NH₂-MIL-53(Al) prepared sustainably and assisted by cationic surfactant as H₂ and CO₂ adsorbent

Abstract

The development of metal–organic frameworks (MOFs) with improved textural properties is a key strategy for enhancing their performance in adsorption. For this purpose, three NH₂-MIL-53(Al) MOFs were synthesized via room temperature and hydrothermal approaches in the presence of cetyltrimethylammonium bromide surfactant using water as the only solvent. Firstly, parent and cetyltrimethylammonium-assisted NH₂-MIL-53(Al) MOFs were synthesized at room temperature in water. In a different route, post hydrothermal treatment of cetyltrimethylammonium-assisted room-temperature-synthesized NH₂-MIL-53(Al) and conventional hydrothermal synthesis of cetyltrimethylammonium-assisted NH₂-MIL-53(Al) were performed at 100 °C in an autoclave. All the as-obtained MOFs were crystalline, except the cetyltrimethylammonium-assisted room-temperature-synthesized NH₂-MIL-53(Al), which was semi-crystalline. The washing protocol using water and ethanol was found to be successful in removing the bulk of surfactant or unreacted precursors from the final products as confirmed by thermal analysis. Room-temperature-synthesized parent MOF showed rice or worm-like shapes and surfactant-templated room-temperature-synthesized MOF displayed shape-less agglomerated morphology as revealed by transmission electron microscopy. Conventionally and post hydrothermally synthesized surfactant-templated MOFs showed well-formed morphology. Textural analysis showed that surfactant tailored the porosity of NH₂-MIL-53(Al) synthesized in water at room temperature by introducing small intercrystalline mesoporosity. The conventional hydrothermally synthesized NH₂-MIL-53(Al) exhibited low surface area compared to the other three samples. The cetyltrimethylammonium-templated room-temperature synthesis in water produced a high-surface area sample with improved hydrogen and carbon dioxide uptakes compared to the parent NH₂-MIL-53(Al).

CTA-templated room-temperature in water synthesis yielded NH₂-MIL-53(Al) with high surface area, high micropore area and small intercrystalline mesoporosity, resulting in enhanced H₂ and CO₂ uptakes compared to CTA-free synthesized NH₂-MIL-53(Al).

1. Introduction

Metal–organic frameworks (MOFs), which are constructed from metal nodes and multitopic organic ligands assembly through coordination chemistry, can be synthesized in multiple ways.¹ The common conventional synthesis of MOFs utilizes high temperature, pressure and corrosive organic solvents such as DMF (N,N-dimethylformamide) and DEF (N,N-diethylformamide).² This creates a problem in large-scale production from an energy demand and environmental perspective.^2,3 In addition, organic solvents pose both environmental and corrosive risks, particularly in large-scale synthesis. The above-mentioned problems have become a bottleneck for the industrial application of MOFs by increasing production costs as well as environmental concerns.^4,5

A sustainable synthesis of MOFs can minimize the costs associated with energy utilization and environmental problems by avoiding the use of high temperature, pressure and toxic solvents.⁴ Sustainable synthesis of MOFs either removes toxic solvents entirely from the methods of synthesis or reduces environmental costs by substituting hazardous organic solvents with benign alternatives such as water. Additionally, lowering the temperature used in MOFs synthesis by implementing room temperature synthesis will enhance economic viability and reduces environmental impact.^5–7 The eco-friendly synthesis of MOFs, particularly those prepared in water solvent at room temperature, often gives rise to agglomerated nanocrystalline MOFs with poor structural ordering.⁶ This contrasts with highly crystalline MOFs synthesized by means of conventional solvothermal-based protocols.⁵ As a consequence, conventionally-synthesized MOFs usually have better textural properties than those synthesized at room temperature.⁸ However, certain sustainable synthesis strategies lead to high quality MOFs, sometimes rich in defects, having outstanding performance in very demanded applications such as catalysis and gas sorption.^6,9,10 Since MOFs' textural properties have direct effect on their gas sorption abilities,¹¹ improving the textural properties of sustainably prepared and scalable MOFs is a key step towards future large-scale applications of MOFs for gas storage/capture/separation.¹²

