Open AccessCCS ChemistryRESEARCH ARTICLE1 May 2022High-Performance Ru2P Anodic Catalyst for Alkaline Polymer Electrolyte Fuel Cells Yuanmeng Zhao†, Fulin Yang†, Wei Zhang†, Qihao Li†, Xuewei Wang, Lixin Su, Xuemei Hu, Yan Wang, Zizhun Wang, Lin Zhuang, Shengli Chen and Wei Luo Yuanmeng Zhao† Hubei Electrochemical Power Sources Key Laboratory, College of Chemistry and Molecular Sciences, Wuhan University, Wuhan, Hubei 430072 †Y. Zhao, F. Yang, W. Zhang, and Q. Li contributed equally to this work.Google Scholar More articles by this author , Fulin Yang† Hubei Electrochemical Power Sources Key Laboratory, College of Chemistry and Molecular Sciences, Wuhan University, Wuhan, Hubei 430072 †Y. Zhao, F. Yang, W. Zhang, and Q. Li contributed equally to this work.Google Scholar More articles by this author , Wei Zhang† Key Laboratory of Automobile Materials MOE, School of Materials Science & Engineering, Electron Microscopy Center, and International Center of Future Science, Jilin University, Changchun 130012 †Y. Zhao, F. Yang, W. Zhang, and Q. Li contributed equally to this work.Google Scholar More articles by this author , Qihao Li† Hubei Electrochemical Power Sources Key Laboratory, College of Chemistry and Molecular Sciences, Wuhan University, Wuhan, Hubei 430072 †Y. Zhao, F. Yang, W. Zhang, and Q. Li contributed equally to this work.Google Scholar More articles by this author , Xuewei Wang Hubei Electrochemical Power Sources Key Laboratory, College of Chemistry and Molecular Sciences, Wuhan University, Wuhan, Hubei 430072 Google Scholar More articles by this author , Lixin Su Hubei Electrochemical Power Sources Key Laboratory, College of Chemistry and Molecular Sciences, Wuhan University, Wuhan, Hubei 430072 Google Scholar More articles by this author , Xuemei Hu Hubei Electrochemical Power Sources Key Laboratory, College of Chemistry and Molecular Sciences, Wuhan University, Wuhan, Hubei 430072 Google Scholar More articles by this author , Yan Wang Key Laboratory of Automobile Materials MOE, School of Materials Science & Engineering, Electron Microscopy Center, and International Center of Future Science, Jilin University, Changchun 130012 Google Scholar More articles by this author , Zizhun Wang Key Laboratory of Automobile Materials MOE, School of Materials Science & Engineering, Electron Microscopy Center, and International Center of Future Science, Jilin University, Changchun 130012 Google Scholar More articles by this author , Lin Zhuang Hubei Electrochemical Power Sources Key Laboratory, College of Chemistry and Molecular Sciences, Wuhan University, Wuhan, Hubei 430072 Google Scholar More articles by this author , Shengli Chen Hubei Electrochemical Power Sources Key Laboratory, College of Chemistry and Molecular Sciences, Wuhan University, Wuhan, Hubei 430072 Google Scholar More articles by this author and Wei Luo *Corresponding author: E-mail Address: [email protected] Hubei Electrochemical Power Sources Key Laboratory, College of Chemistry and Molecular Sciences, Wuhan University, Wuhan, Hubei 430072 Suzhou Institute of Wuhan University, Suzhou, Jiangsu 215123 Google Scholar More articles by this author https://doi.org/10.31635/ccschem.021.202100810 SectionsSupplemental MaterialAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail Exploring efficient and economical electrocatalysts and understanding the mechanism for alkaline hydrogen oxidation reaction (HOR) are crucial to facilitate the development of alkaline polymer electrolyte fuel cells (APEFCs). Herein, Ru2P was synthesized and used as an anodic HOR electrocatalyst for APEFC, achieving a peak power density of 1.3 W cm−2, the highest value among Pt-free anode electrocatalysts reported under the same conditions. From the density functional theory (DFT) calculations and experimental results, it was found that besides the optimized hydrogen binding energy, the enhanced adsorption strength of oxygenated species (OH*) and the reduced work function of Ru2P contributed to the enhanced HOR performance. The normalized exchange current densities of Ru2P/C were 