Selective hydrogenation of cinnamaldehyde over carbon nanotube ...

Jointly published by Akadémiai Kiadó, Budapest and Springer, Dordrecht

React.Kinet.Catal.Lett. Vol. 88, No. 2, 269−275 (2006)

RKCL4806 SELECTIVE HYDROGENATION OF CINNAMALDEHYDE OVER CARBON NANOTUBE SUPPORTED Pd-Ru CATALYST Jieshan Qiua*, Hongzhe Zhanga, Xiuna Wanga, Hongmei Hana, Changhai Lianga,c and Can Lib a

State Key Laboratory of Fine Chemicals, Carbon Research Laboratory, School of Chemical Engineering, Center for Nano-Materials and Science, Dalian University of Technology, 158 Zhongshan Road, P.O .Box 49, Dalian 116012, China, b State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, CAS, Dalian, China, c Laboratory of Industrial Chemistry, Ruhr University Bochum, 44780 Bochum, Germany

Received May 27, 2005, in revised form September 8, 2005, accepted September 13, 2005

Abstract Carbon nanotube supported Pd, Ru and Pd-Ru catalysts have been prepared and tested with the hydrogenation of cinnamaldehyde as a probe reaction. It has been found that the cinnamaldehyde conversion and the selectivity towards the hydrogenation of C=O bond over Pd-Ru/PCNT catalyst could reach 56.6% and 79.1%, respectively, at 120oC and 5.0 MPa, which is better than Pd/PCNT and Ru/PCNT catalysts under the same reaction conditions. It is assumed that the better performance of Pd-Ru/PCNT catalyst for cinnamaldehyde hydrogenation may be due to the synergic effect of Pd and Ru metals or the promoting effect of Ru metal. Keywords: Carbon nanotubes, hydrogenation of cinnamaldehyde, palladium and ruthenium

___________________________ * Corresponding author. Fax: +86-411-88993991; E-mail: [email protected]

0133-1736/2006/US$ 20.00. © Akadémiai Kiadó, Budapest. All rights reserved.

270

JIESHAN QIU et al.: CARBON NANOTUBES

INTRODUCTION The selective catalytic hydrogenation of cinnamaldehyde (CAL) to cinnamyl alcohol (COL) is an important reaction both from academic and industrial point of view because COL is an intermediate or precursor for synthesis of value-added fine chemicals and medicines. In the traditional catalytic process for hydrogenating CAL, the C=C double bond would be hydrogenated more easily than the C=O double bond, leading to low yields of the desired product COL [1], which should be inhibited or avoided to a great extent. Up to now, a number of studies have been performed with an aim of further improving the performance of hydrogenation catalysts, in which the effects of supports [2-5] and metals [6,7] have been addressed. It has been found that some metals are capable of enhancing the hydrogenation rate of the C=O double bond, at the same time, to suppress or inhibit the hydrogenation of the C=C double bond [8]. It is also known that the selectivity to unsaturated alcohols is highly dependent on the property of the metal catalysts and the support used [9]. Of the supports tested, carbonaceous supports are found to be better than the traditional oxide supports in terms of catalytic activity [9]. Carbon nanotube (CNT) is a novel carbon material with unique structure and properties, and has drawn much attention as a potential catalyst support from a viewpoint of both fundamental research and industrial uses [10-11]. It has been found that the CNTs with tubular morphology have a well-graphitized structure and good electronic conductivity [12]. Some previous reports about the hydrogenation of CAL over the CNT-supported catalysts have revealed that the CNT-supported catalysts show high conversion and high selectivity [3-4], in which metal Ru catalysts were mainly focused. In the present work, the CNT supported bi-metallic (Pd-Ru) catalysts are made and tested for the hydrogenation of cinnamaldehyde. The results show that the bi-metallic Pd-Ru/CNTs catalysts are more active and show higher selectivity to COL in comparison to CNT-supported mono-metal catalysts. EXPERIMENTAL The raw carbon nanotube (RCNT) was prepared by the arc discharge method [10], and was treated by refluxing in 65% HNO3 at 120oC for 8 h to get purified nanotube (PCNT). Pd/CNT or Ru/CNT catalysts were prepared by incipient wetness impregnation technique using an aqueous solution of palladium nitrate or RuCl3. After impregnation treatment, the samples were dried overnight at 110oC and calcined in air at 300oC for 3 h. The catalysts were reduced in flowing hydrogen at 400oC for 6 h and then passivated in a flowing gas of O2:N2 (1:99 in volume) at room temperature for 3 h before use. Bi-metallic Pd-Ru/PCNT catalysts were made with the Pd/PCNTs as the starting materials, following the same procedure described above for making Pd/PCNTs or


