Prussian Blue Analogues and Their Derivatives for the Oxygen Evolution Reaction: A Review on Active Site Engineering Strategies

Abstract

The oxygen evolution reaction (OER) is a kinetic bottleneck in electrochemical water splitting, creating an urgent need for the development of efficient electrocatalysts. Prussian blue analogues (PBAs), a significant class of inorganic coordination polymers, have emerged as excellent precursors and pre-catalysts for preparing various OER nanocatalysts, owing to their numerous advantages such as tunable composition, controllable morphology, and structural derivability. This review systematically summarizes recent advances in PBA-based OER electrocatalysts, beginning with two core strategies: enhancing active site accessibility and utilization, and improving the intrinsic activity of each active site. We provide an in-depth discussion of the design principles for enhancing active site accessibility and utilization through constructing porous architectures, creating hierarchical porosity, and improving electrical conductivity. The review also details key approaches for improving intrinsic activity, including regulating electronic structure via elemental doping and optimizing active sites via defect engineering, while examining the underlying mechanisms for performance enhancement. Finally, current challenges and future research directions are outlined, offering a perspective on the potential applications of PBA-based catalysts in sustainable energy conversion systems.

1. Introduction

Coordination polymers have garnered significant attention due to their tunable structures and diverse functionalities, demonstrating broad application potential in fields such as energy conversion, catalysis, and environmental remediation [1,2]. Among them, Prussian blue (PB) and Prussian blue analogues (PBAs) represent one of the earliest and most systematically investigated classes of coordination polymers. PB possesses a characteristic cubic framework structure in which Fe²⁺ and Fe³⁺ sites are alternately interconnected via cyanide ligands (–C≡N–, CN). PBAs retain a similar crystalline architecture to PB, yet their cyanide ligands serve as bridges between different transition metal ions (M²⁺–C≡N–M³⁺, such as Fe, Co, Ni, Zn, Cu, or Mn), conferring high tunability in both composition and morphology [3,4,5,6]. The structure of PBAs has been well established, as illustrated in Figure 1a. Their general chemical formula is represented by A_xM^A[M^B(CN)₆]_y·zH₂O, where A represents cations intercalated within the framework, and M^A and M^B are metal ions coordinated to the CN [7,8,9]. PB was first synthesized as early as 1706, yet the nanoscale development of PBAs occurred relatively late, with stable synthesis methods for nano-sized samples not reported until 1999 by Moulik et al. [10]. In 2007, Qiu et al. pioneered the integration of PBAs with nanoparticles, while in 2012, Yamauchi et al. reported the groundbreaking development of internal etching technology, opening a new chapter for the application and development of PBAs [11,12]. With the advancement of characterization techniques, Itoi et al. utilized ionic liquid-assisted transmission electron microscopy (TEM) to obtain periodic structural images and achieved the first direct atomic-scale observation of the charge transfer-induced spin transition process in coordination polymers [3]. In 2016, Han et al. further acquired clear high-resolution TEM images revealing a checkerboard pattern of alternating Co and Fe atomic arrays, and confirmed the presence of expected cyanide vacancies (V_CN) in the crystal (Figure 1b) [13]. With the diversification of synthesis methods, some researchers have created a high density of vacancies in PBAs to enhance their performance for practical applications [14].

In recent years, PBAs have exhibited remarkable performance across various applications, including ion batteries, supercapacitors, electrocatalysis, optoelectronic devices, and pollutant degradation, owing to their facile synthesis, low cost, and structural adaptability [15,16,17,18,19]. Against the backdrop of escalating global energy demand and mounting environmental challenges, the development of clean and sustainable energy technologies has become a central focus of scientific research [20,21,22,23]. Hydrogen energy, in particular, has emerged as a key topic across multiple disciplines due to its high energy density and zero carbon emissions, positioning it as an ideal energy carrier [24,25]. Electrochemical water splitting offers a promising pathway to convert surplus electricity generated from intermittent renewable sources such as wind and solar into high-value hydrogen, enabling efficient energy storage and utilization [26,27,28].

Electrochemical water splitting for hydrogen production comprises the anodic oxygen evolution reaction (OER, 4OH⁻ → O₂ + 2H₂O + 4e⁻) and the cathodic hydrogen evolution reaction (HER, 4H⁺ + 4e⁻ → 2H₂). The OER, a kinetically sluggish four-electron transfer process, exhibits a thermodynamic equilibrium potential of 1.23 V vs. RHE, significantly higher than the 0 V vs. RHE potential required for the HER [29,30,31]. Moreover, its slow reaction kinetics lead to a high overpotential, making OER the primary contributor to the overall energy consumption and cost of hydrogen production via water electrolysis. The adsorbate evolution mechanism (AEM) is widely regarded as the conventional OER pathway [32,33]. In this mechanism, oxygen evolution proceeds through sequential intermediates adsorbed on metal active sites: HO* → O* → HOO*, culminating in O₂ release [34]. Through DFT calculations, Man et al. demonstrated a universal scaling relation between HOO* and HO* binding energies and established △G⁰_O*−△G⁰_HO* as a generalized activity descriptor. This descriptor successfully reproduced the renowned volcano plot. While this scaling relation constrains the HOO*/HO* binding energy difference to ~3.20 eV for most catalysts, imposing a fundamental overpotential limit of 0.37 V [35], numerous recent studies have reported catalysts that defy this limit, exhibiting significantly lower overpotentials and pointing to the operation of alternative mechanisms beyond AEM [36,37,38,39,40]. Among these, the lattice oxygen oxidation mechanism (LOM) has emerged as a representative pathway, which bypasses the scaling-relation constraints of AEM by directly forming the O–O bond through coupling between a lattice oxygen and an *O, offering a potentially faster kinetic route [41]. This mechanism is often activated in rationally designed catalysts, such as those rich in oxygen vacancies [42], enabling significantly reduced overpotentials. Therefore, understanding these mechanistic pathways and incorporating them into material design is crucial for developing advanced OER electrocatalysts toward efficient hydrogen production. This mechanistic insight guides the design of next-generation electrocatalysts, which must be not only highly active but also stable and cost-effective.

Ir- and Ru-based noble metal catalysts and their compounds have long been regarded as the most advanced electrocatalysts for the OER, demonstrating significant advantages due to their high activity. However, their high cost, scarcity, and potential instability under alkaline conditions pose considerable challenges for large-scale practical applications [43,44]. Consequently, researchers have increasingly focused on exploring non-precious metal-based materials for OER. Various types of non-precious metal catalysts—including alloys, layered double hydroxides (LDHs), perovskites, metal–organic frameworks (MOFs), and other structured compounds—have significantly advanced OER development, serving as precursors, catalysts, or pre-catalysts [29,45,46]. Nevertheless, most of these materials, such as alloys, LDHs, and perovskites, are typically synthesized from metal salts and/or heteroatom-containing precursors, making it difficult to precisely control their composition and structure, which in turn limits their electrocatalytic activity. Meanwhile, the high cost of certain organic ligands and complex synthesis procedures involved in MOF-derived catalysts also restrict their scalability. In this context, PBA emerges as a highly promising candidate. On one hand, PBAs share structural advantages with MOFs: their metal ions are highly tunable, allowing atomic-level precision in active site engineering and rational design of catalytic centers; they also possess a high specific surface area and uniformly distributed active sites, which are conducive to enhancing catalytic performance. On the other hand, as CN-bridged coordination polymers, PBAs benefit from low-cost raw materials and simple synthesis routes, offering great potential for large-scale production. These attributes enable PBA to serve not only as efficient catalysts or pre-catalysts but also as ideal precursors for catalytic materials, opening new avenues for the development of high-performance, low-cost OER catalytic systems [47,48,49].

Given these advantages, considerable efforts have been devoted to enhancing OER performance through in-depth research and structural optimization of PBA-based materials, thereby advancing the development of renewable energy technologies. Numerous comprehensive reviews have summarized the progress in PBA applications [5,19,47,49]. In recent years, the pivotal role of defects in electrocatalytic processes has gained increasing recognition within the scientific community. Defect states not only effectively modulate the adsorption energy barriers of active species and facilitate the desorption of target products but also enhance catalytic selectivity by favoring specific reaction pathways. Furthermore, they improve material stability by reinforcing the catalyst structure and suppressing element leaching [50,51,52]. The engineering of defects significantly elevates the performance ceiling of materials while providing new theoretical perspectives for elucidating reaction mechanisms. In Prussian blue analogs, the inherent lability of CN coordination bonds readily induces the formation of V_CN. These V_CN exhibit electronic modulation capabilities comparable to conventional anionic vacancies and demonstrate remarkable intrinsic catalytic activity. By employing targeted synthesis strategies and plasma surface treatment, defect concentrations can be precisely controlled, thereby substantially expanding the application prospects of PBA materials in the OER [53,54,55]. Indeed, PBA-derived catalyst systems have demonstrated remarkable electrocatalytic performance. For instance, Xu et al. successfully activated the lattice oxygen of CoFe-PBA via a plasma activation strategy, and the resulting catalyst achieved a high current density of 1000 mA cm⁻² at a low overpotential of 276 mV for the OER, with a Tafel slope as low as 22.9 mV dec⁻¹ [56]. The hierarchically structured PBA-derived catalytic electrode reported by Zhang’s group maintained excellent catalytic performance even after a long-term stability test of 2000 h [57]. Compared to noble metal benchmark catalysts, PBA-derived materials also exhibit highly competitive performance. Our previous study revealed that under identical loading (0.2 mg cm⁻²) and testing conditions, both sulfide and phosphide derivatives of CoFe-PBA outperformed IrO_x: the PBA-derived catalyst required only 260 mV overpotential to reach 10 mA cm⁻², whereas IrO_x needed 285 mV to achieve the same current density [58]. To enable systematic comparison, we have compiled performance data of selected noble metal benchmark catalysts reported in the literature (

Table 1. The OER performance of noble metal catalysts reported in some reference.

