Green room-temperature HF-free synthesis of an amorphous chromium terephthalate framework with open Cr sites for highly efficient cyclohexene oxidation

Abstract

Cyclohexene oxidation has been the subject of extensive studies as the products obtained are important synthetic intermediates to produce fine chemicals and can be utilized in polyether polymer synthesis. Different research groups have focused on the development of heterogeneous catalysts for cyclohexene oxidation with environmentally benign oxidants. In this work, we synthesized and characterized a chromium-based MOF, referred to as amorphous Cr-MIL-LT, in an aqueous medium at room temperature without using hydrofluoric acid. The obtained amorphous Cr-MIL-LT possesses high porosity, small particle size and coordinatively unsaturated Cr sites in its structure. As a catalyst, amorphous Cr-MIL-LT exhibited superior catalytic activity in the oxidation of cyclohexene with a high conversion of 97%. The enhancement of the catalytic performance could be attributed to its amorphous structure and furnished open metal sites (Cr-sites), indicating that it can have potential applications in catalytic oxidation reactions. In addition, density functional theory (DFT) calculations show that both H₂O₂ and cyclohexene form stable adsorbed complexes with the MIL-101(Cr) active sites, with the H₂O₂ complex being significantly more stable. The reaction profile for the homolytic O–O bond cleavage of H₂O₂ on Cr³⁺ reveals a kinetically accessible process with modest endothermicity. The formation of oxygen-centered radical species is confirmed by spin-density and Bader charge analyses, which show substantial unpaired electron density on the dissociated ˙OH radical and Cr-bound OH.

Cyclohexene oxidation has been the subject of extensive studies as the products obtained are important synthetic intermediates to produce fine chemicals and can be utilized in polyether polymer synthesis.

1. Introduction

Metal–organic frameworks (MOFs) are a class of crystalline porous hybrid materials constructed from organic ligands and metal nodes.^1–3 In these structures, metal ions coordinate with polycarboxylic organic linkers to form one-, two-, or three-dimensional networks that possess tunable pore and cage sizes as well as excellent hydrothermal and chemical stability.^4,5 Owing to their versatile architecture, MOFs have shown great potential in applications such as gas adsorption and separation, catalysis, and drug delivery.^6–8

Among the reported MOFs, Férey and co-workers first prepared MIL-53(Cr) (Materials of Institute Lavoisier) hydrothermally, in which trimeric chromium(iii) units are interconnected by benzene-1,4-dicarboxylate (BDC) linkers to form a three-dimensional framework.⁹ Notably, MIL-53(Cr) exhibits a “breathing effect,” existing in three structural forms, namely, MIL-53 (Cr^III(OH))·(O₂C–C₆H₄–CO₂) or (HO₂C–C₆H₄–CO₂H)_0.75—pristine, (Cr^III(OH))·(O₂C–C₆H₄–CO₂) at a high temperature (ht), and (Cr^III(OH))·(O₂C–C₆H₄–CO₂)H₂O at a low temperature (lt),¹⁰ and it has been widely studied for applications in gas storage/separation, catalysis,^11,12 adsorption of organics,¹³ and drug delivery.¹⁴ Despite its potential, MIL-53(Cr) typically requires harsh conditions, such as hydrofluoric acid (HF), high temperature (∼493 K), and long reaction times (∼72 h) for its synthesis. Although alternative HF-free methods have been developed, such as the use of excess water or modulators, these approaches still demand elevated temperatures (448–493 K) and extended durations above 15 h.^15,16 Building on this, Férey etal. further developed MIL-101(Cr), another chromium–BDC framework, which offers a very high surface area, large pore volume, and the presence of coordinatively unsaturated Cr sites. These properties enable post-synthetic modification and guest encapsulation, making MIL-101(Cr) highly attractive for gas adsorption/separation, chromatographic separation, and heterogeneous catalysis.¹⁷ However, similar to the synthesis of MIL-53(Cr), the synthesis of MIL-101(Cr) typically requires HF, high temperatures (≥493 K), and prolonged time. Although microwave-assisted and modulator-based methods have been reported, they still rely on energy-intensive and potentially hazardous conditions. Overall, chromium–BDC MOFs, such as MIL-53(Cr) and MIL-101(Cr), are of considerable importance in catalysis and gas separation. However, the reliance on high temperatures and hazardous reagents like HF introduces significant limitations in terms of their energy consumption, safety, and scalability.

Traditional synthesis routes for MIL-101(Cr) heavily rely on hydrofluoric acid (HF) as a mineralizer to facilitate crystallization; however, the extreme toxicity and corrosive nature of HF pose severe environmental and health risks. To address these safety concerns, recent advances have increasingly focused on non-HF or green synthetic strategies for MOF fabrication. These eco-friendly approaches not only eliminate hazardous reagents but also optimize the thermal stability and structural properties.^18,19 To date, no reports have been published on the synthesis of Cr–BDC frameworks under mild or room-temperature conditions.

