NATURAL GAS CONVERSIONV Studies in Surface Science and Catalysis,Vol. 119 A. Parmalianaet al. (Editors) 9 1998 Elsevier Science B.V. All rights reserved.
Effects of Alkali and Rare Earth Metal Oxides on the Thermal Stability and the Carbon-deposition over Nickel Based Catalyst Shenglin Liu Guoxing Xiong* Shishan Sheng Qing Miao Weishen Yang State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, P.O. Box 110, Dalian 116023, P.I~ China ABSTRACT Effects of alkali and rare earth metal oxides on the thermal stability and the carbondeposition over nickel based catalyst for partial oxidation of methane to syngas were investigated by a series of characterization techniques including flow-reaction, TG, TPO, XPS, XRD and BET. The results indicated that the introduction of Li and La oxides could suppress the carbon-deposition on the nickel based catalyst. LiNiLaO/y-Al203 catalyst not only possessed excellent reaction performance (CH4 conversion > 96%, CO selectivity > 98%), carbon-deposition resistance and improved the thermal stability of the nickel based catalyst, but also had comparatively stable porous structure and stable crystallity during the 200h life test experiment under the conditions of reaction temperature at 1123K, CH4/O2 ratio of 1.96 and space velocity of 2.7 xl041/kg.h.
1. INTRODUCTION Partial oxidation of methane to syngas (POM) has received intense attention since 199011,2]. The catalysts used in POM reaction can be classified into two types: (1) noble metal (Pd, Rh, Ru, Pt, It) catalysts; (2) transition metal(Ni, Co, Fe) catalysts. The nickd based catalysts emerged as the most practical one because of their high turnover rates and inexpensive cost. However, nickel based catalysts supported on y-A1203 are usually unstable at high temperature. Preventing the alumina support from thermal deterioration was studied by many authors [4 - 6] and a large number of additives to the y-AI203 support have been shown to inhibit sintering and phase transformation, such as SiO2, BaO, CeO2, La203, etc. The other reason causing the deactivation of NiO/Al203 catalyst is the loss and sintering of nickel that results in certain decreasing of active surface and some change of the interaction between active compound and support. In addition, the carbon-deposition over NiO/AI203 is also often considered as a possible cause of deactivation for the catalyst. Schmidt found the loss of nickel on the 3wt% Ni/AI203 catalyst during the POM reaction . The serious carbon-deposition over Ni/AI203 that filled the catalyst pores and caused the granules to disinergrate into a fine Author to whom all correspondence should be addressed
748 powder at CH4/O2ratio _> 2 was reported by Ltmsford . Hence, the stability of NiO/A1203 catalyst is one of the important considerations before this process can be used on an industrial scale. Previously, we reported that the ABNiO/y-A1203 (A=Li, Na, K; B=La, Sm, Ce, Y) catalysts were excellent POM reaction catalysts[9,10]. Meanwhile, we discussed the oxidative transformation of methane over the nickel based catalysts modified by alkali and rare earth metal oxides. In the present paper, effects of alkali and rare earth metal oxides on the thermal stability and the carbon-deposition over nickel based catalyst and the 200h life test experiment of LiNiLaO/y-A1203 for POM reaction were investigated by a series of characterization techniques including flow-reaction, TG, TPO, XPS XRD and BET.
2.1. Preparation of catalysts The catalysts were prepared by impregnating appropriate amounts of LiNO3, Ni(NO3)2 and La(NO3)3 (for LiNiLaO/y-Al203, Ni:2.7wt%; Li20:1.25 wt%; La203:7.0wt%) and Ni(NO3)2 (for NiO/y-A1203 , Ni:2.Twt%), respectively, on y-A1203 support for 24 h, followed by drying at 393K and calcination in air at 823-1073K for 4 h. 2.2. Test of Catalytic Performance Catalysts were tested in atmosphere pressure fixed-bed microreactors. The carbon-deposition over catalysts was tested using a microreactor with an internal diameter of 4 mm with 100mg catalyst employed. The 80h and 200h life test experiments were performed using a microreactor with an internal diameter of 8 mm with 500rag catalyst employed. The analysis of reaction products was provided elsewhere[ 11]. 2.3. Characterizations of Catalysts TG test was performed with a Perkin-Elmer TGS-2 instrument. The TG profiles were recorded and treated by Perkin-Elmer 3600 work station at a programmed temperature rate of 10K/rain in the air with the flow rate 25ml/min. TPO profiles were tested by on-line Mass Sepectroscopy (m/e: 44(CO2), 28(CO or CO2), 18(H20)) at a programmed temperature rate of 16K/rain in a 5vo1% O2/He flow. X-ray photoelectron spectoscopy (XPS) characterization was performed using a VG ESCA LABMKr spectrometer. The sample chamber was pumped to a pressure of lxl0 9 Tort and monochromatised Mg exciting radiation was used. The Cls peak at 284.6ev due to adventitious carbon was used as an internal standard. The surface relative atomic ratio of catalyst was calculated according to the spectra line peak intensity of XPS. X-ray diffraction(XRD) characterization of the catalysts was performed with a Riguku D/Max-RB X-ray diffractometer using a copper target at 40KV x 100mA and a scanning speed of 8 degree/min. The specific surface area and pore volume of samples were determined by the BET nitrogen method in a volumetric equipment OMNISORP-100CX.
