Catalytic combustion of benzene over metal oxides supported on SBA-15

Catalytic combustion of benzene over metal oxides supported on SBA-15

Available online at www.sciencedirect.com Journal of Industrial and Engineering Chemistry 14 (2008) 779–784 www.elsevier.com/locate/jiec Catalytic c...

2MB Sizes 2 Downloads 48 Views

Recommend Documents

No documents
Available online at www.sciencedirect.com

Journal of Industrial and Engineering Chemistry 14 (2008) 779–784 www.elsevier.com/locate/jiec

Catalytic combustion of benzene over metal oxides supported on SBA-15 Jin Sup Yang b, Won Young Jung b, Gun Dae Lee b, Seong Soo Park b, Euh Duck Jeong a, Hyun Gyu Kim a, Seong-Soo Hong b,* b

a Busan Center, Korea Basic Science Institute, Republic of Korea Division of Applied Chemical Engineering, Pukyong National University, 100 Yongdang-dong, Nam-ku, 608-739 Pusan, Republic of Korea

Received 28 January 2008; accepted 9 May 2008

Abstract The catalytic combustion of benzene over metal oxides supported on SBA-15 was investigated. The catalysts were prepared by the incipient wetness method and characterized by XRD, BET, TEM, ESR and TPR. The calcined siliceous SBA-15 and CuO/SBA-15 samples displayed wellresolved patterns with a sharp peak at about 1.08. It is clear that the loading of CuO on the silica matrix drastically decreases the surface area and pore volume of the catalysts, as would be expected for the incorporation of CuO. Among the supported metal oxides, CuO supported on SBA-15 was found to have the highest activity for benzene oxidation. In addition, copper oxide supported on SBA-15 gives higher catalytic activity than copper oxide supported on MCM-41. From the ESR results, the CuO dispersed on the SBA-15 acts as the active site of the CuO/SBA-15 catalysts in the oxidative decomposition of benzene. The catalytic activity gradually increases with increasing CuO loading on SBA-15. # 2008 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved. Keywords: Copper oxide supported on SBA-15; Catalytic combustion of benzene; VOCs; TPR

1. Introduction Volatile organic compounds (VOCs) include all organic compounds that exist in the gaseous state in ambient air. Offensive odors and toxic air emissions as well as the formation of ground-level ozone and petrochemical smog are environmental problems which are related to the emissions of VOCs. To reduce the environmental impact of these emissions and be able to fulfill the established goals regarding the reduction of VOCs, the existing legislation will likely become more stringent and broader in its application [1]. Benzene is a VOC emitted from different industries, such as the petrochemical, paint and coating and steel manufacturing industries [2,3]. The catalytic combustion of VOCs can operate at a lower temperature than that without a catalyst. The major advantage of catalytic combustion is that it can efficiently treat very dilute pollutants at concentrations of less than 1%, which cannot be thermally combusted without additional fuel. It is more energy efficient and, therefore, has a distinct economic advantage. Hence, catalytic oxidation has broad applicability

* Corresponding author. Tel.: +82 51 620 1465; fax: +82 51 625 4055. E-mail address: [email protected] (S.-S. Hong).

for ‘end of pipe’ pollution clean up than does thermal oxidation [4]. Many different metal oxides are known to be active for the combustion of VOCs, e.g., CuO, Co3O4, Cr2O3, NiO, Fe2O3 and MnO2 [5,6]. The oxide can be deposited on a support to increase its dispersion and, moreover, its specific activity may be increased due to its reaction or interaction with the support. TiO2, which is a widely used support [7], is known to enhance the activity in many cases, due to the active phase-support interaction [8]. Ordered mesoporous materials are considered to be an ideal host for nanoparticles due to their large surface area and welldefined pore structure. More importantly, the pore size of ordered mesoporous materials can be tuned within the nanosize range and the pore surface can be functionalized with various organic materials. In previous studies, various metal or metal oxide particles were confined into nano-channels of the mesoporous materials, MCM-41 and SBA-15 [9,10]. Compared to MCM-41, SBA-15 is a more stable matrix for forming occluded nanoparticles, due to its thicker walls and large pore size compared to MCM-41 [11]. In the present work, we investigated the catalytic combustion of benzene over metal oxides supported on SBA-15. The catalysts were characterized using BET, XRD, TEM, electron

