Efficient Catalytic Upgradation of Bio-Based Furfuryl Alcohol to Ethyl Levulinate Using Mesoporous Acidic MIL-101(Cr).

Enlarging the dispersed micropores of metal-organic frameworks to the well-ordered mesostructure can enhance the accessibility of acidic sites for reactants resulting in the improvement of selectivity. With the regulation of cetyltrimethylammonium bromide/Cr3+ and ClSO3H sulfonation, a series of mesoporous MIL-101(Cr)-SO3H (MMSs) with multiple pore sizes and acidic properties were synthesized and explored to further facilitate the furfuryl alcohol (FA) ethanolization to ethyl levulinate (EL) for the first time. The optimal catalytic activity of 83.8% EL yield at full FA conversion demonstrated that the appropriate mesopore size, acidic density, and accessible -SO3H sites contributed to the superior performance of the MMS(0.3)-0.15. The turnover frequency value (14.8 h-1) obtained with MMS(0.3)-0.15 was comparable to the commercial Nafion NR50 (18.3 h-1) and much higher than the classical Amberlyst-15 (1.9 h-1), confirming the predominant catalytic performance. Furthermore, two coexistent reaction paths with 2-ethoxymethylfuran and 4,5,5-triethoxy-pentan-2-one as intermediates indicated the possibility of producing EL from FA.

Introduction

The valorization of abundant and reproducible biomass into biofuels and platform chemicals caters to the sustainable development strategy alleviating global warming and dependency on fossil fuels. In the past few decades, researchers devoted tremendous attention to the preparation of heterogeneous acid catalysts[1−6] such as N-doped carbon-based acidified IL hollow nanospheres,[7] metal–acid bifunctional catalyst (Au-H4SiW12O40/ZrO2),[8] hollow carbon mesospheres functionalized with arylsulfonic acid (e.g., ArSO3H-HMCSs),[9] and acidic niobia catalysts loaded with Zn-exchanged tungstophosphoric (Zn1TPA/Nb2O5),[10] and utilized them for the fabrication of alkyl levulinates (ALs).[11−14] The platform molecules, ALs, have attracted enormous attention because of their wide applications in fragrance industries and as biofuel additives[15] and precursors to prepare γ-valerolactone.[16] The mentioned catalysts with better acidity and texture properties (e.g., acidic density, good dispersity, and mesoporous and uniform structure) showed good to excellent catalytic efficiency in the alcoholysis of furfuryl alcohol (FA). The literature survey reveals that catalysts with appropriate acidity, excellent texture, low-cost raw materials, and facile preparation protocols are rarely studied, but are of high demand.

Owing to the high specific surface area, adjustable pore size, as well as numerous and versatile Lewis sites or functional sites derived from coordinative unsaturated metal sites (CUS),[17] metal-organic frameworks (MOFs) draw tremendous attention. Nevertheless, most porous MOFs are still confined to the microporous system (pore size < 2 nm), which are widely applied in gas storage,[18] adsorption and separation,[19] and molecular sensing,[20] with the micropores limiting the diffusion and effective mass transfer, especially in catalysts. It is unprecedented to create the mesostructure MOFs for improving the accessibility of bulky molecular to active species, reaction rates,[21] and product yields, as well as to expand the application of MOFs in more areas. The method to lengthen the linker to expand the pore size cannot guarantee the framework stability and interpenetration in the MOFs.[22] As a result, the introduction of hierarchical mesopores into MOFs with mesostructure by quoting structure-directing agents to attain stable mesoporous MOFs with tunable structures and controlled catalytic properties remains a significant challenge.[23,24] Conveniently, P123, F127, and cetyltrimethylammonium bromide (CTAB) were chosen to increase the pore size and adjust the material structures of MOFs. However, P123 and F127 are expensive, which could further affect the development of mesostructured MOFs.[25−27] Thus, it can be seen that CTAB micelles are a promising candidate and the combination of 1,3,5-trimethylbenzene (TMB) to be an structural guide agent to swell CTAB is a potential investigation.

