2D single-faceted IrO2(101) monolayer enabling high-performing proton exchange membrane water electrolysis beyond 8,000 h stability at 1.5 A cm-2

doi:10.21203/rs.3.rs-5187955/v1

Download PDF

Article

2D single-faceted IrO2(101) monolayer enabling high-performing proton exchange membrane water electrolysis beyond 8,000 h stability at 1.5 A cm-2

https://doi.org/10.21203/rs.3.rs-5187955/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Both commercially available and laboratory-synthesized IrO₂ catalysts typically possess rutile-type structures and diverse facet orientations. According to the theoretical results from density functional theory calculations, distinct IrO₂ facets will result in divergent electrocatalytic properties, among which the (101) crystal facet is theoretically predicted as the most energetically favorable for oxygen evolution reaction (OER) owing to its lowest energy barrier. Maintaining a single-unit-cell thickness while exposing a desired facet of 2D IrO₂ presents a significant opportunity and challenge for the development of high-performance OER anode catalysts. Herein, we develop an ammonia-induced facet engineering for oriented modulation of crystal facets in the ultimate limit of monolayer thickness, and successfully synthesize 2D monolayer IrO₂ exposing unique (101) facet. At the current density of 10 mA cm^-2_geo, an ultralow overpotential of 230 mV has been achieved on the highly activated (101) facet in a three-electrode system. More importantly, in a proton exchange membrane (PEM) electrolyzer, the IrO₂ anode reaches a low voltage of 1.74 V at an industrial-level current density of 2 A cm^-2_geo, much lower than that of all commercial IrO₂ electrocatalysts. Though facet engineering primarily contributes to modulating the intrinsic activity rather than stability, the as-prepared IrO₂(101) monolayer performs over 8,000 hours of PEM water electrolysis (PEMWE) stability at constant 1.5 A cm^-2_geo, with a negligible decay rate of 4.0 mV kh^-1. Furthermore, even a long-term PEMWE test of 1000 h using the membrane electrode assembly (MEA) with ultra-low Ir loading of 0.2 mg_Ir cm^-2_geo under fluctuating operating conditions is performed, E_Cell remains highly electrochemically stable over time at 1.5 A cm^-2_geo, without any signs of catalyst degradation. This work proposes that ammonia-induced facet engineering of 2D monolayer IrO₂ could represent a novel approach to selectively expose the desired (101) facet, thereby enabling unique facet-dependent OER performance and ultrahigh stability in industrial-scale PEM electrolysis, even under voltage fluctuations generated by solar and wind power.

Physical sciences/Materials science/Materials for energy and catalysis/Electrocatalysis

Physical sciences/Materials science/Nanoscale materials/Two-dimensional materials

Hydrogen (H₂), benefiting from its low-carbon intensive, high energy capacity and easy recyclability, is widely applied as an ideal energy carrier to ease the burden of energy and climate issues. Large-scale production of green hydrogen is highly dependent on the conversion of surplus electricity from renewable but fluctuating sources like solar, wind, and hydro power¹. Coupling proton exchange membrane water electrolysis (PEMWE) technology with fluctuating renewables presents a promising avenue for producing green hydrogen, which is highly desired to achieve the international goal of ‘‘carbon neutrality’’. The sluggish kinetics of the oxygen evolution reaction (OER) at the anode are significantly hindered by electrocatalysts, posing a primary challenge to PEMWE commercialization^{2, 3}. An even more critical impediment lies in the limited stability of electrocatalysts under both steady-state and fluctuating operational conditions, especially in a strongly acidic environment. To fulfill the industrial demands of PEMWE systems, there is an urgent need to develop ultrahigh stable and active electrocatalysts, particularly anodic OER catalysts that can operate robustly at industrial-level constant current density or even highly volatile current densities, as generated by fluctuating renewables such as solar, wind, and hydro power⁴.

To date, operating current density of 3 A cm^− 2_geo under a low electrolytic voltage of 1.80 V is a basic but harsh guideline on OER electrocatalysts for the US Department of Energy (DOE) 2026 technical target. However, extensive current researches reveal an inevitable trend that most IrO₂-based nanostructured catalysts hardly maintain OER efficiency beyond 120 h even at a low current density of 10 mA cm^− 2_geo in a three-electrode system⁵. According to the DOE 2022 durability targets for both PEM electrolyzer stacks and systems, the commercial use of electrocatalyst must allow for long-term operation exceeding 40,000 hours with a decay rate of less than 4.8 mV kh^− 1. The huge research gap between DOE durability targets and the current status of anode electrocatalysts evaluated in the laboratory becomes the most imposing barrier preventing large-scale PEMWE commercialization⁶. Thus, developing novel IrO₂ system to balance OER activity and stability remains a significant challenge towards electrocatalyst design and application, thus leaving a large room for further study.

Until now, IrO₂ has been widely deemed as the best anodic catalyst considering the balance between OER activity and stability⁷. Doping engineering, support engineering, crystal phase engineering and morphology engineering have been adopted as effective strategies to maximize intrinsic Ir activity and minimize Ir loading amount⁸. For this reason, a series of novel IrO₂-based materials, including nanoparticles⁹, nanoneedles¹⁰, nanotubes¹¹, nanosheets^{12, 13}, nanoribbons¹⁴, and even single atom^{15, 16, 17}, have been designed, synthesized, and applied for acidic OER. Although various morphology engineering strategies have been implemented to develop highly active IrO₂ electrocatalysts, both laboratory-synthesized and commercial IrO₂ mainly exist in a conventional rutile phase with (110), (101), (200) and (211) crystal facets. Selecting an optimal crystal facet as the growth orientation to construct IrO₂ may offer an alternative pathway for the direct synthesis of high-performance anode electrocatalyst. Through 2D edge epitaxial growth, it is possible for us to achieve the unique crystal orientation in the monolayer limit without substrate persist. Accordingly, 2D monolayer IrO₂ with a unique (101) facet (denoted as IrO₂(101) monolayer) hold great promise in OER catalysis due to several structural advantages. First, the 2D monolayer nanostructure makes each Ir atom easily accessible to H₂O molecules during OER. Second, the completely uniform and periodic structure makes all surface atomic sites are equally active and equally accessible, which is close to homogeneous catalyst. Third, the desired (101) facet can be well exposed via altering the ammonia-induced epitaxial growth at a monolayer thickness. As anticipated, the single-faceted rutile structure may simultaneously maximize the intrinsic activity and stability of IrO₂ in both three-electrode cells and PEM device. However, most IrO₂ electrocatalysts are synthesized from the high-temperature oxidation of Ir salts using a traditional Adams combustion method, making facet engineering at a monolayer thickness more challenging¹⁸.

Here, by developing an ammonia-induced facet engineering, we synthesize 2D single-faceted IrO₂(101) monolayer and provide an insight into the long-term PEMWE applications under various operating conditions. Taking advantages of 2D monolayer structure and highly activated (101) facet, IrO₂ electrocatalyst exhibits a lowest overpotential of 237 mV at 10 mA cm^− 2_geo for OER than that of commercial IrO₂ (C-IrO₂). More importantly, the PEM electrolyzer employing IrO₂(101) monolayer as anode catalyst enables over 8,000 h stability test at 1.5 A cm^− 2_geo. Even at a low Ir loading of 0.2 mg_Ir cm^− 2_geo, IrO₂(101) monolayer anode achieves a low overpotential of only 552 mV and maintains stability for over 2,000 hours stability at industrially-relevant 2 A cm^− 2_geo. The ultrahigh OER stability derived from the single (101) facet was also confirmed in the next 1000 h fluctuating test for PEMWE.

Conventional IrO₂ electrocatalysts obtained through the high-temperature Adams fusion method commonly exist as irregular rutile-type nanoparticles with multiple exposed facets¹⁹. The rutile phase of IrO₂ typically exhibits three primary crystallographic facets: (110), (200) and (101) (Fig. 1a). Different catalytic facets will lead to different catalytic pathways, ultimately exhibiting different catalytic performances²⁰. As a result, the observed OER catalytic activity of IrO₂ often represents an average of these various facet-dependent activities. Achieving optimal OER activity through facet engineering of IrO₂ along a preferential orientation remains a significant challenge.

