Effect of Calcination Atmosphere on the Performance of Cu/Al₂O₃ Catalyst for the Selective Hydrogenation of Furfural to Furfuryl Alcohol

Abstract

The selective hydrogenation of the biomass platform molecule furfural (FAL) to produce furfuryl alcohol (FA) is of great significance to alleviate the energy crisis. Cu-based catalysts are the most commonly used catalysts, and their catalytic performance can be optimized by changing the preparation method. This paper emphasized the effect of calcination atmosphere on the performance of a Cu/Al₂O₃ catalyst for the selective hydrogenation of FAL. The precursor of the Cu/Al₂O₃ catalyst prepared by the ammonia evaporation method was treated with different calcination atmospheres (N₂ and air). On the basis of the combined results from the characterizations using in situ XRD, TEM, N₂O titration, H₂-TPR and XPS, the Cu/Al₂O₃ catalyst calcined in the N₂ atmosphere was more favorable for the dispersion and reduction of Cu species and the reduction process could produce more Cu⁺ and Cu⁰ species, which facilitated the selective hydrogenation of FAL to FA. The experimental results showed that the N₂ calcination atmosphere improved the FAL conversion and FA selectivity, and the FAL conversion was further increased after reduction. Cu/Al₂O₃-N₂-R exhibited the outstanding performance, with a high yield of 99.9% of FA after 2 h at 120 °C and an H₂ pressure of 1 MPa. This work provides a simple, efficient and economic method to improve the C=O hydrogenation performance of Cu-based catalysts.

1. Introduction

The energy supply crisis and environmental pollution caused by the excessive consumption of the fossil energy have aroused worldwide concerns [1]. In order to achieve the sustainable development, researchers have spared no efforts to explore and develop new renewable energy as the alternative, which has included solar energy, hydroelectric energy, wind energy, biomass energy, geothermal energy, tidal energy and so on [2,3]. Among them, biomass energy is the only carbon-containing energy and is the fourth largest category after the traditional fossil energy of coal, oil and natural gas [4]. It occupies a very important position in the overall energy system. Compared with other energy products, biomass energy has the advantages of abundant reserves, strong renewable ability and wide distribution and is an ideal substitute for fossil energy [2,5].

Furfural (FAL) is an important biomass platform compound and considered to be one of the most promising platform molecules for sustainable production of fuels and chemicals in the 21st century [6,7]. FAL has a variety of functional groups such as furan ring, aldehyde group and diene ether, and it can generate various derivatives through selective hydrogenation reactions [8,9], as shown in Figure 1. Among these products, furfuryl alcohol (FA), as one of the important value-added intermediates, accounts for about 65% of the total output of FAL derivatives [10]. FA is mainly used in the production of resins [8], plasticizers [11], wood preservatives [12] and as a platform molecule for the synthesis of other chemicals [13]. Due to the complexity of the FAL reaction pathway, obtaining high FAL conversion and high FA selectivity simultaneously remains a great challenge [11].

The CuCr oxide catalyst is mainly used in industry to catalyze FAL hydrogenation to produce FA [14]. However, the high conversion and remarkable yield require high temperature and high pressure, and the employed catalysts also produce the toxic Cr-containing waste for the environment [15]. In recent years, developing the Cr-free catalysts with superior performance as the alternatives has been an interesting topic [16]. In the beginning, most of the catalysts used to catalyze the FAL hydrogenation conversion were noble metals, including Pd [17], Pt [18] and Ru [19]. Although the noble metal catalysts have the high intrinsic hydrogenation activity, the low selectivity for FA and high price limit their large-scale industrial application. Recently, non-noble metals, including Fe [20], Co [21], Ni [22] and Cu [23], have attracted wide interest because of their abundant reserves, low price and simple availability. As long as the reactivity, selectivity and stability can be effectively improved, they can be the promising alternatives for the industrial applications [14]. Among the non-noble metals, Cu-based catalysts have received much attention due to their specific hydrogenation activity towards C=O [24]. Theoretical calculations and experimental results have well demonstrated that FAL is linear chemical adsorbed in a η¹(O) configuration on the active Cu species [14]. This avoids the hydrogenation of the furan ring, resulting in high FA selectivity. However, the low reactivity of Cu-based catalysts urgently needs to be improved [25].

