Gas transformations within metal–organic cages

Abstract

The efficient transformation of gaseous molecules into value-added products remains a central challenge in sustainable chemistry, limited by the low reactivity of some gases and the complexity of achieving product selectivity. Metal–organic cages (MOCs), with their tunable cavities and dynamic host–guest interactions, have emerged as promising platforms for gas conversion, leveraging confinement effects to enhance reactivity and selectivity. This contribution highlights recent advances in MOC-mediated gas transformations—including photocatalytic O₂ reduction, electrochemical CO₂ conversion to combustible gases, H₂S splitting for H₂ generation, and SO₂ oxidation and mineralization—illustrating how spatial arrangement, co-encapsulation of catalysts and substrates, and cavity design unlock new reaction pathways under mild conditions. Mechanistic insights and structural features outline design principles for next-generation cage-based systems. Ultimately, MOCs offer molecular precision to bridge homogeneous and heterogeneous catalysis, with profound implications for the development of complex gas–liquid–solid phase reactions and transformative technologies aimed at addressing some of the most critical challenges in environmental remediation, energy generation, and circular manufacturing.

---

Federico Tzunux-Tzoc

Federico was born in the small town of Momostenango, Guatemala. He earned his undergraduate degree in chemistry from the University of San Carlos of Guatemala and is currently pursuing a Master's in Chemical Sciences at the National Autonomous University of Mexico (UNAM). His academic journey has centered on supramolecular chemistry, with a particular focus on the design and synthesis of metal–organic cages (MOCs). Under the guidance of Dr G. Percástegui, he has worked on methodologies for synthesizing water-soluble MOCs. Beyond the lab, Federico is deeply interested in theoretical chemistry and the broader physical sciences. In his free time, he enjoys athletics and mountaineering.

Edmundo G. Percástegui

Edmundo was born in Mexico City and earned his Ph.D. in chemistry from the Autonomous University of the State of Hidalgo (UAEH) in Mexico. During his doctoral studies, he spent a year in the group of Darren Johnson at the University of Oregon, investigating self-assembled main-group complexes. He held a DGAPA postdoctoral fellowship in Ivan Castillo's group at the National Autonomous University of Mexico (UNAM), where he worked on metallocalixarenes. From 2016 to 2018, he undertook postdoctoral studies with Jonathan Nitschke at the University of Cambridge, focusing on subcomponent self-assembly and molecular receptors. Later, he was appointed Assistant Professor at the Institute of Chemistry at UNAM. His independent research program, started in 2019, investigates the self-assembly of functional metal–organic containers and supramolecular architectures.

---

1. Introduction

As global efforts intensify to mitigate climate change,¹ reduce pollution, and transition toward circular economies,^2,3 the efficient transformation of gaseous molecules into value-added products has emerged as a critical challenge in modern chemistry. Some gases are abundant, yet often chemically inert,⁴ typically requiring high-energy inputs and specialized catalysts for their activation and conversion. Achieving selectivity for the desired products is also challenging, demanding precise control over reaction intermediates and catalytic pathways, particularly in processes involving complex multi-electron and proton transfer steps, or those that encounter competing side reactions. Furthermore, the low solubility of gases in liquids and their slow diffusion in solids limit interactions with catalytic sites.

In recent years, porous materials such as metal–organic frameworks (MOFs)^5,6 and covalent organic frameworks (COFs)⁷ have emerged as promising platforms for gas capture and conversion. Their high surface areas, tunable pore environments, and modular architectures allow for selective adsorption and catalytic activation of gaseous substrates. Reticular MOFs and COFs have demonstrated success in CO₂ reduction,^8,9 O₂ activation,^10–12 and other gas-conversion reactions.^13–17 However, their extended frameworks may suffer from limited accessibility of reactants to deeper pores and often lack the molecular precision and dynamic host–guest interactions required to modulate reactivity at the nanoscale, especially under mild and scalable conditions. The strategic design of novel materials is essential for overcoming these limitations and advancing catalysis, renewable energy generation, green chemical manufacturing, and environmental remediation strategies.

Metal–Organic Cages (MOCs)¹⁸—also referred to as metal–organic polyhedra or coordination capsules—are a notable class of nanomolecular receptors with unique potential for gas capture and transformation. These three-dimensional architectures are formed through the self-assembly of metal ions and organic ligands^19,20 and delineate well-defined internal cavities capable of accommodating a wide range of guest molecules.²¹ The host–guest chemistry of MOCs has been extensively explored for diverse applications in both solution and solid state, including molecular recognition,^22–27 sensing,^28–30 separation,^31–33 drug delivery,^34–37 or removal of pollutants.^38–40 More recently, significant attention has been directed toward their use in gas capture and separation.^41–47 Unlike extended frameworks, MOCs offer molecular-level control over cavity size, shape, and functionality, enabling tailored host–guest interactions and confinement effects that can dramatically influence chemical reactivity.⁴⁸ The confined space within MOCs isolates guests from bulk environments, increases their local concentrations, and preorganizes them in favorable conformations.^49,50 These confinement effects often lead to significantly enhanced reaction rates and altered product distributions,⁵¹ facilitating chemical transformations that are otherwise inefficient or inaccessible in bulk conditions.⁵² Together, these features have recently been harnessed to promote chemical conversions of gaseous substrates within the nanocage spaces. Notably, MOCs have been shown to mediate the photocatalytic reduction of O₂, the electrochemical transformation of CO₂, the light-driven splitting of H₂S, and the oxidation of SO₂.

This minireview highlights the emerging role of MOCs as platforms for gas transformation reactions, focusing on the available examples where molecular confinement within the cage nanocavity plays a decisive role in enhancing reactivity and product selectivity. The selected cases illustrate how the spatial arrangement of catalytic sites, the stabilization of intermediates, the modulation of electronic environments, and the encapsulation of co-catalyst and substrates may lead to the discovery of new reaction pathways and improve catalytic performance. These findings underscore the importance of supramolecular design in achieving efficient gas conversions under mild conditions. Ultimately, the ability of MOCs to confine, activate, and transform gaseous molecules within their nanocavities opens exciting possibilities for future research in catalysis, materials science, and environmental chemistry. As the field continues to evolve, these molecular cages may inspire new strategies for selective gas transformations, bridging the gap between homogeneous and heterogeneous catalysis and offering biomimetic solutions to some of the most pressing challenges in sustainable chemistry.

2. Cavity-mediated gas conversion

The rational design of MOCs, combined with the judicious selection of metal centers, enables the integration of multiple catalytic functions within a single supramolecular framework. In this context, Liu, Lan, and co-workers reported two cobalt-based MOCs—Co₁₄(L-H)₂₄ and Co₁₄(L-CH₃)₂₄—capable of simultaneously realizing the oxygen reduction reaction (ORR) and the water oxidation reaction (WOR) for the efficient photocatalytic production of hydrogen peroxide.⁵³ These MOCs were synthesized via the self-assembly of Co(NO₃)₂·6H₂O with the ligands HL-H (N-benzyl-1-(1H-imidazol-4-yl)methanimine) and HL-CH₃ (1-(1H-imidazol-4-yl)-N-(4-methylbenzyl)methanimine), respectively. Single-crystal X-ray diffraction (SC-XRD) confirmed the formation of rhombic dodecahedral structures with a [Co^II₁₄(L-R)₂₄]⁴⁺ architecture, comprising 14 Co(II) ions and 24 ligands (Fig. 1a). Two distinct Co(II) coordination environments were identified (inset in Fig. 1a): Co(A), coordinated to four imidazole nitrogen atoms and two water molecules in an octahedral geometry, and Co(B), coordinated octahedrally by six nitrogen atoms from three bidentate ligands. Upon dissociation of the axial water ligands, Co(A) adopts square-planar geometry, exposing an open coordination site that serves as the catalytic center. This structural change also increases the internal cavity diameter from 6.61 Å to 10.93 Å in Co₁₄(L-H)₂₄ and from 6.55 Å to 10.88 Å in Co₁₄(L-CH₃)₂₄. In this scenario, water and gaseous O₂ molecules can easily diffuse into the cavity, facilitating the attack of active sites from both inside and outside the cage and promoting improved catalytic activity.

Fig. 1 (a) Schematic representation of the self-assembly of Co₁₄(L-H)₂₄ and Co₁₄(L-CH₃)₂₄ MOCs (hydrogen atoms and counterions are omitted for clarity). (b) Photocatalytic activity for H₂O₂ production in pure water and under O₂ atmosphere. (c) Amounts of H₂O₂ produced in the presence of different agents. (d) The proposed reaction mechanism and the WOR and ORR pathways for photocatalytic synthesis of H₂O₂. Parts (b) and (c) adapted from ref. 53, https://doi.org/10.1002/anie.202308505. Copyright 2023, Wiley-VCH.

The exchange of ligands at the Co(A) sites is favored, as indicated by the adsorption energies. For water molecules adsorbed on Co(A), the energies are −0.61 eV for Co₁₄(L-H)₂₄ and −0.48 eV for Co₁₄(L-CH₃)₂₄, while the adsorption of oxygen molecules involves energy values of −1.76 eV for Co₁₄(L-H)₂₄ and −1.77 eV for Co₁₄(L-CH₃)₂₄. These values reflect the feasibility of ligand exchange, suggesting that the change in coordination geometry at the Co(A) sites is a dynamic process in which axially bound water ligands are easily replaced by oxygen molecules. These oxygen molecules are reduced, released, and continue the catalytic process. Furthermore, this change in temporary geometry can enhance the utilization of photogenerated electrons at the Co(A) sites. As a result, these sites become more accessible to electrons, which improves the production of ˙O₂⁻, a crucial intermediate for the eventual of H₂O₂. UV/Vis-NIR diffuse reflectance spectroscopy revealed that Co₁₄(L-CH₃)₂₄ exhibits broader and stronger light absorption (λ = 300–1100 nm) than Co₁₄(L-H)₂₄. Kubelka–Munk and Mott–Schottky analyses confirmed that both MOCs function as photocatalysts for H₂O₂ generation via WOR and ORR. Notably, Co₁₄(L-CH₃)₂₄ possesses a narrower band gap and exhibits approximately twice the transient photocurrent response of Co₁₄(L-H)₂₄, indicating superior charge separation and transport efficiency. These features contribute to its enhanced photocatalytic activity. Photocatalytic H₂O₂ production was evaluated in O₂-saturated pure water under UV/Vis-NIR irradiation (λ = 300–1100 nm), without sacrificial agents, photosensitizers, or pH adjustment. Both MOCs showed time-linear H₂O₂ generation, with Co₁₄(L-CH₃)₂₄ achieving a production rate of 146.60 μmol g⁻¹ h⁻¹ in 40 min, significantly higher than the 92.91 μmol g⁻¹ h⁻¹ observed for Co₁₄(L-H)₂₄ (Fig. 1b). Noteworthy, the production of Co₁₄(L-CH₃)₂₄ was approximately 1.6 times higher than that of Co₁₄(L-H)₂₄. Moreover, cycling tests were performed for Co₁₄(L-CH₃)₂₄ and it was found that the H₂O₂ production rate remained virtually unchanged over three consecutive cycles, indicating high stability and durability of the cages.

