Peter Danita Patricia
a and
Rajadurai Vijay Solomon
*b
aMCC-MRF Innovation Park, Madras Christian College (Autonomous), East Tambaram, Chennai–600 059, Tamil Nadu, India
bDepartment of Chemistry, Madras Christian College (Autonomous), East Tambaram, Chennai–600 059, Tamil Nadu, India. E-mail: vjsolo@mcc.edu.in; vjsolo@gmail.com
First published on 10th March 2025
Hydrogen is increasingly recognized as a promising clean fuel, offering a sustainable alternative to fossil fuels with water as its only combustion byproduct. Given several hydrogen production methods, photocatalytic water splitting stands out due to its potential for harnessing abundant solar energy to generate hydrogen. Among numerous photocatalysts reported for water-splitting, metal–organic frameworks (MOFs) exhibit excellent photocatalytic activity due to their enormous surface area. In this field, lanthanide-based MOFs (Ln-MOFs) have emerged as exceptional photocatalysts due to their unique properties and customizable structures, enhancing light absorption and charge separation. Recent advancements in the development of Ln-MOFs have demonstrated their potential to achieve notable hydrogen evolution rates under solar irradiation, positioning them at the forefront of renewable energy research. The introduction of Ln-MOFs into photocatalytic water-splitting marks a new era with a multitude of exciting possibilities ahead. In this context, a comprehensive overview of the trends and technologies involved in designing and understanding Ln-MOFs for water splitting is essential to developing efficient catalysts with enhanced properties. Here, we focus exclusively on the role of Ln-MOFs in photocatalytic water splitting, providing an in-depth analysis of their photocatalytic performance and stability. This review systematically classifies Ln-MOFs based on modifications in their frameworks, examining how these changes influence their properties and overall efficiency in hydrogen production. The review highlights the progress made in the field while addressing the gaps in current knowledge, particularly in understanding the mechanisms that govern the performance of Ln-MOFs. Moreover, it outlines future directions for enhancing the efficiency and stability of Ln-MOFs in hydrogen production, offering valuable insights that could guide further research. In summary, this review will aid the naïve and young researchers in the MOF domain to gain comprehensive knowledge on the nuances of lanthanide-based Ln-MOFs and appreciate their significant role in developing new technology for H2 production.
Water-splitting is an environmentally benign method for hydrogen production, offering zero carbon emissions and high efficiency (up to 80%).23 Its only by-product is oxygen, which has no adverse environmental impact, and the water source is abundant. This sustainable process can be driven by thermal, biochemical, photonic, or electrical energy (Fig. 2).20,24 Thermal water splitting occurs at temperatures above 2500 K, but it faces challenges such as complex reaction kinetics and the need for cooling material recovery.19 Biochemical water splitting, involving cyanobacteria and algae in direct photolysis, is greener but hindered by low efficiency, high costs, and large operational areas.18
Electrolysis is another widely known method, where water is split into hydrogen and oxygen by electricity. Hydrogen evolves at the cathode via the hydrogen evolution reaction (HER) and oxygen at the anode via the oxygen evolution reaction (OER).25,26 However, this method requires high energy input and struggles with scalability.24 To overcome these limitations, photocatalytic water-splitting has emerged as an alternative. This process uses solar energy (photons) and a catalyst to generate electrons and holes that split water into hydrogen and oxygen. Photocatalytic water-splitting, or photocatalytic hydrogen evolution (PHE), is highly efficient, as a significant portion of solar photons can be utilized in the presence of an effective catalyst.24 This approach minimizes energy conversion losses and represents a promising path toward sustainable hydrogen production.
Upon light irradiation, electrons (e−) in the photocatalyst's valence band (VB) are excited to the conduction band (CB), leaving holes (h+) behind (I). Holes oxidize water, generating protons (II), which are reduced by CB electrons (III), forming H2 and O2. The overall reaction is summarized below (IV).27–29
On irradiation, 2hv → 2e− + 2h+ | (I) |
Oxidation, 2h+ + H2O(l) → 1/2O2(g) + 2H+ | (II) |
Reduction, 2H+ + 2e− → H2(g) | (III) |
Overall reaction, 2hv + H2O(l) → 1/2O2(g) + H2(g) | (IV) |
Photocatalysts enhance reaction rates by improving photogenerated carrier separation and enabling broad-spectrum light absorption, providing more photons for the reaction. When light energy exceeds the band gap, electron–hole pairs form. For water splitting, the CB must be more negative than the H+ reduction potential and the VB more positive than the water oxidation potential.29 Although water splitting requires 1.23 eV, practical applications require a band gap of 1.6–2.2 eV.30 Smaller band gaps enable wider light absorption, while crystallinity, particle size, and structure affect charge separation. Higher crystallinity reduces defects, improving charge mobility and minimizing recombination, whereas nanometer-sized particles enhance charge transfer but may increase recombination if too small. Cocatalysts are essential for enhancing the performance of semiconductor photocatalysts in H2 evolution. Plasmonic metals like Au and Cu improve visible light absorption through localized surface plasmon resonance (LSPR) effects.31,32 By attracting electrons, cocatalysts facilitate efficient separation of electron–hole pairs and enhance charge carrier transfer when integrated with photocatalyst surfaces.33,34 Additionally, they provide active sites for photocatalytic reactions, functioning as electron sinks and proton reduction sites.35,36 Cocatalysts also improve the stability of photocatalysts, particularly metal sulfides, by extracting photogenerated holes, thereby preventing self-decomposition and supporting oxygen evolution reactions, ultimately boosting overall photocatalytic efficiency.37,38
PSs harvest sunlight and initiate redox reactions for solar fuel generation by absorbing photons and exciting an electron to form PS*. PS* undergoes reductive or oxidative quenching and is regenerated in catalytic reactions, producing solar fuels like hydrogen or reducing CO2. Efficient PSs require long excited state lifetimes for electron transfer and should absorb a broad light spectrum, especially visible light, to optimize sunlight utilization.39–43 SDs are essential in artificial photosynthesis, particularly for water splitting. Acting as electron sources, they sustain electron flow, prevent reverse reactions with oxidized products like oxygen, and ensure continuous hydrogen production. Effective systems require compatible redox potentials between PSs and SDs, accounting for the excited state lifetime and catalytic needs of PSs. SDs are irreversibly oxidized, enabling PS operation but generating chemical waste and requiring replenishment, which reduces sustainability.44–47 PSs also face photobleaching and back electron transfer, causing side reactions and lower efficiency.43 Research aims to develop photocatalysts that drive water splitting independently, eliminating reliance on SDs and PSs, thus improving efficiency and sustainability.
In recent decades, numerous photocatalysts have been explored for water splitting. TiO2, studied since the 1970s, remains a key material due to its abundance, stability, and favorable band alignments. However, its wide band gap (3.2 eV) limits light absorption to the UV region, and issues with conductivity and recombination persist.48 Improvements include band gap modification, nanostructuring, defect control, co-catalyst decoration, and heterojunctions. Other materials like Fe2O3, WO3, ZnO, BiVO4, Cu2O, and CdS offer narrower band gaps (2.0–2.5 eV) but face challenges such as low carrier mobility and crystallographic disorder.49–58 Emerging materials, including graphene-based materials, perovskites (e.g., CH3NH3PbI3, CsPbBr3), and 2D materials (e.g., MoS2, g-C3N4), show promise due to their unique properties. However, they face issues with light absorption, stability, scalability, toxicity, and synthesis.59–63 In addition, covalent organic frameworks (COFs), conjugated porous polymers (CPPs), and porphyrins are promising photocatalysts for water splitting due to their distinct features. COFs offer tunable band gaps, high porosity, and extended π-conjugation for efficient charge separation and light absorption.64,65 CPPs exhibit broad visible-light absorption and delocalized π-electrons, enhancing charge transport.66,67 Porphyrins serve as efficient light-harvesters and electron transfer agents, functioning as both photosensitizers and catalytic sites.68–70 However, metal–organic frameworks (MOFs) surpass these materials with abundant catalytic sites, exceptional structural tunability, and superior photocatalytic performance due to the presence of metal centers, positioning them as frontrunners in water-splitting applications.
