Stability and Performance of BTC-based MOFs for Environmental Applications

Stability and performance of BTC-based MOFs for environmental applications Corina Rodríguez-Esteban a , Regla Ayala a , b , ** , Carlos L´ opez-Cartes a , * a Departamento de Química Inorg´ anica, Facultad de Química, Universidad de Sevilla, Aptdo 1203, 41071, Sevilla, Spain b Instituto de Ciencia de Materiales de Sevilla, Universidad de Sevilla-CSIC, Avda. Am´ erico Vespucio 49, 41092 Sevilla, Spain ARTICLE INFO Keywords: HKUST-1 MIL-100(Fe) Stability and performance Dye adsorption Iodine uptake Moisture adsorption ABSTRACT Two series of open metal site MOFs, HKUST-1 and MIL-100(Fe), have been successfully prepared using different methods of synthesis. Their features depend on the synthetic route as well as their role play in different environmental applications. The stability and performance of these BTC-based MOFs have been tested bearing in mind Congo Red (CR) removal, humidity adsorption and iodine capture and release. HKUST-1 and MIL-100(Fe) samples could offer a remarkable role in the adsorption of CR from aqueous solutions. However, the lability of HKUST-1 in water is revealed as a drawback for its reutilization in both static and agitation conditions. The former contrasts to the stability under ambient moisture. MIL-100(Fe) shows promising properties in both CR adsorption in aqueous solutions and humidity adsorption. Nonetheless, the performance largely depends on the synthesis conditions. Although CR removal is based on surface interaction, there is a relation between the adsorpted quantity and the specific surface area. The size and nature of iodine allows the diffusion in the pores of both HKUST-1 and MIL-100(Fe) MOFs. This way, the uptake of iodine is driving by the porosity and surface area of samples rather than their inherent nature. As a rule, the results of this work indicate that not only is it important the specific nature of the MOF chosen for a given application but also the way in which it has been synthesized and the conditions in which they are used. MIL-100(Fe)-R is revealed as the best suitable candidate to be used as a sorbent for CR in aqueous solutions, moisture and I 2 gas. 1. Introduction Metal Organic Frameworks (MOFs) constitute a relatively new class of crystalline hybrid materials formed by metal species (atoms, ions or clusters) linked by multidentated organic ligands [1,2]. The large versatility of these materials is due to the possibility of selecting all available metals under variable oxidation states, as well as the chance of using diverse organic functional groups as ligands. The former allows tuning both chemical composition and crystalline structure in order to design an almost infinite number of members in MOF collection[3–5]. Moreover, postsynthetic modifications of pristine MOFs extent the possibility of tailoring new systems [6,7]. Their exceptional microporosity and large specific surface areas make them excellent candidates for different technological applications. MOFs have been employed in gas storage and separation [8], sensors [9], catalysis [10–14] and drug delivery [15] among other uses. As porous materials, they have also been studied as alternative systems for multiple adsorption applications in gas and liquid media with the aim to look for improved performances substituting classical adsorbents. Growing population and its associated human activities in terms of energy and resources consumptions, as well as industrial development have dramatically increased pollution [16,17]. One of the main consequences of this environmental degradation is the contamination of water resources. Attending to their chemical composition, common contaminants in water can be divided into inorganic and organic species. Organic pollutants can be diversified in detergents, oils, pesticides, insecticides, pharmaceuticals and dyes among others. Dyes are commonly used in different fabrication processes such as printing, leather, plastic, paper and textile industries. More than one thousand tons of dyes are discharged into water systems every year [18,19]. The vast majority of dyes are an important threat not only for ecosystems but also for human health, presenting carcinogenic and mutagenic characters. Among the available techniques for organic species removal from wastewater, adsorption based technologies are the most competitive methods [20, * Corresponding