Tailoring the Electronic Structure of Ir Alloy Electrocatalysts through Lanthanide (La, Ce, Pr, and Nd) for Acidic Oxygen Evolution Enhancement

The oxygen evolution reaction (OER) is critical for renewable energy conversion and storage devices. However, the rational design of electrocatalysts with suitably high efficiency and stability in strongly acidic electrolytes remains a major challenge. Herein, a solid‐phase synthesis strategy is developed for the preparation of Ir‐Ln (Ln = La, Ce, Pr, Nd) alloy nanoparticles with uniform particle size on carbon supports as superior acidic OER catalysts. Tailoring by the rare earth (RE) elements, Ir2Pr achieves a maximum mass activity of 2.10 A mg−1Ir at 300 mV overpotential and stability over 200 h at 10 mA cm−2 in 0.5 m H2SO4, which is 9.5 and 20 times higher to pure Ir nanoparticles. Furthermore, Ir2Pr alloy nanoparticles exhibit excellent durability in strongly acidic electrolytes. Theoretical calculations have confirmed that the OER performances are strongly related to the RE elements in the alloy, where the d‐band centers show a consistent trend with the overpotential. Moreover, the highest electroactivity of Ir2Pr is attributed to the improved electron transfer by 4f orbitals and the suitable binding strength of intermediates. Herein, a fundamental understanding of the lanthanide–electrochemical performance relationship is provided and will also inspire the rational design of efficient nanoscale RE alloy electrocatalysts.

The oxygen evolution reaction (OER) is critical for renewable energy conversion and storage devices. However, the rational design of electrocatalysts with suitably high efficiency and stability in strongly acidic electrolytes remains a major challenge. Herein, a solid-phase synthesis strategy is developed for the preparation of Ir-Ln (Ln = La, Ce, Pr, Nd) alloy nanoparticles with uniform particle size on carbon supports as superior acidic OER catalysts. Tailoring by the rare earth (RE) elements, Ir 2 Pr achieves a maximum mass activity of 2.10 A mg À1 Ir at 300 mV overpotential and stability over 200 h at 10 mA cm À2 in 0.5 M H 2 SO 4 , which is 9.5 and 20 times higher to pure Ir nanoparticles. Furthermore, Ir 2 Pr alloy nanoparticles exhibit excellent durability in strongly acidic electrolytes. Theoretical calculations have confirmed that the OER performances are strongly related to the RE elements in the alloy, where the d-band centers show a consistent trend with the overpotential. Moreover, the highest electroactivity of Ir 2 Pr is attributed to the improved electron transfer by 4f orbitals and the suitable binding strength of intermediates. Herein, a fundamental understanding of the lanthanide-electrochemical performance relationship is provided and will also inspire the rational design of efficient nanoscale RE alloy electrocatalysts.
partially occupied and the displacement of the d-band center, optimizing its bonding strength for adsorption. [10,14] Despite the enhanced intrinsic activity of each active site in the catalyst, these alloys can suffer from severe dealloying problems, especially in harsh acidic solutions, resulting in diminished long-term stability. In recent years, rare earth (RE) alloys have attracted research interest due to their unusually negative alloy formation energies that are inherently less dealloying and exhibit high stability in acidic media. [15][16][17][18] In addition to the excellent stability, RE elements have lanthanide contraction, and the lanthanide radius decreases with increasing f-shell filling, which allows RE elements to precisely tailor the electronic structure of active metals within a certain range. For example, in the oxygen reduction reaction, lanthanide contraction in Pt-lanthanide alloys controls the lattice spacing of the alloys and tunes the activity, stability, and reactivity of Pt. [15] From an economic point of view, alloying Ir with RE elements can also reduce the use of Ir. For example, the current prices of metals La and Nd are less than 0.001% and 0.14% of Ir metal. [17] However, limited by efficient synthetic methods, only the electrochemical activity of Pt-RE alloys has been investigated in the past decade, and the properties of Ir-RE nanoalloys have largely remained unexplored. [19] Since RE elements are oxophilic, Ir, and RE elements have drastically different standard reduction potentials (e.g., þ1.15 V for Ir 3þ /Ir and À2.38 V for La 3þ /La), which makes synthetic strategies in aqueous or protic solvents difficult to synthesize RE alloys. Therefore, it is highly desirable to develop an effective synthetic strategy for Ir-Ln alloys to achieve efficient and robust OER catalysts while revealing the interaction mechanism between RE and Ir.