Albeit suitable for broadening application in more areas with mesostructure MOFs, the functionalized meso-MOFs are still in blank, especially in catalytic biomass conversion, to the best of our knowledge. A comparison of the materials preparation[28,29] in acidic functionalization highlights that the relatively facile and massive method to adjust the approachable acidic sites is the postsynthesis approach.[30−32]

Herein, we investigate for the first time a facile and feasible strategy that has driven us to reasoningly conceive and prepare hierarchically micro- and mesoporous structure-functionalized MOFs with tunable pore size, pore volume, surface area, and acidic density. Cationic surfactant CTAB was chosen as a primary structure-directing agent, while Cr3+ and terephthalate (BDC2–) ions as framework-building blocks, followed by chlorosulfonic acid (ClSO3H) postsynthetic treatment to design and prepare the expected mesoporous materials MIL-101(Cr)-SO3H (MMS). Specifically, the mesoporous walls in these hierarchical MOFs are composed of crystalline micropores, including three-dimensional channels with 1.41 nm aperture. The novel development of heterogeneous catalytic conversion for the formation of ethyl levulinate (EL) from FA caters to the urgent demand, for avoiding the severe environmental pollution, equipment damage, and difficulty in separation and purification and evaluating the creatively hierarchically micro- and mesoporous structure-functionalized MMS.

Results and Discussion

Characterization of Catalysts

Figure shows the X-ray diffraction (XRD) patterns for the pristine M(0) in the absence of CTAB and the mesoporous samples with different CTAB/Cr3+ molar ratios. The diffraction peaks of M(0) and MS(0) are in accordance with the standard simulated pattern, and no new crystalline phases or impurity spectra were formed. However, the gradually weaker and broader diffraction peaks in Figure d–i reveal that the smaller and lower crystals on nanoscale guide the formation of mesostructure MOFs according to the literature.[33−35] From this, we can obtain that lower crystallinity and more irregular mesostructure grew with increase of the CTAB/Cr3+ molar ratios and the sulfonation treatment to accelerate it, which is consistent with a previous report.[34] All of the results emphasize the key role of CTAB in the hybrid structure composed of micropores and well-defined mesopores.

Figure 1

XRD patterns of (a) simulated MIL-101, (b) M(0), (c) MS(0), (d) MM(0.15), (e) MMS(0.15), (f) MM(0.3), (g) MMS(0.3), (h) MM(0.6), and (i) MMS(0.6).

The field emission scanning electron microscopy (FE-SEM) images of M(0) with the CTAB/Cr3+ molar ratio of 0 show typical octahedra morphologies and possess nanocrystals of size ca. 1 μm (Figure a), which is consistent with the neat MIL-101(Cr).[35] Different morphologies of the mesostructure samples are obtained by increasing the CTAB/Cr3+ molar ratios to 0.15, 0.3, and 0.6, with gradual growth of the ca. 600 nm length and ca. 100 width bunchy crystal assembly of regular crystalline size ranging from 100 to 200 nm (Figure b–d), which indicates the existence of mesoporous samples with crystallinity nanoscale particles, and the mesopore walls were formed from the crystalline microporous framework. The observed results suggested that CTAB showed obvious effects on crystal morphologies of MIL-101, in good agreement with the XRD observation, resulting in nanocrystals with a rough outside surface.

Figure 2

SEM images of (a) M(0), (b) MM(0.15), (c) MM(0.3), (d) MM(0.6), (e) MS(0), (f) MMS(0.15), (g) MMS(0.3), and (h) MMS(0.6).

The sulfonation reaction did not comparably influence the morphology (Figure e–h), but there was a slight reduction of particle size. Nevertheless, obvious smaller crystallite aggregates of UiO-66(Zr) appeared with particle size increasing from 100–200 nm to ca. 1.5 μm (Figure S2), which confirmed the instability of UiO-66(Zr)-SO3H[36] and corroborated well with the XRD results (see Figure S1).

Several representative transmission electron microscopy (TEM) images of the well-defined mesoporous structure with different scale plates are displayed in Figure . The uniform bright spots and the dark spots in Figure a,d in 20 nm exhibit the presence of nanometric crystallites guided by CTAB. The emergence of bunchy crystals demonstrates the construction of mesopore walls from uniform nanocrystal and random connection without long-range and regular order, which is in accordance with the SEM images and the reported disordered wormhole mesopore structures.[26] The diameter distribution of the MM(0.3) and MMS(0.3) has been estimated based on 50 points in the TEM results using software “Nano Measurer 1.2”, and their average particle sizes were found to be ca. 58.5 and 41.1 nm, respectively (Figure ).