271

Ru/CNT catalysts. To be specific, the as-prepared Pd/PCNTs were further impregnated in a solution of RuCl3, then dried overnight at 110oC, and finally calcined and reduced to get bi-metallic Pd-Ru/PCNT catalysts. The raw CNTs and the purified CNTs as well as the CNT-supported catalysts were examined using transmission electron microscopy (TEM, JEOL JEM-2000EX operated at 100 kV) and XRD (D/max-2400, Cu Kα radiation, operated at 40 kV and 100 mA). The metal content in catalyst was analyzed using inductively coupled plasma emission spectrometry (ICP-OES). The hydrogenation of cinnamaldehyde was carried out at 5.0 MPa and 120oC in a magnetically stirred stainless steel autoclave reactor with an inner volume of 300 mL. For each run, the reactor was loaded with 120 mL of solvent (ethanol) containing 12 mL of cinnamaldehyde and 100 mg of catalyst. The hydrogenation products were collected from the reactor at regular intervals through a sampling outlet, and were analyzed using GC-MS (HP6890GC/5973MSD) equipped with a HP-5 capillary column and FID. RESULTS AND DISCUSSION Figure 1a shows the typical TEM image of the CNTs used in the present study. The outer diameters of the CNTs are in a range of 6.0-50.0 nm with the inner diameters being in a range of 3.0-5.0 nm. The BET surface area of the CNTs, determined by nitrogen adsorption at 77 K, is about 22.5 m2/g. Figure 2 shows the XRD patterns of the CNT supported catalysts, in which a prominent peak at 2θ = 26.6º that corresponds to that of graphite can be clearly seen. Both the TEM and XRD studies confirm that the CNTs used in this work are highly graphitic and have a low content of amorphous carbons.

Fig. 1. TEM image of PCNTs (a), Pd-Ru/PCNTs (b)