Sample	Electrolyte	Overpotential ηmA cm−2 [mV]	Tafel [mV dec−1]	Ref.
IrOx	1M KOH	η10 = 285	55.4	[58]
IrO2	1M KOH	η10 = 351	78	[59]
RuO2	1M KOH	η10 = 326	64	[60]
Ir/C	1M KOH	η10 = 325	46.6	[61]
IrO2	1M PBS	η10 = 418	134.6	[62]
Pt/C RuO2	1M KOH	η10 = 566 η10 = 377	151.2 88.6	[63]
RuO2	1M KOH	η10 = 338	120.4	[64]
IrO2	1M KOH	η50 = 460	94	[65]
RuO2	1M KOH	η10 = 303	120.1	[66]
RuO2	1M KOH	η10 = 290	135	[67]

) in the Appendix for reference.

Among the existing high-quality reviews, some studies begin with the preparation methods of PBA materials and, leveraging their structural advantages, comprehensively elaborate on various energy-related applications—including the OER—thereby establishing an integrated conceptual framework for readers. Another category of reviews adopts a more focused perspective, systematically examining PBA synthesis strategies, the elemental composition and types of derived catalysts, as well as morphological control, to provide a detailed analysis of the distinct advantages offered by different categories within electrocatalytic water splitting systems. Although these reviews demonstrate both breadth and depth, there remains a relative scarcity of systematic reviews that specifically address the enhancement mechanisms of PBA-based OER catalysts from the dual perspectives of “increasing the number and utilization efficiency of active sites” and “improving the intrinsic activity”. Based on this identified gap, this review will adopt such a classification as a novel analytical lens to reassess and prospect the research progress of PBA materials in the field of the OER.

As a representative electrocatalytic process, OER catalyst design adheres to general principles governing electrocatalytic reactions. Recent studies indicate that designing highly efficient electrocatalysts requires careful consideration of two fundamental parameters: the number of active sites and their intrinsic activity [68]. Under ideal conditions, these two factors should act synergistically to maximize catalytic performance. Building on previous research, this review focuses on design strategies for PBA-derived OER catalysts, addressing two key aspects: the number and utilization efficiency of active sites, and the modulation of their intrinsic activity. To enhance the number and utilization of active sites, this review examines strategies including the construction porous architectures, creation hierarchical porosity, and improvement electrical conductivity. For improving intrinsic activity, strategies such as elemental doping and defect engineering are discussed (Figure 2).

This review aims to offer a novel perspective on the application of PBA-based materials in OER, systematically elucidating the underlying mechanisms for enhancing catalytic performance. We anticipate that this systematic analysis will provide new insights for researchers, promote the integrated and synergistic application of multiple optimization strategies for PBA-derived catalysts, and ultimately accelerate progress in OER and related energy technologies.

2. Synthesis Method of PBAs

PBA hold significant application value in the OER, serving not only as direct catalysts or pre-catalysts but also as precursors for derived OER catalysts. Consequently, the synthesis methods of PBA are particularly crucial. Before systematically discussing strategies for increasing the number of active sites, improving their utilization efficiency, and enhancing intrinsic activity, it is necessary to briefly outline their synthesis pathways. Given that many cutting-edge reviews have elaborated on this topic in detail, this section provides only a general overview. Currently, PBA synthesis primarily relies on liquid-phase methods, with the most common being liquid-phase coprecipitation, hydrothermal/solvothermal synthesis, and electrodeposition. As the importance of PBAs becomes increasingly prominent and related research deepens, an efficient solid-phase synthesis strategy has also been gradually developed and reported. This approach typically utilizes mechanical grinding to achieve the preparation of PBA nanoparticles.

Liquid-phase coprecipitation is generally conducted at room temperature and atmospheric pressure, involving the coprecipitation reaction of metal salts and cyanometallates in solution to form PBA. This method offers advantages such as simple operation and ease of scaling up. In addition to the facile tunability of elemental composition, researchers often introduce surfactants (e.g., polyvinylpyrrolidone, sodium dodecylbenzenesulfonate, trisodium citrate, etc.) into such reactions to control the morphology of the resulting PBA [69,70,71,72]. This method exhibits strong controllability; adjusting the ratios of different solutions or solutes, or varying the amount of surfactant, can effectively modulate the final morphology of PBA. For instance, Hu et al. achieved a morphological transition of CoZn PBA from microspheres to micropolyhedrons simply by adjusting the feeding ratio of Zn(CH₃COO)₂·2H₂O to K₃[Co(CN)₆] [71]. In the study by Cui et al., trisodium citrate was introduced into pre-synthesized NiCo-PBA nanocubes, and through a heating reaction for etching, nano orthogonal frustums were successfully prepared [72]. In recent years, numerous researchers have successfully prepared hollow-structured PBA materials using the sacrificial template method. Essentially, this approach can be regarded as a specialized form of coprecipitation reaction. Commonly used sacrificial templates include various materials such as oxides, LDHs, and MOFs [13,73,74,75]. The formation process primarily consists of two stages: first, the metal ions released from the sacrificial template coordinate with [M(CN)₆]^x− ions in the solution, forming a PBA shell on the template surface; subsequently, as the reaction proceeds, the template is continuously consumed. Due to the difference in diffusion rates between the released cations and [M(CN)₆]^x− ions, a structurally intact hollow PBA material is ultimately obtained [74,75].

The hydrothermal/solvothermal method is typically conducted in sealed autoclaves under high-temperature and high-pressure conditions to promote crystal growth, generally yielding PBA materials with well-defined morphologies and high crystallinity [76]. Studies have shown that the crystal growth of PBAs—materials with extremely low solubility—follows a non-classical pathway involving mesoscale self-assembly followed by the fusion of adjacent crystalline grains [77]. This mechanism enables diverse structural possibilities in the hydrothermal synthesis of PBA, applicable to the preparation of both nanoparticles and thin films. In the context of nanoparticles, Xu et al. synthesized Ni₂[Fe(CN)₆]·xH₂O nanoparticles via a hydrothermal method. By controlling the composition and concentration of solvents and solutes during the hydrothermal process, they successfully regulated the morphology of the nanoparticle [78]. In the work of Yang et al., a PB thin film was successfully deposited on conductive glass through hydrothermal synthesis. This film allowed for tunable color, structure, and properties by adjusting reactant ratios, reaction temperature, and reaction duration [79].

Electrodeposition typically refers to a technique that directly deposits PBA thin films onto electrode surfaces through electrochemical processes, making it suitable for fabricating integrated electrodes tightly bonded with conductive substrates. This method commonly employs electrochemical modes such as potentiostatic, galvanostatic, or cyclic voltammetry to achieve controlled growth of PBA on various substrates [80]. Electrodeposition also demonstrates remarkable effectiveness in regulating the structure and morphology of PBA. For instance, Baggio et al. successfully achieved preferential crystal facet growth of PB and Prussian White (PW) by modulating the deposition potential [81]. Furthermore, Goux et al. investigated the electrodeposition behavior of PB within extremely narrow channels, providing new insights for morphology control based on confined templates [82].

Although liquid-phase synthesis methods have been extensively studied for the preparation of PBAs, they still present certain limitations, such as prolonged synthesis duration, requirement for multiple chemical reagents, complex process control and operational steps, or reliance on high temperature/pressure or electrochemical assistance. In response, a novel solid-phase synthesis strategy has been developed. This approach involves direct mechanical grinding of metal salts and cyanometallates, enabling one-step reaction completion at room temperature [53,83]. Notably, this method not only simplifies the process but also allows precise modulation of cation and anion defect sites, thereby further endowing the materials with enhanced properties and application potential.

3. Strategies for Enhancing Active Site Accessibility and Utilization

It is well-known that the catalytic performance for the OER is directly related to the number of active sites. Generally, the catalyst performance improves as the number of active sites increases. However, it is important to note that excessively pursuing the quantity of active sites does not always lead to optimal catalytic performance, as limitations such as bubble accumulation and liquid-phase surface tension during the OER process may hinder mass transport. Therefore, current research not only focuses on increasing the number of active sites but also places growing emphasis on optimizing their utilization efficiency. Notably, catalyst systems represented by PBAs and their derivatives have made significant progress in synergistically enhancing both the number and utilization efficiency of OER active sites, providing important support for the development of this field.