Herein, a solid-phase synthesis strategy is developed for the preparation of uniform particle-size Ir-Ln alloys on graphene as superior acidic OER catalysts. Graphene, H 2 IrCl 6 ·6H 2 O, LnCl 3, and cyanamide were uniformly mixed as precursors by grinding, and sodium vapor was used as a reducing agent due to its strong reducing effect. After annealing, washing, and drying, a series of Ir-Ln/G (Ir 2 La/G, Ir 2 Ce/G, Ir 2 Pr/G, and Ir 2 Nd/G) nano-alloy catalysts can be obtained. The catalytic performance of Ir is enhanced by alloying with lanthanides. Benefiting from the lanthanide contraction of La%Nd, the precise customization of the electronic structure of Ir is realized. Most importantly, the overpotential of Ir 2 Pr is only 260 mV at 0.5 M H 2 SO 4 , which exceeds that of Ir/C and other Ir alloy nanocatalysts with different molar ratios. Meanwhile, a representative Ir 2 Pr/G catalyst exhibits a high-quality activity of 2.10 A mg À1 noble metal at 1.53 V, which is 9.5 times that of Ir/G, while the stability reaches 200 h. Benefiting from the well-defined stoichiometric ratio and atomically ordered alloy structure, Ir 2 Ln becomes a reliable model catalyst. Density functional theory (DFT) calculations have demonstrated that the Ir 2 Ln alloys display volcano plots of the OER performances, which are determined by the valley trends of the dband center. The introduced 4f orbitals in Ir 2 Pr can alleviate the barrier of electron transfer from the electroactive Ir sites to the intermediates, leading to the smallest overpotential of OER.

Results and Discussion
The synthesis of nanoscale RE alloys is promising for practical industrial applications, but its synthesis is indeed challenging.
Due to the low reduction potential of RE ions, as shown in Figure S1a, Supporting Information, it is difficult to be simultaneously reduced and alloyed with transition metal ions, which makes the kinetic energy barrier in the synthesis process of RE alloys and late transition metal alloys different. Late transition metal alloys can be synthesized by conventional wet chemical techniques in aqueous or protic solvents, resulting in disordered solid solution alloys under mild conditions. The higher diffusion barrier needs to be overcome only when the synthesis of atomically ordered intermetallic compounds is required ( Figure S1b, Supporting Information). [20] Different from late transition metal alloys, RE elements need to overcome higher energy barriers even to form disordered alloys. This energy barrier is mainly composed of two parts, one part comes from the barrier that RE ions are reduced to metal states, and the other part is attributed to the mutual diffusion barrier of atoms in two metal states. In addition, the electronegativity difference between RE elements and late transition metals is large (e.g., 1.10 for La, 1.9 for Cu, 2.2 for Ir, and 2.2 for Pt), which leads to abnormally negative alloying energies. [21] Therefore, after overcoming the energy barrier of alloying, the RE alloys are thermodynamically more inclined to form intermetallic compounds as shown in Figure S1c, Supporting Information. To overcome the barriers to RE alloying, a solid-phase synthesis strategy was developed for the preparation of Ir-Ln alloys with uniform particle size on graphene ( Figure 1a). A homogeneous mixture of graphene, H 2 IrCl 6 ·6H 2 O, LnCl 3, and cyanamide was used as the precursor, and sodium vapor was employed as the reducing agent due to its strong reducing effect. After annealing, pickling, and drying, a series of Ir-Ln/G nano-alloy catalysts could be obtained.