Mechanistic studies were implemented to elucidate the reaction pathways involved in the production of H₂O₂ (Fig. 1c). For instance, under an Ar-saturated atmosphere (without O₂ gas), only a small amount of H₂O₂ was produced (36.51 μmol g⁻¹ h⁻¹ for Co₁₄(L-H)₂₄ and 50.77 μmol g⁻¹ h⁻¹ for Co₁₄(L-CH₃)₂₄). These and other results corroborated that H₂O and O₂ are essential for H₂O₂ production. In a model experiment, using H₂¹⁸O and ¹⁶O₂, the photocatalytically-synthesized H₂O₂ decomposed by MnO₂ and yielded a product consisting of ¹⁶O₂ and ¹⁸O₂ in a 1 : 1 ratio, indicating that both H₂O and O₂ are the raw materials for the photocatalytic synthesis of H₂O₂. To gain greater insights into the mechanism of WOR and ORR reactions, 1,4-benzoquinone (p-BQ) and isopropanol (IPA) were added to the H₂O₂ synthesis system as reagents to capture ˙O₂⁻ and ˙OH, respectively. In the presence of p-BQ and IPA, the production rate for each of the cages decreased (Fig. 1c), indicating that both ˙O₂⁻ and ˙OH are reactive intermediates, with ˙O₂⁻ playing a fundamental role in the ORR pathway. Density functional theory (DFT) calculations supported that H₂O molecules adsorb onto the imidazole fragments, forming non-conventional hydrogen bonds with the C2–H group (WOR site), while O₂ gas molecules are adsorbed and activated at the Co(A) ions, the ORR catalytically active sites. The mechanism of H₂O₂ synthesis and the WOR and ORR pathways is illustrated in Fig. 1d.

This study demonstrated, for the first time, the potential of MOCs as integrated photocatalysts, leveraging confinement effects, for H₂O₂ synthesis. The spatial arrangement of redox-active sites, for both ORR and WOR processes, within the same cage framework enhances host–guest interactions, facilitates contact between reactants and catalytic sites, and shortens the pathway for photogenerated charges, enabling their rapid transfer. The cage design also gathers six octahedral cobalt(II) Co(A) sites that allow for the simultaneous fixation of multiple gas molecules into the cavity. This arrangement ensures internal oxygen binding, enhances the interaction between the gas and the catalyst, and boosts the production rate of H₂O₂.

The increasing concentration of atmospheric CO₂ is a major contributor to climate change, prompting urgent efforts to develop sustainable strategies for its mitigation. One promising approach is the catalytic conversion of CO₂ into value-added chemicals, which not only reduces greenhouse gas emissions but also contributes to a circular carbon economy and a more sustainable chemical industry. Among emerging approaches, MOCs have demonstrated in recent years their potential as promising platforms for CO₂ transformation due to their tunable structures and ability to promote unique reactivity within their confined molecular environments. Earlier this year, Zhang and coworkers reported that the boron–imidazolate cage BIC-145 facilitates the electrochemical reduction of CO₂ into various products, with ethylene (C₂H₄) being the predominant one.⁵⁴ Ethylene is industrially significant due to its high economic value as a precursor in polymer manufacturing.

Cage BIC-145 was obtained from the solvothermal reaction between the KBH(bim)₃ (bim = benzimidazole) ligand salt (L) and Cu(NO₃)·3H₂O, in the presence of L-tyrosine and a mixed solvent system comprising H₂O, DMSO and t-BuOH, heated at 80 °C. The resulting [Cu^I₆L₄]²⁺ cage was confirmed by SC-XRD to be a regular molecular octahedron defined by the six Cu(I) ions crystallizing in the orthorhombic space group F4̄3c (Fig. 2a). The choice of KBH(bim)₃ was strategic, as it provides a robust Cu–N coordination bond and a strong B–N covalent bond, conferring stability across various solvents (i-PrOH, H₂O, MeOH, DMF, EtOH, t-BuOH) and a broad pH range (1 ≤ pH ≤ 14). BIC-146, an isostructural but less symmetric variant of BIC-145 was obtained by altering the solvent composition to H₂O : DMSO : EtOH during the self-assembly reaction. BIC-146 crystallizes in the P1̄ space group; both MOCs share the [M₆L₄] octahedral array defined by the six Cu atoms. A peculiarity of these two cages is that during formation, the Cu(II) ions reduced to Cu(I) ions as was verified by X-ray photoelectron spectroscopy. Furthermore, BIC-145 was confirmed by mass spectrometry studies in which the signal detected at m/z = 1730.8983 confirmed the {Cu^I₆[BH(bim)₃]₄(H₂O)₃₃-(DMSO)₁₃}²⁺. Electrochemical performance was evaluated in an H-type cell using glassy carbon working electrode and KCl : Cs₂CO₃ (0.1 M/0.05 M) as electrolytes. The main carbon products were CO, CH₄, C₂H₄, and HCOOH, with product distribution strongly dependent on the applied potential (Fig. 2b). At −1.1 V, BIC-145 predominantly produced C₂H₄ and CO, while at −1.2 V, CO formation decreased and C₂H₄ remained the major product (Fig. 2c). BIC-146 exhibited lower selectivity for C₂H₄, with faradaic efficiencies of 24.0% at −1.1 V and 20.6% at −1.2 V, compared to BIC-145. A key difference between the two MOCs lies in their electrochemical stability. At a potential of −1.2 V, BIC-145 displayed exceptional stability, maintaining a stable current density (−3.54 mA cm⁻²) over 5 h of electrolysis, while BIC-146 exhibited declining performance after 2 h under the same conditions. This difference is attributed to the higher symmetry and well-ordered configuration of BIC-145, which enables it to withstand structural deformations caused by the coordination or adsorption of intermediates in its cavity that are produced during CO₂ reduction to C₂H₄. In contrast, the lower structural symmetry of BIC-146 was inferred to result in weaker host–guest associations with the intermediates. Thus, not only the Cu(I) active sites (where the intermediates are coordinated) are key in catalysis, but also the structural configuration of the cage cavity is crucial for catalytic performance, particularly in the adsorption and stabilization of intermediates in the cavity during CO₂ reduction.

Fig. 2 (a) Schematic process of self-assembly of BH(bim)₃⁻ and Cu(II) ions to produce BIC-145. (b) The FE of valuable gas and liquid products of BIC-145 electrocatalyst at different applied potentials. (c) Comparison of valuable carbon products from BIC-145 and BIC-146 at −1.2 and −1.1 V vs. RHE. (d) The possible electrochemical pathways to reduce CO₂ to C₂H₄ showing key intermediates adsorbed on the Cu site inside the BIC-145 cage. Parts (b–d) adapted from ref. 54, https://doi.org/10.1007/s11426-024-2290-5. Copyright 2025, Springer Nature.

Density functional theory (DFT) calculations further elucidated the mechanism. An active Cu(I) site coordinates with two CO molecules to form two activated *CO intermediates, transitioning from a linear Cu(I) to a distorted tetrahedral ion (Fig. 2d); the Cu(I) site exhibits strong affinity for CO. Moreover, DFT suggested that the Cu(I) center, in synergy with the boron-imidazolate ligands, aids the formation of the *COCHO intermediate from *CO and *CHO, a key intermediate influencing the selectivity for C₂H₄. The authors proposed that this intermediate is stabilized by a non-classical hydrogen bond between the C–H bond of the aromatic ring adjacent to the N atoms of the imidazolate ligand and the oxygen of *COCHO, as well as the coordination between the Cu(I) site and a carbon atom of *COCHO within the cage. In this example, the size, polarity, and shape of the cage cavity critically modulate activation barriers for CO₂-to-C₂H₄ conversion by tailoring the microenvironment around the active Cu(I) sites. The large, flexible cavity (∼10 Å diameter) reduces steric hindrance, enabling co-adsorption of *CO and *CHO intermediates and facilitating C–C coupling events, overcoming a relatively low barrier (1.15 eV). The cavity polarity from boron–imidazolate ligands and open Cu coordination sites increases affinity for CO, which in turn, may promote strong σ–π back bonding and non-classical hydrogen bonding, stabilizing key intermediates (e.g., *COCHO) and lowering energy penalties during hydrogenation steps. Thus, the confinement effect operates through (i) spatial restriction, enforcing proximity and orientation of reactants for efficient coupling; (ii) electronic modulation, where ligand fields and Cu–B interactions tune adsorption energies; and (iii) intermediate stabilization, via hydrogen bonding between key and reactive intermediates and cage walls, which minimizes conformational distortion and preserves catalyst integrity. Collectively, these factors reshape the reaction pathway, favoring ethylene formation over competing routes.

