Heterogeneous Bimetallic Cu–Ni Nanoparticle-Supported Catalysts in the Selective Oxidation of Benzyl Alcohol to Benzaldehyde

Three bimetallic Cu–Ni nanoparticle-supported catalysts were synthesized by co-immobilization followed by H2 reduction. A chromium(III) terephthalate metal organic framework (MIL-101), titanium dioxide (TiO2), and carbon (C) with different properties (acidity and Brunauer–Emmett–Teller surface area) were selected as supports for studying the effect of the support nature on the catalytic activity and selectivity in the oxidation of benzyl alcohol. The physicochemical properties of the Cu–Ni-supported catalysts were characterized by XRD, NH3-TPD, nitrogen adsorption/desorption, TEM, EDS, XPS, and ICP-OES. Bimetallic Cu–Ni nanoparticles were highly dispersed on the support. The catalytic activities of CuNi/MIL-101, CuNi/TiO2, and CuNi/C were tested in the selective oxidation of benzyl alcohol to benzaldehyde in the presence of molecular oxygen under mild reaction conditions. The highest benzaldehyde yields were achieved with CuNi/TiO2, CuNi/MIL-101, and CuNi/C catalysts at 100 °C within 4 h under 5, 3, and 3 bar of O2, respectively. The bimetallic Cu–Ni-supported catalysts possessed two types of catalytic active sites: acid sites and bimetallic Cu–Ni nanoparticles. The CuNi/MIL-101 catalyst possessed a high number of acid sites and exhibited high yield during selective benzyl alcohol oxidation to benzaldehyde. Importantly, the catalysts exhibited a high functional group (electron-donating and electron-withdrawing groups) tolerance. Cu–Ni-supported catalysts with an Cu:Ni mole ratio of 1:1 exhibited the highest yield of 47% for the selective oxidation of benzyl alcohol to benzaldehyde. Reusability and leaching experiment results exhibited that CuNi/MIL-101 showed better stability than CuNi/TiO2 and CuNi/C catalysts due to the large porous cavities of MIL-101 support; these cavities can be used to trap bimetallic Cu–Ni nanoparticles and inhibit nanoparticle leaching.


Introduction
Benzaldehyde is a valuable precursor for the production of perfumes, pharmaceuticals, dyestuffs, and agrochemicals [1]. Conventionally, benzaldehyde is synthesized by the selective oxidation of benzyl alcohol in the presence of inorganic oxidants such as KMnO 4 , CrO 3 , MnO 2 , and Br 2 [2]. However, these reagents may produce high amounts of waste and hazardous/toxic compounds during selective oxidation, which may result in intensive environmental problems [3]. Therefore, developing green catalytic processes under mild conditions is necessary to produce clean benzaldehyde by benzyl alcohol oxidation. reached 0.49, 0.45, 0.014, 0.0078, 0.11, and 0.079 cm 3 /g for MIL-101, CuNi/MIL-101, TiO2, CuNi/TiO2, C, and CuNi/C, respectively. The corresponding mean pore diameters were 2.8, 2.5, 4.3, 4.7, 4.6, and 5.2 nm for MIL-101, CuNi/MIL-101, TiO2, CuNi/TiO2, C, and CuNi/C, respectively. Ammonia temperature programmed desorption (NH3-TPD) was used to measure the acidic properties of the MIL-101, TiO2, and C supports, and the results are shown in Figure 2. Three peaks were observed on the spectra of MIL-101 in the range of 40 °C-600 °C, whereas one broad peak was observed on the spectra of TiO2 and C. The low-temperature desorption peak centered at approximately 270 °C was attributed to adsorbed NH3 on Lewis acid sites, whereas the high-temperature desorption peak at approximately 340 °C was assigned to Brønsted acid sites [21][22][23][24][25]. TiO2 contained only Brønsted acid sites, C mainly possessed Lewis acid sites, and MIL-101 preliminarily included Lewis and Brønsted acid sites. For MIL-101, the peak at 443.8 °C was attributed to its structural decomposition. MIL-101 contained more adsorption sites for NH3 than TiO2 and C. The numbers of Lewis and Brønsted acid centers located at 264. 6 °C and 361.7 °C for MIL-101 reached 0.705 and 3.308 mmol/g, respectively. However, extremely low peaks at 336.1 °C and 275.6 °C were noted in the NH3-TPD curves of TiO2 and C, with the amounts of acid sites reaching 0.062 and 0.039 mmol/g, respectively. CuNi/MIL-101, CuNi/TiO2, and CuNi/C catalysts were prepared by a co-immobilization method, followed by treatment with a stream of H2 (12 mL/min) for 2 h at 50 °C. Figure 3 shows the x-ray diffraction (XRD) patterns of MIL-101, CuNi/MIL-101, TiO2, CuNi/TiO2, C, and CuNi/C. MIL-101 and CuNi/MIL-101 presented similar XRD patterns (Figure 3a), thereby proving the preservation of the crystal structures of MIL-101 after loading of the Cu-Ni nanoparticles [26,27]. Based on the XRD patterns, only the rutile phase was observed in TiO2 ( Figure 