Electronic engineering of spinels for advanced electrocatalysis

Abstract

Spinels represent promising candidates for clean energy electrocatalysis due to their abundance and electronic structure adjustability. However, their intrinsic catalytic activity remains limited. This review analyzes the fundamental correlations between the electronic structure and catalytic performance of spinel-based electrocatalysts. It elucidates the critical roles of coordination geometry, e.g., tetrahedral vs. octahedral sites, and the electronic configuration of active metal centers, including the d-band center position and spin state. The applications of these electronic structure modulation strategies across electrocatalytic reactions, encompassing the oxygen evolution reaction, oxygen reduction reaction, hydrogen evolution reaction, nitrogen reduction reaction, nitrate reduction reaction, carbon dioxide reduction reaction, and urea oxidation reaction, were further analyzed. Gaining insights from these diverse reaction systems, this review proposes a generalizable design paradigm for efficient spinel electrocatalysts: coordination engineering–d-band center optimization–spin state modulation. Finally, challenges in electronic-state control and future research frontiers are outlined, providing a robust mechanistic framework for the rational design of spinel electrocatalysts for sustainable energy technologies.

---

1. Introduction

The imperative shift towards sustainable energy systems hinges critically on efficient electrochemical conversions. Key electrocatalytic reactions – including the oxygen evolution (OER), oxygen reduction (ORR), hydrogen evolution (HER), nitrogen reduction (NRR), nitrate reduction (NO₃RR), carbon dioxide reduction (CO₂RR), and urea oxidation (UOR) reactions – represent vital pathways for renewable energy storage and mitigating fossil fuel dependence (Fig. 1).¹ However, their widespread implementation is hampered by inherent challenges such as sluggish kinetics, high overpotentials, and insufficient selectivity/efficiency. Although noble metal-based catalysts (e.g., Ru/Ir oxides, Pt) set performance benchmarks, their scarcity and prohibitive cost severely limit scalability.² Consequently, the development of efficient, durable, and Earth-abundant non-precious electrocatalysts is paramount.

Fig. 1 Schematic diagram for the crystal structure and applications of spinels in electrocatalysis.

Spinels (AB₂X₄; A, B = cations, X anions) emerge as a highly promising class of materials in this context. They exhibit the feature of easy structural regulation and can be constructed with various metals and non-metals (Fig. 1). Their distinctive crystal structure features a close-packed anion (X) sublattice with A cations typically occupying tetrahedral (Td) sites and B cations residing in octahedral (Oh) sites.³ This framework, coupled with extensive compositional flexibility at both A and B sites, enables precise tailoring of physicochemical properties. Leveraging their Earth abundance, structural robustness, and exceptional electronic tunability, AB₂X₄ spinels hold significant promise for diverse electrocatalytic applications. Indeed, spinel-based catalysts have demonstrated notable potential across the OER, ORR, HER, NRR, NO₃RR, CO₂RR, UOR, methanol oxidation,⁴ lithium-ion batteries,⁵ ammonium-ion batteries,⁶ water purification, luminescent materials and even photocatalysis.^7–19 However, their catalytic activity faces challenges such as low intrinsic activity, limited active centers, and poor electrical conductivity.²⁰ Common strategies focus on extrinsic modifications to mitigate these: (i) nanostructuring (e.g., nanosheets,²¹ nanowires,²² nanoflowers,²³ nanorods,²⁴ and nanocubes²⁵) to maximize the electrochemically active surface area (ECSA); (ii) integration with conductive substrates (e.g., carbon cloth,²⁶ doped carbons,²⁷ MOFs,²⁸ and Ni foam²⁹) to improve charge transfer kinetics; and (iii) electronic structure and catalytic pathway optimization through defect engineering, doping heteroatoms, etc. While being effective in boosting apparent performance, these approaches often circumvent the fundamental challenge: optimizing the intrinsic activity of the metal centers.

The core activity limitation of spinels resides in the adsorption/desorption energetics of key intermediates at the active sites. These energetics are fundamentally governed by the local electronic structure of the metal center within their ligand field. Understanding and strategically manipulating key electronic descriptors is therefore essential: (1) the d-band center position, critically controlling adsorbate–catalyst bond strength;³⁰ (2) the electronic configuration and spin state, which directly modulates the interaction strength between the metal center and reactive intermediates;^31–33 and (3) the A–B site interplay, influencing cation oxidation states, site distribution, and charge transfer dynamics.³⁴ Although targeted modulation of these factors, such as d-band shifts,³⁰ e_g filling control,^31,32 and bond engineering,³³ has yielded enhanced activities, a systematic and unified understanding of how the electronic structure of spinels dictates catalytic performance across the OER, ORR, HER, NRR, NO₃RR, CO₂RR and UOR remains fragmented.³⁵ Establishing robust, electronic structure-based structure–property–activity relationships is critically needed to advance rational catalyst design.

Based on the abovementioned criteria, this review provides a comprehensive and in-depth analysis of the fundamental electronic structure principles underpinning spinel electrocatalysis, focusing on: (1) core bonding characteristics and the critical impact of coordination geometry (tetrahedral Td vs. octahedral Oh) on ligand fields and electronic states and (2) d-band center and spin state linking to adsorbate binding energetics. These universal electronic structure engineering strategies in the OER, ORR, HER, NRR, NO₃RR, CO₂RR, and UOR were analyzed. Based on the results, this review proposes a generalizable design paradigm for efficient spinel electrocatalysts: coordination engineering–d-band center optimization–spin state modulation. Finally, it outlines prevailing challenges and future research directions aimed at achieving precise control over spinel electronic states, thereby guiding the rational design of high-performance and non-precious electrocatalysts.

2. Electronic structure of spinel electrocatalysts

2.1 Coupling nature of tetrahedral and octahedral centers

Spinel compounds are characterized by the general formula AB₂X₄, where A represents divalent cations (e.g., Mg²⁺, Mn²⁺, Zn²⁺, Fe²⁺, Co²⁺, and Ni²⁺) occupying tetrahedral interstices, B represents trivalent cations (e.g., Al³⁺, Cr³⁺, Fe³⁺, Mn³⁺, and Co³⁺) occupying octahedral interstices, and X is primarily the O²⁻ or S²⁻ anion. The crystal structure is based on a cubic close-packed framework of anions (X), with cations ordered on tetrahedral (Td) and octahedral (Oh) sites. The Td and Oh polyhedra interconnect three-dimensionally via corner-sharing or edge-sharing, forming a highly symmetric skeleton (Fig. 2).^33,34 The cation occupation pattern defines the spinel type: in normal spinels, A²⁺ cations exclusively occupy Td sites, while B³⁺ cations exclusively occupy Oh sites. In inverse spinels, half of the B³⁺ cations occupy Td sites, while the A²⁺ cations and the remaining half of the B³⁺ cations jointly occupy Oh sites.

Fig. 2 Schematic diagram of the spinel crystal structure.

The functional properties of spinels, particularly their electrocatalytic performance, are profoundly dependent on the connectivity modes of Td and Oh polyhedra: (1) corner-sharing connectivity (Td–O–Oh): this is the primary linkage between Td and adjacent Oh polyhedra. Cations at the Td site (A/B) and Oh site (B/A) are indirectly coupled diagonally through the shared oxygen anion. This connectivity is crucial for maintaining long-range structural stability, effectively dispersing lattice stress, suppressing cation vacancies and lattice distortions, and providing a foundation for the structural integrity of the electrocatalytic surface.^33,34 More importantly, the Td–O–Oh pathway governs the super exchange interaction between cations in Td and Oh sites. This electronic coupling directly influences the material's band structure, degree of electron delocalization, and charge migration efficiency between active sites, serving as a key factor in regulating intrinsic electrical conductivity. (2) Edge-sharing connectivity (Oh–O–Oh): adjacent octahedra are directly connected by sharing two vertex anions (an edge), forming a continuous octahedral network permeating the crystal lattice.³⁵ This Oh–O–Oh connectivity is the core pathway for efficient electronic conduction and ionic diffusion. Edge-sharing significantly shortens the distance between adjacent Oh-site B³⁺ cations (or A²⁺/B³⁺ in inverse spinels), greatly facilitating electron conduction. This occurs via direct d–d orbital overlap or indirect B–O–B pathways, enabling rapid electron migration within the octahedral network, which is vital for charge transfer during electrocatalysis. Simultaneously, it provides low-energy-barrier paths for the bulk diffusion of O²⁻ ions or protons (H⁺), particularly crucial in reactions involving lattice oxygen, such as the OER.

Therefore, the electrocatalytic activities of spinels are intimately linked to their unique cation occupation and connectivity configurations: (1) active site property: electrocatalytic reactions (e.g., OER, ORR, and HER) typically preferentially occur on exposed Oh-site cations.^12,13 Inverse spinels (e.g., NiFe₂O₄ and CoFe₂O₄) often exhibit superior redox activity due to mixed valence states (A²⁺/B³⁺) or variable valence states (B³⁺/B⁴⁺) at the Oh sites. The distribution of cations across the Td/Oh sites directly determines the electronic structure of the active sites (e.g., d-orbital occupancy) and the adsorption strength of the intermediates in catalytic reactions.^16,17 (2) Electronic structure modulation: cations at the Td and Oh sites engage in intricate coupling via Td–O–Oh superexchange. This not only influences the band structure of spinels but also finely tunes the crystal field splitting energy (CFSE) and spin state of the active Oh sites. These parameters directly impact the strength and symmetry matching of interactions between the d-orbitals of the metal center and the molecular orbitals of reactants/intermediates, thereby determining the reaction activation energy. (3) Charge transfer and mass transport: the continuous electron conduction pathways provided by the Oh–O–Oh network ensure efficient charge compensation from the reaction interface to the bulk. Its open ionic diffusion channels facilitate the mass transport of reactants/products and help maintain a stable local chemical environment at the electrode surface under high overpotentials.

Therefore, precise tuning of the electrocatalytic activity, selectivity, and stability of spinels can be achieved by strategically controlling the type and valence state of A/B cations. The occupation of the Td/Oh sites in Td–O–Oh and Oh–O–Oh connectivity modes deeply affects the electronic structure, charge transport, ionic diffusion, and structural stability. This profound structure–property correlation establishes spinels as a versatile platform for designing high-performance electrocatalysts.