Generally, MOFs' high surface area and porosity, reversible gas sorption and tunable nature make them strong candidates for material-based H₂ storage as a clean alternative energy vector and for CO₂ capture to mitigate climate change.¹² However, their implementation in industry still needs improvement for production at scale. There are different ways to increase gas sorption capacity of MOFs through synthetic modifications,¹³ such as ligand functionalization,^14,15 metal node modification,^16,17 MOF based composite formation,¹⁸ post synthetic modification¹⁹ and in situ modification using modulators and surfactants.^20,21 The textural properties of sustainably synthesized MOFs can also be improved by using surfactants.^9,10 Surfactants can control crystal size,¹⁰ morphology²² and pore property of MOFs.⁹ Ionic surfactants are believed to create chemical interactions with the MOF precursor materials in water-based synthesis. Among the ionic surfactants, the cationic surfactant CTA (cetyltrimethylammonium) has been widely used as mesoporous templating agent in MOFs synthesis.^11,23,24

In this study, CTA was employed as structure directing agent in the synthesis of NH₂-MIL-53(Al) to induce mesoporosity within the resultant MOF. The amino-functionalized MIL-53(Al) was selected for this purpose because of its breathing behaviour and because it can be prepared under sustainable conditions. In addition, it gives rise to a nanocrystalline MOF with high surface area and thermal stability.⁶ As a follow up of our previous work,⁹ in this article, the focus is on the sustainable preparation of NH₂-MIL-53(Al) by using salt of the ligand in water at room temperature and in the presence of the surfactant (CTAB) (cetyltrimethylammonium bromide). The impact of some synthesis conditions, viz., pressure and temperature on surfactant-templated synthesis of NH₂-MIL-53(Al) were investigated. This study examined post-hydrothermal treatment of a surfactant-templated MOF synthesized at room temperature, and compared it with a conventional hydrothermal synthesis route for the same surfactant-templated MOF. In addition, the effect of cationic surfactant on the MOF properties, particularly its textural characteristics, was investigated for hydrogen and carbon dioxide sorption applications.

2. Experimental section

2.1. NH₂-MIL-53(Al) synthesized at room temperature

Synthesis of NH₂-MIL-53(Al) was done according to a published approach via a sustainable synthesis route.⁶ In a typical experiment, a solution was prepared in Beaker 1 by adding 13 mL of 1 M NaOH (sodium hydroxide) in 1.087 g (0.006 mol) of the ligand precursor NH₂-H₂BDC (2-aminoterephthalic acid, Sigma-Aldrich, 99%). NaOH was used to deprotonate the carboxylic groups of the linker to yield the water-soluble linker form 2-aminoterephthalate, [NH₂-BDC]²⁻. 12 mL of 1 M NaOH is in principle enough to neutralize both carboxylic groups of the linker; however, one extra mL was added to make sure the ligand is fully deprotonated. In Beaker 2, a metal salt solution was prepared from 1.45 g (0.006 mol) of AlCl₃·6H₂O (aluminum chloride hexahydrate, Sigma-Aldrich, 99%) and 3 mL of H₂O. Both solutions were left under stirring until the precursor was fully dissolved. Finally, the metal solution in Beaker 2 was added to the ligand solution in Beaker 1, which resulted in the rapid formation of a yellow precipitate. The reaction was stirred slowly and maintained at room temperature condition for 24 h. After 24 h, the product was transferred into washing tubes, washed thoroughly with H₂O (four times, 4×) followed by ethanol (4×), centrifuged and dried at room temperature. Sample code A0 was assigned for the synthesized parent NH₂-MIL-53(Al) MOF.

