Stabilization of Cu/ZnO Based Catalysts by Nickel Additive in Methanol Decomposition

Vol. 17No. 42001Chemical Research in Chinese Universities420~ 425Stabil ization of Cu/ ZnO Based Catalysts by NickelAdditive in M ethanol Decomposition*XI Jing-yu, WANG Zhi-fei and LU Gong-xuan(Stute Key Iahoratory for Oxo Synthesis and Selective Oxidation, Lanzhou Inslitute of Chermical Physics,The Chinese Academy of Sciences, Lanzhou 730000, P. R. China)(Received Sept. 19, 2000)A series of Cu/Zn based catalysts with and without Ni, prepared by the co-precipitationmethod , has been studied for methanol decomposition. CO and H2 are the major products. TheCu/Zn catalysts show a low initial activity and a poor stability. The formation of the CuZn alloyswas observed in the dcactivated Cu/Zn catalysts which were used for met hanol decomposition at250 C. When small amounts of Ni were added in the catalyst, the Cu/Zn/Ni(molar ratio 5/4/x) catalysts showed a bigh activity at a low temperature. The activity and the stability of thecatalyst depend on the nickel content. The activity of the Cu/Zn/Ni catalysts was maintained ata relatively stable value of 78% conversion of methanol with 95% selectivity of H2，93%selectivity of CO, and a more than 70% yield of hydrogen was obtained at 250 C when r≥1.The stability of the Cu/Zn/Ni (molar ratio 5/4/x) catalysts showed the maximum(ca 88%)when x= 1. The stabilization effect of nickel on the Cu/Zn based catalysts may lead to theincreasing of the dispersion of active Cu species and the prevention of CuZn alloys formation.Keywords Nickel, Cu/Zn catalysts, Stabilization, Methanol，DccompositionArticle ID 1005-9040(2001)-04-420-06IntroductionHydrogen，whether used directly as a fuel in internal combustion engines[1-41 Oindirectly in supplying electricity by fuel cells(1-)]， is an inherently clean burning andpotentially high energetic fuel. The major impediment of the wide use of hydrogen is thedifficulties in hydrogen distribution and storage[2,3]. Methanol is one of the most hopefulcandidates as an energy carrier for the future. The decomposition of methanol on- board intohydrogen and carbon monoxide(CH:OH→CO十2H2) has drawn a keen interest recently uponaiming at solving the problems in hydrogen usage[5-7).The high activity and stability of the catalyst are particularly needed for the on- boardhydrogen generation from methanol decomposition. A Cu/Zn catalyst has traditionally beenused for the methanol synthesis'[1,3.8]. It may also be used for the decomposition of methanolinto CO and H2. The previous studies have been中国煤化工n Cu/Zn basedcatalysts's59.10]. However, the low activity and the poorGeen the problemsYHCNMHG# Supported by the Foundation of 973(G20000264).* * To whom correspondence should be addressed.No.4XI Jing-yu, WANG Zhi-fei, LU Gong-xuan et al.421when they are used in methanol decomposition[1o,11]. In this work, a new Cu/Zn/Ni catalysthas been developed, in which Ni as a third component can significantly increase the activityand the stability of Cu/ZnO catalysts. The cause of the deactivation of Cu/ Zn-based catalystsfor methanol decomposition has also been discussed.Experimental1 Catalyst PreparationMethanol decomposition catalyst was prepared according to the co-precipitation method.The catalyst precursors were prepared by adding an aqueous solution of copper, zinc and/ornickel nitrate, depending on the indicated catalyst component, to a vigorously stirred solutionof Na2CO3 at room temperature. The precipitate was filtered, washed well with distilled waterand dried in air at 110 C for 10 h. The co-precipitated catalyst precursor was then calcined inair at 400 C for 3 h and crashed into a powder of 40-- 60 mesh. The indicated catalystcompositions are based on the weight ratio of the metals.2 Experimental TechniquesThe XRD patterns of the catalysts were recorded on a D/MAX-RB X-ray diffractometerwith