In this study, we report for the first time the room-temperature HF-free synthesis of a chromium dicarboxylate framework (amorphous Cr-MIL-LT). This work represents a safer and more energy-efficient alternative to conventional methods. The as-synthesized material was thoroughly characterized using X-ray diffraction (XRD), N₂ adsorption, scanning electron microscopy (SEM), Fourier-transform infrared (FT-IR) spectroscopy, and thermogravimetric analysis (TGA), and its properties were compared with those of pristine MIL-53(Cr) and MIL-101(Cr). Furthermore, we evaluated the catalytic performance of amorphous Cr-MIL-LT in the oxidation of allylic compounds, such as cyclohexene. Moreover, the oxidation mechanism over MIL-101(Cr) was investigated by DFT calculations. By demonstrating the feasibility for low-temperature synthesis, this work not only reduces the energy and safety barriers associated with chromium–BDC MOF fabrication but also expands the potential for the practical, large-scale production of these important materials.

2. Experimental

2.1. Materials

Chromium nitrate nonahydrate (Cr(NO₃)₃·9H₂O), chromium chloride hexahydrate (CrCl₃·6H₂O), terephthalic acid (H₂BDC), sodium carbonate (Na₂CO₃), N,N-dimethylformamide (DMF), hydrogen peroxide (H₂O₂, 30 wt%), cyclohexene, acetonitrile, ethanol, sodium benzoate, and deuterated chloroform (CDCl₃) were all purchased from Sinopharm and used as received without further purification.

2.2. Synthesis of disodium terephthalate (Na₂BDC)

Na₂CO₃ (1.06 g, 10 mmol) was dissolved in 50 mL of deionized water at room temperature. Subsequently, H₂BDC (1.66 g, 10 mmol) was added to the sodium carbonate solution. The mixture was continuously stirred at room temperature for 12 h until a clear solution was obtained. The solvent was then evaporated using a rotary evaporator at 80 °C, yielding a white solid of disodium terephthalate (Na₂BDC). The product was dried at 120 °C for 12 h before use.

2.3. Room-temperature synthesis of amorphous Cr-MIL-LT

In a typical procedure, an appropriate amount of Na₂BDC and either Cr(NO₃)₃·9H₂O or CrCl₃·6H₂O were dissolved in deionized water. The mixture was stirred for a fixed period, resulting in the formation of a gray-green precipitate. The solid product was collected by centrifugation and sequentially washed with ethanol and DMF, and 50 mL of ethanol at 70 °C for 3 h. Finally, the sample was dried for 3 h at 100 °C, as shown in Fig. 1. The resulting samples were denoted as amorphous Cr-MIL-LT–x–y–t–T, where x represents the molar ratio of Cr³⁺ to BDC²⁻, y is the volume of deionized water, t is the reaction time, and T indicates the reaction temperature. The optimized sample was further activated under vacuum for 6 h at 150 °C.

Fig. 1. Schematic of the preparation of amorphous Cr-MIL-LT at room temperature.

2.4. Characterization

Powder X-ray diffraction (XRD) patterns were recorded using a Rigaku D/Max-2550 diffractometer equipped with a SolX detector and Cu Ka radiation with a wavelength of λ = 1.5418 Å. Nitrogen sorption isotherms were obtained at −196 °C using a Beishide 3H-2000PS1 gas sorption and porosimetry system. Samples were normally prepared for measurement after degassing at 150 °C under vacuum until a final pressure of 1 × 10⁻³ Torr was reached. Scanning electron microscopy (SEM) images and energy dispersive X-ray spectroscopy (EDX) elemental mapping images were recorded on SUPRA 55, operated at an acceleration voltage of 200 kV. Fourier-transform infrared (FT-IR) spectra were recorded using a Nicolet iS10 spectrometer. Thermogravimetric analysis was carried out on a TGA Q500 instrument from TA Company under an N₂ atmosphere and a heating rate of 5 °C min⁻¹. ¹H nuclear magnetic resonance (¹HNMR) spectra were collected on a Bruker Avance III 400 HD spectrometer. Chemical shifts for ¹HNMR spectra are given in ppm relative to tetramethylsilane (TMS) (δ = 0.00 ppm), and CDCl₃ (0.5 mL), as the solvent and deuterium reagent, was added to the reaction solution.

2.5. Catalytic reaction

The oxidation of cyclohexene was carried out under an air atmosphere in a 25 mL single-neck glass flask equipped with a vigorous magnetic stirrer (1000 rpm) and immersed in an oil bath maintained at 40 °C. In a typical experiment, cyclohexene (2 mmol) and the catalyst (20 mg) were dispersed in 10 mL of acetonitrile, followed by the addition of 1 mL of H₂O₂ as the oxidant. The reaction mixture was stirred at 40 °C for 60 min. Liquid aliquots were withdrawn at regular intervals and analyzed by gas chromatography (Agilent 7890A GC, FID detector) using a 30 m HP-5 packed column, and cyclohexene conversion was calculated using eqn (1).

1

where A_i and A_f are the initial and final concentrations of cyclohexene, respectively. ().

Quantitative ¹HNMR spectroscopy was employed to determine the remaining amount of substrate and evaluate the catalytic conversion. 1,3,5-Trimethoxybenzene (TMB) was utilized as an internal standard. The molar amount of the remaining cyclohexene (n_Cy) in the reaction mixture was calculated using the relative integral areas of the characteristic ¹HNMR signals according to eqn (2).