3.1. Constructing Porous Architectures to Increase Active Site Number

Constructing nanoporous materials is one of the most effective strategies to increase the number of active sites in catalysts. Nano-sized PBAs, as a class of inorganic coordination polymers, possess inherent advantages for forming nanoporous structures. The cyanide (CN) groups in PBAs exhibit behavior similar to organic ligands in MOFs during pyrolysis, decomposing into gaseous species that escape and create pores, while the metal components form corresponding compounds. For example, Chen et al. developed a controllable method for preparing porous NiCoOₓ nanocubes [84]. This material can be easily synthesized by thermal decomposition of Ni-Co PBA in air at different temperatures. The NiCoOₓ-400 catalyst obtained by heat treatment at 400 °C exhibits a porous nanocubic morphology (Figure 3a) and a relatively high specific surface area of 20 m²/g, as revealed by N₂ adsorption–desorption measurements (Figure 3b). Additionally, this nanoporous cube demonstrates excellent OER performance. This strategy of using PBA pyrolysis to easily synthesize catalysts with highly exposed active sites is undoubtedly an effective approach. Similarly, Bhattacharyya et al. prepared porous NiFe oxide nanocubes using this method, which also showed promising OER performance [59]. Fang et al. further improved the morphology of PBA precursors by employing a two-dimensional PBA to prepare oxides, advancing the material’s performance [85]. Besides direct application in OER, PBA-derived nanoporous oxides can also be combined with noble metals to further enhance their OER performance [60,86].

In addition to complete pyrolysis, some studies have adopted a partial decomposition strategy, where low-temperature heat treatment is applied to introduce pores while retaining the basic framework of the PBA. Research shows that the CoFe-PBA porous nanocube treated at 200 °C (CoFe-200) (Figure 3c) exhibits optimal OER activity, even outperforming expensive commercial Ir/C catalysts [61]. To further investigate the origin of the superior performance of CoFe-200, the authors conducted a systematic study. In addition to noble metal catalysts, they included CoFe-PBA derivatives calcined at 550 °C—which feature hierarchical pore channels—in the comparative analysis. N₂ adsorption–desorption tests revealed that this material contains abundant 1.2 nm micropores and possesses a high specific surface area of 421 m² g⁻¹. The study pointed out that although the microporous structure may prolong mass transport pathways, the significantly increased density of active sites still creates favorable conditions for the OER. Based on these findings, the authors concluded that precise regulation of the pore size distribution and micro-/mesopore content in CoFe-based catalysts enables the construction of highly efficient electrocatalytic systems suitable for the OER. Here we must emphasize that while focusing on increasing the number of active sites, it is equally crucial to investigate whether the nature of these sites undergoes fundamental changes. In fact, many cutting-edge studies demonstrate that strategies aimed at regulating the number of active sites often simultaneously enhance their intrinsic activity. These two aspects are never mutually exclusive but rather work synergistically to drive breakthroughs in catalytic performance. For example, Hu’s research team observed in their study of LDHs that OER performance improved progressively with increasing electrochemical active surface area (ECSA) [87]. However, when the LDHs was further fragmented into nanostructures, the performance enhancement far exceeded the increase in ECSA. This is attributed to the generation of a large number of edge active sites with dangling bonds during the fragmentation process, which significantly boosted the intrinsic activity. This case clearly illustrates that the optimization of OER performance often stems from the dual enhancement of both the quantity and intrinsic activity of active sites. Although different studies may emphasize one aspect over the other, it can be unequivocally stated that ideal catalyst design typically requires simultaneous consideration of both the “quantity” and “quality” of active sites.

Furthermore, beyond pyrolysis, the high tunability of PBAs allows for the preparation of nanoporous OER catalysts through liquid-phase conversion. For instance, while Lou’s research group prepared oxides and phosphides under different atmospheres via pyrolysis, they also synthesized porous Ni(OH)₂ as a comparative OER catalyst using a solution-based method (Figure 3d) [88]. It is worth emphasizing that, in addition to oxides and hydroxides, PBA precursors can also be used to synthesize various nanoporous materials such as sulfides, phosphides, and selenides by adjusting the pyrolysis atmosphere or liquid-phase environment [88,89,90,91], greatly expanding their application prospects in OER.

Figure 3. TEM images of (a) NiCoO_x-400; insets are their corresponding high-magnification TEM images; and (b) N₂ adsorption–desorption isotherms of NiCoO_x-400. Reprinted with permission from Ref. [84]. Copyright 2018, Elsevier B.V. (c) Pore-size distributions and BET surface areas (inset) of CoFe-200 composites. The right panels show schematic diagrams of the proposed OER mechanisms CoFe-200 and the OER was favored by microporous CoFe-200. Reprinted with permission from Ref. [61]. Copyright 2020, WILEY-VCH Verlag GmbH & Co. (d) Scheme for the formation of Ni(OH)₂, NiO, and Ni-P porous nanoplates from NiNi-PBA. Reprinted with permission from Ref. [88]. Copyright 2016, The Royal Society of Chemistry.

Figure 3. TEM images of (a) NiCoO_x-400; insets are their corresponding high-magnification TEM images; and (b) N₂ adsorption–desorption isotherms of NiCoO_x-400. Reprinted with permission from Ref. [84]. Copyright 2018, Elsevier B.V. (c) Pore-size distributions and BET surface areas (inset) of CoFe-200 composites. The right panels show schematic diagrams of the proposed OER mechanisms CoFe-200 and the OER was favored by microporous CoFe-200. Reprinted with permission from Ref. [61]. Copyright 2020, WILEY-VCH Verlag GmbH & Co. (d) Scheme for the formation of Ni(OH)₂, NiO, and Ni-P porous nanoplates from NiNi-PBA. Reprinted with permission from Ref. [88]. Copyright 2016, The Royal Society of Chemistry.

3.2. Creating Hierarchical Porosity to Enhance Mass Transport

In the OER, catalytic performance does not always increase linearly with the number of active sites. This is primarily due to the limited mass transfer depth in electrochemical processes and the obstruction of pores by bubbles generated from gas-evolving reactions, which impede effective contact between the electrolyte and the internal active sites, leading to incomplete utilization of the pore channels. To improve the availability of active sites, many studies have focused on constructing hierarchically porous structures to address this issue. This strategy not only preserves mesopores to maintain a high density of active sites, but also incorporates larger pores to alleviate mass transfer limitations and bubble blockage caused by surface tension [92]. Owing to the highly tunable morphology of PBAs, several studies have utilized them to successfully fabricate catalysts with hierarchical porous architectures, providing an effective pathway toward high-performance OER. To construct hierarchically porous structures for high-performance OER catalysts, Lou’s research group etched Ni–Co PBA in an ammonia solution, successfully preparing hollow-structured nanoboxes. Subsequent thermal treatment yielded Ni–Co oxide catalysts. The well-defined structural features of PBA facilitated the investigation of the formation mechanism, revealing that the etching process initiates at coordinatively unsaturated sites and progressively extends inward, ultimately leading to the hollow structure (Figure 4a). After calcination, the resulting product consists of nanoporous boxes with a macroporous hollow architecture (Figure 4b). Performance comparisons demonstrate that the OER activity of this catalyst is significantly higher than that of nanoporous cubes with similar specific surface area but lacking hierarchical pore channels (Figure 4c) [93]. This method has also been employed in other studies to etch Co PBA and Ni–Fe PBA [94,95,96]. The same research group further converted these structures into selenides, which exhibited even more exceptional OER performance [95,96].

In addition to ammonia solution, etching agents such as urea solution, ethanol solution, HCl solution and Na₂S solution have also been used to prepare hierarchically nanoporous materials [97,98,99,100,101]. The etching by urea may be attributed to the release of NH₄⁺ ions into the solution upon heating, with a mechanism analogous to that of ammonia etching. It is particularly noteworthy that ethanol solution, while being the most commonly used solvent, functions similarly to HCl solution in etching PBA [97]. Feng et al. conducted a detailed investigation on this phenomenon, successfully synthesizing hollow-structured PBA by subjecting PBA to a solvothermal reaction in ethanol solution. The ionization equilibrium within the solvothermal system enables ethanol to release sufficient protons—a phenomenon corroborated by the authors through measurements of pH change before and after the reaction, confirming continuous proton consumption during the process. This leads to the conclusion that the hollow architecture results from etching of the PBA framework by protons released from ethanol under solvothermal conditions. During the reaction, in situ dissociated protons diffuse through the intrinsic channels of PBA toward the interior, driven by concentration gradients. Once the local proton concentration at the core reaches a threshold sufficient to break chemical bonds, etching progressively propagates along the high-energy body diagonals of the crystal toward the vertices. Subsequent electrochemical evaluations demonstrate that the hollow-structured material exhibits significantly enhanced OER performance compared to the unetched nanocubes. The formation mechanism of sulfide frameworks via Na₂S solution etching of PBA can be attributed to the synergistic effect of structurally guided anisotropic chemical etching/anion exchange and a nanoscale Kirkendall-type process [100,102]. Typically, the edge and corner regions of PBA nanocubes contain coordinatively unsaturated sites and a high density of defects, which serve as initial active centers for chemical etching and anion exchange. The [M(CN)₆]³⁻ groups at these locations preferentially undergo substitution reactions with S²⁻ ions, forming a thin sulfide shell on the cube surface. As the reaction proceeds, the less-etched lateral regions of the crystal planes are exposed and become secondary reactive sites, promoting the inward progression of etching. According to the Kirkendall effect, when PBA nanocubes react with Na₂S solution, the outward diffusion rate of metal ions exceeds the inward permeation rate of S²⁻ ions. This asymmetric interdiffusion leads to the formation and coalescence of vacancies at the interface. Newly formed sulfide nuclei continuously promote the growth of the preformed framework at the edges, while the core region of the cube is gradually consumed due to vacancy accumulation and progressive etching. The cooperation of these two mechanisms ultimately drives the formation of well-defined sulfide nanoframes.