To gain deeper insights into facet-dependent catalytic performance and subsequently guide the synthesis towards desired facet orientation, we conducted comprehensive density functional theory (DFT) calculations using the Vienna Ab initio Simulation Package (VASP). We constructed adsorption models for the (101), (110), and (200) facets of rutile IrO₂, and calculated the Gibbs free energies (ΔG) of oxygen intermediates (*OH, *O, and *OOH) to quantitatively assess their adsorption strengths based on both the adsorbate evolution mechanism (AEM) and the lattice oxygen mechanism (LOM)^{21, 22}. Our calculations reveal that, for the AEM, the (101) facet exhibits the lowest energy barrier of 0.53 eV (at U = 1.23 V), surpassing the (200) and (110) facets with barriers of 0.64 eV and 0.72 eV, respectively (Fig. 1b). A similar trend is observed in Fig. 1c for the LOM, with the (101) facet again demonstrating the lowest energy barrier. Notably, for the (101) facet, the energy barriers for both AEM and LOM are identical, suggesting that during the OER process on this facet, both reaction mechanisms could potentially occur simultaneously or interchangeably. To elucidate the rate-determining step (RDS) across different facets, we present the reaction pathways in Fig. 1d-f. For the (110) and (200) facets, the RDS in both AEM and LOM is the desorption of O₂ species (step 4 in AEM and step 3 in LOM), indicating strong interactions with O intermediates. In contrast, for the (101) facet, the RDS is identified as the oxidation of *OH to *O (step 2 in both AEM and LOM), while the desorption steps exhibit significantly lower energy barriers compared to the other facets.

Analysis of the d-orbital projected density of states (PDOS) of the exposed Ir atoms in the three models reveals that the Ir atoms in the (101) facet exhibit the lowest d-band center at -2.46 eV, followed by the (200) facet at -2.3 eV, while the (110) facet displays the highest d-band center at -2.13 eV (Supplementary Fig. 1). Previous studies have demonstrated that rutile IrO₂ typically exhibits strong binding to O-based intermediates, which impedes the dissociation and desorption of these species, limiting OER activity^{14, 23, 24}. The downshift of the d-band center can effectively weaken the interaction between the catalytic site and intermediates by increasing the probability of anti-bonding orbital occupation, which is primarily responsible for the optimized OER activity observed on the (101) facet.

Based on these findings, constructing IrO₂ catalysts with a preferential (101) facet orientation holds the potential to significantly enhance performance and provide novel mechanistic insights. Nevertheless, facet-controlled synthesis of 2D IrO₂ materials with the desired orientation remains a challenge, particularly at high calcination temperatures during the traditional molten salt method. Guided by DTF prediction, we propose an ammonia-induced crystal phase engineering approach for the synthesis of the rutile IrO₂ monolayer featuring a single (101) facet.

In this work, we have successfully synthesized 2D rutile IrO₂ in the monolayer limit, uniquely oriented along the (101) facet. As shown in Fig. 2a, as-prepared IrO₂ is a freestanding ultrathin 2D material that can be successfully obtained via ammoniating process with K₂IrCl₆, KNO₃ and NH₃·H₂O. TEM images reveal the uniform planar structure and graphene-like morphology with an average size of 200 nm (Supplementary Fig. 2). The ultrathin character greatly improves their mechanical flexibility, and further causes surface corrugations and crumples to form on 2D IrO₂ (Fig. 2b). The high-resolution transmission electron microscopy (HRTEM) image shown in Fig. 2c further displays the 2D ultrathin structure of rutile IrO₂ with dominant (101) facet. The X-ray diffraction (XRD) technique was then carried out to analyze its crystalline phase and lattice fringe spacing. The selective-area electron diffraction (SAED) pattern marked by red solid box shows a single bright polycrystalline ring (inset in Fig. 2c). D-values were measured in the calibration patterns with a graphical tool by clicking on the two shortest non-collinear spots. Five obvious diffraction spots selected from this diffraction ring correspond to d-values of 0.2518, 0.2513, 0.2530, 0.2526, and 0.2547 nm, respectively, which are in agreement with the (101) lattice spacing²⁵. The corresponding intensity profile along the blue rectangle in high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM) image further exhibits an average space distance of ∼ 2.59 Å, which is almost same as the lattice parameter of 2.58 Å on the rutile-IrO₂ (101) facet (Fig. 2d)^{9, 26}. All the crystalline characteristics of 2D IrO₂ monolayer confirm a preferential crystal orientation growth along the (101) direction. EDS mapping and line-scanning spectra further display Ir, O and N atoms are well-dispersed over the monolayer structure (Fig. 2e and Supplementary Fig. 3). The percentages of Ir, O, and N atoms were measured to be 28.52%, 59.33%, and 12.15%, respectively, with an atomic ratio of Ir and O close to the theoretical value of 1:2 (Supplementary Fig. 4 and Supplementary Table 1). According to atomic force microscopy, 2D IrO₂(101) crystals are approximately 2 ~ 3 nm in height, which is close to the theorical thickness of an IrO₂(101) -(1×1) unit cell, and exhibit a uniform and smooth surface as measured by 3D structure simulation (Fig. 2f and Supplementary Fig. 5).

In general, commercially available IrO₂ exists in the rutile phase with (110), (101), (200) and (211) characteristic peaks in XRD pattern (Fig. 3a)²⁷. Different crystal facets will bring about different OER pathways, and finally display different OER performances²⁸. Obtaining a unique crystal orientation of IrO₂ via the Adams fusion method remains challenging, owing to the isotropic growth that occurs under high-temperature treatment. Interestingly, with the introduction of amine, only a single (101) diffraction peak are observed on the 2D IrO₂(101) monolayer, indicating the successful construction of a single-faceted structure through ammonia-induced preferred orientation growth. From a macroscopic perspective, the catalyst powder of IrO₂(101) monolayer exhibits a distinctly different indigo, green color, which contrasts with the grayish powder of C-IrO₂ purchased from Alfa Aesar and Umicore (Supplementary Fig. 6).

As shown in Fig. 3b and Supplementary Fig. 7, the X-ray photoelectron spectroscopy (XPS) measurements were conducted to compare the electronic structures of as-prepared IrO₂ and C-IrO₂. Except Ir4f and O1s signals, a negligible N1s peak is also detected in XPS spectrum of IrO₂(101) monolayer, revealing trace N dopant during the ammonia-induced growth process. The Ir4f spectra can be deconvoluted into four peaks, including Ir4f_7/2, Ir4f_5/2 and two accompanying satellite peaks (Fig. 3c and Supplementary Fig. 7a)^{29, 30}. The Ir4f binding energies of IrO₂(101) monolayer and IrO₂ NP are almost the same as that of C-IrO₂, indicating that the major oxidation state of Ir is + 4. Based on the O1s XPS spectra in Fig. 3d, the O1s peaks from corresponding lattice oxygen (Ir-O), unsaturated oxygen (Ovancy) and adsorbed oxygen (Oadsorbed) are clearly observed in both as-prepared IrO₂ and C-IrO₂¹⁴. Interestingly, both the Ir4f_7/2 and the O1s binding energies of IrO₂(101) monolayer exhibit obvious positive shifts towards higher binding energy than those of IrO₂ NP and C-IrO₂ (Supplementary Fig. 7b and Supplementary Table 2), suggesting a stronger electron-donating effect of 2D monolayer structure than that of nanoparticle structures.

Raman scattering is a powerful technique to characterize 2D ultrathin materials because of its strong sensitivity to the electronic structure. The peak positions of the Raman features are summarized in Supplementary Table 3. Interestingly, two major Raman peaks, namely the Eg and A1g modes for rutile IrO₂ NP and C-IrO₂ are identified, respectively, while only an Eg peak is detected at 540.09 cm-1 in the Raman spectrum of IrO₂(101) monolayer (Fig. 3e)^{31, 32}. The disappearance of A1g peak clearly reveals that the lattice vibrations of rutile IrO₂ are completely suppressed in the 2D single-faceted monolayer structure. Furthermore, the 2D monolayer nanostructure exhibits highest specific surface area of 353.8 m² g^− 1, almost 3 and 50 times higher than that of nanoparticle counterpart (114.2 m² g^− 1) and C-IrO₂ (7.7 m² g^− 1), respectively. As shown in Fig. 3f, the N₂ adsorption-desorption isotherm of IrO₂(101) monolayer can be categorized as type IV with a hysteresis loop. The pore-size distribution further exhibits the presence of mesopores with an average diameter of 3.29 nm on the 2D monolayer, which is attributed to the ammonia-induced pore-forming process (Supplementary Fig. 8 and Supplementary Table 4)⁹.

