Introduction

The fast-growing development of industrial activities and the relentless growth of the population have put immense additional stress on the resources of energy and the ecosystems, thus an immediate shift to renewable energy systems has become a necessity1,2,3. Conventional fossil fuels, plagued by limited supplies and adverse environmental impacts, have prompted widespread development towards sustainable bio-based substitutes4,5. Of these, fuels derived from lignocellulosic biomass hold a promising route, considering the large-scale availability of the raw material and the reduced competition with food products. A key to this strategy is the production of platform chemicals, including 5-hydroxymethylfurfural (HMF)6,7,8, a multi-potential precursor to valuable chemicals, including 2,5-dimethylfuran, levulinic acid, and biofuels9,10,11. However, commercialization of HMF is hindered by catalytic efficiency, selectivity, and stability issues and the compelling necessity for new catalytic systems. HMF synthesis significantly depends on the acid-catalyzed dehydration of hexoses using either homogeneous or heterogeneous catalysts in the form of mineral acids, heteropoly acids12, zeolites13, and ionic liquids14. Though homogeneous catalysts are useful, they are plagued by numerous limitations related to fast deactivation, tedious separation, and low recyclability. Though heterogeneous catalysts provide the advantages of operating, they tend to possess low acidity and structural instability under reaction conditions. Hence, efficient design of stable, multifunctional catalysts having proper acidic sites along with higher surface accessibility has become a crucial area of study. Natural materials like Boehmite (γ-AlOOH), have attracted interest in this regard due to their peculiar physicochemical properties, thermal stability, and hydroxyl groups on the surface that allow for precise functionalization and catalytic optimization15,16,17,18. Boehmite, B, has drawn significant attention as a catalytic support owing to its low toxicity, economic feasibility, and controllable surface chemistry19,20,21. Surface hydroxyl groups make it feasible to covalently modify B with acidic groups, thus augmenting both Brønsted and Lewis acid functionality while maintaining structural integrity. Current work highlights boehmite’s effectiveness in stabilizing the active species, facilitating dispersion, and inducing synergy between catalytic sites and the support22. Boehmite differs from sulfonated catalysts in inhibiting acidic group leaching, a design necessity for continuous operations. Boehmite’s double acidic functionalities, Brønsted acidity from the hydroxyl groups, and Lewis acidity from the coordinative unsaturation of Al³+ permits exquisite control over reaction pathways, reduce the tendency to form humin, which is a known by-product in the process of HMF synthesis, and increase HMF selectivity.

Considering the abovementioned discussion, in this research, a new heterogeneous catalyst is developed and synthesized using a Boehmite support. To this purpose, Boehmite was first modified with (3-chloropropyl) triethoxysilane (CPTES) to introduce reactive alkyl chloride groups, which facilitated anchoring of an imidazole-containing ionic liquid (IL) moiety.

ILs based on imidazolium cations, in particular, have drawn considerable interest for catalytic uses because of their low volatility, thermal and chemical stability, tunable acidity/basicity, and the capability to form microenvironments favorable to catalysis23,24. Within our design, the introduction of the imidazolium moiety appended a functionalized ionic liquid moiety to the Boehmite surface, providing both structural stability and probable catalytic function.

To further improve the acidity of the catalyst and enhance the feasibility of its usage in acid-catalyzed reactions, sulfonic acid functional groups were incorporated through post-synthetic sulfonation employing chlorosulfonic acid (Fig. 1). The resulting material, designated as B-IL, combined the advantages of ionic liquids, heterogeneous supports, and acidic Brønsted sites. Then, B-IL was used to catalyze the dehydration of fructose to synthesize the useful platform molecule HMF. According to the results, B-IL showed remarkable catalytic performance at optimized conditions, found using Response Surface Method, to reach 97% yield of HMF. The catalyst was further found to be highly reusable and a great potential material for sustainable processes of biomass conversion.

Fig. 1
figure 1

Schematic illustration of B-IL synthesis.

