Next-generation perovskite-metal–organic framework (MOF) hybrids in photoelectrochemical water splitting: a path to green hydrogen solutions

Abstract

Green hydrogen production through sustainable and clean methods, such as photoelectrochemical (PEC) water splitting, has garnered significant attention as a potential solution to the twin issues of fossil fuel depletion and excessive greenhouse gas emissions in energy production. Enabling efficient PEC reactions requires the discovery and utilisation of stable and efficient photoelectrode materials. Perovskite materials, renowned for their exceptional optoelectronic properties, have garnered attention due to their remarkable optoelectronic properties, including tuneable bandgaps, high absorption coefficients, and superior charge transport dynamics. However, their practical use is hindered by inherent stability issues and efficiency bottlenecks. To address these limitations, the integration of perovskites with secondary materials, particularly metal–organic frameworks (MOFs), has emerged as a promising strategy. MOFs, known for their high surface area, tuneable porosity, and catalytic activity, offer a unique opportunity to complement and enhance the performance of perovskites in PEC applications. This review begins by exploring the fundamental properties of perovskites and MOFs, highlighting recent advancements and key challenges in their individual and combined use as photoelectrodes for PEC water splitting. The synergistic interplay within MOF-perovskite hybrids is discussed, focusing on how this interaction improves charge separation while addressing the persistent issues of photocorrosion and operational stability. Finally, this review addresses critical issues that still limit the efficiency of MOF-perovskite photoelectrodes and offers some perspectives on potential solutions. By bridging the unique strengths of perovskites and MOFs, this work underscores the transformative potential of hybrid materials in advancing PEC water splitting toward sustainable and scalable green hydrogen production.

---

Quan Yee Tey

Quan Yee Tey received her bachelor's degree from Monash University Malaysia in 2021. She is currently a PhD candidate at Monash University Malaysia under the supervision of Prof. Chong Meng Nan. Her research interest lies in the field of sustainable energy, specially focusing on the development of high-performance photoelectrodes for solar-driven photoelectrochemical (PEC) water splitting. Her current research endeavours involve the rational design and engineering of semiconductor-based photoelectrodes, with a particular emphasis on PEC tandem cell configurations, aiming to enhance the efficiency and stability of PEC water splitting systems.

Joshua Zheyan Soo

Dr Joshua Zheyan Soo is a postdoctoral research fellow in the Centre for Net-Zero Technology, Monash University Malaysia. He received his PhD from the Australian National University, researching on the scalable synthesis of earth-abundant catalysts for (photo)electrochemical water-splitting. Prior to that, he obtained his MPhil. and BEng (Hons) from the University of Malaya and Multimedia University, respectively. Joshua's current research interests revolve around the area of electrocatalyst synthesis and (photo)electrode device fabrication for green hydrogen and ammonia production, with a focus on using sustainable materials and design processes.

Wen Cai Ng

Dr Ng Wen Cai is an active researcher specializing in renewable energy, solar hydrogen, and nanotechnology. Her work focuses on engineering and developing efficient semiconductor-based photoelectrodes for solar-to-fuel (hydrogen) applications, addressing critical challenges in sustainable energy conversion. She has extensive expertise in materials design, particularly in utilizing advanced nanomaterials for clean energy technologies. She has a strong interest in exploring innovative materials to enhance the efficiency and functionality of photoelectrochemical water splitting systems. With a commitment to advancing green energy solutions, she actively contributes to the development of environmentally sustainable technologies and the global transition to clean energy.

Meng Nan Chong

Professor Ir Dr Meng Nan Chong earned his BEng (Hons.) in Chemical Engineering and PhD from the University of Adelaide. He is a Professor of Chemical Engineering and Director of the Centre for Net-Zero Technology at Monash University Malaysia. His research excellence has earned him prestigious awards, including the UK's Newton Advanced Fellowship, the Green Talents Award, and Top Research Scientist Malaysia. He has been listed among the World's Top 2% Scientists by Stanford University (2020–2023) in career-long and single-year categories, highlighting his global impact in chemical engineering and environmental science research.

---

1 Introduction

The energy industry is at a pivotal moment, as the dominance of fossil fuels in energy production has become unsustainable due to emerging global challenges. Firstly, the pertinent issue of man-made climate change requires a drastic reduction in global greenhouse gas emissions, of which the fossil-fuel-reliant energy sector is a major contributor. At the same time, the possibility of fossil fuel depletion and its current concentration in selected geographical regions raises geopolitical concerns about its impact on global energy security. Therefore, exploring renewable and universally accessible alternatives to fossil fuels is essential to diversify the energy mix and mitigate these challenges, ensuring sustainable long-term energy production.

Amid these considerations, solar energy stands out as an infinite energy resource, with a staggering 173 000 terawatts constantly reaching Earth, a magnitude capable of satisfying global energy needs over 10 000 times.¹ Nevertheless, the intermittent nature of solar energy, dictated by factors like time of day, weather, and geographic location, is a challenge that needs to be overcome to ensure a continuous on-demand supply of energy. A key solution to mitigate this inherent supply variability is converting solar energy into clean, storable fuels. In this regard, hydrogen (H₂), which yields only water as a by-product of combustion, stands out as a prime choice.² Its remarkable energy carrier attributes boast an impressive gravimetric energy density of 141.8 MJ kg⁻¹ (higher heating value, HHV) and 120 MJ kg⁻¹ (lower heating value, LHV) at 298 K, surpassing that of natural gas (54 MJ kg⁻¹) and gasoline (47 MJ kg⁻¹).^3,4 As a result, a litre of H₂ fuel has an energy capacity comparable to that of 0.25 litre of gasoline under ambient conditions.⁵ This high efficiency makes H₂ a compelling fuel option. Beyond its use as a fuel, H₂ is widely employed in industries like petroleum refining (hydrotreating), fertiliser production, food processing (fat and oil conversion), and drug manufacturing.⁶

Currently, steam methane reforming remains the dominant method for industrial H₂ production, which is a high-temperature process that relies on fossil fuels and steam (CH₄ + 2H₂O → 4H₂ + CO₂).⁷ However, this process undermines the low-carbon advantages of using H₂ as a fuel, as it generates substantial amounts of carbon dioxide (CO₂), with estimates ranging from 7.5 to 12 tons of CO₂ produced per ton of H₂.^8,9 Harnessing renewable sources such as solar energy and converting it directly into H₂, particularly through photoelectrochemical (PEC) water splitting, offers a far more sustainable pathway. This approach also addresses the intermittency predicament associated with direct solar energy use, positioning H₂ as a key player in sustainable energy solutions. While photovoltaic-electrocatalysis (PV-EC) is another approach for solar-driven water splitting, the integrated design of PEC systems offers several advantages. PEC systems, which combine light harvesting and water electrolysis in a single component, have vast potential to reduce implementation costs by using fewer materials.¹⁰ Although PV-EC systems have achieved high technological readiness levels (TRL) and demonstrated higher solar-to-H₂ (STH) efficiencies, their high solar cell costs remain a significant limitation.¹¹ This makes PEC research, especially with cost-effective materials like metal oxides, a promising pathway for developing more efficient and affordable solar-driven water splitting systems.¹²

Over the years, a wealth of comprehensive investigations has been conducted to explore a wide range of semiconductor materials (e.g.: TiO₂, Fe₂O₃, Cu₂O, BiVO₄, MoS₂, and g-C₃N₄) for use as PEC photoelectrodes.^13–17 Various iterations of these materials have been explored in these studies, focusing on tuning of properties such as crystallinity, surface behaviour, and heterojunction formation to enhance their intrinsic PEC performance. For example, the porous structure of the ZnO photoanode with oxygen vacancies was constructed which led to an increase in surface area and light harvesting, shortening the charge transport distance and promoting charge transfer and water oxidation kinetics at the electrode/electrolyte interface.¹⁸ However, the reported PEC efficiencies of these materials still fall short of meeting the key benchmarks for large-scale applications, including the 25% STH efficiency target, a 10-year operational lifetime, and H₂ production costs of USD 1–10 per kg. This underscores the ongoing challenge of achieving the necessary efficiency targets to make PEC water splitting a truly transformative technology in the realm of green H₂ production.

Recently, the scientific community has turned its attention to perovskite materials as a compelling alternative to traditional semiconductors, primarily due to their distinctive properties. Perovskites offer exceptional structural flexibility which provides a promising platform for the rational design and engineering of high-efficiency photoelectrodes. By precisely modulating the A and B site elements and composition, the crystal and electronic properties can be meticulously tailored to optimise photoelectrode performance. Additionally, certain perovskites exhibit ferroelectric or piezoelectric properties, which enhance charge carrier separation, reduce recombination and boost the catalytic activity of perovskite-based photoelectrodes for water splitting. Perovskites have already demonstrated remarkable light absorption efficiency in solar cells, reaching up to 26.1%, highlighting their potential as promising photoelectrode materials for PEC water splitting.¹⁹ Despite their remarkable optoelectronic characteristics, perovskites generally suffer from poor stability in water or aqueous solutions, leading to their rapid degradation upon exposure. This instability poses a significant bottleneck to their practical applications, particularly in PEC water splitting, where the photoelectrode must be immersed in an aqueous electrolyte solution. Consequently, researchers are actively exploring various strategies to improve perovskite stability and address this challenge. Among these approaches, encapsulating perovskites with materials such as polymers, metals, and TiO₂ has emerged as the most promising and direct method to reinforce perovskite stability.^20–23

Meanwhile, the versatile potential of metal–organic frameworks (MOFs) has garnered significant attention from researchers, particularly in applications such as gas adsorption, storage, molecular separation, and catalysis. The appeal of MOFs lies in their remarkable characteristics, including a highly porous structure, large specific surface area, and exceptional stability.²⁴ These frameworks, comprising metal nodes intertwined with organic ligands, inherently exhibit tuneable photovoltaic properties such as bandgap, band edge positions, light absorption ability and porosity, making them easily tunable for specific applications. These advantages demonstrate MOFs to be a promising candidate for integration with perovskite-based photoelectrodes to form next-generation perovskite-MOF hybrids for PEC water splitting. In this review, we define next-generation perovskite-MOF hybrids as those that demonstrate significant advancements beyond traditional photoelectrode materials such as Cu₂O, BiVO₄, TiO₂, and ZnO. By synergistically combining these two materials, perovskites can be protected from degradation while simultaneously elevating the efficiency of the photoelectrode. The inherent porosity and stability of MOFs create an environment that safeguards perovskite materials, thus extending their operational lifespan. Furthermore, the synergistic interaction between MOFs and perovskites holds promise for enhancing photoelectrode performance. The stability of MOFs can help mitigate the inherent instability of perovskites, while preserving their exceptional optoelectronic properties. This innovative amalgamation not only provides a solution to the challenges of perovskite stability but also paves the way for optimising the efficiency of energy conversion processes. Furthermore, next-generation perovskite-MOF hybrids often incorporate innovative design strategies, such as multifunctional components (e.g., integrated co-catalysts) and the utilisation of novel MOF structures and perovskite compositions. These materials aim to address the critical challenges faced by the current PEC water splitting technologies and pave the way for the development of highly efficient and sustainable solar-to-H₂ conversion systems.

This review aims to comprehensively examine recent advancements in the strategic integration of MOFs with perovskite-based photoelectrodes for PEC water splitting applications. In this review, we systematically delve into the progress in this emerging field, shedding light on the tactical integration of MOFs and perovskites to improve the efficiency and stability of PEC water splitting processes. We first present the fundamental principles of PEC water splitting, specifically highlighting the role of photoelectrodes in this process. This is followed by a concise overview of the structure and properties of perovskite materials, setting the stage for an in-depth discussion of the recent advancements that have positioned perovskites as promising contenders for driving PEC water splitting reactions. Following this, we also address the stability challenges encountered during aqueous operations, offering insights into the ongoing efforts to overcome these obstacles. Building on this foundation, we then review recent studies of MOFs in the context of PEC water splitting, highlighting their potential as critical components in improving the efficiency of water splitting processes. The discussion culminates in a detailed exploration of the latest developments in MOF-perovskite hybrid photoelectrodes for PEC water splitting, as well as other relevant applications. Here, we emphasise the innovative and tactical design strategies aimed at enhancing stability and overall PEC and photoconversion performance. This comprehensive review provides valuable insights, charting pathways for further refinement and development of MOF-perovskite hybrids. We conclude this review by offering an outlook on the future of this material class, outlining strategies to address current challenges in PEC water splitting. By reviewing and analysing potential advancements, we envision the future trajectory of this dynamic field and highlight how these synergistic combinations could drive transformative progress in sustainable energy technologies.

2 Fundamental principles of PEC water splitting

The mechanism behind water splitting is relatively facile, and involves a redox reaction consisting of two distinct half-reactions: the reductive H₂ evolution reaction (HER) transpiring at the cathode, and the oxidative oxygen evolution reaction (OER) occurring at the anode. The overall water splitting reaction, along with the individual half-reactions, is summarised in eqn (1) as follows:²⁵

Overall 2H₂O → 2H₂ + O₂ ΔE = 1.23 V at pH 0 (1)

The specific steps of both the HER and OER reactions vary depending on the type of electrolyte used in the water splitting cell. This is due to the different majority species (i.e. H⁺, OH⁻) present in the electrolyte, which participate in the water splitting reactions. Eqn (2) and (3) present the general equations for the OER and HER in acidic media, while eqn (4) and (5) present the corresponding half-reactions in basic media.²⁶

OER (acidic medium):

2H₂O → 4H⁺ + O₂ + 4e⁻ ΔE = +1.23 V at pH 0 (2)

HER (acidic medium):

4H⁺ + 4e⁻ → 2H₂ ΔE = 0 V at pH 0 (3)

OER (basic medium):

4OH⁻ → O₂ + 2H₂O + 4e⁻ ΔE = +0.401 V at pH 14 (4)

HER (basic medium):

4H₂O + 4e⁻ → 2H₂ + 4OH⁻ ΔE = −0.828 V at pH 14 (5)

Regardless of the electrolyte properties, water splitting is inherently an energy-consuming process. Under standard conditions (25 °C, 1 atm pressure, pH 0, 1 M concentration), it requires 285.8 kJ mol⁻¹ of energy, which is divided into Gibbs free energy (237.2 kJ mol⁻¹) and heat energy (48.6 kJ mol⁻¹). The Gibbs free energy corresponds to 1.23 V per electron (eV), which represents the thermodynamic potential barrier needed to initiate the water splitting reaction. However, in practice, additional energy is typically required in reality to overcome kinetic losses in the HER and OER reactions at both electrodes of the PEC cell.

In a PEC process, the water splitting reaction is initiated by light illumination, either with or without the need for an external voltage bias. Hence, photoactivity is required at one or both electrodes in the electrolytic cell, which can be enabled by using photoactive semiconductors coupled with a cocatalyst. Additionally, to utilise solar energy to drive the water splitting reaction with a semiconductor photoelectrode, the bandgap of the semiconductor needs to be at least 1.23 eV. Moreover, the band positions of the semiconductor material need to align strategically with the water splitting half-reactions (HER and OER). Ideally, the Fermi level (E_F – a measure of the semiconductor's electron filling) of the photoelectrode should be higher in energy than the H₂ evolution potential for an n-type photoanode, where electrons are the majority carriers. On the other hand, the E_F level of a p-type photocathode should have a lower energy than the O₂ evolution potential, where holes are the majority carriers. Fig. 1 illustrates this process using a single photoanode and a counter cathode as an example. The PEC water splitting reaction inherently involves three fundamental stages. First, the semiconductor photoanode absorbs light with energy greater than its bandgap, leading to the generation of photoexcited charge carriers within the bulk of the photoanode. This photoexcitation creates electron–hole pairs, driving charge carrier separation: electrons (e⁻) are excited from the valence band (VB) to the conduction band (CB), while holes (h⁺) are left behind in the VB. It is worth noting that electron–hole recombination during charge separation can significantly impede efficiency. The final stage involves the participation of photogenerated charge carriers in the HER/OER reactions at the electrode surfaces. Specifically, in this photoanode-dark cathode example, photogenerated electrons travel through the circuit to the counter electrode, where they drive the cathodic HER, reducing H⁺ ions to produce H₂. Simultaneously, holes generated at the photoanode migrate to the interface between the photoanode and electrolyte, catalysing the OER sequence at the interface and producing O₂. In the process, the number of active sites and the extent of surface area play a pivotal role in facilitating efficient redox reactions. Ultimately, improving PEC performance relies on the continuous refinement and optimisation of these three fundamental processes. Enhancing charge separation efficiency and redox reaction effectiveness is crucial for advancing the overall efficiency of PEC water splitting systems.

Fig. 1 Schematic of the PEC cell mechanism with an n-type semiconductor as a working photoanode.

2.1 Semiconductors as photoelectrode materials

Semiconductor photoelectrodes are crucial in PEC water splitting systems, as they generate charge carriers upon radiative illumination and ensure their separation before participating in surface HER or OER reactions. Therefore, the selection of suitable semiconductors as bulk materials for the photoelectrode is paramount for achieving high STH efficiencies in the overall PEC water splitting system.

The efficiency of photoelectrodes in harnessing solar energy directly impacts the overall performance of the system. Since visible light accounts for approximately 45% of the solar spectrum, photoelectrode materials with strong visible light absorption are preferred.²⁷ This characteristic is closely linked to the bandgap of semiconductor photoelectrodes, which determines both the range of absorbable light and the maximum achievable STH efficiency. While a narrower bandgap is generally advantageous, a practical balance must be struck between maximising light absorption and providing sufficient driving force for water splitting. While a theoretical minimum bandgap of 1.23 eV is sufficient to drive water splitting, practical considerations such as thermodynamic overpotentials due to kinetic HER/OER losses dictate a more suitable bandgap range of 2.00–2.25 eV.^28,29 Additionally, the band edge positions must align with the redox potentials (HER and OER) of water for efficient charge transfer. For efficient HER at the photocathode, the CB edge must be more negative than the water reduction potential (H₂/H₂O), while effective OER at the photoanode requires a VB edge more positive than the water oxidation potential (O₂/H₂O).^28,30,31

After photoexcitation, electrons and holes must remain separated enabling them to be transported to the semiconductor surface and participate in their respective redox reactions. The efficiency of charge separation is influenced by the inherent properties of the semiconductor, such as its dielectric constant, carrier mobility and concentration, as well as its diffusion length relative to its distance to the surface.^28,32 These factors must be optimised to minimise recombination losses, which can hinder the carrier activity at the semiconductor surface. The recombination rate is also influenced by the type of semiconductor bandgap (i.e., direct or indirect) and the concentration of defects, which can serve as recombination sites. Efficient charge transfer across the semiconductor–electrolyte interface is crucial to prevent charge carrier depletion within the space charge region.³³

To foster large-scale implementation of PEC water splitting, additional practical considerations must be factored into the selection of photoelectrode materials. Material stability is a key concern, particularly the resilience of semiconductors against corrosion when immersed in non-neutral electrolytes and exposed to radiative illumination. In many cases, (photo)corrosion reactions of certain semiconductors are more thermodynamically favourable than the HER/OER reactions themselves. Additionally, choosing non-toxic and cost-effective materials is also beneficial to support large-scale deployment.

3 Perovskite materials

Perovskites are defined as materials that inherently share the same crystalline unit cell structure as calcium titanate (CaTiO₃), a discovery attributed to Lev Perovski in 1839.³⁴ This commonality anchors perovskite materials following a structural framework represented by the chemical formula ABX₃. In this representation, A and B represent cations of distinct sizes, where A (La³⁺, Sr²⁺, Ba²⁺) typically consists of larger ions compared to B (Bi²⁺, Sn²⁺, Ti⁴⁺). Accompanying these cations are anions (O²⁻, I⁻, Br⁻, Cl⁻) that bond with the two cations.³⁵ As illustrated in Fig. 2a, this arrangement generates a BO₆ octahedral configuration, with the B cation at the centre of the octahedron and surrounded by X anions occupying corner positions.³⁷Fig. 2b visualises the formation of a three-dimensional extended structure, achieved by linking the corners of the octahedra. In this extended structure, the A cations occupy vacant sites between the octahedral units.³⁸

Fig. 2 (a) Cubic ABX₃ perovskite unit cell, and (b) three-dimensional structure of perovskites formed via corner-linked octahedra. Adapted with permission from ref. 36. Copyright (2022), Elsevier.

The Goldschmidt formulation provides a key criterion for determining the formation of a stable cubic perovskite structure. The stability of the ABX₃ structure is predicted based on the calculation of the tolerance factor (t), as defined in eqn (6):³⁹

Here, the ionic radii of the cations at the A and B sites, as well as the anion at the X site (r_A, r_B, and r_O, respectively), play a crucial role in determining the stability of the perovskite structure. A cubic perovskite structure forms and remains stable when the tolerance factor falls within the range of 0.9 to 1. Deviations from this range result in lattice distortions, triggering the emergence of alternative structures such as orthorhombic, monoclinic, and rhombohedral. Therefore, the radii of the cations exert a profound influence on the structural dynamics of perovskites.⁴⁰ The tunability of perovskites becomes a pivotal factor in optimising their photocatalytic properties. Variations in cation types and valence states contribute to differences in ionic radii, which, in turn, induce lattice distortions that significantly impact electronic band structures – a key determinant of the photocatalytic traits of the material.³⁶ These band structures are instrumental in facilitating charge transfer and redox reactions for efficient water splitting. Ultimately, the interplay of cation radii and the tolerance factor governs the structural stability of perovskites. The dynamic flexibility of perovskite properties, driven by cation variation, illustrates how precise structural adjustments in perovskites can be harnessed to achieve optimal photocatalytic properties for water splitting applications.

Perovskites, originally defined as CaTiO₃, now encompass a broad class of materials that share the ABX₃ crystal structure. This versatile structural framework offers significant compositional flexibility, leading to the development of diverse perovskite materials with unique physical and chemical properties. These include perovskite oxides, layered perovskites, and halide perovskites, each of which has distinct characteristics and applications.