The products were analyzed by powder X-Ray diffraction (XRD), inductively coupled plasma atomic emission spectroscopy (ICP-AES), high-resolution transmission electron microscopy (HR-TEM), and elemental mapping. Figure S2, Supporting Information shows the XRD patterns of the four alloy samples. The diffraction peaks of the samples are assigned to (1 1 1), (2 2 0), (3 1 1), (2 2 2), (4 0 0), (3 3 1), (5 1 1), (4 4 0), (5 3 3), and (6 2 2) crystal planes. They match well with the Inorganic Crystal Structure Database (ICSD) card for the ordered intermetallic phase, and the four Ir 2 Ln/G alloys have similar unit cell structures, all of the Cubic Fd-3m space group. [22][23][24] In addition, quantitative ICP-AES analysis was performed on the chemical composition of the nanoalloys (see Table S1, Supporting Information for the results of ICP-AES), and the results confirmed the molar ratio of Ir to RE elements in the samples was very close to 2:1. Figure 1b,c shows typical scanning transmission electron microscopy (STEM) and TEM images of the as-prepared Ir 2 Pr/G. The Ir 2 Pr alloy consists of small particles of a few nanometers in size with an average diameter of 7.3 nm and narrow size distribution ( Figure S3, Supporting Information). As shown in Figure 1d, HR-TEM revealed that the lattice distance of the alloy particles was 0.27 nm and 0.44 nm, and the angle between the intersectant lattice fringes was 90°, corresponding to the (220) and (11) plane of the cubic phase Ir 2 . [24] All Ir 2 Pr nanoparticles showed clear lattice fringes, indicating good crystallinity. The atomic model of Ir 2 Pr further confirms the existence of the (11) plane, indicating the ordered structure of the Ir 2 Pr alloy. Meanwhile, the corresponding element map images of Ir 2 Pr/G confirmed the uniform distribution of Ir and Pr elements (Figure 1e-f ). In addition, TEM analysis of the remaining alloys was performed. All synthesized alloys had uniform shapes and sizes ( Figure S4, Supporting Information). HR-TEM images of Ir 2 La/G, Ir 2 Pr/G, and Ir 2 Nd/G samples all show a lattice spacing of 0.23 nm, corresponding to their (311) planes. In particular, the additionally exposed crystal plane is also consistent with the crystal structure.
To determine the surface electronic structure and chemical state of the fabricated alloy samples and the synthesized Ir/G nanocatalysts, X-Ray photoelectron spectroscopy (XPS) analysis was performed, and Ir 4f and La 3d, Ce 3d, Pr 3d XPS spectra were collected ( Figure 2). Figure 2a shows the Ir 4f XPS spectrum, the two sets of double peaks at 61.3 and 64.3 eV, 62.7 and 65.7 eV, respectively, can be assigned to the Ir 0 and Ir 4þ valence states, indicating that there is partial spontaneous oxidation of Ir on the alloy surface. [25] Compared with the Ir 4f signal of Ir/G that indicates the Ir 4þ valence state, the Ir 4f signal of the Ir alloy shifts to lower binding energy, which indicates the electron transfer from La, Ce, Pr, and Nd to Ir in the alloy. More importantly, as the atomic number of Ln increases, the binding energy of Ir exhibits a regular red negative shift, which decreases by À0.3 eV for Ir 2 Nd/G and À0.1 eV for Ir 2 La/G, indicating that lanthanide contraction provides an effective route for Ir electronic structure design (11). In the Pr 3d spectrum of Ir 2 Pr/G, there are spin doublets at 954.7 eV (Pr 3d 3/2 ) and 934.6 eV (Pr 3d 5/2 ), and the ΔE of the main peak is 20.8 eV, indicating that the Pr is mainly in the metal valence state. [26,27] The two weak peaks at 950.7 eV (Pr 3d 3/2 ) and 929.9 eV (Pr 3d 5/2 ) may be caused by the transfer of electrons from Pr to Ir through metallic bonds. As shown in Figure S5, Supporting Information, the ΔE values of La 3d, Ce 3d, and Nd 3d are 16.4, 18.3, and 22.4 eV, respectively, which are consistent with metallic La, Ce, and Nd, indicating that the Ln in the alloy is mainly zero valence state. [28,29] The above results demonstrate the electronic effect of Ir-Ln alloys, electrons from La, Ce, Pr, or Nd are transferred to Ir, and alloying with lanthanides could effectively tailor the electronic structure of Ir.
