Synthesis and geometric structureInspired by the finding that nitrogen-rich carbon defect trapping Cu(NH3)x15 can be used to prepare a single Cu atom catalyst, we used a vacuum-assisted solid-phase synthetic approach to decorate Pd-Ru@Pt surfaces. As shown in Fig. 2a, we put carbon-supported Pt nanoparticles (Pt/C) on a silicon wafer, and loaded Ruthenium(III) acetylacetonate (Ru(acac)3) and Palladium(II) acetylacetonate (Pd(acac)2) in a combustion boat, and then transferred them into a vacuumed glass tube. The Ru(acac)3 and Pd(acac)2 were chosen as Pd and Ru precursors because of their low evaporation temperatures. The glass tube was heated to 300 °C for 3.0 h. The vaporized Pd and Ru precursors were trapped by the Pt nanoparticles’ surface defects and hollow sites at high temperatures, generating the isolated Pd and Ru atoms on the Pt surface via a catalytic pyrolysis effect of Pt forming Pd-Ru@Pt/C. As reference catalysts, atomic Pd doped Pt/C (Pd@Pt/C) and atomic Ru doped Pt/C (Ru@Pt/C) were also synthesized using the same method.Fig. 2: Preparation and structure characterization of electrocatalysts.a Scheme for synthesizing Pd-Ru@Pt through a chemical vapor deposition (CVD) setup. Normalized Fourier Transform of the k³.χ(k) Extended X-Ray Absorption Fine Structure (EXAFS) spectra of (b) Pt, (c) Pd and (d) Ru. EXAFS of metallic Pt and PtO2 are shown in (b) as ref.37. EXAFS of metallic Pd and PdO are shown in (c) as ref. 38. EXAFS of metallic Ru and RuO2 are shown in (d) as ref. 39. Source data for (b)–(d) are provided as a Source Data file.The morphology of Pd-Ru@Pt was characterized using transmission electron microscopy (TEM). Supplementary Fig. 2 shows the Pd-Ru@Pt catalysts with an average particle size of 2.9 ± 0.3 nm dispersed uniformly on the carbon support. Since the heat treatment only changed the average particle size of the Pt nanoparticle by 0.1 nm from 2.6 ± 0.3 nm (Supplementary Fig. 3) to 2.7 ± 0.4 nm (Supplementary Fig. 4), the thickness changes after surface decoration corresponds to a single particle coating of half atomic layer or less, considering the standard deviation. For the reference catalysts, the Pd@Pt/C (Supplementary Fig. 5) and Ru@Pt/C (Supplementary Fig. 6) showed good dispersion and crystal structure, with average particle sizes 2.9 ± 0.4.Pd and Ru decoration on Pt for Pd-Ru@Pt/C catalysts was directly evidenced by extended X-ray absorption fine structure (EXAFS). The Pt EXAFS in Pd-Ru@Pt/C (Fig. 2b) displays a main peak at ∼2.638 Å, close to that of the Pt–Pt bond length in bulk Pt (2.641 Å in Pt foil). Pt-Pd and Pt-Ru bonds at 2.148 Å are also observed in the EXAFS. The Pd EXAFS in Pd-Ru@Pt/C (Fig. 2c) showed dominant features of a Pd-O peak at ∼1.473 Å and a Pd-Pt peak at ∼2.730 Å over the Pd-Pd bond at ∼2.466 Å (Fig. 2c), while the Ru EXAFS in Pd-Ru@Pt/C (Fig. 2d) showed dominant features of a Ru-Pt peak at ∼2.729 Å over a Ru-Ru bond at ∼2.382 Å. Therefore, Pd-Ru decoration dominated over Pd cluster or Ru cluster formation for Pd-Ru@Pt/C.Aberration-corrected high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) was used to probe the material’s atomic structure. As Pd and Ru have lower Z-contrast than that of Pt due to the large atomic number difference, the darker atomic columns on the edges (Fig. 3a) corresponds to the Pd-Ru decoration on the outmost surface of a Pt nanoparticle. Additionally, the image intensity line profile across the nanoparticle illustrates the significantly low HAADF contrast of the surface Pd/Ru atoms, as highlighted by black arrows in Fig. 3b. As shown in Fig. 3a, the Pd-Ru atoms on the outermost layer are not continuous. This implies that the Pt core may not be entirely covered by foreign doping atoms. After examining multiple particles from randomly picked sample locations (Supplementary Fig. 7a), this representative feature of low-contrast Pd-Ru single-layer shell surrounding high-contrast Pt core was generally observed, and no distinct multi-layer aggregation of Pd-Ru was found. Furthermore, STEM with simultaneously obtained energy-dispersive X-ray spectroscopy (EDS) provides the elemental mapping of Pd, Ru, and Pt (Fig. 3c, Supplementary Fig. 7b), further confirming the Pd-Ru@Pt core-shell structure. Pd and Ru’s relatively uniform spatial distribution implies that most Pd and Ru atoms are likely in the homogeneous and alternate atomic arrangement, confirming the Pd-Ru decoration derived from EXAFS (Fig. 2c, d).Fig. 3: Structural characterization of the Pd-Ru@Pt.a Aberration-corrected high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) image of Pd-Ru@Pt. The surface atoms in darker contrast are Pd-Ru sites. b Line profile of the image intensity along the arrow direction (indicated in the inset) across the individual nanoparticle shown in (a), showing Pd-Ru atoms located on the outmost surface (marked by black arrows). c HAADF-STEM image and the corresponding EDS mapping of individual Pd-Ru@Pt particles with the same type of elemental distribution. Individual elemental distribution of Pd, Ru, and Pt, and the overlaid mixture of Pd+Pt, Ru+Pt, and Pd+Ru+Pt illustrate the spatial distributions of Pd and Ru on the surface and Pt in the core.Structure and thermodynamic stabilityTo gain an atomic-level understanding of the dopant decoration, the preferred adsorption sites of the Pd and Ru atoms at the top (T), short bridge (SB), and long bridge (LB) of the pristine and defective Pt(110) surface were analyzed from a thermodynamic standpoint using DFT calculations (Supplementary Fig. 8a–f). In Supplementary Fig. 8b, the outermost Pt atoms are labeled as the crest, and the bottom accessible Pt atoms are labeled as the trough region. The defective Pt(110) (herein, rPt(110)) was constructed by removing four Pt atoms from a crest row. Details regarding the construction of the surfaces and all sites tested are provided as SI. It is noteworthy that the models were constructed based on the [110] facet since it represents the highly active and stable surface for HOR in alkaline solutions16,17,18. The rPt(110) surface slab is akin to the reconstruction of well-ordered surfaces that become prominent upon annealing. The analysis was restricted to the adsorption of dopants to the detriment of the absorption process. The ground state energy geometry found under vacuum conditions predicted that the optimized site for both Pd and Ru adsorption on top of the pristine Pt(110) surface is the hollow site in the trough region (Supplementary Fig. 8c-f)), for both the pristine and the rPt(110) models. For the adsorption of Pd on the pristine Pt(110), the optimized configuration at the hollow site is 1.82 eV more thermodynamically stable than the adsorption at the top site (the least preferred configuration). Regarding the Ru adsorption, the optimized configuration is 1.03 eV more stable than Ru adsorbing at a top site. On the other hand, for the rPt(110) model, the Pd and Ru adsorptions at the hollow sites are 1.25 eV and 1.72 eV more stable than the least preferred sites, respectively.To explore whether the Pd or Ru atom prefers to deposit on the Pt(110) surface as a dopant or as a clustered dopant, we calculated the binding energy of a Pd (or Ru) dopant on the Pt (110) surface and the energy required to add a dopant once a Pd (or Ru) dopant was already adsorbed at the hollow site in the trough region. The binding energy (Ebinding) per atom (n) was calculated as the difference between the total energy of the optimized substrate-adsorbate and the sum of the relaxed, clean surface (Eslab) and isolated dopants (ETM).$${E}_{{binding}}=\frac{E{total}-(E{slab}+\varSigma {nE}{TM})}{n}$$
(1)
