Ru/UiO-66 Catalyst for the Reduction of Nitroarenes and Tandem Reaction of Alcohol Oxidation/Knoevenagel Condensation.

A 3.1% Ru/UiO-66 material was prepared by adsorption of RuCl3 from ethyl acetate on to MOF UiO-66, followed by reduction with NaBH4. The presence of acid-base and ox-red sites allows this 3.1% Ru/UiO-66 material acting as a bifunctional catalyst for the reduction of nitroarenes and tandem reaction of alcohol oxidation/Knoevenagel condensation. The high efficiency of 3.1% Ru/UiO-66 was demonstrated in the reduction of nitroarenes to amines. This system can be applied as a catalyst for at least six successive cycles without loss of activity. The 3.1% Ru/UiO-66 catalyst also was active in the tandem aerobic oxidation of alcohols/Knoevenagel condensation with malononitrile. However, the activity of this catalyst strongly decreased in the second cycle. A combination of physicochemical and catalytic experimental data indicated that Ru nanoparticles are the active sites both for the catalytic reduction of nitro compounds and the aerobic oxidation of alcohols. The activity for the Knoevenagel condensation reaction was from the existence of the "Zr n+-O2- Lewis acid-base" pair in the framework of UiO-66.

Introduction

Metal organic frameworks (MOFs), being the porous and crystalline coordination polymers formed from the coordination of metal ions or clusters and chelating ligands, have attracted considerable attention in many fields, such as drug delivery systems, sensor, gas storage, as well as separation, electrochemistry, and catalysis, due to their large surface area, high adjustability of metal nodes, and organic ligands.[1−19] In addition, the highly well-organized nanometer-sized nanocages in MOFs allow them to be possible as ideal platforms to anchor metal nanoparticles.[20,21] However, most MOFs have the weakness of rather low thermal, hydrothermal, and chemical stabilities compared to those of zeolites, which will undoubtedly limit their applications on a large scale. Fortunately, a zirconium-based MOF (UiO-66) was confirmed with the character of either large surface area or very high thermal stability.[22] The UiO-66, resulting from the alternating combination of hexanuclear zirconium clusters with bridging ligand 1,4-benzenedicarboxylate, provides a robust, three-dimensional porous structure. It possesses a decomposition temperature above 500 °C and can tolerate many chemicals and water.[23] In addition, there are some hydroxyl groups on the surfaces of hexanuclear zirconium clusters helping the anchoring of other transition metals. Therefore, UiO-66 can be regarded as an ideal support for developing heterogeneous catalysts with special performance.

Up till now many transition metals, such as Pt, Au, Pd, V, and Ti, have been successfully introduced into UiO-66 or its derivatives by the postsynthetic method to obtain catalysts in various catalytic reactions.[20,24−35] For example, Pt nanoparticles were successfully confined inside the cavities of amine-functionalized UiO-66 (UiO-66-NH2), forming a catalyst for selective hydrogenation of cinnamaldehyde, exhibiting high activity and selectivity toward cinnamyl alcohol;[24] Au nanoparticles encapsulated in UiO-66 showed excellent catalytic performance and good reusability for the aerobic oxidation of benzyl alcohol to benzaldehyde under solvent-free conditions.[36] The nanopalladium particles, respectively, supported on UiO-66 and UiO-66-NH2 were used in the catalytic hydrogenation of phenol in aqueous media.[33] Interestingly, Pd–UiO-66 gave higher selectivity toward cyclohexanol but Pd–UiO-66-NH2 showed higher selectivity toward cyclohexanone under the same conditions. Because of the multifunctional catalysis of ruthenium, more attention has been paid to the deposition of ruthenium onto UiO-66 and its derivatives. Ru nanoparticles supported on UiO-66 showed good performance in various reactions, such as the selective hydrogenation of furfural to furfuryl alcohol under mild conditions[37] and the methanation of CO2 in high yield.[38] Ru nanoparticle deposition onto sulfonic acid-functionalized UiO-66 (UiO-66-SO3H) exhibited dual-functional catalysis, converting biomass-derived methyl levulinate into g-valerolactone efficiently under the mild conditions of 80 °C and 0.5 MPa H2.[39]

On the other hand, neat MOFs including UiO-66 have also been used as solid catalysts for various condensation reactions due to the metallic nodes acting as Lewis acid sites.[40] Therefore, the deposition of metal nanoparticles onto MOFs is expected to provide multifunctional catalysts for one-pot cascade reactions. Generally, one-pot cascade reactions can improve efficiency compared to multistep synthetic processes.[41−43]