To further analyze the electronic and local structure in the Ir 2 Pr/G alloy, X-Ray absorption spectra were measured. Figure 2b shows the normalized Ir L 3 edge of the X-Ray nearedge structure absorption (XANES) spectra for Ir foil, Ir 2 Pr/G, and IrO 2 , located around 11215 eV, respectively. The intensity of the white line at the L 3 edge is closely related to the density www.advancedsciencenews.com www.advenergysustres.com of states of the d-band, which is an effective indicator of the valence state of 5d electronic materials. [1] The Ir atom in IrO 2 exhibits the highest valence state and also the highest white line intensity. The white line intensity of Ir 2 Pr/G is weaker than that of IrO 2 , indicating that the average valence state of Ir in Ir 2 Pr/G is lower than þ4 valence, which is in good agreement with the XPS analysis. In addition, the white line intensity of Ir 2 Pr/G is stronger than that of Ir foil, indicating that the alloying of Pr with Ir effectively tunes the d electron configuration of Ir, which helps to optimize the adsorption capacity of active O species during OER. The absorption edge energy of the corresponding sample was extracted using the first derivative of the spectrum. The k 3weighted Fourier transform of the extended X-Ray absorption fine structure (EXAFS) of the corresponding sample is shown in Figure 2c. The k-space EXAFS data are shown in Figure S6, Supporting Information and Table S2, Supporting Information. The presence of the Ir-O peak in the EXAFS curve of Ir 2 Pr/G confirms the partial oxidation of Ir on the alloy surface. The peak at about 2.71 Å is assigned to the Ir-Ir shell, while the peak at about 1.62 Å is assigned to the Ir-O shell. For Ir 2 Pr/G, two additional peaks appear around 3.0 and 3.4 Å, which can be assigned to the Ir-Pr shell and the Ir shell in the Ir 2 Pr alloy. The peak around 1.9 Å can be assigned to the partial oxidation of Ir on the alloy surface. Furthermore, a wavelet transform was performed to visually examine the local coordination of Ir (Figure 2d,e), the maximum at 12.2 Å À1 corresponds to the Ir-Ir coordination shell in the Ir foil, while the values at 9.0 and 13.9 Å À1 correspond to the Ir-Pr coordination shell and Ir-Ir coordination shell in Ir 2 Pr, respectively. These results suggest that the alloying of Pr with Ir modulates the local coordination environment and electronic structure of Ir. A typical three-electrode system was used in 0.5 M H 2 SO 4 to study the effect of lanthanides in RE alloys on the OER properties of Ir, and the synthesized Ir/G was used as a reference. Figure 3a shows the linear sweep voltammetry (LSV) curves of various catalysts at 5 mV s À1 . It can be observed that the overpotentials at Ir-Ln/G current density of 10 mA cm À2 are 318 mV (Ir 2 La/G), 298 mV (Ir 2 Ce/G), 260 mV (Ir 2 Pr/G) and 274 mV (Ir 2 Nd/G), much lower than Ir/G (371 mV), indicating that the presence of alloying Ln and Ir effectively enhances the OER activity in acidic media. Furthermore, as shown in Figure 3b, Ir-Ln/G exhibited lower Tafel slopes of 68.9%82.1 mV dec À1 in the as-prepared electrocatalysts, suggesting that shrinking the lanthanide effectively tunes the reaction kinetics of Ir acidic OER learn. The double-layer capacitance (C dl ) of the catalyst is derived from the cyclic voltammetry (CV) curve (Figure 3c, Figure S7, Supporting Information). The electrochemical active surface area (ECSA) was calculated by dividing the integrated CO oxidation charge by 0.42 Mc cm À2 ( Figure S8, Supporting Information). The ECSA of Ir/G, Ir 2 La/G, Ir 2 Ce/G, Ir 2 Pr/G, Ir 2 Nd/G was 17.4, 59.7, 39.2, 95.2, 52.3 m 2 g À1 , respectively. By alloying with Ln, Ir 2 Ln exposed more active sites. The change in Ir activity comes from the change in the electronic structure of Ln in the alloy, and the OER activity of Ir in the alloy is dependent on the electronic structure of lanthanides. It is worth noting that the order of OER mass activity of the catalyst is Ir 2 Pr/G > Ir 2 Nd/ G > Ir 2 Ce/G > Ir 2 La/G > Ir/G. OER activity showed a volcanic relationship. A