author. ** Corresponding author. Departamento de Química Inorg´ anica, Facultad de Química, Universidad de Sevilla, Aptdo 1203, 41071, Sevilla, Spain. E-mail addresses: [email protected] (R. Ayala), [email protected] (C. L´ opez-Cartes). Contents lists available at ScienceDirect Journal of Solid State Chemistry journal homepage: www.elsevier.com/locate/jssc https://doi.org/10.1016/j.jssc.2024.124956 Received 11 June 2024; Received in revised form 31 July 2024; Accepted 9 August 2024 Journal of Solid State Chemistry 339 (2024) 124956 Available online 10 August 2024 0022-4596/© 2024 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC license ( http://creativecommons.org/licenses/bync/4.0/ ). 21]. MOFs have been presented as a very promising alternative for conventional used adsorbents (activated carbons and zeolites) overcoming the drawbacks of these traditional systems [21–30]. Under the increasing energy demand and considering the greenhouse gas emissions generated by carbon-based sources, the use of nuclear power plants is still considered clean and economically competitive. One of the main problems associated to the operation of these facilities concerns the safety under an accident and the management of used nuclear fuels. The risk of radionuclide emissions in the eventual released gas stream has to be taken into consideration. 129 I and 131 I, which are volatile radionuclides with high concentration in nuclear fuel waste, have serious effect on human health and must be effectively captured from gas release. Porous sorbents, traditionally being activated carbons and Ag-exchanged zeolites, are involved in the main technological processes employed in iodine removal from gas phase [31,32]. MOFs, as a new family of porous materials, have also been probed as good sorbent solids with good recovering and reusing performances [33], overcoming the regeneration limitations of carbons and zeolites [21,34–36]. The adsorption studies of humidity from ambient air using porous materials are also of great interest when capture and release of water are needed. Such processes are crucial for energy-efficient harvesting of water from air in arid earth areas. For that purpose, some MOFs-based adsorbent devices have been recently developed [37–39]. MOFs designed for these applications must capture water at very low relative humidity and must exhibit large uptake capacities. Their stability under water capture and release cycling conditions, usually heating assisted, have to be also taken into account [40,41]. In this context, two series of well-known MOFs (HKUST-1 [42] and MIL-100(Fe) [43]) have been obtained from different synthetic procedures. Both MOFs are based on 1,3,5-benzenetricarboxylic acid (H 3 BTC) and have the so-called coordinative unsaturated positions. According to the hard-soft acid-base (HSAB) principle, metal–ligand (M−L) strengths and stabilities partly depend on the softness or hardness of the components’Lewis acids and bases. In this sense, hard acids like Fe 3+ will form thermodinamically stable bonds with hard bases like BTC ligands whereas the interaction of the former with Cu 2+ would not be so strong and the competition with other harder acids could result in instability. Nevertheless, chemical hardness or softness cannot be considered the only factor to the strength of acid-base interaction, other contributions such as steric effects, medium effects, kinetic control, may also play an important role [44]. Although HKUST-1 is easier to synthesize, MIL-100(Fe) presents higher chemical and thermal stabilities and larger pore volumes. The former can be very profitable in the diffusion process needed for the interaction of guest molecules to exposed centres. At this point it is worth mentioning that for a given application not only is it necessary the presence of unsaturated positions and high pore volumes but also their accessibility that can be ruled out by the left-over solvent molecules from synthesis procedure. This means that although the inherent features of a given MOF are crucial from an Fig. 1. XRD of HKUST-1 (a), and MIL-100(Fe) (b) samples. Fig. 2. FTIR spectra of HKUST-1 and MIL-100(Fe) samples. C. Rodríguez-Esteban et al. Journal of Solid State Chemistry 339 (2024) 124956 2 application point of view, it is also significant to take into account how it has been synthesized [45]. Thus, the preparation route for MOF materials is a key factor for obtaining reproducible