More recently, reduction of nitro compounds with formic acid or its salts as hydrogen source has drawn increasing attention.[44−47] Formic acid as hydrogen source has the advantages of safety, nontoxicity, and easiness of transportation and storage compared with hydrogen. In addition, formic acid is a major byproduct in the conversion of biomass into value-added products. Therefore, the reduction of nitro compounds to amines with formic acid is more efficient and environmentally friendly than the traditional ways. Up till now, several catalysts from the immobilization of noble metals Pt and Pd on various supports have been developed and have showed excellent performance in the hydrogenation of nitroarenes to anilines with formic acid as hydrogen source.[44−47] However, no report has been observed for the reduction of nitro compounds to amines with formic acid activated by Ru nanoparticles deposited onto MOFs. Furthermore, the immobilized Ru nanoparticles can be transformed to different active species under reduction and oxidation conditions due to the presence of Ru/Ru ionic pair, which is expected to be active both in reduction and oxidation reactions as a catalyst.

In this work, a 3.1% Ru/UiO-66 material was prepared from the deposition of Ru nanoparticles onto UiO-66. This material as a catalyst showed good performances for the reduction of nitro compounds with formic acid to amines and one-pot tandem aerobic oxidation/Knoevenagel condensation reaction of various alcohols with malononitrile.

Results and Discussion

Preparation of UiO-66–Ru

UiO-66 was employed as the host matrix to embed Ru nanoparticles due to its high physicochemical stability, resistance to water and various organic solvents, and large surface area. In the beginning, UiO-66 was synthesized according to the reported method.[26] Characterization results confirmed that the structure of the as-synthesized material is in agreement with that of UiO-66. We envisioned the deposition of Ru nanoparticles onto UiO-66 similar to the immobilization of other metal nanoparticles on UiO-66 or UiO-67.[26,27,48] A precursor named as UiO-66–RuCl3 was first synthesized from the reaction of RuCl3 with the hydroxy groups on the Zr node of UiO-66. The stability of RuCl3 and UiO-66–RuCl3 in solvent is of great importance to ensure the adsorption of RuCl3 onto UiO-66. Several solvents, including ethyl acetate, methanol, water, and 0.2 N aqueous solution of hydrochloric acid were tested in the preparation of UiO-66–RuCl3. Except ethyl acetate, the other solvents led to part decomposition of RuCl3 or UiO-66–RuCl3, as shown in Figure a. Therefore, the adsorption of RuCl3 onto UiO-66 was conducted in ethyl acetate. The formed UiO-66–RuCl3 was extracted with ethyl acetate in Soxhlet extractor to remove the free RuCl3 on the surface of the framework. Then, the UiO-66–RuCl3 was reduced with NaBH4 in water, from which Ru3+ was reduced to Ru0, leading to embedment of Ru nanoparticles onto UiO-66 to give UiO-66–Ru. Upon the formation of UiO-66–Ru, it was stable in all of the tested solvents (Figure b).

Figure 1

Stability test in several solvents: UiO-66–RuCl3 (a), UiO-66–Ru (b).

Characterization of Catalysts

The X-ray diffraction (XRD) patterns of UiO-66, UiO-66–RuCl3, and UiO-66–Ru are shown in Figure . The XRD patterns of UiO-66 are matched well with the simulated XRD patterns. Specific characteristic diffraction peaks of UiO-66 structure (Figure ) are same as those reported in the literature.[23,49] The XRD patterns of UiO-66–RuCl3 and UiO-66–Ru are consistent with those of UiO-66, manifesting maintenance of the frame structure of UiO-66 after embedment of Ru species. For all of the samples loaded with Ru species, no characteristic peaks of Ru are observed in their XRD patterns, which indicates that ruthenium species are well dispersed and/or the content of loaded Ru is too low to generate a large cluster to give diffraction signals.[37,50]

Figure 2

XRD patterns of the samples.

The transmission electron microscopy (TEM), scanning transmission electron microscopy (STEM), and high-resolution transmission electron microscopy (HRTEM) images of the samples are shown in Figure . Both the TEM and STEM images revealed that the neat UiO-66 exhibits octahedral morphology and the surfaces of the crystals are very smooth (Figure a,d). After the loading of Ru species, both UiO-66–RuCl3 and UiO-66–Ru remain in octahedral morphology (Figure b,c,e f). The surfaces of the crystals of UiO-66–RuCl3 are still smooth, indicating the uniform dispersion of RuCl3 (Figure b,e). However, some small nanoparticles are clearly observed on the surfaces of the crystals of UiO-66–Ru similar to the case of UiO-66–Pt in the literature[51] (Figure c,f), indicating the depositing of Ru species on the surface of UiO-66. The HRTEM images (Figure g–i) of the fresh and used samples clearly show Ru particles are well dispersed on the surface of the UiO-66. Figure j–l provides the mean Ru particle sizes of the samples of fresh, used UiO-66–Ru in reduction (red.) and used in oxidation (oxi.), which are 1.07, 1.27, and 1.56 nm, respectively. The Ru particle sizes in the used UiO-66–Ru are larger than of those in the fresh one, indicating Ru particle aggregation taking place during the catalytic run, especially in the case of oxidation reaction.