nickel-filtered Cu Ka radiation by scanning 20 angle ranging from 10° to 80°.The BET surface area of the prepared catalyst was determined by a surface area analyzer( Micromeritics Co. ，ASAP 2010). The standard deviations of the BET surface areameasurements were within土3%.The copper surface area of the fresh catalyst was obtained by exploiting the selectivechemisorption of N2O on the surface copper sites according to the equation[l2] :N2O + 2 Cu(s)= Cu2O + N2The catalyst of 200 mg was placed in a“U" type quartz reactor(B4 mm). Previously,the catalyst was reduced in situ by a 10% H2(He balanced) at 200 C for 30 min, at 250 Cfor another 30 min, and by pure H2 at 250 C for the last 30 min. Then, the pre-reducedsample was purged by pure He at 250 C for 60 min and maintained at 40 C in He. The totalflow rate in all the period was 50 mL/min. N2O was fed in pulse(200 μL each). Theevaluated N2 and excess N2O were determined on a quadruple mass spectrometer ( Dycorsystem 1 000).3 Activity-selectivity MeasurementsThe calcined catalyst was reduced in situ by 10% H2 in N2 stream ( flow rate :30 mL/min) at 300 C for 2 h before use. Methanol decomposition was carried out in acontinuous flow quartz reactor(04 mm) with a fixed-bed. Typically, 400 mg of catalyst wasused and MHSV=2.0 h-1. The reaction temperature was ranged from 150 to 300 C. Theeffluents were analyzed by using two gas chromatogr中国煤化工:rmal conductivitydetectors( TCD)，respectively. The gaseous productsMHC N M H GH4 were separatedvia a 13X molecular sieve column. COz, dimethyl ether, methyl formate, methanol and theother products with high boiling temperatures were separated via a Porapak Q column.422Chemical Research in Chinese UniversitiesVol. 17Results and Discussion1 XRD StudyIn Fig. 1, the X-ray diffraction patterns of fresh ( unreduced) Cu/Zn(molar ratio 5/5) .and Cu/Zn/Ni (molar ratio 5/4/1) catalysts are compared. For both the samples, onlycrystalline Zn 0 and CuO phases wereobserved, in agreement with the previousreport of Cheng et al.9. There was nsignificant difference in the ZnO intensitybi i山between the two samples.However, theintensities of CuO peaks at 35. 5° and 38. 7°JN'h(20) of the Cu/Zn (molar ratio 5/5) catalystwere much higher compared with those of the20.0 30. 040. 060.020/<*)Cu/Zn/Ni ( molar ratio 5/4/1 ) catalyst, Fig. 1 The XRD patterns of the fresh Cu/Znsuggesting that the third component Ni tended(molar ratio 5/5) (a) and Cu/Zn/Nito promote the CuO dispersion during the(molar ratio 5/4/1)(b) catalysts.period of the catalyst precursors preparation. The calcination and the subsequent hydrogenreduction of the Cu/Zn/Ni catalyst resulted in highly dispersed Cu° active species[10]. Thehighly dispersed Cu° species are very important to the activities of the Cu-based catalysts formethanol decomposition.The XRD patterns of the Cu/Zn (molar ratio 5/5) and Cu/Zn/Ni (molar ratio5/4/1)catalysts having been used for methanol decomposition for 20 h at 250 C are illustrated inFigs.2 and 3，respectively. ZnO and Cu° phases were observed in the pre reduced and theused catalysts. However, no peak corresponding to CuO was detected. In addition, the newintensitive peaks at 42. 3° and 49. 3° were also observed for the used Cu/Zn (molar ratio 5/5)catalyst, suggesting that the CuZn alloys phase was formed. According to the previous workby Jung et al. [13]，these results indicate that ZnO can be reduced only in the presence ofcopper and methanol via forming CuZn alloys. This may be the major reason why the Cu/Zncatalyst is deactivated quickly in the initial stages of the operation. On the contrary, the。0。Zn0。Cu▲CiZn alayjil20.030.040. 050. 060. 020. 030. 