Supramolecular chemistry-based materials on SO₂ capture: recent advances

Abstract

The urgent need for efficient methods to detect, monitor, and capture toxic and hazardous sulfur dioxide (SO₂) has driven the development of new devices for this purpose. Due to its significant corrosiveness and high toxicity, a key requirement for adsorbent materials is their ability to withstand deterioration, alongside possessing high gas sensitivity and selectivity. Within this domain, supramolecular chemistry has made substantial inroads through the development of innovative porous materials formed by the self-assembly of molecular building blocks, driven by diverse interactions. The complex architectures that result can be categorised as either extended networks, such as metal–organic frameworks (MOFs) and covalent organic frameworks (COFs), or as molecular cages, including metal–organic cages (MOCs) and porous organic cages (POCs). These materials have demonstrated high capture capacities, extremely sensitive detection limits, and even active sites for the catalytic transformation of adsorbed molecules. This review aims to highlight the primary contributions of these supramolecular materials to the experimental adsorption of SO₂, along with their most remarkable results.

---

From left to right: Ricardo Peralta, Pablo Marín-Rosas, Antonio Hernández-Monsalvo, Enrique Lima and Ilich Ibarra

Ricardo Peralta is a Full Professor at UAM who is interested in novel catalytic applications of semi-open metal sites within MOFs. Pablo Marín-Rosas is a PhD Student who is exploring MOF materials for catalytic applications. Antonio Hernández-Monsalvo works on his Master's degree, investigating chemically stable MOFs for the sensing of toxic chemicals. Enrique Lima serves as the Vice-Principal of the Institute for Materials Research at UNAM, and his research focuses on bio-applications of MOF materials. Ilich Ibarra is a Full Professor at UNAM, working on the use of MOFs for the capture and detection of toxic gases such as SO2 and H2S.

Introduction

Today, we are constantly exposed to a vast array of air pollutants. Among these, some pose a particularly significant risk, not only to human health but also to the delicate balance of ecosystems and the natural cycles of the environment. Such is the case of sulphur dioxide (SO₂), a corrosive and colourless gas that the World Health Organization (WHO) classifies as one of the most toxic chemicals for humans.¹ What makes this gas particularly alarming is not just its rapid spread, owing to its high solubility, or its devastating effects on aquifers and plant life through acid rain and the disruption of the natural sulphur cycle.² For example, a brief exposure to trace concentrations of this chemical – a mere 1.5 parts per million (ppm) for as little as 10 minutes – can instantly cause breathing difficulties in a healthy individual.³

In fact, air quality guidelines stipulate that the maximum tolerable human exposure to SO₂ is 175 parts per billion (ppb), which equates to 500 μg m⁻³, for ten-minute intervals, and a daily average of 20 μg m⁻³ (8 ppb).⁴ Exceeding these limits can lead to serious health issues, including chronic inflammation or obstruction in the respiratory system and respiratory tract infections.⁵ Even a few minutes of exposure to 100 ppm can be fatal due to the rapid absorption of SO₂ through the skin and respiratory tract.^1,6 Furthermore, SO₂ acts as a precursor to secondary pollutants, such as fine particulate matter (PM2.5), which are also globally recognized as highly detrimental to health.⁷ The fact that such minute concentrations can be so harmful underscores the urgent need for highly efficient and sensitive methods to detect, monitor, and potentially capture SO₂ in environments that are at risk of contamination.

In recent years, sophisticated porous materials like metal–organic frameworks (MOFs) and molecular cages (MCs) have demonstrated the ability to achieve this crucial and precise detection.⁸ It has been previously demonstrated that these materials exhibit excellent performance in both SO₂ capture and detection, not only maximizing gas uptake and storage, but also enabling cyclic reuse without compromising the inherent properties that make them such effective and sensitive adsorbents.

MOFs represent one class of porous crystalline materials based on supramolecular chemistry; they are characterized by the self-assembly of metal ions with organic ligands via coordination interactions,⁹ whereas a covalent bonded assembly of light elements as building blocks (e.g., C, H, O, N, and B) leads to the formation of covalent organic frameworks (COFs)¹⁰ Similarly, supramolecular porous materials, such as molecular cages, can be generated by an analogous process. Molecular cages have internal cavities suitable for molecular storage, employing a host–guest system, and can also be differentiated by the nature of their constituent bonds. When coordination interactions are involved, it leads to metal–organic cages (MOCs),¹¹ and porous organic cages (POCs) are obtained through the self-assembly of organic monomers linked by covalent bonds.¹²

The reasons why MOFs, COFs, MOCs, and POCs are highly promising candidates for corrosive gas capture and sensing applications lie in the chemistry behind their formation, which is rooted in supramolecular chemistry. One of the fundamental principles of supramolecular chemistry is spontaneous self-assembly, which refers to the ability of molecules to selectively couple, as building blocks, to facilitate the formation of more complex and well-organized structures through forces considerably weaker than covalent bonds (e.g., electrostatic forces, π–π interactions, hydrogen bonds and van der Waals forces).^9,13 The structure they adopt is derived from their self-assembly, which is responsible for their high porosity, providing both high adsorption capacity and a relatively high surface area.

Moreover, the specific types of interactions that maintain the structure of these complex materials lead to the formation of active sites, which function on a lock-and-key principle. When the interaction with the guest molecule (i.e., SO₂) is favourable, the active site (host) will facilitate its binding through non-covalent forces.¹³ This exemplifies another fundamental principle of supramolecular chemistry: molecular recognition (Fig. 1). Thus, the combination of high porosity and molecular recognition provides these materials with significant advantages compared to other adsorbents.

Fig. 1 Molecular recognition interaction between active sites of (a) MOF2⁰ and (b) COF1⁰ building blocks with SO₂ molecules.