1226-086X/$ – see front matter # 2008 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.jiec.2008.05.008

780

J.S. Yang et al. / Journal of Industrial and Engineering Chemistry 14 (2008) 779–784

spin resonance (ESR) and temperature-programmed reduction (TPR), in order to determine the nature of the metal oxide species supported on SBA-15. In addition, their catalytic activity on the combustion of benzene in the presence of oxygen was also examined, along with the effect of the type of support, metal oxide, and copper loading ratio. 2. Experimental The SBA-15 materials were synthesized by the hydrothermal method using tetraethyl orthosilicate (TEOS) as the sources of Si. 4.06 g of Pluronic P123 triblock copolymer surfactant (EO20PO70EO20, Mav = 5800) and 0.68 g of polyethylene-glycol (PEG) were dissolved in 160 mL of distilled water and 10 mL of 2 M HCl was added dropwise to the solution as a catalyst. Under stirring, 22.78 mL of 0.1 M TEOS solution was added and the mixture was stirred continuously for 6 h to make the solution homogeneous. The resultant mixture was then aged in an autoclave at 100 8C for 48 h. The crystallized product was filtered, washed with distilled water, dried at ambient temperature for 24 h, and finally calcined at 500 8C for 3 h. The MCM-41 materials were synthesized according to the literature method [9]. All of the metal oxides supported on SBA-15 used in this experiment were prepared via the incipient wetness method. Aqueous solutions of the metal nitrates were used as precursors for the metal oxides. SBA-15 and MCM-41 were impregnated with appropriate amounts of the precursor solution to incipient wetness, followed by stirring continuously at 80 8C until the total evaporation of water was achieved. Finally, all of the samples were calcined at 500 8C for 12 h. The crystal structures of the prepared supported copper oxides were examined by powder X-ray diffraction (XRD) with Cu Ka radiation (Rigaku Co. Model DMax). Temperature-programmed reduction experiments were carried out using 20 mg of the catalysts under a gas flow (30 mL/ min) of hydrogen (5%) and nitrogen (95%). The temperature of the catalysts was linearly raised at a rate of 10 8C/min. A thermal conductivity detector was used to monitor the hydrogen consumed during the TPR course. ESR measurements were carried out on a JEOL JEX-PX 2000 spectrometer operating in the X-band frequency range and 100-kHz field modulation. Reaction tests were carried out in a continuous flow fixedbed reactor. The reactor was a quartz glass tube with an o.d. of 1 cm and length of 24 cm, mounted in a tubular furnace. A Ktype thermocouple was in contact with the catalyst bed. The reaction was carried out in the temperature range of 200– 500 8C and at a feed flow rate of 100 cm3/min (gas hourly space velocity GHSV = 30,000 h 1) controlled by a mass flow meter/ controller (Tylan). The feed was introduced into the reactor by vaporizing the benzene with helium gas and 80 mg of catalyst was used in all of the experiments. The outlet gases were analyzed using an on-line gas chromatograph equipped with a flame ionization and thermal conductivity detector. The analyses of benzene and CO/CO2 were conducted through Carbowax and Hysep Dip columns, respectively.

Fig. 1. XRD patterns of pure SBA-15 and 5 wt% CuO/SBA-15.

3. Results and discussion Fig. 1 shows the XRD patterns of the SBA-15 and 5 wt% CuO/SBA-15 materials. The calcined siliceous SBA-15 and 5 wt% CuO/SBA-15 samples displayed well-resolved patterns with a sharp peak at about 1.08 that matched well with the reported pattern [12]. The intensity of the (1 0 0) peak shows higher order Bragg reflections in spite of the addition of CuO to SBA-15 and this indicates that the structural integrity of the SBA-15 materials is still maintained. Fig. 2 shows the ESR spectra for SBA-15 loaded with different amounts of CuO. The observed spectral parameters were interpreted as arising from distorted octahedrally coordinated Cu2+ ions [13]. When the copper content is increased, a poorly resolved hyperfine structure is observed and the signal intensities of the ESR spectra also decrease. These changes are due to the dipolar coupling arising from the strong interaction between the nearneighbor copper atoms and the formation of clusters, resulting in antiferromagnetic coupling, which make them undetectable by ESR [14]. These changes also imply that a large part of these isolated Cu2+ ions are buried in multilayered crystallites, whose spatial structure approaches that of a normal CuO lattice.

