nature communications Article https://doi.org/10.1038/s41467-022-32024-6 Sustainable oxygen evolution electro- catalysis in aqueous 1 M H2SO4 with earth abundant nanostructured Co3O4 Received: 1 February 2022 Jiahao Yu1,2, Felipe A. Garcés-Pineda 1, Jesús González-Cobos 1,9 , Marina Peña-Díaz 3, Celia Rogero 3,10, Sixto Giménez4, Accepted: 13 July 2022 Maria Chiara Spadaro 5, Jordi Arbiol 5,6, Sara Barja 7,8 & José Ramón Galán-Mascarós 1,6 Check for updates 1234567890():,; 1234567890():,; Earth-abundant electrocatalysts for the oxygen evolution reaction (OER) able to work in acidic working conditions are elusive. While many first-row transi- tion metal oxides are competitive in alkaline media, most of them just dissolve or become inactive at high proton concentrations where hydrogen evolution is preferred. Only noble-metal catalysts, such as IrO2, are fast and stable enough in acidic media. Herein, we report the excellent activity and long-term stability of Co3O4-based anodes in 1 M H2SO4 (pH 0.1) when processed in a partially hydrophobic carbon-based protecting matrix. These Co3O4@C composites reliably drive O2 evolution a 10 mA cm–2 current density for >40 h without appearance of performance fatigue, successfully passing benchmarking pro- tocols without incorporating noble metals. Our strategy opens an alternative venue towards fast, energy efficient acid-media water oxidation electrodes. Hydrogen is considered the most environmentally friendly alternative Acidic electrolytes offer many advantages for hydrogen pro- fuel to replace traditional fossil energy1,2. However, hydrogen pro- duction, given that high H+ concentration improves the hydrogen duction is still dominated worldwide by natural gas steam reforming, a evolution reaction (HER), and also facilitates fast/low resistance ionic mature technology fixing a very low competitive price in the market. transport9,10. Several electrocatalysts based on low-cost raw materials Electrochemical water splitting powered by renewable sources is have shown great promise to substitute the state-of-the-art platinum regarded as the ideal future technology to produce hydrogen, but electrodes responsible for HER6,7,11–14. On the contrary, no viable costs must be lowered to improve its market penetration3,4. The cat- candidates are known to substitute noble-metal oxides such as IrO2 alysts responsible to improve the efficiency of the process, mainly at the anode where the oxygen evolution reaction (OER) takes relying on noble metals, are part of the high cost of the technology, place1,6,9,15–21. So far, no stable and inexpensive OER catalysts can work and because of this, many laboratories in the world are working to find under high potential and/or current densities in acidic media, where viable solutions to develop effective, earth-abundant, robust and even the highly active RuO2 presents serious deactivation scalable catalyst5–8. problems22,23. 1 Institute of Chemical Research of Catalonia (ICIQ), The Barcelona Institute of Science and Technology (BIST), Avenida Països Catalans 16, 43007 Tarragona, Spain. 2Departament de Química Física i Inorgànica, Universitat Rovira i Virgili, Marcel. lí Domingo 1, 43007 Tarragona, Spain. 3Centro de Física de Materiales, CFM/MPC, (UPV/EHU-CSIC), 20018 San Sebastián, Spain. 4Institute of Advanced Materials (INAM), Universitat Jaume I, 12006 Castelló, Spain. 5 Catalan Institute of Nanoscience and Nanotechnology (ICN2), CSIC and BIST, Campus UAB, Bellaterra, 08193 Barcelona, Catalonia, Spain. 6ICREA, Passeig Lluis Companys, 23, 08010 Barcelona, Spain. 7Departamento de Polímeros y Materiales Avanzados: Física, Química y Tecnología, Centro de Física de Materiales, University of the Basque Country UPV/EHU, 20018 San Sebastián, Spain. 8Donostia International Physics Center, 20018 San Sebastián, Spain. 9 Present address: Institut de Recherches sur la Catalyse et l’Environnement de Lyon, UMR 5256, CNRS, Université Claude Bernard Lyon 1, 2 Avenue A. Einstein, 69626 Villeurbanne, France. 10Present address: Donostia International Physics Center, 20018 San Sebastián, Spain. e-mail: sara.barja@ehu.eus; jrgala- n@iciq.es Nature Communications | (2022)13:4341 1 Article https://doi.org/10.1038/s41467-022-32024-6 Several strategies have been proposed to promote OER at high Results and discussion proton concentration24–28. One strategy deals with the investigation Catalyst and electrode preparation of ternary/complex oxide structures such as nickel-manganese The overall synthetic and processing protocol is summarized in Fig. 1. antimonate. This rutile-type oxide was stable for 168 h at We synthesized carbon-coated Co (Co@C) nanoparticles, starting 10 mA cm–2 operating in 1 M sulfuric acid, although with the penalty from the thermal treatment of Co(bIm)2 (bIm = 2-benzimidazolate), a of requiring a large overpotential (η ≥ 700 mV)26. Cobalt-doped metal-organic framework (MOF) precursor (ZIF-9)37. Then Co@C was hematite thin-film electrocatalysts were also able to sustain a oxidized at low-temperature to achieve its full transformation into geometric current density of 10 mA cm–2 for up to 50 h at pH 0.1, cobalt oxide nanoparticles, covered by an amorphous, nitrogen- but again at large overpotentials (η ≥ 650 mV)27. Indeed, cobalt doped-carbon coating derived from the organic skeleton (Co3O4@C, oxides have shown promising OER electrocatalysis in acidic media, Fig. 1). Powder X-ray diffraction (PXRD) patterns and Raman spectra although highly limited by its redox potential-dependent instabil- confirmed the presence of a Co3O4 phase and the carbon support ity, since CoOx may rapidly dissolve either at open circuit condi- (Supplementary Figs. 1–4). High-resolution transmission electron tions or under high applied potentials29–32. Some interesting microscopy the presence of graphitic-like nanostructures all around advances in long-term stability were achieved by doping or pro- the sample, embedding the Co3O4 nanoparticles. Some of these cessing techniques33–35. C-nanostructures had a nanosheet-like morphology (Fig. 2), while In our previous work with polyoxometalate (POM)-supported some others were folded forming onion-like rings around the Co3O4 catalysts, we disclosed how these molecular catalysts showed pro- nanoparticles (Supplementary Fig. 5). Electron energy loss spectro- mising OER catalytic performance in acidic conditions when scopy in scanning TEM mode (EELS-STEM) confirmed the chemical incorporating active Co centers36. The high activity of these cata- composition of the nanoparticles and surrounding nanostructures lysts, even in heterogeneous conditions, was achieved thanks to the (Supplementary Fig. 6). The Co3O4@C composition was determined as synergic stability offered by a