Upon optimization of the Pd (or Ru) adsorption on the pristine and rPt (110) surfaces (herein, Pd-Pt or Ru-Pt), the average binding energies of Pd-Pt (−3.82 eV) are larger than either of the Pd-Pd (−0.54 eV) and Pd-Ru (−1.07 eV) binding energy. Also, the average binding energy of Ru-Pt (−6.75 eV) is larger than either of the Ru-Ru (−2.80 eV) and Ru-Pd (−1.07 eV) binding energies. Therefore, both transition metals, Pd and Ru, prefer to form bonds with Pt atoms at the Pt(110) surface rather than forming clusters. To further investigate the long-term Pd-Ru@Pt(110) stability, the energetic cost of sequential dopant addition (assuming a Ru or Pd is already adsorbed at the (110) surface) was simulated. For these calculations, the energy of the added dopant was set to its calculated cohesive energy value. The average binding energy to add a Ru atom and form a Ru-Ru on the Pt surface is −0.06 eV. The average binding energy to add a Pd atom and form a Pd-Pd on the Pt surface is −0.19 eV. Meanwhile, the average binding energy to add a Pd atom and form Ru and Pd dopants adsorbed at the Pt (110) surface is −0.77 eV. Inspection of the individual energies of both surfaces shows more negative energy on the pristine Pt(110) surface (−1.42 eV) when compared to the rPt(110) surface (−0.12 eV). In other words, using a Pd-Ru@Pt(110) as a baseline, further stabilizes the decoration system because the atoms might sit in deeper potential wells of the hollow site that inhibits the atoms from clustering.The stability of the adsorbed dopants and the evolution of the Pd-Ru distance on the Pd-Ru@Pt catalyst were investigated using AIMD simulations. The effect of an aqueous solution was considered by constructing a neutral electrode/alkaline electrolyte interface containing thirteen H2O molecules and two K+/OH- ions. The K+ ions were fixed a few angstroms above the surface, while the OH- ions were allowed to evolve with the system. As shown in Fig. 4a–c (top view) and d-f (side view), the Pd and Ru atoms remained adsorbed at a hollow site in the trough region on the outermost surface of the Pd-Ru@Pt (110) over time.Fig. 4: Time evolution and structural dynamics of the Pd-Ru@Pt (110) slab.Snapshots of the time evolution of the Pd-Ru@Pt (110) slab shown at (a–c) for the top view configurations, and side view configurations for (d) 0 fs, (e) 10,000 fs and (f) 20,000 fs. Color code: Gray, blue, yellow, white, red, and purple spheres represent Pt, Pd, Ru, H, O, and K, respectively. The model is represented in a ball and stick style for clarity.Electronic structure and hydrogen adsorption energy as descriptors for the HORAn XPS experiment was used to study the dopant’s electronic effect on Pt (Supplementary Fig. 9 and Supplementary Table 1). The Pt0: Pt2+ ratio of Pd-Ru@Pt/C (83.15: 16.85) from deconvoluted high-resolution Pt XPS spectra was larger than either of Pd@Pt/C (80.34: 19.66) or Ru@Pt/C (80.87: 19.13) Pt0: Pt2+ ratio, which favors Hads activation and subsequent oxidation reaction on Pd-Ru@Pt/C2,19,20. The Pt0 electron binding energy, which is higher in Pd-Ru@Pt (72.05 eV) than in Pt (71.84 eV), enhances the HOR kinetics by weakening hydrogen adsorption on the surface5. In contrast, the Pt0 electron binding energy in Pd@Pt/C negatively shifts to 71.62 eV from 71.84 eV in Pt/C. The Ru-doping on Pt/C does not significantly change the Pt0 electron binding energy, ruling out Ru’s electronic effect on the HOR performance for Ru@Pt/C. Furthermore, after cycling, the high ratio of Pt0: Pt2+ and a Pt0 electron binding energy that is higher for Pd-Ru@Pt/C than for Pt/C are maintained, confirming the stability of as-prepared catalysts. Additionally, Pt, Pd, and Ru’s oxidation states were probed by the white line intensity in the XANES spectra. The Pt L3-edge white line intensity of Pd-Ru@Pt/C is close to that of Pt foil (Fig. 5a), indicating that the average oxidation state of Pt is primarily zero. However, the white line intensity of the Pd K-edge (Fig. 5b) from Pd-Ru@Pt/C features a combination of metallic Pd and PdO. For the Ru K edge (Fig. 5c), the energy absorption edge fell between metallic Ru foil and RuO2, suggesting the co-existence of metallic and oxidized states.Fig. 5: Electronic structure characterization.a Pt L3-edge, (b) Pd K-edge, and (c) Ru K edge X-ray Absorption Near-Edge Structure (XANES) spectra. XANES of metallic Pt and PtO2 are shown in (a) as ref.37. XANES of metallic Pd and PdO are shown in (b) as ref.38. XANES of metallic Ru and RuO2 are shown in (c) as Ref.39. Source data for (a)–(c) are provided as a Source Data file.In addition to XPS and XANES data, the effects