Many strategies, including but not limited to improving the active components [26,27], modulating the supports [28,29], doping the promotors [30,31] and tuning the preparation methods [32,33], have been used to develop the novel Cu-based catalysts. Zhang et al. [26] obtained the highly distributed Cu/MgO catalysts, where the Cu⁰/Cu⁺ molar ratio increased gradually with the upward reduction temperature. The experimental results showed that the Cu/MgO-350 exhibited excellent stability and reusability without significant activity loss. The reduction temperature could modulate the nature of the interface between the Cu⁰/Cu⁺ active species and the metal oxide interface in Cu/MgO catalysts. Gong et al. [34] reported a copper-based catalyst supported by sulfonate group (-SO₃H)-grafted active carbon (AC). The modified Cu/AC-SO₃H catalyst exhibited an enhanced catalytic performance for selective hydrogenation of FAL to FA in the liquid phase. Through grafting the sulfonate group on the support, better metal dispersion, more Cu⁰/Cu⁺ species and stronger FAL adsorption capacity were attained to contribute the high hydrogenation performance. Zhang et al. [30] synthesized a Pd-CuO_x nanocomposite catalyst with outstanding performance for the selective hydrogenation of FAL to FA. The valence state of Cu and Pd-Cu interactions played the critical roles in determining the intrinsic activity of the prepared Pd-Cu catalysts. Various characterizations combined with the kinetic experiments and in situ chemisorption clearly unraveled the adsorption and activation processes of the C=O bond and H₂ molecule on Pd⁰, Cu⁰ and Cu⁺ sites. Zhang et al. [35] provided an efficient strategy of K₂CO₃ assisting the CuO#TiO₂ catalyst for the liquid-phase hydrogenation of FAL. The reason for the increased yield was ascribed to the boosted generation of the surface Cu⁺/Cu⁰ active sites and the promoted gaseous hydrogen dissolution in ethanol media by K₂CO₃. Cheng et al. [36] developed a new catalytic transfer hydrogenation (CTH) strategy for the continuous conversion of FAL into FA with Cu/ZnO/Al₂O₃ as the catalyst under mild conditions. The highly dispersive Cu species (Cu⁰, Cu⁺), with few Lewis acidic sites and η¹(O)-type adsorption, ensured high selectivity and mild conversion. Du et al. [32] investigated the effect of preparation methods on the structure and performance of Cu/SiO₂ catalysts. The results showed that the excellent performance was associated with the highest Cu⁰ surface area, the smallest Cu particle size and the suitable Cu⁺/(Cu⁺ + Cu⁰) ratio. The above studies have confirmed that the efficient Cu-based catalysts are closely related to the abundance of Cu⁺ and Cu⁰ species. Cu⁰ and Cu⁺ active species synergistically catalyze C=O bond hydrogenation, where Cu⁰ promotes the dissociation of H₂, while Cu⁺ adsorbs and activates C=O bonds as the Lewis acid sites [37].

In addition to the above methods, the calcination atmosphere also has great influences on the size of the active metal [38], the metal–support interaction [39] and the surface active species [40], which could also affect the catalytic performance [41]. However, the effect of the calcination atmosphere on the activity of Cu-based catalysts is rarely reported. Moreover, compared with other modification methods, tuning the calcination atmosphere to pretreat catalyst precursor is easier to operate and more economic, which has the potential application prospects.

Herein, Cu/Al₂O₃ catalyst precursor prepared by the ammonia evaporation method was calcined in the different atmospheres and employed for FAL hydrogenation. The physical and chemical properties of Cu/Al₂O₃ samples were systematically characterized by in situ XRD, TEM, N₂O titration, H₂-TPR, and XPS, and the correlation between catalytic activity and catalyst properties was discussed to unravel the effects of calcination atmosphere on the hydrogenation performance.

2. Experimental

2.1. Materials

FAL (99.8%), FA (99.8%), Al₂O₃ and Cu(NO₃)₂·3H₂O were purchased from McLean Chemical Reagent Co., Ltd. (Shanghai, China). NH₃·H₂O (25 wt%) was purchased from Aladdin Reagent Shanghai Co., Ltd. (Shanghai, China). Isopropanol was purchased from Fengchuan Chemical Reagent Technology Co., Ltd. (Tianjin, China). The deionized water was self-prepared. All reagents were used without further purification.