representative Ir 2 Pr/G yielded a current density of 30.15 mA cm À2 and a mass activity of 2.10 A mg À1 Ir at 1.53 V www.advancedsciencenews.com www.advenergysustres.com (vs RHE), respectively, improving the performance of Ir/G (a current density of 3.52 mA cm À2 and a mass activity of 0.22 A mg À1 Ir ) by a factor of 8.6 and 9.5, which verifies the significant enhancement of electrocatalytic activity (Figure 3d). In addition, the EIS spectra of Ir 2 La/G, Ir 2 Ce/G, Ir 2 Pr/G, Ir 2 Nd/ G, and Ir/G exhibit semicircular curves, which are in good agreement with the equivalent circuits (Figure 3e). The semicircle radius of the Ir alloy catalyst is smaller than that of Ir/G, which indicates that the alloying of Ir with Ln effectively accelerates the electron transfer in OER electrolysis. There was almost no difference in solution resistance (R s ) of all samples, among which Ir 2 Pr/G had the smallest charge transfer resistance (R ct ) of 225.3 Ω, (Ir/G was 1548.0 Ω, Ir 2 La/G was 870.2 Ω, Ir 2 Ce/G was 550.8 Ω, and Ir 2 Nd/G was 441.6 Ω), indicating the excellent charge transfer rate and electronic conductivity. The OER activities of Ir 2 La/G, Ir 2 Ce/G, Ir 2 Pr/G, and Ir 2 Nd/G in this work were compared with the mass activities and overpotentials of other previously reported and Ir-based catalysts, as shown in Figure 3f. [5,[30][31][32][33][34][35][36][37][38][39][40][41][42][43] A comparison table with Ir and Ir alloy catalysts recently reported in the literature is shown in Table S4, Supporting Information. It is evident that among the Ir alloy catalysts, Ir 2 Pr/G exhibits relatively high mass activity and low overpotential for OER. The stability of the alloy catalysts and Ir/G catalysts in OER was investigated. After 3000 consecutive cycles, the overpotentials of Ir 2 La/G, Ir 2 Ce/G, Ir 2 Pr/G, and Ir 2 Nd/G catalysts increased by only 6, 5, 4, and 4 mV, much lower than Ir/G (44 mV) ( Figure S9, Supporting Information). After the stability test, the surface and the morphology of Ir 2 Pr/G samples did not obviously change, supporting the robust electroactivity and durability ( Figure S10, S11, Supporting Information). Meanwhile, alloying Ir with Ln achieves the dual improvement of catalytic activity and stability. In addition, Ir 2 Pr/G was supported on carbon fibers to reduce the effect of bubbles on the catalyst surface to evaluate its lifetime. The Ir 2 Pr/G catalyst was found to stably catalyze acidic OER at 10 mA cm À2 for over 200 h (Figure 3g). Tafel plots. c) C dl curves by plotting current density versus scan rate. d) values of overpotential at a current density of 10 mA cm À2 and current density at 1.53 V vs RHE. e) The Nyquist plots, inset is an equivalent circuit. f ) Comparison of the overpotential at a current density of 10 mA cm À2 and mass activity of the Ir-Ln nanoalloys with the representative reported Ir alloy OER catalysts. g) Chronopotentiometry curves for Ir 2 Pr/G and Ir/G. To reveal the electronic structures of Ir 2 Ln alloys and their corresponding influences on the OER process, we have carried out systematic DFT calculations to investigate the electronic modulations and reaction trends. For the electronic distributions near the Fermi level (E F ), we have compared the Ir 2 La and Ir 2 Pr regarding the bonding and antibonding orbitals (Figure 4a). Notably, most of the surface electroactive regions are mostly dominated by the Ir sites, whereas the contributions from La sites are limited. The antibonding orbitals are slightly more evident on the surface, supporting the relatively lowered electroactivity. In comparison, the Pr sites on the surface of Ir 2 Pr are actively involved in the bonding and antibonding orbitals, which are attributed to the appearance of 4f orbitals (Figure 4b). Moreover, the bonding orbitals are enriched on the surface of Ir 2 Pr, leading to improved electron transfer and electroactivity toward the OER process. For both Ir 2 Ln alloys, no evident lattice distortion is noted after the relaxation, which supports potentially high stability. Then, we further interpret