Carbon monoxide-nickel adsorption

Direct measurements on metals such as iron, nickel and stainless steel have shown that adsorption occurs from acid solutions of inhibitors such as iodide ions, carbon monoxide and organic compounds such as amines , thioureas , sulphoxides , sulphidesand mer-captans. These studies have shown that the efficiency of inhibition (expressed as the relative reduction in corrosion rate) can be qualitatively related to the amount of adsorbed inhibitor on the metal surface. However, no detailed quantitative correlation has yet been achieved between these parameters. There is some evidence that adsorption of inhibitor species at low surface coverage d (for complete surface coverage 0=1) may be more effective in producing inhibition than adsorption at high surface coverage. In particular, the adsorption of polyvinyl pyridine on iron in hydrochloric acid at 0 < 0 -1 monolayer has been found to produce an 80% reduction in corrosion rate . 

From the results of other authors should be mentioned the observation of a similar effect, e.g. in the oxidation of olefins on nickel oxide (118), where the retardation of the reaction of 1-butene by cis-2-butene was greater than the effect of 1-butene on the reaction of m-2-butene the ratio of the adsorption coefficients Kcia h/Kwas 1.45. In a study on hydrogenation over C03O4 it was reported (109) that the reactivities of ethylene and propylene were nearly the same (1.17 in favor of propylene), when measured separately, whereas the ratio of adsorption coefficients was 8.4 in favor of ethylene. This led in the competitive arrangement to preferential hydrogenation of ethylene. A similar phenomenon occurs in the catalytic reduction of nitric oxide and sulfur dioxide by carbon monoxide (120a). 

Whereas determination of chemisorption isotherms, e.g., of hydrogen on metals, is a means for calculating the size of the metallic surface area, our results clearly demonstrate that IR studies on the adsorption of nitrogen and carbon monoxide can give valuable information about the structure of the metal surface. The adsorption of nitrogen enables us to determine the number of B5 sites per unit of metal surface area, not only on nickel, but also on palladium, platinum, and iridium. Once the number of B5 sites is known, it is possible to look for other phenomena that require the presence of these sites. One has already been found, viz, the dissociative chemisorption of carbon dioxide on nickel. 

Adsorption of carbon monoxide takes place all over the surface and there is distinct evidence that, at least on nickel, the CO stretching frequency depends upon the coordination number of the nickel atom to which it is attached. Hence, the adsorption of carbon monoxide yields information about the relative numbers of surface atoms with different coordination numbers. This information, howrever, is at best merely of a semiquantitative nature. Steric effects also play a role, as is evidenced by the fact that the subcarbonyl species can be formed only on nickel atoms with a lowr coordination number. 

It is true, however, that many catalytic reactions cannot be studied conveniently, under given conditions, with usual adsorption calorimeters of the isoperibol type, either because the catalyst is a poor heat-conducting material or because the reaction rate is too low. The use of heat-flow calorimeters, as has been shown in the previous sections of this article, does not present such limitations, and for this reason, these calorimeters are particularly suitable not only for the study of adsorption processes but also for more complete investigations of reaction mechanisms at the surface of oxides or oxide-supported metals. The aim of this section is therefore to present a comprehensive picture of the possibilities and limitations of heat-flow calorimetry in heterogeneous catalysis. The use of Calvet microcalorimeters in the study of a particular system (the oxidation of carbon monoxide at the surface of divided nickel oxides) has moreover been reviewed in a recent article of this series (19). 

It must be acknowledged, however, that the determination of the number of the different surface species which are formed during an adsorption process is often more difficult by means of calorimetry than by spectroscopic techniques. This may be phrased differently by saying that the resolution of spectra is usually better than the resolution of thermograms. Progress in data correction and analysis should probably improve the calorimetric results in that respect. The complex interactions with surface cations, anions, and defects which occur when carbon monoxide contacts nickel oxide at room temperature are thus revealed by the modifications of the infrared spectrum of the sample (75) but not by the differential heats of the CO-adsorption (76). Any modification of the nickel-oxide surface which alters its defect structure produces, however, a change of its energy spectrum with respect to carbon monoxide that is more clearly shown by heat-flow calorimetry (77) than by IR spectroscopy. 

Fig. 25. Differential heats of adsorption of carbon monoxide at 30°C on fresh (A) or oxygenated (B) samples of a gallium-doped nickel oxide. Reprinted from (63) with permission J. Chim. Phys.

Carbon monoxide oxidation is a relatively simple reaction, and generally its structurally insensitive nature makes it an ideal model of heterogeneous catalytic reactions. Each of the important mechanistic steps of this reaction, such as reactant adsorption and desorption, surface reaction, and desorption of products, has been studied extensively using modem surface-science techniques.17 The structure insensitivity of this reaction is illustrated in Figure 10.4. Here, carbon dioxide turnover frequencies over Rh(l 11) and Rh(100) surfaces are compared with supported Rh catalysts.3 As with CO hydrogenation on nickel, it is readily apparent that, not only does the choice of surface plane matters, but also the size of the active species.18-21 Studies of this system also indicated that, under the reaction conditions of Figure 10.4, the rhodium surface was covered with CO. This means that the reaction is limited by the desorption of carbon monoxide and the adsorption of oxygen. 