MOFs are crystalline structures formed by coordinating organic linkers with metal ions or clusters that offer high porosity and versatility. Constructed from metals like transition metals, lanthanides, and actinides, mixed-metallic MOFs incorporate multiple metals.71 Organic linkers, including carboxylate, azolate, phosphonate, sulfonate, and pyridyl groups, enable diverse structures from 1D chains to 3D networks.72,73 Some representative examples of linkers from each class are depicted in Fig. 4. The general classification of MOFs based on porosity, dimensionality, type of node, type of linker, and topology is presented in Table 1. Certain MOFs have rigid frameworks, ideal for harsh environments, while flexible ones enable reversible gas adsorption.74 High-valent metals and rigid linkers enhance thermal stability, making MOFs adaptable for addressing limitations in conventional materials.
![]() | ||
Fig. 4 Some representative examples of ligands: (a) carboxylate, (b) phosphonate, (c) azolate, (d) sulphonate and (e) pyridyl. |
Classification | Types | Examples | Ref. |
---|---|---|---|
Dimensionality | 1-Dimensional | IMP-27Na | 75 |
MIL-132 | 76 | ||
Mg-MOF-74 | 77 | ||
CAU-50 | 78 | ||
2-Dimensional | ULMOF-1 | 79 | |
AgPb-MOF | 80 | ||
CTH-15 | 81 | ||
COK-47 | 82 | ||
3-Dimensional | MOF-808 | 83 | |
KGF-1 | 84 | ||
CAUMOF-8 | 85 | ||
CdNa-MOF-1 | 86 | ||
Porosity | Microporous (<2 nm) | ZIF-8 | 87 |
MOF-508 | 88 | ||
[Zn(bdc)(ted)0.5]·2DMF·0.2 H2O | 89 | ||
[Cu(INA)2] | 90 | ||
Mesoporous (2–50 nm) | MIL-100(Cr) | 91 | |
ZIF-100 | 92 | ||
MOF-180 | 93 | ||
NOTT-116(PCN-68) | 94 | ||
Type of organic linker | Carboxylate | FeNi-DOBDC | 95 |
CoNi-MOFNA | 96 | ||
Azolate | ZIF-8 | 97 | |
ZIF-67 | 98 | ||
Pyridyl | {[CuII][CuII(pdc)(H2O)]·1.5MeCN·H2O}n | 99 | |
67BPym-MeI | 100 | ||
Sulphonate | PAMPS@MIL-101-SO3H | 101 | |
UiO-66-SO3H | 102 | ||
Phosphonate | IPCE-1Ni | 103 | |
TUB75 | 104 | ||
Type of metal nodes | Single | MIL-53(Fe) | 105 |
Clusters | UiO-66 | 106 | |
NH2-MIL-125 | 107 | ||
Topology | Simple | MOF-5 | 108 |
HKUST-1 | 109 | ||
ZTF-1 | 110 | ||
Complex | PCN-222 | 111 | |
MOF-74 | 112 |
MOFs are renowned for high porosity, vast internal surface areas, and tunable pore sizes, making them ideal for gas storage, separation, and catalysis.113,114 By varying linker lengths, MOFs achieve selective gas capture, such as CO2 from flue gas, and hydrocarbon separation in petrochemical processes.72,115,116 Their modular synthesis enables functional group introduction during or after synthesis, enhancing trapping and catalytic performance.117–119 Many MOFs are biocompatible for drug delivery, while others excel in heterogeneous catalysis due to large surface areas, tunable pore sizes, and catalytic active sites.72,120–123 Catalytic MOFs integrate metal ions, clusters or functionalized linkers, with porosity allowing easy access to active sites.124 MOFs maintain structural integrity after post-synthetic modifications, enhancing versatility.28
As photocatalysts, MOFs absorb light and generate reactive species for pollutant degradation and water-splitting. Their ability to absorb broad-spectrum light supports electron–hole generation and efficient charge separation.125 The first MOF photocatalyst Al-ATA consists of AlO4(OH)2 octahedral clusters linked by 2-aminoterephthalate (ATA). Incorporating Ni(II) enabled Al-ATA to produce hydrogen at 36.0 μmol h−1, establishing it as the first water-splitting photocatalyst.126 Another breakthrough was MIL-125-CoPi–Pt, with cobalt phosphate (CoPi) and Pt cocatalysts, which reached H2 and O2 production rates of 42.33 μmol h−1 and 21.33 μmol h−1, suppressing electron–hole recombination.127 In another instance, MIL-125-NH2 with Pt and RuOx cocatalysts enhanced H2 and O2 production to 85 μmol g−1 and 218 μmol g−1 over 24 h, achieving 0.32% quantum efficiency.128 Introducing defects, such as plasma-treated Ti-oxo clusters in MIL-125-NH2, improved the photocatalytic activity.129
Despite advancements, transition metal MOFs often show low light response, necessitating cocatalysts. Incorporating lanthanides into MOFs has emerged as a solution. Lanthanide-based MOFs (Ln-MOFs) exhibit high coordination numbers, well-defined energy levels, luminescence, and catalytic properties.130–132 Combining lanthanide properties with MOFs opens opportunities for tailored photocatalytic water splitting.133,134 This review explores Ln-MOFs in photocatalytic water splitting, discussing their structural diversity, optical properties, and catalytic efficiencies.
Review | Pros | Cons | Strengths of this review |
---|---|---|---|
Reddy et al.135 | Comprehensive analysis of MOF-based heterogeneous photocatalysts for various applications, including H2 generation | Limited focus on Ln-MOFs and photocatalytic water splitting | Dedicated focus on Ln-MOFs for hydrogen production, filling the gap in prior studies |
Liu et al.136 | Explores photocatalytic hydrogen production across UV, visible, and near-IR regions | Does not specifically highlight Ln-MOFs or modifications enhancing PHE performance | Provides a detailed exploration of the unique properties of Ln-MOFs for enhanced photocatalysis |
Luo et al.137 | Framework for understanding how MOF modifications enhance photocatalytic performance | General MOF focus with limited mention of Ln-MOF-specific applications | Offers specific insights into Ln-MOF modifications and their effects on water splitting |
Nguyen et al.138 | Identifies key achievements and limitations in MOF photocatalysts for water splitting | Does not detail the unique role of lanthanides in MOF-based water-splitting photocatalysis | Highlights the distinctive advantages of lanthanide metals in water-splitting applications |
Xiao et al.139 | Emphasis on heterostructures and interfacial charge transfer for enhanced photocatalysis | Neglects lanthanide-specific systems and applications in photocatalytic water splitting | Provides an in-depth analysis of Ln-MOF-based heterostructures for PHE performance |
Nordin et al.139 | Extensive overview of synthetic methods and functionalization techniques for MOFs | Limited discussion on lanthanides and their specific photocatalytic capabilities | Explores Ln-MOF-specific synthesis and functionalization for optimal PHE performance |
Sun et al.141 | Highlights advances in water splitting and CO2 reduction using MOF-based materials | Minimal focus on lanthanide-based systems and their unique contributions | Offers a comprehensive review of Ln-MOFs specifically for hydrogen production |
Zhang et al.148 | Detailed synthesis methods and applications of RE-MOFs in various fields | Lacks a focused discussion on Ln-MOFs as catalysts in PHE | Exclusively examines Ln-MOFs for PHE |
Shi et al.,144 Fan et al.,145 Meng et al.146 | Highlights general RE-MOF applications in energy and environmental catalysis | Limited insights into lanthanide-specific photocatalytic water-splitting capabilities | Bridges the gap by providing detailed analysis of Ln-MOFs in water-splitting technologies |
Fan et al. explore RE-modified MOFs for photo/electrocatalysis, emphasizing the theoretical advantages of RE elements in MOF modification but lacking specific discussion on Ln-MOFs for water splitting.145 Meng et al. address photocatalytic and electrocatalytic applications of RE-MOFs, including hydrogen evolution and CO2 reduction, but again do not focus on Ln-MOFs in water splitting.146 Zhang et al. review RE-MOF synthesis and photon-related applications, including fluorescence detection and luminescence, emphasizing Ce, Eu, Tb, Yb, and Gd but only briefly addressing water splitting.147 Zhang et al. also summarize the catalytic applications of Ln-MOFs, including photocatalysis, but further exploration of their role in water splitting is warranted.148 This review is one of the few that specifically addresses the role of lanthanides in MOF-based catalysis, but it still leaves room for a more detailed exploration of their use in water splitting. Here is a table summarizing the pros and cons of the previously existing reviews, along with the strengths of the current manuscript.