the detailed electronic contributions of the electronic structures in the most electroactive Ir 2 Pr (Figure 4c). It is noted that Ir-5d orbitals locate very close to the E F , which play as the main active sites for the OER process. Meanwhile, the Pr-4f orbitals show strong occupation crossing the E F , which further alleviates the energy barriers of electron transfer from Ir sites. The Pr-5d orbitals dominate above the E F , which overlaps with Pr-4f orbitals to support the fast electron transfer and protect Ir sites under the acidic media. Similar to Pr-4f orbitals, the Nd-4f orbitals also show strong occupation crossing the E F , which alleviates the overall energy barriers for d-d electron transfer ( Figure S12, Supporting Information). This leads to relatively good performances of Ir 2 Nd toward OER. The Ce-4f orbitals also locate close to E F, which also facilitates electron transfer with slightly larger energy barriers than Ir 2 Pr and Ir 2 Nd. In contrast, the Ir 2 La shows no contributions from the 4f orbitals, where the significant energy barriers between Ir-5d and La-5d result in the lowest OER performances as experimental characterizations. Since the electrochemical performances of the OER are not linearly changed with the Ln elements, we have correlated the d-band center of Ir active sites with the overpotential (Figure 4d). Notably, the d-band center of Ir shows consistent changes with the overpotential, whereas the lower d-band center leads to the lower overpotential. This indicates that the slightly lowered d-band center of Ir sites potentially optimizes the binding strengths with the protons in the acidic media, which promotes www.advancedsciencenews.com www.advenergysustres.com the conversion of H 2 O to facilitate the OER process. In the meantime, the overall d-band center still locates at a relatively high position to maintain the overall high electroactivity of the Ir 2 Ln alloys. For RE metals, we notice a highly distinct trend in their electronic structures (Figure 4e). La-5d orbitals have shown a broad coverage with very low electron density near the E F , which leads to the lowered electron transfer for the OER. For the 4f orbitals, we notice that Ce-4f orbitals display a much lower electron density than Pr-4f and Nd-4f orbitals. For Pr and Nd, their 4f orbitals are able to alleviate the electron depletion barrier from the alloy surface to the intermediates, which results in the highest electroactivity of the OER process. For different Ir sites within Ir 2 Pr, the site-dependent PDOSs are demonstrated (Figure 4f ). Compared to the Ir metal, the overall 5d orbitals of Ir sites within Ir 2 Pr are much upshifted toward the E F . From the bulk position to the surface, we notice a gradually upshifting trend of the 5d orbitals, which supports the increased electroactivity of the active sites, especially on the surface. In the end, we have compared the reaction trend of the OER from the energetic perspective (Figure 4g). For the water dissociation, it is noted that Ir 2 Pr shows a much smaller energy barrier of 0.28 eV than that of Ir 2 La with a barrier of 0.51 eV, leading to a stronger reaction trend toward the OER. For the OER process without any applied potential (U = 0 V), the OER reactions exhibit a continuous uphill trend, where the conversion from O* to OOH* meets the largest energy barrier as the potentialdetermining step (Figure 4h). Ir 2 Pr displays a smaller barrier than that of the Ir 2 La for the acidic OER. With the applied equilibrium potential of 1.23 V, the overpotential of the OER has been estimated to be 0.26 and 0.47 V for Ir 2 Pr and Ir 2 La, respectively (Figure 4i). Notably, the initial step of forming OH* has shown a smaller barrier for Ir 2 Pr, which is consistent with the water dissociation trends. The much increased overpotential of Ir 2 La is attributed to the overbinding of O* due to the highest d-band center. Therefore, both electronic structure and energetic results have revealed the subtle modulations induced by RE elements and their influences on the acidic OER.