Figure 4. (a) The corresponding schematics illustrate the process of the Ni–Co PBA nanocages. (b) TEM images of the Ni–Co mixed oxide nanocages. (c) Polarization curves of the Ni–Co mixed oxide cages and porous cubes. Reprinted with permission from Ref. [93]. Copyright 2016, WILEY-VCH Verlag GmbH & Co. (d) Schematic illustration of the synthesis process for hierarchically porous NiO/NiFe₂O₄. Reprinted with permission from Ref. [103]. Copyright 2017, The Royal Society of Chemistry. (e) Schematic illustration of the preparation of hollow Fe-CoP nanoprisms. Reprinted with permission from Ref. [75]. Copyright 2021, Elsevier B.V.

Figure 4. (a) The corresponding schematics illustrate the process of the Ni–Co PBA nanocages. (b) TEM images of the Ni–Co mixed oxide nanocages. (c) Polarization curves of the Ni–Co mixed oxide cages and porous cubes. Reprinted with permission from Ref. [93]. Copyright 2016, WILEY-VCH Verlag GmbH & Co. (d) Schematic illustration of the synthesis process for hierarchically porous NiO/NiFe₂O₄. Reprinted with permission from Ref. [103]. Copyright 2017, The Royal Society of Chemistry. (e) Schematic illustration of the preparation of hollow Fe-CoP nanoprisms. Reprinted with permission from Ref. [75]. Copyright 2021, Elsevier B.V.

Other studies have utilized a self-templated epitaxial growth strategy to fabricate polycrystalline hollow nanostructured PBAs. Owing to the well-defined crystal structure of PBA, researchers were able to gain deep insights into the growth kinetics, achieving controlled synthesis of Co–Fe-based PBA cages, frames, and boxes with diverse geometries. The resulting Co–Fe hierarchically nanoporous oxides obtained through subsequent calcination demonstrated superior OER performance [104,105]. In contrast to the approach of first constructing hollow PBAs followed by calcination, hollow porous NiO/NiFe₂O₄ cubes can also be directly prepared by modulating calcination conditions and used as efficient OER catalysts (Figure 4d) [103].

Template-based methods have long been effective for constructing hierarchically porous materials. Accordingly, several studies have employed sacrificial templates to fabricate macrostructured PBA precursors, which were subsequently converted into hierarchically nanoporous OER catalysts. For example, Ding et al. first synthesized Co-acetate-based prism precursors, then induced transformation via ion exchange between cobalt hydroxy species and [Fe(CN)₆]³⁻ to form hollow CoFe–PBA prisms, ultimately obtaining hollow Fe–CoP materials. The resulting Fe–CoP not only retained the hollow prismatic morphology of CoFe–PBA but also developed abundant mesoporous channels (Figure 4e) [75]. Similarly, other coordination compounds such as MOFs and LDHs can also be used to prepare hierarchically nanoporous OER catalysts via this strategy, demonstrating performance far exceeding that of conventional porous materials [74,106,107].

3.3. Improving Electrical Conductivity for Efficient Electron Transfer

In electrochemical reactions, even with sufficient contact between the electrolyte and the catalyst, active sites remain inaccessible if electrons cannot transfer at the contact points. Therefore, many researchers have devoted extensive efforts to improving the electrical conductivity of nanocatalysts. Among various strategies, the integration of carbon materials as excellent conductive carriers with PBAs has attracted significant attention. Carbon nanotubes (CNTs), as highly efficient conductive supports, have been coupled with PBAs and applied in the OER [108,109,110]. For instance, Ho et al. reported an oxygen plasma-activated hybrid structure composed of nickel-iron PBA and CNTs (O–CNT/NiFe) [109]. Their study systematically adjusted the mass ratio between NiFe-PBA and CNTs, not only confirming that the incorporation of CNTs enhances OER performance but also identifying the optimal PBA-to-CNT ratio (Figure 5a,b). The results indicate that CNT-composited PBA holds great promise as a highly active electrocatalyst. Another highly conductive two-dimensional material, graphene, has also been widely employed. Nossol et al., for example, prepared a composite derived from reduced graphene oxide and cobalt-nickel hexacyanoferrate (rGO/CoNiHCF) [111]. Through alkaline electrochemical treatment, the uniformly dispersed globular CoNiHCF particles on rGO sheets were transformed into (oxy)hydroxide species, resulting in a composite exhibiting excellent OER kinetics. Some researchers have also adopted the strategy of first fabricating carbon materials with specific morphologies and then loading PBAs onto them [112,113], which has demonstrated considerable advantages in enhancing OER performance. It is worth emphasizing here that electrospinning technology, due to its capability to produce one-dimensional nanofibers with high specific surface area and excellent electron transport properties, has been widely applied in the OER field. This technique can not only be used to synthesize heteroatom-doped carbon nanofibers but also to load transition metal nanoparticles within them [114]. Leveraging this advantage, researchers have successfully encapsulated PBAs into nanofibers prepared by electrospinning [62,115,116]. For instance, Zhang et al. encapsulated Co-Fe PBA into polyacrylonitrile (PAN) nanofibers via electrospinning (Figure 5c), followed by calcination at different temperatures under an argon atmosphere, ultimately preparing a material consisting of nitrogen-doped carbon nanofibers encapsulating FeCo alloy nanoparticles (FeCo-NCNFs-Ts) (Figure 5d,e) [115]. Systematic research revealed that the addition amount of Co-Fe PBA and the carbonization temperature significantly influence the catalytic performance of the resulting FeCo-NCNFs-Ts.

Although the CN group is an inorganic ligand, it exhibits functional similarities to MOFs in certain aspects. For example, when PBAs are pyrolyzed under an inert atmosphere, a nitrogen-doped graphene layer forms on the surface of the resulting alloy. This feature significantly enhances their OER performance. It is noteworthy to elucidate the formation process of graphene-like carbon shells derived from MOFs. Initially, carbon-containing organic ligands undergo thermal decomposition in an inert atmosphere, leading to in situ carbonization and the formation of amorphous carbon. As the temperature further increases, this amorphous carbon gradually undergoes phase transformation into graphitic carbon materials. Significantly, these carbon species can function as reducing agents at elevated temperatures, converting metal ions into metallic elements. Studies have demonstrated that metal ions with standard reduction potentials below −0.27 V can generally be reduced through this process [117]. Similarly, cyanide groups in PBA can also be initially converted into amorphous carbon under an inert environment, which subsequently graphitizes into carbon materials [118,119]. Li’s research team conducted a systematic investigation using CoCo-PBA as a model system through thermogravimetric analysis, XRD, and Raman spectroscopy [118]. Their findings confirmed that when the temperature reaches 600 °C, the characteristic structure of PBA completely disappears, coinciding with the initiation of graphitic carbon formation. One study utilized a pyrolysis strategy to successfully synthesize Co₃ZnC/Co nanomaterials encapsulated within nitrogen-doped graphene layers (Figure 5f,g) and applied them to OER [63]. The optimal sample, S-650, demonstrated superior performance to state-of-the-art noble metal catalysts, owing to its unique structural features (Figure 5h). In another study, researchers prepared NiFe@C materials and systematically investigated the influence of PBA precursor size on the resulting OER catalyst performance, leveraging the morphological tunability of PBAs [120].

Figure 5. (a) Schematic illustration of the synthesize O−CNT/NiFe and (b) LSV curves. Reprinted with permission from Ref. [109]. Copyright 2020, American Chemical Society. (c) SEM images of Co-Fe PBA@PAN nanofibers. (d,e) TEM images of FeCo-NCNFs-Ts. Reprinted with permission from Ref. [115]. Copyright 2019, American Chemical Society. (f) TEM images and (g) EDS lines profiles of S-650; (h) LSV curves of S-650, RuO₂ and Pt/C. Reprinted with permission from Ref. [63]. Copyright 2016, The Royal Society of Chemistry.

Figure 5. (a) Schematic illustration of the synthesize O−CNT/NiFe and (b) LSV curves. Reprinted with permission from Ref. [109]. Copyright 2020, American Chemical Society. (c) SEM images of Co-Fe PBA@PAN nanofibers. (d,e) TEM images of FeCo-NCNFs-Ts. Reprinted with permission from Ref. [115]. Copyright 2019, American Chemical Society. (f) TEM images and (g) EDS lines profiles of S-650; (h) LSV curves of S-650, RuO₂ and Pt/C. Reprinted with permission from Ref. [63]. Copyright 2016, The Royal Society of Chemistry.

Beyond compositing catalysts with conductive materials, an alternative strategy to address electron transfer issues is the direct growth of PBAs on conductive substrates, such as conductive glass, Si substrate, Ni foam, Cu foil, stainless steel, and carbon cloth [13,57,121,122,123,124,125]. This approach not only improves electrical conductivity but also further enhances the utilization of active sites. A key advantage is the complete avoidance of binders, effectively preventing the coverage of active sites and ensuring more catalytic centers are exposed and participate in reactions. For example, Lei et al. successfully prepared a CoFe-PBA thin film on a silicon substrate using a liquid-phase epitaxy method (Figure 6a) [122]. This preparation method demonstrated good universality, as the research team subsequently achieved uniform growth of CoFe-PBA films on Ni foam and Cu foam substrates, further obtaining an integrated CoFe₂O₄ OER electrode through calcination (Figure 6b). Another similar strategy involves first growing an array structure on the substrate followed by chemical conversion. This approach can construct OER electrodes with a hierarchically porous structure, which optimizes the electron conduction pathway while effectively enhancing the mass transport efficiency of the reaction. For example, Zhang’s research group used cobalt hydroxide or oxide nanoarrays as precursors and structural templates to guide the growth of nano-PBAs (Figure 6c). The resulting catalyst exhibited excellent OER performance, requiring an overpotential of only 258 mV to achieve a current density of 10 mA·cm⁻² (Figure 6d) [57]. The team also conducted comprehensive studies optimizing template selection, conductive supports, elemental composition, and reaction time, providing valuable insights into the design of PBA-based OER electrodes. In another work, they precisely controlled the morphology and growth rate of PBAs by adjusting the pH of the reaction solution [123].