of the dopants on local environment of Pt (110) were studied with a theoretical approach. Specifically, the effect of dopants on the catalyst surface was quantified in terms of hydrogen adsorption energy. The hydrogen adsorption energy is defined as the total energy difference between the pristine and doped surface slab and the sum of the slab with hydrogen atoms adsorbed. Different arrangements of H adsorption at the pristine and doped (110) surface were investigated and described in Supplementary Table 2 and Supplementary Table 3. In this respect, our calculations show that the preferred hydrogen adsorption site at the pristine surface is the top site (T) with an −0.56 eV hydrogen binding energy. The next preferred site is the SB site, with a slightly lower value of −0.54 eV. Finally, LB is the least preferred site with a −0.22 eV. In a reconstructed surface, new sites become available for hydrogen adsorption, as shown in Supplementary Fig. 8b. The trough region becomes the preferred adsorption site with a value of −0.60 eV, and the crest sites (top and short-bridge) have an equal binding energy of −0.43 eV. The T site at the edge atoms is the second preferred site with a −0.50 eV. Upon the Pd-Ru dopant addition, our calculations show that the hydrogen binding energies on the Pd-Ru@Pt(110) catalyst are similar to those obtained for the pristine catalyst. The SB sites show slightly reduced hydrogen binding energy values of −0.50 eV and −0.48 eV in the scenarios where the H atoms are adsorbed at SB sites near the Pd and Ru atoms, respectively. Interestingly, in the defective catalyst (Pd-Ru@rPt(110)), the hydrogen binding energies at the SB sites for Pd and Ru atoms are lowered to −0.24 eV and −0.25 eV, respectively. The energy for hydrogen binding at the top site of the edge row is also reduced to −0.41 eV and −0.35 eV for the Pd and Ru atoms, respectively. Finally, the hydrogen binding energy at the SB and T sites of the crest row was slightly reduced to −0.38 eV and −0.40 eV, respectively.Electrochemical performanceThe effects of surface decoration on electrocatalytic behavior in electrolytes with different pH were explored using a rotating disk electrode (RDE) test in both 0.1 M KOH and 0.1 M HClO4. Supplementary Fig. 10a confirms that the decreased HOR activities of Pt in alkaline electrolytes compared to that in acidic electrolytes is due to the increased HBE2,17,21,22 or destabilized OH adsorption16,17, as evidenced by the 145 mV positive shift of the OH and H associated peak in the cyclic voltammogram (CV)2,23. After the Pd-Ru co-doping on the Pt surface, the OH and H associated peak of the Pd-Ru@Pt/C in the alkaline electrolyte shifted negatively to 122 mV vs. RHE from 274 mV vs. RHE for the Pt/C (Fig. 6a), which is even comparable to that of the Pd-Ru@Pt/C in acidic electrolytes (Supplementary Fig. 10b). In contrast, only Pd or Ru doping on a Pt surface does not significantly shift the peak-current potential of Pt in alkaline electrolytes (Supplementary Fig. 10a, c, d). In 0.1 M KOH, the Pd-Ru co-doping on Pt negatively shifted the desorption peak potential indicating the weakened dissociated Hads or stabilized OH adsorption compared with the reference Pt/C (Fig. 6a), which contributed to HOR performance improvement. From a computational standpoint, we conduct a mechanistic study to shed light on the bond breaking and reforming processes.Fig. 6: Electrochemical performance and mechanism of different catalysts.a CV curves of different catalysts. Test condition: 0.05 to 1.00 V vs. RHE, 50 mV s−1, N2-saturated 0.1 M KOH. b The linear fits (lines) in the micro-polarization region for different catalysts at 10 mV s−1 and 1600 rpm in H2-saturated 0.1 M KOH. All measurements were carried out at 298 K, corrected for the mass-transport and ohmic drop. The resistance of the electrolyte solution in the respective setup was determined to be 51.7 ± 0.5 Ω. c Comparison of ECSA (black arrow, left axis), surface area-based (SA, normalized by ECSA, orange arrow, right axis) and mass-based exchange current density (MA, normalized by