Figure 3

TEM images of (a–c) MM(0.3) and (d–f) MMS (0.3) with different scale plates.

Figure 4

TEM images of (a) MM(0.3) and (b) MMS(0.3) with particle size distribution.

The porosity and the texture property parameters of the mesostructured MOFs affected by various CTAB/Cr3+ molar ratios are investigated by the N2 adsorption–desorption method, and the results are shown in Table . The mesostructure with functionalization by ClSO3H-treated and nontreated MOFs adjusted at CTAB/Cr molar ratios of 0, 0.15, 0.3, and 0.6 exhibits the regulatable pore size, SBET, pore volume, and titration. As listed in Table , the mesoporous pore size of these mesoscale MOFs with and without ClSO3H functionalization increases from 4.57 to 7.33 nm and from 2.41 to 3.23 nm, respectively, by enhancing the molar ratio of CTAB/Cr3+ from 0.15 to 0.60. Therefore, the Brunauer–Emmett–Teller (BET) specific surface area (SBET) and pore volume of the nontreated MOFs decreases rapidly and presents an even more steep glide trend of the functionalized ones; TEM and BET (pore size distribution) analyses manifest the catalyst structure being composed of micropores and well-defined mesopores, matching well with the characterization of XRD and SEM.

Table 1

Physicochemical Properties of Various CTAB-Cr Molar Ratio Catalysts

catalyst	pore size (nm)a	SBET (m2/g)a	pore volume (cm3/g)a	Atitration (mmol(H+)/g)b
MS	1.73	1492	0.86	1.01
M(0)	1.70	2603	1.30
MS(0)	1.40	1825	0.90	0.36
MM(0.15)	2.41	810	0.49
MMS(0.15)	4.57	162	0.19	0.62
MM(0.3)	3.20	638	0.51
MMS(0.3)	7.26	197	0.30	0.65
MM(0.6)	3.23	953	0.77
MMS(0.6)	7.33	23	0.04	0.72

The pore size and SBET were determined by N2 adsorption.

The amount of sulfonic acid sites was tested by acid–base titration.

The mesopore walls were assembled by the crystalline microporous framework within the designed mesostructured samples concluded from the XRD, SEM, and TEM analyses, which is further demonstrated by analysis of the distribution of pore sizes for MM(0.3) and MMS(0.3) (Figure S4). As shown in Figure S4, the solids contain micropores with diameters of 0.42 and 0.49 nm for MM(0.3) and MMS(0.3), respectively, which is much smaller than the micropore diameter (1.41 nm) detected from the pristine M(0)[26] due to the smaller particles and lower crystallinity. The pronounced H4 hysteresis loop located at high P/P0 (P/P0 = 0.3–1.0) revealed the existence of the mesoporous channels in the observed samples in type IV isotherm of Figure . The catalyst pore size distribution (Figure ) is consistent with the H4 hysteresis loop, confirming the presence of mesopores in the materials.

Figure 5

Nitrogen adsorption–desorption isotherms of MM(0.3) and MMS(0.3).

Effect of CTAB-Cr3+ Molar Ratio on the Catalytic Performance of MMS in FA Ethanolization for EL

The aforementioned characterization demonstrated that CTAB and sulfonation process modified and adjusted the porosity and the texture properties of the M(0) for the mesostructured MOFs. The direct relationships between the porosity of MOFs and the catalytic activity are established in Figure ; the gap between the FA conversion and yields might be attributed to the load of trace product in the pores of MOFs (EL conv., 5.7%; room temperature (rt); 2 h) and the generation of intermediates and byproducts (Figure S8 and Table S1).

Figure 6

Influence of CTAB-Cr molar ratio on MMS catalytic activity in FA ethanolization for EL. Reaction conditions: molar ratio of FA/EtOH = 1:60, 50 mg catalyst, 140 °C, and 2 h.