Full size image

Result and discussion

Characterization of B-IL

In Boehmite’s FTIR spectrum, Fig. 2A, there are clear and wide bands of absorption at 3080 and 3380 cm− 1, which relate to the O–H symmetric and asymmetric vibrations on Boehmite’s surface. Bands of absorbance at 480, 605, and 735 cm− 1 relate to the absorption by the Al-O bonds, whereas intense bands at 1070 and 1161 cm− 1 correspond to the hydroxyl group hydrogen bond vibrations. Changes can be seen in the FTIR spectrum upon the grafting of IL onto Boehmite. A new peak is seen at 2509 cm− 1 due to the stretching of –CH2, which indicates the existence of the grafted IL moiety. But it is important to note that the absorbance band due to the –C = N functional group in the structure of the IL overlapped by that of Boehmite, which makes it difficult to distinguish them individually. These FTIR spectral analyses prove that IL is successfully grafted onto Boehmite, confirming the existence of IL functionalities on the surface of the catalyst.

Fig. 2
figure 2

A: FTIR spectra of Boehmite and B-IL, B: XRD patterns of Boehmite and B–IL, C: TG curves of Boehmite and B-IL.

Full size image

To examine the stability of Boehmite through IL grafting process, the XRD pattern of Boehmite was compared to that of B-IL. As seen from Fig. 2B, the XRD pattern of Boehmite showed characteristic reflections at 2θ values of 14.62°, 28.92°, 39.10°, 49.16°, 55.32°, 65.26°, and 72.38°. Interestingly, the XRD pattern of B-IL showed the emergence of the same peaks without any noticeable shift in their positions. This indicates that the Boehmite structure was stable throughout the grafting of IL. Nevertheless, the intensity of the reflections in the B-IL pattern was slightly weaker compared to that of Boehmite, which is due to IL being on the surface of Boehmite25,26.

The thermogram, Fig. 2C, shows that the initial weight loss for Boehmite occurred in the range 100–250 °C, which was due to the gradual loss of absorbed water molecules through dehydration. Weight loss in the range 400–550 °C was possibly due to the decomposition of hydrocarbon. Thermogram of B-IL is distinguished from that of Boehmite. The weight loss observed in the TG curve of B-IL (Fig. 2C) over the temperature range of 100–250 °C is indeed primarily attributed to the evaporation of physically adsorbed and chemically bound water molecules from the catalyst surface, which is consistent with the dehydration behavior typical of Boehmite-based materials. This initial weight change does not reflect decomposition or a loss of the active catalytic components but corresponds to the release of moisture, which is inherent due to the hydrophilic nature of the Boehmite support and the ionic liquid functionalities. Furthermore, a sustained weight loss between 250 and 350 °C can be attributed to the decomposition of the grafted ionic liquid moieties on the Boehmite surface. This observation aligns with the increased acidity and catalytic performance demonstrated by B-IL, confirming that the active acidic sites remain stable below this temperature range. To clarify, the TG profile of the B-IL catalyst (Fig. 2C) demonstrates three distinct weight loss stages. The first weight loss of approximately 13 wt% below 150 °C is ascribed to the removal of physically adsorbed and interstitial water molecules. The second stage, occurring between 250 and 350 °C with a mass reduction of ~ 22 wt%, is attributed to the decomposition of the grafted ionic liquid moieties. A further major loss of ~ 45 wt% between 350 and 550 °C corresponds to the complete degradation of the IL structure. Beyond this temperature, the curve levels off and the catalyst retains ~ 20 wt% of its original mass, indicating the preservation of the inorganic Boehmite backbone.

FESEM images of B-IL, depicted in Fig. 3Aand B, indicated the aggregated-like morphology of the as-prepared catalyst, which is similar to that of pristine Boehmite25. Scrutinizing the surface of B-IL in images with higher magnification confirmed its rough surface.

Elemental mapping and EDS analyses were carried out to further characterize B-IL in Fig. 3C. These analyses confirmed the existence of carbon (C), chlorin (Cl), nitrogen (N), oxygen (O), aluminum (Al), silicon (Si), and sulfur (S) atoms in B-IL. The detection of O, and Al, confirmed the Boehmite structure, whereas the detection of S, C, N, and Cl confirmed the existence of IL. The presence of Si also affirmed successful functionalization with CPTES.

EDS revealed a sulfur (S) content of 13.73 wt% in B-IL, confirming the presence of the -SO3H group in the acid IL. Elemental mapping, Fig. 3D, verified a relatively uniform dispersion of atoms associated with the IL group, confirming an even formation of the IL on the Boehmite surface.