4. Strategies for Improving the Intrinsic Activity

While increasing the number and utilization efficiency of active sites effectively enhances the overall current density, improving the intrinsic activity of each site is equally crucial for fundamentally reducing the OER overpotential and accelerating reaction kinetics. Intrinsic activity is primarily governed by the electronic structure of the catalytic centers, which directly determines the adsorption/desorption behavior of reaction intermediates (e.g., *OH, *O, *OOH). In this context, the d-band center theory offers a robust framework for understanding these effects. This theory posits that the energy position of metal d-orbitals relative to the Fermi level correlates with the adsorption strength of intermediates. Generally, a downshifted d-band center weakens adsorption, whereas an upshifted d-band center strengthens it. Optimal intrinsic activity is achieved when the adsorption energies of these intermediates are balanced, minimizing the activation energy barrier of the rate-determining step [126,127]. Strategies such as metal element doping and defect engineering, employed to enhance the intrinsic activity of PBA-based catalysts, can be viewed as sophisticated approaches to modulate the local coordination environment, induce lattice strain, and trigger charge redistribution. These modifications directly perturb the electronic structure of the material, thereby fine-tuning the d-band center and the charge density at active sites. Ultimately, this leads to a more favorable OER adsorption energy landscape, significantly boosting the intrinsic activity [128]. It is important to emphasize that improving intrinsic activity and fully utilizing active sites are not mutually exclusive; ideally, they can be synergistically combined to significantly enhance the performance.

4.1. Regulating Electronic Structure via Elemental Doping

The intrinsic activity of nanocatalysts can typically be finely modulated by adjusting their surface elemental composition. As a result, various binary, ternary, and even high-entropy nanometallic materials offer unprecedented opportunities for the OER. Studies have demonstrated that, compared to single-metal catalysts, multimetal catalysts exhibit significant advantages in enhancing OER performance—a conclusion that has been extensively validated through both theoretical calculations and experimental investigations [129,130].

In terms of elemental composition regulation, PBAs hold significant advantages. A wide variety of PBAs incorporating different metal elements—such as Co, Ni, Fe, Mn, Cu, Zn, and Ti—have been successfully reported [83,131,132,133]. PBAs not only allow flexible selection of metal species but also enable precise control over their ratios, providing a convenient platform for exploring OER catalysts with optimal compositions. For example, Hu’s group developed a strategy using a bimetallic Co-Fe PBA with homogeneous elemental distribution at the molecular scale as an ideal precursor, from which Co-Fe phosphides (Co₁₋ₓFeₓP) with a continuously adjustable Co/Fe ratio were synthesized [134]. They systematically evaluated the OER activity (Figure 7a) and used this system to construct a composition-activity volcano plot for Co₁₋ₓFeₓP in OER (Figure 7b). It was found that the material exhibited optimal performance at a Co/Fe ratio of 1.63, requiring overpotentials of only 230 mV and 268 mV to achieve current densities of 10 mA cm⁻² and 100 mA cm⁻², respectively. Similarly, Marshall et al. conducted an in-depth study on the Co-Mn component system [135].

Furthermore, it is necessary to provide a more in-depth interpretation of the role of phosphides in the oxygen evolution reaction. Several researchers have successfully integrated anion modulation strategies into multi-element doping systems [136]. In the PBA-driven OER applications discussed earlier, significant research progress has been made involving phosphide-based systems, including nanoporous materials, hierarchical nanostructures, and integrated electrodes, among others. During the OER process, phosphides typically function as pre-catalysts. Compared to pure metals or oxides, phosphides are more prone to undergo surface reconstruction under OER potentials, transforming into highly active (oxy)hydroxides. This reconstruction process often results in an active phase with richer defects, a higher degree of amorphousness, and a larger specific surface area, thereby significantly increasing the exposure of active sites. If an incompletely oxidized phosphide core remains within the material, it can serve as an efficient electron conduction pathway, greatly enhancing charge transfer in the surface active shell and effectively reducing the reaction overpotential. From an electronic structure perspective, the introduction of phosphorus alters the electron cloud distribution of the central metal atoms, inducing a higher apparent oxidation state in the metal. This electronic structure optimization facilitates the formation of key high-valence metal active centers during electrochemical oxidation, which are crucial for the adsorption and conversion of OER intermediates [46,137,138]. Further studies have indicated that PO₄³⁻ groups generated during phosphide oxidation can adsorb onto newly formed active sites, further enhancing the intrinsic catalytic activity by optimizing the adsorption energy of intermediates [139]. Similarly, based on the anion modulation strategy, sulfides and selenides are also fundamentally categorized as pre-catalysts among common types of PBA-derived OER catalysts. Like phosphides, they undergo surface reconstruction under OER potentials to form a highly active surface layer of (oxy)hydroxide. Typically exhibiting metallic or narrow-bandgap semiconductor properties, these materials can maintain partially unconverted sulfide or selenide cores during OER. Such preserved cores not only provide efficient electron conduction pathways but also modulate the electronic states of surface active sites through their distinctive electronic structure. This synergistic effect optimizes the adsorption free energy of oxygen intermediates and significantly enhances the intrinsic catalytic activity [46,140,141]. Furthermore, the in situ generated SO₄²⁻ or SeO₄²⁻ that adsorb onto active sites can effectively regulate the local electronic environment of metal centers. This reinforcement of adsorption stability for key reaction intermediates consequently reduces the overall energy barrier of OER [142].

In addition to bimetallic systems, ternary PBAs can also be readily obtained [143,144,145]. For instance, Chen’s team synthesized a hollow multivoid nanocuboidal catalyst based on a ternary Ni-Co-Fe PBA precursor [143]. By leveraging the differing ion exchange rates of the three transition metals within the PBA precursor, they constructed an interconnected porous structure, achieved heteroatom doping, and optimized the electronic structure. This resulted in excellent electrocatalytic OER performance that surpassed that of binary catalysts prepared by the same method. Guided by a similar rationale, Guo et al. also used a PBA precursor to prepare NiCoFe-Se/CFP, which showed significantly superior OER activity compared to single-metal and binary catalysts [144]. The strong elemental compatibility of PBAs also supports the construction of hierarchical architectures such as core–shell structures, broadening their potential for OER applications [64,146].

In recent years, high-entropy catalysts have emerged as a research hotspot in the OER field, and PBAs have demonstrated unique advantages in the preparation of such materials. For example, Zhang et al. prepared a PBA containing Ni, Fe, La, Mo, and Co (NiFeLaMoCo-PBA), and employed it as an OER catalyst [147]. Elemental mapping images clearly revealed the homogeneous distribution of the elements within the material (Figure 7c). Consequently, the catalyst exhibits superior catalytic activity and reaction kinetics compared to other contrast samples. For the OER, NiFeLaMoCo-PBA requires an overpotential of 244 mV to achieve a current density of 10 mA cm⁻², along with a low Tafel slope of 32 mV dec⁻¹, fully demonstrating its excellent electrocatalytic performance (Figure 7d,e). The significant performance enhancement visually demonstrates the advantages of high-entropy catalysts, yet the mechanism by which they promote the OER is highly complex. The study first confirmed the presence of electronic structure modulation and charge transfer within the material: after the OER, the electron density of Fe and Mo decreased, while the valence state of Co changed, indicating significant electronic interactions within the system. Specifically, Fe/Mo sites may act as electron donors, while Ni/Co sites serve as electron acceptors. This charge complementarity and redistribution mechanism effectively optimizes the adsorption free energy of OER intermediates on the active Ni/Co sites, thereby significantly reducing the energy barrier of the rate-determining step. On the other hand, the introduction of multiple metal ions induces lattice distortion. The larger ionic radius of La³⁺ facilitates the formation of strong bonds with surrounding O²⁻ ions, which helps suppress the disordered expansion of lattice distortion, thereby enhancing the structural stability of the material under harsh electrochemical conditions. More importantly, the random distribution of elements such as Ni, Fe, Mo, and Co within the lattice induces controllable lattice distortion on a global scale. This effect not only increases the defect density of the material, exposing more active sites, but also optimizes the orbital hybridization between metal d-orbitals and reaction intermediates by modulating the electronic states. Unlike traditional catalysts with relatively single active sites, multiple elements in this high-entropy system can potentially serve as active centers, effectively avoiding performance degradation caused by overly strong adsorption or high desorption energy barriers at single sites. This multi-center synergistic mechanism collectively contributes to an efficient and stable OER process. In another study, Xu et al. also prepared a high-entropy PBA; the derived catalyst not only showed good OER activity but also exhibited potential in other electrocatalytic applications [148].

Figure 7. (a) LSV curves of Co_1−xFe_xP. (b) Relationship between Fe doping level and the overpotentials at 10 mA cm⁻² for as-prepared Co_1−xFe_xP with various Co/Fe ratios. Reprinted with permission from Ref. [134]. Copyright 2018, American Chemical Society. (c) HAADF-STEM image and elemental mapping images of NiFeLaMoCo-PBAs. (d) LSV curves and (e) Tafel plots of NiFeLaMoCo-PBA, the contrast samples and RuO₂. Reprinted with permission from Ref. [147]. Copyright 2025, The Royal Society of Chemistry.