272


The catalytic performances of the Pd/RCNT (Pd loading of 5.0 wt.%), Pd/PCNT (Pd loading of 5.0 wt.%), Ru/PCNT (Ru loading of 2.0 wt.%) and Pd-Ru/PCNT (Pd and Ru loading of 5.0 wt.%, 1.0 wt.%, respectively) catalysts for the hydrogenation of cinnamaldehyde (CAL) are shown in Table 1. In the case of Pd/RCNT catalyst, the conversion of CAL increases slowly with the reaction time, from 10.3% at 30 min to 20.5% at 240 min. While in the case of the Pd/PCNT catalyst, as the reaction time increases, the conversion of CAL increases quickly from 5.7% at 30 min to 32.5% at 240 min, however, the selectivity of COL becomes poor with time increasing. For both the Pd/RCNT catalysts and the Pd/PCNT catalysts, the COL yield is quite low, which is still lower than 7.0% even after 240 min of the reaction. While in the case of Ru/PCNT catalyst, a better catalytic performance has been observed, which is evidenced by the fact that over 20.0% of cinnamaldehyde has been converted after 120 min and the selectivity to COL has reached 78.7%, implying clearly that the Ru catalyst favors the hydrogenation of C=O bond in CAL [9]. In the case of bi-metallic Pd-Ru/PCNT catalysts in which the added metal Ru is expected to function as promoter, both the conversion of CAL and the selectivity to COL under identical conditions is higher in comparison to the CNT-supported mono-catalysts. Obviously, the presence of Ru can help to improve the catalytic activity of the Pd/PCNT catalyst. Because of the promoting effect of Ru metal, the COL yield reaches 18.3% in 240 min, which is 2 times higher than the mono-metallic Pd/PCNT catalyst under identical conditions, at the same time, the conversion of CAL reaches over 56.0%. The detailed mechanism involved in the hydrogenation process of cinnamaldehyde and the reasons behind the improvement in performance of Pd-Ru/PCNT catalyst are not clear at the moment, which may be due to a number of factors including the peculiar structure and properties of CNTs such as the higher graphitization degree, good electronic conductivity and the oxygen-containing groups on CNT’s surface. The synergic effect between metal Pd and metal Ru is another factor that cannot be ruled out. To clarify these issues, more detailed work is needed. For graphite-like carbon materials such as CNTs, another important issue that cannot be overlooked is the presence of delocalized π-electrons. Because of this, electron transfer or hydrogen spillover becomes possible between the metal catalyst and the CNT support. This is one of the driving forces behind the research of using CNT as catalyst supports. Richard et al. [13] proposed a scheme of electron transfer from the support to metal crystallites in order to account for the enhancement in activity when platinum was supported on graphite for the hydrogenation of cinnamaldehyde. For the Pd-Ru/PCNT catalyst in our case, the interaction between Pd and Ru particles may result in a synergic effect that leads to the obvious improvements in the catalytic activity of the Pd-Ru/PCNT catalyst for the cinnamaldehyde hydrogenation. Moraweck et al. [14] reported that Pt-Fe/C alloy catalysts exhibited a significant higher activity than Pt/C mono-metal catalysts under the same reaction conditions, which was attributed to the electron transfer from iron to platinum. In the case


273

of CNT supported bimetallic catalysts, the high electrical conductivity of CNTs makes it possible for electrons to be transferred from Pd to Ru metal even though most Pd and Ru particles are not in contact with each other [15]. It has been found that the surface functional groups of CNTs also play an important role in the hydrogenation reactions [2]. As discussed above, the improvement in the catalytic activity of Pd/PCNTs catalysts in comparison to the Pd/RCNT catalysts is obvious, which may be due to the presence of surface functional groups on the CNT supports that function as anchoring sites for Pd particles [2,16]. Table 1 Catalytic activity of Pd/CNTs, Ru/CNTs and Pd-Ru/CNTs for cinnamaldehyde hydrogenation at 120oC and 5.0 MPa of hydrogen

Catalysts

Activity (mol.%)

Reaction time (min) 30

60

90

120

180

240

Pd/RCNTs

Conver.a Select.b Yieldc

10.3 21.3 2.1

11.3 21.2 2.4

13.1 19.9 2.6

13.9 20.1 2.8

16.7 19.2 3.2

20.5 18.0 3.7

Pd/PCNTs

Conver. Select. Yield

5.7 50.9 2.9

7.4 37.1 3.5

17.3 22.5 3.9

22.2 20.0 4.4

28.3 19.9 5.6

32.3 20.0 6.5

Ru/PCNTs


6.7 78.7 5.3

10.8 78.7 8.5

16.7 78.6 13.1

20.8 75.0 15.6

-

-

Pd-Ru/PCNTs


7.3 79.2 5.8

14.0 74.3 10.4

22.8 67.5 15.4

26.6 60.9 16.2

37.9 44.1 16.7

56.6 32.3 18.3

Note: a Conversion of cinnamaldehyde; cinnamyl alcohol

b

the selectivity to cinnamyl alcohol;

c

the yield of

Figure 1b shows the typical TEM image of the Pd-Ru/PCNT catalysts. The HRTEM examination reveals that the size distribution of the metal particles is narrow, which centers at 5.0–10.0 nm. It can be seen from Fig. 1b that the metal catalysts, i.e. the dark spots on the external surface of the carbon tubes, are deposited uniformly on the outer surface of CNTs, though occasionally some large aggregates can also be seen.