0.37 mA cmECSA−2 and 0.27 mA μgRu−1, respectively, both approximately three times higher than those of Ru when conducted in the rotating disk electrode (RDE) system. Our work provides a new pathway for exploring highly active Pt-free HOR electrocatalysts and expanding the family of anodic electrocatalysts for APEFCs. Download figure Download PowerPoint Introduction Proton-exchange membrane fuel cells (PEMFCs) are regarded as the most significant renewable energy conversion technology for efficiently accomplishing transformation from hydrogen to electrical energy, which involves anodic the hydrogen oxidation reaction (HOR), as well as the cathodic oxygen reduction reactions (ORRs).1–3 Currently, Pt remains the most advanced ORR electrocatalyst for PEMFCs, but its use is limited by its high cost and scarcity.4,5 Due to their harsh acidic environment, most nonnoble metals and their alloys are labile and hence lead to low durability in PEMFCs.6 As an alternative to PEMFCs, alkaline polymer electrolyte fuel cells (APEFCs) have received considerable attention by virtue of the development of nonnoble-metal-based electrocatalysts with Pt-like stability and high activity in cathode for ORR in alkaline electrolytes.4,7,8 However, the kinetics of anodic HOR in alkaline solution, even for the most advanced Pt-based catalysts, are approximately two to three orders of magnitude more sluggish than in acidic solution,9,10 which means that more Pt-based catalysts are required to overcome the sluggish anodic HOR in APEFCs. Consequently, developing Pt-free materials as highly active and stable anodic catalysts for APEFCs is highly desirable. Currently, the mechanism for alkaline HOR to guide the catalyst design is still in debate. The hydrogen binding energy (HBE) theory is widely accepted as a valid descriptor to assess the HOR performance by optimizing the strength of adsorbed H (H*) on the surface of electrocatalysts.11–14 However, the most controversial issue is whether HBE is the sole descriptor in alkaline conditions. In fact, Markovic et al.15–18 proposed the bifunctional theory that the strength of both the adsorbed H and adsorbed OH (OH*, a hydroxyl species adsorbed on the active site), are decisive. Furthermore, it has recently been highlighted that the HOR performance of Pt in alkaline media is not only dependent on these thermodynamic characteristics, but also relies on the kinetic factors at electrode/electrolyte interface, for example, the potential of the zero (free) charge (PZC).19,20 Despite the fact that some Ni-based nonprecious metal electrocatalysts have been reported for catalyzing alkaline HOR, their catalytic performances are still far below the practical application in APEFCs.4,6–8,21,22 Most of the APEFCs with Ni-based materials as the anodes only show the peak power density around 0.1 W cm−2.22–24 As the cheapest metal in the platinum group of metals (PGM), Ru-based electrocatalysts have received considerable attention. Ohyama et al.25 reported a Ru/C catalyst applied as the anodic catalyst for APEFC. The peak power density of 0.23 W cm−2 at 323 K was obtained, which is superior to that of their home-made Pt/C (0.20 W cm−2), but still far inferior to that of most commercial Pt/C (0.40–0.82 W cm−2).26–30 According to the electrochemical tests conducted in a rotating disk electrode (RDE) system, the exchange current densities (one of the most vital parameters to assess a catalyst’s intrinsic catalytic activity) of Ru-based catalysts were much lower than that of Pt.31–33 Recently, ruthenium phosphides, such as Ru2P, have been reported as effective electrocatalysts for the hydrogen evolution reaction (HER) due to the optimized HBE induced by the electron transfer between Ru and P atoms.34–36 Some precious transition-metal-based phosphides (TMPs), like palladium phosphides (Pd5P2 and PdP2) and iridium diphosphides (IrP2), have been reported as catalysts for HOR in acidic media.37,38 However, to the best of our knowledge, TMP-based catalysts toward alkaline HOR, especially tested in APEFCs, have so far rarely