metal mass, purple arrow, right axis) for HOR in H2-saturated 0.1 M KOH, obtained by linear fitting in the micro polarization region for all tested catalysts. Error bars represent the standard deviation of three independent tests for each catalyst. d Comparison of limit current density of Pt/C and Pd-Ru@Pt/C under NH3 pump mode in electrochemical cell. Source data for Fig. 6a-d are provided as a Source Data file.The exchange current densities are obtained by linearly fitting the kinetic current versus the mass-transport and ohmic resistance-corrected24,25 potential between -5 and 5 mV vs. RHE (Method part, Fig. 6b). In addition, the ECSAs of catalysts were obtained by measuring the under-potential copper deposition (Cuupd)20 (Supplementary Fig. 11). The similar ECSAs after Pd and Ru decoration on Pt/C (Fig. 6c) prove that the distribution of Pd and Ru dopants does not block the active surface. Furthermore, from Fig. 6c, Supplementary Table 4 and Supplementary Table 5, the surface area and mass-based exchange current density of Pt/C were determined to be 0.42 ± 0.03 mA cm−2Pt and 201 ± 17 A g−1metal, respectively, in good agreement with those reported by other groups17,25,26,27,28. The surface area-specific exchange current density (3.28 ± 0.23 mA cm−2Pt) of Pd-Ru@Pt/C is 7.8 times that of the Pt/C (0.42 ± 0.03 mA cm−2Pt), 3.2 times that of Pd@Pt/C (1.03 ± 0.06 mA cm−2Pt) and 1.9 times that of Ru@Pt/C (1.72 ± 0.13 mA cm−2Pt). Normalized to metal mass, the mass-based exchange current density (1557 ± 85 A g−1metal) of Pd-Ru@Pt/C is 7.7 times that of the Pt/C (201 ± 17 A g−1metal), 3.2 times that of Pd@Pt/C (493 ± 21 A g−1metal) and 1.9 times that of Ru@Pt/C (829 ± 78 A g−1metal). Therefore, compared with either Pd or Ru modification, the Pd-Ru pair more efficiently improves the specific and mass exchange current density of Pt. Moreover, the electrochemical durability of the Pd-Ru@Pt/C in an alkaline electrolyte was also assessed by accelerated stability tests between 0.05 and 0.4 V (vs. RHE) at 100 mV s−1 in N2-saturated 0.1 M KOH. The surface area and mass-based exchange current density of the Pd-Ru@Pt/C after 10,000 cycles were still much higher than those for the Pt/C (8.0 and 8.4 times, respectively) (Supplementary Fig. 12, Supplementary Fig. 13).We also used AIMD simulations combined with the slow-growth sampling approach to shed light on the structure and dynamics of the pristine and Pd-Ru@Pt(110) systems, focusing on assessing the free energy profiles of key reactions. Specifically, we investigated the H2 decomposition process, comparing the behavior at pristine and PdRu-doped Pt(110) surfaces. For these simulations, we used the equilibrated models of the pristine Pt(110) and Pd-Ru@Pt(110) systems (this latter model is shown in Fig. 4). Regarding the H2 decomposition process, we note that the H2 decomposition in a standard AIMD simulation is only a few femtoseconds long, in accordance with the ab-initio studies by Ishikawa and co-workers29. In the following, two slow-growth simulations that compare the H2 decomposition process at the pristine and PdRu-doped Pt(110) surface are discussed in detail. Here, Supplementary Fig. 14a–f summarizes the slow-growth trajectory of a simulation using the pristine Pt(110) surface, where an H2 molecule is adsorbed at the top position of the (110) surface. At the pristine Pt(110) surface, our simulations revealed the steps involved in H2 decomposition, showing an initial elongation of the H–H bond followed by movement towards adjacent Pt atoms. This process exhibited an energy barrier of approximately 0.27 eV. The free energy profile of these dynamics is shown in Supplementary Fig. 14f. The initial H–H bond elongation is facile, but then it overcomes a 0.27 eV barrier. The free energy of the final state shows a value of approximately −0.11 eV: thus, resulting in a strengthened Pt-H interaction.Meanwhile, at the Pd-Ru@Pt(110) surface (Fig. 7), H2 decomposition process overcomes an energy barrier of approximately 0.13 eV. This is a 0.14 eV reduction in the energy barrier when compared to the pristine