Perovskite oxides are a subclass of perovskites in which O²⁻ serves as the anion, with alkaline earth metals and transition metals occupying the A and B sites, respectively. Since the seminal work of Kato et al. in 2002 on their application in photocatalysis, perovskite oxides have drawn extensive interest.⁴¹ Their structural and compositional tunability makes them ideal candidates for optimizing catalytic reactions. In these materials, the CB is typically formed by the d and p orbitals of the B-site cations, while the VB consists of O 2p orbitals. A-site ions, such as alkali metals, alkaline earth metals, and certain lanthanides, generally stabilise the crystal structure without contributing to the bandgap. Advances in defect engineering and ion substitution have played pivotal roles in improving the PEC performance of perovskite oxides, particularly in water-splitting applications.⁴²

Layered perovskites are a structural modification of perovskite oxides with the formula A₂BX₄, which introduces a more pliable structure while retaining excellent electronic and charge transfer properties. These properties make layered perovskites highly advantageous for photocatalytic H₂ production compared to their bulk counterparts. A landmark study in 1997 by Takata et al. demonstrated the use of a hydrated layered perovskite structure for water splitting, highlighting the ability of these compounds to embed reactants within their interlayer spaces, thereby enhancing photocatalytic efficiency.⁴³ Subsequent research has identified several subgroups of layered perovskites, including the Aurivillius phase ((Bi₂O₂)²⁺(B_n−1M_n)²⁻), Dion–Jacobson phase (A₂BX₆), and Ruddlesden–Popper (RP) phase (A_n+1B_nO_3n+1). Recent innovations, such as the fabrication of an S-scheme heterojunction, Fe₂SnO₄, coupled with g-C₃N₄ by Gupta et al., have further enhanced the catalytic performance of the material, achieving notable H₂ production rates with excellent stability over multiple cycles.⁴⁴

Halide perovskites have risen to prominence in the last decade due to their exceptional properties, including a tuneable direct bandgap, long carrier lifetime, high absorption coefficients, and defect tolerance. These characteristics have established halide perovskites as leading materials in solar cells, light-emitting diodes, and photodetectors. Their potential in photocatalytic and PEC applications is equally promising. Halide perovskites exhibit a relatively narrow bandgap compared to conventional photocatalysts, enabling efficient solar energy absorption at lower energies.^45–47 Their conduction band position further supports overall water splitting, making them a compelling option for sustainable energy applications (Fig. 3).

Fig. 3 Energy band position of conventional photocatalysts and lead halide perovskites. Adapted with permission from ref. 48. Copyright (2020), American Chemical Society.

The diversity and adaptability of perovskites—ranging from oxides and layered structures to halides—underscore their importance in advancing technologies across various fields, particularly in catalysis and renewable energy.

3.1 Synthesis strategy of perovskite materials

The synthesis methods of perovskites are fundamental to tailoring their structure and properties. These techniques enable precise control over crystallinity, morphology, and electronic characteristics, significantly influencing their functional performance, especially in photocatalysis and sustainable energy applications. Each method offers unique advantages and limitations, making the choice of synthesis route crucial.

3.1.1 Hydrothermal method. The hydrothermal method is a solution-based approach that relies on creating distinct temperature zones within a sealed autoclave to promote crystal growth. Precursor materials such as metal oxides, sulfides, or acetates are dissolved in solvents at room temperature to form a sol, which is then subjected to elevated temperatures and pressures. The process concludes with cooling, ultrasonication, and centrifugation to remove impurities, yielding high-purity crystalline products. For instance, Gómez-Cuaspud et al. synthesized a phase-pure LaFeO₃ perovskite using this method without requiring calcination.⁴⁸ Similarly, single-crystal BiFeO₃ microplates were developed hydrothermally using C₆H₁₀BiNO₈ as a reductant and surface modifier.⁴⁹ The hydrothermal method stands out for its simplicity in synthesising diverse nanomaterials, providing advantages such as controllable size, low cost, and convenient operation compared to other techniques.⁵⁰

3.1.2 Solid-state method. The solid-state method involves mixing stoichiometric amounts of solid precursors, such as oxides, carbonates, and nitrates, followed by prolonged heating at high temperatures. While cost-effective and straightforward, this approach often produces larger particle sizes and can lead to agglomeration, which may compromise purity. For example, Yang et al. improved the magnetic properties of a Sr₂FeReO₆ double perovskite by combining solid-state synthesis with molten-salt processing.⁵¹

3.1.3 Sol–gel method. The sol–gel method transforms a sol (a stable colloidal suspension) into a gel through hydrolysis and condensation reactions, using precursors like metal alkoxides or chlorides. The resulting gel undergoes drying and calcination, which enhance its structural and crystalline properties. This approach offers significant advantages over traditional solid-state reactions, including: (1) enhanced purification capabilities, (2) readily available starting materials, (3) simplified processing, (4) reduced reaction times, (5) lower reaction temperatures, and (6) lower sintering temperatures.⁵² Despite these benefits, challenges remain, such as substantial shrinkage, residual porosity, high production costs, and the potential for cracking during the drying process. Islam et al. employed this method to produce B-site disordered Y₂FeCrO₆ double perovskite electrodes for the first time.⁵³ Boumaza et al. synthesized a La₂NiO₄ photocatalyst via the sol–gel process, achieving a narrow bandgap and high H₂ production efficiency.⁵⁴

3.1.4 Freeze drying method. Freeze drying begins with preparing a liquid feed that is atomized and rapidly frozen using a cryogenic medium, such as liquid nitrogen. The frozen droplets are then dried under vacuum conditions, leaving behind intact particles with preserved structural integrity. This method effectively produces powdered materials with unique properties while preserving their structural integrity. The success of this technique hinges critically on a careful optimisation of several key factors, including solution composition, concentration, and pH, the spray-freezing method employed, and the precise conditions of calcination.⁵⁵ Feng et al. utilised this technique to fabricate CuI/BaSnO₃-based p–n junctions, achieving high transmittance and enhanced photovoltaic performance.⁵⁶

3.1.5 Pulsed laser deposition method. Pulsed laser deposition is a physical vapor deposition technique where high-energy laser pulses ablate a target material in a vacuum chamber. The resulting plasma plume deposits a thin film on a substrate, enabling controlled stoichiometry and high-quality layer formation. PLD offers exceptional control over film growth by allowing for the manipulation of numerous deposition parameters. This versatility enables precise tailoring of film properties, surpassing the capabilities of many other deposition methods. However, PLD generally exhibits lower deposition rates compared to techniques like sputtering or evaporation, which are better suited for high-volume industrial production.⁵⁷ Sharma et al. used PLD to grow double perovskite thin films of Sr₂FeMoO₆ and Sr₂FeWO₆, achieving nearly single-crystalline films with a dominant c-axis orientation.⁵⁸

3.2 Recent progress of perovskite materials for water splitting

Perovskite materials have garnered significant attention as highly promising candidates for light harvesting in PEC cells and photovoltaic applications. This interest is driven by their exceptional attributes, including high charge carrier mobility, minimal defect density, extended diffusion lengths and lifetimes, and exceptional chemical and structural versatility.^59–65 Impressively, solar cells based on lead halide perovskites have achieved power conversion efficiencies (PCE) of up to approximately 25%.⁶⁶ Beyond their photovoltaic potential, perovskite materials have demonstrated the ability to simultaneously function as charge carrier transporters and generators – by efficiently creating electron–hole pairs under incident light irradiation. This dual functionality further bolsters their viability for PEC water splitting applications.⁵⁹Fig. 3 highlights a key advantage of lead-based iodide perovskites: their narrower bandgaps (1.4–1.8 eV) compared to conventional semiconductors. This feature enhances their light absorption within the visible spectrum, resulting in a higher yield of electron–hole pairs generated from a given quantum of photons.^67,68 As discussed earlier, the tunability of perovskite constituents allows for tailored properties suited to diverse applications. For instance, modifying the anion component can induce a shift in valence band energy with minimal effect on conduction band levels, enabling precise bandgap tuning of the perovskite material.⁶⁹ Similarly, altering the ionic radii of cations introduces lattice distortions, as observed when methylammonium cations are replaced with slightly larger formamidinium cations, resulting in bandgap shifts.⁷⁰

Despite the popularity and promise of lead halide perovskites as high-performance semiconductors, their poor stability in humid environments remains one of the significant limitations for water splitting applications.^71–73 This instability is due to the hygroscopic nature of amine salts, leading to the formation of perovskite hydrates.⁷⁴ The decomposition of methylammonium lead iodide (CH₃NH₃PbI₃), a widely studied perovskite material, in the presence of moisture is represented by the following equations:⁷⁵

CH₃NH₃PbI₃ + H₂O → CH₃NH₃PbI₃·H₂O (7)

4(CH₃NH₃PbI₃) + 2H₂O → (CH₃NH₃)₄PbI₆·2H₂O + 3PbI₂ (8)

Eqn (7) depicts a reversible reaction in which CH₃NH₃PbI₃ can be reformed through a drying process after minimal exposure to moisture. However, with excessive water exposure, the perovskite undergoes structural degradation, resulting in an irreversible transformation to hydrates, as shown in eqn (8). This degradation alters its optical and electrical properties, which subsequently affects the overall efficiency of the PEC device. Moreover, Pb is both toxic and unstable at ambient temperature, prompting researchers to explore lead-free halide and oxide perovskites as alternatives, such as Sn-based, Bi-based, and Cu-based perovskites.^76–79 For example, CsSnX₃ (X = Cl, Br and I) were synthesized in 2016 by Chen et al. via a simple solvothermal process and demonstrated tuneable photoluminescence by substituting different halide ions of CsSnX₃. Chung et al. reported that CsSnI₃ exhibits strong photoluminescence properties and high electrical conductivity, likely due to defect centres in Sn rather than charge transfer between the VB and CB.⁸⁰ This finding highlights the potential of Sn-based perovskites as viable alternatives to lead-based counterparts. Other promising lead-free perovskites include Cs₂PtBr₆, prepared by Peng and co-workers, which has a bandgap of 1.32 eV, excellent conductivity, and achieved a maximum photocurrent density of 335 μA cm⁻² at −0.6 V vs. Ag/AgCl.⁸¹ This Pt-based Cs₂PtBr₆ nanocrystal outperformed CsPbBr₃ and Cs₂PdBr₆-based PEC systems, as shown in Fig. 4a.⁸¹ Bi-based perovskites have also attracted much attention as viable alternatives due to their isoelectric valence shell with Pb,⁸³ while Bi³⁺ ions, with an ionic radius of 1.03 Å, are analogous to those of Pb (1.19 Å) yet offer greater stability.⁸⁴ Nandigana and colleagues synthesized Bi-based perovskites, specifically Cs₂AgBiCl₆ and Cs₃Bi₂Cl₉, using a hydrothermal method to evaluate their properties towards water splitting.⁷⁷ Their study showed that Cs₂AgBiCl₆ is a better photoelectrode material compared to Cs₃Bi₂Cl₉ as the photocurrent densities of water oxidation of Cs₂AgBiCl₆ and Cs₃Bi₂Cl₉ are 10 μA cm⁻² and 6 μA cm⁻² at 0.9 V vs. Ag/AgCl, respectively, with both materials displaying high stability over 10 h (Fig. 4b).⁷⁷ This work suggests the potential of Bi-based perovskites for green H₂ and O₂ production through PEC water splitting. Despite this progress, their material stability remains a significant challenge for long-term PEC water splitting applications, and the power conversion efficiency of lead-free perovskites still falls short of their lead-based counterparts.

Fig. 4 (a) Current densities of different halide perovskite nanocrystal-based PEC systems. Adapted with permission from ref. 81. Copyright (2021), Wiley-VCH. (b) Stability studies of Cs₂AgBiCl₆ and Cs₃Bi₂Cl₉ photoanodes. Adapted with permission from ref. 77. Copyright (2021), Wiley-VCH. (c) Projected density of states for Cs₃Sb₂I₉ and Cs₃Sb₂I₉ doped with (a) Ag, (b) In, (c) Mo, (d) Nb, (e) Sc, and (f) Bi, respectively. Adapted with permission from ref. 82. Copyright (2024), Elsevier.

In order to improve the efficiency of lead-free perovskite materials, one effective strategy is doping with foreign elements, which helps to optimise band structure for water splitting and supports heterojunction formation to enhance charge separation. Another promising approach involves morphology engineering, where the nanostructure is engineered to reduce electron–hole recombination. These methods aim to refine perovskite materials by altering their elemental composition, enabling targeted adjustments to electronic band structures. Such modifications, including ion substitutions within the perovskite lattice, lead to energy band shifts that enhance responsiveness to visible light.⁸⁵ In addition to compositional changes, encapsulation techniques are used to tackle the long-term stability challenges of lead-based perovskites while facilitating heterojunction formation to boost charge transfer efficiency.^86–88 The following sub-sections will further elucidate these approaches.

3.2.1 Doping. As previously discussed, perovskite materials can be refined by manipulating their elemental composition, thus enabling targeted and precise adjustments to electronic band structures. Modifications such as ion substitution within the perovskite lattice can alter energy bands, which facilitate their responsiveness towards visible light.⁸⁵ Beyond the avenue of modifying the elemental composition of perovskites, another effective approach involves doping foreign elements into the perovskite structure. This strategy can introduce the potential for new mid-gap states between the VB and CB, or even a shift to harness better visible light absorption and enhance photocatalytic activity.

Doping different sites within the perovskite structure with foreign elements can effectively tailor the material's energy band positions, further tuning the material's electronic properties for enhanced performance. This alteration stems from the interplay between the energy levels associated with the introduced dopant and those inherent to the host elements.^82,89,90 This was demonstrated in a study performed by Exner et al. where the effects of doping on the optoelectronic properties of Cs₃Sb₂I₉ were demonstrated using density functional theory (DFT).⁸² Doping Cs₃Sb₂I₉ with In and Ag resulted in enhanced light absorption due to a reduced direct bandgap, while Mo and Nb doping increased light absorption by introducing mid-gap states, as illustrated in Fig. 4c. Consequently, doping stands as a powerful approach for tuning the band structure of perovskite materials, allowing them to achieve tailored electronic properties. Specifically, introducing dopants at the A-site impacts the bandgap by altering the distortion level of the octahedral units, with the degree of distortion influenced by the size of the introduced cation. Consequently, the modification of the A-site plays a significant role in enhancing the overall stability by tuning the tolerance factor. This effect was demonstrated in a study by Zhu et al., where Cs-doped FA_0.85Cs_0.15PbI₃, exhibiting a smaller tolerance factor, showed improved stability compared to its FAPbI₃ counterparts.⁹¹ Zou and co-workers have successfully performed water splitting with Zn-doped SrTiO₃ and BaTiO₃, which would otherwise be unattainable with pure SrTiO₃ and BaTiO₃ due to their high recombination rates and inappropriate band alignment for the reaction.⁹² Increasing the Zn concentration narrows the bandgap by contributing Zn 3d orbitals to the CB, which shifts the CB upwards, straddling it with the water splitting potential in BaTiO₃. Additionally, the Zn doping enhances carrier mobility, which is essential to mitigate the high recombination observed in pure SrTiO₃ and BaTiO₃. However, no H₂ production was observed under visible or UV light. In contrast, doping SrTiO₃ with La and Fe at the A-site rendered it a visible light active catalyst.^93,94 This is due to the interaction between the Ti 3d orbitals and Fe d orbitals, which introduces intra-bandgap states and narrows the bandgap, enabling a red-shift in the UV-vis spectra to allow visible light activity.

As for the well-known lead halide perovskites, substituting Pb²⁺ atoms at the B-site can potentially eliminate the material's toxicity. Moreover, phase stability can be enhanced through manipulating the B–O bond length, as substitution at the B-site influences the Goldschmidt's tolerance factor (eqn (6)), which predicts the structural stability of perovskites. Nag and colleagues doped Pb²⁺ (B-site) ions with metal cations such as Mn²⁺, Sn²⁺, Cd²⁺,Sr²⁺, Co²⁺ and Zn²⁺, observing an increase in phase stability as the bond length decreased in the ABX₃ perovskite structure.⁹⁵ However, B-site doping is generally known to be more challenging due to the higher binding energy compared to A- and X-site components.⁹⁶ Furthermore, in the case of well-known lead halide perovskites, substituting Pb²⁺ atoms may compromise their exceptional optoelectronic characteristics. Since the metal cation in the B-site primarily influences these properties, B-site doping has become a preferred approach to achieve highly efficient perovskite solar cells. For instance, the OER performance of LaNiO₃ was improved by incorporating Ni with Fe⁴⁺, which strengthened the hybridisation between Ni/Fe 3d and O 2p orbitals, thereby optimising the surface absorption energetics for reaction intermediates and improving OER activity.⁹⁷ Additionally, doping the B-site of the LaFeO₃ perovskite with a non-metal element such as phosphorus induced a partial valence state change in iron from Fe³⁺ to Fe⁴⁺. This doping modification approach optimised the bandgap electron filling and enhanced HO* intermediate adsorption for better catalytic performance.⁹⁸

Lattice oxygen, present at the X-site of perovskite-type transition metal oxides, plays a crucial role in the OER through the lattice oxygen-mediated mechanism (LOM). This mechanism involves the participation of lattice oxygen and oxygen vacancies as active sites for OH* adsorption and O₂ evolution. The formation of oxygen vacancies during lattice oxygen oxidation is critical for achieving high OER activity. Anion doping at the X-site of the ABX₃ perovskite structure has emerged as an effective strategy to activate lattice oxygen and generate oxygen vacancies. For instance, substituting fluorine (F) for oxygen in ABX₃ perovskites increases the concentration of reactive oxygen species (O₂²⁻/O⁻), thereby enhancing the redox activity of the BO₆ octahedron and improving OER performance.^99–101 Moreover, bandgap tuning, a widely used approach for optimising perovskite properties, can be achieved by varying the ratio of halogen atoms at the anion site.¹⁰² This is due to the influence of anion on the VB maximum causing the alterations in bandgap. For instance, modifying the halogen composition at the X-site of MAPbX₃ perovskites from I to Br and Cl results in bandgap changes from 1.55 eV to 1.80 eV and 2.30 eV, respectively.¹⁰³ Furthermore, bandgap tuning across the visible spectrum has been demonstrated by Sum et al. with MAPb(Br_3−xCl_x) thin films.¹⁰⁴ Notably, X-site doping has a more prominent impact on bandgap modulation compared to doping at the A- and B-sites. Oxynitrides which are known for their ability to absorb visible light for water splitting can be formed by doping the O atom with N³⁻ to form perovskite oxynitrides such as LaTiO₂N, LaTaON₂, or CaNbO₂N. The introduction of N creates a new VB from the N 2p orbital, which effectively reduces the bandgap. However, a narrower bandgap, which increases solar spectrum absorption, does not necessarily result in enhanced H₂ evolution rates. For instance, LaNbON₂ and BaNbO₂N, with bandgaps of 1.6 and 1.7 eV, respectively, exhibit no water reduction and oxidation activity. This lack of redox activity is attributed to insufficient activation energy for the water splitting reaction to proceed.

B-site doping has been shown to enhance OER performance by facilitating interactions between transition metal ions and oxygen-containing adsorbates. Similarly, X-site doping, which introduces oxygen vacancies, has been explored as a strategy to improve OER activity. Moving beyond traditional single-site doping, dual-site doping has emerged as a highly promising approach for developing highly efficient perovskite oxide electrocatalysts for the OER. Using this approach, Liu et al. demonstrated significant enhancements in OER activity for perovskite oxides, including Ba_0.5Sr_0.5(Co_0.8Fe_0.2)_0.9O_3−δ (BSCF), SrCo_0.9O_3−δ (SCO) and SrNi_0.9O_3−δ (SNO), through the incorporation of iodine at both the B and X-sites.¹⁰⁵ Notably, iodine dual-site doped BSCF exhibited superior OER performance with a reduced overpotential of 290 mV (compared to pristine BSCF) and a Tafel slope of 53 mV dec⁻¹ at a current density of 10 mA cm⁻².¹⁰⁵ Fu et al. investigated dual-site doping in perovskite structures by incorporating strontium (Sr) at the A-site and iron (Fe) at the B-site of LaNiO₃.⁹⁷ While Fe doping alone maintained the Fe³⁺ oxidation state, the introduction of Sr induced a transformation to Fe⁴⁺, which was found to be critical for enhancing OER activity. This dual-site doping approach led to a 6-fold increase in current density, whereas Fe³⁺ alone had a negligible effect.⁹⁷ Spectroscopic and theoretical studies attributed this improvement to the formation of Fe⁴⁺–O²–Ni³⁺ configurations, which facilitate electron transfer and accelerate the rate-determining deprotonation step of water oxidation by lowering the energy barrier.⁹⁷

3.2.2 Heterojunction formation. Unleashing the advanced photocatalytic potential of perovskite materials requires exploring the construction of perovskite heterojunctions, offering a pathway beyond the limitations of single-component systems. By combining two distinct materials, these heterojunctions enhance the separation and transfer of charge carriers along the electric potential gradient, thus improving photocatalytic efficiency.^106–109 Moreover, effective charge separation significantly reduces the rate of electron–hole recombination, a common bottleneck in photocatalytic processes.^106–109 Additionally, the integration of a perovskite with narrow bandgap semiconductors within a heterostructure can further reduce in the overall bandgap of the material.^110–112 This bandgap narrowing enhances light absorption across a broader spectrum, boosting photocatalytic performance.

Heterostructures typically exhibit a band alignment that fall into one of three different configurations: type I (straddling gap), type II (staggered gap) and type III (broken gap). Most perovskite-based heterostructures are reported to exhibit type II band alignment, characterised by cascade-like transfer of photogenerated electrons and holes, which effectively addresses the challenge of charge carrier separation in perovskites.¹¹³ For instance, Jang et al. engineered a PbTiO₃ perovskite@TiO₂ heterojunction nanotube, with the PbTiO₃ layer intricately formed along the inner wall of TiO₂ nanotubes.¹¹⁴ This type II heterojunction facilitates efficient separation of light-induced electron–hole pairs. Beyond its role in promoting efficient charge separation, PbTiO₃ functions as a photosensitiser, extending the absorption range of solar light. Consequently, this heterostructure significantly improves the PEC water splitting performance of the photoanode. To overcome the sluggish kinetics of water oxidation, Jayaraman et al. developed a dual-absorber heterojunction photoanode for PEC water oxidation, combining BiVO₄ with the extremely stable Cs₂PtI₆ perovskite.¹¹⁵ This design increased the photocurrent density from 0.63 to 0.92 mA cm⁻² at 1.23 V vs. RHE. The improvement is attributed to enhanced charge separation, as photogenerated holes transfer from the VB of BiVO₄ to the VB of Cs₂PtI₆, while the panchromatic absorption of Cs₂PtI₆ broadens the light absorption range of the photoanode. Similarly, Kumar and Ganguli constructed a type II heterojunction using NaNbO₃@CuS, as illustrated in Fig. 5a.⁷⁹ The combination of n-type NaNbO₃ and p-type CuS forms a p–n junction at the core–shell interface, which enhances charge separation and suppresses recombination. The successful creation of this p–n heterojunction leads to a 6-fold increase in photocurrent density, reaching 4.87 mA cm⁻² at 0.45 V vs. RHE, compared to bare CuS.

