Volume 218, Issue 8

Adsorption of CO and O 2 on Pd particles deposited on a well-ordered Fe 3 O 4 (111) film was studied by temperature programmed desorption. Scanning tunneling microscopy was used to provide structural information. The results show that CO adsorbed on Pd particles reacts with oxygen of the oxide support and forms CO 2 . The reaction occurs at the particle/oxide interface and exhibits a particle size effect such that the smaller particles produce more CO 2 . Oxidation of Pd/Fe 3 O 4 alters the oxide structure due to Pd catalyzed oxygen migration into the film and results in formation of an oxygen reservoir, which favors CO oxidation reaction.

Low energy electron diffraction (LEED) structure determinations have been performed for the p (2×2) structures of pure oxygen and oxygen co-adsorbed with CO on Ni{111}. Optimisation of the non-geometric parameters led to very good agreement between experimental and theoretical IV -curves and hence to a high accuracy in the structural parameters. In agreement with earlier work atomic oxygen is found to adsorb on fcc sites in both structures. In the co-adsorbed phase CO occupies atop sites. The positions of the substrate atoms are almost identical, within 0.02 Å, in both structures, implying that the interaction with oxygen dominates the arrangement of Ni atoms at the surface.

We construct a statistical mechanical model for a polymer brush in θ -solvent. The model is exactly solvable for realistic polymer models ( e.g. the freely rotating chain and the rotational isomeric state models) and it readily accounts for inhomogeneities of the brush with respect to the plane parallel to the grafting surface. The interactions between neighboring chains is mimicked by the presence of a confinement potential centered on the grafting point of each chain. We explore the model to calculate the equilibrium and kinetic properties of the polymer brush. We find that the variations of the brush free energy and height with the coverage depend strongly on the stiffness of the polymer chain (bond angle) and on the severity of the chain confinement. The results for the growth kinetics of the brush are compared with recent measurements on the formation of a PEG2000 brush from solution.

The adsorption of methanol on the clean and oxygen precovered Cu(111) surface was studied by means of high resolution X-ray photoelectron spectroscopy (XPS), ultraviolet photoelectron spectroscopy (UPS), and temperature programmed desorption (TPD). On clean Cu(111) methanol is adsorbed molecularly at 100 K. Only small amounts of methoxy are formed by heating to 160–180 K, whereas the main part of methanol desorbs molecularly. The formation of methoxy drastically increases on the oxygen precovered surface. In the UP spectra the peaks of molecular methanol disappear and the peaks of methoxy develop. TPD measurements show that methoxy desorbs dissociatively as formaldehyde and hydrogen with the rate maxima at around 430 K. The efficiency of the oxidation of methanol to formaldehyde depends on the amount of predosed oxygen, with a maximum methoxy coverage obtained for an oxygen precoverage of 0.18 ML. From the quantitative analysis we propose an additional reaction pathway involving not only the formation, but also the decomposition of an hydroxyl intermediate, parallel to the previously reported pathway, which leads to the desorption of water.

The adsorption of CO and N 2 on KCl(100) single crystal cleavage planes has been investigated by means of low energy electron diffraction (LEED) at primary currents in the nA range as well as polarization infrared spectroscopy (PIRS) in transmittance geometry. These isoelectronic adsorbates behave very similarly, and three different adsorption phases can be distinguished as a function of surface coverage for both of them. Initially a (1×1) structure is observed and assigned to a commensurate monolayer with one molecule per KCl ion pair. The CO infrared spectrum of this ‘phase I’ is characterized by a doublet absorption of the CO stretching vibration, which is discussed in the context of a correlation field splitting. The lack of superstructure diffraction peaks is attributed to a high degree of orientational disorder in this phase. From LEED adsorption isotherms the isosteric heat of adsorption of N 2 has been determined to be 11±3 kJ mol −1 . Upon increase in coverage by 50% ‘phase II’ is formed which exhibits a large number of additional diffraction spots. It is assigned to a bilayer in which the second layer is only half filled. Based on the LEED experiments a structure model is proposed in which the second layer is characterized by growth in rows along the [210] direction and a high degree of roughness in the perpendicular direction. This model can explain all experimentally observed main features. Finally upon further exposure formation of three-dimensional solid is observed, which proceeds via Stransky–Krastanov growth of crystalline 3D clusters with the structure of the low-temperature cubic α -phase. A simulation of the CO cluster infrared spectra within the dipole–dipole coupling approach can reproduce all major observed vibrational features.

Ir(100) is well known to exhibit a (5×1) periodic surface reconstruction denoted as (5×1)-hex. The atoms of the outermost layer are quasi-hexagonally arranged rather than being of quadratic order as the layers below and in the bulk. Accordingly, an excess of 20% of atoms is accommodated in the top layer and the surface is buckled with atomic rows protruding from the surface. It is shown by the combined application of atomically resolved scanning tunnelling microscopy (STM) and quantitative low-energy electron diffraction (LEED) that hydrogen adsorption at sufficiently high temperature induces the surface to restructure. The new phase, (5×1)-H, develops by the extraction of surface protruding and close packed atomic Ir wires from the hexagonal layer. The remaining atoms of the latter reassume quadratic order and the wires expelled reside on this layer. The extraction of the Ir wires proceeds in a zip-like process and starts preferentially at surface defects, in particular at step edges or domain boundaries. The wires’ lateral periodicity normal to their spatial orientation is again 5-fold, but only on average. The precise lateral arrangement depends sensitively on the details of preparation, in particular the amount of hydrogen exposure and the sample’s temperature during, or temperature treatment after, exposure.

Various pseudohomogeneous models for the selective catalytic reduction of NO x by methane were developed by considering reaction schemes and kinetic mechanism. The models were applied to predict the performance of Ce/In-promoted sulfated zirconia catalysts in a wide temperature range from 300 to 700 °C in a fixed bed reactor assuming plug-flow. A nonlinear kinetic model fits the experimental data better than a simple linear model. The mass balance differential equations were solved numerically. The model parameters were estimated by the Simplex-method of Nelder–Mead with a fourth order Runge–Kutta–Merson-method. The finally chosen model is better suited for the prediction of the In-free catalysts than for the mixed oxide catalysts or the Ce-free catalysts. The model suggests that CO formed in the process plays no role in the reduction of NO by CH 4 . This was confirmed by experimental evidence. The investigated Ce/In zirconia system was better suited for NO reduction by CH 4 than those published previously.