The strategic design of MOCs for gas capture and transformation can be significantly enhanced by integrating them into highly porous systems that feature multiple permanent voids. One promising approach involves embedding MOCs within porous liquids (PLs)—a unique class of fluids characterized by intrinsic, permanent cavities. Such a combination offers a compelling solution to the limited gas solubility typically observed in conventional porous solids, particularly in applications involving gas–liquid–solid phase reactions. In a notable example, He et al. reported the synthesis of a porous liquid based on MOCs, capable of storing and activating CO₂ molecules within its permanent cavities, thereby facilitating the synthesis of formamide derivatives.⁵⁵ The porous liquid was obtained by combining the calixarene-like ligand H₄TBSC, Zn(NO₃)₂·6H₂O, and PEG-Im-BDC. Notably, PEG-Im-BDC features long polyethylene glycol (PEG) chains with a molecular weight of ca. 4000 g mol⁻¹ and a length of ≈13 nm, which play a crucial role to preserve the hollow architecture in the liquid phase. The resulting MOC, Im-PL-Cage (Fig. 3a), was obtained as a viscous, paste-like material at room temperature, with a melting point of ≈58 °C, rendering it liquid under mild conditions. FT-IR and UV-vis analyses suggested deprotonated carboxylate groups, demonstrating that the PEG-Im-BDC ligand is fully coordinated to the zinc metal centers. The crystal structure of Im-PL-Cage was not obtained; however, the analogous Zn-Cage structure was elucidated using the ligand (5-Meim-1,3-H₂BDC)⁺(Cl⁻), which lacks the long PEG chains (Fig. 3a). The Zn-Cage was characterized by SC-XRD, PXRD, and mass spectrometry, supporting the proposed model of Im-PL-Cage (Fig. 3a). The barrel-shaped endo cavity of the Zn-Cage features two small portals of 6.8 × 4.5 Ų (including van der Waals radii), which are large enough to permit CO₂ ingress while excluding the bulky PEG chains. This structural array ensures that the cavity of Im-PL-Cage remains permanently unoccupied and that the imidazolium active sites are accessible for CO₂ interaction during catalysis. Thermogravimetric analysis (TGA) further confirmed the thermal stability of Im-PL-Cage up to 200 °C, with no significant weight loss, indicating that no volatile solvent molecules remained within cage cavities. Molecular dynamics simulations corroborated the presence of pores in Im-PL-Cage ranging from 4 to 6 Å, which were readily penetrated by CO₂ molecules (Fig. 3b). The permanent cavity of Im-PL-Cage was reflected in its adsorption CO₂ capacity (1.78 mmol g⁻¹ at 10 bar, Fig. 3c). Importantly, even after the removal of external CO₂ pressure, a substantial fraction of the gas remained sequestered within the porous Im-PL-Cage. In contrast, although Zn-Cage also showed high CO₂ uptake, it released most of the adsorbed CO₂ upon depressurization, rendering it less effective for downstream formylation reactions.

Fig. 3 (a) Schematic process of self-assembly to produce Zn-Cage and porous liquid Im-PL-Cage. (b) Snapshots of simulation boxes for Im-PL-Cage and pore space in Im-PL-Cage. (c) CO₂ adsorption–desorption isotherms of PEG-Im-H₂BDC and Im-PL-Cage at ambient temperature. (d) The formylation reaction of morpholine with CO₂ and phenylsilane. Gray bars represent the transformed amount of CO₂ that occupied the reactor, and red bars represent the transformed amount of CO₂ that was adsorbed on different absorbents. Parts (b–d) adapted from ref. 55, https://doi.org/10.1038/s41467-023-39089-x. Copyright 2023, Springer Nature.

Beyond gas storage, the activation of CO₂ is desirable to form new compounds with added value. For this purpose, the confined CO₂ was used as a gas source for reductive N-formylation reactions. The CO₂ stored in the pores can be easily accessed and activated by the imidazolium sites near the cage walls and the subsequent reaction with amine substrates. To achieve this, first, the CO₂ was loaded into Im-PL-Cage at 20 bar until saturation, followed by equilibration to atmospheric pressure. Subsequently, amine substrates and phenylsilane were introduced, they reacted with the stored CO₂ and then released from the Im-PL-Cage at 60 °C in an air atmosphere. As a proof of concept, morpholine was used as a substrate that reacted with a high concentration of CO₂ (7.74 mmol) in Im-PL-Cage to give N-formylmorpholine in a high yield of 96.7% (Fig. 3d). This high conversion is attributed to the permanent porosity of Im-PL-Cage, which allows CO₂ confinement at levels nearly 15-fold higher than pure PEG-Im-H₂BDC. Conversely, Zn-Cage afforded only 24.5% under identical conditions, likely due to rapid CO₂ desorption during decompression.

The long PEG chains in Im-PL-Cage aid in retaining CO₂ and enhancing its reactivity with morpholine. Additional amine substrates were also tested for formylation, achieving good to excellent conversion rates, indicating that the catalyst has a broad substrate compatibility. Furthermore, Im-PL-Cage displayed good recyclability, maintaining 93% yield of N-formylmorpholine after five consecutive cycles. Hence, the cavity of the MOC governs CO₂ activation by integrating polarity from imidazolium sites with confinement effects. CO₂ is captured near these sites, where spatial restrictions keep the molecule in a reactive state for subsequent conversion and minimize diffusion entropy. Meanwhile, the charged inner walls exert strong electronic modulation, stabilizing CO₂ transition states. Importantly, this confinement reshapes the reaction pathway through synergistic effects that also occur within the pores of the neat liquid, enabling efficient formylation with high yield and material recyclability. With its unique combination of permanent porosity, fluidity, and catalytic efficiency due to molecular confinement, Im-PL-Cage bridges the advantages of heterogeneous and homogeneous systems, offering facile mass transfer, high reactivity, and reusability, thereby expanding the horizon for catalytic applications.

MOCs are emerging and versatile platforms for gas transformation reactions, not only due to their ability to confine and concentrate gaseous molecules within well-defined cavities, but also because they can co-encapsulate other reactive species such as substrates and catalysts. By confining gases, substrates, and catalytic centers into close proximity, these MOCs generate unique nanoenvironments that enhance reaction rates, improve selectivity, and stabilize reactive intermediates, unlocking synergistic effects that extend far beyond simple gas capture. In 2025, Sun, Wang et al. reported an elegant example in which two MOCs, Cu₃VMOP (empty) and its host–guest derivative, the host–guest Cu₄I₄@Cu₃VMOP containing a Cu₄I₄ cluster in its cavity, catalyze the photoreduction of CO₂ with different product selectivity. While empty Cu₃VMOP catalyzes the reduction of CO₂ to HCOOH and CH₃COOH as main products, the host–guest Cu₄I₄@Cu₃VMOP photocatalyst affords exclusively HCOOH in much higher yield than the free-cluster cage.⁵⁶ Typically, the photochemical reduction of CO₂ is fundamentally characterized by a multielectron and proton transfer process, and selectivity remains a crucial challenge.⁵⁷ In this regard, it is essential to obtain highly selective catalysts in the reduction process. Crystals of tetrahedron Cu₃VMOP were obtained through one-pot reaction of VOSO₄·xH₂O, the ligand 1H-pyrazole-4-carboxylic acid (H₂PA), and Cu(NO₃)₂ in DMA : MeOH at 130 °C. The H₂PA ligand was intentionally chosen to enable selective coordination of the carboxylate group with the strongly acidic polyoxovanadates, whereas the pyrazolate moiety coordinates with the less acidic copper ions, enabling the assembly of a heterometallic MOC. To construct the endo-functionalized Cu₄I₄@Cu₃VMOP cage, the Cu₄I₄ cluster was formed in situ within Cu₃VMOP by introducing pyridyl-4-phosphonic acid and an additional copper source, CuI. The molecular framework of Cu₃VMOP contains two types of secondary building units (SBUs): a polyoxovanadate cluster {V₆S} and a trinuclear copper cluster {Cu₃Pz₃} (Fig. 4a). The {Cu₃Pz₃} unit is structurally analogous to 1,3,5-tris(4-carboxyphenyl)benzene (H₃BTB). Overall, Cu₃VMOP features four 3-connected vertices {V₆S} and four 3-connected {Cu₃Pz₃} faces. On the other hand, Cu₄I₄@Cu₃VMOP possesses four 3-connected vertices {V₆P} and four 3-connected {Cu₃Pz₃} faces. The {V₆P} corner is analogous to {V₆S} where the phosphonate groups of pyridine phosphonate ligands replace the sulfate at the center of the {V₆S} cluster. The crystal structures of Cu₃VMOP and Cu₄I₄@Cu₃VMOP were confirmed by SC-XRD (Fig. 4a) and PXRD, with experimental patterns matching simulated ones. X-ray photoelectron spectroscopy (XPS) revealed that the Cu ions in the Cu₄I₄ cluster maintain a +1-oxidation state, whereas those in {Cu₃Pz₃} faces remain at oxidation states of +2. Photocatalytic CO₂ reduction experiments were carried out using Ru(bpy)₃Cl₂ as the photosensitizer and triethanolamine (TEOA) as the sacrificial agent in CH₃CN solution. Noteworthy, no reduction products were detected during control experiments using VMOP-15, an isostructural MOC composed of BTB and copper-free {V₆S}. Upon replacing VMOP-15 with Cu₃VMOP and irradiating for 3 h, formic acid (HCOOH) and acetic acid (CH₃COOH) were detected as the major products at yields of 2.5 and 0.9 mmol g⁻¹, respectively (Fig. 4b). Remarkably, when Cu₄I₄@Cu₃VMOP was used as a catalyst under identical conditions, a selective conversion of CO₂ to HCOOH was noted with 100% selectivity and an enhanced yield of 10.5 mmol g⁻¹, exceeding that of Cu₃VMOP by a factor of 4.2 times. The host–guest catalysts also showed outstanding stability and efficiency over multiple and consecutive catalytic cycles. Undoubtedly, the confinement of the Cu₄I₄ cluster within Cu₃VMOP significantly alters the electronic properties of the catalyst, playing a key role in directed electron/substrate migration and intermediate anchor and stabilization. Mechanistic investigations combining femtosecond transient absorption (fs-TA) and DFT calculations revealed that the Cu₄I₄ cluster enhances localized electronic states and modulates orbital configurations, thereby alleviating charge recombination and extending a long-lived charge-separated state. The fs-TA results reveal a significantly longer absorption decay half-life of the excited state of Cu₄I₄@Cu₃VMOP (τ₁ = 9.38 ps, τ₂ = 104.90 ps) compared to Cu₃VMOP (τ₁ = 0.92 ps, τ₂ = 37.53 ps). This extended lifetime improves photogenerated charge separation efficiency and extends the excited state, ultimately favoring the photocatalytic process. The in situ DRIFTS spectra of Cu₃VMOP showed peaks at 1507 cm⁻¹ for *OCHO, a crucial intermediate for HCOOH, as well as bands at 1558 cm⁻¹ and 1540 cm⁻¹ for *COCO and for *COOH, key intermediates associated with the production of CH₃COOH. In contrast, for Cu₄I₄@Cu₃VMOP, the intensity of the *OCHO peak (1506 cm⁻¹) gradually increased with time, while the signals related to C–C coupling were not observed (Fig. 4c). Electrostatic potential analysis from theoretical simulations indicated that the negative charges were mainly concentrated on {V₆S} clusters in Cu₃VMOP, whereas in Cu₄I₄@Cu₃VMOP, they are distributed across both {V₆P} and Cu₄I₄ clusters, suggesting that Cu₄I₄ may act as an electron trap to generate the localized electronic excited states. HOMO and LUMO calculations identified the {Cu₃Pz₃} fragments as the catalytic site for CO₂ reduction in both cage systems, with the charge recombination being efficiently suppressed by Cu₄I₄. An essential requirement for efficient substrate catalysis is the stereospecific orientation and complementarity between substrates (and/or intermediates) and the catalytic site, which ensures favorable energy barriers. Analogous to enzymatic systems, it was observed that the Cu₄I₄@Cu₃VMOP cluster slightly decreases the activation energy of the rate-determining step for obtaining formic acid (*OCHO → HCOOH), from −4.82 eV in Cu₃VMOP to −5.05 eV. This reduction in the activation energy barrier suggests that Cu₄I₄@Cu₃VMOP, similar to enzymes and with the confined Cu₄I₄ complex makes HCOOH generation more favorable (Fig. 4d). Furthermore, Cu₄I₄@Cu₃VMOP exhibited a significantly higher energy barrier for *CO coupling (+8.03 eV, endothermic), which inhibits acetic acid formation and enhances formic acid selectivity. On the other hand, Cu₃VMOP displayed a lower energy barrier (+4.64 eV), favoring the C–C bond formation (Fig. 4d). The tetrahedral cavity of Cu₃VMOP and its endo-functionalized variant Cu₄I₄@Cu₃VMOP mimic enzymatic active sites, where cavity size and geometry enforce substrate preorganization, while polarity from embedded clusters modulates electron density. This dual control lowers activation barriers for single-carbon products and selectively raises those for C–C coupling, steering the reaction toward formic acid. The endogenous Cu₄I₄ clusters act as electron traps, suppress charge recombination, prolonging charge-separated states, and stabilizing *OCHO intermediates, akin to cofactor-assisted catalysis in metalloenzymes. Such biomimetic confinement parallels molecular sieving phenomena, demonstrating how cavity architecture–through shape, polarity, and electronic tuning–may dictate catalytic selectivity and efficiency in gas conversion reactions.