3b) [28,29]. The main diffraction peak of TiO2 was revealed at 2θ = 27.7°, which represents the rutile (110) phase [28]. The XRD patterns of TiO2 showed no peak corresponding to the anatase (101) phase at 2θ = 25.3° [29]. As shown in the XRD patterns of TiO2, other diffraction peaks appeared at 2θ = 36.4°, 39.5°, 41.6°, 44.4°, 54.7°, 57.0°, 63.1°, 64.3°, 69.2°, 70.1°, and 76.9°, which were consistent with the rutile phase of TiO2. By comparing the XRD pattern of CuNi/TiO2 with that of TiO2, the CuO and NiO nanoparticle loading introduced no significant changes in the TiO2 structure ( Figure 3b). The lattice plane of graphite (JCPDS, File No. 74-2330) was noted in Figure 3c. The characteristic peak at 2θ = 24.8° in the XRD patterns was associated with the (002) of carbon support (Figure 3c) [30][31][32]. Furthermore, characteristic CuO and NiO peaks were absent in the XRD patterns of CuNi/TiO2 and CuNi/C due to the low amounts of Cu and Ni in the catalysts [33,34].       We conducted XPS measurements to investigate the chemical state of Cu and Ni in the CuNi/MIL-101, CuNi/TiO 2 , and CuNi/C catalysts. Figure 4 shows the Cu 2p and Ni 2p XPS spectra of the as-prepared CuNi/MIL-101, CuNi/TiO 2 , and CuNi/C catalysts. For all of the catalysts, two peaks at 932.5 and 952.5 eV were attributed to Cu 2p 3/2 and Cu 2p 1/2 , respectively, thereby suggesting the presence of Cu 2+ in the catalysts [35]. Furthermore, two satellite peaks of Cu 2p 3/2 and Cu 2p 1/2 were detected, thereby further confirming the existence of Cu 2+ [36,37]. The oxidized Ni of Ni 2+ 2p 3/2 and Ni 2+ 2p 1/2 were detected at the binding energies of 856.6 and 874.0 eV, respectively, in the Ni 2p spectra of CuNi/MIL-101, CuNi/TiO 2 , and CuNi/C possibly due to oxidation of the catalysts exposed to air [11,38]. The peaks at 861.8 and 880.2 eV were assigned to satellites in all the catalysts [39].  We conducted XPS measurements to investigate the chemical state of Cu and Ni in the CuNi/MIL-101, CuNi/TiO2, and CuNi/C catalysts. Figure 4 shows the Cu 2p and Ni 2p XPS spectra of the as-prepared CuNi/MIL-101, CuNi/TiO2, and CuNi/C catalysts. For all of the catalysts, two the presence of Cu 2+ in the catalysts [35]. Furthermore, two satellite peaks of Cu 2p3/2 and Cu 2p1/2 were detected, thereby further confirming the existence of Cu 2+ [36,37]. The oxidized Ni of Ni 2+ 2p3/2 and Ni 2+ 2p1/2 were detected at the binding energies of 856.6 and 874.0 eV, respectively, in the Ni 2p spectra of CuNi/MIL-101, CuNi/TiO2, and CuNi/C possibly due to oxidation of the catalysts exposed to air [11,38]. The peaks at 861.8 and 880.2 eV were assigned to satellites in all the catalysts [39].  The morphology of the as-synthesized CuNi/MIL-101 catalyst obtained via co-immobilization followed by H2 reduction was determined by transmission electron microscope (TEM) (Figures 5ac). Line scanning analysis of CuNi/MIL-101 was also determined via energy dispersive x-ray spectrometry (EDS), as observed in Figure 5d. The CuNi/MIL-101 catalyst exhibited a micro sized but well-defined cuboidal morphology. EDS line scanning profiles revealed that Cu and Ni were homogeneously distributed throughout the entire MIL-101, thereby indicating the formation of a CuNi alloy structure. The morphology of the CuNi/TiO2 and CuNi/C catalysts was characterized by The morphology of the as-synthesized CuNi/MIL-101 catalyst obtained via co-immobilization followed by H 2 reduction was determined by transmission electron microscope (TEM) (Figure 5a-c). Line scanning analysis of CuNi/MIL-101 was also determined via energy dispersive x-ray spectrometry (EDS), as observed in Figure 5d. The CuNi/MIL-101 catalyst exhibited a micro sized but well-defined cuboidal morphology. EDS line scanning profiles revealed that Cu and Ni were homogeneously distributed throughout the entire MIL-101, thereby indicating the formation of a CuNi alloy structure. The morphology of the CuNi/TiO 2 and CuNi/C catalysts was characterized by TEM. Figure 6 displays the TEM images of CuNi/TiO 2 and CuNi/C. Figure 6 indicates that the CuNi/TiO 2 particles aggregated, but exhibited remarkable rutile structures and grain interfaces. The TEM image of CuNi/C showed that the catalyst exhibited the morphology of spherical particles with a mean diameter size of 10-30 nm ( Figure 6). TEM. Figure 6 displays the TEM images of CuNi/TiO2 and CuNi/C. Figure 6 indicates that the CuNi/TiO2 particles aggregated, but exhibited remarkable rutile structures and grain interfaces. The TEM image of CuNi/C showed that the catalyst exhibited the morphology of spherical particles with a mean diameter size of 10-30 nm ( Figure 6).