partially hydrophobic carbon-based (Co3O4)(H2O)0.30(OH)0.85C2.00N0.05 by thermogravimetry elemental support. However, this strategy was not successful to achieve long- analysis (Supplementary Fig. 7 and Supplementary Table 1). term stability, since these POM-based electrodes could only survive For the preparation of the electrode composites, Co3O4@C was at low current densities and for a very limited time, given their mixed with graphite (G) and paraffin oil (PO) in the desired ratio (see intrinsic instability to the mechanical stress provoked by gas Methods section) to prepare a homogeneous composite (Co3O4@C/ bubbling. GPO) with the desired Co3O4@C content up to 40% (40-Co3O4@C). In this work, we present a promising processing protocol, which Composites above 40% were mechanically too fragile for further pro- combines in one single anode two powerful strategies: (i) the incor- cessing into the working electrode pocket. HRTEM images and EELS- poration of a nanostructured OER catalyst from earth-abundant metals STEM maps showed similar nanostructures within Co3O4@C/GPO and to maximize active surface area, (ii) supported by a conducting, par- close contact between Co3O4@C and GPO (Fig. 2 and Supplementary tially hydrophobic binder made from paraffin oil and graphite powder. Fig. 8). X-ray photoemission spectroscopy (XPS) analysis was Our processing protocol with nitrogen-doped carbon-coated Co3O4 employed to further identify the surface chemical composition and the nanoparticles (Co3O4@C) delivers robust and scalable anodes that mixed oxidation state Co2+/3+ (Supplementary Fig. 9) consistent with exhibit excellent acidic OER performances, needing a minimum over- the presence of the Co3O4, as confirmed by PXRD and HR-TEM data38. potential (η ≤ 398 mV) to maintain a 10 mA cm–2 current density for XPS spectra from the Co3O4@C/GPO composite show no differences >40 h when working in 1 M sulfuric acid solution, without any sign of respect to the Co3O4@C precursor, demonstrating the absence of fatigue or deactivation. The high activity and also great stability chemical modification during composite preparation. demonstrate a performance superior to any other non-noble catalysts reported. Oxygen evolution quantification confirms the Faradaic effi- OER electrocatalytic activity in 1 M H2SO4 ciency (>96%) of these electrodes towards OER, with negligible parti- The x-Co3O4@C/GPO composites (x corresponds to the % in weight for cipation of other oxidation processes. These results open alternative the metal oxide) were inserted into the pocket of a working electrode opportunities for stable OER electrocatalysis with earth-abundant raw and used as anode during electrochemical water oxidation in 1 M materials. H2SO4 (pH ≈ 0.1). The cyclic voltammetry (CV) showed the appearance Fig. 1 | Electrode preparation. Scheme of the synthesis and processing protocol to obtain a Co3O4@C/GPO electrode. Nature Communications | (2022)13:4341 2 Article https://doi.org/10.1038/s41467-022-32024-6 Fig. 2 | Structural and morphological characterization of Co3O4@C/GPO graphite layers with a ≈ 0.34 nm spacing and here oriented along the [–1–12–3] composites. HAADF STEM (a) and HRTEM (b) micrographs from a fresh Co3O4@C/ (turquoise circles). To highlight the different systems, we show a frequency filtered GPO composite. The power spectrum analysis confirms the presence of Co3O4 map (on the bottom right) where the Co3O4 corresponding lattice fringes are in nanoparticles with a cubic FD3-MZ (s.g. 227) spinel structure, oriented along its yellow and the graphite layer ones in turquoise . Source data are provided as a [011] axis (yellow circles) superimposed to the reflections corresponding to Source Data file. Fig. 3 | Electrocatalytic activity of Co3O4@C/GPO electrodes in 1 M H2SO4 H2SO4 (pH 0.1), at x = 5 (pink), 10 (gray), 20 (orange), 30 (blue), 40 (black) electrolyte. a LSV curves for x-Co3O4@C/GPO electrodes in 1 M H2SO4 (pH 0.1), compared to 40-Co3O4@C/GPO (red); c Tafel plots of IrO2/GPO (black) and at x = 0 (GPO blank, light blue), 5 (blue), 10 (yellow), 20 (grey), 30 (pink), 40 Co3O4@C/GPO (red) extracted from LSV data. Source data are provided as a (red) and with 1 mV s–1 scan rate; b LSV curves of x-IrO2/GPO electrodes in 1 M Source Data file. of a catalytic current density on the Co3O4@C/GPO electrode at rela- (Supplementary Fig. 13). This indicates a greater electrochemical active tively low overpotentials, which was sustained after successive cycling surface area for Co3O4@C/GPO, due to its higher density of active sites curves (Supplementary Figs. 10 and 11). Comparative linear sweep in Co3O4@C/GPO, thanks to its nanostructuration, favouring the voltammetry (LSV) showed an enhanced electrochemical activity upon higher current densities observed in the potential range studied41,42. It increasing Co3O4@C content, reaching a very low onset overpotential is important to point out that nanostructured IrO2 can perform sig- (ηonset = 194 ± 4 mV) for the 40-Co3O4@C/GPO electrode (Fig. 3 and nificantly better than our IrO2/GPO electrodes20,43,44. This must be Supplementary Fig. 12). These electrodes reach 10 mA cm–2 currents at related to the relatively low electrochemical surface area (ECSA) of our just 360 ± 4 mV overpotential. Interestingly, no sign of a transport- IrO2/GPO composite. When we normalize current densities vs. ECSA limited regime appeared in the studied potential range, reaching over (Supplementary Fig. 14), IrO2/GPO electrodes reach higher values at 20 mA cm–2 at η = 397 ± 4 mV. A current density limit of 100 mA cm−2 lower potentials. Still, the Co3O4@C catalyst remains competitive as a was found in these x-Co3O4@C/GPO electrodes. non-noble-metal catalysts, and especially when current density is We prepared analogous IrO2/GPO working electrodes to normalized per gram (Supplementary Fig. 14), as a more practical benchmark our results in the same conditions with the state-of-the- parameter for applications. art IrO2. The IrO2/GPO anodes delivered higher overpotentials, Finally, we measured anodic oxygen evolution during chron- ηj = 10mAcm2 = 368 mV at 10 mA cm–2 and ηj = 20mAcm2 = 396 mV at opotentiometry experiments with Co3O4@C/GPO electrodes (Sup- 20 mA cm–2, slightly above those obtained for the Co3O4@C-based plementary Fig. 15). We found over >96% Faradaic efficiency, electrode (Supplementary Table 2). confirming that OER is the dominant process at these electrodes’ Tafel analyses of the LSV data yielded slopes of 143 mV dec–1 for surface, and confirming no significant oxidation of the carbon-based Co3O4@C and 82 mV dec–1 for IrO2 (Fig. 3c), suggesting a different matrix is taking place in these conditions. reaction mechanism (rate-limiting step) for these