2

where n_TMB is the known molar amount of TMB added as the internal standard (mmol), and I_Cy and I_TMB are the integrated peak areas corresponding to the olefinic protons of cyclohexene and the aromatic protons of TMB, respectively. N_Cy and N_TMB are the number of protons giving rise to the respective resonance signals (N_Cy = 2) for the olefinic protons of cyclohexene, and N_TMB = 3 for the aromatic protons of TMB. Cyclohexene conversion (conversion, %) was subsequently calculated using the initial molar amount of cyclohexene (n_Cy,0) according to eqn (3).

3

2.6. Density functional theory (DFT) simulation

All density functional theory (DFT) calculations were performed using the Vienna Ab initio Simulation Package (VASP 5.4.4)^20,21 with the projector-augmented wave (PAW)²² method and periodic boundary conditions. The Perdew–Burke–Ernzerhof (PBE)²³ exchange–correlation functional within the generalized gradient approximation (GGA) was employed, with Grimme's D3 (ref. 24) dispersion correction and Becke–Johnson damping (PBE-D3(BJ)) to account for the van der Waals interactions. A plane-wave kinetic energy cutoff of 520 eV and convergence criterion of 10⁻⁷ eV for electronic self-consistent iterations were used. The atomic positions were relaxed until all forces were below 0.03 eV Å⁻¹. For the truncated cluster model of the MIL-101-Cr active site (featuring a Cr₃O(OH)(H₂O)₂ core coordinated by six benzoate linkers, oxygen, OH, water molecules and one fluoride atom), calculations were carried out in a cubic supercell of 25 × 25 × 25 Å using a Γ-point-only (1 × 1 × 1) k-point mesh and Gaussian smearing (σ = 0.025 eV).

The DFT simulations were performed using the GGA + U formalism, which accounts for strong on-site Coulomb interactions, following Dudarev et al.'s approach.²⁵ This formalism includes an effective U parameter (U-J), where U represents the on-site Coulomb repulsion and J denotes the on-site Hund exchange coupling. The U parameter is set to 3.7 (ref. 26) for Cr for all simulations.

To model substrate binding and H₂O₂ activation, the terminal fluoride ligand observed in the X-ray structure was replaced by H₂O₂ or cyclohexene (Fig. 2). Gibbs free energies were computed as G = E(DFT) + ZPE − T·S, where E(DFT) is the electronic energy from VASP, and zero-point energy (ZPE) and entropy (S) corrections for the adsorbed species were derived from vibrational frequency calculations using the finite-difference method. Frequencies were calculated using VASP and processed with VASPKIT.²⁷ Transition states were located using the nudged elastic band (NEB) method²⁸ as implemented in VASP with the VTST tools developed by the Henkelman group.^28,29 Atomic charges were analyzed using the Bader charge analysis code developed by the Henkelman group.³⁰

Fig. 2. (a) Primitive unit cell of MIL-101-Cr. (b) Truncated cluster model used for the DFT mechanistic studies, comprising a Cr₃O(OH)(H₂O)₂ core coordinated by six benzoate linkers, oxygen, OH, water molecules and one fluoride atom. (c) DFT-optimized structure of the active site with the terminal fluoride ligand replaced by H₂O₂. (d) DFT-optimized structure with cyclohexene coordinated to the open Cr³⁺ site.

3. Results and discussion

3.1. Optimization of the synthetic parameters

In the synthetic system, the influence of various synthetic parameters, including the chromium source, molar ratio of the chromium precursor/Na₂BDC, crystallization temperature, and time, was investigated to find the optimal synthetic conditions. The samples obtained were mainly characterized using the N₂ sorption technique.

Nitrogen sorption in Fig. 3a shows type IV–like isotherms with a sharp uptake near P/P₀ ≈ 0.9–1.0, consistent with mesoporosity/textural interparticle pores rather than a purely microporous type I profile. Under identical synthesis conditions, the washing protocol strongly impacts accessible porosity: sample (a) adsorbs markedly more N₂ across the range than (b), indicating fewer blocked channels and better pore accessibility after washing. Fig. 3b illustrates the precursor effect: Cr(NO₃)₃·9H₂O delivers a BET surface area of 711 m² g⁻¹versus 232 m² g⁻¹ for CrCl₃·6H₂O. This ≈3× gain is plausibly an anion effect, where weakly coordinating nitrates compete less with BDC during assembly and are more readily removed upon activation, yielding a more open pore framework. In contrast, chloride can slow down hydrolysis/condensation or persist as pore-occluding residues, depressing the surface area. Together, the data indicate that rigorous washing and the use of Cr(NO₃)₃·9H₂O are decisive for maximizing the porosity and surface area of Cr-BDC under these conditions.

Fig. 3. N₂ sorption isotherms of amorphous Cr-MIL-LT prepared with different washing-solvents (a). Specific BET surface area of amorphous Cr-MIL-LT synthesized using different chromium precursors (b).