Conclusion
In conclusion, a rational reduction route was developed for the synthesis of Ir-RE nanoalloys. Ir-Ln (Ln = La, Ce, Pr, Nd) with uniform particle size was prepared on graphene as an efficient OER catalyst. In an acidic environment, the electronic structure of Ir is efficiently tailored by alloying with different RE elements. The relationship between lanthanide contraction and catalyst OER activity follows a "volcanic" relationship, and the catalytic activity and stability of the alloy catalysts are both improved. Ir 2 Pr/G exhibits a mass activity of 2.10 A mg À1 Ir and stability over 200 h, which is 9.5 times and 20 times that of the Ir/G catalyst. DFT calculations have revealed that the superior OER performance of Ir 2 Pr is ascribed to the facilitation by the 4f orbitals in electron transfer and the lowest d-band center of active sites among Ir 2 Ln alloys supply the optimized binding strength of the intermediates, which all guarantee the efficient OER process with lowered overpotential. In this work, the detailed study of the electron transfer mechanism of Ir-Ln nanoalloy catalysts is expected to lay a solid foundation for the novel design of RE alloys in the future. This work will further expand the variety of RE alloy nanoalloy catalysts and paves the way for designing electrocatalysts with high catalytic activity and stability under harsh conditions.
Catalyst Synthesis: 0.3 Â g H 2 IrCl 6 ·6H 2 O, CN 2 H 2 , RE chloride (the mole ratio of Ir, Ln, and CN 2 H 2 were 5:3:80), and 0.45 Â g graphene were mixed in an agate mortar. The obtained mixture was then transferred into a boron nitride (BN) crucible for heat treatment. Then a Na ingot was placed in another BN crucible. These BN crucibles were placed in a big BN crucible (% 20 cm 3 inner volume) so that the sodium vapor produced would only contact the powder mixture when heated. All the heat treatment of the sample was performed under inert gas (argon) using gloveboxes. The temperature was increased to 600°C at the rate of 5°C min À1 . It was kept at this temperature for 2 h, and it is cooled down to room temperature. After the heat treatment, the powder samples were washed with 2-propanol to remove residual sodium, and the remaining sodium in the BN crucible is recycled. The obtained product was transferred to the air and leached in 500 mL of 0.5 M H 2 SO 4 at 40°C for 1 h under continuous stirring. The product was then thoroughly washed with Milli-Q water and vacuum-dried at 70°C for 5 h, yielding the final catalyst. The Ir/G was subjected to the same synthesis conditions but without the addition of LnCl 3 (the mole ratio of Ir and CN 2 H 2 was 5: 80) for comparison.