Hence, more detailed investigations of the effect of the porosity and the texture properties of MOFs on the conversion of FA to EL were performed for comparing the series MMSs and the initial catalyst MS prepared by one-pot hydrothermal synthesis. Without the sulfonation treatment, MM(0.15), MM(0.3), and MM(0.6) showed poor catalytic performance for the transformation of FA, which was comparable to the blank experiment with 18.5% FA conversion. However, catalysis by M(0) achieved high conversion (>40%), which can be ascribed to the Lewis acidic sites for the CUS (coordination of unsaturated metal sites).[37] As can be seen from Table Atitration, the sulfonated treated ones (MS(0), MMS(0.15), MMS(0.3), and MMS(0.6)) showed the highest EL yield, up to 76.3%, with the changeable acidic density from 0.36, 0.62, and 0.65 to 0.72 mmol/g, which indicated the key role of the superior acidic density of MMS(0.3) modified by ClSO3H. Meanwhile, the mesoporous structures of MS(0) (1.40 nm), MMS(0.15) (4.57 nm), MMS(0.3) (7.26 nm), and MMS(0.6) (7.33 nm) are capable of facilitating the diffusion of the reactants to the active center −SO3H, which is well supported by the enhanced catalytic performance (Figure ).

Figure 7

Influence of pore size on the catalytic activity in ethanolysis of FA to EL. Reaction conditions: molar ratio of FA/EtOH = 1:60, 50 mg catalyst, 140 °C, and 2 h.

Conversion of FA into EL with MMS Prepared Using Variable ClSO3H Amounts

Mesostructure MOF MMSs were prepared via postsulfonation treatments of ClSO3H, and the influence of the amount of ClSO3H on the samples properties was investigated. Both Fourier transform infrared (FT-IR) (Figure S3) and acid–base titration results (Table ) verified the successful sulfonation of −SO3H in MMSs. The results exhibited in Table imply that the Brønsted acid density of −SO3H in MOF–SO3Hs is reduced in the order MS (1.01 mmol/g) > MMS(0.3)-0.25 (0.87 mmol/g) > MMS(0.3)-0.20 (0.72 mmol/g) > MMS(0.3)-0.15 (0.65 mmol/g) > MMS(0.3)-0.10 (0.51 mmol/g) > MMS(0.3)-0.05 (0.15 mmol/g), and the homologous ethanolysis of FA to EL in EtOH at 140 °C for 2 h with MOF–SO3Hs show a hill-shaped trend. An initial experiment was conducted to explore the effect of the MS prepared by the one-pot hydrothermal synthesis with the functionalized linker (H2BDC–SO3Na) on the FA-to-EL transformation. Only 59.4% of EL yield was obtained with higher −SO3H density of MS, which is attributed to the speculation that high acid density may have adverse effects on the EL yield via FA conversion.[38] The catalytic performance of MS is consistent with the mesostructured ones, such as MMS(0.3)-0.25 and MMS(0.3)-0.20 with lower EL yields of 66.4 and 69.2% (Table , runs 6 and 7), respectively. For MMS(0.3)-0.15 with a medium Brønsted acidity density of 0.65 mmol/g, an EL yield of 76.3% was observed with a full FA conversion (Table , run 5). Decreasing the −SO3H density to 0.15 and 0.51 mmol/g, the EL yields draw down to 16.6 and 68.9% (Table , runs 3 and 4), respectively. In this regard, the catalytic efficiency of MMSs for the FA-to-EL transformation is a function of the −SO3H functionalization same as the sulfonic acid-site density.

Table 2

Catalytic Transformation of FA into EL with MMS Prepared Using Variable ClSO3H Amountsa

catalyst	Atitration (mmol(H+)/g)b	FA conv. (%)	EL yield (%)	2-EMF yield (%)	TOF (h–1)
		18.5	2.3
MS	1.01	100	59.4	9.3	6.8
MMS(0.3)-0.05	0.15	62.0	16.6	9.1	12.7
MMS(0.3)-0.10	0.51	100	68.9	1.7	15.5
MMS(0.3)-0.15	0.65	100	76.3	3.4	13.5
MMS(0.3)-0.20	0.72	100	66.4	4.6	10.6
MMS(0.3)-0.25	0.87	100	69.2	0	9.1

Reaction conditions: FA/EtOH = 1:60, 50 mg catalyst, 140 °C, 2 h.

The amount of sulfonic acid sites was tested by acid–base titration.