Porous complex materials exhibit pronounced sensitivity, enhanced by guest–guest packing interactions that facilitate a more ordered arrangement of SO₂ within the pores,⁸ enabling the detection of trace amounts of SO₂. Some of these materials even possess robust structural and chemical stability, which supports facile regeneration for cyclical use without loss of crystallinity or porosity (material degradation).^10–12 In addition, the uptake of SO₂ in the internal cavities of these materials can trigger measurable changes in their optical, electrical, or mechanical properties. This effect can be applied in the development of molecular sensors, employing techniques such as spectroscopy or electrical conductivity measurements.¹⁰

Thus, the objective of this work is to discuss the latest contributions to the meticulous design and application of these supramolecular porous materials in SO₂ capture and detection. Furthermore, it aims to encourage other research groups to continue exploring the frontiers of this field, pushing the boundaries of what this branch of chemistry and materials science can achieve in terms of even greater performance.

Advances in chemistry-based materials

The binding of SO₂ to various materials of interest exhibits specific characteristics for each. The interactions governing the adsorption of SO₂ molecules onto the surface are contingent upon the chemical and electronic properties of the surface. The active sites responsible for SO₂ binding on the material's surface are characterised by inherent unsaturation of the binding forces between the constituent atoms of the solid structure.¹⁴ Characteristic features such as pores, surface defects, and even specific functional groups serve as localities where external molecules, such as SO₂, can readily adsorb through tailored interactions favoured by the intrinsic properties of these sites.

The two oxygen atoms within the SO₂ molecule predominantly act as Lewis bases, exhibiting non-bonding electrons that are readily available for donation.¹⁵ Depending on the specific adsorbent material, these oxygen atoms can serve as the primary binding sites for the molecule. Their versatility allows them to engage in a spectrum of interactions, ranging from strong coordination with highly charged, poorly polarizable, low-molecular-mass metal ions (hard metals) to both strong and weak hydrogen bonds with a variety of functional groups.¹⁶

In contrast, the sulphur atom behaves as a Lewis acid, which consequently shows good affinity for accepting external electrons, making it particularly good at generating coordination bonds with electron-rich transition metals and weak S–O interactions. Similarly, if the material is porous and free of metal ions, van der Waals forces, weak noncovalent, and even covalent bonds can occur in the cavities.^16,17 However, these interactions are not mutually exclusive; rather, combinations of different host–guest interactions can coexist within the same material.

The quantity and spatial arrangement of active sites across the solid's surface are intricately linked to the material's electronic and geometric structure. These sites can exhibit preferential affinity for certain molecules over others, thereby imparting specificity and rendering the adsorption process selective. Consequently, this leads to a preferential order in the bonds formed between SO₂ molecules and the binding sites of porous supramolecular materials.

SO₂-MOFs

Given the highly corrosive and reactive nature of SO₂, the adsorbents employed for desulfurization must demonstrate exceptional stability and reversibility. Metal–organic frameworks (MOFs) are no exception to this requirement. The strong coordination that can form between the sulphur atom of SO₂ and the metallic nodes within the MOF structure can, in fact, surpass the strength of the metal–ligand bonds. Consequently, this can lead to irreversible adsorption and even boost the material's degradation.¹⁶ Such deterioration may manifest as a loss of crystallinity or a reduction in surface area stemming from the permanent occupation of active sites.^17,19

Indeed, adsorption reversibility and the prevention of material deterioration are just as critical as gas capture capacity and have posed significant challenges in optimizing the performance of MOFs for SO₂ adsorption, particularly for applications requiring cyclic operation.

The pursuit of SO₂-adsorbing MOFs exhibiting robust stability without compromising high uptake capacities was initially spurred by MOF-177.¹⁷ This material demonstrates an uptake of 25.7 mmol g⁻¹ (at 293 K and 1 bar) but undergoes significant deterioration following SO₂ adsorption. Subsequently, materials like UR3-MIL-101(Cr)¹⁸ and MFM-422¹⁹ also encountered this challenge. While achieving even higher uptakes of 36.7 and 31.3 mmol g⁻¹ (at 293 K and 1 bar), they also suffered from chemical instability, surface degradation, or pore blockage, leading to reduction in porosity after the initial adsorption cycles (Fig. 2).

Fig. 2 Loss in gas maximum uptake for MOFs: (a) MIL-101(Cr)¹⁸ and (b) MOF-177¹⁷ after SO₂ exposure.

More recently, significant contributions to this research have been presented with MOFs MIL-101(Cr)-4F(1%)⁸ and Mg₂(dobpdc).²⁰

These materials share key attributes, including noteworthy SO₂ capture capacities of 18.4 mmol g⁻¹ and 19.5 mmol g⁻¹, respectively (at 293 K and 1 bar) and remarkable resilience, by retaining their structural integrity, crystallinity, and surface area after SO₂ adsorption under both dry and humid conditions, even after 50 cycles (Fig. 3). This underscores complete reversibility in the adsorption process, along with selective adsorption of SO₂ over other gases such as CO₂ or N₂.

Fig. 3 SO₂ experimental adsorption–desorption isotherms and cycles for: (a) and (c) MIL-101(Cr)-4F(1%)⁸ and (b) and (d) Mg₂(dobpdc)²⁰ at 298 K and up to 1 bar.

Additionally, both MIL-101(Cr)-4F(1%) and Mg₂(dobpdc) exhibit sensitive detection limits at low pressures, positioning them as promising candidates for the development of detectors and separators for gas mixtures containing SO₂.

Following this important contribution, further advancements have continued to enhance the performance of SO₂-adsorbing MOFs. State-of-the-art materials, such as the highly porous MOF Zr-TPA²¹ and the MOF designated SJTU-220,²² have achieved impressive sorption capacities of 22.7 mmol g⁻¹ and 29.6 mmol g⁻¹ (at 298 K and 1 bar), respectively, both exhibiting excellent recyclability and sustained stability. Ongoing research in this application of MOFs is significantly contributing to the development and refinement of efficient SO₂ adsorbent solids, while continuous progress takes us deeper into our understanding of the molecular interactions governing SO₂ sorption on these supramolecular materials.

