- Email: [email protected]
Reducibility of Cobalt Oxides over SBA-15 supported Cobalt Catalysts for Fischer-Tropsch Synthesis
Recent Progress in Mesostructured Materials D. Zhao, S. Qiu, Y. Tang and C. Yu (Editors) © 2007 Elsevier B.V. All rights reserved.
685 685
Reducibility of Cobalt Oxides over SBA-15 supported Cobalt Catalysts for Fischer-Tropsch Synthesis Dae Jung Kima*, Brian C. Dunnb, Min Kangc, Jae Eui Yied, Seong-Hyun Kime, Jenifer Gasserb, Eric Fillerupb, Louisa Hope-Weeks1 and Edward M. Eyringb "Department of Chemistry & Biochemistry, Texas Tech Universiy, TX 79409, USA. h Department of Chemistry, University of Utah, Salt Lake City, UT'84112, USA c Department of Chemistry and Sungkyunkwan Advanced Institute ofNano technology, Sungkyunkwan University, Suwon, 440-746, Republic of Korea Department of Applied Chemistry, Division of Biotechnology & Nanotechnology, Ajou University, Suwon, 443-749, Republic of Korea e Department of Environmental Engineering and Biotechnology, Myongji Unversity, Yongin 449-728, Republic of Korea
1. Introduction Fischer-Tropsch (FT) synthesis is a promising pathway to clean alternative fuels derived from coal syngas. The development of active catalysts with high selectivity for the production of long chain hydrocarbons is critical for the further advancement of this technology. Since the catalytic activity depends primarily on the overall amount of exposed metal atoms, an active catalyst requires a high reducibility of metal oxides. Mesoporous silica materials such as SBA-15 have been recently used as supports for cobalt [1-3]. The high surface area (500 - 1500 m2/g) of the mesoporous materials results in increased metal dispersions at higher cobalt loadings compared with conventional amorphous silicas. The objective of this present study is to elucidate the impact of cobalt impregnation method on the reducibility of cobalt oxides and the catalytic activity in Fischer-Tropsch synthesis. The SBA-15 supported cobalt catalysts were prepared by three different cobalt impregnation methods: incipient wetness, post-synthesis, and supercritical solvent. The physical and chemical properties of the catalysts were obtained from N2 adsorption/desorption, XRD
686
and TPR experiments. The catalytic performance in FT synthesis was evaluated with a fixed-bed reactor. 2. Experimental Section The cobalt precursors used in the incipient wetness (IW) and post-synthesis (PS) impregnations were Co(NO3)2-6H2O and (CH3CO2)2Co-4H2O. The impregnation of SBA-15 with cobalt using a supercritical solvent (SS) proceeded as follows: The SBA-15 was added to a 250 ml ethanol solution of Co(NO3)2-6H2O, and stirred at ambient temperature for 1 h. The suspension was transferred to an autoclave placed inside a furnace. The autoclave was purged ten times with 200 psi N2 to remove any oxygen trapped in the system. The autoclave was heated to 350°C at 5°C/min, then held at 350°C for 3 h. The pressure inside the autoclave was maintained at 2000 psi by controlled venting through a high-pressure valve. The system was cooled to 200°C, and the gas inside the autoclave was vented for 1 h. The system was then cooled to ambient temperature. The cobalt impregnated samples were calcined in air at 550°C overnight. For all cobalt catalysts, the cobalt loading was 6 wt. %. The FT synthesis was carried out in a fixed-bed stainless steel reactor (5 mm I.D. and 168 mm length) at 100 psi and 265°C. A H2/CO molar ratio of 2 was used, and the tests for the cobalt catalysts followed the experimental procedures described earlier [4]. 3. Results and Discussion XRD patterns shown in Fig. 1 and nitrogen adsorption isotherms obtained indicated that all the cobalt catalysts had a typical 2-D hexagonal structure of the pure SBA-15. This suggests that the mesopore structure is still retained after cobalt impregnation. Pore structural parameters calculated from nitrogen adsorption/desorption isotherms for the SBA-15 supported catalysts are listed in Table 1. The pore structural parameters of the pure SBA-15 such as surface area, pore volume and average pore size were decreased by the cobalt loading. The three cobalt catalysts having the same loading of cobalt showed similar values in the parameters. The mean CO3O4 crystallite diameters of the three cobalt catalysts deduced from the XRD data using the Scherrer equation are presented in Table 1. The XRD peak of Co3O4 for the PS sample was not detectable. This suggests that most of cobalt oxides are present as cobalt silicates in the framework of the SBA-15, and the crystallite size of CO3O4 on the surface of the SBA-15 is very small. The mean Co3O4 crystallite size on the SS sample is larger than on the IW sample. This result indicates that the crystallite size of CO3O4 is clearly dependent on the cobalt impregnation method. Figure 2 shows TPR profiles of the SBA-15 supported cobalt catalysts. The IW and SS samples exhibited similar TPR profiles with three typical peaks.