2.2 Electronic filling states and spin states of metal centers

The d orbital presents five spatial direction properties and the splitting of d-orbital energies in transition metal cations will induce ligand interactions within the octahedral, tetrahedral, and square planar coordination geometries (Fig. 3a and b). These d-orbital energies are related to the spin state of electrons in d orbitals. The spin state, including low-spin, intermediate-spin and high-spin, of a metal center is primarily determined by the relative magnitudes of the crystal field splitting energy (Δ) and the electron pairing energy (P) (Fig. 3c).³⁶ When Δ > P, electrons preferentially pair up to fill lower-energy orbitals, resulting in a low-spin state. Conversely, when P > Δ, electrons preferentially occupy more orbitals singly, leading to a high-spin state.³⁶ The intrinsic driving force for the high-spin state is the minimization of the system's total energy.

Fig. 3 Spin state of metal centers in a spinel structure. (a) Schematic diagrams of five d-orbitals. (b) Crystal field splitting of d-orbitals in the octahedral, tetrahedral, and square planar geometries. (c) d-orbital energy levels and possible spin states of Co³⁺ as low spin state, intermediate spin state, and high spin state in the octahedral sites of spinel.³⁸ Copyright 2024 Royal Society of Chemistry Publication.

In spinels (AB₂X₄), the electronic filling state of the metal cations is central to their chemical properties.³⁷ Cations located at the tetrahedral sites (A-sites, coordination number = 4) typically exhibit high-spin configurations for their d-electrons due to the smaller crystal field splitting energy (Δ_tetra). This results in higher d-electron occupancy, leading to weaker metal–ligand (e.g., O²⁻) orbital interactions and lower bond covalency. In contrast, cations at the octahedral sites (B-sites, coordination number = 6) experience a larger splitting energy (Δ_oct) and usually adopt low-spin configurations for their d-electrons. In this state, d-electron occupancy is lower, particularly in the higher-energy e_g orbitals. This favors stronger interactions between the metal d-orbitals and ligand orbitals (e.g., O 2p), significantly enhancing the covalency of the metal–ligand bond.^37,38 This enhanced covalency is crucial for material stability and reactivity. Common strategies to modulate metal spin states include altering lattice properties, crystallinity, introducing defects, and heteroatom substitution.³⁸

The d-electron filling property of B-site metal ions in spinels typically dominates their catalytic activity. The low-spin state at the B-site facilitates effective interaction of unpaired electrons with reactant molecules (e.g., O₂), promoting their activation and dissociation.³⁹ Furthermore, unpaired d-electrons also contribute to the material's electronic conductivity. Therefore, a profound understanding and strategic modulation of the spin states and electronic filling configurations of the metal centers in spinels, particularly at the B-sites, are paramount for optimizing their catalytic performance.³⁹ Selecting specific metal ions and controlling their distribution between the four-coordinate A-sites and six-coordinate B-sites provides an effective route to tune the overall catalytic activity and conductivity.

The covalency of the metal–ligand bond arises from direct orbital overlap and interaction between the d-orbitals of the metal cation (A or B) and the p-orbitals of the ligand anion (e.g., O²⁻).³⁰ At the octahedral B-sites, the d-orbitals split into lower-energy t_2g and higher-energy e_g orbitals, with the splitting energy (Δ_oct) depending on the metal ion charge and oxygen coordination. At the tetrahedral A-sites, the d-orbitals split into higher-energy e_g and lower-energy t_2g orbitals, with a smaller splitting energy (Δ_tetra).⁴⁰ Metal–ligand orbital coupling within these coordination environments effectively regulates the electronic filling state of the active centers, thereby: (1) facilitating charge transfer: charge transfer between metal d-orbitals and ligand p-orbitals (O 2p → M d or M d → O 2p) can lower activation energy barriers for reactions and enhance the adsorption of reactants on the surface,⁴⁰ and (2) generating synergistic catalytic effects: the four-coordinate A-sites and six-coordinate B-sites, through orbital coupling with oxygen, form active sites with distinct electronic characteristics. These sites can adapt to different reactant molecules, generating synergistic catalytic effects that achieve multi-functional reaction activities of the spinel.³ It is precisely based on these key mechanisms that the orbital coupling effect is extensively studied in the field of electrocatalysis.

2.3 Metal d-band center

The electronic structure at the catalyst's surface/interface is central to determining electrocatalytic reaction kinetics. In 1995, Jens K. Nørskov and B. Hammer pioneered the concept of the d-band center by correlating the position of the d-electron band center of transition metals with surface adsorption energies using density functional theory (DFT) calculations. The d-band center serves as a key descriptor for the electronic structure of transition metals, quantitatively revealing the mechanism of metal–adsorbate orbital hybridization and providing a theoretical foundation for the rational design of high-performance electrocatalysts. Its physical origin stems from the energy level matching between the transition metal d-orbitals and the molecular orbitals of reactants: when an adsorbate approaches the metal surface, the d-orbitals couple with the adsorbate, forming bonding states below the Fermi level and antibonding states above it (Fig. 4a).⁴¹ Consequently, modulating the d-band center position (ε_d) influences the occupancy of the antibonding states, thereby regulating the adsorption strength of active sites for reactive species: (1) ε_d shifts upwards → antibonding state energy level decreases → electron occupancy increases → adsorption strengthens. (2) ε_d shifts downwards → antibonding state energy level increases → electron occupancy decreases → adsorption weakens. This mechanism aligns with the volcano-shaped activity curve described by the Sabatier principle, where optimal catalytic activity requires moderate reactant adsorption energy (ΔE_ads), avoiding difficulties in product desorption caused by strong adsorption or insufficient reactant activation due to weak adsorption. Therefore, the d-band center acts as a bridge linking the electronic structure to catalytic activity by regulating adsorption bond strength.⁴² As illustrated in Fig. 4b, a higher d-band center position lowers the energy of the antibonding states relative to the Fermi level, leading to stronger adsorption.⁴³

Fig. 4 (a) Schematic illustration of the formation of a chemical bond between an adsorbate valence level and the s and d states of a transition-metal surface.⁴¹ Copyright 2011 National Academy of Sciences. (b) High spin states in tetrahedral coordination and low spin states in octahedral coordination.⁴³ Copyright 2024 Elsevier. (c) d orbital splitting in tetrahedral coordination and octahedral coordination.⁴⁴ Copyright 2024 Elsevier.

Several strategies exist for modulating the d-band center in spinels (AB₂X₄): (1) coordination field effect and cation distribution tuning: catalytic activities in spinels are predominantly governed by the octahedral B-sites. The d-orbitals at these sites split into lower-energy t_2g and higher-energy e_g orbitals. Shortening the B–O bond enhances the ligand field strength, increasing the t_2g–e_g energy gap, which drives ε_d downwards. (2) Heteroatom doping inducing electronic reconstruction: the electronegativity difference between dopant atoms (e.g., P, S, and F) and metal sites (e.g., Fe and Co) triggers localized charge transfer. This alters the hybridization nature between the dopant p-orbitals and the metal d-orbitals, promoting band structure reorganization and thereby modulating ε_d. Furthermore, doping establishes metal–heteroatom charge transfer pathways, optimizing the adsorption/desorption kinetics of intermediates (e.g., *OOH → O₂ in the OER). (3) Modulating electron spin state of metal centers: the spin state of transition metals (e.g., Fe³⁺ and Co³⁺) directly determines ε_d by influencing d-orbital splitting energy and electron occupancy. As shown in Fig. 4c,⁴⁴ doping can alter the charge distribution around Fe atoms due to electronegativity differences, affecting d-orbital filling and spin state, consequently tuning ε_d and optimizing electrocatalytic activity.^45,46

3. Practical applications of spinel electronic state modulation in electrocatalysis

Although fundamental research on spinel electronic structures—such as d-band center regulation, electron spin state modulation, and tuning of tetrahedral/octahedral metal centers—has made significant progress, inherent structural characteristics still pose critical bottlenecks. These include insufficient exposure of active sites, low electrical conductivity, and limited intrinsic activity, which severely constrain electrocatalytic performance. Consequently, developing effective strategies to modulate the electronic states of spinels has become a major research focus. Key strategies for modulating spinel electronic states, such as the coordination environment, valence state, and spin state, have been proposed. These encompass the OER, ORR, HER, NRR, CO₂RR, and UOR. This review offers insights for the precision design of high-performance spinel electrocatalysts from the perspective of electronic structure optimization.

3.1 Oxygen evolution reaction

The oxygen evolution reaction (OER), serving as the anode reaction for metal–air batteries and water electrolysis, represents a critical bottleneck limiting overall energy conversion efficiency due to its sluggish four-electron reaction kinetics.^47,48 Developing highly efficient and stable OER electrocatalysts is paramount to lowering the reaction energy barrier. Spinels are considered promising candidate materials owing to their tunable electronic structures. Therefore, researchers have proposed various electronic state modulation strategies to optimize the OER performance of spinels.

Defect engineering: introducing oxygen vacancies (O_V) into the spinel structure can break the local symmetry of the Oh sites, inducing an upward shift of the metal d-band center, thereby enhancing the adsorption of oxygen-containing intermediates (*OOH and *O). Liu et al.⁴⁹ fabricated amorphous/crystalline hybrid FeNi₂O₄ nanosheet arrays rich in O_Vs (FNO–A/C–OR/NF) (Fig. 5a). Results demonstrated that a high O_V concentration led to an increased Ni²⁺/Ni³⁺ and Fe²⁺/Fe³⁺ ratio, signifying a decrease in metal valence states and an increase in electron filling states. This increased electron filling for Ni and Fe lowered the d-band center. This optimized electronic structure enabled FNO–A/C–OR/NF to achieve an OER current density of 500 mA cm⁻² at a remarkably low overpotential of only 315 mV. DFT calculations further revealed that introducing O_V (C–FNO–O_V) significantly optimized the adsorption energies of oxygen-containing intermediates, positioning it closer to the apex in the overpotential–adsorption energy (ΔG_O–ΔG_OH) volcano plot and substantially boosting intrinsic activity (Fig. 5b).

