Green synthesis of amorphous Ce-MOFs as efficient adsorbents towards Ofloxacin antibiotics

Abstract

This study presents an efficient and green strategy for the synthesis of amorphous metal-organic frameworks (MOFs). Hence, we synthesized a series of cerium (Ce)-based amorphous MOFs at room temperature using water as the sole solvent via screening the undissolved organic ligands. XRD and ATR-FTIR analyses confirmed the highly amorphous nature and preserved chemical structure of the synthesized Ce-MOFs, while N₂ adsorption/desorption and TGA results revealed mesoporosity and good thermal stability. The synthesized small-sized amorphous Ce-MOFs particles exhibited hierarchical pores in the range of 4–15 nm, moderate surface areas (up to 280 m²/g), large pore volumes (up to 0.54 cm³/g), abundant active adsorption sites, and good structural stability, which are all beneficial for the efficient removal of the Ofloxacin (OFL) antibiotic from water. Among different synthesized amorphous Ce-MOFs, the Ce-MOF-A-2 with a surface area of 152 m²/g showed a high affinity toward OFL molecules, resulting in the highest experimental adsorption capacity of 138.98 mg/g under optimal conditions (T = 298 K, t = 4 h, pH ≈ 8). The experimental adsorption data for this adsorbent followed the Langmuir isotherm and pseudo-second-order kinetic models, resulting in a maximum calculated adsorption capacity of 140.84 mg/g, higher than its crystalline counterpart. Additionally, this amorphous adsorbent showed satisfactory regeneration and reusability performances, making it an acceptable adsorbent for water decontamination.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-42188-6.

Introduction

Metal-organic frameworks (MOFs) are crystalline and porous materials in which the clusters/metal ions (inorganic nodes) are connected via multi-functional organic molecules (organic ligands) in an infinite array^1,2. The unique combination of inorganic nodes and organic ligands imparts them with several outstanding properties, including large surface area, simple functionalization, adjustable pore sizes, and tunable chemical structures^3,4. Owing to these unique properties, they have achieved great research interests in the last decades; thereby, they can be applied in different applications such as catalysis, gas adsorption, storage, and separation, biomedical applications, sensing, energy storage, water treatment, etc.^5–7.

MOFs are regarded as an effective and highly promising adsorbent for the removal of different pollutants from water^8–11. Although the research on MOFs in the field of water treatment has significantly focused on the crystalline MOFs, amorphous MOF-based adsorbents have also gained growing attention in recent years¹². This is mainly because amorphous MOF-based adsorbent materials show a relatively good adsorption performance, possess many defect sites, and usually exhibit better mechanical resistance compared to crystalline MOF-based materials, which are all important for practical water treatment applications, mainly in high-pressure packed towers^10,13,14.

Compared with the crystalline MOF-based materials, amorphous MOF-based adsorbent materials may exhibit different guest adsorption behaviors, mainly because of partial collapse of their frameworks^15,16. Additionally, it is easier to synthesize them on a large scale or shape their powders into different macrostructures when using them for practical applications¹⁷. Inspired by these, different research groups prepare amorphous MOF-based materials and use them for different applications^18–20. For instance, Zheng and coworkers synthesized different amorphous MOFs and applied them to remove different organic dyes²¹ and oxo-anions²² efficiently from water. More recently, Zhu et al.¹⁴ directly synthesized an amorphous MOF (DKTA-MOF) via a hydrothermal method to selectively and efficiently remove copper (Cu(II)) ions from water. Under optimal conditions (T = 318 K, pH = 5), it showed the maximum adsorption capacity of 160.5 mg/g for Cu(II) ions, higher than the value obtained for its crystalline counterpart.