Fig. 3
figure 3

A and B, FESEM images of B-IL, C, EDS and D: Elemental mapping analyses of B-IL.

Full size image

It was believed that the introduction and stabilization of acidic IL onto Boehmite could enhance the overall acidity of the resulting B-IL catalyst. To validate this hypothesis, the Brønsted acidity of both Boehmite and B-IL was evaluated by determining their Hammett acidity function H0, as defined in Eq. 1.

$${\text{P}}{{\text{K}}_{\text{a}}}{\left[ {\text{I}} \right]_{{\text{aq}}}}+{\text{ Log }}\left( {\left[ {\text{I}} \right]/\left[ {{\text{H}}{{\text{I}}^+}} \right]} \right)\,=\,{{\text{H}}_0}$$
(1)

In this expression, 4-nitroaniline was employed as the basic indicator [I]. Also, PKa, [HI+], and [I] are the acidic constant of the indicator, protonated, and unprotonated state of the indicator, respectively. According to the Beer-Lambert law, the ratio of [I]/[IH+] can be quantitatively determined through UV-Visible spectroscopic measurements. The maximum absorbance wavelength of the indicator was identified at λmax = 382 nm. Subsequent absorbance measurements of the catalyst samples allowed for the calculation of [I] and [IH+], enabling the determination of H0 values (Table 1). The comparative analysis revealed that B-IL exhibits a significantly higher acidity than pristine Boehmite, confirming the acid-strengthening effect imparted by the acidic IL functionalization.

To further approve the role of IL grafting on the acidity of Boehmite, the catalytic activity of Boehmite and B-IL was examined for conversion of fructose to HMF. To this purpose, the target reaction was conducted in the presence of 35 wt% of the catalyst at 95 °C in 85 min and the yield of HMF in the presence of Boehmite and B-IL was compared. The results demonstrated low activity of Boehmite for fructose conversion, which resulted in only 29% yield of HMF. Upon grafting of the acidic IL, however, the HMF yield remarkably increased to 97%, confirming the pronounced effect of IL on the catalytic performance. In fact, this significant increase of the catalytic activity is ascribed to the higher acidity of B-IL compared to pristine Boehmite.

The values of [I] and [HI+] in Table 1 represent the relative percentages of the unprotonated and protonated forms of the indicator (4-nitroaniline), respectively, derived from UV-Vis absorbance measurements according to the Beer-Lambert law. For the B-IL sample, these values are:

[I] = 2.37% and [HI+] = 97.62% then the ratio of [I]/ [HI+] is 0.0243. According to Eq. 1,

H0 = 0.99 + Log (0.0234) = -0.62.

Table 1 Value of Hammett function (H0) for boehmite and B-IL.
Full size table

Optimization of B-IL catalytic process using RSM

Optimizing reaction conditions is critical for maximizing HMF yield. To systematically assess the synergistic interactions of multiple variables, Response Surface Methodology (RSM), a statistically robust optimization tool, was employed to evaluate key factors influencing fructose-to-HMF conversion. Three pivotal parameters were identified, encompassing reaction temperature (A), time (B), and catalyst loading (C). A quadratic regression model, validated by ANOVA (Table 2), revealed the following relationships. According to the derived equation from RSM, Eq. 2, it is noteworthy that synergistic effects: positive coefficients for the interaction of A (time), C (catalyst) and AB. Antagonistic effects: negative coefficients for B (temperature), AC, BC and quadratic terms (A², B², C²). It was found that the quadratic term C² showed the strongest inhibitory effect on the HMF yield, while the AC interaction was the least significant. These findings highlight the nonlinear interdependencies governing the catalytic performance in this system.