Fig. 5 (a) Alignment of band structures of NaNbO₃ and CuS. Adapted with permission from ref. 79. Copyright (2022), Elsevier; (b) Charge transfer mechanism of the LaFeO₃/RGO S-scheme heterojunction. Adapted with permission from ref. 116. Copyright (2023), Elsevier; (c) FE-SEM images of (i) PMTS (nanospheres), (ii) PMTF (nanoflakes), (iii) PMTH (hierarchical flowers) and (iv) PMTT (thin microbelts) perovskites on the Ni foam respectively. Adapted with permission from ref. 117. Copyright (2017), Elsevier; (d) correlation between current densities of the Co(OH)_x/BaNbO₂N photoanode and BET surface areas and average particle size of the BaNbO₂N powder. Adapted with permission from ref. 118. Copyright (2022), Elsevier.

Recently, Ghosh and colleagues synthesized an S-scheme heterojunction of LaFeO₃/reduced graphene oxide (RGO).¹¹⁶ Unlike the charge transfer mechanism in type II heterojunctions, S-scheme heterojunctions feature a staggered band structure similar to that of type II heterojunctions but with a distinct charge transfer pathway. As shown in Fig. 5b, it can be seen that RGO exhibits a higher CB and VB position compared to LaFeO₃. When LaFeO₃ and RGO come into contact, electrons from RGO spontaneously diffuse into LaFeO₃, creating an electron depletion layer in RGO and an electron accumulation layer near the interface in LaFeO₃. This electron transfer generates an internal electric field that accelerates the transfer of photogenerated electrons from LaFeO₃ to RGO. Additionally, the contact between the materials induces a Fermi level shift, resulting in band bending. This prompts recombination of electrons from LaFeO₃ and holes from RGO, driven by coulombic attraction. This selective recombination eliminates less effective charge carriers, while preserving the more potent photogenerated electrons and holes to actively participate in photocatalytic reactions.¹¹⁹ To address the rapid charge recombination of LaFeO₃, RGO nanosheets have been employed as charge separators. Loading RGO nanosheets onto LaFeO₃ resulted in an approximately 21-fold increase in current density compared to bare LaFeO₃. This enhancement is attributed to the high electrical conductivity and specific surface area of RGO nanosheets, which provide efficient electron channels and facilitate charge carrier separation, hence further improving the photocatalytic performance of the composite catalyst.

3.2.3 Morphological engineering. The morphology of perovskite materials, encompassing factors such as crystal structure, size, crystallinity, and particle shape, plays a pivotal role in determining their photocatalytic activity. By tailoring these morphological characteristics, the overall PEC water splitting performance of perovskite materials can be significantly enhanced. While altering morphology may not necessarily render perovskites active under visible light irradiation, it can significantly impact key properties such as diffusion length and the separation and transfer efficiency of photoinduced charge carriers. These factors are essential in designing efficient photoelectrodes for PEC water splitting. For instance, Chandrasekaran et al. successfully synthesized lead magnesium titanate (PMT) perovskites with four distinct morphologies—spheres, flakes, hierarchical flowers, and thin microbelts—through a facile solution-based method (Fig. 5c).¹¹⁷ PMT with a thin microbelt morphology exhibited a significant increase in electroactive sites and enhanced electrolyte penetration compared to other morphologies, owing to its larger surface area (66.64 m² g⁻¹), higher optical absorbance, and more active sites, which collectively facilitated rapid charge transfer processes.¹¹⁷ The open spaces between adjacent belts further improved the kinetics of redox reactions, promoting increased interaction between the electrolyte and the electrode. These synergistic effects resulted in a remarkable improvement in photocurrent density, achieving 1.77 mA cm⁻² at 1.23 V vs. RHE. This performance is approximately 4.52 times higher than that of PbTiO₃ (0.39 mA cm⁻² at 1.23 V vs. RHE).¹¹⁷ The insights gained from this study provide valuable strategies for the rational design of nanostructured perovskites, paving the way for developing high-performance materials for PEC water splitting applications.

In addition to modifying morphology during initial synthesis, the properties of perovskites can be altered through doping. For instance, the crystal structure of BiFeO₃ can be tailored by the rational introduction of dopants such as La and Se.¹¹⁶ This doping induced a structural phase transformation from a rhombohedral to an orthorhombic crystal structure, corresponding to bandgap energies of 2.04 eV and 1.76 eV, respectively, thereby reducing the energy required to excite electrons under light irradiation. Moreover, the La and Se co-doping transformed the morphology of BiFeO₃ from particles into nanosheets, significantly increasing the surface area available for catalytic reactions. These modifications resulted in La, Se co-doped BiFeO₃ achieving superior photocatalytic activity under visible light. Therefore, the restructuring of the crystal and particle morphology of perovskites has the potential to enhance photoexcitation and improve their catalytic performance.

The crystallite size of a semiconductor plays a crucial role in determining its PEC water splitting efficiency, primarily by influencing the active surface area available for redox reactions. This relationship was exemplified in BaNbO₂N, where increasing calcination temperatures resulted in larger crystallite size.¹¹⁸ Interestingly, higher temperatures led to decreased crystallinity due to the enlargement of oxide particles, which negatively affected the degree of nitridation and hindered the formation of porous BaNbO₂N from Ba₅Nb₄O₁₅. The variations in crystallite size and surface properties of BaNbO₂N were found to significantly impacted PEC water splitting performance, as illustrated in Fig. 5d. Smaller oxynitride crystallites were correlated with a higher BET surface area, signifying more active sites for the water splitting reaction. Additionally, Kim and Choi highlighted a critical aspect of crystallite size in BiVO₄ photoanodes. When the crystal size is smaller than the hole diffusion length, bulk carrier recombination is effectively suppressed, resulting in longer hole lifetimes and improved charge separation efficiency.¹²⁰ These findings highlight the importance of optimising crystallite size to balance surface activity and charge dynamics for enhanced PEC water splitting performance.

Significant advancements have been made in enhancing the performance of perovskites through approaches such as heterojunction formation and morphology engineering, as summarised in Table 1. However, a key challenge persists: the recombination of photogenerated charge carriers, which continues to limit overall efficiency. This issue is particularly pronounced in lead-free perovskites which still fall short of their lead-based counterparts in terms of efficiency. This efficiency gap arises primarily from the inherent material properties and challenges in optimising charge carrier dynamics. Additionally, while lead-free perovskites are generally more stable than their lead-based alternatives, they remain vulnerable to degradation under certain conditions, including combined exposure to moisture, heat, and light. These factors can significantly reduce device lifetime and performance. To address these challenges, integrating hybrid materials such as MOFs with perovskites has emerged as a promising strategy. This integration can enhance perovskite crystallinity, improve charge separation, and boost PEC performance, making it a critical step toward ensuring the long-term stability and efficiency of perovskite-based PEC systems.

Table 1 Summary of perovskite-based photoelectrodes for water splitting and H₂ generation application

Photoelectrode	Configuration/morphology	Synthesis method	Electrolyte	Photocurrent density (mA cm^−2) (at 1.23 V vs. RHE)	Stability	Half-reaction	Ref.
PbTiO3@TiO2	Type II heterojunction	Electron beam vapor deposition, anodic oxidation and spin coating	0.1 M KOH	0.30	Steady-state behaviour throughout 300 s of operation	OER	114
BiVO4/Cs2PtI6	Type II heterojunction	Spin coating, thermal oxidation and drop-casting	KOH solution (pH 13)	0.92	Tested for 2 h, sustained for 2 h	OER	115
Cs2PtI6	—	Solvothermal method	KOH solution (pH 11)	0.80	Tested for 12 h, sustained for 12 h	OER	121
NaNbO3/CuS	Type II heterojunction	Hydrothermal method	0.5 M Na2SO4	4.87 (at 0.45V vs. RHE)	Steady-state behaviour throughout 300 s of operation	HER	79
LaFeO3	—	Hydrothermal method	0.1 M Na2SO4	0.17 μA cm^−2 (at 0.9 V vs. Ag/AgCl)	—	OER	116
LaFeO3/RGO	S-scheme heterojunction			3.67 μA cm^−2 (at 0.9 V vs. Ag/AgCl)	—	OER	116
PbTiO3	—	In situ growth on Ni foam	0.1 M Na2SO4	0.39	Steady-state behaviour throughout 400 s of operation	OER	117
PMTT	Morphological (thin nanobelt)			1.77	Steady-state behaviour throughout 400 s of operation	OER	117
PMTS	Morphological (sphere)			0.52	Steady-state behaviour throughout 400 s of operation	OER	117
PMTF	Morphological (flake)			0.51	Steady-state behaviour throughout 400 s of operation	OER	117
PMTH	Morphological (hierarchical flower)			1.18	Steady-state behaviour throughout 400 s of operation	OER	117
BaNbO2N	—	Thermal annealing and nitridation	0.5 M KBi	5.2	Tested for 1 h, 30% of photocurrent retained	OER	122
BaNbO2N	Morphological (particle size – 1.05 μm)	Polymerised complex method	0.5 M KBi	1.42	—	OER	118
BaNbO2N	Morphological (particle size – 1.93 μm)			2.20	Tested for 1 h, 70% of photocurrent retained	OER	118
BaNbO2N	Morphological (particle size – 4.11 μm)			1.17	—	OER	118
BaNbO2N	Morphological (particle size – 6.92 μm)			0.58	—	OER	118

---

4 MOFs

MOFs, also known as porous coordination polymers (PCPs), are constructed by linking metal nodes with organic ligands through coordination bonds, as illustrated in Fig. 6a. The diverse combinations of metal nodes and ligands give rise to remarkable structural variability, enabling their use in a wide range of applications, such as liquid and gas separation and storage, drug delivery, battery electrodes, and sensing systems.^123–126 One of the main advantages of MOFs is their structural tunability, which allows flexible designs in terms of geometry and chemical functionality.¹²⁷ MOFs also exhibit high porosity, resulting in a large surface area, while their tuneable pore size enhances selectivity for specific separation applications. For instance, MOF pores can be functionalised to target the removal of specific toxic gases. The pore size of MOFs can be tuned from microporous to mesoporous by manipulating the ligand lengths within the framework.¹²⁸ In contrast to conventional semiconductors, MOFs offer superior electrical and optical properties owing to their photoactive organic ligands, which enhance light capture.¹²⁹ Their bandgap can be readily regulated by modifying the building units, as shown in Fig. 6b.^129,130 Additionally, the metal sites in MOFs were reported to act as additional active catalytic centers, which reduces charge diffusion lengths and improves catalytic performance.^127,131 Certain MOFs also demonstrate exceptional stability in water and harsh chemical environments, making them recyclable for repeated use.^129,130,132 When combined with perovskites, MOFs exhibit unique synergistic properties. They can effectively passivate defects on the perovskite surface or at grain boundaries, improving material stability and performance.¹³³ The small apertures and large internal pores of MOFs allow for efficient encapsulation, preventing leaching of encapsulants while permitting reactant molecules to enter the framework, thus enhancing catalytic activity.¹²⁸

Fig. 6 (a) Structure of metal–organic frameworks (MOFs); (b) band gaps of some common MOFs; (c) schematic illustration of the ZnO@M@ZIF-67 photoanode for PEC water splitting. Adapted with permission from ref. 115. Copyright (2017), Royal Society of Chemistry.

4.1 Classification of MOFs

MOFs are a versatile class of crystalline materials formed by connecting metal ions or clusters with multidentate organic linkers. Their modular structure enables the creation of a wide variety of frameworks with customizable properties. Key categories of MOFs include:

Zeolitic Imidazolate Frameworks (ZIFs): ZIFs consist of tetrahedral metal ions, such as Zn²⁺ and Co²⁺, coordinated with imidazolate linkers, resulting in structures resembling zeolites. This similarity endows ZIFs with remarkable thermal and chemical stability. Their large pore sizes and robustness make them ideal for developing advanced MOF composites.

Isoreticular MOFs (IRMOFs): IRMOFs are a series of cubic MOFs featuring high porosity. They are typically constructed using zinc or other divalent metal ions linked with dicarboxylate ligands like terephthalate, making them highly adaptable for various applications.

Materials Institute Lavoisier (MIL) MOFs: the MIL series comprises MOFs based on trivalent metal ions such as Cr³⁺, Fe³⁺, or Al³⁺, combined with carboxylate linkers. These frameworks are known for their high stability, ultrahigh surface areas, uniform pores, and permanent porosity. Such features make them excellent candidates for biomedical and environmental applications, with a unique capacity for transfer between micropores and mesopores under external influences.

UiO Series: developed at the University of Oslo, the UiO series features MOFs based on zirconium (Zr⁴⁺) nodes linked by dicarboxylate ligands. These frameworks exhibit exceptional chemical and thermal stability. For example, UiO-66(Zr), synthesized via a solvothermal method using ZrCl₄ and BDC, demonstrates excellent thermodynamic stability and can withstand highly alkaline conditions (pH 14).

Porous Coordination Networks (PCNs): PCNs are a diverse family of MOFs known for their intricate and complex structures. They exhibit high porosity and include variants like PCN-333, PCN-224, PCN-222, and PCN-57, each offering unique properties for specific applications.

This classification underscores the extensive variety of MOF architectures and compositions, enabling the design and synthesis of materials with tailored properties for targeted applications, including gas storage, drug delivery, and sensing.

4.2 Synthesis of MOFs

The synthesis of MOFs typically involves the self-assembly of metal ions or clusters with organic linkers under solvothermal or hydrothermal conditions. Key factors that influence MOF synthesis include the choice of metal ions, organic linkers, solvents, temperature, and reaction time.

4.2.1 Solvothermal method. The solvothermal approach is the most common method for synthesizing MOFs. It involves reacting metal salts with organic linkers in a solvent, typically a polar organic solvent such as DMF, DEF, or NMP, under high temperatures and pressures in an autoclave. First proposed by Professor Yaghi and the Nalco Chemical Company in 1995, this method leads to MOF crystallization through precipitation.¹³⁴ The solvent may dissolve partially or completely as the temperature approaches its boiling point, allowing for controlled crystal growth with high crystallinity.

4.2.2 Microwave-assisted method. This method employs microwave radiation to heat the reaction mixture, offering advantages such as faster reaction times, higher energy efficiency, and reduced waste compared to conventional heating. It has been used to create numerous MOFs and metal clusters. Key benefits of this approach include shorter synthesis times, higher yields, cost-effectiveness, environmental friendliness, and high-purity products. For example, HKUST-1, with the formula [Cu₃(BTC)₂(H₂O)₃] (where BTC³⁻ is 1,3,5-benzenetricarboxylate), has been synthesized via the microwave method, resulting in enhanced yields. By adjusting parameters like temperature and reactant concentration, crystals with specific shapes, sizes, and nucleation properties can be achieved. However, maintaining precise conditions for crystal growth in this method can be challenging.^135,136

4.2.3 Mechanochemical method. Mechanochemical synthesis involves grinding solid reactants under high pressure, facilitating MOF formation without the use of solvents. This method is an environmentally sustainable alternative to solvothermal and hydrothermal techniques, which often require long reaction times and large amounts of organic solvents, posing significant environmental challenges. Mechanochemical synthesis is particularly useful for producing MOFs that are difficult to obtain through solution-based approaches, making it a promising option for greener and more cost-effective MOF production. A significant challenge associated with mechanochemical methods is the potential for contamination of the final products. This contamination can arise from various sources, including by-product phases generated during the milling process, as well as wear and tear from the milling balls and the milling container itself.¹³⁷

4.3 Application of MOFs in PEC water splitting

Recently, MOFs have garnered attention as promising photoelectrodes for PEC water splitting. Integrating MOFs with semiconductor photoanodes offers several advantages, including enhanced light utilisation, improved charge transport, and increased stability. The heterostructure formed by pairing MOFs with semiconductors significantly boosts light-harvesting efficiency. Photoactive MOFs possess broad absorption bands, allowing them to capture a wide spectrum of wavelengths effectively. Additionally, the high porosity of MOFs does not impede the light absorption path of the semiconductor, further optimising the overall water splitting efficiency.^138,139 Furthermore, MOFs effectively mitigate charge carrier recombination through efficient charge transfer at the semiconductor/MOF interface. Their high porosity also provides a large number of active sites, facilitating catalytic activity and contributing to improved PEC performance.¹³⁹ The stability of the device is significantly enhanced as the ordered structure of MOF could serve as a protective layer.¹³⁸ For instance, a composite system was developed using visible light-active ZIF-67 to encapsulate a noble-metal sensitized semiconductor, ZnO@Au, as depicted in Fig. 6c.¹⁴⁰ This synergistic design resulted in a remarkable improvement in photoconversion efficiency, reaching up to 0.80%, alongside an increased photocurrent density. The combination of the visible light-responsive ZIF-67 and the UV-responsive ZnO enabled efficient light utilisation across a broader spectrum. In terms of stability, the ZnO@Au@ZIF-67 composite demonstrated a consistent O₂ evolution rate and stable current density over 5 cycles of operation. Structural analysis through XRD and SEM further confirmed the maintained integrity of the composite after the 5 cycles, in stark contrast to the decreased crystallinity observed in the unencapsulated ZIF-67 composite. This highlights that the encapsulation of ZnO@Au within the MOF shell not only improved its photoconversion performance but also significantly enhanced its structural integrity and durability.

Ti-based MOFs, such as NH₂-MIL-125, are commonly employed as photosensitisers, where their organic ligands serve as electron donors, transferring electrons to Ti-oxo clusters.¹³⁸ In a study conducted by Yoon et al., a TiO₂/NH₂-MIL-125 heterostructure was developed, achieving a remarkable 2.7-fold increase in photocurrent density compared to bare TiO₂ nanorods.¹⁴¹ NH₂-MIL-125 was incorporated into the pores of TiO₂ nanotube arrays (TNTAs) to overcome the inherent challenges of TNTAs, including their wide bandgap of 3.2 eV and low separation efficiency.¹⁴² This integration resulted in the formation of TNTAs@Ti-MOF, which demonstrated a substantial 14-fold increase in H₂ generation compared to bare TNTAs. This improvement is primarily attributed to the heterostructure's enhanced light utilisation, which facilitated excellent charge separation and stability. Additionally, the study by Song et al. highlight the critical role of the MOF crystal structure in influencing H₂ generation efficiency, further validating the potential of Ti-based MOFs in addressing the limitations of conventional TiO₂ systems¹⁴³ In their study, two distinct MOFs were synthesized (compound 1 – C₄₂H₄₆CuN₆O₂₁S₄ and compound 2 – C₂₁H₁₉CuN₃O₉S₂), both utilising the same copper (Cu) metal node and the ligand 4′-(2,4-disulfophenyl)-3,2′:6′,3′′-terpyridine (H₂DSPTP).¹⁴³ Despite sharing identical metal nodes and ligands, the resulting MOFs exhibited different crystal structures, leading to varied photocatalytic activities under visible and near-infrared-light. Furthermore, the π–π interactions present within the synthesized MOFs were found to facilitate the transfer of photogenerated holes, effectively reducing charge carrier recombination.¹⁴³ Interestingly, this effect was more pronounced in compound 2 compared to compound 1, possibly due to shorter interatomic distances and stronger π–π interactions in compound 2.¹⁴³ Consequently, when designing MOFs for PEC H₂ production, it is crucial to pay careful attention to the crystal structure optimisation to maximise light absorption and charge carrier dynamics.

MOFs offer a unique advantage in designing microstructures for corresponding metal oxides, while the pyrolysis of their organic ligands induces a rich porous structure. This combination results in materials with high surface areas and abundant reaction sites, making them ideal for photoelectrode construction^144–146 An increasing number of MOF-derived materials have been developed for the purpose of constructing photoelectrodes. For instance, a Co₃O₄ thin film derived from ZIF-67 was deposited onto a BiVO₄ photoanode to address its suboptimal OER performance.¹⁴⁷ The resulting Co₃O₄/BiVO₄ photoanode exhibited a significant 3.2-fold improvement in PEC performance compared to pristine BiVO₄, attributed to the increased surface area and additional active sites provided by the MOF-derived Co₃O₄. In another example, a controlled anatase-rutile TiO₂ junction was created using NH₂-MIL-125 as a precursor, producing a TiO₂ photoanode sensitized with CdSe@CdS quantum dots (QDs).¹⁴⁸ The TiO₂-QDs displayed an octahedral morphology similar to the Ti-MOF precursor, resulting in a 47.6% increase in photocurrent density compared to anatase TiO₂. Additionally, the MOF-derived TiO₂ film exhibited improved electron transfer efficiency due to optimised band alignment within the photoelectrode structure. These examples underscore the role of MOFs as effective templates for controlling the morphology and performance of photoelectrode thin films, leading to significant enhancements in PEC efficiency. For a detailed overview of MOF and MOF-derived photoelectrodes and catalysts for water splitting and H₂ generation, please refer to Table 2.

