Vacuum~volume 32/number 5/pages 277 to 281/1982 0042-207X/82/050277-05503.00/0 Printed in Great Britain Pergamon Press Ltd Interactions between co-adsorbed CH 4 and CO on tungsten: ESD and flash desorption study M P Lbpez Sancho and J M Lbpez Sancho, Instituto de Fisica de Materiales, CSIC Serrano 144, Madrid-6, Spain received 11 January 1981 The interactions between methane and carbon monoxide adsorbed species on tungsten have been investigated by electron stimulated desorption and flash desorption techniques. For small CO coverage with respect to CH 4 no changes were found in the CH,L desorption spectrum. It was observed that the second desorption state of CH 4 decreases when the CO/CH,¢ coverage ratio increases disappearing at a CO coverage of 2.4 x 103 molecules cm -=. No changes were found in the first desorption state of CH 4 at any CO/CH= coverage ratios. These results are compared with those found by other investigators in H= and CO co-adsorption in order to clarify unresolved questions in the adsorption of CH,¢. 1. Introduction Although the adsorption of CH4 on tungsten has been studied for several years by a variety of techniques such as FEM 1"4, thermal desorption 5'7, electron stimulated desorption 8 and isotopic exchange experiments 9't2 some questions are still unresolved. Specifically confirmation is desired as to whether dissociative adsorption occurs and on which sites the species are actually adsorbed. As is well known the thermal desorption spectrum presented by a CH 4 layer adsorbed on tungsten [either polycrystalline, or a (100) plane] at room temperature is exclusively formed by molecular hydrogen. A great deal of information is already available about the interactions between co-adsorbed hydrogen and carbon monoxide on the (100) phase of tungsten 13"t6. Consequently we thought it worthwhile to use CO as a chemical probe for the study of methane adsorbed on tungsten. It will be shown that the interactions between CH4 and CO are not the same as those found between adsorbed H E and CO. The study of the interactions between adsorbed CO and CH4 is interesting not only for the understanding of the CH 4 adsorption but also for the study of the economically important CO methanation process (3 H E-I-CO--.CH4 + HEO) in order to see if tungsten is a good catalyst for this reaction. If it is a good catalyst, CH4 desorption will occur and the desorption temperature of the CH4 and HEO will be louver than that of the co-adsorbed HE and CO. The experimental results of Benziger and Madixl 6 showing that CH 4 is desorbed from a CO+H E layer at about 325 K made this possibility more likely but, as will be shown, a CHJW layer inhibits the adsorption of CO, thus preventing the CH4 from reacting. Since the initial sticking coefficient for CH 4 on tungsten is much lower ( ~ 10- a)a than for CO ( ~ 3 x 10- t )22 when the ion pump is turned off, co-adsorption experiments were performed with the ion pump in operation (which increases the CH, sticking coefficient as was seen for CH4 on Ni by Schouten et allT). Since direct ion implantation is prevented in our system by electrical shielding between ionic pump and substrate the increase must be due to a part of the methane molecules being excited, when the ion pump is operating, into a state which adsorbs faster than ground state CH48"17. The sticking coefficient for CH4 excited by the pump operation is ~ 10 -3 a, somewhat closer to that of CO, which allowed us to perform co-adsorption experiments with much lower contamination than when the pump was off. 2. Apparatus The experiments have been performed in a metallic system, evacuated by an ion pump and a titanium sublimation pump immersed in liquid nitrogen. After baking out the system at 400 K, ultimate pressures of about 10-10 torr are reached, the residual vacuum being mainly HE, CH4, CO. The sample was a 2 cm x 0.5 cm x 0.2 x 10- a cm polycrystal- line 99.95~ pure tungsten ribbon, recrystailized by heating in ultra-high vacuum. [X-ray analysis showed that the (100) plane predominated on the surface and no other lines were observed above noise level.] The ribbon shape was chosen so as to have a homogeneous adsorption surface large enough to avoid end effects in the thermal desorption experiments. It could be resistively heated to 2500 K and, being on a Kelvin bridge, the sample temperature was accurately known and stabilized within 2K. The sample, held by a couple of pure tungsten rods, was placed in a home-made desorption cell amply described in the litera- ture ta'19, the electron source being a 99.95~ pure tungsten filament. The surface ion current was measured through a quadrupole mass spectrometer, in line of sight with the sample. 277