2.2. Catalyst Preparation

The Cu/Al₂O₃ catalyst precursor was prepared by the ammonia evaporation method [42]. Firstly, 2.4030 g of Cu(NO₃)₂·3H₂O was dissolved in 140 mL of deionized water. After stirring for 0.5 h, 25 wt% ammonia solution was slowly added to the above solution under stirring to adjust the pH to 12. Then, Al₂O₃ (12 g) was added to the above solution, and the suspension was stirred for another 4 h. Thereafter, the mixture was heated to 90 °C to evaporate ammonia until the pH decreased to 7. Finally, the resultant solid was filtered, washed thoroughly with the deionized water and dried at 120 °C for 12 h to obtain the Cu/Al₂O₃ catalyst precursor.

The Cu/Al₂O₃ catalyst precursor was calcined at 450 °C for 4 h in the different atmospheres, including N₂ and air in the tube furnace. Before each test, Cu/Al₂O₃ catalyst was reduced in 10% vol. H₂/Ar at 350 °C for 4 h. The calcined and reduced catalysts were denoted as Cu/Al₂O₃-N₂, Cu/Al₂O₃-air, Cu/Al₂O₃-N₂-R and Cu/Al₂O₃-air-R, respectively.

2.3. Catalyst Characterization

In situ X-ray diffraction (XRD) measurement was carried out on a SmartLab SE diffractometer (Rigaku Corporation), and the reaction cell was controlled by PTC-EVO. The patterns were recorded with a 2 Theta range of 20–80° and scan rate of 10 °/min. Transmission electron microscopy (TEM) and high-resolution TEM (HRTEM) were taken on a Tecnai G2 S-Twin F20 TEM microscope (FEI Company) at 200 kV. The textural properties were measured on an ASAP 2460 sorptometer (Micromeritics) at −196 °C. The sample was degassed at 150 °C for 3 h before each test. The specific surface area (S_BET) was calculated by the Brunauer–Emmett–Teller (BET) method, and the pore volume, as well as the pore size, was calculated by the Barrett–Joyner–Halenda (BJH) model. The actual Cu loading was determined by an inductively coupled plasma atomic mission spectrometer (ICP-AES) on Shimadzu ICPE-9820. X-ray photoelectron spectra (XPS) and X-ray excited Auger electron spectra (AES) were acquired on a Thermo VG Scientific ESCALAB 250 (UK) equipped with Al Kα X-ray excitation. All binding energies were referenced to C 1s (284.8 eV).

H₂ temperature-programmed reduction (H₂-TPR) and Cu dispersion measurement were performed on AutoChem Ⅱ 2920 chemisorption analyzer (Micromeritics) equipped with an online thermal conductivity detector (TCD). For H₂-TPR, the catalyst (~100 mg) was loaded into the U-shaped quartz tube, and pretreatment occurred under Ar flow at 120 °C for 60 min. The signal of H₂ consumption was subsequently monitored under 10 vol.% H₂/Ar (50 mL/min) at a heating rate of 10 °C/min from room temperature to 800 °C. Cu dispersion was measured by the technique of N₂O titration [43,44]. Firstly, the fresh catalyst (~100 mg) was pretreated at 120 °C for 60 min under Ar flow and cooled to 50 °C. Afterward, the sample was heated to 350 °C under 10 vol.% H₂/Ar gas (50 mL/min) with a rate of 10 °C/min. The area of H₂ consumption was recorded as X. Then, the tube was cooled down to 50 °C and re-oxidized by 10 vol.% N₂O/He for 1 h. Finally, the sample was heated to 350 °C under 10 vol.% H₂/Ar (50 mL/min) with a rate of 10 °C/min. The area of H₂ consumption was recorded as Y. Cu dispersion and the surface area of surface Cu species per gram catalyst (S(m²∙g_cat⁻¹)) were calculated according to the following equations:

$$\mathrm{Cu} \mathrm{dispersion}=[2\mathrm{Y} / \mathrm{X}] \times 100\%$$

(1)

$$\mathrm{S}=(2 \times {\mathrm{N}}_{\mathrm{A}} \times \mathrm{Y}) / (\mathrm{X} \times 1.4 \times {10}^{19}{\times \mathrm{M}}_{\mathrm{Cu}} \times { \mathrm{Wt}}_{\mathrm{Cu}}\%)=(1353 \times \mathrm{Y}) / (\mathrm{X} {\times \mathrm{Wt}}_{\mathrm{Cu}}\%)$$

(2)

where, M_Cu, Wt_Cu% and N_A are the molecular weight of Cu, actual Cu loading and Avogadro’s constant, respectively. In addition, 1.4 × 10¹⁹ is the number of copper atoms per square meter.