been reported. Furthermore, in view of the same intermediate during the reaction process (the adsorbed hydrogen, H*) for hydrogen electrode reaction (HER and HOR), we thus expect that the HER-active Ru2P may be a potential candidate for catalyzing alkaline HOR efficiently. In this work, Ru2P nanoparticles loaded on carbon supports (Ru2P/C) have been prepared. When using Ru2P/C as the anodic electrocatalyst for APEFC, a peak power density of 1.3 W cm−2 was obtained, which is the highest record among Pt-free anodic electrocatalysts under the same condition, and even higher than commercial Pt/C in the small-polarization region. Furthermore, the electrochemical tests under RDE system revealed that Ru2P/C exhibits Pt-like HOR performance under alkaline media with ECSA and mass normalized exchange current densities of 0.37 mA cmECSA−2 and 0.27 mA μgRu−1, respectively, both approximately three times higher than those of Ru/C. Experimental Methods Chemicals and materials Ruthenium acetylacetonate (Ru(acac)3), rhodium acetylacetonate (Rh(acac)3), palladium acetylacetonate (Pd(acac)2), and ruthenium chloride hydrate (RuCl3·nH2O) were obtained from Wuhan Changcheng Chemical Co. Ltd (Wuhan, China). tri-n-octylphosphine (TOP), oleylamine (OAm), oleic acid (OA), 1-octadecene (ODE), dodecyl amine (DDA), and tri-n-octylphosphine oxide (TOPO) were obtained from Aladdin Chemical Ltd. (Shanghai, China) Nickel aceylacetonate (Ni(acac)2), cobalt acetylacetonate (Co(acac)2), potassium hydroxide (KOH), perchloric acid (HClO4), isopropyl alcohol, hexane, and ethanol were bought from Sinopharm Chemical Reagent Co. Ltd. (Shanghai, China). Commercial Pt/C (20 wt %) as well as Nafion (5 wt %) were acquired from Sigma-Aldrich (United States). These chemicals were utilized directly without other treatment. Ultrapure water was used during our whole experimental process. Synthesis of Ru2P nanoparticles 0.1 mmol Ru(acac)3 and 1 g TOPO were added into a 50 mL three-neck flask. This system was vigorously magnetically stirred and heated to 120 °C and maintained for 30 min under vacuum to remove residual low-boiling solvents and oxygen. After degassing, the system was heated up to 320 °C, and 0.5 mL TOP was injected. The mixture solution was held at 320 °C for 2 h. Then the reaction was slowly cooled to room temperature (r.t.). The resultant Ru2P nanoparticles were cleaned by adding ethanol to the reaction mixture, followed by centrifugation at 9000 rpm for 5 min, and thereafter washed with ethanol as well as hexane several times to get rid of organic solvent. Eventually, the product obtained was dried at r.t. under vacuum. Synthesis of Ru nanoparticles The synthesis of Ru nanoparticles was similar to what was reported in the previous literature with slight modification.39 Typically, 0.1037 g RuCl3·nH2O, 0.5 mL OA, and 3 mL DDA were added into a 50 mL three-neck flask. The system was vigorously magnetically stirred and heated to 120 °C, then maintained for 30 min under vacuum to remove residual low-boiling solvents and oxygen. After degassing, the reaction system was heated to 320 °C and maintained for 1 h. The postprocess was the same as the synthesis of Ru2P nanoparticles. Synthesis of Rh2P nanoparticles 3 mL OAm, 0.08 mmol Rh(acac)3, and 3 mL ODE were added into a 50 mL round-bottom flask. This system was vigorously magnetically stirred and heated to 120 °C and maintained for 30 min under vacuum to remove residual low-boiling solvents and oxygen. After degassing, the system was heated up to 240 °C and maintained for 60 min, and then 0.4 mL TOP was injected. Thereafter, the system was heated up to 320 °C and held at 320 °C for 2 h. Then the reaction was slowly cooled to r.t. The postprocess was the same as the synthesis of Ru2P nanoparticles. Synthesis of Pd3P nanoparticles 0.1 mmol Pd(acac)2 and 5 mL OAm were added into a 50 mL three-neck flask. This system was vigorously magnetically stirred and heated to 120 °C and maintained for 30 min under vacuum for removing residual low-boiling solvents and oxygen. After degassing, the system was heated up