催化学报  2014, Vol. 35 Issue (9): 1520-1528   PDF (732 KB)    
扩展功能
加入收藏夹
复制引文信息
加入引用管理器
Email Alert
RSS
本文作者相关文章
刘会敏
李宇明
吴昊
杨维维
贺德华
Effects of Nd, Ce, and La modification on catalytic performance of Ni/SBA-15 catalyst in CO2 reforming of CH4
Huimin Liu, Yuming Li, Hao Wu, Weiwei Yang, Dehua He     
Innovative Catalysis Program, Key Laboratory of Organic Optoelectronics and Molecular Engineering of Ministry of Education, Department of Chemistry, Tsinghua University, Beijing 100084, China
Abstract: Rare-earth metal (Nd, Ce, and La) oxides modified Ni/SBA-15 catalysts were prepared by a β-cyclodextrin-modified impregnation method. The physicochemical properties of the catalysts were characterized using X-ray diffraction, N2 adsorption-desorption, temperature-programmed reduction, and thermogravimetric analysis. The catalytic performance of the catalysts was evaluated in the CO2 reforming of CH4 to syngas. The characterization results showed that Nd, Ce, and La modification had little effect on the textural structures and crystalline phases of the obtained catalysts but influenced the reduction of NiO species. Addition of Nd favored interactions between Ni and SiO2, possibly as a result of the formation of Ni-Nd-O species. The results for the CO2 reforming of CH4 revealed that the addition of suitable amounts of Nd (5-10 wt%) improved the catalytic activity and stability. A small amount of carbon was deposited over the used Nd-modified catalysts. The properties of Ni/SBA-15 catalysts modified with La and Ce were similar to those of the Nd-modified catalysts.
Key words: Rare earth metal     Nd modification     Nickel     SBA-15 catalyst     Carbon dioxide reforming of methane     Carbon deposition    

1. Introduction

CO2 reforming of CH4 (CRM) to syngas can be used to convert two greenhouse gases,i.e.,CH4 and CO2,simultaneously to high-value-added chemical products; this has therefore become a hot topic in catalysis research [ 1 ].

Ni-based catalysts are usually used in CRM to syngas because they have high initial catalytic activities and are cheap [ 2, 3 ]. However,during lengthy stability testing,Ni-based catalysts are gradually deactivated because of carbon deposition [ 4 ]. Several methods have been used to reduce carbon deposition on Ni-based catalysts [ 5, 6 ]. It has been reported that the size of the Ni particles [ 7 ] and the acidic-basic properties both significantly affect the ability of a catalyst to resist carbon deposition [ 8, 9 ].

Alkali metals are used as the basic material. K is an effective promoter,and the catalytic activities of K-modified Ni-K/Al2O3 catalysts increase with increasing K content in the range 1.0-2.9 wt% [ 10 ]. Li modification also improves the catalytic performance of Ni/CeO2 catalysts,and the obtained Ni-Li/CeO2 catalyst has a stronger ability to resist carbon deposition in CRM [ 11 ]. Rare-earth metals have similar abilities to tune the acidic-basic properties of catalysts; Ce [ 12, 13 ],La [ 14 ],Sm [ 15 ],and Gd [ 16, 17 ] are most often used,but the use of Nd as a promoter to modify the properties of Ni-based catalysts in CRM has rarely been reported.

We previously reported that a Ni/SBA-15 catalyst prepared by the β-cyclodextrin-modified impregnation method with small Ni particles exhibited good catalytic performance in CRM under the conditions atmospheric pressure,700 °C,CO2/CH4 = 1,and gas hourly space velocity (GHSV) = 1.28 × 104 mL g-1 h-1 [ 18 ]. However,when the GHSV was increased to 3.5 × 104 mL g-1 h-1,the CH4 and CO2 conversions gradually decreased during lengthy stability tests. In this study,rare-earth metals,i.e.,Nd,La,and Ce,were used to adjust the surface properties of the Ni/SBA-15 catalyst for further improving the catalytic performance at high GHSVs,and the influence of Nd additive on the performance of the catalyst was focused.

2. Experimental
2.1. Catalyst preparation

Poly(ethylene glycol)-block-poly(propylene glycol)-block- poly(ethylene glycol) (P123) of molecular weight 5800 was obtained from Aldrich. Tetraethyl orthosilicate (TEOS) of purity of 99.5% was purchased from Acros. Hydrochloric acid of concentration 36%-38% was purchased from the Beijing Chemical Plant. Ni(NO3)2·6H2O (analytical grade) was provided by the Shantou Xilong Chemical Company. Nd(NO3)3·nH2O,with a Nd2O3 content >37%,was provided by the Shanghai Pharmaceutical Chemical Products Corporation. Analytical-grade Ce(NO3)3·nH2O and La(NO3)3·nH2O were obtained from the Sinopharm Chemical Reagent Co.,Ltd. β-Cyclodextrin was purchased from the Beijing Aobox Biological Technology Company. All reagents were used as received.