structures, topologies and properties in a systematic pattern [46]. Additionally, the post-synthetic activation can have as important implications as the synthetic method itself. It must be also considered the stability of these species in several media and conditions as well as the consequences of their interaction with different solutes. The aim of this work is the comparative study of the sorption capabilities and stabilities of different HKUST-1 and MIL-100(Fe) samples against Congo Red (CR) removal from water solutions, humidity capture from ambient air and iodine uptake and release from gas phase. The former implies a comparison among the capabilities and stabilities of the same MOFs, differently synthesized, and different MOFs, BTC-based, respect to the three applications previously commented. These applications account for different adsorbates species in different media. Moreover, the implications of the presence of water in either liquid or gas phase have also been considered. This study will allow not only the selection of the most suitable type of MOF for a given application but also the synthesis and treatment conditions that must be considered to optimize the results. 2. Experimental section 2.1. Materials All the reactants were used as received without further purification. Specific details about reactants and solvents are provided in the supplementary material. 2.2. Methods Powder X-ray diffraction (PXRD) data were collected on a Panalytical X’PERT PRO diffractometer with Cu-K α at room temperature using a 2θrange 5◦-60◦using a step of 0.05◦and an acquisition time of 200 s per point. Scanning electron microscopy (SEM) images were recorded in a Hitachi 54800 SEM-FE6 electron microscope at an accelerating voltage of 5 kV. Samples were covered with a thin film of chromium before analysis. Fourier transformed infrared (FTIR) analyses were carried out making KBr round pellets from mixtures containing a small amount of the sample and using a Thermo Scientific NICOLET iS5 iD1 Transmission spectrophotometer. Thermogravimetric analyses (TGA) were performed under air atmosphere and at heating rate of 10 ◦C min −1 from room temperature to 500 ◦C, using a Q600 equipment from TA Instruments. Nitrogen adsorption analyses were carried out at liquid nitrogen temperature using a Gemini V-2365//V1.00 Micromeritics equipment. The samples were activated under nitrogen flow at 200 ◦C for 1 h before the analysis. Specific surface areas were calculated from Fig. 3. SEM images of HKUST-1 samples. HKUST-1-S (a,b), HKUST-1-R (c,d), HKUST-1-RT (e,f). C. Rodríguez-Esteban et al. Journal of Solid State Chemistry 339 (2024) 124956 3 the data of adsorption isotherms using BET method. UV–Vis spectra were recorded on a Thermo Scientific Evolution 60S spectrometer. 2.3. Synthesis of MOFs HKUST-1 was synthesized via solvothermal [47], reflux [48] and ageing at room temperature methods. Specific details about the Fig. 4. SEM images of MIL-100(Fe) samples. MIL-100(Fe)–S (a,b), MIL-100(Fe)-R (c,d), MIL-100(Fe) -RT (e,f). Fig. 5. TGA analysis of HKUST-1 and MIL-100(Fe) samples. Fig. 6. Adsorption isotherms of HKUST-1 and MIL-100(Fe) samples. C. Rodríguez-Esteban et al. Journal of Solid State Chemistry 339 (2024) 124956 4 synthesis are collected in the supplementary material. MIL-100(Fe) was prepared using solvothermal [49], reflux [50] methods as reported in bibliography and ageing at room temperature following the procedure in the literature [51] with slight modifications. All details are provided in supplementary material. 2.4. Congo red removal Dye removal experiments were carried out under both static and stirring conditions at room temperature. In general, CR water solutions with concentrations of 50 mg/L were used. 50 mg of the sample were immersed in 50 mL of the organic dye solution and maintained under static or soft magnetically stirring conditions for 120 min. After certain times 3 mL aliquots were taken out and the solid was separated by centrifugation. Two droplets of NaOH aqueous solutions 0.1 M were added to maintain the initial pH of the solutions. The concentration of CR in the liquid phase was determined through UV–Vis spectra measuring the absorbance at λ=497.5 nm and comparing these values with calibration representation. The quantity of adsorbed dye, q t (mg CR / g MOF ) is directly determined through gt=(C0-Ct)⋅V/m, being C0 and Ct the dye concentrations at the beginning