Given the limited focus on lanthanide-based MOFs in the existing literature, there is a clear need for a dedicated review that comprehensively examines their applications in photocatalytic water splitting. Such a review would not only fill a significant gap in the current body of knowledge but also provide valuable guidance for future research in this promising area. Therefore, this review aims to exclusively discuss the application of various Ln-MOFs and their derivatives for hydrogen production through the PHE process, offering a detailed analysis of the latest advancements in the past 10 years and highlighting potential areas for further exploration. Relevant articles were gathered using the Google Scholar search engine. Additionally, special issues focusing on the catalytic applications of MOF materials were instrumental in identifying related studies.149–151
These light-harvesting properties of Ln-MOFs originate from the lanthanide metals present in the MOFs. These lanthanides display special optical properties owing to the 4d electron layer in the metals, thereby activating the framework under photoexcitation conditions. Additionally, the 4f orbitals, shielded by the filled 5s and 5p subshells, create multiple low-lying empty states in the 4f shell.130 These orbitals are responsible for the unique electronic properties of lanthanides, leading to sharp emission lines and long-lived excited states. These characteristics make lanthanides suitable candidates to strengthen the light-harvesting nature of MOFs. Incorporation of lanthanides into MOFs can effectively modify the bandgap and electronic structure of the material. Furthermore, Ln-MOFs exhibit high photostability, ensuring structural integrity and catalytic activity for prolonged periods of irradiation.158 Ln-MOFs hold great promise in the realm of PHE due to their permanent porosity, impressive structural diversity, high coordination number, and the flexible coordination environment of lanthanides.133,159 One of the key features of these MOFs is their ability to incorporate functional organic ligands, support and convert into various derivatives.160,161 The desirable qualities of Ln-MOFs, such as high surface area, tunable pore size, and chemical and thermal stability, make them excellent candidates for catalysts.148 By leveraging the properties of lanthanides, Ln-MOFs can achieve efficient charge separation and transfer, reduced recombination rates, and thereby increased overall efficiency in photocatalytic water-splitting applications.
This review aims to highlight the performance and stability of various Ln-MOFs in PHE applications, drawing from case studies to illustrate their potential. The graphical overview provided in Fig. 5 encapsulates the topics of discussion that follow.
In 2020, a study by A. Melillo et al. examined the catalytic activities of a series of five UiO-66(M: Zr, Zr/Ti, Zr/Ce, Zr/Ce/Ti, Ce) materials for overall water-splitting. It was found that the activity of the trimetallic MOF was seven times higher than that of its single metal analogue that contained only Zr.157 The band gap of the trimetallic MOF was determined to be 3.10 eV with the help of UV-DRS. This makes UiO-66(Zr/Ce/Ti) a suitable catalyst for water splitting, since at pH 7 the CB energy must be higher than −4.03 eV for the evolution of H2 from water. Among the five members, the catalytic activity for overall water splitting followed the order UiO-66(Zr/Ce/Ti) > UiO-66(Zr/Ti) > UiO-66(Zr/Ce), whose activities were higher than that of the single metal MOF, when irradiated with a xenon laser in the presence of a cut-off filter (λ > 450 nm).157 The quantum yields for the reactions were found to be 0.55, 0.055, and 0.1 at 300 nm, 400 nm and 500 nm, respectively, indicating the highest relative efficiency at 300 nm. However, the maximum amount of hydrogen was generated at 400 nm (Fig. 6a). Reusing the UiO-66(Zr/Ce/Ti) catalyst caused a minor decrease in the initial rate of reaction as well as the final volume of hydrogen produced (Fig. 6b). The use of a sacrificial donor, namely methanol, along with the MOF, resulted in the increase of the amount of hydrogen production by two-fold, up to 390 μmol g−1 in 22 hours. This is much higher than that of individual UiO-66 (∼155 μmol g−1).157
![]() | ||
Fig. 6 (a) Diffuse reflectance UV-Vis spectra of (a) UiO-66(Zr), (b) UiO-66(Zr/Ti), (c) UiO-66(Zr/Ce), (d) UiO-66(Zr/Ce/Ti), and (e) UiO-66(Ce). The inset (on the right) corresponds to a magnification of the 400–650 nm region for (a)–(d). (b) Photocatalytic H2 evolution in the overall water splitting for two consecutive uses of UiO-66(Zr/Ce/Ti): first use (■) and second use (○). Reprinted with permission from ref. 157. Copyright 2020 Elsevier. (c) H2 evolution rates of Pr-MOF-Ru(cptpy)2 catalysts under different conditions. (d) Wavelength dependence of the AQE of 0.5% Pt/Eu-MOF-Ru(cptpy)2. Reprinted with permission from ref. 164. Copyright 2023 ACS. |
Cerium plays a multifaceted role in UiO-66(Zr/Ce/Ti), enhancing photocatalytic activity through improved charge separation, increased light absorption, enhanced structural stability, and synergistic interactions with other metals. The improved photocatalytic efficiency of the MOF was attributed to kinetic factors like charge separation and recombination rather than thermodynamic factors like band gap alignment.157
Following this, in 2023, two MOFs, Eu-MOF-Ru(cptpy)2 and Pr-MOF-Ru(cptpy)2, were synthesized using the ruthenium complex Ru(cptpy)2 as the organic linker, with europium (Eu(III)) and praseodymium (Pr(III)) ions serving as the metal nodes, respectively. To enhance their catalytic performance, platinum (Pt) nanoparticles were photodeposited onto the surfaces of the MOFs, acting as co-catalysts. The catalytic performance was evaluated in a triethanolamine aqueous solution, which served as a sacrificial agent. For Pr-MOF-Ru(cptpy)2, an optimal H2 evolution rate of 268 μmol g−1 h−1 was achieved with a 1.5% Pt loading. However, when ascorbic acid (AA) was used as a sacrificial reagent, the H2 evolution rate significantly increased to 1047 μmol g−1 h−1 (Fig. 6c). This enhancement was attributed to the better matching of the redox potential of AA with the VB of the Pr-MOF, facilitating more efficient electron transfer. In contrast, Eu-MOF-Ru(cptpy)2 exhibited superior photocatalytic performance with a lower Pt loading of 0.5%, achieving an impressive H2 evolution rate of 4373 μmol h−1 g−1 in the presence of ascorbic acid. The apparent quantum efficiency (AQE) of this catalyst was measured at 0.79% at 500 nm, indicating its excellent ability to absorb and utilize visible light for photocatalysis (Fig. 6d).164
The Eu-MOF catalyst also demonstrated remarkable stability; it maintained its catalytic activity over 9 hours of continuous reaction, showing little to no degradation (Fig. 7a). In contrast, the Pr-MOF catalyst exhibited a significant decrease in activity over the same period, likely due to instability in the ascorbic acid solution, which led to the decomposition of the MOF structure (Fig. 7c). These findings highlight the potential of Eu-MOF-Ru(cptpy)2 as a highly efficient and stable photocatalyst for hydrogen production under visible light, with a lower requirement for Pt loading and better performance compared to Pr-MOF-Ru(cptpy)2.164
![]() | ||