Table 2 Summary of MOF and MOF-derived photoelectrodes/catalyst for photocatalytic and photoelectrocatalytic application

Photoelectrode	MOF	Electrolyte	Synthesis method	Photocurrent density (mA cm^−2)	H2 production rate (μmol h^−1 g^−1)	Stability	Ref.
TiO2/NH2-MIL-125	NH2-MIL-125	0.5 M Na2SO4	Solvothermal method	0.03 at 0.20 V vs. Ag/AgCl	—	Steady-state behaviour throughout 200 s of operation	149
ZnO–C(ZIF-8)	ZIF-8	0.2 M Na2SO4	Two-step calcination method	0.08 at 1.23 V vs. RHE	—	No obvious deactivation after 5 degradation cycles	150
NH2-UiO-66	NH2-UiO-66	—	Solvothermal method	—	1.72	—	151
Pt/NH2-UiO-66		—		—	50.26	50% decay after the 1st catalytic run
Pt@NH2-UiO-66		—		—	257.38	No obvious change throughout 4 catalytic runs for 10 h
ZnO@Au@ZIF-8	ZIF-8	0.5 M Na2SO4	Direct in situ growth of MOF on noble-metal-sensitized ZnO nanorods	0.69 at 0.60 V vs. SCE	—	Steady-state behaviour throughout 3 h of operation	140
ZnO@Au@ZIF-67	ZIF-67			1.93 at 0.60 V vs. SCE	—	Steady-state behaviour throughout 3 h of operation
Pd/UiO-66	UiO-66	0.2 M Na2SO4	Impregnation reduction method	1.25 at 0.80 V vs. SCE	—	Continuous decrease in H2 production over 4 cycles for 20 h	152
MoS2 QDs/NH2-UiO-66/G	NH2-UiO-66	0.1 M Na2SO4	Solvothermal method	—	0.83	Continuous decrease in H2 production over 4 cycles for 720 min	153
						∼30% decay in current density throughout 400 s of operation
Fe2O3:Ti/NH2-MIL-101(Fe)	NH2-MIL-101	1.0 M NaOH	Surfactant-assisted solvothermal method	2.27 at 1.23 V vs. RHE	—	13.3% decay in current density throughout 2500 s of operation	154
Ni(OH)2/ZIF-8/ZnO/NF	ZIF-8	0.1 M KOH	Seed-mediated hydrothermal growth process	1.95 at 1.23 V vs. RHE	—	Steady state behaviour throughout 10 000 s of operation	155
ZnO/ZIF-8/67	ZIF-8 and ZIF-67	0.5 M Na2SO4	Solvothermal method	0.11 at 1.23 V vs. RHE	—	54% decay in current density throughout 2000 s of operation	156
Fe2O3/Co-MOF	Co-MOF	1.0 M NaOH	Ion-exchange method	2.0 at 1.23 V vs. RHE	—	—	157
PW12@NH2-UiO-66	NH2-UiO-66	0.2 M Na2SO4	One-step solvothermal method	—	72.70	No structural changes after 72 h of immersion in deionised water	158
Ni-MOF/ZnO	ZIF-8	—	Chemical etching method	IPCE: about 11.00% at 355 nm at 0.5 V vs. Ag/AgCl	—	—	159
ZnNi MOF@ZnO	ZnNi MOF	0.5 M Na2SO4	Hydrothermal and ion exchange process	1.40 at 1.23 V vs. RHE	—	Steady state behaviour throughout 100 s of operation	160
NH2-MIL-125/TiO2	NH2-MIL-125	1.0 M NaOH	Hydrothermal method	1.63 at 1.23 V vs. RHE	—	Steady state behaviour throughout 30 days of operation	141
TixFe1−xOy shell/Fe2O3 core	NH2-MIL-125	1.0 M NaOH	Solvothermal process, followed by a two-step calcination process	0.72 at 1.23 V vs. RHE	—	1.1% decay in current density throughout 5 h of operation	161
Co-MOF-BiVO4	ZIF-67	0.5 M KH2PO4	Electrodeposition and in situ growth method	2.35 at 1.23 V vs. RHE	—	Slight decrease in current density throughout 10 000 s of operation	147
UiO-66-NH2@ Au@CdS	NH2-UiO-66	0.1 M Na2SO4	Solvothermal and hydrothermal process	—	39.50 μmol h^−1	Stable H2 generation throughout 24 h of operation and no structural changes observed after	162
TNTAs-Ti-MOF	NH2-MIL-125	—	Hydrothermal method	—	132.86 μmol h^−1 cm^−2	Slight decline in H2 generation after 8 cycles of operation	142
Ag/NH2-MIL-125/TiO2	NH2-MIL-125	0.5 M Na2SO4	Hydrothermal and in situ reduction process	1.06 at 1.23 V vs. RHE	—	Steady state behaviour throughout 3 h of operation	163
MOF-derived TiO2/CdSe@CdS	NH2-MIL-125	0.35 M Na2SO3	MOF-template method	10.72 at 0.9 V vs. RHE	—	2.6% decay in current density throughout 2h of operation	148
CdSe/NH2-MIL-101(Cr)	NH2-MIL-101	—	Hydrothermal method	—	17 664.00 μmol g^−1	7.3% decrease in H2 production after 4 cycles for 8 h	164

---

4.4 Modification strategies of MOFs

MOFs have shown tremendous potential for PEC water splitting due to their customisable structures, extensive surface areas, and versatile functionalities. However, unlocking their full performance requires addressing key challenges and exploring strategies for enhancement. This section delves into various approaches to optimise MOFs for PEC applications, including bandgap engineering to enhance visible light absorption, the selection and optimisation of co-catalysts to boost catalytic activity, morphological engineering to improve material properties, and heterojunction formation to facilitate efficient charge transfer.

4.4.1 Enhancement in visible light absorption ability. The efficiency of PEC water splitting systems largely depends on the light utilisation capabilities of the photoelectrodes, prompting extensive research efforts to develop MOFs with visible-light activity.^165–167 In a pioneering study by Silva et al. in 2010, they first demonstrated the potential of MOFs as photocatalysts for the HER.¹⁶⁵ Using two Zr-based MOFs, UiO-66 and NH₂-UiO-66, the study achieved H₂ generation rates of 248 and 372 μmol h⁻¹ g⁻¹, respectively.¹⁶⁵ The superior performance of NH₂-UiO-66 was attributed to the amine-functionalised ligand, which functions as an auxochromic and bathochromic group in the aromatic ring. This functionalisation shifted the absorption edge of NH₂-UiO-66 into the visible-light region, enhancing its photocatalytic activity.¹⁶⁵ Building on this approach, Nasalevich and colleagues employed amine group functionalisation to extend the absorption edge of MIL-125 from 350 nm to 550 nm (NH₂-MIL-125), enabling the HER under visible light irradiation (Fig. 7a).¹⁷² This phenomenon arises from the interaction between the lone pair of electrons on the amine group and the π*-orbitals of the benzene ring, which donates electron density to the anti-bonding orbitals, thereby effectively reducing the bandgap.¹⁷² Other functional groups, such as –CH₃, –Cl, and –OH, have also been used to functionalise the terephthalate (bdc) ligand in MIL-125, as demonstrated by Hendon and co-workers.¹⁶⁹ As shown in Fig. 7b, DFT calculations revealed that incorporating 10% bdc-R enables flexible bandgap tuning between 3.5 and 2.4 eV, thus enabling the optical response of MIL-125 to be tailored.¹⁶⁹ The extent of bandgap reduction correlates with the electron-donating strength of the substituents. For instance, weak electron donors such as –Cl and –CH₃ moderately decrease the bandgap of MIL-125 to 3.5 eV, while the stronger –NH₂ group further reduces it to 2.6 eV.¹⁶⁹ Beyond ligand functionalisation, Logan et al. extended this approach by decorating the amine group with alkyl chains of varying lengths.¹⁶⁶ Increasing the length of these alkyl substituents led to electron accumulation in the terephthalate ring, resulting in a gradual decrease in the bandgap as illustrated in Fig. 7c and d.¹⁶⁶

Fig. 7 (a) Schematic illustration of the HER over NH₂-MIL-125 under visible light irradiation. Adapted with permission from ref. 168. Copyright (2013), Royal Society of Chemistry; (b) computational studies predicted the bandgaps of MIL-125 and its analogues containing functionalised ligands (inset: molecular structure of the terephthalate (bdc) ligand in MIL-125 with (left) monosubstituted linkers bdc-R and (right) diaminated linker bdc-(NH₂)₂). Adapted with permission from ref. 169. Copyright (2013), American Chemical Society; (c) molecular structure of the bdc ligand in MIL-125 and the N-substituted isoreticular MOFs and (d) its respective measured bandgap. Adapted with permission from ref. 166. Copyright (2017), Royal Society of Chemistry; (e) cycle runs for the photocatalytic H₂ production of 1% MoS₂/(50%) U6–CdS and pure CdS. Adapted with permission from ref. 170. Copyright (2015), Elsevier; (f) hypothetical mechanism of the NiFe-MOF/TiO₂ photoanode for PEC water splitting. Adapted with permission from ref. 171. Copyright (2020), Elsevier.

In addition to ligand modifications in MOFs, the light absorption edge can be extended by incorporating photosensitisers. For example, He and his group embedded CdS nanoparticles into MIL-100 via disproportionation in an aqueous solution.¹⁷³ The introduction of CdS significantly enhanced light absorbance between 300 nm and 500 nm, accompanied by a noticeable red shift toward the absorption edge of bare CdS.¹⁷³ Similarly, semiconductors with narrow bandgaps have been utilised as photosensitisers. For instance, Yuan and co-workers integrated In₂S₃, a semiconductor with a narrow bandgap, into MIL-125 as a core–shell composite, extending the light absorption edge to approximately 600 nm.¹⁷⁴ Dye sensitisation has also gained traction since the first study of dye-sensitised TiO₂ films in 1991, which demonstrated improved light-to-electric energy conversion efficiency.¹⁷⁵ Strong van der Waals interactions and π–π stacking are expected to form between MOFs and dyes, as both contain benzene rings.¹⁷⁶ These interactions are critical for facilitating efficient charge transfer from the dye to the MOF in photocatalytic applications. By constructing dye-sensitised MOF systems, researchers have successfully extended the optical response of MOFs into the visible-light region. For instance, He et al. sensitized Pt-loaded UiO-66 with Rhodamine B (RhB) dye, extending the absorption range to approximately 600 nm and significantly enhancing absorption intensity in the visible-light region.¹⁷⁶ Upon absorbing light energy, RhB transforms into its excited state (RhB*), transferring photogenerated electrons to the VB of UiO-66, which then participates in the HER.¹⁷⁶ However, RhB suffers from poor photostability, limiting its suitability as a long-term photosensitiser.¹⁷⁷ To address this issue, Yuan and his group employed Erythrosin B (ErB) dye as a stabilising and cost-effective light-absorbing antenna.¹⁷⁸ The ErB-sensitised Pt-UiO-66 system exhibited an extended optical response into the visible-light region, reaching approximately 549 nm due to the narrow bandgap of ErB dye.¹⁷⁸ These studies highlight the potential of dye-sensitised MOF systems for H₂ production, particularly given the cost-effectiveness and simplicity of dye sensitization. However, the stability of dyes remains a critical factor for consideration, as their degradation could lead to environmental pollution, emphasizing the quest for more robust and sustainable photosensitisers.

4.4.2 Co-catalyst incorporation. Despite advancements in visible-light-responsive MOF photocatalysts, their photocatalytic activity still requires substantial improvement for broader practical applications. To address this, researchers have focused on integrating various molecular functional components, such as metal nanoparticles (NPs) and metal oxides, which act as co-catalysts, into MOFs.^151,165 Incorporating suitable co-catalysts onto MOF surfaces can significantly enhance catalytic activity, reduce overpotential, and promote efficient charge transfer. For example, Xiao et al. demonstrated improved hydrogen production by incorporating Pt nanoparticles into the cavity of NH₂-UiO-66, rather than merely attaching them to the surface.¹⁵¹ Pt, in its metallic form, is a well-established co-catalyst for H₂ production due to its ability to efficiently trap photogenerated electrons and facilitate charge separation.¹⁷⁹ Although both Pt/NH₂-UiO-66 and Pt@NH₂-UiO-66 exhibit accelerated transient absorption decay kinetics due to the additional electron transport pathway from NH₂-UiO-66 to Pt nanoparticles, the superior H₂ production in Pt@NH₂-UiO-66 can be attributed to the shortened electron diffusion pathway.¹⁵¹ This modification reduces electron recombination and further accelerates transient absorption decay kinetics. Additionally, the high porosity of NH₂-UiO-66 enhances proton accessibility to Pt nanoparticles, thereby further boosting H₂ production.¹⁵¹ Optimising co-catalyst loading is crucial for maximising H₂ production in MOFs. Toyao et al. discovered in the Pt/NH₂-MIL-125 system that Pt suppressed the formation of Ti³⁺ species and inhibited electron–hole recombination, resulting in a high H₂ evolution rate of 500 μmol h⁻¹ g⁻¹ with an optimal Pt loading of 1.5 wt%.¹⁶⁸ Other noble metal co-catalysts, such as Ir, Pd, Au, and Ru, have also been explored for similar enhancements. The interaction between MOFs and noble metals, driven by differences in Fermi levels, generates an electric field that promotes charge separation of photogenerated electron–hole pairs.^180–183 This results in the formation of a Schottky junction, where the positively charged noble metal and negatively charged MOF synergistically enhance charge separation.^180–183 Additionally, the porous nature of MOFs stabilises metal nanoparticles, preventing their aggregation and ensuring efficient diffusion of reactants and products.¹⁸⁴ Common methods for incorporating metal nanoparticles into MOFs include in situ hydrothermal or solvothermal techniques. For instance, Wang et al. immobilised noble metals such as Ir, Pt, Ru, Au, and Pd onto zirconium-porphyrinic MOF nanotubes (HNTM) using a hydrothermal method, resulting in outstanding photocatalytic H₂ evolution.¹⁸⁵ Notably, HNTM loaded with both Ir and Pt nanoparticles achieved a H₂ generation rate of 201.9 μmol g⁻¹ h⁻¹, compared to negligible activity in pristine HNTM. This enhancement was attributed to the synergistic effect of Ir serving as a photosensitiser and Pt acting as a catalyst, which improved charge separation and reduced resistance.¹⁸⁵

While MOFs can prevent nanoparticle aggregation, hydrothermal and solvothermal methods may sometimes impair photocatalytic activity. To overcome this limitation, encapsulating metal nanoparticles within MOFs using a core–shell structure has been explored. Han et al. developed a Pt/Au@Pd@MOF-74 composite, where Au nanoparticles acted as the core for Pd shells (Au@Pd), which were subsequently encapsulated in MOF-74 for applications in CO₂ photoreduction and the reverse water gas shift reaction.¹⁸⁶ In this system, the low concentration of Au@Pd nanoparticles in Au@Pd@MOF-74 limits photon absorption at the Au@Pd active sites.¹⁸⁶ However, introducing Pt nanoparticles onto the surface reduces the composite's dependence on MOF-74 as an electron transfer medium. Furthermore, the higher amount of Pt nanoparticles compared to Au@Pd nanoparticles facilitates more efficient electron capture.¹⁸⁶ As a result, Pt/MOF-74 demonstrates the ability to convert CO₂ to CH₄, whereas Au@Pd@MOF-74 primarily produces CO.¹⁸⁶ The synergistic effect of combining Au, Pd, and Pt to form Pt/Au@Pd@MOF-74 enhances photon adsorption and improves CH₄ selectivity, showcasing the potential of multi-metallic systems for advanced photocatalytic applications.¹⁸⁶

The use of noble metals as co-catalysts in PEC systems poses challenges due to their high costs, limiting their large-scale applications. MoS₂, with its layered hexagonal structure, has emerged as a promising, cost-effective alternative for the HER because of its high abundance compared to Pt. The S atoms at the Mo-edge sites can absorb H₂, with each alternate S atom binding to a single H atom.¹⁸⁷ Additionally, MoS₂ has been shown to serve as active sites for H₂ evolution, with the edges of the nanosized MoS₂ promoting proton reduction.¹⁸⁸ Shen and colleagues demonstrated that depositing MoS₂ onto UiO-66/CdS photocatalysts resulted in a remarkable H₂ production rate of 650 μmol h⁻¹, which is 60 times higher than that of pure CdS.¹⁷⁰ Remarkably, the MoS₂/UiO-66/CdS composite also outperformed the Pt/UiO-66/CdS system under similar conditions (Fig. 7e) in terms of the H₂ evolution rate.¹⁷⁰ The strong interfacial contact between CdS, UiO-66, and MoS₂ facilitated efficient electron transfer, enhancing proton reduction at the MoS₂ nanosheet edges and significantly improving photocatalytic stability and performance.

4.4.3 Heterojunction construction. Coupling two semiconductors with matched band alignment is a proven strategy to reduce photocarrier recombination.^15,189–192 This method relies on constructing potential gradients based on the relative energy band alignment between semiconductors, facilitating efficient charge carrier transfer and separation along the gradient. This approach extends to MOFs, where interfacial effects arise when MOFs are combined with semiconductors or metals.¹⁹³ The exceptional properties of MOF-semiconductor heterostructures have been highlighted in several recent studies.^{141,194–197} The advantages of these heterostructures can be summarised as follows: (i) more efficient light utilisation can be achieved with visible-light-active MOFs; (ii) the porous structure of MOFs provides additional charge transport pathways, allowing photocarriers to be easily transported to the semiconductor;^170,198 (iii) highly dispersed active sites for redox reactions can form due to the high surface area of MOFs; (iv) MOFs can act as protective layers for semiconductors, enhancing the stability of the photoelectrode.^140,141 Beyond enhancing the individual functionalities of MOFs and semiconductors synergistic interactions in MOF-semiconductor heterostructures can also give rise to new and unique functionalities. For instance, Cui et al. synthesized a novel type-II NiFe-MOF/TiO₂ photoanode using a sacrificial template method.¹⁷¹ This photoanode achieved a photocurrent density of 0.77 mA cm⁻² at 1.23 V vs. RHE, which is 3.35 times higher than that of pristine TiO₂.¹⁷¹

Additionally, the IPCE (incident photon-to-current efficiency) at 390 nm increased to 42.10%, with charge injection and separation efficiencies reaching 92.8% and 37.4%, respectively.¹⁷¹ These enhancements are attributed to the NiFe-MOF structure, which shortens the photogenerated hole diffusion length and introduces additional active sites.¹⁹⁹ Theoretical analysis indicates that the CB of NiFe-MOF is more negative than that of TiO₂, allowing electron transfer from the CB of NiFe-MOF to TiO₂.¹⁷¹ Simultaneously, holes in the VB of TiO₂ migrate to the VB of NiFe-MOF, promoting efficient charge separation. As shown in Fig. 7f, Ni ions in NiFe-MOF are oxidised to Ni³⁺ and N⁴⁺via hole capture. The high-valence Ni species oxidise OH⁻/H₂O into O₂, while Ni⁴⁺ is reduced to Ni²⁺, preventing charge carrier recombination on the photoanode surface.¹⁷¹ Similarly, Li et al. developed a Z-scheme composite of Fe-MOF-derived Fe₂O₃ coupled with Ag–ZnO, which exhibited superior photocatalytic performance compared to its single-component counterparts.²⁰⁰ This improvement is attributed to the porous structure of Fe-MOF, which provides a large surface area and abundant active sites, as well as the Z-scheme structure that facilitates efficient separation of photogenerated electron–hole pairs.

A common limitation of most MOFs is their weak absorption of visible light, which reduces their photocatalytic efficiency. Integrating visible-light-responsive semiconductors with MOFs can effectively address this challenge. For example, g-C₃N₄, a visible-light-active semiconductor, is often used to modify MOFs to enhance both light absorption and charge separation efficiency. Zhao et al. developed a method for synthesizing dimension-matched heterojunctions containing MOFs, achieving excellent charge separation for efficient photocatalysis.²⁰¹ The strong photoactivity of the Ni(II) MOF/g-C₃N₄ composite is due to the excellent charge transport properties of ultrathin g-C₃N₄, enhanced charge separation driven by the transfer of excited high-level electrons from g-C₃N₄ to Ni-MOF, and the catalytic functionality of the central Ni ion, which promotes CO₂ activation.²⁰¹ Doping g-C₃N₄/MOF composites with Na further improves their performance by introducing tunable light-harvesting capabilities and optimised band structures.²⁰² For example, in Na-doped g-C₃N₄/Pt@UiO-66, Na doping enhances photocatalytic performance by creating electron transfer sites based on electron traps and a thermodynamic driving force. Moreover, MOFs have been modified with 2D carbon nitride nanosheets (CNNSs), which exhibit good conductivity, enabling rapid photocarrier transfer. Ye et al. constructed a UiO-66/CNNSs heterostructure photocatalyst using electrostatic self-assembly, leveraging the large surface area and strong CO₂ capture ability of UiO-66.²⁰³ This heterostructure demonstrated significantly higher CO₂ reduction activity compared to pristine CNNS.²⁰³ Safarifard and colleagues employed a facile in situ ultrasound synthesis to create a TMU-49/CNNS composite, which exhibited remarkable photocatalytic activity under visible light.²⁰⁴ The composite's lamellar morphology enhanced catalytic behaviour by providing greater access to internal active sites, as demonstrated in benzyl alcohol oxidation via a Z-scheme mechanism.²⁰⁴

Although MOFs hold great promise for PEC water splitting, their efficiency is often constrained by poor charge transport and separation.^130,205 Furthermore, many MOFs suffer from limited stability in aqueous solutions, leading to component leaching and rendering them unsuitable for PEC applications. However, MOFs such as UiO-66, UiO-67, MIL-101, MIL-53, MIL-125, ZIF-8, and SIM-1 have demonstrated operational stability for up to 12 h.^206,207 This stability is attributed to the high-valent metal ions (Zr⁴⁺, Cr³⁺, Ti⁴⁺, Fe³⁺) in these MOFs, which provide high charge density and form strong coordination bonds, contributing to their structural integrity and stability.²⁰⁸ To overcome performance limitations, researchers have employed strategies such as tailoring bandgaps, incorporating co-catalysts, optimising morphology, and forming heterojunctions. These approaches have significantly improved the light absorption, catalytic activity, and charge transfer properties of MOFs. For example, integrating MOFs as additives or interlayers in photoelectrodes has been shown to enhance charge transport and electrical conductivity.^{157,209–212} These advancements in MOF modifications highlight the potential of MOF-perovskite composites as next-generation photoelectrodes for solar-driven water splitting, which will be reviewed and discussed further in the next section.