and at time t, V the volume of solution employed for the experiment and mthe mass of MOF used as adsorbent. After the experiments the solids were recovered by centrifugation, washed with EtOH and dried for further cycles and characterization. The isotherm CR adsorption study of MIL-100(Fe) samples was conducted with initial concentrations ranging from 50 to 450 mg/L and contact time of 24 h what ensures that the equilibrium state has been reached. In order to analyse the control mechanism of the adsorption process pseudo-first-order and pseudo-second-order kinetic models expressed as Eqs. (1) and (2) were employed to test the experimental data of adsorption kinetics. ln(q e -q t )=lnq e -K 1 ⋅t (1) t/q t =1/(K 2⋅ q e 2 )+t/q e (2) Table 1 BET Surface areas obtained for HKUST-1 and MIL-100(Fe) samples. Sample Surface Area (m 2 /g) Sample Surface Area (m 2 /g) HKUST-1-S 1228 MIL-100(Fe)–S 920 HKUST-1-R 890 MIL-100(Fe)-R 1344 HKUST-1-RT 832 MIL-100(Fe)-RT 1291 Fig. 7. Congo Red adsorption capacities for HKUST-1 (a) and MIL-100(Fe) (b) samples under static and agitation conditions. Fig. 8. XRD of HKUST-1 (a) and MIL-100(Fe) (b) samples after CR adsorption experiments under agitation conditions. C. Rodríguez-Esteban et al. Journal of Solid State Chemistry 339 (2024) 124956 5 where, q t and q e (mg⋅g −1 ) are respectively the masses of CR adsorbed per unit mass of the sorbent at a certain time t and equilibrium time; K 1 (min −1 ) and K 2 (g mg −1 ⋅min −1 ) represent the pseudo-first order and the pseudo-second order adsorption rate constants, respectively. 2.5. Humidity capture from ambient air Around 50 mg of MOFs were placed in previously weighted 10 mL empty vials and heated in air at 120 ◦C for 24 h. The weights of these activated samples were determined and then exposed to air at room temperature for 48 h. During the experiments the vials were reweighed at different times and the water uptake efficiency were calculated by gravimetric measurements. Three cycles of essays were repeated. The samples were recuperated and characterized after the whole procedure. 2.6. Iodine capture from vapour phase and subsequent recovery The iodine uptake experiments were carried out following the procedure reported in the literature [34,35]. 100 mg of different samples were placed in previously weighed 10 mL empty vials and heated in air at 120 ◦C for 24 h. After that the weights of these activated samples were determined. These vials were placed joint with solid iodine in a closed 500 mL chamber and heated in oven a 75 ◦C for 24 h. The iodine capture was determined by gravimetric measurements. The procedure was repeated 3 times. The samples were washed in EtOH and centrifuged as previously mentioned to recover the pristine MOFs. To quantify the iodine recovery 10 mg of every charged sample was immersed in 10 mL of ethanol under static conditions. UV–Vis spectra were acquired at certain times between 5 and 120 min to follow the iodine releasing from solid to liquid phase. 3. Results and discussion 3.1. Synthesis and characterization Different HKUST-1 and MIL-100(Fe) samples were obtained through several synthesis procedures such as solvothermal (-S), reflux (-R) and ageing at room temperature (-RT). For the sake of comparison, Fig. 1 collects powder X-ray diffractograms (XRD) for all samples as well as those simulated from the ideal of MOF structures [42,43]. The presence of the characteristic reflections of both MOFs in the experimental records clearly demonstrates the successful synthesis for all the samples. Nevertheless, some extra reflections apart from those of the ideal structure can be observed in the diffractogram of HKUST-1-RT sample. As far as we know, this procedure has not been previously used to obtain HKUST-1. According to literature in which moderate stability of HKUST-1 against water is postulated [41], and taking into account that this sample was prepared using pure water as solvent, the crystalline structure of this secondary phase could appear as a consequence of partially degradation of MOF structure during the synthesis. In fact, this extra reflections were encountered in HKUST-1 degradation studies in water [52] and they were associated to a mixture of crystalline laminar structures of [Cu 2 (BTC)(H 2 O) 6 ] and [Cu 2 (BTC)(OH)(H 2 O)] compositions and an unsolved phase coming from hydrolysis processes undergone by the original structure. For a better understanding, the