Naseem
Kousar
,
Gouthami
Patil
,
Ashwini Chikkabasur
Kumbara
,
Basavesh
Nisty
,
Rajesh
G. H.
and
Lokesh Koodlur
Sannegowda
*
Department of Studies in Chemistry, Vijayanagara Sri Krishnadevaraya University, Jnanasagara Campus, Vinayakanagara, Ballari-583105, Karnataka, India. E-mail: kslokesh@vskub.ac.in; naseem.kousar9999@gmail.com; gouthamipatil6@gmail.com; ashwindilip88@gmail.com; basavesh.nisty@gmail.com; rajeshgh95.rgh@gmail.com; Tel: +91 9035500208
First published on 24th July 2025
Advancements in water splitting technologies are crucial for achieving sustainable hydrogen production. Development of highly efficient and economically viable catalysts is essential for commercialization of water electrolysers. While precious metals like platinum and iridium are renowned for their catalytic capabilities in the hydrogen evolution reaction (HER) and oxygen evolution reaction (OER), their high cost and scarcity present significant challenges. Hence, various metal oxides, carbides, sulfides, phosphides, alloys, metal complexes, and composites have been examined as potential catalysts for water splitting reactions. This review offers a comprehensive analysis of Earth-abundant metal complexes as promising alternatives for water splitting catalysis. The fundamental principles underlying water splitting, including electrochemical dynamics, thermodynamics, and reaction kinetics, and their impact on catalytic performance have been evaluated. Emphasis is placed on the pivotal role of Earth-abundant metals such as manganese, iron, cobalt, nickel, and molybdenum and their recent innovations in catalyst design focussing on composites for enhancing the HER, OER, and integrated dual-function catalysis are discussed. Comparative evaluation related to advantages and limitations of these alternatives with respect to precious catalysts in terms of cost, availability, and environmental impact is presented. To integrate the same catalyst for HER and OER activities, insights into strategies for optimization of the performance are provided. Additionally, the review highlights the contributions of computational chemistry, including density functional theory studies in engineering catalyst design and understanding reaction mechanisms. Finally, an assessment of current challenges and future directions is presented to provide a holistic perspective on the transformative potential of Earth-abundant metal complexes in advancing sustainable water splitting technologies.
In contrast to this, electrochemical water splitting (EWS), in which water splits into molecular hydrogen (H2) and oxygen (O2) utilizing electricity derived from renewable energy sources, has emerged as a monumental approach for generating hydrogen.8 Being naturally carbon-neutral, this green hydrogen pathway presents an opportunity for decentralized hydrogen production by coupling with the periodic renewable energy sources. The hydrogen evolution reaction (HER) at the cathode and the oxygen evolution reaction (OER) at the anode are the two half-reactions responsible for EWS.9,10 Both the reactions, though appearing straightforward, are thermodynamically and kinetically challenging resulting in the requirement of significant activation energies.11 To overcome the slow reaction kinetics along with substantial energy losses, efficient catalysts are used for both the HER and OER and the efficiency of the catalysts has a significant impact on the performance of water electrolyzers.3,7 Although noble metals like platinum (Pt) for the HER and ruthenium (Ru) or iridium (Ir) oxides for the OER have been the standard catalysts for years due to their impressive catalytic activity, they have an array of drawbacks. Since these metals are expensive, rare, uni-functional, and their activity swiftly diminishes by electrochemical processes, their viability and versatility are confined. Furthermore, their comparatively inflexible electronic structure restricts the possibilities of optimization through structural or molecular engineering and in turn limits the ability to tune their catalytic properties.11,12
Beyond molecular engineering, the designed catalysts need to be scalable and cost-effective for practical application of EWS technologies. A recent techno-economic analysis revealed that stack components and catalysts contribute to more than 50% of the overall electrolyzer cost, emphasizing the urgent need for Earth-abundant and high-performance bifunctional materials.13 Additionally, the demand for distinct individual catalysts for the HER and OER causes system construction even more complicated and raises expenses. Bifunctional electrocatalysts can effectively catalyze both HER and OER, and hence a lot of curiosity is drawn as a potential remedy to these challenges.14 These materials advance performance, simplify device construction, and lessen component count for water electrolyzers and unitize renewable fuel cells.15 The design of bifunctional catalysts is inherently difficult because the OER and HER necessitate distinctive active site features and operate via different mechanisms.11 The OER involves a complex four-electron oxidation process with higher thermodynamic and kinetic barriers, whereas the HER usually involves rapid proton-coupled electron transfer (PCET) steps in either acidic or alkaline media.16,17 Therefore, catalysts with adjustable electronic structures, multiple accessible oxidative states, substantial surface area, and robustness in structure under varying redox scenarios are needed to achieve optimal bifunctionality.14
Over the last decade, researchers have focused on electrocatalysts composed of Earth-abundant transition metals such as manganese (Mn), iron (Fe), cobalt (Co), nickel (Ni), and molybdenum (Mo), which have made substantial advances as noble metal substitutes. These transition-metal-based catalysts provide high catalytic efficiency while remaining cost-effective, abundant, and sustainable.18,19 Particularly, the rich redox-chemistry of these transition metals, along with their variety of oxidation states, allow them to take part in multi-electron transfer and PCET processes, which are necessary for EWS reactions. Because of these intrinsic characteristics of transition metals, molecular design techniques such as ligand alterations, doping with heteroatoms, optimization in the coordination environment, and integration with conductive supports can be used to specifically change their electronic and structural properties for tuning the catalytic properties.20,21 For example, Mn is used for the OER, particularly in alkaline media by utilizing its wide range of oxidation states (Mn2+ to Mn7+) through nano-structuring and doping.22 When iron is embedded in N-doped supports or macrocyclic ligands, the redox pair of Fe2+/Fe3+ exhibits increased activity.23 Through hybridization and ligand modulation, Co performance is adjusted to access Co2+/Co3+ states.24 Through alloying (such as NiFe) and support coupling, Ni2+–Ni4+ transitions of Ni are empowered which strengthens the activity for the OER in alkaline medium.25 Mo forms effective bifunctional hybrids like NiMo and CoMo and excels in the HER through active edge sites in MoS2.26,27
Among the various material platforms based on transition metals, metal complexes are unique due to their structural flexibility and molecular reliability. These complexes provide atomically defined coordination spheres around the metal center, offering precise control over the oxidation state, electron density, and geometry, in contrast to bulk metal oxides or alloys.28 This ability to fine-tune the metal center is essential for adjusting the important catalytic descriptors that regulate the effectiveness and selectivity of EWS reactions, such as intermediate adsorption energies, redox potentials, and PCET kinetics.21,29 The molecular makeup of the complex also helps in conducting through structure–activity correlation studies, which yield mechanistic insights that are challenging to discover in heterogeneous systems. The logical design of next-generation catalysts with desired performance characteristics will be greatly benefitted from such insights.30,31 Redox flexibility is another characteristic that makes metal complexes highly desirable. It enables them to participate in multi-electron transformations and stabilize high-energy intermediates during catalytic turnover.32 By integrating them into hybrid systems with conductive frameworks like graphene, carbon nanotubes, or porous matrix structures, their activity can be further increased.11 These hybrids elevate electrochemical active surface area, boost charge transport, and induce synergistic interactions, which ultimately result in increased durability and bifunctionality.3,11 They are especially attractive for practical applications in unitized electrolyzer–fuel cell systems and renewable energy devices due to their configurable and designable architecture.15,33 In view of these beneficial features, Earth-abundant metal complexes have evolved as an exciting category of bifunctional electrocatalysts that can effectively drive both the OER and HER in EWS. This review discusses recent advances in the design of such complexes and hybrid derivatives (Fig. 1). Special attention is paid to the design principles, structure–function relationships, and mechanistic understanding that support their catalytic function. Furthermore, the multi-metallic systems and synergistic approaches that improve stability and activity in a range of operational conditions have been discussed. The ultimate goal of this review is to provide a comprehensive overview and identify future research directions concerning the role of metal complexes in the evolving field of sustainable hydrogen production technologies.
