Abstract

The bimetal catalyst Pt-Cu/SiO2 was prepared by the impregnation method. Its catalytic performance was investigated by the selective hydrogenation of cinnamaldehyde. Pt-Cu/SiO2 exhibited much higher selectivity (64.1%) to cinnamyl alcohol than Pt/SiO2 (3.7%), while they showed similar conversion of cinnamaldehyde. This enhancement was attributed to the increase in the amount of the Pt0 species on the Pt-Cu/SiO2 surface, which is derived from the interaction between Pt and Cu revealed by XRD and XPS.

1. Introduction

Selective catalytic hydrogenation of α,β-unsaturated aldehydes is an important step in industrial production of fine chemicals, such as pharmaceuticals, flavours, and fragrances [15]. Cinnamaldehyde (CAL) is representative α,β-unsaturated aldehydes, which can be hydrogenated first via C=C or C=O double bonds to hydrocinnamaldehyde (HALD) or cinnamyl alcohol (COL), followed by total hydrogenation of these products to hydrocinnamyl alcohol (HALC) (Scheme 1) [1]. In general, the reduction of the C=C group is favored thermodynamically [1, 2]. Thus, it is still a challenging task to enhance the selectivity for the hydrogenation of C=O group to form COL [2, 6, 7].

Nowadays, Pt-based catalyst facilitates the production of COL via the hydrogenation of the C=O bond [812]. The unique advantage of such catalysts is that activity and selectivity can be tuned by controlling the size, shape, and surface species of Pt particles [1, 1315]. In fact, many methods have been developed to enhance the performance of Pt-based catalyst, such as investigating novel support and promoter [1519]. In our previous works, it was found that graphene-supported Pt catalysts showed high selectivity to COL [13]; and the activity and selectivity of Pt/carbon nanosheets (CNS) catalysts increased with the rise of the graphitization degree of CNS [20]. Moreover, the selectivity of Pt catalyst does not depend on particle size, at least within the 1.5–2.5 nm particle size range over Pt/SiO2 [21]. We also found that Mo2N could enhance the efficiency of Pt on ternary catalyst Pt-Mo2N/SBA-15 due to the synergistic effect of Pt and Mo2N [22]. Furthermore, some other studies illustrated that the introduction of a second metal promoter was also a valid way to improve the catalytic properties of Pt nanoparticles. It has been reported that metal promoters (Ga, Fe, Ni, and Ru) play an important role in CAL hydrogenation [7, 23, 24].

In light of the above, a thought stimulates us to investigate whether the nonnoble metal of Cu could promote the catalytic performance of Pt for the hydrogenation of cinnamaldehyde. Thus, the Pt-Cu bimetal catalyst on SiO2 aerogel (Pt-Cu/SiO2) was prepared by a simple impregnation method, and its performance on selective hydrogenation of cinnamaldehyde was investigated. To our delight, the Pt-Cu/SiO2 catalyst showed much higher selectivity than Pt/SiO2. It was found that the addition of Cu enhanced the dispersion of Pt on SiO2. Moreover, there were more Pt0 species on bimetal catalyst, which benefitted the enhancement of the selectivity to COL.

2. Material and Method

2.1. Catalyst Preparation

SiO2 aerogel was purchased from Aladdin. Bimetal catalyst was synthesized by impregnation and hydrogen reduction method. Firstly, 0.4 g Cu(NO3)23H2O was dissolved in 20 mL water at room temperature, and then 2 mol/L ammonia water was dropped to the mixture until there was no precipitation. Secondly, 1.00 g SiO2 aerogel was added to the above solution. After stirring for 4 h, the blue powder was collected by filtration and washing and dried at 80°C. Thirdly, 0.300 g blue powder was impregnated with 1.5 mL H2PtCl6 (10 mmol/L) ethanol solution. The resulting suspension was stirred at room temperature to remove the solvent and dried at 80°C for 12 h. Finally, the sample was placed inside a furnace with a heating rate of 5°C min−1 to 300°C under N2 atmosphere which was maintained at 300°C for 1 h under pure H2 atmosphere. Thus, the obtained sample was denoted as Pt-Cu/SiO2. As references, Pt supported on SiO2 aerogel sample was also prepared under similar condition, which was denoted as Pt/SiO2.