749 3. R E S U L T S A N D D I S C U S S I O N
3.1. Comparison of carbon deposition between NiO/7-A1203 and LiNiLaO/y-Ai203 The introduction of Li and La improves the ability of carbon-deposition resistance, besides the activity of the nickel based catalyst. Tile deposition of surface carbon over the NiO/AI203 catalyst during the POM reaction results in the deactivation of NiO/AI20317,8,12]. It is well known that the acidity of catalyst surface favors carbon-dep0sition and the basity of catalyst surface prevents carbon-deposition. Hence, the addition of Li and La oxides is reasonable to prevent carbon-deposition over catalyst surface. The TG results of the samples after the POM reaction forl0h (CI-h/O2 = 2; GHSV = 27000 1/kg.h; 1123K; latm) indicated that the carbon-deposition resistant ability of LiNiLaO/y-Al203 was much better than that of NiO/y-Al203. During the period of 10 h of POM reaction, there was almost no carbon deposition over LiNiLaO/y-Al203 and the amount of carbon-deposition over LiNiLaO/AI203 was only 0.24% of its net weight, but that over NiO/y-Al203 was about 12% of its net weight. It was shown that the introduction of Li and La oxides could suppress the carbon-deposition of the nickel based catalyst. TPO tests were also performed to demonstrate the above results. According to the TPO results (see Fig. 1), the only gaseous product was CO2 during the TPO tests of NiO/y-A1203 and LiNiLaO/y-Al203 catalysts after POM reaction for 10h under the same reaction conditions, and no CO and H20 were detected. Two CO2 desorption peaks at 813K and 1013K for NiO/yAbO3 and two at 793K and 933K for LiNiLaO/y-Al203 appeared in the profiles, which indicates there were two different kinds of carbon produced during the POM reaction. The amount of carbon over NiO/y-Al203 was far more than that over LiNiLaO/y-A1203 and it was more difficult to be depleted the carbon by oxygen.
678 803 Temperature (K)
Figure 1. TPO profiles of catalysts after POM reaction for lOh (m/e = 44) a: NiO/y-AI203 b" LiNiLaO/y-A1203
750 Another interesting phenomenon was that no H20 was detected, and it was inferred that the carbon species produced during the POM reaction was carbon but no other carbon species with hydrogen atom
3.2. Life test of LiNiLaO/~/-AI203 catalyst 3.2.1 The catalytic performance and stability Previously, we investigated the 50h life test of LiNiLaO/~/-A1203 catalyst for POM reaction and found that CH4 conversion kept exceeding 96% and CO selectivity 95% during the 50h life-test[ 10]. In the present work, we performed the 200h life test of LiNiLaO/~/-AI203 catalyst for POM reaction. The 200h life test experiment was performed using a microreactor with an internal diameter of 8 mm instead of 4 mm (for 50h life test experiment), and the weight of catalyst was 500 mg instead of 100 mg (for 50h life test experiment), while the stability of LiNiLaO/y-Al203 catalyst was studied under the conditions of reaction temperature at 1123K, CH4/O2 ratio of 1.96 and spaces velocity of 2.7 xl04 1/kg.h (see Fig. 2). During the 200 h life test experiment, CI-I4 conversion kept exceeding 96.00%, H2 selectivity and CO selectivity kept exceeding 98.00% and 96.00%, respectively. Those values approached that of the thermodynamic equill"bdun~ It suggested that the LiNiLaO/1r-A1203 catalyst was quite stable during high temperature reaction and the introduction of Li and La could improve the thermal stability of the nickel based catalyst.
tO o} I,_
o -r 0
~ 100 Time
Figure 2. Performance of catalyst as a function of time (CH4/O2=1.96; GHSV =2.7 x 104 l/(kg.h); 1123K; 1 atm)
3.2.2. The catalyst structure and crystaHity analysis After the 80h or 200 h life test experiment, the reactor was cooled to room temperature in N2 gas flow. The catalyst was then investigated by BET method(see table 1). The results of BET showed that the catalyst pore volume changed little. In comparison with the specific
751 surface area of fresh catalyst(Cat-0h), that of Cat-80h decreased by-20%, but the specific surface aYea of Cat-200h decreased only by--,3% (Compared with that of Cat-80h).Those results indicated the catalyst pore structure was comparatively stable during high temperature reaction. The XRD tests were performed to determine the crystallity of those catalysts(see Fig.3). The XRD spectra of Cat-0h (reduced) was similar to that of Cat-200h. Only y-AI203 and reduced nickel appeared, NiO and NiAI204 did not appear, namely, after the 200h life test experiment, the support y-AI203 did not undergo crystallity .transformation into ~-A1203. The results indicated that the reduced nickel was the active center of LiNiLaO/y-A1203 for POM reaction (It was consistent with other paper [ 11]) and the catalyst could kept the acidic property (that favors keeping the reduced nickel[l I ]) during high temperature reaction. It can be concluded that LiNiLaO/y-Al203 catalyst had stable crystallity structure during high temperature reaction.