Benzyl Alcohol Oxidation Using Cu-Ni Bimetallic Catalysts
The catalytic activities of the as-prepared Cu-Ni-supported catalysts for benzyl alcohol oxidation were determined with molecular oxygen as a green oxidant. To avoid producing benzoic acid and benzyl benzoate, we conducted a series of reactions to obtain the optimal reaction conditions to produce benzaldehyde. First, the effects of reaction time on benzyl alcohol conversion, benzaldehyde selectivity, and yield of the CuNi/TiO2, CuNi/MIL-101, and CuNi/C catalysts were

Benzyl Alcohol Oxidation Using Cu-Ni Bimetallic Catalysts
The catalytic activities of the as-prepared Cu-Ni-supported catalysts for benzyl alcohol oxidation were determined with molecular oxygen as a green oxidant. To avoid producing benzoic acid and benzyl benzoate, we conducted a series of reactions to obtain the optimal reaction conditions to produce benzaldehyde. First, the effects of reaction time on benzyl alcohol conversion, benzaldehyde selectivity, and yield of the CuNi/TiO 2 , CuNi/MIL-101, and CuNi/C catalysts were studied. The oxidation reaction was kept at 100 • C under 5 bar of O 2 , and the results are presented in Figure 7. For all of the catalysts, the conversion of benzyl alcohol increased with the prolonged reaction time. However, the selectivity of benzaldehyde decreased at a reaction time above 2 h. The yields of benzaldehyde did show an increase after the reaction time reached up to 4 h for the CuNi/TiO 2 , CuNi/MIL-101, and CuNi/C catalysts, therefore, the optical reaction time was set at 4 h for all of the catalysts. Maximum yields of 32.7%, 41.5%, and 34.8% were obtained within 4 h for CuNi/TiO 2 , CuNi/MIL-101, and CuNi/C, respectively. CuNi/MIL-101 presented a higher benzaldehyde yield than CuNi/TiO 2 and CuNi/C. The acidity of the support substantially influenced the product selectivity. According to the NH 3 -TPD measurement, the values of the acid sites for MIL-101, TiO 2 , and C reached 4.013, 0.062, and 0.039 mmol/g, respectively. The highest yield was observed in CuNi/MIL-101, which contained more acid sites in benzyl alcohol oxidation to benzaldehyde, in accordance with previously reported results [11,14]. Benzaldehyde selectivity was lower in the base N,N-dimethylformamide (DMF) than in the THF and 1,4-dioxane solvents over the AuNi/MIL-101-1 catalyst [11]. The acid sites of the supports also played a crucial role in benzyl alcohol oxidation. A blank experiment showed that benzyl alcohol conversions reached 25.8%, 13.6%, and 12.9% at 100 • C under 5 bar of O 2 within 4 h over the MIL-101, TiO 2 , and C supports, respectively ( Table 1, entries 1-3). MIL-101 with more acid sites showed higher catalytic activity than TiO 2 and C. Notably, the conversions of benzyl alcohol amounted to 54.9%, 62.8%, and 64.3% at 100 • C under 5 bar of O 2 within 4 h for CuNi/TiO 2 , CuNi/MIL-101, and CuNi/C, respectively. Higher benzyl alcohol conversions were observed for Cu-Ni supported catalysts than the supports (MIL-101, TiO 2 , and C) without Cu-Ni particle loading. These results confirm that bimetallic Cu-Ni-supported catalysts possess two types of catalytic active sites: Lewis/Brønsted acid sites and bimetallic Cu-Ni nanoparticles.
