coutanceau, C.; Le´ger, J.-M.; Lamy, C. J. Electroanal. Chem. 2001, 511, 55. (5) (a) Schmidt, T. J.; Noeske, M.; Gasteiger, H. A.; Behm, R. J.; Britz, P.; Brijouz, W.; Bo¨nnemann, H. Langmuir 1997, 13, 2591. (b) Vogel, W.; Britz, P.; Bo¨nnemann, H.; Rothe, J. J. Phys. Chem. B 1997, 101, 11029. (c) Bo¨nnemann, H.; Brinkmann, R.; Britz, P.; Endruschat, U.; Mortel, R.; Paulus, U. A.; Feldmeyer, G. J.; Schmidt, T. J. J. New Mater. Electrochem. Syst. 2000, 3, 199. (6) (a) Zhang, X.; Chan, K.-Y. Chem. Mater. 2003, 15, 451. (b) Liu, Y.; Qiu, X.; Chen, Z.; Zhu, W. Electrochem. Commun. 2002, 4, 550. (7) Boxall, D. L.; Deluga, G. A.; Kenik, E. A.; King, W. D.; Lukehart, C. M. Chem. Mater. 2001, 13, 891.

Hwang et al. (8) (a) Wang, X.; Hsing, I.-M. Electrochim. Acta 2002, 47, 2981. (b) Sarma, L. S.; Lin, T.-D.; Tsai, Y.-W.; Chen, J.-M.; Hwang, B.-J. J. Power Sources 2005, 139, 44. (9) (a) Nashner, M. S.; Frenkel, A. I.; Somerville, D.; Hills, C. W.; Shapley, J. R.; Nuzzo, R. G. J. Am. Chem. Soc. 1998, 120, 8093. (b) Nashner, M. S.; Frenkel, A. I.; Adler, D. L.; Shapley, J. R.; Nuzzo, R. G. J. Am. Chem. Soc. 1997, 119, 7760. (10) Watanabe, M.; Uchida, M.; Motoo, S. J. Electroanal. Chem. 1987, 229, 395. (11) (a) Tsai, Y.-W.; Tseng, Y.-L.; Sarma, L. S.; Liu, D.-G.; Lee, J.-F.; Hwang, B.-J. J. Phys. Chem. B 2004, 108, 8148. (b) Hwang, B.-J.; Tsai, Y.-W.; Sarma, L. S.; Liu, D.-G.; Lee, J.-F. J. Phys. Chem. B 2004, 108, 20427. (c) Chen, C.-H.; Hwang, B.-J.; Wang, G.-R.; Sarma, L. S.; Tang, M.-T.; Liu, D.-G.; Lee, J.-F. J. Phys. Chem. B 2005, 109, 21566. (12) Frenkel, A. I.; Hill, C. W.; Nuzzo, R. G. J. Phys. Chem. B 2001, 105, 12689. (13) Hwang, B.-J.; Sarma, L. S.; Chen, J.-M.; Chen, C.-H.; Shih, S.-C.; Wang, G.-R.; Liu, D.-G.; Lee, J.-F.; Tang, M.-T. J. Am. Chem. Soc. 2005, 127, 11140. (14) Bock, C.; Paquet, C.; Couillard, M.; Botton, G. A.; MacDougall, B. R. J. Am. Chem. Soc. 2004, 126, 8028. (15) (a) Zhang, Z. C.; Lei, G. D.; Sachtler, W. M. H. In Series on Synchrotron Radiation Techniques and Applications; Iwasawa, Y., Ed.; World Scientific: Singapore, 1996; Vol. 2, p 137. (b) Ashcroft, A. T.; Cheetham, A. K.; Harris, P. J. F.; Jones, R. H.; Natarajan, S.; Sankar, G.; Stedman, N. J.; Thomas, J. M. Catal. Lett. 1994, 24, 47. (16) Cao, D.; Bergens, H. J. Power Sources 2004, 134, 170. (17) (a) Teo, B. K. EXAFS. In Basic Principles and Data Analysis; Springer: Berlin, Germany, 1986. (b) Prins, R.; Koningsberger, D. E., Eds. In X-ray Absorption: Principles, Applications, Techniques of EXAFS, SEXAFS, and XANES; Wiley-Interscience: New York, 