Fig. 4 (a) Sequential self-assembly of Cu₃VMOP and Cu₄I₄@Cu₃VMOP. (b) Comparison of yields and selectivity observed for the relevant catalysts during the CO₂ reduction reactions. (c) In situ DRIFTS spectra for detecting the reaction intermediates on Cu₄I₄@Cu₃VMOP and Cu₃VMOP. (d) Free-energy diagrams for HCOOH and CH₃COOH for Cu₃VMOP and Cu₄I₄@Cu₃VMOP. Parts (b–d) adapted from ref. 56, https://doi.org/10.1002/anie.202423018. Copyright 2025, Wiley-VCH.

As demonstrated in previous examples, molecular confinement within well-defined cavities—reminiscent of enzymatic active sites—can significantly enhance substrate activation and electron transfer through synergistic co-encapsulation of substrates and host–guest interactions.^58,59 Another compelling demonstration of this concept is MOC H1, a homogeneous electrocatalyst whose inner nanoenvironment encapsulates and activates the in situ formed ethanolamine-CO₂ adduct. This confinement-driven activation enables CO₂ reduction to methane at a record-low overpotential.⁶⁰ MOC H1 is obtained from the equimolar reaction of Fe(OTf)₂ with a tridentate–bidentate ligand containing guanidine groups under inert conditions (Fig. 5a). SC-XRD analysis evidenced a tetrahedral [Fe^II₄L₄]⁸⁺ molecular array, in which each Fe(II) center is coordinated by three bidentate N,N′-ligand arms, resulting in a symmetric and robust architecture. The molecular array persists in solution as confirmed by NMR and mass spectrometry studies. The guanidine groups in H1 activate hydrogen bonding with guest molecules. In particular, the adduct formed in situ by ethanolamine and CO₂ (MEACO₂) could be stabilized by hydrogen bonds within H1 as confirmed by ESI-MS, which showed a peak at m/z = 784.8828, assigned to [Fe₄L₄(OTf⁻)₂·(MEACO₂⁻)₂]⁴⁺. Likewise, Isothermal Titration Calorimetry (ITC) further validated the encapsulation of two MEACO₂ units, with a favorable Gibbs free energy change (ΔG = −31.53 kJ mol⁻¹). Electrochemical studies using square wave voltammetry (SWV) revealed a dramatic positive shift in the reduction potential of MEACO₂ from −1.10 V to −0.21 V vs. the normal hydrogen electrode (NHE) upon encapsulation by H1 (Fig. 5b), indicating strong host–guest interactions that facilitate electron transfer and activation. Electrocatalytic reduction experiments were conducted under CO₂-saturated and standard conditions (gas bubbled until saturation, DMSO/H₂O = 2 : 1, 0.1 M LiCl, 2.0 M ethanolamine, 0.01 mM Cage H1). CH₄ production was observed across a wide potential range, with significant generation at 0.16 V vs. NHE registered at the cathode and oxygen-evolution taking place at the anode. As the cathodic potential decreased, both faradaic efficiency and turnover frequency (TOF) significantly increased, while CH₄ selectivity remained near 100% (Fig. 5c). For further experiments at −0.04 V vs. RHE, continuous electrocatalysis over 18 hours maintained 100% selectivity and yielded a faradaic efficiency of 69.8% and a turnover number (TON) exceeding 220. To demonstrate the role of confinement and guanidine functionality, a molecular complex M1 was synthesized as a control catalyst, consisting of Fe(II) coordinated by three 2,2′-bipyridine ligands. SWV studies indicate that M1 shows minimal MEACO₂ activation, with a negligible shift in reduction potential (from −1.08 V to −1.02 V vs. NHE), underscoring the essential importance of the H1 nanoenvironment and guanidine groups in catalytic performance. In situ Raman spectroscopy provided mechanistic insights, identifying a signal at 675 cm⁻¹ corresponding to MEA-COO⁻*, which is a key intermediate in the reduction of CO₂ to CH₄. Likewise, absorption peaks were detected at 1047 cm⁻¹ (MEA-COOH*) and at 2921 cm⁻¹ (MEA-CH₃) that increased over time. The proposed mechanism involves electron transfer to MEACO₂ within the H1 cavity, facilitating the gradual hydrogenation to MEACH₃*, a process followed by the cleavage of the C–N to release CH₄, and regeneration of the ethanolamine–CO₂ adduct, completing the catalytic cycle (Fig. 5d). This study highlights how molecular confinement within MOCs, in this case, a guanidine-based tetrahedron, is central to steering the CO₂-to-CH₄ conversion by creating a microenvironment that mimics enzymatic active sites. Spatial restriction inside the cage preorganizes the ethanolamine–CO₂ adduct, ensuring proximity between reactive centers and facilitating stepwise hydrogenation while suppressing competing C–C coupling reactions. Simultaneously, the polar guanidine groups and hydrogen-bonding interactions electronically modulate the substrate, lowering the reduction potential by nearly 0.9 V compared to the free adduct and stabilizing key anionic intermediates, thus offering a biomimetic strategy for selective and efficient carbon conversion.

Fig. 5 (a) Schematic process of self-assembly to produce H1 and the formation of the host–guest 2MEACO₂@H1 adduct. (b) Square wave voltammetry (SWV) of the systems containing MEACO₂ (0.1 mM) at different concentrations of H1. (c) Relationship between TON and selectivity for the electrocatalytic reduction of CO₂ to CH₄ by H1 at different controlled potentials. (d) Schematic diagram of the cavity-mediated mechanism for the conversion of MEACO₂ into CH₄ within tetrahedron H1. Parts (b–d) adapted from ref. 60, https://doi.org/10.1039/d5qi00756a. Copyright 2025, The Royal Society of Chemistry.

Hydrogen sulfide (H₂S) is a toxic and corrosive gas commonly found in industrial waste streams and natural sources such as geothermal vents and crude oil. Its efficient transformation into benign or valuable products is a pressing challenge in environmental chemistry and sustainable catalysis. One promising strategy could involve leveraging confinement effects within supramolecular architectures to enhance reactivity and selectivity. In 2017, Duan, He, Reek et al. informed that the redox-active Ni-TFT cage encapsulates the organic dye fluorescein (FL), such a host–guest complex (FL@Ni-TFT) was applied for light-driven splitting of H₂S, generating molecular hydrogen and elemental sulfur.⁶¹ The combination of 6 equiv. of Ni(CH₃COO)₂·4H₂O with 4 equiv. of the 2,2,2-((nitrilotris(benzene-4,1-diyl))tris(methanelylidene))tris(hydrazinecarbothioamide) ligand (H₃TFT) in DMF afforded Ni-TFT. The resulting molecular octahedron features Ni(II) ions at the vertices of the array and H₃TFT ligands spanning the faces, with two ligands positioned on opposite faces (Fig. 6a). The high coordination capacity of the sulfur-containing H₃TFT ligands confers exceptional stability to MOC Ni-TFT, even in the presence of sulfide ions, which are known to strongly coordinate to metals, leading to poisoning or decomposition of the metal catalyst. Furthermore, the amino groups of the building blocks can act as hydrogen bond acceptors and donors, while phenyl rings enable π–π stacking associations, collectively acting in favor of effective fluorescein binding (Fig. 6a). ESI-MS and NMR analyses supported guest encapsulation, whereas fluorescence indicated the formation of a 1 : 1 host–guest complex with an association constant of ca. 5.4 × 10⁴ M⁻¹. Initial experiments to establish efficient hydrogen production indicated that the proton reduction is more efficient at pH of ≈12.6. Then, in a model reaction, H₂S gas was bubbled into a H₂O/EtOH (1 : 5 v/v) solution containing FL (2.0 mM) and NaOH (0.1 mM, to adjust pH to a value of 12.6); this led to a sulfide ion (S²⁻) concentration of 0.05 M. The mixture was used for hydrogen production under irradiation after addition of the Ni-TFT octahedron. The bright yellow-green solution turned turbid with yellow powder and hydrogen gas formed after 24 h of irradiation (Fig. 6b). The yellow powder was filtered off (ca. 1 mg) and confirmed to be elemental sulfur by PXRD.

