Cu/MgO as an Efficient New Catalyst for the Non-Oxidative Dehydrogenation of Ethanol into Acetaldehyde

Abstract

The non-oxidative dehydrogenation of ethanol into acetaldehyde is one of the efficient solutions for biomass upgrading. In this work, a series of copper catalysts supported on MgO with different Cu loadings ranging from 2.5% to 20% were prepared by an impregnation method. The as-synthesized Cu/MgO catalysts were characterized by N₂ adsorption, XRD, TEM, CO₂-TPD, XPS and TPR. These catalysts were found to be effective for ethanol dehydrogenation into acetaldehyde. As the Cu loading was increased, the ethanol conversion first increased and then leveled off. At a WHSV of 1.5 h⁻¹ and 250 °C, the 20%Cu/MgO catalyst gave an initial conversion of 81.5%, with 97.7% selectivity toward acetaldehyde. Compared to 20%Cu/SiO₂, the 20%Cu/MgO catalyst displayed an equivalent initial acetaldehyde yield, higher acetaldehyde selectivity and longer stability.

1. Introduction

The development of alternative energy sources from renewable biomass and its derivatives contributes to alleviating the dependence on fossil energy. Particularly, ethanol, which can be obtained from crops via fermentation easily and abundantly, has attracted extraordinary attention [1,2]. In the past few decades, in addition to use as a gasoline additive, ethanol was utilized widely as a crucial building block for the production of ethylene, diethyl ether, acetaldehyde, ethyl acetate, propylene, butadiene, butanol, aromatic alcohol and so on [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15]. The non-oxidative of ethanol dehydrogenation into acetaldehyde is regarded as a more promising synthetic route due to its excellent atomic utilization, high selectivity and ease of separation [9].

Copper-based catalysts have been widely recommended for ethanol dehydrogenation into acetaldehyde due to their good activity and relatively low price [9,10,11,12]. The supports can impact the Cu dispersion and valence states, thus adjusting the Cu properties and determining the catalytic performance. The commonly used supports include silica-based materials with relatively weak acidity [16,17,18,19,20,21,22], mixed metal oxides with both acidic and basic properties [23,24,25], and carbon materials with inert surface properties [26,27]. Shen and co-workers reported that copper catalysts supported on basic metal oxides (La₂O₂CO₃ and MgO) could efficiently catalyze the transfer dehydrogenation of primary aliphatic alcohols (i.e., the reaction coupling between primary aliphatic alcohol dehydrogenation and styrene hydrogenation through hydrogen transfer) due to the synergistic effect of basic sites and Cu nanoparticles [28,29]. Consequently, it is envisaged that good performance could be achieved over an MgO-supported Cu catalyst for ethanol dehydrogenation into acetaldehyde.

Inspired by the abovementioned work by Shen’s group [28,29], in this paper, we have developed a new catalyst system, Cu supported on MgO, for the selective dehydrogenation of ethanol into acetaldehyde. Surprisingly, the prepared Cu/MgO catalysts display a superior catalytic performance. The catalytic results are related to the characterization ones and compared with the SiO₂-supported Cu catalyst.

2. Results and Discussion

2.1. Structural and Textural Properties

The XRD patterns of a series of Cu/MgO catalysts with different Cu loadings are presented in Figure 1. The diffraction peaks at 2θ = 36.9°, 42.9°, 62.3°, 74.7° and 78.6° are assigned to the (111), (200), (220), (311) and (222) crystal facets of MgO (PDF #45-0946), while those at 2θ = 43.3°, 50.4° and 74.1° correspond to the Cu (111), Cu (200) and Cu (220) facets (PDF #04-0836). The absence of any diffraction peaks of Cu for the Cu/MgO catalysts with a Cu loading up to 15% implies that the Cu metal is highly dispersed on the MgO support. As the Cu loading is increased to 20%, the weak diffraction peak attributed to Cu crystallites was detected, suggesting that bulk Cu is formed at a high Cu loading. At the same Cu loading of 20%, the 20%Cu/SiO₂ catalyst shows the obviously stronger intensity of the metallic Cu diffraction peaks than 20%Cu/MgO, indicative of a larger Cu crystallite size for the former catalyst than the latter. According to the Scherrer equation, the size of the Cu crystallites in 20%Cu/MgO and 20%Cu/SiO₂ was estimated to be 11.4 and 24.4 nm, respectively. This result is an indication of the better dispersion of Cu on MgO than on SiO₂, although the SiO₂ support has a markedly higher surface area than MgO (381 vs. 95 m²/g), which signifies that the interaction between Cu and the MgO support is stronger than that between Cu and the SiO₂ support.