The coordination information of the 2D single-faceted monolayer structure was also corroborated by synchrotron X-ray absorption spectroscopy using the Ir L₃-edge. As shown in Fig. 4a, the X-ray absorption near-edge spectroscopy (XANES) curve of IrO₂(101) monolayer is in good accordance with that of C-IrO₂, indicating a similarity in the rutile phase. Meanwhile, IrO₂(101) monolayer exhibits a higher absorption energy (11219.7 eV) compared to C-IrO₂ (11219.6 eV), indicating Ir species mainly exist as a more oxidation state. This result is also consistent with those measured from XPS analysis. Figure 4b compares the Fourier transform (FT) k²-weighted EXAFS spectra of IrO₂(101) monolayer, C-IrO₂ and Ir foil. As a reference sample, Ir foil exhibits a strong peak at 2.7 Å, which corresponds to the Ir-Ir coordination³³. In contrast to Ir foil, no obvious peaks of Ir-Ir coordination were detected in IrO₂(101) monolayer. The scattering paths of the Ir centers in IrO₂(101) monolayer and C-IrO₂ exhibit a prominent peak of Ir-O coordination at 2.00 and 1.97 Å, respectively. Accordingly, the Ir L₃-edge EXAFS reveals that the length of the Ir-O bond in IrO₂(101) monolayer is slightly longer than that of C-IrO₂. The increase in Ir-O bond length may be attributed to lattice expansion induced by the ultrathin monolayer structure and unsaturated coordination environment^{34, 35}. Furthermore, the WT-EXAFS spectrum of IrO₂(101) monolayer shows that the shell area for Ir-O scattering at R = 1.50 Å and k = 4.50 Å-1, and no shell area for Ir-Ir scattering is observed, implying neither Ir nanoclusters nor nanoparticles exist (Fig. 4c). The extended X-ray absorption fine structure (EXAFS) results were further Fourier transformed to simulate the real crystal model of IrO₂(101) monolayer. Figure 4d exhibits the fitting curve of single (101) facet, which is perfectly reproduced by the experimental FT-EXAFS data. According to Ir L₃-edge EXAFS fitting parameters in Supplementary Table 5, the first-shell area for Ir-O scatter at R = 2.00 Å, the coordination number (Ir-O) is 5.928 and the bond length is 2.00140 Å. Combining the above XAFS results and previous studies, a schematic model of IrO₂(101) monolayer is presented in Fig. 4e. Each Ir atom coordinates with six O atoms to form a single layer of 2D rutile IrO₂ crystal along (101) orientation.

Although many facet engineering strategies have been proposed for tailoring the crystal facets under mild conditions, such as the additive-assisted solvothermal method and template approach, there are few studies on the facet engineering of rutile IrO₂³⁶. As a result of the traditional Adams fusion method, rutile IrO₂ tends to grow along various orientations at high oxidation temperatures (400–800°C) and eventually expose different crystal facets. When high oxidation temperature is essential, selecting the optimal crystal facet while maintaining the monolayer thickness of IrO₂ becomes an enormous but meaningful challenge. Considering the above inevitable synthetic difficulties, we developed a novel ammonia-induced facet engineering for oriented regulation of the IrO₂ facets at the ultimate monolayer thickness limit.

The key point of this facet-controlled synthesis has been investigated via modulating the synthesis parameters, elucidating that the nucleation and growth of IrO₂(101) monolayer are dominantly guided by NH₃·H₂O and temperature. During the annealing process, NH₃·H₂O is introduced as a potent crystal facet growth inducer for exposing desired (101) facet. As depicted in Supplementary Fig. 9, IrO₂ exhibits a classic rutile structure with strong (110), (101), (200) and (211) diffraction peaks in the absence of NH₃·H₂O. Upon the addition of 3.2 g NH₃·H₂O, the (110), (200) and (211) signals are significantly suppressed, while the (101) facet emerges as the dominant crystal surface, revealing the role of NH₃·H₂O in facilitating the unidirectionally oriented growth of IrO₂ grains. Particularly, as the NH₃·H₂O amount is further increased to 12.8 g, the characteristic peaks corresponding to the (110), (200) and (211) facets almost disappear. Therefore, IrO₂(101) monolayer has been successfully constructed through ammonia-induced growth, in which IrO₂ nucleation preferentially occurs along the (101) direction, further demonstrating the effect of NH₃·H₂O on promoting the preferred growth of the (101) facet. Analogously, the morphological evolution of IrO₂ from irregular nanoparticles to 2D monolayers is accompanied by the concurrent crystal evolution, as evidenced by the increased ammonia amount. The successful synthesis of single-(101)-faceted monolayer validates the intimate correlation between the morphological and crystallographic changes, where the preferential growth along the (101) facet guides the formation of the 2D monolayer structure (Supplementary Fig. 10).

In addition to NH₃·H₂O, the unidirectionally oriented growth is also highly dependent on the annealing temperature, making it challenging to expose the unique (101) facet at temperatures exceeding 500°C. To understand the role of temperature in crystal evolution, XRD characterization was then conducted at different temperatures to simulate the IrO₂ growth (Supplementary Fig. 11). The heat temperatures of 350°C, 450°C and 500°C are crucial to its crystal growth, during which the crystal facet and structure undergo distinct transformations (labeled as IrO₂(101)-350, IrO₂(101) monolayer and IrO₂(101)-500, respectively). According to the kinetic growth of IrO₂, the relative nucleation barriers (ΔE) of other facets are much higher than that of IrO₂(101) under 350°C and 450°C, resulting in the (101) facet becoming the dominating one. As temperature increases to 500°C, the growth velocities of (110), (200) and (211) facets accelerate significantly, which contribute to the isotropic growth behavior. Consequently, a conventional rutile IrO₂ structure with various crystalline facets will evolve preferentially. As evidenced by TEM images in Supplementary Fig. 12, 2D IrO₂ obtained at 350°C, 450°C, and 500°C exhibit average sizes of 100 nm, 500 nm and 2 µm, respectively, which correspond to the bottom-up growth. At 350°C, IrO₂ species tend to first nucleate at the sharp edges of the monolayer. Under a relatively high temperature, the growth rate of the monolayer edge become faster, and then the high-temperature crystal growth process facilitates the precise formation of micrometer-size monolayer structure along the (101) direction. As a result, the single-(101)-faceted structure can be well controlled via 2D edge epitaxial growth in the monolayer limit even without substrate persist. At a temperature of 500°C, the thermodynamically favored isotropic growth poses a major challenge to maintaining the 2D monolayer structure. Consequently, irregular IrO₂ nanoparticles inevitably form on the monolayer IrO₂(101) surface.