Fig. 7 (a) Photocatalytic H2 production amounts in the cycle test and (b) the average rate of H2 evolution during each cycle of 0.5% Pt/EuMOF-Ru(cptpy)2. (c) Photocatalytic H2 production amounts in the cycle test and (d) the average H2 evolution rate during each cycle of 1.5% Pt/Pr-MOF-Ru(cptpy)2. Reprinted with permission from ref. 164. Copyright 2023 ACS. (e) Time course of photocatalytic H2 evolution of CSUST-4 and activated CSUST-4. (f) Energy diagrams of the HOMO and LUMO levels of CSUST-4 and CSUST-Ln (Ln = La, Nd, Eu, Er, Yb). Reprinted with permission from ref. 165. Copyright 2023 Wiley. |
Another recent study by Gu et al. focuses on the PHE activity and stability of a cerium-based MOF called CSUST-4, along with its lanthanide-substituted variants (CSUST-4-Ln, where Ln = La, Nd, Eu, Er, Yb). CSUST-4 was synthesized using a solvothermal method, yielding a 3D porous framework with significant potential for photocatalysis due to its bandgap of 3.04 eV and n-type semiconductor properties. The catalytic activity was assessed under simulated sunlight in the presence of triethylamine (TEOA) as a sacrificial agent and Pt as a co-catalyst. Activated CSUST-4 showed an improved hydrogen evolution of 41 μmol g−1 over 6 hours, compared to 34 μmol g−1 for the as-synthesized version (Fig. 7e).165
This improvement is attributed to the exposure of open Ce(III) sites, which enhanced interactions with water molecules, thereby boosting catalytic efficiency. When examining the stability and performance of lanthanide-substituted variants, CSUST-4-Nd demonstrated the highest hydrogen evolution (71 μmol g−1 in 6 hours), followed by CSUST-4-Er (61 μmol g−1). These MOFs outperformed the base CSUST-4, while CSUST-4-Eu, CSUST-4-La, and CSUST-4-Yb exhibited lower activities, likely due to their limited visible light absorption and differing electronic properties. The band gap values of all the CSUST MOFs studied are provided in Fig. 7f. The study highlights that the catalytic activity and stability of these MOFs are closely linked to the specific lanthanide ion used, with Nd and Er proving most effective for enhancing hydrogen production through water splitting.165
In 2019, Huang et al. developed a new H2BPDYC–Ce complex (UiO-67-Ce) by incorporating Ce(IV) into UiO-67 at a 0.02 Ce/Zr ratio and replaced the H2BPDC ligands with H2BPYDC. As a result, UiO-67-Ce exhibited a much higher PHE rate compared to UiO-67, with UiO-67-Ce achieving 269.6 μmol g−1 h−1, which is more than ten times higher than UiO-67's rate of 26.78 μmol g−1 h−1 under identical experimental conditions. This significant enhancement in photocatalytic activity is attributed to the introduction of the Ce(IV) ion and the BPYDC-Ce ligand into the UiO-67 framework, which created new active sites and promoted efficient energy transfer processes. Despite this remarkable increase in activity under UV/Vis irradiation, neither UiO-67 nor UiO-67-Ce displayed any significant photocatalytic hydrogen production activity under visible light alone (λ > 400 nm), indicating that their performance is primarily driven by UV light (Fig. 8a). The ligand BPYDC-Ce displayed a much stronger UV intensity than BPDC, indicating that Ce coordinates with the N atoms in the ligand and not the carboxylate groups. An EPR signal with a g-value around 2.002, indicating Zr3+ formation via the LMCT process, is observed in both UiO-67-Ce and UiO-67. The signal is weaker in UiO-67-Ce, and no Zr3+ signal is detected in the dark for either material (Fig. 8b). This suggests that the LMCT process under UV/Vis light is reduced in UiO-67-Ce due to BPYDC-Ce. The introduction of BPYDC-Ce is said to weaken LMCT and promote energy transfer from BPDC to BPYDC-Ce, enhancing catalytic H2 evolution.168
![]() | ||
Fig. 8 (a) UV/Vis DRS of UiO-67 and UiO-67-Ce and (b) EPR spectra of UiO-67 and UiO-67-Ce in the presence of UV/Vis illumination. Reprinted with permission from ref. 168. Copyright 2019 Elsevier. High-resolution and separated peak curve XPS spectra of (c) O 1s and (d) C 1s. (e) Collaborative photocatalytic performance of H2 liberation over Pt-loaded Ce-TBAPy in the CH3CHO system under an N2 atmosphere. (f) Time course of H2 liberation and evolution rate curve (inset) of samples. Reprinted with permission from ref. 169. Copyright 2022 Elsevier. |
The stability of UiO-67-Ce during photocatalytic reactions was confirmed by consistent Ce/Zr ratios before and after testing, with no Ce ion leaching observed, indicating the catalyst maintained its structural integrity. However, after 6 hours of continuous UV/Vis irradiation, the H2 evolution rate declined, suggesting reduced stability. XRD and SEM analyses revealed this decline was due to decreased crystallinity and morphological changes, with the cubic particles becoming irregular and larger. This structural degradation was likely caused by hydroxyl nucleophile attack and mechanical stress during the prolonged reaction.168
In the year 2022, Yang et al. studied the photocatalytic performance of a novel Ce-based MOF under visible light. The study presents the synthesis of a cerium-based metal–organic framework (Ce-MOF) named Ce-TBAPy using a solvothermal method. XRD confirmed its high crystallinity, showing Ce atoms coordinated with oxygen from the ligands. The material displayed a cubic morphology with strong Ce–O interactions, contributing to effective charge separation. Characterization techniques, including SEM, FT-IR, and NMR, revealed uniform elemental distribution and high thermal stability. XPS analysis showed a high concentration of Ce3+, linked to oxygen vacancies that enhance photocatalytic performance (Fig. 8c and d). UV-Vis DRS indicated visible light absorption at 406 nm, with a band gap of 2.75 eV. The material demonstrated efficient electron–hole pair separation and improved photoreduction performance, particularly after loading a Pt co-catalyst. DFT calculations confirmed that the –COO–Ce functional groups facilitate electron transfer and inhibit recombination, boosting photocatalytic efficiency.
In terms of photocatalytic performance, Ce-TBAPy demonstrated exceptional hydrogen production under visible light irradiation, achieving hydrogen evolution 3.94 times higher than that of the untreated ligand (Fig. 8e). The material maintained its activity after three rounds of testing, indicating excellent recyclability and stability (Fig. 8f). The bandgap value of Ce-TBAPy (2.75 eV) and its conduction and VB positions made it well-suited for water reduction, with efficient electron transfer facilitated by the Pt co-catalyst. The overlapping structure of polycyclic aromatic conjugated polymers in the ligand further promoted rapid migration of photo-induced electrons, contributing to the overall efficiency of the photocatalytic process.169
An organic dye-like MOF was reported by X. Sun et al. for the first time in 2015, where 3,3′,5,5′-azobenzene tetracarboxylic acid was used as a linker in a Gd-MOF loaded with Ag nanoparticles. The framework was found to be stable in the pH range of 3 to 14 for up to 48 h. UV-Vis spectroscopy and electrochemical analysis revealed a broad absorption band and a favorable energy gap (2.35 eV) for photocatalytic activity. The solid-state cyclic voltammetry showed a suitable redox potential for proton reduction, confirming the potential of the MOF for hydrogen production (Fig. 9a).
