Juan L.
Obeso‡
ab,
Valeria B.
López-Cervantes‡
a,
Catalina V.
Flores‡
ab,
Ana
Martínez
c,
Yoarhy A.
Amador-Sánchez
b,
N. S.
Portillo-Velez
d,
Hugo A.
Lara-García
e,
Carolina
Leyva
a,
Diego
Solis-Ibarra
*b and
Ricardo A.
Peralta
*d
aInstituto Politécnico Nacional, CICATA U. Legaria, Laboratorio Nacional de Ciencia, Tecnología y Gestión Integrada del Agua (LNAgua), Legaria 694, Irrigación, 11500, Miguel Hidalgo, CDMX, Mexico
bLaboratorio de Fisicoquímica y Reactividad de Superficies (LaFReS), Instituto de Investigaciones en Materiales, Universidad Nacional Autónoma de México, Circuito Exterior s/n, CU, Coyoacán, 04510, Ciudad de México, Mexico. E-mail: diego.solis@unam.mx
cDepartamento de Materiales de baja Dimensionalidad. Instituto de Investigaciones en Materiales, Universidad Nacional Autónoma de México. Circuito Interior SN, Ciudad Universitaria, CP 04510, Coyoacán, CDMX, Mexico
dDepartamento de Química, División de Ciencias Básicas e Ingeniería. Universidad Autónoma Metropolitana (UAM-I), 09340, Mexico. E-mail: rperalta@izt.uam.mx
eInstituto de Física, Universidad Nacional Autónoma de México, Apartado Postal 20-364, Mexico City 0100, Mexico
First published on 6th February 2024
The Al(III)-based MOF CYCU-3 exhibits a relevant SO2 adsorption performance with a total uptake of 11.03 mmol g−1 at 1 bar and 298 K. CYCU-3 displays high chemical stability towards dry and wet SO2 exposure. DRIFTS experiments and computational calculations demonstrated that hydrogen bonding between SO2 molecules and bridging Al(III)–OH groups are the preferential adsorption sites. In addition, photoluminescence experiments demonstrated the relevance of CYCU-3 for application in SO2 detection with good selectivity for SO2 over CO2 and H2O. The change in fluorescence performance demonstrates a clear turn-on effect after SO2 interaction. Finally, the suppression of ligand–metal energy transfer along with the enhancement of ligand-centered π* → π electronic transition was proposed as a plausible fluorescence mechanism.
Different strategies have been developed to tackle anthropogenic emissions, and flue-gas desulfurization (FGD) technology is one of the most commonly employed technologies for SO2 removal.9 However, the FGD process generates a large amount of waste and leads to corrosion in pipelines.10 Therefore, the design and implementation of effective absorbent materials are fundamental to implementing new alternatives for SO2 removal. Metal–organic frameworks (MOFs) have emerged as promising candidates for the efficient adsorption of different gases.11–13
MOFs are crystalline hybrid materials formed by organic linkers and metal ions.14 Selected MOFs have shown high chemical and thermal stability, displaying outstanding surface areas, chemical mutability and tunable pore sizes.15 Based on their remarkable properties, MOFs have been widely applied in different research fields, particularly catalysis,16 adsorption of pollutants,17 detection,18 and drug delivery.19 In the case of SO2 removal, MOFs that have been successfully implemented feature high chemical stability toward SO2 and ideally participate in SO2 physisorption processes,20 since, in some cases, metal–oxygen bonds can dissociate after SO2 adsorption. For example, although MOF-177 holds the SO2 uptake record (25.7 mmol g−1 at 298 K and 1 bar), after SO2 adsorption, the crystal structure collapses.21 Therefore, designing chemically stable metal–organic frameworks is fundamental to achieving effective SO2 sequestration. Recently, Mg2(dobpdc) and MIL-101(Cr)-4F(1%) were reported to exhibit outstanding SO2 uptake (19.5 and 18.4 mmol g−1 at 298 K and up to 1 bar, respectively) and stability, even under humid conditions.22,23 The coordination environment of the metal centers (building units) allowed them to interact with SO2 molecules without breaking the fundamental coordination bonds.24 Additionally, some Al(III)-based MOFs have been effective as SO2 adsorbents, showing high chemical stability, low toxicity and sustainability.25
Another relevant property that has recently started to be successfully explored in MOFs is the detection of SO2.26 Specifically, select MOF materials can change their light-emission properties in response to SO2 molecules, the concentration of which can thus be evaluated quantitatively or qualitatively.27 Particularly, the fluorescence phenomenon is a suitable technique for this purpose, either by a turn-on or turn-off mechanism, although the turn-on effect has been reported to be the most accurate.28 In other words, a MOF material can exhibit a response after a stimulus (interaction with a specific molecule), and the resulting change in luminescence can be measured. In addition, this behaviour is observed in MOFs in association with different analytes. For example, Zn2(TCPE) displays a shift in its florescence spectra (related to a turn-on effect) for NH3.29 With regard to SO2, a Tb(III)-based MOF showed a characteristic change in the fluorescence of Tb3+ after interaction with SO2.30 Also, MOF-5-NH2 exhibited a luminescence turn-on phenomenon when exposed to SO2.31 Therefore, implementing certain MOF materials for SO2 adsorption and detection is a promising avenue in the field of environmental remediation and workplace hazard reduction.