Fig. 5 (a) EPR spectra.⁴⁹ (b) Volcano plots.⁴⁹ (a) and (b) Copyright 2024 Wiley Online Library. (c) Density of states of O 2p.⁵⁰ Copyright 2024 Elsevier. (d) Illustration of charge exchange mode of metal cations in Co₂FeO₄/(Co_0.72Fe_0.28)_Td(Co_1.28Fe_0.72)_OctO₄/NCNTs.⁸ Copyright 2024 Wiley Online Library. (e) Co valence state and (f) chronopotentiometric durability of spinels.⁵¹ Copyright 2025 American Chemical Society. (g) and (h) Fourier transformation curves of EXAFS.⁵² Copyright 2023 Wiley Online Library.

Atomic doping: incorporating specific elements enables fine-tuning of the electron density and orbital hybridization at metal sites. For example, the strong-field ligand effect of P³⁻ increases the crystal field splitting energy (Δ_oct) at Oh sites compared to initial O²⁻, promoting a metal spin state transition (HS → LS) and optimizing e_g orbital occupancy. Simultaneously, hybridization between P 3p and O 2p orbitals shifts the O-p band center upwards, facilitating the participation of lattice oxygen in the reaction (Lattice Oxygen Mechanism, LOM). Li et al.⁵⁰ introduced P into NiFe₂O₄ (P–NiFe₂O₄). P doping reduced the electron density around Ni and increased its valence state, strengthening the M–O covalent bond hybridization between Ni 3d and O 2p orbitals, thereby promoting the LOM pathway. DFT calculations showed that P doping shifted the O 2p band center from −3.92 eV in NiFeOOH to −3.62 eV in P–NiFeOOH, significantly lowering the reaction energy barrier for the LOM pathway (Fig. 5c). Consequently, P–NiFe₂O₄ required only 264 mV overpotential to reach 10 mA cm⁻², outperforming the pristine sample by ∼90 mV.

Multi-cation synergy: introducing multiple metal elements to form high-entropy spinels leverages the random distribution of diverse cations at the Oh sites, creating a continuous d-band. Liu et al. introduced Fe³⁺ into Co₃O₄ to create Co₂FeO₄/(Co_0.72Fe_0.28)_Td(Co_1.28Fe_0.72)_OctO₄ nanoparticles on N-doped carbon nanotubes (Co₂FeO₄/(Co_0.72Fe_0.28)_Td(Co_1.28Fe_0.72)_OctO₄/NCNTs) (Fig. 5d).⁸ They found that Fe³⁺ incorporation induces Co 3d electron delocalization and spin-state transition. This synergistic spin and charge effect enables Fe³⁺ ions to activate adjacent Co³⁺ sites, significantly boosting the intrinsic oxygen electrocatalytic activity of the hybrid Co₂FeO₄ spinel. Furthermore, the high-valence elements (e.g., W⁶⁺) can induce charge transfer via O-2p bridges, optimizing the electronic structure of adjacent metals and benefiting long-term stability of the OER. Peng et al. carried out W doping for regulating the electronic environment of NiCo₂O₄, which helps mitigate excessive oxidation and stabilizes the Co–O coordination (Fig. 5e).⁵¹ Such local electronic restructuring not only reinforces the structural framework but also promotes more reversible oxygen redox activity. Therefore, W–NiCo₂O₄ demonstrates exceptional operational durability in 0.5 M H₂SO₄, sustaining over 650 hours of continuous operation with a negligible potential degradation rate of 64.7 μV h⁻¹, significantly surpassing the performance of pristine NiCo₂O₄ and Co₃O₄ (Fig. 5f).

Anion regulation: substituting O²⁻ with S²⁻ reduces the crystal field strength (Δ_oct) at Oh sites, inducing a transition of the metal spin state towards high-spin (HS), thereby enhancing redox flexibility. Furthermore, sulfides undergo electrochemical reconstruction during the OER, forming oxysulfides that create mixed O/S coordination fields, ultimately boosting OER catalytic activity. Sun et al.⁵² synthesized NiCo₂S₄ by substituting O with S anions in NiCo₂O₄. During OER operation, NiCo₂S₄ underwent electrochemical reconstruction involving partial leaching of S from Co–S bonds, forming an amorphous oxysulfide rich in O–S co-regulation (NiCoS₄), where the Oh-field ligands transitioned from S to O. XANES spectra showed an increase in Co valence and a decrease in Ni valence after reconstruction, indicating electron transfer from Co to Ni, reflecting spin rearrangement and charge redistribution (Fig. 5g). EXAFS further confirmed the transformation from Co–S to Co–O bonds (S leaching at Co sites) while Ni–S bonds were partially retained (Fig. 5h). The unique gradient coordination field formed by this O–S heteroanionic structure endowed the reconstructed amorphous NiCoS₄ with exceptional OER performance (258 mV@100 mA cm⁻²) and stability.

3.2 Oxygen reduction reaction

The oxygen reduction reaction (ORR), a crucial cathode reaction in fuel cells and metal–air batteries, is severely constrained by its sluggish four-electron process, limiting battery efficiency.⁵³ Although noble metal-based catalysts exhibit outstanding ORR performance, their high cost and scarcity impede their large-scale application.⁵⁴ Consequently, developing transition metal spinel oxides with tunable electronic structures as efficient ORR catalysts has emerged as a significant research focus. This section analyzes the electronic structure modulation strategies of spinels for the ORR.

Atomic doping: introducing cations with different electronegativities into Oh sites can alter the crystal field splitting energy (Δ_oct), thereby inducing spin state transitions and charge redistribution, ultimately optimizing the adsorption of oxygen-containing intermediates (*O₂ and *OOH).^55,56 Zhao et al.⁵⁷ prepared Ni-doped Co₃O₄ nanosheets (NCO). Ni²⁺ ions, with their d⁸ electron configuration in a low-spin state, occupy Co³⁺ Oh sites. This enhances electronic interactions with neighboring Co³⁺ ions, boosting catalytic performance. As shown by in situ Raman spectra (Fig. 6a and b), only the characteristic peaks of NiOOH (474 and 552 cm⁻¹) appear for NCO at 1.4 V, while CoOOH peaks (700–1000 cm⁻¹) are absent, indicating that Ni sites are the primary OER active centers. Simultaneously, the material exhibits bifunctionality – the introduction of Ni²⁺ increases the average d-electron count at Oh sites, promoting side-on adsorption and dissociation of O₂, achieving an ORR onset potential of 0.92 V.

Fig. 6 (a) and (b) In situ Raman spectra of NCO-2 in 1 M KOH.⁵⁷ Copyright 2022 Elsevier. (c) ORR polarization curves. (d) Schematic diagram of tetrahedral and octahedral interactions and (e) the relationship for the d-band center and catalytic performance.²⁸ Copyright 2022 Wiley Online Library. (f) Relationships for ORR activity of spinels and the e_g occupancy of the octahedral site and (g) relationship for ORR activity of MnCo₂O₄ and the Mn valence state.³⁰ Copyright 2017 Wiley Online Library. (h) Schematic diagram of Mn atom occupancy for CoMn-1.2 and (i) EXAFS k3χ(R) spectra of Mn in CoMn spinel NCs.⁶⁰ Copyright 2024 Wiley Online Library.

Cation regulation: the d-band center theory identifies the d-band center position (ε_d) of metal sites as a key descriptor determining catalyst activity.^58,59 In spinels, the tetrahedral A-site (A_Td) influences the ε_d of the octahedral M-site (M_Oh) via corner-shared A_Td–O–M_Oh bonds, thereby modulating catalytic properties. For instance, Liu et al.²⁸ designed a series of nitrogen-doped carbon nanotube (NCNT)-supported ACo₂O₄ spinels (A = Mn, Co, Ni, Cu, and Zn) for the ORR. Among them, MnCo₂O₄/NCNTs exhibited the best performance with the limiting current density of J_L = −6.06 mA cm⁻² and half-wave potential of E_1/2 = 0.76 V (Fig. 6c). Mechanistic studies revealed that the tetrahedral A-site (A_Td) influences the electronic properties of the octahedral Co site (Co_Oh) through the oxygen bridge (A_Td–O–Co_Oh). DFT calculations confirmed that introducing a high-electronegativity A_Td cation (e.g., Mn) promotes charge transfer from Co_Oh to A_Td, reducing the density of unoccupied orbitals at Co_Oh and shifting ε_d closer to the Fermi level (E_f) (Fig. 6d and e). This enhances the adsorption of oxygen-containing intermediates of MnCo₂O₄/NCNTs, thereby significantly boosting ORR activity. Furthermore, the electron occupancy of the e_g or t_2g orbitals is a key determinant of catalytic activity. In a study by Xu et al.,³⁰ Mn-based spinels with controlled e_g orbital filling and Mn valence states were obtained (Fig. 6f and g). They demonstrated that an e_g occupancy of approximately 0.6 and an average Mn valence of +3.4 at the octahedral sites correlate with the optimal ORR performance.

Nano-size effects: the nanoscale size of spinels significantly impacts cation distribution and electronic structure, consequently regulating the ORR performance. Nanoscaling promotes the formation of metal-enriched octahedral sites (Oh), inducing an upward shift in the d-band center. Concurrently, the small size induces quantum confinement effects, broadening the d-band and enhancing charge transfer. For example, Shang et al.⁶⁰ synthesized CoMn₂O₄ nanocrystals of varying sizes (1.2 nm, 5.5 nm, 8.8 nm, and 50 nm). As the size decreased to 1.2 nm (CoMn-1.2), (1) the Mn 3s XPS spin–orbit splitting energy decreased (minimum 5.15 eV), indicating a significant increase in the average Mn valence state as the Mn⁴⁺ content increased from 0.7% to 27% and the Mn²⁺ content decreased from 24.9% to 5%. (2) Cation distribution changed: Mn showed a stronger preference for occupying Oh sites (Fig. 6h), leading to a reduced proportion of octahedral Co³⁺. (3) Size effects: quantum confinement effects arising from the small size caused an increase in the Mn valence state, optimized the coordination environment, and enhanced charge transfer capability (Fig. 6i). ORR free energy diagrams revealed that reactions involving oxygen intermediates on small nanocrystals were exothermic, accelerating reaction kinetics. Consequently, CoMn-1.2 exhibited an ORR half-wave potential of only 0.88 V. Furthermore, the crystal facet structure also critically influences catalytic performance. Zhou et al.⁶¹ reported CoMn₂O₄ nano-octahedrons exposing {101} facets, achieving a remarkably high mass activity of 60.0 A g⁻¹ at 0.85 V. This further confirms the strong correlation between the ORR activity and the valence state/distribution of Mn at the octahedral sites.^62,63

3.3 Hydrogen evolution reaction

Hydrogen, as a clean energy carrier with high energy density and zero pollution, requires efficient production methods. The electrocatalytic hydrogen evolution reaction (HER) is an effective pathway to obtain high-purity hydrogen.^64,65 However, although noble metal-based catalysts exhibit excellent performance, their high cost and stability issues limit their large-scale application.⁶⁶ Spinels offer a route to overcome the cost limitations of noble metal catalysts by optimizing the hydrogen adsorption free energy (ΔG_*H) through electronic structure modulation of octahedral sites (Oh).⁶⁷ Researchers are dedicated to enhancing the HER performance of spinels via electronic structure tuning.