two catalysts, and indicating a faster increment of current density with the applied OER electrocatalytic stability in 1 M H2SO4 potential for IrO239,40. Interestingly, this is compensated by the lower As mentioned before, stability is a critical issue for earth-abundant OER onset potential of Co3O4@C/GPO. The electrochemical double-layer catalysts in acidic media45–49. To determine the stability of our capacitance (EDLC) of Co3O4@C/GPO and IrO2/GPO were calculated as Co3O4@C/GPO electrodes, we take advantage of the benchmarking 25 and 2 mF cm–2, respectively, with 0.03 mF cm–2 for the blank GPO protocol designed by Jaramillo’s group that uses as figure of merit the Nature Communications | (2022)13:4341 3 Article https://doi.org/10.1038/s41467-022-32024-6 Fig. 4 | Electrochemical stability of Co3O4@C/GPO electrodes. Chron- other OER electrocatalysts (black squares) in the same electrolysis conditions opotentiometry measurements at j = 10 mA cm–2 in 1 M H2SO4 electrolyte. a 40- according to the data from a previous benchmarking study6. Source data are pro- Co3O4@C/GPO (>40 h). b 20-Co3O4@C/GPO (2 h). c 30-Co3O4@C/GPO (2 h). vided as a Source Data file. d Benchmarking of their activity/stability features (red stars) in comparison with overpotential required to achieve and maintain a 10 mA cm–2 current Table 3). We found the presence of Co but at the ppb level, corre- density, as this is approximately the current density expected at the sponding to just ≈0.4 % of the total. Based on this number, we can anode in a 10% efficient solar water-splitting device under 1 sun estimate a 25 S-number, an ASF of 101 and a lifetime of 462 h. These illumination6,50. The corresponding chronopotentiometry data estimations are comparable even to Ir-based catalysts such as SrIrO3 (Fig. 4a–c) show very good stability for all electrodes, independently of in analogous conditions, and confirm the promising performance/ their Co3O4@C content. In all cases, ηj = 10mAcm2 after 2 h shows just a stability of these electrodes. It is worthy to mention that this small small increment. In the case of our best electrodes, the 40-Co3O4@C/ Co loss does not significantly affect performance. This may be due GPO, this increment is of just 3 mV, and the stability is maintained for to Co leaching only from non-catalytically areas, or to the 3D nature long times. After 43 h of continuous electrolysis, the overpotential is of our electrodes. Thus, after leaching, the new surface exposed essentially identical to the starting value (Fig. 4a). Meanwhile, even to brings additional active sites, keeping the performance. Both may keep 100 mA cm–2 current density, only 27 mV increment was needed explain why activity is so stable, but a relatively modest S-number is after 2 h catalysis (Supplementary Fig. 11b). We also applied more obtained. aggressive stability tests under cycling applied potentials. In this accelerated degradation testing (ADT), complete CV cycles were col- Post-electrolysis Co3O4@C/GPO characterization lected, thus exposing the electrodes sequentially to catalytic and non- To further confirm the stability of Co3O4@C as a genuine OER catalyst, catalytic potentials. These conditions simulate better the different we characterized the structural and chemical evolution of the elec- degradation mechanisms during electrode and electrolyzer operation. trodes after these 2 h electrolysis at 10 mA cm–2 in 1 M H2SO4. The The 40-Co3O4@C/GPO electrodes showed also good stability after powder XRD patterns did not show any significant change nor shift in ~600 CV cycles with only a 14 mV increment in the overpotential to the observed peaks, still typical of Co3O4@C and graphite (Supple- reach 10 mA cm–2 current density (Supplementary Fig. 11c). mentary Fig. 16). This suggests no major structural changes are The benchmarking of these electrodes with previous literature is occurring to the bulk of the material Co3O4@C. highlighted in the ηj = 10mAcm2 ,t = 2h vs. ηj = 10mAcm2 ,t = 0h plot (Fig. 4d). We explored potential changes on the chemical composition of This comparative plot illustrates the high activity and stability of our the catalyst due to OER process by XPS characterization of the fresh electrodes. The three of them appear at the diagonal of the graph, as electrode and after water electrolysis under different conditions. The expected for sustainable performance, and very close and competitive Co 2p XPS spectrum of the fresh Co3O4@C/GPO electrode (Fig. 5a) to the results obtained with noble-metal counterparts, indicating that shows two peaks located at 794.9 eV (Co 2p 1/2) and 779.7 eV (Co 2p3/2), earth-abundant anodes may successfully pass this benchmarking corresponding to the spin–orbit splitting of the 2p orbital. Both protocol for OER performance in acidic media. components contain equivalent chemical information. The deconvo- Stability number (S-number) and activity-stability factor (ASF) luted analysis of the peaks reveals the presence of two different che- were also proposed as key metrics for estimating lifetime and long- mical components, which we attribute to the Co3+ (blue) and Co2+ term stability for electrocatalysts51–53. Thus, we analyzed the elec- (green) states, in agreement with the presence of Co3O438. In addition, trolyte after stability tests to check for Co leaching (Supplementary we observe two doubled satellite peaks arising from charge transfer Nature Communications | (2022)13:4341 4 Article https://doi.org/10.1038/s41467-022-32024-6 Fig. 5 | Chemical analysis of the Co3O4@C/GPO catalyst after water electrolysis. GPO electrode after OER chronopotentiometry at 5 mA cm–2 for 24 h. Color a X-ray photoelectron spectra of the Co 2p (left) and O 1s (right) core levels for a 20- assignment for the different areas: Co3+ (blue), Co3+ satellite (pink), Co2+ (green), Co3O4@C/GPO electrode. b Same spectra for a 20-Co3O4@C/GPO electrode after Co2+ satellite (yellow), OH/H2O (purple), O–Co (brown). Source data are provided as OER chronopotentiometry at 10 mA cm–2 for 2 h. c Same spectra for a 20-Co3O4@C/ a Source Data file. and final states effects from Co2+ (satellite A, yellow) and Co3+ (satellite Critical role of GPO B, pink)54, again characteristic of Co3O4. We also analyzed the O 1 s To investigate the actual role of the carbon paste in the stability of the peak (Fig. 5a). In addition to the Co-O component (brown) related to electrodes, we carried out additional alternative experiments. First, we the Co3O4, we observe a higher binding energy component attributed directly deposited Co3O4@C on a glassy carbon (GC) electrode as a to residual OH/H2O (purple). Quantitative analysis of the Co 2p and O Nafion-based ink. This electrode showed a significantly lower electro- 