In order to study the structural and performance advantages, the OER measurements of IrO₂(101) monolayer, IrO₂ NP, C-IrO₂ (Alfa Aesar) and C-Ir black were initially carried out in a three-electrode configuration using 0.1 M HClO₄ as the aqueous electrolyte. All reference electrodes were calibrated before OER tests (Supplementary Fig. 13). As plotted in Fig. 5a, IrO₂(101) monolayer displays the lowest overpotential of 227 mV at a current density of 10 mA cmgeo^− 2, compared to 307, 311, and 301 mV of IrO₂ NP, C-IrO₂ and C-Ir black (Supplementary Fig. 14a). Even at 50 mA cm_geo⁻², IrO₂(101) monolayer maintains an ultralow overpotential of 272 mV, nearly 200 mV lower than its nanoparticle counterparts (Supplementary Fig. 14b). The most active linear sweep voltammetry (LSV) curve confirms that IrO₂(101) facet plays the most vital role in optimizing the OER performance over other facets. Tafel slope is another important parameter to probe the response of IrO₂ catalyst to the interfacial charge transfer towards the working potential. As depicted in Fig. 5b, IrO₂(101) monolayer has the lowest Tafel slope of 44.73 mV dec^− 1 within the potential range of 1.45–1.47 V. For comparison, higher Tafel slopes are obtained by IrO₂ NP, C-IrO₂ and C-Ir black, with delivering 83.06, 70.28 and 80.97 mV dec^− 1, respectively. The OER kinetic on the (101) facet is much faster than those of nanoparticle counterparts, confirming that the 2D single-facet monolayer structure is more favorable for OER pathway³⁷. The kinetics of OER processes on IrO₂(101) monolayer was further evaluated by EIS measurement (Supplementary Fig. 15 and Supplementary Table 6). Fitting EIS curves reveals that IrO₂(101) monolayer has the smallest charge transfer resistance (RCT) of 10.32 ohm, compared to those measured for IrO₂ NP (109.60 ohm), C-IrO₂ (88.49 ohm), and C-Ir/C (18.06 ohm). We attribute the lowest high frequency resistance to the maximized charge transfer process resulted from the monolayer thickness and highly oriented facet, which is the main origin of the superior OER performance. To identify the origin of monolayer structural advantages in mass transfer, we quantified the surface superaerophobicity of IrO₂ catalysts by assessing the adhesive force between bubbles and the surface (Supplementary Fig. 16). Interestingly, the adhesive force measured on the monolayer IrO₂(101) surface was found to be merely 17.7 µN, featuring excellent superaerophobicity in electrolyte. Employing IrO₂ NP and C-IrO₂ as contrastive samples, the values of adhesive force were measured to be 52.4 µN and 32.0 µN, respectively. Therefore, the bubble releases from the MEA surface of monolayer IrO₂(101) seems to be easier in the solution, which can be attributed to the exceptionally low contact region between the bubbles and the uniform monolayer surface and thus low adhesive force^{38, 39}. Above results demonstrate that the fine construction of 2D monolayer interface featuring a specific facet simultaneously optimizes both its kinetic behavior and wetting behavior, thereby facilitating efficient charge and mass transfer from the IrO₂ surface for OER.

We further calculated the specific current densities at 1.50 V vs. RHE to study their OER performances. To present a comprehensive performance evaluation of different catalysts, the OER performances were evaluated by normalizing surface areas obtained using different approaches, such as GCE (glassy carbon electrode) geometric areas and electrochemically active surface areas (ECSA)⁴⁰. IrO₂(101) monolayer exhibits both the highest geometrical activity of 47.75 mA cm_geo⁻² and ECSA-based activity of 0.027 mA cm_ECSA⁻² at 1.50 V vs. RHE, nearly 10 times higher than those of IrO₂ NP, C-IrO₂, and C-Ir/C (Supplementary Fig. 17a, b). The methods to determine the ECSA-based surface areas of different catalysts and clean GCE are summarized in Methods section (Supplementary Figs. 18 and 19). In addition, the mass activities at 1.50 V vs. RHE were calculated to compare their intrinsic activities. By normalizing the Ir loading, rationally designed IrO₂(101) monolayer performs 238.75 mA mg_Ir⁻¹ at 1.5 V vs. RHE, almost 10 times higher than that of other rutile IrO₂ (Fig. 5c). To gain a comprehensive view of the intrinsic performance, the OER performance of IrO₂(101) monolayer was also evaluated using 0.5 M H₂SO₄ as the electrolyte. IrO₂(101) monolayer exhibits the lowest overpotential of 245 mV at the current density of 10 mA cm_geo⁻², compared to 304, 344, and 305 mV for IrO₂ NP, C-IrO₂, and C-Ir/C, respectively (Supplementary Fig. 20a). The Tafel slope of IrO₂(101) monolayer in 0.5 M H₂SO₄ was measured to be only 47.47 mV dec^− 1 (Supplementary Fig. 20b). In addition, both mass activities and EIS results illustrate that IrO₂(101) monolayer shows the highest intrinsic activity and lowest electron transfer resistance (Supplementary Fig. 20c, d). Above OER measurements conducted in 0.5 M H₂SO₄ electrolyte further demonstrates that the IrO₂(101) facet provide an ideal catalytic platform for OER process by favoring mass transport, mechanical behavior, and even electron transfer.

In the nucleation and growth process, both NH₃·H₂O amount and temperature play vital roles in guiding the (101)-oriented growth of IrO₂ nuclei at a monolayer thickness. A perfect IrO₂(101) monolayer is predicted to possess highest electrocatalytic performance for OER compared to other traditional nanostructures, due to the ~ 100% exposure of the (101) facet on the monolayer. LSV curves reveal that IrO₂(101) monolayer exhibits the lowest overpotential of 227 mV to attain a current density of 10 mA cm^− 2_geo, which is 80 mV, 36 mV and 22 mV lower, respectively, than its counterparts synthesized with NH₃·H₂O amounts of 0 g, 3.2 g and 12.8 g (Supplementary Fig. 21). Insufficient ammonization is unfavorable for the preferred orientation growth of the highly active (101) facet, whereas excessive ammonization may disrupt the orderliness and integrity of the (101) crystal facet. In terms of annealing temperature, the overpotential of IrO₂(101)-350 and IrO₂(101)-500 were measured to be 257 mV and 357 mV at 10 mA cm^− 2_geo, respectively, significantly higher than that of IrO₂(101) monolayer (227 mV) (Supplementary Fig. 22). Thus, 450 ^oC is identified as the optimal temperature to favor the (101)-oriented growth of 2D monolayer structure.

In addition to activity, balancing the operating stability in acidic media poses another significant challenge when considering the structural stability at industrial-level current densities under practical OER condition^{41, 42}. In a three-electrode system, both IrO₂ NPs and C-IrO₂ suffer from dramatically activity loss during the 100 h chronopotentiometry (CP) tests while the working potential of IrO₂(101) monolayer at 50 mA cm^− 2_geo maintains a consistent value after 910 h (Fig. 5d). The overpotential decay rate was calculated to be only 15 µV h^− 1, demonstrating the superior stability of IrO₂(101) monolayer. The inset in Fig. 5d further exhibits that the linear polarization curves of IrO₂(101) monolayer before and after the CP test perfectly overlap. It also operates stably at the high current of 200 mA cm^− 2_geo during a 167 h stability test, confirming its great potential for commercial application (Supplementary Fig. 23). We also found that the 2D monolayer structure seems to be more insensitive to the working potential than previously reported nanostructured IrO₂. For instance, a strong redox peak of Ir^Ⅲ/Ⅳ appears at 0.86 V vs. RHE once an anodic potential is applied, indicating that Ir³⁺ has been oxidized to Ir⁴⁺ or an even higher state (Supplementary Fig. 24). Even after 10000 CV cycles, no obvious shift in the redox peak has been detected⁴³. The polarization curves and overpotentials of IrO₂(101) monolayer before and after the 2,000th /10,000th CV cycles tests also prove that its activity loss is negligible despite being a 2D monolayer structure (Supplementary Fig. 25). To provide a comprehensive evaluation of the stability property of IrO₂(101) monolayer, the inductively coupled plasma-optical emission spectrometer (ICP-OES) was further operated to calculate the dissolution amount of the Ir element after 910 h durability test at 50 mA cm^− 2_geo. Form the ICP-OES data in Supplementary Table 7, the S-number for each IrO₂ catalyst was calculated as the ratio of the molar amount of evolved O₂ to the molar amount of dissolved Ir⁴⁴. The S-number of IrO₂(101) monolayer after the 910 h OER was calculated to be 4.64×10⁶, over 11 and 40 times higher than the S-numbers of IrO₂ NP (4.18×10⁵) and C-IrO₂ (1.14×10⁵), respectively (Supplementary Table 8). In addition, IrO₂(101) monolayer exhibits superior stability compared to other recently reported Ir-based catalysts, including pristine Ir or Ir oxides (1 ~ 9.2×10⁵)⁴⁵, Ir-based alloys (5 ~ 2.49×10⁵)⁴¹, and IrO₂@TiO₂ nanoparticles (1.0×10⁴)⁴⁶. Given the high current density and the superlong test time, the performance stability of 2D IrO₂(101) monolayer in acidic media is further highlighted, pointing to its significant potential for commercialization even at monolayer thickness. As shown in Supplementary Figs. 26 and 27, the monolayer structure and uniform (101) facet orientation of the IrO₂(101) monolayer remain intact after ~ 1000 h stability test at 50 mA cm^− 2_geo. TEM analysis reveals an initial sheet-like morphology with no reconstructed nanoparticles or nanoclusters on the surface. XRD characterization further confirms the presence of only a prominent (101) diffraction peak at 34.93°, without any signals from other rutile facets, demonstrating high stability in the crystallographic orientation during long-term operation. To gain insights into the electronic structure changes after the stability test, XPS analysis was performed. As shown in Supplementary Fig. 28, XPS spectrum of the IrO₂(101) monolayer exhibits a slight blue shift of 0.15 eV compared to the initial state, suggesting the partial oxidation of surface Ir atoms under acidic OER conditions. Above characterizations after OER stability test confirm that the single-(101)-facet structure realizes a win-win strategy in terms of activity and stability, positioning IrO₂(101) monolayer as a promising candidate for PEM applications.