EL Production from FA Promoted by Different Catalysts

To explore the optimal catalyst with appropriate physicochemical properties, we compared the activities of the typical and different heterogeneous catalysts for the production of EL from FA (Table ), before which various parameters affecting the conversion were optimized for the optimum EL yield. Under the specified reaction conditions (Figures S6–S8), the conversion of FA into EL was demonstrated to proceed through the relatively stable intermediates, including 2-ethoxymethylfuran (2-EMF), 4,5-diethoxy-5-hydroxypentan-2-one (DHP), and 4,5,5-triethoxy-pentan-2-one (TEP). As the catalyst with moderate acid density, it can facilitate the conversion of FA to the intermediates, thus affording comparable FA conversion to that with more acid-site content. Further increasing the catalyst acid density is able to afford relatively high EL yields, while the use of excess catalyst may generate byproducts with decreased EL yield. Thus, 50 mg of catalyst with a 1:60 mole ratio of EtOH at 160 °C for 2 h was obtained as the best condition.

Table 3

Physicochemical Properties, Conversions, and Yields for FA Conversion to EL over Diverse Catalystsa

catalyst	pore size (nm)b	SBET (m2/g)b	pore volume (cm3/g)b	Atitration (mmol(H+)/g)c	FA conv. (%)	EL yield (%)	2-EMF yield (%)
					36.9	5.5	14.3
MS	1.73	1492	0.86	1.01	100	69.3	0
MMS(0.3)-0.15	7.26	197	0.30	0.65	100	83.8	3.4
Nafion NR50		<1		0.45	100	71.7	0
Amberlyst-15d	>50	0.35	4.8	4.70	100	75.8	0
UiO-66(Zr)-SO3H	3.92	6	0.01	0.95	82	29.3	28.5

Reaction conditions: FA/EtOH = 1:60, 50 mg catalyst, 160 °C, and 2 h.

The pore size, SBET, and pore volume were measured by N2 adsorption.

The amount of sulfonic acid sites was tested by acid–base titration.

Adopted from ref (39).

UiO-66(Zr)-SO3H showed the least activity for the formation of EL and gave 29.3% yield after 2 h, while the MMS(0.3)-0.15 sulfonated by 0.15 mL of ClSO3H mesostructured MOF shows the highest performance compared to the other catalysts used in the system. The latter has more acidity and larger mesopores than the former, and the moderate Brønsted acidity combined with Lewis acidity,[37] derived from the mesopore walls contributed from M(0), has been investigated and found to give a relatively higher EL yield. Nevertheless, MMS(0.3)-0.15 by postsynthesis suffered the drastic reduction of SBET (197 m2/g) compared to the one-pot MS (1492 m2/g). Hence, we can conclude that the catalyst with synergistic physicochemical properties, moderate mesopores, and acidity contribute to the better activity for EL yield. To confirm it, evaluation parameter of various catalysts was correlated with the turnover frequency (TOF) value in Figure . MMS(0.3)-0.15 is comparable to the industrialized Nafion NR50, though only 71.7% for the EL yield.

Figure 8

Yield of EL per acid site (TOF (h–1)). Reaction conditions: molar ratio of FA/EtOH = 1:60, 50 mg MMS(0.3)-0.15, 160 °C, and 2 h.

Catalyst Recycling

To evaluate the stability of the sulfonated solid acid catalyst (MMS(0.3)-0.15), we have performed the five-cycle ethanolysis of FA with EtOH to synthesize EL (Figure ). After the reaction finished, the catalyst has been separated by centrifugation, filtration, washing with EtOH several times, and drying in a vacuum oven at 80 °C for more than 24 h to be reused in the next run. No regeneration was done. The FT-IR bands assigned to −SO3H groups in the fresh and recovered catalysts remained unchanged (Figure S8), confirming the stability of the hybrid framework and the catalytic active sites. The yield of EL was found to decrease significantly from 83.8 to 66.6% after 2 h, and the EMF yields appeared in the fourth and fifth cycles, indicating that the yield change can be correlated with the blockage of pores or active sites. The accumulated oligomeric byproducts generated from the reaction process resulted in the blockage and even the decrease in acidity, which suppresses the access of reactants to the catalytic sites, thus resulting in the reduction of the catalytic activity in the recycles. Moreover, the thermogravimetric (TG) curves of the fresh and recovered catalysts disclose that the recovered MMS(0.3)-0.15 shows more weight loss compared to the fresh one, confirming the accumulation of oligomeric byproducts into the catalyst.