As the Cu loading is increased from 2.5% to 20%, the BET surface area of the Cu/MgO catalysts drops from 63 to 45 m²/g, and the Cu metal dispersion decreases from 52.6% to 26.1% (

Table 1. Physicochemical properties of the different catalysts.

Catalyst	SBET	Basicity (mmol/g)				DCu b	Coke c
	(m2/g)	Weak	Medium	Strong	Total	(%)	(g CHx/mmol Cu)
MgO	95	0.028	0.098	0.027	0.153	−	−
2.5%Cu/MgO	63	0.026	0.078	0.013	0.117	52.6	0.107
5%Cu/MgO	61	0.026	0.070	0.014	0.110	44.8	0.054
10%Cu/MgO	59	0.028	0.068	0.013	0.109	34.4	0.025
15%Cu/MgO	55	0.028	0.064	0.013	0.105	29.2	0.021
20%Cu/MgO	45	0.027	0.063	0.013	0.103	26.1	0.021
20%Cu/SiO2	201 a	0.013	0.017	0.006	0.036	3.6	0.008

^a The BET surface area of the SiO₂ support is 381 m²/g; ^b Cu dispersion; ^c for 2.5%, 5%, 10% and 15% Cu/MgO, reaction from 190 to 290 °C and holding at each temperature for 1 h. For 20%Cu/MgO and 20%Cu/SiO₂, reaction was performed for 48 h at 250 °C.

). This result suggests that the size of the Cu crystallites increases with an increase in the Cu loading. Compared to 20%Cu/MgO, the 20%Cu/SiO₂ catalyst possesses a much higher surface area (201 vs. 45 m²/g). Much higher Cu dispersion was achieved on 20%Cu/MgO than 20%Cu/SiO₂ with the same Cu loading (26.1% vs. 3.6%), which stems from the stronger interaction between Cu species and MgO than between Cu species and SiO₂. This result is consistent with the XRD observation.

The typical TEM images and corresponding elemental mapping of the 20%Cu/MgO and 20%Cu/SiO₂ catalysts are shown in Figure 2 and Figure 3. Analysis of the TEM images reveals the formation of Cu nanoparticles on the MgO and SiO₂ supports, with an average size of ca. 12 nm and 25 nm, respectively, which is consistent with the XRD result calculated using the Scherrer equation. Furthermore, according to the EDS elemental mapping, it can be observed that Cu is distributed uniformly over the MgO and SiO₂ supports. A uniform distribution of Cu for the 2.5–15%Cu/MgO catalysts can be rationally expected, because these catalysts have a higher dispersion of surface Cu than the 20%Cu/MgO catalyst.

2.2. Basic Properties

CO₂-TPD experiments were conducted to determine the surface basicity of the Cu/MgO catalysts. Figure 4 compares the CO₂-TPD curves of the MgO support and the Cu/MgO catalysts with different Cu loadings. All the samples display a broad desorption peak in the temperature range of 80–400 °C. These broad peaks can be deconvoluted into the desorption of CO₂ on weak (peak at ca. 140 °C), medium (peak at ca. 200 °C), and strong basic sites (peak at ca. 280 °C) [25,30,31], which correspond to OH groups, Mg²⁺-O²⁻ pairs, and O²⁻ ions, respectively [28,32]. As presented in Table 1, the Cu/MgO catalysts have more basic sites with medium strength. The number of basic sites present on the Cu/MgO catalysts decreases slightly with the increase in the Cu loading. Compared to MgO, the Cu/MgO catalysts obviously have less basic sites, indicating that the addition of Cu species reduces the number of basic sites on the MgO support.