1 s core levels of Co3O4@C/GPO after chronopotentiometry at catalytic activity when compared with the Co3O4@C/GPO (Figs. 3–4 vs. 10 mA cm–2 for 2 h (Fig. 5b) and 5 mA cm–2 for 24 h (Fig. 5c) revealed no Supplementary Fig. 18). More importantly, after 30-min of the shifts in the binding energies of the components respect to the fresh benchmarking test in 1 M H2SO4, the Co3O4@C/GC electrode is sample. Crucially, the main spectral features attributed to a Co3O4@C apparently deactivated. This suggests that the GPO binder is funda- catalyst remain unaltered after water electrolysis, which demonstrates mental to confer the acidic stability and activity of the Co3O4@C the preservation of the oxidation state of the catalyst. Changes in the component. intensity of the OH/H2O component on the O 1s can be fairly attributed This effect of the GPO binder could be due to a modified local pH to the different environmental conditions of the emersed electrode at the electrode/electrolyte interface55–57. To check this hypothesis, we (see Methods section). XPS analysis of the C component does not decided to investigate the effect of the GPO binder on the local pH evidence significant changes in the oxidation state of the C 1s peak, through the reversible H+/H2 pair as catalyzed with commercial Pt/C58. supporting the preservation of the carbon-based matrix (Supplemen- The reversible potential for this model reaction differs when the Pt/C is tary Fig. 17) in agreement to the obtained Faradaic efficiencies. Nitro- directly deposited on a graphite electrode, or when incorporated into gen detection in the system is below our resolution limit; therefore, no a GPO electrode as observed in their CV plots in a hydrogen saturated discussion is referred to this element. Based on the current analysis, 1 M H2SO4 electrolyte (Supplementary Fig. 19). An average value of the most important finding is that the oxide film is stable and no cobalt –0.002 V vs. ERHE was estimated for the Pt/C catalyst, in good agree- oxide is lost nor further oxidized during the electrolysis process. In ment with the theoretical +0.0 V value. A + 0.031 V vs. ERHE was found summary, XRD and XPS strongly support the bulk and surface stability for the (Pt/C)/GPO electrode. If we associate this potential difference of Co3O4@C before and after acidic OER electrocatalysis, and its only to a theoretical local pH difference, ΔE = 0.059 ΔpH, we can genuine catalytic activity. estimate a pH difference of just 0.52 unit between both electrodes. Of We also investigated the Co3O4@C/GPO composite after 2 h course, this number cannot be assigned to the actual pH at the surface, electrolysis at 10 mA cm–2 by means of HR-TEM (Fig. 6). The images and since this difference may be caused by pH and other parameters, power spectra (FFT) analyses also confirm a high structural and che- including Pt surface states. However, such small difference indicates mical stability. Neither crystallinity nor particle size are affected by the that the higher stability and activity found in GPO electrodes cannot electrochemical process. arise exclusively from the modification of the local pH. Therefore, we Nature Communications | (2022)13:4341 5 Article https://doi.org/10.1038/s41467-022-32024-6 Fig. 6 | Structural and morphological characterization of Co3O4@C/GPO elec- nanoparticles with a cubic FD3-MZ (s.g. 227) spinel structure (yellow cirlces) here trodes after water electrolysis in 1 M H2SO4 electrolyte. HRTEM micrograph (a) oriented along the [2–11] zone axis. No changes in crystallinity or particle size are from a Co3O4@C/GPO composite, recovered after 2 h water electrolysis at observed for the Co3O4@C composite when compared to the fresh samples (Fig. 2). 10 mA cm–2. To highlight the different systems, we show a frequency filtered map Notice that the graphitic layers may show multiple rotated domains as shown by the (b) where the Co3O4 nanoparticle lattice fringes are in yellow and the ones corre- highlighted dotted green circles in the power spectrum, but this effect also hap- sponding to the surrounding graphitic layers in pink and turquoise. The power pened in the as-prepared sample. Source data are provided as a Source Data file. spectra (FFT) confirm the high stability of these electrodes, showing again Co3O4 associate the protective function to the hydrophobic environment, Although previous reports on acidic water splitting with earth- which avoids proper solvation of the oxides, precluding its dissolution. abundant raw materials had achieved either high activity or high sta- The high hydrophobicity of GPO is corroborated by contact angle tests bility, our working anodes fulfill both requirements (Supplementary (Supplementary Fig. 20). A contact angle of 125° was found for a 1 M Table 4). These electrodes operate during more than 40 h, at a rela- H2SO4 solution on GPO, significantly higher than on GC (86°). tively high current density (10 mA cm–2) and at a low overpotential of To prove the general importance of this protective strategy, η < 398 mV, very close to the benchmarking performance of state-of- typical oxides based on Ni, Co, and Fe, which were active in alkaline the-art IrO2. solution but unstable in acid for oxygen evolution, were synthesized The synergy between active catalytic phase, Co3O4 nanoparticles, and then their electrochemical performances were measured in 1 M and the carbon support (doped-carbon cover, graphite and paraffin H2SO4 (Supplementary Fig. 21). It was showed that all of these oxides oil) is crucial to reach this robust performance. We assign this pro- were active for oxygen evolution and kept their performance at tective effect to the hydrophobicity of the electrode surface, which 10 mA cm–2 during 2 h catalysis with the small change of ≤13 mV could avoid proper solvation of the metal oxides precluding their (Supplementary Fig. 22). dissolutions. Indeed, we found preliminary data that supports this We also compared the activity/stability of Co3O4@C vs. Co3O4 approach to be general, also effective when applied to other non- (Supplementary Fig. 10). The corresponding x-Co3O4/GPO electrodes noble-metal-based OER catalysts for working in acid. showed good stability during preliminary CV cycles and chron- Despite the promising performance of our electrodes, a few opotentiometry measurement, but at higher overpotentials. A challenges will need to be addressed before their implementation into 5 mF cm–2 EDLC was determined, just 1/5 that of Co3O4@C/GPO commercial electrolyzers. To start with, the carbon content may (Supplementary Fig. 13). Specific surface area from N2 sorption iso- become an issue at high current densities, as those expected from therm curves for Co3O4@C was also about five times