![]() | (1) |
485 C mol−1). At 298 K and n = 1, eqn (1) simplifies to eqn (2), which demonstrates that while the total cell voltage requirement remains unchanged across pH, each half-reaction shifts by about 59 mV per pH unit.35| E = E° − 0.059 × pH | (2) |
Despite this advantageous thermodynamic start point, kinetic barriers and mass transport constraints usually cause practical electrolyzers to function at 1.6–2.0 V. This extra voltage, known as the overpotential (η, eqn (3)), is the energy required to overcome the catalytic cycle's largest free energy barrier (ΔGmax).36
![]() | (3) |
In order to design effective catalysts, it is essential to minimize ΔGmax. This has been investigated systematically using volcano plots and free energy diagrams based on the binding strengths of intermediates like H, OH, and OOH*. Redox-flexible ligands containing transition metals exhibit improved kinetics by reducing the activation barriers of reactions. Moreover, cooperative interactions in multinuclear metal complexes have demonstrated potential for distributing oxidative load, stabilizing transition states, and adjusting electron density, all of which enhance the activity and long-term resilience in challenging electrochemical environments.37 As a result, effective catalyst design requires identifying these kinetic bottlenecks and creating materials that reduce them.
On the other hand, the OER, which involves a four-electron, four-proton transformation, is intrinsically more complicated (Fig. 4). Water adsorption at the catalytic site initiates the OER in acidic environments. Subsequent deprotonation and oxidation processes result in the production of metal–hydroxyl (OH*), metal–oxo (O*), and metal–hydroperoxo (OOH*) intermediates prior to O2 release. OH− ions adsorb onto the catalyst surface under alkaline conditions, where they undergo oxidation to form similar OH*, O*, and OOH* intermediates prior to oxygen evolution.40 Recent mechanistic studies using isotope labelling and operando spectroscopy have revealed distinct energetic pathways for the OER in alkaline systems, where the deprotonation of OH− and formation of M–OOH intermediates govern the rate.41 Although acidic and alkaline systems share a general mechanistic framework, the nature of crucial intermediary molecules, proton–electron coupling, and interfacial dynamics are affected by the different proton and oxygen sources, requiring tailored catalyst designs that carefully consider these differences.
![]() | (4) |
O species, which may impede turnover and introduce reorganization consequences. In order to reduce these roadblocks and increase efficiency, ligand scaffolds that maintain constant spin states throughout the redox transitions are employed.51
O can be stabilized by solvation and hydrogen bonding effects.55
| Parameter | Definition | HER (acidic/alkaline) | OER (acidic/alkaline) | Overall EWS |
|---|---|---|---|---|
| Thermodynamic potential (E°) | Minimum equilibrium potential under standard conditions (298 K, pH 0 ) | 0.00 V (vs. RHE) | +1.23 V (vs. RHE) | 1.23 V total |
| Overpotential (η) | Additional voltage required beyond E° to drive catalysis | <100 mV (target) | <400 mV (target) | Combined, system-dependent |
| Binding free energy (ΔG) | Optimal adsorption energy of key intermediates | ΔGH ≈ 0.0 eV | ΔGOH, ΔGO, ΔGOOH ≈ 1.23 eV per step | Reflected in cumulative performance |
| Tafel slope | Kinetic indicator of rate-determining step | <60 mV dec−1 | <60 mV dec−1 | Aggregate from half-reactions |
| Exchange current density (j0) | Intrinsic catalytic activity at zero overpotential | ≥10−3 A cm−2 | ≥10−5 A cm−2 | Determined by the slowest step |
| Turnover frequency (TOF) | Catalytic cycles per active site per second | >10 s−1 | >1 s−1 | Limited by OER kinetics |
| Turnover number (TON) | Total cycles before catalyst deactivation | >104 | >104 | Reflects long-term durability |
| Faradaic efficiency (FE) | Fraction of charge converted to target product | >95% | >95% | >95% overall |
| Reorganization energy (λ) | Energy cost of geometric and solvation rearrangements in PCET | Low (favors rapid PCET) | Low (favors redox turnover) | Co-optimized across both |
| d-Band center (electronic descriptor) | Correlates metal electronic structure to adsorbate binding strength | Tuned for balanced H* adsorption | Tuned for balanced OH*/O*/OOH* adsorption | Critical for bifunctional design |
| H2/O2 production rate | Volume or molar rate of gas generated per unit time and area | ≥10 mmol cm−2 h−1 (H2) | ≥5 mmol cm−2 h−1 (O2) | Stoichiometric ratio 2 : 1 (H2 : O2) |
| pH-Dependent potential shift | Theoretical shift per pH unit (Nernst equation) | ∼59 mV pH−1 unit | ∼59 mV pH−1 unit | Overall splitting remains invariant |
To address this duality, bifunctionality in molecular systems is frequently achieved by means of precisely designed ligand fields that stabilize both the HER and OER-relevant intermediates, while modulating electronic density at the metal site.33,62 Further improving PCET efficiency, lowering activation barriers, and stabilizing high-energy transition states can be accomplished by redox-active ligands, proton-responsive moieties, and secondary-sphere functionalities (such as intramolecular base sites or hydrogen-bond donors).30,63 Such cooperative effects are often linked to low Tafel slopes in both systems and high TOFs in both half-reactions. To overcome the limitations of mononuclear systems, researchers have explored multi-nuclear complexes where cooperative metal centers modulate electron density and distribute oxidative stress across the framework. This design improves both catalytic activity and durability under bifunctional conditions.37
In parallel, hybrid frameworks that incorporate molecular catalysts into conductive or porous supports offer synergistic advantages. High surface area matrices such as carbon nanotubes, graphene, or mesoporous scaffolds improve mass transport, enhance electrochemical active surface area, and facilitate electron transfer across active sites.64 In this regard, emerging materials that integrate MOF-derived active centers with molecular catalysts have demonstrated exceptional bifunctional performance. These systems benefits from modular coordination environments, high porosity, and strong metal–ligand covalency that collectively support both HER and OER.41 Such hybridization enables tuning of local electric fields, improves electrolyte ion access, and prevents deactivation via leaching or ligand degradation. Effective bifunctional electrocatalysts must therefore combine redox flexibility, structural logic, interfacial stability, and vibrant proton–electron association within a single catalytic framework.14,65 Rather than being separately optimized for each half-reaction, true bifunctionality arises from the integrated control of these interdependent properties at the molecular and mesoscale levels.