Figure 3

TEM images of the UiO-66, UiO-66–RuCl3, and UiO-66–Ru (a–c); STEM images of the UiO-66, UiO-66–RuCl3, and UiO-66–Ru (d–f), HRTEM images and particle size distributions of the fresh UiO-66–Ru (g, j), used UiO-66–Ru (red.) (h, k), and used UiO-66–Ru (oxi.) (i, l).

In addition, the energy-dispersive spectrometry (EDS) element mappings of UiO-66–Ru (Figure ) disclosed the coexistence of zirconium, ruthenium, carbon, and oxygen. The element mappings of UiO-66–Ru revealed that four elements are uniformly distributed throughout the octahedral crystal.

Figure 4

EDS-mapping analysis of UiO-66–Ru.

Figure shows the SEM images of the fresh and used samples of UiO-66–Ru. The images show that the morphologies of all of the samples are composed of octahedral crystals with different dimensions. No big difference was observed between the fresh and used samples either in nitrobenzene reduction or alcohol oxidation, which indicated that the framework of UiO-66–Ru was stable during the catalytic applications.

Figure 5

SEM images of UiO-66–Ru: fresh UiO-66–Ru (a), used UiO-66–Ru (red.) (b), and used UiO-66–Ru (oxi.) (c).

X-ray photoelectron spectroscopy (XPS) analysis was carried out to determine the ruthenium species and their relative contents in the fresh and used samples of UiO-66–Ru. The XPS wide scan spectra of the samples exhibit distinct C 1s, O 1s, Ru 3p and 3d, and Zr 3d peaks (Figure S1). The fitted Ru 3d spectra are shown in Figure . Deconvolution of the Ru 3d profile of the fresh UiO-66–Ru revealed three binding energy peaks located at 280.3, 281.2, and 282.9 eV (Table ), which could be ascribed to Ru0, RuO2, and RuO2·xH2O, respectively.[37,52] The Ru 3d profile was used to reveal the changes of the three species and their relative contents in the used samples. As shown in Table , both the binding energies of RuO2 and RuO2·xH2O of the used samples changed in reasonable intervals compared to those of the fresh one. The contents of the three species of the sample after use in the reduction of nitrobenzene are almost same as those of the fresh sample. For the sample after use in the oxidation of alcohol, the content of Ru0 decreased by 7%, whereas the content of RuO2 increased by 7%, indicating the transformation of some surface Ru0 into RuO2 under aerobic oxidation conditions.

Figure 6

Fitted XP spectra of the Ru 3d/C 1s and Zr 3d region: fresh UiO-66–Ru (a, d); UiO-66–Ru after use in the reduction of nitrobenzene (b, e); UiO-66–Ru after use in aerobic oxidation reaction of alcohol (c, f).

Table 1

Binding Energies of Ruthenium Species and Zirconium and Relative Contents of Ruthenium Species

sample	Ru0 3d BE (eV)/area %	RuO2 3d BE (eV)/area %	RuO2·xH2O 3d BE (eV)/area %	Zr 3d5/2 BE (eV)	Zr 3d3/2 BE (eV)
fresh UiO-66–Ru	280.3/49	281.2/33	282.8/18	182.1	184.4
used UiO-66–Ru (red.)	280.3/50	281.0/31	282.0/19	182.0	184.3
used UiO-66–Ru (oxi.)	280.3/42	281.4/39	282.8/19	182.3	184.5

In addition, the binding energies of Zr 3d5/2 and Zr 3d3/2 of UiO-66–Ru remained constant before and after the catalytic reactions (Table , Figure ), indicating the stability of UiO-66 as a support to embed ruthenium nanoparticles.

Nitrogen adsorption–desorption isotherms of the samples of UiO-66, UiO-66–RuCl3, and UiO-66–Ru were measured at 77 K, from which the specific surface areas and texture porosities of the samples were obtained. As shown in Figure , all of the samples exhibited typical type I isotherms, indicating the microporous nature of the samples.[53] The Brunauer–Emmett–Teller (BET) surface areas of UiO-66, UiO-66–RuCl3, and UiO-66–Ru are 1276, 789, and 876 m2/g, corresponding to pore volumes of 0.53, 0.34, and 0.39 cm3/g, respectively. The small BET surface areas and the pore volumes of UiO-66–RuCl3 and UiO-66–Ru compared to those of UiO-66 also indicated the deposition of Ru species onto the framework of UiO-66,[20,39,50] which is in accordance with the TEM characterization result. The bigger surface area and pore volume of UiO-66–Ru than those of UiO-66–RuCl3 is due to the reduction of RuCl3 to Ru0, leading to removal of Cl– from the cavities of UiO-66.