Atomic doping: introducing non-metallic elements can regulate the electron density and coordination environment of metal sites, optimizing ΔG_*H. For instance, the strong-field ligand nature of P³⁻ increases the crystal field splitting energy (Δ_oct), forcing Fe_Oh (d⁶) to transition from a high-spin (HS, S = 2) to a low-spin (LS, S = 0) state. This optimizes the σ-bonding capability of the t_2g orbitals with the H 1s orbital. Consequently, Zhang et al.⁶⁸ fabricated P-doped inverse spinel P–Fe₃O₄ (P–Fe₃O₄/IF), achieving a HER current density of 100 mA cm⁻² at a low overpotential of only 138 mV. Mechanistic studies revealed that the strong electronic interaction between Fe and P promotes the preferential occupation of octahedral sites by Fe²⁺ (d⁶) in a low-spin state, enhancing electrical conductivity and lowering the water dissociation energy barrier (Fig. 7a). Furthermore, P substitution for O reduces the electron density around tetrahedral Fe, optimizing ΔG_*H at octahedral Fe and reducing the reaction energy barrier from 0.61 eV to −0.20 eV, while simultaneously mitigating the risk of catalyst poisoning due to strong hydrogen bonding adsorption (Fig. 7b). Zhang's work demonstrates that non-metal doping can construct efficient non-metal–metal dual active sites. Moreover, introducing metal elements can optimize the adsorption kinetics of reaction intermediates. Wang et al.⁶⁹ employed a bulk and surface cooperative Ru doping strategy to modify NiCo₂O₄ (NCRO). Ru⁴⁺ (4d⁴) partially substitutes Co_Oh sites. Its shallow d-orbitals cause an upward shift in the d-band center (ε_d), enhancing orbital overlap with H*, optimizing the adsorption kinetics for the H intermediate, and accelerating the Volmer step (H⁺ + e⁻ → H). This enables NCRO-4 to achieve 100 mA cm⁻² at only 138 mV overpotential.

Fig. 7 (a) The related kinetic energy barriers for water dissociation. (b) Calculated free-energy diagram of the HER.⁶⁸ Copyright 2019 Wiley Online Library. (c) Nyquist plots. (d) Current density as a function of scan rate of CV curves.⁷⁰ (a)–(d) Copyright 2023 Wiley Online Library. (e) DOS curves.⁷¹ Copyright 2018 ACS Publication. (f) Chronoamperometric response of S–NiFe₂O₄ for the HER.⁷² Copyright 2021 Wiley Online Library.

Constructing heterointerfaces: building crystalline/amorphous heterointerfaces can induce the generation of oxygen vacancies (O_V). The formation of O_V reduces the coordination number at Oh sites, promoting d-band broadening and enhanced spin polarization, thereby optimizing electrocatalytic activity. Wang et al.⁷⁰ constructed NiCo₂O₄ nanosheets with crystalline/amorphous heterointerfaces on carbon cloth (NiCo₂O₄-B-CC). This catalyst required only 26 mV overpotential to reach 10 mA cm⁻². The heterointerface led to a high O_V concentration, resulting in: (1) increased content of low-spin Co²⁺ (d⁷) and low-spin Ni²⁺ (d⁸); (2) significantly reduced charge transfer resistance (9.93 Ω); (3) increased electrochemical active surface area (ECSA) (Fig. 7c and 7d); and (4) enhanced stability of the key intermediate *H, promoting HER kinetics. Furthermore, Peng et al.⁷¹ reported necklace-like hollow NiCo₂O₄ (R-NCO) with abundant O_V. R-NCO exhibited a low HER onset potential of 90 mV. DFT calculations showed that the reduction treatment shifted the projected density of states (PDOS) of Co d-orbitals towards lower energy and broadened the d-band peak, increasing Co's spin polarization and making it a more active catalytic center (Fig. 7e).

Anion substitution: introducing weak-field ligands, such as the S²⁻ anion, can regulate the spin rearrangement of metal centers. For example, Jin et al.⁷² synthesized S–NiFe₂O₄ by filling the O_V in inverse spinel NiFe₂O₄ with S atoms. This catalyst required only 61 mV overpotential at 10 mA cm⁻² and demonstrated excellent stability at a high current density (Fig. 7f). Results indicated that S²⁻ filling the oxygen vacancy formed Fe_Oh–S sites. The weak-field ligand character of S reduced Δ_oct, promoting the transition of Fe³⁺ (d⁵) from the LS (S = 1/2) to the HS (S = 5/2) state, thereby enhancing redox activity. This strategy provides a novel approach for optimizing the intrinsic activity of spinels via vacancy.

3.4 Nitrogen reduction reaction

Ammonia (NH₃) is an indispensable basic chemical in agriculture, chemical industries, and manufacturing. The industrial Haber–Bosch process for ammonia synthesis operates under harsh conditions (400–550 °C, 5–30 MPa), consuming substantial energy and generating significant greenhouse gas emissions.^73–75 Electrocatalytic nitrogen reduction (NRR), converting atmospheric N₂ to NH₃ under ambient conditions, is regarded as a promising alternative to the Haber–Bosch process.⁷⁶ However, the activation of N₂ is challenging due to the high dissociation energy of the N≡N bond (941 kJ mol⁻¹). Furthermore, the NRR involves a multi-electron–proton transfer process, resulting in sluggish reaction kinetics. These factors lead to challenges in the NRR, including a low NH₃ yield and insufficient faradaic efficiency.^76,77 Therefore, spinels, through electronic structure engineering of octahedral sites (Oh) to optimize N₂ adsorption and activation, show significant promise in NRR catalyst design.

Atomic doping: introducing large-radius alkaline earth metal ions into spinels can induce lattice expansion and electronic structure reconstruction, enhancing N₂ activation capability. Jiang et al.⁷⁸ reported Sr-doped CoFe₂O₄ (Sr_0.3Co_0.7Fe₂O₄). Doping with large-radius Sr²⁺ induced lattice expansion, reducing the crystal field strength (Δ_oct) at Oh sites and maintaining the high-spin state (HS, S = 5/2) of Fe³⁺. Concurrently, strong Sr–O–Fe bonding facilitated electron injection into the d-orbitals of Fe_Oh, causing a downward shift of the d-band center (ε_d). This optimized the material's ability to adsorb and activate N₂ molecules. The origin of NH₃ from N₂ reduction was confirmed via ¹⁵N₂ isotopic labeling experiments (¹H NMR spectroscopy) (Fig. 8a). This catalyst achieved a high NRR NH₃ yield rate of 36.4 μg h⁻¹ mg⁻¹ and a faradaic efficiency of 76.7% in an acidic electrolyte (Fig. 8b). N₂ temperature-programmed desorption (N₂-TPD) revealed broader desorption peaks (>370 °C), indicating that Sr doping exposed more strong adsorption active sites and lowered the reaction energy barrier (Fig. 8c and d). The catalyst demonstrated stable performance over multiple cycles at −0.3 V vs. RHE (Fig. 8b).

Fig. 8 (a) ¹H NMR spectra of both ¹⁴NH₄⁺ and ¹⁵NH₄⁺. (b) Reusability test. (c) TPD-MS profiles of N₂ (m/z = 28). (d) Free energy diagram for the NRR.⁷⁸ Copyright 2022 Elsevier. (e) Fourier-transformed EXAFS spectra.⁷⁹ Copyright 2021 ACS Publication.

Defect engineering: introducing highly active noble metal single atoms combined with oxygen vacancies (O_V) can synergistically optimize N₂ adsorption sites and the electronic environment. Lee et al.⁷⁹ developed O_V-rich Ru single-atom doped Co₃O₄ (Ru_1.4Co₃O_4−x). This catalyst exhibited a high NH₃ yield rate (39.4 μg h⁻¹ mg⁻¹) and faradaic efficiency (40.2%). Mechanistic analysis revealed that Ru single-atom doping led to an increase in the average Co–O bond length and significant lattice distortion (Fig. 8e), reducing Δ_oct. This structural change, combined with the high activity of Ru atoms and the presence of O_V, collectively provided abundant Lewis acid sites. Ru–Co d-band hybridization increased the density of unoccupied states, enhancing the occupancy of N₂ antibonding orbitals and resulting in a 6.3-fold increase in adsorption capacity.

3.5 Nitrate reduction reaction

The electrocatalytic nitrate reduction reaction (NO₃RR) offers a sustainable pathway for synthesizing high-value-added green NH₃ while simultaneously removing nitrate pollutants from water, presenting an alternative to the energy-intensive Haber–Bosch process.^80,81 However, the NO₃RR involves a complex multi-electron–proton transfer process (NO₃⁻ → NH₄⁺) and faces intense competition from the hydrogen evolution reaction (HER), leading to low ammonia yield and faradaic efficiency. Consequently, developing highly active and selective electrocatalysts is crucial.⁸² Spinel oxides demonstrate significant potential in the NO₃RR due to their tunable cation valence distribution, flexible defect chemistry, and good stability. Researchers have significantly enhanced their performance through various electronic state and structural modulation strategies.