The SBA-15 support was prepared with P123 as the template and TEOS as the precursor [ 19 ]. Rare-earth- metal- modified catalysts,Ni/RE/SBA-15 (RE = Nd,Ce,or La),and unmodified Ni/SBA-15 were prepared using the sequential β- cyclodextrin-modified impregnation method,in which the amount of β-cyclodextrin (CD) added in each step was n(CD)/n(RE) = 1/50 and n(CD)/n(Ni) = 1/50. The catalysts were calcined at 650 °C for 5 h. The theoretical contents of Ni and RExOy were 4.8 wt% and in the range 5-20 wt%,respectively. The obtained Ni/RE/SBA-15 and Ni/SBA-15 catalysts were denoted by 4.8Ni/xRE/SBA-15,in which x represents the RExOy content.

2.2. Catalyst characterization

The specific surface areas of the fresh catalysts were measured using N2 adsorption-desorption with a Micromeritics ASAP 2010 C analyzer. The samples were degassed at 200 °C for 2 h before the measurements.

X-ray diffraction (XRD) was used to investigate the structure of SBA-15 and the crystalline phases of the Ni particles on the catalysts. The XRD patterns were obtained with a Bruker D8 Advance X-ray diffractometer using Ni-filtered Cu radiation at 3 kW. The sizes of the crystalline NiO particles on the catalysts were determined by Scherrer equation.

The Ni contents of the catalysts were analyzed using X-ray fluorescence spectroscopy (XRF; Shimadzu XRF-1800).

The reduction behavior of the catalysts was investigated by H2-temperature-programmed reduction (H2-TPR). TPR profiles were obtained using a Quantachrome adsorption instrument; details of the procedure are available elsewhere [ 18 ].

CO2 temperature-programmed desorption (CO2-TPD) was cattied out to study the basic properties of supports and catalysts,and the samples were flushed at 500 °C for 0.5 h in Ar before analysis.

Thermogravimetric analysis (TGA) was used to measure the amounts of carbon deposited on the spent catalysts and was performed with a Mettler Toledo TGA/SDTA851e instrument under an air flow from room temperature to 800 °C (10 °C/min).

2.3. Catalytic performance tests

CRM to syngas was conducted in a fixed-bed quartz reactor with an inner diameter of 5 mm at atmospheric pressure. Before the reaction,the catalyst (0.20 g) was reduced with 20% H2/Ar at 600 °C for 2 h. The temperature was then raised to 800 °C,and CH4 (99.9%) and CO2 (99.9%) in a molar ratio of 1/1 were introduced into the reactor at a total flow rate of 115 mL/min. The flow rate of the tail gas was measured using a soap flowmeter,and its components were analyzed using on-line gas chromatography every 1 h. The relative amounts of the components in the effluents were calculated by the normalization method. The equations used for calculating the CH4 conversion,CO2 conversion,and selectivities are shown below:

CH4 conversion = 1 − Vout(CH4)/Vin(CH4)

CO2 conversion = 1 − Vout(CO2)/Vin(CO2)

CO selectivity = Vout(CO)/{[Vin(CH4) − Vout(CH4)] + [Vin(CO2) − Vout(CO2)]}

H2 selectivity = Vout(H2)/{2[Vin(CH4) − Vout(CH4)]}

3. Results and discussion
3.1. Physicochemical properties of fresh Ni/RE/SBA-15 catalysts
3.1.1. Textural properties

The textural properties of fresh 4.8Ni/xNd/SBA-15 catalysts with Nd contents in the range 0−20 wt% were characterized using N2 adsorption-desorption; the results are shown in Fig. 1. It can be seen clearly from Fig. 1(a) that the adsorption- desorption isotherms of the 4.8Ni/SBA-15 catalysts,with or without Nd modification,are type IV with H2 hysteresis loops,showing that Nd modification and β-cyclodextrin addition did not destroy the ordered structures of SBA-15. The pore size distributions of all the catalysts were in range 3-7 nm and were very similar.

Fig. 1. N2 adsorption-desorption isotherms (a) and pore size distributions (b) of fresh rare-earth-modified catalysts 4.8Ni/xRE/SBA-15.

The N2 adsorption-desorption isotherms and pore size distributions of the 4.8Ni/10La/SBA-15 and 4.8Ni/10Ce/SBA-15 catalysts were also obtained and compared with those of 4.8Ni/10Nd/SBA-15. Clearly,there were no differences among the shapes of the isotherms and the pore size distribution ranges for the La-,Ce-,and Nd-modified 4.8Ni/10RE/SBA-15 catalysts.