For monitoring ligand field shifts and electronic transitions in molecular complexes like phthalocyanines and porphyrins, in situ ultraviolet-visible (UV–Vis) spectroscopy is promising.70 The development of catalytically active oxidation states or intermediates can be identified by changes in absorption profiles under an applied voltage. In addition, vibrational fingerprints of surface-bound species like M–OH, M
O, and bridging oxygen ligands can be obtained using infrared (IR) spectroscopy, particularly in the attenuated total reflectance (ATR) configuration.71 These vibrational characteristics are important for differentiating between concerted and stepwise reaction pathways, and provide information on PCET activities during the HER and OER. Additional sensitivity to dynamic bonding environments and intermediate formation is provided by Raman spectroscopy, which includes resonance Raman and surface-enhanced Raman spectroscopy (SERS). As an instance, the time-resolved Raman technique has been used to track the O–O bond formation step in the OER via transient *OOH and *O vibrational modes in MOFs.72 Nanogram-level mass changes at the electrode surface during catalytic turnover are monitored using an electrochemical quartz crystal microbalance (EQCM). This method has directly linked structural stability to operational durability by the identification of the mass increase brought about by hydroxide or oxide formation during the OER and mass loss brought about by ligand degradation or metal leaching.73
The study of surface oxidation states, coordination changes, and chemical composition under reaction-relevant conditions has benefited greatly from the use of near-ambient pressure X-ray photoelectron spectroscopy (AP-XPS). AP-XPS can reveal oxidation-induced ligand transformation under bias and distinguish surface-anchored M–N4 sites from adsorbed hydroxide species for metal complexes anchored on conductive supports.74 Kelvin probe force microscopy (KPFM) and scanning electrochemical microscopy (SECM) are spatially resolved methods for the mapping of electronic work function and local reactivity, respectively. While KPFM resolves potential gradients and local dipoles across hybrid catalyst interfaces, SECM enables imaging of active site density and proton reduction zones during the HER.75 These observations are especially pertinent to heterogenized metal complexes, where surface coverage and redox gradients affect bifunctionality. Furthermore, product quantification and intermediate tracking have been made possible by new methods like electrochemical gas chromatography and differential electrochemical mass spectrometry (DEMS).76 These data are helpful for verifying reaction selectivity and FE in intricate catalytic systems.
When combined, these real-time analytical tools have made it possible to directly correlate important electrochemical parameters like the TOF, exchange current density, Tafel slope, and overpotential with spectroscopic observables. Additionally, the direct detection of intermediates and active site motifs have made it possible to validate DFT predictions. Incorporation of these methods will become more and more essential as the field shifts toward logical, mechanism-driven catalyst development. Real-time characterization offers the mechanistic foundation for well-informed catalyst design in molecular and hybrid electrocatalysts, where minor modifications to the ligand environment or nuclearity can significantly alter the performance.
![]() | ||
| Fig. 5 (A) Schematic of Mn–MOF/NF electrocatalyst development; (B) proposed HER mechanism in 0.1 M KOH; (C) proposed OER mechanism in 0.1 M KOH; (D) overall EWS performance: (a) two-electrode setup photograph, (b) LSV polarization curve, and (c) long-term durability test [reproduced from ref. 78 with permission from the American Chemical Society, copyright 2022]. | ||
Recently, Hongxin Guan et al. synthesized FeMn bimetallic MOFs on nickel foam (FeMn–MOF/NF) via a one-step solvothermal method and constructed binder-free electrodes for overall EWS. The optimal Fe
:
Mn (1
:
1) system exhibited excellent bifunctional activity, achieving an OER overpotential of 290 mV and a HER overpotential of −260 mV at 50 mA cm−2, with Tafel slopes of 87 mV dec−1 (OER) and 156.6 mV dec−1 (HER). It showed the highest Cdl (∼3.66 mF cm−2) among the tested variants and retained strong stability over 12 h with minimal performance loss after 1000 cycles. In a two-electrode setup, FeMn–MOF/NF required only 1.70 V at 50 mA cm−2 for overall water splitting, with just a ∼0.02 V increase after 60 days. The outstanding activity was attributed to the Fe/Mn synergistic redox effects, high surface area from the 3D nanoflower morphology, and conductive NF support ensuring efficient charge transfer.79 Furthermore, Liu et al. synthesized two-dimensional CoNi–metal–organic framework (CoNi–MOF) nanoplate arrays on Cu foil through a hydrothermal method, demonstrating excellent catalytic performance for alkaline water splitting. The optimized Co
:
Ni (1
:
1) MOF displayed dominant (220) facets, providing enhanced conductivity along the Z-axis and abundant exposed active sites. In 1.0 M KOH, the CoNi (1
:
1)–MOF achieved an overpotential of 265 mV at 10 mA cm−2 for the OER, with a low Tafel slope of 56 mV dec−1 and minimal charge-transfer resistance (0.924 Ω), outperforming single-metal MOFs and RuO2. The catalyst maintained ∼90% activity over 20 hours. After annealing under NH3, the CoNiN@C hybrid, consisting of metallic nitrides on a porous carbon framework, delivered an efficient HER, requiring only a −120 mV overpotential at 10 mA cm−2, with improved capacitance (39.2 mF cm−2) and lower charge-transfer resistance (4.62 Ω). In a two-electrode setup, combining CoNi(1
:
1)–MOF and CoNiN@C achieved overall water splitting at 1.64 V with ∼99% FE, confirming the designed system's high bifunctional catalytic activity.80 Using the pyrolysis of dual-ligand Co–MOFs assembled from thiophene-2,5-dicarboxylate and 4,4′-bipyridine on graphene oxide at 1000 °C, Zhang et al. developed a bifunctional electrocatalyst composed of Co/Co9S8 core–shell nanoparticles embedded in S,N-doped porous graphene sheets (Co/Co9S8@SNGS). The resultant hybrid endured uniform heteroatom doping, hierarchical porosity, and a high surface area (249.6 m2 g−1). Excellent bifunctional activity was demonstrated by electrochemical studies in 0.1 M KOH, with an OER overpotential of 290 mV at 10 mA cm−2 (Tafel slope: 73 mV dec−1) and a HER overpotential of 350 mV at 20 mA cm−2 (Tafel slope: 96 mV dec−1). With nearly 100% FE, full EWS with identical electrodes produced H2 and O2 evolution rates of 4.87 and 2.48 μmol min−1, respectively, at 1.58 V. Synergistic effects between the metallic Co core, Co9S8 shell, S,N co-doping, and Co–Nx sites are responsible for the enhanced performance. These effects jointly boost conductivity, accessibility of the active sites, and catalytic durability. The potential benefit of dual-heteroatom MOF precursors in designing integrated electrocatalysts for general alkaline water electrolysis is demonstrated in this work.81
Basically, Ni and Fe bimetallic complexes are highly promising electrocatalysts for bifunctional water splitting reactions. Duan et al. developed ultrathin 2D NiFe–MOF nanosheet arrays via a one-step chemical bath deposition using 2,6-naphthalenedicarboxylic acid dipotassium (NDCA) as the organic linker with Ni and Fe salts. The crystalline structure, built from alternating NDCA units and octahedral MO6 (M = Ni, Fe) directly grown on nickel foam (NF), forms meso- and macro-porous architectures. In 0.1 M KOH, NiFe–MOF/NF showed excellent OER activity, requiring only a 240 mV overpotential at 10 mA cm−2, outperforming the Ni–MOF (296 mV), Fe–MOF (324 mV), and NiFe–MOF on glassy carbon (406 mV). For the HER, the NiFe–MOF achieved an overpotential of −134 mV at 10 mA cm−2, superior to those of the Ni–MOF (−177 mV), bulk NiFe–MOF (−196 mV), and calcined NiFe–MOF (−255 mV). Stability testing confirmed consistent HER and OER activity over 20
000 s. As a full water-splitting cell, the NiFe–MOF used as both electrodes delivered 10 mA cm−2 at 1.55 V for 20 h without notable activity loss (Fig. 6). The enhanced catalytic performance arises from the nanosheet architecture exposing abundant Ni/Fe active sites, improving conductivity, and enabling efficient mass transport. However, only one Fe composition (23%) was studied, leaving room for future optimization.82 Using mild heat, Lin et al. phosphated Fe/Ni-based MOF-derived nanosheet arrays on nickel foam to create the bifunctional electrocatalyst (Fe0.1Ni0.9)2P(O)/NF. Ni, Fe, P, and O were uniformly distributed in the final material. The addition of Fe caused lattice distortion, while the addition of oxygen contributed to altered surface chemistry. An overpotential of 240 mV at 100 mA cm−2 for the OER with a Tafel slope of 72 mV dec−1 and 87 mV at 10 mA cm−2 for the HER with stable kinetics were noticed from electrochemical measurements in 1.0 M KOH, which demonstrated efficient activity for both half-reactions. The system demonstrated approximately 100% FE, maintained current stability for 40 hours, and required 1.50 V to sustain 10 mA cm−2 when used as both the anode and cathode in a two-electrode setup. Fe–Ni interactions, high electrochemical active surface area, oxygen doping, and decreased charge-transfer resistance from the precursor transformation process were all associated with the boosting of performance.83
![]() | ||
| Fig. 6 (A) Schematic of NiFe–MOF nanosheet array synthesis; (B) morphological characterization: optical, SEM, TEM, HRTEM, SAED, and AFM images; (C) LSV plots of a two-electrode cell with NiFe–MOF electrodes in 0.1 M KOH, including comparison with Pt/C∥IrO2 and gas bubble evolution at 1.6 V; (D) LSV before and after 20 h chronoamperometric testing at 1.5 V, with the inset showing a stability plot [reproduced from ref. 82 under the terms of the Creative Commons Attribution 4.0 International License (CC BY 4.0)]. | ||
Furthermore, Yang et al. developed the bifunctional NF@FePPc-s/p catalyst by phosphidating surfactant-modified iron polyphthalocyanine grown on nickel foam, producing Fe2P and Ni2P phases with a unique needle mushroom-like morphology. In 1.0 M KOH, it achieved low overpotentials of −190 mV (HER) and 293 mV (OER) at 100 mA cm−2, with Tafel slopes of 123.1 and 82.4 mV dec−1, respectively, outperforming NF@Pt/C and NF@RuO2 references. The catalyst showed the highest ECSA (Cdl: ∼43 mF cm−2) and low charge transfer resistance (3.3 Ω), indicating abundant active sites and fast kinetics. Faradaic efficiencies reached ∼95% for H2 and ∼96% for O2. Long-term tests showed minimal overpotential shifts after 5000–15
000 cycles, with ∼96–98% current retention over 24 h. In a full water-splitting cell, NF@FePPc-s/p∥NF@FePPc-s/p required only 1.433 V at 10 mA cm−2 and 1.743 V at 100 mA cm−2, surpassing Pt/C∥RuO2 benchmarks, with post-cycling XPS confirming Fe2P-to-FeOOH conversion (OER) and stable Fe2P (HER).84 Similarly, Qiuxiang Mou's group developed a NiFe–MOF-5 catalyst with a 3D hierarchical structure grown on Ni foam via a one-step solvothermal method. Nickel(II) nitrate, iron(III) nitrate, and formic acid were dissolved in DMF, and Ni foam was immersed and heated at 100 °C for 12 h in a sealed autoclave. NiFe–MOF-5 exhibited excellent bifunctional activity, with overpotentials of −163 mV (HER) and 168 mV (OER) at 10 mA cm−2, and Tafel slopes of 139 mV dec−1 (HER) and 43 mV dec−1 (OER), indicating fast kinetics. Overall water splitting required just 1.57 V at 10 mA cm−2 in 1.0 M KOH, outperforming many non-precious catalysts, with stable performance maintained over 24 h at 1.8 V and 10