Herein, we report an Al(III)-based MOF named CYCU-3 [Al(OH)(SDC)] (H2SDC = 4,4′-stilbenedicarboxylic acid) (Fig. 1), which displays octahedral Al(III) centres with hydroxo bridging groups (μ2-OH), showing approximately 3 nm wide hexagonal pores.32 CYCU-3 shows promise for SO2 adsorption (even under humid conditions) and detection due to outstanding chemical stability.
![]() | ||
Fig. 1 Structure of CYCU-3 (a) along the channel c-axis and (b) metal cluster and linker arrangement along the b-axis. Atom labels: purple: AlO4(μ-OH)2 octahedra, grey: carbon, and red: oxygen. |
![]() | ||
Fig. 2 SO2 adsorption–desorption isotherm at 298 K and 1 bar for CYCU-3 (solid circles represent adsorption and open circles show desorption). |
Furthermore, the SO2 isotherm shows the absence of hysteresis due to completely reversible adsorption–desorption behaviour, suggesting a relatively weak SO2 interaction. The stability of CYCU-3 after the adsorption process was confirmed since no substantial changes were observed in the PXRD pattern (Fig. S6†). The BET surface area remains unchanged after the SO2 adsorption experiments (Fig. S7†). Also, to further evaluate the structural stability of CYCU-3, an activated sample of CYCU-3 was exposed to humid SO2 using an in-house designed setup (Fig. S10†). The resulting sample, exposed to humid SO2, exhibited a PXRD pattern comparable to that of the pristine material (Fig. S6†).
In order to investigate the host–guest interactions between SO2 and CYCU-3, the heat of adsorption was determined, and DRIFTS experiments were carried out in conjunction with computational calculations. The isosteric heat of adsorption (ΔH) for SO2 adsorption was calculated at low coverage (Fig. S8 and S9†) using the virial method.42 It was estimated that CYCU-3 shows a relatively low heat of adsorption (ΔH = −29.8 kJ mol−1). In this case, the value agrees with the absence of hysteresis in the adsorption isotherm, which is attributed to the saturated Al(III) coordination sphere in CYCU-3, which provides no open metal sites for chemisorption.43 Thus, the low heat of adsorption confirms a physisorption process. The material saturation is not reached under the experimental conditions, characteristic of a mesoporous MOF material. Moreover, this performance is associated with a relatively weak hydrogen bonding interaction between SO2 and the hydroxo-functional groups,44 as it was previously reported for hydroxo-functionalised MOFs. For example, MFM-300(Sc), MFM-300(Al), and DUT-4 displayed negative heat of adsorption values of 36.2,45 30.0,36 and 31.3 kJ mol−1,46 respectively. This also suggests that SO2 directly interacts with CYCU-3 via hydrogen bonding.
To corroborate this, in situ DRIFTS experiments were conducted using CO as a probe molecule to corroborate the role of hydroxyl functional groups in the adsorption mechanism (Fig. 3). This spectroscopic technique is a powerful tool to characterise acid/base sites for different porous materials.47
A characteristic red-shift was observed at 3703 cm−1, which can be attributed to the hydroxyl stretching band (Δν(OH) = 4 cm−1) in the presence of CO due to the bridging OH moieties. In this case, the contribution of Brønsted acid sites can form weak hydrogen bonding.48 Moreover, a broad new band at 2170 cm−1 is observed due to the CO interaction with the –CC– site from the arene rings in the ligand. The π-acid character of the CO molecule is suitable since it possesses a π-acceptor character and can interact with the π-density of the ethylene site.49 In this context, the bridging Al(III)–OH groups and the –C
C– are the preferential adsorption sites interacting with the CO probe molecule. Even though the chemical properties of CO are different from those of SO2, this study proved to be useful for gaining insights into the possible adsorption sites, as previously reported.23
Computational calculations (partially optimised) were performed to gain further insights into the preferential SO2 adsorption sites. The calculations revealed that SO2 molecules interacted in five different scenarios within CYCU-3 (Fig. S11†). The interaction of SO2 through benzene and with the CH site was found to be completely unstable. On the other hand, the other three structures were more stable than the dissociated systems, and in the first scenario, the SO2 molecule can form a hydrogen bond with one H atom of the –CC– group from the ligand arene moiety, displaying an energetic value of 4.7 kcal mol−1. The S atom of SO2 interacts with the hydroxyl O atoms of CYCU-3 with 8.8 kcal mol−1. Finally, the most stable structure was calculated to be 14.4 kcal mol−1, which arises from hydrogen bond formation with the bridging Al(III)–OH groups and one oxygen atom of SO2. Thus, this corroborates the relatively strong hydrogen bonding, confirming that the bridging OH moieties are the preferential adsorption sites.