The textural properties of the fresh 4.8Ni/xRE/SBA-15 catalysts are listed in Table 1. It can be seen that 4.8Ni/SBA-15 had the highest specific surface area,578 m2/g. The specific surface areas of the La-,Ce-,and Nd-modified 4.8Ni/xRE/ SBA-15 catalysts were lower and similar to each other in the range 357-396 m2/g. This might be because,during preparation,4.8Ni/SBA-15 was calcined once,whereas the rare-earth- modified 4.8Ni/xRE/SBA-15 catalysts were calcined twice. Calcination might result in sintering of SBA-15 and Ni particles to some extent.

Table 1
Textural properties of 4.8Ni/xRE/SBA-15 catalysts and amounts of carbon deposited after CRM.
3.1.2. Crystalline structures

The crystalline structures and sizes of the Ni particles on the fresh 4.8Ni/xNd/SBA-15 catalysts were determined using XRD; the results are shown in Fig. 2. All samples had a weak peak at 10°-30°,attributable to amorphous SiO2,and there were also weak peaks at 37.0°,43.2°,and 62.8°,which were ascribed to the NiO crystalline phase. It should have been possible to determine the sizes of the NiO particles on the catalysts using the Scherrer equation. However,the NiO peaks on these catalysts were too weak to be used to calculate the NiO particle sizes; the weak peaks showed that the NiO particles on these catalysts were smaller than 5 nm.

Fig. 2. XRD patterns of rare-earth-modified fresh 4.8Ni/xRE/SBA-15 catalysts.

The XRD patterns of the La- or Ce-modified 4.8Ni/10RE/ SBA-15 catalysts were also obtained. They also had weak peaks attributable to NiO,indicating good dispersion of Ni over the two catalysts.

The Ni and Nd,La,or Ce contents of the catalysts were analyzed using XRF; the results are shown in Table 1. The actual contents of Ni and RExOy were close to the theoretical values.

3.1.3. Reduction behavior of NiO

The reduction behavior of the 4.8Ni/xRE/SBA-15 catalysts was examined using TPR; the results are shown in Fig. 3. There were two clear NiO reduction peaks in all samples,with one minor peak centered at 500-600 °C and another sharp peak at about 800 °C. In addition,compared with those of the 4.8Ni/ SBA-15 catalyst,the NiO reduction peaks of the rare-earth- modified catalysts shifted to lower temperatures,i.e.,the addition of rare-earth metals facilitated the reduction of NiO particles. It was also observed that the NiO reduction temperature shifted to lower temperature with increasing Nd content from 0 to 10 wt%,and then increased slightly as the Nd content was further increased from 10 to 20 wt%. The NiO reduction temperatures for the La- and Ce-modified catalysts were similar to that for the Nd-modified catalyst when the contents of La,Ce,or Nd were 10 wt%.

Fig. 3. TPR profiles of rare-earth-modified fresh 4.8Ni/xRE/SBA-15 catalysts.
3.1.4. Basic properties

CO2-TPD was performed to determine whether the Nd-modified supports,xNd/SBA-15,were basic; the results are shown in Fig. 4. No CO2 desorption peaks were observed in the range 100-500 °C in the 5Nd/SBA-15,10Nd/SBA-15,and 20Nd/ SBA-15,i.e.,Nd modification did not tune the acidic-basic properties of the supports.

Fig. 4. CO2-TPD profiles of xNd/SBA-15 supports.
3.1.5. Effects of rare-earth metals

There are several conflicting opinions on the effects of rare-earth metals. Xu et al. [ 20 ] reported that the interactions between Ni and Al2O3 were much enhanced when La or Ce was doped into Ni/Al2O3. Zhang et al. [ 21 ] found that when Ce was used to modify a Ni/SBA-16 catalyst,Ni−Ce−O species were formed among Ni,Ce,and the SBA-16 support; these species were responsible for the dispersion of NiO particles and facilitated reduction of NiCe/SBA-16. In our case,we suggest that RExOy species were also present in the form of Ni-RE-O because diffraction peaks attributable to RExOy were not be observed even at rare-earth metal loadings as high as 20 wt% (Fig. 2). The Ni-RE-O species improved NiO dispersion and facilitated the reduction of NiO particles.

3.2. Catalytic performance of Ni/RE/SBA-15 catalysts in CRM

The catalytic performance of 4.8Ni/xRE/SBA-15 catalysts in CMR was evaluated in a fixed-bed reactor at atmospheric pressure,800 °C,and GHSV = 3.5 × 104 mL g-1 h-1; the results are shown in Fig. 5. 4.8Ni/SBA-15 gave the lowest catalytic conversion in CRM (CO2 conversion 81% and CH4 conversion 67%). When 5% or 10% Nd was added to modify 4.8Ni/ SBA-15,the initial CO2 and CH4 conversions increased to about 95% and 90%,respectively. However,when the amount of Nd was further increased to 20%,the initial activity declined (on 4.8Ni/20Nd/SBA-15,the initial CO2 and CH4 conversions were 91% and 84%,respectively).