000 LSV cycles.85
Trimetallic MOFs containing Fe, Ni and Co act as efficient multifunctional catalysts. For instance, Farahani et al. developed an innovative and tunable electrodeposition method for in situ growth of a trimetallic Fe–Co–Ni MOF on nickel foam using a layer-by-layer (LbL) reductive approach. This process was carried out in a DMF/H2O solution with metal ions, cetyltrimethylammonium bromide (CTAB), and 2-amino-1,4-benzenedicarboxylic acid yielding a distinct trilayer MOF structure, confirmed by cross-sectional SEM results. In 1.0 M KOH, the Fe–Co–Ni MOF achieved an overpotential of 254 mV at 10 mA cm−2 for OER, outperforming the Fe–Co MOF (280 mV), Fe MOF (290 mV), and bare Ni foam (440 mV), with a Tafel slope of 51.3 mV dec−1 and stability up to 48 h. HER activity followed the same trend, and a two-electrode cell required only 1.6 V at 10 mA cm−2, maintaining stability for 150 h, surpassing Pt/C∥RuO2 (1.62 V). Post-stability XPS revealed a slight increase in trivalent species, indicating possible (oxy)hydroxide formation. DFT calculations showed that the Fe–Co–Ni MOF had more negative H2O adsorption energy, a smaller bandgap, and higher electronic density near the Fermi level compared to mono- and bimetallic MOFs, enhancing conductivity and catalytic performance. Notably, the material also exhibited excellent ORR activity, making it suitable for Zn–air batteries and supercapacitors.86 Besides pyrolyzing MOF precursors, complex metal compounds with varied nanostructures and compositions were also synthesized, serving as efficient bifunctional electrocatalysts for both the HER and OER.87 For instance, Das et al. developed Ni/MoxC–NC, combining nickel and molybdenum carbide nanoparticles on nitrogen-doped graphene/carbon nanotube supports, through thermal treatment of NiMoO4·xH2O and melamine at 850 °C under nitrogen. TEM and HR-TEM confirmed the presence of Mo2C, MoC, and Ni crystalline phases embedded in the carbon matrix. In 1.0 M KOH, Ni/MoxC–NC showed excellent HER and OER performance, with overpotentials of −162 mV and 328 mV at 10 mA cm−2, Tafel slopes of 104 and 74 mV dec−1, and the highest ECSA (29.8 mF cm−2) among the tested catalysts. The system achieved overall water splitting at 1.72 V, comparable to that of Pt/C∥RuO2, and generated H2 and O2 at 0.034 and 0.0165 mL min−1 (Fig. 7). Mechanistic analysis showed that Ni2+ species promote water dissociation (Volmer step), MoC provides Pt-like electronic features, and pyridinic–N enriches the carbon surface, increasing catalytic sites. For the OER, the transformation of Ni(II) to NiOOH plays a key role, with the system delivering better performance and a shifted oxidation peak compared to Ni/NC-100.88
![]() | ||
| Fig. 7 (A) Schematic of Ni/MoxC–NC synthesis; (B) FESEM, TEM, and XRD of NiMoO4·0.7H2O nanobelts, Ni/MoO2–gC3N4 sheets, and Ni/MoxC–NC; (C) (a) polarization curves for overall water splitting in 1.0 M KOH, (b) H2 and O2 production over time; (D) (a and b) high-resolution XPS spectra (Ni 2p, Mo 3d), and (c) schematic of electron transfer pathways in water splitting [reproduced from ref. 88 with permission from the American Chemical Society, copyright 2018]. | ||
Apart from molecularly defined metal complexes, a lot of recent catalyst structures use materials derived from MOFs. In these materials, Earth-abundant metal frameworks, like ZIF-67, act as both structural scaffolds and metal avenues and, undergo regulated post-synthetic morphing to produce highly active electrocatalysts. These systems, which frequently combine the robustness and conductivity of metal phosphides, nitrides, or oxides with the creative tunability of molecular precursors, fill the void between homogeneous molecular catalysts and heterogeneous nanostructures. For instance, Li et al. developed ultrathin Mn-doped CoP nanosheets (Mn–CoP) using an etching-carbonization–phosphidation method starting from rhombic dodecahedral ZIF-67 precursors. These Mn–CoP nanosheets exhibited remarkable bifunctional electrocatalytic performance for the HER and OER, achieving overpotentials of −148 mV (acidic) and −195 mV (alkaline) for the HER, and 290 mV (alkaline) for the OER at 10 mA cm−2, outperforming hollow CoP and Mn–CoP nanoparticles, and even commercial RuO2. The materials showed Tafel slopes of 61 and 85 mV dec−1 (HER) and 76 mV dec−1 (OER), alongside excellent long-term durability up to 30 h. Structural analysis confirmed a large BET surface area (357.7 m2 g−1), meso-porosity, and uniform distribution of Mn, Co, P, C, and N. Electrochemical impedance spectroscopy revealed low charge-transfer resistance (Rct ∼30 Ω acidic, ∼35 Ω alkaline for the HER; ∼20 Ω for the OER), and the double-layer capacitance (Cdl) reached 21.1 mF cm−2, indicating a high electrochemical active surface area. XPS analysis showed Mn2+/Mn4+ states and Co–P bonding, while N-doping enhanced electronic conductivity and optimized the adsorption energies.89