to 320 °C, and 0.5 mL TOP was injected. This mixture solution was held at 320 °C for 2 h. Then this reaction was slowly cooled to r.t. The postprocess was the same as the synthesis of Ru2P nanoparticles. Synthesis of Co2P nanoparticles 0.18 g Co(acac)2 and 5 mL OAm were added into a 50 mL three-neck flask. This system was vigorously magnetically stirred and heated to 120 °C and maintained for 30 min under vacuum for removing residual low-boiling solvents and oxygen. After degassing, the reaction system was heated up to 300 °C, and 0.9 mL TOP was injected. This mixture solution was held at 300 °C for 2 h. The postprocess was the same as the synthesis of Ru2P nanoparticles. Synthesis of Ni2P nanoparticles The synthesis method of Ni2P nanoparticles was analogous to Co2P, except that 0.18 g Ni(acac)2 instead of Co(acac)2 was used, with other experimental parameters unchanged. Materials characterization Powder X-ray diffraction (XRD) patterns were acquired using a Bruker D8-Advance X-ray diffractometer (Bruker, Germany). Catalyst morphologies and sizes were obtained using a Tecnai G20 U-Twin (FEI Co., United States) transmission electron microscopy (TEM) equipped with an energy-dispersive X-ray spectroscopy (EDX). High-angle annular dark-field scanning TEM (HAADF-STEM) image and EDX elemental mapping were acquired by using JEM-ARM300F GRAND ARM (JEOL, Japan) and JEM-2100F (JEOL, Japan). X-ray photoelectron spectroscopy (XPS) was carried out with a Thermo Fischer ESCALAB 250 Xi spectrophotometer (Thermo Fisher Scientific, United States). And inductively coupled plasma atomic emission spectroscopy (ICP-AES) was carried out on a Thermo IRIS Intrepid II XSP atomic emission spectrometer (Thermo Elemental, United States). The H2 temperature programmed desorption (H2-TPD) was conducted with a Micromeritics ChemiSorb 2720 (Micromeritics, United States). Catalyst preparation 5 mg Ru2P nanoparticles and 20 mg XC-72 carbon were added into 20 mL hexane and stirred in a N2 flow for 12 h at r.t. The precipitation was obtained by centrifugation. Then the Ru2P/C was dried under vacuum. Once dry, the Ru2P/C was annealed at 600 °C under 5% H2/N2 to remove surfactants and obtain high crystallinity. Finally, the Ru2P/C catalyst was obtained. The Ru/C catalyst was prepared and postprocessed in the same way. Fifty percentage of loaded catalysts were prepared with 30 mg Ru2P or Ru and 30 mg XC-72 carbon, and postprocessed in the same way for use in single-cell tests. Preparation of working electrodes 4 mg of catalyst was added to 2 mL of mixture solvent containing 5% Nafion/isopropyl alcohol (v/v = 1∶99) with ultrasonic treatment to form a homogeneous solution. Glassy carbon electrode (GCE, 5 mm in diameter) was polished with 0.05 μm gamma alumina powders for a neat surface and then washed. Once the GCE was dried, 5 μL catalyst-ink was drop-coated on the surface of the GCE (loading: ∼0.01 mgPGM cm−2). Electrochemical measurements The electrocatalytic properties were assessed in a three-electrode system using a CHI 760E electrochemical workstation (CH Instruments, Inc., Bee Cave, United States) at 30 °C. The GCE coated with catalyst samples, a graphite rod, and Hg/HgO electrode were applied as the working, counter, and reference electrode, respectively. And the electrolyte was 0.1 M KOH. Before testing the HOR, cyclic voltammetry (CV) was recorded in an Ar-saturated solution to obtain the steady voltammetry curves. Then, polarization curves were obtained by applying a rotation disk electrode (RDE) with a rotation speed of 1600 rpm at a scan rate of 10 mV s−1 in H2-saturated electrolyte. Electrochemical impedance spectra (EIS) tests were carried out after each RDE measurement with the alternating-current (AC) impedance spectra from 200 to 0.1 kHz and a voltage perturbation of 10 mV. All the potential data in this paper were obtained iR-free. Membrane-electrode assembly and single-cell test The as-synthesized Ru2P/C or Ru/C was used as the anode catalyst while the Pt/C (Johnson–Matthey) was used as the cathode catalyst. The home-made quaternary