Nitrogen sorption isotherms of all the samples crystallized for 12–72 h display type IV–like profiles with high P/P₀ uptake, consistent with amorphous Cr-MIL-LT featuring mesoporosity with significant textural contributions (Fig. 4a). BET analysis shows a modest but systematic variation with time: 661, 645, 711, and 700 m² g⁻¹ for 12, 24, 48, and 72 h, respectively (Fig. 4b). The sample prepared at 48 h delivers the maximum surface area (711 m² g⁻¹), indicating an optimal balance between nucleation and growth under these conditions. Shorter preparation times (12–24 h) likely yield incomplete crystallization and residual occluded species, which dampen low-pressure uptake and reduce the accessible surface area. Prolonged time (72 h) maintains the amorphous Cr-MIL-LT isotherm shape but slightly lowers the BET surface area, plausibly due to secondary phase/impurity formation, overgrowth, or partial pore occlusion. Overall, Fig. 4 indicates that time is a sensitive knob, and 48 h is the preferred crystallization time to maximize the porosity and surface area of amorphous Cr-MIL-LT under the stated synthesis conditions.

Fig. 4. N₂ sorption isotherms of amorphous Cr-MIL-LT prepared at different crystallization times (a). Specific BET surface areas of amorphous Cr-MIL-LT prepared at different crystallization times (b).

At a fixed crystallization time of 48 h, all samples obtained between RT and 60 °C exhibit type IV–like N₂ isotherms with pronounced high P/P₀ uptake, consistent with the mesoporosity typical of MIL-101(Cr) (Fig. 5a). The BET surface area decreases monotonically with temperature – 711, 679, and 635 m² g⁻¹ at RT, 40 °C, and 60 °C, respectively (Fig. 5b). Thus, room-temperature synthesis maximizes the accessible surface area, while elevated temperatures progressively erode porosity. The persistence of the isotherm shape across temperatures indicates that the MIL-101(Cr) phase still forms but with a reduced accessible pore volume at higher temperatures. We attribute the decline to thermally driven crystallite coarsening/overgrowth, increased defect healing that collapses textural pores, and/or secondary phase formation, which collectively reduce external and mesopore contributions. Overall, Fig. 5 demonstrates that a lower crystallization temperature (RT) is preferable for achieving high-surface-area amorphous Cr-MIL-LT under these conditions, while higher temperatures are detrimental to porosity.

Fig. 5. N₂ sorption isotherms of amorphous Cr-MIL-LT prepared at different temperatures (a). BET surface areas of amorphous Cr-MIL-LT prepared at different crystallization temperatures (b).

Fig. 6a shows the N₂ sorption isotherms of amorphous Cr-MIL-LT prepared with varying water contents (10, 20, and 40 mL) at otherwise fixed, optimized conditions, and type IV–like N₂ isotherms with pronounced high P/P₀ uptake were observed, consistent with mesoporosity. The BET surface areas are 670, 711, and 690 m² g⁻¹ for the water contents of 10, 20, and 40 mL, respectively, identifying 20 mL as the optimum volume for maximizing the accessible surface (Fig. 6b). Too little water (10 mL) likely limits solvation/hydrolysis and yields incomplete pore development, while excess water (40 mL) can dilute reactants and favor larger crystallites or partial pore occlusion. Fig. 6c and d probe the effect of the Cr/BDC molar ratio and reveal a sharp optimum at Cr/BDC = 1. At Cr/BDC molar ratio = 0.5, the material exhibits a BET surface area of only 261 m² g⁻¹, indicating under-coordinated networks and residual occluded species. At Cr/BDC molar ratio = 2, the measured surface area drops to ∼16 m² g⁻¹, implying a substantial suppression in the formation of amorphous Cr-MIL-L0054 and/or formation of dense/secondary phases. At Cr/BDC molar ratio = 1, the largest uptake is exhibited with a steep capillary condensation near P/P₀ ≈ 0.9–1.0, characteristic of well-developed mesoporosity. The textural properties of amorphous MIL-101(Cr)-LT synthesized under various conditions, including BET surface area and Langmuir surface area and pore volume, are summarized in Table S1. Fig. 6e and f assess the influence of solution pH, adjusted with HNO₃/NaOH, and demonstrate that stronger acidity favors a higher surface area of amorphous Cr-MIL-LT. Lowering the pH from 4.9 to 3.9 increases the BET surface area from ∼340 to ∼711 m² g⁻¹, with the isotherm at pH 3.9 showing a much higher uptake across the pressure range. We attribute this pH effect to a faster and more uniform Cr³⁺ hydrolysis/BDC coordination under acidic conditions, which promotes open frameworks and facilitates the removal of guest species upon activation. Collectively, Fig. 6 shows three convergent knobs water content (20 mL), Cr/BDC molar ratio = 1, and pH (∼3.9)—that jointly maximize the porosity while preserving the amorphous Cr-MIL-LT phase. Integrating these with earlier optimizations yields, the final conditions for preparing optimized amorphous Cr-MIL-LT is as follows: Cr(NO₃)₃·9H₂O as the chromium source, Cr/BDC molar ratio = 1, 48 h crystallization at 20 °C, and 20 mL of water, thereby affording the highest BET surface area under the tested conditions.