2.2. Catalyst Characterization

The X-ray diffraction (XRD) patterns were performed by using a Bruker D8 Advance X-ray diffractometer with Cu Kα ( = 1.5418 Å) radiation (40 kV, 40 mA). The scan range of the wide-angle XRD was 10–80°. Based on XRD results, the average particles sizes can be calculated from Pt(111) by Scherrer’s equation. The average particles sizes were recalculated to dispersion values assuming spherical shapes and using the formula described by Scholten et al.: is dispersion (/), is the atomic weight (195.1 g/mol for Pt), is the platinum surface site density (12.5 Pt atoms/nm2), is particle size (nm), is the metal density (21.45 g/cm3 for Pt), and is the Avogadro constant. Thus, = 1.13/d (nm) for Pt [25, 26].

Properties of surface area were derived from N2 adsorption-desorption isotherms at 77 K with a Micromeritics Tristar II surface area and a porosimetry analyzer. Before the measurement, samples were degassed under vacuum at 150°C for 5 h. The surface area was determined from desorption isotherms by the BET method.

The morphology and structure of the samples were analyzed by using a JEM-2100 transmission electron microscopy (TEM) with an acceleration voltage of 200 kV.

X-ray photoelectron spectroscopy (XPS) was conducted on a Kratos-AXIS ULTRA DLD with an AlKα radiation source. Binding energies (BE) refer to the C (1s) binding energy of carbon taken to be 284.7 eV.

The content of Pt and Cu on the Pt-Cu/SiO2 catalyst was determined by inductively coupled plasma-optical emission spectrometry (ICP-OES), which was performed by using a PerkinElmer Optima 7000DV analyzer. Each sample was dissolved in a diluted HF and chloroazotic acid solution before measurement.

2.3. Selective Hydrogenation of Cinnamaldehyde

The catalyst (50 mg) was mixed with 1.00 g CAL and 30 mL isopropanol. The mixture was subsequently transferred to a 100 mL autoclave. After flushing three times with N2 at 5 bar and three times with H2 at 5 bar, the reaction was allowed to proceed at 80°C under 10 bar of H2 for 2 h.

The products were analyzed by GC (Agilent 7820 A) with flame ionization detection (FID) and a HP-5 capillary column (30 m × 0.32 mm × 0.25 mm). Quantitative analysis of the products was performed via calibration curves using the benzyl alcohol as an external standard. The products were further identified by GC-MS (Agilent 6890/5973N).

3. Results and Discussion

3.1. Characterization of Catalysts

Figure 1 shows the XRD patterns of SiO2, Pt/SiO2, and Pt-Cu/SiO2 samples. All of them displayed a diffuse peak of amorphous SiO2 around 22.5°. For Pt/SiO2, a series of characteristic diffraction lines of platinum at 39.5, 46.4, and 67.4° were observed, which was identified as the (111), (200), and (220) reflections of Pt with face-centered cubic structure [22]. According to Scherrer’s equation, the grain size of Pt nanoparticles on the Pt/SiO2 sample was about 21.54 nm (Table 1). Compared with the Pt diffraction peak of the Pt/SiO2 sample, the Pt (111) diffraction peak of the Pt-Cu/SiO2 sample becomes broadened and shifts to 41.1°, and the reflection of Pt (200) and (220) apparently disappears. Meanwhile, the diffraction peaks assigned to the Cu species are not observed, which may be ascribed to the little amount of Cu (0.33 wt%, Table 1) measured by ICP.