SO₂-COFs

Similar to MOFs, covalent organic frameworks (COFs) are highly porous materials characterised by extended crystalline structures formed by exclusively linking organic molecular building blocks via robust covalent bonds. These inherent bonds endow COFs with exceptional stability, rendering their crystalline structure remarkably resistant to modification, even upon interaction with external agents or under harsh conditions. Consequently, they are inherently appealing for versatile adsorption and storage applications involving hazardous gases, including corrosive ones like SO₂, which contribute to atmospheric pollution.²³

The inherent nature of the cross-linked functional groups within COF structures provides a diverse array of host–guest binding sites for SO₂ molecules. These interactions are not limited to a single type; rather, they concurrently promote a wide spectrum of bonds with different binding strengths. While the π-electron densities present within the building blocks (i.e., aromatic rings and nitrogen atoms' lone pairs) strongly interact with adsorbed SO₂, potentially leading to hysteresis, weaker bonds like hydrogen bonds exert a nearly imperceptible individual effect. However, collectively, these interactions facilitate both in-plane and out-of-plane adsorption along the structure of the mesoporous material.^10,23

Despite these favourable attributes, the exploration of COFs as SO₂ adsorbents has been relatively limited. PI-COF-m10 stands out as an early example in this area; this imine-based COF was functionalized with DMMA as a modulator to lower the energy barrier for SO₂ desorption. While complete reversibility was achieved over 5 cycles with an adsorption capacity of 6.30 mmol SO₂ g⁻¹, the incorporation of the modulator resulted in alterations to the crystal structure, leading to an increase in amorphous regions and a reduction in the overall sorption capacity.^23,24

The integration of active groups or metallic atoms to dope the structure of COFs has also been proposed to improve the efficiency of these materials in capturing specific molecules such as SO₂. In COF-DC-8, the nickel atoms found throughout its 2D structure improve its three-dimensional electronic conductivity, compared to a pure organic solid. This increases the potential to amplify the alignment of π-conjugated units and improves charge transport through space, achieving increased sensitivity and being able to quantify gas molecules adsorbed on its surface within the order of ppb through changes in chemoresistivity.^23,25

In the case of COF-105, density functional theory (DFT) calculations indicate that incorporating a single transition metal atom, Sc(III), within its building blocks can significantly promote the formation of adsorption complexes between the adsorbed SO₂ gas molecule and the metal-doped COF. This doping strategy is predicted to enhance the interaction energy by up to an order of magnitude compared to the undoped material. This enhancement is attributed to donor–acceptor delocalization occurring between a lone pair of electrons on an oxygen atom of SO₂ and the Sc(III) atom within the COF framework.^23,26

Against this scenery, the most recent significant advancement in this field has been the development of SonoCOF-9.¹⁰ Functionalized with imine and aromatic units, this material demonstrates remarkable chemical stability in the presence of SO₂, maintaining a consistent capture capacity of 3.5 mmol g⁻¹ over 50 adsorption–desorption cycles without any loss in crystallinity or porosity. Moreover, its strong interaction energy with SO₂ renders it highly selective and sensitive to this gas, capable of detecting concentrations as low as 6.4 ppb, when THF is used as a dispersing medium.

Additionally, fluorescence spectroscopy can be used to make SonoCOF-9 perform as a sensitive SO₂ sensor. When SO₂ molecules adsorb onto its surface, the intensity of its emission spectrum dramatically decreases (Fig. 4). This contribution underscores the considerable potential of this material for detection, without the need for doping, while also exhibiting notable selectivity for the recognition and separation of gas mixtures, since the emission lessening described does not occur with other adsorbates.

Fig. 4 SonoCOF-9: (a) structural interaction with SO₂ molecules, (b) photoluminescence emission spectrum with air, CO₂, and after SO₂.¹⁰

Research into the application of COFs for the adsorption of corrosive gases such as SO₂ is anticipated to undergo continued expansion and intensification, to achieve optimal performance for these materials in the years ahead. The ongoing exploration of the potential in this class of materials is poised to unlock novel applications and insights that currently remain undiscovered.

SO₂-MOCs

Metal–organic cages (MOCs) are also supramolecular structures formed through the self-assembly of metal ions and organic ligands through coordination interactions. This process yields highly symmetrical, nanometric structures featuring pseudospherical cavities that are ideal for encapsulating spherical guest molecules.²⁷ Moreover, MOCs demonstrate high solubility in polar solvents, including water.²⁸ Consequently, these properties make MOCs excellent candidates for molecular capture and separation applications.

Reports continuously emerge showcasing the enhanced and selective adsorption of gaseous compounds by these materials, whether in the gas phase, in solution, or even when dispersed within polymer membranes.²⁹ Beyond adsorption, their potential for heterogeneous catalysis is also being actively investigated. For instance, recent research highlights the successful application of MOCs in efficiently and selectively catalysing the conversion of chemical species such as iodine and carbon dioxide (CO₂) on their surfaces.^27,30 However, the capture of SO₂ using MOCs has remained largely unexplored to date.

The initial study exploring the interaction of SO₂ with MOCs was published in 2021.²⁸ A palladium-based MOC, specifically [Pd₆L₈](NO₃)₃₆ (referred to as cage 1) was selected for this investigation due to its well-defined hexapalladium(II) matrix. This material, exhibiting a pseudo-cubic molecular structure, demonstrated a remarkable SO₂ adsorption capacity of 6.0 mmol g⁻¹ at 1.0 bar, especially considering its relatively modest surface area of 111 m² g⁻¹. This significant finding underscored the effectiveness of cage 1 for competent SO₂ adsorption.

Further analysis revealed that the Pd(II) binding sites within cage 1 form strong bonds with SO₂ molecules, promoting their oxidation to sulphate species. The presence of hysteresis during gas adsorption, indicative of structural changes, confirmed the chemisorption of SO₂ onto the metal centres (Fig. 5a). The adsorption enthalpy (ΔH_ads), calculated using the Clausius–Clapeyron method, was approximately −50 J mol⁻¹, signifying a strong interaction. However, this robust cage–SO₂ interaction presents a challenge to the cyclability of the material. Cage 1 experienced a drastic loss of over 99% of its reported adsorption capacity between the first and second cycles. This rapid decline strongly suggests that the adsorbate immediately occupies the primary binding sites, even at low SO₂ concentrations.

