We present results of our recent investigations detailing the growth and properties of oxygen phases prepared on Pt(111) and Pt(100) surfaces in ultrahigh vacuum using oxygen atom beams. Our studies reveal common features in the oxidation mechanisms of Pt(111) and Pt(100). On both surfaces, oxygen atoms initially populate a chemisorbed phase, and then incorporate into intermediate phases prior to the growth of bulk-like oxide. The bulk oxide grows on both surfaces as three-dimensional particles with properties similar to those of PtO 2 and decomposes explosively during heating, exhibiting higher thermal stability than the intermediate oxygen phases. Our results suggest that kinetic barriers stabilize the oxide particles against decomposition, thereby producing explosive desorption, and hence also hinder Pt oxide growth at low coverages. We also find that the kinetics of bulk oxide formation on Pt(100) measured as a function of the O atom incident flux and surface temperature is quantitatively reproduced by a model based on a precursor-mediated mechanism. The model assumes that oxygen atoms adsorbed on top of a surface oxide phase act as a precursor species that can either associatively desorb or react with the surface oxide to produce a bulk oxide particle. Similarities in the development of intermediate oxygen phases on Pt(100), Pt(111) and other transition metal surfaces suggest that precursor-mediated kinetics may be a general feature in transition metal oxidation. Finally, we find that Pt oxide particles are less active than lower-coverage oxygen phases on Pt(111) and Pt(100) toward the oxidation of CO, and that the reaction exhibits autocatalytic kinetics that can be explained with a model that treats the reaction as occurring within a dilute oxygen phase that coexists with oxide particles.

No abstract

We utilized temperature-programmed reaction spectroscopy (TPRS) to investigate the reactivity of oxygen-covered Pt(100) toward the oxidation of CO. The reaction is facile on oxygen phases that form below coverages of about 1 ML (monolayers), producing complicated CO2 desorption traces. However, the reaction is inefficient when three-dimensional (3D) oxide particles cover the surface. We observe CO2 production in roughly three temperature regimes during TPRS. Between 120 and 220 K, O atoms react with CO molecules adsorbed within oxygen phase domains. In this regime, the reactivity toward CO is highest for the two-dimensional (2D) surface oxide and decreases in the order of metastable O phase, O chemisorbed in a disordered (3 × 1) phase, and 3D oxide. Decreases in the CO and O binding strengths generally correlate with increasing reactivity of the oxygen phases below 220 K. At temperatures from about 225 to 335 K, reaction involves species at the boundaries of separate O and CO domains. As the temperature surpasses 335 K, CO2 production occurs between CO and disordered or isolated O adatoms. Within the second temperature regime, our data suggests that both Pt and O atoms migrate away from 2D and 3D oxide domains and into CO domains where reaction occurs. Kinetic barriers associated with the creation and transport of Pt adatoms likely inhibit CO oxidation by the 3D oxide. We also find that the generation of Pt adatoms facilitates significant surface defect formation and oxygen loss to the bulk during reaction. The results of this investigation clearly demonstrate that the surface oxygen phase distribution strongly influences the temperature-programmed reaction of coadsorbed CO and O on Pt(100).

We utilized direct rate measurements and temperature programmed desorption to investigate reactions that occur during the collisions of gaseous oxygen atoms with deuterium-covered Pt(100). We find that both D2O and D2 desorb promptly when an oxygen atom beam impinges upon D-covered Pt(100) held at surface temperatures ranging from 90 to 150 K, and estimate effective cross sections of 12 and 1.8 A2, respectively, for the production of gaseous D2O and D2 at 90 K. The yields of D2O and D2 that desorb at 90 K are about 13% and 2%, respectively, of the initial D atom coverage, though most of the D2O product molecules (approximately 80%) thermalize to the surface rather than desorb at the surface temperatures studied. Increasing the surface temperature from 90 to 150 K causes the D2O desorption rate to decay more quickly during O atom exposures to the surface and results in lower yields of gaseous D2O. We attribute the production of D2O and D2 in these experiments to reactions involving intermediates that are not thermally accommodated to the surface, so-called hot precursors. The results are consistent with the production of hot D2O involving first the generation of hot OD groups from the reaction O*+D(a)-->OD*, where the asterisk denotes a hot precursor, followed by the parallel pathways OD*+D(a)-->D2O* and OD*+OD(a)-->D2O*+O(a). The final reaction contributes significantly to hot D2O production only later in the reaction period when thermalized OD groups have accumulated on the surface, and it becomes less important at higher temperature due to depletion of the OD(a) concentration by thermally activated D2O production.

No abstract