Likewise, Xia Shi and co-workers developed Co–Mo2N hollow tube catalysts using a “MOFs plus MOFs” strategy for alkaline (1.0 M KOH) water splitting. One-dimensional Mo–MOFs were first synthesized by reacting imidazole with MoO3, and then coated with ZIF-67 (Co source) via solution-phase assembly, leveraging their ligand compatibility. Upon nitridation, ZIF-67 converted to metallic Co and Mo–MOFs converted to hollow Mo2N and, formed a Co–Mo2N hybrid with a tunable Co/Mo ratio. SEM and TEM confirmed rod-shaped Mo–MOFs decorated with ZIF-67 polyhedrons, while BET analysis showed a high surface area (∼1143 m2 g−1). Electrochemical tests revealed excellent HER performance with an overpotential of −76 mV at 10 mA cm−2, a Tafel slope of 47 mV dec−1, an exchange current density of −0.52 mA cm−2, a TOF of 0.396 s−1, and a FE of 100%. For the OER, the catalyst required 302 mV at 10 mA cm−2 with a Tafel slope of 90 mV dec−1 and an ECSA of 42.8 mF cm−2, and maintained stability for 48 h (HER) and 24 h (OER). A two-electrode Co–Mo2N∥Co–Mo2N system delivered overall water splitting at 1.576 V with 40 h stability (Fig. 8). DFT and XPS data indicated that electron-rich Mo optimizes H adsorption (ΔG* near zero), and late transition metals (LTMs) modulate Mo2N's electronic structure to further lower ΔG*. The hollow, rough-surfaced architecture increases active site exposure and promotes gas release, while Mo2N's excellent conductivity enables fast electron transfer. Scanning Kelvin Probe (SKP) measurements confirmed a work function of 5.56 eV, comparable to that of Pt (5.60 eV), supporting its high catalytic potential.90 Similarly, Jianrui Sun et al. synthesized trimetallic Fe/Ni/Co phosphides (FeCoNiP@NC) via a crystallization method using ZIF-67/GO precursors, forming heterojunctions between FeP, CoP, and Ni2P supported on graphene. Structural analyses (XRD, SEM, TEM, and XPS) confirmed well-dispersed <10 nm phosphide nanoparticles. The catalyst achieved low HER overpotentials of −93 mV (0.5 M H2SO4) and −187 mV (1.0 M KOH), and an OER overpotential of 266 mV (1.0 M KOH) at 10 mA cm−2. A FeCoNiP@NC/Ni two-electrode system delivered 10 mA cm−2 at 1.73 V for overall water splitting with good 10 h stability. The combined roles of FeP and CoP enhanced the HER, while CoP mainly drove the OER, with graphene providing structural support and stability.91
![]() | ||
| Fig. 8 (A) Schematic synthesis of a Co–Mo2N hybrid; (B) XRD, SEM, and TEM of ZIF-67/Mo–MOFs; (C) HER performance: polarization, overpotentials, Tafel plots, stability, and EIS; (D) OER performance: LSV, overpotentials, Tafel plots, and durability; (E) overall water splitting: LSV, cycling, and long-term stability compared to those of Pt/C∥RuO2 [reproduced from ref. 90 with permission from the Royal Society of Chemistry]. | ||
Liang and co-workers developed bifunctional Co–NC@Mo2C composites, where cobalt nanoparticles embedded in nitrogen-doped carbon (Co–NC) are reverse-encapsulated by molybdenum carbide (Mo2C), forming a carbon-based framework with outstanding catalytic efficiency for overall EWS. The synthesis involved preparing ZIF-67 by mixing Co(NO3)2·6H2O and 2-methylimidazole in methanol, followed by annealing with ammonium molybdate tetrahydrate at 700 °C under argon for 3 h. Structural characterization by XRD, XPS, and TEM confirmed the formation of a hollow, well-integrated composite, where the Mo2C shell provides both protection against electrolyte corrosion and additional catalytic sites. In 1.0 M KOH, Co–NC@Mo2C showed excellent HER and OER performance, with low overpotentials of −99 mV and 347 mV at 10 mA cm−2, and Tafel slopes of 65 and 61 mV dec−1, respectively. Impressively, it maintained good HER activity in acidic media (−143 mV at 10 mA cm−2) and delivered an overall water-splitting cell voltage of just 1.685 V at 10 mA cm−2, comparable to those of IrO2–Pt/C systems, while retaining ∼79% activity after 20 h (Fig. 9). Mechanistically, the intimate Co–Mo2C interface facilitates charge transfer, improves intrinsic activity, and ensures long-term durability, demonstrating the promising potential of MOF-derived reverse-encapsulation strategies for designing efficient, non-precious bifunctional electrocatalysts.92 Likewise, Pan et al. designed a bifunctional electrocatalyst consisting of cobalt phosphide (CoP) nanoparticles embedded within nitrogen-doped carbon nanotube hollow polyhedra (CoP/NCNHP), synthesized from a core–shell ZIF-8@ZIF-67 precursor through a controlled pyrolysis, oxidation, and phosphidation process. This unique architecture combines the high catalytic activity of CoP with the excellent conductivity and protective features of the porous N-doped carbon framework, creating abundant accessible active sites and facilitating efficient charge transfer. Electrochemical testing revealed outstanding performance, with overpotentials of −140 mV (HER, 0.5 M H2SO4), −115 mV (HER, 1.0 M KOH), and 310 mV (OER, 1.0 M KOH) at 10 mA cm−2. The catalyst also achieved a low overall water-splitting cell voltage of 1.64 V, maintaining stable operation over 36 hours (Fig. 10). DFT calculations indicated that the N-doped carbon enhances Co d-orbital electron density, optimizing hydrogen adsorption for the HER, while the CoP/CoOOH interface plays a crucial role in OER activity. This work demonstrates the potential of MOF-derived hybrid materials for creating durable, high-performance bifunctional electrocatalysts.93
![]() | ||
| Fig. 9 (A) HER performance in 1.0 M KOH and 0.5 M H2SO4: polarization curves, Tafel plots, long-term stability, and cycling; (B) OER and overall water-splitting performance: polarization, Tafel curves, stability, and LSV; (C) gas collection setup and time-resolved measurements of H2 and O2 evolution in 1.0 M KOH, including device and bubble images, and quantified gas production over time [reproduced from ref. 92 with permission from Elsevier, copyright 2019]. | ||
![]() | ||
| Fig. 10 (A) Schematic of CoP/NCNHP synthesis; (B) LSV, Tafel plots, and stability tests for the HER (0.5 M H2SO4, 1.0 M KOH) and OER (1.0 M KOH); (C) two-electrode water-splitting performance, chronopotentiometry, digital images of gas evolution, and DFT results including DOS, charge density, and HER free-energy diagrams [reproduced from ref. 93 with permission from the American Chemical Society, copyright 2018]. | ||