687
However, the TPR profile for the PS sample with two peaks was significantly different. The maximum intensities of the three TPR peaks for the SS sample were located at lower temperatures than for the IW sample. For the PS sample, the peak at temperatures above 760°C can be assigned to the reduction of cobalt oxides in the framework of SB A-15. The reduction degrees of the cobalt oxides on the three cobalt catalysts at temperatures less than 500°C are presented in Table 1. The SS sample showed the highest reducibility of cobalt oxides among the three catalysts prepared by different cobalt impregnation methods. According to the TPR results, undesireable cobalt oxides such as cobalt silicates (those not easily converted to active cobalt metal at lower temperature) were abundantly produced in the PS sample. Table 1. Physical and chemical properties of the SBA-15 supported cobalt catalysts Co3O4 d Re Catalyst Diameter (nm) . 724 1.243 8.09 Pure SBA-15 IW 465 0.811 8.08 11.1 49 472 PS 0.858 8.08 18 SS 461 0.815 8.08 11.6 63 a BET Surface area, b Total pore volume,c average pore diameter,d Co3O4 crystallite diameter calculated from the widths of XRD peaks using the Scherrer equation (2 theta = 36.68°), e reduction degree of cobalt oxides during TPR at 30 - 500°C,f IW: incipient wetness, PS: post-synthesis, SS: supercritical solvent Table 2 . . CO conversion, hydrocarbon selectivity and chain growth probability of the SBA-15 supported cobalt catalysts Sa (m2/g)
Catalyst
CO conversion
Vt b (cc/g)
Dc (nm)
Product selectivity (C mol%) Cl
C2-C5
C5-C10
C10+
IW 15.7 8.6 38.6 39.0 13.8 4.6 15.8 56.8 24.2 3.2 PS 21.1 32.4 45.3 15.0 7.3 SS a chain growth probability obtained from Anderson-Schulz-Flory equation
a a 0.86 0.82 0.88
Catalytic activities of the SBA-15 supported cobalt catalysts in FT synthesis are summarized in Table 2. CO conversion and higher hydrocarbon selectivity can be related to the reducibility of the cobalt oxides and pore structure of a cobalt catalyst. The three samples (IW, PS, SS) having the same loading of cobalt showed similar values in BET surface area, pore volume and average pore size. However, the three samples showed differences in the reducibility of cobalt oxides. The SS sample exhibited the highest CO conversion, C5+ selectivity and chain growth probability among the three catalysts obtained by three different cobalt impregnation methods. This result is quite consistent with the TPR result.
688
j LJLi
ss
^_3JL
ps
/
100
isity
8
IB
J
i
I
- §
iw
'."
Pure SBA-1S
a u
0
20 / degree
Fig.l XRD patterns of the SBA-15 supported cobalt catalysts
100 200 300 400 500 600 700 BOO 900
Temperature (°C)
Fig.2 TPR profiles of the SBA-15 supported cobalt catalysts
4. Conclusion The supercritical solvent method for cobalt impregnation on the SBA-15 gave the largest crystallite size of Co3O4 and highest reducibility of cobalt oxides on a cobalt catalyst. The cobalt catalyst with cobalt impregnation by the supercritical solvent method showed the highest CO conversion, C5+ selectivity and chain growth probability among the three cobalt catalysts obtained by three different cobalt impregnation methods: incipient wetness, post-synthesis, and supercritical solvent. This result indicates that CO conversion and higher hydrocarbon selectivity can be related to the reducibility of the cobalt oxides. 5. Acknowledgement The U.S. Department of Energy provided financial support to the Consortium for Fossil Fuel Science for this study (contract # DE-FC2602NT41954). 6. References [1] [2] [3] [4]
K. Okabe, M. Wei and H. Arakawa, Energy & Fuels, 17 (2003) 822. A. Martinez, C. Lopez, F. Marquez and I. Diaz, J. Catal., 220 (2003) 486. J. Panpranot, J. G. Goodwin Jr. and A. Sayari, Catal. Today, 77 (2002) 269. D. J. Kim, B. C. Dunn, P. Cole, G. Turpin, R. D. Ernst, R. J. Pugmire, M. Kang, J. M. Kim, and E. M. Eyring, Chem. Commun., (2002) 1462.