greater than that commercial devices (>500 mA cm–2). However, our successful of Co3O4 (Supplementary Fig. 23). These results suggest that the role corrosion-protection opens an interesting strategy that can be trans- of the carbon coating is to improve the nanostructuration of the active lated into full cell devices, looking for alternative approaches to Co3O4 material. incorporate hydrophobic species at the electrode surfaces. In addition, the Co3O4@C/GPO, Co3O4/GPO, and IrO2/GPO elec- In addition, it is worthy to mention that the current densities trodes were studied by Electrochemical Impedance Spectroscopy (EIS) reached and sustained by the Co3O4@C/GPO electrodes are high at different applied potentials. Supplementary Fig. 24 shows the enough to satisfy the needs of photoelectrochemical (PEC) devices, obtained Nyquist plots, which systematically feature two arcs (or dis- well in line with the maximum currents provided by photoanodes. torted arc for IrO2), consistent with two simultaneous/consecutive Investigations to combine these acid-stable electrodes as co-catalyst charge-transfer channels59. Fitting the experimental data to a suitable for photoactive semi-conductors are under way. equivalent circuit model (Supplementary Fig. 25) revealed that the best ohmic contact (reflected by the series resistance, RS) is obtained for Methods the Co3O4@C/GPO (Supplementary Fig. 25a). On the other hand, the Materials and chemicals charge transfer resistance (Rct), scales inversely with the electro- All the chemical reagents and solvents were of commercial grade and catalytic activity of the different electrodes, being the lowest one for used directly without any further purification Experimental Details. the Co3O4@C/GPO electrode (Supplementary Fig. 25b). This is con- sistent with the estimated surface capacitance, which scales with Synthesis electrode performance, as a result of higher surface area and hence, ZIF-9 was prepared via a solvothermal method according to the pre- higher density of catalytic sites (Supplementary Fig. 25c). vious literature with some modifications60,61. Co(NO3)2·6H2O (0.175 g) In summary, we are reporting the activity and promising stability and benzimidazole (0.142 g) were dissolved into 15 mL DMF and then of carbon-decorated Co3O4@C nanoparticles (Co3O4@C) for elec- the homogeneous solution was transferred into a Teflon-lined stainless trocatalytic OER under acidic conditions when protected by a autoclave. The sealed autoclave was put into an oven and kept at hydrophobic binder support. Electrodes built from Co3O4@C, gra- 140 °C for 24 h. When it was cooled down to room temperature, the phite and paraffin oil are able to evolve oxygen from water during the purple product was filtered out and washed with acetone, and then electrolysis of a concentrated (1 M) sulfuric acid solution (pH < 0.1). dried at 60 °C. Nature Communications | (2022)13:4341 6 Article https://doi.org/10.1038/s41467-022-32024-6 To obtain the target material, Co@C was firstly synthesized using open circuit potentials (OCPs) vs. the Ag/AgCl were firstly recorded for ZIF-9 as precursor by heating at 500 and then 750 °C for 2 h, respec- 30 min to reach rather stable values. Combined with above CV mea- tively, under Ar flow while Co3O4 was obtained under air. Afterwards, surements, the 100 mV potential windows centered at OCPs could be the pyrolysis product was oxidized in air at 230 °C for 48 h to generate determined and cyclic voltammetries were then carried out under Co3O4@C. scanning rates of 20, 40, 60, 80, and 100 mV s–1. The current density Mixed metal oxides (NiOx, FeOx, NiFeOx, CoFeOx, and NiCoOx) differences between the minimum and maximum values at OCPs vs. were prepared by modified method available in the literature62. Metal the Ag/AgCl and the corresponding scanning rates were plotted to nitrates with ratio 1:1 were dissolved in 50 mL of distilled water with calculate the EDLC value (1/2 of the slope of current density-scan rate constant stirring until a clear solution was obtained. The whole metal plots)64. Co3O4@C and commercial Pt/C from AlfaAesar were also concentration was fixed to 0.0125 M. Glycine was added into the aqu- deposited on the glassy carbon (GC, 0.07 cm−2) disk electrodes. The eous solution (glycine/metal molar ratio is 1.2) and stirred until total inks were prepared by sonicating 10 mg of catalyst, 25 μL Nafion 117 dissolution. Afterwards, the solution was heated up to 210 °C until total containing solution and 975 μL ethanol aqueous solution (3:1 in solvent evaporation and glycine combustion. The resulting porous dark volume) for 30 min. solid was recovered and calcined at 1100 °C in a tubular oven for 1 h. Electrochemical impedance spectroscopy (EIS) was performed by The composite electrodes were prepared by 2-h ball-milling at means of a typical three-electrode cell in the frequency range from 20 s−1 of a mixture of paraffin oil (20 mg), graphite powder (80 mg) and 100 kHz to 0.1 Hz with 8 points per decade. The AC perturbation was the desired weight of metal oxide (5, 10, 20, 30 or 40 mg), namely, 5 mV. Experimental data were fitted to the selected equivalent circuit x-Co3O4@C/GPO. x-IrO2/GPO (commercial IrO2 from AlfaAesar), model using Zview software (Scriber Associates) for extracting both x-Co3O4/GPO and other oxide electrodes were also prepared using the capacitances and resistances. same process for comparison purposes. Faradaic efficiency Structural characterization In order to evaluate the faradaic efficiency towards oxygen produc- Powder X-ray diffraction (PXRD) data were recorded with a Bruker D8 tion, the chronopotentiometric experiment (Supplementary Fig. 15b) Advance Series equipped with a VANTEC-1 PSD3 detector. Elemental was carried out applying a fixed current (1.4 mA, 2 h) while oxygen analyses were carried out with an Agilent 725-ES inductively coupled concentration in the headspace was in situ measured by using an plasma optical emission spectrometer (ICP-OES) at University of Val- Unisense sensing system equipped with an oxygen microsensor based ladolid (Co) and LECO CHNS-932 elementary microanalyzers (C, H, N) on voltage polarization. The experiment was performed under con- at Complutense University of Madrid. Thermogravimetric analysis was tinuous flow conditions by bubbling Ar as carrier gas in both anodic conducted with a thermogravimetric balance of Mettler Toledo. and cathodic compartments. For this purpose, an H-type cell was used Nitrogen adsorption-desorption isotherms at 77 K were measured on a containing a frit glass separating both compartments, a connection for Quantachrome Autosorb iQ gas adsorption analyzer. Prior to analysis, the sensor to be inserted in the anodic gas headspace, and