Amorphous MOFs are usually produced by introducing various defects into the long-range ordered crystalline structures of crystalline MOFs. Although the structure of amorphous MOFs doesn’t exhibit long-range order, they still show a unique metal-ligand connectivity similar to that of their crystalline counterparts. Generally, disorder and structural defects can be introduced in the structure of MOFs by preventing crystallization and partially collapsing their crystalline structure under harsh conditions such as high temperature, high pressure, and high mechanical force^23–25. Therefore, amorphous MOFs can be prepared via two distinct strategies: direct synthesis from MOF precursors under specific conditions or amorphization of pre-synthesized crystalline MOF through heat treatment, high pressure, mechanical ball-milling, and similar methods^26–28.

Comparing these two synthesis strategies revealed that the direct synthesis of amorphous MOFs from the MOF precursors is an attractive strategy that has recently gained significant attention due to its outstanding advantages, including simplicity, quickness, good controllability, one-step, low-cost, precise control of its influencing parameters, etc.¹²

Recently, the existence of different antibiotics in surface water was frequently observed, mainly due to the growing usage of these materials for agriculture, animal husbandry, treatment of different diseases, etc.^29,30. Among different antibiotics, ofloxacin (OFL, C₁₈H₂₀FN₃O₄), a third-generation quinolone (a type of antibiotic that is more powerful towards gram-positive bacteria than other quinolone generations), has been frequently used in different applications due to its limited side effects and broad antibacterial activity^31,32. It is one of the most frequently used antimicrobial drugs that effectively treats different infections caused by various bacteria. But, similar to other types of the fluoroquinolone family, the injected OFL drug is not entirely absorbed by humans or animals; therefore, a large amount of the injected drug is discharged into the surface water in its original form or as metabolite species^33,34. Thus, it is essential to develop new materials for the efficient removal of this toxic pollutant from water.

Hence, in the present study, we use a unique synthesis strategy to synthesize different amorphous cerium (Ce)-based MOFs (Ce-MOFs), then apply them as adsorbents for the efficient removal of the OFL antibiotic from water. From our previous work on green synthesis of Ce-MOFs, it was found that the amorphous Ce-MOFs can be synthesized by increasing the amount of water as a solvent. It was found that water, as the greenest solvent, could be applied not only as the reaction medium but also as the activation solvent³⁵. However, due to the poor solubility of 1,4-benzene dicarboxylic acid (BDC) in aqueous solutions, even in the presence of sodium perchlorate monohydrate (NaClO₄.H₂O) as a salting-in agent, a large amount of BDC was maintained unchanged and did not participate in the MOF formation reaction³⁶. These undissolved BDC molecules can trap the pores of MOFs, leading to the formation of MOFs with low porosities³⁷. One strategy to overcome this challenge is separating the undissolved BDC molecules from the aqueous MOF precursor solution before starting the MOF formation. Accordingly, in the present study, a series of amorphous Ce-MOFs with a relatively acceptable porosity was synthesized in water at room temperature via a green and simple synthesis strategy.

Experimental section

Materials and reagents

All materials, such as Cerium(IV) ammonium nitrate ((NH₄)₂Ce(NO₃)₆), sodium perchlorate monohydrate (NaClO₄.H₂O), 1,4-benzene dicarboxylic acid (BDC), ethanol, acetone, and N, N-dimethylformamide (DMF), were purchased from Merck Co. (Germany) and used as-received. Moreover, Ofloxacin (OFL, 98%) as a model drug was purchased from Bio Basic Co. (Canada) in crystalline powder form.

Synthesis of amorphous adsorbents

To evaluate the effect of washing solvents on the properties of the resulting adsorbents, a series of Ce-MOFs was synthesized at room temperature based on our previous work³⁵ with significant modifications. Procedure A: 2 g of NaClO₄.H₂O, 33 mg of BDC, and 110 mg of (NH₄)₂Ce(NO₃)₆ were added to 10 mL of water and stirred at room temperature for 18 h. After completion of the reaction, the particles were separated by centrifugation (6000 rpm) and washed three times with 12 mL of water, each washing step performed under sonication for 10 min, to remove the NaClO₄.H₂O and unreacted MOF precursors. Subsequently, to complete the activation process, the water-washed samples were washed with different solvents (DMF or acetone) under sonication for 10 min, soaked in different solvents (water or acetone) for 3 days, and finally dried at 100 ^◦C overnight. As shown in Table 1, the samples synthesized via this procedure were named Ce-MOF-A, whereas those synthesized via procedure B were named Ce-MOF-B.