![]() | ||
Fig. 9 (a) CV curves of the Gd-MOF. (b) Photocatalytic H2 production under UV-Vis light irradiation over H4abtc, Gd-MOF and Ag(X)/Gd-MOF. Reprinted with permission from ref. 175. (c) The pH stability of the dye-based Dy-MOF. (d) The differences in hydrophilicity between the dye-based Dy-MOF and H4abtc ligand using contact angle measurements. Reprinted with permission from ref. 176. Copyright 2015 Royal Society of Chemistry. |
In photocatalytic tests, the Gd-MOF produced H2 at a rate of 7.71 μmol h−1 g−1 under UV-Vis light without a co-catalyst, comparable to that of UiO-66. Adding Ag as a co-catalyst significantly enhanced the performance. The Ag-loaded Gd-MOF (1.5 wt%) achieved a hydrogen production rate of 10.6 μmol h−1 g−1, about 1.5 times higher than that of the Gd-MOF alone (Fig. 9b). This improvement is due to the role of Ag in enhancing charge separation and light absorption through surface plasmon resonance. The photocatalytic activity is maintained even after 5 hours of reaction, and the structure remains stable as confirmed by XRD. The electrochemical impedance spectra and luminescence studies further support the enhanced charge transfer and reduced recombination rates with Ag, leading to superior photocatalytic performance.175
In another similar work in 2018, Yu et al. synthesized a dye-based Dy-MOF through the reaction of DyCl3·6H2O and Na4abtc. The crystal structure, determined via X-ray analysis, exhibits a novel 3D monoclinic framework with dimensions accommodating water channels. The dye-based Dy-MOF demonstrates impressive photocatalytic performance and stability. Under UV-Vis light irradiation, it achieves a high hydrogen production amount of 107.65 μmol g−1 over 5 hours, equivalent to 21.53 μmol g−1 h−1. This rate surpasses that of many MOF-based catalysts without dye-like ligands, such as UiO-66 and ZIF-8, though it is somewhat lower compared to some porphyrin-based MOFs. The enhanced photocatalytic efficiency of the Dy-MOF is attributed to the inclusion of the dye-like ligand (H4abtc), which significantly improves the light-harvesting capability. The UV-Vis absorption spectrum of the Dy-MOF shows broad light absorption with an edge above 570 nm, corresponding to a π–π* transition in organic ligands, indicating effective light capture. Additionally, the incorporation of 0.8% Ag as a co-catalyst markedly boosts activity. The Ag acts as an electron collector, enhancing charge separation and reducing electron–hole recombination, which further increases the hydrogen production rate. In terms of stability, the dye-based Dy-MOF exhibits robust thermal stability, maintaining its structural integrity up to 400 °C before collapsing, indicating its suitability for typical photocatalytic conditions.176
Furthermore, the material demonstrates excellent water stability, retaining its structural characteristics when soaked in solutions with pH values ranging from 3 to 14 for 48 h (Fig. 9c). This broad pH stability is essential for practical photocatalytic applications. Cycling tests reveal that the Dy-MOF maintains consistent hydrogen production performance up to the third cycle, with no significant loss of activity. Post-reaction analysis by X-ray diffraction (XRD) and FT-IR confirms that the material's structure remains largely unchanged, underscoring its durability. Additionally, the Dy-MOF's lower contact angle of 21.5°, compared to 57.0° for the H4abtc ligand, indicates higher hydrophilicity (Fig. 9d). This increased hydrophilicity enhances water adsorption, which is beneficial for efficient proton reduction during photocatalysis.176
Increasing the % weight of the MOF in the ZIS composite showed a decrease in the intensity of photoluminescence, while ZIS/1% (Tm,Gd)-BTC/Pt exhibited the highest intensity of luminescence among the samples analyzed. Pt nanoparticles were shown to decrease the rate of recombination and thus the photoluminescence; however, the addition of (Tm, Gd)-BTC had only a minimal effect in decreasing the luminescence, suggesting a complex interplay in the luminescence behaviour (Fig. 10a). Partial decomposition of ZIS and loss of Pt were observed post-reaction through XRD and XPS studies. The efficiency of catalytic hydrogen generation was analysed using various samples in the aqueous solution of Na2SO3 and Na2S (sacrificial electron donors). The samples used were pure ZIS, ZIS/Pt, ZIS/x% (Tm,Gd)-BTC and ZIS/x% (Tm,Gd)-BTC/Pt. Pure ZIS and ZIS modified with (Tm,Gd)-BTC at 1%, 5% and 10% by weight showed no activity in hydrogen generation, while the samples with Pt (ZIS/Pt and ZIS/x% (Tm,Gd)-BTC/Pt) were active in hydrogen production. ZIS modified with 1% (Tm,Gd)-BTC/Pt produced hydrogen at a rate of 1945.5 μmol g−1 h−1 when irradiated with UV-Vis light, which is much higher than that of pure ZIS, which produced 49.07 μmol g−1 h−1. Increasing the amount of the Ln-MOF, however, decreased the photoactivity of the sample. The highest hydrogen production rate was observed in the presence of both the MOF and the Pt NPs, i.e., for 1 g L−1 ZIS/1% (Tm,Gd)-BTC/Pt where it produced a maximum of 2975 μmol g−1 h−1 of hydrogen when irradiated with UV-Vis. However, under visible light irradiation, its efficiency decreased as it produced only 328.17 μmol g−1 h−1 of hydrogen, which is lower than that of ZIS/Pt which produced 608.32 μmol g−1 h−1. When tested for stability, ZIS/1% (Tm,Gd)-BTC/Pt displayed the highest efficiency in the second cycle, producing 837 μmol g−1 h−1 of hydrogen under visible light (Fig. 10b). The efficiency is said to decrease in the subsequent cycles.183
![]() | ||
Fig. 10 (a) Amount of hydrogen produced for all obtained samples under UV-Vis irradiation (photocatalyst concentration: 1.25 g L−1) and (b) amount of hydrogen produced in five photocatalytic cycles for ZIS/1% (Tm,Gd)-BTC/Pt (1 g L−1) under visible light irradiation (λ > 420 nm). Reprinted with permission from ref. 183. (c) Photonic energy map of the photocatalysts. (d) PXRD patterns of Pr–NO2–TPTC/CZS (1![]() ![]() |
Since ZIS semiconductors face challenges such as low solar utilization, rapid recombination of carrier charges, and low quantum yields, binary metal sulfides such as CdxZn1−xS, especially Cd0.2Zn0.8S, have shown promising stability, excellent carrier transport capacity, and adjustable band structure. Cd0.2Zn0.8S has been identified to have the highest photocatalytic activity among different compositions of CdxZn1−xS;185 however, its application is limited because of the low separation and migration ability of photogenerated electron–hole pairs.
Recently, in 2024, a study by T.-T. Li et al. investigated the catalytic performance of Cd0.2Zn0.8S/Pr–NO2–TPTC. The lanthanide MOF comprises binuclear clusters [Pr2(COO)4] connected by multiple NO2–TPTC4− ligands, creating a three-dimensional coordination framework. The NO2–TPTC4− ligands are said to form hydrogen bonds with the lattice water in the solvent channels of the framework, thus enhancing absorption capacity and promoting water splitting to improve catalytic hydrogen production. The UV-Vis DRS shows absorption bands for Pr–NO2–TPTC around 450 nm and for Cd0.2Zn0.8S around 500 nm. When combining the Pr–NO2–TPTC and the Cd0.2Zn0.8S in various ratios, especially with a higher percentage of Pr–NO2–TPTC, a red shift in the absorption peak is seen, indicating improved light absorption and narrower band gap width.184
Pristine Pr–NO2–TPTC shows no catalytic performance; however, pure Cd0.2Zn0.8S produces 2712 μmol g−1 h−1 of hydrogen. The composite Pr–NO2–TPTC/CZS in a 1:
1 ratio achieves the highest hydrogen production rate of 6321 μmol g−1 h−1, which is 2.33 times higher than that of pure Cd0.2Zn0.8S. Other ratios, namely, 2
:
1 and 4
:
1, showed lower rates of hydrogen generation. The photonic energy map of the compounds is shown in Fig. 10c. The rate of hydrogen produced by Cd0.2Zn0.8S was found to be 2092 μmol g−1 h−1 under visible light. Pr–NO2–TPTC/CZS in a 1
:
1 ratio produced a slightly lower but still significant amount of hydrogen under visible light (5777 μmol g−1 h−1). The catalytic performance of Pr–NO2–TPTC/CZS (1
:
1) was maintained for three consecutive cycles, and there was no significant deterioration of the materials seen (Fig. 10d). The Mott–Schottky analysis revealed a positive slope indicating that the material is of n-type. In the photoluminescence studies, the weakest emission at 560 nm, which corresponded to Pr–NO2–TPTC/CZS(1
:
1), showed a low extent of electron–hole recombination. The transient photocurrent experiments displayed a high response for the material, suggesting enhanced separation of photogenerated carriers. The material also presented reduced charge transfer resistance according to EIS. A heterojunction is said to form between Cd0.2Zn0.8S and Pr–NO2–TPTC, which promotes separation of the photogenerated charges and prolongs the carrier lifetime. UV-Vis light irradiation of both Pr–NO2–TPTC and Cd0.2Zn0.8S leads to the generation of electron–hole pairs. The electrons from the LUMO of Pr–NO2–TPTC migrates to the CB of Cd0.2Zn0.8S, thus reducing H+ in the water to produce hydrogen. While the holes move from the VB of Cd0.2Zn0.8S to the HOMO of Pr–NO2–TPTC and are consumed by the sacrificial agents (Na2S and Na2SO3) present in the medium. The heterojunction that is present in this composite inhibits electron–hole recombination, and hence there are a greater number of electrons available for the reduction of protons.184
The sample demonstrates excellent stability under visible light irradiation for over 24 h across four cycles (Fig. 11a). The slight decrease in activity is due to TEOA consumption. Overall, 0.015Ce–C3N4 shows robust and consistent performance, making it suitable for practical applications. The modifications using NaHCO3 and cellulose acetate also lead to significant improvements in photocatalytic activity. Cellulose acetate-modified samples, in particular, exhibit increased visible light absorption and better light-harvesting efficiency, contributing to improved photocatalytic performance. On the other hand, the incorporation of fluorine shifts both the VB and CB to higher energy levels, which enhances the photocatalytic activity by improving charge carrier dynamics (Fig. 11b). The stability of modified samples, including those doped with NH4F, NaHCO3, and cellulose acetate, aligns with the trends seen in base Ce–C3N4. These modifications generally enhance morphological stability, with NH4F creating finer fibres and cellulose acetate and NaHCO3 improving structural robustness. Consequently, these modifications contribute to more consistent and reliable photocatalytic activity186.