2.3. XPS and TPR

XPS was used to analyze the chemical state of the Cu species in the 20%CuO/MgO and 20%CuO/SiO₂ precursors as well as the reduced 20%Cu/MgO and 20%Cu/SiO₂ catalysts. Figure 5 compares the Cu 2p XPS spectra of the samples. For the 20%CuO/MgO and 20%CuO/SiO₂ samples, the binding energies of Cu 2p_3/2 and 2p_1/2 appear at 933.4 and 953.1 eV, respectively, which fit with those of Cu²⁺ in the CuO species [33,34,35]. Moreover, an accompanying Cu 2p satellite peak centered at around 943 eV can be observed, which is characteristic of the occurrence of Cu²⁺ species. After the reduction of 20%CuO/MgO and 20%CuO/SiO₂ at 300 °C for 1 h under a flow of 10 vol% H₂/Ar, the Cu 2p satellite peak disappeared, indicating the absence of Cu²⁺ species in the reduced 20%Cu/MgO and 20%Cu/SiO₂ catalysts, i.e., Cu²⁺ species has been completely reduced. Moreover, the Cu 2p_3/2 peak of the reduced catalysts is shifted to a lower binding energy (BE) of 932.3 eV, which is assigned to Cu⁰ and/or Cu⁺ species [35,36,37,38,39]. Considering that Cu⁰ and Cu⁺ species present close Cu 2p_3/2 BE, in order to reveal the exact nature of the copper, we further performed Cu LMM X-ray-induced Auger electron spectroscopy (XAES) measurements. Figure 6 shows the Cu LMM XAES spectra of the reduced 20%Cu/MgO and 20%Cu/SiO₂ catalysts. The signal can be deconvoluted into two symmetric peaks. The peak with a kinetic energy (KE) of 918.3 eV is assigned to metallic Cu, while the one with a KE of 914.0 eV is attributed to Cu⁺ species [17,18,40,41]. Based on the peak area percentages, the Cu⁰ and Cu⁺ species in the 20%Cu/MgO and 20%Cu/SiO₂ catalysts were quantified to 81% and 19% as well as 88% and 12%, respectively. This observation indicates that both Cu⁰ and Cu⁺ species can be observed in the catalysts, and most Cu species are retained as Cu⁰. The 20%Cu/MgO catalyst possesses a lower population of Cu⁰ species than 20%Cu/SiO₂ (81% vs. 88%). The presence of Cu⁺ species can be attributed to the strong interaction between Cu species and the MgO and SiO₂ supports via the Cu−O−Mg and Cu−O−Si linkages [19,42,43].

To evaluate the reduction behavior of the Cu species dispersed on the MgO and SiO₂ supports, the CuO/MgO and CuO/SiO₂ precursors were examined by H₂-TPR experiments. As shown in Figure 7, there is only one reduction peak for all the CuO/MgO samples, with the peak temperature ranging from 210 to 237 °C, which is ascribed to the reduction of highly dispersed CuO species similar to those reported for other Cu-based catalysts [17,40,44,45]. Differently, for the 20%CuO/SiO₂ sample (Figure 7f), the TPR profile displays two peaks at 217 and 257 °C, which are assigned to the reduction of well-dispersed CuO with a small size and bulk CuO with a large size, respectively [18,46]. The reducibility of Cu species can be influenced by the particle size, support and preparation method. The TPR results suggest the better dispersion of CuO on MgO than on SiO₂, thereby leading to the better dispersion of Cu on MgO than on SiO₂ in the corresponding reduced Cu catalysts. These findings demonstrate the unique role of MgO in dispersing Cu nanoparticles.

2.4. Catalytic Performance

2.4.1. Effect of Cu Loading and Reaction Temperature

The catalytic performance of a series of Cu/MgO catalysts with different Cu loadings (2.5–20%) was tested in the temperature range of 190 to 290 °C to investigate the influence of the Cu loading and reaction temperature. As shown in Figure 8a, increasing the Cu loading from 2.5% to 15% leads to an increase in the ethanol conversion, which levels off with a further rise from 15% to 20% Cu. A higher Cu loading can provide more Cu sites for the reaction. The enhanced ethanol conversion on the Cu/MgO catalysts with Cu loading can be attributed to an improved number of Cu sites on the surface. As illustrated in Figure 9, the ethanol conversion on the Cu/MgO catalysts achieved at 230 °C is positively correlated with the number of surface Cu atoms.