Figure 7

Representative N2 isotherms for UiO-66, UiO-66–Ru, and UiO-66–RuCl3.

Thermal gravimetric analyses (TGA) were carried out to study the thermal stability of UiO-66 and UiO-66–Ru. As shown in Figure , both UiO-66 and UiO-66–Ru show three weight-loss stages. The initial weight-loss stage occurring in the temperature range of 25–100 °C is due to desorption of physisorbed water, whereas the second weight-loss stage observed in the temperature range of 100–500 °C is related to the removal of dimethylformamide (DMF) and the dehydroxylation of the zirconium oxo-clusters.[53,54] The third weight-loss stage starting at 500 °C is attributed to the decomposition of UiO-66 as a result of the burning of organic-linker molecules in the framework. The results revealed that UiO-66–Ru is thermally stable below 500 °C.

Figure 8

TGA curves of UiO-66 and UiO-66–Ru.

Catalytic Hydrogenation of Nitro Compound

Amines are important intermediates or precursors in the manufacture of pharmaceuticals, agrochemicals, dyes, pigments, and polymers.[55,56] Generally, amines are manufactured by catalytic hydrogenation of nitro compounds. Recently, formic acid as hydrogen source has been successfully applied to the hydrogenation of nitro compounds to amines,[44−47,50] which is safe and environmentally friendly compared with the traditional ways. So we applied UiO-66–Ru to the reduction of nitro compounds to amines, with formic acid as hydrogen source. First, the catalytic performance of UiO-66–Ru was evaluated for the hydrogenation of nitrobenzene to aniline as a model reaction. Initially, the reaction was conducted in a mixture of H2O and 2-propanol (1:9) at room temperature, as described in the literature.[44−47] However, no reaction was observed in this case. With increasing temperature, the reaction took place and was improved gradually (Table , entries 4, 7, 8). When the temperature was increased to 150 °C, nitrobenzene was almost quantitatively converted to aniline at 1.0 mol % of catalyst loading in 3 h (Table , entry 4). As a control experiment, almost no aniline was obtained in the absence of UiO-66–Ru. Replacing UiO-66–Ru with UiO-66 or RuCl3 as catalyst led to very poor results (Table , entries 2, 3). Decreasing UiO-66–Ru loading from 1 to 0.5 mol % led to a big decrease of selectivity of aniline from 99 to 42.8%, although the conversion of nitrobenzene was maintained (Table , entries 4, 5). The reaction was also conducted in different solvents. Both the conversion and selectivity decreased sharply when the reaction was run in neat 2-propanol (Table , entry 9). Excessive increase of water in 2-propanol also led to a sharp decrease of conversion of nitrobenzene (Table , entry 10). The other proton polar solvents, ethanol and H2O–methanol mixture, afforded high conversion of nitrobenzene but low selectivity of aniline (Table , entries 11, 12). High conversion of nitrobenzene and selectivity of aniline were received in toluene but were still inferior to those in H2O/2-propanol (1:9). These results could be ascribed to the fact that the addition of a certain amount of water can greatly improve the conversion rate and selectivity, probably because water can promote the decomposition of formic acid into H+ and HCOO–.[46] Gas chromatography–mass spectrometry (GC–MS) analysis showed that the major byproduct was unconverted nitrosobenzene and hydroxylamine intermediate in other reaction solvents.[46] It is worth noting that the best results were received in H2O/2-propanol (1:9) mixture. The volume ratio of 1:9 in the H2O/2-propanol mixture is close to the azeotropic composition of 2-propanol and water, which allows easy recycling of the solvent in large practice. The loading amount of formic acid was also screened, and the optimal molar ratio of formic acid to substrate was 5:1 (Table S1).