Table 2 The results of the analysis of variance (ANOVA).
Full size table
$$\begin{gathered} {\text{HMF Yield }} = + {\text{ 99}}.{\text{75}} + {\text{3}}.{\text{78 A}} - {\text{3}}.{\text{56 B }} + {\text{ 5}}.{\text{21 C }} - {\text{ 3}}.{\text{56 AB}} \hfill \\ \quad \quad \quad \quad \quad \quad - {\text{ 1}}0.0{\text{1 AC }} + {\text{ 1}}.{\text{59 BC }} - {\text{ 7}}.{\text{58 A}}^{2} {\text{ }} - {\text{ 4}}.{\text{16 B}}^{2} {\text{ }} - {\text{ 2}}0.{\text{24 C}}^{2} \hfill \\ \end{gathered}$$
(2)

The values of correlation coefficient R2, Adjusted R2, and Predicted R2 were 0.96, 0.94, and 0.73, that obtained from RSM data. The 3D surface plots of the interactions between the above parameters (i.e., A, B, and C) as a function of HMF yield are shown in Fig. 4. In Fig. 4a, the effect of reaction temperature vs. reaction time for HMF forming is indicated. These results were displayed by increasing the reaction time to 85 min, the maximum yield of HMF was obtained. By the results presented in Fig. 4b, the best catalyst loading was found to be 35 wt%. According to Fig. 4c, the best reaction temperature was selected as 95 °C.

Fig. 4
figure 4

The 3D surface plots of parameters A (Time), B (temperature) and C (amount of catalyst loading).

Full size image

Proposed mechanism

The catalyst employed in this study is an acidic Brønsted ionic liquid, comprising a as imidazolium cation functionalized with a sulfonic acid group and a chloride anion, which confers ionic stability. In the presence of this Brønsted ionic liquid (B-IL), fructose, a keto-hexose, is protonated, with its hydroxyl groups activated at specific positions (C1 and C2). This protonation facilitates the formation of stabilized ionic intermediates, including carbocations or enols, as shown in Fig. 5.

The reaction proceeds through three successive dehydration steps, during which three water molecules are eliminated, leading to the final product, HMF, features a furan ring substituted at positions 2 and 5 with aldehyde (–CHO) and hydroxymethyl (–CH-OH) groups, respectively. Throughout this process, the B-IL catalyst remains chemically unchanged and is available for reuse in subsequent catalytic cycles.

Fig. 5
figure 5

The plausible mechanism of dehydration of fructose to HMF under B-IL catalysis.

Full size image

Recycling

According to Fig. 6A, which illustrates the yield of HMF over six consecutive reaction cycles using the B-IL catalyst, in the first three runs, the yield remains consistently high at 97%, demonstrating the excellent initial performance and stability of the B-IL catalyst. However, in the fourth run, the HMF yield decreases to 93%, marking the first indication of catalyst deactivation. This decline continues in subsequent cycles, with the yield dropping to 90% in Run 5 and further to 85% in Run 6, reflecting a reduction compared to the initial value.

This trend indicates that while the B-IL catalyst exhibits high activity and stability in the early cycles, its performance gradually deteriorates with repeated use. The approximately 12% decrease in yield over six cycles suggests possible degradation of the catalyst’s active structure or the accumulation of inhibitory substances on its surface.

Fig. 6
figure 6

A: Recyclability of B-IL for dehydration of fructose to HMF, B: Comparison of FTIR spectra of fresh and reused B-IL, C: Hot filtration of B-IL under optimal conditions (Catalyst loading 35 wt% of fructose, 95 °C). D: Comparison of XRD patterns of fresh and reused B-ILs.

Full size image

The recovered B-IL catalyst after six reaction cycles was analyzed using FTIR spectroscopy. A comparative analysis of the FTIR spectra of the fresh and reused B-IL samples, as presented in Fig. 6B, reveals a high degree of similarity between the two. This observation substantiates that multiple recovery and reuse cycles do not induce any significant structural degradation or chemical alteration in the B-IL catalyst. Consequently, these results demonstrate the remarkable structural stability and robustness of B-IL under the applied reaction conditions, underscoring its potential for sustained catalytic performance over repeated cycles. Comparison of XRD patterns of fresh and reused B-IL, Fig. 6D, exhibited no significant changes, confirming the catalyst’s stability. In conclusion, these findings demonstrate the remarkable structural integrity and robustness of B-IL under the applied reaction conditions, emphasizing its capability for sustained and stable catalytic performance over multiple reuse cycles.

