Methanation of Carbon Monoxide over Alumina Supported Ni-Cu Catalysts z P. K. BAJPAI, N. N. BAKHSHI, LIU DAN-CHU" and J. zyxw F. MATHEWS Department oj Chemical Engineering, University of Saskatchewan, Saskatoon, Saskatchewan, S7N OW0 Copper has zyxwvutsrqp been added zyxwvutsrq as a second metal to a nickel-dispersed-on-alunlina methanation catalyst. Compared with the nickcl catalysts without copper. the initial methanation activity was reduced drastically but the tendency to deactivate with time-on-stream was much less, There were also less carbon deposition and less sintcring of metal particles with the copper-added catalysts. ~~ ~ ~ ~~ .... On a ajoute du cuivre conime second metal au catalyseur de methanation forme de nickel disperse sur de I'alumine. Par rapport aux catalyseurs au nickel lie contenant pas dc cuivrc, on a constate que I'activitC initiate dc methanation zy ;I Cte rCduitc de nianih importante. alors que la tendance h la desactivation avec le temps s'est aver& bien moindre. On a aus4 remarqu6 qu'il s'itait fommC moins de tl6pi)t dc carbonc et une agglomeration moindre de particules ni6talliques sur lcs catalyseurs comportant du cuivrc he t'orniation of inethane from synthesis gas mixtures T has been the sub.ject of many studies and reviews of this work are available (Mills & Steffgen [ 19731; Vannice [ 19763). Kecent interest in methanation is for the production of synthetic natural gas from coal or other fossil fuels. In this case thc leed to the methanator consists of a gas mixture containing a large fraction of carbon monoxide. Due to the highly exothermic nature of this reaction a large amount of heat IS generated. causing problems of sintering and carbon formation. Accordingly it would be desirable to develop catalysts which would be less sensitive to deactivation by sintcring or by the forniation of' carbon deposits at the active sites. Ah regards the latter. thc loss in activity is generally attributed to sonie form of carbonacaeous deposit: elemental carbon. a surface carbide or ;I partially hydrogenated spe- cies. Nickel. one of the most commonly used metals for methanation, has a very high hydrogenolysis activity (Sinl'elt. 1974). Since the hydrogenolysis reaction normally leads to the tormation of carbonaceous residue. one would expect that ;i decrease in the hydrogenolysis activity by the formation of' bimetallic clusters would result in a more stable catalyst (Elliot & Lundsford, 1979). Sinfclt ( 1973) has demonstrated that ruthenium and cop- per I'orin iui integrated system when dispersed on silica support although they are immiscible in the bulk. The intlu- ence of copper on the methunation activity of supported ruthenium has been determined by Bond and Turnham (1976) and Elliot and Lunsthrd (1979). In both studies the presence of copper resulted in a decrease in turnover nurn- her. presumably due to the dilution of multicentered cata- lytic sites by the copper atoms. The activity of ruthenium- nickcl clusters has been found to be indepcndent of time on stream at ;I nickel/ruthcnium wtio of 7: I (Elliot & Lunclslord. 1979). Several nickel catalysts using various second metals (iron, cobalt. platinum. palladium and ruthe- niunil have been tested tor methanation activity by Bartholcnicw ct al. ( 1978). Turnover number measured for thew catalysts show that alloying other 'metals with nickel results in ii relatively modest. hut nevertheless significant. effect on activity. Increases or decreases in specific activity __ ~ *Visiting Chinese Scholar. Fu-Dan University, Shanghai. were in the range 0-200%. Nickel-platinum and nickel- cobalt were slightly more resistant to carbon deposition than the case where nickel was present as the only metal. Araki and Ponec ( 1976) have shown that the addition of copper to nickel strongly reduces the rate of methanation on alloy films, but there has been no corresponding work reported on highly dispersed systems. In the present study work has been done on the effect of the presence of copper on nickel/alumina catalysts with respect to: hydrogen and carbon monoxide cheniisorption on the reduced catalyst; methanation activity: and resistance to carbon deposition, sintering and deactivation with time on stream. Experimental Analytically pure nickel nitrate ( Ni(NO1I2 6H20). cupric nitrate (Cu(NOI), 3H20) and q-alumina (Davison Chemical Co.) were used in thc preparation of the catalysts. The catalysts were prepared by simple impregnation of the sup- port to incipient wetness with aqueous metal salt solutions of known concentrations. The alumina was first impreg- nated with nickel nitrate solution to give a nickel concen- tration of - 13 wt%. This first impregnation was dried at 383 K and then re-impregnated with copper nitrate solution to the desired final copper concentration. These re-inipreg- nated catalysts were dried overnight at 383 K and then calcined at 723 K for 16 h. They were then ground to 177-144 k m size, stored in a desiccator and reduced in the reactor before the start of a test run. Details of the catalyst pre-treatment are given elsewhere (Bhatia et al. 1978; Bajpai et al, 1980). Hydrogen (99.99% pure. Matheson Co.) was passed through a Deoxo unit and then through a column containing dehydrated molecular sieve and indicating drierite before use. Carbon monoxide (99.99% pure. Matheson Co.) was dried by passing it through a molecular sieve and drierite column. Argon (99.Y5% pure. Liquid Carbonic Co.) was used as the carrier gas in the chromatographic analyses. Nickel and copper contents of the calcined and unreduced catalysts were measured by atomic absorption spectroscopy. Gas adsorption ineasurenients were carried out in a con- ventional Pyrex glass volumetric adsorption apparatus capa- THE CANADIAN JOURNAL OF CHEMICAL ENGINEERING. VOLIJMI zyxwvutsrq 60. OCTOBER 1982 ooox-4034182'054~6 I3-0icB I oo/ zyxwvutsr IS 613