Second, conditions such as the type of solvent, reaction pressure, and temperature were optimized in the oxidation of benzyl alcohol in the presence of Cu-Ni bimetallic catalysts under magnetic stirring (500 rpm). Table 1 summarizes the results. Selective oxidation of benzyl alcohol in organic solvents such as dioxane, DMF, acetonitrile, and THF were investigated by CuNi/TiO 2 , CuNi/MIL-101, and CuNi/C at 100 • C for 4 h under 5 bar, 3 bar, and 5 bar of O 2 . The best catalytic performance was observed in THF ( Table 2, entries 4-7, 13-16, and [22][23][24][25]. Moreover, in the optimization of the reaction temperature of the CuNi/MIL-101, CuNi/TiO 2 , and CuNi/C catalysts for the oxidation of benzyl alcohol, benzyl alcohol conversion continuously increased with increasing reaction temperature from 80 • C to 120 • C, along with a decreased benzaldehyde selectivity as oxidation is usually endothermic; thus, deep oxidation of benzaldehyde to benzoic acid easily occurred at high reaction temperatures (Table 2, entries 7-9, 16-18, and 26-28) [40]. The highest yield was obtained at 100 • C for all catalysts (Table 2, entries 7-9, 16-18, and 26-28). Oxidation of benzyl alcohol was conducted with different reaction pressures (1-7 bar) to determine the proper reaction pressure for Cu-Ni bimetallic catalysts (Table 2, entries 7, 10-12, 16, 19-20, 25, and 28-30). Increasing the reaction pressure can increase the conversion of benzyl alcohol. However, the selectivity to benzaldehyde decreased due to the appearance of over-oxidized products [41,42]. The highest yields of benzaldehyde were achieved in the CuNi/TiO 2 , CuNi/MIL-101, and CuNi/C catalysts at 100 • C under 5, 3, and 3 bar of O 2 , respectively. Further increasing the reaction pressure to 7 bar of O 2 caused no substantial increase in benzyl alcohol conversion, but significantly decreased the selectivity to benzaldehyde for the CuNi/MIL-101 and CuNi/C catalysts (Table 2,   Second, conditions such as the type of solvent, reaction pressure, and temperature were optimized in the oxidation of benzyl alcohol in the presence of Cu-Ni bimetallic catalysts under magnetic stirring (500 rpm). Table 1 summarizes the results. Selective oxidation of benzyl alcohol in organic solvents such as dioxane, DMF, acetonitrile, and THF were investigated by CuNi/TiO2,  To examine the scope of the selective oxidation reaction, we extended the catalytic protocol to other alcohols including 4-methylbenzyl alcohol, 4-methoxybenzyl alcohol, 4-chlorobenzyl alcohol, and n-butanol, the results of which are listed in Table 3. Aromatic alcohols including those bearing functional groups such as alkoxy, alkyl, and chloro were converted to the corresponding aldehydes, and afforded good catalytic activity in the selective oxidation reaction (Table 3, (Table 3, entries 1-3). The selectivities of the aldehydes reached 24.4%, 28.3%, and 25.2% when 4-methylbenzyl alcohol, 4-methoxybenzyl alcohol, and 4-chlorobenzyl alcohol were used, respectively (Table 3, entries 1-3). The experimental data showed that the functional groups including electron-donating (alkoxy and alkyl) and electron-withdrawing (chloro) groups on the phenyl ring resulted in negligible effects on the conversions and selectivities ( Table 3, entries 1-3). n-Butanol as the linear alcohol was also successfully oxidized to the corresponding compounds. The conversion and selectivity of n-butyl aldehyde reached 74.9% and 26.2%, respectively ( Table 3, entry 4). The experimental data showed that the functional groups including electron-donating (alkoxy and alkyl) and electron-withdrawing (chloro) groups on the phenyl ring resulted in negligible effects on the conversions and selectivities (Table 3, entries 1-3). n-Butanol as the linear alcohol was also successfully oxidized to the corresponding compounds. The conversion and selectivity of n-butyl aldehyde reached 74.9% and 26.2%, respectively (Table 3, entry 4). The catalytic activity of Cu-Ni-supported catalysts with different Cu:Ni mole ratios for the selective oxidation of benzyl alcohol were also investigated at 100 °C within 4 h under 3 bar of O2 in THF ( Table 4). The results showed that the benzyl alcohol conversion, benzaldehyde selectivity, and yield depended strongly on the mole ratio of Cu:Ni for the oxidation of benzyl alcohol. The yields of benzaldehyde amounted to 33.6%, 42.6%, 47.0%, 36.0%, and 30.  