Fig. 6 (a) Self-assembly of Ni-TFT and formation of its host–guest complex with Fl. (b) Photographs showing the reaction mixture containing yellowish powder following irradiation. (c) Light-driven hydrogen evolution activity of systems containing FL (2.0 mM), Na₂S (0.05 M) and Ni-TFT 2.0 mM (black line), 4.0 mM (red line) and 6.0 mM (blue line). (d) Turnover number of hydrogen (per Ni atoms, green bars) and sulfur (per Ni atoms, purple bar) produced of systems containing FL (2.0 mM)/Na₂S (0.05 M) with Ni-TFT (4.0 μM) and Ni-DMT (24 μM). (e) Schematic representation for the encapsulation of FL within Ni-TFT to produce H₂ and sulfide oxidation from H₂S splitting. Parts (b–e) adapted from ref. 61, https://doi.org/10.1002/anie.201704327. Copyright 2017, Wiley-VCH.

When the fluorescein concentration was held constant (2.0 mM), hydrogen generation increased linearly (from 2.0 μM to 6.0 μM) as a function of Ni-TFT concentration (Fig. 6c), indicating that the internal cavity of Ni-TFT is required for the decomposition of H₂S at a turnover number of 2600 per mole of MOC catalyst. The preorganization of the dye within Ni-TFT results in proximity, allowing for a direct photoinduced electron transfer from the excited state of the FL* dye to the redox catalyst, establishing a Ni(II)/Ni(I) redox center whose reduction potential is well-suited for proton reduction in aqueous media; the latter ultimately reduces the nickel complexes of the cage ([Ni-TFT]⁻), and expels the oxidized FL⁺ dye from the molecular cavity. Then, the reduced cage reduces protons to produce molecular hydrogen. With the oxidized dye outside the cage, FL⁺ oxidizes sulfide ions in solution to generate the observed yellowish solid sulfur (Fig. 6e). Importantly, the pre-organization of fluorescein within the cavity facilitates a direct photoinduced electron transfer to the redox center, bypassing diffusion limitations and enabling efficient proton reduction. The achieved TON by the design (2600 per mole of catalyst) rivals the reported values for semiconductor photocatalysts under similar conditions. While semiconductor systems such as TiO₂, ZnO, CdS, and ZnS often require noble metal co-catalysts (Pt, Pd, Ir, Ru, Rh, or Au) to trap photogenerated electrons, prolong electron lifetimes, and enhance hydrogen evolution,⁶² Ni-TFT accomplishes this using an inexpensive, reusable organic dye. This eliminates dependence on costly noble metals, offering a more sustainable approach. Additionally, introducing sulfur atoms into the ligand makes Ni-TFT more stable against sulfide passivation, a common problem in inorganic catalysts. For context, the industrial Claus process for H₂S decomposition requires temperatures above 1000 °C and still achieves low conversion rates, with >50% conversion only at ∼1375 °C.⁶³ In contrast, Ni-TFT enables hydrogen evolution from H₂S at room temperature under visible light, highlighting a significant advantage in energy efficiency and operational simplicity. To validate the role of molecular confinement for fluorescein activation and H₂S splitting, a mono-nuclear Ni-DMT complex was synthesized by reacting Ni(CH₃COO)₂·4H₂O and HDMT. The crystal structure of the complex confirmed that the ligand environment around the nickel(II) atom is similar to that of the metal ions in the Ni-TFT cage. Under the same pH, fluorescein and Na₂S concentrations, and using enough molecular Ni-DMT to ensure the same nickel atom concentration (24 μM), only 0.32 mL of H₂ gas was generated, and no yellow precipitate was observed. In connection to this, a competitive inhibition experiment was carried out by adding glucosamine (GLA), which competes with FL for the cage interior. The addition of GLA to the original FL (2.0 mM)/Ni-TFT (4.0 μM)/Na₂S (0.05 M) reaction system also resulted in only 0.32 mL of H₂ gas and no sulfur formation (Fig. 6d). The confinement effect within the Ni-TFT octahedral cage significantly influences the reaction pathway by organizing the dye and catalytic sites within a spatially restricted, electronically modulated environment. The cavity promotes close proximity between the photosensitizer and nickel centers, enabling a direct photoinduced electron transfer through oxidative quenching, which is more efficient than reductive quenching that occurs in bulk solutions. Additionally, the hydrophobic cavity mimics the active sites found in enzymes, selectively favoring proton reduction within the pocket and expelling the oxidized, more polar, and water-soluble dye for sulfide oxidation outside. Together, these results underscore the power of supramolecular confinement in achieving host–guest-mediated photocatalytic redox reactions of gases.

Duan, Jing, and co-workers demonstrated the utility of the host–guest catalyst approach for gas conversion in synthetic chemistry. In this study, the light-driven splitting of H₂S, yielding hydrogen gas, was combined with the subsequent reduction of nitroaromatics in a one-pot process (Fig. 7).⁶⁴ Two helical [Fe^II₂L₃]⁶⁺ metal–organic capsules, C1 and C2, were synthesized using Fe(II) ions and ligands with or without hydrogen-bonding capabilities. Capsule C1, featuring amide groups, was able to encapsulate fluorescein (FL) and form the stable host–guest C1F complex through hydrogen bonding (Fig. 7a). Although capsule C2 lacked such interactions, it also formed the corresponding host–guest C2F complex upon encapsulation of fluorescein. Upon photoirradiation, both capsules catalyzed the transformation of H₂S into hydrogen gas and elemental sulfur. The authors hypothesized that incorporating unsaturated substrates within the capsule cavities could enable direct hydrogenation via proximity to the generated active hydrogen species. Hence, the produced active hydrogen was utilized for hydrogenation of nitroaromatics, to aniline derivatives (Fig. 7b). Under optimized conditions for hydrogen generation, the reduction of nitrobenzene (0.50 mmol) to aniline was first studied, which directly resulted in the formation of aniline (up to 99% yield) within 60 min. When the concentrations of Na₂S and nitrobenzene were fixed and the C1F loading was varied from 0.20 mM to 1.0 mM, the initial reduction rates of nitrobenzene exhibited first-order dependence on the C1F concentration. With varying the size of the nitroaromatic from nitrobenzene to 1-nitropyrene, the initial rate decreased from 1.87 mM min⁻¹ to 0.28 mM min⁻¹, showing the size-dependent kinetics of the catalytic process, suggesting that the reduction process occurs within the helicate cavity. As evidence of this, when 2,4,6-triphenylnitrobenzene, which has a molecular size larger than the cavity of C1F, was used as the substrate, no amine product was observed under the optimized conditions, and only a small amount of hydrogen gas was generated.

Fig. 7 (a) Schematic representation of self-assembly of C1 and the photocatalytic system of C1F for the combination of H₂S splitting and hydrogenation of nitroaromatics. (b) Photocatalytic reduction of different nitroaromatics into aminoarenes under standard conditions: substrate (0.50 mmol).

Although C2 showed higher overall photocatalytic efficiency, C1 exhibited superior performance in the reduction of nitroaromatics, a reaction highly sensitive to substrate size and cavity compatibility. The observed size-dependent kinetics and Michaelis–Menten behavior confirmed that the reactions occurred within the confined space of the capsules, mimicking enzyme-like selectivity. These findings further attest to the potential of molecular confinement in enhancing reactivity and selectivity. The encapsulation of substrates within tailored cavities facilitates pseudo-intramolecular electron transfer, stabilizes reactive intermediates, and enables controlled redox transformations. In particular, the ability of C1 to selectively bind and transform nitroaromatics highlights how hydrogen bonding and cavity architecture can be leveraged to tune catalytic outcomes. This work exemplifies how supramolecular confinement within MOCs can be harnessed for efficient gas transformation and substrate-specific catalysis, paving the way for coupling cage-based gas conversions to sustainable production of complex and fine chemicals.

Sulfur dioxide (SO₂) is an irritating and highly toxic gas that poses severe health risks to humans, with exposure levels exceeding 100 ppm potentially being fatal. SO₂ is categorized as one of the most hazardous air pollutants,⁶⁵ influencing climate change and playing a major role in atmospheric chemistry. Its emissions contribute to the presence of particulate matter (PM_2.5) and the formation of acid rain, causing damage to crops and forests. Our group has recently noted the essential role of the molecular cavity within MOC-Pd-NO₃ in enhancing the efficiency of the transformation of dry or humid SO₂ gas into sulfate, further emphasizing the importance of cage cavities in optimizing chemical processes.⁶⁶ The nanometric MOC-Pd-NO₃ cage results from the self-assembly of the threefold-symmetric ligand L(NO₃)₃ with palladium(II) nitrate in water.⁶⁷ SC-XRD analysis revealed a pseudo-cubic [Pd^II₆L₈]³⁶⁺ molecular framework, which consists of eight corners defined by an aryl ligand moiety and six faces covered by a Pd^II ion coordinated by four pyridyl rings; in this array, the Pd^II⋯Pd^II separations range from 1.80 to 2.14 nm. The positive cage charge is balanced by 36 nitrate counterions that reside within the cage interior (Fig. 8a). Activated MOC-Pd-NO₃ can capture up to 6.00 mmol g⁻¹ of SO₂ at 298 K and 1.0 bar. Notably, 94% of the captured SO₂ is irreversibly retained (ca. 5.70 mmol g⁻¹), as evidenced by the pronounced and broad hysteresis loop observed in the SO₂ isotherms (Fig. 8b). FT-IR, XPS, and NMR analyses confirmed that the retained gas oxidizes to sulfate due to its reaction with cage nitrate ions; all nitrate is consumed. During this process, one cage unit lowers the harmful effects of ca. 42 moles of SO₂, in other words, 1.0 gram of cage material can inactivate ca. 360 mg of hazardous SO₂ at room temperature. Isostructural MOC-M-NO₃ cages assembled through Co(II), Cu(II), or Zn(II) exhibited SO₂-chemisorption; however, only Pd(II) ions induce quantitative SO₂ conversion. While MOC-Pd-NO₃ oxidizes 94% (ca. 5.7 mmol g⁻¹) of its uptake, MOC-Zn-NO₃ decomposes upon exposure to SO₂, and the MOC-Co-NO₃ and MOC-Cu-NO₃ cages achieve partial irreversible adsorption; from what they capture, solely 57% (ca. 1.8 mmol g⁻¹) and 46% (ca. 3.0 mmol g⁻¹) are chemisorbed, respectively.