Table 1.

The amount of material used for the synthesis of different adsorbents and the properties of the resulting adsorbents.

Sample codes	NaClO4.H2O (g)	Used solvents	Yield a (%)	Surface area (m2/g)	Pore volume (cm3/g)	Average pore diameter (nm)
Ce-MOF-A-1	2	Water, DMF, and acetone	78 ± 14	282	0.292	4.14
Ce-MOF-A-2	2	Water	81 ± 13	152	0.185	4.86
Ce-MOF-B-3	2	Water	24 ± 7	27	0.104	15.39
Ce-MOF-B-4	2	Water and ethanol	36 ± 12	-	-	-
Ce-MOF-B-5	1	Water	33 ± 16	179	0.543	12.10
Ce-MOF-B-6	0.5	Water	21 ± 8	126	0.142	4.50

^a The production yield was calculated as the ratio of the mass of the obtained dry product to the initial mass of the organic ligand (BDC) used in the synthesis.

Procedure B: A different amount of NaClO₄.H₂O and 33 mg of BDC were added to 10 mL of water and stirred vigorously for 20 min, followed by sonicating at room temperature for 20 min to guarantee the dissolution of BDC in water. Subsequently, to screen the undissolved BDC molecules within water, the aqueous BDC mixture was filtered using filter paper. Then, 110 mg of (NH₄)₂Ce(NO₃)₆ was directly added to the resulting filtered BDC aqueous solution and stirred at room temperature for 18 h. After completion of the reaction, the particles were separated by centrifugation (6000 rpm) and washed three times with 12 mL of water, each washing step performed under sonication for 10 min, to remove the NaClO₄.H₂O and unreacted MOF precursors. Subsequently, the washed samples were soaked in water or ethanol (15 mL) for 3 days and finally dried at 100 ^◦C overnight. The main difference between procedures A and B is that, in procedure B, the undissolved BDC ligands are removed by filtration prior to the MOF formation, whereas in procedure A the reaction proceeds in the presence of undissolved BDC. A schematic illustration of the synthesis procedures A and B is provided in Scheme 1.

Scheme 1.

Schematic illustration of the synthesis procedures A and B for Ce–MOFs.

Characterization

Details related to the characterization methods and used equipment are expressed in the supporting information file.

Batch adsorption experiments

For the kinetic test, 20 mg of the amorphous MOF was dispersed in 80 mL of aqueous solution with the OFL concentration of 100 mg/L and stirred in dark conditions for 240 min, in which, after each time interval, 5 mL of OFL solution was taken away, and the concentration of OFL in aqueous solutions was determined using UV–Vis spectroscopy at a wavelength of 330 nm. For the isotherm test, 5 mg of the amorphous adsorbent was dispersed in 20 mL of aqueous solutions with the OFL concentration in the range of 25–400 mg/L and stirred in dark conditions for 240 min. If not stated otherwise, all adsorption tests were carried out by dissolving 5 mg of adsorbents in 20 mL of 100 mg/L OFL solution for 240 min at different temperatures (25–45 ^◦C), various pH values (2–10), and different salt concentrations (NaCl, 0.1–1 mol/L (M)), to investigate the impact of temperature, pH, and ionic strength, respectively. Usually, each adsorption test was repeated at least two times, and the average values were used to calculate the equilibrium OFL adsorption capacity (Q_e (mg/g)) using Eq. 1:

1

Where C₀ and C_e are the OFL concentrations (mg/L) at initial and after 240 min of the adsorption process, respectively. V is the volume (L) of OFL solutions and M is the mass (g) of adsorbents³⁸.