5 MOF-perovskite hybrids

Perovskites have garnered significant attention in recent years as a cost-effective and efficient alternative to silicon in solar cells. However, their application in PEC water splitting remains relatively underexplored. On the other hand, MOFs are widely utilised across various fields, including gas storage, catalysis and drug delivery, due to their tuneable porosity, high surface area and structural versatility.^123–126 These unique properties make MOFs highly promising for PEC water splitting, as they can enhance light absorption, facilitate charge transfer, and provide abundant active sites for efficient H₂ production, particularly when integrated with perovskites.²¹³ A significant milestone in MOF-perovskite integration was achieved in 2014 by Vinogradov et al., who synthesized a perovskite/TiO₂-MIL-125 heterojunction for a perovskite solar cell (PSC).²¹⁴ This development opened new avenues for leveraging MOFs in perovskite-based systems. This section will examine the advantages of integrating MOFs with perovskites and their current applications in PEC water splitting. Additionally, related applications such as PSC devices will be discussed, with an emphasis on key properties that are beneficial for PEC water splitting. Various architectures for integrating MOFs into perovskite-based PEC cells will also be explored, highlighting their potential to advance solar-driven water splitting technologies.

While significant progress has been made in recent years, it is crucial to acknowledge that the literature on MOF-perovskite composites for PEC water splitting remains relatively limited. This presents a significant opportunity for further exploration and innovation in this field, particularly so given the known symbiotic benefits of this hybrid composite that are discussed in this section.

5.1 Key properties of MOF-perovskite hybrids

MOF-perovskite hybrids offer a unique combination of complementary properties that synergistically enhance their performance in PEC water splitting. These hybrids demonstrate notable improvements in light absorption and charge carrier transport, leading to higher catalytic activity.^128,215 The interface between MOFs and perovskites plays a crucial role by facilitating efficient charge transfer and minimising recombination losses.^216–218 The structural and functional versatility of MOFs allows for precise tailoring of their properties, enabling the design of hybrids optimised for PEC performance. Moreover, MOFs can act as protective layers around perovskites, significantly enhancing their stability and durability under operational conditions.^216,219 This protective function can also address environmental concerns associated with lead-based perovskites by encapsulating them in non-toxic, eco-friendly MOF matrices.¹²⁸ These unique attributes position MOF-perovskite hybrids as highly promising candidates for the development of advanced and next-generation photoelectrodes in solar-driven water splitting systems.

5.1.1 Improved stability. As previously mentioned, the unique capability to chemically and structurally tune MOFs offers a significant advantage in adjusting internal pore sizes. This flexibility, which surpasses that of other porous materials like zeolites or SiO₂, makes MOFs highly advantageous in terms of controllable porosity. This tuneable porosity renders MOFs particularly suitable as encapsulating materials to protect the inherently poor stability of perovskites. For instance, Mollick et al. successfully grew MAPbBr₃ crystals (6–8 nm in size) within the pores of ZIF-8, as confirmed through high-angle annular dark-field scanning electron microscopy (HAADF-STEM) imaging.²²⁰ Precise size and shape matching between the guest and the host MOF cavity is critical for preserving the inherent lattice parameters of the perovskites without undergoing any structure perturbations of both host MOF frameworks and guest crystals. This is evident in larger-sized MAPbBr₃ crystals occupying the small cavities of ZIF-8, which leads to defects such as missing ligands in the MOF framework. Nevertheless, the MaPbBr₃@ZIF-8 composite demonstrated exceptional long-term environmental stability, maintaining consistent luminescence intensity over several months. It also exhibited stable photoluminescence (PL) intensity at elevated temperatures for 20 days, a stark contrast to bare perovskite, which degraded within just 5 hours. The remarkable stability of MAPbBr₃@ZIF-8 can be attributed to the hydrophobic MOF walls, which function as protecting layers for the perovskite nanoparticles. Similarly, studies have utilised PCN-221(Fe) to encapsulate MAPbI₃, safeguarding it in aqueous environments for photocatalytic carbon dioxide reduction.²²¹ Dong et al. addressed the issue of Pb leakage in PSCs by employing MOF@covalent organic frameworks (COFs), supported by a detailed investigation using density functional theory (DFT) calculations.²²² These calculations provided critical insights into the mechanisms on how the MOF@COF structure inhibited lead leakage. Specifically, the study revealed that the C–N, –COO⁻, and C=O active groups in MOF@COFs chemically fix uncoordinated Pb²⁺ ions in situ, thereby enhancing the internal stability of the perovskite film by modifying its local chemical environment.²²² Furthermore, even if perovskite degradation occurs, the biomimetic nanoparticle structure of MOF@COF functions as an adsorbent, trapping Pb²⁺ ions via in situ chemical adsorption, similar to a spider web.²²² This synergistic effect of chemical fixation and adsorption not only mitigates lead leakage but also underscores the potential of MOF@COF hybrids to optimise the electronic structure of perovskites and improve stability, as elucidated through DFT modelling and experimental approaches. MOFs have been utilised to form heterojunctions with perovskites in PSCs and as scaffolds for the growth of perovskite layers. In 2014, Vinogradov et al. first demonstrated that incorporating MOFs into PSCs improved the crystallinity of perovskite films.²¹⁴ The primary motivation for integrating MOFs into PSCs is to produce high-quality perovskite films and enhance the overall stability of PSCs. For instance, Chang et al. incorporated a Zr-based porphyrin MOF (MOF-525) as an additive in the perovskite precursor solution to fabricate a MOF-525/CH₃NH₃PbI_3-xCl_x heterojunction, resulting in a perovskite material with enhanced crystallinity.²²³ Since the crystallisation of perovskite begins at the top of the MOF/perovskite thin film, the ordered scaffold of the microporous MOF provides a regular arrangement of perovskite crystallites during the initial crystallisation stage. However, the device's stability against moisture did not improve due to the use of Spiro-OMeTAD as the hole transport layer. Further discussion of this hole transport layer and potential MOF-based modifications is provided in Section 5.3.1.2. Nevertheless, an average PCE of 12% was achieved with the MOF-perovskite hybrid layer, highlighting its potential for performance enhancement. Chang et al. further suggested that the pore size of MOFs plays a critical role in improving device performance.²²³ Incorporating UiO-67 into the perovskite layer resulted in improved performance compared to UiO-66, primarily due to the large pore size of UiO-67, which facilitated more efficient diffusion of the perovskite precursor solution. A similar study by Lee et al. involved blending a Zr-based MOF into the perovskite layer of a PSC.¹³³ This integration created a “grain-locking” effect, where Zr atoms were distributed along the grain boundaries of the perovskite in the hybrid film, reinforcing the robustness of the perovskite layer.¹³³ PL spectra confirmed the impact of MOFs on the perovskite layer (Fig. 8a), with hybrid films exhibiting higher PL intensity than pristine perovskite films.¹³³ This increase indicates defect passivation induced by the hybrid MOFs.¹³³ Moreover, the coordination between the lone pairs of oxygen in the MOF and Pb²⁺ ions in the perovskite provided additional protection against defects and moisture, enhancing the stability of the hybrid film. Both MOF-perovskite hybrids delivered higher PCE values compared to using singular UiO-66 and MOF-808 as interlayers, thereby underscoring their significant potential in PEC water splitting applications.

Fig. 8 (a) SEM-EDS images (red is I and green is Zr) and the PL spectra of MOF-808 and UiO-66 hybrid perovskite film. Adapted with permission from ref. 133. Copyright (2019), Wiley-VCH; (b) possible photocatalytic mechanism of the LaFeO₃/MIL-125-NH₂ composite. Adapted with permission from ref. 224. Copyright (2023), with permission from Elsevier; (c) schematic illustration of Co₃O₄/NaTaO₃ composite photocatalysts with H₂ production under simulated sunlight irradiation and production rates of H₂ of pure NaTaO₃ and Co₃O₄/NaTaO₃ composite photocatalysts under simulated sunlight. Adapted with permission from ref. 225. Copyright (2019), Elsevier; (d) the charge density of Cs₃Cu₂I₅@ZIF-9-III with an isosurface value of 0.25 e Å⁻³. Adapted with permission from ref. 226. Copyright (2024), Elsevier; (e) diagram of the light-harvesting and carrier separation mechanism in the MOF-derived TiO₂(N–C)/CPB/TiO₂ NR photoanode. Adapted with permission from ref. 227. Copyright (2021), Elsevier; (f) schematic synthesis process of CPB@Co₃O₄/N–C. Adapted with permission from ref. 228. Copyright (2020), Elsevier.

5.1.2 Enhanced charge separation. Coupling perovskites with MOFs not only enhances the stability of perovskite materials but also significantly improves photocatalytic performance by promoting greater separation of photo-excited charge carriers. For instance, BiFeO₃ was coupled with a Cu/Ni-based MOF to serve as a recoverable catalyst for alcohol oxidation.²²⁹ The high surface area of the Cu/Ni-MOF synergised with BiFeO₃, effectively suppressing electron–hole recombination and significantly enhancing the reaction efficiency.²²⁹ Similarly, BaTiO₃ was incorporated into ZIF-8 for the degradation of methylene blue and Congo Red dyes under solar irradiation.²³⁰ The BaTiO₃@ZIF-8 nanocomposite exhibited exceptional catalytic performance through a unique charge transfer mechanism between the semiconductor and the MOF. Another example involves the formation of a Z-scheme heterojunction when LaFeO₃ is combined with NH₂-MIL-125 (Fig. 8b). This hybrid structure exhibited lower PL intensity compared to LaFeO₃ and NH₂-MIL-125 individually, indicating a longer charge carrier lifetime. This reduction in charge recombination resulted in higher photocatalytic activity²²⁴ Such synergistic effects are also evident when components are integrated via encapsulation. For instance, LaMnO₃ encapsulated within MOF-5 was used to hydrogenate 4-nitrophenol. In this case, MOF-5 not only provided a scaffold for the catalytic species but also enhanced the overall efficiency of the process.²³¹ Moreover, the encapsulation of LaMnO₃ into MOF-5 introduced numerous crystal defects in the MOF, which served as active sites.²³² Consequently, the LaMnO₃@MOF-5 composite exhibited significantly higher catalytic activity for reducing 4-NP compared to either LaMnO₃ or MOF-5 alone.

Beyond the direct utilisation of MOFs, researchers have developed heterostructures combining MOF-derived metal oxides with other semiconductors to enhance water splitting activity. Metal oxides derived from MOFs are particularly advantageous because they retain the porous structure and large specific surface area of the parent MOF. For example, Xu et al. constructed Co₃O₄ from Co-MOF through thermal annealing and attached it to the surface of NaTaO₃ through electrostatic attraction.²²⁵ The Co₃O₄ nanoparticles exhibited light absorption across a broad range of the light spectrum ranging from 200 to 700 nm, whereas NaTaO₃ alone showed no visible light activity. As a p-type semiconductor, the MOF-derived Co₃O₄ formed a p–n junction with the n-type NaTaO₃, resulting in effective carrier separation and an 8.2-fold enhancement in photocatalytic activity compared to bare NaTaO₃ (Fig. 8c).²²⁵ Similarly, Zhu and colleagues developed a BiFeO₃@Bi₅O₇I composite with an n–n heterojunction derived from a MOF for tetracycline degradation.²³³ The MOF derived composite exhibited a higher specific surface area than pure BiFeO₃ and Bi₅O₇I, providing more catalytic centres on the surface for more efficient degradation. Moreover, the n–n junction between BiFeO₃ and Bi₅O₇I improved charge transfer kinetics and reduced charge carriers' recombination through the internal electric field generated between the two compounds.

To further enhance our comprehension of the interaction between perovskites and the MOF hybrid, Sharma et al. performed DFT calculation on Cs₃Cu₂I₅@ZIF-9-III hybrids and water concerning the pathway for the HER.²²⁶ The calculations revealed that water molecules preferentially bind to Cs sites on the Cs₃Cu₂I₅ surface, with adsorption energies indicating favourable interactions. Under electron-rich conditions, the dissociation of water becomes energetically feasible, facilitating the HER.²²⁶ This process involves electron trapping at Cu sites, leading to structural rearrangements that promote water dissociation and H₂ generation. Fig. 8d provides evidence of type II heterojunction formation between Cs₃Cu₂I₅ and ZIF-9-III. Charge density analysis reveals electron accumulation near the phenyl rings of ZIF-9-III, while a concomitant decrease in electron density is observed around the iodine atoms of Cs₃Cu₂I₅. This suggests potential electron transfer from the conduction band of the perovskite to the ZIF-9-III, facilitated by a weak interfacial interaction. This charge transfer mechanism is consistent with the formation of a type II heterojunction, effectively separating photogenerated electron–hole pairs and enhancing charge transport efficiency.²²⁶ Electrons excited under ultraviolet light transfer from the conduction band (CB) of Cs₃Cu₂I₅ to ZIF-9-III, while holes move in the reverse direction, enabling photocatalytic reactions.²²⁶ These findings underscore the synergistic interactions within Cs₃Cu₂I₅/ZIF-9-III composites, offering critical insights into their potential for sustainable energy applications.

5.2 Recent advancement of MOF-perovskite composites in PEC water splitting

Perovskites have shown tremendous potential in PEC water splitting, and their integration with MOFs represents a promising strategy to further enhance performance. However, the application of MOF-perovskite composites in PEC water splitting remains relatively underexplored, presenting significant opportunities for further research and innovation. Leveraging the broad light absorption range of inorganic perovskite materials, Tang et al. improved the performance of TiO₂ through interface engineering (Fig. 8e).²²⁷ Specifically, they developed a MOF-derived TiO₂/CsPbBr₃ (CPB)/TiO₂ composite, where the optical response range of the MOF-derived TiO₂ (N–C)/CPB/TiO₂ photoanode was extended from approximately 400 nm to 530 nm.²²⁷ This extension in light absorption, combined with enhanced charge separation efficiency, led to a substantial increase in photocurrent density, reaching 4.5 mA cm⁻² at 1.2 V vs. RHE.²²⁷ The bonding interaction between CPB and the MOF-derived N–C framework further enhanced system stability by anchoring CPB nanoparticles to the TiO₂ structure.²²⁷ In addition to coating perovskites with MOF-derived semiconductors to improve stability, Tang and colleagues utilised MOF-derived Co₃O₄ to encapsulate CPB (Fig. 8f), achieving a significantly boosted photocurrent density of 2.50 mA cm⁻² and carrier migration of 41.40%—an improvement of 9.61 and 12.24 times, respectively, compared to pristine CPB.²²⁸ The exceptional PEC performance and stability of the CPB@Co₃O₄/N–C can be attributed to several factors: (1) the formation of N–Br halogen bonds at the interface between the Co₃O₄/N–C framework and CPB chemically immobilises the CPB, preventing its decomposition and significantly improving its stability in an aqueous environment. (2) The unique type-II heterojunction at the CPB/Co₃O₄/N–C interface enhances carrier separation performance. (3) The MOF-derived Co₃O₄/N–C serves as a co-catalyst, significantly improving the surface water oxidation kinetics of pristine CPB.²²⁸ These combined effects collectively contribute to the superior PEC performance and stability of the CPB@Co₃O₄/N–C system. Sharma et al. recently developed a novel Cu-based MOF integrated with Ba_0.85Ca_0.15Zr_0.1Ti_0.9O₃ (BCZT) for PEC applications, demonstrating significant improvements in performance.²³⁴ Specifically, the photocurrent density of the Cu-MOF/BCZT composite increased fourfold, from 20 μA cm⁻² to 80 μA cm⁻², accompanied by a reduction in the bandgap from 2.8 eV to 2.2 eV.²³⁴ This enhancement in photoelectrochemical properties can be attributed to several factors: increased surface charge availability, faster separation and higher transfer rates of photogenerated charges, and improved optical absorption in the visible region. These improvements arise from the highly conductive Cu-MOF at the interface, which plays a crucial role in boosting overall PEC performance.²³⁴ In a recent study, Li and colleagues developed a plasmonic perovskite oxide-based Z-scheme catalyst, TiO₂@NH₂-MIL-125@reduced-SrTiO₃ (R-STO), through in situ derivation of NH₂-MIL-125 followed by oxygen-vacancy engineering.²³⁵ When applied as a photoanode for PEC water splitting, this catalyst achieved a remarkable photocurrent density of 4.4 mA cm⁻² at 1.23 V vs. RHE—three times higher than that of pure TiO₂, and achieved a H₂ generation rate of 58.5 μmol h⁻¹ cm⁻².²³⁵ These enhancements are attributed to the synergistic effects of the MOF-based Z-scheme architecture and the surface plasmon resonance (SPR) properties of the non-noble R-STO.²³⁵

To summarise, Tang et al. achieved a high photocurrent density of 2.5 mA cm⁻² by leveraging a type-II heterojunction for efficient charge separation and utilising halogen bonding for stability in CPB@Co₃O₄/N–C.²²⁸ Sharma et al. demonstrated a fourfold increase in photocurrent density by integrating a highly conductive Cu-MOF with BCZT, thereby enhancing optical absorption and charge transfer.²³⁴ Li et al. developed a Z-scheme catalyst with oxygen-vacancy engineering and utilised the SPR properties of R-STO, achieving a remarkable photocurrent density of 4.4 mA cm⁻² and a high H₂ evolution rate.²³⁵ These studies collectively showcase the versatility of MOF-perovskite hybrids, with each approach offering unique advantages depending on the specific design and optimisation strategies employed.

Table 3 below summarises the MOF-perovskite composites for PEC water splitting.

Table 3 Summary of MOF-perovskite composites for PEC water splitting

Composite	Role/application	Synthesis	Performance	Stability	Ref.
[MAPbBr3@MA-M(HCOO)3] (M = Mn and Co)	OER	Single-step, solid-state mechanochemical grinding process	N/A	The MOF template also imparts extra stability to the MAPbBr3 which is evident in the superior solvent stability over 15 days	236
				No structural changes after 10 days of immersion in methanol
MOF-derived TiO2(N–C)/CsPbBr3/TiO2	OER + glycerol oxidation	Solvothermal method	4.50 mA cm^−2 photocurrent density at 1.2 V vs. RHE	Stable photocurrent density throughout 0.5 h of operation	227
CsPbBr3@MOF-derived Co3O4/N-doped C	OER	Co-precipitation-calcination method	1.36 mA cm^−2 photocurrent density at 1.23 V vs. RHE	Steady-state behaviour throughout 24 h of operation	228
TiO2@NH2-MIL-125@R-STO	OER	Solvothermal method	4.4 mA cm^−2 photocurrent density at 1.23 eV vs. RHE + H2 generation rate of 58.5 μmol h^−1 cm^−2	Steady-state behaviour throughout 18 000 s of operation and no structural changes observed after 5 h of operation	235
Pt/Cs2Bi2I9@NH2-UiO-66	HER	Ligand-assisted reprecipitation (LARP) method	H2 production of 141.87 μmol g^−1 h^−1	No structural changes observed after 5 h of operation	237
La0.6Sr0.4Co0.8Fe0.2O3@Ni3(HITP)2	OER	Electrospinning method	Low overpotential of 272 mV achieved at 10 mA cm^−2	Slight decay of 2.9% observed after 12 h of operation	238
Cu-MOF/Ba0.85Ca0.15Zr0.1Ti0.9O3	OER	Solvothermal method	80 μA cm^−2 photocurrent density at 1 V vs. Ag/AgCl with an IPCE of ∼7.5%	N/A	234

---

5.3 Architecture of perovskite-based cells with MOFs

While MOFs have shown significant advantages when combined with perovskites, the precise methods of integrating them into MOF-perovskite hybrids remain a topic of particular interest. Determining and optimising the architecture of this offers several potential benefits, including targeted modification of specific cell components for improved properties and performance, as well as addressing other pertinent issues relating to reducing manufacturing complexity. This subsection will explore various reported architecture designs in which MOFs have been incorporated into perovskite-based structures.

5.3.1 MOF/perovskite heterojunction. Analogous to PSCs, perovskite-based PEC systems for solar-assisted water splitting typically employ a layered structure comprising an electron transport layer (ETL) and a hole transport layer (HTL) that sandwich the perovskite layer. The differences in bandgap and Fermi levels between the narrow bandgap perovskites and the wider bandgap ETL or HTL lead to the formation of heterojunctions at their interfaces. These heterojunctions generate built-in electric fields that promote the separation of photogenerated charge carriers within the perovskite layer, efficiently directing electrons towards the ETL and holes towards the HTL. Favourable band alignment further facilitates the efficient transport of electrons and holes through the ETL and HTL to the catalyst layer, thereby improving photovoltage and reducing charge carrier recombination. The interfaces between the perovskite, ETL and HTL are critical for carrier transport and PEC performance, directly influencing key parameters such as photovoltage, photocurrent, and fill factor.
5.3.1.1 MOFs in/as the electron transport layer. As a result, interface engineering, which focuses on optimising the properties of these interfaces, is essential for improving charge carrier extraction and transport. The main functions of interface engineering include: (a) ensuring optimal energy band alignment and charge transport, (b) passivating defects and minimising ion migration, and (c) enhancing device stability by forming protective layers. Thus, MOFs or MOF-derived materials can be strategically introduced into the perovskite-based photoelectrode forming a layered structure akin to stacking two sheets of paper to improve the functional layers, as illustrated in Fig. 9a. In this arrangement, the MOF and perovskite layers are physically distinct, with one layer positioned on top of the other. While in close contact, they maintain their individual structural integrity, forming a heterojunction. This layered structure enables interactions at the interface between the materials, which can facilitate improved charge transport and enhance the overall performance of the system.

Fig. 9 (a) Diagram of the possible locations of MOFs in perovskite-based cells; (b) schematic diagram of the Co-doped TiO₂ ETL in PSCs and the Nyquist plot characteristic curves of films based on TiO₂ and 1 wt% Co-doped TiO. Adapted with permission from ref. 239. Copyright (2020), American Chemical Society; (c) Schematic diagrams of the formation of MOF-derived ZnO polyhedra and their use as ETLs to enhance light harvesting and electron extraction. Adapted with permission from ref. 240. Copyright (2020), Elsevier; (d) comparison of the SEM images of the MAPbI₃-based layer formed on the surface of mp-TiO₂ with different ZIF-8 coating durations: (i) 0, (ii) 10, (iii) 20, and (iv) 40 min. Adapted with permission from ref. 241. Copyright (2018), Royal Society of Chemistry; (e) UV stability of perovskite film under continuous 365 nm UV illumination with the total dosage of 100 mW cm⁻² for 120 min. Adapted with permission from ref. 242. Copyright (2024), Wiley.