At 230 °C, the productivity (expressed as the ethanol converted per gram of Cu per hour) drops from 26.8 to 4.8 g_ethanol/g_Cu/h as the dispersion of Cu particles on the MgO support is decreased from 52.6% to 26.1% (Figure 10). Small Cu particles (i.e., higher Cu dispersion) possess higher densities of coordinated unsaturated sites, such as corner and kink sites, which are more active than step or terrace sites [47]. Our finding is in accordance with the result for the Cu/SiO₂ catalysts reported by Zhang et al. [17].

With the reaction temperature increased from 190 to 290 °C, the ethanol conversion on the 10%Cu/MgO, 15%Cu/MgO and 20%Cu/MgO catalysts improves progressively. However, the 2.5%Cu/MgO and 5%Cu/MgO catalysts show a trend in ethanol conversion with the temperature that first increases and then decreases (Figure 8a). A similar phenomenon was reported by Zhang et al. for ethanol dehydrogenation over Cu/SiO₂ catalysts with 0.5% and 1% Cu loadings [17]. As shown in Figure 11, the ethanol conversion on 2.5%Cu/MgO at 270 °C diminishes from 64.8% to 44.6% after 1.5 h on stream. After regenerating the used catalyst by burning off the carbon species deposited on the catalyst in an air stream at 500 °C for 1 h followed by a reduction in the flow of 10 vol% H₂/Ar at 300 °C for 1 h, the activity of the 2.5%Cu/MgO catalyst is fully restored. This observation indicates that carbon deposition is the cause of the deactivation. Since the catalytic activity is dependent on the surface Cu sites, the deactivation caused by carbon deposition will be more severe for the samples with low Cu loadings than those with high loadings [48]. The amount of coke (CH_x) deposited on the catalyst after the reaction was quantified by TGA [17]. As presented in Table 1, obviously higher values are observed for 2.5%Cu/MgO and 5%Cu/MgO (0.107 and 0.054 g CH_x/mmol Cu, respectively) than 10% and 15% Cu/MgO (0.021–0.025 CH_x/mmol Cu) after several hours of the reaction, suggesting that the coke formation on Cu/MgO is pronounced at low Cu loadings. Thus, it can be concluded that the decrease in activity at higher temperatures for 2.5%Cu/MgO and 5%Cu/MgO can be attributed to the catalyst deactivation caused by carbon deposition.

The main reaction product is acetaldehyde, with the minor by-products being ethyl acetate, acetic acid, acetone, butyraldehyde and ethylene. The acetaldehyde selectivity for all the Cu/MgO catalysts is >97% at low temperatures, and it declines slightly with the temperature due to the generation of more ethyl acetate (Figure 8b). The selectivity to acetaldehyde can still reach higher than 95% at high temperatures.

2.4.2. Effect of WHSV

The dependence of the activity and selectivity on the space velocity was also investigated at 250 °C over 20%Cu/MgO as a representative catalyst. The results are shown in Figure 12. With the increment of the WHSV from 0.5 h⁻¹ to 6 h⁻¹, the conversion of ethanol obviously declines from 87.5% to 44.4%. The selectivity of acetaldehyde enhances from 94.7% at 0.5 h⁻¹ to 97.7% at 1.5 h⁻¹, followed by a slight increase to 98.2% at 2.5 h⁻¹, and then levels off above 2.5 h⁻¹.