Fig. 5 (a) Change in FTIR spectra before (blue) and after (black) SO₂ exposure²⁸ and (b) catalytic cycle of the Pd–NO₃ MOC (cage 1). (c) SO₂ adsorption–desorption isotherms for palladium-based MOCs with different counterions.¹¹

Following this strong binding of SO₂ gas within cage 1, a significant process of oxidation occurred. The presence of highly oxidising nitrate ions facilitated anion metathesis reactions, effectively converting the initial nitrate salt, [Pd₆L₈](NO₃)₃₆, into a sulphate salt, [Pd₆L₈](SO₄)₁₈. Thus, the overarching chemisorption mechanism observed in cage 1 can be summarised as a two-step process: (1) the strong direct interaction of SO₂ molecules directly with the Pd(II) metal centres, and (2) subsequent oxidation to sulphate species, driven by the nitrate counterions (see Fig. 5b).

Subsequent research reported in 2024¹¹ further investigated the application of this Pd(II)-based cage as a catalyst for the surface-mediated conversion of SO₂ into less hazardous and more manageable sulphate salts. This later work explored the impact of various counterions (PF₆⁻, BF₄⁻, or SO₄²⁻) on the binding and subsequent oxidation process, moving beyond the initially studied nitrate (NO₃⁻).

Interestingly, while NO₃⁻ proved to be the most effective counterion for promoting catalysis, the palladium MOCs incorporating these new counterions still exhibited a high adsorption capacity for SO₂. Crucially, with these alternative counterions, the adsorption process became reversible (physisorption), significantly enhancing the material's cyclability (Fig. 5c). This allowed for up to five effective cycles of SO₂ capture and release.

Another important contribution of this study was the demonstration of binding selectivity for SO₂ over CO₂ during surface capture and transformation. A notable selectivity of 78.3 (molar ratio 10 : 90) was achieved, which highlights the potential of these MOCs in targeted SO₂ removal in mixed gas streams.

The successive investigation into cage 1 for the capture and posterior oxidation of SO₂ has unveiled previously undocumented characteristics for MOCs interacting with this chemical species (see Table 1). These findings emphasize several key properties that position these MOCs as potential catalysts for SO₂ transformation, even within complex gas mixtures. The MOCs demonstrated efficient capture and conversion of SO₂, with the rate being dependent on the specific counterions incorporated.

Table 1 Comparison of the experimental performance of various stable supramolecular materials for sulfur dioxide (SO₂) adsorption

Material	Uptake (0.01–1 bar) at 298 K (mmol g^−1)	Surface area (m^2 g^−1)	Adsorption type	Reversibility	ΔH° ads (kJ mol^−1)	Selectivity (SO2 : CO2) molar ratio	No. of cycles	Material deterioration	Ref.
MOF: MIL-101(Cr)-4F(1%)	4.6–18.4	2176	Physisorption (hysteresis at low pressures)	Reversible	−54.3	Not compared	50	Not observed	8
MOF: Mg2(dobpdc)	3.37–19.5	3300	Physisorption	Reversible	−36.6	Greater affinity for SO2	50	Water retention (wet conditions)	20
COF: SonoCOF-9	0.9–3.5	1147	Physisorption	Reversible	−42.3	12.93 (10 : 90) 25.44 (1 : 99)	50	Not observed	10
MOC: Pd6L8 Cage 1	3.2–6.0	111	Chemisorption/physisorption	Irreversible with NO3 counterions. Reversible with other counterions.	−48.8 to −64.4	78.3 (10 : 90)	5	Not observed	11 and 28
POC: 6FT-RCC3	3.57–13.78	396	Physisorption	Reversible	−43.03	Not compared	50	Loss in crystallinity	12
HOF: HOF-NKU-1	3.02–6.74	200	Physisorption	Reversible	−50	7331 (10 : 90)	3	Not observed	40

---

The aptitude of these MOCs as effective catalysts for SO₂ transformation, endorsed by the considerable improvement in the MOCs’ ability to be recycled, their resistance to degradation, which is crucial for long-term applications, their selectivity for adsorbing SO₂ in gas mixtures and simple synthesis methods in water, paves the way for exciting future research. Exploring and analysing MOCs will undoubtedly continue to reveal their fascinating nature and diverse potential applications in environmental remediation and catalysis.

SO₂-POCs

As previously discussed, discrete crystalline porous materials offer distinct advantages over extended systems, with their processability in solution being a primary benefit. Porous organic cages (POCs) are constructed through the covalent bonding of organic building blocks, which endows these materials with remarkable chemical and thermal stability.^31,32

Similar to metal–organic cages, POCs exhibit internal cavities and windows that allow access to guest molecules. They also exhibit solubility in a wide range of common organic solvents. This inherent processability facilitates the development of POC-based interphases, such as thin films or composites, which might not be achievable with other types of porous materials.³² Consequently, their most conceivable applications lie in gas separation and storage.

Due to their distinctive advantages, POCs have been studied as adsorbents, particularly for gases and pollutants. Among these, the imine known as CC3 stands out. CC3 is a tetrahedral cage synthesized from 1,3,5-triformylbenzene and (R,R)-1,2-cyclohexanediamine. Its derivatives include the reduced secondary amine version (RCC3) and a more rigid tertiary amine functionalization (6FT-RCC3). They are recognised as flexible amine cages with morphological stability under both acidic and basic conditions, and for maintaining porosity in the solid state.³³

These POCs have demonstrated to retain porosity even after adsorbing gas molecules,³¹ and remarkably, even when incorporated into thin films and membranes. Therefore, impressive sorption and permeability capacities have been reported for various gases, such as CO₂, CH₄, N₂, and H₂, across different exposure interfaces.

However, experimental studies for the corrosive gas SO₂ with POCs had been limited. With research primarily confined to in silico modeling of the CC3 cage. Through canonical Monte Carlo simulations, employing the vdW3 force field with double hybrid functional and dispersion correction (B2PLYP-D3), researchers were able to simulate the adsorbate–adsorbent interactions with SO₂ gas molecules.³⁴ In 2021 the first experimental study specifically focused on SO₂ adsorption by these covalent cages was reported.¹² In that work, CC3, RCC3, and 6FT-RCC3 were selected as SO₂ adsorbents due to the presence of nitrogen in their functional groups, which confers high affinity for interaction with this gas.