Results and discussion

Characterization of amorphous MOFs

To determine whether the synthesized adsorbents are crystalline or amorphous, XRD was performed on all prepared samples. As observed from Fig. 1A, all prepared samples only show several broad peaks at 2θ < 10^°, 2θ = 25–35^°, and 2θ > 50^°, demonstrating that all synthesized samples are basically amorphous³⁹. The diffraction patterns of samples synthesized via procedure A (Ce-MOF-A-1 and Ce-MOF-A-2) also exhibit a sharp peak at around 2θ = 17^°, which may originate from the crystallization of unreacted BDC molecules on the surface of MOFs or within their porous structures³⁰. It is clearly obvious that this peak had disappeared when the adsorbents were synthesized via procedure B. Since in this procedure, the undissolved BDC molecules were completely separated from the aqueous MOF precursor solution before starting the MOF formation, the crystallization of free BDC molecules is impossible.

Fig. 1.

(A) XRD patterns of the synthesized adsorbents in different conditions. (B) ATR-FTIR spectra of the selected adsorbent before and after adsorption of the model pollutant.

Moreover, after screening the undissolved BDC molecules via filtration, the weak and broad diffraction peak at 2θ < 10^° disappeared entirely, indicating the crystallinity of adsorbents is further reduced via this procedure, resulting in the formation of highly amorphous MOFs. Additionally, as shown in Fig. 1B, nearly most of the characteristic peaks of Ce-MOFs, such as the Ce–O stretching vibration in the range of 500–800 cm^− 1, C = C at around 1400–1600 cm^− 1, and C = O at around 1640–1740 cm^− 1, are observed in the spectrum of the selected adsorbent Ce-MOF-A-2⁴⁰. The symmetric and asymmetric stretching vibration modes of Ce-O-C have been observed at 551 and 513 cm^− 1, respectively. Moreover, the absorption peak at around 746 cm^− 1 is corresponded to the Ce-O bond, while the peak at 670 cm^− 1 is assigned to the bending vibrations of O-H bonds⁴¹. These amorphous samples exhibited the same ATR-FTIR spectrum as their crystalline counterparts, indicating that the chemical composition as well as the unique building blocks of these amorphous samples are similar to crystalline samples³⁵. However, it was found that the Ce content (≈ 49%) of this sample was slightly lower than those values measured for its crystalline counterpart⁴², further indicating the presence of more defects (particularly missing-node defects) in the structure of the synthesized amorphous sample.

The ATR-FTIR spectra of other synthesized samples (Figure S1) show the same result, confirming the successful synthesis of all samples via these different two synthesis strategies. After the OFL adsorption over this adsorbent, the ATR-FTIR spectrum of this adsorbent not only shows the characteristic peaks of Ce-MOFs but also shows some new peaks, indicating the adsorption of this model pollutant on the surface of amorphous particles³³. The presence of this pollutant on the surface of this adsorbent was also confirmed by its elemental mapping (Figure S2) after the adsorption process.

The porosity of synthesized adsorbents was studied by preparing N₂ adsorption/desorption isotherms, where their N₂ adsorption/desorption isotherms are plotted in Fig. 2. As observed from Fig. 2A, the N₂ adsorption/desorption isotherms of most samples belong to type IV, demonstrating the presence of mesopores in the structure of the synthesized adsorbents. Both samples synthesized via procedure A (Ce-MOF-A-1 and Ce-MOF-A-2) exhibit nearly the same N₂ adsorption/desorption isotherms. However, Ce-MOF-A-1, which was washed with DMF and acetone, exhibited higher surface area (282 m²/g) and total pore volume (0.292 cm³/g) in comparison with that washed only with water (Table 1). The same results were also observed when the crystalline samples were washed with water, DMF, and acetone³⁵. Due to the high solubility of BDC in DMF, most of the free BDC molecules trapped within the porous structure of adsorbents were removed. Moreover, using acetone as an exchange solvent leads to a better activation process, mainly due to the low boiling point as well as the low polarity of acetone compared to water. Accordingly, the trapped DMF and water molecules within the pores of Ce-MOF-A-1 were completely exchanged with acetone, leading to the formation of a sample with a higher porosity⁴³. However, all synthesized amorphous adsorbents showed lower surface areas compared to those values reported for their crystalline counterparts³⁵.