The catalyst research evolving from the simple modulation of IrO₂ composition/size on the macroscopic scale to the precise design of IrO₂ structure/morphology at the nanoscale suffer from the same corrosion problem stemming from the dissolution of IrO₄²⁻ anions into the electrolyte⁴². In general, most previously reported high-performance OER electrocatalysts can only survive in a three-electrode system, while their activity loss usually become magnified in PEM devices after 100 h, which starkly contrasts with the stability results observed on GCE⁴⁷. For the industrial application of PEM electrolyzers, searching an advanced IrO₂ material structure that can meet the technical targets for PEM electrolyzer—high current densities of 2 A cm^− 2_geo at low overpotentials (≤ 1.9 V) and long-term stability exceeding 40,000 h at industrial-scale current density with a decay rate below 4.8 µV h^− 1—presents a great opportunity and challenge for a sustainable PEM future^{6, 48, 49}. As a proof-of-concept of its commercial value, the OER performance of IrO₂(101) monolayer was evaluated in PEM devices under various operating conditions (Supplementary Fig. 29). As depicted in Fig. 6a, the catalyst-coated membranes (CCMs) for PEMWE were prepared using IrO₂(101) monolayer as the anode catalyst (0.2 mg_Ir cm^− 2_geo or 1.5 mg_Ir cm^− 2_geo) and commercial Pt/C (60 wt % Pt, Tanaka Kikinzoku Group) as the cathode catalyst via spray coating on both sides of the Nafion 115 membrane. After assembly, MEAs with IrO₂(101) monolayer were tested in a 4 cm² active area PEMWE single cell at 80°C and ambient pressure, with a preheated Milli-Q ultrapure water flow of 10 mL min^− 1 onto the anode.

IrO₂(101) monolayer performs remarkable PEM activity, achieving the standard current density of 2 A cm^− 2 _geo at a low potential of 1.74 V, significantly lower than E_cell of IrO₂ NP (1.82 V@2 A cm^− 2 _geo) and C-IrO₂ (1.84 V@2 A cm^− 2 _geo), respectively (Fig. 6b). Additionally, lowing the Ir loading in CCMs below 0.5 mg_Ir cm^− 2 _geo is another crucial issue to meet the deployment and upscaling of PEM electrolyzers, which can be ascribed to the ultralow-earth abundance and extremely high cost ($4700 per oz in 2024)^{18, 37}. Thus, PEMWEs with ultralow-Ir-loading MEA (0.2 mg_Ir cm^− 2 _geo) were recorded to evaluate catalyst performance. As shown in the inset of Fig. 6b, E_cell reaches 1.78 V at 2 A cm^− 2 _geo using ultralow-Ir-loading MEA, which is comparable to the best PEMWEs employing same MEAs with substantially higher Ir loadings (1 ~ 2 mg_Ir cm^− 2 _geo)^{50, 51, 52}.

Demonstrating the catalyst stability in PEM device is a more challenging task than achieving high PEM activity. To date, most OER catalyst stabilities are still tested in a simple three-electrode system utilizing a low working current density below 200 mA cm^− 2_geo, which perform far from their practical working conditions. Although some recently reported catalysts have demonstrated operation in PEM system for over 100 h at a constant current density of 1 ~ 2 A cm^− 2_geo, evaluating their comprehensive working stabilities in all aspects including the stability response to fluctuations in wind/solar energy inputs or ultralow Ir loading, is easy to ignore⁶. Therefore, many high-performance OER electrocatalysts developed from academia often rest in the laboratory, while commercially available IrO₂ is exploring to address technical problems encountered from real applications⁵³. Compared to the steady-state condition, voltage fluctuation caused by wind or solar power can exacerbate Ir loss and even catalytic deterioration, ultimately compromising the overall PEM efficiency^{54, 55}. To bridge the research gap of stability tests, we use a square wave voltage toggling between 1.6 and 2.3 V to simulate the fluctuating power patterns of wind or solar photovoltaic sources. As depicted in Fig. 6c, E_Cell over time was measured at 1.6 and 2.3 V for 1 minute each. Notably, voltage switching exhibits little negative effect on PEM durability, suggesting IrO₂(101) monolayer remains both activity and structural advantages even after 1,000 h fluctuating testing.

As for steady-state condition, stability assessments of OER catalysts are typically conducted in a simple three-electrode or PEM system, where constant voltages are applied from external electrochemical workstation. This is a key criteria before we can commercialize it. Figure 6d firstly compares the stability data of low-Ir-loading PEM electrolyzers. Even with Ir and Pt loading as low as 0.2 mg cm^− 2_geo, E_cell achieves over 2,000 h stability without obvious activity loss at 1.5 A cm^− 2_geo. As shown in Fig. 6e, IrO₂(101) monolayer not only meets the DOE technical standards but also significantly outperforms previously reported catalysts in terms of both voltage efficiency and stability^{10, 12, 33, 46, 47, 56, 57, 58, 59, 60, 61}. As evidenced in Fig. 6f, the IrO₂(101) monolayer anode cell (1.5 mg_Ir cm^− 2_geo) maintains an ultrahigh OER stability of 8,000 h at 1.5 A cm^− 2_geo with a negligible degradation rate of 4 µV h^− 1. Up to now, the PEM system is still operating stably. In contrast, for C-IrO₂ and IrO₂ NPs, their efficiency loss is drastically raised during the 500 h OER test, which may be attributed to the accelerated catalyst degradation. All these comprehensive stability tests confirm that IrO₂(101) monolayer suffers less catalytic deterioration under both dynamic and constant power operation, despite its monolayer rutile (101) structure.

Guided by theoretical calculations, we have successfully synthesized 2D single-faceted IrO₂(101) monolayer based on an ammonia-induced preferred orientation growth. Both XRD characterization and TEM observation confirm that IrO₂(101) monolayer possesses a monolayer rutile structure featuring a single (101) facet. The highly activated (101) facet presents an ultralow overpotential of 237 mV at 10 mA cm^− 2_geo in a three-electrode cell, and performs excellent PEM activity with a low E_cell of 1.74 V at 2 A cm^− 2_geo. Importantly, the practical operating stability of the IrO₂(101) monolayer anode cell has been tested under various working conditions. Although IrO₂ crystal is constructed in the monolayer limit along a unique growth orientation, IrO₂(101) monolayer still meets the DOE durability targets for PEMWE. Under steady-state condition, the long-term operation exceeds 8,000 hours with a negligible decay rate of less than 4.0 mV kh^− 1. Even at an ultralow Ir loading of 0.2 mg cm^− 2_geo, the PEMWE test for IrO₂(101) monolayer anode maintains stability for over 2,000 h at a constant 1.5 A cm^− 2_geo. When coupled with simulated fluctuating electricity from solar/wind power sources, IrO₂(101) monolayer anode cell also achieves long-term stability exceeding 1,000 h. This work proposes a novel phase engineering strategy for the design and synthesis of 2D single-(101)-faceted IrO₂ at monolayer thickness, demonstrating outstanding OER activity and long-term PEMWE stability under both steady-state condition and fluctuating condition.