2.4.3. Comparison of Performance for Cu/MgO with Cu/SiO₂ in a Prolonged Time

The 20%Cu/MgO catalyst was chosen for comparison because it exhibits higher ethanol conversion than the other catalysts (2.5–15%Cu/MgO) due to its higher number of surface Cu sites. Considering that silica-supported Cu is a conventional catalyst for the non-oxidative dehydrogenation of ethanol, we compared the catalytic performance of the 20%Cu/MgO and 20%Cu/SiO₂ catalysts with the same Cu loading at 250 °C over a prolonged reaction time. The results are presented in Figure 13. For the 20%Cu/MgO catalyst, as the reaction of ethanol dehydrogenation goes on, the conversion of ethanol diminishes from 81.5% at 1 h to 72.6% at 10 h and to 61.9% at 24 h, which decreases relatively by ca. 24%. After regeneration of the spent catalyst in flowing air at 500 °C for 1 h, the activity in the initial 10 h can be completely restored, followed by a fast deactivation where the ethanol conversion drops from ca. 73% at 10 h to ca. 52% at 24 h. A relative reduction in the ethanol conversion of ca. 36% is achieved in the second run. This finding suggests that the reason for the catalyst deactivation for 20%Cu/MgO in the first run (24 h) as well as in the initial 10 h of the second run is carbon deposition. The faster deactivation between 10 h and 24 h observed in the second run than in the first run is caused by sintering of the Cu particles. An increase in the intensity of the diffraction peaks of the Cu phase can be found for the used 20%Cu/MgO catalyst compared to the fresh one (Figure 14), signifying the growth of the Cu particles after the second run (from 11.4 to 15.0 nm, as determined by the Scherrer equation).

For the 20%Cu/SiO₂ catalyst, the ethanol conversion decreases from 87.7% at 1 h to 52.8% at 24 h, corresponding to a relative reduction of ca. 40%, which is indicative of a faster catalyst deactivation for 20%Cu/SiO₂ than 20%Cu/MgO. After pretreatment of the used catalyst in an air flow at 500 °C for 1 h, the ethanol conversion on the regenerated catalyst is 84.7% at 1 h, which is lower than the initial conversion, indicating that the original activity of 20%Cu/SiO₂ cannot be fully restored. This result suggests that, in addition to carbon deposition, there is another reason for the catalyst deactivation in the first run of 20%Cu/SiO₂, i.e., the growing of the Cu particles. The ethanol conversion on 20%Cu/SiO₂ in the second run drops from 84.7% at 1 h to 47.0% at 24 h, which drops relatively by ca. 45%. The intensity of the diffraction peaks of metallic Cu on the spent 20%Cu/SiO₂ catalyst is stronger than that on the fresh one (Figure 14), implying the aggregation of the Cu particles after reuse for 2 runs (from 24.4 to 33.0 nm, as determined by the Scherrer equation). Taking into account that less coke is formed on 20%Cu/SiO₂ than 20%Cu/MgO (0.008 vs. 0.021 g CH_x/mmol Cu), the quicker catalyst deactivation observed for the former catalyst than the latter one is attributed to the more severe Cu sintering for 20%Cu/SiO₂ as a consequence of the weaker interfacial interaction between Cu species and the SiO₂ support than between Cu species and the MgO support.

As the ethanol dehydrogenation reaction goes on, the acetaldehyde selectivity on 20%Cu/MgO increases slightly from 97.7% to 98.8%, whereas that on 20%Cu/SiO₂ improves from 91.6% to 96.4%. Thus, both catalysts exhibit a comparable initial acetaldehyde yield (ca. 80%). More ethylene and ethyl acetate, which are derived from ethanol dehydration and the coupling between ethanol and acetaldehyde, respectively [36,49], are formed on 20%Cu/SiO₂, leading to lower selectivity to acetaldehyde for 20%Cu/SiO₂ than 20%Cu/MgO. The generation of more ethylene and ethyl acetate on 20%Cu/SiO₂ might be related to the weak acid sites present on SiO₂, i.e., surface—OH groups [26]. In summary, compared to 20%Cu/SiO₂, the 20%Cu/MgO catalyst displays an equivalent initial acetaldehyde yield, higher acetaldehyde selectivity and longer stability.