Hot filtration

To evaluate the effectiveness of heterogeneous catalysis, a hot filtration experiment was performed (Fig. 6C). To this purpose, the catalytic dehydration of fructose to HMF was carried out under the optimal conditions (95 °C, 0.035 g catalyst). 50 min after the start of the experiment, the reaction was stopped, the catalyst was separated from the reaction mixture, and the yield (62%) of the reaction was determined. Subsequently, the reaction was continued to 85 min, and the progress of HMF yield was monitored. All measurements were performed in triplicate to ensure reproducibility and reliability of the observed catalytic behavior. The results showed that the HMF yield remained at 62% and no more HMF was produced in the absence of the catalyst, proving the heterogeneous nature of the catalyst.

Kinetic study

According to research, fructose is assumed to be converted to HMF through a first-order process27,28. As a result, the reaction rate constant remains constant but is temperature-dependent. The reaction rate can be determined using Eq. (3)29.

In Eq. (3), the kinetics, the sign (-) indicates the consumption of the reactant and the sign (+) indicates the formation of the HMF. The symbol (r) serves as a synonym for reaction rate, which represents the change in reactant (fructose consumption) or change in product (HMF formation) over time.

$$- {\text{ }}r\left[ {fructose} \right]\,=\,+\,d\left[ {HMF} \right]{\text{ }}/dt\,=\, - \,d\left[ {fructose} \right]{\text{ }}/dt\,=\,+\,K\left[ {fructose} \right]$$
(3)

Equation (4) will be obtained if X is used as the conversion rate. In addition, the integration of Eq. (4) in Eq. (3) gives rise to the generation of Eq. (5).

$$\left[{fructose} \right]={\left[ {fructose} \right]_0} \times \left( {1- X} \right)$$
(4)
$$-ln\left({1-X} \right)=kt+C$$
(5)

By evaluating the conversion rates at four distinct temperatures (70, 85, 90, and 95 °C), Eq. (5) can be effectively applied to analyze the temperature-dependent reaction kinetics. This approach enables the construction of the -Ln(1-X) diagram as a function of time for four distinct temperatures, where the speed constant for each temperature can be determined from the slope of the line, as depicted in Fig. 7A. The relationship expressed in Eq. 5 is evident in Fig. 7A, where it can be observed that the rate constant value, k, increases with an increase in temperature. The temperature values of 70, 85, 90, and 95 °C correspond to rate constant values of k equal to 4.59, 5.55, 6.09, and 6.21, respectively. This establishes a direct correlation between the rate constant and temperature.

Fig. 7
figure 7

A: Determination of the value of the rate constant, k (performed for four temperatures as part of this research study), B: Calculating of the amount of Ea for the dehydration of fructose to HMF. C: Calculating the amount of ΔH≠ and ΔS≠ for the dehydration of fructose to HMF.

Full size image

Furthermore, Ea can be calculated by utilizing the Arrhenius Equation, represented by Eq. (6).

$$lnk=-{E_a}/RT+lnA$$
(6)

To achieve this objective, a graph has been constructed by plotting Lnk against 1000/T(K). In Fig. 7B, the inclination of the resultant line corresponds to -Ea/R, which has been derived using the value of R, the universal gas constant, equal to 8.314 J/(mol.K). Based on this computation, the Ea value has been determined to be 13.29 kJ/mol (Fig. 7B).

Moreover, the thermodynamic characteristics of the reaction, including the activation enthalpy ΔH≠, activation entropy ΔS≠, and activation Gibbs energy ΔG≠, were determined through calculations by utilizing Eqs. (7) and (8). These equations contain variables that hold specific physical meanings. The symbol kb represents the Boltzmann constant (1.38 × 10− 23 j K− 1), and the symbol h represents Planck’s constant (6.626 × 10− 36 j.s)30.

$$k={\text{ }}\left( {{k_b}T/h} \right){e^-}^{{\Delta G \ne/RT}}$$
(7)
$$\Delta G^{ \ne } = \Delta H^{ \ne } - {\text{T}}\Delta S^{ \ne }$$
(8)

Equation (9) was obtained using Eqs. (7) and (8). Additionally, a plot of Ln k/T vs. 1000/T was created, as shown in Fig. 7C. From this plot, the activation enthalpy of 10.34 kJ/mol and activation entropy of -241.5 J/mol were determined by analyzing the slope (-ΔH≠/R) and the intercept ( Ln kb/h + ΔS≠/R ), respectively30.

$$Ln\frac{k}{T} = \left( {\frac{{\Delta H^{ \ne } }}{2}} \right)\left( {\frac{1}{T}} \right) + \left( {Ln\frac{{K_{b} }}{h} + \frac{{\Delta S^{ \ne } }}{R}} \right)$$
(9)

Finally, the Gibbs activation energy was determined to be equal to 99.21 kJ/mol by utilizing Eq. (8).