Chemicals. Dipotassium hexachloroiridate (K₂IrCl₆, 99.9%) was purchased from Alfa Aesar Co. Iridium(IV) oxide (IrO₂, 99%) was purchased from Alfa Aesar and Umicore. Ir Black (Ir, 99.9%) was purchased from Apinno Co. Ammonium hydroxide (NH₃·H₂O, 25.0–30.0%) and isopropanol were purchased from Beijing InnoChem Science & Technology Co., Ltd. Sulphuric acid (H₂SO₄, 95.0–98.0%), perchloricacid (HClO₄, 70.0–72.0%), and Potassium nitrate (KNO₃, ≥ 99.0) were obtained form Sinopharm Chemical Reagent Co. Nafion solution (5 wt %) and Nafion N115 membrane was purchased from DuPont Co. Other reagents used were of analytical grade without further purification. Deionized water, supplied by Thermo Scientific Smart2Pure 6 UV Water Purification System, was used throughout the experiments. The PEM device was purchased from China Titanium Guochuang (Qingdao) Technology Co.

Synthesis of 2D IrO ₂ (101) monolayer. The synthesis of IrO₂ was carried out in a muffle furnace. 240 mg of K₂IrCl₆ and 6.4 g of NH₃·H₂O were completely dissolved in 40 mL deionized water. After the mixture had been reacting for 3 h ultrasonically, 4.848g of KNO₃ was added and stirred for 1 h. After that, the mixture was heated at 80°C and magnetically stirred until nearly complete evaporation of water was achieved. The resulting powder was then placed in a quartz boat and heated to 450°C at a rate of 5°C min^− 1 and held at this temperature for 0.5 h in air. It was then allowed to cool naturally to room temperature. Finally, IrO₂(101) monolayer was obtained by filtering and washing with deionized water, respectively, followed by drying at 60°C in a vacuum oven overnight. The comparison materials were prepared by altering the calcination temperatures to 350°C and 500°C, and by modulating the volume of ammonia solution to 0 mL, 3.2 mL, and 12.8 mL. Notably, IrO₂ NP was synthesized specifically using 0 mL of ammonia solution.

Characterizations. The XRD patterns of all samples were recorded by X-ray powder diffraction (XRD, D8 advance) at an operation voltage of 40 kV and a current of 40 mA, using a Cu Kα radiation source. The TEM images and EDX data of all samples were characterized by a JEOL JEM-F200 transmission electron microscope with an accelerating voltage of 200 kV. AFM images were captured by the atomic force microscopy (AFM, Dimenson ICON). The XPS results were tested by X-ray photoelectron spectrometer (Thermo Fischer, ESCALAB250Xi) using an operation voltage of 12.5 kV and a filament current of 16 mA, with the excitation source using Alka rays (hv = 1486.8 eV). And the charge correction is based on C1s at 284.80 eV as the energy reference. The BET specific surface areas were characterized using American Micromeritics 3Flex Version 5.00. The Ir dissolution concentration was determined by Agilent 7800 ICP-MS. The Raman spectroscopies were measured using a Laser Microscopic Confocal Raman Spectrometer (inVia™, Renishaw) under excitation by laser light at λ = 532 nm. The adhesion was recorded using an adhesion tester (DCAT 25, DataPhysics) at a line voltage of 24 V DC and an input power of 70 W. XAFS spectra were collected in fluorescence mode at the 1W1B X-ray absorption beamline of the Beijing Synchrotron Radiation Facility (BSRF) using a double-crystal Si (111) monochromator. Iridium foil was measured in transmission mode and used as a reference for energy calibration. Data were analyzed with the ATHENA and ARTEMIS (version 0.9.26) modules of the IFEFFIT software package. The EXAFS functions were Fourier transformed to R-space with a k-weight of 2, within the range of 3.16–12.40 Å⁻¹, using a Hanning window (dk = 1.0 Å⁻¹).

Electrochemical measurements. All electrochemical experiments in the three electrode system were carried out using the CHI760E electrochemical workstation (Shanghai Chenhua, China). The glass carbon electrode (GCE, area: 0.196 cm²) was used as the working electrode, calomel electrode was used as the reference electrode, and platinum gauze was used as the counter electrode. OER measurements were conducted in O₂-saturated 0.1 M HClO₄ or 0.5 M H₂SO₄ electrolyte. For measurement with iR-correction, R referred to the ohmic resistance arising from the electrolyte/contact resistance of the setup. The scan rate was set to 2 mV s⁻¹ for both LSV measurements ranging from 1.1 to 1.6 V vs. SCE, and Tafel plot measured from 1.1 to 1.4 V vs. SCE. The Electrochemical Impedance Spectroscopy (EIS) tests were conducted with a frequency range from 10 Hz (low frequency) to 100 kHz (high frequency). Chronopotentiometry measurements were performed using the Chronopotentiometry mode on the CHI760 electrochemical workstation, with the current set to constant values of 50 or 200 mA/cm². Cyclic voltammetry was performed using the CV mode on the CHI760 electrochemical workstation, with the voltage range set from 0.05 to 1.25 V vs SCE. The mass activities and ECSA-based specific activities of the catalysts were calculated based on the catalyst mass and electrochemical active surface areas (ECSAs) on the glassy carbon electrodes, respectively. The catalysts solutions were obtained by mixing 4.0 mg IrO₂ catalyst in a solution of 800 µL isopropanol, 200 µL deionized water and 40 µL Nafion solution (5 wt%), with sonication to form the homogenous catalyst ink. Then, 10 µL of the ink was dispersed on the GCE with drying naturally for testing. In the stability test, Ti felt (area: 1×1 cm²) was used as the working electrode, and 0.1M HClO₄ saturated with O₂ was used as the electrolyte. The homogeneous catalyst ink was formed by ultrasonic mixing of 5.0 mg catalyst in 500 µL isopropanol, 500 µL deionized water and 10 µL Nafion solution (5 wt%). 200 µL of ink was dispersed on the Ti felt.

For the PEM experiments, the anode catalyst ink was prepared by mixing 10 mg of IrO₂ catalyst with 108 µL of deionized water, 1327 µL of isopropanol, and 44.4 µL of 5 wt% Nafion solution. The cathode catalyst ink was prepared by mixing 10 mg of Pt/C catalyst with 128 µL of deionized water, 1528 µL of isopropanol, and 95.3 µL of 5 wt% Nafion solution. Both inks were then thoroughly mixed using an ultrasonic bath and a cell disruptor. After stirring for over 24 hours, the catalyst ink was uniformly sprayed on the central 2.5×2.5 cm² area of the N115 proton exchange membrane at 90 ℃ using ultrasonic spraying machine. Carbon paper and Ti felt were pressed on the outside of the catalytic layer using a hot press at a pressure of 450 kg and a temperature of 140 ℃ for ten minutes. The anode and cathode loadings of MEAs were calibrated using X-ray fluorescence (XRF) spectroscopy.

After the MEA was prepared, PTFE sealing materials and Ti bipolar plates were assembled around the MEA to form an electrolyzer. For all PEM tests, deionized water was used as the electrolyte at a flow rate of 18.2 ml min^− 1, with a working temperature of 80°C, and voltage was supplied by an NGI programmable DC power source. The I-V curve was measured from 0 to 3 A cm⁻², with 25 current points, each held for 30 seconds. Steady-state tests were conducted using a cross-flow configuration at 1.5 A cm⁻². Non-steady-state tests were conducted using a square wave pattern with voltages of 1.6 V and 2.3 V, each held for 1 minute. The XRD, XPS, and TEM results of IrO₂(101) monolayer after the stability test were obtained by analyzing the used MEAs.

The ECSAs of the IrO₂-NS were estimated from double layer capacitance (C_DL) and specific capacitance (C_S) using the following equation:

$$\:\text{ECSA=}{\text{C}}_{\text{DL}}/{\text{C}}_{\text{S}}$$

The electrochemical double-layer capacitance (C_DL) at non-Faradic potential range was obtained by measuring the capacitance of double layer at solid-liquid interface employing cyclic voltammetry (CV) with different scan rates (5, 10, 20, 30, 40, 50, 60 and 70 mV s^− 1) in a range from 1.25 to 1.35 V vs. RHE.