Fig. 2. ESR spectra at 77 K of CuO/SBA-15 catalysts with different loading ratio.

J.S. Yang et al. / Journal of Industrial and Engineering Chemistry 14 (2008) 779–784 Table 1 Textural properties of CuO-loaded SBA-15 materials. Materials

Surface area (m2/g)

Micropore volume (10 8 m3/g)

Pore diametera (nm)

SBA-15 1 wt% CuO/SBA-15 5 wt% CuO/SBA-15 7 wt% CuO/SBA-15 10 wt% CuO/SBA-15

869 793 506 504 387

9.4 8.7 5.0 4.9 4.4

4.3 4.3 3.9 3.9 3.6

a

BJH adsorption average pore diameter.

Nevertheless, the hyperfine structure can still be observed in the ESR spectrum of the 7 wt% CuOx/SBA-15 catalyst. Similar results were also found by Dow et al. over CuO/g-alumina catalysts [13]. These results strongly suggest that the CuO dispersed on the SBA-15 surface acts as the active site of the CuO/SBA-15 catalysts in the oxidative decomposition of benzene. Table 1 shows the textural properties of the CuO loaded on the SBA-15 samples. It is clear that a CuO loading of more than 5 wt% on the silica matrix drastically decreases the surface area of the support, as would be expected for CuO incorporation. A rapid decrease in their pore volumes is also observed, when CuO is loaded on the SBA-15 samples. This reduction in the pore volume indicates that most of the CuO particles are loaded within the ordered channels of the support. This result is confirmed by the TEM micrograph of the 10 wt% loaded CuO on SBA-15 (Fig. 4), in which the highly ordered mesopore structure disappeared due to the clogging of the mesopore channels by the copper oxide particles. The conversion of benzene over various metal oxide catalysts (7 wt% metal loading) supported on SBA-15 is shown in Fig. 3 as a function of the reaction temperature. Fig. 3 shows the XRD patterns of the CuO/SBA-15 samples prepared by the impregnation method. Clearly, for the CuO/ SBA-15 sample with a loading of less than 3%, we cannot observe any diffraction peaks corresponding to copper oxide phase, indicating either that the structure of the copper is amorphous or that the size of CuO is smaller than the XRD detection limit. In contrast, for the CuO/SBA-15 sample with a loading of more than 5%, very strong and sharp XRD peaks at

Fig. 3. XRD patterns of CuO/SBA-15 catalysts with different loading ratios.

781

2u = 35.58 and 38.88 were observed. These two peaks are all characteristic of tenorite CuO [14]. Fig. 4 shows the TEM images of the SBA-15 and CuO/SBA15 samples prepared with different CuO loading ratios. The highly ordered structure of SBA-15 is shown as described in the literature [12]. Moreover, the regular silica morphology is maintained in the case of the 5 wt% CuO/SBA-15 sample, but the highly ordered hexagonal structure of the CuO/SBA-15 sample loaded with more than 7 wt% is partially destroyed and the highly ordered mesopore structure disappeared due to the clogging of the mesopore channels by the copper oxide particles. When metallic particles are loaded onto ordered mesoporous supports, deposition onto the external surface or confinement into the nano-channels can occur. As the loading of CuO on SBA-15 is increased, severe sintering during thermal treatment is often inevitable, due to the absence of strong metalsupport interaction, and this will lead to the formation of large particles. A series of SBA-15-supported metal oxide catalysts were prepared by the impregnation method and their catalytic activities are shown in Fig. 5. The conversion of benzene shows typical S-shaped dependences for the conversion as a function of temperature. The steep rise in the conversion is similar to the results observed in many other oxidation reactions [5,15]. In a blank experiment, the empty quartz glass tube gave only 2% benzene conversion at 500 8C. All of the catalysts studied in the experiment produced carbon oxides and no detectable Ccontaining by-products were found. As shown in Fig. 5, among the supported the metal oxides, CuO supported on SBA-15 is found to have the highest activity for benzene oxidation. The conversion of the benzene oxidation on the CuO/SBA-15 catalyst reaches 100% at about 250 8C. The activity is in the order of CuO > MnO > FeO > NiO and the pure SBA-15 shows very low activity for benzene oxidation. This result indicates that the metal oxides supported on SBA-15 play an important role in and have an influence on the catalytic activity of benzene oxidation. Fig. 6 shows the reduction behaviors of these catalysts. For the CuO/SBA-15 catalyst, the reduction peaks appear at about 210 8C and 330 8C, which are lower than the corresponding temperatures of the other catalysts. In addition, the catalytic activity increases as the reduction peak temperature decreases. Thus, for the CuO/SBA-15 catalyst, more oxygen can be used in the redox cycle. This is likely due to the highly dispersed CuO on SBA-15. However, the higher temperature of the reduction peak for the MnO/SBA-15 catalyst is attributed to the low dispersion of MnO on SBA-15. This result suggests that the CuO/SBA-15 catalyst has the best redox property, which must be one of the reasons for it exhibiting the highest activity for the catalytic combustion of benzene. The conversion of the benzene oxidation over the CuO catalysts (7 wt% Cu loading) supported on SBA-15 and MCM41 is shown in Fig. 7 as a function of temperature. As shown in Fig. 7, copper oxide supported on SBA-15 is found to have higher activity for benzene oxidation. The conversion of benzene oxidation reaches almost 100% at 350 8C. At this temperature, the benzene conversion for the CuO/MCM-41