In situ diffuse reflectance infrared Fourier transform (DRIFT) spectra of FAL adsorption and hydrogenation were conducted on INVENIO S FTIR (Bruker) equipped with a diffuse reflectance cell and mercury-cadmium telluride (MCT) detector. Before each test, the sample was reduced under 10 vol.% H₂/Ar at 350 °C for 1 h. After cooling down to 30 °C under N₂, the background signal was collected. Then, FAL was introduced into the cell with a bubbler using N₂ as the carrier. Thereafter, the temperature was heated to 120 °C to desorb the physically adsorbed FAL. When the desorption was completed, 10 vol.% H₂/Ar was introduced for in situ hydrogenation reaction, and the spectra were finally collected every 10 min.

2.4. Catalyst Evaluation

The activity of the catalysts was evaluated in a 100 mL stainless-steel batch reactor, which was equipped with mechanical stirrer and temperature controller. Typically, catalyst (0.1 g), FAL (0.5 g) and solvent (30 mL) were added into the reactor. After sealing, H₂ was charged to replace the internal air for 5 times. Then, the reactor was filled with the desired H₂ pressure and heated to the predetermined temperature under the continuous mechanical stirring (800 rpm). After the required time, the reactor was naturally cooled down to room temperature, and the mixture was separated by filtration and collected. The products were analyzed by a gas chromatographer (GC9600) equipped with flame ionization detector (FID), and the carrier gas was ultra-high-purity N₂. The standard solutions of FAL and FA with different concentrations were precisely prepared. External standard method was used to quantify FAL conversion and product selectivity as follows [45,46]:

$$\mathrm{Conversion} (\%)=\frac{\mathrm{Mole} \mathrm{of} \mathrm{FAL} \mathrm{reacted}}{\mathrm{Mole} \mathrm{of} \mathrm{initial} \mathrm{FAL}} \times 100\%$$

(3)

$$\mathrm{Selectivity} (\%)=\frac{\mathrm{Mole} \mathrm{of} \mathrm{FA} }{\mathrm{Mole} \mathrm{of} \mathrm{FAL} \mathrm{converted}} \times 100\%$$

(4)

$$\mathrm{Yield} (\%)= \mathrm{Conversion} \times \mathrm{selectivity}$$

(5)

3. Results and Discussion

3.1. Structure and Morphology

Figure 2a–f show the in situ XRD patterns of the Cu/Al₂O₃ catalyst precursor during calcination in N₂ and air and the following reduction in H₂/Ar. As shown in Figure 2a, with the increased temperature from 50 to 450 °C in the N₂ atmosphere, only the CuO species with the characteristic diffraction peaks at 35.6° and 38.7° were observed [43,47]. Here, the CuO species might be derived from the thermal decomposition of Cu(OH)₂ and [Cu(NH₃)_x]²⁺ in the precursor. Figure 2b displays the further thermal treatment in N₂ atmosphere at 450 °C. The peaks at 43.3° and 50.5° corresponded to the (111) and (200) crystalline planes of Cu⁰, respectively [48]. Notably, the intensity of the diffraction peaks ascribed to the CuO species decreased gradually with the extended time (after 2 h, the peaks disappear), while the diffraction peaks indexed to the Cu⁰ species appeared after 1 h and became sharper with the prolonging time. Figure 2c displays the reduction process of the Cu/Al₂O₃-N₂ catalyst in H₂/Ar atmosphere with the increased temperature from 50 to 350 °C. Observably, the intensity of the diffraction peaks assigned to Cu⁰ gradually enhanced with increasing the reduction temperature, implying a further reduction of the CuO species. Figure 2d shows the patterns of the Cu/Al₂O₃-N₂ catalyst reduced for another 4 h at 350 °C. It was noted that the diffraction peaks of Cu⁰ were almost unchanged, suggesting the high anti-sintering ability of the Cu/Al₂O₃ catalyst [49].

From in situ XRD patterns in Figure 2e,f during the calcination in air, one can note that only the diffraction peaks of CuO species were present, and no diffraction peaks of Cu⁰ appeared. When the Cu/Al₂O₃-air catalyst was reduced by H₂/Ar (Figure 2g), the intensity of the diffraction peaks attributed to CuO decreased with increasing the reduction temperature, and the diffraction peaks of Cu⁰ species came to exist at 200 °C. When the reduction temperature was increased to 300 °C, the diffraction peaks of CuO completely disappeared. The intensity of the diffraction peaks corresponding to Cu⁰ continuously increased when the reduction temperature rose to 350 °C. As shown in Figure 2h, the diffraction peaks intensity of Cu⁰ remained stable after 3 h reduction at 350 °C. The aforementioned analysis indicates that the presence of Cu⁰ species during the calcination in N₂ may have been to the generation of Cu(NH₃)₄(NO₃)₂ in the catalyst precursor prepared by ammonia evaporation method. The thermal decomposition of Cu(NH₃)₄(NO₃)₂ at a high temperature can produce reduction by NH₃, which could reduce CuO to Cu⁰. Therefore, the Cu/Al₂O₃ precursor was easier to reduce to the lower valence state in the N₂ atmosphere.