2.2. CTA templated NH₂-MIL-53(Al) synthesized at room temperature

CTAB (cetyltrimethylammonium bromide, Sigma-Aldrich) was used to investigate the surfactant templating effect in the synthesis of NH₂-MIL-53(Al) at room temperature in water. In this regard, 1.087 g (0.006 mol) of NH₂-H₂BDC (2-aminoterephthalic acid, Sigma-Aldrich, 99%) was added in 13 mL of 1 M NaOH (Beaker 1). After the ligand was perfectly dissolved and formed a clear solution, 1.97 g (0.0054 mol) of CTAB (Sigma-Aldrich) was added. This amount corresponds to a CTAB/AlCl₃·6H₂O or CTAB/NH₂-H₂BDC molar ratio of 0.9. CTAB/metal precursor or CTAB/ligand precursor ratio of 0.9 was chosen for the preparation of the experiment as it has been reported to be ideal for the formation of CTA micelles.⁹ A metal salt solution was prepared with 3 mL of H₂O and 1.45 g (0.006 mol) of AlCl₃·6H₂O (Beaker 2).⁹ Instantly, a yellowish precipitate was formed when the solution in Beaker 2 was mixed to the solution in Beaker 1 under stirring at room temperature. The reaction was stirred and maintained at room temperature condition for 24 h. The product formed was then transferred into washing tubes, washed thoroughly with H₂O (4×) followed by ethanol (4×) via centrifugation and dried at room temperature. The sample synthesized with surfactant at room temperature in water was labeled ACS.

2.3. Hydrothermally synthesized CTA templated NH₂-MIL-53(Al)

Conventional hydrothermal synthesis was also carried out to examine the effect of CTA on NH₂-MIL-53(Al). Accordingly, 1.087 g (0.006 mol) of NH₂-H₂BDC (2-aminoterephthalic acid, Sigma-Aldrich, 99%) was added in 13 mL of 1 M NaOH (Beaker 1). After the ligand was fully dissolved and formed a clear solution, 1.97 g (0.0054 mol) of CTAB (Sigma-Aldrich) was added. A metal salt solution was prepared by dissolving 1.45 g (0.006 mol) of AlCl₃·6H₂O with 3 mL of H₂O (Beaker 2).⁹ Instant formation of a yellowish precipitate was observed when the solution in Beaker 2 was added to the solution in Beaker 1, under constant stirring. The formed precipitate was immediately put into a 60 mL Teflon-lined autoclave and heated at 100 °C for 24 h without prior solvent separation. After 24 h being in the autoclave, the sample was centrifugated and then washed first with H₂O (4×) and subsequently with ethanol (4×) and finally dried at room temperature. The sample prepared in this protocol was coded AC100A.

2.4. Post hydrothermal treatment of CTA templated NH₂-MIL-53(Al)

1.087 g (0.006 mol) of NH₂-H₂BDC (2-aminoterephthalic acid, Sigma-Aldrich, 99%) was added to 13 mL of 1 M NaOH (Beaker 1). After the ligand was completely dissolved and formed a clear solution, 1.97 g (0.0054 mol) of CTAB (Sigma-Aldrich) was added. 1.45 g (0.006 mol) of AlCl₃. 6H₂O and 3 mL of H₂O was used to prepare a metal salt solution (Beaker 2).⁹ Instantly, a yellowish precipitate was formed when the solution in Beaker 2 was added to the solution in Beaker 1 under constant stirring. The reaction was stirred slowly and maintained at room temperature condition for 24 h and then put into a 60 mL Teflon-lined autoclave and heated at 100 °C for 24 h without prior solvent separation. Finally, the precipitate was washed with H₂O (4×) and ethanol (4×) in a subsequent manner followed by centrifugation. Finally, the product obtained was dried at room temperature. The sample synthesized following this method was labeled AC100B. Fig. 1 summarizes the syntheses procedures discussed.

Fig. 1. Scheme for the syntheses of parent and CTA templated NH₂-MIL-53(Al) MOFs.