connections the sample was degassed in vacuum. The BET method was applied to for the inlet and outlet Ar streams in both compartments (Supple- calculate the total surface area. mentary Fig. 26). The oxygen microsensor was in situ two-point cali- brated by feeding to the H-cell with certified standard of Ar (≥99.999%) Electrochemistry and compressed air, being the gas flow rate controlled by a set of mass All electrochemical experiments were performed under ambient con- flowmeters (Bronkhorst EL-FLOW). ditions (≈293 K) with a Bio-Logic VMP3 multichannel potentiostat and After purging the cell headspace with argon, the chron- implemented with a three-electrode configuration using 1 M H2SO4 opotentiometry test was started and the oxygen concentration was (pH 0.1) as filling solution, Pt mesh as counter electrode, Ag/AgCl (3 M monitored until reaching stabilization. The expected faradaic oxygen KCl) as reference electrode and a pocket working electrode (0.07 cm2 production rate (F O2 ,far in mol s−1) is calculated with the following surface area and 4 mm depth) filled with the GPO composites. The equation: actual mass amounts of the x-Co3O4@C/GPO composites in the elec- 1 trode pocket were measured with a weight balance and are indicated in F O2 ,far = I n1 e F ð1Þ Supplementary Table 5. Although it is difficult to estimate the actual active layer, an estimation is suggested that 1/8 of the total electrode where I is the applied current (in A), ne is the number of mols of pocket volume is used as the maximum limit in contact with the electrons involved in the water oxidation reaction to generate one mol solution in order to determine the mass loading for comparison of oxygen (4) and F is the Faraday constant (96,485 C mol–1). (Supplementary Table 5)36. All potentials were measured vs. Ag/AgCl The experimental O2 flow rate (F O2 ,exp in mol s–1) was calculated electrode and converted to the RHE reference scale using ERHE = considering ideal gas behavior with the following equation: EAg/AgCl + 0.21 + 0.059 pH (V) while overpotentials η = ERHE – 1.229 V. All current densities were calculated based on the geometrical surface F O2 ,exp = P C O2 F Ar ð100  CO2 Þ1 R1 T 1 ð2Þ area of the electrodes. Ohmic drop (R) was determined by using the automatic current interrupt (CI) software and the corresponding where P is the total gas pressure (1 atm), C (O2) is the steady-state ohmic drops were included in Supplementary Table 6. iR- oxygen concentration provided by the sensor (in %), FAr is the Ar compensations were applied to all electrochemical data. CV experi- carrier flow (in L s–1), R is the ideal gas constant (0.082 atm L K–1 mol–1) ments were carried out with 10 or 100 mV s–1 scan rates. Single LSV and T is the cell temperature (293 K). curve was recorded with a 1 mV s–1 scan rate for activity comparison Then Faradaic efficiency (in %), FE, is calculated as follows: after 10-CV cycle activation in the 0–1.4 V vs. Ag/AgCl potential range. Tafel slopes were estimated from the LSV curves by plotting over- 100 × F O2 ,exp FE = ð3Þ potential η vs. log j (j = current density). The potential vs. RHE to drive F O2 ,far 1 mA cm–2 was used to define onset potential and corresponding ηonset. Chronopotentiometry tests were carried out at fixed current densities of 10 or 100 mA cm–2. Accelerated degradation testing (ADT) was X-ray photoemission spectroscopy measured between 0–1.4 V vs. Ag/AgCl using 100 mV s−1 scan rate63. For X-Ray photoemission (XPS) experiments were performed inside an the electrochemical double-layer capacitance (EDLC) measurements, ultra-high vacuum chamber (base pressure of 10–10 mbar) using a Nature Communications | (2022)13:4341 7 Article https://doi.org/10.1038/s41467-022-32024-6 Phoibos 100 photoelectron spectrometer equipped with an Al Kα 12. Schipper, D. E. et al. Effects of catalyst phase on the hydrogen X-ray source (16 mA, 1486.6 V) as the incident photon radiation. XPS evolution reaction of water splitting: preparation of phase-pure spectra of Co 2p, O 1s, N 1s, and C 1s core levels were measured for as films of FeP, Fe2P, and Fe3P and their relative catalytic activities. received samples deposited on top of indium tape. The spectra are well Chem. Mater. 30, 3588–3598 (2018). described by the superposition of several Doniach-Sunjic curve-com- 13. Sun, J. et al. Robust hydrogen-evolving electrocatalyst from het- ponents. The intensities of the XPS core levels were evaluated by the erogeneous molybdenum disulfide-based catalyst. ACS Catal. 10, peak areas, after a standard background subtraction according to 1511–1519 (2020). Shirley procedure65. The spin–orbit splitting for every component into 14. Yu, J., Cheng, G. & Luo, W. Ternary nickel-iron sulfide microflowers the Co-2p core level has been set to D = 15.2 eV with a branching ratio of as a robust electrocatalyst for bifunctional water splitting. J. Mater. 0.5. The metallic cobalt peak, Co 2p3/2 = 779.7 eV, was used for a final Chem. A 5, 15838–15844 (2017). calibration of the spectra38. Co3O4@C/GPO after OER chron- 15. Kuo, D. Y. et al. Influence of surface adsorption on the oxygen opotentiometry at 10 mA cm–2 for 2 h at was washed with acetone in evolution reaction on IrO2 (110). J. Am. Chem. Soc. 139, order to remove the paraffin oil prior to XPS measurements. As 3473–3479 (2017). expected, the treatment do not affect the oxidation state of the 16. Cherevko, S. et al. Oxygen and hydrogen evolution reactions on Ru, composite, as evidenced by the measured spectra. RuO2, Ir, and IrO2 thin film electrodes in acidic and alkaline elec- trolytes: a comparative study on activity and stability. Catal. Today Transmission electron microscopy 262, 170–180 (2016). High-resolution transmission electron microscopy (HRTEM) and 17. Chen, S. et al. Mn-doped RuO2 nanocrystals as highly active elec- scanning transmission electron microscopy (STEM) investigations trocatalysts for enhanced oxygen evolution in acidic media. ACS were performed on a field emission gun FEI Tecnai F20 microscope. Catal. 10, 1152–1160 (2020). High-angle annular dark-field (HAADF) STEM was combined with 18. Shan, J., Zheng, Y., Shi, B., Davey, K. & Qiao, S. Z. Regulating electron energy loss spectroscopy (EELS) in the Tecnai microscope by electrocatalysts via surface and interface engineering for acidic using a GATAN QUANTUM energy filter in order to obtain composi- water electrooxidation. ACS Energy Lett. 4, 2719–2730 (2019). tional maps. STEM-EELS maps were performed using the O K-edge at 19. Escudero-Escribano, M. et al. Importance of surface IrOx in stabi- 532 eV (green), the Co L-edge at 779 eV (red) and C K-edge at lizing RuO2 for oxygen evolution. J. Phys. Chem. B 122, 284 eV (blue). 947–955 (2018). 