To study the textural properties of the Cu/Al₂O₃ precursor and the employed catalysts, N₂ physical adsorption-desorption experiments were carried out (

Table 1. Physicochemical properties of the Cu/Al₂O₃ precursor and the employed catalysts.

Sample	SBET (m2/g) a	Vpore (cm3/g) a	Dpore (nm) a	Cu Loading (wt.%) b	DCu (%) c	SCu (m2Cu/gcat) c
Cu/Al2O3 precursor	131.5	0.80	20.5	-	-	-
Cu/Al2O3-N2	125.1	0.80	20.7	4.3	-	-
Cu/Al2O3-N2-R	128.5	0.76	19.9	4.3	48.3	14.1
Cu/Al2O3-air	125.0	0.78	21.1	4.3	-	-
Cu/Al2O3-air-R	127.2	0.77	20.8	4.3	21.5	6.3

^a: Measured by N₂ adsorption; ^b: Determined by ICP-AES; ^c: Calculated by N₂O titration.

and Figure 3). According to the IUPAC classification [50], all the isotherms exhibited type IV with H3 hysteresis loop (Figure 3a), indicating the typical mesoporous structure [48,51]. As shown in Figure 3b and Table 1, the specific surface areas, pore volumes and average pore sizes of the precursor and the catalysts, including the calcined and reduced samples, were very similar. It illustrates that the calcination atmosphere and further reduction did not affect the textural properties of the Cu/Al₂O₃ catalyst [52].

TEM and HRTEM were employed to investigate the size of Cu nanoparticles and the morphology of the reduced Cu/Al₂O₃ catalysts. As shown in Figure 4a,b, the average size of Cu nanoparticles was 4.9 and 6.6 nm for Cu/Al₂O₃-N₂-R and Cu/Al₂O₃-air-R, respectively. That is, the average particle size of the Cu species on Cu/Al₂O₃-N₂-R was smaller and more uniformly dispersed than that on Cu/Al₂O₃-air-R. The lattice fringe spacing was measured to be 0.21 and 0.18 nm in the HRTEM images (Figure 4c,d), corresponding to the (111) and (200) crystalline plane of Cu, respectively [48]. The assignment of the above crystalline planes is consistent with the XRD results. Moreover, the high-angle annular darkfield scanning transmission electron microscopy (HAADF-STEM) and the corresponding elemental mapping images (Figure 4f–h,j–l) confirm that Cu, O and Al were homogeneously distributed on the Cu/Al₂O₃ catalysts [53].

The actual Cu loadings on the catalysts determined by ICP-AES are listed in Table 1. The actual loading of 4.3 wt.% was lower than the theoretical loading of 5 wt.%, which may have been caused by the incomplete precipitation and washing during the preparation [42]. The actual loadings on the different catalysts were the same, which suggests that the calcination atmosphere and reduction treatment could not have resulted in the metal loss and thus did not affect the experimental results and conclusions. Moreover, the dispersion (D_Cu) and specific surface area of Cu (S_Cu) on the catalysts were calculated (Table 1). The results show that the dispersion on Cu/Al₂O₃-N₂-R catalyst was 48.3%, which was much higher than that on Cu/Al₂O₃-air-R catalyst, and the specific surface area of Cu was 14.1 m²_Cu/g_cat. The Cu dispersion and specific surface area on Cu/Al₂O₃-air-R catalyst were only 21.5% and 6.3 m²_Cu/g_cat, respectively. The data are consistent with the TEM results. In conclusion, the catalyst calcined in the N₂ atmosphere was more favorable for the dispersion and reduction of Cu species.