![]() | ||
Fig. 11 (a) The 24 h cycling measurement of H2 evolution from 0.015Ce–C3N4. (b) Photocatalytic H2 evolution of CA-CN, NF-CN, and NHC-CN. Reprinted with permission from ref. 184. Copyright 2019 MDPI. (c) Gibbs free energy profile of the hydrogen evolution reaction for CeO2@N,S–C HN (blue), N,S–C (black), and CeO2 with the (111) crystal plane (pink). (d) The 40 h cycling measurement of H2 evolution from CeO2@N,S–C HN. Reprinted with permission from ref. 184. Copyright 2019 ACS. |
In the following year, Hao et al. successfully synthesized N,S-codoped C-encapsulated CeO2 with a hinge-like structure through the thermal decomposition of sulfanilic acid-modified Ce-based MOFs. The CeO2@N,S–C HN catalyst demonstrated outstanding photocatalytic performance in PHE reactions. Its mass-normalized hydrogen production rate reached 555 μmol h−1 g−1, surpassing those of CeO2@C HN (405 μmol h−1 g−1), CeO2 HN (325 μmol h−1 g−1), and commercial CeO2 (195 μmol h−1 g−1). The Gibbs free energy profile of the PHE reaction for all three compounds is given in Fig. 11c. This exceptional activity is largely due to the combination of the N,S-codoped carbon layer and the hinge-like porous structure of the catalyst. The N,S-codoped carbon enhances visible light absorption and facilitates efficient separation and transport of photogenerated charge carriers. The hinge-like structure further improves light trapping and enhances the photocatalytic process by allowing multiple reflections and increasing the interaction surface. Regarding stability, CeO2@N,S–C HN exhibited impressive performance consistency over multiple cycles (Fig. 11d). In PHE tests under simulated sunlight irradiation, the catalyst maintained high activity over four cycles, with no significant loss in performance.
The stability was assessed through repeated cycling, and the catalyst showed negligible deactivation, maintaining its structural and compositional integrity throughout. Characterization techniques, such as XRD, SEM, and EDX, confirmed that the catalyst retained its morphology and elemental distribution after several cycles. This robust stability is attributed to the effective encapsulation of CeO2 nanoparticles by the N,S-doped carbon layer, which protects the core material from degradation and preserves its catalytic properties over extended use. Overall, the CeO2@N,S–C HN catalyst combines high photocatalytic efficiency with long-term stability, making it a promising candidate for sustainable photocatalytic applications.187
For the [Fe–Fe]-based catalyst (C1), the addition of NiPr2EtH·OAc as a sacrificial electron donor led to increased hydrogen production rates with higher catalyst concentrations. Initially, hydrogen production rates showed first-order dependence on C1 concentration (Fig. 12a), but this linear increase stalled at concentrations above 10 mM due to the poor solubility of C1. After 6 hours of irradiation, a notable drop in hydrogen production indicated potential decomposition of a system component. However, reintroducing C1 and NiPr2EtH·OAc to the filtered MOF successfully resumed hydrogen production. Over a 40-hour period with four rounds of continuous irradiation, a total of 15 mL of hydrogen was produced. In contrast, the cobalt-based catalyst [Co(bpy)3]Cl2 (C2) demonstrated significant quenching of the emission of Gd-TCA but showed notable catalytic activity. Under alkaline conditions, Gd-TCA with C2 achieved a turnover frequency of 320 h−1 per molecule of C2 within the first hour and a quantum yield of 0.21%. While the hydrogen evolution rate increased with C2 concentration up to 50 μM, further increases did not enhance the rate linearly, likely due to catalyst decomposition (Fig. 12b). As with C1, the system's activity was restored with the addition of fresh C2 and base. This system also demonstrated impressive longevity, producing 22 mL of hydrogen over 20 hours and five rounds.
![]() | ||
Fig. 12 (a) H2 evolution of Gd-TCA (1 mg) in 5 mL of solution containing NiPr2EtH·OAc (0.8 M) and C1 with various concentrations. (b) H2 evolution of Gd-TCA (1 mg) in 5 mL of a solution containing Et3N (2.5%) and C2 ([Co(bpy)3]Cl2) with various concentrations. Reprinted with permission from ref. 188. Copyright 2016 ACS. (c) Time-dependent PHE TONs of Ce6-BTB-Ir and Ce6-BTB-Ru along with homogeneous controls. (d) PXRD patterns of Ce6-BTB (red), Ce6-BTB-Ir (blue), Ce6-BTB-Ru (green), and Ce6-BTB-Ir after reaction (purple), and Ce6-BTB-Ir after reaction (khaki) in comparison to that simulated for Hf6-BTB MOL (black). Reprinted with permission from ref. 134. Copyright 2020 ACS. |
Additionally, Gd-TCA was utilized in the form of films supported by α-Al2O3, known for its high affinity towards carboxylic groups. The initial hydrogen production rate from a 1.5 × 0.5 cm2 film was 1.71 mL h−1, yielding 3.8 mL of hydrogen in the first 5 hours under alkaline conditions with 50 μM C2. The film-based system maintained hydrogen production for over 40 hours, generating a total of 33.5 mL of hydrogen. This study marks the first reported use of MOF films for the photochemical reduction of water, highlighting the exceptional performance and stability of Gd-TCA framework in both powder and film forms.188
In addressing the limitations of traditional MOFs in photocatalysis, such as light scattering at the nanoscale and inefficient light penetration in bulk forms, researchers have developed metal–organic layers (MOLs). MOLs, as monolayer versions of MOFs, offer improved photocatalytic performance by reducing light scattering and enhancing diffusion of reaction components. This innovation aims to overcome the constraints imposed by MOF symmetry and channel diffusion issues. The first Ce-based MOL was successfully synthesized and studied by Song et al. in 2020. The SBUs are made of Ce6 clusters, and they are linked using BTB molecules.134 The MOLs are capped with photosensitizing molecules such as [(HMBA)Ir(ppy)2]Cl and [(HMBA)Ru(bpy)2]Cl2. The study evaluates the PHE activities of two metal–organic layers (MOLs): Ce6-BTB-Ir and Ce6-BTB-Ru. These MOLs were tested in an oxygen-free acetonitrile solution with acetic acid as the proton source and 1,3-dimethyl-2-phenyl-2,3-dihydro-1H-benzo[d]imidazole (BIH) as the sacrificial agent. The performance was assessed by quantifying hydrogen production through gas chromatography. Both Ce6-BTB-Ir and Ce6-BTB-Ru exhibited impressive PHE activities, with turnover numbers (TONs) of 1357 and 484, respectively (Fig. 12c), following 72 hours of photoirradiation using a solid-state plasma light source. The apparent quantum yields were 4.8% for Ce6-BTB-Ir and 3.8% for Ce6-BTB-Ru. These results highlight the superior photocatalytic performance of these MOLs compared to their homogeneous counterparts, which demonstrated significantly lower TONs.134
Stability was a crucial aspect of the study. Both Ce6-BTB-Ir and Ce6-BTB-Ru maintained their structural integrity after PHE, as evidenced by consistent PXRD patterns (Fig. 12d) and HRTEM images, with less than 3% leaching of Ce into the solution. The MOLs also retained their photocatalytic activity over at least three consecutive runs, demonstrating their durability under reaction conditions. This structural stability and enduring activity underscore the effectiveness of MOLs in photocatalytic applications.134