Figure 9

Catalyst recycling experiments of MMS(0.3)-0.15. Reaction conditions: molar ratio of FA/EtOH = 1:60, 50 mg MMS(0.3)-0.15, 160 °C, and 2 h.

Possible Mechanism

Two most feasible ways for the FA ethanolization to EL are proposed (Scheme ), considering the indented intermediates and byproducts (Figure S10 and Table S1), according to the reported literature.[40,41] A possible mechanism with two coexisting paths of FA transformation to EL is as follows: (1) (i) The protonation of the oxygen atom of furan ring, (ii) the protonated one further undergoes etherification reaction to form 2-EMF, and (iii) the acidification of which is then transferred toward EL, as exhibited in path 1 (Scheme S1). (2) Another path course contains the step of the protonation as well: (i) The protonation of hydroxyl group of FA toward intermediates 4,5-diethoxy-5-hydroxypentan-2-one (DHP) and 4,5,5-triethoxy-pentan-2-one (TEP), (ii) DHP attacked by EtOH again to convert into TEP with the delivery of water, and (iii) DHP and TEP all can release micromolecule to give the desired product EL. 2-EMF, DHP, and TEP tested by gas chromatography-mass spectrometry (GC-MS) all exist in this investigated system. Hence, from this investigation, it is clear that the coexistence of the two paths has the highest possibility.

Scheme 1

Possible Reaction Pathways for the Conversion of Furfuryl Alcohol (FA) to Ethyl Levulinate (EL) in EtOH over the Acidic Catalyst

Furthermore, the relative amount of the main product and intermediates for different times at 120 °C was identified by GC-MS (Table S2). The data analysis suggests that there is the formation of 2-EMF in path 1, the content of which is higher than the total intermediates of DHP and TEP in path 2 for the conversion of TEP to EL, which is endothermic as evidenced by thermodynamics,[42] in the preparation of EL. Meanwhile, the presence of 2-EMF, DHP, and TEP indicates the possible reaction course that the conversion mostly proceeds via paths 1 and 2, simultaneously.

Conclusions

Various MMSs were developed to improve the potential as efficient heterogeneous catalysts for the FA-to-EL conversion process. MMS(0.3)-0.15 possessed enhanced mesopore size and appropriate acidic density and exhibited superior catalytic performances with 100% FA conversion and 83.8% yield of EL at 160 °C in 2 h. The catalytic activity suggested a complicated interaction between tunable texture and acidity of the catalysts, where the increased mesostructure may enhance the effective approachability of the catalytic sites during the reaction course. The high TOF value of MMS(0.3)-0.15 demonstrated the overall property, including the physicochemical properties. Catalyst recycling studies of the spent MMS(0.3)-0.15 catalyst suggested a fairly good catalyst stability. The exploration of reaction mechanism revealed the existence of the two possible paths.

Methods

Materials

2-Sulfoterephthalic acid monosodium salt (>98%) was obtained from Shanghai J&K Chemical Ltd; CTAB (>99%) was bought from Youpu Chemical Reagent Co., Ltd (Tianjin, China); protonated Nafion NR50 was offered by Tianjin Alfa Aesar; FA (98%) was obtained from Maya Reagent Co. Ltd.; and ClSO3H was obtained from Beijing Chemical Agents Company. Chromic nitrate nonahydrate (99%), terephthalic acid (99%), TMB (>98%), hydrofluoric acid (HF, 49% in water), Amberlyst-15 (AR), and EL (AR) were bought from Aladdin Industrial Inc.; 2-EMF (>98%) was provided by Shanghai Bide Pharmatech Ltd; CrO3, HCl, and EtOH of analytical grade were obtained from Chongqing Chuandong Chemical Reagent Ltd. without any purification. Deionized water was prepared by Milli-Q Advantage A10 instrument.