The primary role of the ETL is to transport electron charge carriers efficiently, making it essential for the material used to construct the ETL to possess high carrier mobility, a large surface area, and minimal defects.²⁴ Non-toxic TiO₂, known for its fast charge injection rate and excellent thermal stability, is one of the most commonly used ETL materials.^243,244 Non-toxic TiO₂, known for its fast charge injection rate and excellent thermal stability, is one of the most commonly used ETL materials.^243,244 However, TiO₂ suffers from low electron mobility, which increases the likelihood of charge carrier recombination. Similarly, ZnO, which shares comparable properties with TiO₂, is another widely used ETL material. While ZnO offers higher charge mobility than TiO₂, its chemical instability and poor compatibility with perovskite films pose significant challenges. Chemical residues from the manufacturing process can further exacerbate this issue by accelerating the decomposition of perovskite materials.^240,243,245 Therefore, to achieve efficient and stable PSCs, modifications to TiO₂ and ZnO ETLs or the development of alternative ETL materials is imperative. Moreover, utilising a Ti-based MOF template resulted in a highly porous structure, both internally and on the surface, due to the thermal decomposition of the MOF template. This structural improvement contributed to an increase in PCE from 12.32% to 15.75%.²³⁹ The performance enhancement was brought about by improved electron transfer of Co doping, which reduced charge carrier recombination. This was further confirmed by a significant decrease in charge transport resistance and charge recombination, as shown in Fig. 9b.²³⁹ Additionally, the high temperature (>450 °C) required for synthesizing TiO₂ ETLs often limits their application in flexible perovskite solar cells. MOFs can overcome this limitation as demonstrated by Ryu et al.,²⁴⁶ where nanocrystalline formation of MIL-125(Ti) was employed at room temperature. This MOF-based ETL was successfully used in both rigid PSCs on glass substrates and flexible PSCs on polyethylene naphthalate (PEN) substrates, respectively.²⁴⁶ The electrical conductivity of Ti-MOF as an ETL was initially found to be lower than that of TiO₂, primarily due to microcracks in the Ti-MOF structure. However, this issue was significantly mitigated with the incorporation of [6,6]-phenyl-C61-butyric acid methyl ester (PCBM).²⁴⁶ As a result, flexible PSCs displayed exceptional mechanical stability, maintaining an average PCE of 16.57% after 700 bending cycles.²⁴⁶ This enhanced performance was made possible by the additional H bonds provided by the organic linkers in the MOF framework.¹²⁸ The small-sized nanocrystals of Ti-MOF were highly dispersed in organic solvents, which led to the simple fabrication of uniform and ultrathin films. Moreover, the electronic structure of Ti-MOF (CBM = −4.12 eV) is better suited for charge injection and transfer from the perovskite layer (CBM = −3.8 eV) to the ITO substrate compared to that of TiO₂ (CBM = −3.98 eV). To overcome the instability issues caused by chemical residues, Zhang et al.²⁴⁰ proposed the use of porous MOF-derived ZnO as an ETL. The porous structure of the MOF-derived ZnO increased the contact area between the ETL and the perovskite layer, thereby suppressing charge carrier recombination and enhancing carrier extraction efficiency.²⁴⁰ Additionally, the poriferous structure of MOF-derived ZnO improved light utilisation by acting as light scattering sites, as illustrated in Fig. 9c.²⁴⁰ Zhang et al. also reported a reduction in trap states and photoluminescence decay, indicating a significant inhibition of charge carrier recombination in the presence of MOF-derived ZnO.²⁴⁰

5.3.1.2 MOFs in/as the hole transport layer. In addition to facilitating charge transport, the addition of a HTL in an n–i–p device configuration prevents direct contact between the anode (Au, Ag, or Al) and the perovskite layer. This minimises the transfer of electrons to the anode, thereby suppressing the recombination of photogenerated charge carriers.²⁴ Common HTL materials include Spiro-OMeTAD, PEDOT: PSS, and CuI.²⁴³ However, despite being one of the best-performing HTLs, Spiro-OMeTAD raises concerns regarding its stability and its impact on the quality of the resulting perovskite layer.²⁴⁷ To remediate these issues, researchers have introduced multiple buffer layers to prevent the ingress of moisture and oxygen and have designed novel HTL materials.^248,249 For instance, Zhang et al. introduced an interfacial carbon layer as an HTL using a simple and cost-effective glucose-derived carbonization process. This approach effectively suppressed undesirable charge recombination, thereby reducing charge losses and enhancing overall performance.²⁵⁰ While effective, this approach increases the complexity of the fabrication process, which is non-ideal for commercialisation.²⁵¹ An alternative strategy is to directly incorporate MOFs into existing HTL materials. Li et al. demonstrated this approach by doping indium-based MOFs (In₂) into Spiro-OMeTAD to improve the light responsiveness of PSCs and provide a buffer to block unwanted material penetration into the perovskite layer.²⁵¹ The In₂ MOF demonstrated strong UV light absorption and visible light emission. Furthermore, In₂ with its cube-like structure acted as light scattering centres, extending the light path length and enhancing light utilisation.²⁵¹ By sealing the gaps in Spiro-OMeTAD with In₂, a denser HTL was formed, thus effectively preventing Au from the electrode and oxygen from penetrating the perovskite layer. Furthermore, the increased hydrophobicity imparted by the In₂ protected the PSC device against moisture ingress significantly enhancing its stability.

In addition to improving HTL stability, the incorporation of MOFs has been reported to improve the conductivity of HTLs. As Spiro-OMeTAD suffers from low conductivity, lithium bis(trifluoromethane) sulfonimide (Li-TFSI) and 4-tert-butylpyridine (t-BP) were often added as standard bi-dopants to enhance the performance of Spiro-OMeTAD.²⁵² The role of Li-TFSI is to act as an oxidising agent to enhance Spiro-OMeTAD conductivity, while t-BP is used to prevent charge carrier recombination. However, both dopants present significant drawbacks: Li-TFSI is hygroscopic and highly reactive to water, which results in the degradation of PSCs, while the low boiling point of t-BP causes it to volatilise during the fabrication process.²⁵³ Over time, t-BP forms salt aggregates on the surface, which reduces the mobility of charge carriers and accelerates the degradation of PSCs.²⁵⁴ MOFs have been proposed as alternative oxidisers to combat these drawbacks of Li-TFSI and t-BP. Wang et al. demonstrated the use of a three-dimensional (3D) MOF, {[Zn(Hcbob)]·(solvent)}_n (Zn-CBOB), as a dopant for Spiro-OMeTAD. In their previous studies, Zn-CBOB was found to oxidise Spiro-OMeTAD effectively in inert environments while simultaneously addressing the uncoordinated Pb²⁺ ions in the perovskite layer. The carbonyl groups in Zn-CBOB with Lewis basicity interact with these uncoordinated Pb²⁺ ions, passivating surface traps and improving charge carrier extraction.²⁵³ Additionally, Wang et al. explored the possibility of combining Li-TFSI with MOFs to enhance both the electron transfer and oxidation processes of Spiro-OMeTAD (eqn (9) and (10)).

Spiro-OMeTAD + Zn-CBOB ⇌ Spiro-OMeTADs˙⁺Zn-CBOB˙⁻ (9)

Spiro-OMeTADs˙⁺Zn-CBOB˙⁻ + Li⁺TFSI⁻ → Spiro-OMeTADs˙⁺Zn-CBOB˙⁻ + Li⁺Zn-CBOB˙⁻ (10)

Zn-CBOB not only facilitates the rapid oxidation of Spiro-OMeTAD but also enhances the overall hole mobility and conductivity of the PSC. The improved stability of the device was attributed to the hydrophobicity of Zn-CBOB, as revealed by SEM studies, which showed severe damage to the HTL in the absence of Zn-CBOB.²⁵³ Additionally, the porosity of MOFs contribute to uniform coverage of the HTL, preventing salt aggregation and blocking the penetration of undesired materials into the perovskite layer. Similarly, Zhou and co-workers developed an In(III)-based MOF (In-Pyia) as a substitute for t-BP in HTLs.²⁵⁵ The inherent porosity of the In-Pyia, along with its open pyridine nitrogen sites, allowed effective dispersion of Li-TFSI, resulting in a more uniform HTL film. Additionally, the open pyridine nitrogen atoms in In-Pyia coordinated with Li⁺ ions, creating a dense and crack-free HTL layer.²⁵⁵ In contrast, HTL films prepared with t-BP showed significant drawbacks, as the rapid volatilisation of the t-BP during fabrication led to the formation of numerous cracks on the film surface.²⁵⁵ The In-Pyia-modified HTL film also exhibited significantly reduced hygroscopicity, which is crucial for minimising hydration and permeation behaviours that could otherwise damage the perovskite layer.²⁵⁵ As a result, the In-Pyia modified HTL demonstrated significantly improved long-term stability, maintaining performance for 16 days in an ambient environment.²⁵⁵ Furthermore, the well-aligned energy levels and their positive impact on the oxidation of Spiro-OMeTAD make In-Pyia an excellent additive for HTLs, promoting more efficient hole extraction and conduction.²⁵⁵

5.3.1.3 MOFs as interlayers. MOFs have also been employed as interlayers in PSCs, where they enhance charge carrier mobility and to suppress charge recombination. Similar to blending MOFs into the perovskite layer, interlayering MOF beneath the perovskite layer can also improve the crystallinity of the perovskite film. For instance, Shen et al. observed significant improvement after introducing ZIF-8 as an interlayer between the perovskite and TiO₂, which serves as the ETL, as illustrated in Fig. 9d.²⁴¹ The introduction of the ZIF-8 interlayer eliminated pinholes in the perovskite films. Moreover, as the coating time of ZIF-8 increased, the grain size of the perovskite film grew while its surface roughness decreased. This suggests that the ZIF-8 interlayer not only preserved the morphology of the pristine film but also influenced the crystallisation dynamics, resulting in improved structural properties. The simultaneous increase in grain size and reduction in grain boundaries enhanced electron injection efficiency, reduced recombination rate, and ultimately improved device performance. While studies on MOFs as interlayers between the ETL and perovskite layer are more common, their use as interlayers between the HTL and perovskite layer is less prevalent. However, Lee's group provided an example of this application by utilising a MOF interlayer above an HTL, demonstrating its potential in further optimising PSC architectures. Specifically, UiO-66 and MOF-808 were specifically used to modify the surface of the NiO_x HTL, effectively regulating the crystallisation of the perovskite film. The full width at half maximum (FWHM) values of the perovskite films grown on UiO-66- and MOF-808-modified NiO_x films were 0.301° and 0.281°, respectively—both narrower than the 0.311° FWHM observed for films grown on unmodified NiO_x.¹³³ These results highlight the improved crystallinity of the perovskite films achieved through MOF modification. Additionally, the grain sizes of perovskite films grown on MOF-modified NiO_x films were significantly larger, increasing from approximately 480 nm for unmodified samples to 720 nm and 640 nm for UiO-66- and MOF-808-modified samples, respectively.¹³³ This demonstrates the beneficial role of the MOF scaffold in improving the crystallinity and grain size of the deposited perovskite film, which helps reduce defect density and grain boundaries, ultimately promoting charge transfer at the interface. Moreover, the perovskite films grown on MOF-modified layers exhibited a degree of PL quenching, which can be attributed to the 3D porous scaffold of the MOFs. This porous architecture facilitates the infiltration of perovskite precursors, leading to the formation of perovskite nanocrystals and improving interfacial compatibility. As a result, charge extraction efficiency is significantly enhanced, further contributing to improved device performance.^214,251

5.3.2 MOF-perovskite composite. In addition to using MOFs as interlayers, directly incorporating them into the perovskite layer by blending MOFs and perovskite components at the nanoscale within a single composite layer has also been demonstrated to improve perovskite crystallinity and overall device performance. Chang et al. demonstrated this by blending Zr-based MOF-525 into the perovskite precursor solution.²²³ The MOF-525 acted as a scaffold, facilitating the ordered arrangement of crystallites and improving overall device performance.²²³ However, the use of Spiro-OMeTAD as the HTL limited device stability due to its poor mechanical toughness.²²³ Similarly, Lee et al. observed a similar grain-locking effect in Zr-based MOF (UiO-66 or MOF-808)-perovskite composites, where Zr atoms reinforced the mechanical stability and robustness of the perovskite layer.¹³³ The coordination between oxygen atoms in the MOFs and Pb²⁺ ions in the perovskites provided protection against defects and moisture, enhancing the durability of the composite film.¹³³ Both MOFs used to form the MOF-perovskite hybrid film in this study demonstrated superior performance compared to when the MOFs were employed as interlayers.¹³³ Wu et al. conducted an in-depth study on the crystallisation kinetics of perovskites enhanced by the addition of yttrium (Y)-MOF to the perovskite layer in PSCs, combining advanced in situ characterization techniques and theoretical insights.²⁴²In situ UV-vis spectroscopy revealed that Y-MOF significantly slowed down the crystallisation process of perovskites, promoting the formation of larger grains, which are critical for improved charge transport and reduced defects.²⁴² Grazing incidence wide-angle X-ray scattering (GIWAXS) analysis provided further understanding of the microscopic mechanisms, indicating that Y-MOF crystals primarily accumulate at the bottom of the perovskite film.²⁴² This strategic positioning enhances stability by shielding the active layer from ultraviolet light, thereby mitigating UV-induced degradation, as shown in Fig. 9e where the UV absorption of the pristine sample noticeably weakened after 40 min while the Y-MOF-assisted perovskite maintained stable absorption characteristics even after 120 min of continuous light exposure. The Y-MOF-assisted PSCs achieved an impressive power conversion efficiency (PCE) of 24.05%, alongside significantly enhanced stability.²⁴² Notably, unencapsulated cells retained 84% of their initial PCE after exposure to UV light for 300 h.²⁴² This work underscores the potential of MOF frameworks to enhance both the performance and durability of PSCs. This phenomenon can be attributed to several factors: (1) the highly ordered porous structure of Y-MOF, which introduces micropores and mesopores within the PbI₂ films, thereby enhancing the porosity of the perovskite film; (2) the ordered channels and pores of the MOF, which guide the arrangement of PbI₂ grains around them, resulting in more uniform and well-distributed pore formation within the PbI₂ film; and (3) the influence of Y-MOF channels on interactions during grain growth, leading to a highly porous structure. These findings highlight the potential of MOF-perovskite hybrid films for photoelectrochemical water splitting applications, showcasing their ability to achieve improved efficiency and stability and paving the way for future advancements in this field. While the addition of Zr-based MOFs has been shown to improve the crystallinity of perovskite films, their use is limited by inflexible synthesis conditions, which can complicate the overall manufacturing process.^256–258 In contrast, the In₂ reported by Li et al. as an HTL modifier offers more flexible synthesis conditions.²⁵¹ When In₂ is blended into the perovskite precursor solution, it has been found to prevent perovskite degradation by converting unstable Pb²⁺ into a more stable Pb species.²⁵⁹ Recently, Zhou et al. fabricated a MOF-perovskite hybrid using [In₁₂O(OH)₁₆(H₂O)₅(btc)₆]_n (In-BTC), which was synthesized under mild conditions.²⁶⁰ The polar sites of In-BTC modulated the crystallisation of the perovskite, reducing grain defects and further enhancing the crystallinity of the film.²⁶⁰ Additionally, In-BTC nanocrystals exhibited strong absorption in the UV region, suggesting that high-energy photons, known to degrade perovskite film stability, could be effectively filtered by introducing In-BTC as an additive.²⁶¹ Notably, the In-BTC-modified PSC displayed a significantly enhanced photoresponse in the 300 to 550 nm range compared to the pristine device. This improvement is attributed to the Förster resonance energy transfer (FRET) between the In-BTC and the perovskite layer.²⁶⁰

6 Conclusion and future outlook

Perovskites have garnered significant attention for their exceptional optoelectronic properties, as evidenced by their record-breaking efficiencies in solar cells and LEDs. This remarkable performance has spurred interest in their application for PEC water splitting. However, conventional perovskite oxides are hindered by limited light absorption and poor stability, posing significant challenges for their practical use in PEC devices. To address these challenges, this review explores strategies to enhance PEC performance, focusing on the integration of perovskites with MOFs to form next-generation hybrid materials. Unlike perovskites, MOFs offer unique structural and functional advantages, such as tuneable porosity, high surface area, and a wide range of metal centres. By combining the strengths of these two materials, MOF-perovskite hybrids have demonstrated vast potential to overcome the individual shortcomings of their components. MOFs can enhance the stability of perovskites by shielding them from degradation while also providing additional active sites for catalytic reactions. Furthermore, the porous structure of MOFs facilitates efficient mass transport of reactants and products, thereby improving the overall system efficiency.

Despite promising advancements, the field of MOF-perovskite hybrids for PEC water splitting remains relatively nascent. To date, integrating MOFs with perovskites has been shown to significantly enhance the stability and mechanical properties of perovskite-based materials. The inherent porosity of certain MOFs effectively prevents moisture infiltration, further improving perovskite stability. While the specific MOF structures employed may vary, their integration with perovskites consistently improves photoelectrode performance by enhancing perovskite film quality and crystallinity, facilitating efficient charge transfer, suppressing charge recombination, and boosting device stability. The synergy between perovskites, with their excellent light-harvesting capabilities and high charge carrier mobility, and MOFs, renowned for their tuneable porosity, large surface area, and catalytic versatility, provides a strong platform for developing next-generation photoelectrodes. These combined attributes position MOF-perovskite hybrids as highly promising candidates for high-efficiency and long-term stable PEC water-splitting systems.

The future of MOF-perovskite hybrid materials for PEC water splitting holds immense promise, but significant challenges remain. To fully unlock their potential, future research directions should prioritize the following.

6.1 Enhancing fundamental understanding

• Develop a robust theoretical framework: a comprehensive understanding of the optoelectronic properties of these hybrid materials is crucial. This necessitates the development of robust theoretical frameworks, including advanced computational models and simulations, to predict and understand their electronic structures, band alignments, charge carrier dynamics, and interfacial interactions.

• Elucidate synergistic mechanisms: in-depth investigations with advanced technologies such as in situ characterization techniques are needed to elucidate the fundamental mechanisms underlying the synergistic interactions between MOFs and perovskites. This includes understanding how MOFs stabilize perovskites, influence charge carrier transport, and facilitate interfacial charge transfer.

6.2 Optimising material design and performance

• Fine-tune MOF properties: balancing MOF properties is crucial. While MOFs can enhance perovskite stability by providing a protective environment, their hydrophobic nature and narrow pore apertures can hinder mass transport. Future research should focus on fine-tuning MOF frameworks to achieve an optimal balance between perovskite protection and efficient reactant diffusion.

• Explore novel hybrid architectures: drawing inspiration from successful designs in perovskite solar cells (PSCs), researchers can explore novel hybrid architectures, such as 3D hierarchical structures and the integration of co-catalysts within the MOF framework.

• Leverage AI-powered materials discovery: the emergence of artificial intelligence (AI) and machine learning offers exciting opportunities for accelerating materials discovery. AI-powered algorithms can be utilised to predict and design optimal MOF-perovskite combinations with enhanced performance.

6.3 Technological integration

• Develop scalable synthesis methods: unlike MOFs, the commercial use of perovskites is still restricted due to low STH efficiency and poor material stability. It is evident that solely focusing on the performance of the material itself is insufficient. As research progresses, we should consider the preparation process, equipment development, and the full life cycle costs associated with industrialization and commercial applications. Furthermore, current research has primarily focused on material composition and structure modifications, with a notable absence of innovative preparation methods.

• Real-world operation: most of the reported perovskites have only been tested at a laboratory scale, and the potential challenges and implications of scaling up these materials for commercial application remain largely unexplored. Therefore, the large-scale preparation processes and operation that extend beyond laboratory conditions warrant further investigation.

Ultimately, the successful integration of MOFs and perovskites holds the potential to revolutionise PEC water-splitting technology, providing a sustainable and efficient solution for green H₂ production. With ongoing advancements in this field, MOF-perovskite hybrids could lead to the development of high-efficiency PEC devices, making a significant contribution to the global renewable energy landscape.

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.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

We acknowledge financial support from the Fundamental Research Grant Scheme (FRGS) (Project Reference Code: FRGS/1/2020/STG05/MUSM/02/1) provided by the Ministry of Higher Education (MOHE), Malaysia, which supported the PhD stipend for Quan Yee Tey.