Atomic doping: introducing highly active single atoms can precisely modulate the local electronic environment of spinels, optimizing intermediate adsorption. Guan et al.⁸³ doped Ru single atoms into hollow Co₃O₄ (Ru₁–Co/HCO), forming atomically dispersed Ru–Co pairs. This catalyst achieved a maximum faradaic efficiency of ∼100% and an NH₃ yield rate of 7.02 mg h⁻¹ mg⁻¹. XANES analysis indicated that Ru doping increased the valence state of Co and thereby the upward shift of Co's ε_d (Fig. 9a). This unique Ru–Co d-band hybridization induced local charge rearrangement and electronic structure modifications, significantly promoting the hydrogenation of reaction intermediates (e.g., *NO₂⁻) and their conversion to NH₃, resulting in exceptional NO₃RR performance.

Fig. 9 (a) Co K-edge normalized XANES spectra.⁸³ Copyright 2024 Elsevier. (b) Free energy diagrams for the NO₃RR to NH₃.⁸⁴ Copyright 2022 Elsevier. (c) Performance descriptors in CO₂ hydrogenation on different Co-based catalysts.⁸⁸ Copyright 2024 Elsevier. (d) Co 2p and (e) Mn 2p XPS spectra.⁸⁹ Copyright 2019 Elsevier.

Supporter engineering: loading spinels onto highly conductive substrates effectively addresses their intrinsic conductivity limitations and leverages the confinement effect to optimize the reaction microenvironment. Niu et al.⁸⁴ supported CuCo₂O₄ spinel on porous carbon nanofibers (CuCo₂O₄/CFs), achieving an NH₃ yield rate of 394.5 mmol h⁻¹ g⁻¹ and a faradaic efficiency of 83.6%. The CF supporter provided excellent electrical conductivity and a large specific surface area, effectively suppressing the HER and promoting the reduction of NO₃⁻ to NO₂⁻. Simultaneously, the carbon fiber confinement inhibited spinel particle agglomeration, maintaining the high-spin state of Cu_Oh–Co_Oh pairs and lowering the *NH₃ desorption energy barrier (Fig. 9b).

Defect engineering: constructing oxygen vacancies (O_V) in spinels optimizes their surface electronic state, enhances reactant adsorption, and inhibits the HER. Zhao et al.⁸⁵ fabricated a Co₃O₄/Co catalyst with an interleaved nanosheet structure, achieving an NH₃ yield rate of 4.43 mg h⁻¹ cm⁻² and a faradaic efficiency of 88.7% in a neutral electrolyte. Mechanistic studies revealed that O_V promoted the formation of low-coordination Co_Oh sites, causing an upward shift of the d-band center and increasing the density of unoccupied states. This enhanced NO₃⁻ adsorption, while the O_V sites, acting as strong Lewis acid sites, suppressed the HER. This work highlights the importance of synergistic modulation across multiple scales for boosting NO₃RR activity.

3.6 Carbon dioxide reduction reaction

Electrocatalytic CO₂ reduction (CO₂RR) enables the conversion of CO₂ into high-value-added chemicals such as CO, CH₄, and HCOOH under mild conditions, representing a key technology for achieving carbon neutrality.^86,87 However, the CO₂RR faces bottlenecks of low product selectivity and slow reaction kinetics due to the high bond energy of the C=O bond in CO₂ (∼750 kJ mol⁻¹), the complexity of the multi-electron/proton transfer processes, and the competitive hydrogen evolution reaction (HER). Spinels, with their tunable cation valences and diverse coordination structures, hold potential as efficient CO₂RR catalysts. Consequently, researchers employ various strategies—including defect engineering, cation substitution, and anion modification—to optimize the electronic structure and catalytic performance of spinels.

Defect engineering: the formation of oxygen vacancies (O_V) creates five-coordinated low-coordination Co_Oh sites. This causes an upward shift of the d-band center (ε_d), enhancing *COOH adsorption. Simultaneously, O_V reduces the crystal field splitting energy (Δ_oct), maintaining the high-spin (HS) state of Co²⁺ (d⁷ HS), which promotes deep hydrogenation. Li et al.⁸⁸ significantly enhanced CO₂ hydrogenation activity by controlling the O_V concentration in CoAl₂O₄ spinel (Fig. 9c). A high O_V concentration not only acts as an electron reservoir enhancing H₂ activation capability but also induces changes in the local coordination environment of Co sites, such as forming low-coordination Co_Oh sites. This electronic structure reconstruction shifts the reaction pathway towards deep hydrogenation, significantly increasing CH₄ selectivity and overcoming the limitation of traditional spinels primarily producing C₁ products (e.g., CO and HCOOH). However, spinel catalysts generally exhibit low selectivity for multi-carbon products. This limitation may be attributed to two main factors: the strong competition from hydrogen evolution and the inherently low affinity of their metal sites for crucial reaction intermediates (e.g., *CO), which impedes subsequent C–C coupling.

Cation engineering: Stangeland et al.⁸⁹ designed a series of mesoporous Mn–Co spinels (Mn_xCo_3−xO₄). XPS revealed that Mn doping systematically increased the Co valence state (Fig. 9d and e). This occurs because the highly electronegative Mn_Td withdraws electrons via Mn_Td–O–Co_Oh bonds, leading to an increased valence state of Co_Oh and an upward shift in ε_d. This optimizes *OCHO adsorption, thereby boosting activity for CO₂ hydrogenation to methanol. Furthermore, the Mn–Co synergy, combined with enhanced surface basicity, suppressed the HER side reaction. The 20% Mn-doped sample (20 MnO_x–Co₃O₄) achieved the highest methanol selectivity (22.1%) and CO₂ conversion (45.1%). This work elucidates the complete structure–activity chain: cation modification → metal valence state → surface properties → product selectivity.

Anion engineering: introducing weak-field ligands like S²⁻ reduces Δ_oct in spinels, optimizing the d-orbital electron filling state. This promotes an upward d-band center shift and enhances catalytic activity. Simon et al.⁹⁰ synthesized pure-phase spinel-type sulfide Ni₂FeS₄ nanosheets, achieving an efficient electrocatalytic CO₂RR for the co-production of CO/H₂. They found that S²⁻ substitution profoundly altered the material's intrinsic properties: (1) it induced a transition of Ni/Fe_Oh towards high-spin states (Ni²⁺: d⁸ HS → LS; Fe³⁺: d⁵ HS), shifting the d-band center upward and optimizing *COOH adsorption. (2) Compared to spinels, sulfur substitution narrowed the band gap, improving charge transport efficiency. (3) Regarding hydrophilicity, surface hydrophilic groups enhanced mass transfer at the electrolyte–catalyst interface. Consequently, the S sites in Ni₂FeS₄ are disadvantageous for C–C coupling and deep reduction, enhancing selectivity towards CO and H₂ products.

3.7 Urea oxidation reaction

The urea oxidation reaction (UOR, CO(NH₂)₂ + 6OH⁻ → N₂ + 5H₂O + 6e⁻) offers a promising alternative to the oxygen evolution reaction (OER) due to its low theoretical potential (0.37 V vs. RHE), significantly reducing the energy consumption for hydrogen production via electrocatalytic water splitting.^91–93 However, the UOR involves a kinetically sluggish six-electron process and complex intermediates, necessitating the development of highly efficient electrocatalysts.⁹⁴ Although noble metals (Pt, Rh) exhibit excellent activity, their scarcity and high cost hinder their industrial-scale application.³ Recently, spinels, particularly Ni/Co-based spinels, have emerged as ideal UOR catalysts owing to their tunable electronic structure, multivalent synergy, and low cost. Researchers optimize their performance through the following strategies:

Cation engineering: Zn²⁺ (d¹⁰) occupying tetrahedral sites (Td) acts as an electron donor. Through Zn_Td–O–Co_Oh bonds, it promotes electron delocalization at octahedral Co³⁺ (d⁶) sites, causing a negative shift of the d-band center. This weakens *CO(NH₂)₂ adsorption and accelerates dehydrogenation. Consequently, ZnCo₂O₄ on Ni foam reported by Jiang et al. achieved the optimal reaction kinetics compared to FeCo₂O₄, MnCo₂O₄, and CuCo₂O₄ (Fig. 10a).⁹⁵ Additionally, the three-dimensional porous structure provided high exposure of active sites. Combined with rapid electron transfer between ZnCo₂O₄ and the Ni foam, it conferred exceptional stability, exhibiting only a 4% activity decay after 40 h at 10 mA cm⁻² (Fig. 10b).

Fig. 10 (a) Tafel curves. (b) Chronoamperometry.⁹⁵ (a) and (b) Copyright 2023 Elsevier. (c) Statistical analysis using XPS.⁹⁶ Copyright 2024 Elsevier. (d) Comparison of the adsorption energy of *NHCONH₂. (e) The center position of the Ni d band. (f) DOS spectra.⁹⁷ Copyright 2025 Wiley Online Library.

Anion doping: Wan et al. achieved precise doping of P, B, and S atoms into two-dimensional porous NiCo₂O₄.⁹⁶ The low electronegativity and large ionic radius of P enhanced P–Co covalent bonding, expanded the lattice parameter, increased the Ni³⁺/Co³⁺ ratio, and optimized the d-band center (Fig. 10c). As a result, P–NiCo₂O₄ required only 1.27 V to reach 10 mA cm⁻², surpassing the performance of B/S-doped samples. Zhang et al.⁹⁷ also reported excellent UOR performance for a P–NiCo₂O₄ catalyst. DFT simulations revealed that P substitution for O increased the charge density at Ni sites, leading to stronger *NHCONH₂ adsorption energy on P–NiCo₂O₄ (0.59 eV) compared to pristine NiCo₂O₄ (NCO, 0.35 eV) (Fig. 10d). Furthermore, P doping increased the degree of electron delocalization at Ni, causing a downward d-band center shift (Fig. 10e), enhancing carrier concentration and electrical conductivity (Fig. 10f), and promoting urea N–H bond cleavage, thereby achieving superior UOR activity.