274


Graphite (004)

Pd (111) Graphite (100)

Intensity(a.u.)

Graphite (002)

In the XRD patterns of different catalysts shown in Fig. 2, the remarkable feature is the peaks corresponding to Pd metal with relatively high intensity and the peaks corresponding to Ru metal with low intensity, for which two possible reasons can be envisioned. One reason is that in the Pd-Ru/PCNTs catalyst, the loading amount of Pd catalyst (5.0 wt.%) is higher than Ru catalyst (only 1.0 wt.%); another one lies in the fact that the Ru metal is present in high and homogeneous dispersion.

c b a

10

20

30 40 50 o 2 Theta ( )

60

70

Fig. 2. XRD patterns of CNT supported catalysts (a) PCNTs, (b) Pd/PCNTs and (c) Pd-Ru/PCNTs

The results presented here clearly show that high performance catalysts for hydrogenation reactions can be prepared using carbon nanotubes (CNTs) as support, which may be attributed to the peculiar structure and properties of CNTs. For the hydrogenation of CAL, bimetallic catalysts Pd-Ru supported on purified CNTs show remarkably high catalytic activities in comparison to mono-metallic Pd/PCNTs, which is believed to be due to the synergic effect between Pd and Ru metals. Acknowledgements. This work was partly supported by the National Natural Science Foundation of China (Nos. 20376011, 50472082), the Natural Science Foundation of Liao-ning Province of China (No. 9810300701, 2001101003), the National Basic Research Program of China (Grant No. G2003CB615806) and the Program for New Century Excellent Talents in Universities of China. CH Liang thanks the Alexander von Humboldt Stiftung for a fellowship.


275

REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.

P.N. Rylander: Catalytic Hydrogenation in Organic Synthesis. p. 72. Academic Press, New York 1979. M.L. Toebes, F.F. Prinsloo, J.H. Bitter, A. Jos van Dillen, Krijn P. de Jong: J. Catal., 214, 78 (2003). J.M. Planeix, N. Coustel, B. Coq, V. Brotons, P.S. Kumbhar, R. Dutartre, P. Geneste, P. Bernier, P.M. Ajayan: J. Am. Chem. Soc., 116, 7935 (1994). V. Lordi, N. Yao, J. Wei: Chem. Mater., 13, 733 (2001). F. Salman, C. Park, R.T.K. Baker: Catal. Today, 53, 385 (1999). Z. Poltarzewski, S. Galvagno, P. Pietropaolo, P. Staiti: J. Catal., 102, 190 (1986). U. Schroder, L. De Verdier: J. Catal., 166, 25 (1993). P. Gallezot, D. Richard: Catal. Rev.- Sci. Eng. 40, 81 (1998). A. Giroir-Fendler, D. Richard, P. Gallezot: Stud. Surf. Sci. Catal., 41, 171 (1988). S. Iijima: Nature, 354, 56 (1991). Walt A. de Heer, A. Châtelain, D. Ugarte: Science, 270, 1179 (1995). R. Saito, M. Fujita, G. Dresselhaus, M. S. Dresselhaus: Phys. Rev., 46, 1804 (1992). P. Gallezot, C. Leclercq, I. Mutin, C. Nicot, D. Richard: J. Microsc. Spectrosc. Electron. 10, 479 (1985). B. Moraweck, P. Bondot, D. Goupil, P. Fouilloux, A. Renouprez: J. Phys. (Paris), C 8 287 (1987). Changhai Liang, Zhonglai Li, Jieshan Qiu, Can Li: J. Catal., 211, 278 (2002). Hongzhe Zhang, Jieshan Qiu, Changhai Liang, Zhonglai Li, Xiuna Wang, Yunpeng Wang, Zhaochi Feng, Can Li: Catal. Lett., 211, 101 (2005).