model. In this decomposition process, the moving H atom is quickly located at a bridge-like position between a top Pt atom and the Pd atom. Thus, the presence of the Pd atom modifies the bonding environment of H atoms. This finding suggests that Pd dopants aid in the H2 decomposition.Fig. 7: Mechanism of H2 dissociation and mobility on Pd-Ru@Pt(110) catalyst with free energy landscape.a–e The H2 dissociation and mobility reaction mechanism across the Pd-Ru@Pt(110) catalyst. f The free energy associated with this reaction mechanism. Color code: Gray, white, red, purple, yellow, and blue spheres represent Pt, H, O, K, Ru, and Pd atoms, respectively. The model is represented in a ball and stick style for clarity. Source data for (f) are provided as a Source Data file.Supplementary Fig. 15a–e show the entire H2 decomposition and Supplementary Fig. 15f shows the free energy profile of this reaction. Considering the weakened adsorption of hydrogen atoms on the Pd-Ru@rPt(110) catalyst, we extended our simulations to investigate H2 dissociation at this surface, revealing similarities to the process observed in the Pd-Ru@Pt(110) catalyst. Despite a slightly larger energy barrier of approximately 0.16 eV, the final state energy value was comparable.Due to these similarities, next, we calculated the H2O formation at the Pd-Ru@Pt(110) catalyst only. Moving to the formation of water (H2O), we observed differences in the behavior of OH molecules at the two surfaces. At the pristine Pt(110) catalyst surface, OH was adsorbed in Supplementary Fig. 16a, with subsequent water formation exhibiting a high energy barrier compared to the Pd-Ru@Pt(110) surface (0.91 eV versus 0.48 eV shown in Supplementary Fig. 16). We note that the OH is adsorbed at the catalyst’s surface (Supplementary Fig. 16a). However, Supplementary Fig. 16b–d shows the OH molecule slightly away from the surface. Thus, this Pt-O bond is dynamic, and the formed water molecule can stay adsorbed at the surface, as shown in Supplementary Fig. 16e, f. However, due to the absence of a potential in the calculation, we will not discuss the rate-limiting steps in the HOR. Instead, based on the relative reduction in the energy barrier observed at the Pd-Ru@Pt(110) surface we speculate that the dopants may form catalytic centers able to enhance the adsorption of OH molecules thus facilitating the HOR. It is noteworthy that our simulations at the Pt(110) surface reveal at some instances the interactions between an adsorbed OH and a water molecule in the interface region. This dynamic interaction leads to a proton from water bonding to the adsorbed OH forming adsorbed water and an OH molecule in the interface region. Thus, we speculate that the adsorbed H atom may compete with this exchange for an available OH molecule, making the water formation more difficult at this catalyst surface.Meanwhile, at the Pd-Ru@Pt(110) catalyst, one feature that differentiates this catalytic surface from the pristine one is that the Ru atom attracts OH molecules. As shown in Supplementary Fig. 17, the adsorbed OH molecule does not need to interact with Pt atoms to form a water molecule. However, it is important to point out that although our simulations did not reveal proton exchange with adsorbed OH molecules on the Pd-Ru@Pt(110) catalyst, the possibility of water molecule adsorption remains. Furthermore, regardless of the absence of an applied potential in our calculation, the observed differences in energy barriers for key reactions between the pristine and doped surfaces suggest potential enhancements in catalytic activity due to the presence of dopants.Regarding the water adsorption strength, its increase on the Pd-Ru@Pt surface in alkaline electrolytes was validated by Fourier-transform infrared spectroscopy (FTIR). The relative absorbance of the OH stretching bands can be used to estimate the overall water organization changes. Water adsorption changes OH stretching modes in the region 3000 − 3500 cm−1, highly sensitive to water’s hydrogen bond (HB) network. The broad band consists of two main components