Fig. 5. Catalytic performance of Nd-modified 4.8Ni/xNd/SBA-15 catalysts in CRM. Reaction conditions: atmospheric pressure,800 °C,V(CH4)/V(CO2) = 1,GHSV = 3.5 × 104 mL g-1 h-1. (1) 4.8Ni/SBA-15; (2) 4.8Ni/5Nd/SBA-15; (3) 4.8Ni/10Nd/SBA-15; (4) 4.8Ni/20Nd/SBA-15.

The stabilities of these catalysts over 50 h were evaluated to determine their long-term stabilities; the results are also shown in Fig. 5. The Nd-modified catalysts were more stable than 4.8Ni/SBA-15; 4.8Ni/5Nd/SBA-15 and 4.8Ni/10Nd/ SBA-15 were the most stable. The reason for the stabilities of the Nd-modified catalysts in CRM to syngas might be that Nd modification decreased the NiO particle sizes,which improved the abilities of the catalysts to eliminate deposited carbon [ 22 ].

The catalytic performance of 4.8Ni/10Nd/SBA-15 was also compared with those of 4.8Ni/10La/SBA-15 and 4.8Ni/10Ce/ SBA-15; the results are shown in Fig. 6. It can be seen that the initial CH4 (84%-88%) and CO2 (92%-94%) conversions were similar over these rare-earth-modified catalysts,and all three catalysts were stable in CRM under the reaction conditions used.

Fig. 6. Catalytic performance of different rare-earth-modified 4.8Ni/10RE/SBA-15 catalysts in CRM. Reaction conditions: atmospheric pressure,800 °C,V(CH4)/V(CO2) = 1,GHSV = 3.5 × 104 mL g-1 h-1. (1) 4.8Ni/SBA-15; (2) 4.8Ni/10Nd/SBA-15; (3) 4.8Ni/10La/SBA-15; (4) 4.8Ni/10Ce/SBA-15.

There have been many reports on rare-earth-modified Ni-based catalysts in CRM to syngas. Wang et al. [ 23 ] used a subsequent impregnation method to prepare CaO-,La2O3-,or CeO2-modified Ni/γ-Al2O3 catalysts and used these catalysts in CRM. The results showed that under the conditions 700 °C,V(CH4)/V(CO2) = 1,and GHSV = 1.8 × 104 mL g-1 h-1,CO2 conversions were about 80% and stable for 24 h. Slagtern et al. [ 24 ] prepared a catalyst modified with a rare-earth mixture,Ni/Ln/Al2O3 (Ln = rare-earth mixture,elemental composition 66.1 wt% La,25.4 wt% Nd,and 7.7 wt% Pr),and found that CH4 conversion reached about 90% under the conditions 800 °C,1 atm,CH4:CO2:N2 = 2:2:1,and total flow rate 50 mL/min. However,the CH4 conversion dropped by more than 30% after testing for 600 h. Guo et al. [ 17 ] reported that the addition of Gd2O3 greatly improved the catalytic activity of Ni/SiO2 in CRM,but the modified catalyst was not stable during long-term evaluation,with the CH4 conversion decreasing from 63.9% to 54.3% within 6 h. Li et al. [ 25 ] used Ni-0.75 wt% La-BaTiO3 in CRM to study the effects of La addition; the results showed th at the resistance of Ni-0.75 wt% La-BaTiO3 to carbon deposition was higher than that of Ni-BaTiO3.

3.3. Physicochemical properties of spent Ni/RE/SBA-15 catalysts

The crystalline structures of the spent catalysts were determined using XRD. The wide-angle XRD patterns (Fig. 7) showed wide,weak peaks attributable to amorphous SiO2 at 10°-30° and weak peaks at 43.0° ascribed to Ni,i.e.,during the reaction,slight sintering of Ni particles occurred. In the XRD profiles,sharp peaks from carbon appeared over the 4.8Ni/ SBA-15 catalyst,but these were not observed over the other catalysts,showing that more carbon was deposited on the spent 4.8Ni/SBA-15 catalyst.

Fig. 7. XRD patterns of the spent 4.8Ni/xRE/SBA-15 catalysts.