Fig. 2.

(A) N₂ adsorption/desorption isotherms and (B) pore size distribution of selected adsorbents.

Unlike samples synthesized via procedure A, those samples synthesized via procedure B showed distinctive N₂ adsorption/desorption behaviors. For example, Ce-MOF-B-5 and Ce-MOF-B-6 showed type IV N₂ adsorption/desorption isotherms with H3 hysteresis loops, suggesting the presence of wedge-shaped pores. Whereas, Ce-MOF-B-3 exhibited type III N₂ adsorption/desorption isotherms based on the classification accomplished by IUPAC⁴⁴. This type of isotherm is usually seen in non-porous or low-surface-area materials with very weak interactions. Accordingly, as reported in Table 1, this adsorbent exhibited the lowest specific surface area of 27 m²/g and total pore volume of 0.104 cm³/g, compared to other adsorbents.

Additionally, Fig. 2B exhibits the significant differences in the pore size distribution of the synthesized adsorbents. As expected, due to the non-porosity of Ce-MOF-B-3, its pore size distribution profile doesn’t have any peak in the region of micropores. Whereas other adsorbents show several peaks with different intensities in the region of micro- and mesopores, indicating the mesoporous nature of these adsorbents. Accordingly, the mesopores with larger sizes observed for these adsorbents are typically due to the formation of secondary pores between the agglomerated particles, which is consistent with their FE-SEM images (Fig. 3). As shown in Fig. 3, all synthesized samples show irregular shapes with different particle sizes, which may originate from the agglomeration of many small-sized particles. However, most of the particles were observed to be smaller than 100 nm.

Fig. 3.

FE-SEM images of the synthesized adsorbents at different scales.

Ce-MOF-A-2 and Ce-MOF-B-5, which have relatively the same surface areas, were selected as representative samples, and their thermal stability was investigated using thermogravimetric analysis (TGA). Figure 4 exhibits the TGA profiles of both samples, in which their profiles show a slight weight loss (~ 5–10%) at a temperature lower than 200 ^◦C originating from the evaporation of guest solvents, demonstrating that a large amount of solvents (especially water) were removed from the pores of MOFs during thermal activation. The other weight loss (~ 15–20%) originating from the evaporation of free BDC molecules within the porous structure of adsorbents, is observed at the temperature of 300–400 ^◦C. Although the pore volume as well as the average pore size in Ce-MOF-B-5 are large, its weight loss in this region is not high, demonstrating that filtration of undissolved BDC molecules can significantly remove most of free BDC molecules from the porous structure of frameworks. This phenomenon results in the formation of an adsorbent with higher thermal stability, in which its thermal stability is comparable to that of the crystalline counterpart^35,45. The final and sharp weight reduction beginning at 450 ^◦C (~ 25%) may be because of the decomposition of frameworks and the formation of metal oxide. More importantly, it was found that the regenerated sample exhibited excellent thermal stability, even better than the pristine Ce-MOF-A-2. This observation indicates the good stability of this adsorbent during adsorption/desorption cycles, making it an ideal candidate for practical water treatment applications.

Fig. 4.

(A) TGA and (B) DTG profiles of selected adsorbents.

Adsorption performance of amorphous MOFs

To determine the adsorption performance of the synthesized amorphous MOFs, all samples were applied as adsorbents to remove OFL from water. It was found from Fig. 5 that most of the synthesized amorphous adsorbents showed good adsorption performances towards OFL, in which their OFL adsorption capacities are in the following order: Ce-MOF-A-2 (81.2 mg/g) > Ce-MOF-A-1 (60.2 mg/g) > Ce-MOF-B-6 (57.3 mg/g) > Ce-MOF-B-4 (35.6 mg/g) > Ce-MOF-B-5 (23.9 mg/g) > Ce-MOF-B-3 (21.0 mg/g), which are nearly in line with their porosities (Table 1). The high adsorption capacity of Ce-MOF-A-2 may be due to its larger average pore diameters (4.86 nm), which allow OFL molecules to penetrate into the interior pores of the framework easily. Hence, this sample was selected as the optimum adsorbent, and further investigations were performed only on it.

