3.1 Functionalization of commercial carbon XC-72R
Functionalization of commercial carbon XC-72R was conducted by immersing XC-72R into pure sulfuric acid (97 wt%) and nitric acid (61 wt%) at 3:1 volume ratio for 1 h at 25 ℃. The pretreatment was imposed to increase the adhesion of chemically reduced Ru nanoparticles by providing suitable functional groups on the carbon support [39].
3.2 Fabrication of carbon supported Ru nanoparticles
Ru nanoparticles were deposited onto functionalized commercial carbon XC-72R via a chemical reduction method. Mixture 0.2 g of RuCl3and 0.8 g of functionalized XC-72R were dissolved in excess of de-ionized water. Appropriate amount of NaBH4 was then dissolved with de-ionized water and slowly added into the RuCl3/XC-72R mixture. Then, the mixture was filtered and washed by de-ionized water, followed by 80 ℃ heating in oven to remove the remaining solvent.
3.3 Samples for displacement reaction
The carbon supported Ru nanoparticles were deposited on a 2 × 2 cm2 carbon cloth (E-TEK) by an ink method. The ink was composed of 8 mg carbon supported Ru nanoparticles, 5 mg PTFE (30 wt%) and 5 mL 99.5 wt% ethanol, the ink was well-dispersed by supersonication for 30 minutes.
Subsequently, the ink dispersion was drop-wisely deposited on a 2 × 2 cm2 carbon cloth which was kept atop a hot plate at 80 ℃. After the residual solvent was dried, the Ru/C coated carbon clothes were immersed into 5 mM intrinsic hexachloroplatinic acid aqueous solution with three different pH values (1.0, 2.2 and 8.0). The nature for 5 mM hexachloroplatinic acid was pH 2.2, and the pH value was adjusted to pH1 by perchloric acid and pH 8 by potassium hydroxide, respectively. The immersion lasted for 24 hours at 40 ℃. The as-prepared samples were thoroughly rinsed with
16
de-ionized water, noted as group A. The second identical set of samples was fabricated, defined as group B, with the same fabrication process except the Ru/C was subjected to hydrogen reduction before the immersion in hexachloroplatinic acid solution. The third set of samples undergoing a hydrogen reduction after the immersion in hexachloroplatinic acid solution was labeled as group C.
The hydrogen reduction was conducted with pure hydrogen under 1 atm at 80 ℃ for two hr. For samples in reference groups, the immersion baths were replaced by perchloric acid (pH 1), potassium hydroxide (pH 8) and de-ionized water, respectively.
3.4 XRD, TEM, EDX and ICP-MS measurements
X-ray diffraction (XRD) patterns for the as-prepared samples were obtained by Max Science-M18XHF KXY-8019-1 with a Cu Kα source of 1.54 Å . The X-ray diffractogram was received at a scan speed of 4 degrees/second for 2θ values between 30° and 90°. Transmission Electron Microscopy (TEM) was performed to observe the morphology of the as-prepared nanoparticles using JEOL JEM-3000F with an accelerating voltage of 300 kV. Approximate element composition was acquired with energy-dispersive X-ray spectroscopy (EDX) equipped on the TEM. The exact metal loadings were determined by an inductively coupled plasma mass spectrometry (ICP-MS).
3.5 XAS measurements
X-ray absorption spectra (XAS) for Pt L3-edge (11,564 eV) and Ru K-edge (22,117 eV) were obtained at end stations BL01C1 and BL17C1 of the Taiwan Light Source (TLS), National Synchrotron Radiation Research Center (NSRRC), Hsinchu, Taiwan. The storage ring energy was 1.5 GeV, and the stored top-up ring current was in the range of 300-360 mA. A double Si(111) crystal monochromator was employed for energy selection with a resolution ΔE/E better than 2 × 10-4 at both end stations. Rh or Pt-coated mirrors were adopted to reject high-order harmonics,
17
collimate (upstream) and refocus (downstream). The XAS measurements were carried out in fluorescence mode of detection at 25 ℃. A Lytle fluorescence detector along with three gas-filled ionization chambers were used to collect the intensities of the X-ray fluorescence photons from an as-prepared sample (If), the incident beam (I0), the transmitted beam through the sample (It), and the transmitted beam through reference metal foil or powder (Ir). A Pt foil was served as reference for Pt L3-edge measurements and Ru powder was served as that for Ru K-edge measurements.
3.6 EXAFS data analysis
Extended X-ray absorption fine structure (EXAFS) data analysis and fitting were processed by IFEFFIT 1.2.11c data analysis package (Athena, Artemis, and FEFF6) [14-15]. The raw data were calibrated by aligning the scans against the reference and merging multiple scans as an average to achieve better signal quality. X-ray absorption near edge structure (XANES) spectra were obtained after appropriate process by Athena software. The EXAFS function was obtained by standard procedures with Athena including pre-edge and post-edge background subtractions, and normalization with respect to the edge jump. The detailed procedure has been reported elsewhere [16]. The resultant EXAFS function, χ(E), was then transformed from energy space to k-space. The value k refers to photoelectron wave vector. At the high k-region of χ(k) data, multiplication by k3 was conducted to compensate the damping of EXAFS oscillations. Next, k3 weighted χ(k) data were Fourier transformed to r-space. Specific ranges in k-space for the Fourier transformation were selected from 3.32 to 12.74 Å-1 for the Pt L3-edge and 4.01 to 13.42 Å-1 for the Ru K-edge. The EXAFS curve fitting in r-space was applied by a nonlinear least-square algorithm. Also, the r-space ranges for the curve fitting were established from 1.29 to 3.12 Å (without phase correction) for Pt and 1.32 to 2.73 Å for Ru. The structural parameters were fitted by Artemis with the theoretical standards generated by FEFF6 code [17]. The fitted structural parameters include the coordination number (N), bond distance (R), Debye–Waller factor (Δσj2), and inner potential shift (ΔE0). Two
18
assumptions were added in the EXAFS fitting for hexachloroplatinic acids. First, we assumed that the Pt ion (IV) complexes were octahedrally coordinated, which suggested that the sum of Pt-Cl and Pt-O cooridination numbers were kept at six for all complexes. Second, the Debye-Waller factor differences of Pt-Cl and Pt-O were assumed to be identical [12]. The amplitude reduction factor (S02
) for Ru were obtained by analyzing Ru powder and found to be 0.79.
19