685 685
Reducibility of Cobalt Oxides over SBA-15 supported Cobalt Catalysts for Fischer-Tropsch Synthesis Dae Jung Kima*, Brian C. Dunnb, Min Kangc, Jae Eui Yied, Seong-Hyun Kime, Jenifer Gasserb, Eric Fillerupb, Louisa Hope-Weeks1 and Edward M. Eyringb "Department of Chemistry & Biochemistry, Texas Tech Universiy, TX 79409, USA. h Department of Chemistry, University of Utah, Salt Lake City, UT'84112, USA c Department of Chemistry and Sungkyunkwan Advanced Institute ofNano technology, Sungkyunkwan University, Suwon, 440-746, Republic of Korea Department of Applied Chemistry, Division of Biotechnology & Nanotechnology, Ajou University, Suwon, 443-749, Republic of Korea e Department of Environmental Engineering and Biotechnology, Myongji Unversity, Yongin 449-728, Republic of Korea
1. Introduction Fischer-Tropsch (FT) synthesis is a promising pathway to clean alternative fuels derived from coal syngas. The development of active catalysts with high selectivity for the production of long chain hydrocarbons is critical for the further advancement of this technology. Since the catalytic activity depends primarily on the overall amount of exposed metal atoms, an active catalyst requires a high reducibility of metal oxides. Mesoporous silica materials such as SBA-15 have been recently used as supports for cobalt [1-3]. The high surface area (500 - 1500 m2/g) of the mesoporous materials results in increased metal dispersions at higher cobalt loadings compared with conventional amorphous silicas. The objective of this present study is to elucidate the impact of cobalt impregnation method on the reducibility of cobalt oxides and the catalytic activity in Fischer-Tropsch synthesis. The SBA-15 supported cobalt catalysts were prepared by three different cobalt impregnation methods: incipient wetness, post-synthesis, and supercritical solvent. The physical and chemical properties of the catalysts were obtained from N2 adsorption/desorption, XRD
686
and TPR experiments. The catalytic performance in FT synthesis was evaluated with a fixed-bed reactor. 2. Experimental Section The cobalt precursors used in the incipient wetness (IW) and post-synthesis (PS) impregnations were Co(NO3)2-6H2O and (CH3CO2)2Co-4H2O. The impregnation of SBA-15 with cobalt using a supercritical solvent (SS) proceeded as follows: The SBA-15 was added to a 250 ml ethanol solution of Co(NO3)2-6H2O, and stirred at ambient temperature for 1 h. The suspension was transferred to an autoclave placed inside a furnace. The autoclave was purged ten times with 200 psi N2 to remove any oxygen trapped in the system. The autoclave was heated to 350°C at 5°C/min, then held at 350°C for 3 h. The pressure inside the autoclave was maintained at 2000 psi by controlled venting through a high-pressure valve. The system was cooled to 200°C, and the gas inside the autoclave was vented for 1 h. The system was then cooled to ambient temperature. The cobalt impregnated samples were calcined in air at 550°C overnight. For all cobalt catalysts, the cobalt loading was 6 wt. %. The FT synthesis was carried out in a fixed-bed stainless steel reactor (5 mm I.D. and 168 mm length) at 100 psi and 265°C. A H2/CO molar ratio of 2 was used, and the tests for the cobalt catalysts followed the experimental procedures described earlier [4]. 3. Results and Discussion XRD patterns shown in Fig. 1 and nitrogen adsorption isotherms obtained indicated that all the cobalt catalysts had a typical 2-D hexagonal structure of the pure SBA-15. This suggests that the mesopore structure is still retained after cobalt impregnation. Pore structural parameters calculated from nitrogen adsorption/desorption isotherms for the SBA-15 supported catalysts are listed in Table 1. The pore structural parameters of the pure SBA-15 such as surface area, pore volume and average pore size were decreased by the cobalt loading. The three cobalt catalysts having the same loading of cobalt showed similar values in the parameters. The mean CO3O4 crystallite diameters of the three cobalt catalysts deduced from the XRD data using the Scherrer equation are presented in Table 1. The XRD peak of Co3O4 for the PS sample was not detectable. This suggests that most of cobalt oxides are present as cobalt silicates in the framework of the SBA-15, and the crystallite size of CO3O4 on the surface of the SBA-15 is very small. The mean Co3O4 crystallite size on the SS sample is larger than on the IW sample. This result indicates that the crystallite size of CO3O4 is clearly dependent on the cobalt impregnation method. Figure 2 shows TPR profiles of the SBA-15 supported cobalt catalysts. The IW and SS samples exhibited similar TPR profiles with three typical peaks.