Fig. 6. N₂ sorption isotherms of amorphous Cr-MIL-LT prepared with different water contents (a). Corresponding specific BET surface area (b). N₂ sorption isotherms of amorphous Cr-MIL-LT prepared with different Cr/BDC molar ratios (c). Corresponding specific BET surface areas (d). N₂ sorption isotherms of amorphous Cr-MIL-LT prepared at different pH values (e). Corresponding specific BET surface areas (f).

3.2. Structural integrity of amorphous Cr-MIL-LT

Considering that the composition of amorphous Cr-MIL-LT is very similar to that of MIL-53(Cr) and MIL-101(Cr), a detailed structural comparison among the three materials was conducted. The XRD patterns of MIL-53(Cr), MIL-101(Cr),³¹ and amorphous MIL-101(Cr)-LT are presented in Fig. 7a. As observed in the figure, the XRD pattern of amorphous MIL-101(Cr)-LT exhibits a broad diffraction peak at approximately 2θ = 8°, suggesting its tendency to adopt an amorphous structure. In contrast, both MIL-53(Cr) and MIL-101(Cr) display highly crystalline characteristics. These findings confirm that the three samples possess distinctly different structural features.

Fig. 7. XRD patterns (a) and N₂ sorption isotherms (b) of amorphous Cr-MIL-LT, MIL-101(Cr) and MIL-53(Cr).

Fig. 7b shows the N₂ sorption isotherms of the samples. Both amorphous Cr-MIL-LT and MIL-53(Cr) exhibit typical type IV isotherms at low relative pressures, confirming their mesoporosity, while MIL-101(Cr) displays type I isotherms with additional uptake at P/P₀ ≈ 0.1–0.2, indicative of its mesoporous cages. As summarized in Table 1, amorphous Cr-MIL-LT has a lower BET surface area (711 m² g⁻¹) and micropore volume (0.29 cm³ g⁻¹) than those of MIL-53(Cr) (1192 m² g⁻¹, 0.52 cm³ g⁻¹) and MIL-101(Cr) (3348 m² g⁻¹, 1.19 cm³ g⁻¹), which can be attributed to its amorphous structure. Additionally, the relatively low micropore volume of 0.29 cm³ g⁻¹ compared with the total pore volume of 1.11 cm³ g⁻¹ of amorphous Cr-MIL-LT indicates that a substantial fraction of porosity arises from textural/interparticle voids and larger non-microporous pores rather than intrinsic microporosity. This behavior is consistent with the amorphous nature of the material, where structural disorder and particle aggregation can generate additional meso-/macro-porous void spaces.

Table 1. N₂ sorption data of amorphous Cr-MIL-LT, MIL-101(Cr) and MIL-53(Cr).

Sample	S BET (m2 g−1)	S Langmuir (m2 g−1)	V Micro a (cm3 g−1)	V Total (cm3 g−1)
Amorphous Cr-MIL-LT	711	957	0.29	1.11
MIL-101(Cr)	3348	4239	1.19	1.32
MIL-53(Cr)	1192	1510	0.52	0.64

^aHK method.

Fig. 8 shows the FT-IR spectra of the samples. The band at 580 cm⁻¹ corresponds to the Cr–O stretching in the MIL series, while the peak at 730 cm⁻¹ indicates the out-of-plane bending of the hydrogen on para-disubstituted benzene. Additional benzene-related vibrations appear in the 700–900 cm⁻¹ range, including the 830 cm⁻¹ C–H bending band, and at 1130 and 1260 cm⁻¹, which are characteristic of terephthalic acid groups. Multiple benzene absorptions are also observed between 1450 and 1620 cm⁻¹.^32,33 In MIL-53(Cr), a weak band at 3540 cm⁻¹ reflects its residual –OH groups, while amorphous MIL-101(Cr)-LT lacks this feature, confirming the full coordination of hydroxyls. Likewise, MIL-53(Cr) shows H₂BDC vibrations at 1550 and 1390 cm⁻¹, which are absent in amorphous MIL-101(Cr)-LT, indicating the complete incorporation of the BDC linker. For MIL-101(Cr), the bands at ∼1550 and 1430 cm⁻¹ confirm its coordinated dicarboxylates, while these are not detected in amorphous MIL-101(Cr)-LT, further evidencing the absence of free dicarboxylic acid and the formation of coordinatively unsaturated Cr-open metal sites.

Fig. 8. FT-IR spectra of amorphous Cr-MIL-LT, MIL-101(Cr), and MIL-53(Cr).

The SEM images reveal that amorphous Cr-MIL-LT exhibits an aggregated morphology composed of nearly spherical nanoparticles, with particle sizes ranging from approximately 40 to 100 nm (Fig. 9a–c). The surface texture appears relatively rough, which may enhance the surface area and accessibility of the active sites of the material. The elemental mapping images (Fig. 9d–f) further demonstrate that Cr, O, and C are homogeneously distributed throughout the material, indicating a uniform composition and the successful integration of the organic linker with chromium nodes. The EDX spectrum (Fig. 9g) confirms the presence of Cr, O, and C as key elements in the framework with contents of 10.28%, 35.69%, and 54.03%, respectively. This composition is consistent with the expected stoichiometry of a Cr-based BDC coordination material (Table S2). Overall, the SEM–EDX results confirm that amorphous Cr-MIL-LT consists of nanosized spherical particles with uniform elemental dispersion, which is favorable for catalytic and adsorption applications due to the higher surface area-to-volume ratio and homogeneous distribution of active sites.