Catalyst Preparation

Preparation of Mesoporous MIL-101(Cr) (MM)

Cr(NO3)3·9H2O (3 mmol, 1.2 g), H2BDC (3 mmol, 0.5 g), the supramolecular templating agent CTAB/TMB (0–1.8 mmol, 0–0.66 g/0 or 0.9 mmol, 0 or 0.126 mL with changeable synthetic molar ratio), HF (0.1 mL), and deionized water (15 mL) were placed into a 25 mL stainless steel autoclave, which was heated to 493 K and kept for 8 h. After cooling to rt, the reaction product was extracted three times with ethanol to remove the template (CTAB/TMB) from the framework (ca. 1 g of powder and 100 mL of ethanol). Ultimately, the obtained solid was dried at 423 K in a vacuum oven to achieve the constant weight, powderlike MM. To explore the structure effect of CTAB, a mixture of CTAB/Cr(NO3)3·9H2O with different molar ratios of 0, 0.15, 0.3, and 0.6 was prepared and the resulting sample was coined as M(0) or MM(x) (x denotes the CTAB/Cr3+ molar ratio).

Sulfonation of MIL-101(Cr) and Mesoporous MIL-101(Cr)

The activated MIL-101(Cr) (0.25 g) was stirred in 7.5 mL of CH2Cl2 for 20 min at rt; then, ClSO3H in 2.5 mL of CH2Cl2 was added under even stirring conditions.[29] After 4 h, the resulting solid was filtered off and flushed with CH2Cl2 and ethanol before the neutral filter liquor was obtained and dried to give mesoporous MIL-101(Cr)-SO3H. UiO-66(Zr)-SO3H was synthesized following the same synthetic procedure. The amount of the ClSO3H for the preparation of MMS(0.3) with 0.05, 0.1, 0.15, 0.2, and 0.25 mL per 0.25 g of MM was varied to further investigate the acid density effect in the reaction system, and the corresponding catalysts are denoted as MMS(0.3)-y (y denotes the used amount of ClSO3H). The general MMS(x) were sulfonated by 0.15 mL of ClSO3H per 0.25 g of MM, i.e., MMS(0.3) denotes the sulfonation of mesoporous MIL-101(Cr)(0.3) with different amounts of ClSO3H, while MMS(0.3)-0.15 denotes the treatment of 0.25 g of MM(0.3) with 0.15 mL of ClSO3H. For comparison, MIL-101(Cr)-SO3H (MS) and UiO-66(Zr) were also synthesized by following the reported methods.[43−45]

Catalyst Characterization

The crystal structures were demonstrated by X-ray diffraction (XRD, Shimadzu XRD-6000) with Cu Kα radiation (λ = 0.1541 nm) at a scanning rate of 0.5 °/min. The porosity specialities of the MOFs were measured by nitrogen adsorption–desorption isotherms at −196 °C by a Micromeritics ASAP 2020 system. FT-IR spectra were confirmed using a Nicolet 360 FT-IR spectrometer, and the samples were prepared by the KBr pellet. The extent of sulfonation and the sulfonic acid amounts of MMS were detected by acid–base titration using saturated NaCl solution to confirm the ion exchange. MMS (0.5 g) was suspended in 20 g of aqueous NaCl saturated solution. The mixed suspension was stirred at rt for more than 24 h to reach equilibrium, using 30 mL of deionized water for the next filtration and washing. Eventually, the achieved filtrate was titrated by 0.1 M NaOH solution. The morphologies of the MOFs were characterized by a field emission scanning electron microscope (Zeiss SUPRA 55) operating at 20 kV. Transmission electron microscopy (TEM) results were provided by combined Philips Tecnai 20 and JEOL JEM-2010 HR-TEM at 200 kV.

Typical Process for Furfuryl Alcohol Transformation into Ethyl Levulinate

FA (0.1 mL, 1.15 mmol), MMS (50 mg), and EtOH (4 mL, 69.00 mmol) were mixed into a 25 mL stainless steel-coated liner with a screw cap. The reactor was placed into an oil bath and kept at 160 °C for 2 h with even stirring. After the designated reaction time, the reactor was shifted out from the oil bath, quenching the reaction with tap water. After dilution and centrifugation, the supernatant was filtered through a Nylon 0.45 μm syringe filter before analysis with Agilent 7890 B gas chromatography (GC).

Analysis Methods

Quantitative detection of FA, 2-EMF, EL, and liquid products was performed using Agilent 7890 B GC and GC-MS (Agilent 6890-5973) similar to the previous method reported by Liu and co-workers.[44] Naphthalene was chosen as the internal standard.