XRD of this degraded HKUST-1 phase was also included in Fig. 1. Fig. 2 shows the FTIR spectra for the HKUST-1 and MIL-100(Fe) samples. For the sake of completeness, the spectrum of the corresponding organic ligand was also included. The typical bands of the trisubstituted aromatic ring (730 and 1278 cm −1 ), the C–H and C=C stretch vibrations of the aromatic ring (1450-1400 cm −1 ), the C–H bending modes (760 and 707 cm −1 ) appear in both HKUST-1 [53] and MIL-100(Fe) [54] samples. The characteristic band for the C=C stretching of the aromatic ring is also observed at 1583 cm −1 as long Fig. 9. PH evolution for suspensions of HKUST-1 and MIL-100(Fe) samples in water under agitation conditions. Fig. 10. Reusability of MIL-100(Fe) samples in Congo Red adsorption under agitation conditions. Fig. 11. Congo Red adsorption capacity of MIL-100(Fe) vs initial dye concentration. C. Rodríguez-Esteban et al. Journal of Solid State Chemistry 339 (2024) 124956 6 with the C=O symmetric (1450 cm −1 ) and asymmetric (1374 cm −1 ) stretching. The narrowing of the wide band at 1710 cm −1 in the prepared samples suggests the lack of unreacted organic linker and, therefore, a good purification of the final products. The signal related to the Cu-OOC isobidentante interaction (1700-1300 cm −1 ) and the Cu–O stretching (491 cm −1 ) are well-recognized in HKUST-1 samples. Additionally, the Fe–O stretching (464 cm −1 ) is present for MIL-100(Fe) preparations. SEM images corresponding to HKUST-1 samples are shown in Fig. 3. As revealed by the direct inspection of these pictures, particle morphologies present a clear dependence on the synthesis procedure. Wellshaped octahedral particles are obtained when solvothermal synthesis is used while quite irregular and smaller rounded particles are observed on the images of the sample prepared by reflux method. Two types of particles are found in the images coming from the sample prepared at room temperature in water. On one hand, well-shaped particles that could be assigned to the HKUST-1 fraction of this sample, and on the other hand, needle-like particles that should correspond to the secondary crystalline phase identified by previously commented X-Ray diffraction studies. SEM images of MIL-100(Fe) samples are shown in Fig. 4. In general, the samples are more homogeneous and no differences associated to the synthesis can be identified. In all cases, no well-faceted particles with rounded shapes are observed. However, the solvothermal sample presents smaller particle sizes than the reflux and room temperature preparations. Fig. 5 shows the TGA analyses of all the samples. Both, HKUST-1 and MIL-100(Fe) samples, show TGA profiles similar to those found for these MOFs in literature [50,55]. This means a first drop at a temperature lower than 200 ◦C corresponding to the removal of physically adsorbed moieties and a second stage attributed to the collapse of the structure around 300 and 350 ◦C for HKUST-1 and MIL-100(Fe), respectively. These results indicate the higher thermal stability of MIL-100(Fe) vs. HKUST-1 materials [56]. The type of porosity and surface area are important properties for adsorption applications. Fig. 6 collects the N 2 adsorption isotherms for HKUST-1 and MIL-100(Fe) samples. The BET surface areas are displayed in Table 1. Both HKUST-1 and MIL-100(Fe) samples show type I isotherm and high surface area in accordance with literature [50,57]. The main contribution in such a high surface area is due to micropores. HKUST-1 samples present a wider range of surface areas that can be related to the different morphologies found in the SEM images. As expected HKUST-1-RT sample has the lowest porosity in agreement with the idea of partial degradation of the MOF structure during the synthesis. As a rule, the larger surface areas are shown by the MIL-100(Fe) samples. 