Since the 21st century, the world energy crisis has become increasingly urgent mainly due to the growing shortage of non-renewable energy reserves such as coal, oil, and natural gas ^[1][2][3][1,2,3]. It is crucial to find clean renewable energy, especially one that can be produced on an industrial scale ^[4][5][4,5]. Hydrogen (H₂) is regarded as one of the most ideal substitutes for carbon-based energy sources because of its high calorific value and lack of pollution from its combustion products ^[6][7][6,7]. At present, the electrocatalytic decomposition of water is mostly used in industrial hydrogen production, but this undoubtedly brings a large amount of waste of electrical energy ^[8][9][8,9]. As solar energy is inexhaustible, it would be a satisfactory advantage to fully utilize and convert solar energy to catalyze the hydrogen evolution reaction ^[10][11][10,11]. In the early days of this field, some complexes of precious metals were explored as photosensitizers. Although they were highly capable of capturing light, these complexes were too expensive to meet the economic principle ^[12][13][14][12,13,14], which promoted the researchers to develop some low-cost materials.

With the continuous in-depth research of the theory and practice of photoelectric chemistry, a multitude of semiconductor materials with superb light trapping abilities emerged, such as CdS ^{[15][16][17][18][19]}[15,16,17,18,19], TiO₂ ^[20][21][20,21], C₃N₄ ^{[22][23][24][25]}[22,23,24,25], etc. However, these classical semiconductor materials also had some inevitable shortcomings and deficiencies, despite that they were generally low toxicity, cheap, and easy to synthesize. In the case of CdS, serious photocorrosion would occur in the process of photocatalysis, which reduced the available components and made it difficult to maintain a high efficiency catalysis for a long-term reaction ^[26][27][28][26,27,28]. As far as TiO₂ was concerned, it mainly absorbed ultraviolet light and was not sensitive to visible light, which was obviously not the optimal way to utilize solar energy ^[29][30][29,30]. C₃N₄ did not have the above-mentioned disadvantages as a kind of novel organic semiconductor material, but its apparent quantum yields (AQYs) in photocatalysis were not satisfactory, usually no more than 10% ^[31][32][31,32]. Therefore, it is a prerequisite for researchers to make unremitting efforts to extend such semiconductor materials from design to practice.

In the past three decades, metal–organic frameworks (MOFs) have attracted the attention of many chemists due to their controllable pore size, redox ability of central ions, and changeful ligands ^{[33][34][35][36][37][38][39][40][41][42][43][44]}[33,34,35,36,37,38,39,40,41,42,43,44]. Many literature reports in the fields of gas separation and adsorption ^{[45][46][47][48]}[45,46,47,48], fluorescence recognition ^{[49][50][51][52]}[49,50,51,52], electrocatalysis ^[53][54][55][53,54,55], and photocatalysis ^[56][57][58][56,57,58] have confirmed this hotspot. In the aspect of photocatalytic (hydrogen evolution reaction) HER, MOFs are particularly outstanding, and the orientations are mainly focused on the following three aspects. Firstly, the ligands and central ions of MOFs usually play various roles in the photocatalytic process, which makes reasonable design and adjustment of internal structures become one of the most common modification methods for MOFs. Secondly, the integration of MOFs and other types of materials is one of the frequent techniques, which can not only make full use of the stable skeleton of MOFs, but also make the supported substances evenly dispersed to improve the utilization of the cocatalyst. Thirdly, MOFs can be appointed as precursors to synthesize and prepare some substances with special morphology or composition. These corresponding measures give MOFs and their derivatives or composites better development potential and vitality in the field of photocatalytic HER for practical application.

2. Internal Modification and Structure Optimization of MOFs

Reasonable internal modification and structure optimization of MOFs may not only make the photosensitive units orderly, but also shorten the distance between the photosensitive units and the catalytic centers, so as to improve the efficiency of photogenerated electrons separation and accelerate the rate of hydrogen evolution ^[59][60][59,60]. Jiang et al. proposed a facile strategy for the construction of a deployable coordination microenvironment based on MOFs [61]. In this work, the earth-abundant metals (such as Ni²⁺, Co²⁺, and Cu²⁺) were fixed to the Zr₆-oxo cluster in the form of single-atom catalysts (SACs) via a rapid and facile microwave-assisted method. The adjacent -O/OH_x groups in the Zr₆-oxo cluster of MOFs provided the lone pair electrons and the charge balance to anchor the additional single metal atoms. The atomically dispersed metal site was in close proximity to the photosensitive unit (the linker), which greatly accelerated the transfer of photogenerated electrons and thus facilitated the redox reactions. Therefore, the optimized Ni₁-S/MOF had a unique Ni(I) microenvironment and exhibited excellent photocatalytic H₂ evolution performance, which was 270 times higher than that of the pure MOF and far exceeded the other Ni₁-X/MOF counterparts. This work unequivocally demonstrated the great advantage of MOFs in preparing high-content SACs with the proximity of variable microenvironments to photosensitive junctions, thereby promoting electrons transfer and facilitating photocatalysis. In another example of structure optimization of MOFs, Jiang et al. discussed the role of linker engineering in MOFs for dark photocatalysis [62]. The dynamic and thermodynamic investigations manifested that the generation and the lifetime of Ti³⁺ intermediates were the most critical factors affecting their properties, due to the electron-donating/-withdrawing effect of the functional groups. The time-dependent density-functional theory (TD-DFT) calculation indicated that the introduction of an electron-donating group was beneficial, which helped to lengthen the distance between the photogenerated electrons and holes and improved the separation of these pairs. According to the investigation, this was the first study to systematically regulate the dark photocatalytic hydrogen production process of MOFs-based materials. The relevant works gave a new reference to the lifetime adjustment of electron relay for enhancing the dark photocatalysis. The stripping of MOFs single crystals into 2D layered materials could increase the surface area and shorten the transfer path of electrons, which was regarded as an excellent medication method. Zhang and his co-authors [63] reported a water-stable nickel-based MOF single crystal (Ni-TBAPy-SC) and its exfoliated nanobelts (Ni-TBAPy-NB), which could bear a wide range of pH environments. The optimized hydrogen production rate of Ni-TBAPy-NB could reach 98 μmol·h⁻¹ (5 mmol·h⁻¹·g⁻¹) with an apparent quantum efficiency (AQE) of 8.0% at 420 nm, which was 164 times higher than that of Ni-TBAPy-SC. Based on the DFT calculations, the transfer of photogenerated electrons from the H₄TBAPy to the [Ni₃O₁₆] cluster was thermodynamically permissible. In addition, both the reduced [Ni₃O₁₆]⁻ and [Ni₃O₁₆] cluster had excellent properties of water absorption. In another report about 2D MOFs, Duan et al. [64] synthesized 2D indium-based porphyrin MOF cubic nanosheets (2D In-TCCP NS) via a surfactant-assisted method. The 2D In-TCCP NS showed great chemical stability in the range of 2-11 in acidic and basic solutions. In the photocatalytic tests, the 2D In-TCCP NS exhibited a hydrogen evolution rate of 67.97 μmol·g⁻¹·h⁻¹, which was 11.5 times higher than that of 3D In-TCCP bulk (5.87 μmol·g⁻¹·h⁻¹). It was worth noting that there was not an obvious activity decrease after 40 h of photocatalysis, which reflected the practical commercial application.