Fig. 5.

The OFL adsorption capacities of the synthesized amorphous adsorbents.

Adsorption kinetics and isotherms

Adsorption kinetics is an essential factor in evaluating adsorbent materials in practical wastewater treatment applications and shows the time needed for the adsorption process to reach the equilibrium. As shown in Fig. 6A, the adsorption amount of OFL over Ce-MOF-A-2 was dramatically increased, primarily because of abundant free adsorption centers on the adsorbent material. Subsequently, the OFL adsorption capacity was gradually increased over time (from 30 to 120 min). Then it tends to be relatively stable at 120–240 min, suggesting that the adsorption equilibrium is reached, and 240 min is an adequate time for the adsorption process. Moreover, different kinetic models (details are in the supporting information file) were also applied to study the adsorption kinetics and determine the rate-limiting step. According to the fitted curves (Fig. 6B-D, using Eqs. S1-3) and the calculated correlation coefficients (R², listed in Table 2), it can be concluded that the OFL adsorption over Ce-MOF-A-2 fits well with the pseudo-second-order kinetic model, demonstrating that the chemical adsorption is the rate-controlling step in this adsorption process.

Fig. 6.

(A) The OFL adsorption capacity over the Ce-MOF-A-2 as a function of time, and the fitting to the (B) pseudo-first-order, (C) pseudo-second-order, and (D) intra-particle diffusion models. (E) The adsorption isotherm of OFL over the Ce-MOF-A-2, and the fitting to the (F) Langmuir, (G) Harkins-Jura, and (H) Freundlich isotherm models.

Table 2.

Calculated kinetics and isotherm parameters for the adsorption of OFL over the Ce-MOF-A-2.

Kinetic parameters	R 2						qe (mg/g)				K2 (g/mg min)
	Pseudo-first-order		Pseudo-second-order		Intra-particle diffusion		Experimental		Calculated
	0.9891		0.9991		0.9428		73.22		78.12		7.17*10− 4

Isotherm parameters	Langmuir				Harkins-Jura				Freundlich
	R2	qmax (mg/g)		KL (L/mg)	R2	A (mg/g)2		B	R2	n		KF (mg/g)
	0.9866	140.84		0.0227	0.8731	2000.0		2.60	0.8495	5.74		14.48

Following, the effect of initial OFL concentrations as well as adsorption isotherms was also studied on the OFL adsorption capacity of Ce-MOF-A-2 to determine the maximum adsorption capacity. It is found from Fig. 6E that the OFL adsorption capacity over Ce-MOF-A-2 was gradually increased with increasing the initial concentration of OFL from 25 to 200 mg/L. At the initial OFL concentration of over 200 mg/L, the OFL adsorption capacity tends to be nearly stable, and the maximum adsorption capacity of 115.71 mg/g is obtained as the initial OFL concentration is about 400 mg/L. The results of isotherm model fitting (using Eqs. S4-6) plotted in Figs. 6F-H revealed that the adsorption of OFL over Ce-MOF-A-2 better fits the Langmuir isotherm model compared to other isotherm models (details are in the supporting information file), indicating that the adsorption is monolayer and all adsorption sites have the same binding energy^46,47. The maximum adsorption capacity of 140.84 mg/g was calculated using the Langmuir model (Table 2), which is slightly higher than the obtained value for crystalline Ce-UiO-66-NH₂ MOFs⁴⁸, further indicating the superiority of amorphous adsorbents over their crystalline counterparts. Additionally, its OFL adsorption capacity is slightly higher or comparable to other values reported for the adsorption of this antibiotic by previous adsorbent materials (Table S1), indicating the outstanding potential of this amorphous adsorbent to remove OFL from water^33,49–51. More importantly, it was found that the OFL adsorption capacity of this sample was significantly higher than (approximately 2.4 times) the value measured for activated carbon (≈ 58 mg/g), a benchmark adsorbent, further confirming the potential of this amorphous Ce-MOF as an efficient adsorbent.