Table 2

Reaction Condition Optimization for the Reduction of Nitrobenzene to Aniline with Formic Acida

entry	catalyst	catalyst loading (mol %)	solvent	temperature (°C)	conversion (%)b	selectivity (%)b
1		0	2-PrOH/H2O (9/1)	150	20.2	2.5
2c	UiO-66		2-PrOH/H2O (9/1)	150	28.5	2.8
3	RuCl3	1.0	2-PrOH/H2O (9/1)	150	28.2	1.8
4	UiO-66–Ru	1.0	2-PrOH/H2O (9/1)	150	100	>99
5	UiO-66–Ru	0.5	2-PrOH/H2O (9/1)	150	100	42.8
6	UiO-66–Ru	0.7	2-PrOH/H2O (9/1)	150	100	89.6
7	UiO-66–Ru	1.0	2-PrOH/H2O (9/1)	110	60.2	54.0
8	UiO-66–Ru	1.0	2-PrOH/H2O (9/1)	130	66.9	75.3
9	UiO-66–Ru	1.0	2-PrOH	150	62.8	39.4
10	UiO-66–Ru	1.0	2-PrOH/H2O (4/1)	150	72.4	34.1
11	UiO-66–Ru	1.0	EtOH	150	91.0	12.7
12	UiO-66–Ru	1.0	MeOH/H2O (4.5/1)	150	90.8	59.8
13	UiO-66–Ru	1.0	toluene	150	84.1	95.0

Reaction conditions: nitrobenzene 0.5 mmol, formic acid 2.5 mmol, solvent 3 mL, reaction temperature 150 °C, reaction time 3 h.

Conversation and selectivity were determined by GC on the basis of area %.

UiO-66 17.0 mg.

According to the above results in combination with the results reported in the literature, the optimal reaction conditions were obtained, which are nitrobenzene 0.5 mmol, UiO-66–Ru 1.0 mol %, molar ratio of formic acid to nitrobenzene of 5:1, 2-PrOH/H2O (9/1) 3 mL, and reaction temperature 150 °C.

A filtration test was carried out to evaluate the heterogeneity of the hydrogenation of nitro compounds to amines catalyzed by UiO-66–Ru, with nitrobenzene as the substrate. After 1 h of reaction, the UiO-66–Ru was removed by centrifugation and the liquid phase was stirred further under identical conditions for 2 h. As shown in Figure , the conversion remained constant throughout the process, which indicated that no active species leaking took place and UiO-66–Ru belongs to a heterogeneous catalyst.

Figure 9

Filtration test for the reduction of nitrobenzene over UiO-66–Ru.

To evaluate the versatility of this catalytic hydrogenation system, the reduction of various nitro compounds were explored under the optimized conditions. As shown in Table , various substituted nitrobenzenes with both electron-withdrawing and electron-donating groups were selectively converted to the corresponding anilines in excellent yields. The time to finish the reaction changed with the substituents, but no regularity was observed from the viewpoint of electronic effect (Table , entries 2, 5–7, 9). The position of the substituent related to the nitro group has some effects on the reaction. The substrate with an o-substituent gave lower selectivity and isolated yield than that with a para- or meta-substituent due to the steric hindrance (Table , entries 2–4). Large substituent generally gave low yield compared with small substituent of the same kind (Table , entries 5–7 and 9). Delightedly, no dehalogenation was detected in the cases of various halogenated nitrobenzenes as substrates, which generally occurred in the catalytic hydrogenation of halogenated nitro compounds. Similar to the catalyst from deposition of Pd nanoparticles onto UiO-66,[44] UiO-66–Ru showed high activity and selectivity in the reduction of substrates with ketone and carboxyl groups (Table , entries 10, 11). Besides, nitronaphthalene was almost quantitatively converted to aminonaphthalene (Table , entry 12). However, it failed in the reduction of p-nitrobenzaldehyde (Table , entry 13). Some oligomers rather than the target product were obtained, which might be due to the condensation of formyl group with amines at 150 °C. The catalyst also failed in the reduction of nitrocyclohexane, being a representative of aliphatic nitro compounds. An oily mixture of unidentified compounds was obtained in this case (Table , entry 14).

Table 3

Reduction of Nitro Compounds to Corresponding Aminesc

Conversation and selectivity were determined by GC and calculated on the basis of initial mmol of benzyl alcohol.

Isolated yield.

Reaction conditions: substrate (0.5 mmol), FA (2.5 mmol), UiO-66–Ru (1 mol %), 2-PrOH/H2O (9:1) (3 mL), reaction temperature 150 °C.

Catalytic Alcohol Oxidation/Knoevenagel Condensation Reaction

Tandem reactions, which combine two or more synthetic steps in one pot, are attractive and practical tools in organic synthesis. Compared to the corresponding multistep synthetic processes, they have the advantages of avoiding isolation of intermediates, reducing production of wastes, and easily available starting materials.[57,58] It is known that supported Ru nanoparticles are efficient catalysts for the aerobic oxidation of alcohols[59−62] and UiO-66 is active in the Knoevenagel condensation reaction due to the existence of the “Zr–O2– Lewis acid–base” pair in its framework.[63] Therefore, the UiO-66–Ru was also evaluated in the one-pot tandem aerobic alcohol oxidation/Knoevenagel condensation reaction. First, we examine the catalysis of UiO-66–Ru on the aerobic oxidation of alcohols with benzyl alcohol as a model substrate under the atmospheric oxygen with a catalyst loading of 3.6 mol % at 100 °C in toluene (Table S2). The reaction proceeded smoothly and completed almost quantitatively in 1 h. Replacing UiO-66–Ru with UiO-66 or UiO-66–RuCl3 led to very low conversion of benzyl alcohol, which indicated that the embedded Ru nanoparticles were the active species for the aerobic oxidation of benzyl alcohol.