Fig. 9. (a–c) SEM images, (d–f) EDX elemental mapping images and (g) EDX spectrum of amorphous Cr-MIL-LT.

The XPS survey spectrum of amorphous Cr-MIL-LT (Fig. 10a) confirms the presence of Cr, O, and C, consistent with the expected framework composition. The high-resolution C 1s spectrum (Fig. 10b) can be deconvoluted into peaks assigned to C–C (284.7 eV), O–C–O– (286.2 eV), and –O–C O (288.6 eV), which are characteristic of the terephthalic acid linker. The Cr 2p spectrum (Fig. 10c) shows two main peaks at 576.6 eV and 577.9 eV, corresponding to Cr³⁺, along with a minor peak at 579.6 eV attributed to Cr⁶⁺, indicating that chromium predominantly exists in the +3 oxidation state with slight surface oxidation.³⁴ The O 1s spectrum (Fig. 10d) further reveals the contributions from carboxyl oxygen (532.5 eV), carbonyl oxygen (531.9 eV), and hydroxyl oxygen (531.2 eV). The relatively weak –OH signal suggests that most hydroxyl groups in the carboxyl linker are coordinated with chromium ions. These findings are consistent with the FT-IR results, confirming the coordination environment of the Cr–O bonds and verifying the chemical stability of the amorphous Cr-MIL-LT framework. XPS analysis also estimates the atomic weight percentage of Cr, O, and C in the amorphous Cr-MIL-LT framework, as represented in Table S2.

Fig. 10. XPS survey spectra of amorphous Cr-MIL-LT (a). High-resolution C 1s (b), Cr 2p (c) and O 1s (d) spectra.

The thermogravimetric analysis (TGA) curves illustrate the thermal stability of MIL-53(Cr) and amorphous MIL-LT under an N₂ atmosphere (Fig. 11). Both materials show an initial minor weight loss below 150 °C, which is attributed to the removal of adsorbed water and residual solvent. For MIL-53(Cr), the framework remains stable up to ∼350 °C, after which a sharp weight loss occurs due to the decomposition of the organic linker, leaving a final residue content of ∼40%. In contrast, amorphous MIL-LT undergoes gradual weight loss beginning at lower temperatures and exhibits complete decomposition at ∼400 °C, with a smaller residue content of ∼25%. These results indicate that MIL-53(Cr) has higher thermal stability than amorphous MIL-LT, while the amorphous structure of amorphous MIL-LT contributes to its reduced resistance to thermal degradation.

Fig. 11. TGA curves of MIL-53(Cr) and amorphous Cr-MIL-LT under an N₂ atmosphere.

3.3. Catalytic evaluation

To evaluate the catalytic applicability of amorphous Cr-MIL-LT in oxidation reactions, cyclohexene oxidation was employed as a probe reaction, which is of significant relevance in pharmaceutical and agrochemical processes. This reaction typically yields products such as 2-cyclohexene-1-ol (Cy-ol), 2-cyclohexene-1-one (Cy-one), and 2-cyclohexene-1-hydroperoxide (Cy-HP). As shown in Fig. 12a, amorphous Cr-MIL-LT achieved a cyclohexene conversion of 97% within 20 min, while MIL-53(Cr) and MIL-101(Cr) achieved only 72.3% and 65.5%, respectively, after 60 min. These results highlight the superior catalytic activity of amorphous Cr-MIL-LT, which can be attributed to the combined effects of its amorphous structure and the Lewis acidity of the coordinatively unsaturated open Cr sites. Comparatively, Fig. 12b and S1 shows the comparison between the amorphous Cr-MIL-LT/H₂O₂ system and other catalytic systems under the studied conditions. The complete catalytic system achieved the highest conversion of 97%, confirming the synergistic importance of both the catalyst and H₂O₂. In the absence of H₂O₂ or the catalyst, the conversion decreased to 74% and 69%, respectively, indicating that both components contribute substantially to the reaction efficiency. In comparison, homogeneous Cr(NO₃)₃ afforded only 47% conversion, demonstrating the superior catalytic performance of the amorphous Cr-MIL-LT catalyst.

Fig. 12. Catalytic activity of MIL-101(Cr), MIL-53(Cr) and amorphous Cr-MIL-LT (a) and the comparison of activity using different catalytic systems (b) in the oxidation reaction of cyclohexene.

