Photocatalytic CO₂ reduction is a sustainable strategy that converts solar energy into chemical energy, addressing both global warming and the fossil fuel crisis^[1-3]. However, the dissociation energy of the C=O bond in the CO₂ molecule is as high as 805 kJ·mol^-1[4], making CO₂ highly stable and difficult to activate. In recent years, Metal-organic frameworks (MOFs) have attracted significant research interest in photocatalytic CO₂ reduction due to their highly adjustable structures and versatile compositional characteristics. MOFs can not only efficiently capture CO₂^[5] but also optimize their band structures and facilitate the separation of photogenerated charge carriers through strategies such as linker engineering^[6], metal node engineering^[7], and defect engineering^[8]. Compared with traditional inorganic semiconductors, MOFs offer unique advantages in photocatalytic CO₂ reduction^[9,10]. Currently, the main product of photocatalytic CO₂ reduction with MOF systems is mostly CO^[11] or CH₄^[12]. Limited research focuses on the CH₃OH product^[13], and the yield and selectivity still need to be improved. Moreover, the use of sacrificial agents and the lack of rigorous validation raise concerns about the true source of CH₃OH product^[14].

Recently, Dr. Yuanyuan Zhao, Prof. Guangshan Zhu, Prof. Hongzhu Xing, and coworkers reported a site-specific asymmetric coordination engineering strategy to construct a high-performance defective MIL-125-NH₂ (D-MIL) photocatalyst with Cu single-atom (SA) sites for highly selective CO₂ reduction to CH₃OH^[15]. First, the researchers prepared D-MIL with unsaturated Ti-oxo clusters via low-temperature pyrolysis. Then, the flexible N-heteromacrocyclic ligand tetraxetan (DOTA) was anchored at the defective sites to repair the missing-linker defects. Finally, Cu(II) ions were chelated with DOTA to form the target catalyst Cu-DOTA@D-MIL [Figure 1A]. X-ray absorption fine structure (XAFS) analysis and density functional theory (DFT) simulations confirm that Cu SA is coordinated with four N atoms of DOTA to form an asymmetric Cu-N₄ configuration with a dihedral angle of 45° [Figure 1B], which effectively suppresses the aggregation of Cu species and stabilizes the active sites.

Notably, the ultrahigh formation constant of Cu-DOTA (log β_Cu = 22.3) endows the catalyst with excellent structural stability. The D-MIL support not only provides suitable anchoring sites for the DOTA but also maintains the original crystallinity and disk-like morphology. Meanwhile, the incorporation of Cu-DOTA alters the conduction band position of the catalyst, making it closer to the reduction potential of CO₂-to-CH₃OH (-0.38 V vs. NHE) and more favorable for the conversion. In situ electron paramagnetic resonance results under light irradiation, along with cyclic voltammetry measurements, indicate that the catalyst follows a cascade electron transfer pathway. The photogenerated electrons reduce Cu(II) to stable Cu(I) active sites, which further mediate fast electron transfer to activate CO₂ and are themselves oxidized back to Cu(II). The asymmetric Cu-N₄ coordination microenvironment sustains the Cu(I)/Cu(II) redox cycle and avoids over-reduction to Cu(0), providing a steady electron supply for selective CO₂ conversion [Figure 1C].

Under visible-light irradiation without any sacrificial agent, Cu-DOTA@D-MIL exhibits excellent photocatalytic CO₂ reduction performance, achieving a CH₃OH yield of 229.0 μmol·g^-1·h^-1 with a selectivity of up to 95.9% [Figure 2A]. Control experiments confirm the photocatalytic property of the conversion and the indispensability of CO₂. Isotopic labeling experiments with ¹³CO₂ further verify that the produced CH₃OH is exclusively derived from CO₂. The catalyst also shows excellent stability, maintaining about 93.8% of its initial CH₃OH yield after 12 consecutive cyclic runs [Figure 2B]. Powder X-ray diffraction, transmission electron microscopy, X-ray photoelectron spectroscopy, and XAFS characterizations prove the robustness of Cu-DOTA@D-MIL. The Cu-DOTA complex alone shows no catalytic activity, while the combination of these components significantly boosts the CH₃OH production rate, which is 70.5 times higher than that of pristine MIL-125-NH₂ [Figure 2C]^[15].

Mechanistic investigations reveal that the superior catalytic results should be attributed to the asymmetric Cu-N₄ coordination microenvironment, which modulates the electronic structure of Cu SA and facilitates the formation and stabilization of Cu(I) active sites. As a result, the Gibbs free energies to yield key reaction intermediates such as *CO and *CHO are significantly reduced by 0.07 and 0.50 eV, respectively, while the formation of byproducts including HCOOH and CO is suppressed [Figure 3]. In contrast, the Cu(II) site proceeds via a thermodynamically unfavorable pathway, as evidenced by the highly endothermic *OCH₂-to-*CH₂OH rate-determining step (ΔG = +1.53 eV) and the transition from Cu-O to Cu-C adsorption state.

In summary, MOF defect engineering and site-specific macrocyclic ligand functionalization are innovatively integrated to design a novel photocatalyst with an asymmetric coordination environment. The emerging and rare research strategy is deployed in photocatalytic CO₂ to CH₃OH reduction without a sacrificial agent. The high yield and selectivity of CH₃OH highlight the importance of precise integration of each functional module in the catalyst design. This work not only addresses the critical issues of metal aggregation and active site oxidation in SA catalysts but also provides a new approach to regulate the catalytic reaction pathway by engineering the local coordination field of atomic sites. The asymmetric coordination engineering offers a new perspective on the development of high-selectivity, high-stability photocatalysts for CO₂ conversion, and deepens the mechanistic understanding of solar-driven CO₂ reduction catalysis.