The reversibility of the adsorption process was also a key focus, and to complement the experimental findings, computational calculations were performed using density functional theory (DFT) methods, incorporating appropriate approximations, corrections, and parameters. These calculations were performed with Gaussian 16 software and aimed to provide a deeper understanding of the host–guest dynamics. This comprehensive approach allowed for a direct comparison between the experimental results, the computational calculations, and the prior simulations.

The experimental results from the three discrete crystalline molecules—CC3, RCC3, and 6FT-RCC3—tested for SO₂ adsorption revealed varying degrees of capture capacity and reversibility. At 1 bar and 298 K, the amine CC3 demonstrated a modest and reversible SO₂ capture of 2.78 mmol g⁻¹. This figure closely aligned with the previously calculated adsorption capacity of 3.6 mmol g⁻¹ determined by the B2PLYP-D3 method.³⁴ In contrast, the secondary amine cage RCC3 achieved a higher capture of 12.34 mmol g⁻¹. However, this process proved to be irreversible, characterised by open-loop hysteresis (Fig. 6).

Fig. 6 Three types of amine cages CC3, RCC3, and 6FT-RCC3 (a) interaction with SO₂ and (b) their corresponding adsorption–desorption isotherms at 298 K and 1 bar.¹²

The most promising results were observed with the tertiary amine cage, 6FT-RCC3. This POC achieved the highest amount of adsorbed SO₂ at 13.78 mmol g⁻¹ and exhibited excellent reversibility for 50 adsorption–desorption cycles. Despite its relatively small surface area of 396 m² g⁻¹, this material demonstrated impressive capture capability, coupled with high sensitivity to trace amounts of SO₂ at low partial pressures (0.1 bar). 6FT-RCC3 is positioning as a strong competitor to higher surface area materials, such as some MOFs,¹⁶ for practical SO₂ capture applications.

Both 6FT-RCC3 and RCC3 exhibited a moderate degree of similar hysteresis, suggesting a chemisorption process. This chemisorption is driven by a strong bond formed between the R₂N–H amino functional group on the cages and the SO₂ molecules.³⁵ The key difference, however, lies in their regeneration capacities. Vacuum reactivation of 6FT-RCC3 at 80 °C completely restores its original SO₂ adsorption capacity, which boosts its recyclability.

Another phenomenon shared by both RCC3 and 6FT-RCC3 after SO₂ exposure is a significant loss of crystallinity. This is attributed to the inherent flexibility of the material combined with the strength of the newly formed N–SO₂ bond.¹² To further understand these interactions, theoretical DFT calculations were performed to determine the host–guest binding energies between the POCs and SO₂. A clear trend of increasing experimental adsorption heats was observed from CC3 (49.7 kJ mol⁻¹) to 6FT-RCC3 (68.6 kJ mol⁻¹) and RCC3: 86.4 kJ mol⁻¹. This demonstrates that the amine groups provide strong binding sites for SO₂ molecules.

The loss of structural crystallinity in porous organic cages is not necessarily a drawback. In fact, amorphous POCs can become even more porous than their crystalline counterparts, being able to reach 898 m² g⁻¹ (Fig. 7).³² This increased porosity, combined with their high affinity for SO₂, allows the guest molecules to optimize their packing within the porous structure of 6FT-RCC3. A crucial feature of these discrete covalent materials is that, unlike metal–organic frameworks and covalent–organic frameworks, POCs do not lose porosity even when they become amorphous.³³ This unique characteristic, coupled with their excellent processability, synthetic scalability, recyclability, and high sensitivity to SO₂, positions POCs with N-rich functional groups, such as 6FT-RCC3, as promising adsorbents for both SO₂ capture and sensing.

Fig. 7 Structural simulation of (a) crystalline and (b) amorphous CC3 packing, with N₂ molecules (blue) adsorbed. (c) SEM image of CC3 crystals. (d) N₂ adsorption–desorption isotherm of CC3 cage (red) and other amorphous POC, ASPOC (blue).³²

The distinct adsorption behaviors and affinities observed among different POCs make them suitable for a wide array of applications. These discrete covalent materials exemplify the vast potential of supramolecular chemistry and highlight the numerous unexplored opportunities this field offers for developing novel compounds and devices directly applicable to solving real-world problems (Table 2).

Table 2 Timeline showing milestones in supramolecular chemistry-based materials for the SO₂ adsorption study in recent years

UNAM-1	MOF-177	Mg-gallate	CC3 Family	PI-COF-m10	SJTU-220	SonoCOF-9
First hydrogen-bonded material demonstrates reversible SO2 adsorption	Initial attempt to adsorb SO2 without MOF deterioration	Selectivity record (325) in SO2/CO2 desulfurization processes by MOFs.	First SO2 adsorption by POCs’ experimental study.	First attempt for cyclic adsorption by COFs, with loss of crystallinity	SO2 uptake capacity record by MOFs (29.6 mmol g^−1)	Extremely sensitive COFs withstand cyclical use without deterioration
2019	2020	2021	2022	2023	2024	2025
COF-DC-8	HOF CB6	MOC Cage 1	Mg2(dobpdc)		MOC Cage 1	HOF-NKU-1
First COF with SO2 ppb sensitivity	Reusable HOF particles for separating SO2 from binary gas mixtures	First study exploring the interaction of SO2 with MOCs	Successful SO2 separation from gas mixtures using MOFs		Refinement of MOCs as an adsorbent and a catalyst for SO2 using different counterions	Ultra-selective (7331) SO2 separation from gas mixtures by HOFs
(2014) CC3 Family In silico simulation of SO2 interaction with POCs

---

SO₂-HOFs

The hydrogen bonding interactions that form hydrogen-bonded organic frameworks (HOFs) are weaker than typical coordination or covalent bonds. While HOFs can be constructed from a wide variety of light-element molecules, diaminotriazine (DAT) and carboxylic acid (–COOH) remain the most reported building blocks.³⁶ The network structure is primarily formed by the N–H bonds and O–H⋯O interactions from these functional groups. The assembly of small organic molecules through these directional hydrogen bonds results in highly crystalline 2D and 3D supramolecular networks. Even more importantly, this bonding provides a high degree of flexibility to the porous framework, allowing it to deform and regenerate through simple steps.