Fig. 8 (a) Aqueous self-assembly of MOC-Pd-NO₃. The cationic part of the crystal structure of cage is depicted displaying the encapsulated nitrate counterions in space-filling mode. (b) Comparison of the SO₂ uptake of Face-Pd-NO₃ molecular complex against its MOC-Pd-NO₃ cage counterpart; adsorption isotherms at 298 K and up to 1 bar (left). Comparison of the SO₂-oxidation ability of the six-faced MOC-Pd-NO₃ against six disassembled Face-Pd-NO₃ edges (right). (c) Schematic representation of the cavity-mediated mechanism of SO₂ to SO₄²⁻ transformation leading to the formation of MOC-Pd-SO₄ containing trapped sulfate. Part (b) adapted from ref. 66, https://doi.org/10.1002/anie.202421169. Copyright 2025, Wiley-VCH.

The cage cavity was identified as the most significant factor in efficient sulfate formation. A molecular Face-Pd-NO₃ complex, representing one edge of the cage, was prepared and exposed to SO₂, resulting in an adsorption of 6.91 mol of SO₂ per mol of complex. As expected, this amount is one-sixth of the uptake observed with MOC-Pd-NO₃ (45.0 mol mol⁻¹). In this context, six moles of Face-Pd-NO₃ complex, which are equivalent to one mol of the MOC-Pd-NO₃ cage and provide a total of 36 nitrates, collectively achieve an uptake of 41.4 moles of SO₂. However, only about 39% of the SO₂ adsorbed by Face-Pd-NO₃ was chemisorbed, and a significant amount of unreacted NO₃⁻ was detected. Although six molecular edges adsorb virtually the same amount of gas as the six faces forming the cage (41.4 compared to 45 moles), only 16.1 moles of SO₂ would be oxidized by six separated edges. This yield is quite poor when compared to the 42.3 moles of gas converted by the pseudo-cube (94% yield of S(IV) to S(VI) oxidation, Fig. 8b).

Based on the results and the established chemistry surrounding the reaction of SO₂ by nitrate,⁶⁸ a cavity-mediated mechanism for in-cage sulfate production was proposed (Fig. 8c). Initially, SO₂ diffuses into the cage, where it may react with the confined nitrates according to the standard 1 SO₂ : 2 NO₃⁻ stoichiometry, yielding one equivalent of sulfate and two of NO₂. The NO₂ gaseous product is a strong SO₂ oxidant and can react with additional gas, producing more sulfate and N^III-reduced species, which may continue to oxidize SO₂ until cage saturation. The nanoconfined environment of the cage enforces highly effective concentrations of reactants. It is estimated that the effective concentrations of nitrate and SO₂ could reach up to 11.4 and 13.3 M per cage, respectively, based on a molecular volume of 5264 ų. Likewise, the inner cage space may act as a reservoir for the gaseous byproduct NO₂ and stabilize reactive intermediates, many of which are likely to be radicals. The synergistic operation of these effects resulted in the observation of an unprecedented ca. 2.3 SO₂ : 2 NO₃⁻ stoichiometry (42 SO₂ : 36 NO₃⁻), significantly exceeding the established 1 SO₂ : 2 NO₃⁻ standard ratio. Finally, immersing the solid product MOC-Pd-SO₄ in water, followed by the addition of Ba(NO₃)₂, brought about the regeneration and release of caged sulfate into solution, its precipitation as BaSO₄, and the recycling of former nitrate cage, MOC-Pd-NO₃.

From the discussed examples, it is evident that metal–organic cages can be regarded as emerging catalytic platforms for gas conversion reactions, leveraging their confined cavities to influence reactivity and selectivity. The spatial confinement within these discrete constructs enables unique mechanistic pathways, stabilizes reactive intermediates, and enhances local concentrations of substrates. Table 1 summarizes the features of the examples discussed, highlighting their catalytic roles, operating conditions, and key mechanistic insights that underpin their performance in gas transformations.

Table 1 Key aspects and mechanistic insights of gas conversion reactions within MOCs

MOC catalyst	Gas transformation	General conditions	Yields/selectivity	Key mechanistic insights	Ref.
FE: faradaic efficiency
Co14(L-CH3)24 (a [Co^II14L24]^4+ MOC)	Photocatalytic	UV/Vis–NIR (λ = 300–1100 nm), O2-saturated water, no sacrificial agent	H2O2 (production rate = 146.6 mol g^−1 h^−1)	Photogenerated e^− reduce O2 (ORR); h^+ oxidize H2O (WOR); both via 2e^− processes	53
	O2 + H2O → H2O2
BIC-145 (a [Cu^I6L4]^2+ MOC)	Electrochemical reduction	Electrocatalysis in KCl : CS2CO3, CO2 flow, −1.2 V	FE: 28% for C2H4 and current density of −3.54 mA cm^−2	Cu(I) sites enable *CO/*CHO coupling; cage stabilizes intermediates	54
	CO2 → C2H4 (main)		Minor products: CO, CH4, HCOOH
Im-PL-Cage (Zn-calixarene/imidazolium MOC as a porous liquid)	N-formylation	1.0 bar CO2, phenylsilane, DMF, 60 °C, 10 h	7.74 mmol of CO2 to formylation product (96.7% yield)	PEG chains enhance CO2 transport and storage; imidazole sites activate CO2	55
	CO2 + morpholine → N-formylmorpholine
Cu4I4@Cu3VMOP (host–guest catalyst [Cu3VMOP(Cu4I4)]^4−)	Photoreduction	CH3CN, Ru(bpy)3Cl2, triethanolamine, light	10.5 mmol g^−1 of HCOOH product; 100% selectivity	Cu4I4 cluster improves charge separation; lowers energy barrier for HCOOH formation over CH3COOH	56
	CO2 → HCOOH
H1 (host–guest catalyst, [Fe^II4L4(MEACO)2]^8+)	Electrochemical reduction	H-type cell, RT, ethanolamine, LiCl, DMSO/H2O (2 : 1), CO2 (saturation)	CH4 selectivity ≈100%; FE = 69.8%; TON > 220; TOF ≈ 35 h^−1	H1 encapsulates ethanolamine, yielding H1-MEACO2 guest adduct and lowering the reduction potential of CO2. Guest hydrogenation forms MEACH3*. The C–N bond is broken releasing CH4.	60
	CO2 → CH4
Ni-TFT ([Ni^II6L4(fluorescein)] host–guest catalyst)	Photocatalytic splitting and H2 generation	pH ≈ 12.6, Ni-TFT (4 μM), fluorescein, Et3N, H2O/EtOH, Na2S (0.05 M)	H2 and elemental sulfur obtained (1 mg). TON: 1600 for H2	Light-activated fluorescein is expelled from the cavity and oxidizes S^2−; Ni center reduces protons to H2	61
	H2S → H2 + S
C1F (a triple helical [Fe^II2L2(fluorescein)]^4+ host–guest catalyst)	Photocatalytic H2 generation	CH3CN/H2O, Na2S, Xe lamp	H2 (TON: 276.7; TOF: 19.7 h^−); 0.05 mmol aniline products	Fluorescein activation; H2 reduces nitroaromatics; ouside of the cavity, the oxidized fluorescein oxidizes S^2−	64
	H2S → H2 + S; then, nitroaromatic reduction
MOC-Pd-NO3 (a pseudocubic [Pd^II6L8]^36+ catalyst)	Redox gas reaction	Solid-state, 1.0 bar SO2, 298 K	5.7 mmol of SO4^2− produced per gram of MOC (94% conversion efficiency)	NO3^− ions oxidize SO2, yielding SO4^2− and reduced nitrogenate species. Efficiency given by both Pd and MOC cavity (increases reactant concentrations)	66
	SO2 + NO3^− → SO4^2−

---

Considering the discussion presented, it is important to note that reticular MOFs and COFs have been studied more extensively than discrete MOCs concerning gas capture and their chemical transformation, and each of these systems brings distinctive strengths and faces specific limitations in gas transformation. Undoubtedly, MOFs have been protagonist materials in this century, they are known for their ultrahigh surface areas, crystalline long-range order, and versatile metal chemistry that facilitates adsorption and heterogeneous catalysis, though their performance can be constrained by moisture/chemical stability, activation requirements, as well as gas diffusion and mass transport into the deepest pores. COFs exhibit covalent backbone robustness, thermal stability, and organic pore environments that can be tailored for selective adsorption and photocatalysis; yet, challenges remain in achieving the required high crystallinity, controlling interlayer stacking, and enhancing the catalytic conversion of gases, as they lack metal sites.

Meanwhile, MOCs can be solubilized in organic solvent or water, ideal for homogeneous catalysis and monitoring of host–guest interactions. As porosity is intrinsic to the molecule, crystalline material is not mandatory for gas capture and conversion; activity is seen for MOCs as dissolved molecules, crystalline-ordered, or amorphous solids. However, concerns regarding scalability and long-term cage integrity persist today. MOCs usually display lower thermal, hydrolytic, and chemical stability than MOFs and COFs; current and continuous efforts are helping mitigate these limitations. Rather than positioning MOCs as “better” materials, this contribution also intended to point out that MOCs complement current reticular materials and broaden the toolbox for gas conversion using porous systems.