Table 1 Comparison of Porous volume, Specific surface area and Surface composition of catalysts Sample*
Specific surface area
ml/g 0.60 0.62 0.58
m2/g 120 96 93
Cat-0h Cat-80h Cat-200h
Surface relative atomic ratio (XPS) C 0.45
Cat- Oh: The fresh LiNiLaO/y-A1203 catalyst Cat - 80h: The catalyst that has reacted for 80h, then was cooled to room temperature in N2 gas flow Cat- 200h: The catalyst that has reacted for 200h, then was cooled to room temperature in N2 gas flow
a . . . . w ~"~:'~"'.~."',~"~'~'#'~'v'"~"~ tm'~l~'~tc't~
A [_--] .
~r .... ~/ ~"~.,,,,~..~..~.~.~,,~,..~:'~":w~.'~ 9!
, ,, ~" n~
20 Figure 3. XRD spectra of LiNiLaO/y-AlzO3 before and after POM reaction a: Cat-0h(reduced) b: Cat-200h Cat -Oh (reduced): The LiNiLaOh/-A1203 catalyst that was cooled to room temperature h~ Ar gas flow, after it was pretreated by 5% H2/Ar
752 3.2.3 The carbon deposition and surface composition of the catalysts After the 200 h life test experiment, the reactor was cooled to room temperature in N2 gas flow. The catalyst was then investigated by TG (no shown) and XPS(see Table 1), the results of TG and XPS all indicated that the used catalyst had almost no carbon-deposition. The results demonstrated that the LiNiLaO/T-A1203 catalyst had excellent resistance to carbon deposition during high temperature reaction The surface relative atomic ratio of catalyst was investigated by XPS(see Table 1). The results indicated that the loss of Li on the surface of the catalyst was not serious and the intensity of La changed slightly. And the surface nickel lost obviously, but the reaction performance did not decrease during 200h life test experiment, which suggested that this amount of lost nickel was not enough to cause any change of reaction performance at low space velocity such as 2.7 x l 0 4 l/kg.h. Previously, the influence of space velocity on the performance of LiNiLaO/y-A1203 (Ni:2.Twt%)for POM reaction was reported. The results indicated that CH4 conversion decreased at space velocity >__9 x 104 l/kg.h[ 10]. In the present work, the effect of the amount of nickel loaded on the reaction performance was discussed. The similar results were obtained, that was, CH4 conversion decreased at space velocity >_9 x 104 l/kg.h when the amount of nickel over y-A12Os was 2.7wt%. However, when the amount of nickel increased to 10.0 wt%, the reaction performance did not decreased (CH4 conversion 96%), even though the space velocity was increased to 5 xl 05 l/kg.h. So it could be concluded that the appropriate amount of nickel over T-AI203 must be preserved in order to keep high performance for long period at high space velocity and temperature. The work will be published in other articles. ACKNOWLEDGEMENT The financial support of the National Natural Science Foundation of China is gratefully acknowledged. REFFRENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.
A.T. Ashcrofi, A. K. Cheetham, and P. D. F. Vernon, Nature, 344(1990)319. S.C.Tsang, J. B. Claridge, M. L. H. Green, Catal.Today, 23(1995)3. W.G. Schlaffer, C. Z. Morgan and J. N.Wilson, J. Phys. Chem., 61(1957)714. S. Blonski, S. H. Garofalim~ Catal. Lett., 25(1994) 324. M. Pijolat, M. Dauzat, and M. Soustelle, Solid State Ionics, 50(1992)31. Z. 1L Ismagilov, 1L A. Shkrabina, N. A. Koryabkina, Catal. Today, 24(1995)269. P.M.Tomiainen, X. Chu, L. D. Schmidt, J. Catal., 146(1994) 1. D. Dissanayake, M. P. Rosynek, J. H. Lunsford, J. Catal., 132(1991) 117. G.X. Xiong, Q. Miao, China Patent, Appl. No. 95110071.8. Q. Miao, G. X. Xiong, and X. X. Guo, Appl. Catal. A, 154(1997) 17. Q. Miao, G. X. Xiong, and X. X. Guo, Stud. Surs Sci. Catal., 101(1996)453. J. B. Claridge, M. L. H. Green, and P. D. Battle, Catal. Lett., 22(1993)299. D. L. Trimm~ Design of Industrial Catalyst, Elsevier Scientific Publishing Company, Amsterdam-Oxford-New York, 1980. 14. C. H. Bartholomew, Catal. Rev. -Sci. Eng., 24(1982) 67.