782

J.S. Yang et al. / Journal of Industrial and Engineering Chemistry 14 (2008) 779–784

Fig. 4. TEM images of 1 wt% CuO/SBA-15 (a), 5 wt% CuO/SBA-15 (b), 7 wt% CuO/SBA-15 (c), and 10 wt% CuO/SBA-15 (d) materials.

Fig. 5. Benzene conversion vs. reaction temperature over different metal oxide supported on SBA-15 catalysts; benzene = 10,000 ppm, O2 = 20%, GHSV = 30,000 h 1.

Fig. 6. TPR profiles measured for different metal oxides supported on SBA-15 catalysts: heating rate = 10 K/min, gas mixture = 7.5% H2/N2.

J.S. Yang et al. / Journal of Industrial and Engineering Chemistry 14 (2008) 779–784

Fig. 7. Benzene conversion vs. reaction temperature over copper oxide supported on SBA-15 and MCM-41 catalysts; benzene = 10,000 ppm, O2 = 20%, GHSV = 30,000 h 1.

Fig. 8. TPR profiles measured for CuO/SBA-15 and CuO/MCM-41 catalysts: heating rate = 10 K/min, gas mixture = 7.5% H2/N2.

catalyst is about 40%. For the CuO/MCM-41 catalyst, the temperature required for the 80% conversion of benzene is at least 500 8C, which is 200 8C higher than that for the CuO/ SBA-15 catalyst.

783

Fig. 10. TPR profiles measured for CuO/SBA-15 catalysts with different loading: heating rate = 10 K/min, gas mixture = 7.5% H2/N2.

Fig. 8 shows the TPR profiles of these catalysts. For the CuO/SBA-15 catalyst, the reduction peaks appear at about 210 8C and 330 8C. The CuO/MCM-41 catalyst shows reduction peaks at 350 8C and 500 8C. When copper oxide is supported on SBA-15 and MCM-41, two TPR peaks are observed. It is thought that the peak at low temperature is due to the reduction of the highly dispersed copper oxide species and the peak at high temperature to the reduction of bulk CuO. Thus, the reduction temperature peak appeared at lower temperature for CuO/SBA-15 is larger compared to that for CuO/MCM-41, resulting in the higher redox property and the higher activity for the catalytic oxidation of benzene. To make further investigation, a series of SBA-15-supported copper oxide catalysts were prepared with different Cu loadings. Their catalytic activities are shown in Fig. 9 as a function of the reaction temperature. The pure SBA-15 catalyst shows very poor catalytic activity compared with the CuO/ SBA-15 catalyst. This result indicates that the CuO acts as an active site in the combustion of benzene. In addition, the catalytic activity gradually increases with increasing Cu loading on SBA-15. When the Cu loading reaches 7 wt%, the total conversion temperature is lowered to 400 8C. The TPR profiles for the series of CuO/SBA-15 catalysts with Cu loadings ranging from 1 to 10 wt% are shown in Fig. 10. At the lower Cu loadings of less than 1 wt%, no reduction peak appears until 650 8C. As the Cu loading is increased, the reduction peak, which is attributed to the reduction of highly dispersed copper oxide, moves to a lower temperature. This result suggests that the CuO/SBA-15 catalyst has the best redox property, which must be one of the reasons for it exhibiting the highest activity for the catalytic combustion of benzene. 4. Conclusions