Defect engineering: defect engineering serves as a pivotal strategy for tailoring material properties. By inducing the formation of localized magnetic moments through unpaired d electrons in transition metal ions, it enables precise modulation of electronic spin states. This process not only enhances spin polarization but also optimizes magnetic interaction mechanisms. Furthermore, defect-induced charge redistribution and alterations in coordination number can trigger a shift in the d-band center, thereby significantly regulating the catalytic adsorption energy and activation barriers for reactants. Oxygen vacancies (O_Vs) create low-coordination Ni/Co_Oh sites, leading to an upward d-band center shift and enhancing *CO(NH₂)₂ adsorption. Xu et al. designed nanowire-like NiCo₂O₄ rich in O_Vs and possessing multiple redox couples (Ni²⁺/Ni³⁺, Co²⁺/Co³⁺).⁹⁸ This increased the proportion of high-spin Ni²⁺ (d⁸ HS) and Co²⁺ (d⁷ HS), accelerating proton-coupled electron transfer (PCET). Consequently, NiCo₂O₄ required only 1.343 V to achieve 100 mA cm⁻² and maintained stability for >20 h.

4. Generalizable strategies for optimizing the spinel electronic structure

Based on the preceding comprehensive review of spinels in electrocatalysis, it is concluded that the d-band center, spin state and geometrized sites are the key factors affecting the performance of spinels (Table 1). The generalizable strategies for optimizing the spinel electronic structure were proposed to guide the precision design of high-performance spinels (Fig. 11).

Table 1 The effects of the d-band center, spin state and geometrized sites of spinels for the OER

Materials	Active sites	d-Band center (eV)	Shift of the d-band center	ηj=10 (mV)	Ref.
Ce–Co3O4	Co	−3.06	Downshift	340	99
Vo–Fe0.4Co2.6O4	Co	−3.12	Downshift	288	100
VCo–Co3O4	Co	−2.36	Downshift	240	101
Fe–Co3O4	Co	−2.54	Middle	384	102

---

Materials	Active sites	Spin state	Electron-filled state	ηj=10	Ref.
Co3O4@NiFe–LDH/CC	Co^2+	High spin	t^32ge^4g	217	103
Zn0.8Co2.2O4	Co^3+	High spin	t^42ge^2g	254	104
LiCoO2	Co^3+	Intermediate spin	t^52ge^1g	374	105
Ru/Mn–Co3O4	Co^3+	High spin	t^42ge^2g	384	106

---

Materials	Active sites	Optimizing strategy	ηj=10	Ref.
Co2FeO4/NCNTs	FeOct/Td–O–CoOct/Td	Fe^3+ doping in Co3O4	420	8
(Ni,Mn)–(Co)tet(Co2)octO4	(Ni,Mn)–(Co)tet(Co2)oct	Ni, Mn doping in Co3O4	242	107
CoFe0.25Al1.75O4	CoOct–O–CoTd	—	361	108
Zn0.3–Co3O4	CoOct–O–CoTd	Zn^2+ doping in Co3O4	293	109
CoFe2O4−Cd	Co–O–Fe	N atom modification	254	110
NiFe2O4–M	Ni^2+(Td)	Addition of a magnetic field	322	111
B–MnFe2O4	FeOh	B modified MnFe2O4	212	112

---

Fig. 11 Universal strategies for optimizing the electronic structure of spinels.

Regulation of A/B site metal coupling: optimizing the spatial connectivity modes and electronic interactions between tetrahedrally coordinated A-site and octahedrally coordinated B-site metals enhances charge transfer pathways and reaction kinetics. Specifically, corner-sharing connectivity shortens A–O–B bond lengths, promotes electron delocalization, and facilitates synergistic redox processes at bimetallic sites. Conversely, edge-sharing connectivity activates the lattice oxygen via strong orbital overlap, lowering the energy barrier for oxygen intermediate formation. Consequently, selecting A/B site metal pairs with significant electronegativity differences leverages their strong electronic coupling to reduce reactant binding energies.

Optimization of the metal d-band center position: modulating the transition metal d-band center's energy position effectively tunes reactant adsorption/desorption strength. An upward shift fills antibonding states, weakening the strong adsorption of oxygen-containing intermediates and lowering reaction overpotential. A moderate downward shift, however, enhances reactant activation capability. Key implementation pathways include introducing high-electronegativity A-site metal ions or utilizing lattice distortion and heterointerface construction to shift bonding states, thereby positioning reactant adsorption energy near the volcano plot apex.

Modulation of the metal center electron spin state: modulating the crystal field splitting energy (Δ) controls the metal ion spin state. A high-spin state favors stronger adsorption and activation of reactants at the metal center. A low-spin state, in contrast, facilitates lower activation energy barriers and accelerates proton-coupled electron transfer processes. Implementation strategies involve systematic defect etching, atomic doping, and anion/cation substitution.

Key strategies to optimize spinel properties—including A/B site metal coupling, d-band center position, and electron spin state—are defect engineering, atomic doping, anion/cation substitution, nanostructure modulation, and heterointerface construction. These approaches should target precise tuning of the d-band center and electron spin state at A/B site metals to enhance catalytic activity. Critically, synergistic regulation of A/B site coupling, the d-band center, and metal spin states—achieved via the above strategies—provides an effective pathway to optimize the electronic structure. This optimization ultimately boosts intrinsic activity, reaction efficiency, and conductivity.

5. Conclusion and outlook

In summary, this review examines the electronic structure–performance relationship in spinel electrocatalysts. The key factors analyzed include coordination geometry (Td/Oh sites), the electronic configuration of metal centers (d-band center and spin state), and their orbital coupling effects with reaction intermediates. The application of electronic structure modulation strategies in the OER, ORR, HER, NRR, NO₃RR, CO₂RR and UOR is also discussed. Ultimately, a generalizable design paradigm is proposed: “coordination engineering → d-band center optimization → spin state modulation”. Despite significant advances in intrinsic activity and charge transfer efficiency through electronic structure engineering, three core challenges and further efforts remain for practical spinel applications (Fig. 12).

Fig. 12 Schematic diagram for further efforts of the spinel electrocatalyst.

Resolving dynamic reaction surfaces: current in situ techniques (e.g., XAS and Raman) capture steady-state electron transfer and bonding changes. However, they struggle to reveal transient reaction pathways on the femtosecond-to-picosecond timescale during multi-electron/proton transfers. Future efforts should develop time-resolved coupled techniques like transient X-ray emission spectroscopy and surface-enhanced infrared spectroscopy. These techniques need to track intermediate evolution in real-time. Integrating them with machine learning molecular dynamics simulations will help build dynamic interaction models between A or B sites in spinels and intermediates.

Quantifying activity influencing factors: doping with transition metals (Fe/Co/Ni) or non-metals (P/B/S) can optimize lattice distortion and spin states. However, a systematic understanding is lacking. Specifically, how intrinsic element properties, such as electronegativity, ionic radius and orbital energy levels, quantitatively affect catalytic descriptors needs clarification. Future work requires constructing a comprehensive spinel element-descriptor database, including the d-band center, ΔG_*ads, and e_g or t_2g orbital occupancy. High-throughput calculations coupled with graph neural networks should then predict optimal compositions of spinel materials and types of catalytically active sites.

Achieving industrial-grade current stability: at high current densities (>500 mA cm⁻²), spinels often suffer from active site dissolution, phase transformation, and surface reconstruction. This leads to rapid cycling stability loss. Moreover, due to their instability in acidic media, spinel oxides can only be used in alkaline water electrolyzers. In order to enhance the commercial applicability of spinels, future designs should prioritize those with optimized gradient electronic structures and architecture that ensure both efficient catalytic activity and robust stability. Key strategies include: (i) surface-bulk engineering: creating rich active sites through oxygen vacancies, atomic doping, anion/cation substitution, nanostructure modulation, or heterointerface construction based on spinels to enhance reactant adsorption; (ii) self-healing coatings: developing dynamic phosphate protective layers on catalyst surfaces to suppress transition metal leaching and enhance the stability; and (iii) system validation: rigorously testing long-term catalyst performance in integrated membrane electrode assembly (MEA) systems within industrial electrolyzers. Tests must cover wide pH ranges, high salinity, and impurity tolerance.

Conflicts of interest

There are no conflicts to declare.

Data availability

No primary research results, software or code have been included and no new data were generated or analysed as part of this review.

Acknowledgements

This work was supported by the Natural Science Foundation of China (No. U24A20541, 22278094, and W2421038 and 22308070), the Jiangxi Province “Double Thousand Plan” (No. jxsq2023102142), the Basic and Applied Basic Research Program of Guangzhou (No. SL2024A03J00499), the Guangdong Basic and Applied Basic Research Foundation (2024A1515140005 and 2024B1515120017) and the Foshan Xianhu Laboratory Project (XHD2024-31000000-06).