The amounts of carbon deposited on the spent catalysts were determined using TGA; the profiles are shown in Fig. 8. There were two weight loss peaks for all the spent catalysts at 100-200 and 500-700 °C. The weight-loss peaks at 100-200 °C resulted from the removal of moisture and impurities,whereas those at 500-700 °C were caused by the removal of deposited carbon [ 26 ]. Generally,carbon species removed at temperatures below 600 °C are active carbons,whereas those removed above 600 °C are inactive carbons such as nanotubes and nanofibers [ 27 ]. It is therefore speculated that,in our case,both active and inactive carbons were deposited over these spent catalysts. The amounts of carbon deposited over these catalysts were calculated based on the weight loss peaks at 500-700 °C; the results are listed in Table 1. It can be seen that the amounts of carbon deposited on the rare-earth-modified 4.8Ni/xRE/ SBA-15 catalysts were smaller than that on 4.8Ni/SBA-15,probably because the small NiO particles promoted carbon elimination. For the catalysts modified with different contents of Nd,less carbon was deposited over the spent 4.8Ni/5Nd/ SBA-15 and 4.8Ni/10Nd/SBA-15 catalysts,which indicated that appropriate amounts of rare-earth metals improved carbon removal during CRM to syngas.

4. Conclusions

Rare-earth metal (Nd,Ce,and La) oxides were used to modify a Ni/SBA-15 catalyst. The results showed that Nd,Ce,and La modification had little effect on the textural properties and crystalline phases of the obtained catalysts but influenced the reduction of NiO species. The addition of Nd favored interactions between Ni and support,possibly by formation of Ni-Nd-O species. Catalysts modified with 5% Nd,10% Nd,10% La,and 10% Ce exhibited higher initial catalytic activities in CRM and showed stronger resistance to carbon deposition.

References
[1] Choudhary V R, Mondal K C, Mamman A S, Joshi U A. Catal Lett, 2005, 100: 271
[2] Zhang M L, Ji S F, Hu L H, Yin F X, Li C Y, Liu H. Chin J Catal (张美丽, 季生福, 胡林华, 银凤翔, 李成岳, 刘辉. 催化学报), 2006, 27: 777
[3] Sutthiumporn K, Kawi S. Int J Hydrogen Energy, 2011, 36:14435
[4] Natesakhawat S, Watson R B, Wang X Q, Ozkan U S. J Catal, 2005, 2: 496
[5] Pereniguez R, Gonzalez-Delacruz V M, Holgado J P, Cabollero A. Appl Catal B, 2010, 93: 346
[6] Serrano-Lotina A, Martin A J, Folgado M A, Daza L. Int J Hydrogen Energy, 2012, 37: 12342
[7] Hally W, Bitter J H, Seshan K, Lerchert J A, Ross J R H. Stud Surf Sci Catal, 1994, 88: 167
[8] Yuan C Y, Wei Y X, Li J Z, Xu S T, Chen J R, Zhou Y, Wang Q Y, Xu L, Liu Z M. Chin J Catal (袁翠峪, 魏迎旭, 李金哲, 徐舒涛, 陈景润, 周游, 王全义, 许磊, 刘中民. 催化学报), 2012, 33: 367
[9] Xu L L, Song H L, Chou L J. Appl Catal B, 2011, 108: 177
[10] Nandini A, Pant K K, Dhingra S C. Appl Catal A, 2005, 290: 166
[11] Barroso-Quiroga M M, Castro-Luna A E. Int J Hydrogen Energy, 2010, 35: 6052
[12] Ocsachoque M, Bengoa J, Gazzoli D, Gonzalez M G. Catal Lett, 2011, 141: 1643
[13] Daza C E, Cabrera C R, Moreno S, Molina R. Appl Catal A, 2010, 378: 125
[14] Zhu J Q, Peng X X, Yao L, Shen J, Tong D, Hu C. Int J Hydrogen Energy, 2011, 36: 7094
[15] Ikeguchi M, Mimura T, Sekine Y, Kikuchi E, Matsukata M. Appl Catal A, 2005, 290: 212
[16] Guo J Z, Hou Z Y, Gao J, Zheng X M. Fuel, 2008, 87: 1348
[17] Guo J Z, Hou Z Y, Zheng X M. React Kinet Mech Catal, 2010, 101: 129
[18] Liu H M, Li Y M, Wu H, Takayama H, Miyake T, He D H. Catal Commun, 2012, 28: 168
[19] Zhou W, He D H. Green Chem, 2009,11: 1146
[20] Xu L H, Mi W L, Su Q Q. J Nat Gas Chem, 2011, 20: 287
[21] Zhang S H, Muratsugu S, Ishiguro N, Tada M. ACS Catal, 2013, 3: 1855
[22] Yamazaki O, Nozaki T, Omatak Y K, Fujimoto K. Chem Lett, 1992: 1953
[23] Wang S, Lu G Q. J Chem Technol Biotechnol, 2000, 75: 589
[24] Slagtern A, Olsbye U, Blom R, Dahl I M, Fjellvag H. Appl Catal A, 1997, 165: 379
[25] Li X, Li S, Yang Y, Wu M, He F. Catal Lett, 2007, 118: 59
[26] Liu B S, Au C T. Appl Catal A, 2003, 244: 181
[27] Wang N, Chu W, Zhang T, Zhao X S. Int J Hydrogen Energy, 2012, 37: 19
Nd, Ce和La改性对Ni/SBA-15催化剂在CH4/CO2重整反应中性能的影响
刘会敏, 李宇明, 吴昊, 杨维维, 贺德华     
清华大学化学系, 有机光电子与分子工程教育部重点实验室, 北京100084
摘要:以稀土金属Nd,Ce或La的氧化物为助剂,采用β-环糊精浸渍法对Ni/SBA-15催化剂进行了改性,并运用X射线衍射、N2吸附-脱附、程序升温还原和热重等手段考察了改性的催化剂在CO2重整CH4制合成气反应中的催化性能. 结果表明,Nd等稀土金属氧化物的添加对催化剂孔结构和晶相结构等性质影响不大,但可影响NiO的还原; Nd的添加使NiO与载体之间以Ni-Nd-O形式相互作用,促进了活性组分NiO的还原. 其中,Nd的添加量为5-10 wt%时所制备的催化剂在重整反应中的催化活性最高,且具有很强的抗积碳性能. La和Ce氧化物促进的Ni催化剂也表现出类似的性质和催化性能.
关键词稀土金属     钕改性          SBA-15催化剂     甲烷/二氧化碳重整     积碳    