The experimental data showed that the functional groups including electron-donating (alkoxy and alkyl) and electron-withdrawing (chloro) groups on the phenyl ring resulted in negligible effects on the conversions and selectivities (Table 3, entries 1-3). n-Butanol as the linear alcohol was also successfully oxidized to the corresponding compounds. The conversion and selectivity of n-butyl aldehyde reached 74.9% and 26.2%, respectively (Table 3, entry 4). The catalytic activity of Cu-Ni-supported catalysts with different Cu:Ni mole ratios for the selective oxidation of benzyl alcohol were also investigated at 100 °C within 4 h under 3 bar of O2 in THF ( Table 4). The results showed that the benzyl alcohol conversion, benzaldehyde selectivity, and yield depended strongly on the mole ratio of Cu:Ni for the oxidation of benzyl alcohol. The yields of benzaldehyde amounted to 33.6%, 42.6%, 47.0%, 36.0%, and 30.  alkyl) and electron-withdrawing (chloro) groups on the phenyl ring resulted in negligible effects on the conversions and selectivities (Table 3, entries 1-3). n-Butanol as the linear alcohol was also successfully oxidized to the corresponding compounds. The conversion and selectivity of n-butyl aldehyde reached 74.9% and 26.2%, respectively (Table 3, entry 4). The catalytic activity of Cu-Ni-supported catalysts with different Cu:Ni mole ratios for the selective oxidation of benzyl alcohol were also investigated at 100 °C within 4 h under 3 bar of O2 in THF ( Table 4). The results showed that the benzyl alcohol conversion, benzaldehyde selectivity, and yield depended strongly on the mole ratio of Cu:Ni for the oxidation of benzyl alcohol. The yields of benzaldehyde amounted to 33.6%, 42.6%, 47.0%, 36.0%, and 30.  alkyl) and electron-withdrawing (chloro) groups on the phenyl ring resulted in negligible effects on the conversions and selectivities (Table 3, entries 1-3). n-Butanol as the linear alcohol was also successfully oxidized to the corresponding compounds. The conversion and selectivity of n-butyl aldehyde reached 74.9% and 26.2%, respectively (Table 3, entry 4). The catalytic activity of Cu-Ni-supported catalysts with different Cu:Ni mole ratios for the selective oxidation of benzyl alcohol were also investigated at 100 °C within 4 h under 3 bar of O2 in THF ( Table 4). The results showed that the benzyl alcohol conversion, benzaldehyde selectivity, and yield depended strongly on the mole ratio of Cu:Ni for the oxidation of benzyl alcohol. The yields of benzaldehyde amounted to 33.6%, 42.6%, 47.0%, 36.0%, and 30.  successfully oxidized to the corresponding compounds. The conversion and selectivity of n-butyl aldehyde reached 74.9% and 26.2%, respectively (Table 3, entry 4). The catalytic activity of Cu-Ni-supported catalysts with different Cu:Ni mole ratios for the selective oxidation of benzyl alcohol were also investigated at 100 °C within 4 h under 3 bar of O2 in THF ( Table 4). The results showed that the benzyl alcohol conversion, benzaldehyde selectivity, and yield depended strongly on the mole ratio of Cu:Ni for the oxidation of benzyl alcohol. The yields of benzaldehyde amounted to 33.6%, 42.6%, 47.0%, 36.0%, and 30.  successfully oxidized to the corresponding compounds. The conversion and selectivity of n-butyl aldehyde reached 74.9% and 26.2%, respectively (Table 3, entry 4). The catalytic activity of Cu-Ni-supported catalysts with different Cu:Ni mole ratios for the selective oxidation of benzyl alcohol were also investigated at 100 °C within 4 h under 3 bar of O2 in THF ( Table 4). The results showed that the benzyl alcohol conversion, benzaldehyde selectivity, and yield depended strongly on the mole ratio of Cu:Ni for the oxidation of benzyl alcohol. The yields of benzaldehyde amounted to 33.6%, 42.6%, 47.0%, 36.0%, and 30.   The catalytic activity of Cu-Ni-supported catalysts with different Cu:Ni mole ratios for the selective oxidation of benzyl alcohol were also investigated at 100 °C within 4 h under 3 bar of O2 in THF ( Table 4). The results showed that the benzyl alcohol conversion, benzaldehyde selectivity, and yield depended strongly on the mole ratio of Cu:Ni for the oxidation of benzyl alcohol. The yields of benzaldehyde amounted to 33.6%, 42.6%, 47.0%, 36.0%, and 30.   