Figure 7. (a) LSV curves of Co_1−xFe_xP. (b) Relationship between Fe doping level and the overpotentials at 10 mA cm⁻² for as-prepared Co_1−xFe_xP with various Co/Fe ratios. Reprinted with permission from Ref. [134]. Copyright 2018, American Chemical Society. (c) HAADF-STEM image and elemental mapping images of NiFeLaMoCo-PBAs. (d) LSV curves and (e) Tafel plots of NiFeLaMoCo-PBA, the contrast samples and RuO₂. Reprinted with permission from Ref. [147]. Copyright 2025, The Royal Society of Chemistry.

4.2. Optimizing Active Sites via Defect Engineering

Building upon the effectiveness of elemental doping in enhancing the activity of OER catalysts, further design and modulation of material structures can contribute to deeper optimization of catalytic performance. Studies have shown that constructing vacancy defects on the catalyst surface is a key strategy for regulating its electronic structure, among which oxygen vacancies (V_O) and metal vacancies (V_M) are the most common types [149,150,151,152,153,154,155]. DFT calculations indicate that the presence of V_O significantly reduces the positive charge of adjacent metal atoms, leading to electron accumulation at metal sites and an increased density of electronic states near the Fermi level, thereby promoting the adsorption and dissociation of H₂O molecules [149,150]. Furthermore, Wang’s research team employed a series of in situ characterization techniques to track the evolution of Co valence states, changes in the local coordination environment, and the formation of Co(IV) active species during the OER, revealing the kinetic mechanism by which V_O facilitates the pre-oxidation of Co²⁺ and the formation of reaction intermediates [150]. Regarding V_M, theoretical analyses suggest they also possess the potential to enhance OER performance. V_M can alter the local coordination environment of surrounding metal atoms, promoting their transition to higher valence states. Experimental results demonstrate that higher-valence Co³⁺ species with lower coordination numbers exhibit superior intrinsic catalytic activity [151]. Li et al. further investigated the mechanism of dual-cation vacancy systems [152]. Using in situ Raman spectroscopy, they observed that within the same potential window (1.35–1.45 V), the oxidation of Ni²⁺ occurs synergistically with the electrocatalytic water oxidation process, confirming that the highly disordered structure induced by dual-cation vacancies facilitates a reduction in the energy barrier for the Ni²⁺/Ni³⁺ transition. DFT calculations further reveal that dual-cation vacancies can shift the rate-determining step from O₂ desorption to a deprotonation process. The fundamental reason is that in the O₂ desorption step, the O₂ molecule in the dual-vacancy model binds to only one metal site, requiring the breaking of fewer M–O bonds, thereby significantly lowering the energy barrier of this step.

In the application of PBAs for OER, the predominant type of vacancy defect is the CN vacancy (V_CN). This is closely related to the chemical nature of the CN group. Although the CN triple bond itself is highly stable, its coordination bond with metal centers is relatively weak. Consequently, when PBA materials undergo certain experimental treatments, the entire CN group tends to desorb and escape from the crystal lattice, leading to the formation of V_CN vacancies. In an earlier study, Guo et al. utilized a mild air-plasma approach to prepare PBAs with missing coordination units (Figure 8a) [14]. The reactive oxygen species generated by the air plasma bonded to the open sites of Co, elevating its oxidation state to Co³⁺, while the Co centers retained two open coordination sites occupied by water molecules in a cis configuration. The nanoporous framework structure was preserved after plasma treatment. Benefiting from the highly active and uniformly distributed metal catalytic sites within the nanoporous framework, the resulting OER catalyst required a low overpotential of only 330 mV to achieve a high current density of 100 mA cm⁻² (Figure 8b).

During the same period, studies revealed that when NiFe-PBA is used as an OER catalyst, it essentially functions as a pre-catalyst. Under the oxidative conditions of OER, it undergoes structural reconstruction to form corresponding (oxy)hydroxides, which are the true active species [156]. This behavior is quite similar to that of phosphides, sulfides, and other analogous catalysts. Our previously reported phosphides and sulfides derived from PBA precursors also transformed into amorphous (oxy)hydroxides after the OER process [58]. Based on this, it can be inferred that the PBA-derived phosphides, sulfides, and selenides mentioned earlier likely undergo a similar transformation. However, unlike the above materials, when PBA is directly used as a pre-catalyst for OER, it is often accompanied by the leaching of Fe species. Interestingly, PBA containing V_CN can avoid this issue during OER catalysis. Yu’s research group utilized N₂-plasma to prepare Ni-Fe PBA with V_CN [157]. Test results showed that samples treated with plasma for different durations all exhibited excellent OER performance (Figure 8c). Furthermore, they conducted an in-depth investigation into the mechanism behind the enhanced OER performance due to V_CN. The presence of V_CN in PBAs leads to the formation of numerous coordinatively unsaturated Fe sites (Figure 9c). These unsaturated Fe centers readily bond with oxygen during OER to form Fe–O bonds, which effectively inhibits the leaching of Fe into the electrolyte. In contrast, pristine PBA nanomaterials undergo complete anion exchange between [Fe(CN)₆]⁴⁻ and OH⁻ in alkaline solution, resulting in Fe leaching and subsequent degradation of OER performance (Figure 8d). In subsequent research, Jiang et al. combined DFT calculations with XPS characterization to reveal that the formation of V_CN induces electron transfer between Ni and Fe. Their findings align with those of Yu’s group, demonstrating that V_CN effectively suppresses Fe leaching while promoting the generation of highly active NiOOH/NiFeOOH species, thereby enhancing both OER activity and stability [158]. Ma’s team further validated the argument that “cyanogen vacancies promote the formation of active species” by tracking the surface reconstruction of V_CN-rich N₂-NiCoFe-PBA during OER using in situ Raman spectroscopy. Their study revealed that the characteristic CN peak (2050–2100 cm⁻¹) disappears at 1.12 V vs. RHE in the V_CN-containing sample, whereas a higher potential is required for pristine NiCoFe-PBA. This indicates that V_CN enables CN ligands to be replaced by OH⁻ at lower potentials, significantly accelerating the dynamic reconstruction of the catalyst into the true active phase—metal oxyhydroxides (MOOH) [159].

Figure 8. (a) Schematic diagram of the preparation of PBA containing V_CN using an air plasma strategy. (b) LSV curves of different studied catalysts. Reprinted with permission from Ref. [14]. Copyright 2018, WILEY-VCH Verlag GmbH & Co. (c) LSV curves of PBA samples treated with plasma for different durations. (d) Illustrations of the surface reconstruction on the Ni-Fe PBA and Ni-Fe PBA with V_CN. Reprinted with permission from Ref. [157]. Copyright 2019, Springer Nature.

Figure 8. (a) Schematic diagram of the preparation of PBA containing V_CN using an air plasma strategy. (b) LSV curves of different studied catalysts. Reprinted with permission from Ref. [14]. Copyright 2018, WILEY-VCH Verlag GmbH & Co. (c) LSV curves of PBA samples treated with plasma for different durations. (d) Illustrations of the surface reconstruction on the Ni-Fe PBA and Ni-Fe PBA with V_CN. Reprinted with permission from Ref. [157]. Copyright 2019, Springer Nature.

With advancing research, the construction of V_CN in PBA has become a highly focused research direction in recent years. A significant number of studies have focused on using plasma treatment under different atmospheres (such as O₂, H₂, Ar, N₂, and air) to create V_CN in PBAs and apply them in OER catalysis [14,54,55,56,157,160]. Beyond the plasma approach, researchers have also found that V_CN can be effectively generated by controlling thermal decomposition conditions [65,66,158]. Furthermore, by adjusting the types of ions in the interstitial sites of PBAs, pyrolysis can also fine-tune the catalytic activity of the resulting products [161]. This pyrolysis-based preparation method offers mild conditions and a simple process, enabling the efficient production of high-performance catalysts and providing a highly practical technical pathway for OER research.

Certainly, beyond constructing V_CN, PBAs can also be utilized to introduce other types of defects. For example, both vapor-phase heating and liquid-phase transformation routes can convert PBAs into amorphous compounds. Amorphous materials possess disordered crystal structures containing numerous dangling bonds, which also provides an effective strategy for constructing highly active OER catalysts [162,163]. Among the strategies for improving the intrinsic activity of OER catalysts, constructing heterojunction interfaces is another important approach. The presence of an interface often leads to lattice mismatch, generating defect vacancies that facilitate the adsorption of OER intermediates and enhance catalytic activity. Lin et al. embedded graphene quantum dots (GQDs) into a PBA matrix, successfully preparing GQD-PBA composites (Figure 9a,b) [164]. In this composite catalyst system, the heterogeneous nucleation of GQDs within the NiFe PBA matrix facilitates the construction of abundant defect structures. The lattice constant mismatch between GQDs and NiFe PBA induces significant lattice distortion and mismatch at their interface, thereby introducing numerous grain boundaries, vacancies, and edge sites. This effectively enhances the material’s electron transport capability and increases the number of reactive sites. As previously discussed, multiple improvement strategies can synergistically enhance OER performance. The carboxyl and hydroxyl functional groups abundant on the GQD surface can effectively adsorb Ni²⁺ ions through electrostatic interactions. This not only disperses the nucleation centers of PBA, suppressing its rapid growth, but also significantly reduces the particle size of NiFe PBA cubes, thereby further exposing more active sites. Furthermore, the inherent excellent conductivity of GQDs provides efficient electron transport pathways for the composite system. The defect structures induced by the combination of GQDs and PBA, coupled with the multiple advantages conferred by GQDs themselves, collectively indicate that this composite material is destined to exhibit outstanding electrocatalytic OER performance. In performance studies, after optimizing the GQD ratio in the PBA matrix, the optimal sample required an overpotential of only 259 mV to reach a current density of 10 mA cm⁻² (Figure 9c). In addition to defect vacancies, interfacial interactions can induce electron redistribution at the heterojunction interface, thereby modulating the spin state and valence state of local metal sites. For instance, Lu et al. prepared a MoS₂-coated PBA catalyst (Co^IIIFe-PBA/MoS₂) [165]. The in situ grown MoS₂ shell modulates the electronic structure of Co via interfacial electron regulation, resulting in a high valence state (Figure 9d). This catalyst required overpotentials of only 265 mV and 306 mV to achieve current densities of 10 mA cm⁻² and 100 mA cm⁻², respectively, in OER (Figure 9e).