ammonia poly(N-methyl-piperidine-co-p-terphenyl) (QAPPT) was used for the ionomer and the membrane in electrodes. More details of QAPPT can be found in previous literature40,41 ( Supporting Information Table S1). The ink was prepared by mixing Ru2P/C, Ru/C, and Pt/C with QAPPT ionomer solution (20 mg mL−1) in n-propanol, which contained 80 wt % of catalyst and 20 wt % ionomer. Then the ink was ultrasonic for 40 min and then sprayed on the QAPPT APEs (25 ± 3 μm in dry state) to form a catalyst-coated membrane (CCM) with an electrode area of 4 cm−2. The metal loading was controlled to be 0.4 mg cm−2 for anode and 0.4 mg cm−2 for the cathode. Next, the prepared CCM was soaked in 2 M KOH at 80 °C for 24 h for the purpose of changing the anion into OH−, and then washed with distilled water to remove the excess KOH. To make the membrane electrode assembly (MEA) in situ, the CCM was placed between two pieces of carbon paper (AvCarb GDS3250), and the hot-pressing was not used. H2/O2 single-cell APEFCs were tested by an 850e Multi-Range Fuel Cell Test Station (Scribner Associates, Inc., Southern Pines, NC) in a galvanic mode at 80 °C. H2 and O2 were humidified at 80 °C (100% RH) and fed with a flow rate of 1000 sccm with a backpressure of 0.1 MPa symmetrically on both sides. The fuel cell was activated at a constant current for a moment, and then the cell voltage at every current density was recorded. Computational details and theoretical models All the DFT calculations applied in this paper were performed by the PWscf code of the Quantum ESPRESSO package. And the generalized gradient approximation (GGA) method with the Perdew–Burke–Ernzerh (PBE) functional was applied to illustrate the exchange and correlation interactions.42 The interaction between valence electrons and ionic cores was described by Ultrasoft pseudopotential.43 The kinetic energy cutoff was 33 Ry while the charge density cutoff was 330 Ry. The Brillouin zone was sampled using the Monkhorst-Pack method, and the number of k points was set according to this model, of which 2 × 4 × 1 was set for Ru2P geometric optimization and 6 × 6 × 1 for Ru. The convergence threshold was set as 1.0e−6 Ry. In this study, we constructed correlative theoretical models to simulate Ru2P, as well as Ru and Pt for comparison. According to the TEM image (see below), the Ru2P (112) facet was exposed. And the (112) facet showed the highest intensity among all the facets as indicated by the XRD results of Ru2P. Thus, we chose (112) surface as the model of Ru2P for calculation, and Pt (111) and Ru (001) for comparison. For all structure models, four layers were used for calculations, and the top two layers and adsorbate were fully relaxed, with a vacuum gap of ∼15 Å. A slab of (2 × 1 × 1) was used for Ru2P, and slabs of (3 × 3 × 1) were used for Pt and Ru. The calculation of adsorption free energies The adsorption free energy of H (ΔGH*) is usually considered as a descriptor to assess the HER/HOR performance (a catalyst with ΔGH* ≈ 0 could be an excellent candidate for HER).44 The different adsorption free energies of H were determined by the following formula ΔGH* = ΔEH* + 0.24 eV, where ΔEH* represents the binding energy of H* adsorption. The calculation of OH binding energies ΔEOH* = E(OH*) − E(*) − (E(H2O) − 1/2E(H2)),45 for which E(OH*) and E(*) are the ground-state energies of the surfaces with OH* and the clean surface, respectively. E(H2O) and E(H2) are the calculated DFT energies of H2O and H2 molecules in the gas phase. Results and Discussion Amorphorous Ru2P nanoparticles (NPs) were initially prepared by a colloidal method, in which ruthenium acetylacetonate [Ru(acac)3], tri-n-octylphosphine oxide (TOPO), and tri-n-octylphosphine (TOP) were used as Ru precursor, solvent, and P precursor, respectively. The obtained NPs were supported on a certain amount of XC-72 carbon and annealed at 600 °C under 5% H2/N2 to remove the residual surfactants and obtain high crystallinity (denoted as Ru2P/C). X-ray powder diffraction (XRD) was carried out to investigate the crystal structure