XRD analysis technique was performed to verify the crystal phase of Pt-Ce nanoalloy. Figure 1a shows XRD pattern of the Pt-Ce nanoalloy under four different annealing temperatures. Three main peaks at 39.76°, 46.24° and 67.45° were indexed as (111), (200) and (220) crystal faces respectively, for crystalline face centered cubic (fcc) Pt (PDF 04-0802). The obvious shifts of Pt peaks are observed. It is most likely due to the CeCl3 was firstly reduced to Ce element and then Ce atoms diffused into the Pt crystal to expand the lattice of Pt. Additionally, for the two samples with the annealing temperatures of 800℃ and 900℃, there are many weaker peaks. After matching with the standard PDF files, we found that some of these peaks originate from Pt2Ce (PDF #17-0010), some from Pt5Ce (PDF #17-0071), but the rest peaks locate between the two standard card. Interestingly, the other pattern in Fig. 1b indicates the different annealing time hardly affected the Pt–Ce nanoalloy phase structure under annealing temperature 800 ℃. The average sizes of all samples were calculated by Scherrer’s equation. The results in Table 1 manifest the size of the Pt–Ce nanoalloy is significantly larger than that of Pt particles and the particle sizes increase with increasing annealing temperatures.
TEM was used to analyze the particle size, the dispersion and the structure of Pt–Ce nanoalloy. Figure 2 shows TEM images and size distribution histograms of the Pt–Ce nanoalloy catalysts prepared under the annealing temperatures of 600 °C and 800 °C. It can be seen that the particle sizes of Pt–Ce alloy catalysts prepared at different reduction temperatures have wide distribution. Correspondingly, the size distribution histograms in Fig. 2b,d also clearly proved this. The size of Pt–Ce alloy catalysts which ranged from 4.5 to 9.5 nm with average size of 5.0 nm at annealing temperature 600 °C. Similarly, the size distribution changed from 6.5 nm to 13.5 nm with average size of 7.0 nm when the annealing temperature increased to 800 °C. In addition, the results have also been obtained for the samples which were annealed under 700 °C and 900 °C. The corresponding statistics of Pt–Ce alloy catalysts were listed in the Table 1. It obviously displays the particle sizes of all Pt–Ce alloy catalysts obtained from the TEM are agreed well with that from XRD.
To further analyze the electronic structure and surface composition of Pt-Ce alloy catalyst, XPS spectrum of PtCe/C-800 ℃ sample was carried out. As shown in Fig. 3a, the Pt 4f XPS spectrum can be split into three pairs of peaks: the strongest couple emerged at 71.69 eV (Pt 4f7/2) and 75.21 eV (Pt 4f5/2) corresponds to Pt0 and the stronger pair presented at 72.52 eV (Pt 4f7/2) and 75.98 eV (Pt 4f5/2) is derived from Pt2+, while the weakest pair located at 73.68 eV (Pt 4f7/2) and 77.04 eV (Pt 4f5/2) is assigned to Pt4+ species21. Meanwhile, the Ce 3d XPS spectra in Fig. 3b is disintegrated into two couples of peaks: one pair located at 884.59 eV (Ce 3d3/2) and 902.33 eV (Ce 3d5/2) is attributed to metal Ce22,23, while the couples at 886.65 eV (Ce 3d3/2) and 904.17 eV (Ce 3d5/2) were stemmed from Ce3+. According to the data, the vast majority of metallic Ce0 was the dominant state of cerium for Pt–Ce alloy catalyst. A trace amount of cerium oxide remained, it is mainly because cerium, as one of the most active elements among the rare earth metals, is particularly easy to be oxidized to Ce3+ even under hypoxic atmosphere.
The elemental composition of Pt–Ce nanoalloy was detected by the energy dispersive X-ray spectroscopy (EDX) and inductively coupled plasma optical emission spectrometer (ICP-OES). As listed in Table 2, it reveals that the Pt contents are higher than 97% when the annealing temperature below 700 °C. It means that a small amount of Ce have been reduced and diffused into the Pt nanoparticles. The Ce contents increased to 13.7% and 14.5% when the reduction temperature increased to 900 °C and 800 °C, the ratio is higher than the Ce contents in Pt5Ce (12.5%) but lower than that in Pt2Ce (26%). It suggests that the alloy catalysts are mixed alloy phases of Pt2Ce and Pt5Ce. This is agree well with the results of XRD analysis.
CV tests were performed in an aqueous N2-saturated 0.1 M HClO4 solution with a scan rate of 100 mV s−1. As presents in Fig. 4a, a well-defined under potential deposition Hydrogen domain (Hupd) between − 0.20 and − 0.05 V was exhibited. With the annealing temperature increases, the current density increases non-linearly. Although the current density reaches the maximum when annealing temperature reaches 700 ℃ or 800 ℃, there are no apparent changes between the four samples. At the same time, Fig. 4b shows the electrochemical active surface area (ECSA) of the four Pt–Ce nanoalloy catalysts, which determined by measuring the charge collected in the hydrogen desorption region and assuming a value of 210 µC cm−2 for a monolayer hydrogen adsorption. Obviously, the ECSA of the four samples changed in the range of 24 m2 gPt−1 to 30 m2 gPt−1, which smaller than that of Com Pt/C (70.63 m2 gPt−1) catalyst tested separately as a reference. And the variation trend of ECSA is agree well with the change trend of Ce concentration in the Pt–Ce nanoalloy catalysts. It further indicates the higher Ce concentration in Pt–Ce nanoalloy catalysts will enhance the ECSA of the nanoally.
Linear sweep voltammetry (LSV) measurements were also collected to confirm the ORR activities of the Pt-Ce alloy catalysts and Com Pt/C. Figure 5a shows the LSV curves of the catalysts in solution of O2-saturated 0.1 M HClO4 at a scan rate of 10 mV s−1 at 1,600 rpm. The half-wave potential are 0.45, 0.45, 0.45, 0.50 and 0.48 V (vs. Ag/AgCl) for Com Pt/C, PtCe-600 ℃, PtCe-700 ℃, PtCe-800 ℃ and PtCe-900 ℃ respectively. The mass activity (MA) in Fig. 5b of Com Pt/C, PtCe-600 °C and PtCe-700 ℃ are about 160 A gPt−1 (at 0.5 V vs. Ag/AgCl), while that of PtC-800 ℃ and PtCe-900 ℃ increase to 286 and 209 A gPt−1 respectively. It indicates that Pt–Ce alloy-structure catalysts can effectively improve the ORR activity, and the catalytic activity increase with the increasing of the Ce content. Correspondingly, Fig. 5c illustrates that all Pt–Ce catalysts have higher specific activity (SA) than Com Pt/C, and the SA also increases with increasing Ce content. In addition, the electrochemical stability of all samples and Com Pt/C was studied by Chronoamperometry (CA) technique in 0.1 M HClO4 at 0.5 V (vs. Ag/AgCl). As shown in Fig. 5d, all the curves show a sharp initial current drop in the first 300 s and then decay very slowly until 5000 s. In contrast to the previous reports on Pt5M polycrystalline24,25 (M = La, Ce or Gd etc.), the Ce doping did not significantly improve the stability of the Pt–Ce alloy catalyst in this work. This is mostly because the reduced Ce atoms will diffuse from the edge of the Pt particles into the nucleus. With the diffusion increases, the concentration of Ce atoms at the edge of the Pt particle gradually decrease, eventually a Ce skin is formed.