Knoevenagel condensation reaction of benzylaldehyde with malononitrile as a model reaction was conducted over bare UiO-66 and UiO-66–Ru, respectively, to reveal whether the catalytic activity was from UiO-66 or Ru nanoparticles. The reaction was performed in 1.5 mL of toluene at 100 °C with 0.25 mmol benzaldehyde and 0.3 mmol malononitrile. As shown in Figure , bare UiO-66 exhibited higher activity than UiO-66–Ru, which indicated that the catalytic activity was from UiO-66 due to the existence of the Zr+–O2– Lewis acid–base pair in its framework.[63] The low activity of UiO-66–Ru compared to that of bare UiO-66 could be ascribed to the deposition of Ru nanoparticles on the Zr nodes of UiO-66, reducing the active sites exposed to the reactants.

Figure 10

Knoevenagel condensation of benzaldehyde with malononitrile over UiO-66 and UiO-66–Ru.

Thereafter, UiO-66–Ru was applied to the one-pot tandem aerobic oxidation/Knoevenagel condensation reaction of alcohols with malononitrile. Benzyl alcohol was converted to the condensation product benzylidene malononitrile in an isolated yield of 89.3% in a total reaction time of 6 h (Table , entry 1). Various para- or meta-substituted benzyl alcohols with either an electron-withdrawing or an electron-donating group were smoothly converted to the corresponding condensation products in high yields. Unexpectedly, random results were received in view of the electronic effect of the substituent (Table , entries 2–9), which is different with those of the NH2-MIL-101(Fe)-catalyzed photo-oxidation/Knoevenagel condensation at room temperature.[64] Herein, p-methoxy, p-methyl, and p-nitro substituted benzyl alcohols were transformed to the desired products with similar yields (Table , entries 2, 5, 9), although nitro group is a strong electron-withdrawing group and methoxy and methyl groups belong to good electron-donating groups. In addition, p-fluorobenzyl alcohol gave the highest yield among all halogenated benzyl alcohols (Table , entries 6–8). Steric hindrance has a large effect on the tandem reaction because ortho-substituted benzyl alcohols exhibited poor results compared to their para- or meta-substituted counterparts, and no desired condensation product was received in the case of α-phenethyl alcohol as substrate. These results indicated that the dehydration of the addition product from aldehyde and malononitrile to benzylidene malononitrile was the rate-determining step in the absence of steric hindrance; in the presence of steric hindrance, the addition of malononitrile to aldehyde became the rate-determining step.

Table 4

Tandem Aerobic Oxidation/Knoevenagel Condensation of Alcohols with Malononitrile Catalyzed by UiO-66–Rud

Conversation and selectivity were determined by GC on the basis of area %.

Isolated yield, calculated on the basis of initial mmol of benzyl alcohol.

Independent oxidation reaction, reaction temperature 120 °C, O2 0.4 MPa.

Reaction conditions: benzyl alcohol (0.25 mmol), UiO-66–Ru (3.6 mol % Ru), toluene (1.5 mL), O2 1 atm, malononitrile (0.3 mmol), reaction temperature 100 °C.

Poor results were obtained in the cases of primary heterocyclic alcohols as substrates. Although 2-thiophenemethanol and 2-furanmethanol were smoothly oxidized to their corresponding aldehydes, low yields (41.3 and 56.0%) of the final condensation products were obtained due to the formation of some unidentified impurities during the condensation process (Table , entries 11, 12). This catalytic system was also less effective in the tandem reaction of trans-cinnamyl alcohol due to the low conversion of the intermediate cinnamaldehyde to the condensation product, in which the isolated yield of the desired product was only 51.2% (Table , entry 13). Unfortunately, no Knoevenagel condensation product was detected in the cases of 1-hexanol and cyclohexanol as substrates, which are the representatives of aliphatic alcohols as substrates, presumably due to the inactivity of the intermediate aldehyde and ketone in the nucleophilic addition by the malononitrile.