1988. (18) (a) Camara, G. A.; Giz, M. J.; Paganin, V. A.; Ticianelli, E. A. J. Electroanal. Chem. 2002, 537, 21. (b) Russell, A. E.; Rose, A. Chem. ReV. 2004, 104, 4613. (c) Hills, C. W.; Nashner, M. S.; Frenkel, A. I.; Shapley, J. R.; Nuzzo, R. G. Langmuir 1999, 15, 690. (d) Toshima, N.; Harada, M.; Yamazaki, Y.; Asakura, K. J. Phys. Chem. 1992, 96, 9927. (e) Toshima, N.; Harada, M.; Yonezawa, T.; Kushihashi, K.; Asakura, K. J. Phys. Chem. 1991, 95, 7448. (f) Bian, C.-R.; Suzuki, S.; Asakura, K.; Ping, L.; Toshima, N. J. Phys. Chem. B 2002, 106, 8587. (19) (a) Ankudinov, A. L.; Rehr, J. J.; Low, J. J.; Bare, S. R. J. Chem. Phys. 2002, 116, 1911. (b) Ankudinov, A. L.; Rehr, J. J.; Low, J.; Bare, S. R. Phys. ReV. Lett. 2001, 86, 1642. (c) Bazin, D.; Sayers, D.; Rehr, J.; Mottet, C. J. Phys. Chem. 1997, 101, 5332. (20) Bazin, D.; Rehr, J. J. J. Phys. Chem. B 2003, 107, 12398. (21) Dassenoy, F.; Casanove, M.-J.; Lecante, P.; Pan, C.; Philippot, K.; Amiens, C.; Chaudret, B. Phys. ReV. B 2001, 63, 235407. (22) (a) See, for example, the guidelines for data collection modes for EXAFS measurements and user controlled parameters: http://ixs.iit.edu/ subcommittee_reports/sc/sc00report.pdf. (b) See, for example, the guidelines for error reporting: http://ixs.iit.edu/subcommittee_reports/sc/err-rep.pdf. (23) Stern, E. A.; Newville, M.; Ravel, B.; Yacoby, Y.; Haskel, D. Phys. B (Amsterdam. Neth.) 1995, 208-209, 117. (24) Zabinsky, S. I.; Rehr, J. J.; Anukodinov, A. L.; Albers, R. C.; Eller, M. J. Phys. ReV. B 1995, 52, 2995. (25) Horsely, J. A. J. Chem. Phys. 1982, 76, 1451. (26) Ankudinov, A. L.; Rehr, J. J.; Simon, R. B. Chem. Phys. Lett. 2000, 316, 495. (27) Ayala, R.; Sa´nchez Marcos, E.; Dı´az-Moreno, S.; Armando Sole´, V.; Mun˜oz-Pa´ez, A. J. Phys. Chem. B 2001, 105, 7598. (28) Herron, M. E.; Doyle, S. E.; Pizzini, S.; Roberts, K. J.; Robinson, J.; Hards, G.; Walsh, F. C. J. Electroanal. Chem. 1992, 324, 243. (29) Petrow, H. G.; Allen, R. J. U.S. Patent No. 4044193, 1977. (30) Paˆrvulescu, V. I.; Coman, S.; Palade, P.; Macovei, D.; Teodorescu, C. M.; Filoti, G.; Molina, R.; Poncelet, G.; Wagner, F. E. Appl. Surf. Sci. 1999, 141, 164. (31) Arico`, A. S.; Creti, P.; Kim, H.; Mantegna, R.; Giordano, N.; Antonucci, V. J. Electrochem. Soc. 1996, 143, 3047. (32) Warren, B. E. In X-ray Diffraction; Addison-Wesley: Reading, MA, 1996. (33) (a) Via, G. H.; Sinfelt, J. H. In X-ray Absorption Fine Structure for Catalysts and Surfaces; Iwasawa, Y., Ed.; World Scientific: London, UK, 1996. (b) Bazin, D.; Sayers, D.; Rehr, J. J. J. Phys. Chem. 1997, 101, 11040.