Notes and references

1. W. Wang, M. O. Tadé and Z. Shao, Chem. Soc. Rev., 2015, 44, 5371–5408.
2. M. Ni, M. K. H. Leung, D. Y. C. Leung and K. Sumathy, Renewable Sustainable Energy Rev., 2007, 11, 401–425.
3. S. Satyapal, J. Petrovic, C. Read, G. Thomas and G. Ordaz, Catal. Today, 2007, 120, 246–256.
4. W. C. Ng, C. S. Yaw, S. N. A. Shaffee, N. A. A. Samad, Z. K. Koi and M. N. Chong, Sustainable Mater. Technol., 2024, 40, e00972.
5. M. Aravindan, V. M. Kumar, V. S. Hariharan, T. Narahari, A. P. Kumar, K. Madhesh, G. P. Kumar and R. Prabakaran, Renewable Sustainable Energy Rev., 2023, 188, 113791.
6. S. Lavate, H. P. S. Yalamati and R. Srivastava, Solar-Driven Green Hydrogen Generation and Storage, 2023, pp. 507–524.
7. J. Moore, J. Durham, A. Eijk, E. Karakas, R. Kurz, J. Lesak, M. McBain, P. McCalley, L. Moroz, Z. Mohamed, B. Pettinato, G. Phillippi, H. Watanabe and B. Williams, Machinery and Energy Systems for the Hydrogen Economy, 2022, pp. 333–424.
8. P. Sun, B. Young, A. Elgowainy, Z. Lu, M. Wang, B. Morelli and T. Hawkins, Environ. Sci. Technol., 2019, 53, 7103–7113.
9. M. Katebah, M. Al-Rawashdeh and P. Linke, Clean. Eng. Technol., 2022, 10, 100552.
10. Z. Wang, Y. Gu and L. Wang, Front. Energy, 2021, 15, 596–599.
11. P. Zhai, S. Haussener, J. Ager, R. Sathre, K. Walczak, J. Greenblatt and T. McKone, Energy Environ. Sci., 2013, 6, 2380–2389.
12. Y. Chen, Y. Liu, F. Wang, X. Guan and L. Guo, J. Energy Chem., 2021, 61, 469–488.
13. S. Balachandran, N. Prakash, K. Thirumalai, M. Muruganandham, M. Sillanpää and M. Swaminathan, Ind. Eng. Chem. Res., 2014, 53, 8346–8356.
14. J. Wang, T. van Ree, Y. Wu, P. Zhang and L. Gao, Metal Oxide Semiconductors for Solar Water Splitting, Elsevier Inc., 2018, pp. 205–249.
15. C. S. Yaw, Q. Ruan, J. Tang, A. K. Soh and M. N. Chong, Chem. Eng. J., 2019, 364, 177–185.
16. D. Chen and Z. Liu, ChemSusChem, 2018, 11, 3438–3448.
17. W. Ni, D. Yan, S. Fu, Y. Wei, H. Yang, X. Zhang, F. Zhao, K. You and H. Luo, Chem. Eng. J., 2023, 465, 142905.
18. C. Zhong, T. Duan, D. Yan, S. Zhang, H. Yang, Y. Pei, J. Yan, K. You, Y. Li and H. Luo, Sep. Purif. Technol., 2024, 331, 125728.
19. NREL, Best Research Cell-Efficiencies, https://www.nrel.gov/pv/cell-efficiency.html.
20. Y.-F. Xu, X.-D. Wang, J.-F. Liao, B.-X. Chen, H.-Y. Chen, D.-B. Kuang, Y.-F. Xu, X.-D. Wang, J.-F. Liao, B.-X. Chen, H.-Y. D. Chen and B. Kuang, Adv. Mater. Interfaces, 2018, 5, 1801015.
21. K. Ma, X. Y. Du, Y. W. Zhang and S. Chen, J. Mater. Chem. C, 2017, 5, 9398–9404.
22. M. Crespo-Quesada, L. M. Pazos-Outón, J. Warnan, M. F. Kuehnel, R. H. Friend and E. Reisner, Nat. Commun., 2016,(7), 1–7.
23. K. Rokesh, M. Sakar and T. O. Do, Mater. Lett., 2019, 242, 99–102.
24. D. Y. Heo, H. H. Do, S. H. Ahn and S. Y. Kim, Polymers, 2020, 12, 2061.
25. T. Himanshu, A. K. Agarwal, P. R. Chakraborty and S. Powar, Advances in Solar Energy Research, Springer US, 2018.
26. A. Raveendran, M. Chandran and R. Dhanusuraman, RSC Adv., 2023, 13, 3843–3876.
27. Y. Sang, H. Liu and A. Umar, ChemCatChem, 2015, 7, 559–573.
28. C. Jiang, S. J. A. Moniz, A. Wang, T. Zhang and J. Tang, Chem. Soc. Rev., 2017, 46, 4645–4660.
29. O. K. Varghese and C. A. Grimes, Sol. Energy Mater. Sol. Cells, 2008, 92, 374–384.
30. Z. Chen, T. F. Jaramillo, T. G. Deutsch, A. Kleiman-Shwarsctein, A. J. Forman, N. Gaillard, R. Garland, K. Takanabe, C. Heske, M. Sunkara, E. W. McFarland, K. Domen, E. L. Milled and H. N. Dinh, J. Mater. Res., 2010, 25, 3–16.
31. T. Hisatomi, J. Kubota and K. Domen, Chem. Soc. Rev., 2014, 43, 7520–7535.
32. J. W. Jang, D. Friedrich, S. Müller, M. Lamers, H. Hempel, S. Lardhi, Z. Cao, M. Harb, L. Cavallo, R. Heller, R. Eichberger, R. van de Krol and F. F. Abdi, Adv. Energy Mater., 2017, 7, 1701536.
33. Y. Lu, Y. Yang, X. Fan, Y. Li, D. Zhou, B. Cai, L. Wang, K. Fan and K. Zhang, Adv. Mater., 2022, 34, 2108178.
34. G. Rose, in De Novis Quibusdam Fossilibus Quae in Montibus Uraliis Inveniuntur, 1839.
35. K. Bienkowski, R. Solarska, L. Trinh, J. Widera-Kalinowska, B. Al-Anesi, M. Liu, G. K. Grandhi, P. Vivo, B. Oral, B. Yılmaz and R. Yıldırım, ACS Catal., 2024, 14, 6603–6622.
36. D. Saikia, A. Betal, J. Bera and S. Sahu, Mater. Sci. Semicond. Process., 2022, 150, 106953.
37. Y. Huang, J. Liu, Y. Deng, Y. Qian, X. Jia, M. Ma, C. Yang, K. Liu, Z. Wang, S. Qu and Z. Wang, J. Semicond., 2020, 41, 011701.
38. J. Zhou and J. Huang, Adv. Sci., 2018, 5, 1700256.
39. V. M. Goldschmidt, Naturwissenschaften, 1926, 14, 477–485.
40. M. Moniruddin, B. Ilyassov, X. Zhao, E. Smith, T. Serikov, N. Ibrayev, R. Asmatulu and N. Nuraje, Mater. Today Energy, 2018, 7, 246–259.
41. H. Kato, H. Kobayashi and A. Kudo, J. Phys. Chem. B, 2002, 106, 12441–12447.
42. B. Weng, C. R. Grice, J. Ge, T. Poudel, X. Deng and Y. Yan, Adv. Energy Mater., 2017, 8, 1701655.
43. T. Takata, K. Shinohara, A. Tanaka, M. Hara, J. N. Kondo and K. Domen, J. Photochem. Photobiol., A, 1997, 106, 45–49.
44. A. Gupta, U. Alam and N. Verma, Int. J. Hydrogen Energy, 2024, 58, 69–78.
45. B. Du, W. Yang, Q. Jiang, H. Shan, D. Luo, B. Li, W. Tang, F. Lin, B. Shen, Q. Gong, X. Zhu, R. Zhu and Z. Fang, Adv. Opt. Mater., 2018, 6, 1701271.
46. A. Fakharuddin, M. K. Gangishetty, M. Abdi-Jalebi, S. H. Chin, A. R. bin Mohd Yusoff, D. N. Congreve, W. Tress, F. Deschler, M. Vasilopoulou and H. J. Bolink, Nat. Electron., 2022,(5), 203–216.
47. K. Shum and A. Tsatskina, Nat. Energy, 2016,(1), 1–2.
48. J. A. Gómez-Cuaspud, E. Vera-López, J. B. Carda-Castelló and E. Barrachina-Albert, React. Kinet. Mech. Catal., 2017, 120, 167–179.
49. X. Yang, G. Xu, Z. Ren, X. Wei, C. Chao, S. Gong, G. Shen and G. Han, CrystEngComm, 2014, 16, 4176–4182.
50. E. Grabowska, Appl. Catal., B, 2016, 186, 97–126.
51. L. Yang, S. Fu, K. Leng, Q. Tang, Z. Wu, K. Yi and X. Zhu, Open Ceram., 2022, 9, 100238.
52. H. Li, C. Lai, Z. Wei, X. Zhou, S. Liu, L. Qin, H. Yi, Y. Fu, L. Li, M. Zhang, F. Xu, H. Yan, M. Xu, D. Ma and Y. Li, Chemosphere, 2023, 344, 140395.
53. M. A. I. Islam, T. Sato, F. Ara and M. A. Basith, J. Alloys Compd., 2023, 944, 169066.
54. S. Boumaza, R. Brahimi, L. Boudjellal, A. Belhadi and M. Trari, Synthesis and Characterization of the Double Perovskite La₂NiO₄-Application for Hydrogen Production, ed. A. Khellaf, Springer Nature Singapore, Singapore, 2021, pp. 339–346.
55. J. Kirchnerova and D. Klvana, Int. J. Hydrogen Energy, 1994, 19, 501–506.
56. S. Feng, Y. Zhang, P. Zhang, J. Cao, J. Wang, Y. Zheng, L. Shi, J. Pan and C. Li, ACS Appl. Nano Mater., 2023, 6, 14469–14477.
57. T. Soto-Montero and M. Morales-Masis, ACS Energy Lett., 2024, 9, 4199–4208.
58. S. Sharma, D. Kumar, A. Manori Dhyani and C. Bhardwaj, Mater. Today Proc., 2023, 73, 118–121.
59. B. M. Bresolin, Y. Park and D. W. Bahnemann, Catalysts, 2020, 10, 709.
60. T. Bu, J. Li, Q. Lin, D. P. McMeekin, J. Sun, M. Wang, W. Chen, X. Wen, W. Mao, C. R. McNeill, W. Huang, X. L. Zhang, J. Zhong, Y. B. Cheng, U. Bach and F. Huang, Nano Energy, 2020, 75, 104917.
61. Q. Dong, Y. Fang, Y. Shao, P. Mulligan, J. Qiu, L. Cao and J. Huang, Science, 2015, 347, 967–970.
62. D. W. DeQuilettes, S. M. Vorpahl, S. D. Stranks, H. Nagaoka, G. E. Eperon, M. E. Ziffer, H. J. Snaith and D. S. Ginger, Science, 2015, 348, 683–686.
63. C. C. Stoumpos, C. D. Malliakas and M. G. Kanatzidis, Inorg. Chem., 2013, 52, 9019–9038.
64. Y. Gao, K. Wang, M. Wang, J. I. Khan, A. H. Balawi, W. Liu, S. De Wolf and F. Laquai, Sol. RRL, 2020, 4, 2000072.
65. X. Wan, Z. Yu, W. Tian, F. Huang, S. Jin, X. Yang, Y. B. Cheng, A. Hagfeldt and L. Sun, J. Energy Chem., 2020, 46, 8–15.
66. National Renewable Energy Laboratory, Research Cell Efficiency Record, 2019.
67. Q. Jiang, Y. Zhao, X. Zhang, X. Yang, Y. Chen, Z. Chu, Q. Ye, X. Li, Z. Yin and J. You, Nat. Photonics, 2019, 12, 460–466.
68. H. Huang, B. Pradhan, J. Hofkens, M. B. J. Roeffaers and J. A. Steele, ACS Energy Lett., 2020, 5, 1107–1123.
69. P. Schulz, E. Edri, S. Kirmayer, G. Hodes, D. Cahen and A. Kahn, Energy Environ. Sci., 2014, 7, 1377–1381.
70. G. E. Eperon, S. D. Stranks, C. Menelaou, M. B. Johnston, L. M. Herz and H. J. Snaith, Energy Environ. Sci., 2014, 7, 982–988.
71. S. Wang, Y. Jiang, E. J. Juarez-Perez, L. K. Ono and Y. Qi, Nat. Energy, 2017, 2, 16195.
72. Y. Xie, Y. Yu, J. Gong, C. Yang, P. Zeng, Y. Dong, B. Yang, R. Liang, Q. Ou and S. Zhang, Opt. Mater. Express, 2018, 8, 3494–3505.
73. L. Gao, S. Huang, L. Chen, X. Li, B. Ding, S. Huang and G. Yang, Sol. RRL, 2018, 2, 1800088.
74. M. Xie and J. Tian, J. Phys. Chem. Lett., 2022, 13, 1962–1971.
75. M. Shirayama, M. Kato, T. Miyadera, T. Sugita, T. Fujiseki, S. Hara, H. Kadowaki, D. Murata, M. Chikamatsu and H. Fujiwara, J. Appl. Phys., 2016, 119, 115501.
76. S. J. A. Moniz, C. S. Blackman, P. Southern, P. M. Weaver, J. Tang and C. J. Carmalt, Nanoscale, 2015, 7, 16343–16353.
77. P. Nandigana, S. Pari, D. Sujatha, M. Shidhin, C. Jeyabharathi and S. K. Panda, ChemistrySelect, 2023, 8, e202204731.
78. J. Li, Z. Tan, M. Hu, C. Chen, J. Luo, S. Li, L. Gao, Z. Xiao, G. Niu and J. Tang, Front. Optoelectron., 2019, 12, 352–364.
79. S. Kumar and A. K. Ganguli, Appl. Surf. Sci. Adv., 2022, 9, 100239.
80. I. Chung, J. H. Song, J. Im, J. Androulakis, C. D. Malliakas, H. Li, A. J. Freeman, J. T. Kenney and G. Kanatzidis, J. Am. Chem. Soc., 2012, 134, 8579–8587.
81. H. Peng, L. Xu, Y. Sheng, W. Sun, Y. Yang, H. Deng, W. Chen and J. Liu, Small, 2021, 17, 2102149.
82. A. Exner, R. A. Gouvêa, A. Köche, S. Khan, J. F. L. Santos and M. José Leite Santos, J. Sci.:Adv. Mater. Devices, 2024, 9, 100700.
83. A. M. Ganose, C. N. Savory and D. O. Scanlon, Chem. Commun., 2016, 53, 20–44.
84. O. A. Lozhkina, A. A. Murashkina, V. V. Shilovskikh, Y. V. Kapitonov, V. K. Ryabchuk, A. V. Emeline and T. Miyasaka, J. Phys. Chem. Lett., 2018, 9, 5408–5411.
85. T. Kako, N. Kikugawa and J. Ye, Catal. Today, 2008, 131, 197–202.
86. F. Matteocci, L. Cinà, E. Lamanna, S. Cacovich, G. Divitini, P. A. Midgley, C. Ducati and A. Di Carlo, Nano Energy, 2016, 30, 162–172.
87. R. Cheacharoen, C. C. Boyd, G. F. Burkhard, T. Leijtens, J. A. Raiford, K. A. Bush, S. F. Bent and M. D. McGehee, Sustainable Energy Fuels, 2018, 2, 2398–2406.
88. F. Corsini and G. Griffini, JPhys Energy, 2020, 2, 031002.
89. R. Konta, T. Ishii, H. Kato and A. Kudo, J. Phys. Chem. B, 2004, 108, 8992–8995.
90. T. Ohno, T. Tsubota, Y. Nakamura and K. Sayama, Appl. Catal., A, 2005, 288, 74–79.
91. Z. Li, M. Yang, J. S. Park, S. H. Wei, J. J. Berry and K. Zhu, Chem. Mater., 2016, 28, 284–292.
92. J. P. Zou, L. Z. Zhang, S. L. Luo, L. H. Leng, X. B. Luo, M. J. Zhang, Y. Luo and G. C. Guo, Int. J. Hydrogen Energy, 2012, 37, 17068–17077.
93. M. Abdi, V. Mahdikhah and S. Sheibani, Opt. Mater., 2020, 102, 109803.
94. T. Oshikiri, X. Shi, K. Ueno, A. -, T. Yu, X. Chen, Z. Zhao, X. Zhang, Y. Lei, P. Yang, J. Fan, B. Zhang and S. Yin, Mater. Res. Express, 2020, 7, 15034.
95. A. Swarnkar, W. J. Mir and A. Nag, ACS Energy Lett., 2018, 3, 286–289.
96. G. E. Eperon and D. S. Ginger, ACS Energy Lett., 2017, 2, 1190–1196.
97. G. Fu, W. Li, J.-Y. Zhang, M. Li, C. Li, N. Li, Q. He, S. Xi, D. Qi, J. L. MacManus-Driscoll, J. Cheng, K. Hongliang Zhang, G. Fu, J. Zhang, J. Cheng, K. H. Zhang, W. Li, J. L. MacManus-Driscoll, M. Li, C. Li, N. Li, Q. He, S. Xi and D. Qi, Small, 2021, 17, 2006930.
98. Z. Li, L. Lv, J. Wang, X. Ao, Y. Ruan, D. Zha, G. Hong, Q. Wu, Y. Lan, C. Wang, J. Jiang and M. Liu, Nano Energy, 2018, 47, 199–209.
99. B. Hua, M. Li, W. Pang, W. Tang, S. Zhao, Z. Jin, Y. Zeng, B. Shalchi Amirkhiz and J. L. Luo, Chem, 2018, 4, 2902–2916.
100. J. Xiong, H. Zhong, J. Li, X. Zhang, J. Shi, W. Cai, K. Qu, C. Zhu, Z. Yang, S. P. Beckman and H. Cheng, Appl. Catal., B, 2019, 256, 117817.
101. W. Wang, Y. Yang, D. Huan, L. Wang, N. Shi, Y. Xie, C. Xia, R. Peng and Y. Lu, J. Mater. Chem. A, 2019, 7, 12538–12546.
102. C.-H. Lu, G. V Biesold-Mcgee, Y. Liu, Z. Kang and Z. Lin, Chem. Soc. Rev., 2020, 49, 4953.
103. A. Buin, R. Comin, J. Xu, A. H. Ip and E. H. Sargent, Chem. Mater., 2015, 27, 4405–4412.
104. G. Xing, N. Mathews, S. Sien Lim, N. Yantara, X. Liu, D. Sabba, M. Grätzel, S. Mhaisalkar and T. Chien Sum, Nat. Mater., 2014, 13, 476–480.
105. Y. Liu, C. Ye, S. N. Zhao, Y. Wu, C. Liu, J. Huang, L. Xue, J. Sun, W. Zhang, X. Wang, P. Xiong and J. Zhu, Nano Energy, 2022, 99, 107344.
106. J. K. Cooper, S. Gul, F. M. Toma, L. Chen, P. A. Glans, J. Guo, J. W. Ager, J. Yano and I. D. Sharp, Chem. Mater., 2014, 26, 5365–5373.
107. J. H. Kim, Y. Jo, J. H. Kim, J. W. Jang, H. J. Kang, Y. H. Lee, D. S. Kim, Y. Jun and J. S. Lee, ACS Nano, 2015, 9, 11820–11829.
108. Y. Pihosh, I. Turkevych, K. Mawatari, T. Asai, T. Hisatomi, J. Uemura, M. Tosa, K. Shimamura, J. Kubota, K. Domen and T. Kitamori, Small, 2014, 10, 3692–3699.
109. L. Han, F. F. Abdi, R. Van De Krol, R. Liu, Z. Huang, H. J. Lewerenz, B. Dam, M. Zeman and A. H. M. Smets, ChemSusChem, 2014, 7, 2832–2838.
110. Y. Yang, D. Xu, Q. Wu and P. Diao, Sci. Rep., 2016, 6, 35158.
111. J. Wen, Y. Zhao, P. Wu, Y. Liu, X. Zheng, R. Lin, S. Wan, K. Li, H. Luo, Y. Tian, L. Li and H. Tan, Nat. Commun., 2023, 14, 1–10.
112. H. Li, Y. Zhou, W. Tu, J. Ye and Z. Zou, Adv. Funct. Mater., 2015, 25, 998–1013.
113. B. Y. Cheng, J. S. Yang, H. W. Cho and J. J. Wu, ACS Appl. Mater. Interfaces, 2016, 8, 20032–20039.
114. J. S. Jang, C. W. Ahn, S. S. Won, J. H. Kim, W. Choi, B.-S. Lee, J.-H. Yoon, H. G. Kim and J. S. Lee, J. Phys. Chem. C, 2017, 121, 15063–15070.
115. J. P. Jayaraman, M. Hamdan, M. Velpula, N. S. Kaisare and A. K. Chandiran, ACS Appl. Mater. Interfaces, 2021, 13, 16267–16278.
116. S. Samajdar, S. Bera, P. S. Das, H. Finch, V. R. Dhanak, S. Chakraborty, T. Maiyalagan, K. Annapurna and S. Ghosh, Int. J. Hydrogen Energy, 2023, 48, 17838–17851.
117. S. Chandrasekaran, E. J. Kim, J. S. Chung, I. K. Yoo, V. Senthilkumar, Y. S. Kim, C. R. Bowen, V. Adamaki and S. Hyun Hur, Chem. Eng. J., 2017, 309, 682–690.
118. T. T. T. Tran, S. Kim and J. Seo, J. Catal., 2022, 406, 157–164.
119. Q. Xu, L. Zhang, B. Cheng, J. Fan and J. Yu, Chem, 2020, 6, 1543–1559.
120. T. W. Kim and K. S. Choi, Science, 2014, 343, 990–994.
121. M. Hamdan and A. K. Chandiran, Angew. Chem. Int. Ed., 2020, 59, 16033–16038.
122. J. Seo, T. Hisatomi, M. Nakabayashi, N. Shibata, T. Minegishi, M. Katayama, K. Domen, J. Seo, T. Hisatomi, T. Minegishi, M. Katayama, K. Domen, M. Nakabayashi and N. Shibata, Adv. Energy Mater., 2018, 8, 1800094.
123. J. R. Long and O. M. Yaghi, Chem. Soc. Rev., 2009, 38, 1213–1214.
124. R. Kaur, K. H. Kim, A. K. Paul and A. Deep, J. Mater. Chem. A, 2016, 4, 3991–4002.
125. S. Lin, P. M. Usov and A. J. Morris, Chem. Commun., 2018, 54, 6965–6974.
126. A. Crake, K. C. Christoforidis, A. Gregg, B. Moss, A. Kafizas and C. Petit, Small, 2019, 15, 1805473.
127. X. Feng, Y. Pi, Y. Song, C. Brzezinski, Z. Xu, Z. Li and W. Lin, J. Am. Chem. Soc., 2020, 142, 690–695.