The heterogeneity of the oxidation reaction was examined in the oxidation of benzyl alcohol. After 18 min of reaction, the UiO-66–Ru was removed by centrifugation and the liquid phase was stirred further under identical conditions for 42 min. As shown in Figure a, the conversion remained constant throughout the process. Similarly, the heterogeneity of the Knoevenagel condensation was confirmed in the condensation of benzaldehyde with malononitrile. After 2 h of reaction, the UiO-66–Ru was removed by centrifugation and the liquid phase was stirred further under identical conditions for 3 h. As shown in Figure b, the conversion did not change with time. Both filter tests indicated that no active species leaking took place during the reactions and UiO-66–Ru belongs to a heterogeneous catalyst.

Figure 11

Filtration test for the tandem reactions over UiO-66–Ru: oxidation reaction (a), Knoevenagel condensation reaction (b).

Further experiments were performed to evaluate the reusability of UiO-66–Ru in the nitrobenzene reduction. After reaction, UiO-66–Ru was recovered by centrifugation and subjected to the next run. As shown in Figure , the conversion and selectivity were well maintained in the consecutive runs, which manifested the good recyclability of UiO-66–Ru in the nitrobenzene reduction. The SEM images disclosed the framework of UiO-66–Ru did not change obviously after catalytic runs. The XPS characterization revealed that the relative contents of Ru0, RuO2, and RuO2·xH2O in the UiO-66–Ru did not change after application in the reduction of nitrobenzene, indicating the stability of the active species in the reaction. Besides, the molar ratio of zirconium to ruthenium in the sample of UiO-66–Ru disclosed by XPS and inductively coupled plasma (ICP) almost remained constant before and after the reaction (Tables S3 and S4), demonstrating little leaching of ruthenium occurring in the catalytic run. On the basis of the above results, it can be concluded that UiO-66–Ru is a stable and recyclable catalyst in the hydrogenation of nitrobenzene to aniline with formic acid.

Figure 12

Recyclability of the UiO-66–Ru catalyst for the reduction of nitrobenzene.

Unfortunately, UiO-66–Ru showed very poor reusability in the tandem aerobic oxidation/Knoevenagel condensation of alcohols with malononitrile. The deactivation of UiO-66–Ru occurred in the oxidation step. As shown in Figure S2, the conversion of benzyl alcohol was only 56% in the second run of the aerobic oxidation of benzyl alcohol. As shown in the SEM images (Figure ), the morphology of the sample after use in alcohol oxidation did not change obviously compared to that of the fresh one, which indicated that the deactivation of UiO-66–Ru was not due to the collapse of the framework of UiO-66. ICP analysis revealed that no big leaching occurred in the reaction (Table S4) and the molar ratio of Zr to Ru disclosed by XPS changed slightly after catalytic run (Table S3). These results indicated that the deactivation of UiO-66–Ru in the aerobic oxidation of alcohol was not caused by the leaking of Ru in the reaction. As shown in Figure , obvious aggregation of Ru particles took place during the aerobic oxidation of alcohols, which could be ascribed to the main reason causing the deactivation of the catalyst in the tandem aerobic oxidation/Knoevenagel condensation of alcohols with malononitrile.

Conclusions

In summary, Ru nanoparticles were successfully embedded onto the framework of UiO-66 with the postsynthesis method to afford UiO-66–Ru. The characterization results revealed that most parts of the Ru nanoparticles were deposited on the inner surfaces of the cavities of UiO-66. UiO-66–Ru as a catalyst showed high activity and selectivity in the reduction of nitroarenes to amines at elevated temperature and can be recycled without obvious deactivation of its catalysis. UiO-66–Ru was also active in the tandem aerobic oxidation/Knoevenagel condensation of alcohols with malononitrile; however, it cannot be recycled in this case due to the transformation of surface Ru0 to RuO2 of Ru nanoparticles under the aerobic oxidation conditions. The surface Ru0 of Ru nanoparticles were regarded as the active species for both the catalytic reduction of nitro compounds and the aerobic oxidation of alcohols. The activity for the Knoevenagel condensation reaction was from the existence of the Zr–O2– Lewis acid–base pair in the framework of UiO-66.

Experimental Section

Materials

Terephthalic acid and ZrCl4 were obtained from J&K Scientific. RuCl3·3H2O was supplied by Xi’an Catalyst Chemical Co., Ltd. The nitro compounds and alcohols were obtained from Alfa Aesar China (Tianjin) Co., Ltd. All reagents were of analytical grade and used without further purification.

Synthesis of UiO-66

UiO-66 was prepared in accordance with the methods described previously.[26] Briefly, in a 3000 mL round-bottom flask, ZrCl4 (3.72 g, 16 mmol) was dissolved under stirring in a mixture of DMF (1000 mL) and glacial acetic acid (288 g, 274.6 mL, 4.8 mol). In another 2000 mL round-bottom flask, terephthalic acid (2.66 g, 16 mmol) was dissolved completely in DMF (1000 mL). The terephthalic acid solution was slowly added into the ZrCl4 solution, and the resulting mixture was stirred to a homogeneous solution. The solution was separated into 100 vials. The vials were closed and heated at 120 °C for 24 h, then cooled to room temperature. The reaction mixture in the vials was gathered, and the white powder UiO-66 product was collected through centrifugation and rinsed with methanol to get a white solid (3.0 g, 70% yield). The UiO-66 product was activated at 150 °C under high vacuum for 12 h, stored at room temperature.