The catalytic activity of Cu-Ni-supported catalysts with different Cu:Ni mole ratios for the selective oxidation of benzyl alcohol were also investigated at 100 °C within 4 h under 3 bar of O2 in THF ( Table 4). The results showed that the benzyl alcohol conversion, benzaldehyde selectivity, and yield depended strongly on the mole ratio of Cu:Ni for the oxidation of benzyl alcohol. The yields of benzaldehyde amounted to 33.6%, 42.6%, 47.0%, 36.0%, and 30.   The reusability of catalysts in the oxidation reaction of benzyl alcohol was tested at 100 • C under 5 bar of O 2 (Figure 8). A slight decrease in catalytic activity occurred after the fourth run, thereby demonstrating the good stability of CuNi/MIL-101 under the reaction conditions (Figure 8a). However, the catalytic activity was considerably affected by the reuse of the CuNi/TiO 2 and CuNi/C catalysts (Figure 8b,c). Benzyl alcohol conversion decreased from 54.9% to 23.7% in four consecutive runs over the CuNi/TiO 2 catalyst, whereas the selectivity toward benzaldehyde approximated 99% in the second, third, and fourth cycles (Figure 8b). A significant decrease was also observed in the conversion of benzyl alcohol from 64.3% to 12.2% for CuNi/C after four reaction cycles. The BET surface areas measured 1186.0, 15.2, and 94.9 m 2 /g for MIL-101, TiO 2 , and C, respectively. Total pore volume reached 0.49, 0.014, and 0.11 cm 3 /g for MIL-101, TiO 2 , and C, respectively. The large porous cavities of MIL-101 can be used to trap bimetallic Cu-Ni nanoparticles and inhibit nanoparticle leaching. CuNi/MIL-101 displayed better stability than the CuNi/TiO 2 and CuNi/C catalysts. The decline in the catalytic activity was possibly due to the leaching of metals (Cu and Ni) under experimental conditions (Figure 8b,c). The stability of the Cu-Ni-supported catalysts was studied. The XRD patterns of fresh and recycled CuNi/MIL-101, CuNi/TiO2, and CuNi/C are shown in Figure 9. The characteristic peak The stability of the Cu-Ni-supported catalysts was studied. The XRD patterns of fresh and recycled CuNi/MIL-101, CuNi/TiO2, and CuNi/C are shown in Figure 9. The characteristic peak intensity of the recycled CuNi/MIL-101 decreased severely, suggesting that the structure of the The stability of the Cu-Ni-supported catalysts was studied. The XRD patterns of fresh and recycled CuNi/MIL-101, CuNi/TiO 2 , and CuNi/C are shown in Figure 9. The characteristic peak intensity of the recycled CuNi/MIL-101 decreased severely, suggesting that the structure of the recycled CuNi/MIL-101 was slightly destroyed during the oxidation of benzyl alcohol under present reaction conditions (Figure 9a). The XRD patterns of CuNi/TiO 2 and CuNi/C after four runs were similar to those of the as-prepared CuNi/TiO 2 and CuNi/C (Figure 9b,c). Such a similarity illustrates that recycled CuNi/TiO 2 and CuNi/C retained their structure integrity after four runs under the present reaction condition [43]. Figure 10 shows the N 2 adsorption-desorption isotherms for recycled CuNi/MIL-101, CuNi/TiO 2 , and CuNi/C. All of the recycled catalysts displayed a type IV isotherm. The BET surface areas of the recycled CuNi/MIL-101, CuNi/TiO 2 , and CuNi/C were 632.0, 7.2, 21.3 m 2 /g, respectively. The BET surface area decreased significantly for the recycled Cu-Ni-supported catalysts when compared with that of the as-prepared. The decrease in the BET surface area could be due to the collapse of the structure or/and the substrate/solvents blocking the pores of the support [10]. that recycled CuNi/TiO2 and CuNi/C retained their structure integrity after four runs under the present reaction condition [43]. Figure 10 shows the N2 adsorption-desorption isotherms for recycled CuNi/MIL-101, CuNi/TiO2, and CuNi/C. All of the recycled catalysts displayed a type IV isotherm. The BET surface areas of the recycled CuNi/MIL-101, CuNi/TiO2, and CuNi/C were 632.0, 7.2, 21.3 m 2 /g, respectively. The BET surface area decreased significantly for the recycled Cu-Ni-supported catalysts when compared