128. J. Hou, Z. Wang, P. Chen, V. Chen, A. K. Cheetham and L. Wang, Angew. Chem., Int. Ed., 2020, 59, 19434–19449.
129. Y. Shi, A. F. Yang, C. S. Cao and B. Zhao, Coord. Chem. Rev., 2019, 390, 50–75.
130. Y. Xiao, X. Guo, N. Yang and F. Zhang, J. Energy Chem., 2021, 58, 508–522.
131. Y. Pi, X. Feng, Y. Song, Z. Xu, Z. Li and W. Lin, J. Am. Chem. Soc., 2020, 142, 10302–10307.
132. S. Yuan, L. Feng, K. Wang, J. Pang, M. Bosch, C. Lollar, Y. Sun, J. Qin, X. Yang, P. Zhang, Q. Wang, L. Zou, Y. Zhang, L. Zhang, Y. Fang, J. Li and H. C. Zhou, Adv. Mater., 2018, 30, 1704303.
133. C. C. Lee, C. I. Chen, Y. Te Liao, K. C. W. Wu and C. C. Chueh, Adv. Sci., 2019, 6, 1801715.
134. J. Tian, J. Yu, Q. Tang, L. Zhang, F. Liand and L. Luo, IOP Conf. Ser. Earth Environ. Sci., 2018, 108, 042104.
135. Z. Ni and R. I. Masel, J. Am. Chem. Soc., 2006, 128, 12394–12395.
136. Y. K. Seo, G. Hundal, I. T. Jang, Y. K. Hwang, C. H. Jun and J. S. Chang, Microporous Mesoporous Mater., 2009, 119, 331–337.
137. T. Tsuzuki, Commun. Chem., 2021,(4), 1–11.
138. M. Ali, E. Pervaiz, T. Noor, O. Rabi, R. Zahra and M. Yang, Int. J. Energy Res., 2021, 45, 1190–1226.
139. Z. Zhuang and D. Liu, Nano-Micro Lett., 2020, 12, 132.
140. Y. Dou, J. Zhou, A. Zhou, J. R. Li and Z. Nie, J. Mater. Chem. A, 2017, 5, 19491–19498.
141. J. W. Yoon, D. H. Kim, J. H. Kim, H. W. Jang and J. H. Lee, Appl. Catal., B, 2019, 244, 511–518.
142. H. Song, Z. Sun, Y. Xu, Y. Han, J. Xu, J. Wu, T. Sun, H. Meng and X. Zhang, Sep. Purif. Technol., 2019, 228, 115764.
143. T. Song, L. Zhang, P. Zhang, J. Zeng, T. Wang, A. Ali and H. Zeng, J. Mater. Chem. A, 2017, 5, 6013–6018.
144. X. Cao, C. Tan, M. Sindoro and H. Zhang, Chem. Soc. Rev., 2018, 47, 5997.
145. S. Dang, Q. L. Zhu and Q. Xu, Nat. Rev. Mater., 2018, 3, 17075.
146. C. Guan, X. Liu, W. Ren, X. Li, C. Cheng, J. Wang, C. Guan, X. M. Liu, X. Li, J. Wang, W. N. Ren and C. W. Cheng, Adv. Energy Mater., 2017, 7, 1602391.
147. D. Xu, T. Xia, W. Fan, H. Bai, J. Ding, B. Mao and W. Shi, Appl. Surf. Sci., 2019, 491, 497–504.
148. L. Shi, D. Benetti, F. Li, Q. Wei and F. Rosei, Appl. Catal., B, 2020, 263, 118317.
149. L. Zhang, P. Cui, H. Yang, J. Chen, F. Xiao, Y. Guo, Y. Liu, W. Zhang, F. Huo and B. Liu, Adv. Sci., 2015, 3, 1500243.
150. L. Pan, T. Muhammad, L. Ma, Z. F. Huang, S. Wang, L. Wang, J. J. Zou and X. Zhang, Appl. Catal., B, 2016, 189, 181–191.
151. J. D. Xiao, Q. Shang, Y. Xiong, Q. Zhang, Y. Luo, S. H. Yu and H. L. Jiang, Angew. Chem. Int. Ed., 2016, 55, 9389–9393.
152. Z. Jin and H. Yang, Nanoscale Res. Lett., 2017, 12, 539.
153. X. Hao, Z. Jin, H. Yang, G. Lu and Y. Bi, Appl. Catal., B, 2017, 210, 45–56.
154. Y. J. Dong, J. F. Liao, Z. C. Kong, Y. F. Xu, Z. J. Chen, H. Y. Chen, D. Bin Kuang, D. Fenske and C. Y. Su, Appl. Catal., B, 2018, 237, 9–17.
155. X. Li, S. Liu, K. Fan, Z. Liu, B. Song and J. Yu, Adv. Energy Mater., 2018, 8, 1800101.
156. G. Jia, L. Liu, L. Zhang, D. Zhang, Y. Wang, X. Cui and W. Zheng, Appl. Surf. Sci., 2018, 448, 254–260.
157. Q. Zhang, H. Wang, Y. Dong, J. Yan, X. Ke, Q. Wu and S. Xue, Sol. Energy, 2018, 171, 388–396.
158. P. Tian, X. He, W. Li, L. Zhao, W. Fang, H. Chen, F. Zhang, W. Zhang and W. Wang, J. Mater. Sci., 2018, 53, 12016–12029.
159. Z. Q. Liu, K. Qiu, H. Bai, F. Wang, Y. Ge, W. Cui, G. Zheng, J. Cui and W. Fan, Funct. Mater. Lett., 2018, 11, 1850085.
160. Z. Peng, S. C. Abbas, J. Lv, R. Yang, M. Wu and Y. Wang, Int. J. Hydrogen Energy, 2019, 44, 2446–2453.
161. C. H. Li, C. L. Huang, X. F. Chuah, D. Senthil Raja, C. T. Hsieh and S. Y. Lu, Chem. Eng. J., 2019, 361, 660–670.
162. X. Hou, L. Wu, L. Gu, G. Xu, H. Du and Y. Yuan, J. Mater. Sci. Mater. Electron., 2019, 30, 5203–5211.
163. W. Cui, J. Shang, H. Bai, J. Hu, D. Xu, J. Ding, W. Fan and W. Shi, Chem. Eng. J., 2020, 388, 124206.
164. J. Han, Y. Wang, Y. Liu, J. Bai, R. Zhao and L. Wang, Catal. Lett., 2021, 151, 2560.
165. C. G. Silva, I. Luz, F. X. Llabrés, I. Xamena, A. Corma and H. García, Chem.–Eur. J., 2010, 16, 11133–11138.
166. M. W. Logan, S. Ayad, J. D. Adamson, T. Dilbeck, K. Hanson and F. J. Uribe-Romo, J. Mater. Chem. A, 2017, 5, 11854–11863.
167. Y. Li, Y. Fu, B. Ni, K. Ding, W. Chen, K. Wu, X. Huang and Y. Zhang, AIP Adv., 2018, 8, 035012.
168. T. Toyao, M. Saito, Y. Horiuchi, K. Mochizuki, M. Iwata, H. Higashimura and M. Matsuoka, Catal. Sci. Technol., 2013, 3, 2092–2097.
169. C. H. Hendon, D. Tiana, M. Fontecave, C. Sanchez, L. D'Arras, C. Sassoye, L. Rozes, C. Mellot-Draznieks and A. Walsh, J. Am. Chem. Soc., 2013, 135, 10942–10945.
170. L. Shen, M. Luo, Y. Liu, R. Liang, F. Jing and L. Wu, Appl. Catal., B, 2015, 166–167, 445–453.
171. W. Cui, H. Bai, J. Shang, F. Wang, D. Xu, J. Ding, W. Fan and W. Shi, Electrochim. Acta, 2020, 349, 136383.
172. M. A. Nasalevich, M. Van Der Veen, F. Kapteijn and J. Gascon, CrystEngComm, 2014, 16, 4919–4926.
173. J. He, H. Yang, Y. Chen, Z. Yan, Y. Zeng, Z. Luo, W. Gao and J. Wang, Water Air Soil Pollut., 2015, 226, 197.
174. H. Wang, X. Yuan, Y. Wu, G. Zeng, H. Dong, X. Chen, L. Leng, Z. Wu and L. Peng, Appl. Catal., B, 2016, 186, 19–29.
175. B. O'Regan and M. Grätzel, Nature, 1991, 353, 737–740.
176. J. He, J. Wang, Y. Chen, J. Zhang, D. Duan, Y. Wanga and Z. Yan, Chem. Commun., 2014, 50, 7063–7066.
177. H. Tributsch, Coord. Chem. Rev., 2004, 248, 1511–1530.
178. Y. P. Yuan, L. S. Yin, S. W. Cao, G. S. Xu, C. H. Li and C. Xue, Appl. Catal., B, 2015, 168–169, 572–576.
179. B. Ma, F. Wen, H. Jiang, J. Yang, P. Ying and C. Li, Catal. Lett., 2010, 134, 78–86.
180. Y. Fu, D. Sun, Y. Chen, R. Huang, Z. Ding, X. Fu and Z. Li, Angew. Chem. Int. Ed., 2012, 51, 3364–3367.
181. M. Sun, Q. Q. Wang, C. Qin, C. Y. Sun, X. L. Wang and Z. M. Su, Chem.–Eur. J., 2018, 25, 2824–2830.
182. S.-Y. Han, D.-L. Pan, H. Chen, X.-B. Bu, Y.-X. Gao, H. Gao, Y. Tian, G.-S. Li, G. Wang, S.-L. Cao, C.-Q. Wan and G.-C. Guo, Angew. Chem., 2018, 31, 10012–10017.
183. S. M. Liu, Z. Zhang, X. Li, H. Jia, M. Ren and S. Liu, Adv. Mater. Interfaces, 2018, 5, 1801062.
184. Q. Yang, Q. Xu and H. L. Jiang, Chem. Soc. Rev., 2017, 46, 4774–4808.
185. T. He, S. Chen, B. Ni, Y. Gong, Z. Wu, L. Song, L. Gu, W. Hu and X. Wang, Angew. Chem. Int. Ed., 2018, 57, 3493–3498.
186. Y. Han, H. Xu, Y. Su, Z. liang Xu, K. Wang and W. Wang, J. Catal., 2019, 370, 70–78.
187. B. Hinnemann, P. G. Moses, J. Bonde, K. P. Jørgensen, J. H. Nielsen, S. Horch, I. Chorkendorff and J. K. Nørskov, J. Am. Chem. Soc., 2005, 127, 5308–5309.
188. Y. Hou, B. L. Abrams, P. C. K. Vesborg, M. E. Björketun, K. Herbst, L. Bech, A. M. Setti, C. D. Damsgaard, T. Pedersen, O. Hansen, J. Rossmeisl, S. Dahl, J. K. Nørskov and I. Chorkendorff, Nat. Mater., 2011, 10, 434–438.
189. K. Arifin, R. M. Yunus, L. J. Minggu and M. B. Kassim, Int. J. Hydrogen Energy, 2021, 46, 4998–5024.
190. T. F. Hou, M. A. Johar, R. Boppella, M. A. Hassan, S. J. Patil, S. W. Ryu and D. W. Lee, J. Energy Chem., 2020, 49, 262–274.
191. T. N. Q. Trang, L. T. N. Tu, T. V. Man, M. Mathesh, N. D. Nam and V. T. H. Thu, Composites, Part B, 2019, 174, 106969.
192. S. Li, W. Xu, L. Meng, W. Tian and L. Li, Small Sci., 2022, 2, 2100112.
193. C. V. Reddy, K. R. Reddy, V. V. N. Harish, J. Shim, M. V. Shankar, N. P. Shetti and T. M. Aminabhavi, Int. J. Hydrogen Energy, 2020, 45, 7656–7679.
194. G. E. M. Schukraft, B. Moss, A. G. Kafizas and C. Petit, ACS Appl. Mater. Interfaces, 2022, 14, 19342–19352.
195. S. Guo, L. Chi, T. Zhao, Y. Nan, X. Sun, Y. Huang, B. Hou and X. Wang, J. Electroanal. Chem., 2021, 880, 114915.
196. W. W. Dong, J. Jia, Y. Wang, J. R. An, O. Y. Yang, X. J. Gao, Y. L. Liu, J. Zhao and D. S. Li, Chem. Eng. J., 2022, 438, 135622.
197. T. Zeng, D. Shi, Q. Cheng, G. Liao, H. Zhou and Z. Pan, Environ. Sci.:Nano, 2020, 7, 861–879.
198. R. Peng, C. M. Wu, J. Baltrusaitis, N. M. Dimitrijevic, T. Rajh and R. T. Koodali, Chem. Commun., 2013, 49, 3221–3223.
199. Y. Horiuchi, T. Toyao, K. Miyahara, L. Zakary, D. Do Van, Y. Kamata, T. H. Kim, S. W. Lee and M. Matsuoka, Chem. Commun., 2016, 52, 5190–5193.
200. Y. Li, Y. Xia, K. Liu, K. Ye, Q. Wang, S. Zhang, Y. Huang and H. Liu, ACS Appl. Mater. Interfaces, 2020, 12, 25494–25502.
201. L. Zhao, Z. Zhao, Y. Li, X. Chu, Z. Li, Y. Qu, L. Bai and L. Jing, Nanoscale, 2020, 12, 10010–10018.
202. Y. Pan, D. Li and H. L. Jiang, Chem.–Eur. J., 2018, 24, 18403–18407.
203. L. Shi, T. Wang, H. Zhang, K. Chang and J. Ye, Adv. Funct. Mater., 2015, 25, 5360–5367.
204. S. M. Hosseini, M. Karimi and V. Safarifard, New J. Chem., 2021, 45, 17674–17682.
205. V. R. Bakuru, M. E. DMello and S. B. Kalidindi, ChemPhysChem, 2019, 20, 1177–1215.
206. K. Meyer, M. Ranocchiari and J. A. Van Bokhoven, Energy Environ. Sci., 2015, 8, 1923–1937.
207. S. Wang and X. Wang, Small, 2015, 11, 3097–3112.
208. C. Zhang, W. Li and L. Li, Angew. Chem. Int. Ed., 2020, 60, 7488–7501.
209. L. Li, H. Zhang, C. Liu, P. Liang, N. Mitsuzaki and Z. Chen, Energy Technol., 2019, 7, 1801069.
210. Z. Jiao, J. Zheng, C. Feng, Z. Wang, X. Wang, G. Lu and Y. Bi, ChemSusChem, 2016, 9, 2824–2831.
211. W. Zhang, R. Li, X. Zhao, Z. Chen, A. W. K. Law and K. Zhou, ChemSusChem, 2018, 11, 2710–2716.
212. H. Yang, J. Bright, S. Kasani, P. Zheng, T. Musho, B. Chen, L. Huang and N. Wu, Nano Res., 2019, 12, 643–650.
213. W. Wang, X. Xu, W. Zhou and Z. Shao, Adv. Sci., 2017, 4, 1600371.
214. A. V. Vinogradov, H. Zaake-Hertling, E. Hey-Hawkins, A. V. Agafonov, G. A. Seisenbaeva, V. G. Kessler and V. V. Vinogradov, Chem. Commun., 2014, 50, 10210–10213.
215. R. Dutta, R. Shrivastav, M. Srivastava, A. Verma, S. Saxena, N. K. Biswas, V. R. Satsangi and S. Dass, Int. J. Hydrogen Energy, 2022, 47, 5192–5210.
216. S. K. Yadav, G. K. Grandhi, D. P. Dubal, J. C. de Mello, M. Otyepka, R. Zbořil, R. A. Fischer and K. Jayaramulu, Small, 2020, 16, 1–32.
217. J. Lee, N. Tsvetkov, S. R. Shin and J. K. Kang, ACS Appl. Mater. Interfaces, 2022, 14, 35495–35503.
218. R. Stanton and D. J. Trivedi, Nano Lett., 2023, 23, 11932–11939.
219. Y. Ye, Y. Yin, Y. Chen, S. Li, L. Li and Y. Yamauchi, Small, 2023, 19, 2208119.
220. S. Mollick, T. N. Mandal, A. Jana, S. Fajal, A. V. Desai and S. K. Ghosh, ACS Appl. Nano Mater., 2019, 2, 1333–1340.
221. L. Y. Wu, Y. F. Mu, X. X. Guo, W. Zhang, Z. M. Zhang, M. Zhang and T. B. Lu, Angew. Chem. Int. Ed., 2019, 58, 9491–9495.
222. Y. Dong, J. Zhang, H. Zhang, W. Wang, B. Hu, D. Xia, K. Lin, L. Geng and Y. Yang, Nano-Micro Lett., 2024, 16, 1–14.
223. T. H. Chang, C. W. Kung, H. W. Chen, T. Y. Huang, S. Y. Kao, H. C. Lu, M. H. Lee, K. M. Boopathi, C. W. Chu and K. C. Ho, Adv. Mater., 2015, 27, 7229–7235.
224. H. A. Younes, M. Taha, R. Khaled, H. M. Mahmoud and R. M. Abdelhameed, J. Alloys Compd., 2023, 930, 167322.
225. D. Xu, L. Li, H. Xu, J. Zhu, W. Fan, J. Ding and W. Shi, J. Solid State Chem., 2019, 280, 120986.
226. S. Sharma, N. Jacob, G. K. Grandhi, M. B. Choudhary, S. Ippili, V. R. Hathwar, P. Vivo, R. Lo, M. Motapothula and K. Jayaramulu, iScience, 2024, 27, 110924.
227. R. Tang, L. Wang, Z. Zhang, W. Yang, H. Xu, A. Kheradmand, Y. Jiang, R. Zheng and J. Huang, Appl. Catal., B, 2021, 296, 120382.
228. R. Tang, S. Zhou, H. Li, R. Chen, L. Zhang and L. Yin, Appl. Catal., B, 2020, 265, 118583.
229. H. Ramezanalizadeh and F. Manteghi, RSC Adv., 2016, 6, 99096–99104.
230. J. Swain, A. Priyadarshini, S. Hajra, S. Panda, J. Panda, R. Samantaray, Y. Yamauchi, M. Han, H. J. Kim and R. Sahu, J. Alloys Compd., 2023, 965, 171438.
231. A. Afzalinia and M. Mirzaee, J. Mol. Struct., 2023, 1283, 135322.
232. X. Kang, K. Lyu, L. Li, J. Li, L. Kimberley, B. Wang, L. Liu, Y. Cheng, M. D. Frogley, S. Rudić, A. J. Ramirez-Cuesta, R. A. W. Dryfe, B. Han, S. Yang and M. Schröder, Nat. Commun., 2019, 10, 1–9.
233. L. Zhang, Z. Shen, D. Li, J. Lin, Y. Chen, C. Ma, R. Pu, L. Wu, Y. Zhu, C. Li, D. Hou, G. Gao, W. Luo, Z. Duan, T. Zhang, Q. Xv, Y. Wang and J. Tang, Nanotechnology, 2022, 33, 205601.
234. D. Sharma, Sungjemmenla, D. Kumar, D. Takhar, B. Birajdar, V. Kumar and N. Khare, Mater. Sci. Eng., B, 2023, 297, 116769.
235. C. Li, T. Hang, H. Zhou, P. Ge, S. Li and X. D. Zhu, Inorg. Chem. Front., 2023, 10, 3375–3382.
236. D. Rambabu, S. Bhattacharyya, T. Singh, M. L. C and T. K. Maji, Inorg. Chem., 2020, 59, 1436–1443.
237. Y. Y. Wang, X. Y. Ji, M. Yu and J. Tao, Mater. Chem. Front., 2021, 5, 7796–7807.
238. Z. Li, J. G. Li, X. Ao, H. Sun, H. Wang, M. F. Yuen and C. Wang, Electrochim. Acta, 2020, 334, 135638.
239. T. M. H. Nguyen and C. W. Bark, ACS Omega, 2020, 5, 2280–2286.
240. Y. N. Zhang, B. Li, L. Fu, Q. Li and L. W. Yin, Electrochim. Acta, 2020, 330, 135280.
241. D. Shen, A. Pang, Y. Li, J. Dou and M. Wei, Chem. Commun., 2018, 54, 1253–1256.
242. J. Wu, X. Liang, J. Liu, L. yu, F. Wang, T. Wang, Y. Sun, X. Zhou, X. yuan Liu, H. Lin, Q. Zhu, H. Wang, G. Li and H. Hu, Adv. Funct. Mater., 2024, 34, 2407381.
243. D. Kim, D. K. Lee, S. M. Kim, W. Park and U. Sim, Materials, 2020, 13, 210.
244. A. Nikolskaia, M. Vildanova, S. Kozlov, N. Tsvetkov, L. Larina and O. Shevaleevskiy, Sustainability, 2020, 12, 788.
245. J. Ma, Z. Lin, X. Guo, L. Zhou, J. Su, C. Zhang, Z. Yang, J. Chang, S. Liu and Y. Hao, Sol. RRL, 2019, 3, 1900096.
246. U. J. Ryu, S. Jee, J. S. Park, I. K. Han, J. H. Lee, M. Park and K. M. Choi, ACS Nano, 2018, 12, 4968–4975.
247. L. K. Ono, S. R. Raga, M. Remeika, A. J. Winchester, A. Gabe and Y. Qi, J. Mater. Chem. A, 2015, 3, 15451–15456.
248. T. Shi, J. Chen, J. Zheng, X. Li, B. Zhou, H. Cao and Y. Wang, Sci. Rep., 2016, 6, 39132.
249. S. Guarnera, A. Abate, W. Zhang, J. M. Foster, G. Richardson, A. Petrozza and H. J. Snaith, J. Phys. Chem. Lett., 2015, 6, 432–437.
250. S. Zhang, T. Duan, D. Yan, H. Yang, L. Cheng, Y. Pei, K. You and H. Luo, Chem. Eng. J., 2024, 486, 150176.
251. M. Li, D. Xia, Y. Yang, X. Du, G. Dong, A. Jiang and R. Fan, Adv. Energy Mater., 2018, 8, 1702052.
252. S. Wang, W. Yuan and Y. S. Meng, ACS Appl. Mater. Interfaces, 2015, 7, 24791–24798.
253. J. Wang, J. Zhang, Y. Yang, S. Gai, Y. Dong, L. Qiu, D. Xia, X. Fan, W. Wang, B. Hu, W. Cao and R. Fan, ACS Appl. Mater. Interfaces, 2021, 13, 5235–5244.
254. B. Xu, Z. Zhu, J. Zhang, H. Liu, C. C. Chueh, X. Li and A. K. Y. Jen, Adv. Energy Mater., 2017, 7, 1700683.
255. X. Zhou, L. Qiu, R. Fan, H. Ye, C. Tian, S. Hao and Y. Yang, J. Power Sources, 2020, 473, 228556.
256. L. Valenzano, B. Civalleri, S. Chavan, S. Bordiga, M. H. Nilsen, S. Jakobsen, K. P. Lillerud and C. Lamberti, Chem. Mater., 2011, 23, 1700–1718.
257. S. Yuan, J. S. Qin, C. T. Lollar and H. C. Zhou, ACS Cent. Sci., 2018, 4, 440–450.
258. Y. Bai, Y. Dou, L. H. Xie, W. Rutledge, J. R. Li and H. C. Zhou, Chem. Soc. Rev., 2016, 45, 2327–2367.
259. M. Li, D. Xia, A. Jiang, X. Du, X. Fan, L. Qiu, P. Wang, R. Fan and Y. Yang, Energy Technol., 2019, 7, 1900027.
260. X. Zhou, L. Qiu, R. Fan, J. Zhang, S. Hao and Y. Yang, Nano-Micro Lett., 2020, 12, 80.
261. S. Yoon, S. J. Kim, H. S. Kim, J. S. Park, I. K. Han, J. W. Jung and M. Park, Nanoscale, 2017, 9, 16305–16312.

---