3.2. Congo red removal The evolution of the UV–Vis spectra with time for samples containing aqueous CR solutions in contact with HKUST-1 and MIL-100(Fe) samples under agitation conditions are included in the Supplementary Material (Fig. S1). In all cases, a decrease of absorbance with time evidences the adsorption of dye by MOFs. As mentioned in the Material section, aqueous NaOH 0.1 M was added after centrifugation. Without the former addition, a shift of the maximum towards higher wavelengths (λ =570 nm) was observed as time proceeds what implies a change in the colour of the samples from red to violet (see Fig. S2 in Supplementary Material). This change is due to a variation of the solution pH. One should recall that CR is an acid-base indicator that remains blue below pH 3.0 and gradually turns to a brilliant red above pH 5.2. The acid conditions can be related to MOF hydrolysis[58–60] what can provoke the degradation of its structure. Thus, it is important to study in detail the consequences of hydrolysis to establish the convenience of using this type of materials to adsorption processes in aqueous solutions. The adsorption capacity of the samples is plot in Fig. 7. It can be seen that the amount of CR adsorbed per g of MOF is rather similar in all cases. The high adsorption rate at the beginning can be attributed to the large number of accessible active sites on the surface of MOFs. Then, as the adsorption process proceeds, the vast majority of active sites on the surface were occupied by the CR molecules, and the adsorption rate slowed down gradually until an equilibrium adsorption state was achieved. Additionally, the vast majority of the CR in solution was transferred to the MOFs after only 120 min. The FT_IR spectra recorded after the experiment confirm the adsorption by both HKUST-1 and MIL-100 (Fe) samples (Fig. S3 in the supplementary material). The shifting of the peaks in the range 1582-1562 and 1391-1368 cm −1 indicates the π - π Fig. 12. The pseudo-first-order plot of Congo Red adsorption for MIL-100(Fe) and HKUST-1 samples under static and agitation conditions. C. Rodríguez-Esteban et al. Journal of Solid State Chemistry 339 (2024) 124956 7 interaction between the benzene rings of the MOF framework and the aromatic backbone of the organic dye. Although there is a relation between the specific surface areas and the adsorption, the size of CR molecule (around 21 Å) makes difficult the diffusion along the porous. Thus, it can be assumed that the adsorption takes place in the surface of the MOFs. This way the samples with larger particle sizes as estimated by SEM provide better adsorption capabilities. Moreover, it is worth mentioning the appearance of a precipitate in the solution in acid conditions related to the dye. This could be interpreted as CR is extracted from the aqueous solutions not only adsorbed on the MOF but also as an independent solid. At this point, it is necessary to study the chemical stability of the samples after CR adsorption to evaluate their reusability. To asses this proposal, Fig. 8 shows the XRD of HKUST-1 and MIL-100(Fe) after the adsorption in comparison with the simulated diagram for each MOF and the degraded HKUST-1 phase [52]. The results indicate that HKUST-1 is spoilt and new peaks assigned to laminar structures of [Cu 2 (BTC) (H 2 O) 6 ] and [Cu 2 (BTC)(OH)(H 2 O)] compositions appear in the diagram whereas no obvious changes in structure of the MIL-100(Fe) samples were appreciated. The change of colour of the aqueous solutions of CR in contact with MOFs and the degradation of HKUST-1 motivate the study of aqueous suspensions of MOFs. Fig. 9 displays the evolution of water pH in contact with HKUST-1 and MIL-100(Fe). In all cases, there is a decrease of the water pH but the acidity of the media is MOF dependant. This way, HKUST-1 samples diminish water pH in more than 2 units, being more pronounced in the case of HKUST-1 –R and -RT. For MIL-100(Fe), the decrease of the water pH is almost immediately and keeps constant along the whole experiment. The acidity of the samples justifies the change of colours observed in the CR adsorption. MOFs were also characterized after the study of pH evolution (Fig. S4 in supplementary material). MIL-100(Fe) samples keep their identity while HKUST-1 samples are degraded in the same way as in the presence of CR. Hence, it can be concluded that, the loss of HKUST-1 structure is not responsibility of the dye but it is due to the hydrolysis processes motivated by the MOF presence in aqueous medium. It is also worth estimating if the agitation has any implication in the degradation of HKUST- 1, the CR removal were repeated in static conditions (Fig. 7). As expected, the absorption of CR in static conditions is smaller due to a lesser contact between MOF and dye. The XRDs of the HKUST-1 samples after CR adsorption in static conditions (Fig. S5 in supplementary material) show differences from the as-prepared samples, namely, the appearance of new peaks that remind those found