1. 前言

CH4/CO2重整制合成气(CRM)反应是以CH4和CO2这两种温室气体为原料制备具有高附加值化工产品的过程, 为催化领域的研究热点之一[ 1 ].  在CRM反应中, Ni基催化剂因价格低廉且具有优异的初始催化活性而被广泛研究[ 2, 3 ], 但易因积碳而失活[ 4 ].  目前, 有关抑制Ni基催化剂积碳的研究较多[ 5, 6 ].  除Ni晶粒尺寸外[ 7 ], 催化剂的酸碱性也会对催化剂的抗积碳性能有很大影响[ 8, 9 ].  

碱金属和碱土金属常作为碱性助剂添加于催化剂中.  当K含量在1.0-2.9 wt%时, Ni-K/Al2O3催化剂的活性随K含量增加而提高[ 10 ], 而Li修饰的Ni/CeO2催化剂表现出更强的抗积碳性能[ 11 ].  稀土金属在调变催化剂酸碱性方面表现出类似的性质, 如Ce [ 12, 13 ], La [ 14 ], Sm[ 15 ]和Gd [ 16, 17 ]等. 而有关Nd修饰Ni基催化剂并应用于CH4/CO2重整的报道较少.  

本课题组曾采用β-环糊精(CD)改性浸渍法制得具有介孔结构的Ni/SBA-15-CD(1/50)催化剂, 活性组分Ni晶粒小, 可较好地分散于载体SBA-15上, 在CH4/CO2重整制合成气反应中(常压, 700 °C, CO2/CH4 = 1, GHSV = 1.28× 104 mL g-1 h-1)显示出较高的催化活性及稳定性[ 18 ].  然而, 当GHSV提高至3.5×104 mL g-1 h-1时, 由于接触时间较短, Ni/SBA-15-CD催化剂上初始CH4转化率和初始CO2转化率有所降低, 且随着反应时间的延长而逐渐下降.  为了进一步提高催化剂在高空速下的催化性能, 本文采用稀土金属Nd, La和Ce作为助剂, 调变催化剂表面性质, 着重研究了Nd助剂对Ni基催化剂性能的影响.  

2. 实验部分
2.1. 催化剂制备

以硅酸四乙酯(TEOS)为硅源, P123 (分子量5800, Aldrich)为模板制备载体SBA-15[ 19 ].  

稀土金属(RE)改性的Ni/RE/SBA-15 (RE = Nd, Ce或La)催化剂和未修饰的Ni/SBA-15催化剂均采用分步的β-环糊精(CD, 北京生物科技公司)改性的浸渍法制备.  投料量为n(CD)/n(RE) = 1/50和n(CD)/ n(Ni) = 1/50.  浸渍所得催化剂前驱体于650 °C焙烧5 h即得到相应催化剂.  Ni和RExOy的理论负载量分别为4.8 wt%和5-20 wt%, 催化剂样品记为4.8Ni/xRE/SBA-15, 其中x为RExOy含量.  