Notes and references

1. T. Wen, Y. Zheng, H. Wang and J. Zou, Appl. Catal., B, 2025, 378, 125588.
2. Y. Cheng, P. Fu, Z. Yu, X. Yang, Y. Zhang, A. Yuan, H. Liu, J. Du and L. Chen, Carbon Neutralization, 2024, 3, 873–887.
3. Q. Zhao, Z. Yan, C. Chen and J. Chen, Chem. Rev., 2017, 117, 10121–10211.
4. S. Huang, F. Feng, R. Huang, T. Ouyang, J. Liu and Z. Liu, Adv. Mater., 2022, 34, 2208438.
5. M.-S. Balogun, H. Yang, Y. Luo, W. Qiu, Y. Huang, Z.-Q. Liu and Y. Tong, Energy Environ. Sci., 2018, 11, 1859–1869.
6. K. Xiao, B.-H. Xiao, J.-X. Li, S. Cao and Z.-Q. Liu, Proc. Natl. Acad. Sci. U. S. A., 2024, 121, e2409201121.
7. X. Wang, T. Ouyang, L. Wang, J. Zhong and Z. Liu, Angew. Chem. Int. Ed., 2020, 59, 6492–6499.
8. X. Wang, T. Ouyang, L. Wang, J. Zhong, T. Ma and Z. Liu, Angew. Chem. Int. Ed., 2019, 58, 13291–13296.
9. L. Tian, Z. Tang, L. Hao, T. Dai, J. Zou and Z. Liu, Angew. Chem. Int. Ed., 2024, 136, e202401434.
10. H. Cheng, J.-M. Chen, Q.-J. Li, C.-Y. Su, A.-N. Chen, J.-X. Zhang, Z.-Q. Liu and Y. Tong, Chem. Commun., 2017, 53, 11596–11599.
11. J. Li, X. Wang, H. Fang, X. Guo, R. Zhou, C. Wang, J. Li, M. N. Ghazzal and Z. Rui, J. Mater. Chem. A, 2023, 11, 25639–25649.
12. K. Xiao, Y. Wang, P. Wu, L. Hou and Z. Liu, Angew. Chem. Int. Ed., 2023, 135, e202301408.
13. H. Cheng, M. Li, C. Su, N. Li and Z. Liu, Adv. Funct. Mater., 2017, 27, 1701833.
14. H. Cheng, Y. Su, P. Kuang, G. Chen and Z. Liu, J. Mater. Chem. A, 2015, 3, 19314–19321.
15. Z. Wang, J. Huang, L. Wang, Y. Liu, W. Liu, S. Zhao and Z. Liu, Angew. Chem. Int. Ed., 2022, 134, e202114696.
16. Z. Liu, H. Cheng, N. Li, T. Y. Ma and Y. Su, Adv. Mater., 2016, 28, 3777–3784.
17. Z. Wang, P. Wu, X. Zou, S. Wang, L. Du, T. Ouyang and Z. Liu, Adv. Funct. Mater., 2023, 33, 2214275.
18. Z. Li, Y. Zhang, Y. Feng, C. Cheng, K. Qiu, C. Dong, H. Liu and X. Du, Adv. Funct. Mater., 2019, 29, 1903444.
19. D. Ghosh, U. Dutta, A. Haque, N. E. Mordvinova, O. I. Lebedev, K. Pal, A. Gayen, Md. M. Seikh and P. Mahata, Dalton Trans., 2018, 47, 5011–5018.
20. J. Bao, X. Zhang, B. Fan, J. Zhang, M. Zhou, W. Yang, X. Hu, H. Wang, B. Pan and Y. Xie, Angew. Chem. Int. Ed., 2015, 127, 7507–7512.
21. Z. Wang, H. Liu, R. Ge, X. Ren, J. Ren, D. Yang, L. Zhang and X. Sun, ACS Catal., 2018, 8, 2236–2241.
22. L. Leng, X. Zeng, H. Song, T. Shu, H. Wang and S. Liao, J. Mater. Chem. A, 2015, 3, 15626–15632.
23. J. Zhang, X. Shang, H. Ren, J. Chi, H. Fu, B. Dong, C. Liu and Y. Chai, Adv. Mater., 2019, 31, 1905107.
24. P. Sekar, L. Calvillo, C. Tubaro, M. Baron, A. Pokle, F. Carraro, A. Martucci and S. Agnoli, ACS Catal., 2017, 7, 7695–7703.
25. S. Li, J. Fan, S. Li, Y. Ma, J. Wu, H. Jin, Z. Chao, D. Pan and Z. Guo, J. Nanostruct. Chem., 2021, 11, 735–749.
26. L. Lyu, S. Zheng, F. Wang, Y. Liu and J. Liu, J. Colloid Interface Sci., 2021, 602, 197–206.
27. P. Tan, B. Chen, H. Xu, W. Cai, W. He and M. Ni, Appl. Catal., B, 2019, 241, 104–112.
28. Z. Wang, J. Huang, L. Wang, Y. Liu, W. Liu, S. Zhao and Z. Liu, Angew. Chem. Int. Ed., 2022, 134, e202114696.
29. Y. Sun, X. Ren, S. Sun, Z. Liu, S. Xi and Z. J. Xu, Angew. Chem. Int. Ed., 2021, 133, 14657–14665.
30. C. Wei, Z. Feng, G. G. Scherer, J. Barber, Y. Shao-Horn and Z. J. Xu, Adv. Mater., 2017, 29, 1606800.
31. H. Li, S. Sun, S. Xi, Y. Chen, T. Wang, Y. Du, M. Sherburne, J. W. Ager, A. C. Fisher and Z. J. Xu, Chem. Mater., 2018, 30, 6839–6848.
32. C. Xie, D. Yan, W. Chen, Y. Zou, R. Chen, S. Zang, Y. Wang, X. Yao and S. Wang, Mater. Today, 2019, 31, 47–68.
33. B. Liu, J. Cheng, H. Peng, D. Chen, X. Cui, D. Shen, K. Zhang, T. Jiao, M. Li, C. Lee and W. Zhang, J. Mater. Chem. A, 2019, 7, 775–782.
34. T. Zhao, R. Abdurahman, R. Aiwaili, S. Wu and X.-B. Yin, Coord. Chem. Rev., 2023, 488, 215171.
35. R. J. Hill, J. R. Craig and G. V. Gibbs, Phys. Chem. Miner., 1979, 4, 317–339.
36. R. G. Burns, Mineralogical Applications of Crystal Field Theory, Cambridge University Press, Cambridge, 2nd edn, 1993, pp. 7–43.
37. N. Zhang and Y. Chai, Energy Environ. Sci., 2021, 14, 4647–4671.
38. Y. Zhang, Q. Wu, J. Z. Y. Seow, Y. Jia, X. Ren and Z. J. Xu, Chem. Soc. Rev., 2024, 53, 8123–8136.
39. J. Wu, X. Wang, W. Zheng, Y. Sun, Y. Xie, K. Ma, Z. Zhang, Q. Liao, Z. Tian, Z. Kang and Y. Zhang, J. Am. Chem. Soc., 2022, 144, 19163–19172.
40. K. A. Moltved and K. P. Kepp, J. Phys. Chem. C, 2019, 123, 18432–18444.
41. J. K. Nørskov, F. Abild-Pedersen, F. Studt and T. Bligaard, Proc. Natl. Acad. Sci. U. S. A., 2011, 108, 937–943.
42. B. Hammer and J. K. Norskov, Nature, 1995, 376, 238–240.
43. Z. Li and C. Yu, in Nanostructured Materials, Elsevier, 2024, pp. 43–92.
44. X. Wang, Z. Kang, D. Wang, Y. Zhao, X. Xiang, H. Shang and B. Zhang, Nano Energy, 2024, 121, 109268.
45. K. Liu, J. Fu, Y. Lin, T. Luo, G. Ni, H. Li, Z. Lin and M. Liu, Nat. Commun., 2022, 13, 2075.
46. X. Li, T. Chen, B. Yang and Z. Xiang, Angew. Chem. Int. Ed., 2023, 62, e202215441.
47. K. Zhang and R. Zou, Small, 2021, 17, 2100129.
48. N. T. Suen, S. F. Hung, Q. Quan, N. Zhang, Y. J. Xu and H. M. Chen, Chem. Soc. Rev., 2017, 46, 337–365.
49. Y. Liu, L. Xiao, H. Tan, J. Zhang, C. Dong, H. Liu, X. Du and J. Yang, Small, 2024, 20, 2401504.
50. Y. Li, Z. Zhang, C. Li, Y. Zhou, X. Chen, H. Lu, Z. Shi and S. Feng, Appl. Catal., B, 2024, 355, 124116.
51. S. Li, S. Zhao, S.-F. Hung, L. Deng, L. Wang, F. Shi, A. Dong, Y. Zhang, T.-Y. Chen, F. Hu, L. Li, S. Ramakrishna, Y. Wu and S. Peng, J. Am. Chem. Soc., 2025, 147, 33770–33779.
52. Y. Y. Sun, X. Y. Zhang, J. Tang, X. Li, H. Q. Fu, H. G. Xu, F. Mao, P. Liu and H. G. Yang, Small, 2023, 19, 2207965.
53. X. Yan, Y. Jia and X. Yao, Chem. Soc. Rev., 2018, 47, 7628–7658.
54. Y. Wang, J. Li and Z. Wei, J. Mater. Chem. A, 2018, 6, 8194–8209.
55. L. Zhang, Y. Hu, K. Jiang, K. Li, Y.-Q. Liu, D. Wang and Y. Ye, J. Electrochem. Soc., 2021, 168, 054523.
56. Z. Zhang, G. Tan, A. Kumar, H. Liu, X. Yang, W. Gao, L. Bai, H. Chang, Y. Kuang, Y. Li and X. Sun, Mol. Catal., 2023, 535, 112852.
57. J. Zhao, Y. He, J. Wang, J. Zhang, L. Qiu, Y. Chen, C. Zhong, X. Han, Y. Deng and W. Hu, Chem. Eng. J., 2022, 435, 134261.
58. B. Hammer, Y. Morikawa and J. K. Nørskov, Phys. Rev. Lett., 1996, 76, 2141–2144.
59. V. Stamenkovic, B. S. Mun, K. J. J. Mayrhofer, P. N. Ross, N. M. Markovic, J. Rossmeisl, J. Greeley and J. K. Nørskov, Angew. Chem. Int. Ed., 2006, 45, 2897–2901.
60. L. Shang, Y. Ni, Y. Wang, W. Yang, L. Wang, H. Li, K. Zhang, Z. Yan and J. Chen, Adv. Mater., 2024, 36, 2413141.
61. M. Zhou, D. Yoon, Y. Yang, L. Zhang, C. Li, H. Wang, A. Sharma, S. Jiang, D. A. Muller, H. D. Abruña and J. Fang, ACS Energy Lett., 2023, 8, 3631–3638.