To maintain their properties and find practical applications, research has focused on standardizing the design of HOFs to preserve both their shape and crystallinity. Solvated crystals typically crystallize slowly, forming HOFs. They are also highly sensitive to changes in synthesis parameters like precursor concentration, reaction time, or temperature. However, several studies have proposed decreasing the crystallization rate or increasing the thermal energy as potential solutions for obtaining a more thermodynamically stable phase.³⁶

The reversibility of the bonds in HOFs facilitates the regeneration of the system after surface saturation. Recrystallization restores the original porosity and crystallinity of the material.³⁷ While HOFs have been widely used for gas separation and adsorption, specifically for hydrocarbons, the capture and separation of toxic gases like SO₂ remains a comparatively under-researched topic. Pioneering work in this field aims to address this gap by studying and reporting significant advancements in using HOFs as an adsorbent and cyclical material for SO₂ capture.

In 2019, UNAM-1³⁸ was introduced, which became the cornerstone of this experimental research. UNAM-1 incorporates metal complexes ([Cu^IICu^I₁₂(HL)₁₂]¹⁰⁻) as structural elements, which form a perfect octahedral arrangement with [Me₂NH₂]⁺ (DMAH) cations that interact via hydrogen bonds. Most notably, it was the first hydrogen-bonded material to demonstrate reversible SO₂ adsorption. The material, when activated under mild conditions, achieved a uptake of 3.5 mmol g⁻¹ at 1 bar. The calculated heat of adsorption for SO₂ was −17.52 kJ mol⁻¹. This weak interaction results in complete desorption without hysteresis. This work enriched design strategies for multifunctional HOF materials and paved the way for the development of new HOFs specifically designed for this application.

A year later, it was reported that cucurbit[6]uril (CB6)³⁹ could be used to separate SO₂ from CO₂. CB6 is formed from formaldehyde and glycoluril under acidic conditions. Its main characteristic is a hydrophobic cavity with two symmetrical hydrophilic windows of a smaller diameter. This structure makes it easier for polar molecules like SO₂ to access the pores, especially with short diffusion paths, i.e., using small particles is better for maximum accessibility.

Nanometric CB6 particles were reported to achieve a capture of 4.98 mmol g⁻¹ at 1 bar after the first exposure to dry SO₂. They exhibited a selectivity of 120 for SO₂ in mixtures with CO₂ (10 : 90 molar ratio), which was explained by the difference in adsorption enthalpy for both gases: a calculated −48 kJ mol⁻¹ for SO₂ and −35 kJ mol⁻¹ for CO₂. The particles also proved to be highly reusable, maintaining their selectivity and capture capacity for up to 10 adsorption cycles. This reusability makes CB6 a promising material for efficiently separating SO₂ from binary gas mixtures.

In recent years, advancements in HOFs have focused on improving chemical stability even under harsh conditions. For example, HOF-NKU-1,⁴⁰ a material presented in 2025, was activated using a supercritical method, which resulted in a unique and valuable property: dynamic tautomerism. This HOF-NKU-1 interconverts between different structural forms in response to external stimuli; the material can remain deformed after a stimulus and then revert to its original shape. This property is unprecedented for traditional rigid or flexible frameworks.

In the case of HOF-NKU-1, the stimulus is the exposure to SO₂ itself. As with previous examples, the calculated adsorption enthalpy indicates a much higher-energy interaction (>10 kJ mol⁻¹ difference) with SO₂ molecules than with other gases. The strong interactions result in the deformation of the structure, causing it to display a dominant preference for capturing SO₂ over any other gas (N₂, CO₂, and O₂). A selectivity of 7331 was reported for binary SO₂/CO₂ (10 : 90 molar ratio), one of the highest for porous adsorbents. In addition to reaching 6.74 mmol g⁻¹ of SO₂ capture at 1 bar. Most impressive is that using a mixture of SO₂/CO₂/O₂/N₂ (0.3/15/3.5/81.2, v/v/v/v) with 3000 ppm of SO₂ at a flow rate of 20 ml min⁻¹, selective SO₂ capture was achieved, and its desorption could even be programmed. These advances will mark a turning point in gas separation and storage.

Contrast with other material classes

Surface modification is one of the most widely used techniques for achieving significant and selective capture.⁴¹ In the case of activated carbon, modifying the surface with amine groups has been shown to promote the chemisorption of SO₂ in a directly proportional way to the amount of nitrogen added. Activated carbon derived from cellulose-rich sources and doped with NH₃ groups exhibits the highest saturation capacity, at 6.25 mmol g⁻¹.^42,43 This material is also processable; for instance, membranes made from fibers of the same activated carbon type have been reported to continuously remove about 65% of the SO₂ from a SO₂/CO₂ gas mixture at 3000 ppm over long periods.⁴⁴

In the context of zeolites, modifying them with metal oxides has been used to improve their chemical activity for contaminant removal. Titanium dioxide (TiO₂), which is widely studied as a photoactive material, has been incorporated into zeolites as a surface dopant to facilitate the removal of SO₂ from contaminated environments. The combination of the photoactive nature of TiO₂ and the adsorptive capacity of zeolites makes heating the zeolites unnecessary to achieve removal, unlike typical methods.⁴⁵ The highest SO₂ capture capacity recorded for zeolites is 1.75 mmol g⁻¹, which was achieved by K-NaX and F700-0.8-6. Zeolites can remove around 78% of SO₂ from gas mixtures (20% SO₂, 40% CO₂, and 40% CO), through dual surface interactions. Based on Lewis base donor–acceptor interactions and hydrogen bonding, these allow for both physisorption and chemisorption, depending on the exposure conditions.⁴⁶

While porous materials like activated carbon and zeolites have long been the industry standard due to their high thermal stability, well-defined pores, and low production costs, recent research highlights their limitations. They tend to be sensitive to moisture, which can reduce both their adsorption capacity and stability. This makes them difficult to use in realistic, humid flue gas flows.^44,46

In contrast, supramolecular materials can be custom-designed by selecting different organic links and metallic nodes, allowing for the fine-tuning of both pore size and chemical function. This key property enables the pore structures—which are typically fixed in other materials—to be tailored specifically for SO₂ capture.