in the assays under agitation. These results suggest that although HKUST-1 is as good adsorbent for CR in aqueous solutions as MIL-100(Fe), its integrity is not maintained due to hydrolysis events no matter the conditions stablished in the experiment. MIL-100(Fe) samples were recycled and reutilised in several adsorption cycles. The results shown in Fig. 10 indicate, on one hand, that rather similar behaviour appears no matter the sample considered. On the other hand, there is no a significant performance decay along the cycles. In addition, Fig. S6 illustrate that there is no change in the structure of the MOF due to the recycling and reutilization. Thereby these results confirm the good response of repeatability and good longterm stability of MIL-100(Fe) samples respect to CR adsorption in aqueous solutions. To evaluate in more detail, the sensitivity of the different MIL-100 (Fe) samples towards dye elimination, the effect of the initial concentration across the range of 50–450 mg/L was studied. Fig. 11 depicts the amount of adsorbed CR per g of MOF v.s. the CR initial concentration in solution. Up to 100 mg/L the adsorption capacity of MIL-100(Fe) increased with the initial concentration for all the samples. At higher concentrations, the trend depends on the method of synthesis. MIL-100 (Fe)-R follows a non-stop increasing behaviour until a lowering is observed. This tendency can be explained in terms of exchange ligands of the MOF with dissolved species in the medium [61]. Hence, CR may replace BTC ligands due to the amino sulphate groups present in the dye [62]. MIL-100(Fe)–S and MIL-100(Fe)-RT present a lower performance for the CR elimination. Nonetheless, according to the former reasoning their structural integrity seems to be maintained at high concentrations. These results point out that the performance of MIL-100(Fe) is rather dependant on how it has been synthesized. A conclusion based only on the type of MOF should be misleading. Recently, kinetic studies have been widely performed to the adsorption mechanism of dyes from aqueous solutions onto adsorbents Table 2 Parameters of the kinetic models used to fit the adsorption of CR onto MIL-100 (Fe) and HKUST-1 samples. Pseudo first-order kinetic q exp (mg/ g) intercept q cal (mg/g) slope k 1 (min −1 ) R 2 MIL-100 (Fe)– S_Agit. 38.6 4.443 85.0 −0.0422 0.0422 0.5720 MIL-100 (Fe)- R_Agit. 46.7 4.357 78.0 −0.0460 0.0460 0.7666 MIL-100 (Fe)- RT_Agit. 40.3 2.194 9.0 −0.0179 0.0179 0.2461 MIL-100 (Fe)– S_Stat. 15.4 1.570 4.8 −0.0167 0.0167 0.5305 MIL-100 (Fe)- R_Stat. 10.8 3.268 26.3 −0.0453 0.0453 0.7671 MIL-100 (Fe)- RT_Stat. 29.9 4.773 118.3 −0.0640 0.0640 0.5367 HKUST-1- S-Agit. 47.0 4.742 114.7 −0.0560 0.0560 0.6679 HKUST-1- R_Agit. 46.9 0.404 1.5 −0.0034 0.0034 −0.1086 HKUST-1- RT_Agit. 47.0 3.067 21.5 −0.0334 0.0334 0.2826 HKUST-1- S_Stat. 21.0 3.225 25.2 −0.0460 0.0460 0.6596 HKUST-1- R_Stat. 28.1 4.066 58.3 −0.0558 0.0558 0.8216 HKUST-1- RT_Stat. 25.3 3.917 50.2 −0.0442 0.0442 0.6281 Pseudo second-order kinetic intercept k 2 (g mg −1 min −1 ) slope q cal (mg/ g) R 2 MIL-100(Fe)– S_Agit. 1.042 0.0004 0.0207 48.40 0.8280 MIL-100(Fe)- R_Agit. 0.448 0.0007 0.0183 54.73 0.9831 MIL-100(Fe)- RT_Agit. 0.073 0.0085 0.0249 40.16 0.9930 MIL-100(Fe)– S_Stat. 0.320 0.0137 0.0661 15.13 0.9879 MIL-100(Fe)- R_Stat. 10.567 0.0000 −0.0144 −69.25 −0.1346 MIL-100(Fe)- RT_Stat. 1.293 0.0005 0.0258 38.83 0.8872 HKUST-1-S- Agit. 0.527 0.0006 0.0172 58.00 0.9626 HKUST-1- R_Agit. −0.020 −0.0255 0.0229 43.73 0.9974 HKUST-1- RT_Agit. 0.164 0.0027 0.0209 47.89 0.9724 HKUST-1- S_Stat. 0.498 0.0040 0.0445 22.47 0.9944 HKUST-1- R_Stat. 0.740 0.0012 0.0292 34.20 0.9744 HKUST-1- RT_Stat. 1.615 0.0005 0.0271 36.95 0.8011 C. Rodríguez-Esteban et al. Journal of Solid State Chemistry 339 (2024) 124956 8 [63]. In order to obtain information about the adsorption kinetic, the experimental data were fitted using the analysis model detailed in the Material Section. The results are collected in Fig. 12 and Table 2. The correlation coefficients (R 2 ) of pseudo first-order kinetic model was in the range [0.24–0.77] what indicates that it is not appropriate to predict the adsorption of CR on MIL-100(Fe) or HKUST-1. On the other hand, as shown in Fig. 13 the pseudo-second-order model was fitted well to the kinetic data. Additionally, the calculated value of the q e parameter from the pseudo-second-order kinetic model exhibited great accordance with the experimental values. The results of kinetic analysis showed that the adsorption kinetics conformed to the pseudo-second order kinetic model. K 2 constants collected in Table 2 show a variety of values that not only depend on the type of MOF but also on the preparation method. Again, it is highlighted that the performance of a given MOF does not only relies on its inherent nature. 