Synthesis of UiO-66–RuCl3

In a 100 mL round-bottom flask, RuCl3·3H2O (700 mg, 2.68 mmol) was dissolved in ethyl acetate (30 mL), followed by the addition of the activated UiO-66 (1.5 g). The resulting mixture was heated to 50 °C and maintained at this temperature for 48 h in an oil bath. The reaction mixture was cooled to room temperature, and the solid was collected through centrifugation. The unreacted RuCl3 was removed by extraction with ethyl acetate in a Soxhlet extractor for 24 h. The extracted solid was dried under air to give UiO-66–RuCl3.

Synthesis of UiO-66–Ru

The UiO-66–Ru was synthesized referencing a previous method.[65] NaBH4 (130 mg) was added to 70 mL of deionized water in a round-bottom flask, after which 1.0 g of UiO-66–RuCl3 was added. The pH was controlled in the range of 6–8 throughout the addition of NaBH4. The obtained mixture was stirred for 30 min, then was subject to filtration. The collected solid was washed thoroughly with deionized water to give black powder of UiO-66–Ru. Finally, the UiO-66–Ru was activated at 150 °C for 12 h under high vacuum. The content of Ru in UiO-66–Ru is 3.1 wt %, determined by ICP-optical emission spectrometry (OES).

Catalysts Characterization

Powder X-ray diffraction patterns of the samples were recorded on a D8 Advance X-ray diffractometer (Bruker Corporation, Germany) with Cu Kα radiation (λ = 0.15406 nm) and scanning speed of 12°/min from 5 to 50° at ambient temperature. Transmission electron microscopy (TEM) was carried out on a FEI Tecnai G2 F20 microscope operating at an accelerating voltage of 200 kV. The samples were prepared by dispersing the catalyst powder in ethanol under ultrasound and then supported on a carbon film of copper grid. X-ray photoelectron spectroscopy (XPS) was conducted on an ESCALAB 250Xi system (Thermo Fisher Scientific) with a standard Al Kα X-ray source. The contents of Ru were analyzed by inductively coupled plasma optical emission spectrometry (ICP-OES, PerkinElmer). The specific surface areas of the catalyst samples were measured by N2 adsorption method at 77 K using ASAP2020M+C instrument (Micromeritics Company). Prior to the measurement, the samples were degassed at 423 K for 12 h under high vacuum. Thermogravimetric Analysis (TGA) was evaluated using an SDT Q-600 thermogravimetric analyzer from 30 to 800 °C with a temperature rate of 20 °C/min in air. SEM images were taken with a Nova Nano SEM450 instrument.

1H NMR spectra were recorded using a Bruker 400 DRX spectrometer in CDCl3 or DMSO-d6 with tetramethylsilane as an internal standard.

Reduction of Nitro Compounds with Formic Acid

In a typical process, substrate nitro compound (0.5 mmol), catalyst (16.3 mg, 1 mol % Ru), formic acid (2.5 mmol), and 3 mL of isopropanol/water (9:1) were successively charged into an autoclave, after which the autoclave was sealed and purged with nitrogen. Then, the mixture was stirred at 150 °C until the end of the reaction. The reaction progress was monitored by a gas chromatograph (Shandong Lunan Ruihong SP-7800A) equipped with a SE-54 column of 30 m. Because of very weak response of formic acid on GC, the reaction mixture was directly subjected to GC analysis without needing pretreatment to remove the formic acid. The yield of the product was obtained by column chromatography separation.

General Procedure for One-Pot Tandem Oxidation/Knoevenagel Condensation

Into a round-bottom flask, alcohol (0.5 mmol), catalyst (3.6 mol % Ru), and toluene (1.5 mL) were added. The flask was purged with O2 from a top balloon, and then the mixture was stirred at 100 °C under 1 atm of oxygen to start the aerobic oxidation of alcohol. The oxidation was monitored by GC. After completion of the reaction, the reaction mixture was cooled down to room temperature. Then, malononitrile (0.55 mmol) was added directly into the above reaction mixture and the resulting mixture was stirred at 100 °C further to finish the reaction. After reaction, the solvent was removed by rotary evaporation under vacuum. The crude product was subject to silica gel column chromatography eluted with petroleum ether–EtOAc (5:1) to afford pure product.