Effect of background

To better evaluate the ability of synthesized amorphous adsorbents in practical wastewater treatment applications, the impact of various parameters, including pH, salt concentration, and temperature, on the adsorption capacity of Ce-MOF-A-2 for OFL was systematically investigated. Since the pH of aqueous solutions can significantly affect the surface charge of adsorbents as well as their stability, the pH of the OFL solution was changed from 2 to 10. The OFL adsorption capacity depicted in Fig. 7A was gradually increased with rising the concentration of OH⁻ ions, and showed the highest value of 138.98 mg/g at the pH of 8.05. This observation may be because of changes in the electrostatic interactions between OFL molecules and the Ce-MOF-A-2 particles, originating from the changes in the surface charge of Ce-MOF-A-2 particles (Fig. 7B).

Fig. 7.

Effect of pH on the (A) OFL adsorption capacity and (B) zeta potential of Ce-MOF-A-2. Impact of (C) NaCl concentration and (D) temperature on the OFL adsorption capacity over Ce-MOF-A-2.

Since pKa values of the OFL molecule were 5.77 and 8.44³³, in pH < 5.7 it exists as the cationic form. On the other hand, as depicted in Fig. 7B, at pH < 4.5, the adsorbent has a positive surface charge. Therefore, under these conditions, the dominant interaction between the adsorbent surface and OFL molecules will be electrostatic repulsion. However, in the pH range of 5.7–8.4, OFL molecules mainly exist as zwitterionic forms, while the adsorbent has a negative surface charge. Thus, the highest OFL adsorption capacity was obtained at this range of pH, mainly due to the electrostatic interactions between both materials and the hydrophobicity of adsorbent particles that promotes the adsorption of OFL molecules⁴⁹. At pH > 8.4, the OFL molecule exists as the anionic form, and the adsorbent has a negative surface charge. Hence, the reduction in the OFL adsorption capacity at basic conditions may be because of increasing the electrostatic repulsion between anionic OFL molecules and negatively charged Ce-MOF-A-2 particles, along with the partial decomposition of frameworks as confirmed by FTIR spectra (Figure S4)⁴³. These observations indicated that the neutral solution is the optimal condition for the efficient removal of OFL from water by Ce-MOF-A-2 particles, mainly due to the contribution of different adsorption mechanisms under this condition (Scheme 2).

Scheme 2.

The proposed mechanisms for the adsorption of OFL over the Ce-MOF-A-2 particles.

The presence of electrostatic interaction between both materials was further confirmed by investigating the effect of ionic strength. As observed from Fig. 7C, the OFL adsorption capacity was significantly increased with rising NaCl concentration, indicating the enhancement of OFL-Ce-MOF-A-2 particle interactions mainly due to the presence of Cl⁻ ions. Under these conditions, Cl⁻ ions act as the salting-out agent and reduce the solubility of OFL molecules in water, thereby increasing the OFL adsorption capacity⁵². Besides electrostatic interactions, the adsorption of OFL over the Ce-MOF-A-2 particles involves other mechanisms, including hydrogen bonding, π-π stacking, and pore-filling, in which some of which had already been confirmed by ATR-FTIR analysis (Fig. 1B).

The thermodynamic experiment was performed to study the relationship between temperature and adsorption capacity. In this regard, the OFL adsorption was performed at a temperature range of 25 to 45 °C, and the obtained results are depicted in Fig. 7D. As described in Fig. 7D, the OFL adsorption capacity over the Ce-MOF-A-2 continuously increased with rising reaction temperature, mainly due to increasing the molecule movements and the collisions at the interface of solid-liquid¹⁴. This observation also indicates that the adsorption of OFL molecules over the Ce-MOF-A-2 particles is endothermic in nature and the adsorption is promoted by increasing the temperature⁵³.