3.4. DFT mechanistic insights

DFT calculations were performed to investigate the adsorption thermodynamics of H₂O₂ and cyclohexene on the open Cr³⁺ site of MIL-101-Cr, as well as the activation of H₂O₂ to generate hydroxyl radicals for cyclohexene oxidation. All calculations employed the truncated cluster model shown in Fig. 2b. The three Cr³⁺ cores were treated in their high-spin electronic configuration. Each Cr³⁺ ion (d³) adopts a local spin of S = 3/2, yielding a total quartet ground state achieved through the antiferromagnetic coupling of two chromium centers with the third unsaturated Cr³⁺ in its high-spin state. Spin polarization and on-site Coulomb repulsion were accounted for using the DFT + U method (U_eff = 3.7 eV).²⁶

The adsorption of both substrates is thermodynamically favorable. The calculated Gibbs free energy of adsorption (ΔG, 298 K) is −1.03 eV for H₂O₂ and −0.60 eV for cyclohexene. The coordination of H₂O₂ to the unsaturated Cr³⁺ site forms a strong Cr–O bond of 2.20 Å, comparable to the experimental Cr–F distance (2.05 Å). In contrast, cyclohexene binds more weakly through its C C double bond, with Cr⋯C distances of 3.22 Å and 3.33 Å. The overall geometry of the Cr core remains largely unperturbed upon substrate coordination (Fig. 2c and d). The observed π-coordination mode of cyclohexene is fully consistent with the initial step of its oxidation by the generated oxygen radical species, such as ˙OH.

The homolytic fission of H₂O₂ on the unsaturated Cr³⁺ site of MIL-101(Cr) begins with the formation of a reactant complex, in which H₂O₂ coordinates to the Cr³⁺ ion. The O–O bond of this complex undergoes homolytic cleavage, generating two adsorbed hydroxyl radicals (˙OH*): one directly bound to the Cr³⁺ site and the second hydrogen-bonded to the first.

Following fission, the Cr³⁺–˙OH bond is shortened to 1.90 Å, compared with the Cr³⁺–O distance of 2.20 Å in the initial H₂O₂ complex. The transition state lies 0.47 eV above the reactant complex, while the product complex is 0.57 eV higher in energy, indicating a slightly endothermic reaction. Nevertheless, the modest activation barrier renders the process kinetically feasible.

These results are consistent with those reported by Tian et al.^13,35 for homolytic H₂O₂ cleavage on CuFe₂O₄ surfaces. Spin density analysis along the reaction coordinates (Fig. 13) clearly shows the progressive localization of unpaired electrons on both oxygen atoms upon O–O bond cleavage. In the final product state, spin density is localized on one ˙OH radical, while the other remains coordinated to the Cr center. Notably, the Cr oxidation state increases from Cr³⁺ to Cr⁴⁺, reflecting electron transfer from the Cr 3d orbital to the π* antibonding orbital of H₂O₂, an electronic rearrangement that drives radical formation.

Fig. 13. Reaction energy profile of the homolytic O–O bond cleavage of H₂O₂ at the Cr(iii) site of MIL-101(Cr). Optimized structures are shown along the reaction coordinates, with key distances in Å. Spin-density isosurfaces (0.01 e/ų) illustrate the formation of the ˙OH radicals.

Spin-density analysis reveals that after O–O cleavage, unpaired electron densities of ∼0.75 e⁻ resides on the terminal ˙OH fragment, ∼0.06 e⁻ on the Cr-bound oxygen, and the remainder on chromium. Further, Bader charge analysis confirms the formal Cr³⁺ → Cr⁴⁺ oxidation, with the absolute charge of the reactive Cr site increasing from +1.25e (R) → +1.81e (TS) → +1.87e (P). This is accompanied by a corresponding increase in the total negative charge on the H₂O₂-derived fragment, which progresses as −0.036e → −0.424e → −0.574e. Together, these changes clearly evidence the transfer of electrons from the chromium center to the oxidant. The Bader atomic charges are reported in Table S3.

In summary, DFT calculations reveal that both H₂O₂ and cyclohexene form thermodynamically stable complexes on the Cr³⁺ sites of MIL-101, with the H₂O₂ adduct being significantly more stable. The homolytic cleavage of the O–O bond in the coordinated H₂O₂ is kinetically favorable, proceeding via a low-energy transition state and exhibiting only mild endothermicity. The radical nature of the resulting oxygen-centered species is clearly confirmed by spin-density analysis, which shows approximately 0.75 e⁻ localized on the terminal ˙OH radical, with the remaining spin density predominantly residing on the Cr center. This is further supported by Bader charge analysis, which tracks the progressive electron transfer from the metal to the OH fragments along the reaction coordinates.

3.5. Comparison of the catalytic performance

Table 2 shows the comparison of the catalytic performance of different heterogeneous catalysts for the oxidation of cyclohexene, which varies widely depending on the reaction conditions, catalyst type, reaction temperature, and reaction time. Many previously reported catalysts require either high temperatures, long reaction times, or large catalyst amounts to achieve cyclohexene oxidation conversion in the range of 65–91.4%. In contrast, the amorphous Cr-MIL-LT in this work achieves a high conversion of 97% under relatively mild conditions (40 °C, 20 min), demonstrating competitive or superior activity compared with more expensive catalysts. These results highlight the strong efficiency and practical advantages of the amorphous Cr-MIL-LT catalyst, especially considering its low cost, waste-derived nature, and ease of preparation.