20. Seitz, L. C. et al. A highly active and stable IrOx/SrIrO3 catalyst for Reporting summary the oxygen evolution reaction. Science 353, 1011–1014 (2016). Further information on research design is available in the Nature 21. Ouattara, L., Fierro, S., Frey, O., Koudelka, M. & Comninellis, C. Research Reporting Summary linked to this article. Electrochemical comparison of IrO2 prepared by anodic oxidation of pure iridium and IrO2 prepared by thermal decomposition of Data availability H2IrCl6 precursor solution. J. Appl. Electrochem. 39, All data is available in the main text or in the supplementary materi- 1361–1367 (2009). als. Source data are provided with this paper. 22. Lyons, M. E. G. & Floquet, S. Mechanism of oxygen reactions at porous oxide electrodes. Part 2—Oxygen evolution at RuO2, IrO2 References and IrxRu1-xO2 electrodes in aqueous acid and alkaline solution. 1. Wei, C. et al. Recommended practices and benchmark activity for Phys. Chem. Chem. Phys. 13, 5314–5335 (2011). hydrogen and oxygen electrocatalysis in water splitting and fuel 23. Stoerzinger, K. A. et al. Orientation-dependent oxygen evolution cells. Adv. Mater. 31, 1–24 (2019). on RuO2 without lattice exchange. ACS Energy Lett. 2, 2. Staffell, I. et al. The role of hydrogen and fuel cells in the global 876–881 (2017). energy system. Energy Environ. Sci. 12, 463–491 (2019). 24. Rodríguez-García, B. et al. Cobalt hexacyanoferrate supported on 3. Singh, S. et al. Hydrogen: a sustainable fuel for future of the Sb-doped SnO2 as a non-noble catalyst for oxygen evolution in transport sector. Renew. Sustain. Energy Rev. 51, 623–633 (2015). acidic medium. Sustain. Energy Fuels 2, 589–597 (2018). 4. Sharma, S. & Ghoshal, S. K. Hydrogen the future transportation fuel: 25. Li, A. et al. Stable potential windows for long-term electrocatalysis from production to applications. Renew. Sustain. Energy Rev. 43, by manganese oxides under acidic conditions. Angew. Chem. Int. 1151–1158 (2015). Ed. 58, 5054–5058 (2019). 5. Seh, Z. W. et al. Combining theory and experiment in electro- 26. Moreno-Hernandez, I. A. et al. Crystalline nickel manganese anti- catalysis: Insights into materials design. Science 355, 1–12 (2017). monate as a stable water-oxidation catalyst in aqueous 1.0 M 6. McCrory, C. C. L. et al. Benchmarking hydrogen evolving reaction H2SO4. Energy Environ. Sci. 10, 2103–2108 (2017). and oxygen evolving reaction electrocatalysts for solar water 27. Kwong, W. L., Lee, C. C., Shchukarev, A. & Messinger, J. Cobalt- splitting devices. J. Am. Chem. Soc. 137, 4347–4357 (2015). doped hematite thin films for electrocatalytic water oxidation in 7. Zhu, J., Hu, L., Zhao, P., Lee, L. Y. S. & Wong, K. Y. Recent advances in highly acidic media. Chem. Commun. 55, 5017–5020 (2019). electrocatalytic hydrogen evolution using nanoparticles. Chem. 28. Zhou, L. et al. Rutile alloys in the Mn–Sb–O system stabilize Mn3+ to Rev. 120, 851–918 (2020). enable oxygen evolution in strong acid. ACS Catal. 8, 8. Lyu, F., Wang, Q., Choi, S. M. & Yin, Y. Noble-metal-free electro- 10938–10948 (2018). catalysts for oxygen evolution. Small 15, 1–17 (2019). 29. Chatti, M. et al. Intrinsically stable in situ generated electrocatalyst 9. Spöri, C., Kwan, J. T. H., Bonakdarpour, A., Wilkinson, D. P. & for long-term oxidation of acidic water at up to 80 °C. Nat. Catal. 2, Strasser, P. The stability challenges of oxygen evolving catalysts: 457–465 (2019). towards a common fundamental understanding and mitigation of 30. Bloor, L. G., Molina, P. I., Symes, M. D. & Cronin, L. Low pH elec- catalyst degradation. Angew. Chem. Int. Ed. 56, 5994–6021 (2017). trolytic water splitting using earth-abundant metastable catalysts 10. Han, B. et al. Screening oxide support materials for OER catalysts in that self-assemble in situ. J. Am. Chem. Soc. 136, acid. J. Electrochem. Soc. 165, 813–820 (2018). 3304–3311 (2014). 11. Chung, D. Y. et al. Large-scale synthesis of carbon-shell-coated FeP 31. Mondschein, J. S. et al. Crystalline cobalt oxide films for sustained nanoparticles for robust hydrogen evolution reaction electro- electrocatalytic oxygen evolution under strongly acidic conditions. catalyst. J. Am. Chem. Soc. 139, 6669–6674 (2017). Chem. Mater. 29, 950–957 (2017). Nature Communications | (2022)13:4341 8 Article https://doi.org/10.1038/s41467-022-32024-6 32. Etzi Coller Pascuzzi, M., van Velzen, M., Hofmann, J. P. & Hensen, E. 52. Kim, Y. T. et al. Balancing activity, stability and conductivity of J. M. On the stability of Co3O4 oxygen evolution electrocatalysts in nanoporous core-shell iridium/iridium oxide oxygen evolution acid. ChemCatChem 13, 459–467 (2021). catalysts. Nat. Commun. 8, 1–8 (2017). 33. Yang, X. et al. Highly acid-durable carbon coated Co3O4 nanoar- 53. An, L. et al. Recent development of oxygen evolution electro- rays as efficient oxygen evolution electrocatalysts. Nano Energy catalysts in acidic environment. Adv. Mater. 33, 1–31 (2021). 25, 42–50 (2016). 54. Díaz-Fernández, D. et al. The growth of cobalt oxides on HOPG 34. Yan, K. L. et al. Probing the active sites of Co3O4 for the acidic and SiO2 surfaces: A comparative study. Surf. Sci. 624, oxygen evolution reaction by modulating the Co2+/Co3+ ratio. J. 145–153 (2014). Mater. Chem. A 6, 5678–5686 (2018). 55. Wang, J., Musameh, M. & Mo, J. W. Acid stability of carbon paste 35. Li, A. et al. Enhancing the stability of cobalt spinel oxide towards enzyme electrodes. Anal. Chem. 78, 7044–7047 (2006). sustainable oxygen evolution in acid. Nat. Catal. 5, 109–118 (2022). 56. Fang, Z. et al. Hybrid organic-inorganic gel electrocatalyst for 36. Arens, J. T. et al. Water oxidation electrocatalysis in acidic media stable acidic water oxidation. ACS Nano 13, 14368–14376 (2019). with Co-containing polyoxometalates. J. Catal. 389, 57. Sempionatto, J. R. et al. Enzymatic biofuel cells based on protec- 345–351 (2020). tive hydrophobic carbon paste electrodes: towards epidermal 37. Park, K. S. et al. Exceptional chemical and thermal stability of bioenergy harvesting in the acidic sweat environment. Chem. zeolitic imidazolate frameworks. Proc. Natl Acad. Sci. USA 103, Commun. 56, 2004–2007 (2020). 10186–10191 (2006). 58. Liang, Y. et al. Co3O4 nanocrystals on graphene as a synergistic 38. Biesinger, M. C. et al. Resolving surface chemical states in XPS catalyst for oxygen reduction reaction. Nat. Mater. 10, analysis of first row transition metals, oxides and hydroxides: Cr, 780–786 (2011). Mn, Fe, Co and Ni. Appl. Surf. Sci. 257, 2717–2730 (2011). 59. Han, L. et al. Cobalt hexacyanoferrate as a selective and high 39. Anantharaj, S. et al. Recent trends and perspectives in electro- current density formate oxidation electrocatalyst. ACS Appl. chemical water splitting with an emphasis on sulfide, selenide, and Energy Mater. 