Research on gas conversion using MOCs remains scarce compared to MOFs or COFs; however, MOCs have demonstrated strong potential as catalytic platforms, not only complementing but often matching or even surpassing the activity of reticular materials. Their unique molecular nature and solubility introduce additional effects not observed in solid frameworks. In particular, the solubility of many MOCs enhances host–guest interactions with gases and other substrates, as exemplified by systems involving fluorescein and nitroaromatics. Furthermore, cage solubility facilitates guest diffusion and retention within the cavity, enabling in-cage formation of cocatalysts, such as the Cu₄I₄ cluster or the ethanolamine–CO₂ adduct, whose confinement within the cage significantly improves gas conversion efficiency.

Such possibilities have been illustrated in this contribution. For example, CO₂ reduction using Cu₄I₄@Cu₃VMOP (containing the endogenous Cu₄I₄ cluster) achieved 100% selectivity toward HCOOH. This performance surpasses that of other MOFs, such as the Sn(II)-doped ZIF-8 reported by Geng et al.,⁶⁹ which reached 74% selectivity for HCOOH formation, and the dual-MOF copper-based catalytic electrode developed by Liu et al.,⁷⁰ which exhibited 90% efficiency for selective HCOOH production.

Only a few MOFs have been reported to adsorb H₂S and convert it into sulfides–typically as an unintended consequence. Most MOFs capable of adsorbing this polar, acidic gas suffer from poor chemical stability, often forming irreversible metal–sulfur bonds and metal sulfides (evidenced by black solids such as ZnS, CuS, or FeS), leading to cluster disintegration or partial structural collapse. These degradations are generally accelerated by the presence of water.^71,72 Distinct from this behavior, the molecular cage Ni-TFT photocatalytically decomposes H₂S in aqueous solution into hydrogen and elemental sulfur without cage decomposition or NiS formation. Another example of how MOCs can exhibit unique reactivity compared to MOFs or COFs is MOC-Pd-NO₃, where nitrate counterions and confined molecular space promote SO₂ oxidation to sulfate. Cage integrity is maintained thanks to its solubility and anion-exchange capability. Currently, the efficient, quantitative, and clean conversion of SO₂ into more benign and useful byproducts, while allowing for adsorbent recovery, is unparalleled in the state-of-the-art of reticular or molecular gas adsorbents such as MOFs, COFs, HBFs, or POCs.

3. Conclusions

The examples discussed in this contribution underscore the remarkable potential of MOCs as platforms for the capture and chemical transformation of gases into value-added products. By mimicking the selective and efficient microenvironments of enzymes, MOCs offer a unique opportunity to extend host–guest chemistry to gaseous substrates, permitting transformations that are both catalytically improved and spatially controlled, and may take place in gas–liquid–solid phase. Across the systems examined, it is evident that molecular confinement and the ability of cage cavities to encapsulate gases, catalysts, and co-substrates are central to increasing reactivity and selectivity. These confined environments stabilize reactive intermediates, preorganize molecular components, and enforce intramolecular electron transfer, all of which contribute to improved catalytic performance. The size, shape, and functionalization of the cavity directly influence substrate accessibility, intermediate stabilization, and reaction pathways, making confinement a key design principle for gas-phase catalysis.

The examples discussed herein highlight key structural features that render MOCs suitable for gas conversions. These features can be regarded as design principles for developing novel gas transformations within the molecular interiors of MOCs: (1) cavity: accessible internal cavities of appropriate dimensions enable gas diffusion, substrate accommodation, and size selectivity. Larger ligands can increase pore volume, improving catalytic performance for bulkier substrates and allowing co-catalyst encapsulation. A balance between size and polarity is crucial: enclosed polar environments within cages promote CO₂ capture, enhance affinity for polar gases (e.g., H₂S, SO₂), and influence selectivity for polar intermediates and products.

(2) Stability: robust frameworks that maintain integrity under harsh catalytic conditions-including aqueous environments, wide pH ranges, high temperature, or corrosive species-are essential. Metal nodes dictate coordination geometry and stability; thus, kinetically inert metals should be considered to ensure resistance under acidic, basic, or high-temperature conditions. Rigid, bulky metal nodes further prevent degradation in the working medium.⁷³

(3) Metal centers: in some cases, metal sites of the cage framework have been observed to play a pivotal role in gas conversion by activating substrates, stabilizing intermediates, and enabling redox processes. Open-metal sites that coordinate and immobilize gaseous guests within the cage should allow labile metal–ligand bonds and ligand exchange without compromising structural integrity. This can be achieved using metals with flexible coordination geometries (e.g., Zn(II), Co(II), Cu(II)). Metals that exhibit multiple oxidation states (such as Fe(III)/Fe(II), Ni(II)/Ni(I), or Cu(II)/Cu(I)) and have labile coordination sites may facilitate electron transfer and substrate exchange.

(4) Functionalized ligands: organic ligands define the shape and polarity, and influence the stability of MOCs through the number of their coordination sites (e.g., ditopic, tritopic, tetratopic). Functionalization is thus essential for host–guest recognition and molecular preorganization. Introducing hydrogen-bond donors/acceptors, aromatic centers, or charged groups enhances substrate binding via hydrogen bonding or π-stacking interactions. Additionally, ligands should possess a degree of flexibility to allow for dynamic complex formation; incorporating methylene groups or weakly coordinating amines can provide this necessary adaptability. In summary, rational MOC design requires balancing structural stability with dynamic adaptability, optimizing metal–ligand interactions, and tailoring cavity size and polarity to achieve high catalytic efficiency and selectivity.

While the use of MOCs for gas conversion is still in its early stages, the results presented here highlight their promise and versatility. We anticipate that future advances will focus on stimuli-responsive or adaptive cages, capable of dynamically adjusting cavity size, polarity, or functionality to accommodate diverse gases and intermediates. Moreover, the development of multifunctional cages that co-encapsulate catalysts, photosensitizers, and substrates could allow tandem or cascade reactions, further expanding the scope of MOC-based catalysis. In this sense, the integration of computational tools for guided design will be helpful to predict host–guest interactions and optimize cavity environments for specific gas reactions. Additionally, finding methods for embedding MOCs into hybrid catalytic materials—such as porous liquids, catalytic membranes, flow reactors or electrodes^74,75—will enhance scalability, recyclability, and process integration,⁷⁶ moving beyond gas capture toward gas conversions.

Ultimately, the field of gas transformations using porous materials and molecular cages holds immense potential for environmental remediation, energy storage, and sustainable chemical synthesis. By enabling the selective conversion of reactive, harmful, pollutant gases into useful products such as fuels, fine chemicals, or fertilizers, MOCs offer a promising route toward green and circular chemical processes, positioning them as versatile platforms for addressing some of the global challenges in pollution control and renewable energy.

Conflicts of interest

There are no conflicts to declare.

Data availability

No primary research results, software or code have been included and no new data were generated or analyzed as part of this review.

Acknowledgements

This work was supported by the Dirección General de Asuntos del Personal Académico from the DGAPA UNAM-PAPIIT program, Mexico (grant IN212223). F. T. T. is grateful to SECIHTI Mexico for a graduate scholarship (CVU 2038873, grant number 4047239).