The H2SDC ligand shows strong bands at 465 and 440 nm, related to the presence of conjugation, which leads to ligand-centred π* → π electronic transitions. Moreover, the solid-state emission spectra of CYCU-3 display an intensity decrease and a slight blue shift, showing fluorescence quenching. This behaviour can be associated with the charge transfer between the H2SDC ligand and Al(III) centres.34
Thus, an activated sample of CYCU-3 was saturated with SO2 in our in-house designed ex situ adsorption system (Fig. S10†) to measure the photoluminescence properties of SO2-saturated CYCU-3. Remarkably, the photoluminescence increases considerably (2.22-fold increase) in emission intensity (Fig. 4). Additionally, the quantum yields (QY) of CYCU-3 before and after its exposure to SO2 were determined, and an increase from 18.65% to 26.81% was observed; such an increment in luminescence QY after SO2 exposure reinforces the fact that its adsorption on the framework has a positive impact on the light emission efficiency of CYCU-3. Furthermore, this behaviour can be related to the electronic effect that SO2 generates on CYCU-3. In this case, the hydrogen bonding between SO2 and bridging Al(III)–OH groups can suppress the energy transfer between the ligand and Al(III) centres, magnifying the π* → π electronic transition.50,51
![]() | ||
Fig. 4 Solid-state emission spectra of activated CYCU-3 (purple line) and after exposure to SO2 (pink line), CO2 (pink fuchsia line), and H2O (blue line). The excitation wavelength was set at 343 nm. |
Thus, to further test the selectivity of CYCU-3, the fluorescence of CYCU-3 upon exposure to H2O and CO2 was investigated (Fig. 4). Interestingly, no significant changes in the shape or intensity of the photoluminescence spectrum were produced by any of these molecules compared to the spectrum of activated CYCU-3. Since the change in fluorescence is only generated by SO2 exposure, a reproducibility test was conducted. Then, 6 independent experiments were undertaken, and the same SO2-saturated CYCU-3 sample was re-activated. In all cases, the sample was saturated with SO2 to obtain the photoluminescence absorption properties during the cycling experiment. The same emission spectrum was consistently observed, with an average 2.22 ± 0.11-fold increase in emission intensity (Fig. S13†).
Once the SO2 detection capability of CYCU-3 and the reproducibility of the detection of SO2-saturated CYCU-3 samples were demonstrated, the detection properties of this Al(III)-based material at low SO2 pressure were investigated. Since low SO2 pressure could be associated with SO2 detection at low SO2 concentrations, exploring any photoluminescence response at only 0.1 bar of SO2 is fundamental.
Then, an activated sample of CYCU-3 was placed in direct contact with 0.1 bar of SO2 (in our in-house designed ex situ adsorption system, Fig. S10†) to obtain the corresponding photoluminescence absorption spectrum (Fig. 5). Interestingly, the broad photoluminescence peaks remained centered at λmax = 450 and 427 nm, with a 1.65-fold increase in emission intensity recorded. Additionally, the reproducibility of this experiment was estimated with 6 independent experiments (re-activation of the sample and re-exposure to 0.1 bar of SO2), discovering an average 1.65 ± 0.14-fold increase in emission intensity. Importantly, these experiments only evaluate the SO2 detection by CYCU-3 at 0.1 bar, which cannot be directly interpreted as SO2 sensing. However, the consistent and reproducible response of the Al(III)-based MOF material towards low-pressure SO2 is very promising and validates ongoing efforts to develop MOF-based SO2 sensors.
![]() | ||
Fig. 5 Solid-state emission spectra of activated CYCU-3 (purple line) and after exposure to SO2 at 0.1 bar (pink line). The excitation wavelength was set at 343 nm. |
Lastly, time-resolved photoluminescence (TRPL) experiments were conducted using a 340 nm picosecond-pulsed LED as the excitation source (Fig. S14†). Thus, an activated sample and an SO2-saturated sample were analysed. TRPL decay was measured at an emission wavelength of 450 nm (Table S2†). It was observed that the decay lifetimes slightly decreased after SO2 interaction with CYCU-3. It is worth mentioning that fluorescence lifetime is independent of fluorescence brightness,52 as observed in this work. Thus, the small decay in fluorescence lifetime can be attributed to the cooperative effect of the intramolecular rotational constraint of the framework53 and the emission ligand-centered π* → π electronic transitions.54
Footnotes |
† Electronic supplementary information (ESI) available: Instrumental techniques, characterization, and experimental data. See DOI: https://doi.org/10.1039/d3dt04073a |
‡ These authors contributed equally to this work. |
This journal is © The Royal Society of Chemistry 2024 |