3.2. Surface Chemical State

H₂-TPR was performed to study the reduction behavior of the calcined Cu/Al₂O₃ catalysts, and the profiles are given in Figure 5a. Apparently, the calcination atmosphere significantly affected the reduction behaviors of the CuO species. The reduction peaks centered at 208 °C/218 °C (a) and 245 °C/290 °C (b) were observed for the Cu/Al₂O₃ catalysts [28]. The peak a was attributed to the reduction of the highly dispersed CuO species, and the peak b was related to the reduction of the small-sized CuO species [54,55]. The temperature of the peak a for Cu/Al₂O₃-N₂ was lower than that for Cu/Al₂O₃-air. This indicates that the former was more easily reduced, which may have been due to its higher Cu dispersion. Peak fitting and integration were performed to better quantify the percentage of different CuO species, and the quantitative results are plotted in Figure 5b. Cu/Al₂O₃-N₂ had a higher proportion of highly dispersed CuO, accounting for 95.1% of all Cu species, while Cu/Al₂O₃-air only contained 45.4% of highly dispersed CuO. This again verifies that for the Cu/Al₂O₃ precursor, the calcination in N₂ was more favorable for the dispersion of Cu species, which are easier to reduce. The analysis is in good agreement with the TEM and N₂O titration results.

To further examine the chemical state of Cu species, XPS was also conducted, and the spectra are displayed in Figure 6a. The spectra at the region of 930.0–937.9 eV were fitted to two characteristic peaks, where the feature at 932.5 eV corresponded to the low oxidation state species (Cu⁺/Cu⁰), and the one at 934.4 eV was assigned to the Cu²⁺ species [54]. According to the fitting results (

Table 2. Surface element compositions on the Cu/Al₂O₃ catalysts.

Catalyst	(Cu0 + Cu+)/Cu(%) a	Cu+/Cu(%) b	Cu0/Cu(%) b
Cu/Al2O3-N2	37.1	27.4	9.7
Cu/Al2O3-N2-R	60.8	45.3	15.5
Cu/Al2O3-air	21.8	13.8	8.0
Cu/Al2O3-air-R	48.8	38.6	10.2

^a: calculated from the XPS data. ^b: calculated from the AES data.

), the ratio of (Cu⁺/Cu⁰)/Cu on the surface of Cu/Al₂O₃-air was only 21.8%, but the Cu/Al₂O₃-N₂ reached 37.1%. After reduction in H₂/Ar, the ratio of (Cu⁺/Cu⁰)/Cu increased to 48.8% and 60.8%, respectively. This suggests that the Cu/Al₂O₃ catalyst calcined in the N₂ atmosphere was more easily reduced to the low oxidation state, which is consistent with the H₂-TPR results.

It is difficult to make a clear distinction between Cu⁺ and Cu⁰ on the basis of XPS analysis [56]. Thus, the Cu AES spectra were also recorded (Figure 6b). Three peaks were extracted by peak fitting in the Kinetic energy (KE) range of 903-922 eV, including the characteristic of Cu⁰ (KE = 918.3 eV), Cu²⁺ (KE = 916.3 eV) and Cu⁺ (KE = 914.3 eV) species [57]. Further quantitative analysis of the fitted peaks is listed in Table 2. The Cu⁺ and Cu⁰ ratio of Cu/Al₂O₃ catalyst calcined in air atmosphere were 13.8% and 8.0%, respectively. However, the Cu⁺ and Cu⁰ content of Cu/Al₂O₃-N₂ reached higher levels of 27.4% and 9.7%. It further indicates that the calcination in the N₂ atmosphere was more favorable for the generation of low valence Cu species, which is in accordance with the results of in situ XRD and TPR. After reduction by H₂/Ar atmosphere, the Cu⁺ and Cu⁰ content on both Cu/Al₂O₃-air-R and Cu/Al₂O₃-N₂-R further increased, where on the former, an increase to 38.6% and 10.2% was observed, respectively, and on the latter, the Cu⁺ and Cu⁰ ratio had the highest values of 45.3% and 15.5%, respectively. According to previous reports, high Cu⁺ and Cu⁰ ratio facilitates the selective hydrogenation of FAL [24,48,54].

Full-spectrum analysis of XPS spectra was performed. The surface Cu/Al atomic ratios of Cu/Al₂O₃-N₂-R and Cu/Al₂O₃-air-R were 0.056 and 0.092, respectively. The surface Cu/Al atomic ratio of Cu/Al₂O₃-N₂-R was lower, indicating that Cu nanoparticles were highly dispersed on the Al₂O₃ support. The small Cu nanoparticles easily entered the support pores, leading to a decrease in the surface Cu/Al atomic ratio. However, the Cu/Al₂O₃-air-R dispersion was poor. A large number of Cu nanoparticles aggregated on the support surface, resulting in a high surface Cu/Al atomic ratio. The above results are in good agreement with the N₂O titration and TEM results, which prove that Cu/Al₂O₃-N₂-R had higher dispersion.