00.050.028/(*)中国煤化工Fig.2 The XRD pattern of the used Cu/Zn Fig.MHCNMHGthe used Cu/Zn/Ni(molar ratio 5/5) catalyst for 20 h at(molar ratio 5/4/1 catalyst for 20 h at250 C.250 C. .No.4XI Jing-yu, WANG Zhi-fei, LU Gong -xuan et al.formation of CuZn alloys was not observed in the used Cu/Zn/Ni (molar ratio 5/4/1 )catalyst, suggesting that the nickel additive can maintain the stability of Cu° species byinhibiting the formation of CuZn alloys efficiently. Accordingly, the Cu/ Zn/Ni catalyst couldkeep a high activity for a long period in the methanol decomposition reaction.For all the Cu/Zn/Ni(molar ratio 5/4/x) catalysts, whether fresh or used, no peakscorresponding to metallic nickel and nickel oxide were detected，suggesting that both the Niand/or NiOx were amorphous or highly dispersed in the catalysts.2 BET Surface Area and Copper Surface AreaThe BET surface area and the copper surface area of the Cu/Zn (molar ratio 5/5) andCu/Zn/Ni (molar ratio 5/4/1) catalysts are listed in Table 1. The particle size of Cu° wasestimated from the copper surface area under considering the sphere particle mode![12]. TheBET surface areas of the Cu/Zn-based catalysts prepared according to the co-precipitationmethod are similar to those reported in the previous work[13]. Although the BET areadecreased a little when nickel was added into the Cu/ Zn catalyst, the measured copper surfacearea of the Cu/Zn/Ni(molar ratio 5/4/1) catalyst was 12. 4 m2/g, while the value of the Cu/Zn (molar ratio 5/5) catalyst was only 10. 3 m2/g. Accordingly, the Cu/Zn/Ni (molar ratio5/4/1) catalyst shows higher dispersion of the Cu° active species than the Cu/Zn(molar ratio5/5) catalyst. Those copper surface area and copper dispersity results are consisent well withthose obtained from the XRD study.Table 1 The composition, BET surface area and copper surface area of the catal ystsWeight ratioBET surfaccCopper surfacc Copper dispersityCatalystIn/ nmof the metalarea/(m°.g_")_ area/(m'.g")__ in surfacc(% )Cu/Zn(molar ratio 5/5)51531.710.332. 525. 97Cu/Zn/Ni(molar ratio 5/4/1)514:126. 712.446. 421.613 Catalytic Activity and SelectivityThe previous work in our laboratory about the effects of additives on the performance ofthe Cu/ZnO catalyst for methanol decomposition indicated that nickel is helpful to improve theactivity and the stability of the Cu/ ZnO catalystl14].The effects of the nickel content on methanol conversion, CO selectivity, H2 selectivityand the yield of H2 in methanol decomposition at 250 C over Cu/ Zn/Ni(molar ratio 5/4/x)are shown in Fig. 4，respectively. The beneficial effect of the Ni component is dependent onthe nickel concentration. The promotive effect of nickel was obvious when x increased from0.1 to 1, and the conversion of methanol raised from 28. 5% to 77. 3%. However, when thenickel content further increased, the activities of the Cu/Zn/Ni catalysts maintained arelatively stable value with a 78% of methanol conversion, a 93% of CO selectivity and a95% of H2 selectivity. When x≥> 1, the yield of hydrogen exceeded 70%. The comparativelyhigh selectivity for CO is very important for methal中国煤化工btain maximumhydrogen evolution and improve the combustion valufMYHCNMHGmallamountofmethyl formate was the intermediate product of methanol decomposition, which could befurther decomposed to methanol and CO at high temperatures according to the followingequatiopn[10]，方方数据424Chemical Research in Chinese UniversitiesVol.172CHzOH一+HCOOCH3+2H2-+CHgOH+CO十2H2Fig.5 shows the effect of the nickel concentration on the stability of the Cu/Zn/Nicatalysts in methanol decomposition. In this work, the stability of the catalyst is expressed asthe ratio of the catalytic activity after 20 h operation to the initial activity. The catalyticstability increased with the increase of nickel content while x was smaller than 1 and a .maximum(ca 88%) was obtained when the nickel content was 1. However, the stabilitydecreased to the range from 64% to 79% when x was larger than 1.120 r。Methanol Con. = CO Sel. Hz Sel. aYield Hz1008060 t400t20 t)0.10.5123Nickel Content(x)Nickel Conten1