The summary of all Ln-MOFs discussed above is given in Table 3. Apart from all these studies reported regarding the efficiency of Ln-MOFs in photocatalytic water-splitting processes, various computational studies suggest that Ln-MOFs are promising candidates to be employed in photocatalytic water-splitting. For instance, in 2020, Anderson et al. presented two series of lanthanide-based MOFs (Ln-SION1 and Ln-SION-2) and found that the Ln-SION1 series shows photoconductivity due to its desirable orbital structure. They conclude that the latter members of the series have the potential to act as photocatalysts for water-splitting, with the help of a photocatalyst.189 Additionally, Hidalgo-Rosa et al., in 2023, discussed the significance of functional groups on enhancing the light-harvesting nature of rare earth MOFs and suggested that these materials could show photocatalytic properties upon solar light irradiation.190 All these studies emphasize that numerous opportunities lie ahead of researchers in the field of photocatalytic water-splitting when it comes to lanthanide-based MOFs. Furthermore, their structural tunability and reusability align with the goals of sustainable hydrogen production. However, economic feasibility remains a significant challenge. The high cost of lanthanides, due to their scarcity and complex extraction processes, and the intricate synthesis methods required for Ln-MOFs make them expensive to produce. Scaling up production while maintaining performance is also challenging, further hindering their large-scale adoption. Additionally, these materials face competition from more cost-effective alternatives such as transition metal-based catalysts like TiO2 or ZnO which are widely available and easy to produce. Real-world deployment of Ln-MOFs requires further validation of their efficiency and stability under practical conditions, as laboratory performance may not directly translate to operational environments. Despite these challenges, strategies such as materials optimization—using mixed-metal MOFs or doping to reduce lanthanide content—can lower costs while maintaining functionality. Advances in synthesis techniques, including greener and scalable methods, along with recycling and recovery of lanthanides from spent MOFs, could further enhance economic viability. As global demand for sustainable energy solutions grows, increased research and industrial collaboration may drive cost reductions and facilitate commercialization. Although currently less economically competitive, Ln-MOFs remain a promising avenue for green hydrogen production with further technological and economic advancements.
S. no. | Ln-MOF photocatalyst | Surface area of catalyst (m2 g−1) | SDs and their concentration | Light sources | PHE rate (μmol g−1 h−1) | TON | AQE | Stability | Ref. |
---|---|---|---|---|---|---|---|---|---|
1 | UiO-66(Ce/Zr/Ti) | 1019 | Methanol | 150 mW cm−2 Xe lamp (λ > 450 nm) | 17.7 μmol g−1 | — | 5.5% | Stable up to 2 cycles | 157 |
2 | 0.5%Pt/Eu-MOF-Ru(cptpy)2 | 4.0 | AA (0.1 M) | 300 W Xe lamp (λ ≥ 420 nm) | 4373 μmol g−1 | — | 0.79% | Stable up to 3 cycles (9 h) | 164 |
3 | 1.5%Pt/Pr-MOF-Ru(cptpy)2 | 9.9 | AA (0.1 M) | 300 W Xe lamp (λ ≥ 420 nm) | 1047 μmol g−1 | — | — | Not very stable (3 h) | 164 |
4 | CSUST-4-Nd | — | TEOA (0.65 M) | 300 W Xe lamp | 71 μmol g−1 | — | — | Moderate thermal stability (up to 350 °C) | 165 |
5 | CSUST-4-Er | — | TEOA (0.65 M) | 300 W Xe lamp | 61 μmol g−1 | — | — | 165 | |
6 | Activated CSUST-4 | — | TEOA (0.65 M) | 300 W Xe lamp | 41 μmol g−1 | — | — | 165 | |
7 | UiO-67-Ce | 1545 | Methanol | 300 W Xe lamp | 269.6 μmol g−1 | — | — | Stable up to 6 h | 168 |
8 | Ce-TBAPy | — | TEOA (0.65 M) | 300 W Xe lamp (λ ≥ 420 nm) | 375.1 μmol g−1 | — | — | Good stability and recyclability up to 12 h | 169 |
7 | Ag(1.5)/Gd-ABTC | — | TEOA (∼3 M) | 300 W Xe lamp | 10.6 μmol g−1 | — | — | Stable for 48 h at pH 3–14 | 175 |
8 | 0.8Pt/Dy-ABTC | — | TEOA (2.26 M) | 300 W Xe lamp (λ > 320 nm) | 21.5 μmol g−1 | — | — | 176 | |
9 | ZIS/1% (Tm,Gd)-BTC/Pt | — | Na2SO3/Na2S (0.25 M) | 300 W Xe lamp (λ ≥ 420 nm) | 1945.5 μmol g−1 | — | — | Poor stability | 183 |
10 | Pr–NO2–TPTC/CZS (1![]() ![]() |
3.4480 | Na2SO3/Na2S | Full light | 6321 μmol g−1 | — | — | Stable up to 12 h | 184 |
11 | CeO2@N,S-C HN | 26.9 | — | Full light | 555 μmol g−1 | — | — | Good stability up to 40 h | 186 |
12 | Gd-TCA/C1 | — | Nipr2EtH·OAc (0.8 M) | 500 W Xe lamp | 16.7 μmol g−1 | — | — | Poor stability | 188 |
13 | Gd-TCA/C2 | — | Et3N (0.179 M) | 500 W Xe lamp | 49.1 μmol g−1 | — | 0.21% | Poor stability | 188 |
14 | Gd-TCA film | — | Et3N (0.179 M) | 500 W Xe lamp | 76.3 μmol g−1 | — | Stable up to 40 h | 188 | |
15 | Ce6-BTB-Ir | — | BIH | 13.9 W 350–700 nm solid-state plasma light source | — | 1357 | 4.8% | Good stability | 134 |
16 | Ce6-BTB-Ru | — | BIH | 13.9 W 350–700 nm solid-state plasma light source | — | 484 | 3.8% | Good stability | 134 |
Ln-MOFs have gained significant attention in photocatalysis, particularly for hydrogen production via water splitting, due to their unique ability to form mixed-metal frameworks without altering the MOF structure. This flexibility is made possible by the similar crystal structures of adjacent rare earth ions, which allows for the introduction of active sites and the enhancement of catalytic performance through electron relay between different metal nodes. The result is a material that can be finely tuned for optimal photocatalytic efficiency. Recent advancements in the field have demonstrated that trimetallic Ln-MOFs, which integrate three different metals into a single framework, can achieve a significant boost in catalytic activity compared to their single-metal counterparts. This improvement is primarily due to the broader absorption spectrum and reduced band gap that these mixed-metal systems offer, enabling more efficient light harvesting and energy transfer. Additionally, lanthanide doping, particularly in combination with Pt co-catalysts, has proven to be an effective strategy for further enhancing the photocatalytic efficiency and stability of these materials. The introduction of Pt helps to improve charge separation and transfer, leading to more efficient hydrogen production. To overcome traditional limitations of MOFs, such as light scattering and inefficient light penetration, researchers have developed MOLs. These thinner materials, combined with photosensitizing molecules, exhibit high PHE activities while maintaining their structural integrity over multiple cycles. Gd-based Ln-MOFs have also shown promise in this area, demonstrating effective electron transfer and strong photocatalytic activity, resulting in sustained hydrogen production over extended periods. Ln-MOFs have also been used as sacrificial templates to create more efficient photocatalysts. These materials exhibit enhanced visible light absorption and charge separation, leading to superior photocatalytic activity and stability. Furthermore, the combination of Ln-MOFs with catalytic semiconductors has shown significant potential in further enhancing hydrogen production. Ln-MOF–semiconductor composites take advantage of the synergistic effects of both materials, resulting in improved light absorption, charge separation, and overall photocatalytic efficiency. In addition, numerous studies support the fact that lanthanide-based MOFs show efficient LMCT and upconversion properties. Nd based complexes and MOFs have shown photon upconversion properties through which they emit blue light.191–193 Up-conversion displayed by lanthanides represents a powerful approach to enhance photocatalytic water-splitting.