of the product. As expected, distinct characteristic diffraction peaks corresponding to Ru2P (PDF#89-3031) were observed, which further confirmed the successful formation of Ru2P/C (Figure 1a). Ru nanoparticles were prepared through a similar approach reported previously with some modifications39 and following the same high-temperature annealing process (denoted as Ru/C). Characteristic XRD diffraction peaks of Ru (PDF#88-1734) further confirmed the successful synthesis of Ru/C (Figure 1a). Figure 1 | (a) XRD patterns of Ru2P/C and Ru/C. (b) Cell voltage and power density versus current density curves obtained from APEFC with Ru2P/C (red), Ru/C (blue), and Pt/C (black) as the anode catalysts, respectively. Pt/C was used as the cathode catalyst. All of the metal or metal phosphide loadings in the anodes and the cathodes are controlled to be 0.4 mg cm−2. Single-cells test was conducted at 80 °C. (c) The expanded region of the cell voltage versus current density curves for Ru2P/C and Pt/C from (b). Download figure Download PowerPoint The APEFC performances of Ru2P/C, Ru/C, and commercial Pt/C were first conducted. Figure 1b shows the corresponding plot of cell voltage and power density versus current density. The exact loadings of Ru in Ru2P and Ru were confirmed by ICP-AES with the contents of 43.76 and 48.12 wt %, respectively ( Supporting Information Table S2). When tested with an anode metal (or metal phosphide) loading of 0.4 mg cm−2, Ru2P/C achieved a peak power density of 1.3 W cm−2 (at the current density of 3.0 A cm−2) at 80 °C with 0.1 MPa backpressure, which is the highest value among the reported Pt-free anode catalysts under this condition ( Supporting Information Table S3). For comparison, Ru/C only achieved a peak power density of 0.7 W cm−2 (at the current density of 1.4 A cm−2). Even though the peak power density of Pt/C (1.4 W cm−2, at the current density of 3.2 A cm−2) was a bit higher than Ru2P/C, the current density of Ru2P/C was larger than Pt/C till the cell voltage decrease to 0.87 V, as shown in Figure 1c, suggesting the higher apparent activity of Ru2P/C than Pt/C in the small-polarization region. Due to the same cathodic catalysts, we could confirm that Ru2P/C is a better HOR catalyst, superior to Ru/C, and even comparable with the commercial Pt/C for APEFC. The durability of the Ru2P-catalyzed APEFC single cell was tested in H2/air at 80 °C with 0.2 MPa backpressure. The cell voltage could be maintained at above 0.7 V for 100 h, indicating the stability of the Ru2P-catalyzed APEFC single cell ( Supporting Information Figure S1). To evaluate the alkaline HOR catalytic performance in detail, RDE measurements were conducted in 0.1 M KOH, and the catalyst film-modified glass carbon electrode (GCE) was invoked as the working electrode with a total loading of ∼50 μgcat. cm−2. The exact loadings of Ru in Ru2P and Ru were acquired by ICP-AES with the content of 19.11 and 20.26 wt %, respectively ( Supporting Information Table S4). Commercial Pt/C (20 wt %) was also studied for comparison ( Supporting Information Figure S2). CVs were obtained in an Ar-saturated electrolyte (Figure 2a). The CV peaks of Ru2P/C and Ru/C at about 0.1 V stemmed from the stripping process of the underpotentially deposited hydrogen (H-UPD), accompanied by the oxidation of the Ru surface.33,46–48 And the polarization curves of the three catalysts with iR-compensation tested in the H2-saturated electrolyte are shown in Figure 2b. Similar phenomena can be observed that show that the anodic currents for Ru-based catalysts rise initially from the potential at 0 to around 0.1 V and decline afterward. The decreased currents were illustrated by the excessive oxidation of Ru-species that led to the inhibited behaviors of H adsorption and H2 oxidation.33,46 The HOR/HER current density of Ru2P/C increased much faster than that of Pt/C and Ru/C at around 0 V, suggesting the highest apparent HOR activity of Ru2P/C near the equilibrium potential. It should be noted that due to the surface oxidation, the HOR current density of Ru2P/C