3, 9198–9207 (2020). phosphide catalysts of Fe, Co, and Ni: a review. ACS Catal. 6, 60. Yang, F., Chen, Y., Cheng, G., Chen, S. & Luo, W. Ultrathin nitrogen- 8069–8097 (2016). doped carbon coated with CoP for efficient hydrogen evolution. 40. Reier, T., Oezaslan, M. & Strasser, P. Electrocatalytic oxygen evo- ACS Catal. 7, 3824–3831 (2017). lution reaction (OER) on Ru, Ir, and pt catalysts: a comparative 61. Yang, F. et al. Design of active and stable oxygen reduction reaction study of nanoparticles and bulk materials. ACS Catal. 2, catalysts by embedding CoxOy nanoparticles into nitrogen-doped 1765–1772 (2012). carbon. Nano Res. 10, 97–107 (2017). 41. Li, D., Shi, J. & Li, C. Transition-metal-based electrocatalysts as 62. Chen, J. Y. C., Miller, J. T., Gerken, J. B. & Stahl, S. S. Inverse spinel cocatalysts for photoelectrochemical water splitting: a mini review. NiFeAlO4 as a highly active oxygen evolution electrocatalyst: Small 14, 1–22 (2018). Promotion of activity by a redox-inert metal ion. Energy Environ. 42. Tamirat, A. G., Rick, J., Dubale, A. A., Su, W. N. & Hwang, B. J. Using Sci. 7, 1382–1386 (2014). hematite for photoelectrochemical water splitting: a review of 63. Petzoldt, P. J., Kwan, J. T. H., Bonakdarpour, A. & Wilkinson, D. P. current progress and challenges. Nanoscale Horiz. 1, Deconvoluting reversible and irreversible degradation phenom- 243–267 (2016). ena in OER catalyst coated membranes using a modified RDE 43. Oh, H. S., Nong, H. N., Reier, T., Gliech, M. & Strasser, P. Oxide- technique. J. Electrochem. Soc. 168, 026507 (2021). supported Ir nanodendrites with high activity and durability for the 64. Morales, D. M. & Risch, M. Seven steps to reliable cyclic voltam- oxygen evolution reaction in acid PEM water electrolyzers. Chem. metry measurements for the determination of double layer capa- Sci. 6, 3321–3328 (2015). citance. J. Phys. Energy 3, 34013 (2021). 44. Oh, H. S. et al. Electrochemical catalyst-support effects and their 65. Shirley, D. A. High-resolution X-ray photoemission spectrum of the stabilizing role for IrOx nanoparticle catalysts during the oxygen valence bands of gold. Phys. Rev. B 5, 4709–4714 (1972). evolution reaction. J. Am. Chem. Soc. 138, 12552–12563 (2016). 45. Reier, T., Nong, H. N., Teschner, D., Schlögl, R. & Strasser, P. Acknowledgements Electrocatalytic oxygen evolution reaction in acidic environments The authors thank the support from projects RED2018-102459-T, —reaction mechanisms and catalysts. Adv. Energy Mater. 7, RTI2018-095618-B-I00, PID2020-116093RB-C41,C43&C44, PID2019- 1601275 (2017). 107338RB-C63 and PID2020-114252GB-I00 funded by MCIN/AEI/ 46. Ashton, S. J. & Arenz, M. A DEMS study on the electrochemical 10.13039/501100011033/ and by “ERDF A way of making Europe”; the oxidation of a high surface area carbon black. Electrochem. Generalitat de Catalunya (2017-SGR-1406 and 2017-SGR-327), and Commun. 13, 1473–1475 (2011). the Basque Government (grant IT1591-22). J.Y. thanks the China Scho- 47. Hu, Q. et al. Coupling pentlandite nanoparticles and dual-doped larship Council (CSC) for predoctoral fellowships (File No. carbon networks to yield efficient and stable electrocatalysts for 201806270234). S.B. acknowledges Grant RYC-2017-21931 funded by acid water oxidation. J. Mater. Chem. A 7, 461–468 (2019). MCIN/AEI/ 10.13039/501100011033 and by ESF Investing in your future, 48. Cheng, W. et al. A metal-vacancy-solid-solution NiAlP nanowall and UPV/EHU project EHUrOPE19/01 and IKERBASQUE, Basque Foun- array bifunctional electrocatalyst for exceptional all-pH overall dation for Science. ICIQ and ICN2 are supported by the Ministerio de water splitting. J. Mater. Chem. A 6, 9420–9427 (2018). Ciencia e Innovación through the Severo Ochoa Excellence Accredita- 49. Mondschein, J. S. et al. Intermetallic Ni2Ta electrocatalyst for the tions CEX2019-000925-S (MIC/AEI) and SEV-2017-0706; and by the oxygen evolution reaction in highly acidic electrolytes. Inorg. CERCA Programme/Generalitat de Catalunya. M.C.S. has received Chem. 57, 6010–6015 (2018). funding from the European Union’s Horizon 2020 research and innova- 50. McCrory, C. C. L., Jung, S., Peters, J. C. & Jaramillo, T. F. Bench- tion programme under Marie Skodowska-Curie grant 754510 (PROBIST) marking heterogeneous electrocatalysts for the oxygen evolution and the Severo Ochoa programme. M.C.S. has received funding from reaction. J. Am. Chem. Soc. 135, 16977–16987 (2013). the postdoctoral fellowship Juan de la Cierva Incorporation from 51. Geiger, S. et al. The stability number as a metric for electrocatalyst MICINN (JCI-2019) and the Severo Ochoa programme. M.P.D. acknowl- stability benchmarking. Nat. Catal. 1, 508–515 (2018). edges funding for a Ph.D. fellowship from the MPC Foundation. Nature Communications | (2022)13:4341 9 Article https://doi.org/10.1038/s41467-022-32024-6 Author contributions Peer review information Nature Communications thanks the anon- J.R.G.M. conceived the original concept and with S.B. designed the ymous reviewers for their contribution to the peer review of this work. project and coordinated the efforts of the research team. J.Y. synthe- sized, processed and structurally characterized the materials and elec- Reprints and permission information is available at trodes. J.Y. and F.G.P. characterized the compounds electrochemically http://www.nature.com/reprints and analyzed the data. J.Y. and J.G.C. measured the oxygen evolution. M.P.D., C.R., and S.B. collected and analyzed the XPS data. M.C.S. and J. Publisher’s note Springer Nature remains neutral with regard to jur- A. collected and analyzed the TEM data. S.G. collected and analyzed the isdictional claims in published maps and institutional affiliations. EIS data. J.Y., S.B., and J.R.G.M. co-wrote the manuscript with con- tributions from all authors. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as Competing interests long as you give appropriate credit to the original author(s) and the The authors declare no competing interests. source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless Additional information indicated otherwise in a credit line to the material. If material is not Supplementary information The online version contains supplementary included in the article’s Creative Commons license and your intended material available at use is not permitted by statutory regulation or exceeds the permitted https://doi.org/10.1038/s41467-022-32024-6. use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/ Correspondence and requests for materials should be addressed to licenses/by/4.0/. Sara Barja or José Ramón. Galán-Mascarós. © The Author(s) 2022, corrected publication 2022 Nature Communications | (2022)13:4341 10