with that of the as-prepared. The decrease in the BET surface area could be due to the collapse of the structure or/and the substrate/solvents blocking the pores of the support [10]. that recycled CuNi/TiO2 and CuNi/C retained their structure integrity after four runs under the present reaction condition [43]. Figure 10 shows the N2 adsorption-desorption isotherms for recycled CuNi/MIL-101, CuNi/TiO2, and CuNi/C. All of the recycled catalysts displayed a type IV isotherm. The BET surface areas of the recycled CuNi/MIL-101, CuNi/TiO2, and CuNi/C were 632.0, 7.2, 21.3 m 2 /g, respectively. The BET surface area decreased significantly for the recycled Cu-Ni-supported catalysts when compared with that of the as-prepared. The decrease in the BET surface area could be due to the collapse of the structure or/and the substrate/solvents blocking the pores of the support [10].  The leaching of the CuNi/MIL-101, CuNi/TiO2, and CuNi/C catalysts was tested to further investigate the stability of the catalysts. This process was conducted by stopping the benzyl alcohol oxidation reaction after 2 h at 100 °C under 5 bar of O2 using THF as the solvent and by separating the reaction mixture from the catalyst via centrifugation. Benzyl alcohol conversions reached 22.9%, 32.8%, and 31.9% at 100 °C under 5 bar of O2 within 2 h over CuNi/MIL-101, CuNi/TiO2, and CuNi/C, respectively. Furthermore, the reaction solution was reused and the entire reaction was continued for another 2 h under the same reaction conditions. Benzyl alcohol conversion increased by 4.8%, 14.6%, and 8.9% after reacting for another 2 h for CuNi/MIL-101, CuNi/TiO2, and CuNi/C, respectively, thereby implying the leaching of active species after the removal of the solid catalyst. The Cu and Ni contents in the reaction solutions after removal of CuNi/MIL-101, CuNi/TiO2, and CuNi/C were 0.00014 and 0.0024, 0.11 and 0.10, and 0.010 and 0.060 wt%, respectively, as determined by ICP-OES, which were consistent with the leaching results.

Preparation of Catalysts
MIL-101 was prepared following a method described in the literature [23]. Cu-Ni bimetallic catalysts were prepared by co-immobilization followed by H2 reduction. A typical procedure for the preparation of the CuNi/MIL-101 catalyst is described as follows. First, Cu(NO3)2·3H2O (0.4 mmol, 96.6 mg) and Ni(NO3)2·6H2O (0.4 mmol, 116.3 mg) were dissolved in 2 mL ethanol. Subsequently, the solution was dropwise added to MIL-101 (0.5 g) and sonicated for 1 h in an ultrasonic cleaner. Finally, the mixture was left to stand for 12 h at room temperature, dried under vacuum for 8 h at 50 °C, and reduced at 50 °C in H2 flow (12 mL/min) for 2 h. The obtained catalyst was labeled CuNi/MIL-101. CuNi/TiO2 and CuNi/C catalysts were prepared using the same procedure, but differed due to changing the MIL-101 support to TiO2 and C. The actual Cu and Ni contents in the CuNi/MIL-101, CuNi/TiO2, and CuNi/C catalysts were 3.15 and 2.80, 3.53 and 3.46, and 2.62 and 2.33 wt%, respectively, as determined by inductively coupled plasma (ICP) optical emission spectroscopy. Four The leaching of the CuNi/MIL-101, CuNi/TiO 2 , and CuNi/C catalysts was tested to further investigate the stability of the catalysts. This process was conducted by stopping the benzyl alcohol oxidation reaction after 2 h at 100 • C under 5 bar of O 2 using THF as the solvent and by separating the reaction mixture from the catalyst via centrifugation. Benzyl alcohol conversions reached 22.9%, 32.8%, and 31.9% at 100 • C under 5 bar of O 2 within 2 h over CuNi/MIL-101, CuNi/TiO 2 , and CuNi/C, respectively. Furthermore, the reaction solution was reused and the entire reaction was continued for another 2 h under the same reaction conditions. Benzyl alcohol conversion increased by 4.8%, 14.6%, and 8.9% after reacting for another 2 h for CuNi/MIL-101, CuNi/TiO 2 , and CuNi/C, respectively, thereby implying the leaching of active species after the removal of the solid catalyst. The Cu and Ni contents in the reaction solutions after removal of CuNi/MIL-101, CuNi/TiO 2 , and CuNi/C were 0.00014 and 0.0024, 0.11 and 0.10, and 0.010 and 0.060 wt%, respectively, as determined by ICP-OES, which were consistent with the leaching results.