Comparing the XRD patterns of the Pt-Cu/SiO2 (Figure 1(c)) with those of the Pt/SiO2 (Figure 1(b)), on one hand, the broadening peak of Pt (111) and the disappearance of the Pt (200) and Pt (220) peaks for the Pt-Cu/SiO2 imply that the addition of Cu is able to decrease the size of Pt particles and improve the dispersion of Pt. In view of the calculation using Scherrer’s equation, the Pt particle size for the Pt-Cu/SiO2 sample (11.05 nm) was much smaller than that of the Pt/SiO2 (21.54 nm), which also implies that the addition of Cu benefits gaining relatively small Pt particles. On the other hand, the diffraction peak of the Pt (111) shifted to higher angle with the addition of Cu. It is likely that the Cu atom is incorporated into the crystal lattice of Pt and then occupies the position of Pt atoms. The atomic radius of Cu (0.128 nm) is a little smaller than Pt (0.139 nm); the substitution of Cu for Pt results in the crystal cell of Pt shrinking and the parameters of crystal cell decreasing. The difference of XRD patterns illustrates that there are interactions between Pt and Cu in the Pt-Cu/SiO2 sample.

The textural properties of the samples were investigated by N2 adsorption-desorption isotherms. The BET specific surface area is summarized in Table 1. It was worthwhile to note that the surface area significantly decreased after metal modification. The BET surface area of the Pt/SiO2 decreased to 364 m2 g−1 from 550 m2 g−1 for SiO2. For the Pt-Cu/SiO2, the corresponding surface area decreased to 410 m2 g−1. This indicated that Pt and Cu were loaded on the SiO2 support and occupied its surface. According to the results revealed by XRD patterns, the particle size of Pt-Cu/SiO2 was smaller than that of Pt/SiO2; therefore the surface area of the Pt-Cu/SiO2 was slightly larger than that of the Pt/SiO2.

Figure 2 shows the TEM images of Pt/SiO2 and Pt-Cu/SiO2. Unfortunately, we failed to take the HRTEM images of the samples, maybe due to the wrapping of SiO2 aerogel. From Figures 2(a)2(c), it can be seen that the dispersion of Pt nanoparticles on Pt/SiO2 catalysts was poor. The Pt nanoparticles size was not uniform, and the aggregation is serious (Figure 2(b)). Thus, it is difficult to gain its average particle size determined by statistical analysis. With regard to the Pt-Cu/SiO2 sample in Figures 2(d)-2(e), the Pt nanoparticles were highly dispersed on SiO2 supporter with uniform size ( = 10.25 ± 0.30 nm), which was in line with the XRD results. Although the average particle size of the Pt/SiO2 was not available, it is obvious that the Pt-Cu/SiO2 sample exhibits much smaller particle size and far better dispersion. The observations disclose that the addition of Cu helps to disperse Pt and effectively avoids agglomeration of Pt.

XPS was conducted to investigate the surface properties and the chemical states of Pt. Figure 3 shows the Pt 4f XPS of Pt/SiO2 and Pt-Cu/SiO2 together with corresponding deconvolution by fitting Gaussian peaks after Shirley-background subtraction. As shown in Figure 3, the Pt XPS of Pt/SiO2 and Pt-Cu/SiO2 presented a doublet corresponding to Pt 4f7/2 and Pt 4f5/2, and the corresponding binding energy (BE) is listed in Table 2. As for the Pt/SiO2 sample, the Pt 4f7/2 BE at 71.14 eV was ascribed to platinum in the metallic state (Pt0); the Pt 4f7/2 peak around 73.34 eV could be assigned to the Pt2+ species and the Pt 4f7/2 peak at 75.52 eV was attributed to the Pt4+ species. But there were only two kinds of Pt species in the Pt-Cu/SiO2 sample. As shown in Figure 3(b), the peak located at 70.93 eV was attributed to Pt0, while the peak around 72.99 eV corresponded to the Pt2+ species. Interestingly, the BE of Pt0 and Pt2+ species for the Pt-Cu/SiO2 sample exhibited a negative shift compared with that for Pt/SiO2 sample. This shift should be attributed to the electrons transfer from Cu to Pt by their intimate contact. In other studies [10, 11, 22, 23, 28], this electron transfer phenomenon between precious and promoter was also found, and the direction of this transfer changed with the nature of precious metal and promoter.