This design flexibility allows for the creation of materials with exceptionally high selectivity. For example, Mg-gallate,⁴⁷ which holds the record for SO₂/CO₂ selectivity (325), was designed with a large number of hydrogen bond donors lining its pore channels. Additionally, this approach provides a tailored affinity for these molecules. The design of ECUT-11⁴⁸ features two distinct cavity apertures (5.4 and 10.2 Å), which allow it to selectively surpass the SO₂ capture capacity of the most commonly reported adsorbents.

Ultimately, while zeolites and other porous materials are currently the leading commercially viable and effective materials for many applications, supramolecular materials represent the next generation of porous adsorbents. Their ability to be designed at the molecular level enables superior performance in terms of selectivity, capture capacity, and reusability even under challenging conditions.

Conclusions

This perspective synthesises recent and salient experimental contributions in the use of supramolecular chemistry-based materials towards SO₂ capture and detection. Each material exhibits distinct characteristics, which in turn lead to powerful and unique interactions with the guest molecule. The diverse nature of these interactions enriches this field of research, highlighting how each supramolecular material holds promise for tailored applications, all centered on the interaction with SO₂.

These materials share key properties like recyclability and porosity, which means their application potential is primarily determined by their specific sensitivity or high capture capacity. Their versatility is highly valued, from use in highly effective sensors that work at low or trace pressures, to gas capture and storage, and their most promising application: the separation of gas mixtures, all while maintaining robust chemical stability. In this ongoing research, discrete materials often offer superior processability, allowing for integration into various interfaces. This opens the door to the design and development of new devices, which is especially useful for flowing gases.

The combined advancements in enhanced stability, record-breaking capture capacities, and improved in situ chemical activity are collectively opening a vast new horizon for research. These promising avenues extend beyond merely refining the next generation of materials to include innovative technologies for their design and improvement.

In the near future, the synergy between emerging technologies will result in a vast list of prospective research directions for this application. For example, machine learning could be applied to guide pore engineering or to devise strategies for combining the best qualities into composite materials by new synthesis routes. Furthermore, supramolecular chemistry will be leveraged to create new techniques and devices for a deeper understanding of the adsorption mechanisms of SO₂ and other substances. This field holds immense potential for future growth.

Author contributions

All authors contributed to writing the original draft, reviewing and editing the manuscript.

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 analysed as part of this review.

Acknowledgements

This work was supported by PAPIIT-UNAM (Grant IIN201123), Mexico. A. H. M. acknowledges SECIHTI for the MSc fellowship (1336791). V. M. T. acknowledges the financial support from SECIHTI-Mexico through the program “Convocatoria Ciencia Básica y de Frontera 2023–2024,” grant number CBF2023-2024-2725. P. M. R., thank SECIHTI for the PhD fellowship (1277642).