Table 2. Comparison of the catalytic activity of different heterogeneous catalysts in the oxidation of cyclohexene.

Catalyst	Temp. (° C)	Time (h)	Conversion (%)	Ref.
Ni-MOF-74	80	4	81	36
Cu-MOF	80	96	65	37
CoFe2O4	100	6	91.4	38
Cu-trz MOF	70	24	66	39
SBA-15-DAFO-Pd(ii)	60	12	85.5	40
Ti-MCM-41	70	4	90	41
Amorphous Cr-MIL-LT	40	0.33	97	This work

3.6. Reusability test

To evaluate the recyclability of amorphous Cr-MIL-LT, consecutive catalytic runs were performed over five cycles (Fig. 14a). After each run, the catalyst was recovered by centrifugation and directly reused in a fresh cyclohexene reaction without any regeneration treatment. Cyclohexene conversion was quantified by ¹HNMR spectroscopy using 1,3,5-trimethoxybenzene as an internal standard (Fig. 14b). Remarkably, the catalyst retained a conversion of approximately 58% even after the fifth cycle without regeneration, indicating excellent economical operational stability and reusability. This high recyclability reduces catalyst consumption and waste generation, supporting the economic and environmental advantages of amorphous Cr-MIL-LT.

Fig. 14. Reusability of amorphous Cr-MIL-LT in the oxidation reaction of cyclohexene (a). ¹HNMR spectra of amorphous Cr-MIL-L after each cycle (b).

3.7. Leaching and radical scavenger tests

To verify the heterogeneous nature of the catalyst, a hot-filtration test was performed by removing the solid catalyst after 15 min of reaction and then allowing the filtrate to react further under identical conditions (Fig. 15a). In the presence of the catalyst, the conversion continued to increase from ∼60% at 15 min to nearly 98% after 60 min, while after catalyst removal, the conversion remained essentially constant at ∼60%. This lack of further conversion in the catalyst-free filtrate indicates that the active species remain associated with the solid catalyst, supporting the heterogeneous nature of the catalytic process.

Fig. 15. Leaching test (a) and the oxidation reaction of cyclohexene in the absence and presence of a scavenger (b).

Further, to assess the structural stability of the catalyst and exclude chromium leaching under the reaction conditions, the reaction filtrate was analyzed by inductively coupled plasma–optical emission spectroscopy (ICP-OES) using the Cr emission line at 267.71 nm. Prior to analysis, the filtrate was digested with aqua regia (3 : 1 HCl/HNO₃) to ensure the complete dissolution of any chromium-containing species. Chromium concentration was found to be below 1.2 ppm, indicating negligible metal leaching. Triplicate measurements showed good reproducibility. These results demonstrate the high stability of amorphous Cr-MIL-LT and support the conclusion that the observed catalytic activity originates predominantly from the solid catalyst, confirming its heterogeneous nature under the applied reaction conditions.

To identify the reactive radical species involved in the catalytic pathway, radical-scavenging experiments were conducted under the optimized reaction conditions. The addition of sodium benzoate, a hydroxyl-radical (˙OH) scavenger, markedly decreased the substrate conversion from 97% to 54% (Fig. 15b and S2). This substantial inhibition indicates that ˙OH radicals play a major role in the catalytic reaction pathway. In contrast, the comparatively limited effects observed with other scavengers further support hydroxyl radicals as the predominant reactive species, consistent with previous reports on related Cr-based catalytic systems.^42–45

4. Conclusion

Amorphous Cr-MIL-LT was successfully synthesized through a green, HF-free, and room-temperature route. The optimized amorphous framework exhibits high porosity (711 m² g⁻¹), a particle size of 40–100 nm, and coordinatively unsaturated Cr sites. Systematic tuning of the precursor ratio, pH, water volume, and reaction time established an efficient synthesis protocol under mild conditions. This strategy maximized pore accessibility while avoiding the use of hazardous reagents and high-temperature treatment conditions. XRD, FT-IR spectroscopy, N₂ sorption, XPS, and SEM/EDX analyses confirmed the formation of a homogeneous amorphous Cr-MIL-LT network dominated by Cr³⁺ species. Despite its lower surface area relative to crystalline MIL-53(Cr) and MIL-101(Cr), this material showed superior catalytic activity. A 97% cyclohexene oxidation yield was achieved within only 20 min, highlighting its remarkable efficiency. This enhanced performance is attributed to the synergy between Lewis-acidic open Cr centers and rapid diffusion through the nanostructured textural porosity. DFT calculations revealed the favorable adsorption of H₂O₂ and cyclohexene on the Cr³⁺ active site and a low-energy homolytic O–O bond cleavage pathway. Spin-density and Bader charge analyses confirmed ˙OH radical formation and progressive electron transfer, demonstrating the strong potential of this amorphous MOF as a sustainable catalyst platform for green oxidation reactions.

Conflicts of interest

There are no conflicts to declare.

Data availability

All data are available in the main text and supplementary information (SI). Supplementary information is available. See DOI: https://doi.org/10.1039/d6ra06771a.