2.5. Characterization techniques

PXRD (powder X-ray diffraction) was performed at high angle to determine the crystalline phases found on the parent and surfactant templated samples using X’PERT diffractometer instrument equipped with X'Celerator detector (X’PERT Pro PANalytical) with Cu Kα radiation (λ = 1.5406 Å). Low-angle PXRD was employed to investigate mesoscopic order of the four samples under study. An aluminum sample holder was used for high-angle PXRD analysis and a zero-background silicon sample holder was used for low-angle PXRD analysis. Thermal degradation measurements were conducted using TGA (thermogravimetric analysis) with platinum pan sample holder from 25 to 900 °C at 20 °C min⁻¹ with PerkinElmer TGA7 instrument under air flow. The morphology and crystal size of the materials was examined by TEM (transmission electron microscopy) analysis with a JEOL 2100F electron microscope at operating voltage of 200 kV. Micrometrics ASAP 2420 was employed for the BET (Brunauer–Emmet–Teller) surface area calculation with nitrogen adsorption–desorption isotherm attained at −196 °C, with degassing temperature of 150 °C for 16 h for all samples. The PSD (pore size distribution) was determined using the BJH (Barrett–Joyner–Halenda) model on the adsorption branch of the N₂ isotherm with Kruk–Jaroniec–Sayari correction. A Micrometrics 3Flex surface analyzer was used for H₂ and CO₂ gas sorption experiments at −196 °C and 25 °C, respectively using ultra-high purity gases. Gas sorption capacities were evaluated after degassing the samples for 16 h at 150 °C.

3. Results and discussion

3.1. Powder X-ray diffraction analysis

The crystalline samples high-angle XRD patterns of parent (A0) and surfactant templated NH₂-MIL-53(Al) (AC100A and AC100B) matched the simulated MIL-53(Al) XRD pattern corresponding to CCDC (Cambridge Crystallographic Data Center) deposition number 220475 (Fig. 2).²⁵ This unactivated MIL-53(Al) XRD pattern is consistent with a narrow or intermediate pore state of the MOF, which is usually found to contain unreacted linker in the pores.^12,25

Fig. 2. High-angle (a) and low-angle (b) XRD patterns for parent and surfactant-templated synthesized NH₂-MIL-53(Al) MOFs.

The XRD pattern of the non-templated parent MOF with sample code A0, confirms nanocrystalline NH₂-MIL-53(Al) was synthesized in water as a sole solvent and at room temperature.⁶ The MOF formation under these conditions was instantaneous and substantial, as if the process was precipitation rather than crystallization, in such a way that the resultant MOF material (A0) was very nanocrystalline, as evidenced by the significantly broad XRD peaks.¹⁰ NH₂-MIL-53(Al) synthesized at room temperature in the presence of the surfactant CTA (sample ACS) became much more nanocrystalline, as its XRD peaks are much wider than those for the sample A0. The broader peaks for ACS sample indicate that the presence of CTA affected the formation of NH₂-MIL-53(Al), leading to crystalline domains that were much smaller, and hence an XRD pattern typical of a semi-crystalline material.

The high-temperature hydrothermal synthesis of the surfactant-templated samples AC100A and AC100B yielded highly crystalline materials, as evidenced by their better resolved XRD patterns with distinct diffraction peaks. Notably, AC100A exhibits sharper and higher-intensity peaks than AC100B, indicating a higher degree of structural ordering and crystallinity (Fig. 2a). The pronounced intensity of the peaks in the XRD pattern in the case of AC100A is likely attributable to its synthesis procedure, in which MOF formation occurred in the autoclave at 100 °C for 24 h. In contrast, sample AC100B was prepared at room temperature for the initial 24 h and then subjected to an additional 24 h autoclave treatment at 100 °C. This sequential approach effectively served as a post heat-and-pressure treatment, which may alter nucleation and crystal growth relative to the continuous high-temperature synthesis used for AC100A. The preferential crystal growth for the AC100A sample better matches with the simulated MIL-53(Al) in the relative intensity of the peak at 2θ = 9.0°, suggesting that the MOF formation under high temperature and pressure also affected the crystal morphology. This demonstrates that the presence or absence of heat during the reaction can affect crystal growth rates and produce crystal phases, which alter relative peak intensities as shown for non-templated room-temperature synthesized A0 sample and conventionally synthesized AC100A sample.^25,26 AC100B sample showed a good crystalline XRD pattern compared to the room-temperature-synthesized parent (A0) and the one assisted by CTA at room temperature (ACS sample).