687
However, the TPR profile for the PS sample with two peaks was significantly different. The maximum intensities of the three TPR peaks for the SS sample were located at lower temperatures than for the IW sample. For the PS sample, the peak at temperatures above 760°C can be assigned to the reduction of cobalt oxides in the framework of SB A-15. The reduction degrees of the cobalt oxides on the three cobalt catalysts at temperatures less than 500°C are presented in Table 1. The SS sample showed the highest reducibility of cobalt oxides among the three catalysts prepared by different cobalt impregnation methods. According to the TPR results, undesireable cobalt oxides such as cobalt silicates (those not easily converted to active cobalt metal at lower temperature) were abundantly produced in the PS sample. Table 1. Physical and chemical properties of the SBA-15 supported cobalt catalysts Co3O4 d Re Catalyst Diameter (nm) . 724 1.243 8.09 Pure SBA-15 IW 465 0.811 8.08 11.1 49 472 PS 0.858 8.08 18 SS 461 0.815 8.08 11.6 63 a BET Surface area, b Total pore volume,c average pore diameter,d Co3O4 crystallite diameter calculated from the widths of XRD peaks using the Scherrer equation (2 theta = 36.68°), e reduction degree of cobalt oxides during TPR at 30 - 500°C,f IW: incipient wetness, PS: post-synthesis, SS: supercritical solvent Table 2 . . CO conversion, hydrocarbon selectivity and chain growth probability of the SBA-15 supported cobalt catalysts Sa (m2/g)
Catalyst
CO conversion
Vt b (cc/g)
Dc (nm)
Product selectivity (C mol%) Cl
C2-C5
C5-C10
C10+
IW 15.7 8.6 38.6 39.0 13.8 4.6 15.8 56.8 24.2 3.2 PS 21.1 32.4 45.3 15.0 7.3 SS a chain growth probability obtained from Anderson-Schulz-Flory equation
a a 0.86 0.82 0.88
Catalytic activities of the SBA-15 supported cobalt catalysts in FT synthesis are summarized in Table 2. CO conversion and higher hydrocarbon selectivity can be related to the reducibility of the cobalt oxides and pore structure of a cobalt catalyst. The three samples (IW, PS, SS) having the same loading of cobalt showed similar values in BET surface area, pore volume and average pore size. However, the three samples showed differences in the reducibility of cobalt oxides. The SS sample exhibited the highest CO conversion, C5+ selectivity and chain growth probability among the three catalysts obtained by three different cobalt impregnation methods. This result is quite consistent with the TPR result.
688
j LJLi
ss
^_3JL
ps
/
100
isity
8
IB
J
i
I
- §
iw
'."
Pure SBA-1S
a u
0
20 / degree
Fig.l XRD patterns of the SBA-15 supported cobalt catalysts
100 200 300 400 500 600 700 BOO 900
Temperature (°C)
Fig.2 TPR profiles of the SBA-15 supported cobalt catalysts
4. Conclusion The supercritical solvent method for cobalt impregnation on the SBA-15 gave the largest crystallite size of Co3O4 and highest reducibility of cobalt oxides on a cobalt catalyst. The cobalt catalyst with cobalt impregnation by the supercritical solvent method showed the highest CO conversion, C5+ selectivity and chain growth probability among the three cobalt catalysts obtained by three different cobalt impregnation methods: incipient wetness, post-synthesis, and supercritical solvent. This result indicates that CO conversion and higher hydrocarbon selectivity can be related to the reducibility of the cobalt oxides. 5. Acknowledgement The U.S. Department of Energy provided financial support to the Consortium for Fossil Fuel Science for this study (contract # DE-FC2602NT41954). 6. References [1] [2] [3] [4]
K. Okabe, M. Wei and H. Arakawa, Energy & Fuels, 17 (2003) 822. A. Martinez, C. Lopez, F. Marquez and I. Diaz, J. Catal., 220 (2003) 486. J. Panpranot, J. G. Goodwin Jr. and A. Sayari, Catal. Today, 77 (2002) 269. D. J. Kim, B. C. Dunn, P. Cole, G. Turpin, R. D. Ernst, R. J. Pugmire, M. Kang, J. M. Kim, and E. M. Eyring, Chem. Commun., (2002) 1462.