The C_S value was found to be 0.036 mF cm^− 2 in 0.1 M HClO₄ electrolyte using a polished glassy carbon-rotating disk electrode from the CV curves in Supplementary Fig. 18. The C_DL values of IrO₂(101) monolayer, IrO₂ NP, C-IrO₂ and C-Ir black were estimated from the CV curves in Supplementary Figs. 18 and 19.

DFT calculations

Density functional theory (DFT) calculations were conducted using the Vienna Ab initio Simulation Package (VASP)^{62, 63, 64, 65}. The interactions between ion cores and valence electrons were described using the projector augmented wave (PAW) method^{66, 67}. For the exchange-correlation functional, we employed the generalized gradient approximation (GGA) with the Perdew-Burke-Ernzerhof (PBE) formulation⁶⁸. The wave functions at each k-point were expanded using a plane wave basis set, with the kinetic energy cutoff set to 500 eV. Geometric optimizations were performed using a force-based conjugate gradient method until the energy converged to 1.0×10^− 5 eV per atom, with the maximum force on each atom less than 0.05 eV Å⁻¹. To account for long-range van der Waals interactions, Grimme’s D3 dispersion correction was implemented^{69, 70}.

The crystal structures of rutile-IrO₂ were used to construct slab models for the electrocatalytic study. Periodic slab models of the IrO₂(110), (200), and (101) facets were constructed using 2×1, 2×2, and 1×2 surface supercells, respectively, each comprising four metal-oxide layers. The bottom layer was constrained to maintain bulk-like properties. A vacuum space of 15 Å was introduced between the slab and its periodic image to minimize inter-slab interactions. To evaluate the activity of the electrocatalysts, we estimated the free energy diagram using the following Eq. 7¹:

$$\:\text{ΔG=ΔE+ΔZPE-TΔS}$$

where ΔG is the change in Gibbs free energy, ΔE represents the total energy change derived from DFT calculations, ΔZPE is the change in zero-point energy, ΔS denotes the entropy change, and T is the temperature (298.15 K).

Competing interests

The authors declare no competing financial interests.

Additional information

Supplementary Information is available for this paper.

Reprints and permissions information is available.

Correspondence and requests for materials should be addressed to D.Y. or J.L.

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Author contributions

J.L. led the whole project. D.Y. and C.Z. contributed equally to this work. D.Y. guided synthetic process, performed data analysis and wrote the manuscript. C.Z. synthesized the catalysts, performed TEM, XRD, XPS, Raman, ICP and electrochemical measures. A.T. performed DFT calculation and analysis. S.Z. performed XAFS measurements and analysis.

Acknowledgements

This work was supported by the National Key R&D Plan of China (2021YFB4000101), Interdisciplinary Innovation Program of North China Electric Power University, the National Natural Science Foundation of China (22205061), Hebei Province Outstanding Youth Fund (B2024502007), the Double First-class University Construction Project of North China Electric Power University (XM2412302) and the Fundamental Research Funds for the Central Universities (JB2024087).

Data availability.

The data that support the plots within this paper and other finding of this study are available from the corresponding author upon reasonable request.

Received: ((will be filled in by the editorial staff))

Published online: ((will be filled in by the editorial staff))