Furthermore, what is noticeable is the difference in the quantitation of the surface Pt species. As listed in Table 2, Pt0, Pt2+, and Pt4+ species simultaneously existed in the Pt/SiO2 sample and the corresponding proportion was 28%, 53%, and 19%, respectively. Obviously, for the Pt/SiO2, metal Pt exists mostly in the form of the Pt2+ species.

As far as the Pt-Cu/SiO2 was concerned, there were only Pt0 and Pt2+ species, and the ratios were 60% and 40%, respectively. Compared with the Pt/SiO2, the Pt-Cu/SiO2 contains more Pt0 content and less Pt4+ amount. This suggests that the strengthening of metallic characteristics of Pt in Pt-Cu/SiO2 may be attributed to the presence of the transition metal Cu.

3.2. Catalytic Performance

The selective hydrogenation of CAL was applied to evaluate the catalytic performance of the Pt/SiO2 and Pt-Cu/SiO2, and the results are shown in Table 3. It can be seen that SiO2 and Cu/SiO2 showed little activity to the selective hydrogenation of CAL. And Pt/SiO2 showed 38.3% conversion of CAL, and the selectivity to COL, HALD, HALC, and other products was 3.7%, 1.3%, 35.4%, and 59.6%, respectively. 10.2% CAL could be converted over Cu/SiO2, and the selectivity to COL, HALD, and HALC was 1.4%, 5.9%, and 9.3%, respectively. However, the Pt-Cu/SiO2 catalyst performed 37.9% CAL conversion and 64.1% selectivity to COL, while the selectivity to HALC, HALD, and other side products was 5.4%, 11.3%, and 19.2%, respectively. And the TOF of Pt to form COL increased to 3076 h−1 from 329 h−1 for the Pt/SiO2. Therefore, it is clear that the Pt-Cu/SiO2 catalyst displays much higher selectivity to COL than Pt/SiO2, while they showed similar conversion of CAL. This result is different from a past report supporting Pt and Cu on carbon nanotube by microwave-assisted polyol reduction [23]. The different results imply that the interaction between Pt and promoter may be different due to the distinctive preparation strategies. In this work, the enhanced selectivity to COL caused by promoter Cu could be attributed to the increased fraction of the metallic Pt0 species on the surface of the Pt-Cu/SiO2 catalyst. According to our previous studies, the amount of Pt0 on the catalyst has a much larger influence on the selectivity of COL, while the Pt2+ and Pt4+ species are responsible for the high-molecular-weight secondary products [13, 2022]. For Pt/SiO2, there are more Pt2+ and Pt4+ species and less Pt0 species, so it exhibits less selectivity to COL and higher selectivity to those condensation products. On the other hand, the Pt-Cu/SiO2 catalyst possesses more Pt0 species and less species on its surface; thus it displays much higher selectivity to COL than Pt/SiO2.

4. Conclusions

The Pt-Cu bimetal catalyst was prepared on SiO2 aerogel. The Pt-Cu/SiO2 showed uniform Pt NPs benefitted from the introduction of Cu. The intensive interaction between Pt and Cu was observed in the Pt-Cu/SiO2 catalyst, which results in the difference of Pt species on its surface and the electron transfer between Pt and Cu. The Pt-Cu/SiO2 catalyst exhibited a remarkably enhanced selectivity to cinnamyl alcohol for the liquid-phase hydrogenation of cinnamaldehyde. This promotion can be ascribed to the fact that Cu increases the amount of the Pt0 species in the surface of the Pt-Cu/SiO2 catalyst.

Conflicts of Interest

The authors declare no conflicts of interest.

Acknowledgments

This work was supported by Natural Sciences Fund of Heilongjiang Province (B2015009) and the Ph.D. Fund of Hanshan Normal University (no.QD20170503).