at the potential above 0.1 V was inferior to that of Pt/C, which was in accordance with the single-cell test (Figure 1c). Figure 2 | HOR performances of Ru2P/C, Ru/C, and Pt/C. (a) CVs curves of Ru2P/C, Ru/C, and commercial Pt/C. (b) HOR polarization curves and (c) linear-current potential region around the equilibrium potential of HOR/HER of Ru2P/C, Ru/C, and commercial Pt/C after iR-compensation. (d) MA and SA of Ru-based catalysts by rotating disk electrode in 0.1 M KOH. The data of different colors come from different references (dark gray, orange, violet, green, magenta, blue, and red represent refs 33, 49–53 and this work, respectively). Download figure Download PowerPoint Exchange current density (j0) was assessed by the linear-current potential region (−5 to 5 mV) (Figure 2c), using the approximate Butler–Volmer equation, j = j0(ηF/RT),31,33 where j, η, F, R, and T represent the current density, overpotential, Faraday’s constant (96,485 C mol−1), universal gas constant (8.314 J mol−1 K−1), and temperature on the Kelvin scale, respectively. Obviously, the apparent j0 of Ru2P/C was higher than that of Ru/C and Pt/C. In addition, to eliminate interruptions from the currents resulting from the oxidation of Ru or P, the steady-state polarization curve was acquired by the chronoamperometry test ( Supporting Information Figure S3). In addition, similar polarization behaviors between the steady-state polarization curve obtained from the chronoamperometry test and the transient polarization curve acquired from linear scanning voltammetry (LSV) at the micropolarization regions were observed, suggesting that the employment of a transient polarization curve to calculate the j0 does not lead to overestimation. After normalizing by the weight of the PGM ( Supporting Information Table S5), the obtained mass-specific exchange current density (j0,m) of Ru2P/C (0.27 mA μg−1) was a little higher than that of commercial Pt/C (0.21 mA μg−1), and about three times higher than that of Ru/C (0.10 mA μg−1). To further clarify the intrinsic activity of the identified active sites, we tried to estimate the electrochemically active surface areas (ECSAs) of the three samples by the Cu underpotential deposition (Cu-UPD) stripping method ( Supporting Information Figure S4 and Table S5).33 The ECSA-normalized exchange current density (j0,s, specific activity) of Ru2P/C was measured to be 0.37 mA cm−2, which is about three times higher than that of Ru/C (0.12 mA cm−2) and is comparable with that of commercial Pt/C (0.41 mA cm−2). Figure 2d and Supporting Information Tables S5 and S6 show the mass activity (MA) as well as the specific activity (SA) of our samples and other Ru-based catalysts reported previously.33,49–53 It is clear that Ru2P exhibits remarkable HOR performance among the reported Ru-based catalysts, close to the benchmarked Pt-Ru system. The long-term durability of Ru2P/C was probed via accelerated degradation tests (ADTs) ( Supporting Information Figure S5). The CV curves ( Supporting Information Figure S5a) and HOR polarization curves ( Supporting Information Figure S5b) of Ru2P/C were conducted before and after 1000 CVs between 0 and 0.72 V versus RHE. Then the exchange current densities were evaluated (inset of Supporting Information Figure S5b) by means of the approximate Butler–Volmer equation. And after the ADTs, we can see that the apparent exchange current density descended a bit after 1000 cycles but was still higher than the fresh Ru/C, indicating its good HOR stability. Figure 3 | (a) HRTEM image of a single Ru2P nanoparticle of Ru2P/C. Inset shows the SAED pattern of Ru2P/C. (b) HAADF-STEM image and EDX elemental mapping of Ru and P in Ru2P/C. Ru 3p (c) and P 2p (d) core-level XPS spectra of Ru2P/C and Ru/C. Download figure Download PowerPoint The size effect of Ru2P nanoparticles for HOR performances was also examined. Typically, Ru2P/C with average sizes at about 3.3, 4.5, and 6.6 nm were obtained by annealing at 500, 600, and 700 °C, respectively ( Supporting Information Figures S6–S8). The XRD patterns sho