2.2. 催化剂表征

载体及催化剂的比表面积、孔体积及孔径分布在Micromeritics ASAP-2010C型仪器上通过N2物理吸附法测定, 测定前样品在200°C用N2处理2 h.  催化剂的规整介孔结构及其晶相结构采用X射线衍射法(XRD)在Bruker D8 Advance型X射线衍射仪上测定, Cu Kα辐射源, Ni滤色片.  活性组分及载体的晶粒大小利用Scherrer公式计算得到.  催化剂中各元素的实际含量在Shimadzu XRF-1800型X射线荧光光谱仪上测定.  催化剂上活性组分的还原特性在Quantachrome公司的CHEMBET 3000型吸附仪上通过H2程序升温还原法(H2-TPR)测定[ 18 ].  使用后催化剂上的积碳量通过热重(TGA)分析, 所用仪器为Mettler Toledo TGA/SDTA851e;  称取适量催化剂, 在空气气氛下以10 °C/min从室温升至800 °C, 记录样品在升温过程中的质量损失.  

2.3. 催化剂评价

CH4/CO2重整反应在常压石英管(内径5 mm)固定床微型反应器中进行.  反应前先用20% H2/Ar混合气将催化剂(0.20 g)在600 °C还原2 h, 然后升温至800 °C并切换为CO2/CH4混合气(n(CO2)/n(CH4) = 1, 115 mL/min)进行反应.  反应过程中, 每隔1 h对尾气用气相色谱进行在线分析, 并通过皂沫流量计测定尾气流速.  相应反应物转化率和产物选择性计算如下:  

CH4转化率 = 1 − Vout(CH4)/Vin(CH4)

CO2转化率 = 1 − Vout(CO2)/Vin(CO2)

CO选择性 = Vout(CO)/{[Vin(CH4) − Vout(CH4)] + [Vin(CO2) − Vout(CO2)]}

H2选择性 = Vout(H2)/{2[Vin(CH4) − Vout(CH4)]}

其中, VinVout分别为原料气和尾气中各气体的流速.  

3. 结果与讨论
3.1. 新鲜催化剂Ni/RE/SBA-15的物理化学性质
3.1.1. 织构性质

图1为新鲜4.8Ni/xNd/SBA-15催化剂吸脱附等温线及孔径分布.  可以看出, 所有催化剂都具有IV型吸脱附等温线及H2型回滞环, 说明Nd和β-CD的添加并未破坏SBA-15的规整介孔结构.  此外, 所有催化剂的孔径分布在3-7 nm, 差别不大.  此外, 4.8Ni/10La/SBA-15和4.8Ni/ 10Ce/SBA-15催化剂的吸脱附等温线形状及孔径分布与4.8Ni/10Nd/SBA-15差别不大.  

表1为新鲜4.8Ni/xRE/SBA-15催化剂的织构性质.  由表可见, 未改性的4.8Ni/SBA-15催化剂的比表面积最大(578.3 m2/g), 经La, Ce和Nd改性后, 催化剂的比表面积均有所下降(357-396 m2/g).  这可能是由于4.8Ni/ SBA-15经高温焙烧一次, 而4.8Ni/xRE/SBA-15为两次, 从而导致SBA-15和Ni晶粒不同程度的烧结.  

3.1.2. 晶相结构

图2为新鲜4.8Ni/xNd/SBA-15催化剂的XRD谱.  由图可见, 除了SiO2的衍射峰(10°-30°)外, 各样品均出现NiO的衍射峰(37.0°, 43.2°和62.8°).  由于这些衍射峰都比较微弱, 不能用Scherrer公式计算NiO的晶粒大小, 这意味着这些催化剂上的NiO颗粒较小(小于5 nm), 具有较好的分散.  

La, Ce修饰的4.8Ni/10RE/SBA-15催化剂上也观察到微弱的NiO衍射峰, 表明Ni颗粒分散也较好.  

如表1所示, 几种催化剂上Ni和稀土金属的实际含量都与理论投料量相近.  

3.1.3. NiO的还原行为

图3为4.8Ni/10RE/SBA-15催化剂的H2-TPR谱.  可以看出, 各样品的TPR谱形状比较接近, 都出现了两个NiO还原峰, 分别位于500-600和800 °C.  此外, RE的加入使得催化剂上NiO的还原峰向低温移动, 即促进了NiO的还原.  还可以看出, 随着Nd含量增至10 wt%, 样品的NiO还原峰明显向低温移动;  至10-20 wt%时则变化不大.  此外, 10 wt%的La和Ce修饰催化剂的TPR谱与10 wt% Nd修饰的接近.  

3.1.4. 碱性

图4为xNd/SBA-15载体的CO2-TPD谱.  可以看出, 各样品在100-500 °C均未出现明显的CO2脱附峰, 即Nd的添加对载体碱性影响不大.  