Preparation of Catalysts
MIL-101 was prepared following a method described in the literature [23]. Cu-Ni bimetallic catalysts were prepared by co-immobilization followed by H 2 reduction. A typical procedure for the preparation of the CuNi/MIL-101 catalyst is described as follows. First, Cu(NO 3 ) 2 ·3H 2 O (0.4 mmol, 96.6 mg) and Ni(NO 3 ) 2 ·6H 2 O (0.4 mmol, 116.3 mg) were dissolved in 2 mL ethanol. Subsequently, the solution was dropwise added to MIL-101 (0.5 g) and sonicated for 1 h in an ultrasonic cleaner. Finally, the mixture was left to stand for 12 h at room temperature, dried under vacuum for 8 h at 50 • C, and reduced at 50 • C in H 2 flow (12 mL/min) for 2 h. The obtained catalyst was labeled CuNi/MIL-101. CuNi/TiO 2 and CuNi/C catalysts were prepared using the same procedure, but differed due to changing the MIL-101 support to TiO 2 and C. The actual Cu and Ni contents in the CuNi/MIL-101, CuNi/TiO 2 , and CuNi/C catalysts were 3.15 and 2.80, 3.53 and 3.46, and 2.62 and 2.33 wt%, respectively, as determined by inductively coupled plasma (ICP) optical emission spectroscopy. Four different bimetallic Cu:Ni catalysts supported on MIL-101 were also prepared by co-immobilization followed by H 2 reduction, with a total metal loading of ca. 6%. The as-synthesized catalysts were labelled as Cu(3)Ni/MIL-101, Cu(2)Ni/MIL-101, CuNi(2)/MIL-101, and CuNi(3)/MIL-101 on the basis of the mole ratio of Cu and Ni.

Catalyst Characterization
The details of the characterizations with XRD, XPS, nitrogen adsorption/desorption, and ICP atomic emission spectroscopy analysis are the same as those in our previous work [10]. TEM images were recorded with a Jeol JEM-1210 transmission electron microscope (Jeol, Tokyo, Japan) with a high-angle annular dark-field detector. EDS line-scanning profile analysis was performed on an Oxford X-MaxN 80T IE250 spectroscopy instrument (Oxford, UK). NH 3 -TPD analyses of MIL-101, TiO 2 , and C supports were conducted on a MicrotracBEL AutoChem BELCAT II Chemisorption Analyzer (Bel, Osaka, Japan).

Selective Oxidation of Benzyl Alcohol
Selective oxidation of benzyl alcohol with molecular oxygen was conducted in a stainless-steel high-pressure reactor equipped with magnetic stirring and a temperature controller under mild conditions. In a typical experiment, 0.2 mmol benzyl alcohol, 40 mg catalyst, and 1.5 mL solvent were mixed in a 10 mL stainless-steel high-pressure reactor. The system was pumped five times with pure O 2 to remove air prior to the reaction. Subsequently, the reactor was immersed in a methylsilicon oil bath at 100 • C under 5 bar of O 2 to initiate the reaction. After the reaction, the mixture was centrifuged to remove the catalyst particles completely. The remaining solution was analyzed with a gas chromatograph (GC-6890, Purkinje General instrument Co., Ltd., Beijing, China) equipped with a SE-54 capillary column and flame ionization detector. Reaction products were identified by comparison with known standard samples, and an external standard method was used for the qualitative analysis of the amounts of the reactants consumed (conversion) and products generated (selectivity). In the recycling experiment, the catalysts were separated from the mixture by centrifugation after each run, and washed with ethanol twice. Finally, the catalysts were dried at 50 • C for 4 h in a vacuum at 0.1 MPa for the next reaction.

Conclusions
A series of bimetallic Cu-Ni nanoparticles supported on MIL-101, TiO 2 , and C was successfully prepared using co-immobilization followed by H 2 reduction for the selective oxidation of benzyl alcohol to benzaldehyde using molecular oxygen as the oxidizing agent. The acidity of the supports and Cu:Ni mole ratios substantially influenced the catalytic performance of the oxidation of benzyl alcohol into benzaldehyde. The CuNi/MIL-101 catalyst possessing more acid sites showed the highest yield for the selective oxidation of benzyl alcohol to the corresponding aldehyde at 100 • C. Additionally, the CuNi/MIL-101 catalyst exhibited a high functional group (electron-donating and electron-withdrawing groups) tolerance. The CuNi/MIL-101 catalyst also displayed better stability than the CuNi/TiO 2 and CuNi/C catalysts due to the large porous cavities of the MIL-101 support. These findings emphasize the potential of developing new heterogeneous catalytic systems for the oxidation of benzyl alcohol under mild conditions.