3.3. Catalytic Performance

The selective hydrogenation of FAL over the employed Cu/Al₂O₃ catalysts was examined in the stainless-steel batch reactor, and the obtained data are given in

Table 3. Hydrogenation performance of FAL to FA over the employed Cu/Al₂O₃ catalysts.

Entry	Catalyst	FAL Conversion (%)	FA Selectivity (%)	FA Yield (%)
1	Cu/Al2O3-N2	42.7	99.9	42.7
2	Cu/Al2O3-N2-R	99.9	99.9	99.9
3	Cu/Al2O3-air	14.3	62.2	8.9
4	Cu/Al2O3-air-R	26.7	85.8	22.9

Reaction conditions: (0.1 g), FAL (0.5 g), isopropanol (30 mL), temperature (120 °C), time (2 h), H₂ pressure (1 MPa).

. FA yield over the Cu/Al₂O₃ catalysts showed the following trend: Cu/Al₂O₃-N₂-R > Cu/Al₂O₃-N₂ > Cu/Al₂O₃-air-R > Cu/Al₂O₃-air. Cu/Al₂O₃ precursor calcined in air (Cu/Al₂O₃-air) had very low FAL conversion and FA selectivity. Even after reduction (Cu/Al₂O₃-air-R), the improvement of the catalytic activity was still very limited, where FAL conversion and FA selectivity were only 26.7% and 85.8%, respectively. Interestingly, the precursor calcined in N₂ had the relatively superior activity. FAL conversion reached 42.7%, and FA selectivity was even 99.9% over Cu/Al₂O₃-N₂. After reduction, the activity was greatly enhanced. Specifically, FAL conversion and FA selectivity reached 99.9% over Cu/Al₂O₃-N₂-R.

The evaluation data indicate that the calcination atmosphere significantly affected the performance of Cu/Al₂O₃ catalyst for FAL hydrogenation. Compared with the Cu/Al₂O₃-air-R, the FAL conversion and FA selectivity of Cu/Al₂O₃-N₂-R were greatly improved. In general, among the Cu⁰ and Cu⁺ active hydrogenation species on the catalyst surface, the Cu⁰ species contributed the dissociation and activation of H₂ [27]. The above in situ XRD, TEM, N₂O titration and H₂-TPR characterization results demonstrate that Cu/Al₂O₃ catalyst calcined in the N₂ atmosphere was more favorable for the dispersion and reduction of Cu species. This facilitated the reduction of CuO to the lowest valence state of the Cu⁰ species. The XPS results were the same as the above conclusion. The Cu⁰ content over Cu/Al₂O₃-N₂-R was the highest, accounting for 15.5%. Then, Cu⁺ species could act as Lewis acid sites and polarize the carbonyl group (C=O), thus promoting the formation of FA [24]. The Cu AES spectra fitting results show that Cu/Al₂O₃-N₂-R also contained the highest content of Cu⁺ species, accounting for 45.3%. Compared to Cu/Al₂O₃-air, the Cu/Al₂O₃ catalyst calcined in the N₂ atmosphere had higher Cu dispersion and higher content of Cu⁺/Cu⁰, resulting in the improved FA conversion and FAL selectivity. After reduction, Cu/Al₂O₃-N₂-R obtained the highest content of Cu⁺ and Cu⁰ active hydrogenated species, and the FAL conversion was further improved. Cu⁰ and Cu⁺ synergistically catalyzed the selective hydrogenation of FAL to prepare FA. Therefore, Cu/Al₂O₃-N₂-R had the highest catalytic activity.

The reusability of the Cu/Al₂O₃-N₂-R catalyst for FAL hydrogenation was also tested. After every cycle, the catalyst was washed with isopropanol and water to remove the adsorbed organics on the surface and dried overnight in a vacuum oven at 60 °C. As displayed in Figure 7, FAL conversion and FA selectivity was not obviously declined after four recycles. These results imply that the catalyst possessed a good reusability performance.

Table 4. A brief comparison of the performance over the representative Cu-based catalysts for FAL hydrogenation to FA using the batch reactor.