Comparative study

To further appraise the performance of B-IL for HMF synthesis from fructose, its catalytic activity was compared with some selected catalysts, which were reported in the literature, Table 3. As listed, various types of catalysts, ranging from magnetic catalysts to metal-organic frameworks and denritic ones have been reported for this key reaction. As known, synthesis of dendritic catalysts included multi-step processes, which makes procedure time-consuming. Some other catalysts, such as PW-[VBIm]@Ca-Alg 25, exhibited excellent catalytic activity. However, this catalyst is composed of both heteropolyacid, and IL, which are encapsulated in alginate bead. Hence, the synthetic procedure is more complicated compared to B-IL. Regarding CDNS-SO42−/ZrO2, which is based on immobilization of SO42−/ZrO2 on a cyclodextrin polymer, comparable activity was observed.

In the case of B-IL, use of Boehmite, which is a naturally occurring and low-cost clay and its chemical modification with a recognized IL via a facile procedure resulted in a catalyst with a comparable activity with some other reported catalysts, indicating the efficiency of this bio-based catalyst for HMF production.

Table 3 Comparison of the performance of B-IL for conversion of Fructose to HMF with other catalysts.
Full size table

Experimental

Materials

All chemicals and reagents for the synthesis of the catalyst, including Boehmite (B), dichloromethane (CH2Cl2), imidazole (IM, 99%), chlorosulfuric acid (97%), (3-chloropropyl) triethoxysilane (CPTES, 95%), 1-propyl-1H-imidazole (97%), and toluene (> 99%), were sourced from Sigma-Aldrich and utilized without additional purification. Fructose (> 99%) and dimethyl sulfoxide (DMSO, > 99%), provided from Sigma-Aldrich, were employed as reagents to assess the catalytic performance of the catalysts.

Catalyst synthesis

This work entailed the synthesis of a heterogeneous catalyst B-IL that involved immobilizing an acidic ionic liquid on Boehmite through covalent attachment. The synthesis process, illustrated in Fig. 1, includes the following steps:

Preparation of B-Cl

Boehmite (1.0 g) was reacted with CPTES (3.0 mmol) in dry toluene (50 mL) under a N2 atmosphere and refluxed condition for 24 h. Afterwards, the resulting solid (B-Cl) was isolated by centrifugation, washed thrice with dry toluene (3 × 20 mL) and dried overnight at 70 °C.

Grafting of imidazole (IM) onto B-Cl and synthesis of B-IM

B-Cl (1.0 g) was dispersed in dry toluene (30 mL) and subjected to ultrasonic irradiation for 10 min to form a homogeneous suspension. Subsequently, IM (3.0 mmol) was added, and the suspension was refluxed for 12 h. Then, B-IM product was collected via centrifugation, washed with toluene (3 × 20 mL), and dried at 70 °C for 12 h.

Sulfonation of B-IM with chlorosulfonic acid and synthesis of B-IL

B-IM (1.0 g) was suspended in dry toluene (30 mL) under stirring condition at 0 °C. Then, a chlorosulfonic acid (5.0 mmol) solution in toluene (10 mL) was added in a dropwise manner over 30 min. The mixture was stirred for 24 h at room temperature, after which the final solid was separated by centrifugation, washed with toluene (3 × 20 mL), and dried at 70 °C for 12 h, Fig. 1.

Catalyst characterization

The crystalline structure of B and B-IL was examined using a Siemens D5000 X-ray diffractometer with Cu Kα radiation, scanning over a 2θ range from 10° to 90°. Functional group identification was conducted via Fourier Transform Infrared (FT-IR) spectroscopy using a PERKIN-ELMER Spectrum 65 instrument. Samples were prepared as KBr pellets, and spectra were collected between 400 and 4000 cm− 1 at a resolution of cm− 1, providing detailed information on the chemical functionalities.