Notes and references

1. Z. Meng, Y. Liu and D. Wu, Inhalation Toxicol., 2005, 17, 303–307.
2. K. T. Kuwata, E. J. Guinn, M. R. Hermes and A. B. Fernandez, J. Phys. Chem. A, 2015, 119, 10316–10335.
3. A. L. Reno, E. G. Brooks and B. T. Ameredes, Environ. Health Insights, 2015, 9s1, EHI.S15671.
4. World Health Organization, WHO guidelines for indoor air quality: selected pollutants. WHO Regional Office for Europe, Copenhagen, 2010.
5. M. Matooane and R. Diab, Arch. Environ. Health, 2003, 58, 763–770.
6. Z. Meng, Y. Liu and D. Wu, Inhalation Toxicol., 2005, 17, 303–307.
7. J. Saravia, G. I. Lee, S. Lomnicki, B. Dellinger and S. A. Cormier, J. Biochem. Mol. Toxicol., 2013, 27, 56–68.
8. E. Martínez-Ahumada, M. L. Díaz-Ramírez, H. A. Lara-García, D. R. Williams, V. Martis, V. Jancik, E. Lima and I. A. Ibarra, J. Mater. Chem. A, 2020, 8, 11515.
9. K. Biradha, A. Ramanan and J. J. Vittal, Cryst. Growth Des., 2009, 9, 2969–2970.
10. W. Zhao, J. L. Obeso and V. B. López-Cervantes, Angew. Chem. Int. Ed., 2025, 64, 1–7.
11. E. G. Percástegui, E. Sánchez-González, S. de J. Valencia-Loza, S. Cruz-Nava, V. Jancik and D. Martínez-Otero, Angew. Chem. Int. Ed., 2024, e202421169.
12. E. Martínez-Ahumada, D. He, V. Berryman, A. López-Olvera, M. Hernandez, V. Jancik, V. Martis, M. A. Vera, E. Lima, D. J. Parker, A. I. Cooper, I. A. Ibarra and M. Liu, Angew. Chem. Int. Ed., 2021, 60, 17556–17563.
13. F. Huang and E. V. Anslyn, Chem. Rev., 2015, 115(15), 6999–7000.
14. A. Lund, G. V. Manohara, A. Y. Song, K. M. Jablonka, C. P. Ireland, L. A. Cheah, B. Smit, S. Garcia and J. A. Reimer, Chem. Mater., 2022, 34, 3893–3901.
15. K. Tan, P. Canepa, Q. Gong, J. Liu, D. H. Johnson, A. Dyevoich, P. K. Thallapally, T. Thonhauser, J. Li and Y. J. Chabal, Chem. Mater., 2013, 25, 4653–4662.
16. N. K. Gupta, A. López-Olvera, E. González-Zamora, E. Martínez-Ahumada and I. A. Ibarra, ChemPlusChem, 2022, 87(e202200006), 1–9.
17. P. Brandt, A. Nuhnen, M. Lange, J. Möllmer, O. Weingart and C. Janiak, ACS Appl. Mater. Interfaces, 2019, 11, 17350–17358.
18. N. Tannert, Y. Sun, E. Hastürk, S. Nießing, C. Janiak and Z. Anorg, Allg. Chem., 2021, 647, 1124–1130.
19. J. Li, G. L. Smith, Y. Chen, Y. Ma, M. Kippax-Jones, M. Fan, W. Lu, M. D. Frogley, G. Cinque, S. J. Day, S. P. Thompson, Y. Cheng, L. L. Daemen, A. J. Ramirez-Cuesta, M. Schröder and S. Yang, Angew. Chem., Int. Ed., 2022, 61, e202207259.
20. E. Martínez-Ahumada, D. Kim, M. Wahiduzzaman, P. Carmona-Monroy, A. López-Olvera, D. R. Williams, V. Martis, H. A. Lara-García, S. López-Morales, D. Solis-Ibarra, G. Maurin, I. A. Ibarra and C. S. Hong, J. Mater. Chem. A, 2022, 10, 18636.
21. W. Gong, Y. Xie, A. Yamano, S. Ito, X. Tang, E. W. Reinheimer, C. D. Malliakas, J. Dong, Y. Cui and O. K. Farha, J. Am. Chem. Soc., 2023, 145(49), 26890–26899.
22. W. Gong, P. Gao, Y. Gao, Y. Xie, J. Zhang, X. Tang, K. Wang, X. Wang, X. Han, Z. Chen, J. Dong and Y. Cui, J. Am. Chem. Soc., 2024, 146(46), 31807–31815.
23. S. Ge, K. Wei, W. Peng, R. Huang and E. Akinlabi, Chem. Soc. Rev., 2024, 53, 11259.
24. G.-Y. Lee, J. Lee, H. T. Vo, S. Kim, H. Lee and T. Park, Sci. Rep., 2017, 7.
25. Z. Meng, R. M. Stolz and K. A. Mirica, J. Am. Chem. Soc., 2019, 141, 11929–11937.
26. J. Wang, J. Wang, W. Zhuang, X. Shi and X. Du, J. Chem., 2018, 1–8.
27. 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, e202506838, 1–9.
28. 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, 18658–18665.
29. Z. Yang, G. Liu, Y. D. Yuan, S. B. Peh, Y. Ying, W. Fan, X. Yu, H. Yang, Z. Wu and D. Zhao, J. Membr. Sci., 2021, 636, 119564.
30. Z. Y. Zeng, Z. C. Lou, L. Hu, W. Dou, X. Zhao, X. Li, J. Fang, X. Qian, H. B. Yang and L. Xu, Chem. Eng. J., 2024, 496, 154091.
31. T. Tozawa, J. T. Jones, S. I. Swamy, S. Jiang, D. J. Adams, S. Shakespeare, R. Clowes, D. Bradshaw, T. Hasell, S. Y. Chong, C. Tang, S. Thompson, J. Parker, A. Trewin, J. Bacsa, A. M. Slawin, A. Steiner and A. I. Cooper, Nat. Mater., 2009, 8, 973–978.
32. Q. Song, S. Jiang, T. Hasell, M. Liu, S. Sun, A. K. Cheetham, E. Sivaniah and A. I. Cooper, Adv. Mater., 2016, 28, 2629–2637.
33. M. Liu, M. A. Little, K. E. Jelfs, J. T. A. Jones, M. Schmidtmann, S. Y. Chong, T. Hasell and A. I. Cooper, J. Am. Chem. Soc., 2014, 136, 7583–7586.
34. W. Li and J. Zhang, J. Comput. Chem., 2014, 35, 174–180.
35. Z. Xu, Y. Ye, Y. Liu, H. Liu and S. Jiang, Chem. Commun., 2024, 60, 2261–2282.
36. L. Chen, B. Zhang, L. Chen, H. Liu, Y. Hua and S. Qiao, Mater. Adv., 2022, 3, 3680.
37. B. Han, H. Wang, C. Wang, H. Wu, W. Zhou, B. Chen and J. Jiang, J. Am. Chem. Soc., 2019, 141, 8737–8740.
38. R. Dominguez-Gonzalez, I. Rojas-Leon, E. Martinez Ahumada, D. Martinez-Otero, H. A. Lara-Garcia, J. Balmaseda-Era, I. A. Ibarra, E. G. Percastegui and V. Jancik, J. Mater. Chem. A, 2019, 7, 26812–26817.
39. J. Liang, S. Xing, P. Brandt, A. Nuhnen, C. Schlüsener, Y. Sun and C. Janiak, J. Mater. Chem. A, 2020, 8, 19799–19804.
40. L. Li, X. Zhang, X. Lian, L. Zhang, Z. Zhang, X. Liu, T. He, B. Li, B. Chen and X. H. Bu, Nat. Chem., 2025, 17, 727–733.
41. A. A. Abdulrasheed, A. A. Jalil, S. Triwahyono, M. A. A. Zaini, Y. Gambo and M. Ibrahim, Renewable Sustainable Energy Rev., 2018, 94, 1067–1085.
42. J. P. Boudou, Carbon, 2003, 41, 1955–1963.
43. J. P. Boudou, M. Chehimi, E. Broniek, T. Siemieniewska and J. Bimer, Carbon, 2003, 41, 1999–2007.
44. V. Gaur, R. Asthana and N. Verma, Carbon, 2006, 44, 46–60.
45. J. Yu, J. Kiwi, T. Wang, C. Pulgarin and S. Rtimi, J. Photochem. Photobiol., A, 2019, 375, 270–279.
46. D. Tefera-Zewdie, Y. Desta-Bizualem and A. Gashu-Nurie, Discover Appl. Sci., 2024, 6, 331.
47. F. Chen, D. Lai, L. Guo, J. Wang, P. Zhang, K. Wu, Z. Zhang, Q. Yang, Y. Yang, B. Chen, Q. Ren and Z. Bao, J. Am. Chem. Soc., 2021, 143, 9040–9047.
48. M. J. Yin, X. H. Xiong, X. F. Feng, W. Y. Xu, R. Krishna and F. Luo, Inorg. Chem., 2021, 60, 3447–3451.

---