Composite architectures of this type demonstrate remarkable compositional diversity. Beyond the previously mentioned quantum dots and sulfides, PBA can also be combined with various components such as oxides, phosphides, metal single atoms, and polyoxometalates, leveraging their strong material compatibility [67,166,167,168,169]. This type of composite not only provides rich interfacial effects through core–shell structures but also, due to the coupling of different materials with PBA, offers multi-dimensional support for the OER in aspects such as optics and strain response [169,170,171]. Taking the research by Du et al. as an example, they successfully prepared PBA/sulfide nanoboxes as OER electrocatalysts, demonstrating significantly enhanced catalytic performance [169]. This improvement primarily stems from the strong synergistic effect between PBA and sulfides, the enlarged electrochemical active surface area, and the abundant electron transport and mass transfer pathways within the structure. It is particularly noteworthy that under light irradiation, the optimized CoFe PBA/CoS₂ nanoboxes achieved a multiplicative enhancement in OER performance. Mechanistic studies reveal that this enhancement is attributed to the efficient migration of photogenerated electrons from CoFe PBA to CoS₂. This process not only enriches the PBA surface with a large number of photogenerated holes that promote water oxidation but also significantly reduces the reaction activation energy of the CoS₂ component, thereby collectively driving a substantial acceleration in the kinetics of the water oxidation reaction. Furthermore, synergistic effects between heterogeneous materials and interfacial strain also significantly influence catalytic activity [170,171]. In summary, the synthesis of these interfacial-structured PBA nanocomposite catalysts and their optimization of local electronic structures provide new insights for developing highly efficient catalysts.

Figure 9. TEM and HR-TEM images of (a,b) GQD-NiFe PBA. (c) LSV curves of the samples. Reprinted with permission from ref. [164]. Copyright 2023, Elsevier B.V. (d) Schematic of the synthesis of Co³⁺-rich CoFe PBA coated with controllable MoS₂ shell heterostructures. (e) LSV curves of the nickel foam substrate, RuO₂, CoFe-PBA and Co^IIIFe-PBA/MoS₂. Reprinted with permission from Ref. [165]. Copyright 2023, Elsevier B.V.

Figure 9. TEM and HR-TEM images of (a,b) GQD-NiFe PBA. (c) LSV curves of the samples. Reprinted with permission from ref. [164]. Copyright 2023, Elsevier B.V. (d) Schematic of the synthesis of Co³⁺-rich CoFe PBA coated with controllable MoS₂ shell heterostructures. (e) LSV curves of the nickel foam substrate, RuO₂, CoFe-PBA and Co^IIIFe-PBA/MoS₂. Reprinted with permission from Ref. [165]. Copyright 2023, Elsevier B.V.

5. Prospects

PBAs, as classical coordination compounds with a history of several hundred years, have been widely recognized and studied as nanomaterials for decades. In the face of current challenges such as energy shortages and environmental degradation, PBAs continue to demonstrate significant application value, which stems from their multiple advantages: firstly, low raw material costs and simple synthesis methods; secondly, well-defined structures and controllable morphologies enabling the precise construction and regulation of diverse nanostructures; thirdly, a wide variety of transition metals capable of forming PBAs, allowing for multi-metal combinations and precise tuning of their ratios; fourthly, the distinctive role of the CN⁻ ligand, which can not only readily escape during thermal treatment to create vacancies but also participate in forming nitrogen-doped carbon materials, thereby optimizing the properties of active sites; fifthly, ease of combination with other materials to form nanocomposites. Leveraging these unique advantages, PBAs and their derivatives play an important role in the anodic OER of electrochemical water splitting, making the development of higher-performance PBA-based nanocatalytic materials a key goal for many researchers.

In recent years, researchers have developed a variety of methods for the preparation and modulation of PBAs and their derivatives, aiming to tailor them into ideal OER catalytic materials. This review systematically explores pathways for enhancing the OER performance of PBA-based catalysts along two main lines: regulating the number and utilization efficiency of active sites, and enhancing the intrinsic activity of active sites. The discussion covers key strategies such as increasing the exposure of active sites, optimizing site accessibility, improving electrical conductivity, adjusting elemental composition, and constructing vacancy defects. To provide a clear comparison of the OER performance across various PBA-based catalysts, we have compiled key parameters such as overpotential, Tafel slope, and stability in

Table 2. OER performance of reported electrocatalysts derived from PBAs.