In addition, Wenxia Chen et al. reported a bifunctional CoPO/NF (cobalt hollow phosphorus polyhedral nanostructures on Ni foam) catalyst synthesized by transforming ZIF-67/Ni foam through high-temperature annealing, Ar–N2 plasma treatment with a phosphorus precursor, and subsequent oxygen incorporation at lower temperatures. This process yielded a hollow, porous Co–P nanocage structure with a high surface area (BET: 1530 m2 g−1), abundant active sites (ECSA: 11.4 mF cm−2, ∼3.8 times higher than ZIF-67), and reduced charge-transfer resistance (18.8 Ω). In 1.0 M KOH, CoPO/NF exhibited excellent HER and OER performance, requiring overpotentials of only −105 mV and 275 mV at 10 mA cm−2, with Tafel slopes of 48 and 52 mV dec−1, respectively. The catalyst achieved H2 and O2 production rates of 1031 and 515 μmol h−1 with nearly 100% FE and demonstrated stable operation over 45 hours. DFT calculations indicated significant charge transfer from CoO to phosphorus, generating electron-rich P regions that enhanced water adsorption (−1.18 eV), raised the Fermi level, and produced an optimal hydrogen adsorption free energy (ΔGH* ≈ 0 eV) (Fig. 11). Post-reaction XPS analysis showed some oxidation of Co to Co3O4 and partial phosphorus loss, although the catalyst maintained its structural integrity and catalytic performance.94
![]() | ||
| Fig. 11 (A) Synthesis of the CoPO/NF bifunctional electrocatalyst; (B) OER performance: polarization curves, Tafel slopes, stability, chronoamperometry, O2 generation, and FE; (C) HER performance: polarization, Tafel slopes, stability, Cdl comparison, H2 generation, and FE; (D) bifunctional water splitting: device setup, overall performance, gas evolution, and current stability; (E) formation mechanism of CoPO; (F) mechanistic insights from atomistic models: optimized structures, charge distributions, water adsorption energies, density of states, and HER free energy profiles [reproduced from ref. 94 with permission from the Royal Society of Chemistry]. | ||
Using a direct selenization route, Meng et al. fabricated a bifunctional catalyst (Co0.85Se@NC) by thermally converting ZIF-67 into a composite of ultrafine Co0.85Se nanoparticles which are uniformly confined within a nitrogen-doped carbon matrix. The resulting material retained the parent MOF's polyhedral geometry, developed hierarchical mesopores (2–20 nm), and achieved a surface area of 55 m2 g−1. Structural analyses confirmed the formation of Co–Nx moieties, which played a key role in enhancing the catalytic activity. Electrochemical tests in alkaline medium (1.0 M KOH) showed excellent OER performance, with an onset potential of 1.49 V, a low overpotential of 320 mV at 10 mA cm−2, and a Tafel slope of 75 mV dec−1, along with a FE of 97.5% and long-term operational stability over 1000 cycles. In HER studies, the same material delivered an overpotential of −240 mV at 10 mA cm−2 and a Tafel slope of 135 mV dec−1. When assembled into a symmetrical electrolyzer configuration, Co0.85Se@NC required just 1.76 V to achieve 10 mA cm−2, maintaining steady operation for 35 hours. The enhanced activity and durability were attributed to the synergistic interaction between nanoscale Co0.85Se and the conductive N-doped carbon framework, combined with high nitrogen incorporation (5.26 wt%) and electronic modulation through Co–Nx coordination.95Table 2 summarizes reported Earth-abundant metal (Mn, Fe, Co, Ni, and Mo) complexes and MOF-derived systems for the HER, OER, and bifunctional overall water splitting.
| Catalyst | Electrolyte | HER (η10, mV) | OER (η10, mV) | Cell voltage (V) | Ref. |
|---|---|---|---|---|---|
TImCoPc/KB (3.5 : 1.5) |
0.5 M H2SO4 | −108 | — | — | 52 |
| NENU-500 | 0.5 M H2SO4 | −237 | — | — | 96 |
| AB&CTGU-5 | 0.5 M H2SO4 | −44 | — | — | 97 |
| MoC@GS(700) | 1.0 M KOH | −77 | — | — | 98 |
poly[CoOTPc] + KB (3.5 : 1.5) |
0.5 M H2SO4 | −79 | — | — | 3 |
| UiO-66–NH2–Mo-5 | 0.5 M H2SO4 | −200 | — | — | 99 |
| Pd(II)TNPc + KB | 0.5 M H2SO4 | −187 | — | — | 100 |
| Co–N–GA | 1.0 M PBS | −299 | — | — | 101 |
| NiS2 HMSs | 1.0 M KOH | −219 | — | — | 102 |
Poly[Co(II)THTPc] : KB (4 : 1) |
1.0 M KOH | — | 359 | — | 103 |
| Mn0.52Fe0.71Ni–MOF-74 | 1.0 M KOH | 267 (at η100) | — | 104 | |
| Ni/FeVAPc | 1.0 M KOH | — | 312 | — | 105 |
| CoMM | 1.0 M KOH | 351 | 106 | ||
| CoTTPc/MWCNTs | 1.0 M KOH | 305 | 107 | ||
| Co–MOF | 1.0 M KOH | 280 | — | 108 | |
CoOBrPc + KB (4 : 1) |
1.0 M KOH | — | 381 (GCE) | 62 | |
| 330 (NF) | |||||
| HQCoPc + KB | HER: 0.5 M H2SO4 | −76 (onset) | 360 | — | 109 |
| OER: 1.0 M KOH | |||||
The rGO : FeSPc/GCE |
HER: 0.5 M H2SO4 | −93 | 350 | — | 110 |
| OER: 1.0 M KOH | |||||
| Co/Co9S8@ SNGS-1000 | 0.1 M KOH | −350 (at η20) | 290 | 1.58 (at η20) | 81 |
| 3D-CNTA | 1.0 M KOH | −185 | 360 | 1.68 | 111 |
| Co0.85Se@NC | 1.0 M KOH | −230 | 320 | 1.76 | 95 |
| MSZIF-900 | HER: 0.5 M H2SO4 | −233 | 337 | — | 112 |
| OER: 1.0 M KOH | |||||
| Ni3ZnC0.7-550 | 1.0 M KOH | −93 | 320 | 1.65 | 113 |
| (Fe0.1Ni0.9)2P(O)/NF | 1.0 M KOH | −87 | 240 (at η100) | 1.50 | 83 |
| UTBrImPc–MWCNT | HER: 0.5 M H2SO4 | −15 (onset) | 368 | — | 11 |
| OER: 1.0 M KOH | |||||
| NiFe–MOF array | 0.1 M KOH | −134 | 240 | 1.55 | 82 |
| CoS2 NTA/CC | 1.0 M KOH | −193 | 276 | 1.67 | 114 |
| CoSe2/CF | 1.0 M KOH | −95 | 297 | 1.63 | 115 |
| NiFe–Se/C | 1.0 M KOH | −160 | 240 | 1.68 | 116 |
| (Ni,Co) Se2/C–HRD | 1.0 M KOH | −87 | 245 | 1.58 | 117 |