Low-angle XRD patterns of parent and surfactant-assisted-synthesized NH₂-MIL-53(Al) MOFs are exhibited in Fig. 2b. Low-angle XRD peaks are an indication of an ordered mesoporous structure in a material.²⁷ There is a low-angle XRD peak for samples A0, ACS and AC100B but this peak is absent for the AC100A sample. This low-angle peak appears as a broad feature, which likely indicates the presence of a mesoporous structure within these samples.²⁸ However, it is not ordered or homogenous, rather it is a single broad peak which indicates a heterogenous mesoporous structure.²⁹ The formed mesoporosity is intercrystalline, which ensued because of the nanoscale size of the MOFs as can be deduced from the broader high-angle XRD peaks for the samples A0, ACS and AC100B.⁶ Room temperature condition for MOF synthesis by precipitation techniques frequently forms intercrystalline mesoporosity. This is a result of void formation in the crystalline agglomerate of nanocrystalline MOF during nucleation and growth of phases.^6,9,12 For the sample AC100A, the absence of the low-angle XRD peak can be well explained by the synthesis conditions used. The synthesis for this specific sample was done through a conventional hydrothermal synthesis method. The precursor materials were fed directly into an autoclave after mixing, and placed in an oven at 100 °C for MOF formation, unlike the other three samples that formed at room temperature. The high-angle XRD pattern for the AC100A sample in Fig. 2a also showed relatively high intensity peaks which indicate large crystals compared to the other three samples. This can be the reason why A0, ACS and AC100B likely possessed voids of mesoporous sizes as revealed by the presence of the low-angle peaks, but not sample AC100A (Fig. 2b).⁶ The post temperature and autogenous pressure employed in the synthesis of the sample AC100B assisted to form a more crystalline MOF with larger intercrystalline spaces. This difference in intercrystalline mesoporosity is further supported by means of TEM and N₂ adsorption/desorption isotherms, as explained below.

3.2. Thermogravimetry analysis

Fig. 3a presents the TGA graphs of the four samples, and the corresponding derivative plots are provided in Fig. 3b. The weight losses can be sectioned and described accordingly. The weight loss ascribed to the solvent, which is water in these cases, appears below 150 °C for all synthesized samples. The second weight loss observed between 250 and 350 °C for A0, AC100A and AC100B samples is due to entrapped or unreacted linker within the pores of the MOFs, which is well known in NH₂-MIL-53(Al)-type MOFs.^6,9,12,25 This second weight loss is not detected in the ACS sample (Fig. 3a), which is corroborated by the derivative graph (Fig. 3b). Indeed, the first weight loss assigned to water is much more significant in the sample ACS, as more of the template and linker is removed from the pores, which is consistent with the greater amount of water uptake. This indicates the washing protocol removed most of the unreacted or entrapped linker or CTA from the ACS sample. The semi-crystalline nature of NH₂-MIL-53(Al) in the case of the sample ACS (Fig. 2a) could be responsible for the enhanced removal of unreacted linker molecules, probably because its really short pore length would allow water/ethanol used for washing to efficiently remove much of the unreacted linker. The absence of sharp weight losses around 250 and 350 °C for A0, AC100A and AC100B samples indicates that the washing protocol was also effective in removing most of the CTA. The last weight loss for all samples depicted after 400 °C represents the degradation of the MOF structure and the formation of inorganic residues. It is remarkable that the amount of the residual Al₂O₃ resulting at the end of heating in the TGA experiment is practically equal for all samples. However, the amount of linker is lower for the sample ACS, the semi-crystalline NH₂-MIL-53(Al), suggesting that the external surface of the MOF crystal is made of Al rather than linker (Fig. 3a).

Fig. 3. TGA graphs of parent and surfactant-assisted-synthesized NH₂-MIL-53(Al) MOFs (a) and the respective derivative plots (b).