Overall, incorporation of lanthanides into frameworks increases their photocatalytic activity, somewhere from 2.3 in UiO-67-Ce to 10 times in Pr–NO2–TPTC/CZS (1:
1), by altering their band gap values and hence improving light absorption. Among all the lanthanides discussed in this review, Ce, Eu, Pr, Tm and Gd outperform other lanthanides by improving charge separation and light absorption, owing to their unique electronic structure. Although almost all of the studies involve the use of SDs, Ce-doped g-CN seems to have performed the best in our opinion, exhibiting an excellent hydrogen production rate in the absence of an SD and appreciable stability up to 40 h; in addition to Ce-doped g-CN, Pr–NO2–TPTC/CZS (1
:
1) and Eu-MOF-Ru(cptpy)2 also offer notable efficiency and hence can be used in practical applications. The incorporation of dye-like ligands, multiple lanthanides, and semiconductors are indeed innovative approaches to enhance the PHE performance of Ln-MOFs.
With all this being said, future research on Ln-MOFs should focus on several key areas to fully unlock their potential in photocatalysis, particularly for hydrogen production. One critical area is enhancing the quantum efficiency of Ln-MOFs through optimizing photon absorption, energy transfer processes, and minimizing non-radiative losses, which will be crucial for boosting the overall photocatalytic performance of these materials. Exploring new lanthanide elements and their combinations within MOFs is another promising avenue. Less commonly studied lanthanides could offer unique optical and electronic properties, potentially leading to improved band gaps and more efficient energy transfer mechanisms. The creation of mixed-metal MOFs, leveraging the identical crystal structures of adjacent lanthanide ions, allows for the introduction of active sites and enhances catalytic performance through electron relay between different metal nodes. The integration of Ln-MOFs with emerging technologies, such as artificial photosynthesis and solar fuel production, could open new avenues for clean energy applications. However, most of the Ln-MOFs still face challenges in effectively utilizing sunlight. Most of the Ln-MOFs reported derive their photocatalytic activity to a major extent from UV light that constitutes only about 5%. Hence the visible-light response of these catalysts has to be improved. Furthermore, improving the stability and durability of Ln-MOFs under prolonged irradiation and harsh reaction conditions remains a challenge. Future research should focus on developing more robust materials that maintain their performance over extended periods, addressing issues like structural degradation and the stability of doped and composite MOFs. Furthermore, with the advancements in artificial intelligence (AI) technologies, machine learning (ML) has become a great tool with vast applications. ML has the potential to revolutionize the design of Ln-MOFs for PHE. By rapidly predicting key properties such as band gaps, charge carrier mobility, and photocatalytic activity, ML models can accelerate the screening and optimization of Ln-MOFs. These models can be trained on existing data from quantum chemical calculations like density functional theory or experimental results, allowing researchers to explore vast compositional spaces and identify the most promising materials. By integrating ML with high-throughput experiments, materials synthesis, and computational chemistry, researchers can significantly reduce the time and cost involved in discovering and optimizing new Ln-MOFs for hydrogen production. This interdisciplinary approach fosters the development of high-performance photocatalysts, making ML an essential tool in advancing sustainable energy solutions. Finally, assessing the environmental and economic impacts of Ln-MOFs in large-scale applications will be crucial for ensuring their sustainability and informing their adoption in practical systems for energy and environmental applications. Developing frameworks resistant to hydrolytic and photochemical degradation will enhance stability. Replacing expensive co-catalysts like Pt with earth-abundant alternatives (e.g., Ni, Co) is crucial for cost reduction. Research into dye-based MOFs and composites can improve visible-light activity, while scalable and cost-effective synthesis methods will facilitate the transition from lab-scale research to industrial applications. Continued research and development in this field, focusing on new materials, synthesis methods, and real-world applications, will be crucial for unlocking the full potential of Ln-MOFs and addressing global energy and environmental challenges.
AA | Ascorbic acid |
Ag | Silver |
Al | Aluminium |
AQE | Apparent quantum efficiency |
ATA | 2-Aminoterephthalate |
Au | Gold |
BiVO4 | Bismuth vanadate(V) |
BPDC | Biphenyl-4,4′-dicarboxylic acid |
Bpy | 2,2′-Bipyridine |
BPYDC | 2,2′-Bipyridine-4,4′-dicarboxylate |
BTC | Benzene-1,3,5-tricarboxylic acid |
CB | Conduction band |
CdS | Cadmium(II) sulfide |
Ce | Cerium |
CeO2 | Cerium(IV) oxide |
CH3NH3PbI3 | Methylammonium lead iodide |
CoPi | Cobalt phosphate |
COF | Covalent organic framework |
CPP | Conjugated porous polymer |
CSUST | Changsha University of Science and Technology |
Cu | Copper |
Cu2O | Copper(I) oxide |
CZS | Cadmium zinc sulfide |
DyCl3·6H2O | Dysprosium(III) chloride |
Eu | Europium |
Fe2O3 | Iron(III) oxide |
Gd | Gadolinium |
g-C3N4 | Graphitic carbon nitride |
H3TCA | Taurocholic acid |
H4abtc | 3,3′,5,5′-Azobenzene-tetracarboxylic acid |
H4TPTC | [1,1′:4′,1′′]Terphenyl-3,3′′,5,5′′-tetracarboxylic acid |
Hcptpy | 4′-(4-Carboxyphenyl)-2,2′:6′,2′′-terpyridine |
HER | Hydrogen evolution reaction |
HOMO | Highest occupied molecular orbital |
La | Lanthanum |
Ln | Lanthanide |
Ln-MOFs | Lanthanide-metal–organic frameworks |
LMCT | Ligand-to-metal charge transfer |
LSPR | Localized surface plasmon resonance |
LUMO | Lowest unoccupied molecular orbital |
MIL | Materials of Institute Lavoisier |
MOL | Metal–organic layer |
MoS2 | Molybdenum(IV) sulfide |
Na2S | Sodium sulfide |
Na2SO3 | Sodium sulfite |
NH4F | Ammonium fluoride |
Ni | Nickel |
OER | Oxygen evolution reaction |
PHE | Photocatalytic hydrogen evolution |
Pr | Praseodymium |
PS | Photosensitizer |
Pt | Platinum |
RE-MOFs | Rare earth metal–organic frameworks |
RuOx | Ruthenium oxide |
SBU | Secondary building unit |
SD | Sacrificial donor |
Tb | Terbium |
TBAPy | 4,4′,4′′-(Pyridine-2,4,6-triyl)tribenzaldehyde |
TEOA | Triethanolamine |
TiO2 | Titanium(IV) oxide |
Tm | Thulium |
VB | Valence band |
WO3 | Tungsten(VI) oxide |
Yb | Ytterbium |
ZIS | Zinc indium sulfide |
ZnO | Zinc(II) oxide |
This journal is © The Royal Society of Chemistry 2025 |