Fig. 9. Benzene conversion vs. reaction temperature over copper oxide supported on SBA-15 with different loading ratio; benzene = 10,000 ppm, O2 = 20%, GHSV = 30,000 h 1.

The catalytic combustion of benzene over metal oxides supported on SBA-15 was examined. The catalysts were characterized using BET, XRD, TEM, ESR and TPR, in order to determine the nature of the metal oxide species supported on SBA-15. The calcined siliceous SBA-15 and 5 wt% CuO/SBA15 samples displayed well-resolved patterns with a sharp peak

784

J.S. Yang et al. / Journal of Industrial and Engineering Chemistry 14 (2008) 779–784

at about 1.08. The loading of CuO on the silica matrix drastically decreases the surface area and pore volume of the catalysts, as would be expected for the incorporation of CuO. CuO supported on SBA-15 was found to have the highest activity for benzene oxidation and it gives higher catalytic activity than copper oxide supported on MCM-41. The catalytic activity gradually increases with increasing Cu loading on SBA-15. From the ESR results, the CuO dispersed on the SBA15 acts as the active site of the CuO/SBA-15 catalysts in the oxidative decomposition of benzene. Acknowledgement This work was supported by the Korea Research Foundation Grant funded by the Korea Government (MOEHRD) (KRF2006-311-D00396). References [1] G.H. Lee, M.S. Lee, G.D. Lee, Y.H. Kim, S.S. Hong, J. Ind. Eng. Chem. 8 (2002) 572.

[2] S.J. Cho, M.W. Ryoo, K.S. Soun, S.K. Lee, K.S. Kang, Korean J. Chem. Eng. 16 (1999) 478. [3] R.S.G. Ferreira, P.G.P. Oliveira, F.B. Noronha, Appl. Catal. B 29 (2001) 275. [4] G.J. Hutchings, C.S. Heneghan, I.D. Hudson, S.H. Taylor, Nature 384 (1996) 341. [5] P. Larsson, A. Andersson, J. Catal. 179 (1998) 72. [6] M.I. Vass, V. Georgescu, Catal. Today 29 (1996) 463. [7] (a) C.N. Satterfield, Heterogeneous Catalysis in Industrial Practice, 2nd ed., McGraw-Hill, New York, 1991; (b) M.K. Jeon, M.K. Yeo, H.J. Choi, J. Kim, S.T. Choung, J.B. Kim, D. Jeong, M. Kang, J. Ind. Eng. Chem. 13 (2007) 827. [8] G. Cordoba, M. Viniegra, J.L.G. Fierro, J. Padolla, R. Arroyo, J. Solid State Chem. 138 (1998) 1. [9] H. Balcar, P. Topka, N. Zilkova, J. Perez-Pariente, J. Cejka, Stud. Surf. Sci. Catal. 156 (2005) 795. [10] Y.J. Do, J.H. Kim, J.H. Park, S.S. Park, S.S. Hong, C.S. Suh, G.D. Lee, Catal. Today 101 (2005) 299. [11] D. Zhao, Q. Huo, J. Feng, B.F. Chmelka, G.D. Stucky, J. Am. Chem. Soc. 120 (1998) 6024. [12] G. Li, X.S. Zhao, Ind. Eng. Chem. Res. 45 (2006) 3569. [13] W. Dow, Y. Wang, T. Huang, Catal. J. 160 (1996) 155. [14] A. Gervasini, S. Bennici, Appl. Catal. A 281 (2005) 199. [15] P. Larsson, A. Andersson, L.R. Wallenberg, B. Svensson, J. Catal. 163 (1996) 279.