3.3. Transmission electron microscopy

Fig. 4 displays TEM images of the synthesized samples at different magnifications. TEM images of parent A0 show agglomerated particles of NH₂-MIL-53(Al) with worm/rice like structures. For ACS, imaging revealed a greater extent of particle agglomeration, with no discernible, well-defined morphology under the same magnification conditions as the parent MOF. This ACS sample, which is semi-crystalline as aforementioned, is made up of nanocrystals that are very small and homogeneous in size. The sample A0 is composed of nanocrystals of larger size, and the nanocrystals of the sample AC100B are even larger, in good agreement with the observation from XRD patterns (Fig. 2a), presumably because the treatment at 100 °C induces crystal growth. Obviously, the larger crystals would generate higher void volume within the agglomeration. The TEM images of conventionally synthesized sample, AC100A, reveal a well-formed crystal shape, in agreement with the high crystallinity depicted by the high-intensity peaks in the XRD patterns (Fig. 2a).

Fig. 4. TEM images of parent and surfactant-assisted-synthesized NH₂-MIL-53(Al) MOFs.

3.4. N₂ adsorption–desorption isotherms

N₂ adsorption–desorption isotherms of all as-synthesized samples are shown in Fig. 5a, and the pore size distributions are presented in Fig. 5b. From Fig. 5a, the isotherm type for the samples A0, ACS and AC100B could be classified as a mixture of Type I (indicative of microporosity)¹² and Type IV(a) (indicative of mesoporosity) according to IUPAC classification.³⁰ Type IV(a) isotherm is characterized by the presence of mesopores greater than 4 nm which leads to the observed type of hysteresis loop.^31,32 The pore diameter in Fig. 5b for each sample is greater than 4 nm, confirming the presence of large mesopores in the three samples with the highest pore diameter observed for AC100B sample. Mesopore size and geometry depend on the relative pressure values at which capillary condensation occurs.^33,34 The hysteresis loop for ACS sample starts at a lower relative pressure compared to the other two samples A0 and AC100B. Accordingly, the hysteresis loop ends at relative pressure of 0.71 for ACS compared to 0.85 for A0 and 0.91 for AC100B. This shift of the hysteresis loop is indicative of small mesopore size and volume for ACS when compared to A0 and AC100B, which is in good agreement with the values presented in Table 1.³⁵ A flatter plateau observed for ACS sample is a confirmation of the successful mesopore filling.³² This indicates the enhanced removal of unreacted precursors or CTA from the formed mesopores as also evident from TGA in Fig. 3a and b, which might contribute to the higher surface area of the sample as depicted in Table 1. From Table 1, more microporosity was also accessed by nitrogen gas during the measurement for the ACS sample, in spite of their semi-crystalline nature. Surely, the greater removal of protonated linker within the pores accounts for the highest microporosity observed for sample ACS among the three analyzed samples.

Fig. 5. N₂ adsorption–desorption isotherms at −196 °C (a) and pore size distribution (b) of parent NH₂-MIL-53(Al) MOF and CTA-assisted-synthesized MOFs. Full circles represent adsorption data and open circles represent desorption.

Table 1. Surface area and pore property values of parent and CTA-templated-synthesized MOFs obtained from N₂ adsorption/desorption isotherm.

Sample	S BET a (m2 g−1)	S microp b (m2 g−1)	V p c (cm3 g−1)	D p d (nm)
A0	850	407	0.82	7.7
ACS	870	466	0.54	5
AC100B	480	182	0.64	11.5
AC100A	270	126	0.53	—

^a S _BET (total surface area) is the sum of external surface.

^b S _microp (microporous area) obtained from t-plot data.

^c V _p (pore volume) at p/p₀ = 0.98.

^d D _p (average pore diameter) is calculated from PSD which is measured by BJH method on the adsorption branch of the N₂ isotherm with Kruk–Jaroniec–Sayari correction.