Liu R-T et al (2023) Recent advances in proton exchange membrane water electrolysis. Chem Soc Rev 52:5652–5683
Song J et al (2020) A review on fundamentals for designing oxygen evolution electrocatalysts. Chem Soc Rev 49:2196–2214
Suen N-T, Hung S-F, Quan Q, Zhang N, Xu Y-J, Chen HM (2017) Electrocatalysis for the oxygen evolution reaction: recent development and future perspectives. Chem Soc Rev 46:337–365
Chatenet M et al (2022) Water electrolysis: from textbook knowledge to the latest scientific strategies and industrial developments. Chem Soc Rev 51:4583–4762
Li R et al (2021) IrW nanochannel support enabling ultrastable electrocatalytic oxygen evolution at 2 A cm^– 2 in acidic media. Nat Commun 12
Tao HB et al (2024) The gap between academic research on proton exchange membrane water electrolysers and industrial demands. Nat Nanotechnol 19:1074–1076
Shi Q, Zhu C, Du D, Lin Y (2019) Robust noble metal-based electrocatalysts for oxygen evolution reaction. Chem Soc Rev 48:3181–3192
Xu H et al (2023) Iridium-based electrocatalysts for the acidic oxygen evolution reaction: engineering strategies to enhance the activity and stability. Mater Chem Front 7:1248–1267
Li G, Li S, Xiao M, Ge J, Liu C, Xing W (2017) Nanoporous IrO₂ catalyst with enhanced activity and durability for water oxidation owing to its micro/mesoporous structure. Nanoscale 9:9291–9298
Lim J et al (2017) Ultrathin IrO₂ Nanoneedles for Electrochemical Water Oxidation. Adv Funct Mater 28
Zhu J-Y et al (2020) Iridium Nanotubes as Bifunctional Electrocatalysts for Oxygen Evolution and Nitrate Reduction Reactions. ACS Appl Mater Interfaces 12:14064–14070
Dang Q et al (2021) Iridium metallene oxide for acidic oxygen evolution catalysis. Nat Commun 12
Jiang B et al (2018) Mesoporous Metallic Iridium Nanosheets. J Am Chem Soc 140:12434–12441
Liao F et al (2023) Iridium oxide nanoribbons with metastable monoclinic phase for highly efficient electrocatalytic oxygen evolution. Nat Commun 14
Hua K et al (2023) Catalytic Activity of Nanometer-Sized Ir–O_x Catalysts with Different Coordination Numbers for Electrocatalytic Oxygen Evolution. ACS Appl Nano Mater 7:487–497
Hua K et al (2024) Integrating Atomically Dispersed Ir Sites in MnCo₂O_4.5 for Highly Stable Acidic Oxygen Evolution Reaction. ACS Catal 14:3712–3724
Zhu Y et al (2022) Iridium single atoms incorporated in Co₃O₄ efficiently catalyze the oxygen evolution in acidic conditions. Nat Commun 13
Ouimet RJ, Glenn JR, De Porcellinis D, Motz AR, Carmo M, Ayers KE (2022) The Role of Electrocatalysts in the Development of Gigawatt-Scale PEM Electrolyzers. ACS Catal 12:6159–6171
Lee Y, Suntivich J, May KJ, Perry EE, Shao-Horn Y (2012) Synthesis and Activities of Rutile IrO₂ and RuO₂ Nanoparticles for Oxygen Evolution in Acid and Alkaline Solutions. J Phys Chem Lett 3:399–404
Stoerzinger KA, Qiao L, Biegalski MD, Shao-Horn Y (2014) Orientation-Dependent Oxygen Evolution Activities of Rutile IrO₂and RuO₂. J Phys Chem Lett 5:1636–1641
Zagalskaya A, Alexandrov V (2020) Role of Defects in the Interplay between Adsorbate Evolving and Lattice Oxygen Mechanisms of the Oxygen Evolution Reaction in RuO₂ and IrO₂. ACS Catal 10:3650–3657
Yang L et al (2023) A Highly Active, Long-Lived Oxygen Evolution Electrocatalyst Derived from Open‐Framework Iridates. Adv Mater 35
Kim K-S et al (2022) Promoting Oxygen Evolution Reaction Induced by Synergetic Geometric and Electronic Effects of IrCo Thin-Film Electrocatalysts. ACS Catal 12:6334–6344
Li M et al (2021) Exclusive Strain Effect Boosts Overall Water Splitting in PdCu/Ir Core/Shell Nanocrystals. Angew Chem Int Ed 60:8243–8250
Sun W, Song Y, Gong X-Q, Cao L-m, Yang J (2015) An efficiently tuned d-orbital occupation of IrO₂ by doping with Cu for enhancing the oxygen evolution reaction activity. Chem Sci 6:4993–4999
Zhuang L et al (2021) Porous Structure Engineering of Iridium Oxide Nanoclusters on Atomic Scale for Efficient pH-Universal Overall Water Splitting. Small 17
Hao S et al (2021) Torsion strained iridium oxide for efficient acidic water oxidation in proton exchange membrane electrolyzers. Nat Nanotechnol 16:1371–1377
Kwon S, Stoerzinger KA, Rao R, Qiao L, Goddard WA, Shao-Horn Y (2024) Facet-Dependent Oxygen Evolution Reaction Activity of IrO₂ from Quantum Mechanics and Experiments. J Am Chem Soc 146:11719–11725
Sanchez Casalongue HG, Ng ML, Kaya S, Friebel D, Ogasawara H, Nilsson A (2014) In Situ Observation of Surface Species on Iridium Oxide Nanoparticles during the Oxygen Evolution Reaction. Angew Chem Int Ed 53:7169–7172
Kasian O et al (2019) Degradation of iridium oxides via oxygen evolution from the lattice: correlating atomic scale structure with reaction mechanisms. Energy Environ Sci 12:3548–3555
Liao PC, Chen CS, Ho WS, Huang YS, Tiong KK (1997) Characterization of IrO₂ thin films by Raman spectroscopy. Thin Solid Films 301:7–11
Korotcov AV, Huang YS, Tiong KK, Tsai DS (2007) Raman scattering characterization of well-aligned RuO₂ and IrO₂ nanocrystals. J Raman Spectrosc 38:737–749
Wang C et al (2021) Engineering Lattice Oxygen Activation of Iridium Clusters Stabilized on Amorphous Bimetal Borides Array for Oxygen Evolution Reaction. Angew Chem Int Ed 60:27126–27134
Yu H et al (2024) Strain-Triggered Distinct Oxygen Evolution Reaction Pathway in Two-Dimensional Metastable Phase IrO₂ via CeO₂ Loading. J Am Chem Soc 146:20251–20262
Meng G et al (2019) Strain Regulation to Optimize the Acidic Water Oxidation Performance of Atomic-Layer IrO_x. Adv Mater 31
Li L, Xia Y, Zeng M, Fu L (2022) Facet engineering of ultrathin two-dimensional materials. Chem Soc Rev 51:7327–7343
Retuerto M et al (2022) Highly active and stable OER electrocatalysts derived from Sr₂MIrO₆ for proton exchange membrane water electrolyzers. Nat Commun 13
Liang J et al (2024) Efficient bubble/precipitate traffic enables stable seawater reduction electrocatalysis at industrial-level current densities. Nat Commun 15
Li H et al (2018) Systematic design of superaerophobic nanotube-array electrode comprised of transition-metal sulfides for overall water splitting. Nat Commun 9
Dang Q et al (2021) Iridium metallene oxide for acidic oxygen evolution catalysis. Nat Commun 12:6007
Li R et al (2021) IrW nanochannel support enabling ultrastable electrocatalytic oxygen evolution at 2 A cm^-2 in acidic media. Nat Commun 12:3540
Loncar A, Escalera-Lopez D, Cherevko S, Hodnik N (2022) Inter-relationships between Oxygen Evolution and Iridium Dissolution Mechanisms. Angew Chem Int Ed Engl 61:e202114437
Suen NT, Hung SF, Quan Q, Zhang N, Xu YJ, Chen HM (2017) Electrocatalysis for the oxygen evolution reaction: recent development and future perspectives. Chem Soc Rev 46:337–365
Knöppel J et al (2021) On the limitations in assessing stability of oxygen evolution catalysts using aqueous model electrochemical cells. Nat Commun 12
Geiger S et al (2018) The stability number as a metric for electrocatalyst stability benchmarking. Nat Catal 1:508–515
Pham CV et al (2020) IrO₂ coated TiO₂ core-shell microparticles advance performance of low loading proton exchange membrane water electrolyzers. Appl Catal B 269:118762
Wu Q et al (2023) Advances and status of anode catalysts for proton exchange membrane water electrolysis technology. Mater Chem Front 7:1025–1045
Spori C, Kwan JTH, Bonakdarpour A, Wilkinson DP, Strasser P (2017) The Stability Challenges of Oxygen Evolving Catalysts: Towards a Common Fundamental Understanding and Mitigation of Catalyst Degradation. Angew Chem Int Ed Engl 56:5994–6021
DOE U. DOE Technical Targets for Hydrogen Production from Electrolysis.) (2018)
Stiber S et al (2022) A high-performance, durable and low-cost proton exchange membrane electrolyser with stainless steel components. Energy Environ Sci 15:109–122
Oh HS, Nong HN, Reier T, Gliech M, Strasser P (2015) Oxide-supported Ir nanodendrites with high activity and durability for the oxygen evolution reaction in acid PEM water electrolyzers. Chem Sci 6:3321–3328
Hegge F et al (2020) Efficient and Stable Low Iridium Loaded Anodes for PEM Water Electrolysis Made Possible by Nanofiber Interlayers. ACS Appl Energy Mater 3:8276–8284
Rakousky C et al (2017) Polymer electrolyte membrane water electrolysis: Restraining degradation in the presence of fluctuating power. J Power Sources 342:38–47
Honsho Y, Nagayama M, Matsuda J, Ito K, Sasaki K, Hayashi A (2023) Durability of PEM water electrolyzer against wind power voltage fluctuation. J Power Sources 564:232826
Siracusano S, Trocino S, Briguglio N, Pantò F, Aricò AS (2020) Analysis of performance degradation during steady-state and load-thermal cycles of proton exchange membrane water electrolysis cells. J Power Sources 468:228390
Zhuang L et al (2021) Porous Structure Engineering of Iridium Oxide Nanoclusters on Atomic Scale for Efficient pH-Universal Overall Water Splitting. Small 17:2100121
Puthiyapura VK, Pasupathi S, Su H, Liu X, Pollet BG, Scott K (2014) Investigation of supported IrO₂ as electrocatalyst for the oxygen evolution reaction in proton exchange membrane water electrolyser. Int J Hydrog Energy 39:1905–1913
Retuerto M et al (2022) Highly active and stable OER electrocatalysts derived from Sr₂MIrO₆ for proton exchange membrane water electrolyzers. Nat Commun 13:7935
Hu C et al (2023) Misoriented high-entropy iridium ruthenium oxide for acidic water splitting. Sci Adv 9:eadf9144
Wu H et al (2022) Recent developments of iridium-based catalysts for the oxygen evolution reaction in acidic water electrolysis. J Mater Chem A 10:13170–13189
Cruz JC et al (2011) Nanosized IrO₂ electrocatalysts for oxygen evolution reaction in an SPE electrolyzer. J Nanopart Res
Kresse G, FurthmГјller J (1996) Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set. Comput Mater Sci 6:15–50
Kresse G, Hafner J (1993) Ab initio molecular dynamics for liquid metals. Phys Rev B Condens Matter 47:558–561
Kresse G, Hafner J (1994) Ab initio molecular-dynamics simulation of the liquid-metal-amorphous-semiconductor transition in germanium. Phys Rev B Condens Matter 49:14251–14269
Kresse G, FurthmГјller J (1996) Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set. Phys Rev B 54:11169
BlГ¶chl PE (1994) Projector augmented-wave method. Phys Rev B 50:17953
Kresse G, Joubert D (1999) From ultrasoft pseudopotentials to the projector augmented-wave method. Phys Rev B 59:1758
Perdew JP, Burke K, Ernzerhof M (1996) Generalized gradient approximation made simple. Phys Rev Lett 77:3865
Grimme S, Antony J, Ehrlich S, Krieg H (2010) A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu. J Chem Phys 132
Grimme S, Ehrlich S, Goerigk L (2011) Effect of the damping function in dispersion corrected density functional theory. J Comput Chem 32:1456–1465
Man IC et al (2011) Universality in oxygen evolution electrocatalysis on oxide surfaces. ChemCatChem 3:1159–1165

There is NO Competing Interest.

SupportingInformation.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

2D single-faceted IrO2(101) monolayer enabling high-performing proton exchange membrane water electrolysis beyond 8,000 h stability at 1.5 A cm-2

Status:

Version 1

Abstract

Figures

1 Introduction

2 DFT-guided facet selection

3 Characterization of 2D IrO(101) monolayer

4 The synthesis of IrO(101) monolayer

5 OER performance in three-electrode cell

6 OER performance in PEM cell

Conclusion

Methods

Declarations

Competing interests

Additional information

Author contributions

Acknowledgements

Data availability.

References

Additional Declarations

Supplementary Files

Status:

Version 1