Sample	Morphological/Structural Characteristics	Electrolyte	Overpotential ηmA cm−2 [mV]	Tafel [mV dec−1]	Stability	Ref.
Co3O4/Co-Fe oxide DSNBs	Hierarchical Nanoporous Structure	1 M KOH	η10 = 297	61	10	[74]
Fe-CoP prisms	Hierarchical Nanoporous Structure	1 M KOH	η10 = 235	32.9	45	[75]
Ni2[Fe(CN)6]	Defects	1 M KOH	η20 = 288	86	40	[53]
NiCoOx-400	Porous Structure	0.1 M KOH	η10 = 280	74	0.83	[84]
NiFe-NCs	Porous Structure	1 M KOH	η10 = 271	48	18	[59]
Ultrathin CoFe2O4 Nanosheets	2D and Porous Structure	1 M KOH	η10 = 275	42.1	10	[85]
RuO2/Co3O4	Interface and Porous Structure	1 M KOH	η10 = 302	74.37	1000C	[86]
Ni-Fe-Pt NCs	Interface and Porous Structure	1 M KOH	η10 = 333	64	12 1000C	[60]
CoFe-200/GCE	Porous Structure	1 M KOH	η10 = 316	49.6	10,000C	[61]
Ni-P	2D and Porous Structure	1 M KOH	η10 = 300	64	--	[88]
NiCoFeP	Hierarchical Nanoporous Structure	1 M KOH	η10 = 273	35	--	[89]
FeSe-NiSe/NF	Self-supported Electrode	1 M KOH	η10 = 234	22	120	[90]
FeCoSx-PBA	Hierarchical Nanoporous Structure and amorphous sites	1 M KOH	η10 = 266	33	--	[91]
Ni–Co mixed oxide cages	Hierarchical Nanoporous Structure	1 M KOH	η10 = 380	50	10	[93]
Co3O4 microframes	Hierarchical Nanoporous Structure	1 M KOH	η10 = 370	53	1000C	[94]
Ni-Fe-Se cages	Hierarchical Nanoporous Structure	1 M KOH	η10 = 240	22	10 500C	[95]
Ni-Fe-Se cages	Hierarchical Nanoporous Structure	1 M KOH	η10 = 249	35	144	[96]
NiFe hollow cages	Hierarchical Nanoporous Structure	1 M KOH	η10 = 313	20	5000C	[97]
Etched-PBA-MoS2	Hierarchical Nanoporous Structure	1 M KOH	η10 = 260	55	1000C	[98]
Co0.6Fe0.4P	Hierarchical Nanoporous Structure	1 M KOH	η10 = 298	48	120	[99]
A-CoS4.6O0.6 PNCs	Hierarchical Nanoporous Structure and Amorphous sites	1 M KOH 0.1 M PBS	η10 = 290 ηonset = 270	67 164	--	[100]
Mesoporous Nanostructured CoFe−Se−P	Hierarchical Nanoporous Structure	0.1 M KOH	η10 = 210	98	40 1000C	[101]
NiO/NiFe2O4 multicomposite hollow NCs	Hierarchical Nanoporous Structure	1 M KOH	η10 = 303	58.5	12	[103]
Co-Fe mixed oxide NAFSs	Hierarchical Nanoporous Structure	1 M KOH	η10 = 340	57	8	[105]
Fe-CoxP hollow nanosphere	Hierarchical Nanoporous Structure	1 M KOH	η10 = 345	49	12	[106]
t-CoIICoIII	Hierarchical Structure	1 M KOH	η10 = 220	79	50	[107]
CoFeP NFs/NPCNT.	High Electrical Conductivity	1 M KOH	η10 = 278	39.5	40 3000C	[108]
O-CNT/NiFe 1:18	High Electrical Conductivity and Defect	1 M KOH	η10 = 279	42.8	--	[109]
O-PBA/N-CNT 1:2	High Electrical Conductivity and Defect	1 M KOH	η10 = 280	48	24 1000C	[110]
rGO/CoNiPBd-OOH	High Electrical Conductivity	1 M KOH	η10 = 346	33	15	[111]
CoFe/CoFeOx@3DNC	Porous Structure and High Electrical Conductivity	1 M KOH	η10 = 335	50.5	20 2000C	[112]
PB-Co/Co−N-PHCS	Interface and High Electrical Conductivity	0.1 M KOH	η10 = 370	92	28	[113]
FeCo-NCNFs-800	High Electrical Conductivity	0.1 M KOH	η10 = 456	105.48	1000C	[115]
CoP@CNF	High Electrical Conductivity	1 M KOH	η10 = 300	73.8	24 2000C	[116]
NiCo@NC-900	Porous Structure and High Electrical Conductivity	1 M PBS	η10 = 396	89.2	25 3000C	[62]
Co3ZnC/Co nanojunctions@NC	High Electrical Conductivity and Interface	1 M KOH	η10 = 366	81	5000C	[63]
M-NiFe-700@C	High Electrical Conductivity	1 M KOH	η10 = 281	53	2000C	[120]
NiFeP/CC	High Electrical Conductivity and Self-supported Electrode	1 M KOH	η200 = 260	39.4	300	[121]
CoFe2O4 Thin Film	High Electrical Conductivity and Self-supported Electrode	1 M KOH	η10 = 266	53	24	[122]
nPBA@Co(OH)2/NF	Hierarchical Structure and Self-supported Electrode	1 M KOH	η10 = 256	46	2000	[57]
CuFe Oxide/CF	Hierarchical Structure and Self-supported Electrode	1 M KOH	η10 = 294	68	600	[123]
Fe–NiO/CC	High Electrical Conductivity and Hierarchical porous Structure	1 M KOH	η10 = 218	47	50	[124]
c-Ti-Fe-S boxes	Metal element doping and Porous Structure	1 M KOH	η10 = 350	55	12	[131]
Co0.62Fe0.38P	Metal element doping	1 M KOH	η10 = 230	51	15	[134]
FeCo0.41Mn0.42	Metal element doping	1 M KOH	η10 = 261	47.7	72	[135]
NCF MOF	Metal element doping and Hierarchical nanoporous Structure	0.1 M KOH	η10 = 190	49	5.5	[143]
NiCoFe-Se/CFP	Metal element doping and High Electrical Conductivity	1 M KOH	η10 = 221	38.58	40 1000C	[144]
Fe-Mn-Co/PBA	Metal element doping	1 M NaOH	η10 = 260	48	20	[145]
PBA-Se 350	Metal element doping	1 M KOH	η10 = 184	43.4	30	[64]
Ni-doped CoFe2O4 hollow nanocubes	Metal element doping and Hierarchical porous Structure	1 M KOH	η10 = 330	72.6	12	[146]
NiFeLaMoCo-PBA	High-Entropy Materials	1 M KOH	η10 = 244	32	48	[147]
HE-PBA-e	High-Entropy Materials and Defect	1 M KOH	η10 = 332	86.3	100	[148]
Co-PBA-plasma-2h	Defect	1 M KOH	η10 = 274	53	16	[14]
Co0.4Fe0.28P	Metal element doping and Porous Structure	1 M KOH	η10 = 270	25.6	3	[58]
VCN-mediated Ni–Fe PBA	CN vacancies	1 M KOH	η10 = 283	54	25	[157]
Oxygen Plasma-CoFe-PBA	Lattice oxygen activation	1 M KOH	η10 = 218 η1000 = 276	26.7 22.9	400	[56]
NiFe PBA-H2 4min	CN vacancies	1 M KOH	η10 = 400	104	100	[54]
PBA-plasma 3h-Air	CN vacancies	1 M KOH	η10 = 251	62.1	100	[55]
N90-NiCoPBA	Oxygen vacancies	1 M KOH	η10 = 289	70	24	[160]
Double-shelled NiFe PBAs with VCN	CN vacancies	1 M KOH	η20 = 267	79	80	[158]
N2-NiCoFe-PBA	CN vacancies	1 M KOH 1 M KOH/0.5M NaCl 1 M KOH/seawater	η100 = 286 η100 = 293 η100 = 323	41.5 41.9 54.8	120	[159]
Ni Fe PBAs with VCN	CN vacancies	1 M KOH	η50 = 270	53	80	[65]
v-NiFe PBA@rGO	[Fe(CN)6]3− vacancies	1 M KOH	η10 = 251	36.2	200	[66]
Ar-U-CoFe PBA	Amorphous Structure	1 M KOH	η10 = 305	36.1	20	[161]
NiFe PBA-N2 300 °C	Amorphous Structure	1 M KOH	η10 = 330	73	10	[162]
Boron-modified PBA/NFs 2h	Amorphous Structure	1 M KOH	η100 = 311	88	50	[163]
O-GQD-NiFe PBA	Interface and Defects	1 M KOH	η10 = 259	52.5	100	[164]
CoIIIFe-PBA/MoS2−x	Interface	1 M KOH	η100 = 306	36.2	30	[165]
PBA@POM	Interface and Hierarchical Porous Structure	1 M KOH	η10 = 440	23.45	16	[166]
Co-PB/Pt	Interface and Hierarchical Structure	1 M KOH	η10 = 300	68	500C	[167]
NC@CeO2-CoFe	Interface	1 M KOH	η10 = 255	47	1000 4000C	[168]
CoFe PBA/CoS2-12 CNBs	Interface	1 M KOH	η10 = 301	59.2	45	[169]
CoFe PBA@CoP/NF	Interface and Hierarchical Structure	1 M KOH	η10 = 171	75.7	15	[67]
AuNSt@PBA	Strain	0.1 M PBS 1 M NaNO3	η10 = 800	~190	--	[170]
Au@CoFe	High Electrical Conductivity	1 M KOH	η10 = 300	63	24	[171]

Stability: The I-T and E-T curves use hours, while the CV curves use cycle number (denoted with a suffix ‘C’).

at the end of this review. Although significant progress has been made in using PBAs and their derivatives as catalysts or pre-catalysts in OER research, and catalytic performance has been greatly improved through the integrated application of various modulation strategies, the practical application of these materials still faces numerous challenges due to the inherent complexity of the OER process, which involves a four-electron transfer reaction.

First, PBAs used in the OER, particularly under alkaline conditions, predominantly serve as precursors or pre-catalysts. This approach fails to leverage their advantages of well-defined crystal structures and atomically dispersed metal sites within the framework, while also leading to the formation of active sites that are difficult to predict and characterize. There is a pressing need to develop more advanced regulation strategies to create PBA materials that maintain structural stability during OER. This would enable a clear understanding of the mechanism of individual active sites and the synergistic effects between different sites, which holds significant importance not only for elucidating OER mechanisms but also for the broader fields of electrocatalysis and energy materials.

Second, current mechanistic investigations predominantly focus on PBA derivatives or the active species generated post-reaction, which may lead to a biased understanding of the actual catalytic process. There remains a lack of in-depth, quantitative comprehension regarding the dynamic evolution of PBAs as precursors during the critical reconstruction phase where they transform into the true active sites. Advanced in situ techniques developed in recent years will provide robust support for elucidating this mechanism, facilitating the precise identification of active centers and clarification of reaction pathways. Gaining insights into the transformation process will enable more targeted and rational design of catalysts.

Third, theoretical research on different types of PBA models remains relatively limited. Since PBAs and their derivatives often act as pre-catalysts—especially high-performance catalysts containing V_CN that have been experimentally verified—key questions are still unclear. These include how the elemental composition influences the pathway of V_CN transformation into the true active site during OER, and the specific impact of V_CN concentration on the reaction kinetics. Future work needs to more deeply integrate theoretical calculations with experimental analysis, establishing more accurate models and optimizing experimental design to provide a theoretical basis for guiding the synthesis of high-performance PBA-based OER catalysts.

Fourth, although research on PBA-based OER is extensive, some repetitive work is inevitable, leading to resource wastage, while certain detailed issues remain unexplored systematically. Introducing machine learning methods and promoting the establishment of large-scale, shared data platforms through multi-team collaboration can help avoid redundant studies, precisely address research gaps, and potentially enable the prediction and targeted design of material properties.

Fifth, the challenges of scalable fabrication and long-term operational stability represent the primary bottlenecks hindering the industrial application of PBA-based electrodes. Currently, the synthesis of most high-performance PBA-based catalysts remains confined to the laboratory scale, where intricate morphology control and multi-step reaction processes pose significant hurdles for large-scale, low-cost production. Furthermore, the long-term stability of PBAs and their derivatives under the harsh conditions of strong alkalinity and high OER potentials remains inadequate for practical implementation. Although strategies like defect engineering or carbon compositing can mitigate issues like active component dissolution, detachment, or phase transformation to some extent, the performance still falls short of the industrial requirement for multi-year stability. Future research should focus on developing more intrinsically stable PBA material systems. Coupled with advanced in situ characterization techniques, a deeper investigation into their failure mechanisms under realistic operating conditions is essential. This will provide critical guidance for designing practical electrodes that combine high activity with industrial-grade durability.

In conclusion, although research on PBAs and their derivatives still faces numerous challenges and unknowns, these difficulties also present new opportunities and impetus for the development of the OER field. It is undeniable that research based on PBAs has significantly advanced OER catalytic systems. We sincerely appreciate the contributions of researchers in this field and look forward to the future development of practical, industrially viable PBA-based catalytic materials.