peaking at 3250 (ν1) and 3400 cm−1 (ν2), which relate to water molecules in tetrahedral-bonded ice-like organization and an HB-distorted liquid environment, respectively30. The increase of the bands absorbance at 3250 cm−1 demonstrates a higher coordination of water molecules. As shown in Supplementary Fig. 18, the ratio R of ν1: ν2 increases in the order: R(0.1 M KOH) < R(Pt/C 0.1 M KOH) < R(Pd-Ru@Pt/C 0.1 M KOH), indicating the increased coordination and water-substance interaction strength.It is important to note, in light of the fundamental insights provided by Koper et al.14, which underscore the enthalpic barriers on polycrystalline Pt, our work extends these findings by employing AIMD simulations to elucidate the synergistic effects of Pd-Ru pair modifications on Pt within an aqueous environment. This approach not only complements the microsolvation method but also provides a comparative baseline analysis of the HOR process, thereby enhancing our understanding of the dynamic interactions at play and contributing to a broader perspective on catalytic performance in alkaline media. Our results, which highlight the relative improvements in catalytic activity in the presence of Pd and Ru, serve to reinforce the potential of such modified catalysts for practical applications in fuel cell technologies.In a proof-of-principle demonstration of practical application, we compared the bifunctional hydrogen electrocatalytic activity of Pt/C and Pd-Ru@Pt/C using a 5 cm2 NH3 pump cell (Supplementary Fig. 19). The Pd-Ru@Pt/C showed a 1.4 times higher steady current value than the Pt/C in the NH3 pump mode in alkaline environments (Supplementary Figs. 20 and 6d), which is consistent with the advantage of optimized surface micro-environments via Pd-Ru doping in the presence of OH-.In this work, Pd-Ru decoration on Pt nanoparticles (Pd-Ru@Pt) was synthesized by trapping the high-temperature-evaporated Pd and Ru atoms from precursors onto the Pt nanoparticle surface. The stable Pd-Ru@Pt structure is verified by both EXAFS and DFT simulations. The surface area and mass specific exchange current density of Pd-Ru pair were 7.8 and 7.7 times those of the Pt/C. DFT and AIMD simulations indicate potential dopant stability and preferred adsorption mode in the presence of a liquid environment. The AIMD simulations, coupled with slow-growth sampling, offered insights into the structural and dynamic differences between pristine and Pd-Ru@Pt(110) systems, especially concerning key reactions like H2 decomposition and subsequent H2O formation. The decomposition of H2 on pristine and Pd-Ru doped Pt(110) surfaces shows different energy barriers, with the pristine system encountering a barrier of approximately 0.27 eV and the doped system a reduced barrier of about 0.13 eV. Similarly, the energy barrier for water formation varies markedly between the two surfaces. On the pristine Pt(110) catalyst, the process exhibits a high barrier of 0.91 eV, which is substantially lowered to 0.48 eV on the Pd-Ru@Pt(110) catalyst. The simulations revealed variations in energy barriers between the two surfaces, which may indicate potential catalytic benefits of Pd and Ru dopants in facilitating these reactions. While we acknowledge that the presence of an applied potential in the simulation would provide additional input into the electron transfer steps, our findings highlight the importance of dopants in modifying the bonding environment and potentially enhancing catalytic activity. Further investigations may shed more light on these mechanisms and their implications for practical applications in catalysis. Practically, the enhanced HOR and HER performances of the Pd-Ru@Pt/C is validated by NH3 electrochemical pump performance. The demonstrated electrochemical device can easily be scaled up on demand. With renewable energy as the power input, our method represents an efficient, distributed, and sustainable NH3 separation and/or compression with broad scientific and technical impacts.
Pd-Ru pair on Pt surface for promoting hydrogen oxidation and evolution in alkaline media