3.1.5. 稀土金属对Ni/RE/SBA-15性质的影响

对稀土金属的修饰效果有几种不同的观点.  Xu等[ 20 ]发现La和Ce的添加使得Ni/Al2O3上Ni与Al2O3间相互作用大大增强.  Zhang等[ 21 ]发现Ce修饰的Ni/SBA-16催化剂上形成了Ni−Ce−O物种, 该物种会促进NiO颗粒的分散及NiCe/SBA-16的还原.  由于本文中RExOy含量高达20 wt%时依旧观察不到RExOy的衍射峰(图2), 因此也可推断RExOy同样以Ni−RE−O的形式存在, 从而促进NiO颗粒的分散及NiO的还原.  

3.2. Ni/RE/SBA-15催化剂在CH4/CO2重整反应中的催化性能

图5为各4.8Ni/xNd/SBA-15催化剂上CH4/CO2重整反应结果.  可以看出, 4.8Ni/SBA-15催化剂的初始活性最低:  CO2和CH4转化率分别为81%和67%;  而5 wt% Nd的添加使得CO2和CH4初始转化率分别增至95%和90%.  随着Nd含量继续增加到20%时, 催化剂CO2和CH4初始转化率分别逐渐降至91%和84%.  

由图5还可见, Nd的加入使得Ni/SBA-15催化剂更稳定.  其中, 当Nd含量为5 wt%和10 wt%时最为稳定.  这可能是由于Nd的添加降低了催化剂上NiO的晶粒大小, 提高了催化剂的抗积碳性能[ 22 ].  

图6比较了4.8Ni/10La/SBA-15, 4.8Ni/10Ce/ SBA-15和4.8Ni/10Nd/SBA-15的催化性能.  可以看出, 各催化剂上CO2和CH4初始转化率都很接近, 且非常稳定.  

Wang等[ 23 ]采用分步浸渍法制备了CaO, La2O3或CeO2修饰的Ni/γ-Al2O3催化剂并用于CRM反应 (700°C, V(CH4)/V(CO2) = 1, GHSV = 1.8×104 mL g-1 h-1), 在24 h内CO2转化率保持在80%左右.  Slagtern等[ 24 ]制备的多种稀土金属混合物修饰的Ni/Ln/Al2O3 (Ln为稀土金属混合物, 包含66.1 wt% La, 25.4 wt% Nd和7.7 wt% Pr)在CRM反应中CH4初始转化率在90%左右, 但反应 600 h后活性降低约30%.  Guo等[ 17 ]发现Gd2O3的添加显著增加了Ni/SiO2在CRM反应中的催化活性, 但催化剂并不稳定, 反应6 h后CH4转化率从63.9%降到54.3%.  Li等[ 25 ]将Ni-0.75wt%La-BaTiO3催化剂应用于CRM反应, 发现0.75 wt% La的加入提高了催化剂的抗积碳性能.  

3.3. 反应后的催化剂的物理化学性质

图7为反应后催化剂的XRD谱.  可以看出, 除了SiO2的衍射峰(10°−30°)外, 各样品都出现了弱的Ni衍射峰(43.0°), 表明在反应过程中Ni颗粒有一定程度的烧结.  此外, 4.8Ni/SBA-15上还出现了尖锐的碳衍射峰, 说明该催化剂上的积碳量较大.  

图8为反应后催化剂的TGA曲线.  可以看出, 所有催化剂样品均出现两个失重峰, 其中100−200 °C的失重峰归属于水汽和杂质的移除, 而500−700 °C的失重峰则对应于积碳的移除[ 26 ].  一般来说, 在600 °C以下可以被消除的碳为活泼碳, 而600 °C以上的是惰性碳, 如碳纳米管和碳纳米纤维[ 27 ].  可以推测, 本文催化剂上的积碳既有活泼碳, 又有惰性碳.  根据500−700 °C的失重峰计算了相应催化剂上的积碳量, 结果见表1.  未改性的Ni/SBA-15催化剂上的积碳量最大, 而经稀土金属改性后, 催化剂上的积碳量均有所下降.  这可能是由于稀土金属的添加降低了催化剂上活性组分的晶粒大小, 提高了消碳能力.  另外, 4.8Ni/5Nd/SBA-15和4.8Ni/10Nd/ SBA-15上的积碳量更少, 表明适量Nd的添加有利于催化剂消碳.  

4. 结论

制备了Nd, Ce和La氧化物修饰的4.8Ni/SBA-15催化剂.  结果显示, Nd, Ce和La的修饰对催化剂孔结构和晶相结构等性质影响不大, 但对NiO的还原有影响;  其中Nd的添加使NiO与载体之间以Ni−Nd−O形式相互作用, 促进了活性组分NiO的还原.  用5% Nd, 10% Nd, 10%La和10% Ce修饰的催化剂在CH4/CO2重整反应中表现出更高的初活性和更强的抗积碳性能.