Kebulkova, L., Novakova,)., Jaeger, N.I., and Schultz-Ekloff, G. (1993) Characterization of nickel species at Ni/7-Al203 and Ni/faujasite catalysts by carbon monoxide adsorption. Appl Catal. A,... 

Amorphous (rapidly quenched) metals and alloys have been investigated as catalysts (Schlogl, 1985). It has been found that adsorption characteristics of carbon monoxide on metallic glasses are different from those on crystalline materials. For example CO is found to dissociate readily on the surface of Ni76Bi2Si,2 metglass, but it is always molecularly adsorbed on metallic nickel (Prabhakaran Rao, 1985). 

If it is assumed that the mobile oxygen differs from the extralattice oxygen by the absence of an additional electron supplied by the solid, it is quite likely that modifications of the electronic levels of nickel oxide by impurities will not affect substantially the low-temperature rate of carbon monoxide oxidation. Indeed, the rate depends on surface diffusion with subsequent reaction of the adsorbed partners if our scheme is correct. On the contrary such modifications might affect the rate of the high-terapera-ture process insofar as it depends on the availability and heat of adsorption of the extralattice oxygen. As will be seen later, this prediction is correct. 

Fig. 26. Resistance increase of a transparent nickel film (90 X 10 atoms/sq. cm.) on the adsorption of carbon monoxide at T = 90.4°K. [according to (18)].

In carbon monoxide the bond between the atoms depends, as in the N2O molecule, on an asymmetrical electron shift, electrons of the 0 atom moving toward the C atom, and the CO molecule having a dipole character. In this case, too, metal electrons are displaced toward the adsorbed molecule and taken from the electron gas, as shown by the change of the electrical resistance of thin nickel films on carbon monoxide adsorption (18). 

Even the precovering with hydrogen is able to block the surface against electronic interaction. In Fig. 27 a nickel surface was precovered with hydrogen at 3 X 10 mm. Hg to saturation (0 = 0.39), causing an irreversible resistance decrease of 1%. After pumping off at B, carbon monoxide of 6 X 10 mm. Hg was added at Aco- At a pure nickel surface the carbon monoxide influence would have effected an increase of the resistance by 0.8%. At the surface precovered with hydrogen, neither a resistance effect nor a carbon monoxide adsorption is to be observed. 

With the exception of the high initial heat of adsorption of CO on NiO(200), the differential heats of adsorption as a function of the amount of CO adsorbed are similar for both catalysts. Metallic nickel which exists in the sample prepared at 250°C. may chemisorb carbon monoxide (15). However, the metal content is small and cannot account for the heat released in these experiments on NiO(250), since the heat of chemisorption of CO on metallic nickel is still higher (42 kcal. per mole) than the heat registered during adsorption of the first dose (29 kcal. per mole). 

Sequence II O2—CO. Oxygen is first adsorbed on NiO(250) at 30°C. The sample is then evacuated at 30°C. (amount of irreversibly adsorbed oxygen, 1.9 cc. per gram), and carbon monoxide is adsorbed at the same temperature (Figure 3). The electrical conductivity of nickel oxide containing preadsorbed oxygen 1.8 X 10 5 (ohm cm.)"1 decreases during the adsorption of CO, and at the end of the adsorption, is identical to the conductivity of the pure oxide. Moreover, carbon dioxide is condensed in the cold trap. This shows that all ionized species are transformed into neutral species at the end of the interaction. 

Figure 3. Adsorption of carbon monoxide on nickel oxide containing preadsorbed oxygen...

Thus, again the limiting step of both Interactions 2 and 4 cannot be the adsorption of carbon monoxide (Equation 3) since the rates of adsorption of CO, without interaction with preadsorbed species, on nickel oxides containing nearly the same quantity of 0"(ads) ions or C03"(ads) ions should be the same for the same coverage of the surface and the rate of production of heat would be the same. 

A systematic attempt to correlate the catalytic effect of different surfaces with their adsorptive capacity was made by Taylor and his collaborators. Taylor and Burns, for example, investigated the adsorption of hydrogen, carbon dioxide, and ethylene by the six metals nickel, cobalt, palladium, platinum, iron, and copper. All these metals are able to catalyse the hydrogenation of ethylene to ethane, while nickel, cobalt, and palladium also catalyse the reduction of carbon monoxide and of carbon dioxide to methane. 

Spectra of CO adsorbed on nickel and copper films obtained by Bailey and Richards (23) are shown in Figure 11. Carbon monoxide adsorption at 77°K resulted in the appearance of only a single band at 2097 cm l for nickel and 2109 cm l for copper. Further CO exposure of the sample at 1.5°K resulted in the observation of a second band near 2143 cm l on both metals. The low frequency band in each case was identified with chemisorbed CO while the high frequency band was associated with physisorbed CO. Bailey and Richards (23) have also used their apparatus to obtain spectra of N2 chemisorbed on nickel, and benzene and... 

---