62. Y. Zhou, S. Xi, J. Wang, S. Sun, C. Wei, Z. Feng, Y. Du and Z. J. Xu, ACS Catal., 2018, 8, 673–677.
63. Y. Wang, Y. Yang, S. Jia, X. Wang, K. Lyu, Y. Peng, H. Zheng, X. Wei, H. Ren, L. Xiao, J. Wang, D. A. Muller, H. D. Abruña, B. J. Hwang, J. Lu and L. Zhuang, Nat. Commun., 2019, 10, 1506.
64. X. H. Chen, T. Li, X. Li, J. Lei, N. B. Li and H. Q. Luo, ACS Catal., 2023, 13, 6721–6729.
65. Z. Zeb, Y. Huang, L. Chen, W. Zhou, M. Liao, Y. Jiang, H. Li, L. Wang, L. Wang, H. Wang, T. Wei, D. Zang, Z. Fan and Y. Wei, Coord. Chem. Rev., 2023, 482, 215058.
66. H. Zhang, W. Li, X. Feng, L. Zhu, Q. Fang, S. Li, L. Wang, Z. Li and Z. Kou, J. Colloid Interface Sci., 2022, 607, 281–289.
67. S. Ma, P. Yang, J. Chen, Z. Wu, X. Li and H. Zhang, J. Colloid Interface Sci., 2023, 642, 604–611.
68. J. Zhang, X. Shang, H. Ren, J. Chi, H. Fu, B. Dong, C. Liu and Y. Chai, Adv. Mater., 2019, 31, 1905107.
69. D. Wang, Y. Chen, L. Fan, T. Xiao, T. Meng, Z. Xing and X. Yang, Appl. Catal., B, 2022, 305, 121081.
70. M. Wang, X. Feng, S. Li, Y. Ma, Y. Peng, S. Yang, Y. Liu, H. Lei, J. Dang, W. Zhang, R. Cao and H. Zheng, Adv. Funct. Mater., 2024, 34, 2410439.
71. S. Peng, F. Gong, L. Li, D. Yu, D. Ji, T. Zhang, Z. Hu, Z. Zhang, S. Chou, Y. Du and S. Ramakrishna, J. Am. Chem. Soc., 2018, 140, 13644–13653.
72. J. Jin, J. Yin, H. Liu, B. Huang, Y. Hu, H. Zhang, M. Sun, Y. Peng, P. Xi and C. Yan, Angew. Chem. Int. Ed., 2021, 133, 14236–14242.
73. R. Nittoor-Veedu, X. Ju and M. Pumera, Adv. Energy Mater., 2024, 2402205.
74. H. Liu, Chin. J. Catal., 2014, 35, 1619–1640.
75. H. Wang, R. R. Seemakurthi, G. Chen, V. Strauss, O. Savateev, G. Hai, L. Ding, N. López, H. Wang and M. Antonietti, Nat. Commun., 2023, 14, 5668.
76. Q. Wang, J. Guo and P. Chen, J. Energy Chem., 2019, 36, 25–36.
77. K. Ithisuphalap, H. Zhang, L. Guo, Q. Yang, H. Yang and G. Wu, Small Methods, 2019, 3, 1800352.
78. Y. Jiang, M. Wang, L. Zhang, S. Liu, Y. Cao, S. Qian, Y. Cheng, X. Xu, C. Yan and T. Qian, Chem. Eng. J., 2022, 450, 138226.
79. J. Lee, A. Kumar, M. G. Kim, T. Yang, X. Shao, X. Liu, Y. Liu, Y. Hong, A. R. Jadhav, M. Liang, N. Q. Tran and H. Lee, ACS Energy Lett., 2021, 6, 4299–4308.
80. X. Tong, Z. Zhang, Z. Fang, J. Guo, Y. Zheng, X. Liang, R. Liu, L. Zhang and W. Chen, J. Phys. Chem. C, 2023, 127, 5262–5270.
81. Z. Deng, J. Liang, Q. Liu, C. Ma, L. Xie, L. Yue, Y. Ren, T. Li, Y. Luo, N. Li, B. Tang, A. Ali Alshehri, I. Shakir, P. O. Agboola, S. Yan, B. Zheng, J. Du, Q. Kong and X. Sun, Chem. Eng. J., 2022, 435, 135104.
82. M. Liu, Q. Mao, K. Shi, Z. Wang, Y. Xu, X. Li, L. Wang and H. Wang, ACS Appl. Mater. Interfaces, 2022, 14, 13169–13176.
83. J. Guan, L. Cai, W. Li, H. Zhou and Y. Huang, Appl. Catal., B, 2024, 358, 124387.
84. Z. Niu, S. Fan, X. Li, P. Wang, Z. Liu, J. Wang, C. Bai and D. Zhang, Chem. Eng. J., 2022, 450, 138343.
85. F. Zhao, G. Hai, X. Li, Z. Jiang and H. Wang, Chem. Eng. J., 2023, 461, 141960.
86. M. Liu, M. Liu, X. Wang, S. M. Kozlov, Z. Cao, P. De Luna, H. Li, X. Qiu, K. Liu, J. Hu, C. Jia, P. Wang, H. Zhou, J. He, M. Zhong, X. Lan, Y. Zhou, Z. Wang, J. Li, A. Seifitokaldani, C. T. Dinh, H. Liang, C. Zou, D. Zhang, Y. Yang, T.-S. Chan, Y. Han, L. Cavallo, T.-K. Sham, B.-J. Hwang and E. H. Sargent, Joule, 2019, 3, 1703–1718.
87. M. A. Tekalgne, H. H. Do, A. Hasani, Q. Van Le, H. W. Jang, S. H. Ahn and S. Y. Kim, Mater. Today Adv., 2020, 5, 100038.
88. Y. Li, Z. Zhao, M. Zhao, H. Zhu, X. Ma, Z. Li, W. Lu, X. Chen, L. Ying, R. Lin, Y. Meng, Y. Lyu, L. Yan and Y. Ding, Appl. Catal., B, 2024, 347, 123824.
89. K. Stangeland, D. Y. Kalai, Y. Ding and Z. Yu, J. CO₂ Util., 2019, 32, 146–154.
90. C. Simon, J. Zander, T. Kottakkat, M. Weiss, J. Timm, C. Roth and R. Marschall, ACS Appl. Energy Mater., 2021, 4, 8702–8708.
91. X. Gao, X. Bai, P. Wang, Y. Jiao, K. Davey, Y. Zheng and S.-Z. Qiao, Nat. Commun., 2023, 14, 5842.
92. S.-K. Geng, Y. Zheng, S.-Q. Li, H. Su, X. Zhao, J. Hu, H.-B. Shu, M. Jaroniec, P. Chen, Q.-H. Liu and S.-Z. Qiao, Nat. Energy, 2021, 6, 904–912.
93. R. Lin, L. Kang, T. Zhao, J. Feng, V. Celorrio, G. Zhang, G. Cibin, A. Kucernak, D. J. L. Brett, F. Corà, I. P. Parkin and G. He, Energy Environ. Sci., 2022, 15, 2396.
94. G. Ma, Q. Xue, J. Zhu, X. Zhang, X. Wang, H. Yao, G. Zhou and Y. Chen, Appl. Catal., B, 2020, 265, 118567.
95. L. Jiang, X. Cheng, H. Zhang, Q. Cao, K. Song and J. He, J. Colloid Interface Sci., 2023, 643, 403–408.
96. J. Wan, Z. Wu, G. Fang, J. Xian, J. Dai, J. Guo, Q. Li, Y. You, K. Liu, H. Yu, W. Xu, H. Jiang, M. Xia and H. Jin, J. Energy Chem., 2024, 91, 226–235.
97. H. Zhang, Z. Chen, X. Ye, K. Xiao and Z. Liu, Angew. Chem. Int. Ed., 2025, 64, e202421027.
98. X. Xu, Y. Zhang, L. Su, H. Zhang and X. Miao, Int. J. Hydrogen Energy, 2023, 48, 9669–9681.
99. J. Qian, X. Liu, C. Zhong, G. Xu, H. Li, W. Zhou, B. You, F. Wang, D. Gao and D. Chao, Adv. Funct. Mater., 2022, 33, 2212021.
100. P. Guo, D. Liu, Q. Wang, P. Chen, H. Yang, M. Zhang, C. Zheng, L. Jiang, H. Pan and R. Wu, Nano Lett., 2025, 25, 10209–10217.
101. S. Pan, L. Zhang, M. Liu, X. Pan, M. Bi, T. Guo, Y. Zhang, J. Sun, A. Vasiliev, X. Ouyang, X. Wang, J. Zhu and Y. Fu, ACS Sustainable Chem. Eng., 2022, 11, 290–299.
102. X. Wu, Z. Shao, Q. Zhu, X. Hou, C. Wang, J. Zeng, K. Huang and S. Feng, ACS Catal., 2024, 14, 5888–5897.
103. Z. Xue, B. Wu, Z. Zhang, C. Lin, X. Li, Q. Zhang and K. Tao, ACS Catal., 2024, 14, 5685–5695.
104. R. M. Ramsundar, V. K. Pillai and P. A. Joy, Phys. Chem. Chem. Phys., 2018, 20, 29452–29461.
105. T. Maiyalagan, K. A. Jarvis, S. Therese, P. J. Ferreira and A. Manthiram, Nat. Commun., 2014, 5, 3949.
106. X. Wen, L. Ren, D. Du, Y. Yan, H. Xu, T. Zeng, G. Tian, X. Wang, S. Liu and C. Shu, Energy Fuels, 2022, 37, 735–745.
107. L. An, H. Zhang, J. Zhu, S. Xi, B. Huang, M. Sun, Y. Peng, P. Xi and C. H. Yan, Angew. Chem. Int. Ed., 2023, 62, e202214600.
108. A. Zou, C. Wu, Q. Zhang, Y. Tang, J. Li, H. Meng, Z. Wang, C. Diao, Z. Yu, J. Xue, S. Xi, X. Wang and J. Wu, Angew. Chem. Int. Ed., 2024, 63, e202409912.
109. G. Chen, Q. Yin, X. Li, X. Han, C. Qin, Y. Ji, Y. Chen and W. Zhang, Fuel, 2024, 378.
110. Y. Wang, J. Jia, X. Zhao, W. Hu, H. Li, X. Bai, J. Huang, J. Zhang, J. Li, X. Tang, Y. Peng, J. Huang and C. Xu, ACS Catal., 2024, 14, 2313–2323.
111. X. Lyu, Y. Zhang, Z. Du, H. Chen, S. Li, A. I. Rykov, C. Cheng, W. Zhang, L. Chang, W. Kai, J. Wang, L. Zhang, Q. Wang, C. Huang and E. Kan, Small, 2022, 18, e2204143.
112. M. Chen, N. Kitiphatpiboon, C. Feng, Q. Zhao, A. Abudula, Y. Ma, K. Yan and G. Guan, Appl. Catal., B, 2023, 330, 122577.

---