Furthermore, the integration of in situ spectroscopic tools reveals real-time structural dynamics under electrochemical conditions. In an impressive example, Jia et al. created a hybrid catalyst by decorating a cobalt-based metal–organic framework (Co–MOF-74) with cobalt phthalocyanine (CoPc), resulting in a composite known as Co–MOF-74@CoPc. The production of high-valent cobalt species, which are normally thermodynamically disfavoured under OER conditions, was greatly accelerated by this π-conjugated molecular modification technique. With a robust durability of over 210 hours in 1.0 M KOH and a low overpotential of 291 mV at 10 mA cm−2, the resultant hybrid demonstrated exceptional electrocatalytic performance. Strong electronic coupling between CoPc and the Co–MOF scaffold was confirmed by XPS and XAS, with the former modifying the local coordination environment surrounding Co centers. Interestingly, in situ Raman spectroscopy showed that CoPc helped the MOF change into α-Co(OH)2 at low potentials, which then changed into disordered CoOOH and Co(IV)O2 species when anodic polarization occurred. Dynamic reconstruction and catalytic activity were directly correlated by spectroscopically confirmed structural transitions. This study emphasizes the crucial role of in situ spectroscopy in revealing real-time active-site evolution.118 Similarly, Linke et al. used in situ XAS and XRD to study Ni–MOF-74 under electrochemical OER conditions, offering a complementary perspective on catalyst reconstruction. They discovered that during anodic polarization, Ni–MOF-74 experiences irreversible amorphization, resulting in the formation of an active NiOOH–metal organic phase (Ni–MOC*), which is responsible for the observed catalytic enhancement. A progressive oxidation of Ni centers identified by XAS and a loss of crystallinity in real-time XRD verified the transformation. Surprisingly, the reconstructed phase maintained operational stability for more than 100 hours at 500 mA cm−2 in an AEM electrolyzer and delivered a high current density of 14 A g Ni−1 at 1.5 V vs. RHE. This clear link between structural reorganization and electrochemical activation supports the usefulness of in situ spectroscopy in determining the functional states of MOF-based electrocatalysts and identifies amorphization as a feasible method for producing catalytically competent phases.119
A compelling extension of such in situ investigations was provided by Wu et al., who developed a MOF-derived bifunctional electrocatalyst (C@NiCoP/NF) for efficient urea-assisted hydrogen production. A Ni–Co bimetallic phosphide composite (C@NiCoP/NF) with superior electrical conductivity and hierarchical porosity was produced by phosphidating ZIF-67 grown on nickel foam and then coating it with carbon. In situ spectroscopic investigations and DFT analyses verified the dynamic surface reconstruction of NiCoP into NiOOH and Co(OH)2 during the urea oxidation reaction (UOR) and HER, forming the real catalytically active phase. In addition to increasing the electrochemical active surface area and improving the adsorption–desorption behaviour of urea-derived intermediates, the reconstructed surface also improved HER kinetics by optimizing H adsorption energies* and enhancing water molecule activation. With the catalyst requiring only 1.34 V to reach 100 mA cm−2 for the UOR and exhibiting a low HER overpotential of 168 mV at −100 mA cm−2, the synergistic effects allowed for superior bifunctional performance. Compared to traditional OER-driven electrolysis, C@NiCoP/NF∥C@NiCoP/NF required only 1.51 V to deliver 100 mA cm−2 in a full cell consisting of a urea-assisted water splitting setup. By highlighting the importance of electrochemical reconstruction as a catalyst design strategy in dual-function electrolyzers and offering mechanistic clarity into the operando formation of NiOOH/Co(OH)2, this work emphasizes the significance of in situ spectroscopy in revealing transient phase transformations and active site evolution in MOF-derived metal phosphides.120 Collectively, these studies demonstrate the revolutionary potential of in situ spectroscopy in revealing the electronic evolution and dynamic restructuring of electrocatalysts under operating conditions which is frequently unavailable using static ex situ techniques. These methods are still underutilized in molecularly defined metal-complex catalysts for water electrolysis, where structural flux and low intermediate lifetimes present particular difficulties, despite their success in MOF-derived systems. To close this gap, a concerted effort must be made to develop multimodal, operando techniques that combine computational, spectroscopic, and electrochemical tools to reveal active-site dynamics and guide the predictive design of catalysts of the future.
In addition, despite the great value of DFT in clarifying adsorption energetics, PCET kinetics, and d-band tuning techniques, it has not been widely incorporated into experimental procedures. Data-driven, feedback-optimized design cycles are not possible because theoretical predictions are typically separated from material synthesis and testing. Combining computational and experimental frameworks to close this gap could greatly speed up catalyst discovery and logical optimization. Apart from mechanistic clarity, stability and scalability problems interfere with the practical implementation of these materials. Even though a lot of systems show promise in the short term, they frequently degrade over time, especially at high current densities that are pertinent to industrial water electrolysis. Long-term durability is compromised by irreversible phase changes, ligand dissociation, and metal leaching. However, the synthetic procedures for these complexes often entail complex, multi-step processes that are not suitable for large-scale production, such as template-directed growth, high-temperature pyrolysis, or multi-day solvothermal procedures. Addressing these issues in the future will call for a diversified approach. This includes adopting scalable synthetic routes that are compatible with industrial processes, expanding compositional space to include underutilized metals like Mn and Mo, and using operando characterization extensively to track active-state transitions. Furthermore, design rules that go beyond trial-and-error discovery can be unlocked by combining data-driven catalyst screening with high-throughput DFT modelling. For laboratory-scale advancements to be translated into reliable, inexpensive, and scalable technologies for sustainable hydrogen production, these approaches must converge.
| This journal is © The Royal Society of Chemistry 2025 |