Electrochemical Measurements: 2 mg synthesized catalysts were dispersed in a mixture of 490 μL ultrapure water, 490 μL isopropanol, and 20 μL Nafion solution, after sonication for 1 h. Electrochemical measurements were conducted on a CHI 660 E Electrochemical Workstation (Shanghai Chenhua Instrument Corporation, China) in a conventional three-electrode system. The graphite rod electrode is the counter electrode, a mercurous sulfate electrode is the reference electrode, and 0.5 M H 2 SO 4 solution is the electrolyte (pH = 0.35). Every working electrode was activated by CV for 10 cycles with a sweep rate of 100 mV s À1 before electrochemical measurements. All of the potentials were converted into the potential against reversible hydrogen electrode (RHE) based on the formulas, E(RHE) = E(Hg/Hg 2 SO 4 ) þ0.656 þ 0.0592 pH. Bubble high-purity Ar in the electrolyte for 30 min before measurement to exclude dissolved oxygen. The working electrode was a glassy carbon electrode (GCE, diameter: 5 mm, area: 0.196 cm 2 ). 10 μL of the catalyst was dropped onto the GCE surface for further electrochemical tests. All the potentials reported in this work were converted to the RHE. OER performance was investigated by LSV at the scan rate of 5.0 mV s À1 . All the polarization curves were obtained with an ohmic potential drop (iR) correction arising from the solution resistance. Ohmic losses were corrected by subtracting the ohmic voltage drop from the measured potential, using the electrolyte resistance determined by the electrochemical impedance spectroscopy (EIS) at a high frequency where iR-corrected potentials were denoted as E À iR s (i as the current and R s as the electrolyte resistance). The Tafel slopes were derived from LSV curves at low overpotential fitted to the Tafel equation: η = a þ b logj, where η is the overpotential, b is the Tafel slope, j is the current density, and a is the constant. The long-term durability of the catalyst was examined through chronopotentiometry at a constant current density of 10 mA cm À2 to evaluate the durability of the samples supported on the carbon cloth. CV curves were collected at different scan rates (5,10,15,20, and 25 mV s À1 ) in a non-Faradaic potential window, which was measured from 1. 21  Physical Characterization: The powder XRD measurements were carried out on Rigaku Smart Lab 3 kW, an X-Ray diffraction diffractometer using monochromatized Cu Kα radiation (λ = 1.5418 Å). XPSwere conducted using a Thermo Scientific ESCALAB 250Xi instrument equipped with an Al X-Ray excitation source. Transmission electron microscope (TEM) and high-angle annular dark-field scanning TEM images and energy-dispersive spectroscopy (EDS) were obtained using a field-emission TEM (JEM-2800, Japan). The contents of the electrocatalyst were analyzed by inductively coupled plasma mass spectrometry (Elan drc-e, USA). Data reduction, data analysis, and EXAFS fitting were performed and analyzed with the Athena and Artemis programs of the Demeter data analysis packages that utilize the FEFF6 program to fit the EXAFS data. [44,45] The energy calibration of the sample was conducted through a standard Ir foil, which as a reference was simultaneously measured. A linear function was subtracted from the pre-edge region, then the edge jump was normalized using Athena software. The χ(k) data were isolated by subtracting a smooth, three-stage polynomial approximating the absorption background of an isolated atom. The k 3 -weighted χ(k) data were Fourier transformed after applying a Kaiser-Bessel window function (Δk = 1.0). For EXAFS modeling, the global amplitude EXAFS (CN, R, σ 2 , and ΔE 0 ) was obtained by nonlinear fitting, with least-squares refinement, of the EXAFS equation to the Fourier-transformed data in R-space, using Artemis software, EXAFS of the Ir foil is fitted and the obtained amplitude reduction factor S 0 2 value (0.829) was set in the EXAFS analysis to determine the coordination numbers (CNs) in the Ir-O/Pr/Ir scattering path in the sample.
Calculation Setup: In this work, we select the DFT within the CASTEP to investigate the electronic modulations induced by different RE elements in the Ir 2 Ln alloys. [46] The generalized gradient approximation and Perdew-Burke-Ernzerhof have been chosen to supply accurate descriptions of the exchange-correlation energy. [47][48][49] Then, the cutoff energy has been set to 380 eV based on the ultrasoft pseudopotential scheme. [50] The Broyden-Fletcher-Goldfarb-Shannon algorithm is applied with the coarse setting of the k-point mesh. [51,52] The Ir 2 Ln model has been built based on experimental characterizations with exposed (311) surfaces for the OER process. The following convergence criteria have been applied to guarantee the geometry optimizations including the Hellmann-Feynman forces should not exceed 0.001 eV Å À1 ; the total energy and the interionic displacement should be smaller than 5 Â 10 À5 eV atom À1 and 0.005 Å, respectively.

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.