Regeneration and reusability

The recycling adsorption performance of the amorphous adsorbent was also studied after washing the OFL-loaded Ce-MOF-A-2 particles with ethanol and water. As observed from Fig. 8A, the adsorption capacity of OFL over Ce-MOF-A-2 was gradually decreased with increasing the cycle time (from 72.58 to 35.33 mg/g), reaching 35.33 mg/g after five consecutive adsorption/desorption cycles, indicating that desorption and regeneration did not occur entirely⁴⁶. The presence of OFL molecules on the adsorbent after five successive adsorption/desorption cycles (Figure S3) confirms the fact that the desorption of OFL was not entirely. However, its TGA profile (Fig. 4), N₂ adsorption/desorption isotherms (Fig. 8B), and FE-SEM images (Fig. 8C and D) show that the thermal stability, porous structure, and morphology of the recycled adsorbent did not significantly change during consecutive adsorption/desorption cycles. Accordingly, the specific surface area of the recycled sample reduced from 152 to 74 m²/g, exhibiting a 51.32% reduction in its surface area. This observation may be because of the partial desorption of adsorbed OFL molecules, in which the adsorbed OFL molecules can partially block the pores of adsorbent particles, leading to a reduction in the surface area of regenerated adsorbent particles. Moreover, the structural stability of the regenerated adsorbent was further confirmed by ATR-FTIR analysis, which shows the preservation of the main functional groups after the regeneration process (Figure S5). Furthermore, it was found that the Ce leaching of sample during repeated adsorption/desorption cycles is negligible, in which its Ce content was reduced from 49% to 43% after five successive adsorption/desorption cycles.

Fig. 8.

(A) Recyclability of Ce-MOF-A-2 towards OFL adsorption in the cyclic adsorption/desorption test. (B) N₂ adsorption/desorption isotherms of regenerated Ce-MOF-A-2. (C and D) FE-SEM images of regenerated Ce-MOF-A-2.

Conclusion

In conclusion, a series of amorphous Ce-MOFs was synthesized via a novel water-based room temperature synthesis method using water as the sole solvent for synthesis and activation processes. The synthesized amorphous adsorbents were characterized by XRD, ATR-FTIR, N₂ adsorption/desorption, FE-SEM, and TGA analyses, in which the obtained results demonstrated the successful synthesis of these amorphous frameworks. All synthesized amorphous Ce-MOFs were applied as adsorbents for water decontamination, exhibiting good adsorption capacities (21–141 mg/g) for OFL antibiotic, slightly better than their crystalline counterparts. The highest OFL adsorption capacity of 138.98 mg/g was measured for the adsorbent synthesized in water and activated by only water, indicating the great potential of this synthesis strategy in the economical and large-scale production of amorphous adsorbents. Adsorption mechanism investigations revealed that the adsorption of OFL over the Ce-MOF-A-2 particles occurs via various mechanisms, such as hydrogen bonding, π-π stacking, electrostatic interactions, and pore-filling. The pH effect study indicated that the neutral solution is the optimal condition for the efficient removal of OFL from water by Ce-MOF-A-2 particles, mainly due to the contribution of many adsorption mechanisms under this condition. Moreover, the amorphous Ce-MOF-A-2 sample exhibited reasonable regeneration and reusability characteristics, making it an acceptable candidate for water decontamination. We believe this study presents an efficient and green strategy for the synthesis of amorphous adsorbents, increases our knowledge about Ce-MOFs, improves our understanding about the performance of amorphous Ce-MOFs in water treatment applications, and paves the way for extending the practical water treatment applications of amorphous Ce-MOFs.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Somayeh Saeedi and Amirhosein Ghorbani carried out the experimental works, performed the characterizations, and prepared the original draft. Hossein Molavi conceived the project, discussed the results, and thoroughly revised the manuscript.

Funding

This work was financially supported by the Iran National Science Foundation (INSF) with grant No: 40402113 and 40407665. The authors gratefully appreciate their support.

Data availability

All data relevant to this study are included in the manuscript and supporting information file.

Declarations

Competing interests

The authors declare no competing interests.