Surface morphology and elemental composition were analyzed using a TESCAN MIRA 3 LMU Field Emission Scanning Electron Microscope (FESEM) coupled with Energy-Dispersive X-ray Spectroscopy (EDS). FESEM imaging revealed the catalysts’ surface features, while EDS enabled qualitative and quantitative elemental analysis. Thermal stability assessments were performed through Thermogravimetric Analysis (TGA) under an oxygen atmosphere with a heating rate of 10 °C/min, utilizing a METTLER TOLEDO apparatus. The Brønsted acidity of the catalyst was evaluated via Hammett method.

Synthesis of 5-hydroxymethylfurfural

Fructose (1 mg) was dissolved in 4 mL of DMSO, and the catalyst (0.02–0.05 mg) was added under atmospheric conditions. The reaction was systematically investigated over a temperature range of 65–125 °C in 55–115 min to identify optimal conditions. After each reaction cycle, the catalyst was recovered by centrifugation, washed with DMSO, and dried at 70 °C overnight for reuse. Under the optimized parameters (Temp. = 95 °C, time = 85 min, and 35 wt% catalyst loading), the B-IL catalyst resulted in a 97% yield of HMF from fructose.

The purification method for HMF

For the purification of HMF, a method adapted from prior literature was implemented35. Initially, the catalyst was removed by filtration from the reaction mixture. Subsequently, 9 mL of a saturated sodium chloride solution was added to the filtrate, inducing phase separation into an aqueous lower layer and an organic upper layer. The majority of HMF partitioned into the organic phase, whereas residual DMSO and trace amounts of HMF remained in the aqueous phase. The organic layer was then subjected to rotary evaporation to isolate HMF. To further recover HMF, 35 mL of diethyl ether was introduced into the aqueous phase, followed by distillation of the organic solvent to collect the purified product. This procedure ensures efficient separation of HMF from the reaction medium with minimal loss.

Analysis of HMF

The formation of HMF was confirmed by proton nuclear magnetic resonance (1H NMR) and gas chromatography-mass spectrometry (GC-MS). 1HNMR spectra were recorded using a Bruker DRX 400 MHz spectrometer with deuterated DMSO solvent, Figure S1. Quantitative analysis of HMF was performed by gas chromatography (GC) using an Agilent 6890 system equipped with a flame ionization detector (FID) and a G&W HP-5 ms capillary column. Nitrogen served as the carrier gas with a 100:1 split ratio. The GC and GC-MS operating parameters included an inlet temperature of 275 °C and a detector temperature of 285 °C. The oven temperature program started at 60 °C (for 1 min), increased at a rate of 20 °C/min to 280 °C, and held for 5–12 min, corresponding to a total run time of 17 and 24 min, respectively. The HMF production rate was calculated according to the Eq. 10:

where Mole (0) represents the initial moles of the substrate.

$$HMF{\text{ }}yield{\text{ }}\left( \% \right)\,=\,Mole{\text{ }}\left( {HMF} \right){\text{ }}/{\text{ }}Mol{e_{(0)}} \times {\text{ }}100$$
(10)

Conclusion

In this research, Boehmite, as an abundant and low-cost clay was chemically modified by an acidic IL to furnish an acidic heterogeneous catalyst for the conversion of fructose to HMF. Experimental results approved increase of the acidity of Boehmite upon conjugation of IL. To maximize the yield of HMF, RSM was utilized to optimize the reaction variables, including reaction time, temperature and catalyst loading and the results revealed that using 35wt% B-IL at 95 °C resulted in 97% yield of HMF in 85 min. Hot filtration test was also conducted to approve the heterogeneous nature of the catalysis. Noteworthy, B-IL exhibited high recyclability and preserved its activity for three successive runs. The kinetic study show that the Ea was 13.29 kJ/mol. Furthermore, thermodynamic parameters, i.e. ΔH≠ and ΔS≠, and Gibbs free energy were estimated as10.34 kJ/mol, -241.5 J/mol and 99.21 kJ/mol respectively.