The obtained pore diameter values are from 5 to 11.5 nm as shown in Fig. 5b and Table 1, which lies in the mesoporous range. It has already been established that the very small crystallite and crystal size of the sample ACS (Fig. 2 and 4), leads to very uniform aggregates of the nanocrystals, which creates uniform intercrystalline voids associated with mesoporosity. This is also made clear by the shape of the isotherm of this sample (Fig. 5a), as well as the pore size distribution presented in Fig. 5b, which indeed shows a relatively narrow distribution centered at 5 nm for the sample ACS. It is remarkable that approximately half of the S_BET area is accounted for by the external surface area (Table 1). The pore size distribution is much larger in the sample A0 (7.7 nm) and in the sample AC100B (11.5 nm), undoubtedly due the larger crystal size of these samples (Fig. 4) which led to the formation of higher intercrystalline mesoporosity compared to ACS sample.³⁶ Using a deprotonating agent for the linker and water as sole solvent for the hydrothermal methodology assisted by CTA (sample AC100A) resulted in poor porosity (Fig. 5a). There is no well-defined peak observed in the pore size distribution plot, which indicates no ordered mesoporosity (Fig. 5b). This was expected as above discussed using XRD results (Fig. 2), as the high intensity of the XRD peaks (Fig. 2a) of the sample AC100A indicates crystals of notable size, whose agglomeration, if any, does not produce ordered mesoporosity (Fig. 5b).

3.5. Hydrogen and carbon dioxide storage

The gas sorption study was performed for A0 and ACS samples which have shown the highest total (BET) surface area and micropore surface area among the synthesized samples. Gas sorption in MOFs is generally governed by physisorption and it is primarily related to surface area and particularly micropore surface area at low pressures. Gas sorption measurements for AC100A and AC100B samples were not performed due to relatively lower BET surface area and micropore surface area in comparison to A0 and ACS samples. In addition, the observed intercrystalline mesoporosity of samples A0, ACS and AC100B (Fig. 4 and 5b) will have a minimal contribution to H₂ and CO₂ adsorption at low pressures. The pore size for AC100B is even much larger than for ACS and A0 samples as shown in Table 1. The large mesopore in AC100B is unlikely to contribute to gas uptake at low pressures.¹² Accordingly, Fig. 6a presents an investigation of hydrogen uptake for room-temperature water-synthesized parent NH₂-MIL-53(Al) (A0) and CTA-assisted NH₂-MIL-53(Al) (ACS). From the graph, it is observed that the templated sample ACS stored more H₂ than the non-templated parent A0 sample. The H₂ uptake of ACS was 0.57 wt% which is higher than the uptake reached by A0 (0.49 wt%) sample at −196 °C and 1 bar pressure. Fig. 6b depicts the CO₂ sorption of the non-templated and CTA-assisted NH₂-MIL-53(Al). Similar to H₂ uptake, the CO₂ sorption of the surfactant-templated-synthesized ACS sample was higher than that of the non-templated parent MOF, A0. At 25 °C and 1 bar, ACS exhibited a CO₂ uptake of 5.76 wt%, whereas A0 showed a lower uptake of 4.42 wt%. It is evident that gas sorption of the MOFs is directly affected by the surface area and porosity. The ACS sample has a higher BET surface area and micropore surface area, and gas molecules are favorably adsorbed in micropores at low pressures, which results in high H₂ and CO₂ uptakes for ACS sample. The gas uptake for ACS sample might be further improved if measurements are performed at high pressures, due to the presence of small intercrystalline mesopores that can be accessed during high-pressure gas sorption.¹² The overall gas sorption experiment proved that the surfactant-assisted room temperature MOF synthesis improved the gas uptake of NH₂-MIL-53(Al) MOF. The accessible microporosity of templated room-temperature-synthesized NH₂-MIL-53(Al) enhanced the gas uptake of the sample,³⁷ resulting in its better performance compared to the parent MOF, A0. The gas adsorption was completely reversible in both cases, consistent with a physisorption process.

Fig. 6. Hydrogen (a) and carbon dioxide (b) uptakes of parent NH₂-MIL-53(Al) (A0) and CTA-assisted-synthesized NH₂-MIL-53(Al) (ACS). Closed circles are adsorption and open circles are desorption.