3. Materials and Methods

3.1. Reagent and Materials

The Mg(NO₃)₂·6H₂O, (NH₄)₂C₂O₄·H₂O and ethanol were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). The Cu(NO₃)₂·3H₂O was purchased from Macklin Biochemical Co., Ltd. (Shanghai, China). The SiO₂ support was purchased from Qingdao Haiyang Chemical Co., Ltd. (Qingdao, China). All the chemicals were of analytical grade and were used without further purification.

3.2. Catalyst Preparation

The MgO support was synthesized via a precipitation method using Mg(NO₃)₂·6H₂O and (NH₄)₂C₂O₄·H₂O as a precursor and a precipitating agent, respectively. In a typical synthetic procedure, 50 mL of 0.8 M Mg(NO₃)₂ aqueous solution was added dropwise to 200 mL of 0.3 M (NH₄)₂C₂O₄ aqueous solution under stirring, followed by aging at room temperature for 12 h. The obtained white precipitate was washed with deionized water, dried at 110 °C overnight, and then calcined at 600 °C in an air stream for 6 h. The Cu/MgO catalysts with various Cu loadings (2.5%, 5%, 10%, 15% and 20%, by weight) were prepared by a conventional impregnation method. Different amounts of Cu(NO₃)₂·3H₂O were dissolved in deionized water, and then a certain amount of MgO was added. After drying under an infrared lamp, the samples were dried at 110 °C overnight, and then calcined at 550 °C in an air stream for 6 h to yield the CuO/MgO precursors. The CuO/MgO precursors were reduced at 300 °C for 1 h under a flow of 10 vol% H₂/Ar to yield the x%Cu/MgO catalysts, where x% represents the weight percentage of Cu in the catalysts. Accordingly, the corresponding CuO/MgO precursors were labelled as x%CuO/MgO, where x% represents the weight percentage of Cu in the final Cu/MgO catalysts. For comparison, the 20%Cu/SiO₂ catalyst was prepared in the same way but employing SiO₂ as the support.

3.3. Catalyst Characterization

The X-ray diffraction (XRD) patterns were acquired on a D2 PHASER diffractometer (Brucker, Madison, WI, USA) using Cu Kα radiation at 30 kV and 10 mA. The BET surface areas were determined by N₂ adsorption at −196 °C using a Micromeritics Tristar 3020 apparatus (Micromeritics, Atlanta, GA, USA). Before the measurement, the catalyst was pretreated at 300 °C under a vacuum for 5 h. The transmission electron microscopy (TEM) images and high-angle annular dark field scanning transmission electron microscopy (HAADF-STEM), together with the elemental mapping, were obtained on an FEI Tecnai G² F20 S-TWIN microscope (FEI, Hillsboro, OR, USA). The X-ray photoelectron spectroscopy (XPS) measurements were carried out on a Perkin-Elmer PHI 5000C spectrometer (Perkin-Elmer, Waltham, MA, USA) using an Mg Kα radiation source. All the binding energy values were referenced to the C 1s peak at 284.6 eV. The thermogravimetric analysis (TGA) was conducted in an air stream on a Perkin-Elmer TGA8000 instrument (Perkin-Elmer, Waltham, MA, USA) for measuring the amount of coke deposited on the spent catalyst.

The basicity of the catalysts was determined via the temperature-programed desorption of CO₂ (CO₂-TPD) on an AutoChem II apparatus (Micromeritics, Atlanta, GA, USA). The catalyst (0.1 g, 40–60 mesh) was pretreated at 500 °C in He flow for 1 h. Afterwards, CO₂ was adsorbed at 80 °C for 2 h under 5 vol% CO₂/He flow (30 mL/min), followed by purging with He (30 mL/min) for 2 h. At last, the catalyst was heated to 500 °C with a ramp of 10 °C /min in an He flow (30 mL/min), and the desorption curve was recorded by a thermal conductivity detector. The H₂ temperature-programed reduction (H₂-TPR) characterization was carried out on a Micromeritics AutoChem II apparatus (Micromeritics, Atlanta, GA, USA). Before the TPR measurement, 0.1 g of the CuO/MgO precursor (40–60 mesh) was pretreated at 300 °C in an He flow for 1 h. After cooling to 60 °C in an He flow, the flow was changed to 10 vol% H₂/Ar (30 mL/min), and then the temperature was raised to 500 °C with a ramp of 10 °C /min. The dispersion of Cu was determined by N₂O oxidation and then by H₂ titration [50]. In detail, 0.1 g of the CuO/MgO precursor (40–60 mesh) was first reduced at 300 °C for 1 h under a flow of 10 vol% H₂/Ar (30 mL/min). After cooling to 60 °C in an He flow, the flow was switched to an N₂O flow (30 mL/min) and maintained at 60 °C for 0.5 h to oxidize the surface Cu atoms to Cu₂O, followed by flushing with He (30 mL/min) for 1 h to remove any residual N₂O. Finally, another H₂-TPR experiment was conducted as above, and the hydrogen consumption was recorded as Y. The hydrogen consumption in the first H₂-TPR experiment was denoted as X. The Cu dispersion was calculated using the following equations:

Reduction of all the Cu atoms:

CuO + H₂ = Cu + H₂O, hydrogen consumption = X

Reduction of the surface Cu atoms only:

Cu₂O + H₂ = 2Cu + H₂O, hydrogen consumption = Y

The dispersion of the surface Cu was calculated by the following formula:

$$\mathrm{D}(\%)={\displaystyle \frac{2\mathrm{Y}}{\mathrm{X}}}\times 100\%$$

3.4. Catalytic Test

Ethanol dehydrogenation was carried out in a flow fixed-bed microreactor under ambient pressure. The CuO/MgO precursor (0.3 g, 40–60 mesh) was in situ reduced at 300 °C for 1 h in a flow of 10 vol% H₂/Ar. After the reactor was cooled down to the desired reaction temperature, the flow was changed to the feed gas consisting of 5 vol% ethanol and 95 vol% N₂. Ethanol was introduced into the reactor using a bubbler. The products were analyzed online by a gas chromatograph equipped with a flame ionization detector and a FFAP capillary column (30 m × 0.32 mm × 0.25 μm). The conversion of ethanol, selectivity and yield of acetaldehyde were calculated according to the following formulas:

$$\mathrm{Conversion}(\%)={\displaystyle \frac{{\left[\mathrm{ethanol}\right]}_{\mathrm{in}}{-\left[\mathrm{ethanol}\right]}_{\mathrm{out}}}{{\left[\mathrm{ethanol}\right]}_{\mathrm{in}}}}\times 100\%$$

$$\mathrm{Selectivity}(\%)={\displaystyle \frac{{\left[\mathrm{acetaldehyde}\right]}_{\mathrm{out}}}{{\left[\mathrm{ethanol}\right]}_{\mathrm{in}}-[{\mathrm{ethanol}]}_{\mathrm{out}}}}\times 100\%$$

$$\mathrm{Yield}(\%)={\displaystyle \frac{{\left[\mathrm{acetaldehyde}\right]}_{\mathrm{out}}}{{\left[\mathrm{ethanol}\right]}_{\mathrm{in}}}}\times 100\%$$

4. Conclusions

In this work, a series of Cu/MgO catalysts with different Cu loadings (2.5–20%) were prepared by an impregnation method. The as-synthesized Cu/MgO catalysts display a superior catalytic performance for the non-oxidative dehydrogenation of ethanol into acetaldehyde. The ethanol conversion is positively correlated with the number of surface Cu atoms. The productivity (expressed as the ethanol converted per gram of Cu per hour) is positively correlated with the dispersion of Cu particles on the MgO support. The 20%Cu/MgO catalyst affords an initial 81.5% conversion, with 97.7% acetaldehyde selectivity, at a WHSV of 1.5 h⁻¹ and 250 °C. In comparison with 20%Cu/SiO₂, the 20%Cu/MgO catalyst exhibits an equivalent initial acetaldehyde yield, higher acetaldehyde selectivity and longer stability. The faster catalyst deactivation observed for the former catalyst than the latter one is attributed to the more severe Cu sintering for 20%Cu/SiO₂ as a consequence of the weaker interfacial interaction between Cu species and the SiO₂ support than between Cu species and the MgO support.