3.3. Humidity adsorption The results of moisture adsorption are collected in Fig. 14. MIL-100 (Fe) samples present a similar adsorption capacity whereas the behaviour of HKUST_1 samples is more unalike. On one hand, the interaction with the water molecules in both cases is mainly through the open metal centre saturating the coordination sphere of the metal centre. In general, the affinity of MIL-100(Fe) for water molecules is higher and can be explained in terms of larger bond enthalpy (−75 kJ/mol vs. −47.3 kJ/ mol for MIL-100(Fe) and HKUST-1, respectively) [64,65]. On the other hand, water molecules also interact with ligands and with other water molecules [66]. To do so, they need to diffuse through the structure, so that pore sizes and specific surface areas are features that play a significant role in moisture adsorption. In all cases, the adsorption process is reversible in the sense that the samples recover their initial weight after heating at 120 ◦C for 24 h. Two additional cycles were performed in order to evaluate the recycling and reuse aspects of the process. The analysis of Fig. 14 b) and c) shows that MIL-100(Fe) samples rather maintain the adsorption properties whilst the performance of HKUST-1 materials decreases in each cycle. The characterization of the samples after moisture adsorption indicates that there is not any change in the diffractograms of all the samples respect to those of the pristine MOF (Fig. S7). In order to analyse in a greater extent the stability of HKUST-1 respect to moisture adsorption, the HKUST-1 samples were exposed to ambient conditions for 7 days. XRDs patterns were recorded at different time intervals and collected in Fig. S8 of the supplementary material. The analysis indicates that the structure of HKUST-1 remains after 7 days of exposure without the appearance of new phases due to the hydrolysis of the MOF. Nevertheless, it is clear that the diminishing in the adsorption could be interpreted as a first step in the degradation of these MOFs, especially in the case of HKUST-1 –R. 3.4. Iodine capture Table 3 collects the results of the adsorption capacity of vapour Iodine. The samples show a great affinity to the Iodine with an adsorption larger than 100 wt% in 24 h. This uptake is an order of magnitude larger than traditional zeolite sorption materials [34]. The size of the I 2 molecules (2.6 Å) allows the diffusion through the pores of both MOFs. In this case, the adsorption capacity depends on the surface areas being higher for HKUST-1-S and MIL-100(Fe) samples. When the adsorption is considered in terms of mole of I 2 per mole of MOF (see Fig. S9 in the supplementary material), the behaviour of MIL-100(Fe) is in agreement with the presence of larger pores in this MOF. The reutilization of the samples after their washing and drying gives rise to the maintenance of the level of adsorption. The PXRDs of samples after three cycles of iodine adsorption experiments do not show any evidence of structural changes (see Fig. S10 in the supplementary material). The positions of the main reflections from MIL-100(Fe) and HKUST-1 structures remain unchanged in the diagrams. It is well-known the stability of these MOFs in ethanol media [27,52,67]. However, it is worth noting that in the HKUST-1 diagrams, the intensity of the reflections is highly modified compared to those in the pristine samples (see Fig. 1). This may be related to the plausible ordered iodine adsorption inside the micropore structure, using the MOF structure as a template. Moreover, the characterization of the materials using XRD techniques indicate that the removal of I 2 does not affect MOF structures in both MIL-100(Fe) and HKUST-1 samples. The desorption of I 2 was studied in a qualitative way because the extraction with ethanol gives rise to the formation of different species Fig. 13. The pseudo-second-order plot of Congo Red adsorption for MIL-100(Fe) and HKUST-1 samples under static and agitation conditions. C. Rodríguez-Esteban et al. Journal of Solid State Chemistry 339 (2024) 124956 9