The H₂ uptake of ACS sample was comparable with a recently published report at 1 bar and −196 °C for a sample which was synthesized by a conventional method.³⁸ The CO₂ uptake was also equivalent to reported data obtained, under similar adsorption conditions, for conventionally synthesized samples.^39–41 However, in these reports, the samples were synthesized using solvothermal conditions (i.e. use of DMF solvent at high temperatures), as opposed to room temperature and water employed to synthesize ACS sample in this work. Additionally, for those reported samples, evacuation of unreacted linker from pores was performed through either high temperature calcination or washing with corrosive organic solvent for a prolonged time. One of the advantages of preparing the NH₂-MIL-53(Al) materials at room temperature is that the pH is so high that the linker is mostly deprotonated. Consequently, the amount of protonated linker found in the pores of NH₂-MIL-53(Al) is much lower than in NH₂-MIL-53(Al) prepared at high temperatures as was discussed in the weight loss analysis. It is very important to highlight that the material preparation in these experiments is not only sustainable in its synthesis but also in its activation. A calcination step which is done at 275 °C is avoided, a treatment that is used to remove unreacted linker in the pores of MIL-53(Al) type MOFs. A calcination treatment cannot be safely used for NH₂-MIL-53(Al) because the amino-functionalized MOF is less stable than MIL-53(Al).^9,25 Overall, the synthesis method employed in this study offers the advantage of sustainability and produces MOFs with appreciable gas uptake.

4. Conclusions

The surfactant cetyltrimethylammonium (CTA)-assisted sustainable synthesis (in water as solvent) was conducted for the breathing MOF, NH₂-MIL-53(Al). Four MOFs were synthesized at room temperature and hydrothermally. These are parent NH₂-MIL-53(Al), CTA-templated room-temperature-synthesized MOF, hydrothermally treated CTA-templated room-temperature synthesized MOF and conventional hydrothermally synthesized CTA-templated MOF. Crystalline MOFs were formed in all cases except CTA-templated MOF synthesized at room temperature. The semi-crystalline CTA-templated room-temperature-synthesized MOF transformed to crystalline MOF when treated hydrothermally after MOF formation at room temperature. Comparatively, the most highly crystalline MOF obtained was the CTA-templated MOF, which was synthesized through conventional hydrothermal synthesis. Thermal degradation of the synthesized MOFs showed weight losses for the removal of water, unreacted linker or CTA and degradation of the MOF structure. The CTA or unreacted linker weight loss was absent for the CTA-templated-synthesized MOF at room temperature, which indicates the greater extent of removal of the unreacted linker or CTA from the structure. The same sample showed more textural porosity and hence higher total surface area compared to all synthesized MOFs. This is due to the accessible micro and mesoporosity of the sample due to the increased removal of CTA or unreacted linker as confirmed from weight loss analysis. A decrease in the intercrystalline mesoporosity was observed for the same sample compared to the parent MOF, both of which were synthesized in a similar way. This indicates that the intercrystalline mesoporosity for CTA-templated sample comes from the CTA micelles after much of the surfactant was removed by washing processes. The intercrystalline mesoporosity increased when the CTA-assisted room-temperature synthesized NH₂-MIL-53(Al) was treated hydrothermally at 100 °C after the MOF formation. The CTA-assisted conventional synthesis of NH₂-MIL-53(Al) exhibited poor textural porosity due to the loss of the intercrystalline mesoporosity during high temperature MOF formation. The gas uptakes were enhanced for CTA-assisted room-temperature-formed NH₂-MIL-53(Al) compared to the non-templated parent MOF. The parent NH₂-MIL-53(Al) stored 0.49 wt% H₂ at 1 bar and −196 °C and 4.42 wt% CO₂ at 25 °C and 1 bar, whereas the surfactant-templated room-temperature-synthesized sample stored 0.57 wt% H₂ and 5.76 wt% CO₂ under the same conditions.

Author contributions

Nejat Redwan Habib: conceptualization, methodology, formal analysis, visualization, investigation, validation, and writing – original draft. Manuel Sanchez-Sanchez: resources, and writing – review & editing. Isabel Diaz: resources, and writing – review & editing. Abi M. Taddesse: resources, and writing – review & editing. Henrietta W. Langmi: conceptualization, methodology, resources, funding acquisition, project administration, supervision, and writing – review & editing.

Conflicts of interest

There are no conflicts to declare.

Data availability

The experimental data that supports the findings of this study will be available on request.