References

1. Y.-L. Zhao, X. Zhang, M.-Z. Li and J.-R. Li, Chem. Soc. Rev., 2024, 53, 2056.
2. L. F. Vega, D. Bahamon and I. I. I. Alkhatib, ACS Sustainable Chem. Eng., 2024, 12, 5357.
3. M. Yousaf, M. Zaman, A. Mahmood, M. Imran, A. Elkamel, M. Rizwan, T. Wilberforce and F. Riaz, Energy Sci. Eng., 2022, 10, 4890.
4. B. Yao, T. Xiao, O. A. Makgae, X. Jie, S. Gonzalez-Cortes, S. Guan, A. I. Kirkland, J. R. Dilworth, H. A. Al-Megren, S. M. Alshihri, P. J. Dobson, G. P. Owen, J. M. Thomas and P. P. Edwards, Nat. Commun., 2020, 11, 6395.
5. L. Zhang, C. Wang and C. Zhan, Mater. Chem. Front., 2025, 9, 1650.
6. A. Jana, A. Maity, A. Adalder, S. Saha and A. Bhunia, Nanoscale, 2025, 17, 12235.
7. W. Shan, Z. Wang, C. Gao, X. Li, W. Zhuang, R. Liu, C. Shi, H. Qin, X. Li and D. Shi, Green Chem., 2024, 26, 5744.
8. J. Gandara-Loe, L. Pastor-Perez, L. F. Bobadilla, J. A. Odriozola and T. R. Reina, React. Chem. Eng., 2021, 6, 787.
9. J. G. Doremus, B. Lotsi, A. Sharma and P. L. McGrier, Nanoscale, 2024, 16, 21619.
10. Q. Song, Z. Li, Q. Peng, X. Ou, K. Zhao, Z. Dong, L. Ye, C. Liu, Y. Liu, J. He, L.-H. Chung and B. Han, Appl. Catal., B, 2025, 378, 125619.
11. Y. Li, J. Zhang, K. Zuo, Z. Li, Y. Wang, H. Hu, C. Zeng, H. Xu, B. Wang and Y. Gao, Angew. Chem., Int. Ed., 2024, 63, e202405476.
12. S. Zhang, Y. Zhuo, C. I. Ezugwu, C.-C. Wang, C. Li and S. Liu, Environ. Sci. Technol., 2021, 55, 8341.
13. Z. Alsudairy, N. Brown, A. Campbell, A. Ambus, B. Brown, K. Smith-Petty and X. Li, Mater. Chem. Front., 2023, 7, 3298.
14. J. Song, Z. Zhang, S. Hu, T. Wu, T. Jiang and B. Han, Green Chem., 2009, 11, 1031.
15. H. Xu, X.-F. Liu, C.-S. Cao, B. Zhao, P. Cheng and L.-N. He, Adv. Sci., 2016, 3, 1600048.
16. D. Zhao, X.-H. Liu, C. Zhu, Y.-S. Kang, P. Wang, Z. Shi, Y. Lu and W.-Y. Sun, ChemCatChem, 2017, 9, 4598.
17. W.-Y. Gao, H. Wu, K. Leng, Y. Sun and S. Ma, Angew. Chem., Int. Ed., 2016, 55, 5472.
18. R. Chakrabarty, P. S. Mukherjee and P. J. Stang, Chem. Rev., 2011, 111, 6810.
19. T. R. Cook and P. J. Stang, Chem. Rev., 2015, 115, 7001.
20. A. J. McConnell, Chem. Soc. Rev., 2022, 51, 2957.
21. E. G. Percástegui, T. K. Ronson and J. R. Nitschke, Chem. Rev., 2020, 120, 13480.
22. F. J. Rizzuto, L. K. S. von Krbek and J. R. Nitschke, Nat. Rev. Chem., 2019, 3, 204.
23. E. G. Percástegui, Eur. J. Inorg. Chem., 2021, 2021, 4425.
24. M. D. Ward, C. A. Hunter and N. H. Williams, Acc. Chem. Res., 2018, 51, 2073.
25. T. Y. Kim, R. A. S. Vasdev, D. Preston and J. D. Crowley, Chem. – Eur. J., 2018, 24, 14878.
26. S. Cruz-Nava, S. de J. Valencia-Loza and E. G. Percástegui, Eur. J. Org. Chem., 2022, e202200844.
27. M. Yoshizawa and L. Catti, Acc. Chem. Res., 2019, 52, 2392.
28. A. Brzechwa-Chodzyńska, W. Drożdż, J. Harrowfield and A. R. Stefankiewicz, Coord. Chem. Rev., 2021, 434, 213820.
29. A. J. Plajer, E. G. Percástegui, M. Santella, F. J. Rizzuto, Q. Gan, B. W. Laursen and J. R. Nitschke, Angew. Chem., Int. Ed., 2019, 58, 4200.
30. N. Dey and C. J. E. Haynes, ChemPlusChem, 2021, 86, 418.
31. C. García-Simón, M. Garcia-Borràs, L. Gómez, T. Parella, S. Osuna, J. Juanhuix, I. Imaz, D. Maspoch, M. Costas and X. Ribas, Nat. Commun., 2014, 5, 5557.
32. K. Wu, K. Li, Y.-J. Hou, M. Pan, L.-Y. Zhang, L. Chen and C.-Y. Su, Nat. Commun., 2016, 7, 10487.
33. T. Grancha, A. Carné-Sánchez, L. Hernández-López, J. Albalad, I. Imaz, J. Juanhuix and D. Maspoch, J. Am. Chem. Soc., 2019, 141, 18349.
34. S. K. Samanta, D. Moncelet, V. Briken and L. Isaacs, J. Am. Chem. Soc., 2016, 138, 14488.
35. Y.-R. Zheng, K. Suntharalingam, T. C. Johnstone and S. J. Lippard, Chem. Sci., 2015, 6, 1189.
36. G. Montà-Gongález, D. Bastante-Rodríguez, A. García-Fernández, P. J. Lusby, R. Martínez-Máñez and V. Martí-Centelles, Chem. Sci., 2024, 15, 10010.
37. G. Montà-González, R. Martínez-Máñez and V. Martí-Centelles, Targets, 2024, 2, 372.
38. E. G. Percástegui, Chem. Commun., 2022, 58, 5055.
39. Y. Gao, S.-Q. Deng, X. Jin, S.-L. Cai, S.-R. Zheng and W.-G. Zhang, Chem. Eng. J., 2019, 357, 129.
40. D. Camdzic, H. K. Welgama, M. R. Crawley, A. Avasthi, T. R. Cook and D. S. Aga, ACS Appl. Eng. Mater., 2024, 2, 87.
41. Q.-W. Zeng, L. Hu, Y. Niu, D. Wang, Y. Kang, H. Jia, W.-T. Dou and L. Xu, Chem. Commun., 2024, 60, 3469.
42. W. Li, L. Wang, B. E. Keshta, Y. Bi, Y. Zhang and B. Chen, Coord. Chem. Rev., 2025, 545, 217032.
43. J.-L. Zhu, D. Zhang, T. K. Ronson, W. Wang, L. Xu, H.-B. Yang and J. R. Nitschke, Angew. Chem., Int. Ed., 2021, 60, 11789.
44. C. A. Rowland, G. R. Lorzing, A. J. Gosselin, B. A. Trump, G. P. A. Yap, C. M. Brown and E. D. Bloch, J. Am. Chem. Soc., 2018, 140, 11153.
45. A. Carné-Sánchez, J. Martínez-Esaín, T. Rookard, C. J. Flood, J. Faraudo, K. C. Stylianou and D. Maspoch, ACS Appl. Mater. Interfaces, 2023, 15, 6747.
46. G. R. Lorzing, A. J. Gosselin, B. A. Trump, A. H. P. York, A. Sturluson, C. A. Rowland, G. P. A. Yap, C. M. Brown, C. M. Simon and E. D. Bloch, J. Am. Chem. Soc., 2019, 141, 12128.
47. X. Feng, X. Wang, H. Yan, H. Liu, X. Liu, J. Guan, Y. Lu, W. Fan, Q. Yue and D. Sun, Angew. Chem., Int. Ed., 2024, 63, e202407240.
48. W. Liu and J. F. Stoddart, Chem, 2021, 7, 919.
49. H. Takezawa, K. Shitozawa and M. Fujita, Nat. Chem., 2020, 12, 574.
50. M. Morimoto, S. M. Bierschenk, K. T. Xia, R. G. Bergman, K. N. Raymond and F. D. Toste, Nat. Catal., 2020, 3, 969.
51. L.-D. Syntrivanis and K. Tiefenbacher, Angew. Chem., Int. Ed., 2024, 63, e202412622.
52. S. Bhattacharyya, M. R. Black and D. S. Pilgrim, Chem. Sci., 2025, 16, 21238.
53. J.-N. Lu, J.-J. Liu, L.-Z. Dong, J.-M. Lin, F. Yu, J. Liu and Y.-Q. Lan, Angew. Chem., Int. Ed., 2023, 62, e202308505.
54. J.-Q. Chen, Q.-H. Li, Q.-L. Hong, P. Shao, H.-X. Zhang and J. Zhang, Sci. China: Chem., 2025, 68, 943.
55. C. He, Y.-H. Zou, D.-H. Si, Z.-A. Chen, T.-F. Liu, R. Cao and Y.-B. Huang, Nat. Commun., 2023, 14, 3317.
56. Y. Yang, M. Dong, Q. Wu, C. Qin, W. Chen, Y. Geng, S. Wu, C. Sun, K. Shao, Z. Su and X. Wang, Angew. Chem., Int. Ed., 2025, 64, e202423018.
57. Y. Wang, E. Chen and J. Tang, ACS Catal., 2022, 12, 7300.
58. Y.-L. Lu, Y.-P. Wang, K. Wu, M. Pan and C.-Y. Su, Acc. Chem. Res., 2024, 57, 3277.
59. S. Ganta, J.-H. Borter, C. Drechsler, J. J. Holstein, D. Schwarzer and G. H. Clever, Org. Chem. Front., 2022, 9, 5485.
60. Z. Zhang, Y. Yang, J. Wang, X. Jing and C. Duan, Inorg. Chem. Front., 2025, 12, 5389.
61. X. Jing, Y. Yang, C. He, Z. Chang, J. N. H. Reek and C. Duan, Angew. Chem., Int. Ed., 2017, 56, 11759.
62. G. Ma, H. Yan, X. Zong, B. Ma, H. Jiang, F. Wen and C. Li, Chin. J. Catal., 2008, 29, 313.
63. W. Mulewa and M. Tahir, Energy Fuels, 2024, 38, 15972.
64. Y. Yang, H. Li, X. Jing, Y. Wu, Y. Shi and C. Duan, Chem. Commun., 2022, 58, 807.
65. S. J. Smith, J. van Aardenne, Z. Klimont, R. J. Andres, A. Volke and S. Delgado Arias, Atmos. Chem. Phys., 2011, 11, 1101.
66. E. G. Percástegui, E. Sánchez-González, S. de J. Valencia-Loza, S. Cruz-Nava, V. Jancik and D. Martínez-Otero, Angew. Chem., Int. Ed., 2025, 64, e202421169.
67. S. de J. Valencia-Loza, A. López-Olvera, E. Martínez-Ahumada, D. Martínez-Otero, I. A. Ibarra, V. Jancik and E. G. Percástegui, ACS Appl. Mater. Interfaces, 2021, 13, 18658.
68. T. Liu, A. W. H. Chan and J. P. D. Abbatt, Environ. Sci. Technol., 2021, 55, 4227.
69. W. Geng, W. Chen, G. Li, X. Dong, Y. Song, W. Wei and Y. Sun, ChemSusChem, 2020, 13, 4035.
70. S. Liu, L. Song, R. Liu, L. Li, D. Yang, S. Yuan and X. Dai, Small, 2023, 19, 2304808.
71. E. Martínez-Ahumada, A. López-Olvera, V. Jancik, J. E. Sánchez-Bautista, E. González-Zamora, V. Martis, D. R. Williams and I. A. Ibarra, Organometallics, 2020, 39, 883.
72. A. López-Olvera, J. G. Flores, J. Aguilar-Pliego, C. K. Brozek, A. Gutiérrez-Alejandre and I. A. Ibarra, Chem. Mater., 2021, 33, 6269.
73. M.-T. Lv, M.-D. Cui, K.-P. Bai, Y. Jiang, W.-P. Chen, N.-N. Sun, X.-Q. Hu, C. Huang, Q.-Y. Yang and Y.-Z. Zheng, Angew. Chem., Int. Ed., 2025, 64, e202506838.
74. L. Rao, X. Zhang, P. Liu, W.-T. Dou, H.-Y. Lin, F. Zhao, W. Zhu, X. Qian, H.-B. Yang and L. Xu, J. Am. Chem. Soc., 2025, 147, 31618.
75. H.-Y. Lin, L.-Y. Zhou, F. Mei, W.-T. Dou, L. Hu, H.-B. Yang and L. Xu, Angew. Chem., Int. Ed., 2023, 62, e202301900.
76. T. Jin, K. Zeng, X. Zhang, W.-T. Dou, L. Hu, D. Zhang, W. Zhu, X. Qian, H.-B. Yang and L. Xu, Angew. Chem., Int. Ed., 2024, 63, e202409878.

---