Catalyst	Reaction Conditions	Con. (%)	Sel. (%)	Yield (%)	Ref.
Cu2Ni1AlOy	120 °C, 1.6 MPa H2, 1.5 h	98.0	99.0	97.0	[58]
10Cu-MCM-41-GLY	150 °C, 2 MPa H2, 2 h	98.1	80.0	78.0	[59]
Cu/SiO2	110 °C, 1 MPa H2, 4 h	66.0	100.0	66.0	[60]
Cu/C-400	170 °C, 2 MPa H2, 3 h	99.6	99.3	98.9	[61]
Cu/MgO	180 °C, 0.9 MPa	95.0	94.0	89.3	[62]
CuAlOx	220 °C, 3.5 MPa H2, 5 h	91.0	98.4	89.5	[63]
Na-Cu@TS-1	110 °C, 1 MPa H2, 2 h	93.0	98.1	91.2	[64]
Cu/Co3O4	170 °C, 2 MPa H2, 1 h	100.0	67.0	67.0	[65]
NiCu0.33/C	120 °C, 1.5 MPa H2, 12 h	96.7	93.8	89.6	[66]
7NiCu/ZrO2	200 °C, 1.5 MPa H2, 4 h	99.5	93.0	93.0	[67]
Ni-Cu	110 °C, 6 MPa H2, 2.3 h	50.0	100.0	50.0	[68]
Cu/γ-Al2O3	130 °C, 4 MPa H2, 4 h	64.2	72.3	46.4	[69]
Cu/CeO2	210 °C, 0.1 MPa H2	83	67	55.6	[70]
Cu/Al2O3-N2-R	120 °C, 1 MPa H2, 2 h	99.9	99.9	99.9	This work

gives a brief comparison of the performance over the representative Cu-based catalysts for FAL hydrogenation to FA using the batch reactor. Notably, the present work provides an efficient FAL hydrogenation catalyst under the relatively mild conditions. Therefore, the strategy in this work has certain advantages for the selective hydrogenation of FAL catalyzed by non-noble metal catalysts.

3.4. Reaction Mechanisms

Figure 8a shows the in situ DRIFT spectra of FAL adsorption at 120 °C. The observed bands located at 1643 cm⁻¹ were attributed to the C=O stretching vibration [71], and the bands at 1592 cm⁻¹ and 1452 cm⁻¹ corresponded to the furan ring breath and C=C characteristic vibrations, respectively [51,71]. Compared with the v(C=O) for gaseous FAL, the significant red shift in v(C=O) from 1670 cm⁻¹ to 1643 cm⁻¹ implies that FAL was linearly chemical adsorbed in the η¹(O) configuration on the catalysts [52]. This facilitates FAL hydrogenation towards the production of FA. Cu/Al₂O₃-N₂-R had larger C=O peak intensity, which indicates that Cu/Al₂O₃-N₂-R was more favorable for C=O chemical adsorption. Moreover, in situ DRIFT spectra of FAL hydrogenation are also provided in Figure 8b to further understand the evolution of C=O hydrogenation. Apparently, the strong characteristic bands assigned to C=O declined faster on Cu/Al₂O₃-N₂-R with the continuous inflow of H₂ than the Cu/Al₂O₃-air-R catalyst, indicating the higher catalytic activity of Cu/Al₂O₃-N₂-R. This is consistent with experimental data on FAL hydrogenation.

4. Conclusions

In summary, the Cu/Al₂O₃ catalysts pretreated by calcining in the N₂ and air atmosphere were prepared and applied for FAL liquid-phase hydrogenation. The activity of the catalyst calcined in air was lower than that calcined in N₂, which was related to the generated lower valence state during the calcination process on the latter. After reduction, the performance of FAL selective hydrogenation to FA was proliferated. Specially, the Cu/Al₂O₃-N₂-R catalyst exhibited the most excellent performance. At 120 °C, 1 MPa H₂, 2 h, both FAL conversion and FA selectivity over Cu/Al₂O₃-N₂-R reached 99.9%, which was comparable to the reported Cu-based catalysts. The comprehensive characterization results based on in situ XRD, TEM, N₂O titration, H₂-TPR and XPS demonstrate that the Cu/Al₂O₃ catalyst calcined in the N₂ atmosphere was more favorable for the dispersion and reduction of Cu species and the reduction process could produce more Cu⁺ and Cu⁰ species. Thus, Cu/Al₂O₃-N₂-R had the highest content of Cu⁺/Cu⁰ active hydrogenation species among the employed catalysts, which provides a reasonable explanation for its efficient activity under the relatively mild conditions. This work designs a simple strategy for optimizing the reactivity of Cu-based catalysts and is conducive to efficiently utilizing biomass and its derivatives to develop a renewable energy system.